Browse the corpus

Introduction Secondary prevention following an acute coronary syndrome (ACS) event is key as further ischemic events are common following the index event. Risk prediction tools have identified a number of factors which impact on risk of death and myocardial infarction (MI) following an ACS event. However, patient prognosis at hospital discharge continues to vary markedly, and post-discharge mortality remains a concern.1 Most risk scores include hospital mortality in their estimations.2–4 There are no tools for risk calculation of one-year mortality in hospital survivors. It is usually at the time of discharge that patients are asking about their prognosis. Therefore, there is a need for a reliable prediction tool to identify patients with high mortality risk, which may ultimately allow tailored treatment decisions and improve prognosis. For instance, patients identified as at high risk may receive more frequent follow-up visits to facilitate their optimal care. For patients experiencing an acute coronary event, a crucial time to assess their prognosis and future management is at discharge from hospital. Hence, there is merit in developing a risk model that utilizes all the patient data on demographics, medical history, and patient status at, and during, admission, and at discharge. From a large representative international cohort study of consecutive patients with ACS who survived to discharge, we have related such detailed patient records to their subsequent follow-up for one year, expressing prognosis in terms of one-year mortality.

nd patient status at, and during, admission, and at discharge. From a large representative international cohort study of consecutive patients with ACS who survived to discharge, we have related such detailed patient records to their subsequent follow-up for one year, expressing prognosis in terms of one-year mortality. While ST segment elevation myocardial infarction (STEMI) and non-ST-elevation ACS (NSTE-ACS) patients have very different management and prognosis patterns during the in-hospital phase, from the moment of hospital discharge there is sufficient common ground and similarity of the key risk factors to combine both sets of patients into a single overall risk model. There is an extensive literature on risk scores in ACS,5 and their use is advocated by the European Society of Cardiology (ESC) guidelines for the management of NSTE-ACS.6 However, relatively little attention has been paid to risk assessment at hospital discharge, with just one previous risk score to date regarding six-month mortality post discharge.7 This is a valuable opportunity to quantify individual patient risk of mortality to one year after discharge following an acute coronary event hospitalization. Methods EPICOR (long-tErm follow uP of antithrombotic management patterns In acute CORonary syndrome patients) is a prospective, international, observational, real-world practice, cohort study (NCT01171404) comprising consecutive patients, hospitalized for ACS within 24 h of symptom onset, who survived to hospital discharge.

s EPICOR (long-tErm follow uP of antithrombotic management patterns In acute CORonary syndrome patients) is a prospective, international, observational, real-world practice, cohort study (NCT01171404) comprising consecutive patients, hospitalized for ACS within 24 h of symptom onset, who survived to hospital discharge. In total, 10,568 patients with non-fatal ACS who survived until hospital discharge were enrolled between September 2010–March 2011 from 555 hospitals in 20 countries across Europe and Latin America. A detailed account of the methodology of the study is described elsewhere.8,9 For external validation of our risk model, we used data from the EPICOR-Asia study10 (NCT01361386) which enrolled 12,993 patients from eight Asian countries from June 2011–April 2012. This Asian study has followed an almost identical protocol and case record forms as in our EPICOR study.

In total, 10,568 patients with non-fatal ACS who survived until hospital discharge were enrolled between September 2010–March 2011 from 555 hospitals in 20 countries across Europe and Latin America. A detailed account of the methodology of the study is described elsewhere.8,9 For external validation of our risk model, we used data from the EPICOR-Asia study10 (NCT01361386) which enrolled 12,993 patients from eight Asian countries from June 2011–April 2012. This Asian study has followed an almost identical protocol and case record forms as in our EPICOR study. Statistical methods We identified over 50 candidate variables for prediction (patient history, at admission, during admission, and at discharge), and these are listed in Appendix 1. From these a new risk score for one-year mortality post discharge was developed using Cox proportional hazard models. The statistical approach for model building was forward stepwise variable selection, with a criterion of p<0.01 for variable inclusion. For continuous predictors, checks were undertaken for non-linearity and, if found appropriate, re-modelling of such variables was conducted e.g. either using a binary cut-off (e.g. hemoglobin, blood glucose) or by expressing as a linear trend above a certain level (e.g. serum creatinine). In combining predictors for patients with STEMI and with NSTE-ACS it is important to explore evidence of statistical interactions with other predictors. On the whole, most variables selected showed a similar magnitude of risk prediction for both STEMI and NSTE-ACS patients. The one exception was that the increased risk of not receiving coronary revascularization during hospitalization was more marked in NSTE-ACS patients.

statistical interactions with other predictors. On the whole, most variables selected showed a similar magnitude of risk prediction for both STEMI and NSTE-ACS patients. The one exception was that the increased risk of not receiving coronary revascularization during hospitalization was more marked in NSTE-ACS patients. Some prognostic variables were missing in a small minority of patients. To overcome this problem, thereby enabling all patients’ available data to be validly used, a multiple imputation method was used based on a recently developed extension of the chained equations approach.11 Most predictor variables identified (see Table 1) are well understood, but the novel use of the EuroQoL EQ-5D requires explanation. This questionnaire evaluates five issues: patient mobility, self-care, usual activities, pain/discomfort and anxiety/depression. For each there is specific wording to elicit whether the patient has no, moderate, or severe limitation. For each we have scored 0, 1 or 2 points respectively, and adding up these scores yields a simple overall score ranging from 0 points up to a maximum of 10 points. While there do exist more complex weighted schemes for handling the EQ-5D,12 we feel that for practical use in our context of user-friendly risk prediction this required the adoption of such simple scoring. Table 1. Descriptive statistics for key baseline variables.

Most predictor variables identified (see Table 1) are well understood, but the novel use of the EuroQoL EQ-5D requires explanation. This questionnaire evaluates five issues: patient mobility, self-care, usual activities, pain/discomfort and anxiety/depression. For each there is specific wording to elicit whether the patient has no, moderate, or severe limitation. For each we have scored 0, 1 or 2 points respectively, and adding up these scores yields a simple overall score ranging from 0 points up to a maximum of 10 points. While there do exist more complex weighted schemes for handling the EQ-5D,12 we feel that for practical use in our context of user-friendly risk prediction this required the adoption of such simple scoring. Table 1. Descriptive statistics for key baseline variables. STEMI patients NSTE-ACS patients All Deaths No. of patients 4943 5625 10,568 3.9% STEMI 4943 3.1% NSTE-ACS 5625 4.5% Age, years, mean (SD) 59.4 (12.1) 63.8 (12.1) 61.8 (12.3) Gender Male 3924 3996 7920 3.7% Female 1019 1629 2648 4.3% Ejection fraction at admissiona Normal ≥40% 4035 4641 8676 2.9% Moderately reduced 30–39% 459 329 788 9.0% Severely reduced <30% 112 126 238 22.7% Cardiac complications in hospital MI or recurrent ischemia 258 342 600 6.7% Cardiogenic shock 85 24 109 7.3% Heart failure 327 289 616 12.8% Any arrhythmia 589 425 1014 6.4% Any of the above 1019 915 1934 7.6% Serum creatinine at admission,a mg/dl, mean (SD) 0.96 (0.42) 1.04 (0.59) 1.00 (0.52) ≥1.2 mg/dl 650 1060 1710 8.9% High blood glucose (≥160 mg/dl) at admissiona 1134 1035 2169 6.0% Low hemoglobin (<13 g/dl) at admissiona 891 1328 2219 6.9% COPD or other chronic lung disease 256 427 683 8.8% Peripheral vascular disease 145 384 529 11.0% On diuretics at discharge 683 1283 1966 8.5% Interventions during admission CABG or PCI 3863 3285 7148 2.6% Neither 1080 2340 3420 7.1% Simple EQ-5D score at dischargea 0, no problems 2392 2382 4774 2.4% 1 1049 1157 2206 3.2% 2 576 785 1361 4.4% 3 335 485 820 4.7% 4 211 325 536 8.4% ≥5, severe problems 226 345 571 11.9% Geographic region Northern Europe 1608 2174 3782 2.5% Southern Europe 1124 1213 2337 3.6% Eastern Europe 1145 1235 2380 4.8% Latin America 1066 1003 2069 5.5% CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; MI: myocardial infarction; NSTE-ACS: non-ST-elevation ACS; PCI: percutaneous coronary intervention; SD: standard deviation; STEMI: ST-segment elevation myocardial infarction.

ern Europe 1145 1235 2380 4.8% Latin America 1066 1003 2069 5.5% CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; MI: myocardial infarction; NSTE-ACS: non-ST-elevation ACS; PCI: percutaneous coronary intervention; SD: standard deviation; STEMI: ST-segment elevation myocardial infarction. a Indicates variables with missing data as follows: ejection fraction (8.2% missing), serum creatinine (5.6%), blood glucose (13.2%), hemoglobin (6.7%), EQ-5D (2.8%). Multiple imputation was used to overcome this: see statistical methods section. The multiple imputations were performed using Stata 12.0 while all other analyses used SAS version 9.2. Results The study cohort comprises 10,568 consecutive hospital survivors after an ACS event (4943 STEMI and 5625 NSTE-ACS). Four hundred and seven patients (3.9%) died within one year of discharge while 242 (2.3%) were lost to follow-up. From all of the candidate variables available, a Cox proportional hazard model was used with forward stepwise variable selection to identify 12 highly significant independent predictors of one-year mortality. These are described in Table 1 along with geographic region.

Results The study cohort comprises 10,568 consecutive hospital survivors after an ACS event (4943 STEMI and 5625 NSTE-ACS). Four hundred and seven patients (3.9%) died within one year of discharge while 242 (2.3%) were lost to follow-up. From all of the candidate variables available, a Cox proportional hazard model was used with forward stepwise variable selection to identify 12 highly significant independent predictors of one-year mortality. These are described in Table 1 along with geographic region. Table 2 presents the multivariable predictive model which simultaneously uses all 12 risk variables to produce an overall risk score. Variables in Table 2 are listed in order of their statistical significance (age is the strongest predictor) and each hazard ratio is adjusted for all the other variables. One statistical interaction was identified: for NSTE-ACS patients only, those who received percutaneous coronary intervention (PCI) or coronary artery bypass graft (CABG) during this admission had a lower mortality than those on medication only. For continuous variables, potential non-linearity in the prediction of survival was explored. Hence the increasing impact of serum creatinine on mortality was confined to values above 1.2 mg/dl while for blood glucose and hemoglobin binary cut-offs of ≥160 mg/dl and <13 g/dl were respectively used. Table 2. Multivariate analysis of one-year mortality: final model for all patients (with missing data imputed).

Table 2 presents the multivariable predictive model which simultaneously uses all 12 risk variables to produce an overall risk score. Variables in Table 2 are listed in order of their statistical significance (age is the strongest predictor) and each hazard ratio is adjusted for all the other variables. One statistical interaction was identified: for NSTE-ACS patients only, those who received percutaneous coronary intervention (PCI) or coronary artery bypass graft (CABG) during this admission had a lower mortality than those on medication only. For continuous variables, potential non-linearity in the prediction of survival was explored. Hence the increasing impact of serum creatinine on mortality was confined to values above 1.2 mg/dl while for blood glucose and hemoglobin binary cut-offs of ≥160 mg/dl and <13 g/dl were respectively used. Table 2. Multivariate analysis of one-year mortality: final model for all patients (with missing data imputed). Variable All patients Coefficient HR 95% CI p-value Age (per 10 years) 0.43 1.54 1.40–1.70 <0.00001 Ejection fraction <40%a 0.62 1.87 1.42–2.46 <0.0001 Ejection fraction <30%a 1.35 3.84 2.80–5.27 <0.0001 EQ-5D score (per unit) 0.15 1.16 1.10–1.21 <0.0001 Serum creatinine (per unit ≥1.2 mg/dl)a 0.22 1.25 1.13–1.38 <0.0001 Cardiac complication in hospital 0.41 1.50 1.21–1.86 0.0002 Blood glucose ≥160 mg/dla 0.39 1.48 1.19–1.84 0.0004 COPD 0.52 1.68 1.26–2.24 0.0004 Male gender 0.40 1.49 1.18–1.89 0.0009 NSTE-ACS with meds onlyb 0.39 1.47 1.17–1.86 0.0012 NSTE-ACS with PCI/CABGb −0.22 0.80 0.61–1.05 0.1117 Hemoglobin <13 g/dla 0.35 1.42 1.13–1.80 0.0029 Peripheral vascular disease 0.45 1.57 1.17–2.10 0.0029 On diuretics at discharge 0.30 1.35 1.08–1.70 0.0095 CABG: coronary artery bypass graft; CI: confidence interval; COPD: chronic obstructive pulmonary disease; HR: hazard ratio; NSTE-ACS: non-ST-elevation ACS; PCI: percutaneous coronary intervention; STEMI: ST-segment elevation myocardial infarction.

57 1.17–2.10 0.0029 On diuretics at discharge 0.30 1.35 1.08–1.70 0.0095 CABG: coronary artery bypass graft; CI: confidence interval; COPD: chronic obstructive pulmonary disease; HR: hazard ratio; NSTE-ACS: non-ST-elevation ACS; PCI: percutaneous coronary intervention; STEMI: ST-segment elevation myocardial infarction. a At admission; bas compared to STEMI. Figure 1 displays the independent impact of each predictor on mortality risk. In addition, there remain substantial regional differences in one-year mortality not explained by these predictors: Eastern Europe and Latin American have adjusted hazard ratios of 2.15 and 2.10, respectively, compared with Western Europe (North). Figure 1. Mortality hazard ratios for each variable in the predictive model. CABG: coronary artery bypass graft; CI: confidence interval; COPD: chronic obstructive pulmonary disease; NSTE-ACS: non-ST-elevation acute coronary syndrome; PCI: percutaneous coronary intervention; STEMI: ST segment elevation myocardial infarction. From the risk coefficients in Table 2, the multivariable risk score is readily calculated for each patient and its distribution (×10) is shown in Figure 2. The curve in Figure 2 relates a patient’s score to the probability of dying within one year of discharge. Good discrimination is achieved with c-statistic=0.81. Figure 2. Risk score distribution (and predicted mortality risk).

From the risk coefficients in Table 2, the multivariable risk score is readily calculated for each patient and its distribution (×10) is shown in Figure 2. The curve in Figure 2 relates a patient’s score to the probability of dying within one year of discharge. Good discrimination is achieved with c-statistic=0.81. Figure 2. Risk score distribution (and predicted mortality risk). Figure 3 shows the cumulative mortality over one year for patients classified into six risk groups. Groups 1–4 comprise the bottom four quintiles of risk while groups 5 and 6 are the top two deciles of risk. While all six groups are clearly separated, the absolute magnitude of differences between risk groups is much more marked for the top two deciles, with 6.3% and 18.2% one-year mortality, respectively. This contrasts with 0.5% one-year mortality in the lowest quintile. Figure 3. Cumulative mortality in six risk groups. Risk groups 1–4 correspond to quintiles 1–4, with the fifth quintile subdivided into two deciles (risk groups 5 and 6). Regarding model goodness-of-fit, Figure 4(a) compares observed and model-predicted one-year mortality risk across the six risk groups. Figure 4. Assessment of risk discrimination and model goodness-of-fit in six groups from low to very high risk (a) In original EPICOR (long-tErm follow uP of antithrombotic management patterns In acute CORonary syndrome patients) study and (b) In EPICOR Asia (validation cohort). For both plots, risk groups 1–4 correspond to quintiles 1–4, with the fifth quintile subdivided into two deciles (risk groups 5 and 6).

very high risk (a) In original EPICOR (long-tErm follow uP of antithrombotic management patterns In acute CORonary syndrome patients) study and (b) In EPICOR Asia (validation cohort). For both plots, risk groups 1–4 correspond to quintiles 1–4, with the fifth quintile subdivided into two deciles (risk groups 5 and 6). Table 3 shows two separate models for STEMI and NSTE-ACS patients. For nearly all predictors, the strength of mortality association is similar in both subgroups. However, the lower risk if coronary revascularization occurred during admission is more notable in NSTE-ACS patients, and this statistical interaction is captured in the main predictive model in Table 2. Table 3. Separate models for ST-segment elevation myocardial infarction (STEMI) and non-ST-elevation acute coronary syndrome (NSTE-ACS) patients.

Table 3 shows two separate models for STEMI and NSTE-ACS patients. For nearly all predictors, the strength of mortality association is similar in both subgroups. However, the lower risk if coronary revascularization occurred during admission is more notable in NSTE-ACS patients, and this statistical interaction is captured in the main predictive model in Table 2. Table 3. Separate models for ST-segment elevation myocardial infarction (STEMI) and non-ST-elevation acute coronary syndrome (NSTE-ACS) patients. Variable STEMI patients NSTE–ACS patients HR 95% CI HR 95% CI Age (per 10 years) 1.56 1.34–1.80 1.52 1.34–1.73 Ejection fraction <40%a 1.42 0.89–2.27 2.29 1.62–3.24 Ejection fraction <30%a 3.73 2.17–6.41 4.03 2.69–6.02 EQ-5D score (per unit) 1.18 1.09–1.28 1.14 1.07–1.22 Serum creatinine (per unit ≥1.2 mg/dl)a 1.27 1.04–1.55 1.23 1.09–1.38 Cardiac complication in hospital 1.15 0.80–1.65 1.73 1.33–2.27 Blood glucose ≥160 mg/dla 1.29 0.91–1.84 1.64 1.24–2.16 COPD 1.60 0.96–2.68 1.71 1.20–2.43 Male gender 1.47 0.98–2.22 1.54 1.14–2.06 PCI/CABG during admission 0.73 0.52–1.04 0.52 0.40–0.69 Hemoglobin <13 g/dla 1.57 1.05–2.35 1.33 1.00–1.78 Peripheral vascular disease 1.47 0.76–2.86 1.55 1.10–2.18 On diuretics at discharge 1.43 0.97–2.11 1.29 0.97–1.73 CABG: coronary artery bypass graft; CI: confidence interval; COPD: chronic obstructive pulmonary disease; HR: hazard ratio; PCI: percutaneous coronary intervention. aAt admission.

7 1.05–2.35 1.33 1.00–1.78 Peripheral vascular disease 1.47 0.76–2.86 1.55 1.10–2.18 On diuretics at discharge 1.43 0.97–2.11 1.29 0.97–1.73 CABG: coronary artery bypass graft; CI: confidence interval; COPD: chronic obstructive pulmonary disease; HR: hazard ratio; PCI: percutaneous coronary intervention. aAt admission. For STEMI patients we investigated the impact of rapid time to admission (or time to reperfusion) on reducing mortality after discharge. There were 697 STEMI patients (14%) admitted within one hour of symptom onset: hazard ratio 0.44 (p=0.026) compared to other STEMI patients. Also, 1316 STEMI patients (27%) had reperfusion within two hours of symptom onset: hazard ratio 0.64 (p=0.053) compared to other STEMI patients. These findings were of borderline statistical significance so these two variables were not included in the main predictive model. In order to validate our main model on an external cohort, we used the 9907 patients in the EPICOR Asia registry who had complete data on all variables listed in Table 2, of whom 3.1% died within one year of hospital discharge. Figure 4(b) compares the observed and model-predicted mortality in six risk groups (from lowest quintile to top decile). The model fit and extent of risk discrimination is very similar to what was found in our original cohort. The c-statistic in EPICOR Asia patients is 0.784, only slightly less than the c=0.81 achieved in model development.

he observed and model-predicted mortality in six risk groups (from lowest quintile to top decile). The model fit and extent of risk discrimination is very similar to what was found in our original cohort. The c-statistic in EPICOR Asia patients is 0.784, only slightly less than the c=0.81 achieved in model development. Discussion The findings we present are based on a large international prospective real-world cohort study comprising consecutive patients hospitalized from an ACS event within 24 h of symptoms onset who survived to hospital discharge. Such a representative population across Europe and Latin America is therefore uniquely well placed to quantify the independent determinants of mortality risk over one year post-discharge. The 12 highly significant predictors we identified should all be readily available in routine clinical practice. To facilitate the quantification of individual risk we provide a web calculator www.acsrisk.org thus avoiding the burden of numerical calculations. There is a marked identifiable variation in individual patient risk (see Figure 4). This means a sizeable proportion of patients can be classified as low risk, e.g. around half have a one-year mortality risk <1%. On the other hand 10% of patients have a high one-year mortality risk (see Figures 2 and 3). Knowing this fact, based on our risk model, should help in supporting appropriate patient management.

s means a sizeable proportion of patients can be classified as low risk, e.g. around half have a one-year mortality risk <1%. On the other hand 10% of patients have a high one-year mortality risk (see Figures 2 and 3). Knowing this fact, based on our risk model, should help in supporting appropriate patient management. The contributions made by each specific predictor are worth noting. Not surprisingly, age has the most profound influence on mortality risk, followed by reduced ejection fraction. A more novel contributor to risk assessment is quality of life at discharge, using a simple score derived from the EuroQoL EQ-5D.13 Across five aspects (mobility, self-care, usual activities, pain/discomfort, anxiety/discomfort) we add one point for moderate impairment or two points for severe impairment. Patients scoring four points or more (11%) had more than double the mortality risk of patients with no impairment (45%), with a gradient of risk for patients in between these two extremes. Thus, a poor functional quality of life may be expressing some level of frailty and a mortality risk that is not captured by other predictors.

oring four points or more (11%) had more than double the mortality risk of patients with no impairment (45%), with a gradient of risk for patients in between these two extremes. Thus, a poor functional quality of life may be expressing some level of frailty and a mortality risk that is not captured by other predictors. The 4943 STEMI patients had a lower one-year mortality after discharge compared to the 5625 NSTE-ACS patients: 3.1% vs 4.5% died, respectively. However, after adjustment for the other 11 risk factors, the hazard ratio became 1.00 (95% CI 0.80–1.24). This reflects that NSTE-ACS had a higher prevalence of other risk factors (see Table 1). Indeed, NSTE-ACS contributes over twice as many patients in the top decile of risk compared to STEMI. However, for NSTE-ACS patients only, one notable contributor to risk was not having PCI or CABG during hospital stay: hazard ratio 1.84 after adjustment for other risk factors. Thus, lack of coronary revascularization reflects an anticipated poorer prognosis post-discharge. This may be explained by either the actual risk benefit of coronary revascularization or selection bias (i.e. poor risk patients are deemed not appropriate for intervention).14

1.84 after adjustment for other risk factors. Thus, lack of coronary revascularization reflects an anticipated poorer prognosis post-discharge. This may be explained by either the actual risk benefit of coronary revascularization or selection bias (i.e. poor risk patients are deemed not appropriate for intervention).14 From blood samples at admission, contributions to higher risk are represented by raised creatinine, raised glucose and lower hemoglobin. For disease history, both chronic obstructive pulmonary disease and peripheral vascular disease increased risk, indicating that conditions other than cardiac disease carry a mortality risk. Cardiac complications during the admission were associated with a 50% increase in mortality risk. Also, men had a 50% higher risk than women after all other risk factors were accounted for. In univariate analysis women have a higher one-year mortality than men (4.3% vs 3.7%). But women are more prone to having other risk factors (e.g. older age) so that in the multivariable model being female independently predicts a lower risk. Lastly, being on diuretics at discharge was an indicator of 35% higher mortality risk.

iate analysis women have a higher one-year mortality than men (4.3% vs 3.7%). But women are more prone to having other risk factors (e.g. older age) so that in the multivariable model being female independently predicts a lower risk. Lastly, being on diuretics at discharge was an indicator of 35% higher mortality risk. Confidence in the generalizability of any new risk model is much enhanced if it is validated on an external population. Here, the EPICOR Asia study has provided similar risk discrimination and goodness of fit (compare the two plots in Figure 4), as summarized by the c-statistic of 0.78 in Asian patients compared to 0.81 in the original cohort. Given that the two studies were from different geographic regions, this provides assurance that our risk model may well be of global applicability. In external validation some reduction in c-statistic is always to be expected on statistical grounds i.e. risk coefficients in any model are optimized by the maximum likelihood principle of any model fit, so the true strength of prediction is inevitably slightly less in another independent data set. To further explore model fit in the Asian cohort we did another Cox regression model with the same 12 predictor variables: the hazard ratios of all but one variable were very similar to those reported in our original EPICOR model (Table 2). The one exception was peripheral vascular disease which was very uncommon in the Asian cohort so that its hazard ratio had a wide CI. This consistency of findings suggests that there is little effect of ethnic diversity on risk prediction.

able were very similar to those reported in our original EPICOR model (Table 2). The one exception was peripheral vascular disease which was very uncommon in the Asian cohort so that its hazard ratio had a wide CI. This consistency of findings suggests that there is little effect of ethnic diversity on risk prediction. While there exist several other risk scores for patients with ACS, most do not focus on risk from the time of hospital discharge and hence are not appropriate for comparison here. However, Eagle et al. have used the Global Registry of Acute Coronary Events (GRACE) registry to estimate mortality risk six months post discharge.7 Their risk calculator includes nine predictors: age, history of congestive heart failure, history of MI, increased heart rate at admission, lower systolic blood pressure at admission, elevated serum creatinine at admission, elevated cardiac enzymes, ST-segment depression, and no in-hospital PCI. This achieved a similar predictive strength to the current model (c-statistic = 0.81 at development, 0.75 at validation). However, the shorter time period could be a limiting factor. Their six-month mortality rate (4.8%) is higher than our one-year mortality rate (3.9%), perhaps reflecting the fact that their cohort is from around 10 years ago. Also, the mortality rate in EPICOR might not include high-risk patients transferred to other units for non-cardiac complications or needing longer-term care. It would be useful if the GRACE and EPICOR risk models were directly compared in an independent cohort of ACS patients followed from hospital discharge.

ears ago. Also, the mortality rate in EPICOR might not include high-risk patients transferred to other units for non-cardiac complications or needing longer-term care. It would be useful if the GRACE and EPICOR risk models were directly compared in an independent cohort of ACS patients followed from hospital discharge. There are some limitations inherent to our risk model. Being a study on hospital survivors, blood pressure and heart rate at admission were not recorded in our database and hence could not be included in the model. Certain other variables (e.g. probrain natiuretic peptide, incomplete revascularization) were also not available for inclusion. Even after taking account of our 12 highly predictive risk factors, there persist substantial unexplained geographic differences in post-discharge mortality risk. In Eastern Europe and Latin America one-year mortality is markedly higher than in Western Europe, and further investigation is needed to clarify why this discrepancy exists. This geographic heterogeneity could be perceived as a limitation but given the intention of any risk model is that it be useful in many different countries we feel our population’s geographic diversity is an asset in enhancing generalizability. We intend to publish further on the geographic regional differences in the distribution (prevalence) of risk factors, both in the EPICOR and EPICOR-Asia cohorts.

on of any risk model is that it be useful in many different countries we feel our population’s geographic diversity is an asset in enhancing generalizability. We intend to publish further on the geographic regional differences in the distribution (prevalence) of risk factors, both in the EPICOR and EPICOR-Asia cohorts. One could argue that STEMI and NSTE-ACS are sufficiently different conditions that two separate risk models should be developed. However, as shown in Table 3 there is substantial consistency of risk prediction for the 12 variables (i.e. mostly similar hazard ratios) so that for practical purposes we feel a single overall risk model is desirable. Another limitation is that with over 50 candidate predictor variables there is a risk of a “false positive” predictor entering the risk model. However with p<0.01 as entry criterion this risk is relatively low. The one novel predictor is the EQ-5D score, but this is very highly significant. The rest are to be expected on the basis of prior studies of mortality risk in ACS patients. While the design of EPICOR was geared to recruiting representative patients from representative centers in each country, we cannot directly verify that centers are indeed representative in respect to adherence to treatment guidelines and other aspects of patient management. Thus some caution is warranted with respect to extrapolation of findings to the overall population of ACS patients.

ients from representative centers in each country, we cannot directly verify that centers are indeed representative in respect to adherence to treatment guidelines and other aspects of patient management. Thus some caution is warranted with respect to extrapolation of findings to the overall population of ACS patients. In conclusion, we have documented how post-discharge mortality risk after an ACS event varies markedly. Individual one-year mortality can be reliably estimated using 12 readily available items, and the consequent risk discrimination and model fit are good. User-friendly access to our risk model is available via the web: www.acsrisk.org. We feel this tool can help influence appropriate patient management post discharge, especially in identifying individuals at higher risk for whom more intensive monitoring may be appropriate. The authors are grateful to Jonathan Bartlett for carrying out the complex multiple imputation procedures, and Laura Emerson from Worldwide Clinical Trials for performing the statistical analyses. Editorial assistance was provided by Rob Campbell of Prime Medica Ltd (Knutsford, Cheshire, UK), funded by AstraZeneca.

In conclusion, we have documented how post-discharge mortality risk after an ACS event varies markedly. Individual one-year mortality can be reliably estimated using 12 readily available items, and the consequent risk discrimination and model fit are good. User-friendly access to our risk model is available via the web: www.acsrisk.org. We feel this tool can help influence appropriate patient management post discharge, especially in identifying individuals at higher risk for whom more intensive monitoring may be appropriate. The authors are grateful to Jonathan Bartlett for carrying out the complex multiple imputation procedures, and Laura Emerson from Worldwide Clinical Trials for performing the statistical analyses. Editorial assistance was provided by Rob Campbell of Prime Medica Ltd (Knutsford, Cheshire, UK), funded by AstraZeneca. Conflict of interest: S Pocock has received research funding from AstraZeneca; H Bueno has received advisory/consulting fees from AstraZeneca, Bayer, BMS, Daichii-Sankyo, Eli-Lilly, Novartis, Pfizer, Sanofi, and Roche, and grants from AstraZeneca; M Licour, J Medina, and L Zhang are employees of AstraZeneca; L Annemans has received consulting and lecture fees from AstraZeneca; N Danchin has received consulting or speaking fees from AstraZeneca, BMS, Boehringer-Ingelheim, GSK, MSD-Schering Plough, Novartis, Pierre Fabre, Pfizer, Roche, Sanofi-Aventis, Servier, Takeda, and The Medicines Company; Y Huo has nothing to disclose; F Van de Werf has received consulting fees and research grants from Boehringer Ingelheim and Merck, and consulting fees from Roche, Sanofi-Aventis, AstraZeneca, and The Medicines Company.

h, Novartis, Pierre Fabre, Pfizer, Roche, Sanofi-Aventis, Servier, Takeda, and The Medicines Company; Y Huo has nothing to disclose; F Van de Werf has received consulting fees and research grants from Boehringer Ingelheim and Merck, and consulting fees from Roche, Sanofi-Aventis, AstraZeneca, and The Medicines Company. Funding: EPICOR is funded by AstraZeneca. Appendix Appendix 1. A list of candidate predictor variables.

h, Novartis, Pierre Fabre, Pfizer, Roche, Sanofi-Aventis, Servier, Takeda, and The Medicines Company; Y Huo has nothing to disclose; F Van de Werf has received consulting fees and research grants from Boehringer Ingelheim and Merck, and consulting fees from Roche, Sanofi-Aventis, AstraZeneca, and The Medicines Company. Funding: EPICOR is funded by AstraZeneca. Appendix Appendix 1. A list of candidate predictor variables. Demographics and medical history Variables collected during admission Gender Time from symptom onset to admission Age Time from admission to reperfusion Race Time from symptom onset to reperfusion Education level Length of hospital stay Professional status Killip class Height Diagnosis (STEMI, NSTEMI, unstable angina) Weight Left bundle branch block Body mass index Ejection fractiona Hypertension White blood counta Hypercholesterolemia Creatininea Diabetes Glucosea Family history of CAD Hemoglobina Smoking PCI during admission Previous MI CABG during admission Prior PCI Reperfusion (PCI at thrombolysis) Prior CABG No. of dilated vessels Chronic angina Any drug eluting stent Prior heart failure No. of antiplateletsb Prior atrial fibrillation Anticoagulantb Prior transient ischemic attack/stroke Beta blockerb Prior peripheral vascular disease Angiotensin-converting enzyme inhibitor/angiotensin receptor blockerb Chronic renal failure Diureticsb COPD or other chronic lung disease Aldosterone inhibitorb Calcium-channel blockerb Ischemic complications Cardiogenic shock Heart failure Dyspnea Arrhythmia Dependence at discharge EQ-5D overall health state at discharge EQ-5D simple score at discharge COPD: chronic obstructive pulmonary disease;

nal failure Diureticsb COPD or other chronic lung disease Aldosterone inhibitorb Calcium-channel blockerb Ischemic complications Cardiogenic shock Heart failure Dyspnea Arrhythmia Dependence at discharge EQ-5D overall health state at discharge EQ-5D simple score at discharge COPD: chronic obstructive pulmonary disease; a At admission; bat discharge.

Introduction Chest pain and related complaints are estimated to account for 6% of all attendances to UK Emergency Departments (EDs).1 Determining which of these presentations represent an acute coronary syndrome, quickly and with high sensitivity and specificity, is an everyday challenge. The measurement of cardiac-specific biomarkers released into the circulation is invaluable, and the measurement of cardiac troponin (cTn) I and T is engrained in the universal definition of myocardial infarction.2 However, the slow release of cTn, in combination with the relative analytic insensitivity of conventional cTn assays, has necessitated serial measurements separated by at least six hours to increase both sensitivity and specificity. This period of diagnostic uncertainty prolongs the patient’s hospital stay, delays their treatment and has an associated fiscal cost. The advent of high sensitivity troponin assays has encouraged investigators to examine shorter intervals between repeat troponin estimations. The high sensitivity assays have also allowed the testing of diagnostic cut-off concentrations well below the population defined 99th centile to rapidly rule out acute myocardial injury. These innovations culminated in the European Society of Cardiology (ESC) releasing new guidelines in September 2015 for the management of patients without persistent ST elevation.3–6 These guidelines adopt a ‘rule-out’ troponin value significantly below the 99th centile and a ‘rule-in’ value well above the 99th centile. Between these values of diagnostic clarity, the change in troponin level over the course of one hour can guide further rule-in or rule-out. In October 2015 we proposed introduction of the 0 hour rule-in/rule-out component of the ESC algorithm at St Thomas’ Hospital (based in central London and home to a tertiary cardiac unit) and adopted the guideline, following an internal consultation process, during December 2015–January 2016. This internal consultation process also involved extension of teaching to ED staff, both nursing and physician, as to the appropriate use of the algorithm.

sed in central London and home to a tertiary cardiac unit) and adopted the guideline, following an internal consultation process, during December 2015–January 2016. This internal consultation process also involved extension of teaching to ED staff, both nursing and physician, as to the appropriate use of the algorithm. All ‘post-intervention’ data were collected after implementation and associated staff training. Whilst the ESC guidelines help to streamline the diagnostic pathway, there has been little information regarding their impact on front-line medical services. The present study, based in the ED of a large Central London hospital, aims to (a) prospectively assess the risk classification of patients based on 0 hour hs-cTnT measurement, and (b) examine the effect of clinical implementation of the 0 hour component of the ESC guideline on the patient pathway. In particular, we document changes in the pattern of repeat troponin measurements and overnight admission.

spectively assess the risk classification of patients based on 0 hour hs-cTnT measurement, and (b) examine the effect of clinical implementation of the 0 hour component of the ESC guideline on the patient pathway. In particular, we document changes in the pattern of repeat troponin measurements and overnight admission. Methods Data was prospectively collected on all high-sensitivity cardiac troponin T (hs-cTnT) assays performed on serum from patients presenting to the ED of St Thomas’ Hospital, between September 2015–March 2016. This time-period of data collection spans the pre-intervention (September-November), transition (December), and post-intervention (January-March) phases of algorithm implementation. hs-cTnT assays were performed using the Roche Elecsys platform (using a high-sensitivity reagent instead of a contemporary: 99th percentile of a healthy reference population reported at 14 ng/l, imprecision corresponding to 10% coefficient of variation (CV) at 13 ng/l, limit of blank at 3 ng/l, limit of detection at 5 ng/l). The hs-cTnT value measured in the ED was matched to any subsequent hs-cTnT measurement on the same patient within 24 h. Further information on admission, admitting specialty, and length of stay was collected from electronic discharge records. Data on presenting symptom was obtained from the system used for triage and clinical tracking in the ED (Ascribe Symphony); this captures the prime medical complaint but, however, it does not encompass a physician’s interpretation. Discharge diagnoses are locally recorded according to the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10) and were subsequently categorised into diagnostic groups by two adjudicators (JM and TEK).

however, it does not encompass a physician’s interpretation. Discharge diagnoses are locally recorded according to the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10) and were subsequently categorised into diagnostic groups by two adjudicators (JM and TEK). The new algorithm for the diagnostic management of possible Non-ST elevation Acute Coronary Syndrome (NSTE-ACS) can be summarised as follows: hs-cTnT is measured on arrival to ED for patients with a history suggestive of Acute Coronary Syndrome (ACS), and an electrocardiogram (ECG) without persistent ST elevation. ACS can be ‘ruled-out’ in low-risk patients with a hs-cTnT on presentation of <5 ng/l, and ‘ruled-in’ for those patients with an initial hs-cTnT of >50 ng/l (Figure 1). Although not adopted into our algorithm, the ESC advises that in patients with an initial hs-cTnT of 5-51 ng/l, a repeat hs-cTnT at one hour is performed, with rule-out if the initial hs-cTnT is <12 ng/l and if a change in hs-cTnT (ΔTnT) is <3 ng/l, and rule-in if ΔTnT is ≥5 ng/l. For the purposes of our analysis, a patient was considered to have had a repeat hs-cTnT if a second sample was measured within 24 h of the first. Patients were excluded from analysis if the first sample haemolysed. Those hs-cTnT measurements returned below the limit of blank (<3 ng/l) were all ascribed a value of 2.99 ng/l to allow for data analysis. Continuous variables were assessed for normality using Shapiro-Wilk Test. All data are expressed as medians (1st quartile, 3rd quartile) or means (standard deviation) for continuous variables (compared with the Mann-Whitney-U test or student’s t-test), and for categorical variables as numbers and percentages (compared with Pearson chi-square). Hypothesis testing was two-tailed, and p values <0.05 were considered statistically significant. Statistical analysis was conducted using SPSS version 22 (IBM Corp., Armonk, New York, USA) and R, version 3.3.0 GUI 1.68 (The R Foundation for Statistical Computing), including ggplot2.

(compared with Pearson chi-square). Hypothesis testing was two-tailed, and p values <0.05 were considered statistically significant. Statistical analysis was conducted using SPSS version 22 (IBM Corp., Armonk, New York, USA) and R, version 3.3.0 GUI 1.68 (The R Foundation for Statistical Computing), including ggplot2. Figure 1. The high-sensitivity troponin T (hs-cTnT) rapid diagnostic algorithm introduced at St Thomas’ Hospital. ACS: Acute Coronary Syndrome; LBBB: Left Bundle Branch Block; ECG: electrocardiogram; PCI: Percutaneous Coronary Intervention. Results Over a period of 213 days, spanning the introduction of the new diagnostic protocol, a total of 4644 patients had a hs-cTnT measurement in the ED. A summary of the presenting complaint of all patients with hs-cTnT measurements in the study period (September 2015 –March 2016) is presented in Table 1. In short, of the patients with a measured hs-cTnT (n=4644), chest pain was the primary presenting symptom in 45.7% (n=2120), and shortness of breath in 8.2% (n=382) – see Figure 2. Median age was 54 years (interquartile range (IQR), 41–70). Table 1. Summary of the presenting complaint of all patients with high-sensitivity troponin T (hs-cTnT) measurements in the study period (September 2015–March 2016). Presenting complaint All attendances (%) Abdominal pain 141 (3.0) Back pain 55 (1.2) Chest pain 2120 (45.7) Collapsed adult 211 (4.5) Falls 91 (2.0) Shortness of breath 382 (8.2) Other 437 (9.4) Unwell adult 1207 (26.0)

Table 1. Summary of the presenting complaint of all patients with high-sensitivity troponin T (hs-cTnT) measurements in the study period (September 2015–March 2016). Presenting complaint All attendances (%) Abdominal pain 141 (3.0) Back pain 55 (1.2) Chest pain 2120 (45.7) Collapsed adult 211 (4.5) Falls 91 (2.0) Shortness of breath 382 (8.2) Other 437 (9.4) Unwell adult 1207 (26.0) Total n=4644 Frequencies quoted as number (%); sample selection: all patients presenting to the Emergency Department with a hs-cTnT measured as part of their assessment between September 2015 and March 2016, age ≥18 years; ‘Other’ summarises non-cardiac presentations such as ‘overdose’ and ‘limb problems’. Figure 2. Bar graph summarising the presenting complaint of all patients (n=4644) with a measured high-sensitivity troponin T (hs-cTnT) in the entire study period; frequencies quoted as percentage of the cohort. SOB: Shortness of Breath. 0 hour risk stratification for whole sample period Of the entire cohort, 40.4% had an initial hs-cTnT concentration below the ‘rule-out’ value of 5 ng/l at presentation, and 7.6% had a concentration above the ‘rule-in’ value of 50 ng/l (Figure 3). Of the patients presenting with chest pain (n=2120), 1026 (48.4%) had an initial hs-cTnT concentration below the ‘rule-out’ threshold, 107 (5%) had a concentration above the ‘rule-in’ threshold. Of the patients presenting with shortness of breath (n=382), 89 (23.3%) had an initial hs-cTnT concentration below the ‘rule-out’ threshold, 74 (19.4%) had a concentration above the ‘rule-in’ threshold.

ial hs-cTnT concentration below the ‘rule-out’ threshold, 107 (5%) had a concentration above the ‘rule-in’ threshold. Of the patients presenting with shortness of breath (n=382), 89 (23.3%) had an initial hs-cTnT concentration below the ‘rule-out’ threshold, 74 (19.4%) had a concentration above the ‘rule-in’ threshold. Figure 3. Graph outlining the distribution of all high-sensitivity troponin T (hs-cTnT) values measured on patients presenting to the Emergency Department during the monitoring period (September 2015–March 2016; n=4644); the following thresholds applied: <5 ng/l ‘rule-out’, 5-50 ng/l ‘observe’, >50 ng/l ‘rule-in’. Retrospective analysis of deltas for all presentations Although our algorithm incorporates only the rule-in/rule-out classification based on a 0 hour hs-cTnT measurement, retrospective analysis of the entire cohort demonstrates that 10.6% of those at intermediate risk (0 hour hs-cTnT 5–50 ng/l) could have been ruled-in on repeat testing with a ΔTnT ≥5 ng/l, and 45.1% could have been ruled-out on the basis of an initial TnT<12ng/l and ΔTnT<3 ng/l.

n based on a 0 hour hs-cTnT measurement, retrospective analysis of the entire cohort demonstrates that 10.6% of those at intermediate risk (0 hour hs-cTnT 5–50 ng/l) could have been ruled-in on repeat testing with a ΔTnT ≥5 ng/l, and 45.1% could have been ruled-out on the basis of an initial TnT<12ng/l and ΔTnT<3 ng/l. Discharge diagnosis Altogether 1876 patients were admitted from the ED during the entire study period. Amongst these, the prevalence of ischaemic heart disease in the discharge diagnosis was 21.2% (n=397); congestive cardiac failure was the discharge diagnosis in 5.8%; pulmonary embolism in 1.5%. Of those patients admitted with a troponin value above the rule-in threshold (50 ng/l), 35.6% were diagnosed with ischaemic cardiac pathology (see Table 2, Figure 4 for details on all admitted patients, Figure 5 for subgroup analysis on all patients with a hs-cTnT at presentation >50 ng/l). Table 2. Details of admitted patients. Coding diagnosis All admitted patients hs-cTnT >50 ng/L Aortic dissection 8 (0.4) 0 (0) IHD 397 (21.2) 88 (35.6) Arrhythmia 159 (8.5) 17 (6.9) CCF 108 (5.8) 26 (10.5) Cardiac other 106 (5.7) 20 (8.1) PE 28 (1.5) 5 (2.0) OAD 100 (5.3) 7 (2.8) Resp other 24 (1.3) 3 (1.2) Infectious 189 (10.1) 17 (6.9) Renal 52 (2.8) 15 (6.1) GI 124 (6.6) 8 (3.2) MSK 100 (5.3) 9 (3.6) Other 481 (25.6) 32 (13.0) Total n=1876 n=247 CCF: congestive cardiac failure; GI: gastrointestinal disorders; IHD: ischaemic heart disease; MSK: musculo-skeletal disorders; OAD: obstructive airways disease PE: pulmonary embolism.

Resp other 24 (1.3) 3 (1.2) Infectious 189 (10.1) 17 (6.9) Renal 52 (2.8) 15 (6.1) GI 124 (6.6) 8 (3.2) MSK 100 (5.3) 9 (3.6) Other 481 (25.6) 32 (13.0) Total n=1876 n=247 CCF: congestive cardiac failure; GI: gastrointestinal disorders; IHD: ischaemic heart disease; MSK: musculo-skeletal disorders; OAD: obstructive airways disease PE: pulmonary embolism. ‘Cardiac other’ includes myocarditis, valvular heart and pericardial disease; ‘Resp other’ includes pleural effusion; Infectious includes lobar pneumonia, urinary tract infection and influenza; GI includes gastro-oesophageal reflux disease, gastroenteritis and symptomatic cholelithiasis; MSK includes costochondritis, bony fractures and other injuries; ‘Other’ includes sickle-cell anaemia, malignancy and mental health disorder. Frequencies quoted as number (%); Sample representative of the entire study period (September 2015 – March 2016) and comprises of all patients admitted from the Emergency Department. Figure 4. Bar graph summarising the discharge diagnosis of all admitted patients in the monitoring period (September 2015–March 2016; n=1876); frequencies quoted as percentage of the overall number of patients admitted following high-sensitivity troponin T (hs-cTnT) testing. CCF: congestive cardiac failure; GI: gastrointestinal disorder; IHD: ischaemic heart disease; MSK: musculo-skeletal disorder; OAD: obstructive airways disease; PE: pulmonary embolism.

Figure 4. Bar graph summarising the discharge diagnosis of all admitted patients in the monitoring period (September 2015–March 2016; n=1876); frequencies quoted as percentage of the overall number of patients admitted following high-sensitivity troponin T (hs-cTnT) testing. CCF: congestive cardiac failure; GI: gastrointestinal disorder; IHD: ischaemic heart disease; MSK: musculo-skeletal disorder; OAD: obstructive airways disease; PE: pulmonary embolism. ‘Cardiac other’ includes myocarditis, valvular heart, conduction tissue and pericardial disease; ‘Resp other’ includes pleural effusion; GI includes gastro-oesophageal reflux disease, gastroenteritis and symptomatic cholelithiasis; ‘Infection’ includes lobar pneumonia, urinary tract infection and influenza; MSK includes costochondritis, bony fractures and other injuries; ‘Other’ includes sickle-cell anaemia, malignancy and mental health disorder. Figure 5. Bar graphs summarising the discharge diagnosis of all admitted patients in the monitoring month with an initial high-sensitivity troponin T (hs-cTnT) level >50 ng/l (n=247); frequencies quoted as percentage. CCF: congestive cardiac failure; GI: gastrointestinal disorder; IHD: ischaemic heart disease; MSK: musculo-skeletal disorder; OAD: obstructive airways disease; PE: pulmonary embolism.

Figure 5. Bar graphs summarising the discharge diagnosis of all admitted patients in the monitoring month with an initial high-sensitivity troponin T (hs-cTnT) level >50 ng/l (n=247); frequencies quoted as percentage. CCF: congestive cardiac failure; GI: gastrointestinal disorder; IHD: ischaemic heart disease; MSK: musculo-skeletal disorder; OAD: obstructive airways disease; PE: pulmonary embolism. ‘Cardiac other’ includes myocarditis, valvular heart, conduction tissue and pericardial disease; ‘Resp other’ includes pleural effusion; GI includes gastro-oesophageal reflux disease, gastroenteritis and symptomatic cholelithiasis; ‘Infection’ includes lobar pneumonia, urinary tract infection and influenza; MSK includes costochondritis, bony fractures and other injuries; ‘Other’ includes sickle-cell anaemia, malignancy and mental health disorder. Repeat troponin samples in the post-intervention period In the three months following introduction of the algorithm (i.e. the ‘post-intervention period’), 946 patients (50.2%) had an initial hs-cTnT in the 5-50 ng/l zone of diagnostic uncertainty – of these, 443 (46.8%) had a repeat measurement within 24 h. Of the patients undergoing further testing, 189 (42.7%) had a repeat measurement within 1.5 h. Median time to repeat hs-cTnT measurement was 1.6 h (1.3, 2.2) for the entire post-intervention period.

ial hs-cTnT in the 5-50 ng/l zone of diagnostic uncertainty – of these, 443 (46.8%) had a repeat measurement within 24 h. Of the patients undergoing further testing, 189 (42.7%) had a repeat measurement within 1.5 h. Median time to repeat hs-cTnT measurement was 1.6 h (1.3, 2.2) for the entire post-intervention period. Eight hundred and ninety-two patients presented with chest pain in the post-intervention period. Of these, 390 patients (43.7%) were in the observational group, of which 222 (56.9%) had a repeat measurement within 24 h. Of the patients undergoing further testing, 106 (47.7%) had a repeat measurement within 1.5 h. The median time to repeat hs-cTnT measurement in the group presenting with chest pain was 1.5 h (1.3, 2). One hundred and fifty-four patients presented with shortness of breath in the post-intervention period. Of these, 87 patients (56.5%) were in the observational group, of which 29 (33.3%) had a repeat hs-cTnT within 24 h. Of the patients undergoing further testing, 10 (34.5%) had a repeat measurement within 1.5 h. Median time to repeat in the group presenting with shortness of breath was 1.8 h (1.4, 2.1).

rvention period. Of these, 87 patients (56.5%) were in the observational group, of which 29 (33.3%) had a repeat hs-cTnT within 24 h. Of the patients undergoing further testing, 10 (34.5%) had a repeat measurement within 1.5 h. Median time to repeat in the group presenting with shortness of breath was 1.8 h (1.4, 2.1). Comparison of pre- and post-intervention periods for all presentations Over the timeframe of implementation of the new algorithm we have demonstrated a gradual rise in the proportion of patients in the intermediate risk group (all presenting complaints) who had a repeat hs-cTnT measured within 1.5 h. At month 1 (pre-implementation), only 3.3% of repeat hs-cTnT measurements in the intermediate-risk patients were within 1.5 h, rising to 40.8% by month 7 (post-implementation) (p<0.001). In tandem, the median time to repeat troponin has fallen from 7.8 h (4.7, 11.1) to 1.7 h (1.3, 2.4) (p<0.001). This has been accompanied by a non-significant trend towards reduced overnight admissions in the low-risk group. In a month prior to implementation, of all patients with a hs-cTnT measurement <5 ng/l on presentation to ED, 12.7% were admitted for at least one night. This figure fell to 9.5% by month 7 (p=0.26, n=525). Early outcome data demonstrates that 30-day mortality in all patients with suspected ACS was not different before and after implementation of the new algorithm (1.8% versus 1.4% respectively, p=0.38).

on presentation to ED, 12.7% were admitted for at least one night. This figure fell to 9.5% by month 7 (p=0.26, n=525). Early outcome data demonstrates that 30-day mortality in all patients with suspected ACS was not different before and after implementation of the new algorithm (1.8% versus 1.4% respectively, p=0.38). Discussion This study documents the rate of adoption of a rapid rule-in/rule-out algorithm for the routine clinical care of patients presenting with suspected NSTE-ACS, based on a single blood test at presentation. In this large cohort of over 4600 patients, 48% of all patients and 53% of patients with chest pain could be dichotomised into high- or low risk groups on the basis of a single hs-cTnT measured on presentation.

e clinical care of patients presenting with suspected NSTE-ACS, based on a single blood test at presentation. In this large cohort of over 4600 patients, 48% of all patients and 53% of patients with chest pain could be dichotomised into high- or low risk groups on the basis of a single hs-cTnT measured on presentation. Multiple studies have prospectively validated the sensitivity and specificity of diagnostic algorithms based on high-sensitivity troponin assays.4–8 The unifying aim is to rapidly identify patients with ACS, facilitating prompt therapeutic intervention for those who need it, and prompt discharge for those who do not. However, since the ESC guidelines have been established, there is a dearth of studies that have addressed the fundamental question – can such an algorithm be implemented into routine clinical practice? As we have incorporated our algorithm into clinical practice, we have seen an increased rate of repeat testing, and a trend to faster repeats, in patients classified into the intermediate risk group on presentation. Whilst this algorithm should lead to empowerment of clinicians to exclude NSTE-ACS and discharge during the early stages of presentation, we have yet to observe a significant reduction in overnight admissions in the low-risk group.

faster repeats, in patients classified into the intermediate risk group on presentation. Whilst this algorithm should lead to empowerment of clinicians to exclude NSTE-ACS and discharge during the early stages of presentation, we have yet to observe a significant reduction in overnight admissions in the low-risk group. Whilst there is a clear trend in uptake of the protocol following its implementation, it is evident that it is still not being used universally across the services. This may reflect hesitancy amongst clinicians to discharge patients soon after presentation, without a significant period of monitoring. It is of paramount importance to involve all staff in understanding the rationale for change, optimising operation procedures to ensure rapid turn-around times for sequential blood draws and to streamline a rapid assessment process; in order to reap the benefits of an earlier rule-out. This study looks predominantly at the rule-in/rule-out power of the ESC algorithm at 0 hour, based on a hs-cTnT measurement at presentation. Whilst we have been able to retrospectively quantify the risk classification of patients based on ΔTnT, the translation of ΔTnT values into prospective clinical practice needs further evaluation. Although the ESC recommends a 0 h hs-cTnT ≥52 ng/l as a rule-in threshold, our algorithm defines rule in as >50 ng/l for ease of clinical implementation.

etrospectively quantify the risk classification of patients based on ΔTnT, the translation of ΔTnT values into prospective clinical practice needs further evaluation. Although the ESC recommends a 0 h hs-cTnT ≥52 ng/l as a rule-in threshold, our algorithm defines rule in as >50 ng/l for ease of clinical implementation. Chest pain is clearly the typical presentation of NSTE-ACS. However, the ESC guideline appreciates that ACS can present atypically as ‘epigastric pain, indigestion-like symptoms and isolated dyspnea’.3 The 0–1 h ESC algorithm suggests progression to biomarker risk stratification in the patient with ‘suspected Non-ST elevation Myocardial Infarction’ and does not delineate that this suspicion must arise from the presence of typical chest pain. As such, presenting complaints like isolated shortness of breath and abdominal pain, that feature in Figure 1, can reasonably enter the troponin algorithm if the clinician has a high index of suspicion for ACS. Nonetheless, a limitation of this study is the underlying assumption that all patients who had a hs-cTnT measured in the ED correctly entered the diagnostic algorithm, i.e. had a clinical presentation compatible with a NSTE-ACS. Further, the ‘presenting complaint’ entered on the ED triage system is more a clerical than a medically driven assessment and captures only the main complaint, and not a complex presentation. This may explain why a significant number of patients with an initial troponin in the 5–50 ng/l group did not go on to have a repeat (as it became evident that they should not have entered the algorithm in the first place). Nonetheless, this is likely to represent the reality of a patient’s clinical pathway in ED. The ESC guideline acknowledges that deviation from the protocol is appropriate in circumstances of clinical concern, and rapid rule-out is inappropriate for patients presenting very early after the onset of chest pain. Our study does not account for these possible extenuating circumstances. Importantly, despite our clinical practice moving toward faster repeat troponin measurements, the current study of ΔTnT is based on the repeat troponin at any time within 24 h, whereas the ESC guideline is predicated on a repeat at one hour. In keeping with previously published observations,9 approximately 12% of initial troponin samples taken in the ED were haemolysed.

ward faster repeat troponin measurements, the current study of ΔTnT is based on the repeat troponin at any time within 24 h, whereas the ESC guideline is predicated on a repeat at one hour. In keeping with previously published observations,9 approximately 12% of initial troponin samples taken in the ED were haemolysed. These samples were excluded from analysis as they inevitably lead to deviation from the algorithm, and this study aimed to look at the routine functioning of the algorithm in clinical practice. However, it is important to acknowledge that, in the real-world setting, haemolysis is likely to affect the timings of samples. Finally, the troponin values available electronically to clinicians are rounded to the nearest integer, which may lead to some discrepancy between the true risk bracket that the patient belonged to and the risk bracket that they were ascribed to clinically in ED. Conclusions A 0 hour rule-in/rule-out algorithm, modelled on the 2015 ESC guideline, can be implemented with good uptake within the first few months of introduction. Although this has failed to demonstrate reduced overnight admission in the low-risk group, the algorithm clarifies the appropriate clinical pathway for up to 53% of chest pain patients at presentation. Further studies are needed to address the implications of one-hour repeat testing in routine clinical practice. The authors wish to express their thanks to David Steed and Geoff Martin for their contribution to data collection.

Conclusions A 0 hour rule-in/rule-out algorithm, modelled on the 2015 ESC guideline, can be implemented with good uptake within the first few months of introduction. Although this has failed to demonstrate reduced overnight admission in the low-risk group, the algorithm clarifies the appropriate clinical pathway for up to 53% of chest pain patients at presentation. Further studies are needed to address the implications of one-hour repeat testing in routine clinical practice. The authors wish to express their thanks to David Steed and Geoff Martin for their contribution to data collection. Conflict of interest: MS Marber is named as an inventor on a patent held by King’s College London for the detection of cardiac myosin binding protein-C as a biomarker of myocardial injury. The remaining authors have no conflict of interest to report. Funding: Supported by the UK Department of Health through the National Institute for Health Research Biomedical Research Centre award to Guy’s & St Thomas’ National Health Service Foundation Trust, and the British Heart Foundation (FS/15/13/31320).

Introduction The extent to which first acute myocardial infarction (AMI) with or without ST-elevation is heralded by previous symptomatic atherosclerotic disease, major risk factors, or symptoms has important implications for understanding the aetiology of each phenotype, as well as the provision of optimal services. Studies which retrospectively evaluate medical history suggest that prior atherosclerotic disease is common in people with AMI,1–5 and patterns differ according to ST-elevation myocardial infarction (STEMI) and non-ST-elevation myocardial infarction (NSTEMI). Patients with NSTEMI tend to have higher levels of angina,6,7 heart failure symptoms,8 coronary artery bypass graft (CABG) and percutaneous coronary intervention (PCI),6,9,10 and peripheral vascular disease7 compared to patients with STEMI (Supplementary Table 1, available online). However many of these studies take a single, retrospective snapshot of medical history and have important limitations. They may underestimate the burden of prior disease (i.e. falsely inflating the estimate of unheralded AMI) and may poorly reflect the timing of initial and subsequent manifestations of disease. To our knowledge, no large-scale study to date has evaluated the extent and nature of STEMI and NSTEMI heralding using prospectively collected information on the onset of atherosclerotic disease (in coronary, cerebral, and peripheral circulations) and other risk factors.

nitial and subsequent manifestations of disease. To our knowledge, no large-scale study to date has evaluated the extent and nature of STEMI and NSTEMI heralding using prospectively collected information on the onset of atherosclerotic disease (in coronary, cerebral, and peripheral circulations) and other risk factors. Therefore, this paper aims to compare the evolution of atherosclerotic disease and cardiovascular risk between people going on to experience STEMI and NSTEMI. Using prospectively collected longitudinal primary care data linked to detailed hospital data on acute coronary syndromes, we describe the initial manifestation, distribution and timing of different atherosclerotic presentations before first STEMI and NSTEMI, and the proportion of AMIs that occur without any previously diagnosed atherosclerotic disease, cardiovascular risk factors, or chest pain. Methods Study design As part of the CALIBER research programme (Cardiovascular disease research using Linked Bespoke studies and Electronic Records, www.caliberresearch.org),11 the records of patients presenting with STEMI and NSTEMI in the Myocardial Ischaemia National Audit Project (MINAP) were linked to longitudinal electronic health records from primary care from the General Practice Research Database (GPRD).

using Linked Bespoke studies and Electronic Records, www.caliberresearch.org),11 the records of patients presenting with STEMI and NSTEMI in the Myocardial Ischaemia National Audit Project (MINAP) were linked to longitudinal electronic health records from primary care from the General Practice Research Database (GPRD). MINAP MINAP is the national registry of patients admitted to hospitals in England and Wales with acute coronary syndrome (ACS).12 The MINAP dataset records timing of symptom onset and admission, clinical features and investigations (including ECG results and cardiac biomarkers), past medical history, hospital treatment, and discharge diagnosis.12 GPRD The GPRD is a primary care database containing anonymized patient records from general practices for approximately 8% of the UK population (5.2 million patients).13 General practitioners (GPs) play a key role in the UK healthcare system as they are responsible for primary health care and specialist referrals. Patients are affiliated to a practice, which centralizes the medical information from the GP (diagnoses, symptoms, prescriptions, treatments, and health behaviours), specialist referrals, and hospitalizations, so that GP data provide a comprehensive longitudinal health record. Around 40% of the general practices in GPRD permit linkage of individual patient records with other data sources.14 Data from these practices, all in England, are used in the current study.

ealth behaviours), specialist referrals, and hospitalizations, so that GP data provide a comprehensive longitudinal health record. Around 40% of the general practices in GPRD permit linkage of individual patient records with other data sources.14 Data from these practices, all in England, are used in the current study. Linkage Linkage of MINAP with GPRD permits researchers to establish a longitudinal patient journey before and after ACS, while providing greater clinical detail on ACS events than is reliably available within GPRD. All linked patients had general practice data and were a representative 4% sample of AMI cases from MINAP. The pseudoanonymized dataset was created using a Trusted Third Party to perform the linkage, based on patient NHS number, date of birth, gender, and postcode.15

n ACS events than is reliably available within GPRD. All linked patients had general practice data and were a representative 4% sample of AMI cases from MINAP. The pseudoanonymized dataset was created using a Trusted Third Party to perform the linkage, based on patient NHS number, date of birth, gender, and postcode.15 Definition of acute myocardial infarction STEMI or NSTEMI was defined by details recorded in MINAP, following the joint American Heart Association/European Society of Cardiology definition.16 In order to confine the analysis to first AMI, we excluded patients with a history of AMI noted in their MINAP record, or with evidence of AMI in their GPRD record prior to the first AMI recorded in MINAP. We included patients fulfilling the following criteria: at least 18 years of age at AMI, first AMI occurring between 1 January 2003 and 31 December 2008; registered with the GPRD practice at the time of AMI, with at least one year of observation before AMI and at least one consultation during pre-AMI follow up to allow prevalent diagnoses to be recorded once a patient joins a practice, as patients can register with a practice and not attend for many months or years (Supplementary Figure 1).

PRD practice at the time of AMI, with at least one year of observation before AMI and at least one consultation during pre-AMI follow up to allow prevalent diagnoses to be recorded once a patient joins a practice, as patients can register with a practice and not attend for many months or years (Supplementary Figure 1). Identifying atherosclerotic cardiovascular disease and risk factors in the linked data MINAP and GPRD data were used to identify AMI, other atherosclerotic disease and cardiovascular risk factors among study patients. Any disease, risk factor, or chest pain recorded at any time prior to AMI was defined as ‘heralding’ the AMI. Atherosclerotic disease included cardiac disease (stable angina, unstable angina, cardiac arrest, heart failure, coronary heart disease (CHD) not otherwise specified, receipt of PCI and CABG), ischaemic cerebrovascular disease, including stroke, non-stroke cerebrovascular disease, and transient ischaemic attack, and peripheral arterial disease (PAD), including abdominal aortic aneurysm.

able angina, cardiac arrest, heart failure, coronary heart disease (CHD) not otherwise specified, receipt of PCI and CABG), ischaemic cerebrovascular disease, including stroke, non-stroke cerebrovascular disease, and transient ischaemic attack, and peripheral arterial disease (PAD), including abdominal aortic aneurysm. Risk factors investigated were smoking (categorized as non, ex, current, or unknown at the time of AMI), hypertension (either diagnosed hypertension or three consecutive raised (>140/90 mmHg) measurements), dyslipidaemia (abnormal lipid measurements or management of high lipids), and diabetes (diagnosed diabetes or insulin prescription) and were defined by codes in the primary care or MINAP hospital record. We also determined whether patients had been prescribed blood pressure-lowering, lipid-lowering, or antiplatelet medications in the 6 months before AMI. Missing data from MINAP variables and absence of any diagnostic codes in the GPRD were taken to indicate absence of the risk factor or morbidity.

INAP hospital record. We also determined whether patients had been prescribed blood pressure-lowering, lipid-lowering, or antiplatelet medications in the 6 months before AMI. Missing data from MINAP variables and absence of any diagnostic codes in the GPRD were taken to indicate absence of the risk factor or morbidity. Determining onset and duration of diagnosed atherosclerotic disease before AMI For patients whose AMI was heralded by a diagnosis of atherosclerotic disease, we took the earliest record of any atherosclerotic disease before AMI in the GPRD to be the date of onset (data on timing of prior disease are not recorded in MINAP). Where this code was for a prevalent diagnosis (e.g. ‘history of stroke’) or the morbidity was recorded only in MINAP, the date of onset was recorded as missing. The earliest date of each subtype of atherosclerotic disease (coronary, cerebrovascular, peripheral arterial) was ascertained using the same method. This allowed calculation of the duration of diagnosed disease before AMI and the rate of diagnosis before STEMI and NSTEMI. Consultation or admission for chest pain in the linked data In patients without diagnosed atherosclerotic disease, we assessed the frequency of primary care consultations for chest pain.

Determining onset and duration of diagnosed atherosclerotic disease before AMI For patients whose AMI was heralded by a diagnosis of atherosclerotic disease, we took the earliest record of any atherosclerotic disease before AMI in the GPRD to be the date of onset (data on timing of prior disease are not recorded in MINAP). Where this code was for a prevalent diagnosis (e.g. ‘history of stroke’) or the morbidity was recorded only in MINAP, the date of onset was recorded as missing. The earliest date of each subtype of atherosclerotic disease (coronary, cerebrovascular, peripheral arterial) was ascertained using the same method. This allowed calculation of the duration of diagnosed disease before AMI and the rate of diagnosis before STEMI and NSTEMI. Consultation or admission for chest pain in the linked data In patients without diagnosed atherosclerotic disease, we assessed the frequency of primary care consultations for chest pain. Statistical analysis The proportions with diagnosed atherosclerotic disease and risk factors were calculated for STEMI and NSTEMI patients. Since the age and sex profiles of STEMI and NSTEMI patients differed, we included each atherosclerotic disease/risk factor in turn in an age- and sex-adjusted logistic regression model to determine whether the odds of prior disease/risk factor differed between STEMI and NSTEMI patients, after accounting for age and sex differences. We used the models to assess interaction between age and sex. We also calculated the age- and sex-standardized prevalences of each atherosclerotic disease subtype and risk factor for STEMI and NSTEMI patients, using the age and sex distribution of the study population as the standard.

nting for age and sex differences. We used the models to assess interaction between age and sex. We also calculated the age- and sex-standardized prevalences of each atherosclerotic disease subtype and risk factor for STEMI and NSTEMI patients, using the age and sex distribution of the study population as the standard. To investigate the timing of disease prior to AMI, we calculated rates of new coronary, cerebrovascular, and peripheral arterial disease in 1-year time bands in the period before AMI, and rates of new coronary diagnoses and chest pain consultations in 1-month time bands in the period before AMI. We used Poisson regression to calculate rate ratios and 95% confidence intervals, comparing the rate of coronary diagnosis in the year before AMI to the rate in the previous 9 years, and also to test for linear trend in the rate of diagnosis in the years leading to AMI. All analyses were performed in STATA. The study details are registered online at clinicaltrials.gov (NCT01379131) and a time-stamped detailed analytic protocol is available on request. CALIBER has received ethics approval (ref. 09/H0810/16) for creation of linked pseudoanonymized data encompassing GPRD and MINAP. Results We identified 8174 first AMI patients who met the eligibility criteria. Their median age was 71 years (IQR 59–80), 2946 (36%) were women, and 3780 (46%) had STEMI. The median duration of follow up before AMI was 8.7 years (overall 77,228 person-years of follow up). Table 1 shows the demographic and hospital admission characteristics of patients by AMI type.

ho met the eligibility criteria. Their median age was 71 years (IQR 59–80), 2946 (36%) were women, and 3780 (46%) had STEMI. The median duration of follow up before AMI was 8.7 years (overall 77,228 person-years of follow up). Table 1 shows the demographic and hospital admission characteristics of patients by AMI type. Table 1. Demographics and hospital admission characteristics of STEMI and NSTEMI patients at the time of hospital admission. Characteristic STEMI (n=3780) NSTEMI (n=4394) Age (years) 67.0 (57.0–77.0) 74.0 (63.0–82.0) Female 1172 (31.0) 1774 (40.4) Ethnicity White 3146 (83.2) 3739 (85.1) South Asian 13 (0.3) 18 (0.4) Other 59 (1.6) 63 (1.4) Unknown 562 (14.9) 574 (13.1) ECG at admission ST-segment elevation 3552 (94.0) 0 (0) Left bundle branch block 87 (2.3) 246 (5.6) ST-segment depression 0 (0.0) 1144 (26) T-wave changes only 0 (0.0) 1024 (23.3) Other abnormality 0 (0.0) 836 (19) Normal ECG 0 (0.0) 473 (10.8) Unknown 141 (3.7) 671 (15.3) Peak troponin at admission (µg/l)a 5.2 (1.2–25.0) 1.0 (0.3–3.9) Heart rate at admission (bpm)a 76.0 (63.0–90.0) 80.0 (68.0–98.0) Systolic BP at admission (mmHg)a 138.0 (120.0–157.0) 140.0 (121.0–160.0) Values are median (interquartile range) or n (%). As shown in Table 3, previous treatment with cardiovascular medication was different in STEMI and NSTEMI. a Completeness in peak troponin, heart rate, and systolic BP was 85, 77, and 77%, respectively. BP, blood pressure.

Characteristic STEMI (n=3780) NSTEMI (n=4394) Age (years) 67.0 (57.0–77.0) 74.0 (63.0–82.0) Female 1172 (31.0) 1774 (40.4) Ethnicity White 3146 (83.2) 3739 (85.1) South Asian 13 (0.3) 18 (0.4) Other 59 (1.6) 63 (1.4) Unknown 562 (14.9) 574 (13.1) ECG at admission ST-segment elevation 3552 (94.0) 0 (0) Left bundle branch block 87 (2.3) 246 (5.6) ST-segment depression 0 (0.0) 1144 (26) T-wave changes only 0 (0.0) 1024 (23.3) Other abnormality 0 (0.0) 836 (19) Normal ECG 0 (0.0) 473 (10.8) Unknown 141 (3.7) 671 (15.3) Peak troponin at admission (µg/l)a 5.2 (1.2–25.0) 1.0 (0.3–3.9) Heart rate at admission (bpm)a 76.0 (63.0–90.0) 80.0 (68.0–98.0) Systolic BP at admission (mmHg)a 138.0 (120.0–157.0) 140.0 (121.0–160.0) Values are median (interquartile range) or n (%). As shown in Table 3, previous treatment with cardiovascular medication was different in STEMI and NSTEMI. a Completeness in peak troponin, heart rate, and systolic BP was 85, 77, and 77%, respectively. BP, blood pressure. Acute myocardial infarction occurring with and without heralding As shown in Figure 1, among patients with STEMI, 29% had prior atherosclerotic disease, 56% had no prior atherosclerotic disease diagnosis but at least one cardiovascular risk factor, and 0.6% experienced only chest pain, leaving 14% (95% CI 13–16%) unheralded by these factors. In NSTEMI patients, 50% had previous disease, 40% had no previous disease but at least one cardiovascular risk factor, 0.7% reported only chest pain, and 9% (95% CI 9–10%) experienced AMI unheralded by these factors. Thus NSTEMIs were more often heralded by prior atherosclerotic disease rather than other risk factors only. STEMIs were more likely to be unheralded than NSTEMIs, but the absolute proportions of AMIs unheralded by these factors were low for both types.

and 9% (95% CI 9–10%) experienced AMI unheralded by these factors. Thus NSTEMIs were more often heralded by prior atherosclerotic disease rather than other risk factors only. STEMIs were more likely to be unheralded than NSTEMIs, but the absolute proportions of AMIs unheralded by these factors were low for both types. Figure 1. Previous atherosclerotic disease and risk factors in patients with first ST-elevation myocardial infarction (STEMI, n=3780) and non-ST-elevation myocardial infarction (NSTEMI, n=4394). Diagnosed atherosclerotic disease before first AMI As shown in Table 2, 3326 (41%) of patients had previously diagnosed atherosclerotic disease. Patients with NSTEMI experienced more disease (STEMI 29%, NSTEMI 50%, age- and sex-standardized values 32% and 47%, respectively, p<0.001) and this pattern was consistent across age groups, for men and women and for different atherosclerotic disease manifestations, even after standardizing for age and sex. There was no age–sex interaction. Coronary disease was the most common presentation before AMI, diagnosed in 21% of STEMI patients and 41% of NSTEMI; most of these patients had stable angina (16% in STEMI and 33% in NSTEMI). Although most patients with previous disease had a coronary diagnosis, 9% of STEMIs and 10% of NSTEMIs were heralded only by PAD and/or atherosclerotic cerebrovascular disease.

on before AMI, diagnosed in 21% of STEMI patients and 41% of NSTEMI; most of these patients had stable angina (16% in STEMI and 33% in NSTEMI). Although most patients with previous disease had a coronary diagnosis, 9% of STEMIs and 10% of NSTEMIs were heralded only by PAD and/or atherosclerotic cerebrovascular disease. Table 2. Prevalence (unstandardized and age- and sex-standardized) and duration of diagnosed atherosclerotic disease in patients with first STEMI and NSTEMI, recorded over a median 8.7 years follow up before myocardial infarction, including patients with atherosclerotic disease at more than one site.

on before AMI, diagnosed in 21% of STEMI patients and 41% of NSTEMI; most of these patients had stable angina (16% in STEMI and 33% in NSTEMI). Although most patients with previous disease had a coronary diagnosis, 9% of STEMIs and 10% of NSTEMIs were heralded only by PAD and/or atherosclerotic cerebrovascular disease. Table 2. Prevalence (unstandardized and age- and sex-standardized) and duration of diagnosed atherosclerotic disease in patients with first STEMI and NSTEMI, recorded over a median 8.7 years follow up before myocardial infarction, including patients with atherosclerotic disease at more than one site. STEMI (N=3780) NSTEMI (N=4394) p-value n (%) Standardized prevalence (95% CI) Median disease duration (IQR) n (%) Standardized prevalence (95% CI) Median disease duration (IQR) Any atherosclerotic disease 1112 (29.4) 32.0 (30.5–33.5) 6.2 (2.2–11.7) 2214 (50.4) 47.2 (45.8–48.5) 7.6 (3.2–13.4) <0.001 Coronary disease 788 (20.8) 22.7 (21.3–24) 4.5 (1–8.9) 1795 (40.9) 38.2 (36.8–39.5) 4.2 (1.1–9.3) <0.001 Stable angina 587 (15.5) 16.9 (15.7–18.2) 6.3 (1.4–11.4) 1442 (32.8) 30.8 (29.5–32.1) 7.2 (2.5–13.2) <0.001 Unstable angina 46 (1.2) 1.4 (1.0–1.9) 4.6 (1.8–7.9) 172 (3.9) 3.8 (3.2–4.3) 2.7 (0.3–6.9) <0.001 PCI or CABG 99 (2.6) 2.6 (2.1–3.1) 6.5 (1.5–10.7) 281 (6.4) 6.4 (5.7–7.2) 7.4 (2.0–13.1) <0.001 CHD not otherwise specified 404 (10.7) 11.7 (10.6–12.7) 7.3 (2.8–12.2) 969 (22.1) 20.5 (19.4–21.7) 8.1 (3.5–13.7) <0.001 Heart failure 142 (3.8) 4.6 (3.9–5.4) 4.5 (1.5–9.5) 498 (11.3) 9.9 (9.1–10.7) 4.1 (1.2–7.9) <0.001 Cardiac arrest 3 (0.1) 0.1 (0–0.1) 0.1 (0–8.3) 7 (0.2) 0.2 (0–0.3) 2.3 (0.4–18.8) 0.277 Other atherosclerotic disease 537 (13.9) 15.6 (14.4–16.8) 4.8 (1.8–9.3) 1036 (23.6) 21.7 (20.5–22.8) 5.6 (2.6–9.7) <0.001 Cerebrovascular disease 276 (7.3) 9.5 (8.5–10.5) 5.3 (2.2–11.3) 554 (12.6) 12.6 (11.7–13.5) 6.1 (2.8–10.9) <0.001 Peripheral arterial disease 261 (6.9) 7.7 (6.8–8.6) 4.4 (1.7–8.4) 565 (12.9) 12.0 (11.1–13.0) 6.1 (2.9–10.5) <0.001 Unknown initial presentationa 6 (0.2) 0.2 (0–0.3) 15.8 (12.7–17.6) 23 (0.5) 0.5 (0.3–0.8) 4.4 (2.1–8.1) 0.009 p-values for the association between MI subtype and each presentation (adjusted for age and sex).

ipheral arterial disease 261 (6.9) 7.7 (6.8–8.6) 4.4 (1.7–8.4) 565 (12.9) 12.0 (11.1–13.0) 6.1 (2.9–10.5) <0.001 Unknown initial presentationa 6 (0.2) 0.2 (0–0.3) 15.8 (12.7–17.6) 23 (0.5) 0.5 (0.3–0.8) 4.4 (2.1–8.1) 0.009 p-values for the association between MI subtype and each presentation (adjusted for age and sex). CABG, coronary artery bypass graft; CHD, coronary heart disease; PCI, percutaneous coronary intervention. a Where the only code indicating atherosclerotic disease was unspecific. As shown in Figure 2, 30% of patients were diagnosed with disease in only one arterial bed, 9% in two, and 2% in three. The extent of disease differed by AMI type (age- and sex-adjusted logistic regression p<0.001); overall, 15% of patients with NSTEMI had disease at more than one site, compared to 6% in STEMI. Figure 2. Proportions of patients with ST-elevation myocardial infarction (STEMI, n=3780) and non-ST-elevation myocardial infarction (NSTEMI, n=4394) with different combinations of disease in one, two, or three arterial beds; 71% of STEMI patients and 50% of NSTEMI patients were unheralded by atherosclerotic disease at any site. CHD, coronary heart disease; CVD, cerebrovascular disease; PAD, peripheral arterial disease.

evation myocardial infarction (NSTEMI, n=4394) with different combinations of disease in one, two, or three arterial beds; 71% of STEMI patients and 50% of NSTEMI patients were unheralded by atherosclerotic disease at any site. CHD, coronary heart disease; CVD, cerebrovascular disease; PAD, peripheral arterial disease. Of the 3326 patients with atherosclerotic disease diagnoses, we were able to estimate a date of disease onset for 2891 (87%; 84% STEMI, 89% NSTEMI). Throughout the 10 years preceding infarction, the rates of diagnosis of coronary, cerebrovascular, and peripheral disease were higher in NSTEMI than STEMI (Figure 3). The rates of cerebrovascular disease and PAD remained stable throughout follow up, with an upward trend towards AMI over time (average increase in rate per 1-year time band: 1.06, 95% CI 1.03–1.10; p<0.001). Rates of coronary disease were higher than rates of peripheral or cerebrovascular disease throughout follow up, consistent with the higher prevalence of coronary disease at the time of AMI. Figure 3. Rates of coronary heart disease (CHD), peripheral arterial disease (PAD) and cerebrovascular disease (CVD) in the 10 years before diagnosis of ST-elevation myocardial infarction (STEMI) and non-ST-elevation myocardial infarction (NSTEMI), with 95% confidence intervals. Each time point covers a 1-year time band (1=0–1 years before AMI, 2=1–2 years before AMI, etc).

ral arterial disease (PAD) and cerebrovascular disease (CVD) in the 10 years before diagnosis of ST-elevation myocardial infarction (STEMI) and non-ST-elevation myocardial infarction (NSTEMI), with 95% confidence intervals. Each time point covers a 1-year time band (1=0–1 years before AMI, 2=1–2 years before AMI, etc). In contrast to the patterns observed in cerebrovascular and peripheral diseases, the rate of coronary disease diagnosis rose rapidly in the year before AMI (Figures 3 and 4). Compared to the rate in the previous 9 years, the rate of coronary diagnosis was 4.1-times higher (95% CI 3.3–5.0) in the year before STEMI and 3.6-times higher (3.1–4.2) in the year before NSTEMI. Figure 4A shows that these increases were largely restricted to the 3 months before infarct, during which 159 (2%) patients were first diagnosed with coronary disease or received a coronary intervention (102 stable angina, 17 unstable angina, 26 CHD of unspecified type, 14 PCI/CABG). A similar pattern was observed in the rate of chest pain consultations in patients without diagnosed atherosclerotic disease (Figure 4B).

ich 159 (2%) patients were first diagnosed with coronary disease or received a coronary intervention (102 stable angina, 17 unstable angina, 26 CHD of unspecified type, 14 PCI/CABG). A similar pattern was observed in the rate of chest pain consultations in patients without diagnosed atherosclerotic disease (Figure 4B). Figure 4. Rates of coronary diagnosis (A) and chest pain consultations (B) in the months leading to ST-elevation myocardial infarction (STEMI) and non-ST-elevation myocardial infarction (NSTEMI), with 95% confidence intervals. Consultations for chest pain are only in those without diagnosed atherosclerotic disease. Each time point covers a 1-month time band (0–1 months, 1–2 months, etc). NB: Figure 4B describes chest pain consultations in only patients without previously diagnosed atherosclerotic disease, therefore describing that although these patients have not received a coronary disease diagnosis, they may well be heralded by possible coronary symptoms. Among patients with prior atherosclerotic disease, the median duration between first diagnosis and STEMI was 6.2 years (IQR 2.2–11.7) and in NSTEMI 7.6 years (3.2–13.4) (Table 2). The median duration of all atherosclerotic diseases combined was longer in NSTEMI at all age groups and for men and women (Supplementary Table 2). Importantly, the duration of diagnosed disease tended to be long: 26% of atherosclerotic disease heralding in STEMI and 35% in NSTEMI was 10 or more years’ duration (48% and 57% 5 or more years, respectively).

erotic diseases combined was longer in NSTEMI at all age groups and for men and women (Supplementary Table 2). Importantly, the duration of diagnosed disease tended to be long: 26% of atherosclerotic disease heralding in STEMI and 35% in NSTEMI was 10 or more years’ duration (48% and 57% 5 or more years, respectively). Use of cardiovascular medications in patients with atherosclerotic disease Of those with previously diagnosed atherosclerotic disease, 87% were being prescribed one or more of aspirin, statins, and blood-pressure-lowering treatment in the 6 months before AMI, but only 34% were receiving all three.

erotic diseases combined was longer in NSTEMI at all age groups and for men and women (Supplementary Table 2). Importantly, the duration of diagnosed disease tended to be long: 26% of atherosclerotic disease heralding in STEMI and 35% in NSTEMI was 10 or more years’ duration (48% and 57% 5 or more years, respectively). Use of cardiovascular medications in patients with atherosclerotic disease Of those with previously diagnosed atherosclerotic disease, 87% were being prescribed one or more of aspirin, statins, and blood-pressure-lowering treatment in the 6 months before AMI, but only 34% were receiving all three. Cardiovascular risk factors and medications in patients without previous atherosclerotic disease Fifty-nine per cent of AMIs were unheralded by previously diagnosed atherosclerotic disease (71% STEMI, 95% CI 69–72%; 50% NSTEMI, 95% CI 48–51%). Overall, 79% of these patients had at least one elevated or treated risk factor (ever had a record of diabetes, hypertension, dyslipidaemia, current smoking, or a prescription for statins, blood pressure-lowering, or antiplatelets in the 6 months before AMI). This was the same in STEMI (79%) and NSTEMI (80%) (Table 3). The most common risk factors were diagnosed hypertension or recent use of blood pressure-lowering drugs (42% of STEMI patients and 53% of NSTEMI) and current smoking (39.8% STEMI, 28.3% NSTEMI); one or both of these risk factors was present in 70% of STEMI and 80% of NSTEMI patients. STEMI patients tended to have a slightly lower burden of cardiovascular risk factors than NSTEMI (median two risk factors in STEMI patients, three in NSTEMI).

nts and 53% of NSTEMI) and current smoking (39.8% STEMI, 28.3% NSTEMI); one or both of these risk factors was present in 70% of STEMI and 80% of NSTEMI patients. STEMI patients tended to have a slightly lower burden of cardiovascular risk factors than NSTEMI (median two risk factors in STEMI patients, three in NSTEMI). Table 3. Prospectively collected evaluation of prevalence (standardized and unstandardized) of cardiovascular risk factors and cardiovascular medications in STEMI (n=2268) and NSTEMI (n=2180) patients without previously diagnosed atherosclerotic disease.

nts and 53% of NSTEMI) and current smoking (39.8% STEMI, 28.3% NSTEMI); one or both of these risk factors was present in 70% of STEMI and 80% of NSTEMI patients. STEMI patients tended to have a slightly lower burden of cardiovascular risk factors than NSTEMI (median two risk factors in STEMI patients, three in NSTEMI). Table 3. Prospectively collected evaluation of prevalence (standardized and unstandardized) of cardiovascular risk factors and cardiovascular medications in STEMI (n=2268) and NSTEMI (n=2180) patients without previously diagnosed atherosclerotic disease. STEMI (n=2668) NSTEMI (n=2180) p-value n (%) Age- and sex-standardized prevalence (95% CI) n (%) Age- and sex-standardized prevalence (95% CI) Smoking Non 339 (12.7) 13.4 (12.1–14.6) 338 (15.5) 15.0 (13.5–16.5) <0.001 Former 1261 (47.3) 48.8 (47–50.5) 1219 (55.9) 54.0 (52–56) Current 1063 (39.8) 37.4 (35.7–39) 616 (28.3) 30.5 (28.6–32.3) Unknown 5 (0.2) 0.2 (0–0.4) 7 (0.3) 0.3 (0.1–0.5) Hypertension 1083 (40.6) 41.9 (40.1–43.7) 1082 (49.6) 47.7 (45.7–49.7) <0.001 Dyslipidaemia 569 (21.3) 21.4 (19.9–22.9) 459 (21.1) 21.4 (19.7–23.1) 0.913 Diabetes 276 (10.3) 10.6 (9.4–11.7) 302 (13.9) 13.4 (12–14.7) 0.002 Blood pressure-loweringa 806 (30.2) 31.8 (30.1–33.5) 880 (40.4) 38.3 (36.4–40.2) <0.001 Statinsa 430 (16.1) 16.3 (14.9–17.6) 404 (18.5) 18.4 (16.8–20) 0.043 Antiplateletsa 420 (15.7) 16.3 (14.9–17.7) 509 (23.3) 22.9 (21.1–24.6) <0.001 Chest pain consultationb in 90 days before MI 122 (4.6) 4.7 (3.9–5.4) 158 (7.2) 7.4 (6.3–8.5) <0.001 Without any of these risk factors or cardiovascular medications (% of unheralded MI)a 567 (21.3) 21.4 (19.8–22.9) 443 (20.3) 20.3 (18.6–22) 0.318 Without any of these risk factors or cardiovascular medications (% of all MI)a 567 (15) 14.7 (13.6–15.8) 443 (10.1) 10.8 (9.8–11.7) <0.001 p-values for the association of risk factor with MI subtype (adjusted for age and sex).

medications (% of unheralded MI)a 567 (21.3) 21.4 (19.8–22.9) 443 (20.3) 20.3 (18.6–22) 0.318 Without any of these risk factors or cardiovascular medications (% of all MI)a 567 (15) 14.7 (13.6–15.8) 443 (10.1) 10.8 (9.8–11.7) <0.001 p-values for the association of risk factor with MI subtype (adjusted for age and sex). a Prescribed in the 6 months before MI. b Excluding consultations recorded for administrative and prescription purposes only. Patients without heralding by previous atherosclerotic disease, risk factors, medications, or chest pain In STEMI patients, 546 (14%) were unheralded by all of the factors discussed and in NSTEMI patients, 413 (9%). These patients were more likely to be younger and men than those who were heralded (heralded median age 71 years (IQR 60–80), 63% men; unheralded median age 67 years (IQR 58–77), 72% men). They were also likely to have a lower rate of consultation with the GP in the period leading to AMI (median 7 consultations per year in patients heralded by anything, compared to 4 per year in those unheralded). The proportions of patients by age group are shown in Supplementary Figure 2. We conducted a sensitivity analysis including family history of cardiovascular disease and obesity as cardiovascular risk factors. This reduced the proportion of STEMIs unheralded by disease, risk factors, or chest pain from 14% to 9% and NSTEMIs from 9% to 7%.

Patients without heralding by previous atherosclerotic disease, risk factors, medications, or chest pain In STEMI patients, 546 (14%) were unheralded by all of the factors discussed and in NSTEMI patients, 413 (9%). These patients were more likely to be younger and men than those who were heralded (heralded median age 71 years (IQR 60–80), 63% men; unheralded median age 67 years (IQR 58–77), 72% men). They were also likely to have a lower rate of consultation with the GP in the period leading to AMI (median 7 consultations per year in patients heralded by anything, compared to 4 per year in those unheralded). The proportions of patients by age group are shown in Supplementary Figure 2. We conducted a sensitivity analysis including family history of cardiovascular disease and obesity as cardiovascular risk factors. This reduced the proportion of STEMIs unheralded by disease, risk factors, or chest pain from 14% to 9% and NSTEMIs from 9% to 7%. Discussion We found that heart attack without previous clinically ascertained and recorded atherosclerotic disease was common, but due to the high prevalence of elevated cardiovascular disease risk factors, it was rare for heart attack to occur without warning by disease, risk factors, or chest pain. In the first large-scale evaluation of coronary, cerebral, and peripheral atherosclerotic disease manifestations, risk factors, and symptoms prior to STEMI and NSTEMI using prospectively collected data, we found large differences in the pattern of diagnosed atherosclerotic disease by AMI type in the period leading up to AMI. While the proportion of AMI that occurs without disease, cardiovascular risk factors, medications, or chest pain (i.e. ‘out of the blue’) was slightly higher in STEMI, an important proportion of unheralded AMIs are NSTEMI (14% vs. 9%, respectively). We also found that there was a premonitory period for both AMI types during which the rates of both coronary disease diagnosis and chest pain consultation were raised, but there was no equivalent increase in the rate of peripheral artery or cerebrovascular disease diagnoses.

AMIs are NSTEMI (14% vs. 9%, respectively). We also found that there was a premonitory period for both AMI types during which the rates of both coronary disease diagnosis and chest pain consultation were raised, but there was no equivalent increase in the rate of peripheral artery or cerebrovascular disease diagnoses. Previous atherosclerotic disease and risk factors Uniquely, our study provided prospective data on the rate of onset of different subtypes of atherosclerotic disease in the years leading to AMI. Patients with NSTEMI had a consistently higher rate of coronary, cerebrovascular and peripheral disease diagnosis throughout follow up compared to STEMI. This is in line with other studies showing that patients with NSTEMI are more likely to have prior atherosclerotic disease than STEMI patients (Supplementary Table 1).6–10,17,18 Our results describing the extent of disease across vascular territories are also similar to published findings for NSTEMI,5 and we have shown that in STEMI patients, disease in two or more sites is less common. This is consistent with the idea that NSTEMI patients tend to be a sicker group overall. The widely different pattern in the prevalence and rate of onset of atherosclerotic disease between AMI types lends support to the hypothesis that STEMI and NSTEMI are two different pathophysiological entities (NSTEMI is more often caused by a non-occlusive thrombus and STEMI is more often caused by a complete occlusion19).

idely different pattern in the prevalence and rate of onset of atherosclerotic disease between AMI types lends support to the hypothesis that STEMI and NSTEMI are two different pathophysiological entities (NSTEMI is more often caused by a non-occlusive thrombus and STEMI is more often caused by a complete occlusion19). If AMI occurs without prior symptomatic atherosclerotic disease, to what extent can it be considered to occur ‘out of the blue’? Although a substantial proportion of infarcts were unheralded by diagnosed atherosclerotic disease, the majority of these had at least one cardiovascular risk factor (smoking, hypertension, dyslipidaemia, diabetes) or were being treated with a cardiovascular medication. Our sensitivity analysis indicated that inclusion of a family history of cardiovascular disease and obesity as risk factors increased this majority. The relatively high prescription of antiplatelets before both STEMI and NSTEMI indicates that GPs suspected a high risk of atherosclerotic disease in many of these patients. Our findings are consistent with other prospective studies showing high risk factor burdens in AMI patients overall (Supplementary Table 3).20,21 To our knowledge, there are no other estimates for the proportion of STEMI and NSTEMI occurring without heralding. We have shown that unheralded AMI is uncommon, occurring in roughly one in 10 patients in our study, and more often in younger men. The true prevalence of unheralded AMI is likely to be lower than our data suggest because our data were from general practice where risk factors are recorded opportunistically during patient consultations.

t unheralded AMI is uncommon, occurring in roughly one in 10 patients in our study, and more often in younger men. The true prevalence of unheralded AMI is likely to be lower than our data suggest because our data were from general practice where risk factors are recorded opportunistically during patient consultations. Premonitory period Clinical experience and retrospective studies have long suggested that AMI might be preceded by premonitory symptoms of chest pain presenting to a family physician or ambulatory care.22,23 Our study extends knowledge in several respects. First, we confirmed this association with prospective data. Second, we found that there were increases in coronary disease diagnoses and chest pain consultations in both STEMI and NSTEMI. This is in contrast to the widely held view that STEMI is usually of sudden onset. Third, we showed that the increases were specific to coronary diagnoses and chest pain, rather than disease in cerebral or peripheral circulations, suggesting a local rather than systemic pro-thrombotic state.

s in both STEMI and NSTEMI. This is in contrast to the widely held view that STEMI is usually of sudden onset. Third, we showed that the increases were specific to coronary diagnoses and chest pain, rather than disease in cerebral or peripheral circulations, suggesting a local rather than systemic pro-thrombotic state. Clinical implications and missed opportunities for care? In patients with previously diagnosed atherosclerotic disease or risk factors, AMI represents the unmet potential of secondary or primary prevention, respectively. Despite a clear premonitory period where many patients were diagnosed with coronary disease shortly before AMI, the majority of disease was diagnosed long in advance of both AMI types. Therefore, there is an extended period during which secondary prevention could be implemented. Our data describing the use of secondary prevention measures in the 6 months before AMI showed that most patients with diagnosed atherosclerotic disease were receiving one of either statins, aspirin, or blood pressure-lowering drugs, but only a third were in receipt of all three, indicating that there are likely to be missed opportunities for secondary prevention in this group.

in the 6 months before AMI showed that most patients with diagnosed atherosclerotic disease were receiving one of either statins, aspirin, or blood pressure-lowering drugs, but only a third were in receipt of all three, indicating that there are likely to be missed opportunities for secondary prevention in this group. Interestingly, while coronary disease was the most common pre-AMI presentation, 10% of both STEMIs and NSTEMIs were heralded by peripheral artery disease and/or cerebrovascular disease alone. This emphasizes the importance of further efforts to improve secondary prevention following diagnoses in the cerebral and peripheral arteries in order to prevent an important proportion of AMI. The high prevalence of risk factors in both STEMI and NSTEMI suggests the importance of tackling the widely reported missed opportunities for implementation of existing interventions known to be effective.24–27 Additionally, the categorization of continuous measures in this analysis may have been an over-simplification of cardiovascular risk. Although a binary indicator is simple to interpret in studies and a useful basis on which to prescribe treatment, it does not reflect the continuum of risk over the full range of measurements. A more detailed investigation of these risk factors might reveal borderline raised risk in many patients and lowering blood pressure and lipids in those not diagnosed as hypertensive or dyslipidaemic may also prevent AMI. However, the implications of our analysis are limited by a lack of comparison to AMI-free controls. Such a comparison may allow further conclusions to be drawn from these data.

ised risk in many patients and lowering blood pressure and lipids in those not diagnosed as hypertensive or dyslipidaemic may also prevent AMI. However, the implications of our analysis are limited by a lack of comparison to AMI-free controls. Such a comparison may allow further conclusions to be drawn from these data. Strengths The main strength of this study is the quality of data from the linked MINAP and GPRD records. MINAP collects data from all hospitals in England and undergoes annual assessments to ensure the data are of research quality.28 ECG and cardiac marker results are recorded and our STEMI and NSTEMI case definitions were based on the international definition of AMI.16 The recording of admission date in MINAP allowed us to interpret the timing of previous atherosclerotic disease diagnoses in relation to AMI. The GPRD is representative of the UK population13 and roughly half of GPRD practices consented to linkage with MINAP; patients in practices that participated in the linkage were representative of the GPRD as a whole.14

allowed us to interpret the timing of previous atherosclerotic disease diagnoses in relation to AMI. The GPRD is representative of the UK population13 and roughly half of GPRD practices consented to linkage with MINAP; patients in practices that participated in the linkage were representative of the GPRD as a whole.14 The primary care GPRD data are collected prospectively as part of usual clinical care and therefore are not subject to recall bias or differential error related to outcome. Data regarding new diagnoses and treatment of disease were available for a median of 8.7 years before AMI, allowing sufficient time for incident diagnoses to arise and be recorded. The GPRD closely monitors data quality and the recording of a wide range of atherosclerotic disease outcomes have undergone validation in GPRD studies, which, for example, have compared the electronic data to paper-based medical records or compared the rate of a condition in the GPRD to an external source. These have shown most diagnoses to be of high quality.29–32 For the recording of cardiovascular disease, concordance between GPRD and MINAP was over 90%, and for risk factors and medications was over 80% (assuming missingness in MINAP was concordant with a complete GPRD record); these values represent further evidence of data quality.

The primary care GPRD data are collected prospectively as part of usual clinical care and therefore are not subject to recall bias or differential error related to outcome. Data regarding new diagnoses and treatment of disease were available for a median of 8.7 years before AMI, allowing sufficient time for incident diagnoses to arise and be recorded. The GPRD closely monitors data quality and the recording of a wide range of atherosclerotic disease outcomes have undergone validation in GPRD studies, which, for example, have compared the electronic data to paper-based medical records or compared the rate of a condition in the GPRD to an external source. These have shown most diagnoses to be of high quality.29–32 For the recording of cardiovascular disease, concordance between GPRD and MINAP was over 90%, and for risk factors and medications was over 80% (assuming missingness in MINAP was concordant with a complete GPRD record); these values represent further evidence of data quality. Our analysis was based on patients with ‘definite’ atherosclerotic disease diagnoses, using diagnostic codes which had been rated by two clinicians as being indicative of disease. If ‘possible’ diagnoses were included, the proportion with previous disease rose from 41% to 44%; this small change indicates that our ‘definite’ atherosclerotic disease definition had high sensitivity.

otic disease diagnoses, using diagnostic codes which had been rated by two clinicians as being indicative of disease. If ‘possible’ diagnoses were included, the proportion with previous disease rose from 41% to 44%; this small change indicates that our ‘definite’ atherosclerotic disease definition had high sensitivity. Weaknesses This study included only patients hospitalized with their first AMI recorded in MINAP data and therefore our results cannot be generalized to patients who die outside hospital or those with recurrent AMI, in whom the prevalence of heralding factors is likely to differ. Data were not available in this study to describe the numbers of patients who had out of hospital fatal AMI, but data from a further CALIBER study33 suggest that MINAP captures 30% of patients with AMI recorded as their cause of death.

with recurrent AMI, in whom the prevalence of heralding factors is likely to differ. Data were not available in this study to describe the numbers of patients who had out of hospital fatal AMI, but data from a further CALIBER study33 suggest that MINAP captures 30% of patients with AMI recorded as their cause of death. Because our analyses of heralding are based largely on general practice data, symptomatic atherosclerotic disease may be undiagnosed if patients do not consult their GP. We excluded only fourteen patients without any consultations as these patients never had an opportunity for measurement of risk factors or morbidity. However, introducing a minimum consultation rate could introduce a bias towards sicker patients. The data available for this analysis did not allow us to differentiate between patients with a low consultation rate because of good health and those that did not consult despite symptomatic disease. Excluding patients with less than one year of follow up prior to AMI may also have introduced a selection bias if patients who tend to move practices more frequently are different to patients who stay in a practice for longer periods. However, shortening this time period would likely lead to misclassification of disease and an absence of cardiovascular risk factor records as the GP would not have sufficient time to record these.

f patients who tend to move practices more frequently are different to patients who stay in a practice for longer periods. However, shortening this time period would likely lead to misclassification of disease and an absence of cardiovascular risk factor records as the GP would not have sufficient time to record these. Implications for research A small but important proportion of STEMI and NSTEMI do appear to occur with no recognized heralding signs and further research is warranted to better characterize these phenotypes, their causes and their prognosis. For patients with different forms of heralding the challenge remains to better characterize short term risk of coronary events in order to identify for which patients this represents a (potentially remediable) premonitory period. A research priority would be an analysis comparing AMI patients to a control group without AMI, including a comparison of missed opportunities for care in measuring and controlling elevated risk. Patients with NSTEMI were more likely to have been in receipt of cardiovascular medications but the causal relationship between medications and severity of AMI is unclear. CALIBER data present a new opportunity to study the effects of medications in a population for whom detailed information regarding risk factors and prescription of medications are collected. Conclusion The majority of STEMIs and NSTEMI were heralded by prior disease or at least one other risk factor, suggesting that opportunities for prevention may be being missed.

Patients with NSTEMI were more likely to have been in receipt of cardiovascular medications but the causal relationship between medications and severity of AMI is unclear. CALIBER data present a new opportunity to study the effects of medications in a population for whom detailed information regarding risk factors and prescription of medications are collected. Conclusion The majority of STEMIs and NSTEMI were heralded by prior disease or at least one other risk factor, suggesting that opportunities for prevention may be being missed. AT acknowledges the support of Barts and the London Cardiovascular Biomedical Research Unit, which is funded by the National Institute for Health Research. Conflict of interest: The authors declare that there is no conflict of interest.

Conclusion The majority of STEMIs and NSTEMI were heralded by prior disease or at least one other risk factor, suggesting that opportunities for prevention may be being missed. AT acknowledges the support of Barts and the London Cardiovascular Biomedical Research Unit, which is funded by the National Institute for Health Research. Conflict of interest: The authors declare that there is no conflict of interest. Funding: This work was supported by the UK National Institute for Health Research (grant number RP-PG-0407-10314) and the Wellcome Trust (086091/Z/08/Z). EH is supported by a Medical Research Council studentship. LS is supported by a senior clinical fellowship from the Wellcome Trust. KB is supported by a post-doctoral fellowship from the National Institute for Health Research. JG is supported by a doctoral fellowship from the National Institute for Health Research (DRF-2009-02-50). The work of SD, AT, HH, and LS is supported by CHAPTER (Centre for Health service and Academic Partnership in Translational E-Health Research), part of Health eResearch Centre Network (HeRC-UK), funded by the Medical Research Council, in partnership with Arthritis Research UK, the British Heart Foundation, Cancer Research UK, the Economic and Social Research Council, the Engineering and Physical Sciences Research Council, the National Institute of Health Research, the National Institute for Social Care and Health Research (Welsh Assembly Government), the Chief Scientist Office (Scottish Government Health Directorates), and the Wellcome Trust.

e Economic and Social Research Council, the Engineering and Physical Sciences Research Council, the National Institute of Health Research, the National Institute for Social Care and Health Research (Welsh Assembly Government), the Chief Scientist Office (Scottish Government Health Directorates), and the Wellcome Trust. Department of Health disclaimer: The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR PHR Programme or the Department of Health.

Introduction The occurrence of dyspnoea in acute coronary syndrome (ACS) patients has always been considered a challenging diagnostic and therapeutic clinical scenario. P2Y12 platelet receptor inhibitors (i.e., clopidogrel, prasugrel and ticagrelor) are currently the cornerstone of treatment of ACS patients. Thus, in the last few years, the potential association between ACS and dyspnoea has become also more challenging with the increasing use of ticagrelor in these patients,1,2 due to its beneficial effects on ischaemic event prevention and mortality, since ticagrelor can induce dyspnoea as a side effect.3 The present article is intended to review the current literature regarding dyspnoea in ACS patients, especially those treated with ticagrelor, and to propose ticagrelor-associated dyspnoea management recommendations based on current knowledge.

ntion and mortality, since ticagrelor can induce dyspnoea as a side effect.3 The present article is intended to review the current literature regarding dyspnoea in ACS patients, especially those treated with ticagrelor, and to propose ticagrelor-associated dyspnoea management recommendations based on current knowledge. Epidemiology of dyspnoea in acute coronary syndrome patients Dyspnoea is one of the most common and distressing symptoms experienced by patients and can result from a variety of conditions, including cardiac, pulmonary, renal and liver diseases, anaemia and metabolic abnormalities. A substantial proportion (at least 25%) of patients with ACS may present with dyspnoea as the predominant symptom.4 Moreover, ACS patients may develop dyspnoea during the index hospitalization or in the following weeks due to the development of heart failure, lung infection, adverse reaction to beta-blockers, recurrent ischaemia, anaemia or other potential complications. Patients with ACS who present with dyspnoea as their principal symptom are less likely to be recognized as having a coronary event, less likely to receive evidence-based treatments and more likely to experience poor outcomes.5 In the Prospective Registry Evaluating Myocardial Infarction: Events and Recovery (PREMIER),6 1835 unselected patients who survived an acute myocardial infarction had 1-month dyspnoea assessment using the Rose Dyspnoea Scale. In this study, 863 (47%) patients reported experiencing dyspnoea, with 340 (19%) noting moderate to severe limitation due to dyspnoea. Dyspnoea scores at 1 month were associated with an increased risk of rehospitalization and mortality at long-term follow-up. Moreover, although risk adjustment attenuated this association, even after adjustment for all relevant clinical and sociodemographic factors, the relationship between dyspnoea and impaired quality of life scores remained robust.

ssociated with an increased risk of rehospitalization and mortality at long-term follow-up. Moreover, although risk adjustment attenuated this association, even after adjustment for all relevant clinical and sociodemographic factors, the relationship between dyspnoea and impaired quality of life scores remained robust. In a real-world setting, 17% of patients with ST-elevation acute myocardial infarction present with heart failure symptoms at admission and, despite optimal reperfusion, further patients may develop heart failure during hospitalization.7 The CRUSADE (Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA guidelines) Investigators reported that about 25% of patients with non-ST segment elevation (NSTE) ACS in contemporary practice in the United States present with signs of congestive heart failure or develop in-hospital heart failure.8 Finally, data from the PLATelet inhibition and patient Outcome (PLATO) study3 reported that a significant proportion (more than 20%) of enrolled patients had dyspnoea prior to ACS onset, and those reporting dyspnoea prior to the index event were more likely also to experience dyspnoea after randomization. Dyspnoea during the study period was more likely to occur in the elderly, obese or smokers, and in patients with a history of congestive heart failure, asthma, chronic obstructive pulmonary disease or renal disease, and thus in subjects with dyspnoea prior to enrolment. Again, dyspnoea post randomization in clopidogrel-treated patients was associated with a poor outcome, including increased mortality, as expected with the presence of comorbidities.

tive heart failure, asthma, chronic obstructive pulmonary disease or renal disease, and thus in subjects with dyspnoea prior to enrolment. Again, dyspnoea post randomization in clopidogrel-treated patients was associated with a poor outcome, including increased mortality, as expected with the presence of comorbidities. Dyspnoea with ticagrelor Dyspnoea is a very common ticagrelor side effect (see Table 1). In phase 2 studies, ticagrelor was associated with a dose-dependent incidence of dyspnoea of 10 to 20%, compared with 0–6.4% in patients treated with clopidogrel.9,10 Ticagrelor-related dyspnoea is generally described as sudden and unexpected air hunger or unsatisfied inspiration. Its pattern may vary widely, from very brief episodes lasting minutes, generally starting in the first week of treatment, to sustained or intermittent episodes occurring over several weeks, with most episodes being reported as mild.3 In the ONSET/OFFSET study,11 only 18% of dyspnoea episodes occurring in patients treated with ticagrelor were reported as moderate (no severe dyspnoea was reported). In this study, only three (14%) episodes of dyspnoea in ticagrelor-treated patients were persistent and only three patients required drug discontinuation due to dyspnoea. Generally, ticagrelor-related dyspnoea is not associated with wheezing, orthopnoea, paroxysmal nocturnal dyspnoea, or chest tightness or pain. Moreover, it usually occurs at rest, and is typically not related to exertion and does not limit exercise capacity. The exact mechanism of ticagrelor-related dyspnoea has not been definitively proven. Current hypotheses include stimulation of pulmonary vagal C fibres by increased levels of extracellular adenosine due to ticagrelor’s known antagonism of adenosine reuptake via equilibrative nucleoside transporter-1 (ENT-1)3,11 or the inhibition of P2Y12 receptors located on C fibres of sensory neurons.12 The reversible nature of sensory neuron P2Y12 receptor inhibition could play a role, since cangrelor and elinogrel (other similar reversible agents) also increase dyspnoea occurrence.12

via equilibrative nucleoside transporter-1 (ENT-1)3,11 or the inhibition of P2Y12 receptors located on C fibres of sensory neurons.12 The reversible nature of sensory neuron P2Y12 receptor inhibition could play a role, since cangrelor and elinogrel (other similar reversible agents) also increase dyspnoea occurrence.12 Table 1. Comparison of dyspnoea incidence in patients treated with ticagrelor and clopidogrel. Study (reference) Patients (number) Ticagrelor dose Duration of treatment Dyspnoea with ticagrelor (%) Dyspnoea with clopidogrel (%) DISPERSE9 Patients with atherosclerosis (200) 50 mg bid 4 weeks 10 0 100 mg bid 10 0 200 mg bid 16 0 400 mg bid 20 0 DISPERSE-210 NSTE-ACS (990) 90 mg bid 12 weeks 10.5 6.4 180 mg bid 15.8 6.4 ONSET/OFFSET11 Stable CAD (123) 90 mg bid 6 weeks 38.6 9.3 RESPOND14 Stable CAD (98) 90 mg bid 2 weeks 13 4 PLATO3 ACS (18624) 90 mg bid 12 months 13.8 7.8 ACS: acute coronary syndrome; bid: twice daily; CAD: coronary artery disease; NSTE: non-ST segment elevation; RESPOND: Response to Ticagrelor in Clopidogrel Nonresponders and Responders and Effect of Switching Therapies. Current evidence for and against increased extracellular adenosine due to ENT-1 inhibition by ticagrelor as the underlying mechanism for ticagrelor-induced dyspnoea is presented in Table 2. Table 2. Evidence for and against the hypothesis that adenosine reuptake inhibition is the underlying mechanism for ticagrelor-related dyspnoea.

Current evidence for and against increased extracellular adenosine due to ENT-1 inhibition by ticagrelor as the underlying mechanism for ticagrelor-induced dyspnoea is presented in Table 2. Table 2. Evidence for and against the hypothesis that adenosine reuptake inhibition is the underlying mechanism for ticagrelor-related dyspnoea. Topic For adenosine hypothesis Against adenosine hypothesis Effects of adenosine Intravenous adenosine infusion induces similar dyspnoea to ticagrelor-induced dyspnoea and ticagrelor augments adenosine-induced dyspnoea Pulmonary vagal C fibres These fibres are believed to mediate adenosine-induced dyspnoea and likely mediate ticagrelor-induced dyspnoea The hypothetical presence of P2Y12 receptors on these fibres could directly mediate effects of reversibly binding inhibitors Thienopyridines Thienopyridines (ticlopidine, prasugrel, clopidogrel) do not cause dyspnoea despite often achieving high levels of P2Y12 inhibition The short plasma half-lives of thienopyridine active metabolites may limit their ability to inhibit P2Y12 receptors on C fibres, particularly if these are constantly recycling, explaining why thienopyridines do not induce dyspnoea Effects of cangrelor and elinogrel The principal metabolite of cangrelor also inhibits adenosine reuptake and is present at a higher plasma concentration than cangrelor; the effects of elinogrel and its metabolites on adenosine reuptake have not been reported Cangrelor and elinogrel belong to different chemical classes from ticagrelor yet also induce the same type of dyspnoea Dipyridamole Multiple adenosine receptors mediate effects of different concentrations of adenosine and are susceptible to heterologous desensitization, so higher levels of extracellular adenosine with dipyridamole might hypothetically abolish dyspnoea Dipyridamole is a more potent inhibitor of adenosine reuptake than ticagrelor yet has not been reported to cause dyspnoea How to detect ticagrelor-related dyspnoea It is important to note that, in the context of clinical practice where clinicians are not blinded to the prescribed medications, there may be a propensity to attribute dyspnoea to ticagrelor when in reality it is due to clinical factors, such as pulmonary congestion in a patient with new-onset heart failure after myocardial infarction. Moreover, anaemia, intercurrent pneumonia, worsening of pre-existing chronic pulmonary diseases, and metabolic abnormalities may also contribute to the development of dyspnoea.

n in reality it is due to clinical factors, such as pulmonary congestion in a patient with new-onset heart failure after myocardial infarction. Moreover, anaemia, intercurrent pneumonia, worsening of pre-existing chronic pulmonary diseases, and metabolic abnormalities may also contribute to the development of dyspnoea. For proper management of patients, it is therefore crucial to consider all clinically plausible causes of dyspnoea before attributing it to the medication itself.

n in reality it is due to clinical factors, such as pulmonary congestion in a patient with new-onset heart failure after myocardial infarction. Moreover, anaemia, intercurrent pneumonia, worsening of pre-existing chronic pulmonary diseases, and metabolic abnormalities may also contribute to the development of dyspnoea. For proper management of patients, it is therefore crucial to consider all clinically plausible causes of dyspnoea before attributing it to the medication itself. The evaluation of a patient with dyspnoea continues to be dependent on a thorough history and physical examination. For initial assessment of dyspnoea, it is pivotal to ask the patient whether the identical symptom was present before starting ticagrelor. If this is the case, the relationship between ticagrelor and dyspnoea becomes unlikely and alternative causes of dyspnoea should be considered and assessed. Then, it is very important to assess dyspnoea characteristics. Typical ticagrelor-related dyspnoea characteristics have been described in detail in the previous section. Frequently, the diagnosis of ticagrelor-related dyspnoea is based on exclusion. In addition to the patient’s interview, the exclusion of alternative dyspnoea causes may be performed by physical examination and other analyses and tests. The clinical examination, which can be easily performed during hospital stay and as an outpatient, is able to assess the likelihood of conditions such as heart failure or significant pulmonary disease associated with bronchospasm, emphysema and pneumonia. Simple blood sample examinations, especially during the initial phase of treatment occurring during hospitalization, are able to confirm heart failure (increased N-terminal of the prohormone brain natriuretic peptide (NT-pro-BNP)), anaemia (low red blood cell count) or respiratory failure (low oxygenation with or without hypercapnia). Despite the excess of dyspnoea in the ticagrelor group in PLATO, quality of life scores were no different between the ticagrelor and clopidogrel groups after adjustment for the improved life expectancy in the ticagrelor group.13 It has been clearly documented that ticagrelor does not induce any adverse change in cardiac or pulmonary function that may cause dyspnoea either in patients with stable coronary artery disease or in patients with acute coronary syndromes.11,14 The assessment of left ventricular systolic and diastolic function, as well as valvular morphology and function, by transthoracic echocardiography provides further information and help in heart failure diagnosis of patients with ACS.

with stable coronary artery disease or in patients with acute coronary syndromes.11,14 The assessment of left ventricular systolic and diastolic function, as well as valvular morphology and function, by transthoracic echocardiography provides further information and help in heart failure diagnosis of patients with ACS. Finally, chest X-ray and spirometry are useful tests to assess patients with dyspnoea and can be selectively used in difficult cases reporting dyspnoea after an ACS. When to consider ticagrelor discontinuation Ongoing studies have been designed to determine the main mechanism of ticagrelor-related dyspnoea. Other studies are also evaluating the possibility of pharmacologically treating this side effect in order to increase patient compliance: for example, using caffeine and other xanthine derivatives that oppose adenosine effects, hypothesizing a role of adenosine in the development of dyspnoea. However, until new data are available, currently the only way to manage persistent and intolerable ticagrelor-related dyspnoea is drug discontinuation.

compliance: for example, using caffeine and other xanthine derivatives that oppose adenosine effects, hypothesizing a role of adenosine in the development of dyspnoea. However, until new data are available, currently the only way to manage persistent and intolerable ticagrelor-related dyspnoea is drug discontinuation. In the case of an ACS patient treated with ticagrelor and reporting typical ticagrelor-related dyspnoea, after excluding alternative dyspnoea causes and pre-existing dyspnoea, we recommend taking some time to allow the possibility of spontaneous improvement in dyspnoea. In many cases, ticagrelor-related dyspnoea is transient and lasts a few hours or days, generally occurring within the first week of treatment. Thus, if the clinical conditions allow it, we can wait 3–4 days while observing the patient to understand whether the dyspnoea is transient or permanent. During this time period, alternative causes of dyspnoea can also emerge. In the case of permanent dyspnoea as a ticagrelor side effect, the following question should be considered: can the patient tolerate the dyspnoea, accepting the potential benefit in terms of reduced mortality risk? Ticagrelor-related dyspnoea is generally mild, sometime moderate and very occasionally severe.3 In the majority of cases, patients tolerate the mild discomfort associated with the dyspnoea, and it is reasonable not to discontinue the drug, allowing the patient to continue to benefit from ticagrelor therapy and to have an optimal secondary prevention strategy. Only in the case of either persistent and intolerable ticagrelor-related dyspnoea or severe initial dyspnoea deemed likely to be provoked by ticagrelor should drug discontinuation be considered. Ticagrelor-related dyspnoea, if sustained during treatment, tends to resolve after the medication is discontinued, and there has been no evidence of any compromise to pulmonary or cardiac function that would lead to sustained dyspnoea after drug discontinuation.11 In the case of drug discontinuation, if dyspnoea improvement occurs within a few days, compatible with ticagrelor clearance time, this clearly supports the diagnosis of ticagrelor-related dyspnoea. In Figure 1, we present a proposed algorithm for dyspnoea management in ACS patients treated with ticagrelor.

tinuation.11 In the case of drug discontinuation, if dyspnoea improvement occurs within a few days, compatible with ticagrelor clearance time, this clearly supports the diagnosis of ticagrelor-related dyspnoea. In Figure 1, we present a proposed algorithm for dyspnoea management in ACS patients treated with ticagrelor. Figure 1. Dyspnoea diagnostic flow-chart. Consequences of switching from ticagrelor to clopidogrel or prasugrel The substantially lower mortality rates in ticagrelor-treated patients with dyspnoea compared with clopidogrel-treated patients with dyspnoea observed in the PLATO study3 are consistent with three effects: first, the treatment benefit of ticagrelor compared with clopidogrel seen in the overall trial appears to be preserved in ticagrelor-treated patients with dyspnoea compared with those without dyspnoea; second, there is a favourable mortality prognosis in patients with ticagrelor-related dyspnoea compared with other causes of dyspnoea; and, third, high-risk patients such as those reporting dyspnoea need to be treated with the most effective pharmacological strategies. The fact that there was no evidence that the mortality benefit associated with ticagrelor in the PLATO trial was attenuated in the subgroup of patients with dyspnoea suggests that patients with tolerable dyspnoea should be encouraged to continue ticagrelor, while patients who cannot tolerate dyspnoea that is believed to be an adverse effect of the drug may be switched to either prasugrel or clopidogrel.

lor in the PLATO trial was attenuated in the subgroup of patients with dyspnoea suggests that patients with tolerable dyspnoea should be encouraged to continue ticagrelor, while patients who cannot tolerate dyspnoea that is believed to be an adverse effect of the drug may be switched to either prasugrel or clopidogrel. Switching patients with ticagrelor-related persistent and intolerable dyspnoea to clopidogrel may be an option. In the medical literature there are very few data regarding the strategy of switching ACS patients from ticagrelor to clopidogrel. Consequently, the clinical consequences of this procedure are poorly explored. Platelet function tests may be considered in selected patients at about 5 days after switching from ticagrelor to clopidogrel. Currently, the optimal length of a dual antiplatelet therapy after an ACS is unknown; however, in the PLATO study ticagrelor benefits over clopidogrel were evident in the first 30 days but clearly increased from 1 to 12 months, including in patients reporting dyspnoea.3 We should keep in mind that, if a patient was initially selected for ticagrelor therapy due to a perceived high-risk profile, there are no meaningful reasons to downgrade the antiplatelet strategy to clopidogrel, except in the case of either a dangerous or an intolerable side effect or development of a contraindication to ticagrelor.

in mind that, if a patient was initially selected for ticagrelor therapy due to a perceived high-risk profile, there are no meaningful reasons to downgrade the antiplatelet strategy to clopidogrel, except in the case of either a dangerous or an intolerable side effect or development of a contraindication to ticagrelor. In the RESPOND study,15 ticagrelor therapy was associated with greater platelet inhibition as compared with clopidogrel treatment in both ‘responders’ and ‘non-responders’ to clopidogrel, as defined by the pharmacodynamic response to a clopidogrel loading dose. Moreover, switching to clopidogrel therapy was associated with a reduction in drug-induced platelet inhibition. Finally, ticagrelor was extremely effective in reducing the prevalence of high residual platelet reactivity using previously defined cutoffs; nearly all patients during ticagrelor therapy, irrespective of clopidogrel response status, had platelet reactivity below the cutoffs associated with increased ischaemic risk. Thus, switching from a new and more potent antiplatelet agent such as ticagrelor (or prasugrel) to clopidogrel is ‘navigating in unknown waters’,16 since it is currently unknown whether the subsequent increase in platelet reactivity soon after an acute coronary event might potentially lead to ischaemic events following recovery of platelet reactivity. As a matter of fact, switching patients perceived as being at high risk of bleeding from prasugrel to clopidogrel led to a 10-fold increase in average platelet aggregation, frankly unmasking poor responders to clopidogrel,17 a subset of patients known to be at higher risk of thrombotic events, including coronary stent thrombosis.18

f fact, switching patients perceived as being at high risk of bleeding from prasugrel to clopidogrel led to a 10-fold increase in average platelet aggregation, frankly unmasking poor responders to clopidogrel,17 a subset of patients known to be at higher risk of thrombotic events, including coronary stent thrombosis.18 Eventually, in the case of a non-haemorrhagic ticagrelor side-effect, such as dyspnoea, there is a greater rationale to switch many ACS patients to prasugrel as compared with downgrading to clopidogrel, assuming no contraindication to prasugrel. The Switching AntiPlatelet-2 (SWAP-2) trial was a pharmacodynamic study that addressed the issue of switching coronary artery disease patients from ticagrelor to prasugrel.19 The study results suggested a pharmacodynamic interaction when switching from ticagrelor to prasugrel that is only partially mitigated when a prasugrel loading dose is used. In fact, during the early switching phase and up to 7 days of study treatment, prasugrel was associated with significantly higher platelet reactivity as compared with ticagrelor, consistent with observed differences in mean platelet reactivity between prasugrel and ticagrelor reported in other studies. The results of all switching pharmacodynamic studies should be confirmed by pharmacokinetic and, more importantly, by larger-scale clinical studies. However, a negative interaction between ticagrelor and either clopidogrel or prasugrel is highly probable: it is possible that occupancy of P2Y12 receptors by ticagrelor might prevent the active metabolites of clopidogrel or prasugrel from binding to the receptor during the early switching phase, as has been demonstrated for the reversibly binding P2Y12 inhibitor cangrelor.20 Hypothetically, the prasugrel active metabolite may not be able to bind to the receptor until ticagrelor has dissociated. Ticagrelor might also induce a change in receptor conformation that temporarily precludes the clopidogrel or prasugrel active metabolite from binding. Given the SWAP-2 study results, it seems reasonable to start prasugrel at least 24 h after the last ticagrelor intake using a prasugrel loading dose. However, more data are needed to determine the optimal strategy and timing of switching to clopidogrel or prasugrel in the uncommon case in which ticagrelor should be discontinued due to a relevant, permanent and intolerable side effect.

rel at least 24 h after the last ticagrelor intake using a prasugrel loading dose. However, more data are needed to determine the optimal strategy and timing of switching to clopidogrel or prasugrel in the uncommon case in which ticagrelor should be discontinued due to a relevant, permanent and intolerable side effect. Conclusions Some facts should be considered in our clinical practice: In a patient with new onset of dyspnoea, the history and physical examination remain the mainstays of diagnostic evaluation. Assessing cause of dyspnoea may be difficult in ACS patients treated with ticagrelor. Dyspnoea is a very common ticagrelor side effect (>10%), but a genuine need for drug discontinuation is rare. ACS patients reporting dyspnoea have a high-risk profile and should be managed with the most effective treatment strategies. Balancing side effects and therapeutic advantages of each single drug is needed. Only in the case of persistent and intolerable ticagrelor-related dyspnoea should drug discontinuation be considered. Conflict of interest: Guido Parodi reported receiving consulting or lecture fees from AstraZeneca, Bayer, Daiichi Sankyo/Eli Lilly and The Medicines Company. Robert F. Storey reported receiving consulting fees or honoraria from AstraZeneca, Accumetrics, Daiichi Sankyo, Eli Lilly, Medscape, Merck, Novartis, PlaqueTec, Roche, Regeneron, Sanofi Aventis and The Medicines Company, and institutional grants or support from AstraZeneca, Accumetrics, Daiichi Sankyo, Eli Lilly and Merck.

rt F. Storey reported receiving consulting fees or honoraria from AstraZeneca, Accumetrics, Daiichi Sankyo, Eli Lilly, Medscape, Merck, Novartis, PlaqueTec, Roche, Regeneron, Sanofi Aventis and The Medicines Company, and institutional grants or support from AstraZeneca, Accumetrics, Daiichi Sankyo, Eli Lilly and Merck. Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Introduction To improve outcomes of sudden out-of-hospital cardiac arrest (OHCA), a novel citizen alert system was implemented in several regions in the Netherlands. Besides activating two ambulances, the dispatch centre also notifies citizen volunteers by text message (TM). Within their zip code vicinity, those volunteers are requested to go to the presumed arrest and either start basic life support (BLS) or first get an automated external defibrillator (AED). Recently, we performed a study in the Dutch province Limburg to assess the value of this system.1 If the system was activated but no volunteer responded (Scenario 1) then only standard care was given, and therefore this scenario was used as the reference group. It was found that survival to hospital discharge substantially increased from 16.0% to 27.1%, when at least one volunteer responded (Scenario 2) to the notification.1

tem was activated but no volunteer responded (Scenario 1) then only standard care was given, and therefore this scenario was used as the reference group. It was found that survival to hospital discharge substantially increased from 16.0% to 27.1%, when at least one volunteer responded (Scenario 2) to the notification.1 In the study at hand we aimed to explore the contribution of the alert system to survival specifically in situations with prolonged delay of start of resuscitation. The rationale behind the system is that responders can contribute to survival because they help to reduce the period between onset of the arrest and start of cardiopulmonary resuscitation (CPR) sufficiently soon after the collapse. Therefore, it was hypothesised that the system is most effective in situations where there may be a delay in response time, such as in the home or at night, and longer ambulance arrival times. Furthermore, a reduction of response time was expected to be especially effective in witnessed victims, because in unwitnessed victims help and support may often come too late anyway. Methods Setting The details of the study design and system have been published previously.1 From April 2012–April 2014, a prospective registry included all OHCAs in the Dutch province of Limburg. The study region Limburg has an area of approximately 2153 km2 (831 mi2) and consists of 1.12m inhabitants. Approval for the study was obtained from the medical ethics committee of the Maastricht University Medical Centre (project number 114029).

prospective registry included all OHCAs in the Dutch province of Limburg. The study region Limburg has an area of approximately 2153 km2 (831 mi2) and consists of 1.12m inhabitants. Approval for the study was obtained from the medical ethics committee of the Maastricht University Medical Centre (project number 114029). Resuscitation volunteer network in the study region Throughout the Netherlands, two ambulances are dispatched in the case of an (suspected) OHCA, each ambulance including one paramedic and a driver with CPR skills. A network of BLS/AED certified volunteers was implemented throughout Limburg and other regions in the Netherlands in order to reduce the delay in response time to start CPR. Furthermore, registered network AEDs were placed specifically in residential areas. Using the zip code derived location of the arrest location and volunteers, the dispatch centre notifies volunteers, close to the OHCA, simultaneously with two ambulances. In a 1:2 fashion, zip code identified volunteers within a radius of 1 km (0.62 mi) of the OHCA are notified to either start BLS or to get an AED first by the nearest network. During the study period, the alert system was active in 17 of the 24 Dutch dispatch centres and included 61,000 registered volunteers. The system was implemented in both dispatch centres in Limburg with more than 9000 volunteers (8.3/1000). The response rate of volunteers is not predictable and depends on the number of volunteers in the specific zip code area and their actual availability. A maximum of 30 volunteers are notified to make sure a sufficient but not excessive number of volunteers responds to the notification.

than 9000 volunteers (8.3/1000). The response rate of volunteers is not predictable and depends on the number of volunteers in the specific zip code area and their actual availability. A maximum of 30 volunteers are notified to make sure a sufficient but not excessive number of volunteers responds to the notification. Data collection Data were used from a registry of all consecutive OHCAs which occurred during a two-year period (April 2012–April 2014) in the Dutch province of Limburg. Data were collected according to the Utstein template.2–4 On a daily basis, all emergency calls were screened for suspected OHCAs. The data consisted of notification time, ambulance departure time and arrival time at the location, departure time to and arrival time at the hospital, patient’s condition and treatment. Information was also obtained from the paramedic notes about the resuscitation scenario (e.g. whether the OHCA was witnessed or not, who started CPR and the sequence of laymen and professionals that attended the OHCA). The alert system organisation (Hartslagnu) provided information about the activation of the system, such as the time the TM was sent, the number of notified volunteers and AEDs, and type of notification (start BLS or first get an AED). All notified volunteers received a questionnaire to obtain information about their attendance and, if applicable, about details of the scenario. Information included the presence of a witness and the start of CPR by the witness or by a bystander. Importantly, a witness was defined as the one who saw, heard or monitored the OHCA. A bystander was defined as the one who did not witness the event but was at the scene as well (e.g. a neighbour called by the witness). From the six hospitals in the province of Limburg information was gathered about post-resuscitation treatment, clinical outcome and discharge date and, if available, the medical history before OHCA.

s defined as the one who did not witness the event but was at the scene as well (e.g. a neighbour called by the witness). From the six hospitals in the province of Limburg information was gathered about post-resuscitation treatment, clinical outcome and discharge date and, if available, the medical history before OHCA. In this study, survival was compared between two resuscitation scenarios. In Scenario 1, the system was activated but no volunteer attended at the scene. In this situation, survival depended on standard care available from the two ambulances directed to the OHCA. In Scenario 2, the system was activated and volunteers indeed responded. The primary outcome measure was the proportion of patients surviving until discharge from hospital.

t no volunteer attended at the scene. In this situation, survival depended on standard care available from the two ambulances directed to the OHCA. In Scenario 2, the system was activated and volunteers indeed responded. The primary outcome measure was the proportion of patients surviving until discharge from hospital. Statistical analysis OHCA cases were categorised into subgroups according to whether the OHCA was witnessed, the location of the arrest, the time until arrival of the first ambulance and the time of day. Proportions of patients surviving until hospital discharge and relative risk estimates of survival with 95% confidence intervals (CIs), using Scenario 1 as the reference category, were calculated within subgroups (strata) which are referred to as stratum-specific odds ratios (ORs). Multivariable logistic regression analyses including scenario, the covariate coding for resuscitation setting and an interaction term for both variables were used to test for effect modification. Exponentiation of the regression coefficient corresponding with the interaction term gives the interaction OR. The interaction OR indicates whether the gain in survival due to the volunteer system differs between resuscitation settings (witnessed or not, location, arrival time of ambulance and time of day). An interaction OR=1 indicates equal survival benefit across strata. An interaction OR=2 indicates doubling of survival benefit compared to the reference category and for example an interaction OR=0.50 indicates halving of survival benefit compared to the reference category. Values of p⩽0.05 were considered statistically significant. For the analyses the software package of SPSS (SPSS for Windows, version 22.0, SPSS Inc., Chicago, Illinois, USA) was used.

the reference category and for example an interaction OR=0.50 indicates halving of survival benefit compared to the reference category. Values of p⩽0.05 were considered statistically significant. For the analyses the software package of SPSS (SPSS for Windows, version 22.0, SPSS Inc., Chicago, Illinois, USA) was used. Results The study population has been described previously.1 During the 24-month study period a total of 833 victims had (presumed) cardiac arrest. The system was activated in 422 (50.7%) cases and not activated in 411 (49.3%) cases. If the system was not activated, this was mostly because an ambulance was nearby or present at the scene, or because the OHCA occurred in a (closed) public place with an on-site AED (such as shopping malls). For this study, only data from system-activated cases were used where one or more volunteers responded in 291 cases (Scenario 2) and no volunteers responded in 131 cases (Scenario 1) (see Figure 1). Scenario 1 was used as the reference group. Figure 1. Flowchart of patient inclusion. Scenario 1, system activated 0 responders; Scenario 2, system activated ⩾1 responder. DNR: do not resuscitate policy; OHCA: out-of-hospital cardiac arrest.

Results The study population has been described previously.1 During the 24-month study period a total of 833 victims had (presumed) cardiac arrest. The system was activated in 422 (50.7%) cases and not activated in 411 (49.3%) cases. If the system was not activated, this was mostly because an ambulance was nearby or present at the scene, or because the OHCA occurred in a (closed) public place with an on-site AED (such as shopping malls). For this study, only data from system-activated cases were used where one or more volunteers responded in 291 cases (Scenario 2) and no volunteers responded in 131 cases (Scenario 1) (see Figure 1). Scenario 1 was used as the reference group. Figure 1. Flowchart of patient inclusion. Scenario 1, system activated 0 responders; Scenario 2, system activated ⩾1 responder. DNR: do not resuscitate policy; OHCA: out-of-hospital cardiac arrest. Distribution of resuscitation settings Mean age was 68.1 years (standard deviation (SD)±13.6) and 71.6% of OHCA victims were male. OHCA was witnessed in 75.1% of cases, and took place in the home situation in 82.5% of the cases. About 53.1% of the OHCAs occurred during the day vs 46.9% at evening or night (Table 1). In about 75% of cases the ambulance arrived after six minutes. The mean number of responding volunteers was 2.8 at daytime vs 2.9 at evening/night. Table 1. Percentages of survivors per scenario and total numbers (%) within strata according to witness status, location, time until arrival of first ambulance and time of day.

Distribution of resuscitation settings Mean age was 68.1 years (standard deviation (SD)±13.6) and 71.6% of OHCA victims were male. OHCA was witnessed in 75.1% of cases, and took place in the home situation in 82.5% of the cases. About 53.1% of the OHCAs occurred during the day vs 46.9% at evening or night (Table 1). In about 75% of cases the ambulance arrived after six minutes. The mean number of responding volunteers was 2.8 at daytime vs 2.9 at evening/night. Table 1. Percentages of survivors per scenario and total numbers (%) within strata according to witness status, location, time until arrival of first ambulance and time of day. Scenario 1 Scenario 2 Number (%) Witnessed, no. (%), n=422 No 2/32 (6.3) 3/73 (4.1) 105 (24.9) Yes 19/99 (19.2) 76/218 (34.9) 317 (75.1) Location of the arrest, no. (%), n=422 Outside the home 7/26 (26.9) 16/48 (33.3) 74 (17.5) Inside the home 14/105 (13.3) 63/243 (25.9) 348 (82.5) Time until arrival of first ambulance, no. (%), n=412 ⩽6 min 9/36 (25.0) 30/76 (39.5) 112 (27.2) 7–10 min 7/60 (11.7) 33/128 (25.8) 188 (45.6) ⩾11 min 4/31 (12.9) 13/81 (16.0) 112 (27.2) Time of day, no. (%), n=422 Day 13/62 (21.0) 42/162 (25.9) 224 (53.1) Evening/night 8/69 (11.6) 37/129 (28.7) 198 (46.9) Scenario 1 indicates that the system was activated but no volunteers responded. Scenario 2 indicates that the system was activated and at least 1 volunteer responded. In case OHCA was witnessed, the majority of the OHCAs (92.7%) occurred in at least one of the following settings: (a) in the home or (b) the arrival time of the first ambulance was between 6–11 min or (c) during the evening/night.

d. Scenario 2 indicates that the system was activated and at least 1 volunteer responded. In case OHCA was witnessed, the majority of the OHCAs (92.7%) occurred in at least one of the following settings: (a) in the home or (b) the arrival time of the first ambulance was between 6–11 min or (c) during the evening/night. Contribution of the responder to survival in different situations Figures 2(a)–(d), and Table 1 display the percentages of survival until discharge within strata of victims according to whether the OHCA was witnessed (yes or no), the location (inside vs outside the home), arrival time of the first ambulance (6, 7–10 or ⩾11 min) and time of day (08:00–18:00 vs 18:00–08:00). Table 2 shows stratum-specific and interaction odds ratios. The data show that the system leads to survival benefit within all strata except for the subgroup of non-witnessed arrests. Figure 2. Percentages of survivors in the cardiopulmonary resuscitation (CPR) groups among the subgroups. Scenario 1 indicates the system activated no responders; Scenario 2 indicates the system activated with ⩾1 responder. Table 2. Relative risk estimates and interaction odds ratios (ORs) of survival at hospital discharge in Scenario 2 according to witnessed arrest (yes or no), location, time until arrival of first ambulance and period of the day.

Figure 2. Percentages of survivors in the cardiopulmonary resuscitation (CPR) groups among the subgroups. Scenario 1 indicates the system activated no responders; Scenario 2 indicates the system activated with ⩾1 responder. Table 2. Relative risk estimates and interaction odds ratios (ORs) of survival at hospital discharge in Scenario 2 according to witnessed arrest (yes or no), location, time until arrival of first ambulance and period of the day. Setting Stratum specific OR (95% CI) p-Value Interaction OR (95% CI) p-Value Witnessed No 0.64 (0.10–4.05) 0.638 1.00 (reference) – Yes 2.25 (1.27–4.00) 0.005 3.51 (0.51–24.07) 0.202 Location Outside the home 1.36 (0.47–3.89) 0.570 1.00 (reference) – Home 2.28 (1.21–4.28) 0.011 1.68 (0.49–5.73) 0.410 Arrival times ⩽6 min 1.96 (0.81–4.73) 0.137 1.00 (reference) – 7–10 min 2.63 (1.09–6.35) 0.032 1.34 (0.39–4.69) 0.642 ⩾11 min 1.29 (0.39–4.31) 0.679 0.66 (0.15–2.94) 0.585 Period of the day Day 1.32 (0.65–2.67) 0.441 1.00 (reference) – Evening/night 3.07 (1.34–7.03) 0.008 2.33 (0.78–6.91) 0.129 CI: confidence interval.

Arrival times ⩽6 min 1.96 (0.81–4.73) 0.137 1.00 (reference) – 7–10 min 2.63 (1.09–6.35) 0.032 1.34 (0.39–4.69) 0.642 ⩾11 min 1.29 (0.39–4.31) 0.679 0.66 (0.15–2.94) 0.585 Period of the day Day 1.32 (0.65–2.67) 0.441 1.00 (reference) – Evening/night 3.07 (1.34–7.03) 0.008 2.33 (0.78–6.91) 0.129 CI: confidence interval. Witnessed and non-witnessed arrests In both scenarios, witnessed arrests were associated with a better survival probability compared to non-witnessed OHCA (Table 1). In the presence of volunteers the survival rate of witnessed OHCA increased from 19.2% (Scenario 1) to 34.9% (Scenario 2) corresponding with an OR=2.25 (95% CI: 1.27–4.00; p=0.005). During a non-witnessed arrest the attendance of volunteers was not associated with gain in survival (Table 1) corresponding with an OR=0.64 (95% CI: 0.10–4.05; p=0.638). The OR for interaction is 3.51 (95% CI: 0.51–24.07) meaning that the survival benefit due to the volunteer system is 3.5 times higher for witnessed arrests than for non-witnessed arrests. The p-value for interaction is 0.202.

ed with gain in survival (Table 1) corresponding with an OR=0.64 (95% CI: 0.10–4.05; p=0.638). The OR for interaction is 3.51 (95% CI: 0.51–24.07) meaning that the survival benefit due to the volunteer system is 3.5 times higher for witnessed arrests than for non-witnessed arrests. The p-value for interaction is 0.202. Location of the arrest As expected, the system was mainly activated in cases occurring in the home (348/422) but activation also occurred in 74 cases outside the home. For both Scenario 1 and Scenario 2 survival was higher outside the home than in the home (Table 1, Figure 2(b)). However, within the home, survival in Scenario 2 almost doubles compared to Scenario 1 (25.9% vs 13.3%) whereas outside the home survival in Scenario 2 is not much increased (33.3% vs 26.9%). As depicted in Table 2, stratum-specific relative risk estimates (favouring Scenario 2) were 2.28 (95% CI: 1.21–4.28; p=0.011) and 1.36 (95% CI: 0.47–3.89; p=0.570), respectively. The OR for interaction is 1.68 (95% CI: 0.49–5.73) meaning that survival benefit due to the volunteer system is more than 1.5 times higher for arrests occurring in the home than for arrests outside the home. The p-value for interaction is 0.410.

: 1.21–4.28; p=0.011) and 1.36 (95% CI: 0.47–3.89; p=0.570), respectively. The OR for interaction is 1.68 (95% CI: 0.49–5.73) meaning that survival benefit due to the volunteer system is more than 1.5 times higher for arrests occurring in the home than for arrests outside the home. The p-value for interaction is 0.410. Ambulance arrival times With respect to time of arrival of the first ambulance, a trend was found towards lower survival probability with increasing delay. However, survival in Scenario 2 was higher compared to Scenario 1 within each stratum of ambulance arrival time since notification (Table 1). Importantly, in Scenario 2 the decrease in survival with increasing delay was less substantial than in Scenario 1 (Figure 2(c)). Strong effects of the system on survival were observed for cases where the first ambulance arrived with slight delay. The relative risk estimate associated with a 7–10 min interval between notification and arrival of the ambulance was 2.63 (95% CI: 1.09–6.35; p=0.032). The volunteer system is especially effective when the ambulance arrives with a slight delay (7–10 min) as indicated by the OR for interaction of 1.34. When the delay increases to 11 min or more there is still survival advantage in Scenario 2 compared to Scenario 1, 16% vs 12.9% respectively, but the stratum-specific OR of 1.29 (Table 2) is no longer statistically significant (p=0.679).

ves with a slight delay (7–10 min) as indicated by the OR for interaction of 1.34. When the delay increases to 11 min or more there is still survival advantage in Scenario 2 compared to Scenario 1, 16% vs 12.9% respectively, but the stratum-specific OR of 1.29 (Table 2) is no longer statistically significant (p=0.679). Time of the day When no volunteers attended (Scenario 1) survival was higher during daytime (21.0%) than at evening/night (11.6%). In the presence of volunteers (Scenario 2) survival percentages were higher than in Scenario 1 and at evening/night survival was even slightly higher than during the day (28.7% and 25.9% respectively), as depicted in Table 1. The decrease in survival of arrests during evening/night in Scenario 1 combined with the slight increase in survival in Scenario 2 is consistent with a stratum specific OR=3.07 (95% CI: 1.34–7.03; p=0.008), favouring Scenario 2. During daytime the contribution to survival was lower with OR=1.32 (95% CI: 0.65–2.67; p=0.441). The interaction OR was 2.33 with p=0.129 which indicated that the benefit of the system during the evening/night is 2.33 times higher compared to the benefit during daytime.

5% CI: 1.34–7.03; p=0.008), favouring Scenario 2. During daytime the contribution to survival was lower with OR=1.32 (95% CI: 0.65–2.67; p=0.441). The interaction OR was 2.33 with p=0.129 which indicated that the benefit of the system during the evening/night is 2.33 times higher compared to the benefit during daytime. Adjustment for potential confounders During the evening or night the distribution of ambulance arrival times differs from that during the day with a higher frequency of longer delays. Distribution of the other effect modifiers (presence of witness and location) may also be different. For this reason, multivariable logistic regression analyses were performed including scenario, all effect modifiers and their terms for interaction with scenario. Age and sex as potential confounders were also added to the model. These analyses gave similar results (not shown). Discussion Recently we reported that survival to hospital discharge in resuscitated out-of-hospital (presumed) cardiac arrest substantially increases by the involvement of citizen responders notified by the ambulance dispatch centre by a text message. In the current study, the hypotheses were tested that the system was especially effective in (a) witnessed OHCA, (b) in the home situation, (c) at longer ambulance delay times and (d) during the evening/night-time.

ses by the involvement of citizen responders notified by the ambulance dispatch centre by a text message. In the current study, the hypotheses were tested that the system was especially effective in (a) witnessed OHCA, (b) in the home situation, (c) at longer ambulance delay times and (d) during the evening/night-time. Main findings It was found that the contribution of the system was most pronounced if the OHCA was witnessed (OR=2.25), occurred in the home situation (OR=2.28), when the ambulance arrived with a slight delay i.e. 7–10 min (OR=2.63) and when the OHCA occurred at evening/night (OR=3.07). After adjustment for other effect modifiers, age and gender, results were similar. Witnessed and non-witnessed arrests One of the most pronounced predictors of survival is OHCA being witnessed.5 Also, in this study witnessed arrests had a higher survival probability in both scenarios. The attendance of volunteers in case of a witnessed arrest had an additional positive effect on survival. Volunteers apparently effectively shorten the delay time to start CPR before emergency medical services (EMS) arrival. Unwitnessed arrest carries a poor prognosis anyway and volunteers cannot contribute much to improve this.

ance of volunteers in case of a witnessed arrest had an additional positive effect on survival. Volunteers apparently effectively shorten the delay time to start CPR before emergency medical services (EMS) arrival. Unwitnessed arrest carries a poor prognosis anyway and volunteers cannot contribute much to improve this. Location of the arrest Higher survival in OHCA outside the home is related to the higher probability that the collapse is witnessed and that witnesses and/or bystanders will start CPR before the arrival of an ambulance. In this study we found that OHCAs outside the home were witnessed in 81.1% of cases and that CPR was started by a witness or bystander in 84.7%. In OHCAs inside the home, these percentages were 73.9% and 50.0%, respectively. Due to lower survival probability of OHCA inside the home there is considerable potential for an alert system to contribute to survival. Rapid arrival of volunteers can compensate for the longer delay times until the start of resuscitation. The higher survival gain in the home situation is reflected by the results in this study, where the OR of 2.28 in the home situation is higher than the OR of 1.36 for OHCAs occurring outside the home. These results are promising because the large majority of OHCAs occur in the home, supporting the value of this citizen volunteer system.

l gain in the home situation is reflected by the results in this study, where the OR of 2.28 in the home situation is higher than the OR of 1.36 for OHCAs occurring outside the home. These results are promising because the large majority of OHCAs occur in the home, supporting the value of this citizen volunteer system. Ambulance arrival times Survival is known to be negatively related to longer arrival times of the ambulance.6,7 Optimal gain in survival by the system can therefore be achieved specifically in settings with more delay until the arrival of healthcare professionals; at short first ambulance arrival times, the ambulance could even arrive before the responders. Importantly, as shown in Figure 2(c), the contribution of the system was typically seen at ambulance arrival times between 6–11 min, which occurred in 44.5% of the cases. Apparently this is the window of opportunity where volunteers contribute mostly to survival. At later arrival times (11 min or later) this benefit and survival decreased, likely due to the overly long time between onset of the arrest and onset of professional care.8 Although volunteers can provide good quality CPR, early stabilisation of the patient by the EMS is crucial for survival of an OHCA.

y to survival. At later arrival times (11 min or later) this benefit and survival decreased, likely due to the overly long time between onset of the arrest and onset of professional care.8 Although volunteers can provide good quality CPR, early stabilisation of the patient by the EMS is crucial for survival of an OHCA. Time of the day During daytime, patients in Scenario 2 had a higher probability of survival compared to Scenario 1 (25.9% vs 21.0%). This difference was even greater in the evening/night and amounts to 17.1% (28.7% vs 11.6%, Table 1). These results suggest that gain in survival due to the system is more evident during the evening/night than during the day. There was no difference between the mean number of responders during daytime and evening/night and therefore the gain in survival during night cannot be attributed to better availability and/or preparedness of volunteers during night-time. Instead a lower activation state of the dispatch/ambulance system and/or less availability of ambulances in the evening/night have to be considered, given the decrease in survival rate in Scenario 1, comparing OHCA at evening/night with daytime. This possibility is supported by our data showing that during evening/night the ambulance arrival time >11 min was 34.5% in contrast to 20.6% during the day (p<0.001). During the evening/night the system could therefore more effectively compensate for the longer delay time of the ambulance, and contributed to a higher survival rate.

ty is supported by our data showing that during evening/night the ambulance arrival time >11 min was 34.5% in contrast to 20.6% during the day (p<0.001). During the evening/night the system could therefore more effectively compensate for the longer delay time of the ambulance, and contributed to a higher survival rate. Comparison with other community responder systems In different countries several strategies exist to involve citizen volunteers to improve survival of out-of-hospital circulatory arrest. Comparable to the Dutch alert system is the Mobile Life Service (MLS) in Stockholm, Sweden.9 In Denmark a volunteer-based network of AEDs (accessible to lay persons) is active where the dispatcher guides bystanders to a close by AED.10 Also mobile phone applications are used such as the GoodSAM app in the UK, enabling a call to the dispatch centre and alert to nearby registered first aiders. All these systems have in common that they all rely on trained citizen rescuers who are already nearby the OHCA. These trained citizen rescuers can potentially decrease the time between onset of the arrest and time of starting CPR. Every citizen can be a potential rescuer. However, because of the voluntary nature of these systems, it is hard to predict whether volunteers actually will respond to a notification.

y nearby the OHCA. These trained citizen rescuers can potentially decrease the time between onset of the arrest and time of starting CPR. Every citizen can be a potential rescuer. However, because of the voluntary nature of these systems, it is hard to predict whether volunteers actually will respond to a notification. Legal issues with regard to the implementation and use of citizen rescuers in case of emergencies differ between countries and should always be explored. To our knowledge up till now no data on their contribution to survival have been published. A previous study in another region in the Netherlands reported that this alert system contributes to earlier defibrillation in sudden cardiac arrest (SCA)11 but did not report on survival. Although no outcome data were reported, the benefit of the alert system was suggested by a reduced time to defibrillation by citizen responders with AEDs, compared to time to defibrillation by the EMS. Limitations A limitation of the study is the small sample size within specific subgroups that likely resulted in limited power to detect significant interaction. Nevertheless, lack of significance does not indicate absence of interaction and the higher contribution to survival of the alert system in the case of witnessed arrests, in the home, in situations with some delay in arrival of the first ambulance and during the evening/night, is consistent with the a priori hypotheses.

Nevertheless, lack of significance does not indicate absence of interaction and the higher contribution to survival of the alert system in the case of witnessed arrests, in the home, in situations with some delay in arrival of the first ambulance and during the evening/night, is consistent with the a priori hypotheses. Conclusion The contribution of the system to survival of OHCA is most pronounced when OHCAs are witnessed, occur in the home, the ambulance arrives with a delay between 6–11 minutes and the OHCA occurs in the evening or night. Taking only the witnessed arrests into account, the majority of the OHCAs (92.7%) occurred in at least one of the three other conditions (in the home, a delay between 6–11 min or in the evening), indicating that many OHCA victims can benefit from the system.

een 6–11 minutes and the OHCA occurs in the evening or night. Taking only the witnessed arrests into account, the majority of the OHCAs (92.7%) occurred in at least one of the three other conditions (in the home, a delay between 6–11 min or in the evening), indicating that many OHCA victims can benefit from the system. The authors are greatly indebted to the Province Limburg and the Mercurius Foundation for the financial support of this study; FW Lindemans and HJJ Wellens for their support and suggestions; the staff of the participating hospitals, other institutions and medical students for helping in collecting the data: Zuyderland Hospital Sittard/Heerlen; D van Kraaij, H Kragten and the R&D Cardiology; Laurentius Hospital Roermond, C Werter and M Janssen; Sint Jans Gasthuis Weert, H Klomps and Viecuri Venlo; the emergency medical services of GGD South-Limburg (N Otten) and AmbulanceZorg Limburg-North (L Triepels), Hartslagnu and Ocean (Theo Schrijnemaekers); police department district Limburg South, AED solutions (R Henderikx), BHV-competent (J Hoofs), Vivon (M van Gorp). Last, but not least, all volunteers helping to increase survival of their fellow citizens with OHCA are gratefully acknowledged. Conflict of interest: The authors declare that there is no conflict of interest. Funding: This work was supported by the Province Limburg (SAS-202-01794) and the Mercurius Foundation (30957210N).

Introduction Electrical storm is a state of electrical instability and is characterised by several episodes of ventricular tachycardia (VT) or ventricular fibrillation (VF). The implantable cardioverter defibrillator (ICD) can effectively terminate ventricular arrhythmia (VA); however, it will not eliminate or modify the trigger or substrate of electrical storm.1 Electrical storm patients usually present as a severe medical emergency characterised by multiple ICD shocks and haemodynamic instability. Because of the infrequent nature and unpredictability of electrical storm associated with a potential lethal outcome many physicians feel uncertain in the acute setting. Mortality in the early and subacute phase is high.2,3 Several factors are associated with a negative outcome in electrical storm patients: severely impaired left ventricular ejection fraction (LVEF),4 pre-existing advanced New York Heart Association class, cardiogenic shock5 and older age. Electrical storm can be a distressing experience for patients and their families, leading to significant psychological consequences. Effective management of electrical storm is crucial, and a collaborative hospital network with a dedicated electrical storm team has been suggested as beneficial.6,7 Treatment of electrical storm can be very complex and consists of the administration of anti-arrhythmic drugs (AADs), suppression of sympathetic tone, device re-programming and sometimes urgent catheter ablation (Table 1).

tive hospital network with a dedicated electrical storm team has been suggested as beneficial.6,7 Treatment of electrical storm can be very complex and consists of the administration of anti-arrhythmic drugs (AADs), suppression of sympathetic tone, device re-programming and sometimes urgent catheter ablation (Table 1). Definition of electrical storm: diversity in the literature The clinical syndrome of electrical storm has been defined empirically. In the past a variety of definitions were used. In those early definitions the VT episodes ranged between two and 20 within 24 hours.5,8 At present, in the era of ICDs the most commonly accepted definition is three or more separate arrhythmia episodes leading to ICD therapy occurring over a single 24-hour time period.9 The episodes of VT must be separate, meaning that the persistence of VT following unsuccessful ICD therapy is not considered as a second episode.10 Incessant VT is a condition in which a sustained VT resumes within 5 minutes after successful ICD therapy and continues for over 12 hours. No study to date has determined a certain threshold burden of ICD therapy that begins to confer an adverse outcome.

unsuccessful ICD therapy is not considered as a second episode.10 Incessant VT is a condition in which a sustained VT resumes within 5 minutes after successful ICD therapy and continues for over 12 hours. No study to date has determined a certain threshold burden of ICD therapy that begins to confer an adverse outcome. Mechanisms underlying electrical storm Crucial for the occurrence of electrical storm is an interplay between the autonomic nervous system, cellular milieu and a predisposing electrophysiological substrate. Both the trigger and the substrate may change over time influenced by the progression of scarring, left ventricular remodelling and the progression of heart failure. The critical role of an increased activation of the sympathetic nervous system in initiating and maintaining electrical storm is demonstrated in electrical storm patients who have exacerbation of heart failure.11

fluenced by the progression of scarring, left ventricular remodelling and the progression of heart failure. The critical role of an increased activation of the sympathetic nervous system in initiating and maintaining electrical storm is demonstrated in electrical storm patients who have exacerbation of heart failure.11 Electrical storm: disease or symptom Although electrical storm directly affects the patients’ prognosis, by preventing the next episode of electrical storm the mortality does not necessarily decrease.12 Electrical storm often represents part of the natural history of advanced cardiac disease and may predict a serious deterioration in the underlying processes. It can even be debated if electrical storm is a marker for mortality in the near future and accordingly functions as a major bystander. This raises the question of whether all electrical storm patients would be potential candidates for catheter ablation. It is also a valid question as to whether a severe disbalance in the cellular milieu could outweigh a modification of the substrate? At the other end of the spectrum is those presenting with a first episode of electrical storm, who may benefit much more from a catheter ablation intervention and have a possible survival benefit.13 Therefore, every patient that presents with even a single ICD shock should be considered as a possible electrical storm, whereas it may be preceded by multiple episodes of VT successfully treated by antitachycardia pacing (ATP).

fit much more from a catheter ablation intervention and have a possible survival benefit.13 Therefore, every patient that presents with even a single ICD shock should be considered as a possible electrical storm, whereas it may be preceded by multiple episodes of VT successfully treated by antitachycardia pacing (ATP). Treatment of electrical storm: corresponding to the mechanism and trigger Searching for and correction of reversible factors In the majority of cases, no clear cause for electrical storm can be identified. Triggers such as electrolyte imbalance, acute ischaemia, exacerbation of heart failure, adjustment of or non-compliance to anti-arrhythmic medication1 and recent introduction to biventricular pacing have been identified.14 They should be actively searched for and promptly corrected in each electrical storm patient. Flow limiting coronary artery disease and volume overload should be adequately treated. Decreased left ventricular wall stress can be achieved with non-invasive and invasive haemodynamic support including a left ventricular assist device (LVAD), venoarterial extracorporeal membrane oxygenation (ECMO) and continuous flow percutaneous ventricular assist devices. Fever is a more rare trigger of electrical storm, and is especially important in patients with Brugada syndrome, in whom unsuppressed fever may lead to medically resistant incessant polymorphic and possibly fatal VT.15 Device programming Shocks delivered for self-limiting haemodynamically tolerable arrhythmias ought to be avoided.

Treatment of electrical storm: corresponding to the mechanism and trigger Searching for and correction of reversible factors In the majority of cases, no clear cause for electrical storm can be identified. Triggers such as electrolyte imbalance, acute ischaemia, exacerbation of heart failure, adjustment of or non-compliance to anti-arrhythmic medication1 and recent introduction to biventricular pacing have been identified.14 They should be actively searched for and promptly corrected in each electrical storm patient. Flow limiting coronary artery disease and volume overload should be adequately treated. Decreased left ventricular wall stress can be achieved with non-invasive and invasive haemodynamic support including a left ventricular assist device (LVAD), venoarterial extracorporeal membrane oxygenation (ECMO) and continuous flow percutaneous ventricular assist devices. Fever is a more rare trigger of electrical storm, and is especially important in patients with Brugada syndrome, in whom unsuppressed fever may lead to medically resistant incessant polymorphic and possibly fatal VT.15 Device programming Shocks delivered for self-limiting haemodynamically tolerable arrhythmias ought to be avoided. Detection time can be prolonged and ATP can be given as an initial therapy.16 Augmentation of ATP attempts, when feasible, is encouraged especially when shown previously to be successful.17 During an electrical storm an effort should be made to avoid further conscious shocks,18 and temporary disabling of shock therapy may be considered.

be prolonged and ATP can be given as an initial therapy.16 Augmentation of ATP attempts, when feasible, is encouraged especially when shown previously to be successful.17 During an electrical storm an effort should be made to avoid further conscious shocks,18 and temporary disabling of shock therapy may be considered. Anti-arrhythmic drugs Frequently, the first step in the treatment of electrical storm is the administration of beta-blockers. Beta-blockers play a fundamental role in the management of electrical storm by blocking the sympathetic system. Adding beta-blockers intravenously in electrical storm patients already on oral beta-blocker therapy may help to keep an electrical storm episode under control.19 Propranolol, a lipophilic unselective beta-blocker that penetrates the central nervous system, has been demonstrated to be effective in suppressing VAs as compared to metoprolol and amiodarone.20 In the presence of structural heart disease amiodarone is one of the most frequently used drugs for the treatment of electrical storm. Procainamide, a class 1C AAD, has demonstrated its superiority compared to amiodarone for the treatment of haemodynamically tolerated monomorphic VT in the PROCAMIO trial.21 However, it has been investigated only in patients without manifest heart failure and without severely depressed LVEF, in whom it is considered safe. The incidence of IV-amiodarone-refractory electrical storm is approximately 30%. IV-amiodarone-refractory VT storms are frequently induced by triggering premature ventricular contractions (PVCs) with a narrow QRS complex,22 and may be successfully suppressed with additional administration of mexiletine, a class 1B AAD.23 Reperfusion often leads to the development of automaticity or delayed afterdepolarisations originating from the Purkinje network,24 which in fact is sodium channel mediated.25 Lidocaine, a class 1B AAD is used in the setting of acute ischaemia.26

uppressed with additional administration of mexiletine, a class 1B AAD.23 Reperfusion often leads to the development of automaticity or delayed afterdepolarisations originating from the Purkinje network,24 which in fact is sodium channel mediated.25 Lidocaine, a class 1B AAD is used in the setting of acute ischaemia.26 There is no consensus on the optimal drug treatment for refractory malignant VA, and AADs may be given in a manner of trial and error. Drug combinations are sometimes necessary to alter electrical instability. AADs carry the risk of decreasing the cycle length of re-entry VAs and make VT more stable, which may precipitate to incessant VT. AADs should be given individually, taking into account not only the efficacy but also the increased risk of drug-related pro-arrhythmia and other side effects. Overdrive pacing and sedation Temporary (atrial) overdrive pacing may help to interrupt an incessant or re-occurring VA, especially in conditions such as Brugada and early repolarisation syndrome.26 Overdrive pacing helps by preventing PVCs from occurring and reduces early afterdepolarisation.27 As the sympathetic nervous system plays a major role in the initiation but also the maintenance of VAs,11 sedation and/or intubation may be needed in order to suppress the sympathetic tone. A complete sympathetic blockade can be performed by left cardiac sympathetic denervation.28

Overdrive pacing and sedation Temporary (atrial) overdrive pacing may help to interrupt an incessant or re-occurring VA, especially in conditions such as Brugada and early repolarisation syndrome.26 Overdrive pacing helps by preventing PVCs from occurring and reduces early afterdepolarisation.27 As the sympathetic nervous system plays a major role in the initiation but also the maintenance of VAs,11 sedation and/or intubation may be needed in order to suppress the sympathetic tone. A complete sympathetic blockade can be performed by left cardiac sympathetic denervation.28 Radiofrequency catheter ablation In the majority of electrical storm patients the episodes are characterised by a monomorphic VT based on re-entry. Therefore catheter ablation, targeting the substrate in which re-entry has formed, is an important treatment option for electrical storm. Table 1. Learning objective. Learning objective 1 How the mechanism and trigger of electrical storm can guide electrical storm treatment 2 To learn about the importance of the sympathetic nervous system in the initiation and maintenance of electrical storm 3 To tailor AAD treatment considering efficacy, drug-related pro-arrhythmia and other side effects 4 How to programme an ICD to avoid recurrent shocks 5 To learn about the indication and timing of catheter ablation AAD: anti-arrhythmic drug; ICD: implantable cardioverter defibrillator.

2 To learn about the importance of the sympathetic nervous system in the initiation and maintenance of electrical storm 3 To tailor AAD treatment considering efficacy, drug-related pro-arrhythmia and other side effects 4 How to programme an ICD to avoid recurrent shocks 5 To learn about the indication and timing of catheter ablation AAD: anti-arrhythmic drug; ICD: implantable cardioverter defibrillator. In a pooled meta-analysis29 of 471 electrical storm patients who underwent catheter ablation, catheter ablation had a high success rate with a low rate of recurrent electrical storm. Acute procedural success was 72% and procedural failure was 9%. During a follow-up of 15 months, 60% of patients were free of VA recurrences and 94% were free of electrical storm. Since then ablation of VT has evolved, and new approaches and technologies, such as the substrate approach,30 remote magnetic navigation,31 and a combined endo-epicardial substrate ablation,32 have improved the outcome of VT ablation (Table 2).33 Table 2. Outcome of ventricular tachycardia ablation in electrical storm patients.

In a pooled meta-analysis29 of 471 electrical storm patients who underwent catheter ablation, catheter ablation had a high success rate with a low rate of recurrent electrical storm. Acute procedural success was 72% and procedural failure was 9%. During a follow-up of 15 months, 60% of patients were free of VA recurrences and 94% were free of electrical storm. Since then ablation of VT has evolved, and new approaches and technologies, such as the substrate approach,30 remote magnetic navigation,31 and a combined endo-epicardial substrate ablation,32 have improved the outcome of VT ablation (Table 2).33 Table 2. Outcome of ventricular tachycardia ablation in electrical storm patients. Number of electrical storm patients Population Control group Follow-up duration, months Free of recurrence, % Survival, % Nayyar et al., 201330 471 Meta-analysis 68% ICM 37% incessant VA NA 15 60 83 Di Biase et al., 201231 92 ICM Limited substrate ablation vs. endo-epicardial homogenisation 25 Limited substrate: 53 endo-epi: 81 P=0.006 Limited substrate: 98 endo-epi: 98 P= NS Izquierdo et al., 201212 23 ICM Catheter ablation: 83 MT: 66 MT: 23 28 Catheter ablation: 68 MT 71 P=NS Catheter ablation: 70 MT: 48 P=NS Özcan et al., 201639 44 ICM, drug refractory electrical storm NA 28 55 82 Muser et al., 201740 267 ICM and non-ICM 22% endo-epicardial ablation NA 45 67 71 Jin et al., 201732 54 ICM, ablation with RMN NA 17 50 80 Morawski et al., 201713 28 81% ICM MT: 42 28 MT: 26 Catheter ablation: 43 P=NS Catheter ablation: 86 MT: 62 P=0.03 Kumar et al., 201741 287 64% ICM ICM vs. non-ICM 12 ICM: 51 Non-ICM: 36 P=0.007 ICM: 75 Non-ICM: 72 P=NS ICM: ischaemic cardiomyopathy; MT: medical therapy; NA: not applicable; RMN: remote magnetic navigation.

wski et al., 201713 28 81% ICM MT: 42 28 MT: 26 Catheter ablation: 43 P=NS Catheter ablation: 86 MT: 62 P=0.03 Kumar et al., 201741 287 64% ICM ICM vs. non-ICM 12 ICM: 51 Non-ICM: 36 P=0.007 ICM: 75 Non-ICM: 72 P=NS ICM: ischaemic cardiomyopathy; MT: medical therapy; NA: not applicable; RMN: remote magnetic navigation. There is also a role for catheter ablation in patients who suffer from recurrent VF episodes. In 29 patients with ischaemic heart disease, recurrent VF was triggered by monomorphic ventricular extrasystole that originated from the fibrous peri-infarction zone. In eight patients with drug refractory electrical storm, ablation of the ventricular extrasystole was successfully performed, and control of electrical storm was achieved.34

ischaemic heart disease, recurrent VF was triggered by monomorphic ventricular extrasystole that originated from the fibrous peri-infarction zone. In eight patients with drug refractory electrical storm, ablation of the ventricular extrasystole was successfully performed, and control of electrical storm was achieved.34 Compared to medical therapy catheter ablation reduces the number of subsequent VT episodes especially when VT ablation is performed within one month of electrical storm.35 VT ablation in patients with a LVEF of 25% or greater is shown most beneficial.12 Freedom from recurrent VT after catheter ablation has been associated with an improved survival.13,36 Morawski et al.13 showed that in a first time electrical storm population, VT ablation was significantly more effective than any other form of therapy in reducing death at any time, even though the recurrence rate was not lower in the catheter ablation group. Yet, it is also known that patients with electrical storm have an increased risk of non-cardiac death. In other studies a mortality benefit from VT ablation in electrical storm patients was not shown.12 This underlines the importance of the selection of patients as potential candidates for ablation.

ter ablation group. Yet, it is also known that patients with electrical storm have an increased risk of non-cardiac death. In other studies a mortality benefit from VT ablation in electrical storm patients was not shown.12 This underlines the importance of the selection of patients as potential candidates for ablation. The timing of catheter ablation, the approach and support should be tailored. Patients with incessant drug refractory VT who fail on haemodynamic support can benefit from a rescue VT ablation.37 Patients with advanced heart failure and unstable VTs are at highest risk of haemodynamic collapse during the ablation procedure; they can benefit from mechanical support during catheter ablation.38 Alternatively, the ablation can be confined to a substrate approach only. Consequent fluid overload related to irrigated catheter ablation may precipitate acute decompensation,39 and preventive measures such as LVAD or ECMO may still be indicated in patients with severely depressed left ventricular function. In a small proportion of patients there is such a limited reserve in cardiac output that limited ablation should be aimed for, targeting only the critical isthmus of the clinical VT. Conclusion Electrical storm is a critical condition and even after successful catheter ablation patients continue to bear an increased burden of morbidity and mortality. Early recognition and referral to a tertiary electrophysiology centre is mandatory. Electrical storm should be treated by a team that offers a structured and tailored approach.

m is a critical condition and even after successful catheter ablation patients continue to bear an increased burden of morbidity and mortality. Early recognition and referral to a tertiary electrophysiology centre is mandatory. Electrical storm should be treated by a team that offers a structured and tailored approach. Key points Early recognition of electrical storm and referral to a tertiary electrophysiology centre is mandatory. Electrical storm should be treated by an experienced team that offers a structured and tailored approach. An increased activation of the sympathetic nervous system is critical in the initiation and maintenance of electrical storm. Electrical storm often represents part of the natural history of advanced cardiac disease and may be a predictor of serious deterioration of the underlying disease. By treating electrical storm we attempt to eliminate the trigger and modify the substrate of the ventricular arrhythmia. Treatment of electrical storm is complex and consists of the administration of anti-arrhythmic drugs, suppression of sympathetic tone, device re-programming and catheter ablation. Anti-arrhythmic drugs should be given individually, taking into account not only the efficacy but also an increased risk of drug-related pro-arrhythmia and other side effects. Electrical storm is a critical condition and even after successful catheter ablation patients continue to bear an increased risk of morbidity and mortality. Conflict of interest: The authors declare that there is no conflict of interest.

Anti-arrhythmic drugs should be given individually, taking into account not only the efficacy but also an increased risk of drug-related pro-arrhythmia and other side effects. Electrical storm is a critical condition and even after successful catheter ablation patients continue to bear an increased risk of morbidity and mortality. Conflict of interest: The authors declare that there is no conflict of interest. Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Introduction Due to its low cost, its bedside applicability and its non-invasiveness, cardiac ultrasound has developed into a widely used diagnostic tool in many heart diseases. The diagnostic possibilities increased even further with the introduction of microbubbles. In 1968, Gramiak and Shah were the first to report the use of microbubbles.1 In general, the microbubble currently consists of a high-molecular-weight, gas-filled core to prevent it from dissolving in the pulmonary vasculature. This gas is encapsulated in a lipid or albumin shell, depending on the manufacturer.2 The composition of microbubbles makes them very good scatterers of ultrasound in contrast to erythrocytes, resulting in enhancement of the ultrasound image, thereby increasing diagnostic possibilities.3 Recently, attention has shifted to potential therapeutic applications of contrast ultrasound not only in the field of cardiology, but also in cerebrovascular4 and peripheral vascular disease.5 It was known that local application of ultrasound can cause bioeffects.6,7 The addition of microbubbles lowers the threshold for these effects to take place, thereby potentially enhancing these bioeffects. It has been demonstrated in vitro and in vivo that contrast ultrasound has a potential therapeutic effect in acute ST elevation myocardial infarction (STEMI).

sound can cause bioeffects.6,7 The addition of microbubbles lowers the threshold for these effects to take place, thereby potentially enhancing these bioeffects. It has been demonstrated in vitro and in vivo that contrast ultrasound has a potential therapeutic effect in acute ST elevation myocardial infarction (STEMI). Coronary thrombosis STEMI is most often caused by the acute formation of a thrombus blocking an epicardial coronary artery, and current treatment strategies are focused on the prompt revascularisation of the occluded coronary artery. According to American College of Cardiology/American Heart Association (ACC/AHA) and European Society of Cardiology (ESC) guidelines, the first choice of treatment is primary percutaneous coronary intervention (PCI).8,9 However, in rural areas, primary PCI is not available within 90 minutes of first patient contact. In those areas, thrombolysis is the next best therapy. Worldwide, thrombolysis remains the most used therapy in STEMI patients, despite its lower recanalisation rate and higher haemorrhagic complication rate compared to PCI. Hence, there is a continuing search for an easily applicable, non-invasive treatment strategy resulting in higher recanalisation rates with lower complication rates.

mbolysis remains the most used therapy in STEMI patients, despite its lower recanalisation rate and higher haemorrhagic complication rate compared to PCI. Hence, there is a continuing search for an easily applicable, non-invasive treatment strategy resulting in higher recanalisation rates with lower complication rates. Pathophysiology Low-frequency, high-intensity ultrasound Hong et al. were one of the first to prove the potential for ultrasound-induced cavitation to dissolve thrombi in vitro.10 They used low-frequency, high-intensity ultrasound with different-sized probes to treat blood clots, and observed that all clots were disrupted in less than 3 minutes. The size of the particles of the disrupted blood clot ranged from 2.5 to 80 µm. The addition of Streptokinase®, a thrombolytic agent, did not seem to have a measurable effect on the particle size distribution. Another study performed by Rosenschein and colleagues confirmed these results in vitro.11 They also conducted an in vivo study in dogs, which led to a significant decrease in thrombus obstruction, without an increase in wall injury of the treated vessel.12 After these successful in vitro and in vivo studies with high-intensity ultrasound, two small non-randomized studies presented good results regarding safety, feasibility and thrombus dissolution based on Thrombolysis in Myocardial Infarction (TIMI) flow.13,14 However, since it was an invasive approach that did not significantly improve recanalisation rates compared to PCI, further development of high-intensity ultrasound to dissolve thrombi came to a halt.

regarding safety, feasibility and thrombus dissolution based on Thrombolysis in Myocardial Infarction (TIMI) flow.13,14 However, since it was an invasive approach that did not significantly improve recanalisation rates compared to PCI, further development of high-intensity ultrasound to dissolve thrombi came to a halt. Low-intensity, high-frequency ultrasound A disadvantage of transcutaneous high-intensity, low-frequency ultrasound is the fact that it is not readily available in daily clinical practice. At lower intensities and long pulse durations, a number of studies have been conducted with these ultrasound settings to evaluate the effect on thrombi. It was demonstrated that low-intensity ultrasound alone was unable to enhance clot dissolution. However, on top of a thrombolytic agent (e.g. tissue plasminogen activator [t-PA]), it was observed that thrombolysis was accelerated compared to treatment of thrombus with t-PA alone.15

gs to evaluate the effect on thrombi. It was demonstrated that low-intensity ultrasound alone was unable to enhance clot dissolution. However, on top of a thrombolytic agent (e.g. tissue plasminogen activator [t-PA]), it was observed that thrombolysis was accelerated compared to treatment of thrombus with t-PA alone.15 Cavitation Thus, both high- and low-intensity ultrasound seem to be able to enhance clot dissolution.16 Nevertheless, both approaches have an essentially different working mechanism. Where high-intensity ultrasound is capable of clot dissolution on its own under the influence of inertial cavitation, this effect does not occur using low-intensity ultrasound. Instead, it was thought that low-intensity ultrasound causes stable cavitation, thus creating local microstreaming. While lower-intensity ultrasound on its own may not be capable of enhancing clot dissolution, it is effective at increasing local concentrations of exogenously administered t-PA, thereby enhancing thrombolysis. This was nicely shown in an in vitro study performed by Sakharov and colleagues.17 They demonstrated when ultrasound was applied without stirring that it accelerated lysis by about two-fold. However, in the presence of mild or strong stirring around the thrombus, ultrasound only slightly enhanced thrombolysis. The fact that ultrasound still enhanced thrombolysis in the presence of strong stirring was largely explained by the local temperature rise, which is another feature of the application of low-intensity ultrasound.18

presence of mild or strong stirring around the thrombus, ultrasound only slightly enhanced thrombolysis. The fact that ultrasound still enhanced thrombolysis in the presence of strong stirring was largely explained by the local temperature rise, which is another feature of the application of low-intensity ultrasound.18 Contrast ultrasound Tachibana and Tachibana hypothesised that the addition of microbubbles would lower the threshold for cavitation when using low-intensity ultrasound.19 They tested this by creating fresh thrombi using whole venous blood drawn from a healthy volunteer. These thrombi were divided in three groups, with each group receiving a different treatment strategy: either Urokinase® alone, Urokinase® and ultrasound or Urokinase®, ultrasound and microbubbles. Low-intensity ultrasound was applied for 3 minutes at 170 kHz and 0.5 W/cm2 in an on–off sequence (2 seconds on and 4 seconds off). After treatment, the thrombi were incubated and weighed at different time periods. After 60 minutes, a significant difference was found between the three groups. Treatment with ultrasound alone resulted in an increase in thrombolysis, as expected. However, the addition of microbubbles on top of ultrasound and Urokinase® enhanced the thrombolytic effect even further.19

ighed at different time periods. After 60 minutes, a significant difference was found between the three groups. Treatment with ultrasound alone resulted in an increase in thrombolysis, as expected. However, the addition of microbubbles on top of ultrasound and Urokinase® enhanced the thrombolytic effect even further.19 In 2007, Prokop et al. carried out an experiment to test which form of cavitation results in ultrasound-accelerated, t-PA-mediated fibrinolysis.20 Although inertial cavitation was detected at the start of the experiment with contrast ultrasound on top of t-PA, this was not the dominant mechanism for enhanced fibrinolysis. When the microbubbles were pre-treated with ultrasound to prevent inertial cavitation, the reduction in clot weight was the same without pre-treatment, confirming that stable cavitation is the key mechanism for ultrasound-enhanced fibrinolysis after the addition of microbubbles.20 The different composure of microbubbles might also influence the results. However, this has never been fully explored. Therefore, clinical trials today are performed using commercially available microbubbles.

In 2007, Prokop et al. carried out an experiment to test which form of cavitation results in ultrasound-accelerated, t-PA-mediated fibrinolysis.20 Although inertial cavitation was detected at the start of the experiment with contrast ultrasound on top of t-PA, this was not the dominant mechanism for enhanced fibrinolysis. When the microbubbles were pre-treated with ultrasound to prevent inertial cavitation, the reduction in clot weight was the same without pre-treatment, confirming that stable cavitation is the key mechanism for ultrasound-enhanced fibrinolysis after the addition of microbubbles.20 The different composure of microbubbles might also influence the results. However, this has never been fully explored. Therefore, clinical trials today are performed using commercially available microbubbles. Effect of pulse duration These results were confirmed in a study performed by Petit et al.21 Additionally, they also demonstrated that, in order for stable cavitation to enhance fibrinolysis, a longer pulse length is needed. A longer pulse duration results in a higher temporal average intensity, which can sustain any induced microbubble cavitational activity. These in vitro results were confirmed in an in vivo study performed by Xie et al.22 Using a diagnostic ultrasound system, they demonstrated in an atherosclerotic pig model with STEMI that even a slightly longer pulse duration (20 µs) resulted in higher recanalisation rates (Figure 1). When using a diagnostic ultrasound system, it was observed that thrombus age and the angle of the transducer also had influences on the thrombolytic effect of contrast ultrasound.7

atherosclerotic pig model with STEMI that even a slightly longer pulse duration (20 µs) resulted in higher recanalisation rates (Figure 1). When using a diagnostic ultrasound system, it was observed that thrombus age and the angle of the transducer also had influences on the thrombolytic effect of contrast ultrasound.7 Figure 1. Macrovascular effect of ultrasound and microbubbles, resulting in increased concentrations of t-PA due to microstreaming and thrombus disruption due to cavitation. (A) illustrates a total occlusion of a coronary artery due to thrombus formation. (B) illustrates improvement of flow under the influence of ultrasound and cavitating microbubbles. t-PA: tissue plasminogen activator. All of these studies were performed with intermittent applications of ultrasound in order for microvascular replenishment to occur in between the high mechanical index (MI) impulses. The pulse durations were not prolonged further than 20 µs in order to prevent unwanted side effects, such as vascular wall damage and haemolysis, seen with longer pulse durations or continuous-wave ultrasound.12 There is another advantage of using image-guided, interrupted ultrasound, as it permits the visual assessment of the effect of treatment (Figure 2). Figure 2. Change in replenishment kinetics during treatment of an acute ST elevation myocardial infarction, confirmed on coronary angiography (total occlusion of the left anterior descending artery (LAD), pointed out by the arrow).21 (Courtesy of Dr TR Porter, University of Nebraska Medical Center, Omaha, Nebraska, USA). MI: mechanical index.

enishment kinetics during treatment of an acute ST elevation myocardial infarction, confirmed on coronary angiography (total occlusion of the left anterior descending artery (LAD), pointed out by the arrow).21 (Courtesy of Dr TR Porter, University of Nebraska Medical Center, Omaha, Nebraska, USA). MI: mechanical index. Despite these successful preliminary studies, the optimal ultrasound settings remain a matter of debate. In addition, when applying it clinically in acute STEMI, attenuation plays an important role, making it difficult to establish the optimal settings when translating these results to human studies.23 Targeted contrast ultrasound Another potential advantage of contrast ultrasound is the possibility of targeting microbubbles to a site of interest, by conjugating the microbubble surface with agents that would adhere to upregulated receptors. At first, it was thought that microbubbles passed freely through the microcirculation. However, it was discovered that this was not the case in inflamed or injured tissue. Activated neutrophils and monocytes cause microbubbles with an albumin or phosphatidyl serine-containing lipid shell to bind and become phagocytosed. In this circumstance, the microbubble remains acoustically active and thus creates an increased ultrasound signal locally.24

s not the case in inflamed or injured tissue. Activated neutrophils and monocytes cause microbubbles with an albumin or phosphatidyl serine-containing lipid shell to bind and become phagocytosed. In this circumstance, the microbubble remains acoustically active and thus creates an increased ultrasound signal locally.24 More specific targeting strategies can be utilised to enhance thrombus dissolution with ultrasound and microbubbles (sonothrombolysis). By incorporating a single-chain antibody specific for activated glycoprotein IIb/IIIa into the lipid shell of the microbubble, the interaction with activated platelets can be enhanced, thereby locally increasing the concentration of microbubbles and thus increasing signal intensity.25 Positive results with targeted contrast ultrasound have already been demonstrated in an in vivo pig study by Xie et al.26 A recent study with both targeted ultrasound and targeted fibrinolytic agents in an in vivo mouse model demonstrated promising results not only with respect to increased fibrinolysis, but also with respect to lower bleeding risks.27,28

st ultrasound have already been demonstrated in an in vivo pig study by Xie et al.26 A recent study with both targeted ultrasound and targeted fibrinolytic agents in an in vivo mouse model demonstrated promising results not only with respect to increased fibrinolysis, but also with respect to lower bleeding risks.27,28 Clinical trials The first sonothrombolysis trials in humans were performed with non-diagnostic ultrasound systems at low frequencies. Cohen and colleagues were the first to investigate the effect of ultrasound-enhanced thrombolysis in 25 patients with an acute STEMI.29 After patients were pre-treated with thrombolytics, low-frequency ultrasound was applied for 60 minutes, followed by a coronary angiogram. With 64% of the patients having TIMI III flow after treatment, results compared favourably with historical data. However, the PLUS trial, a large, multicentre, randomised, clinical trial examining the effect of low-frequency ultrasound added to fibrinolytic agents, was halted after an interim analysis in 360 patients demonstrated a lack of treatment efficacy, both in TIMI flow grade as in ST segment resolution.30 No microbubbles or image-guided therapy were used in both trials.

ndomised, clinical trial examining the effect of low-frequency ultrasound added to fibrinolytic agents, was halted after an interim analysis in 360 patients demonstrated a lack of treatment efficacy, both in TIMI flow grade as in ST segment resolution.30 No microbubbles or image-guided therapy were used in both trials. Our group was the first to conduct a feasibility study using diagnostic ultrasound and commercially available microbubbles to the treatment protocol. Although it involved only 10 patients, this first trial demonstrated that sonothrombolysis was safe and feasible.31 Subsequent to this, Mathias Jr et al. published the results of a large clinical trial using contrast ultrasound with different ultrasound settings in acute STEMI patients.32 Prior to PCI, there was a significantly higher epicardial coronary recanalisation rate in the patient group receiving image-guided, high-mechanical index impulses. Interestingly, in contrast to in vitro and in vivo studies, even patients treated with a short pulse duration (3-4 µs) had a higher epicardial recanalisation rate. Microvascular flow also appeared improved, indicating that other factors, in addition to the sonothrombolytic effect, may be improving early and longer-term outcomes.32

y, in contrast to in vitro and in vivo studies, even patients treated with a short pulse duration (3-4 µs) had a higher epicardial recanalisation rate. Microvascular flow also appeared improved, indicating that other factors, in addition to the sonothrombolytic effect, may be improving early and longer-term outcomes.32 Microvascular obstruction The introduction of primary PCI improved revascularisation rates significantly for patients with STEMI. However, in up to 50% of patients, myocardial perfusion has remained poor in the area at risk.33 The causes of this persistent microvascular obstruction (MVO) can be categorised into four different pathogenetic components, which are distal embolisation, injury caused by ischemia, injury caused by reperfusion and individual predisposition. Contrast ultrasound may improve outcomes by affecting one or all of these components, and may potentially reduce the frequency of MVO in STEMI.

e categorised into four different pathogenetic components, which are distal embolisation, injury caused by ischemia, injury caused by reperfusion and individual predisposition. Contrast ultrasound may improve outcomes by affecting one or all of these components, and may potentially reduce the frequency of MVO in STEMI. Pathophysiology During a study aiming for epicardial recanalisation using contrast ultrasound in combination with thrombolysis, Xie et al. encountered an improvement in microvascular flow despite an absence of epicardial coronary recanalisation.34 It was therefore hypothesised that contrast ultrasound has a potential therapeutic effect on microvascular perfusion, reducing the effect of MVO. This was confirmed in a microvascular rat model by Leeman et al.35 In this study, the authors demonstrated that both intermittently applied higher mechanical index and longer pulse duration improved microvascular blood flow in a pure model of microvascular thrombotic obstruction. In a later study, using the same model, t-PA was added, resulting in an improvement of microvascular reperfusion.36 Recently, Belcik et al. confirmed these results in a hind limb ischemia mouse model using contrast ultrasound, and the authors observed that this improvement in microvascular flow lasted for more than 24 hours.37 Even though the exact mechanism underlying this improvement in microvascular flow might not be known, a fair amount of studies with contrast ultrasound have been conducted in order to explore the effect on cellular level and thereby its potential beneficial effect on MVO.

Pathophysiology During a study aiming for epicardial recanalisation using contrast ultrasound in combination with thrombolysis, Xie et al. encountered an improvement in microvascular flow despite an absence of epicardial coronary recanalisation.34 It was therefore hypothesised that contrast ultrasound has a potential therapeutic effect on microvascular perfusion, reducing the effect of MVO. This was confirmed in a microvascular rat model by Leeman et al.35 In this study, the authors demonstrated that both intermittently applied higher mechanical index and longer pulse duration improved microvascular blood flow in a pure model of microvascular thrombotic obstruction. In a later study, using the same model, t-PA was added, resulting in an improvement of microvascular reperfusion.36 Recently, Belcik et al. confirmed these results in a hind limb ischemia mouse model using contrast ultrasound, and the authors observed that this improvement in microvascular flow lasted for more than 24 hours.37 Even though the exact mechanism underlying this improvement in microvascular flow might not be known, a fair amount of studies with contrast ultrasound have been conducted in order to explore the effect on cellular level and thereby its potential beneficial effect on MVO. Nitric oxide It is known that when the balance between nitric oxide (NO) synthesis and superoxide production falls in favour of superoxide production after reperfusion, this will result in vasoconstriction and exacerbation of the inflammatory state.38 Increasing the amount of NO release might counteract this imbalance after reperfusion. Therefore, it was hypothesised that vasodilation due to the release of NO was one of the main contributors to this improvement in local perfusion. This was subsequently demonstrated by pre-treating animals with L-Nω-nitroarginine methyl ester (L-NAME), a strong inhibitor of NO synthase. This time, tissue perfusion units (TPU) and pH did not improve under the influence of ultrasound.39 These results were confirmed in another in vivo study where the coronary arteries of nine dogs and five pigs were occluded, and local myocardial perfusion improved in both groups after treatment with low-frequency ultrasound.40

me, tissue perfusion units (TPU) and pH did not improve under the influence of ultrasound.39 These results were confirmed in another in vivo study where the coronary arteries of nine dogs and five pigs were occluded, and local myocardial perfusion improved in both groups after treatment with low-frequency ultrasound.40 In a rat study using the same pure microvascular thrombosis protocol used by Leeman et al.,35 intermittent high-mechanical index impulses and 20-µs pulse duration ultrasound were applied through a tissue-mimicking phantom in order to simulate transthoracic attenuation, and this proved effective at improving microvascular flow during a commercially available microbubble infusion.41 In this study, when the NO synthase inhibitor L-NAME was administered, there was a higher perivascular haemorrhage rate. Thus, NO production induced by diagnostic ultrasound high MI impulses may be preventing unwanted bioeffects.41

at improving microvascular flow during a commercially available microbubble infusion.41 In this study, when the NO synthase inhibitor L-NAME was administered, there was a higher perivascular haemorrhage rate. Thus, NO production induced by diagnostic ultrasound high MI impulses may be preventing unwanted bioeffects.41 In an animal study by Xie and colleagues, microbubbles were added on top of ultrasound and t-PA.26,34 They also detected an increase in microvascular flow in the peri-infarct zone, despite the absence of epicardial coronary recanalisation. Since ultrasound-induced microbubble cavitation causes local shear stress on both thrombus and endothelial borders, it was postulated that local mechanical forces stimulate endothelial cells to produce NO.42 This mechanism of stimulating the release of NO might be due to the production of intracellular hydrogen peroxide (H2O2). In a study performed by Juffermans et al., the production of intracellular H2O2 was detected in vitro after treatment of the endothelial cell with contrast ultrasound.43 This intracellular H2O2 has been shown to stimulate the production of NO by the activation of endothelial NO synthase, counteracting the imbalance between NO synthesis and superoxide production that occurs after reperfusion (Figure 3).44 The duration and intensity of cavitation induced by ultrasound, as well as the cellular environment in which cavitation occurs, may influence what bioeffect is observed. For example, if microbubbles are phagocytosed by neutrophil granulocytes, there may be amplification of superoxide generation, even prior to ultrasound-induced cavitation.45,46 However, once high-intensity ultrasound with a long pulse duration was applied, a decrease in superoxide generation was observed. Nevertheless, the application of high-intensity ultrasound caused an increase in apoptosis, membrane injury and complete cell destruction. This effect was stronger using albumin microbubbles compared to lipid microbubbles.45 Therefore, further investigation of the optimal ultrasound intensity setting in combination with microbubbles is needed, in addition to exploring what effect that surrounding cellular environment has on outcome.

lete cell destruction. This effect was stronger using albumin microbubbles compared to lipid microbubbles.45 Therefore, further investigation of the optimal ultrasound intensity setting in combination with microbubbles is needed, in addition to exploring what effect that surrounding cellular environment has on outcome. Figure 3. Microvascular effect of ultrasound and microbubbles on an endothelial cell level, resulting in a decreased inflammatory response, leading to vasodilation through: (1) activation of shear stress resulting in an increase in intracellular H2O2, which results in the production of extracellular NO; (2) a decrease of superoxide; and (3) a decrease in rolling/adhered leukocytes. (A) The untreated situation without ultrasound; (B) treatment with ultrasound. NO: nitric oxide; H2O2: hydrogen peroxide; O2−: superoxide.

Figure 3. Microvascular effect of ultrasound and microbubbles on an endothelial cell level, resulting in a decreased inflammatory response, leading to vasodilation through: (1) activation of shear stress resulting in an increase in intracellular H2O2, which results in the production of extracellular NO; (2) a decrease of superoxide; and (3) a decrease in rolling/adhered leukocytes. (A) The untreated situation without ultrasound; (B) treatment with ultrasound. NO: nitric oxide; H2O2: hydrogen peroxide; O2−: superoxide. Inflammatory response Another important mechanism resulting in reperfusion injury is the activation of the inflammatory response. Under the influence of changes in the local mechanical forces and the increased local release of reactive oxygen species after reperfusion, there is an increase in leukocyte adhesion. These activated leukocytes in turn enhance the inflammatory response, can form aggregates with platelets and release vasoconstrictors like superoxide.47 Decreasing the inflammatory response might have a positive effect on the microcirculation after ischemia/reperfusion, resulting in a lower prevalence of MVO (Figure 3). In several in vivo studies, it has been demonstrated that application of ultrasound alone at the area where ischemia/reperfusion was created resulted in local vasodilation and a decrease in the number of rolling and adhered leukocytes. When microbubbles were added in this setting, ultrasound application further enhanced vasodilation and decreased the number of leukocytes.48–50

pplication of ultrasound alone at the area where ischemia/reperfusion was created resulted in local vasodilation and a decrease in the number of rolling and adhered leukocytes. When microbubbles were added in this setting, ultrasound application further enhanced vasodilation and decreased the number of leukocytes.48–50 Interestingly, in an in vivo mouse model, no clear disruption of leukocytes could be observed after application of ultrasound with a mechanical index of 0.9 to phagocytosed microbubbles.51 Clinical trials The first-in-human, Phase II trial has been performed by Roos et al.52 In this study aiming at improving microvascular reperfusion with a longer pulse duration (20 µs), ultrasound in combination with microbubbles was used to treat patients at up to a maximum of 15 minutes prior to PCI and 30 minutes post-PCI. Unfortunately, the study was discontinued after six patients, as the investigators encountered coronary spasm of the culprit vessel in 50% of the patients. These results were confirmed in a pig model when a blockage of the LAD was created by balloon injury and thrombus injection and treated with long-pulse duration ultrasound.52

ely, the study was discontinued after six patients, as the investigators encountered coronary spasm of the culprit vessel in 50% of the patients. These results were confirmed in a pig model when a blockage of the LAD was created by balloon injury and thrombus injection and treated with long-pulse duration ultrasound.52 Although contrast ultrasound has been shown to improve microvascular perfusion despite absent epicardial coronary recanalisation, the full mechanism of this effect still remains unclear. Furthermore, pulse duration seems to play an important role in enhancing microvascular perfusion. Nevertheless, it also results in local vasoconstriction. Further research is warranted in order to investigate the working mechanism, as well as the optimal settings. Safety and feasibility Concerns have also risen regarding possible adverse effects of the treatment of reperfusion injury using contrast ultrasound. It has been demonstrated that high-intensity, focused ultrasound results in platelet activation, especially in the presence of inertial cavitation.53,54 When microbubbles were added, it was demonstrated that only at very high microbubble concentrations could platelets be activated, and when ultrasound was applied, there was a significant further increase in platelet activation. It must be emphasised that these results were obtained in platelet-rich plasma and with high concentrations of microbubbles. With normal concentrations of microbubbles added to samples of whole blood, no significant increase in platelet activation was observed, not even after ultrasound application.55

telet activation. It must be emphasised that these results were obtained in platelet-rich plasma and with high concentrations of microbubbles. With normal concentrations of microbubbles added to samples of whole blood, no significant increase in platelet activation was observed, not even after ultrasound application.55 Furthermore, it was also shown that high-intensity ultrasound can perforate normal vessel walls.12 Nevertheless, this effect was only observed at very high intensities in an in vitro study using tissue from a necropsy. In vivo studies did not encounter this adverse effect. Another safety issue encountered was coronary spasm during contrast ultrasound with longer pulse durations during a Phase II trial, which was therefore halted at an early stage.52 The intensity and pulse duration seem to play important roles in both mechanical effects as well as adverse effects. Nevertheless, several in-human studies have been performed thus far, demonstrating safety and feasibility (Table 1). Table 1. Studies assessing the effect of ultrasound on epicardial recanalisation in ST elevation myocardial infarction patients.

Another safety issue encountered was coronary spasm during contrast ultrasound with longer pulse durations during a Phase II trial, which was therefore halted at an early stage.52 The intensity and pulse duration seem to play important roles in both mechanical effects as well as adverse effects. Nevertheless, several in-human studies have been performed thus far, demonstrating safety and feasibility (Table 1). Table 1. Studies assessing the effect of ultrasound on epicardial recanalisation in ST elevation myocardial infarction patients. First author/reference No. of patients Contrast used Drugs US settings Outcome Cohen et al.29 25 No Reteplase or tenecteplase 27 kHz, continuous wave, 0.9 W/cm2 Safe and feasible Hudson et al.30 360 No ASA, heparin or enoxaparin, tenecteplase 28 kHz, pulsed wave, 0.38 W/cm2 No improvement Slikkerveer et al.31 10 Yes ASA, heparin, alteplase 1.6 MHz, MI 1.18, pulse duration 5 µs Safe and feasible Roos et al.52 6 Yes ASA, heparin, ticagrelor 1.6 MHz, MI 1.3, pulse duration 20 µs Coronary spasm Mathias Jr et al.32 30 Yes ASA, heparin, clopidogrel 1.3/1.8 MHz, MI 1.1–1.3, pulse duration 3/5/20 µs Safe and feasible ASA: acetylsalicylic acid; US: ultrasound; MI: mechanical index.

ulse duration 5 µs Safe and feasible Roos et al.52 6 Yes ASA, heparin, ticagrelor 1.6 MHz, MI 1.3, pulse duration 20 µs Coronary spasm Mathias Jr et al.32 30 Yes ASA, heparin, clopidogrel 1.3/1.8 MHz, MI 1.1–1.3, pulse duration 3/5/20 µs Safe and feasible ASA: acetylsalicylic acid; US: ultrasound; MI: mechanical index. Future directions With the recent publication of the positive results by Mathias Jr et al.,32 the next step is to initiate a larger, multicentre, randomised, clinical trial aiming to improve both epicardial coronary recanalisation as well as MVO using contrast ultrasound in STEMI patients. Since the group treated with high-mechanical index impulses and shorter pulse durations demonstrated a higher epicardial recanalisation rate without signs of coronary spasm, it seems to be appropriate to use these settings in future studies. In addition to ST resolution on an electrocardiogram and epicardial recanalisation on angiography, magnetic resonance imaging and myocardial contrast echocardiography might also be performed in order to examine the effects of these impulse on the subsequent frequency of MVO at specific time points following treatment.

In addition to ST resolution on an electrocardiogram and epicardial recanalisation on angiography, magnetic resonance imaging and myocardial contrast echocardiography might also be performed in order to examine the effects of these impulse on the subsequent frequency of MVO at specific time points following treatment. Technical issues When taking all of the different settings into consideration for reaching an optimal therapeutic effect and preventing serious side effects, it raises the question as to whether this new therapy is limited to centres with extensive experience. Since the largest improvement in epicardial thrombus dissolution can be obtained in the prehospital setting, care should be taken with future research that a protocol will be developed that can be performed in this prehospital setting by non-experienced users. The same goes for treatment of MVO using contrast ultrasound after primary PCI has been performed. This also calls for the development of a device, such as an ultrasound vest, which could be transported in the prehospital setting and is easy to use.

can be performed in this prehospital setting by non-experienced users. The same goes for treatment of MVO using contrast ultrasound after primary PCI has been performed. This also calls for the development of a device, such as an ultrasound vest, which could be transported in the prehospital setting and is easy to use. Summary We focused on the therapeutic application of contrast ultrasound in STEMI. A considerable amount of research has been performed to demonstrate that contrast ultrasound enhances thrombolysis. Although the exact working mechanism remains to be elucidated, it is known that destruction of microbubbles and local application of ultrasound induces several bioeffects, resulting in enhanced thrombolysis. Multiple animal infarct studies confirmed the therapeutic application of contrast ultrasound in STEMI, and the first human studies show promising results.

ns to be elucidated, it is known that destruction of microbubbles and local application of ultrasound induces several bioeffects, resulting in enhanced thrombolysis. Multiple animal infarct studies confirmed the therapeutic application of contrast ultrasound in STEMI, and the first human studies show promising results. The aforementioned animal studies also demonstrated that despite an absence of epicardial coronary recanalisation, the peripheral perfusion of the area at risk did improve, indicating microvascular effects that are independent of upstream vascular flow and obstruction. This may be mediated by cavitation-induced activation of purinergic pathways, leading to prolonged increases in NO production that have a positive influence on the imbalance between NO and superoxide production and subsequent inflammatory responses. There does appear to be an ultrasound dose–response curve in the setting of a microbubble infusion, in that inertial cavitation at a longer pulse duration may cause unwanted vasospasm in the presence of microbubbles. Pulse duration seems to be playing a major role in these side effects. Nevertheless, the local destruction of microbubbles under the influence of ultrasound might have a positive effect on MVO. Therefore, further research is warranted in order to explore the potential application of contrast ultrasound in patients suffering MVO after successful primary PCI. Conflict of interest: The authors declare that there is no conflict of interest.

The aforementioned animal studies also demonstrated that despite an absence of epicardial coronary recanalisation, the peripheral perfusion of the area at risk did improve, indicating microvascular effects that are independent of upstream vascular flow and obstruction. This may be mediated by cavitation-induced activation of purinergic pathways, leading to prolonged increases in NO production that have a positive influence on the imbalance between NO and superoxide production and subsequent inflammatory responses. There does appear to be an ultrasound dose–response curve in the setting of a microbubble infusion, in that inertial cavitation at a longer pulse duration may cause unwanted vasospasm in the presence of microbubbles. Pulse duration seems to be playing a major role in these side effects. Nevertheless, the local destruction of microbubbles under the influence of ultrasound might have a positive effect on MVO. Therefore, further research is warranted in order to explore the potential application of contrast ultrasound in patients suffering MVO after successful primary PCI. Conflict of interest: The authors declare that there is no conflict of interest. Funding: This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Introduction Left ventricular free-wall rupture (LVFWR) may represent a dramatic and life-threatening event, occurring in up to 2% of patients with acute myocardial infarction (AMI).1,2 The clinical presentation varies from a catastrophic blowout type characterized by cardiogenic shock and eventually cardiac arrest, to the oozing type with haemodynamic instability and pericardial effusion.3 Prompt diagnosis and management can lead to successful treatment for LVFWR. Conservative strategies have been described,4,5 but the poor results of this treatment make surgical intervention generally mandatory. Despite different surgical approaches having been proposed, in-hospital mortality continues to be high6,7 and the most appropriate surgical management remains still controversial, particularly in terms of rupture recurrence or other type of post-operative complications. This review describes the approaches reported in the literature regarding the management of this mechanical complication of AMI. History The first free-wall rupture of the heart after AMI was described by William Harvey in 1647.8 In 1769, Morgagni collected and reported 11 cases of myocardial rupture discovered postmortem9 (Figure 1). Ironically, Morgagni himself later died of ventricular rupture. Duaine, in 1871, reported the first large series of patients with cardiac rupture and concluded that rupture never occurs spontaneously.10 However, it was not until 1925, in the report of Krumbhaar and Crowell, that the relationship of LVFWR to myocardial infarction was first pointed out.11

died of ventricular rupture. Duaine, in 1871, reported the first large series of patients with cardiac rupture and concluded that rupture never occurs spontaneously.10 However, it was not until 1925, in the report of Krumbhaar and Crowell, that the relationship of LVFWR to myocardial infarction was first pointed out.11 Figure 1. Morgagni’s description of a case of ventricular wall rupture. Frontispiece of De sedibus, et causis morborum per anatomen indagatis (JB Morgagni, 1765), with a portrait of Morgagni and Morgagni’s description (‘Epistola anatomico-medica XXVII’) of a case of ventricular wall rupture (by courtesy of the Medical Library of Spedali Civili, Brescia, Italy). Hatcher and associates, from Emory University, in 1970 described the first successful operation for free-wall rupture of the right ventricle.12 Two years later, Fitzgibbon et al. reported the successful repair of LVFWR associated with ischaemic heart disease.13 Cardiac rupture was treated, on cardiopulmonary bypass (CPB), with infarctectomy and closure of the myocardial defect. In the same year, Montegut described the history of a 58-year-old man who was successfully treated for post-infarction LVFWR with direct suture.10

ir of LVFWR associated with ischaemic heart disease.13 Cardiac rupture was treated, on cardiopulmonary bypass (CPB), with infarctectomy and closure of the myocardial defect. In the same year, Montegut described the history of a 58-year-old man who was successfully treated for post-infarction LVFWR with direct suture.10 In 1973, Calick et al. reported the surgical repair of ruptured myocardium with a Dacron® patch placed over the area of perforation.14 Ten years later, Nunez and colleagues described post-AMI free-wall ventricular rupture in seven patients who underwent surgical intervention.15 The control of haemorrhage was obtained by covering the ventricular tear and the surrounding infarcted myocardium with a Teflon® patch fixed on epicardium by a continuous Prolene® suture.15 The first attempt to repair LVFWR with a patch fixed with glue over the area of rupture was reported by Löfström and associates in 1972, but the Dacron® patch loosened soon after the operation, causing the patient’s death.16 In 1993, the first series of patients successfully treated with a sutureless epicardial patch technique was described.3 Padró and colleagues successfully treated 13 patients with subacute LVFWR using a Teflon® patch fixed onto the heart surface with only surgical glue.3

ter the operation, causing the patient’s death.16 In 1993, the first series of patients successfully treated with a sutureless epicardial patch technique was described.3 Padró and colleagues successfully treated 13 patients with subacute LVFWR using a Teflon® patch fixed onto the heart surface with only surgical glue.3 Operative strategies Clinical presentation and course are variable, depending on the location, the size and the time-course of the rupture.17 Although acute LVFWR is often fatal, some patients with subacute or contained rupture present with a window of opportunity for intervention.18,19 Thereby, the diagnosis or even a high suspicion of cardiac rupture represents an indication for emergency surgery without further delay. The most important diagnostic method for LVFWR is transthoracic echocardiography: the presence of reduced myocardial wall thickness, haemopericardium or epicardial clots and cardiac tamponade are the most relevant findings.20 In suitable patients, cardiac magnetic resonance can complement the diagnosis by identifying contained ventricular rupture.21 Pericardiocentesis prior to surgery, which confirms a haemorrhagic effusion, may further support the diagnosis, but the definitive diagnosis is usually made at surgery.

most relevant findings.20 In suitable patients, cardiac magnetic resonance can complement the diagnosis by identifying contained ventricular rupture.21 Pericardiocentesis prior to surgery, which confirms a haemorrhagic effusion, may further support the diagnosis, but the definitive diagnosis is usually made at surgery. Rarely, the rupture of the free-wall of the left ventricle is contained by epicardial clots or pericardial adhesions, leading to the formation of a pseudoaneurysm. Left ventricular pseudoaneurysm (LVPA) that occurs a few days after AMI may be unstable and tends to rupture.22 Since the clinical presentation of this condition is variable, ranging from uneventful condition to congestive heart failure, or even sudden death,23 a high index of suspicion for this serious complication of AMI is paramount. Diagnosis can be made by several imaging techniques, including echocardiography, computed tomography, ventriculography and magnetic resonance imaging (MRI). Echocardiography is the first-line imaging modality in suspected LVPA. Colour-Doppler can help to demonstrate the discontinuity of the ventricular wall and highlight the distinctive bi-directional flow between the extracardiac echo-free space and the left ventricle.24 Contrast ventriculography has the likelihood of establishing a definitive diagnosis23 and can help along with coronary angiography in surgical planning; however, it is an invasive procedure that might precipitate ventricular rupture. Owing to its ability to visualize the heart in precise detail, MRI can provide essential information to confirm diagnosis and clearly elucidate the characteristics of the pseudoaneurysm and guide surgical intervention.25 However, MRI can be performed only in patients with LVPA who remain haemodynamically stable. When the diagnosis of LVPA is established, surgical correction is usually mandatory because onset of rupture is unpredictable.

s and clearly elucidate the characteristics of the pseudoaneurysm and guide surgical intervention.25 However, MRI can be performed only in patients with LVPA who remain haemodynamically stable. When the diagnosis of LVPA is established, surgical correction is usually mandatory because onset of rupture is unpredictable. On the way to surgery, restoration of a satisfactory haemodynamic state may require inotropic support, intra-venous fluid, intra-aortic balloon pump (IABP) and pericardial drainage, as guided by clinical status (Figure 2). Pericardiocentesis may relieve cardiac tamponade and improve circulation in a critical situation; however, it is often unhelpful because much of the pericardial space is taken up by non-drainable clots. Decompression of the pericardium may also be obtained by subxiphoid drainage or by a classic sternotomy. In the presence of refractory cardiac arrest, extracorporeal membrane oxygenation might provide a chance to perform definitive surgical treatment.26 Figure 2. Pre-operative management. An algorithm that can be applied in cases of left ventricular free-wall rupture. AMI: acute myocardial infarction; LVFWR: left ventricular free-wall rupture; IABP: intra-aortic balloon pump.

On the way to surgery, restoration of a satisfactory haemodynamic state may require inotropic support, intra-venous fluid, intra-aortic balloon pump (IABP) and pericardial drainage, as guided by clinical status (Figure 2). Pericardiocentesis may relieve cardiac tamponade and improve circulation in a critical situation; however, it is often unhelpful because much of the pericardial space is taken up by non-drainable clots. Decompression of the pericardium may also be obtained by subxiphoid drainage or by a classic sternotomy. In the presence of refractory cardiac arrest, extracorporeal membrane oxygenation might provide a chance to perform definitive surgical treatment.26 Figure 2. Pre-operative management. An algorithm that can be applied in cases of left ventricular free-wall rupture. AMI: acute myocardial infarction; LVFWR: left ventricular free-wall rupture; IABP: intra-aortic balloon pump. Different opinions are published about the opportunity to perform coronary angiograms or avoid this investigation in order to ‘save time’.17,27 Some authors believe that coronary angiography should be promptly performed as soon as pericardial effusion is noted in AMI patients, before they deteriorate.28,29 The knowledge of coronary status is of great help in deciding where and how to place the sutures during surgery; in addition, a proper revascularization of the diseased vessels supplying the non-infarcted area at the time of LVFWR repair (concomitant coronary artery bypass grafting) has been shown to exert a positive impact on survival and freedom from angina.29,30

lp in deciding where and how to place the sutures during surgery; in addition, a proper revascularization of the diseased vessels supplying the non-infarcted area at the time of LVFWR repair (concomitant coronary artery bypass grafting) has been shown to exert a positive impact on survival and freedom from angina.29,30 Patients being operated for post-AMI LVFWR are generally unstable, so exposure of the femoral vessels at the groin and circuit preparation for rapid institution of CPB should be considered before thoracic incision. The use of CPB support, however, is a matter of controversy. Some authors, in fact, believe that in certain circumstances (anterior defect, haemodynamic stability, oozing type rupture, sub-acute course), and with the application of particular surgical techniques (epicardial patch covering, sutureless repair), LVFWR can be safely treated without utilizing CPB.28

r of controversy. Some authors, in fact, believe that in certain circumstances (anterior defect, haemodynamic stability, oozing type rupture, sub-acute course), and with the application of particular surgical techniques (epicardial patch covering, sutureless repair), LVFWR can be safely treated without utilizing CPB.28 The post-operative use of IABP, and other mechanical circulatory support, should always be taken into consideration because they can reduce the intra-cavitary pressure of the left ventricle, increase coronary blood flow and limit or prevent the development of low cardiac output syndrome.17,31 Timóteo and colleagues reported that the use of IABP in patients with post-AMI mechanical complications was associated with improved in-hospital outcome.32 However, the European Society of Cardiology/European Association for Cardio-Thoracic Surgery Guidelines on myocardial revascularization allocate short-term mechanical circulatory support, in the presence of post-AMI mechanical complications, in Class of recommendation IIb.33 Further investigations are required to better evaluate the efficacy and safety of mechanical devices in the setting of post-infarction LVFWR repair. General principles of post-operative care, after LVFWR repair, are shown in Figure 3. Figure 3. Post-operative management. General principles of post-operative care after left ventricular free-wall rupture repair. LVFW: left ventricular free-wall; ICU: intensive care unit; IABP: intra-aortic balloon pump.

The post-operative use of IABP, and other mechanical circulatory support, should always be taken into consideration because they can reduce the intra-cavitary pressure of the left ventricle, increase coronary blood flow and limit or prevent the development of low cardiac output syndrome.17,31 Timóteo and colleagues reported that the use of IABP in patients with post-AMI mechanical complications was associated with improved in-hospital outcome.32 However, the European Society of Cardiology/European Association for Cardio-Thoracic Surgery Guidelines on myocardial revascularization allocate short-term mechanical circulatory support, in the presence of post-AMI mechanical complications, in Class of recommendation IIb.33 Further investigations are required to better evaluate the efficacy and safety of mechanical devices in the setting of post-infarction LVFWR repair. General principles of post-operative care, after LVFWR repair, are shown in Figure 3. Figure 3. Post-operative management. General principles of post-operative care after left ventricular free-wall rupture repair. LVFW: left ventricular free-wall; ICU: intensive care unit; IABP: intra-aortic balloon pump. Surgical materials In the past, surgical treatment of LVFWR was generally achieved using only Prolene® sutures, buttressed or not with Teflon® felt. In recent years, however, cardiac procedures have employed patches and surgical glues. Different materials have been proposed in patch-based repairs. The most common applied are: Teflon®,15 Goretex®,34 Dacron®35 and pericardium,36 either autologous or xenograft. Collagen sponges, or fleeces, have been utilized in other surgical specialties, ranging from gynaecology, urology and liver surgery.37,38 The use of collagen sponges in cardiac surgery is not new; its safety has been assessed in different cardio-thoracic procedures.39 However, the indication for using TachoComb® or TachoSil® patches in LVFWR repair has become more common only in recent years, based on preliminary favourable results.40,41

er surgery.37,38 The use of collagen sponges in cardiac surgery is not new; its safety has been assessed in different cardio-thoracic procedures.39 However, the indication for using TachoComb® or TachoSil® patches in LVFWR repair has become more common only in recent years, based on preliminary favourable results.40,41 Adhesives reported to be successful in the treatment of LVFWR have been of several types and include the biologic glues (fibrin based or gelatin hydrogels) as well as the synthetic cyanoacrylate monomers. Fibrin glues function by reproducing the normal clotting cascade and result in a stable fibrin matrix after the degradation of exogenous fibrinogen. The main advantage is their lack of toxicity and complete biocompatibility.42 Gelatin-based glues have greater bonding strength than fibrin glues; however, cytotoxic effect has been shown, due to the release of formaldehyde during degradation, raising concerns regarding potential long-term complications.43 The main limitation of both these biologic glues is that they are effective only in the absence of bleeding. Synthetic glues, such as cyanoacrylate, are monomers that polymerize in an exothermic reaction when brought into contact with fluid. Although they are cytotoxic and potentially mutagenic, acetylation has greatly reduced these concerns.44 Using Histoacryl® to secure a patch of Teflon® onto the myocardium, Padró and coworkers reported a 100% survival rate among 13 patients treated for subacute LVFWR.3

mic reaction when brought into contact with fluid. Although they are cytotoxic and potentially mutagenic, acetylation has greatly reduced these concerns.44 Using Histoacryl® to secure a patch of Teflon® onto the myocardium, Padró and coworkers reported a 100% survival rate among 13 patients treated for subacute LVFWR.3 Surgical techniques As previously mentioned, several different techniques can be applied to treat post-infarction LVFWR. Depending on the optional use of sutures to treat the rupture of the ventricular wall, two different kinds of surgical approach can be proposed: sutureless and sutured repair. Initially, the sutured techniques were the only ones used. Such approaches included: i) linear closure, ii) infarctectomy and closure, and iii) patch covering. More recently, the availability of tissue adhesive materials and surgical glues have allowed the wide diffusion of the sutureless technique. Principles of surgical treatment of LVFWR are to relieve tamponade, close the tear and/or stop the bleeding, anchor the repair on healthy tissue and minimize distortion of heart geometry, while preventing recurrence of rupture or pseudoaneurysm formation. The chosen method for surgical repair is usually dictated by the type of rupture, its surrounding tissues and the presence of concomitant lesions.

and/or stop the bleeding, anchor the repair on healthy tissue and minimize distortion of heart geometry, while preventing recurrence of rupture or pseudoaneurysm formation. The chosen method for surgical repair is usually dictated by the type of rupture, its surrounding tissues and the presence of concomitant lesions. Linear closure The first successful repairs of LVFWR with linear closure were reported by Montegut10 and Cobbs and colleagues.45 In this technique, the ventricular tear is closed with Prolene® horizontal mattress sutures buttressed by Teflon® felt. The sutures should be along the non-ischaemic area to avoid suture in the necrotic myocardium, which usually generates myocardial tearing and increased bleeding. Additionally, an over and over suture can be taken, approximating the edges of the Teflon® felts, to achieve a satisfactory haemostasis45 (Figure 4). Better results have been reported using linear closure between strips of Teflon® felt, if the ventricular tear and surrounding infarct area are covered with a patch secured to the heart surface with stitches, sutures or surgical glue. Figure 4. Linear closure. The ventricular tear is closed with Prolene® horizontal mattress sutures with two supporting Teflon® felt; an over and over suture is taken to achieve haemostasis. This technique is usually not feasible in the presence of a large necrotic area, based on the potential distortion and excessive reduction of the residual left ventricular cavity.

The ventricular tear is closed with Prolene® horizontal mattress sutures with two supporting Teflon® felt; an over and over suture is taken to achieve haemostasis. This technique is usually not feasible in the presence of a large necrotic area, based on the potential distortion and excessive reduction of the residual left ventricular cavity. Infarctectomy and closure Infarct excision and closure of the defect, with either direct closure or prosthetic patch, has fallen into disuse as a result of the increased application of biological glues. These procedures are possibly now reserved for the treatment of acute massive ruptures (blow-out type rupture) or in the presence of concomitant lesions such as a post-infarction ventricular septal defect. In the conventional approach, the excision of necrotic myocardial tissue is followed by replacement using a prosthetic patch, carefully fashioned to reestablish the geometry of the left ventricular chamber17 (Figure 5), or by direct closure of the myocardial defect with interrupted mattress sutures reinforced with Teflon® felt.46 As described by Reardon et al., once the prosthetic patch is sutured with pledgeted interrupted sutures, the edge can be oversewn to the myocardium with a continuous Prolene® running suture to achieve haemostasis.17 Figure 5. Infarctectomy and closure of the defect with a prosthetic patch. After infarctectomy, a prosthetic patch is fashioned to fit this space and is sutured with pledgeted interrupted sutures.

In the conventional approach, the excision of necrotic myocardial tissue is followed by replacement using a prosthetic patch, carefully fashioned to reestablish the geometry of the left ventricular chamber17 (Figure 5), or by direct closure of the myocardial defect with interrupted mattress sutures reinforced with Teflon® felt.46 As described by Reardon et al., once the prosthetic patch is sutured with pledgeted interrupted sutures, the edge can be oversewn to the myocardium with a continuous Prolene® running suture to achieve haemostasis.17 Figure 5. Infarctectomy and closure of the defect with a prosthetic patch. After infarctectomy, a prosthetic patch is fashioned to fit this space and is sutured with pledgeted interrupted sutures. Such surgical techniques are demanding. Transmural stitches are placed in a healthy but friable myocardial tissue, so a tear could occur, and the suture line, placed along the non-ischaemic myocardium, can lead to heart geometry distortion and further deterioration of ventricular function, due to new iatrogenic myocardial infarction areas. Furthermore, heparinization associated with CPB increases the risk of bleeding.

rdial tissue, so a tear could occur, and the suture line, placed along the non-ischaemic myocardium, can lead to heart geometry distortion and further deterioration of ventricular function, due to new iatrogenic myocardial infarction areas. Furthermore, heparinization associated with CPB increases the risk of bleeding. Patch covering Some authors considered this technique the method of choice when LVFWR shows no active bleeding (or direct tear).47 In this method, ventricular rupture and the surrounding infarcted muscle are covered with a patch grafted to the healthy epicardium by Prolene® running sutures (Figure 6). Meticulous attention should be paid to avoiding coronary involvement. Glue, injected underneath the patch, is a mandatory adjunct to increase the compression strength on the myocardium and prevent blood leakage to the epicardial surface along the suture line.28,29 Figure 6. Patch covering technique. A patch of pericardium is sealed on the infarcted area using surgical glue and fixed with a running Prolene® suture on the surrounding healthy myocardium.

Patch covering Some authors considered this technique the method of choice when LVFWR shows no active bleeding (or direct tear).47 In this method, ventricular rupture and the surrounding infarcted muscle are covered with a patch grafted to the healthy epicardium by Prolene® running sutures (Figure 6). Meticulous attention should be paid to avoiding coronary involvement. Glue, injected underneath the patch, is a mandatory adjunct to increase the compression strength on the myocardium and prevent blood leakage to the epicardial surface along the suture line.28,29 Figure 6. Patch covering technique. A patch of pericardium is sealed on the infarcted area using surgical glue and fixed with a running Prolene® suture on the surrounding healthy myocardium. Because the anchoring sutures are placed only in the epicardium, myocardial damage is minimal. Furthermore, in active bleeding cardiac ruptures, fixing the patch with running sutures appears to be more effective to prevent re-rupture than a sutureless technique. Although this type of repair can be performed avoiding CPB, some surgeons prefer to perform it on CPB with aortic cross clamp to reduce the tension on the epicardial tissue and cardiac movement, thereby favouring a quicker and more effective running suture and reducing the risk of further myocardial tearing.48

chnique. Although this type of repair can be performed avoiding CPB, some surgeons prefer to perform it on CPB with aortic cross clamp to reduce the tension on the epicardial tissue and cardiac movement, thereby favouring a quicker and more effective running suture and reducing the risk of further myocardial tearing.48 Pocar and colleagues reported a modified patch covering technique in which a Tachosil® fleece was applied to widely cover the ventricular tear and the adjacent infarcted tissue; thereafter, a generous pericardial patch was fixed with a few separate Prolene® stitches and fibrin glue was injected to seal the two layers.40 Sutureless repair Availability of tissue adhesive materials has allowed a sutureless patch technique of repair for the oozing type of ventricular rupture. A prosthetic patch, adequately fashioned to cover the area of haematoma and muscle necrosis, its borders placed on healthy myocardium in the peri-necrotic area, is fixed onto the myocardial surface with surgical glue. The glued patch must be held in place for 1 to 3 min (depending on the type of glue used), until it becomes firmly adherent and haemostatic. As reported above, a wide range of synthetic and biological glues can be used for this purpose. The obvious limitation of the biological glues is that they are effective only in the absence of active bleeding. So glues should preferably be used if the tear is sealed or the lesion is of the oozing type.

and haemostatic. As reported above, a wide range of synthetic and biological glues can be used for this purpose. The obvious limitation of the biological glues is that they are effective only in the absence of active bleeding. So glues should preferably be used if the tear is sealed or the lesion is of the oozing type. Lachapelle et al. suggested that even patients with actively bleeding lesions can be treated using sutureless repair, provided they are on CPB with total decompression of the heart.49 Nevertheless, concerns remain over the safety of this technique when applied to treat blow-out ruptures. In this setting, the risk of recurrent re-rupture and pseudoaneurysm formation is high, since it is very unlikely that a simply glued patch can withstand the intraventricular pressure when there is a direct communication between the left ventricular cavity and pericardial space. Recent reports are likely to shift toward the use of collagen sponge patch (Figure 7). A TachoComb® or Tachosil® patch is placed and pressed with wet gauze to the dry surface of the ventricle for a few minutes. When a simple patch is not large enough to cover the entire damage surface, a second patch is applied. Again, surrounding healthy myocardium not involved in the necrotic process is important to allow better anchoring and avoid re-rupture. This surgical approach, particularly when reserved to treat patients with oozing type rupture, has shown satisfactory clinical results.50,51 Figure 7. Sutureless repair.

Recent reports are likely to shift toward the use of collagen sponge patch (Figure 7). A TachoComb® or Tachosil® patch is placed and pressed with wet gauze to the dry surface of the ventricle for a few minutes. When a simple patch is not large enough to cover the entire damage surface, a second patch is applied. Again, surrounding healthy myocardium not involved in the necrotic process is important to allow better anchoring and avoid re-rupture. This surgical approach, particularly when reserved to treat patients with oozing type rupture, has shown satisfactory clinical results.50,51 Figure 7. Sutureless repair. A TachoSil® patch is applied to widely cover the ventricular wall rupture and the adjacent infarcted tissues. In conclusion, the sutureless technique is simple, fast and can be done without CPB; moreover, it preserves left ventricular geometry and, leaving the necrotic tissue untouched, provides complete haemostasis. However, before considering this approach to be safe and suitable for all types of LVFWR, further investigations are required.

the sutureless technique is simple, fast and can be done without CPB; moreover, it preserves left ventricular geometry and, leaving the necrotic tissue untouched, provides complete haemostasis. However, before considering this approach to be safe and suitable for all types of LVFWR, further investigations are required. Percutaneous intra-pericardial fibrin-glue injection therapy Recently, a new therapeutic option, namely percutaneous intra-pericardial fibrin-glue injection therapy (PIFIT), has been introduced to the clinical setting. The use of PIFIT to enhance haemostasis in post-infarction LVFWR was first described by Ogiwara et al. in 1995.52 Fibrin-glue, comprising fibrinogen and thrombin, when topically applied, exerts a local haemostatic and sealant effect. It is widely used in various surgical procedures.53 PITIF for LVFWR was reported in several case reports with reasonable short- and medium-term clinical outcomes.54,55 The use of intra-pericardial thrombin injection, as an alternative sealant, has also been reported.56 Despite an initial concern about uncontrolled pericardial adhesion after fibrin-glue application, reports, using serial echocardiographic follow-up, have not shown development of left ventricular restriction.54 However, Terashima et al. reported an in-hospital mortality of 25%,54 thereby, at present, such a technique should be applied only to patients with LVFWR and high surgical risk due to poor general condition.

reports, using serial echocardiographic follow-up, have not shown development of left ventricular restriction.54 However, Terashima et al. reported an in-hospital mortality of 25%,54 thereby, at present, such a technique should be applied only to patients with LVFWR and high surgical risk due to poor general condition. Conservative management Conservative management, in patients with sub-acute LVFWR who have recovered from cardiac tamponade (with or without pericardiocentesis) and have a prohibitive surgical risk, or when emergency surgical repair is not available, has been described. Management includes maintenance of fluid infusion and inotropic support as needed, with early attempts at weaning within the ensuing 24 h. This is followed by institution of beta-blockade therapy and strict blood pressure control. Insertion of IABP, reducing left ventricular wall stress and intra-cavitary pressure, could limit infarct extension and avoid re-rupture. Prolonged bed rest has also been included in the conservative management to prevent arterial hypertension or hypertensive crisis, often considered the precipitating causes of re-rupture.57 Subjects who survive generally have small leaks that might close spontaneously by epicardial fibrin deposits. Obviously, this kind of treatment should be considered as unusual and non-standard. Blinc et al. conducted a retrospective cohort study of 107 patients who developed LVFWR; survival for the conservatively treated group was only 10%.58 By contrast, surgically treated patients had acceptable long-term survival (50%). So, surgical repair has to be considered undoubtedly the treatment of choice.

standard. Blinc et al. conducted a retrospective cohort study of 107 patients who developed LVFWR; survival for the conservatively treated group was only 10%.58 By contrast, surgically treated patients had acceptable long-term survival (50%). So, surgical repair has to be considered undoubtedly the treatment of choice. Conclusion Despite the first successful LVFWR repair dating back to the year 1972, this surgical procedure continues to be associated with high in-hospital mortality rates.6 Different techniques have been developed over the years for the management of LVFWR. Nevertheless, the optimal surgical treatment for this post-AMI mechanical complication remains controversial. Although the technical strategy varies, a basic principle that appears to remain unchanged is that surgeons should put the stitches, or fixed patches, in the healthy myocardial tissue. To date, sutured techniques should be considered the procedure of choice for surgical repair of LVFWR with active bleeding at the site of rupture. In the presence of oozing rupture, the most common operative finding, the sutureless technique represents a safe and effective alternative option, demonstrating satisfactory outcome results. This field still has room for cardiac surgeons to improve surgical strategies and techniques. Conflict of interest: RL is the Principal Investigator of PERSIST-AVR Trial sponsored by LivaNova and he is consultant for Medtronic. Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Introduction Mechanical circulatory support devices provide support to the heart and overall circulation. Several percutaneous support devices are available, including the Impella devices (Abiomed Inc., Massachusetts, USA).1 The Impella device is an axial pump placed across the aortic valve which retrieves blood from the left ventricle and expels it through a cannula into the ascending aorta. In patients with cardiogenic shock (CS), the aim of Impella treatment is to support the heart and failing circulation by increasing mean arterial pressure (MAP) and cardiac output. Moreover, the Impella unloads the left ventricle by volume unloading, and reduces left ventricular wall stress which reduces myocardial oxygen consumption and improves myocardial perfusion.2, 3 The Impella 2.5 and Impella CP can be inserted percutaneously and provide a maximum support of 2.5 and 3.7 L/minute, respectively. The larger Impella 5.0 can provide 5.0 L/minute but its introduction requires surgical cut-down of the femoral or axillary artery.4 These three types of Impella provide haemodynamic support to the left ventricle while the Impella RP provides circulatory support to the right ventricle.

.5 and 3.7 L/minute, respectively. The larger Impella 5.0 can provide 5.0 L/minute but its introduction requires surgical cut-down of the femoral or axillary artery.4 These three types of Impella provide haemodynamic support to the left ventricle while the Impella RP provides circulatory support to the right ventricle. The recently published 2017 European Society of Cardiology (ESC) guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation recommend that mechanical circulatory support may be considered in patients in whom therapy with vasopressor and inotropes is inadequate.5 However, this recommendation is based on only limited data of small randomised controlled trials.6, 7 Therefore, data on the real-life use of mechanical support devices in patients with CS after acute myocardial infarction, including outcomes, complication rates and predictors of adverse outcomes are important. The aim of this study is to report our outcomes and complications over 12 years of clinical experience with left ventricular Impella device, in patients with CS after acute myocardial infarction. The secondary aim is to identify predictors of 6-month mortality. Methods Patient population The data analysed in this study were obtained from patients who received an Impella at the Academic Medical Center in Amsterdam between October 2004 and December 2016. Our centre is a high volume tertiary referral hospital with on-site cardiac surgery.

The recently published 2017 European Society of Cardiology (ESC) guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation recommend that mechanical circulatory support may be considered in patients in whom therapy with vasopressor and inotropes is inadequate.5 However, this recommendation is based on only limited data of small randomised controlled trials.6, 7 Therefore, data on the real-life use of mechanical support devices in patients with CS after acute myocardial infarction, including outcomes, complication rates and predictors of adverse outcomes are important. The aim of this study is to report our outcomes and complications over 12 years of clinical experience with left ventricular Impella device, in patients with CS after acute myocardial infarction. The secondary aim is to identify predictors of 6-month mortality. Methods Patient population The data analysed in this study were obtained from patients who received an Impella at the Academic Medical Center in Amsterdam between October 2004 and December 2016. Our centre is a high volume tertiary referral hospital with on-site cardiac surgery. The Impella programme at our hospital started with the placement of the Impella 2.5 in patients undergoing elective high-risk percutaneous coronary intervention (PCI) with the aim of preventing haemodynamic compromise during complex PCI procedures.8–12 After gaining experience with the Impella in this controlled elective setting in 24 patients, we expanded Impella usage into the acute setting in four patients with a large anterior myocardial infarction without major haemodynamic compromise. Only thereafter did we start using Impella in patients with CS.13–16 More Impella devices have become available over the years. Initially, only the percutaneous Impella 2.5 and the Impella 5.0, which needs a surgical cut-down of the femoral artery, were available. After our first report on the outcomes of CS patients treated with an Impella 2.5 or Impella 5.0, we adhered to the strategy of either placing an Impella 5.0 immediately or, if not possible, initially to insert an Impella 2.5 and upgrade to an Impella 5.0 before the patient was transferred to the intensive care unit, with the aim of giving the patient more haemodynamic support than the 2.5 L/minute of the Impella 2.5.15 In 2012, the percutaneous Impella CP, which supports up to 3.7 L/minute, became available whereafter patients were routinely treated with the Impella CP. As we deemed the difference in support provided by the Impella CP and the Impella 5.0 to be approximately 1 L/minute, the need to upgrade the Impella CP to a surgical Impella 5.0 was infrequent. However, the actual support provided by the Impella devices is also determined by patient characteristics (e.g. native heart pulsatile flow and vascular resistance).Veno-arterial extracorporeal life support was not readily available in our institution during the inclusion period of this study. Also, our institution is not a durable left ventricular assist device (LVAD) or heart transplantation centre. However, these patients are generally not deemed candidates for this therapy during the acute phase of CS. The study was approved by the Academic Medical Center’s institutional review board and complies with the Declaration of Helsinki.

ot a durable left ventricular assist device (LVAD) or heart transplantation centre. However, these patients are generally not deemed candidates for this therapy during the acute phase of CS. The study was approved by the Academic Medical Center’s institutional review board and complies with the Declaration of Helsinki. Treatment Implantation of the Impella was according to instructions for use, and all operators received training prior to the start of the Impella programme. The use of the Impella device and the timing of the initiation of Impella therapy (before or after revascularisation) were left to the discretion of the treating cardiologist or cardiac surgeon.

pella was according to instructions for use, and all operators received training prior to the start of the Impella programme. The use of the Impella device and the timing of the initiation of Impella therapy (before or after revascularisation) were left to the discretion of the treating cardiologist or cardiac surgeon. After implantation, the Impella performance level was set to a maximum level without suction or position console alarms, usually P7–8. In case of positioning alarms, echocardiography was performed to verify the position of the Impella device. Duration of Impella support was at the discretion of the treating physician. Upgrade to an impella device with more haemodynamic support was considered when the device used was deemed to provide insufficient support. This was the case in patients who exhibited a combination of worsening haemodynamics and increased need for inotropes and vasopressors (dosages) despite high Impella performance, together with an overall assessment of the patient and his/her neurological status. During Impella support, all patients were treated with unfractionated heparin in order to maintain an activated clotting time level between 160 and 180 seconds. All patients were treated with heparin (5000 IU), and aspirin (500 mg) pre-PCI. Adjunctive treatment with glycoprotein IIb/IIIa inhibitors was at the discretion of the interventional cardiologists. Post-PCI dual antiplatelet therapy was prescribed in all patients in accordance with the guidelines.

160 and 180 seconds. All patients were treated with heparin (5000 IU), and aspirin (500 mg) pre-PCI. Adjunctive treatment with glycoprotein IIb/IIIa inhibitors was at the discretion of the interventional cardiologists. Post-PCI dual antiplatelet therapy was prescribed in all patients in accordance with the guidelines. Weaning was not formally protocolled, but was evaluated daily by the treating physician and typically started on signs of haemodynamic recovery, usually 12–24 hours after PCI, when inotropes and vasopressors were reduced. Weaning usually occurred in two steps: from maximum possible support (P7–8) to approximately half support (P4–5) (if necessary patients were observed for several hours, typically overnight), to low-level Impella support (P2–3) before device removal. Device removal was typically also two-staged. First the device is pulled back from the left ventricle into the descending aorta. The device is not switched off, but set to level P1 in order to prevent thrombus formation. After 45–60 minutes of heparin cessation, the device is removed, followed by approximately 30 minutes of femoral compression. Study population We included all consecutive patients presenting with CS in the setting of acute myocardial infarction who underwent Impella 2.5, Impella CP or Impella 5.0 therapy and were treated with primary PCI. Patients were excluded from analysis if they were referred to our hospital while already on Impella support or received Impella therapy after revascularisation with coronary artery bypass grafting.

cute myocardial infarction who underwent Impella 2.5, Impella CP or Impella 5.0 therapy and were treated with primary PCI. Patients were excluded from analysis if they were referred to our hospital while already on Impella support or received Impella therapy after revascularisation with coronary artery bypass grafting. Data source We retrieved baseline demographic variables, procedural and angiographic information that had been prospectively collected from our local electronic database. Data on complications were obtained by a dedicated researcher who performed a retrospective in-depth chart review for each individual patient, including the daily clinical course reports. Follow-up data were completed with information obtained from discharge letters and inpatient and outpatient charts from the hospitals or referring centres.

ere obtained by a dedicated researcher who performed a retrospective in-depth chart review for each individual patient, including the daily clinical course reports. Follow-up data were completed with information obtained from discharge letters and inpatient and outpatient charts from the hospitals or referring centres. Definitions CS was defined as a clinical diagnosis made by the treating physician, based on blood pressure criteria from the SHOCK trial, i.e. systolic blood pressure (SBP) of 90 mmHg or less for at least 30 minutes or the need for vasopressors to maintain a SBP greater than 90 mmHg.17 Survival was defined as survival within the hospital admission or up to 30 days, whichever was longer. A device-related vascular complication was defined as limb ischaemia requiring extraction of the device, an access site infection, or an access site-related bleed. Access site-related bleeding was subdivided into minor and major bleeding. Major bleeding was defined as bleeding associated with a serum haemoglobin level decrease of 3.1 mmol/L (5 g/dL), a bleed necessitating a minimum of two packed cells of blood product transfusion or the need for surgery to control the bleeding.16 Access site bleeds that were reported on the patient’s chart or the hospital discharge letter, but did not fit the definition of a major bleed, were noted as minor bleeds. Haemolysis was defined as clinically relevant haemolysis requiring extraction of the device or blood transfusion. Haemorrhagic or ischaemic stroke was confirmed by a neurologist and a concurring computed tomography scan. Renal insufficiency on admission was defined by means of the clinical threshold for impaired renal function (creatinine >95 µmol/L for women and >110 μmol/L for men). Low haemoglobin on admission was defined using the clinical threshold for anaemia (7.5 mmol/L for women and 8.5 mmol/L for men).

omputed tomography scan. Renal insufficiency on admission was defined by means of the clinical threshold for impaired renal function (creatinine >95 µmol/L for women and >110 μmol/L for men). Low haemoglobin on admission was defined using the clinical threshold for anaemia (7.5 mmol/L for women and 8.5 mmol/L for men). Analysis Normally distributed continuous variables are reported as mean ± standard deviation (SD) and compared with analysis of variance (ANOVA) corrected for multiple testing by Bonferroni. Skewed distributed variables are presented as median (25th–75th percentile) and compared with the Wilcoxon rank sum test. Categorical variables are presented as proportions and compared using chi-square tests. Kaplan–Meier analyses were calculated and a log-rank test was used to compare the clinical outcomes between groups. Patients were compared according to the outcome (survivors vs. non-survivors) and the type of Impella device the patients first received. Confidence intervals (95% CIs) were calculated based on the Pearson–Clopper method.

yses were calculated and a log-rank test was used to compare the clinical outcomes between groups. Patients were compared according to the outcome (survivors vs. non-survivors) and the type of Impella device the patients first received. Confidence intervals (95% CIs) were calculated based on the Pearson–Clopper method. Univariate Cox proportional hazard analyses were performed to identify predictors for 6-month mortality. The following parameters were included: age, laboratory values on admission (lactate, glucose, pH), haemodynamic variables on admission (MAP, SBP and heart rate (HR)), all continuous variables. Sex, low haemoglobin on admission, renal insufficiency on admission, cardiac arrest before Impella placement and traumatic injuries on admission were included as dichotomous variables. Variables that were significant in univariate analysis (P<0.10) were entered in a stepwise backward multivariate Cox regression analysis. A covariate was removed from the model if its significance level exceeded P=0.10. Analyses were performed using SPSS (version 23.0; Chicago, IL, USA).

luded as dichotomous variables. Variables that were significant in univariate analysis (P<0.10) were entered in a stepwise backward multivariate Cox regression analysis. A covariate was removed from the model if its significance level exceeded P=0.10. Analyses were performed using SPSS (version 23.0; Chicago, IL, USA). Results Patient population Between October 2004 and December 2016, a total of 250 patients received Impella 2.5, Impella CP or Impella 5.0 at our institution (Figure 1). A total of 112 patients with CS in the setting of acute myocardial infarction were treated with primary PCI and Impella. Baseline characteristics are summarised in Table 1. In the total population, patients were 60±10 years old and 80% were men. A total of 60% of patients had experienced a cardiac arrest before Impella placement. A total of 89% of patients were mechanically ventilated and 87% were treated with catecholamines or inotropes during primary PCI. The median ischaemic time was 153 minutes and 81% had an anterior myocardial infarction. Angiographic success was achieved in 98% of patients, defined as thrombolysis in myocardial infarction (TIMI) flow post-PCI of 2/3. Figure 1. Flow diagram of the patients treated with Impella therapy at the Academic Medical Center, Amsterdam. Table 1. Baseline characteristics of patients with Impella support for acute myocardial infarction. All patients Survivors Non-survivors P value (n=112) (n=47) (n=65) Clinical characteristics and risk factors

Results Patient population Between October 2004 and December 2016, a total of 250 patients received Impella 2.5, Impella CP or Impella 5.0 at our institution (Figure 1). A total of 112 patients with CS in the setting of acute myocardial infarction were treated with primary PCI and Impella. Baseline characteristics are summarised in Table 1. In the total population, patients were 60±10 years old and 80% were men. A total of 60% of patients had experienced a cardiac arrest before Impella placement. A total of 89% of patients were mechanically ventilated and 87% were treated with catecholamines or inotropes during primary PCI. The median ischaemic time was 153 minutes and 81% had an anterior myocardial infarction. Angiographic success was achieved in 98% of patients, defined as thrombolysis in myocardial infarction (TIMI) flow post-PCI of 2/3. Figure 1. Flow diagram of the patients treated with Impella therapy at the Academic Medical Center, Amsterdam. Table 1. Baseline characteristics of patients with Impella support for acute myocardial infarction. All patients Survivors Non-survivors P value (n=112) (n=47) (n=65) Clinical characteristics and risk factors Age (years) 60.1 ± 10.6 59.3 ± 9.5 60.7 ± 11.4 0.503 Male sex, n (%) 90 (80.4) 40 (85.1) 50 (76.9) 0.282 Body mass index (kg/m2) 26.0 (23.7–27.8) 26.0 (23.4–27.8) 25.8 (24.2–27.8) 0.850 Cardiovascular risk factors, n (%) Current smoking 41 (42.7) 21 (48.8) 19 (35.8) 0.131 Hypertension 38 (35.2) 14 (29.8) 24 (39.3) 0.302 Hypercholesterolemia 15 (14.2) 6 (12.8) 9 (15.3) 0.715 Diabetes mellitus 17 (15.3) 6 (12.8) 11 (17.2) 0.070 Prior myocardial infarction, n (%) 17 (15.7) 4 (8.5) 13 (21.3) 0.070 Prior TIA or stroke, n (%) 4 (3.7) 2 (4.3) 2 (3.2) 0.777 Known peripheral arterial disease, n (%) 6 (5.7) 1 (2.1) 5 (8.6) 0.154 Prior PCI or CABG, n (%) 15 (13.6) 5 (10.6) 10 (15.9) 0.429

5.3) 0.715 Diabetes mellitus 17 (15.3) 6 (12.8) 11 (17.2) 0.070 Prior myocardial infarction, n (%) 17 (15.7) 4 (8.5) 13 (21.3) 0.070 Prior TIA or stroke, n (%) 4 (3.7) 2 (4.3) 2 (3.2) 0.777 Known peripheral arterial disease, n (%) 6 (5.7) 1 (2.1) 5 (8.6) 0.154 Prior PCI or CABG, n (%) 15 (13.6) 5 (10.6) 10 (15.9) 0.429 Clinical characteristics on admission Cardiac arrest, n (%) 67 (59.8) 28 (59.6) 39 (60.0) 0.998 Out of hospital cardiac arrest, n (%) 49 (74.2) 22 (78.6) 27 (71.1) 0.490 Witnessed arrest, n (%) 58 (90.6) 27 (96.4) 31 (86.1) 0.160 First rhythm VT/VF/AED, n (%) 56 (86.2) 26 (92.9) 30 (81.1) 0.173 Time till return of spontaneous circulation (min) 21 (11–50) 16 (10–25) 30 (19–54) 0.025 Traumatic injuries at admission, n (%) 7 (6.3) 3 (6.4) 4 (6.2) 0.961 Primary percutaneous coronary intervention

Cardiac arrest, n (%) 67 (59.8) 28 (59.6) 39 (60.0) 0.998 Out of hospital cardiac arrest, n (%) 49 (74.2) 22 (78.6) 27 (71.1) 0.490 Witnessed arrest, n (%) 58 (90.6) 27 (96.4) 31 (86.1) 0.160 First rhythm VT/VF/AED, n (%) 56 (86.2) 26 (92.9) 30 (81.1) 0.173 Time till return of spontaneous circulation (min) 21 (11–50) 16 (10–25) 30 (19–54) 0.025 Traumatic injuries at admission, n (%) 7 (6.3) 3 (6.4) 4 (6.2) 0.961 Primary percutaneous coronary intervention Ischaemic time (min) 153 (107–240) 140 (95–266) 161 (119–232) 0.517 Infarct-related artery, n (%) 0.879 Left main 29 (25.9) 14 (29.8) 15 (23.1) Left anterior descending 62 (55.4) 25 (53.2) 37 (56.9) Left circumflex 13 (11.6) 5 (10.2) 8 (12.3) Right coronary artery 8 (7.1) 3 (6.4) 5 (7.7) Multi-vessel disease, n (%)a 74 (66.1) 28 (59.6) 46 (70.8) 0.217 Mechanical complications, n (%) 3 (2.7) 0 (0) 3 (4.6) 0.135 TIMI flow 0/1 pre-PCI, n (%) 90 (81.8) 35 (77.8) 55 (84.6) 0.361 TIMI flow 2/3 post-PCI, n (%) 101 (91.0) 43 (93.5) 58 (89.2) 0.441 Cardiogenic shock during primary PCI 103 (92.0) 44 (93.6) 59 (90.8) 0.584 Catecholamines or inotropes, n (%) 94 (83.9) 37 (78.7) 57 (87.7) 0.202 Mechanical ventilation, n (%) 98 (87.5) 41 (87.2) 57 (87.7) 0.942 Primary PCI in other hospital 9 (8.0) 5 (10.6) 4 (6.2) 0.389 Before device placement

Ischaemic time (min) 153 (107–240) 140 (95–266) 161 (119–232) 0.517 Infarct-related artery, n (%) 0.879 Left main 29 (25.9) 14 (29.8) 15 (23.1) Left anterior descending 62 (55.4) 25 (53.2) 37 (56.9) Left circumflex 13 (11.6) 5 (10.2) 8 (12.3) Right coronary artery 8 (7.1) 3 (6.4) 5 (7.7) Multi-vessel disease, n (%)a 74 (66.1) 28 (59.6) 46 (70.8) 0.217 Mechanical complications, n (%) 3 (2.7) 0 (0) 3 (4.6) 0.135 TIMI flow 0/1 pre-PCI, n (%) 90 (81.8) 35 (77.8) 55 (84.6) 0.361 TIMI flow 2/3 post-PCI, n (%) 101 (91.0) 43 (93.5) 58 (89.2) 0.441 Cardiogenic shock during primary PCI 103 (92.0) 44 (93.6) 59 (90.8) 0.584 Catecholamines or inotropes, n (%) 94 (83.9) 37 (78.7) 57 (87.7) 0.202 Mechanical ventilation, n (%) 98 (87.5) 41 (87.2) 57 (87.7) 0.942 Primary PCI in other hospital 9 (8.0) 5 (10.6) 4 (6.2) 0.389 Before device placement Catecholamines or inotropes, n (%) 102 (91.1) 41 (87.2) 61 (93.8) 0.226 Mechanical ventilation, n (%) 100 (89.3) 42 (89.4) 58 (89.2) 0.982 Intra-aortic balloon pump before Impella placement, n (%) 22 (19.6) 6 (12.8) 16 (24.6) 0.119 Blood pressure values Mean arterial pressure (mmHg) 67 (56–77) 68 (57–80) 66 (52–76) 0.289 Systolic blood pressure (mmHg) 86 (73–102) 89 (79–104) 83 (70–100) 0.202 Diastolic blood pressure (mmHg) 58 (44–65) 60 (48–66) 56 (40–65) 0.215 Heart rate (beats per minute) 96 (78–113) 95 (75–108) 97 (80–115) 0.274 Blood values Lactate (mmol/L) 6.2 (3.6–9.7) 4.2 (2.2–8.1) 7.6 (4.1–10.9) 0.012 Haemoglobin (mmol/L) 8.4 (7.5–9.4) 8.8 (7.4–9.5) 8.3 (7.5–9.1) 0.285 Creatinine (µmol/L) 114 (90–136) 104 (87–129) 123 (95–140) 0.080 Glucose (mmol/L) 13.4 (9.8–18.3) 11.5 (8.9–17.0) 14.1 (11.7–20.6) 0.028 Arterial pH 7.21 (7.07–7.30) 7.26 (7.17–7.35) 7.14 (6.94–7.25) 0.002 Data are displayed as count (percentage), mean ± standard deviation or median (25th percentile to 75th percentile).

reatinine (µmol/L) 114 (90–136) 104 (87–129) 123 (95–140) 0.080 Glucose (mmol/L) 13.4 (9.8–18.3) 11.5 (8.9–17.0) 14.1 (11.7–20.6) 0.028 Arterial pH 7.21 (7.07–7.30) 7.26 (7.17–7.35) 7.14 (6.94–7.25) 0.002 Data are displayed as count (percentage), mean ± standard deviation or median (25th percentile to 75th percentile). P value for the comparison between survivors versus non-survivors. TIA: transient ischaemic attack; PCI: percutaneous coronary intervention; CABG: coronary artery bypass grafting; VT: ventricular tachycardia; VF: ventricular fibrillation; AED: automated external defibrillator;TIMI: thrombolysis in myocardial infarction. a >50% stenosis in non-culprit vessel.

ittee of the Philipps University of Marburg, which waived the need for written informed consent, as renal Doppler ultrasonography is an existing feature of our clinical practice and the augmentation of Impella flow level was performed in stable patients without any alterations in the systematic haemodynamic parameters. Statistical analysis Data are presented as absolute variables and percentages (%) for categorical variables and either median with interquartile range (IQR: 25th–75th percentile) or mean with standard deviation according to the distribution of the variables. We assessed normality using the Shapiro–Wilk test as well as Pearson tests. After testing for normal distribution, Student’s t-test or Mann–Whitney test was implemented to test for differences between the various characteristics. Intraobserver variability was calculated based on the ICC and its 95% CI.

P value for the comparison between survivors versus non-survivors. TIA: transient ischaemic attack; PCI: percutaneous coronary intervention; CABG: coronary artery bypass grafting; VT: ventricular tachycardia; VF: ventricular fibrillation; AED: automated external defibrillator;TIMI: thrombolysis in myocardial infarction. a >50% stenosis in non-culprit vessel. Clinical course Characteristics of the clinical course are summarised in Table 2. The initial Impella strategy consisted of Impella 2.5 in 40 patients (35.7%), Impella CP in 52 patients (46.4%) and Impella 5.0 in 20 patients (17.9%). The Impella device was placed before primary PCI in 18.8% of the patients. In 58%, the Impella was placed directly after the primary PCI (during the initial procedure), and in 23.2% of patients the Impella was placed in a separate procedure (after having left the cardiac catheterisation laboratory). The median Impella support time was 53 hours. A total of 12 patients (10.7%) underwent an upgrade to a higher-flow support device (Impella 5.0 or veno-arterial extracorporeal membrane oxygenation (ECMO)). One patient received a durable LVAD after Impella and ECMO treatment. The majority of the patients were treated with inotropic or vasopressor agents (95%), mechanical ventilation (95%), and were admitted to the intensive care unit (ICU) (89%) (Table 3). Renal replacement therapy was necessary in 38% of patients and 59% required blood products. Table 2. Clinical course of patients with cardiogenic shock after acute myocardial infarction. All patients Survivors Non-survivors P value (n=112) (n=47) (n=65)

Clinical course Characteristics of the clinical course are summarised in Table 2. The initial Impella strategy consisted of Impella 2.5 in 40 patients (35.7%), Impella CP in 52 patients (46.4%) and Impella 5.0 in 20 patients (17.9%). The Impella device was placed before primary PCI in 18.8% of the patients. In 58%, the Impella was placed directly after the primary PCI (during the initial procedure), and in 23.2% of patients the Impella was placed in a separate procedure (after having left the cardiac catheterisation laboratory). The median Impella support time was 53 hours. A total of 12 patients (10.7%) underwent an upgrade to a higher-flow support device (Impella 5.0 or veno-arterial extracorporeal membrane oxygenation (ECMO)). One patient received a durable LVAD after Impella and ECMO treatment. The majority of the patients were treated with inotropic or vasopressor agents (95%), mechanical ventilation (95%), and were admitted to the intensive care unit (ICU) (89%) (Table 3). Renal replacement therapy was necessary in 38% of patients and 59% required blood products. Table 2. Clinical course of patients with cardiogenic shock after acute myocardial infarction. All patients Survivors Non-survivors P value (n=112) (n=47) (n=65) Mechanical circulatory support

Clinical course Characteristics of the clinical course are summarised in Table 2. The initial Impella strategy consisted of Impella 2.5 in 40 patients (35.7%), Impella CP in 52 patients (46.4%) and Impella 5.0 in 20 patients (17.9%). The Impella device was placed before primary PCI in 18.8% of the patients. In 58%, the Impella was placed directly after the primary PCI (during the initial procedure), and in 23.2% of patients the Impella was placed in a separate procedure (after having left the cardiac catheterisation laboratory). The median Impella support time was 53 hours. A total of 12 patients (10.7%) underwent an upgrade to a higher-flow support device (Impella 5.0 or veno-arterial extracorporeal membrane oxygenation (ECMO)). One patient received a durable LVAD after Impella and ECMO treatment. The majority of the patients were treated with inotropic or vasopressor agents (95%), mechanical ventilation (95%), and were admitted to the intensive care unit (ICU) (89%) (Table 3). Renal replacement therapy was necessary in 38% of patients and 59% required blood products. Table 2. Clinical course of patients with cardiogenic shock after acute myocardial infarction. All patients Survivors Non-survivors P value (n=112) (n=47) (n=65) Mechanical circulatory support First Impella device 0.053 Impella 2.5 40 (35.7) 13 (27.7) 27 (41.5) Impella CP 52 (46.4) 21 (44.7) 31 (47.7) Impella 5.0 20 (17.9) 13 (27.7) 7 (10.8) Change of mechanical support device, n (%) 12 (10.7) 4 (8.5) 8 (12.3) 0.521 Upgrade to Impella 5.0 9 (75) 3 (75.0) 6 (75.0) Upgrade to ECMO 3 (25) 1 (25.0) 2 (25.0) Device replacement by similar device, n (%) 2 (1.8) 1 (2.1) 1 (1.5) 0.816 Time of device placement 0.546 Impella placement before primary PCI, n (%) 21 (18.8) 11 (23.4) 10 (15.4) Impella placement directly after primary PCI, n (%) 67 (59.8) 26 (55.3) 41 (63.1) Impella placement in separate procedure after primary PCI, n (%) 24 (21.4) 10 (21.3) 14 (21.5) Time between revascularisation and Impella placement (hours) 13 (8–23) 13 (10–29) 14 (7–20) 0.752 IABP between primary PCI and Impella placement, n (%) 10 (41.7) 5 (50.0) 5 (35.7) 0.484 Duration of Impella support (hours)a 52 (22 – 122) 80 (51–150) 36 (12–72) <0.001 Device failure requiring extraction of the device, n (%) 1 (0.9) 1 (2.1) 0 (0) 0.237

and Impella placement (hours) 13 (8–23) 13 (10–29) 14 (7–20) 0.752 IABP between primary PCI and Impella placement, n (%) 10 (41.7) 5 (50.0) 5 (35.7) 0.484 Duration of Impella support (hours)a 52 (22 – 122) 80 (51–150) 36 (12–72) <0.001 Device failure requiring extraction of the device, n (%) 1 (0.9) 1 (2.1) 0 (0) 0.237 During admission Inotropic or vasopressor therapy, n (%) 106 (94.6) 42 (89.4) 64 (98.5) 0.035 Renal replacement therapy, n (%) 43 (38.4) 19 (40.4) 24 (36.9) 0.707 Mechanical ventilation, n (%) 106 (94.6) 43 (91.5) 63 (96.9) 0.208 Peak CKMB (μmol/L) 457 (184 – 934) 354 (120–781) 623 (251–1029) 0.051 Blood products, n (%) 68 (60.7) 29 (61.7) 39 (60.0) 0.856 Number of patients in the intensive care unit, n (%) 100 (89.3) 42 (89.4) 58 (89.2) 0.982 Days on the intensive care unit 5 (3–15) 12 (7–25) 3 (2–7) <0.001 P value for the comparison between survivors versus non-survivors. a Sum of support duration of all given support devices, including upgrades. ECMO: extracorporeal membrane oxygenation; PCI: percutaneous coronary intervention; IABP: intra-aortic balloon pump; CKMB: creatine kinase myocardial type. Table 3. Clinical outcome for patients with cardiogenic shock after acute myocardial infarction. All patients Survivors Non-survivors P value (n=112) (n=47) (n=65) In-hospital outcome

the reference standard. Studies were excluded if they missed diagnostic accuracy data relevant to our research questions or if there were insufficient data for the derivation of 2×2 contingency tables. The screening process was performed with the reference manager Endnote X7 using the method proposed by Bramer et al.8 Methodological quality assessment The methodological quality of the included articles was independently assessed by two investigators (MA and EAD) using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool.9 In case of disagreement, consensus was reached by joint reading or by consulting a third investigator (AD). Data extraction and statistical analysis Using a standardised data extraction form, one investigator (MA) extracted relevant details concerning the study design (e.g. study population, inclusion period, target condition and reference standard), the patient characteristics and study results relevant to our research questions. The extracted data were then verified by two investigators (EAD and AD).

ECMO: extracorporeal membrane oxygenation; PCI: percutaneous coronary intervention; IABP: intra-aortic balloon pump; CKMB: creatine kinase myocardial type. Table 3. Clinical outcome for patients with cardiogenic shock after acute myocardial infarction. All patients Survivors Non-survivors P value (n=112) (n=47) (n=65) In-hospital outcome In-hospital mortality, n (%) 65 (58.0% CI 48.3–67.3) 0 (0% CI 0.0–7.5) 65 (100% CI 94.5–100) Refractory cardiogenic shock 44 (67.7) – 44 (67.7) Post-anoxic brain injury 13 (20.0) – 13 (20.0) Other reason 8 (12.3) – 8 (12.3) Stroke, n (%) 4 (3.6% CI 1.0–8.9) 0 (0% CI 0.0–7.5) 4 (6.2% CI 1.7–15.0) 0.083 Haemorrhagic stroke 1 (25.0) 0 (0) 1 (25.0) Ischaemic stroke 3 (75.0) 0 (0) 3 (75.0) Device-related vascular complication, n (%) 19 (17.0% CI 10.5–25.2) 8 (17.0% CI 7.6–30.8) 11 (16.9% CI 8.8–28.3) 0.989 Limb ischaemia 4 (21.1) 3 (37.5) 1 (9.1) Access site-related bleeding 14 (73.7) 4 (50.0) 10 (90.9) Major bleeding 11 (78.6) 3 (75.0) 8 (80.0) Minor bleeding 3 (21.4) 1 (25.0) 2 (20.0) Access site infection 1 (5.3) 1 (12.5) 0 (0) Non-device-related bleeding 14 (12.5% CI 7.0–20.1) 7 (14.9% CI 6.2–28.3) 7 (10.8% CI 4.4–20.9) 0.280 Gastrointestinal bleeding 6 (42.9) 4 (57.1) 2 (28.6) Other location 8 (57.1) 4 (42.9) 5 (71.4) Clinically relevant haemolysis, n (%) 8 (7.1% CI 3.1–13.6) 6 (12.8% CI 4.8–25.7) 2 (3.1% CI 0.4–10.7) 0.049 Surgical LVAD placement, n (%) 1 (0.9% CI 0.0–4.9) 1 (2.1% CI 0.1–11.3) 0 (0% CI 0.0–5.5) 0.237 Heart transplantation, n (%) 0 (0% CI 0.0–3.2) 0 (0% CI 0.0–7.5) 0 (0% CI 0.0–5.5) – P value for the comparison between survivors versus non-survivors.

%) 8 (7.1% CI 3.1–13.6) 6 (12.8% CI 4.8–25.7) 2 (3.1% CI 0.4–10.7) 0.049 Surgical LVAD placement, n (%) 1 (0.9% CI 0.0–4.9) 1 (2.1% CI 0.1–11.3) 0 (0% CI 0.0–5.5) 0.237 Heart transplantation, n (%) 0 (0% CI 0.0–3.2) 0 (0% CI 0.0–7.5) 0 (0% CI 0.0–5.5) – P value for the comparison between survivors versus non-survivors. LVAD: left ventricular assist device; CI: confidence interval was calculated based on the Pearson–Clopper method. Complications and adverse outcome The clinical outcomes for patients with CS after acute myocardial infarction are summarised in Table 3. A total of 65 patients (58%) died during the admission. The cause of death was refractory CS (67.7%), post-anoxic brain injury (20.0%), or other reasons (12.3%). Four patients were diagnosed with stroke during admission (3.6%). Device-related vascular complications occurred in 19 patients (17%), of whom 14 had an access site-related bleed (11 major and three minor bleeds), four patients had limb ischaemia requiring surgery and one patient had an access site infection requiring surgery. Clinically relevant haemolysis occurred in 7.1% of patients. Non-device-related bleeding occurred in 14 patients (12.5%).

7%), of whom 14 had an access site-related bleed (11 major and three minor bleeds), four patients had limb ischaemia requiring surgery and one patient had an access site infection requiring surgery. Clinically relevant haemolysis occurred in 7.1% of patients. Non-device-related bleeding occurred in 14 patients (12.5%). Survivors versus non-survivors Non-survivors had lower pH levels (7.14 (6.94–7.25) vs. 7.26 (7.17–7.35), P=0.002), and higher lactate (7.6 (4.1–10.9) vs. 4.2 (2.2–8.1) mmol/L, P=0.012) and glucose levels (14.1 (11.7–20.6) vs. 11.5 (8.9–17.0) mmol/L, P=0.028) before device placement (Table 1). However, non-survivors did not differ in age (60.7±11.4 vs. 59.3±9.5, P=0.503), MAP (66 (52–76) vs. 68 (57–80), P=0.289), cardiac arrest (60.0% vs. 59.6%, P=0.998), ischaemic time (161 (119–232) vs. 140 (95–266), P=0.517), mechanical ventilation (89.2% vs. 89.4, P=0.982). Survivors required less inotropic or vasopressor therapy during admission (98.5% vs. 89.4%, P=0.035), a longer duration of Impella support (80 (51–150) vs. 36 (12–72) hours, P<0.001), and had a longer length of ICU stay (12 (7–25) vs. 3 (2–7) days, P<0.001) (Table 2). All strokes occurred in the non-survivors (6.2% vs. 0%, P=0.083) (Table 3). There were no differences in device-related vascular complications (17.0% in survivors vs. 16.9% in the non-survivors, P=0.989), non-device-related bleeding (14.9% vs. 10.8%, P=0.280), but higher rates of clinically relevant haemolysis in the survivors (12.8% vs. 3.1%, P=0.049).

vivors (6.2% vs. 0%, P=0.083) (Table 3). There were no differences in device-related vascular complications (17.0% in survivors vs. 16.9% in the non-survivors, P=0.989), non-device-related bleeding (14.9% vs. 10.8%, P=0.280), but higher rates of clinically relevant haemolysis in the survivors (12.8% vs. 3.1%, P=0.049). 6-Month mortality The mortality at 6 months follow-up was 60.7%. There was no difference within age groups of 10 years, or tertiles of peak creatine kinase myocardial type, lactate, MAP and HR (Figure 3). There was a higher mortality when patients had lower pH levels or higher glucose levels before Impella insertion. Placement of the Impella device before revascularisation compared with directly after the revascularisation did not show a significant difference in 6-month mortality (52.4% vs. 64.2%, HR 1.45, 95% CI 0.75–2.81, P=0.273) (Table 4). The type of Impella device was not associated with a significant difference in mortality. In a Cox univariate analysis, lactate, glucose, pH and renal insufficiency before Impella insertion were predictors of 6-month mortality (Table 5). In a multivariate Cox regression analysis, only pH before Impella insertion was a predictor of 6-month mortality. Table 4. Mortality at 6 months according to Impella device, time of Impella placement, sex, cardiac arrest, traumatic injuries, renal impairment and haemoglobin on admission. n 6-Month mortality Hazard ratio (95% CI) P value Impella device Impella 2.5 40 70.0 Reference – Impella CP 52 61.5 0.84 (0.51–1.39) 0.838 Impella 5.0 20 40.0 0.46 (0.21–1.00) 0.455 Timing of Impella placement

Table 4. Mortality at 6 months according to Impella device, time of Impella placement, sex, cardiac arrest, traumatic injuries, renal impairment and haemoglobin on admission. n 6-Month mortality Hazard ratio (95% CI) P value Impella device Impella 2.5 40 70.0 Reference – Impella CP 52 61.5 0.84 (0.51–1.39) 0.838 Impella 5.0 20 40.0 0.46 (0.21–1.00) 0.455 Timing of Impella placement Before revascularisation 21 52.4 Reference – Directly after revascularisation 67 64.2 1.45 (0.75–2.81) 0.273 Delayed (in separate procedure) 24 58.3 1.31 (0.59–2.88) 0.510 Sex Male 90 57.8 Reference – Female 22 72.7 1.56 (0.89–2.73) 0.123 Cardiac arrest Yes 67 59.7 Reference – No 45 62.2 1.03 (0.63–1.67) 0.912 Traumatic injuries before admission Absent 105 61.0 Reference – Present 7 57.1 0.875 (0.32–2.40) 0.796 Renal impairment Creatinine lower than normal reference value 46 50.0 Reference – Creatinine higher than normal reference value 57 70.2 1.68 (1.01–2.82) 0.046 Haemoglobin Higher than normal reference value 53 54.7 Reference – Lower than normal reference value 49 65.3 1.30 (0.78–2.14) 0.312 CI: confidence interval. Table 5. Univariate and multivariate Cox regression of parameters on the association with mortality at 6-month follow-up. Parameter n Univariate analysis Multivariate analysis HR 95% CI P value HR 95% CI P value Age 112 1.015 0.99–1.04 0.224 – – – Male sex 112 1.556 0.89–2.73 0.123 – – – Lactate (mmol/L) 81 1.071 1.01–1.14 0.021 – – – Glucose (mmol/L) 99 1.046 1.01–1.09 0.035 – – – pH 99 0.087 0.02–0.34 <0.001 0.087 0.02–0.34 <0.001

Table 5. Univariate and multivariate Cox regression of parameters on the association with mortality at 6-month follow-up. Parameter n Univariate analysis Multivariate analysis HR 95% CI P value HR 95% CI P value Age 112 1.015 0.99–1.04 0.224 – – – Male sex 112 1.556 0.89–2.73 0.123 – – – Lactate (mmol/L) 81 1.071 1.01–1.14 0.021 – – – Glucose (mmol/L) 99 1.046 1.01–1.09 0.035 – – – pH 99 0.087 0.02–0.34 <0.001 0.087 0.02–0.34 <0.001 Low haemoglobin on admission 102 1.296 0.78–2.14 0.155 – – – Renal insufficiency on admission 103 1.688 1.01–2.82 0.046 – – – MAP before Impella placement 107 0.996 0.98–1.01 0.531 – – – SBP before Impella placement 108 0.998 0.99–1.01 0.675 – – – HR before Impella placement 105 1.073 0.79–1.01 0.650 – – – Cardiac arrest 112 1.028 0.63–1.67 0.912 – – – Traumatic injury on admission 112 0.947 0.34–2.61 0.916 – – – MAP: mean arterial blood pressure (mmHg); SBP: systolic blood pressure (mmHg); HR: heart rate (beats/min).

la placement 108 0.998 0.99–1.01 0.675 – – – HR before Impella placement 105 1.073 0.79–1.01 0.650 – – – Cardiac arrest 112 1.028 0.63–1.67 0.912 – – – Traumatic injury on admission 112 0.947 0.34–2.61 0.916 – – – MAP: mean arterial blood pressure (mmHg); SBP: systolic blood pressure (mmHg); HR: heart rate (beats/min). Differences between Impella devices The initial Impella strategy consisted of Impella 2.5 in 40 patients (35.7%), Impella CP in 52 patients (46.4%) and Impella 5.0 in 20 patients (17.9%) (Supplementary Table 1). There were some differences in the baseline characteristics of patients with Impella 2.5, CP and 5.0 (Supplementary Table 2). Patients treated with Impella 2.5 experienced an out-of-hospital cardiac arrest (OHCA) less frequently. In the Impella 5.0 group, biochemical values at admission were compatible with a less severe state of CS, although there was no difference in MAP. Also, patients who had undergone primary PCI at another centre more often received an Impella 5.0 on arrival at our institution (Supplementary Table 3). Impella 5.0 was more frequently placed at a separate procedure and not at primary PCI. There were differences in the number of patients who were upgraded to another support device, the number of patients receiving blood products and the number of days on the ICU. There was no difference in stroke, device-related vascular complications or haemolysis between patients treated with Impella 2.5, CP or 5.0 (Supplementary Table 4).

s in the number of patients who were upgraded to another support device, the number of patients receiving blood products and the number of days on the ICU. There was no difference in stroke, device-related vascular complications or haemolysis between patients treated with Impella 2.5, CP or 5.0 (Supplementary Table 4). Discussion This analysis describes the largest single-centre experience with Impella technology in CS over 12 years. It provides an insight into the treatment strategy, outcomes and complications of patients treated with Impella at an experienced centre. We describe an overall 30-day mortality of 56.2% (Figure 2), of which 67.7% was due to refractory CS. Complications consisted of device-related vascular complications (17.0%), non-device-related bleeding (12.5%), haemolysis (7.1%) and stroke (3.6%). In a multivariate analysis, pH before Impella placement is a predictor of 6-month mortality. Figure 2. Kaplan–Meier curve for patients treated with Impella for cardiogenic shock after acute myocardial infarction. Figure 3. Mortality at 6 months according to age, peak creatine kinase myocardial type (CKMB), pH, glucose, lactate, mean arterial blood pressure (MAP) and heart rate (HR) before Impella placement and time to return of spontaneous circulation (ROSC). Glucose, pH, lactate, MAP, HR and peak CKMB were dichotomised by dividing them into tertiles. Age was dichotomised per 10 years of age and time to ROSC by 10 minutes. Comparison between groups was made by Pearson chi-square analysis.

Figure 3. Mortality at 6 months according to age, peak creatine kinase myocardial type (CKMB), pH, glucose, lactate, mean arterial blood pressure (MAP) and heart rate (HR) before Impella placement and time to return of spontaneous circulation (ROSC). Glucose, pH, lactate, MAP, HR and peak CKMB were dichotomised by dividing them into tertiles. Age was dichotomised per 10 years of age and time to ROSC by 10 minutes. Comparison between groups was made by Pearson chi-square analysis. Impella has been on the market since 2004 but with little (randomised) evidence on its effectiveness in CS. Three small and underpowered randomised trials compare the Impella with intra-aortic balloon pump (IABP) in CS, and although the Impella can provide more haemodynamic support than an IABP, this was not translated into reduced mortality in randomised trials.6, 16, 18, 19 However, comparing the outcomes is hampered by the fact that a large percentage of the randomly assigned patients had had a cardiac arrest before admission, resulting in a high percentage of neurological damage. This might have resulted in the treatment effect of Impella support being underestimated. Furthermore, one study was prematurely discontinued due to slow inclusion.18

at a large percentage of the randomly assigned patients had had a cardiac arrest before admission, resulting in a high percentage of neurological damage. This might have resulted in the treatment effect of Impella support being underestimated. Furthermore, one study was prematurely discontinued due to slow inclusion.18 Mortality rates in real-world cohorts are higher than in randomised controlled trials of mechanical circulatory support in CS patients.16, 19–21 Registries that describe the real-world usage of devices actually describe an unselected patient cohort.15, 22–31 Randomised controlled trials are particularly difficult to conduct in critically ill patients in an emergency situation. Also, severely ill patients with a very poor prognosis are often excluded from randomised studies. This is why registries are of interest in these severely ill patients, as they provide important hypothesis-generating rationales for future clinical trials. Furthermore, as data from randomised trials are sparse, real-world registries have an important role in reporting on complications and other safety outcomes.

ies. This is why registries are of interest in these severely ill patients, as they provide important hypothesis-generating rationales for future clinical trials. Furthermore, as data from randomised trials are sparse, real-world registries have an important role in reporting on complications and other safety outcomes. We described the largest single-centre cohort on Impella therapy in CS patients after myocardial infarction. Several multi-centre registries also reported on Impella therapy in this particular patient group, but only described the use of Impella 2.5 (O’Neill N=154; Lauten N=120) or Impella CP (Basir N=287). 24, 25, 32 It is also important to report results on outcomes and complications in patients treated with Impella CP, and especially Impella 5.0. Patient selection may result in more severely ill patients treated with an Impella CP or 5.0, and the Impella 5.0 requires surgical cut-down of the artery, which may lead to more complications. The largest Impella cohort to our knowledge (Basir; multi-centre N=287) described the use of Impella CP in CS patients after myocardial infarction.32 However, they only reported the rate of vascular complications requiring surgical repair, and did not provide complication rates on other important complications such as leg ischaemia, haemolysis or the need for renal replacement therapy. Other studies that described the outcomes and complications of Impella CP or 5.0 therapy consisted of a small number of patients, a mixed population, or both.15, 22, 27–29

and did not provide complication rates on other important complications such as leg ischaemia, haemolysis or the need for renal replacement therapy. Other studies that described the outcomes and complications of Impella CP or 5.0 therapy consisted of a small number of patients, a mixed population, or both.15, 22, 27–29 We describe an in-hospital mortality rate of 58%, with a high percentage of patients having experienced cardiac arrest before Impella treatment (59.8%). Comparable mortality rates are described in other registries: Basir et al.32 describe a 56% in-hospital mortality rate in patients with acute myocardial infarction but with 40% of patients with cardiac arrest, and Lackermair et al.22 describe a 30-day mortality rate of 64% in a mixed patient cohort. Although the aim of Impella therapy is to provide systemic haemodynamic support, the majority of the patients may still die from refractory CS (67%). Refractory shock is a complex disease in which haemodynamic support only is unlikely to be the complete answer, especially once the inflammatory shock reaction emerges and multiple organ failure becomes severe. The Impella strategy at our hospital has changed over time. The Impella 2.5 was the standard therapy in patients with CS, until the Impella CP became available in 2012. The Impella CP requires a slightly larger insertion sheath than the Impella 2.5 (14 F versus the 13 F), but provides more flow (3.7 L/minute vs. 2.5 L/minute).

le scholar, without a restriction on the publication date. Only articles in English were included. The strategies for the literature search are provided in the Supplementary Material Table S1. Reference lists of relevant papers including systematic reviews were hand searched for potentially relevant additional studies. Study inclusion Titles and abstracts were independently screened by two investigators (MA and EAD) and selected for further evaluation if they met the following criteria: (a) the publication was a prospective cohort study published in a peer-reviewed journal; (b) patients were adults; (c) patients presented to the ED with symptoms suggestive of an acute coronary syndrome; (d) the diagnostic accuracy of the Roche Elecsys hs-cTnT was evaluated; (e) the primary endpoint was an admission diagnosis of AMI based on the universal definition of AMI.7 Full-text articles were then retrieved and independently screened for inclusion by both investigators (MA and EAD). In case of disagreement, a consensus was reached by joint reading. There were no restrictions on the type of troponin assay used as part of the reference standard. Studies were excluded if they missed diagnostic accuracy data relevant to our research questions or if there were insufficient data for the derivation of 2×2 contingency tables. The screening process was performed with the reference manager Endnote X7 using the method proposed by Bramer et al.8

lla strategy at our hospital has changed over time. The Impella 2.5 was the standard therapy in patients with CS, until the Impella CP became available in 2012. The Impella CP requires a slightly larger insertion sheath than the Impella 2.5 (14 F versus the 13 F), but provides more flow (3.7 L/minute vs. 2.5 L/minute). There was a trend towards a lower mortality rate in patients treated with larger Impella devices (Impella 2.5 70.0%, Impella CP 61.5%, Impella 5.0 40.0% at 6 months). However, comparison of baseline characteristics between patients treated Impella 2.5 and patients treated with Impella CP shows a much more extensive use of Impella CP, resulting in the treatment of more severely ill patients, who may already have more severe neurological damage on admission. Despite this, the mortality rates of the patients treated with the Impella CP are numerically lower than the mortality rates in patients treated with Impella 2.5. Also, the Impella 5.0 is more often used when the Impella was inserted in a separate procedure, because of the need for surgical cut-down of the femoral or axillary artery in order to introduce a 21 F catheter. This delayed Impella placement induces patient selection bias, as the most severely ill patients will be treated with a percutaneous Impella at primary PCI because they may have been deemed too ill to wait for surgical cut-down. Patients admitted to the ICU without an Impella may be deemed to be less ill, and may either recover or deteriorate and require delayed mechanical support. Unfortunately, our sample size is too small to take all possible confounders into account in a multivariate analysis and therefore we cannot fully evaluate our hypothesis.

s admitted to the ICU without an Impella may be deemed to be less ill, and may either recover or deteriorate and require delayed mechanical support. Unfortunately, our sample size is too small to take all possible confounders into account in a multivariate analysis and therefore we cannot fully evaluate our hypothesis. The introduction of Impella 2.5 and CP requires 13 F and 14 F sheaths and therefore some vascular complications may be expected. In our cohort, device-related vascular complications occurred in 19 patients (17%), of which the majority had access site-related bleeding (n=14). Limb ischaemia occurred in four patients (3.6%). The largest Impella cohort that reports complications (n=154) describes 9.7% vascular complications requiring surgery, 3.9% limb ischaemia and 17.5% bleeding requiring transfusion.10 Other smaller cohorts report limb ischaemia of 25%,27 12%,22, 28 10%31 and 3%.26

(n=14). Limb ischaemia occurred in four patients (3.6%). The largest Impella cohort that reports complications (n=154) describes 9.7% vascular complications requiring surgery, 3.9% limb ischaemia and 17.5% bleeding requiring transfusion.10 Other smaller cohorts report limb ischaemia of 25%,27 12%,22, 28 10%31 and 3%.26 Access site-related bleeding occurred in 14 patients (12.5%), of which 11 were a major bleed. Non-device-related bleeding occurred in 14 patients (12.5%). During mechanical support, patients receive heparin in addition to standard dual antiplatelet therapy after PCI (aspirin and a P2Y12 receptor blocker), which, in combination with larger-bore sheaths, facilitates bleeding. In a registry of post-cardiac arrest patients (n=78), the bleeding rate was 26% and three CS registries (n=120, n=154, n=66) describe bleeding rates of 24%, 18% and 35%.24–26, 28 Jensen et al. describe groin bleeding in 13% and higher rates of minor (29%), moderate (19%) and severe (5%) bleeding.31 Haemolysis occurred in 7.1% of treated patients. Earlier reports describe haemolysis in 6.0%, 7.5% and 10.3% of patients treated with Impella 2.5.24–26 Our cohort describes a stroke rate of 3.6%, which is comparable with other cohorts (5%,28 1.9%,24 0%,26 1.7%).25

minor (29%), moderate (19%) and severe (5%) bleeding.31 Haemolysis occurred in 7.1% of treated patients. Earlier reports describe haemolysis in 6.0%, 7.5% and 10.3% of patients treated with Impella 2.5.24–26 Our cohort describes a stroke rate of 3.6%, which is comparable with other cohorts (5%,28 1.9%,24 0%,26 1.7%).25 Previous described cohorts suggest a favorable outcome in patients in whom the Impella devices is placed before the revascularisation.16, 23, 31, 33, 34 Impella placement before primary PCI may enable stable haemodynamics during the intervention. It may prevent deterioration during the procedure and when opening the occluded vessel. Several animal studies have shown that unloading the left ventricle before reperfusion reduces infarct size despite the longer ischaemic time.35–37 These studies demonstrate that the use of Impella before revascularisation activates the neurohormonal cascade associated with reperfusion injury. This results in a cardioprotective signalling cascade which limits myocardial damage. In our cohort, 21 of the patients with acute myocardial infarction received an Impella before revascularisation (19%), 60% received it directly after the revascularisation and 21% received the Impella in a separate procedure. Our registry did not show a difference in 6-month mortality (52.4% vs. 64.2% durable, P=0.273). The time of device placement was at the discretion of the operator and therefore might be biased by the severity of the patient’s condition.

y after the revascularisation and 21% received the Impella in a separate procedure. Our registry did not show a difference in 6-month mortality (52.4% vs. 64.2% durable, P=0.273). The time of device placement was at the discretion of the operator and therefore might be biased by the severity of the patient’s condition. Currently, there is no (randomised) evidence that either Impella or ECMO support is associated with improved clinical outcomes in CS patients after acute myocardial infarction. In our institution, Impella was the support device of choice, based on local availability and expertise. A total of four patients were transferred for ECMO (N=3) and LVAD placement (N=1). Although only a few patients required ECMO/LVAD/heart transplantation in another centre, it is possible that these techniques would have been deployed earlier if available in our centre. Analysis of the PROTECT II trial, comparing IABP with Impella 2.5 in the setting of high-risk PCI, suggests a learning curve associated with the introduction of the Impella.12 Our experience describes a stepwise introduction of the Impella in the setting of elective high-risk PCI, followed by the use of Impella in the emergency setting of CS and placement of Impella prior to emergency revascularisation. A stepwise introduction is important to allow the successful introduction of a new technology into the clinical setting.

wise introduction of the Impella in the setting of elective high-risk PCI, followed by the use of Impella in the emergency setting of CS and placement of Impella prior to emergency revascularisation. A stepwise introduction is important to allow the successful introduction of a new technology into the clinical setting. There are several limitations to consider. This is an observational study with its associated limitations. Selection bias might play a significant role in the selection of patients to receive an Impella, the time of Impella placement, the selection of the device and the course of treatment. Multivariate analysis was limited due to the sample size of our cohort and did not include all variables that influence mortality in the model, such as the timing of device placement and myocardial infarction location. In addition, there are many factors that might have influenced the results, such as experience of the device, change of therapy over time, improvement of general treatment of CS and ST-segment elevation myocardial infarction patients over time, and change in patient selection over time. Moreover, we performed two randomised controlled trials comparing Impella with IABP.16, 18 During the inclusion period of these trials, half the patients were randomly assigned to IABP and the type of Impella therapy was defined by the study protocol.

ction patients over time, and change in patient selection over time. Moreover, we performed two randomised controlled trials comparing Impella with IABP.16, 18 During the inclusion period of these trials, half the patients were randomly assigned to IABP and the type of Impella therapy was defined by the study protocol. Also, we could only retrieve complications that were noted in the patient records and only reported on haemolysis that led to device removal or transfusion. However, we are aware of the fact that important complications, such as bleeding or haemolysis, are not always captured well in the patient records. Therefore it is very likely that we underestimated the rate of some complications. This registry shows that in patients with CS due to acute myocardial infarction, mechanical circulatory support with Impella is feasible, although mortality and complication rates remain high. Future studies should focus on the selection of the patient population that may have the most benefit from this therapy. Supplemental Material Supplemental_Material – Supplemental material for Real-life use of left ventricular circulatory support with Impella in cardiogenic shock after acute myocardial infarction: 12 years AMC experience Click here for additional data file.

This registry shows that in patients with CS due to acute myocardial infarction, mechanical circulatory support with Impella is feasible, although mortality and complication rates remain high. Future studies should focus on the selection of the patient population that may have the most benefit from this therapy. Supplemental Material Supplemental_Material – Supplemental material for Real-life use of left ventricular circulatory support with Impella in cardiogenic shock after acute myocardial infarction: 12 years AMC experience Click here for additional data file. Supplemental material, Revised_supplementary_data_[04-06-18] for Real-life use of left ventricular circulatory support with Impella in cardiogenic shock after acute myocardial infarction: 12 years AMC experience by Dagmar M Ouweneel, Justin de Brabander, Mina Karami, Krischan D Sjauw, Annemarie E Engström, M Marije Vis, Joanna J Wykrzykowska, Marcel A Beijk, Karel T Koch, Jan Baan, Robbert J de Winter, Jan J Piek, Wim K Lagrand, Thomas GV Cherpanath, Antoine HG Driessen, Riccardo Cocchieri, Bas AJM de Mol, Jan GP Tijssen and José PS Henriques in European Heart Journal: Acute Cardiovascular Care Conflict of interest: The authors declare that there is no conflict of interest. Funding: This work was supported by the Amsterdam UMC, University of Amsterdam, The Netherlands.

Introduction In recent years, increasing number of hospitals worldwide have adopted the high-sensitivity cardiac troponin T (hs-cTnT) assay as the reference marker of myocardial necrosis. These new cardiac biomarkers have proven to be more sensitive and faster in detecting myocardial damage.1 These features are especially desirable in patients who present at the emergency department (ED) with non-differentiated acute chest pain, where a timely diagnosis is essential. Acute chest pain is the second most common reason for visits to the ED.2,3 Only 10–20% of these patients are eventually diagnosed with acute myocardial infarction (AMI). However, missing this diagnosis may have grave consequences.1,4 Various strategies have been suggested for safe and early discharge of patients based on serial or a single low value of hs-cTnT balanced with the clinical presentation. In conjunction with these rule-out strategies, a direct rule-in strategy has also been proposed for patients with highly abnormal baseline hs-cTnT values.5 The aim of this systematic review and meta-analysis was to determine (a) the ability of serial hs-cTnT measurements to correctly rule out AMI and (b) the ability of a single high baseline hs-cTnT measurement to correctly rule in AMI in patients presenting to the ED with acute chest pain. Methods We conducted a systematic review and meta-analysis of the literature in agreement with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.6

The aim of this systematic review and meta-analysis was to determine (a) the ability of serial hs-cTnT measurements to correctly rule out AMI and (b) the ability of a single high baseline hs-cTnT measurement to correctly rule in AMI in patients presenting to the ED with acute chest pain. Methods We conducted a systematic review and meta-analysis of the literature in agreement with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.6 Search strategy The following databases were searched on 12 April 2017: Embase, Medline Ovid, Cochrane CENTRAL, Web of Science and Google scholar, without a restriction on the publication date. Only articles in English were included. The strategies for the literature search are provided in the Supplementary Material Table S1. Reference lists of relevant papers including systematic reviews were hand searched for potentially relevant additional studies.

orm, one investigator (MA) extracted relevant details concerning the study design (e.g. study population, inclusion period, target condition and reference standard), the patient characteristics and study results relevant to our research questions. The extracted data were then verified by two investigators (EAD and AD). First, we were interested in the capability of serial hs-cTnT measurements to correctly rule out AMI. For studies reporting the diagnostic accuracy of serial measurements, the timing of the troponin measurements and the reported cut-offs were extracted. Subsequently, the extracted data were assessed on appropriateness for quantitative analysis. If appropriate, 2×2 contingency tables were constructed for the individual studies and thereafter the sensitivity was calculated with 95% confidence intervals (CIs). In the case of overlapping samples, only the publication with the largest cohort was included in the quantitative analysis. Second, we were interested in the capability of a single high baseline measurement to correctly rule in AMI. A single high baseline measurement was defined as a baseline hs-cTnT value> 50 ng/l. This cut-off was chosen because it resembles the cut-off point for direct rule-in recommended by the European Society of Cardiology (ESC).5 After constructing 2×2 contingency tables, the specificity for each study was calculated with 95% CI.

le high baseline measurement was defined as a baseline hs-cTnT value> 50 ng/l. This cut-off was chosen because it resembles the cut-off point for direct rule-in recommended by the European Society of Cardiology (ESC).5 After constructing 2×2 contingency tables, the specificity for each study was calculated with 95% CI. Statistical analysis Because of our dual research question, we were primarily interested in obtaining the summary estimate of sensitivity of serial hs-cTnT measurements to rule out AMI, and the summary estimate of specificity of a single high hs-cTnT value to rule in AMI. In addition, we also calculated the other parameters of diagnostic accuracy for both research questions, i.e. negative predictive value (NPV), specificity and positive predictive value (PPV) for serial hs-cTnT measurements; and PPV, sensitivity and NPV for a single high baseline hs-cTnT value. To this end, a meta-analysis for proportions was performed by applying random effects models. Briefly, the Freeman-Tukey double arcsine method was used to transform the sensitivity, NPV, specificity and PPV estimates for each study.10,11 These were then used to calculate weighted summary estimates and their 95% CIs under the random effects model.12 Heterogeneity was assessed with the I2 statistic.13,14

models. Briefly, the Freeman-Tukey double arcsine method was used to transform the sensitivity, NPV, specificity and PPV estimates for each study.10,11 These were then used to calculate weighted summary estimates and their 95% CIs under the random effects model.12 Heterogeneity was assessed with the I2 statistic.13,14 Due to the limited number of studies in the quantitative analyses (<10), subgroup analyses and meta-regression were not performed. Publication bias was not investigated, because of suboptimal performance of standard tests and funnel plots in diagnostic test accuracy (DTA) studies and little evidence of the existence of publication bias in DTA studies to this date.15 All statistical analyses were performed using Microsoft Excel 2010 (Microsoft Corporation, Redmond, Washington State, USA) and MedCalc statistical software version 18 (MedCalc Software bvba, Ostend, Belgium). Results The systematic literature search generated 625 potentially relevant citations. An additional article was identified through a hand search of reference lists of relevant papers. After titles and abstracts screening, 539 studies were excluded. The remaining 87 articles were assessed in full-text, after which 66 studies were excluded for various reasons (Figure 1). Of the 21 articles discussed in the present systematic review (Table 1),16–36 14 studies were included in the meta-analysis with a total of 11,929 patients and an overall prevalence of AMI of 13.0% (range 3.6–56%)17,19,20,24–27,29–33,35,36 Figure 1. Flow diagram: study inclusion process for the systematic review and meta-analysis.

Results The systematic literature search generated 625 potentially relevant citations. An additional article was identified through a hand search of reference lists of relevant papers. After titles and abstracts screening, 539 studies were excluded. The remaining 87 articles were assessed in full-text, after which 66 studies were excluded for various reasons (Figure 1). Of the 21 articles discussed in the present systematic review (Table 1),16–36 14 studies were included in the meta-analysis with a total of 11,929 patients and an overall prevalence of AMI of 13.0% (range 3.6–56%)17,19,20,24–27,29–33,35,36 Figure 1. Flow diagram: study inclusion process for the systematic review and meta-analysis. AMI: acute myocardial infarction; hs-cTnT: high-sensitivity cardiac troponin T. Table 1. Study and patient characteristics of all studies included in the systematic review.

49.6%) <0.001 <2 hours 2/229 (0.9%) 0/41 (0.0%) 2/188 (1.1%) 1 <6 hours 7/229 (3.1%) 1/41 (2.4%) 6/188 (3.2%) 0.8 <24 hours 44/229 (19.2%) 2/41 (4.9%) 42/188 (22.3%) 0.02 <72 hours 132/229 (57.6%) 9/41 (22.0%) 123/188 (65.4%) <0.001 <96 hours 132/229 (57.6%) 9/41 (22.0%) 123/188 (65.4%) <0.001 Percutaneous coronary int ervention Coronary angiography with PCI 63/229 (27.5%) 6/41 (14.6%) 57/188 (30.3%) 0.07 Coronary angiography without PCI 166/229 (72.5%) 35/41 (85.4%) 131/188 (69.7%) 0.07 Mortality 30 days 7 (0.4%) 1 (0.1%) 6 (1.6%) <0.001 1 year 35 (2.2%) 8 (0.6%) 27 (7.1%) <0.001 UA: unstable angina; NSTEMI: non-ST-segment elevation myocardial infarction; STEMI: ST-segment elevation myocardial infarction; ACS: acute coronary syndrome; ED: emergency department; eGFR: estimated glomerular filtration rate; hsTnT: high-sensitivity troponin T; PCI: percutaneous coronary intervention. a Patients with STEMI were registered but excluded for the analysis. After discharge, only one patient died 13 days post-discharge, yielding an all-cause mortality rate of 0.08% (Figure 2). This 89-year-old woman presented with typical chest pain onset over 6 hours before presentation and had a GRACE score of 136 points. NSTEMI was ruled out based on two serial hsTnT values 2 hours apart (both 12 ng/L). Investigations identified urinary tract infection and mild hyponatremia. Work-up revealed chest radiography suspicious for lung cancer. The patient left hospital at her own request despite a recommendation for hospitalisation.

Results The systematic literature search generated 625 potentially relevant citations. An additional article was identified through a hand search of reference lists of relevant papers. After titles and abstracts screening, 539 studies were excluded. The remaining 87 articles were assessed in full-text, after which 66 studies were excluded for various reasons (Figure 1). Of the 21 articles discussed in the present systematic review (Table 1),16–36 14 studies were included in the meta-analysis with a total of 11,929 patients and an overall prevalence of AMI of 13.0% (range 3.6–56%)17,19,20,24–27,29–33,35,36 Figure 1. Flow diagram: study inclusion process for the systematic review and meta-analysis. AMI: acute myocardial infarction; hs-cTnT: high-sensitivity cardiac troponin T. Table 1. Study and patient characteristics of all studies included in the systematic review. Study Author, year Study type, country Number of patients included Target condition Reference assay Age, year mean±SD or median (IQR) Male sex % Time to presentation+, hours mean±SD or median (IQR) Study group Christchurch Aldous et al., 201216 Single-centre, New Zealand 385 AMI cTnI 65 (56–76) 60.5 2.7 (2.0–3.3) Christchurch Aldous et al., 201117 Single-centre, New Zealand 332 AMI cTnI 64.3 (52.8–73.5) 60.2 4.0 (2.0-8.6) Christchurch Aldous et al., 201118 Single-centre, New Zealand 939 NSTEMI cTnI 65 (56-76) 59.7 6.3 (3.3-13.3) Christchurch Aldous et al., 201219 Single-centre, New Zealand 939 NSTEMI cTnI 65 (56–76) 59.7 6.3 (3.3–13.3) Nuremberg Bahrmann et al., 201320 Single-centre, Germany 306 NSTEMI hs-cTnT 81±6 49 NA Heidelberg Biener et al., 201321 Single-centre, Germany 572 NSTEMI hs-cTnT 72.7 64.0 NA Heidelberg Biener et al., 201522 Single-centre, Germany 658 NSTEMI hs-cTnT 70.6 63.4 NA Heidelberg Biener et al., 201323 Single-centre, Germany 459 NSTEMI hs-cTnT 72.0 61.9 NA Chenevier-Gobeaux et al., 201324 Multi-centre, France 84 (subgroup >70 years) AMI cTnI 81±8 51 NA TI-AMO Goorden et al., 201625 Single-centre, The Netherlands 1490 AMI hs-cTnT 69 (57–80) 50 NA Stockholm Melki et al., 201126 Single-centre, Sweden 233 NSTEMI hs-cTnT, (cTnT, cTnI) 65 (55–76) 67 5.3 (3.3–7.5) TRAPID-AMI Mueller et al., 201627 Multi-centre, Switzerland, Germany, Spain etc.a 1282 AMI cTnI 62 (50–74) 62.8 1.8 (1.0–2.9) Heidelberg Mueller et al., 201228 Single-centre, Germany 784 NSTEMI hs-cTnT 72.7 51 NA TRAPID-AMI Mueller-Hennessen et al., 201729 Multi-centre, Switzerland, Germany, Spain etc.a 1282 AMI cTnI 62 (50–74) 62.8 1.8 (1.0–2.9) Brisbane Parsonage et al., 201430 Single-centre, Australia 764 AMI cTnI 55.3±15.1 61.3 4.97 (1.63–20.60) APACE Reichlin et al., 201231 Multi-centre, Switzerland 872 NSTEMI hs-cTnT 64 (51–75) 67 NA APACE Reichlin et al., 201532 Multi-centre, Switzerland, Spain and Italy 1320 NSTEMI hs-cTnT 60 (49–73) 69.3 NA APACE Reiter et al., 201133 Multi-centre, Switzerland, Spain 406 (subgroup >70 years) AMI cTnI, cTnT 78 (74–82) 54 NA TUSCA Santalo et al., 201334 Multi-centre, Spain 358 NSTEMI cTnT 69 (27–93) 67.9 NA Slagman et al., 201735 Single-centre,

lti-centre, Switzerland, Spain and Italy 1320 NSTEMI hs-cTnT 60 (49–73) 69.3 NA APACE Reiter et al., 201133 Multi-centre, Switzerland, Spain 406 (subgroup >70 years) AMI cTnI, cTnT 78 (74–82) 54 NA TUSCA Santalo et al., 201334 Multi-centre, Spain 358 NSTEMI cTnT 69 (27–93) 67.9 NA Slagman et al., 201735 Single-centre, Germany 3423 NSTEMI hs-cTnT, POC-TnT 61 (45–73) 57.2 NA Lund Thelin et al., 201336 Single-centre, Sweden 478 NSTEMI hs-cTnT 66 (55–76) 63 NA AMI: acute myocardial infarction; cTnI: cardiac troponin I; cTnT: cardiac troponin T; hs-cTnT: high-sensitivity cardiac troponin T; IQR: interquartile range; NA: not available; NSTEMI: non-ST segment elevation myocardial infarction; POC-TnT: point-of-care troponin; SD: standard deviation; T; +time from chest pain onset to ED presentation; TI-AMO: High sensitive Troponin T and I to diagnose Acute Myocardial Infarction, a prospective Observational study; TRAPID-AMI: High Sensitivity Cardiac Troponin T Assay for Rapid Rule-out of Acute Myocardial Infarction; APACE: Advantageous Predictors of Acute Coronary Syndromes Evaluation; TUSCA: Ultrasensitive Troponin in Acute Coronary Syndrome. a For a more detailed list see Supplementary Material Tables S2 and S3.

Germany 3423 NSTEMI hs-cTnT, POC-TnT 61 (45–73) 57.2 NA Lund Thelin et al., 201336 Single-centre, Sweden 478 NSTEMI hs-cTnT 66 (55–76) 63 NA AMI: acute myocardial infarction; cTnI: cardiac troponin I; cTnT: cardiac troponin T; hs-cTnT: high-sensitivity cardiac troponin T; IQR: interquartile range; NA: not available; NSTEMI: non-ST segment elevation myocardial infarction; POC-TnT: point-of-care troponin; SD: standard deviation; T; +time from chest pain onset to ED presentation; TI-AMO: High sensitive Troponin T and I to diagnose Acute Myocardial Infarction, a prospective Observational study; TRAPID-AMI: High Sensitivity Cardiac Troponin T Assay for Rapid Rule-out of Acute Myocardial Infarction; APACE: Advantageous Predictors of Acute Coronary Syndromes Evaluation; TUSCA: Ultrasensitive Troponin in Acute Coronary Syndrome. a For a more detailed list see Supplementary Material Tables S2 and S3. Study and patient characteristics In studies that reported time to presentation, the median time to presentation ranged from one hour27 to 6.3 h (Table 1).18,19 The proportion of women varied from 30.7%32 to 51%.20 Fourteen out of 21 studies excluded patients with ST-segment elevation myocardial infarction (STEMI) from the analysis. Four studies had a hs-cTnT threshold>14 ng/l as part of their inclusion criteria21–23,28 (see Supplementary Material Table S2). Eleven studies used the hs-cTnT assay as part of their reference standard.20–23,25,26,28,31,32,35,36 Additional characteristics of all included studies, such as the reported cut-offs and the timing of the troponin measurements for serial measurement strategies can be found in Supplementary Material Tables S2, S3 and S5.

leven studies used the hs-cTnT assay as part of their reference standard.20–23,25,26,28,31,32,35,36 Additional characteristics of all included studies, such as the reported cut-offs and the timing of the troponin measurements for serial measurement strategies can be found in Supplementary Material Tables S2, S3 and S5. Methodological quality assessment The results of the QUADAS-2 methodological quality assessment are provided in Supplementary Material Table S4. Eight studies consecutively enrolled patients presenting to the ED.17,20–24,34,35 Eleven studies used the hs-cTnT assay as part of their reference standard and were considered as high risk for incorporation bias.20–23,25,26,28,31,32,35,36 Seven studies did not exclude STEMI patients,16,17,25,27,29,30,33 thus raising concerns about applicability. All studies formally re-adjudicated the final diagnoses, except the study by Slagman et al.35 in which the initial clinical diagnosis was used to establish the endpoints. Reference assay cut-offs used for defining the endpoint differed between the various papers that used a standardised adjudication process (see Supplementary Material Table S4).

udicated the final diagnoses, except the study by Slagman et al.35 in which the initial clinical diagnosis was used to establish the endpoints. Reference assay cut-offs used for defining the endpoint differed between the various papers that used a standardised adjudication process (see Supplementary Material Table S4). Meta-analysis Serial measurements of hs-cTnT for rule-out of AMI Out of the 14 studies that were included in the meta-analysis, nine studies reported the diagnostic accuracy of serial measurements.17,19,20,26,27,30–32,36 Six of these studies presented the sensitivity of serial hs-cTnT measurements <14 ng/l (99th percentile).17,19,20,26,30,36 The median time of serial troponin measurement was 2.5 h (range, two to 6–24 h). The prevalence of AMI ranged from 7.3–56%. Applying the 99th percentile as cut-off classified 60.1% (range, 32.0–77.7%) of patients as rule-out (see Supplementary Material Table S6). The summary sensitivity of serial hs-cTnT measurements <14 ng/l was 96.7% (95% CI: 92.3–99.3; I2=82.1) (Figure 2). The NPV’s of the individual studies varied from 94.7% to 100% (see Supplementary Material Figure S1). For completeness, the summary estimates of NPV, specificity and PPV are provided in Supplementary Material Figures S1–S3. Figure 2. Forest plot displaying the summary estimate of sensitivity of serial high-sensitivity cardiac troponin T (hs-cTnT) measurements <14 ng/l (99th percentile).

Meta-analysis Serial measurements of hs-cTnT for rule-out of AMI Out of the 14 studies that were included in the meta-analysis, nine studies reported the diagnostic accuracy of serial measurements.17,19,20,26,27,30–32,36 Six of these studies presented the sensitivity of serial hs-cTnT measurements <14 ng/l (99th percentile).17,19,20,26,30,36 The median time of serial troponin measurement was 2.5 h (range, two to 6–24 h). The prevalence of AMI ranged from 7.3–56%. Applying the 99th percentile as cut-off classified 60.1% (range, 32.0–77.7%) of patients as rule-out (see Supplementary Material Table S6). The summary sensitivity of serial hs-cTnT measurements <14 ng/l was 96.7% (95% CI: 92.3–99.3; I2=82.1) (Figure 2). The NPV’s of the individual studies varied from 94.7% to 100% (see Supplementary Material Figure S1). For completeness, the summary estimates of NPV, specificity and PPV are provided in Supplementary Material Figures S1–S3. Figure 2. Forest plot displaying the summary estimate of sensitivity of serial high-sensitivity cardiac troponin T (hs-cTnT) measurements <14 ng/l (99th percentile). AMI: acute myocardial infarction; CI: confidence interval; cTnI: cardiac troponin I; FN: false negative; NSTEMI: non-ST segment elevation myocardial infarction; TP: true positive.

Figure 2. Forest plot displaying the summary estimate of sensitivity of serial high-sensitivity cardiac troponin T (hs-cTnT) measurements <14 ng/l (99th percentile). AMI: acute myocardial infarction; CI: confidence interval; cTnI: cardiac troponin I; FN: false negative; NSTEMI: non-ST segment elevation myocardial infarction; TP: true positive. The remaining three studies used a one-hour algorithm for serial measurements and presented the sensitivity of a baseline hs-cTnT value<12 ng/l and delta (Δ) 1 h<3 ng/l.27,31,32 The prevalence of AMI was comparable between the studies, ranging from 16.6–17.3%. The one-hour algorithm classified 60.2% (range, 56.3–63.4%) of patients as rule-out (see Supplementary Material Table S6). The pooled sensitivity for this algorithm was 98.9% (95% CI: 96.4–100; I2=77.5%) (Figure 3). The NPV’s of the individual studies varied from 99.1% to 100% (see Supplementary Material Figure S4). For completeness, the summary estimates of NPV, specificity and PPV are provided in Supplementary Material Figures S4–S6. Figure 3. Forest plot displaying the summary estimate of sensitivity of high-sensitivity cardiac troponin T (hs-cTnT) <12 ng/l and Δ1 h<3 ng/l. AMI: acute myocardial infarction; CI: confidence interval; cTnI: cardiac troponin I; FN: false negative; NSTEMI: non-ST segment elevation myocardial infarction; TP: true positive.

The remaining three studies used a one-hour algorithm for serial measurements and presented the sensitivity of a baseline hs-cTnT value<12 ng/l and delta (Δ) 1 h<3 ng/l.27,31,32 The prevalence of AMI was comparable between the studies, ranging from 16.6–17.3%. The one-hour algorithm classified 60.2% (range, 56.3–63.4%) of patients as rule-out (see Supplementary Material Table S6). The pooled sensitivity for this algorithm was 98.9% (95% CI: 96.4–100; I2=77.5%) (Figure 3). The NPV’s of the individual studies varied from 99.1% to 100% (see Supplementary Material Figure S4). For completeness, the summary estimates of NPV, specificity and PPV are provided in Supplementary Material Figures S4–S6. Figure 3. Forest plot displaying the summary estimate of sensitivity of high-sensitivity cardiac troponin T (hs-cTnT) <12 ng/l and Δ1 h<3 ng/l. AMI: acute myocardial infarction; CI: confidence interval; cTnI: cardiac troponin I; FN: false negative; NSTEMI: non-ST segment elevation myocardial infarction; TP: true positive. Single high baseline measurement of hs-cTnT for rule-in of AMI Six out of 14 studies reported the specificity of a single high baseline measurement24,25,29,32,33,35 (Figure 4). The prevalence of AMI differed considerably between the studies, ranging from 3.6–24%. The pooled specificity of a high baseline hs-cTnT value was 94.6% (95% CI: 91.5–97.1; I2=95.5%). Sensitivity analysis, performed by removing the two studies that used cut-offs other than 50 ng/l, produced a pooled specificity of 95.2% (91.6–97.9%). The PPV’s of the individual studies varied from 28.3–86.5% (see Supplementary Material Figure S7). For completeness, the summary estimates of PPV, sensitivity and NPV are provided in Supplementary Material Figures S7–S9.

studies that used cut-offs other than 50 ng/l, produced a pooled specificity of 95.2% (91.6–97.9%). The PPV’s of the individual studies varied from 28.3–86.5% (see Supplementary Material Figure S7). For completeness, the summary estimates of PPV, sensitivity and NPV are provided in Supplementary Material Figures S7–S9. Figure 4. Forest plot displaying the summary estimate of specificity of a baseline high-sensitivity cardiac troponin T (hs-cTnT) value>50 ng/l. AMI: acute myocardial infarction; CI: confidence interval; cTnI: cardiac troponin I; cTnT: cardiac troponin T; FP: false positive; NSTEMI: non-ST segment elevation myocardial infarction; POC-TnT: point-of-care troponin T; TN: true negative. Discussion Novel cardiac biomarkers such as the hs-cTnT have become increasingly important in the diagnostic pathway and risk stratification of patients presenting with acute chest pain to the ED. They are a central part of clinical decision algorithms recommended by the current ESC guidelines.5 The present systematic review and meta-analysis demonstrates that: (a) the two most frequently reported serial hs-cTnT measurement strategies have a high sensitivity to rule out AMI, while triaging a similarly large proportion of patients towards rule-out and (b) a direct rule-in strategy with a single baseline hs-cTnT value>50 ng/l has a high specificity.

-analysis demonstrates that: (a) the two most frequently reported serial hs-cTnT measurement strategies have a high sensitivity to rule out AMI, while triaging a similarly large proportion of patients towards rule-out and (b) a direct rule-in strategy with a single baseline hs-cTnT value>50 ng/l has a high specificity. Serial measurements for rule-out of AMI Several meta-analyses have previously assessed the ability of a hs-cTnT assay to rule out AMI. Four articles examined a single baseline hs-cTnT measurement for rule-out of AMI at various diagnostic cut-offs.37–40 The 99th percentile was shown to have modest sensitivity for rule-out of AMI,37,39,40 however when the cut-off for rule-out of AMI was set below the limit of detection, i.e. <5 ng/l for the hs-cTnT assay, the sensitivities were generally high.38–40 It has been suggested that serial measurement of hs-cTnT is more accurate and informative than a single measurement for rule-out of AMI. As serial measurements provide information on rise and fall patterns, they are more informative for discrimination of acute from chronic myocardial injury.19 A serial measurement strategy is particularly necessary in early presenters as they might have normal initial troponin values due to the time dependency of troponin release.41,42 A subgroup analysis performed by Mueller et al. showed that in early presenters (chest pain onset to presentation <2 h) the one-hour algorithm reached a NPV comparable to late presenters (chest pain onset to presentation ⩾2 h) to the emergency room.27 Adding copeptin, a marker which is released very early after onset of symptoms, to hs-cTnT has also been suggested for early presenters in particular.43 A recent meta-analysis by Shin et al. showed that adding copeptin to hs-cTnT improved the sensitivity for rule-out of AMI.44

resentation ⩾2 h) to the emergency room.27 Adding copeptin, a marker which is released very early after onset of symptoms, to hs-cTnT has also been suggested for early presenters in particular.43 A recent meta-analysis by Shin et al. showed that adding copeptin to hs-cTnT improved the sensitivity for rule-out of AMI.44 Our meta-analysis exclusively investigated the ability of the hs-cTnT assay to rule out AMI with the two most frequently reported serial measurement strategies. We found similar diagnostic accuracy for rule-out of AMI with serial high-sensitivity troponin measurements <99th percentile compared to the study by Lipinski et al.37 Recently, Badertscher et al. directly compared the one-hour algorithm with the three-hour algorithm, which uses a fixed cut-off (the 99th percentile) at presentation and three hours in conjunction with clinical criteria (Global Registry of Acute Events (GRACE) score <140 and the requirement to be pain free). While both ESC recommended algorithms had comparable NPVs for rule-out, the one-hour algorithm allowed the rule-out of significantly more patients.45 Contrastingly, the results of our study do not suggest a difference between the two rule-out strategies in the proportion of patients that are triaged towards rule-out. However, it is important to note that the serial measurement strategy with a fixed 99th percentile cut-off described in our study lacked the clinical criteria which are a key part of the ESC three-hour algorithm.5

rence between the two rule-out strategies in the proportion of patients that are triaged towards rule-out. However, it is important to note that the serial measurement strategy with a fixed 99th percentile cut-off described in our study lacked the clinical criteria which are a key part of the ESC three-hour algorithm.5 It is important to highlight that the majority of the studies, including the studies in this meta-analysis, have only examined the performance of the hs-cTnT assay in patients presenting with acute chest pain and free from major comorbidities. Few prospective studies have assessed the performance of this assay in lower-risk patients, e.g. women presenting with atypical symptoms, and higher-risk patients, e.g. patients with renal failure, to better reflect the cohort of patients in clinical practice. Biener et al. demonstrated that in patients presenting with atypical symptoms the sensitivity of a rule-out strategy with hs-cTnT was lower than in patients with typical chest pain.22 Twerenbold et al. showed that in patients with renal failure a rule-out strategy based on hs-cTnT had a comparably high sensitivity and NPV to patients without renal failure, however they found that the efficacy of the strategy was substantially lower.46 The underlying cause is the increased baseline troponin value in patients with renal failure, which decreases the possibility of a rule-out when using the same cut-off values.47

ly high sensitivity and NPV to patients without renal failure, however they found that the efficacy of the strategy was substantially lower.46 The underlying cause is the increased baseline troponin value in patients with renal failure, which decreases the possibility of a rule-out when using the same cut-off values.47 Single high baseline measurement for rule-in of AMI To our knowledge, this is the first systematic review and meta-analysis to report the specificity and PPV of a direct rule-in strategy with baseline hs-cTnT value>50 ng/l. The prevalence of AMI differed considerably between the studies; in the study by Slagman et al. the prevalence of AMI was 3.6%, whereas in the study by Reiter et al. the prevalence was 24%.33,35 The specificity of the direct rule-in strategy was consistently high between the individual studies. At the same time the PPV differed considerably; in studies with a low prevalence of AMI, we observed a lower PPV, which is in concordance with the theorem of Bayes.48 We conclude that the applicability of a direct rule-in strategy with baseline hs-cTnT value>50 ng/l is highly dependent on the pre-test probability of disease and this stresses the importance of assessing the individual pre-test probability for clinicians. Patients with a low pre-test probability or a very atypical presentation might suffer from other conditions that also give rise to hs-cTnT. In such cases, serial hs-cTnT measurements can increase the probability of AMI when a rise and fall pattern is present.49 If after serial sampling AMI is deemed unlikely, other causes of troponin elevations should be investigated as elevated troponins are associated with an unfavourable prognosis even in the absence of an AMI.50

ses, serial hs-cTnT measurements can increase the probability of AMI when a rise and fall pattern is present.49 If after serial sampling AMI is deemed unlikely, other causes of troponin elevations should be investigated as elevated troponins are associated with an unfavourable prognosis even in the absence of an AMI.50 Recommendations for further research The 2015 ESC one-hour algorithm has distinct cut-offs for ‘rule-out’ and ‘rule-in’ and patients not meeting these criteria are placed in the ‘observational zone’. This concerns a considerable number of patients (20–30%) who are known to have an unfavourable prognosis.51 They are now faced with prolonged observational periods in the hospital, with or without invasive testing. Further research is needed to determine the optimal diagnostic approach. It has been suggested that advanced cardiac imaging may be useful for better risk stratification in these patients.52,53

urable prognosis.51 They are now faced with prolonged observational periods in the hospital, with or without invasive testing. Further research is needed to determine the optimal diagnostic approach. It has been suggested that advanced cardiac imaging may be useful for better risk stratification in these patients.52,53 Study limitations Our study has several limitations. The studies included in the current analysis had different methods for adjudication of the final diagnosis. Some used conventional assays, whereas others used high-sensitivity assays for the reference standard. In addition, studies that used the same assay frequently had different cut-offs for adjudication of the final diagnosis. The use of high-sensitivity assays as opposed to conventional assays as the reference standard can influence the diagnostic accuracy parameters of the test being evaluated. Because of its higher sensitivity, the high-sensitivity assay detects more patients with myocardial injury not due to NSTEMI when compared to conventional assays. Moreover, we were not able to validate and examine the performance of the official ESC 0/1 hour algorithm for rule-out and rule-in, because of the limited number of distinct cohorts that were in studies with this algorithm. Due to lack of individual patient data and the limited number of studies we could not perform meta-regression or subgroup analysis to investigate the effects of certain patient and study characteristics, such as age, cardiac risk factors, target condition, time of chest pain onset to obtaining the first blood sample, assay used for reference standard and prevalence of AMI.

number of studies we could not perform meta-regression or subgroup analysis to investigate the effects of certain patient and study characteristics, such as age, cardiac risk factors, target condition, time of chest pain onset to obtaining the first blood sample, assay used for reference standard and prevalence of AMI. Conclusion Serial hs-cTnT measurement strategies to rule out AMI perform well and can classify 60% of patients for rule-out, while a direct rule-in strategy with a baseline hs-cTnT value >50 ng/l has a high specificity. Supplemental Material ACC819421_supplementary_material – Supplemental material for Serial high-sensitivity cardiac troponin T measurements to rule out acute myocardial infarction and a single high baseline measurement for swift rule-in: A systematic review and meta-analysis Click here for additional data file. Supplemental material, ACC819421_supplementary_material for Serial high-sensitivity cardiac troponin T measurements to rule out acute myocardial infarction and a single high baseline measurement for swift rule-in: A systematic review and meta-analysis by M Arslan, A Dedic, E Boersma and EA Dubois in European Heart Journal: Acute Cardiovascular Care The authors would like to thank Wichor Bramer (Biomedical Information Specialist, Erasmus MC) for his aid in the development and implementation of the search strategy. Conflict of interest: The authors declare that there is no conflict of interest. Funding: This work was supported by a research grant from the Erasmus MC Thorax Foundation (project grant B4).

The authors would like to thank Wichor Bramer (Biomedical Information Specialist, Erasmus MC) for his aid in the development and implementation of the search strategy. Conflict of interest: The authors declare that there is no conflict of interest. Funding: This work was supported by a research grant from the Erasmus MC Thorax Foundation (project grant B4). Supplemental material: Supplemental material for this article is available online.

Introduction Accelerated 0/1 or 0/2-hour diagnostic protocols for the diagnosis of non-ST-segment elevation acute myocardial infarction (NSTEMI) are being recommended by the 2015 European Society of Cardiology (ESC) guidelines1 as an alternative to the established ESC 0/3-hour protocol. Although accelerated protocols have been validated in numerous observational cohorts,2–11 implementation is very low worldwide.12 Reasons for the limited use of fast protocols are multifactorial and include fear of litigation after missed myocardial infarction (MI) in early presenters, the presence of comorbidities that were excluded or were under-represented in observational studies and the limited clinical experience in patients presenting with symptoms other than chest pain or angina.13–16

re multifactorial and include fear of litigation after missed myocardial infarction (MI) in early presenters, the presence of comorbidities that were excluded or were under-represented in observational studies and the limited clinical experience in patients presenting with symptoms other than chest pain or angina.13–16 Another issue is the lack of a clinically validated acceptable event rate after discharge. All-cause mortality rates between 0.1% and 2% have been suggested to be acceptable based on a survey among physicians who were asked to give their expectation on 30-day mortality rates.17 In observational studies physicians were usually unaware of investigational biomarker results and protocols, and patients were discharged at the discretion of the attending physician. Adding to this dilemma, there are only few randomised biomarker-based trials evaluating the safety of discharge in low-risk patients, either using high-sensitivity troponin (hsTn) assays in combination with validated clinical scores,18,19 a dual biomarker strategy combining copeptin with cardiac troponin,8 or an accelerated diagnostic protocol using hsTn I measurements 2 hours apart, together with electrocardiography (ECG) and either the thrombolysis in myocardial infarction (TIMI)9,20 or the emergency department assessment of chest pain (EDACS) score.20 Moreover, incremental evidence comes from a large pre and post-implementation study on 31,332 patients providing findings on the lower length of emergency department (ED) stay and increased rates of discharge within 6 hours, without an adverse event when clinical pathways were correctly used.21

st pain (EDACS) score.20 Moreover, incremental evidence comes from a large pre and post-implementation study on 31,332 patients providing findings on the lower length of emergency department (ED) stay and increased rates of discharge within 6 hours, without an adverse event when clinical pathways were correctly used.21 In the light of sparse real world evidence, our prospective pre/post-implementation study sought to evaluate the feasibility, efficacy and safety of ESC recommended fast diagnostic protocols using hsTnT in a consecutive all-comer cohort with suspected acute coronary syndrome (ACS) based on a broad spectrum of symptoms.

In the light of sparse real world evidence, our prospective pre/post-implementation study sought to evaluate the feasibility, efficacy and safety of ESC recommended fast diagnostic protocols using hsTnT in a consecutive all-comer cohort with suspected acute coronary syndrome (ACS) based on a broad spectrum of symptoms. Methods In this prospective single centre study at Heidelberg University Hospital, we screened all consecutive patients with suspected ACS between 1 July 2016 and 30 June 2017. In this period, 7668 patients presented to the ED. Patients were managed in a chest pain unit (CPU), which represents a specialised ED that is led by a cardiologist and requires certification by the German Cardiac Society (Deutsche Gesellschaft für Kardiologie; DGK). In Germany, more than 320 certified CPUs are distributed across the country and represent the preferred facilities for the evaluation of patients with suspected ACS. The median number of patients per day (25th percentile–75th percentile) was 20 (17–24) patients. The ED of the department of cardiology is part of the internal medicine ED and follows the CPU quality criteria that are audited by the German Cardiac Society. The team consists of experienced resident physicians in training for cardiology working a three-shift schedule on weekdays (day shift 2, swing shift 1 and night shift 1 physician) and a two-shift schedule during weekends (day shift 1 and night shift 1 physician). The nursing team consists of experienced nurses working in a three-shift system (day shift 3, swing shift 3, night shift 2 nurses), with a ratio of one nurse per five patients. The ED is under the permanent supervision of a senior cardiologist who is responsible for the decision to admit or discharge, and for the indication and timing of an invasive strategy. There is unlimited access to coronary angiography or other diagnostic resources as per the required criteria for certification of a CPU.22 Patient disposition, times and treatments were collected in a 6-month pre-implementation period followed by the implementation on 1 January 2017 that encouraged the use of the ESC 0/1-hour algorithm as the primary diagnostic strategy, and subsequently a post-implementation period of another 6 months to demonstrate changes.

ent disposition, times and treatments were collected in a 6-month pre-implementation period followed by the implementation on 1 January 2017 that encouraged the use of the ESC 0/1-hour algorithm as the primary diagnostic strategy, and subsequently a post-implementation period of another 6 months to demonstrate changes. Exclusion criteria comprised the following: (a) repeated presentations beyond the index admission (‘frequent flyer’); (b) patients referred from other hospitals for early or primary percutaneous coronary intervention (PCI) without receiving a standard diagnostic work-up; (c) diagnostic set of hsTnT samples not available (e.g. missing initial or consecutive blood sample); (d) patients with ST-segment elevation myocardial infarction (STEMI) were registered but were excluded for this analysis. A consort diagram illustrates the screening process (Figure 1). Figure 1. Standards for reporting diagnostic accuracy studies statement (STARD) patient inclusion flow diagram.

Exclusion criteria comprised the following: (a) repeated presentations beyond the index admission (‘frequent flyer’); (b) patients referred from other hospitals for early or primary percutaneous coronary intervention (PCI) without receiving a standard diagnostic work-up; (c) diagnostic set of hsTnT samples not available (e.g. missing initial or consecutive blood sample); (d) patients with ST-segment elevation myocardial infarction (STEMI) were registered but were excluded for this analysis. A consort diagram illustrates the screening process (Figure 1). Figure 1. Standards for reporting diagnostic accuracy studies statement (STARD) patient inclusion flow diagram. Patients qualified for enrolment with initial presentation of clinically suspected ACS, based on a broad spectrum of symptoms including atypical symptoms and dyspnoea. Patients on chronic haemodialysis were not included. Previously published findings23 in symptomatic patients with atrioventricular nodal re-entrant tachycardia demonstrate ST-segment depressions and relevant troponin kinetics rendering the unequivocal differentiation between type 1 MI, type 2 MI, MI with non-obstructive coronary arteries, or acute myocardial injury impossible. Therefore, we decided not to include these patients. In the present study population, there are no patients with the following initial presentation: acute heart failure due to already known non-coronary heart disease without suspected ACS; confirmed primary pulmonary disease without suspected ACS; traumatic chest pain with preceded thorax injury without suspected ACS. Patients were not excluded for severe chronic kidney disease, older age, chronic heart failure, or atrial fibrillation. Patients were not included in the case of inappropriate command of the English/German language or permanent residence in a foreign country.

hest pain with preceded thorax injury without suspected ACS. Patients were not excluded for severe chronic kidney disease, older age, chronic heart failure, or atrial fibrillation. Patients were not included in the case of inappropriate command of the English/German language or permanent residence in a foreign country. Reliable electronic time stamps were available for the time of arrival, referral and discharge, blood draws (between specimen intervals) and reporting of blood test results (‘turn-around-times’), and all diagnostic and therapeutic interventions. Acute MI was diagnosed in-hospital by treating clinicians based on all clinical information, using the diagnostic criteria of the 3rd universal MI definition.24 Patients were categorised using the validated biomarker criteria for classification into rule-out, observe or rule-in as proposed by the 2015 ESC guidelines on non-ST elevation (NSTE)-ACS.1 Levels of cardiac troponin were measured at presentation, after 1 or 3 hours and thereafter as long as clinically indicated. A small proportion of patients were categorised into ruled-out using hsTnT and copeptin at presentation, with biomarker cut-offs based on the BIC-8 trial.8 HsTnT was measured on Cobas E 411 (Roche Diagnostics Ltd., Rotkreuz, Switzerland) and copeptin was measured on Kryptor (Thermo Fisher Scientific, BRAHMS GmbH, Hennigsdorf, Germany).

n of patients were categorised into ruled-out using hsTnT and copeptin at presentation, with biomarker cut-offs based on the BIC-8 trial.8 HsTnT was measured on Cobas E 411 (Roche Diagnostics Ltd., Rotkreuz, Switzerland) and copeptin was measured on Kryptor (Thermo Fisher Scientific, BRAHMS GmbH, Hennigsdorf, Germany). On 1 January 2017, the ESC 0/1-hour algorithm was officially implemented. Prior to this date, every staff member in the ED (nurses and physicians) had received training for the use and interpretation of the ESC 0/1-hour algorithm using hsTnT from 1 January 2017. This included formal education, posters and bedside cards based on the algorithm shown in Figure 3 of the 2015 ESC guidelines on NSTE-ACS.1 The training was also implemented in the initial training for newly rotating personnel.

or the use and interpretation of the ESC 0/1-hour algorithm using hsTnT from 1 January 2017. This included formal education, posters and bedside cards based on the algorithm shown in Figure 3 of the 2015 ESC guidelines on NSTE-ACS.1 The training was also implemented in the initial training for newly rotating personnel. All patients underwent a clinical assessment that included medical history, physical examination, 12-lead ECG, continuous ECG monitoring, pulse oximetry and standard blood tests. Results were reported on the electronic patient record and were communicated to the clinicians responsible for patient care. Patients received treatment at the discretion of the attending physician, and all decisions to admit or discharge, or on the need and timing for invasive coronary angiography were made based on available information during the ED stay. The standard 12-lead ECG included routinely precordial leads V7–V9. The decision to discharge comprised clinical judgement from individual risk variables, or the GRACE score25 that was generated by an electronic calculator embedded into the electronic file, and was thus accessible for all physicians. Adjudication of final diagnoses in the ED was made prospectively in clinical routine by attending physicians and responsible cardiologists on duty, while confirmation of ED diagnoses for research purposes was done retrospectively by two cardiologists and a third cardiologist in case of discordance.

All patients underwent a clinical assessment that included medical history, physical examination, 12-lead ECG, continuous ECG monitoring, pulse oximetry and standard blood tests. Results were reported on the electronic patient record and were communicated to the clinicians responsible for patient care. Patients received treatment at the discretion of the attending physician, and all decisions to admit or discharge, or on the need and timing for invasive coronary angiography were made based on available information during the ED stay. The standard 12-lead ECG included routinely precordial leads V7–V9. The decision to discharge comprised clinical judgement from individual risk variables, or the GRACE score25 that was generated by an electronic calculator embedded into the electronic file, and was thus accessible for all physicians. Adjudication of final diagnoses in the ED was made prospectively in clinical routine by attending physicians and responsible cardiologists on duty, while confirmation of ED diagnoses for research purposes was done retrospectively by two cardiologists and a third cardiologist in case of discordance. Definition of fast protocols According to the ESC 0/1-hour protocol, patients with a 0-hour value below the limit of detection (LoD) (hsTnT <5 ng/L) and interval from the last chest pain episode exceeding 3 hours were classified as rule-out. Patients with a 0-hour value of hsTnT of 5 ng/L or greater and less than 12 ng/L and difference between the 0-hour and 1-hour value of hsTnT (Δ0–1 h) less than 3 ng/L were classified as rule-out. Patients with a 0-hour value of hsTnT of 52 ng/L or greater or difference between the 0-hour and 1-hour value of hsTnT (Δ0–1 h) of 5 ng/L or greater were classified as rule-in. Patients qualified for the ESC 0/1-hour protocol if the time window between the first and the second blood draw was between 30 and 90 minutes. According to the ESC 0/3-hour protocol, patients with a 0-hour value at or below the upper limit of normal (ULN) (hsTnT ≤14 ng/L) and interval from the last chest pain episode exceeding 6 hours were classified as rule-out. Patients with a 0-hour value at or below the ULN (hsTnT ≤14 ng/L) and absolute concentration change between the 0-hour and 3-hour value of hsTnT (Δ0–3 h) of 7 ng/L or less (defined as 50% of the ULN) were classified as rule-out. If the 0-hour value exceeded the ULN, patients with a relative concentration change between the 0-hour and 3-hour value of hsTnT (Δ0–3 h) of less than 20% of the 0-hour value were classified as rule-out. Patients qualified for the ESC 0/3-hour protocol if the second blood draw was between 150 and 210 minutes after the initial blood sample. Per protocol, blood samples for the baseline hsTnT value had to be obtained within 45 minutes according to laboratory time stamps and the follow-up specimens had to be obtained 1 or 3 hours ± 30 minutes after the initial specimen. Time sampling intervals between 90 and 150 minutes and blood draws beyond 210 minutes were summarised as ‘other’ protocols. These patients were commonly diagnosed using the criteria of either the ESC 0/1 or the ESC 0/3-hour protocol, whatever protocol came closer.

d 1 or 3 hours ± 30 minutes after the initial specimen. Time sampling intervals between 90 and 150 minutes and blood draws beyond 210 minutes were summarised as ‘other’ protocols. These patients were commonly diagnosed using the criteria of either the ESC 0/1 or the ESC 0/3-hour protocol, whatever protocol came closer. According to the 2015 ESC guidelines for the management of NSTE-ACS,1 repeat blood sampling was performed at the discretion of the treating physician, based on clinical assessment. The final diagnosis before admission or discharge was based on the complete clinical information and all hsTnT measurements. Endpoints The primary endpoints were: (a) the temporal change of implementation rate of the 0/1-hour algorithm and (b) 30-day all-cause mortality among patients discharged from the ED after rule-out. Secondary endpoints included: (a) the safety of discharge after 12 months; (b) the temporal trends of lengths of stay in the ED; and (c) the change of discharge rates before and after implementation. Additional endpoints included the prognostic role of baseline hsTnT concentrations on outcomes, as well as the impact of the severity of ED crowding on the length of ED stay and outcomes. Follow-up was accomplished using telephone, questionnaire, patient’s hospital notes, the family physician’s records and the municipal registry on vital status. The study was approved by the ethics committee of the University of Heidelberg, and performed in accordance with the Declaration of Helsinki. Informed consent of the individual patients was not required.

questionnaire, patient’s hospital notes, the family physician’s records and the municipal registry on vital status. The study was approved by the ethics committee of the University of Heidelberg, and performed in accordance with the Declaration of Helsinki. Informed consent of the individual patients was not required. Statistical analysis Continuous variables were tested for normal distribution and were presented either as means with 95% confidence intervals, or as medians with 25th/75th percentiles (interquartile range). The normality of data distribution was assessed by the Kolmogorov–Smirnov test. Groups were compared using the χ2 test for categorical variables and the Mann–Whitney U-test for continuous variables. Absolute changes between baseline and follow-up samples were calculated by Ct2−Ct1 (where C is troponin concentration and t1 and t2 represents the time-point of blood draw, respectively) and relative changes with the formula (Ct2−Ct1/Ct1) × 100 (expressed as a percentage). Kaplan–Meier curves and the log-rank test were used to assess differences in outcomes between groups. A multivariate Cox proportional hazards regression was performed to determine predictors for discharge. All hypothesis testing was two-tailed and P values less than 0.05 were considered statistically significant. All statistical analyses were performed using MedCalc 11.1, R 3.5.1 (The R Foundation for Statistical Computing) and SAS 9.3 (SAS Institute Inc.).

regression was performed to determine predictors for discharge. All hypothesis testing was two-tailed and P values less than 0.05 were considered statistically significant. All statistical analyses were performed using MedCalc 11.1, R 3.5.1 (The R Foundation for Statistical Computing) and SAS 9.3 (SAS Institute Inc.). Results During the 12-month study period, 2525 patients met inclusion criteria, while 910 patients were excluded due to presentation as STEMI, repeated presentations, or inappropriate set of troponin (Figure 1). Patients were followed up for all-cause mortality for a median of 400 (316–459) days, and follow-up was complete for 98.7% (missing follow-up in 34 of 2525 cases). Discordance between the final diagnosis in the ED and the retrospective adjudication occurred in only eight NSTEMI cases (2.4%) and in 11 unstable angina cases (3.9%). Demographic characteristics for the entire study cohort, split by inclusion period are shown in Table 1. Table 1. Demographic characteristics for the entire study cohort by study period.

Results During the 12-month study period, 2525 patients met inclusion criteria, while 910 patients were excluded due to presentation as STEMI, repeated presentations, or inappropriate set of troponin (Figure 1). Patients were followed up for all-cause mortality for a median of 400 (316–459) days, and follow-up was complete for 98.7% (missing follow-up in 34 of 2525 cases). Discordance between the final diagnosis in the ED and the retrospective adjudication occurred in only eight NSTEMI cases (2.4%) and in 11 unstable angina cases (3.9%). Demographic characteristics for the entire study cohort, split by inclusion period are shown in Table 1. Table 1. Demographic characteristics for the entire study cohort by study period. Entire cohort (N=2525) Period 1 (N=1243) Period 2 (N=1282) P value Age, years 62 ± 18 63 ± 18 61 ± 18 0.004 Sex, male/female 1465/1060 738/505 727/555 0.2 Presenting symptom: Chest pain 1164 (46.1%) 611 (49.2%) 553 (43.1%) 0.002 Dyspnoea 335 (13.3%) 173 (13.9%) 162 (12.6%) 0.3 Atypical 1021 (40.4%) 459 (36.9%) 562 (43.8%) <0.001 Else or missing 5 (0.2%) − 5 (0.4%) − Time from onset of symptoms to admission 0–3 hours 541 (21.4%) 249 (20.0%) 292 (22.8%) 0.09 3–6 hours 253 (10.0%) 128 (10.0%) 125 (9.7%) 0.6 >6 hours 1021 (40.4%) 459 (36.9%) 562 (43.8%) <0.001 Unknown 721 (28.6%) 372 (29.9%) 349 (27.2%) 0.1 Final diagnoses UA 280 (11.1%) 146 (11.7%) 134 (10.5%) 0.3 NSTEMI 330 (13.1%) 166 (13.4%) 164 (12.8%) 0.7 STEMIa 133a 58a 75a 0.6 Non-ACS 1915 (75.8%) 931 (74.9%) 984 (76.8%) 0.3 Renal function Serum creatinine (mg/dL) 0.98 ± 0.4 1.00 ± 0.5 0.95 ± 0.4 <0.001 eGFR (mL/min/1.73 m2) 81 ± 25 79 ± 26 82 ± 25 <0.001 hsTnT at admission (ng/L) 11 (6–25) 12 (6–29) 10 (6–22) 0.5 hsTnT <5 ng/L 247 (9.8%) 131 (10.5%) 116 (9.0%) 0.2 hsTnT 5–14 ng/L 1175 (46.5%) 509 (40.9%) 666 (51.9%) <0.001 hsTnT 15–51 ng/L 918 (36.4%) 506 (40.7%) 412 (32.1%) <0.001 hsTnT ≥52 ng/L 314 (12.4%) 162 (13.0%) 152 (11.8%) 0.4 hsTnT at 1 hour (ng/L) 9 (6–16) 8 (6–13) 9 (6–18) 0.07 Copeptin at admission (n=464) (pmol/L) 5.5 (3.4–11.2) 5.4 (3.3–10) 5.5 (3.5–12) 0.9 Copeptin <10 pmol/L 336 (72.4%) 221 (73.9%) 115 (69.7%) 0.3 GRACE score 98 (75–121) 99 (78–123) 95 (72–118) 0.003 Low 1466 (58.0%) 686 (55.2%) 780 (60.8%) 0.005 Intermediate 756 (29.9%) 394 (31.7%) 362 (28.2%) 0.05 High 275 (10.9%) 146 (11.8%) 129 (10.0%) 0.2 ECG Sinus rhythm 2175 (86.1%) 1061 (85.4%) 1114 (86.9%) 0.3 Atrial tachycardia 241 (9.5%) 118 (9.5%) 123 (9.6%) 0.9 Paced rhythm 58 (2.3%) 32 (2.6%) 25 (2.0%) 0.4 LBBB/RBBB 264 (10.4%) 139 (11.2%) 125 (9.8%) 0.3 ST depression 83 (3.3%) 31 (2.5%) 52 (4.1%) 0.04 T-wave inversion 521 (20.6%) 320 (25.7%) 201 (15.7%) <0.001 History Coronary artery disease 827 (32.8%) 447 (36.0%) 380 (29.6%) <0.001 Myocardial infarction 430 (17.0%) 247 (19.9%) 183 (14.3%) <0.001 Coronary intervention 655 (25.9%) 371 (29.8%) 284 (22.2%) <0.001 Coronary bypass surgery 166 (6.6%) 95 (7.6%) 71 (5.5%) 0.04 Coron

ersion 521 (20.6%) 320 (25.7%) 201 (15.7%) <0.001 History Coronary artery disease 827 (32.8%) 447 (36.0%) 380 (29.6%) <0.001 Myocardial infarction 430 (17.0%) 247 (19.9%) 183 (14.3%) <0.001 Coronary intervention 655 (25.9%) 371 (29.8%) 284 (22.2%) <0.001 Coronary bypass surgery 166 (6.6%) 95 (7.6%) 71 (5.5%) 0.04 Coron ary angiography 1039 (41.1%) 551 (44.3%) 488 (38.1%) 0.002 Risk factors Hypertension 1652 (65.4%) 845 (68.0%) 807(62.9%) <0.001 Hypercholesterolemia 1135 (44.9%) 580 (46.7%) 555 (43.3%) <0.001 Diabetes mellitus 535 (21.2%) 290 (23.3%) 245 (19.1%) <0.001 Active smoker 551 (21.8%) 253 (20.4%) 298 (23.2%) 0.9 Family history 665 (26.3%) 349 (28.1%) 316 (24.6%) <0.001 UA: unstable angina; NSTEMI: non-ST-segment elevation myocardial infarction; STEMI: ST-segment elevation myocardial infarction; ACS: acute coronary syndrome; eGFR: estimated glomerular filtration rate; hsTnT: high-sensitivity troponin T. a Patients with STEMI were registered but excluded for the analysis.

ary angiography 1039 (41.1%) 551 (44.3%) 488 (38.1%) 0.002 Risk factors Hypertension 1652 (65.4%) 845 (68.0%) 807(62.9%) <0.001 Hypercholesterolemia 1135 (44.9%) 580 (46.7%) 555 (43.3%) <0.001 Diabetes mellitus 535 (21.2%) 290 (23.3%) 245 (19.1%) <0.001 Active smoker 551 (21.8%) 253 (20.4%) 298 (23.2%) 0.9 Family history 665 (26.3%) 349 (28.1%) 316 (24.6%) <0.001 UA: unstable angina; NSTEMI: non-ST-segment elevation myocardial infarction; STEMI: ST-segment elevation myocardial infarction; ACS: acute coronary syndrome; eGFR: estimated glomerular filtration rate; hsTnT: high-sensitivity troponin T. a Patients with STEMI were registered but excluded for the analysis. Primary endpoints Utilisation of ESC 0/1-hour algorithm Trends for utilisation of diagnostic algorithms changed significantly after the implementation of the ESC 0/1-hour protocol. In particular, there was an increase of the ESC 0/1-hour algorithm by 270% and a concomitant decrease of the ESC 0/3-hour algorithm by 62%. The algorithms based on a single baseline hsTnT of less than 5 ng/L (LoD) remained almost stable. In parallel, the median interval between the initial and the first follow-up hsTnT specimen shortened by a median of 45 minutes from 2.2 (1.48–3.08) to 1.45 (1.15–2.03) hours post implementation. The proportion of patients categorised into rule-out, observational zone and rule-in was 62.9%, 16.2% and 20.9%, respectively.

parallel, the median interval between the initial and the first follow-up hsTnT specimen shortened by a median of 45 minutes from 2.2 (1.48–3.08) to 1.45 (1.15–2.03) hours post implementation. The proportion of patients categorised into rule-out, observational zone and rule-in was 62.9%, 16.2% and 20.9%, respectively. Safety of discharge after rule-out The overall discharge rate was 58.5% (1476 of 2525 patients), and discharge rates increased from 53.9% to 62.8% post implementation (P<0.0001). Among the 1588 patients who were classified as rule-out, 76.1% (n=1209) were discharged directly from the ED. The baseline characteristics of patients admitted to hospital versus discharged from the ED are displayed in Table 2. Briefly, patients admitted to hospital were 15 years older (54 ± 17 vs. 69 ± 14, P<0.0001), more frequently had unstable angina (1.9% vs. 41.2%, P<0.0001), more often had hypertension, diabetes and hypercholesterolemia, a history of cardiovascular disease and higher GRACE scores (80 ± 27 vs. 106 ± 27, P<0.0001). Table 2. Baseline characteristics of patients admitted to hospital versus discharged from emergency department.

Safety of discharge after rule-out The overall discharge rate was 58.5% (1476 of 2525 patients), and discharge rates increased from 53.9% to 62.8% post implementation (P<0.0001). Among the 1588 patients who were classified as rule-out, 76.1% (n=1209) were discharged directly from the ED. The baseline characteristics of patients admitted to hospital versus discharged from the ED are displayed in Table 2. Briefly, patients admitted to hospital were 15 years older (54 ± 17 vs. 69 ± 14, P<0.0001), more frequently had unstable angina (1.9% vs. 41.2%, P<0.0001), more often had hypertension, diabetes and hypercholesterolemia, a history of cardiovascular disease and higher GRACE scores (80 ± 27 vs. 106 ± 27, P<0.0001). Table 2. Baseline characteristics of patients admitted to hospital versus discharged from emergency department. All rule-out (N=1588) Discharged (N=1209) Admitted (N=379) P value Age, years 57 ± 18 54 ± 17 69 ± 14 <0.001 Sex, male/female 874/714 646/563 228/151 0.02 GRACE score 86 ± 29 80 ± 27 106 ± 27 <0.001 Low 1153 (72.6%) 970 (80.0%) 183 (48.3%) <0.001 Intermediate 365 (22.9%) 208 (17.2%) 157 (41.4%) <0.001 High 48 (3.0%) 12 (1.0%) 36 (9.5%) <0.001 Final diagnoses UA 179 (11.3%) 23 (1.9%) 156 (41.2%) <0.001 NSTEMI 14 (0.9%) 1 (0.08%) 13 (3.4%) <0.001 STEMIa 0 0 0 NA Non-ACS 1395 (87.8%) 1185 (98.0%) 210 (55.4%) <0.001 Time from onset of symptoms to admission 0–3 hours 712 (44.8%) 502 (41.5%) 210 (55.4%) <0.001 3–6 hours 141 (8.9%) 65 (5.4%) 76 (20.1%) <0.001 >6 hours 731 (46.0%) 639 (52.9%) 92 (24.3%) <0.001 Unknown 404 (25.4%) 273 (22.6%) 131 (34.6%) <0.001 Length of ED stay (hours) 3.9 (2.8–5.3) 3.5 (2.7–4.8) 5 (3.9–6.5) <0.001 History Myocardial infarction 214 (13.5%) 131 (10.8%) 83 (21.9%) <0.001 Coronary bypass surgery 71 (4.5%) 36 (3.0%) 35 (9.2%) <0.001 Coronary angiography 580 (36.5%) 356 (29.4%) 224 (59.1%) <0.001 Coronary intervention 352 (22.2%) 204 (16.9%) 148 (39.1%) <0.001 Left ventricular dysfunction 318 (20.0%) 212 (17.5%) 106 (27.9%) <0.001 Risk factors Hypertension 943 (59.4%) 635 (52.5%) 308 (81.3%) <0.001 Hypercholesterolemia 638 (40.2%) 418 (34.6%) 220 (58.0%) <0.001 Diabetes mellitus 249 (15.7%) 127 (10.5%) 122 (32.2%) <0.001 Active smoker 388 (24.4%) 314 (25.9%) 74 (19.5%) 0.2 Family history 451 (28.4%) 340 (28.1%) 111 (29.3%) 0.003 Laboratory values eGFR (mL/min/1.73 m2) 87 ± 23 91 ± 21 73 ± 24 <0.001 Serum creatinine (mg/dL) 0.9 ± 0.3 0.9 ± 0.2 1.0 ± 0.4 <0.001 hsTnT at admission (ng/L) 7 (5–10) 6 (5–9) 11 (7–20) <0.001 hsTnT at 1 hour (ng/L) 6 (5–8) 6 (5–8) 8 (6–10– <0.001 Copeptin at admission (pmol/L) 4.7 (3.3–7.7) 4.3 (2.9–6.5) 7.1 (3.9-13.8) <0.001 Coronary angiography within 30 days 229/1588 (14.4%) 41/1209 (3.4%) 188/379 (49.6%) <0.001 <2 hours 2/229 (0.9%) 0/41 (0.0%) 2/188 (1.1%) 1 <6 hours 7/229 (3.1%) 1/41 (2.4%) 6/188 (3.2%) 0.8 <24 hours 44/229 (19.2%) 2/41 (4.9%) 42/188 (22.3%) 0.02 <72 hours 132/229 (57.6%) 9/41 (22.0%) 123/188 (65.4%) <0.001 <96 hours 132/229 (57.6%) 9/41 (22.0%) 123/188 (65.4%) <0.001 Percutaneous coronary int

6 points. NSTEMI was ruled out based on two serial hsTnT values 2 hours apart (both 12 ng/L). Investigations identified urinary tract infection and mild hyponatremia. Work-up revealed chest radiography suspicious for lung cancer. The patient left hospital at her own request despite a recommendation for hospitalisation. Figure 2. Kaplan–Meier estimates of 30-day mortality in patients with rule-out of acute myocardial infarction by hospital admission: discharged from emergency department (blue) and admitted to hospital (red). In Cox regression analysis, several variables listed in Table 3 were independently associated with a lower probability of discharge. These variables included unstable angina, dyspnoea as the primary presenting symptom, history of diabetes, intermediate or high GRACE score, hsTnT concentration above the 99th percentile and history of left ventricular dysfunction. Table 3. Independent factors associated with a lower probability of discharge. Hazard ratio 95% CI P value Age (for 1 year higher) 0.9881 0.9742-1.0022 0.0984 Diabetes mellitus 0.6019 0.3977-0.9108 0.0163 Previous myocardial infarction 1.0932 0.6456-1.8510 0.7403 Unstable angina 0.0232 0.0133-0.0402 <0.0001 Dyspnoea as primary presenting symptom 0.3136 0.1927-0.5105 <0.0001 Intermediate or high GRACE score 0.6012 0.4134-0.8743 0.0078 Baseline hsTnT ≥14 ng/L 0.2830 0.1778-0.4507 <0.0001 Left ventricular dysfunction 0.4640 0.2863-0.7521 0.0018 High or very high crowding 0.8785 0.5883-1.3118 0.5265 Study period 1 0.8650 0.6073-1.2321 0.4217 hsTnT: high-sensitivity troponin T.

-0.5105 <0.0001 Intermediate or high GRACE score 0.6012 0.4134-0.8743 0.0078 Baseline hsTnT ≥14 ng/L 0.2830 0.1778-0.4507 <0.0001 Left ventricular dysfunction 0.4640 0.2863-0.7521 0.0018 High or very high crowding 0.8785 0.5883-1.3118 0.5265 Study period 1 0.8650 0.6073-1.2321 0.4217 hsTnT: high-sensitivity troponin T. Secondary endpoints Length of stay in ED The median length of stay (LoS) in the ED was 3.9 (2.8–5.3) hours for the entire study population, and significantly shorter for patients discharged from the ED vs. admitted to hospital (3.5 (2.7–4.8) vs. 5.0 (3.9–6.5) hours, P<0.001). The LoS grouped by different rule-out protocols is displayed in Figure 3. Median LoS was 2.9 hours with the single hsTnT less than LoD, 3.2 hours with the ESC 0/1-hour and 5.3 hours with the 0/3-hour protocol. Figure 3. Length of stay in the emergency department by protocol for rule-out of acute myocardial infarction. Safety of discharge within 30 days and one year by diagnostic categories At 30 days, total all-cause mortality across all diagnostic categories at 30 days was 1.9%. Mortality rates were 0.4%, 1.2% and 6.3% for rule-out, observe and rule-in (Figure 4). Figure 4. Kaplan–Meier estimates of 30-day mortality by diagnostic rule classification: rule-out (blue), observational zone (orange) and rule-in (red). At one year, the total all-cause mortality rate across diagnostic rules was 5.7%, with corresponding rates of 2.2%, 6.1% and 16.1% for rule-out, observe and rule-in (Supplementary Figure 1).

Figure 4. Kaplan–Meier estimates of 30-day mortality by diagnostic rule classification: rule-out (blue), observational zone (orange) and rule-in (red). At one year, the total all-cause mortality rate across diagnostic rules was 5.7%, with corresponding rates of 2.2%, 6.1% and 16.1% for rule-out, observe and rule-in (Supplementary Figure 1). Misclassifications Among the 2525 patients, a total of 4872 hsTnT samples (median 2 (2–2), min-max 1–4) were collected. Beyond the minimally required diagnostic set of hsTnT measurement(s), 944 additional samples were collected (one additional sample in 846 cases and two additional samples in 49 cases), allowing a reclassification of patients from the rule-out category to the rule-in category, and from the observational zone to the rule-in zone. There were 19 reclassifications from rule-out to rule-in including five cases of missed NSTEMI and 12 reclassifications from the observational zone to rule-in including four missed NSTEMI. Selected clinical, laboratory, angiographic parameters and revascularisation procedures, disposition and outcomes are displayed in Supplementary Tables 1 and 2. All-cause mortality by baseline hsTnT As displayed in Supplementary Figure 2, all-cause mortality increased in proportion to the baseline hsTnT concentration demonstrating no death among patients presenting with a hsTnT of less than 5 ng/L and the highest all-cause mortality among those presenting with a hsTnT of 52 ng/L or greater.

ality by baseline hsTnT As displayed in Supplementary Figure 2, all-cause mortality increased in proportion to the baseline hsTnT concentration demonstrating no death among patients presenting with a hsTnT of less than 5 ng/L and the highest all-cause mortality among those presenting with a hsTnT of 52 ng/L or greater. Additional findings Kaplan–Meier survival plots for all-cause death at 30 days and one year according to different diagnostic rules for rule-in are illustrated in Supplementary Figures 3 and 4. After rule-out of MI (n=1588 patients), a total of 379 patients (23.9%) were admitted to hospital. The use of diagnostic work-up during index admission and subsequent 30 days following discharge is displayed in Supplementary Table 3, split by study period. Supplementary Figure 5 shows the steep uptake of the ESC 0/1-hour algorithm at the cost of the ESC 0/3-hour algorithm starting immediately after the official implementation date (1 January 2017 – 6 months after the start of the study). Small changes of implementation rates before the official implementation were presumably motivated by the 2015 ESC guideline recommendation on the usefulness of the ESC 0/1-hour algorithm as an alternative to the ESC 0/3-hour algorithm.1

cial implementation date (1 January 2017 – 6 months after the start of the study). Small changes of implementation rates before the official implementation were presumably motivated by the 2015 ESC guideline recommendation on the usefulness of the ESC 0/1-hour algorithm as an alternative to the ESC 0/3-hour algorithm.1 Additional blood sampling beyond the minimal diagnostic set was obtained in 30% of the patients and reflects clinical practice, in which decisions to extend observation or continue blood sampling is left at the discretion of the physician, rather than on the protocol of a clinical trial. The additional measurements allowed us to obtain information on potential misclassifications, e.g. transition from rule-out to rule-in. Such reclassifications – from rule-out to rule-in – occurred in 19 patients, of whom five patients would have a missed NSTEMI diagnosis. In these five cases, maximal hsTnT values were low (maximal hsTnT in the ED ranging from 21 to 29 ng/L) suggesting a very small infarct size that would have been missed by a conventional sensitive and probably also by a contemporary sensitive cardiac troponin assay. Consistently, only two of the five patients with theoretically missed MI underwent PCI and there were no mortality events at 30 days among these five patients. Additional information on outcomes, diagnostic rules, timing of coronary angiography, rates of revascularisation are provided in Supplementary Table 4.

Additional blood sampling beyond the minimal diagnostic set was obtained in 30% of the patients and reflects clinical practice, in which decisions to extend observation or continue blood sampling is left at the discretion of the physician, rather than on the protocol of a clinical trial. The additional measurements allowed us to obtain information on potential misclassifications, e.g. transition from rule-out to rule-in. Such reclassifications – from rule-out to rule-in – occurred in 19 patients, of whom five patients would have a missed NSTEMI diagnosis. In these five cases, maximal hsTnT values were low (maximal hsTnT in the ED ranging from 21 to 29 ng/L) suggesting a very small infarct size that would have been missed by a conventional sensitive and probably also by a contemporary sensitive cardiac troponin assay. Consistently, only two of the five patients with theoretically missed MI underwent PCI and there were no mortality events at 30 days among these five patients. Additional information on outcomes, diagnostic rules, timing of coronary angiography, rates of revascularisation are provided in Supplementary Table 4. Discussion There is a striking contrast between the excellent diagnostic performance of fast protocols,2–11 the consistent findings for lower observation times in the ED,8,19,21 relevant cost savings26–29 and the limited adoption of fast diagnostic protocols despite an urgent need to decongest crowded EDs.12,30 Uncertainties that reduce the enthusiasm of physicians to implement single biomarker or accelerated protocols are mostly driven by the fear of litigation in case of missed MI or death.13 Previously, warnings were expressed that clinicians should apply the 1-hour algorithm with caution and only in low-risk patients,14,15 and that decisions should rather be based on rising or falling patterns of troponin than on single cut-off values.14 In support of the former, 2014 American College of Cardiology (ACC)/American Heart Association (AHA) guidelines16 cautioned against early troponin testing given that some values may not become abnormal for up to 12 hours. However, at the time of publication, high sensitivity tests were not available in the USA. This fear is further fuelled by the paucity of evidence from randomised trials evaluating the safety of discharge using accelerated protocols and hsTn assays, as well as issues to extrapolate findings derived from observational studies that enrolled patients with higher pre-test probabilities for an ACS. Currently, most evidence from randomised trials has focused on the implementation of hsTn in combination with validated clinical scores,18,19 a dual biomarker strategy combining copeptin with cardiac troponin,8 or a discharge of low-risk patients based on a normal hsTnI measurement 2 hours apart, together with a normal ECG and either a TIMI score of 1 point or less,9,10,21 or a low EDACS score.20 A large pre and post-implementation study on 31,332 patients with suspected ACS demonstrated a reduced LoS and increased proportions of patients discharged from the ED within 6 hours, without an adverse event when clinical pathways were correctly applied.21

TIMI score of 1 point or less,9,10,21 or a low EDACS score.20 A large pre and post-implementation study on 31,332 patients with suspected ACS demonstrated a reduced LoS and increased proportions of patients discharged from the ED within 6 hours, without an adverse event when clinical pathways were correctly applied.21 The data from this large study confirm previous observations and add information on feasibility, efficacy and safety of discharge using the ESC 0/3-hour but most importantly the ESC 0/1-hour rule-out protocols in an all-comers registry with broad inclusion criteria. The study provides several important findings.

TIMI score of 1 point or less,9,10,21 or a low EDACS score.20 A large pre and post-implementation study on 31,332 patients with suspected ACS demonstrated a reduced LoS and increased proportions of patients discharged from the ED within 6 hours, without an adverse event when clinical pathways were correctly applied.21 The data from this large study confirm previous observations and add information on feasibility, efficacy and safety of discharge using the ESC 0/3-hour but most importantly the ESC 0/1-hour rule-out protocols in an all-comers registry with broad inclusion criteria. The study provides several important findings. First, our findings suggest that the ESC recommended 0/1-hour algorithm can be implemented as the predominant diagnostic algorithm, and is clinically feasible. Second, discharge after rule-out is safe with or without the use of the GRACE score, with 30-day all-cause mortality rates less than 0.1%. In this cohort, mortality was defined as all-cause death, and the only fatality occurred 13 days after discharge due to lung cancer. There was no statistically significant difference regarding the safety of the ESC 0/1-hour algorithm compared to the standard ESC 0/3-hour protocol. Even the rule-out strategy based on a single low baseline hsTnT below the LoD (<5 ng/L), or a normal hsTnT (≥5 but ≤14 ng/L) together with a normal copeptin (<10 pmol/L) at presentation were not associated with a higher risk of all-cause mortality after discharge among 1309 patients primarily discharged from the ED. Third, lengths of stay in the ED were closely related to the specific protocols and were as low as a median of 2.9 hours using a single rule-out hsTnT, 3.2 hours using the 0/1-hour protocol and 5.3 hours with the 0/3-hour protocol. Likewise, the interval between the first and the second blood draw shortened by 45 minutes after implementation of the ESC 0/1-hour protocol. Fourth, there is no overuse of resources for diagnostic work-up in order to facilitate earlier discharge, which is important as others14,15 have warned about an increase of unnecessary and costly diagnostic work-up before discharge. Conversely, utilisation rates of stress testing, imaging and coronary angiography were in the range of 10–20% among patients discharged, and thus very similar to rates of investigations performed during the index visit or 30-day follow-up reported by Mokhtari et al.7 Among 1038 patients with suspected ACS, exercise ECG, echocardiography, computed tomography coronary angiography and coronary angiography were performed in 13.8%, 18.3%, 0.6% and 13.3% of patients, respectively.

to rates of investigations performed during the index visit or 30-day follow-up reported by Mokhtari et al.7 Among 1038 patients with suspected ACS, exercise ECG, echocardiography, computed tomography coronary angiography and coronary angiography were performed in 13.8%, 18.3%, 0.6% and 13.3% of patients, respectively. Interestingly, rates of utilisation of stress testing and imaging before discharge vary considerably between studies showing low rates19,26,31 or very high utilisation rates of investigations31 suggesting different adherence to guideline recommendations1 and local practice across institutions and continents.32

to rates of investigations performed during the index visit or 30-day follow-up reported by Mokhtari et al.7 Among 1038 patients with suspected ACS, exercise ECG, echocardiography, computed tomography coronary angiography and coronary angiography were performed in 13.8%, 18.3%, 0.6% and 13.3% of patients, respectively. Interestingly, rates of utilisation of stress testing and imaging before discharge vary considerably between studies showing low rates19,26,31 or very high utilisation rates of investigations31 suggesting different adherence to guideline recommendations1 and local practice across institutions and continents.32 The consistencies and differences of our study compared to other observational studies require detailed discussion. Our study is distinct to other observational studies regarding several aspects. First, we enrolled a consecutive cohort of all-comers with a broad spectrum of presenting symptoms without exclusion for older age, heart failure or chronic kidney disease, with the consequence of a further lowering of clinical specificity. Accordingly, rates of NSTEMI within the rule-in and observational zone were 57.5% and 3.2%, and thus considerably lower than reported in other observational trials.2–4,6,7 Second, LoS in the ED in our study are very short, ranging from 2.9 hours with rule-out based on a hsTnT less than LoD to 5.3 hours with the 0/3-hour diagnostic algorithm. In the literature, reported LoS vary between 4 hours,8,26 5.3 hours,38 5.5 hours,39 6.4 hours19 to 26.3 hours,31 with the shortest observation times (median LoS 4 hours in both studies) using the dual biomarker strategy,8 or the 0/1-hour rule-out in the TRAPID study.26 Conversely, the longest LoS have been reported in an Australasian cohort,31 in which LoS was associated with high utilisation rates of investigations before discharge. As compared to most other observational studies in which treating physicians were blinded to the investigational hsTn results and patients were not managed in accordance with these results, physicians in the present study reacted on the fast protocols thus accelerating the disposition of patients.

ons before discharge. As compared to most other observational studies in which treating physicians were blinded to the investigational hsTn results and patients were not managed in accordance with these results, physicians in the present study reacted on the fast protocols thus accelerating the disposition of patients. Third, post-implementation discharge rates after rule-out were 62.8% and thus at the upper end of the reported range. Discharge rates in other observational or randomised studies are heterogeneous with low discharge rates among low-risk patients between 18.4% and 26%,9,18,21 intermediate discharge rates between 42.3% and 55%19,39,40 and high discharge rates between 67.8% and 72%.8,38,41 Only one study42 reported significantly longer hospitalisation stays by 35%, fewer early discharges after a negative result (7% vs. 22%, P=0.0001), more coronary angiograms (77% vs. 55%, P=0.0001) and revascularisations (45% vs. 31%, P=0.0001) after implementation of hsTn assays by 35%.

es between 67.8% and 72%.8,38,41 Only one study42 reported significantly longer hospitalisation stays by 35%, fewer early discharges after a negative result (7% vs. 22%, P=0.0001), more coronary angiograms (77% vs. 55%, P=0.0001) and revascularisations (45% vs. 31%, P=0.0001) after implementation of hsTn assays by 35%. Our study findings are consistent with other observational studies with regard to important aspects. First, the prevalence of NSTEMI was 13.1% in our study and was thus similar to reported rates between 7.0% and 23.3%,33,34 with an overall prevalence in the pooled population of 9241 patients of 15.4%.33 Second, total mortality rates at 30 days and one year across all diagnostic categories were 1.9% and 5.7%, respectively, indicating enrolment of a risk that is at least as high as in other observational studies,2–5,21 and the need for risk stratification before discharge, even among patients ruled-out for MI. Along with others, total mortality rates in the TRAPID acute MI study,4 a study that used a similar protocol, were lower than in our cohort, highlighting the broader inclusion of consecutive patients. At 30 days, total mortality rates were 0.2%, ranging between 0.1%, 0.4% and 0.5% for those ruled-out, in the observational zone, and ruled-in. At one year, total mortality rates were 4.1%, ranging between 0.7%, 9.7% and 9.3% for those ruled-out, in the observational zone and ruled-in, respectively. Consistently, a recent meta-analysis on 9241 patients in 11 cohort studies noted no death occurring within 30 days among low-risk patients defined as hsTnT of less than 5 ng/L without ECG changes.33 Thus it appears that fast rule-out protocols allow safe classification and risk stratification of patients with suspected ACS, with or without the use of clinical scores, enabling the safe discharge of a low-risk cohort, with an all-cause mortality rate below 0.01%. Third, rates of missed MI following fast rule-out were 0.3% (five of 1588 patients), which is at the lower end of reported rates between 0.2% and 1.2% using hsTn and fast protocols.4,5,35,36 There were no deaths or MI, all underwent coronary angiography, but only two patients (10.6%) received a PCI within 30 days, supporting the claim of safe discharge after rule-out using hsTn and accelerated protocols.

ich is at the lower end of reported rates between 0.2% and 1.2% using hsTn and fast protocols.4,5,35,36 There were no deaths or MI, all underwent coronary angiography, but only two patients (10.6%) received a PCI within 30 days, supporting the claim of safe discharge after rule-out using hsTn and accelerated protocols. Finally, rates for the diagnostic categories rule-out, rule-in and observe were 62.9%, 20.9% and 16.2%, which is very consistent with other observational studies on the 0/1-hour protocol using hsTnT or hsTnI.37 Of interest, the inclusion of patients with a broad spectrum of symptoms with higher prevalence of myocardial injury unrelated to ischaemia did not augment the observational zone but – as expected – fewer patients received a diagnosis of NSTEMI within the rule-in and observational zone.2–4 The yield of NSTEMI was similar to Mokhtari et al.7 reporting 50% NSTEMI among rule-in but only 1.9% in the observational zone. These findings extend previous studies on the safety of discharge of low-risk patients after rule-out of MI using hsTnT in general and the ESC 0/1-hour protocol in particular. Our findings might increase the confidence of physicians to apply accelerated diagnostic protocols and are likely to speed up the slow implementation of accelerated diagnostic algorithms in crowded EDs.

scharge of low-risk patients after rule-out of MI using hsTnT in general and the ESC 0/1-hour protocol in particular. Our findings might increase the confidence of physicians to apply accelerated diagnostic protocols and are likely to speed up the slow implementation of accelerated diagnostic algorithms in crowded EDs. Limitations Our findings suggest that the use of a fast rule-out using hsTnT in a 0/1-hour protocol allows safe discharge without the need for extensive pre or post-discharge work-up, provided the residual risk is estimated appropriately. We identified variables that were independently associated with a hospital admission despite rule-out. These variables included individual risk factors, hsTnT level and a high GRACE score. However, an extremely low all-cause mortality rate after discharge prohibits any meaningful speculations on the utility of clinical scores in the era of hsTn assays. Chapman et al.43 recently reported that when clinical risk scores were applied in the High-STEACS pathway at thresholds to rule out MI that were considerably lower than the 99th percentile value, the proportion of patients ruled out halved without improving safety. In our institution, the use of the GRACE score, which represents a validated and objective tool for estimation of risk for death or MI,25 was not mandatory for the decision to discharge, highlighting the importance of the physicians’ experience to estimate future risk. Accordingly, our liberal discharge policy might not be transferrable to other EDs with lower levels of physicians’ experience.

and objective tool for estimation of risk for death or MI,25 was not mandatory for the decision to discharge, highlighting the importance of the physicians’ experience to estimate future risk. Accordingly, our liberal discharge policy might not be transferrable to other EDs with lower levels of physicians’ experience. Second, we cannot exclude under-reporting of diagnostic work-up within 30 days that might have escaped our follow-up. Third, as a consequence of early discharge after rule-out without a gold standard diagnosis based on a 6-hour algorithm, we cannot exclude the later development of rule-in or NSTEMI. In order to reduce numbers of false negatives, we strictly followed ESC guideline recommendations and refrained from rule-out and discharge of patients based on a single hsTnT less than LoD, if the interval from the last pain episode was unclear, equivocal, or shorter than 3 hours. Moreover, physicians were instructed to continue hsTnT measurement in patients with refractory or recurrent symptoms, a persistent high level of clinical suspicion for evolving MI. Additional blood draws beyond the adequate diagnostic set were collected in 30% of patients and allowed us to identify 19 of 1588 patients (1.2%) who changed the diagnostic category from rule-out to rule-in. Among these, five patients had a missed NSTEMI (0.3%), characterised by small hsTnT between 17 and 23 ng/L, none died within 30 days, all received coronary angiography but only two patients required PCI. The numbers of missed MI wrongly classified as rule-out are very similar in other observational studies ranging from 7/813 (0.9%) in the TRAPID study,4 to 5/2488 (0.2%) in the pooled cohorts of APACE and BACC trials,36 20 of 2533 (0.8%) in the APACE study,35 and between four of 342 (1.2%) and 12 of 2160 (0.6%) in two High-STEACS study reports.5,35

gly classified as rule-out are very similar in other observational studies ranging from 7/813 (0.9%) in the TRAPID study,4 to 5/2488 (0.2%) in the pooled cohorts of APACE and BACC trials,36 20 of 2533 (0.8%) in the APACE study,35 and between four of 342 (1.2%) and 12 of 2160 (0.6%) in two High-STEACS study reports.5,35 Finally, the present study was performed in a single centre – the ED of the department of cardiology, which is certified as a CPU. Our ED is constantly supervised by a cardiologist and has a volume of patients that is representable compared to other CPUs. The number inclines to more than 40 patients per day when patients admitted to the entire internal medicine ED are also considered. While we cannot fully exclude a referral bias, patients with suspected ACS are almost exclusively referred to the CPU. This has the advantage to evaluate consecutive patients across almost the entire spectrum of risk. Our study findings are consistent regarding the prevalence of ACS and rates of death or MI during follow-up as compared with other European observational studies.3,6,7 Given that our findings were obtained in a German CPU, led by a cardiologist and certified by the German Cardiac Society only after operating criteria were met, our results cannot be automatically generalised to other EDs outside Germany, unless they provide similar infrastructural characteristics or tight interdisciplinary collaboration with cardiologists.

n a German CPU, led by a cardiologist and certified by the German Cardiac Society only after operating criteria were met, our results cannot be automatically generalised to other EDs outside Germany, unless they provide similar infrastructural characteristics or tight interdisciplinary collaboration with cardiologists. Conclusions Implementation of the ESC 0/1-hour algorithm is feasible and is associated with very low mortality of discharged patients after rule-out. Furthermore, the ESC 0/1-hour algorithm is associated with a significantly shorter length of ED stay than the ESC 0/3-hour protocol. After implementation of the ESC 0/1-hour algorithm, discharge rates increased significantly, without excessive use of diagnostic resources. Supplemental Material Supplemental_Material – Supplemental material for RAPID-CPU: a prospective study on implementation of the ESC 0/1-hour algorithm and safety of discharge after rule-out of myocardial infarction Click here for additional data file. Supplemental material, Supplemental_Material for RAPID-CPU: a prospective study on implementation of the ESC 0/1-hour algorithm and safety of discharge after rule-out of myocardial infarction by Kiril M Stoyanov, Hauke Hund, Moritz Biener, Jochen Gandowitz, Christoph Riedle, Julia Löhr, Matthias Mueller-Hennessen, Mehrshad Vafaie, Hugo A Katus and Evangelos Giannitsis in European Heart Journal: Acute Cardiovascular Care

e ESC 0/1-hour algorithm and safety of discharge after rule-out of myocardial infarction by Kiril M Stoyanov, Hauke Hund, Moritz Biener, Jochen Gandowitz, Christoph Riedle, Julia Löhr, Matthias Mueller-Hennessen, Mehrshad Vafaie, Hugo A Katus and Evangelos Giannitsis in European Heart Journal: Acute Cardiovascular Care The authors would like to thank the patients who participated in the study and the staff of the ED. They also express their gratitude to the study nurses Heidi Deigentasch, Melanie Hütter and Elisabeth Mertz. Conflict of interest: MB reports grants and non-financial support from AstraZeneca, non-financial support from Thermo Fisher. MMH reports grants and speaker honoraria from Roche Diagnostics; grants and non-financial support from BRAHMS Thermo Scientific. HAK received honoraria for lecturers from Roche Diagnostics, AstraZeneca, Bayer Vital, Daiichi-Sankyo, and held a patent on cTnT that has expired. EG received honoraria for lectures from Roche Diagnostics, AstraZeneca, Bayer Vital, Daiichi-Sankyo, Eli Lilly Deutschland. He serves as a consultant for Roche Diagnostics, BRAHMS Thermo Fisher, Boehringer Ingelheim, and has received research funding from BRAHMS Thermo Fisher, Roche Diagnostics, Bayer Vital and Daiichi Sankyo. All other authors have no conflicts of interest to declare. Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: this work was supported by a research grant from Roche Diagnostics International Ltd.

Conflict of interest: MB reports grants and non-financial support from AstraZeneca, non-financial support from Thermo Fisher. MMH reports grants and speaker honoraria from Roche Diagnostics; grants and non-financial support from BRAHMS Thermo Scientific. HAK received honoraria for lecturers from Roche Diagnostics, AstraZeneca, Bayer Vital, Daiichi-Sankyo, and held a patent on cTnT that has expired. EG received honoraria for lectures from Roche Diagnostics, AstraZeneca, Bayer Vital, Daiichi-Sankyo, Eli Lilly Deutschland. He serves as a consultant for Roche Diagnostics, BRAHMS Thermo Fisher, Boehringer Ingelheim, and has received research funding from BRAHMS Thermo Fisher, Roche Diagnostics, Bayer Vital and Daiichi Sankyo. All other authors have no conflicts of interest to declare. Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: this work was supported by a research grant from Roche Diagnostics International Ltd. Supplemental material: Supplemental material for this article is available online.

Introduction Patients with symptoms of a possible acute coronary syndrome (ACS) are frequently seen at the emergency department (ED).1–3 Physicians strive to achieve an effective but safe diagnostic work-up, as misdiagnoses can have serious consequences.2–4 The HEART score, a clinical tool for rapid risk stratification, has been proposed to improve decision making in patients suspected of ACS.5–8 Based on history, electrocardiogram (ECG), age, risk factors and initial troponin levels, the HEART score provides the physician with recommendations for further management. Recent studies suggest that the HEART score permits safe discharge of a considerable number of patients, effectively reducing downstream testing.7 At the same time, several randomised trials have shown that coronary computed tomography angiography (CCTA) allows safe and early discharge from the ED providing valuable prognostic information as well.5,9,10 However, CCTA is a costly test and requires radiation exposure to the patient. Combining the HEART score with CCTA may provide a more efficient diagnostic work-up, where CCTA can be reserved for a subset of patients. The aim of this study was to investigate the diagnostic value and efficiency of the HEART score before CCTA in patients suspected of ACS in the ED.

radiation exposure to the patient. Combining the HEART score with CCTA may provide a more efficient diagnostic work-up, where CCTA can be reserved for a subset of patients. The aim of this study was to investigate the diagnostic value and efficiency of the HEART score before CCTA in patients suspected of ACS in the ED. Methods Patient population We conducted a secondary analysis of two prospective studies of patients presenting to the ED with symptoms suggestive of ACS. The methods, including study designs, inclusion and exclusion criteria have previously been published.9,11 In the current analysis, we included patients who underwent CCTA of diagnostic image quality. Both studies were performed according to the principles of the Declaration of Helsinki, approved by the local institutional review boards and all patients provided written informed consent. CCTA Image acquisition was performed on 64-slice or newer computed tomography systems, using ECG-synchronised axial or spiral scan protocols combined with radiation minimising measures, depending on local practices, available technology, and patient characteristics. Results of CCTA were reported by certified radiologists with a minimum of two years of experience reading CCTA. The presence of coronary plaque and the degree of stenosis was assessed for each evaluable coronary segment. The degree of stenosis was quantified as: no stenosis, ≤50% stenosis (non-obstructive plaque) or >50% stenosis (obstructive plaque).

certified radiologists with a minimum of two years of experience reading CCTA. The presence of coronary plaque and the degree of stenosis was assessed for each evaluable coronary segment. The degree of stenosis was quantified as: no stenosis, ≤50% stenosis (non-obstructive plaque) or >50% stenosis (obstructive plaque). HEART score The HEART score, a clinical risk tool for rapid risk stratification of patients with acute chest pain, was calculated for each patient. The score consists of five components: History, ECG, Age, Risk factors and Troponin. Each of these components may be scored with 0, 1 or 2 points with a maximum score of 10 points.8 Detailed information on the composition of the HEART score and how each component is scored can be found in Supplemental Material Table S1. Information regarding all components were retrieved from hospital records from the day of index presentation. As suggested by the original authors, patients were also categorised as: low risk (HEART ≤3), intermediate risk (HEART 4–6) and high risk (HEART ≥7).8

t is scored can be found in Supplemental Material Table S1. Information regarding all components were retrieved from hospital records from the day of index presentation. As suggested by the original authors, patients were also categorised as: low risk (HEART ≤3), intermediate risk (HEART 4–6) and high risk (HEART ≥7).8 Clinical endpoints The primary outcome was occurrence of major adverse cardiac events (MACEs) within 30 days by analogy with prior publications on the HEART score;12,13 a composite of all-cause mortality, ACS or coronary revascularisation (emergent or elective within 30 days). ACS was defined as acute myocardial infarction or unstable angina according to the universal definition of acute myocardial infarction.14,15 All clinical endpoints were adjudicated by two cardiologists who independently reviewed medical records of patients. The result of the CCTA was blinded to the cardiologists performing the event adjudication.

cardial infarction or unstable angina according to the universal definition of acute myocardial infarction.14,15 All clinical endpoints were adjudicated by two cardiologists who independently reviewed medical records of patients. The result of the CCTA was blinded to the cardiologists performing the event adjudication. Statistical analysis Continuous data are presented as mean±standard deviation (SD) or median (interquartile ranges), and categorical data are presented as proportions (percentages). Differences between independent groups were compared using analysis of variance or the Kruskal-Wallis test for continuous variables, and the Fisher’s exact test or the Pearson’s chi-square test for categorical variables. Parameters of diagnostic accuracy, i.e. sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV) for the prediction of 30-day MACEs were calculated with their corresponding 95% confidence intervals using exact binomial confidence intervals. When evaluating the CCTA, >50% stenosis was considered a positive test. Areas under the curve (AUCs) were calculated and compared using the test of DeLong et al.16 All statistical analyses were performed using MedCalc Statistical Software version 18.10 (MedCalc Software bvba, Ostend, Belgium) and SPSS version 24.0 (IBM, Armonk, New York, USA). All tests were two-tailed and a p-value <0.05 was considered statistically significant.

ted and compared using the test of DeLong et al.16 All statistical analyses were performed using MedCalc Statistical Software version 18.10 (MedCalc Software bvba, Ostend, Belgium) and SPSS version 24.0 (IBM, Armonk, New York, USA). All tests were two-tailed and a p-value <0.05 was considered statistically significant. Results Baseline characteristics and clinical endpoints Of 500 patients included in the Better Evaluation of Acute Chest Pain with Computed Tomography Angiography (BEACON) trial, 229 patients underwent CCTA and had diagnostic image quality.9 Additionally, 111 patients in the Rotterdam Acute Chest Pain cohort underwent CCTA of diagnostic image quality.11 In total, 340 patients met eligibility and were included for the current study (Figure 1). The mean age was 56±10 years and the proportion of women was 44.7%. MACEs occurred in 45 (13.2%) patients within 30 days (Table 1). The adjudicated diagnosis of ACS was established in 42 (12.4%) patients: 27 (7.9%) had myocardial infarction and 15 (4.4%) had unstable angina pectoris. Coronary revascularisation was performed in 38 (11.2%) patients. Seven patients with an adjudicated diagnosis of ACS did not undergo revascularisation within 30 days; Of these, two were managed medically and underwent revascularisation after 30 days and five were found to have no significant stenosis on invasive coronary angiography (ICA). Additionally, three patients underwent elective percutaneous coronary intervention (PCI) for stable angina pectoris. Cardiac troponins were available in all patients. Troponins were measured with high-sensitive troponin assays in 180 (53%) patients, of whom 177 patients with the high-sensitive Troponin T assay (Roche diagnostics). In the remaining 160 (47%) patients cardiac troponins were measured with conventional troponin assays. Supplemental Material Table S2 lists all troponin assays used, their characteristics and the algorithm in which they were implemented.

of whom 177 patients with the high-sensitive Troponin T assay (Roche diagnostics). In the remaining 160 (47%) patients cardiac troponins were measured with conventional troponin assays. Supplemental Material Table S2 lists all troponin assays used, their characteristics and the algorithm in which they were implemented. Figure 1. Flow diagram shows the enrolment process for the study population. CCTA: coronary computed tomography angiography. Table 1. Baseline patient characteristics.

of whom 177 patients with the high-sensitive Troponin T assay (Roche diagnostics). In the remaining 160 (47%) patients cardiac troponins were measured with conventional troponin assays. Supplemental Material Table S2 lists all troponin assays used, their characteristics and the algorithm in which they were implemented. Figure 1. Flow diagram shows the enrolment process for the study population. CCTA: coronary computed tomography angiography. Table 1. Baseline patient characteristics. Total (n=340) HEART Low risk (0–3) (n=119) HEART intermediate (4–6) (n=193) HEART high risk (7–10) (n=28) p-Value Mean age, years 55.6±10.1 51.3±9.4 57.2±9.5 63.3±9.9 <0.001 Women 152 (44.7) 52 (43.7) 90 (46.6) 10 (35.7) 0.53 Cardiovascular risk factors Hypertension 170 (50.0) 39 (32.8) 110 (57.0) 21 (75.0) <0.001 Dyslipidaemia 116 (34.1) 14 (11.8) 80 (41.5) 22 (78.6) <0.001 Diabetes mellitus 44 (12.9) 5 (4.1) 32 (16.6) 7 (25.0) <0.001 Smoking 131 (38.5) 40 (33.6) 78 (40.4) 13 (46.4) 0.33 Family history positive for CAD 139 (40.9) 43 (36.1) 82 (42.5) 14 (50.0) 0.32 Prior atherosclerotic disease 40 (11.8) 4 (3.4) 28 (14.5) 8 (28.6) <0.001 Blood pressure Systolic 141.7±21.1 137.1±18.6 143.9±21.2 145.4±27.0 0.01 Diastolic 81.9±13.6 81.5±12.7 82.7±14.3 78.7±12.1 0.32 CCTA assessment for CAD No stenosis 151 (44.4) 74 (62.2) 76 (39.4) 1 (3.6) <0.001 1–50% stenosis 103 (30.3) 32 (26.9) 67 (34.7) 4 (14.3) 0.05 >50% stenosis 86 (25.3) 13 (10.9) 50 (25.9) 23 (82.1) <0.001 Radiation dose, mSv 4.9 (3.1–8.8) 4.5 (2.7–8.1) 5.3 (3.3–9.4) 4.7 (3.3–6.5) 0.05 Occurrence of MACEs within 30 days of index visit MACEs 30 days 45 (13.2) 4 (3.4) 24 (12.4) 17 (60.7) <0.001 All-cause mortality 1 (0.3) 0 (0) 0 (0) 1 (3.6) 0.08 ACS 42 (12.4) 3 (2.5) 23 (11.9) 16 (57.1) <0.001 Unstable angina 15 (4.4) 2 (1.7) 6 (3.1) 7 (25.0) <0.001 Myocardial infarction 27 (7.9) 1 (0.8) 17 (8.8) 9 (32.1) <0.001 Coronary revascularisation 38 (11.2) 4 (3.4) 21 (10.9) 13 (46.4) <0.001 ACS: acute coronary syndrome; CAD: coronary artery disease; CCTA: coronary computed tomography angiography; MACE: major adverse cardiac event; mSv: millisievert; SD: standard deviation.

yocardial infarction 27 (7.9) 1 (0.8) 17 (8.8) 9 (32.1) <0.001 Coronary revascularisation 38 (11.2) 4 (3.4) 21 (10.9) 13 (46.4) <0.001 ACS: acute coronary syndrome; CAD: coronary artery disease; CCTA: coronary computed tomography angiography; MACE: major adverse cardiac event; mSv: millisievert; SD: standard deviation. MACEs defined as all-cause mortality, ACS or coronary revascularisation. Values are mean±SD, median (interquartile ranges) or n (%). CCTA When assessed for coronary artery disease (CAD), 151 (44.4%) patients had no stenosis, 103 (30.3%) patients had 1–50% stenosis and 86 (25.3%) patients had >50% stenosis on CCTA (Table 1). The incidence of 30-day MACEs in patients with no stenosis, 1–50% stenosis and >50% stenosis was 0% (n=0), 1.9% (n=2) and 50% (n=43), respectively. Sensitivity, specificity, NPV and PPV of >50% stenosis on CCTA for the prediction of 30-day MACEs was 95.6% (84.9–99.5), 85.4% (80.9–89.2), 99.2% (97.0–99.8) and 50.0% (43.0–57.0), respectively. The AUC of >50% stenosis on CCTA for prediction of 30-day MACEs was 0.91 (0.87–0.93).

.9% (n=2) and 50% (n=43), respectively. Sensitivity, specificity, NPV and PPV of >50% stenosis on CCTA for the prediction of 30-day MACEs was 95.6% (84.9–99.5), 85.4% (80.9–89.2), 99.2% (97.0–99.8) and 50.0% (43.0–57.0), respectively. The AUC of >50% stenosis on CCTA for prediction of 30-day MACEs was 0.91 (0.87–0.93). HEART score The HEART score classified 119 (35.0%) patients as low-risk, 193 (56.8%) as intermediate-risk and 28 (8.2%) as high-risk. The incidence of 30-day MACEs in patients stratified as low-risk, intermediate-risk and high-risk was 3.4% (n=4), 12.4% (n=24), and 60.7% (n=17), respectively (Table 1). All patients (n=4) in the low-risk category with 30-day MACEs had a HEART score of three (Figure 2). Table 2 shows detailed characteristics of patients with a low HEART score (≤3) and MACEs within 30 days. All low-risk patients with 30-day MACEs had >50% stenosis on CCTA. Sensitivity, specificity, NPV and PPV of the HEART score for the prediction of 30-day MACEs at different cut-offs are shown in Table 3. The AUC of the HEART score for prediction of 30-day MACEs was 0.83 (0.78–0.87). Figure 2. Frequency of 30-day major adverse cardiac events (MACEs) according to HEART score. MACEs defined as all-cause mortality, acute coronary syndrome or coronary revascularisation. Table 2. Detailed characteristics of patients with a low HEART score (≤3) or coronary computed tomography angiography (CCTA)≤50% stenosis and major adverse cardiac events (MACEs) within 30 days.

Figure 2. Frequency of 30-day major adverse cardiac events (MACEs) according to HEART score. MACEs defined as all-cause mortality, acute coronary syndrome or coronary revascularisation. Table 2. Detailed characteristics of patients with a low HEART score (≤3) or coronary computed tomography angiography (CCTA)≤50% stenosis and major adverse cardiac events (MACEs) within 30 days. Patient Age, years Sex HEART Type of troponin assay Initial troponin Highest troponin MACE >50% stenosis on CCTA Additional information History ECG Age Risk factors Initial troponin Total 1 46 Male 2 0 1 0 0 3 TnT Negative Negative UA; PCI Yes Patient admitted for ICA after positive ExECG 2 63 Female 1 0 1 1 0 3 Hs-TnT 4 ng/l 4 ng/l UA; PCI Yes Patient admitted for ICA after obstructive plaque on CCTA 3 63 Female 1 0 1 1 0 3 Hs-TnT 5 ng/l 5 ng/l PCI Yes PCI after elective ICA 4 45 Male 2 0 1 0 0 3 TnT Negative 0.69 μg/l NSTEMI; PCI Yes Patient admitted for ICA after significant rise of troponin 5 69 Male 1 0 2 2 1 6 Hs-TnT 24 ng/l 30 ng/l MINOCA No Patient admitted for ICA. No significant stenosis detected during ICA ECG: electrocardiogram; ExECG: exercise stress electrocardiography; Hs-TnT: high-sensitivity troponin T; ICA: invasive coronary angiography; MACE: major adverse cardiac events; MINOCA: myocardial infarction with nonobstructive coronary artery disease; NSTEMI: non-ST segment elevation myocardial infarction; PCI: percutaneous coronary intervention; TnT: troponin T; UA: unstable angina. MACEs defined as all-cause mortality, acute coronary syndrome or coronary revascularisation.

Patient Age, years Sex HEART Type of troponin assay Initial troponin Highest troponin MACE >50% stenosis on CCTA Additional information History ECG Age Risk factors Initial troponin Total 1 46 Male 2 0 1 0 0 3 TnT Negative Negative UA; PCI Yes Patient admitted for ICA after positive ExECG 2 63 Female 1 0 1 1 0 3 Hs-TnT 4 ng/l 4 ng/l UA; PCI Yes Patient admitted for ICA after obstructive plaque on CCTA 3 63 Female 1 0 1 1 0 3 Hs-TnT 5 ng/l 5 ng/l PCI Yes PCI after elective ICA 4 45 Male 2 0 1 0 0 3 TnT Negative 0.69 μg/l NSTEMI; PCI Yes Patient admitted for ICA after significant rise of troponin 5 69 Male 1 0 2 2 1 6 Hs-TnT 24 ng/l 30 ng/l MINOCA No Patient admitted for ICA. No significant stenosis detected during ICA ECG: electrocardiogram; ExECG: exercise stress electrocardiography; Hs-TnT: high-sensitivity troponin T; ICA: invasive coronary angiography; MACE: major adverse cardiac events; MINOCA: myocardial infarction with nonobstructive coronary artery disease; NSTEMI: non-ST segment elevation myocardial infarction; PCI: percutaneous coronary intervention; TnT: troponin T; UA: unstable angina. MACEs defined as all-cause mortality, acute coronary syndrome or coronary revascularisation. Table 3. Predictive value of the HEART score for 30-day major adverse cardiac events (MACEs) at various cut-offs.

Patient Age, years Sex HEART Type of troponin assay Initial troponin Highest troponin MACE >50% stenosis on CCTA Additional information History ECG Age Risk factors Initial troponin Total 1 46 Male 2 0 1 0 0 3 TnT Negative Negative UA; PCI Yes Patient admitted for ICA after positive ExECG 2 63 Female 1 0 1 1 0 3 Hs-TnT 4 ng/l 4 ng/l UA; PCI Yes Patient admitted for ICA after obstructive plaque on CCTA 3 63 Female 1 0 1 1 0 3 Hs-TnT 5 ng/l 5 ng/l PCI Yes PCI after elective ICA 4 45 Male 2 0 1 0 0 3 TnT Negative 0.69 μg/l NSTEMI; PCI Yes Patient admitted for ICA after significant rise of troponin 5 69 Male 1 0 2 2 1 6 Hs-TnT 24 ng/l 30 ng/l MINOCA No Patient admitted for ICA. No significant stenosis detected during ICA ECG: electrocardiogram; ExECG: exercise stress electrocardiography; Hs-TnT: high-sensitivity troponin T; ICA: invasive coronary angiography; MACE: major adverse cardiac events; MINOCA: myocardial infarction with nonobstructive coronary artery disease; NSTEMI: non-ST segment elevation myocardial infarction; PCI: percutaneous coronary intervention; TnT: troponin T; UA: unstable angina. MACEs defined as all-cause mortality, acute coronary syndrome or coronary revascularisation. Table 3. Predictive value of the HEART score for 30-day major adverse cardiac events (MACEs) at various cut-offs. HEART score Number of patients ruled-out (%) Sensitivity (95% CI) Specificity (95% CI) PPV (95% CI) NPV (95% CI) ≥0 0 (0) 100.0 (92.1–100.0) 0.0 (0.0–1.2) 13.2 (13.2–13.2) NA <1 3 (0.9) 100.0 (92.1–100.0) 1.0 (0.2–2.9) 13.4 (13.2–13.5) 100.0 <2 15 (4.4) 100.0 (92.1–100.0) 5.1 (2.9–8.3) 13.9 (13.5–14.2) 100.0 <3 47 (13.8) 100.0 (92.1–100.0) 15.9 (11.9–20.6) 15.4 (14.7–16.0) 100.0 <4 119 (35.0) 91.1 (78.8–97.5) 39.0 (33.4–44.8) 18.6 (16.7–20.6) 96.6 (91.8–98.7) <5 205 (60.3) 80.0 (65.4–90.4) 66.4 (60.7–71.8) 26.7 (22.6–31.1) 95.6 (92.4–97.5) <6 271 (79.7) 68.9 (53.4–81.8) 87.1 (82.8–90.7) 44.9 (36.4–53.8) 94.8 (92.2–96.6) <7 312 (91.8) 37.8 (23.8–53.5) 96.3 (93.4–98.1) 60.7 (43.7–75.5) 91.0 (89.0–92.7) <8 332 (97.6) 15.6 (6.5–29.5) 99.7 (98.1–100.0) 87.5 (46.9–98.2) 88.6 (87.2–89.8) <9 339 (99.7) 2.2 (0.1–11.8) 100.0 (98.8–100.0) 100.0 87.0 (86.5–87.5) >10 340 (100) 0.0 (0.0–7.9) 100.0 (98.8–100.0) NA 86.8 (86.8–86.8) CI: confidence interval; NA: not applicable; NPV: negative predictive value; PPV: positive predictive value.

97.6) 15.6 (6.5–29.5) 99.7 (98.1–100.0) 87.5 (46.9–98.2) 88.6 (87.2–89.8) <9 339 (99.7) 2.2 (0.1–11.8) 100.0 (98.8–100.0) 100.0 87.0 (86.5–87.5) >10 340 (100) 0.0 (0.0–7.9) 100.0 (98.8–100.0) NA 86.8 (86.8–86.8) CI: confidence interval; NA: not applicable; NPV: negative predictive value; PPV: positive predictive value. MACEs defined as all-cause mortality, acute coronary syndrome or coronary revascularisation.

97.6) 15.6 (6.5–29.5) 99.7 (98.1–100.0) 87.5 (46.9–98.2) 88.6 (87.2–89.8) <9 339 (99.7) 2.2 (0.1–11.8) 100.0 (98.8–100.0) 100.0 87.0 (86.5–87.5) >10 340 (100) 0.0 (0.0–7.9) 100.0 (98.8–100.0) NA 86.8 (86.8–86.8) CI: confidence interval; NA: not applicable; NPV: negative predictive value; PPV: positive predictive value. MACEs defined as all-cause mortality, acute coronary syndrome or coronary revascularisation. HEART score and CCTA The association between HEART risk categories and CCTA findings are shown in Table 1. In intermediate-risk patients, CCTA reclassified 143 (74.1%) patients to low-risk (<50% stenosis with a 30-day MACE rate 0.7%) and 50 (25.9%) patients to high-risk (>50% stenosis with a 30-day MACE rate 46%). One intermediate-risk patient (HEART score six) with non-obstructive plaque on CCTA had an adjudicated diagnosis of myocardial infarction, however this was considered a myocardial infarction with nonobstructive coronary arteries (MINOCA) with a minimal rise pattern in cardiac troponin and no significant stenosis on subsequent ICA (Table 2). The addition of CCTA to the HEART score was associated with a significant improvement of the diagnostic accuracy for 30-day MACEs (AUC 0.95 (0.92–0.97) vs 0.83 (0.78–0.87); p<0.001) (Figure 3). Sensitivity, specificity, NPV and PPV for the prediction of 30-day MACEs of an algorithm where CCTA is reserved for intermediate HEART scores (4–6) was 88.9% (76.0–96.3), 87.1% (82.8–90.7), 98.1% (95.7–99.2) and 51.3% (43.5–59.0), respectively. This algorithm reduces the need for CCTA by 43% (n=147). An algorithm where CCTA is reserved for HEART score 3–6 patients had a sensitivity, specificity, NPV and PPV for the prediction of 30-day MACEs of 97.8% (88.2–99.9), 84.1% (79.4–88.1), 99.6% (97.3–99.9) and 48.4% (41.8–55.0), respectively (Figure 4). This algorithm reduces the need for CCTA by 22% (n=75).

43% (n=147). An algorithm where CCTA is reserved for HEART score 3–6 patients had a sensitivity, specificity, NPV and PPV for the prediction of 30-day MACEs of 97.8% (88.2–99.9), 84.1% (79.4–88.1), 99.6% (97.3–99.9) and 48.4% (41.8–55.0), respectively (Figure 4). This algorithm reduces the need for CCTA by 22% (n=75). Figure 3. Predictive value of coronary computed tomography angiography (CCTA), HEART score and HEART score combined with CCTA for 30-day major adverse cardiac events (MACEs). Receiver-operating-characteristic curves show the predictive value of CCTA, the HEART score and the HEART score combined with CCTA for 30-day MACEs. MACEs defined as all-cause mortality, acute coronary syndrome or coronary revascularisation. AUC: area under the curve. Figure 4. Predictive value of the HEART score combined with coronary computed tomography angiography (CCTA) assessment in HEART scores 3–6 for 30-day major adverse cardiac events (MACEs). MACEs defined as all-cause mortality, acute coronary syndrome or coronary revascularisation. *One patient with a HEART score of six and ≤50% stenosis on CCTA had an adjudicated diagnosis of myocardial infarction, however this was considered a myocardial infarction with nonobstructive coronary arteries (MINOCA) with a minimal rise pattern in cardiac troponin and no significant stenosis on subsequent invasive coronary angiography. NPV: negative predictive value; PPV: positive predictive value.

diagnosis of myocardial infarction, however this was considered a myocardial infarction with nonobstructive coronary arteries (MINOCA) with a minimal rise pattern in cardiac troponin and no significant stenosis on subsequent invasive coronary angiography. NPV: negative predictive value; PPV: positive predictive value. Discussion In the current study, we investigated the predictive value and efficiency of the HEART score before CCTA for 30-day MACEs in suspected ACS patients in the ED and report several important findings. First, CCTA is a good predictor of 30-day MACEs in suspected ACS patients in the ED (AUC 0.91). Second, rule-out of 30-day MACEs based on the originally proposed low-risk HEART category (HEART score ≤3) is suboptimal (sensitivity 91.1% and NPV 96.6%). Third, addition of CCTA to the HEART score significantly improves the diagnostic accuracy for 30-day MACEs (AUC: 0.83 to 0.95; p<0.001). Finally, an algorithm where CCTA is reserved for patients with HEART score 3–6 reduces the need for CCTA by 22% (n=75) without compromising diagnostic accuracy or safety.

NPV 96.6%). Third, addition of CCTA to the HEART score significantly improves the diagnostic accuracy for 30-day MACEs (AUC: 0.83 to 0.95; p<0.001). Finally, an algorithm where CCTA is reserved for patients with HEART score 3–6 reduces the need for CCTA by 22% (n=75) without compromising diagnostic accuracy or safety. HEART score In our study, the HEART score identified a large proportion (35%) of low-risk patients proposed for early discharge. However, the incidence of MACEs in low-risk patients was higher (3.4%) compared to previous reports, where the incidence ranged from 0.4–2.5%.8,12,13,17–19 Using the originally proposed score of ≤3 resulted in a generally unacceptable sensitivity and NPV in this population.20 Notably, all four low-risk patients with 30-day MACEs had a score of three, of whom only one was diagnosed with non-ST segment elevation myocardial infarction (NSTEMI). Lowering the cut-off value for discharge to HEART scores ≤2 increased the diagnostic accuracy to acceptable levels in our study, something that has been proposed previously.21 Further improvement of the diagnostic accuracy can probably be achieved by modifying the HEART score to incorporate serial troponin measurements.22–24

ng the cut-off value for discharge to HEART scores ≤2 increased the diagnostic accuracy to acceptable levels in our study, something that has been proposed previously.21 Further improvement of the diagnostic accuracy can probably be achieved by modifying the HEART score to incorporate serial troponin measurements.22–24 CCTA following HEART score The addition of CCTA to the HEART score resulted in a substantial improvement in diagnostic accuracy, mainly by reclassifying intermediate-risk patients to their appropriate risk group. At the same time, using the HEART score to select patients that will benefit most from CCTA can result in a more efficient approach. Very low-risk HEART patients (score ≤2) did not experience 30-day MACEs in the current study and can be discharged safely from the ED, with further screening in an outpatient setting. High-risk HEART score patients, of whom 60.7% experienced 30-day MACEs in the current study, probably benefit most from an approach with early ICA. The algorithm HEART 3–6+CCTA reduced the number of needed coronary computed tomography angiograms while maintaining a high diagnostic accuracy and identifying a large proportion (73%) of patients who are eligible for safe and early discharge from the ED. In a similar fashion to the PRospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) minimal risk tool in suspected stable angina patients, which identifies individuals with low risk of CAD, the HEART score is able to reduce the need for non-invasive testing without comprising safety.25

from the ED. In a similar fashion to the PRospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) minimal risk tool in suspected stable angina patients, which identifies individuals with low risk of CAD, the HEART score is able to reduce the need for non-invasive testing without comprising safety.25 Limitations The current study is a secondary analysis of patients suspected of ACS that underwent CCTA in the ED and should therefore be regarded as hypothesis generating. Our study population, which consisted mostly of low- to intermediate-risk patients, may not be representative of other populations of patients presenting with suspected ACS. The study population also consisted of patients in whom results of CCTA were used as part of their clinical work-up which in turn might have introduced a work-up bias. Furthermore, due to the heterogeneity of troponin assays implemented in current study, the results may be less applicable to individual troponin assays in clinical practice. In the current analysis, we were unable to investigate the diagnostic accuracy of the HEART pathway, an algorithm which incorporates serial troponin measurements into the HEART score, as serial troponin measurements were available in a minority of the patients. A disadvantage of CCTA is the exposure to radiation, however recent developments in scanner technology and dose-reducing protocols have led to a reduction in radiation exposure.9 Furthermore, in the current analysis the HEART score helps reduce the number of coronary computed tomography angiograms performed, which also minimises the number of patients that are exposed to radiation.

ent developments in scanner technology and dose-reducing protocols have led to a reduction in radiation exposure.9 Furthermore, in the current analysis the HEART score helps reduce the number of coronary computed tomography angiograms performed, which also minimises the number of patients that are exposed to radiation. Conclusion The predictive value of CCTA for 30-day MACEs in suspected ACS patients is good and reserving CCTA for HEART score 3–6 patients reduces the number of needed coronary computed tomography angiograms without affecting diagnostic accuracy. Supplemental Material Supplemental_Material – Supplemental material for HEART score improves efficiency of coronary computed tomography angiography in patients suspected of acute coronary syndrome in the emergency department Click here for additional data file. Supplemental material, Supplemental_Material for HEART score improves efficiency of coronary computed tomography angiography in patients suspected of acute coronary syndrome in the emergency department by Murat Arslan, Jeroen Schaap, Pleunie PM Rood, Koen Nieman, Ricardo PJ Budde, Mohamed Attrach, Eric A Dubois and Admir Dedic in European Heart Journal: Acute Cardiovascular Care Conflict of interest: KN reports unrestricted institutional research support from Siemens Healthineers, Bayer, GE and HeartFlow, outside the submitted work. All other authors declared no conflict of interest.

Supplemental material, Supplemental_Material for HEART score improves efficiency of coronary computed tomography angiography in patients suspected of acute coronary syndrome in the emergency department by Murat Arslan, Jeroen Schaap, Pleunie PM Rood, Koen Nieman, Ricardo PJ Budde, Mohamed Attrach, Eric A Dubois and Admir Dedic in European Heart Journal: Acute Cardiovascular Care Conflict of interest: KN reports unrestricted institutional research support from Siemens Healthineers, Bayer, GE and HeartFlow, outside the submitted work. All other authors declared no conflict of interest. Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a grant from Erasmus MC and a research grant from the Erasmus MC Thorax Foundation (project grant B4).

Introduction Prompt treatment with primary percutaneous coronary intervention (PCI) increases the likelihood of survival for patients presenting with ST-segment elevation myocardial infarction (STEMI). Given the close relationship between ischaemia time and hypoxia-induced loss of contractile myocardium, recent advances in the therapy of STEMI patients have mainly focused on the time to reperfusion as a key performance indicator for better outcomes in terms of morbidity and mortality.1–10 Rapid and accurate interpretation of pre-hospital electrocardiograms in combination with bypassing the emergency department (ED) has been proposed as a feasible strategy to optimise and shorten re-perfusion times by preventing treatment delays from hospital arrival to balloon inflation and coronary stenting.11–21

.1–10 Rapid and accurate interpretation of pre-hospital electrocardiograms in combination with bypassing the emergency department (ED) has been proposed as a feasible strategy to optimise and shorten re-perfusion times by preventing treatment delays from hospital arrival to balloon inflation and coronary stenting.11–21 As the successful restoration of antegrade coronary flow by early reperfusion therapy improves the prognosis of STEMI, the current guidelines of the European Society of Cardiology (ESC) for the treatment of STEMI patients underscore the significance of fast treatment pathways to primary PCI for emergency revascularisation by recommending fast tracking from the field directly to the catheterisation laboratory of a PCI-capable hospital.22 However, clinical evidence for this approach is rather limited and the American College of Cardiology (ACC)/American Heart Association (AHA) has not issued comparable guidance on this matter.23 The rather weak ESC recommendation is based mainly on the results from a large US registry which, however, reported similar adjusted mortality risks between patients bypassing and not bypassing the ED.18 A recently published systematic review on the impact of direct admission to the catheterisation laboratory as compared to transport to the ED confirmed the reduced delay to the start of revascularisation but did not produce a clear evidence-based benefit with respect to outcome.24 Thus, conflicting data exist as to whether a direct transfer without stopping at the ED will lower the incidence of adverse events during in-hospital treatment.10,11,12,25,26

he ED confirmed the reduced delay to the start of revascularisation but did not produce a clear evidence-based benefit with respect to outcome.24 Thus, conflicting data exist as to whether a direct transfer without stopping at the ED will lower the incidence of adverse events during in-hospital treatment.10,11,12,25,26 Recently, we reported in data from the ongoing FITT–STEMI (Feedback Intervention and Treatment Times in ST-Elevation Myocardial Infarction) trial that shortened contact-to-balloon times were associated with improved survival, particularly in patients with cardiogenic shock.27,28 Using key driver analysis, we found that direct transmission to the catheterisation laboratory was a significant determinant of time to PCI treatment. However, neither from this study nor the existing literature is it clear whether ED bypass has any significant impact on mortality after adjustment to clinically relevant confounders. In the present paper, we addressed this important clinical issue in both haemodynamically stable patients and unstable patients presenting with cardiogenic shock.

study nor the existing literature is it clear whether ED bypass has any significant impact on mortality after adjustment to clinically relevant confounders. In the present paper, we addressed this important clinical issue in both haemodynamically stable patients and unstable patients presenting with cardiogenic shock. Methods Participating hospitals The present paper reports outcome data from the prospective and ongoing multicentre FITT–STEMI study. The study protocol included systematic data analysis and standardised feedback interventions on treatment times and mortality in STEMI patients treated in different regional cardiac care networks.29,30 All consecutive PCI-treated STEMI patients presenting within 24 hours of symptom onset from 1 January 2006 to 31 December 2015, who were transported directly from the field by emergency medical transportation and treated with PCI within 360 minutes after first medical contact, were considered eligible for inclusion in this analysis. The existing study protocol encouraged immediate activation of the catheterisation laboratory by a single call, when the patient was triaged to primary PCI, combined with a direct transfer to the laboratory in order to prevent treatment delays in the time to reperfusion.27 The participating 48 PCI-capable hospitals, which are located all over Germany, endorsed the key strategies of the ACC initiative for the management of STEMI patients, including the establishment of multidisciplinary treatment teams for around the clock invasive treatment. For enrollment in the FITT-STEMI consortium, the candidate PCI centre had to fulfill the following requirements: a 24-hour PCI accessibility for at least one year before study participation, affiliation of at least two interventional cardiologists qualified to take the incoming calls of the emergency medical transportation team, a minimum of 250 PCI procedures per year and more than 50 annual PCI treatments in STEMI patients.

ts: a 24-hour PCI accessibility for at least one year before study participation, affiliation of at least two interventional cardiologists qualified to take the incoming calls of the emergency medical transportation team, a minimum of 250 PCI procedures per year and more than 50 annual PCI treatments in STEMI patients. Among the total study participants with STEMI (n=20,964), the majority of patients (n=15,153) were directly transported from the scene to the PCI hospital by emergency medical services (EMS), and of those n=13,365 underwent prompt interventional reperfusion therapy with primary PCI (Figure 1). Not eligible for this outcome analysis were patients with: (a) interfacility transfer from a non-PCI-capable hospital to a receiving on-site study centre (n=3136); (b) self-admission to the PCI hospital (n=2174); and (c) myocardial infarction during hospital treatment at the PCI hospital (n=501). Complete records were available for n=13,219 patients (98.9%) transported by EMS transportation and were treated for reperfusion by primary PCI within 360 minutes after the first medical contact. Among them, n=6740 patients (51.0%) were directly transported from the pre-hospital setting to the catheterisation laboratory, while the other half (n=6479) had a stop at the ED on their way to PCI treatment. Cardiogenic shock was diagnosed in 1625 patients (12.3%), among them 684 patients (42.1%) were transported directly to the catheterisation laboratory. Figure 1. Flow diagram of the FITT–STEMI study cohort.

Among the total study participants with STEMI (n=20,964), the majority of patients (n=15,153) were directly transported from the scene to the PCI hospital by emergency medical services (EMS), and of those n=13,365 underwent prompt interventional reperfusion therapy with primary PCI (Figure 1). Not eligible for this outcome analysis were patients with: (a) interfacility transfer from a non-PCI-capable hospital to a receiving on-site study centre (n=3136); (b) self-admission to the PCI hospital (n=2174); and (c) myocardial infarction during hospital treatment at the PCI hospital (n=501). Complete records were available for n=13,219 patients (98.9%) transported by EMS transportation and were treated for reperfusion by primary PCI within 360 minutes after the first medical contact. Among them, n=6740 patients (51.0%) were directly transported from the pre-hospital setting to the catheterisation laboratory, while the other half (n=6479) had a stop at the ED on their way to PCI treatment. Cardiogenic shock was diagnosed in 1625 patients (12.3%), among them 684 patients (42.1%) were transported directly to the catheterisation laboratory. Figure 1. Flow diagram of the FITT–STEMI study cohort. Outcome measures In order to achieve continuous high-quality management, ongoing outcome monitoring with respect to treatment times was performed using an electronic case report form for data collection and a web-based data transfer system to the principal coordinating centre. For each consecutive STEMI patient, the participating PCI centres collected detailed information on pre- and in-hospital treatment times from initial contact with the medical system to balloon inflation. Key time points for out-of-hospital treatment included time of arrival at the field by EMS transportation usually staffed in Germany with a trained physician and experienced paramedics, treatment time at the scene, and transport time to the hospital. Assessment of in-hospital treatment times included information on whether there was a direct transfer to the catheterisation laboratory or, alternatively, an indirect transfer by the ED, as well as the time on arrival at the catheterisation laboratory, and the times of puncture and first balloon inflation at the culprit lesion. For each STEMI patient, the following interventional variables were documented as predefined key quality indicators by ambulance personnel or attending physicians: time of pre-hospital ECG recording within or longer than 10 minutes after arrival at the scene, telephone announcement in advance and telemetry-ECG transmission, as well as average and median components of times to treatment, including those from the first medical contact to balloon inflation. Information about in-hospital mortality after PCI was obtained for each patient. Additional data were obtained on medical history, cardiac risk factors, prior medication, medical comorbidity, Thrombolysis In Myocardial Infarction (TIMI) risk score as well as results from coronary angiography and the PCI procedure.

ion. Information about in-hospital mortality after PCI was obtained for each patient. Additional data were obtained on medical history, cardiac risk factors, prior medication, medical comorbidity, Thrombolysis In Myocardial Infarction (TIMI) risk score as well as results from coronary angiography and the PCI procedure. Patients were classified as being in cardiogenic shock (Killip class IV) when the following clinical criteria were fulfilled and confirmed by experienced cardiologists: hypotension (systolic blood pressure of <90 mmHg or the need for supportive measures to maintain a systolic blood pressure of ⩾90 mmHg), signs of critical end-organ hypoperfusion, and a heart rate of 60 beats/minute or greater.31 The study protocol was approved by the ethics committee of the medical faculty at the University of Göttingen and the local ethics committees of all participating PCI centres.

to maintain a systolic blood pressure of ⩾90 mmHg), signs of critical end-organ hypoperfusion, and a heart rate of 60 beats/minute or greater.31 The study protocol was approved by the ethics committee of the medical faculty at the University of Göttingen and the local ethics committees of all participating PCI centres. Data assessment Outcome data for each local study site were provided through a cooperative agreement which included regular and systematic data analysis and formalised feedback interventions in order to implement procedures of standardised quality management for timely reperfusion therapy. For this purpose, out-of-hospital emergency ambulance teams consisting of qualified physicians, trained in emergency medicine, and the medical staff working in the ED of the participating centres were regularly instructed by the local principal investigators to make precise diagnoses of the acute coronary syndrome and early PCI team activation in order to increase the frequency of direct transports to the catheterisation laboratory for STEMI patients. Feedback interventions were given on a quarterly basis during the first month of each quarter beginning in the third quarter after attending the FITT–STEMI consortium. At each local study site, the formalised feedback presentations were discussed in periodic and interactive sessions with members of the participating interdisciplinary STEMI treatment teams, including staff from the local EMS, physicians and nurses working in the ED and the emergency responding system, as well as staff from the catheterisation laboratory and interventional cardiologists. Particular attention was given to site-specific descriptive statistics with regard to improvements in treatment times elapsed from the first medical contact to direct hand-off in the catheterisation laboratory.

ncy responding system, as well as staff from the catheterisation laboratory and interventional cardiologists. Particular attention was given to site-specific descriptive statistics with regard to improvements in treatment times elapsed from the first medical contact to direct hand-off in the catheterisation laboratory. Data from two other infarct surveys were used to validate the completeness of our recruitment strategy, namely (a) insurance reimbursement data based on the International Statistical Classification of Diseases and Related Health Problems 10 (ICD-10) codes I21.0 to I21.3 for acute and subacute transmural infarction; and (b) data from the mandatory German hospital quality report on PCI procedures for the indication ‘ST-segment elevation myocardial infarction within 24 hours after ECG diagnosis’.27 The latter survey also included subacute myocardial infarctions, and the participation in the survey was compulsory for all certified PCI-capable catheterisation laboratories up to the year 2016. The stable percentages of annually included STEMI patients for the ICD-coded diagnosis of transmural infarction (69.7 ± 6.8 percentage points) as well as the considerable consilience with routine PCI procedures (95.6 ± 11.5 percentage points) underscore the overall integrity and completeness of the enrollment strategy used in the FITT–STEMI study.

ually included STEMI patients for the ICD-coded diagnosis of transmural infarction (69.7 ± 6.8 percentage points) as well as the considerable consilience with routine PCI procedures (95.6 ± 11.5 percentage points) underscore the overall integrity and completeness of the enrollment strategy used in the FITT–STEMI study. Statistical analysis Statistical analyses were performed using the SAS system (version 9.4). Raw data were used to calculate time intervals along the treatment pathway for each patient. Continuous data given as means and standard deviations were compared between the two groups of direct and non-direct transfer to the catheterisation laboratory using Student’s t tests. Categorical variables were analysed using chi-square tests. To assess whether bypassing the ED impacted on survival, a series of different logistic regression models were constructed with in-hospital mortality as the dependent variable and direct transfer as the independent variable adjusted for potential confounders known to be associated with outcomes. To search for potential factors that may account for the beneficial effect of direct transfer on survival, we optionally included door-to-balloon time as an interventional variable for in-hospital treatment time. For the models, a backward selection method was used to enter factors into the regression models. The results from these regression models are presented as odds ratios (ORs) with 95% confidence intervals (CIs). All analyses were two-tailed and a P value of 0.05 was considered statistically significant.

treatment time. For the models, a backward selection method was used to enter factors into the regression models. The results from these regression models are presented as odds ratios (ORs) with 95% confidence intervals (CIs). All analyses were two-tailed and a P value of 0.05 was considered statistically significant. Results Treatment times in patients with and without ED bypass Although STEMI patients transported directly from the scene to the catheterisation laboratory differed with respect to age, sex and TIMI risk group from those not directly transported, there was a great overlap between the two groups (Supplementary Figure 1). In the total study cohort, angiographic results showed a high rate of successful revascularisation, as TIMI angiographic flow grade scoring 3 was achieved in more than nine out of 10 patients in the two groups (94.1% vs. 92.1%, P<0.0001). A detailed description of the total study population including the comparison between the groups with and without direct transfer is given in Table 1. Table 1. Baseline characteristics of PCI-treated STEMI patients transported by emergency medical services and stratified by direct transfer to the catheterization laboratory (ED bypass) versus non-direct transfer by the emergency department (non-ED bypass).

Results Treatment times in patients with and without ED bypass Although STEMI patients transported directly from the scene to the catheterisation laboratory differed with respect to age, sex and TIMI risk group from those not directly transported, there was a great overlap between the two groups (Supplementary Figure 1). In the total study cohort, angiographic results showed a high rate of successful revascularisation, as TIMI angiographic flow grade scoring 3 was achieved in more than nine out of 10 patients in the two groups (94.1% vs. 92.1%, P<0.0001). A detailed description of the total study population including the comparison between the groups with and without direct transfer is given in Table 1. Table 1. Baseline characteristics of PCI-treated STEMI patients transported by emergency medical services and stratified by direct transfer to the catheterization laboratory (ED bypass) versus non-direct transfer by the emergency department (non-ED bypass). Total study population (n=13,219) ED bypass (n=6740; 51%) Non-ED bypass (n=6479; 49%) P value Demographic data Male gender 9733 (74%) 5080 (75%) 4653 (72%) <0.0001 Age ± SD 63.6 ± 12.9 63.0 ± 12.6 64.3 ± 13.2 <0.0001 Age >80 years 1375 (10%) 600 (9%) 775 (12%) <0.0001 Body mass index (kg/m²) (mean, SD) 27.5 ± 4.6 27.5 ± 4.5 27.4 ± 4.6 0.2161 Clinical data Hypertension 7845 (59%) 3954 (59%) 3891 (60%) 0.1036 Diabetes mellitus 2281 (17%) 1106 (16%) 1175 (18%) 0.0087 Prior angina pectoris 1709 (13%) 895 (13%) 814 (13%) 0.2205 Hyperlipoproteinaemia 3849 (29%) 1911 (28%) 1938 (30%) 0.0486 Family history 2511 (19%) 1291 (19%) 1220 (19%) 0.6347 Current smoker 5577 (42%) 2930 (43%) 2647 (41%) 0.0023 Previous myocardial infarction 1450 (11%) 651 (10%) 799 (12%) <0.0001 Previous stroke 550 (4%) 249 (4%) 301 (5%) 0.0062 Previous angioplasty 1,474 (11%) 687 (10%) 787 (12%) 0.0004 Previous CABG 301 (2%) 120 (2%) 181 (3%) 0.0001 Renal failure 619 (5%) 248 (4%) 371 (6%) <0.0001 Cardiopulmonary resuscitation 1256 (10%) 528 (8%) 728 (11%) <0.0001 Cardiogenic shock 1625 (12%) 684 (10%) 941 (15%) <0.0001 Intra-aortic balloon counterpulsation 264 (2%) 156 (2%) 108 (2%) 0.0078 Off hours (nights/weekends) 7607 (58%) 3265 (48%) 4342 (67%) <0.0001 Pre-hospital ECG 12,202 (92%) 6604 (98%) 5598 (86%) <0.0001 Telemetry ECG 3072 (23%) 1924 (29%) 1148 (18%) <0.0001 Pre-announcement by telephone 10,877 (82%) 6513 (97%) 4364 (67%) <0.0001 TIMI risk score 0–2 4679 (35%) 2592(38%) 2087 (32%) <0.0001 3–4 3787 (29%) 1964 (29%) 1823 (28%) 5–8 4102 (31%) 1889 (28%) 2213 (34%) >8 651 (5%) 295 (4%) 356 (5%) Treatment with glycoprotein IIb/IIIa receptor blockers (n=11,942) 5143 (43%) 2567 (42%) 2576 (44%) 0.0063 Angiographic results No.

(97%) 4364 (67%) <0.0001 TIMI risk score 0–2 4679 (35%) 2592(38%) 2087 (32%) <0.0001 3–4 3787 (29%) 1964 (29%) 1823 (28%) 5–8 4102 (31%) 1889 (28%) 2213 (34%) >8 651 (5%) 295 (4%) 356 (5%) Treatment with glycoprotein IIb/IIIa receptor blockers (n=11,942) 5143 (43%) 2567 (42%) 2576 (44%) 0.0063 Angiographic results No. of coronary arteries narrowed: 0 31 (0.2%) 8 (0.1%) 23 (0.4%) <0.0001 1 5274 (40%) 2748 (41%) 2526 (39%) 2 4072 (31%) 2143 (32%) 1929 (30%) 3 3742 (28%) 1796 (27%) 1946 (30%) LMCA 99 (0.8%) 45 (0.7%) 54 (0.8%) CTO in NIRA 1494 (11%) 691 (10%) 803 (12%) Recanalisation vessel LAD 5814 (44%) 2912 (43%) 2902 (45%) <0.0001 RCA 5470 (41%) 2939 (44%) 2531 (39%) LCX 1676 (13%) 786 (12%) 890 (14%) LMCA 124 (1%) 54 (0.8%) 70 (1.1%) Graft 134 (1%) 49 (0.7%) 85 (1.3%) ECG (STEMI site) Anterior 5841 (44%) 2927 (43%) 2914 (45%) <0.0001 Inferior 6577 (50%) 3477 (52%) 3100 (48%) Lateral 672 (5%) 298 (4%) 374 (6%) LBBB 129 (1%) 38 (1%) 91 (1%) TIMI angiographic flow grade before PCI Score 0–2 12,227 (92%) 6281 (93%) 5946 (92%) 0.0022 Score 3 991 (7%) 459 (7%) 532 (8%) TIMI angiographic flow grade after PCI Score 0–2 905 (7%) 395 (6%) 510 (8%) <0.0001 Score 3 12,313 (93%) 6345 (94%) 5968 (92%) Outcome In-hospital mortality rate 1062 (8.0%) 417 (6.2%) 645 (10.0%) <0.0001 Data are presented as percentages or means and standard deviations. P-values refer to the comparisons between the two groups.

graphic flow grade after PCI Score 0–2 905 (7%) 395 (6%) 510 (8%) <0.0001 Score 3 12,313 (93%) 6345 (94%) 5968 (92%) Outcome In-hospital mortality rate 1062 (8.0%) 417 (6.2%) 645 (10.0%) <0.0001 Data are presented as percentages or means and standard deviations. P-values refer to the comparisons between the two groups. STEMI: ST-segment elevation myocardial infarction; ED: emergency department; PCI: percutaneous coronary intervention; CABG: coronary artery bypass grafting; CTO: chronic total occlusion; LBBB: left bundle branch block; LCA: left coronary artery; LCX: left circumflex artery; LMCA: left main coronary artery; NIRA: non-infarct-related artery; RCA: right coronary artery; SD: standard deviation; TIMI: Thrombolysis In Myocardial Infarction. Whereas scene-to-door time was longer in the group bypassing versus not bypassing the ED (18.3 ± 11.5 vs. 16.6 ± 22.8 minutes, P<0.0001), the mean door-to-catheterisation time was considerably shorter in the ED bypass group (5.4 ± 5.6 vs. 47.8 ± 38.8 minutes, P<0.0001) (Table 2). Likewise, the mean door-to-balloon time in the direct transfer group was 45 minutes shorter than in the non-bypassing group (P<0.0001; Figure 2). Overall, this resulted in a gain of treatment time from the first medical contact to balloon inflation of more than 44 minutes for the direct transfer group as compared to the group with a stop at the ED (76.5 ± 22.3 vs. 121.2 ± 47.3 minutes, P<0.0001). Notably, there was no difference in the cath-to-puncture time between directly and not directly transported patients (11.6 ± 6.6 vs. 11.8 ± 7.2, P=0.1937).

oon inflation of more than 44 minutes for the direct transfer group as compared to the group with a stop at the ED (76.5 ± 22.3 vs. 121.2 ± 47.3 minutes, P<0.0001). Notably, there was no difference in the cath-to-puncture time between directly and not directly transported patients (11.6 ± 6.6 vs. 11.8 ± 7.2, P=0.1937). Table 2. Relevant interventional time intervals in the two groups with direct and non-direct transportation to the catheterisation laboratory (ED bypass versus non-ED bypass). Total study population (n=13,219) ED bypass (n=6740) Non-ED bypass (n=6479) P value Symptom-to-contact (min) 153.7 ± 221.4 152.6 ± 217.6 154.8 ± 225.3 0.5568 Time at scene (min) 22.9 ± 22.1 22.0 ± 11.4 23.8 ± 29.4 <0.0001 Transport time (min) 17.4 ± 17.9 18.3 ± 11.5 16.6 ± 22.8 <0.0001 Door-to-cath time (min) 26.2 ± 34.7 5.4 ± 5.6 47.8 ± 38.8 <0.0001 Cath-to-puncture time (min) 11.7 ± 6.9 11.6 ± 6.6 11.8 ± 7.2 0.1937 Puncture-to-balloon time (min) 20.6 ± 12.9 19.2 ± 12.0 22.0 ± 13.6 <0.0001 Door-to-balloon time (min) 58.5 ± 39.4 36.2 ± 14.8 81.6 ± 43.4 <0.0001 Contact-to-balloon time (min) 98.4 ± 43.0 76.5 ± 22.3 121.2 ± 47.3 <0.0001 Data are given as means and standard deviations. ED: emergency department.

Total study population (n=13,219) ED bypass (n=6740) Non-ED bypass (n=6479) P value Symptom-to-contact (min) 153.7 ± 221.4 152.6 ± 217.6 154.8 ± 225.3 0.5568 Time at scene (min) 22.9 ± 22.1 22.0 ± 11.4 23.8 ± 29.4 <0.0001 Transport time (min) 17.4 ± 17.9 18.3 ± 11.5 16.6 ± 22.8 <0.0001 Door-to-cath time (min) 26.2 ± 34.7 5.4 ± 5.6 47.8 ± 38.8 <0.0001 Cath-to-puncture time (min) 11.7 ± 6.9 11.6 ± 6.6 11.8 ± 7.2 0.1937 Puncture-to-balloon time (min) 20.6 ± 12.9 19.2 ± 12.0 22.0 ± 13.6 <0.0001 Door-to-balloon time (min) 58.5 ± 39.4 36.2 ± 14.8 81.6 ± 43.4 <0.0001 Contact-to-balloon time (min) 98.4 ± 43.0 76.5 ± 22.3 121.2 ± 47.3 <0.0001 Data are given as means and standard deviations. ED: emergency department. Figure 2. Frequencies of door-to-balloon time intervals as demonstrated by histograms separately for percutaneous coronary intervention (PCI)-treated ST-segment elevation myocardial infarction (STEMI) patients with direct transfer to the catheterisation laboratory bypassing the emergency department (ED) (a) and with indirect transfer to the catheterisation laboratory due to a transient stop in the ED (b).

parately for percutaneous coronary intervention (PCI)-treated ST-segment elevation myocardial infarction (STEMI) patients with direct transfer to the catheterisation laboratory bypassing the emergency department (ED) (a) and with indirect transfer to the catheterisation laboratory due to a transient stop in the ED (b). In patients presenting with cardiogenic shock, symptom-to-contact (P=0.1654) and contact-to-door times (P=0.0692) did not significantly differ with regard to ED bypass (Supplementary Table 1). However, the mean contact-to-balloon time was 49 minutes shorter in haemodynamically unstable patients with ED bypass as compared to their not directly transported counterparts (P<0.0001). For stable patients with no clinical signs of cardiogenic shock, this gain in time to PCI treatment was in the same range (43 minutes).

ver, the mean contact-to-balloon time was 49 minutes shorter in haemodynamically unstable patients with ED bypass as compared to their not directly transported counterparts (P<0.0001). For stable patients with no clinical signs of cardiogenic shock, this gain in time to PCI treatment was in the same range (43 minutes). Reasons for non-direct transfer to the catheterisation laboratory Despite a universal strategy among all study sites for pre-arrival notification of the catheterisation laboratory by the emergency medical transportation team and an endorsement for direct transport to the catheterisation laboratory, approximately half of the total study cohort was not directly transferred (49%). Using a standardised item on the case report form allowing multiple answers, we assessed the reasons for failed direct transfer. Direct transport was failed due to late arrival of the catheterisation laboratory staff (27.8%), no announcement or failure of timely announcement by EMS (20.0%), ambiguous pre-hospital STEMI diagnosis (17.1%), occupancy of the catheterisation laboratory by another patient at the time of arrival (16.6%), time-consuming primary care in the ED (10.2%), elaborate diagnostic procedures to exclude significant comorbidity including computed tomography (0.7%) and miscellaneous causes (14.9%). In summary, the reasons for failed direct transmission were often multifactorial, most probably due to problems in pre-hospital diagnosis and delays in catheterisation laboratory team readiness.

agnostic procedures to exclude significant comorbidity including computed tomography (0.7%) and miscellaneous causes (14.9%). In summary, the reasons for failed direct transmission were often multifactorial, most probably due to problems in pre-hospital diagnosis and delays in catheterisation laboratory team readiness. Overall beneficial impact of ED bypass on survival Among the study participants who went through the ED, there were 645 deaths (10.0%), whereas mortality was significantly lower in the patient group brought directly to the catheterisation laboratory (417 deaths, 6.2%, P<0.0001). The improved survival related to ED bypass was observed for both haemodynamically stable (n =11,594, 2.8% vs. 3.8%, P=0.0024) and unstable patients with cardiogenic shock (n=1625, 36.3% vs. 46.2%, P<0.0001) (Figure 3(a)). Corroborating these findings, in all four predefined TIMI risk score subgroups, mortality was lower in patients bypassing the ED (Figure 3(b)). Using a logistic regression model with in-hospital mortality as the dependent variable adjusted to the TIMI risk score, we confirmed that direct transport to the catheterisation laboratory was a statistically significant and independent predictor of better survival (OR 0.64, 95% CI 0.56–0.74, P<0.0001; Table 3).

3(b)). Using a logistic regression model with in-hospital mortality as the dependent variable adjusted to the TIMI risk score, we confirmed that direct transport to the catheterisation laboratory was a statistically significant and independent predictor of better survival (OR 0.64, 95% CI 0.56–0.74, P<0.0001; Table 3). Figure 3. Mortality rates in percutaneous coronary intervention (PCI)-treated ST-segment elevation myocardial infarction (STEMI) patients with (blue columns) and without (red columns) emergency department bypass by cardiogenic shock (a) and predefined Thrombolysis In Myocardial Infarction (TIMI) risk score intervals (b). Significant group differences are marked with asterisks. Table 3. Logistic regression model with in-hospital mortality as dependent variable and contact-to-door time as independent variable adjusted for TIMI risk score. Variable Odds ratio 95% CI P value ED bypass 0.639 0.555–0.736 <0.0001 Contact-to-door time 1.030 1.026–1.034 <0.0001 TIMI risk score 3–4 vs. ⩽2 3.817 2.687–5.424 <0.0001 5–8 vs. ⩽2 17.164 12.492–23.582 >8 vs. ⩽2 71.341 50.518–100.748 ED: emergency department; TIMI: Thrombolysis In Myocardial Infarction; CI: confidence interval.

dds ratio 95% CI P value ED bypass 0.639 0.555–0.736 <0.0001 Contact-to-door time 1.030 1.026–1.034 <0.0001 TIMI risk score 3–4 vs. ⩽2 3.817 2.687–5.424 <0.0001 5–8 vs. ⩽2 17.164 12.492–23.582 >8 vs. ⩽2 71.341 50.518–100.748 ED: emergency department; TIMI: Thrombolysis In Myocardial Infarction; CI: confidence interval. The reduced mortality in ED bypass patients is achieved through shorter treatment times We then considered whether the positive impact of ED bypass on outcome resulted from a gain in time to reperfusion therapy. To this end, we computed a second regression model again using the TIMI risk score and contact-to-door time as confounders and, in addition to ED bypass, entered door-to-balloon time as an independent and competing variable. In contrast to the first model, direct transfer completely lost its significance in predicting outcome (OR 0.88, 95% CI 0.74–1.04, P=0.1313), whereas in this model shorter door-to-balloon time was a highly significant predictor of better survival (P<0.0001; Table 4). Table 4. Similar model to that demonstrated in Table 3 except that door-to-balloon time was additionally entered as a clinically relevant confounder. Variable OR 95% CI P value ED bypass 0.877 0.740–1.040 0.1313 Contact-to-door time 1.030 1.026–1.033 <0.0001 Door-to-balloon time 1.006 1.004–1.008 <0.0001 TIMI risk score 3–4 vs. ⩽2 3.700 2.603–5.258 <0.0001 5–8 vs. ⩽2 16.365 11.908–22.491 >8 vs. ⩽2 66.964 47.392–94.620 ED: emergency department; TIMI: Thrombolysis In Myocardial Infarction; OR: odds ratio; CI: confidence interval.

ontact-to-door time 1.030 1.026–1.033 <0.0001 Door-to-balloon time 1.006 1.004–1.008 <0.0001 TIMI risk score 3–4 vs. ⩽2 3.700 2.603–5.258 <0.0001 5–8 vs. ⩽2 16.365 11.908–22.491 >8 vs. ⩽2 66.964 47.392–94.620 ED: emergency department; TIMI: Thrombolysis In Myocardial Infarction; OR: odds ratio; CI: confidence interval. Similar results were obtained in two independent models when a set of clinically relevant confounders was substituted for the TIMI risk score. Again, ED bypass was a highly significant predictor of better survival (OR 0.77, 95% CI 0.65–0.91, P=0.0027, Supplementary Table 2), but completely lost its predictive role (P=0.6897) when door-to-balloon time was additionally entered as a significant variable (P<0.0001; Supplementary Table 3). These observations suggest that the beneficial effect of ED bypass resulted simply from the prevention of a treatment delay during the time from hospital arrival to restoring coronary blood flow. Similar findings as for the total study population were also revealed in adjusted models including only patients with cardiogenic shock. Again, ED bypass was a significant predictor of better survival in shock patients (OR 0.69, 95% CI 0.54–0.88, P=0.0028), but lost its significance for predicting death (P=0.8756) when competing with door-to-balloon time entered as an additional and highly significant confounder (P<0.0001).

ients with cardiogenic shock. Again, ED bypass was a significant predictor of better survival in shock patients (OR 0.69, 95% CI 0.54–0.88, P=0.0028), but lost its significance for predicting death (P=0.8756) when competing with door-to-balloon time entered as an additional and highly significant confounder (P<0.0001). Discussion The present paper demonstrates outcome data of the FITT–STEMI trial, a prospective and multicentre study designed to evaluate the association between treatment delays from first medical contact to PCI and in-hospital mortality in a large, unselected cohort of consecutive STEMI patients. Results showed that mortality was significantly lower in the group of STEMI patients with direct transmission to the catheterisation laboratory compared with their non-directly transmitted counterparts. Notably, the gain in survival resulting from ED bypass was observed for both haemodynamically stable and unstable patients presenting with cardiogenic shock. In particular, the latter group benefitted most from direct transport to the catheterisation laboratory, as determined by the low number needed to be treated to save one additional life.

vival resulting from ED bypass was observed for both haemodynamically stable and unstable patients presenting with cardiogenic shock. In particular, the latter group benefitted most from direct transport to the catheterisation laboratory, as determined by the low number needed to be treated to save one additional life. Although the decision for direct transport was generally favoured in younger STEMI patients with less medical comorbidity and a lower TIMI risk score, the highly significant relationship between survival and direct transport remained stable also in models adjusted for these clinically relevant confounders. Another important finding of our study is that, when door-to-balloon time was entered in these models as an independent variable to compete with direct transport in its relevance for in-hospital survival, the bypassing of the ED completely lost its predictive value, whereas shorter treatment time from hospital arrival to balloon inflation became a highly associated predictor of better survival. The prognosis with regard to in-hospital mortality was much better in the ED bypass group, because time-consuming delays associated with the transient stop at the ED were prevented. This underscores the assumption that the beneficial effect of bypassing the ED on survival simply resulted from a gain in time to revascularisation.

sis with regard to in-hospital mortality was much better in the ED bypass group, because time-consuming delays associated with the transient stop at the ED were prevented. This underscores the assumption that the beneficial effect of bypassing the ED on survival simply resulted from a gain in time to revascularisation. By collecting detailed information on treatment times along the entire pre- and intra-hospital treatment pathways, we found that nearly all STEMI patients bypassing the ED with direct transport to the catheterisation laboratory had received a pre-hospital ECG recording and were pre-announced by the EMS. We observed that this did not prolong treatment times at the scene, whereas transport times were minimally longer in this group. This finding could be related to the fact that longer transport times increased the prospects of direct transport to the catheterisation laboratory due to timely activation of the catheterisation laboratory team. In total, in STEMI patients bypassing the ED, we measured no delay in pre-hospital treatment times (contact-to-door), whereas in-hospital times (door-to-balloon) were markedly reduced. This gain in time to reperfusion was exclusively related to the reduction in the time from arrival at the hospital to admission to the catheterisation laboratory.

bypassing the ED, we measured no delay in pre-hospital treatment times (contact-to-door), whereas in-hospital times (door-to-balloon) were markedly reduced. This gain in time to reperfusion was exclusively related to the reduction in the time from arrival at the hospital to admission to the catheterisation laboratory. Our data are in line with recently published studies demonstrating that field triage and the bypassing of the ED in STEMI patients transported by EMS resulted in reduced in-hospital treatment times. Although several studies have supported the idea that a direct-access catheterisation laboratory pathway for timely PCI treatment shortens the time interval to reperfusion therapy and ameliorates hypoxia-induced left ventricular dysfunction, the impact of this approach on survival is still controversial.26 Most studies failed to show a significant effect on mortality in adjusted models or did not exclude transfer patients and patients who did not receive PCI treatment.12,13,15,17 In a single-centre study, Farshid and colleagues observed an improved prognosis in those 190 PCI-treated STEMI patients in whom catheterisation laboratory activation was initiated prior to hospital arrival (24% of all study participants) compared with those with in-hospital catheterisation laboratory activation.21 In a total of 1859 study participants, Estévez-Loureiro et al. demonstrated a long-term prognostic benefit of field triage and direct transfer to the catheterisation laboratory when compared with patients transmitted to the ED.19 However, the proportion of patients transferred by the two treatment pathways was imbalanced as only 425 patients (23%) of the total study cohort were transported directly to PCI treatment and, in addition, the use of the glycoprotein IIb/IIIa receptor antagonist abciximab was higher in the direct transfer group.19 In a report from the AHA Mission Lifeline Program with data from 12,581 STEMI patients with a pre-hospital ECG diagnosis, ED bypass occurred only in 1316 patients (10.5%) and was associated numerically with a lower mortality. The mortality in multivariate analysis, however, was similar in the two groups with and without ED bypass.18

om the AHA Mission Lifeline Program with data from 12,581 STEMI patients with a pre-hospital ECG diagnosis, ED bypass occurred only in 1316 patients (10.5%) and was associated numerically with a lower mortality. The mortality in multivariate analysis, however, was similar in the two groups with and without ED bypass.18 Both the number of study participants with STEMI diagnosis and the ED bypass rate were considerably higher in our large multicentre study compared with previous reports. Our findings confirmed that the algorithm to shorten intervention times from the first medical contact in the field to in-hospital mechanical reperfusion by regular feedback-driven quality controls is feasible and successful in local networks of STEMI care.27 As shown in our previous feasibility studies, formalised interactive data feedback led to an increased number of STEMI patients bypassing the ED in both single-centre and multicentre settings.29,30 However, only half of our total STEMI population was transported directly to the catheterisation laboratory, despite the general endorsement of preventing delays by activating the cardiac catheterisation call team already in the pre-hospital setting.

bypassing the ED in both single-centre and multicentre settings.29,30 However, only half of our total STEMI population was transported directly to the catheterisation laboratory, despite the general endorsement of preventing delays by activating the cardiac catheterisation call team already in the pre-hospital setting. The bypassing of the ED is a complex process and requires strong collaboration between different systems and treatment groups, which includes correct ECG-based STEMI diagnosis in the field as well as the bypassing of non-PCI hospitals and pre-activation of catheterisation laboratory teams. If the patient arrives at the PCI hospital during off hours before the cardiac intervention team, there is a need for the availability of in-hospital urgent care teams to receive the patient at the catheterisation laboratory. For the first time, our study protocol systematically determined both system- and patient-related reasons for the non-bypassing of the ED. Failed pre-announcement by EMS and late arrival of the catheterisation laboratory staff during off hours were identified as relevant factors impeding direct transfer, which raises the potential for further improvements.

systematically determined both system- and patient-related reasons for the non-bypassing of the ED. Failed pre-announcement by EMS and late arrival of the catheterisation laboratory staff during off hours were identified as relevant factors impeding direct transfer, which raises the potential for further improvements. Although the percentage of direct transfers achieved in unstable STEMI patients was lower compared with stable patients (42.1% vs. 52.2%), they benefitted considerably from ED bypass with respect to outcome. In the group of shock patients, ED bypass was significantly associated with better survival, as revealed in univariate comparison and multivariate regression models. One additional life out of 10 PCI-treated patients with cardiogenic shock can be saved when these patients are directly transported to the catheterisation laboratory. Given this low number needing to be treated, more effort should be made to increase the number of direct transports particularly in this high-risk STEMI group.

ditional life out of 10 PCI-treated patients with cardiogenic shock can be saved when these patients are directly transported to the catheterisation laboratory. Given this low number needing to be treated, more effort should be made to increase the number of direct transports particularly in this high-risk STEMI group. Limitations The findings from this prospective study need to be interpreted carefully in the light of several limitations, which mainly result from the observational design as a randomised controlled trial not being feasible due to ethical considerations. In the absence of a randomisation procedure, causality between direct transport to the catheterisation laboratory and mortality cannot formally be concluded. Although we noted that indirect transfer by way of the ED was associated with a considerable delay from hospital arrival to PCI treatment, unmeasured confounding variables and selection bias may impact on decision-making for direct/indirect transfer. To address this important limitation, we used a standardised questionnaire to assess reasons for system delays in those patients not directly transported to the catheterisation laboratory. In addition, the prognostic value of ED bypass may be less pronounced in STEMI management care systems in other countries, in which physicians experienced in emergency medicine do not participate in EMS as they do in Germany.28 However, the beneficial effect of ED bypass can most likely be extrapolated to other health services with similar sophisticated EMS systems and high numbers of PCI centres as in most other developed countries, regardless of whether the EMS are physician-manned or exclusively paramedic-staffed. Finally, only in-hospital mortality data were available in our study, and future studies are needed to examine the association between in-hospital delays before PCI treatment and long-term outcomes.

s in most other developed countries, regardless of whether the EMS are physician-manned or exclusively paramedic-staffed. Finally, only in-hospital mortality data were available in our study, and future studies are needed to examine the association between in-hospital delays before PCI treatment and long-term outcomes. In summary, outcome data from the prospective FITT–STEMI trial indicate the prognostic significance of ED bypass to improve the process of care in STEMI patients treated by PCI. Our adjusted models support the conclusion that the improved prognosis observed for those patients directly transferred to the catheterisation laboratory simply results from their reduced time to PCI treatment. Based on this finding, we encourage existing PCI-capable networks to promote their healthcare system readiness and response to STEMI by pre-hospital announcement of STEMI diagnosis and direct access to the catheterisation laboratory. Protocols favouring the transport of patients from the field directly to the catheterisation laboratory may improve the survival of both haemodynamically stable and unstable patients in established local STEMI treatment networks. In our view, these recommendations do not only apply to the healthcare system in Germany where the data were collected, but can be extrapolated to other settings as faster treatment is very likely to result in improved patient outcomes.

mically stable and unstable patients in established local STEMI treatment networks. In our view, these recommendations do not only apply to the healthcare system in Germany where the data were collected, but can be extrapolated to other settings as faster treatment is very likely to result in improved patient outcomes. Supplemental Material Scholz_et_al_Supplemental_data – Supplemental material for Prognostic significance of emergency department bypass in stable and unstable patients with ST-segment elevation myocardial infarction Click here for additional data file. Supplemental material, Scholz_et_al_Supplemental_data for Prognostic significance of emergency department bypass in stable and unstable patients with ST-segment elevation myocardial infarction by Karl Heinrich Scholz, Tim Friede, Thomas Meyer, Claudius Jacobshagen, Björn Lengenfelder, Jens Jung, Claus Fleischmann, Hiller Moehlis, Hans G Olbrich, Rainer Ott, Albrecht Elsässer, Stephen Schröder, Christian Thilo, Werner Raut, Andreas Franke, Lars S Maier and Sebastian KG Maier in European Heart Journal: Acute Cardiovascular Care Conflict of interest: Tim Friede reports personal fees for consultancies (including data monitoring committees) from Novartis, Bayer, Biogen, AstraZeneca, Janssen, Grünenthal, Pharmalog, SGS and Roche, all outside the submitted work. Furthermore, he has received research funding by the European Commission for statistical analyses on the EUTrigTreat (NCT01209494) and EU-CERT-ICD (NCT02064192) clinical studies. All relationships declared are modest. All other authors declare no conflict of interest.

log, SGS and Roche, all outside the submitted work. Furthermore, he has received research funding by the European Commission for statistical analyses on the EUTrigTreat (NCT01209494) and EU-CERT-ICD (NCT02064192) clinical studies. All relationships declared are modest. All other authors declare no conflict of interest. Funding: This study was supported by a grant from the German Heart Foundation and the Arbeitsgemeinschaft Leitender Kardiologischer Krankenhausärzte to KHS.

log, SGS and Roche, all outside the submitted work. Furthermore, he has received research funding by the European Commission for statistical analyses on the EUTrigTreat (NCT01209494) and EU-CERT-ICD (NCT02064192) clinical studies. All relationships declared are modest. All other authors declare no conflict of interest. Funding: This study was supported by a grant from the German Heart Foundation and the Arbeitsgemeinschaft Leitender Kardiologischer Krankenhausärzte to KHS. List of contributors: (in order of number of patients included up to 31 December 2015) Universitätsklinikum Göttingen (Claudius Jacobshagen, Kristina Schröder, Swetlana Hartmann, Lars S. Maier); Universitätsklinikum Würzburg (Björn Lengenfelder, Verena Reinhart, Sebastian K.G. Maier); St. Bernward-Krankenhaus Hildesheim (Karl H. Scholz, Dorothee Ahlersmann); Helios Klinikum Krefeld (Rainer Ott, Heinrich G. Klues, Alexander Bufe); Klinikum Oldenburg (Albrecht Elsässer, Susanne Grafmüller, Annette Schütz); Klinikum Wolfsburg (Claus Fleischmann, Rolf Engberding, Rüdiger Becker); Klinikum Darmstadt (Gerald S. Werner, Hiller Moehlis); Asklepios Klinik Langen (Hans G. Olbrich, Kerstin Eck); Städtisches Klinikum München Neuperlach (Harald Mudra, Martin Hug, Anamaria Stote); Klinikum Ingolstadt (Harald Franck, Monika-Krista Zackl, Karlheinz Seidl); Klinik am Eichert Göppingen (Stephen Schröder, Marion Steindl, Josef Steindl, Sophia Atseles); Klinikum Worms (Jens Jung, Birgit Nicklas); Klinikum Augsburg (Christian Thilo, Georg Waidhauser, Wolfgang v. Scheidt); Universitätsklinikum Jena (Attila Yilmaz, Hans R. Figulla, Daniel Kretzschmar, Corinna Schneider, Christian Schulze); Klinikum Lüneburg (Christian Weiß, Claus H. Müller); Krankenhaus Buchholz (Werner Raut, Klaus Hertting); Krankenhaus Landshut-Achdorf (Bernhard Zrenner, Josef Haimerl, Ute Zrenner); KRH Klinikum Hannover-Siloah (Andreas Franke, Jan Fürste); Klinikum Lippe-Detmold (Dirk Härtel, Melanie Kriete, Ulrich Tebbe, Stephan Gielen); Klinikum Leer (Christian Vahlhaus, Ralf-G. Pretzsch); Klinikum Ludwigsburg (Ralph Berroth, Joachim Geiger, Friederike Wunsch, Christian Wolpert); Robert-Bosch-Krankenhaus Stuttgart (Stephan Hill, Andrea Bullinger, Udo Sechtem); Klinikum Deggendorf (Edmond Skenderaj, Ulrich Valta-Seufzer, Martin Giesler); Kreiskrankenhaus Eschwege (Marco Lubitz, Peter Schott); Regio Klinikum Pinneberg (Konrad Gorski, Thomas Hofmann); Klinikverbund Kempten-Oberallgäu (Carsten Bauer, Wulf Ito); Klinikum Viersen (Nicolas v. Beckerath); Medizinische Hochschule Hannover (Jörn Tongers, Benedikta Ritter, Karin Hohenleitner); Sana Kliniken Lübeck (Hans Martin Grusnick, Joachim Weil); Klinikum St.

); Regio Klinikum Pinneberg (Konrad Gorski, Thomas Hofmann); Klinikverbund Kempten-Oberallgäu (Carsten Bauer, Wulf Ito); Klinikum Viersen (Nicolas v. Beckerath); Medizinische Hochschule Hannover (Jörn Tongers, Benedikta Ritter, Karin Hohenleitner); Sana Kliniken Lübeck (Hans Martin Grusnick, Joachim Weil); Klinikum St. Elisabeth Straubing (Sebastian K.G. Maier, Elke Grassl); Klinikum Regiomed-Kliniken Coburg (Caroline Kleinecke, Andrea Linss, Kerstin Truthan, Hans-Joachim Goller, Johannes Brachmann); Asklepios Harzklinik Goslar (Gaby Lehnert, Stefan Lange, Tobias Steffen, Arnd B. Buchwald, Christoph Engelhardt); Kliniken Ostallgäu-Kaufbeuren, Füssen (Simon Delladio, Martin Hinterseer, Myriam Parvanov); Hermann-Josef-Krankenhaus Erkelenz (Klaus Dieter Winter, Christina Ziesen); Kliniken Maria Hilf Mönchengladbach (Jürgen vom Dahl, Dierk Rulands); SLK Kliniken Heilbronn (Marcus Hennersdorf, Jens Martin Maier, Eva Schropp); Krankenhaus Rothenburg ob der Tauber (Christian Wacker); Kreiskrankenhaus Dormagen (Benjamin Orth, Georg Haltern); Marienkrankenhaus Soest (Roland Bürger, Markus Flesch); SLK Kliniken Am Plattenwald Bad Friedrichshall (Thomas Dengler); Universitätsklinikum Regensburg (Christina Strack, Dierk Endemann, Lars S. Maier); Klinikum Landkreis Erding (Lorenz Bott-Flügel); Klinikum Neumarkt (Veronika Lingg); Krankenhaus Bethanien Moers (Alexander Donath, Stefan Möhlenkamp); Vinzenzkrankenhaus Hannover (Beate Bugdoll, Petra Wucherpfennig, Jan-Bernd Schüttert, Christian Zellerhoff); Krankenhaus Henriettenstift Hannover (Thomas Weiss, Thorsten Grundmann); Evangelisches Krankenhaus Bethesda Mönchengladbach (Thomas Lickfeld); DRK-Krankenhaus Clementinenhaus Hannover (Heinz-Peter Remmlinger).

Introduction Acute kidney injury (AKI) is a frequent complication of cardiogenic shock, which is promoted by low cardiac output (CO) and subsequent organ hypoperfusion, i.e. kidney, liver and brain leading to sustained high morbidity and mortality.1–3 Thus the administration of fluids and inotropes to maintain CO and organ perfusion is the hallmark of cardiogenic shock therapy. However, increasing doses of vasopressors develop deleterious effects on organ perfusion, i.e. promote acute renal failure by increasing renal vascular resistance. Therefore, we used a microaxial mechanical circulatory support (MCS) device to maintain haemodynamic stability by augmenting CO and reducing vasoconstrictor demand hoping to improve organ perfusion.4–6 Monitoring of critical organ dysfunction, i.e. AKI, includes urine production as an index of renal perfusion and creatinine clearance as an index of glomerular filtration. However, these parameters do not reflect acute renal haemodynamic changes in the renal vasculature and are useless for the short-term management of mechanical circulatory devices, i.e. the Impella microaxial pump.7,8 Therefore, we here evaluated as an indicator for renal haemodynamics the renal resistive index (RRI) determined by intrarenal artery Doppler measurements. Even though there are controversial data on the pathophysiological relevance of RRI, it is significantly influenced by systemic haemodynamic parameters (i.e. in cardiogenic shock patients), it correlates with renal vascular resistance depicting changes in renal blood flow,7,9,10 while several data indicate that it can predict the occurrence and reversibility of kidney failure in critically ill patients.7,9–11 Thus the aim of this study was to investigate the effect of left ventricular mechanical support using the Impella microaxial pump on the RRI in otherwise stable patients with cardiogenic shock.

several data indicate that it can predict the occurrence and reversibility of kidney failure in critically ill patients.7,9–11 Thus the aim of this study was to investigate the effect of left ventricular mechanical support using the Impella microaxial pump on the RRI in otherwise stable patients with cardiogenic shock. Materials and methods The study was conducted during a 6-month period (May 2018 to October 2018). We included consecutive patients with cardiogenic shock supported with MCS by the Impella microaxial pump in this single-centre study. Cardiogenic shock was defined as systolic blood pressure less than 90 mmHg for more than 30 minutes or catecholamines required to maintain systolic blood pressure at more than 90 mmHg plus clinical signs of pulmonary congestion and impaired end-organ perfusion (at least one of the following: altered mental status, cold and clammy skin, oliguria with urine output <30 ml/hour or serum lactate >2.0 mmol/L). The RRI was obtained in every haemodynamically stable patient using Doppler ultrasound. The two measurements were performed within 6 hours of admission and within the time frame of one hour. The first measurement was obtained when haemodynamic stability of the patient with Impella support was achieved. Haemodynamic stability was defined as mean arterial pressure of 60 mmHg or greater for more than one hour with no changes of Impella MCS level, catecholamine doses or fluid administration rates. After an additional 30 minutes of MCS support the RRI was measured at a different support level. Between the two RRI measurements only the Impella MCS level was changed, whereas all other therapeutic interventions, especially fluid management and doses of catecholamines, remained unchanged.

fluid administration rates. After an additional 30 minutes of MCS support the RRI was measured at a different support level. Between the two RRI measurements only the Impella MCS level was changed, whereas all other therapeutic interventions, especially fluid management and doses of catecholamines, remained unchanged. RRI was determined using Doppler ultrasound at the patient’s bedside according to standard procedures (Figure 1).12,13 A transparietal 2–6 MHz pulsed-wave Doppler probe (Philips Sparq) was used. Kidneys and interlobar arteries were localised using sonography and colour Doppler. Pulse-wave Doppler measurements in the interlobar arteries were then obtained. On each kidney three pulse-wave measurements were performed and RRI values were averaged to obtain mean values. RRI was defined as (peak systolic velocity – end diastolic velocity)/ peak systolic velocity. All RRI measurements were performed by one investigator experienced in kidney Doppler ultrasonography and certified in echocardiography. Normal values for native kidneys are reported between 0.6 and 0.7. In order to assess the intraobserver variability, the RRI was measured previously in a separate cohort of 10 healthy volunteers by the same investigator. The intraclass correlation coefficient (ICC) was then calculated and had a value of 0.997 (95% confidence interval (CI) 0.991–0.999) with a variance of 0.008. Figure 1. Renal Doppler ultrasound with renal resistive index (RRI) measurement.

RRI was determined using Doppler ultrasound at the patient’s bedside according to standard procedures (Figure 1).12,13 A transparietal 2–6 MHz pulsed-wave Doppler probe (Philips Sparq) was used. Kidneys and interlobar arteries were localised using sonography and colour Doppler. Pulse-wave Doppler measurements in the interlobar arteries were then obtained. On each kidney three pulse-wave measurements were performed and RRI values were averaged to obtain mean values. RRI was defined as (peak systolic velocity – end diastolic velocity)/ peak systolic velocity. All RRI measurements were performed by one investigator experienced in kidney Doppler ultrasonography and certified in echocardiography. Normal values for native kidneys are reported between 0.6 and 0.7. In order to assess the intraobserver variability, the RRI was measured previously in a separate cohort of 10 healthy volunteers by the same investigator. The intraclass correlation coefficient (ICC) was then calculated and had a value of 0.997 (95% confidence interval (CI) 0.991–0.999) with a variance of 0.008. Figure 1. Renal Doppler ultrasound with renal resistive index (RRI) measurement. The RRI is calculated from the peak systolic and end-diastolic velocities of arterial blood flow in the renal cortex (RRI = peak systolic velocity – end diastolic velocity/ peak systolic velocity).

RRI was determined using Doppler ultrasound at the patient’s bedside according to standard procedures (Figure 1).12,13 A transparietal 2–6 MHz pulsed-wave Doppler probe (Philips Sparq) was used. Kidneys and interlobar arteries were localised using sonography and colour Doppler. Pulse-wave Doppler measurements in the interlobar arteries were then obtained. On each kidney three pulse-wave measurements were performed and RRI values were averaged to obtain mean values. RRI was defined as (peak systolic velocity – end diastolic velocity)/ peak systolic velocity. All RRI measurements were performed by one investigator experienced in kidney Doppler ultrasonography and certified in echocardiography. Normal values for native kidneys are reported between 0.6 and 0.7. In order to assess the intraobserver variability, the RRI was measured previously in a separate cohort of 10 healthy volunteers by the same investigator. The intraclass correlation coefficient (ICC) was then calculated and had a value of 0.997 (95% confidence interval (CI) 0.991–0.999) with a variance of 0.008. Figure 1. Renal Doppler ultrasound with renal resistive index (RRI) measurement. The RRI is calculated from the peak systolic and end-diastolic velocities of arterial blood flow in the renal cortex (RRI = peak systolic velocity – end diastolic velocity/ peak systolic velocity). The study was approved by the local ethics committee of the Philipps University of Marburg, which waived the need for written informed consent, as renal Doppler ultrasonography is an existing feature of our clinical practice and the augmentation of Impella flow level was performed in stable patients without any alterations in the systematic haemodynamic parameters.

f the variables. We assessed normality using the Shapiro–Wilk test as well as Pearson tests. After testing for normal distribution, Student’s t-test or Mann–Whitney test was implemented to test for differences between the various characteristics. Intraobserver variability was calculated based on the ICC and its 95% CI. Results The study included 15 patients with infarct-related cardiogenic shock supported with an Impella. The demographics and baseline characteristics of these patients are reported in Table 1. Mean age was 66.7 ± 14 years and 73% were men. Mean vasopressor and inotropes doses were 8.9 ± 14.7 µg/min noradrenaline and 233 ± 200 µg/min dobutamine. The systolic left ventricular ejection fraction was 31 ± 7%. Doppler ultrasonography was performed within 6 hours after admission on the intensive care unit. The RRI could be calculated for both kidneys in 13 patients and for one kidney in two patients. The mean difference between right and left RRI was 0.026 ± 0.023, P=0.72. No patient had a difference greater than 0.05. The RRI decreased significantly from 0.66 ± 0.08 to 0.62 ± 0.06 (P<0.001), when increasing the Impella support by a mean of 0.44 L/min (±0.2 L/min) (Table 2), while both systolic and diastolic blood pressure remained unchanged. The decreasing tendency in RRI was consistent in each individual patient (Figure 2). Moreover, intra-renal peak systolic or peak diastolic velocity (Figures 3 and 4) remained unchanged. Table 1. Demographics and baseline characteristics.

Results The study included 15 patients with infarct-related cardiogenic shock supported with an Impella. The demographics and baseline characteristics of these patients are reported in Table 1. Mean age was 66.7 ± 14 years and 73% were men. Mean vasopressor and inotropes doses were 8.9 ± 14.7 µg/min noradrenaline and 233 ± 200 µg/min dobutamine. The systolic left ventricular ejection fraction was 31 ± 7%. Doppler ultrasonography was performed within 6 hours after admission on the intensive care unit. The RRI could be calculated for both kidneys in 13 patients and for one kidney in two patients. The mean difference between right and left RRI was 0.026 ± 0.023, P=0.72. No patient had a difference greater than 0.05. The RRI decreased significantly from 0.66 ± 0.08 to 0.62 ± 0.06 (P<0.001), when increasing the Impella support by a mean of 0.44 L/min (±0.2 L/min) (Table 2), while both systolic and diastolic blood pressure remained unchanged. The decreasing tendency in RRI was consistent in each individual patient (Figure 2). Moreover, intra-renal peak systolic or peak diastolic velocity (Figures 3 and 4) remained unchanged. Table 1. Demographics and baseline characteristics. Age (years) 67.81 ± 14.18 ΒΜΙ (kg/m²) 26.4 ± 2.6 LVEF (%) 31 ± 7 Male/female 11/4 Cause of CS AMI 14 Acute myocarditis 1 Creatinine (mg/dl) 1.286 ± 0.684 Heart rate (bpm) 102 ± 21 SAP (mmHg) 109.3 ± 17.19 DAP (mmHg) 60 ± 10 MAP (mmHg) 85.9 ± 13.2 Noradrenaline (µg/min) 8.9 ± 14.7 Dobutamine (µg/min) 233 ± 200 Renal longitudinal length (cm) 9.58 ± 0.9 Renal parenchymal thickness (cm) 2 ± 0.3 BMI: body mass index; LVEF: left ventricular ejection fraction; CS: cardiogenic shock; AMI: acute myocardial infarction; SAP: systolic arterial pressure; DAP: diastolic arterial pressure; MAP: mean arterial pressure.

e (µg/min) 233 ± 200 Renal longitudinal length (cm) 9.58 ± 0.9 Renal parenchymal thickness (cm) 2 ± 0.3 BMI: body mass index; LVEF: left ventricular ejection fraction; CS: cardiogenic shock; AMI: acute myocardial infarction; SAP: systolic arterial pressure; DAP: diastolic arterial pressure; MAP: mean arterial pressure. Table 2. RRI values at different Impella support levels. Patient Impella flow (L/min) RRI 1 1.4 0.76 1.9 0.71 2 1.4 0.73 2.4 0.62 3 1.3 0.76 2 0.69 4 1.6 0.55 1.9 0.52 5 2 0.56 2.2 0.54 6 3 0.63 3.4 0.59 7 2.1 0.74 2.5 0.65 8 2.2 0.78 2,6 0.74 9 1.5 0.63 2 0.58 10 2.2 0.74 2.5 0.69 11 2 0.56 2.3 0.55 12 1.6 0.64 2 0.63 13 1.6 0.62 1.9 0.61 14 2.8 0.62 3.4 0.58 15 1.8 0.64 2.1 0.59 RRI: renal resistive index. Figure 2. Individual RRI profiles in relation to Impella support. In every patient a reduction of the RRI was observed after increasing Impella support. RRI: renal resistive index. Figure 3. (a) Impact of the augmentation of the Impella flow level on the peak systolic velocity in each patient and between the two time points. The peak systolic velocity was increased only in three patients, there was no significant difference in the levels of the peak systolic velocity between baseline and after augmentation of the Impella flow level. (b) Impact of the augmentation of the Impella flow level on the peak diastolic velocity in each patient and between the two time points. The peak diastolic velocity was increased only in five patients, there was no significant difference in the levels of the peak diastolic velocity between baseline and after augmentation of the Impella flow level.

(b) Impact of the augmentation of the Impella flow level on the peak diastolic velocity in each patient and between the two time points. The peak diastolic velocity was increased only in five patients, there was no significant difference in the levels of the peak diastolic velocity between baseline and after augmentation of the Impella flow level. Figure 4. Impact of the augmentation of the Impella flow level on mean arterial pressure. The mean arterial pressure did not change after the increase in the Impella flow level compared to baseline. Discussion The RRI has been studied intensively not only to gain diagnostic and prognostic insights into a variety of renal pathologies (such as the progression of chronic kidney disease and renal allograft rejection), but also for the prediction of renal outcomes in critically ill patients.7,11 Darmon and colleagues found that RRI values greater than 0.75 predicted persistent AKI with a good sensitivity and specificity in critically ill patients with mechanical ventilation.11 In this study, the performance of the RRI was better than urinary indices for predicting AKI.11 Moreover, the RRI has been shown to predict AKI with high sensitivity and specificity in the immediate postoperative period after cardiac surgery.14

y and specificity in critically ill patients with mechanical ventilation.11 In this study, the performance of the RRI was better than urinary indices for predicting AKI.11 Moreover, the RRI has been shown to predict AKI with high sensitivity and specificity in the immediate postoperative period after cardiac surgery.14 Here we investigated for the first time the effects of left ventricular mechanical support (MCS) using the Impella microaxial pump on the RRI in patients with cardiogenic shock. The present study shows that a significant decrease in RRI can be observed when increasing CO by Impella MCS without any changes in systolic or diastolic blood pressure (Figure 2).

e the effects of left ventricular mechanical support (MCS) using the Impella microaxial pump on the RRI in patients with cardiogenic shock. The present study shows that a significant decrease in RRI can be observed when increasing CO by Impella MCS without any changes in systolic or diastolic blood pressure (Figure 2). The RRI is used for assessing instant renal perfusion7 and is one of the most sensitive parameters of renal vascular resistance, which in turn depicts alterations of renal blood flow.10 Therefore, analysing the intrarenal arterial waveforms obtained by Doppler ultrasonography might be useful in patients with cardiogenic shock for the detection of renal hypoperfusion. Prompting then adequate treatment decisions in order to improve renal perfusion may prevent or attenuate persistent AKI.15 Such a prompt response would not be possible if therapeutic manoeuvres are based on delayed criteria of AKI such as serum creatinine or low urine output.11,14 AKI, which often develops in critically ill patients such as in cardiogenic shock, is associated with increased morbidity and mortality.1,2,16,17 Therefore, monitoring kidney function and the early detection of renal hypoperfusion in patients with cardiogenic shock is crucial for implementation of therapeutic measures and adjusting haemodynamic strategies in cardiogenic shock.

s in cardiogenic shock, is associated with increased morbidity and mortality.1,2,16,17 Therefore, monitoring kidney function and the early detection of renal hypoperfusion in patients with cardiogenic shock is crucial for implementation of therapeutic measures and adjusting haemodynamic strategies in cardiogenic shock. Decreased renal blood flow and renal venous congestion are independent determinants of worsening renal function in patients with heart failure in addition to neurohormonal activation, including activation of the sympathetic nervous system.18 A decrease in CO will cause the autoregulatory mechanisms of renal perfusion to reduce renal vascular resistance in order to maintain renal perfusion.19 In heart failure and cardiogenic shock the hyperactivation of the sympathetic nervous system increases vascular resistance and may lead to a decrease in renal perfusion, especially in the presence of reduced CO.19 Moreover, vasopressors, which are often used in cardiogenic shock, may further reduce renal perfusion and increase RRI by direct vasoconstriction.20 In particular, vasopressors, such as norepinephrine, may have vasoconstrictive effects on renal vessels as doses increase, inducing an increase in vascular resistance and thereby reducing renal blood flow.

often used in cardiogenic shock, may further reduce renal perfusion and increase RRI by direct vasoconstriction.20 In particular, vasopressors, such as norepinephrine, may have vasoconstrictive effects on renal vessels as doses increase, inducing an increase in vascular resistance and thereby reducing renal blood flow. On the other hand, the maintenance of continuous flow during Impella support in cardiogenic shock may increase CO, reduce vasopressor doses4–6,21 and thereby improve renal perfusion and decrease RRI. In patients undergoing high-risk percutaneous coronary intervention, including patients with severely depressed systolic left ventricular function and cardiogenic shock, Impella support significantly reduced the risk of AKI.22 The repetitive finding of RRI decrease after augmentation of the support through the microaxial Impella pump underlines a causality, which may suggest the importance of Impella support as part of a renal protective strategy. In conclusion, increasing Impella support in patients with cardiogenic shock led to a significant reduction of the RRI, suggesting improved renal perfusion. Determining the optimal haemodynamic support in patients with cardiogenic shock not only on systemic haemodynamic parameters but also on regional perfusion indices such as the RRI may be beneficial in optimising end-organ perfusion. Whether RRI may in future be a relevant endpoint to titrate Impella support in patients with cardiogenic shock or not remains to be answered in future studies.

c shock not only on systemic haemodynamic parameters but also on regional perfusion indices such as the RRI may be beneficial in optimising end-organ perfusion. Whether RRI may in future be a relevant endpoint to titrate Impella support in patients with cardiogenic shock or not remains to be answered in future studies. Limitations Our observations are obviously limited by the retrospective and non-randomised and open-label design of our study. However, this is the first study to investigate the effects of Impella support on the RRI in patients with cardiogenic shock. Detailed right heart catheter haemodynamic data before implantation of the Impella device were not available for all patients, but in emergency situations extensive invasive haemodynamic measurements are often not routinely performed. Another limitation of our study is the small number of patients included. However, the purpose of our investigation was to assess the effects of Impella support on the RRI and was not powered to evaluate renal outcomes. Larger studies with longer periods of assessment are needed to determine the effect of titrating Impella support using the RRI on the prevention of acute renal injury. Conflict of interest: BS, KK, BM and UL have received speaker’s honoraria from Abiomed. NP, GC and HA have no conflicts of interest to disclose. Funding: The authors received no financial support for the research, authorship, and/or publication of this article.