Browse the corpus

Hypertension is known as the “silent killer”: it rarely causes symptoms and is the chief modifiable risk factor driving a global health crisis of cardiovascular and renal disease.1 Hypertension exerts a major socioeconomic burden, costing in the United States, for example, more than 46 billion dollars per year to manage in terms of direct health care and associated costs.2 At the millennium it was estimated that 972 million adults had hypertension; a 60% increase is predicted by 2025, which means that approximately 1.5 billion adults will have hypertension within the next decade.3 To arrest this increase in prevalence, the World Health Organization (WHO) proposes a dual strategy of improving access to inexpensive and effective therapeutic agents alongside education to improve long-term cardiovascular health through lifestyle changes. A key target here is an approximately 30% relative reduction in mean population salt intake. In most countries the average salt (NaCl) intake is 9 to 12 g per person per day, against the WHO recommended intake of 5 g per day.4 Although the relationship between salt intake and cardiovascular mortality is u-shaped, targeting salt reduction toward the recommended daily allowance (RDA) would be beneficial5: based on the INTERSALT study, long-term compliance would lower blood pressure and significantly reduce cardiovascular events later in life.6

Although the relationship between salt intake and cardiovascular mortality is u-shaped, targeting salt reduction toward the recommended daily allowance (RDA) would be beneficial5: based on the INTERSALT study, long-term compliance would lower blood pressure and significantly reduce cardiovascular events later in life.6 How Does Salt Increase Blood Pressure? What is the relationship between salt intake that is habitually high and long-term blood pressure? This is difficult to gauge, as most clinical studies estimate salt intake from measurement of 24-hour urinary sodium excretion and recent research clearly shows that this is not an effective index of intake.7 Nonetheless, hypertension research has been strongly influenced by the computational modeling of Guyton et al., which placed renal function at the center of long-term blood pressure homeostasis.8 The control of effective intravascular volume through urinary excretion of salt and water offsets any perturbation, stabilizing blood pressure around the individual’s set point. For example, an increase in arterial pressure will increase renal arterial pressure and, in turn, will cause blood flow through the medullary vasa recta to rise. This hemodynamic effect promotes the release of a variety of paracrine factors, such as adenosine triphosphate (ATP) and nitric oxide, which directly inhibits sodium transport in the proximal tubule, thick limb of Henle, and distal nephron. This vascular−tubular cross-talk underpins the pressure natriuresis response, that is, the direct relationship between sodium excretion and renal perfusion pressure.9 If pressure natriuresis achieves long-term stability of blood pressure, then it follows that hypertension can be sustained only if the renal response to elevated blood pressure is impaired: that is, hypertension is caused by renal dysfunction.10 Indeed, it is well documented that the acute pressure natriuresis curve is right-shifted in hypertension; more importantly, the gradient is often blunted. How such acute “loss of function” integrates into chronic blood pressure control is not very well defined.

d: that is, hypertension is caused by renal dysfunction.10 Indeed, it is well documented that the acute pressure natriuresis curve is right-shifted in hypertension; more importantly, the gradient is often blunted. How such acute “loss of function” integrates into chronic blood pressure control is not very well defined. It may initially manifest as loss of the normal nocturnal dip, with blood pressure remaining high to facilitate sodium excretion and maintain balance.11 It is also argued that the acute pressure natriuresis mechanism is not the only—or indeed, not the most important—mechanism of sodium and fluid homeostasis12: a large body of work in humans, recently reviewed,13 finds that high salt intake can increase blood pressure without inducing volume expansion. We also find this in murine models of glucocorticoid hypertension.14, 15 In both situations, hypertension seems to reflect increased vascular tone and/or enhanced activity of the sympathetic nervous system rather than sodium retention and volume expansion.

can increase blood pressure without inducing volume expansion. We also find this in murine models of glucocorticoid hypertension.14, 15 In both situations, hypertension seems to reflect increased vascular tone and/or enhanced activity of the sympathetic nervous system rather than sodium retention and volume expansion. It is certain that chronically elevated blood pressure reflects interactions between multiple systems (Figure 1), but this article will not discuss the relative merits of renocentric, vasculocentric, or neurogenic views of hypertension; rather, we focus on the evolution of hypertension, discussing emerging concepts presented at the ISN Forefront Meeting “Immunomodulation of Cardio-renal Function,” held in Shenzhen, China. First, we examine the hypothesis that the hypertension pandemic reflects a discord between our ancestral genes and our current high-salt environment. What can “Evolutionary Medicine” tell us about high blood pressure? Second, the molecular pathways controlling salt reabsorption in the kidney are expressed in other areas important for salt balance, including the brain. We discuss recent research showing that these central pathways can influence salt intake and blood pressure without altering renal function.Figure 1 The hypertensive storm. There is strong evidence for a dynamic interaction among hormonal (chiefly the renin−angiotensin−aldosterone system), immune, and autonomic nervous systems in the physiological regulation of blood pressure. In the context of a high-salt diet, these interactions may become maladaptive, causing hypertension. Antagonists of the RAAS are front-line antihypertensive treatments, and device-based interventions are targeting the nervous system. The World Health Organization advocates reducing dietary sodium intake to ∼5 g/d. ACEIs, angiotensin-converting enzyme inibitors; ARBs, angiotensin receptor blockers; RAAS, renin−angiotensin−aldosterone system.

are front-line antihypertensive treatments, and device-based interventions are targeting the nervous system. The World Health Organization advocates reducing dietary sodium intake to ∼5 g/d. ACEIs, angiotensin-converting enzyme inibitors; ARBs, angiotensin receptor blockers; RAAS, renin−angiotensin−aldosterone system. Figure 1

are front-line antihypertensive treatments, and device-based interventions are targeting the nervous system. The World Health Organization advocates reducing dietary sodium intake to ∼5 g/d. ACEIs, angiotensin-converting enzyme inibitors; ARBs, angiotensin receptor blockers; RAAS, renin−angiotensin−aldosterone system. Figure 1 Ancestral Genes and the Evolution of Hypertension As recently reviewed,16 the processes through which sodium balance is regulated have evolved over millennia. The renin−angiotensin system, for example, emerged approximately 400 million years ago in the Paleozoic era, as marine organisms moved to the land and faced a strong selection pressure to conserve an essential micronutrient. It is proposed that these genes no longer fit: the Ancestral-Susceptibility hypothesis posits that hypertension, like other complex modern conditions, is a “disease of civilization” because of a mismatch between ancient genomes and current environment.17 According to this theory, ancestral gene variants that promoted efficient sodium retention accrued a selection advantage as humans first developed in the hot and dry African savannah with sodium chloride a scarce nutrient.18 The selection pressure changed following the African diaspora, but the latitudinal cline of heat adaptation remained a strong driver of natural selection.19 This hypothesis20 is broadly supported by the difference in hypertension prevalence in populations: Individuals of white ethnicity have a lower prevalence of salt-sensitive hypertension than do African Americans; populations from hot climates are more susceptible to hypertension than those from cold climates.19 At the gene level, there is evidence that ancestral “sodium-conserving” variants contribute to the phenotypic variability of blood pressure. For example, variants in the AGT promotor that increase circulating angiotensinogen encoding gene are found at higher frequency in African populations21; loss of function variants expected to reduced salt avidity have risen to a higher frequency outside of Africa.21 Similar observations have been made for single nucleotide polymorphisms (SNPs) in the genes encoding the α and γ subunits of ENaC19 and for kinases regulating major sodium transport proteins.22

opulations21; loss of function variants expected to reduced salt avidity have risen to a higher frequency outside of Africa.21 Similar observations have been made for single nucleotide polymorphisms (SNPs) in the genes encoding the α and γ subunits of ENaC19 and for kinases regulating major sodium transport proteins.22 How should these data be interpreted? One possible inference is that hypertension arises from a mismatch between environment and ancestral salt-conserving variants. Indeed, the gain-of-function variants would impair the pressure natriuresis response, promoting sodium retention. Indeed, such variants in angiotensinogen, for example, are associated with hypertension in the general population.23 The persistence of ancestral variants in the molecular machinery for salt conservation becomes deleterious when the environment changes. In our salt-saturated society, a genome aligned with sodium avidity is maladaptive, increasing blood pressure and cardiovascular risk.

d with hypertension in the general population.23 The persistence of ancestral variants in the molecular machinery for salt conservation becomes deleterious when the environment changes. In our salt-saturated society, a genome aligned with sodium avidity is maladaptive, increasing blood pressure and cardiovascular risk. However, to contextualize hypertension as a misalignment of ancestrally favorable “blood pressure” variants is too narrow a view. Recent studies suggest that hypertension is a modern bystander effect of selective pressure imposed to conserve other desirable traits. For example, an ancestral variant in the APOL1 gene, which encodes apolipoprotein-L1, is observed in higher frequency in African Americans, contributing to higher rates of cardiovascular and renal disease.24 The disease-causing mechanism is not defined, but it is likely that the positive evolutionary selection pressure on the ancestral variant reflects improved protection against infection by Trypanosoma brucei, which causes sleeping sickness, rather than a beneficial effect on blood pressure homeostasis.

renal disease.24 The disease-causing mechanism is not defined, but it is likely that the positive evolutionary selection pressure on the ancestral variant reflects improved protection against infection by Trypanosoma brucei, which causes sleeping sickness, rather than a beneficial effect on blood pressure homeostasis. A similar picture is emerging for gain-of-function variants in UMOD, the gene encoding uromodulin (Tamm-Horsfall) protein. Common gain-of-function SNPs in UMOD associate with hypertension, low glomerular filtration rate, and risk of renal disease.25, 26, 27, 28 The evolutionary genetics of 1 “risk” SNP (rs4293393) was examined further and identified as the ancestral allele based on expression in the genomes of nonhuman primates.29 Further sequencing of ancient hominid genomes identified a protective allele, but this is now found only at low frequency. Overall, this indicates that the evolution of modern man placed a strong selection pressure in favor of the ancestral, risk-associated allele, probably because this confers protection against bacterial urinary tract infection and regulates the innate immune system.29 What does Evolutionary Medicine tell us about hypertension? First, these studies underscore the necessity to return to a more “primitive” diet, low in sodium and high in potassium. Second, they can provide new mechanistic insights into blood pressure control: the study of ancestral variants is identifying new loci associated with hypertension, including kinases that regulate ENaC and sodium chloride co-transporter (NCC).22

to return to a more “primitive” diet, low in sodium and high in potassium. Second, they can provide new mechanistic insights into blood pressure control: the study of ancestral variants is identifying new loci associated with hypertension, including kinases that regulate ENaC and sodium chloride co-transporter (NCC).22 Salt Appetite and Blood Pressure Early terrestrial animals evolved highly effective strategies to conserve sodium, and it is evident that both afferent (i.e., salt taste and hunger; gastrointestinal absorption) and efferent (i.e., renal excretion; sweat) arms of salt homeostasis engage a conserved molecular framework of sodium transport proteins and regulatory kinases (Figure 2). Mutations in the genes encoding these key proteins cause Mendelian (monogenic) blood pressure disorders, all of which have an impact on sodium homeostasis.30Figure 2 Molecular mechanisms under the control of the renin−angiotensin−aldosterone (RAAS) system are expressed in multiple sites that influence sodium homeostasis and blood pressure. BBB, blood−brain barrier; CSF, cerebrospinal fluid; ENaC, epithelial sodium channel; NTS, nucleus of the solitary tract. Figure 2

Salt Appetite and Blood Pressure Early terrestrial animals evolved highly effective strategies to conserve sodium, and it is evident that both afferent (i.e., salt taste and hunger; gastrointestinal absorption) and efferent (i.e., renal excretion; sweat) arms of salt homeostasis engage a conserved molecular framework of sodium transport proteins and regulatory kinases (Figure 2). Mutations in the genes encoding these key proteins cause Mendelian (monogenic) blood pressure disorders, all of which have an impact on sodium homeostasis.30Figure 2 Molecular mechanisms under the control of the renin−angiotensin−aldosterone (RAAS) system are expressed in multiple sites that influence sodium homeostasis and blood pressure. BBB, blood−brain barrier; CSF, cerebrospinal fluid; ENaC, epithelial sodium channel; NTS, nucleus of the solitary tract. Figure 2 This framework is exemplified in the principal cell of the distal nephron (Figure 3). Aldosterone is a major regulator of sodium balance: activation of the mineralocorticoid receptor stimulates sodium transport through coordinated activation of the Na,K-ATPase in the basolateral membrane and the epithelial sodium channel (ENaC) in the apical membrane. This process is underpinned by activation of serum and glucocorticoid induced kinase 1 (sgk1) to promote ENaC insertion and to suppress ubiquitination and retrieval through Nedd4-2, prolonging ENaC retention in the apical membrane. Additional regulation is achieved by the enzyme 11β-hydroxysteroid-dehydrogenase type 2 (11βHSD2), which converts “active” glucocorticoids into derivatives that do not activate the mineralocorticoid receptor (MR).31Figure 3 The molecular framework for sodium transport in the principal cell of the collecting duct. Aldosterone is a regulator via the mineralocorticoid receptor, activating transport through a network of regulatory kinases. The enzyme 11β-hydroxysteroid dehydrogenase metabolizes cortisol to cortisone, which does not activate the mineralocorticoid receptor (MR). Glucocorticoids can activate the epithelial sodium channel (ENaC) when in excess.

via the mineralocorticoid receptor, activating transport through a network of regulatory kinases. The enzyme 11β-hydroxysteroid dehydrogenase metabolizes cortisol to cortisone, which does not activate the mineralocorticoid receptor (MR). Glucocorticoids can activate the epithelial sodium channel (ENaC) when in excess. Figure 3 The same molecular machinery contributes to sodium absorption in the GI tract, activation of salt-taste receptors on the tongue,32 control of sodium appetite and sympathetic drive by the brain,33, 34, 35 and also endothelial function/peripheral vascular tone.36 Conservation of mechanisms across multiple systems makes evolutionary sense, but how these integrate to control blood pressure is not well defined. In the rest of this article, we focus on mineralocorticoid receptors and the control of salt intake.

in,33, 34, 35 and also endothelial function/peripheral vascular tone.36 Conservation of mechanisms across multiple systems makes evolutionary sense, but how these integrate to control blood pressure is not well defined. In the rest of this article, we focus on mineralocorticoid receptors and the control of salt intake. Salt appetite is often overlooked as a factor for hypertension and, despite evidence of benefit, compliance with restricted salt intake is poor.37 This, of course, reflects the abundance of salt in modern foods, but there may also be a physiological context: heart failure patients show increased preference for salty foods,38 and mammals have evolved pathways in the brain that evoke salt appetite in response to sodium/volume depletion.39 For example, intracerebrovascular infusion of aldosterone increases blood pressure without altering renal function in rats.40 Similarly, intracerebrovascular infusion of 11βHSD2 inhibitors causes hypertension in rats, without any measurable effect on whole-body sodium balance.41 11βHSD2 metabolizes cortisol and protects MR from overactivation by glucocorticoids.31 Null mutations in the encoding gene cause the Mendelian syndrome of Apparent Mineralocorticoid Excess. Hypertension in this setting is severe and considered to be renal in origin, as the enzyme is abundantly expressed in the distal nephron.31 Nevertheless, patients also have strong salt appetite despite suppressed aldosterone,42 and in the general population loss-of-function variants in HSD11B2 positively associate with sodium intake.43 MR co-localizes with 11βHSD2 in only a few areas of the brain, populations of neurons that can be considered as classical “aldosterone” target cells.44 Several lines of evidence suggest that these 11βHSD2-expressing neurons contribute importantly to salt appetite and blood pressure control. In rats, 11βHSD2-expressing neurons in the nucleus of the solitary tract (nucleus tractus solitarius; NTS) are selectively activated by sodium depletion and rapidly inactivated when salt appetite is satiated.45 In mice, global genetic deletion of 11βHSD2 causes hypertension and renal sodium retention.46 Deleting the enzyme in the NTS alone does not change blood pressure and does not impair renal function.47 Nevertheless, these “brain knockout” mice provide strong evidence indicating that that 11βHSD2 activity in the NTS normally exerts a significant influence on sodium homeostasis and BP control by regulating MR activation.

ng the enzyme in the NTS alone does not change blood pressure and does not impair renal function.47 Nevertheless, these “brain knockout” mice provide strong evidence indicating that that 11βHSD2 activity in the NTS normally exerts a significant influence on sodium homeostasis and BP control by regulating MR activation. Deletion of the enzyme only in the NTS uncovers an innate salt preference such that salt intake increased 3-fold in the absence of any overt physiological driver to consume salt. Furthermore, this increased salt intake induced hypertension: the effect was permissive, as BP did not rise in control mice fed the same high level of salt. This suggests that aldosterone-sensitive neurons in the NTS normally co-ordinate a response to increased salt intake such that blood pressure does not rise. The mechanisms are not yet resolved, but the knockout animals had an exaggerated pressor response to catecholamine and an impaired baroreflex. Other studies show that increased ENaC expression in the choroid plexus and in neurons promotes an exaggerated pressor response to dietary salt.34 Important here may be the homeostatic regulation of sodium concentration in the cerebrospinal fluid: small elevations (∼5 mmol/l) increase sympathetic outflow, which causes hypertension by peripheral vasoconstriction and by direct activation of sodium transporters including NCC48 and ENaC,49 shifting the pressure natriuresis curve to the right and reducing the slope of the response.

tration in the cerebrospinal fluid: small elevations (∼5 mmol/l) increase sympathetic outflow, which causes hypertension by peripheral vasoconstriction and by direct activation of sodium transporters including NCC48 and ENaC,49 shifting the pressure natriuresis curve to the right and reducing the slope of the response. It is not certain whether MR activation in the CNS is determined by aldosterone penetrating the blood−brain barrier or whether aldosterone is also produced in the brain.50 Recent evidence suggests that central synthesis is promoted by peripheral aldosterone excess to amplify the hypertensive response to salt.35

tration in the cerebrospinal fluid: small elevations (∼5 mmol/l) increase sympathetic outflow, which causes hypertension by peripheral vasoconstriction and by direct activation of sodium transporters including NCC48 and ENaC,49 shifting the pressure natriuresis curve to the right and reducing the slope of the response. It is not certain whether MR activation in the CNS is determined by aldosterone penetrating the blood−brain barrier or whether aldosterone is also produced in the brain.50 Recent evidence suggests that central synthesis is promoted by peripheral aldosterone excess to amplify the hypertensive response to salt.35 Dysregulation of central salt-regulating pathways can compromise long-term adherence to restricted sodium intake and promote hypertension. Can these processes be targeted to improve cardiovascular outcome? Certainly the brain and the kidney “talk” to each other, and bilateral renal denervation has gained considerable traction as a device-based approach to manage blood pressure in patients with resistant hypertension.51, 52 Interest has waned considerably since the first randomized, double-blinded, placebo-controlled trial (Symplicity HNT-3) failed to reach its primary endpoint and was halted early in 2014.53 Experimental hypertension, a setting in which effective denervation can be rigorously determined, shows that interrupting brain−kidney communication is more effective in some models (e.g., obesity) than others (e.g., Dahl salt-sensitive). Post hoc analysis of Symplicity-HTN3 found benefit of denervation in hypertensive patients with obstructive sleep apnea,54 suggesting that patient stratification might lead to a renaissance of this therapeutic approach.

munication is more effective in some models (e.g., obesity) than others (e.g., Dahl salt-sensitive). Post hoc analysis of Symplicity-HTN3 found benefit of denervation in hypertensive patients with obstructive sleep apnea,54 suggesting that patient stratification might lead to a renaissance of this therapeutic approach. Summary For most of mankind’s existence, sodium chloride was a scarce nutrient. This scarcity gave great economic value to salt and shaped the formation and customs of our societies, both ancient and modern. It is also reflected in our DNA, encoding the multiple interlocking pathways that efficiently control salt balance. However, our salt intake is now habitually high, and these genes no longer fit: blood pressure rises, and cardiovascular disease is the leading cause of global mortality. It is clear that BP homeostasis is intimately associated with sodium homeostasis and the distribution of sodium between fluid compartments and within tissues. Research has given us a more sophisticated understanding of blood pressure control, revealing a dynamic interplay among hormonal, neuronal, and immune systems. Our habitually high salt intake promotes abnormal interactions and causes hypertension. This improved understanding may help us to develop therapeutic and lifestyle interventions to tame the “silent killer.” Disclosure All the authors declared no conflict of interest. Acknowledgments The authors thank The British Heart Foundation and Kidney Research UK for research funding.

Introduction Antineutrophil cytoplasmic antibody (ANCA)−associated vasculitis is a multisystem autoimmune disease that often involves the lungs. It can be serious and sometimes fatal, and requires prompt recognition and treatment. Cyclophosphamide is a well-established alkylating agent that is widely used in the treatment of ANCA vasculitis. The most common side effects of cyclophosphamide are infection and the risk of malignancy.

that often involves the lungs. It can be serious and sometimes fatal, and requires prompt recognition and treatment. Cyclophosphamide is a well-established alkylating agent that is widely used in the treatment of ANCA vasculitis. The most common side effects of cyclophosphamide are infection and the risk of malignancy. Case Presentation A 61-year-old man, a lifelong smoker, was investigated in the respiratory outpatient clinic for progressive breathlessness. Pulmonary function tests at the time revealed a forced expiratory volume in 1 second (FEV1) of 2.84 L (predicted 3.2) and a forced vital capacity (FVC) of 4.9 L (predicted 5.1) with an FEV1/VC of 0.58. Based on these values, he was diagnosed with chronic obstructive pulmonary disease. His breathlessness continued to worsen and was associated with reduced exercise tolerance. As a result he underwent a high-resolution computed tomography (HRCT) scan of the lungs 2 months later. This showed extensive emphysema and bi-basal peripheral–ground-glass changes with possible honeycomb cyst formation (Figure 1a). Serum creatinine was normal at this time. The patient re-presented 2 months later with symptoms of lethargy, worsening breathlessness, and numbness affecting his left foot.Figure 1 (a) Initial computed tomogram with mild changes, notably extensive emphysema, and bi-basal peripheral–ground-glass changes. (b) Computed tomogram from 4 months later (after initiation of cyclophosphamide), with severe centrilobar and paraseptal emphysema with new inflammatory changes. (c) Improvement in the changes seen in (b), 3 months after cyclophosphamide withdrawal.

bly extensive emphysema, and bi-basal peripheral–ground-glass changes. (b) Computed tomogram from 4 months later (after initiation of cyclophosphamide), with severe centrilobar and paraseptal emphysema with new inflammatory changes. (c) Improvement in the changes seen in (b), 3 months after cyclophosphamide withdrawal. Physical examination revealed bi-basal fine expiratory crackles in the lungs alongside a mononeuritis multiplex. Renal function was severely impaired with a serum creatinine of 13.01 mg/dl (normal range 0.60–1.10) and C-reactive protein (CRP) was elevated at 93 mg/dl (normal range 0–5). Myeloperoxidase antineutrophil cytoplasmic antibody (ANCA) titers were raised at >100 IU/ml (normal range 0–5). The patient went on to have a renal biopsy. This showed an active segmental and necrotizing glomerulonephritis with evidence of significant tubular atrophy and interstitial fibrosis. Overall, the clinical diagnosis was of an ANCA-associated systemic vasculitis, most in keeping with microscopic polyangiitis. Despite the significant chronic renal damage, the patient was treated with a combination of prednisone (1 mg/kg/d), plasmapheresis, and i.v. cyclophosphamide in addition to hemodialysis. His clinical condition improved significantly, in particular his shortness of breath. However, he remained dialysis dependent. He was discharged home with a plan to continue outpatient treatment with i.v. cyclophosphamide alongside a reduction of his prednisone dose. A repeat HRCT before discharge showed no new changes compared to the HRCT from a few months earlier.

n particular his shortness of breath. However, he remained dialysis dependent. He was discharged home with a plan to continue outpatient treatment with i.v. cyclophosphamide alongside a reduction of his prednisone dose. A repeat HRCT before discharge showed no new changes compared to the HRCT from a few months earlier. The patient re-presented 6 weeks later with marked dyspnea. Despite 14.5 L of fluid removal on hemodialysis and treatment with broad-spectrum antibiotics, he remained dyspneic. Results of sputum and blood cultures as well as respiratory viral swabs were negative. Results of investigations for Pneumocystis jirovecii pneumonia were also negative. Diffuse alveolar hemorrhage was considered, although imaging was not consistent with this, hemoglobin was stable, and there was no externalization of blood. The patient was not fit to undergo bronchoscopy. An HRCT scan of the lungs was reported as showing severe centrilobar and paraseptal emphysema with super-added inflammatory changes (Figure 1b), significantly worse compared to a scan 2 months earlier. The patient had received 2 doses of i.v. cyclophosphamide, and it was believed that this was implicated. Following its withdrawal, the patient’s symptoms and exercise tolerance gradually improved. An HRCT 3 months later showed resolution of the lung changes seen on CT immediately before cyclophosphamide cessation (Figure 1c). The patient was diagnosed with cyclophosphamide-induced acute pneumonitis. His prednisone dose was increased transiently following withdrawal of cyclophosphamide.

erance gradually improved. An HRCT 3 months later showed resolution of the lung changes seen on CT immediately before cyclophosphamide cessation (Figure 1c). The patient was diagnosed with cyclophosphamide-induced acute pneumonitis. His prednisone dose was increased transiently following withdrawal of cyclophosphamide. Discussion Cyclophosphamide is an established alkylating agent that is widely used in the treatment of ANCA-associated vasculitis and in hematological malignancies as part of a chemotherapy regimen. Its most commonly recognized side effects are those of infection and the long-term risk of malignancy. Pulmonary side effects are rare (<1%) and are dose related.1 They manifest as either an early-onset pneumonitis, with patients presenting with cough and dyspnea within 6 months of starting treatment, or as a late fibrosis with gradual worsening dyspnea and a nonproductive cough (Table 1).2Table 1 Teaching points related to the case

e effects are rare (<1%) and are dose related.1 They manifest as either an early-onset pneumonitis, with patients presenting with cough and dyspnea within 6 months of starting treatment, or as a late fibrosis with gradual worsening dyspnea and a nonproductive cough (Table 1).2Table 1 Teaching points related to the case 1. Cyclophosphamide, a commonly used alkylating agent for the treatment of ANCA-associated vasculitis, can cause direct lung toxicity 2. Lung damage can manifest as early-onset pneumonitis (likely reversible) or late-onset lung fibrosis (usually irreversible) 3. When considering patients undergoing hemodialysis who present with symptoms and signs of cyclophosphamide-associated lung injury, it is important to first exclude pulmonary infection and edema 4. In patients with a pneumonitis-type picture, as presented here, withdrawal of cyclophosphamide can lead to resolution of symptoms and radiographic evidence of disease ANCA, antineutrophil cytoplasmic antibody.

of cyclophosphamide-associated lung injury, it is important to first exclude pulmonary infection and edema 4. In patients with a pneumonitis-type picture, as presented here, withdrawal of cyclophosphamide can lead to resolution of symptoms and radiographic evidence of disease ANCA, antineutrophil cytoplasmic antibody. The diagnosis of cyclophosphamide-related lung injury is made clinically on the basis of symptoms, history of cyclophosphamide use, compatible findings on chest imaging studies, and the absence of an alternative diagnosis. Early drug toxicity may respond to discontinuation of cyclophosphamide alongside glucocorticoids, whereas late toxicity is usually untreatable. It is unclear whether a higher total dose of cyclophosphamide is a risk factor for lung toxicity, as affected patients have received doses ranging from 150 mg to 81 g.2, 3 As demonstrated here, HRCT may be a better imaging modality than plain chest X-ray for detecting these changes (serial chest X-rays for the patient described here are shown in Figure 2).Figure 2 (a) X-ray taken before cyclophosphamide treatment. (b) X-ray taken during treatment. (c) X-ray taken 2 months after cyclophosphamide was discontinued. The pretreatment X-ray was reported as showing a largely peripheral and basal pattern of fibrosis, which then improved significantly following drug withdrawal.

.Figure 2 (a) X-ray taken before cyclophosphamide treatment. (b) X-ray taken during treatment. (c) X-ray taken 2 months after cyclophosphamide was discontinued. The pretreatment X-ray was reported as showing a largely peripheral and basal pattern of fibrosis, which then improved significantly following drug withdrawal. The importance of recognizing the pulmonary toxicity associated with cyclophosphamide is underscored by the fact that the prevalence rates of ANCA-associated vasculitis, for which cyclophosphamide is the main therapeutic agent, are increasing, as are the global rates of cancer, another indication for treatment with this agent.4, 5 Furthermore, the sales of cyclophosphamide in the United States alone for the year ending September 2014 amounted to ∼$420 million according to IMS Health. There have been few data published over the past 20 years detailing the important pulmonary toxicity associated with the drug described here. When considering cyclophosphamide-related lung injury in those patients receiving maintenance hemodialysis, fluid overload should first be excluded. Patient Perspective Diagnosis and Initial Treatment “I remember feeling constantly tired and exhausted… I could hardly walk the length of my front room without feeling breathless. The treatment worked really well for my breathing… But of course, what was really beginning to worry me at that time was being started on dialysis…”

Perspective Diagnosis and Initial Treatment “I remember feeling constantly tired and exhausted… I could hardly walk the length of my front room without feeling breathless. The treatment worked really well for my breathing… But of course, what was really beginning to worry me at that time was being started on dialysis…” Re-admission “Life was much more difficult after I made it home the first time… The diagnosis began to sink in, I still had to travel to the hospital three times a week for dialysis, and I came to rely on those around me for a lot of things that I could do myself before, like shopping and cooking… After a while, though, I noticed I was becoming more short of breath and struggling whenever I left the house… It got to the stage where I could hardly speak without feeling short of breath. That’s when I decided enough was enough and I needed help again. Things have taken a lot longer to get better this time, but it still feels like things are going in the right direction…” Disclosure All the authors declared no competing interests. Acknowledgments ND is supported by a British Heart Foundation Intermediate Clinical Research Fellowship (FS/13/30/29994). TEF is funded by an MRC Clinical Fellowship (MR/R017840/1). Authorship All authors contributed equally to the case report.

Nonadherence is common in both HD patients and kidney transplant recipients. Hemodialysis regimens are complex and demanding, necessitating attendance at HD sessions, adherence to prescribed medications, and fluid and dietary restrictions.1, 2 Poor adherence can lead to poor clinical outcomes and increased risk of mortality,3, 4 as well as adding to health care costs.5, 6 There is a lack of consensus on definitions, which contributes to widely varying nonadherence rates in the HD population. Rates of skipping HD sessions vary between 0% and 32.3%, medication nonadherence between 1.2% and 81%, fluid restriction nonadherence from 3.4% to 74%, and nonadherence to dietary restrictions from 1.2% to 82.4%.2 There have been some attempts to achieve consensus. A United States Renal Data System (USRDS) study3 defined nonadherence in 4 ways: (i) skipping HD sessions; (ii) shortening HD sessions by 10 minutes or more; (iii) an interdialytic weight gain of more than 5.7% of patient dry weight; and (iv) serum phosphate of greater than 7.5 mg/dl (2.42 mmol/l). Using these definitions, the highest rates of nonadherence were found for shortening HD sessions (20.3%) and serum phosphate (22.1%). Applying the same criteria to data from the Dialysis Outcomes and Practice Patterns Study (DOPPS), levels of nonadherence across these 4 areas were 3.8%–19.6% for the overall international sample and 0.6%–20% for the European sample,4 and from 0.6%–23.8% for the overall European sample and 0.8%–21.9% for the UK sample.7

pplying the same criteria to data from the Dialysis Outcomes and Practice Patterns Study (DOPPS), levels of nonadherence across these 4 areas were 3.8%–19.6% for the overall international sample and 0.6%–20% for the European sample,4 and from 0.6%–23.8% for the overall European sample and 0.8%–21.9% for the UK sample.7 Compared with HD patients, kidney transplant recipients have improved quality of life, a less restrictive diet, longer life expectancy,8 and fewer psychological symptoms such as depression.9 Nonetheless, these patients are also required to make adjustments in their lifestyle such as adherence to immunosuppressants to prevent rejection, alongside attending clinic for regular check-ups and generally maintaining a good diet and activity level.10 The length of time that a patient waits for a kidney transplant varies in the UK across transplanting centers; however, the average wait is 2.5 to 3 years.11 For adult patients registered for a deceased donor transplant from April 2011 to March 2014, the median waiting time was 829 days and the median time from start of dialysis to kidney transplant from April 2016 to March 2017 was 1148 days.12

UK across transplanting centers; however, the average wait is 2.5 to 3 years.11 For adult patients registered for a deceased donor transplant from April 2011 to March 2014, the median waiting time was 829 days and the median time from start of dialysis to kidney transplant from April 2016 to March 2017 was 1148 days.12 Nonadherence is a major risk factor for poor outcomes including graft survival.13 Reported rates of nonadherence to immunosuppressants in transplant patients vary, and again, definitions and methods of assessment are inconsistent. Methods such as self-report, electronic monitoring, reports from family, or health care professional observations are most common.14 Use of clinical data to assess nonadherence in this setting is less frequently used. A review of nonadherence to immunosuppressants14 reported nonadherence rates ranging from 2% to 67%, depending on definitions and methods deployed, with an average rate of 28% when adherence was measured by self-report.

most common.14 Use of clinical data to assess nonadherence in this setting is less frequently used. A review of nonadherence to immunosuppressants14 reported nonadherence rates ranging from 2% to 67%, depending on definitions and methods deployed, with an average rate of 28% when adherence was measured by self-report. It is important to consider how adherence behavior may transfer from 1 modality to another. For example, if poor adherence in HD patients pretransplantion could be identified and addressed, could posttransplantation adherence be improved and, with it, the risks of graft loss? Douglas et al.,15 conducted a longitudinal retrospective chart audit to examine this relationship, specifically examining pretransplantation adherence and posttransplantation outcomes in 126 renal transplant recipients. They defined nonadherence as having at least 1 chart note indicating pre- or posttransplantation nonadherence with a therapeutic regimen. Findings showed that 61% of patients identified as nonadherent before transplantation experienced graft loss or died. This research indicated a potential relationship between pretransplantation adherence and posttransplantation outcomes; however, the method of defining nonadherence was not particularly stringent.

imen. Findings showed that 61% of patients identified as nonadherent before transplantation experienced graft loss or died. This research indicated a potential relationship between pretransplantation adherence and posttransplantation outcomes; however, the method of defining nonadherence was not particularly stringent. More recently, Dobbels et al.16 prospectively followed 141 heart, liver and lung transplant recipients, examining pretransplantation predictors of posttransplantation outcomes. Independent predictors of nonadherence to immunosuppressants 1 year posttransplantation were pretransplantation nonadherence to taking medication, having less social support with medication taking, having higher education status, and having lower scores for the personality trait “conscientiousness.” A meta-analysis of 122 studies reporting associations between social support and patient adherence also suggested that poor social support is a key determinant of nonadherence to medical treatment regimens.17 In addition, pretransplant medication adherence was found to be the only predictor of late acute rejection. Although this research was not conducted in the kidney transplant population, it signals the importance of attending to pretransplantation behavioral patterns in predicting posttransplantation outcomes.

gimens.17 In addition, pretransplant medication adherence was found to be the only predictor of late acute rejection. Although this research was not conducted in the kidney transplant population, it signals the importance of attending to pretransplantation behavioral patterns in predicting posttransplantation outcomes. To our knowledge, there is little previous literature examining the relationship between clinical measures of pretransplantation adherence to HD and posttransplantation adherence in the renal transplant population. Although previous literature has highlighted predictors of nonadherence to HD and posttransplantation adherence separately, there is little exploration of whether there is a relationship that could help clinicians to identify aspects of pretransplantation nonadherence that act as potential risk factors for posttransplantation nonadherence. In addition, this could highlight patients who need to be targeted for intervention to address adherence concerns before transplantation to modify adherence behavior posttransplantation. It is clear from the literature that nonadherence both pretransplantation when on HD, and posttransplantation, are major risk factors for poor clinical outcomes and hence important to address. This retrospective study addresses whether nonadherent HD patients become nonadherent transplant recipients. It also considers whether there are particular patterns of nonadherence to HD that are more likely to associate with poor adherence after transplantation.

r poor clinical outcomes and hence important to address. This retrospective study addresses whether nonadherent HD patients become nonadherent transplant recipients. It also considers whether there are particular patterns of nonadherence to HD that are more likely to associate with poor adherence after transplantation. Methods This was a retrospective study carried out in a subregional renal unit. Data were collected from the electronic patient system about adherence to HD regimens in the 6 months before transplantation, and for 1 year posttransplantation after return transfer to the posttransplantation clinic from the transplanting center. Participants The study population consisted of 88 adult (aged 18 years and more) kidney transplant recipients. Patients were eligible for inclusion if they had (i) received their transplant between 2006 and 2016, and (ii) had data available for a minimum of 6 months of HD before transplantation. We also included only those patients who were prescribed tacrolimus as their posttransplantation immunosuppressant, as this was the most common immunosuppressant across patients. Exclusion criteria included patients who were transferred later than 1 year posttransplantation back to the transplant clinic from the transplanting center. Further exclusions included patients who received HD for less than 6 months before transplantation, and patients who received home HD or peritoneal dialysis in the 6-month period before transplantation.

o were transferred later than 1 year posttransplantation back to the transplant clinic from the transplanting center. Further exclusions included patients who received HD for less than 6 months before transplantation, and patients who received home HD or peritoneal dialysis in the 6-month period before transplantation. Of a sample of 204 kidney transplant patients who had received HD as a treatment at some point before transplantation, 88 were eligible for inclusion in the study analysis. The remaining 116 patients were excluded. Figure 1 includes details of patient exclusions.Figure 1 Flow diagram of participant study inclusion and exclusion.

e of 204 kidney transplant patients who had received HD as a treatment at some point before transplantation, 88 were eligible for inclusion in the study analysis. The remaining 116 patients were excluded. Figure 1 includes details of patient exclusions.Figure 1 Flow diagram of participant study inclusion and exclusion. Data Retrieved Demographic data retrieved included age, sex, ethnicity, age at first dialysis session, age at transplantation, and Index of Multiple Deprivation (IMD) score from patient postcodes. Clinical data collected as part of routine care were retrieved that could provide indicators of nonadherence. Pretransplantation measures included the followiing: variance from dialysis prescription (in minutes of session length), missed number of dialysis sessions, dialysis vintage, residual kidney function (KRU), serum phosphate levels, parathyroid hormone (PTH), and interdialytic weight gain (IDWG). Missed dialysis due to hospitalization was not included as missed sessions. Means were calculated for the 6-month period before transplantation for pretransplantation measures, using measurements recorded as part of routine medical care (with the exception of missed dialysis sessions, which was a count). Posttransplantation measures included tacrolimus levels and their SD, number of missed clinic appointments, and donor type.

onth period before transplantation for pretransplantation measures, using measurements recorded as part of routine medical care (with the exception of missed dialysis sessions, which was a count). Posttransplantation measures included tacrolimus levels and their SD, number of missed clinic appointments, and donor type. Defining Adherence There is no universally agreed way of defining adherence pre- and posttransplantation; therefore, based on previous literature3, 4, 7, 18 and clinical expected ranges, different cut-off points were applied to assess the data to explore potential relationships between pre- and posttransplantation adherence. Pretransplantation definitions of nonadherence included patients: (i) on average, shortening their dialysis prescription by more than 10 minutes; (ii) on average, shortening their dialysis prescription by more than 15 minutes; (iii) missed 2 or more HD sessions; and (iv) had a mean serum phosphate level of 1.8 mmol/l or more. Posttransplantation definitions of nonadherence included: (i) mean tacrolimus levels outside of the expected range within the first 2 years of 5 to 10 ng/ml after transplantation; and (ii) missing 1 or more posttransplantation clinic appointments.

Pretransplantation definitions of nonadherence included patients: (i) on average, shortening their dialysis prescription by more than 10 minutes; (ii) on average, shortening their dialysis prescription by more than 15 minutes; (iii) missed 2 or more HD sessions; and (iv) had a mean serum phosphate level of 1.8 mmol/l or more. Posttransplantation definitions of nonadherence included: (i) mean tacrolimus levels outside of the expected range within the first 2 years of 5 to 10 ng/ml after transplantation; and (ii) missing 1 or more posttransplantation clinic appointments. For this study, a 6-month period was used to assess nonadherence pretransplantation, whereas previous studies3, 4, 7 defined nonadherence in a narrower time frame of 1 month. In addition to using shortening dialysis prescription by more than 10 minutes as used in previous research as a definition of nonadherence, more than 15 minutes was also used as a cut-off point, as this represented the top 25% of variance in dialysis prescription times for patients in this study. Previous studies used a higher nonadherence cut-off point for serum phosphate levels of 7.5 mg/dl (∼2.42 mmol/l). We chose to use a cut-off point in line with previous research using serum phosphate as a measure of nonadherence19 and in line with the recommended Renal Association serum phosphate levels of 1.1 to 1.7 mmol/l in the United Kingdom. Hence we use serum phosphate levels of 1.8 mmol/l or more to define nonadherence.20 Posttransplantation tacrolimus therapeutic range of 5 to 10 ng/ml was determined on the basis of clinical advice and previous research that used this range.18

ssociation serum phosphate levels of 1.1 to 1.7 mmol/l in the United Kingdom. Hence we use serum phosphate levels of 1.8 mmol/l or more to define nonadherence.20 Posttransplantation tacrolimus therapeutic range of 5 to 10 ng/ml was determined on the basis of clinical advice and previous research that used this range.18 Statistical Analysis As different markers were used to measure adherence pretransplantation compared to post- transplantation, a narrative comparison of the data is reported. Demographic data are reported for the study sample using means and frequencies. Comparisons of the data were completed using the McNemar test and Cochran Q for categorical data and t tests for continuous data. All tests were 2-tailed, and P values of less than 0.05 were considered to be significant. Logistic regressions were used to determine potential predictors of nonadherence both pre- and posttransplantation. The McNemar test was used to explore relationships between pre- and posttransplantation adherence. The data were analysed using SPSS version 25 (IBM SPSS, Armonk, NY). Approvals This study was considered by the institutional review team at East and North Hertfordshire NHS Trust (RD2016-82) and was determined to be a service evaluation. Departmental agreement was provided for the service evaluation to be completed.

Statistical Analysis As different markers were used to measure adherence pretransplantation compared to post- transplantation, a narrative comparison of the data is reported. Demographic data are reported for the study sample using means and frequencies. Comparisons of the data were completed using the McNemar test and Cochran Q for categorical data and t tests for continuous data. All tests were 2-tailed, and P values of less than 0.05 were considered to be significant. Logistic regressions were used to determine potential predictors of nonadherence both pre- and posttransplantation. The McNemar test was used to explore relationships between pre- and posttransplantation adherence. The data were analysed using SPSS version 25 (IBM SPSS, Armonk, NY). Approvals This study was considered by the institutional review team at East and North Hertfordshire NHS Trust (RD2016-82) and was determined to be a service evaluation. Departmental agreement was provided for the service evaluation to be completed. Results Patient Characteristics Of the 88 patients, 62.5% were male and 37.5% were female. Mean age at transplantation for the overall sample was 48.5 years (SD = 12.7), with no significant difference between sexes. The majority of patients (54.5%) were from white ethnic backgrounds, although a considerable proportion (45.5%) were also from other ethnic groups (Table 1).Table 1 Demographic comparison of adherent and nonadherent patients pretransplantation

le was 48.5 years (SD = 12.7), with no significant difference between sexes. The majority of patients (54.5%) were from white ethnic backgrounds, although a considerable proportion (45.5%) were also from other ethnic groups (Table 1).Table 1 Demographic comparison of adherent and nonadherent patients pretransplantation n = 88 (%) Pretransplantation shortening dialysis prescription >10 min Pretransplantation shortening dialysis prescription >15 min Pretransplantation missed dialysis sessions ≥2 Phosphate ≥1.8 mmol/l Adherent Nonadherent Adherent Nonadherent Adherent Nonadherent Adherent Nonadherent n = 53 (60.2%) n = 35 (39.8%) n = 66 (75%) n = 22 (25%) n = 64 (72.7%) n = 24 (27.3%) n = 51 (58%) n = 37 (42%) Age at transplant, mean (SD) 48.5 (12.7) 49.2 (13.4) 47.5 (11.6) 49.1 (13.1) 46.9 (11.4) 48.4 (12.8) 48.9 (12.6) 50.8 (11.7) 45.4 (13.4)a Age at first dialysis, mean (SD) 44.9 (13.0) 45.3 (13.7) 44.3 (12.0) 45.1 (13.6) 44.2 (11.2) 44.7 (13.3) 45.5 (12.4) 47.2 (12.3) 41.7 (13.4) Sex, n (%) Male 55 (62.5) 32 (60.4) 23 (65.7) 41 (62.1) 14 (63.6) 39 (60.9) 16 (66.7) 31 (60.8) 24 (64.9) Female 33 (37.5) 21 (39.6) 12 (34.3) 25 (37.9) 8 (36.4) 25 (39.1) 8 (33.3) 20 (39.2) 13 (35.1) Ethnicity, n (%) White 48 (54.5) 29 (54.7) 19 (54.3) 39 (59.1) 9 (40.9) 32 (50) 16 (66.7) 28 (54.9) 20 (54.1) Nonwhite 40 (45.5) 24 (45.3) 16 (45.7) 27 (40.9) 13 (59.1) 32 (50) 8 (33.3) 23 (45.1) 17 (45.9) Index of Multiple Deprivation, mean (SD) 5.5 (2.9) 5.3 (3.0) 5.9 (2.8) 5.6 (3.0) 5.2 (2.7) 5.3 (3.1) 5.9 (2.6) 6.0 (3.0) 4.8 (2.7) Dialysis Dialysis vintage, mo Median (IQR) 26 (16, 49) 26 (16.3, 54) 26 (15, 40) 26 (17, 51) 24 (13, 39) 25 (16, 48) 27 (16, 50) 28 (15, 55) 24 (17, 40) KRU median (IQR) 0.42 (0.01, 2.2) 0.30 (0.01, 1.2) 1.28 (0.07, 2.7)a 0.41 (0.01, 1.9) 0.64 (0.01, 2.6) 0.39 (0.01, 2.2) 0.81 (0.01, 2.3) 1.1 (0.01, 2.8) 0.10 (0.01, 0.86)a PTH median (IQR) 37 (25, 57) 35 (22.2, 57) 43 (25, 59) 36 (25, 57) 43 (25, 60) 38 (26, 57) 30 (14, 70) 30 (17, 51) 50 (34, 72)a IDWG median (IQR) 1.79 (1.24, 2.4) 1.8 (1.4, 2.4) 1.8 (1.1, 2.4) 1.8 (1.2, 2.4) 1.8 (1.1, 2.5) 1.8 (1.2, 2.4) 1.9 (1.3, 2.5) 1.7 (.94, 2.4) 1.8 (1.6, 2.5) IDWG, interdialytic weight gain (kg); IQR, interquartile range; KRU, urea clearance (ml/min); PTH, parathyroid hormone (pmol/l).

30 (14, 70) 30 (17, 51) 50 (34, 72)a IDWG median (IQR) 1.79 (1.24, 2.4) 1.8 (1.4, 2.4) 1.8 (1.1, 2.4) 1.8 (1.2, 2.4) 1.8 (1.1, 2.5) 1.8 (1.2, 2.4) 1.9 (1.3, 2.5) 1.7 (.94, 2.4) 1.8 (1.6, 2.5) IDWG, interdialytic weight gain (kg); IQR, interquartile range; KRU, urea clearance (ml/min); PTH, parathyroid hormone (pmol/l). a P < 0.05. Indices of deprivation rank every postcode in England from 1 (most deprived) to 32, 844 (least deprived). These are split into deciles of 1 to 10 from most deprived to least deprived, dividing them into 10 equal groups, ranging from 1 = from the most deprived 10%, to 10 = the least deprived 10%. Across the overall sample, 23.9% (n = 21) of patients lived in neighborhoods that fall in the 20% most deprived small areas in England. Findings differed across ethnicity, with 40% (n = 16) of nonwhite patients shown to live in neighborhoods that fall in the 20% most deprived small areas in England as compared to only 10.4% (n = 5) of white patients. Chronic glomerulonephritis (21.6%: n = 19), diabetic nephropathy (20.5%: n = 18), polycystic kidney disease (13.6%: n = 12), chronic pyelonephritis (9.1%, n = 8), and hypertension (4.5%: n = 4) accounted for the majority of cases. Etiology was uncertain for more than a quarter of the sample (28.4%, n = 25).

e patients. Chronic glomerulonephritis (21.6%: n = 19), diabetic nephropathy (20.5%: n = 18), polycystic kidney disease (13.6%: n = 12), chronic pyelonephritis (9.1%, n = 8), and hypertension (4.5%: n = 4) accounted for the majority of cases. Etiology was uncertain for more than a quarter of the sample (28.4%, n = 25). Clinical Pretransplantation Data Nonadherence ranged from 25% to 42%, depending on how it was operationalized. There were no significant demographic differences between groups across measures, with the exception being that patients categorized as nonadherent based on phosphate levels were significantly younger at transplantation (t[86] = 1.99, P = 0.049) than those categorized as adherent (Table 1). Cochran Q was used to determine whether there were any differences in patients identified as nonadherent across the 4 pretransplantation measures. There was a statistically significant difference in the proportion of nonadherent patients across the 4 nonadherence measures (χ2[3] = 9.79, P = 0.020).

adherent (Table 1). Cochran Q was used to determine whether there were any differences in patients identified as nonadherent across the 4 pretransplantation measures. There was a statistically significant difference in the proportion of nonadherent patients across the 4 nonadherence measures (χ2[3] = 9.79, P = 0.020). We explored whether pretransplantation clinical data predicted pretransplantation adherence. Dialysis vintage, KRU, serum phosphate, PTH, and IDWG were compared independently across adherent and nonadherent patients using the 4 pretransplantation adherence measures. All measures were skewed with the exception of serum phosphate. There was a significant difference for KRU between adherent and nonadherent patients, when adherence was defined as shortening dialysis by more than 10 minutes (U = 684.5, P = 0.035), with adherent patients having lower residual kidney function than nonadherent patients (Table 1). No other significant differences were observed between adherent and nonadherent patients when adherence was defined as shortening dialysis by more than 10 minutes or by more than 15 minutes. Significant differences were observed when adherence was defined using serum phosphate levels. Patients who were categorized as nonadherent based on their phosphate levels had lower KRU (U = 648, P = 0.011), and higher parathyroid hormone levels (U = 545.5, P = 0.002) (Table 1).

n 10 minutes or by more than 15 minutes. Significant differences were observed when adherence was defined using serum phosphate levels. Patients who were categorized as nonadherent based on their phosphate levels had lower KRU (U = 648, P = 0.011), and higher parathyroid hormone levels (U = 545.5, P = 0.002) (Table 1). Logistic regression analyses were conducted to identify possible predictors of nonadherence among pretransplantation patients. Factors included in the models were age at transplantation, sex, ethnicity, Index of Multiple Deprivation score, and dialysis vintage. No significant predictors of nonadherence were identified for any of the nonadherence measures. Clinical Posttransplantation Data Patients with mean tacrolimus levels outside the range expected within the first 2 years of 5 to 10 ng/l were highlighted. Of the 88 patients, 14 (15.9%) had tacrolimus levels outside the expected range of 5 to 10 ng/l. Ten patients were male and 4 were female. There were equal numbers of white and nonwhite patients (n = 7). There were no significant demographic or clinical differences between adherent and nonadherent patients defined in this way.

8 patients, 14 (15.9%) had tacrolimus levels outside the expected range of 5 to 10 ng/l. Ten patients were male and 4 were female. There were equal numbers of white and nonwhite patients (n = 7). There were no significant demographic or clinical differences between adherent and nonadherent patients defined in this way. When nonadherence was defined using the number of missed posttransplantation clinic appointments as 1 or more, 20 patients (22.7%) were identified as nonadherent. No significant demographic differences were observed between groups when posttransplantation adherence was defined in this way, except that nonadherent patients were significantly younger when they underwent transplantation (t[86] = 2.14, P = 0.035) and significantly younger when starting dialysis (t[86] = 2.07, P = .041), than those categorized as adherent (Table 2).Table 2 Demographic comparison of adherent and nonadherent patients posttransplantation

at nonadherent patients were significantly younger when they underwent transplantation (t[86] = 2.14, P = 0.035) and significantly younger when starting dialysis (t[86] = 2.07, P = .041), than those categorized as adherent (Table 2).Table 2 Demographic comparison of adherent and nonadherent patients posttransplantation Posttransplantation tacrolimus levels Posttransplantation missed clinic appointments ≥1 Adherent Nonadherent Adherent Nonadherent n = 74 (84.1%) n = 14 (15.9%) n = 68 (77.3%) n = 20 (22.7%) Age at transplant, mean (SD) 48.8 (12.9) 46.8 (11.7) 50.0 (12.1) 43.3 (13.4)a Age at first dialysis, mean (SD) 45.1 (13.3) 43.50 (11.7) 46.4 (12.2) 39.7 (14.6)a Dialysis vintage, mo, mean (SD) 35.9 (27.6) 29.8 (16.7) 35.4 (26.4) 33.1 (26.2) Gender, n (%) Male 45 (60.8) 10 (71.4) 42 (61.8) 13 (65.0) Female 29 (39.2) 4 (28.6) 26 (38.2) 7 (35.0) Ethnicity, n (%) White 41 (55.4) 7 (50) 37 (54.4) 11 (55.0) Nonwhite 33 (44.6) 7 (50) 31 (45.6) 9 (45.0) Index of Multiple Deprivation, mean (SD) 5.7 (2.9) 4.6 (2.9) 5.69 (3.1) 4.9 (2.4) a P < 0.05. Logistic regression analyses were conducted to identify possible predictors of nonadherence among posttransplantation patients in this study. No significant predictors for nonadherence to tacrolimus immunosuppressant medication were identified. Phosphate levels of 1.8mmol/l or more pretransplantation were identified as predicting higher odds of nonattendance at posttransplantation clinic appointments (Table 3).Table 3 Predictors of nonadherence posttransplantation

nificant predictors for nonadherence to tacrolimus immunosuppressant medication were identified. Phosphate levels of 1.8mmol/l or more pretransplantation were identified as predicting higher odds of nonattendance at posttransplantation clinic appointments (Table 3).Table 3 Predictors of nonadherence posttransplantation Odds ratio (95% confidence interval) by nonadherence measure Posttransplantation tacrolimus levels Posttransplantation missed clinic appointments ≥1 Age at transplant, yr 0.99 (0.93, 1.05) 0.98 (0.94, 1.04) Male vs. female 2.09 (0.49, 8.88) 0.80 (0.23, 2.86) White vs. nonwhite 0.89 (0.23, 3.46) 1.40 (0.34, 5.82) Index of Multiple Deprivation 0.85 (0.66, 1.09) 0.89 (0.70, 1.14) Dialysis vintage, mo 0.99 (0.96, 1.02) 1.00 (0.97, 1.02) Variance in dialysis time from prescription, min 1.01 (0.97, 1.04) 1.05 (0.99, 1.11) Missed dialysis sessions >2 vs. <2 0.86 (0.19, 3.84) 0.60 (0.14, 2.58) Phosphate ≥1.8 mmol/l vs. <1.8 mmol/l 0.69 (0.17, 2.80) 4.19 (1.15, 15.24)a Donor type deceased vs. living 0.28 (0.06, 1.19) 2.42 (0.41, 14.29) a P < 0.05.

7, 1.02) Variance in dialysis time from prescription, min 1.01 (0.97, 1.04) 1.05 (0.99, 1.11) Missed dialysis sessions >2 vs. <2 0.86 (0.19, 3.84) 0.60 (0.14, 2.58) Phosphate ≥1.8 mmol/l vs. <1.8 mmol/l 0.69 (0.17, 2.80) 4.19 (1.15, 15.24)a Donor type deceased vs. living 0.28 (0.06, 1.19) 2.42 (0.41, 14.29) a P < 0.05. In addition to looking at tacrolimus using the mean levels, the SD and coefficient of variation (CV) was also calculated to examine the variation in tacrolimus levels for each patient for the 1-year period recorded posttransplantation. A nonadherence cut-off point of SD of greater than 2.0 was used in line with previous research,21 and a tacrolimus CV% cut-off point of 41% was used, again in line with previous research.22 Logistic regression analyses were conducted to identify potential predictors of these parameters. No significant predictors were identified. There was no difference between the proportion of nonadherent patients defined in terms of highly variable tacrolimus levels (tacrolimus CV% > 41%) and defined in terms of missed clinic appointments (P = 0.47 by McNemar test).

identify potential predictors of these parameters. No significant predictors were identified. There was no difference between the proportion of nonadherent patients defined in terms of highly variable tacrolimus levels (tacrolimus CV% > 41%) and defined in terms of missed clinic appointments (P = 0.47 by McNemar test). Comparing Pre- and Posttransplantation Adherence In general, the prevalence of nonadherence was greater pretransplantation than posttransplantation. The prevalence of pretransplantation nonadherence defined by shortened dialysis by more than 10 minutes was greater than posttransplantation nonadherence adherence defined by both tacrolimus levels and by missed posttransplantation clinic appointments (P = 0.001 and 0.029, respectively; McNemar test). Likewise, the prevalence of pretransplantation nonadherence determined by phosphate levels was greater than the prevalence of posttransplantation nonadherence determined by both tacrolimus levels (P < 0.001) and by missed posttransplantation clinic appointments (P = 0.003). No other significant differences were found when comparing pre- and posttransplantation groups (Table 4).Table 4 Comparing pretransplantation adherence to posttransplantation adherence measures

nadherence determined by both tacrolimus levels (P < 0.001) and by missed posttransplantation clinic appointments (P = 0.003). No other significant differences were found when comparing pre- and posttransplantation groups (Table 4).Table 4 Comparing pretransplantation adherence to posttransplantation adherence measures Pretransplantation shortening dialysis prescription >10 min P Pretransplantation shortening dialysis prescription >15 min P Pretransplantation missed dialysis sessions ≥2 P Phosphate ≥1.8 mmol/l P Adherent Nonadherent Adherent Nonadherent Adherent Nonadherent Adherent Nonadherent n = 53 (60.2%) n = 35 (39.8%) n = 66 (75%) n = 22 (25%) n = 64 (72.7%) n = 24 (27.3%) n = 51 (58%) n = 37 (42%) Posttransplantation determined by tacrolimus levels Adherent, n = 74 45 29 .001a 55 19 0.20 53 21 0.11 43 31 <.001a Nonadherent, n = 14 8 6 11 3 11 3 8 6 Posttransplantation missed clinic appointments ≥1 Adherent, n = 68 40 28 0.029a 48 20 0.87 49 19 0.61 45 23 0.003a Nonadherent, n = 20 13 7 18 2 15 5 6 14 a P < 0.05.

Posttransplantation determined by tacrolimus levels Adherent, n = 74 45 29 .001a 55 19 0.20 53 21 0.11 43 31 <.001a Nonadherent, n = 14 8 6 11 3 11 3 8 6 Posttransplantation missed clinic appointments ≥1 Adherent, n = 68 40 28 0.029a 48 20 0.87 49 19 0.61 45 23 0.003a Nonadherent, n = 20 13 7 18 2 15 5 6 14 a P < 0.05. We explored the relationship between pretransplantation demographic and clinical data to posttransplantation adherence. Of the 28 patients categorized as nonadherent to either one (n = 22) or both (n = 6) posttransplantation measures, 46.4% (n = 13) were nonadherent to 2 or more pretransplantation measures, compared with 32.2% (n = 9) who were nonadherent on a single pretransplantation measure. The remaining 21.4% (n = 6) of nonadherent patients posttransplantation were adherent to all pretransplantation measures. In general, there was only a weak relationship between pretransplantation data and posttransplantation adherence. The exception was that patients who had missed 1 or more posttransplantation clinic appointments had higher mean pretransplantation phosphate levels (mean = 1.92, SD = 0.41) compared with those who had missed none (mean = 1.69, SD = 0.40; t[86] = 2.25, P = 0.027). This finding suggests that patients with higher phosphate levels pretransplantation are more likely to miss clinic appointments posttransplantation. There was no relationship of interdialytic weight gain with posttransplantation adherence, even when the analysis was confined to patients with no residual kidney function pretransplantation.

s that patients with higher phosphate levels pretransplantation are more likely to miss clinic appointments posttransplantation. There was no relationship of interdialytic weight gain with posttransplantation adherence, even when the analysis was confined to patients with no residual kidney function pretransplantation. Discussion The primary aim of this single-center retrospective study was to explore whether patterns of adherence behavior in patients on HD relate to posttransplantation adherence. Our findings do not support the likelihood of a strong direct relationship between these behaviors. However, the possibility remains of some overlap of nonadherent behaviors in these 2 settings, as evidenced by our finding of a relationship between pretransplantation phosphate control and subsequent attendance at posttransplantation follow-up. The number of patients categorized as nonadherent was dependent on how nonadherence was defined. Pretransplantation nonadherence ranged from 25% to 42%, and posttransplantation nonadherence ranged from 15.9% to 22.7%, depending on definition. Our finding of a higher nonadherence rate for phosphate control than that quoted in the literature3, 4 is highly likely to be due to our using a lower cut-off point for nonadherence, in line with previous research from our unit19 and UK clinical practice guidelines.20

anged from 15.9% to 22.7%, depending on definition. Our finding of a higher nonadherence rate for phosphate control than that quoted in the literature3, 4 is highly likely to be due to our using a lower cut-off point for nonadherence, in line with previous research from our unit19 and UK clinical practice guidelines.20 Defining pretransplantation adherence in terms of phosphate control, nonadherent patients were found to be of younger age at transplantation, to have less residual kidney function, and to have higher PTH levels. Both latter factors have well-established effects on phosphate control. Previous studies have also demonstrated negative correlations between age and phosphate control. Longer dialysis vintage was also associated with lower phosphate levels.2 Nevertheless, our findings may help to define a group of patients in whom targeted intervention may improve aspects of adherence pretransplantation and potentially posttransplantation.

ated negative correlations between age and phosphate control. Longer dialysis vintage was also associated with lower phosphate levels.2 Nevertheless, our findings may help to define a group of patients in whom targeted intervention may improve aspects of adherence pretransplantation and potentially posttransplantation. Posttransplantation predictors of nonadherence, defined in terms of number of missed clinic appointments, similarly indicated that nonadherent patients were significantly younger at transplantation and at dialysis initiation. These patients also had higher phosphate levels pretransplantation, above the Renal Association−recommended range, indicating inadequate phosphate control. Previous literature using serum phosphate as a clinical measure of nonadherence also indicated phosphate control as a major issue for HD patients.19, 23 This is similar to our findings, which showed a mean phosphate level of 1.74 (SD = 0.41) for the overall sample. Logistic regression identified that phosphate levels of 1.8 mmol/l or more predicted higher odds of nonattendance at posttransplantation clinic appointments.

e control as a major issue for HD patients.19, 23 This is similar to our findings, which showed a mean phosphate level of 1.74 (SD = 0.41) for the overall sample. Logistic regression identified that phosphate levels of 1.8 mmol/l or more predicted higher odds of nonattendance at posttransplantation clinic appointments. However, serum phosphate levels are affected by clinical variables and diet, and therefore their reliability as a measure of nonadherence should be interpreted with caution.23 There are multiple factors that could influence adherence to phosphate treatment, such as complex treatment regimen, high pill burden, side effects, and lack of immediate symptomatic benefit.23 It has been suggested that dietary and fluid restrictions may require more patient willpower in order to adhere. A study found that 57.6% of patients reported difficulty adhering to dietary prescription, and 56.3% reported that this was due to an inability to resist favorite foods.2 In addition, in the same study, 62% of patients reported some difficulty adhering to fluid restrictions, and 43.7% were unable to control their desire for fluid. This suggests that these aspects of the treatment regimen may be more challenging to adhere to and could explain why nonadherence rates are higher for these measures of pretransplantation nonadherence. This suggests that phosphate may be a better indicator of pretransplantation nonadherence than other measures because of the multifaceted nature of the behaviors required to manage phosphate levels, which encompass dietary restriction, phosphate binder medication adherence, and adherence to dialysis protocols. Age may be a factor in this relationship, as nonadherent patients judged by this parameter pretransplantation were significantly younger, as were those who missed posttransplantation clinic appointments. On the other hand, the absence of other significant predictors of posttransplantation nonadherence increases the possibility that this association was a chance finding.

rent patients judged by this parameter pretransplantation were significantly younger, as were those who missed posttransplantation clinic appointments. On the other hand, the absence of other significant predictors of posttransplantation nonadherence increases the possibility that this association was a chance finding. We found that the prevalence of nonadherence was greater pretransplantation by some but not all measures. This may suggest that adherence with treatment posttransplantation is more manageable than adherence to treatment pretransplantation, which involves both HD and associated medications. However, the measures of adherence used in these setting are, of necessity, very different, so this interpretation needs to be treated with caution.

est that adherence with treatment posttransplantation is more manageable than adherence to treatment pretransplantation, which involves both HD and associated medications. However, the measures of adherence used in these setting are, of necessity, very different, so this interpretation needs to be treated with caution. Overall, these findings suggest that pre- and posttransplantation adherence are only weakly associated. The relationship is complex. The challenges that patients experience with adherence pretransplantation may be different posttransplantation. There are multiple potential factors. For example, whether a patient has received a living or deceased donor organ may play a role in behavior modification.24, 25 In addition, anxiety, depression,9, 26 and socio-economic factors12 have been associated with nonadherence. Pretransplantation, patients on HD are usually entitled to free prescriptions. However, although posttransplantation clinic appointments are covered via the National Health Service (NHS), posttransplantation medication is not covered (unless patients meet the criteria for prescription payment exemption). This additionally could contribute to differences in nonadherence rates. Our findings do support existing literature pertaining to adherence in specific renal replacement therapy (RRT) modalities, for example, the relationship between younger age and nonadherence.

meet the criteria for prescription payment exemption). This additionally could contribute to differences in nonadherence rates. Our findings do support existing literature pertaining to adherence in specific renal replacement therapy (RRT) modalities, for example, the relationship between younger age and nonadherence. This study has both strengths and limitations. Although our study suggests the possibility of a complex relationship between pre- and posttransplantation adherence, the findings should be interpreted with caution. The sample size was small and single-centered. Younger age predicted high phosphate levels pretransplantation. We found no other major associations. However, the time frame in which nonadherence was assessed pretransplantation was 1 month in the previous literature,3, 4, 7 whereas this study assessed nonadherence over the 6 months pretransplantation. Our 6-month assessment of nonadherence may provide a more stable picture of patient behavioral patterns. In addition, a follow-up of 1 year may be too short a period for exploring posttransplantation nonadherence, as research suggests that rates of nonadherence can increase as time from transplantation increases.27 Finally, although we have attempted to delineate clinically relevant indices of nonadherence in the transplant population, there may be other parameters that may be more relevant. This could indicate the need to identify clinically relevant definitions that accurately measure nonadherence rates so as to ensure that this is reported reliably in future research. Notwithstanding these limitations, our study is 1 of the few to consider how patterns of adherence vary within patient groups as they transition between RRT modalities.

identify clinically relevant definitions that accurately measure nonadherence rates so as to ensure that this is reported reliably in future research. Notwithstanding these limitations, our study is 1 of the few to consider how patterns of adherence vary within patient groups as they transition between RRT modalities. Conclusion Poor phosphate control pretransplantation was associated with some aspects of adherence posttransplantation. However, our findings do not indicate a strong direct relationship between pre- and posttransplantation adherence. Whatever measure of adherence used pretransplantation, nonadherence is less posttransplantation and, in some cases, significantly so. However, the only adherence parameter that predicted posttransplantation adherence was pretransplantation phosphate control. Although some patients do improve adherence to treatment posttransplantation, nonadherence remains an issue for a proportion of patients posttransplantation. Nonadherent patients pretransplantation should be reviewed on a case-by-case basis for transplant eligibility, to determine whether adherence behavior could change posttransplantation or whether interventions are needed pretransplantation before wait listing. These findings require confirmation and further work to assess whether interventions in relation to pretransplantation adherence may enhance adherence posttransplantation and hence improve outcomes. Furthermore, enhancing patient understanding about the importance of medication and engaging in treatment regimens could help to improve adherence posttransplantation.

r work to assess whether interventions in relation to pretransplantation adherence may enhance adherence posttransplantation and hence improve outcomes. Furthermore, enhancing patient understanding about the importance of medication and engaging in treatment regimens could help to improve adherence posttransplantation. Disclosure All the authors declared no competing interests. Author Contributions KF conceived the study concept. All authors (AH, CL, SS, KF) contributed to the design of the study. KF and CL provided expertise in nephrology, including hemodialysis and renal transplant populations. AH conducted the evaluation. AH retrieved and analyzed the data and wrote the first draft of the manuscript. All authors provided feedback during the development of the manuscript and approved the final manuscript.

For the majority of the more than 600,000 patients in the United States with end-stage renal disease (ESRD),1 kidney transplantation is the preferred treatment, providing longer survival, better quality of life, lower hospitalization rates, and substantial cost savings compared to dialysis.2, 3

For the majority of the more than 600,000 patients in the United States with end-stage renal disease (ESRD),1 kidney transplantation is the preferred treatment, providing longer survival, better quality of life, lower hospitalization rates, and substantial cost savings compared to dialysis.2, 3 Patients who receive dialysis have an expected remaining lifetime of 5.9 years versus 16.4 years for transplant recipients, yet a large number of patients have not actively pursued wait-listing for a transplant.1 However, the relative risk of death varies substantially depending on individual characteristics, including age, race, or comorbidities.4 Prior studies suggest that most patients want information about treatment options and want to participate in the selection of treatment.5, 6 However, current literature suggests that ESRD patients have very limited knowledge about their mortality rate on dialysis versus transplant, and that not all patients are aware of the survival benefit of transplantation.7, 8, 9 In dialysis facilities, only 18% of centers reported having detailed discussions about the risks and benefits of living- and deceased-donor transplant.10 In addition, dialysis facility transplant educators have been found to need improved education on the benefits and process of renal transplantation.11 While transplant education ideally should start prior to ESRD and referral to a transplant center, providing information about the survival benefit of transplant versus dialysis, and in particular living- versus deceased-donor transplant, could help to increase patients’ knowledge and preferences to get a transplant.

1 While transplant education ideally should start prior to ESRD and referral to a transplant center, providing information about the survival benefit of transplant versus dialysis, and in particular living- versus deceased-donor transplant, could help to increase patients’ knowledge and preferences to get a transplant. We previously developed and validated a novel, shared patient–provider clinical decision aid called iChoose Kidney (iPad, iPhone, and website: www.ichoosekidney.emory.edu). iChoose Kidney is an electronic application that compares mortality on dialysis versus kidney transplantation to translate medical evidence into terms understandable to patients. Models of mortality for patients receiving dialysis versus deceased-donor (DD) or living-donor (LD) kidney transplantation were developed using a cohort of more than 700,000 patients in the nationally representative United States Renal Data System (2000–2011 data).12 The intention of this clinical decision aid is to help facilitate patient–provider discussions about the risks and benefits of transplantation versus dialysis and to support informed decision making among patients with ESRD.

0 patients in the nationally representative United States Renal Data System (2000–2011 data).12 The intention of this clinical decision aid is to help facilitate patient–provider discussions about the risks and benefits of transplantation versus dialysis and to support informed decision making among patients with ESRD. While mobile health application production has increased substantially in the last several years,13 with more than 100,000 iOS and Android health-related applications14, 15 currently in the marketplace, few methodologically rigorous studies have been conducted to confirm the efficacy or effectiveness of mobile health applications in improving access and outcomes of health. To our knowledge, no studies have been conducted to examine whether a mobile health application, designed to influence the shared clinical decision-making process between patient and provider to choose a treatment option, influences knowledge about treatment options and access to kidney transplantation.

omes of health. To our knowledge, no studies have been conducted to examine whether a mobile health application, designed to influence the shared clinical decision-making process between patient and provider to choose a treatment option, influences knowledge about treatment options and access to kidney transplantation. The purpose of this paper is to describe the study protocol used to design the iChoose Kidney randomized controlled trial to test the clinical efficacy of a clinical decision aid in improving knowledge about the survival benefit of transplantation versus dialysis. We will also examine secondary end points of decreased decisional conflict in choosing treatment options for ESRD, changing treatment preference from dialysis to transplant, and access to kidney transplantation. Planned subgroup analyses will also examine whether the efficacy of iChoose Kidney varies by health literacy, numeracy, and race. Finally, this study will seek to evaluate the usability of the iChoose Kidney tool among clinical providers (transplant nephrologists and surgeons).

nt, and access to kidney transplantation. Planned subgroup analyses will also examine whether the efficacy of iChoose Kidney varies by health literacy, numeracy, and race. Finally, this study will seek to evaluate the usability of the iChoose Kidney tool among clinical providers (transplant nephrologists and surgeons). Materials and Methods Study Overview The iChoose Kidney Study is a 2-arm randomized trial to test the efficacy of a mobile health clinical decision aid on improving patient knowledge about the survival benefit of transplantation versus dialysis. Prior to initiation of study activities, the iChoose Kidney Study was registered on ClinicalTrials.gov (protocol NCT02235571). This study was approved by Institutional Review Boards at Emory University, Columbia University, and Northwestern University. All patients will be consented for participation in the study before study involvement. Target Population, Setting, and Inclusion and Exclusion Criteria The study will be conducted in 3 US kidney transplant centers with a total target enrollment of 450 (150 at each site): Emory Transplant Center in Atlanta, Georgia, Columbia University Medical Center in New York, New York, and Northwestern University Transplant Center in Chicago, Illinois. ESRD patients will be recruited into the study during transplant medical evaluation if they meet inclusion criteria: (i) 18 to 70 years of age; (ii) no previous solid organ or multiorgan transplant; (iii) English speaking; and (iv) no severe cognitive or visual impairment.

Target Population, Setting, and Inclusion and Exclusion Criteria The study will be conducted in 3 US kidney transplant centers with a total target enrollment of 450 (150 at each site): Emory Transplant Center in Atlanta, Georgia, Columbia University Medical Center in New York, New York, and Northwestern University Transplant Center in Chicago, Illinois. ESRD patients will be recruited into the study during transplant medical evaluation if they meet inclusion criteria: (i) 18 to 70 years of age; (ii) no previous solid organ or multiorgan transplant; (iii) English speaking; and (iv) no severe cognitive or visual impairment. Study Arms At kidney transplant evaluation, the control group will receive the usual center-specific standard-of-care education about renal transplantation, without the iChoose Kidney decision aid. The intervention group will also receive the usual center-specific standard-of-care education. However, it will be supplemented by the use of the iChoose Kidney decision aid (either iPad or iPhone version). Among patients randomized to the intervention study arm, a nephrologist or transplant surgeon will use iChoose Kidney to provide individualized risk estimates of mortality or survival by treatment (dialysis vs. transplant; LD vs. DD transplant) based on a patient’s demographic and clinical characteristics. Using visual displays, the iChoose Kidney decision aid communicates both absolute and relative risk estimates in several messaging frames to increase patient and provider understanding of treatment benefit (Figure 1). The provider will have the option of which format to display for the patient when discussing the risks and benefits of each treatment option.Figure 1 Screenshots of the iChoose Kidney decision aid (iPad version), which communicate both absolute and relative risk estimates in several messaging frames. By entering a patient’s clinical information (sex, age, race, ethnicity, time on dialysis, and several comorbidities), the risk prediction calculator generates individualized 1- and 3-year mortality and survival risk estimates for (i) dialysis versus kidney transplant and (ii) deceased- versus living-donor transplant.

By entering a patient’s clinical information (sex, age, race, ethnicity, time on dialysis, and several comorbidities), the risk prediction calculator generates individualized 1- and 3-year mortality and survival risk estimates for (i) dialysis versus kidney transplant and (ii) deceased- versus living-donor transplant. Study Procedures At each site, research assistants will identify patients who meet inclusion criteria at the time of the patient’s evaluation visit. Patients who agree to participate will be provided with detailed study information, and informed consent with written documentation will be obtained from the research participant or appropriate representative prior to initiation in the iChoose Kidney Study. Research assistants will then generate study identification numbers for consented participants and randomize patients to 1 of 2 study arms (the intervention arm or the control arm) with a random number generator application via iPad or iPhone. Patients will be surveyed on the same day before (baseline) and after (follow-up) meeting with the provider, completing a total of 2 surveys. Providers will take a baseline survey prior to the start of the study, a follow-up survey directly after meeting with each study patient, and a final poststudy completion survey (Figure 2). All patient and provider surveys will be administered through SurveyMonkey, an electronic surveying tool (or paper surveys later entered into SurveyMonkey if Internet access is unavailable). Patients will be offered a $10 gift card for their participation.Figure 2 iChoose Kidney Study schema shows the study process and points of data collection for both control and intervention patients, and clinical providers (i.e., nephrologists or surgeons). All patients will receive informed consent and a baseline survey before being evaluated by a transplant provider. Patients will complete a postconsultation survey after the provider consultation. Providers will receive a baseline survey prior to patient recruitment. After consulting with each patient, providers will take a postconsultation survey. At study completion, providers will complete a poststudy survey.

ransplant provider. Patients will complete a postconsultation survey after the provider consultation. Providers will receive a baseline survey prior to patient recruitment. After consulting with each patient, providers will take a postconsultation survey. At study completion, providers will complete a poststudy survey. In the 2 months prior to initiation of the study, we will conduct pilot testing among the 3 sites to finalize study design and inform a power calculation. Pilot testing of the study will enable us to modify any issues with study design and data capture, and to ensure that all sites are conducting and recording data uniformly. Surveys Provider Baseline Survey Prior to using the iChoose tool in the RCT, providers will be surveyed regarding their professional background and experience with educational tools (mobile or other). Surveys will also assess the amount of time providers spend with patients during evaluation appointments, how often they discuss survival benefit of transplant versus dialysis, and how they communicate mortality risk with patients during the kidney pretransplant evaluation appointment. Patient Baseline Survey On the day of evaluation, research assistants will administer baseline surveys to patients prior to the commencement of their nephrology or transplant surgery consultation. Baseline surveys include questions about patient demographics, exposure to transplant education, transplant knowledge, preferences for treatment, and decisional conflict.

n, research assistants will administer baseline surveys to patients prior to the commencement of their nephrology or transplant surgery consultation. Baseline surveys include questions about patient demographics, exposure to transplant education, transplant knowledge, preferences for treatment, and decisional conflict. Provider Follow-up Survey Directly following each nephrology or surgery transplant evaluation, research assistants will administer a short survey to providers to assess satisfaction with and usefulness of the tool. Providers will report whether or not they discussed the survival benefit of transplant versus dialysis and report other conversation topics. Providers who use the tool with intervention patients will be asked whether they used mortality or survival estimates. They will also report whether they perceived that the patient gained transplant knowledge through the use of the iChoose Kidney tool. Providers who do not use the tool (if the patient was a control, or for some other reason) will be asked whether they felt their conversation with the patient could have benefited from using iChoose Kidney. Patient Follow-up Survey Following the provider consultation, patients will be given a follow-up survey, which consists of questions similar to the baseline survey, including transplant knowledge, preferences of treatment, decisional conflict, and health literacy and numeracy.

Provider Follow-up Survey Directly following each nephrology or surgery transplant evaluation, research assistants will administer a short survey to providers to assess satisfaction with and usefulness of the tool. Providers will report whether or not they discussed the survival benefit of transplant versus dialysis and report other conversation topics. Providers who use the tool with intervention patients will be asked whether they used mortality or survival estimates. They will also report whether they perceived that the patient gained transplant knowledge through the use of the iChoose Kidney tool. Providers who do not use the tool (if the patient was a control, or for some other reason) will be asked whether they felt their conversation with the patient could have benefited from using iChoose Kidney. Patient Follow-up Survey Following the provider consultation, patients will be given a follow-up survey, which consists of questions similar to the baseline survey, including transplant knowledge, preferences of treatment, decisional conflict, and health literacy and numeracy. Provider Poststudy Survey Within 1 month of study completion, providers will be surveyed again on their conversations with patients during the pretransplant evaluation and opinions on the iChoose Kidney application. Specifically, they will be asked about the average time spent during an evaluation visit, how often they discuss survival benefit of transplant versus dialysis, words used to communicate mortality risk, and perceived barriers to using mobile technology. They will also be asked their opinions of iChoose Kidney: its effect on patient knowledge and uncertainty regarding treatment, challenges to using the tool, how it could be improved, how it would best be used in practice (e.g., the time point when it should be used, who should administer it, which patients would benefit most), and whether they intend to use it in the future.

its effect on patient knowledge and uncertainty regarding treatment, challenges to using the tool, how it could be improved, how it would best be used in practice (e.g., the time point when it should be used, who should administer it, which patients would benefit most), and whether they intend to use it in the future. Outcome Measures Transplant Knowledge The primary outcome of the iChoose Kidney Study is change in patient knowledge about the survival benefits of transplantation. Improvement in knowledge will be measured using 8 survey questions in the pre- and postassessment. First, patients will be asked 2 knowledge questions in the baseline and follow-up survey: “On average, dialysis patients live: 1) longer than transplant patients, 2) about the same time as transplant patients, 3) a shorter time than transplant patients, or 4) unsure”; and “On average, living donor transplant patients live 1) a longer time than deceased donor transplant patients, 2) about the same time as deceased donor transplant patients, 3) a shorter time than deceased donor transplant patients, or 4) unsure.” Patients in both study arms will also be asked to estimate their absolute chance of 3-year survival on dialysis, transplant overall, deceased-donor transplant, and living-donor transplant (4 items) both pre- and postassessment. Lastly, patients will be asked to estimate their relative risk of mortality with dialysis versus kidney transplant (on a scale from 1 to 9) and whether they are more, less, or equally likely to die with dialysis compared to with a kidney transplant (2 items) both pre- and postassessment.

) both pre- and postassessment. Lastly, patients will be asked to estimate their relative risk of mortality with dialysis versus kidney transplant (on a scale from 1 to 9) and whether they are more, less, or equally likely to die with dialysis compared to with a kidney transplant (2 items) both pre- and postassessment. Decisional Conflict Decisional conflict was measured in both the pre- and postsurvey by a validated scale of 10 items that assess personal perceptions of uncertainty in choosing options, modifiable factors contributing to uncertainty, and effective decision making (Table 1).16 This particular version of the decisional conflict scale was selected for the study given its recommendation to be used on individuals with “limited reading or response skills.” The subscales for decisional conflict include uncertainty (patient feels certain about choice), informed status (patient informed of treatment options), values clarity (patient clear regarding personal values), and support (patient feels supported in decision making). We used the scale to determine whether or not patients had lower decisional conflict after using iChoose Kidney during the evaluation appointment.Table 1 Description of scales and tests administered to patients, pre– and post–provider consultation

ual organ, and that while one part of it acts, so to speak, passively in response to force impressed upon it from without, another part, endowed with sensibility, reacts to external forces in a direction which may be opposed to these forces, but is in all cases the appropriate one for the welfare of the whole organism. In 1963, Borst and Borst-de Geus2 discussed hypertension in light of Starling’s theory of circulatory homeostasis and postulated that blood pressure rises as a homeostatic reaction to deficient sodium excretion; that is, pressure rises to reestablish sodium balance (at the expense of persistently elevated blood pressure).2 This response, known as pressure natriuresis, has been reviewed from various perspectives by many investigators.3, 4, 5, 6, 7 Figure 1 shows central blood pressure as a function of the ECV and cardiac output as well as the kidneys’ regulation of sodium chloride and water reabsorption. A kidney’s decision to excrete sodium chloride and water is a function of mediators and controls, both extrarenal and intrarenal, that affect the rise in blood pressure, as well as the extravascular storage of sodium (discussed in an accompanying report in this series).8Figure 1 Central blood pressure shown as a function of the effective circulating volume and cardiac output as well as the kidneys’ regulation of sodium chloride (NaCl) and water (H2O) reabsorption, adapted from Starling1 and Borst and Borst-De Geus.2 A kidney’s willingness to excrete NaCl and water when blood pressure rises is a key controller of the effective circulating volume and blood pressure. This article examines factors, both extrarenal and intrarenal, that reduce this willingness in the proximal tubule (and mechanisms involved), leading to an elevation in blood pressure and activation of pressure natriuresis mechanisms in the proximal tubule that contribute to maintenance of effective circulating volume homeostasis.

sonal values), and support (patient feels supported in decision making). We used the scale to determine whether or not patients had lower decisional conflict after using iChoose Kidney during the evaluation appointment.Table 1 Description of scales and tests administered to patients, pre– and post–provider consultation Item Description of purpose How measured Number of items Knowledge of survival benefit To assess patient knowledge of survival benefit of treatment options (dialysis, deceased-donor transplant, and living-donor transplant), patient-reported survival estimates of each treatment, and relative risk of mortality with dialysis versus transplant Patient baseline and follow-up survey; scored on a scale by assessing whether survival benefit questions were correct or incorrect 8 Decisional conflict scale16 To assess personal perceptions of uncertainty in choosing options, modifiable factors contributing to uncertainty, and effective decision making Patient baseline and follow-up surveys, and scored on a scale from 1 to 100 10 Patient treatment preferences To assess treatment preferences for end-stage renal disease Patient baseline and follow-up surveys; patients asked what type of treatment they prefer and whether they want a kidney transplant (yes or no) 2 Newest Vital Signs17 To assess general literacy and numeracy skills as applied to health information, yielding an overall estimate of health literacy Patient follow-up survey; patients given a nutrition label and asked a series of free-response questions; responses scored and categorized into low, medium, or high literacy 6 Lipkus numeracy test18 To assess numeracy, or the ability to understand and use numeric information Patient baseline survey; scored on a scale from 1 to 11 and categorized into low, medium, or high 11

el and asked a series of free-response questions; responses scored and categorized into low, medium, or high literacy 6 Lipkus numeracy test18 To assess numeracy, or the ability to understand and use numeric information Patient baseline survey; scored on a scale from 1 to 11 and categorized into low, medium, or high 11 Patient Treatment Preferences To assess whether the tool affected treatment preferences (from dialysis to kidney transplantation), patients will be asked in both the pre- and postsurvey what type of treatment they prefer for their kidney disease (hemodialysis, peritoneal dialysis, transplant, or unsure). Participants will also be asked whether they want a kidney transplant (yes or no) (Table 1). Provider Opinions We will use qualitative and quantitative methods to assess provider preferences, opinions, and satisfaction in order to evaluate usability of the iChoose Kidney tool among providers. We will also compare providers’ intent to use mobile technology and the length of time providers report discussing the survival benefit of transplant versus dialysis pre- versus poststudy.

assess provider preferences, opinions, and satisfaction in order to evaluate usability of the iChoose Kidney tool among providers. We will also compare providers’ intent to use mobile technology and the length of time providers report discussing the survival benefit of transplant versus dialysis pre- versus poststudy. Access to Transplant Transplant access is a combined end point measured as completion of the transplant evaluation, number of living-donor inquiries, and wait-listing or transplant receipt. We will collect information on these measures from the patient medical record through data extraction at 6 months and 1 year from the patient’s evaluation appointment. Variables collected will include dates relevant to the transplant process (evaluation start and end date, wait-listing date, and transplant date) and whether the patient received a living-donor inquiry since evaluation (Table 2).Table 2 Variables collected for iChoose Kidney randomized control trial

uation appointment. Variables collected will include dates relevant to the transplant process (evaluation start and end date, wait-listing date, and transplant date) and whether the patient received a living-donor inquiry since evaluation (Table 2).Table 2 Variables collected for iChoose Kidney randomized control trial Variable name Mode of collection Patient baseline survey Research assistant observation EMR data abstraction or extraction Patient demographic and socioeconomic factors Age x Race x x x Hispanic ethnicity x x x Sex x x x Income x Marital status x Education level x Health insurance x Employment status x Self-rated health x Internet access x Social support x x Health literacy x Health numeracy x Prior exposure to transplant x Time point first educated about transplant x Patient clinical factors Body mass index x History of hypertension x History of diabetes x History of cardiovascular disease x Low albumin levels x Date of dialysis start x Time on dialysis x x Dialysis modality x x Outcome measures Knowledge about transplant x Decisional conflict x Treatment preferences x Access to transplant measures Date of transplant evaluation x Transplant evaluation end date x Date of wait-listing x Date of transplant x Number of living-donor inquiries x All variables collected were measures at the time of kidney transplant evaluation. EMR, electronic medical record.

s transplant, ultimately helping to facilitate more informed decision making among ESRD patients about their treatment options. However, prior to dissemination for use in a clinical setting, the efficacy of the clinical decision aid in improving patient knowledge about the survival benefit of transplantation is needed. While transplantation is the preferred treatment for most ESRD patients, significant barriers exist in access to multiple steps of the kidney transplantation process. Patients who are minorities, of lower socioeconomic status,22 and of lower health literacy23 are less likely to receive a kidney transplant. Education about transplantation as a treatment option may also play a role in disparities in access to kidney transplantation. According to the US Agency for Healthcare Research and Quality, disparities in health outcomes are due, in part, to differences in access to health care, provider biases, poor patient–provider communication, and poor health literacy.24 Mobile health technology can be a useful way to deliver interventions and may have a high potential for reducing health disparities. Effective risk communication strategies must consider patients with varying degrees of health literacy, numeracy, and education levels to ensure that the information provided by the tool is relevant and understandable to patients from diverse backgrounds. The iChoose Kidney decision aid communicates risks of mortality on dialysis versus transplant in both absolute and relative terms to meet best practices in conveying health risks.25 Evidence-based research supports presenting absolute risk in visuals to emphasize the clinical significance and size of risks,26 which iChoose Kidney utilizes.

Variable name Mode of collection Patient baseline survey Research assistant observation EMR data abstraction or extraction Patient demographic and socioeconomic factors Age x Race x x x Hispanic ethnicity x x x Sex x x x Income x Marital status x Education level x Health insurance x Employment status x Self-rated health x Internet access x Social support x x Health literacy x Health numeracy x Prior exposure to transplant x Time point first educated about transplant x Patient clinical factors Body mass index x History of hypertension x History of diabetes x History of cardiovascular disease x Low albumin levels x Date of dialysis start x Time on dialysis x x Dialysis modality x x Outcome measures Knowledge about transplant x Decisional conflict x Treatment preferences x Access to transplant measures Date of transplant evaluation x Transplant evaluation end date x Date of wait-listing x Date of transplant x Number of living-donor inquiries x All variables collected were measures at the time of kidney transplant evaluation. EMR, electronic medical record. Other Covariates Patient Factors We will collect the following demographic, socioeconomic, and clinical variables from patient surveys: age, race, sex, income, marital status, education level, health insurance, time on dialysis, dialysis type, self-rated health, and Internet access. A clinical data warehouse for all 3 transplant centers will be used to collect data on patient race, Hispanic ethnicity, age, body mass index, history of comorbidities (diabetes, hypertension, cardiovascular disease), low albumin level (<3.5 g/dl), and dialysis start date. Research assistants will also track whether patients had a social support member with them at the evaluation appointment (Table 2).

n patient race, Hispanic ethnicity, age, body mass index, history of comorbidities (diabetes, hypertension, cardiovascular disease), low albumin level (<3.5 g/dl), and dialysis start date. Research assistants will also track whether patients had a social support member with them at the evaluation appointment (Table 2). Prior Exposure to Transplant Patients will be asked to check a list of several ways that they may have been exposed to information about transplant (e.g., through brochures, websites, by attending kidney support groups, etc.). They will also be asked when they were first educated about transplant as a treatment option (Table 2). Health Literacy and Numeracy Health literacy will be measured using the Newest Vital Signs assessment17 (6 items), and numeracy will be assessed using the Lipkus test18 (11 items) (Table 1). To administer Newest Vital Signs, patients will be given a nutrition label and asked a series of free-response questions. Responses will then be scored and categorized into low, medium, or high literacy. The Lipkus test assesses the patient’s ability to understand and use numeric information, and includes questions on risks and percentages. Numeracy will be scored continuously on a scale from 1 to 11.

a series of free-response questions. Responses will then be scored and categorized into low, medium, or high literacy. The Lipkus test assesses the patient’s ability to understand and use numeric information, and includes questions on risks and percentages. Numeracy will be scored continuously on a scale from 1 to 11. Provider Factors The following provider factors will be collected: time practicing medicine or surgery, use of patient educational tools, perceived barriers to using mobile technology, average time spent during an evaluation visit, how often the survival benefit of transplant versus dialysis is discussed, and specific reasons why the provider might not discuss the survival benefit of transplant versus dialysis. Data Management Research assistants will track study participants uniformly across sites, using a worksheet they will later transcribe into a Microsoft Excel spreadsheet. Patient demographics, social support, and any deviations from the study protocol will be recorded in this spreadsheet (e.g., if the provider performed the intervention with a control patient). At each site, patient demographics (race, age, sex, ethnicity), dialysis start date, and comorbidities included in the iChoose Kidney application will be abstracted or extracted from patient electronic medical records. To ensure data accuracy, self-reported patient survey data will be compared to data collected from electronic medical records.

ographics (race, age, sex, ethnicity), dialysis start date, and comorbidities included in the iChoose Kidney application will be abstracted or extracted from patient electronic medical records. To ensure data accuracy, self-reported patient survey data will be compared to data collected from electronic medical records. The limited de-identified electronic medical record–collected data from all 3 sites, as well the de-identified SurveyMonkey patient and provider survey data, will be merged, cleaned, and operationalized at Emory University. Microsoft Excel and SAS 9.4 (SAS Institute, Cary, NC) will be used to prepare and merge study data. Statistical Analyses Descriptive Analyses All analyses will be conducted using an intention-to-treat approach where patients will remain assigned to the treatment condition they were randomized to regardless of whether they receive the intervention (e.g., patients randomized to the intervention arm but who did not receive the iChoose Kidney clinical decision aid will still be considered as intervention participants). Descriptive analyses of transplant center–level baseline variables (demographic and clinical characteristics and transplant access measures) will be compared. To evaluate the differences between study arms at baseline, Pearson’s χ2 tests and t-tests, or their non-parametric equivalents, will be performed for categorical and continuous variables, respectively. Statistically significant differences in baseline characteristics will be adjusted when assessing the overall intervention effect as described above.

een study arms at baseline, Pearson’s χ2 tests and t-tests, or their non-parametric equivalents, will be performed for categorical and continuous variables, respectively. Statistically significant differences in baseline characteristics will be adjusted when assessing the overall intervention effect as described above. Change in Knowledge To assess change in knowledge from pre- to postassessment, the change in score from the knowledge questions will be calculated for each patient. A paired t-test will be used to determine whether this difference in proportions is significant in intervention versus control patients. Change in Decisional Conflict Each of 10 items in the decisional conflict scale will be given a value of 0, 2, or 4 for responses “yes,” “no,” or “unsure,” respectively. The items will then be summed, divided by 10, and multiplied by 25 to determine the total score of the decisional conflict variable. Decisional conflict will then be scored on a scale from 0 (no decisional conflict) to 100 (high decisional conflict) (Table 1). The change in decisional conflict will be calculated for each patient from pre– to post–provider consultation to determine whether the change is significantly higher in intervention patients versus control patients using a χ2 test or multivariable logistic regression.

flict) to 100 (high decisional conflict) (Table 1). The change in decisional conflict will be calculated for each patient from pre– to post–provider consultation to determine whether the change is significantly higher in intervention patients versus control patients using a χ2 test or multivariable logistic regression. Change in Treatment Preferences We will calculate the proportion of patients who initially preferred dialysis to transplantation but changed to transplantation during the postsurvey. A χ2 test will be used to determine whether this difference is significant in intervention versus control patients. Access to Transplant We will create a composite measure of living-donor inquiries, placement on the wait list, and receipt of transplant to determine long-term outcomes at 1 year after inclusion in the study. We will use χ2 tests or multivariable logistic regression to determine whether positive outcomes are significantly different among intervention patients versus control patients. Planned Subgroup Analyses To determine whether the effect of iChoose Kidney varies by health literacy, numeracy, and race, we will conduct similar analyses for knowledge and decisional conflict but across different subgroups of patients.

Access to Transplant We will create a composite measure of living-donor inquiries, placement on the wait list, and receipt of transplant to determine long-term outcomes at 1 year after inclusion in the study. We will use χ2 tests or multivariable logistic regression to determine whether positive outcomes are significantly different among intervention patients versus control patients. Planned Subgroup Analyses To determine whether the effect of iChoose Kidney varies by health literacy, numeracy, and race, we will conduct similar analyses for knowledge and decisional conflict but across different subgroups of patients. Provider Attitudes To evaluate whether providers believe the tool impacted study patients, we will use descriptive analyses to assess variables collected directly after the patient appointment. We will use χ2 tests to assess whether there was a difference in the number of times providers had the conversation about the survival benefit of transplant versus dialysis, and of living- versus deceased-donor transplants with intervention versus control patients. We will also compare differences between provider responses at baseline and poststudy regarding use of mobile decision aids in general and intention to use iChoose Kidney using χ2 tests.

al benefit of transplant versus dialysis, and of living- versus deceased-donor transplants with intervention versus control patients. We will also compare differences between provider responses at baseline and poststudy regarding use of mobile decision aids in general and intention to use iChoose Kidney using χ2 tests. Power and Sample Size Calculations Sample size calculations will be based on our primary aim to improve patient knowledge about the survival benefit of transplantation versus dialysis. To test the primary null hypothesis of no difference in referral proportions between the control and intervention groups, a sample size of 450 patients (225 patients per study arm) achieves 80% power to detect a moderate knowledge difference of 13% between the 2 groups at 5% significance level.

ntation versus dialysis. To test the primary null hypothesis of no difference in referral proportions between the control and intervention groups, a sample size of 450 patients (225 patients per study arm) achieves 80% power to detect a moderate knowledge difference of 13% between the 2 groups at 5% significance level. Discussion Critically important treatment decisions are often made without evidence-based information about a patient’s prognosis. Typically, average, population-based, non-tailored prognosis estimates of mortality on dialysis versus transplant are the only types of estimates communicated to individual ESRD patients, if these estimates are communicated at all.19 Despite overwhelming evidence to support transplantation for certain ESRD patients,20 patient-specific prognostic information is rarely used to calculate a patient's individualized prognosis.21 The iChoose Kidney clinical decision aid12 is a tool that clinical providers, including nephrologists, surgeons, and other ESRD educators, could potentially use with patients to explain their chance of mortality or survival on dialysis versus transplant, ultimately helping to facilitate more informed decision making among ESRD patients about their treatment options. However, prior to dissemination for use in a clinical setting, the efficacy of the clinical decision aid in improving patient knowledge about the survival benefit of transplantation is needed.

cision aid communicates risks of mortality on dialysis versus transplant in both absolute and relative terms to meet best practices in conveying health risks.25 Evidence-based research supports presenting absolute risk in visuals to emphasize the clinical significance and size of risks,26 which iChoose Kidney utilizes. Upon completion of this research, we will have assessed the effect of a decision aid in improving knowledge about the survival benefits of kidney transplantation versus dialysis. We will also have assessed providers’ attitudes toward clinical decision tools and their impact on the patient–provider interaction. If found to be effective in improving knowledge about the survival benefit of transplant, the clinical decision aid could be a useful tool for improving access to transplantation and improving decisional conflict with regard to treatment options. A strength of this randomized controlled trial is the conduct of the research among a diverse, multicenter population of ESRD patients, as results may be broadly generalizable to other ESRD patients who have been referred for transplantation to the more than 250 transplant centers across the United States. Further, with additional testing in a population of incident ESRD patients, the iChoose Kidney decision aid may also be applicable for use across the more than 5000 dialysis facilities or chronic kidney disease clinics in the United States. Future studies could also assess the applicability of the use of such a tool outside of the United States. Disclosure All the authors declared no competing interests.

A strength of this randomized controlled trial is the conduct of the research among a diverse, multicenter population of ESRD patients, as results may be broadly generalizable to other ESRD patients who have been referred for transplantation to the more than 250 transplant centers across the United States. Further, with additional testing in a population of incident ESRD patients, the iChoose Kidney decision aid may also be applicable for use across the more than 5000 dialysis facilities or chronic kidney disease clinics in the United States. Future studies could also assess the applicability of the use of such a tool outside of the United States. Disclosure All the authors declared no competing interests. Acknowledgments This trial is registered as NCT02235571 in ClinicalTrials.gov. We thank the Norman S. Coplon Satellite Healthcare Foundation for providing funding for this clinical trial. This work is also supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award number KL2TR000455. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

In 1909, Ernest Starling published his lectures in The Fluids of the Body.1 In this monograph, Starling outlines his thoughts on the relationship between body fluid balance, blood circulation, central venous pressure, and cardiac performance. Importantly, renal control of body fluid balance is viewed as key for circulatory and cardiac performance:1Change in the blood flow through the kidney may bring about alterations in the flow of urine quite irrespective of the composition of the blood or of the tissues. The occurrence of these two classes of phenomena seems to be determined by the fact that the kidney is a dual organ, and that while one part of it acts, so to speak, passively in response to force impressed upon it from without, another part, endowed with sensibility, reacts to external forces in a direction which may be opposed to these forces, but is in all cases the appropriate one for the welfare of the whole organism.

examines factors, both extrarenal and intrarenal, that reduce this willingness in the proximal tubule (and mechanisms involved), leading to an elevation in blood pressure and activation of pressure natriuresis mechanisms in the proximal tubule that contribute to maintenance of effective circulating volume homeostasis. Hypertension is the leading cause of stroke and cardiovascular diseases affecting 30% of the adult population in Western cultures.9 Blood pressure can be elevated by vasoconstriction or by increasing ECV. Excess sodium reabsorption raises ECV and blood pressure, yet, according to Guyton,3 kidneys have the capacity via pressure natriuresis to excrete enough sodium and volume to normalize blood pressure in the face of expanded ECV. Hypertension was classically viewed as a failure of pressure natriuresis; however, a recent discussion of the role of kidneys in the pathogenesis of hypertension10 concluded that for hypertension to become chronic there must be impairment of both renal output of salt and water as well as dysfunction of peripheral vascular tone; for example, a failure of peripheral vasodilation due to arterial stiffness. Support for the latter is provided by recent studies illustrating a positive feedback loop wherein arterial stiffening leads to more arterial stiffening.11 Although appreciating these complex interactions of body fluids, cardiac output, vascular stiffness, and blood pressure, this report will focus on the regulation of the renal proximal tubule NHE3 as a case-in-point mediator of the pressure natriuresis response, specifically regulation of NHE3 trafficking and abundance, our understanding of how renal dysfunction resets NHE3 regulation to higher pressures, and strategies that may be exploited to improve pressure natriuresis.

tion of the renal proximal tubule NHE3 as a case-in-point mediator of the pressure natriuresis response, specifically regulation of NHE3 trafficking and abundance, our understanding of how renal dysfunction resets NHE3 regulation to higher pressures, and strategies that may be exploited to improve pressure natriuresis. Proximal Tubule NHE3 Regulated by Redistribution Within the MV As reviewed by Palmer and Schnermann,12 the proximal tubule reabsorbs two-thirds of the salt and water filtered at the glomerulus (120 ml/min) and NHE3 is the main sodium transporter driving transcellular reabsorption in this region. The proximal tubule is a leaky epithelium well built to reabsorb the ∼80 ml/min filtrate. As illustrated in the cross-section of the electron micrograph in Figure 2, the proximal tubule has an apical pole covered with a tall brush border of MV each scaffolded by an actin filament core bundled by villin. This specialization increases the surface area for reabsorption more than 30-fold.13 The apical MV contain water channels as well as many different transporters to reabsorb cations, anions, and substrates from the filtrate. Importantly, a significant fraction of the filtered salt and water is reabsorbed via a paracellular route by claudins.14Figure 2 Elevating salt intake from 0.4% to 4% does not change the total abundance of proximal tubule sodium hydrogen exchanger 3 (NHE3), but redistributes NHE3 to the base of the proximal tubule microvilli. In salt-resistant animals this occurs without an elevation in blood pressure. (a) Immunoblots of renal cortical homogenates from rats fed 0.4%- or 4%-salt diet.18 (b) Cross-section of proximal tubule shown in a simple model and electron micrograph illustrating dense apical microvilli. NHE3 redistribution along the proximal tubule microvilli is detected by colabeling the actin bundling protein villin (red V) and NHE3 (green circle) with specific antibodies.18 Left half of the proximal tubule model represents the proximal tubule at a normal-salt diet with the villin and NHE3 colocalized, yielding microvilli stained yellow. The right half of the proximal tubule represents the proximal tubule during a high-salt diet with NHE3 retracted to the base of the microvilli exposing red V in the body of the villi and green/yellow at the base of the microvilli.

ormal-salt diet with the villin and NHE3 colocalized, yielding microvilli stained yellow. The right half of the proximal tubule represents the proximal tubule during a high-salt diet with NHE3 retracted to the base of the microvilli exposing red V in the body of the villi and green/yellow at the base of the microvilli. (c) Colabeling of the proximal tubule from rats fed 0.4% sodium chloride (left half-tubule) with both NHE3 and villin in the body of the microvilli, and 4.0% sodium chloride (right half-tubule) with NHE3 concentrated at the base of the microvilli. (d) Myosin VI, an atypical molecular motor implicated in the redistribution of NHE3 and sodium phosphate within the plane of the microvilliar membrane,17, 19 is located in the body of the microvilli with a 0.4% sodium chloride diet and redistributes to the base of the microvilli during a 4% sodium chloride diet.

villi. (d) Myosin VI, an atypical molecular motor implicated in the redistribution of NHE3 and sodium phosphate within the plane of the microvilliar membrane,17, 19 is located in the body of the microvilli with a 0.4% sodium chloride diet and redistributes to the base of the microvilli during a 4% sodium chloride diet. Membrane transporters and channels can be regulated by trafficking between plasma membrane and intracellular membranes, altered total pool size, covalent modifications such as cleavage or phosphorylation, or protein–protein interaction. Once NHE3 is localized to the proximal tubule MV, there is scant in vivo evidence for regulated trafficking between MV and intracellular pools. Rather, NHE3, localized to ordered lipid domains (rafts) in the MV, redistribute between the body and the base of the MV, moving in the plane of the microvillar membranes, likely driven by the atypical molecular motor myosin VI.15, 16, 17 This redistribution from one location to another, rather than degradation and synthesis, facilitates rapid continuous adaptation to changing salt intake, ECV, and/or blood pressure. Figure 2 illustrates the simple case of NHE3 regulation in the transition between normal and high-salt diets in the absence of any change in blood pressure.18 Figure 2a illustrates that this natriuresis occurs without any change in NHE3 total abundance. Figure 2b shows cross-sections of proximal tubules in a model and in an electron micrograph illustrating organization of dense apical MV. NHE3 redistribution along the proximal tubule MV is detected by colabeling the actin bundling protein villin (red V) and NHE3 (green circle) with specific antibodies.18 The left half of the model represents a proximal tubule from a normal-salt-diet-fed rat with villin and NHE3 colocalized in the MV, yielding a yellow stain. The right half of the proximal tubule model represents a proximal tubule from a high-salt-diet-fed rat wherein NHE3 is retracted to the base of the MV, exposing a red V in body of the villi and green/yellow at the base of the MV. Figure 2c demonstrates NHE3 in the body of the proximal tubule MV in rats fed 0.4% sodium chloride (left half-tubule), and to the base of the MV, in rats fed 4.0% sodium chloride (right half-tubule).

n NHE3 is retracted to the base of the MV, exposing a red V in body of the villi and green/yellow at the base of the MV. Figure 2c demonstrates NHE3 in the body of the proximal tubule MV in rats fed 0.4% sodium chloride (left half-tubule), and to the base of the MV, in rats fed 4.0% sodium chloride (right half-tubule). Myosin VI, an atypical molecular motor implicated in the redistribution of NHE3 and sodium phosphate transporter within the plane of the microvillar membrane,17, 19 also redistributes from the body to the base of the MV during high-salt diet, presumably driving the NHE3 (Figure 2d). As discussed below, when NHE3 is clustered at the base of the MV, its activity is predicted to be inhibited by unfavorable pH gradients.20 Recent studies have shown that when the excretory function of the kidney is chronically impaired by inhibiting nitric oxide synthase activity, the resultant renal inflammation blunts the depression in sodium transport during high-salt diet, leading to a rise in blood pressure; that is, the renal dysfunction leads to salt-sensitive hypertension.21

at when the excretory function of the kidney is chronically impaired by inhibiting nitric oxide synthase activity, the resultant renal inflammation blunts the depression in sodium transport during high-salt diet, leading to a rise in blood pressure; that is, the renal dysfunction leads to salt-sensitive hypertension.21 Evidence for a Role of Proximal Tubule in Acute Pressure Diuresis The classic acute pressure natriuretic protocol developed by Roman and Cowley22 and a typical response is illustrated in Figure 3a: In inactin-anesthetized male rats, raising mean atrial blood pressure from 87 to 130 mm Hg rapidly increases urine output more than 10-fold.23, 24 Because the pressure–natriuresis response is very large and rapid, the proximal tubule was identified as a good candidate region for natriuresis—it reabsorbs the bulk of the filtered load. In the mid-1980s, Chou and Marsh25, 26 developed a video-densitometric approach to analyze tubular flow in real time and found that acutely raising blood pressure rapidly increased end proximal tubule flow rate by 50%. Because they also demonstrated this occurred without appreciable changes in glomerular filtration rate (GFR) or renal blood flow (due to autoregulation), they concluded that proximal tubule sodium transport was inhibited during acute hypertension, and that this response contributed not only to pressure diuresis, but also to the autoregulation of GFR and renal blood flow (mediated by increasing salt delivery to the macula densa). This response confirmed Starling’s assertions that “the mechanisms, which determine the adaptation of the organism to changes in the total volume of its fluid content, must come into play with every rise or fall in the general blood pressure.”1Figure 3 Acute hypertension produced by raising total peripheral resistance rapidly increases urine salt and volume output (pressure natriuresis/diuresis) and redistributes the sodium hydrogen exchanger 3 (NHE3) to the base of the proximal tubule microvilli. (a) In inactin-anesthetized male rats, raising mean atrial blood pressure (BP) from 87 to 130 mm Hg according to the protocol of Roman and Cowley22 rapidly increases urine output more than 10-fold.23 (b) The cross-section model of proximal tubule as described in Figure 2.

to the base of the proximal tubule microvilli. (a) In inactin-anesthetized male rats, raising mean atrial blood pressure (BP) from 87 to 130 mm Hg according to the protocol of Roman and Cowley22 rapidly increases urine output more than 10-fold.23 (b) The cross-section model of proximal tubule as described in Figure 2. Colabeling of the proximal tubule from rats at baseline BP (left half-tubule) with both NHE3 and villin in the body of microvilli, and at elevated BP (right half-tubule) with NHE3 concentrated at the base of the microvilli.16 (c) When angiotensin II (AngII) levels are clamped by preinfusion with an angiotensin-converting enzyme inhibitor to prevent AngII production along with a constant infusion of AngII to maintain baseline BP, the redistribution of NHE3 to the base of the microvilli is significantly blunted.32 The findings suggest that a drop in local AngII may contribute to the NHE3 redistribution during acute hypertension, and that high local AngII can blunt pressure natriuresis.

g with a constant infusion of AngII to maintain baseline BP, the redistribution of NHE3 to the base of the microvilli is significantly blunted.32 The findings suggest that a drop in local AngII may contribute to the NHE3 redistribution during acute hypertension, and that high local AngII can blunt pressure natriuresis. NHE3 Regulated by Redistribution Within the MV During Acute Hypertension The renal natriuretic and diuretic responses to acute or chronic increases in blood pressure are referred to very generally as pressure natriuresis. The responses to acute hypertension in the proximal tubule (Figure 3) are analogous to those observed during a high-salt diet (Figure 2). The pressure–natriuretic signals, discussed below, provoke the dynamic redistribution of apical transporters, including NHE3 and sodium phosphate cotransporter II (NaPiII),16 driven by molecular motors (eg, myosin VI and IIA) and cytoskeleton-associated proteins, to the base of the proximal tubule MV (Figure 3b).17, 19 The lipid raft-associated NHE3 remains at the base15 and the nonraft-associated NaPiII is endocytosed, culminating in decreased sodium transport activity and increased proximal tubule flow rate.7, 27, 28 Recently, with Brasen et al.,20 we visualized the hypertension-stimulated redistribution of NHE3 to the base of the MV in vivo using 2-photon microscopy with the pH indicator BCECF. Mathematic modeling of this redistribution suggests that NHE3 clustering produces unfavorable pH microdomains near the bottom of the brush border sufficient to inhibit NHE3 activity.20 This conclusion helps to explain how NHE3 redistribution can contribute to natriuretic responses during a high-salt diet and acute hypertension.

c modeling of this redistribution suggests that NHE3 clustering produces unfavorable pH microdomains near the bottom of the brush border sufficient to inhibit NHE3 activity.20 This conclusion helps to explain how NHE3 redistribution can contribute to natriuretic responses during a high-salt diet and acute hypertension. The signaling that provokes the rapid decrease in proximal tubule sodium and volume reabsorption, and retraction of NHE3 to the base of the MV in the face of autoregulated renal blood flow and GFR appears to involve many layers of regulation by both intrinsic and extrinsic factors. Summarizing the findings of multiple labs, it appears that the initial natriuretic response is driven by rapid local generation of 20-hydroxyeicosatetraeonic acid and nitric oxide (NO),28, 29, 30 (perhaps involving nonautoregulating vasculature that senses the hypertension), and that the response in the proximal tubule involves the production of cyclic guanosine monophosphate,31 which plays a role in depressing sodium transport. We found that clamping AngII levels at a nonpressor level by coinfusion of both the ACE inhibitor captopril and AngII before the acute hypertension protocol significantly blunted the pressure diuresis as well as the redistribution of NHE3 to the base of the MV (Figure 3c).32, 33 The decrease in AngII is key to not only allow NHE3 redistribution to the base of the MV, but also to sustain the response by reducing sodium transport in AngII-sensitive regions all along the nephron, including the distal tubule.34 Not the focus of this review, but important to discuss in light of signaling, is the fact that the medullary loop of Henle also clearly participates in pressure natriuresis during hypertension: medullary blood flow, NO, and reactive oxygen species participate as signals.35, 36 A recent report from Crowley et al.37 suggests that inflammatory accumulation of interleukin-1 during hypertension activates loop of Henle Na+-K+- 2Cl- cotransporter (NKCC2), which blunts the natriuresis and raises blood pressure. Eliminating interleukin-1 limits the blood pressure elevation by reducing NKCC2 sodium reabsorption. This study provides another example of how injury signals contribute to hypertension by blunting pressure natriuresis.

of Henle Na+-K+- 2Cl- cotransporter (NKCC2), which blunts the natriuresis and raises blood pressure. Eliminating interleukin-1 limits the blood pressure elevation by reducing NKCC2 sodium reabsorption. This study provides another example of how injury signals contribute to hypertension by blunting pressure natriuresis. In summary, regarding natriuretic signaling during hypertension, there is consensus that multiple signals have the potential to act all along the nephron; thus, the pressure natriuresis response is the sum of the prevailing natriuretic and antinatriuretic influences.

of Henle Na+-K+- 2Cl- cotransporter (NKCC2), which blunts the natriuresis and raises blood pressure. Eliminating interleukin-1 limits the blood pressure elevation by reducing NKCC2 sodium reabsorption. This study provides another example of how injury signals contribute to hypertension by blunting pressure natriuresis. In summary, regarding natriuretic signaling during hypertension, there is consensus that multiple signals have the potential to act all along the nephron; thus, the pressure natriuresis response is the sum of the prevailing natriuretic and antinatriuretic influences. Renin Angiotensin System Regulates Proximal Tubule NHE3 Distribution and Abundance The renin angiotensin aldosterone system is the most powerful controller of ECV and blood pressure, as recently reviewed by Rossier.38 AngII increases sodium reabsorption in the proximal tubule mediated by AT1 receptors. Building on the findings of the suppression of pressure natriuresis during the AngII clamp, we investigated the acute effects of adding or inhibiting AngII without changing blood pressure. Because ACE inhibitors are among the most popular drugs prescribed to lower blood pressure and slow the progression of renal and heart disease, it is key to understand how they regulate the proximal tubule NHE3. We addressed this issue by merging physiology and proteomics.39 Leong et al.39 identified a dose of the ACE inhibitor captopril that, when infused for 20 minutes into anesthetized rats, did not change blood pressure or GFR, but did significantly increase urine output (12 μg/min). With the Yip lab, they demonstrated that this dose rapidly increased proximal tubule flow rate, evidence for ACE inhibitor suppression of proximal tubule sodium reabsorption; Figure 4b shows that NHE3 is retracted to the microvillar base after 20 minutes of ACE inhibitor treatment, evidence that basal AngII is responsible, at least in part, for the location of NHE3 within the MV at baseline.39 With the Klein lab, Leong et al.39 applied a limited proteomic approach and discovered that other brush border proteins that redistribute with captopril include myosin VI, dipeptidyl peptidase VI, NHERF-1, ezrin, megalin, vacuolar H+-ATPase, aminopeptidase N, and clathrin.Figure 4 Angiotensin II (AngII) regulates sodium hydrogen exchanger 3 (NHE3) distribution in the proximal tubule microvilli. Renal cortex examined by colabeling NHE3 (green) and villin (red) with specific antibodies as described in Figure 1. (a) NHE3 in body of the microvilli in an inactin-anesthetized rat at baseline.

Figure 4 Angiotensin II (AngII) regulates sodium hydrogen exchanger 3 (NHE3) distribution in the proximal tubule microvilli. Renal cortex examined by colabeling NHE3 (green) and villin (red) with specific antibodies as described in Figure 1. (a) NHE3 in body of the microvilli in an inactin-anesthetized rat at baseline. (b) Infusing captopril for 20 minutes at a dose that does not lower blood pressure (12 μg/min) redistributes NHE3 to the base of the microvilli.39 (c) Infusing captopril for 20 minutes followed by AngII infusion for 20 minutes (20 ng/kg/min), at doses that do not change blood pressure, moves NHE3 from the base to the body of the microvilli.40 (d) Coinfusing a β–arrestin-biased AngII receptor-(AT1R) specific agonist along with captopril for 20 minutes (TRV120023) moved NHE3 to the base of the microvilli and blocked the effects of subsequent AngII infusion.41 (e) Infusion of AngII into rats for 14 days (400 ng/kg/min) raised the mean arterial pressure from 120 to 160 mm Hg and decreases the total abundance of renal cortical NHE3 by 20%, yet despite the hypertension NHE3 is retained in the body of the microvilli.35 Images in panels a and e were collected with the same settings, demonstrating that the ratio of NHE3 to the villin signal has decreased during AngII.

sure from 120 to 160 mm Hg and decreases the total abundance of renal cortical NHE3 by 20%, yet despite the hypertension NHE3 is retained in the body of the microvilli.35 Images in panels a and e were collected with the same settings, demonstrating that the ratio of NHE3 to the villin signal has decreased during AngII. Starting with ACE inhibitor-infused rats, Riquier-Brison et al.40 determined the effects of acute AngII infused at a rate that did not alter blood pressure for another 20 minutes (20 ng/kg/min) and found that the NHE3 completely returned to the body of the MV (AngII + captopril) (Figure 4c). Leong et al.39 previously reported that AngII + captopril infusion also decreased proximal tubule flow rate, providing evidence that the NHE3 redistribution is antinatriuretic. In collaboration with Carneiro de Morais et al.,41 we recently investigated the effects of a β–arrestin-biased AT1 receptor (AT1R) agonist (TRV120023) on proximal tubule NHE3. Stimulating β–arrestin-biased pathways promotes G-protein receptor internalization and desensitization and can activate G-protein independent responses.42 Perfusing proximal tubules in vivo with the TRV compound reduced bicarbonate reabsorption, a surrogate measure of NHE3 transport, and also moved NHE3 to the bottom of the MV.41 TRV can also prevent the effects of AngII activation: Coinfusing the TRV compound with ACE inhibitors for 20 minutes blocks the effects of subsequent 20-minute AngII infusion and redistributes the NHE3 to the base (Figure 4d).

rption, a surrogate measure of NHE3 transport, and also moved NHE3 to the bottom of the MV.41 TRV can also prevent the effects of AngII activation: Coinfusing the TRV compound with ACE inhibitors for 20 minutes blocks the effects of subsequent 20-minute AngII infusion and redistributes the NHE3 to the base (Figure 4d). Experimental AngII hypertension involves infusing AngII continuously at a dose that is initially subpressor but eventually provokes hypertension. Using immunoblots of the renal cortex, we determined in rats that after 3 days of AngII infusion (200 ng/kg/min), before blood pressure increase, the total abundance of cortical NHE3 increases around 50% above baseline,43 and then by 14 days of AngII infusion (at 400 ng/kg/min), when blood pressure is chronically elevated to 160 mm Hg, the NHE3 abundance is depressed to 20% below baseline.35 Figure 4e shows a proximal tubule from a 14-day AngII-infused rat with hypertension in which NHE3 is retained in the body of the MV.35 Images in Figure 4a and e were collected with the same settings, demonstrating that the ratio of NHE3 to villin signal decreased in the MV during AngII hypertension. Overall, these findings suggest that AngII “fixes” NHE3 in the MV and blunts redistribution, both acutely (Figure 3c) and chronically (Figure 4e). In lieu of redistribution, a compensatory decrease in NHE3 abundance in the MV becomes evident as blood pressure increases. The biphasic effects of AngII on proximal tubule reabsorption reported in rodents44 have led to the suggestion that the decreased NHE3 abundance may be due to high-dose AngII inhibition of NHE3, rather than hypertension. However, this inhibitory effect is observed when AngII was directly applied to tubules at doses of (>10–7 mol/l), and not observed during systemic AngII infusion (where the tubular AngII concentration only reaches the nanomoles per liter range).45 Interestingly, human proximal tubules studied in vitro only exhibit a stimulatory transport response to AngII.46

ved when AngII was directly applied to tubules at doses of (>10–7 mol/l), and not observed during systemic AngII infusion (where the tubular AngII concentration only reaches the nanomoles per liter range).45 Interestingly, human proximal tubules studied in vitro only exhibit a stimulatory transport response to AngII.46 Role of the Proximal Tubule in NHE3 Regulation in Experimental AngII Hypertension Blood vessels, the kidneys, and the central nervous system are all implicated in the genesis of experimental hypertension, and T-cells may provide a key link. Animal models of chronic hypertension exhibit increased immune infiltration into the vascular adventitia and kidney.47, 48, 49 Mice and rats lacking T-lymphocytes exhibit blunted hypertensive responses to experimental hypertension, restored by adoptive transfer of T-cells.50, 51 Evidence suggests that the initiating insult (whether angiotensin II infusion or other), increases nicotinamide adenine dinucleotide phosphate oxidase-mediated reactive oxygen species generation, which stimulates sympathetic nervous system activity and norepinephrine release in tissues, which can mediate tissue T-cell activation, producing local proinflammatory molecules (eg, reactive oxygen species, neoantigens, and cytokines).52, 53 In the kidney, these processes are reported to activate local accumulation of AngII, even when systemic levels of AngII are very low.54, 55 The local AngII is antinatriuretic, produces reactive oxygen species locally, and can attract more immune cells, creating a positive feedback loop that manifests as chronic inflammation and elevated blood pressure.49, 52 In rats, we determined that AngII-infusion hypertension activates distal transporters and channels, specifically, increasing abundance and phosphorylation of NKCC2 and Na+-Cl– cotransporters and/or activating proteolytic cleavage of epithelial sodium channels. In contrast, AngII hypertension suppresses proximal tubule and loop of Henle transporters, including NHE3, NaPiII, medullary NKCC2, and medullary Na,K-ATPase.35 In summary, the systemic AngII, inflammation, and intrarenal production of AngII stimulate distal transporters and contribute to hypertension, whereas the resultant hypertension counteracts the effects of the AngII and depresses proximal and loop of Henle sodium transport, facilitating pressure natriuresis.56

ATPase.35 In summary, the systemic AngII, inflammation, and intrarenal production of AngII stimulate distal transporters and contribute to hypertension, whereas the resultant hypertension counteracts the effects of the AngII and depresses proximal and loop of Henle sodium transport, facilitating pressure natriuresis.56 Many mouse genetic models have been subjected to experimental AngII hypertension and several exhibit a blunted hypertensive response to AngII infusion. We tested the hypothesis that we could identify the locus of the blunting of the hypertension along the nephron: Either reduced AngII stimulation of distal transporters or augmented pressure-natriuretic depression of transporter abundance. This section will focus specifically on the responses of NHE3. In most cases, an augmented suppression of NHE3 abundance accompanied the lower blood pressure, consistent with the notion that local AngII locks NHE3 in the MV and counters the responses to hypertension.

ure-natriuretic depression of transporter abundance. This section will focus specifically on the responses of NHE3. In most cases, an augmented suppression of NHE3 abundance accompanied the lower blood pressure, consistent with the notion that local AngII locks NHE3 in the MV and counters the responses to hypertension. Figure 5 summarizes the measurements of renal NHE3 abundance by immunoblot in wild type (WT) C57Bl/6J mice and genetically modified mice in response to 14-day AngII infusion (490 ng/kg/min). The WT and genetically modified samples were studied at the same time, and results are normalized to mean baseline results defined as 1.0. In the analyses of different sets of WT mice, we did not observe the significant suppression of NHE3 abundance that we measured in rats,35 nor did we detect evidence for AngII stimulation of NHE3; thus, the counteracting influences of hypertension and AngII may balance at the baseline levels in these WT mice. We have not yet analyzed the NHE3 distribution in the MV, nor have we determined whether NHE3 is stimulated during short-term AngII infusion. In mice with a specific genetic deletion of AT1R from the renal proximal tubule (ATIR genetically modified mice), generated by Gurley et al.,57 the baseline levels of NHE3 were unaltered, yet during AngII infusion NHE3 abundance was decreased 40%, associated with a 20-mm Hg lower blood pressure. In collaboration with the Harrison and Madhur groups58 we analyzed mice lacking the ability to synthesize the cytokine interleukin-17A or the cytokine interferon-γ during AngII hypertension. In both genetically modified mouse strains, AngII decreased abundance of NHE3 (25%-40%), myosin VI (25%), and NaPiII (50%) associated with 20- to 25-mm Hg lower blood pressure and improved natriuretic response to saline infusion. A subsequent study led by the Madhur lab59 demonstrated that the interleukin-17A strain had little or no increase in urinary albumin or angiotensinogen during AngII infusion, suggesting that interleukin-17A may reduce intrarenal AngII. Interestingly, the study also demonstrated that interleukin-17A stimulated NHE3 expression in cultured kidney cells mediated by serum and glucocorticoid regulated kinase 1 phosphorylation.Figure 5 In mouse genetic models that exhibit a blunted hypertensive response to angiotensin II (AngII) or Nω-nitro-l-arginine methyl ester hydrochloride (L-NAME) experimental hypertension, sodium hydrogen exchanger 3 (NHE3) abundance is significantly depressed.

rum and glucocorticoid regulated kinase 1 phosphorylation.Figure 5 In mouse genetic models that exhibit a blunted hypertensive response to angiotensin II (AngII) or Nω-nitro-l-arginine methyl ester hydrochloride (L-NAME) experimental hypertension, sodium hydrogen exchanger 3 (NHE3) abundance is significantly depressed. Models treated for 14 days with AngII infusion by osmotic minipumps include proximal tubule AT1R knockout (KO),57 mice with whole body KO of interleukin-17A,58 mice with whole body knockout of interferon-γ,58 and in mice with no kidney angiotensin converting enzyme (ACE 10/10).60 Immunoblots of a constant amount of renal homogenate are shown for each set of wild-type (WT) and KO mice infused with or without AngII. The last set shows the ACE 10/10 mice infused with L-NAME.55 These common findings in disparate models implicate a decrease in NHE3 abundance in the improved pressure natriuresis and blunted hypertension. Additionally, the findings suggest that elevated AngII, cytokines, and reactive oxygen species prevent the redistribution and/or decreased abundance of NHE3 in response to chronic sodium transport stimulation.

e models implicate a decrease in NHE3 abundance in the improved pressure natriuresis and blunted hypertension. Additionally, the findings suggest that elevated AngII, cytokines, and reactive oxygen species prevent the redistribution and/or decreased abundance of NHE3 in response to chronic sodium transport stimulation. With Gonzalez-Villalobos et al.60 we investigated the sodium transporter responses in mice that were engineered to express normal systemic ACE but no kidney ACE (ACE 10/10 mice). During AngII infusion, these mice filter and sense-infused AngII, but this AngII nor accompanying inflammatory cytokines cannot stimulate additional local intrarenal production of AngII. In AngII-infused ACE 10/10 mice blood pressure was about 20-mm Hg lower, yet this was not accompanied by a fall in NHE3 abundance. Rather, there was suppression of AngII activation of distal NKCC2 and Na+-Cl– cotransporters (NCC).60 This same strain was also subject to another distinct model of experimental hypertension caused by inhibition of nitric oxide synthase with the inhibitor Nω-nitro-l-arginine methyl ester hydrochloride (L-NAME). In WT mice L-NAME raises blood pressure 25 mm Hg; suppresses systemic AngII; and through local inflammation, stimulates intrarenal production of AngII. In the ACE 10/10 mice L-NAME did not raise blood pressure, but did significantly suppress NHE3 and sodium phosphate abundance by 50% and 30%, respectively.55 A third study in the ACE 10/10 mice demonstrated that this strain is resistant to salt-sensitive hypertension: After washout of the L-NAME treatment, blood pressure returned to normal in both genotypes, yet transporters remained lower in abundance in the ACE 10/10 mice. When both genotypes were subsequently fed a high-salt diet, the WT mice developed a salt-sensitive rise of 20 mm Hg in blood pressure, whereas the ACE 10/10 mice maintained baseline blood pressure.21 This differential response can be attributed to the higher inflammation and local production of AngII in the WT mice, and lower NHE3 in the ACE 10/10 mice. Taken together, the findings in these hypertension-resistant mouse models suggest that elevated AngII, cytokines, and/or reactive oxygen species maintain the NHE3 in the MV and blunt the redistribution or decreased abundance of NHE3 in response to sodium transport stimulation along the nephron.

3 in the ACE 10/10 mice. Taken together, the findings in these hypertension-resistant mouse models suggest that elevated AngII, cytokines, and/or reactive oxygen species maintain the NHE3 in the MV and blunt the redistribution or decreased abundance of NHE3 in response to sodium transport stimulation along the nephron. In other words, in both AngII and L-NAME hypertension, elevated local AngII production put the brakes on pressure natriuretic adjustments by activating transporters; thus, a further increase in pressure, and perhaps a different signaling path, is required to decrease transporters.

3 in the ACE 10/10 mice. Taken together, the findings in these hypertension-resistant mouse models suggest that elevated AngII, cytokines, and/or reactive oxygen species maintain the NHE3 in the MV and blunt the redistribution or decreased abundance of NHE3 in response to sodium transport stimulation along the nephron. In other words, in both AngII and L-NAME hypertension, elevated local AngII production put the brakes on pressure natriuretic adjustments by activating transporters; thus, a further increase in pressure, and perhaps a different signaling path, is required to decrease transporters. Summary and Future Directions Figure 6 provides a simplified overview of the connections among sodium transport stimulation by AngII; cytokines; reactive oxygen species; another important activator, renal sympathetic nerve stimulation; rise in effective circulating volume and blood pressure; and suppression of proximal and loop sodium transporters during hypertension. The results presented in this report suggest that, ultimately, the magnitude of hypertension is determined by the strength of the blood pressure signal(s) required to reduce proximal nephron sodium reabsorption enough to maintain effective circulating volume near baseline. Our conclusion is not intended to ignore the importance of the neuro and vascular aspects of hypertension, but to focus on renal sodium handling. Based on these findings, it is worth considering strategies to facilitate proximal tubule natriuresis as an approach to counteract the stimulatory influences of local AngII. Three candidate pathways appear promising.Figure 6 Simplified overview of the connections between sodium transport stimulation by angiotensin II (AngII); cytokines; reactive oxygen species (ROS); and another important activator, renal sympathetic nerve stimulation (RSNA); rise in effective circulating volume and blood pressure; and suppression of proximal and loop sodium transporters during hypertension. The local activation of proximal tubule NHE3 by intrarenal AngII, cytokines, ROS, or RSNA can blunt pressure natriuresis responses and raise blood pressure. Ultimately, the magnitude of hypertension is determined by the strength of the blood pressure signal required to reduce proximal nephron sodium reabsorption (by sodium hydrogen exchanger 3 redistribution to the base of the microvilli or depressed abundance) to restore effective circulating volume. The signals connecting hypertension to antinatriuresis are discussed in the text.

he strength of the blood pressure signal required to reduce proximal nephron sodium reabsorption (by sodium hydrogen exchanger 3 redistribution to the base of the microvilli or depressed abundance) to restore effective circulating volume. The signals connecting hypertension to antinatriuresis are discussed in the text. AT2R-Mediated Natriuresis AT1R and AT2R share similar affinity for AngII, yet AT2R stimulation counteracts the effects of AT1R by increasing bradykinin and nitric oxide release, reducing inflammation, promoting vasodilation and natriuresis.61, 62 AngIII is the likely ligand for AT2R.63 Hilliard et al.62, 64 showed that direct stimulation of AT2R with the selective agonist C21 increases natriuresis and diuresis without changes in GFR or RBF, evidence for tubular actions. The Carey group confirmed these findings in volume expanded female rats, also showing C21 may move NHE3 to the base of the MV, and that the natriuresis was dependent on nitric oxide and bradykinin.65 Hilliard et al.64 demonstrated that the lower blood pressure and more sensitive pressure natriuresis observed in WT female rodents is associated with 4-fold higher AT2R mRNA; and that this sex advantage disappears with age and in global AT2R KO mice.66 This fast-growing area full of therapeutic potential61, 62 has a large gap in knowledge about how AT2R signaling affects sodium transporters/channels and intrarenal renin angiotensin system in male and female humans.

4-fold higher AT2R mRNA; and that this sex advantage disappears with age and in global AT2R KO mice.66 This fast-growing area full of therapeutic potential61, 62 has a large gap in knowledge about how AT2R signaling affects sodium transporters/channels and intrarenal renin angiotensin system in male and female humans. Glucagon-Like Peptide–1-Mediated Natriuresis Glucagon like peptide-1 (GLP-1) is an incretin hormone constantly secreted from the intestine at low basal levels in the fasted state. Plasma concentrations rise rapidly after nutrient ingestion. Upon release, GLP-1 exerts insulinotropic effects via a G protein-coupled receptor, stimulation of adenylyl cyclase, and cyclic adenosine monophosphate generation. Although primarily involved in glucose homeostasis, GLP-1 can induce diuresis and natriuresis when administered in pharmacologic doses in humans and rodents. Carraro-Lacroix et al.67 and Crajoinas et al.68 defined the chronic effects of stimulation of the incretin receptor GLP-1 in kidney and discovered that GLP-1 has diuretic and natriuretic effects mediated by changes in both renal hemodynamic parameters and by downregulation of proximal tubule NHE3 activity.68 Recently, Farah et al.69 demonstrated that endogenous baseline GLP-1 plays a significant role in regulating renal function. They blocked the GLP-1 receptor with the antagonist exendin-9 in overnight-fasted anesthetized rats. Exendin–9-infused (30 minutes) rats exhibited reduced GFR, lithium clearance, urinary volume flow, and sodium excretion compared with vehicle-infused controls. NHE3 phosphorylation at a site associated with retraction to the base was also increased. Collectively, these results provide novel evidence that GLP-1 is a physiologically relevant natriuretic factor that contributes to sodium balance, in part, via tonic modulation of sodium transport activity in the proximal tubule.

NHE3 phosphorylation at a site associated with retraction to the base was also increased. Collectively, these results provide novel evidence that GLP-1 is a physiologically relevant natriuretic factor that contributes to sodium balance, in part, via tonic modulation of sodium transport activity in the proximal tubule. Sodium-Glucose-Cotransporter 2 Inhibitors The sodium-glucose-cotransporter inhibitors used to treat diabetes directly target the proximal tubule, provoke natriuresis and diuresis, and may lower blood pressure.70 Pessoa et al.71 recently reported that NHE3 activity is stimulated by luminal glucose, and that NHE3 colocalizes and may functionally interact with sodium-glucose-cotransporter 2 in the proximal tubule. Thus, it is possible that sodium-glucose-cotransporter 2 inhibition may inhibit NHE3 transport activity.

pressure.70 Pessoa et al.71 recently reported that NHE3 activity is stimulated by luminal glucose, and that NHE3 colocalizes and may functionally interact with sodium-glucose-cotransporter 2 in the proximal tubule. Thus, it is possible that sodium-glucose-cotransporter 2 inhibition may inhibit NHE3 transport activity. Dopamine Receptor-Mediated Natriuresis The intrarenal dopaminergic system is an important determinant of the blood pressure set point: Reduced dopamine signaling is associated with hypertension.72, 73, 74 There are 5 dopamine receptors in the kidney: D1R and D5R (D1likeR) physically interact and their activation inhibits NHE3 and Na,K-ATPase75; thus, counteracting the effects of AngII via AT1R.63 D1likeR activation increases salt and water excretion mediated by inhibition of proximal tubule NHE3, sodium phosphate, and Na,K-ATPase as well as loop of Henle NKCC2, the proximal sites that can elicit pressure natriuresis.76 Additionally, dopamine receptor activation is known to antagonize AT1R signaling, reduce intrarenal renin angiotensin system components,63, 76, 77 and stimulate AT2R signaling.78 Numerous labs have provided evidence for functional complexes between Ang AT1R and DA D1R79 in which activation of one attenuates the expression of the other. Likewise, AT2Rs may oppose AT1Rs by protein–protein interaction.63 Thus, effects of dopamine deficiency include AT1R activation and vice versa.

AT2R signaling.78 Numerous labs have provided evidence for functional complexes between Ang AT1R and DA D1R79 in which activation of one attenuates the expression of the other. Likewise, AT2Rs may oppose AT1Rs by protein–protein interaction.63 Thus, effects of dopamine deficiency include AT1R activation and vice versa. Determining how the pressor and natriuretic arms of the renin angiotensin system, in parallel with the interacting dopaminergic system, regulate transporters and channels may fill important gaps in understanding the sexual dimorphism of blood pressure and provide new and sex-specific therapeutic approaches to treat resistant hypertension. Disclosure The author declared no competing interests. Acknowledgments Arvid Maunsbach generated the proximal tubule electron micrograph in Figure 2. This work was supported by National Institutes of Health National Institute of Diabetes and Digestive and Kidney Diseases grant No. R01DK083785 and AHA Grant in Aid Western States Affiliate (No. 15GRNT23160003).

with the clinical features, supported an active immune complex–mediated glomerulonephritis with a lupus-like pattern of immune complex deposition. There was also a background of moderate glomerular and tubulointerstitial scarring, possibly from chronic injury from hepatorenal syndrome (Table 4 and Figure 2b, e, and h). She was treated with methylprednisolone 250 mg i.v. for 3 days, followed by prednisone, 60 mg daily (1 mg/kg), and mycophenolate mofetil, 500 mg twice daily. Her renal function rapidly improved and serum creatinine decreased to 1.40 mg/dl within 3 days. Significant delirium and agitation developed, and the dose of prednisone was rapidly decreased to 20 mg daily and then tapered by 5 mg/wk. Because of the tubulointerstitial inflammation, trimethoprim/sulfamethoxazole and allopurinol were permanently discontinued. At time of discharge 1 week after kidney biopsy, the serum creatinine level improved to 1.0 mg/dl and then returned to 1.4 mg/dl once diuretics were reintroduced. Delirium resolved and the patient reported improvement in joint pain and increased mobility. Unfortunately, the HCV viral load became detectable 4 weeks after treatment, indicating relapse. Three months later, she began receiving on SOF, 400 mg, plus ledipasvir, 90 mg, fixed-dose combination therapy with the addition of RBV, 200 mg, every alternate day for 12 weeks. However, approximately 8 weeks into therapy, her joint pain relapsed with prominent synovitis in the bilateral wrists, metacarpal joints, and knees, and she required a course of prednisone therapy starting at 40 mg and tapered by 10 mg every 2 weeks. Mycophenolate mofetil was discontinued. Azathioprine was substituted as a steroid-sparing agent; however, her joint pain continued, so she was switched back to mycophenolate mofetil (Figure 2b). By 12 weeks after treatment, her arthritis improved and there was no evidence of relapse of glomerulonephritis. Her viral load was negative, indicating cure of HCV infection. She continues to receive mycophenolate mofetil, 500 mg twice daily, and low-dose prednisone that is slowly being tapered. She is without joint pain or worsening renal function.

Introduction It is estimated that between 2.2 and 3.2 million people are infected with hepatitis C virus (HCV) in the United States.1 For more than 20 years, the mainstay of HCV treatment has been interferon (IFN) alfa, a nonspecific immunomodulatory antiviral cytokine, and ribavirin (RBV). This regimen had limited efficacy and was poorly tolerated. Because of this, therapy was not widely deployed and the majority of patients with HCV infection remained untreated.2 In December 2013, sofosbuvir (SOF), an inhibitor of the HCV NS5B RNA polymerase, was approved by the U.S. Food and Drug Administration, and thus began an era of all-oral IFN-free HCV therapies. Currently approved direct-acting antiviral (DAA) therapies target 1 of 3 viral proteins, the NS5B polymerase, the NS3/4A protease, or the NS5A protein (Table 1). These agents profoundly inhibit viral replication and, when used in combination, can produce viral clearance without dependence on IFN.3Table 1 Currently FDA-approved direct-acting antiviral agents NS5B polymerase inhibitors NS3/4A protease inhibitors NS5A inhibitors Sofosbuvir Simeprevir Daclatasvir Dasabuvir Paritaprevir Grazoprevir Ledipasvir Ombitasvir Elbasvir FDA, U.S Food and Drug Administration.

Introduction It is estimated that between 2.2 and 3.2 million people are infected with hepatitis C virus (HCV) in the United States.1 For more than 20 years, the mainstay of HCV treatment has been interferon (IFN) alfa, a nonspecific immunomodulatory antiviral cytokine, and ribavirin (RBV). This regimen had limited efficacy and was poorly tolerated. Because of this, therapy was not widely deployed and the majority of patients with HCV infection remained untreated.2 In December 2013, sofosbuvir (SOF), an inhibitor of the HCV NS5B RNA polymerase, was approved by the U.S. Food and Drug Administration, and thus began an era of all-oral IFN-free HCV therapies. Currently approved direct-acting antiviral (DAA) therapies target 1 of 3 viral proteins, the NS5B polymerase, the NS3/4A protease, or the NS5A protein (Table 1). These agents profoundly inhibit viral replication and, when used in combination, can produce viral clearance without dependence on IFN.3Table 1 Currently FDA-approved direct-acting antiviral agents NS5B polymerase inhibitors NS3/4A protease inhibitors NS5A inhibitors Sofosbuvir Simeprevir Daclatasvir Dasabuvir Paritaprevir Grazoprevir Ledipasvir Ombitasvir Elbasvir FDA, U.S Food and Drug Administration. DAA regimens offer significantly increased efficacy, shorter duration of therapy, and dramatically improved side effect profiles. Real-world data demonstrate cure rates of more than 90% with combination therapy consisting of SOF and simeprevir for genotype 1 infections, which is similar to the results from the phase III clinical trials.4, 5, 6 The IFN-free DAA combinations of SOF plus simeprevir and SOF plus RBV are extremely well tolerated both in clinical trials and in real-world patient experiences, with the most common side effects being headache, fatigue, and nausea.6 Mild anemia was observed only in regimens containing RBV.

III clinical trials.4, 5, 6 The IFN-free DAA combinations of SOF plus simeprevir and SOF plus RBV are extremely well tolerated both in clinical trials and in real-world patient experiences, with the most common side effects being headache, fatigue, and nausea.6 Mild anemia was observed only in regimens containing RBV. Autoimmune phenomena are well described sequela of IFN-based therapies for HCV infection7, 8, 9; IFN may aggravate preexisting autoimmunity, unmask previously silent autoimmune processes, or even cause the emergence of de novo autoimmune disease.8, 10 The mechanism is thought to be mediated by the effect of IFN on modulation of the host’s immune system or as a result of molecular mimicry.11 However, autoimmune syndromes are not currently described in patients receiving all-oral IFN-free DAA therapies. These regimens directly target viral replication and therefore have not been considered “immunomodulatory.” In this series we report on 3 cases of patients undergoing HCV therapy with SOF-based DAA therapy in whom a biopsy-proven lupus-like immune complex–mediated glomerulonephritis developed.

l IFN-free DAA therapies. These regimens directly target viral replication and therefore have not been considered “immunomodulatory.” In this series we report on 3 cases of patients undergoing HCV therapy with SOF-based DAA therapy in whom a biopsy-proven lupus-like immune complex–mediated glomerulonephritis developed. Case Presentation Case 1 The first patient is a 51-year-old white female with a medical history significant for HCV genotype 3a infection complicated by cirrhosis with a Model for End-Stage Liver Disease (MELD) score of 10, Child-Pugh class A, who was referred for treatment of HCV infection (Figure 1a and Tables 2 and 3). Her baseline kidney function was normal; her serum creatinine level was 0.72 mg/dl, and her estimated glomerular filtration rate was >60 ml/min per 1.73 m2. Cirrhosis was diagnosed on the basis of a liver biopsy and previously failed attempts to eradicate HCV with pegylated IFN and RBV in 2002. The patient began antiviral therapy with SOF, 400 mg daily, and weight-based RBV (1000 mg daily) for a total of 24 weeks. Her other medications were spironolactone, 50 mg daily, and furosemide, 20 mg taken as needed for edema. During the course of antiviral therapy, she experienced mild nausea and diarrhea. Otherwise, no other adverse effects were noted. Twelve weeks after treatment she achieved sustained virological response with an undetectable viral load. Three months after completing treatment, she presented to her primary doctor with complaints of rash and pruritus of the arms, abdomen, and thighs. The rash was morbilliform with papules involving both thighs. There were no pustules, vesicles, or ulcerations noted on physical examination. Her liver function test results were elevated, with an alanine aminotransferase level of 1194 U/l, aspartate aminotransferase level of 1608 U/l, alkaline phosphatase level of 179 U/l, and total bilirubin of 6.0 U/l. She was admitted to the hospital and a liver biopsy was performed; it demonstrated panlobular hepatitis against a background of cirrhosis. The differential diagnosis of this histopathologic lesion included acute viral injury, immune-mediated drug reaction, and autoimmune hepatitis. Her serologic work-up results were abnormal, with a positive antinuclear antibody test result (a titer of 1:320 titer) and positive smooth muscle antibody test result (a titer of 1:80). The results of all available serological tests before and after treatment are shown in Table 3.

ug reaction, and autoimmune hepatitis. Her serologic work-up results were abnormal, with a positive antinuclear antibody test result (a titer of 1:320 titer) and positive smooth muscle antibody test result (a titer of 1:80). The results of all available serological tests before and after treatment are shown in Table 3. Tests for hepatitis A virus, hepatitis B virus, hepatitis E virus, cytomegalovirus, Epstein–Barr virus and HCV RNA were performed and all the results were negative. The episode of acute hepatitis began to resolve spontaneously. While she was hospitalized, however, her serum creatinine level began to rise, with 3+ proteinuria detected by urinary dipstick. She had hematuria with dysmorphic red blood cells and cellular casts containing leukocytes and erythrocytes on microscopic examination of her urine.Figure 1 Creatinine trend over time in cases 1, 2, and 3. The clinical course of the first (a), second (b), and third (c) cases with immune complex–mediated glomerulonephritis are shown. The time during which the patient was receiving the initial course of direct-acting antiviral (DAA) is shown in shaded gray. AZA, azathioprine; CVVH, continuous venovenous hemofiltration; ICU, intensive care unit; LDV, ledipasvir; MMF, mycophenolate mofetil; Pred, prednisone; RBV, ribavirin; SIM, simeprevir; SOF, sofosbuvir. Table 2 Summary of cases of lupus-like immune complex glomerulonephritis Case Patient characteristics Clinical presentation Immunosuppression Outcome 1 • 51-year-old Asian female • Genotype 3a, viral load 3.6 × 105 • Treatment experienced (PEG-IFN/RBV)

Tests for hepatitis A virus, hepatitis B virus, hepatitis E virus, cytomegalovirus, Epstein–Barr virus and HCV RNA were performed and all the results were negative. The episode of acute hepatitis began to resolve spontaneously. While she was hospitalized, however, her serum creatinine level began to rise, with 3+ proteinuria detected by urinary dipstick. She had hematuria with dysmorphic red blood cells and cellular casts containing leukocytes and erythrocytes on microscopic examination of her urine.Figure 1 Creatinine trend over time in cases 1, 2, and 3. The clinical course of the first (a), second (b), and third (c) cases with immune complex–mediated glomerulonephritis are shown. The time during which the patient was receiving the initial course of direct-acting antiviral (DAA) is shown in shaded gray. AZA, azathioprine; CVVH, continuous venovenous hemofiltration; ICU, intensive care unit; LDV, ledipasvir; MMF, mycophenolate mofetil; Pred, prednisone; RBV, ribavirin; SIM, simeprevir; SOF, sofosbuvir. Table 2 Summary of cases of lupus-like immune complex glomerulonephritis Case Patient characteristics Clinical presentation Immunosuppression Outcome 1 • 51-year-old Asian female • Genotype 3a, viral load 3.6 × 105 • Treatment experienced (PEG-IFN/RBV) • Cirrhosis diagnosed by biopsy • Morbilliform rash • Elevated liver function test results • Rising creatinine level • Proteinuria and hematuria • Methylprednisolone, 500 mg i.v. × 3 d, followed by prednisone, 60 mg daily • Rapid improvement in renal function • Achieved SVR12 • Death from sepsis, fungal pneumonia, and empyema 2 • 56-year-old white female

• Cirrhosis diagnosed by biopsy • Morbilliform rash • Elevated liver function test results • Rising creatinine level • Proteinuria and hematuria • Methylprednisolone, 500 mg i.v. × 3 d, followed by prednisone, 60 mg daily • Rapid improvement in renal function • Achieved SVR12 • Death from sepsis, fungal pneumonia, and empyema 2 • 56-year-old white female • Genotype 1b, viral load 2.8 × 106 • Treatment experienced (PEG-IFN/RBV) • Cirrhosis diagnosed by allograft biopsy • Joint pain • Rising creatinine level • Hematuria and leukocyturia • Methylprednisolone, 250 mg i.v. × 3 d, followed by prednisone, 60 mg daily • Prednisone rapidly decreased to 20 mg daily because of delirium, then tapered off over the next 4 wk • MMF maintenance therapy, 250 mg twice daily × 5 mo • Rapid improvement in renal function • Relapse of HCV viremia, retreated with sofosbuvir and ledipasvir • Relapse of joint pain with retreatment, steroids and MMF restarted with improvement • Achieved SVR12 3 • 47-year-old Hispanic female • Genotype 1, viral load 8.4 × 104 • Treatment experienced (IFNα) • Cirrhosis clinically diagnosed • Rash (face, chest, and hands) • Back pain, fatigue • Rising creatinine level • Proteinuria and hematuria • Methylprednisolone, 250 mg i.v. × 3 d, followed by prednisone, 60 mg daily • Rituximab, 1000 mg i.v. weekly x 3 • Rapid improvement in renal function • Achieved SVR12 • Death from bacteremia, bacterial pneumonia

• Cirrhosis clinically diagnosed • Rash (face, chest, and hands) • Back pain, fatigue • Rising creatinine level • Proteinuria and hematuria • Methylprednisolone, 250 mg i.v. × 3 d, followed by prednisone, 60 mg daily • Rituximab, 1000 mg i.v. weekly x 3 • Rapid improvement in renal function • Achieved SVR12 • Death from bacteremia, bacterial pneumonia d, day(s); HCC, hepatocellular carcinoma; HCV, hepatitis C virus; IFNα, interferon alfa; LT, liver transplant; MMF, mycophenolate mofetil; PEG-IFN, pegylated interferon; mo, month(s); RBV, ribavirin; SVR12, sustained virological response at 12 weeks; wk, week(s). Table 3 Comparison of autoimmune serologic findings before and after DAA treatment

d, day(s); HCC, hepatocellular carcinoma; HCV, hepatitis C virus; IFNα, interferon alfa; LT, liver transplant; MMF, mycophenolate mofetil; PEG-IFN, pegylated interferon; mo, month(s); RBV, ribavirin; SVR12, sustained virological response at 12 weeks; wk, week(s). Table 3 Comparison of autoimmune serologic findings before and after DAA treatment Case Baseline serologic findings Post-DAA serologic findings 1 Cryoglobulin: negative Remainder not available ANA: 1:320 Ds-DNA: negative Anti-histone: ND Anti-RNP: negative Anti-Smith: negative Complement levels C3: 74 (low) C4: 14 Cryoglobulin: negative Rheumatoid factor: Negative Anti–smooth muscle Ab: 1:80 ANCA: Negative 2 ANA: 1:640 in 2007 Cryoglobulin: negative Rheumatoid factor: Positive in 2007 Remainder not available ANA: >1:5120 Ds-DNA: Positive 1:80 Anti-histone: >7.0 Anti-Ro/La: negative Anti-Smith: negative Complements levels C3: 54 C4: 9 (low) Cryoglobulin: negative Rheumatoid factor: negative 3 ANA: 1:160 Ds-DNA: negative Anti Ro/La: negative Anti-Smith: negative Cryoglobulin level: 5% Complement levels C3: 87 C4: 20 (normal) Rheumatoid factor: negative Remainder not available ANA: 1:160 Ds-DNA: Negative Anti-histone: 1.5 Units Anti Ro/La: positive Anti-Smith: Negative Cryoglobulin level: 1% Complement levels C3: 42 (low) C4: 7 (low) Rheumatoid factor: negative Ab, antibody; ANCA, antineutrophil cytoplasmic antibody; ANA, antinuclear antibody; DAA, direct-acting antiviral; Ds-DNA, double-stranded DNA; ND, not done; RNP, ribonucleoprotein.

its Anti Ro/La: positive Anti-Smith: Negative Cryoglobulin level: 1% Complement levels C3: 42 (low) C4: 7 (low) Rheumatoid factor: negative Ab, antibody; ANCA, antineutrophil cytoplasmic antibody; ANA, antinuclear antibody; DAA, direct-acting antiviral; Ds-DNA, double-stranded DNA; ND, not done; RNP, ribonucleoprotein. Normal range for complements levels: C3, 81–157 mg/dl; C4, 12–39 mg/dl. Normal range for antihistone antibody, <1.0 U. Her blood urea nitrogen level peaked at 59 mg/dl and her serum creatinine level peaked at 1.65 mg/dl. The results of cryocrit, double-stranded DNA, and rheumatoid factor tests were all negative. Her complement C3 level was 74 mg/dl (reference range 81–157 mg/dl), and her C4 level was 14 mg/dl (reference range 12–39 mg/dl).

blood urea nitrogen level peaked at 59 mg/dl and her serum creatinine level peaked at 1.65 mg/dl. The results of cryocrit, double-stranded DNA, and rheumatoid factor tests were all negative. Her complement C3 level was 74 mg/dl (reference range 81–157 mg/dl), and her C4 level was 14 mg/dl (reference range 12–39 mg/dl). She underwent an ultrasound-guided renal biopsy that showed 19 nonsclerotic glomeruli, all showing marked endocapillary hypercellularity characterized by endothelial cell swelling, intracapillary neutrophils and macrophages, apoptotic debris, and mild mesangial hypercellularity. A single pseudothrombus was seen in 1 glomerulus. This was associated with moderate tubulointerstitial nephritis. Rare red cell casts were seen. Focal vasculitis was present. There was no glomerulosclerosis or chronic vascular disease. Interstitial fibrosis and tubular atrophy were minimal. Immunofluorescence showed granular mesangial and glomerular basement membrane (GBM) staining for IgG, IgA, IgM, C3, C1q, and kappa and lambda light chains. Immunoflourescent staining of the tubular basement membranes and vessels were negative. Many mesangial, paramesangial, and subendothelial electron-dense deposits without substructure were present. Tubuloreticular inclusions were seen. These features were consistent with an acute immune complex glomerulonephritis with a lupus-like pattern of immune complex deposition (Table 4 and Figure 2a, d, and g).Figure 2 Light microscopy of the 3 cases of immune complex–mediated glomerulonephritis. The cases show varying degrees of glomerular activity from (a) global endocapillary hypercellularity with endothelial cell swelling, intracapillary mononuclear cells, neutrophils, and apoptotic debris (case 1) and (b) segmental endocapillary hypercellularity and moderate mesangial hypercellularity (case 2) to (c) global endocapillary and mesangial hypercellularity with double contours (case 3). A full house immunofluorescence pattern was present in all cases (case 3 had only minimal C1q). (d) Granular glomerular basement membrane (GBM) and segmental mesangial staining for C3 (case 1). (e) Granular mesangial and segmental GBM staining for IgG (case 2). (f) Global granular mesangial and GBM staining for IgG (case 3). Electron microscopy showing (g) subendothelial and mesangial electron-dense deposits and activated endothelium in a glomerular capillary loop (case 1) and (h,i) mesangial and paramesangial deposits (cases 2 and 3, respectively).

staining for IgG (case 2). (f) Global granular mesangial and GBM staining for IgG (case 3). Electron microscopy showing (g) subendothelial and mesangial electron-dense deposits and activated endothelium in a glomerular capillary loop (case 1) and (h,i) mesangial and paramesangial deposits (cases 2 and 3, respectively). Table 4 Summary of pathologic characteristics Parameter Case 1 Case 2 Case 3 Light microscopy Number of glomeruli 19 9 33 Global glomerulosclerosis 0% 67% 9% Segmental glomerulosclerosis 0% 0% 0% Endocapillary hypercellularity +++ ++ +++ Pseudothrombi + – – Mesangial hypercellularity + ++ ++ Segmental GBM duplication + + +++ Crescents – – – Fibrosis and tubular atrophy <5% ∼50% 20% Interstitial inflammation + + ± Tubular injury ± + + Vascular disease ± ± ++ Vasculitis + hilar arteriole – – Immunofluorescence microscopy results Glomeruli IgG ++ ++ +++ IgA ++ +++ ++++ IgM + ++ ++ C3 +++ ++ +++ C1q ++ +++ ± Fibrin – – +++ Kappa light chain ++ +++ +++ Lambda light chain ++ +++ ++++ Tubules – Focal, broad linear IgG and granular C3 – Interstitium – – – Vessels – focal C3 focal C3 Electron microscopy Foot process effacement segmental global segmental Electron-dense deposits mesangial, paramesangial, subendothelial mesangial, paramesangial, subendothelial, intramembranous mesangial, subendothelial, and intramembranous Substructure – – – Tubuloreticular inclusions + +++ – GBM duplication ± + +++ Tubular basement membrane deposits – – – GBM, glomerular basement membrane. In case 3, ± staining for C1q indicates present but minimal staining.

Parameter Case 1 Case 2 Case 3 Light microscopy Number of glomeruli 19 9 33 Global glomerulosclerosis 0% 67% 9% Segmental glomerulosclerosis 0% 0% 0% Endocapillary hypercellularity +++ ++ +++ Pseudothrombi + – – Mesangial hypercellularity + ++ ++ Segmental GBM duplication + + +++ Crescents – – – Fibrosis and tubular atrophy <5% ∼50% 20% Interstitial inflammation + + ± Tubular injury ± + + Vascular disease ± ± ++ Vasculitis + hilar arteriole – – Immunofluorescence microscopy results Glomeruli IgG ++ ++ +++ IgA ++ +++ ++++ IgM + ++ ++ C3 +++ ++ +++ C1q ++ +++ ± Fibrin – – +++ Kappa light chain ++ +++ +++ Lambda light chain ++ +++ ++++ Tubules – Focal, broad linear IgG and granular C3 – Interstitium – – – Vessels – focal C3 focal C3 Electron microscopy Foot process effacement segmental global segmental Electron-dense deposits mesangial, paramesangial, subendothelial mesangial, paramesangial, subendothelial, intramembranous mesangial, subendothelial, and intramembranous Substructure – – – Tubuloreticular inclusions + +++ – GBM duplication ± + +++ Tubular basement membrane deposits – – – GBM, glomerular basement membrane. In case 3, ± staining for C1q indicates present but minimal staining. She began receiving on i.v. methylprednisolone, 500 mg daily for 3 days, and was then transitioned to prednisone 60 mg (1 mg/kg). This resulted in the normalization of her serum creatinine level (0.8 mg/dl) over the next 3 weeks. At the same time, however, she indicated that she had begun feeling confused and anxious; she was found to have a plasma ammonia level of 165 μmol/l and was admitted to the hospital for treatment of hepatic decompensation with lactulose and rifaximin, resulting in resolution of encephalopathy. During this hospitalization, ascites and a urinary tract infection were diagnosed, and the patient was subsequently treated with ceftriaxone and discharged to a rehabilitation facility. Over the next 4 days leukocytosis, hyponatremia, decreased urinary output, abdominal pain, and rising serum creatinine level (a rise from 0.77 mg/dl to 0.95 mg/dl) developed, and she was transferred back to the hospital. She was again given ceftriaxone for presumed urinary tract infection due to hematuria and leukocyturia, although urine cultures demonstrated no growth. During her hospital stay, she had persistent leukocytosis and a chest radiograph revealed a right lower lobe opacity for which she began treatment with i.v. vancomycin and piperacillin-tazobactam; prednisone was rapidly tapered. Her serum creatinine level began to rise and peaked at 1.3 mg/dl. Vancomycin was stopped on account of elevated trough values (37μg/ml [upper limit of normal 20 μg/ml]), and her creatinine level stabilized. She was discharged to a rehabilitation hospital. However, 48 hours later she was readmitted to the medical intensive care unit with hypoxic reparatory failure and septic shock due to pneumonia with empyema and oliguric acute kidney injury with a serum creatinine level of 2.34 mg/dl secondary to presumed acute tubular necrosis. She was intubated and placed on continuous venovenous hemofiltration on account of acidosis and worsening volume overload. Cultures from a bronchial alveolar lavage and of the pleural fluid revealed growth of Aspergillus fumigatus, and amphotericin was started; however, she died the following day.

ute tubular necrosis. She was intubated and placed on continuous venovenous hemofiltration on account of acidosis and worsening volume overload. Cultures from a bronchial alveolar lavage and of the pleural fluid revealed growth of Aspergillus fumigatus, and amphotericin was started; however, she died the following day. Case 2 The second patient is a 56-year-old white female with HCV genotype 1b infection who had received a liver transplant because of hepatocellular carcinoma in 2009 (Figure 1b and Tables 2 and 3). Before receiving the transplant, she was treated with pegylated IFN and RBV with an initial response followed by a relapse after completion of therapy. Cirrhosis developed in the transplanted liver and was complicated by chronic ascites and hepatic encephalopathy that were controlled medically. Her Model for End-Stage Liver Disease score was 17, Child-Pugh class B. At baseline she had chronic kidney disease stage 3 with a serum creatinine ranging from 1.4 to 1.7 mg/dl and an estimated glomerular filtration rate of 31 to 38 ml/min per 1.73m2 without significant proteinuria, hematuria, or leukocyturia detected on pretreatment urinalysis. Her renal insufficiency had been attributed to chronic calcineurin use and decompensated cirrhosis (hepatorenal syndrome type II), although a biopsy had not been performed. She began therapy with SOF, 400 mg daily, and RBV, 200 mg twice daily (dose reduced for her renal dysfunction), for 24 weeks. Her other medications included allopurinol 100, mg twice daily; trimethoprim/sulfamethoxazole, single-strength tablet once daily; bumetanide, 2 mg twice daily; colchicine; 0.6 mg daily; cyclosporine, 25 mg twice daily; erythropoietin, 2000 units weekly; labetalol. 200 mg 3 times per day; lactulose, 30 ml 3 times per day; magnesium oxide, 800 mg twice daily; metolazone, 5 mg daily; omeprazole, 20 mg twice daily; potassium chloride extended release, 40 milliequivalents daily; and rifaximin, 550 mg twice per day.

ine, 25 mg twice daily; erythropoietin, 2000 units weekly; labetalol. 200 mg 3 times per day; lactulose, 30 ml 3 times per day; magnesium oxide, 800 mg twice daily; metolazone, 5 mg daily; omeprazole, 20 mg twice daily; potassium chloride extended release, 40 milliequivalents daily; and rifaximin, 550 mg twice per day. Her viral load was undetectable by week 4 of treatment and remained undetected for the duration of therapy. Her serum creatinine fluctuated between 1.2 and 1.9 mg/dl throughout therapy. Her RBV was dose reduced to 200 mg once daily because of development of anemia. While receiving treatment she complained of new onset of joint pain involving her hands and wrists, decreased mobility, and worsening lower extremity edema and ascites. Two weeks after completing therapy, she was referred to a nephrologist when it was discovered that her serum creatinine level had risen to 2.6 mg/dl and her urinalysis showed leukocytosis and trace proteinuria on dipstick. Urine microscopy was notable for the presence of white blood cell casts. A serologic profile was notable for a positive antinuclear antibody test result with a titer >1:5120 and a histone antibody test result positive at >7.0 Units and double-stranded DNA antibody positive at a 1:10 titer. Her complement levels were low; her C3 level was 54 (reference range 81–157 mg/dl) and her C4 level was was 9 (reference range 12–39 mg/dl). Cryoglobulins were not detected in her serum. The results of all available serological tests before and after treatment are shown in Table 3.

NA antibody positive at a 1:10 titer. Her complement levels were low; her C3 level was 54 (reference range 81–157 mg/dl) and her C4 level was was 9 (reference range 12–39 mg/dl). Cryoglobulins were not detected in her serum. The results of all available serological tests before and after treatment are shown in Table 3. She underwent a renal biopsy that showed mild endocapillary hypercellularity characterized by intracapillary mononuclear cells and swollen endothelial cells and moderate mesangial hypercellularity. Focal GBM duplication was present. There was moderate tubulointerstitial inflammation. Leukocyte and granular casts were seen. There was no vasculitis. Immunofluorescence showed granular mesangial and GBM staining for IgG, IgA, IgM, C3, C1q, and kappa and lambda light chains. Mesangial, paramesangial, subendothelial, and intramembranous electron-dense deposits without substructure were present. Many tubuloreticular inclusions were seen. These findings, along with the clinical features, supported an active immune complex–mediated glomerulonephritis with a lupus-like pattern of immune complex deposition. There was also a background of moderate glomerular and tubulointerstitial scarring, possibly from chronic injury from hepatorenal syndrome (Table 4 and Figure 2b, e, and h).

r arthritis improved and there was no evidence of relapse of glomerulonephritis. Her viral load was negative, indicating cure of HCV infection. She continues to receive mycophenolate mofetil, 500 mg twice daily, and low-dose prednisone that is slowly being tapered. She is without joint pain or worsening renal function. Case 3 The third patient is a 48-year-old Hispanic female with a medical history of HCV genotype 1 infection and cirrhosis who was initiated on a 12-week course of DAA therapy with SOF, 400 mg ,and simeprevir, 150 mg daily (Figure 1c and Tables 2 and 3). She had contracted HCV infection at age 18 and her medical history included idiopathic thrombocytopenic purpura treated by splenectomy in her mid-20s and systemic lupus erythematous diagnosed in 1998 (at age 31) that had been precipitated by IFN alfa use for HCV infection. Lupus manifestations included hemolytic anemia; arthritis of the hands, knees, and ankles; and central nervous system manifestations, including headache, vertigo, and diplopia. Initially, she had multiple flares and had been treated with azathioprine, hydroxychloroquine, and prednisone until 2009. At that point, because of disease quiescence, prednisone and azathioprine were discontinued and she continued to receive hydroxychloroquine, 200 mg daily. She continued to have quiescent disease and did not experience a lupus flare between 2009 and when DAAs were started in 2014. A diagnosis of cirrhosis was made clinically on the basis of recent imaging showing a nodular liver. She was listed for a liver transplant and had a Model for End-Stage Liver Disease score of 11, Child-Pugh class B, before beginning DAA therapy. Her baseline serum creatinine level was 0.95 mg/dl, which corresponded to an estimated glomerular filtration rate of 67 ml/min per 1.73m2. Her other medications included aspirin, 81 mg daily, clonidine, 0.1 mg twice per day, folic acid, 1 mg daily, furosemide, 80 mg daily, hydroxychloroquine, 200 mg daily, spironolactone, 50 mg daily, and valsartan 320, mg daily.

mg/dl, which corresponded to an estimated glomerular filtration rate of 67 ml/min per 1.73m2. Her other medications included aspirin, 81 mg daily, clonidine, 0.1 mg twice per day, folic acid, 1 mg daily, furosemide, 80 mg daily, hydroxychloroquine, 200 mg daily, spironolactone, 50 mg daily, and valsartan 320, mg daily. Two weeks into DAA treatment a triangular dark red plaque on the upper chest with hemorrhagic small bulla on the inferior edge, and erythema and erosions around the nose and cheeks developed along with erythematous and edematous plaques on the lateral part of the wrists and thumbs after a brief exposure to the sun, and she was treated with topical steroids. During the fourth week of DAA treatment, she complained of several days of increased somnolence, fatigue, lower abdominal and back discomfort, and dark urine. Laboratory studies demonstrated a rise in serum creatinine level to 1.43 mg/dl and her urinalysis showed red blood cells, white blood cells, and hyaline casts. She was referred to the emergency department, where she was treated empirically with i.v. hydration and ciprofloxacin for presumed pyelonephritis, although the results of a urine culture were later negative. Her serum creatinine level decreased from 1.43 mg/dl to 1.11 mg/dl and she was discharged with a 10-day prescription of ciprofloxacin. Furosemide and spironolactone were temporarily discontinued. Four weeks later, after a total of 9 weeks of DAA therapy she again presented to the hepatologist with back pain and dark urine. Laboratory studies were notable for hyperbilirubinemia (5.4 mg/dl), prolonged prothrombin time (international normalized ratio 1.7), hypoalbuminemia (1.8 g/dl), and a serum creatinine level of 1.45 mg/dl. SOF and simeprevir were discontinued at this time and she was treated with 2 additional courses of antibiotics for possible urinary tract infection. Because of persistent elevation of her serum creatinine level, hematuria, and proteinuria, she underwent renal biopsy.

8 g/dl), and a serum creatinine level of 1.45 mg/dl. SOF and simeprevir were discontinued at this time and she was treated with 2 additional courses of antibiotics for possible urinary tract infection. Because of persistent elevation of her serum creatinine level, hematuria, and proteinuria, she underwent renal biopsy. The renal biopsy showed enlarged, lobular glomeruli with marked endocapillary with intracapillary mononuclear cells and marked mesangial hypercellularity and expansion. The GBMs were thickened and showed double contours. There was no tubulointerstitial inflammation and no vasculitis. Immunofluorescence microscopy showed granular mesangial and GBMs staining for IgG, IgA, IgM, C3, C1q, kappa and lambda light chains, and fibrinogen. Numerous large and confluent subendothelial, mesangial, paramesangial and intramembranous electron dense deposits were present and did not show substructures. These findings were consistent with a severe active and chronic immune complex–mediated glomerulonephritis resembling a diffuse proliferative lupus nephritis (Table 4 and Figure 2c, f, and i).

bendothelial, mesangial, paramesangial and intramembranous electron dense deposits were present and did not show substructures. These findings were consistent with a severe active and chronic immune complex–mediated glomerulonephritis resembling a diffuse proliferative lupus nephritis (Table 4 and Figure 2c, f, and i). Her serologic work-up results were notable for a positive antinuclear antibody test result at a titer of 1:160, her histone antibody level was 1.5 U, her cryoglobulin percentage was 1%, and she had low serum complement (C3 and C4) levels (Table 3). Serum HCV RNA was undetectable more than 3 months after completion of therapy, indicating that HCV had been eradicated despite the fact that her regimen had been shortened to 9 weeks. Three weeks after the renal biopsy, after completing the antibiotic course prescribed to treat a positive urine culture demonstrating Klebsiella infection, she was given a pulse dose of methylprednisolone, 250 mg i.v. for 3 days, and then transitioned to prednisone 60 mg daily (1 mg/kg) followed by weekly rituximab infusions for 3 doses. Her serum creatinine stabilized at 2.3 mg/dl and she was discharged.

treat a positive urine culture demonstrating Klebsiella infection, she was given a pulse dose of methylprednisolone, 250 mg i.v. for 3 days, and then transitioned to prednisone 60 mg daily (1 mg/kg) followed by weekly rituximab infusions for 3 doses. Her serum creatinine stabilized at 2.3 mg/dl and she was discharged. However, 1 week later she presented to the hospital with lethargy and back pain and was admitted to the medical intensive care unit with septic shock from Serratia bacteremia with pancolitis seen on imaging. Oliguric acute kidney injury developed, and she required continuous venovenous hemofiltration. She had progressive respiratory failure from Stenotrophomonas pneumonia and required intubation. Her liver function worsened, disseminated intravascular coagulation developed, and she died on day 30 of hospitalization.

aging. Oliguric acute kidney injury developed, and she required continuous venovenous hemofiltration. She had progressive respiratory failure from Stenotrophomonas pneumonia and required intubation. Her liver function worsened, disseminated intravascular coagulation developed, and she died on day 30 of hospitalization. Discussion Autoimmune phenomena are well-described complications of IFN-based therapies for HCV infection. Here we have described 3 cases of lupus-like immune complex glomerulonephritis with “full house” immunofluorescence in patients with cirrhosis who were treated with IFN-free all-oral DAA regimens. It is possible that this autoimmune phenomenon represents an immune reconstitution syndrome. In the first 2 cases, these patients had no known autoimmune history, nor suggestive symptoms before initiation of DAAs. Case 2 had a prior positive antinuclear antibody test result (Table 3) that was performed as a part of routine evaluation before liver transplantation and was not due to suggestive clinical symptoms. In the third case, DAA therapy rapidly aggravated the patient's preexisting IFN alfa–induced lupus that had been diagnosed 17 years prior. In this case, the timing of symptom onset and the fact that her disease had been clinically quiescent for more than 5 years suggest that clearance of her HCV aggravated preexisting autoimmunity. In this case C1q was minimal and tubuloreticular inclusions were not detected, in contrast to the majority of cases of lupus nephritis.12 Lupus nephritis is generally characterized by immune deposits that stain dominantly or codominantly for IgG. In this series, although IgG staining was strong, it was not dominant (Table 4). We note that IgA was codominant in case 2 and 3 and that this may be due to known association of mesangial IgA deposition in advanced liver disease.13, 14, 15

characterized by immune deposits that stain dominantly or codominantly for IgG. In this series, although IgG staining was strong, it was not dominant (Table 4). We note that IgA was codominant in case 2 and 3 and that this may be due to known association of mesangial IgA deposition in advanced liver disease.13, 14, 15 Two patients received SOF paired with RBV and 1 received SOF and simeprevir; the fact that this occurred with the use of agents that purely target steps in the HCV life cycle raises the possibility that an autoimmune diathesis was unmasked with removal of HCV, even in the absence of IFN or RBV use. It is tempting to speculate, given the known effects of chronic HCV on suppression of host IFN responses, that removal of viral antigen in otherwise predisposed patients may elicit sensitivity to endogenous IFN effects and resultant unmasking of autoimmunity.16, 17 Two of the 3 patients had antihistone antibody titers determined and in both cases this autoantibody was detected in the serum (Table 3). Although a drug-induced lupus may be an alternative explanation in these patients with presence of antihistone antibodies, in 1 of the cases with a positive antihistone antibody test result, systemic lupus erythematous had been a long-standing condition. However, drug-induced lupus is only rarely associated with glomerulonephritis. In case 2, the recurrence of severe joint pain during retreatment with SOF/ledipasvir and RBV strengthens the relationship between treatment and emergence of lupus-like symptoms. Finally, all patients had remote previous exposures to IFN; it is possible that a smoldering lupus syndrome existed but was not clinically detected; however, the dramatic emergence of symptomatic lupus with both renal and nonrenal manifestations in all cases (Table 2) argues for the relationship between IFN-free DAA treatment and disease.

s had remote previous exposures to IFN; it is possible that a smoldering lupus syndrome existed but was not clinically detected; however, the dramatic emergence of symptomatic lupus with both renal and nonrenal manifestations in all cases (Table 2) argues for the relationship between IFN-free DAA treatment and disease. Although the pattern of glomerular injury seen in lupus nephritis and mixed cryoglobulinemia can result in identical morphological changes, the lack of substructures in the deposits by electron microscopy in all cases, absence of large intracapillary deposits in the glomeruli (pseudothrombi), and negative HCV RNA test result at the time of biopsy for all cases favors a lupus-like nephritis over an HCV-associated immune complex disease. Furthermore, HCV-associated renal disease typically improves with HCV eradication.18, 19, 20, 21, 22

bsence of large intracapillary deposits in the glomeruli (pseudothrombi), and negative HCV RNA test result at the time of biopsy for all cases favors a lupus-like nephritis over an HCV-associated immune complex disease. Furthermore, HCV-associated renal disease typically improves with HCV eradication.18, 19, 20, 21, 22 All 3 patients experienced rapid improvement in serum creatinine level with the initiation of high-dose (1 mg/kg) corticosteroids with or without the addition of other immunosuppressive agents (Figure 1a–c). All 3 experienced normalization of serum creatinine level within 1 to 3 weeks. Unfortunately, 2 of the 3 patients died of infectious complications while receiving high-dose steroids. The lone surviving patient had been rapidly tapered off steroids early in her course on account of delirium and was maintained on low-dose mycophenolate for 3 to 6 months, which suggests that this glomerular lesion may respond to a short course of steroids; it is unclear whether or not maintenance therapy is needed. The lone surviving patient continues to receive mycophenolate mofetil to manage joint pains and to prevent relapse of glomerulonephritis. Thus, a key take-away point from this series is that these patients, who each had baseline cirrhosis in addition to new-onset glomerulonephritis, are susceptible to adverse outcomes from immunosuppression. Recent evidence in the literature highlights the occurrence of severe infections after treatment with rituximab in patients with kidney disease.23, 24, 25 Thus, aggressive immunosuppressive therapy should be used in these cases with extreme caution, particularly if cirrhosis is present.

outcomes from immunosuppression. Recent evidence in the literature highlights the occurrence of severe infections after treatment with rituximab in patients with kidney disease.23, 24, 25 Thus, aggressive immunosuppressive therapy should be used in these cases with extreme caution, particularly if cirrhosis is present. We have reported 3 cases of immune complex–mediated glomerulonephritis with full house immunofluorescence occurring in patients being treated for HCV with novel all-oral direct-acting antiviral therapies who presented with joint pain or rash in addition to hematuria, pyuria, and rising creatinine level. This phenomenon was not reported in phase II/III studies and appears to have affected a very small minority of the patients treated in our health care system. Although this possible reaction is rare, it is important that HCV providers and nephrologists be aware of it and consider kidney biopsy in any patient with worsening renal function, proteinuria, or development of active urine sediment during or after treatment with DAAs. Further study is needed to determine the mechanism of autoimmunity in patients treated with IFN-free regimens. Disclosures MES received a research grant from Gilead Sciences and is an Abbvie advisory board member. ALL is a consultant to Genzyme. RIT is a Merck advisory board member. RTC has received research grant support from Gilead, Merck, Abbvie, BMS, Janssen, and Mass Biologics. All the other authors declared no competing interests.

es MES received a research grant from Gilead Sciences and is an Abbvie advisory board member. ALL is a consultant to Genzyme. RIT is a Merck advisory board member. RTC has received research grant support from Gilead, Merck, Abbvie, BMS, Janssen, and Mass Biologics. All the other authors declared no competing interests. Acknowledgments RTC was supported by a National Institutes of Health grant (DK078772). JAH was supported by an American Association for the Study of Liver Diseases grant, a National Institutes of Health Opportunity Fund Grant AI082630, and a National Health and Medical Research Council Early Career Fellowship. RIT was supported by a National Institutes of Health grant (DK094872-04).

Chronic kidney disease (CKD) is a growing public health issue, which today concerns around 8% to 16% of individuals worldwide, and a further disproportional increase is expected in developing countries.1 As the awareness of CKD in population is relatively low,2, 3 early detection of decreased kidney function is of major importance, not the least because CKD is associated with increased risk of cardiovascular events, hospitalization, and death.4 Creatinine and cystatin C are commonly used as markers for kidney function,5, 6 and to estimate the glomerular filtration rate (eGFR), which together with albuminuria is used to determine the stage of CKD.1 However, as both eGFR and albuminuria are relatively insensitive measures of early kidney injury, more sensitive biomarkers are needed to identify at-risk individuals earlier in the disease process to facilitate prevention of the progression to CKD.7

ich together with albuminuria is used to determine the stage of CKD.1 However, as both eGFR and albuminuria are relatively insensitive measures of early kidney injury, more sensitive biomarkers are needed to identify at-risk individuals earlier in the disease process to facilitate prevention of the progression to CKD.7 Very recently, elevated plasma levels of the soluble urokinase-type plasminogen activator receptor (suPAR) were shown to be strongly associated with an increased decline in kidney function and incidence of CKD in patients with cardiovascular disease (CVD) undergoing cardiac catheterization.8 Both suPAR and the membrane-bound form of the uPAR are known to be involved in the regulation of cell adherence and migration through binding of integrins,8, 9 and highly elevated suPAR levels have been implicated as potentially causal in the pathogenesis of focal segmental glomerulosclerosis through activation of β3 integrin that, when sufficiently activated, can lead to interference with podocyte migration and apoptosis.10, 11, 12, 13 In addition to kidney function, earlier studies have reported elevated suPAR levels in association with increased risk for several adverse health conditions, such as CVD,14, 15, 16 inflammation,17 and cancer.18

t, when sufficiently activated, can lead to interference with podocyte migration and apoptosis.10, 11, 12, 13 In addition to kidney function, earlier studies have reported elevated suPAR levels in association with increased risk for several adverse health conditions, such as CVD,14, 15, 16 inflammation,17 and cancer.18 Earlier evidence hence indicates a role for elevated serum suPAR levels in focal segmental glomerulosclerosis and CKD in severely ill patients, but information about if circulating suPAR may play a role in kidney function of healthy individuals is lacking. Therefore, we challenged this question among generally healthy participants of the Malmö Diet and Cancer Study, a large population-based cohort from Southern Sweden, and aimed to investigate if circulating suPAR levels are associated with a longitudinal decline in kidney function, incidence of CKD, or hospitalization due to impaired kidney function, during a median follow-up time of more than 18 years. Methods Malmö Diet and Cancer Study The Malmö Diet and Cancer Study (MDCS) is a population-based cohort study including men and women living in Malmö and born between 1923–1945 and 1923–1950. The participation rate was 40.8%.19 Written informed consent was given, and the MDCS was approved by the ethic committee at Lund University (LU 51-90). A detailed description of the cohort has been published elsewhere.20

ased cohort study including men and women living in Malmö and born between 1923–1945 and 1923–1950. The participation rate was 40.8%.19 Written informed consent was given, and the MDCS was approved by the ethic committee at Lund University (LU 51-90). A detailed description of the cohort has been published elsewhere.20 For this study, we included individuals from the MDCS-Cardiovascular Cohort—a subcohort of the MDCS of 6103 randomly selected participants who underwent additional phenotyping, designed to study epidemiology of carotid artery disease, in between 1991 and 1994. Information on suPAR was available for 5381 individuals. We excluded participants with data missing for smoking (n = 137), fasting glucose (n = 20), body mass index (BMI) (n = 4), and creatinine (n = 85), leading to a final study population of, in total, 5135 participants. Between 2007 and 2012, 3734 of those individuals who were alive and had not emigrated from Sweden (n = 4924) attended the follow-up re-examination, which has been described previously.21

For this study, we included individuals from the MDCS-Cardiovascular Cohort—a subcohort of the MDCS of 6103 randomly selected participants who underwent additional phenotyping, designed to study epidemiology of carotid artery disease, in between 1991 and 1994. Information on suPAR was available for 5381 individuals. We excluded participants with data missing for smoking (n = 137), fasting glucose (n = 20), body mass index (BMI) (n = 4), and creatinine (n = 85), leading to a final study population of, in total, 5135 participants. Between 2007 and 2012, 3734 of those individuals who were alive and had not emigrated from Sweden (n = 4924) attended the follow-up re-examination, which has been described previously.21 Clinical Examination and Assays During baseline examination, all participants underwent a physical examination and anthropometrics measurements were obtained by trained nurses. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured. BMI was calculated as weight/height2 (kg/m2). Questions concerning socioeconomic status, lifestyle factors, and medical history were assessed via a self-administered questionnaire.20 Fasting blood samples were drawn and immediately frozen to –80 °C and stored in a biological bank.22 Creatinine (μmol/l) was measured in plasma and analyzed with the Jaffé method, and traceable to the International Standardization with isotope dilution mass spectrometry. Cystatin C was measured using a particle-enhanced immunonephelometric assay (N Latex Cystatin; Dade Behring, Deerfield, IL). The values of cystatin C were not standardized because they were analyzed before the introduction of the world calibrator in 2010. The reference value for the method was 0.53 to 0.95 mg/l. The eGFR was calculated according to the previously reported CKD-Epidemiology Collaboration creatinine equation23 as follows: for males, eGFR = 141 × (Scr/0.9)α × 0.993age, where α is –0.411 if creatinine is ≤0.9 and –1.209 if creatinine is >0.9; for females, eGFR = 144 × (Scr/0.7)α × 0.993age, where α is –0.329 if creatinine is ≤0.7 and –1.209 if creatinine is >0.7. A factor of 0.0113 was included to convert creatinine levels measured in μmol/l into mg/dl. The concentration of suPAR (ng/ml) was analyzed in 2012 from frozen plasma blood samples, which were stored at –80 °C. A commercial ELISA suPARnostic® kit (Virogates, Copenhagen, Denmark) was used, which had an intraassay coefficient of variation of 3% and an interassay coefficient of variation of 5%. It has been shown previously that the stability of suPAR in plasma samples is high and it remains stable throughout several cycles of freezing and thawing.24

PARnostic® kit (Virogates, Copenhagen, Denmark) was used, which had an intraassay coefficient of variation of 3% and an interassay coefficient of variation of 5%. It has been shown previously that the stability of suPAR in plasma samples is high and it remains stable throughout several cycles of freezing and thawing.24 During the follow-up re-examination (2007–2012), anthropometric characteristics, and SBP and DBP were measured following similar approaches as at baseline. Furthermore, using the same analytical methods as at baseline, the plasma concentrations of glucose (mmol/l), creatinine (μmol/l), and cystatin C (mg/l) were measured in fasting blood samples. Renal Outcomes Data on eGFR at follow-up were available for 3193 participants, with measured baseline levels of suPAR; CKD at follow-up re-examination was defined as an eGFR of <60 ml/min per 1.73 m2. For the analysis of incident CKD, we excluded all participants with prevalent CKD at the baseline examination (n = 582).

Renal Outcomes Data on eGFR at follow-up were available for 3193 participants, with measured baseline levels of suPAR; CKD at follow-up re-examination was defined as an eGFR of <60 ml/min per 1.73 m2. For the analysis of incident CKD, we excluded all participants with prevalent CKD at the baseline examination (n = 582). In addition, we linked the MDCS cohort to the Swedish patient register to obtain information on hospitalization due to impaired kidney function. The Swedish patient register covers all hospitalizations in Sweden since 1987 and hospital outpatient visits from 2001 onward. The register has been previously described and validated for outcome classification.25 All participants were followed up until the occurrence of a hospital diagnosis of impaired kidney function, death, emigration from Sweden, or until 31 December 2013. In this study, we considered participants with admission to the hospital due to impaired kidney function defined as 585-586 according to International Classification of Diseases (ICD) 9, and N18 and N19 according to ICD10. We differentiated between impaired kidney function as the main diagnosis, registered as the first diagnosis in the patient record (110 cases), and impaired kidney function as the main or contributing diagnosis cases (patients with the above-mentioned ICD codes at any position, in total 207 cases). For the analysis of incident hospitalization due to impaired kidney function, we excluded all participants with prevalent impaired kidney function (n = 6).

cases), and impaired kidney function as the main or contributing diagnosis cases (patients with the above-mentioned ICD codes at any position, in total 207 cases). For the analysis of incident hospitalization due to impaired kidney function, we excluded all participants with prevalent impaired kidney function (n = 6). Statistics To reach normal distribution, we log-transformed suPAR levels and analyzed per 1 SD increment of the log value. In addition, the study sample was categorized into equal quartiles according to suPAR levels. We tested the association between suPAR and baseline characteristics using a general linear model (for continuous variables) adjusted for age and sex, and the χ2 test (for categorical variables). Furthermore, we tested the relationship between suPAR and eGFR at follow-up re-examination, for which we calculated the annual change in eGFR by dividing the variable “mean change over time” (value at follow-up minus value at baseline) by follow-up time in years to account for different lengths of follow-up.

ariables). Furthermore, we tested the relationship between suPAR and eGFR at follow-up re-examination, for which we calculated the annual change in eGFR by dividing the variable “mean change over time” (value at follow-up minus value at baseline) by follow-up time in years to account for different lengths of follow-up. Logistic regression was used to estimate the odds ratio and 95% confidence interval (CI) for incident CKD (eGFR at follow-up <60 ml/min per 1.73 m2) cases. For the basic model, we included age, sex, and follow-up time as covariates. We calculated the Net Reclassification Improvement (NRI)26 using the nri STATA command for the package idi from http://personalpages.manchester.ac.uk/staff/mark.lunt, and model discrimination was tested by calculating C-statistic using the roccomp command in STATA for models using risk factors with and without suPAR. Cox proportional hazard regression was used to estimate the hazard ratio (HR) and 95% CI for incident CKD cases. Age was used as the underlying time variable. Proportional hazard assumption was tested using Schoenfeld residuals (estat STATA command) and graphically (stphplot STATA command). The hazard function was graphically examined by plotting the Kaplan-Meier failure function (sts graph STATA command) according to the quartiles of suPAR.

as the underlying time variable. Proportional hazard assumption was tested using Schoenfeld residuals (estat STATA command) and graphically (stphplot STATA command). The hazard function was graphically examined by plotting the Kaplan-Meier failure function (sts graph STATA command) according to the quartiles of suPAR. In the basic model, we included age and sex as covariates. The final model for both Cox and logistic regression was adjusted for further risk factors of CKD: fasting glucose levels, eGFR, BMI, SBP, smoking status (current, former, or never smokers), and use of antihypertensive treatment (yes/no) at baseline. A P value of ≤0.05 was considered to be statistically significant. SPSS (version 21, IBM Corporation, Armonk, NY) and STATA version 13 (StataCorp LP, College Station, Texas, USA) were used for analysis. Results Baseline Characteristics According to the suPAR Concentrations in the MDCS-Cardiovascular Cohort Table 1 shows the baseline characteristics of the 5135 participants according to the quartiles of suPAR concentration at baseline. A positive association was observed between increased levels of suPAR with increased age, weight, BMI, SBP, DBP, waist, and glucose, and with female sex, antihypertensive treatment, and current smoking.Table 1 Clinical characteristicsa of the Malmö Diet and Cancer Study participants at baseline examination (1991–1996) according to the quartiles of suPAR

between increased levels of suPAR with increased age, weight, BMI, SBP, DBP, waist, and glucose, and with female sex, antihypertensive treatment, and current smoking.Table 1 Clinical characteristicsa of the Malmö Diet and Cancer Study participants at baseline examination (1991–1996) according to the quartiles of suPAR N All Q1 Q2 Q3 Q4 P valueb Mean (range) suPAR (ng/ml) 5135 3.0 (0.0–35.9) 2.1 (0.0–2.4) 2.6 (2.4–2.8) 3.1 (2.8–3.4) 4.4 (3.4–35.9) <0.001 Age (yr) 5135 57.6 (5.9) 55.7 (5.7) 57.34 (5.9) 58.7 (5.7) 58.8 (5.9) <0.001 Male sex, n (%) 5135 2114 (41.2%) 642 (50.0%) 551 (42.9%) 450 (35.1%) 471 (36.7%) <0.001 Height (cm) 5135 169 (8.9) 171 (9.04) 169 (8.9) 168 (8.83) 168 (8.7) 0.04 Weight (kg) 5135 74 (13.6) 73 (13.09) 73 (12.63) 74 (13.87) 74 (14.59) <0.001 BMI (kg/m2) 5135 25.7 (3.9) 25.1 (3.4) 25.5 (3.6) 26.0 (4.0) 26.2 (4.6) <0.001 SBP (mm Hg) 5135 142 (19.0) 138 (17.2) 141 (18.8) 143 (19.6) 144 (19.9) <0.001 DBP (mm Hg) 5135 87 (9.3) 86 (8.7) 87 (9.5) 87 (9.5) 87 (9.7) 0.028 Waist (cm) 5134 83.8 (12.9) 83.2 (12.5) 83.3 (12.5) 83.7 (13.0) 84.9 (13.6) <0.001 Fasting glucose (mmol/l) 5135 5.2 (1.4) 5.0 (0.8) 5.1 (1.2) 5.2 (1.4) 5.4 (1.9) <0.001 Cystatin C (mg/l) 4814 0.8 (0.2) 0.7 (0.1) 0.8 (0.1) 0.8 (0.1) 0.9 (0.2) <0.001 P-Creatinine (μmol/l) 5135 84.8 (16.4) 85.3 (14.6) 85.4 (14.4) 83.6 (15.0) 85.0 (20.5) 0.829 eGFR (ml/min per 1.73 m2) 5135 76.0 (13.7) 78.1 (13.3) 75.7 (13.1) 75.3 (13.4) 75.1 (14.9) 0.145 AHT, n (%) 5135 889 (17.8%) 166 (12.9%) 196 (15.3%) 252 (19.6%) 275 (21.4%) <0.001 Current smoking, n (%) 5135 1377 (26.8%) 192 (15.0%) 226 (17.6%) 316 (24.6%) 643 (50.1%) <0.001 AHT, antihypertensive treatment; BMI, body mass index; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology Collaboration creatinine equation23; SBP, systolic blood pressure; suPAR, soluble urokinase-type plasminogen activator receptor.

%) <0.001 AHT, antihypertensive treatment; BMI, body mass index; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology Collaboration creatinine equation23; SBP, systolic blood pressure; suPAR, soluble urokinase-type plasminogen activator receptor. a Data shown as mean ± SD or n (%). b P value adjusted for age and sex, or χ2 (1df) test (categorical variables). suPAR Levels and Longitudinal Change in Kidney Function At follow-up re-examination, the mean eGFR was 70.15 (SD, 15.21) ml/min per 1.73 m2 in the 3193 participants. Compared with participants within the lowest quartile of suPAR, participants in the top quartile had a significantly lower eGFR (73.95 vs. 67.75 ml/min per 1.73 m2; P-trend = 4.3 × 10–7). Participants within the top quartile of baseline suPAR had a higher annual decline in eGFR than those within the lowest quartile of baseline suPAR (P-trend = 2.9 × 10–8) (Figure 1).Figure 1 Annual change in eGFR between baseline and follow-up re-examination in 3193 participants of the Malmö Diet and Cancer Study-Cardiovascular Cohort according to the quartiles of suPAR concentration at baseline. The general linear model was adjusted for age, sex, baseline levels of eGFR, and follow-up time. Concentration of suPAR at baseline Q1: 0.03–2.36, Q2: 2.36–2.73, Q3: 2.73–3.26, and Q4: 3.26–15.64. eGFR, estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology Collaboration creatinine equation23; suPAR, soluble urokinase-type plasminogen activator receptor.

eGFR, and follow-up time. Concentration of suPAR at baseline Q1: 0.03–2.36, Q2: 2.36–2.73, Q3: 2.73–3.26, and Q4: 3.26–15.64. eGFR, estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology Collaboration creatinine equation23; suPAR, soluble urokinase-type plasminogen activator receptor. suPAR and Incidence of CKD During Follow-up During a mean follow-up time of 16.6 (range, 13.3–20.2) years, incident CKD (eGFR < 60 ml/min per 1.73 m2 at follow-up re-examination) was present in 561 (231 men and 330 women) participants. In the basic model, the incidence of CKD was increased by 23% per 1 SD increase of log suPAR concentration at baseline (odds ratio, 1.23; 95% CI, 1.10–1.38). After additional adjustment for known risk factors (baseline levels of eGFR, fasting glucose, BMI, SBP, antihypertensive treatment, smoking), the risk increase remained similar (odds ratio, 1.25; 95% CI, 1.10–1.41). Compared with participants within the lowest quartile of suPAR, participants within the highest quartile had 69% increased incidence of CKD (Q4 odds ratio, 1.69; 95% CI, 1.25–2.30) (Figure 2).Figure 2 Risk for incident chronic kidney disease (eGFR ≤ 60 ml/min per 1.73 m2) at follow-up re-examination in 2851 participants of the Malmö Diet and Cancer Study-Cardiovascular Cohort according to the quartile of suPAR concentration at baseline. The logistic regression model was adjusted for age, sex, baseline levels of eGFR, fasting glucose, body mass index, systolic blood pressure, antihypertensive treatment (yes/no), smoking (current, ex, former), and follow-up time. Concentration of suPAR at baseline Q1: 0.03–2.35, Q2: 2.35–2.70, Q3: 2.71–3.25, and Q4: 3.25–15.64. N/cases: Q1: 712/94, Q2: 713/132, Q3: 713/163, and Q4: 713/172. CI, confidence interval; eGFR, estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology Collaboration creatinine equation23; OR, odds ratio; suPAR, soluble urokinase-type plasminogen activator receptor.

and Q4: 3.25–15.64. N/cases: Q1: 712/94, Q2: 713/132, Q3: 713/163, and Q4: 713/172. CI, confidence interval; eGFR, estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology Collaboration creatinine equation23; OR, odds ratio; suPAR, soluble urokinase-type plasminogen activator receptor. Risk Discrimination and NRI We added suPAR to a model with conventional risk factors (i.e., sex and baseline age, eGFR, fasting glucose, SBP, antihypertensive medication, BMI, and smoking) and follow-up time to test the incremental value in discriminating between participants with and without CKD at follow-up re-examination. The area under the curve of the receiver operating characteristics was only marginally improved (area under the curve without suPAR 0.729 vs. area under the curve with suPAR 0.733, P = 0.17). However, adding suPAR to the risk model led to a significant NRI for 15.5% of the individuals (P = 0.0010). Model calibration was acceptable (Hosmer-Lemeshow’s P > 0.05) for both models with and without suPAR.

nally improved (area under the curve without suPAR 0.729 vs. area under the curve with suPAR 0.733, P = 0.17). However, adding suPAR to the risk model led to a significant NRI for 15.5% of the individuals (P = 0.0010). Model calibration was acceptable (Hosmer-Lemeshow’s P > 0.05) for both models with and without suPAR. suPAR and Hospitalization due to Impaired Kidney Function During a mean follow-up time of 19.04 (range, 0–22.25) years, 110 individuals (63 men and 47 women) were admitted to the hospital due to impaired kidney function as the main diagnosis. The event rate was 1.13 per 1000 person-years. The HR for incidence of impaired kidney function per 1 SD increase of log suPAR was 1.59 (95% CI, 1.41–1.79) in the basic model and remained significant in the fully adjusted model (HR = 1.49; 95% CI, 1.27–1.74). Hospitalization due to impaired renal function was significantly more common among participants within the highest quartile of baseline suPAR (Table 2). Adjusted for sex, participants within the highest suPAR quartile had an HR of 6.89 (95% CI, 3.11–15.28) compared with those within the lowest quartile of suPAR. After adjusting for further risk factors, the HR remained significantly increased (HR = 3.73; 95% CI, 1.65–8.44) (Figure 3).Figure 3 Cumulative incidence of hospitalization due to impaired renal function (main diagnosis, n = 110) during follow-up according to the quartiles of suPAR in 5129 participants in the Malmö Diet and Cancer Study-Cardiovascular Cohort. In the final model, male sex (HR: 2.53; 95% CI, 1.69–3.79), BMI (HR: 1.10; 95% CI, 1.05–1.15), baseline glucose (HR: 1.21; 95% CI, 1.14–1.30), AHT (HR: 1.59; 95% CI, 1.05–2.41), eGFR (HR: 0.96; 95% CI, 0.95–0.98), and current smoking (HR: 2.23; 95% CI, 1.34–3.73) were significantly associated with hospitalization due to impairment of renal function, in addition to suPAR. The Kaplan-Meier plot shows cumulative percentages of main cases of impaired renal function during follow-up in quartiles: first (lowest values) to fourth (highest values) quartile of the baseline suPAR concentration. Median (range) concentrations of the quartiles 1 to 4 are shown in Table 2. The numbers at risk are shown at 5-year intervals. Cox regression adjusted for sex, fasting glucose levels, eGFR, BMI, systolic blood pressure, smoking status (current, former, or never smokers), and use of antihypertensive treatment (AHT) (yes/no) at baseline. Age was used as an underlying time variable.

n in Table 2. The numbers at risk are shown at 5-year intervals. Cox regression adjusted for sex, fasting glucose levels, eGFR, BMI, systolic blood pressure, smoking status (current, former, or never smokers), and use of antihypertensive treatment (AHT) (yes/no) at baseline. Age was used as an underlying time variable. BMI, body mass index; CI, confidence interval; eGFR, estimated glomerular filtration rate; HR, hazard ratio; suPAR, soluble urokinase-type plasminogen activator receptor. Table 2 Incidence of impaired kidney function (main diagnosis) during a mean follow-up time of 19 yr, in relation to baseline concentration of suPAR in the Malmö Diet and Cancer-Cardiovascular Cohort

BMI, body mass index; CI, confidence interval; eGFR, estimated glomerular filtration rate; HR, hazard ratio; suPAR, soluble urokinase-type plasminogen activator receptor. Table 2 Incidence of impaired kidney function (main diagnosis) during a mean follow-up time of 19 yr, in relation to baseline concentration of suPAR in the Malmö Diet and Cancer-Cardiovascular Cohort Per quartile of suPAR Q1 Q2 Q3 Q4 Per 1 SD log suPAR Mean (range) suPAR (ng/ml) 2.12 (0.03–2.43) 2.62 (2.43–2.83) 3.10 (2.83–3.41) 4.36 (3.41–35.86) Impaired kidney function as the main diagnosis N/casesa 5129/110 1282/7 1282/24 1283/32 1282/47 Sex-adjusted HR (95% CI) 1.69 (1.40–2.04) 1.0 3.17 (1.37–7.37) 4.18 (1.84–9.50) 6.89 (3.11–15.28) 1.59 (1.41–1.79) Risk factorb-adjusted HR (95% CI) 1.37 (1.22–1.67) 1.0 2.67 (1.15–6.22) 3.11 (1.36–7.10) 3.73 (1.65–8.44) 1.49 (1.27–1.74) All impaired kidney function cases N/casesc 5129/207 1282/23 1282/47 1283/57 1282/80 Sex-adjusted HR (95% CI) 1.46 (1.28–1.67) 1.0 1.84 (1.12–3.03) 2.20 (1.35–3.59) 3.50 (2.19–5.57) 1.46 (1.32–1.61) Risk factorb-adjusted HR (95% CI)c 1.24 (1.08–1.43) 1.0 1.55 (0.94–2.56) 1.69 (1.04–2.76) 2.13 (1.32–3.46) 1.35 (1.18–1.53) AHT, antihypertensive treatment; BMI, body mass index; CI, confidence interval; eGFR, estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology Collaboration creatinine equation23; HR, hazard ratio; SBP, systolic blood pressure, suPAR, soluble urokinase-type plasminogen activator receptor. a Admission to the hospital due to impaired kidney function as main diagnosis.

Per quartile of suPAR Q1 Q2 Q3 Q4 Per 1 SD log suPAR Mean (range) suPAR (ng/ml) 2.12 (0.03–2.43) 2.62 (2.43–2.83) 3.10 (2.83–3.41) 4.36 (3.41–35.86) Impaired kidney function as the main diagnosis N/casesa 5129/110 1282/7 1282/24 1283/32 1282/47 Sex-adjusted HR (95% CI) 1.69 (1.40–2.04) 1.0 3.17 (1.37–7.37) 4.18 (1.84–9.50) 6.89 (3.11–15.28) 1.59 (1.41–1.79) Risk factorb-adjusted HR (95% CI) 1.37 (1.22–1.67) 1.0 2.67 (1.15–6.22) 3.11 (1.36–7.10) 3.73 (1.65–8.44) 1.49 (1.27–1.74) All impaired kidney function cases N/casesc 5129/207 1282/23 1282/47 1283/57 1282/80 Sex-adjusted HR (95% CI) 1.46 (1.28–1.67) 1.0 1.84 (1.12–3.03) 2.20 (1.35–3.59) 3.50 (2.19–5.57) 1.46 (1.32–1.61) Risk factorb-adjusted HR (95% CI)c 1.24 (1.08–1.43) 1.0 1.55 (0.94–2.56) 1.69 (1.04–2.76) 2.13 (1.32–3.46) 1.35 (1.18–1.53) AHT, antihypertensive treatment; BMI, body mass index; CI, confidence interval; eGFR, estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology Collaboration creatinine equation23; HR, hazard ratio; SBP, systolic blood pressure, suPAR, soluble urokinase-type plasminogen activator receptor. a Admission to the hospital due to impaired kidney function as main diagnosis. b Adjusted for sex, fasting glucose levels, eGFR, BMI, SBP, smoking status (current, former, or never smokers), and use of AHT (yes/no) at baseline. Age was used as an underlying time variable. c Admission to the hospital due to impaired kidney function as the main diagnosis (n = 110) and the secondary diagnosis (n = 97, in total n = 207).

b Adjusted for sex, fasting glucose levels, eGFR, BMI, SBP, smoking status (current, former, or never smokers), and use of AHT (yes/no) at baseline. Age was used as an underlying time variable. c Admission to the hospital due to impaired kidney function as the main diagnosis (n = 110) and the secondary diagnosis (n = 97, in total n = 207). Similar results were observed when using a broader endpoint, including also individuals who were hospitalized with impairment of kidney function as a secondary diagnosis, which identified 97 additional cases, leading to 207 cases in total (128 men and 79 women) and an event rate of 2.12 per 1000 person-years (Table 2 and Figure 4).Figure 4 Cumulative incidence of hospitalization due to impaired renal function (all cases, n = 207) during follow-up according to the quartiles of suPAR in 5129 participants in the Malmö Diet and Cancer Study-Cardiovascular Cohort. The Kaplan-Meier plot shows cumulative percentages of all cases of impaired kidney function (n = 207) during follow-up in quartiles: first (lowest values) to fourth (highest values) quartile of the baseline suPAR concentration. Median (range) concentrations of the quartiles 1 to 4 are shown in Table 2. The numbers at risk are shown at 5-year intervals. Cox regression adjusted for sex, fasting glucose levels, estimated glomerular filtration rate, body mass index, systolic blood pressure, smoking status (current, former, or never smokers), and use of antihypertensive treatment (yes/no) at baseline. Age was used as underlying time variable. suPAR, soluble urokinase-type plasminogen activator receptor.

fasting glucose levels, estimated glomerular filtration rate, body mass index, systolic blood pressure, smoking status (current, former, or never smokers), and use of antihypertensive treatment (yes/no) at baseline. Age was used as underlying time variable. suPAR, soluble urokinase-type plasminogen activator receptor. We tested the proportional hazard assumption of all Cox regression models, and the global P value for the Schoenfeld residuals in the final model for hospitalization due to impaired renal function as the main diagnosis was 0.099 and as the secondary diagnosis was 0.18, indicating no deviation from proportional hazard for the overall models. However, the assumption was violated by eGFR with a P value of 0.0010 and 0.0006, respectively. Therefore, because eGFR is known to decline by age, we formally tested interaction between age and eGFR on incidence of hospitalization due to impaired kidney function, and observed a strong interaction (P < 0.0001). However, adding such interaction term to the Cox models produced a similar estimated risk increase per 1 SD of log suPAR in the full model for hospitalization due to impaired kidney function as the main diagnosis, 1.51 (1.26–1.82) and as the secondary diagnosis, 1.37 (1.18–1.58), respectively.

interaction (P < 0.0001). However, adding such interaction term to the Cox models produced a similar estimated risk increase per 1 SD of log suPAR in the full model for hospitalization due to impaired kidney function as the main diagnosis, 1.51 (1.26–1.82) and as the secondary diagnosis, 1.37 (1.18–1.58), respectively. Sensitivity Analyses In sensitivity analyses, we excluded all cases classified as unspecified impaired kidney function (ICD9 586 or ICD10 N19) (n = 15) at first and last admission. HR estimates for main impaired kidney function (n = 109) were unchanged in the remaining 5114 participants (sex-adjusted HR = 1.59 [95% CI, 1.42–1.79] and fully adjusted HR = 1.49 [95% CI, 1.28–1.75] per increase in 1 SD log suPAR). Likewise, the estimates for all cases of impaired kidney function (in total n = 192) remained similar (sex-adjusted HR = 1.48 [95% CI, 1.34–1.64] and fully adjusted HR = 1.38 [95% CI, 1.21–1.57] per increase in 1 SD log suPAR). In addition, we further excluded participants with prevalent diabetes or CVD at baseline (n = 325). Estimates remained unchanged (fully adjusted HR for impaired kidney function as the main diagnosis [n = 86] 1.45 [95% CI, 1.22–1.73] and all cases of impaired kidney function [n = 166] 1.32 [95% CI, 1.15–1.52] per 1 SD increment of suPAR, respectively).

participants with prevalent diabetes or CVD at baseline (n = 325). Estimates remained unchanged (fully adjusted HR for impaired kidney function as the main diagnosis [n = 86] 1.45 [95% CI, 1.22–1.73] and all cases of impaired kidney function [n = 166] 1.32 [95% CI, 1.15–1.52] per 1 SD increment of suPAR, respectively). Lastly, given the known association between baseline levels of suPAR and gender, we created sex-specific quartiles and tested the association between suPAR and hospitalization due to impaired kidney function. Hazard estimates remained similar (per increases in sex-specific quartile of suPAR main cases of impaired kidney function: sex-adjusted HR = 1.66 [95% CI, 1.37–2.01] and fully adjusted HR = 1.32 [95% CI, 1.09–1.62]; and all cases of impaired kidney function: sex-adjusted HR = 1.45 [95% CI, 1.26–1.65] and fully adjusted HR = 1.20 [95% CI, 1.05–1.39], respectively).

es in sex-specific quartile of suPAR main cases of impaired kidney function: sex-adjusted HR = 1.66 [95% CI, 1.37–2.01] and fully adjusted HR = 1.32 [95% CI, 1.09–1.62]; and all cases of impaired kidney function: sex-adjusted HR = 1.45 [95% CI, 1.26–1.65] and fully adjusted HR = 1.20 [95% CI, 1.05–1.39], respectively). Discussion The main findings from our population-based prospective cohort demonstrate that increased suPAR levels are associated with a decline in kidney function and hospitalization due to impaired kidney function in middle-aged men and women from southern Sweden, independently of traditional risk factors. It has recently been reported that elevated suPAR levels are associated with kidney function decline in patients with CVD undergoing cardiac catheterization,8 and that highly elevated suPAR may be causally linked to focal segmental glomerulosclerosis,10, 11, 12, 13 yet to our knowledge, our study is the first to show that suPAR predicts incidence of CKD and a longitudinal decline in kidney function in a generally healthy population. Thereby, our results are not only in line with the previously reported findings in severely ill patients,8 but also provide novel evidence for a broader application of suPAR as a predictive biomarker for declining kidney function. Awareness is one of the key issues for individuals at risk for CKD or with CKD.27 Thus, early detection would be of immense importance as it may open opportunities for pharmacological and/or lifestyle-related preventive interventions.27, 28

Discussion The main findings from our population-based prospective cohort demonstrate that increased suPAR levels are associated with a decline in kidney function and hospitalization due to impaired kidney function in middle-aged men and women from southern Sweden, independently of traditional risk factors. It has recently been reported that elevated suPAR levels are associated with kidney function decline in patients with CVD undergoing cardiac catheterization,8 and that highly elevated suPAR may be causally linked to focal segmental glomerulosclerosis,10, 11, 12, 13 yet to our knowledge, our study is the first to show that suPAR predicts incidence of CKD and a longitudinal decline in kidney function in a generally healthy population. Thereby, our results are not only in line with the previously reported findings in severely ill patients,8 but also provide novel evidence for a broader application of suPAR as a predictive biomarker for declining kidney function. Awareness is one of the key issues for individuals at risk for CKD or with CKD.27 Thus, early detection would be of immense importance as it may open opportunities for pharmacological and/or lifestyle-related preventive interventions.27, 28 In our study cohort, suPAR was not observed to add discriminative value on the top of the already established risk factors, as the C-statistics only marginally increased when suPAR was added to the model. Receiver operating characteristic is commonly used to evaluate how well a test or model can distinguish between a diseased and a nondiseased status (i.e., CKD in this study). One important aspect to keep in mind in regard to receiver operating characteristic analysis is that the effect on change in area under the curve depends on both the predictive ability of the “traditional risk model” and the strength of the new marker, and also on a potential correlation between them, and C-statistics may often be an insensitive measure.29, 30 Another aspect to consider is that suPAR levels may change over time, which in the scenario of a long follow-up most likely would reduce the observed associations between suPAR and CKD, and reduce the discrimination ability of rather insensitive measures like C-statistics. However, lack of information on suPAR levels at follow-up excludes such investigation in our study. Generally, the approach of reclassification is different, with the ability to determine how many individuals would be classified into the clinically relevant risk strata, thus directly comparing the clinical impact of 2 models.30 The results from our NRI analysis demonstrated that suPAR was able to significantly reclassify individuals into the correct risk direction. Hence, suPAR could be a potentially useful novel biomarker to identify individuals at high risk for developing CKD. Advantageously, we observed that elevated suPAR predicts future kidney dysfunction already a relatively long time before (on average >16 years) detection of declined eGFR at follow-up re-examination or hospitalization due to impaired kidney function.

el biomarker to identify individuals at high risk for developing CKD. Advantageously, we observed that elevated suPAR predicts future kidney dysfunction already a relatively long time before (on average >16 years) detection of declined eGFR at follow-up re-examination or hospitalization due to impaired kidney function. Our results from the follow-up re-examination are supported by the additional analysis using data obtained from inpatient and outpatient registries on hospitalization due to impaired renal function (i.e., ICD9/10 585&586/N18&N19), which can be considered as a more strict definition, and also reflected in the lower number of cases in these analyses (110 cases with impaired renal function as the main diagnosis vs. 561 individuals classified with incident CKD at follow-up re-examination). We observed the expected effect modification between age and eGFR. However, as the estimates for suPAR remained unchanged after adding an interaction term (age × eGFR) to the used Cox models, the relationship between age and eGFR seemed not to markedly influence the interpretation of our results.

ow-up re-examination). We observed the expected effect modification between age and eGFR. However, as the estimates for suPAR remained unchanged after adding an interaction term (age × eGFR) to the used Cox models, the relationship between age and eGFR seemed not to markedly influence the interpretation of our results. The clinical usability of novel biomarkers is of major importance as accurate early stratification of individuals at increased risk could enable targeting of preventive health resources. Previously, it has been shown that suPAR fulfills critical requirements for being a clinically usable biomarker of CKD in patients with CVD,8 considering that it has been shown to be stable in plasma,8, 31 and it is associated with incident CKD already in patients with still normal eGFR.8 Furthermore, suPAR has been shown to add prognostic value in patients with CVD and among subgroups with diabetes or hypertension,8 as well as in both whites and blacks regardless of the clear differences in prevalence of CKD between these ethnic groups.8 The close relation between heart and kidney disease makes our findings clinically useful in predicting risk of CKD in patients with CVD. In contrast to the study by Hayek et al.,8 our study population was generally healthy with a follow-up of more than 16 years, and expands the previous results, now highlighting its potential as an early risk marker also in the general population. In addition, the cNRI analysis demonstrates that suPAR could correctly reclassify individuals with respect to risk for future CKD, on top of other risk factors. Altogether, the results suggest that suPAR could be of relevance in a clinical setting.

g its potential as an early risk marker also in the general population. In addition, the cNRI analysis demonstrates that suPAR could correctly reclassify individuals with respect to risk for future CKD, on top of other risk factors. Altogether, the results suggest that suPAR could be of relevance in a clinical setting. suPAR has been suggested to have a specific role in the kidney, particularly in anchoring of podocytes in the basement membrane of the glomerulus through β3 integrin activation.12, 13 In line with earlier studies in severely ill patients,8 and in patients with focal segmental glomerulosclerosis,10, 11, 12, 13 our results now suggest that suPAR may also play a role in deterioration of kidney function in a generally healthy population. Therefore, more studies are needed to further explore the role of suPAR in kidney function in functional, preventive, and therapeutic settings.

focal segmental glomerulosclerosis,10, 11, 12, 13 our results now suggest that suPAR may also play a role in deterioration of kidney function in a generally healthy population. Therefore, more studies are needed to further explore the role of suPAR in kidney function in functional, preventive, and therapeutic settings. Strengths and Limitations In our study, we used kidney function data from several sources to investigate the role of circulating suPAR levels in a longitudinal decline of kidney function during a long follow-up with convincing results toward similar conclusions for all approaches. Yet, we need to note that all the participants of our study are Caucasians limiting the generalizability of our observations; however, the study by Hayek et al.8 reported that elevated suPAR was associated with a decline in kidney function similarly among Caucasians and Afro-American patients with CVD. Furthermore, it is important to note that all our main observations were detected independently of known risk factors, and that thorough sensitivity analyses (excluding either patients with unspecified impaired renal function, prevalent CVD, or diabetes, and using sex-specific tertiles of suPAR) provided similar estimates for hazard for hospitalization due to impaired renal function, demonstrating that most likely none of these significantly accounted for the observed associations.

ng either patients with unspecified impaired renal function, prevalent CVD, or diabetes, and using sex-specific tertiles of suPAR) provided similar estimates for hazard for hospitalization due to impaired renal function, demonstrating that most likely none of these significantly accounted for the observed associations. Our study has some limitations that need to be discussed. Unfortunately, we do not have data for albuminuria in our cohort at baseline, which is a key limitation in the assessment of CKD stages 1 and 2. Furthermore, we have only measurements of cystatin C and creatinine at 2 time points, and more measurements would have been desirable as also recommended in the current KDIGO 2012 CKD guidelines.32 The Kidney Disease Improving Global Outcomes criteria for the classification of CKD require an eGFR of ≤60 ml/min per 1.73 m2 for a duration of >3 months; as there was only 1 follow-up visit, this was not possible to obtain. Yet, a similar definition for CKD was used by Hayek et al.8 Nevertheless, it needs to be pointed out that the long follow-up time of our study increases the confidence in assessing CKD progression.32 Furthermore, the results using the Swedish registry data on hospitalization due to impaired kidney function strongly supported the results observed based on changes in eGFR, not the least because this endpoint was assessed independently of the CKD definition at follow-up re-examination. Lastly, our study design provides evidence for the association between suPAR and kidney function decline, yet we cannot prove causality.

ion strongly supported the results observed based on changes in eGFR, not the least because this endpoint was assessed independently of the CKD definition at follow-up re-examination. Lastly, our study design provides evidence for the association between suPAR and kidney function decline, yet we cannot prove causality. Conclusion The findings of this prospective cohort study of generally healthy middle-aged participants from Sweden indicate that increased circulating suPAR level is associated with increased risk for the future decline of kidney function, and hospitalization due to impaired kidney function. Overall, the results of our study highlight circulating suPAR as a potential novel biomarker to identify individuals at increased risk for the decline in kidney function in the general population. Disclosure This study was supported by the Swedish Research Council, the Swedish Heart and Lung Foundation, the Novo Nordic Foundation, the Swedish Diabetes Foundation, and the Påhlsson Foundation, Skåne University Hospital, Ernhold Lundström, and by equipment grants from the Knut and Alice Wallenberg Foundation, the Region Skåne, Skåne University Hospital, Ernhold Lundström Foundation, the Linneus Foundation for the Lund University Diabetes Center, and the European Research Council (Consolidator grant number 649021, to MO-M). Acknowledgments We thank all participants of the Malmö Diet and Cancer Study.

Worldwide, the prevalence of obesity (body mass index [BMI] ≥ 30 kg/m2) in the general population has risen dramatically over the past few decades; this trend is paralleled in the chronic kidney disease (CKD) population.1, 2 Among adults with CKD in the United States, for example, the prevalence of obesity increased from 38.1% in 1999 to 2002 to 44.1% in 2011 to 2014 (P = 0.004 for linear trend) (Figure 1).2 The increase in obesity prevalence occurred primarily in World Health Organization (WHO) class II and III obesity3 (BMI ≥ 35 and ≥ 40 kg/m2, respectively), which increased from 17.2% in 1999 to 2002 to 22.2% in 2011 to 2014 (P = 0.01 for linear trend).Figure 1 Trends* in class I, II, and III obesity over time in the US adult chronic kidney disease (CKD) population (National Health and Nutrition Examination Survey [NHANES] 1999−2014). *P values for linear trends over time ≤ 0.01 for body mass index (BMI) ≥ 30, BMI ≥ 35, and BMI ≥ 40 kg/m2.

(P = 0.01 for linear trend).Figure 1 Trends* in class I, II, and III obesity over time in the US adult chronic kidney disease (CKD) population (National Health and Nutrition Examination Survey [NHANES] 1999−2014). *P values for linear trends over time ≤ 0.01 for body mass index (BMI) ≥ 30, BMI ≥ 35, and BMI ≥ 40 kg/m2. An increasing prevalence of obesity in the CKD population is of particular concern due to evidence of associations between higher BMI and adverse renal outcomes. In observational studies, obesity has been associated with higher risk of incident CKD and end-stage renal disease (ESRD), as well as nephrolithiasis and renal cell cancer.4, 5, 6, 7 Potential mechanisms explaining the increased risk for CKD and ESRD include obesity-mediated hypertension, insulin resistance, glomerular hyperfiltration, activation of the renin−angiotensin−aldosterone system, inflammation, and adipocytokine dysregulation.8, 9 On the other hand, the risk associated with obesity may be reversible: a randomized controlled trial of patients with type 2 diabetes demonstrated that weight loss decreased the risk of adverse CKD outcomes.10 However, achieving sustained weight loss through lifestyle modification is challenging.

ine dysregulation.8, 9 On the other hand, the risk associated with obesity may be reversible: a randomized controlled trial of patients with type 2 diabetes demonstrated that weight loss decreased the risk of adverse CKD outcomes.10 However, achieving sustained weight loss through lifestyle modification is challenging. Bariatric surgery is a proven, effective method for sustained weight loss and is becoming more commonplace for patients with morbid or severe obesity. Consensus guidelines from a National Institutes of Health (NIH) conference include BMI ≥ 40 kg/m2 or BMI ≥ 35 kg/m2 with obesity-related comorbidity as approved clinical indications for bariatric surgery.11 As a significant proportion of patients with CKD may qualify for bariatric surgery, it is increasingly important to understand the potential benefits and risks of bariatric surgery in regard to kidney function and other outcomes.

with obesity-related comorbidity as approved clinical indications for bariatric surgery.11 As a significant proportion of patients with CKD may qualify for bariatric surgery, it is increasingly important to understand the potential benefits and risks of bariatric surgery in regard to kidney function and other outcomes. Overview of Bariatric Surgery Types Contemporary bariatric surgery techniques are very effective in achieving sustained weight loss, with total weight loss averaging 20% to 35% of total body weight.12, 13 Several surgical procedures to promote weight loss have been developed over the past few decades. These procedures vary in terms of the amount of gastric surface area restriction, intended nutrient malabsorption, effects on gastrointestinal hormones, weight loss outcomes, and risk of complications (Table 1).14 Initial efforts in bariatric surgery started in the 1970s with the jejunoileal bypass, which was a purely malabsorptive procedure, bypassing most of the small intestine.15 The jejunoileal bypass has since been abandoned due to the high rate of complications, which included deficiency of fat-soluble vitamins, bacterial overgrowth, calcium oxalate nephrolithiasis, and kidney and liver failure.Table 1 Comparison of the most common surgical procedures for weight loss

f the small intestine.15 The jejunoileal bypass has since been abandoned due to the high rate of complications, which included deficiency of fat-soluble vitamins, bacterial overgrowth, calcium oxalate nephrolithiasis, and kidney and liver failure.Table 1 Comparison of the most common surgical procedures for weight loss RYGB LSG LAGB Weight loss Highest Moderate Lowest Gastric emptying ↑ or ↓ ↑ No change Plasma GLP-1 levels ↑ ↑ No change Plasma PYY levels ↑ ↑ No change Plasma ghrelin levels Variable ↑ ↓ Plasma leptin levels ↓ ↓ ↓ Plasma bile acid levels ↑ ↑ No change Fat malabsorption/fat-soluble vitamin deficiency ↑ No change No change Nephrolithiasis risk ↑ No change No change Diabetes remission Highest Moderate Lowest Short-term complications Higher Lower Lower Need for reoperation Lower Lower Higher GLP-1, glucagon-like peptide-1; LAGB, laparoscopic-assisted gastric banding; LSG, laparoscopic sleeve gastrectomy; PYY, peptide YY; RYGB, Roux-en-y gastric bypass. Adapted from previously published literature.14, 74, 75

RYGB LSG LAGB Weight loss Highest Moderate Lowest Gastric emptying ↑ or ↓ ↑ No change Plasma GLP-1 levels ↑ ↑ No change Plasma PYY levels ↑ ↑ No change Plasma ghrelin levels Variable ↑ ↓ Plasma leptin levels ↓ ↓ ↓ Plasma bile acid levels ↑ ↑ No change Fat malabsorption/fat-soluble vitamin deficiency ↑ No change No change Nephrolithiasis risk ↑ No change No change Diabetes remission Highest Moderate Lowest Short-term complications Higher Lower Lower Need for reoperation Lower Lower Higher GLP-1, glucagon-like peptide-1; LAGB, laparoscopic-assisted gastric banding; LSG, laparoscopic sleeve gastrectomy; PYY, peptide YY; RYGB, Roux-en-y gastric bypass. Adapted from previously published literature.14, 74, 75 Currently, the most common bariatric procedures performed worldwide are laparoscopic Roux-en-Y gastric bypass (RYGB) and laparoscopic vertical sleeve gastrectomy (LSG).16, 17 The RYGB surgery involves both malabsorption and restriction. First, the stomach is divided into an upper stomach pouch (15−30 ml) and a lower gastric remnant (Figure 2). The stomach pouch is then anastomosed to the mid-jejunum, and a jejuno-jejunal anastomosis is created to reconnect the biliopancreatic limb and the gastric remnant, thereby allowing gastric, pancreatic, and biliary secretions to mix with food in the jejuno-jejunal anastomosis.14 LSG is a restrictive surgery that involves the removal of 70% to 80% of the lateral stomach. Due to its success in achieving weight loss and perhaps better safety profile compared to RYGB, LSG has become more common in the past few years and has eclipsed RYGB in the United States (Figure 3).18 However, some patients who undergo LSG may require subsequent conversion to RYGB or duodenal switch surgery, in which biliopancreatic secretions are diverted from the food until the last portion of the small bowel to increase malabsorption. Reported reasons for conversion of LSG to RYGB or duodenal switch surgery include weight regain and intractable acid reflux.19Figure 2 Surgical procedures for weight loss include (a) laparoscopic adjustable gastric banding, (b) sleeve gastrectomy, (c) Roux-en-Y gastric bypass, and (d) biliopancreatic diversion with duodenal switch. See text for details of these procedures.

nal switch surgery include weight regain and intractable acid reflux.19Figure 2 Surgical procedures for weight loss include (a) laparoscopic adjustable gastric banding, (b) sleeve gastrectomy, (c) Roux-en-Y gastric bypass, and (d) biliopancreatic diversion with duodenal switch. See text for details of these procedures. Reprinted with permission from Maria EJ. Bariatric surgery for morbid obesity. N Engl J Med. 2007;356:2176–2183.76Figure 3 Estimated number of surgical procedures for weight loss in the United States from 2011 to 2015. Adapted from the American Society of Metabolic Surgery (ASMBS) estimations.18 BPD/DS, biliopancreatic diversion with duodenal switch; LAGB, laparoscopic-assisted gastric banding; LSG, laparoscopic sleeve gastrectomy; RYGB, Roux-en-y gastric bypass.

Reprinted with permission from Maria EJ. Bariatric surgery for morbid obesity. N Engl J Med. 2007;356:2176–2183.76Figure 3 Estimated number of surgical procedures for weight loss in the United States from 2011 to 2015. Adapted from the American Society of Metabolic Surgery (ASMBS) estimations.18 BPD/DS, biliopancreatic diversion with duodenal switch; LAGB, laparoscopic-assisted gastric banding; LSG, laparoscopic sleeve gastrectomy; RYGB, Roux-en-y gastric bypass. Laparoscopic adjustable gastric banding (LAGB) is another purely restrictive procedure that involves the insertion of an adjustable ring immediately below the gastroesophageal junction on the proximal stomach. Due to lower success in achieving weight loss and high risk of reoperation, LAGB has fallen out of favor during the past few years and is now much less commonly performed than RYGB or LSG. All 3 procedures are now almost exclusively done laparoscopically; the proportion of laparoscopic bariatric surgery procedures in a worldwide survey increased from 63% in 2003 to 96% in 2013.16, 20 Unlike gastric banding, RYGB and LSG have favorable effects on various hormones linked to hunger, satiety, and food preferences (Table 1).14 Other procedures that induce more weight loss by increased malabsorption are less commonly performed and include the biliopancreatic diversion with duodenal switch (Figure 2). A variation of the RYGB involves increasing the length of the Roux limb to upwards of 200 cm, resulting in increased malabsorption.21 Although increasing malabsorption results in greater weight loss, risks of nephrolithiasis and oxalate nephropathy may be increased.

pancreatic diversion with duodenal switch (Figure 2). A variation of the RYGB involves increasing the length of the Roux limb to upwards of 200 cm, resulting in increased malabsorption.21 Although increasing malabsorption results in greater weight loss, risks of nephrolithiasis and oxalate nephropathy may be increased. Effect of Bariatric Surgery on CKD Risk Factors Obesity greatly increases the risk of hypertension and diabetes, the 2 most common reported causes of ESRD.22 A systematic review of bariatric surgery studies with long-term follow-up reported remission rates for type 2 diabetes of 66.7% and 28.6% for RYGB and LAGB, respectively.23 In a randomized trial of obese patients with uncontrolled type 2 diabetes who were randomized to either bariatric surgery plus intensive medical therapy or intensive medical therapy alone, bariatric surgery resulted in better glycemic control, defined by achievement of glycated hemoglobin level of 6% or less after 3 years (RYGB, 38%; sleeve gastrectomy, 24%; medical therapy, 5%; P ≤ 0.01 for both comparisons).24 Bariatric surgery also reduced the number of glucose-lowering and antihypertensive medications, and improved quality of life compared to intensive medical therapy (P < 0.05 for all comparisons). Impressive effects on blood pressure have also been reported. Among adolescents undergoing bariatric surgery, elevated blood pressure remitted in 75% of patients.25 In the aforementioned systematic review of bariatric surgery studies with long-term follow-up, remission rates for hypertension were 60.4% for RYGB and 22.7% for LAGB.23

ects on blood pressure have also been reported. Among adolescents undergoing bariatric surgery, elevated blood pressure remitted in 75% of patients.25 In the aforementioned systematic review of bariatric surgery studies with long-term follow-up, remission rates for hypertension were 60.4% for RYGB and 22.7% for LAGB.23 Associations Between Bariatric Surgery and Long-term Risk of Death Over the long-term, observational studies suggest that bariatric surgery is associated with a 30% to 45% lower risk of death compared to similar patients with severe obesity who did not undergo bariatric surgery.26, 27, 28 Most of the observed reduction in mortality risk is due to lower risks of deaths from cardiovascular disease, diabetes, and cancer. However, caution should be advised in interpreting these observational findings, as they are subject to selection bias: patients who undergo bariatric surgery are likely healthier than their severely obese peers. Patients undergoing surgery go through a rigorous, multistep process that includes evaluations and multiple appointments with a multidisciplinary team, and require successful weight loss prior to bariatric surgery, thereby displaying their commitment and motivation. It is important to note that 1 study found a 58% higher rate of non−disease-related deaths (e.g., accidents, suicide) among patients who underwent bariatric surgery compared to matched controls.27 Further research from randomized trials, if possible, is needed to evaluate whether bariatric surgery provides a long-term mortality benefit as well as to elucidate the long-term risks of surgery.

se-related deaths (e.g., accidents, suicide) among patients who underwent bariatric surgery compared to matched controls.27 Further research from randomized trials, if possible, is needed to evaluate whether bariatric surgery provides a long-term mortality benefit as well as to elucidate the long-term risks of surgery. Bariatric Surgery and Risk of Perioperative and Postoperative Complications There are nontrivial perioperative and postoperative risks of bariatric surgery, including infection, respiratory failure, acute kidney injury (AKI), and death. Recent reports from a prospective observational cohort study in the United States and an Italian national registry found that modern bariatric surgery mortality rates were approximately 0.3%, similar to those for laparoscopic cholecystectomy.29, 30 In an analysis of 27,736 bariatric surgery patients from 2006 to 2008 in the American College of Surgeons National Surgical Quality Improvement Program, the risk of 30-day mortality was not significantly different across levels of estimated glomerular filtration rate (eGFR) (eGFR ≥ 90, 0.1%; eGFR 60−89, 0.2%; eGFR 30−59, 0.3%; eGFR <30, 0.0%; P = 0.2); however, postoperative complications were more common with lower levels of kidney function (4.6%, 6.1%, 7.7%, 7.5%, and 9.9%, respectively; P < 0.001 for linear trend).31 After multivariate adjustment, each increase in CKD stage was associated with an 18% higher odds of postoperative complications. Another American College of Surgeons National Surgical Quality Improvement Program study examined outcomes in 138 dialysis-dependent ESRD patients who underwent bariatric surgery betweeen 2006 and 2011 (34% LAGB, 49% RYGB, 17% LSG). Reassuringly, this study found that 30-day mortality was relatively low (0.7%) and noted a shift from LABG to LSG in more recent years, similar to overall trends in bariatric surgery.32

outcomes in 138 dialysis-dependent ESRD patients who underwent bariatric surgery betweeen 2006 and 2011 (34% LAGB, 49% RYGB, 17% LSG). Reassuringly, this study found that 30-day mortality was relatively low (0.7%) and noted a shift from LABG to LSG in more recent years, similar to overall trends in bariatric surgery.32 Effect of Bariatric Surgery on Measured Kidney Function in Patients With Preserved GFR Change in kidney function after bariatric surgery was first described in 1980 in a study in which 8 obese patients had GFR measured (mGFR) using [51Cr] ethylenediaminetetraacetic acid (EDTA) before and 12 months after jejunoileal bypass (Table 2).33 Mean unindexed mGFR significantly decreased from 153 to 123 ml/min, but because body surface area (BSA) also decreased from 2.33 m2 to 1.93 m2, there was no change in mGFR indexed to BSA (114 to 110 ml/min/1.73 m2). Similar findings have been seen in other studies measuring GFR or creatinine clearance in individuals with preserved kidney function.34, 35, 36, 37, 38, 39 Since nephron number is fixed at birth, a decrease in unindexed GFR in this high range may be interpreted as improvement in single nephron glomerular hyperfiltration.34, 40 A meta-analysis of surgical weight loss studies found that bariatric surgery reduced unindexed mGFR in patients with normal GFR or high GFR by a mean of 25.6 ml/min.41 In support of a beneficial effect of bariatric surgery on future renal risk, a number of studies and meta-analyses have shown that albuminuria and proteinuria decrease after bariatric surgery, although it is unclear whether this is a direct effect of weight loss or mediated by improvements in blood pressure and insulin resistance.42, 43 Two case reports observed complete resolution of proteinuria after bariatric surgery in patients with obesity-related FSGS.44, 45 However, all of these studies were of short duration (1−2 years), and often lacked comparison groups.Table 2 Short-term studies measuring glomerular filtration rate by exogenous filtration markers or creatinine clearance (24-h urine) before and after bariatric surgery

urgery in patients with obesity-related FSGS.44, 45 However, all of these studies were of short duration (1−2 years), and often lacked comparison groups.Table 2 Short-term studies measuring glomerular filtration rate by exogenous filtration markers or creatinine clearance (24-h urine) before and after bariatric surgery First author, year n Type of surgery (n) Follow-up (mo) GFR assessment Presurgery GFR, CrCl, albuminuria Follow-up GFR, CrCl, albuminuria Patients with normal or increased GFR Brochner, 1980 8 Intestinal bypass 12 mGFR (EDTA) GFRunindexed 153 ml/min GFRindexed 114 ml/min/1.73 m2 GFRunindexed 123 ml/min GFR 110indexed ml/min/1.73 m2 Chagnac, 2004 8 Gastroplasty 12 mGFR (inulin) GFRunindexed 145 ml/min UAE 16 μg/min GFRunindexed 110 ml/min UAE 5 μg/min Navarro-Diaz, 2006 61 Gastric bypass 24 24-h CrCl CrCl 140 ml/min UAE ≥ 30 mg/d 42.6% CrCl 118 ml/min UAE ≥ 30 mg/d 14.8% Serpa, 2009 140 RYGB 8 CrCl CrCl 148 ml/min UAE ≥ 30 mg/d 43.6% CrCl 114 ml/min UAE ≥ 30 mg/d 21.4% Saliba, 2010 35 RYGB 12 CrCl Diabetic patients: CrCl 155 ml/min UAE 26 mg/d Nondiabetic patients: CrCl 148 ml/min UAE 10 mg/d Diabetic patients: CrCl 132 ml/min UAE 15 mg/d Nondiabetic patients: CrCl 117 ml/min UAE 14 mg/d Lieske, 2014 11 RYGB (9), BPD/DS (2) 12 mGFR (iothalamate), CrCl GFRunindexed 121 ml/min GFRindexed 95 ml/min/1.73 m2 CrCl 120 ml/min UAE 20.5 mg/d GFRunindexed 90 ml/min GFRindexed 85 ml/min/1.73 m2 CrCl 98 ml/min UAE 17.1 mg/d Friedman, 2014 36 Gastric bypass Mean 10 mGFR (iohexol) GFRunindexed 117 ml/min GFRindexed 87 ml/min/1.73 m2 GFRunindexed 100 ml/min GFRindexed 87 ml/min/1.73 m2 Patients with CKD Navaneethan, 2015 15 RYGB (7), LAGB (3), LSG (3) 12 mGFR (iothalamate) GFRunindexed 82 ml/min GFRindexed 50 ml/min/1.73 m2 Proteinuria 0.60 g/d GFRunindexed 81 ml/min GFRindexed 64 ml/min/1.73 m2 Proteinuria 0.43 g/d BPD/DS, biliopancreatic diversion with duodenal switch; CKD, chronic kidney disease; CrCl, creatinine clearance; EDTA, ethylenediaminetetraacetic acid; GFR, glomerular filtration rate; LAGB, laparoscopic-assisted gastric banding; LSG, laparoscopic sleeve gastrectomy; mGFR, measured GFR; RYGB, Roux-en-y gastric bypass; UAE, urinary albumin excretion.

on with duodenal switch; CKD, chronic kidney disease; CrCl, creatinine clearance; EDTA, ethylenediaminetetraacetic acid; GFR, glomerular filtration rate; LAGB, laparoscopic-assisted gastric banding; LSG, laparoscopic sleeve gastrectomy; mGFR, measured GFR; RYGB, Roux-en-y gastric bypass; UAE, urinary albumin excretion. Effect of Bariatric Surgery on Measured Kidney Function in Patients With CKD A limited number of studies have examined the effect of bariatric surgery in patients with CKD. In a study led by Navaneethan et al., 13 patients with serum creatinine ≥ 1.3 mg/dl underwent measurement of GFR using iothalamate clearance before surgery and 3, 6, and 12 months after bariatric surgery (RYGB, 7; LAGB, 3; LSG, 3).47 In contrast to the studies in patients with preserved GFR, unindexed mGFR in this CKD cohort remained unchanged at 12 months (82.0−80.5 ml/min, P = 0.3). When mGFR was indexed to BSA, mGFR actually increased from 50 to 64 ml/min/1.73 m2 over the 12-month period (P = 0.02). Interestingly, increased indexed and unindexed mGFR were significantly associated with decreases in leptin. A randomized trial comparing LSG to optimal medical management was conducted in 11 patients with stage 3 to 4 CKD (LSG, 5; control, 6), but was unable to draw firm conclusions due to small sample size and lack of measured GFR.46

ncreased indexed and unindexed mGFR were significantly associated with decreases in leptin. A randomized trial comparing LSG to optimal medical management was conducted in 11 patients with stage 3 to 4 CKD (LSG, 5; control, 6), but was unable to draw firm conclusions due to small sample size and lack of measured GFR.46 Matched Observational Cohort Studies Examining Long-term Outcomes After Bariatric Surgery The association between bariatric surgery and kidney outcomes was evaluated in a large health system from central and northeast Pennsylvania with up to 9 years of follow-up.48 Using propensity scores that included demographic, comorbidity, laboratory, and previous nutrition clinic use data, 985 patients who underwent bariatric surgery (97% RYGB) were matched with 985 morbidly obese patients. Bariatric surgery was associated with a 58% lower risk of eGFR decline of ≥ 30% and a 57% lower risk of doubling of serum creatinine or ESRD (Figure 4). The beneficial association of bariatric surgery with kidney outcomes was similar among patients with and without eGFR < 90 ml/min/1.73 m2, hypertension, and diabetes. However, only 91 patients had eGFR <60 ml/min/1.73 m2, limiting power to examine CKD patients. Investigators at Kaiser Permanante reported a similar association between bariatric surgery and improved eGFR outcomes after bariatric surgery (RYGB, 58%; LSG, 42%) in 714 surgery patients and 714 matched controls with stage 3 or 4 CKD.49 Patients who underwent RYGB experienced greater weight loss and greater improvements in eGFR compared to LSG patients.Figure 4 Kaplan−Meier curves estimating time to kidney outcomes by surgery group (n = 985) and control group (n = 985). Figure from Chang et al.48 Estimated glomerular filtration (eGFR) decline ≥ 30% outcome was defined as having a follow-up outpatient eGFR ≥ 30% lower than the baseline eGFR value. End-stage renal disease (ESRD) was defined as eGFR < 15 ml/min/1.73 m2 or treated ESRD per US Renal Data System Registry. Shaded areas represent 95% confidence interval bounds.

glomerular filtration (eGFR) decline ≥ 30% outcome was defined as having a follow-up outpatient eGFR ≥ 30% lower than the baseline eGFR value. End-stage renal disease (ESRD) was defined as eGFR < 15 ml/min/1.73 m2 or treated ESRD per US Renal Data System Registry. Shaded areas represent 95% confidence interval bounds. Filtration Markers in Estimating GFR After Bariatric Surgery There are important limitations of all observational studies of kidney disease and bariatric surgery, including potential residual confounding and the use of creatinine-based eGFR, which correlates with muscle mass. Loss of muscle mass with massive weight loss might result in overestimation of eGFR after bariatric surgery.35, 36, 50 Friedman et al. measured GFR by iothalamate clearance, serum creatinine, and cystatin C in 33 patients with normal or supranormal kidney function.36 Cystatin C correlated better with mGFR than creatinine both before and after surgery, although neither had very good accuracy in estimating GFR alone. Use of the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) combined creatinine-cystatin C equation estimated GFR within 30% of mGFR more than 80% of the time before and after surgery. This suggests that the use of multiple filtration markers could be beneficial in future research studying kidney function changes after bariatric surgery.

ase Epidemiology Collaboration (CKD-EPI) combined creatinine-cystatin C equation estimated GFR within 30% of mGFR more than 80% of the time before and after surgery. This suggests that the use of multiple filtration markers could be beneficial in future research studying kidney function changes after bariatric surgery. Renal Risks of Bariatric Surgery There exist a number of renal risks of bariatric surgery, including perioperative AKI, and long-term risks of nephrolithiasis and oxalate nephropathy. AKI is fairly common after bariatric surgery, with reports ranging from 2.9% to 8.5% in published studies, which have used varying definitions of AKI.51, 52, 53 Risk factors for AKI after bariatric surgery include higher BMI, lower eGFR, preoperative use of angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and intraoperative hypotension. Bariatric surgery patients may be prone to dehydration and higher risk for prerenal AKI in the long term, although this risk has not been quantified.

atric surgery include higher BMI, lower eGFR, preoperative use of angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and intraoperative hypotension. Bariatric surgery patients may be prone to dehydration and higher risk for prerenal AKI in the long term, although this risk has not been quantified. Obesity is a well-recognized risk factor for nephrolithiasis,54, 55 and thus one would hope that weight loss may reduce the risk of nephrolithiasis. However, risk of kidney stones may increase after certain types of bariatric surgery and appears to be related to the degree of fat malabsorption achieved. This complication was well recognized with the jejunoileal bypass: the 15-year risk of developing renal stones was 29%.15 Although less fat malabsorption occurs in RYGB than in the jejunoileal bypass, fecal fat has also been shown to increase 6 and 12 months after RYGB.56 Steatorrhea is thought to result in hyperoxaluria by increasing formation of calcium fatty acids salts, leading to decreased binding of calcium to oxalate, and then increased oxalate absorption.56 Several studies have found that patients who undergo RYGB have greater urinary risk factors for nephrolithiasis after surgery compared to before surgery.21, 57, 58, 59 Urinary oxalate and calcium oxalate supersaturation increase, whereas urinary citrate and total urine volume decrease. Supplementation with calcium citrate is recommended routinely after RYGB surgery60 and would be expected to reduce risk of oxaluria. However, to our knowledge, no data exist on the effect of calcium supplementation on urinary supersaturation for calcium oxalate and risk of nephrolithiasis and oxalate nephropathy.

ease. Supplementation with calcium citrate is recommended routinely after RYGB surgery60 and would be expected to reduce risk of oxaluria. However, to our knowledge, no data exist on the effect of calcium supplementation on urinary supersaturation for calcium oxalate and risk of nephrolithiasis and oxalate nephropathy. In an observational study of 762 patients who underwent bariatric surgery and 762 matched nonsurgery patients in Olmsted County, Minnesota, the risk of nephrolithiasis was 11.1% in bariatric surgery patients compared to 4.3% in nonsurgery patients.61 When examined by type of bariatric surgery, the risk of nephrolithiasis was 315% higher for the procedures causing the most malabsorption (biliopancreatic diversion with duodenal switch or very long limb RYGB), and 113% higher for standard RYGB, compared to that in nonsurgery patients. In contrast, restrictive procedures (LAGB or LSG) were not associated with increased risk of nephrolithiasis. This study also found a 96% increased risk of CKD for biliopancreatic diversion with duodenal switch or very long limb RYGB surgeries, as well as a nonsignificant trend toward reduced risk of CKD for RYGB. Important limitations of this study include the use of International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes to ascertain nephrolithiasis and CKD, as well as the possibility that bariatric surgery patients may receive more medical care and thus more recorded diagnoses of nephrolithiasis and CKD than nonsurgery patients.

s study include the use of International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes to ascertain nephrolithiasis and CKD, as well as the possibility that bariatric surgery patients may receive more medical care and thus more recorded diagnoses of nephrolithiasis and CKD than nonsurgery patients. Oxalate nephropathy is the most severe renal complication of bariatric surgery and has been reported in patients after jejunoileal bypass and RYGB surgery. A case series of 11 patients with oxalate nephropathy after RYGB demonstrated acute and chronic tubulointerstitial nephropathy and calcium oxalate deposits (mean, 3.5 deposits per glomerulus).62 Time from surgery to AKI ranged from 4 to 96 months, and the majority of patients progressed to ESRD. More research is needed to determine how commonly oxalate nephropathy occurs after RYGB, and whether compliance with calcium citrate supplementation can prevent this complication or calcium oxalate nephrolithiasis. Reversal of RYGB may be considered to reduce hyperoxaluria, although it is unclear whether this improves long-term outcomes.63

ed to determine how commonly oxalate nephropathy occurs after RYGB, and whether compliance with calcium citrate supplementation can prevent this complication or calcium oxalate nephrolithiasis. Reversal of RYGB may be considered to reduce hyperoxaluria, although it is unclear whether this improves long-term outcomes.63 Bariatric Surgery in ESRD Patients and Kidney Transplantation Candidates Although policies vary by kidney transplantation center, BMI ≥ 35 to 40 kg/m2 is generally considered a contraindication for transplantation due to worse outcomes compared to those in patients with lower BMI.64 Therefore, bariatric surgery may play an important role in improving access to kidney transplantation for severely obese patients. In a 2004 retrospective study, investigators at the University of Cincinnati reported that RYGB (open, 97%; laparoscopic, 3%) was safe and effective at achieving weight loss in 30 morbidly obese patients with CKD or ESRD.65 Of the 10 patients who were on dialysis before RYGB, 3 patients received a kidney transplant, 4 were scheduled for a living donor transplant, and 3 were on the waitlist. Only 1 complication was reported (abdominal wound infection). There were no perioperative deaths; the only death reported was due to cardiovascular disease, occurring 7.9 years after RYGB and 6.1 years after transplantation.

kidney transplant, 4 were scheduled for a living donor transplant, and 3 were on the waitlist. Only 1 complication was reported (abdominal wound infection). There were no perioperative deaths; the only death reported was due to cardiovascular disease, occurring 7.9 years after RYGB and 6.1 years after transplantation. The same group conducted a prospective study from 2011 to 2014, during which all kidney transplant candidates meeting National Institutes of Health (NIH) criteria for bariatric surgery were referred to a multidisciplinary clinic that included a bariatric surgeon, dietitian, and coordinator.66 These individuals received concurrent workup for laparoscopic LSG and renal transplantation, and were required to undergo a 6-month medical weight loss program, involving monthly physician-supervised visits with the dietitian and bariatric surgery coordinator. Of 170 patients deemed appropriate for both LSG and kidney transplantation, 102 were in evaluation for LSG, and 52 patients (47 dialysis, 5 CKD stage 4) had undergone LSG by the end of 2014. Among the 52 patients who underwent LSG, the mean BMI decreased from 43.0 to 36.4 kg/m2. The majority of patients (55.8%) achieved BMI < 35 kg/m2 and were able to be placed on the waitlist, with 6 patients receiving renal transplants after LSG. The only reported complication was an episode of supraventricular tachycardia, and no perioperative deaths (<30 days) occurred.

I decreased from 43.0 to 36.4 kg/m2. The majority of patients (55.8%) achieved BMI < 35 kg/m2 and were able to be placed on the waitlist, with 6 patients receiving renal transplants after LSG. The only reported complication was an episode of supraventricular tachycardia, and no perioperative deaths (<30 days) occurred. Another single-center study reported similar weight loss outcomes with LSG in 9 CKD (5 dialysis) patients but more common postoperative complications.67 The median BMI decreased from 44.2 kg/m2 to 34.7 kg/m2, and 4 of 5 dialysis patients achieved their target weight, enabling them to be placed on the waitlist. One of these listed patients was later made inactive due to infection and gastric leak. Other complications included fistula thrombosis on postoperative day 1, which was stented; AKI due to dehydration 2 weeks postoperatively, requiring hospitalization; and myocardial infarction 3 weeks postsurgery.

ed on the waitlist. One of these listed patients was later made inactive due to infection and gastric leak. Other complications included fistula thrombosis on postoperative day 1, which was stented; AKI due to dehydration 2 weeks postoperatively, requiring hospitalization; and myocardial infarction 3 weeks postsurgery. Lentine et al. identified Medicare billing claims for bariatric surgery among renal transplant candidates and recipients in the United States Renal Data System registry from 1991 to 2004.68 Of 188 cases of bariatric surgery identified, 72 surgeries were performed pre-listing, 29 while on the waitlist, and 87 posttransplantation. Of 29 waitlisted patients, 20 later proceeded to kidney transplantation after bariatric surgery. Perioperative mortality (30-day) was 3.5% for patients on the waitlist at the time of bariatric surgery and 3.5% for posttransplantation patients. Mortality between 30 and 90 days after bariatric surgery was 3.5% for posttransplantation patients, and 0% for patients on the waitlist at the time of bariatric surgery. However, all of these procedures were open procedures, and most were RYGB surgeries. Because the vast majority of surgeries are now laparoscopic LSG, further research is needed to examine outcomes of transplantation candidates and patients using these methods. Whether the pharmacokinetics of immunosuppressive drugs is altered in the setting of bariatric surgery is another open question and requires further study to draw firm conclusions.69, 70, 71, 72, 73

copic LSG, further research is needed to examine outcomes of transplantation candidates and patients using these methods. Whether the pharmacokinetics of immunosuppressive drugs is altered in the setting of bariatric surgery is another open question and requires further study to draw firm conclusions.69, 70, 71, 72, 73 Conclusions With the rising prevalence of morbid obesity overall and in the CKD population, there is an urgent need to better understand the effects of bariatric surgery on kidney-related outcomes. Bariatric surgery is extremely effective for achieving weight loss and improving CKD risk factors such as hypertension and diabetes. It also reduces proteinuria and glomerular hyperfiltration, which, in the long term, could impart beneficial effects. In observational studies, bariatric surgery is associated with improvement in creatinine-based renal outcomes, although additional long-term studies are needed using other filtration markers less correlated with body mass. In any case, other risks associated with surgery such as AKI and nephrolithiasis should be weighed before referring patients for surgery. Bariatric surgeries featuring higher malabsorption appear to increase the risk of nephrolithiasis and oxalate nephropathy, and further research is needed to identify those individuals who are at highest risk, as well as effective management strategies. Disclosure All the authors declared no competing interests. Acknowledgments ARC is supported by National Institutes of Health (NIH)/National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) grant K23 DK106515-01.

Patients with advanced chronic kidney disease (CKD), that is, with an estimated glomerular filtration rate (eGFR) of <30 ml/min/1.73 m2 body surface area, including those with end-stage renal disease (ESRD) who receive maintenance dialysis therapy, have a substantially high annual mortality of 10% to 20%.1 Indeed the mortality is even higher in the first several months of transitioning to dialysis therapy, and the annualized death rate may approach 30% to 40% or higher.2 This excessively high death risk of advanced CKD is worse than that of most cancers,3 in which the leading causes of death are cardiovascular and infectious.1 Hospitalizations, too, are exceptionally high in these patients, and their health-related quality of life is low. The etiology of such exceptionally poor clinical outcomes have remained obscure.

of advanced CKD is worse than that of most cancers,3 in which the leading causes of death are cardiovascular and infectious.1 Hospitalizations, too, are exceptionally high in these patients, and their health-related quality of life is low. The etiology of such exceptionally poor clinical outcomes have remained obscure. For decades, management efforts and strategies have focused on targeting the well-known and conventional risks factors of poor clinical outcomes in the general population such as hyperlipidemia, hypertension and obesity. However, these strategies, which were based on the extrapolation of findings from the general population, have not resulted in major improvements in survival. Furthermore, targeting CKD-specific factors including anemia, iron deficiency, hyperphosphatemia, hyperparathyroidism, vitamin D deficiency, hypercalcemia, and dialysis dose have also not led to improved clinical outcomes. Randomized clinical trials have failed to show any survival benefit with the normalization of hemoglobin level,4 increase in dialysis dose of hemodialysis5 or peritoneal dialysis,6 controlling hyperparathyroidism by calcimimetics,7 or supplementation by vitamin D analogues.8 Lowering blood pressure9 or managing hyperlipidemia with statins have failed to improve outcomes, especially in dialysis patients.10 Although not all of these trials have examined survival as a primary endpoint, there is no meaningful survival differential in their primary and secondary analyses.

vitamin D analogues.8 Lowering blood pressure9 or managing hyperlipidemia with statins have failed to improve outcomes, especially in dialysis patients.10 Although not all of these trials have examined survival as a primary endpoint, there is no meaningful survival differential in their primary and secondary analyses. The Obesity Paradox in the Context of Reverse Epidemiology Over the past 1 to 2 decades, a large number of observational studies with very large sample sizes (usually more than 10,000 patients) have consistently indicated seemingly counterintuitive associations between the traditional risk factors for cardiovascular disease, in particular obesity as well as hypertension and hyperlipidemia, and paradoxically better survival.11 These and other risk factor survival paradoxes, including the adiponectin paradox12 and uric acid paradox,13 have been collectively referred to as the “reverse epidemiology” phenomenon, or altered risk factor patterns, to highlight the associations that are in sharp contradistinction to conventional patterns.14 Reverse epidemiology has also been observed in persons with heart failure,15 chronic obstructive lung disease, liver cirrhosis, and metastatic cancer, as well as in the geriatric population.16 Data on the reverse epidemiology of obesity have been remarkably consistent in showing that a lower body mass index (BMI) or weight loss over time are associated with poor outcomes, whereas higher BMI or gaining solid weight have been protective and associated with better survival (Figure 1). This phenomenon has been referred to as the “obesity paradox.”17 Studies by different investigators have shown rather consistent and uniform findings on the obesity paradox in advanced CKD, especially in dialysis patients. Many recent studies have also confirmed the presence of the obesity paradox in contemporary cohorts across different ethnicities and races as well as geographic regions of the world.18 Indeed these epidemiologic associations have been robust to many different types of statistical analyses, including marginal structural models, tempering concerns about substantial residual confounding and other biases.11, 17 A deeper understanding of the phenomenon of the obesity paradox in CKD patients is important, considering that the poor outcomes in this population may improve if any gain in solid weight is associated with greater survival.

tural models, tempering concerns about substantial residual confounding and other biases.11, 17 A deeper understanding of the phenomenon of the obesity paradox in CKD patients is important, considering that the poor outcomes in this population may improve if any gain in solid weight is associated with greater survival. In this review, we summarize data on the obesity paradox and relate them to clinical practice and public health.Figure 1 Reverse association of body mass index (BMI) and survival in patients with advanced chronic kidney disease (CKD) as compared to the general population.

tural models, tempering concerns about substantial residual confounding and other biases.11, 17 A deeper understanding of the phenomenon of the obesity paradox in CKD patients is important, considering that the poor outcomes in this population may improve if any gain in solid weight is associated with greater survival. In this review, we summarize data on the obesity paradox and relate them to clinical practice and public health.Figure 1 Reverse association of body mass index (BMI) and survival in patients with advanced chronic kidney disease (CKD) as compared to the general population. Is Obesity Good or Bad for CKD? Data are relatively consistent in showing that obesity is associated with higher risk of incident CKD. Large cohort studies suggest that obesity, that is, BMI > 30 kg/m2, especially in the context of metabolic syndrome and insulin resistance, is associated with higher risk of de novo CKD.19 In a national cohort of more than 3 million US veterans without previously known renal insufficiency (eGFR >60 ml/min/1.73 m2), higher BMI > 30 kg/m2 was associated with loss of kidney function across different ages.20 The lowest risk for loss of kidney function was noted in patients with BMI levels between 25 and 30 kg/m2, whereas a consistent U-shaped association between BMI and rapid loss of kidney function was noted for BMI levels <25 kg/m2 and >30 kg/m2, which was more prominent with advanced age, except in the patients who were younger than 40 years, in whom BMI was not predictive of renal function impairment.20 The investigators concluded that obesity, defined by a BMI of >30 kg/m2, was associated with a rapid loss of kidney function in patients with eGFR > 60 ml/min/1.73 m2.20 Emerging data suggest that weight loss interventions may prevent de novo CKD or may slow or reverse early CKD progression, although some bariatric surgical interventions may result in an initial drop in eGFR, which may be due to improvement in glomerular hyperfiltration and hence favorable sequelae.21, 22 Although the pathogenesis of CKD in obesity remains obscure, studies indicate that excess body fat can result in kidney disease by means of different mechanisms including secondary focal segmental glomerulosclerosis.23

FR, which may be due to improvement in glomerular hyperfiltration and hence favorable sequelae.21, 22 Although the pathogenesis of CKD in obesity remains obscure, studies indicate that excess body fat can result in kidney disease by means of different mechanisms including secondary focal segmental glomerulosclerosis.23 Meta-analyses suggest that once CKD develops, overweight and obese ranges of BMI are paradoxically associated with greater survival in advanced predialysis (eGFR<30 ml/min/1.73 m2) and dialysis-dependent CKD patients,24 whereas a pooled analysis showed that higher pretransplantation BMI was associated with higher mortality in kidney transplantation recipients.25 In dialysis patients, the obesity paradox data are quite consistent, especially in maintenance hemodialysis patients, as has been reviewed elsewhere.11, 26 Hence, it is important to acknowledge the role of obesity as an important risk factor for de novo CKD. However, once CKD has occurred, there appears to be a consistent association between obesity and better outcomes including lower mortality in those with advanced CKD, particularly among patients receiving hemodialysis therapy, suggesting that the reverse epidemiology of obesity is robust (Figure 2).Figure 2 Obesity is a risk factor for chronic kidney disease (CKD), yet it protects against CKD-associated death. ESRD, end-stage renal disease.

ower mortality in those with advanced CKD, particularly among patients receiving hemodialysis therapy, suggesting that the reverse epidemiology of obesity is robust (Figure 2).Figure 2 Obesity is a risk factor for chronic kidney disease (CKD), yet it protects against CKD-associated death. ESRD, end-stage renal disease. What Components of Body Mass Are Increased in Weight Gain? Having a larger body size means having either greater solid weight or water weight. It is relatively well known that higher fluid retention is associated with poorer outcomes, particularly in dialysis patients.27 A 2-year cohort of 34,107 hemodialysis patients who had an average weight gain of at least 0.5 kg above their postdialysis dry weight by the time of their subsequent hemodialysis treatment showed that higher weight gain increments were associated with higher risk of all-cause and cardiovascular mortality, so that the hazard ratios of cardiovascular death for weight gains of <1.0 kg and >4.0 kg (compared with 1.5−2.0 kg as the reference) were 0.67 (95% confidence interval: 0.58−0.76) and 1.25 (1.12−1.39), respectively.27 The mechanisms by which fluid retention influences cardiovascular death in hemodialysis patients may be similar to that of the heart failure population and warrants further research.27

4.0 kg (compared with 1.5−2.0 kg as the reference) were 0.67 (95% confidence interval: 0.58−0.76) and 1.25 (1.12−1.39), respectively.27 The mechanisms by which fluid retention influences cardiovascular death in hemodialysis patients may be similar to that of the heart failure population and warrants further research.27 Because the bone and viscera typically cannot expand, having a larger solid weight or weight gain is due to having or gaining more skeletal muscle mass or larger fat mass. A gain in fat mass is often the dominating development after a hypercatabolic event or acute illness has resolved or upon higher protein and calorie intake.28 Indeed, the Minnesota study in volunteer soldiers who agreed to starve for days showed that, after losing weight with proportional losses of fat and muscle, regaining the same weight back to baseline was associated with disproportionally higher fat versus muscle regain.29 Dullo et al. showed that in so-called yo-yo dieting, losing and gaining back the same amount of weight is invariably associated with more fat and less muscle mass accumulation, and is often associated with an even higher risk of insulin resistance, metabolic syndrome, and diabetes mellitus.30 Gaining muscle is much more difficult and requires resistance exercise along with anabolic support such as high protein intake with high biologic value and sometimes anabolic steroids in chronic disease populations and those of older age.

n higher risk of insulin resistance, metabolic syndrome, and diabetes mellitus.30 Gaining muscle is much more difficult and requires resistance exercise along with anabolic support such as high protein intake with high biologic value and sometimes anabolic steroids in chronic disease populations and those of older age. If Fat Is Good, Muscle Is Better Several studies have shown that any gain in body weight is associated with better survival in CKD, whereas both fat mass and fat-free lean body mass, the latter of which is essentially representative of muscle mass, also confer survival advantage. There remains considerable challenge in differentiating fat and muscle mass routinely in the clinical setting. Fat mass can be assessed using dual energy x-ray absoptiometry (DEXA)31, 32 or near-infrared interactance.33 Lean body mass can be estimated using imaging studies, anthropometry such as mid-arm muscle circumference,34, 35 or equations based on serum creatinine.36 Serum creatinine has been shown to correlate closely with muscle mass, especially in dialysis patients,37 and equations have been created that use serum creatinine and certain demographic data to estimate lean body mass, as published by Noori et al.36

uscle circumference,34, 35 or equations based on serum creatinine.36 Serum creatinine has been shown to correlate closely with muscle mass, especially in dialysis patients,37 and equations have been created that use serum creatinine and certain demographic data to estimate lean body mass, as published by Noori et al.36 In a study in 535 adult hemodialysis patients whose body fat was directly measured with near-infrared interactance, low baseline body fat percentage and fat loss over time were independently associated with higher mortality even after adjustment for demographics and surrogates of muscle mass and inflammation, whereas a tendency toward a worse quality of life was seen with a higher body fat percentage.38 In a cohort of 742 hemodialysis patients comprising 391 males and 351 females who were separately divided into 4 quartiles of near-infrared interactance−measured lean body mass and fat mass, the highest versus lowest quartiles of fat mass and lean body mass were strongly associated with lower mortality in women, whereas the highest versus lowest quartiles of fat mass and percentage fat but not of lean body mass were associated with greater survival in men.39 Cubic spline survival analyses showed greater survival with higher fat mass percentage and higher “fat mass minus lean body mass percentiles” in both sexes, whereas a higher lean body mass was protective in women. This study suggested that the survival advantage of fat mass was superior to that of lean body mass.39 There are, however, other studies suggesting that both higher lean body mass and BMI are related to greater survival in hemodialysis patients. In a large cohort of 117,683 hemodialysis patients, higher estimated lean body mass, defined by creatinine based equations developed by Noori et al.,36 was linearly associated with lower mortality.40 Compared with the reference group (48.4 to <50.5 kg), patients with the lowest estimated lean body mass (<41.3 kg) had a 1.4-fold higher risk of mortality. A similar linear association was seen among patients with BMI < 35 kg/m2 and in non-Hispanic Caucasian and African American subgroups. However, higher estimated lean body mass was not associated with improved survival in Hispanic patients or those with BMI > 35 kg/m2.40 To better examine the role of different types of fat, a landmark study was conducted by Italian colleagues led by Zoccali et al.

non-Hispanic Caucasian and African American subgroups. However, higher estimated lean body mass was not associated with improved survival in Hispanic patients or those with BMI > 35 kg/m2.40 To better examine the role of different types of fat, a landmark study was conducted by Italian colleagues led by Zoccali et al. in a prospective cohort of 537 dialysis patients, in whom waist circumference was used as surrogate of intra-abdominal or visceral (truncal) fat.41 In this study, each 10-cm increase in waist circumference was associated with 10% and 37% higher all-cause and cardiovascular death.41

non-Hispanic Caucasian and African American subgroups. However, higher estimated lean body mass was not associated with improved survival in Hispanic patients or those with BMI > 35 kg/m2.40 To better examine the role of different types of fat, a landmark study was conducted by Italian colleagues led by Zoccali et al. in a prospective cohort of 537 dialysis patients, in whom waist circumference was used as surrogate of intra-abdominal or visceral (truncal) fat.41 In this study, each 10-cm increase in waist circumference was associated with 10% and 37% higher all-cause and cardiovascular death.41 To determine whether dry weight gain accompanied by an increase in muscle mass is associated with a survival benefit in a nationally representative 5-year cohort of 121,762 maintenance hemodialysis patients, 3-month averaged serum creatinine levels and their changes over time were used as muscle mass and as muscle mass change, respectively.42 Dry weight loss or gain over time exhibited a graded association with higher rates of mortality or survival, respectively, as did changes in serum creatinine level over time. Among a subcohort of 50,831 patients who survived the first 6 months, those who lost weight but had an increased serum creatinine level had a greater survival rate than those who gained weight but had a decreased creatinine level.42 These data suggest that there is a superiority of lean body mass to fat mass, in that larger body size with more muscle mass was associated with better survival, whereas a discordant muscle gain with weight loss over time conferred greater survival benefit as compared with weight gain while losing muscle.42 Additional analyses of the same cohort using more sophisticated analytic techniques confirmed the superiority of muscle mass while overall weight gain or loss maintained parallel associations with survival and mortality, respectively.43 A decline in muscle mass appeared to be a stronger predictor of mortality than weight loss. These studies suggest that a considerable proportion of the obesity paradox in dialysis patients might be explained by the survival benefits of greater muscle mass.43 In a large epidemiologic study by Beddhu et al.,44 24-hour urinary creatinine excretion was used as a measure of muscle mass in 70,028 patients who initiated hemodialysis in the US over 5 years (January 1995 to December 1999), and the outcomes of hemodialysis patients with high BMI and normal or high muscle mass (inferred low body fat) and high BMI and low muscle mass (inferred high body fat) were compared.

s used as a measure of muscle mass in 70,028 patients who initiated hemodialysis in the US over 5 years (January 1995 to December 1999), and the outcomes of hemodialysis patients with high BMI and normal or high muscle mass (inferred low body fat) and high BMI and low muscle mass (inferred high body fat) were compared. The investigators found that patients with high BMI (>25 kg/m2) had 15% lower hazard of death, but that patients who had more muscle mass had even greater survival, whereas patents with high BMI but lower muscle mass had a 14% to 19% higher all-cause and cardiovascular mortality.44 According to the authors’ interpretation of their data, the protective effect conferred by high BMI is limited to higher muscle mass, as patients with higher BMI with inferred high body fat exhibited increased and not decreased mortality.44 Hence, given the commonality of the muscle mass superiority despite mixed data about fat, controlled trials of muscle-enhancing interventions in patients receiving dialysis are warranted.

ed to higher muscle mass, as patients with higher BMI with inferred high body fat exhibited increased and not decreased mortality.44 Hence, given the commonality of the muscle mass superiority despite mixed data about fat, controlled trials of muscle-enhancing interventions in patients receiving dialysis are warranted. It is important to note that obesity paradox associations are not only observed in hemodialysis patients but have also been seen in peritoneal dialysis patients.45 In a cohort of 10,896 peritoneal dialysis patients, the association of baseline serum creatinine level as a surrogate of muscle mass and its change during the first 3 months thereafter with all-cause mortality was examined.46 Compared with patients with serum creatinine levels of 8.0 to <10 mg/dl, patients with serum creatinine levels of <4.0 mg/dl and 4.0 to <6 mg/dl had 36% and 19% higher risks of death, respectively, whereas patients with serum creatinine levels of 10.0 to <12 mg/dl, 12.0 to <14 mg/dl, and >14.0 mg/dl had 12%, 29%, and 36% lower risks of death, respectively. Decreases in serum creatinine level exceeding 1.0 mg/dl during the 3 months predicted an additional increased risk of death. The investigators concluded that muscle mass reflected in serum creatinine levels were associated with survival in peritoneal dialysis patients.46

29%, and 36% lower risks of death, respectively. Decreases in serum creatinine level exceeding 1.0 mg/dl during the 3 months predicted an additional increased risk of death. The investigators concluded that muscle mass reflected in serum creatinine levels were associated with survival in peritoneal dialysis patients.46 Is the Obesity Paradox a Statistical Fallacy? It has been argued that the inverse association between BMI and mortality observed under the obesity paradox in dialysis patients may be a consequence of residual confounding. Thus, marginal structural model analysis, a technique that accounts for time-varying confounders, may be more appropriate to investigate this association.47, 48, 49, 50 In a recent study of the associations between BMI and all-cause mortality among 123,624 adult hemodialysis patients comprising 45% women and 32% African Americans, BMI showed a linear incremental inverse association with mortality across all models.18 Compared with the reference (BMI 25 to <27.5 kg/m2), a BMI of <18 kg/m2 was associated with a 3.2-fold higher death risk (hazard rate [HR] = 3.17, 95% confidence interval [CI] = 3.05−3.29).18 Furthermore, mortality risk was incrementally lower with increasing BMI levels, with the greatest survival advantage observed with a BMI of 40 to <45 kg/m2 (HR = 0.69, 95% CI = 0.64−0.75).18 This study suggested that the linear inverse relationship between BMI and mortality is robust across models, including marginal statistical model analyses that more completely account for time-varying confounders and biases.18

survival advantage observed with a BMI of 40 to <45 kg/m2 (HR = 0.69, 95% CI = 0.64−0.75).18 This study suggested that the linear inverse relationship between BMI and mortality is robust across models, including marginal statistical model analyses that more completely account for time-varying confounders and biases.18 Changes in Body Weight and Mortality CKD patients, and in particular, incident dialysis patients, may experience rapid weight loss in the first several months of starting dialysis. However, there are limited data on trends in weight changes over time and their associations with mortality in CKD patients. In a large contemporary cohort of 58,106 patients who initiated hemodialysis over the 5-year period of January 2007 to December 2011 and survived the first year of dialysis therapy, trends in weight changes during the first year of treatment, as well as associations of postdialysis weight change with all-cause mortality were examined.51 Patients' postdialysis weights rapidly decreased and reached a nadir at the 5th month of dialysis with an average decline of 2% from baseline, whereas obese patients, defined as those with a BMI > 30 kg/m2, did not reach a nadir and lost approximately 3.8% of their weight by the 12th month. Compared with the reference group (−2% to +2% change in weight), the mortality HRs (95% CI) of patients with −6% to −2% and greater than or equal to −6% weight loss during the first 5 months were 1.08 (1.02−1.14) and 1.14 (1.07−1.22), respectively.51 Moreover, the mortality HRs (95% CI) with +2 to +6% and +>6% weight gain during the 5th to 12th months were 0.91 (0.85−0.97) and 0.92 (0.86−0.99), respectively.51 The study concluded that in hemodialysis patients who survive the first year of hemodialysis, a decline in postdialysis weight is observed and reaches a nadir at the 5th month. In addition, an incrementally larger weight loss during the first month is associated with higher death risk, whereas weight gain is associated with greater survival during the 5th to 12th month but not in the first 5 months of dialysis therapy.51

in postdialysis weight is observed and reaches a nadir at the 5th month. In addition, an incrementally larger weight loss during the first month is associated with higher death risk, whereas weight gain is associated with greater survival during the 5th to 12th month but not in the first 5 months of dialysis therapy.51 Does Race Influence the Obesity Paradox? An interesting issue is whether the obesity paradox is affected by race or ethnicity in CKD patients or differs across geographic regions. Glanton et al.52 performed a historical cohort study in 151,027 incident dialysis patients and found that the obesity paradox was even stronger in African Americans. Johansen et al.53 has also examined whether BMI is associated with better survival in Asian Americans, Caucasians, African Americans, and Hispanics. Rick et al.54 evaluated whether higher BMI is more strongly associated with lower mortality among African Americans and Hispanics versus non-Hispanic Caucasians. In a cohort of 109,605 hemodialysis patients who comprised 39,090 African Americans, 17,417 Hispanics, and 53,098 non-Hispanic Caucasians, a higher BMI was linked with lower mortality across all racial/ethnic categories. Notably, a more potent association between higher BMI category and greater survival was observed among African American and Hispanic patients versus non-Hispanic Caucasians. Park et al.26 also sought to determine whether the body size−mortality association among hemodialysis patients is uniform across different races, particularly East Asian versus Caucasian and African American patients. Among 20,818 South Korean hemodialysis patients who were matched to 20,000 US hemodialysis patients (10,000 Caucasian and 10,000 African American patients), the investigators found that BMI level was inversely and linearly associated with mortality even among East Asian hemodialysis patients. In addition, the strength of the association between BMI and mortality was similar across the 3 racial/ethnic groups, suggesting that the obesity paradox is a universal phenomenon, irrespective of race, in hemodialysis patients.

versely and linearly associated with mortality even among East Asian hemodialysis patients. In addition, the strength of the association between BMI and mortality was similar across the 3 racial/ethnic groups, suggesting that the obesity paradox is a universal phenomenon, irrespective of race, in hemodialysis patients. Biologic Plausibility for the Obesity Paradox and Causality The obesity paradox is not restricted to advanced CKD and has also been observed in other populations including the elderly individuals16, 55 and in those with chronic heart failure,15, 56 among others. It has been argued that the obesity paradox, along with other paradoxes such as the lipid paradox, are a hallmark of chronic disease states or conditions that are associated with wasting, sarcopenia, and full-blown cachexia.57, 58 Although we argue that weight loss is causally related to death in CKD and other similar conditions, others have questioned whether weight loss is truly in the causal pathway between CKD-associated protein−energy wasting and death, as shown in Figure 3. According to the alternative hypothesis, weight loss and gain are epiphenomena in that they occur when a patient does poorly or favorably, respectively, whereas changes in weight or body composition are not causally related to survival (Figure 3, models 2 and 3). Hence, the inability of observational studies to prove causality is massively limited, no matter what kind of multivariate techniques are used.59Figure 3 Three hypothetical “causal” models of the weight loss and death associations in chronic kidney disease (CKD) and the role of protein−energy wasting (PEW).

s 2 and 3). Hence, the inability of observational studies to prove causality is massively limited, no matter what kind of multivariate techniques are used.59Figure 3 Three hypothetical “causal” models of the weight loss and death associations in chronic kidney disease (CKD) and the role of protein−energy wasting (PEW). Despite the foregoing view about causal inference in epidemiological studies of the obesity paradox, there are some criteria required for making the leap from associations to causation, the most well-known of which were presented in the 1965 article of Sir Austin Bradford Hill, “The Environment and Disease: Association or Causation,”60 in which several benchmarks—subsequently refined and expanded to 9 criteria—that “suggest” causality were listed (Table 1). The most important one is the “temporal relationship,” which indicates that the cause or “exposure,” say, weight loss to overt cachexia, should precede the effect or “outcome,” say, death. However, an inherent problem in studying the causes of death is the fact that death is inherently the final event preceding any risk factor or condition. Hence, temporality is universally present in this association and cannot discern causality.Table 1 Hill’s considerations60 for the inference of causality in the obesity paradox

n inherent problem in studying the causes of death is the fact that death is inherently the final event preceding any risk factor or condition. Hence, temporality is universally present in this association and cannot discern causality.Table 1 Hill’s considerations60 for the inference of causality in the obesity paradox Benchmark Definition/comments Application to the obesity paradox 1. Temporalitya The cause (exposure) must precede the effect (outcome). PRO: Dialysis patients who gain more edema-free (dry) weight live longer. Wasting and weight loss precedes death and is associated with higher death risk. CON: Death is inherently the final event. Death may also occur during weight gain (e.g., upon nutritional support). 2. Strength of association Stronger association may make causality more likely. PRO: Most studies indicate strong and consistent associations between higher BMI and greater survival, especially in dialysis patients. CON: The reported strengths of the associations are not consistent. CKD stage and renal insufficiency severity confound the association. 3. Biological gradient (dose−response) Greater exposure increases the incidence or magnitude of the effect. PRO: Greater weight loss may be associated with higher likelihood of death, whereas incrementally higher BMI is associated with better survival. CON: The wasting severity is inconsistent and in some studies even has a weak association with death risk especially in less severe stages of CKD. 4. Consistency The association can be replicated in studies in different settings using different methods. PRO: The obesity paradox is observed in both hemodialysis and peritoneal dialysis as well as in more advanced stages of NDD-CKD. CON: Some types of weight loss such as intentional weight loss may not be associated with higher death risk. 5. Biologic plausibility The association is consistent with known biological or pathological processes. PRO: Higher fat and muscle mass may provide better cardiovascular profiles, whereas weight loss may lead to thromboembolic events, arrhythmia, sudden cardiac death, immune system disorders, and higher rates of cardiovascular and infectious disease events and death. CON: There is essentially no clear pathophysiologic pathway to explain the protective effects of higher muscle and fat mass. 6. Experimentation The putative effect can be altered (prevented or mitigated) by an experimental regimen. PRO: In some animal models of CKD, starvation and weight loss can lead to death.

death. CON: There is essentially no clear pathophysiologic pathway to explain the protective effects of higher muscle and fat mass. 6. Experimentation The putative effect can be altered (prevented or mitigated) by an experimental regimen. PRO: In some animal models of CKD, starvation and weight loss can lead to death. Improving wasting in human dialysis patients appears to improve survival. CON: The current CKD animal models are scarce and not convincing. There are very few trials of nutritional support interventions in human CKD subjects, and survival is rarely examined. 7. Specificity A single cause produces the effect without other pathways. PRO: Preceding wasting and weight loss can fully explain death events. CON: Weight loss is only 1 of the correlates of protein−energy wasting and may be an epiphenomenon. It is not clear how to explain survival advantage of obesity or weight gain. 8. Biologic coherence The association is consistent with the natural history of the disease or laboratory findings. PRO: A lower risk of death should result from preventing weight loss or by nutritional support in CKD patients. CON: Death events in CKD are mainly cardiovascular or infectious and have little to do with wasting and weight loss. 9. Analogy The effect of similar factors may be considered in other populations or under different settings. PRO: Wasting, fat, and muscle mass loss precede death in other chronic disease states such as heart failure, COPD, and metastatic cancer. CON: There is little biologically plausible analogy in death due to other conditions, such as cardiovascular (atherosclerosis) or cancer death. Each causality benchmark is examined for the cachexia-death association. BMI, body mass index; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; NDD, non–dialysis dependent.

biologically plausible analogy in death due to other conditions, such as cardiovascular (atherosclerosis) or cancer death. Each causality benchmark is examined for the cachexia-death association. BMI, body mass index; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; NDD, non–dialysis dependent. a Temporality is the only requisite condition of causality.

biologically plausible analogy in death due to other conditions, such as cardiovascular (atherosclerosis) or cancer death. Each causality benchmark is examined for the cachexia-death association. BMI, body mass index; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; NDD, non–dialysis dependent. a Temporality is the only requisite condition of causality. It is important to note that even though Hill’s criteria can be used to carefully shift interpretations from associations toward that of causation, epidemiology can never prove causality. Table 1 lists pros and cons pertaining to each of Hill’s 9 considerations. Given the observational nature of most of these benchmarks in the obesity paradox, the current state of knowledge does not sufficiently confirm that a higher body mass or even weight loss or gain are the main drivers of the longevity in advanced CKD. Nevertheless, the obesity paradox is not restricted to CKD but also exists in chronic heart failure, chronic obstructive pulmonary disease, cancer, AIDS, rheumatoid arthritis, and in elderly individuals. These populations apparently have slowly progressive to full-blown wasting and significantly greater short-term mortality than the general population.57 Hence, the consistency of the associative data, the remarkable strength of the obesity paradox, the early occurrence of death following progressive weight loss in CKD, and emerging evidence from basic science and animal models suggest that the causality element may be present and may soon be identified. Nevertheless, it is important to note that Hill’s criteria have been applied primarily in cases in which clinical trials are not ethically or logistically feasible, such as smoking and lung cancer,61 whereas this is not quite the case with the obesity question in CKD, where the plausibility of Hill’s criteria should not dissuade from conducting randomized controlled trials related to weight management or other nutritional interventions.62

re not ethically or logistically feasible, such as smoking and lung cancer,61 whereas this is not quite the case with the obesity question in CKD, where the plausibility of Hill’s criteria should not dissuade from conducting randomized controlled trials related to weight management or other nutritional interventions.62 Putative Pathophysiology of the Obesity Paradox Several hypotheses have been advanced to explain a biologically plausible model for the obesity paradox in CKD (Figure 4). The leading hypothesis pertains to protein−energy wasting (PEW), which is frequently observed in patients with advanced CKD.63, 64 The pathophysiology of PEW in CKD is related to the induction of inflammatory processes,65, 66, 67 such as activation of inflammatory cytokines including interleukin-6 and/or tumor necrosis factor−α, that suppress appetite and promote muscle breakdown and subsequent hypoalbuminemia.68 Loss of muscle and fat mass and inflammation may subsequently lead to heightened risk of cardiovascular disease and death via pathways related to vascular endothelial damage.69, 70, 71 Animal models also suggest that malnutrition may precipitate inflammation.72 As such, the malnutrition−inflammation−cachexia syndrome is thought to contribute to the obesity and other paradoxes in CKD and other chronic disease states.73Figure 4 Putative mechanisms of the survival advantages of obesity in chronic kidney disease (CKD). BP, blood pressure.

nutrition may precipitate inflammation.72 As such, the malnutrition−inflammation−cachexia syndrome is thought to contribute to the obesity and other paradoxes in CKD and other chronic disease states.73Figure 4 Putative mechanisms of the survival advantages of obesity in chronic kidney disease (CKD). BP, blood pressure. Obesity may potentially attenuate the magnitude of PEW and thereby provide protection against the downstream sequelae of inflammation such as cardiovascular disease. For example, patients with greater adipose tissue mass may be at lower risk of developing PEW in the context of malnutrition due to greater energy and/or protein reserves. In contrast, patients with poor nutrition may be more vulnerable to the ill effects of inflammation.64, 74 Lowrie et al.75 proposed that in the setting of inflammation, protein stores may be harnessed to restore injured tissues and mitigate inflammatory conditions. This may in part explain the lower mortality observed among dialysis patients with a high BMI or creatinine concentration but with low nutritional reserves.

, 74 Lowrie et al.75 proposed that in the setting of inflammation, protein stores may be harnessed to restore injured tissues and mitigate inflammatory conditions. This may in part explain the lower mortality observed among dialysis patients with a high BMI or creatinine concentration but with low nutritional reserves. It is important to highlight that the time discrepancies observed among competing risk factors for death are a likely an explanatory factor for the obesity paradox in advanced CKD. In long-term studies of the general population residing in industrialized nations, overnutrition has been observed to be a predictor of cardiovascular disease and death. However, in developing countries, which comprise a large proportion of the global population, undernutrition is a powerful determinant of reduced life expectancy.76 Similarly, survival advantages that exist in obese CKD patients may, in the short term, outweigh the harmful effects of obesity on cardiovascular disease in the long term. Indeed even in non-CKD patients, obesity may confer certain short-term benefits even though it is associated with poorer outcomes long term.77 Given that dialysis patients have an extremely poor short-term survival, with a large proportion of deaths occurring within the first 5 years of initiating dialysis,78, 79 the long-term effects of obesity as a traditional cardiovascular and mortality risk factor may be overwhelmed by the short-term ill effects of malnutrition and inflammation. It is important to note that in peritoneal dialysis patients, the obesity paradox has been less consistent, which may be related to the said differences in the follow-up time. For instance, Snyder et al.80 showed that in a 5-year (1995−2000) cohort of 418,021 US dialysis patients including 11% peritoneal dialysis patients, the likelihood of peritoneal dialysis modality assignment at dialysis initiation was 23% to 27% lower in overweight (BMI 25 to <30 kg/m2) and obese (BMI > 30 kg/m2) patients, respectively, whereas overweight and obese peritoneal dialysis patients still exhibited better survival in the first several years than those with lower BMI.

od of peritoneal dialysis modality assignment at dialysis initiation was 23% to 27% lower in overweight (BMI 25 to <30 kg/m2) and obese (BMI > 30 kg/m2) patients, respectively, whereas overweight and obese peritoneal dialysis patients still exhibited better survival in the first several years than those with lower BMI. However, in an 11-year (April 1991 to March 2002) cohort of 9679 Australian peritoneal dialysis patients, McDonald et al81 reported that obesity was associated with 36% higher mortality and 17% higher dialysis technique failure except among patients of New Zealand Maori/Pacific Islander origin, for whom there was no significant relationship between BMI and death during peritoneal dialysis treatment. Obesity may also be associated with better short-term hemodynamic stability. Many CKD patients on dialysis experience heart failure and/or fluid overload. Despite having similar pulmonary capillary wedge pressure and cardiac indices, overweight and obese patients with heart failure tend to have higher systolic blood pressure values,56 and thus may have better resilience against large volumes and faster rates of ultrafiltration during dialysis and lower likelihood of transient hypotension. This may attenuate sympathetic and renin−angiotensin−aldosterone activity82 which are linked with poor outcomes in heart failure and CKD patients.83 This bears particular relevance, as hypotension and subsequent myocardial stunning during the hemodialysis procedure may contribute to the extremely high cardiovascular mortality of ESRD patients.84, 85, 86

renin−angiotensin−aldosterone activity82 which are linked with poor outcomes in heart failure and CKD patients.83 This bears particular relevance, as hypotension and subsequent myocardial stunning during the hemodialysis procedure may contribute to the extremely high cardiovascular mortality of ESRD patients.84, 85, 86 Cytokine alterations may also contribute to better outcomes in obese patients. Adipose tissue produces tumor necrosis factor−α receptors, which are elevated in CKD patients and may lead to cardiac insult via pro-apoptotic and negative inotropic effects.71, 87 Conversely, increased tumor necrosis factor−α receptors may also play a cardioprotective role by neutralizing the adverse effect of tumor necrosis factor−α.88 In the context of adipose accumulation, uremic toxins may also be more effectively sequestered in these tissues. In addition, loss of weight and adipose tissue were found to be associated with increased release of circulating lipophilic hexachlorobenzene and other chlorinated hydrocarbons.89 This may in part explain why loss of body fat is associated with higher risk of death in ESRD patients.38 On average, obese patients have higher lipid and lipoprotein concentrations; given that lipopolysaccharide levels are elevated in fluid overload,90, 91 a higher concentration of lipoproteins (which bind to lipopolysaccharides) may mitigate the adverse sequelae of circulating endotoxins.91 Finally, platelet activation may be associated with high mortality risk in dialysis patients with PEW; it has been suggested that relative thrombocytosis in the context of an unfavorable malnutrition−inflammation−cachexia syndrome profile may lead to greater thromboembolism, cardiovascular disease, and death.92

xins.91 Finally, platelet activation may be associated with high mortality risk in dialysis patients with PEW; it has been suggested that relative thrombocytosis in the context of an unfavorable malnutrition−inflammation−cachexia syndrome profile may lead to greater thromboembolism, cardiovascular disease, and death.92 Concluding Remarks The seemingly counterintuitive obesity paradox is commonly observed in chronic disease states and conditions associated with wasting, such as advanced CKD. Studying similarities between CKD and other populations with a reverse epidemiology of cardiovascular risk may help to reveal common pathophysiologic mechanisms of the obesity paradox, leading to a major shift in clinical medicine and public health beyond conventional paradigms. Future studies that will advance our understanding of the existence, etiology, and components of the obesity paradox, as well as the role of PEW and the malnutrition−inflammation−cachexia syndrome in its development in advanced CKD, remain of paramount importance. Malnutrition and inflammation may be potentially modifiable, and as such may result in improved clinical outcomes. More research is needed to define all of the populations with versus without the obesity paradox, as this will drive future nutritional and therapeutic management decisions in patients at risk. These efforts may eventually lead to novel therapeutic approaches, including nutritional interventions that would improve the short-term and long-term survival of CKD and other vulnerable populations.

ut the obesity paradox, as this will drive future nutritional and therapeutic management decisions in patients at risk. These efforts may eventually lead to novel therapeutic approaches, including nutritional interventions that would improve the short-term and long-term survival of CKD and other vulnerable populations. Disclosure KKZ has received honoraria and/or other support from Abbott, Abbvie, Alexion, Amgen, American Society of Nephrology, AstraZeneca, AVEO Oncology, Chugai, DaVita, Fresenius, Genentech, Haymarket Media, Hofstra Medical School, International Federation of Kidney Foundations, International Society of Hemodialysis, International Society of Renal Nutrition & Metabolism, Japanese Society of Dialysis Therapy, Hospira, Kabi, Keryx, Novartis, National Institutes of Health, National Kidney Foundation, OPKO, Pfizer, Relypsa, Resverlogix, Sandoz, Sanofi, Shire, Vifor, UpToDate, and ZS-Pharma. All the other authors declared no competing interests. The findings and conclusions in this report are those of the authors and not necessarily of the World Kidney Day, of which KKZ serves as a member of the steering committee. Acknowledgments KKZ is supported by the National Institute of Diabetes, Digestive and Kidney Disease grants K24-DK091419, and philanthropist grants from Dr. Joseph Lee, Mr. Lois Chang, Dr Nick Vaziri, and AVEO. CMR is supported by the National Institute of Diabetes, Digestive and Kidney Disease grants K23-DK102903 and a philanthropist grant from Dr. Joseph Lee.

Measures of protein and energy stores such as body size,1, 2, 3, 4, 5, 6, 7 muscle mass,8, 9 fat mass,8, 9 serum albumin,10, 11, 12 and cholesterol13 levels are strong predictors of survival in dialysis patients. The International Society of Renal Nutrition and Metabolism (ISRNM) developed objective criteria for the definition of the Protein-Energy Wasting (PEW) syndrome in dialysis and chronic kidney disease (CKD) patients.14 Since PEW syndrome imposes high morbidity and mortality in dialysis patients, these efforts to define specific criteria for this syndrome are laudable. The PEW syndrome might be considered a specific “cachectic” condition of relevance for kidney disease and related chronic conditions. Nonetheless, whether these criteria are reflective of protein or energy stores or whether they associate with mortality in the general and moderate CKD populations have not been examined. Therefore, we tested the hypothesis that the PEW syndrome criteria reflect protein and/or energy wasting and associate with increased mortality in the general and CKD populations using the 1999 to 2004 National Health and Nutrition Examination Survey (NHANES) data. Materials and Methods Study Population and Baseline Data The National Center for Health Statistics (NCHS) is conducting NHANES to sample a representative population of the noninstitutionalized US population. NHANES data collection details have been published elsewhere.15

Measures of protein and energy stores such as body size,1, 2, 3, 4, 5, 6, 7 muscle mass,8, 9 fat mass,8, 9 serum albumin,10, 11, 12 and cholesterol13 levels are strong predictors of survival in dialysis patients. The International Society of Renal Nutrition and Metabolism (ISRNM) developed objective criteria for the definition of the Protein-Energy Wasting (PEW) syndrome in dialysis and chronic kidney disease (CKD) patients.14 Since PEW syndrome imposes high morbidity and mortality in dialysis patients, these efforts to define specific criteria for this syndrome are laudable. The PEW syndrome might be considered a specific “cachectic” condition of relevance for kidney disease and related chronic conditions. Nonetheless, whether these criteria are reflective of protein or energy stores or whether they associate with mortality in the general and moderate CKD populations have not been examined. Therefore, we tested the hypothesis that the PEW syndrome criteria reflect protein and/or energy wasting and associate with increased mortality in the general and CKD populations using the 1999 to 2004 National Health and Nutrition Examination Survey (NHANES) data. Materials and Methods Study Population and Baseline Data The National Center for Health Statistics (NCHS) is conducting NHANES to sample a representative population of the noninstitutionalized US population. NHANES data collection details have been published elsewhere.15 In brief, trained study personnel conducted a home interview followed by an examination at a mobile examination center. Demographic and comorbidity data over the past year were by self-report. Participants were asked about their current and past weight 1 year ago (in pounds, without clothes or shoes) and whether any weight loss was intentional. Height, weight, mid-arm circumference, and triceps skinfold thickness were measured following standardized protocols.16 Body mass index (BMI) was calculated from measured height and weight. Mid-arm muscle circumference (MAMC) was calculated as mid-arm circumference (mm) – (3.14 × triceps skin fold [mm]). Whole-body dual-energy x-ray absorptiometry (DXA) scans were performed with a Hologic QDR-4500A fanbeam densitometer (Hologic, Inc., Bedford, MA).17 Hologic software version 8.26:a3* was used to administer all scans. A computer-assisted dietary interview (CADI) system was administered by trained study personnel to obtain 24-hour dietary recalls.18 The interview files were imported into the University of Texas Food Intake Analysis System (FIAS) for coding. FIAS version 3.99 with the USDA 1994−1998 Survey Nutrient Database was used to code and report the dietary data. Serum albumin was measured with a bromcresol purple method using a Hitachi 917 analyzer (Roche Diagnostics, Indianapolis, IN), serum C-reactive protein using a Dade Behring Nephelometer II Analyzer (Dade Behring Diagnostics Inc., Somerville, NJ), and serum total cholesterol using a Hitachi 704 analyzer in the NHANES central laboratories following standardized protocols. The most recent Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation was used to estimate glomerular filtration rate (GFR).19

zer (Dade Behring Diagnostics Inc., Somerville, NJ), and serum total cholesterol using a Hitachi 704 analyzer in the NHANES central laboratories following standardized protocols. The most recent Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation was used to estimate glomerular filtration rate (GFR).19 Definitions of Individual Criteria and Categories for PEW Syndrome Definitions of individual criteria and categories for PEW syndrome are summarized in Table 1. The phrase “PEW criteria” refers to each individual variable, whereas the phrase “PEW category” refers to the grouping of each individual variable as serum chemistry, body mass, muscle mass, and dietary intake categories. Per ISRNM, the presence of at least 1 criteria in 3 of the 4 categories constitutes PEW syndrome. The ISRNM serum albumin criterion was based on the bromcresol green method measurement. In NHANES, serum albumin was measured by the bromcresol purple method, which is ∼0.55 g/dl lower than that obtained with the bromcresol green method.20 Hence, we used bromcresol purple serum albumin < 3.25 g/dl (which is approximate bromcresol green serum albumin < 3.8 g/dl) for the definition of PEW in this analysis. Self-reported weight (current and 1 year prior) and whether there was an intention to lose weight were used to estimate unintentional weight loss; >10% unintentional loss over the past year was used as a PEW criterion. The ISRNM panel considered BMI < 23 kg/m2 as a PEW criterion in dialysis patients, whereas BMI < 18.5 kg/m2 was considered undernutrition in the general population.21 There were only 197 participants with BMI < 18.5 kg/m2 in this population. The fifth percentile of BMI in this cohort was 19.93 kg/m2. As a result, we chose to use a cut-off of BMI < 20 kg/m2 for PEW definition in this study. In additional sensitivity analyses, we also examined a BMI definition of <18.5 kg/m2 as a PEW criterion. Total body fat% as measured by DXA was used. The ISRNM panel used total body fat percentage (fat%) of <10% as a PEW criterion. Because none of the study population had total body fat% of <10%, we dropped this criterion in this analysis. Based on the reported protein and calorie intake from the 24-hour dietary recall data and measured body weight, dietary protein intake and dietary energy intake were defined as ratios of reported dietary intakes and body weight in accordance with the ISRNM criteria (Table 1).Table 1 ISRNM PEW syndrome criteria and categoriesa

n the reported protein and calorie intake from the 24-hour dietary recall data and measured body weight, dietary protein intake and dietary energy intake were defined as ratios of reported dietary intakes and body weight in accordance with the ISRNM criteria (Table 1).Table 1 ISRNM PEW syndrome criteria and categoriesa Categories Criteria within categories Low serum chemistry Serum albumin < 3.8 g/dlb Serum cholesterol < 100 mg/dl Serum prealbumin < 30 mg/dlc Low body mass BMI < 23 kg/m2d Unintentional weight loss over time: 10% over 6 moe Body fat percentage < 10%f Low muscle mass Muscle wasting: reduced muscle mass 5% over 3 mo or 10% over 6 moc Reduced mid-arm muscle circumference area (reduction >10% in relation to 50th percentile of reference population) Creatinine appearancec Low dietary intakeˆ Dietary protein intake < 0.60 g/kg/d Dietary energy intake < 25 kcal/kg/d BMI, body mass index; CKD, chronic kidney disease; ISRNM, International Society of Renal Nutrition and Metabolism; NHANES, National Health and Nutrition Examination Survey; PEW, protein–energy wasting. a The term “criteria” refers to each individual variable, whereas “category” refers to grouping of criteria as serum chemistry, body mass, muscle mass, and dietary intake categories. b ISRNM panel used bromcresol green method. In NHANES, serum albumin was measured by bromcresol purple method (see text for details). c Data were not available in NHANES; hence, this criterion was not used in this analysis.

a The term “criteria” refers to each individual variable, whereas “category” refers to grouping of criteria as serum chemistry, body mass, muscle mass, and dietary intake categories. b ISRNM panel used bromcresol green method. In NHANES, serum albumin was measured by bromcresol purple method (see text for details). c Data were not available in NHANES; hence, this criterion was not used in this analysis. d ISRNM panel used BMI < 23 kg/m2 in dialysis patients. The current study used BMI < 20 kg/m2 (∼fifth percentile) in non-CKD and moderate CKD populations. e ISRNM panel used 10% weight loss over 6 months. f There were no patients with body fat < 10%; hence, this criterion was not used in this analysis. Protein and Energy Stores The ISRNM panel defined PEW syndrome as “the state of decreased body stores of protein and energy fuels (that is, body protein and fat masses).”14 Therefore, the face validity of the criteria used to define PEW syndrome requires that these criteria reflect either protein wasting or energy wasting. Muscle is the largest protein store, and fat is the largest energy store in the body. Hence, we examined the associations of each of the criteria with protein stores (as estimated by lean body mass measured by DXA scan) and energy stores (as estimated by fat mass measured by DXA scan) in the entire cohort and the CKD subpopulation. Mortality Data A Linked Mortality File through 31 December 2011 was created by NCHS using a probabilistic match between NHANES and National Death Index death certificate records.22

Protein and Energy Stores The ISRNM panel defined PEW syndrome as “the state of decreased body stores of protein and energy fuels (that is, body protein and fat masses).”14 Therefore, the face validity of the criteria used to define PEW syndrome requires that these criteria reflect either protein wasting or energy wasting. Muscle is the largest protein store, and fat is the largest energy store in the body. Hence, we examined the associations of each of the criteria with protein stores (as estimated by lean body mass measured by DXA scan) and energy stores (as estimated by fat mass measured by DXA scan) in the entire cohort and the CKD subpopulation. Mortality Data A Linked Mortality File through 31 December 2011 was created by NCHS using a probabilistic match between NHANES and National Death Index death certificate records.22 Statistical Analyses NHANES is based on a complex probability sample design. We used the svy suite of commands in Stata 13 (Stata Corporation, College Station, TX) and followed the NHANES analytical guidelines.23 Hence, all reported results (including descriptive statistics, linear, and Cox regression models) take survey weights into account. Descriptive statistics for baseline clinical characteristics and each PEW criterion are reported as means, SDs, or medians, 25th and 75th percentiles for numeric variables and as proportions for categorical variables.

Statistical Analyses NHANES is based on a complex probability sample design. We used the svy suite of commands in Stata 13 (Stata Corporation, College Station, TX) and followed the NHANES analytical guidelines.23 Hence, all reported results (including descriptive statistics, linear, and Cox regression models) take survey weights into account. Descriptive statistics for baseline clinical characteristics and each PEW criterion are reported as means, SDs, or medians, 25th and 75th percentiles for numeric variables and as proportions for categorical variables. Associations of Variables Contributing to the Definition of PEW Syndrome With Lean Body Mass and Fat Mass Linear regression analyses were used to relate lean body mass and fat mass separately to each individual variable contributing to the definition of PEW syndrome with adjustment for the covariates age, gender, race, education, smoking, and alcohol use in the entire cohort and CKD subpopulation. Relationships of the Number of PEW Syndrome Categories With Body Size and Body Composition As per ISRNM, a category is considered to be present if any 1 of the individual criteria within that category is present (Table 1). It has been suggested that inadequate dietary intake rarely contributes to protein or energy wasting in uremia.14 Hence the distributions of body size and body composition parameters by the number of PEW categories defined by nondietary categories alone were first examined. These were also re-examined by including the dietary category in the number of PEW categories.

ake rarely contributes to protein or energy wasting in uremia.14 Hence the distributions of body size and body composition parameters by the number of PEW categories defined by nondietary categories alone were first examined. These were also re-examined by including the dietary category in the number of PEW categories. In multivariate linear regression models, the number of PEW categories defined by nondietary categories alone were related to lean body mass and fat mass (measured by DXA scans). These analyses were repeated by including dietary category in the number of PEW categories. These models were adjusted for age, gender, race, education, smoking, and alcohol use. All analyses were performed in the entire cohort and the CKD subpopulation. Relationships of the Number of PEW Syndrome Categories With Mortality The associations of the number of nondietary categories with mortality in the entire and CKD subpopulations were examined. These analyses were repeated by including dietary category. These models were adjusted for age, gender, race, education, smoking, alcohol use, myocardial infarction (MI), congestive heart failure (CHF), stroke, diabetes, hypertension, lung disease, cancer, serum C-reactive protein, and serum bicarbonate. Proportional hazards assumptions for the Cox regression analyses were evaluated by comparing Cox regression coefficients for the first 18 months to Cox regressions after 18 months in time-dependent analyses.

re (CHF), stroke, diabetes, hypertension, lung disease, cancer, serum C-reactive protein, and serum bicarbonate. Proportional hazards assumptions for the Cox regression analyses were evaluated by comparing Cox regression coefficients for the first 18 months to Cox regressions after 18 months in time-dependent analyses. Results Of the 15,332 participants in 1999 to 2004 NHANES with age >20 years, 11,834 individuals with nonmissing data for estimated glomerular filtration rate, serum albumin and total cholesterol, MAMC, and mortality status were included. All of the reported results (means, medians, proportions, regression coefficients, hazard ratios, SDs, and confidence intervals) are survey weight adjusted. The mean age was 46.0 ± 13.4 years. Of the participants, 50.3% were men, and 9.5% were African American. The prevalence of CKD (estimated glomerular filtration rate < 60 ml/min/1.73 m2) was 6.7%. Of those with CKD, 91.6% were in stage 3. Baseline clinical characteristics of the study population by the number of PEW syndrome categories are summarized in Table 2. In general, those with ≥3 PEW categories were older and had a higher prevalence of comorbid conditions. However, those with 1 PEW category had the highest prevalence of diabetes, higher serum C-reactive protein levels, and higher BMI as well as body fat%.Table 2 Baseline clinical characteristicsa by number of PEW syndrome categories in entire cohort (N = 11,834)

e older and had a higher prevalence of comorbid conditions. However, those with 1 PEW category had the highest prevalence of diabetes, higher serum C-reactive protein levels, and higher BMI as well as body fat%.Table 2 Baseline clinical characteristicsa by number of PEW syndrome categories in entire cohort (N = 11,834) Number of PEW syndrome categories 0 (43.15%) 1 (46.85%) 2 (9.41%) ≥3 (0.69%) Demographics Age (yr) 43.2 ± 11.8 48.4 ± 14.3 46.8 ± 14.5 50.8 ± 16.6 Male (%) 60.8% 43.9% 35.7% 32.3% African American race (%) 8.4% 10.8% 8.1% 13.6% ≥High school education (%) 82.2% 78.2% 78.2% 60.0% Clinical parameters Myocardial infarction (%) 2.6% 4.2% 4.6% 6.6% Congestive heart failure (%) 1.1% 3.0% 2.1% 4.6% Stroke (%) 1.3% 3.1% 2.6% 2.5% Diabetes (%) 5.0% 9.7% 7.4% 6.9% Smoking (%) 50.8% 50.6% 47.6% 61.7% Alcohol use (%) 74.6% 65.9% 64.7% 63.1% Hypertension (%) 23.4% 35.4% 25.9% 35.2% Lung disease (%) 5.6% 7.9% 10.5% 16.9% Cancer (%) 6.4% 9.2% 9.2% 8.1% C-reactive protein (mg/l) 1.6 (0.7−3.5) 2.3 (0.9−5.0) 1.3 (0.5−3.2) 1.5 (0.4−6.1) Serum bicarbonate (mmol/l) 24.1 ± 1.7 24.1 ± 1.9 24.2 ± 1.9 24.6 ± 2.1 eGFR (ml/min/1.73 m2) 95.6 ± 15.0 92.0 ± 18.4 94.6 ± 19.3 89.7 ± 23.7 PEW variables Serum albumin (g/dl) 4.4 ± 0.2 4.3 ± 0.3 4.3 ± 0.3 4.2 ± 0.5 Total cholesterol (mg/dl) 202.6 ± 32.9 204.1 ± 35.2 196.2 ± 32.6 194.2 ± 51.7 Body mass index (kg/m2) 27.0 ± 3.2 28.5 ± 4.8 22.7 ± 3.7 19.4 ± 1.7 Weight change compared to last yr (%) 1.1 ± 5.8 0.9 ± 7.8 −1.8 ± 7.5 −2.2 ± 6.7 Body fat (%) 30.3 ± 6.0 35.2 ± 8.1 28.7 ± 7.3 26.0 ± 6.5 Mid-arm muscle circumference (cm) 27.3 ± 2.9 26.3 ± 3.4 22.0 ± 2.4 21.1 ± 1.8 Dietary protein intake (g/kg/d) 1.2 (1.0−1.6) 0.8 (0.6−1.1) 0.9 (0.7−1.4) 0.7 (0.5−0.9) Dietary energy intake (kcal/kg/d) 33.5 (28.8−40.8) 20.7 (16.3−24.7) 23.8 (19.3−38.3) 19.9 (18.0−22.3) Data are mean ± SD, median (interquartile range), or proportion as appropriate, adjusted by survey weight. All P values <0.001, except smoking (P = 0.14) and serum bicarbonate (P = 0.04). eGFR, estimated glomerular filtration rate; PEW, protein–energy wasting.

.8−40.8) 20.7 (16.3−24.7) 23.8 (19.3−38.3) 19.9 (18.0−22.3) Data are mean ± SD, median (interquartile range), or proportion as appropriate, adjusted by survey weight. All P values <0.001, except smoking (P = 0.14) and serum bicarbonate (P = 0.04). eGFR, estimated glomerular filtration rate; PEW, protein–energy wasting. a Survey weight−adjusted means, medians, or proportions. The prevalence of the individual PEW criteria in the entire cohort and CKD subpopulation are summarized in Table 3, Table 4, respectively. Low dietary energy intake and protein intake were among the most common conditions, whereas no participant had body fat < 10%. Low serum albumin, low serum cholesterol, low BMI, unintentional weight loss, and low MAMC criteria tended to be associated with lower protein stores and lower energy stores, although statistical significance was not achieved in all cases. In contrast, low dietary protein and energy intakes were associated with significantly higher protein and energy stores.Table 3 Prevalence and associationsa of individual PEW criteria with lean body mass and fat mass in entire cohort (N = 11,834)

stores, although statistical significance was not achieved in all cases. In contrast, low dietary protein and energy intakes were associated with significantly higher protein and energy stores.Table 3 Prevalence and associationsa of individual PEW criteria with lean body mass and fat mass in entire cohort (N = 11,834) Prevalence Lean body mass (kg) regression coefficientb (95% CI) Fat mass (kg) regression coefficientb (95% CI) Low serum albumin (<3.25 g/dl) 0.6% –3.5 (–7.5 to 0.4) –3.7 (–7.0 to –0.3) Low serum cholesterol (<100 mg/dl) 0.2% –2.5 (–6.8 to 1.8) –4.6 (–9.7 to 0.5) Low body mass index (<20 kg/m2) 5.9%c –9.7 (–10.3 to –9.1) –14.2 (–14.5 to –13.8) Unintentional weight loss (>10% over 1 yr) 2.4% –2.2 (–3.3 to –1.0) –3.4 (–4.8 to –1.9) Low body fat % (<10%) 0 NA NA Low mid-arm muscle circumference area 17.1% –8.6 (–9.0 to –8.3) –8.3 (–8.8 to –7.9) Low dietary protein intake (<0.60 g/kg/d) 15.2% 3.6 (3.0 to 4.2) 6.4 (5.8 to 7.1) Low dietary energy intake (<25 kcal/kg/d) 42.7% 4.3 (3.9 to 4.7) 7.3 (6.9 to 7.7) CI, confidence interval; NA, not applicable as none had body fat % <10%; NHANES, National Health and Nutrition Examination Survey; PEW, protein–energy wasting. a NHANES survey weight adjusted. b Each cell represents a separate model adjusted for age, gender, race, education, smoking, and alcohol use. c Fifth percentile of BMI was 19.93 kg/m2. Table 4 Prevalence and associationsa of individual PEW criteria with lean body mass and fat mass in CKD subpopulation (n = 1156)

Prevalence Lean body mass (kg) regression coefficientb (95% CI) Fat mass (kg) regression coefficientb (95% CI) Low serum albumin (<3.25 g/dl) 0.6% –3.5 (–7.5 to 0.4) –3.7 (–7.0 to –0.3) Low serum cholesterol (<100 mg/dl) 0.2% –2.5 (–6.8 to 1.8) –4.6 (–9.7 to 0.5) Low body mass index (<20 kg/m2) 5.9%c –9.7 (–10.3 to –9.1) –14.2 (–14.5 to –13.8) Unintentional weight loss (>10% over 1 yr) 2.4% –2.2 (–3.3 to –1.0) –3.4 (–4.8 to –1.9) Low body fat % (<10%) 0 NA NA Low mid-arm muscle circumference area 17.1% –8.6 (–9.0 to –8.3) –8.3 (–8.8 to –7.9) Low dietary protein intake (<0.60 g/kg/d) 15.2% 3.6 (3.0 to 4.2) 6.4 (5.8 to 7.1) Low dietary energy intake (<25 kcal/kg/d) 42.7% 4.3 (3.9 to 4.7) 7.3 (6.9 to 7.7) CI, confidence interval; NA, not applicable as none had body fat % <10%; NHANES, National Health and Nutrition Examination Survey; PEW, protein–energy wasting. a NHANES survey weight adjusted. b Each cell represents a separate model adjusted for age, gender, race, education, smoking, and alcohol use. c Fifth percentile of BMI was 19.93 kg/m2. Table 4 Prevalence and associationsa of individual PEW criteria with lean body mass and fat mass in CKD subpopulation (n = 1156) Prevalence Lean body mass (kg) regression coefficientb (95% CI) Fat mass (kg) regression coefficientb (95% CI) Low serum albumin (<3.25 g/dl) 1.3% –1.1 (–10.2 to 7.9) 0.3 (–6.8 to 7.3) Low serum cholesterol (<100 mg/dl) 0.03% –6.9 (–8.6 to –5.2) –8.6 (–10.4 to –6.8) Low body mass index (<20 kg/m2)c 4.9% –9.3 (–10.5 to –8.1) –15.4 (–16.9 to –13.8) Unintentional weight loss (>10% over 1 yr) 5.4% –1.8 (–4.6 to 1.1) –4.9 (–8.7 to –1.1) Low body fat % (<10%) 0 NA NA Low mid-arm muscle circumference area 15.8% –7.7 (–9.0 to –6.4) –8.8 (–10.1 to –7.5) Low dietary protein intake (<0.60 g/kg/d) 25.8% 3.4 (2.3 to 4.5) 6.0 (4.5 to 7.4) Low dietary energy intake (<25 kcal/kg/d) 65.4% 4.7 (3.6 to 5.8) 7.5 (6.1 to 9.0) CI, confidence interval; CKD, chronic kidney disease; NA, not applicable as none had body fat % <10%; NHANES, National Health and Nutrition Examination Survey; PEW, protein–energy wasting.

0 g/kg/d) 25.8% 3.4 (2.3 to 4.5) 6.0 (4.5 to 7.4) Low dietary energy intake (<25 kcal/kg/d) 65.4% 4.7 (3.6 to 5.8) 7.5 (6.1 to 9.0) CI, confidence interval; CKD, chronic kidney disease; NA, not applicable as none had body fat % <10%; NHANES, National Health and Nutrition Examination Survey; PEW, protein–energy wasting. a NHANES survey weight adjusted. b Each cell represents a separate model adjusted for age, gender, race, education, smoking, and alcohol use. c Fifth percentile of body mass index in entire cohort was 19.93 kg/m2. When only nondietary categories were considered, 22.2% of the population had 1 or more nondietary categories (Table 5). BMI and fat and lean body masses measured by DXA and MAMC were progressively lower with the presence of additional nondietary categories.Table 5 Body size and body composition characteristics by number of PEW categories defined by nondietary categories alone (N = 11,834) 0 Nondietary (78.83%)a 1 Nondietary (16.57%)a 2 Nondietary (4.57%)a 3 Nondietary (0.03%)a Body mass index 28.5 ± 4.0 23.8 ± 2.9 19.0 ± 1.4 17.2 ± 0.8 Mid-arm muscle circumference (cm2) 27.3 ± 3.1 22.8 ± 2.5 21.0 ± 1.9 20.3 ± 0.8 Lean body mass (kg) 52.3 ± 9.4 43.2 ± 7.0 37.6 ± 5.5 37.8 ± 2.9 Fat mass (kg) 28.3 ± 7.8 21.8 ± 5.4 13.6 ± 2.7 9.2 ± 3.1 PEW, protein–energy wasting. a Survey weight−adjusted proportions.

0 Nondietary (78.83%)a 1 Nondietary (16.57%)a 2 Nondietary (4.57%)a 3 Nondietary (0.03%)a Body mass index 28.5 ± 4.0 23.8 ± 2.9 19.0 ± 1.4 17.2 ± 0.8 Mid-arm muscle circumference (cm2) 27.3 ± 3.1 22.8 ± 2.5 21.0 ± 1.9 20.3 ± 0.8 Lean body mass (kg) 52.3 ± 9.4 43.2 ± 7.0 37.6 ± 5.5 37.8 ± 2.9 Fat mass (kg) 28.3 ± 7.8 21.8 ± 5.4 13.6 ± 2.7 9.2 ± 3.1 PEW, protein–energy wasting. a Survey weight−adjusted proportions. In contrast, when only dietary category was present, the mean BMI was 30.2 ± 4.4 kg/m2, which is in the obesity range (Table 6). Fat mass measured by DXA was also higher with only dietary category present, compared to when none of the categories were present (Table 6). The further presence of nondietary categories in addition to the dietary category was associated with lower BMI, fat and lean body masses, and MAMC (Table 6).Table 6 Body size and body composition of patients with dietary category alone compared to those with none of the categories or those with dietary category with additional nondietary categories (n = 10,137)a 0 Dietary or nondietary (50.65%)b Dietary alone (42.10%)b Dietary + 1 nondietary (6.47%)b Dietary + 2 or more nondietary (0.78%)b Body mass index (kg/m2) 27.0 ± 3.1 30.2 ± 4.4 25.4 ± 3.4 19.5 ± 1.6 Mid-arm muscle circumference (cm2) 27.3 ± 2.9 27.3 ± 3.2 22.8 ± 2.6 21.1 ± 1.8 Lean body mass (kg) 52.6 ± 8.5 51.9 ± 10.2 43.0 ± 7.1 37.0 ± 5.4 Fat mass (kg) 25.0 ± 6.0 32.3 ± 8.4 25.1 ± 6.0 14.7 ± 2.8 a Patients with only nondietary categories without dietary category being present were not included in this table. b Survey weight−adjusted proportions.

0 Dietary or nondietary (50.65%)b Dietary alone (42.10%)b Dietary + 1 nondietary (6.47%)b Dietary + 2 or more nondietary (0.78%)b Body mass index (kg/m2) 27.0 ± 3.1 30.2 ± 4.4 25.4 ± 3.4 19.5 ± 1.6 Mid-arm muscle circumference (cm2) 27.3 ± 2.9 27.3 ± 3.2 22.8 ± 2.6 21.1 ± 1.8 Lean body mass (kg) 52.6 ± 8.5 51.9 ± 10.2 43.0 ± 7.1 37.0 ± 5.4 Fat mass (kg) 25.0 ± 6.0 32.3 ± 8.4 25.1 ± 6.0 14.7 ± 2.8 a Patients with only nondietary categories without dietary category being present were not included in this table. b Survey weight−adjusted proportions. These relationships are more evident in the multivariate regression models relating the number of PEW syndrome categories with lean body and fat masses (Figure 1) in the entire cohort. When the numbers of PEW syndrome categories were defined by nondietary categories alone, there was a monotonic inverse relationship with lean body mass (Figure 1a) and fat mass (Figure 1c). However, when dietary category was included, the presence of any 1 of the categories was associated with higher lean body mass (Figure 1b) and higher fat mass (Figure 1d). The further presence of additional categories was associated with lower lean body mass (Figure 1b) and fat mass (Figure 1c). Results were similar in the CKD subpopulation (Figure 2a−d).Figure 1 Associations of the number of protein−energy wasting (PEW) syndrome categories with lean body mass and fat mass measured by dual-energy x-ray absorptiometry scans in multivariate linear regression models in the entire cohort (N = 11,834). (a) Nondietary categories alone and lean body mass. (b) Dietary category included and lean body mass. (c) Nondietary categories alone and fat mass. (d) Dietary category included and fat mass.

s measured by dual-energy x-ray absorptiometry scans in multivariate linear regression models in the entire cohort (N = 11,834). (a) Nondietary categories alone and lean body mass. (b) Dietary category included and lean body mass. (c) Nondietary categories alone and fat mass. (d) Dietary category included and fat mass. Figure 2 Associations of the number of protein−energy wasting (PEW) syndrome categories with lean body mass and fat mass measured by dual-energy x-ray absorptiometry scans in multivariate linear regression models in the chronic kidney disease (CKD) subpopulation (n = 1156). (a) Nondietary categories alone and lean body mass. (b) Dietary category included and lean body mass. (c) Nondietary categories alone and fat mass. (d) Dietary category included and fat mass.

-ray absorptiometry scans in multivariate linear regression models in the chronic kidney disease (CKD) subpopulation (n = 1156). (a) Nondietary categories alone and lean body mass. (b) Dietary category included and lean body mass. (c) Nondietary categories alone and fat mass. (d) Dietary category included and fat mass. Compared to those individuals with none of the nondietary categories, the presence of even 1 nondietary category was associated with increased mortality in the entire cohort (Figure 3a) and the CKD subpopulation (Figure 3c). On the other hand, the presence of any 1 of the dietary or nondietary categories was not associated with increased mortality in the entire population (Figure 3b) or the CKD subpopulation (Figure 3d). The presence of additional categories conferred higher mortality risk, but these relationships appear to be stronger when nondietary categories alone were used (Figure 3a−d).Figure 3 Mortality associations of the number of protein−energy wasting (PEW) syndrome categories with mortality in multivariate Cox regression models in the entire cohort (N = 11,834) and chronic kidney disease (CKD) subpopulation (n = 1156). (a) Nondietary categories alone and mortality in the entire cohort. (b) Dietary category included and mortality in the entire cohort. (c) Nondietary categories alone and mortality in the CKD subpopulation.* (d) Dietary category included and mortality in the CKD subpopulation.$∗There was only 1 observation with 3 nondietary categories present in the CKD subpopulation and hence this observation was included as ≥2 categories in this figure). $There were only 19 observations with ≥3 categories (1 dietary and 2 or 3 nondietary) present in the CKD subpopulation and hence these observations were included as ≥2 categories in this figure.

dietary categories present in the CKD subpopulation and hence this observation was included as ≥2 categories in this figure). $There were only 19 observations with ≥3 categories (1 dietary and 2 or 3 nondietary) present in the CKD subpopulation and hence these observations were included as ≥2 categories in this figure. In sensitivity analyses, when defining BMI criteria as <18.5 kg/m2, the relationships of the number of PEW syndrome nondietary categories with lean body mass (Supplementary Figure 1a), fat mass (Supplementary Figure 1b), and mortality (Supplementary Figure 1c) were similar. Discussion The ISRNM panel defined PEW syndrome as “the state of decreased body stores of protein and energy fuels (that is, body protein and fat masses).”14 The results of this study indicate that nondietary definitions are reflective of protein and/or energy wasting in the general and moderate CKD populations. However, dietary criteria as defined by the ISRNM panel do not appear to be reflective of either protein wasting or energy wasting in the general or moderate CKD population.

ts of this study indicate that nondietary definitions are reflective of protein and/or energy wasting in the general and moderate CKD populations. However, dietary criteria as defined by the ISRNM panel do not appear to be reflective of either protein wasting or energy wasting in the general or moderate CKD population. The strong associations of low dietary protein and energy intakes estimated from 24-hour dietary recalls with higher lean body mass and fat mass are likely because of mathematical coupling; as the dietary intakes were normalized to body weight, lower dietary protein and energy intakes reflect higher body weight and hence, higher lean body mass and fat mass. It is possible that other dietary cut-offs normalized to body weight or height might be related to protein or energy wasting, and those need to be examined in future studies in the general and moderate CKD populations. Furthermore, whether the ISRNM dietary definitions are indicative of protein or energy wasting in the dialysis population also needs to be tested in future studies. In contrast, low serum albumin, low serum cholesterol, low BMI, unintentional weight loss, and low MAMC were associated with lower lean body mass and/or fat mass, suggesting the face validity of these variables as indicators of PEW syndrome.

The strong associations of low dietary protein and energy intakes estimated from 24-hour dietary recalls with higher lean body mass and fat mass are likely because of mathematical coupling; as the dietary intakes were normalized to body weight, lower dietary protein and energy intakes reflect higher body weight and hence, higher lean body mass and fat mass. It is possible that other dietary cut-offs normalized to body weight or height might be related to protein or energy wasting, and those need to be examined in future studies in the general and moderate CKD populations. Furthermore, whether the ISRNM dietary definitions are indicative of protein or energy wasting in the dialysis population also needs to be tested in future studies. In contrast, low serum albumin, low serum cholesterol, low BMI, unintentional weight loss, and low MAMC were associated with lower lean body mass and/or fat mass, suggesting the face validity of these variables as indicators of PEW syndrome. The ISRNM panel definition of PEW syndrome requires the presence of all 3 nondietary categories or dietary category plus 2 nondietary categories. As is evident from Table 5 and Figures 1 and 2, presence of all 3 nondietary categories was associated with significantly lower lean body mass and fat mass. It was also very strongly associated with increased mortality (Figure 3).

presence of all 3 nondietary categories or dietary category plus 2 nondietary categories. As is evident from Table 5 and Figures 1 and 2, presence of all 3 nondietary categories was associated with significantly lower lean body mass and fat mass. It was also very strongly associated with increased mortality (Figure 3). The presence of dietary category in addition to 2 nondietary categories was also associated with significantly lower lean body mass and fat mass and mortality (Table 6, Figure 1, Figure 2, Figure 3). However, as is evident from Table 6, inclusion of dietary category undermines the face validity of the PEW syndrome criteria as a measure of protein or energy wasting. When dietary category alone was present, the mean BMI was in the obesity range. The association of dietary category plus 2 additional nondietary categories with lower protein or energy stores was driven by the presence of the 2 nondietary categories (Table 6). Indeed, as shown in Figure 1, the lean body mass (−11.5 kg) and fat mass (−14.8 kg) were lower when 2 of the nondietary criteria were present than when 3 categories including dietary category were present (lean body mass –9.3 kg and fat mass –11.7 kg). Therefore, a modified definition PEW syndrome as the presence of 2 of 3 nondietary categories is likely a better indicator of protein or energy wasting than a definition of 3 of 4 categories that include dietary category.

categories including dietary category were present (lean body mass –9.3 kg and fat mass –11.7 kg). Therefore, a modified definition PEW syndrome as the presence of 2 of 3 nondietary categories is likely a better indicator of protein or energy wasting than a definition of 3 of 4 categories that include dietary category. Both muscle mass and serum albumin are commonly considered nutritional markers and are known to be associated with better survival in the general, CKD, and dialysis populations.10, 24, 25, 26 Even though much of the focus is on the associations of higher BMI with increased mortality BMI in the general population, it is also known to have a “U”-shaped association with mortality in that population.27, 28 Similarly, the U-shaped association of total cholesterol with mortality in the general population is also well known.29 The results of the current study show that the presence of PEW syndrome defined by a combination of the above criteria was strongly associated with increased mortality in the entire cohort and the moderate CKD subpopulation. Furthermore, the valid assessment of nutrient intake with dietary recalls or questionnaires even in research settings might be difficult. This might be even more difficult in routine clinical practice. Thus, using dietary variables as diagnostic criteria for PEW syndrome lacks face validity.

Both muscle mass and serum albumin are commonly considered nutritional markers and are known to be associated with better survival in the general, CKD, and dialysis populations.10, 24, 25, 26 Even though much of the focus is on the associations of higher BMI with increased mortality BMI in the general population, it is also known to have a “U”-shaped association with mortality in that population.27, 28 Similarly, the U-shaped association of total cholesterol with mortality in the general population is also well known.29 The results of the current study show that the presence of PEW syndrome defined by a combination of the above criteria was strongly associated with increased mortality in the entire cohort and the moderate CKD subpopulation. Furthermore, the valid assessment of nutrient intake with dietary recalls or questionnaires even in research settings might be difficult. This might be even more difficult in routine clinical practice. Thus, using dietary variables as diagnostic criteria for PEW syndrome lacks face validity. It is generally considered that the hypercatabolism of uremia (induced by uremic toxins, inflammation,30 oxidative stress, insulin resistance,31 metabolic acidosis,32 and the dialysis procedure itself33) is the cause of PEW syndrome in the dialysis population.34, 35, 36 It is unclear to what extent these factors play a role in the causation of PEW syndrome in the general and moderate CKD populations. Further studies are warranted to determine the causes of PEW syndrome in these populations.

dialysis procedure itself33) is the cause of PEW syndrome in the dialysis population.34, 35, 36 It is unclear to what extent these factors play a role in the causation of PEW syndrome in the general and moderate CKD populations. Further studies are warranted to determine the causes of PEW syndrome in these populations. The strengths of the study include the use of NHANES, a national survey designed to obtain a representative sample of the noninstitutionalized US population. Furthermore, data collection in NHANES was rigorous. The limitations include the observational nature of the study. In summary, the dietary variables used in the ISRNM PEW syndrome criteria do not appear to reflect either protein wasting or energy wasting in the general or moderate CKD populations. A definition of PEW syndrome without the dietary variables in these populations has better face validity and might be a better reflection of protein and energy wasting. Disclosure All the authors declared no competing interests. Supplementary Material Figure S1 Associations of number of protein–energy wasting (PEW) syndrome categories (defined with body mass index < 18.5 kg/m2 as a PEW criterion) with lean body mass, fat mass measured by dual-energy x-ray absorptiometry scans, and mortality in multivariate linear regression models and Cox regression models in the entire cohort (N = 11,834). (a) Nondietary categories alone and lean body mass; (b) nondietary categories alone and fat mass; (c) nondietary categories alone and mortality.

at mass measured by dual-energy x-ray absorptiometry scans, and mortality in multivariate linear regression models and Cox regression models in the entire cohort (N = 11,834). (a) Nondietary categories alone and lean body mass; (b) nondietary categories alone and fat mass; (c) nondietary categories alone and mortality. Acknowledgment This work is supported by grants from the National Institute of Diabetes and Digestive and Kidney Diseases (R01 DK091437 and RO1-DK077298) and the University of Utah Study Design and Biostatistics Center (funded in part from the Public Health Services research grant numbers UL1-RR025764 and C06-RR11234 from the National Center for Research Resources). The funding sources had no role in the design and conduct of the study; the collection, management, analysis, and interpretation of the data; the preparation, review, or approval of the manuscript; and the decision to submit the manuscript for publication. The findings and conclusions in this article are those of the authors and not necessarily those of the Centers for Disease Control and Prevention. Figure S1. Associations of number of protein–energy wasting (PEW) syndrome categories (defined with body mass index < 18.5 kg/m2 as a PEW criterion) with lean body mass, fat mass measured by dual-energy x-ray absorptiometry scans, and mortality in multivariate linear regression models and Cox regression models in the entire cohort (N = 11,834). (a) Nondietary categories alone and lean body mass; (b) nondietary categories alone and fat mass; (c) nondietary categories alone and mortality.

at mass measured by dual-energy x-ray absorptiometry scans, and mortality in multivariate linear regression models and Cox regression models in the entire cohort (N = 11,834). (a) Nondietary categories alone and lean body mass; (b) nondietary categories alone and fat mass; (c) nondietary categories alone and mortality. Supplementary material is linked to the online version of the paper at http://www.kireports.org.

Current literature documents substantial disparities in access to kidney transplantation waitlisting, including variation in waitlisting across the United States.1, 2, 3, 4, 5, 6 To address some of the disparities in transplantation access, the United Network for Organ Sharing (UNOS) implemented a new kidney allocation system (KAS) in December 2014 that changed how kidneys are allocated to potential recipients across the United States.7 Under the previous KAS, the most important determinant of receiving a new organ was time spent on the waiting list, with the clock starting when the transplantation center placed the patient on the waitlist, rather than when the patient started dialysis treatment. Under the new allocation system, the waiting time reverts back to the time of dialysis treatment initiation for all dialysis patients. Because African Americans on average spend a longer time on dialysis before referral for transplantation evaluation compared with white patients,8 this is 1 major aspect of the policy that is expected to reduce racial disparities in access to multiple transplantation steps. However, nephrologists and other dialysis staff may not be aware that patients with a longer time on dialysis who are not yet on the waitlist may receive a kidney transplantation more quickly under the new KAS.9 Because most patients with end-stage renal disease (ESRD) in the United States are initially treated at a dialysis facility,9 these facilities play a key role in educating patients and referring them to a transplantation center to undergo a transplantation evaluation.

ey transplantation more quickly under the new KAS.9 Because most patients with end-stage renal disease (ESRD) in the United States are initially treated at a dialysis facility,9 these facilities play a key role in educating patients and referring them to a transplantation center to undergo a transplantation evaluation. Previous research suggested that multicomponent dialysis facility−based interventions conducted with the support of government agencies, such as Centers for Medicare & Medicaid and/or ESRD Networks, might be effective in improving dialysis access,10 which would increase vascular access11 and increase referral for transplantation. Audit and feedback reports, otherwise known as performance feedback reports,12 were used in poorly performing dialysis facilities. Furthermore, research showed that when clinical interventions had a substantial evidence base, and there was need for expediency in ensuring the intervention was rapidly translated from research into practice, an effectiveness-implementation hybrid study design might be particularly useful to increase the usefulness and policy relevance of clinical research.13 Such a hybrid model allows for evaluating the effectiveness of a multicomponent intervention in a real-life setting while also assessing the implementation and potential sustainability of the intervention.

study design might be particularly useful to increase the usefulness and policy relevance of clinical research.13 Such a hybrid model allows for evaluating the effectiveness of a multicomponent intervention in a real-life setting while also assessing the implementation and potential sustainability of the intervention. In our planned ASCENT (Allocation System for Changes in Equity in Kidney Transplantation) study, we will test the effectiveness of educating dialysis physicians, staff, and patients on this recent KAS policy change on waitlisting using this effectiveness-implementation study framework to more quickly implement the intervention into practice if it is deemed effective. We will create a multicomponent intervention consisting of a webinar for dialysis facility medical directors, an educational video for patients, an educational video for dialysis facility staff, and a dialysis facility−specific transplantation performance feedback report for medical directors detailing the transplantation performance of the facility and communicating key relevant aspects of the new KAS in context with the data of the facility. An estimated 600 dialysis facilities across the United States with low kidney transplantation waitlisting in all 18 ESRD networks will be randomized to receive either the multicomponent intervention (intervention) or a UNOS brochure describing the recent KAS change (control). We will use a randomized effectiveness-implementation study design to test the effectiveness of the multicomponent intervention among dialysis facilities with low waitlisting, with a goal of increasing access to the deceased donor kidney waitlist and reducing racial disparities in waitlisting.

ent KAS change (control). We will use a randomized effectiveness-implementation study design to test the effectiveness of the multicomponent intervention among dialysis facilities with low waitlisting, with a goal of increasing access to the deceased donor kidney waitlist and reducing racial disparities in waitlisting. Study Design and Methods Study Overview A dissemination advisory board (DAB), including relevant stakeholders within the kidney health care system, will be convened to develop, finalize, and disseminate intervention materials among an estimated national sample of ∼600 dialysis facilities with low waitlisting. Co-primary outcomes will include (i) change in proportion of patients waitlisted, and (ii) disparity reduction in proportion of patients waitlisted in a dialysis facility after 1 year. Secondary outcomes include changes from baseline to 3 months in medical director knowledge about transplantation and KAS, as well as the intent of the medical director to refer patients for transplantation.

aitlisted, and (ii) disparity reduction in proportion of patients waitlisted in a dialysis facility after 1 year. Secondary outcomes include changes from baseline to 3 months in medical director knowledge about transplantation and KAS, as well as the intent of the medical director to refer patients for transplantation. Eligibility Criteria and Description of Potential Study Population All 18 ESRD networks will be contacted and invited to participate in this study. To encourage participation of ESRD networks across the nation, we will develop annual transplantation performance reports with tailored feedback detailing the performance of each participating network in waitlisting and transplantation compared with other ESRD networks across the United States, as well as some of the key features that will be included in the dialysis facility−specific reports. These network-level feedback reports will be shared only with ESRD network staff, rather than the dialysis facilities within their respective network.

antation compared with other ESRD networks across the United States, as well as some of the key features that will be included in the dialysis facility−specific reports. These network-level feedback reports will be shared only with ESRD network staff, rather than the dialysis facilities within their respective network. Facilities with low waitlisting, which have at least 11 patients overall and at least 4 African American patients, will be eligible for participation, because measured outcomes focus on disparity reduction and facilities with small proportions of African Americans; a small number of patients may be difficult to classify as a facility with a disparity. Low waitlisting will be defined as the lowest national tertile for 2014 (most recent data available) at randomization. Of the 1529 dialysis facilities meeting eligibility criteria across the United States, we estimate ∼40% of those invited (600 facilities) will agree to participate (Figure 1).Figure 1 Selection criteria for dialysis facilities eligible to participate among 18 end-stage renal disease networks invited to ASCENT Study.

n. Of the 1529 dialysis facilities meeting eligibility criteria across the United States, we estimate ∼40% of those invited (600 facilities) will agree to participate (Figure 1).Figure 1 Selection criteria for dialysis facilities eligible to participate among 18 end-stage renal disease networks invited to ASCENT Study. Study Procedures Dissemination Advisory Board The DAB of partnering stakeholders will be created among study co-investigators and national partners, including the National Kidney Foundation and the American Association of Kidney Patients, dialysis facility medical directors, nephrologists, social workers, ESRD patients, researchers, key policy partners, including ESRD Network 6 leadership and staff, UNOS, and regional members of the Southeastern Kidney Transplant Coalition (an academic−community collaboration among partners in North Carolina, South Carolina, and Georgia committed to eliminating health disparities in kidney transplantation). Stakeholder feedback regarding patient barriers to kidney transplantation and development of educational materials will ensure that these intervention materials are appropriate for dialysis facilities to understand the recently changed KAS and to help communicate information to facility staff and ESRD patients to encourage improved access to kidney transplantation. The volunteer DAB will meet via conference phone calls monthly for ∼6 months to develop the multicomponent intervention and finalize surveys. After materials are created, the DAB will review materials and provide feedback for improvement. Detailed information about the intervention material and the role of the DAB in developing these is described below.

ference phone calls monthly for ∼6 months to develop the multicomponent intervention and finalize surveys. After materials are created, the DAB will review materials and provide feedback for improvement. Detailed information about the intervention material and the role of the DAB in developing these is described below. Intervention Materials Transplantation Performance Feedback Reports The transplantation performance feedback report will reflect the performance of a dialysis facility with respect to kidney transplantation waitlisting and racial disparities in waitlisting, and will be provided to facility medical directors. The report will note information about the recent changes in KAS that are most relevant for the dialysis facility and will display facility-specific transplantation access performance measures, such as facility-specific waitlisting and racial disparity in waitlisting data, comparing the performance of the facility to the national average. An example potential feedback report is provided in Figure 2. The DAB will review several versions of the feedback report and discuss which layouts, content, and messages are best tailored to dialysis facilities with low waitlisting. Individualized reports will be emailed to intervention-assigned dialysis facility staff by their respective ESRD networks.Figure 2 Example of ASCENT feedback report for dialysis facility medical directors.

rt and discuss which layouts, content, and messages are best tailored to dialysis facilities with low waitlisting. Individualized reports will be emailed to intervention-assigned dialysis facility staff by their respective ESRD networks.Figure 2 Example of ASCENT feedback report for dialysis facility medical directors. Educational Video for ESRD Patients An ∼10-minute educational video will be produced for dialysis patients, highlighting the benefits of kidney transplantation, disputing common misconceptions about transplantation, and motivating patients through real patient stories on overcoming barriers to transplantation. The video is intended to educate and encourage patients to talk to their providers about being referred for kidney transplantation. The DAB will help recruit patients for this video and provide input on the video script, length, content, and format, as well as feedback on future revisions of the video.

transplantation. The video is intended to educate and encourage patients to talk to their providers about being referred for kidney transplantation. The DAB will help recruit patients for this video and provide input on the video script, length, content, and format, as well as feedback on future revisions of the video. Educational Video for Dialysis Facility Staff An ∼10-minute educational video targeted to dialysis facility staff (nephrologists, nurses, and social workers) will be created that describes racial disparities in transplantation, recent changes in the KAS and its effort to reduce disparities, and the important role of dialysis staff in educating patients about kidney transplantation and being involved with patients throughout the entire transplantation process. The video will feature clinical staff, such as a social worker, nurse, and nephrologist, as well as patient testimonials to emphasize the great impact that proactive dialysis staff have on their patients’ transplantation journeys. The DAB will help create the video script content, select graphics, provide feedback on video length, and review the video to provide feedback for future edits.

and nephrologist, as well as patient testimonials to emphasize the great impact that proactive dialysis staff have on their patients’ transplantation journeys. The DAB will help create the video script content, select graphics, provide feedback on video length, and review the video to provide feedback for future edits. Educational Webinar for Medical Directors and Facility Staff The DAB will work with a UNOS physician representative to create and present an ∼30-minute webinar targeted to dialysis facility medical directors, physicians, and other staff involved in transplantation education at the dialysis units. The webinar will discuss benefits of kidney transplantation, recent changes in KAS, implications of KAS on reducing racial disparities in waitlisting, and how dialysis facility staff can assist patients throughout the transplantation process. The webinar will be presented live with a question-and-answer session, and will also be recorded for those who cannot attend the live session. It will be hosted on the study website for ASCENT intervention facilities to access. Attendees who view the webinar will have an opportunity to receive continuing medical education credit. Many members of the DAB have experience with developing educational webinars and will ensure content is appropriate for dialysis facility medical directors and staff.

ite for ASCENT intervention facilities to access. Attendees who view the webinar will have an opportunity to receive continuing medical education credit. Many members of the DAB have experience with developing educational webinars and will ensure content is appropriate for dialysis facility medical directors and staff. Formative Evaluation To study the implementation of the intervention, we will conduct in-person and online formative testing of intervention materials in 3 geographically diverse dialysis facilities to ensure that these materials are appropriate for their target populations (dialysis facility medical directors, staff, and patients). Medical directors will review and provide feedback on the following: (i) the transplantation performance feedback report; (ii) the webinar; and (iii) a baseline survey (Supplementary Appendix S1) for medical directors for use in the clinical effectiveness study. A structured interview will be conducted to receive feedback on these materials and assess whether there are any missing educational domains from the transplantation performance feedback report or webinar, and if the survey contains items relevant to medical directors and other clinicians involved in transplantation education within the dialysis facility. During formative testing, we will also discuss with dialysis facility medical directors how long they believe it will take to educate staff and patients about the KAS to ensure that we select an appropriate time for follow-up to measure outcomes, using 3 months as an estimate based on previous conversations with members of the DAB.

ive testing, we will also discuss with dialysis facility medical directors how long they believe it will take to educate staff and patients about the KAS to ensure that we select an appropriate time for follow-up to measure outcomes, using 3 months as an estimate based on previous conversations with members of the DAB. For formative testing of the educational patient and staff videos, research staff will conduct structured interviews either in person or will administer surveys via email using a Health Insurance Portability and Accountability Act−compliant SurveyMonkey (San Mateo, CA) link. Medical directors will identify staff who will be asked to view the ∼10-minute staff educational video in person (either on an iPad [Apple, Cupertino, CA] or in a lunch-and-learn setting) or via the ASCENT website video link, followed by a structured, in-person interview or SurveyMonkey survey, depending on study site, to assess overall content and style, as well as any missing educational pieces or points of concern. Dialysis patients will be identified by dialysis facility medical directors or staff and will be asked to watch the patient education video on an iPad or computer during their regularly scheduled dialysis appointment. After viewing videos, structured interviews will be conducted to assess patients’ satisfaction and understanding of the video, the impact of the video on patient intent to discuss transplantation with providers, and other ways to improve the video.

iPad or computer during their regularly scheduled dialysis appointment. After viewing videos, structured interviews will be conducted to assess patients’ satisfaction and understanding of the video, the impact of the video on patient intent to discuss transplantation with providers, and other ways to improve the video. Randomized Effectiveness-Implementation Study We will test the effectiveness and implementation of the intervention materials14 among approximately one-half of the estimated 600 randomized dialysis facilities in US ESRD networks to examine whether this intervention improves dialysis facility waitlisting and reduces racial disparity in waitlisting. Because there may be significant heterogeneity in dialysis facilities and patient and staff populations across the participating ESRD networks, we will randomize dialysis facilities that were not included in formative testing within each ESRD network region 1:1 to either the multicomponent intervention (transplantation performance feedback report, webinar, and educational videos) or control group (UNOS educational brochure) (Supplementary Appendix S2). At baseline, all eligible dialysis facility medical directors in both the intervention and control groups will receive an email from their ESRD network with a link to a web-based survey (Health Insurance Portability and Accountability Act−compliant SurveyMonkey) with informed consent as the first page. We will randomize facilities to either the control or intervention group, and in cases in which 1 dialysis facility medical director or nurse manager oversees multiple facilities that are included in the study, we will assign these facilities to the same study group to avoid cross contamination. Within 1 week of completing the baseline survey, all facility medical directors and/or nurse managers from participating facilities will be emailed and mailed materials associated with their study group assignment, and instructed to share with staff. Intervention dialysis facilities will receive an email containing all intervention materials (transplantation performance feedback report, link to webinar, patient educational video, and staff educational video), and will also be mailed hard copies of the educational videos in DVD format and performance feedback reports. Control facilities will receive the UNOS pamphlet by e-mail and mail.

tervention materials (transplantation performance feedback report, link to webinar, patient educational video, and staff educational video), and will also be mailed hard copies of the educational videos in DVD format and performance feedback reports. Control facilities will receive the UNOS pamphlet by e-mail and mail. After ∼3 months following the baseline survey, all participating facility medical directors and/or nurse managers will be emailed follow-up surveys by their respective ESRD network contacts to assess secondary outcomes. Staff will be offered the option of a $10 gift card as incentive for participation for each survey. Surveys Dialysis Facility Medical Director Baseline Survey The medical director will answer items regarding their kidney transplantation knowledge and knowledge of KAS, staff training and patient education activities, and intent to refer patients for kidney transplantation evaluation (Table 1; Appendix S1).Table 1 Description of baseline dialysis facility medical director survey for ASCENT Study

r will answer items regarding their kidney transplantation knowledge and knowledge of KAS, staff training and patient education activities, and intent to refer patients for kidney transplantation evaluation (Table 1; Appendix S1).Table 1 Description of baseline dialysis facility medical director survey for ASCENT Study Scales Description Number of Questions Dialysis facility characteristics Assess dialysis facility characteristics, such as size, number of patients and staff, and amenities for patients 14 Perceived staff knowledge and KAS training Assess staff knowledge of transplant education and training provided, including proportion of staff trained on KAS and delivery of training 4 Perceived patient knowledge, transplantation education, and barriers to transplantation Assess patient knowledge of transplantation, education provided, including proportion of patients educated about transplantation delivery of education, and patient barriers 4 Medical director knowledge of transplantation, KAS, and racial disparity in transplantation Assess medical director knowledge of transplantation; knowledge of KAS; and awareness about racial disparities and waitlisting performance at their own facility and nationally 9 Medical director referral practices Assess medical director’s perceived referral practices (demographics of patients referred by race and time on dialysis, and estimates of proportion of patients eligible for, interested in, referred for, and waitlisted for kidney transplantation. 10 KAS, Kidney Allocation System.

ly 9 Medical director referral practices Assess medical director’s perceived referral practices (demographics of patients referred by race and time on dialysis, and estimates of proportion of patients eligible for, interested in, referred for, and waitlisted for kidney transplantation. 10 KAS, Kidney Allocation System. Dialysis Facility Medical Director Follow-up Survey Approximately 3 months after receiving educational materials, medical directors of both intervention and control facilities will receive a follow-up survey with similar questions to the baseline survey to assess knowledge about kidney transplantation and KAS, staff training on the allocation policy, patient education of transplantation, intent to refer patients for kidney transplantation, and uptake of intervention and control materials. Intervention and control facilities will also be asked several questions related to implementation (e.g., whether they used each intervention material) corresponding to their study group. The time frame of 3 months for a follow-up of secondary outcomes will be finalized by DAB members and medical directors during formative testing.

and control facilities will also be asked several questions related to implementation (e.g., whether they used each intervention material) corresponding to their study group. The time frame of 3 months for a follow-up of secondary outcomes will be finalized by DAB members and medical directors during formative testing. Co-Primary Outcome Measures and Statistical Analyses Change in Waitlisting and Waitlisting Disparity We will calculate change in the proportion of patients waitlisted at facilities at 1 year preintervention and 1 year postintervention to determine if intervention facilities had higher waitlisting poststudy compared with control facilities. We will calculate facility racial disparity in waitlisting 1 year preintervention and 1 year postintervention as the difference between the proportions of African American patients versus white patients who were waitlisted within a facility. We chose the period of 1 year for 2 major reasons. First, national surveillance data on waitlisting is only available on an annual basis. Second, we expect the impact of the intervention to be strongest within a timeframe closest to the delivery of the intervention (i.e., within a year of the intervention). To determine if there is a difference in either of these 2 co-primary outcomes among the intervention versus control facilities, we will use generalized linear models15 to account for potential correlation of facilities within networks and 2 sample t-tests.

Co-Primary Outcome Measures and Statistical Analyses Change in Waitlisting and Waitlisting Disparity We will calculate change in the proportion of patients waitlisted at facilities at 1 year preintervention and 1 year postintervention to determine if intervention facilities had higher waitlisting poststudy compared with control facilities. We will calculate facility racial disparity in waitlisting 1 year preintervention and 1 year postintervention as the difference between the proportions of African American patients versus white patients who were waitlisted within a facility. We chose the period of 1 year for 2 major reasons. First, national surveillance data on waitlisting is only available on an annual basis. Second, we expect the impact of the intervention to be strongest within a timeframe closest to the delivery of the intervention (i.e., within a year of the intervention). To determine if there is a difference in either of these 2 co-primary outcomes among the intervention versus control facilities, we will use generalized linear models15 to account for potential correlation of facilities within networks and 2 sample t-tests. Secondary Outcome Measures and Statistical Analyses Change in Knowledge About Kidney Transplantation and KAS At baseline and 3 months, we will assess change in transplantation and KAS knowledge among medical directors to determine the degree of knowledge improvement pre- versus poststudy. Items will include general transplantation knowledge, knowledge of KAS, and knowledge about racial disparities and waitlisting performance at their own facility and nationally (Table 1). The knowledge items will be summed, and each dialysis provider will receive a score between 0 and 9. We will calculate average change in knowledge from pre- to postintervention by study group, using t-tests to determine if medical directors from intervention facilities were more likely to improve in knowledge compared with providers from control facilities after receiving the intervention.

ive a score between 0 and 9. We will calculate average change in knowledge from pre- to postintervention by study group, using t-tests to determine if medical directors from intervention facilities were more likely to improve in knowledge compared with providers from control facilities after receiving the intervention. Change in Staff Training About Kidney Transplantation and KAS We will assess at baseline and at 3 months what percentage of staff medical directors have been trained about kidney transplantation and KAS, as well as how the training was delivered (e.g., did they hold a training session, send an email, watch video presentations, and so on). We will evaluate changes in how knowledgeable medical providers perceived their staff were (on a scale from 1 [not at all] to 5 [extremely]) on KAS pre- to postintervention. We will conduct paired t-tests to determine if differences in the proportion of correct items were greater for intervention facilities versus control facilities. Change in Patient Education About Kidney Transplantation We will ask providers at baseline and at 3-month follow-up whether they educated patients on kidney transplantation and how this information was delivered. We will also track visits to the educational video website to determine intervention dose and usage statistics. We will conduct similar analyses to determine if there was a change in the proportion of patients educated about KAS.

r they educated patients on kidney transplantation and how this information was delivered. We will also track visits to the educational video website to determine intervention dose and usage statistics. We will conduct similar analyses to determine if there was a change in the proportion of patients educated about KAS. Change in Intent to Refer Patients to Kidney Transplantation We will assess current referral practices of facilities by surveying the facility medical director about the estimated proportion of patients interested, eligible, and referred for transplantation in their facility at baseline and at 3 months postintervention. We will also ask questions about the estimated percentage of patients referred for transplantation by race/ethnicity and time on dialysis. We will conduct paired t-tests to determine if differences in the proportion of referred patients was greater for intervention facilities versus control facilities. Other Covariates To explore potential modifiers of the effectiveness of this system-level intervention, we will examine facility characteristics (region, facility size, profit status, and so on), characteristics of patients in facilities (e.g., race, insurance status, comorbid conditions, and so on), and contextual neighborhood characteristics such as poverty, education, or income level. We will include process measures for the intervention (receipt of intervention and self-report) to evaluate the potential for future dissemination of interventions to other US dialysis facilities.

, comorbid conditions, and so on), and contextual neighborhood characteristics such as poverty, education, or income level. We will include process measures for the intervention (receipt of intervention and self-report) to evaluate the potential for future dissemination of interventions to other US dialysis facilities. Implementation Effect Measures We will use an adaptation of the RE-AIM (Reach, Effectiveness, Adoption, Implementation, and Maintenance) framework16 for evaluating the public health impact of this health policy change.14 This framework builds upon the conceptual models of Rogers17 and Green and Krueter18 in this hybrid effectiveness-implementation study. Adoption will be assessed by participation and use of any intervention materials. Implementation will be assessed by calculating a composite measure, or “crude implementation index” for each facility as the sum of each secondary outcome (dichotomized at the median) of receipt and/or use of the feedback report and conduct of staff and patient education. We will explore barriers and facilitators to the use of the reports and education. We will conduct qualitative analyses of select medical directors that were successful intervention implementers (n = 3) and non-implementers (n = 3) at 1 year via phone interviews and online surveys with medical directors from implementers and non-implementers to assess RE-AIM measures.

se of the reports and education. We will conduct qualitative analyses of select medical directors that were successful intervention implementers (n = 3) and non-implementers (n = 3) at 1 year via phone interviews and online surveys with medical directors from implementers and non-implementers to assess RE-AIM measures. Sample Size and Power Based on 2014 data, if all 18 ESRD networks participate in the ASCENT study, a total of 1529 dialysis facilities will be potentially eligible for participation, of which 368 have a waitlisting racial disparity (Figure 1). For the primary outcome of overall waitlisting proportion, if an estimated 600 facilities (300 facilities in each study group, with an average of 70 patients per facility) respond, we will be adequately powered (80% at α = 0.05) to detect a small difference of 1.9% in the intervention group versus the control group based on a common waitlisting proportion of 10% at baseline (i.e., a waitlisting difference of 10% in the control group and 11.9% in the intervention group). A 2-sided Z-test (pooled) statistic and an intraclass correlation coefficient of 0.06 will be used.

nce of 1.9% in the intervention group versus the control group based on a common waitlisting proportion of 10% at baseline (i.e., a waitlisting difference of 10% in the control group and 11.9% in the intervention group). A 2-sided Z-test (pooled) statistic and an intraclass correlation coefficient of 0.06 will be used. For our other outcome of waitlisting disparity reduction among facilities with a racial disparity at baseline, our sample of 300 in each control and intervention group (total N = 600), will achieve 80% power (at α = 0.05) to detect a minimum difference of 11% in the waitlisting disparity proportion (percentage of facilities with African American racial disparity) between the intervention and the control groups after 1 year (i.e., a disparity proportion of 21.4% in the intervention group vs. 24.0% in the control group). This calculation assumes a common baseline disparity proportion of 0.24 (24% of facilities have a disparity) at baseline and an intraclass correlation coefficient of 0.06 among patients in a facility. We will use the 2-sided Likelihood Score Test (Farrington & Manning)19; the significance level of the test is 0.05.

p). This calculation assumes a common baseline disparity proportion of 0.24 (24% of facilities have a disparity) at baseline and an intraclass correlation coefficient of 0.06 among patients in a facility. We will use the 2-sided Likelihood Score Test (Farrington & Manning)19; the significance level of the test is 0.05. Discussion Previous research documented substantial decreased access to kidney transplantation waitlisting and racial disparities in access to kidney transplantation.6 A major policy change in the national kidney transplantation allocation system in December 2014 aimed in part at reducing racial disparities among patients waitlisted for transplantation.7 Preliminary results suggested that racial disparities might have been reduced in transplantation rates following the implementation of KAS.20 However, due to substantial disparities that existed before waitlisting,2, 4, 21 there were more dialysis patients who could potentially benefit from the changes in KAS by increased access to the deceased donor kidney waitlist.

ties might have been reduced in transplantation rates following the implementation of KAS.20 However, due to substantial disparities that existed before waitlisting,2, 4, 21 there were more dialysis patients who could potentially benefit from the changes in KAS by increased access to the deceased donor kidney waitlist. Previous ESRD network−led quality improvement interventions were successful in helping to improve ESRD patient outcomes, including increasing influenza and pneumococcal vaccination rates,22 fistula placement through the Fistula First initiative,11 and kidney transplantation referrals.23 Although the support of ESRD networks is a strength for this study, there are several potential limitations of the study design. Network leadership will send both the baseline and follow-up surveys to medical directors to help with study recruitment and data collection, but it is possible that some medical directors will have lower than expected response rates due to differential network responsiveness and because the project is not mandatory, unlike other previous dialysis-facility based projects we have conducted with success.10, 22, 24 To address this issue, the ASCENT research staff will follow up with dialysis facilities that are unresponsive after the initial and reminder emails from their network with additional emails and phone calls to achieve maximum participation. It is also possible that medical directors may forward surveys to nurse managers. We will capture role/title within the survey to address this possibility. In addition, because ESRD network leadership is sending surveys to medical directors, a positive response bias in which facilities report that they improve but may not actually change practice may occur. To minimize this bias, we will ensure that medical directors know that facility-identifiable data are blinded to networks.

ition, because ESRD network leadership is sending surveys to medical directors, a positive response bias in which facilities report that they improve but may not actually change practice may occur. To minimize this bias, we will ensure that medical directors know that facility-identifiable data are blinded to networks. An additional limitation could be difficulty to accurately measure uptake of the intervention because of the large-scale nature of the study. For example, dialysis facility staff may report sharing patient videos with patients, but we have no way to track whether patients watched the video and/or were educated by clinicians about transplantation other than through the medical director survey. However, this study is designed to be an effectiveness-implementation study, with the goal of real-world pragmatic implementation rather than measuring efficacy of the intervention in a controlled setting in which all participants are confirmed to have received the intervention. A strength of this approach is that we will have an estimate of the effectiveness of this intervention approach in the real world, which will provide insight into whether the intervention should be disseminated to all U.S. dialysis facilities through the support of their ESRD networks. An additional potential pitfall of our study is the possibility that knowledge about KAS may increase among medical directors, but that this will not translate into changes in referral and waitlisting for the patient population. Using our process and evaluation measures, we hope to be able to hypothesize reasons for any limits to the success of the intervention.

is the possibility that knowledge about KAS may increase among medical directors, but that this will not translate into changes in referral and waitlisting for the patient population. Using our process and evaluation measures, we hope to be able to hypothesize reasons for any limits to the success of the intervention. Despite these limitations, we consider delivery of information about transplantation and the new KAS as a first step toward increasing waitlisting overall and reducing disparities in access to transplantation in the United States. In addition, we will gain essential information from our analyses and implementation measures from surveys and interviews to inform future implementation of the intervention materials to other dialysis facilities across the country. For example, some components of the intervention, such as the patient and staff videos and the medical director webinar, will be made publicly available on a website after study end. If the intervention is effective in improving waitlisting or reducing disparity in waitlisting, ESRD networks could implement the intervention among control dialysis facilities and/or other dialysis facilities not selected for participation in the study.

ar, will be made publicly available on a website after study end. If the intervention is effective in improving waitlisting or reducing disparity in waitlisting, ESRD networks could implement the intervention among control dialysis facilities and/or other dialysis facilities not selected for participation in the study. In conclusion, if effective, the ASCENT study interventions could help extend the reach of a national kidney allocation policy by educating dialysis facility medical directors, staff, and patients about transplantation about the new KAS and thereby increasing the potential impact of KAS on disparity reduction. Conducting this research among dialysis facilities with low waitlisting across the U.S. could help to ensure equitability by reducing racial disparities in, and increasing access to, kidney transplant waitlisting. Disclosure SOP is a minority shareholder in Fresenius Dialysis, College Park, Georgia. All the other authors declared no competing interests.

In conclusion, if effective, the ASCENT study interventions could help extend the reach of a national kidney allocation policy by educating dialysis facility medical directors, staff, and patients about transplantation about the new KAS and thereby increasing the potential impact of KAS on disparity reduction. Conducting this research among dialysis facilities with low waitlisting across the U.S. could help to ensure equitability by reducing racial disparities in, and increasing access to, kidney transplant waitlisting. Disclosure SOP is a minority shareholder in Fresenius Dialysis, College Park, Georgia. All the other authors declared no competing interests. Author Contributions REP participated in study design, intervention development, planning for data collection, helped draft the manuscript, and is the principal investigator. KDS participated in intervention development, planning for data collection, and helped draft the manuscript. MB participated in intervention development, planning for data collection, study design, and helped draft the manuscript. JG and LP participated in intervention development, study design, and planning for data collection. SM, CE, SK, and SP participated in study design and intervention development. TM, GG, AB, GR, TB, and NT participated in intervention development. SC participated in study design. Supplementary Material Appendix S1 ASCENT Baseline Survey for Medical Directors. Appendix S2 United Network for Organ Sharing (UNOS) Brochure for Control Facilities.

Author Contributions REP participated in study design, intervention development, planning for data collection, helped draft the manuscript, and is the principal investigator. KDS participated in intervention development, planning for data collection, and helped draft the manuscript. MB participated in intervention development, planning for data collection, study design, and helped draft the manuscript. JG and LP participated in intervention development, study design, and planning for data collection. SM, CE, SK, and SP participated in study design and intervention development. TM, GG, AB, GR, TB, and NT participated in intervention development. SC participated in study design. Supplementary Material Appendix S1 ASCENT Baseline Survey for Medical Directors. Appendix S2 United Network for Organ Sharing (UNOS) Brochure for Control Facilities. Acknowledgments We would like to thank the National Institute on Minority Health and Health Disparities for funding this project (grant no. R01MD010290). Trial Registration: ClinicalTrials.gov number: NCT02879812. Appendix S1. ASCENT Baseline Survey for Medical Directors. Appendix S2. United Network for Organ Sharing (UNOS) Brochure for Control Facilities. Supplementary material is linked to the online version of the paper at www.kireports.org.

There exists a global pandemic in diabetes mellitus.1 Diabetic nephropathy (DN) develops in as many as 25% to 40% of diabetic patients after 25 years of uncontrolled and even treated diabetes. This makes DN the leading reason for the development of late-stage renal disease especially in the Western world2 where it accounts for approximately 22% of patients starting dialysis in Denmark3 and 44% in the USA.4 DN is characterized clinically by the increased blood pressure, occurrence of albuminuria, and a continual decrease of kidney function,2 and is associated with a remarkable increase in cardiovascular diseases5 and mortality.6 The annual mortality rate for patients with DN-induced renal disease is around 20%.7 During the last decade, there has been an 11-fold increase in spending for the treatment of patients with diabetes mellitus with concomitant DN and chronic kidney disease in the USA.8

ncrease in cardiovascular diseases5 and mortality.6 The annual mortality rate for patients with DN-induced renal disease is around 20%.7 During the last decade, there has been an 11-fold increase in spending for the treatment of patients with diabetes mellitus with concomitant DN and chronic kidney disease in the USA.8 The diffuse thickening of glomerular basement membrane together with nodular glomerulosclerosis is the major pathological feature of DN. During the initial stage of diabetic kidney disease, most patients only present modest proteinuria that deteriorates as the disease progresses. More advanced DN leads to primary and secondary pathological changes in the tubulointerstitial and vascular compartments, which is harmful to the maintenance of renal function. In a proportion of patients with clinical symptoms of DN, additional primary renal diseases (e.g., IgA nephropathy and renal arterial disease) may result from the diabetes mellitus-driven pathological abnormalities. Although the pathophysiology of DN is mainly due to hyperglycemia, it is also related to the wider network involving local and systemic processes,9, 10 some of which have been identified with cell and animal models, tissue samples, and human studies. For instance, in renal parenchymal cells, hyperglycemia induces abnormal activation of protein kinase C together with the increased expression of transforming growth factor β (TGFβ), matrix proteins fibronectin and collagen type IV, the dysregulation of nitric oxide, endothelial dysfunction and activation of the nuclear factor kappa B, and mitogen-activated protein kinase signaling.10, 11, 12 In addition, hyperglycemia can result in the overproduction of advanced glycation end products and induce overexpression of TGFβ.13 In spite of certain available treatments, the progression of DN continues to increase worldwide and thus novel therapeutic strategies are urgently needed.

otein kinase signaling.10, 11, 12 In addition, hyperglycemia can result in the overproduction of advanced glycation end products and induce overexpression of TGFβ.13 In spite of certain available treatments, the progression of DN continues to increase worldwide and thus novel therapeutic strategies are urgently needed. Nuclear receptors are transcription factors that play various roles in embryo development, maintenance of the differentiated cellular phenotype, and manipulation of cell metabolism and death. This review mainly discusses the association between the pathogenesis of DN and nuclear receptors, including peroxisome proliferator–activated receptors (PPARs) α (NR1C1), β/δ (NR1C2), and γ (NR1C3); farnesoid X receptor (FXR, NR1H4); liver X receptors (LXRs, NR1H2, NR1H3); vitamin D receptor (VDR, NR1I1); hepatocyte nuclear factor 4α (HNF4α, NR2A1); retinoid X receptors (RXR, NR1F1, NR1F2, NR1F3); retinoid acid receptors (NR1B1, NR1B2, NR1B3); estrogen receptor (ER, NR3A1); and mineralocorticoid receptor (MR, NR3C2). Several studies have suggested that activation or inhibition of specific receptors could prevent the progression of DN, which implies that targeting nuclear receptors may be a potential therapeutic strategy for DN.

ceptors (NR1B1, NR1B2, NR1B3); estrogen receptor (ER, NR3A1); and mineralocorticoid receptor (MR, NR3C2). Several studies have suggested that activation or inhibition of specific receptors could prevent the progression of DN, which implies that targeting nuclear receptors may be a potential therapeutic strategy for DN. Nuclear Receptors PPAR PPARs are ligand-activated transcriptional factors and include 3 related forms PPARα, PPARβ/δ, and PPARγ. Although they all have different tissue distributions, ligand selectivities, and biological effects, they play an important role in modulating lipid metabolism, adipogenesis, insulin sensitivity, inflammation, and blood pressure. Renal PPARα and PPARγ modulate energy utilization in the kidney by regulating fatty acid oxidation.14 Activated PPARα can stimulate fatty acid β-oxidation that can reduce the lipid content of tissues and blood, prevent the accumulation of lipid, and ameliorate lipotoxicity.15 Several kinases, including protein kinase A, protein kinase C, mitogen-activated protein kinases, and adenosine monophosphate kinase, were shown to phosphorylate PPARs resulting in changes in DNA-binding activity, ligand affinity, recruitment of transcriptional cofactors, and proteasome degradation in both a ligand-dependent or -independent manner.16 Phosphorylation by adenosine monophosphate kinase leads to increased PPARα and PPARγ signaling and enhances renal function in a type 2 diabetes mouse model by removing lipid accumulation in the kidney.15 Furthermore, the activation of PPARγ suppresses the renal expression of an α(1D)-adrenergic receptor that is overexpressed in the diabetic kidney.17

ophosphate kinase leads to increased PPARα and PPARγ signaling and enhances renal function in a type 2 diabetes mouse model by removing lipid accumulation in the kidney.15 Furthermore, the activation of PPARγ suppresses the renal expression of an α(1D)-adrenergic receptor that is overexpressed in the diabetic kidney.17 Chronic inflammation and oxidative stress play a pivotal role in the pathogenesis of chronic kidney disease. Activated PPARα can prevent overexpression of proinflammatory molecules.18 It was shown that the ligand activation of PPARα will increase the expression of fibroblast growth factor-21 (FGF-21), enhance the phosphatidylinositol-3 kinase/protein kinase B (AKT)/glycogen synthase kinase 3β (GSK-3β)/Fyn-mediated nuclear factor (erythroid-derived 2)-like 2 signal, and prevent the development of DN.19 PPARα activation improves lipotoxicity by activating adenosine monophosphate kinase-peroxisome proliferator-activated receptor g coactivator-1α (PGC-1α)-estrogen-related receptor-1α (ERR-1α)-forkhead box 03a (Fox03a) signaling and ameliorating glucose-induced matrix production and mesangial cell proliferation by inhibiting extracellular signal–regulated kinase 1/2 and phosphatidylinositol-3′-kinase/AKT activation, suggesting its potential for the treatment of DN.20, 21 In the absence of PPARα, the glomerular lesions displayed enhanced type IV collagen and TGFβ levels in DN, indicating that PPARα agonists can prevent glomerular matrix expansion together with apoptosis and the infiltration of inflammatory cells within the glomerulus.22 A recent study found that Huangkui capsule, an extract from Abelmoschus manihot (L.) medic, can ameliorate DN by increasing PPARα/PPARγ signaling leading to lowered endoplasmic reticulum (ER) stress in rats.23 It was reported that fenofibrate, a PPARα agonist, can dramatically decrease the excretion of urinary albumin and reduce mesangial matrix expansion and glomerular hypertrophy in the db/db diabetic mice model.24 Fenofibrate also improved insulin resistance and glomerular lesions in db/db mice,24 thus suggesting a renal protective role for fenofibrate in DN via the activation of PPARα in mesangial cells. A Fenofibrate Intervention and Event Lowering in Diabetes study further suggested that the early use of fenofibrate may prevent or postpone the development of DN.25

olecule-1, which is increased in a high glucose environment in mesangial cells.142 Blockage of HNF4α in mesangial cells might be a candidate therapeutic strategy for DN, as the stromal interacting molecule-1-gated store-operated Ca(2+) entry pathway in mesangial cells was recently found to be antifibrotic.142, 143, 144 HNF1α is a homeodomain-containing transcription factor that plays an important role for modulating different metabolic functions in the liver, pancreatic islet, kidney, and intestine.145 Maturity-onset diabetes of the young-3 result from rare mutations in HNF1A.146, 147 Although genetic variants in HNF1β are not a major cause of maturity-onset diabetes of the young or DN, they might lead to the manifestation of disease in Chinese.148

tance and glomerular lesions in db/db mice,24 thus suggesting a renal protective role for fenofibrate in DN via the activation of PPARα in mesangial cells. A Fenofibrate Intervention and Event Lowering in Diabetes study further suggested that the early use of fenofibrate may prevent or postpone the development of DN.25 The protection provided by activated PPARγ is partially mediated by downregulating the level of renal disintegrin and metalloprotease-17 (ADAM17) and angiotensin-converting enzyme-2 (ACE2) shedding.26 Increased fibrosis in glomerular microenvironment is a remarkable characteristic of DN. Strong evidence suggests that PPARγ plays an important role during the pathogenesis of glomerulosclerosis. Treatment with PPARγ agonist ameliorated the hyperglycemia-mediated cannabinoid receptor type 1 (CB1R) signaling, inflammation, and glomerular fibrosis in diabetic animals.27, 28 PPARγ could prevent protein kinase A signaling, the activation of rat intraglomerular mesangial cells, TGFβ-induced accumulation of p-cyclic-AMP-responsive element binding protein and collagen-IV.29 PPARγ also negatively regulates inflammation through binding to the MIP3A promoter and downregulating the expression of macrophage inflammatory protein-3α (MIP-3α), a pathogenic mediator playing a crucial role in inflammation of DN.30 Other studies showed that PPARγ provides renoprotective action by negatively regulating the microsomal prostaglandin E synthase-1 (mPGES-1)/prostaglandin E2/prostaglandin E2 receptor 4 (EP4) pathway and restoring expression of the klotho axis in a PPARγ-dependent manner.31, 32 PPARγ may enhance the function of the angiotensin II receptor blocker by downregulating thioredoxin-interacting protein.33 PPARγ activated by pigment epithelium-derived factor could suppress the expression of the receptor for advanced glycation end products and decrease the reactive oxygen species (ROS), which subsequently prevents advanced glycation end product-induced apoptotic cell death in podocytes.34 Many studies were performed to separate the insulin sensitizing effects of PPARγ agonists from the transcriptional activation of genes that result in untoward side effects. This was achieved to some degree by using partial agonists that, compared with a full agonist, only partially activated the transcription of select genes.35

ies were performed to separate the insulin sensitizing effects of PPARγ agonists from the transcriptional activation of genes that result in untoward side effects. This was achieved to some degree by using partial agonists that, compared with a full agonist, only partially activated the transcription of select genes.35 Among patients with type 2 diabetes, the polymorphism within PPARγ2 (Pro12Ala) provides protection against nephropathy progression and deterioration of renal function, independent of major confounders.36 However, the PPARγ2 (Pro12Ala) polymorphism may not be associated with the progression of DN in patients with type 1 diabetes.37 A meta-analysis showed that the PPARγ (Pro/Pro) genotype presented close association with DN risk in Caucasians, but the Ala/Ala genotype and Ala allele did not.38 Conversely, another meta-analysis indicated that the polymorphism in PPARγ (Pro12Ala) gene has no relationship with DN risk in Asians.39 The rs1801282 C>G variant in PPARγ was closely associated with decreased DN risk.40 However, further studies revealed that the PPARγ2 Ala12 variant provided renal protection by reducing the occurrence of albuminuria among patients with type 2 diabetes.41, 42

(Pro12Ala) gene has no relationship with DN risk in Asians.39 The rs1801282 C>G variant in PPARγ was closely associated with decreased DN risk.40 However, further studies revealed that the PPARγ2 Ala12 variant provided renal protection by reducing the occurrence of albuminuria among patients with type 2 diabetes.41, 42 PPARβ/δ agonist treatment inhibited glomerular mesangial expansion, albuminuria, and the accumulation of type IV collagen with no effect on blood glucose levels in streptozotocin-treated diabetic mice.43 The activation of PPARβ/δ is necessary for treating DN by preventing inflammation and activating of its downstream receptor for advanced glycation end product or nuclear factor kappa B signals.43, 44 PPARβ/δ agonist could postpone diabetes-induced nephrin loss, enhance podocyte integrity, and prevent albuminuria subsequently.45 LXR LXRs were first identified as orphan receptors when discovered, and then subsequently found to be targets of oxysterol metabolites of cholesterol.46 LXRs include LXRα and LXRβ that have different tissue distribution patterns, but have been most extensively studied in the liver.

PPARβ/δ agonist treatment inhibited glomerular mesangial expansion, albuminuria, and the accumulation of type IV collagen with no effect on blood glucose levels in streptozotocin-treated diabetic mice.43 The activation of PPARβ/δ is necessary for treating DN by preventing inflammation and activating of its downstream receptor for advanced glycation end product or nuclear factor kappa B signals.43, 44 PPARβ/δ agonist could postpone diabetes-induced nephrin loss, enhance podocyte integrity, and prevent albuminuria subsequently.45 LXR LXRs were first identified as orphan receptors when discovered, and then subsequently found to be targets of oxysterol metabolites of cholesterol.46 LXRs include LXRα and LXRβ that have different tissue distribution patterns, but have been most extensively studied in the liver. LXRs might have a role in regulating lipid metabolism and maintaining the function of proximal tubule as well as podocytes by downregulating the expression of nephrin.47 The administration of the LXR agonist T0901317 could increase cholesterol efflux via activating the ATP-binding cassette transporter A1 (ABCA1) in cultured glomerular mesangial cells, and enhance the expression of stearoyl-CoA desaturase-1 through increasing the level of sterol regulatory element-binding protein 1c (SREBP-1c) within proximal tubules.48, 49 LXRα/SREBP-1 signaling also has the capability of regulating the expression of many genes involved in fatty acid and triglyceride synthesis.50 Nε-(carboxymethyl) lysine, a member of the advanced glycation end product family, modulates cholesterol metabolism through stimulating LXR and SREBP-2, which resulted in a reduction in ABCA1-mediated cholesterol efflux and the accumulation of lipid in human kidney-2 (HK-2) cells.51 Bilirubin improved dyslipidemia and renal disfunction via suppressing the expression of LXRα and SREBP-1 and decreasing ROS.52 Furthermore, the activation of LXR may prevent inflammation and the development of DN.46, 53 T0901317 could prevent the development of albuminuria, glomerular mesangial expansion, and interstitial fibrosis by decreasing osteopontin level, macrophage infiltration, and expression of inflammatory genes, such as monocyte chemoattractant protein-1 (MCP-1), tumor necrosis factor α (TNFα), and TGFβ, in the diabetic kidney.46 Knockdown of LXRα expression resulted in loss of the anti-inflammatory effect of anthocyanins, and further studies demonstrated that LXRα might participate in the anthocyanin-induced action of decreasing intercellular adhesion molecule 1, MCP1, and TGFβ1 via inhibiting the nuclear translocation of nuclear factor kappa B protein.54 Expression of LXRα in macrophage of transgenic mice markedly ameliorated hyperlipidemic-hyperglycemic nephropathy by suppressing glycated or acetylated low-density lipoprotein-induced cytokines and ROS in macrophages.55 Recently, accelerated mesangial matrix expansion and glomerular lipid accumulation were observed in Lxra/Lxrb-null diabetic mice, in coupling w

nic mice markedly ameliorated hyperlipidemic-hyperglycemic nephropathy by suppressing glycated or acetylated low-density lipoprotein-induced cytokines and ROS in macrophages.55 Recently, accelerated mesangial matrix expansion and glomerular lipid accumulation were observed in Lxra/Lxrb-null diabetic mice, in coupling w ith the enrichment of oxidative stress and inflammatory markers. Moreover, treatment with a synthetic oxysterol, N,N-dimethyl-3beta-hydroxycholenamide, an LXR agonist, dramatically ameliorated the excretion of albumin and nephrin, the levels of glomerular lipids and plasma triacylglycerol and cholesterol. In addition, the decreased level of kidney inflammatory and oxidative stress markers was observed upon N,N-dimethyl-3beta-hydroxycholenamide treatment.47 Together, these results indicate that the activity of LXR is necessary for both normal and diabetic kidney.

merular lipids and plasma triacylglycerol and cholesterol. In addition, the decreased level of kidney inflammatory and oxidative stress markers was observed upon N,N-dimethyl-3beta-hydroxycholenamide treatment.47 Together, these results indicate that the activity of LXR is necessary for both normal and diabetic kidney. FXR FXR was first thought to be an orphan receptor when discovered. However, further studies revealed that bile acid-induced activation of FXR is important for bile-acid synthesis and transport in the liver and intestine.56, 57, 58 Endogenous ligands for FXR include the primary bile acids, taurocholic acid, chenodeoxycholic acid, and cholic acid. FXR is expressed at highest levels in the liver and intestine, and at lower levels in adrenal gland and other tissues; it is also highly expressed in the kidney.59 It also plays a pivotal role in lipid, glucose, and bile acid homeostasis in the enterohepatic system.60, 61 Furthermore, FXR agonists may provide protection against liver fibrosis.62, 63 FXR agonists downregulate renal overexpression of SREBP-1 that could lead to lipid accumulation during the development of nephropathy and regulating renal lipid metabolism. The activation of FXR could prevent the induction of profibrotic growth factors, proinflammatory cytokines, and oxidative stress-related enzymes in the kidney, and thus improve glomerulosclerosis and proteinuria.64 Furthermore, the activation of FXR could suppress the development of nephropathy in type 1 diabetes via blocking diabetes-induced dysregulation of lipid metabolism, fibrosis, inflammation, and oxidative stress in the kidney.65 Recently, the adipocytokine visfatin was found to have a crucial role in the development of DN, at least partly, through enhancing high glucose-induced human mesangial cell inflammation, fibrosis, and proliferation in the absence of FXR.66

lipid metabolism, fibrosis, inflammation, and oxidative stress in the kidney.65 Recently, the adipocytokine visfatin was found to have a crucial role in the development of DN, at least partly, through enhancing high glucose-induced human mesangial cell inflammation, fibrosis, and proliferation in the absence of FXR.66 VDR Vitamin D is necessary for the metabolism of calcium and bone. It was reported that vitamin D deficiency was closely associated with increased risk for diabetes development, diabetes complications, and cardiovascular disease.67 A meta-analysis including 5 observational studies suggested that children treated with vitamin D are less likely to develop type 1 diabetes mellitus.68 The fact that the lack of vitamin D impairs insulin synthesis and secretion suggested its close association with the pathogenesis of type 2 diabetes, although the mechanistic link has not been well established.69

udies suggested that children treated with vitamin D are less likely to develop type 1 diabetes mellitus.68 The fact that the lack of vitamin D impairs insulin synthesis and secretion suggested its close association with the pathogenesis of type 2 diabetes, although the mechanistic link has not been well established.69 The protective activities of VDR may result from the inhibition of the renin-angiotensin system, reduction of proteinuria, and regulation of cell proliferation and differentiation. A recent study suggested that vitamin D and its receptor might modulate the progression of DN via regulating the TGFβ levels, the expression of angiotensinogen, and apoptosis of podocytes through the nuclear factor kappa B pathway.70 Activated macrophages 1 (M1) and activated macrophages 2 (M2) have opposing roles in inflammation. M1 activation was inhibited by 1,25-dihydroxyvitamin D3, a VDR agonist, while M2 was activated.71 Another study reported that vitamin D can switch the M1 phenotype to M2 via activating the VDR-PPARγ pathway.72 Diabetic Vdr null mice developed more severe nephropathy than wild-type mice as renin-angiotensin system activation was enhanced, suggesting that VDR protects the kidney from hyperglycemia-induced injury through inhibiting renin-angiotensin system activity.73 These data indicated that the combination of renin-angiotensin system inhibitors and a VDR activator might be of value to improve DN-induced albuminuria.74 A randomized clinical trial revealed that daily treatment of paricalcitol, a selective VDR agonist, could ameliorate residual albuminuria in ACE inhibitor (ACEI)- or angiotensin II type 1 receptor blockade (ARB)-treated DN patients, especially in those with high dietary sodium intake. These data suggested that the combination of paricalcitol and ACEI or ARBs could effectively reduce residual albuminuria, which may be applied as a new strategy in the treatment of DN.75 Wnt/β-catenin signal-related epithelial-mesenchymal transition was reportedly involved in the pathogenesis of DN.76 A recent study documented that VDR could decrease the expression of β-catenin by replacing β-catenin complexing with transcription factor 4 (TCF-4), therefore blocking Wnt/β-catenin signaling.77

of DN.75 Wnt/β-catenin signal-related epithelial-mesenchymal transition was reportedly involved in the pathogenesis of DN.76 A recent study documented that VDR could decrease the expression of β-catenin by replacing β-catenin complexing with transcription factor 4 (TCF-4), therefore blocking Wnt/β-catenin signaling.77 Podocyte injury is one of the causes of DN.78 VDR activation in podocytes plays an important role in preventing the kidney from diabetic damage.79 Calcitriol or a vitamin D analog can improve podocyte damage by inhibiting the expression of transient receptor potential cation channel. subfamily C. member 6 (TRPC6) during the early stage of DN in a rat model.78 1,25-D3 treatment ameliorated proteinuria in 25-hydroxy-1α-hydroxylase conventional knockout mice coupled with increasing heparanase expression, suggesting that vitamin D mediated the emergence of proteinuria by reducing heparanase levels in podocytes.80 Furthermore, vitamin D analogs provide protection against lesion of renal barrier by maintaining and reactivating the expression of podocalyxin, a specialized component of podocytes.81

ng heparanase expression, suggesting that vitamin D mediated the emergence of proteinuria by reducing heparanase levels in podocytes.80 Furthermore, vitamin D analogs provide protection against lesion of renal barrier by maintaining and reactivating the expression of podocalyxin, a specialized component of podocytes.81 The anti-inflammatory action of vitamin D is due to its influence on the crosstalk between signal transducer and activator of transcription 5 and VDR.82 A functional polymorphism of the VDR gene may result in individual susceptibility to DN, and a meta-analysis suggested the correlation of a Fok1 single-nucleotide polymorphism with DN susceptibility in Caucasians.83 Another study showed that a BsmI single-nucleotide polymorphism polymorphism in Han Chinese people was responsible for the type 2 diabetes-related albuminuria.84

ndividual susceptibility to DN, and a meta-analysis suggested the correlation of a Fok1 single-nucleotide polymorphism with DN susceptibility in Caucasians.83 Another study showed that a BsmI single-nucleotide polymorphism polymorphism in Han Chinese people was responsible for the type 2 diabetes-related albuminuria.84 MR MR regulates the reabsorption of sodium and water and secretion of potassium via control of the epithelial ion channel. The representative agonist and antagonist of MR are respectively aldosterone and spironolactone. However, mineralocorticoids could not only regulate the transport of epithelial salt, extracellular volume, and blood pressure, but also inflammation and fibrosis either directly or indirectly. Emerging evidence indicates that aldosterone participates in the pathogenesis of kidney disease in a nonepithelial MR-dependent manner.85 Some studies also reported that aldosterone impairs insulin sensitivity through MR activation in adipocytes in vitro, which indicates that aldosterone may play an important role in the development of diabetes.86, 87 Interestingly, leptin, which is upregulated in diabetic obese models, stimulates aldosterone production in vitro in human adrenocortical cells and in vivo in mice. In addition, aldosterone increases fibrosis by upregulating the production of TGFβ1, ROS, plasminogen activator inhibitor 1 (PAI-1), and the enrichment of collagen protein, which can be blocked by MR antagonist.88 Integrin β1 and β3 expression in podocytes is essential to the integrity of a glomerular structure.

mice. In addition, aldosterone increases fibrosis by upregulating the production of TGFβ1, ROS, plasminogen activator inhibitor 1 (PAI-1), and the enrichment of collagen protein, which can be blocked by MR antagonist.88 Integrin β1 and β3 expression in podocytes is essential to the integrity of a glomerular structure. In a high glucose environment, the expression of integrin β1 in cultured podocytes is markedly decreased, accompanied with an increase of integrin β3, and a recent study suggested that spironolactone inhibited cell motility and stabilized podoctyes cultured in a high glucose environment, in part by normalizing the level of integrin β1 and β3.89 Treatment with spironolactone provides protection for podocytes and inhibits the development of morphological changes associated with DN, probably by the inhibition of TGFβ1 mRNA expression.90 Spironolactone could inhibit MR-induced ROS production and hyperglycemia-mediated podocyte lesions in diabetics.91 Recent studies revealed a crucial role for aldosterone in the pathogenesis of DN, which has no effect on angiotensin II and blood pressure levels.92 Another study enrolling type 2 diabetic patients also demonstrated that patients who developed aldosterone escape, an increase in aldosterone levels during long-term treatment of ACEIs, suffered more severe albuminuria than did patients without aldosterone escape. However, in combination with spironolactone treatment a further decrease in albuminuria was noted in these patients.92 Furthermore, the incidence of severe hyperkalemia, which is the major side effect of spironolactone treatment in clinical trials, is low, probably resulting from the monitoring of dietary intake of potassium and diuretics in clinical observation. However, the liberalized usage of spironolactone is strictly forbidden for patients whose kidney function was reduced.92 It was suggested that alterations of Na/K ATPase levels might be a new pathophysiological feature for DN. The ability of aldosterone antagonists to decrease Na/K ATPase protein levels and enzyme mislocation that are increased in diabetes may suggest a new pharmaceutical use in the treatment of DN.93

tion was reduced.92 It was suggested that alterations of Na/K ATPase levels might be a new pathophysiological feature for DN. The ability of aldosterone antagonists to decrease Na/K ATPase protein levels and enzyme mislocation that are increased in diabetes may suggest a new pharmaceutical use in the treatment of DN.93 Other Nuclear Receptors The sex hormone estrogen has several functions including control of bone growth, modulation of differentiation and function of the reproductive tract, and memory storage.94, 95 Estrogen exerts its biological activity through the interaction with classic estrogen receptors, ERα and ERβ.96 It is generally known that females have a lower chance of suffering from nondiabetic chronic kidney disease than males.97, 98, 99, 100 Although the contribution of gender to the progression of type 1 or type 2 diabetic renal disease is still uncertain,100, 101 some studies suggested that DN even progresses faster in males than females.102, 103, 104, 105, 106, 107, 108 However, other results indicated an acceleration of disease progression in females,109, 110, 111, 112 whereas some studies reported no difference between men and women.113, 114, 115 Because ERβ can regulate cell apoptosis and cycle in tumor cells116 and ERβ protein expression is increased in podocytes treated with estrogen,117 estrogens could protect against podocytes apoptosis.117 The fact that podocytes isolated from estrogen-treated diabetic mice showed an increase in the level of AKT phosphorylation indicates that estrogen may achieve such an effect by activating the phosphatidylinositol-3′-kinase-AKT axis.117 The increased ERβ protein level in podocytes could manipulate the cell cycle and increase cell survival rates, suggesting that estrogen has the capability of preventing podocyte loss during diabetes-mediated kidney disease.117 Several lines of evidence revealed that TGFβ promotes diabetic kidney disease, at least partly through inducing cell apoptosis and podocyte clearance.118, 119, 120, 121 Relevant data showed that E2 treatment provides protection for podocytes against TGFβ or (TNFα)-induced apoptosis in vitro. Other studies suggested that treatment with E2 could be helpful to prevent albuminuria, glomerulosclerosis, and tubulointerstitial fibrosis in the initial stages of diabetes.122, 123, 124 However, some studies did not support the protective effects of estrogens for the patients with diabetic kidney.

apoptosis in vitro. Other studies suggested that treatment with E2 could be helpful to prevent albuminuria, glomerulosclerosis, and tubulointerstitial fibrosis in the initial stages of diabetes.122, 123, 124 However, some studies did not support the protective effects of estrogens for the patients with diabetic kidney. A recent study found that elevated serum concentrations of phytoestrogens are positively correlated with the severity of diabetic renal disease, suggesting the potential harmful effect of phytoestrogens.125

apoptosis in vitro. Other studies suggested that treatment with E2 could be helpful to prevent albuminuria, glomerulosclerosis, and tubulointerstitial fibrosis in the initial stages of diabetes.122, 123, 124 However, some studies did not support the protective effects of estrogens for the patients with diabetic kidney. A recent study found that elevated serum concentrations of phytoestrogens are positively correlated with the severity of diabetic renal disease, suggesting the potential harmful effect of phytoestrogens.125 Retinoic acid is the active metabolite of vitamin A, which plays a pivotal role in many physiological processes including but not limited to energy metabolism. Retinoic acid can facilitate the formation of retinoic acid receptor/RXR heterodimers or RXR/RXR homodimers, which could bind to the retinoic acid response element upstream of retinoic acid target gene promoters and modulate their transcription in the presence of specific ligands.126, 127 PPARs or other nuclear receptors can also form heterodimers with RXR, and modulate the biological function of several hormones and drugs.128, 129 For example, the RXRα:RXRα homodimer and RXRα:PPARγ are needed to recruit their coactivators to initiate the transcription of target genes130 through binding to their response elements.131 Considering that PPARγ is a key target in the treatment of DN, RXRα targeting may become a new treatment strategy. Furthermore, RXRs can be used as permissive heterodimers with LXR, FXR, PXR, and constitutive androstane receptor (CAR), or as nonpermissive heterodimer interacting with VDR, and as conditional heterodimers together with retinoid acid receptor or thyroid receptor (TR).132 On the other hand, because of the nature of its partners, the activation state of RXR changes in different heterodimers.133 Three RXR subtypes were identified as RXRα, RXRβ, and RXRγ.134, 135 As compared with the universal distribution of RXRα and RXRβ, RXRγ is only detected in some specific tissues.136 RXRα also showed antioxidants properties and played an important role in the pathogenesis of diabetic retinopathy.137 RXRγ encoded by RXRG gene was also involved in the pathogenesis of DN.138

d RXRγ.134, 135 As compared with the universal distribution of RXRα and RXRβ, RXRγ is only detected in some specific tissues.136 RXRα also showed antioxidants properties and played an important role in the pathogenesis of diabetic retinopathy.137 RXRγ encoded by RXRG gene was also involved in the pathogenesis of DN.138 Orphan Receptors HNF4α is expressed at high levels in the liver, kidney, and intestine, and controls the expression of a large gene set including those involved in glucose and fatty acid metabolism, urea biosynthesis, cholesterol metabolism, blood coagulation, hepatitis B virus infection, and hepatocyte differentiation.139 Dysfunction of HNF4α can lead to metabolic disease. Notably, genetic mutations in HNF4α result in maturity-onset diabetes of the young-1.140 The expression of the HNF4α gene is significantly decreased in the kidney and liver in 2 diabetic rodent models.141Additionally, HNF4α is decreased in kidneys of patients with DN. HNF4α negatively regulates the transcription of stromal interacting molecule-1, which is increased in a high glucose environment in mesangial cells.142 Blockage of HNF4α in mesangial cells might be a candidate therapeutic strategy for DN, as the stromal interacting molecule-1-gated store-operated Ca(2+) entry pathway in mesangial cells was recently found to be antifibrotic.142, 143, 144

c functions in the liver, pancreatic islet, kidney, and intestine.145 Maturity-onset diabetes of the young-3 result from rare mutations in HNF1A.146, 147 Although genetic variants in HNF1β are not a major cause of maturity-onset diabetes of the young or DN, they might lead to the manifestation of disease in Chinese.148 Conclusion The systematic and renal effects of nuclear hormone receptor activation in the context of diabetic nephropathy are shown in Table 1. Among the highlights, the fibrate class of PPARα agonists have long been prescribed to reduce triglyceride (TG), increase high-density lipoprotein-C (HDL-C), and improve cardiovascular outcomes in diabetic patients,149, 150 mainly by activating the expression of genes involved in lipid homeostasis.151 PPARα agonists also have the ability to improve renal lesion in DN animal models; however, whether a similar efficacy is also observed in diabetic patients remains to be determined.152 The VDR agonist calcitriol might ameliorate albuminuria by reducing urinary angiotensinogen levels.80 Furthermore, a combination treatment of mineralocorticoid receptor blockers with ACEI or ARB therapy has recently emerged, but the long-term efficacy and safety of such treatment has not been established.153 Although only a few nuclear receptors were evaluated as potential targets for the treatment of DN, clinical trials and animal studies have put more focus into the function of nuclear hormone receptors for protection against kidney disease. Identifying the mechanism by which activation of nuclear hormone receptors modulate kidney disease and determining their roles in the pathogenesis of DN, and the ultimate application of nuclear receptor targeting as a therapeutic strategy require considerably more experimentation.Table 1 Systematic and renal effects of nuclear hormone receptor activation in the context of diabetic nephropathy

modulate kidney disease and determining their roles in the pathogenesis of DN, and the ultimate application of nuclear receptor targeting as a therapeutic strategy require considerably more experimentation.Table 1 Systematic and renal effects of nuclear hormone receptor activation in the context of diabetic nephropathy Nuclear hormone receptor Affected genes, proteins, and processes Agonists Outcomes of receptor activation Mechanism of action PPARα14, 15, 19, 20, 21 (highly expressed in proximal tubule epithelium and medullary thick ascending limbs, with lower levels in glomerular mesangial cells) Fatty acid oxidation, NFκB, FGF21, PI3K/Akt/GSK-3β/Fyn-Nrf2 signaling, AMPK-PGC-1α-ERR-1α-FoxO3a signaling, PI3K/AKT, and ERK1/2 signaling Fibrates ↓Mesangial expansion, ↓matrix production, ↓proteinuria Systemic: ↓insulin resistance, ↓lipid, ↓hypertension Renal: anti-inflammatory, antifibrotic, antiapoptotic properties PPARγ14, 15, 26, 27, 29, 30, 31, 32, 33, 34 (primarily expressed in the epithelium of distal medullary collecting ducts and to a lesser extent in the glomerular mesangial cells, endothelial cells and podocytes, proximal tubular cells, endothelial cells of renal microvasculature, and interstitial fibroblast cells) Renal disintegrin, metalloprotease-17, angiotensin-converting enzyme-2, CB1R signaling, protein kinase A, pCREB, collagen-IV, MIP-3α, mPGES-1/PGE2/EP4 pathway, klotho axis, thioredoxin-interacting protein, RAGE, ROS Glitazones ↓Proteinuria; ↓glomerulosclerosis, ↓tubulointerstitial fibrosis Systemic: ↓insulin resistance, ↓lipid, ↓hypertension Renal: anti-inflammatory, antifibrotic, antioxidative, antiapoptotic properties PPARβ/δ14, 17, 43, 44 (highly expressed in medullary interstitial and stromal cells) α(1D)-adrenergic receptor, collagen-IV, RAGE, NFκB GW0742 ↓Proteinuria, ↓mesangial expansion, ↓tubulointerstitial fibrosis Renal: anti-inflammatory, antifibrotic LXR46, 48, 49, 50, 53, 55 (expressed in all major renal cells including mesangial cells, endothelial cells, and podocytes) ABCA1, SREBP-1c, OPN, MCP-1, TNFα, TGFβ, ROS T0901317 ↓Proteinuria, ↓mesangial expansion, ↓tubulointerstitial fibrosis, ↓macrophage infiltration in kidney Systemic: ↓lipid Renal: anti-inflammatory, antifibrotic, antioxidative FXR56, 57, 58, 64, 65, 66 (expressed in isolated glomeruli and proximal tubules, cultured mesangial cells, and podocytes) SREBP-1, visfatin Chenodeoxycholic acid, cholic acid ↓Proteinuria, ↓glomerulosclerosis, ↓mesangial cell inflammation, ↓mesangial expansion

lipid Renal: anti-inflammatory, antifibrotic, antioxidative FXR56, 57, 58, 64, 65, 66 (expressed in isolated glomeruli and proximal tubules, cultured mesangial cells, and podocytes) SREBP-1, visfatin Chenodeoxycholic acid, cholic acid ↓Proteinuria, ↓glomerulosclerosis, ↓mesangial cell inflammation, ↓mesangial expansion , ↓tubulointerstitial fibrosis Systemic: ↓lipid Renal: anti-inflammatory, antifibrotic, antioxidative VDR69, 70, 72, 73, 78, 80, 81, 82, 154, 155, 156 (expressed in the proximal and distal tubular epithelial cells, glomerular parietal epithelial cells, collecting duct cells and cultured podocytes, and mesangial cells) VDR-PPARγ pathway, TRPC6, heparanase, podocalyxin, STAT5, TGFβ, angiotensinogen, NF-κB 1,25-Dihydroxyvitamin D3, calcitriol ↓Proteinuria, ↓glomerulosclerosis, ↓macrophage infiltration in kidney Systemic: ↓insulin resistance, ↓RAS Renal: anti-inflammatory, antifibrotic, antiapoptotic MR85, 86, 87, 88, 89, 90, 91, 92, 93 (expressed in the ortical collecting duct cells of distal nephrons) TGFβ1, ROS, PAI-1, collagen, integrin β1, integrin β3, NKA, ROS Aldosterone ↑Proteinuria, ↓glomerular structural integrity Systemic: ↑hypertension, ↑insulin resistance Renal: inflammation, ↑fibrosis, ↑oxidation Estrogen receptors α and β117, 122, 123, 124, 157, 158, 159 (expressed in glomeruli, isolated mesangial cells and podocytes) PI3K-AKT signaling, TGFβ, TNFα Estrogen ↑Glomerular structural integrity, ↓protenuria, ↓glomerulosclerosis, ↓tubulointerstitial fibrosis Renal: anti-inflammatory, antifibrotic, antiapoptotic properties RXR131, 137 (distribution unknown in kidney) PPAR, LXR, FXR, VDR Magnolol NA Renal: antioxidative HNF4α142, 143, 144 (distribution unknown in kidney) STIM1 NA ↓Glomerulosclerosis Renal: antifibrotic ABCA1, ATP-binding cassette transporter A1; AKT, protein kinase B; AMPK, adenosine monophosphate kinase; CB1R, cannabinoid receptor type 1; ERK, extracellular signal–regulated kinase; ERR, estrogen-related receptor; FGF, fibroblast growth factor; FoxO3, forkhead box 03; FXR, farnesoid X receptor; GSK, glycogen synthase kinase 3β; LXR, liver X receptor; MIP, macrophage inflammatory protein; NA, not available; Nrf2, nuclear factor (erythroid-derived 2)-like 2; NKA, Na/K ATPase; NF-κB, nuclear factor κB; OPN, osteopontin; pCREB, p-cyclic-AMP-responsive element binding protein; PGCg-1α, peroxisome proliferator-activated receptor g coactivator-1α; PI3K, phosphatidylinositol-3′-kinase; PPAR, peroxisome proliferator–activated receptors; RAGE, receptor

erythroid-derived 2)-like 2; NKA, Na/K ATPase; NF-κB, nuclear factor κB; OPN, osteopontin; pCREB, p-cyclic-AMP-responsive element binding protein; PGCg-1α, peroxisome proliferator-activated receptor g coactivator-1α; PI3K, phosphatidylinositol-3′-kinase; PPAR, peroxisome proliferator–activated receptors; RAGE, receptor for advanced glycation end products; RAS, renin-angiotensin system; ROS, reactive oxygen species; SREBP-1c, sterol regulatory element-binding protein 1c; STAT5, signal transducer and activator of transcription 5; STIM1, stromal interacting molecule-1;TGFβ, transforming growth factor β; TNFα, tumor necrosis factor-α; TRCP6, transient receptor potential cation channel, subfamily C, member 6; VDR, vitamin D receptor. Disclosure All the authors declared no competing interests.

The incidence of acute kidney injury (AKI) is increasing around the world.1–4 The ongoing search for supporting procedures and interventions has produced improved guidelines and recommendations.5,6 Demonstration of increasing AKI incidence has led to an emphasis on prevention or early intervention,5 but unfortunately, analytical methods that predict AKI, or preventive and therapeutic approaches to accelerate recovery or prevent progression to chronic kidney disease (CKD), are only beginning to be understood.7–9 Early recognition of AKI is essential to ensure prompt and appropriate management, and to avoid progression to deadlier stages of the disease10,11 (Figure 1). In the appropriate context, early detection requires a high degree of suspicion that AKI is occurring. Diagnosis requires a combination of a clinical history, a thorough physical examination, an accurate assessment of kidney function, appropriate imaging, and when indicated, a kidney biopsy. In low- and middle-income countries (LMICs), early detection is impaired by limited resources and poor understanding of the condition.1,2,9,12–15 Such limited understanding—to a large extent determined by inadequate reporting and education—limits awareness and early recognition, and delays the implementation of measures that permit early and adequate management.16

arly detection is impaired by limited resources and poor understanding of the condition.1,2,9,12–15 Such limited understanding—to a large extent determined by inadequate reporting and education—limits awareness and early recognition, and delays the implementation of measures that permit early and adequate management.16 To address this goal, the steering committee of the 18th Acute Dialysis Quality Initiative (ADQI) conference dedicated a work group with the task to identify what elements affect the recognition of AKI within the limited resource constraints prevalent in LMICs. Using a modified Delphi process, this group reached consensus regarding strategies to recognize and diagnose AKI focusing on low resource countries. The group addressed the following 3 questions that served as the basis for accompanying consensus statements: When should AKI be suspected? What tests are needed when AKI is suspected? How do we confirm the diagnosis of AKI in patients with an initially elevated serum creatinine (Scr) level?

To address this goal, the steering committee of the 18th Acute Dialysis Quality Initiative (ADQI) conference dedicated a work group with the task to identify what elements affect the recognition of AKI within the limited resource constraints prevalent in LMICs. Using a modified Delphi process, this group reached consensus regarding strategies to recognize and diagnose AKI focusing on low resource countries. The group addressed the following 3 questions that served as the basis for accompanying consensus statements: When should AKI be suspected? What tests are needed when AKI is suspected? How do we confirm the diagnosis of AKI in patients with an initially elevated serum creatinine (Scr) level? Methods The ADQI process has been described previously.17,18 Complete ADQI methodology description is available at www.adqi.org and in the editorial accompanying the ADQI 18 conference papers.19 The broad objective of ADQI is to provide expert-based statements and interpretation of current knowledge for use by clinicians according to professional judgment, and to identify clinical research priorities to address these gaps. The 18th ADQI Consensus Conference Chairs convened a diverse panel that represented relevant disciplines (i.e., adult and pediatric nephrology, critical care, and renal pathology) from several continents (e.g., Africa, Asia, North America, Latin America, and Europe) around the theme of “Management of Acute Kidney Injury in the Developing World” for a 2-1/2–day consensus conference in Hyderabad, India on September 27 to 30, 2016.

adult and pediatric nephrology, critical care, and renal pathology) from several continents (e.g., Africa, Asia, North America, Latin America, and Europe) around the theme of “Management of Acute Kidney Injury in the Developing World” for a 2-1/2–day consensus conference in Hyderabad, India on September 27 to 30, 2016. The preconference activities involved a search of the literature for evidence on the epidemiology, recognition, and management of AKI in developing countries and their differences with developed countries. A literature search was conducted using the following terms: recognition; awareness; diagnosis; point of care; and low income countries or developing countries, together with either acute kidney injury and acute renal failure in PubMed. This work group was also tasked to summarize the scope, implementation, and evaluative strategies for AKI recognition and diagnosis based on the location, resource availability, and a critical evaluation of the relevant literature. A series of phone conferences and emails that involved work group members before the meeting identified current knowledge to enable the formulation of main questions from which discussion and consensus would be developed. A formal systematic review was not conducted. During the conference, the work group developed consensus positions, and plenary sessions that involved all ADQI contributors were used to present, debate, and refine these positions. Following the meeting, this summary report was generated, revised, and approved by all members of the ADQI participants. All the participants interacted throughout the meeting in the general session, and all group deliberations were subjected to review and consensus agreement in the final versions. In addition, all participants discussed and approved the contents of this paper. The participants did not represent specific societies, but were invited because they had domain knowledge expertise. Their affiliations are provided in the Supplementary Appendix.

ed to review and consensus agreement in the final versions. In addition, all participants discussed and approved the contents of this paper. The participants did not represent specific societies, but were invited because they had domain knowledge expertise. Their affiliations are provided in the Supplementary Appendix. For the purposes of all work group discussions, we used the current Kidney Disease Improving Global Outcomes (KDIGO) definitions for AKI and stages of AKI, which defines AKI as an episode that occurred within a 7-day timeframe.5 Community-acquired AKI was defined as an episode of AKI when the initial event occurred outside of the hospital setting and where the patient was admitted to the hospital with AKI; hospital-acquired AKI was defined as an episode of AKI due to a kidney insult that occurred to hospitalized patients who developed de novo AKI during their hospital stay.15 Q1: When Should AKI Be Suspected? Consensus Statement 1 In the appropriate clinical context, AKI should be suspected in patients who present with the signs and symptoms listed in Table 1.

For the purposes of all work group discussions, we used the current Kidney Disease Improving Global Outcomes (KDIGO) definitions for AKI and stages of AKI, which defines AKI as an episode that occurred within a 7-day timeframe.5 Community-acquired AKI was defined as an episode of AKI when the initial event occurred outside of the hospital setting and where the patient was admitted to the hospital with AKI; hospital-acquired AKI was defined as an episode of AKI due to a kidney insult that occurred to hospitalized patients who developed de novo AKI during their hospital stay.15 Q1: When Should AKI Be Suspected? Consensus Statement 1 In the appropriate clinical context, AKI should be suspected in patients who present with the signs and symptoms listed in Table 1. During the initial interaction of a patient with the health care system, the diagnosis of AKI is influenced by the clinical presentation and the context of the encounter11,20 (Figure 2). Improved awareness that the presenting symptoms and signs might correspond to AKI is the first step toward timely recognition. Unfortunately, AKI is frequently not recognized or is recognized too late, at a more severe stage.21 Failure to recognize early AKI is frequently associated with disease progression that requires more aggressive therapies and support when recovery is less likely and mortality is heightened.22

toward timely recognition. Unfortunately, AKI is frequently not recognized or is recognized too late, at a more severe stage.21 Failure to recognize early AKI is frequently associated with disease progression that requires more aggressive therapies and support when recovery is less likely and mortality is heightened.22 In LMICs, because of the common absence of access to specialized nephrology care, increased awareness of the clinical situations associated with AKI, and the implications of failing to detect it, AKI must be more understood at all levels of the health care system.23 A practical and easily accessible educational strategy focused on providers at the forefront of health care delivery is indispensable to achieve this goal. Providers must be trained to consider AKI in patients who present with certain signs and symptoms (Table 1)24 in the right clinical context. For example, in areas where infectious diseases (e.g., severe malaria, leptospirosis, or dengue) are endemic and associated with high rates of AKI,20,25–27 a febrile patient should elicit concern for renal injury.28–48 Similarly, in patients with severe volume depletion due to gastrointestinal loss, volume resuscitation is central to care and to prevent renal injury—preferably before the onset of persistent oliguria.15 Management must be appropriate to the clinical condition.49,50

patient should elicit concern for renal injury.28–48 Similarly, in patients with severe volume depletion due to gastrointestinal loss, volume resuscitation is central to care and to prevent renal injury—preferably before the onset of persistent oliguria.15 Management must be appropriate to the clinical condition.49,50 The development of AKI as a maternal and neonatal complication deserves special consideration in the LMIC environment,51–62 because failure to recognize renal injury frequently leads to significant consequences for both the mother and child. Successful efforts to improve early recognition have clearly demonstrated benefit, especially by reducing some of the more dreaded consequences such as cortical necrosis.63,64 In some areas of the world, exposure to snake venom represents a frequent cause of AKI.65,66 Administration of herbs by traditional healers has been associated with nephrotoxicity, and must be considered when confronted with AKI of unclear etiology.12,21,67 Increased availability and use of over-the-counter allopathic medications (e.g., nonsteroidal anti-inflammatory drugs) significantly contribute to a rising incidence of AKI. In LMIC, recognition of AKI in the hospital faces challenges that are akin to those seen in the developed world; hospitalized patients demonstrate a high incidence of AKI related to exposure to nephrotoxic medications, antibiotics, intravascular administration of iodinated radiocontrast, and surgical procedures.68,69 Consensus Statement

In LMIC, recognition of AKI in the hospital faces challenges that are akin to those seen in the developed world; hospitalized patients demonstrate a high incidence of AKI related to exposure to nephrotoxic medications, antibiotics, intravascular administration of iodinated radiocontrast, and surgical procedures.68,69 Consensus Statement 2 Evaluation for AKI should be incorporated into the diagnosis and management of specific endemic conditions associated with a high AKI risk (e.g., severe malaria, leptospirosis, dengue, and HIV). Endemic infections contribute significantly to the burden of AKI in LMICs. Much remains to be learned about the prevalence of AKI, the clinical characteristics that predispose to the onset of AKI, and the impact of AKI on the management of patients with those infections. Thus, the HIV epidemic in Sub-Saharan Africa has contributed to the rising burden of AKI, either as a direct result of the viral infection or as an unintended consequence of antiretroviral therapy.16,26,70–73 Other infectious diseases in LMICs have not received the same level of attention, and much remains to be understood about the nature of AKI associated with these conditions.48,49,74,75 Research Recommendation In LMICs, efforts must be directed to a better understanding of the epidemiology and management of infection-related AKI. Q2. What Tests Are Needed When AKI Is Suspected? Consensus Statement We recommend that patients suspected to have AKI should have an estimation of urinary output, a measurement of SCr levels, and a thorough urinalysis.

In LMICs, efforts must be directed to a better understanding of the epidemiology and management of infection-related AKI. Q2. What Tests Are Needed When AKI Is Suspected? Consensus Statement We recommend that patients suspected to have AKI should have an estimation of urinary output, a measurement of SCr levels, and a thorough urinalysis. Whenever possible, the performance of urine microscopy and urine biochemistry is essential to elucidate the underlying etiology and to assess severity. We recommend that point-of-care testing (POCT) technologies should be made available for the diagnosis of AKI in low resource settings. In hospitalized patients, we recommend additional testing, including renal imaging and renal biopsy, as indicated. The use of newer biomarkers of structural injury in economically constrained environments should await demonstration of efficacy. Confirmation of AKI The diagnosis and staging of AKI using current KDIGO definitions rests upon changes in serum creatinine and/ or urinary output.5 Additional testing and urinary microscopy are necessary to identify the underlying etiology.

In hospitalized patients, we recommend additional testing, including renal imaging and renal biopsy, as indicated. The use of newer biomarkers of structural injury in economically constrained environments should await demonstration of efficacy. Confirmation of AKI The diagnosis and staging of AKI using current KDIGO definitions rests upon changes in serum creatinine and/ or urinary output.5 Additional testing and urinary microscopy are necessary to identify the underlying etiology. Urinary Output In patients with developing AKI, urine output is a sensitive functional marker of kidney dysfunction.76–80 Unfortunately, oliguria may be easily confounded in its significance79 and can be difficult to record accurately, thereby limiting its reliability as a marker of AKI. In the community setting, diuresis is often unknown or inaccurately recorded, which limits its usefulness.5 In LMICs, oliguria is usually an accurate marker of AKI severity in children and neonates, and is associated with patient outcomes.39,81–83 Urinalysis When available, use of urine dipsticks and measurement of urinary indices such as urinary sodium,84,85 fractional excretion of sodium,86–88 fractional excretion of urea,84–86,89–91 urine plasma creatinine ratio,84 urine concentration (osmolality or specific gravity),84,85,92 and protein are useful for the initial evaluation of AKI.

e, use of urine dipsticks and measurement of urinary indices such as urinary sodium,84,85 fractional excretion of sodium,86–88 fractional excretion of urea,84–86,89–91 urine plasma creatinine ratio,84 urine concentration (osmolality or specific gravity),84,85,92 and protein are useful for the initial evaluation of AKI. The performance of basic urine microscopy,85,91,93–96 which focuses on the presence of erythrocytes, leukocytes, eosinophils, and casts in the sediment,60,97,98 is invaluable to assess the initial presentation of the patient with AKI (Table 2). We recommend that training in microscopic urine examination and availability of basic examination equipment for such testing should be promoted as a key, low-resource test for detection of AKI in LMICs. Although the usefulness of urinary indices (Table 3) in the critically ill patient with sepsis has been questioned,86,99 and may be confused by the use of diuretics, the combination of these tests with a thorough patient history, physical examination, and urinalysis will increase the sensitivity and specificity of AKI prediction and severity.100

urinary indices (Table 3) in the critically ill patient with sepsis has been questioned,86,99 and may be confused by the use of diuretics, the combination of these tests with a thorough patient history, physical examination, and urinalysis will increase the sensitivity and specificity of AKI prediction and severity.100 Serum Creatinine Despite limitations in the use of serum creatinine as a marker of renal function, changes in SCr and/or urine output form the basis of all AKI diagnostic criteria. SCr is a frequently inaccurate biomarker due to the need for a baseline and/or historical value to provide context101–104 and the limitations of a delayed diagnosis.105–108 Serum creatinine concentrations are affected by age, sex, and muscle mass109; they can change in response to certain drugs and are unreliable in patients with liver dysfunction or fluid overload. Serum levels take 24 to 36 hours to rise after a definite insult.110–113 In addition, although changes in creatinine concentration remain central to the diagnosis of AKI, differences in individual body composition that result in differences in creatinine production and volume of distribution across populations, as well as variations in dietary composition, have largely been ignored,102 and may be different from current estimates originated in the developed world.

o the diagnosis of AKI, differences in individual body composition that result in differences in creatinine production and volume of distribution across populations, as well as variations in dietary composition, have largely been ignored,102 and may be different from current estimates originated in the developed world. Until recently, the most common assay for measurement creatinine was the alkaline picrate (Jaffé) assay. However, chromogens other than creatinine interfere with the assay, giving rise to errors in up to 20% in subjects with a normal glomerular filtration rate (GFR). Modern assays do not detect noncreatinine chromogens and yield lower levels of creatinine. The lack of standardization to adjust for this interference affects the ability to estimate kidney function based on SCr concentration by different laboratories, especially at higher levels of estimated GFR. Standardization will reduce but not completely eliminate this error.114

ns and yield lower levels of creatinine. The lack of standardization to adjust for this interference affects the ability to estimate kidney function based on SCr concentration by different laboratories, especially at higher levels of estimated GFR. Standardization will reduce but not completely eliminate this error.114 Blood and Saliva Urea Nitrogen Serum urea and blood urea nitrogen (BUN) levels must be carefully interpreted as markers of kidney function in view of the numerous non-GFR factors that influence their blood concentrations. Levels of urea and/or BUN depend on protein intake, endogenous urea production, and tubular reabsorption. Reduced kidney perfusion in the setting of volume depletion enhances reabsorption of urea, which may lead to an elevation of BUN disproportionate to the concomitant decrease in GFR. Conversely, decreased protein intake or underlying liver disease can prevent the expected rise in BUN, whereas increased urea production (gastrointestinal bleeding, hypercatabolic status) or impaired protein anabolism (corticosteroid administration) can increase BUN in the absence of increased urea reabsorption.84,115 Because of multiple confounding, the use of BUN as an isolated marker of kidney injury may be unreliable. Additional POCT tools such as saliva urea nitrogen have been recently proposed and may be effective to screen patients with elevated urea nitrogen levels when blood tests may be unavailable or unaffordable.116

,115 Because of multiple confounding, the use of BUN as an isolated marker of kidney injury may be unreliable. Additional POCT tools such as saliva urea nitrogen have been recently proposed and may be effective to screen patients with elevated urea nitrogen levels when blood tests may be unavailable or unaffordable.116 Serum Cystatin C Currently, cystatin C is not being widely used. The absence of a relationship with body composition makes this marker an interesting alternative, but its value is limited by changes in concentration in response to inflammation, lung disease, and cigarette smoking.117

,115 Because of multiple confounding, the use of BUN as an isolated marker of kidney injury may be unreliable. Additional POCT tools such as saliva urea nitrogen have been recently proposed and may be effective to screen patients with elevated urea nitrogen levels when blood tests may be unavailable or unaffordable.116 Serum Cystatin C Currently, cystatin C is not being widely used. The absence of a relationship with body composition makes this marker an interesting alternative, but its value is limited by changes in concentration in response to inflammation, lung disease, and cigarette smoking.117 Point-of-Care Testing POCT for creatinine measurements occurs close to the patient instead of in a central laboratory (Table 4). It can be performed by nonlaboratory trained individuals, thus eliminating delays in testing and reporting of results.118 Although POCT is a particularly attractive option in remote and low resource environments, it requires the implementation of a quality assurance program that ensures accurate and reliable results. Several POCTs for Scr are available in the market across the world116,118–124 and can be classified into blood gas analyzers and nonblood gas analyzers. They also vary with respect to the types of samples that can be processed—whole blood, plasma, or serum. Other specific requirements include a power source, availability of deionized water, specific consumables (which sometimes require refrigeration), space, and requirements for calibration and disposal as a biohazard waste. As a result, most POCTs for SCr are not yet cost-effective and must be further tested for their usefulness in the detection of AKI.119 The failure of most of POCT creatinine devices to be in full alignment with isotope dilution mass spectrometry equivalent standards is another limitation.118,125 Definitive studies to determine the best practices to incorporate POCTs in low-resource health care settings are needed.

sefulness in the detection of AKI.119 The failure of most of POCT creatinine devices to be in full alignment with isotope dilution mass spectrometry equivalent standards is another limitation.118,125 Definitive studies to determine the best practices to incorporate POCTs in low-resource health care settings are needed. Novel Biomarkers As discussed, SCr as the current gold standard remains a flawed marker of renal dysfunction. Newer biomarkers are being developed, but even in high-income countries their use is yet to become a standard of care; their application in the developing world is even more challenging.126 Because of their simplicity of use and limited requirement for technological support, dipsticks are one of the most widely used tools to assess renal injury. Although traditional dipsticks allow the assessment of renal injury by primarily testing glomerular integrity (albuminuria and/or proteinuria), newer devices have more recently been modified as markers of renal dysfunction by estimating elevated BUN using saliva, or novel blood or urine markers of tubular injury such as kidney injury molecule-1 and neutrophil gelatinase-associated lipocalin.123,127 Recently, newer biomarkers in dipstick format have been made commercially available.112

ecently been modified as markers of renal dysfunction by estimating elevated BUN using saliva, or novel blood or urine markers of tubular injury such as kidney injury molecule-1 and neutrophil gelatinase-associated lipocalin.123,127 Recently, newer biomarkers in dipstick format have been made commercially available.112 AKI etiologies in low-resource rural areas, where volume depletion, infection, and nephrotoxic agents are leading causes of AKI,12,26 are usually different from those seen in the developed world.12 Such differences pose a challenge in our understanding of how potential novel biomarkers can be deployed. The ideal biomarker would facilitate the distinction between AKI due to volume depletion and AKI due to intrinsic kidney injury, and must be able to distinguish transient elevations in SCr from persistent changes consistent with injury. Such markers should allow early detection of the most likely cause of AKI, facilitate a diagnosis in the absence of historical information on baseline renal function, and support early therapeutic intervention.12,127 Unfortunately, novel AKI biomarkers remain poorly studied in clinical conditions commonly associated with AKI in LMICs; such limitations raise questions about their potential usefulness and practical implementation in those areas.128

ormation on baseline renal function, and support early therapeutic intervention.12,127 Unfortunately, novel AKI biomarkers remain poorly studied in clinical conditions commonly associated with AKI in LMICs; such limitations raise questions about their potential usefulness and practical implementation in those areas.128 Newer AKI Definitions, Staging Criteria, and Recent Uncertainties Although newer AKI definitions and staging criteria such as KDIGO; acute kidney injury network (AKIN); and risk, injury, failure, loss, and end-stage kidney disease (RIFLE)5,129–131 are appropriate to define AKI epidemiology and to design clinical trials, questions have been raised on their clinical application to the individual patient.111,112 The classification of AKI and its various stages has been validated in multiple hospitalized populations by demonstrating a strong association with short- and long-term outcomes,13,132 but significant problems in the usefulness of this classification persist.110 Because they rely on renal function changes, current AKI definitions only permit a relatively late diagnosis hours or days after the risk of injury or when the actual lesion began. As discussed previously, efforts to achieve an earlier diagnosis have led to the development of biomarkers of injury and are currently in progress.133 It is expected that newer biomarkers may detect kidney damage before the SCr and GFR become abnormal, but it is unclear how accurately those biomarkers will measure kidney damage instead of the severity of disease.134–139

er diagnosis have led to the development of biomarkers of injury and are currently in progress.133 It is expected that newer biomarkers may detect kidney damage before the SCr and GFR become abnormal, but it is unclear how accurately those biomarkers will measure kidney damage instead of the severity of disease.134–139 Because of current uncertainties on the correlation among AKI definitions, biomarker data, and histopathology,140 better availability of histopathologic data in LMICs provides a unique opportunity to probe such correlation, and begins to close the gap between our understanding of actual human histopathology, the pathogenesis of AKI, and our current, strictly functional KDIGO, AKIN, and RIFLE definitions.5,129–131 Histopathology in AKI A better understanding of the histopathology and pathogenesis of AKI is indispensable to continue to unveil the process of kidney injury,141 and by developing bench-to-bedside processes, to foster a better understanding on how to avoid and how to treat kidney injury.142

Because of current uncertainties on the correlation among AKI definitions, biomarker data, and histopathology,140 better availability of histopathologic data in LMICs provides a unique opportunity to probe such correlation, and begins to close the gap between our understanding of actual human histopathology, the pathogenesis of AKI, and our current, strictly functional KDIGO, AKIN, and RIFLE definitions.5,129–131 Histopathology in AKI A better understanding of the histopathology and pathogenesis of AKI is indispensable to continue to unveil the process of kidney injury,141 and by developing bench-to-bedside processes, to foster a better understanding on how to avoid and how to treat kidney injury.142 During the evaluation of patients with renal injury, a diagnosis based on histopathology remains important because it not only provides insight into the injury pattern, but often guides patient management. Multiple causes of AKI require histopathological diagnosis, but unfortunately, the number of biopsies and publications on the histopathology of AKI is declining.110,143 Concerns about procedural complications, including the risk of bleeding and the perception that AKI is commonly the result of acute tubular necrosis, appear to contribute to the reluctance to perform biopsies in the acute setting, despite evidence to the contrary.144–148

n the histopathology of AKI is declining.110,143 Concerns about procedural complications, including the risk of bleeding and the perception that AKI is commonly the result of acute tubular necrosis, appear to contribute to the reluctance to perform biopsies in the acute setting, despite evidence to the contrary.144–148 Kidney biopsies are indicated when: (i) The clinical presentation suggests that biopsy findings will likely lead to important therapeutic changes, an improved probability of recovery, and avoidance of further injury; (ii) when the magnitude of benefit is assessed to be greater than the risk of the procedure; and (iii) when the temporal course of the disease and delayed recovery dictates the need for further ascertainment of histopathologic diagnosis and prognosis. Multiple old and new studies have reviewed the indications and attested to the safety and usefulness of percutaneous kidney biopsies in the management of kidney disease.149–161

the temporal course of the disease and delayed recovery dictates the need for further ascertainment of histopathologic diagnosis and prognosis. Multiple old and new studies have reviewed the indications and attested to the safety and usefulness of percutaneous kidney biopsies in the management of kidney disease.149–161 Currently, kidney biopsies in patients with AKI are more common in LMICs than in high-income countries; thus, there is a greater appreciation of the relative incidence of multiple etiologies and the value of a renal biopsy to guide management.1,15,20,21,162 Although results from biopsy series are likely confounded by indication bias, those studies suggest that the role of a renal biopsy must be reconsidered in the diagnosis and management of AKI of unclear etiology, such as: unexplained AKI; acute interstitial nephritis60,163,164 acute or chronic glomerulonephritis, or rapidly progressive glomerulonephritis165 ; interstitial or tubular injury due to drug toxicity, or exposure to traditional herbal remedies21,166–170; thrombotic microangiopathies171 ; or leptospirosis.172–176 Because of current uncertainties on the relationship among AKI definitions, biomarker data and renal histopathology, and their effects on treatment and prognosis,140 we strongly recommend that kidney biopsies be considered in patients with AKI, whenever appropriate and feasible.

Currently, kidney biopsies in patients with AKI are more common in LMICs than in high-income countries; thus, there is a greater appreciation of the relative incidence of multiple etiologies and the value of a renal biopsy to guide management.1,15,20,21,162 Although results from biopsy series are likely confounded by indication bias, those studies suggest that the role of a renal biopsy must be reconsidered in the diagnosis and management of AKI of unclear etiology, such as: unexplained AKI; acute interstitial nephritis60,163,164 acute or chronic glomerulonephritis, or rapidly progressive glomerulonephritis165 ; interstitial or tubular injury due to drug toxicity, or exposure to traditional herbal remedies21,166–170; thrombotic microangiopathies171 ; or leptospirosis.172–176 Because of current uncertainties on the relationship among AKI definitions, biomarker data and renal histopathology, and their effects on treatment and prognosis,140 we strongly recommend that kidney biopsies be considered in patients with AKI, whenever appropriate and feasible. We further recommend that in LMIC settings, basic training be provided to local pathologists on renal histopathology, understanding that even the limited information provided by light microscopy may provide invaluable guidance in patient management. Training of members of the health care team in simple imaging, including ultrasonography, when feasible, is also desirable. Research Recommendation We recommend the development, validation, and standardization of POCT to facilitate the diagnosis of AKI in the community.

We further recommend that in LMIC settings, basic training be provided to local pathologists on renal histopathology, understanding that even the limited information provided by light microscopy may provide invaluable guidance in patient management. Training of members of the health care team in simple imaging, including ultrasonography, when feasible, is also desirable. Research Recommendation We recommend the development, validation, and standardization of POCT to facilitate the diagnosis of AKI in the community. Q3: How Do We Confirm the diagnosis of AKI in Patients With an Initially Elevated SCr Level? Consensus Statement 1 We recommend that patients with an isolated (single) elevated creatinine or oliguria be considered to have AKI until proven otherwise, to ensure rapid implementation of effective treatment measures. Concerns that the initially elevated SCr may be due to CKD may unnecessarily delay the initiation of urgent therapeutic measures. We strongly recommend that patients with apparently acute, severe dysfunction be emergently treated as if they had AKI, until proven otherwise (see the following). Consensus Statements 2 We recommend that the presence of CKD be evaluated using clinical history, urinalysis, renal imaging, and biopsy when indicated. 3 We recommend that the diagnosis of AKI should be confirmed by repeat assessment of renal function at no later than 7 days. We recommend that the frequency of repeat assessment of renal function be guided by the clinical context and response to intervention.

Consensus Statements 2 We recommend that the presence of CKD be evaluated using clinical history, urinalysis, renal imaging, and biopsy when indicated. 3 We recommend that the diagnosis of AKI should be confirmed by repeat assessment of renal function at no later than 7 days. We recommend that the frequency of repeat assessment of renal function be guided by the clinical context and response to intervention. Differentiation Between AKI and CKD When a patient without historical information presents in the community center with clinical features and/or an elevated creatinine consistent with a diagnosis of kidney injury, distinguishing isolated AKI from AKI superimposed on CKD or baseline CKD can be challenging. We believe this distinction should not be immediately relevant to the initial management, which should focus on the amelioration of the urgent metabolic and/or volume imbalances and on the correction of all known precipitating factors (Table 5).

lated AKI from AKI superimposed on CKD or baseline CKD can be challenging. We believe this distinction should not be immediately relevant to the initial management, which should focus on the amelioration of the urgent metabolic and/or volume imbalances and on the correction of all known precipitating factors (Table 5). We suggest that all patients without a known history of renal disease who present with a first episode of kidney injury must be presumed to have potentially reversible AKI, until proven otherwise. Moreover, even when the presence of CKD is demonstrated, modifiable factors that could have led to potentially reversible acute deterioration of renal function should be identified and corrected. This distinction becomes very relevant in certain regions of the world, where decisions are made on resource allocation in countries where public health care systems only offer support for the dialytic management of potentially reversible AKI, but frequently deny dialysis if renal failure is irreversible.

nd corrected. This distinction becomes very relevant in certain regions of the world, where decisions are made on resource allocation in countries where public health care systems only offer support for the dialytic management of potentially reversible AKI, but frequently deny dialysis if renal failure is irreversible. In patients presenting with kidney failure, all attempts should be made to explore whether previous measures of kidney function are available. This information can be part of previous encounters in the health care system, such as during pregnancy; presurgical screening; evaluation during an unrelated illness; or as part of medical screening before employment, insurance, or during school, corporate, or community health checks. In the fragmented LMIC health care systems, records are often unavailable, so when consulting, patients should be encouraged to bring all records of previous encounters with the health care system, which is a common practice in LMICs. Certain symptoms, signs, and laboratory or imaging findings (Table 6) can increase the suspicion of preexisting kidney disease, but should not be used to exclude the presence of coexisting AKI.

In patients presenting with kidney failure, all attempts should be made to explore whether previous measures of kidney function are available. This information can be part of previous encounters in the health care system, such as during pregnancy; presurgical screening; evaluation during an unrelated illness; or as part of medical screening before employment, insurance, or during school, corporate, or community health checks. In the fragmented LMIC health care systems, records are often unavailable, so when consulting, patients should be encouraged to bring all records of previous encounters with the health care system, which is a common practice in LMICs. Certain symptoms, signs, and laboratory or imaging findings (Table 6) can increase the suspicion of preexisting kidney disease, but should not be used to exclude the presence of coexisting AKI. In high-income countries, the first 48 hours of the SCr trajectory of patients hospitalized with initially elevated SCr has been used to evaluate the rate of AKI development and to assess whether kidney injury is transient or persistent.177 In this approach, the attainment of peak SCr after the initial creatinine elevation is considered an indication of persistent AKI. In LMICs, when community patients reach hospitals with established AKI such time-course information is usually not available. In those situations, excluding the possibility of the preexisting presence of CKD on a clinical basis may not be possible. Diagnosis may require either a kidney biopsy or be made retrospectively, when kidney function fails to improve despite appropriate supportive therapy.

e-course information is usually not available. In those situations, excluding the possibility of the preexisting presence of CKD on a clinical basis may not be possible. Diagnosis may require either a kidney biopsy or be made retrospectively, when kidney function fails to improve despite appropriate supportive therapy. Limitations The recommendations in this paper should not be limited to LMICs, but extended to all areas where nephrology resources are not widely available due to a variety of reasons, including cultural, geographic, or religious limitations. The World Bank country economic classification does not necessarily reflect either the health care structure or health care investment of each country. Many countries included in the LMIC category offer universal health care coverage, whereas some subpopulations in high-income countries may not have access to primary care, such as refugees, minorities, aboriginal peoples, or persons without health care coverage. Efforts should be directed toward a more granular analysis of the impact of health care investment and delivery on the recognition and management of AKI. Current limitations in the understanding of the epidemiology of AKI in LMICs are only beginning to be understood; continuously improving information will be necessary to enable the development of more accurate recommendations.

f the impact of health care investment and delivery on the recognition and management of AKI. Current limitations in the understanding of the epidemiology of AKI in LMICs are only beginning to be understood; continuously improving information will be necessary to enable the development of more accurate recommendations. Conclusions Measures to increase AKI awareness and recognition are essential to improve the treatment and prognosis of AKI in all regions of the world. To ensure a prompt to potentially reversible AKI, once a preliminary diagnosis is obtained by the demonstration of an elevated SCr, patients must be managed as if they had AKI until proven otherwise. Whenever possible, we recommend the pursuit of a diagnostic strategy geared toward the identification of the etiology of AKI to guide therapeutic options. This is particularly important in LMICs, where various endemic infections and toxicities often underlie renal damage. AKI is potentially treatable and reversible, and treatment is often specific to the underlying condition. To enhance AKI recognition, it is necessary to promote a better understanding of this epidemiological association of AKI with highly prevalent conditions, including endemic diseases, and to promote widespread education on AKI at all levels and to all members of the health care system. Supplementary Material 1 DISCLOSURE All the authors declared no competing interests. AUTHOR CONTRIBUTIONS JC, SM, GG, VJ, SS, SG, RC, and RM all participated in the consensus-building process and drafting of this paper. RM, RC, and AB provided a critical review of this paper.

To enhance AKI recognition, it is necessary to promote a better understanding of this epidemiological association of AKI with highly prevalent conditions, including endemic diseases, and to promote widespread education on AKI at all levels and to all members of the health care system. Supplementary Material 1 DISCLOSURE All the authors declared no competing interests. AUTHOR CONTRIBUTIONS JC, SM, GG, VJ, SS, SG, RC, and RM all participated in the consensus-building process and drafting of this paper. RM, RC, and AB provided a critical review of this paper. SUPPLEMENTARY MATERIAL Appendix S1. ADQI participants’ names and affiliations. Supplementary material is linked to the online version of the paper at www.kireports.org. Supported through the UAB-UCSD O’Brien Center NIH-NIDDK Grant DK079337. Figure 1 Acute kidney injury (AKI) recognition: the process and its modifiers. In addition to the usual AKI trajectory from clinical suspicion to confirmation to diagnosis, other factors modify the process. The degree of AKI awareness, the context in which the patient is encountered, and the available diagnostic resources may facilitate, delay, or impede the achievement of early AKI diagnosis. CKD, chronic kidney disease; KDIGO, Kidney Disease: Improving Global Outcomes; POC, point of care. Figure 2 Main components of the acute kidney injury diagnostic context. Table 1 Signs and symptoms leading to suspicion of acute kidney injury in low- to middle-income countries

Figure 1 Acute kidney injury (AKI) recognition: the process and its modifiers. In addition to the usual AKI trajectory from clinical suspicion to confirmation to diagnosis, other factors modify the process. The degree of AKI awareness, the context in which the patient is encountered, and the available diagnostic resources may facilitate, delay, or impede the achievement of early AKI diagnosis. CKD, chronic kidney disease; KDIGO, Kidney Disease: Improving Global Outcomes; POC, point of care. Figure 2 Main components of the acute kidney injury diagnostic context. Table 1 Signs and symptoms leading to suspicion of acute kidney injury in low- to middle-income countries In the community setting History of kidney disease Oliguria Total body swelling Hypotension Dehydration GI loss of volume and electrolytes Dark, concentrated urine Sepsis syndrome Fever in the context of prevalent endemic disease Exposure to potential nephrotoxins Pregnancy-related complications Plus, in the hospital setting, Multiple organ failure Nephrotoxic medication exposure GI, gastrointestinal. Table 2 Urine microscopy Reference Test Patients (n) Sensitivity (%) Specificity (%) PPV (%) NPV (%) Comments Bagshaw86 UMS ≥3 83 0.67 (0.39–0.86) 0.95 (0.84–0.99) 0.80 (0.49–0.94) 0.91 (0.78–0.96) The UMS was compared between septic and nonseptic AKI and correlated with NGAL, worsening AKI, RRT, and hospital mortality. UMS correlates with uNGAL, but not with pNGAL; a UMS score ≥ 3 was associated with increased odds of worsening AKI (AOR: 8.0; 95% CI: 1.03–62.5; P = 0.046).

0.99) 0.80 (0.49–0.94) 0.91 (0.78–0.96) The UMS was compared between septic and nonseptic AKI and correlated with NGAL, worsening AKI, RRT, and hospital mortality. UMS correlates with uNGAL, but not with pNGAL; a UMS score ≥ 3 was associated with increased odds of worsening AKI (AOR: 8.0; 95% CI: 1.03–62.5; P = 0.046). Perazella93 USS ≥2 267 0.76 0.86 100 44 Using the final diagnosis as the gold standard, the ability of the urine microscopy diagnosis to distinguish ATN from prerenal AKI was fair (sensitivity 0.76; specificity 0.86; positive LR 5.75). However, the scoring system was highly predictive of the final diagnosis of ATN Chawla96 CSI 30 Gold standard was patients with AKI consistent with the syndrome of acute tubular necrosis. The patients with nonrenal recovery had a higher CSI compared to those patients who did recover renal function (2.55 ± 0.93 vs. 1.57 ± 0.79; P = 0.04) Limitations: small sample, lack of control for urine osmolality and pH, the number of reviewers and the variation in the reviewer’s training was relatively limited.

sis. The patients with nonrenal recovery had a higher CSI compared to those patients who did recover renal function (2.55 ± 0.93 vs. 1.57 ± 0.79; P = 0.04) Limitations: small sample, lack of control for urine osmolality and pH, the number of reviewers and the variation in the reviewer’s training was relatively limited. Carvounis89 Scr ≥1.1 mg/dl 363 84.2 (74.4–90.7) 77.7 (72.5–82.1) 50.0 94.9 Renal epithelial cells or epithelial/granular casts 22.4 (14.5–32.9) 91.3 (87.5–94.0) 40.5 81.6 NGAL (ng/ml) ≥ 42.71 64.5 (53.3–74.3) 64.5 (58.8–69.8) 32.5 87.3 AKI, acute kidney injury; AOR, adjusted odds ratio; ATN, acute tubular necrosis; CI, confidence interval; CSI, Cast score index; LR, likelihood ratio; NGAL, neutrophil gelatinase-associated lipocalin; NPV, negative predictive value; PPV, positive predictive value; RRT, renal replacement therapy; UMS, urine microscopy score; USS, urinary scoring system. Table 3 Urine biochemistry Reference Test Patients (n) AUC Sensitivity (%) Specificity (%) PPV (%) NPV (%) Comments Carvounis89 FeU vs. FENa 102 90 96 99 75 Gold standard clinical grounds; more sensitive and specific index than FENa in differentiating between ARF due to prerenal azotemia and that due to ATN, especially if diuretics have been administered; in osmotic diuresis, the proximal tubular absorption of salt and water is impaired; thus, increased FEUN is expected despite renal hypoperfusion. A similar picture emerges in patients given a high protein diet or having excessive catabolism. PR = 50 Prdiu = 27 ATN = 25

N, especially if diuretics have been administered; in osmotic diuresis, the proximal tubular absorption of salt and water is impaired; thus, increased FEUN is expected despite renal hypoperfusion. A similar picture emerges in patients given a high protein diet or having excessive catabolism. PR = 50 Prdiu = 27 ATN = 25 Pepin90 FeNA vs. Feur in transient AKI (prerenal) 99 Gold standard clinical context and whether serum creatinine level returned to baseline within 7 days. In patients without diuretic use, FENa is better able to distinguish transient from persistent AKI. In patients administered diuretics, this distinction cannot be made accurately by means of FENa. FEur cannot be used as an alternative tool because it lacks specificity. FENa 0.83 ± 0.07 78 75 86 64 FENA + dir 0.75 ± 0.06 68 81 86 49 FEur 0.56 ±0.11 48 75 79 43 FEur + diur 0.57 ± 0.08 79 33 71 44

t from persistent AKI. In patients administered diuretics, this distinction cannot be made accurately by means of FENa. FEur cannot be used as an alternative tool because it lacks specificity. FENa 0.83 ± 0.07 78 75 86 64 FENA + dir 0.75 ± 0.06 68 81 86 49 FEur 0.56 ±0.11 48 75 79 43 FEur + diur 0.57 ± 0.08 79 33 71 44 Bagshaw86 FeU ≤35% n = 28 0.54 (0.42–0.67) 40 59 Gold standard: uNGAL. In sepsis, FeNa and FEUN are not reliable markers of renal hypoperfusion. Urine biochemical profiles and microscopy do not discriminate septic and non-septic AKI. UNa, FeNa, and FeU do not reliably predict biomarker release, worsening AKI, RRT or mortality. These data imply limited utility for these measures in clinical practice in critically ill patients FeNa <1% n = 47 0.54 (0.42–0.67) 50 58 AKI, acute kidney injury; ARF, acute renal failure; ATN, acute tubular necrosis; AUC, area under the curve; diur, diuretics; FENa, fractional excretion of sodium; FeU, fractional excretion of urea; FEUN, fractional excretion of urea nitrogen; NPV, negative predictive value; PPV, positive predictive value; RRT, renal replacement therapy; UNa, urine sodium; uNGAL, urine neutrophil gelatinase-associated lipocalin. Table 4 Issues that must be considered when selecting a point-of-care test

Bagshaw86 FeU ≤35% n = 28 0.54 (0.42–0.67) 40 59 Gold standard: uNGAL. In sepsis, FeNa and FEUN are not reliable markers of renal hypoperfusion. Urine biochemical profiles and microscopy do not discriminate septic and non-septic AKI. UNa, FeNa, and FeU do not reliably predict biomarker release, worsening AKI, RRT or mortality. These data imply limited utility for these measures in clinical practice in critically ill patients FeNa <1% n = 47 0.54 (0.42–0.67) 50 58 AKI, acute kidney injury; ARF, acute renal failure; ATN, acute tubular necrosis; AUC, area under the curve; diur, diuretics; FENa, fractional excretion of sodium; FeU, fractional excretion of urea; FEUN, fractional excretion of urea nitrogen; NPV, negative predictive value; PPV, positive predictive value; RRT, renal replacement therapy; UNa, urine sodium; uNGAL, urine neutrophil gelatinase-associated lipocalin. Table 4 Issues that must be considered when selecting a point-of-care test Ease of use Accuracy Low error rate (imprecision + bias) Consumable need: strips, cassettes, cartridges, rotor system, etc. Portability (handheld vs. bench top); different models may be appropriate for field vs. hospital settings Power source (battery vs. mains) Scalability Processing time Sample source and volume Connectivity (e.g., Bluetooth integration) Ability for integration into electronic decision support systems Possibility to do >1 test Cost of the device and consumables Table 5 Factors that can cause worsening renal function in a patient with preexisting kidney disease

alability Processing time Sample source and volume Connectivity (e.g., Bluetooth integration) Ability for integration into electronic decision support systems Possibility to do >1 test Cost of the device and consumables Table 5 Factors that can cause worsening renal function in a patient with preexisting kidney disease Systemic infection Infection of the urinary tract Volume deficit Urinary tract obstruction Uncontrolled hypertension Unrecognized renovascular disease Drug-induced (hemodynamic, interstitial nephritis) Table 6 Features that indicate the presence of preexisting kidney disease in a patient presenting with kidney injury History of long-standing nocturia History of edema, hematuria or renal stones History of long-term intake of painkillers, herbal medicines, over-the-counter drugs History of recurrent dehydration Family history of kidney disease Urinalysis showing broad casts Musculoskeletal manifestations: growth retardation, rickets, or proximal myopathy Anemia out of proportion to the duration of symptoms in the absence of another cause Elevated phosphate and/or PTH levels Characteristic imaging abnormalities (e.g., renal cysts or obstruction) Small and/or highly echogenic kidneys on ultrasound PTH, parathyroid hormone.

Introduction Carfilzomib is a selective proteasome inhibitor approved in 2012 for the treatment of relapsed and refractory multiple myeloma. It was developed with the aim of achieving improved safety profile and greater efficacy in patients who failed conventional treatments. A phase II trial for single-agent carfilzomib analyzed safety data in 526 treated patients and reported a rise in serum creatinine in 127 (24.1%) patients.1 In 73.2% of these 127 patients, the rise in serum creatinine was attributed to the carfilzomib with no other precipitating event identified.1 These data suggest that carfilzomib may be a cause of acute kidney injury (AKI), although the mechanism has not been determined. There have been several case reports providing evidence of AKI secondary to carfilzomib.2, 3, 4, 5, 6, 7 Two recent reports describe thrombotic microangiopathy associated with carfilzomib administration, although causality was not definitively established.4, 5 To our knowledge this is the first case report of biopsy-proven acute tubular necrosis (ATN) in a patient with multiple myeloma who was treated with carfilzomib.

7 Two recent reports describe thrombotic microangiopathy associated with carfilzomib administration, although causality was not definitively established.4, 5 To our knowledge this is the first case report of biopsy-proven acute tubular necrosis (ATN) in a patient with multiple myeloma who was treated with carfilzomib. Case Presentation A 60-year-old man with IgG-λ multiple myeloma who had received autologous stem cell transplantation 2 years prior and suffered a recent relapse presented to the hospital with shortness of breath and chest discomfort. Past medical history was also notable for atrial fibrillation and congestive heart failure with preserved ejection fraction. In the emergency department he appeared to be in mild distress with blood pressure of 141/74 mm Hg, heart rate 83 bpm, respirations 16 per minute, and an oxygen saturation of 97% on room air. Physical examination revealed clear lungs, normal S1 and S2 without murmur, and pitting edema of both legs. Electrocardiogram revealed normal sinus rhythm with peaked T waves in the anterior leads with right bundle branch block. Laboratory data, which are summarized in Table 1, were significant for serum sodium of 131 mmol/l, potassium of 6.3 mmol/l, and creatinine of 3.4 mg/dl. Free serum κ light chains were 5.55 mg/l, and serum λ light chains were 3630 mg/l (ratio 0.0015). Twenty days earlier, the patient had a baseline serum creatinine of 0.8 mg/dl, a serum free λ light chain level of 1800 mg/l, and a serum free κ light chain level of 3.96 mg/l (ratio 0.0022). Notably, the patient was given 2 consecutive injections of carfilzomib with decadron at a dose of 20 mg/m2 7 days prior to presentation. He denied nonsteroidal anti-inflammatory drug use, radiocontrast exposure, or any other changes in medications. His outpatient medications included acyclovir, warfarin, fentanyl patch, furosemide, gabapentin, digoxin, metoprolol, olanzapine, ramipril, potassium chloride, bupropion, and alprazolam. A urine sample obtained by bladder catheterization revealed pH 6.0, specific gravity 1.008, 1+ protein, 2+ blood, and 2 white blood cells and 63 red blood cells per high-power field. The spot urine protein/creatinine ratio was 3 g/g, and the albumin/creatinine ratio was 0.14 g/g. Urine κ light chains were 13.5 mg/l, and urine λ light chains were 8190.0 mg/l, yielding a urine κ/λ ratio of 0.0016. Renal ultrasound revealed no hydronephrosis and normal kidney size (right kidney 12.8 cm and left kidney 12 cm).

urine protein/creatinine ratio was 3 g/g, and the albumin/creatinine ratio was 0.14 g/g. Urine κ light chains were 13.5 mg/l, and urine λ light chains were 8190.0 mg/l, yielding a urine κ/λ ratio of 0.0016. Renal ultrasound revealed no hydronephrosis and normal kidney size (right kidney 12.8 cm and left kidney 12 cm). Several days later the creatinine stabilized at 2.6 mg/dl, at which point a kidney biopsy was performed.Table 1 Summary of laboratory results

urine protein/creatinine ratio was 3 g/g, and the albumin/creatinine ratio was 0.14 g/g. Urine κ light chains were 13.5 mg/l, and urine λ light chains were 8190.0 mg/l, yielding a urine κ/λ ratio of 0.0016. Renal ultrasound revealed no hydronephrosis and normal kidney size (right kidney 12.8 cm and left kidney 12 cm). Several days later the creatinine stabilized at 2.6 mg/dl, at which point a kidney biopsy was performed.Table 1 Summary of laboratory results Laboratory variable Prior to carfilzomib After carfilzomib References White blood cells 8.2 K/μl 9.3 K/μl 3.9–11.0 K/μl Hb 8.8 g/dl 9.1 g/dl 12.7–18.0 g/dl Platelets 100 K/μl 122 K/μl 160–392 K/μl Haptoglobin 150 mg/dl 151 mg/dl 40–290 mg/dl Lactate dehydrogenase Unavailable 144 IU/liter 100–250 IU/liter Sodium 137 mEq/l 131 mEq/l 138–145 mEq/l Potassium 4.1 mEq/l 6.3 mEq/l 3.7–5.2 mEq/l Creatinine 0.8 mg/dl 3.4 mg/dl 0.6–1.2 mg/dl Calcium 8.0 mg/dl 8.6 mg/dl 8.6–10.3 mg/dl Albumin 3.2 g/dl 3.9 g/dl 3.5–4.8 g/dl Urine protein/creatinine ratio Unavailable 3 g/g <0.2 g/g Urine albumin/creatinine ratio Unavailable 0.14 g/g <0.03 g/g Serum κ light chains 3.96 mg/l 5.55 mg/l 1.35–24.19 mg/l Serum λ light chains 1880 mg/l 3630 mg/l 0.24–6.66 mg/l Serum κ/λ light chain ratio 0.002 0.0015 0.26–1.65 Urine κ light chains Unavailable 13.50 mg/l 1.35–24.19 mg/l Urine λ light chains Unavailable 8190.00 mg/l 0.24–6.66 mg/l Urine κ/λ ratio Unavailable 0.0016 2.04–10.37 Urine sodium Unavailable 49 mEq/l Urine potassium Unavailable 32 mEq/l Urine chloride Unavailable 41 mEq/l Urine osmolarity Unavailable 352 mOsm/l Urine creatinine Unavailable 49.8 mg/dl Fractional excretion of sodium Unavailable 2.55% Urinalysis pH 7.0 6.0 5.0–8.0 Specific gravity 1.011 1.008 1.002–1.035 Protein Negative 1+ Negative Blood 1+ 2+ Negative Red blood cell number 0/hpf 63/hpf <3/hpf White blood cell number 0/hpf 2/hpf <3/hpf hpf, high-power field.

/l Urine creatinine Unavailable 49.8 mg/dl Fractional excretion of sodium Unavailable 2.55% Urinalysis pH 7.0 6.0 5.0–8.0 Specific gravity 1.011 1.008 1.002–1.035 Protein Negative 1+ Negative Blood 1+ 2+ Negative Red blood cell number 0/hpf 63/hpf <3/hpf White blood cell number 0/hpf 2/hpf <3/hpf hpf, high-power field. Renal Biopsy Findings The 7 glomeruli sampled for light microscopy were unremarkable. The major histologic finding was diffuse acute tubular injury involving 100% of the cortical parenchyma, affecting both proximal and distal tubules, associated with mild interstitial edema and sparse interstitial inflammation. The cortical tubules exhibited luminal ectasia, attenuation of brush border, focal coarse clear cytoplasmic vacuolization, and enlarged reparative nuclei containing nucleoli (Figure 1). A minority of distal tubules contained atypical hard crystalline casts of the myeloma type with giant cell reaction (Figure 2). By immunofluorescence microscopy, the casts revealed restricted 3+ staining for λ light chain, with negative staining for κ light chain. Congo red stain for amyloid was negative. The presence of diffuse acute tubular injury out of proportion to the sparse crystalline casts suggested ischemic or toxic ATN superimposed on mild myeloma cast nephropathy. The close temporal association with the initiation of carfilzomib and the absence of other obvious recent insults suggested that the medication had a role in development of the severe acute tubular injury.Figure 1 The major finding was diffuse acute tubular injury affecting both proximal and distal tubules with epithelial simplification, luminal ectasia, attenuation of brush border, coarse clear cytoplasmic vacuolization, and focal shedding of degenerating epithelial cells into the lumen. These degenerative tubular changes were present in tubules lacking myeloma-type casts (hematoxylin and eosin, original magnification X400).

epithelial simplification, luminal ectasia, attenuation of brush border, coarse clear cytoplasmic vacuolization, and focal shedding of degenerating epithelial cells into the lumen. These degenerative tubular changes were present in tubules lacking myeloma-type casts (hematoxylin and eosin, original magnification X400). Figure 2 The biopsy contained several atypical casts surrounded by multinucleated giant cells and dehisced tubular epithelial cells, typical of myeloma casts (hematoxylin and eosin, original magnification X600). The patient’s light-chain burden continued to increase, and he was treated with cyclophosphamide. His kidney function progressively worsened and he eventually required renal replacement therapy. Discussion Carfilzomib is a relatively new agent approved for the treatment of relapsed and refractory multiple myeloma. It has been associated with AKI as an adverse event in a phase II trial.1 Most of the cases of AKI in this phase II trial were attributed to carfilzomib, as no other precipitating cause could be identified; however, the mechanism of AKI was not determined. There have also been a number of case reports2, 3, 4, 5, 6, 7 attributing AKI to carfilzomib, some suggesting that thrombotic microangiopathy may have been the mechanism of injury based on clinical presentation and evidence from kidney biopsies, but definitive causality was not established.4, 5

AKI was not determined. There have also been a number of case reports2, 3, 4, 5, 6, 7 attributing AKI to carfilzomib, some suggesting that thrombotic microangiopathy may have been the mechanism of injury based on clinical presentation and evidence from kidney biopsies, but definitive causality was not established.4, 5 It can be difficult to determine the mechanism of AKI in patients with multiple myeloma because the differential is typically broad and includes a prerenal state from nausea and vomiting, hypercalcemia leading to renal vasoconstriction, monoclonal Ig deposition disease, myeloma cast nephropathy, infections, and drug-induced toxicity, among others (Table 2). A prerenal state was unlikely in our patient because he did not present with clinical signs of volume depletion, he had no vomiting, diarrhea, or hypercalcemia, and his fractional excretion of sodium was >2% approximately 24 hours after he last received furosemide. He presented with a rise in serum λ light chains and markedly reduced κ/λ ratio that can be explained by both worsening myeloma with increased production of light chains and decreased excretion due to kidney failure. Several case reports in the literature have suggested an association between carfilzomib and thrombotic microangiopathy.4, 5, 7 However, our patient had no clinical manifestations of thrombotic microangiopathy, such as hemolytic anemia, thrombocytopenia, or schistocytes, and no histologic evidence of thrombosis. There was no evidence on kidney biopsy of acute interstitial nephritis, which has been described in a single patient treated with bortezomib, a similar proteasome inhibitor.8Table 2 Teaching points

ic microangiopathy, such as hemolytic anemia, thrombocytopenia, or schistocytes, and no histologic evidence of thrombosis. There was no evidence on kidney biopsy of acute interstitial nephritis, which has been described in a single patient treated with bortezomib, a similar proteasome inhibitor.8Table 2 Teaching points Acute kidney injury in patients with multiple myeloma has many potential etiologies, including direct consequences of the hematologic malignancy and nephrotoxicity of therapeutic agents. Carfilzomib has been associated with acute kidney injury, but few patients have been subjected to diagnostic kidney biopsy. The patient presented here, who had multiple myeloma for years, developed acute tubular necrosis and mild myeloma cast nephropathy 1 week following exposure to carfilzomib. It is plausible that carfilzomib may promote acute tubular necrosis by direct cellular toxicity, possibly exacerbated by the toxic effects of monoclonal light chains.

ed here, who had multiple myeloma for years, developed acute tubular necrosis and mild myeloma cast nephropathy 1 week following exposure to carfilzomib. It is plausible that carfilzomib may promote acute tubular necrosis by direct cellular toxicity, possibly exacerbated by the toxic effects of monoclonal light chains. The patient presented herein suffered AKI 1 week after receiving 2 consecutive doses of carfilzomib. In the phase II trial,1 the incidence of first episodes of worsening renal function was evenly distributed across earlier and later time points, suggesting that a high cumulative exposure is not required for development of toxic AKI.1 The renal biopsy revealed ATN. While the renal biopsy also showed mild focal myeloma cast nephropathy, the degree of acute tubular injury appeared far out of proportion to the few myeloma casts. To our knowledge this is the first case report of biopsy-proven ATN in a patient with multiple myeloma treated with carfilzomib.

renal biopsy revealed ATN. While the renal biopsy also showed mild focal myeloma cast nephropathy, the degree of acute tubular injury appeared far out of proportion to the few myeloma casts. To our knowledge this is the first case report of biopsy-proven ATN in a patient with multiple myeloma treated with carfilzomib. Although not previously demonstrated, it is plausible that carfilzomib could cause ATN by its cellular effects on renal tubular epithelium. Carfilzomib is a selective proteasome inhibitor similar to bortezomib. Both drugs target the ubiquitin–proteasome system and inhibit the 20s proteasome. The ubiquitin–proteasome system is an intracellular degradation pathway in eukaryotic cells that normally leads to degradation of proteins such as p53 and nuclear factor-κB, which are involved in apoptosis, inflammation, senescence, and angiogenesis.9 Normal function of the 20s proteasomal system is critical to cellular maintenance and survival pathways. Inhibition of the 20s proteasome system, by reducing the degradation of proteins such as p53, could enhance apoptosis. While this is a desirable result for malignant cells, proteasome inhibition could exert harmful effects in renal tubular epithelial cells and potentially other cell types, resulting in AKI. In a murine model of ischemia–reperfusion injury, mice receiving the 20s proteasome inhibitor bortezomib experienced a significant increase in tubular cell apoptosis and greater decline in kidney function compared to control mice subjected to ischemia–reperfusion injury alone.9 Bortezomib and carfilzomib have similar mechanisms of action. However, acute kidney injury is less frequently reported following exposure to bortezomib as compared to carfilzomib. One explanation could be that carfilzomib is used in refractory or relapsed multiple myeloma where some tubular injury might already have occurred secondary to monoclonal light chains, thereby predisposing to further tubular injury by the drug.

ently reported following exposure to bortezomib as compared to carfilzomib. One explanation could be that carfilzomib is used in refractory or relapsed multiple myeloma where some tubular injury might already have occurred secondary to monoclonal light chains, thereby predisposing to further tubular injury by the drug. While carfilzomib may have been the etiologic factor causing ATN, we must also consider that the ATN could result from a combined effect of the drug and monoclonal light chains. Excessive production of monoclonal light chains may be directly toxic to tubular epithelial cells.10 For example, in an in vitro study, exposure to λ light chains induced a 6-fold increase in the number of apoptotic cultured human proximal tubule cells.10 Monoclonal light chains have been shown to generate intracellular oxidative stress in the form of hydrogen peroxide, which in turn promotes synthesis of chemokines and cytokines that lead to inflammation.11 In particular, monoclonal light chains activate apoptosis signal–regulating kinase 1, which is a key mediator of oxidative stress–induced apoptosis.11

wn to generate intracellular oxidative stress in the form of hydrogen peroxide, which in turn promotes synthesis of chemokines and cytokines that lead to inflammation.11 In particular, monoclonal light chains activate apoptosis signal–regulating kinase 1, which is a key mediator of oxidative stress–induced apoptosis.11 Given the potential for tubular toxicity from light chains, one must consider at least 2 additional possible mechanisms for ATN in this clinical setting. First, it is plausible that our patient suffered light chain–induced tubular injury that was then compounded by the “second hit” of carfilzomib, with the combined insult being sufficient to cause ATN. Second, it is plausible that the carfilzomib caused acute tubular injury, which in turn suddenly compromised the ability of the proximal tubules to endocytose and catabolize the high filtered load of monoclonal light chains, resulting in the development of cast nephropathy. However, it should be noted that there was no evidence of light chain proximal tubulopathy, in which crystalline intracytoplasmic inclusions develop within proximal tubular cells. Whether acting alone or in combination with nephrotoxic light chains, the close temporal relationship between carfilzomib therapy and AKI suggests that the drug played some pathogenetic role in the development of AKI.

l tubulopathy, in which crystalline intracytoplasmic inclusions develop within proximal tubular cells. Whether acting alone or in combination with nephrotoxic light chains, the close temporal relationship between carfilzomib therapy and AKI suggests that the drug played some pathogenetic role in the development of AKI. Conclusion To our knowledge this is the first report of biopsy-proven ATN occurring after carfilzomib treatment for multiple myeloma. Although our case demonstrates an association of carfilzomib administration and ATN, an exact mechanism of injury remains to be determined. This adverse event could be the result of a combined cellular effect of the drug itself and nephrotoxicity of monoclonal light chains. Greater use of renal biopsy in this setting may provide insight into the prevalence of ATN in multiple myeloma patients treated with carfilzomib. Disclosures All the authors declared no competing interests.

Acute kidney injury (AKI) has been recognized as a major public health problem.1, 2 The epidemiology, management, and prognosis of AKI vary considerably worldwide. Renal replacement therapy (RRT) is acutely applied to 20% to 25% of critically ill patients with AKI, but major variations in practice have been seen. In 2016, the Acute Disease Quality Initiative (ADQI) published consensus recommendations for the management of continuous renal replacement therapy (CRRT) to develop best clinical practice and standards of care.3, 4, 5, 6 However, clinicians in developing countries face additional challenges due to limited resources, reduced availability of trained staff and equipment, cultural and socioeconomic aspects, and administrative and governmental barriers, all of which affect patient selection, choice of RRT modality, and management of RRT.7, 8 Although some facilities for RRT are available in most metropolitan cities in these regions, children usually receive hemodialysis or peritoneal dialysis in adult units, whereas input from a dedicated pediatric team involved in multidisciplinary care are limited. Guidelines and recommendations for acute RRT need to incorporate these particular aspects of the condition.

ost metropolitan cities in these regions, children usually receive hemodialysis or peritoneal dialysis in adult units, whereas input from a dedicated pediatric team involved in multidisciplinary care are limited. Guidelines and recommendations for acute RRT need to incorporate these particular aspects of the condition. Methods This consensus meeting followed the established ADQI process, as previously described.9 The broad objective of ADQI is to provide expert-based statements and interpretation of current knowledge for use by clinicians according to professional judgment, as well as identify evidence care gaps to establish research priorities. The 18th ADQI Consensus Conference focused on “Management of AKI in the Developing World,” convening a diverse panel for a 2-1/2 day meeting in Hyderabad, India from September 27 to 30, 2016. The consensus-building process was informed by preconference, conference, and postconference activities. Before the conference, the workgroup searched PubMed for English language articles on dialysis support for AKI. This search included the terms “acute kidney injury” and “acute renal failure,” combined with “renal replacement therapy,” “continuous venovenous hemodialysis,” “continuous venovenous haemofiltration,” “hemodialysis,” “peritoneal dialysis,” “sustained low efficiency dialysis,” “CRRT,” “PIRRT,” “SLED” and “extracorporeal therapy.”

included the terms “acute kidney injury” and “acute renal failure,” combined with “renal replacement therapy,” “continuous venovenous hemodialysis,” “continuous venovenous haemofiltration,” “hemodialysis,” “peritoneal dialysis,” “sustained low efficiency dialysis,” “CRRT,” “PIRRT,” “SLED” and “extracorporeal therapy.” A preconference series of emails that involved the work group members was used to identify the current state of knowledge and enable the formulation of key questions. At the in-person meeting, the work group developed consensus statements through a series of alternating breakout and plenary sessions. In each breakout session, the work group refined the key questions, identified the supporting evidence, and generated consensus statements. Work group members presented the results for feedback to all ADQI participants during the plenary sessions, and then revised the drafts based on the plenary comments until a final version was accepted. We developed recommendations and consensus of expert opinion with evidence, where possible, to distill the current literature. To address important unanswered questions, we articulated a research agenda. Following the conference, this summary report was generated, revised, and approved by all members of the work group.

A preconference series of emails that involved the work group members was used to identify the current state of knowledge and enable the formulation of key questions. At the in-person meeting, the work group developed consensus statements through a series of alternating breakout and plenary sessions. In each breakout session, the work group refined the key questions, identified the supporting evidence, and generated consensus statements. Work group members presented the results for feedback to all ADQI participants during the plenary sessions, and then revised the drafts based on the plenary comments until a final version was accepted. We developed recommendations and consensus of expert opinion with evidence, where possible, to distill the current literature. To address important unanswered questions, we articulated a research agenda. Following the conference, this summary report was generated, revised, and approved by all members of the work group. Q1: What Are the Minimal Infrastructure Requirements for RRT? Consensus Statements 1.1. We recommend the availability of an essential core team of trained personnel, consisting of at least 1 physician and healthcare professional dedicated to managing the dialysis therapy. If intermittent hemodialysis (IHD), prolonged intermittent renal replacement therapy (PIRRT), sustained low efficiency dialysis (SLED), and/or CRRT are used, a technician for machine maintenance should be available. 1.2. We recommend the availability of peritoneal dialysis (PD) catheters and vascular access catheters for PD and hemodialysis (HD) techniques.

Q1: What Are the Minimal Infrastructure Requirements for RRT? Consensus Statements 1.1. We recommend the availability of an essential core team of trained personnel, consisting of at least 1 physician and healthcare professional dedicated to managing the dialysis therapy. If intermittent hemodialysis (IHD), prolonged intermittent renal replacement therapy (PIRRT), sustained low efficiency dialysis (SLED), and/or CRRT are used, a technician for machine maintenance should be available. 1.2. We recommend the availability of peritoneal dialysis (PD) catheters and vascular access catheters for PD and hemodialysis (HD) techniques. 1.3. We recommend the availability of appropriate fluid bags and tubing in case of PD, and appropriate filters, circuits, and fluids in case of extracorporeal RRT. 1.4. We recommend that the essential core team and equipment be available at all times. 1.5. We recommend that units managing children who need acute RRT have the appropriate infrastructure, equipment, and trained personnel to provide appropriate standards of care.

1.3. We recommend the availability of appropriate fluid bags and tubing in case of PD, and appropriate filters, circuits, and fluids in case of extracorporeal RRT. 1.4. We recommend that the essential core team and equipment be available at all times. 1.5. We recommend that units managing children who need acute RRT have the appropriate infrastructure, equipment, and trained personnel to provide appropriate standards of care. Context Barriers to providing RRT in developing countries or resource-limited regions can be due to regional impediments, RRT-related aspects, and patient-related factors (Table 1). Examples of regional barriers include environmental challenges, logistics, and inadequate administrative or policy support by the government or institution. Delivery of RRT can be hindered by decreased availability of equipment, lack of trained healthcare personnel, absence of a regulatory framework to ensure quality of dialysis, decreased availability of laboratory tests for monitoring of RRT, and financial costs.8 Inadequate technical support leads to poor equipment maintenance, frequent machine breakdowns, and interruptions or delay in treatment. Patient barriers to RRT include cultural beliefs and socioeconomic aspects that influence the decision to start RRT and the type of modality, including the ability to pay for such services. Important geographic factors are the availability of transportation and the distance patients would have to travel to receive RRT, because RRT is usually only obtainable in larger cities for those who can afford to pay.Table 1 Barriers at several levels for receiving renal replacement therapy in developing countries

s. Important geographic factors are the availability of transportation and the distance patients would have to travel to receive RRT, because RRT is usually only obtainable in larger cities for those who can afford to pay.Table 1 Barriers at several levels for receiving renal replacement therapy in developing countries Population Sociocultural: Customs, health beliefs, accessibility and beliefs in other health systems Policy and financial: Lack of legislation to provide health care Medical and scientific: Lack of scientific data from the developing countries, scepticism to accept the scientific data derived from developed countries Socioeconomic: Lack of infrastructure, such as continuous provision of electricity, good quality water, and sanitation Healthcare system Lack of administrative support at the level of hospital, local, state, and national governments Lack of physicians trained to provide RRT Density of physicians and geographic distance from centers providing RRT Existence of several different health systems, especially indigenous systems Healthcare provider Lack of infrastructure to provide RRT Wide variation in the quality of care and infrastructure to provide RRT Lack of trained personnel to provide RRT at all times; limited training to manage RRT in children Lack of technical support to maintain and service the dialysis machines High cost of RRT Lack of laboratory facilities and high cost of laboratory tests Late referral of patients to centres providing RRT Patients Fear and anxiety of the patient and family regarding RRT Health insurance availability, access and coverage Financial constraints RRT, renal replacement therapy.

dialysis machines High cost of RRT Lack of laboratory facilities and high cost of laboratory tests Late referral of patients to centres providing RRT Patients Fear and anxiety of the patient and family regarding RRT Health insurance availability, access and coverage Financial constraints RRT, renal replacement therapy. For RRT to be safe and effective, a minimal infrastructure has to be in place. This can only be achieved with full local commitment, a viable financial model, and the availability of skilled staff and equipment. All dialytic devices need to function properly at all times, and trained personnel and equipment should be available on a 24-hour basis.10 An essential core of trained personnel has to have the expertise to prescribe, provide, manage and monitor the dialytic therapy. The number of personnel should be sufficient to ensure adequate patient care and safety. If the dialytic therapy involves the use of IHD, PIRRT, or CRRT machines, a qualified technician or engineer for maintenance and regular preventive servicing of the machines should be available. Portable water and engineering systems for the production of pure water need to be in place.10 The water used to produce the dialysate should be treated to achieve the standards of the Association for the Advancement of Medical Instrumentation.11

nce and regular preventive servicing of the machines should be available. Portable water and engineering systems for the production of pure water need to be in place.10 The water used to produce the dialysate should be treated to achieve the standards of the Association for the Advancement of Medical Instrumentation.11 Essential equipment for the delivery of PD includes catheters with appropriate fluid bags and tubing. Adapting nasogastric tubes for PD should be discouraged. For IHD, PIRRT, and CRRT, vascular access catheters should be available, as well as appropriate dialyzers, circuits, fluids, and emergency electric power supply for life-saving equipment in case of power failure. Written protocols for all procedures, including cleaning and disinfection of surfaces and equipment should be available.

T, and CRRT, vascular access catheters should be available, as well as appropriate dialyzers, circuits, fluids, and emergency electric power supply for life-saving equipment in case of power failure. Written protocols for all procedures, including cleaning and disinfection of surfaces and equipment should be available. Reusing dialyzers and tubes is common practice in developing countries and usually follows manual reprocessing. A study from Sri Lanka showed that the reuse of hemodialyzers in patients with end-stage renal failure (ESRF) resulted in 40% cost saving of consumables and a reduction in the hourly dialysis expense by one-third.12 However, concerns have been raised about reduced dialyzer efficiency and an increased mortality risk with reuse of dialyzers.13 Regions that reuse tubing and dialyzers should have an appropriate protocol for reprocessing, facilities for cleaning, and reliable monitoring systems. Only dialyzers labeled for multiple uses should be used. The chemical quality of water used for dialyzer reprocessing should meet the same Advancement of Medical Instrumentation standards as for dialysate.11 Agents used for disinfecting dialyzers are sodium hypochlorite, formaldehyde, glutaraldehyde, or peracetic acid. Technicians and other personnel responsible for the reprocessing of dialyzers should receive proper training, including training for infection control.

f Medical Instrumentation standards as for dialysate.11 Agents used for disinfecting dialyzers are sodium hypochlorite, formaldehyde, glutaraldehyde, or peracetic acid. Technicians and other personnel responsible for the reprocessing of dialyzers should receive proper training, including training for infection control. Most children in developing countries receive HD in adult units, often with limited input from a pediatric multidisciplinary team trained in the relevant medical, nursing, developmental, and psychosocial issues. It is recommended that units managing children who need acute RRT should have the appropriate infrastructure and equipment, as well as trained personnel to ensure standards of care.14 Physicians taking care of children in developing countries should receive appropriate training and acquire requisite knowledge and skills to meet the specific needs of children on RRT.14 Saving Young Lives is an initiative of 3 international societies (i.e., the International Society of Nephrology, International Pediatric Nephrology Association, and International Society for Peritoneal Dialysis) and the Sustainable Kidney Care Foundation. With a focus on education and training, it has successfully developed sustainable programs for treatment of AKI in sub-Saharan Africa and South East Asia.15 Research Recommendations • To compile a registry for dialysis availability in different regions and countries of the world that can be used to study practice patterns, identify barriers to the use of dialysis, and determine outcomes in low resource regions.

Saving Young Lives is an initiative of 3 international societies (i.e., the International Society of Nephrology, International Pediatric Nephrology Association, and International Society for Peritoneal Dialysis) and the Sustainable Kidney Care Foundation. With a focus on education and training, it has successfully developed sustainable programs for treatment of AKI in sub-Saharan Africa and South East Asia.15 Research Recommendations • To compile a registry for dialysis availability in different regions and countries of the world that can be used to study practice patterns, identify barriers to the use of dialysis, and determine outcomes in low resource regions. • To develop strategies for training of healthcare workers to provide RRT in low resource regions. • To develop more techniques for reliable, affordable, and cost-effective RRT. Q2: Who Should Be Considered for RRT? Consensus Statements 2.1. We recommend RRT should be initiated emergently when life-threatening changes in fluid, electrolytes, and acid-base balance are unresponsive to medical therapy. 2.2. We recommend RRT should be considered when metabolic and fluid demands exceed total kidney capacity. 2.3. We suggest that factors such as patient preference, quality of life, comorbid conditions, severity of acute illness, expected prognosis, urine output, logistics, and social and cultural issues should be considered when deciding whether to start RRT. 2.4. We suggest that in severely ill patients a shared decision-making process with the physician, patient, and family should be undertaken to decide whether to start RRT.

2.3. We suggest that factors such as patient preference, quality of life, comorbid conditions, severity of acute illness, expected prognosis, urine output, logistics, and social and cultural issues should be considered when deciding whether to start RRT. 2.4. We suggest that in severely ill patients a shared decision-making process with the physician, patient, and family should be undertaken to decide whether to start RRT. 2.5. We suggest that a palliative care program should be available for supportive care.

2.3. We suggest that factors such as patient preference, quality of life, comorbid conditions, severity of acute illness, expected prognosis, urine output, logistics, and social and cultural issues should be considered when deciding whether to start RRT. 2.4. We suggest that in severely ill patients a shared decision-making process with the physician, patient, and family should be undertaken to decide whether to start RRT. 2.5. We suggest that a palliative care program should be available for supportive care. Context Multiple factors should be taken into consideration when deciding whether to initiate RRT for AKI (Figure 1). It is well accepted that dialysis should be initiated emergently in patients with life-threatening indications, such as severe hyperkalemia, severe acidosis, pulmonary edema, and uremic complications refractory to medical management.16 However, beyond these absolute indications, the Kidney Disease Improving Global Outcome (KDIGO) clinical practice guidelines recommend considering the broader clinical context of the patient.16 This includes taking into account the severity of the underlying disease, degree of dysfunction of other organs, severity of fluid overload, solute burden and urine output, and the likelihood of recovery of kidney function.4, 16 In the last 2 years, 3 randomized controlled trials (RCTs) compared early initiation of RRT versus late initiation of RRT.17, 18, 19 In 2 studies, 36% and 50% of the patients assigned to late RRT had spontaneous renal recovery without ever receiving RRT,17, 19 which highlights that efforts should be made to identify patients with a high probability of early renal recovery in whom RRT may be avoidable.20 The decision to start acute RRT should be individualized, and RRT should be considered when the capacity of the kidneys cannot meet the demands being placed on them4, 21 (Figure 2).Figure 1 Factors to consider for renal replacement therapy (RRT) initiation in acute kidney injury.

ecovery in whom RRT may be avoidable.20 The decision to start acute RRT should be individualized, and RRT should be considered when the capacity of the kidneys cannot meet the demands being placed on them4, 21 (Figure 2).Figure 1 Factors to consider for renal replacement therapy (RRT) initiation in acute kidney injury. Figure 2 Demand versus capacity paradigm. The upper left quadrant represents the normal condition in which metabolic and fluid demand on the kidneys is low, and the kidneys have full capacity. The upper right quadrant represents the situation in which renal capacity is preserved but demand is high, and the lower left quadrant represents the situation in which renal capacity is decreased but the demand on the kidney is also low; in these 2 situations the kidneys can be managed conservatively without renal replacement therapy (RRT). The right lower quadrant represents the situation in which the demand on the kidneys is high and the capacity and/or function of the kidneys is low. In this situation, RRT should be initiated.

he kidney is also low; in these 2 situations the kidneys can be managed conservatively without renal replacement therapy (RRT). The right lower quadrant represents the situation in which the demand on the kidneys is high and the capacity and/or function of the kidneys is low. In this situation, RRT should be initiated. Other factors relevant to the decision-making process are patient preference, quality of life, comorbid conditions, expected prognosis, logistics, and social and cultural issues. In developing countries or resource-limited regions, physicians may be faced with the ethical dilemma of the appropriateness of starting RRT in patients with a poor prognosis due to significant acute and chronic comorbid conditions. If the patient is critically ill or has otherwise a poor prognosis, a shared decision process with the patient and family should be undertaken. When it is unclear if a patient will benefit from dialysis, a time-limited trial may be used to determine the benefit of treatment versus the burden of treatment.22 For those who decide not to start dialysis, it is important to have formal conservative care programs available. Palliative care should be offered to all patients with AKI regardless of whether they start or decline RRT. A formal palliative program should consist of a team of clinicians and trained personnel who provide expert management of pain and other symptoms, emotional and spiritual support, and guidance with difficult treatment choices. The goal of palliation would be to improve quality of life for both the patient and the family, and to help the patient and family understand the treatment options and goals.

rsonnel who provide expert management of pain and other symptoms, emotional and spiritual support, and guidance with difficult treatment choices. The goal of palliation would be to improve quality of life for both the patient and the family, and to help the patient and family understand the treatment options and goals. Research Recommendations • To develop a clinical decision system that helps healthcare workers deciding when and how to implement RRT. • To determine reproducible criteria for the demand to capacity paradigm to inform the decision to start RRT. Q3: How Should RRT Be Delivered in AKI Patients? Q3a: What Are the Goals of RRT? Consensus Statements 3a.1. We recommend the short-term goal of RRT for AKI is to support the kidneys’ capacity to overcome the metabolic and fluid demands and to achieve control of azotemia, acid-base and electrolyte derangement, and fluid overload. 3a.2. We recommend the long-term goals of RRT for AKI are to improve survival and promote renal recovery. Context The concept of demand capacity balance in AKI was originally described by Macedo and Mehta,21 and recently recommended in a consensus ADQI meeting.4 The demand is determined by severity of the acute illness, and the solute and fluid burden. The demand capacity balance is dynamic in nature and varies with the course of critical illness. When renal capacity decreases and fails to cope with the demands, initiation of RRT should be considered (Figure 2).

ADQI meeting.4 The demand is determined by severity of the acute illness, and the solute and fluid burden. The demand capacity balance is dynamic in nature and varies with the course of critical illness. When renal capacity decreases and fails to cope with the demands, initiation of RRT should be considered (Figure 2). The aim of acute RRT is to support native kidney function in controlling acid base and electrolyte derangements, as well as fluid overload, and to reduce the effects of AKI on nonrenal organs. Monitoring of serum creatinine, electrolytes, and cumulative fluid balance is necessary to adjust RRT according to the needs of the patient. The long-term goals are patient survival and renal recovery. The latter is relevant to low-resource settings. To date, there are insufficient data to recommend specific RRT techniques to facilitate renal and patient recovery.23 There is also a lack of evidence on the optimal timing and mode of discontinuation4 (see Q4: When Should RRT Be Transitioned or Stopped?) Recommendations for Future Research • To evaluate whether fluid overload at initiation and during RRT affects renal recovery. • To investigate whether the degree of control of azotemia during RRT affects mortality and renal recovery. • To investigate the optimal method of delivering acute RRT in clinical settings relevant to developing countries, including poisoning or obstetric AKI. • To include renal recovery as an outcome in clinical trials and cost utility analyses of RRT, especially if conducted in developing countries.

• To investigate whether the degree of control of azotemia during RRT affects mortality and renal recovery. • To investigate the optimal method of delivering acute RRT in clinical settings relevant to developing countries, including poisoning or obstetric AKI. • To include renal recovery as an outcome in clinical trials and cost utility analyses of RRT, especially if conducted in developing countries. Q3b: What Is the Most Appropriate Type of RRT? Consensus Statements 3b.1. We suggest that the choice of the initial RRT modality is primarily based on the local availability and experience with a specific treatment and the patient’s clinical status. 3b.2. We recommend IHD for life-threatening emergent indications. 3b.3. We recommend IHD and PIRRT when mobilization is the priority, and fluid and metabolic control can be obtained. 3b.4. In patients with acute brain injury or increased intracranial pressure, we recommend the use of CRRT or PD, if available. 3b.5. We recommend the use of CRRT, PD, or PIRRT in situations where fluctuations in fluid balance and solutes are poorly tolerated. 3b.6. For patients with an increased catabolic state, we suggest CRRT, PIRRT, or IHD over PD. 3b.7. All dialysis modalities provide particular benefits and should be used accordingly to optimize care. Transition of modality should be considered when the patient’s condition allows, and adequate infrastructure and trained personnel are available.

3b.6. For patients with an increased catabolic state, we suggest CRRT, PIRRT, or IHD over PD. 3b.7. All dialysis modalities provide particular benefits and should be used accordingly to optimize care. Transition of modality should be considered when the patient’s condition allows, and adequate infrastructure and trained personnel are available. Context Current RRT modalities for AKI include IHD, PIRRT (including SLED), CRRT, and PD. Table 2 shows the advantages and disadvantages of the different techniques. There are at least 10 RCTs that compared IHD versus CRRT.24, 25, 26, 27, 28, 29, 30, 31 However, several studies were limited by restricted patient selection, protocol deviations, and the need for crossover treatment. Three systematic reviews and meta-analyses were also published, all of which found no significant differences in mortality or recovery of kidney function between patients treated with intermittent or continuous modalities.32, 33, 34 The most recent meta-analysis by the Cochrane Collaboration, which included 15 RCTs in 1550 critically ill patients with AKI, concluded that there was no significant difference in mortality in hospital and in the intensive care unit (ICU), length of hospitalization, and chances of renal recovery in survivors between patients treated with CRRT and IHD.34 However, in patients who received CRRT, the mean arterial pressure (MAP) was significantly higher at the end of the treatment, and the number of patients who required escalation of vasopressor therapy was significantly lower.Table 2 Advantages and disadvantages of dialysis modalities

tients treated with CRRT and IHD.34 However, in patients who received CRRT, the mean arterial pressure (MAP) was significantly higher at the end of the treatment, and the number of patients who required escalation of vasopressor therapy was significantly lower.Table 2 Advantages and disadvantages of dialysis modalities Factors IHD PIRRT CRRT PD Need for vascular access +++ +++ +++ − Need for anticoagulation + ++ +++ − Need for peritoneal integrity − − − +++ Impact on diaphragm movement − − − + Speed of toxin removal +++ ++ + + Risk of cerebral edema +++ ++ + + Hemodynamic tolerance + ++ +++ +++ Solute and balance stability + ++ +++ +++ Removal of nutrients and drugs + ++ +++ ++ Complexity +++ +++ ++ + Time for mobilization +++ ++ + ++ Financial costs ++ ++ +++ + +, weakly relevant; ++, moderately relevant; +++, very relevant; –, not relevant; CRRT, continuous renal replacement therapy; IHD, intermittent hemodialysis; PD = peritoneal dialysis; PIRRT, prolonged intermittent renal replacement therapy. Based on these data, the KDIGO Clinical Practice Guidelines for AKI in adults recommends considering continuous and intermittent RRT modalities as complementary, except for 2 specific patient groups for whom CRRT is recommended over standard intermittent RRT: patients with intracranial hypertension and/or acute brain injury, and patients with hemodynamic instability.16

idelines for AKI in adults recommends considering continuous and intermittent RRT modalities as complementary, except for 2 specific patient groups for whom CRRT is recommended over standard intermittent RRT: patients with intracranial hypertension and/or acute brain injury, and patients with hemodynamic instability.16 Hybrid treatments, such as PIRRT and SLED, incorporate the advantages of both CRRT and IHD and are used worldwide in many ICUs.35, 36, 37 They may be considered for hemodynamically unstable patients in situations where other forms of CRRT are not available, but data on comparative efficacy and harm are limited.38, 39, 40, 41 A systematic review and meta-analysis including 17 studies from 2000 to 2014 (7 RCTs and 10 observational studies involving 533 and 675 patients, respectively) focused on the impact of PIRRT and CRRT on mortality and renal recovery.42 Meta-analysis of the RCTs only showed no difference in mortality between both modalities (relative risk [RR]: 0.90; 95% confidence interval [CI]: 0.74–1.11; P = 0.3). However, when using data from observational studies, PIRRT was associated with lower mortality compared with CRRT (RR: 0.86; 95% CI: 0.74–1.00; P = 0.05). In RCTs and observational studies, there were no significant differences in recovery of kidney function, fluid removal, days in the ICU, and biochemical efficacy. The meta-analysis concluded that PIRRT was associated with similar outcomes to CRRT.

er mortality compared with CRRT (RR: 0.86; 95% CI: 0.74–1.00; P = 0.05). In RCTs and observational studies, there were no significant differences in recovery of kidney function, fluid removal, days in the ICU, and biochemical efficacy. The meta-analysis concluded that PIRRT was associated with similar outcomes to CRRT. Some studies suggested that the choice of initial RRT modality might affect renal recovery and risk of dialysis dependence after AKI, which has implications for patients and families, as well as healthcare systems, in terms of survival, quality of life, and financial costs.42, 43, 44, 45, 46 A meta-analysis included 7 RCTs and 16 observational studies and showed that based on pooled analysis of data from observational studies, dialysis dependence was higher among survivors who initially received IHD versus CRRT (RR: 1.99; 95% CI: 1.53–2.59).46 However, analysis of the RCTs only demonstrated no difference in dialysis dependence among survivors (RR: 1.15; 95% CI: 0.78–1.68).

wed that based on pooled analysis of data from observational studies, dialysis dependence was higher among survivors who initially received IHD versus CRRT (RR: 1.99; 95% CI: 1.53–2.59).46 However, analysis of the RCTs only demonstrated no difference in dialysis dependence among survivors (RR: 1.15; 95% CI: 0.78–1.68). Experience with PD in AKI is limited, except in the pediatric setting and in regions with limited resources.47, 48, 49, 50, 51, 52, 53, 54 Gravity-driven PD is particularly attractive because it provides RRT without the need for machines and electricity. In most countries, PD is underused despite advantages such as lower costs (as little as US $150 to save 1 life).51 Technical advances (i.e., flexible and cuffed catheters, automatic cycling, and high and continuous flow PD) have made PD an acceptable alternative to other forms of acute RRT.48 The International Society for Peritoneal Dialysis (ISPD) firmly recommends that PD is a suitable modality for patients with AKI, especially in developing countries.55 Recent reports have confirmed a fall in mortality and complication rates in units where acute PD is performed regularly.47, 48, 49, 50, 51, 52, 53 IHD is the preferred treatment in situations in which immediate removal of small solutes is required, such as severe hyperkalemia, poisoning, and tumor lysis syndrome. IHD and PIRRT have a particular role in situations in which rehabilitation and mobilization are priorities, and fluid and metabolic fluctuations can be tolerated.4

preferred treatment in situations in which immediate removal of small solutes is required, such as severe hyperkalemia, poisoning, and tumor lysis syndrome. IHD and PIRRT have a particular role in situations in which rehabilitation and mobilization are priorities, and fluid and metabolic fluctuations can be tolerated.4 Continuous types of RRT are recommended for patients who may not tolerate rapid shifts in fluid balance, including those with severe hemodynamic instability.4, 16 However, PIRRT might also have a role in this situation, in particular because there was no significant difference in mortality, hemodynamic stability, and solute clearance in studies that compared PIRRT with CRRT.37, 38, 39, 40, 41 Intracranial hypertension and/or acute brain injury are specific situations in which CRRT or PD are preferred over IHD.16 The KDIGO guideline cited observational studies that reported increases in intracranial pressure with IHD56 and increases in brain water content after IHD, whereas such changes were not observed after CRRT.57 Since then, further case reports raised concerns about the potential risk of brain herniation due to rapid changes in osmolytes, falls in cerebral oxygen saturation, and negative effects on cerebrovascular autoregulation with IHD, all of which support the current KDIGO recommendation.58, 59, 60, 61 Conditions associated with extremely high catabolism should be treated with CRRT, IHD, or PIRRT rather than PD.47, 48, 62

Intracranial hypertension and/or acute brain injury are specific situations in which CRRT or PD are preferred over IHD.16 The KDIGO guideline cited observational studies that reported increases in intracranial pressure with IHD56 and increases in brain water content after IHD, whereas such changes were not observed after CRRT.57 Since then, further case reports raised concerns about the potential risk of brain herniation due to rapid changes in osmolytes, falls in cerebral oxygen saturation, and negative effects on cerebrovascular autoregulation with IHD, all of which support the current KDIGO recommendation.58, 59, 60, 61 Conditions associated with extremely high catabolism should be treated with CRRT, IHD, or PIRRT rather than PD.47, 48, 62 In children, the choice of the initial RRT modality is predominantly based on patient age, underlying illness and clinical status, expertise and experience with the modality, and cost of therapy. Although recent CRRT technology has been developed for neonates and small infants,63 CRRT is rarely available in developing regions with limited resources. Instead, PD is the first choice in most countries. It is relatively inexpensive and easy to initiate and monitor, especially in infants and in children younger than 3 years old. Stylet-based rigid catheters are still commonly used for the first session of PD, although the use of soft catheters has increased. Older children, especially those with severe metabolic complications or fluid overload, are best managed by IHD.

nd monitor, especially in infants and in children younger than 3 years old. Stylet-based rigid catheters are still commonly used for the first session of PD, although the use of soft catheters has increased. Older children, especially those with severe metabolic complications or fluid overload, are best managed by IHD. Recommendations for Clinical Practice All RRT modalities have particular advantages and offer clinicians options to manage patients and optimize care. Based on the existing evidence, the choice of RRT modality should be based on the clinical status of the patient (hemodynamic stability, catabolic state, need for removal of large amounts of fluid, presence of life-threatening complications, or acute brain injury), availability of modalities, clinical experience, and financial cost of therapy (Figure 3). For young children (younger than 5 years), PD is often the first choice because of its availability and the ease of initiation. Transition of modality should be considered when the option exists, and adequate infrastructure and trained personnel are available.Figure 3 Clinical scenarios for choosing renal replacement therapy techniques. CRRT, continuous renal replacement therapy; IHD, intermittent hemodialysis; PD, peritoneal dialysis; PIRRT, prolonged intermittent renal replacement therapy. Research Recommendations • To assess the safety of different RRT modalities in developing countries. • To perform cost-effectiveness studies of acute RRT.

Recommendations for Clinical Practice All RRT modalities have particular advantages and offer clinicians options to manage patients and optimize care. Based on the existing evidence, the choice of RRT modality should be based on the clinical status of the patient (hemodynamic stability, catabolic state, need for removal of large amounts of fluid, presence of life-threatening complications, or acute brain injury), availability of modalities, clinical experience, and financial cost of therapy (Figure 3). For young children (younger than 5 years), PD is often the first choice because of its availability and the ease of initiation. Transition of modality should be considered when the option exists, and adequate infrastructure and trained personnel are available.Figure 3 Clinical scenarios for choosing renal replacement therapy techniques. CRRT, continuous renal replacement therapy; IHD, intermittent hemodialysis; PD, peritoneal dialysis; PIRRT, prolonged intermittent renal replacement therapy. Research Recommendations • To assess the safety of different RRT modalities in developing countries. • To perform cost-effectiveness studies of acute RRT. Q3c: What Is the Most Appropriate Prescription for Acute RRT? Peritoneal Dialysis: Access Consensus Statements 3c.1. We recommend that flexible PD catheters should be used for acute PD where resources and expertise exist. We suggest that alternatives such as rigid stylet catheters be used if flexible catheters are unavailable.

t Is the Most Appropriate Prescription for Acute RRT? Peritoneal Dialysis: Access Consensus Statements 3c.1. We recommend that flexible PD catheters should be used for acute PD where resources and expertise exist. We suggest that alternatives such as rigid stylet catheters be used if flexible catheters are unavailable. 3c.2. We recommend that healthcare professionals receive training to insert these catheters to ensure timely dialysis in the emergency setting. 3c.3. We recommend the use of prophylactic antibiotics before PD catheter insertion. Context It should be recognized that the volumes of fluid used in acute PD are significantly greater than those in the chronic setting. Thus, flow rates need to be high, and the catheters need to tolerate them. Time spent filling and draining is effectively lost dwell time. As such, time for effective clearance can be seriously influenced by the performance of the catheter, especially if cycle time is short. For this reason, flexible PD catheters (e.g., Tenckhoff catheters) have an advantage because they have a larger lumen and side holes compared with rigid catheters.

ly lost dwell time. As such, time for effective clearance can be seriously influenced by the performance of the catheter, especially if cycle time is short. For this reason, flexible PD catheters (e.g., Tenckhoff catheters) have an advantage because they have a larger lumen and side holes compared with rigid catheters. Nonflexible catheters are still frequently used and can be life-saving. The rigid stylet catheter, which is introduced with the aid of a trocar through a skin incision sub-umbilically, is the most widely used nontunneled catheter. However, it has major drawbacks. First, it is produced from rigid nylon, and injury to the visceral organs may occur during insertion or later. Second, it may become obstructed with fibrin and therefore needs to be flushed regularly. The break in the sterile circuit necessary to perform regular flushes may explain the higher rates of peritonitis.64

s. First, it is produced from rigid nylon, and injury to the visceral organs may occur during insertion or later. Second, it may become obstructed with fibrin and therefore needs to be flushed regularly. The break in the sterile circuit necessary to perform regular flushes may explain the higher rates of peritonitis.64 Knowledge of the key aspects related to PD has improved significantly because of initiatives like the Saving Young Lives Campaign and industry-sponsored training sessions, but the task of teaching and training is enormous. The ISPD guidelines strongly recommend that flexible PD catheters are inserted by nephrologists at the bedside to avoid delays in initiating treatment.55 In the early stages of PD, there is a risk of leakage of peritoneal fluid. To prevent this, the most important step is to keep the patient at bed rest while the abdomen is full. Tunneled catheters inserted by the Seldinger technique have been shown to have a lower risk of leaking compared with surgically placed catheters.64 Videos demonstrating different techniques of catheter placement are available.65, 66 Peritoneal Dialysis: Fluid Delivery Consensus Statements 3c.4. We recommend that, when possible, a closed system be used (Y connection). 3c.5. We suggest that automatic and manual PD should be considered as equivalent.

Knowledge of the key aspects related to PD has improved significantly because of initiatives like the Saving Young Lives Campaign and industry-sponsored training sessions, but the task of teaching and training is enormous. The ISPD guidelines strongly recommend that flexible PD catheters are inserted by nephrologists at the bedside to avoid delays in initiating treatment.55 In the early stages of PD, there is a risk of leakage of peritoneal fluid. To prevent this, the most important step is to keep the patient at bed rest while the abdomen is full. Tunneled catheters inserted by the Seldinger technique have been shown to have a lower risk of leaking compared with surgically placed catheters.64 Videos demonstrating different techniques of catheter placement are available.65, 66 Peritoneal Dialysis: Fluid Delivery Consensus Statements 3c.4. We recommend that, when possible, a closed system be used (Y connection). 3c.5. We suggest that automatic and manual PD should be considered as equivalent. Context The technique of fluid delivery in acute PD can increase the risk of peritonitis because there are significantly more connections and disconnections compared with the 3 to 4 exchanges in chronic continuous ambulatory PD. In low-resource environments, makeshift and proprietary circuits are used, with each bag of dialysate being attached using a spike. By gravity, the fluid flows into the peritoneal cavity through a 3-way tap. Drainage into a bag or bucket also occurs through gravity. Although this is an effective and inexpensive system, there is a risk of contamination every time the bag is spiked. In addition, the open drainage system risks retrograde travel of bacteria into the peritoneum. Disconnecting systems with a Y-set and double bag, as used in chronic PD, are associated with lower peritonitis rates.48, 67, 68, 69, 70 There is no reason to suspect this would not also apply to acute PD.

the bag is spiked. In addition, the open drainage system risks retrograde travel of bacteria into the peritoneum. Disconnecting systems with a Y-set and double bag, as used in chronic PD, are associated with lower peritonitis rates.48, 67, 68, 69, 70 There is no reason to suspect this would not also apply to acute PD. Automated cycler PD uses a mechanized device to deliver and drain the dialysate. It can be set up by a trained staff member once per day, which reduces the risk of complications, including contamination. Nursing time is also reduced because all cycles occur automatically. There are conflicting reports related to the incidence of peritonitis with cyclers, but there appears to be no difference compared with the manual system used in chronic PD. Cyclers also offer tidal PD in which a small volume of fluid is left in the abdomen at all times, which may reduce mechanical complications and discomfort. Tidal PD also has the theoretical benefit of increased solute clearance because fluid continuously dwells in the peritoneal space, including during the fill and drain portion of the cycle. Automated cyclers have been used extensively for PD in AKI, but they may prove to be too expensive in low-resource settings. Peritoneal Dialysis: Solutions Consensus Statements 3c.6. We suggest that patients with shock or liver failure should be treated with bicarbonate-containing solutions. When these solutions are not available, the use of lactate-containing solutions is an alternative.

Automated cycler PD uses a mechanized device to deliver and drain the dialysate. It can be set up by a trained staff member once per day, which reduces the risk of complications, including contamination. Nursing time is also reduced because all cycles occur automatically. There are conflicting reports related to the incidence of peritonitis with cyclers, but there appears to be no difference compared with the manual system used in chronic PD. Cyclers also offer tidal PD in which a small volume of fluid is left in the abdomen at all times, which may reduce mechanical complications and discomfort. Tidal PD also has the theoretical benefit of increased solute clearance because fluid continuously dwells in the peritoneal space, including during the fill and drain portion of the cycle. Automated cyclers have been used extensively for PD in AKI, but they may prove to be too expensive in low-resource settings. Peritoneal Dialysis: Solutions Consensus Statements 3c.6. We suggest that patients with shock or liver failure should be treated with bicarbonate-containing solutions. When these solutions are not available, the use of lactate-containing solutions is an alternative. 3c.7. We suggest that commercially prepared solutions should be used. However, when resources do not permit this, custom-prepared fluids may be life-saving. 3c.8. Once serum potassium level falls to <4 mmol/L, potassium should be added to dialysate using a sterile technique.

Peritoneal Dialysis: Solutions Consensus Statements 3c.6. We suggest that patients with shock or liver failure should be treated with bicarbonate-containing solutions. When these solutions are not available, the use of lactate-containing solutions is an alternative. 3c.7. We suggest that commercially prepared solutions should be used. However, when resources do not permit this, custom-prepared fluids may be life-saving. 3c.8. Once serum potassium level falls to <4 mmol/L, potassium should be added to dialysate using a sterile technique. Context The ISPD guidelines recommend the use of commercially produced PD solutions.55 Although dialysate solutions are manufactured in a number of developing countries, their availability continues to be difficult in many regions of the world. Because they are too heavy to be delivered by air, they often need to pass through several countries before they reach their destination. As a result, a number of PD units produce their own solutions using a mixture of modified Ringer’s lactate and glucose, both of which are readily available in most hospitals. The potential risks are contamination and infection.

air, they often need to pass through several countries before they reach their destination. As a result, a number of PD units produce their own solutions using a mixture of modified Ringer’s lactate and glucose, both of which are readily available in most hospitals. The potential risks are contamination and infection. There has been much interest in the composition of dialysate or replacement fluid used for RRT in critically ill patients, in particular because patients with shock or liver failure may not be able to convert lactate to bicarbonate. In RCTs that compare lactate-based replacement fluids versus bicarbonate-based replacement fluids for CRRT, patients randomized to bicarbonate-buffered solutions had more rapid correction of acidosis and less cardiovascular instability.71 In PD, the evidence is limited to 1 small RCT that also showed that acidosis in patients with shock or liver failure was corrected significantly faster if bicarbonate-containing solutions were used rather than lactate-based fluids.72 The ISPD guidelines advocate bicarbonate-containing solutions for those with shock or liver failure, but not for other patient groups.55

also showed that acidosis in patients with shock or liver failure was corrected significantly faster if bicarbonate-containing solutions were used rather than lactate-based fluids.72 The ISPD guidelines advocate bicarbonate-containing solutions for those with shock or liver failure, but not for other patient groups.55 Standard PD solutions do not contain any potassium. As a result, a significant number of chronic PD patients develop hypokalemia (potassium <3.5 mmol/L) or require potassium supplementation, especially because hypokalemia is a risk factor for peritonitis and death in chronic PD patients.73, 74 In acute PD, potassium loss can be particularly high because each 2-L exchange has the potential to remove up to 2 times the serum potassium concentration. Such rapid potassium loss can be prevented or corrected by adding potassium to the dialysis solution (4 mmol/L).75 Ponce et al. and Gabriel et al.75, 76, 77, 78 demonstrated that control of serum potassium was obtained after a 1-day session of high-volume PD. In case serum potassium fell to <4 mmol/L, potassium 3.5 to 5 mmol/L was added to the dialysis solutions. Strict adherence to an aseptic technique and attention to detail are important when adding fluids or drugs to the dialysis solution.53 The process should be undertaken in a clean environment using a minimum number of punctures and involve the least number of steps to reduce the risk of infection and error. The fluids should be used immediately.

ptic technique and attention to detail are important when adding fluids or drugs to the dialysis solution.53 The process should be undertaken in a clean environment using a minimum number of punctures and involve the least number of steps to reduce the risk of infection and error. The fluids should be used immediately. Peritoneal Dialysis: Prescription Consensus Statements 3c.9. We recommend continuous PD until metabolic and fluid control are achieved. 3c.10. We suggest targeting a minimal weekly Kt/V urea of 2.1 in noncritically ill patients. 3c.11. We suggest targeting a weekly Kt/V urea of 3.5 in critically ill patients. 3c.12. We suggest prescribing 1 to 2 L of dialysate per cycle and 24 to 36 L per session with 1 session lasting 24 hours. For a 70-kg patient, the minimal volume prescribed would be 24 L per session. 3c.13. To correct fluid overload, we suggest raising the concentration of dextrose and/or shortening the cycle duration. When the patient is euvolemic, the dextrose concentration and cycle time should be adjusted to ensure a neutral fluid balance. 3c.14. We suggest measuring effluent concentrations to determine delivered Kt/V urea at least once a week.

3c.13. To correct fluid overload, we suggest raising the concentration of dextrose and/or shortening the cycle duration. When the patient is euvolemic, the dextrose concentration and cycle time should be adjusted to ensure a neutral fluid balance. 3c.14. We suggest measuring effluent concentrations to determine delivered Kt/V urea at least once a week. Context PD has been shown to provide comparable outcomes to IHD in appropriately selected patients, but several areas of uncertainty remain. The dose and/or efficacy of PD can be assessed by measuring urea clearance over time as Kt/V urea where: K = volume of dialysate drained multiplied by dialysate/plasma urea concentration; t = duration of dialysis; and V = volume of distribution of urea (total body water ∼ 0.5 [female] or 0.6 [male] multiplied by body weight).

e and/or efficacy of PD can be assessed by measuring urea clearance over time as Kt/V urea where: K = volume of dialysate drained multiplied by dialysate/plasma urea concentration; t = duration of dialysis; and V = volume of distribution of urea (total body water ∼ 0.5 [female] or 0.6 [male] multiplied by body weight). The most appropriate dose of PD for patients with AKI is unknown, mainly due to a limited number of trials, the existence of methodological flaws in some studies, and the fact that the doses of dialysis used varied widely. In the most thorough study by Ponce et al., acute PD using a cuffed catheter (36–44 L per session, 18 to 22 cycles, 2 L per cycle, weekly delivered Kt/V urea of 3.6) was compared with daily HD.77 Clinical outcomes were comparable. Other studies have also shown good outcomes with much lower doses (16–24 L per session, 8–16 cycles, 1–2 L per cycle).77, 78 However, because these studies were nonrandomized, the problem of a positive reporting bias needs to be kept in mind. Ponce et al. followed up their initial report with a study that compared high volume with lower volume acute PD and showed no clinical benefit with higher volumes; the lower dose group achieved a weekly Kt/V urea of 3.43 and did as well as the higher dose group with a Kt/V of 4.13.77

ting bias needs to be kept in mind. Ponce et al. followed up their initial report with a study that compared high volume with lower volume acute PD and showed no clinical benefit with higher volumes; the lower dose group achieved a weekly Kt/V urea of 3.43 and did as well as the higher dose group with a Kt/V of 4.13.77 By inference from data from extracorporeal blood therapies, it has been suggested that a targeted PD dose of a weekly Kt/V urea of 2.1 may represent a reasonable goal as the minimum dose, but the optimal dose for an individual patient remains unknown.62, 79, 80, 81 It is certainly possible that higher small-solute clearance is necessary for patients with more complex catabolic illnesses.62 It also remains uncertain whether removal of small molecules (e.g., urea, creatinine) or larger molecules (e.g., cytokines, soluble receptors) is more important. According to the ISPD Guideline PD for AKI, a weekly Kt/V urea target of 3.5 provides outcomes comparable to that of daily HD.55 Higher doses are not associated with better outcomes. For noncritically ill AKI patients, a weekly Kt/V target of 2.1 may be acceptable, but this suggestion is not evidence-based.81

mportant. According to the ISPD Guideline PD for AKI, a weekly Kt/V urea target of 3.5 provides outcomes comparable to that of daily HD.55 Higher doses are not associated with better outcomes. For noncritically ill AKI patients, a weekly Kt/V target of 2.1 may be acceptable, but this suggestion is not evidence-based.81 Much attention has focused on solute clearances, but there is increasing evidence that fluid overload is also harmful and should be avoided or corrected. In principle, regular assessment of volume status and the prescription of clear ultrafiltration and fluid balance targets are necessary for all patients receiving RRT, including PD. Relatively large amounts of fluid can be removed by PD (i.e., up to 1 L in 4 hours when using a 4.25% PD solution). Although this may cause hyperglycemia, the risks of hypertonic solutions are negligible in the short term.

afiltration and fluid balance targets are necessary for all patients receiving RRT, including PD. Relatively large amounts of fluid can be removed by PD (i.e., up to 1 L in 4 hours when using a 4.25% PD solution). Although this may cause hyperglycemia, the risks of hypertonic solutions are negligible in the short term. For children treated with PD, not enough information is available regarding dosing. However, it has been suggested that the Kt/V target should exceed that of the adult standards because daily protein intake per kilogram is higher in children.82 Exchange volumes of 20 to 30 ml/kg have been traditionally applied. In infants, the peritoneal surface area per unit body weight is twice that of adults, whereas the relationship between body surface area and peritoneal membrane surface area is constant and age-independent. Therefore, an exchange volume of 1,100 ml/m2 of body surface area (equivalent to 2000 ml/1.73 m2) might be a better suggestion.83 If possible, intra-abdominal pressure should be measured to detect the fill volume limit, which has been reported up to 1400 ml/m2, leading to an intra-abdominal pressure of 18 cm H2O.84 Recommendations for Clinical Practice • During the initial 24 hours of acute PD, the duration of the cycle time needs to be determined based on the clinical circumstances. Short cycle times (every 1–2 hours) may be necessary in the first 24 to 48 hours to correct hyperkalemia, fluid overload, and/or metabolic acidosis. Thereafter, the cycle time may be increased to 4 to 6 hours depending on the clinical circumstances.

cycle time needs to be determined based on the clinical circumstances. Short cycle times (every 1–2 hours) may be necessary in the first 24 to 48 hours to correct hyperkalemia, fluid overload, and/or metabolic acidosis. Thereafter, the cycle time may be increased to 4 to 6 hours depending on the clinical circumstances. • To treat or avoid fluid overload, ultrafiltration can be increased by raising the concentration of dextrose and/or shortening the cycle duration. When the patient is euvolemic, the dextrose concentration and cycle time should be adjusted to ensure a neutral fluid balance. Research Recommendations • To focus on comparing higher intensity PD versus lower intensity PD in lower demand/capacity settings. • To perform risk−benefit analyses of more frequent versus less frequent monitoring. • To identify clinical parameters to guide ultrafiltration. • To identify the most appropriate fluid delivery method and solutions for acute PD. Extracorporeal Renal Replacement Therapies Consensus Statements 3c.15. We recommend ultrasound guidance for vascular catheter placement. If not available, blinded puncture is acceptable. 3c.16. We recommend using either the right jugular or right femoral site as the first option for vascular access in non-obese patients. 3c.17. We suggest using bicarbonate-based solutions. 3c.18. We recommend implementing water quality measurements and providing adequate equipment, including either commercially available fluids or reverse osmosis. 3c.19. For IHD or PIRRT, we suggest a minimum urea reduction ratio of 60% or Kt/V urea of 1.2 per treatment.

3c.17. We suggest using bicarbonate-based solutions. 3c.18. We recommend implementing water quality measurements and providing adequate equipment, including either commercially available fluids or reverse osmosis. 3c.19. For IHD or PIRRT, we suggest a minimum urea reduction ratio of 60% or Kt/V urea of 1.2 per treatment. 3c.20. For CRRT, we recommend using citrate for anticoagulation. If not available, CRRT can be delivered without anticoagulation in patients at high risk of bleeding or heparin for low-risk patients. 3c.21. For CRRT, we suggest delivering a minimum effluent volume of 20 to 25 ml/kg per hour; however, the dose should be dynamic and adapted to the metabolic demands placed on the patient. Context Good vascular access is essential for adequate delivery of extracorporeal RRT. Ultrasound guidance for catheter placement has reduced the risk of catheter placement failure (RR: 0.12; 95% CI: 0.04–0.37; P < 0.001) and arterial puncture (RR: 0.22; 95% CI: 0.06–0.81; P = 0.02), as well as the number of attempts.85 However, even in nonlimited resource settings, ultrasound machines may not always be available.

guidance for catheter placement has reduced the risk of catheter placement failure (RR: 0.12; 95% CI: 0.04–0.37; P < 0.001) and arterial puncture (RR: 0.22; 95% CI: 0.06–0.81; P = 0.02), as well as the number of attempts.85 However, even in nonlimited resource settings, ultrasound machines may not always be available. An evaluation of sites for acute temporal catheter placement showed that circuit life was comparable between jugular and femoral access (17.1 hours vs. 20.2 hours, respectively).86 A RCT that included 750 patients showed a higher incidence of hematomas with jugular access (3.6% vs. 1.1%; P = 0.03).87 There was no difference in catheter-related bloodstream infection (2.3 per 1000 catheter-days vs. 1.5 per 1000 catheter-days; P = 0.42), but there was a trend to higher colonization in patients with a body mass index >28.4 kg/m2 who were randomized to femoral access. Choosing the right length of catheter is also important because it influences blood flow, recirculation, filter survival, and ultimately, the dose of RRT.88, 89 Tunneled catheters are associated with a reduced infection risk, but they require special training. Switching to tunneled catheters should be considered when prolonged RRT is anticipated.

catheter is also important because it influences blood flow, recirculation, filter survival, and ultimately, the dose of RRT.88, 89 Tunneled catheters are associated with a reduced infection risk, but they require special training. Switching to tunneled catheters should be considered when prolonged RRT is anticipated. For catheter care, 2 meta-analyses suggested that low-dose citrate lock solutions might help to reduce the risk of catheter malfunction and catheter-related bacteremia.90 Due to the potential of systemic exposure, neither antibiotic nor high-dose heparin (5000 U) are recommended as locking solutions.91, 92, 93 Chlorhexidine-impregnated dressing seems to reduce catheter-related bacteremia, but if not available, standard polyurethane dressing is recommended over using the gauze and tape approach, no dressing, or any highly adhesive strategies.94, 95

nor high-dose heparin (5000 U) are recommended as locking solutions.91, 92, 93 Chlorhexidine-impregnated dressing seems to reduce catheter-related bacteremia, but if not available, standard polyurethane dressing is recommended over using the gauze and tape approach, no dressing, or any highly adhesive strategies.94, 95 In 2015, a Cochrane systematic review assessed the composition of dialysate and replacement solutions and compared bicarbonate-buffered solutions versus lactate-buffered solutions for CRRT.96 Analysis of 4 clinical trials that included 171 patients revealed no significant differences in mortality and acid-base or electrolyte parameters, except for higher serum lactate levels in the lactate group. Only 1 study reported fewer cardiovascular events and fewer hypotensive events in the bicarbonate group,71 whereas a different study found a higher mean arterial pressure using bicarbonate.97 Therefore, we consider bicarbonate-based fluids to be the first option; if not available, lactate solutions are acceptable for extracorporeal RRT, as long it is recognized that they may cause a rise in serum lactate in patients with liver failure.

fferent study found a higher mean arterial pressure using bicarbonate.97 Therefore, we consider bicarbonate-based fluids to be the first option; if not available, lactate solutions are acceptable for extracorporeal RRT, as long it is recognized that they may cause a rise in serum lactate in patients with liver failure. Anticoagulation is a potential limitation of extracorporeal RRT. Regional anticoagulation with citrate has emerged as an effective method to maintain circuit patency.16 A recent meta-analysis that included 14 RCTs and 1134 patients showed significantly longer circuit life with citrate, with a mean difference of 15.69 hours (range: 9.3–22.08 hours) and a significantly lower bleeding risk (RR: 0.31; 95% CI: 0.19–0.51) compared with heparin.98 A systematic review of observational studies in children found a circuit survival of almost 70% at 60 hours and a decreased risk of bleeding with citrate.99 However, metabolic alkalosis was common, affecting 20% to 100% of patients. Citrate accumulation was reported in only 1 study and reported in 20% of children.99 If citrate is not available, extracorporeal RRT without anticoagulation is feasible in patients at high risk of bleeding, especially when using IHD. For PIRRT, filter loss due to clotting has been reported in approximately 25% of sessions beyond 6 hours.100 Other methods of keeping the circuit patent include using predilution fluid and keeping a filtration fraction of <20%.101, 102 Finally, based on a meta-analysis, filter life may be better with hemodiafiltration compared with hemofiltration.103

otting has been reported in approximately 25% of sessions beyond 6 hours.100 Other methods of keeping the circuit patent include using predilution fluid and keeping a filtration fraction of <20%.101, 102 Finally, based on a meta-analysis, filter life may be better with hemodiafiltration compared with hemofiltration.103 Increasing the dose of extracorporeal RRT in AKI patients has not been associated with improved survival.104, 105 A study that compared daily versus alternate day intermittent RRT showed a reduction in mortality with daily treatment together with better control of uremia, fewer hypotensive episodes, and more rapid resolution of AKI.106 Although these results have not been replicated, it seems reasonable that more frequent intermittent support would allow better control of fluid balance, regardless of small-solute clearance. With regards to CRRT, high volume hemofiltration (>50 ml/kg per hour) has not been associated with a survival benefit but with a higher risk of inadvertent nutrient and antibiotic loss.104, 107 No study has demonstrated a substantial benefit with CRRT doses in the range of 20 to 50 ml/kg per hour, but not enough information is available to recommend a dose <20 ml/kg per hour.108 Although Jiang et al. randomized patients with pancreatitis to 1000 ml/h (approximately equivalent to 14 ml/kg per hour) and reported a higher mortality in the lower dose group,109 a retrospective analysis by Uchino et al. found no difference in mortality between patients who received 14.3 ml/kg per hour versus 20 to 25 ml/kg per hour.110 Based on the existing data and in support of the KDIGO recommendations,16 the ADQI consensus recommendation suggests a target dose between 20 and 25 ml/kg per hour, recognizing that the dose may need to be increased or reduced to meet changes in demand or capacity.3 Importantly, reuse of filters confers a risk of reduced dose delivery, but variations exist due to different reuse techniques.111, 112

I consensus recommendation suggests a target dose between 20 and 25 ml/kg per hour, recognizing that the dose may need to be increased or reduced to meet changes in demand or capacity.3 Importantly, reuse of filters confers a risk of reduced dose delivery, but variations exist due to different reuse techniques.111, 112 No evidence-based recommendations can be made for children receiving CRRT. Doses between 2000 and 3000 ml/h per 1.73 m2 have been used.113, 114 Research Recommendations • To design a RCT comparing a low dose versus a standard dose of extracorporeal RRT in low-income countries. • To evaluate different ways of dosing beyond classical small-solute clearance. • To conduct an RCT to evaluate citrate dosing for specific populations (pediatrics, patients with liver failure, and patients with hypoperfusion). • To test surrogate outcomes for different prescriptions based on filtration fraction. • To compare different dose regimens for extracorporeal RRT in children. Q3d. How Should RRT Be Monitored? Consensus Statements 3d.1. We recommend that standardized protocols for prescribing and delivering RRT are developed, adopted, and continuously reviewed. 3d.2. We recommend the standardized documentation of RRT treatments received by the patient. 3d.3. We suggest that patient-related parameters such as hemodynamics, volume status, temperature, and nutrition be monitored during RRT. 3d.4. We suggest monitoring of delivered dose of RRT at least once a week. 3d.5. We suggest reassessing the delivered RRT dose when significant changes in prescription or in the clinical status of patients occur.

3d.3. We suggest that patient-related parameters such as hemodynamics, volume status, temperature, and nutrition be monitored during RRT. 3d.4. We suggest monitoring of delivered dose of RRT at least once a week. 3d.5. We suggest reassessing the delivered RRT dose when significant changes in prescription or in the clinical status of patients occur. Context Recent reports indicate that the care received by patients with AKI and RRT for AKI is suboptimal even in developed countries, which may be due to variations in the practices of RRT worldwide and a general lack of consensus.33, 115, 116 Standardized protocols of RRT help to improve the delivery, quality, and safety of RRT. Centers that provide RRT for AKI patients should develop protocols for the initiation, monitoring, and termination of RRT according to their local needs based on the patient case mix, economics, and resource availability. In developing countries, protocols for RRT should also take into consideration the availability of resources and equipment, as well as financial costs. Protocols should be reviewed periodically to identify any deficiencies and improve clinical care. Clear roles and lines of responsibility should be set. The documentation should be standardized and include patient specific information and data related to the machine, extracorporeal circuit, fluids used, anticoagulation, and complications and interventions performed during dialysis.

encies and improve clinical care. Clear roles and lines of responsibility should be set. The documentation should be standardized and include patient specific information and data related to the machine, extracorporeal circuit, fluids used, anticoagulation, and complications and interventions performed during dialysis. Hypotension is a common complication of RRT and may contribute to morbidity and delay in recovery from AKI. Akhoundi et al. reported that 43% of patients on CRRT developed hypotension within 1 hour of initiation.117 The Acute Renal Failure Trial Network (ATN) study compared high intensive RRT versus less intensive RRT in critically ill patients with AKI and reported significantly more hypotension in the high intensity group (14.4% vs. 10%; P = 0.02).104 Gaudry et al. compared early and late initiation of RRT and reported serious cardiac arrhythmias in 3.2% and moderately serious arrhythmias in 15.7% of patients.17

less intensive RRT in critically ill patients with AKI and reported significantly more hypotension in the high intensity group (14.4% vs. 10%; P = 0.02).104 Gaudry et al. compared early and late initiation of RRT and reported serious cardiac arrhythmias in 3.2% and moderately serious arrhythmias in 15.7% of patients.17 Fluid overload is common in critically ill patients, especially in those with AKI who receive RRT. The importance of monitoring fluid balance has been recognized, especially because fluid overload of >10% of body weight and prolonged duration of fluid accumulation has been found to be associated with an increased risk of complications and mortality.118, 119, 120, 121 A recent multicenter study also showed that the speed of fluid accumulation was independently associated with ICU mortality.121 Rate of fluid removal during RRT depends on the degree of fluid overload and hemodynamic stability. Several methods have been proposed to guide fluid removal, including chest radiography, measurement of natriuretic peptides, bioimpedance analysis, thoracic ultrasound, and ultrasonic measurement of the vena cava. Other factors should also be monitored during RRT, including body temperature and nutritional values.

ity. Several methods have been proposed to guide fluid removal, including chest radiography, measurement of natriuretic peptides, bioimpedance analysis, thoracic ultrasound, and ultrasonic measurement of the vena cava. Other factors should also be monitored during RRT, including body temperature and nutritional values. The dose of RRT delivered to the patient needs to be monitored, especially because it is often 15% to 30% lower than the prescription.122, 123, 124 Hence, it is recommended to prescribe a dose that is 25% higher than the required dose of 20 ml/kg per minute.16 The discrepancy between prescribed and delivered solute clearance is due to 2 main reasons: (i) down time effect, in which CRRT is provided for <24 h/d; and (ii) progressive reduction in efficiency of the filter over time due to clogging of the hollow fibers. The down time effect is common and usually due to circuit clotting, poor vascular access, and patient-related factors, such as need for investigations and procedures.125 Monitoring of the down time effect and periodic measurement of solute clearance should be part of routine monitoring in CRRT.

ng of the hollow fibers. The down time effect is common and usually due to circuit clotting, poor vascular access, and patient-related factors, such as need for investigations and procedures.125 Monitoring of the down time effect and periodic measurement of solute clearance should be part of routine monitoring in CRRT. We recommend that the delivered dose of RRT should be measured at least once a week and every time after significant changes in prescriptions or in the clinical status of patients have occurred, to ensure that the changing demands are met by the therapy. Several methods have been proposed to measure solute clearance during CRRT. Claure-de Granado et al. determined solute clearance from the blood-side and dialysate-side kinetics.126 They recommended using dialysate-side measurements in CRRT (in milliliters per minute) and blood-side kinetics for clearance measurement in IHD and hybrid therapies in Kt/V urea or equivalent renal urea clearance. Clearance of middle molecules is not generally measured. Consensus Statement 3d.6. We suggest that the frequency of blood tests should be based on the clinical state of the patient.

We recommend that the delivered dose of RRT should be measured at least once a week and every time after significant changes in prescriptions or in the clinical status of patients have occurred, to ensure that the changing demands are met by the therapy. Several methods have been proposed to measure solute clearance during CRRT. Claure-de Granado et al. determined solute clearance from the blood-side and dialysate-side kinetics.126 They recommended using dialysate-side measurements in CRRT (in milliliters per minute) and blood-side kinetics for clearance measurement in IHD and hybrid therapies in Kt/V urea or equivalent renal urea clearance. Clearance of middle molecules is not generally measured. Consensus Statement 3d.6. We suggest that the frequency of blood tests should be based on the clinical state of the patient. Context Electrolyte and acid-base abnormalities are common in patients with AKI. IHD allows more rapid correction of life-threatening abnormalities, whereas CRRT takes longer.127 Hypophosphatemia and hypokalaemia are commonly observed on CRRT. The renal replacement therapy study RENAL reported a 59.5% incidence of hypophosphatemia in patients who underwent CRRT.105 Hypocalemia was observed in 22% to 63% of patients on CRRT, commonly in the context of citrate-based anticoagulation.117 The frequency of laboratory tests depends mainly on the condition of the patient, but other factors (e.g., availability and cost) play an important role. More serious derangements require more frequent monitoring.

s observed in 22% to 63% of patients on CRRT, commonly in the context of citrate-based anticoagulation.117 The frequency of laboratory tests depends mainly on the condition of the patient, but other factors (e.g., availability and cost) play an important role. More serious derangements require more frequent monitoring. Monitoring of anticoagulation is important during RRT. However, the relationship between heparin dose, activated prothrombin time (APTT), filter survival, and bleeding complications is not straightforward.128 Hence, monitoring of APTT during heparin anticoagulation during RRT should be individualized. Routine monitoring of APTT is not essential during IHD and PIRRT when heparin anticoagulation is used, but measurement of APTT should be considered in case of premature filter clotting or bleeding complications. When using heparin anticoagulation for CRRT, APTT may be measured at 6- to 8-hour intervals during the first 24 hours and subsequently at least twice daily. Monitoring of citrate anticoagulation in CRRT is more complex and requires more frequent monitoring of serum electrolytes, ionized calcium and serum calcium, and arterial blood gases.129 We suggest that ionized calcium be measured at 6- to 8-hour intervals, and the total calcium to ionized calcium ratio and ABG be measured at least once a day to monitor efficacy and safety of citrate-based therapy. Consensus Statement 3d.7. For patients who receive RRT, we suggest monitoring of drug levels when possible and to adjust drug doses accordingly.

Monitoring of anticoagulation is important during RRT. However, the relationship between heparin dose, activated prothrombin time (APTT), filter survival, and bleeding complications is not straightforward.128 Hence, monitoring of APTT during heparin anticoagulation during RRT should be individualized. Routine monitoring of APTT is not essential during IHD and PIRRT when heparin anticoagulation is used, but measurement of APTT should be considered in case of premature filter clotting or bleeding complications. When using heparin anticoagulation for CRRT, APTT may be measured at 6- to 8-hour intervals during the first 24 hours and subsequently at least twice daily. Monitoring of citrate anticoagulation in CRRT is more complex and requires more frequent monitoring of serum electrolytes, ionized calcium and serum calcium, and arterial blood gases.129 We suggest that ionized calcium be measured at 6- to 8-hour intervals, and the total calcium to ionized calcium ratio and ABG be measured at least once a day to monitor efficacy and safety of citrate-based therapy. Consensus Statement 3d.7. For patients who receive RRT, we suggest monitoring of drug levels when possible and to adjust drug doses accordingly. Context Appropriate delivery of drugs, especially antibiotics, is crucial. There is a large gap in our understanding of the pharmacokinetics and pharmacodynamics of many drugs in patients with AKI and multi-organ failure. As a result, data to guide drug dosing in patients who receive RRT are limited, and patients are at risk of both drug underdosing and overdosing.130 The removal of drugs during RRT depends on several factors, such as blood concentration, sieving coefficient, degree of protein binding, dialysis dose, and duration. Therapeutic drug monitoring by prospective measurement of serum drug concentrations should be used whenever possible, but the necessary laboratory assays are rarely available and generally expensive.

veral factors, such as blood concentration, sieving coefficient, degree of protein binding, dialysis dose, and duration. Therapeutic drug monitoring by prospective measurement of serum drug concentrations should be used whenever possible, but the necessary laboratory assays are rarely available and generally expensive. Guidance for drug dosing during RRT is available (e.g., British National Formulary, Martindale: The Complete Drug Reference, and American Hospital Drug Information). However, they vary in the source of information and recommendations.131 Table 3 lists the recommended doses of common drugs during RRT, as compiled from several sources.132, 133, 134Table 3 Recommendations for dose adjustment of common drugs during renal replacement therapy

nce, and American Hospital Drug Information). However, they vary in the source of information and recommendations.131 Table 3 lists the recommended doses of common drugs during RRT, as compiled from several sources.132, 133, 134Table 3 Recommendations for dose adjustment of common drugs during renal replacement therapy Antimicrobial drug Normal dose IHD or PIRRT CRRT PD Acyclovir (IV) 5–10 mg/kg q8h 5–10 mg/kg q48h, dose after dialysis 5–7.5 mg q24h 2.5–5 mg/kg q24h Amikacin (IV)a 7.5 mg/kg q12h 7.5 mg/kg q48–72h, additional dose of 3.5 mg/kg after each HD 7.5 mg/kg q24–48h 15–20 mg/L/d Amphotericin (IV) 0.5–1.5 mg/kg/d Normal dose Normal dose Normal dose Amoxicillin (IV) 1–2 g q6h 1–2 g q12h, dose after dialysis 1–2 g q8h 250 mg q12h Cefazolin (IV) 1–2 g q8h 1–2 g q12–24h, 0.5–1 gm after dialysis 1–2 g q12h 0.5 gm q12h Ceftazidime (IV) 1–2 g q8h 1 g q24h, and 1 g post-HD 1–2 g q12h 0.5 gm q12h Cetrioxone (IV) 1–2 g q24h Normal dose, post-HD 0.75 g q12h Normal dose Cefotoxime (IV) 1–2 g q6–8h 1–2 g q12–24h, and 1 g post-HD 1 g q24h 1 g q12h Cefoperazone (IV) 1–2 g q12h Normal dose and 1 g post-HD Normal dose Normal dose Colistin (IV) 2.5 mg/kg q12h 1.5 mg/kg q36h 2.5 mg/kg q48h 1.5 mg/kg q36h Ciprofloxacin (IV) 200–400 mg q12h 200 mg q24h 200 mg q12h 200 mg q24h Fluconozole (IV) 200–800 mg q24h 100–400 mg q24h, 200 mg after dialysis 200–800 mg q24h 100–400 mg q24h Gancyclovir (IV) 5 mg/kg q12h 2.5 mg/kg q24h, dose after dialysis 2.5 mg/kg q24h 2.5 mg/kg q24h Gentamicin (IV)a 1.7 mg/kg q8h 1.7 mg/kg q24h, half the dose post-HD 1–2.5 mg/kg q24–48h 3–4 mg/L/d Imipenem/cilastatin (IV) 250–500 mg q6h 250 mg q12h, dose after dialysis 250–500 mg q8–12h 250 mg q12h Levofloxacin (IV) 500–750 mg q24h 500 mg q48h 250–750 mg q24h 250 mg q24h Meropenem (IV) 1 g q8h 0.5–1 g q24h, dose after dialysis 0.5–1 g q12h 0.5–1 g q24h Penicillin G (IV) 1–2 million U q4h 1–2 million U q8h, dose post-HD 1–2 million U q6h 1–2 million U q8h Piperacillin/tazobactam (IV) 3.375 g q6h 3.375 g q12h, dose post-HD 3.375 g q8h 3.375 g q12h Valacyclovir (PO) 1 g q8h 0.5 g q24h, dose post-HD 1 g q12–24h 0.5 g q24h Voriconozole (IV) 200 mg q12h Normal dose Normal dose Normal dose Vancomycin (IV)a 1 g q12h 1 g q48–72h 1 g q48–72h 1 g q48h CRRT, continuous renal replacement therapy; HD, hemodialysis; IHD, intermittent hemodialysis; PD, peritoneal dialysis; PIRRT, prolonged intermittent renal replacement therapy; PO, oral; q, every.

Voriconozole (IV) 200 mg q12h Normal dose Normal dose Normal dose Vancomycin (IV)a 1 g q12h 1 g q48–72h 1 g q48–72h 1 g q48h CRRT, continuous renal replacement therapy; HD, hemodialysis; IHD, intermittent hemodialysis; PD, peritoneal dialysis; PIRRT, prolonged intermittent renal replacement therapy; PO, oral; q, every. a Monitoring of serum drug levels and dose adjustment accordingly is recommended. Consensus Statement 3d.8. We suggest implementing an infection control plan. Context Dialysis catheter-related infections can contribute significantly to morbidity and cost of hospitalization. Hoste et al. reported an 8.8% incidence of bloodstream infection in patients who received acute RRT compared with 3.5% observed in non-AKI patients in the same unit.135 Sixteen percent of bloodstream infections in the dialysis population were related to the dialysis catheter. The risk is higher with femoral vein catheters.87 To reduce the risk, infection control protocols, education of staff with periodic reinforcement, periodic review of infection rates, and regular feedback are advised.136 Dialysis catheters should be removed as soon as dialysis is no longer necessary or when infections are suspected or proven.137

moral vein catheters.87 To reduce the risk, infection control protocols, education of staff with periodic reinforcement, periodic review of infection rates, and regular feedback are advised.136 Dialysis catheters should be removed as soon as dialysis is no longer necessary or when infections are suspected or proven.137 The role of antimicrobial catheter lock solutions is controversial. A recent meta-analysis that included 23 studies concluded that their use reduced the risk of catheter-related bloodstream (CRBS) infections by 69%.138 However, several guidelines do not recommend routine use of antimicrobial catheter lock solutions because of the potential risk of fungal infections, antimicrobial resistance, and systemic toxicity.16, 139, 140 We suggest considering the use of antimicrobial catheter lock solutions in specific patient groups, that is, in ICU patients with an increased risk of CRBS infections, patients in whom CRBS infections are likely to have devastating consequences, and in those with a previous CRBS infection. The choice of antimicrobial lock depends on the local prevalence of bacterial isolates in the hospital. Citrate-based catheter lock solutions are effective in preventing CRBS infections, but may increase the risk of thrombotic complications.141

have devastating consequences, and in those with a previous CRBS infection. The choice of antimicrobial lock depends on the local prevalence of bacterial isolates in the hospital. Citrate-based catheter lock solutions are effective in preventing CRBS infections, but may increase the risk of thrombotic complications.141 The risk of acute peritonitis related to acute PD is relatively high in developing countries. Ponce et al. reported peritonitis in patients who underwent high volume PD for AKI; 18 of 204 patients (12%) had peritonitis, of whom 8 (61%) underwent catheter removal.47 There is a risk of hospital-acquired bacterial peritonitis in children if the stylet catheter is used beyond 36 to 48 hours, which may limit the duration of PD. Patients initiated on RRT should be screened for hepatitis B virus (HBV), hepatitis C virus (HCV), and human immunodeficiency virus (HIV). We suggest that dialyzers not be reused if patients are seropositive for HBV, HCV, or HIV. Standard disinfection procedures for the dialysis machines should be carried out after each therapy, irrespective of the infectious status of the patient. It may not be practical to use dedicated machines for patients who are seropositive for HBV, HCV, or HIV, but in units where segregation of machines for seropositive long-term dialysis patients is practiced, the same policy should apply to patients with AKI. Research Recommendations • To study the impact of frequent laboratory tests versus infrequent laboratory tests on the rate of complications and therapeutic goals in a randomized fashion.

Patients initiated on RRT should be screened for hepatitis B virus (HBV), hepatitis C virus (HCV), and human immunodeficiency virus (HIV). We suggest that dialyzers not be reused if patients are seropositive for HBV, HCV, or HIV. Standard disinfection procedures for the dialysis machines should be carried out after each therapy, irrespective of the infectious status of the patient. It may not be practical to use dedicated machines for patients who are seropositive for HBV, HCV, or HIV, but in units where segregation of machines for seropositive long-term dialysis patients is practiced, the same policy should apply to patients with AKI. Research Recommendations • To study the impact of frequent laboratory tests versus infrequent laboratory tests on the rate of complications and therapeutic goals in a randomized fashion. • To compare online monitoring of solute clearance by analyzing ionic changes and actual measurement of small-solute clearance. • To develop models to predict the risk of circuit clotting based on pressure changes within the extracorporeal circuit. • To study the impact of ultrafiltration rates based on hemodynamic parameters, such as mean arterial pressure, and other technologies, such as bioimpedance on renal recovery, hospital stay, time on mechanical ventilation, and hospital mortality. • To study quality control initiatives to reduce the risk of bacterial infections during PD and IHD.

• To study the impact of ultrafiltration rates based on hemodynamic parameters, such as mean arterial pressure, and other technologies, such as bioimpedance on renal recovery, hospital stay, time on mechanical ventilation, and hospital mortality. • To study quality control initiatives to reduce the risk of bacterial infections during PD and IHD. Q4: When Should RRT Be Transitioned or Stopped? Consensus Statements 4.1. We recommend the transition of RRT should depend on patient-related factors, such as physiological status, degree of discrepancy between demand and capacity, chances of renal recovery, and technical considerations (e.g., equipment availability and cost). 4.2. When the demand-to-capacity ratio increases and the hemodynamic condition of the patient worsens, escalation of RRT should be considered. 4.3. When the demand-to-capacity ratio improves, de-escalation of RRT to a therapy that places less strain on resources and cost should be considered.

Q4: When Should RRT Be Transitioned or Stopped? Consensus Statements 4.1. We recommend the transition of RRT should depend on patient-related factors, such as physiological status, degree of discrepancy between demand and capacity, chances of renal recovery, and technical considerations (e.g., equipment availability and cost). 4.2. When the demand-to-capacity ratio increases and the hemodynamic condition of the patient worsens, escalation of RRT should be considered. 4.3. When the demand-to-capacity ratio improves, de-escalation of RRT to a therapy that places less strain on resources and cost should be considered. Context The transition from one initial RRT modality to another modality later in the course of AKI is common in clinical practice. The main reasons for switching are changes in the clinical condition of the patient or adverse events. In an RCT that compared CRRT and IHD, 20% of patients switched from initial CRRT to IHD, and 18% transitioned from initial IHD to CRRT.30 Higher rates were reported in a different RCT, in which 20% of the patients assigned to IHD group switched to CRRT, and 46% of the patients who were randomized to CRRT later changed to IHD.31 The transition from IHD to CRRT occurred earlier (mean time: 4.4 ±12 days) compared with the switch from CRRT to IHD (mean time: 6.2 ± 5.6 days). In the recent early versus late initiation of RRT in critically ill patients with AKI (ELAIN) study, 26% of patients switched from CRRT to SLED, 2% transitioned from CRRT to IHD, and 6% changed from CRRT to IHD and SLED.18

(mean time: 4.4 ±12 days) compared with the switch from CRRT to IHD (mean time: 6.2 ± 5.6 days). In the recent early versus late initiation of RRT in critically ill patients with AKI (ELAIN) study, 26% of patients switched from CRRT to SLED, 2% transitioned from CRRT to IHD, and 6% changed from CRRT to IHD and SLED.18 Transition from hybrid therapy to CRRT has also been reported. Fieghen et al. analyzed the data of 158 critically ill patients whose initial RRT modality was CRRT, and 74 patients who were initiated on SLED.38 Within 3 days of RRT initiation, 15% of patients who were initially started on SLED were changed to CRRT and 15% switched from IHD to SLED. Using data from 146 critically ill patients on RRT, Khanal et al. reported that 81% received PIRRT as the initial mode of RRT, of whom 21.2% also had exposure to CRRT.142 Annigeri et al. reported transition from PIRRT to CRRT within 24 hours of RRT initiation due to hemodynamic intolerance in 5% of patients.143 Ponce et al. had the largest experience using acute PD and recently reported that 51 of 301 patients (16.9%) were transferred from initial acute PD to IHD.48

exposure to CRRT.142 Annigeri et al. reported transition from PIRRT to CRRT within 24 hours of RRT initiation due to hemodynamic intolerance in 5% of patients.143 Ponce et al. had the largest experience using acute PD and recently reported that 51 of 301 patients (16.9%) were transferred from initial acute PD to IHD.48 Based on the existing data, we propose a schema to guide the appropriate transition of RRT (Table 4 and Figure 4). If the demand-to-capacity balance worsens or side effects related to a particular RRT modality occur, it is reasonable to consider escalation of RRT. Similarly, if the demand-to-capacity ratio improves, it is prudent to consider de-escalating to a RRT modality that places less strain on cost and resources.Figure 4 Schematic representation of proposed guide for consideration of transition of renal replacement therapy modality in acute kidney injury. CRRT, continuous renal replacement therapy; IHD, intermittent hemodialysis; PD, peritoneal dialysis; PIRRT, prolonged intermittent renal replacement therapy. Table 4 Factors that influence the transition of modality of renal replacement therapy in acute kidney injury

Based on the existing data, we propose a schema to guide the appropriate transition of RRT (Table 4 and Figure 4). If the demand-to-capacity balance worsens or side effects related to a particular RRT modality occur, it is reasonable to consider escalation of RRT. Similarly, if the demand-to-capacity ratio improves, it is prudent to consider de-escalating to a RRT modality that places less strain on cost and resources.Figure 4 Schematic representation of proposed guide for consideration of transition of renal replacement therapy modality in acute kidney injury. CRRT, continuous renal replacement therapy; IHD, intermittent hemodialysis; PD, peritoneal dialysis; PIRRT, prolonged intermittent renal replacement therapy. Table 4 Factors that influence the transition of modality of renal replacement therapy in acute kidney injury Patient-related factors: 1. Change in the physiologic status of the patient 2. Change in the metabolic demand and capacity ratio (azotemia, acidosis, and electrolyte and divalent ion balance and fluid balance) 3. To facilitate mobility of patient 4. To facilitate better renal recovery Factors related to technology, technical capacity, and availability: 1. Availability of technology 2. Availability for human resources and expertise to provide therapy 3. Backup technical support 4. Technical failure and complications related to technology Factors related to cost and resource constraints: 1. Cost of therapy 2. Resource allocation issues

nical capacity, and availability: 1. Availability of technology 2. Availability for human resources and expertise to provide therapy 3. Backup technical support 4. Technical failure and complications related to technology Factors related to cost and resource constraints: 1. Cost of therapy 2. Resource allocation issues Consensus Statement 4.4. RRT should be discontinued if kidney function has recovered sufficiently to reduce the demand-to-capacity imbalance (current and expected) to acceptable levels or the overall goals of treatment have changed. Context Discontinuation of RRT may be considered when there is sufficient improvement in renal function to meet the metabolic and fluid demands, or there is an improvement in the demand-to-capacity balance that favors weaning patients from RRT. When only fluid demand exists, it is reasonable to consider a trial of diuretics at a higher dose, in view of the reduced GFR.144 Common triggers that prompt a trial of RRT discontinuation are a decline in serum creatinine while on a constant dose of RRT and a progressive increase in urine output.145, 146 In a large multicenter observational study, Uchino et al. showed that a spontaneous urine output of >400 ml/d was associated with a 80.9% chance of successful transition off RRT.145 In patients who received diuretics, a urine output >2330 ml/d had a positive predictive value of 87.9% for transition off RRT. In a recent analysis of 67 patients, Aniort et al. concluded that daily urine urea excretion was superior to urine output in predicting successful weaning from IHD.147

ul transition off RRT.145 In patients who received diuretics, a urine output >2330 ml/d had a positive predictive value of 87.9% for transition off RRT. In a recent analysis of 67 patients, Aniort et al. concluded that daily urine urea excretion was superior to urine output in predicting successful weaning from IHD.147 Consensus Statement 4.5. We suggest considering discontinuation of RRT after a time trial of 48 hours in cases of deteriorating or non-improving clinical status. Context As stated previously (see section Q2), if a time trial of RRT was implemented following a shared decision process, lack of clinical improvement and worsening multiple organ failure after 48 hours could prompt the decision to discontinue RRT. Ferreira et al. showed that a rise in the sequential organ failure assessment score by 30% was associated with a mortality risk of >50%.148 When the goal of therapy is palliation, it is also rational to discontinue RRT. Research Recommendations • To prospectively study the common reasons for transition of RRT modality to determine clinical practice patterns across the world. • To evaluate a strategy of using CRRT as the initial modality followed by a rapid switch to hybrid therapy after 36 to 72 hours. This has the advantage of offering the most tolerated modality when the demand is maximum and switching to a less expensive modality as soon as the clinical situation allows.

Research Recommendations • To prospectively study the common reasons for transition of RRT modality to determine clinical practice patterns across the world. • To evaluate a strategy of using CRRT as the initial modality followed by a rapid switch to hybrid therapy after 36 to 72 hours. This has the advantage of offering the most tolerated modality when the demand is maximum and switching to a less expensive modality as soon as the clinical situation allows. • To evaluate a strategy of using PD as the primary RRT with supplementary hybrid therapy as directed by the demand-to-capacity balance instead of switching from PD to IHD. Such a strategy may save costs and improve the chances of renal recovery. Summary Providing RRT for patients with AKI in developing countries is challenging due to limited resources, cost constraints, and sociocultural aspects. Our consensus recommendations provide minimum requirements for use of RRT in resource-limited countries. Future research should focus on the innovations in RRT to provide optimum care and maximum outcomes in these settings. Disclosure All the authors declared no competing interests. Author Contributions RA, AT, AVR, DP, RC, and RM all participated in the consensus-building process and drafting of this paper. MO, RM, RC, and AB provided a critical review of this paper. Acknowledgment Supported through the UAB-UCSD O’Brien Center NIH-NIDDK Grant DK079337.

The choice of dialysis modality is one of the most important decisions for dialysis patients and their families. Research has demonstrated that home-based peritoneal dialysis (PD), compared to in-center hemodialysis (HD), has significant clinical advantages. Specifically, PD treatment preserves residual renal function better,1 requires a lower dose of erythropoiesis-stimulating agent (ESA) to treat anemia2 and avoids the need for a vascular access, thus reducing infections, which are the leading cause of hospitalization and mortality among end-stage renal disease (ESRD) patients.3, 4 Studies have also shown favorable effects of PD compared to HD in health-related quality of life,5 treatment satisfaction,6 and survival in the first 1 to 2 years of dialysis among ESRD patients when age, diabetes, and cardiovascular disease are considered.7, 8, 9 In addition, PD is less costly than HD, as home-based dialysis saves costs on staff overhead and dialysis supplies.10 The average per-treatment costs for delivering HD and PD, estimated by the United States Government Accounting Office (GAO), were $251 and $94, respectively.11

diovascular disease are considered.7, 8, 9 In addition, PD is less costly than HD, as home-based dialysis saves costs on staff overhead and dialysis supplies.10 The average per-treatment costs for delivering HD and PD, estimated by the United States Government Accounting Office (GAO), were $251 and $94, respectively.11 Despite clinical benefits and economic advantages, the present use of PD in the USA is low, being approximately 6.8% among prevalent dialysis patients in 2013,12 which is far less than the optimal rate (35%) recommended by nephrologists,13 as well as PD use in other industrialized countries (20%-81%).14 It has been suggested that nonmedical factors—especially financial incentives—contribute to the low use of PD in the USA.14, 15 Prior to the development of the End Stage Renal Disease (ESRD) Prospective Payment System (PPS), Medicare paid separately for injectable medications based on dose administered. HD patients received ESAs i.v. compared to PD patients, who received ESAs subcutaneously, requiring 2 to 4 times more ESA dose.16 Therefore, dialysis facilities were able to increase their profits by having a majority of patients on HD.17, 18 The new PPS bundled routine dialysis services including injectable drugs into an equivalent Medicare payment for both PD and HD, and thus removed the financial barrier to the use PD. To further promote home dialysis, the Centers for Medicare and Medicaid Services (CMS) imposed additional reforms in PPS such as raising payment for home dialysis training by 60%.19

s including injectable drugs into an equivalent Medicare payment for both PD and HD, and thus removed the financial barrier to the use PD. To further promote home dialysis, the Centers for Medicare and Medicaid Services (CMS) imposed additional reforms in PPS such as raising payment for home dialysis training by 60%.19 Rigorous studies have not been conducted to examine whether PPS has effectively increased PD use. In this study, we evaluated whether PPS is associated with increased PD use among incident dialysis patients in the USA using a quasi-experimental design that examined dialysis modality choice among all Medicare ESRD patients initiating treatment 2 years before and 2 years after PPS implementation. Methods Data Sources We used data from United States Renal Data System (USRDS) from January 2009 to December 2012, 2 years before and 2 years after PPS, to conduct this study. The variables included in the USRDS Standard Analytical Files (SAFs), as well as the data source, collection methods, and validation studies, are described on the USRDS website (http://www.usrds.org). As a special data request, USRDS gave us the provider number for each patient at dialysis initiation, thereby allowing us to conduct subgroup analysis by provider characteristics. By cross-referencing facility data and patient-level data, a patient−provider file was constructed for analysis.

ite (http://www.usrds.org). As a special data request, USRDS gave us the provider number for each patient at dialysis initiation, thereby allowing us to conduct subgroup analysis by provider characteristics. By cross-referencing facility data and patient-level data, a patient−provider file was constructed for analysis. Study Measures Our primary study measure was the rate of monthly PD use, calculated as a percentage of total continuous ambulatory PD (CAPD) and continuous cycling PD (CCPD) among all dialysis modalities for each month in the 48-month study period. Choice of modality at dialysis initiation was determined from the Medical Evidence Form (CMS-2728), mandatory for every new ESRD patient at initiation of renal replacement therapy in the USA. The modality variable in the MEF has been widely used in previous studies.7, 9 Because our focus was difference in in-center HD versus continuous PD after PPS, home HD or intermittent PD patients (∼3%) were excluded from this study. As a secondary analysis, we used the USRDS definition to ascertain dialysis modality; PD was determined by identifying PD use on day 90 after dialysis initiation with continuous treatment using PD in the subsequent 60 days (known as the 60-day rule).

e HD or intermittent PD patients (∼3%) were excluded from this study. As a secondary analysis, we used the USRDS definition to ascertain dialysis modality; PD was determined by identifying PD use on day 90 after dialysis initiation with continuous treatment using PD in the subsequent 60 days (known as the 60-day rule). Patient demographics, comorbid conditions, and laboratory values including hemoglobin, serum albumin, and glomerular filtration rate (GFR) were collected at dialysis initiation. Patients who were unable to ambulate or to transfer, those who needed assistance with daily activities, and those who lived in a nursing home, assisted living, or other institutions were categorized as “inability to ambulate or institutionalized.” Chain and profit status associated with the facility in which the patient initiated dialysis were also determined using USRDS Facility file. Interrupted Time Series Analysis We used interrupted time series (ITS) regression models (segmented regression analysis)20, 21 with maximum likelihood method to evaluate changes in rate of PD use that occurred after PPS, controlling for the baseline pre-PPS period. The ITS model in this study was the following: Y[t]=β0+β1∗timebeforePPS+β2∗PPS+β3∗timeafterPPS+et where β0 estimates PD rate at the beginning of the study period; β1 estimates the change in PD rate each month before PPS; β2 estimates the level change in PD rate immediately after PPS; and β3 estimates the change in the trend in PD rate after PPS.

was the following: Y[t]=β0+β1∗timebeforePPS+β2∗PPS+β3∗timeafterPPS+et where β0 estimates PD rate at the beginning of the study period; β1 estimates the change in PD rate each month before PPS; β2 estimates the level change in PD rate immediately after PPS; and β3 estimates the change in the trend in PD rate after PPS. Serial autocorrelation was tested by the Durbin−Watson (DW) statistic by using backward elimination. Heteroscedasticity was tested by the Q statistic in the regression model. We also accessed seasonality by white noise test using the Fisher κ and Bartlett Kolmogorov−Smirnov (BKS) statistic.22 We adjusted both seasonality and lagged intervention effects in our final factorial autoregressive models. We further stratified our analyses by patient demographics, clinical history, predialysis care, and dialysis facility chain and profit status to determine whether the effect of PPS differed among patient subgroups. Lagged effects and seasonality were assessed using stepwise autoregressive analyses. After adjustment, the Durbin−Watson statistic for our final factored autoregressive model was 1.803 (P value for hypothesis of positive autocorrelation = 0.127, P value for hypothesis of negative autocorrelation = 0.877) with R2Autoreg of 0.90, indicating no autocorrelation and good model fit. All analyses were conducted using SAS (SAS Institute Inc, Cary NC), mainly AUTOREG and SPECTEA procedures.

ve model was 1.803 (P value for hypothesis of positive autocorrelation = 0.127, P value for hypothesis of negative autocorrelation = 0.877) with R2Autoreg of 0.90, indicating no autocorrelation and good model fit. All analyses were conducted using SAS (SAS Institute Inc, Cary NC), mainly AUTOREG and SPECTEA procedures. Results The study population across the 48 months (N = 430,927) comprised 57% male and 43% female participants. Of the participants, 50% were aged ≥65 years; 66% were white; 56% were diabetic; 87% had hypertension; 54% had cardiovascular disease; 8% had cancer; 10% had chronic obstructive pulmonary disease; 8% were employed full time; and 18% were unable to ambulate or were institutionalized. Approximately 42% had no predialysis nephrology care, and 63% underwent dialysis in large dialysis organizations (LDOs). Although statistically significant because of our large sample size, there were only minor differences in patient age, sex, presence of comorbidities, laboratory values, and dialysis facility characteristics between the baseline pre-PPS and post-PPS periods (Table 1).Table 1 Characteristics of patients initiating dialysis before and after the January 2011 Prospective Payment System (PPS)

here were only minor differences in patient age, sex, presence of comorbidities, laboratory values, and dialysis facility characteristics between the baseline pre-PPS and post-PPS periods (Table 1).Table 1 Characteristics of patients initiating dialysis before and after the January 2011 Prospective Payment System (PPS) Entire study period Pre-PPS Post-PPS P value Number of patients 430,927 217,867 213,060 Age group (yr) % % % 18−44 11.4 11.6 11.3 <0.0001 45−64 38.6 38.3 39.0 ≥65 49.9 50.1 49.7 Sex Male 56.9 56.7 57.1 0.006 Female 43.1 43.3 42.9 Race White 65.6 65.6 65.7 0.458 Nonwhite 34.4 34.4 34.3 Ethnicity Hispanic 9.3 8.9 9.7 <0.0001 Non-Hispanic 86 86.3 85.7 Employed full-time 8 8 8 0.739 Inability to ambulate/institutionalized 17.8 17.6 18.1 <0.0001 Primary cause of renal failure Hypertension/large vessel disease 29.4 29.2 29.5 <0.0001 Diabetes 45.9 45.6 46.3 Glomuleronephritis 5.8 5.8 5.8 Other 18.9 19.4 18.4 Diabetes 55.9 55.2 56.6 <0.0001 Hypertension 86.6 86.1 87.1 <0.0001 Cardiovascular disease 53.6 54.2 53.1 <0.0001 Atherosclerotic heart disease 20.1 21.1 19.2 <0.0001 Congestive heart failure 31.9 32.5 31.3 <0.0001 Other cardiac disease 18.3 18 18.6 <0.0001 Cerebrovascular disease 9.4 9.5 9.2 0.000 Peripheral vascular disease 13.3 13.7 12.8 <0.0001 Cancer 7.7 7.7 7.6 0.2575 Chronic obstructive lung disease 9.7 9.6 9.8 0.030 BMI (kg/m2, mean ± SD) 29.5 ± 8.1 29.5 ± 8.1 29.6 ± 8.1 <0.0001 Hemoglobin (g/dl, mean ± SD) 10.2 ± 16.9 10.3 ± 16.7 10 ± 17.1 0.000 Serum albumin (g/dl, mean ± SD) 3.2 ± 4.4 3.2 ± 4.7 3.2 ± 4 0.201 GFR (ml/min/1.73 m2, mean ± SD)a 12 ± 5.4 12.1 ± 5.5 11.9 ± 5.4 <0.0001 Nephrology care No care 41.7 42.9 40.6 <0.0001 0−12 mo 32.4 32.5 32.2 >12 mo 25.9 24.6 27.2 Facility chain status <0.0001 LDO 63.2 61.8 64.6 SDO 11.3 12.1 10.6 Nonchain 25.5 26.2 24.8 Facility profit status <0.0001 For profit 83.6 82.7 84.4 Nonprofit 16.4 17.3 15.6 Pre-PPS period is from January 2009 to December 2010. Post-PPS period is from January 2011 to December 2012. P value for Pearson χ2 test or t test was based on the difference between pre- and post-PPS periods. BMI, body mass index; GFR, glomerular filtration rate; LDO, large dialysis chain; SDO, small dialysis chain.

6 Pre-PPS period is from January 2009 to December 2010. Post-PPS period is from January 2011 to December 2012. P value for Pearson χ2 test or t test was based on the difference between pre- and post-PPS periods. BMI, body mass index; GFR, glomerular filtration rate; LDO, large dialysis chain; SDO, small dialysis chain. a GFR was calculated using the Modification of Diet in Renal Disease Study (MDRD) equation.

6 Pre-PPS period is from January 2009 to December 2010. Post-PPS period is from January 2011 to December 2012. P value for Pearson χ2 test or t test was based on the difference between pre- and post-PPS periods. BMI, body mass index; GFR, glomerular filtration rate; LDO, large dialysis chain; SDO, small dialysis chain. a GFR was calculated using the Modification of Diet in Renal Disease Study (MDRD) equation. Unadjusted Analyses and Covariate Effects Overall, the average PD use rate increased from 6.4% to 7.9% between the 2-year pre- and 2-year post-PPS periods for all dialysis patients (Table 2). Throughout the study, participants who were younger (18−44 years), female, white, employed full-time, and those with more than 12 months of nephrology care were more likely to start dialysis with PD compared to their counterparts (P < 0.05). Furthermore, healthier patients were more likely to start dialysis with PD (P < 0.05), including those without diabetes, cardiovascular disease, cancer, or chronic obstructive pulmonary disease, and those with higher values of hemoglobin (≥12 g/dl), serum albumin (≥3.5 g/dl), glomerular filtration rate (GFR) (≥8 ml/min/1.73 m2), and body mass index (BMI) (≥18.5 kg/m2). Patients who were unable to ambulate or who were institutionalized had the lowest rates of PD across the study period. Over the 4-year period, patients who had more than 12 months of predialysis care were more likely to start their dialysis with PD than those with 0 to 6 months of care or those without predialysis nephrology care. Patients who underwent dialysis in small dialysis organizations (SDOs) were more likely to start with PD as compared to those in nonchain facilities and LDOs (Table 2).Table 2 Rate of peritoneal dialysis (PD) use as proportion of all dialysis modality by patient characteristics from January 2009 to December 2012

care. Patients who underwent dialysis in small dialysis organizations (SDOs) were more likely to start with PD as compared to those in nonchain facilities and LDOs (Table 2).Table 2 Rate of peritoneal dialysis (PD) use as proportion of all dialysis modality by patient characteristics from January 2009 to December 2012 Entire study period Pre-PPS Post-PPS % % % All 7.1 6.4 7.9 Demographics Age, yr 18−44 11.8 10.4 13.3 45−64 8.2 7.5 9.0 65+ 5.2 4.6 5.7 Sex 7.0 6.2 7.7 Male Female 7.3 6.6 8.0 Race White 7.6 6.9 8.3 Nonwhite 6.2 5.4 6.9 Employed full-time Yes 19.2 17.3 21.0 No 6.1 5.4 6.7 Comorbid conditions Cardiovascular disease Yes 4.7 4.3 5.2 No 9.9 8.8 10.9 Diabetes Yes 6.3 5.6 7.0 No 8.1 7.3 9.0 Hypertension Yes 7.3 6.6 8.0 No 6.0 5.2 6.8 Cancer Yes 4.7 4.1 5.3 No 7.3 6.6 8.1 Chronic obstructive lung disease Yes 3.4 3.1 3.6 No 7.5 6.7 8.3 Inability to ambulate/institutionalized Yes 2.0 1.7 2.2 No 8.2 7.4 9.1 Laboratory values BMI (kg/m2) <18.5 4.4 3.8 5.0 18.5 to <30 7.2 6.5 8.0 ≥30 7.2 6.4 7.9 Hemoglobin (g/dl) <10 4.8 4.0 5.5 10 to <12 9.6 8.6 10.7 ≥12 11.4 10.8 12.2 Serum albumin (g/dl) <3.5 4.1 3.6 4.6 ≥3.5 13.5 12.2 14.9 GFR (ml/min/1.73 m2)a <5 4.2 3.5 4.9 5 to <8 6.4 5.6 7.1 ≥8 7.7 6.9 8.5 Dialysis care Nephrology care No care 2.3 2.0 2.6 0−6 mo 9.6 8.8 10.4 >12 mo 11.9 10.9 12.7 Facility chain status LDO 6.9 6.1 7.7 SDO 8.0 6.9 9.4 Nonchain 7.3 6.9 7.7 Facility profit status For-profit 7.1 6.4 7.9 Nonprofit 7.0 6.3 7.8 Pre-PPS period is from January 2009 to December 2010. Post-PPS period is from January 2011 to December 2012. P value for Pearson χ2 test or t test was based on the difference between pre- and post-PPS periods. BMI, body mass index; GFR, glomerular filtration rate; LDO, large dialysis chain; PPS, Prospective Payment System; SDO, small dialysis chain.

2009 to December 2010. Post-PPS period is from January 2011 to December 2012. P value for Pearson χ2 test or t test was based on the difference between pre- and post-PPS periods. BMI, body mass index; GFR, glomerular filtration rate; LDO, large dialysis chain; PPS, Prospective Payment System; SDO, small dialysis chain. a GFR was calculated using the Modification of Diet in Renal Disease Study (MDRD) equation. ITS Analysis of PD Rate Figure 1 shows the actual, predicted, and mean rates of PD use for each month from January 2009 (4.8% use) to December 2012 (7.8% use), the end of the study period. ITS analysis indicated that PPS implementation resulted in increased use of PD in the 2-year period after PPS (change in slope = 0.04, 95% confidence interval [CI] = 0.03−0.06, P < 0.0001). The trend of increasing PD use began in the 2-year period prior to 2011 (trend [or slope] before bundling 0.04, 95% CI = 0.03−0.06, P < 0.0001) and accelerated in the 2-year follow-up. There was no immediate change in level of PD use after PPS (P = 0.512).Figure 1 Time series of monthly peritoneal dialysis (PD) use from January 2009 to December 2012. Fitted trend line shows predicated values from the segmented regression analysis. PPS, Prospective Payment System.

accelerated in the 2-year follow-up. There was no immediate change in level of PD use after PPS (P = 0.512).Figure 1 Time series of monthly peritoneal dialysis (PD) use from January 2009 to December 2012. Fitted trend line shows predicated values from the segmented regression analysis. PPS, Prospective Payment System. Stratified ITS analyses indicated PPS led to an increased PD use across all age, race, and sex groups (P < 0.05) except in females (P = 0.086) after PPS (Table 3). Increased PD use was also found regardless of employment status, dialysis chain, or profit status. It appears that SDOs had a higher rate of PD increase compared to LDOs (change in slope = 0.12, 95% CI = 0.07−0.17, vs. 0.03, 95% CI = 0−0.05, respectively) (Figure 2). Similarly, it appears that nonprofit organizations had a higher rate of PD increase compared to for-profit organizations (change in slope = 0.08, 95% CI = 0.04−0.12 vs. 0.04, 95% CI = 0.02−0.06, respectively). The extent of nephrology care before dialysis had a unique pattern in terms of influencing PD use after PPS; no care and ≥12 months of predialysis nephrology care resulted in similar changes in slope after PPS (0.05, with 95% CI = 0.04−0.07, and 0.07, with 95% CI = 0.02−0.11, respectively), but PD use in patients with 0 to 12 months of nephrology care did not change (−0.03 with 95% CI = −0.06 to 0.01) (Table 3 and Figure 2). The rate of PD use significantly increased after PPS among ESRD patients with diabetes and hypertension as well as those whose with BMI ≥ 30 or GFR ≥ 5 ml/min at dialysis initiation (Supplementary Table S1).Figure 2 Time series of monthly peritoneal dialysis (PD) use by predialysis nephrology care or dialysis facility chain status from January 2009 to December 2012. Fitted trend line shows predicated values from the segmented regression analysis. LDO, large dialysis organization; SDO, small dialysis organization.

able S1).Figure 2 Time series of monthly peritoneal dialysis (PD) use by predialysis nephrology care or dialysis facility chain status from January 2009 to December 2012. Fitted trend line shows predicated values from the segmented regression analysis. LDO, large dialysis organization; SDO, small dialysis organization. Table 3 Changes in rate of peritoneal dialysis (PD) use after PPS assessed using interrupted time series stratified by selected characteristics

able S1).Figure 2 Time series of monthly peritoneal dialysis (PD) use by predialysis nephrology care or dialysis facility chain status from January 2009 to December 2012. Fitted trend line shows predicated values from the segmented regression analysis. LDO, large dialysis organization; SDO, small dialysis organization. Table 3 Changes in rate of peritoneal dialysis (PD) use after PPS assessed using interrupted time series stratified by selected characteristics Trend before bundling Change in level after PPS Change in trend after PPS β1 P value β2 P value β3 P value All 0.04 (0.03,0.06) <0.0001 –0.09 (–0.38 to 0.19) 0.512 0.04 (0.03–0.06) <0.0001 Age, yr 18–44 0.05 (–0.01 to 0.11) 0.085 0.49 (–0.68 to 1.67) 0.393 0.10 (0.03–0.17) 0.007 45–64 0.05 (0.03–0.08) 0.0001 –0.24 (–0.77 to 0.29) 0.354 0.04 (0.01–0.07) 0.020 ≥65 0.04 (0.02–0.06) <0.0001 –0.22 (–0.55 to 0.11) 0.186 0.03 (0.01–0.05) 0.009 Sex Female 0.05 (0.02–0.08) 0.001 0.02 (–0.55 to 0.60) 0.934 0.03 (0–0.07) 0.086 Male 0.04 (0.02–0.06) <0.0001 –0.21 (–0.59 to 0.18) 0.286 0.05 (0.03–0.07) <0.0001 Race White 0.03 (0.02–0.05) <0.0001 –0.06 (–0.44 to 0.31) 0.729 0.05 (0.03–0.07) <0.0001 Nonwhite 0.06 (0.04–0.07) <0.0001 –0.14 (–0.51 to 0.23) 0.454 0.03 (0.01–0.05) 0.016 Employed full-time No 0.04 (0.03–0.05) <0.0001 –0.16 (–0.39 to 0.08) 0.192 0.03 (0.02–0.05) <0.0001 Yes 0.10 (0.04–0.17) 0.001 0.26 (–1.01 to 1.52) 0.683 0.08 (0.001–0.16) 0.04 Nephrology care No care 0.01 (0–0.02) 0.204 –0.20 (–0.44 to 0.03) 0.090 0.05 (0.04–0.07) <0.0001 0–12 mo 0.08 (0.05–0.11) <0.0001 0.07 (–0.47 to 0.62) 0.786 –0.03 (–0.06 to 0.01) 0.122 >12 mo 0.05 (0.01–0.09) 0.008 –0.35 (–1.12 to 0.42) 0.358 0.07 (0.02–0.11) 0.005 Chain status LDO 0.05 (0.03–0.07) <0.0001 0.07 (–0.29 to 0.44) 0.684 0.03 (0–0.05) 0.020 SDO 0.03 (0–0.07) 0.071 0.23 (–0.54 to 0.99) 0.549 0.12 (0.07–0.17) <0.0001 Nonchain 0.037 (0.01–0.06) 0.01 –0.91 (–1.49 to 0.33) 0.003 0.06 (0.03–0.10) 0.001 Profit/nonprofit For profit 0.05 (0.03–0.06) <0.0001 –0.17 (–0.49 to 0.16) 0.296 0.039 (0.02–0.06) 0.0002 Nonprofit 0.01 (–0.02 to 0.04) 0.486 0.263 (–0.37 to 0.90) 0.407 0.078 (0.04–0.12) 0.0002 LDO, large dialysis chain; PPS, Prospective Payment System; SDO, small dialysis chain.

to 0.33) 0.003 0.06 (0.03–0.10) 0.001 Profit/nonprofit For profit 0.05 (0.03–0.06) <0.0001 –0.17 (–0.49 to 0.16) 0.296 0.039 (0.02–0.06) 0.0002 Nonprofit 0.01 (–0.02 to 0.04) 0.486 0.263 (–0.37 to 0.90) 0.407 0.078 (0.04–0.12) 0.0002 LDO, large dialysis chain; PPS, Prospective Payment System; SDO, small dialysis chain. In a secondary analysis ascertaining PD modality at month 3 versus at initiation of dialysis, the total PD population in the ITS analysis was 199,937 (less than one-half of the PD population identified using the 2728 form). Because “claims data” are required to identify PD patients at month 3, patients who were Medicare Secondary Payor (MSP) and those enrolled in health maintenance organization plans were not included in this secondary analysis. However, results from this secondary analysis are similar to the results reported above. The ITS analysis indicated that PPS implementation resulted in increased use of PD in the 2-year period after PPS (change in slope = 0.03, 95% CI = 0.02−0.05, P < 0.003). The trend of increasing PD use also began in the 2-year period prior to 2011 (trend [or slope] before bundling = 0.07, 95% CI = 0.05−0.09, P < 0.0001). There was no immediate change in level of PD use after PPS (P = 0.902) (Supplementary Figure S1).

iod after PPS (change in slope = 0.03, 95% CI = 0.02−0.05, P < 0.003). The trend of increasing PD use also began in the 2-year period prior to 2011 (trend [or slope] before bundling = 0.07, 95% CI = 0.05−0.09, P < 0.0001). There was no immediate change in level of PD use after PPS (P = 0.902) (Supplementary Figure S1). Discussion Our study confirmed the previously unexamined hypothesis that implementation of the CMS ESRD Prospective Payment System is associated with an increase in PD use in incident dialysis patients in the USA. This trend started prior to 2011 (implementation of PPS) but accelerated in the 2 years after PPS was launched. Our findings are consistent with USRDS Annual Data Report,23 and recent studies24, 25, 26 showing an increase in rate of PD use over time. However, these previous studies and reports were descriptive in nature, whereas our study, using the census of dialysis patients initiating PD therapy in the USA was the first to use a causal, quasi-experimental design to demonstrate that PPS itself was effective in increasing PD use.

rease in rate of PD use over time. However, these previous studies and reports were descriptive in nature, whereas our study, using the census of dialysis patients initiating PD therapy in the USA was the first to use a causal, quasi-experimental design to demonstrate that PPS itself was effective in increasing PD use. Historically, financial considerations have played an important role in many clinical decisions among ESRD patients.27 In this case, choice of PD shifted from a financial disincentive before PPS to an incentive after PPS. Specifically, under the past payment structure, injectable medications such as ESAs were paid based on the total amount administered; because a PD patient tended to use less i.v. medication than an HD patient, potential revenues and profits generated from larger dosage of injectable drugs given at HD treatment outweighed the less costly PD use.28 Consequently, providers might have been discouraged to prescribe PD as a dialysis treatment option. Conversely, PPS applies a fixed payment covering all dialysis services including injectable drugs. Given the equal payments for both HD and PD modalities, savings from significantly fewer requirements of expensive ESA doses makes delivery of PD treatment more profitable to the provider compared to HD treatment. Combined with fewer staffing requirements and less expensive supplies, providers could administer more PD care with less use of resources and could take advantage of the inherent profitability of PD under the new bundle.

ESA doses makes delivery of PD treatment more profitable to the provider compared to HD treatment. Combined with fewer staffing requirements and less expensive supplies, providers could administer more PD care with less use of resources and could take advantage of the inherent profitability of PD under the new bundle. In addition to injectable medications, other factors were important in the growth of home dialysis before and after PPS. According to Mark Neumann, Editor-in-Chief at Nephrology News & Issues (NN&I), who conducted an annual ranking of dialysis providers in the USA,29 the percentage of patients on home therapies, particularly PD, has been growing since 2010. Incentives offered by Medicare such as the Comprehensive ESRD Care Initiative demonstration played an important role in helping patients choose PD therapy. Educational efforts, such as the NN&I-produced webinar “Home Dialysis: Next Steps” have also been touted as increasing visibility and decreasing barriers associated with home dialysis. Numerous courses made available by universities, the American Society of Nephrology (ASN), the National Kidney Foundation (NKF), the International Society for Peritoneal Dialysis (ISPD), and other renal organizations were designed to expand the knowledge base and comfort level of physicians to perform PD. A movement to offer patients “urgent” PD therapy instead of the traditional route from the emergency department of an HD catheter and in-center HD has garnered interest and increased initiation of PD.30 Other home dialysis programs have succeeded using a Web-based project funded by Baxter that offers expertise on setting up a home program. Finally, increasing emphasis on predialysis education paid by Medicare—the Kidney Disease Education (KDE) benefit—has helped more patients learn about PD.

reased initiation of PD.30 Other home dialysis programs have succeeded using a Web-based project funded by Baxter that offers expertise on setting up a home program. Finally, increasing emphasis on predialysis education paid by Medicare—the Kidney Disease Education (KDE) benefit—has helped more patients learn about PD. During the past 2 decades, the ESRD industry has undergone tremendous market structural changes, with an influx of large, for-profit, multi-unit dialysis chains. Our finding that SDOs and nonprofit organizations appear to have increased use of PD after PPS compared to their LDO and for-profit counterparts may have been anticipated given their PD use prior to PPS; that is, historically, PD use has been significantly lower in for-profit units compared with not-for-profit units,31 and lower in large dialysis chain facilities (mostly for-profit) than in smaller units. Although the potential profitability of PD after PPS is anticipated to increase its use in all facilities, studies have shown that for-profit facilities appear to use fewer resources32, 33 to deliver hemodialysis services and therefore might not benefit financially to the same extent as nonprofit organizations and SDOs.

lthough the potential profitability of PD after PPS is anticipated to increase its use in all facilities, studies have shown that for-profit facilities appear to use fewer resources32, 33 to deliver hemodialysis services and therefore might not benefit financially to the same extent as nonprofit organizations and SDOs. Although PPS has increased rates of PD use, these rates remain low compared to other industrialized countries and the National Kidney Foundation (NKF) goals. Several significant barriers might need to be addressed in order to promote PD use in the future. First, long-existing HD facilities in practice may impede investing in construction of PD infrastructure.34 As a matter of fact, lack of PD supplies might be a bottleneck limiting PD use as reported, a plunge in PD rate in 2014 due to dialysate shortage.35 Rising of PD also calls upon adequate training of personnel to prepare them to implement PD. Currently, PD training during nephrology fellowship in the USA is very limited because it is not required by the Accreditation Council for Graduate Medical Education (ACGME).36 One study revealed 29% of nephrology training programs in the USA had fewer than 5 PD patients for 1 fellow; 14% of these programs spent less than 5% of the training time on PD training.37 Consequently, when surveyed, only one-half of nephrologists felt prepared to use PD and comfortable using it.38 Moreover, our study showed that length of predialysis care is associated with increased PD use, which is consistent with previous results that patient misconception and lack of knowledge of PD are strong barriers to PD use.39 and patients who are referred early to a nephrologist are more likely to choose PD.40 Referring patients in the late stages of chronic kidney disease to a nephrologist is an important key to improving patient PD use.

revious results that patient misconception and lack of knowledge of PD are strong barriers to PD use.39 and patients who are referred early to a nephrologist are more likely to choose PD.40 Referring patients in the late stages of chronic kidney disease to a nephrologist is an important key to improving patient PD use. A previous systematic review of 58 studies on Medicare’s use of Prospective Payment Systems by the Agency for Healthcare Research and Quality (AHRQ) concluded that bundled payment programs have successfully reduced costs without incurring major compromises on quality of care.41 Other studies, similar to ours, that have examined changes in practice patterns after implementation of ESRD PPS have shown that dialysis providers are now motivated to adopt less expensive strategies given bundling for injectable medications; for example, using less expensive oral and i.v. iron to substitute for ESA to treat anemia, increasing use of subcutaneous ESA route of administration (which requires a one-third to one-half the dose of i.v. ESA administration), and using less expensive oral vitamin D versus i.v. vitamin D.42, 43 Overall Medicare spending for dialysis drugs has been estimated to be reduced by $25 per dialysis session per patient after PPS.24 However, it might be too early to conclude that ESRD PPS represents a successful policy reform, without evidence that these substantial changes in care patterns have not adversely affected survival among dialysis patients.

lysis drugs has been estimated to be reduced by $25 per dialysis session per patient after PPS.24 However, it might be too early to conclude that ESRD PPS represents a successful policy reform, without evidence that these substantial changes in care patterns have not adversely affected survival among dialysis patients. Our study has several limitations. First, we were unable to include a comparable control group to investigate effects of other factors on PD use during the 4 years of the study, as PPS is a universal payment reform affecting ∼95% of dialysis patients in the USA enrolled in the Medicare ESRD Program. However, use of a quasi-experimental interrupted time series design enabled us to adjust for baseline trend and autocorrelation to improve internal validity.44 Moreover, the large sample size and tight indicator trend lines used in this study provide compelling evidence for the association of PPS with increased PD use. Future studies will determine whether this trend of increasing PD use is sustained by PPS and whether it will ameliorate the high mortality rates (nearly 17% annually) found among the dialysis population in the USA.12

lines used in this study provide compelling evidence for the association of PPS with increased PD use. Future studies will determine whether this trend of increasing PD use is sustained by PPS and whether it will ameliorate the high mortality rates (nearly 17% annually) found among the dialysis population in the USA.12 It is hoped that our results will inform policymakers inside and outside of the ESRD community regarding the possible effects of changes in financial incentives. We used an innovative time series model that uses Medicare ESRD payment reform as a “natural experiment” to study the impact of PPS on PD modality. Information provided herein is useful as Medicare continues to implement payment reforms that shift reimbursement from fee-for-service toward episode-based or capitated payments. With the passage of the Patient Protection and Affordable Care Act (ACA), policymakers face the challenge of minimizing health care costs while maintaining or improving quality of care. In conclusion, the CMS PPS has led to an increase in the use of PD among incident ESRD patients. Our findings highlight the role of financial incentives in changing practice patterns, in this case to increase the use of a dialysis modality considered by many to be both more cost-effective and empowering to ESRD patients. Disclosure All the authors declared no competing interests.

In conclusion, the CMS PPS has led to an increase in the use of PD among incident ESRD patients. Our findings highlight the role of financial incentives in changing practice patterns, in this case to increase the use of a dialysis modality considered by many to be both more cost-effective and empowering to ESRD patients. Disclosure All the authors declared no competing interests. Supplementary Material Figure S1 Time series of monthly peritoneal dialysis (PD) use at month 3 after dialysis initiation from January 2009 to December 2012. Fitted trend line shows predicated values from the segmented regression analysis. PPS, Prospective Payment System. Table S1 Changes in rate of peritoneal dialysis (PD) use after Prospective Payment System (PPS) assessed using interrupted time series stratified by comorbidities. Acknowledgments This study was funded in part by an Agency for Healthcare Research and Quality (AHRQ) grant R03 HS22931-01. The data reported herein have been supplied by the US Renal Data System. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as the official policy or interpretation of the US government. All authors met each of the 3 authorship requirements as stated in the Uniform Requirements for Manuscripts Submitted to Biomedical Journals. Furthermore, there are no disclaimers or ethical concerns for any of the authors to report. The study findings were presented in part at the Agency for Healthcare Research and Quality (AHRQ) 2015 Annual Conference, Crystal City, Virginia (6 October 2015).

m Requirements for Manuscripts Submitted to Biomedical Journals. Furthermore, there are no disclaimers or ethical concerns for any of the authors to report. The study findings were presented in part at the Agency for Healthcare Research and Quality (AHRQ) 2015 Annual Conference, Crystal City, Virginia (6 October 2015). Figure S1. Time series of monthly peritoneal dialysis (PD) use at month 3 after dialysis initiation from January 2009 to December 2012. Fitted trend line shows predicated values from the segmented regression analysis. PPS, Prospective Payment System. Table S1. Changes in rate of peritoneal dialysis (PD) use after Prospective Payment System (PPS) assessed using interrupted time series stratified by comorbidities. Supplementary material is linked to the online version of the paper at http://www.kireports.org.

Introduction Leukemia is among the most common malignancies associated with hospital admissions for acute kidney injury (AKI).1, 2 AKI develops in up to one-third of patients affected by hematologic malignancy, particularly in older patients with prior chronic kidney disease (CKD),3 and has been shown to have an adverse effect on long-term prognosis in this patient population.4 Intravascular volume depletion, tumor lysis syndrome, and drug-induced acute tubular injury are among the most common causes of leukemia-associated AKI.4, 5 Radiologic studies may be useful to exclude obstructive uropathy secondary to retroperitoneal lymphadenopathy or tumorous masses. Direct infiltration of the renal parenchyma by leukemic cells, although a common finding at autopsy,5 is only rarely associated with the development of symptomatic AKI.4 Rare cases of AKI secondary to intravascular leukostasis have also been described, particularly in patients with white blood cell counts greater than 100,000/mm3.6 Lysozyme-induced nephropathy (LyN) is a rare and underrecognized complication of chronic myelomonocytic leukemia (CMML) and other forms of monocytic leukemia in which lysozyme, a small cationic protein, is released into the circulation, filtered by the glomerulus and reabsorbed by the proximal tubule, causing toxic tubular injury.7

ced nephropathy (LyN) is a rare and underrecognized complication of chronic myelomonocytic leukemia (CMML) and other forms of monocytic leukemia in which lysozyme, a small cationic protein, is released into the circulation, filtered by the glomerulus and reabsorbed by the proximal tubule, causing toxic tubular injury.7 Case Presentation Clinical History and Initial Laboratory Data A 69-year-old white man with a history of CKD (baseline serum creatinine, 2.0 mg/dl; estimated glomerular filtration rate, 35 ml/min/1.73 m2) and well-controlled HIV infection (CD4 count, 500 cells/mm3; viral load, <20 copies/ml) presented to the emergency department after 10 days of watery diarrhea. Two months prior, he had been diagnosed with CMML, but had not commenced treatment. Maintenance antiretroviral therapy included darunavir, emtricitibine, and ritonavir. Physical examination findings were notable for a blood pressure of 119/63 mm Hg, splenomegaly, and the absence of edema.

er 10 days of watery diarrhea. Two months prior, he had been diagnosed with CMML, but had not commenced treatment. Maintenance antiretroviral therapy included darunavir, emtricitibine, and ritonavir. Physical examination findings were notable for a blood pressure of 119/63 mm Hg, splenomegaly, and the absence of edema. Initial laboratory evaluation (Table 1) revealed markedly elevated serum creatinine (10.9 mg/dl; estimated glomerular filtration rate, 5 ml/min/1.73 m2) associated with oliguria. Complete blood count revealed leukocytosis, anemia, and thrombocytopenia. Urinalysis revealed 3+ protein by dipstick. Urine protein:creatinine ratio was 6.7 g/g on a spot measurement. Urinary microscopy, performed on a specimen collected after Foley catheter insertion, showed 21 to 30 red blood cells and 31 to 40 white blood cells per high-power field, but no cellular casts or crystals. Serologic workup results were negative (Table 1). The patient showed no improvement in renal function with volume resuscitation. In light of the additional laboratory findings of hyperuricemia with hyperphosphatemia, hypocalcemia, and elevated lactate dehydrogenase, the patient was started on dialysis and allopurinol for suspected tumor lysis syndrome. A kidney biopsy was performed on the 10th hospital day.Table 1 Initial laboratory findings

on. In light of the additional laboratory findings of hyperuricemia with hyperphosphatemia, hypocalcemia, and elevated lactate dehydrogenase, the patient was started on dialysis and allopurinol for suspected tumor lysis syndrome. A kidney biopsy was performed on the 10th hospital day.Table 1 Initial laboratory findings Parameter Value (reference range) SCr (mg/dl) 10.9 (0.6–1.2) eGFR (ml/min/1.73 m2) 5 (>90) Serum urea nitrogen (mg/dl) 72 (8–20) Serum potassium (mmol/l) 4.4 (3.5–5.5) Serum uric acid (mg/dl) 25 (3.8–8.0) Serum phosphorus (mg/dl) 9.7 (2.4–4.1) Serum calcium (mg/dl) 7.1 (8.5–10.2) Serum albumin (g/dl) 2.5 (3.5–5.1) LDH (IU/l) 1100 (140–280) Hemoglobin (g/dl) 9 (12.0–16.0) WBC count (× 103/μl) 67.4 (4.5–13.5) Differential blood count (%) Neutrophils 48 (40–70) Monocytes 39 (0–10) Lymphocytes 9 (20–50) Eosinophils 0 (0–6) Urine dipstick protein 3+ Urine RBC (/hpf) 21–30 (none)a Urine WBC (/hpf) 31–40 (0–2)a Spot urine PCR (g/g) 6.7 (<0.3) Urine culture No growth C3 (mg/dl) 124 (88–165) C4 (mg/dl) 37 (14–44) ANA Neg (neg) MPO-ANCA <1:20 (<1:20) PR3-ANCA <1:20 (<1:20) Hepatitis C antibody Neg (neg) Anti-GBM antibody Neg (neg) Hepatitis B core antigen Neg (neg) Serum cryoglobulins Neg (neg) SPEP No M-spike ANA, anti−nuclear antibody; ANCA, anti−neutrophil cytoplasmic antibody; anti-GBM, anti−glomerular basement membrane; eGFR, estimated glomerular filtration rate; hpf, high-power field; LDH, lactate dehydrogenase; MPO, myeloperoxidase; Neg, negative; PCR, protein:creatinine ratio; RBC, red blood cell; SCr, serum creatinine; SPEP, serum protein electrophoresis WBC, white blood cell.

antibody; anti-GBM, anti−glomerular basement membrane; eGFR, estimated glomerular filtration rate; hpf, high-power field; LDH, lactate dehydrogenase; MPO, myeloperoxidase; Neg, negative; PCR, protein:creatinine ratio; RBC, red blood cell; SCr, serum creatinine; SPEP, serum protein electrophoresis WBC, white blood cell. Conversion factors for units: SCr in mg/dl to μmol/l, ×88.4; SUN in mg/dl to mmol/l, ×0.357. a Results obtained from Foley catheter−collected urine.

antibody; anti-GBM, anti−glomerular basement membrane; eGFR, estimated glomerular filtration rate; hpf, high-power field; LDH, lactate dehydrogenase; MPO, myeloperoxidase; Neg, negative; PCR, protein:creatinine ratio; RBC, red blood cell; SCr, serum creatinine; SPEP, serum protein electrophoresis WBC, white blood cell. Conversion factors for units: SCr in mg/dl to μmol/l, ×88.4; SUN in mg/dl to mmol/l, ×0.357. a Results obtained from Foley catheter−collected urine. Kidney Biopsy Results The sampling for light microscopy included 16 glomeruli, 2 of which were globally sclerotic. The remaining 14 glomeruli appeared largely unremarkable, and no lesions of focal segmental glomerulosclerosis were identified. The predominant abnormalities involved proximal tubular epithelial cells, which exhibited widespread degenerative changes including luminal ectasia, cytoplasmic simplification and vacuolization, irregular luminal profiles, loss of brush border, enlarged nuclei with prominent nucleoli, and focal apoptotic figures (Figure 1a). The proximal tubular injury was accompanied by mild interstitial edema and mild interstitial inflammation composed of lymphocytes and monocytes. Many proximal tubular cells had hypereosinophilic granular cytoplasm owing to the presence of abundant intracytoplasmic, PAS-positive granules (Figure 1b). Scattered larger rounded, eosinophilic inclusions that were moderately periodic acid–Schiff positive and nonargyrophilic with the Jones methenamine silver stain were also seen (Figure 1h). Immunohistochemical staining for lysozyme (Lysozyme EP134; RabMAb, Rocklin, CA) revealed strong positivity in the distribution of the proximal tubular cell cytoplasm (Figure 1c, d). No atypical casts or intracytoplasmic crystalline-type inclusions were seen. Mild tubular atrophy and interstitial fibrosis involved 15% of the cortex sampled.Figure 1 A low-power view demonstrates widespread proximal tubular degenerative changes and interstitial edema. (a) A glomerulus appears unremarkable (hematoxylin and eosin, original magnification ×200). (b) At higher magnification, many proximal tubular cells are distended by numerous small intracytoplasmic PAS+ granules (arrow; periodic acid–Schiff, original magnification ×400). (c) Immunohistochemial staining for lysozyme shows intense granular reactivity in the distribution of proximal tubular cell cytoplasm (immunoperoxidase, original magnification ×100). (d) No significant staining is seen in a paired negative control obtained from an allograft postreperfusion biopsy (immunoperoxidase, original magnification ×100).

ining for lysozyme shows intense granular reactivity in the distribution of proximal tubular cell cytoplasm (immunoperoxidase, original magnification ×100). (d) No significant staining is seen in a paired negative control obtained from an allograft postreperfusion biopsy (immunoperoxidase, original magnification ×100). (e) On ultrastructural evaluation, proximal tubular cells contain abundant membrane-bound vacuoles (arrow; original magnification ×8000). (f) On closer inspection, the vacuoles in proximal tubules contain clumped, degenerating organellar debris, consistent with autophagolysosomes (arrow; original magnification ×25,000). (g) In rare cells, the autophagolysomes appear to form larger, membrane-bound aggregates (arrow; original magnification ×6,000), corresponding to the (h) scattered large round eosinophilic inclusions seen in a minority of cells (arrow; hematoxylin and eosin, original magnification ×600). Immunofluorescence staining for IgG, IgM, IgA, C3, C1, fibrinogen, albumin, and kappa (κ) and lambda (λ) light chains revealed no significant positivity. Immunofluorescence staining for κ and λ light chains repeated on pronase-digested paraffin tissue sections again was negative, providing evidence against light chain proximal tubuloapthy.

aining for IgG, IgM, IgA, C3, C1, fibrinogen, albumin, and kappa (κ) and lambda (λ) light chains revealed no significant positivity. Immunofluorescence staining for κ and λ light chains repeated on pronase-digested paraffin tissue sections again was negative, providing evidence against light chain proximal tubuloapthy. On ultrastructural evaluation, glomeruli exhibited no significant abnormalities. Specifically, there was only 10% foot process effacement, and no electron-dense deposits were seen. Proximal tubules displayed diffuse degenerative changes including loss of apical brush border, cytoplasmic simplification, dilatation of the endoplasmic reticulum, and intraluminal cellular debris. The most distinctive finding was abundant membrane-bound cytoplasmic vacuoles containing clumped, electron-dense, degenerating cellular organellar debris, consistent with autophagolysosomes (Figure 1e, f). The autophagolysosomes were diffusely distributed in proximal tubular epithelia and in some instances formed larger, membrane-bound aggregates that approached the size of the nucleus (Figure 1g). No intracellular crystals or dysmorphic mitochondria were identified. Diagnosis A diagnosis of lysozyme-induced nephropathy (LyN) was made.

On ultrastructural evaluation, glomeruli exhibited no significant abnormalities. Specifically, there was only 10% foot process effacement, and no electron-dense deposits were seen. Proximal tubules displayed diffuse degenerative changes including loss of apical brush border, cytoplasmic simplification, dilatation of the endoplasmic reticulum, and intraluminal cellular debris. The most distinctive finding was abundant membrane-bound cytoplasmic vacuoles containing clumped, electron-dense, degenerating cellular organellar debris, consistent with autophagolysosomes (Figure 1e, f). The autophagolysosomes were diffusely distributed in proximal tubular epithelia and in some instances formed larger, membrane-bound aggregates that approached the size of the nucleus (Figure 1g). No intracellular crystals or dysmorphic mitochondria were identified. Diagnosis A diagnosis of lysozyme-induced nephropathy (LyN) was made. Clinical Follow-up Additional testing revealed markedly elevated serum lysozyme levels (101 μg/ml; nl range 2.7–9.4) and lysozymuria (>11 μg/ml; nl <3), supporting a diagnosis of lysozyme-induced nephropathy. The patient was started on chemotherapy with azacitidine, resulting in normalization of his white blood cell count. Serum lysozyme levels remained elevated 4 months after initiation of chemotherapy, but ultimately decreased to within reference range. The patient remained dialysis dependent from the time of kidney biopsy, and expired 18 months later in the setting of relapsing leukemia with blast crisis.

f his white blood cell count. Serum lysozyme levels remained elevated 4 months after initiation of chemotherapy, but ultimately decreased to within reference range. The patient remained dialysis dependent from the time of kidney biopsy, and expired 18 months later in the setting of relapsing leukemia with blast crisis. Discussion CMML is a rare, aggressive, malignant, hematopoietic neoplasm that most commonly affects patients over the age of 65 years and is characterized by peripheral blood monocytosis with myelodysplastic features. Lysozyme, also known as muramidase, is a lytic enzyme with bactericidal properties that is synthesized by monocytes and can be produced in large amounts by neoplastic cells of monocyte lineage.7 Increased serum and urine lysozyme levels (lysozymuria) were first described in patients with monocytic or myelomonocytic leukemia, including both acute and chronic forms, in the late 1960s.8 Lysozyme-induced nephropathy (LyN) is a rare and underrecognized cause of AKI in a subset of these patients, particularly those with CMML.3, 7, 9

nd urine lysozyme levels (lysozymuria) were first described in patients with monocytic or myelomonocytic leukemia, including both acute and chronic forms, in the late 1960s.8 Lysozyme-induced nephropathy (LyN) is a rare and underrecognized cause of AKI in a subset of these patients, particularly those with CMML.3, 7, 9 Lysozyme is a small cationic protein (molecular weight, 15 kDa) that is freely filtered by the glomerulus.8 Lysozyme is reabsorbed in the proximal tubule, where it is taken up by endocytosis and catabolized in phagolysosomes.10 Despite the proximal tubules’ high absorptive capacity for lysozyme,11, 12 marked overproduction of lysozyme in patients with monocytic leukemias may exceed the transport maximum, leading to nephrotic-range nonalbumin proteinuria, as appears to have occurred in this case.13, 14 By protein electrophoresis, lysozyme migrates in the gamma (γ) region. Therefore, the presence of nonalbumin proteinuria with an increased γ-globulin fraction but without detectable monoclonal bands by immunofixation electrophoresis may be a useful indicator of lysozymuria in the appropriate clinical context.15 Unfortunately, these studies were not performed.

ates in the gamma (γ) region. Therefore, the presence of nonalbumin proteinuria with an increased γ-globulin fraction but without detectable monoclonal bands by immunofixation electrophoresis may be a useful indicator of lysozymuria in the appropriate clinical context.15 Unfortunately, these studies were not performed. In the setting of monocytic and myelomonocytic leukemias, including chronic and acute forms, a steady proportion of the filtered load of lysozyme accumulates in proximal tubular cells, allowing the kidney to act as a reservoir for circulating lysozyme.12 There appears to be a threshold above which the concentration of lysozyme, a lytic enzyme with bactericidal properties,7 becomes toxic to proximal tubular cells, leading to the development of acute tubular injury and AKI. Indeed, lysozymuria can impair proximal tubular cell function,16 leading to renal insufficiency, hypokalemia secondary to renal potassium wasting, and low-level tubular albuminuria, but not Fanconi syndrome.9, 17 However, elevated serum and urine lysozyme levels commonly occur in patients with monocytic and myelomonocytic leukemias and are usually not associated with the development of AKI.16, 18 Therefore, kidney biopsy is needed to establish a definitive diagnosis of LyN.

el tubular albuminuria, but not Fanconi syndrome.9, 17 However, elevated serum and urine lysozyme levels commonly occur in patients with monocytic and myelomonocytic leukemias and are usually not associated with the development of AKI.16, 18 Therefore, kidney biopsy is needed to establish a definitive diagnosis of LyN. Descriptions of the pathologic findings in LyN are limited. An early report described tubular degenerative changes accompanied by hyaline droplets in proximal tubules, but also described hyaline droplets in patients with monocytic or myelomonocytic leukemia and intact renal function.16 A more recent report described coarse protein granules in proximal tubules by light microscopy, corresponding with the ultrastructural finding of large and prominent but relatively isomorphic lysosomes.7 In the case reported herein, the direct toxicity of lysozyme was associated with the distinctive ultrastructural finding of abundant autophagolysosomes containing degenerating organellar debris and strong immunostaining of proximal tubules for lysozyme. Similar ultrastructural findings have been described in rats transplanted with chloroleukemia, a myelomonoblastic tumor that also secretes large amounts of lysozyme, leading to lysozymuria.19

bundant autophagolysosomes containing degenerating organellar debris and strong immunostaining of proximal tubules for lysozyme. Similar ultrastructural findings have been described in rats transplanted with chloroleukemia, a myelomonoblastic tumor that also secretes large amounts of lysozyme, leading to lysozymuria.19 Other causes of AKI should be excluded in patients with CMML. Laboratory evaluation of our patient also revealed hyperuricemia, mild hyperphosphatemia, and mild hypocalcemia, raising the possibility of concurrent tumor lysis syndrome. However, no intratubular deposition of uric acid crystals or calcium phosphate salts was seen on kidney biopsy. Nonetheless, elevated serum uric acid levels may still have contributed to AKI in our patient, as hyperuricemia may reduce renal plasma flow and may inhibit proximal tubular cell proliferation.20 Although unlikely in the case reported herein, alternative etiologies of lysozymuria should be considered. Heavy lysozymuria may occur rarely as a form of overflow proteinuria in the setting of other disease states associated with macrophage activation and overproduction of lysozyme (including sarcoidosis) or at very low levels in the setting of tubular proteinuria related to proximal tubular cell dysfunction.7, 9

nsidered. Heavy lysozymuria may occur rarely as a form of overflow proteinuria in the setting of other disease states associated with macrophage activation and overproduction of lysozyme (including sarcoidosis) or at very low levels in the setting of tubular proteinuria related to proximal tubular cell dysfunction.7, 9 The treatment of CMML is initially supportive; however, the development of leukemia-associated organ damage may be an indication for more aggressive treatment, including cytoreductive therapy or use of hypomethylating agents.21 Variable improvement in kidney function has been described in patients receiving CMML-specific chemotherapy who have direct leukemic infiltration or overproduction of lysozyme by the malignant cells.22 Although our patient had a good hematologic response to chemotherapy, he remained dialysis dependent.

lating agents.21 Variable improvement in kidney function has been described in patients receiving CMML-specific chemotherapy who have direct leukemic infiltration or overproduction of lysozyme by the malignant cells.22 Although our patient had a good hematologic response to chemotherapy, he remained dialysis dependent. Conclusion In conclusion, acute and chronic monocytic and myelomonocytic neoplasms, most notably CMML, are associated with overproduction of lysozyme, resulting in elevated serum and urine levels. Lysozyme is freely filtered by the glomerulus, and can be associated with nephrotic-range lysozymuria. Lysozyme accumulates in proximal tubular cells, and there is a threshold at which this accumulation is associated with renal potassium wasting, toxic proximal tubular injury, and AKI. LyN is a rare and likely underrecognized etiology of AKI that has a distinctive pathologic appearance characterized by acute tubular necrosis, with increased cytoplasmic granules in proximal tubular cells at the light microscopic level, corresponding to abundant phagolysosomes containing partially digested organellar debris at the ultrastructural level, with strong proximal tubular cytoplasmic staining for lysozyme (Table 2). Recognition of this rare etiology of AKI may guide chemotherapeutic intervention in the management of acute and chronic monocytic and myelomonocytic leukemias.Table 2 Lysozyme-induced nephropathy (LyN): teaching points • Lysozyme is a small cationic protein produced by monocytes that is freely filtered by the glomerulus and reabsorbed by proximal tubules.

Conclusion In conclusion, acute and chronic monocytic and myelomonocytic neoplasms, most notably CMML, are associated with overproduction of lysozyme, resulting in elevated serum and urine levels. Lysozyme is freely filtered by the glomerulus, and can be associated with nephrotic-range lysozymuria. Lysozyme accumulates in proximal tubular cells, and there is a threshold at which this accumulation is associated with renal potassium wasting, toxic proximal tubular injury, and AKI. LyN is a rare and likely underrecognized etiology of AKI that has a distinctive pathologic appearance characterized by acute tubular necrosis, with increased cytoplasmic granules in proximal tubular cells at the light microscopic level, corresponding to abundant phagolysosomes containing partially digested organellar debris at the ultrastructural level, with strong proximal tubular cytoplasmic staining for lysozyme (Table 2). Recognition of this rare etiology of AKI may guide chemotherapeutic intervention in the management of acute and chronic monocytic and myelomonocytic leukemias.Table 2 Lysozyme-induced nephropathy (LyN): teaching points • Lysozyme is a small cationic protein produced by monocytes that is freely filtered by the glomerulus and reabsorbed by proximal tubules. • Overproduction of lysozyme in patients with chronic monocytic or myelomonocytic leukemia may lead to nephrotic-range proteinuria (lysozymuria). • Lysozyme-induced nephropathy is a rare cause of acute kidney injury in patients with chronic monocytic or myelomonocytic leukemia.

• Lysozyme is a small cationic protein produced by monocytes that is freely filtered by the glomerulus and reabsorbed by proximal tubules. • Overproduction of lysozyme in patients with chronic monocytic or myelomonocytic leukemia may lead to nephrotic-range proteinuria (lysozymuria). • Lysozyme-induced nephropathy is a rare cause of acute kidney injury in patients with chronic monocytic or myelomonocytic leukemia. • Lysozyme-induced nephropathy is characterized by acute tubular injury with abundant cytoplasmic granular inclusions that stain strongly for lysozyme and have an ultrastructural appearance consistent with that of autophagolysosomes. Disclosure All the authors declared no competing interests.

Acute kidney injury (AKI) is a common complication in hospitalized patients, and has been consistently associated with an increased risk of death, de novo or worsening chronic kidney disease (CKD), and end-stage renal disease (ESRD).1, 2, 3, 4, 5 Patients discharged after an episode of AKI have a 40% increased risk of death in the 2 years following hospitalization,6 and a 50% to 60% increased risk of cardiovascular events, compared to patients who do not develop AKI.7 There are currently no effective therapies targeting established AKI; however, identifying and treating patients with CKD following an episode of AKI may improve health outcomes. Although recently published data suggested that nephrologist follow-up was associated with a 24% reduction in risk of death after hospitalization with severe AKI requiring dialysis,8 little is known about processes of care that modify outcomes after episodes of AKI. Statins are effective for reducing cardiovascular morbidity and mortality in patients with CKD.8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 We sought to evaluate whether the use of statins was associated with better outcomes among patients with CKD after AKI.

Acute kidney injury (AKI) is a common complication in hospitalized patients, and has been consistently associated with an increased risk of death, de novo or worsening chronic kidney disease (CKD), and end-stage renal disease (ESRD).1, 2, 3, 4, 5 Patients discharged after an episode of AKI have a 40% increased risk of death in the 2 years following hospitalization,6 and a 50% to 60% increased risk of cardiovascular events, compared to patients who do not develop AKI.7 There are currently no effective therapies targeting established AKI; however, identifying and treating patients with CKD following an episode of AKI may improve health outcomes. Although recently published data suggested that nephrologist follow-up was associated with a 24% reduction in risk of death after hospitalization with severe AKI requiring dialysis,8 little is known about processes of care that modify outcomes after episodes of AKI. Statins are effective for reducing cardiovascular morbidity and mortality in patients with CKD.8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 We sought to evaluate whether the use of statins was associated with better outcomes among patients with CKD after AKI. Methods Study Population and Data Sources We used the Alberta Kidney Disease Network population-based database, which has been described in detail elsewhere.19 The study cohort20 included all adults 18 years or older residing in Alberta who were admitted to the hospital between 1 July 2008 and 31 March 2011 and had an episode of AKI during hospitalization. To be eligible for inclusion, patients had to have at least 1 outpatient serum creatinine (Scr) measurement within 180 days prior to hospitalization to establish baseline kidney function, ≥ 1 measurement during the hospitalization to establish AKI, and ≥ 1 SCr, urine dipstick (udip), urine albumin to serum creatinine ratio (ACR), or urine protein to serum creatinine ratio (PCR) measurement in the follow-up period after hospital discharge to establish CKD. If participants had more than 1 hospitalization during this period, only the first was considered (index hospitalization). Participants who died or progressed to ESRD (estimated GFR [eGFR] < 15 ml/min per 1.73 m2, chronic dialysis, prior kidney transplantation) before or during the index hospitalization were excluded. The cohort was restricted to patients who had CKD after an AKI episode. All subjects were followed up from the discharge date of their index hospitalization until 31 March 2013, with a minimum follow-up of 2 years.

chronic dialysis, prior kidney transplantation) before or during the index hospitalization were excluded. The cohort was restricted to patients who had CKD after an AKI episode. All subjects were followed up from the discharge date of their index hospitalization until 31 March 2013, with a minimum follow-up of 2 years. The study was reviewed and approved by the institutional review boards at the Universities of Alberta and Calgary. Assessment of Baseline Kidney Function The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation was used to determine the estimated glomerular filtration rate (eGFR).21 Baseline kidney function was defined as the mean outpatient Scr in the 180 days prior to the index hospitalization. Identification of Acute Kidney Injury AKI was identified by changes between baseline (pre-hospital) and peak in-hospital Scr. AKI was defined as an increase in serum creatinine by 50% or greater within 7 days or by 26.5 μmol/L within 48 hours and/or requirement for acute dialysis within 7 days of the index hospitalization. AKI severity was determined using the consensus criteria for AKI staging from the recently published Kidney Disease Improving Global Outcomes (KDIGO) AKI guidelines.22 Requirement for acute dialysis for AKI was determined using a validated approach, based on diagnosis and procedural administrative codes.23

italization. AKI severity was determined using the consensus criteria for AKI staging from the recently published Kidney Disease Improving Global Outcomes (KDIGO) AKI guidelines.22 Requirement for acute dialysis for AKI was determined using a validated approach, based on diagnosis and procedural administrative codes.23 Assessment of CKD After AKI The presence of CKD was assessed using Scr, urine ACR, urine PCR, or udip measured 90 days or more after the AKI episode to allow sufficient time for recovery of renal function (Supplementary Figure S1). A 90-day time frame for recovery was chosen based on the Kidney Disease Outcomes Quality Initiative guidelines,24 which define CKD as a persistent decline in kidney function lasting >90 days. If subjects had more than 1 outpatient SCr measurement, the measurement closest to 90 days was considered. Postdischarge CKD was defined as eGFR < 60 ml/min/1.73 m2, ACR > 30 mg/g, PCR > 150 mg/g, or Udip > trace, consistent with the KDIGO CKD guidelines.25

D as a persistent decline in kidney function lasting >90 days. If subjects had more than 1 outpatient SCr measurement, the measurement closest to 90 days was considered. Postdischarge CKD was defined as eGFR < 60 ml/min/1.73 m2, ACR > 30 mg/g, PCR > 150 mg/g, or Udip > trace, consistent with the KDIGO CKD guidelines.25 Assessment of Medication Use After Discharge Prescription drug information was obtained from the Pharmaceutical Information Network (PIN) database. We classified statin exposure into the following groups: no previous prescription, new prescription (defined as at least 1 prescription within 2 years after discharge from the index hospitalization), stopping a previous prescription, and continuing a prescription. Patients were classified in the continuing prescription group if they had at least 1 prescription in the 6 months prior to admission and at least 1 prescription within 2 years postdischarge. High-dose statin was defined as the highest dosage for each statin drug (Supplementary Table S1).

and continuing a prescription. Patients were classified in the continuing prescription group if they had at least 1 prescription in the 6 months prior to admission and at least 1 prescription within 2 years postdischarge. High-dose statin was defined as the highest dosage for each statin drug (Supplementary Table S1). Assessment of Comorbid Conditions Relevant demographic characteristics, preexisting comorbid conditions (defined using validated algorithms),20, 26 hospitalizations and outpatient physician visits (general practitioner as well as specialist visits), details of the index hospitalization including primary admission diagnosis, and intensive care unit stay were obtained using hospitalization data, claims files, and ambulatory care classification system files. We obtained primary International Classification of Diseases (10th revision) codes and used these to classify primary admission diagnoses using a previously published approach.26 Resource intensity weight, similar to diagnostic related group weight, was used to categorize acuity and severity of illness.27, 28 Cholesterol level was defined as the mean outpatient total cholesterol in the 1 year prior to the index hospitalization. The cholesterol levels were classified into 5 risk categories according to the Framingham coronary heart disease risk score.29 Patients who did not have a cholesterol measurement during this time period were classified in an unknown group.

n outpatient total cholesterol in the 1 year prior to the index hospitalization. The cholesterol levels were classified into 5 risk categories according to the Framingham coronary heart disease risk score.29 Patients who did not have a cholesterol measurement during this time period were classified in an unknown group. Outcomes The primary outcome was mortality; secondary outcomes included all-cause re-hospitalization and cardiovascular events. All outcomes were assessed after the discharge date of the index hospitalization. All-cause re-hospitalization was defined as any hospitalization occurring after the index hospitalization. All hospitalizations were identified using AH data. All-cause mortality was identified using administrative data sources. Cardiovascular events were defined as myocardial infarction, stroke, or revascularization procedure, as defined by validated algorithms.30, 31, 32 Statistical Analysis Continuous variables were described using the mean and SD or the median with interquartile range as appropriate. Categorical variables were described as proportions of the cohort.

Outcomes The primary outcome was mortality; secondary outcomes included all-cause re-hospitalization and cardiovascular events. All outcomes were assessed after the discharge date of the index hospitalization. All-cause re-hospitalization was defined as any hospitalization occurring after the index hospitalization. All hospitalizations were identified using AH data. All-cause mortality was identified using administrative data sources. Cardiovascular events were defined as myocardial infarction, stroke, or revascularization procedure, as defined by validated algorithms.30, 31, 32 Statistical Analysis Continuous variables were described using the mean and SD or the median with interquartile range as appropriate. Categorical variables were described as proportions of the cohort. Multivariable Cox proportional hazard models with time-varying covariates were used to estimate the association between use of statin following index hospitalization and all-cause mortality, re-hospitalization, and cardiovascular events. The adjusted models included terms for age, sex, income quintile, aboriginal race, location of residency, health care use preceding the index hospitalization, Canadian Institute for Health Information resource intensity weight, intensive care unit admission during the index hospitalization, primary diagnostic code for hospitalization, procedures associated with AKI (cardiac catheterization, cardiac and abdominal aortic surgery), comorbid conditions, baseline kidney function (based on eGFR), cholesterol level, statin, angiotensin-converting enzyme inhibitor (ACEi), angiotensin receptor blocker (ARB), and β-blocker use in the 6 months preceding admission and following discharge (Supplementary Figure S2). Time of origin was the discharge date for the index hospitalization. The proportional hazard assumption was evaluated and satisfied by examining plots of the log-negative-log within-group survivorship functions versus log-time.

r use in the 6 months preceding admission and following discharge (Supplementary Figure S2). Time of origin was the discharge date for the index hospitalization. The proportional hazard assumption was evaluated and satisfied by examining plots of the log-negative-log within-group survivorship functions versus log-time. First use of statin was used to update the exposure status during the course of the follow-up (i.e., a person on a statin would contribute person time to the “no statin use” before the first statin was prescribed and contribute person time to the “statin use” group after the first statin was prescribed). Patients were censored if they moved out of the province or reached the end of the study date (31 March 2013) for all outcomes. For the secondary outcomes, patients were censored if they died.

re the first statin was prescribed and contribute person time to the “statin use” group after the first statin was prescribed). Patients were censored if they moved out of the province or reached the end of the study date (31 March 2013) for all outcomes. For the secondary outcomes, patients were censored if they died. Analyses were repeated after further categorizing statin users into the following groups: new prescription postdischarge, continuing a previous prescription within 2 years postdischarge, and stopping use of a pre-hospital admission prescription. In the sensitivity analysis, we excluded patients who required acute dialysis to determine whether associations were similar in patients with less severe AKI. We also performed a sensitivity analysis excluding patients with proteinuria at any time during the study period to assess whether associations were similar in patients with non-proteinuric CKD. We repeated analyses after excluding patients who had any pre-existing cardiovascular disease, to examine findings in patients less likely to have pre-existing established cardiovascular disease. We compared the effect of statin use after stratifying the cohort according to whether patients had pre-existing CKD or de novo CKD (post-AKI) patients. We also compared outcomes in patients on high-dose versus low-dose statins.

e findings in patients less likely to have pre-existing established cardiovascular disease. We compared the effect of statin use after stratifying the cohort according to whether patients had pre-existing CKD or de novo CKD (post-AKI) patients. We also compared outcomes in patients on high-dose versus low-dose statins. Results Patient Characteristics Between 1 July 2008 and 31 March 2011, there were 35,086 patients 18 years of age or older residing in Alberta with hospital-acquired AKI (Figure 1). Of these participants, 26,724 (76.2%) survived to 90 days after their AKI episode. The study cohort included 19,707 patients with CKD after the AKI episode: 12,148 (61.6%) had prior CKD and 7,559 (38.4%) had de novo CKD (Figure 1). The mean patient age was 69.9 years, 52.5% were men, and 83.7% lived in an urban location (Table 1). The mean number of hospitalizations during the 3 years preceding the index hospitalization was 1.7 (interquartile range [IQR] 0−2), and 18.4% of the cohort had a cardiovascular diagnostic code as the diagnosis most responsible for the index hospitalization. The majority of participants (80.9%) had hypertension, and a large number of patients had diabetes (42.6%), chronic heart failure (33.4%), and history of stroke or TIA (23.1%).Figure 1 Selection of study population after episode of hospital-acquired acute kidney injury (AKI). CKD, chronic kidney disease; ESRD, end-stage renal disease. Table 1 Characteristics of statin users and non−statin users

Results Patient Characteristics Between 1 July 2008 and 31 March 2011, there were 35,086 patients 18 years of age or older residing in Alberta with hospital-acquired AKI (Figure 1). Of these participants, 26,724 (76.2%) survived to 90 days after their AKI episode. The study cohort included 19,707 patients with CKD after the AKI episode: 12,148 (61.6%) had prior CKD and 7,559 (38.4%) had de novo CKD (Figure 1). The mean patient age was 69.9 years, 52.5% were men, and 83.7% lived in an urban location (Table 1). The mean number of hospitalizations during the 3 years preceding the index hospitalization was 1.7 (interquartile range [IQR] 0−2), and 18.4% of the cohort had a cardiovascular diagnostic code as the diagnosis most responsible for the index hospitalization. The majority of participants (80.9%) had hypertension, and a large number of patients had diabetes (42.6%), chronic heart failure (33.4%), and history of stroke or TIA (23.1%).Figure 1 Selection of study population after episode of hospital-acquired acute kidney injury (AKI). CKD, chronic kidney disease; ESRD, end-stage renal disease. Table 1 Characteristics of statin users and non−statin users All subjects Statin users Non−statin users P value Number of subjects (%) 19,707 7539 (38.3) 12,168 (61.7) Age, yr, mean (SD) 69.9 (14.9) 70.7 (12) 69.3 (16.5) <0.01 Sex, male (%) 52.5 57.6 49.4 <0.01 Aboriginal (%) 3.8 3.2 4.2 <0.01 Income quintile 0.59 Lowest (level = 1) (%) 24.9 24.6 25.1 Highest (level = 5) (%) 15.9 15.6 16.0 Urban location (%) 83.7 84.3 83.2 0.04 Healthcare access 3-year preceding hospital admission, mean (median, IQR) Number of hospitalizations 1.7 (1, 0–2) 1.7 (1, 0–2) 1.8 (1, 0–2) 0.02 Number of GP visits 50.8 (36, 21–64) 48.2 (37, 22–61) 52.4 (36, 20–66) <0.01 Number of nephrologist visits 1.1 (0, 0–0) 1.2 (0, 0–0) 1 (0, 0–0) <0.01 Number of cardiology visits 3.1 (0, 0–2) 4.6 (0, 0–5) 2.2 (0, 0–1) <0.01 Number of internist visits 7.2 (3, 0–8) 7.7 (3, 1–9) 6.9 (2, 0–8) <0.01 Number of emergency visits 5.8 (3, 1–6) 5.5 (3, 1–6) 6 (3, 1–7) <0.01 CIHI resource intensity weight, mean (SD) 3.1 (6.3) 2.7 (4.4) 3.4 (7.2) <0.01 Intensive care unit during hospitalization (%) 18.9 24.8 15.3 <0.01 Primary diagnostic code for hospitalization (%) Cardiovascular 18.4 26.0 13.6 <0.01 Respiratory 9.1 8.9 9.3 0.245 Gastrointestinal 10.3 8.4 11.4 <0.01 Infectious disease 4.6 3.8 5.1 <0.01 Cancer 7.6 5.4 9.0 <0.01 Orthopedic 4.6 4.9 4.4 0.09 Hematologic 5.8 6.2 5.5 0.04 Genitourinary 11.1 10.4 11.5 0.02 Injury/poisoning 5.1 4.5 5.4 <0.01 Other disease 23.5 21.4 24.7 <0.01 Procedure or condition during index hospitalization (%) Sepsis 4.8 3.8 5.4 <0.01 Cardiac surgery 2.7 4.7 1.5 <0.01 Cardiac catheterization 3.7 6.5 1.9 <0.01 Abdominal aortic aneurysm repair 0.5 0.8 0.3 <0.01 Pneumonia 10.7 10.0 11.2 0.01 Liver failure 0.7 0.2 1.0 <0.01 Acute myocardial infraction 10.1 16.3 6.3 <0.01 Noncardiac surgery 17.0 15.3 18.1 <0.01 Comorbid disease (%) Diabetes 42.6 55.7 34.5 <0.01 Hypertension 80.9 90.5 74.9 <0.01 Myocardial infarction 13.0 19.8 8.8 <0.01 Chronic heart failure 33.4 38.3 30.4 <0.01 Stroke or TIA 23.1 25.9 21.4 <0.01 Cancer 9.0 7.8 9.6 <0.01 Liver disease 2.4 1.0 3.2 <0.01 Peripheral vascular disease 7.2 10.0 5.4 <0.01 Kidney function Baseline eGFR, ml/min/1.73 m2, mean (SD) 62.3 (25.8) 58.8 (23.1) 64.4 (27.1) <0.01 Prior CKD (%) 61.6 66.2 58.8 <0.0

<0.01 Chronic heart failure 33.4 38.3 30.4 <0.01 Stroke or TIA 23.1 25.9 21.4 <0.01 Cancer 9.0 7.8 9.6 <0.01 Liver disease 2.4 1.0 3.2 <0.01 Peripheral vascular disease 7.2 10.0 5.4 <0.01 Kidney function Baseline eGFR, ml/min/1.73 m2, mean (SD) 62.3 (25.8) 58.8 (23.1) 64.4 (27.1) <0.01 Prior CKD (%) 61.6 66.2 58.8 <0.0 1 Prior CKD defined by eGFR 33.3 34.0 32.9 <0.01 Prio CKD defined by proteinuria 11.6 11.8 11.5 0.02 Prior CKD defined by eGFR and proteinuria 16.7 20.3 14.5 <0.01 AKI stage (%) <0.01 AKI stage 1 75.9 77.9 74.6 <0.01 AKI stage 2 14.8 13.5 15.6 <0.01 AKI stage 3 (no dialysis) 7.1 6.2 7.6 <0.01 Dialysis 2.3 2.4 2.2 0.37 Baseline total cholesterol, mmol/l <0.01 <4.1 29.7 40.6 23.0 <0.01 4.15−5.17 18.7 20.6 17.5 <0.01 5.18−6.21 10.2 10.1 10.3 0.67 6.22−7.24 3.3 3.6 3.1 0.05 ≥7.25 1.3 1.9 1.0 <0.01 Unknown 37 23 45 <0.01 AKI, acute kidney injury; CIHI, Canadian Institute for Health Information; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; TIA, transient ischemic attack. Primary Outcomes More than one-half (54.4%) of the cohort were never prescribed a statin, 12.0% received a new prescription, and 26.3% were continued on a statin within 2 years after hospital discharge. Only 7.3% of previous statin prescriptions were not restarted after hospital discharge (Table 2).Table 2 Number of patients (%) who were using statin before and after index hospitalization

escribed a statin, 12.0% received a new prescription, and 26.3% were continued on a statin within 2 years after hospital discharge. Only 7.3% of previous statin prescriptions were not restarted after hospital discharge (Table 2).Table 2 Number of patients (%) who were using statin before and after index hospitalization Within 6 months before admission 2 yr After discharge Statin 6620 (33.6%) 7539 (38.3%) Never prescribed New prescription Stopping previous prescription Continuing prescription Statin 10,729 (54.4%) 2358 (12.0%) 1439 (7.3%) 5181 (26.3%) Over a total of 53,700 person-years of follow up, the adjusted hazard ratio for mortality associated with statin use after hospital discharge, compared with no statin use, was 0.74 (95% confidence interval [CI], 0.69, 0.79). Statin use after hospitalization was also associated with a lower risk of all cause re-hospitalization (hazard ratio [HR], 0.90; 95% CI, 0.85, 0.94) (Table 3).Table 3 Hazard ratios of statin use after hospital discharge Outcome Adjusted hazard ratio (95% CI) Number of events Follow-up time in person-yr Crude hazard ratio (95% CI) Survival 0.74 (0.69–0.79) 6758 53700.94 0.75 (0.71–0.79) All cause re-hospitalization 0.90 (0.85–0.94) 15,256 25144.77 0.94 (0.9–0.97) Cardiovascular event 0.95 (0.87–1.04) 3493 48764.46 1.54 (1.43–1.65) CI, confidence interval.

ard ratio (95% CI) Number of events Follow-up time in person-yr Crude hazard ratio (95% CI) Survival 0.74 (0.69–0.79) 6758 53700.94 0.75 (0.71–0.79) All cause re-hospitalization 0.90 (0.85–0.94) 15,256 25144.77 0.94 (0.9–0.97) Cardiovascular event 0.95 (0.87–1.04) 3493 48764.46 1.54 (1.43–1.65) CI, confidence interval. Adjusted factors: Angiotensin-converting enzyme inhibitor (ACEI)/angiotensin receptor blocker (ARB), β-blocker, and statin use within 6 months before admission, ACEI/ARB and β-blocker after discharge, age, sex, income quintile, urban location, health care use 3 years before hospital admission, Canadian Institute for Health Information resource intensity weight, intensive care unit, primary diagnostic code for hospitalization, procedure or condition during index hospitalization, comorbid disease, baseline kidney function (estimated glomerular filtration rate), and total cholesterol risk categories. Separate Cox proportional hazard models with medication use as time-varying covariates were fit for the outcome of all-cause mortality, all-cause re-hospitalization, and cardiovascular events. All patients were followed up for at least 2 years starting at the date of hospital discharge, with further censoring for death, outmigration from Alberta, and the end of study (31 March 2013) in the model fit of all-cause re-hospitalization and cardiovascular events, and for outmigration from Alberta and end of study in the model fit for mortality.

or at least 2 years starting at the date of hospital discharge, with further censoring for death, outmigration from Alberta, and the end of study (31 March 2013) in the model fit of all-cause re-hospitalization and cardiovascular events, and for outmigration from Alberta and end of study in the model fit for mortality. Both a new statin prescription (HR, 0.73; 95% CI, 0.66, 0.80) and continuing a previous prescription (HR, 0.76; 95% CI, 0.71, 0.81) after hospital discharge was associated with better survival compared to no statin use. There was also lower all-cause re-hospitalization in both patients who were given a new statin prescription or continued on a previous statin prescription after hospital discharge, compared to no statin use (Table 4).Table 4 Hazard ratios of statin use for never prescribed, new prescription, stopping previous prescription, or continuing previous prescription