Browse the corpus

Introduction Neonatal hyperbilirubinemia and neurological sequelae resulting from kernicterus pose a serious medical burden in both developing and developed countries [13, 15, 16, 23]. However, severe hyperbilirubinemia in the newborn is preventable through appropriate follow-up, diagnosis, and treatment, such as phototherapy and exchange transfusions [8, 9, 13, 23]. The American Academy of Pediatrics (AAP) recommends that all neonates undergo total serum bilirubin (TSB) or transcutaneous bilirubin (TcB) measurements at least once before hospital discharge to assess their risk of hyperbilirubinemia [2, 17]. TcB nomogram is a useful tool for neonatal hyperbilirubinemia screening [10]. However, the TcB value, rate of rise and peak varies with country and ethnicity [5], suggesting that each country and ethnic group has specific risk factors for hyperbilirubinemia. Few studies on the TcB nomogram have been done for Mongolian neonates in a resource-limiting setting [1]. The aim of this study was to create an hour-specific TcB nomogram for healthy Mongolian neonates, which was achieved by obtaining TcB measurements in the first 144 h after birth.

TcB nomogram is a useful tool for neonatal hyperbilirubinemia screening [10]. However, the TcB value, rate of rise and peak varies with country and ethnicity [5], suggesting that each country and ethnic group has specific risk factors for hyperbilirubinemia. Few studies on the TcB nomogram have been done for Mongolian neonates in a resource-limiting setting [1]. The aim of this study was to create an hour-specific TcB nomogram for healthy Mongolian neonates, which was achieved by obtaining TcB measurements in the first 144 h after birth. Methods Patients This was a prospective cohort study at the National Center for Maternal and Child Health (NCMCH), a tertiary care hospital in Ulaanbaatar, Mongolia. Neonates who were eligible for enrollment were healthy newborns ≥35 weeks of gestation with a birth weight ≥2000 g who were delivered in the hospital on a weekday (from Monday to Friday morning) and whose address were in the Bayangol district. Exclusion criteria included the following: infants with birth asphyxia, respiratory distress, NICU admission, congenital infection, Rh isoimmunization, jaundice within 24 h of birth, skin infection, congenital spinal anomaly, those with no TcB measurement, and cases with missing information. The study was carried out from October 15, 2012 to October 4, 2013 but was temporarily suspended between June 28 and August 2, 2013 when the maternity unit was closed for renovation. Perinatal information on the mother and infant were extracted from their medical charts including sex, gestational age, birth weight, mode of delivery, duration of hospitalization, maternal age, parity, gravidity, maternal blood type, family history of siblings, and feeding type before and after discharge.

renovation. Perinatal information on the mother and infant were extracted from their medical charts including sex, gestational age, birth weight, mode of delivery, duration of hospitalization, maternal age, parity, gravidity, maternal blood type, family history of siblings, and feeding type before and after discharge. This study was approved by the NCMCH ethics committee in Mongolia and the National Center for Global Health and Medicine (NCGM) human investigation ethics committee in Japan, in accordance with the Declaration of Helsinki. Informed consent was obtained from all individual participants (mothers of the infants) included in the study.

s approved by the NCMCH ethics committee in Mongolia and the National Center for Global Health and Medicine (NCGM) human investigation ethics committee in Japan, in accordance with the Declaration of Helsinki. Informed consent was obtained from all individual participants (mothers of the infants) included in the study. Measurements TcB measurements were performed between 9 am and 5 pm twice daily (once in the morning and once in evening) from Monday to Friday. The Jaundice Meter JM-103 (Konica Minolta, Osaka, Japan) was used by properly trained pediatricians or a research technician to measure TcB. The operation of the JM-103 and its measurement technique have been described previously [1]. Briefly, TcB levels were measured at the forehead and at the midsternum three times at each site, and the median value was defined as the TcB at each measurement site. The highest value of two sites was used as the TcB for each infant. Because NCMCH applies a discharge policy of ≥48 h for vaginally delivered infants and ≥72 h for those delivered by cesarean section, the parents were advised to return to the outpatient clinic for follow-up measurements. Enrolled neonates returned for a follow-up visit between days 3–6 or earlier if the parents noticed worsening jaundice in their child. Parents who did not bring their child for the visit were reminded by telephone. An enrollment number was given to each neonate and recorded in a log book to identify the study neonate during the visit. TcB values were recorded on a flow sheet designed for the study and attached to the signed informed consent form in the medical record of each infant.

child for the visit were reminded by telephone. An enrollment number was given to each neonate and recorded in a log book to identify the study neonate during the visit. TcB values were recorded on a flow sheet designed for the study and attached to the signed informed consent form in the medical record of each infant. TcB measurements obtained at 6 ± 2 h intervals up to 144 h were examined [12, 18]. TcB measurements not taken between these time measurements were omitted from analysis. The decision regarding the need for phototherapy was made by the attending pediatrician based on TSB levels, in accordance with NCMCH’s protocol and/or AAP guidelines [2]. If TcB values were obtained, measurement values obtained after the initiation of phototherapy were excluded from analysis. The characteristics of mothers and children are presented as percentages for binary and categorical data, and as means and standard deviations for continuous data. The data analysis was performed by using SPSS 15.0 for Windows (SPSS, Chicago, IL). TcB percentiles (95th, 90th, 75th, 50th, 25th, 10th, and 5th percentiles) were calculated for each designated time. A TcB nomogram with smoothed percentile curves for the period 6–144 postnatal hours was prepared using Microsoft Excel 2010 (Microsoft, Redmond, WA).

The characteristics of mothers and children are presented as percentages for binary and categorical data, and as means and standard deviations for continuous data. The data analysis was performed by using SPSS 15.0 for Windows (SPSS, Chicago, IL). TcB percentiles (95th, 90th, 75th, 50th, 25th, 10th, and 5th percentiles) were calculated for each designated time. A TcB nomogram with smoothed percentile curves for the period 6–144 postnatal hours was prepared using Microsoft Excel 2010 (Microsoft, Redmond, WA). Results Of 1846 live births in NCMCH during the study period, 549 neonates did not meet the enrollment criteria (Fig. 1), resulting in 1297 healthy term and late-preterm neonates from whom 5084 TcB measurements were obtained. Among these TcB measurements, 3487 that were taken at the designated time were used for analysis. Of the 1297 enrolled neonates, 1137 (87.7 %) returned for the follow-up visit after hospital discharge.Fig. 1 Study flow Demographic and background data of mothers and neonates are summarized in Table 1. Of note, all neonates were Mongolian, 27.1 % were delivered by a cesarean section, ≥90 % were breastfed, and 17.8 % had a birth weight ≥4000 g. Regarding the family history of jaundice, 851 of 1297 enrolled infants had siblings and 163 of 851 siblings (19.2 %) had a history of treatment for jaundice according to an interview with the mother. Details of the jaundice could not be obtained to clarify the etiology of jaundice.Table 1 Characteristics of the enrolled neonates and their mothers

of jaundice, 851 of 1297 enrolled infants had siblings and 163 of 851 siblings (19.2 %) had a history of treatment for jaundice according to an interview with the mother. Details of the jaundice could not be obtained to clarify the etiology of jaundice.Table 1 Characteristics of the enrolled neonates and their mothers Characteristic n = 1297 (%) 1. Sex Male 662 (51.0) Female 635 (49.0) 2. Gestational age 35–36 weeks 16 (1.2) 37–40 week 1230 (94.9) 41–42 weeks 51 (3.9) 3. Birth weight 2000–2999 g 92 (7.1) 3000–3999 g 975 (75.1) 4000–4999 g 228 (17.6) >5000 g 2 (0.2) 4. Mode of delivery Vaginal 945 (72.9) Cesarean section 352 (27.1) 5. Duration of hospitalization 1–2 day (s) 907 (69.9) 3–4 days 355 (27.4) 5–7 days 35 (2.7) 6. Maternal age 15–24 years old 293 (22.6) 25–34 years old 748 (57.7) 35–45 years old 256 (19.7) 7. Parity 1 320 (24.7) 2 411 (31.7) 3 258 (19.9) 4 169 (13.0) 5 to 9 139 (10.7) 8. Gravidity 1 451 (34.8) 2 541 (41.7) 3 230 (17.7) 4 60 (4.6) 5 to 9 15 (1.2) 9. Maternal blood type O 366 (28.2) A 347 (26.8) B 456 (35.2) AB 116 (8.9) Unknown 11 (0.8) No description 1 (0.1) 10. Family history of siblings First baby 446 (34.4) Phototherapy 161 (12.4) Exchange transfusion 2 (0.2) No jaundice 688 (53.0) 11. Feeding (before discharge) Breastfeeding 1264 (97.4) Formula 3 (0.2) Mixed 30 (2.3) 12. Feeding (follow-up visit) No follow-up visit 160 Attended follow-up visit 1137 Breastfeeding 1025 (90.1) Formula 1 (0.1) Mixed 111 (9.8)

baby 446 (34.4) Phototherapy 161 (12.4) Exchange transfusion 2 (0.2) No jaundice 688 (53.0) 11. Feeding (before discharge) Breastfeeding 1264 (97.4) Formula 3 (0.2) Mixed 30 (2.3) 12. Feeding (follow-up visit) No follow-up visit 160 Attended follow-up visit 1137 Breastfeeding 1025 (90.1) Formula 1 (0.1) Mixed 111 (9.8) A TcB nomogram with smoothed percentile curves (95th, 90th, 75th, 50th, 25th, 10th, and 5th percentiles) for the time period 6–144 postnatal hours is presented in Fig. 2. The numbers of TcB measurements are shown for each time point. TcB levels increased in a linear fashion most rapidly in the first 24 h and less rapidly from 24 to 78 h, reaching a plateau after 78 h for the 50th percentile.Fig. 2 An hour-specific TcB nomogram for Mongolian neonates with smoothed curves for the 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles. Numbers in parentheses indicate the number of TcB measurements at each time point Discussion We created an hour-specific TcB nomogram for the first 144 h after birth for Mongolian healthy term and late preterm neonates. The characteristics of this group of Mongolian neonates included delivery by cesarean section in 26.9 % of cases and exclusive breastfeeding in >90 %. The TcB values and rate of rise at each designated time point resulted in a TcB nomogram with higher values than those reported for other countries and ethnicities in previous studies. This result indicates that this group might have risk factors and a higher likelihood of developing neonatal hyperbilirubinemia.

>90 %. The TcB values and rate of rise at each designated time point resulted in a TcB nomogram with higher values than those reported for other countries and ethnicities in previous studies. This result indicates that this group might have risk factors and a higher likelihood of developing neonatal hyperbilirubinemia. Many TcB nomogram studies have been performed in North America, Europe, and Asia in different groups [3, 6, 7, 11, 12, 14, 18, 19, 21, 25, 28] since the second generation of the TcB meter became available. The risk factors of hyperbilirubinemia were determined to be Asian ethnicity, exclusive breastfeeding, weight loss, and a sibling with neonatal jaundice [17]. In a North American study investigating an hour-specific TcB nomogram in the first 96 h using the JM-103 to measure levels in 3984 neonates, 73.1 % were Caucasian, 45.1 % were delivered by cesarean section, and 66.2 % were breastfed [18]. TcB nomogram for Mongolian neonates showed much higher values than the North American nomogram at each time point. This might be due to the predominant Caucasian ethnicity of the study group [22]. In a study on infants of Asian ethnicity, a TcB nomogram was developed using JM-103 measurements taken from 6035 Chinese infants [28]. These Chinese infants were delivered by cesarean section in 55.5 % of cases, 25.1 % were breastfed, and 9.4 % had a birth weight >4000 g. In contrast, our study of Mongolian neonates showed higher TcB values at each time point, suggesting that exclusive breastfeeding might have contributed to this result [4]. A recent study on the influence of dehydration on body weight loss at 72 h after birth in Taiwan revealed the optimum body weight loss cutoff points for predicting hyperbilirubinemia [27]. Unfortunately, the body weights on each day of TcB measurement were not determined in our study. Therefore, it remains unclear whether the acceleration of hyperbilirubinemia of Mongolian neonates is caused by exclusive breastfeeding itself or by dehydration secondary to inappropriate breastfeeding.

rbilirubinemia [27]. Unfortunately, the body weights on each day of TcB measurement were not determined in our study. Therefore, it remains unclear whether the acceleration of hyperbilirubinemia of Mongolian neonates is caused by exclusive breastfeeding itself or by dehydration secondary to inappropriate breastfeeding. This prospective cohort study was a tertiary hospital-based study that enrolled neonates from one district, and TcB measurements were done either in the maternity unit or in the outpatient clinic [26]. Our study method for creating a TcB nomogram is practical and can be applied in countries with limited resources. This encourages resource-limited countries to create country- and ethnic-specific TcB nomograms.

m one district, and TcB measurements were done either in the maternity unit or in the outpatient clinic [26]. Our study method for creating a TcB nomogram is practical and can be applied in countries with limited resources. This encourages resource-limited countries to create country- and ethnic-specific TcB nomograms. This study has several limitations. First, there were insufficient numbers of TcB measurements at the follow-up visit on day 3–6, only one to two measurements. Second, the designated times for TcB measurements were difficult to establish, leading to the omission of a certain number of TcB measurements. However, it was impossible to fix a designated TcB measurement time for each neonate in ordinary practice. Third, the glucose-6-phosphate dehydrogenase (G6PD) and direct Coombs test are not routinely performed in prenatal care screening in Mongolia so G6PD deficiency and ABO incompatibility could not be ruled out in this group. Fourth, cephalohematoma, a risk factor for hyperbilirubinemia, might be increased, considering the high percentage of vaginal deliveries and heavy infants (17.8 % had birth weight >4000 g) [20, 24]. Although cephalohematoma should have been identified as a risk factor, this does not affect the validity of the nomogram. Fifth, because infants with significant hyperbilirubinemia were referred to the hospital for further work-up and treatment and were later excluded from the study, the TcB measurements and curves we report here might be lower than the actual TcB values.

s a risk factor, this does not affect the validity of the nomogram. Fifth, because infants with significant hyperbilirubinemia were referred to the hospital for further work-up and treatment and were later excluded from the study, the TcB measurements and curves we report here might be lower than the actual TcB values. The findings in this study contribute to improving appropriate diagnosis and treatment of neonatal hyperbilirubinemia, and might reduce the incidence of neurological impairment in this high-risk ethnic group. Conclusion We provide data on TcB levels for the first 144 postnatal hours that were derived from a group of Mongolian term and late preterm neonates. The higher values of the TcB nomogram derived from Mongolian neonates may be due to their Asian ethnicity and exclusive breastfeeding. Abbreviations AAPAmerican Academy of Pediatrics G6PDGlucose-6 pyruvate deficiency GAGestational age TcBTranscutaneous bilirubin TSBTotal serum bilirubin We thank Dr. Gerentuya Dashzeveg for performing TcB measurements and maternity unit staff for their collaboration in this research project. Role of the funding source This work was supported in part by Grants-in-Aid for Research from the National Center for Global Health and Medicine (24A-1, 26A-201 and 27A-1404). Conflict of interest The authors declare that they have no conflicts of interest.

We thank Dr. Gerentuya Dashzeveg for performing TcB measurements and maternity unit staff for their collaboration in this research project. Role of the funding source This work was supported in part by Grants-in-Aid for Research from the National Center for Global Health and Medicine (24A-1, 26A-201 and 27A-1404). Conflict of interest The authors declare that they have no conflicts of interest. Compliance with ethical standards This study was approved by the National Center for Maternal and Child Health ethics committee in Mongolia and the National Center for Global Health and Medicine human investigation ethics committee in Japan. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. A written informed consent was obtained from all mothers.

ll procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. A written informed consent was obtained from all mothers. Authors’ contributions MAA conceptualized and designed the study, carried out the initial analyses, drafted the initial manuscript, and approved the final manuscript as submitted. NY conceptualized and advised on the analysis, reviewed the manuscript, and approved the final manuscript as submitted. RM conceptualized and reviewed the manuscript, and approved the final manuscript as submitted. SH, TM reviewed the manuscript, and approved the final manuscript as submitted. KS, GN designed the data collection instruments, coordinated and reviewed the manuscript, and approved the final manuscript as submitted. BB designed the data collection instruments, coordinated and supervised data collection, reviewed the manuscript, and approved the final manuscript as submitted. ES reviewed the manuscript and approved the final manuscript as submitted.

Introduction Since the invention of the Macintosh and Miller laryngoscope blades in the 1940s, direct laryngoscopy (DL) has been considered as the “gold standard” of endotracheal intubation (ETI). However, according to scientific studies, the effectiveness of the ETI on children performed by paramedics using a standard laryngoscope in pre-hospital care is insufficient and ranges from 63.4 to 77 % [8, 9]. In the light of the fact that one in four children requiring ETI and adequate ventilation is not intubated, or the endotracheal tube is incorrectly inserted [8, 9]. The ETI is considered the standard for securing the airway of severely ill or injured patients [1, 33]. According to current guidelines on the treatment of severely injured patients [2], emergency ETI should be performed immediately on all patients with a Glasgow Coma Scale (GCS) < 9 by emergency medical service (EMS) providers. Also, the 2010 European Resuscitation Council (ERC) [1] and American Heart Association (AHA) resuscitation guidelines [11] emphasize ETI as an airway management method during cardiopulmonary resuscitation (CPR). The ERC guidelines for CPR recommend that chest compressions are continued and interruptions are minimized during CPR, and ETI during resuscitation should be performed quickly and efficiently by an experienced operator, while interruptions to chest compressions should be avoided where possible.

ary resuscitation (CPR). The ERC guidelines for CPR recommend that chest compressions are continued and interruptions are minimized during CPR, and ETI during resuscitation should be performed quickly and efficiently by an experienced operator, while interruptions to chest compressions should be avoided where possible. Securing the airway using a tracheal tube brings many benefits. Firstly, it allows the use of asynchronous resuscitation while eliminating chest compression interruption for performing rescue breaths [1, 33]. It is also possible to use positive end-expiratory pressure (PEEP), as well as the constant measurement of the concentration of carbon dioxide in exhaled air [30]. Opinions on the intubation of children by paramedics in the prehospital care are varied [10, 22, 38]. However, paramedic working in EMS in Poland must have the ability to intubation, both children and adults. Philip Ragg noticed the benefits of using video laryngoscopy during child intubation and suggested an extension to the algorithm of Difficult Airway Society (DAS) on the use of video laryngoscopes in “Plan A” [33]. Several studies indicate that the use of video laryngoscopy in emergency situations can increase the effectiveness of intubation [12, 14, 36]. The aim of the study was to compare time and success rates of the TruView PCD video laryngoscope and the Macintosh laryngoscope (MAC) for pediatric emergency intubation with three airway scenarios on a standardized manikin model.

Securing the airway using a tracheal tube brings many benefits. Firstly, it allows the use of asynchronous resuscitation while eliminating chest compression interruption for performing rescue breaths [1, 33]. It is also possible to use positive end-expiratory pressure (PEEP), as well as the constant measurement of the concentration of carbon dioxide in exhaled air [30]. Opinions on the intubation of children by paramedics in the prehospital care are varied [10, 22, 38]. However, paramedic working in EMS in Poland must have the ability to intubation, both children and adults. Philip Ragg noticed the benefits of using video laryngoscopy during child intubation and suggested an extension to the algorithm of Difficult Airway Society (DAS) on the use of video laryngoscopes in “Plan A” [33]. Several studies indicate that the use of video laryngoscopy in emergency situations can increase the effectiveness of intubation [12, 14, 36]. The aim of the study was to compare time and success rates of the TruView PCD video laryngoscope and the Macintosh laryngoscope (MAC) for pediatric emergency intubation with three airway scenarios on a standardized manikin model. Methods This open, prospective, randomized, crossover manikin study was approved by the Program Committee of the International Institute of Rescue Research and Education (Head: Dr. A. Kurowski, 10.2014.05.15 on September 3rd, 2014). Prior to the study commencing, it was registered at the ClinicalTrials register (www.clinicaltrials.gov, identifier NCT02289872). With voluntary written, informed consent, 120 paramedics were recruited that satisfied the following inclusion criteria: (1) they had not performed more than 100 clinical adult (human) intubations by DL and no experience with clinical pediatric (human) intubation, and (2) they had not received any training in ETI using TruView device prior to the study. The study was conducted between November and December 2014.

owing inclusion criteria: (1) they had not performed more than 100 clinical adult (human) intubations by DL and no experience with clinical pediatric (human) intubation, and (2) they had not received any training in ETI using TruView device prior to the study. The study was conducted between November and December 2014. Simulation of the scenario Each participant performed orotracheal intubations on a PediaSIM CPR training manikin (FCAE HealthCare, Sarasota, FL, USA). Subjects participated in three airway scenarios:The control scenario, in which neither chest compression nor cervical stabilization was applied during intubation. The chest compression scenario, in which continuous chest compression was applied using the LUCAS-2 chest compression system (Physio-Control, Redmond, WA, USA). Chest compression was provided at a rate of 100 min−1 to a depth of 5–6 cm during all intubation procedures. The chest compression with cervical stabilization scenario, in which both chest compression using Lucas-2 and cervical stabilization were applied. A correctly fitting standard cervical immobilization collar (StifNeck Select, Laerdal, Stavanger, Norway) was applied to the manikin’s neck to prevent movement of the cervical spine. In each scenario, the manikin was placed in a neutral position on the floor of a well-lit room. The elevation of the head or the upper body was not allowed.

The chest compression with cervical stabilization scenario, in which both chest compression using Lucas-2 and cervical stabilization were applied. A correctly fitting standard cervical immobilization collar (StifNeck Select, Laerdal, Stavanger, Norway) was applied to the manikin’s neck to prevent movement of the cervical spine. In each scenario, the manikin was placed in a neutral position on the floor of a well-lit room. The elevation of the head or the upper body was not allowed. Devices All the participants completed a 45-min training program prior the study, including an introduction to the anatomy and physiology of the airway and the techniques of ETI using a laryngoscope with Macintosh blade no. 2 (MAC; HEINE Optotechnik, Munich, Germany) and the TruView PCD video laryngoscope (TruView; Blade # 2, Truphatek Int.; Netanya, Israel) (Fig. 1). The Truview PCD video is intended to enable the medical professional to perform routine and difficult oral intubation cases while using a minimal amount of force and with a reduced rate of side effects to the patient, such as sore throat or soft tissue damage. TruView blade can be connected to a dedicated 5-in. LCD monitor via a unique camera for obtaining clear visual pictures of the intubation process. In this way, clinical safety is greatly improved and the incidence of incorrectly positioned endotracheal tube is reduced. The addition of oxygen during the intubation procedure via the unique oxygen port on the Truview PCD blades serves to slow the rate of desaturation, prevents the accumulation of mist and secretions on the lenses, and ensures a clear visual picture of the entire procedure [7, 18]. All intubations were performed using a tracheal tube with 5.0-mm internal diameter (ID). Lubricant was pre- applied to the tracheal tube, and a 10 mL syringe to block the tube’s cuff as well as an AMBU resuscitator bag (AMBU, Copenhagen, Denmark) were readily available and within range of the participants. After the training section, the participants were given 10 min to practice ETI with the three laryngoscopes.Fig. 1 Laryngoscopes used for this study were a Macintosh Laryngoscope, b TruView PCD Videolaryngoscope

BU resuscitator bag (AMBU, Copenhagen, Denmark) were readily available and within range of the participants. After the training section, the participants were given 10 min to practice ETI with the three laryngoscopes.Fig. 1 Laryngoscopes used for this study were a Macintosh Laryngoscope, b TruView PCD Videolaryngoscope Study protocol A Research Randomizer program was used [www.researchrandomizer.com] to divide the participants into six groups and determine the order in which the different ETI devices were applied within each group. The first group attempted ETI using the MAC in scenario A, the second using the MAC in scenario B, the third using the MAC in scenario C, the fourth using the TruView in scenario A, the fifth using the TruView in scenario B, and the sixth using the TruView in scenario C (Fig. 2). After completing the ETI procedure, the participants had a 10 min break before performing intubation using another laryngoscope. The participants had a maximum of three attempts for ETI with each intubation method.Fig. 2 Flow chart of design and recruitment of participants according to CONSORT statement The participants were reminded before each attempt that the “patient” needs emergency ETI as quickly as possible to give them the feeling of time pressure that would be present in real emergency patients. Measurements and outcomes The primary outcome of the study was time to intubation, defined as the time from insertion of the laryngoscope blade between the teeth to the first manual ventilation of the manikin’s lungs.

The participants were reminded before each attempt that the “patient” needs emergency ETI as quickly as possible to give them the feeling of time pressure that would be present in real emergency patients. Measurements and outcomes The primary outcome of the study was time to intubation, defined as the time from insertion of the laryngoscope blade between the teeth to the first manual ventilation of the manikin’s lungs. The secondary outcome was success of the intubation attempt (i.e., tracheal or oesophageal placement of the tube) which was recorded when the success of the ventilation attempt was confirmed by the manikin’s ventilation indicators. After each attempt, the participants were asked to rate the glottic view they had during the attempt using a Cormac and Lehane Grade [4]. The severity of the potential dental trauma was calculated based on a previously described [29] grading scale of the pressure on the teeth (0 = none, 1 = mild, 2 = moderate, and 3 = severe) by the same investigator. To access subjective opinions about the difficulty of the each intubation method, the participants were asked to give a rating on a visual analogue scale (VAS) with a score from 1 (extremely easy) to 10 (extremely difficult).

sure on the teeth (0 = none, 1 = mild, 2 = moderate, and 3 = severe) by the same investigator. To access subjective opinions about the difficulty of the each intubation method, the participants were asked to give a rating on a visual analogue scale (VAS) with a score from 1 (extremely easy) to 10 (extremely difficult). Statistical analysis Times needed to successful intubation were compared using the Wilcoxon signed-rank test. McNemar’s test was used to detect possible differences in success rates for ETI. For all statistical analysis, the R statistical package version 3.0.0 for Windows was used. P < 0.05 was considered as statistically significant. For comparisons of VAS, a one-way analysis of variance with a post hoc (Scheffé’s) test was used. Results are shown as means ± standard deviation (SD) or absolute numbers and percentages. Results Demographic testing One hundred twenty paramedics (46 female, 38.3 %) participated in this study. No participant had previously performed a pediatric intubation with any laryngoscope. Eighty-seven participants (21 female, 24.1 %) worked in EMS teams, 33 participants (25 female, 75.5 %) worked in hospital emergency units. Mean age was 27.5 ± 5.8 years, and mean work experience was 3.7 ± 2.1 years.

ted in this study. No participant had previously performed a pediatric intubation with any laryngoscope. Eighty-seven participants (21 female, 24.1 %) worked in EMS teams, 33 participants (25 female, 75.5 %) worked in hospital emergency units. Mean age was 27.5 ± 5.8 years, and mean work experience was 3.7 ± 2.1 years. Scenario A: the control scenario In the control scenario, overall effectiveness of intubation using the MAC and TruView was 100 %. However, the success rate after the first attempt using the MAC and TruView varied and amounted to 95.8 vs. 100 %, respectively. The average times to successful intubation using MAC and TruView are presented in Fig. 3. Time to intubation was achieved fastest with TruView (18.5 ± 4.5 s) and was significantly slower with MAC (24.3 ± 6.2 s, p = 0.017).Fig. 3 Comparison of time to intubation of the study devices in seconds Scenario B: the chest compression scenario In the chest compression scenario, the difference in time to intubation between MAC and TruView was significant (25.7 ± 7.1 s vs. 21.6 ± 4.1 s; p = 0.023). The success rate after the first attempt using the distinct laryngoscopes varied and amounted to 57.5 vs. 100 % (MAC and TruView, respectively). There was a statistically significant difference between MAC and TruView in first intubation attempt effectiveness (p < 0.001) and overall effectiveness (p = 0.011).

= 0.023). The success rate after the first attempt using the distinct laryngoscopes varied and amounted to 57.5 vs. 100 % (MAC and TruView, respectively). There was a statistically significant difference between MAC and TruView in first intubation attempt effectiveness (p < 0.001) and overall effectiveness (p = 0.011). Scenario C: the chest compression with cervical stabilization scenario In the chest compression with cervical stabilization scenario, time to intubation was significantly longer with the MAC (45.4 ± 10.3 s) than the TruView (28.9 s ± 5.6 s; p < 0.001). First-attempt success was achieved in 98.3 % of the TruView group compared with 45.8 % in the MAC group (p < 0.001). Overall effectiveness of TruView was 100 %, which was significantly higher than that of the MAC (68.4 %, p < 0.001) (Table 1).Table 1 Intubation success for all scenarios Success rate Scenario A Scenario B Scenario C MAC TruView MAC TruView MAC TruView First (%) 115 (95.8 %) 120 (100 %) 69 (57.5 %) 120 (100 %) 55 (45.8 %) 118 (98.3 %) Second (%) 120 (100 %) 120 (100 %) 90 (75 %) 120 (100 %) 78 (65.0 %) 120 (100 %) Third (%) 120 (100 %) 120 (100 %) 98 (81.7 %) 120 (100 %) 82 (68.3 %) 120 (100 %) Failed (%) 0 (0.0 %) 0 (0.0 %) 22 (18.3 %) 0 (0.0 %) 38 (31.7 %) 0 (0.0 %) MAC Macintosh Laryngoscope, TruView TruView PCD Video-laryngoscope. Scenario A The chest compression scenario, Scenario B The chest compression scenario, Scenario C The chest compression with cervical stabilization scenario

Success rate Scenario A Scenario B Scenario C MAC TruView MAC TruView MAC TruView First (%) 115 (95.8 %) 120 (100 %) 69 (57.5 %) 120 (100 %) 55 (45.8 %) 118 (98.3 %) Second (%) 120 (100 %) 120 (100 %) 90 (75 %) 120 (100 %) 78 (65.0 %) 120 (100 %) Third (%) 120 (100 %) 120 (100 %) 98 (81.7 %) 120 (100 %) 82 (68.3 %) 120 (100 %) Failed (%) 0 (0.0 %) 0 (0.0 %) 22 (18.3 %) 0 (0.0 %) 38 (31.7 %) 0 (0.0 %) MAC Macintosh Laryngoscope, TruView TruView PCD Video-laryngoscope. Scenario A The chest compression scenario, Scenario B The chest compression scenario, Scenario C The chest compression with cervical stabilization scenario Measures of difficulty in intubation Glottic view quality was better with TruView than MAC in all scenarios, p < 0.001 (Table 2). Dental compression was also significantly lower with the TruView compared to MAC in all scenarios, p < 0.001 (Table 2). The ETI was most easily achieved with TruView compared to MAC in all scenarios: 2.4 vs. 2.6 points, for scenario A (p = 0.75); 2.7 vs. 3.9 points, for scenario B (p = 0.012); and 3.4 vs. 5.6 points, for scenario C (p < 0.001), respectively.Table 2 Measures of difficulty in intubation

all scenarios, p < 0.001 (Table 2). The ETI was most easily achieved with TruView compared to MAC in all scenarios: 2.4 vs. 2.6 points, for scenario A (p = 0.75); 2.7 vs. 3.9 points, for scenario B (p = 0.012); and 3.4 vs. 5.6 points, for scenario C (p < 0.001), respectively.Table 2 Measures of difficulty in intubation Parameter Scenario A Scenario B Scenario C MAC TruView MAC TruView MAC TruView Reported Cormack-Lehane grade I 118 (98.3 %) 120 (100 %) 79 (65.8 %) 120 (100 %) 41 (34.2 %) 114 (95.0 %) II 2 (1.7 %) 0 (0.0 %) 27 (22.5 %) 0 (0.0 %) 56 (46.7 %) 6 (5.0 %) III 0 (0.0 %) 0 (0.0 %) 14 (11.7 %) 0 (0.0 %) 23 (19.2 %) 0 (0.0 %) IV 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) Dental compression scale 0 34 (28.3 %) 107 (89.2 %) 14 (11.7 %) 99 (82.5 %) 11 (9.2 %) 59 (49.2 %) 1 56 (46.7 %) 13 (10.8 %) 69 (57.5 %) 21 (17.5 %) 41 (34.2 %) 49 (40.8 %) 2 27 (22.5 %) 0 (0.0 %) 30 (25.0 %) 0 (0.0 %) 49 (40.8 %) 12 (10.0 %) 3 3 (2.5 %) 0 (0.0 %) 7 (5.8 %) 0 (0.0 %) 19 (15.8 %) 0 (0.0 %) MAC Macintosh Laryngoscope, TruView TruView PCD Video-laryngoscope, Scenario A The chest compression scenario, Scenario B The chest compression scenario, Scenario C The chest compression with cervical stabilization scenario

Parameter Scenario A Scenario B Scenario C MAC TruView MAC TruView MAC TruView Reported Cormack-Lehane grade I 118 (98.3 %) 120 (100 %) 79 (65.8 %) 120 (100 %) 41 (34.2 %) 114 (95.0 %) II 2 (1.7 %) 0 (0.0 %) 27 (22.5 %) 0 (0.0 %) 56 (46.7 %) 6 (5.0 %) III 0 (0.0 %) 0 (0.0 %) 14 (11.7 %) 0 (0.0 %) 23 (19.2 %) 0 (0.0 %) IV 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) Dental compression scale 0 34 (28.3 %) 107 (89.2 %) 14 (11.7 %) 99 (82.5 %) 11 (9.2 %) 59 (49.2 %) 1 56 (46.7 %) 13 (10.8 %) 69 (57.5 %) 21 (17.5 %) 41 (34.2 %) 49 (40.8 %) 2 27 (22.5 %) 0 (0.0 %) 30 (25.0 %) 0 (0.0 %) 49 (40.8 %) 12 (10.0 %) 3 3 (2.5 %) 0 (0.0 %) 7 (5.8 %) 0 (0.0 %) 19 (15.8 %) 0 (0.0 %) MAC Macintosh Laryngoscope, TruView TruView PCD Video-laryngoscope, Scenario A The chest compression scenario, Scenario B The chest compression scenario, Scenario C The chest compression with cervical stabilization scenario Discussion The DL using a laryngoscope with either a Miller or Macintosh blade is the main method of child intubation. The MAC is suitable for the treatment of children over 2 years [23, 37]. Therefore, a training minikin resembling a 6 years old was used in the study and intubation was performed with Macintosh blade. However, it should be noted that the effectiveness of intubation on children performed by paramedics in prehospital conditions using a laryngoscope with Miller or Macintosh blades is varied and ranges from 63.4 to 77 % [8, 9, 35, 38]. The problem of unsatisfactory efficacy of the first attempts of pediatric intubation applies not only to paramedics, but also to doctors who are not anesthesiologists [3]. Due to this, video laryngoscopy may be an alternative to DL, both for children and adults ETI [19, 28].

and ranges from 63.4 to 77 % [8, 9, 35, 38]. The problem of unsatisfactory efficacy of the first attempts of pediatric intubation applies not only to paramedics, but also to doctors who are not anesthesiologists [3]. Due to this, video laryngoscopy may be an alternative to DL, both for children and adults ETI [19, 28]. Following a single training session, the 120 paramedics recruited for our study had more pediatric intubation success with the TruView than the MAC. Previously, no study had compared TruView and MAC in pediatric intubations performed by paramedics in simulated chest compression scenarios or chest compression with cervical stabilization scenarios. Chest compressions increased the time to intubation for both devices: a mean time of 3.4 s for MAC and 3.1 s for TruView. Other studies have also shown that time to intubation using DL increases when uninterrupted chest compressions are applied [13, 34, 39].

Following a single training session, the 120 paramedics recruited for our study had more pediatric intubation success with the TruView than the MAC. Previously, no study had compared TruView and MAC in pediatric intubations performed by paramedics in simulated chest compression scenarios or chest compression with cervical stabilization scenarios. Chest compressions increased the time to intubation for both devices: a mean time of 3.4 s for MAC and 3.1 s for TruView. Other studies have also shown that time to intubation using DL increases when uninterrupted chest compressions are applied [13, 34, 39]. The overall effectiveness of intubation using the MAC in our study was 100 % for the control scenario, 81.7 % for the chest compression scenario, and 68.3 % for the chest compression with cervical stabilization scenario. Time to intubation in these scenarios varied and amounted to 22.3 vs. 25.7 s vs. 45.4 s, respectively. Mutlak et al. showed that effectiveness of the MAC for routine tracheal intubation in infants with normal airways was 100 %, and time to intubation was 26 s [20]. The study by Rodríguez-Núñez et al. [27] evaluating the intubation time using a Miller laryngoscope and GlideScope videolaryngoscope performed by 23 residents, indicated that the videolaryngoscope Glidescope® does not improve performance in this setting, and the time to intubation using Miller laryngoscope was 28.2 s (20.4–34.4). In our study with normal airway intubation, overall effectiveness of using the MAC was also 100 %. Nileshwar and Garg showed that the success ratio of orotracheal intubation in pediatric patients with simulated restriction of cervical spine movements using a short-handled MAC by anaesthesiologists was 100 % [21].

4.4). In our study with normal airway intubation, overall effectiveness of using the MAC was also 100 %. Nileshwar and Garg showed that the success ratio of orotracheal intubation in pediatric patients with simulated restriction of cervical spine movements using a short-handled MAC by anaesthesiologists was 100 % [21]. The mean intubation time using TruView during the chest compression scenario was 21.6 s, which was comparable to the results obtained in another study (20.1 s; IQR 18–23.3 s) [33]. Overall, the effectiveness of intubation using TruView was 100 % for the control scenario, the chest compression scenario, and the chest compression with cervical stabilization scenario. Time to intubation in these scenarios varied and amounted to 18.5 s vs. 21.6 s vs. 28.9 s, respectively. The success ratio of intubation during chest compression using TruView in study by Szarpak et al. was also 100 % [33]. In the study by Riveros et al. concerning patients (neonate up to 10 years of age) who were scheduled for general surgical procedures, times to intubation were 44 s and 23 s with the Truview PCD and DL, respectively [30]. In the case of TruView PCD, time to intubation was shorter in each scenario than in the study by Riveros et al. [26]. This may be due to the fact that intubation was performed on a minikin, not on a human.

eneral surgical procedures, times to intubation were 44 s and 23 s with the Truview PCD and DL, respectively [30]. In the case of TruView PCD, time to intubation was shorter in each scenario than in the study by Riveros et al. [26]. This may be due to the fact that intubation was performed on a minikin, not on a human. In our study, the Cormack-Lehane graded views attained using the TruView PCD video laryngoscope were superior to the views attained using Macintosh laryngoscopy in all scenarios. The study by Riveros et al. showed no differences in Cormac-Lehane views between TruView and MAC [26]. Many studies have shown the superiority of video laryngoscopy over DL, especially in emergency intubation for both pediatric and adult patients [5, 6, 16, 25, 31].

tained using Macintosh laryngoscopy in all scenarios. The study by Riveros et al. showed no differences in Cormac-Lehane views between TruView and MAC [26]. Many studies have shown the superiority of video laryngoscopy over DL, especially in emergency intubation for both pediatric and adult patients [5, 6, 16, 25, 31]. Several limitations have to be noted. First, the procedures were performed on manikins, not on live subjects. Manikin studies can never fully replace studies on humans; however, the decision to use a standardized airway model was made intentionally as manikin studies allow researchers to simulate clinical practice conditions with strict standardization, thus allowing them to investigate thoroughly [24]. Several studies, on the other hand, have shown that the manikin used in considered to be the best manikin overall for the tasks performed in this study [15, 32]. Besides, these devices have not been compared in this situation in a randomized, controlled trial. Moreover, according to the International Liaison Committee on Resuscitation (ILCOR), randomized clinical trials for cases of cardiac arrest are unethical and cannot determine the expected benefits of CPR [17]. The second limitation is that we used inexperienced intubators; therefore, the results may have been less pronounced in more experienced hands. However, we believed that novice intubators would offer a more reliable comparison because they had little prior experience in pediatric intubation with either technique and would be less likely to display any bias. Although all participants prior the study received an 45-min standardized intubation training session. The strengths of this study include the use of a highly advanced patient simulator for performing pediatric advanced life support and the randomized crossover procedure.

e and would be less likely to display any bias. Although all participants prior the study received an 45-min standardized intubation training session. The strengths of this study include the use of a highly advanced patient simulator for performing pediatric advanced life support and the randomized crossover procedure. The results from our study showing higher efficiency intubation using Truview PCD by paramedics show that short training is sufficient in order in this professional group to performed highly proficient with TruView PCD during intubation manikin. Further clinical studies are necessary to confirm these initial positive findings. Conclusions The TruView offers better intubation conditions than the MAC on a pediatric manikin in all the scenarios test. The TruView may be used to elevate the epiglottis for orotracheal intubation. Further clinical studies are necessary to confirm these initial positive findings. Abbreviations CPRCardiopulmonary resuscitation DASDifficult airway society DLDirect laryngoscopy ERCEuropean Resuscitation Council EMSEmergency Medical Service ETIEndotracheal intubation GCSGlasgow Coma Scale MACMacintosh laryngoscope SDStandard deviation PEEPPositive end-expiratory pressure VASVisual Analogue Scale TruView PCD video laryngoscope used was lent from Cardinal Wyszynski National Institute of Cardiology (Warsaw, Poland) and was returned. The authors alone are responsible for the content and writing of this paper. The authors want to thank all paramedics for their participation in this study.

PEEPPositive end-expiratory pressure VASVisual Analogue Scale TruView PCD video laryngoscope used was lent from Cardinal Wyszynski National Institute of Cardiology (Warsaw, Poland) and was returned. The authors alone are responsible for the content and writing of this paper. The authors want to thank all paramedics for their participation in this study. Conflict of interest statement All authors have no conflicts of interest to declare. None of authors involved in this study have any financial relationship with any manufacturers of intubation devices. Author contributions LS, LC, AK, and ZT contributed significantly to the planning of the study and the study design. LS, LC, AK, and ZT recruited participants and collected data. LS was principal investigators of this study and did major manuscript preparation. LS and LC performed statistical analysis. LS, LC, AK, and ZK contributed significantly for manuscript editing and expertise. Source of support No sources of financial and material support to be declared.

Introduction The ductus arteriosus fails to close after birth in 30 to 60 % of prematurely born infants [16]. This condition—patent ductus arteriosus (PDA)—is associated with a prolonged ventilation need and carries an increased risk of morbidity (i.e., necrotizing enterocolitis, chronic lung disease) and even mortality [10, 15, 18]. Pharmacological closure with non-steroidal anti-inflammatory drugs (NSAIDs), mainly ibuprofen and indomethacin, is currently the standard of care [27]. NSAIDs are not effective in around 30 % of patients, however, and can have side effects such as gastrointestinal bleeding and perforation, diminished platelet aggregation, hyperbilirubinemia, and transient renal function impairment [22, 23]. Moreover, NSAIDs are contraindicated in a considerable proportion of newborns, notably those with renal failure, intracerebral hemorrhage, gastrointestinal problems, and thrombocytopenia. If NSAIDs fail or are contraindicated, the only currently available solution is surgical ligation, which is associated with the risks of cardiothoracic surgery and impaired neurological outcome [17]. Therefore, alternative pharmacological interventions are needed.

rhage, gastrointestinal problems, and thrombocytopenia. If NSAIDs fail or are contraindicated, the only currently available solution is surgical ligation, which is associated with the risks of cardiothoracic surgery and impaired neurological outcome [17]. Therefore, alternative pharmacological interventions are needed. Paracetamol has been suggested as an alternative drug for PDA closure [21]. More than 10 observational and retrospective studies have described oral or intravenous high-dose paracetamol treatment with varying effectiveness [1, 8, 9, 11, 12, 19, 21, 26, 29] (see Table 1). Two recent prospective randomized controlled trials comparing oral paracetamol with ibuprofen both showed a slightly favorable effect of paracetamol (closure rate 81.2 versus 78.8 % for ibuprofen) [5]. The other trial by Oncel et al. even showed a higher closure success rate in the paracetamol group (97.5 versus 95 % in the ibuprofen group) after a maximum of two courses of ibuprofen or paracetamol [20].Table 1 Literature review on PDA treatment with paracetamol

ol (closure rate 81.2 versus 78.8 % for ibuprofen) [5]. The other trial by Oncel et al. even showed a higher closure success rate in the paracetamol group (97.5 versus 95 % in the ibuprofen group) after a maximum of two courses of ibuprofen or paracetamol [20].Table 1 Literature review on PDA treatment with paracetamol Author Dose (mg/kg/day) Treatment interval (h) Route Treatment PCM (days) Gestational age (weeks) PN age start PCM (days) Patients that achieved ductal closure/no surgical ligation 1. Hammerman 2011 15 6 Oral Max. 7 26–296/7 3–35 5/5 2. Yurttutan 2013 15 6 Oral Max. 6 26–30 3–7 5/6 3. Oncel 2013 15 6 Intravenous 3–6 24–29 2–15 10/10 4. Alan 2013 15 6 Intravenous Max. 19 262/7–335/7 8–19 0/3 5. Özmert 2013 15 6 Oral 3–6 23–32 20–47 5/7 6. Sinha 2013 15 8 Oral 2 27–33 4–7 10/10 7. Kessel 2013 15 6 Oral 3–11 26–30 ? 7/7 8. Jasani 2013 15 6 Oral 2.3–4.3 28.5–31.1 2.6–8.9 9. Dang 2013 RCT 15 6 Oral 3 31.2 ± 1.8 65/80 10. Oncel 2013 RCT 15 6 Oral 3–6 ≤26 2–3 12/13 15 6 Oral 3–6 <28 2–3 22/23 11. Tekgunduz 2014 15 6 Intravenous 1 29 3 0/1a 10 8 Intravenous 1–4 24–31 2–9 10/12b 12. Nadir 2014 15 6 Oral Max. 7 24–27 2–22 4/7 13. el-Khuffash 2014 15 6 Oral 2 26–33 14–56 0/5 15 6 Oral 7 26–30 8–35 6/7 15 6 Intravenous 2–5 26–32 3–41 9/9 14. Terrin 2014 7.5–15 max. 60 4–6 Intravenous 3 26 ± 2 2.8 ± 1.2 6/8 15. Roofthooft 2014 15 6 Intravenous 3–7.5 236/7–264/7 3–33 6/33 16. el-Khuffash 2015 15 6 Intravenous Max. 6 24.6–27.9 16–39 24/30 aTransaminases elevated: paracetamol treatment stopped after three doses, ductal closure with oral ibuprofen

12. Nadir 2014 15 6 Oral Max. 7 24–27 2–22 4/7 13. el-Khuffash 2014 15 6 Oral 2 26–33 14–56 0/5 15 6 Oral 7 26–30 8–35 6/7 15 6 Intravenous 2–5 26–32 3–41 9/9 14. Terrin 2014 7.5–15 max. 60 4–6 Intravenous 3 26 ± 2 2.8 ± 1.2 6/8 15. Roofthooft 2014 15 6 Intravenous 3–7.5 236/7–264/7 3–33 6/33 16. el-Khuffash 2015 15 6 Intravenous Max. 6 24.6–27.9 16–39 24/30 aTransaminases elevated: paracetamol treatment stopped after three doses, ductal closure with oral ibuprofen bDuctal closure of two remaining PDAs with oral ibuprofen after paracetamol After publication of the first studies on paracetamol and PDA closure, we added intravenous paracetamol to our PDA treatment guideline. However, the results in the first patients did not match the high closure rates of other studies, and only 20 % of patients did not need further PDA treatment after paracetamol [24]. Based on the promising results of other published studies, we decided to continue paracetamol treatment for PDA closure in preterm infants with ibuprofen contraindications or ibuprofen treatment failure. In the current study, we describe the evaluation of the efficacy of intravenous paracetamol on PDA closure in very low birth weight (VLBW) infants admitted to our hospital.

After publication of the first studies on paracetamol and PDA closure, we added intravenous paracetamol to our PDA treatment guideline. However, the results in the first patients did not match the high closure rates of other studies, and only 20 % of patients did not need further PDA treatment after paracetamol [24]. Based on the promising results of other published studies, we decided to continue paracetamol treatment for PDA closure in preterm infants with ibuprofen contraindications or ibuprofen treatment failure. In the current study, we describe the evaluation of the efficacy of intravenous paracetamol on PDA closure in very low birth weight (VLBW) infants admitted to our hospital. Methods In a prospective observational, single center study performed from December 2012 until September 2014, at the level III NICU of the Erasmus MC—Sophia Children’s Hospital in Rotterdam, the Netherlands, we included all neonates with a gestational age of less than 28 weeks and a birth weight of less than 1500 g, diagnosed with a hemodynamically significant patent ductus arteriosus (hsPDA) using clinical and cardiac ultrasound evaluation. Findings in the first nine patients in the current study were also presented in a preliminary report in 2014 [4].

gestational age of less than 28 weeks and a birth weight of less than 1500 g, diagnosed with a hemodynamically significant patent ductus arteriosus (hsPDA) using clinical and cardiac ultrasound evaluation. Findings in the first nine patients in the current study were also presented in a preliminary report in 2014 [4]. The first drug of choice in our department for PDA treatment was intravenous ibuprofen (Neobrufen©; Pedea ©), a single daily dose for 3 days (10 mg/kg on the first day, 5 mg/kg on the second and third days), and a repeated 3-day course if closure was not yet obtained. Paracetamol (Perfalgan©; Bristol-Meyers Squibb) was given if two courses of ibuprofen had no effect or if ibuprofen was contraindicated. Intravenous paracetamol 15 mg/kg every 6 h (60 mg/kg/day) was administered for a minimum of 3 days and a maximum of 7 days. Based on the indication for paracetamol, we created three groups. Group A: ibuprofen contraindicated and paracetamol as first drug of choice (primary contraindication); group B: development of a contraindication for ibuprofen during treatment with ibuprofen as first choice and switch to paracetamol (incomplete ibuprofen); and group C: two complete courses of ibuprofen failed to achieve closure and switch to paracetamol (complete ibuprofen) (see Fig. 1).Fig. 1 Flowchart included patients

oup B: development of a contraindication for ibuprofen during treatment with ibuprofen as first choice and switch to paracetamol (incomplete ibuprofen); and group C: two complete courses of ibuprofen failed to achieve closure and switch to paracetamol (complete ibuprofen) (see Fig. 1).Fig. 1 Flowchart included patients Contraindications for ibuprofen treatment were active intracerebral hemorrhage, thrombocytopenia or other known clotting disorders, severe sepsis, suspected or confirmed necrotizing enterocolitis (NEC), intestinal perforation, liver and kidney function disorders (oliguria <1.0 ml/kg/h, serum creatinine >110 μmol/l) and severe hyperbilirubinemia. Echocardiographic examination was performed by the echocardiosonographer or pediatric cardiologist before the start of paracetamol treatment, after 3 days and after 7 days. Measurements included ductus diameter, PDA diameter/LPA (left pulmonary artery) diameter, and LA/Ao (left atrial to aortic root) ratio. Two-dimensional color Doppler echocardiography was performed using a Vivid-S6 (GE Health Care) ultrasound system equipped with a 10-MHz transducer.

ment, after 3 days and after 7 days. Measurements included ductus diameter, PDA diameter/LPA (left pulmonary artery) diameter, and LA/Ao (left atrial to aortic root) ratio. Two-dimensional color Doppler echocardiography was performed using a Vivid-S6 (GE Health Care) ultrasound system equipped with a 10-MHz transducer. Cardiac ultrasound studies were done at the bedside by different echocardiosonographers, and measurements were checked by one pediatric cardiologist (I. M. v B.). HsPDA closure was defined as no flow through the duct. An open duct with diameter <1.5 mm, without significant left-to-right shunt, PDA:LPA diameter <0.5 and LA/Ao ratio <1.4 was defined as small PDA and not hemodynamically significant. HsPDA was defined as a ductal diameter of >2.0 mm, PDA:LPA diameter >0.8 and/or LA/Ao ratio >1.6. Statistical analysis Patient characteristics are presented as medians (IQR: interquartile range) in case of non-normally distributed variables and as means (standard deviations) in case of normally distributed variables. PDA diameters, PDA:LPA ratio and LA/Ao ratio before, during, and after treatment with paracetamol were compared using paired t tests. Fisher exact tests were used in case of categorical variables. Data analyses were performed with SPSS version 21.0 (SPSS Inc.). A p value of 0.05 was set as statistically significant.

s. PDA diameters, PDA:LPA ratio and LA/Ao ratio before, during, and after treatment with paracetamol were compared using paired t tests. Fisher exact tests were used in case of categorical variables. Data analyses were performed with SPSS version 21.0 (SPSS Inc.). A p value of 0.05 was set as statistically significant. Results A total of 33 VLBW infants with a median gestational age of 251/7 weeks (range 236/7–266/7, IQR 1.66) and a median birth weight of 750 g (range 365–1130, IQR 327) were included. Intravenous paracetamol was started at a median postnatal age of 14 days (IQR 12). Background characteristics and clinical outcome are shown in Table 2.Table 2 Background characteristics Characteristics Group A N = 13 Group B N = 8 Group C

Results A total of 33 VLBW infants with a median gestational age of 251/7 weeks (range 236/7–266/7, IQR 1.66) and a median birth weight of 750 g (range 365–1130, IQR 327) were included. Intravenous paracetamol was started at a median postnatal age of 14 days (IQR 12). Background characteristics and clinical outcome are shown in Table 2.Table 2 Background characteristics Characteristics Group A N = 13 Group B N = 8 Group C N = 12 Gestational age (weeks); median 25.2 24.3 25.8 Range 24.0–26.4 24.0–26.3 23.6–26.6 IQR 0.8 0.7 1.0 Birth weight (g): median 650 730 868 Range 400–1130 365–820 480–990 IQR 360.0 305.0 231.3 Small for gestational age: n (%) 5 (38.5) 3 (37.5) 2 (16.7) Gender Male: n (%) 9 (69.2) 4 (50) 6 (50) Female: n (%) 4 (30.8) 4 (50) 6 (50) Died: n (%) 4 (30.8) 4 (50) 0 (0) Post natal age: median/IQR 51/24.5 30/34.3 Antenatal steroids: n (%) 11 (84.6) 8 (100) 12 (100) PIH: n (%) 3 (23.1) 2 (25) 2 (16.7) Cesarean section: n (%) 9 (69.2) 5 (62.5) 7 (58.3) Mechanical ventilation: n (%) 12 (92.3) 7 (87.5) 11 (91.7) Surfactant treatment 10 (76.9) 7 (87.5) 11 (91.7) Diuretics 11 (84.6) 4 (50) 11 (91.7) Fluid restriction 9 (69.2) 3 (37.5) 8 (66.7) PNA start PCM: median (days) 12.0 12.5 16.5 IQR 11.5 14.75 10.75 Paracetamol treatment 6.0 6.5 5.5 Days in total (median/IQR) 3 2.75 4 Surgical ligation: n (%) 5 (38.5) 6 (75.0) 12 (100) PDA diameter before start PCM (mm): median/IQR 2.4/1.30 1.9/1.13 .4/0.83 PDA diameter after 3 days PCM (mm): median/IQR 1.9/0.90 2.1/1.13 2.1/1.08 PDA diameter after 7 days PCM (mm): median/IQR 1.8/1.28 2.0/0.40 2.6/1.60 PDA:LPA ratio before start PCM median/IQR 0.85/0.55 0.85/0.45 0.90/0.30 PDA:LPA ratio after 3 days PCM median/IQR 0.90/0.60 0.80/0.15 0.95/0.50 PDA:LPA ratio after 7 days PCM median/IQR 0.75/0.58 0.80/0.30 0.85/0.28 LA/Ao ratio before start PCM median/IQR 1.6/0.30 1.7/0.53 1.75/0.30 LA/Ao ratio after 3 days PCM median/IQR 1.64/0.60 1.7/0.70 1.8/0.55 LA/Ao ratio after 7 days PCM median/IQR 1.4/0.55 1.9/0.40 1.7/0.65 NT-proBNP (pmol/l)a: median 1097 2102 3078 IQR 3849.3 5832.0 4981.8

days PCM median/IQR 0.75/0.58 0.80/0.30 0.85/0.28 LA/Ao ratio before start PCM median/IQR 1.6/0.30 1.7/0.53 1.75/0.30 LA/Ao ratio after 3 days PCM median/IQR 1.64/0.60 1.7/0.70 1.8/0.55 LA/Ao ratio after 7 days PCM median/IQR 1.4/0.55 1.9/0.40 1.7/0.65 NT-proBNP (pmol/l)a: median 1097 2102 3078 IQR 3849.3 5832.0 4981.8 PIH pregnancy induced hypertension syndrome, PNA postnatal age, PCM paracetamol, LPA left pulmonary artery, LA-Ao left atrium-Aorta, NT-proBNP N-terminal pro-brain natriuretic peptide aAll NT-proBNP values were determined on day 3 PNA Median duration of paracetamol treatment was 6 days (IQR 3); the median cumulative dose was 360 mg (IQR 180). Ductal closure or no hsPDA after treatment was achieved in six of the 33 patients (18 %). In total, 23 patients (76.7 %) needed surgical ligation for hsPDA with clinical symptoms. Findings in the three different groups (see Fig. 1) are detailed below. Patients in group A (= primary contraindication for ibuprofen; N = 13, 39.4 %) received the first dose of paracetamol after a median of 12 (IQR 11.5) postnatal days and the median duration of the course was 6 (IQR 3) days. In six patients (46 %), the ductus arteriosus was closed or not hemodynamically significant after paracetamol and further treatment was not indicated. Two of the seven patients who did not respond to paracetamol treatment were successfully treated with ibuprofen (one and two courses, respectively) afterwards because the contraindication for ibuprofen had disappeared. Surgical ligation was performed in the other five (54 %).

ol and further treatment was not indicated. Two of the seven patients who did not respond to paracetamol treatment were successfully treated with ibuprofen (one and two courses, respectively) afterwards because the contraindication for ibuprofen had disappeared. Surgical ligation was performed in the other five (54 %). Patients in group B (= incomplete ibuprofen; N = 8, 24.2 %) received the first dose of paracetamol after a median of 12.5 (IQR 14.75) postnatal days and the median duration of the course was 6.5 (IQR 2.75) days. Two patients died on the 24th and 34th postnatal day (i.e., sepsis with extension of bilateral intraventricular hemorrhage with venous infarction, gram-negative bacterial infection) after paracetamol treatment (started on day 8 and 13, respectively) before a decision could be made to treat the persistent PDA with surgical closure. Paracetamol treatment was ineffective in the six remaining patients and all underwent surgical closure of the duct. Patients in group C (= complete ibuprofen; N = 12, 36.4 %) received paracetamol for a median of 5.5 days (IQR 4) after two courses of ibuprofen. At the start of paracetamol treatment, their median postnatal age was 16.5 (IQR 10.75) days. Paracetamol treatment was ineffective in all patients, and consequently, they all underwent surgical ligation.

ete ibuprofen; N = 12, 36.4 %) received paracetamol for a median of 5.5 days (IQR 4) after two courses of ibuprofen. At the start of paracetamol treatment, their median postnatal age was 16.5 (IQR 10.75) days. Paracetamol treatment was ineffective in all patients, and consequently, they all underwent surgical ligation. Both the surgical ligation rate and the mortality rate differed between the three groups. Surgical ligation was performed in 5/13 (38.5 %) patrients in group A; 6/8 (75 %) in group B; and 12/12 (100 %) in group C (Fisher exact test for surgical ligation p = 0.001). In total, eight patients died (Fisher exact test for death p = 0.025): 4/13 (31 %) in group A (median 51 days; IQR 24.5); 4/8 (50 %) in group B (median 30 days; IQR 34.3); and none in group C. Two of the four non-survivors in group B died from the consequences of NEC before surgical closure of the PDA could be performed. Cardiac ultrasound studies showed a statistically significant decrease in ductal diameter after paracetamol treatment only in group A, from median 2.4 to 1.8 mm after 7 days of treatment (p = 0.035, paired t test) (Fig. 2).Fig. 2 Change in ductus arteriosus diameter after 3 to maximum 7 days of intravenous PCM treatment for the three different groups (group A: primary contraindication for ibuprofen; group B: paracetamol after early stop of ibuprofen; and group C: paracetamol after complete ibuprofen treatment) All pre- and posttreatment measurements of kidney function (urea and creatinine) and liver function (transaminases, conjugated bilirubin) were normal.

Cardiac ultrasound studies showed a statistically significant decrease in ductal diameter after paracetamol treatment only in group A, from median 2.4 to 1.8 mm after 7 days of treatment (p = 0.035, paired t test) (Fig. 2).Fig. 2 Change in ductus arteriosus diameter after 3 to maximum 7 days of intravenous PCM treatment for the three different groups (group A: primary contraindication for ibuprofen; group B: paracetamol after early stop of ibuprofen; and group C: paracetamol after complete ibuprofen treatment) All pre- and posttreatment measurements of kidney function (urea and creatinine) and liver function (transaminases, conjugated bilirubin) were normal. Discussion In this study, high-dose intravenous paracetamol for the treatment of hsPDA in VLBW infants was overall effective in only 18 %. Looking at the subgroups of patients, paracetamol treatment was completely ineffective after previous ibuprofen treatment failure. However, it was effective in 46 % of the newborns with primary contraindications for ibuprofen.

us paracetamol for the treatment of hsPDA in VLBW infants was overall effective in only 18 %. Looking at the subgroups of patients, paracetamol treatment was completely ineffective after previous ibuprofen treatment failure. However, it was effective in 46 % of the newborns with primary contraindications for ibuprofen. The variability in success rates of this paracetamol therapy for PDA closure is hard to explain. Other studies reported success rates of 0 [7] up to 100 % [12, 19]. Success rates in two RCTs comparing oral paracetamol versus oral ibuprofen were 81 [5] and 94 % [20], respectively, thus, much higher than in our study. Still, in all but one of those studies, the same dosing regimen of 60 mg/kg/24 h paracetamol was used. In the study by Tekgunduz et al. the dose was lowered to 30 mg/kg/day after a patient showed elevated transaminases [13]. The duration of paracetamol treatment in our study was 6.0 days compared to 4.1 days [29] and 3.9 days [19] in the studies with high success rates. In previous studies, except in the two RCTs [5, 20], the indication for paracetamol was the same as in our study (treatment failure or contraindication for ibuprofen). The interpretation of different studies on this subject seems to be hindered by the lack of an international guideline on the definition of hsPDA and cutoff values for a small, medium, or large PDA (cardiac ultrasound measurements).

mol was the same as in our study (treatment failure or contraindication for ibuprofen). The interpretation of different studies on this subject seems to be hindered by the lack of an international guideline on the definition of hsPDA and cutoff values for a small, medium, or large PDA (cardiac ultrasound measurements). It is remarkable that Oncel et al. reported a 100 % success rate of PDA closure with intravenous paracetamol [19], the same administration route as in our study. Although unlikely, the route of administration could in part explain the variety in success rates; as long as data on bioavailability of the drug in the different routes and enterohepatic recirculation are lacking, we cannot tell which route of administration is superior. Still, ibuprofen studies also tend to find better results with oral administration [27]. Intravenous therapy probably leads to high peak plasma levels, but on the other hand, also to a relatively fast decrease in levels. Oral therapy might result in lower but more steady plasma levels. It can be hypothesized that PDA closure relies more on continuous prostaglandin inhibition than on intermittent high paracetamol or NSAIDs levels.

ably leads to high peak plasma levels, but on the other hand, also to a relatively fast decrease in levels. Oral therapy might result in lower but more steady plasma levels. It can be hypothesized that PDA closure relies more on continuous prostaglandin inhibition than on intermittent high paracetamol or NSAIDs levels. The relatively late start of paracetamol administration might be the most important explanation for our disappointing results compared to other studies. A second likely reason is the low gestational ages in our study. Other studies with better results [12, 19, 21, 29] included older infants with gestational ages >28 weeks. PDAs in this age group are generally less hemodynamically significant and tend to close spontaneously and respond better to pharmacological treatment. Ibuprofen therapy failure was previously found to be 17 % in infants with gestational ages of 26–27 weeks versus 62 % in 23–25-week-old infants [6]. PDAs in more preterm born infants are probably less responsive to cyclooxygenase inhibitors due to higher expression of prostaglandin receptors [3]. Next to gestational age and postnatal age, clinical factors such as the amount of fluids given, incidence of infections or sepsis, type of respiratory support, and use of co-medication might be influential factors for PDA closure success in extreme preterm infants.

e to higher expression of prostaglandin receptors [3]. Next to gestational age and postnatal age, clinical factors such as the amount of fluids given, incidence of infections or sepsis, type of respiratory support, and use of co-medication might be influential factors for PDA closure success in extreme preterm infants. Third, selection bias may have occurred, in that, we included patients in whom ibuprofen therapy had failed prior to paracetamol treatment. As NSAIDs are more potent prostaglandin synthesis inhibitors than paracetamol [8], resulting in lower peripheral PGE2 levels as is shown in orthodontic studies [25], the a priori probability of success of paracetamol in this group of patients was already relatively low.

herapy had failed prior to paracetamol treatment. As NSAIDs are more potent prostaglandin synthesis inhibitors than paracetamol [8], resulting in lower peripheral PGE2 levels as is shown in orthodontic studies [25], the a priori probability of success of paracetamol in this group of patients was already relatively low. Pharmacokinetics and pharmacodynamics of paracetamol for PDA closure have been hardly studied. Consequently, the effective plasma level of paracetamol to achieve PDA closure is unknown. The currently used dosages are already much higher than recommended for analgesia, and might be unsafe. Increasing the dose to improve closure rates is unadvisable. In a study by Kessel et al., plasma levels of paracetamol (15 mg/kg/6 h by nasogastric tube) did not exceed the recommended plasma levels of 10–20 mg/l for pain and fever control [14]. Based on predictive modeling, plasma levels will accumulate with the 15 mg/kg/6 h regimen, reaching peaks of nearly 25 mg/kg after four doses [1]. In very preterm infants and murine studies, the effectiveness of paracetamol on PDA closure was suggested to depend on dosing, duration of treatment (>2 days course), and mode of administration [8].

ing, plasma levels will accumulate with the 15 mg/kg/6 h regimen, reaching peaks of nearly 25 mg/kg after four doses [1]. In very preterm infants and murine studies, the effectiveness of paracetamol on PDA closure was suggested to depend on dosing, duration of treatment (>2 days course), and mode of administration [8]. Likewise, safety of paracetamol in very preterm infants (gestational age <28 weeks) has been little studied. Allegaert et al. showed no hemodynamic alterations during and following an intravenous loading dose of paracetamol and afebrile neonates maintained normothermia [2]. Paracetamol-induced hepatotoxicity is the most important concern; this is caused by the formation of a highly active metabolite N-acetyl-p-benzoquinone imine (NAPQ1) by the hepatic cytochrome P450-dependent CYP2E1 enzyme system [28]. The formation of NAPQ1 is suggested to be low due to the immaturity of the hepatic enzymes, although increased susceptibility to toxicity from supratherapeutic paracetamol is described in infants and younger children with fever [13]. CYP2E1 activity has not yet been quantified in neonates and the correlation between paracetamol concentration and increased NAPQ1 production is unknown.

y of the hepatic enzymes, although increased susceptibility to toxicity from supratherapeutic paracetamol is described in infants and younger children with fever [13]. CYP2E1 activity has not yet been quantified in neonates and the correlation between paracetamol concentration and increased NAPQ1 production is unknown. Several limitations of this study should be addressed. First, this was an observational single center study with relatively few patients per group, without a matched control group to rule out spontaneous PDA closure. Firm statements are therefore difficult to make. Second, the echocardiosonographer and pediatric cardiologist were not blinded for the PDA treatment. Third, only patients were included in our study after ibuprofen failure or with a primary or secondary contraindication for ibuprofen; this selection bias might have contributed to our results.

refore difficult to make. Second, the echocardiosonographer and pediatric cardiologist were not blinded for the PDA treatment. Third, only patients were included in our study after ibuprofen failure or with a primary or secondary contraindication for ibuprofen; this selection bias might have contributed to our results. Conclusions In view of the findings from this study, we do not recommend the use of intravenous paracetamol for hsPDA closure in VLBW infants after ibuprofen failure. Still, as we did not rule out effectiveness of paracetamol as early PDA treatment, it might be recommended when started in the first week of life. On the other hand, as long as data on long term safety are lacking, high dosages of paracetamol should be used with caution. Better designed PK/PD studies are needed to shed a light on safety aspects and the optimal dose-concentration-effect relationship. PK modeling of available data on PDA treatment with paracetamol in different gestational age groups can lead to different dosing recommendations per age group; a trial with these dose recommendations of paracetamol for PDA is also an option for future research. Abbreviations hsPDAHemodynamically significant patent ductus arteriosus IQRInterquartile range LA/Ao ratioLeft atrial to aortic root ratio LPALeft pulmonary artery NECNecrotizing enterocolitis NSAIDsNon-steroidal anti-inflammatory drugs PDAPatent ductus arteriosus VLBWVery low birth weight We thank J. Hagoort (Erasmus MC—Sophia Children’s Hospital) for editorial assistance. Conflict of interest All authors have nothing to disclose of potential conflicts of interest.

LA/Ao ratioLeft atrial to aortic root ratio LPALeft pulmonary artery NECNecrotizing enterocolitis NSAIDsNon-steroidal anti-inflammatory drugs PDAPatent ductus arteriosus VLBWVery low birth weight We thank J. Hagoort (Erasmus MC—Sophia Children’s Hospital) for editorial assistance. Conflict of interest All authors have nothing to disclose of potential conflicts of interest. Compliance with ethical standards The research involved human participants. Paracetamol treatment is part of our routine treatment guidelines for PDA treatment, and this study was an observational study. With the approval of the Medical Ethical Committee of our hospital (registration number MEC-2013-138), no informed consent was asked. To perform ongoing safety surveillance and to perform interim analyses on the safety data, an independent Data Safety Monitoring Board was assigned. The committee consisted of a pediatric intensivist, a hospital pharmacist and a neonatologist (from another NICU in the Netherlands). They had no involvement in or contact with the patients. Authors’ contribution Dr. Roofthooft conceptualized and designed the study, designed the data collection instruments, coordinated and supervised data collection at the site, carried out the initial analyses, drafted the initial manuscript, and approved the final manuscript as submitted. Dr. van Beynum and Dr. de Klerk carried out the cardiac ultrasounds and analyzed them. Furthermore, they reviewed and revised the manuscript, and approved the final manuscript as submitted.

Authors’ contribution Dr. Roofthooft conceptualized and designed the study, designed the data collection instruments, coordinated and supervised data collection at the site, carried out the initial analyses, drafted the initial manuscript, and approved the final manuscript as submitted. Dr. van Beynum and Dr. de Klerk carried out the cardiac ultrasounds and analyzed them. Furthermore, they reviewed and revised the manuscript, and approved the final manuscript as submitted. Dr. Tibboel, Dr. van den Anker, and Dr Reiss critically reviewed and revised the manuscript, and approved the final manuscript as submitted. Dr. van Dijk and Dr Simons conceptualized and designed the study, designed the data collection instruments, carried out the initial analyses, critically reviewed and revised the manuscript, and approved the final manuscript as submitted.

Introduction Noroviruses (NoVs) cause great burden of acute gastroenteritis worldwide in children under 5 years of age with approximately 200,000 deaths annually [1]. The virus is genetically highly diverse with more than 30 genotypes belonging to six genogroups (GI–GVI) [2]. NoV infections in children are associated with GI and GII, with genotype GII.4 being predominant for a long period of time [2]. These infections are acquired early in life, and there is limited knowledge on protective immune responses and duration of protection in natural infection in children. A longitudinal study of NoV infection in young children has recently demonstrated that reinfections with distinct genotypes commonly happen [3,4], but the immunity was not directly measured. In a recent case report [5], NoV-specific mucosal antibodies did not protect a child from re-infection with heterologous NoV. In here, we report a child followed from birth to 2 years of age with four NoV infections and emergence of acquired NoV-specific serum IgG and blocking antibodies. NoV-specific serum antibodies, which block binding of NoV capsid-derived virus-like particles (VLPs) to the host cell attachment factors, histo-blood group antigens (HBGAs), are considered as correlates of protection to NoV infection [6].

ections and emergence of acquired NoV-specific serum IgG and blocking antibodies. NoV-specific serum antibodies, which block binding of NoV capsid-derived virus-like particles (VLPs) to the host cell attachment factors, histo-blood group antigens (HBGAs), are considered as correlates of protection to NoV infection [6]. Materials and methods A healthy newborn was recruited in 2001 into the prospective Type 1 Diabetes Prediction and Prevention (DIPP) Study starting at birth [7]. The study was approved by the ethics committee of the University of Oulu and Oulu University Hospital and a written informed consent was obtained from the parents. Serum samples were collected at 0 (cord blood), 3, 6, 12, 18, and 25 months of age, and stool samples were collected monthly from 4 to 19 months of age (Table 1).Table 1 Detection of norovirus in stool samples and GI.3- and GII.4-specific antibodies in serum samples of a child from 0 to 25 months of age Age (month) Stool collection Serum collection RT-PCR (stool samples) Serum IgG ELISA GI.3-specific GII.4-specific 0 (cord blood) No Yes + + 3 No Yes + + 4 Yes No GII.6 5 Yes No − 6 Yes Yes − ++ ++ 7 Yes No − 8 Yes No − 10 Yes No − 11 Yes No − 12 No Yes + + 13 Yes No GI.3 14 Yes No − 15 Yes No GII.4 16 Yes No − 17 Yes No − 18 No Yes ++ ++ 19 Yes No GII.2 25 No Yes + + − negative, + positive, ++ indicates fourfold increase in GI.3- and GII.4-specific end-point titer

+ + 4 Yes No GII.6 5 Yes No − 6 Yes Yes − ++ ++ 7 Yes No − 8 Yes No − 10 Yes No − 11 Yes No − 12 No Yes + + 13 Yes No GI.3 14 Yes No − 15 Yes No GII.4 16 Yes No − 17 Yes No − 18 No Yes ++ ++ 19 Yes No GII.2 25 No Yes + + − negative, + positive, ++ indicates fourfold increase in GI.3- and GII.4-specific end-point titer Reverse transcription-PCR (RT-PCR) and ORF1 polymerase (region A) sequencing were used for detecting NoV genotype from stool suspensions according to the well-established methods [8]. NoV GI.3 and GII.4 VLPs cloned from original patient sequences from 2002 (GI.3, GenBank reference strain accession no. AF414403) and 1999 (GII.4, AF080551) were produced in baculovirus-insect cell expression system as earlier described [9]. These VLPs were chosen as antigens to detect NoV GI- and GII-specific IgG antibody responses using an ELISA method [10] as these sequences originated near the time of serum collections in this study (2001–2003, respectively). Briefly, VLPs were coated in phosphate-buffered saline (0.5 μg/ml) on 96-well polystyrene plates (Costar, Corning, NY, USA). Twofold serial dilutions of serum samples, starting at 1:100, were incubated on blocked plates for 1 h at 37 °C. Bound antibodies of serially diluted sera were detected with goat anti-human IgG-HRP (Invitrogen, CA, USA) followed by o-phenylenediamine dihydrochloride (OPD) substrate (Sigma-Aldrich, MO, USA). Optical density (OD) was measured, and a mean OD ≥0.100 was considered positive. End-point titer was expressed as a reciprocal of final serum dilution having positive OD. Seroconversion was defined as at least fourfold increase in successive serum end-point titer. Serum antibodies able to block binding of GI.3 and GII.4 VLPs to HBGAs in human saliva and therefore potentially neutralize the virus were tested in blocking assay as previously described [10]. Briefly, 96-well plates were coated with type A saliva from a secretor-positive adult at 1:3000 dilution. Serially, twofold diluted sera (starting dilution 1:50) preincubated with GI.3 or GII.4 VLPs (0.1 μg/ml) for 1 h at 37 °C were added to the plates. VLPs without serum were used as maximum binding controls. Bound VLPs were detected with NoV genotype-specific mouse sera and anti-mouse IgG-HRP (Sigma-Aldrich) and OPD substrate. Blocking index (%) was calculated as 100 % − [(OD wells with VLP serum mix/OD wells without serum; maximum binding) × 100].

ed to the plates. VLPs without serum were used as maximum binding controls. Bound VLPs were detected with NoV genotype-specific mouse sera and anti-mouse IgG-HRP (Sigma-Aldrich) and OPD substrate. Blocking index (%) was calculated as 100 % − [(OD wells with VLP serum mix/OD wells without serum; maximum binding) × 100]. Blocking titer 50 (BT50) was determined as the reciprocal of the final serum dilution that blocked at least 50 % of VLPs binding to the HBGA.

ed to the plates. VLPs without serum were used as maximum binding controls. Bound VLPs were detected with NoV genotype-specific mouse sera and anti-mouse IgG-HRP (Sigma-Aldrich) and OPD substrate. Blocking index (%) was calculated as 100 % − [(OD wells with VLP serum mix/OD wells without serum; maximum binding) × 100]. Blocking titer 50 (BT50) was determined as the reciprocal of the final serum dilution that blocked at least 50 % of VLPs binding to the HBGA. Results Four different NoV infections, three with GII (GII.6, GII.4, and GII.2) and one with GI (GI.3) NoVs, were detected by RT-PCR during a period of 15 months (Table 1). The child acquired febrile primary infection with GII.6 already at 4 months indicating that maternal antibodies although present in cord blood (end-point titer 800 to GI.3 and 3200 to GII.4, respectively) did not protect the child from NoV infection (Table 1 and Fig. 1). However, the cord blood antibodies completely lacked blocking potential of both VLPs (Fig. 1). Following the first infection, IgG seroconversion to GI.3 and GII.4 was detected at 6 months (Table 1 and Fig. 1). Although the levels of antibodies pertained until 12 months of age, they did not confer protection from a subsequent infection with GI.3 NoV at 13 months of age, and even though GII.4-specific antibodies detected at 12 month at a serum dilution 1:50 blocked 98 % of GII.4 VLPs binding to the HBGA receptor/s (Fig. 1), they did not protect the child from acquiring GII.4 infection at 15 months of age (Table 1). Clinically, the child experienced a prolonged diarrhea for a period of 3–4 weeks at the age of 13–14 months. During that period, the child might have experienced the two sequential infections with GI.3 and GII.4 NoVs. After the infections with GI.3 and GII.4, serum antibody levels to both VLPs measured at 18 months increased 16 times (end-point titers 51,200 and 102,400, respectively) compared to the pre-infection titers at 12 months (end-point titers 3200 and 6400) (Fig. 1). The high antibody level to GII.4 genotype (end-point titer 102,400 with 97 % blocking) did not prevent further infection with heterologous GII.2 NoV genotype already in the following month at the age of 19 months (Table 1 and Fig. 1), when the child again had typical clinical symptoms of gastroenteritis. These results show complete lack of cross-protection between the GII NoVs. In 7 months’ time (from 18–25 months), the child had retained only a half of GI.3 (end-point titer 25,600) and 1/8 of GII.4-specific antibodies (end-point titer 12,800) in the serum (Fig. 1).Fig.

had typical clinical symptoms of gastroenteritis. These results show complete lack of cross-protection between the GII NoVs. In 7 months’ time (from 18–25 months), the child had retained only a half of GI.3 (end-point titer 25,600) and 1/8 of GII.4-specific antibodies (end-point titer 12,800) in the serum (Fig. 1).Fig. 1 Norovirus-specific serum IgG antibodies in a child from 0 to 25 months of age. Serum samples collected at 0, 3, 6, 12, 18, and 25 months were tested for GI.3- and GII.4-specific antibodies in ELISA and blocking assays. End-point titer is expressed as a reciprocal of final serum dilution giving an OD ≥ 0.100. Blocking index (%) is shown for serum dilutions 1:50. Blocking titer 50 (BT50), reciprocal of the final serum dilution with 50 % blocking activity. Approximate time points of norovirus infections are indicated by arrows. GE stands for gastroenteritis symptoms reported

rocal of final serum dilution giving an OD ≥ 0.100. Blocking index (%) is shown for serum dilutions 1:50. Blocking titer 50 (BT50), reciprocal of the final serum dilution with 50 % blocking activity. Approximate time points of norovirus infections are indicated by arrows. GE stands for gastroenteritis symptoms reported Discussion Our study shows that a child can acquire as many as four consecutive NoV infections before the age of 2 years. Upon primary infection with GII.6 at 4 months, the child presented fever, but as there were no signs of vomiting or diarrhea, the infection was probably mild or asymptomatic, as recently reported [3]. Maternal antibodies present prior to the infection might have protected from severe disease but not the infection. A seroconversion to GII.4 was detected after the primary infection already at 6 months, but increase in blocking antibodies to GII.4 was detected at 12 months, suggesting that heterologous blocking antibodies may take longer time to develop. Although the child generated remarkable homologous serum antibody titer (GII.4-specific end-point titer 102,400, respectively) with high blocking activity (BT50 200), the immunity was not cross-protective and the child remained prone to infections with new GII.2 NoV genotype. The results presented here are in contrast to the findings observed in rotavirus infections, where protective antibody titers were achieved after two consecutive symptomatic or asymptomatic rotavirus infections in children less than 2 years of age [11]. The results in the present study suggest that NoV protective immunity in young children is type-specific. However, further studies in large pediatric populations are needed to address the magnitude, specificity, and duration of immune responses required for protection to NoV infection in young children. As children up to 2 years of age are highly vulnerable to infection and severe medically attended gastroenteritis, NoV vaccination of this target population is needed [12].

ions are needed to address the magnitude, specificity, and duration of immune responses required for protection to NoV infection in young children. As children up to 2 years of age are highly vulnerable to infection and severe medically attended gastroenteritis, NoV vaccination of this target population is needed [12]. Abbreviations DIPPType 1 diabetes prediction and prevention ELISAEnzyme-linked immunosorbent assay GGenogroup HBGAHisto-blood group antigens HRPHorseradish peroxidase IgGImmunoglobulin G NoVNorovirus ODOptical density OPDo-Phenylenediamine dihydrochloride ORFOpen reading frame PBSPhosphate-buffered saline RT-PCRReverse transcription polymerase chain reaction VLPVirus-like particle We thank the personnel of the Vaccine Research Center and Virology Department at the University of Tampere for the technical assistance and the DIPP Study personnel at the Department of Pediatrics, University of Oulu, for the acquisition of the clinical samples. Compliance with ethical standards ᅟ Funding This study was financially partly supported by Sigrid Juselius Foundation, Oulu University Hospital Research Funds, and Juvenile Diabetes Research Foundation. There was no other external funding received for this study. Conflict of interest All authors declare that they have no competing interests. Ethical approval All procedures performed in the studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Conflict of interest All authors declare that they have no competing interests. Ethical approval All procedures performed in the studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent Informed consent was obtained from the parents or legal guardians of all individual participants included in the study. Authors’ contributions Dr. V. Blazevic is the head of the laboratory of Vaccine Research Center, who designed this study, interpreted the results, and was the main author of this study. Dr. M. Malm was responsible for the immunological laboratory analysis and processing the data. M. Salminen was responsible for genotyping the noroviruses from the stool samples. S. Oikarinen coordinated the processing of DIPP biobank samples for virus analyses and was responsible for sample maintenance. Professor of Virology H. Hyöty, one principal investigator of the DIPP Study, and Dr. T. Vesikari, MD, the director of the Vaccine Research Center, critically reviewed the manuscript. Professor R. Veijola, MD, is one principal investigator of the DIPP Study and was responsible for the recruitment of the study subject and provided the samples and the data on clinical symptoms. All authors participated in writing and critical review and revision of the manuscript and approved the final manuscript as submitted.

Introduction Human parechoviruses (HPeV) are RNA viruses belonging to the family of Picornaviridae. Formerly described as echovirus 22 and 23 in the Enterovirus genus, HPeV were reclassified into their own genus, Parechovirus, in the 1990s based on genetic differences and biological properties. Then, they were renamed as HPeV-1 and HPeV-2, respectively [13, 25]. Recently, additional types of HPeV have been reported, and a total of 16 different types have been recognized to date (HPeV-1 to 16). The most common genotype detected worldwide is HPeV-1 followed by HPeV-3. Other types such as HPeV-2 and HPeV-4 are less common. Infections with HPeV are prevalent in young children and have been associated with mild diseases of the respiratory and gastrointestinal tract but also with serious diseases such as meningitis, encephalitis and sepsis in young infants [3, 12, 13]. According to recent studies, HPeV-3 might be one of the main agents causing severe neonatal neurological infections in Europe [5, 14, 17], although its real incidence is unknown since HPeV detection is not routinely performed in the majority of clinical microbiology laboratories. Clinical data specific for neonates are also scarce. This study describes the results obtained about the prevalence of HPeV infections in children up to 1 month of age in Spain and the clinical and epidemiological characteristics of the infected patients over a 1-year period. Furthermore, HPeV and enterovirus (EV) infections were compared.

Infections with HPeV are prevalent in young children and have been associated with mild diseases of the respiratory and gastrointestinal tract but also with serious diseases such as meningitis, encephalitis and sepsis in young infants [3, 12, 13]. According to recent studies, HPeV-3 might be one of the main agents causing severe neonatal neurological infections in Europe [5, 14, 17], although its real incidence is unknown since HPeV detection is not routinely performed in the majority of clinical microbiology laboratories. Clinical data specific for neonates are also scarce. This study describes the results obtained about the prevalence of HPeV infections in children up to 1 month of age in Spain and the clinical and epidemiological characteristics of the infected patients over a 1-year period. Furthermore, HPeV and enterovirus (EV) infections were compared. Methods Patients and samples In 2013, a prospective study was conducted in collaboration with 10 Spanish hospitals. The project was supported with a grant from the Health System (AES; PI12-00904). The personal data of the patients were protected, and the study was approved by the Ethics Committee of the Health Institute Carlos III. Criteria for inclusion were infants <1 month of age with fever without source (FWS), clinical sepsis disease or meningitis/encephalitis admitted to the participant centres, whose parents agreed to sign the informed consent. During the hospital stay, a physician filled out a study questionnaire with the clinical characteristics of the patients. FWS was defined as axillary temperature greater than 37.9 °C, which after an initial examination and laboratory evaluation has no apparent cause. Aseptic meningitis was defined by a neonate with fever, irritability, poor feeding, vomiting, or bulging fontanelle with pleocytosis >30 cells/mm3 in cerebrospinal fluid (CSF), with negative CSF culture for bacteria. Encephalitis was defined as a clinical diagnosis by the attending neurologist. It may be accompanied or not by CSF pleocytosis. Electroencephalogram and MRI compatible support the diagnosis but were not considered essential to it, and a clinical criterion of the attending neonatologist was considered enough for inclusion. Meningoencephalitis shares data from the two latter ones. Clinical sepsis is defined as an infant with an unwell appearance, with alteration of the paediatric assessment triangle (appearance, respiratory and circulatory items). Exclusion criteria were absence of consent or insufficient clinical sample.

h for inclusion. Meningoencephalitis shares data from the two latter ones. Clinical sepsis is defined as an infant with an unwell appearance, with alteration of the paediatric assessment triangle (appearance, respiratory and circulatory items). Exclusion criteria were absence of consent or insufficient clinical sample. From January to December 2013, a total of 84 individual specimens were collected from 84 children up to the age of 1 month, who met the clinical inclusion criteria and were included in the study. Samples were 71 CSFs (85 %), 12 sera (14 %) and 1 throat swab (1 %). Mean age of the children was 13 + 10 days and 46 of them were males (55 %). Viral detection and typing Enterovirus (EV) and herpes simplex virus (HSV-1 and 2) and varicella zoster virus (VZV) screening was performed in the microbiology laboratory from the hospital where the patients were admitted or in the National Centre for Microbiology (CNM), Enterovirus Laboratory, using commercial molecular methods (Xpert EV, Cepheid, CA, USA) or “in-house” RT-PCRs as previously described [8, 21]. HPeV was tested at the CNM in those specimens negative for EV, HSV and VZV, using a real-time RT-PCR designed in the 5-NCR of the genome [7]. Molecular typing of detected EV and HPeV was carried out by 3′-VP1 or VP3/VP1 amplification, respectively, and sequencing [6, 12].

Viral detection and typing Enterovirus (EV) and herpes simplex virus (HSV-1 and 2) and varicella zoster virus (VZV) screening was performed in the microbiology laboratory from the hospital where the patients were admitted or in the National Centre for Microbiology (CNM), Enterovirus Laboratory, using commercial molecular methods (Xpert EV, Cepheid, CA, USA) or “in-house” RT-PCRs as previously described [8, 21]. HPeV was tested at the CNM in those specimens negative for EV, HSV and VZV, using a real-time RT-PCR designed in the 5-NCR of the genome [7]. Molecular typing of detected EV and HPeV was carried out by 3′-VP1 or VP3/VP1 amplification, respectively, and sequencing [6, 12]. Statistical analysis Clinical and laboratory characteristics of the HPeV-positive patients were compared with those of the children infected with EV. A descriptive analysis was performed, expressing qualitative variables as proportions and quantitative variables as mean and standard deviation or median and interquartile range when appropriate. The chi-square test was used to compare groups. A difference with p value <0.05 was considered to be significant.

d with EV. A descriptive analysis was performed, expressing qualitative variables as proportions and quantitative variables as mean and standard deviation or median and interquartile range when appropriate. The chi-square test was used to compare groups. A difference with p value <0.05 was considered to be significant. Results Baseline data and frequency of EV and HPeV infections Out of 84 infants included in the study, 32 were positive for EV (38 %). No HSV-1 and 2 and VZV were detected. The 52 negative specimens (47 CSFs, 4 sera and 1 throat swab) were tested for HPeV, with positive detection in 9 (17 %) of them. Overall, HPeV detection frequency was 11 %. During the clinical course, other diagnosis were made (urinary infection, bronchiolitis, gastroenteritis or non-infectious neurologic disease) for 13 of the infants, and those were finally excluded from the epidemiological and clinical data analysis; all were EV and HPeV negative (Fig. 1).Fig. 1 Flow chart of enrolled and tested patients Viral detections were not found in any patient with suspected early-onset sepsis (11 newborns). In addition, four patients had bacteria positive blood culture, one Staphylococcus epidermidis (contaminated culture), two Streptococcus agalactiae, and one had a coinfection of Pasteurella multocida in blood and echovirus 9 in CSF. It was the only detected virus-bacteria coinfection [1].

ly-onset sepsis (11 newborns). In addition, four patients had bacteria positive blood culture, one Staphylococcus epidermidis (contaminated culture), two Streptococcus agalactiae, and one had a coinfection of Pasteurella multocida in blood and echovirus 9 in CSF. It was the only detected virus-bacteria coinfection [1]. Epidemiological and clinical associations of HPeV infections HPeV-positive samples were from four males and five females aged between 6 and 31 days (mean age, 17.6 days). Regarding monthly distribution, most of the HPeV infections (67 %) were detected between April and July, with another peak in October (22 %) (Fig. 2). All viruses were successfully characterized. Eight strains were HPeV-3 and the other was HPeV-5. All but one HPeV-positive specimens were found at CSFs. One of the HPeV-3 was detected in a serum sample. Clinically, six cases were diagnosed as FWS, two as clinical sepsis and one as encephalitis (Table 1). Fever was present in all cases (100 %), 38.6 ± 0.4 °C and lasted a mean of 1.3 ± 1.1 days. Although all except one of the HPeV were detected in CSF, none of the positive cases had cells in the range for pleocytosis in children <1 month of age (up to 30 cells/mm3). Protein and glucose were also normal in CSF. Leucocytes in blood were 8500 ± 5696 cells/mm3. C-reactive protein and procalcitonine were normal in all patients. Four of the nine HPeV-infected patients (44 %) were admitted to the neonatal intensive care unit (NICU). One of them, diagnosed with encephalitis, had seizures. None of them needed mechanical ventilation, and all experienced total recovery except the infant with encephalitis who had persistent seizures. All patients received antibiotics until the bacterial aetiology could be ruled out.Fig. 2 Distribution by month of the enterovirus (EV)- and parechovirus (HPeV)-positive samples detected in this study

anical ventilation, and all experienced total recovery except the infant with encephalitis who had persistent seizures. All patients received antibiotics until the bacterial aetiology could be ruled out.Fig. 2 Distribution by month of the enterovirus (EV)- and parechovirus (HPeV)-positive samples detected in this study Table 1 Clinical characteristics of infants with human enterovirus (EV) and parechovirus (HPeV) infections Clinical feature EV HPeV p (n = 32) (n = 9) Male 17 (53 %) 4 (44 %) NS Age average (days) 17.2+ 7.9 17.6+ 9 NS Temperature >37.9 °C 31 (97 %) 9 (100 %) NS Highest temperature 38.5 ± 0.6 38.6 ± 0.4 NS Fever duration (days) 1.6 ± 1.1 1.3 ± 1.1 NS Rash 4 (12.5 %) 1 (11 %) NS Antibiotic treatment 23 (72 %) 9 (100 %) NS NICU admission 3 (9 %) 4 (44 %) 0.001* OR = 5.2(CI:1.9–14) Leucocytes (cells/mm3) 10147 + 4693 8500 + 5696 NS Serum CRP (mg/L) 10.8 + 27 2.6 + 3.9 NS Procalcitonine (ng/mL) 0.8 + 2.7 0.2 + 0.09 NS CSF cells/mm3 250 + 415a 5.7 + 7 0.015* OR = 1.6(CI:1.1–2.1) CSF proteins (mg/dL) 104 ± 117a 49 ± 17 0.030* CSF glucose (mg/dL) 53 + 10a 50 + 8 NS Diagnosis Fever without source 12 (37.5 %) 6 (67 %) NS Clinical sepsis 1 (3 %) 2 (22 %) NS Meningitis 19 (59.4 %) 0 0.001* Encephalitis 0 1 (11 %) NS Quantitative variables are expressed as mean and standard deviation aExcluded a patient coinfected with EV and Pasteurella multocida NS not significant, NICU neonatal intensive care unit, CSF cerebrospinal fluid, CRP C-reactive protein, OR odds ratio, CI confidence interval *p values <0.05

(n = 32) (n = 9) Male 17 (53 %) 4 (44 %) NS Age average (days) 17.2+ 7.9 17.6+ 9 NS Temperature >37.9 °C 31 (97 %) 9 (100 %) NS Highest temperature 38.5 ± 0.6 38.6 ± 0.4 NS Fever duration (days) 1.6 ± 1.1 1.3 ± 1.1 NS Rash 4 (12.5 %) 1 (11 %) NS Antibiotic treatment 23 (72 %) 9 (100 %) NS NICU admission 3 (9 %) 4 (44 %) 0.001* OR = 5.2(CI:1.9–14) Leucocytes (cells/mm3) 10147 + 4693 8500 + 5696 NS Serum CRP (mg/L) 10.8 + 27 2.6 + 3.9 NS Procalcitonine (ng/mL) 0.8 + 2.7 0.2 + 0.09 NS CSF cells/mm3 250 + 415a 5.7 + 7 0.015* OR = 1.6(CI:1.1–2.1) CSF proteins (mg/dL) 104 ± 117a 49 ± 17 0.030* CSF glucose (mg/dL) 53 + 10a 50 + 8 NS Diagnosis Fever without source 12 (37.5 %) 6 (67 %) NS Clinical sepsis 1 (3 %) 2 (22 %) NS Meningitis 19 (59.4 %) 0 0.001* Encephalitis 0 1 (11 %) NS Quantitative variables are expressed as mean and standard deviation aExcluded a patient coinfected with EV and Pasteurella multocida NS not significant, NICU neonatal intensive care unit, CSF cerebrospinal fluid, CRP C-reactive protein, OR odds ratio, CI confidence interval *p values <0.05 Comparison between EV and HPeV infections Similar to HPeV infections, 66 % of the total of EV-positive samples were detected between April and July (Fig. 2). Twenty-nine of the 32 EV (91 %) were successfully characterized. All serotypes found belonged to EV species B. Echovirus (E)-18 was the most frequently detected type (7/29, 24 %), followed by E-3 and E-5 (4/29, 14 % each) and coxsackievirus (CV)-B3 (3/29, 10 %). Other types also detected in minor proportion were E-11, E-30, E-6, E-9, E-7, E-17 and CV-B2. The diagnoses for EV-infected patients were meningitis in 19 cases (59.4 %), FWS in 12 cases (37.5 %), and clinical sepsis in 1 case (3 %). Clinical data of the infants are shown in Table 1.

oxsackievirus (CV)-B3 (3/29, 10 %). Other types also detected in minor proportion were E-11, E-30, E-6, E-9, E-7, E-17 and CV-B2. The diagnoses for EV-infected patients were meningitis in 19 cases (59.4 %), FWS in 12 cases (37.5 %), and clinical sepsis in 1 case (3 %). Clinical data of the infants are shown in Table 1. When clinical characteristics of EV-infected patients were compared with those of HPeV-positive children, we observed significant differences in the pleocytosis in CSF (250 ± 415 cells/mm3) p = 0.015. The odds ratio (OR) for pleocytosis and EV infection was OR = 1.5 (confidence interval CI:1.1–2.1). CSF protein were also higher in EV group (104 ± 117 vs. 49 ± 17 mg/dL, p = 0.03). Furthermore, meningitis diagnosis was significantly more frequent in EV-positive patients than in HPeV-infected children (19/32 vs. 0/9, p < 0.001). Out of the 32 EV-positive patients, 23 (72 %) received antibiotic treatment. Only three EV-positive patients (9 %) needed admission to the NICU and no sequelae were described in this group. However, HPeV infection was a risk factor for admission to the NICU (OR = 5.2 (CI: 1.9–14)) (3/32 vs. 4/9, p < 0.001).

p < 0.001). Out of the 32 EV-positive patients, 23 (72 %) received antibiotic treatment. Only three EV-positive patients (9 %) needed admission to the NICU and no sequelae were described in this group. However, HPeV infection was a risk factor for admission to the NICU (OR = 5.2 (CI: 1.9–14)) (3/32 vs. 4/9, p < 0.001). Discussion We present a prospective, multicentre study performed during 1 year in Spanish infants up to 1 month of age with suspected sepsis or meningo-encephalitis or fever without source. EV was detected in 38 % of the patients and HPeV in 11 %. With respect to EV infections, other authors found EV incidences in neonates between 9.5 and 24 % [2, 16, 18, 26]. The high detection frequency observed in our series might be due to the study being performed in clinically ill infants with a high suspicion of viral infection and not as active surveillance. Besides, not all EV-negative samples collected by the hospitals could be included in the final study, due to absence of available samples or signed consent forms.

ed in our series might be due to the study being performed in clinically ill infants with a high suspicion of viral infection and not as active surveillance. Besides, not all EV-negative samples collected by the hospitals could be included in the final study, due to absence of available samples or signed consent forms. On the other hand, few HPeV epidemiological studies focused on newborns (infants <30 days) have been published. Most of them were performed in children (<18 years), and the presence of the virus in CSF varies from 2.3 to 7 %, whereas in the general population the percentage decreases to 0.5–0.8 % [9–11, 23]. As far as we know, there is only one study from Italy in which 7 (11.6 %) cases of EV and 3 (5 %) cases of HPeV infections were detected out of 60 neonates with suspected sepsis or neurological infection [18]. Our higher HPeV prevalence in the same type of patients (11 %) might be due to the fact that the Italian study [18] was conducted at the NICU, reducing the study population and leaving out the analysis of children who do not require admission to the NICU (56 % in our series).

cted sepsis or neurological infection [18]. Our higher HPeV prevalence in the same type of patients (11 %) might be due to the fact that the Italian study [18] was conducted at the NICU, reducing the study population and leaving out the analysis of children who do not require admission to the NICU (56 % in our series). In most of the previous studies, HPeV-3 infections have been described as clearly associated with central nervous system infections and sepsis-like illness in neonates and young infants [4, 15, 22]. Our data supported these findings, as HPeV-3 was also the predominant genotype identified, representing 89 % of the total HPeV detected. Only one strain was identified as HPeV-5, which is an infrequent type reported in the literature [9–11, 18, 23]. In this study, this type was detected in a case with FWS, so we cannot conclude that HPeV type 5 has a proved association with neurological or systemic diseases in newborns.

89 % of the total HPeV detected. Only one strain was identified as HPeV-5, which is an infrequent type reported in the literature [9–11, 18, 23]. In this study, this type was detected in a case with FWS, so we cannot conclude that HPeV type 5 has a proved association with neurological or systemic diseases in newborns. Regarding seasonal distribution, in Europe, HPeV infections are more common in spring and summer, similar to the EV infections [9–11, 23]. In our series, there is also a predominance of HPeV detection in spring and summer, although we have only 1 year of enrollment period. Besides, HPeV has been described to present in biannual cycles in Europe [9, 11, 15]. Previous data in Spain (not published) are consistent with those findings as the detection frequency was significantly higher in 2011 and 2013 than in 2012, but further longer surveillance studies are necessary to confirm the seasonality and periodicity of the HPeV-3 infections. This circulation pattern might also explain the high HPeV incidence found in this study, in comparison with the previously mentioned Italian one [18].

ficantly higher in 2011 and 2013 than in 2012, but further longer surveillance studies are necessary to confirm the seasonality and periodicity of the HPeV-3 infections. This circulation pattern might also explain the high HPeV incidence found in this study, in comparison with the previously mentioned Italian one [18]. Recently, Sharp et al [23], in Kansas, found that up to 17 % of the CSF-positive samples of children (<18 years) tested between June and October 2009 were positive for HPeV, with HPeV-3 being the predominant type. Clinically, they described HPeV infections with irritability and longer fever, lower peripheral leucocytes, absence of pleocytosis, lower CSF protein and higher CSF glucose compared to EV infections. In our series, some of these clinical characteristics were also observed as HPeV-infected patients had lower blood leucocytes, CSF protein and CSF cells values than those with EV infection. Furthermore, several studies including the American one mentioned above reported that admission to intensive care units were more frequent in HPeV-positive cases than in those infected by EV, but the prognosis was generally good [9, 22, 23]. We observed the same in our series. The higher proportion of HPeV-infected children who required admission to NICU was probably determined by the low age in an infant with an unwell appearance rather than by the clinical severity, as there was only one patient with encephalitis who had sequelae. In the literature, however, some fatal cases were described [19, 20], encephalitis being the major complication that caused sequelae and deaths.

obably determined by the low age in an infant with an unwell appearance rather than by the clinical severity, as there was only one patient with encephalitis who had sequelae. In the literature, however, some fatal cases were described [19, 20], encephalitis being the major complication that caused sequelae and deaths. Another recent study reported that between 50 and 100 % of HPeV-positive cases had been associated with erythematous rash on the extremities, especially palms and soles, not always present in the first hours but during evolution [24]. This dermatologic manifestation would be very suspicious data of HPeV infection in neonates and very young infants with febrile syndrome. However, these data are not consistent in the literature; the rash might have gone unnoticed or showed a different prevalence depending on the area. In the present study, only 12.5 % of patients showed rash. In summary, HPeV-3 infections should be suspected in infants younger than 1 month, with fever but without leucocytosis and pleocytosis in CSF. In these cases, HPeV screening in CSF or blood should be incorporated in the routine viral diagnosis to prevent unnecessary antibiotic treatment and prolonged hospitalization. The low number of samples included in this study was a limitation that affects the incidence data of the EV and HPeV infections, but not the results obtained in relation to the clinical and epidemiological differences found between both infections. Further studies could provide more information about the burden of HPeV infections in newborns. Abbreviations CSFCerbroespinal fluid

In summary, HPeV-3 infections should be suspected in infants younger than 1 month, with fever but without leucocytosis and pleocytosis in CSF. In these cases, HPeV screening in CSF or blood should be incorporated in the routine viral diagnosis to prevent unnecessary antibiotic treatment and prolonged hospitalization. The low number of samples included in this study was a limitation that affects the incidence data of the EV and HPeV infections, but not the results obtained in relation to the clinical and epidemiological differences found between both infections. Further studies could provide more information about the burden of HPeV infections in newborns. Abbreviations CSFCerbroespinal fluid EVEnterovirus HPeVHuman parechovirus FWSFever without source NICUNeonatal intensive care unit This study was partially supported by a grant from the Spanish National Health Institute PI12-00904. Conflict of interest The authors declare that they have no conflict of interest and they do not have a financial relationship with the organization that sponsored the research.

FWSFever without source NICUNeonatal intensive care unit This study was partially supported by a grant from the Spanish National Health Institute PI12-00904. Conflict of interest The authors declare that they have no conflict of interest and they do not have a financial relationship with the organization that sponsored the research. Author’s contribution María Cabrerizo and Gloria Trallero performed the detection and genotyping of HPeV. María Cabrerizo and Cristina Calvo designed the study, analysed the clinical data and wrote the manuscript. María José Pena Gregoria Megías, María Carmen Muñoz Almagro, María Pilar Romero, Sara Sanbonmatsu-Gámez, Mercedes Pérez-Ruiz, Manuel Omeñaca, Juan García Costa, Antonio Moreno-Docón and Nuria Rabella performed the local virological study, drafted the initial manuscript and approved the final manuscript as submitted. Amaia Cilla Eva del Amo, Diana Roda, María José Santos-Muñoz, Ana Isabel Mensalvas and Cristina Calvo recruited the patients and performed the data collection, contributed to the discussion of the results, and approved the final manuscript as submitted.

Introduction Until now, the proportion of cancers in children and adolescents attributable to an underlying genetic syndrome or inherited susceptibility is unclear. In the early 1990s, the inherited fraction of childhood cancer was estimated at 1–10 % [29]. A recent report from the Pediatric Cancer Genome Project/St. Jude Children’s Research Hospital determined an incidence of 16.0 % in patients with solid tumors, 8.6 % with brain tumors, and 3.9 % with leukemias. The report initially focused on 23 well-known cancer predisposition genes and 8 genes that predispose to pediatric cancer with a high penetrance [47]. The most frequently affected genes included TP53, APC, and BRCA2. Additional analyses were expanded to 565 genes that are known to play a role in various steps and pathways of cellular transformation. Identified variants were classified as pathologic, likely pathologic, uncertain significance, likely benign, and benign. Taking the larger gene-set into account, the overall prevalence of an inherited mutation increased only slightly, with a pathologic or likely pathologic variant being detected in 8.6 % of all patients and 4.6 % of patients with leukemia. However, the spectrum of tumors sequenced was not numerically representative of the spectrum of childhood tumors, and the mutation frequencies may be skewed accordingly. In a hereditary cancer risk assessment study in survivors of childhood cancer, a genetic counselor considered 29 % of the survivors as eligible for further genetics evaluation [19].

sequenced was not numerically representative of the spectrum of childhood tumors, and the mutation frequencies may be skewed accordingly. In a hereditary cancer risk assessment study in survivors of childhood cancer, a genetic counselor considered 29 % of the survivors as eligible for further genetics evaluation [19]. However, in the era of high-throughput sequencing in which new cancer susceptibility syndromes (CSS) and mechanisms are increasingly discovered—did we so far maybe just see the tip of the iceberg? Current clinical approach to CSS Pediatric oncologists consider children diagnosed with cancer and their families for inherited cancer susceptibility according to well-established criteria [20]. These comprise patient-specific constellations including (i) rare tumors commonly associated with cancer predisposition (e.g., adrenocortical carcinoma), (ii) bilateral or multifocal tumors (e.g., Wilms’ tumor), (iii) cancer diagnosis at a younger than expected age (e.g., thyroid carcinoma), (iv) multiple synchronous or metachronous tumors, (v) additional conditions (e.g., axillary freckling) indicative of an underlying syndrome, and (vi) suspicious family features. These might include (a) familial clustering of the same or closely related cancers, (b) cancer diagnoses in two or more first-degree relatives, (c) tumor patterns associated with a specific cancer predisposition syndrome, (d) exceptional young age at diagnosis, (e) sibling with childhood cancer, and (f) consanguineous parents.

ght include (a) familial clustering of the same or closely related cancers, (b) cancer diagnoses in two or more first-degree relatives, (c) tumor patterns associated with a specific cancer predisposition syndrome, (d) exceptional young age at diagnosis, (e) sibling with childhood cancer, and (f) consanguineous parents. Li-Fraumeni syndrome (LFS) is one of the most striking familial cancer predisposition syndromes. It is clinically and genetically heterogeneous and characterized by autosomal dominant inheritance and early onset of tumors, multiple tumors within one individual, and multiple affected family members. LFS presents with a variety of tumor types with soft tissue sarcomas, osteosarcomas, breast cancer, brain tumors, leukemia, and adrenocortical carcinoma being the most common tumor types. Comprehensive surveillance protocols have been implemented and proven efficiency in terms of superior survival [46]. Table 1 lists common hereditary cancer susceptibility syndromes sorted by the underlying mechanism. The American College of Medical Genetics and Genomics and the National Society of Genetic Counselors just published the latest referral indications for cancer predisposition assessment [13]. However, due to de novo mutations, incomplete penetrance of inherited mutations, and variable phenotype/genotype correlations, the family history may not in all cases be helpful. For example, up to 25 % de novo events of TP53 mutations are reported in Li-Fraumeni syndrome [6]. In most other cases of CSS, however, the proportion of inherited susceptibility versus de novo mutations remains unknown.Table 1 Examples of common hereditary cancer predisposition syndromes

y not in all cases be helpful. For example, up to 25 % de novo events of TP53 mutations are reported in Li-Fraumeni syndrome [6]. In most other cases of CSS, however, the proportion of inherited susceptibility versus de novo mutations remains unknown.Table 1 Examples of common hereditary cancer predisposition syndromes Syndrome Gene(s) Inheritance Clinical characteristics Tumor types Cancer risk DNA damage repair defects/genetic instability Ataxia telangiectasia (AT) * ATM AR Progressive ataxia, central nervous system degeneration, growth deficiency, ocular and facial telangiectasias, immunodeficiency, infertility, premature aging Leukemia, lymphoma, carcinoma 10–38 % overall cancer risk 70-fold increased leukemia risk (T-ALL, T-PLL) 250-fold increased lymphoma risk (B cell) Bloom syndrome (BS) BLM AR Short stature, immunodeficiency, malar rash, microcephaly, high-pitched voice, hypogonadism Leukemia, lymphoma 50 % overall cancer risk 15 % leukemia risk 15 % lymphoma risk Constitutional mismatch repair-deficiency syndrome (CMMR-D) MLH1, MSH2, MSH6, PMS2 AR Multiple café au lait (CAL) spots, features reminiscent of NF1 Pediatric brain tumors, colorectal cancers, ALL, AML, lymphoma, early onset gastrointestinal or gynecological cancers Biallelic mutations at very high risk Fanconi anemia (FA) FANCA, C, D1, D2, E, F, G, I, J, L, M, RAD51C, SLX4/BTBD12 FANCB AR X-linked Bone marrow failure, growth failure, radial ray abnormalities, renal abnormalities, CAL spots, hypopigmentation, congenital heart disease, microphthalmia, ear anomalies/deafness, hypogonadism; up to 25 % phenotypically normal Leukemia (MDS, AML), squamous cell carcinoma, gynecological tumors, brain tumors, Wilms tumor, neuroblastoma 25 % cumulative risk of hematologic malignancy by age 45 7 % MDS 9 % (500-fold increased risk of) AML Li-Fraumeni syndrome (LFS) TP53 AD up to 25 % de novo mutations beyond classical LFS malignancies phenotypically normal Soft tissue sarcoma, osteosarcoma, breast cancer, adrenocortical carcinoma (ACC), leukemia, brain tumors (glioblastoma multiforme, high-grade astrocytoma/primitive neuroectodermal tumor, medulloblastoma, choroid plexus carcinoma) 90 % lifetime risk to develop cancer 1–3 % ALL (hypodiploid ALL) Nijmegen breakage syndrome (NBS) NBS1 AR Microcephaly, prominent midface, receding mandible, CAL, recurrent infections, bone marrow failure NHL, DLBCL, Burkitt lymphoma, T-LBL/-ALL, AML, Hodgkin lymphoma, medulloblastoma, rhabomyosarcoma 40 % cancer risk by the age of 20 years Bo

elop cancer 1–3 % ALL (hypodiploid ALL) Nijmegen breakage syndrome (NBS) NBS1 AR Microcephaly, prominent midface, receding mandible, CAL, recurrent infections, bone marrow failure NHL, DLBCL, Burkitt lymphoma, T-LBL/-ALL, AML, Hodgkin lymphoma, medulloblastoma, rhabomyosarcoma 40 % cancer risk by the age of 20 years Bo ne marrow failure (BMF) syndromes: ribosome biogenesis and/or telomere maintenance anomalies Congenital amegakaryocytic thrombocytopenia (CAMT) type I / II MPL AR Thrombocytopenia and megakaryocytopenia with no physical anomalies MDS/AML Unknown Diamond blackfan anemia (DBA) RPS19, RPS24, RPS17, RPL35A, RPL5, RPL11, RPS7, RPS26, RPS10, GATA1 AD Majority sporadic Normochromic macrocytic anemia, reticulocytopenia, and nearly absent erythroid progenitors in the bone marrow, growth retardation, craniofacial, upper limb, heart, and urinary system congenital malformations, persistence of hemoglobin F Adenocarcinoma of the colon, sarcoma, genital cancer, MDS/AML, ALL 5.4 %-fold increased cancer risk Dyskeratosis congenital (DC) DKC1, CTC1, TERC, TERT, TINF2, NOP10, NHP2, WRAP53 X-linked Triad of abnormal skin pigmentation, nail dystrophy, and leukoplakia of the oral mucosa MDS/AML 3–33 % leukemia risk Shwachman-Diamond syndrome (SDS) SBDS AR (considered) Exocrine pancreatic insufficiency, hematologic abnormalities (pancytopenia), skeletal abnormalities MDS/AML, ALL 5–24 % leukemia risk Severe congenital neutropenia (SCN) (Kostmann syndrome) * ELANE, HAX1 AD AR Congenital neutropenia, recurrent/persistent infections, omphalitis MDS/AML 8–25 % leukemia risk Thrombocytopenia and absent radii syndrome (TAR) RBM8A and/or microdeletion 1q21.1 Unclear Reduction in the number of platelets and absence of the radius MDS/AML Unknown Cell cycle/differentiation defects (RAS pathway dysfunction) CBL syndrome CBL AD Dysmorphic facial features, short neck, developmental delay, hyperextensible joints, and thorax abnormalities with widely spaced nipples JMML Unknown Neurofibromatosis type I (NF1) NF1, SPRED1 AD CAL, axillary/inguinal frecking, Lisch nodules, bony dysplasia, seizures, learning difficulties, sphenoid wing abnormalities CMML/JMML, AML, neurofibroma, optic pathway glioma, peripheral nerve sheath tumor, astrocytoma, paraganglioma/pheochromocytoma 200–500-fold increased JMML risk 11 % MDS 5-fold increased brain tumor risk almost 100 % neurofibroma risk Noonan/Noonan-like syndrome PTPN11, HRAS, KRAS, NRAS, RAF1, SOS1, BRAF, SHOC2, MEPK1 AD Short stature, short webbed neck, lym

hway glioma, peripheral nerve sheath tumor, astrocytoma, paraganglioma/pheochromocytoma 200–500-fold increased JMML risk 11 % MDS 5-fold increased brain tumor risk almost 100 % neurofibroma risk Noonan/Noonan-like syndrome PTPN11, HRAS, KRAS, NRAS, RAF1, SOS1, BRAF, SHOC2, MEPK1 AD Short stature, short webbed neck, lym phedema, hypertelorism, coarse facies, CAL, pulmonary valve stenosis, pectus excavatum, wide and low-set nipples, cardiomyopathy, bleeding disorders Self-resolving myeloproliferative disease (MPD/TMD) and JMML, CMML, ALL, neuroblastoma, rhabdomyosarcoma MPD/JMML in pts with PTPN11 Transcription factors/pure familial leukemia syndromes Familial CEBPA leukemia CEBPA AD None MDS/AML FAB M1/M2 highly penetrant Familial ETV6 / ALL syndrome ETV6 AD Thrombocytopenia MDS/AML, MPAL, ALL, multiple myeloma, colon cancer Unknown Familial platelet disorder with predisposition to myeloid malignancy (FPD/AML) RUNX1 (dominant) AD Mild to moderate thrombocytopenia, platelet function abnormalities MDS/AML 35 % AML risk Familial PAX5 syndrome PAX5 AD None ALL 30 % penetrance in PAX5 SNP allele carriers PAX5 c.547G > A MonoMac GATA2 AD Monocytopenia, NK cell lymphopenia, infections MDS/AML 50 % leukemia risk Immunodeficiencies Wiskott-Aldrich syndrome (WAS) WAS X-linked Eczema, thrombocytopenia, immunodeficiency Diffuse large B cell lymphomas, non-Hodgkin’s lymphoma of larynx, leukemia, cerebellar astrocytoma, Kaposi sarcoma, smooth muscle tumors 5–13 % lymphoid malignancies X-linked lymphoproliferative (XLP) syndrome type I / II SH2D1A XIAP, SAP X-linked Severe immune dysregulation often after viral infection, typically with Epstein-Barr virus (EBV), severe or fatal mononucleosis, acquired hypogammaglobulinema, (HLH), lymphomatoid granulomatosis Hemophagocytic lymphohistiocytosis (HLH), non-Hodgkin lymphoma Unknown Autoimmune lymphoproliferative syndrome (ALPS) type IA/B/II CD95 CD95L CASP10 IL12RB1 AD AR in ALPS1A Lymphadenopathy with hepatosplenomegaly and autoimmune cytopenias, hypergammaglobulinema Hodgkin (HL) and non-Hodgkin (NHL) lymphoma, carcinoma (thyroid, breast, skin, tongue, liver), multiple neoplastic lesions (thyroid/breast adenomas, gliomas) 14-fold NHL risk 51-fold HL risk IL2-inducible T cell kinase deficiency ITK AR Fever, lymphadenopathy, splenomegaly, EBV associated lymphoproliferation Hodgkin lymphoma, Unknown Unknown Familial mosaic monosomy 7 Unknown Unknown Early-childhood onset of bone marrow insufficiency / failure MDS, AML Very high, fatal outcome Conge

HL risk 51-fold HL risk IL2-inducible T cell kinase deficiency ITK AR Fever, lymphadenopathy, splenomegaly, EBV associated lymphoproliferation Hodgkin lymphoma, Unknown Unknown Familial mosaic monosomy 7 Unknown Unknown Early-childhood onset of bone marrow insufficiency / failure MDS, AML Very high, fatal outcome Conge nital syndromes/aneuploidy Beckwith-Wiedemann syndrome (BWS) p57, H19, LIT1, ICR1, CDKN1C, NSD1 complex (AD, genomic imprinting, pUPD) Overgrowth syndrome, marcoglossia, omphalocele, hemihypertrophy, neonatal hypoglycemia Wilms tumor, hepatoblastoma, adrenal carcinoma, rhabdomyosarcoma 8.6 % cancer risk, depending on subtypes highest risk in patients with hemihypertrophy and organomegaly Cowden syndrome type I-VI (CWS) PTEN, SDHB, SDHD, KLLN AD Hamartomatous polyps of the gastrointestinal tract, mucocutaneous lesions, cobblestone-like papules of the gingiva and buccal mucosa, multiple facial trichilemmomas Dysplastic gangliocytoma of the cerebellum (Lhermitte-Duclos), colon, breast and thyroid cancer Lifetime risk 25–30 % breast cancer 10 % thyroid cancer 5–10 % endometrial/uterine cancer Denys-Drash syndrome (DDS) WT1 (dominant) Usually sporadic Diffuse mesangial sclerosis leading to early endstage renal disease, disorder of sexual development in XY patients Wilms tumor, gonadoblastoma Almost 100 % Wilms tumor Down syndrome (DS) Trisomy 21 n.a.

ncer 10 % thyroid cancer 5–10 % endometrial/uterine cancer Denys-Drash syndrome (DDS) WT1 (dominant) Usually sporadic Diffuse mesangial sclerosis leading to early endstage renal disease, disorder of sexual development in XY patients Wilms tumor, gonadoblastoma Almost 100 % Wilms tumor Down syndrome (DS) Trisomy 21 n.a. Facial dysmorphism, mental retardation, hypotonia, congenital heart disease TMD, AML, ALL 10 % TMD, 1–2 % ALL/AML 10–20-fold increased leukemia risk 500-fold increased risk of AMKL Familial Adenomatous Polyposis (FAP) syndrome APC AD Intestinal polyposis, osteomas, fibromas, sebaceous cysts, dental abnormalities Colon, thyroid, stomach, and intestinal cancer, hepatoblastoma, desmoid tumors, medulloblastoma Almost 100 % colorectal cancer Familial neuroblastoma ALK, PHOX2B AD None Neuroblastoma Unknown Familial Pleuropulmonary blastoma tumor predisposition syndrome DICER1 AD Pulmonary cysts, multinodular goiter PPB, cystic nephroma, Sertoli-Leydig cell tumors, rhabdomyosarcoma, supratentorial primitive neuroectodermal tumor, intraocular medulloepithelioma Variable penetrance, exact rest unknown Hereditary paragangliomas and pheochromocytoma syndrome (HPPS) SDHB AD None Paraganglioma, pheochromocytoma, renal, thyroid >70 % with metastatic disease 12 % GISTs Multiple endocrine naeoplasia type I (MEN1) MEN1 AD None Pancreatic islet cell tumor, pituitary adenoma, parathyroid adenoma 10 % carcinoid tumors Multiple endocrine neoplasia type II (MEN2A, MEN2B) RET AD Mucosal neuroma (intestinal tract, tongue, lips), marfanoid habitus Medullary thyroid carcinoma, pheochromocytoma, parathyroid hyperplasia 100 % risk of developing medullary thyroid carcinoma in MEN2A Nevoid basal cell carcinoma syndrome (NBCCS) / Gorlin syndrome PTCH1, 2, SUFU AD Macrocephaly, hypertelorism, palmar or plantar pits, rib abnormalities, ectopic calcification of the falx cerebri Basal cell carcinoma, desmoplastic medulloblastoma, ovarian fibromas 90 % basal-cell carcinoma, 5 % medulloblastoma Peutz Jeghers syndrome (PJS) STK11 AD Melanocytic macules of the lips, buccal mucosa, digits, multiple gastrointestinal hamartomatous polyps Intestinal, ovarian, pancreatic, breast cancers 55 % gastrointestinal cancer 45 % breast cancer Familial retinoblastoma syndrome (RB) RB1 AD Leukocoria Retinoblastoma, osteosarcoma, melanoma, glioma, carcinoma 80 % bilateral retinoblastoma 20 % unilateral retinoblastoma Rhabdoid tumor predisposition syndrome SMARCB1/INI1 Unclear, up to 21 % de novo mutations None Rhabdo

gastrointestinal cancer 45 % breast cancer Familial retinoblastoma syndrome (RB) RB1 AD Leukocoria Retinoblastoma, osteosarcoma, melanoma, glioma, carcinoma 80 % bilateral retinoblastoma 20 % unilateral retinoblastoma Rhabdoid tumor predisposition syndrome SMARCB1/INI1 Unclear, up to 21 % de novo mutations None Rhabdo id tumor, medulloblastoma, choroidplexus tumor, schwannoma Penetrance unclear Rubinstein-Taybi syndrome (RSTS) CREBBP AD Short stature, learning difficulties, distinctive facial features, broad thumbs and first toes, microcephaly, growth retardation Neuroblastoma, medulloblastoma, oligodendroglioma, meningeoma, pheochromocytoma, rhabdomyosarcoma, leiomyosarcoma, leukemia, lymphoma Unknown Tuberous sclerosis complex (TSC) TSC1/TSC2 AD Tubers, heterotopia, central nervous system migrational/psychomotor delay, seizures, renal/bone cysts Subependymal giant cell astrocytoma, hamartoma, renal angiomyolipoma, renal cell carcinoma, cardial rhabdomyoma, renal angiomyolipoma 4 % renal cell carcinoma 14 % giant cell astrocytoma Lynch syndrome type I / II MLH1, MSH2, MSH6, PSM2 AD CAL Colorectal cancer, glioblastoma multiforme, medulloblastoma Depending on subtype 50.4 % cumulative risk for colorectal cancer at the age of 70 WAGR syndrome WT1 AD Aniridia, genitourinary abnormalities, mental retardation Wilms tumor, gonadoblastoma High percentage of bilateral Wilms tumors *And cell cycle regulation

L Colorectal cancer, glioblastoma multiforme, medulloblastoma Depending on subtype 50.4 % cumulative risk for colorectal cancer at the age of 70 WAGR syndrome WT1 AD Aniridia, genitourinary abnormalities, mental retardation Wilms tumor, gonadoblastoma High percentage of bilateral Wilms tumors *And cell cycle regulation Personalized medicine With the ongoing efforts to personalize cancer medicine, comprehensive genome analyses will increasingly find its way into daily clinical routine in pediatric oncology. In the recently established German INdividualized therapy FOr Relapsed Malignancies in childhood (INFORM) project, this idea has been introduced for pediatric patients with relapsed or refractory high-risk disease without further standard of care therapy options. Individual tumor samples are characterized on the molecular level by next-generation sequencing (NGS) to establish a “fingerprint” of the tumor to identify promising targets for a successful relapse therapy [10].

ic patients with relapsed or refractory high-risk disease without further standard of care therapy options. Individual tumor samples are characterized on the molecular level by next-generation sequencing (NGS) to establish a “fingerprint” of the tumor to identify promising targets for a successful relapse therapy [10]. Other such examples in which the detection of specific mutations has already led to a change of therapy of course also exist. Recently, a new leukemia subtype of high-risk B-precursor acute lymphoblastic leukemia (ALL), called Ph-like ALL, was characterized. Besides its Ph- or BCR-ABL-like transcriptional profile, no translocation t(9;22) or BCR/ABL rearrangement, respectively, is present. Instead, multiple other genetic alterations can be detected, which are potentially druggable by tyrosine kinase inhibitors or other targeted therapies [18, 24, 36, 37]. In pediatric low-grade astrocytoma, the BRAF V600E-mutation was identified as a frequent genomic aberration activating the MAPK pathway. Tumors carrying this mutation show significantly increased BRAF and CCND1 levels [33]. Since its discovery, the BRAF V600E-mutation has been described in an increasing number of pediatric central nervous system (CNS) tumors [8, 11, 40, 41]. Targeted therapies such as the BRAF inhibitor vemurafenib and MEK1/2 inhibitors are available and some encouraging examples of effective therapies even in very aggressive tumor types have already been reported, such as the successful treatment of a 12-year-old child with relapsed glioblastoma multiforme with vemurafenib [38]. With the identification of a highly recurrent genetic alteration and its resulting fusion protein in ependymoma, the C11orf95-RELA protein, a further potentially druggable target was identified and specific therapy will hopefully be available in the near future [31]. We might also hypothesize that children with hereditary cancer syndromes like the so-called rasopathies might soon benefit from targeted therapy, as the underlying genetic alterations are highly recurrent [1, 9].

y druggable target was identified and specific therapy will hopefully be available in the near future [31]. We might also hypothesize that children with hereditary cancer syndromes like the so-called rasopathies might soon benefit from targeted therapy, as the underlying genetic alterations are highly recurrent [1, 9]. Next-generation sequencing Due to rapid technical advances in the field of NGS, tumor (including leukemia) genomes can nowadays comprehensively be analyzed within few days. Today’s state of the art in high-throughput sequencing already allows the usage of whole genome sequencing for research projects and of whole exome sequencing for daily clinical routine. However, the likelihood of identifying contemplable mutations is highly dependent on the relative ability of the sequencing approach to find these mutations. Computational processing, analyzing, and interpreting the massive amounts of data and genetic variants produced by NGS still remains challenging and requires comparisons with databases such as dbSNP and 1000 genomes project [3, 16]. Another valuable resource in interpreting own experimental data is the ExAC browser provided by the Broad Institute at www.exac.broadinstitute.org. It meanwhile provides exome data from more than 60,000 unrelated individuals. Before definitive conclusions can be drawn, the functional consequences of identified mutations on protein structure and function often have to be demonstrated experimentally [43]. In addition, a frequent conceptual misunderstanding relates to the fact that even a mutation with profound impact on protein function does not automatically proves its pathogenicity and disease-causing effect.

s of identified mutations on protein structure and function often have to be demonstrated experimentally [43]. In addition, a frequent conceptual misunderstanding relates to the fact that even a mutation with profound impact on protein function does not automatically proves its pathogenicity and disease-causing effect. Each of us carries an average of approximately 3000 single nucleotide polymorphisms (SNPs) in terms of individual SNPs. To generate a personal cancer genome signature for molecular targeted therapy, it is important to discriminate between these individual SNPs and somatic (tumor) mutations. Thus, comparing the NGS data of tumor versus germline DNA is a condition sine qua non to identify the somatically acquired genetic variants of the tumor. However, NGS not only generates focused genetic results with precise clinical implications for treatment but also so-called incidental findings with possible, limited, or unknown clinical impact or might even uncover an underlying susceptibility to cancer and other hereditary diseases. Such incidental findings are divided into “anticipatable” and “unanticipatable” ones. The former is a finding that is known to be associated with the test and is possible to be found. The latter could not have been anticipated given the current state of scientific knowledge [17]. Hence, treating physicians will increasingly be faced with such incidental genetic findings and the difficulties of interpreting and reporting these results.

known to be associated with the test and is possible to be found. The latter could not have been anticipated given the current state of scientific knowledge [17]. Hence, treating physicians will increasingly be faced with such incidental genetic findings and the difficulties of interpreting and reporting these results. Moreover, the pediatric oncologist is confronted with one new situation in particular: the fact that genetic events within the tumor genome itself can point toward underlying germline cancer susceptibility. Thus, even if not initially aimed to detect a CSS, the somatic “molecular make-up” of the tumor genome may suggest a germline mutation in a CSS gene. Up to now, there are two well-known findings in tumor genetics which can be indicative that the tumor arose on the basis of the child’s germline alteration, (a) chromothripsis and (b) a high total number of mutational events which exceeds that of other samples of the same tumor type.The phenomenon of chromothripsis was first reported by Stephens in 2011 [44]. The term “chromothripsis” (“chromo” from chromosome; “thripsis” for shattering into pieces) describes the shattering of a chromosome or a chromosomal region into tens to hundreds of pieces and locally clustered reassembling of some of the genomic fragments while others are lost to the cell.

ed by Stephens in 2011 [44]. The term “chromothripsis” (“chromo” from chromosome; “thripsis” for shattering into pieces) describes the shattering of a chromosome or a chromosomal region into tens to hundreds of pieces and locally clustered reassembling of some of the genomic fragments while others are lost to the cell. According to Stephens [44], chromothripsis is defined by six features: (1) rearrangements localized within the genome, (2) oscillating changes of the copy number profile between one and two copies, whereby (3) loss of heterozygosity (LOH) causes a copy number of one, and retaining heterozygosity a copy number of two, (4) clustering of breakpoints across the chromosome, (5) conjunction of two remote chromosome fragments, and (6) joining rearrangements between two chromosome arms with clustering at the breakpoints. Rapid oscillations between copy number states one and two within the whole or parts of the chromosome characterizes the copy number profile in case of chromothripsis.

osome, (5) conjunction of two remote chromosome fragments, and (6) joining rearrangements between two chromosome arms with clustering at the breakpoints. Rapid oscillations between copy number states one and two within the whole or parts of the chromosome characterizes the copy number profile in case of chromothripsis. In contrast to common theories of cancer evolution through progressive accumulation of genomic alterations such as oncogene activation and tumor suppressor loss through environmental and lifestyle factors in adults, chromothripsis as a single catastrophic event might be involved in the development of a variety of cancers in childhood. It can cause the formation of new gene fusions, disruption of tumor suppressors, and amplification of oncogenes [35, 44]. In adults, 2–3 % of all cancers show evidence of chromothripsis; in bone cancers, this incidence is especially high with 25 % [44]. The impact of chromothripsis on cancer gene function and cancer development in childhood has already been demonstrated for many different tumor entities, e.g., ALL, AML, ependymoma, medulloblastoma, neuroblastoma, and retinoblastoma [4, 23, 26, 28, 30, 31, 35]. In addition, chromothripsis has been associated with poor prognosis in neuroblastoma [28]. A list of pediatric tumors, in which chromothripsis has been described, is given in Table 2. Conversely, alterations in TP53 have been shown for low-hypodiploid ALL but without chromothriptic pattern [15].Table 2 Examples of (pediatric) tumors associated with chromothripsis

ed with poor prognosis in neuroblastoma [28]. A list of pediatric tumors, in which chromothripsis has been described, is given in Table 2. Conversely, alterations in TP53 have been shown for low-hypodiploid ALL but without chromothriptic pattern [15].Table 2 Examples of (pediatric) tumors associated with chromothripsis Tumor References Burkitt lymphoma * Sarova et al., Cancer Genet 2014 Brain tumors • Ependymoma • High-grade gliomas • Medulloblastoma -Sonic-Hedgehog -Group 3 • Parker et al., Nature 2014 • Zhao et al., Neuro Oncol 2014 -Rausch et al., Cell 2012 -Northcott et al., Nature 2012 Hodgkin lymphoma * Nagel et al., Genes Chromosomes Cancer 2013 Leukemia • AML • ALL (iAMP21) • Rausch et al., Cell 2012 • Harrison et al., Blood 2015; Li et al., Nature 2014 Neuroblastoma Ambros et al., Frontiers in Oncology 2014; Boeva et al., PLoS One 2013; Molenaar et al., Nature 2012 Osteosarcoma * Stephens et al., Cell 2011 Phaeochromocytoma (PCC) / Paraganglioma (PGL) * Flynn et al., J Pathol 2014 Retinoblastoma McEvoy et al., Oncotarget 2014 *Described in adult tumor samples

re 2014 Neuroblastoma Ambros et al., Frontiers in Oncology 2014; Boeva et al., PLoS One 2013; Molenaar et al., Nature 2012 Osteosarcoma * Stephens et al., Cell 2011 Phaeochromocytoma (PCC) / Paraganglioma (PGL) * Flynn et al., J Pathol 2014 Retinoblastoma McEvoy et al., Oncotarget 2014 *Described in adult tumor samples (b) To provide a comprehensive landscape of somatic genomic alterations (termed mutational signatures) in cancer genomes, numerous cancers have been profiled by DNA sequencing [2, 34, 45]. The occurring genomic alterations are presumably caused by defective DNA replication or repair and exogenous or endogenous mutagen exposure and include substitutions, insertions or deletions, rearrangements, copy number alterations, completely new sequences from exogenous sources, and combinations of all these possibilities. The prevalence of such mutations is highly variable between cancer (sub)types [2, 22]. Due to extensive exposure to carcinogens, small cell lung cancer (tobacco) and malignant melanoma (ultraviolet light) show the highest somatic mutation prevalence (over 100/Megabase (Mb)). In contrast, the mutation rate in pediatric cancers is lowest (0.1/Mb; approximately one change across the entire exome) as chronic mutagenic exposure plays a minor part in carcinogenesis in childhood [22]. An outline of mutation frequencies in various (pediatric) cancer types is given in Table 3.Table 3 Examples of mutation frequencies in (pediatric) tumors

ric cancers is lowest (0.1/Mb; approximately one change across the entire exome) as chronic mutagenic exposure plays a minor part in carcinogenesis in childhood [22]. An outline of mutation frequencies in various (pediatric) cancer types is given in Table 3.Table 3 Examples of mutation frequencies in (pediatric) tumors Malignancy Mutations (range) Reference AMLa 0.37 per Mb (0.01–10) of coding sequence Lawrence et al., Nature 2013 Ependymoma, intracranialb 12.8 ± 10.6 mutations (range 5 to 34) per tumor Bettegowda et al., Oncotarget 2013 Ependymoma, spinal cordb 12.9 ± 6.4 mutations (range 2 to 23) per tumor Bettegowda et al., Oncotarget 2013 Ewinga 0.15 per Mb of coding sequence Brohl et al., PLoS Genet 2014 Glioblastoma multiformeb 1.4 per Mb Cancer Genome Atlas Research Network, Nature 2008 Glioblastoma, non-brainstem pediatric 23.5 ± 11.2 mutations (range 4–46) per tumor Bettegowda et al., Oncotarget 2013 MDSb 3 (0–12) mutations per sample in 104 cancer genes Haferlach et al., Leukemia 2014 Medulloblastoma 8.3 non-synonymous SNVs per sample 0.35 non-silent mutations per megabase Parsons et al., Science 2011 Pugh et al., Nature 2012 Neuroblastoma 0.60 per Mb of coding regions Pugh et al., Nature Genet 2013 Rhabdoid cancers 0.19 per Mb (0–0.45) of coding regions Lee et al., J Clin Invest 2012 Xanthoastrocytoma, pleomorphicb 9.5 ± 8.5 mutations (range 1 to 28) per tumor Bettegowda et al., Oncotarget 2013 aTumor samples not specified bDescribed in adult tumor samples

Malignancy Mutations (range) Reference AMLa 0.37 per Mb (0.01–10) of coding sequence Lawrence et al., Nature 2013 Ependymoma, intracranialb 12.8 ± 10.6 mutations (range 5 to 34) per tumor Bettegowda et al., Oncotarget 2013 Ependymoma, spinal cordb 12.9 ± 6.4 mutations (range 2 to 23) per tumor Bettegowda et al., Oncotarget 2013 Ewinga 0.15 per Mb of coding sequence Brohl et al., PLoS Genet 2014 Glioblastoma multiformeb 1.4 per Mb Cancer Genome Atlas Research Network, Nature 2008 Glioblastoma, non-brainstem pediatric 23.5 ± 11.2 mutations (range 4–46) per tumor Bettegowda et al., Oncotarget 2013 MDSb 3 (0–12) mutations per sample in 104 cancer genes Haferlach et al., Leukemia 2014 Medulloblastoma 8.3 non-synonymous SNVs per sample 0.35 non-silent mutations per megabase Parsons et al., Science 2011 Pugh et al., Nature 2012 Neuroblastoma 0.60 per Mb of coding regions Pugh et al., Nature Genet 2013 Rhabdoid cancers 0.19 per Mb (0–0.45) of coding regions Lee et al., J Clin Invest 2012 Xanthoastrocytoma, pleomorphicb 9.5 ± 8.5 mutations (range 1 to 28) per tumor Bettegowda et al., Oncotarget 2013 aTumor samples not specified bDescribed in adult tumor samples Alexandrov et al. [2] described a mutational signature with very large numbers of substitutions and small indels, the latter at short nucleotide repeats and with overlapping microhomology at breakpoint junctions, termed “microsatellite instability,” which is characteristic of cancers with defective DNA mismatch repair and may suggest constitutional mismatch repair-deficiency syndrome (CMMR-D) in childhood.

utions and small indels, the latter at short nucleotide repeats and with overlapping microhomology at breakpoint junctions, termed “microsatellite instability,” which is characteristic of cancers with defective DNA mismatch repair and may suggest constitutional mismatch repair-deficiency syndrome (CMMR-D) in childhood. As was shown by Rausch et al. [35], the single nucleotide variant (SNV) rate of children with Sonic-Hedgehog medulloblastoma (SHH-MB) is clearly higher (24 tumor-specific SNVs) in the case of inherited TP53 mutations compared to sporadic pediatric medulloblastoma samples (average 5.7 non-synonymous SNVs per sample; [32]). Thus, comparing the patient’s SNV with the average SNV rate of a given tumor entity, an increased mutation frequency (SNV rate) detected by NGS of the tumor again may point to an underlying CSS (Li-Fraumeni syndrome). Ethical and legal issues “Are our other children at an increased risk of developing cancer?” Parents of a child diagnosed with cancer frequently raise this question. Up to now, pediatric oncologists mostly reassure them that cancer in children usually is not hereditary but an exceptionally bad stroke of fate. However, will this statement still hold true in the future with ever-increasing knowledge about underlying cancer predisposition syndromes and inherited cancer susceptibilities in childhood?

atric oncologists mostly reassure them that cancer in children usually is not hereditary but an exceptionally bad stroke of fate. However, will this statement still hold true in the future with ever-increasing knowledge about underlying cancer predisposition syndromes and inherited cancer susceptibilities in childhood? The incidental finding of chromothripsis and its association with Li-Fraumeni syndrome in SHH-MB patients very well demonstrates the far-reaching consequences of translational research and genetic testing in pediatric oncology with its challenges for scientists, treating physicians, and the affected child and his entire family. By detecting chromothripsis in a tumor, further genetic testing for germline p53 mutations is highly advisable as this phenomenon might be attributable to an underlying Li-Fraumeni syndrome. The latter obviously represents an important piece of clinical information as it will guide treatment, surveillance, and further early cancer screening programs [21, 46].

urther genetic testing for germline p53 mutations is highly advisable as this phenomenon might be attributable to an underlying Li-Fraumeni syndrome. The latter obviously represents an important piece of clinical information as it will guide treatment, surveillance, and further early cancer screening programs [21, 46]. According to the recommendations of national and international human genetic societies and the legislation of most European countries, prior to genetic testing, the child (wherever possible) and the parents must be informed in detail, preferences as to which findings should be reported must be assessed, and written informed consent must be obtained. This is a well-established standard of care for targeted molecular testing an affected individual or suspected carrier for a specific hereditary condition. However, NGS is likely, apart from the initial indication to perform it, to uncover incidental findings, such as an underlying CSS as well as non-cancer-related germline mutations (e.g., CFTR, Huntington’s disease) with varying clinical importance for the patient. In order to comply with the aforementioned recommendations, this would require extensive genetic counseling of the child/parents of a child diagnosed with cancer undergoing NGS of the tumor prior to testing, which would have to encompass both incidental findings with possible, limited, or unknown clinical impact and numerous results unrelated to the indication for NGS [42]. We believe that this is highly impractical in the daily life of a pediatric hemato-oncologist as disclosing the diagnosis of cancer itself is overwhelming and dramatically limits the child’s/parents’ receptivity, and NGS of the tumor often has to be initiated at the time of diagnosis. However, whenever NGS is initiated, the treating physician has an obligation to discuss the full range of generated data and the possibility of incidental findings and its disclosure with the child/parents. Furthermore, the ordering physician is responsible for obtaining informed consent and providing pre- and post-test counseling. Thus, regarding the child’s/parents’ autonomy and both their right to access all NGS data and their “right not to know,” they should be informed of the benefits, risks, and alternatives of genetic testing in detail [5, 7, 12].

an is responsible for obtaining informed consent and providing pre- and post-test counseling. Thus, regarding the child’s/parents’ autonomy and both their right to access all NGS data and their “right not to know,” they should be informed of the benefits, risks, and alternatives of genetic testing in detail [5, 7, 12]. When the patient/parent refuses to be informed about incidental findings, even if disclosure leads to beneficial interventions, the physician must ensure that adequate information has preceded this refusal. However, most clinicians do not have sufficient training in NGS and need to be extensively trained for clinical translation and reporting of NGS data. In contrast to the standards for genetic testing in adults, predictive testing in pediatric patients is only recommended when the disease is associated with childhood onset and only with available effective screening and/or intervention options [7, 39]. Refraining from predictive testing of children allows them to autonomously make this decision once they reach adulthood.

ults, predictive testing in pediatric patients is only recommended when the disease is associated with childhood onset and only with available effective screening and/or intervention options [7, 39]. Refraining from predictive testing of children allows them to autonomously make this decision once they reach adulthood. Last but not least, identifying children with hereditary cancer predispositions has immediate consequences for the entire family (siblings, parents, and extended family) [20, 25, 42]. Due to the young age of the index patient, potentially affected relatives might as well be young and yet asymptomatic. Having been tested themselves might—depending on the outcome—influence their family planning but will of course also provide an excellent opportunity to initiate early cancer surveillance programs which they will benefit from. However, genetic testing and tumor surveillance can have deeply affecting psychological consequences for the child and the family, emotional support should thus be in place for the families.

will of course also provide an excellent opportunity to initiate early cancer surveillance programs which they will benefit from. However, genetic testing and tumor surveillance can have deeply affecting psychological consequences for the child and the family, emotional support should thus be in place for the families. Clear legislation on returning genetic results in oncology are still missing. Lolkema et al. have thoroughly addressed the accompanying ethical, legal, and counseling challenges [25]. Comprehensive ethical recommendations on how to report research results to patients and parents are, for example, given by the American College of Medical Genetics and Genomics, the Boston Children’s Hospital, the American Academy of Pediatrics, the “EURAT” (Ethical and legal aspects of whole human genome sequencing) project of the Marsilius Kolleg of Heidelberg University, and the Leopoldina National Academy of Sciences Germany [5, 12, 14, 27, 39]. However, their practical implementation in day-to-day clinical life remains challenging. Conclusions Genetic testing and translational research in pediatric oncology provides new and exciting insights into the evolution and pathogenesis of childhood cancer. On the other hand, it can incidentally uncover an underlying cancer susceptibility syndrome with implications not only for the child but also for the entire family. Pediatric oncologists should therefore increase their awareness of chances and risks that accompany the increasingly wide clinical implementation of NGS platforms [42, 43]. Abbreviations ALLAcute lymphoblastic leukemia

Conclusions Genetic testing and translational research in pediatric oncology provides new and exciting insights into the evolution and pathogenesis of childhood cancer. On the other hand, it can incidentally uncover an underlying cancer susceptibility syndrome with implications not only for the child but also for the entire family. Pediatric oncologists should therefore increase their awareness of chances and risks that accompany the increasingly wide clinical implementation of NGS platforms [42, 43]. Abbreviations ALLAcute lymphoblastic leukemia CMMR-DConstitutional mismatch repair-deficiency CSSCancer susceptibility syndromes DGVDatabase of genomic variants INFORMIndividualized therapy for relapsed malignancies in childhood LFSLi-Fraumeni syndrome LOHLoss of heterozygosity MbMegabase NGSNext-generation sequencing SHH-MBSonic-Hedgehog medulloblastoma SNPsSingle nucleotide polymorphisms SNVsSingle nucleotide variants Revisions received: 18 04 April 2015 / 06 May 2015 The authors thank Dr. Jessica I. Höll for critical reading of the manuscript and editorial assistance. AB is supported by the German Research Consortium of Translational Cancer Research, DKTK. Conflict of interest The authors declare that they have no conflict of interest. Authors’ contributions MK screened the literature, collected the data, and drafted the manuscript. AB revised the manuscript. Both authors read and approved the final manuscript.

Introduction The main cause of cardiac arrest in pediatric patients is respiratory failure [11, 14, 23]. Ensuring an adequate airway and adequate oxygenation of the patient is a key element of CPR. The 2010 European Resuscitation Council (ERC) guidelines emphasize that interruptions in chest compression should be minimized during cardiopulmonary resuscitation (CPR) [1]. These guidelines also suggest that the intubator should be able to secure the airway without interrupting chest compression (CC). There are many reports in the literature regarding the effectiveness of intubation of children in hospital settings [2, 10], in which the intubator—usually the anesthetist—has the appropriate hardware facilities, including a full range of sedative and relaxative drugs. For Emergency Medical Services (EMS) teams, when the paramedic cannot count on the help of an experienced anesthetist, child intubation during CPR can cause many problems. According to Gerritse et al., child intubation effectiveness when performed by paramedics in prehospital settings was insufficient and ranges from 63.4 to 77 % [7, 8, 23]. This means that one in four children requiring airway protection and proper ventilation is not intubated, or the endotracheal tube is inserted incorrectly. Endotracheal intubation, which is currently the gold standard for airway management, is not currently performed only with Macintosh or Miller blades. In “Plan A” of his algorithm for airway management in children, Dr. Philip Ragg [13] indicates the possibility of using video laryngoscopes as a method of intubation.

ly. Endotracheal intubation, which is currently the gold standard for airway management, is not currently performed only with Macintosh or Miller blades. In “Plan A” of his algorithm for airway management in children, Dr. Philip Ragg [13] indicates the possibility of using video laryngoscopes as a method of intubation. The aim of the study was to compare time and success rates of different available laryngoscopes for pediatric emergency intubation during resuscitation with uninterrupted chest compression on a standardized pediatric manikin model. Methods This manuscript reports on our randomized controlled trial in accordance with the CONsolidated Standards of Reporting Trials (CONSORT) statement [16]. Approval was granted by the International Institute of Rescue Research (Warsaw, Poland) Institutional Review Board (Approval 11.2014.02.19, November 5th, 2014). Prior to the study, commencing it was registered at the ClinicalTrials register (www.clinicaltrials.gov, identifier NCT02289664). Study design We conducted a randomized crossover trial comparing the effects of the Macintosh, Intubrite®, Coopdech®, and CoPilot® laryngoscopes on intubation parameters including time-to-intubation and number of intubation attempts. With voluntary written, informed consent, 107 paramedics were recruited that satisfied the following inclusion criteria: (1) they had not performed more than 50 clinical child intubations by direct laryngoscopy, and (2) they had not received any training in endotracheal intubation using the Intubrite®, Coopdech®, and CoPilot® devices prior to the study.

nsent, 107 paramedics were recruited that satisfied the following inclusion criteria: (1) they had not performed more than 50 clinical child intubations by direct laryngoscopy, and (2) they had not received any training in endotracheal intubation using the Intubrite®, Coopdech®, and CoPilot® devices prior to the study. The devices used for the study were (Fig. 1):Fig. 1 Laryngoscopes used for this study were a standard Macintosh laryngoscope, b Intubrite® laryngoscope, c (3) Coopdech® video laryngoscope portable VLP-100, and d (4) CoPilot® video laryngoscope Standard Macintosh laryngoscope, blade 2 (Macintosh) (HEINE Optotechnik, Munich, Germany). Intubrite® with Macintosh #2 blade (Intubrite®) (Intubrite®, LLC; Vista, CA, USA). Coopdech® Video Laryngoscope Portable VLP-100 (Coopdech®) (Daiken Medical CO., LTD.; Osaka, Japan). CoPilot® Video Laryngoscope (CoPilot®) (Magaw Medical; Fort Worth, TX, USA). All intubations were performed using a Magill tracheal tube with 5.0-mm internal diameter (ID). Lubricant was already applied to the tracheal tube, and a 10-mL syringe to block the tube’s cuff, as well as an Ambu® resuscitator bag (Ambu, Copenhagen, Denmark), was readily available and within range of the participant.

CoPilot® Video Laryngoscope (CoPilot®) (Magaw Medical; Fort Worth, TX, USA). All intubations were performed using a Magill tracheal tube with 5.0-mm internal diameter (ID). Lubricant was already applied to the tracheal tube, and a 10-mL syringe to block the tube’s cuff, as well as an Ambu® resuscitator bag (Ambu, Copenhagen, Denmark), was readily available and within range of the participant. Prior to the trial, all participants were given a 30-min standardized training session and a 10-min practice session. The standardized training session consisted of a PowerPoint (Microsoft, Redmond, WA, USA) presentation with voice-over and a series of technique videos. The PowerPoint presentation reviewed airway anatomy, features of the laryngoscopes, and study protocol, while the technique videos demonstrated the use of all three laryngoscopes. The participants were given 10 min to practice with the three laryngoscopes on a manikin until they were comfortable with the devices; then, each participant had a maximum of three successful intubation attempts per laryngoscope. The manikin head used in the practice session and timed trials was a PediaSIM CPR training manikin (FCAE HealthCare, Sarasota, FL, USA), which is designed to be an accurate representation of a six-year-old child. Chest compression was performed using LUCAS-2 (Physio-Control, Redmond, WA, USA). Participants were allowed to ask questions at any time during the practice session and constructive feedback was provided by an instructor.

e, Sarasota, FL, USA), which is designed to be an accurate representation of a six-year-old child. Chest compression was performed using LUCAS-2 (Physio-Control, Redmond, WA, USA). Participants were allowed to ask questions at any time during the practice session and constructive feedback was provided by an instructor. A Research Randomizer program was used [www.randomizer.com] to divide the volunteers into four groups and to determine the order in which to apply the different endotracheal intubation (ETI) devices within each group. The first group attempted ETI using the Macintosh laryngoscope, the second using the Intubrite®, the third using the Coopdech®, and the fourth using the CoPilot® (Fig. 2). After completing the ETI procedure, participants had a 30-min break before performing intubation using another laryngoscope.Fig. 2 Flow chart of design and recruitment of participants according to CONSORT statement Measurements The primary endpoint of the study was the success rate of intubation. The secondary endpoint was defined as the time from insertion of the blade between the teeth to the first manual ventilation of the manikin’s lungs. If the examinee failed at all attempts, the case was excluded from the time calculations. After each attempt, participants were asked to rate the glottic view they had during the attempt using a Cormack–Lehane grade [4]. Quantitative data are presented as mean and standard deviation.

ventilation of the manikin’s lungs. If the examinee failed at all attempts, the case was excluded from the time calculations. After each attempt, participants were asked to rate the glottic view they had during the attempt using a Cormack–Lehane grade [4]. Quantitative data are presented as mean and standard deviation. Statistical analysis The R statistical package for Windows (version 3.0.0) was used for statistical analysis. Results were reported as mean and standard deviation (±SD) or absolute numbers and percentages. The Kolmogorov–Smirnov test was used to assess the normality of the distributions. As data were found not to be normally distributed, non-parametric tests were applied. We used a median test for continuous variables and an uncertainty coefficient test for categorical data. The cumulative success rate associated with time to complete tracheal intubation was analyzed using Kaplan–Meier analysis. p value <0.05 was considered to be statistically significant.

parametric tests were applied. We used a median test for continuous variables and an uncertainty coefficient test for categorical data. The cumulative success rate associated with time to complete tracheal intubation was analyzed using Kaplan–Meier analysis. p value <0.05 was considered to be statistically significant. Results Study collective One hundred and seven paramedics (42 female, 39.3 %) participated in this study. They each had 1 to 2 years of clinical experience and had performed about 50 tracheal intubations each. None had previously used Coopdech®, CoPilot®, or Intubrite® laryngoscopes. No complications such as dental compression were noted in the trial. No data were excluded from analysis. Seventy-three paramedics (27 female, 36.9 %) worked in teams of emergency medical services (EMS), 34 (15 female, 44.1 %) in hospital emergency units. Mean age was 31.2 ± 7.5 years. Average experience of clinical children intubation was 27 ± 5 intubations.

on were noted in the trial. No data were excluded from analysis. Seventy-three paramedics (27 female, 36.9 %) worked in teams of emergency medical services (EMS), 34 (15 female, 44.1 %) in hospital emergency units. Mean age was 31.2 ± 7.5 years. Average experience of clinical children intubation was 27 ± 5 intubations. Success rate The success rate after the first attempt using the Macintosh, Intubrite®, Coopdech®, and CoPilot® laryngoscopes varied and amounted to 58.9 vs. 69.1 % vs. 100 vs. 100 %. The overall effectiveness of intubation is presented in Table 1. There was a statistically significant difference in intubation success rate between Macintosh and Intubrite® (p = 0.019), as well as Coopdech® (p < 0.001) and CoPilot® (p < 0.001). There was also a statistically significant difference between Intubrite® and Coopdech® (p < 0.001) and CoPilot® (p < 0.001). Among the four analyzed laryngoscopes, intubation was most effective with the Coopdech® and Copilot® laryngoscopes.Table 1 Time to and success of intubation Type of laryngoscope blade Time to intubation (s) [mean (SD)] Tracheal intubation attempts First (%) Second (%) Third (%) Failed (%) Macintosh 29.4 ± 8.2 58.9 79.4 80.4 19.6 Intubrite 25.4 ± 10.5 69.1 85.0 89.7 10.3 Coopdech 21.6 ± 6.2 100 100 100 0.0 CoPilot 25.6 ± 7.4 100 100 100 0.0 Macintosh standard Macintosh laryngoscope, Intubrite Intubrite® laryngoscope, Coopdech Coopdech® video laryngoscope portable VLP-100, CoPilot CoPilot® video laryngoscope

Type of laryngoscope blade Time to intubation (s) [mean (SD)] Tracheal intubation attempts First (%) Second (%) Third (%) Failed (%) Macintosh 29.4 ± 8.2 58.9 79.4 80.4 19.6 Intubrite 25.4 ± 10.5 69.1 85.0 89.7 10.3 Coopdech 21.6 ± 6.2 100 100 100 0.0 CoPilot 25.6 ± 7.4 100 100 100 0.0 Macintosh standard Macintosh laryngoscope, Intubrite Intubrite® laryngoscope, Coopdech Coopdech® video laryngoscope portable VLP-100, CoPilot CoPilot® video laryngoscope Time to first ventilation The average times to successful intubation using Macintosh, Intubrite®, Coopdech®, and Copilot® are presented in Fig. 3. Analysis showed that the shortest average time of child intubation during uninterrupted chest compressions was achieved when using Coopdech® (21.6 ± 6.2 s), and the longest when using Macintosh (29.4 ± 8.2 s). A statistically significant difference was noticed between Coopdech® and Macintosh (p < 0.001) and Coopdech® and CoPilot® (p = 0.007) as well as between Coopdech® and Intubrite® (p = 0.006). A statistically significant difference was also observed between Intubrite® and Macintosh (p = 0.019) and between CoPilot® and Macintosh (p < 0.001).Fig. 3 Time required for tracheal intubation with four types of laryngoscopes. MacIntosh standard Macintosh laryngoscope, Intubrite Intubrite® laryngoscope, Coopdech Coopdech® video laryngoscope portable VLP-100, CoPilot CoPilot® video laryngoscope

nd Macintosh (p = 0.019) and between CoPilot® and Macintosh (p < 0.001).Fig. 3 Time required for tracheal intubation with four types of laryngoscopes. MacIntosh standard Macintosh laryngoscope, Intubrite Intubrite® laryngoscope, Coopdech Coopdech® video laryngoscope portable VLP-100, CoPilot CoPilot® video laryngoscope Quality of glottic view Glottic view quality was best with Coopdech® and CoPilot®, where 100 % reported a quality of glottic view corresponding to a Cormack–Lehane classification of I (Table 2).Table 2 Grade of glottic view according to the Cormack–Lehane grading that was achieved with the different ETI devices Device C/L I C/L II C/L III C/L IV Macintosh 78 (72.9 %) 26 (24.3 %) 3 (2.8 %) 0 (0.0 %) Intubrite® 84 (78.5 %) 23 (11.5 %) 0 (0.0 %) 0 (0.0 %) Coopdech® 107 (100 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) CoPilot® 107 (100 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) Data is given in absolute numbers and percentage

Quality of glottic view Glottic view quality was best with Coopdech® and CoPilot®, where 100 % reported a quality of glottic view corresponding to a Cormack–Lehane classification of I (Table 2).Table 2 Grade of glottic view according to the Cormack–Lehane grading that was achieved with the different ETI devices Device C/L I C/L II C/L III C/L IV Macintosh 78 (72.9 %) 26 (24.3 %) 3 (2.8 %) 0 (0.0 %) Intubrite® 84 (78.5 %) 23 (11.5 %) 0 (0.0 %) 0 (0.0 %) Coopdech® 107 (100 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) CoPilot® 107 (100 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) Data is given in absolute numbers and percentage Discussion The main cause of sudden cardiac arrest in children is respiratory failure and not heart disease as is the case with adults [1, 11, 14, 23]. The 2010 European Resuscitation Council (ERC) and American Heart Association (AHA) resuscitation guidelines emphasize the importance of minimizing interruptions to chest compression during cardiopulmonary resuscitation (CPR) [4, 17, 23]. These guidelines also suggest that the intubator should be able to secure the airway without interrupting chest compression. However, direct laryngoscopy and tracheal intubation during emergencies remain a challenge to medical practitioners who do not have clinical experience with the techniques or who are expected to perform intubation in difficult situations.

bator should be able to secure the airway without interrupting chest compression. However, direct laryngoscopy and tracheal intubation during emergencies remain a challenge to medical practitioners who do not have clinical experience with the techniques or who are expected to perform intubation in difficult situations. In the present study conducted under simulated resuscitation, intubation efficiency with Macintosh was 80.4 % and 89.7 % with Intubrite®. The higher efficiency of Intubrite® may be due to the different profiles of its handle and better lighting placement on the laryngoscope blade. The effectiveness of direct intubation on children using a laryngoscope with Macintosh or Miller blades performed by paramedics in out-of-hospital conditions is varied and ranges from 63.4 to 77 % in the study by Gerritse et al. [7, 8], 79.8 % in the study by Tollefsen et al. [19], and 82 % in the study by Vilke et al. [20]. Research indicates that low first attempt efficiency of intubation of children does not apply only to intubation performed by paramedics and activities under prehospital care. The study performed in 13 emergency departments by Choi et al. indicates that the first intubation attempt effectiveness with laryngoscope with Macintosh or Miller blades, for doctors with a specialization in emergency medicine, was 74.4 and 50 % for people with other specialties [3]. The poor efficiency of the first intubation attempts performed by paramedics and emergency physicians is also indicated in a study by Ehrlich et al. [5], in which the first intubation attempt effectiveness was 45 %, when performed by paramedics and 67 % [5] for doctors with emergency medicine specialization outside trauma centers. Eich et al. also showed in their study that intubation effectiveness increases with the age of the child. In prospective studies examining prehospital intubation of children by doctors with a specialization in the field of emergency medicine, the first intubation attempt effectiveness through the mouth was varied and was 53.9 % for infants, 68.2 % for children under 5 years of age, and 95.7 for children aged 6–14 years [6].

n prospective studies examining prehospital intubation of children by doctors with a specialization in the field of emergency medicine, the first intubation attempt effectiveness through the mouth was varied and was 53.9 % for infants, 68.2 % for children under 5 years of age, and 95.7 for children aged 6–14 years [6]. Alternative methods of airway management may be useful, especially for those who do not have regular exposure to situations in which airway management skills are required. In these situations, it is vital that an effective airway can be obtained quickly and without extensive prior training using the airway device. The authors of several studies indicate the higher efficiency of video laryngoscopes on direct laryngoscopy child intubation [9, 18, 21, 22]. The main findings of the study are that video laryngoscopes (Coopdech® and CoPilot®) may provide benefits regarding time to final placement compared to direct laryngoscopy. There is no significant difference in time to being operational in the case of those laryngoscopes. The first intubation attempt effectiveness using Coopdech® and CoPilot® in our study was 100 % for both devices. Average intubation time was shorter when using Coopdech® (21.6 ± 6.2 s) than using CoPilot® (25.6 ± 7.4 s.).

here is no significant difference in time to being operational in the case of those laryngoscopes. The first intubation attempt effectiveness using Coopdech® and CoPilot® in our study was 100 % for both devices. Average intubation time was shorter when using Coopdech® (21.6 ± 6.2 s) than using CoPilot® (25.6 ± 7.4 s.). This study is the first study showing intubation effectiveness on manikins resembling a six-year-old child using the CoPilot® and Coopdech® video laryngoscopes. Saito et al. evaluated the intubation effectiveness of Coopdech® VLP-100 under simulated difficult airways. He found vocal cord visibility and intubation efficiency to be better with the Coopdech® than with the Macintosh laryngoscope [15]. Video laryngoscopes are supposed to help during intubation. As demonstrated in the study, the effectiveness of pediatric intubation while using the Coopdech® and CoPilot® laryngoscopes was 100 %. These devices offered the most effective “method for pediatric intubation during resuscitation” out of all the tested ETI devices. Analysis showed that according to the researched paramedics, video laryngoscopes provided a better view of the larynx area.

n while using the Coopdech® and CoPilot® laryngoscopes was 100 %. These devices offered the most effective “method for pediatric intubation during resuscitation” out of all the tested ETI devices. Analysis showed that according to the researched paramedics, video laryngoscopes provided a better view of the larynx area. This study has a number of significant limitations. Firstly, it is a manikin study and does not involve real patients, and a recognized problem with manikin studies is that the times required to perform airway interventions are generally quicker than in patients. However, according to the International Liaison Committee on Resuscitation (ILCOR), randomized clinical trials for cases of cardiac arrest are unethical and cannot determine the expected benefits of CPR [12]. In both video laryngoscope and optical laryngoscope (e.g., AirTraq), reduce hampering their use may be the presence of fluids (including blood) in the oral cavity which reduces or completely prevents the visibility of the glottis. The strengths of this study include the use of a highly advanced patient simulator for performing pediatric advanced life support and the randomized crossover procedure. Conclusions We conclude that in child simulations managed by paramedics, the Coopdech® and Copilot® video laryngoscopes performed better than the standard Macintosh or Intubrite® laryngoscopes for endotracheal intubation during child chest compression. Further validation of the Intubrite®, Coopdech®, and Copilot® laryngoscopes in a clinical setting is required. Abbreviations AHAAmerican Heart Association

Conclusions We conclude that in child simulations managed by paramedics, the Coopdech® and Copilot® video laryngoscopes performed better than the standard Macintosh or Intubrite® laryngoscopes for endotracheal intubation during child chest compression. Further validation of the Intubrite®, Coopdech®, and Copilot® laryngoscopes in a clinical setting is required. Abbreviations AHAAmerican Heart Association CCChest compression CPRCardiopulmonary resuscitation EMSEmergency medical services ERCEuropean Resuscitation Council ETIEndotracheal intubation ILCORInternational Liaison Committee on Resuscitation We would like to thank all paramedic providers for their participation in our study. Funding None. Source of support No sources of financial and material support to be declared. Conflict of interests The authors declare that they have no competing interests. Author’s contributions LS, AK, ZT, and TG contributed significantly to the planning of the study and the study design. LS, AK, ZT, and LC recruited the participants and collected the data. LS and AK were the principal investigators of this study and did major manuscript preparation. LS and LC prepared the statistical analysis. LS, AK, LC, ZT, and TG contributed significantly for manuscript editing and expertise. Compliance with ethical standards ᅟ Ethical approval Approval was granted by the International Institute of Rescue Research (Warsaw, Poland) Institutional Review Board (Approval 11.2014.02.19, November 5th, 2014). Prior to the study, commencing it was registered at the ClinicalTrials register (www.clinicaltrials.gov, identifier NCT02289664).

Compliance with ethical standards ᅟ Ethical approval Approval was granted by the International Institute of Rescue Research (Warsaw, Poland) Institutional Review Board (Approval 11.2014.02.19, November 5th, 2014). Prior to the study, commencing it was registered at the ClinicalTrials register (www.clinicaltrials.gov, identifier NCT02289664). Informed consent Informed consent was obtained from all individual participants included in the study.

Introduction The rate of diagnosis and prevalence of hypertension (HTN) in children and adolescents appear to show a steady upward trend. Recent reports have shown that prevalence of primary hypertension in the under-18 population varies from 5 to 20 % worldwide [27, 38, 84] and from 5 to 12 % in Poland [58, 60, 76]. Persistent elevation of arterial blood pressure (BP) is an independent risk factor for myocardial infarction and heart failure, stroke and end-stage renal disease. Hypertension and specific morbid sequelae have emerged as leading causes of premature death among adults worldwide [1]. Although it is rare for the young to develop atherosclerotic cardiovascular disease (CVD), the cumulative long-term effects of high BP may have multiple acute and chronic complications. There is evidence that high BP in young age is associated with early markers of cardiovascular abnormalities such as left ventricular hypertrophy and atherosclerosis [3, 6, 28]. Comorbid conditions include obesity, high blood lipid levels, learning and attention problems and type 2 diabetes [2, 13]. However, the main complication of persistent high blood pressure in young age is its progress to adulthood, making it the greatest cardiovascular disease risk [4]. Several lines of evidence suggest that young people who enter adulthood with higher BP parameters are more likely than their normotensive peers to be affected with HTN and its morbidity sequelae [19, 36]. Thus, earlier stages of life seem to be critical to HTN [20].

king it the greatest cardiovascular disease risk [4]. Several lines of evidence suggest that young people who enter adulthood with higher BP parameters are more likely than their normotensive peers to be affected with HTN and its morbidity sequelae [19, 36]. Thus, earlier stages of life seem to be critical to HTN [20]. Hypertension is a common condition of multifactorial determination. It is suggested that HTN develops from a complex interplay of genetic, developmental, environmental and behavioural factors [38, 57, 61]. Heredity is a predisposing factor; almost half of young people with primary HTN have a positive family history of this condition, but environmental and individual contextual conditions may also play an important role in the development of HTN [61, 67]. A number of factors have been identified as predictors of elevated blood pressure in children and adolescents. They include maternal, birth and early-life characteristics, such as maternal weight status (BMI), smoking during pregnancy, low birth weight, breast feeding duration and childhood obesity [65, 70, 84]. Life cycle approach indicates that adolescence and especially puberty is a critical stage for adult BP. Longitudinal data showed that during puberty, BP may increase more than before it [81]. Rate of change in BP is likely to be synchronized with rapid somatic growth and adolescent growth spurt in height and weight [81, 86]. Moreover, changes in stature during puberty are closely linked to Tanner staging for sexual maturation. Gonadal hormones with possibly a preponderant effect of testosterone may affect BP levels and emergence of BP sexual dimorphism well manifest in adulthood [17, 29]. Additionally, adolescence is marked by increasing involvement in health risk behaviours often continuing into adult life [32]. Unfavourable effects of sedentary lifestyle, lack of vigorous or moderate physical activity, obesity, lack of a proper nutritionally balanced diet, high salt intake, low potassium and low calcium intake, tobacco use, alcohol intake and high stress may increase the risk for the development of adolescent hypertension [62, 63]. Furthermore, many of these factors are additive, such as unhealthy diet, insufficient physical activity and obesity, and vary in their propensity to contribute to the elevation of BP.

lcium intake, tobacco use, alcohol intake and high stress may increase the risk for the development of adolescent hypertension [62, 63]. Furthermore, many of these factors are additive, such as unhealthy diet, insufficient physical activity and obesity, and vary in their propensity to contribute to the elevation of BP. The literature on the potential confounders and mediators of childhood and adolescent hypertension has emphasized the role family level of socioeconomic status (SES) plays in the development of this condition [15, 55]. The association between parental SES and their offspring’s health outcomes has been well established [5, 16, 25]. These studies suggest that children of low-SES families are likely to have worse health outcomes. They are at a higher risk of CVD, elevated BP, metabolic syndrome, greater BMI and other negative health outcomes [43, 45]. They are also more likely to engage in risk-for-health behaviours than their better-off peers [34, 52]. Unlike heredity, ethnicity and geographic location, parental SES and adolescents’ behavioural factors are potentially modifiable. Elucidating the pathways by which these factors influence BP levels and health consequences (ischemic heart disease, stroke and others) may help in understanding the health gaps between different social groups and in developing a public health programme to counteract the health inequality [41].

potentially modifiable. Elucidating the pathways by which these factors influence BP levels and health consequences (ischemic heart disease, stroke and others) may help in understanding the health gaps between different social groups and in developing a public health programme to counteract the health inequality [41]. In Poland, there is no data available for parental SES as a potent risk factor in adolescent hypertension, although social differences in somatic growth, development and maturation have been recorded since the 1980s [8, 10, 9]. This study aimed to fill that knowledge gap by focusing on SES differences in blood pressure (BP) levels among Polish adolescents 10–18 years old. The specific aims for this study were (i) to calculate prevalence of elevated blood pressure in relation to selected indicators of parental SES and (ii) to establish relative importance of SES-related factors on the development of high blood pressure. The study hypothesis was that adolescents living in low-SES families might be at a higher relative risk of elevated BP than their better-off counterparts.

essure in relation to selected indicators of parental SES and (ii) to establish relative importance of SES-related factors on the development of high blood pressure. The study hypothesis was that adolescents living in low-SES families might be at a higher relative risk of elevated BP than their better-off counterparts. Materials and methods Study design and sampling A cross-sectional survey was carried out between February 2009 and September 2010 on a representative, randomly selected sample of adolescents, aged 10–18 years, participants in the ADOPOLNOR project, a transdisciplinary study on adolescent health and quality of life. It was an ethnically homogeneous group of students in grades 5 through 6 of primary school, 1 through 3 of junior secondary and 1 to 2 of senior secondary schools in the Wielkopolska province and its capital, the city of Poznań. Sample size was calculated using the formula for quantitative variable and a single cross-sectional survey [66]. The number of selected subjects was 5400.

Materials and methods Study design and sampling A cross-sectional survey was carried out between February 2009 and September 2010 on a representative, randomly selected sample of adolescents, aged 10–18 years, participants in the ADOPOLNOR project, a transdisciplinary study on adolescent health and quality of life. It was an ethnically homogeneous group of students in grades 5 through 6 of primary school, 1 through 3 of junior secondary and 1 to 2 of senior secondary schools in the Wielkopolska province and its capital, the city of Poznań. Sample size was calculated using the formula for quantitative variable and a single cross-sectional survey [66]. The number of selected subjects was 5400. Sampling procedure was a stratified two-stage cluster sample design. For the first sampling stage, schools were sampled from the sampling frame provided by the Ministry of Education for the Wielkopolska province via the Poznań Board of Education. Sampling was stratified by rural and urban areas as provided by Rogacki [79] and Central Statistical Office of Poland 2008 (www.stat.gov.pl). In this way, 52 schools were selected. The second sampling stage consisted of the selection of classes from the target grade of each participating school. In this procedure, if the number of classes was more than one, the class was randomly selected (as, for example, one class out of every six). In most villages, however, the students were assigned to only one class of each year level group.

of the selection of classes from the target grade of each participating school. In this procedure, if the number of classes was more than one, the class was randomly selected (as, for example, one class out of every six). In most villages, however, the students were assigned to only one class of each year level group. The study design and study protocol were approved by the Bioethics Commission of the Poznań University of Medical Sciences (Resolution no. 311/07) and the Poznań Board of Education (Resolution WAF-405/1/JM/07). The survey was carried out in compliance with principles outlined in the Helsinki Declaration and subsequent amendments [90]. Schools’ headmasters received an invitation letter and an information brochure about the research project. They approved the study protocol and gave permission to run the study in their schools. Furthermore, in collaboration with them, subjects’ parents were informed about the goals of the study and possibility of refusing the participation of their children in the study. Enrolled for the study were those students whose parents had given a written consent for them to participate. In addition, students who had attained the legal age for consent (16 years in Poland) gave assent for their participation in the study. Almost all parents (97.1 %) provided written informed consent for their children to participate in the ADOPOLNOR research project and 96.7 % of young people aged between 16 and 18 gave us their written consent to be participants of the study.

onsent (16 years in Poland) gave assent for their participation in the study. Almost all parents (97.1 %) provided written informed consent for their children to participate in the ADOPOLNOR research project and 96.7 % of young people aged between 16 and 18 gave us their written consent to be participants of the study. Complete data on parental characteristics (demographic, socioeconomic, behavioural) and adolescent characteristics at time of investigation (medical examination, anthropometry, arterial blood pressure and physical fitness) were obtained for 2451 male and 2490 female students, the total of 4941. All examinations were performed in school nursery rooms during morning hours (up to noon). The study protocol included medical examination, anthropometric measurements, and parental and self-reported background data questionnaires. Detailed description of the ADOPOLNOR study is available elsewhere [51]. General health status Health status of each subject was assessed by general practitioners (GPs) during general medical examinations, via self-report and proxy reports from parents.

All examinations were performed in school nursery rooms during morning hours (up to noon). The study protocol included medical examination, anthropometric measurements, and parental and self-reported background data questionnaires. Detailed description of the ADOPOLNOR study is available elsewhere [51]. General health status Health status of each subject was assessed by general practitioners (GPs) during general medical examinations, via self-report and proxy reports from parents. Anthropometric measurements Body height and weight were measured by well-trained researchers according to standard procedures [56]. The subject, wearing light gym exercise clothes and without shoes, was standing in an upright position with heels together, arms to the side, legs straight, shoulders relaxed and the head positioned in the Frankfurt plane. The height was measured with a portable Swiss-made Gneupel Precision Mechanics (GPM) anthropometer to the nearest 1 mm from the highest point on the midline vault (vertex) to the floor a subject was standing on. Body weight was measured to the nearest 0.1 kg on a calibrated electronic scale (Precision Health Scale). Then, BMI was calculated by taking a subject’s weight (kg) and dividing it by his/her height squared (m2). Following the IOTF recommendation, Cole’s cutoff values were used to determine the weight status [23, 24].

Body weight was measured to the nearest 0.1 kg on a calibrated electronic scale (Precision Health Scale). Then, BMI was calculated by taking a subject’s weight (kg) and dividing it by his/her height squared (m2). Following the IOTF recommendation, Cole’s cutoff values were used to determine the weight status [23, 24]. Chronological age was calculated in decimal values by subtracting the date of examination from the date of birth. The age groups were divided by years, defined in terms of the whole year; e.g., 10 years old group involved subjects between 10.00 and 10.99 years old. Blood pressure measurements Blood pressure was measured by school nurses strictly following the guidelines of the Fourth Protocol of the American Working Group of High Blood Pressure in Children and Adolescents [74]. A fully calibrated TECH MED TM-Z mercury gauge sphygmomanometer with sets of exchangeable cuffs and a clinical stethoscope was used for all BP measurements.

ed by school nurses strictly following the guidelines of the Fourth Protocol of the American Working Group of High Blood Pressure in Children and Adolescents [74]. A fully calibrated TECH MED TM-Z mercury gauge sphygmomanometer with sets of exchangeable cuffs and a clinical stethoscope was used for all BP measurements. Blood pressure, systolic and diastolic, was measured in duplicate on each of the three occasions separated by a 2-day interval. Measurements were taken on the right arm with the subjects sitting for at least a 5-min rest, and the average of the two measurements was the final result for the given day as it was suggested in the Seventh Report for adults [22]. The systolic and diastolic BP measurements corresponded to the reading on the sphygmomanometer at the first and fifth phases of the Korotkoff sounds, respectively. The scale on the sphygmomanometer was graduated in 2-mmHg divisions. The readings were made to the nearest millimeter Hg. Calculated intra-observer technical error (intra-TEM) equalled 1.3 mmHg and inter-observer technical error (inter-TEM) equalled 2.3 mmHg [59]. The BP classification was determined using the surveillance method. For each participant, the mean of measurements taken on three occasions was calculated. The values of mean systolic blood pressure (SBP) and diastolic blood pressure (DBP) were adjusted by sex, age and height percentile using current reference data for Polish children and adolescents [59]. Normal BP was defined as systolic and diastolic BP less than 90th percentile, prehypertension (high normal BP) was defined as an average systolic or diastolic BP of greater than or equal to 90th percentile but less than 95th percentile, and hypertension was defined as an average systolic or diastolic BP of greater than or equal to 95th percentile [39].

and diastolic BP less than 90th percentile, prehypertension (high normal BP) was defined as an average systolic or diastolic BP of greater than or equal to 90th percentile but less than 95th percentile, and hypertension was defined as an average systolic or diastolic BP of greater than or equal to 95th percentile [39]. Socioeconomic status Socioeconomic status was assessed through a self-reported family wealth using the Family Affluence Scale II (FAS II) and reports from parents using the ADOPOLNOR-R survey instrument. The SES indicators used in the study were the place of residence categorized according to the urbanization level (village with population of less than 1000 inhabitants, mainly engaged in farm work and this work is a source of income, small- to medium-sized city with population of less than 100,000, large-sized city with a population of 100,000 or more) [79], paternal and maternal educational attainment (the number of years of schooling completed and equalled to educational level: less than 12 years = primary/vocational level, 12 years = secondary level, more than 12 years = third level) and occupation status, family size (number of children in family), family finance-related burden referred to as income adequacy indicative of the objective financial situation, dwelling conditions and others (rated as an ordinal measure of more than enough, just enough or not enough money to cover expenses each month reported by study participants’ parents).

er of children in family), family finance-related burden referred to as income adequacy indicative of the objective financial situation, dwelling conditions and others (rated as an ordinal measure of more than enough, just enough or not enough money to cover expenses each month reported by study participants’ parents). The FAS II, a four-item measure of family wealth, provided by students, was reported by number of cars in family, asking if the respondent have one’s own bedroom, number of family’s vacation travels during the past 12 months and number of computers in the household. The FAS II total score could range from 0 to 9, with higher scores indicating higher level of family wealth. In the study, it was scored as a composite score and classified into three categories: low affluence (0–2), middle affluence (3–5) and high affluence (6–9).

st 12 months and number of computers in the household. The FAS II total score could range from 0 to 9, with higher scores indicating higher level of family wealth. In the study, it was scored as a composite score and classified into three categories: low affluence (0–2), middle affluence (3–5) and high affluence (6–9). Data analysis The outcome of interest was demographic and parental socioeconomic factors associated with BP status (0 = normotension, 1 = prehypertension and 2 = hypertension) in adolescent males and females after controlling for parental hypertension, sex, age and weight status. At first, multiple correspondence analysis (MCA) was used to determine whether the explanatory variables for the BP status were associated to each other and which of them might potentially operate in an additive way [85]. Crude associations of BP status and all potential covariate variables were evaluated individually using the chi-square Pearson test. Multiple multinomial logistic regression analyses (MLRA) were used to assess the association between BP status and the variables in question. The dependent outcome variable was a dichotomous variable of BP status. Two models were evaluated: model 1 involving normotensive vs. prehypertensive BP status and model 2 involving normotensive vs. hypertensive status after adjustment for all potentially confounding variables simultaneously. The odds ratio was used as a measure of association. A final explanatory model with a subset and relative odds ratio (OR) of the factors associated with BP status was obtained using a stepwise procedure with backward elimination and rejection criterion of the p value greater than 0.05.

ounding variables simultaneously. The odds ratio was used as a measure of association. A final explanatory model with a subset and relative odds ratio (OR) of the factors associated with BP status was obtained using a stepwise procedure with backward elimination and rejection criterion of the p value greater than 0.05. Statistical analyses were performed using the STATISTICA 10.0 data analysis software system (StatSoft Inc. Tulsa, OK, USA). All significance tests comprised two-way determinations. A value of p < 0.05 was considered statistically significant. Results The social background of the sample is shown in Table 1.Table 1 Characteristics of study participants on family history of hypertension, weight status and indicators of parental SES Variables N = 4941

Statistical analyses were performed using the STATISTICA 10.0 data analysis software system (StatSoft Inc. Tulsa, OK, USA). All significance tests comprised two-way determinations. A value of p < 0.05 was considered statistically significant. Results The social background of the sample is shown in Table 1.Table 1 Characteristics of study participants on family history of hypertension, weight status and indicators of parental SES Variables N = 4941 n (%) Family history of hypertensiona 1018 (20.6) Weight status—BMI (kg/m2) Underweight 534 (10.8) Normal weight 3548 (71.8) Overweight 697 (14.1) Obesity 162 (3.3) Place of residence Rural areas 1897 (38.3) Urban areas <100,000 inhabitants 2021 (40.9) Urban areas ≥100,000 inhabitants 1023 (20.8) Paternal education <12 years (primary, vocational) 2658 (53.8) 12 years (secondary) 1603 (32.5) >12 years (university degree or above) 680 (13.7) Maternal education <12 years (primary, vocational) 1947 (39.4) 12 years (secondary) 2013 (40.7) >12 years (university degree or above) 981 (19.9) Paternal occupation Economically inactiveb 180 (3.6) UB/PTJ/pension/others 513 (10.4) Employed—full-time job 2974 (60.2) Own business 727 (14.7) Farming 547 (11.1) Maternal occupation Economically inactive 763 (15.4) UB/PTJ/pension/others 521 (10.5) Employed—full-time job 2849 (57.7) Own business 327 (6.6) Farming 481 (9.8) Number of children in family 1 child 752 (15.2) 2 children 2312 (46.8) 3 and more children 1877 (38.0) Income adequacyc Not enough 574 (11.6) Enough 1832 (37.1) More than enough 2535 (51.3) Family affluence (FAS II)d

n (%) Family history of hypertensiona 1018 (20.6) Weight status—BMI (kg/m2) Underweight 534 (10.8) Normal weight 3548 (71.8) Overweight 697 (14.1) Obesity 162 (3.3) Place of residence Rural areas 1897 (38.3) Urban areas <100,000 inhabitants 2021 (40.9) Urban areas ≥100,000 inhabitants 1023 (20.8) Paternal education <12 years (primary, vocational) 2658 (53.8) 12 years (secondary) 1603 (32.5) >12 years (university degree or above) 680 (13.7) Maternal education <12 years (primary, vocational) 1947 (39.4) 12 years (secondary) 2013 (40.7) >12 years (university degree or above) 981 (19.9) Paternal occupation Economically inactiveb 180 (3.6) UB/PTJ/pension/others 513 (10.4) Employed—full-time job 2974 (60.2) Own business 727 (14.7) Farming 547 (11.1) Maternal occupation Economically inactive 763 (15.4) UB/PTJ/pension/others 521 (10.5) Employed—full-time job 2849 (57.7) Own business 327 (6.6) Farming 481 (9.8) Number of children in family 1 child 752 (15.2) 2 children 2312 (46.8) 3 and more children 1877 (38.0) Income adequacyc Not enough 574 (11.6) Enough 1832 (37.1) More than enough 2535 (51.3) Family affluence (FAS II)d Low 672 (13.6) Medium 2764 (55.9) High 1505 (30.5) UB/PTJ/pension/others unemployment benefits/part-time job/life annuity/all others aFirst-degree family history of hypertension: maternal and/or paternal hypertension bA category that includes people who voluntarily remain out of the active workforce, those raising a family at home and/or those who are unemployed

Low 672 (13.6) Medium 2764 (55.9) High 1505 (30.5) UB/PTJ/pension/others unemployment benefits/part-time job/life annuity/all others aFirst-degree family history of hypertension: maternal and/or paternal hypertension bA category that includes people who voluntarily remain out of the active workforce, those raising a family at home and/or those who are unemployed cAn ordinal measure of more than enough, just enough or not enough money to cover expenses each month reported by study participants’ parents dFamily affluence evaluated by adolescent participants in the study The majority of families were urban residents (61.7 %) of working parents (60.7 % for both parents combined), fathers having fewer than 12 years of schooling (53.8 %) and mothers with 12 years of schooling (40.7 %), without financial strain, i.e., with income adequacy (51.3 % with more than enough income). Family affluence level was self-rated by study participants and the majority of them (55.9 %) rated it as medium. Families with two children accounted for almost a half of the sample (46.8 %); three and more children (38.0 %) were next in order of frequency.

, with income adequacy (51.3 % with more than enough income). Family affluence level was self-rated by study participants and the majority of them (55.9 %) rated it as medium. Families with two children accounted for almost a half of the sample (46.8 %); three and more children (38.0 %) were next in order of frequency. Structural relationships among indicators of parental socioeconomic status—explanatory variables in the MCA (data not shown but available upon request from authors)—revealed that rural setting, low parental educational attainment, maternal economic inactivity, large families with three and more children, income inadequacy and low family wealth appeared to cluster closely together. Urban residence (<100,000 population) was associated with parental employment, income adequacy, medium family wealth and two children in family. One-child families were related to large city settings (≥100,000 population). The high affluent families markedly outlaid from other clusters indicating that the wealth was not associated with a specific setting or parental SES indicator. These associations conform to the pattern of additive nature of urbanization and parental SES factors. The rural-urban disparities in parental SES are shown in Fig. 1. Proportion of parents with low education level was significantly higher for rural areas as it was for income inadequacy and low family wealth (maternal education <12 years, 52.8 vs. 29.3 %; paternal education <12 years, 52.8 vs. 31.2 %; income inadequacy 14.8 vs. 10.7 %; low family wealth 21.7 vs. 9.9 % for rural and urban settings, respectively). There was also a significant rural vs. urban difference in the adolescent weight status. Prevalence of obese adolescents was higher in rural areas (4 vs. 3 %). Parental hypertension was equally distributed among inhabitants of rural and urban areas.Fig. 1 Rural-to-urban differences in selected indicators of parental socioeconomic status and family history of hypertension. For urban category small, medium and large cities combined. Abbreviations: FAS II Family Affluence Scale II, PE paternal education, ME maternal education, FHH family history of hypertension

reas.Fig. 1 Rural-to-urban differences in selected indicators of parental socioeconomic status and family history of hypertension. For urban category small, medium and large cities combined. Abbreviations: FAS II Family Affluence Scale II, PE paternal education, ME maternal education, FHH family history of hypertension Table 2 shows the prevalence of BP status (SBP and DBP separately) in the study sample according to potential covariate variables. Crude associations between BP status and covariates are expressed in terms of chi-square test.Table 2 Prevalence of prehypertension and hypertension among adolescent students by all factors involved in analysis

valence of BP status (SBP and DBP separately) in the study sample according to potential covariate variables. Crude associations between BP status and covariates are expressed in terms of chi-square test.Table 2 Prevalence of prehypertension and hypertension among adolescent students by all factors involved in analysis Variables Blood pressure level Normal Pre-HTN HTN p value SBP DBP SBP DBP SBP DBP SBP DBP Sex 0.009 0.612 Male 89.9 91.3 3.5 3.5 6.6 5.2 Female 90.8 90.7 4.4 5.3 4.8 4.0 Age (years) 0.042 0.006 10 90.5 92.9 3.9 3.4 5.6 3.7 18 88.7 88.2 3.4 5.6 7.9 6.2 Family history of hypertension <0.001 0.026 Yes 84.5 88.7 5.7 5.6 9.8 5.7 No 92.5 92.5 4.2 4.2 3.3 3.3 Place of residence <0.001 <0.001 Rural areas 85.3 86.0 5.7 6.8 9.0 7.2 Urban <100,000 inhabitants 90.2 91.2 4.3 3.9 5.5 4.9 Urban ≥100,000 inhabitants 92.2 94.3 3.4 3.0 4.4 2.7 Paternal education <0.001 <0.001 <12 years 88.5 89.1 4.5 5.1 7.0 5.8 12 years 92.1 93.4 3.4 3.9 4.5 2.7 >12 years 93.4 95.9 3.9 2.4 2.7 1.7 Maternal education 0.002 <0.001 <12 years 88.9 87.7 3.9 6.2 7.2 6.1 12 years 90.6 93.0 4.1 3.4 5.3 3.6 >12 years 92.9 95.2 3.6 2.3 3.5 2.5 Paternal occupation 0.005 0.003 Economically activea 91.3 92.7 6.0 4.3 2.7 3.0 UB/PTJ/pension/others 89.1 90.0 7.6 5.8 3.3 4.2 Economically inactive 86.1 87.4 8.2 7.3 5.7 5.3 Maternal occupation 0.04 <0.001 Economically activea 94.0 94.1 4.0 3.1 2.0 2.8 UB/PTJ/pension/others 89.1 87.9 6.6 5.1 4.3 7.0 Economically inactive 86.2 84.3 7.9 8.9 5.9 6.8 Number of children in family 0.05 0.004 1 child 91.9 93.8 3.4 2.8 4.7 3.4 2 children 90.1 92.2 3.9 3.9 6.0 3.9 3 and more 88.5 89.4 4.3 5.4 7.2 5.2 Income adequacy 0.047 0.041 More than enough 92.5 92.5 4.1 4.0 3.4 3.5 Enough 89.9 89.4 5.3 5.1 4.8 5.5 Not enough 88.8 90.5 5.9 4.4 5.3 5.1 Family affluence (FAS II) 0.532 0.151 High 89.4 91.9 6.7 4.4 3.9 3.7 Low 89.4 89.8 7.1 5.0 3.5 5.2 Medium 89.2 90.0 6.9 4.7 3.9 5.3 Weight status—BMI (kg/m2) <0.001 <0.001 Underweight 97.2 96.0 2.4 2.8 0.4 1.2 Normal weight 92.3 92.9 4.4 3.8 3.3 3.3 Overweight 83.0 82.9 9.1 7.7 7.9 9.4 Obesity 58.2 72.1 15.1 14.3 26.7 13.6 Values are in percentage

32 0.151 High 89.4 91.9 6.7 4.4 3.9 3.7 Low 89.4 89.8 7.1 5.0 3.5 5.2 Medium 89.2 90.0 6.9 4.7 3.9 5.3 Weight status—BMI (kg/m2) <0.001 <0.001 Underweight 97.2 96.0 2.4 2.8 0.4 1.2 Normal weight 92.3 92.9 4.4 3.8 3.3 3.3 Overweight 83.0 82.9 9.1 7.7 7.9 9.4 Obesity 58.2 72.1 15.1 14.3 26.7 13.6 Values are in percentage aEconomically active category includes employed/own business/farming

32 0.151 High 89.4 91.9 6.7 4.4 3.9 3.7 Low 89.4 89.8 7.1 5.0 3.5 5.2 Medium 89.2 90.0 6.9 4.7 3.9 5.3 Weight status—BMI (kg/m2) <0.001 <0.001 Underweight 97.2 96.0 2.4 2.8 0.4 1.2 Normal weight 92.3 92.9 4.4 3.8 3.3 3.3 Overweight 83.0 82.9 9.1 7.7 7.9 9.4 Obesity 58.2 72.1 15.1 14.3 26.7 13.6 Values are in percentage aEconomically active category includes employed/own business/farming In univariate analysis, the SBP levels were associated with all but the family affluence factor. Like systolic, diastolic BP was associated with all but sex and family affluence factors. There was a clear gradient in socioeconomic factors with a tendency of the disadvantage to locate in rural areas, parental low educational attainment, unemployment or farming and in income inadequacy. Higher prevalence of SBP as well as DBP HTN was found for participants with positive family history of hypertension (FHH) (SBP 9.8 vs. 3.3 % and DBP 5.7 vs. 3.3 % for yes and no, respectively) being at older age (7.9 vs. 5.6 % for 18 and 10 years for SBP and 6.2 vs. 3.7 % for DBP) and for male sex (SBP 6.6 vs. 4.8 %) with obese weight status (SBP 26.7 vs. 3.3 % and DBP 13.6 vs. 3.3 % for obese and normal weight status, respectively), living in rural than urban areas (SBP 9.0 vs. 4.4 % and DBP 7.2 vs. 2.7 % for rural and urban settings, respectively), having parents with low educational attainment (7.0 vs. 2.7 % for SBP and 5.8 vs. 1.7 % for DBP for fathers and 7.2 vs. 3.5 % for SBP and 6.1 vs. 2.5 % for DBP for mothers) and economically inactive (5.7 and 5.9 % for SBP and 5.3 and 6.8 % for DBP for fathers and mothers, respectively) with income inadequacy (5.3 % for SBP and 5.1 % for DBP) as compared to their better-off peers.

for SBP and 5.8 vs. 1.7 % for DBP for fathers and 7.2 vs. 3.5 % for SBP and 6.1 vs. 2.5 % for DBP for mothers) and economically inactive (5.7 and 5.9 % for SBP and 5.3 and 6.8 % for DBP for fathers and mothers, respectively) with income inadequacy (5.3 % for SBP and 5.1 % for DBP) as compared to their better-off peers. At multivariate level, only selected factors remained in their significance. The adjusted odds ratios for parental SES-related risk factors of pre-HTN and HTN after controlling for FHH, sex, age and weight status are presented in Table 3.Table 3 Multiple/multinomial logistic regression analysis of most parsimonious set of factors affecting the likelihood of developing prehypertension and hypertension in adolescent students Variable Stepwise MLRA with backward elimination Systolic blood pressure Diastolic blood pressure Prehypertension OR (95 % CI) Hypertension OR (95 % CI) Prehypertension OR (95 % CI) Hypertension OR (95 % CI) Sex Male (reference category) 1 1 Female 1.24 (1.01; 1.54) 0.77 (0.59; 0.91) p value for trend 0.039 0.045 Age (years) 10 years (reference category) 1 1 1 1 18 years 1.23 (1.06; 2.01) 1.39 (1.09; 2.13) 2.30 (2.01; 2.97) 1.72 (1.05; 2.81) p value for trend 0.046 0.038 0.005 0.029 Family history of hypertension No (reference category) 1 1 1 1 Yes 1.81 (1.31; 2.51) 1.72 (1.34; 2.20) 1.39 (1.01; 1.91) 1.43 (1.02; 1.99)

p value for trend 0.039 0.045 Age (years) 10 years (reference category) 1 1 1 1 18 years 1.23 (1.06; 2.01) 1.39 (1.09; 2.13) 2.30 (2.01; 2.97) 1.72 (1.05; 2.81) p value for trend 0.046 0.038 0.005 0.029 Family history of hypertension No (reference category) 1 1 1 1 Yes 1.81 (1.31; 2.51) 1.72 (1.34; 2.20) 1.39 (1.01; 1.91) 1.43 (1.02; 1.99) p value for trend 0.0003 <0.0001 0.039 0.036 Place of residence Rural areas (reference category) 1 1 1 1 Urban areas <100,000 inhabitants 0.82 (0.72; 0.94) 0.74 (0.66; 0.82) 0.64 (0.56; 0.73) 0.55 (0.48; 0.64) Urban areas ≥100,000 inhabitants 0.56 (0.37; 0.82) 0.40 (0.29; 0.55) 0.26 (0.17; 0.39) 0.17 (0.11; 0.27) p value for trend 0.004 <0.0001 <0.0001 <0.0001 Maternal education <12 years (reference category) 1 1 1 1 12 years 0.75 (0.69; 0.92) 0.73 (0.62; 0.86) 0.66 (0.54; 0.81) 0.60 (0.49; 0.74) >12 years 0.60 (0.48; 0.91) 0.54 (0.39; 0.75) 0.44 (0.29; 0.66) 0.36 (0.24; 0.55) p value for trend 0.002 0.0002 <0.0001 <0.0001 Paternal occupation Economically active (reference category) 1 UB/PTJ/pension/others 1.24 (1.02; 1.50) Economically inactive 1.53 (1.04; 2.25) p value for trend 0.029 Income adequacy More than enough (reference group) 1 Enough 1.27 (1.12; 1.73) Not enough 1.40 (1.17; 1.94) p value for trend 0.019 Weight status—BMI (kg/m2) Normal weight (reference category) 1 1 1 1 Overweight 2.9 (2.31; 3.64) 3.12 (2.63; 3.71) 1.89 (1.51; 2.38) 2.59 (2.10; 3.19) Obesity 8.42 (5.33; 12.28) 9.75 (6.91; 13.75) 3.59 (2.28; 5.65) 6.75 (4.43; 10.15) p value for trend <0.0001 <0.0001 <0.0001 <0.0001

p value for trend 0.029 Income adequacy More than enough (reference group) 1 Enough 1.27 (1.12; 1.73) Not enough 1.40 (1.17; 1.94) p value for trend 0.019 Weight status—BMI (kg/m2) Normal weight (reference category) 1 1 1 1 Overweight 2.9 (2.31; 3.64) 3.12 (2.63; 3.71) 1.89 (1.51; 2.38) 2.59 (2.10; 3.19) Obesity 8.42 (5.33; 12.28) 9.75 (6.91; 13.75) 3.59 (2.28; 5.65) 6.75 (4.43; 10.15) p value for trend <0.0001 <0.0001 <0.0001 <0.0001 The likelihood of developing pre-HTN and HTN among adolescents from families with parental hypertension was almost twice as high as among those from families without FHH (OR = 1.81, 95 % CI 1.31; 2.51, ptrend = 0.0003 and OR = 1.72, 95 % CI 1.34; 2.20, ptrend < 0.0001) for SBP and 1.4 times as high for DBP (OR = 1.39, 95 % CI 1.01; 1.91, ptrend = 0.039 and OR = 1.43, 95 % CI 1.02; 1.99, ptrend = 0.036). Adolescent females were 1.2 times (OR = 1.24, 95 % CI 1.01; 1.54, ptrend = 0.039) more likely than males to develop systolic pre-HTN and 1.3 times less likely to develop systolic HTN (OR = 0.77; 95 % CI 0.59; 0.91, ptrend = 0.045). Adolescents at age 18 as compared to those at age 10 were 2.3 times (OR = 2.30, 95 % CI 2.01; 2.97, ptrend = 0.005) and almost 2 times (OR = 1.72, 95 % CI 1.05; 2.81, ptrend = 0.029) more likely to develop diastolic pre-HTN and HTN. They were 1.2 times (OR = 1.23, 95 % CI 1.06; 2.01, ptrend = 0.046) more likely to develop systolic pre-HTN and 1.4 times (OR = 1.39, 95 % CI 1.09; 2.13, ptrend = 0.038) more likely to develop systolic HTN.

almost 2 times (OR = 1.72, 95 % CI 1.05; 2.81, ptrend = 0.029) more likely to develop diastolic pre-HTN and HTN. They were 1.2 times (OR = 1.23, 95 % CI 1.06; 2.01, ptrend = 0.046) more likely to develop systolic pre-HTN and 1.4 times (OR = 1.39, 95 % CI 1.09; 2.13, ptrend = 0.038) more likely to develop systolic HTN. Residents of large cities were almost twice less likely than their rural counterparts to develop systolic pre-HTN (OR = 0.56, 95 % CI 0.37; 0.82, ptrend = 0.004) and systolic HTN (OR = 0.40, 95 % CI 0.29; 0.55, ptrend < 0.0001). They were almost 4 times less likely to develop diastolic pre-HTN (OR = 0.26, 95 % CI 0.17; 0.39, ptrend < 0.0001) and almost 5 times to develop diastolic HTN (OR = 0.17, 95 % CI 0.11; 0.27, ptrend < 0.0001). Compared to adolescents whose mothers had low level of education, peers having mothers with high/academic education level were 1.7 times less likely to develop systolic pre-HTN (OR = 0.60, 95 % CI 0.48; 0.91, ptrend = 0.002) and 1.8 times less likely to develop systolic HTN (OR = 0.54, 95 % CI 0.39; 0.75, ptrend = 0.0002). In addition, they were 2.3 times less likely to develop diastolic pre-HTN (OR = 0.44, 95 % CI 0.29; 0.66, ptrend < 0.0001) and 2.8 times less likely to develop diastolic HTN (OR = 0.36, 95 % CI 0.24; 0.55, ptrend < 0.0001).

1.8 times less likely to develop systolic HTN (OR = 0.54, 95 % CI 0.39; 0.75, ptrend = 0.0002). In addition, they were 2.3 times less likely to develop diastolic pre-HTN (OR = 0.44, 95 % CI 0.29; 0.66, ptrend < 0.0001) and 2.8 times less likely to develop diastolic HTN (OR = 0.36, 95 % CI 0.24; 0.55, ptrend < 0.0001). Paternal occupation and income adequacy were two other factors associated with systolic pre-HTN. Using the employment status as reference category, the adjusted odds ratio for systolic pre-HTN risk from unemployment was OR = 1.53 (95 % CI 1.04; 2.25, ptrend = 0.029). Using the better-off financial situation (more than enough money) as a reference category, the adjusted odds ratio from income inadequacy for systolic pre-HTN risk was OR = 1.40 (95 % CI 1.17; 1.94, ptrend = 0.019). Using the normal BMI for age as reference category, the adjusted odds ratios for systolic pre-HTN and HTN risks from obesity were OR = 8.42 (95 % CI 5.33; 12.28, ptrend < 0.0001) and OR = 9.75 (95 % CI 6.91; 13; 75, ptrend < 0.0001) whereas for diastolic pre-HTN and HTN were OR = 3.59 (95 % CI 2.28; 5.65, ptrend < 0.0001) and OR = 6.75 (95 % CI 4.43; 10.15, ptrend < 0.0001), respectively.

atios for systolic pre-HTN and HTN risks from obesity were OR = 8.42 (95 % CI 5.33; 12.28, ptrend < 0.0001) and OR = 9.75 (95 % CI 6.91; 13; 75, ptrend < 0.0001) whereas for diastolic pre-HTN and HTN were OR = 3.59 (95 % CI 2.28; 5.65, ptrend < 0.0001) and OR = 6.75 (95 % CI 4.43; 10.15, ptrend < 0.0001), respectively. Discussion The present study provides the first data documenting social disparities in blood pressure levels among Polish adolescents. The findings revealed the multifactorial dependency of BP levels on geographic, i.e., rural or urban dwelling and SES-related familial influences at adolescence. The clustering structure of all factors involved in the analysis indicated that residential location might be operating through differential parental SES. The underlying pathways by which parental SES may influence their offspring BP levels include modifiable factors, such as the level of maternal education, status of paternal occupation and income interrelated to urbanization category of the place of residence after adjustment for FHH, subject’s sex, age and weight status. Consistent rural-to-urban and socioeconomic gradients were found in prevalence of elevated blood pressure, which increased with continuous lines from large cities through small- to medium-sized cities to villages and from high-SES to low-SES familial environments. Furthermore, the adjusted likelihood of developing HTN decreased with each step increase in maternal educational attainment, and pre-HTN decreased with increased maternal education, paternal employment status and income adequacy. The relationship between parental SES and BP levels as a gradient confirms persistence of social gradients that have been observed in Poland since the 1980s in other indicators of physical health [8, 10, 9].

inment, and pre-HTN decreased with increased maternal education, paternal employment status and income adequacy. The relationship between parental SES and BP levels as a gradient confirms persistence of social gradients that have been observed in Poland since the 1980s in other indicators of physical health [8, 10, 9]. Adverse consequences of low SES on BP levels and cardiovascular functions have been widely demonstrated in adults [77]. The findings of this study showed that social inequalities in BP levels manifest at adolescence. This is in line with the adolescent-emergent model (AEM) which states that relationships between SES and health outcomes are rather weak earlier in life but strengthen during adolescence when young people begin to be influenced by peers in their health behaviour [20, 30, 42]. Adolescence and, especially, puberty seems to be critical for the appearance of sexual dimorphism in BP which persists throughout adulthood. This finding is also in line with AEM and is most likely due to the activation of gonadal hormones with possibly a preponderant effect of testosterone involved during sexual maturation as well as acceleration in somatic growth during pubertal growth spurt [29, 81, 86]. The direct association of male sex with HTN and inverse association with pre-HTN found in our study need further analysis of data from longitudinal study.

possibly a preponderant effect of testosterone involved during sexual maturation as well as acceleration in somatic growth during pubertal growth spurt [29, 81, 86]. The direct association of male sex with HTN and inverse association with pre-HTN found in our study need further analysis of data from longitudinal study. As expected, elevated BP was independently associated with age. The likelihood of developing systolic pre-HTN and HTN increased twice with each year increase. Slightly weaker though significant association was observed between diastolic BP levels and age. Chronological age is a proxy for developmental trajectories. Its contribution varies in importance during each period prior to adulthood, so it does for BP level [18].

re-HTN and HTN increased twice with each year increase. Slightly weaker though significant association was observed between diastolic BP levels and age. Chronological age is a proxy for developmental trajectories. Its contribution varies in importance during each period prior to adulthood, so it does for BP level [18]. Not unexpectedly, our findings confirmed that parental HTN would be a major determinant of adolescent pre-HTN and HTN [54, 73]. The contribution of genetic determinants in developing high BP is accounted for 27 % of diastolic and 36 % of systolic BP [11]. Environmental exposures to permissive/adverse conditions via parental SES can be targeted in order to improve community and individual health. However, SES per se does not directly impact the physical status and physiological and functional capacity of growing individuals [12] neither health outcomes and so cannot be regarded as a treatable risk factor of elevated BP. There are several causal pathways that have been hypothesized for understanding the mechanisms that transfer geographical location and social and economic environment to health disparities at the community and individual levels. Rural communities are likely to be socially disadvantaged, facing job and neighbourhood strain, having low educational attainment, limited access to culture and the Internet and, in consequence, limited health literacy, which drive them to unhealthy behaviours ultimately resulting in chronic ill health, which, in turn, coupled with limited access to health care and low education level, may limit job opportunities. This is a vicious circle. Living in urban areas typically offers opportunities for better education, employment, better accessibility to health care services and adherence to medical treatment. On the other hand, rapid urban growth may generate numerous stressors resulting from population density, pollution, noise, unemployment and poverty [35, 53].

g in urban areas typically offers opportunities for better education, employment, better accessibility to health care services and adherence to medical treatment. On the other hand, rapid urban growth may generate numerous stressors resulting from population density, pollution, noise, unemployment and poverty [35, 53]. In Poland, disparities in somatic growth and selected health outcomes in the young age due to place of residence is a well-known phenomenon that has been reported by numerous studies. Despite inconsistencies as to the benefits of either environment, previous studies have shown that young people from urban areas are likely to be taller, thinner and earlier maturing as well as of better general health status as compared to their peers living in rural areas [8, 34, 48, 75]. Adverse effects of rural environment and its social structure on health outcomes found in our study are consistent with the results from the CBOS August 2013 report “Profile of the rural population” [78]. According to this report, the population living in rural areas showed that the better demographic situation characterized by positive demographic balance (the net population growth rate was higher in villages than in towns and cities—1.2 and 0.6 per 1000, in 2009, respectively) is accompanied by worse economic situation—a disposable per capita income of urban residents was twice as high as that of rural residents.

situation characterized by positive demographic balance (the net population growth rate was higher in villages than in towns and cities—1.2 and 0.6 per 1000, in 2009, respectively) is accompanied by worse economic situation—a disposable per capita income of urban residents was twice as high as that of rural residents. Importantly, maternal education, parental occupation, income adequacy and adolescent obesity remained significant when adjusted for all other relevant parent-related risk factors and place of residence. All these variables appear to act synergistically on adolescent BP levels via the acquisition of knowledge and skills that promote health associated with a higher level of schooling and the indirect effects of education on earnings and employment prospects [26, 87]. Wilson and colleagues, in their study on 76 black adolescents, revealed that adolescents who lived in poorer neighbourhoods had lower diastolic BPs if their mothers were more (vs. less) educated and their family had a higher (vs. lower) annual income [89]. The association between adolescent BP and maternal educational level has been demonstrated extensively [68, 83]. In our study, however, maternal occupational status interrelated with education and paternal occupation had no effect on blood pressure. Neither had family size. This finding is inconsistent with previous studies showing an independent effect of family size on offspring somatic growth [10, 48, 49]. At present study, parental educational attainment and earnings become more important for offspring BP than number of children at home.

ect on blood pressure. Neither had family size. This finding is inconsistent with previous studies showing an independent effect of family size on offspring somatic growth [10, 48, 49]. At present study, parental educational attainment and earnings become more important for offspring BP than number of children at home. The prevalence of elevated BP that has been reported in paediatrics recently varies substantially across countries [27, 31, 71]. The regional-wide variation in elevated BP prevalence is largely attributed to differences in geographic location, age range and methodology [39]. The overall prevalence of elevated BP was 6.6 % for pre-HTN and 8.9 % for HTN (SBP and/or DBP combined). These data indicate that the prevalence of systemic hypertension in the juvenile population in Poland has doubled over the last decade. Krzyżaniak and colleagues, in the national study of BP conducted in 2000 among Polish school children (7–19 years), reported the prevalence rate of HTN ∼4 % [58]. Similar figures, 4.9 % for HTN and 11.1 % for pre-HTN, were found in a large sample study of children and adolescents, aged 7 to 19 years in the city of Lodz, Poland [76].

colleagues, in the national study of BP conducted in 2000 among Polish school children (7–19 years), reported the prevalence rate of HTN ∼4 % [58]. Similar figures, 4.9 % for HTN and 11.1 % for pre-HTN, were found in a large sample study of children and adolescents, aged 7 to 19 years in the city of Lodz, Poland [76]. This upward trend in HTN is attributed at least in part to the rapid increase in adolescent overweight and obesity [44, 50] and the high prevalence of sedentary behaviours, physical inactivity and unhealthy dietary habits [37]. According to recent data from the national survey in Poland, the prevalence of overweight and obesity in 6–19-year-old children and adolescents is 16.4 % (18.7 and 14.3 %, boys and girls, respectively) and underweight—12.0 % total (10.0 and 13.7 %, boys and girls, respectively) [44]. The findings of the present study showed similar figures: 16.8 % of overweight/obese adolescents in total sample (20.1 % for boys and 14.4 % for girls). Prevalence of obesity was found to be twice higher in boys (4.2 %) than in girls (2.4 %), and in rural (2.9 %) than in urban (1.9 %) residence areas. Although these figures are not top ranked among European adolescents (HBSC), predictions based on the worldwide trends suggest that it may change in the near future resulting in increasing risk of developing elevated BP and undesirable cardiovascular consequences [27, 72].

(2.9 %) than in urban (1.9 %) residence areas. Although these figures are not top ranked among European adolescents (HBSC), predictions based on the worldwide trends suggest that it may change in the near future resulting in increasing risk of developing elevated BP and undesirable cardiovascular consequences [27, 72]. Brummett and colleagues revealed obesity and increased heart rate as key modifiable correlates of higher SBP and lower SES [15]. In the present study, weight status (BMI) appeared to be an independent most significant risk factor, suggesting its major role in mediating effects of parental SES on their offspring BP levels. In addition, it was found that higher parental SES was associated with lower BMI as it was claimed by Brummett and colleagues. There are many other studies identifying overweight and obesity as an independent significant factor of increased BP [21, 33, 37, 46, 80, 82].

ects of parental SES on their offspring BP levels. In addition, it was found that higher parental SES was associated with lower BMI as it was claimed by Brummett and colleagues. There are many other studies identifying overweight and obesity as an independent significant factor of increased BP [21, 33, 37, 46, 80, 82]. This study is not without its limitations. A cross-sectional design makes it difficult to assess the direction and causality. This design, however, was methodologically appropriate for solving the research question, i.e., evaluating the association between BP levels (outcome variable) and exposures (parental SES, weight status) [69]. There is also a possibility that confounding factors operating earlier in life and not included to this analysis introduced bias into the study results. However, studies of the association between prenatal factors and offspring BP have yielded mixed results indicating direct or inverse associations and null results as well [7, 64]. An example of maternal smoking during pregnancy may prove that bias of present results, if any, can be neglected. Smoking during pregnancy is more common among women with low SES [47]; therefore, it was argued that the relation between smoking during pregnancy and offspring BP is due to SES confounding rather than a true intrauterine effect [14, 47]. Another bias may be produced by errors in recall of the exposure and possible outcome. It would have been useful to have longitudinal, prospective information. The reliability of self-reported data has widely been discussed in the literature and involved in the premises of this study [40]. Finally, SES indicators do not include disposable per capita income, but we share the opinion that income adequacy reflects more adequately the families’ purchasing power [88].

information. The reliability of self-reported data has widely been discussed in the literature and involved in the premises of this study [40]. Finally, SES indicators do not include disposable per capita income, but we share the opinion that income adequacy reflects more adequately the families’ purchasing power [88]. The strengths of this study include a population-based cohort study of healthy adolescents, the clustering structure of parental SES reflecting that of the general population (30), a multivariate approach and integration of multiple factors hypothesized as to be associated with the outcome variable—adolescent arterial BP level. Conclusions The findings of the present study confirmed socioeconomic inequalities in blood pressure levels among adolescents. Young people living in rural areas are likely to be at a higher risk to develop elevated blood pressure than their better-off peers from urban areas. Weight status appeared to be the strongest confounder of adolescent blood pressure level and, at the same time, a mediator between their blood pressure and parental SES. This calls for regularly performed BP assessment and monitoring in this population. Effective strategies aimed at reducing global CVD risk should focus on obesity prevention and socioeconomic health inequalities as early as at adolescence by further trying to improve living and working conditions in rural areas. Abbreviations BMIBody mass index BPArterial blood pressure CIConfidence interval DBPDiastolic blood pressure FAS IIFamily Affluence Scale II FHHFamily history of hypertension GPGeneral practitioner

Weight status appeared to be the strongest confounder of adolescent blood pressure level and, at the same time, a mediator between their blood pressure and parental SES. This calls for regularly performed BP assessment and monitoring in this population. Effective strategies aimed at reducing global CVD risk should focus on obesity prevention and socioeconomic health inequalities as early as at adolescence by further trying to improve living and working conditions in rural areas. Abbreviations BMIBody mass index BPArterial blood pressure CIConfidence interval DBPDiastolic blood pressure FAS IIFamily Affluence Scale II FHHFamily history of hypertension GPGeneral practitioner GPMGneupel Precision Mechanics HTNHypertension Intra-TEMIntra-observer technical error Inter-TEMInter-observer technical error IOTFInternational Obesity Task Force MCAMultiple correspondence analysis MLRAMultiple multinomial logistic regression analysis OROdds ratio Pre-HTNPrehypertension SBPSystolic blood pressure SESSocioeconomic status Revisions received: 18 March 2015/21 April 2015 The authors would like to express their sincere gratitude to the anonymous reviewers for their helpful comments and valuable suggestions on this paper. Conflict of interest The authors declare that they have no competing interests. Funding This work was supported by research funds 2004–2008 and co-financed by a grant from Iceland, Liechtenstein and Norway through the EEA Financial Mechanism and the Norwegian Financial Mechanism. Project PL0255 ADOPOLNOR.

The authors would like to express their sincere gratitude to the anonymous reviewers for their helpful comments and valuable suggestions on this paper. Conflict of interest The authors declare that they have no competing interests. Funding This work was supported by research funds 2004–2008 and co-financed by a grant from Iceland, Liechtenstein and Norway through the EEA Financial Mechanism and the Norwegian Financial Mechanism. Project PL0255 ADOPOLNOR. Author’s contributions MK was the project manager, conceived the study design, coordinated the research, conceived the paper, performed the statistical computations and drafted the manuscript; AK coordinated the medical research; and BS-W, AK, MK-W and AS made contributions to the design of the medical research, analysed and interpreted the data, and revised the content of the manuscript. All authors critically read and approved the final manuscript.

men become more aware of the health aspects of nutrition during pregnancy and seek for more nutrition-related information. Compared to the period before conception and pregnancy, pregnant women are more interested in healthy food and may be more receptive to behaviour change and lifestyle interventions [2, 26, 32, 33]. The increased nutrient requirements during pregnancy are mostly covered by a balanced diet, but dietary supplements are often taken to improve maternal or foetal health status [1]. Maternal (health) behaviour and micronutrient status during pregnancy have been linked to the health status of the child [4, 5, 17, 19, 21]. Moreover, aberrant microbiota compositions have been detected during critical periods when early programming occurs, including pregnancy and early neonatal life [17, 19, 25]. Manipulation of the maternal microbiota composition through the use of probiotics may have subsequent consequences for the health of the offspring, as the presence of bacteria in human milk implicates that modulation of maternal gut microbiota during pregnancy and lactation could have an effect on infant health [10, 13, 14, 23]. Improvement of maternal intestinal microbiota composition, relief of possible gastrointestinal complaints, reduced infant’s risk of developing atopic dermatitis, atopic sensitization and gastrointestinal symptoms as well as changes in foetal and infant’s growth have been reported as positive health effects of probiotics [9, 12, 19, 22, 24]. In western societies, a substantial percentage of pregnant women appear to use probiotic supplements [4, 7]. Because of the potential positive effects for the health of the woman and her neonate, pregnancy is an opportune time for probiotic use. Both beliefs and knowledge seem to strongly affect the mother’s behaviour [6]. Review of the literature shows that ingestion of probiotics (combination of strains of Lactobacillus and Bifidobacterium) for a limited period of time during (late) pregnancy appears to be low risk, as it does not increase the rate of adverse pregnancy outcomes and seems to be well tolerated [30].

Electronic supplementary material Below is the link to the electronic supplementary material. Abbreviations ACE-IAngiotensin-converting enzyme inhibitor ARBAdrenergic receptor blocker BARß-Adrenergic receptor BNPBrain natriuretic peptide cHFChronic heart failure CHFCongestive heart failure DCMDilative cardiomyopathy HFpEFHeart failure with preserved ejection fraction HLHSHypoplastic left heart syndrome HTXHeart transplantation LVLeft ventricle NONitric oxide NYHANew York Heart Association RAA-SRenin-angiotensin-aldosterone system RVRight ventricle VSDVentricle septum defect VVIVentricular-ventricular interaction Authors’ contributions Dietmar Schranz conceptualized and drafted the initial manuscript and approved the final manuscript. Norbert Voelkel critically reviewed, revised, and approved the final manuscript as submitted. Compliance with ethical standards This article does not contain any studies with human participants performed by any of the authors. Conflict of interest The authors declare that they have no competing interests.

Introduction Adequate nutrition is of major importance for one’s health and well-being, especially during preconception and pregnancy [3, 5, 17, 18, 21]. Women become more aware of the health aspects of nutrition during pregnancy and seek for more nutrition-related information. Compared to the period before conception and pregnancy, pregnant women are more interested in healthy food and may be more receptive to behaviour change and lifestyle interventions [2, 26, 32, 33].

ther’s behaviour [6]. Review of the literature shows that ingestion of probiotics (combination of strains of Lactobacillus and Bifidobacterium) for a limited period of time during (late) pregnancy appears to be low risk, as it does not increase the rate of adverse pregnancy outcomes and seems to be well tolerated [30]. The aim of our study was to investigate characteristics and health behaviour patterns of mothers who use probiotics during pregnancy. As a secondary aim, we studied the effects of maternal use of probiotics on the offspring’s health during the first year of life. Materials and methods Study design and study population Subjects of the present study were mothers, with their child, participating in the ongoing Wheezing Illnesses Study Leidsche Rijn (WHISTLER) study. WHISTLER is a large prospective birth cohort study that started in December 2001 [15]. Baseline pre-pregnancy data of these parents were available from the Utrecht Health Project [11]. At the infant’s age of 3–8 weeks, information on pre- and post-natal risk factors is obtained by questionnaires and the infant’s birth weight and height, as well as gestational age and gender are recorded at an outpatient visit. Health parameters during the infant’s first year of life are followed in the WHISTLER study through linkage with the computerized medical files recorded by general practitioners.

tained by questionnaires and the infant’s birth weight and height, as well as gestational age and gender are recorded at an outpatient visit. Health parameters during the infant’s first year of life are followed in the WHISTLER study through linkage with the computerized medical files recorded by general practitioners. Definitions of outcomes General characteristics and behaviour patterns of the mother (and father) were extracted from the WHISTLER database. History of smoking was defined as smoking ever, prior to pregnancy (without limitation in months/years ago). A positive history of parental allergy was defined as questionnaire-reported allergy to pollen, house dust mite, pets or food. Maternal higher education was defined as higher vocational or university education. Maternal paid occupation was defined as having a paid job (yes or no) at time of completing the questionnaire.

positive history of parental allergy was defined as questionnaire-reported allergy to pollen, house dust mite, pets or food. Maternal higher education was defined as higher vocational or university education. Maternal paid occupation was defined as having a paid job (yes or no) at time of completing the questionnaire. At the visit shortly after birth, maternal use of probiotics was asked as follows: Did you use probiotics during pregnancy, either as in a probiotic milk or yoghurt product and/or probiotic-containing supplements? If yes, how many portions did you (on average) use per week? One portion was defined as use of one sachet or one capsule or use of one serving of a known probiotic-containing milk or yoghurt product. Active maternal smoking during pregnancy was considered present if the mother smoked at least one cigarette per day during pregnancy. Exposure to smoke during pregnancy was defined present when the mother smoked herself and/or if she reported being exposed to environmental cigarette smoke for at least 2 h per week. Use of supplements without doctor’s prescription was defined as maternal-reported use of at least one of the following, during the past 3 months:Vitamins, minerals, iron substitutes or resistance-increasing substitutes; Substitutes for other gastrointestinal complaints; Substitutes against cough and cold; Laxatives or sedatives. Use of homeopathic or herbal products during the past 3 months was also recorded.

At the visit shortly after birth, maternal use of probiotics was asked as follows: Did you use probiotics during pregnancy, either as in a probiotic milk or yoghurt product and/or probiotic-containing supplements? If yes, how many portions did you (on average) use per week? One portion was defined as use of one sachet or one capsule or use of one serving of a known probiotic-containing milk or yoghurt product. Active maternal smoking during pregnancy was considered present if the mother smoked at least one cigarette per day during pregnancy. Exposure to smoke during pregnancy was defined present when the mother smoked herself and/or if she reported being exposed to environmental cigarette smoke for at least 2 h per week. Use of supplements without doctor’s prescription was defined as maternal-reported use of at least one of the following, during the past 3 months:Vitamins, minerals, iron substitutes or resistance-increasing substitutes; Substitutes for other gastrointestinal complaints; Substitutes against cough and cold; Laxatives or sedatives. Use of homeopathic or herbal products during the past 3 months was also recorded. Eating fruits and/or vegetables on a regular basis was considered as a parameter for a healthy lifestyle, and the variables were defined as eating five or more pieces of fruit a week and preparing fresh vegetables seven or more times a week.

Use of homeopathic or herbal products during the past 3 months was also recorded. Eating fruits and/or vegetables on a regular basis was considered as a parameter for a healthy lifestyle, and the variables were defined as eating five or more pieces of fruit a week and preparing fresh vegetables seven or more times a week. To analyze the effects of the use of probiotics on the offspring’s health, we used follow-up data from the WHISTLER study. In this study data on respiratory symptoms, disease episodes and day-care attendance are recorded during the first year of life using monthly questionnaires. Furthermore, GP diagnoses on upper respiratory tract infections, lower respiratory tract infections, gastrointestinal tract infections and constitutional eczema are recorded using International Classification of Primary Care (ICPC) codes.

tendance are recorded during the first year of life using monthly questionnaires. Furthermore, GP diagnoses on upper respiratory tract infections, lower respiratory tract infections, gastrointestinal tract infections and constitutional eczema are recorded using International Classification of Primary Care (ICPC) codes. Analysis We compared mothers that used probiotics during pregnancy to non-users. In order to assess differences between groups, chi-square tests and independent samples t tests were used where appropriate. For all the analyses, firstly, the univariable association with use of probiotics during pregnancy was estimated using logistic regression. Secondly, we extended to multivariable logistic regression to adjust for maternal characteristics or behaviour patterns that were significantly associated with use of probiotics or showed a trend towards significance in the univariable analysis. A cut-off p value of <0.30 in the univariable association was used to insert variables into the multivariate model. Results are presented as odds ratios, with 95 % confidence intervals and p values. Associations were considered statistically significant if p values were ≤0.05. Analyses were run using SPSS version 20.0 (SPSS Inc., Chicago, IL, USA).

the univariable association was used to insert variables into the multivariate model. Results are presented as odds ratios, with 95 % confidence intervals and p values. Associations were considered statistically significant if p values were ≤0.05. Analyses were run using SPSS version 20.0 (SPSS Inc., Chicago, IL, USA). Results Data of 2491 mothers were used for analysis of their lifestyle and behaviour during pregnancy. Group characteristics and maternal attitudes between mothers who did and did not use probiotics during pregnancy are shown in Table 1. Of the total group, 13.7 % of the mothers reported use of probiotics during pregnancy. The mean usage per week was 3.5 portions (ranging from 1 to 15), in which there were no differences between the first and the second half of their pregnancy (data not shown). No differences were shown between both groups for gestational age, birth weight, maternal age at time of delivery, education and ethnicity of the mother. Probiotic-using mothers more often had a history of smoking, compared to non-users (p = 0.003). There was also a significant association between fathers with a history of smoking and the use of probiotics by the mother (p = 0.01). Maternal use of probiotics during pregnancy was significantly associated with use of other supplements and substitutes without doctor’s prescription (p = 0.02) and the use of homeopathic products (p < 0.001).Table 1 Characteristics of the study group Parental characteristics Total group Probiotic use Non-probiotic use p value

Results Data of 2491 mothers were used for analysis of their lifestyle and behaviour during pregnancy. Group characteristics and maternal attitudes between mothers who did and did not use probiotics during pregnancy are shown in Table 1. Of the total group, 13.7 % of the mothers reported use of probiotics during pregnancy. The mean usage per week was 3.5 portions (ranging from 1 to 15), in which there were no differences between the first and the second half of their pregnancy (data not shown). No differences were shown between both groups for gestational age, birth weight, maternal age at time of delivery, education and ethnicity of the mother. Probiotic-using mothers more often had a history of smoking, compared to non-users (p = 0.003). There was also a significant association between fathers with a history of smoking and the use of probiotics by the mother (p = 0.01). Maternal use of probiotics during pregnancy was significantly associated with use of other supplements and substitutes without doctor’s prescription (p = 0.02) and the use of homeopathic products (p < 0.001).Table 1 Characteristics of the study group Parental characteristics Total group Probiotic use Non-probiotic use p value n = 2491 (13.7 %) (86.3 %) Maternal age at time of delivery (mean, in years) (SD) 32.7 (3.9) 32.8 (3.7) 32.7 (4.0) 0.63a Maternal weight (mean, in kg) (SD) 71.4 (12.5) 70.2 (12.4) 71.5 (12.3) 0.12a Ethnicity mother (% western) 89.9 91.0 89.7 0.51b Maternal higher education (%) 67.3 70.0 66.9 0.31b Maternal paid occupation (%) 89.8 89.9 89.8 0.99b

n = 2491 (13.7 %) (86.3 %) Maternal age at time of delivery (mean, in years) (SD) 32.7 (3.9) 32.8 (3.7) 32.7 (4.0) 0.63a Maternal weight (mean, in kg) (SD) 71.4 (12.5) 70.2 (12.4) 71.5 (12.3) 0.12a Ethnicity mother (% western) 89.9 91.0 89.7 0.51b Maternal higher education (%) 67.3 70.0 66.9 0.31b Maternal paid occupation (%) 89.8 89.9 89.8 0.99b Maternal history of smoking (prior to pregnancy ever) (%) 35.4 43.3 34.1 0.003 b Maternal smoking during pregnancy (%) 6.2 7.0 6.1 0.51b Maternal smoke exposure during pregnancy (%) 14.9 17.0 14.5 0.23b Current smoking mother (%) 8.2 9.8 7.9 0.30b Paternal history of smoking (ever) (%) 40.6 48.3 39.4 0.01 b Current smoking father (%) 17.5 16.5 17.7 0.64b Pet ownership during pregnancy (%) 39.3 41.6 38.9 0.34b Use of alcohol (in general) (%) 79.6 81.2 79.4 0.50b Use of substitutes/supplements without doctor’s prescription during the past 3 months (%)c 75.9 81.5 75.0 0.02 b Use of homeopathic substitutes/herbal medicines during the past 3 months (%) 22.3 30.5 20.9 <0.001 b Use of fruits (5 or more pieces a week) (%) 64.8 68.0 64.3 0.24b Use of fresh vegetables (7 or more times a week) (%) 40.9 44.0 40.4 0.25b Maternal allergy (%)d 35.1 34.1 35.2 0.71b Maternal allergy (%)e 48.2 50.9 47.7 0.33b Paternal allergy (%)e 43.7 41.6 44.0 0.47b Children’s day-care visit during the first 6 months of life (%) 65.3 70.8 64.4 0.03 b p values in bold are statistically significant a t test bChi-square test ce.g. vitamins, minerals, iron substitutes, resistance-increasing substitutes, substitutes for other gastrointestinal complaints, substitutes against cough and cold, laxatives and sedatives

Children’s day-care visit during the first 6 months of life (%) 65.3 70.8 64.4 0.03 b p values in bold are statistically significant a t test bChi-square test ce.g. vitamins, minerals, iron substitutes, resistance-increasing substitutes, substitutes for other gastrointestinal complaints, substitutes against cough and cold, laxatives and sedatives dAllergy to pollen, dust, house mite and pets eAllergy to pollen, dust, house mite, pets, food or other Table 2 shows characteristics of the offspring of both probiotic-using and non-probiotic-using mothers. In the probiotic group, mothers more frequently gave birth to a boy (p = 0.04). Otherwise, no differences between the groups were demonstrated.Table 2 Characteristics of the study group Infant characteristics Total group Probiotic use group Non-probiotic use group p value n = 2491 (13.7 %) (86.3 %) Gestational age (mean, in weeks) (SD) 39.4 (1.4) 39.3 (1.5) 39.4 (1.4) 0.13a Birth weight (mean, in g) (SD) 3526 (514) 3529 (539) 3525 (510) 0.87a Gender (% boys within the group) 49.3 54.5 48.5 0.04 b Siblings (% with at least one) 54.0 50.9 54.5 0.22b Upper respiratory tract infectionsc (%) 46.9 46.6 47.0 0.88b Lower respiratory tract infectionsc (%) 9.0 11.0 8.6 0.15b Gastrointestinal tract infectionsc (%) 16.8 17.3 16.7 0.79b Constitutional eczemac (%) 12.9 14.3 12.7 0.39b p values in bold are statistically significant a t test bChi-square test cDuring the first year of life, doctor’s diagnosis

Upper respiratory tract infectionsc (%) 46.9 46.6 47.0 0.88b Lower respiratory tract infectionsc (%) 9.0 11.0 8.6 0.15b Gastrointestinal tract infectionsc (%) 16.8 17.3 16.7 0.79b Constitutional eczemac (%) 12.9 14.3 12.7 0.39b p values in bold are statistically significant a t test bChi-square test cDuring the first year of life, doctor’s diagnosis The results of the multivariable analysis on the association between the use of probiotics and other parental characteristics are shown in Table 3. The use of probiotics during pregnancy was increased in mothers who reported use of homeopathic substitutes or herbal medicines [odds ratio (OR) 1.65, 95 % confidence interval (CI) 1.17–2.33, p = 0.005]. Use of probiotics by the mother was also significantly associated with a higher frequency of history of smoking of both mother (OR 1.72, 95 % CI 1.25–2.37, p = 0.001) and father (OR 1.39, 95 % CI 1.01–1.89, p = 0.05).Table 3 Associations between parental characteristics and use of probiotics during pregnancy Multivariable analysis OR (95 % CI) p value Maternal history of smoking (prior to pregnancy ever) 1.72 (1.25–2.37) 0.001 Maternal smoke exposure during pregnancy 1.06 (0.67–1.70) 0.79 Use of substitutes/supplements without doctor’s prescription during the past 3 monthsa 1.05 (0.70–1.58) 0.82 Use of homeopathic substitutes/herbal medicines during the past 3 months 1.65 (1.17–2.33) 0.005 Use of fruits (5 or more pieces a week) 1.14 (0.82–1.60) 0.44 Use of fresh vegetables (7 or more times a week) 0.96 (0.70–1.32) 0.81 Paternal history of smoking (ever) 1.39 (1.01–1.89) 0.05

Maternal smoke exposure during pregnancy 1.06 (0.67–1.70) 0.79 Use of substitutes/supplements without doctor’s prescription during the past 3 monthsa 1.05 (0.70–1.58) 0.82 Use of homeopathic substitutes/herbal medicines during the past 3 months 1.65 (1.17–2.33) 0.005 Use of fruits (5 or more pieces a week) 1.14 (0.82–1.60) 0.44 Use of fresh vegetables (7 or more times a week) 0.96 (0.70–1.32) 0.81 Paternal history of smoking (ever) 1.39 (1.01–1.89) 0.05 Maternal higher education 1.27 (0.88–1.83) 0.20 Children’s day-care visit during the first 6 months of life 1.31 (0.92–1.88) 0.13 p values in bold are statistically significant aSee Table 1 for definition Table 4 shows the results of the multivariable analysis on the association of maternal use of probiotics during pregnancy and offspring’s disease symptoms during the first year of life. These symptoms did not differ between infants from probiotic-using mothers and non-probiotic-using mothers in this population.Table 4 Associations between maternal use of probiotics during pregnancy and infant characteristics Multivariable analysis OR (95 % CI) p value Upper respiratory tract infectionsa 0.97 (0.77–1.22) 0.79 Lower respiratory tract infectionsa 1.31 (0.90–1.91) 0.16 Gastrointestinal tract infectionsa 1.03 (0.76–1.40) 0.86 Constitutional eczemaa 1.15 (0.82–1.60) 0.42 aDuring the first year of life, doctor’s diagnosis

Table 4 shows the results of the multivariable analysis on the association of maternal use of probiotics during pregnancy and offspring’s disease symptoms during the first year of life. These symptoms did not differ between infants from probiotic-using mothers and non-probiotic-using mothers in this population.Table 4 Associations between maternal use of probiotics during pregnancy and infant characteristics Multivariable analysis OR (95 % CI) p value Upper respiratory tract infectionsa 0.97 (0.77–1.22) 0.79 Lower respiratory tract infectionsa 1.31 (0.90–1.91) 0.16 Gastrointestinal tract infectionsa 1.03 (0.76–1.40) 0.86 Constitutional eczemaa 1.15 (0.82–1.60) 0.42 aDuring the first year of life, doctor’s diagnosis Discussion This study shows that about one out of seven mothers in our population used probiotics during pregnancy. Use of probiotics during pregnancy was independently associated with use of homeopathic products and with a history of smoking of both mother and father.

Multivariable analysis OR (95 % CI) p value Upper respiratory tract infectionsa 0.97 (0.77–1.22) 0.79 Lower respiratory tract infectionsa 1.31 (0.90–1.91) 0.16 Gastrointestinal tract infectionsa 1.03 (0.76–1.40) 0.86 Constitutional eczemaa 1.15 (0.82–1.60) 0.42 aDuring the first year of life, doctor’s diagnosis Discussion This study shows that about one out of seven mothers in our population used probiotics during pregnancy. Use of probiotics during pregnancy was independently associated with use of homeopathic products and with a history of smoking of both mother and father. To our knowledge, to date, no other studies analyzed the association between maternal use of probiotic supplements and other behaviour patterns during pregnancy. The number of mothers that reported consumption of probiotics during pregnancy in our cohort corresponds reasonably with previous estimates [4, 7]. In our study, mothers that used probiotics during pregnancy were not characterized by specific maternal features (gestation, age, ethnicity, education) compared to mothers that did not use probiotics during pregnancy, although the literature shows that generally the adequacy of micronutrient intake during pregnancy is related to environmental, cultural and demographic variables [4, 5, 16].

racterized by specific maternal features (gestation, age, ethnicity, education) compared to mothers that did not use probiotics during pregnancy, although the literature shows that generally the adequacy of micronutrient intake during pregnancy is related to environmental, cultural and demographic variables [4, 5, 16]. To many, probiotics, homeopathic products and nutritional and dietary supplements belong to the category of complementary medicines. Pregnancy is a time to become more aware of a healthy lifestyle including healthy nutrition. Taking any form of supplement may be part of such a (change in) lifestyle. We hypothesized that next to the health-promoting properties, that are suggested for probiotics, mothers may use probiotics during pregnancy to compensate for adverse (prior) habits of themselves or their partners, for instance smoking. We showed comparable disease symptoms during the first year of life in the offspring from probiotic-using and non-probiotic-using mothers. Reviews and a meta-analysis demonstrated that current evidence on the effects of probiotics on the offspring’s health is fairly inconclusive [8, 25, 27]. Our data do not add evidence for a beneficial effect. The main strength of this study was the sample size which was large enough to estimate correlates of probiotic use during pregnancy. Our data have been prospectively documented, and all extensive parental characteristics and behaviour patterns could be aggregated from the database. Former studies of our group have demonstrated that the results may be generalized to other populations [28].

ugh to estimate correlates of probiotic use during pregnancy. Our data have been prospectively documented, and all extensive parental characteristics and behaviour patterns could be aggregated from the database. Former studies of our group have demonstrated that the results may be generalized to other populations [28]. However, there are also some limitations. Use of supplements, and especially probiotics, may have been underreported due to non-recall or format of the questions, as has been reported in the literature [31]. Nevertheless, we cannot conceive that non-recall of probiotic use would be related to use of other supplements or history of smoking and, therefore, is unlikely to have caused real bias. Also, neither the type of probiotic supplement nor the regularity of intake was specified and we were not able to investigate the use of probiotics by the mothers before and after pregnancy, which would have helped to discriminate mothers based on their using habits. There is emerging evidence that the effect of probiotics is strain specific and that timing, administration route and the applied dose do affect the outcomes. We consider the current reported conclusions valid and reliable because of the standardized manner of data collection, correction for potential confounders and presence of the unselected population. Moreover, we consider our population size sufficiently large to render our results statistically robust.

ct the outcomes. We consider the current reported conclusions valid and reliable because of the standardized manner of data collection, correction for potential confounders and presence of the unselected population. Moreover, we consider our population size sufficiently large to render our results statistically robust. Thirdly, as reported earlier, in the study population of the Utrecht Health Project and WHISTLER study, a vast percentage of participants completed higher vocational or university education [20, 29]. High socio-economic status and ethnicity might have played a role in parents’ decision to participate, which results in a not entirely unselected study population. This effect will be mediated in the population but has to be taken into account when results are generalized to lower class (young) families.

tion [20, 29]. High socio-economic status and ethnicity might have played a role in parents’ decision to participate, which results in a not entirely unselected study population. This effect will be mediated in the population but has to be taken into account when results are generalized to lower class (young) families. Conclusion This study shows that about one out of seven mothers in our population use probiotics during pregnancy. Probiotic-using mothers are not characterized by specific maternal features (gestation, age, ethnicity, education) compared to non-users. Use of probiotics during pregnancy is independently associated with use of homeopathic products and with parental history of smoking. According to common doctors’ diagnosed disease symptoms in the offspring the first year of life, no differences between groups are observed. Using probiotics and/or other health-related products without doctor’s prescription during pregnancy, for their health-promoting properties, might point to compensation for types of less favourable behaviour such as parental smoking. Caregivers and people concerned with pregnant women should be aware of this effect when discussing (nutritional) behaviour. Abbreviations ICPCInternational Classification of Primary Care WHISTLERWheezing Illnesses Study Leidsche Rijn

Conclusion This study shows that about one out of seven mothers in our population use probiotics during pregnancy. Probiotic-using mothers are not characterized by specific maternal features (gestation, age, ethnicity, education) compared to non-users. Use of probiotics during pregnancy is independently associated with use of homeopathic products and with parental history of smoking. According to common doctors’ diagnosed disease symptoms in the offspring the first year of life, no differences between groups are observed. Using probiotics and/or other health-related products without doctor’s prescription during pregnancy, for their health-promoting properties, might point to compensation for types of less favourable behaviour such as parental smoking. Caregivers and people concerned with pregnant women should be aware of this effect when discussing (nutritional) behaviour. Abbreviations ICPCInternational Classification of Primary Care WHISTLERWheezing Illnesses Study Leidsche Rijn We would like to thank all the parents and their children for their willingness to participate in the ongoing WHISTLER study and the Utrecht Health Project. The WHISTLER study is supported by grants from the Netherlands Organization for Health Research and Development (ZON-MW, no. 2001-1-1322), by the University Medical Center Utrecht, and by an unrestricted research grant from GlaxoSmithKline, The Netherlands. This part of the study was also partially funded by the Netherlands Enterprise Agency (Agentschap NL, nowadays RVO, grant number FND-06015).

r Health Research and Development (ZON-MW, no. 2001-1-1322), by the University Medical Center Utrecht, and by an unrestricted research grant from GlaxoSmithKline, The Netherlands. This part of the study was also partially funded by the Netherlands Enterprise Agency (Agentschap NL, nowadays RVO, grant number FND-06015). Conflict of interest Kors van der Ent has received unrestricted research grants from Grünenthal and GlaxoSmithKline. The other authors declare that they have no conflicts of interest to disclose. The funding agencies did not have any role in the study design, data collection, analysis and interpretation of data, or in the writing of the article and the decision to submit it for publication. Compliance with ethical standards ᅟ Ethical approval This study was approved by the medical ethics committee of the University Medical Center Utrecht, Utrecht, The Netherlands, and has, therefore, been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Informed consent Written informed consent was obtained prior to inclusion in the study.

Ethical approval This study was approved by the medical ethics committee of the University Medical Center Utrecht, Utrecht, The Netherlands, and has, therefore, been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Informed consent Written informed consent was obtained prior to inclusion in the study. Authors’ contribution NR designed research questions, analyzed data, wrote the paper and had primary responsibility for the final content. AG conducted research, analyzed data and wrote the paper. CU planned and set up the WHISTLER cohort and supervised statistical analysis. AV and GR designed research questions and critically reviewed the manuscript. KE designed research questions, planned and set up the WHISTLER cohort and supervised writing of the paper. All authors read and approved the final manuscript.

Introduction The febrile child is a common presentation to emergency departments (ED) with 10 to 20 % of all paediatric patients due to febrile illness alone [14, 17, 28]. Most children suffering from simple self-limiting infections do not need treatment. However, a small proportion will have serious bacterial infections (SBI) which require investigation, hospital admission, antibiotics and in some cases intensive care admission. Understanding health care professionals’ decision making, particularly regarding to diagnosis, treatment and follow-up is of vital importance, particularly as ED’s become increasingly overcrowded [33, 34]. Moreover, diagnostic errors, especially in infectious diseases, are amongst the most common medical misadventures of malpractice lawsuits in paediatrics [16]. To support decision making in febrile children, different clinical prediction models have been developed in the past decade [4, 7, 12, 19, 30, 31]. Although most studies on prediction models report good accuracy and high compliance, implementation in paediatric emergency care is limited. One of the reasons might be that clinicians’ intuitive estimation of probabilities may be as good as, or better than, prediction models [15, 21, 27]. Moreover, the lack of evidence on clinically based decision thresholds makes the application process of prediction models in clinical practice complex.

y care is limited. One of the reasons might be that clinicians’ intuitive estimation of probabilities may be as good as, or better than, prediction models [15, 21, 27]. Moreover, the lack of evidence on clinically based decision thresholds makes the application process of prediction models in clinical practice complex. The aim of this study was to estimate risk thresholds at which children would be managed as SBI according to clinicians’ judgement by assessment of video vignettes of febrile children visiting the ED. Secondary measures included determining the effect of investigations by recording risk estimations after information on C-reactive protein value, determining the presenting characteristics that influence these risks and comparing clinician perceived risk with risk estimates using a validated prediction model (Feverkidstool) [19]. Methods Study design and setting We performed a cross-sectional study with real life video vignettes of febrile children who presented themselves to the children’s ED of the Leicester Royal Infirmary in Leicester, UK. All parents had given formal consent for the video images to be viewed by healthcare professionals under trust policy guidelines via previously published process [13]. Ethical consent for the collection of video images process had been granted by the National Research Ethics Committee East Midlands.

y in Leicester, UK. All parents had given formal consent for the video images to be viewed by healthcare professionals under trust policy guidelines via previously published process [13]. Ethical consent for the collection of video images process had been granted by the National Research Ethics Committee East Midlands. Study population Paediatricians and paediatric emergency clinicians from the source population of the REPEM network (Research in Paediatric Emergency Medicine, Europe; www.pemdatabase.org/REPEM.html), and Paediatricians at teaching hospitals with an interest in acute and emergency care in the Netherlands and United Kingdom, were invited (104 invitations). Non-responders were sent reminders at 4-week intervals, for a maximum of four mailings per subject. Study intervention—video vignettes Twenty-one online video vignettes of febrile children were shown to the study participants. The vignettes were a mix of children in different age categories with potential SBI and children with simple self-limiting problems reflecting the different levels of severity in febrile child presentations in practice. The videos, with a mean duration of about 30 s, were originally recorded for educational purposes of paediatricians in training as part of the REMIT (Refining Evaluation Methodologies for Practice Changing Interventions) study (ISRCTN94772165). Background history and vital signs were reported as added text or could easily be interpreted from the video vignettes.

s, were originally recorded for educational purposes of paediatricians in training as part of the REMIT (Refining Evaluation Methodologies for Practice Changing Interventions) study (ISRCTN94772165). Background history and vital signs were reported as added text or could easily be interpreted from the video vignettes. Initially, the participants were asked if they should manage the febrile child as having a SBI based on the vignette and background history (e.g. duration of fever) alone. Next, they were asked to assess the actual risk of the child having a SBI on a visual analogue scale (VAS1). Finally, we add different values of C-reactive protein (CRP) and asked if their risk assessment would have changed (VAS2). The online vignettes and the respondents were hosted on a secure password protected server.

t, they were asked to assess the actual risk of the child having a SBI on a visual analogue scale (VAS1). Finally, we add different values of C-reactive protein (CRP) and asked if their risk assessment would have changed (VAS2). The online vignettes and the respondents were hosted on a secure password protected server. Data collection All data collected online was exported in an anonymised format as an Excel file. We collected answers on the following questions: (1) Would you manage this child as having a serious bacterial infection? (Answers: yes/no). (2) Which diagnostics or therapeutics would you perform? (Options: no action and/or discharge; antipyretic; fluid trial; blood tests; chest-radiography; lumbar puncture; urine dipstick; oral antibiotics; intravenous antibiotics; admission). Study participants could tick as many items as they judged relevant. (3) What is the chance of SBI in this child? (Answer: 0–100 % on a VAS (VAS1)) [1]. As CRP is the strongest predictor of the Feverkidstool, we studied the additional value of CRP in clinicians’ management decision, with the following question: (4) A CRP is taken and returns at (continuous value) mg/l. What is the chance of SBI in this child? (Answer: 0–100 % VAS (VAS2)).

0–100 % on a VAS (VAS1)) [1]. As CRP is the strongest predictor of the Feverkidstool, we studied the additional value of CRP in clinicians’ management decision, with the following question: (4) A CRP is taken and returns at (continuous value) mg/l. What is the chance of SBI in this child? (Answer: 0–100 % VAS (VAS2)). Participant’s background information was collected after finishing the video vignettes. These questions included (1) Are you a: Emergency Medicine clinician/Paediatrician; (2) How long have you been working as an Emergency Medicine clinician/paediatrician? (Options: <5 years; 5–10 years; 10–15 years; >15 years); (3) Have you ever missed/recognised a serious infection too late? (Options: yes/no). Definitions and outcome measures All participants were informed about the predefined SBI definition in the letter for the study invitation: culture or radiographically proven bacterial infection (e.g. meningitis, sepsis, bacteremia, pneumonia, urinary tract infection, bacterial gastroenteritis, osteomyelitis or ethmoiditis). The outcome SBI in the vignettes was defined as management of the child as having a SBI.

tion in the letter for the study invitation: culture or radiographically proven bacterial infection (e.g. meningitis, sepsis, bacteremia, pneumonia, urinary tract infection, bacterial gastroenteritis, osteomyelitis or ethmoiditis). The outcome SBI in the vignettes was defined as management of the child as having a SBI. Detailed descriptions on the Feverkidstool development and validation have been published earlier [19]. The originally reported discriminative ability according to the area under the receiver operating characteristic curve (AUC) of the model to predict pneumonia was 0.81 (standard error 0.04) and for other SBI 0.86 (standard error: 0.03) [19]. As the Feverkidstool was based on a polytomous logistic regression model, two risk scores were calculated, one for pneumonia and one for other SBI (e.g. urinary tract infection). We used the highest risk score in the comparison with the VAS risk scores of the video vignettes. We dichotomised the outcome of performed diagnostics and/or therapeutics. This outcome was scored ‘present’ if participants ticked fluid trial, blood tests, chest-radiography, lumbar puncture, urine dipstick, administration of oral/intravenous antibiotics and/or admission. When ‘no action and/or discharge and/or antipyretics’ was chosen, the outcome was scored as ‘not present’.

therapeutics. This outcome was scored ‘present’ if participants ticked fluid trial, blood tests, chest-radiography, lumbar puncture, urine dipstick, administration of oral/intravenous antibiotics and/or admission. When ‘no action and/or discharge and/or antipyretics’ was chosen, the outcome was scored as ‘not present’. All vignettes had a statement on age, temperature and duration of fever. Abnormal clinical signs and symptoms were distributed amongst the different vignettes, with ten vignettes having one alarming sign, four vignettes with two alarming signs and seven vignettes having three or more alarming signs. Statistical analysis First, we assessed the range of estimated median risks by clinical judgement (VAS) and the risk with the added value of CRP. Second, we measured the patient characteristics which enact SBI management with discrete choice experiment (DCE) analysis. Finally, we compared VAS risk scores with prediction model based judgement (Feverkidstool).

the range of estimated median risks by clinical judgement (VAS) and the risk with the added value of CRP. Second, we measured the patient characteristics which enact SBI management with discrete choice experiment (DCE) analysis. Finally, we compared VAS risk scores with prediction model based judgement (Feverkidstool). DCEs are a quantitative approach to assess preferences for e.g. medical interventions and are increasingly used in health care [10]. In DCEs, it is assumed that important items influencing medical interventions, such as vital signs, can be described by its characteristics (i.e. attributes) [24]. Those characteristics are further specified by variants of that characteristics (i.e. attribute levels). A second assumption is that the levels of those attributes are determined by the individuals’ preference for a medical intervention [24]. We studied the clinical variables of the Feverkidstool (www.erasmusmc.nl/feverkidstool) as attributes to the decision whether or not to manage febrile children of the vignettes as a SBI [13]. All DCE data was analysed by taking each choice amongst the two management alternatives as an observation. Using the Nlogit software http://www.limdep.com/ to the next sentence, the observations were analysed by a logit model. As there was a lack of diversity amongst the clinical variables ‘oxygen saturation’ and ‘tachypnoea’ between the vignettes, we could not analyse these variables accordingly. The variables tachycardia and prolonged capillary refill were taken together as one clinical variable as their correlation was too high. The influence of the different variable coefficients was tested for statistical significance (p value ≤0.05). As at this moment, no formal statistical methods to determine sample sizes for DCE exist; our study strived to reach at least 40 respondents in line with previous studies [6, 26].

heir correlation was too high. The influence of the different variable coefficients was tested for statistical significance (p value ≤0.05). As at this moment, no formal statistical methods to determine sample sizes for DCE exist; our study strived to reach at least 40 respondents in line with previous studies [6, 26]. Results Of the 104 invited participants, 50.4 % agreed to participate and 42 (40.4 %) participants finished the online video vignettes. The 42 final participants included 83 % paediatricians and 17 % paediatric emergency medicine physicians. Fifty per cent of the participants had a working experience of more than 10 years. Almost half of the participants had at least once missed or delayed recognised serious infection (Table 1).Table 1 Demographics Participants (n = 42) Specialisma Paediatric emergency medicine clinician 7 (16.7) Paediatrician 35 (83.3) Years of working experiencea <5 years 4 (9.5) 5–10 year 17 (40.5) 10–15 years 9 (21.4) >15 years 12 (28.6) Missed/recognised a serious infection too latea Yes 19 (45.2) No 23 (54.8) aAbsolute number (percentage)

Results Of the 104 invited participants, 50.4 % agreed to participate and 42 (40.4 %) participants finished the online video vignettes. The 42 final participants included 83 % paediatricians and 17 % paediatric emergency medicine physicians. Fifty per cent of the participants had a working experience of more than 10 years. Almost half of the participants had at least once missed or delayed recognised serious infection (Table 1).Table 1 Demographics Participants (n = 42) Specialisma Paediatric emergency medicine clinician 7 (16.7) Paediatrician 35 (83.3) Years of working experiencea <5 years 4 (9.5) 5–10 year 17 (40.5) 10–15 years 9 (21.4) >15 years 12 (28.6) Missed/recognised a serious infection too latea Yes 19 (45.2) No 23 (54.8) aAbsolute number (percentage) Study intervention—video vignettes In Table 2, clinical characteristics of the video vignettes are summarised. Median age of the children was 12.0 months (interquartile range (IQR) 2.0–72.0), 57 % were boys and the median C-reactive protein level (CRP) was 60 mg/l (IQR 10.0–110.0). Answers on the four questions of the video vignettes are summarised in Table 3. Forty-one per cent of the video vignettes are managed as having a SBI according to the participants. Diagnostics and/or therapeutics were started in 77 % of the video vignettes. Median risk before the knowledge of CRP (VAS1) was 20.0 % (IQR 9.0–50.0) and with CRP information the risk (VAS2) increased to 30.0 % (IQR 10.0–60.0). As CRP values were already available in the first video for vignette 3 and 21, no change in risk could be measured. Details of performed diagnostics, therapeutics and follow-up are described in Table 4. More diagnostics and/or therapeutics were performed when the child was managed as SBI. Antipyretics were given in 65 % of the video vignettes with no differences when stratifying by outcome (SBIM). In 94 % of the video vignettes who were managed as SBI, blood tests were done and 71 % were hospitalised (Table 4).Table 2 Clinical variables

d/or therapeutics were performed when the child was managed as SBI. Antipyretics were given in 65 % of the video vignettes with no differences when stratifying by outcome (SBIM). In 94 % of the video vignettes who were managed as SBI, blood tests were done and 71 % were hospitalised (Table 4).Table 2 Clinical variables Video vignettes (n = 21) Clinical variables Age (months)a 12.0 (2.0–72.0) ≤3 months 4 (19.0) >3 months–<1 year 6 (28.6) ≥1 year–≤18 months 5 (23.8) >18 months 6 (28.6) Sex, male* 12 (57.1) Temperaturea(°C) 38.7 (38.5–40.2) 38.5–38.9 °C 12 (57.1) 39.0–39.9 °C 7 (33.3) ≥40.0 °C 2 (9.5) Duration fevera (days) 2.0 (1.0–3.0) Prolonged capillair refill* (>2 s) 4 (19.0) Chest wall retractions* 3 (14.3) Ill appearance* 7 (33.3) Saturation (<94 % O2)* 1 (4.8) Respiratory ratea (/minute) 32.0 (20.0–60.0) Tachypnoea 1 (4.8) Heart ratea (/minute) 132.0 (100.0–172.0) Tachycardia 4 (19.0) CRPa (mg/L) 60.0 (10.0–110.0) <40 mg/l 8 (38.1) ≥40 mg/l 7 (33.3) ≥80 mg/l 6 (28.6) Presence of no. alarming symptomsa ≤1 11 (0–1) >1 10 (2–5) *Absolute number (percentage) aMedian (min; max) Table 3 Answers of 42 participants on 21 video vignettes (n total = 882)

Video vignettes (n = 21) Clinical variables Age (months)a 12.0 (2.0–72.0) ≤3 months 4 (19.0) >3 months–<1 year 6 (28.6) ≥1 year–≤18 months 5 (23.8) >18 months 6 (28.6) Sex, male* 12 (57.1) Temperaturea(°C) 38.7 (38.5–40.2) 38.5–38.9 °C 12 (57.1) 39.0–39.9 °C 7 (33.3) ≥40.0 °C 2 (9.5) Duration fevera (days) 2.0 (1.0–3.0) Prolonged capillair refill* (>2 s) 4 (19.0) Chest wall retractions* 3 (14.3) Ill appearance* 7 (33.3) Saturation (<94 % O2)* 1 (4.8) Respiratory ratea (/minute) 32.0 (20.0–60.0) Tachypnoea 1 (4.8) Heart ratea (/minute) 132.0 (100.0–172.0) Tachycardia 4 (19.0) CRPa (mg/L) 60.0 (10.0–110.0) <40 mg/l 8 (38.1) ≥40 mg/l 7 (33.3) ≥80 mg/l 6 (28.6) Presence of no. alarming symptomsa ≤1 11 (0–1) >1 10 (2–5) *Absolute number (percentage) aMedian (min; max) Table 3 Answers of 42 participants on 21 video vignettes (n total = 882) Alarming symptoms Question 1 Question 2 Question 3 Question 4 Video vignette No. SBIM Dx/Tx* VAS1a (%) CRPa (mg/l) VAS2a (%) 1 2 3 (7.1) 16 (38.1) 10.0 (4.8–20.0) 85 26.5 (10.0–44.8) 2 1 29 (69.0) 42 (100.0) 30.0 (20.0–50.3) 70 54.5 (30.0–79.3) 3 1 11 (26.2) 26 (61.9) 16.0 (7.8–32.8) 38 10.0 (4.8–23.0) 4 3 27 (64.3) 39 (92.9) 27.0 (10.0–51.8) 100 60.0 (30.8–76.0) 5 3 41 (97.6) 42 (100.0) 81.0 (60.0–90.0) 65 71.5 (50.0–90.0) 6 3 13 (31.0) 36 (85.7) 20.5 (10.0–40.0) 90 44.0 (20.0–69.3) 7 1 23 (54.8) 33 (78.6) – 10 30.5 (11.0–60.3) 8 1 27 (64.3) 41 (97.6) 30.0 (14.0–50.0) 25 17.0 (10.0–29.3) 9 1 4 (9.5) 25 (59.5) 10.0 (4.0 21.0) 30 9.5 (4.0–21.0) 10 2 41 (97.6) 42 (100.0) 80.0 (62.5–90.0) 50 69.5 (40.0–90.0) 11 4 9 (21.4) 38 (90.5) 10.5 (5.0–21.0) 90 40.5 (21.0–69.0) 12 1 5 (11.9) 32 (76.2) 10.5 (5.8–21.0) 28 6.0 (4.0–14.5) 13 1 0 (0) 11 (26.2) 5.0 (2.8–15.5) 36 4.0 (0.8–12.0) 14 6 16 (38.1) 38 (90.5) 16.0 (9.8–40.0) 60 30.0 (16.3–50.0) 15 3 32 (76.2) 42 (100.0) 41.5 (20.0–69.3) 75 62.5 (38.5–80.0) 16 1 1 (2.4) 15 (35.7) 8.5 (2.8–15.8) 10 1.0 (0.0–6.0) 17 3 41 (97.6) 42 (100.0) 82.5 (69.8–93. 3) 48 81.5 (49.8–91.8) 18 2 7 (16.7) 32 (76.2) 11.5 (7.8–25.3) 110 60.0 (31.0–80.0) 19 1 9 (21.4) 24 (57.1) 15.5 (8.3–30.0) 75 30.5 (19.3–50.0) 20 1 16 (38.1) 35 (83.3) 21.0 (10.0–45.5) 35 13.5 (8.0–36.3) 21 2 10 (23.8) 29 (69.0) – 100 19.5 (6.8–30.3) Total 365/882 (41.4) 680/882 (77.1) 20.0 (9.0–50.0) 60.0 (35.0–85.0) 30.0 (10.0–61.0) *Absolute number (percentage); aMedian (25–75 percentile)

.0–80.0) 19 1 9 (21.4) 24 (57.1) 15.5 (8.3–30.0) 75 30.5 (19.3–50.0) 20 1 16 (38.1) 35 (83.3) 21.0 (10.0–45.5) 35 13.5 (8.0–36.3) 21 2 10 (23.8) 29 (69.0) – 100 19.5 (6.8–30.3) Total 365/882 (41.4) 680/882 (77.1) 20.0 (9.0–50.0) 60.0 (35.0–85.0) 30.0 (10.0–61.0) *Absolute number (percentage); aMedian (25–75 percentile) Question 1: Would you manage this child as having a serious bacterial infection? SBIM: child is managed as having SBI according to participant Question 2: Which diagnostics or therapy would you perform? Dx/Tx: diagnostics and/ or therapy done (defined as: fluid trial; blood tests; chest-radiography; lumbar puncture; urine dipstick; administration of oral/ intravenous antibiotics or admission) Question 3: What is the chance of SBI in this child? (Answer: 0–100 % on a VAS (VAS1)) VAS1: risk assessment without knowledge of CRP (0–100 % VAS) Question 4: A CRP is taken and returns at (continuous value) mg/l. What is the chance of SBI in this child? (Answer: 0–100 % VAS (VAS2) VAS2: risk assessment with knowledge of CRP (0–100 % VAS) Table 4 Diagnostics, therapy and follow-up

Question 3: What is the chance of SBI in this child? (Answer: 0–100 % on a VAS (VAS1)) VAS1: risk assessment without knowledge of CRP (0–100 % VAS) Question 4: A CRP is taken and returns at (continuous value) mg/l. What is the chance of SBI in this child? (Answer: 0–100 % VAS (VAS2) VAS2: risk assessment with knowledge of CRP (0–100 % VAS) Table 4 Diagnostics, therapy and follow-up Diagnostics SBIM yes n = 365 SBIM no n = 517 N total = 882 No diagnostics 4 (1.1) 100 (19.3) 104 (11.8) Urine dipstick 252 (69.0) 134 (25.9) 386 (43.8) Fluid trial 135 (37.0) 73 (14.1) 208 (23.6) Blood tests 344 (94.2) 180 (34.8) 524 (59.4) Chest-radiography 112 (30.7) 76 (14.7) 188 (21.3) Lumbar puncture 140 (38.4) 9 (1.7) 149 (16.9) Therapy and follow-up SBIM yes n = 365 SBIM no n = 517 N total = 882 Antipyretics 244 (66.8) 330 (63.8) 574 (65.1) No therapy 74 (20.3) 404 (78.1) 478 (54.2) Oral antibiotics 11 (3.0) 16 (3.1) 27 (3.1) Intravenous antibiotics 209 (57.3) 4 (0.8) 213 (24.1) Admission 258 (70.7) 96 (18.6) 354 (40.1) Discharge 75 (20.5) 405 (78.3) 480 (54.4)

up SBIM yes n = 365 SBIM no n = 517 N total = 882 Antipyretics 244 (66.8) 330 (63.8) 574 (65.1) No therapy 74 (20.3) 404 (78.1) 478 (54.2) Oral antibiotics 11 (3.0) 16 (3.1) 27 (3.1) Intravenous antibiotics 209 (57.3) 4 (0.8) 213 (24.1) Admission 258 (70.7) 96 (18.6) 354 (40.1) Discharge 75 (20.5) 405 (78.3) 480 (54.4) Clinical judgement versus different levels of CRP In Fig. 1, the differences in clinical risk scores are visualised versus different levels of CRP values. The median clinical risk differences (VAS2-VAS1) were positively correlated with a higher level of CRP (SBIM yes: Pearson correlation 0.53 (p = 0.000) and SBIM no: Pearson correlation 0.68 (p = 0.000)). Risk scores of children classified initially already as being managed as SBI were influenced only by high levels of CRP (>65 mg/l), whereas children not managed initially as SBI were influenced by lower CRP levels (>40 mg/l) (Fig. 1).Fig. 1 Relation video vignettes risk difference and C-reactive protein (mg/l)

0.000)). Risk scores of children classified initially already as being managed as SBI were influenced only by high levels of CRP (>65 mg/l), whereas children not managed initially as SBI were influenced by lower CRP levels (>40 mg/l) (Fig. 1).Fig. 1 Relation video vignettes risk difference and C-reactive protein (mg/l) Discrete choice experiment—video vignettes Discrete choice experiment was based upon 20 video vignettes as the clinical variables of one video were too correlated. Almost all clinical variables of the Feverkidstool could be tested with DCE analysis, except for CRP, oxygen saturation and tachypnoea. Ranking and coefficients of influencing variables on management decision of febrile children according to the DCE analysis are presented in Table 5. All tested clinical variables influenced the decision on management of febrile children significantly. Ill appearance and the combined variables of prolonged capillary refill and tachycardia were the most influencing factors and age and duration of fever the least influencing factors.Table 5 Influencing variables on management decisions in febrile children (SBIM): a discrete choice experiment (n total = 882)

nificantly. Ill appearance and the combined variables of prolonged capillary refill and tachycardia were the most influencing factors and age and duration of fever the least influencing factors.Table 5 Influencing variables on management decisions in febrile children (SBIM): a discrete choice experiment (n total = 882) Clinical variables Ranking Coefficients (SE) p value Intercept −0.92 (0.37) 0.013 Ill appearance 1 1.15 (0.13) <0.001 Prolonged capillary refill (>2 s) and/or tachycardia 2 0.99 (0.17) <0.001 Chest wall retractions 3 −0.97 (0.22) <0.001 Temperature (≥39.0 °C) 4 0.77 (0.12) <0.001 Sex (male) 5 0.63 (0.11) <0.001 Duration fever (days) 6 0.51 (0.20) 0.009 Age (≥1 year) 7 −0.42 (0.12) 0.001 Saturation (<94 % O2) NA NA NA Tachypnoea NA NA NA SBIM: child is managed as having SBI according to participant NA not applicable, items could not been tested with DCE analyses

Clinical variables Ranking Coefficients (SE) p value Intercept −0.92 (0.37) 0.013 Ill appearance 1 1.15 (0.13) <0.001 Prolonged capillary refill (>2 s) and/or tachycardia 2 0.99 (0.17) <0.001 Chest wall retractions 3 −0.97 (0.22) <0.001 Temperature (≥39.0 °C) 4 0.77 (0.12) <0.001 Sex (male) 5 0.63 (0.11) <0.001 Duration fever (days) 6 0.51 (0.20) 0.009 Age (≥1 year) 7 −0.42 (0.12) 0.001 Saturation (<94 % O2) NA NA NA Tachypnoea NA NA NA SBIM: child is managed as having SBI according to participant NA not applicable, items could not been tested with DCE analyses Risk scores video vignettes—risk scores Feverkidstool The median clinical risk score (VAS2) according to the participants amongst those video vignettes who were assigned as managed as SBI was 60.0 % (IQR 30.0–80.5) compared to a risk score according to the Feverkidstool of 12.7 % (IQR 7.7–28.1) (Table 6). When the video vignettes were not managed as SBI, the clinical risk score (VAS2) amounted to 16.0 % (5.0–32.0) compared to a risk of 7.3 % (5.7–16.3) according the Feverkidstool (Table 7). The largest risk score differences between the vignettes and risk scores according to the Feverkidstool were seen for video vignettes with (various levels of) decreased consciousness or agitation. This item is clearly observed when watching the video vignettes, but this clinical variable is not included in the predictors of the Feverkidstool. Finally, no differences were found in median clinical risk scores when stratified for previously missed diagnoses of the participant (p = 0.218).Table 6 Clinical risk scores (video vignettes) versus prediction model risk scores (Feverkidstool) in children managed as SBI (SBIM = yes)

the predictors of the Feverkidstool. Finally, no differences were found in median clinical risk scores when stratified for previously missed diagnoses of the participant (p = 0.218).Table 6 Clinical risk scores (video vignettes) versus prediction model risk scores (Feverkidstool) in children managed as SBI (SBIM = yes) VAS2 (%)a Feverkidstool (%)a Video vignettes (no.) SBIM yes n = 365 n = 365 Risk ≤10 % 12 5.0 (2.0–9.5) 16.3 13 – – Risk 10–50 % 16 15.0 (15.0–15.0) 2.0 8 20.0 (12.0–30.0) 8.9 3 23.0 (9.0–61.0) 7.2 20 29.0 (12.5–61.8) 3.8 9 30.5 (8.3–66.3) 11.6 14 47.0 (32.0–76.8) 36.9 Risk ≥50 % 21 54.0 (17.8–80.3) 12.7 19 59.0 (45.0–90.0) 7.3 2 60.0 (30.0–80.0) 38.2 7 60.0 (30.0–72.0) 2.3 1 62.0 (50.0–62.0) 20.6 6 68.0 (35.0–83.0) 19.0 15 68.0 (52.3–80.8) 50.5 10 70.0 (44.5–90.0) 7.7 11 70.0 (57.5–81.0) 4.8 4 71.0 (35.0–80.0) 9.7 5 72.0 (50.0–90.0) 22.2 18 80.0 (21.0–82.0) 6.6 17 83.0 (50.0–92.5) 28.1 Total 60.0 (30.0–80.5) 12.7 (7.7–28.1) Table 7 Clinical risk scores (video vignettes) versus prediction model risk scores (Feverkidstool) children not managed as SBI (SBIM = no)

0 70.0 (44.5–90.0) 7.7 11 70.0 (57.5–81.0) 4.8 4 71.0 (35.0–80.0) 9.7 5 72.0 (50.0–90.0) 22.2 18 80.0 (21.0–82.0) 6.6 17 83.0 (50.0–92.5) 28.1 Total 60.0 (30.0–80.5) 12.7 (7.7–28.1) Table 7 Clinical risk scores (video vignettes) versus prediction model risk scores (Feverkidstool) children not managed as SBI (SBIM = no) VAS1 (%)aFeverkidstool (%)aVideo vignettes (no.) SBIM no n = 517 n = 517 Risk ≤10 % 16 1.0 (0.0–5.5) 2.0 13 4.0 (0.8–12.0) 5.7 12 6.0 (4.0–16.5) 16.3 9 8.5 (4.0–16.3) 11.6 3 10.0 (3.0–17.0) 7.2 8 10.0 (6.0–18.0) 8.9 20 10.0 (7.0–20.3) 3.8 Risk 10–50 % 10 13.0 (13.0–13.0) 7.7 21 15.5 (5.3–28.0) 12.7 7 17.0 (10.0–28.0) 2.3 1 20.0 (10.0–39.0) 20.6 14 20.0 (10.0–31.3) 36.9 17 20.0 (20.0–20.0) 28.1 19 25.0 (15.5–48.0) 7.3 11 30.0 (20.0–53.5) 4.8 5 40.0 (40.0–40.0) 22.2 6 40.0 (17.5–57.5) 19.0 2 42.0 (33.0–74.5) 38.2 4 46.0 (22.0–60.0) 9.7 15 46.5 (26.0–64.5) 50.5 Risk ≥50 % 18 60.0 (31.0–71.0) 6.6 Total 16.0 (5.0–32.0) 7.3 (5.7–16.3)

0) 2.3 1 20.0 (10.0–39.0) 20.6 14 20.0 (10.0–31.3) 36.9 17 20.0 (20.0–20.0) 28.1 19 25.0 (15.5–48.0) 7.3 11 30.0 (20.0–53.5) 4.8 5 40.0 (40.0–40.0) 22.2 6 40.0 (17.5–57.5) 19.0 2 42.0 (33.0–74.5) 38.2 4 46.0 (22.0–60.0) 9.7 15 46.5 (26.0–64.5) 50.5 Risk ≥50 % 18 60.0 (31.0–71.0) 6.6 Total 16.0 (5.0–32.0) 7.3 (5.7–16.3) Discussion Main findings This is the first study on real life video vignettes to determine febrile child characteristics which enact clinicians’ management decisions. High clinical risk scores to manage febrile children as SBI were created by clinicians. All tested clinical variables of the Feverkidstool influenced clinicians’ management decisions of febrile children significantly with ill appearance and aberrant circulatory signs being the most important. Moderate CRP levels influenced risk scores in children who were initially not managed as SBI whereas high CRP levels were needed to influence risk scores in children who were initially already managed as SBI. In children managed as SBI risk thresholds judged by the clinician were higher compared with predicted risk thresholds according to the Feverkidstool. Clinical risk thresholds of children not managed as having a SBI were more comparable to prediction model-based risk thresholds.

o were initially already managed as SBI. In children managed as SBI risk thresholds judged by the clinician were higher compared with predicted risk thresholds according to the Feverkidstool. Clinical risk thresholds of children not managed as having a SBI were more comparable to prediction model-based risk thresholds. Comparison with literature In this study, we aimed to get insight in patient characteristics and contextual factors influencing management decisions of the febrile child at the ED. One way to approach this process of diagnostic reasoning is decision making [11]. Decision making has been influenced by statistical models of reasoning under uncertainty using pre- and post-test probability according to Bayes’ theorem. This model deals with two major classes of errors in clinical reasoning: in the assessment of either pretest probability or the strength of the evidence [11]. Although the pretest probability of having SBI (prevalence of disease) is depending on several factors as for example age and relevant medical history, the pretest probability determined by health care setting was considered stable in the vignettes. However, we focused on the interpretation of clinicians’ strengths of evidence of the probability of a serious infection. For this decision process, we performed discrete choice experiment (DCE) analysis, which is an increasingly used method applied in studies where clinicians weigh clinical information in the diagnostic work-up [3].

used on the interpretation of clinicians’ strengths of evidence of the probability of a serious infection. For this decision process, we performed discrete choice experiment (DCE) analysis, which is an increasingly used method applied in studies where clinicians weigh clinical information in the diagnostic work-up [3]. In literature on diagnostic reasoning, evidence-based medicine is the most successful educational method in the translation of statistical decision theory into clinical practice [25]. Within this translation, we aimed to elaborate on the determination of quantitative decision thresholds that proved to be a complex topic. Most studies used optimised performance measures as area under the receiver operating characteristic curve (AUC) or sensitivity/specificity to establish these thresholds. Other studies described Delphi procedures to determine their clinical based cutoff points [5, 18, 20, 22, 32]. In our study, we described clinicians’ assigned median risk estimates according to which patients would have been managed as SBI. We observed agreement on clinical and prediction model-based risk thresholds when clinicians decided not to manage the febrile child as a SBI. However, the clinical risk threshold to manage the child as SBI was much higher compared with prediction model-based judgement. This phenomenon is well recognised, as clinicians don’t want to miss serious, but treatable diseases, there is a tendency to overestimate the probability of these diseases [11].

hild as a SBI. However, the clinical risk threshold to manage the child as SBI was much higher compared with prediction model-based judgement. This phenomenon is well recognised, as clinicians don’t want to miss serious, but treatable diseases, there is a tendency to overestimate the probability of these diseases [11]. Clinical and research implications The most important finding of this study includes the high risk scores clinicians assigned to those children who they would have managed as SBI (median risk 60.0 % (IQR 30–80.5)). This observation is in contrast to our hypothesis that very low risk thresholds might be chosen for specific diagnosis with high morbidity/mortality (e.g. meningitis). Apparently, clinicians create more dichotomous risk estimations (high risk or low risk) for the management of specific serious infections with reassessment of risk estimates after every diagnostic step. Clinicians used a stepwise approach in the management of febrile children, rather than considering one risk thresholds for SBI in general. We observed agreement in predictive value of all tested clinical predictor variables in the detection of children with SBI, for both clinical-based as prediction model-based judgement. Clinicians were guided by ill appearance and aberrant circulatory signs in their febrile child evaluation, which were not the most influencing factors according to the Feverkidstool. For the Feverkidstool respiratory predictors as chestwall retractions and oxygen saturation were more powerful influencing factors. Furthermore, we found that CRP levels influenced clinical risk scores differently in children with or without initial SBI management, with higher influence of clinical factors than of CRP value. In our study population, this approach was not enhanced by experiences of errors in the past. These insights in influencing factors in the clinical prediction of febrile children at risk for SBI helps us to understand, review and evaluate clinical management decisions.

r influence of clinical factors than of CRP value. In our study population, this approach was not enhanced by experiences of errors in the past. These insights in influencing factors in the clinical prediction of febrile children at risk for SBI helps us to understand, review and evaluate clinical management decisions. Compared to prediction model based risk scores, thresholds of children who were not managed as having a SBI were more comparable, ranging from 7 to 16 %. We might have to conclude that this risk threshold is justified as SBI rule-out threshold, but no agreement can be defined on rule-in thresholds as there appears too much difference between prediction model and the clinical stepwise risk assessment in children managed as SBI. Strengths and limitations The main strength of this study is the use of real-life videos instead of paper-case patients. This approach is a more representative way of portraying real life, and there is an evolving evidence base on the use of patient video cases as educational interventions [8, 23].

Compared to prediction model based risk scores, thresholds of children who were not managed as having a SBI were more comparable, ranging from 7 to 16 %. We might have to conclude that this risk threshold is justified as SBI rule-out threshold, but no agreement can be defined on rule-in thresholds as there appears too much difference between prediction model and the clinical stepwise risk assessment in children managed as SBI. Strengths and limitations The main strength of this study is the use of real-life videos instead of paper-case patients. This approach is a more representative way of portraying real life, and there is an evolving evidence base on the use of patient video cases as educational interventions [8, 23]. A second strength of the study is the use of the Feverkidstool as an arithmetic model to compare the subjective overall assessment of the clinician when evaluating the febrile child. In a review describing vignette studies on medical decision behaviour, it was concluded that most studies on this topic did not compare their results to some sort of normative benchmark [3]. Moreover, the role of prediction models becomes greater, as clinicians may increasingly rely on alarming signs and symptoms described in (inter)national clinical guidelines and prediction models due to decreasing incidence of SBI. Although, there was a discrepancy in risk assessment of some video vignettes (e.g. vignettes 7, 11 and 18), probably due to the absence of variables as decreased consciousness or agitation in the Feverkidstool.

ms described in (inter)national clinical guidelines and prediction models due to decreasing incidence of SBI. Although, there was a discrepancy in risk assessment of some video vignettes (e.g. vignettes 7, 11 and 18), probably due to the absence of variables as decreased consciousness or agitation in the Feverkidstool. There are some other limitations in this study. Videos still lack some aspects of real life such as observation time or concise descriptions of patients’ history. However, from literature, we know that more detailed case descriptions will be assigned a higher subjective probability of disease than a brief abstract of the same case, even if they contain the same disease information [11]. Another limitation includes the determination of some clinical variables by the clinicians’ judgement (ill appearance, chestwall retractions and capillary refill time). In this way, misclassification of these clinical predictors could have occurred. However, this approach does reflect clinical practice and therefor may just strengthen generalisability of our results.

some clinical variables by the clinicians’ judgement (ill appearance, chestwall retractions and capillary refill time). In this way, misclassification of these clinical predictors could have occurred. However, this approach does reflect clinical practice and therefor may just strengthen generalisability of our results. Next, the DCE analysis had to be performed within the availability of a limited number of video vignettes. As a consequence, we were forced to exclude or merge some predictor variables (e.g. oxygen saturation and tachypnoea) to meet the DCE theory design. Second, although a response rate of 50 % for clinicians was similar to other DCE studies, this response rate is not optimal [2, 9, 29]. However, due to the experienced background of all participants, we assume limited answer variability resulting in representative study results. Conclusion In this study on real-life video vignettes, we observed high risk scores in clinicians’ risk estimation of SBI management in febrile children, and these risks are mostly influenced by the clinical characteristics ill appearance and aberrant circulatory signs. Uniform risk thresholds at which one should start SBI management in febrile children remains unclear, as the concept of clinicians’ dichotomous risk thresholds was hardly comparable to the overall SBI risk assessment of the prediction model. However, more consistent results were found for clinical and prediction model-based risk thresholds at which we refrain from SBI management in the febrile child visiting the emergency department. Abbreviations CIconfidence interval

Conclusion In this study on real-life video vignettes, we observed high risk scores in clinicians’ risk estimation of SBI management in febrile children, and these risks are mostly influenced by the clinical characteristics ill appearance and aberrant circulatory signs. Uniform risk thresholds at which one should start SBI management in febrile children remains unclear, as the concept of clinicians’ dichotomous risk thresholds was hardly comparable to the overall SBI risk assessment of the prediction model. However, more consistent results were found for clinical and prediction model-based risk thresholds at which we refrain from SBI management in the febrile child visiting the emergency department. Abbreviations CIconfidence interval CRPC-reactive protein EDemergency department SBIserious bacterial infections VASvisual analogue scale We gratefully acknowledge all participants of the video vignettes study for their time, patience and complete participation in our study. We want to thank Paul Muston for the collaboration and the development of the video vignettes for practical use. We acknowledge Johan van der Lei, Ewout Steyerberg and Yvonne Vergouwe for discussion on the approach towards DCE analyses.

o vignettes study for their time, patience and complete participation in our study. We want to thank Paul Muston for the collaboration and the development of the video vignettes for practical use. We acknowledge Johan van der Lei, Ewout Steyerberg and Yvonne Vergouwe for discussion on the approach towards DCE analyses. Authors’contributions EdVK, DR, ML and HAM substantially contributed to the conception and design of the study. DR collected the original video vignettes as used in the study. He monitored participant response rates and undertook data extraction. EdVK actively enrolled study participants and monitored response rates. She undertook data extraction and performed data analysis. She drafted the initial manuscript. EdBG was responsible for the DCE analysis and interpretation. RO and HAM participated and supervised analysis and interpretation of the data. All authors reviewed and revised the manuscript and approved the final manuscript as submitted. Compliance with ethical standards Conflict of interest The authors declare that they have no competing interests. Funding source EK is supported by ZonMW, a Dutch organisation for health research and development. The study sponsor had no role in study design, in the collection, analysis, and interpretation of data; in the writing of the report; nor in the decision to submit the paper for publication.

Compliance with ethical standards Conflict of interest The authors declare that they have no competing interests. Funding source EK is supported by ZonMW, a Dutch organisation for health research and development. The study sponsor had no role in study design, in the collection, analysis, and interpretation of data; in the writing of the report; nor in the decision to submit the paper for publication. Financial disclosure Dr. Damian Roland is the co-director of QuackApps, a mobile applications company which designed the online risk assessment system. No payments were made for the delivery of the video vignettes. The other authors have no financial disclosures relevant to this article.

Considerations “Science is neither dogmatic nor democratic!” Heart failure is the final stage of a wide variety of cardiac diseases. Symptoms of HF develop, once the heart becomes unable to meet the metabolic demands of the body. In adults, heart failure is frequently called the new epidemic of the twenty-first century [36]. Causes of heart failure in childhood are associated to congenital heart diseases, cardiomyopathies, and arrhythmias as well as acquired heart and circulatory diseases [21]. Epidemiological studies covering the pediatric age group are very rare. More than 20 years ago, Rodeheffer et al. [46] reported an incidence of four infants <1 year of age in 1000 person years and a prevalence in children <10 years of age of 1.3 in 1000. Hsu and Pearson summarized the various causes of pediatric heart failure [21] as well as the current treatment and future directions [22].

In adults, heart failure is frequently called the new epidemic of the twenty-first century [36]. Causes of heart failure in childhood are associated to congenital heart diseases, cardiomyopathies, and arrhythmias as well as acquired heart and circulatory diseases [21]. Epidemiological studies covering the pediatric age group are very rare. More than 20 years ago, Rodeheffer et al. [46] reported an incidence of four infants <1 year of age in 1000 person years and a prevalence in children <10 years of age of 1.3 in 1000. Hsu and Pearson summarized the various causes of pediatric heart failure [21] as well as the current treatment and future directions [22]. Considering that cardiac output (flow) is based on heart rate, myocardial contractility, preload, afterload, synchrony, and ventricular-ventricular interaction (VVI), a low cardiac output might be associated with heart failure due to one or all of these variables (Fig. 1).Fig. 1 Parameters responsible for cardiac output. In addition to heart rate, contractility, preload, and afterload, cardiac synchrony as well as ventricular-ventricular interaction (VVI) defines cardiac function. CaO2 oxygen content, Ca-vDO2 arterial-venous oxygen content difference, CO cardiac output, DO2 oxygen delivery, HR heart rate, SaO2 arterial oxygen saturation, SvO2 venous oxygen saturation in superior and inferior caval vein, SV stroke volume, VO2 oxygen consumption

tricular interaction (VVI) defines cardiac function. CaO2 oxygen content, Ca-vDO2 arterial-venous oxygen content difference, CO cardiac output, DO2 oxygen delivery, HR heart rate, SaO2 arterial oxygen saturation, SvO2 venous oxygen saturation in superior and inferior caval vein, SV stroke volume, VO2 oxygen consumption If there is no successful strategy to remove the root cause, the aims of chronic congestive heart failure (CHF) therapy are to modify the neuroendocrine responses that worsen CHF and their pathophysiological consequences and to stimulate endogenous repair mechanisms. Established therapies in adults aim at reducing preload, afterload, and neuro-humoral activation and to halt the ongoing loss of cardiomyocytes, which gives way to replacement fibrosis [7]. In Table 1, therapeutic goals for chronic HF treatment in children are summarized.Table 1 Therapeutic goals for chronic HF treatment in children 1. Preload optimization by avoiding intravascular, in particular intra-arterial, volume depletion 2. Reduction of the systemic vascular resistance without jeopardizing the coronary perfusion pressure 3. Optimizing myocardial oxygen consumption and re-establishing myocardial synchrony as well as VVI 4. Allowing time to establish endogenous and exogenous repair mechanisms

n particular intra-arterial, volume depletion 2. Reduction of the systemic vascular resistance without jeopardizing the coronary perfusion pressure 3. Optimizing myocardial oxygen consumption and re-establishing myocardial synchrony as well as VVI 4. Allowing time to establish endogenous and exogenous repair mechanisms In this context, tachy- and brady-arrhythmias need to be prevented and if present effectively treated; sinus rhythm heart rate needs to be adjusted to the lowest effective level in order to reduce myocardial oxygen demand and to optimize the diastolic ventricular filling time. Chronic stimulation of a dysfunctional myocardium is counterproductive; all exogenous therapeutic strategies that stimulate the neuro-humoral system have been repeatedly analyzed and ought to be, whenever possible, omitted (for example: chronic treatment with loop-diuretics); if inotropic agents and vasoconstrictors become necessary, they should be used as short as possible or as a bridge towards heart transplant, if cardiac assist devices are not the better option.

ave been repeatedly analyzed and ought to be, whenever possible, omitted (for example: chronic treatment with loop-diuretics); if inotropic agents and vasoconstrictors become necessary, they should be used as short as possible or as a bridge towards heart transplant, if cardiac assist devices are not the better option. Considering the balance of oxygen delivery and consumption, positive inotrope vasodilators (milrinone, levosimendane) ought to be preferentially used as long as myocardial perfusion pressures are not compromised. During catecholamine infusion therapy, strategies designed to protect the myocardium should be considered, for example: a combination of epinephrine or norepinephrine infusion with ß1-receptor blockers. In decompensated systolic heart failure, which often occurs in infants and children with dilated cardiomyopathy, short-term epinephrine infusion combined with intravenous or oral ß1-receptor blocking agents (metoprolol, bisoprolol) is not a contradiction, but an actually used and recommended strategy [43]. In addition, it has to emphasize that the right and left heart does not act in isolation; cardiac re-synchronization [35] as well as strategies to improve the ventricular-ventricular interaction (VVI) is desirable in order to allow endogenous cardiac repair, in particular in younger patients [53, 61].

ed strategy [43]. In addition, it has to emphasize that the right and left heart does not act in isolation; cardiac re-synchronization [35] as well as strategies to improve the ventricular-ventricular interaction (VVI) is desirable in order to allow endogenous cardiac repair, in particular in younger patients [53, 61]. Bridging to heart transplantation (HTX), with or without the use of assist devices, or transiting the patient from an acute to a chronic heart failure status might be the therapeutically strategy of choice, if no satisfactory return to normal function can be achieved [53]. Differences of chronic HF treatment in children and adults Age- and perhaps mechanism-independent chronic HF is associated with neuro-humoral activation and increased levels of circulating neuro-hormones (noradrenaline, adrenaline, renin, angiotensin II, aldosterone, vasopressin), which lead to vasoconstriction, sodium, and water retention. Continuous endogenous neuro-humoral activation leads to myocardiocyte apoptosis, necrosis, and cardiac fibrosis, the main causes of chamber dilatation and progressive dysfunction, culminating in a vicious cycle of ever worse quality of myocardium and heart function [51].

o vasoconstriction, sodium, and water retention. Continuous endogenous neuro-humoral activation leads to myocardiocyte apoptosis, necrosis, and cardiac fibrosis, the main causes of chamber dilatation and progressive dysfunction, culminating in a vicious cycle of ever worse quality of myocardium and heart function [51]. Medical treatment recommendations for chronic HF in adults have been based on controlled, randomized studies [29]. Large cohort studies were necessary to pinpoint that among a group of HF treatment drugs as ß-adrenergic receptor blockers [2, 3, 40], angiotensin-converting enzyme (ACE) inhibitors [1] and aldosterone receptor antagonists [41] significantly reduce the mortality by counteracting the neuro-humoral overdrive of chronic HF. Such large cohort clinical trials cannot realistically be conducted in children with chronic heart failure due to the small patient numbers and the heterogeneity of the HF causes. Clinical trials in the pediatric age group are generally underpowered and cannot detect significant impact differences on survival rates. Several review articles address the differences in responses to medications in relation to the age of patients, pharmacokinetic/pharmacodynamic characteristics, and underlying causes of CHF and their molecular characteristics [48, 51]; however, the overall strategy to block the neuro-humoral axis is not disputed. Rossano and Shaddy [48] pointed to the missing data in children and emphasized that extrapolating evidence from adult patients to children with heart failure may have limited utility. However, given the state of our current situation of pediatric heart failure therapy, this statement lacks a forward-looking attitude and encourages a not-justified therapeutic “nihilism” when it comes to chronic HF treatment in children. Unfortunately, this entirely uncreative state of the affairs has been cultivated over decades [49].

state of our current situation of pediatric heart failure therapy, this statement lacks a forward-looking attitude and encourages a not-justified therapeutic “nihilism” when it comes to chronic HF treatment in children. Unfortunately, this entirely uncreative state of the affairs has been cultivated over decades [49]. In adults with chronic HF, new therapeutic strategies like angiotensin-neprilysin inhibition have been lauded as a paradigm shift in HF therapy [30], while in children, the use of ACE inhibitors, ß-adrenergic receptor blockers, and mineralocorticoid-receptor blockers still remains controversial and—if recommended—is rarely used [60]. This then begs the question, why effective therapy is being withheld in infants and children despite well-designed and sufficiently powered randomized trials in adults, which have been published more than a decade ago? In fact, drugs shown to be highly effective in adult heart failure patients have been discredited because of their use in studies with an unfortunate design, despite or because of their administration in controlled, randomized, double-blind trials which are acknowledged as the gold standard in evidence-based medicine. In one single controlled randomized study, the ACE inhibitor enalapril was judged to be not effective for the treatment of univentricular heart failure [23]; this study has been frequently cited to support the notion that ACE inhibitors are ill-advised in the treatment of all children with HF. ß-Adrenergic receptor (BAR) blockers have also been labeled to be of no benefit for children with chronic HF; the potential use of highly selective or non-specific BAR blocker has not been considered. In one, highly cited BAR-blocker study in children, the effect of carvedilol was assessed in 106 patients, while placebo was administered to 54 children with functional classes II and III [56]. This study, performed in 26 North American pediatric heart centers, was not stratified according to the causes and severity of HF and terminated after 5 years; in addition, there was no control for co-medications; the dosage of diuretic drugs and whether the BAR-blocker dose was adequate had not been or monitored by simply observing the heart rate response. Remarkably, during the same time period in US pediatric heart centers, BAR blockers were routinely used only in 4 % of children with cHF [60]. This fact illustrates what little clinical experience there is with BAR-blocker therapy in children with cHF.

not been or monitored by simply observing the heart rate response. Remarkably, during the same time period in US pediatric heart centers, BAR blockers were routinely used only in 4 % of children with cHF [60]. This fact illustrates what little clinical experience there is with BAR-blocker therapy in children with cHF. The third thrust of HF therapy should decrease the endogenous neuro-hormonal response by mineralocorticoid antagonism. In the RALES trial [41], treatment with spironolactone, in addition to conventional therapy, led to a relative reduction of the risk of death by 30 % and of cHF-related hospitalizations by 35 % in adults. In pediatric patients, spironolactone is a common component of diuretic regimens due to its potassium-sparing property but seldom used as a tissue aldosterone antagonist to influence cardiac re-remodeling, in particular myocardial fibrosis. The Canadian guideline reports that therapy with drugs that block the effects of aldosterone is well established in adults with systolic HF; “data regarding the role of spironolactone or related agents in the treatment of children with cHF are very limited” [25]. However, the few pediatric data published by Masutani et al.—with a focus on the importance of aldosterone-blocking drugs like spironolactone in children with HF, and in particular with preserved ejection fraction (HFpEF), have been widely ignored [28]. On the other hand, since 1978, Engle et al. [16] had administered 137 courses of furosemide to 106 hospitalized pediatric patients with salt and water retention associated with cardiac or renal disease; furosemide is by far the most commonly used and recommended drug, not only for acute but also for chronic heart failure in children. The diuretics are recommended as effective and safe in the pediatric age group when administered acutely as a parenteral medication and over a long-term course by the oral route in the doses and at the time intervals used in this study [16]. Yet, despite a lack of evidence-based studies, the use of diuretics is supported by Cochrane’s systematic review [18] as sufficient for the routine use in children with chronic CHF. While there is a recognized lack of evidence with regard to the use of ACE inhibitors, BAR blockers, and mineralocorticoid antagonists in pediatric patients, the chronic use of diuretics has never been scrutinized by pediatric opinion leaders [25].

ew [18] as sufficient for the routine use in children with chronic CHF. While there is a recognized lack of evidence with regard to the use of ACE inhibitors, BAR blockers, and mineralocorticoid antagonists in pediatric patients, the chronic use of diuretics has never been scrutinized by pediatric opinion leaders [25]. A similar situation prevails with regard to digoxin treatment of chronic heart failure in infants and children. Again, a randomized, placebo controlled study is only available for adult patients with ischemic or cardiomyopathic CHF. Digoxin had shown weak, but positive effects on left ventricular ejection fraction (EF), exercise capacity, quality of life, and reduction of hospitalizations, but the survival rate in the whole cohort was not significantly improved [14]. In the pediatric literature, we find several early clinical observational studies, which demonstrated a beneficial effect on CHF, including improved contractility and a decreased neuro-hormonal stimulation. Perhaps the Na/K-ATPase inhibitor digoxin, as a positive inotropic drug might be indicated in younger patients. On the other hand, a recently published study of 48 infants with chronic HF secondary to left-to-right shunt lesions who were randomized to treatment with enalapril and furosemide ±, digoxin did not find any clinical improvement. It was concluded that digoxin does not provide any extra benefit in the treatment of such patients [15]. In considering the results of this latter study, the Canadian guidelines for pediatric chronic HF did not recommend digoxin in children with chronic HF [25]. However, digoxin is a weak vasoconstrictor and co-therapy with a neuro-hormonal axis-stimulating diuretic is from our point of view not indicated for HF due to a left-right shunt because the shunt flow is favored.

e Canadian guidelines for pediatric chronic HF did not recommend digoxin in children with chronic HF [25]. However, digoxin is a weak vasoconstrictor and co-therapy with a neuro-hormonal axis-stimulating diuretic is from our point of view not indicated for HF due to a left-right shunt because the shunt flow is favored. In conclusion: HF drugs need a clearly defined indication and a convincing rationale. Simply, a randomized, placebo controlled study design—which is highly desirable—however, lacking a crisp definition of the clinical target and the underlying disease mechanism does not allow us to generate the needed outcome information. One wonders also why the effectiveness of a chronic HF drug is not reported, as it is customary for the drugs targeting pulmonary arterial hypertension [29]; i.e., if the drug does not significantly alter the mortality, then surrogate endpoints which demonstrate improvement of the clinical status should be obtained to support their use. PH-targeting drugs are not being withheld if they do not affect mortality but improve the clinical functional class [26, 61].

tension [29]; i.e., if the drug does not significantly alter the mortality, then surrogate endpoints which demonstrate improvement of the clinical status should be obtained to support their use. PH-targeting drugs are not being withheld if they do not affect mortality but improve the clinical functional class [26, 61]. Recommendations to improve chronic HF treatment in infants and children Systolic dysfunction related to left ventricular dilated cardiomyopathy The incidence of childhood cardiomyopathy is in the range of 1.13–1.24 per 100,000 children according to the data collected by two population-based registries, where dilated cardiomyopathy accounts for more than half of the primary causes [6, 27]. Dilated cardiomyopathy (DCM) is a serious disease. The 1-year and 5-year rates of death or transplantation were first described to be 35 and 49 %, respectively, in a population-based study in 1998 and 31 and 46 % in the largest population study in 2006 [39, 56]. Recently, we described in a single-center study a lower rate for death of 18 % and transplantation 31 % [50], which is perhaps the result of an improved treatment strategy (Fig. 2). While the overall goal of the treatment of children with DCM is to avoid death or delay heart transplantation, different survival rates after the diagnosis of DCM may be attributed to different treatment strategies. Prior to treatment, a detailed diagnostic workup is needed for left ventricular (LV)-DCM: this includes the medical history, clinical examination, echocardiogram, X-ray, MRI, coronary-angiography, and myocardial biopsy, as well as laboratory data (BNP or NT-proBNP, CRP, hemoglobin, sodium, potassium, creatinin, albumin, and if possible, plasma levels of aldosterone, norepinephrine, angiotensin, renin, respectively). Treatment success is monitored by the clinical functional (NYHA, Ross I–IV) and nutritional status, heart and respiratory rate, intermittent diastolic and systolic blood pressure, and SaO2 measurements. Follow-up laboratory and imaging data should be acquired.Fig. 2 Estimated survival and freedom of death in children with DCM [50]. Shown is the median follow-up 16 months (range 2–80 months) of 38 children in an age less than 3 years admitted at the Pediatric Heart Center Giessen. The Kaplan-Meier survival curve after the diagnosis of dilated cardiomyopathy revealed a 1-year survival of 97 % and a 5-year survival of 86 % [50]

ildren with DCM [50]. Shown is the median follow-up 16 months (range 2–80 months) of 38 children in an age less than 3 years admitted at the Pediatric Heart Center Giessen. The Kaplan-Meier survival curve after the diagnosis of dilated cardiomyopathy revealed a 1-year survival of 97 % and a 5-year survival of 86 % [50] Taking into account-specific exclusion criteria, at our institution, the chronic, age-independent heart failure therapy consists of specific long-acting ß1-adrenoreceptor blocker (bisoprolol), long-acting tissue angiotensin-converting enzyme inhibitor (lisinopril), and mineralocorticoid-receptor blocker (spironolactone) and applying a goal-oriented drug dosage [43]. Digoxin is the fourth-line HF drug with a target of a plasma level of 0.5–0.9 ng/ml. Chronic treatments with loop-diuretics are avoided, and hydrochloro-thiazide in low dosages of 0.5–1 mg/kg are applied once or twice per day, if really needed. If there are signs of inappropriate ADH secretion with severe hyponatremia, we currently use the V2-receptor antagonist, tolvaptan, once per day in a dosage of 0.1–(0.3) mg/kg, targeting a sodium serum level of 140–145 mmol/l. The hypothesized treatment goals for treatment of LV-DCM with a highly specific ß1-BAR blocker, tissue ACE inhibitor, and aldosterone-antagonist are summarized in Table 2.Table 2 Summarizes the hypothetical aim of therapy and the used heart failure drugs

.1–(0.3) mg/kg, targeting a sodium serum level of 140–145 mmol/l. The hypothesized treatment goals for treatment of LV-DCM with a highly specific ß1-BAR blocker, tissue ACE inhibitor, and aldosterone-antagonist are summarized in Table 2.Table 2 Summarizes the hypothetical aim of therapy and the used heart failure drugs 1. Heart rate control in order to improve the ratio of (myocardial) oxygen consumption to demand and prolong the time for diastolic ventricular filling (ß1-specific ß-blocker) 2. Diminishing apoptosis and myocytes necrosis (ß1-specific ß-blocker) 3. Diminishing interstitial fibrosis by blocking sympathetic- and RAA-S (three first-line drugs) 4. Reduction of cardiac afterload together with preservation of the coronary perfusion pressure by adequate (preload) intravascular, in particular, arterial vascular filling (avoidance of diuretics, effectively dosed ACE-I + ß1-specific blocker, preserving the beta 2 receptor function) 5. Basis for additional strategies to re-establish ventricular synchrony and re-establish VVI as prerequisite for cardiac regeneration (all three first-line drugs) 6. Low risk-benefit ratio and high parental compliance by daily single-dose therapy together with easy dosing of 0.1–0.2 (0.3) mg/kg × day for both bisoprolol and lisinopril and once application of spironolactone 1–2 mg/kg × day, respectively

ish VVI as prerequisite for cardiac regeneration (all three first-line drugs) 6. Low risk-benefit ratio and high parental compliance by daily single-dose therapy together with easy dosing of 0.1–0.2 (0.3) mg/kg × day for both bisoprolol and lisinopril and once application of spironolactone 1–2 mg/kg × day, respectively Considering the (patho-)physiology of the ß-adrenergic receptors (ß-AR) in pediatric DCM, highly selective ß1-AR blockers are recommended in children to treat chronic HF caused by LV-DCM [34, 43, 50]. It is generally accepted that chronic stimulation of the cardiac β1-adrenergic system is toxic to the heart and contributes to the pathogenesis of congestive heart failure [11]. We could show that children that underwent open heart surgery with cardiac arrest demonstrated a decreased ß-adrenoceptor-mediated adenylate cyclase activation in a manner compatible with an uncoupling of ß-adrenoceptors from the Gs-protein-adenylate cyclase complex [52]. Cardiac arrest and HF liberate myocardially stored norepinephrine [47] and cause ß1-receptor desensitization. As stated, chronic stimulation of β1-receptors is cardiotoxic while β2-receptor stimulation might be cardio-protective [8, 11]. The elegant study of Miyamoto et al. [33, 34] not only demonstrated the differences of BAR pathophysiology between pediatric and adult patients with idiopathic dilated cardiomyopathy but also recommended the use of ß1-selective BAR blocker in children. The data by Miyamoto et al. might also explain the failure of non-specific ß-blocking therapy in previous clinical pediatric HF trials. Miyamoto et al. showed that ß1-ARs are downregulated in both adults and children with chronic HF but that the ß2-AR is downregulated only in pediatric DCM. She stated “Further inhibition of the already downregulated ß2-ARs may override the benefit of ß1-AR inhibitor therapy, because preservation of ß2-AR function is beneficial.” In addition, several studies support the cardio-protective effect of short-term beta-2 stimulation [4, 5, 9, 17, 31, 32, 38, 45, 63]. Bisoprolol is a long-acting, highly cardio-selective BAR-blocker, with a low side effect profile. Bisoprolol has a beta1/beta2-specific binding ratio of about 125, metoprolol of about 80, and propranolol and carvedilol of about 5 [63]. Dual mechanism of action includes selective beta1-receptor blockade and stimulation of endothelial NO production [62].

y cardio-selective BAR-blocker, with a low side effect profile. Bisoprolol has a beta1/beta2-specific binding ratio of about 125, metoprolol of about 80, and propranolol and carvedilol of about 5 [63]. Dual mechanism of action includes selective beta1-receptor blockade and stimulation of endothelial NO production [62]. Additionally, ß1-specific BAR-blockers can block renal ß1-adrenergic receptors and concomitant renin release caused by HF or in association with ACE inhibitor therapy [10]. Combined with ß1-adrenoreceptor blockers, long-acting tissue angiotensin-converting enzyme inhibitors [57] like lisinopril or ramipril are our age-independent HF drugs of choice, based on the hypothesis that stimulated RAAS can be blocked by both drugs acting synergistically to lower systemic vascular resistance and avoid myocardial fibrosis, the latter together with the mineralocorticoid-receptor blocker spironolactone.

7] like lisinopril or ramipril are our age-independent HF drugs of choice, based on the hypothesis that stimulated RAAS can be blocked by both drugs acting synergistically to lower systemic vascular resistance and avoid myocardial fibrosis, the latter together with the mineralocorticoid-receptor blocker spironolactone. In order to improve systemic blood flow, these drugs have not only been recommended in HF patients with stable hemodynamics [25], but also to transit unstable patients to a stable hemodynamic status [43, 50, 53]. More recently, we are treating all LV-DCM patients with the triple drug combination in preparation for reversible pulmonary artery banding [43, 53]. Only patients with volume depletion due to overdiuresis are excluded, but ongoing inotropic support is not a contraindication for additional ß1-adrenoreceptor blockade, which may facilitate successful weaning of inotropes. Clinical parameters and biomarkers monitor treatment success (Fig. 3). Treatment compliance together with very low side effects, can be impressive, as we could observe and document prior to and immediately after reversible pulmonary arterial banding (rPAB) as a novel therapy to treat LV-DCM [43, 53].Fig. 3 Shown is the mean brain natriuretic peptide (BNP) value of 20 infants and children younger than 3 years with left ventricular dilative cardiomyopathy (LV-DCM) and preserved right ventricular ejection fraction (pRV-EF), which were admitted for heart transplantation. The extremely high BNP values at admission decreased significantly [we believe as a consequence of the administered ‘triple therapy’] during the period prior to surgical pulmonary banding (PAB). At admission, all patients had been treated with in some high dosages of furosemide independent of continuously administered inotropic treatment. Furosemide was stopped and bisoprolol (B), lisionopril (L), and spironolactone (S) were started. The goal was to achieve a resting heart rate (HR) of less than 120/min with an adequate systemic blood pressure to sufficient diuresis. The inotropic treatment was continued, but if dobutamine had been part of the pre-admission regiment, it was changed to the inodilator milrinone

), and spironolactone (S) were started. The goal was to achieve a resting heart rate (HR) of less than 120/min with an adequate systemic blood pressure to sufficient diuresis. The inotropic treatment was continued, but if dobutamine had been part of the pre-admission regiment, it was changed to the inodilator milrinone Additional questions arise: why is captopril typically recommended as the first choice for HF treatment in infants and enalapril an appropriate choice for those older than the age of 2 years [24]? In infants and young children, bronchoconstriction and bronchiolitis are common and often associated with HF. Thus, it is our view that non-specific BAR blockers with a ß2-blocking effect and less-specific ACE-I with a higher risk of bradykinin-dependent cough and bronchoconstriction should be avoided. High renin and creatinin levels together with hyponatremia are classical signs not only of severe HF but also of inadequate diuretic therapy. To lower the risk of a blood pressure drop after initiating vasodilator therapy, there should not be a dosage reduction or withdrawal of ACE-I therapy, but rather implementation of diuretics [19, 20, 37, 44, 59].

gether with hyponatremia are classical signs not only of severe HF but also of inadequate diuretic therapy. To lower the risk of a blood pressure drop after initiating vasodilator therapy, there should not be a dosage reduction or withdrawal of ACE-I therapy, but rather implementation of diuretics [19, 20, 37, 44, 59]. Hyperdynamic HF related to a ventricular left-to-right shunt Surgical or transcatheter closure of a hemodynamically relevant ventricle septum defect (VSD) is the treatment of choice. However, there may be several reasons why such a curative approach cannot be performed or need to be delayed. Diuretics, digoxin, and fluid restriction are usually recommended to treat chronic HF based on a left-right shunt, although there is no data that document efficacy [12, 15, 42, 55]. This above management strategy is ill conceived because diuretics, fluid restriction, and digoxin do not achieve the goal of increasing systemic blood flow but instead favor the imbalance of pulmonary and systemic blood flow [12, 13]. Diuretic- and fluid restriction-dependent neuro-humoral stimulation and digoxin-caused vasoconstriction increase the left-right shunt; in mid-term, such treatment might be associated by cardiac cachexia (Fig. 4). And as a reminder, in congenital heart defects with left-right shunt, the systolic function of the left ventricle is mostly preserved [13], whereas diastolic function may be impaired. Older patients with a left-right shunt develop commonly diastolic dysfunction caused by the compensatory mechanism of left ventricular volume depletion due to sympathetic and RAA stimulation. Again, based on our own institutional experience of more than a decade, bisoprolol without blocking the systemic vasodilation favoring ß2-adrenoreceptor, highly specific tissue ACE-I, like lisinopril and spironolactone, if diastolic dysfunction with preserved ejection fraction are present, should be and are used to influence the balance of left-to-right shunt in order to favor systemic blood flow. The improvement of chronic HF symptoms can be easily monitored by the decrease in heart and respiratory rates, signs of less severe pulmonary congestion, and improved food intake in infants. This therapeutic concept avoids cardiac cachexia, which was usually observed in almost all patients in the past and is still observable today.Fig.

f chronic HF symptoms can be easily monitored by the decrease in heart and respiratory rates, signs of less severe pulmonary congestion, and improved food intake in infants. This therapeutic concept avoids cardiac cachexia, which was usually observed in almost all patients in the past and is still observable today.Fig. 4 Represents L/R shunt of hemodynamically relevant VSD; infants and young children not early surgically corrected, but mid-term treated by diuretics and fluid restriction develop as pictured severe cachexia because L/R shunt is favored by increased systemic vascular resistance and high oxygen consumption by compensating increased heart and breath rate and in particular of malnutrition

young children not early surgically corrected, but mid-term treated by diuretics and fluid restriction develop as pictured severe cachexia because L/R shunt is favored by increased systemic vascular resistance and high oxygen consumption by compensating increased heart and breath rate and in particular of malnutrition HF prophylaxis in hypoplastic left heart syndrome after hybrid stage I The right ventricle as the systemic heart chamber is usually found in congenital malformations like congenital corrected transposition of the great arteries (ccTGA) and also in the hypoplastic left heart syndrome (HLHS) and in the HLH-complex; in HLHC, the right ventricle works in parallel with an obstructed or borderline left ventricle. There is growing evidence that RV dysfunction develops in many of those patients and accounts for the considerable morbidity and mortality. Therefore, systemic RV function needs close surveillance and sufficient timing of an appropriate intervention to optimize outcome. Almost dogmatically, HF medications, which have been pronounced to be effective for the treatment of a failing left ventricle, are judged to be ineffective for treating a failing systemic right ventricle [24, 25]. However, we wish to point out that the subpulmonary right or left ventricle is metabolically differently active and this may have substantial implications for the pharmacotherapy. Genes encoding drug-metabolizing enzymes, like cytochrome P450 mono-oxygenases, are predominantly expressed in the subpulmonary heart chamber; this might explain the lack of efficacy of drugs like angiotensin-converting enzyme inhibitors and angiotensin receptor blockers on a subpulmonary RV or LV. An atrial switch of the venous connection to the right and left atria reverses the messenger RNA expression profiles. An anatomical left, but subpulmonary positioned, ventricle shows the expression of cytochrome P450 genes normally found in the subpulmonary RV. These facts highlight the importance of the subpulmonary ventricle and pulmonary circulation for the metabolic breakdown of drugs [58]. Thus, for the pharmacological response more important than the morphology of the ventricle appears to be its position. Such data are not available for univentricular hearts, in particular prior to Fontan completion.

tance of the subpulmonary ventricle and pulmonary circulation for the metabolic breakdown of drugs [58]. Thus, for the pharmacological response more important than the morphology of the ventricle appears to be its position. Such data are not available for univentricular hearts, in particular prior to Fontan completion. Considering HLHS, the right ventricle is responsible for the systemic and pulmonary circulation, but following the surgical completion of the Fontan circulation, the distally to the pulmonary circulation positioned RV supports the systemic circulation.

tance of the subpulmonary ventricle and pulmonary circulation for the metabolic breakdown of drugs [58]. Thus, for the pharmacological response more important than the morphology of the ventricle appears to be its position. Such data are not available for univentricular hearts, in particular prior to Fontan completion. Considering HLHS, the right ventricle is responsible for the systemic and pulmonary circulation, but following the surgical completion of the Fontan circulation, the distally to the pulmonary circulation positioned RV supports the systemic circulation. At our institution, neonates born with HLHS and some patients with HLHC are palliated using the Giessen Hybrid approach which consists of bilateral pulmonary banding, duct stenting, and, if necessary, atrial septum manipulation (Fig. 5). This approach has been established more than 17 years ago as an alternative to the Norwood stage I operation [54]. It is our goal to reduce inter-stage morbidity and mortality; one part of this inter-stage strategy is based on the use of dual or triple therapy of bisoprolol with lisinopril and spironolactone. The use of ß1-selective BAR blockers is further supported by recently published data from Miyamoto et al. [34], which stresses the altered BAR signaling in HLHS [34]. During the last 5 years, we have routinely avoided diuretics or digoxin when treating newborns or young infants discharged home after the hybrid procedure. We monitor heart and respiratory rate (Video 1) and, intermittently, systolic and diastolic blood pressure. The mean blood pressure should not be relied on simply because a diminished systemic blood flow might remain undetected. Echocardiographic and serum biomarkers (BNP, ProBNP) ought to be used for additional, intermittent monitoring. The blood pressure amplitude, together with the echocardiographically estimated degree of diastolic left-to-right shunt across the stented duct is important and supports our medical strategy (Fig. 5). The effect of the drug therapy on the systemic vascular resistance can be easily monitored. Low heart rate, prolonged diastolic filling time, and reduction of systemic vascular resistance (without increasing retrograde aortic blood flow and jeopardizing myocardial perfusion pressure) are the therapeutic goals. Excellent parental compliance with a single-dose regiment and the monitoring of body weight prior to hybrid stage II are usually obtained. Whether out-of-proportion myocardial hypertrophy of the systemic right ventricle or interstitial fibrosis can also be reduced by our treatment strategy remains to be examined in future studies.Fig.

al compliance with a single-dose regiment and the monitoring of body weight prior to hybrid stage II are usually obtained. Whether out-of-proportion myocardial hypertrophy of the systemic right ventricle or interstitial fibrosis can also be reduced by our treatment strategy remains to be examined in future studies.Fig. 5 Shows the schematic picture of Hybrid stage I consisting of duct stenting (DA stent), bilateral pulmonary banding (bPAB), and interatrial septum manipulation by stent placement. Left atrial (LAP) decompression to a pressure level of the right atrium (RAP) is an important part of a balanced parallel turned pulmonary to systemic circulation. The echocardiography shows an effective bPAB by its typical flow-pattern, but despite an effective bPAB, the systolic right-to-left flow through the duct is accompanied by a diastolic left-to-right reflow; one important indication to reduce the systemic vascular resistance without jeopardizing coronary blood flow. Additionally, bisoprolol reduces heart rate, which improves single ventricle filling; all factors together diminish pulmonary congestion, reduce total and myocardial oxygen consumption, and improve the baby’s functional class

ndication to reduce the systemic vascular resistance without jeopardizing coronary blood flow. Additionally, bisoprolol reduces heart rate, which improves single ventricle filling; all factors together diminish pulmonary congestion, reduce total and myocardial oxygen consumption, and improve the baby’s functional class Video 1 With permission of the mother shown here with her premature baby treated with the “Giessen Hybrid” method (see text), the video demonstrates how the parents are advised to monitor her babies that have a high risk of acute clinical deterioration. If the age-dependent respiratory rate at rest is normal the baby is stable; if the respiratory rate becomes fast, for example: double the normal rate, the parents have been advised to contact an experienced doctor or the responsible institution (MOV 2982 kb)

babies that have a high risk of acute clinical deterioration. If the age-dependent respiratory rate at rest is normal the baby is stable; if the respiratory rate becomes fast, for example: double the normal rate, the parents have been advised to contact an experienced doctor or the responsible institution (MOV 2982 kb) In conclusion Our clinical practice and institutional experience in chronic HF treatment in infants and children with left ventricular systolic heart failure, high-output failure due to significant left-right shunting congenital heart diseases, as well as HLHS and HLH-complex after hybrid stage I, are in clear contrast to the official guidelines for chronic HF therapy in infants and children. We acknowledge as a substantial weakness that we are missing randomized study results; however, in the absence of such trials, observational studies should receive credit as a first step. In addition, the here illustrated chronic HF therapy can lay the ground for a randomized multicenter studies designed to analyze the efficacy of other endogenous repair mechanisms supporting therapies as we advocated by pulmonary arterial banding in selected young patients with DCM and preserved right ventricular function. Electronic supplementary material Below is the link to the electronic supplementary material. Abbreviations ACE-IAngiotensin-converting enzyme inhibitor ARBAdrenergic receptor blocker BARß-Adrenergic receptor BNPBrain natriuretic peptide cHFChronic heart failure CHFCongestive heart failure DCMDilative cardiomyopathy HFpEFHeart failure with preserved ejection fraction

Introduction The advances in perinatal care have led to improved outcome in extreme prematurity. Gestational ages (GA) at which active treatment can be considered have decreased worldwide; however, a “gray zone” still remains [2, 6, 31, 35]. Therefore, in daily practice, several decisions have to be made about the initiation of care at 23–25 weeks GA. The key question is whether to initiate comfort care or active care. Factors that could be of influence are parental preferences and individual maternal or fetal characteristics. To support clinicians with decision-making in daily practice, several national guidelines on perinatal care are developed. Pignotti (2008), Gallagher (2014), and Guillen (2015) reviewed these guidelines and showed that 23 to 24 weeks of gestation are regarded as the gray zone of viability. In this gray zone, treatment decisions may be made using an individual approach and/or taking parental preferences into account. In some countries, this gray zone extends through to 25+6⁄7 weeks [10, 13, 28]. Not only guidelines differ, but also international, national, and local variations in actual practice do exist [4, 7, 12, 14, 17, 22, 23, 25, 30]. Furthermore, surveys amongst perinatal professionals revealed variation in (preferred) treatment decisions at the lower limits of viability, for example, decisions on performing a caesarian section (CS) and resuscitation [5, 9, 20, 24, 26, 32, 33].

iations in actual practice do exist [4, 7, 12, 14, 17, 22, 23, 25, 30]. Furthermore, surveys amongst perinatal professionals revealed variation in (preferred) treatment decisions at the lower limits of viability, for example, decisions on performing a caesarian section (CS) and resuscitation [5, 9, 20, 24, 26, 32, 33]. Accompanying the decision whether or not to initiate care, several other choices have to be made, such as transfer to a specialized hospital, antenatal administration of corticosteroids, monitoring of the fetus, delivery mode, presence of the neonatologist at birth, and the extent of potential resuscitation. Guidelines do not always cover all these aspects [10, 13, 28]. In 2010, the national Dutch guideline on perinatal practice in extremely premature delivery lowered the limit offering intensive care from 25+0/7 to 24+0/7 weeks GA [8]. Unlike in other countries, in the Netherlands, this lower limit is rather strict (in contrast to, i.e., the American AAP guideline advocating for an individualized approach) and, in general, intensive care will not be offered below 24 weeks GA [4, 8, 10, 13, 28]. The Dutch guideline states “informed consent of parents is prerequisite in the decision whether or not to initiate care at 24 weeks GA”. It indicates that prognostic factors (such as weight, gender) can be taken into account in decision-making in individual cases. However, because the prognostic value of these factors is unknown for the Dutch population, no specific recommendations are given. Furthermore, the Dutch guideline recommends transfer to a tertiary center from 23+4/7 weeks GA for counseling, administration of corticosteroids from 23+5/7 weeks GA, and fetal monitoring with a CS can be considered in case of suspected fetal distress from 24+0/7 GA in which the specific risks and benefits for current and future pregnancies and delivery need to be discussed.

tertiary center from 23+4/7 weeks GA for counseling, administration of corticosteroids from 23+5/7 weeks GA, and fetal monitoring with a CS can be considered in case of suspected fetal distress from 24+0/7 GA in which the specific risks and benefits for current and future pregnancies and delivery need to be discussed. Like in most guidelines, variable operationalization is imaginable. Disagreement between perinatal professional (individuals and/or groups) on treatment decisions in extremely preterm gestations could potentially lead to a conflict in perinatal care [5, 9, 10, 32]. Therefore, our primary goal was to investigate Dutch physicians’ preferences on decisions about treatment options for an extremely premature neonate against the background of this guideline. Our secondary goal was to study potential differences between neonatologists and obstetricians. Materials and methods Study design Cross-sectional, multicenter study (PreCo survey) using an online survey. Setting and study population This study, the PreCo survey, is part of the larger PreCo study, which evaluates Dutch care on different levels in (imminent) extreme preterm birth, e.g., prenatal counseling and treatment decisions (clinicaltrials.gov, NCT02782650 & NCT02782637). This PreCo study is supported and followed by both the national associations of neonatology and obstetrics as well as the patient association.

evaluates Dutch care on different levels in (imminent) extreme preterm birth, e.g., prenatal counseling and treatment decisions (clinicaltrials.gov, NCT02782650 & NCT02782637). This PreCo study is supported and followed by both the national associations of neonatology and obstetrics as well as the patient association. The Dutch care for extreme premature births is centralized in 10 level III centers for perinatal care which all gave informed consent to participate in the current study. Surveys were sent to all fellows and senior staff members in both obstetrics and neonatology. Data were collected from July 2012 through October 2013, approximately 2 to 3 years after the introduction of the new guideline on perinatal practice in extreme premature delivery in the Netherlands.

pate in the current study. Surveys were sent to all fellows and senior staff members in both obstetrics and neonatology. Data were collected from July 2012 through October 2013, approximately 2 to 3 years after the introduction of the new guideline on perinatal practice in extreme premature delivery in the Netherlands. Survey design and data collection The PreCo survey was developed in three stages. The first stage consisted of the development of a draft of the survey based on international literature about prenatal counseling; several prenatal counseling surveys that have kindly been shared with us [1, 3, 5, 18, 19, 27], observations from our previous study [11], and on public discussions generated by the Dutch guideline on perinatal practice in extreme premature delivery [8]. In the second stage, the survey was improved in two Delphi rounds; in the first round, the concept survey was extensively evaluated by four team members (two neonatologists, one obstetrician, and one pediatrician) and in the second round, independent perinatal experts (two neonatologists) pilot-tested the survey for clarity and content. In the third stage, the survey was adapted for both profession groups to exclude irrelevant questions and to optimize the participation rate. The PreCo survey required ~20 min to complete.

ician) and in the second round, independent perinatal experts (two neonatologists) pilot-tested the survey for clarity and content. In the third stage, the survey was adapted for both profession groups to exclude irrelevant questions and to optimize the participation rate. The PreCo survey required ~20 min to complete. We were particularly interested in physicians’ preferences on treatment decisions. Therefore, we designed three questions: the first asking is for recommendations on whether or not to initiate intensive care at several extreme preterm GAs, the second asking is the personal lower limits in GA for various interventions potentially associated with extreme prematurity, and the third is on the importance of associated factors in recommending active treatment or not. We used a fictitious case of an “uncomplicated” extreme premature delivery to examine all these preferences. Finally, the last section of the entire survey contained demographical items such as age and years of experience.

and the third is on the importance of associated factors in recommending active treatment or not. We used a fictitious case of an “uncomplicated” extreme premature delivery to examine all these preferences. Finally, the last section of the entire survey contained demographical items such as age and years of experience. An individual link to the online survey was sent to all participants. Three reminders were sent to non-responders. Survey results were anonymized before analysis. This study was waived by the local institutional review board.Characteristics of the fictitious case A consultation for prenatal counseling with an impending extreme premature delivery, singleton fetus, unremarkable history of pregnancy, average estimated fetal birth weight, unknown gender, no known congenital abnormalities, unremarkable social and medical history of parents, antenatal corticosteroids have been administered, and normal fetal heart rate registration. Data analysis Descriptive statistics were given as proportions of the respondents for that specific question. For comparison between obstetricians and neonatologists Chi-square (Ӽ2), Fisher exact, or Mann-Whitney U tests were used when applicable. Exact p values were provided, values <0.05 were considered significant. Statistical analyses were conducted using IBM SPSS Statistics (Version 20.0. Armonk, NY: IBM Corp).

question. For comparison between obstetricians and neonatologists Chi-square (Ӽ2), Fisher exact, or Mann-Whitney U tests were used when applicable. Exact p values were provided, values <0.05 were considered significant. Statistical analyses were conducted using IBM SPSS Statistics (Version 20.0. Armonk, NY: IBM Corp). Results We received 122 surveys from 205 eligible professionals; a response rate of 60 %. Each perinatal center was represented. Of those, 45 were from obstetricians and 77 from neonatologists. Of all 122 returned surveys, eight were only partially completed. Obstetricians had fewer years of experience than the neonatologists (Table 1). There was no influence of age or years of experience on the results, but some differences based on the institute of the participant did exist.Table 1 Characteristics of professionals Obstetricians (n = 84 sent) Neonatologists (n = 121 sent) Response rate 54 % 64 % Gender, % male 32 % 69 % Having children (parent) % of those: parent of premature child (<27 weeks) 91 % 83 % 0 % 2 % Median age in years (q25–75) 40 (38–47) 45 (37–50) Years of experience, median (q25–75) 5 (1–10)* 9 (4–17) *p 0.02 (MWU)

Results We received 122 surveys from 205 eligible professionals; a response rate of 60 %. Each perinatal center was represented. Of those, 45 were from obstetricians and 77 from neonatologists. Of all 122 returned surveys, eight were only partially completed. Obstetricians had fewer years of experience than the neonatologists (Table 1). There was no influence of age or years of experience on the results, but some differences based on the institute of the participant did exist.Table 1 Characteristics of professionals Obstetricians (n = 84 sent) Neonatologists (n = 121 sent) Response rate 54 % 64 % Gender, % male 32 % 69 % Having children (parent) % of those: parent of premature child (<27 weeks) 91 % 83 % 0 % 2 % Median age in years (q25–75) 40 (38–47) 45 (37–50) Years of experience, median (q25–75) 5 (1–10)* 9 (4–17) *p 0.02 (MWU) Professionals gave their preferred recommendations about whether or not to initiate intensive care at each week of gestation, ranging from providing comfort care through providing intensive care (Fig. 1). At 22+0–6/7 weeks GA, recommending comfort care was the only option. For a birth at 23+0–6/7 weeks GA, 82 % recommended comfort care only, but some professionals (16 %) also agreed with intensive care at parental request. At 24+0–6/7 and 25+0–6/7 weeks GA, the majority (54 and 64 %, respectively) recommended intensive care with the ability of comfort care at parental request; however, there was variation in the given preferences. At 26+0–6/7 and 27+0–6/7 weeks GA, the vast majority (89 and 96 %, respectively) recommended intensive care without the possibility of comfort care—however, a minority would agree with comfort care at parental request (11 and 4 %, respectively). No significant differences were found for any of the GAs between obstetricians and neonatologists.Fig. 1 Recommendations for a pregnant woman who is about to deliver a fetus of the following gestational ages neonatologist = * obstetrician = o (individuals) xx %/xx % = proportion of neonatologist/proportion of obstetricians

ignificant differences were found for any of the GAs between obstetricians and neonatologists.Fig. 1 Recommendations for a pregnant woman who is about to deliver a fetus of the following gestational ages neonatologist = * obstetrician = o (individuals) xx %/xx % = proportion of neonatologist/proportion of obstetricians Professionals were asked for their personal lower limits for certain interventions or decisions at extreme prematurity (Fig. 2). Answer options ranged from “starting at 22+0/7 weeks GA” through “starting at 26+0/7 weeks GA”; only at 23 weeks more detailed answer options were available (23+0/7, 23+4/7, and 23+5/7 weeks GA). There was variation between individuals up to 4 weeks. Medians in weeks GA (for this lower limit) were as mentioned below and interquartile ranges (IQR) are provided (obstetricians and neonatologists had the same median except where otherwise stated):Fig. 2 Personal limits of various interventions that could be taken around a possible premature delivery neonatologist = * obstetrician = o xx %/xx % = proportion of neonatologist/proportion of obstetricians Transfer pregnant woman with imminent premature delivery to tertiary hospital with NICU: 23+4/7IQRobstetricians23+0/7–23+5/7neonatologists23+4/7–23+5/7. Antenatal administration of corticosteroids: 23+.5/7IQRobstetricians23+4/7–23+5/7neonatologists23+5/7–23+5/7. Use of fetal monitor during delivery for monitoring of the fetal condition: 24+0/7IQRboth24+0/7–25+0/7. Perform a CS on fetal indication:25+0/7IQRobstetricians24+0/7–25+0/7neonatologists25+0/7–26+0/7.

Transfer pregnant woman with imminent premature delivery to tertiary hospital with NICU: 23+4/7IQRobstetricians23+0/7–23+5/7neonatologists23+4/7–23+5/7. Antenatal administration of corticosteroids: 23+.5/7IQRobstetricians23+4/7–23+5/7neonatologists23+5/7–23+5/7. Use of fetal monitor during delivery for monitoring of the fetal condition: 24+0/7IQRboth24+0/7–25+0/7. Perform a CS on fetal indication:25+0/7IQRobstetricians24+0/7–25+0/7neonatologists25+0/7–26+0/7. Neonatologists have to be present at the delivery: 24+0/7IQRboth23+5/7–24+0/7. Intubate after birth when necessary: 24+0/7IQRboth24+0/7–24+0/7. Chest compressions after birth when necessary: obstetricians 25+0/7 and neonatologists 26+0/7p<0.01IQRobstetricians24+0/7–25+0/7neonatologists25+0/7–26+0/7. Administration of epinephrine after birth when necessary: obstetricians 25+0/7 and neonatologists 26+0/7p<0.01IQRobstetricians24+0/7–25+0/7neonatologists25+0/7–26+0/7. There was a disagreement between both professional groups with obstetricians having a significant lower threshold on providing chest compressions and administering epinephrine than neonatologist. Furthermore, significant differences between institutes were found on fetal monitoring, CS, chest compressions, and epinephrine, which may reflect local policies.

professional groups with obstetricians having a significant lower threshold on providing chest compressions and administering epinephrine than neonatologist. Furthermore, significant differences between institutes were found on fetal monitoring, CS, chest compressions, and epinephrine, which may reflect local policies. Figure 3 shows how certain factors could alter potential recommendations on initiating intensive care at 24+0/7 weeks GA. Ranked by the proportions of subjects being less likely to advise intensive treatment, congenital disorders was the strongest factor, followed by small for gestational age (SGA) infant, no antenatal administration of corticosteroids, and a disturbed fetal heart-rate. There were two items that differed between the two professional groups: first “no administration of corticosteroids”; this factor made 63 % of obstetricians versus 40 % of neonatologists less likely to advise intensive treatment (p = 0.033). Second for “SGA infant”, 92 % of neonatologists versus 76 % of obstetricians were less likely to advise intensive treatment (p = 0.028). Significant differences between institutes were found on male gender, multiple pregnancy, and disturbed fetal monitor, so again, some local preferences seem to exist.Fig. 3 Factors influencing potential recommendation towards parents for initiating intensive treatment. (*)Significantly different between neonatologists and obstetricians: no corticosteroids p 0.012, SGA p 0.028

ale gender, multiple pregnancy, and disturbed fetal monitor, so again, some local preferences seem to exist.Fig. 3 Factors influencing potential recommendation towards parents for initiating intensive treatment. (*)Significantly different between neonatologists and obstetricians: no corticosteroids p 0.012, SGA p 0.028 Discussion This is the first study to assess professional preferences on treatment decisions in extreme prematurity in the Netherlands, particularly of neonatologists and obstetricians. There was a wide variation in preferred treatment decisions at the limits of viability, mostly when aspects were not covered in the Dutch national guideline on perinatal practice in extreme prematurity. This variation was shown between individual perinatal professionals on (a) the attitude towards the GA at which active treatment should be started, (b) the individual preferential lower limits of GA for certain interventions, and (c) the influence of additional patient characteristics on initiating care or not. Neonatologists’ and obstetricians’ opinions differed on the latter two.

als on (a) the attitude towards the GA at which active treatment should be started, (b) the individual preferential lower limits of GA for certain interventions, and (c) the influence of additional patient characteristics on initiating care or not. Neonatologists’ and obstetricians’ opinions differed on the latter two. Recommendations on comfort versus intensive care Regarding the large variation between perinatal professionals, our findings are comparable to Kaempf (2006) and Tomlinson (2010) who also discovered variety at various GA [18, 33]. At <24+0/7 weeks, comfort care is preferred to be recommended, consistent with the Dutch guideline. However, for some physicians, intensive care is an option at 23 weeks of gestation at parental request and, although being internationally practiced, it is not supported in the Dutch national guideline [8, 10, 28]. At 24 weeks, the Dutch guideline requests agreement of parents when initiating intensive care, and only one participant preferred recommending intensive care treatment only without accepting a potential parental request for comfort care. However, the majority did give the recommendation to provide intensive care with comfort care only as an option at parental request. The Dutch guideline describes that “intensive care can be offered” at 24 weeks GA, without explicitly giving an advice on whether to present this as default or present as neutral option next to comfort care. Our results show that at 24 weeks, variety in preferences exists. It is known that presenting delivery room options for extremely premature infants as default exerts a significant effect on decision makers [16]. At 25 weeks gestation, there is similar variety.

default or present as neutral option next to comfort care. Our results show that at 24 weeks, variety in preferences exists. It is known that presenting delivery room options for extremely premature infants as default exerts a significant effect on decision makers [16]. At 25 weeks gestation, there is similar variety. Notable is that some physicians approve comfort care on parental request at 26 and 27 weeks. The Dutch guideline does not explicitly cover this GA [8]. Also, internationally, 26 and 27 weeks of gestation are in general not being seen as the gray zone of viability anymore and the initiation of intensive care at these GA is being seen as standard of care [6, 10, 28, 31]. Kipnis argues that though treatment may offer a reasonable chance of a good outcome, there are situations in which neonatologists should nonetheless defer to parental nontreatment decisions’; however, no specific recommendation for 26 and 27 weeks GA is given [21]. Haward suggests that professional organizations should make guidelines on treatment decisions in extreme prematurity based on the best-interest principle [15]. To the best of our knowledge, there are no papers commenting on an upper limit in terms of GA where parental preferences can be followed including the legal aspects of this.

ts that professional organizations should make guidelines on treatment decisions in extreme prematurity based on the best-interest principle [15]. To the best of our knowledge, there are no papers commenting on an upper limit in terms of GA where parental preferences can be followed including the legal aspects of this. Decisions or interventions associated with prematurity The personal lower limit for certain treatment decisions associated with prematurity varied up to 4 weeks between individuals. It is known that personal preferences can influence counseling and decision-making [34]. For the interventions covered in the Dutch guideline, our results are fairly consistent with the guideline recommendations (transfer to a tertiary center at 23+4/7 weeks GA and administration of corticosteroids at 23+5/7 weeks GA). For these, 59 and 95 %, respectively, of all perinatal professionals have their lower limit at our below that mentioned GA, fulfilling the requirements of guideline. It should although be taken into account that the referring gynecologists (non-third line) were not questioned. Other interventions/decisions show large variation. According to the Dutch guideline, a CS at 24+0/7 weeks GA on fetal indication can be considered only in case of a spontaneously started delivery and after discussion with parents. Participants indicated 25+0/7 weeks of GA as median lower limit, with a wide variation. Tucker Edmonds recently showed that obstetricians had a personal cutoff for performing a CS at a later GA (median 25+0/7 weeks GA) than the institutional cutoff (median 24+0/7 weeks GA) [34]. Having a neonatologist present at delivery and intubation after birth when necessary both showed a median lower limit at 24+0/7 weeks of gestation, with relatively little variation; probably because these two items are seen as the minimum conditions that must be met when offering intensive care at 24+0/7 weeks (and implicitly covered in our guideline). However, there was a much wider range for providing chest compressions (when applicable) and administration of epinephrine. The reason for this wide range might be that they are not covered in the Dutch guideline. Compared to surveys from Finland, UK, and the USA, Dutch physicians seem to prefer almost all interventions at a later GA than their colleagues from mentioned countries [5, 32, 34].

n applicable) and administration of epinephrine. The reason for this wide range might be that they are not covered in the Dutch guideline. Compared to surveys from Finland, UK, and the USA, Dutch physicians seem to prefer almost all interventions at a later GA than their colleagues from mentioned countries [5, 32, 34]. Regarding the differences between obstetricians and neonatologists, for two items, obstetricians preferred significant lower limits than neonatologists (both chest compressions and epinephrine: median lower limit 25+0/7 weeks GA for obstetricians and 26+0/7 weeks GA for neonatologists). Apparently, obstetricians believe that intensive resuscitation can be provided at an earlier GA than their neonatal counterparts, which is contradictory to findings from England by Chan et al. (neonatal staff wished to be more interventional at earlier GA) and also from Finland by Taittonen et al. (pediatric personnel demonstrated more proactive attitudes to the treatment of a premature birth and baby than obstetric personnel) [5, 32]. Obstetricians do not normally perform these parts of resuscitation, so perhaps it was harder for them to comment on these items. However, obstetricians and neonatologists should participate as a team in taking care for mothers and newborns at the limits of viability; including agreement on the extent of potential neonatal resuscitation. Differences in opinions should be solved and no conflicts on care should arise [5, 10, 32].

nt on these items. However, obstetricians and neonatologists should participate as a team in taking care for mothers and newborns at the limits of viability; including agreement on the extent of potential neonatal resuscitation. Differences in opinions should be solved and no conflicts on care should arise [5, 10, 32]. Role of associated factors Individual prognostic factors do play a role in prenatal counseling at 24 weeks of GA for Dutch physicians, mainly a congenital disorder. Tyson et al. described four factors next to GA to have an impact on predicting outcome: birth weight, sex, (non)exposure to antenatal corticosteroids, and single/multiple gestation. These factors have a varying impact in this survey; for (male) gender and a multiple pregnancy <15 % of participants are less likely to advise intensive treatment. The two other factors from Tyson’s prediction model seem to have a greater impact; being SGA and no corticosteroids administered [35]. Although these two have an impact for both specialties, the proportion that agrees differs between them. We do not know why “no corticosteroids administered” is more important for obstetricians, and a “SGA infant” is more important for neonatologists. It is speculating, however, that these factors might be more visible within their specific expertise/field of work.

ialties, the proportion that agrees differs between them. We do not know why “no corticosteroids administered” is more important for obstetricians, and a “SGA infant” is more important for neonatologists. It is speculating, however, that these factors might be more visible within their specific expertise/field of work. In contrast to Tyson, we did not try to quantify the prognostic factors and only asked for an influence. The Dutch guideline states that the factors from the Tyson model “can be taken into account”; however, since the value of these prognostic factors is unknown for the Dutch population, no specific recommendations are given. In 2014 though, 3 years after introduction of the guideline and after this survey was done, Dutch data showed that no antenatal corticosteroids, male gender, maternal age >35 years, Caucasian ethnicity, non-cephalic presentation, and birth outside a level III hospital were predictors for mortality (together with GA) in a prematurity cohort of GA 250–316 [29]. This survey shows that some factors were taken into account in prenatal consultation at 24 weeks GA. Nevertheless, we should allow for an individual approach to these prognostic factors, since the models are mostly not developed for counseling decisions and, most important, they do not predict an individual course. However, these factors might be helpful in addition to the GA to identify the potential range of outcome.

evertheless, we should allow for an individual approach to these prognostic factors, since the models are mostly not developed for counseling decisions and, most important, they do not predict an individual course. However, these factors might be helpful in addition to the GA to identify the potential range of outcome. Strengths and limitations The strongest aspects of this study are the national level of the survey (all Dutch tertiary centers where included) representing our national situation. Also, the fact that most of the questions are directly related to content of the national guideline on perinatal practice makes it relevant for daily practice. This study also has limitations. Some degree of selection-bias cannot be ruled out. The character of the survey (asking for recommendations to parents and personal lower limits for certain decisions) might not be representative for actual practice; however, no less relevant since it is known that, despite guidelines or local policies, personal preferences influence decision-making [34]. Because of the long inclusion period (16 months) effects of experience or learning cannot be ruled out, and it is unsure to what extent results from this Dutch cohort can be generalized to the international situation. However, many countries do have guidelines, so the general conclusion on variety between individuals and between professions despite guidelines might be applicable.

perience or learning cannot be ruled out, and it is unsure to what extent results from this Dutch cohort can be generalized to the international situation. However, many countries do have guidelines, so the general conclusion on variety between individuals and between professions despite guidelines might be applicable. Conclusions and future perspectives This is the first study to asses physicians’ opinions on treatment decisions at the threshold of viability in the Netherlands. There was a wide variety in preferred treatment decisions at the limits of viability and in perceived lower limits of treatment between individual professionals. This variation was especially observed when aspects were not covered in the Dutch national guideline on perinatal practice in extreme prematurity. Furthermore, obstetricians and neonatologists disagreed on some aspects, particularly lower limits of GA for cardiac resuscitation and the influence of patient characteristics on initiating care. This variety and disagreement can lead to unwanted practice variation.

ideline on perinatal practice in extreme prematurity. Furthermore, obstetricians and neonatologists disagreed on some aspects, particularly lower limits of GA for cardiac resuscitation and the influence of patient characteristics on initiating care. This variety and disagreement can lead to unwanted practice variation. When items are covered in a guideline it seems to reduce, but not to exclude, variation. Especially when a guideline leaves room for interpretation, personal opinions will become more important. Revision of guidelines to cover more aspects might be a solution. However, more strict guidelines and recommendations that are based on national consensus need not interfere with an individualized approach, since making different choices based on patient characteristics and parental preferences are part of this consensus. The current study showed that in similar cases, dealing with different caregivers, different decisions can be made. At the limits of viability, it covers, by definition, decisions about life and death and practice variation is therefore even more unwanted. Abbreviations CSCaesarian section GAGestational age(s) IQRInterquartile range SGASmall for gestational age The authors would like to thank all authors who shared their survey with us [1, 3, 5, 18, 19, 27] The authors would like to thank all participating Dutch obstetricians and neonatologists.

When items are covered in a guideline it seems to reduce, but not to exclude, variation. Especially when a guideline leaves room for interpretation, personal opinions will become more important. Revision of guidelines to cover more aspects might be a solution. However, more strict guidelines and recommendations that are based on national consensus need not interfere with an individualized approach, since making different choices based on patient characteristics and parental preferences are part of this consensus. The current study showed that in similar cases, dealing with different caregivers, different decisions can be made. At the limits of viability, it covers, by definition, decisions about life and death and practice variation is therefore even more unwanted. Abbreviations CSCaesarian section GAGestational age(s) IQRInterquartile range SGASmall for gestational age The authors would like to thank all authors who shared their survey with us [1, 3, 5, 18, 19, 27] The authors would like to thank all participating Dutch obstetricians and neonatologists. Authors’ contributions RG, JD, AvH, and MH had the core idea for this study. RG, JD, AvH, MW, MH, and RH prepared the questionnaire. All authors either analyzed the data or interpreted the results. RG wrote the draft of the article. All other authors commented on the manuscript.

The authors would like to thank all participating Dutch obstetricians and neonatologists. Authors’ contributions RG, JD, AvH, and MH had the core idea for this study. RG, JD, AvH, MW, MH, and RH prepared the questionnaire. All authors either analyzed the data or interpreted the results. RG wrote the draft of the article. All other authors commented on the manuscript. Compliance with ethical standards This study was not funded. All procedures performed in this study were in accordance with the ethical standard. In the Netherlands, there is no need for a survey study to be validated by an ethics committee. The local institutional review board confirmed this. Conflict of interest The authors declare that they have no conflict of interest.

Introduction The recurrence of primary renal disease after renal transplantation is a risk factor of inferior outcome [23]. Drug-resistant nephrotic syndrome leading to end-stage renal failure is one the most aggresive glomerular diseases recurring after renal transplanation [1]. Immediate recurrence suggests the role of preformed and circulating protein permeability factor(s), which bind to specific targets in transplanted kidney. Identification of responsible factor(s) is an ongoing subject of clinical investigations for years, and several hypotheses have been raised, including protein permeability factor suggested by Savin’s group [21], soluble urokinase type plasminogen activator receptor (suPAR) [24], cardiothropin-like cytokine 1(CLC-1) [13], and anti-CD40 antibody [4]. The hypothesis of circulating factor(s)-related mechanism of post-transplant recurrence of nephrotic syndrome is translated into therapeutic protocols, including plasmapheresis (to remove the factor(s)) [17, 18], corticosteroids (to stabilize the cytoskeleton of podocytes and reduce podocytes apoptosis) [19], cyclosporine A (to stabilize the cytoskeleton of podocytes) [5], and rituximab (monoclonal antibody depleting B CD20 cells) to deplete the cytokines-producers [7, 15]. There are conflicting data on efficacy of rituximab used for treatment of post-transplant recurrence of nephrotic syndrome in adult and pediatric patients. The purpose of our study was to present the summary of our experience with the efficacy and safety of rituximab in five children after first renal transplantation, who presented rapid recurrence of nephrotic syndrome.

sed for treatment of post-transplant recurrence of nephrotic syndrome in adult and pediatric patients. The purpose of our study was to present the summary of our experience with the efficacy and safety of rituximab in five children after first renal transplantation, who presented rapid recurrence of nephrotic syndrome. Patients and methods A retrospective chart review was performed and included five children at the age from 5 to 12 years (at the day of renal transplantation), who presented immediate (day 1 or 2) recurrence of nephrotic syndrome (NS) post-transplant. The criterion of chart selection was the use of rituximab for the treatment of NS recurrence after transplantation. The drug-resistant nephrotic syndrome was the underlying cause of end-stage renal disease. Focal segmental glomerulosclerosis (FSGS) was present in native renal biopsy in two patients, mesangial-proliferative glomerulonephritis (MesPGN) in two, and minimal change nephrotic syndrome (MCNS) in one case. One patient had confirmed heterozygous podocine (NPHS2) genetic mutation (p.R229Q). There was 3 to 8 years interval between introducing dialysis and primary renal transplantation. There were four deceased donor-related and one living-related transplantation performed, the last one in a patient with genetic podocine mutation (patient’s father was a donor). All patients received triple immunosuppression (CsA + MMF + Pred). All five demonstrated immediate recurrence of nephrotic syndrome (four on the day 1 and one on the second day post-transplant), with gross proteinuria and oliguria. Plasmapheresis was introduced in all five patients, and in four cases, rituximab was directly added to the protocol at the dose of 375 mg/m2 on weekly basis. In one case, the introduction of rituximab was delayed due to primary promising effect of plasmapheresis; however, once the patient became plasmapheresis-dependent, the drug was added during fourth month post-transplant [8]. Four patients received four weekly doses, and in one case the drug administration was limited to two doses due to severe relapsing infections The count of B CD19 cells was monitored on weekly and then monthly basis by flow cytometry. The clinical characteristics and treatment course in summarized in Table 1. The complete remission was achieved in two patients (no. 1 and no. 3), with complete depletion of B CD19 for up to 4 and 2 months (after last dose of rituximab), respectively, and with no adverse drug-related adverse events.

by flow cytometry. The clinical characteristics and treatment course in summarized in Table 1. The complete remission was achieved in two patients (no. 1 and no. 3), with complete depletion of B CD19 for up to 4 and 2 months (after last dose of rituximab), respectively, and with no adverse drug-related adverse events. The remission was sustained within 7 and 5 years of follow-up, respectively, with no specific treatment beyond standard triple immunosuppression (CsA + MMF + Pred). Only partial remission (significant, however floating proteinuria; from 0.4 to 3.5 g and from 0.5 to 2.0 g/24 hours; respectively) was seen in two cases (no. 4 and no. 5), presenting complete B CD19 depletion up to 7 and 4 months after treatment, respectively. Serious adverse events were present in these two cases, including relapsing severe infections (blood culture was positive for Klebsiella pneumoniae) (case no. 4), which limited the number of scheduled rituximab doses to 2 and severe acute lung injury (RALI) (case no. 5). Patient no. 4 lost the graft 2.5 years after transplantation due to chronic nephropathy. Patient no. 5 died due to non-infectious rapid lung fibrosis. There was no therapeutic effect at all in the case no. 2, who was also finally given oral galactose; however, the graft was never functioning and was removed 7 months post-transplant.Table 1 Clinical chacteristics and treatment course of five children with post-transplant recurrence of nephrotic syndrome

lung fibrosis. There was no therapeutic effect at all in the case no. 2, who was also finally given oral galactose; however, the graft was never functioning and was removed 7 months post-transplant.Table 1 Clinical chacteristics and treatment course of five children with post-transplant recurrence of nephrotic syndrome Patient no. Age at diagnosis of NS (years), gender renal biopsy Age at trans-plantation (years) origin of the graft Timing of NS recurrence post-transplant (days) Treatment day of rituximab introduction after transplantation Duration of B CD19 depletion (months)/correlation of number of B CD19 cells to clinical effect Clinical effects renal graft biopsy SAE 1 2, male FSGS 5.5 deceased donor 1 PF + CsA + MMF + MP + rituximab 4 × 375 mg/m2 day 133 4 Yes Complete remission (7 years of follow-up) MCNS No 2 2, female MCNS 5 deceased donor 1 PF + CsA + MMF + MP + rituximab 4 × 375 mg/m2 + galactose day 25 3 No Failure; graft removed 7 months later FSGS No 3 2, female MesPGN (NPHS2 gene mutation-single heterozygous mutation; p.R229Q) 10 living-related donor 1 PF + CsA + MMF + MP + rituximab 4 × 375 mg/m2 day 15 2 Yes Complete remission (5 years of follow-up) FSGS No 4 6, female FSGS 12 deceased donor 1 PF + CsA + MMF + MP + rituximab 2 × 375 mg/m2 day 21 7 no Partial remission; graft removed 2.5 years later FSGS Relapsing severe infections 5 6, male MesPGN 10 deceased donor 2 PF + CsA + MMF + MP + rituximab 4 × 375 mg/m2 day 15 4 no Partial remission MPGN RALI (fatal)

-up) FSGS No 4 6, female FSGS 12 deceased donor 1 PF + CsA + MMF + MP + rituximab 2 × 375 mg/m2 day 21 7 no Partial remission; graft removed 2.5 years later FSGS Relapsing severe infections 5 6, male MesPGN 10 deceased donor 2 PF + CsA + MMF + MP + rituximab 4 × 375 mg/m2 day 15 4 no Partial remission MPGN RALI (fatal) MesPGN mesangial-proliferative glomerulonephritis, FSGS focal segmental glomerulosclerosis, MCNS minimal change nephrotic syndrome, NPHS2 podocine, CsA cyclosporine A, MMF mycophenolate mofetil, MP methylprednisolone, PF plasmapheresis, SAE serious adverse events, RALI rituximab-related acute lung injury Discussion Rituximab has been used for treatment of post-transplant nephrotic syndrome for a decade, and initial reports were in favor to the drug efficacy [7, 12, 16]. Our preliminary experience (case no. 1) was also very positive [8] and encouraged us to use this protocol. The reports published later on have revealed the conflicting data on clinical efficacy. Published 13 studies (mainly case reports or case series reports) included overall 44 pediatric patients treated with rituximab for post-transplant recurrence of nephrotic and among those 27 patients (55 %) presented complete remission, 14 (28.5 %) partial remission of nephrotic syndrome, and no effect was present in the remaining 8 (16.5 %) cases [14].

eports or case series reports) included overall 44 pediatric patients treated with rituximab for post-transplant recurrence of nephrotic and among those 27 patients (55 %) presented complete remission, 14 (28.5 %) partial remission of nephrotic syndrome, and no effect was present in the remaining 8 (16.5 %) cases [14]. Rituximab was always used in combination with plasmapheresis and other drugs It must be noted that most of the reported patients, treated for recurrence of nephrotic syndrome after renal transplantation with rituximab, underwent the combined therapy with plasma exchange and post-transplant triple immunosuppression including calcineurine inhibitors, steroids, and antiproliferative drugs, so the final clinical effect (if positive) might be the summary result of several therapeutic mechanisms. Moreover, the timing of introduction and the number of rituximab doses used in this setting was variable, as the drug was given also preemptively (as prophylaxis) at the day of transplantation [6] or as rescue therapy of overt recurrence of nephrotic syndrome [7, 12, 16, 22]. The number of doses ranged from single to six weekly doses of 375 mg/m2. The correlation between B CD20/CD19 cells depletion and clinical effect was also variable [14]. Among patients described in our report, four (out of five) have received rituximab as first-line therapy, and in remaining one, the drug was prescribed as second-line treatment due to persistent dependency from plasmapheresis. In four of our cases, four fixed weekly doses of 375 mg/m2 were given, and in remaining one, the number of scheduled doses was limited to two, due to relapsing infections. It is not clear whether the time interval (shorter versus longer) between post-transplant NS recurrence and introduction of rituximab is important for the overall efficacy of this intervention. Some authors expressed such suggestion in the discussion of their reports; however, there is no evidence so far. Almost universally, rituximab therapy was combined with plasmapheresis, and other suggestions stated that rituximab should be introduced early after the evidence of plasmapheresis failure [10, 12, 22].

tion. Some authors expressed such suggestion in the discussion of their reports; however, there is no evidence so far. Almost universally, rituximab therapy was combined with plasmapheresis, and other suggestions stated that rituximab should be introduced early after the evidence of plasmapheresis failure [10, 12, 22]. Why the efficacy of rituximab is variable It is not clear, whether the lack of significant clinical effect in three (of five) of our cases is related to specificity of this particular setting of the patients, in whom the recurrence was immediate (day 1 or 2), which may suggest the high concentration/activity of hypothetic circulating factors. Unfortunately, at the time of this treatment the evaluation of suPAR concentration in serum or urine was not available. The clinical course of these cases also show that long waiting time on dialysis (from 3 to 8 years) has no protective effect in terms of risk of rapid recurrence of nephrotic syndrome, as was carefully suggested based on some clinical observations [2]. Using similar approach, we faced two successful and three disappointing cases. B cells depletion was achieved in all cases, irrespective from the clinical outcome, so apparently rituximab was effective in terms of pharmacodynamics activity, independent from the absence of clinical effect in those patients, who did not respond. This might correspond with the data on B cell-independent mechanism of rituximab, which may act via regulation (preservation) of sphingomyelin phosphodiesterase acid-like 3b (SMLPD-3b), a protein which is involved in the podocyte cytoskeleton activity [6]. The anecdotal use of galactose in case no. 2, based on hypothesis of local inhibition of protein permeability factor in podocytes by galactose, was not effective either [20, 21]. Probably achieving (or not) the complete remission in particular cases is the effect of random targeting (or not) the patient, in whom the mechanism of nephrotic syndrome is directly related to abnormal activity of B cells. It is also possible that partial efficacy is related to B-independent manner of rituximab action [6]. Interestingly, the complete depletion of B cells was also present in our case no. 2, who showed complete resistance to rituximab, suggesting that there are specific cases, in whom neither B cell-dependent nor B cell-independent (local) mechanisms of the drug are relevant to reduce heavy post-transplant proteinuria.

6]. Interestingly, the complete depletion of B cells was also present in our case no. 2, who showed complete resistance to rituximab, suggesting that there are specific cases, in whom neither B cell-dependent nor B cell-independent (local) mechanisms of the drug are relevant to reduce heavy post-transplant proteinuria. This is possible that this patient had T cell-dependent mechanism of the disease; however, plasmapheresis, which theoretically should be helpful in removing T cell driven cytokines (such as IL-13 stimulating local expression of CD80 molecule in podocytes) [3], was not effective either. Current experience on potential of use of abatacept (specific inhibitor of CD80) in post-transplant recurrence of NS, resistant to plasmapheresis, and rituximab [25] was not known at the time when our case was treated. Unfortunately, we are still at an early stage of distinguishing the different patterns of the mechanisms of post-transplant nephrotic syndrome; however, recent data on the use of diagnostic tool to identify patients at high risk of FSGS recurrence are encouraging [4].

This is possible that this patient had T cell-dependent mechanism of the disease; however, plasmapheresis, which theoretically should be helpful in removing T cell driven cytokines (such as IL-13 stimulating local expression of CD80 molecule in podocytes) [3], was not effective either. Current experience on potential of use of abatacept (specific inhibitor of CD80) in post-transplant recurrence of NS, resistant to plasmapheresis, and rituximab [25] was not known at the time when our case was treated. Unfortunately, we are still at an early stage of distinguishing the different patterns of the mechanisms of post-transplant nephrotic syndrome; however, recent data on the use of diagnostic tool to identify patients at high risk of FSGS recurrence are encouraging [4]. One of our patients had heterozygous mutation of podocine and received living-related transplant. There were some reports on recurrence of nephrotic syndrome after renal transplantation in children with podocine genetic mutations, with high variety in terms of post-transplant time to proteinuria (range 4 to 10 years). The final outcome of reported six cases was good [11]. As the recurrence was immediate (first day) in our case, we presume that she presented combined, genetic defect and immunological injury, and the latter one was the major cause of rapid recurrence.

6 % at the end of the study period in 2012. A survey conducted in 2002, comprising 576 pregnant women at the same hospital, had shown that 21.5 % smoked during pregnancy (unpublished data). Thus, the prevalence of smoking a few years before the commencement of the present investigation appears to have been even higher. The most recent data in Austria—derived from 2010—show that 20–30 % of pregnant women smoked [10]. However, the rates registered in the present study were consistently lower than 20 %. Although maternal smoking has been decreasing in other countries over the last few years [5, 25] and a steady decline was also noted in the present investigation, smoking rates remain high among young mothers below 20 years of age (43.7 %) and those aged 20–25 years (28.8 %). In our population, smoking rates among women older than 36 years were below the overall prevalence. The high prevalence of smoking among very young pregnant women has been attributed to the high-risk behavior and lack of responsibility at this age, which might have led to pregnancy in the first place [7, 12]. This may well be true in view of the fact that the smoking prevalence among pregnant women below the age of 20 years, registered in the present study, was approximately 20 % higher than that in the general population.

terms of post-transplant time to proteinuria (range 4 to 10 years). The final outcome of reported six cases was good [11]. As the recurrence was immediate (first day) in our case, we presume that she presented combined, genetic defect and immunological injury, and the latter one was the major cause of rapid recurrence. Safety concern Another issue is safety of this protocol. We have faced two cases of severe adverse events, including severe recurrent bacterial infections in case no. 4 suggesting overimmunosuppression. The most significant adverse event was RALI (rituximab-associated acute lung injury) in case no. 5, which was reported in details as case report [9]. The risk/benefit ratio is then individual and must be taken into consideration. In summary, we present disappointing results of rituximab plasmapheresis-based rescue protocol of treating immediate recurrence of post-transplant nephrotic syndrome. It is unclear, whether aggressive pattern of immediate recurrence was the reason of the treatment failure in three cases. Center experience was the reason of the evolution of our attitude in terms of using this protocol, from enthusiastic to reluctant. Abbreviations B CD19, B CD20Lymphocytes B expressing specific receptors bound by monoclonal antibody (rituximab) CsACyclosporine A MMFMycophenolate mofetil MPMethylprednisolone PFPlasmapheresis NSNephrotic syndrome FSGSFocal segmental glomerulosclerosis MCNSMinimal change nephrotic syndrome MesPGNMesangial-proliferative glomerulonephritis IL-13Interleukin 13 CD80T-cell co-stimulatory molecule Revisions received: 18 May 2016; 13 June 2016

Abbreviations B CD19, B CD20Lymphocytes B expressing specific receptors bound by monoclonal antibody (rituximab) CsACyclosporine A MMFMycophenolate mofetil MPMethylprednisolone PFPlasmapheresis NSNephrotic syndrome FSGSFocal segmental glomerulosclerosis MCNSMinimal change nephrotic syndrome MesPGNMesangial-proliferative glomerulonephritis IL-13Interleukin 13 CD80T-cell co-stimulatory molecule Revisions received: 18 May 2016; 13 June 2016 Author’s contribution RG was a senior consultant in all medical cases, created the idea, and has written the manuscript. WJ was a member of medical team involved in therapeutic process and reviewed the manuscipt. JR was a member of medical team involved in therapeutic process and reviewed the manuscipt. BP performed immunological monitoring in all cases and reviewed the manuscript. SP was a member of medical team, involved in therapeutic process, and reviewed the manuscipt. Compliance with ethical standards All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Funding The study did not receive fundings. Conflict of interest The authors declare that they have no conflict of interest. Informed consent Informed consent was obtained from all individual participants included in this study.

Introduction Although the harmful effects of smoking during pregnancy are well known and have been investigated for more than 50 years, a significant number of women continue to smoke during pregnancy. Thus, smoking still is a public health concern. In addition to causing the fetal tobacco syndrome [10], it has been held responsible for the stagnation of infant mortality in the last decade in Austria, along with alcohol consumption and obesity in pregnant women [26]. Secondary complications due to maternal smoking lead to higher hospital admission rates of infants from the postnatal phase to the age of 5 years, which signifies higher health-care-related costs [2, 16]. Smoking in the general population in Austria With a smoking prevalence of 34 % in 2009, Austria ranks among the top six countries in the European Union [23], assumes the fourth highest rank in daily smoking (26 % of adult women smoked on a daily basis in 2008) [17], and is among the countries with the highest prevalence of teenage smoking. The overall prevalence of teenage smoking is 25.4 %; among girls, it ranges from 21.1 to 30.4 % depending on the age group [9].

on [23], assumes the fourth highest rank in daily smoking (26 % of adult women smoked on a daily basis in 2008) [17], and is among the countries with the highest prevalence of teenage smoking. The overall prevalence of teenage smoking is 25.4 %; among girls, it ranges from 21.1 to 30.4 % depending on the age group [9]. Smoking during pregnancy in Europe and Austria A study comprising data from six European countries (Belgium, Bulgaria, Germany, Greece, Ireland, and Portugal) showed that the prevalence of smoking at the time of delivery varied between 7 % (Bulgaria) and 52.5 % (Ireland), with a mean smoking prevalence of 19.6 % [8]. Twenty-one percent of pregnant women smoked in Germany [18], while 20–30 % of pregnant women smoked in Austria in 2010 [10]. A study conducted in Spain in 2013, addressing the effects of smoking on fetal biometry, comprised 2478 women and revealed an overall smoking prevalence of 32 % [4]. These data are consistent with similar studies in Europe.

nt women smoked in Germany [18], while 20–30 % of pregnant women smoked in Austria in 2010 [10]. A study conducted in Spain in 2013, addressing the effects of smoking on fetal biometry, comprised 2478 women and revealed an overall smoking prevalence of 32 % [4]. These data are consistent with similar studies in Europe. Harmful effects of smoking on the fetus Smoking exerts harmful effects in every stage of pregnancy. It is the most important preventable risk factor for an adverse pregnancy outcome [1, 11]. Horak et al. [10] summarized the known effects as the fetal tobacco syndrome. Cigarette smoke contains toxins in quantities that affect placental and fetal cell growth, proliferation, and differentiation and exert harmful effects on the development of organ systems including the cardiovascular, respiratory, and nervous system. Smoking reduces blood flow to the placenta and accumulates carboxyhemoglobin in the fetus, both of which cause chronic hypoxic stress for the fetus [1]. Placental metabolism is altered in smoking women, and their placenta also revealed structural differences as well as oxidative damage of the placental tissue [21]. These effects might explain the higher risk of miscarriage, fetal growth restriction, stillbirth, preterm birth, and placenta praevia [11]. Fetal growth restriction results in small for gestational age neonates. These neonates are burdened with both short- and long-term health effects, including higher risk for hyperglycemia during the neonatal period, failing to thrive during childhood, and obesity, early-onset type 2 diabetes, and arterial hypertension during adulthood [19].

ion results in small for gestational age neonates. These neonates are burdened with both short- and long-term health effects, including higher risk for hyperglycemia during the neonatal period, failing to thrive during childhood, and obesity, early-onset type 2 diabetes, and arterial hypertension during adulthood [19]. Aims of the study The aims of the study were the following:To determine a statistically significant change, if any, in the prevalence of smoking (yes/no) among pregnant women from 2007 to 2012. To elucidate risk factors for maternal smoking, specifically whether maternal age and parity influenced its prevalence. Besides, neonatal outcome data were analyzed to confirm the harmful effects of maternal smoking on newborns and determine whether changes in prevalence would be reflected in neonatal parameters.

Aims of the study The aims of the study were the following:To determine a statistically significant change, if any, in the prevalence of smoking (yes/no) among pregnant women from 2007 to 2012. To elucidate risk factors for maternal smoking, specifically whether maternal age and parity influenced its prevalence. Besides, neonatal outcome data were analyzed to confirm the harmful effects of maternal smoking on newborns and determine whether changes in prevalence would be reflected in neonatal parameters. Patients and Methods The data for the study were collected at the obstetric department of Sozialmedizinisches Zentrum (SMZ) Ost, a tertiary-care medical center providing a wide spectrum of mother/child health care. Data were collected from 2007 to 2012. Every woman with a singleton pregnancy, who had delivered her child after the 22nd week of gestation, was included in this study. The women’s general and pregnancy-related medical history was registered when they booked their delivery at the hospital. Their current consumption of tobacco (yes/no; if yes number of cigarettes/day) was registered in a personal conversation by a midwife or a doctor and entered in an electronic data sheet. Women were classified into (0) non-smokers or (1) smokers; and the numbers of cigarettes per day were documented. Smoking behavior was registered once during pregnancy. Maternal age was divided into the following categories: (0) younger than 20 years, (1) 20–25 years, (2) 26–30 years, (3) 31–36 years, and (4) over 36 years. Neonatal characteristics included birthweight and weight percentile. The duration of pregnancy in weeks and days was also registered. The children were classified as preterm births (37 + 0 weeks or below) or small for gestational age (below the 10th percentile). Women who gave birth more than once in the study period were treated as separate individuals in the respective study year. Data obtained from a survey conducted in 2002, comprising 576 women in the identical hospital setting, were used to compare the prevalence of smoking. All data were pseudonymized.

10th percentile). Women who gave birth more than once in the study period were treated as separate individuals in the respective study year. Data obtained from a survey conducted in 2002, comprising 576 women in the identical hospital setting, were used to compare the prevalence of smoking. All data were pseudonymized. Statistical analysis Basic patient and neonatal data were collected at the initial exploratory data analysis. Multiple logistic regressions were used (a) to explore the association between the year of documentation and the occurrence of smoking, (b) to assess upward and downward trends in the association between smoking during pregnancy and maternal age, (c) to test for the occurrence of small for gestational age children in connection with maternal smoking, and (d) to test for the age distribution throughout the study years. Linear regression was used to assess the association between maternal age and the number of cigarettes smoked per day. The χ2 test was used to test for significance as to whether parity is associated with smoking during pregnancy. The test was also used to determine the significance of relative risk and the number of cigarettes causing preterm births and small for gestational age newborns.

e and the number of cigarettes smoked per day. The χ2 test was used to test for significance as to whether parity is associated with smoking during pregnancy. The test was also used to determine the significance of relative risk and the number of cigarettes causing preterm births and small for gestational age newborns. Kendall’s correlation coefficient was used to determine whether smoking behavior changed between pregnancies. The analysis was performed with IBM SPSS Statistics 21. The relative risk and the number of cigarettes needed to cause harm were calculated by using a modified Microsoft Excel sheet by Dr. Georg Heinze available at http://www.meduniwien.ac.at/user/georg.heinze/mb2. The level of significance was set to p < 0.05. Results Of 12,302 women who underwent their first antenatal investigation at SMZ Ost Hospital, 11,142 were included in the study. Of the records, 1160 were excluded because of twin/triplet pregnancies (n = 823 single records), missing information in the charts, not completing the 22nd week of gestation, or being lost to follow-up (n = 337).

n who underwent their first antenatal investigation at SMZ Ost Hospital, 11,142 were included in the study. Of the records, 1160 were excluded because of twin/triplet pregnancies (n = 823 single records), missing information in the charts, not completing the 22nd week of gestation, or being lost to follow-up (n = 337). Characteristics of women and newborns In the final dataset, 2022 (18.1 %) women had smoked during the study years. The mean age of non-smokers was 30.33 years (SD = 5.574) and that of smokers was 27.56 years (SD = 5.983). A total of 394 women were below 20 years of age, of whom 222 (1.9 %) were non-smokers and 172 (1.5 %) smokers. The largest non-smoking group was 31–36 years old (34.5 %) and the largest smoking group 20–25 years old (32.1 %). The mean number of cigarettes smoked per day was 8.84 (SD = 5.268), which corresponded to the category of 5–10 cigarettes per day (43.9 %). Of the smokers, 36.2 % smoked 1–5 cigarettes per day, and 19.9 % smoked more than 10 cigarettes per day. Parity was rather evenly distributed among both groups, with slightly more primiparous women in both groups. The mean duration of pregnancy in weeks was slightly higher in the non-smoking group (38.85 % [SD = 2.384] vs. 38.69 % [SD = 2.47]; Table 1).Table 1 Maternal characteristics Characteristics of women Non-smokers Smokers

Characteristics of women and newborns In the final dataset, 2022 (18.1 %) women had smoked during the study years. The mean age of non-smokers was 30.33 years (SD = 5.574) and that of smokers was 27.56 years (SD = 5.983). A total of 394 women were below 20 years of age, of whom 222 (1.9 %) were non-smokers and 172 (1.5 %) smokers. The largest non-smoking group was 31–36 years old (34.5 %) and the largest smoking group 20–25 years old (32.1 %). The mean number of cigarettes smoked per day was 8.84 (SD = 5.268), which corresponded to the category of 5–10 cigarettes per day (43.9 %). Of the smokers, 36.2 % smoked 1–5 cigarettes per day, and 19.9 % smoked more than 10 cigarettes per day. Parity was rather evenly distributed among both groups, with slightly more primiparous women in both groups. The mean duration of pregnancy in weeks was slightly higher in the non-smoking group (38.85 % [SD = 2.384] vs. 38.69 % [SD = 2.47]; Table 1).Table 1 Maternal characteristics Characteristics of women Non-smokers Smokers n (%) 9120 (81.9 %) 2022 (18.1 %) Year of delivery 2007 1441 (80.9 %) 341 (19.1 %) 2008 1503 (80.7 %) 360 (19.3 %) 2009 1490 (80.7 %) 357 (19.3 %) 2010 1479 (81.5 %) 336 (18.5 %) 2011 1593 (82.8 %) 330 (17.2 %) 2012 1614 (84.4 %) 298 (15.6 %) Age (mean, standard deviation) 30.33 ± 5.574 27.56 ± 5.983 Age categories Non-smokers Smokers Younger than 20 years 222 (2.4 %a, 1.9 %c) 172 (8.5 %b, 1.5 %c) 20–25 years 1606 (17.6 %a, 14.4 %c) 649 (32.1 %b, 5.8 %c) 26–30 years 2837 (31.1 %a, 25.5 %c) 581 (28.7 %b, 5.2 %c) 31–36 years 3149 (34.5 %a, 28.5 %c) 448 (22.2 %b, 4.2 %c) Over 36 years 1306 (14.3 %a, 11.7 %c) 172 (8.5 %b, 1.5 %c) Number of cigarettes smoked per day (mean, standard deviation) 0 8.84 ± 5.268 Number of cigarettes smoked per day Non-smokers Smokers 0 cigarette/day 9120 0 (81.9 %) 1–5 cigarettes/day 0 732 (36.2 %b, 6.6 %c) 5–10 cigarettes/day 0 887 (43.9 %b, 7.9 %c) More than 10 cigarettes/day 0 403 (19.9 %b, 3.6 %c) Parity Non-smokers Smokers 4329 (47.5 %a, 38.9 %c) 915 (45.3 %b, 8.2 %c) 4791 (52.5 %a, 42.9 %c) 1107 (54.7 %b, 9.9 %c) Primiparous Multiparous Duration of pregnancy in weeks 38.85 ± 2.384 38.69 ± 2.47

732 (36.2 %b, 6.6 %c) 5–10 cigarettes/day 0 887 (43.9 %b, 7.9 %c) More than 10 cigarettes/day 0 403 (19.9 %b, 3.6 %c) Parity Non-smokers Smokers 4329 (47.5 %a, 38.9 %c) 915 (45.3 %b, 8.2 %c) 4791 (52.5 %a, 42.9 %c) 1107 (54.7 %b, 9.9 %c) Primiparous Multiparous Duration of pregnancy in weeks 38.85 ± 2.384 38.69 ± 2.47 aWithin smokers bWithin non-smokers cWithin study population To determine changes in smoking behavior between pregnancies, women who had delivered more than one baby at the hospital during the study period were viewed separately for the purpose of the investigation. Of 1255 women who had two babies at the study hospital, 212 smoked during one or both pregnancies. Seventy-six (35.8 %) women increased their cigarette consumption, 43 (20.3 %) smoked the same number of cigarettes, and the remaining women (n = 93; 43.9 %) reduced or quit smoking. Of 45 women who had three babies at the hospital, 15 had smoked during one or more pregnancies. Data concerning the newborns (Table 2) were divided according to maternal smoking behavior. The infants of non-smoking women had a higher mean birthweight in grams (3352.7 [SD = 615.57] vs. 3158 [SD = 604.36]). The mean weight percentile of infants born to smokers was 34.57 (SD = 25.49) and that of non-smokers was 43.94 (SD = 26.91). Preterm birth rates were 8.3 % among non-smoking women and 9.5 % among smoking women. The incidence of small for gestational age children was slightly higher among smokers than among non-smokers,Table 2 Characteristics of newborns

ts born to smokers was 34.57 (SD = 25.49) and that of non-smokers was 43.94 (SD = 26.91). Preterm birth rates were 8.3 % among non-smoking women and 9.5 % among smoking women. The incidence of small for gestational age children was slightly higher among smokers than among non-smokers,Table 2 Characteristics of newborns Neonatal outcome Non-smokers Smokers Weight in grams (mean, standard deviation) 3352.7 ± 615.57 3158.39 ± 604.36 Weight percentile (mean, standard deviation) 43.94 ± 26.91 34.57 ± 25.49 Non-smokers Smokers Preterm births (n, %) 758 (8.3 %a, 6.8 %c) 194 (9.5 %b, 1.7 %c) Non-smokers Smokers Small for gestational age (n, %) 1037 (11.4 %a, 9.3 %c) 347 (17.2 %b, 3.1 %c) aWithin non-smokers bWithin smokers cWithin study population Risk factors for maternal smoking The overall percentage of smokers during the study years was 18.1 %, increased from 19.1 % in 2007 to 19.3 % in 2008, then remained constant at 19.3 % for 1 year and fell steadily thereafter, reaching the lowest figure of 15.6 % in 2012. The decline was significant in the logistic regression analysis with an odds ratio (OR) of 0.952 (95 % confidence interval (CI) 0.926–0.980, p = 0.00; Fig. 1).Fig. 1 Change of maternal smoking in the study years in percent and absolute numbers The age distribution of pregnant women did not change significantly throughout the study years (−0.22, 95 % CI −0.46–1.52, p = 0.42).

Risk factors for maternal smoking The overall percentage of smokers during the study years was 18.1 %, increased from 19.1 % in 2007 to 19.3 % in 2008, then remained constant at 19.3 % for 1 year and fell steadily thereafter, reaching the lowest figure of 15.6 % in 2012. The decline was significant in the logistic regression analysis with an odds ratio (OR) of 0.952 (95 % confidence interval (CI) 0.926–0.980, p = 0.00; Fig. 1).Fig. 1 Change of maternal smoking in the study years in percent and absolute numbers The age distribution of pregnant women did not change significantly throughout the study years (−0.22, 95 % CI −0.46–1.52, p = 0.42). With regard to the age distribution, smokers accounted for 43.7 % of women younger than 20 years and 28.8 % of those aged 20–25 years. The percentage of maternal smoking decreased with age; the lowest percentage of 13.2 % was observed in women older than 36 years of age (Fig. 2). The overall logistic regression function showed that the prevalence of smoking fell steadily with increasing maternal age; the OR was 0.917 (95 % CI 0.908–0.925, p = 0.00).Fig. 2 Maternal age and smoking Linear regression, which only took smokers (n = 2022) into account, revealed no linear relationship between maternal age and cigarette consumption. The model showed a R2 and adjusted R2 of 0.01, with a standard error of estimate of 5.266 (p = 0.00). The weak R2 clearly indicates the absence of a linear relationship between the two variables.

h only took smokers (n = 2022) into account, revealed no linear relationship between maternal age and cigarette consumption. The model showed a R2 and adjusted R2 of 0.01, with a standard error of estimate of 5.266 (p = 0.00). The weak R2 clearly indicates the absence of a linear relationship between the two variables. With regard to the number of births and smoking during pregnancy, the χ2 test revealed a significant decline in smoking with increasing numbers of births (31.359 [critical value 18.307], df = 10, p = 0.001). Kendall’s correlation coefficient τb was used to assess changes in smoking behavior between pregnancies. With a correlation coefficient τb of 0.365 (p = 0.00), no strong correlation was registered between smoking during the first and the second pregnancies. The analyses of three consecutive pregnancies revealed a moderately positive correlation (τb = 0.611, p = 0.00) between the first and the second pregnancies but an insignificant weak positive correlation between the second and the third pregnancies (τb 0.394, p = 0.106). Neonatal outcome Of 952 preterm infants, 194 (20.4 %) were born to smoking mothers. In the risk analysis, the risk of bearing preterm infants was just slightly higher among smokers than among non-smoking women (9.6 vs. 8.3 %). The relative risk was 1.15 (95 % CI 0.99–1.34), and the number of cigarettes needed to cause harm was 12.9 (95 % CI −832.4–37.2). The χ2 test for 1 df was 3.49 (p = 0.0619).

rs. In the risk analysis, the risk of bearing preterm infants was just slightly higher among smokers than among non-smoking women (9.6 vs. 8.3 %). The relative risk was 1.15 (95 % CI 0.99–1.34), and the number of cigarettes needed to cause harm was 12.9 (95 % CI −832.4–37.2). The χ2 test for 1 df was 3.49 (p = 0.0619). Of 1424 small for gestational age (SGA) children, 27.2 % (n = 387) were born to smoking mothers. In the risk analysis, the risk of bearing an SGA infant was 19.21 % (95 % CI 17.51–21) for smokers in contrast to 11.42 % for non-smokers (95 % CI 10.78–12.10). Thus, the relative risk of a smoker having a low-birthweight child increased by 1.68 (95 % CI 1.51–1.87), and the number of cigarettes needed to cause harm was 12.9 (95 % CI 10.4–16.38). This becomes significant in the χ2 test, with a value of 89.23 for 1 df (p < 0.0001). The number of SGA children did not change statistically in the study period. The odds ratio for the decline was 1.033 (95 % CI 0.966–1.041, p = 0.887) among smokers and 1.033 (95 % CI 0.968–1.104, p = 0.328) among non-smokers.

Of 1424 small for gestational age (SGA) children, 27.2 % (n = 387) were born to smoking mothers. In the risk analysis, the risk of bearing an SGA infant was 19.21 % (95 % CI 17.51–21) for smokers in contrast to 11.42 % for non-smokers (95 % CI 10.78–12.10). Thus, the relative risk of a smoker having a low-birthweight child increased by 1.68 (95 % CI 1.51–1.87), and the number of cigarettes needed to cause harm was 12.9 (95 % CI 10.4–16.38). This becomes significant in the χ2 test, with a value of 89.23 for 1 df (p < 0.0001). The number of SGA children did not change statistically in the study period. The odds ratio for the decline was 1.033 (95 % CI 0.966–1.041, p = 0.887) among smokers and 1.033 (95 % CI 0.968–1.104, p = 0.328) among non-smokers. Discussion This study, conducted at a tertiary perinatal center in Vienna, revealed that 18.1 % of pregnant women smoked during pregnancy; 19.1 % smoked in 2007, the highest prevalence of 19.3 % was registered in 2008/2009, and a steady decline was observed thereafter, culminating in 15.6 % at the end of the study period in 2012. A survey conducted in 2002, comprising 576 pregnant women at the same hospital, had shown that 21.5 % smoked during pregnancy (unpublished data). Thus, the prevalence of smoking a few years before the commencement of the present investigation appears to have been even higher.

and lack of responsibility at this age, which might have led to pregnancy in the first place [7, 12]. This may well be true in view of the fact that the smoking prevalence among pregnant women below the age of 20 years, registered in the present study, was approximately 20 % higher than that in the general population. The present study revealed no association between age and the number of cigarettes smoked per day. The authors of previous studies mention that heavy smokers were largely older mothers because of their prolonged exposure to tobacco before pregnancy and therefore greater addiction to nicotine [9, 15]; both of these studies comprised much larger sample sizes. We registered a reduction in smoking rates during pregnancy with the number of births. A positive correlation was noted between smoking during the first and second pregnancies but not between the second and third pregnancies. This might indicate rising awareness of the harmful effects of smoking on the part of the 15 women who experienced three pregnancies during the study period. However, it may also have been due to underreporting, secondary to social stigma experienced in previous pregnancies. Although the present study did indicate a higher risk of preterm births in conjunction with smoking, the increased risk was not statistically significant. This may have been due to the sample size; the majority of studies addressing preterm births and maternal smoking comprised much larger sample sizes [3, 6, 13].

We registered a reduction in smoking rates during pregnancy with the number of births. A positive correlation was noted between smoking during the first and second pregnancies but not between the second and third pregnancies. This might indicate rising awareness of the harmful effects of smoking on the part of the 15 women who experienced three pregnancies during the study period. However, it may also have been due to underreporting, secondary to social stigma experienced in previous pregnancies. Although the present study did indicate a higher risk of preterm births in conjunction with smoking, the increased risk was not statistically significant. This may have been due to the sample size; the majority of studies addressing preterm births and maternal smoking comprised much larger sample sizes [3, 6, 13]. SGA is another birth outcome related to maternal smoking during pregnancy [24]. The present study clearly showed a higher risk of SGA among smoking mothers; the risk was 7.8 % higher and the number of cigarettes needed to cause harm (in respect of giving birth to an SGA child) was 12.9.

Although the present study did indicate a higher risk of preterm births in conjunction with smoking, the increased risk was not statistically significant. This may have been due to the sample size; the majority of studies addressing preterm births and maternal smoking comprised much larger sample sizes [3, 6, 13]. SGA is another birth outcome related to maternal smoking during pregnancy [24]. The present study clearly showed a higher risk of SGA among smoking mothers; the risk was 7.8 % higher and the number of cigarettes needed to cause harm (in respect of giving birth to an SGA child) was 12.9. Although maternal smoking clearly declined in the study period (19.1 to 15.6 %), the decline in SGA children was not statistically significant. This may have been due to the transfer of high-risk pregnancies to the perinatal center and their impact on birth outcome statistics. Another explanation could be that smoking, in fact, did not decrease and the lower figures indirectly reflected underreporting in the last few years. As there have been no intensive campaigns denouncing smoking in Austria and smoking is commonly regarded as a habit rather than a health-damaging phenomenon, the underreporting thesis seems unlikely. Currently, we have no data concerning changes in underreporting rates.

y reflected underreporting in the last few years. As there have been no intensive campaigns denouncing smoking in Austria and smoking is commonly regarded as a habit rather than a health-damaging phenomenon, the underreporting thesis seems unlikely. Currently, we have no data concerning changes in underreporting rates. The strength of the present study is its large sample size; 11,142 complete maternal records comprising self-reported smoking behavior, maternal characteristics, and neonatal outcome were studied. The most relevant limitation is that the information on smoking status is based on self-reporting alone. This may have led to underreporting because of the stigma associated with smoking, especially during pregnancy. Furthermore, underreporting may have been common in women who had experienced previous pregnancies with unfavorable outcomes and had been told that smoking cessation may prevent problems in the forthcoming pregnancy. Although a number of authors registered rather high discrepancies in self-reported smoking status and cotinine measurement [22], many of them agree that self-reporting is a reliable means of determining smoking status during pregnancy [14]. In a Spanish study [4], the authors observed accurate self-reporting (3.9 % misreporting) and a very similar overall prevalence of smoking as in the present study (18.5 %).

atus and cotinine measurement [22], many of them agree that self-reporting is a reliable means of determining smoking status during pregnancy [14]. In a Spanish study [4], the authors observed accurate self-reporting (3.9 % misreporting) and a very similar overall prevalence of smoking as in the present study (18.5 %). Another limitation of the present study is that no socioeconomic data were collected. However, given the fact that the entire investigation was performed in a single region and the same hospital setting, socioeconomic factors would probably have not changed considerably during the study years. In contrast to many countries, the absence of social security or access to prenatal and perinatal care is no hindrance in Austria; 98 % of the population have excellent social security coverage [20]. The high density of hospitals providing prenatal and perinatal care and the “mother-and-child card” (Mutter-Kind-Pass) is an established social security measure. This mother-and-child card includes free but mandatory examinations during pregnancy and until the child’s second birthday. The first mandatory examination must be completed before the 16th gestational week. This checkup is taken quite seriously because of its ensuing health benefits and child support payment.

measure. This mother-and-child card includes free but mandatory examinations during pregnancy and until the child’s second birthday. The first mandatory examination must be completed before the 16th gestational week. This checkup is taken quite seriously because of its ensuing health benefits and child support payment. A further limitation of the study is that the women’s smoking status was only documented once during pregnancy. For various reasons—such as the woman’s condition being graded as a high-risk pregnancy some time later during her pregnancy—this documentation was not performed at the same time point in all women. The time of documentation varied from the first trimester to the late third trimester. However, most of the information was obtained from early pregnancy. Data concerning the cessation of smoking were not registered. A meta-analysis by the CDC showed high variability in cessation rates [25]. The present study did not address smoking cessation, although cessation rates would probably have not altered the results substantially.

nformation was obtained from early pregnancy. Data concerning the cessation of smoking were not registered. A meta-analysis by the CDC showed high variability in cessation rates [25]. The present study did not address smoking cessation, although cessation rates would probably have not altered the results substantially. Conclusion Once inquired, self-reported smoking during pregnancy declined from 19.1 to 15.6 % in the study years. Urine cotinine was not measured. Although the prevalence of smoking among young women is still alarmingly high, the largest number of smoking women is in the 22–31-year age category. Hence, both age groups should be the focus of attention of anti-smoking campaigns targeted at pregnant women. Primary educational programs should focus on the youth in order to prevent addiction at an early age. Further educational programs, including the use of social media and online courses, should be implemented. Abbreviations CIConfidence interval HBSCHealth behavior in school-aged children NICUNeonatal Intensive Care Unit OB/GynObstetrics and gynecology OROdds ratio SDStandard deviation SGASmall for gestational age SMZSozialmedizinisches Zentrum WHOWorld Health Organization Open access funding provided by Medical University of Vienna, Austria. We thank Professor Georg Heinze for his support in planning the statistical analysis and the obstetric department at SMZ Ost Hospital for providing the data.

SDStandard deviation SGASmall for gestational age SMZSozialmedizinisches Zentrum WHOWorld Health Organization Open access funding provided by Medical University of Vienna, Austria. We thank Professor Georg Heinze for his support in planning the statistical analysis and the obstetric department at SMZ Ost Hospital for providing the data. Author’s contribution Angelika Schultze, MD: at the time of the study, she was medical student at Medical University of Vienna. Data collection, methods, statistical analysis, and manuscript writing Herbert Kurz, MD: aims of the study, study planning, and manuscript writing Ingrid Stümpflen, BSc: data acquisition, patient documentation, and data management Erich Hafner, MD, Prof., head of obstetric clinic: data acquisition, documentation, and manuscript writing (perinatological aspects) Compliance with ethical standards Angelika Schultze declares that she has no conflict of interest. Herbert Kurz declares that he has no conflict of interest. Ingrid Stümpflen and Erich Hafner declare that they have no conflict of interest. Ethical approval Ethical approval has been obtained from the Ethical Committee of Vienna. Patients gave informed consent to the collection of their data.

Introduction Functional defecation disorders (FDDs) are common in children with an estimated prevalence of functional constipation (FC) in children worldwide ranging from 0.7 to 29.6 % [18]. FDD include FC, FC with fecal incontinence (FC + FI), and functional nonretentive fecal incontinence (FNRFI) [23]. The main characteristics of FC are infrequent, and hard and painful defecations often accompanied by the involuntary loss of feces in the underwear. FDDs seem to be particularly common in children with behavioral and developmental disorders, such as attention deficit hyperactivity disorder (ADHD) and autism spectrum disorders (ASDs) [7, 14, 16]. In an effort to better understand the association between FDDs and ASD, Peeters et al. used two validated ASD screening surveys (Social Responsiveness Scale (SRS) and the Social Communication Questionnaire–Lifetime (SCQ-L)) to prospectively screen children with functional defecation disorders (FDDs) [20]. They found that 29 % of children with FDDs had ASD symptoms. It is uncertain if these children indeed had ASD, since study patients were not referred for behavioral and psychological assessment.

munication Questionnaire–Lifetime (SCQ-L)) to prospectively screen children with functional defecation disorders (FDDs) [20]. They found that 29 % of children with FDDs had ASD symptoms. It is uncertain if these children indeed had ASD, since study patients were not referred for behavioral and psychological assessment. ASD is a neurodevelopmental disorder, defined by persistent deficits in social communication and interaction, as well as by restricted and repetitive patterns of behaviors and activities according to the Diagnostic and Statistical Manual of mental disorders–Fifth Edition (DSM-5) [1]. The diagnosis of ASD cannot be based on high rates of ASD symptoms in screening surveys; a full child psychiatric assessment is needed. Whether a positive ASD screen in children with FDD accurately identifies children with ASD is unknown. Our primary objective was therefore to prospectively assess whether positive ASD screening surveys (SRS and SCQ-L) in children with FDDs accurately identify ASD. While the SRS and SCQ-L [8, 20, 24] are often used to screen for ASD in children, the results from these screens may be influenced by the presence of internalizing and externalizing behaviors not associated with an ASD [13]. Interestingly, studies assessing psychosocial factors associated with FDDs in children have found an increase of both internalizing and externalizing behaviors in these children [22, 28]. The presence of these behaviors, in particular externalizing behaviors, was associated with differences in response to treatments for FDDs [28].

y, studies assessing psychosocial factors associated with FDDs in children have found an increase of both internalizing and externalizing behaviors in these children [22, 28]. The presence of these behaviors, in particular externalizing behaviors, was associated with differences in response to treatments for FDDs [28]. Our secondary aims were to evaluate if other DSM-5 disorders are present in children with positive ASD screenings surveys, but not having ASD, as well as whether results of ASD screens varied among various FDD diagnoses and between responders and nonresponders 1 year after recruitment. Methods A prospective cohort study was carried out between August 2012 and February 2013. Children (4–12 years), who met criteria for a diagnosis of FC, FC with FI, or FNRFI based upon Rome III criteria [23], were recruited from the tertiary outpatient clinic of pediatric gastroenterology of the Nationwide Children’s Hospital (Columbus, OH). Patients were included regardless of previous treatments for constipation or previously diagnosed behavioral and developmental health problems. Patients were excluded if they had a medical disease that could have contributed to the development of constipation, such as inflammatory bowel disease, celiac disease, congenital anorectal malformation, history of colonic surgery, cerebral palsy, and spina bifida. After informed consent was obtained, parents of caregivers were asked to complete the two ASD screening surveys: SRS and SCQ-L. This study was approved by the Nationwide Children’s Hospital Institutional Review Board.

ease, congenital anorectal malformation, history of colonic surgery, cerebral palsy, and spina bifida. After informed consent was obtained, parents of caregivers were asked to complete the two ASD screening surveys: SRS and SCQ-L. This study was approved by the Nationwide Children’s Hospital Institutional Review Board. Social Responsiveness Scale The SRS is a 65-item scale that requires parents to rate the child’s behaviors in the previous 6 months in 4- to 18-year olds. The questionnaire focuses on behaviors grouped into categories of social cognition, social motivation, social communication, social awareness, and autistic mannerisms [9]. Each question is rated on a four-point Likert scale ranging from 0 (never true) to 3 (almost always true), generating a total score in the range from 0 to 195. Total raw scores can be transformed into gender-normed T-scores with higher scores indicating a greater severity of behavioral symptoms. A total score ≥51 is suggestive for the presence of ASD [8]. Social Communication Questionnaire–Lifetime The SCQ-L consists of 40 yes/no questions to evaluate behaviors associated with social interaction, communication, abnormal language, and stereotypical behaviors. Patients should be over 4 years, with a minimum mental age of 2 years. Total scores range from 0 to 40. A total score of ≥15 is suggestive for the presence of ASD [24].

onsists of 40 yes/no questions to evaluate behaviors associated with social interaction, communication, abnormal language, and stereotypical behaviors. Patients should be over 4 years, with a minimum mental age of 2 years. Total scores range from 0 to 40. A total score of ≥15 is suggestive for the presence of ASD [24]. Behavioral evaluations Children who met the preestablished cutoff scores for one or both questionnaires were referred to a clinical psychologist or clinical social worker from the Child Developmental Center at Nationwide Children’s Hospital for a comprehensive behavioral health diagnostic assessment. Patients were not only evaluated for whether they met criteria for ASD but also for other DSM-5 behavioral disorders. Patients were only included when all questionnaires were completed. Clinical course A year after enrollment, follow-up surveys were mailed to the home of all participants with a self-addressed stamped envelope for returning the survey to clinical investigators. The survey contained the questions about bowel movements from the Rome III Diagnostic Questionnaire for the Pediatric Functional GI Disorders that were utilized for making the initial FDD diagnoses. The survey also asked if the children were taking medications or supplements for treatment of constipation. Based upon the responses, children were categorized as those whose bowel patterns still met criteria for an FDD and those whose bowel patterns no longer met criteria for an FDD.

making the initial FDD diagnoses. The survey also asked if the children were taking medications or supplements for treatment of constipation. Based upon the responses, children were categorized as those whose bowel patterns still met criteria for an FDD and those whose bowel patterns no longer met criteria for an FDD. Outcomes The primary outcome was the number of children with ASD diagnoses, confirmed by psychological evaluation, within the group of positive SRS and SCQ-L scores. Secondary outcome was the number of children with positive SRS and SCQ-L scores with other DSM disorders, as well as whether differences exist in scores of ASD screening surveys among different FDD diagnoses and between the outcome groups of responders and nonresponders. We hypothesized that if the severity of abnormal behavior is associated with fecal incontinence, scores for SRS and SCQ-L would be higher for those with FC+FI and/or FNRFI than those with FC. In addition, if abnormal behaviors are associated with poorer prognosis, then one would expect SRS and SCQ-L scores to be higher for those who still met criteria for an FDD (nonresponders) as compared to those who were improved (responders) a year after enrollment in the study and no longer met criteria for an FDD.

FC. In addition, if abnormal behaviors are associated with poorer prognosis, then one would expect SRS and SCQ-L scores to be higher for those who still met criteria for an FDD (nonresponders) as compared to those who were improved (responders) a year after enrollment in the study and no longer met criteria for an FDD. Statistical analyses For statistical analyses, Stata S.E. version 13.1 was used. Total raw scores on the SRS and total raw scores on the SCQ-L were analyzed as continuous variables. Independent sample t tests were used to examine differences in means in continuous variables across groups of children who improved compared to those that did not. Chi-squared analyses with Fisher’s exact tests were performed to compare proportions of categorical variables. The significance level was set at p < 0.05. Results Characteristics of the study sample A total of 127 consecutive patients referred for evaluation and treatment of FDDs were asked to participate. Thirty patients were excluded, because they did not meet inclusion criteria (n = 16) or parents were not interested to participate (n = 14). Parents of 97 patients completed both ASD screening questionnaires (Table 1). Among the 97 study participants, 63 % (n = 61), 31 % (n = 30), and 6 % (n = 6) fulfilled the Rome III criteria for FC, FC+FI, and FNRFI, respectively.Table 1 Patient characteristics

16) or parents were not interested to participate (n = 14). Parents of 97 patients completed both ASD screening questionnaires (Table 1). Among the 97 study participants, 63 % (n = 61), 31 % (n = 30), and 6 % (n = 6) fulfilled the Rome III criteria for FC, FC+FI, and FNRFI, respectively.Table 1 Patient characteristics Total Negative screens Positive SRS Positive SCQ-L Number of participants 97 70 27 8 Gender, n (%) Male 53 (55) 39 (56) 14 (52) 6 (75) Female 44 (45) 31 (44) 13 (48) 2 (25) Mean age in years (SD) 7.8 (±3.0) 8 (±3.2) 7.4 (±2.3) 7.3 (±2.3) Diagnosis FDD, n (%) FC 30 (31) 25 (36) 5 (9) 0 (0) FC with FI 61 (63) 40 (57) 21 (78) 8* (100) FNRFI 6 (6) 5 (7) 1 (4) 0 (0) Diagnoses of ASD Prior diagnosis of ASD 1 0 1 1 ASD suspected 1 0 1 1 Previously diagnosed behavior disorder (parent reported) 14 4 10# 2 Taking medication for behavioral issues 8 3 5** 2 As compared to participants with negative screens, higher percentage of positive SCQ-L participants were diagnosed with FC with FI (*p value = 0.024) and higher percentage of positive SRS participants had behavioral disorders (#p value = 0.000) and took medications for behavioral issues (**p value = 0.036) ASD autism spectrum disorder, FC functional constipation, FC with FI functional constipation with fecal incontinence, FNRFI functional nonretentive fecal incontinence, SRS Social Responsiveness Scale, SCQ-L Social Communication Questionnaire–Lifetime

Total Negative screens Positive SRS Positive SCQ-L Number of participants 97 70 27 8 Gender, n (%) Male 53 (55) 39 (56) 14 (52) 6 (75) Female 44 (45) 31 (44) 13 (48) 2 (25) Mean age in years (SD) 7.8 (±3.0) 8 (±3.2) 7.4 (±2.3) 7.3 (±2.3) Diagnosis FDD, n (%) FC 30 (31) 25 (36) 5 (9) 0 (0) FC with FI 61 (63) 40 (57) 21 (78) 8* (100) FNRFI 6 (6) 5 (7) 1 (4) 0 (0) Diagnoses of ASD Prior diagnosis of ASD 1 0 1 1 ASD suspected 1 0 1 1 Previously diagnosed behavior disorder (parent reported) 14 4 10# 2 Taking medication for behavioral issues 8 3 5** 2 As compared to participants with negative screens, higher percentage of positive SCQ-L participants were diagnosed with FC with FI (*p value = 0.024) and higher percentage of positive SRS participants had behavioral disorders (#p value = 0.000) and took medications for behavioral issues (**p value = 0.036) ASD autism spectrum disorder, FC functional constipation, FC with FI functional constipation with fecal incontinence, FNRFI functional nonretentive fecal incontinence, SRS Social Responsiveness Scale, SCQ-L Social Communication Questionnaire–Lifetime Initial medical histories revealed that 16 of the 97 (16.5 %) participants presented with a previously diagnosed or suspected developmental and/or behavioral disorder by medical providers from other institutions. One patient presented with an ASD diagnosis of pervasive developmental disorder—not otherwise specified (PDD-NOS). An ASD diagnosis was suspected in another participant. One or more behavioral disorders were reported in another 14 patients, including ADHD (n = 9), oppositional defiant disorder (ODD) (n = 5), anxiety disorders (n = 4), obsessive compulsive disorder (OCD) (n = 1), and disruptive behavior disorder–NOS (n = 1) (see “previously diagnosed behavior disorders” in Table 2). Of the 14 patients with a previously diagnosed disorder not fitting under ASD, 8 were taking medications for their behavioral issues.Table 2 Results of behavioral evaluation

essive compulsive disorder (OCD) (n = 1), and disruptive behavior disorder–NOS (n = 1) (see “previously diagnosed behavior disorders” in Table 2). Of the 14 patients with a previously diagnosed disorder not fitting under ASD, 8 were taking medications for their behavioral issues.Table 2 Results of behavioral evaluation Patient Gender Age Rome III diagnosis SRS score SCQ-L score Previously diagnosed behavior disorder Behavior diagnosis based on study evaluation 1 F 5 FC with FI 112 29 PDD-NOS PDD-NOS 2 M 10 FC with FI 96 17 None Asperger’sa, ADHDa, DBD-NOSa 3 F 4 FC with FI 92 16 ASD (suspected) 22q11.2 deletiona, learning disordera, DBD-NOSa 4 M 8 FC with FI 124 18 ADHD, OCD, GAD, ODD ADHD, ODD 5 M 6 FC with FI 141 22 ADHD, ODD, PTSD ADHD, ODD, PTSD 6 M 9 FC with FI 127 21 None ADHDa, DBD-NOSa 7 M 6 FC with FI 102 16 None Learning disordera 8 M 10 FC with FI 104 19 None No diagnosis 9 M 10 FC with FI 71 3 ADHD, ODD ADHD, ODD 10 F 5 FC with FI 60 5 GAD GAD, separation anxiety disordera 11 F 5 FNRFI 84 9 None DBD-NOSa 12 F 5 FC with FI 59 4 None ADHD, ODDa 13 F 8 FC with FI 54 6 None Anxiety disorder NOSa 14 F 7 FC with FI 83 8 None Adjustment disordera, ADHDa 15 F 8 FC with FI 91 6 None ADHDa, anxiety disordera Values in italics are positive SCQ-L scores

8 M 10 FC with FI 104 19 None No diagnosis 9 M 10 FC with FI 71 3 ADHD, ODD ADHD, ODD 10 F 5 FC with FI 60 5 GAD GAD, separation anxiety disordera 11 F 5 FNRFI 84 9 None DBD-NOSa 12 F 5 FC with FI 59 4 None ADHD, ODDa 13 F 8 FC with FI 54 6 None Anxiety disorder NOSa 14 F 7 FC with FI 83 8 None Adjustment disordera, ADHDa 15 F 8 FC with FI 91 6 None ADHDa, anxiety disordera Values in italics are positive SCQ-L scores ASD autism spectrum disorder, FC functional constipation, FC with FI functional constipation with fecal incontinence, FNRFI functional nonretentive fecal incontinence, PDD-NOS pervasive developmental disorder–NOS, ADHD attention deficit hyperactivity disorder, OCD obsessive compulsive disorder, GAD generalized anxiety disorder, ODD oppositional defiant disorder, PTSD posttraumatic stress disorder, DBD-NOS disruptive behavioral disorder–NOS aNew diagnosis The majority of patients was referred (n = 83) by their primary care physicians, five by the emergency department, and the other nine by other specialists within the hospital; none was referred by another pediatric gastroenterologist.

ASD autism spectrum disorder, FC functional constipation, FC with FI functional constipation with fecal incontinence, FNRFI functional nonretentive fecal incontinence, PDD-NOS pervasive developmental disorder–NOS, ADHD attention deficit hyperactivity disorder, OCD obsessive compulsive disorder, GAD generalized anxiety disorder, ODD oppositional defiant disorder, PTSD posttraumatic stress disorder, DBD-NOS disruptive behavioral disorder–NOS aNew diagnosis The majority of patients was referred (n = 83) by their primary care physicians, five by the emergency department, and the other nine by other specialists within the hospital; none was referred by another pediatric gastroenterologist. Autism spectrum disorder screening questionnaires Among the 97 study participants, 27 (27.8 %) scored above the cutoff value for one of the two ASD screening questionnaires (Table 1). All 27 had SRS scores ≥51, and 8 of the 27 had a SCQ-L ≥15. The most common FDD among the 27 with a positive SRS was FC+FI (n = 21), followed by FC (n = 5) and FNFRI (n = 1). All eight participants with a positive SCQ-L had FC+FI. The distribution of FDD diagnosis did not significantly differ between patients with positive and negative screening questionnaires (p = 0.168). Compared to those who scored <51, children with SRS scores ≥51 had a higher percentage of previously diagnosed behavioral disorders and of medications use for behavioral issues (Table 1). As shown in Table 1, four patients with previously known behavioral disorders scored negative on ASD screening questionnaires.

168). Compared to those who scored <51, children with SRS scores ≥51 had a higher percentage of previously diagnosed behavioral disorders and of medications use for behavioral issues (Table 1). As shown in Table 1, four patients with previously known behavioral disorders scored negative on ASD screening questionnaires. All 27 children with positive ASD questionnaires were referred for further evaluation. Parents of 12 patients were not interested in a referral and were therefore lost to follow-up. Half of these 12 patients were previously diagnosed with behavior disorders: 3 patients had ADHD and used medication, 1 was diagnosed with disruptive behavior disorder–NOS, and 2 had an anxiety disorders, of which 1 was seeing a therapist. Results of behavior evaluations Subsequently, 15 participants (55.6 %) were seen by a psychologist or clinical social worker to undergo evaluations for behavioral and developmental disorders (Table 2). Previously known diagnoses The histories obtained from the parents/caretakers for these 15 children revealed that providers from other institutions had previously diagnosed 5 with a developmental or behavioral disorder (1 with ASD and 4 with behavioral disorders). A primary care physician had recently referred one child for evaluation of a suspected ASD. The remaining nine had never been diagnosed with or previously suspected to have a behavioral or developmental disorder (Table 2).

nosed 5 with a developmental or behavioral disorder (1 with ASD and 4 with behavioral disorders). A primary care physician had recently referred one child for evaluation of a suspected ASD. The remaining nine had never been diagnosed with or previously suspected to have a behavioral or developmental disorder (Table 2). New or confirmed diagnoses by developmental evaluation Developmental evaluations found that 14 of the 15 patients met criteria for developmental, behavioral, and/or learning disorders. New diagnoses were made in ten patients. However, developmental evaluations demonstrated that only 2 of the 15 children met criteria for an ASD. One of the two was the participant who was previously diagnosed with ASD at an outside institution. The other one had never been suspected of having an ASD and was diagnosed with Asperger’s disorder. As for the child who was previously suspected to have an ASD, the evaluation resulted in a diagnosis of disruptive behavioral disorder–NOS and a learning disorder. In addition, genetic testing discovered a 22q11 deletion.

he other one had never been suspected of having an ASD and was diagnosed with Asperger’s disorder. As for the child who was previously suspected to have an ASD, the evaluation resulted in a diagnosis of disruptive behavioral disorder–NOS and a learning disorder. In addition, genetic testing discovered a 22q11 deletion. Next to the two ASD diagnoses, positive ASD screens led to diagnosis of various types of other disorders. Disorders of externalizing behaviors were identified in ten patients, including ADHD (n = 8), ODD (n = 4), and disruptive behaviors–NOS (n = 4). Anxiety disorders were identified in three patients, including generalized anxiety disorder (n = 1), anxiety disorder (n = 2), and separation anxiety (n = 1). Other diagnosis made were learning disorders (n = 2), adjustment disorder (n = 1), and posttraumatic stress disorder (n = 1)(Table 2). Scores according to the functional defecation disorder The average scores for SRS (49.2 ± 33.7) and SCQ-L (6.7 ± 3.4) of children diagnosed with FC+FI were statistically higher than the average scores for SRS (26.3 ± 22.3, p = 0.0011) and SCQ-L (3.4 ± 2.8, p = 0.0053) of children diagnosed with FC (Fig. 1). The average scores of children with FNRFI did not differ significantly from the average scores of either FC+FI or FC groups of children. Of note, all eight participants who were positive for both SRS and SCQ were diagnosed with FC+FI.Fig. 1 Scores ASD screening questionnaires compared to FDD diagnosis

with FC (Fig. 1). The average scores of children with FNRFI did not differ significantly from the average scores of either FC+FI or FC groups of children. Of note, all eight participants who were positive for both SRS and SCQ were diagnosed with FC+FI.Fig. 1 Scores ASD screening questionnaires compared to FDD diagnosis Scores according to clinical course—1-year follow-up Among the initial 97 participants, parents/caregivers for 52 (53.6 %) responded to the survey sent them 1 year after enrollment. Parental responses to questions on bowel patterns indicated that bowel patterns for 26 of the 52 (50 %) no longer met criteria for an FDD while the bowel patterns for the other 26 (50 %) continued to meet criteria for an FDD. Of the 26 whose bowel patterns no longer met criteria for an FDD, 9 reported that they were taking medications for constipation while 17 reported not being on any laxatives/supplements for constipation. Of the 26 who still met criteria for an FDD, 17 reported that they were taking medications for constipation while 9 reported not being on any laxatives/supplements for constipation. Comparison of scores between groups revealed that at the time of enrollment the average scores for the SRS (46.7 ± 30.8 vs. 31.6.2 ± 27.6, p = 0.034, one-tailed test) and the SCQ-L (6.3 ± 4.7 vs. 4.2 ± 4.2, p = 0.049, one-tailed test) were significantly higher in those who still met criteria for an FDD after 1 year as compared to those who no longer met criteria for an FDD after 1 year.

llment the average scores for the SRS (46.7 ± 30.8 vs. 31.6.2 ± 27.6, p = 0.034, one-tailed test) and the SCQ-L (6.3 ± 4.7 vs. 4.2 ± 4.2, p = 0.049, one-tailed test) were significantly higher in those who still met criteria for an FDD after 1 year as compared to those who no longer met criteria for an FDD after 1 year. Discussion The 27.8 % positive ASD screens in the current study population of children with FDDs closely match previous findings [20]. However, formal psychological evaluations found that positive SRS and SCQ-L screens were not able to correctly identify an ASD diagnosis in our study population, but did identify children with previously undiagnosed behavioral disorders. Findings from the current study indicate that children with FDDs display behaviors that are common to ASD and other behavioral disorders. The use of ASD screening surveys resulted in ten new DSM diagnosis that were unknown or unrecognized by parents and previous healthcare professionals. Furthermore, our findings indicate that children with FC+FI have a higher severity of abnormal behaviors that result in higher survey scores than children with other FDDs. In addition, comparison of scores indicate a higher prevalence of abnormal behaviors in children who do not improve with standard medical interventions as compared to those that do improve.

hildren with FC+FI have a higher severity of abnormal behaviors that result in higher survey scores than children with other FDDs. In addition, comparison of scores indicate a higher prevalence of abnormal behaviors in children who do not improve with standard medical interventions as compared to those that do improve. Findings from the current study, along with previous reports, provide strong evidence that many children with FC+FI share a common behavioral phenotype that results in positive ASD screens. Prior studies have found that the presence of increased internalizing and externalizing behaviors are associated with higher SRS scores [13]. Attention deficit and disruptive behavior disorders, which were the most common group of disorders found in the current study, are disorders of externalizing behaviors. Externalizing behaviors have also been associated with fecal incontinence in children with ASD [21] and in the general pediatric population [15, 28]. Many of the abnormal behaviors for which the SRS and SCQ-L screen commonly occur not only in children with ASD but also in children with other behavior disorders. Towbin et al. [25] demonstrated that a substantial proportion of children with mood and anxiety disorders scored in the ASD-likely range of the SRS. Similarly, children with ADHD may find some aspects of social interaction and communication more difficult than healthy children [17]. In addition, when ASD is compared to disruptive behavior disorders, the sensitivity and specificity of the SRS is considerably lower [17].

ers scored in the ASD-likely range of the SRS. Similarly, children with ADHD may find some aspects of social interaction and communication more difficult than healthy children [17]. In addition, when ASD is compared to disruptive behavior disorders, the sensitivity and specificity of the SRS is considerably lower [17]. While use of SRS and SCQ-L may not accurately identify ASD in children with FDDs, the current study indicates, in accordance with previous studies [20, 28], that behavior problems are common in children with FDDs. Even though this study could overestimate the prevalence of behavior disorders in children with FDDs, due to the fact that parents concerned for the presence of behavior problems in their children are more likely to participate in this study, we believe that behavioral screening should be incorporated into the diagnostic workup of children with FDDs. Since screening could result in early diagnosis and treatment of previously unknown behavior disorders and may have therapeutic consequences. Based on our results, we cannot advise which screening surveys should be used. We can only speculate that it might be better to use more general surveys that screen for both internalizing and externalizing disorders, rather than ASD-specific screens. However, future research is necessary to answer this question.

ences. Based on our results, we cannot advise which screening surveys should be used. We can only speculate that it might be better to use more general surveys that screen for both internalizing and externalizing disorders, rather than ASD-specific screens. However, future research is necessary to answer this question. In addition to being associated with a particular type of FDD, the results from the 1-year follow-up survey indicate that behaviors that the ASD screens measure are associated with poorer outcomes. This observation confirms previous reports that indicate that increased behavioral and social problems are associated with the need for longer treatment and poorer outcome in children with FDDs [16, 28, 29]. Identifying previously unrecognized behavioral disorders would provide opportunity to address such issues that may be contributing to the poorer response to routine medical interventions in this patient population. However, it is possible that the poorer outcomes are related to noncompliance with FDD treatment regimens.

dentifying previously unrecognized behavioral disorders would provide opportunity to address such issues that may be contributing to the poorer response to routine medical interventions in this patient population. However, it is possible that the poorer outcomes are related to noncompliance with FDD treatment regimens. While prior studies have shown mixed results, the overall use of behavioral interventions in addition to common laxative therapies appear to produce slightly higher rates of improvement in the treatment of FDD, particularly when treating FC+FI [2, 4, 6, 19, 27, 29]. However, even in studies where combined behavioral and laxative therapy produced the best outcomes over other treatment regiments, the rates of cure at 12-month follow-up are at best 51 % [4, 6, 19]. In this study, we also reported that 50 % no longer met the Rome III criteria for FDDs at follow-up. Such disappointing cure rates suggest the need to better understand the clinical characteristics of children at risk of not responding to standard behavioral and medical therapies. The use of behavioral screens such as those in this study may not only help clarify specific behavioral phenotypes associated with FDDs, but may also lead to the development of specific behavioral and/or cognitive therapies needed to improve current outcome rates for FDDs. The relatively ineffectiveness of psychological treatment for FDDs could also be explained by the focus of behavioral therapy on fear of defecation. Based on the assumption that constipation often has to do with fear of defecation and consequently to withholding of stools, behavior therapy is mainly focused on these problems. Findings from current study along with previously published reports showed that children with FDDs also have externalizing behavior problems. Treatment results might improve if behavioral treatments are focused on both internalizing and externalizing behavior problems.

avior therapy is mainly focused on these problems. Findings from current study along with previously published reports showed that children with FDDs also have externalizing behavior problems. Treatment results might improve if behavioral treatments are focused on both internalizing and externalizing behavior problems. To date, it has remained uncertain whether behavioral problems are primary or secondary to FDDs. It can be hypothesized that preexisting behavior problems may lead to a complicated toilet training, which is known as a critical phase in the development of FDDs [5]. On the other hand, the presence of FDDs might give rise to a considerable level of stress and embarrassment for the child and family. Fecal incontinence triggers parental stress because of the dishonesty of the child about fecal accidents and the burden of cleaning clothes. In addition, the majority of parents assume that fecal incontinence is caused by the child’s laziness, carelessness, and stubbornness [3, 10, 11, 26]. This influences the parent-child interaction negatively and could therefore be related with the onset or maintenance of FDDs in children.

the burden of cleaning clothes. In addition, the majority of parents assume that fecal incontinence is caused by the child’s laziness, carelessness, and stubbornness [3, 10, 11, 26]. This influences the parent-child interaction negatively and could therefore be related with the onset or maintenance of FDDs in children. This study has several strengths. Diagnoses of FDDs in children were made according to the internationally accepted Rome III criteria [23]. Additionally, the diagnosis of ASD and other DSM diagnoses were made after behavioral and psychological assessment by experienced behavioral health clinicians. Nevertheless, there are limitations that should be taken into account when interpreting our results and extrapolating them to other patient cohorts. First, there could have been a selection bias toward more severe cases of FDDs because the study patients were recruited in a tertiary care center. This could have resulted in a higher percentage of behavioral health problems as children presenting at a tertiary care center might be more difficult to treat. Care should therefore be taken before generalizing these results to children who have FDDs and are treated in a general pediatric practice. However, most of prior healthcare provided was not from tertiary care centers as the majority (86 %) was referred and presenting from general pediatric practice. Another limitation was the relatively small sample size and the low response rate in the 1-year follow-up. Finally, the use of parents as informants to fill out the ASD screens and Rome III questions may be a confounder of this study, since parental report may also be biased by the parents’ own psychological or health status. However, the ASD screens used in this study have been previously validated and parent report of GI problems has been highly correlated with GI diagnoses made by physicians [8, 12, 24].

estions may be a confounder of this study, since parental report may also be biased by the parents’ own psychological or health status. However, the ASD screens used in this study have been previously validated and parent report of GI problems has been highly correlated with GI diagnoses made by physicians [8, 12, 24]. Conclusion Our findings draw attention to the fact that behavioral health problems are often unrecognized in children with FDD. While positive ASD screening surveys did not correctly identify ASD confirmed by psychological evaluation in the majority of children, it did help to identify other unrecognized behavior disorders. Presence of the behaviors uncovered by the ASD screens appears to be more common in children with FC+FI and in children with poorer responses to current medical treatments. Future studies are needed to determine whether early identification and therapy for externalizing behaviors associated with ASD and other behavioral disorders could help improve treatment outcomes for children with FDDs. Abbreviations ADHDAttention deficit hyperactivity disorder ASDAutism spectrum disorder FCFunctional constipation FDDFunctional defecation disorders FIFecal incontinence FNRFIFunctional nonretentive fecal incontinence SRSSocial Responsiveness Scale SCQ-LSocial Communication Questionnaire–Lifetime The authors thank the clinicians for their help with the inclusion of patients at the outpatient clinic.

ASDAutism spectrum disorder FCFunctional constipation FDDFunctional defecation disorders FIFecal incontinence FNRFIFunctional nonretentive fecal incontinence SRSSocial Responsiveness Scale SCQ-LSocial Communication Questionnaire–Lifetime The authors thank the clinicians for their help with the inclusion of patients at the outpatient clinic. Author’s contribution Sophie Kuizenga-Wessel: Participated in the design of the study and contributed to developing of the research protocols, collected data, drafted the initial manuscript, and approved the final manuscript as submitted. Carlo Di Lorenzo: Participated in the design of the study and contributed to developing of the research protocols, collected data, supervised drafting of the manuscript, and approved the final manuscript as submitted. Lisa Nicholson: Performed data analysis, supervised drafting of the manuscript, and approved the final manuscript as submitted. Eric Butter: Participated in the design of the study and contributed to developing of the research protocols, collected data, and approved the final manuscript as submitted. Karen Ratliff-Schaub: Participated in the design of the study and contributed to developing of the research protocols, collected data, and approved the final manuscript as submitted. Marc Benninga: Participated in the design of the study and contributed to developing of the research protocols, approved the final manuscript as submitted, supervised drafting of the manuscript, and approved the final manuscript as submitted.

Karen Ratliff-Schaub: Participated in the design of the study and contributed to developing of the research protocols, collected data, and approved the final manuscript as submitted. Marc Benninga: Participated in the design of the study and contributed to developing of the research protocols, approved the final manuscript as submitted, supervised drafting of the manuscript, and approved the final manuscript as submitted. Kent Williams: Collected data, supervised drafting of the manuscript, and approved the final manuscript as submitted. Compliance with ethical standards Funding No external funding was secured for this study. Conflict of interest The authors declare that they have no conflict of interest. Financial disclosure The authors have no financial relationships relevant to this article to disclose. Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent Informed consent was obtained from all individual participants included in the study.

Introduction Phenylketonuria (PKU; OMIM: #261600) is an inborn error of phenylalanine (Phe) metabolism with an estimated average prevalence in Europe of 1/10,000 live births [16]. Timely newborn screening and life-long Phe-restricted diet enable the severe outcomes of untreated PKU to be avoided [2]. Despite the cost-effective nature of newborn screening, it has been reported that several countries either do not offer this service or can offer it to only part of their population [3–5, 9, 10, 15, 20, 22–24, 30]. Furthermore, diagnostic and management practices for PKU vary between countries. The present study was conducted to describe the management and treatment practices for PKU in an area comprising much of South and Eastern Europe. The majority of these centres were not captured by a previous European survey [1]. Based on socioeconomic factors such as gross domestic product per capita, some of these countries might be expected to employ less comprehensive management approaches for PKU than are in use in more affluent regions of Western Europe. This survey was designed to highlight the shortcomings in PKU management and to facilitate the targeting of future initiatives.

ss domestic product per capita, some of these countries might be expected to employ less comprehensive management approaches for PKU than are in use in more affluent regions of Western Europe. This survey was designed to highlight the shortcomings in PKU management and to facilitate the targeting of future initiatives. Methods Questionnaire development A questionnaire consisting of 52 closed (answer choices provided) and 29 open answer questions was developed covering the following topics: (1) general information; (2) screening procedures and confirmatory diagnosis procedures; (3) treatment practices; (4) follow-up; (5) constitution of treatment team; (6) existing guidelines and protocols; (7) services offered to patients; and (8) challenges and areas for improvement in PKU management. The full survey is available in Online Resource 1. Invited health care professionals An invitation with a link to the online questionnaire (Survey Monkey: www.surveymonkey.com) was emailed to 80 health care professionals (HCP) working in the field of PKU from 59 centres in 22 countries in South and Eastern Europe (see Fig. 1 and Appendix). Belarus was not included in this study because no HCP contacts were known. The Russian Federation was not included due to difficulties in accessing health professionals, although many PKU treatment centres exist. The questionnaire was only provided in English.

ries in South and Eastern Europe (see Fig. 1 and Appendix). Belarus was not included in this study because no HCP contacts were known. The Russian Federation was not included due to difficulties in accessing health professionals, although many PKU treatment centres exist. The questionnaire was only provided in English. Data were collected from February to August 2014. A maximum of three HCPs per centre were invited to participate. To avoid inconsistencies in responses, all contributing HCPs from the same centre were requested to complete a single questionnaire. Data analysis Returned questionnaires were analysed for completeness: questionnaires with incomplete answers beyond the ‘general information’ section were excluded from the main results analysis, and the demographic data provided by these questionnaires were evaluated separately (see Sect. “Centres providing demographic data only/e-mail responses”). Information provided by centres exclusively in e-mail form was also evaluated separately (see Sect. “Centres providing demographic data only/e-mail responses”). Data were analysed using descriptive statistics (percent of total responses or medians). Prior to analysis, responses to some open answer questions were grouped or categorised according to the answers received. Answers were not systematically quality checked with the participating centres. Questions and answers not included in this report are summarised in Online Resource 2.

of total responses or medians). Prior to analysis, responses to some open answer questions were grouped or categorised according to the answers received. Answers were not systematically quality checked with the participating centres. Questions and answers not included in this report are summarised in Online Resource 2. Results Contributing centres and HCPs In total, information on the prevalence and management of PKU was obtained from 37/59 (63 %) centres in 19/22 (86 %) contacted countries (Fig. 1). Complete questionnaires were returned from 31 centres involving 60 HCPs in 15 countries. Incomplete questionnaires, which included only demographic data, and/or e-mail-only responses were returned from six centres involving an additional four countries (Fig. 1; see Sect. “Centres providing demographic data only/e-mail responses” for more details). Contributing HCPs and centres are listed in Appendix.Fig. 1 Countries and centres participating in the survey. Information was returned from 19/22 contacted countries, ordered in the list by the total number of responding centres per country (high to low). ‘–’ data not provided. aCentre provided only an incomplete questionnaire response; bCentre provided only an email response with limited information; cNumber of patients in centre providing the incomplete questionnaire, not included in main results’ analysis

total number of responding centres per country (high to low). ‘–’ data not provided. aCentre provided only an incomplete questionnaire response; bCentre provided only an email response with limited information; cNumber of patients in centre providing the incomplete questionnaire, not included in main results’ analysis Among the 60 HCPs returning complete questionnaires, most were either metabolic paediatricians (43.3 %) or paediatricians (30.0 %). Other professions included ‘clinical geneticist’ (6.7 %); ‘dietician/nutritionist’ (6.7 %); ‘research scientist’ (3.3 %); and ‘dietician, metabolic diseases’, ‘laboratory technician’, ‘physician’, ‘adult metabolic doctor’, ‘nurse specialising in PKU’ and ‘clinical biochemist’ (each 1.7 %). These HCPs had a median (range) of 17 (1–41) years of PKU clinical experience, and 75 % cared for patients of all ages with PKU or maternal PKU. Ten percent of centres reported that their PKU diagnosis and treatment team consisted of physicians, clinical biochemists, psychologists, dieticians and specialist nurses who care for the PKU population, and 16.1 % of centres reported having a specialised adult PKU clinic (Table 1). A dedicated dietician was part of the team in 65 % of centres, 70 % of whom have a university degree.Table 1 Summary of patients with PKU and management team profile (N = 31)

dieticians and specialist nurses who care for the PKU population, and 16.1 % of centres reported having a specialised adult PKU clinic (Table 1). A dedicated dietician was part of the team in 65 % of centres, 70 % of whom have a university degree.Table 1 Summary of patients with PKU and management team profile (N = 31) Question and answers by category Median (range) How many PKU patients are currently followed at your clinic/centre? 102 (3–2500) 1–25 26–50 51–100 101–200 201–500 ≥501 12.9 % 9.7 % 22.6 % 16.1 % 22.6 % 12.9 % What is the number of new PKU patients followed over the course of a typical year at your clinic/centre? 6 (1–50) 0 1–5 6–10 11–20 ≥21 No answer 0 % 46.7 % 20.0 % 13.3 % 13.3 % 6.7 % What is the number of maternal PKU pregnancies followed over the course of a typical year at your clinic/centre? 2 (0–17) 0 1–5 6–10 11–20 ≥21 No answer 19.4 % 58.1 % 6.5 % 3.2 % 0 % 12.9 % Approximately what percentage of PKU patients currently followed at your centre were late diagnosed? 10 (0–100) 0 % 1–5 % 6–10 % 11–20 % ≥21 % No answer 6.5 % 29.0 % 19.4 % 25.8 % 12.9 % 6.5 % The PKU team at your centre includes which of the following?a N/A Physicians (any type) Clinical biochemists Psychologists Dieticians/nutritionists (any type) Nurses specialising in PKU Research scientists Otherb 100 % 58.1 % 54.8 % 71.0 % 25.8 % 29.0 % 9.7 % Who cares for PKU patients from the age of 18 years?a N/A PKU paediatric clinic PKU adult clinic Otherc 71.0 % 16.1 % 16.1 % N/A not applicable, PKU phenylketonuria aMultiple answers were possible, and therefore, the total exceeds 100 %

Question and answers by category Median (range) How many PKU patients are currently followed at your clinic/centre? 102 (3–2500) 1–25 26–50 51–100 101–200 201–500 ≥501 12.9 % 9.7 % 22.6 % 16.1 % 22.6 % 12.9 % What is the number of new PKU patients followed over the course of a typical year at your clinic/centre? 6 (1–50) 0 1–5 6–10 11–20 ≥21 No answer 0 % 46.7 % 20.0 % 13.3 % 13.3 % 6.7 % What is the number of maternal PKU pregnancies followed over the course of a typical year at your clinic/centre? 2 (0–17) 0 1–5 6–10 11–20 ≥21 No answer 19.4 % 58.1 % 6.5 % 3.2 % 0 % 12.9 % Approximately what percentage of PKU patients currently followed at your centre were late diagnosed? 10 (0–100) 0 % 1–5 % 6–10 % 11–20 % ≥21 % No answer 6.5 % 29.0 % 19.4 % 25.8 % 12.9 % 6.5 % The PKU team at your centre includes which of the following?a N/A Physicians (any type) Clinical biochemists Psychologists Dieticians/nutritionists (any type) Nurses specialising in PKU Research scientists Otherb 100 % 58.1 % 54.8 % 71.0 % 25.8 % 29.0 % 9.7 % Who cares for PKU patients from the age of 18 years?a N/A PKU paediatric clinic PKU adult clinic Otherc 71.0 % 16.1 % 16.1 % N/A not applicable, PKU phenylketonuria aMultiple answers were possible, and therefore, the total exceeds 100 % b‘Other’ was specified as ‘molecular geneticist’, ‘laboratory technician’, and ‘neurologist’ in three surveys c‘Other’ was specified as (number of questionnaires): ‘Department of Genetics’ (3), ‘adult neurologists (1), ‘metabolic physicians’ (1) and ‘general practitioner’ (1)

aMultiple answers were possible, and therefore, the total exceeds 100 % b‘Other’ was specified as ‘molecular geneticist’, ‘laboratory technician’, and ‘neurologist’ in three surveys c‘Other’ was specified as (number of questionnaires): ‘Department of Genetics’ (3), ‘adult neurologists (1), ‘metabolic physicians’ (1) and ‘general practitioner’ (1) Patients In total, the 31 centres who returned complete questionnaires followed a total of 8573 patients with PKU, of whom approximately 75 % were followed in two countries, Turkey and Poland (Fig. 1). Data on patients and the HCPs working in the treatment team at each centre are presented in Table 1. The median number of patients per centre was 102 (range 3–2500), with the largest centres in Turkey and Poland (Fig. 1). The median number of new patients followed per year per centre was 6 (range 1–50). A median of 10 % (range 0–100) of patients were late diagnosed, and only two centres had no late diagnosed patients. Across the region, patients were estimated to travel a median of 110 km (range 10–500 km) to their centre.

and (Fig. 1). The median number of new patients followed per year per centre was 6 (range 1–50). A median of 10 % (range 0–100) of patients were late diagnosed, and only two centres had no late diagnosed patients. Across the region, patients were estimated to travel a median of 110 km (range 10–500 km) to their centre. Screening and confirmatory diagnosis procedures Among the completed questionnaires, only one centre in Malta had no newborn screening programme (Fig. 1, further information on screening practices presented in Table 2; see Sect. “Centres providing demographic data only/e-mail responses” for responses from other countries lacking newborn screening). Seventy-one percent of centres screened at 3 days of age (median 3 days; range 2–7 days), and 77 % saw positively screened newborns within the first 15 days of life (median 10 days; range 3–30 days). The most common upper threshold for a positive newborn screening test was a blood Phe concentration of 120 μmol/L (48.4 %), followed by 180 μmol/L (25.8 %).Table 2 Summary of screening and confirmatory diagnostic practices (N = 31)

ly screened newborns within the first 15 days of life (median 10 days; range 3–30 days). The most common upper threshold for a positive newborn screening test was a blood Phe concentration of 120 μmol/L (48.4 %), followed by 180 μmol/L (25.8 %).Table 2 Summary of screening and confirmatory diagnostic practices (N = 31) Question and answers by category Median (range) At what age is the heel prick test performed at your centre? 3 (2–7) 0–1 days 2 days 3 days 4 days ≥5 days No answer 0 % 19.4 % 71.0 % 3.2 % 3.2 % 3.2 % At what age are positively screened newborns seen in the medical service? 10 (3–30) 0–9 days 10–15 days 16–28 days ≥29 days No answer 41.9 % 35.5 % 16.1 % 3.2 % 3.2 % What is the blood Phe level cut-off for a positive neonatal screening test in your country? N/A 120 μmol/L 180 μmol/L 240 μmol/L 360 μmol/L Othera No answer 48.4 % 25.8 % 9.7 % 3.2 % 9.7 % 3.2 % For which patient age groups does your centre perform the BH4 loading test?b N/A Newborns Infants Young children Older children Adolescents Adults No answer 38.7 % 9.7 % 25.8 % 32.3 % 22.6 % 3.2 % 35.5 % What is the duration of the BH4 loading test? N/A 8 h 24 h 48 h 72 h 7 days 30 days No answer 0 % 35.5 % 9.7 % 9.7 % 0 % 3.2 % 41.9 % What percentage of patients at your centre are BH4 responders? 15 (0–31) 0 % 1–5 % 6–10 % 11–20 % ≥21 % No answer 3.2 % 9.7 % 6.5 % 9.7 % 12.9 % 58.1 % BH4 sapropterin dihydrochoride, N/A not applicable, Phe phenylalanine a‘Other’ was specified by three questionnaires as 80, 132 and 150 μmol/L bMultiple answers were possible, and therefore, the total exceeds 100 %

Question and answers by category Median (range) At what age is the heel prick test performed at your centre? 3 (2–7) 0–1 days 2 days 3 days 4 days ≥5 days No answer 0 % 19.4 % 71.0 % 3.2 % 3.2 % 3.2 % At what age are positively screened newborns seen in the medical service? 10 (3–30) 0–9 days 10–15 days 16–28 days ≥29 days No answer 41.9 % 35.5 % 16.1 % 3.2 % 3.2 % What is the blood Phe level cut-off for a positive neonatal screening test in your country? N/A 120 μmol/L 180 μmol/L 240 μmol/L 360 μmol/L Othera No answer 48.4 % 25.8 % 9.7 % 3.2 % 9.7 % 3.2 % For which patient age groups does your centre perform the BH4 loading test?b N/A Newborns Infants Young children Older children Adolescents Adults No answer 38.7 % 9.7 % 25.8 % 32.3 % 22.6 % 3.2 % 35.5 % What is the duration of the BH4 loading test? N/A 8 h 24 h 48 h 72 h 7 days 30 days No answer 0 % 35.5 % 9.7 % 9.7 % 0 % 3.2 % 41.9 % What percentage of patients at your centre are BH4 responders? 15 (0–31) 0 % 1–5 % 6–10 % 11–20 % ≥21 % No answer 3.2 % 9.7 % 6.5 % 9.7 % 12.9 % 58.1 % BH4 sapropterin dihydrochoride, N/A not applicable, Phe phenylalanine a‘Other’ was specified by three questionnaires as 80, 132 and 150 μmol/L bMultiple answers were possible, and therefore, the total exceeds 100 % After a positive newborn screening test, diagnosis was confirmed by tandem mass spectrometry (39 %), amino acids chromatography (26 %), and/or a fluorescent, enzymatic or colorimetric method (58 %). No centres reported relying on the Guthrie test for confirmatory diagnosis. Routine genetic analysis was performed at 64.5 % of centres as part of their diagnostic procedure (centres in Croatia (1/1), Romania (3/3), Turkey (5/7) and Poland (1/7) did not conduct genetic analysis). Of the centres, 51.6 % performed routine sapropterin dihydrochloride (tetrahydrobiopterin (BH4)) loading tests (newborns, 38.7 %; older children, 32.3 %; adults, 3.2 %). This test is used to investigate a diagnosis of BH4-responsive PKU and/or to rule out BH4 deficiency in newborns. At these centres, the BH4 loading test dose was 20 mg/kg, BH4 responsiveness in PKU was consistently defined as a reduction in Phe concentration of ≥30 %, and the most common duration of the test was 24 h (35.5 %; range 24 h–30 days). A median of 15 % (range 0–31) of patients were classified as BH4 responders. Other tests used to distinguish BH4 deficiency from PKU included pterins and dihydropteridine reductase (DHPR) analysis (45.2 %).

on in Phe concentration of ≥30 %, and the most common duration of the test was 24 h (35.5 %; range 24 h–30 days). A median of 15 % (range 0–31) of patients were classified as BH4 responders. Other tests used to distinguish BH4 deficiency from PKU included pterins and dihydropteridine reductase (DHPR) analysis (45.2 %). Treatment practices and reimbursement All centres advocated life-long treatment with a low-Phe diet (based on Phe-free amino acid supplements and special low-protein foods) (Table 3), whereas BH4 and large neutral amino acids were available in 48 and 32 % of centres, respectively (in Poland BH4 treatment was only available for patients with BH4 deficiency).

ll centres advocated life-long treatment with a low-Phe diet (based on Phe-free amino acid supplements and special low-protein foods) (Table 3), whereas BH4 and large neutral amino acids were available in 48 and 32 % of centres, respectively (in Poland BH4 treatment was only available for patients with BH4 deficiency). Blood Phe thresholds for starting low-Phe diet treatment in different patient groups are presented in Fig. 2. For newborns, the most common blood Phe threshold for initiation of treatment was ≥360 μmol/L (65 %; other thresholds used were ≥240 μmol/L [19 %], ≥400 μmol/L [6 %], ≥600 μmol/L [10 %]). For young and older children, this was also ≥360 μmol/L (64 and 56 %, respectively), whereas for adolescents and adults, it was ≥600 μmol/L (52 and 65 %, respectively). Phe thresholds to commence treatment were not always the same between centres in the same country (data not shown).Fig. 2 Blood Phe threshold levels for starting low-Phe diet treatment in each patient group (consistently elevated levels), ‘n’ represents the number of questionnaires which provided an answer for each group (total responses); ‘Other’ was specified for adolescents as ‘480 μmol/L’, and for adults as ‘130–900’ and ‘900 μmol/L’; age groups were as follows: newborns and infants, <1 year of age; young children, 1–4 years of age; older children, 5–10 years of age; adolescents, 11–17 years of age; adults, ≥18 years of age

h group (total responses); ‘Other’ was specified for adolescents as ‘480 μmol/L’, and for adults as ‘130–900’ and ‘900 μmol/L’; age groups were as follows: newborns and infants, <1 year of age; young children, 1–4 years of age; older children, 5–10 years of age; adolescents, 11–17 years of age; adults, ≥18 years of age Seventy-one percent of centres started administering low-Phe diet treatment before 15 days of a patient’s life, with 26 % reporting this to be >15 days (Table 3). The median dose of total protein (natural and Phe-free L-amino acid supplement) prescribed to newborns and infants with classical PKU was 2.5 g/kg/day (median [range] lower limit 2.50 [1.0–3.0] g/kg/day, median upper limit 2.50 [2.0–3.5] g/kg/day), which decreased steadily with increasing patient age to 1 g/kg/day in adults (median [range] lower limit 1.0 [0.7–1.5] g/kg/day, median upper limit 1.0 [0.7–1.8] g/kg/day) (Fig. 3).Fig. 3 Dose of total protein (g/kg/day) prescribed to patients with classical PKU by each centre per patient age group. Box and whisker plot, with boxes representing interquartile range (thick line indicating the median, with value indicated in the plot), dashed whiskers minimum and maximum values. ‘n’ represents the number of questionnaires which provided an answer for each group; age groups were as follows: newborns and infants, <1 year of age; young children, 1–4 years of age; older children, 5–10 years of age; adolescents, 11–17 years of age; adults, ≥18 years of age

ashed whiskers minimum and maximum values. ‘n’ represents the number of questionnaires which provided an answer for each group; age groups were as follows: newborns and infants, <1 year of age; young children, 1–4 years of age; older children, 5–10 years of age; adolescents, 11–17 years of age; adults, ≥18 years of age Low-Phe diet was available to treat maternal PKU at all centres, although only 22 centres indicated in their questionnaire the Phe level thresholds used to initiate treatment in maternal PKU. This was either ≥240 μmol/L (74 %) or ≥360 μmol/L (26 %; Fig. 2). With respect to reimbursement and social support, governments contributed to costs of Phe-free L-amino acid supplements in the majority of centres (97 %), although there was less governmental support to cover costs of special low-protein foods (59 %; Table 3). Patients at most centres were entitled to a disability allowance and/or a disability certificate to cover out of pocket costs (81 %), although there was no social support available for patients at three centres (one in each of Bulgaria, Hungary and Turkey) (Table 3).Table 3 Summary of treatment and reimbursement information (N = 31) What treatment options are available in your country? Low-Phe diet BH4 Large neutral amino acids Othera 100 % 48.4 % 32.3 % 3.2 % At what age does your centre introduce low-Phe diet in newly diagnosed newborns? 0–9 days 10–15 days 16–28 days ≥29 days No answer 19.4 % 51.6 % 22.6 % 3.2 % 3.2 % Who contributes to the costs of Phe-free protein substitutes?b

What treatment options are available in your country? Low-Phe diet BH4 Large neutral amino acids Othera 100 % 48.4 % 32.3 % 3.2 % At what age does your centre introduce low-Phe diet in newly diagnosed newborns? 0–9 days 10–15 days 16–28 days ≥29 days No answer 19.4 % 51.6 % 22.6 % 3.2 % 3.2 % Who contributes to the costs of Phe-free protein substitutes?b Government Private health insurance Parents/patients 96.8 % 12.9 % 22.6 % Who contributes to the cost of special low-protein foods (flour, pasta)?b Government Private health insurance Parents/patients Otherc 59.4 % 12.5 % 75.0 % 6.3 % What kind of social support is offered to PKU patients in your country?b Disability allowance Disability certificate which helps with education, employment, travel expenses Dietary allowance to go to summer camps Reimbursement of travel expenses Home support or compensation to parents for decreasing working hours No support available 80.6 % 45.2 % 32.3 % 25.8 % 12.9 % 9.7 % BH4 sapropterin dihydrochoride, Phe phenylalanine, PKU phenylketonuria a‘Other’ was specified as ‘BH4 only for BH4-deficient patients’ bMultiple answers were possible, and therefore, the total exceeds 100 % c‘Other’ was specified as: ‘sellers’, ‘non-governmental organisation’ and ‘parents/patients association’

Disability allowance Disability certificate which helps with education, employment, travel expenses Dietary allowance to go to summer camps Reimbursement of travel expenses Home support or compensation to parents for decreasing working hours No support available 80.6 % 45.2 % 32.3 % 25.8 % 12.9 % 9.7 % BH4 sapropterin dihydrochoride, Phe phenylalanine, PKU phenylketonuria a‘Other’ was specified as ‘BH4 only for BH4-deficient patients’ bMultiple answers were possible, and therefore, the total exceeds 100 % c‘Other’ was specified as: ‘sellers’, ‘non-governmental organisation’ and ‘parents/patients association’ Follow-up practices A specific follow-up protocol for patients with PKU was used in 77.4 % of centres (Table 4). Most centres collected blood samples from the patient’s home (74.2 %) and in outpatient clinics (67.7 %), and 61 % returned blood Phe results within 4 days (range 1–8 days), using e-mail, phone, letter and clinic visits (data not shown).Table 4 Summary of follow-up practices (N = 31)

sed in 77.4 % of centres (Table 4). Most centres collected blood samples from the patient’s home (74.2 %) and in outpatient clinics (67.7 %), and 61 % returned blood Phe results within 4 days (range 1–8 days), using e-mail, phone, letter and clinic visits (data not shown).Table 4 Summary of follow-up practices (N = 31) Question and answers by category Median (range) Does your centre follow a specific follow-up protocol for PKU patients? N/A Yes No No answer 77.4 % 19.4 % 3.2 % Where does your centre collect samples for monitoring Phe levels in treated PKU patients?a N/A Home Outpatient clinic Hospital Family doctor clinic (general practitioner) 74.2 % 67.7 % 35.5 % 29.0 % What is the average return time of routine Phe control results to your patients, after blood sampling? N/A 1–2 days 3–4 days 5–6 days 7–8 days No answer 25.8 % 35.5 % 16.1 % 19.4 % 3.2 % What proportion of your patients is lost to follow-up (defined as patient was not seen for two years)? 10 (0–21) 0 % 1–5 % 6–10 % 11–20 % ≥21 % No answer 9.7 % 22.6 % 22.6 % 25.8 % 3.2 % 16.1 % N/A not applicable, Phe phenylalanine, PKU phenylketonuria aMultiple answers were possible, and therefore, the total exceeds 100 %

Question and answers by category Median (range) Does your centre follow a specific follow-up protocol for PKU patients? N/A Yes No No answer 77.4 % 19.4 % 3.2 % Where does your centre collect samples for monitoring Phe levels in treated PKU patients?a N/A Home Outpatient clinic Hospital Family doctor clinic (general practitioner) 74.2 % 67.7 % 35.5 % 29.0 % What is the average return time of routine Phe control results to your patients, after blood sampling? N/A 1–2 days 3–4 days 5–6 days 7–8 days No answer 25.8 % 35.5 % 16.1 % 19.4 % 3.2 % What proportion of your patients is lost to follow-up (defined as patient was not seen for two years)? 10 (0–21) 0 % 1–5 % 6–10 % 11–20 % ≥21 % No answer 9.7 % 22.6 % 22.6 % 25.8 % 3.2 % 16.1 % N/A not applicable, Phe phenylalanine, PKU phenylketonuria aMultiple answers were possible, and therefore, the total exceeds 100 % The median percentage of patients ‘lost to follow-up’ (as defined by a patient not being seen for over 2 years) was 10 %. There was a general trend for higher lost to follow-up rates in the larger centres. Among the centres providing this data, the 13 largest centres (range of patients 130–2500) had a median of 15 % patients lost to follow-up, whereas this was 5 % for the 13 smallest centres (range of patients 3–123).

r 2 years) was 10 %. There was a general trend for higher lost to follow-up rates in the larger centres. Among the centres providing this data, the 13 largest centres (range of patients 130–2500) had a median of 15 % patients lost to follow-up, whereas this was 5 % for the 13 smallest centres (range of patients 3–123). Blood Phe level target ranges are presented in Fig. 4a. With a few exceptions, the lower Phe target was 120 μmol/L across all age groups. The median (range) upper Phe level target was 240 μmol/L (180–600) in newborns, 360 μmol/L (180–900) in children, and 600 μmol/L (360–1200) in teenagers and adults. Upper Phe targets were more variable across centres than lower Phe levels targets, with the interquartile range spanning >100 μmol/L for newborns and infants, young children and adults. With respect to the frequency of Phe monitoring, there was a general trend towards less frequent monitoring with increasing patient age (Fig. 4b).Fig. 4 Blood Phe target ranges (μmol/L) and frequency of blood Phe monitoring for different patient groups. a Box and whisker plot of blood Phe target ranges per patient group: boxes represent interquartile range (thick line indicating the median, with value indicated in the plot), and dashed whiskers minimum and maximum values. b Frequency histogram of blood Phe monitoring frequency. ‘Other’ was as follows: newborns and infants: ‘sometimes weekly, sometimes 2-weekly’; young children: ‘1–2 years: 2-weekly; 2–3 years: 3-weekly; 3–4 years: monthly’; older children: not specified; adolescents: ‘2-weekly or monthly’ ‘depends on patient collaboration’; not specified; adults: ‘It depends on the patient collaboration’, not specified; maternal PKU: ‘sometimes weekly, sometimes 2X per week’; not specified (2). ‘n’ represents the number of questionnaires which provided an answer for each group. Age groups were as follows: newborns and infants, <1 year of age; young children, 1–4 years of age; older children, 5–10 years of age; adolescents, 11–17 years of age; adults, ≥18 years of age

sometimes 2X per week’; not specified (2). ‘n’ represents the number of questionnaires which provided an answer for each group. Age groups were as follows: newborns and infants, <1 year of age; young children, 1–4 years of age; older children, 5–10 years of age; adolescents, 11–17 years of age; adults, ≥18 years of age In maternal PKU, the median (range) of the upper blood Phe target while on treatment was 240 μmol/L (120–360) (Fig. 4a), and Phe level monitoring was most commonly on a weekly basis (13/23; 57 %), with 3/23 (13 %) centres monitoring every 2 weeks (Fig. 4b). Guidelines, registries and organisations devoted to PKU A variety of guidelines and protocols for PKU diagnosis and treatment were in use across the centres: while 32 % of centres used only published guidelines/protocols, 32 % used only their own unpublished guidelines/protocols and 29 % of centres used a mixture of both types. Eighty-five percent of centres were aware of either a local or national PKU registry, and 94 % were aware of patient/family organisations. Challenges and areas for improvement Participants were asked to describe the main challenges that they face in terms of screening, diagnosis and treatment of PKU; a summary of this feedback is presented in Online Resource 3.

Guidelines, registries and organisations devoted to PKU A variety of guidelines and protocols for PKU diagnosis and treatment were in use across the centres: while 32 % of centres used only published guidelines/protocols, 32 % used only their own unpublished guidelines/protocols and 29 % of centres used a mixture of both types. Eighty-five percent of centres were aware of either a local or national PKU registry, and 94 % were aware of patient/family organisations. Challenges and areas for improvement Participants were asked to describe the main challenges that they face in terms of screening, diagnosis and treatment of PKU; a summary of this feedback is presented in Online Resource 3. Centres providing demographic data only/e-mail responses Four centres (Bosnia and Herzegovina, Republic of Macedonia, Romania and Serbia) provided only demographic data. These centres collectively cared for 27 patients with PKU, a high proportion of whom were late diagnosed (range 20–90 % of patients). The Republic of Macedonia was previously reported to have no newborn screening for PKU [9], and the contacted centre indicated via e-mail that they manage the sporadic cases they encounter with an occupational therapy plan. Two centres (Albania and Montenegro) provided only short responses by e-mail, each reporting that there was no PKU screening in their respective countries.

Centres providing demographic data only/e-mail responses Four centres (Bosnia and Herzegovina, Republic of Macedonia, Romania and Serbia) provided only demographic data. These centres collectively cared for 27 patients with PKU, a high proportion of whom were late diagnosed (range 20–90 % of patients). The Republic of Macedonia was previously reported to have no newborn screening for PKU [9], and the contacted centre indicated via e-mail that they manage the sporadic cases they encounter with an occupational therapy plan. Two centres (Albania and Montenegro) provided only short responses by e-mail, each reporting that there was no PKU screening in their respective countries. Discussion Challenges in diagnosis and treatment A well-documented issue for PKU care is that management guidelines differ between countries, often in important areas such as newborn screening, target blood Phe levels in different patient groups, frequency of Phe monitoring and duration of treatment [29]. Our survey adds to a growing body of evidence showing a difference in PKU care between Western and South and Eastern Europe and provides further reason to support efforts to raise and standardise treatment across Europe.

Phe levels in different patient groups, frequency of Phe monitoring and duration of treatment [29]. Our survey adds to a growing body of evidence showing a difference in PKU care between Western and South and Eastern Europe and provides further reason to support efforts to raise and standardise treatment across Europe. The most critical disparity concerns the lack of newborn screening in several centres, thereby leading to late commencement of treatment. Results from this survey and complementary studies [9, 11] highlight that although newborn screening is widespread, it is not yet implemented in all European countries. This includes five of the target countries of this survey, namely Albania, Kosovo (from where no information was returned), Malta, Montenegro and the Republic of Macedonia (Fig. 1). Furthermore, in some countries with established newborn screening programmes, including Bulgaria and Romania, up to 10 % of newborns are not screened [9]. Based on these data and current birth rates [6, 9], there were approximately 170,000 live births unscreened in 2013 across the target countries of this survey. Considering an estimated PKU prevalence of 1/10,000 live births [16], 17 newborns per year may face a delayed diagnosis across these countries.

ot screened [9]. Based on these data and current birth rates [6, 9], there were approximately 170,000 live births unscreened in 2013 across the target countries of this survey. Considering an estimated PKU prevalence of 1/10,000 live births [16], 17 newborns per year may face a delayed diagnosis across these countries. Our results from 31 centres providing complete questionnaires (of which 30 had newborn screening) suggest that diagnostic testing is not always comprehensive or optimally managed in the region. For example, fewer than half of the centres reported using specific tests, such as pterins and DHPR analysis, to diagnose BH4 deficiency. Also, the routine use of BH4 loading tests was linked to the availability of BH4 treatment for patients with PKU, and a lack of BH4 testing and treatment were highlighted as main challenges/areas for improvement (Online Resource 3). Nevertheless, use of some advanced diagnostic techniques was surprisingly widespread: genetic analysis was routinely used in 13 of the countries represented in this survey, higher than reported elsewhere [1]. Another concern is the time taken for positively screened patients to be seen at some centres, which was over 2 weeks in 20 % of centres and over 1 month in one centre, and was mirrored by a similar delay in starting treatment at these centres. By contrast, US guidelines recommend that initiation of treatment for PKU should preferably occur within the first week of life, with the aim to have achieved target blood Phe concentrations within the first 2 weeks of life [30].

in one centre, and was mirrored by a similar delay in starting treatment at these centres. By contrast, US guidelines recommend that initiation of treatment for PKU should preferably occur within the first week of life, with the aim to have achieved target blood Phe concentrations within the first 2 weeks of life [30]. Regarding the professionals within the treatment teams, there was a notable lack of dedicated dieticians in 30 % of the centres in this survey (compared with only 6 % in the predominantly Western European survey [1]). Dieticians are key to ensuring best management of the complex dietary requirements of patients with PKU [7, 17]. The general lack of all core professionals within treatment teams and of specialised adult PKU clinics has also been reported elsewhere, including in countries with advanced management approaches [1, 26, 27].

ieticians are key to ensuring best management of the complex dietary requirements of patients with PKU [7, 17]. The general lack of all core professionals within treatment teams and of specialised adult PKU clinics has also been reported elsewhere, including in countries with advanced management approaches [1, 26, 27]. A striking statistic was the relatively large number of patients per centre who lived within large catchment areas. There were approximately two times more patients per centre than in a predominantly Western European survey [1], and the median travel distance to the centre was over 100 km. This could impact care in two ways. Firstly, an increased patient load may put increased strain on resources (additional information on patient staffing levels will be needed to assess this possibility, see ‘Sect. Limitations of study’). Secondly, living further than approximately 160 km from the care centre has been associated with significantly fewer Phe samples for analysis [8], which raises the possibility that distance may impact outcomes. The proportion of patients lost to follow-up (median of 10 % in this survey), might also be influenced by the size of these catchment areas and patient loads. This situation might be caused by countries adopting a ‘tertiary referral hospital’ model, which might not be optimal in all countries. Notably, it could add to out-of-pocket costs of patients travelling to the centres. Individual countries should review whether it is more effective to redistribute resources at a more local level for managing PKU care.

countries adopting a ‘tertiary referral hospital’ model, which might not be optimal in all countries. Notably, it could add to out-of-pocket costs of patients travelling to the centres. Individual countries should review whether it is more effective to redistribute resources at a more local level for managing PKU care. Commencement and management of treatment was also not always optimal. Inconsistencies across centres in the Phe level thresholds to start treatment and upper Phe level targets during treatment is a universal issue [1, 11] and may lead to different outcomes for patients with an otherwise similar condition [12]. Monitoring frequency followed the expected trend of being most frequent in newborns and declining with increasing patient age, although this was also variable across centres. Importantly, in maternal PKU, 10 % of centres monitored Phe levels only once every 2 weeks, which is sub-optimal given the changing Phe requirements during pregnancy and risks associated with both high and low Phe concentrations: very low levels are associated with intrauterine growth retardation; high levels (>360 μmol/L) carry a high risk of maternal PKU syndrome [13, 25]. It is hoped that several of these issues will be addressed in the near future by the release of Pan-European guidelines [11].

ted with both high and low Phe concentrations: very low levels are associated with intrauterine growth retardation; high levels (>360 μmol/L) carry a high risk of maternal PKU syndrome [13, 25]. It is hoped that several of these issues will be addressed in the near future by the release of Pan-European guidelines [11]. A lack of treatment options (i.e. availability of BH4, variety of protein substitutes and foods) featured as a main challenge, as noted elsewhere [1, 11]. About half of centres relied only on low-Phe diet, and some felt that the variety of Phe-free protein substitutes and low-protein foods available to patients was limited due to a lack of reimbursement. These issues could decrease adherence to a low-Phe diet [18]. Given the relatively low income of many families in the surveyed countries, the cost and availability of low-Phe foods is an important issue that must be addressed by healthcare systems and governments. BH4 treatment is expected to become more widely available in the coming years, although reimbursement issues are likely to affect its impact.

income of many families in the surveyed countries, the cost and availability of low-Phe foods is an important issue that must be addressed by healthcare systems and governments. BH4 treatment is expected to become more widely available in the coming years, although reimbursement issues are likely to affect its impact. Limitations of study This analysis is subject to several limitations. Unfortunately, of the five countries known to lack newborn screening, only one centre in Malta provided a completed questionnaire. Centres in The Republic of Macedonia, Albania and Montenegro provided limited information and no response was received from Kosovo, meaning that we were unable to comprehensively evaluate standard of care in these locations. The number of patients with PKU in the region who remain undiagnosed is also unknown and is likely to be a major issue in areas where newborn screening was only introduced recently (e.g. in Romania, nationwide newborn screening was only introduced in 2011 [9]). Thus, there may be severely intellectually disabled patients living without a correct diagnosis or appropriate medical care not recognised by this survey [19, 21, 28].

jor issue in areas where newborn screening was only introduced recently (e.g. in Romania, nationwide newborn screening was only introduced in 2011 [9]). Thus, there may be severely intellectually disabled patients living without a correct diagnosis or appropriate medical care not recognised by this survey [19, 21, 28]. Several targeted centres did not answer the questionnaire. Literature-based estimates put the population with PKU in the surveyed region at around 19,000 patients [16], which suggests a 45 % coverage by this survey. A lack of translation into the local language or poor access to computers and/or the internet may have hampered questionnaire completion by some centres. Additionally, contacts were lacking for Belarus, and the list of contacts for other countries may not have been exhaustive (although in an additional question, the 31 centres in the main results section indicated that a cumulative total of 45 centres existed across their countries, comparable to the 46 that were contacted, so it is likely that very few were omitted; see Online Resource 1 and 2). The centres in this survey care for different numbers of patients with PKU, and therefore, some apply a management approach which affects far more patients than others (whereas the size of each centre is not weighted in the above analysis). It should also be noted that neither the extent to which each member of the treatment team is dedicated to PKU, nor staff to patient ratios were established. These two factors play an important role in the delivery of care.

re patients than others (whereas the size of each centre is not weighted in the above analysis). It should also be noted that neither the extent to which each member of the treatment team is dedicated to PKU, nor staff to patient ratios were established. These two factors play an important role in the delivery of care. Finally, the questionnaire used non-validated questions, which were open to misinterpretation. Conclusion Results of this survey point to several important areas for improvement in PKU diagnostic and management practices across Southern and Eastern Europe. Given that there remain parts of Europe where there is no routine newborn screening (or it has only been recently introduced), older patients with PKU who can benefit from low-Phe diet treatment [14, 31] are likely under-diagnosed. Interestingly, coverage of newborn screening and diagnostic techniques may not necessarily be related to the economic standing of the country [9]. Furthermore, the diagnostic and treatment packages offered to patients may be more influenced by the interests and skills of the PKU teams rather than the direct needs of the patients. New evidence-based Pan-European guidelines are currently in development with the aim of encouraging a common standard of care [11]. It is important that development of new guidelines coincides with efforts by HCPs and governments across the region to ensure all patients receive the best possible care. Electronic supplementary material ESM. 1 (PDF 559 kb) ESM. 2 (PDF 644 kb) ESM. 3 (PDF 445 kb) Appendix Contributing HCPs and centres.

Conclusion Results of this survey point to several important areas for improvement in PKU diagnostic and management practices across Southern and Eastern Europe. Given that there remain parts of Europe where there is no routine newborn screening (or it has only been recently introduced), older patients with PKU who can benefit from low-Phe diet treatment [14, 31] are likely under-diagnosed. Interestingly, coverage of newborn screening and diagnostic techniques may not necessarily be related to the economic standing of the country [9]. Furthermore, the diagnostic and treatment packages offered to patients may be more influenced by the interests and skills of the PKU teams rather than the direct needs of the patients. New evidence-based Pan-European guidelines are currently in development with the aim of encouraging a common standard of care [11]. It is important that development of new guidelines coincides with efforts by HCPs and governments across the region to ensure all patients receive the best possible care. Electronic supplementary material ESM. 1 (PDF 559 kb) ESM. 2 (PDF 644 kb) ESM. 3 (PDF 445 kb) Appendix Contributing HCPs and centres. Bulgaria: Radka Tincheva, Aleksei Savov, Adil Kadam (University Pediatric Hospital Sofia); Croatia: Vladimir Sarnavka, Ivo Barić (University Hospital Centre Zagreb); Czech Republic: Dagmar Prochazkova (I.dětská interní klinika FN Brno), Renata Pazdírková, Jana Komárková, Hana Kothánková (Klinika dětí a dorostu FNKV, Praha); Estonia: Katrin Ounap, Karit Reinson, Mari-Liis Uudelepp (Tartu University Clinicum, Tartu); Hungary: László Szőnyi, Erika Kiss, Péter Reismann (Semmelweis University, Hungary); Latvia: Rita Lugovska, Parsla Vevere (Children clinical university hospital, Riga); Lithuania: Loreta Cimbalistiene (Vilnius University, Vilnius); Malta: Simon Attard Montalto (Mater Dei Hospital, University of Malta); Poland: Agnieszka Chrobot, Izabela Horka (Bydgoszcz), Bożena Didycz, Mirosław Bik-Multanowski (Kraków), Bożena Mikołuć, Maria Jolanta Piotrowska-Depta, Ewa Samocik (Białystok), Ewa Starostecka, Agata Lange (Łódź), Jolanta Wierzba, Joanna Jagłowska (Gdańsk), Kalina Plutowska-Hoffmann, Joanna Zarębska (Katowice), Maria Giżewska, Hanna Romanowska, Elżbieta Krzywińska-Zdeb (Szczecin), Maria Nowacka, Joanna Żółkowska, Dorota Korycińska-Chaaban (Warszawa); Republic of Moldova: Usurelu Natalia (Institute of Mother and Child, Chisinau); Romania: Anton-Paduraru Dana-Teodora (Spitalul clinic de urgenta pentru copii “Sf.

ska (Katowice), Maria Giżewska, Hanna Romanowska, Elżbieta Krzywińska-Zdeb (Szczecin), Maria Nowacka, Joanna Żółkowska, Dorota Korycińska-Chaaban (Warszawa); Republic of Moldova: Usurelu Natalia (Institute of Mother and Child, Chisinau); Romania: Anton-Paduraru Dana-Teodora (Spitalul clinic de urgenta pentru copii “Sf. Maria”, Iași), Mariana Muresan (Clinica de Pediatrie III, Cluj Napoca), Nanu Michaela Iuliana, Moldovanu Florentina, Iorgulescu Daniela (IOMC Bucuresti, Bucharest); Serbia: Maja Đorđević, Božica Kecman, Adrijan Sarajlija (Institut za zdravstvenu zaštitu majke i deteta Srbije, Novi Beograd), Slovakia: Katarína Hálová (Detská fakultná nemocnica, Banska Bystrica); Slovenia: Mojca Zerjav Tansek (University Clinical Center Ljubljana); Turkey: Burcu Öztürk Hişmi (Çocuk Metabolizma Hastalıkları ve Beslenme Uzmanı Gaziantep Çocuk, Gaziantep); Serap Sivri (Hacettepe University Faculty of Medicine, Hacettepe); Isil Ozer (Medeniyet University, Istanbul, Turkey); Mahmut Coker, Sema Kalkan Ucar (Ege University, Izmir); Neslihan önenli Mungan, Deniz Kör, Berna şeker Yilmaz (Cukurova University, Adana), Nur Arslan, Yesim Ozturk (Dokuz Eylul University, Izmir); Selda Bulbul (Kirikkale University, Turkey). Centres providing e-mail/incomplete questionnaire responses.

Maria”, Iași), Mariana Muresan (Clinica de Pediatrie III, Cluj Napoca), Nanu Michaela Iuliana, Moldovanu Florentina, Iorgulescu Daniela (IOMC Bucuresti, Bucharest); Serbia: Maja Đorđević, Božica Kecman, Adrijan Sarajlija (Institut za zdravstvenu zaštitu majke i deteta Srbije, Novi Beograd), Slovakia: Katarína Hálová (Detská fakultná nemocnica, Banska Bystrica); Slovenia: Mojca Zerjav Tansek (University Clinical Center Ljubljana); Turkey: Burcu Öztürk Hişmi (Çocuk Metabolizma Hastalıkları ve Beslenme Uzmanı Gaziantep Çocuk, Gaziantep); Serap Sivri (Hacettepe University Faculty of Medicine, Hacettepe); Isil Ozer (Medeniyet University, Istanbul, Turkey); Mahmut Coker, Sema Kalkan Ucar (Ege University, Izmir); Neslihan önenli Mungan, Deniz Kör, Berna şeker Yilmaz (Cukurova University, Adana), Nur Arslan, Yesim Ozturk (Dokuz Eylul University, Izmir); Selda Bulbul (Kirikkale University, Turkey). Centres providing e-mail/incomplete questionnaire responses. Albania: Lindita Grimci (University Hospital Center “Mother Teresa”, Tirana); Bosnia and Herzegovina: Smail Zubcevic (Clinical Center University of Sarajevo, Faculty of Medicine); Montenegro: Mira Samardzic (Institute for Sick Children, Podgorica); Republic of Macedonia: Elena Sukarova Angelovska, Natalija Angelkova (University Children’s Hospital Skopje); Romania: Otilia Marginean, Marinela Lesovici (Spitalului Clinic de Urgenta Pentru Copii “Louis Turcanu”, Timisoara); Serbia: Jovanovic Privrodski Jadranka, Kavecan Ivana (Institut Za Decu i Omladinu, Novi Sad). Abbreviations BH4Tetrahydrobiopterin or sapropterin dihydrochloride

Albania: Lindita Grimci (University Hospital Center “Mother Teresa”, Tirana); Bosnia and Herzegovina: Smail Zubcevic (Clinical Center University of Sarajevo, Faculty of Medicine); Montenegro: Mira Samardzic (Institute for Sick Children, Podgorica); Republic of Macedonia: Elena Sukarova Angelovska, Natalija Angelkova (University Children’s Hospital Skopje); Romania: Otilia Marginean, Marinela Lesovici (Spitalului Clinic de Urgenta Pentru Copii “Louis Turcanu”, Timisoara); Serbia: Jovanovic Privrodski Jadranka, Kavecan Ivana (Institut Za Decu i Omladinu, Novi Sad). Abbreviations BH4Tetrahydrobiopterin or sapropterin dihydrochloride DHPRDihydropteridine reductase HCPHealth care professional N/ANot applicable PhePhenylalanine PKUPhenylketonuria This project was conceived and discussed by the authors during European Phenylketonuria Group (EPG) meetings, an advisory board to Merck Serono. Editorial and data analysis support was provided by Neil Burton working for Fishawack Communications GmbH, funded by Merck Serono. The authors acknowledge the invaluable support of their colleagues who took their time to respond to this survey (listed in Appendix below).

PG) meetings, an advisory board to Merck Serono. Editorial and data analysis support was provided by Neil Burton working for Fishawack Communications GmbH, funded by Merck Serono. The authors acknowledge the invaluable support of their colleagues who took their time to respond to this survey (listed in Appendix below). Conflict of interest All authors are members of the EPG, an advisory board to Merck Serono, for which they have received consultancy fees from Merck Serono. Anita MacDonald and Amaya Belanger-Quintana have participated in the European Nutritionist Expert Panel in PKU (ENEP), an Advisory Board to Merck Serono. Additionally, Maria Giżewska has participated in strategic advisory boards for NUTRICIA Advanced Medical Nutrition and has received speaker honoraria from Merck Serono and NUTRICIA Advanced Medical Nutrition; Anita MacDonald has received research funding and honoraria from NUTRICIA, Vitaflo International and Merck Serono, and has participated in advisory boards to NUTRICIA and Arla; Amaya Bélanger-Quintana has received speaker honoraria from Merck Serono; Alberto Burlina has participated in advisory boards to NUTRICIA; Ania C. Muntau has participated in advisory boards to Arla, has received lecture fees from Merck Serono and has received research funding from Vitaflo, Merck Serono and NUTRICIA. Authors’ contributions All authors contributed to the design of the questionnaire, analysis of the results and the development of the manuscript, and have read and approved the final version of the manuscript.

Conflict of interest All authors are members of the EPG, an advisory board to Merck Serono, for which they have received consultancy fees from Merck Serono. Anita MacDonald and Amaya Belanger-Quintana have participated in the European Nutritionist Expert Panel in PKU (ENEP), an Advisory Board to Merck Serono. Additionally, Maria Giżewska has participated in strategic advisory boards for NUTRICIA Advanced Medical Nutrition and has received speaker honoraria from Merck Serono and NUTRICIA Advanced Medical Nutrition; Anita MacDonald has received research funding and honoraria from NUTRICIA, Vitaflo International and Merck Serono, and has participated in advisory boards to NUTRICIA and Arla; Amaya Bélanger-Quintana has received speaker honoraria from Merck Serono; Alberto Burlina has participated in advisory boards to NUTRICIA; Ania C. Muntau has participated in advisory boards to Arla, has received lecture fees from Merck Serono and has received research funding from Vitaflo, Merck Serono and NUTRICIA. Authors’ contributions All authors contributed to the design of the questionnaire, analysis of the results and the development of the manuscript, and have read and approved the final version of the manuscript. Ethical standards The manuscript does not contain clinical studies or patient data.

Introduction Acute gastroenteritis (AGE) is one of the most frequent reasons for young children to visit the emergency department (ED). Clinical dehydration scores are often used to assess severity of dehydration. These scores attempt to differentiate children without signs of dehydration from those with moderate dehydration or those with severe dehydration with signs of hypovolemic shock [10, 11]. The most commonly used clinical dehydration scale (CDS) is a 4-point scale, which includes four clinical signs (i.e., general appearance, eyes, mucous membranes, and tears) [10] The CDS has been incorporated in several clinical guidelines for appropriately managing acute gastroenteritis or dehydration [1, 13, 17, 25]. Although clinical guidelines aim to assist standardized assessment and treatment of dehydration, clinicians often do not adhere to the guidelines’ recommendations. Incorporating a guideline in an electronic, easily accessible clinical decision support system can improve guideline-adherence and therefore quality of care [18] In this study, we aimed to evaluate the feasibility of an electronic, easily accessible, guideline-based clinical decision support system, as well as the impact of this nurse-guided clinical decision support system for managing children with AGE at the ED compared to usual care on diagnostics, treatment and costs.

Introduction Acute gastroenteritis (AGE) is one of the most frequent reasons for young children to visit the emergency department (ED). Clinical dehydration scores are often used to assess severity of dehydration. These scores attempt to differentiate children without signs of dehydration from those with moderate dehydration or those with severe dehydration with signs of hypovolemic shock [10, 11]. The most commonly used clinical dehydration scale (CDS) is a 4-point scale, which includes four clinical signs (i.e., general appearance, eyes, mucous membranes, and tears) [10] The CDS has been incorporated in several clinical guidelines for appropriately managing acute gastroenteritis or dehydration [1, 13, 17, 25]. Although clinical guidelines aim to assist standardized assessment and treatment of dehydration, clinicians often do not adhere to the guidelines’ recommendations. Incorporating a guideline in an electronic, easily accessible clinical decision support system can improve guideline-adherence and therefore quality of care [18] In this study, we aimed to evaluate the feasibility of an electronic, easily accessible, guideline-based clinical decision support system, as well as the impact of this nurse-guided clinical decision support system for managing children with AGE at the ED compared to usual care on diagnostics, treatment and costs. Methods Design We conducted a randomized controlled trial comparing management of children with AGE at risk for dehydration by clinical decision support recommendations with usual care (Nederlands Trial Register (NTR), http://www.trialregister.nl/trialreg/index.asp; NTR2304).

In this study, we aimed to evaluate the feasibility of an electronic, easily accessible, guideline-based clinical decision support system, as well as the impact of this nurse-guided clinical decision support system for managing children with AGE at the ED compared to usual care on diagnostics, treatment and costs. Methods Design We conducted a randomized controlled trial comparing management of children with AGE at risk for dehydration by clinical decision support recommendations with usual care (Nederlands Trial Register (NTR), http://www.trialregister.nl/trialreg/index.asp; NTR2304). Patients and setting We included children with acute vomiting and/ or diarrhea, aged 1 month to 5 years, who visited the ED of the Erasmus MC-Sophia, Rotterdam between May 2010 and December 2012. The Erasmus MC-Sophia is an inner-city pediatric university hospital, with annually 9000 children presenting at the ED [19]. About 35% of the ED population has chronic co-morbidity. [23] We excluded children with chronic diarrhea (>7 days), severe dehydration with hypovolemic shock, children with vomiting/diarrhea with a focus for another infectious disease (e.g., otitis media, urinary tract infection) and chronically ill children with complex needs. Ethical approval for this study was obtained by the institutional review board (IRB) of the Erasmus MC. Informed consent was required and obtained from all parents (MEC-2008-071).

Patients and setting We included children with acute vomiting and/ or diarrhea, aged 1 month to 5 years, who visited the ED of the Erasmus MC-Sophia, Rotterdam between May 2010 and December 2012. The Erasmus MC-Sophia is an inner-city pediatric university hospital, with annually 9000 children presenting at the ED [19]. About 35% of the ED population has chronic co-morbidity. [23] We excluded children with chronic diarrhea (>7 days), severe dehydration with hypovolemic shock, children with vomiting/diarrhea with a focus for another infectious disease (e.g., otitis media, urinary tract infection) and chronically ill children with complex needs. Ethical approval for this study was obtained by the institutional review board (IRB) of the Erasmus MC. Informed consent was required and obtained from all parents (MEC-2008-071). Standard of care: initial patient assessment and treatment Following the standard of care, during the process of triage, trained nurses registered vital signs and weight, as well as signs and symptoms and risk factors for dehydration for all patients [24].

Ethical approval for this study was obtained by the institutional review board (IRB) of the Erasmus MC. Informed consent was required and obtained from all parents (MEC-2008-071). Standard of care: initial patient assessment and treatment Following the standard of care, during the process of triage, trained nurses registered vital signs and weight, as well as signs and symptoms and risk factors for dehydration for all patients [24]. Patients randomized to usual care were evaluated by the attending physician who subsequently decided on further rehydration management based on the patients clinical assessment and estimated level of dehydration. Our current guidelines advised rehydration treatment of 10–20 ml/kg/h during the length of stay at ED, followed by parental guidance on fluid maintenance as well as treatment of ongoing losses was advised [8] If oral rehydration did not succeed due to refusal of oral intake ór persistent vomiting, secondly, rehydration by a nasogastric tube was started. As rehydration therapy already was mostly based on oral rehydration using ORS, they received ORS or any rehydration fluid as prescribed by the attending physician. Anti-emetics were not a part of our national guideline, as evidence on the use of ant-emetics, as well as on safety of ondansetron, was lacking. The guideline could be retrieved from the protocol server of the Erasmus MC website on initiative of the clinician.

rehydration fluid as prescribed by the attending physician. Anti-emetics were not a part of our national guideline, as evidence on the use of ant-emetics, as well as on safety of ondansetron, was lacking. The guideline could be retrieved from the protocol server of the Erasmus MC website on initiative of the clinician. The intervention The clinical dehydration scale and current guidelines on treatment of AGE were incorporated in an electronic, easily accessible clinical decision support system, available at each desktop at the ED [10] (Fig. 1) [1, 13, 17, 25]. At the time, anti-emetics were not a part of our national guidelines, as evidence on the use of ant-emetics, as well as on safety of ondansetron, was lacking. Therefore, we did not incorporate it in our electronic clinical decision support system. Through the clinical decision support system, structured data were collected by the nurses on clinical signs and symptoms of all included patients. As the actual intervention included a treatment advice, most important difference with the usual care concerned a standardized amount of ORS for every dehydration level.

The intervention The clinical dehydration scale and current guidelines on treatment of AGE were incorporated in an electronic, easily accessible clinical decision support system, available at each desktop at the ED [10] (Fig. 1) [1, 13, 17, 25]. At the time, anti-emetics were not a part of our national guidelines, as evidence on the use of ant-emetics, as well as on safety of ondansetron, was lacking. Therefore, we did not incorporate it in our electronic clinical decision support system. Through the clinical decision support system, structured data were collected by the nurses on clinical signs and symptoms of all included patients. As the actual intervention included a treatment advice, most important difference with the usual care concerned a standardized amount of ORS for every dehydration level. If randomized to the intervention, the decision support system generated a guideline-based rehydration advice corresponding to the level of dehydration of the patient (Fig. 1). The nurse generated the rehydration advice by the clinical decision support system and started the rehydration. Children with mild or moderate dehydration received orally ORS 15 ml/kg/h. Children with signs of moderate dehydration and/ or persistent vomiting received 80 ml/kg ORS per nasogastric tube in 3 h. Children without clinical signs of dehydration also started treatment to prevent dehydration and to assess tolerability of oral fluids, whilst assessing volume of ongoing losses.

ly ORS 15 ml/kg/h. Children with signs of moderate dehydration and/ or persistent vomiting received 80 ml/kg ORS per nasogastric tube in 3 h. Children without clinical signs of dehydration also started treatment to prevent dehydration and to assess tolerability of oral fluids, whilst assessing volume of ongoing losses. The randomization was computer-generated and integrated in the clinical decision support system (randomly assigned to both groups depending on even and odd seconds of the digital computer clock). All patients, irrespective of randomization, were evaluated within the time frame generated by the triage system as well as discharged after rehydration by the attending physician. All the clinical dehydration score items had to be completed in order to get a rehydration advice in the CDS. Nurses were blinded for the contribution of predictors on the risk score. If nurse were in doubt on diagnosis and/or starting treatment, they could, at any stage, overrule the advice of the intervention and consult the attending physician. Creating an optimal environment for implementation of the decision support system, we created group lectures for nurses at the start of their shift, repeated individual briefings and reminders by posters, email and newsletters periodically. The implementation process was closely monitored and evaluated [4].

The randomization was computer-generated and integrated in the clinical decision support system (randomly assigned to both groups depending on even and odd seconds of the digital computer clock). All patients, irrespective of randomization, were evaluated within the time frame generated by the triage system as well as discharged after rehydration by the attending physician. All the clinical dehydration score items had to be completed in order to get a rehydration advice in the CDS. Nurses were blinded for the contribution of predictors on the risk score. If nurse were in doubt on diagnosis and/or starting treatment, they could, at any stage, overrule the advice of the intervention and consult the attending physician. Creating an optimal environment for implementation of the decision support system, we created group lectures for nurses at the start of their shift, repeated individual briefings and reminders by posters, email and newsletters periodically. The implementation process was closely monitored and evaluated [4]. Data collection We prospectively collected patient characteristics, data on signs and symptoms, vital signs, diagnostic tests, presumed diagnoses, treatment, referral, and discharge in a structured electronic hospital patient record system [21] During the study period, compliance of CDS recommendations was measured and checked with digital logbook information generated by the clinical decision support system.

ms, vital signs, diagnostic tests, presumed diagnoses, treatment, referral, and discharge in a structured electronic hospital patient record system [21] During the study period, compliance of CDS recommendations was measured and checked with digital logbook information generated by the clinical decision support system. To ensure correct diagnosis and ruling-out the possibility a complicated disease course, and be informed about revisits, telephonic follow-up was performed in all patients with standardized questionnaires 3 days after ED-discharge. Outcome measures Feasibility was measured by compliance of the nurses to the recommendations generated by the clinical decision support system. Outcome measures included length of stay (LOS, based on triage registration until the moment of ED-discharge) at the ED, the number of diagnostic tests (electrolytes, acid-base analysis), treatment and follow-up (telephonic consultation, outpatient clinic visit, ED revisit, hospitalization), and costs. In order to determine the association between level of dehydration and weight change, the ED nurse measures the weight of all included patients at triage and intended weight measurement 24 h after discharge.

atment and follow-up (telephonic consultation, outpatient clinic visit, ED revisit, hospitalization), and costs. In order to determine the association between level of dehydration and weight change, the ED nurse measures the weight of all included patients at triage and intended weight measurement 24 h after discharge. Statistical analysis Power analysis Based on previous research at our ED, inclusion of 450 children with acute vomiting/diarrhea in 24 months was expected [22] Initial power estimates were based on the number of correct diagnosed dehydrated children, based on weight change, and its associated false positives and false negatives as described in the trial register. Despite extensive efforts, we did not succeed in repetitive weight measurements in our population, due to lack of cooperation of parents to determine weight after 24 h, neither at the ED nor elsewhere. Therefore, we were forced to recalculate power on LOS at the ED. In order to detect a reduction of 10 min consultation time (30 min standard deviation), 99 patients had to be included in each group for reliable assessment of the actual impact on LOS with a power of 0.8 and an alpha of 0.05 (one-sided test). Evaluation of the clinical decision system Being an randomized controlled trial an intention-to-treat analysis was performed.

Statistical analysis Power analysis Based on previous research at our ED, inclusion of 450 children with acute vomiting/diarrhea in 24 months was expected [22] Initial power estimates were based on the number of correct diagnosed dehydrated children, based on weight change, and its associated false positives and false negatives as described in the trial register. Despite extensive efforts, we did not succeed in repetitive weight measurements in our population, due to lack of cooperation of parents to determine weight after 24 h, neither at the ED nor elsewhere. Therefore, we were forced to recalculate power on LOS at the ED. In order to detect a reduction of 10 min consultation time (30 min standard deviation), 99 patients had to be included in each group for reliable assessment of the actual impact on LOS with a power of 0.8 and an alpha of 0.05 (one-sided test). Evaluation of the clinical decision system Being an randomized controlled trial an intention-to-treat analysis was performed. Feasibility was measured by comparing treatment advice generated by the clinical decision support system with the actual treatment using chi-square analysis. Outcome measures were evaluated using chi-square and Student’s t test. Due to failure of measurement of post ED weight, we had to delete the outcome for correct diagnosis of dehydration. A p value <0.05 was considered statistically significant. We used SPSS version 20.0 for Windows.

treatment using chi-square analysis. Outcome measures were evaluated using chi-square and Student’s t test. Due to failure of measurement of post ED weight, we had to delete the outcome for correct diagnosis of dehydration. A p value <0.05 was considered statistically significant. We used SPSS version 20.0 for Windows. Cost analysis Cost analysis was performed from the hospital perspective (Appendix) [14]. Medical costs were calculated by multiplying the volumes of health care use with unit prices. We used real unit prices when available; otherwise, charges were used as a proxy for real costs. Salary schemes were used to calculate costs per hour for each health care worker. In-hospital medical costs included costs of initiated diagnostics and treatment, length of stay at the ED, hospitalization, and revisits. Volumes of diagnostics and treatment were measured according to the computer-based hospital information system. Effects of the clinical decision support system were defined as the differences in the number of false positive and false negative errors. Because the clinical decision support system resulted in comparable patient outcomes, a cost-minimization study was required. In a sensitivity analysis, we considered variation of doctor’s time and variation in costs of diagnostic tests and therapy.

differences in the number of false positive and false negative errors. Because the clinical decision support system resulted in comparable patient outcomes, a cost-minimization study was required. In a sensitivity analysis, we considered variation of doctor’s time and variation in costs of diagnostic tests and therapy. Results Of 915 eligible children visiting the ED with vomiting/diarrhea, 693 (75%) children were not included due to early assessment of the physician before randomization and lack of time for the nurses to obtain informed consent. We could include 222 children with informed consent in the randomized controlled trial (Fig. 2). Compared with the included population, the eligible population included more highly urgent patients according to the Manchester Triage System (indicating physicians’ evaluation within 10 min) and more patients with increased heart rate.Fig. 2 Patient flow chart Fig. 1 Clinical decision support system The intervention (N = 113) and the control group (N = 109) were comparable with respect to age (median age 1.5 years (IQR 0.9–2.4) versus 1.3 years (IQR 0.8–2.4) in the usual care group), gender, triage urgency, vital signs, and CDS-items (Table.1).Table 1 Patient characteristics Intervention N = 113 (100%)a Usual care N = 109 (100%)a

The intervention (N = 113) and the control group (N = 109) were comparable with respect to age (median age 1.5 years (IQR 0.9–2.4) versus 1.3 years (IQR 0.8–2.4) in the usual care group), gender, triage urgency, vital signs, and CDS-items (Table.1).Table 1 Patient characteristics Intervention N = 113 (100%)a Usual care N = 109 (100%)a Age (years)b 1.5 (0.9–2.4) 1.3 (0.8–2.4) Sex, male 61 (54.0) 57 (52.3) Vital signs Temperaturec (°C) 37.6 (1) 37.7 (1) Heart ratec (beats per minute) 127 (24) 128 (21) Respiratory ratec (breaths per minute) 30 (8) 30 (8.4) MTS urgency Emergent 0 (0) 1 (0.9) Very urgent 13 (11.5) 11 (10.2) Urgent 45 (39.8) 57 (52.8) Standard 53 (46.9) 39 (36.1) Non-urgent missing 2 (1.8) 0 (0.0) 0 (0) 1 (0.9) Referral Own initiative 72 (63.7) 59 (54.1) Primary care 31 (27.4) 35 (32.1) Ambulance 4 (3.5) 1 (0.9) Othersd 6 (5.4) 14 (12.9) Clinical dehydration score overall Mild 72 (63.7) 61 (56.0) Moderate 41 (36.3) 47 (43.1) Severe 0 (0) 1 (0.9) Clinical dehydration score variables General appearance Normal 101 (89.4) 96 (88.1) Thirsty/ restless/ irritable 11 (9.7) 13 (11.9) Drowsy/ comatose 1 (0.9) 0 (0) Eyes Normal 91 (80.5) 79 (72.5) Slightly sunken 22 (19.5) 27 (24.8) Very sunken 0 (0) 3 (2.8) Mucous membranes Moist 101 (89.4) 91 (83.5) Dry 12 (10.6) 18 (16.5) Very dry 0 (0) 0 (0) Tears Present 98 (86.7) 90 (82.6) Decreased 10 (8.8) 15 (13.8) Absent 5 (4.4) 4 (3.7) aAbsolute number (percentage) bMedian (IQR) cMean (SD) dOthers include secondary care and after telephone contact

Age (years)b 1.5 (0.9–2.4) 1.3 (0.8–2.4) Sex, male 61 (54.0) 57 (52.3) Vital signs Temperaturec (°C) 37.6 (1) 37.7 (1) Heart ratec (beats per minute) 127 (24) 128 (21) Respiratory ratec (breaths per minute) 30 (8) 30 (8.4) MTS urgency Emergent 0 (0) 1 (0.9) Very urgent 13 (11.5) 11 (10.2) Urgent 45 (39.8) 57 (52.8) Standard 53 (46.9) 39 (36.1) Non-urgent missing 2 (1.8) 0 (0.0) 0 (0) 1 (0.9) Referral Own initiative 72 (63.7) 59 (54.1) Primary care 31 (27.4) 35 (32.1) Ambulance 4 (3.5) 1 (0.9) Othersd 6 (5.4) 14 (12.9) Clinical dehydration score overall Mild 72 (63.7) 61 (56.0) Moderate 41 (36.3) 47 (43.1) Severe 0 (0) 1 (0.9) Clinical dehydration score variables General appearance Normal 101 (89.4) 96 (88.1) Thirsty/ restless/ irritable 11 (9.7) 13 (11.9) Drowsy/ comatose 1 (0.9) 0 (0) Eyes Normal 91 (80.5) 79 (72.5) Slightly sunken 22 (19.5) 27 (24.8) Very sunken 0 (0) 3 (2.8) Mucous membranes Moist 101 (89.4) 91 (83.5) Dry 12 (10.6) 18 (16.5) Very dry 0 (0) 0 (0) Tears Present 98 (86.7) 90 (82.6) Decreased 10 (8.8) 15 (13.8) Absent 5 (4.4) 4 (3.7) aAbsolute number (percentage) bMedian (IQR) cMean (SD) dOthers include secondary care and after telephone contact Compliance to the clinical decision support system In the intervention group, 72/113 patients (64%) were mildly and 41/113 (36%) moderately dehydrated children, compared with 61/109 (56%) and 47/109 ( 43%) in the control group. In the intervention group, 4/88 (4.5%) children received oral rehydration solution (ORS) via a nasogastric tube. These four children refused to drink ORS (Table 2). Twelve of 25 (48.0%) children assigned to nasogastric tube rehydration by the clinical decision support system, drank the ORS themselves instead. Compliance of the nurse to the treatment advice occurred in 97/113 patients (86%; CI 95% 0.78–0.92).Table 2 Outcome measures

ese four children refused to drink ORS (Table 2). Twelve of 25 (48.0%) children assigned to nasogastric tube rehydration by the clinical decision support system, drank the ORS themselves instead. Compliance of the nurse to the treatment advice occurred in 97/113 patients (86%; CI 95% 0.78–0.92).Table 2 Outcome measures Intervention N = 113 (100%)a Usual care N = 109 (100%)a p value Compliance Advice oral rehydration 88 Not applicable Compliance 84 Non-complianceb 4 Advice nasogastric tube rehydration 25 Compliance 13 Non- compliancec 12 Patient consultation time Time spent at the ED (min)d 136 (98–206) 133 (92–184) NS Diagnostic procedures performed NS Electrolytes Acid-base balance 15 (13.3) 13 (11.5) 23 (21.1) 16 (14.7) Treatment procedures performed 0.01 ORS oral 73 (64.6) 57 (52.3) 0.04 ORS nasogastric tube 17 (15.0) 9 (8.3) NS Intravenous rehydration 0 (0.0) 2 (1.8) NS Liquid other 18 (15.9) 30 (27.5) 0.02 Unknown 5 (4.4) 11 (10.1) NS Follow-up NS No 57 (50.4) 59 (54.1) Hospitalization 10 (8.8) 7 (6.4) Outpatient clinic 25 (22.1) 26 (23.9) Telephonic consultation 21 (18.6) 15 (13.8) Other 0 (0.0) 2 (1.8) Revisits 30 (26.5) 20 (27.5) Hospitalization after revisit 1 (0.9) 4 (3.7) aAbsolute number (percentage) bNasogastric tube by nurse in patient with oral ORS advice cOral rehydration in patient with nasogastric tube rehydration advice dMedian (IQR)

p value Compliance Advice oral rehydration 88 Not applicable Compliance 84 Non-complianceb 4 Advice nasogastric tube rehydration 25 Compliance 13 Non- compliancec 12 Patient consultation time Time spent at the ED (min)d 136 (98–206) 133 (92–184) NS Diagnostic procedures performed NS Electrolytes Acid-base balance 15 (13.3) 13 (11.5) 23 (21.1) 16 (14.7) Treatment procedures performed 0.01 ORS oral 73 (64.6) 57 (52.3) 0.04 ORS nasogastric tube 17 (15.0) 9 (8.3) NS Intravenous rehydration 0 (0.0) 2 (1.8) NS Liquid other 18 (15.9) 30 (27.5) 0.02 Unknown 5 (4.4) 11 (10.1) NS Follow-up NS No 57 (50.4) 59 (54.1) Hospitalization 10 (8.8) 7 (6.4) Outpatient clinic 25 (22.1) 26 (23.9) Telephonic consultation 21 (18.6) 15 (13.8) Other 0 (0.0) 2 (1.8) Revisits 30 (26.5) 20 (27.5) Hospitalization after revisit 1 (0.9) 4 (3.7) aAbsolute number (percentage) bNasogastric tube by nurse in patient with oral ORS advice cOral rehydration in patient with nasogastric tube rehydration advice dMedian (IQR) There were significant differences in rehydration treatment between the intervention group and the usual care group (p 0.01). In the intervention group, 90/113 (80%) patients received ORS, compared with 66/109 (61%) in the usual care group, fewer patients received other liquids instead of ORS ( 18/113 (16%) vs 30/109 (28%) in the usual care group), and no patients received intravenous rehydration compared with 2/109(1.8%) patients in the usual care group. In the intervention group, 10/113 (8.8%) patients were hospitalized compared with 7/109 (6.4%) in the usual care group (p 0.47). After discharge from the ED, 30/113 (26.5%) of the intervention group revisited the ED; one (0.9%) of these patients was hospitalized. In the usual care group, 30/109 (27.5%) revisited the ED; 4/109 (3.7%) patients were hospitalized.

atients were hospitalized compared with 7/109 (6.4%) in the usual care group (p 0.47). After discharge from the ED, 30/113 (26.5%) of the intervention group revisited the ED; one (0.9%) of these patients was hospitalized. In the usual care group, 30/109 (27.5%) revisited the ED; 4/109 (3.7%) patients were hospitalized. Impact of the clinical decision support system We did not find differences between the intervention and the control group for LOS at the ED (Table 2). We observed a non-significant trend in reduction of laboratory tests, which were decreased with 50%. The number of revisits or hospitalization did not differ. All parents of children were contacted for follow-up, with a median follow-up time of 72 h. We did not observe adverse events.

group for LOS at the ED (Table 2). We observed a non-significant trend in reduction of laboratory tests, which were decreased with 50%. The number of revisits or hospitalization did not differ. All parents of children were contacted for follow-up, with a median follow-up time of 72 h. We did not observe adverse events. Costs In the cost minimization study, no differences in costs in the intervention group compared with the usual care group were detected (Table 3). The differences in patient outcome between the intervention group and the usual care group consisted of more rehydration by nasogastric tube, without adverse events, and were therefore not regarded as long-term effect. Because nasogastric tubes and diagnostic tests only account for a very small part of (health care) costs, differences did not significantly influence total costs. A cost-minimization study showed comparable costs for the intervention group and the usual care group: mean costs per patient were 346 euro in the intervention group and 350 euro in the usual care group. Total (once-only) costs for development and implementation of the clinical decision support system amounted to 7000 euro/US dollar 7770 (EUR/USD 1.11; currency rate dated 8th Nov 2016).Table 3 Cost-analysis (Euro)

roup: mean costs per patient were 346 euro in the intervention group and 350 euro in the usual care group. Total (once-only) costs for development and implementation of the clinical decision support system amounted to 7000 euro/US dollar 7770 (EUR/USD 1.11; currency rate dated 8th Nov 2016).Table 3 Cost-analysis (Euro) Intervention (N = 113) Usual care (N = 109) Cost price Volume Costs Volume Costs CDSS Development (h) 36 144 5184 – – Implementation Researcher (h) 54 4 216 – – Nurse (number × h) 40 20 × 2 1600 – – Total costs CDSS implementation 7000 ED visit Physician (h × visit number) 68 0.33 × 113 2535 0.33 × 109 2445 Nurse (h × visit number) 40 0.5 × 113 2260 0.5 × 109 2180 Hospital costs 114 113 12,882 109 12,426 Diagnostics Electrolytes 3.8 15 57 23 87 Acid-base balance 5.10 13 66 16 82 Treatment Unknown 6 11 ORS portion 0.3 73 22 57 17 Liquid other NA 18 NA 30 NA ORS nasogastric tube 28.2 17 479 9 254 IV rehydration 5.0 0 5 2 10 Follow-up/hospitalization Hospitalization patients × (LOS days) 575 10 × 2 11,500 7 × 2 8050 Outpatient clinic 129 25 3225 26 3354 Telephonic follow-up 20 21 420 15 300 Costs of missed diagnoses/adverse events Revisit ED 144 30 4464 30 4320 Admission after revisit (LOS days × patients) 575 2 × 1 1150 2 × 4 4600 Mean costs per patient (including CDSS) 408 350 Mean costs per patient (without CDSS) 346 350 Currency rate EUR/USD 1.11; 8th Nov 2016 LOS length of stay

Intervention (N = 113) Usual care (N = 109) Cost price Volume Costs Volume Costs CDSS Development (h) 36 144 5184 – – Implementation Researcher (h) 54 4 216 – – Nurse (number × h) 40 20 × 2 1600 – – Total costs CDSS implementation 7000 ED visit Physician (h × visit number) 68 0.33 × 113 2535 0.33 × 109 2445 Nurse (h × visit number) 40 0.5 × 113 2260 0.5 × 109 2180 Hospital costs 114 113 12,882 109 12,426 Diagnostics Electrolytes 3.8 15 57 23 87 Acid-base balance 5.10 13 66 16 82 Treatment Unknown 6 11 ORS portion 0.3 73 22 57 17 Liquid other NA 18 NA 30 NA ORS nasogastric tube 28.2 17 479 9 254 IV rehydration 5.0 0 5 2 10 Follow-up/hospitalization Hospitalization patients × (LOS days) 575 10 × 2 11,500 7 × 2 8050 Outpatient clinic 129 25 3225 26 3354 Telephonic follow-up 20 21 420 15 300 Costs of missed diagnoses/adverse events Revisit ED 144 30 4464 30 4320 Admission after revisit (LOS days × patients) 575 2 × 1 1150 2 × 4 4600 Mean costs per patient (including CDSS) 408 350 Mean costs per patient (without CDSS) 346 350 Currency rate EUR/USD 1.11; 8th Nov 2016 LOS length of stay Considering a reduction of 50% doctor’s consultation time, sensitivity analysis showed a reduction of costs by 3% (12 euro/13.2 USD). Further sensitivity analyses were not performed due to relatively low impact of diagnostic and treatment costs compared with total costs.

Intervention (N = 113) Usual care (N = 109) Cost price Volume Costs Volume Costs CDSS Development (h) 36 144 5184 – – Implementation Researcher (h) 54 4 216 – – Nurse (number × h) 40 20 × 2 1600 – – Total costs CDSS implementation 7000 ED visit Physician (h × visit number) 68 0.33 × 113 2535 0.33 × 109 2445 Nurse (h × visit number) 40 0.5 × 113 2260 0.5 × 109 2180 Hospital costs 114 113 12,882 109 12,426 Diagnostics Electrolytes 3.8 15 57 23 87 Acid-base balance 5.10 13 66 16 82 Treatment Unknown 6 11 ORS portion 0.3 73 22 57 17 Liquid other NA 18 NA 30 NA ORS nasogastric tube 28.2 17 479 9 254 IV rehydration 5.0 0 5 2 10 Follow-up/hospitalization Hospitalization patients × (LOS days) 575 10 × 2 11,500 7 × 2 8050 Outpatient clinic 129 25 3225 26 3354 Telephonic follow-up 20 21 420 15 300 Costs of missed diagnoses/adverse events Revisit ED 144 30 4464 30 4320 Admission after revisit (LOS days × patients) 575 2 × 1 1150 2 × 4 4600 Mean costs per patient (including CDSS) 408 350 Mean costs per patient (without CDSS) 346 350 Currency rate EUR/USD 1.11; 8th Nov 2016 LOS length of stay Considering a reduction of 50% doctor’s consultation time, sensitivity analysis showed a reduction of costs by 3% (12 euro/13.2 USD). Further sensitivity analyses were not performed due to relatively low impact of diagnostic and treatment costs compared with total costs. Discussion We observed good compliance to the recommendations of the clinical decision support system for early rehydration in children with AGE by ED nurses and a significant increase in appropriate use of ORS compared to usual care. We did not observe adverse events. However, despite a stricter rehydration policy, we did not observe more successful rehydration in this mildly dehydrated study population, as expressed in (lack of) differences in revisits, hospitalization rates or costs.

cant increase in appropriate use of ORS compared to usual care. We did not observe adverse events. However, despite a stricter rehydration policy, we did not observe more successful rehydration in this mildly dehydrated study population, as expressed in (lack of) differences in revisits, hospitalization rates or costs. Adherence to the treatment advice generated by the intervention was good; however, compliance to use the application of the clinical decision support system can be influenced by several factors, such as crowding hours and increased time needed to use a clinical decision support system. In our study, we indeed observed that, during crowded ED hours, the nurses did not complete all clinical dehydration score items, a prerequisite for including patients. This may point out the limitations of the implementation process. Notably, we noticed reduced inclusion of children with more severe dehydration. The substantial loss of eligible patients, and in particular those with more severe dehydration, may reflect problems of performing research and obtaining informed consent in emergency care settings as was noted by others previously [16]. Given the randomized controlled character of our trial, we believe our results remain valid. Non-included patients, however, might affect generalization of our results to more severe dehydrated children. Therefore, extrapolating our results to all dehydrated children with signs of dehydration should be done with care. The high compliance to recommendations on treatment of the clinical decision support system in the included population is encouraging for further implementation. Factors contributing to this high compliance can be explained: first, our nursing staff consists of trained pediatric nurses, who who were skilled and experienced in the clinical assessment of dehydrated children. Second, an implementation program was used [12] Third, treatment recommendations of the clinical decision support system were based on a pre-existing rehydration protocol, with which the medical staff was already familiar. Last, the clinical decision support system was designed for easy use with readily available standardized clinical items, and was accessible from every computer at the ED [3, 5].

of the clinical decision support system were based on a pre-existing rehydration protocol, with which the medical staff was already familiar. Last, the clinical decision support system was designed for easy use with readily available standardized clinical items, and was accessible from every computer at the ED [3, 5]. According to the rehydration treatment guidelines underlying the decision support system, oral rehydration with ORS is recommended and biochemical testing is only indicated in severely dehydrated children. In the study population of mildly dehydrated children, the clinical decision support system showed a trend to fewer diagnostic tests and more frequent use of standardized amounts of ORS. We did not find a beneficial effect on costs. A major reason is our low hospitalization rate, as hospitalization dominates the costs in AGE management. Nurse-guided patient assessment and treatment in the intervention group suggest that patient flow can be managed more efficiently. Sensitivity analysis on the reduction of doctor’s consultation time by 50% showed a reduction of costs by 3% (12 euro). Although this estimation is hypothetical, this might imply lower costs due to a reallocation of tasks as already described in emergency medicine, HIV care, and mental health care [2, 6, 7, 9].

iently. Sensitivity analysis on the reduction of doctor’s consultation time by 50% showed a reduction of costs by 3% (12 euro). Although this estimation is hypothetical, this might imply lower costs due to a reallocation of tasks as already described in emergency medicine, HIV care, and mental health care [2, 6, 7, 9]. Strength and limitations The main strength of this study is a randomized clinical trial on the impact of a clinical decision support system in pediatric emergency care. Impact analysis completes the final step in the translation process of clinical decision rules to routine practice [20]. We applied the decision support system in an electronic environment enabling easy access in routine practice and easy adaptation by other EDs.

cal decision support system in pediatric emergency care. Impact analysis completes the final step in the translation process of clinical decision rules to routine practice [20]. We applied the decision support system in an electronic environment enabling easy access in routine practice and easy adaptation by other EDs. One of our limitations is the absence of inclusion criteria based on strict signs and symptoms. Rather, we aimed for a pragmatic approach, including children with vomiting and/or diarrhea in the absence of another clear infectious focus, instead of using a strict definition based on signs and symptoms [13]. Secondly, despite our best efforts, we could not recruit the calculated number of patients. During the study period, we observed no epidemic of acute gastroenteritis. Also, we were confronted with higher than anticipated exclusion rates for serious co-morbidity as well as a high number of non-included eligible patients. Hence, we abandoned our envisioned primary endpoint and recalculated power for length of stay at the ED (LOS). One could argue that our main outcome measure LOS is subject to influence of other factors, such as admission method, discharge destination, provider (hospital) type, and specialty (acute vs non-acute). Especially, discharge destination and admission methods appeared important influencing factors [15]. As we observed hospitalization in only resp. 10 (8.8%) and 7 (6.4%) of all patients and all other patients are discharged for further treatment at home directly after finishing rehydration treatment, we think LOS was a valid outcome in our study. We expect that the presence of dedicated research personnel at the ED would have improved the whole research process, but especially the inclusion of eligible patients during crowding at the ED. Third, we could not evaluate impact on diagnosis due to failure of measurement of post ED weight. Fourth, the limited impact on (secondary) process outcome measures may be caused by the population of rather mildly dehydrated children, evaluated at a relative low-volume university hospital, and the high experienced medical and nursing staff in evaluating children. Although we proved a significant increase in appropriate use of ORS in our study, we expect larger impact also on process outcome measures in more severely dehydrated patients, or in settings with high volume or less experienced personnel.

ital, and the high experienced medical and nursing staff in evaluating children. Although we proved a significant increase in appropriate use of ORS in our study, we expect larger impact also on process outcome measures in more severely dehydrated patients, or in settings with high volume or less experienced personnel. Finally, 12 out of 25 children (48%) assigned to nasogastric route in the intervention group drank their ORS and therefore did not receive nasogastric tube. The decision support system may induce excessive use of nasogastric route. This result highlights the moderate validity of the clinical decision support system and the risks associated with the indiscriminate use of the system. The clinical decision support system must therefore be considered only as an additional tool and should not replace common sense. Conclusion Implementation of a nurse-guided clinical decision support system on rehydration treatment in children with AGE showed high compliance and an increase standardized use of ORS, without differences the number of diagnostic tests, LOS, revisits, and hospitalization or costs. Abbreviations AGEAcute gastroenteritis CDSClinical dehydration Scale EDEmergency department LOSLength of stay ORSOral rehydration solution Revisions received: 29 September 2016; 15 November 2016 Acknowledgement Nurses at Emergency Department ErasmusMC-Sophia and M. de Wilde. Authors’ Contributions Dorien Geurts developed the research protocol, implemented the intervention, collected and analyzed the data, and is the main author of this manuscript.

ORSOral rehydration solution Revisions received: 29 September 2016; 15 November 2016 Acknowledgement Nurses at Emergency Department ErasmusMC-Sophia and M. de Wilde. Authors’ Contributions Dorien Geurts developed the research protocol, implemented the intervention, collected and analyzed the data, and is the main author of this manuscript. Evelien de Vos-Kerkhof contributed to the protocol development, participated substantially in the data collection and analysis, and writing of the manuscript. Suzanne Polinder contributed to and supervised the economic evaluation. Henriette Moll, Ewout Steyerberg, and Johan van der Lei supervised the development of the research protocol, interpretation of the results, and writing of the manuscript. Rianne Oostenbrink was primarily responsible for the study and was responsible for obtaining funding. She supervised development of the research protocol, data analysis, and interpretation of the results, and writing of the manuscript. Dorien Geurts and Rianne Oostenbrink had full access to all the data and take responsibility for the integrity of the data and the accuracy of the data analysis. All authors read and approved the final manuscript. Compliance with ethical standards Funding Dorien Geurts is supported by Kroger Foundation and Sophia Research Foundation. Evelien de Vos-Kerkhof is supported by Erasmus MC Grant/ Dutch National Health Council (ZonMW). Rianne Oostenbrink is supported by a fellowship grant of the European Society of Pediatric Infectious Diseases in 2010. Conflict of interest The authors declare that they have no conflict of interest.

Compliance with ethical standards Funding Dorien Geurts is supported by Kroger Foundation and Sophia Research Foundation. Evelien de Vos-Kerkhof is supported by Erasmus MC Grant/ Dutch National Health Council (ZonMW). Rianne Oostenbrink is supported by a fellowship grant of the European Society of Pediatric Infectious Diseases in 2010. Conflict of interest The authors declare that they have no conflict of interest. Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent Informed consent was obtained from all individual participants included in the study.

Case description A 34-year-old man with dysmorphic features, osteoporosis and recurrent fragility fractures with non-union was referred to the department of clinical genetics. At age 9, a clinical diagnosis of calcinosis cutis, osteoma cutis, poikiloderma and skeletal abnormalities (COPS syndrome) was made and reported in this journal [11]. In summary, he had dysmaturity (birth weight 2400 g at 40 weeks gestational age) and severe diarrhoea requiring parenteral feeding. At the age of 3, he suffered from meningitis due to mumps infection. Furthermore, at the age of 4, subcutaneous tumours, osteomas, with a maximal diameter of 3 cm were removed from the ankles, knees and forehead. Skeletal abnormalities were observed with hypoplastic patellae and delayed bone maturation. At the age of 15 years, a diagnosis of coeliac disease was made for which he was started on a diet and vitamin D supplementation. He suffered from multiple fragility fractures of both the tibiae, the right elbow, the left patella and the metatarsal bone V of his right foot, complicated by pseudo arthroses. A bone mass density measurement was performed at age 27, showing a T-score of −2.6 femur and 2.4 lumbar vertebrae, consistent with a diagnosis of osteoporosis. Treatment with alendronate was initiated and while on treatment his bone mass increased and no fractures occurred. Treatment was discontinued at the age of 30 and at age 34. He suffered from a mild intellectual disability for which he attended special educational programs.

, consistent with a diagnosis of osteoporosis. Treatment with alendronate was initiated and while on treatment his bone mass increased and no fractures occurred. Treatment was discontinued at the age of 30 and at age 34. He suffered from a mild intellectual disability for which he attended special educational programs. At age 34, he was referred to our hospital for a second opinion regarding his osteoporosis and non-union of a tibia fracture. Physical examination showed a slender man with a short stature (height 167 cm, (−2.1 SDS), with a saddle nose, absence of eyelashes and eyebrows and facial poikiloderma, (Fig. 1). He had sparse hair on the scalp, with two spots of alopecia areata and multiple small hyperpigmented macules on the trunk and arms. He had small but normal hands (Fig. 2); at the right foot, a partial 2–3 syndactyly of the right foot was observed.Fig. 1 Facial features at age 34. Note: absent eye brows and eye lashes, small nose Fig. 2 Hands at age 34. Note: relatively small hands, with small though normally shaped nails, normal thumbs His 5 years older brother has a milder phenotype, with a similar physical appearance and a mild intellectual disability. He was diagnosed with osteopenia and recurrent fragility fractures with non-union, but not with celiac disease (Table 1). Their parents were non-consanguineous.Table 1 Clinical features in our patient and his brother compared with the frequencies of these features among previously reported Rothmund-Thomson patients Our patient Brother of the patient Reported frequency of RTS featuresa

His 5 years older brother has a milder phenotype, with a similar physical appearance and a mild intellectual disability. He was diagnosed with osteopenia and recurrent fragility fractures with non-union, but not with celiac disease (Table 1). Their parents were non-consanguineous.Table 1 Clinical features in our patient and his brother compared with the frequencies of these features among previously reported Rothmund-Thomson patients Our patient Brother of the patient Reported frequency of RTS featuresa Skin Poikiloderma + + All Hyperpigmentation + + + Hypopigmentation + + + Calcinosis cutis + − Uncommon Osteoma cutis + − − Palmoplantar hyperkeratosis − − 30% Photosensitivity − − +/− Hair 50% Sparse hair + + Absent eyelashes + + Sparse/absent eyebrows + + Alopecia areata + − Dental abnormalities + − 27–59% Growth Low birth weight + + + Short stature + + + Skeleton 68–75% Radial ray defects − − 20% Metaphyseal changes + − + Osteopenia/osteoporosis + + + Small patellae + − + Ocular lesions Cataract − − 10–50% Gastrointestinal features 17% Oesophageal or pyloric stenosis + − Feeding problems + − Chronic emesis/diarrhoea + − Hematologic abnormalities Hodgkin’s lymphoma − Occasionally Cancer Osteosarcoma − − 30% Skin cancer − − 5% Lymphoma + − Two casesb Neurocognitive development No specific data available Mild intellectual disability + + Cytogenetic abnormalities Mosaic trisomy 8 (15%) Mosaic trisomy 8 (13%) Cases reported with Mosaic trisomy 2, 7 or 8 Mosaic isochromosome 8q (13%)c Mosaic isochromosome 8q (9%)c Mosaic isochromosome 2, 7 or 8 RECQL4 gene mutations + + 66%

Neurocognitive development No specific data available Mild intellectual disability + + Cytogenetic abnormalities Mosaic trisomy 8 (15%) Mosaic trisomy 8 (13%) Cases reported with Mosaic trisomy 2, 7 or 8 Mosaic isochromosome 8q (13%)c Mosaic isochromosome 8q (9%)c Mosaic isochromosome 2, 7 or 8 RECQL4 gene mutations + + 66% aFrequencies derived from [2, 8, 16, 17] bCases reported by Siitonen 2009 and Simon 2010 [13, 14] cBased on interphase FISH (2 probes: LSI MYC, 8q24) on 400 lymphocytes nuclei in blood Altogether, this presentation was compatible with a clinical diagnosis of Rothmund-Thomson syndrome (RTS, OMIM 268400). Cytogenetic testing at the age of 13 showed a normal male karyotype in a total of 50 analysed nuclei without signs of chromosomal instability, while a CytoScan HD Array (Affymetrix) at the age of 34 showed a slight excess of chromosome 8q, suggestive for a mosaic chromosome 8q duplication (presumably between 12 and 18%). Subsequently, karyotyping and FISH analysis were performed on cultured lymphocytes from both brothers, showing a mosaicism for trisomy 8, isochromosome 8q and a normal karyotype (Table 1). Sanger sequencing of the RECQL4 gene (OMIM 603780) showed two compound heterozygous recurrent pathogenic mutations in both brothers: one frame shift mutation: c.1048_1049delAG (p.(Arg350fsX21)) and one splice site mutation c.1391-1G>A (p.(?)). Carrier testing in the parents confirmed biallelic inheritance. Altogether, these findings confirmed the clinical diagnosis of RTS in both brothers.

eterozygous recurrent pathogenic mutations in both brothers: one frame shift mutation: c.1048_1049delAG (p.(Arg350fsX21)) and one splice site mutation c.1391-1G>A (p.(?)). Carrier testing in the parents confirmed biallelic inheritance. Altogether, these findings confirmed the clinical diagnosis of RTS in both brothers. One year later, the index patient was diagnosed with a stage I Hodgkin’s lymphoma in the neck for which he was started on three cycles of adriamycin, bleomycin, vinblastin, darcabazin and prednisone in combination with involved node radiation with 12 × 1.8 Gy resulting in a complete remission. He developed fever and neutropenia after the first chemotherapy. The neutropenia has resolved after a temporary break of chemotherapy and treatment with antibiotics. Neutropenia did not return after continuation of the ABVD therapy. The radiotherapy did not lead to considerable side effects. Discussion In this report, we presented a case of RTS alternatively diagnosed as COPS. RTS is a rare autosomal recessive genodermatosis with a distinctive phenotype, characterised by poikiloderma, sparse hair, skeletal abnormalities and an increased risk for osteosarcoma [8]. RTS had been considered, but was assumed less likely, due to the absence of cataract and photophobia, a major sign, and the presence of osteoma cutis, whereas genetic confirmation was not possible in that period [11]. However re-evaluation of the patient and his brother changed the diagnosis to RTS and later the cataract was not associated with RECQL4 mutations anymore.

, due to the absence of cataract and photophobia, a major sign, and the presence of osteoma cutis, whereas genetic confirmation was not possible in that period [11]. However re-evaluation of the patient and his brother changed the diagnosis to RTS and later the cataract was not associated with RECQL4 mutations anymore. The absence of the cataract makes the distinction between the two clinical variants of RTS: the form with poikiloderma and ocular defects, named RTSI, and poikiloderma, skeleton defects, predisposition to cancer and RECQL4 mutations, named RTSII, which accounts for approximately 66% of RTS patients (Table 1). Osteoma cutis has not been described before; calcinosis cutis has been linked to RTS [1, 4].

of RTS: the form with poikiloderma and ocular defects, named RTSI, and poikiloderma, skeleton defects, predisposition to cancer and RECQL4 mutations, named RTSII, which accounts for approximately 66% of RTS patients (Table 1). Osteoma cutis has not been described before; calcinosis cutis has been linked to RTS [1, 4]. The RECQL4 gene on the long arm of chromosome 8 (8q24.3) codes for an ATP-dependent DNA helicase, which plays a role in regulating DNA replication, DNA repair and chromosomal integrity [6, 13]. RECQL4-deficient mice show abnormal karyotypes and aneuploidy [10], as well as defects in osteoblast progenitors [18]. In patients with RTS, these defects in osteoblast progenitors make them prone to osteosarcoma and low-turnover osteoporosis with a predisposition for fracture non-union [9, 18]. Chromosomal abnormalities, like the mosaic trisomy 8 and i(8q) have been reported in RTS patients [7]. The specific RECQL4 mutations, c.1048_1049delAG in exon 5 and c.1391-1G>A in intron 7, were both previously reported in patients with RTS [3, 8]. One patient was reported with the exact combination of mutations; contrary to our patient, this patient showed additional humoral immune deficiency and granulomatous skin lesions [3].

fic RECQL4 mutations, c.1048_1049delAG in exon 5 and c.1391-1G>A in intron 7, were both previously reported in patients with RTS [3, 8]. One patient was reported with the exact combination of mutations; contrary to our patient, this patient showed additional humoral immune deficiency and granulomatous skin lesions [3]. Generally, intellectual disability is not considered to be a feature of RTS [8, 13, 17]. As a consequence, the number of RTS patients with a mild intellectual disability, like our patients, may be underestimated. Mild to moderate intellectual disability has been reported in a small number of cases [5, 8, 15, 16]. Co-occurring features like hydrocephalus and craniosynostosis may have played a role in the ID [16], and in some cases the diagnosis of RTS was not confirmed [5]. In our case, the history of meningitis may have attributed to some part of the intellectual disability; however, the brother was affected as well, implying a relationship with RTS. We hypothesised that the mosaic chromosomal abnormalities observed are responsible for the development of intellectual disability. However, the chromosomal imbalances were apparently acquired. Additionally, mutations in RecQ DNA helicase genes may mildly impact intellectual development, such as observed in Bloom syndrome (OMIM 210900), a chromosomal breakage syndrome, caused by mutations in the RECQL3 gene (OMIM 604610). Although most affected individuals with Bloom syndrome have normal intellectual ability, many exhibit learning disability [12]. Likewise, a relative mild intellectual disability, such as in our patients, may be underreported in RTS patients and considered to be normal variability of intelligence. Modifier genes and co-morbidity may also play a role in the variability and atypical expression with or without ID; these may explain the phenotypic differences between the brothers.

ectual disability, such as in our patients, may be underreported in RTS patients and considered to be normal variability of intelligence. Modifier genes and co-morbidity may also play a role in the variability and atypical expression with or without ID; these may explain the phenotypic differences between the brothers. In conclusion, we reclassify a patient with COPS as RTS with osteoma cutis and a mild intellectual disability, refuting COPS as a separate entity, since there were no reports in literature after 1991. To link mild intellectual disability to RTS, more studies are needed. Abbreviations BMDBone marrow density COPS syndromeCalcinosis cutis, osteoma cutis, poikiloderma and skeletal abnormalities DNADeoxyribonucleic acid FISHFluorescent in situ hybridisation IDIntellectual disability LSI MYCLocus-specific identifier (LSI) MYC, 2 FISH probes used to visualise locus 8q24 on chromosome 8 RECQL4DNA helicase, RECQ-like, type 4 RTSRothmund-Thomson syndrome Revisions received: 1 November 2016; 13 December 2016; 8 December 2016 Arnold P. Oranje has passed away shortly before submitting the final version of the paper. We are grateful to the patients and their family for their cooperation in this publication. Additionally, we thank Prof. Dr. M. Mearadji, Paediatric Radiologist, Erasmus Medical Centre Rotterdam, for proofreading and commenting on the radiologic findings, and Dr. Laura J.C.M. van Zutven, Department of Clinical genetics, Erasmus Medical Centre, Rotterdam, The Netherlands, for providing additional details on the chromosomal analysis performed at childhood.

Paediatric Radiologist, Erasmus Medical Centre Rotterdam, for proofreading and commenting on the radiologic findings, and Dr. Laura J.C.M. van Zutven, Department of Clinical genetics, Erasmus Medical Centre, Rotterdam, The Netherlands, for providing additional details on the chromosomal analysis performed at childhood. Authors’ contribution M.C. van Rij: involved in writing the case report, literature research. M.L. Grijsen: dermatologist in training who was involved in the patients’ dermatologic follow-up and description of the phenotype. N.M. Appelman-Dijkstra: internist involved in treatment of osteoporosis of the patient, intellectually contributing to the article, and proofreading the english language. K.B.M. Hansson: cytogeneticist involved in diagnosing the chromosomal abnormalities of both brothers and contributing to the description of these findings in the article. C.A.L. Ruivenkamp: molecular geneticist involved in the array analysis and proofreading the article. K. Mulder: dermatologist in training involved in the intellectual workup for the article. R. van Doorn: supervising dermatologist specialised in geno-dermatologic syndromes, and contributing to the intellectual content of the article. A.P. Oranje: dermatologist who wrote the initial case report (1991), now involved from a distance by proofreading the article. Sadly, this co-author passed away very recently (Oct 19, 2016). S.G. Kant: clinical geneticist who first diagnosed the patient with RTS syndrome, and who contributed to the article by critical reading.

A.P. Oranje: dermatologist who wrote the initial case report (1991), now involved from a distance by proofreading the article. Sadly, this co-author passed away very recently (Oct 19, 2016). S.G. Kant: clinical geneticist who first diagnosed the patient with RTS syndrome, and who contributed to the article by critical reading. Compliance with ethical standards Conflict of interest The authors declare that they have no conflict of interest. Funding There was no funding applied to complete this case report. Informed consent of the patient involved was provided, including permission to publish pictures.

Introduction Childhood bladder and/or bowel dysfunctions (CBBD) are common worldwide [1–5]. Bladder dysfunctions are daytime urinary incontinence (DUI), enuresis, and overactive bladder syndrome. The prevalence of DUI and enuresis decreases with age. The peak prevalence of DUI in girls is 8.4% at the age of 7 years, gradually decreasing to 4% in adolescence, whereas the corresponding prevalence rates for boys are 1.4 and 0.9%, respectively [6–8]. Enuresis is a complex condition, involving multiple pathogenic factors [9]. Prevalence’s vary, depending on the definition, approximately 10–20% of all 5-year-olds regularly wet their beds and the prevalence decreases by about 15% each year [10]. Generally accepted is that enuresis is more common in boys than in girls but only until the teenage years [11, 12]. The overactive bladder syndrome is found in 60 to 70% of children with urinary incontinence [13]. Bowel dysfunctions constitute of constipation and fecal incontinence (FI). Estimates of constipation in the general pediatric population range from 0.3 to 8%, with boys and girls equally affected. FI is one of the most common presentations of constipation and is found in up to 84% of constipated children at presentation [4]. FI is estimated to affect 0.8 to 7.8% children in Western societies with reported boys to girls’ ratio’s ranging from 3:1 to 6:1 [14–17]. CBBD have a major impact on a child’s psychosocial functioning. Comorbid behavioral disorders in about 20 to 40% of children with CBBD affect the everyday life of the children and their family [13, 14, 18–21]. The first treatment option of CBBD is a relaxed toileting posture and effective straining to defecate, which requires sufficient locomotor skills [4, 22]. The pelvic floor muscles (PFM) assist in maintaining urinary and fecal continence and, opposite to it, in adequate urination and defecation. Moreover, the PFM cooperate in close synergy with the diaphragm and the abdominal muscles. Therefore, the PFM are also involved in breathing and stabilizing connecting joints and the lower back [23–27].

s (PFM) assist in maintaining urinary and fecal continence and, opposite to it, in adequate urination and defecation. Moreover, the PFM cooperate in close synergy with the diaphragm and the abdominal muscles. Therefore, the PFM are also involved in breathing and stabilizing connecting joints and the lower back [23–27]. This means that an unstable or tensed posturing on the toilet, in which the PFM are unable to relax properly, can cause an inadequate urinary flow or bowel movement. From this point of view, CBBD might be related to impaired locomotor skills, although evidence for this hypothesis is lacking. In the Netherlands, standard medical care for CBBD is initially delivered by the general practitioner, but patients are also allowed to visit a private pelvic physiotherapist (PPT) without referral from a general practitioner or medical specialist. In case of unsatisfactory results, the child can be referred to a pediatrician at a general hospital (secondary healthcare). University-hospital care (tertiary healthcare) is required when secondary healthcare fails. Next, to support SMC, patients can be referred from medical doctors (primary, secondary, and tertiary healthcare) to primary healthcare PPT. Consequently, children visiting a PPT form a heterogeneous group of all ages and CBBD problems.

ty-hospital care (tertiary healthcare) is required when secondary healthcare fails. Next, to support SMC, patients can be referred from medical doctors (primary, secondary, and tertiary healthcare) to primary healthcare PPT. Consequently, children visiting a PPT form a heterogeneous group of all ages and CBBD problems. Limited data is available on the clinical characteristics and complexity of (concomitant) CBBD (physician and parent-reported) in primary and secondary healthcare, while most CBBD studies conducted in tertiary healthcare settings especially focus on treatment effects [6, 28, 29]. The lack of knowledge of patient characteristics, severity of symptoms and co-morbidities throughout healthcare settings may hamper targeting effective treatments. Furthermore, discrepancies are described between physicians’ diagnoses and parent-reported daily symptoms regarding CBBD [30, 31]. The aims of this pragmatic study are to describe (i) the clinical characteristics of CBBD in, and between, primary, secondary and tertiary healthcare settings, (ii) the level of agreement between referring physicians’ diagnoses and questionnaire-based symptoms, reported by parents and (iii) the relation between CBBD and locomotor problems. It was hypothesized that the prevalence’s of CBBD, comorbidities, and locomotor problems would increase, going from lower-to-higher-level healthcare settings.

reement between referring physicians’ diagnoses and questionnaire-based symptoms, reported by parents and (iii) the relation between CBBD and locomotor problems. It was hypothesized that the prevalence’s of CBBD, comorbidities, and locomotor problems would increase, going from lower-to-higher-level healthcare settings. Methods Study design and population We performed a cross-sectional study in a sample of children, aged 1–16 years, affected with varying forms of bladder and/or bowel dysfunctions, irrespective of the cause or presence of comorbidity and/or behavioral problems. Except age (1–16 years), no exclusions were made. Children from across the Netherlands and visiting primary healthcare PPT-practices were enrolled. They came on their own initiative (self-initiated visit; primary healthcare) or were referred by either the general practitioner (primary), district hospitals (secondary), or university hospitals (tertiary healthcare settings). Participating PPT’s are all expert pelvic physiotherapists who had completed a professional master’s degree in PPT. Physicians’ diagnoses were established based on patient history and additional assessments (e.g., physical examination, flowmetry, etc.), as documented in accepted pediatric Dutch guidelines [32–35]. Prior to the first visit at the PPT, the parents reported symptoms by completing the Childhood Bladder and Bowel Questionnaire (CBBDQ). Data were retrieved from the electronic patient records of the children.

ts (e.g., physical examination, flowmetry, etc.), as documented in accepted pediatric Dutch guidelines [32–35]. Prior to the first visit at the PPT, the parents reported symptoms by completing the Childhood Bladder and Bowel Questionnaire (CBBDQ). Data were retrieved from the electronic patient records of the children. Ethics statement Informed consent was obtained from all parents and children, aged 12 years and older included in the study. Web-based electronic patient records Prior to the first visit at the PPT-practice, parents completed the electronic patient records at home. These included the following components: Patient history Age, sex, physicians’ diagnoses (possible diagnoses listed with check boxes and an “other options”-text box), chronicity of the CBBD, medication use, comorbidities, and family history. The parent-reported Strength and Difficulties Questionnaire (SDQ), a brief validated screening questionnaire, for children age 4–17 was used to assess emotional and behavioral problems in child’s daily life [36–38].

an “other options”-text box), chronicity of the CBBD, medication use, comorbidities, and family history. The parent-reported Strength and Difficulties Questionnaire (SDQ), a brief validated screening questionnaire, for children age 4–17 was used to assess emotional and behavioral problems in child’s daily life [36–38]. Childhood Bladder and Bowel Dysfunction Questionnaire The CBBDQ is a recently developed evaluative symptom questionnaire based on International Children’s Continence Society recommendations and Rome III criteria for functional gastrointestinal disorders [1, 4, 39, 40], with excellent content and construct validity [40, 41]. The CBBDQ consists of two subscales: (1) the bladder symptoms scale (10 items) and (2) the bowel symptoms scale, including abdominal pain and bloated belly (8 items). The parents were asked to indicate the presence of the symptoms, using a five-point Likert scale [0 (never) to 4 (almost daily or daily)]. Locomotor problems A seven items questionnaire was developed by experts (PPT’s and pediatric physiotherapists) [42] and used as a measure to report problems in locomotor control and motor learning, motor skills (ability to learn to tie shoelaces, cycle or swim) and starting and performing a task, motor control (core stability), and musculoskeletal problems.

onnaire was developed by experts (PPT’s and pediatric physiotherapists) [42] and used as a measure to report problems in locomotor control and motor learning, motor skills (ability to learn to tie shoelaces, cycle or swim) and starting and performing a task, motor control (core stability), and musculoskeletal problems. Data analyses For the descriptive analyses, data are expressed as means and standard deviations for continuous variables or as frequencies and percentages for categorical variables. Comparisons between healthcare settings and gender are made, using analysis of variance for continuous variables and the χ 2 test for categorical variables. The SDQ items are coded as “not true”, “somewhat true”, and “certainly true”. The total difficulties score ranges from 0 to 40. Two categories are distinguished: “close to average to slightly raised” (0–16) and “high to very high” (17–40). To examine the symptom prevalence rates of the CBBDQ, the outcomes were dichotomized, with “never or once in the preceding month” recorded as non-symptomatic and “more than once to (almost) every day in the preceding month” as symptomatic. Possible missing items were imputed as “non-symptomatic”.

The SDQ items are coded as “not true”, “somewhat true”, and “certainly true”. The total difficulties score ranges from 0 to 40. Two categories are distinguished: “close to average to slightly raised” (0–16) and “high to very high” (17–40). To examine the symptom prevalence rates of the CBBDQ, the outcomes were dichotomized, with “never or once in the preceding month” recorded as non-symptomatic and “more than once to (almost) every day in the preceding month” as symptomatic. Possible missing items were imputed as “non-symptomatic”. To compare in individual children for their referring physicians’ diagnoses versus the parent-reported symptoms (as determined by the CBBDQ), the Cohen’s kappa coefficient is calculated with regard to the categories as follows: “≥1 no BBD”, “≥1 bladder dysfunction”, “≥1 bowel dysfunction”, and “≥1 concomitant CBBD”. A kappa coefficient of 0 to 0.4 is interpreted as poor agreement, 0.41 to 0.75 as “fair to good agreement” and above 0.75 as “excellent agreement” [43]. A P-value <0.05 was considered to indicate statistical significance. Statistical analyses were performed with SPSS software, version 23 (IBM Corporation, Somers, NY, USA).

To compare in individual children for their referring physicians’ diagnoses versus the parent-reported symptoms (as determined by the CBBDQ), the Cohen’s kappa coefficient is calculated with regard to the categories as follows: “≥1 no BBD”, “≥1 bladder dysfunction”, “≥1 bowel dysfunction”, and “≥1 concomitant CBBD”. A kappa coefficient of 0 to 0.4 is interpreted as poor agreement, 0.41 to 0.75 as “fair to good agreement” and above 0.75 as “excellent agreement” [43]. A P-value <0.05 was considered to indicate statistical significance. Statistical analyses were performed with SPSS software, version 23 (IBM Corporation, Somers, NY, USA). Results Baseline patient characteristics Participants Table 1 presents the baseline characteristics of the 1748 children (855 boys; mean age 7.6 years [SD 2.8], 893 girls; mean age 7.7 years [SD 2.9]) included from May 2010 to May 2015. No significant differences were found in age and gender between the healthcare settings. One thousand five hundred children (87%) were referred to PPT by a general practitioner or a medical specialist, like a pediatrician, urologist, nephrologist, or pediatric gastroenterologist, while 13% were self-initiated visits.Table 1 Baseline patient characteristics between healthcare settings Total 1-HC 2-HC 3-HC P valuea

Results Baseline patient characteristics Participants Table 1 presents the baseline characteristics of the 1748 children (855 boys; mean age 7.6 years [SD 2.8], 893 girls; mean age 7.7 years [SD 2.9]) included from May 2010 to May 2015. No significant differences were found in age and gender between the healthcare settings. One thousand five hundred children (87%) were referred to PPT by a general practitioner or a medical specialist, like a pediatrician, urologist, nephrologist, or pediatric gastroenterologist, while 13% were self-initiated visits.Table 1 Baseline patient characteristics between healthcare settings Total 1-HC 2-HC 3-HC P valuea n = 1748 n = 731 n = 906 n = 111 (41.7) (51.9) (6.4) Participants Age (years) Mean ± SD 7.7 ± 2.9 7.5 ± 2.9 7.8 ± 2.8 7.9 ± 3.2 0.27 Boys 855 (48.9) 385 (52.6) 414 (45.4) 56 (51.3) Age (years) Mean ± SD 7.6 ± 2.8 7.4 ± 2.8 7.9 ± 2.8 7.6 ± 2.9 0.25 Girls 893 (51.1) 345 (47.3) 493 (53.6) 55 (48.7) Age (years) Mean ± SD 7.7 ± 2.9 7.6 ± 3.0 7.8 ± 2.8 8.2 ± 3.5 0.27 Parentsb Single parent 113 (6.7) 46 (6.7) 59 (6.7) 8 (7.1). Two parents 1429 (85.1) 589 (85.7) 749 (84.8) 91 (83.5) Newly formed family 137 (8.2) 52 (7.6) 75 (8.5) 10 (7.3) Siblings 1481 (88.2) 603 (87.8) 779 (88.2) 99 (90.8) 0.39 Problems at homec 236 (13.5) 96 (13.2) 128 (14.1) 12 (5.1) 0.27 Childhood bladder and bowel dysfunctions (physicians’ diagnoses) Daytime urinary incontinence 602 (34.3) 274 (37.6) 296 (32.6) 32 (28.3) 0.039*

wly formed family 137 (8.2) 52 (7.6) 75 (8.5) 10 (7.3) Siblings 1481 (88.2) 603 (87.8) 779 (88.2) 99 (90.8) 0.39 Problems at homec 236 (13.5) 96 (13.2) 128 (14.1) 12 (5.1) 0.27 Childhood bladder and bowel dysfunctions (physicians’ diagnoses) Daytime urinary incontinence 602 (34.3) 274 (37.6) 296 (32.6) 32 (28.3) 0.039* Constipation 549 (31.4) 164 (22.5) 341 (37.6) 44 (38.9) <0.001* Enuresis 493 (28.2) 243 (33.3) 229 (25.2) 21 (18.9) <0.001* Fecal incontinence 362 (20.7) 159 (21.8) 178 (19.6) 25 (22.1) 0.51 Increased voiding frequency 267 (15.3) 114 (15.7) 141 (15.5) 12 (10.6) 0.37 Urinary tract infections 210 (12.0) 48 (6.6) 142 (15.7) 20 (17.7) <0.001* Abdominal pain 457 (26.1) 163 (22.4) 259 (28.6) 35 (31.0) 0.009* Withholding behavior At least one bladder dysfunction 1122 (64.2) 486 (66.8) 572 (63.1) 64 (55.6) 0.069 At least one bowel dysfunction 793 (45.4) 286 (36.1) 445 (49.1) 62 (54.9) <0.001* CBBD 321 (18.4) 116 (15.9) 183 (20.2) 22 (19.5) 0.083 Other health problemsd 126 (7.4) 26 (3.7) 80 (8.8) 20 (18.7) <0.001* Medicatione Bladder 183 (10.5) 77 (10.6) 89 (9.8) 17 (15.0) 0.23 Bowel 787 (45.4) 287 (39.7) 440 (48.9) 60 (53.1) <0.001* Parent-completed SDQ, age 4–16 years Analyzed = 1559 Range close to average to slightly raised 0–16 1386 (88.9) 547 (87.1) 749 (90.0) 90 (90.9) high to very high 17–40 173 (11.1) 81 (12.9) 83(10.0) 9 (9.1) Total difficulties score 0–40 9.2 ± 5.7 9.5 ± 5.8 9.0 ± 5.7 8.5 ± 5.5 0.15

287 (39.7) 440 (48.9) 60 (53.1) <0.001* Parent-completed SDQ, age 4–16 years Analyzed = 1559 Range close to average to slightly raised 0–16 1386 (88.9) 547 (87.1) 749 (90.0) 90 (90.9) high to very high 17–40 173 (11.1) 81 (12.9) 83(10.0) 9 (9.1) Total difficulties score 0–40 9.2 ± 5.7 9.5 ± 5.8 9.0 ± 5.7 8.5 ± 5.5 0.15 HC healthcare setting (1 primary; 2 secondary; 3 tertiary); SDQ Strength and Difficulties Questionnaire; CBBD Both, bladder and bowel dysfunction. Data are presented as number (n) and percentage (%) (or otherwise circumscribed, mean and ± standard deviation (SD)) aANOVA or Pearson’s chi-square bmissing 4.9% (P = 0.93) cDivorce of parents, death of a parent or grandparent, sibling or pet, moving house, stress at home dPulmopathy/asthma (2.2%); feeling sick/failure to thrive (1.7%); allergy/eczema (0.8%); neurological disorders (0.7%); endocrine or metabolic disorders (0.5%); musculoskeletal disorders (0.5%); gynecologic disorders (0.4%); complex disorders (0.3%); otopathy (0.2%); cardiopathy (0.2%); psychiatric disorders (0.2%); cystic fibrosis (0.1%); other gastrointestinal tract disorders (0.1%); not otherwise specified (0.4%) eBladder: overactive bladder, enuresis and current urinary tract infection; bowel: laxatives *Significant at the P < 0.05 level

dPulmopathy/asthma (2.2%); feeling sick/failure to thrive (1.7%); allergy/eczema (0.8%); neurological disorders (0.7%); endocrine or metabolic disorders (0.5%); musculoskeletal disorders (0.5%); gynecologic disorders (0.4%); complex disorders (0.3%); otopathy (0.2%); cardiopathy (0.2%); psychiatric disorders (0.2%); cystic fibrosis (0.1%); other gastrointestinal tract disorders (0.1%); not otherwise specified (0.4%) eBladder: overactive bladder, enuresis and current urinary tract infection; bowel: laxatives *Significant at the P < 0.05 level Referring physicians’ diagnoses The most common referred diagnoses were daytime urinary incontinence (DUI; 34.3%), constipation (31.4%), enuresis (28.2%), and fecal incontinence (FI; 20.7%) while 26.1% of the children were referred with abdominal pain and 12.0% with urinary tract infections. No differences between health care settings were found with respect to chronicity of any complaints. Laxative use increased going from primary to tertiary care, and these agents were prescribed to 45.4% of the children, whereas medication for bladder dysfunctions was prescribed to 10.5% of the children. A total of 1122 (64.2%) of the children were referred with at least one bladder dysfunction, 793 (45.4%) with at least one bowel dysfunction, and 321 children (18.4%) with both, bladder, and bowel dysfunctions.

o 45.4% of the children, whereas medication for bladder dysfunctions was prescribed to 10.5% of the children. A total of 1122 (64.2%) of the children were referred with at least one bladder dysfunction, 793 (45.4%) with at least one bowel dysfunction, and 321 children (18.4%) with both, bladder, and bowel dysfunctions. Significantly, more DUI (P = 0.039) and enuresis (P < 0.001) were diagnosed in primary healthcare. Urinary tract infection, especially among girls, increased going from primary to tertiary healthcare settings (P = <0.001). Boys had less frequent bladder dysfunctions (P = 0.022) in tertiary care and less bowel dysfunctions in primary healthcare (P < 0.001). Whereas constipation (P < 0.001) and abdominal pain (P = 0.009) were more diagnosed in secondary and tertiary healthcare and more prevalent among girls (P < 0.001) (Table 2). FI was more common among boys in all settings (P < 0.001). The number of children referred with “at least one bowel dysfunction” increased, going from primary to tertiary healthcare settings (P < 0.001). Other health problems, such as lung diseases, also increased significantly from 3.7 to 18.7% (P < 0.001).Table 2 Gender-specific childhood bladder and bowel dysfunctions between healthcare setting (physicians’ diagnosis) Gender Total 1-HC 2-HC 3-HC P valuea n = 1748 n = 731 n = 906 n = 111 (41.8) (51.9) (6.4) Childhood Bladder and Bowel Dysfunctions (physicians’ diagnoses) Daytime urinary incontinence B versus Gb 0.65 boys 229 (35.0) 138 (36.0) 147 (35.5) 14 (24.1) 0.20 girls 303 (33.9) 136 (39.4) 149 (30.2) 18 (32.7) 0.022*

Significantly, more DUI (P = 0.039) and enuresis (P < 0.001) were diagnosed in primary healthcare. Urinary tract infection, especially among girls, increased going from primary to tertiary healthcare settings (P = <0.001). Boys had less frequent bladder dysfunctions (P = 0.022) in tertiary care and less bowel dysfunctions in primary healthcare (P < 0.001). Whereas constipation (P < 0.001) and abdominal pain (P = 0.009) were more diagnosed in secondary and tertiary healthcare and more prevalent among girls (P < 0.001) (Table 2). FI was more common among boys in all settings (P < 0.001). The number of children referred with “at least one bowel dysfunction” increased, going from primary to tertiary healthcare settings (P < 0.001). Other health problems, such as lung diseases, also increased significantly from 3.7 to 18.7% (P < 0.001).Table 2 Gender-specific childhood bladder and bowel dysfunctions between healthcare setting (physicians’ diagnosis) Gender Total 1-HC 2-HC 3-HC P valuea n = 1748 n = 731 n = 906 n = 111 (41.8) (51.9) (6.4) Childhood Bladder and Bowel Dysfunctions (physicians’ diagnoses) Daytime urinary incontinence B versus Gb 0.65 boys 229 (35.0) 138 (36.0) 147 (35.5) 14 (24.1) 0.20 girls 303 (33.9) 136 (39.4) 149 (30.2) 18 (32.7) 0.022* Constipation B versus Gb <0.001* boys 225 (26.3) 56 (14.5) 149 (36.0) 20 (35.7) <0.001* girls 324 (36.6) 108 (31.3) 192 (38.9) 24 (43.6) 0.039* Enuresis B versus Gb <0.001* boys 320 (37.4) 160 (41.6) 147 (35.5) 13 (23.2) 0.016* girls 173 (19.4) 83 (24.1) 82 (16.6) 8 (14.5) 0.018* Fecal incontinence B versus Gb <0.001*

n = 1748 n = 731 n = 906 n = 111 (41.8) (51.9) (6.4) Childhood Bladder and Bowel Dysfunctions (physicians’ diagnoses) Daytime urinary incontinence B versus Gb 0.65 boys 229 (35.0) 138 (36.0) 147 (35.5) 14 (24.1) 0.20 girls 303 (33.9) 136 (39.4) 149 (30.2) 18 (32.7) 0.022* Constipation B versus Gb <0.001* boys 225 (26.3) 56 (14.5) 149 (36.0) 20 (35.7) <0.001* girls 324 (36.6) 108 (31.3) 192 (38.9) 24 (43.6) 0.039* Enuresis B versus Gb <0.001* boys 320 (37.4) 160 (41.6) 147 (35.5) 13 (23.2) 0.016* girls 173 (19.4) 83 (24.1) 82 (16.6) 8 (14.5) 0.018* Fecal incontinence B versus Gb <0.001* boys 232 (27.1) 104 (27.0) 113 (27.3) 15 (26.8) 0.048* girls 130 (14.6) 56 (16.2) 65 (13.2) 9 (16.4) 0.22 Increased voiding frequency B versus Gb 0.51 boys 136 (15.9) 56 (14.6) 74 (17.9) 6 (10.3) 0.23 girls 131 (14.7) 58 (16.8) 67 (13.6) 6 (10.9) 0.32 Urinary tract infections B versus Gb <0.001* boys 18 (2.1) 6 (1.6) 11 (2.7) 1 (1.7) 0.63 girls 192 (21.5) 42(12.2) 131(26.6) 19 (34.5) <0.001* Abdominal pain B versus Gb <0.001* boys 167 (19.5) 61 (15.9) 93 (22.5) 13 (22.4) 0.056 girls 290 (32.5) 102 (29.6) 166 (33.7) 22 (40.0) 0.22 Withholding behavior B versus Gb 0.71 boys 234 (27.4) 121 (31.4) 102 (24.6) 11 (19.6) 0.040* girls 252 (28.2) 124 (35.9) 110 (22.3) 18 (32.7) <0.001* At least one bladder dysfunction B versus Gb 0.32 boys 554 (64.8) 265 (68.8) 260 (62.8) 29 (51.8) 0.022* girls 568 (63.6) 222 (64.3) 312 (63.3) 34 (61.8) 0.91 At least one bowel dysfunction B versus Gb 1.0 boys 388 (45.4) 142 (36.9) 214 (51.7) 32 (56.9) <0.001*

boys 167 (19.5) 61 (15.9) 93 (22.5) 13 (22.4) 0.056 girls 290 (32.5) 102 (29.6) 166 (33.7) 22 (40.0) 0.22 Withholding behavior B versus Gb 0.71 boys 234 (27.4) 121 (31.4) 102 (24.6) 11 (19.6) 0.040* girls 252 (28.2) 124 (35.9) 110 (22.3) 18 (32.7) <0.001* At least one bladder dysfunction B versus Gb 0.32 boys 554 (64.8) 265 (68.8) 260 (62.8) 29 (51.8) 0.022* girls 568 (63.6) 222 (64.3) 312 (63.3) 34 (61.8) 0.91 At least one bowel dysfunction B versus Gb 1.0 boys 388 (45.4) 142 (36.9) 214 (51.7) 32 (56.9) <0.001* girls 405 (45.4) 145 (42.0) 231 (46.9) 29 (52.7) 0.20 Both, bladder and bowel dysfunction B versus Gb 0.76 boys 160 (18.7) 63 (16.4) 88 (21.3) 9 (16.1) 0.18 girls 161 (18.0) 54 (15.7) 95 (19.3) 12 (21.8) 0.30 HC healthcare setting (1 primary; 2 secondary; 3 tertiary). Data are presented as number (n) and percentage (%) aPearson’s chi-square or Fisher’s exact test bB versus G; boys versus girls between healthcare settings *Significant at the P < 0.05 level Strength and Difficulties Questionnaire As depicted in Table 1, no significant differences in SDQ were found between healthcare settings. On a range of 0–40, the mean total SDQ difficulties scores were 9.5 (SD 5.8), 9.0 (SD 5.7), and 8.5 (SD 5.5) for primary, secondary, and tertiary healthcare, respectively, and 11.1% of the children had a total SDQ-difficulties score over 16, indicating the children had emotional or behavioral problems.

thcare settings. On a range of 0–40, the mean total SDQ difficulties scores were 9.5 (SD 5.8), 9.0 (SD 5.7), and 8.5 (SD 5.5) for primary, secondary, and tertiary healthcare, respectively, and 11.1% of the children had a total SDQ-difficulties score over 16, indicating the children had emotional or behavioral problems. Parent-reported symptoms Parent-reported symptoms are described in Table 3 (proportions between healthcare settings) and Table 4 (gender-specific). Based on the parent-reported symptoms, bladder problems like DUI occurred more often in primary healthcare than in secondary and tertiary healthcare (P = 0.023). High prevalence of ignoring the urge to urinate (53.0%) and urgency (56.6%) were found in all healthcare settings, whereby “ignoring” decreased significantly (P = 0.023) from primary to tertiary healthcare setting. Boys have more DUI than girls (P = 0.006), especially in secondary and tertiary healthcare settings and post-micturition dribble (P = 0.003), decreasing from primary to tertiary healthcare settings. Boys were more likely to suffer from enuresis than girls (P < 0.001). In contrast, girls wake up at night to urinate more often (P = 0.045).Table 3 Childhood bladder and bowel dysfunctions (parent-reported) between HC-settings Total 1-HC 2-HC 3-HC P valuea n = 1748 n = 730 n = 907 n = 111 Items of the Childhood Bladder and Bowel Dysfunction Questionnaireb

Parent-reported symptoms Parent-reported symptoms are described in Table 3 (proportions between healthcare settings) and Table 4 (gender-specific). Based on the parent-reported symptoms, bladder problems like DUI occurred more often in primary healthcare than in secondary and tertiary healthcare (P = 0.023). High prevalence of ignoring the urge to urinate (53.0%) and urgency (56.6%) were found in all healthcare settings, whereby “ignoring” decreased significantly (P = 0.023) from primary to tertiary healthcare setting. Boys have more DUI than girls (P = 0.006), especially in secondary and tertiary healthcare settings and post-micturition dribble (P = 0.003), decreasing from primary to tertiary healthcare settings. Boys were more likely to suffer from enuresis than girls (P < 0.001). In contrast, girls wake up at night to urinate more often (P = 0.045).Table 3 Childhood bladder and bowel dysfunctions (parent-reported) between HC-settings Total 1-HC 2-HC 3-HC P valuea n = 1748 n = 730 n = 907 n = 111 Items of the Childhood Bladder and Bowel Dysfunction Questionnaireb 1 Passes urine more than 8 times during the day 676 (38.7) 289 (39.6) 355 (39.1) 32 (28.8) 0.087 2 Wets underwear and /or outer clothing during the day 859 (49.1) 383 (52.5) 430 (47.4) 46 (41.4) 0.031* 3 Loses some drops of urine immediately after urinating has finished 684 (39.1) 302 (41.4) 347 (38.3) 35 (31.5) 0.40 4 Loses urine within the hour after urinating has finished 518 (29.6) 225 (30.8) 267 (29.4) 26 (23.4) 0.28 5 Seems to ignore the urge to urinate 927 (53.0) 407 (55.8) 473 (52.1) 47 (42.3) 0.023* 6 Uses tricks to stay dry, like wriggling or forcefully crossing the legs 597 (34.2) 261 (35.8) 303 (33.4) 33 (29.7) 0.36 7 Experiences a sudden uncontrollable urge to urinate 989 (56.6) 422 (57.8) 510 (56.2) 57 (51.4) 0.42 8 Postpones first urination in the morning 505 (28.9) 203 (27.8) 268 (29.5) 34 (30.6) 0.68 9 Wets the bed or diaper during sleeping periods 742 (42.4) 339 (46.4) 362 (39.9) 41 (36.9) 0.014* 10 Wakes up at night to urinate 353 (20.2) 128 (17.5) 203 (22.4) 22 (19.8) 0.052 11 Has two or fewer bowel movements per week 373 (21.4) 169 (23.2) 176 (19.4) 28 (25.2) 0.11 12 Stains or soils the underwear with stools 778 (44.5) 335 (45.9) 385 (42.4) 58 (52.3) 0.09* 13 Has hard stools or painful bowel movements 414 (23.7) 150 (20.5) 232 (25.6) 32 (28.8) 0.025* 14 Has large amounts of stool (that may obstruct the toilet) 345 (19.7) 147 (20.1) 168 (18.5) 30 (27.0) 0.10 15 Postpones bowel movements 616 (35.2) 292 (40.0) 286 (31.5) 38 (34.2) 0.002* 16 Experiences a sudden uncontrollable urge to defecate 763 (43.7) 326 (44.7) 389 (42.9) 48 (43.6) 0.77 17 Has abdominal pain 699 (40.1) 270 (37.1) 383 (42.3) 46 (41.4) 0.10 18 Has a bloated belly 415 (23.7) 157 (21.5) 214 (23.6) 44 (39.6) <0.001* At least one bladder symptom 1566 (89.6) 668 (91.5) 805 (88.8) 93 (83.8) 0.023* At least one bowel symptom 1404 (80.3) 589 (80.7) 722 (79.6) 93 (83.8) 0.55 Combined bladder and bowel symptom 1266 (72.4) 539 (73.8) 647 (71.3) 80 (72.1) 0.53

) 0.10 18 Has a bloated belly 415 (23.7) 157 (21.5) 214 (23.6) 44 (39.6) <0.001* At least one bladder symptom 1566 (89.6) 668 (91.5) 805 (88.8) 93 (83.8) 0.023* At least one bowel symptom 1404 (80.3) 589 (80.7) 722 (79.6) 93 (83.8) 0.55 Combined bladder and bowel symptom 1266 (72.4) 539 (73.8) 647 (71.3) 80 (72.1) 0.53 HC healthcare setting (1 primary, 2 secondary, 3 tertiary). Data are presented as number (n) and percentage (%) aPearson’s chi-square bLikert scale for symptoms on all items are (never–(nearly) every day) dichotomized: “Never or once in the past month” classified as “non-symptomatic” (no); “more than once in the past month to (nearly) every day in the past month” classified as “symptomatic” (yes). Missing items were imputed as “non-symptomatic” *Significant at the P < 0 05 level Table 4 Childhood Bladder and Bowel Dysfunction Questionnaire (parent-reported symptoms) gender-specific Gender Total 1-HC 2-HC 3-HC P valuea n = 1748 n = 730 n = 907 n = 111 Items of the Childhood Bladder and Bowel Dysfunction Questionnaireb 1 Passes urine more than 8 times during the day B versus Gc 0.057 boys 350 (40.9) 163 (42.3) 170 (41.1) 17 (30.4) 0.23 girls 326 (36.5) 126 (36.5) 185 (37.5) 15 (27.3) 0.33 2 Wets underwear and /or outer clothing during the day B versus Gc 0.006* boys 449 (52.2) 213 (47.4) 212 (48.8) 32 (57.1) 0.17 girls 410 (45.9) 170 (49.3) 218 (44.2) 22 (40.0) 0.23 3 Loses some drops of urine immediately after urinating finished B versus Gc 0.003*

1 Passes urine more than 8 times during the day B versus Gc 0.057 boys 350 (40.9) 163 (42.3) 170 (41.1) 17 (30.4) 0.23 girls 326 (36.5) 126 (36.5) 185 (37.5) 15 (27.3) 0.33 2 Wets underwear and /or outer clothing during the day B versus Gc 0.006* boys 449 (52.2) 213 (47.4) 212 (48.8) 32 (57.1) 0.17 girls 410 (45.9) 170 (49.3) 218 (44.2) 22 (40.0) 0.23 3 Loses some drops of urine immediately after urinating finished B versus Gc 0.003* boys 490 (57.3) 173 (44.9) 173 (41.8) 19 (33.9) 0.26 girls 319 (35.7) 129 (37.4) 174 (35.3) 16 (29.1) 0.47 4 Loses urine within the hour after urinating has finished B versus Gc 0.08 boys 270 (31.6) 122 (31.7) 131 (31.6) 17 (30.4) 0.98 girls 248 (27.8) 103 (29.9) 136 (27.6) 9 (16.4) 0.12 5 Seems to ignore the urge to urinate B versus Gc 0.006* boys 482 (56.4) 227 (59.0) 234 (56.5) 21 (37.5) 0.010* girls 445 (49.8) 180 (52.2) 239 (48.5) 26 (47.3) 0.53 6 Uses tricks to stay dry, like wriggling or forcefully crossing legs B versus Gc 0.76 boys 295 (34.5) 147 (38.2) 136 (32.9) 12 (21.4) 0.030* girls 302 (33.8) 114 (33.0) 167 (33.9) 21 (38.2) 0.76 7 Experiences a sudden uncontrollable urge to urinate B versus Gc 0.04* boys 505 (59.1) 232 (60.3) 244 (58.9) 27 (48.2) 0.48 girls 484 (54.2) 190 (55.1) 266 (54.0) 28 (50.9) 0.84 8 Postpones first urination in the morning B versus Gc 0.09 boys 231 (27.0) 105 (27.3) 112 (27.1) 14 (25.0) 0.94 girls 274 (30.7) 98 (28.4) 156 (31.6) 20 (36.4) 0.39 9 Wets the bed or diaper during sleeping periods B versus Gc <0.001*

60.3) 244 (58.9) 27 (48.2) 0.48 girls 484 (54.2) 190 (55.1) 266 (54.0) 28 (50.9) 0.84 8 Postpones first urination in the morning B versus Gc 0.09 boys 231 (27.0) 105 (27.3) 112 (27.1) 14 (25.0) 0.94 girls 274 (30.7) 98 (28.4) 156 (31.6) 20 (36.4) 0.39 9 Wets the bed or diaper during sleeping periods B versus Gc <0.001* boys 445 (52.0) 210 (54.5) 208 (46.7) 29 (48.2) 0.40 girls 297 (33.3) 129 (37.4) 154 (31.2) 14 (25.5) 0.079 10 Wakes up at night to urinate B versus Gc 0.01* boys 151 (17.7) 56 (14.5) 87 (21.0) 8 (14.3) 0.045* girls 202 (22.6) 72 (20.9) 116 (23.5) 14 (25.5) 0.58 11 Has two or fewer bowel movements per week B versus Gc 0.44 boys 176 (20.6) 77 (20.0) 85 (20.5) 14 (25.0) 0.014* girls 197 (22.1) 92 (26.7) 91 (18.5) 14 (25.5) 0.69 12 Stains or soils the underwear with stools B versus Gc 0.001* boys 440 (51.5) 183 (47.5) 202 (48.8) 30 (53.6) 0.69 girls 363 (40.6) 152 (44.1) 183 (37.1) 28 (50.9) 0.037* 13 Has hard stools or painful bowel movements B versus Gc <0.001* boys 164 (19.2) 58 (15.1) 91 (22.0) 15 (26.8) 0.015* girls 250 (28.0) 92 (26.7) 141 (28.6) 17 (30.9) 0.73 14 Has large amounts of stool (that may obstruct the toilet) B versus Gc 0.57 boys 164 (19.2) 65 (16.9) 84 (20.3) 15 (26.8) 0.16 girls 181 (20.3) 82 (23.8) 84 (17.0) 15 (23.7) 0.024* 15 Postpones bowel movements B versus Gc <0.001*

8 (15.1) 91 (22.0) 15 (26.8) 0.015* girls 250 (28.0) 92 (26.7) 141 (28.6) 17 (30.9) 0.73 14 Has large amounts of stool (that may obstruct the toilet) B versus Gc 0.57 boys 164 (19.2) 65 (16.9) 84 (20.3) 15 (26.8) 0.16 girls 181 (20.3) 82 (23.8) 84 (17.0) 15 (23.7) 0.024* 15 Postpones bowel movements B versus Gc <0.001* boys 337 (39.4) 172 (44.7) 150 (36.2) 15 (26.8) 0.021* girls 279 (31.2) 120 (34.8) 136 (27.6) 23 (41.8) 0.002* 16 Experiences a sudden uncontrollable urge to defecate B versus Gc 0.06 boys 402 (47.0) 187 (48.6) 192 (46.4) 23 (41.1) 0.54 girls 361 (40.5) 139 (40.3) 197 (40.0) 25 (46.3) 0.66 17 Has abdominal pain B versus Gc <0.001* boys 299 (35.0) 128 (33.3) 150(36.2) 21 (37.5) 0.22 girls 400 (44.9) 142 (41.3) 233 (47.5) 25 (45.5) 0.64 18 Has a bloated belly B versus Gc <0.001* boys 162 (18.9) 67 (17.1) 77 (18.6) 18 (32.1) 0.030* girls 253 (28.3) 90 (26.1) 137 (27.8) 26 (47.3) 0.005* At least one bladder symptom B versus Gc 0.27 boys 773 (90.4) 355 (92.2) 371 (89.6) 47 (83.9) 0.11 girls 793 (88.8) 313 (90.7) 493 (88.0) 46 (83.6) 0.22 At least one bowel symptom B versus Gc 0.93 boys 686 (80.2) 312 (81.0) 331 (80.0) 43 (76.8) 0.74 girls 718 (80.4) 227 (80.3) 391 (79.3) 50 (90.9) 0.12 Combined bladder and bowel symptom B versus Gc 0.61 boys 624 (73.0) 286 (74.3) 300 (72.5) 38 (67.9) 0.57 girls 642 (71.9) 253 (73.3) 347 (70.4) 42 (76.4) 0.48 HC healthcare setting (1 primary, 2 secondary, 3 tertiary). Data are presented as number (n) and percentage (%) aPearson’s chi-square

boys 162 (18.9) 67 (17.1) 77 (18.6) 18 (32.1) 0.030* girls 253 (28.3) 90 (26.1) 137 (27.8) 26 (47.3) 0.005* At least one bladder symptom B versus Gc 0.27 boys 773 (90.4) 355 (92.2) 371 (89.6) 47 (83.9) 0.11 girls 793 (88.8) 313 (90.7) 493 (88.0) 46 (83.6) 0.22 At least one bowel symptom B versus Gc 0.93 boys 686 (80.2) 312 (81.0) 331 (80.0) 43 (76.8) 0.74 girls 718 (80.4) 227 (80.3) 391 (79.3) 50 (90.9) 0.12 Combined bladder and bowel symptom B versus Gc 0.61 boys 624 (73.0) 286 (74.3) 300 (72.5) 38 (67.9) 0.57 girls 642 (71.9) 253 (73.3) 347 (70.4) 42 (76.4) 0.48 HC healthcare setting (1 primary, 2 secondary, 3 tertiary). Data are presented as number (n) and percentage (%) aPearson’s chi-square bLikert scale for symptoms (never–(nearly) every day) dichotomized: “Never or once in the past month” classified as “non-symptomatic” (no); “more than once in the past month to (nearly) every day in the past month” classified as “symptomatic” (yes). Missing items were imputed as “non-symptomatic” cB versus G; boys versus girls between healthcare settings *Significant at the P < 0.05 level

bLikert scale for symptoms (never–(nearly) every day) dichotomized: “Never or once in the past month” classified as “non-symptomatic” (no); “more than once in the past month to (nearly) every day in the past month” classified as “symptomatic” (yes). Missing items were imputed as “non-symptomatic” cB versus G; boys versus girls between healthcare settings *Significant at the P < 0.05 level Locomotor problems Locomotor problems prevailed in all healthcare settings and the prevalence increased with higher-level healthcare settings (Table 5). Parents indicated that 14.5% of all children have had problems in motor learning, 9.2% had problems in core stability, and 10.8% had an increased muscle tension. Children referred from tertiary healthcare settings experienced more problems than those from primary and secondary settings. Significant differences between healthcare settings were found in motor skills (P = 0.041) and core stability problems (P = 0.015). No differences were found with respect to having problems starting or performing a task or musculoskeletal problems.Table 5 Locomotor problems (parent-reported) Total 1-HC 2-HC 3-HC P valuea n = 1674 n = 684 n = 881 n = 109 Motor learning Reduced manual dexterityb 175 (10.5) 73 (10.7) 88 (10.0) 14 (12.8) 0.64 Problems of motor skillsc 286 (17.1) 99 (14.6) 164 (19.0) 23 (21.3) 0.041* Problems starting and performing a taskc 266 (15.9) 101 (14.8) 145 (16.5) 20 (18.5) 0.50 Motor control Problems in core stabilityc 154 (9.2) 51 (7.5) 86 (9.8) 17 (15.7) 0.015*

n = 1674 n = 684 n = 881 n = 109 Motor learning Reduced manual dexterityb 175 (10.5) 73 (10.7) 88 (10.0) 14 (12.8) 0.64 Problems of motor skillsc 286 (17.1) 99 (14.6) 164 (19.0) 23 (21.3) 0.041* Problems starting and performing a taskc 266 (15.9) 101 (14.8) 145 (16.5) 20 (18.5) 0.50 Motor control Problems in core stabilityc 154 (9.2) 51 (7.5) 86 (9.8) 17 (15.7) 0.015* Musculoskeletal Increased muscle tension 180 (10.8) 76 (12.1) 94 (11.3) 10 (10.1) 0.83 Reduced muscle tension 28 (1.7) 13 (1.8) 13 (1.4) 2 (1.8) 0.91 HC healthcare setting (1 primary, 2 secondary, 3 tertiary). Data are presented as number (n) and percentage (%) aPearson’s chi-square bCoding of dummy variables: Manual dexterity: normal/increased = 0; reduced = 1 cCoding of dummy variables: Locomotor regarding stability and skills (such as tying shoelaces, swimming), starting or performing a task never/sometimes = 0; often = 1 *Significant at the P < 0.05 level

HC healthcare setting (1 primary, 2 secondary, 3 tertiary). Data are presented as number (n) and percentage (%) aPearson’s chi-square bCoding of dummy variables: Manual dexterity: normal/increased = 0; reduced = 1 cCoding of dummy variables: Locomotor regarding stability and skills (such as tying shoelaces, swimming), starting or performing a task never/sometimes = 0; often = 1 *Significant at the P < 0.05 level Discussion To our knowledge, this cross-sectional study is the first to describe clinical characteristics of (concomitant) CBBD in different healthcare settings, in a large sample of 1748 affected children, visiting primary PPT-practices. In this pragmatic study, all referred children, irrespective of age, complexity of complaints, or comorbidities were included, yielding a heterogeneous cohort reflecting routinely PPT-practice. Although we have hypothesized that the prevalence’s of CBBD, comorbidities, and locomotor problems would increase going from lower-to-higher-level healthcare settings, the results of our study could only be confirmed for the physicians’ diagnoses constipation and abdominal pain and the parent-reported symptoms hard stools, abdominal pain, bloating, problems in core stability, and the existence of other health problems. In contrast, DUI, ignoring the urge to urinate and enuresis decreased going from lower-to-higher-level healthcare settings. Poor agreement exists between referred physicians’ diagnoses and questionnaire-based parent-reported symptoms. Locomotor problems prevailed in all healthcare settings. Children referred from tertiary healthcare settings experienced more problems in motor skills and core stability than those from primary and secondary settings.

eement exists between referred physicians’ diagnoses and questionnaire-based parent-reported symptoms. Locomotor problems prevailed in all healthcare settings. Children referred from tertiary healthcare settings experienced more problems in motor skills and core stability than those from primary and secondary settings. Some striking gender differences appeared when comparing our results with earlier studies. Significantly, more girls were suffering from constipation, abdominal pain, and urinary tract infection (physician’ diagnosis), whereas boys experienced more DUI, hard stools or painful bowel movements (parent-reported symptoms) and fecal incontinence and enuresis (physician’s diagnosis and parent-reported symptoms). Overall, estimates of presented prevalence figures differ greatly and depend not only on the clinical setting, but also on the heterogeneity of the criteria used for defining or diagnosing bladder or bowel dysfunctions. Standardized use of the accepted International Children’s Continence Society recommendations and/or the Rome III criteria for functional gastrointestinal disorders would facilitate study comparability.

setting, but also on the heterogeneity of the criteria used for defining or diagnosing bladder or bowel dysfunctions. Standardized use of the accepted International Children’s Continence Society recommendations and/or the Rome III criteria for functional gastrointestinal disorders would facilitate study comparability. Concomitant CBBD was equally distributed over all three healthcare settings and diagnosed by physicians in 18.4% of the children compared to 72.4% when considering parent-reported symptoms. In fact, all parent-reported symptoms occurred more frequently than indicated by the physicians’ diagnoses, especially when considering combined bladder and bowel symptoms. This discrepancy between physicians’ diagnoses and CBBDQ outcomes may due to both, physicians and parents. When physicians focus on questioning for bladder symptoms or bowel symptoms, then concomitant CBBD will be missed. Next, is the lack of parent’s knowledge of linking complaints of CBBD. Besides, filling in questionnaires raise the attention to certain symptoms. Therefore, when a physician does not explicitly ask for all CBBD symptoms, the parents or children most likely will not report them. Therefore, caregivers should be made aware of this discrepancy, to prevent the risk of inadequately diagnosing CBBD, to promote favorable therapy outcomes, and to reduce the risk of relapses. Using a CBBD questionnaire might facilitate elicitation of all relevant symptoms.

or children most likely will not report them. Therefore, caregivers should be made aware of this discrepancy, to prevent the risk of inadequately diagnosing CBBD, to promote favorable therapy outcomes, and to reduce the risk of relapses. Using a CBBD questionnaire might facilitate elicitation of all relevant symptoms. Locomotor problems prevailed in all healthcare settings. Epidemiological studies have shown that about 6% of all school-age children are described by experts and parents as uncoordinated in their fine and gross motor skills [44]. In our study, parents indicated that 14.5% of all children have had problems in motor learning and 9.2% had problems in core stability. This finding could not be explained by concomitant physical problems or comorbidities, as the number of these children was low and did not differ significantly between healthcare settings. Dysfunction of the pelvic floor muscles (PFM) and cooperating abdominal muscles is an integral component of the pathophysiology of CBBD [45–47]. Only Chase et al. [48] have examined whether different trunk musculoskeletal characteristics might be related to defecation difficulties. In agreement with our results, they found rather high prevalence rates for core stability and motor skills problems in children, supporting the hypothesis, that locomotor problems exists, indicating that dysfunctions of the muscles may be associated with CBBD [23–25, 27, 49–51]. Hence, pelvic physiotherapists, as musculoskeletal specialists, might play a role in treating children with CBBD [45, 48, 52–55].

and motor skills problems in children, supporting the hypothesis, that locomotor problems exists, indicating that dysfunctions of the muscles may be associated with CBBD [23–25, 27, 49–51]. Hence, pelvic physiotherapists, as musculoskeletal specialists, might play a role in treating children with CBBD [45, 48, 52–55]. Studies have reported that children with CBBD are at increased risk of psychosocial, behavioral, or psychological disorders [13, 56]. Although emotional or behavioral problems were present in 11.1% of the children, no association between behavioral problems and CBBD could be confirmed in all three healthcare settings. The SDQ scores did not deviate far from the norm scores reported in the literature, indicating that our sample appeared to be representative of the normal Dutch population of children aged 1–16 years.

the children, no association between behavioral problems and CBBD could be confirmed in all three healthcare settings. The SDQ scores did not deviate far from the norm scores reported in the literature, indicating that our sample appeared to be representative of the normal Dutch population of children aged 1–16 years. Some limitations might affect the interpretation of our findings. First, diagnoses used by physicians were based on heterogeneity of criteria. Secondly, although CBBDQ has been evaluated for validity aspects, further research is required to define its psychometric properties and to justify its use in research and clinical practice. Moreover, symptoms, indicated by the CBBDQ, were not verified by means of diagnostic testing. Then, generalization of our findings may be hampered by the fact that healthcare systems and therewith referral patterns may differ per country. Next, it is unclear if this sample of children is a typical subset of the broader population, and whether the medical doctors have referred all children with functional BBD, or only the children who failed SMC. Finally, data on the locomotor problems were obtained through parental reports and were not confirmed by a questionnaire of adequate psychometric evaluation or by musculoskeletal examination. Further, well-designed studies are necessary to assess whether children with CBBD have more locomotor problems compared to their healthy peers.

n the locomotor problems were obtained through parental reports and were not confirmed by a questionnaire of adequate psychometric evaluation or by musculoskeletal examination. Further, well-designed studies are necessary to assess whether children with CBBD have more locomotor problems compared to their healthy peers. Despite the aforementioned limitations, we feel that our study has strong points, such as a large sample that approximates the average patient in all healthcare settings with no restrictions regarding CBBD definition and comorbidities, and our study is one of first taking in account the motor control problems in relation to CBBD.

aforementioned limitations, we feel that our study has strong points, such as a large sample that approximates the average patient in all healthcare settings with no restrictions regarding CBBD definition and comorbidities, and our study is one of first taking in account the motor control problems in relation to CBBD. Conclusion The present pragmatic study is one of the first to report the clinical characteristics of children with various forms of CBBD referred to PPT from primary, secondary, and tertiary healthcare settings. The results indicate that our hypothesis could only be confirmed for the physicians’ diagnoses constipation, abdominal pain, the existence of other health problems, and the parent-reported symptoms hard stools, bloating, and problems in core stability. More girls were suffering from constipation, abdominal pain, and urinary tract infections (physicians’ diagnoses), boys from DUI and experiencing hard stools or painful bowel movements (parent-reported symptoms) and fecal incontinence and enuresis (physicians’ diagnoses and parent-reported symptoms). The major discrepancy between physicians’ diagnoses and the symptoms, reported by the parents, raises the question whether parents are aware that their child has concomitant bladder and bowel dysfunctions when visiting a physician. Using a combined CBBD questionnaire might reduce the risk of inadequate diagnosing CBBD. Finally, locomotor problems prevailed in all healthcare settings. Since both the PFM (contributing to urination, defecation, continence, intra-abdominal pressure generation, antigravity support, and lumbo-pelvic stability) and locomotor problems might be a part of CBBD, pelvic physiotherapists can be considered to be involved in the healthcare of children affected with CBBD.

ettings. Since both the PFM (contributing to urination, defecation, continence, intra-abdominal pressure generation, antigravity support, and lumbo-pelvic stability) and locomotor problems might be a part of CBBD, pelvic physiotherapists can be considered to be involved in the healthcare of children affected with CBBD. Abbreviations CBBDchildhood bladder and bowel dysfunctions CBBDQChildhood Bladder and Bowel Dysfunctions Questionnaire DUIDaytime urinary incontinence FIFecal incontinence ICCSInternational Children’s Continence Society SDQStrengths and Difficulties Questionnaire SMCStandard medical care PFMPelvic floor muscles PPTPelvic physiotherapy Authors’ contributions Dr. M. L. van Engelenburg–van Lonkhuyzen conceptualized and designed the study, designed the data collection instrument, coordinated and supervised data collection, acquired data, analyzed and interpreted the data and was the main person responsible for analyzing the data and writing the manuscript. Dr. E.M.J. Bols conceptualized and designed the study, analyzed and interpreted the data and critically revised the manuscript. Prof. Dr. M.A. Benninga conceptualized and designed the study, analyzed and interpreted the data and critically revised the manuscript. Dr. W.A. Verwijs conceptualized and designed the study, analyzed and interpreted the data and critically revised the manuscript. Prof. Dr. R.A. de Bie conceptualized and designed the study, analyzed and interpreted the data, critically revised the manuscript and supervised the study. All authors have approved the final manuscript as submitted and agreed to be accountable for all aspects of the work.

Dr. W.A. Verwijs conceptualized and designed the study, analyzed and interpreted the data and critically revised the manuscript. Prof. Dr. R.A. de Bie conceptualized and designed the study, analyzed and interpreted the data, critically revised the manuscript and supervised the study. All authors have approved the final manuscript as submitted and agreed to be accountable for all aspects of the work. Compliance with ethical statement The Medical Ethics Committee of the Maastricht University Medical Centre approved the study (MEC 15-4-117). Informed consent was obtained from all parents and children, aged 12 years and older included in the study. Funding source No external funding. Financial disclosure The authors have indicated they have no financial relationships relevant to this article to disclose.

Introduction Preterm birth accounts for 11% of all live-births worldwide and is increasing in most countries [4]. In recent years, survival rates of even the most immature preterm infants have markedly improved. Nonetheless, these infants are still at risk for neurologic sequelae that extend beyond the neonatal period [21]. About one fourth of survivors of extremely preterm birth suffer from relevant neurological impairment [9]. Adverse functional consequences persist into adolescence and early adulthood and pose a growing social and economic burden on families and society [1, 13, 16, 20]. Strong efforts are being made to identify markers for early prediction of motor and cognitive outcome in order to aid parental counseling, initiate early support, and thereby improve long-term morbidity. Originally being developed as a bedside monitor in adult intensive care [15], the use of amplitude-integrated EEG (aEEG) has become more and more widespread in neonatal intensive care units (NICU) for continuous monitoring of cerebral function in neonates in recent years [19]. The main field of clinical use of aEEG is the prediction of cerebral outcome after birth asphyxia in term infants [2, 12, 22], the detection of cerebral seizure activity and the surveillance of antiepileptic drug treatment [11].

its (NICU) for continuous monitoring of cerebral function in neonates in recent years [19]. The main field of clinical use of aEEG is the prediction of cerebral outcome after birth asphyxia in term infants [2, 12, 22], the detection of cerebral seizure activity and the surveillance of antiepileptic drug treatment [11]. Also in the population of preterm infants, aEEG is increasingly being used. It is generally well accepted by the NICU staff and has been proven to be a safe method for cerebral function monitoring even in extremely preterm infants [7]. Recent studies have shown a correlation between aEEG recordings obtained during the first weeks of life and short- and long-term neurodevelopmental outcome of preterm infants [3, 14, 18, 25, 27, 29]. In order to become a practical tool for clinical routine use, it is necessary to find easy-to-use schemes for assessment of aEEG tracings in preterm infants.

on between aEEG recordings obtained during the first weeks of life and short- and long-term neurodevelopmental outcome of preterm infants [3, 14, 18, 25, 27, 29]. In order to become a practical tool for clinical routine use, it is necessary to find easy-to-use schemes for assessment of aEEG tracings in preterm infants. Electrocortical activity, and thus the aEEG background pattern, depends on the infant’s gestational age [11]. Several authors have made suggestions for classification and scoring of aEEGs. One of the first publications for the classification of background patterns, sleep-wake cycling and seizure activity in preterm and newborn infants was published by Hellström-Westas et al. [11]. Burdjalov et al. made an attempt to quantify brain maturation by aEEG between 24 and 39 weeks with a scoring system [6]. This score has been cited and modified by several authors describing maturational changes of electrocortical function in premature infants [6, 11, 17, 23, 24]. Another system using reference values for the corresponding gestational age was developed by Olischar et al. [17]. Later, these reference values were used to build a score based on the combination of background activity, sleep-wake cycling, and seizure activity [14]. In recent years, several studies correlated aEEG patterns with clinical conditions (e.g., patent ductus arteriosus, small for gestational age) [5, 10]. Electronical assessment of aEEG is possible but less common, even though a few publications used this approach [5, 26, 28].

eep-wake cycling, and seizure activity [14]. In recent years, several studies correlated aEEG patterns with clinical conditions (e.g., patent ductus arteriosus, small for gestational age) [5, 10]. Electronical assessment of aEEG is possible but less common, even though a few publications used this approach [5, 26, 28]. The aim of our study was to compare two common classifications and scoring systems (Hellström-Westas and Burdjalov), regarding their value for the prediction of survival and of mental and psychomotor outcome. The hypothesis of our study is that survival and long-term outcome following preterm birth may be predicted by early aEEG recordings. Therefore, in our single center cohort of non-sedated infants, we investigated the correlation between aEEG parameters as defined by two common scores obtained within the first 72 h of life and outcome as quantified by survival and by the Bayley II testing at 24 months corrected age. Survival and long-term outcome following preterm birth can be predicted by early aEEG recordings. Both methods proved to be valuable for the assessment of aEEG tracings.

wo common scores obtained within the first 72 h of life and outcome as quantified by survival and by the Bayley II testing at 24 months corrected age. Survival and long-term outcome following preterm birth can be predicted by early aEEG recordings. Both methods proved to be valuable for the assessment of aEEG tracings. Methods Patient recruitment All preterm infants (gestational age ≤ 32 weeks) who were treated in our unit between 01/2009 and 12/2012 and received both aEEG recordings of at least 4 h duration within 72 h of life and the Bayley II at 24 months corrected age or died during their stay in the NICU were eligible for our retrospective study. From a total of 402 patients born ≤32 weeks gestational age, 65 infants were included into the study (Fig. 1). Gestational age was assessed based on the best obstetric estimate (the first day of the last menstruation and ultrasound). Sedation or opiate medication within 12 h before or during the recording was an exclusion criterion. Between 06/2009 and 10/2010, infants were recruited for aEEG recordings as part of the NEOBRAIN study, which was a European multicenter prospective trial. The study included infants <28 weeks gestational age, but in our center, we sought informed consent for all infants ≤32 weeks gestational age in order to conduct aEEG tracings. After the end of recruiting for NEOBRAIN, aEEGs became part of the clinical routine in our NICU. For infants born after 10/2010, parents signed a written consent that clinical data may be used for scientific retrospective analysis upon admission of each infant. The study was approved by the local ethics committee and in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.Fig. 1 Patient recruitment

rents signed a written consent that clinical data may be used for scientific retrospective analysis upon admission of each infant. The study was approved by the local ethics committee and in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.Fig. 1 Patient recruitment aEEG recording All aEEGs were recorded by the nurse or the doctor in charge of the infant within the first 72 h of life. The aEEG was recorded as a two-channel EEG using needle electrodes or gold caps with BRM2 and BRM3 monitors (BrainZ Instruments, New Zealand). Electrodes were placed on the scalp corresponding to the positions C3, P3, C4, and P4 of the international 10–20 system. A reference electrode was placed on the back of the infant. aEEG interpretation All aEEGs were analyzed by two independent investigators (N.B. and H.M.) blinded for the patients’ outcome and using predefined criteria [6, 11]. The first section with continuously good quality of tracing (no artifacts, impedance <15 kΩ) for 4 h within each day of life was selected for assessment. The aEEG recordings were independently scored by each person. In case of disagreeing scores, the tracing was reassessed by both investigators together and consent was sought. The strength of interrater agreement was good (Hellström-Westas: weighted κ = 0.7589 (CI 0.7177, 0.8001); Burdjalov: weighted κ = 0.6265 (CI 0.5626, 0.6904)).

rdings were independently scored by each person. In case of disagreeing scores, the tracing was reassessed by both investigators together and consent was sought. The strength of interrater agreement was good (Hellström-Westas: weighted κ = 0.7589 (CI 0.7177, 0.8001); Burdjalov: weighted κ = 0.6265 (CI 0.5626, 0.6904)). aEEG classifications For the assessment of the Burdjalov score, we evaluated the tracings following the criteria from the original publication [6]. The total score adds up from visual classification of four criteria: continuity, the expression of sleep-wake cycling (SWC), the amplitude of the lower border (LBA), and the bandwidth (BW). The maximum score is 13 and should be reached at a corrected gestational age of 36/37 weeks. Additionally, we adapted the Burdjalov total score (see Table 1) and calculated an “adapted Burdjalov total score”. Furthermore, we assessed the recordings following the classification by Hellström-Westas which additionally includes criteria for pathologic findings [11].Table 1 Overview over parameters for interpretation of aEEG

ly, we adapted the Burdjalov total score (see Table 1) and calculated an “adapted Burdjalov total score”. Furthermore, we assessed the recordings following the classification by Hellström-Westas which additionally includes criteria for pathologic findings [11].Table 1 Overview over parameters for interpretation of aEEG Very immature/depressed pattern More mature/developed pattern Hellström-Westas background pattern Burst suppression pattern/flat trace Discontinuous/continuous pattern Hellström-Westas sleep-wake cycling No cycling Imminent/immature cycling Burdjalov continuity Discontinuous (a0 point; b0 point) Somewhat continuous or better (a1 to 2 points; b1 point) Burdjalov sleep-wake cycling No cycling (a0 point; b0 point) Waves first appear or better (a1 to 5 points; b1 point) Burdjalov lower border < 5 μV (a0 to 1 point; b0 point) >5 μV (a2 to 4 points; b1 point) Burdjalov bandwidth Very depressed/very immature (a0 to 1 point; b0 point) Immature or better (a2 to 4 points; b1 point) abBurdjalov total score (according to [6]), adapted Burdjalov total score

Very immature/depressed pattern More mature/developed pattern Hellström-Westas background pattern Burst suppression pattern/flat trace Discontinuous/continuous pattern Hellström-Westas sleep-wake cycling No cycling Imminent/immature cycling Burdjalov continuity Discontinuous (a0 point; b0 point) Somewhat continuous or better (a1 to 2 points; b1 point) Burdjalov sleep-wake cycling No cycling (a0 point; b0 point) Waves first appear or better (a1 to 5 points; b1 point) Burdjalov lower border < 5 μV (a0 to 1 point; b0 point) >5 μV (a2 to 4 points; b1 point) Burdjalov bandwidth Very depressed/very immature (a0 to 1 point; b0 point) Immature or better (a2 to 4 points; b1 point) abBurdjalov total score (according to [6]), adapted Burdjalov total score Outcome and Bayley II testing Testing was performed by trained staff as part of the national routine follow-up program for preterm infants. MDI scores were determined in 59 infants and PDI scores in 53 infants. Six of the 65 included infants died during their stay in the NICU, thus no MDI and PDI scores were available. Favorable outcome was defined as both MDI and PDI ≥ 70, poor outcome was defined either one or both MDI and PDI < 70. Infants who died during the neonatal period were summed up in a third group. MDI or PDI scores < 50 were considered as 45 to enable statistical analysis.

NICU, thus no MDI and PDI scores were available. Favorable outcome was defined as both MDI and PDI ≥ 70, poor outcome was defined either one or both MDI and PDI < 70. Infants who died during the neonatal period were summed up in a third group. MDI or PDI scores < 50 were considered as 45 to enable statistical analysis. Statistical analysis For quantitative variables, mean and standard error of the mean (SEM) have been calculated; ordinal scaled parameters are presented by median and range; for qualitative factors, absolute and relative frequencies are given. To correlate the outcome (good–poor–dead) with aEEG parameters, findings of each parameter were divided into two categories: very immature/depressed versus more mature/developed (Table 1). For each day, we evaluated whether aEEG parameters showed a significant difference between the groups using the chi-square-test, Fisher’s exact test, Kruskal-Wallis, or Mann Whitney U test, as appropriate. In order to compare two mean values resulting from data which are approximately normally distributed, two sample t tests has been performed. In order to compare three outcome groups regarding gestational age and birth weight, a one-way ANOVA has been performed followed by the Tukey-Kramer post hoc tests in the case of a significant test result. To evaluate the association between aEEG parameters and MDI/PDI, we performed multiple regression analyses adjusting for birth weight and gestational age.

regarding gestational age and birth weight, a one-way ANOVA has been performed followed by the Tukey-Kramer post hoc tests in the case of a significant test result. To evaluate the association between aEEG parameters and MDI/PDI, we performed multiple regression analyses adjusting for birth weight and gestational age. The strength of interrater agreement concerning aEEG recordings has been assessed using weighted Kappa coefficients using weight factors according to Cicchetti Allison [8]. All statistical calculations have been done using SAS software, release 9.3 (SAS Institute Inc., Cary, NC, USA). The result of a statistical test has been considered as significant for p < 0.05. Results Sixty-five infants met the inclusion criteria. Mean birth weight was 1022 ± 47 g (mean ± SEM; range 440–1880 g; median 985 g) with a mean gestational age of 27.3 ± 0.3 weeks (mean ± SEM; range 23–32 weeks; median 27 weeks). Six infants died during treatment in our NICU. For the clinical details of the cohort, we refer to Tables 2 and 3. The median aEEG recording time was 63 h and 41 min (range 4:55–91:16) with 50 utilizable aEEGs on day 1, 61 on day 2, and 52 on day 3. Bayley testing was performed in the surviving infants at a mean age of 24.0 ± 0.5 months (mean ± SEM), (range 15–31 months, quartiles 23 and 27 months).Table 2 Clinical data of cohort

. The median aEEG recording time was 63 h and 41 min (range 4:55–91:16) with 50 utilizable aEEGs on day 1, 61 on day 2, and 52 on day 3. Bayley testing was performed in the surviving infants at a mean age of 24.0 ± 0.5 months (mean ± SEM), (range 15–31 months, quartiles 23 and 27 months).Table 2 Clinical data of cohort Outcome GA: mean ± SEM (range; median) [weeks] BW: mean ± SEM (range; median) [g] Gender [male/female] CA [n] SGA <10. P. [n] Caffeine therapy [n] Indo-methacin therapy [n] IVH III/IV [n] PVL [n] Severe BPDa [n] Mechanical ventilation within first 72 h of life [n] Catech within first 72 h of life [n] Good (n = 40) 28.1 ± 0.3 (24–32; 28.0) 1125 ± 52 (610–1880; 1060) 19/21 9 4 39 17 1 1 2 9 3 Poor (n = 19) 26.7 ± 0.6 (23–31; 27.0) 959 ± 95 (450–1810; 985) 5/14 5 3 18 13 0 2 6 11 0 Dead (n = 6) 23.7 ± 0.3 (23–25; 24.0) 534 ± 33 (440–649; 523) 0/6 1 1 6 1 4 0 – 6 0 Total (n = 65) 27.3 ± 0.3 (23–32; 27.0) 1022 ± 47 (440–1880; 985) 24/41 15 8 63 32 5 3 8 26 3 GA gestational age, BW birth weight, CA chorioamnionitis, SGA small for gestational age, BPD bronchopulmonary dysplasia, catech catecholamines aO2 > 36 weeks postmenstrual age; mechanical ventilation within first 72 h of life inludes only ventilation longer than 1 h Table 3 Clinical data of deceased infants

Outcome GA: mean ± SEM (range; median) [weeks] BW: mean ± SEM (range; median) [g] Gender [male/female] CA [n] SGA <10. P. [n] Caffeine therapy [n] Indo-methacin therapy [n] IVH III/IV [n] PVL [n] Severe BPDa [n] Mechanical ventilation within first 72 h of life [n] Catech within first 72 h of life [n] Good (n = 40) 28.1 ± 0.3 (24–32; 28.0) 1125 ± 52 (610–1880; 1060) 19/21 9 4 39 17 1 1 2 9 3 Poor (n = 19) 26.7 ± 0.6 (23–31; 27.0) 959 ± 95 (450–1810; 985) 5/14 5 3 18 13 0 2 6 11 0 Dead (n = 6) 23.7 ± 0.3 (23–25; 24.0) 534 ± 33 (440–649; 523) 0/6 1 1 6 1 4 0 – 6 0 Total (n = 65) 27.3 ± 0.3 (23–32; 27.0) 1022 ± 47 (440–1880; 985) 24/41 15 8 63 32 5 3 8 26 3 GA gestational age, BW birth weight, CA chorioamnionitis, SGA small for gestational age, BPD bronchopulmonary dysplasia, catech catecholamines aO2 > 36 weeks postmenstrual age; mechanical ventilation within first 72 h of life inludes only ventilation longer than 1 h Table 3 Clinical data of deceased infants No. GA [weeks] BW [g] Umbilical arterial blood pH APGAR ’10 Day of life when died Cause of death 1 23 470 Not determinable 7 3 Severe IVH, respiratory failure 2 23 495 7.41 7 7 Septic shock 3 24 610 7.29 8 4 Cardiorespiratory failure 4 24 440 7.29 8 9 Multi organ failure 5 23 550 6.60 3 2 Cardiorespiratory failure, asphyxia after placentar disruption 6 25 640 7.21 7 9 Severe IVH, intestinal perforation, cardiorespiratory failure GA gestational age, BW birth weight, IVH intraventricular hemorrhage

No. GA [weeks] BW [g] Umbilical arterial blood pH APGAR ’10 Day of life when died Cause of death 1 23 470 Not determinable 7 3 Severe IVH, respiratory failure 2 23 495 7.41 7 7 Septic shock 3 24 610 7.29 8 4 Cardiorespiratory failure 4 24 440 7.29 8 9 Multi organ failure 5 23 550 6.60 3 2 Cardiorespiratory failure, asphyxia after placentar disruption 6 25 640 7.21 7 9 Severe IVH, intestinal perforation, cardiorespiratory failure GA gestational age, BW birth weight, IVH intraventricular hemorrhage The first day of life: The distribution of background patterns and sleep-wake cycling was very similar between the groups with good and poor outcome following both the definitions of Hellström-Westas and Burdjalov. However, in the group of deceased infants very immature/depressed findings for both background pattern and cycling were more prevalent, but reached no statistical significance (Fig. 2, see Table 4 for p values).Fig. 2 Distribution of very immature/depressed (left side) and of mature/developed (right side) aEEG patterns on each day in all three outcome groups (good outcome–poor outcome–dead) is demonstrated in percent of each outcome group. Adverse short-term outcome (death) is reflected by the absence of mature/developed patterns during the first 72 h of life. For each day, we evaluated whether aEEG parameters showed a significant difference between groups (good–poor–death) using Fisher’s exact test: Cycling on day 2 (Hellström-Westas, p = 0.0140; Burdjalov, p = 0.0041) was significant; for details see Table 4 Table 4 aEEG patterns in the different outcome groups on day 1–3

The first day of life: The distribution of background patterns and sleep-wake cycling was very similar between the groups with good and poor outcome following both the definitions of Hellström-Westas and Burdjalov. However, in the group of deceased infants very immature/depressed findings for both background pattern and cycling were more prevalent, but reached no statistical significance (Fig. 2, see Table 4 for p values).Fig. 2 Distribution of very immature/depressed (left side) and of mature/developed (right side) aEEG patterns on each day in all three outcome groups (good outcome–poor outcome–dead) is demonstrated in percent of each outcome group. Adverse short-term outcome (death) is reflected by the absence of mature/developed patterns during the first 72 h of life. For each day, we evaluated whether aEEG parameters showed a significant difference between groups (good–poor–death) using Fisher’s exact test: Cycling on day 2 (Hellström-Westas, p = 0.0140; Burdjalov, p = 0.0041) was significant; for details see Table 4 Table 4 aEEG patterns in the different outcome groups on day 1–3 Day 1 in all outcome groups Day 2 in all outcome groups ay 3 in all outcome groups Good (n = 32) Poor (n = 14) Dead (n = 4) p value Good (n = 37) Poor (n = 19) Dead (n = 5) p value Good (n = 34) Poor (n = 16) Dead (n = 2) p value HW background In % (n): pathological (burst suppression/flat trace) vs. discontinuous/continuous 31 (10) vs 69 (22) 29 (4) vs 71 (10) 75 (3) vs 25 (1) 0.2548 a 16 (6) vs 84 (31) 11 (2) vs 89 (17) 60 (3) vs 40 (2) 0.0567 a 12 (4) vs 88 (30) 6 (1) vs 94 (15) 0 (0) vs 100 (2) 1.000

Day 1 in all outcome groups Day 2 in all outcome groups ay 3 in all outcome groups Good (n = 32) Poor (n = 14) Dead (n = 4) p value Good (n = 37) Poor (n = 19) Dead (n = 5) p value Good (n = 34) Poor (n = 16) Dead (n = 2) p value HW background In % (n): pathological (burst suppression/flat trace) vs. discontinuous/continuous 31 (10) vs 69 (22) 29 (4) vs 71 (10) 75 (3) vs 25 (1) 0.2548 a 16 (6) vs 84 (31) 11 (2) vs 89 (17) 60 (3) vs 40 (2) 0.0567 a 12 (4) vs 88 (30) 6 (1) vs 94 (15) 0 (0) vs 100 (2) 1.000 a HW cycling In % (n): no cycling vs. immature/mature 31 (10) vs 69 (22) 36 (5) vs 64 (9) 75 (3) vs 25 (1) 0.2744 a 16 (6) vs 84 (31) 21 (4) vs 79 (15) 80 (4) vs 20 (1) 0.0140 a 12 (4) vs 88 (30) 19 (3) vs 81 (13) 50 (1) vs 50 (1) 0.2632 a Burdjalov continuity In % (n): discontinuous vs. somewhat continuous/continuous 41 (13) vs 59 (19) 50 (7) vs 50 (7) 100 (4) vs 0 (0) 0.0851 a 35 (13) vs 65 (24) 53 (10) vs 47 (9) 80 (4) vs 20 (1) 0.1260 a 26 (9) vs 74 (25) 44 (7) vs 56 (9) 100 (2) vs 0 (0) 0.0876 a Burdjalov cycling In % (n): no cycling vs. waves first appear or better 44 (14) vs 56 (18) 36 (5) vs 64 (9) 100 (4) vs 0 (0) 0.0931 a 24 (9) vs 76 (28) 37 (7) vs 63 (12) 100 (5) vs 0 (0) 0.0041 a 15 (5) vs 85 (29) 19 (3) vs 81 (13) 100 (2) vs 0 (0) 0.0504 a Burdjalov lower border In % (n): <5 μV vs. >5 μV 84 (27) vs 16 (5) 86 (12) vs 14 (2) 100 (4) vs 0 (0) 1.000 a 76 (28) vs 24 (9) 84 (16) vs 16 (3) 100 (5) vs 0 (0) 0.6146 a 71 (24) vs 29 (10) 75 (12) vs 25 (4) 100 (2) vs 0 (0) 1.000 a

a 24 (9) vs 76 (28) 37 (7) vs 63 (12) 100 (5) vs 0 (0) 0.0041 a 15 (5) vs 85 (29) 19 (3) vs 81 (13) 100 (2) vs 0 (0) 0.0504 a Burdjalov lower border In % (n): <5 μV vs. >5 μV 84 (27) vs 16 (5) 86 (12) vs 14 (2) 100 (4) vs 0 (0) 1.000 a 76 (28) vs 24 (9) 84 (16) vs 16 (3) 100 (5) vs 0 (0) 0.6146 a 71 (24) vs 29 (10) 75 (12) vs 25 (4) 100 (2) vs 0 (0) 1.000 a Burdjalov bandwidth In % (n): very depressed/very immature vs. immature or better 84 (27) vs 16 (5) 86 (12) vs 14 (2) 100 (4) vs 0 (0) 1.000 a 68 (25) vs 32 (12) 84 (16) vs 16 (3) 100 (5) vs 0 (0) 0.2327 a 53 (18) vs 47 (16) 81 (13) vs 19 (3) 100 (2) vs 0 (0) 0.0810 a Burdjalov total score d Score: range; median 1–8; 3.5 1–9; 3.5 0–2; 2 0.0871 b 1–10; 4.0 2–10; 3.0 2–3; 3.0 0.0284 b 1–9; 5.0 2–9; 4.0 2.0; 2.0 0.0526 b Burdjalov total score (adapted e) Score: range; median 0–4; 1.5 0–4; 1.5 0–0; 0 0.0769 b 0–4: 2.0 0–4; 1.0 0–1; 0 0.0183 b 0–4; 2.0 0–4; 2.0 0–0; 0 0.0472 b Birth weight [g] mean ± SEM (range) 1112 ± 56 (610–1730) 976 ± 121 (450–1810) 524 ± 37 (440–610) 0.0007 c 1134 ± 54 (610–1880) 959 ± 95 (450–1810) 531 ± 40 (440–640) 0.0018 c 1102 ± 59 (610–1880) 889 ± 93 (450–1490) 625 ± 15 (610–640) 0.0447 c Gestational age [weeks] mean ± SEM (range) 27.9 ± 0.4 (24.0–32.0) 26.9 ± 0.7 (23.0–31.0) 23.5 ± 0.3 (23.0–24.0) <0.0001 c 28.1 ± 0.4 (24.0–32.0) 26.7 ± 0.6 (23.0–31.0) 23.8 ± 0.4 (23.0–25.0) 0.0004 c 27.8 ± 0.4 (24.0–32.0) 26.3 ± 0.6 (23.0–30.0) 24.5 ± 0.5 (24.0–25.0) 0.0238 c HW Hellström-Westas, vs versus, SEM standard error of the mean

Gestational age [weeks] mean ± SEM (range) 27.9 ± 0.4 (24.0–32.0) 26.9 ± 0.7 (23.0–31.0) 23.5 ± 0.3 (23.0–24.0) <0.0001 c 28.1 ± 0.4 (24.0–32.0) 26.7 ± 0.6 (23.0–31.0) 23.8 ± 0.4 (23.0–25.0) 0.0004 c 27.8 ± 0.4 (24.0–32.0) 26.3 ± 0.6 (23.0–30.0) 24.5 ± 0.5 (24.0–25.0) 0.0238 c HW Hellström-Westas, vs versus, SEM standard error of the mean aFisher’s exact test, b Kruskal-Wallis test, c one-way ANOVA, d Burdjalov total score according to [6], e adapted Burdjalov total score (see Table 1) There was a statistical difference when comparing the birth weights or gestational age between the three outcome groups (gestational age p < 0.0001; birth weight p = 0.0007 using one-way ANOVAs). Post hoc tests according to Tukey-Kramer revealed significant differences between good and bad outcome (p < 0.0001 and p = 0.0006, accordingly) as well as between poor and bad outcome (p = 0.0140 and p = 0.0278, accordingly). The differences between good and poor outcome failed to be significant (p = 0.0597 and 0.2028, accordingly)

s according to Tukey-Kramer revealed significant differences between good and bad outcome (p < 0.0001 and p = 0.0006, accordingly) as well as between poor and bad outcome (p = 0.0140 and p = 0.0278, accordingly). The differences between good and poor outcome failed to be significant (p = 0.0597 and 0.2028, accordingly) Second day of life: 16–21% of infants with good and poor outcome each presented very immature/depressed findings in cycling (Hellström-Westas), whereas 24% of the infants with good and 37% of infants with poor outcome showed very immature/depressed characteristics in cycling by Burdjalov. Again, the group of deceased infants showed very immature/depressed patterns in a higher percentage of cases: no cycling could be found in 80% (Hellström-Westas) and 100% (Burdjalov). The difference for both parameters was significant (p = 0.0140 and p = 0.0041; Fig. 2, Table 4). There was no significance for background pattern (Hellström-Westas), continuity, lower border, and bandwidth (all Burdjalov). Third day of life: On day 3, we found no difference between aEEG parameters, but the absolute number in the group of deceased infants was small (n = 2). For the details for all 3 days of life refer to Table 4.

Second day of life: 16–21% of infants with good and poor outcome each presented very immature/depressed findings in cycling (Hellström-Westas), whereas 24% of the infants with good and 37% of infants with poor outcome showed very immature/depressed characteristics in cycling by Burdjalov. Again, the group of deceased infants showed very immature/depressed patterns in a higher percentage of cases: no cycling could be found in 80% (Hellström-Westas) and 100% (Burdjalov). The difference for both parameters was significant (p = 0.0140 and p = 0.0041; Fig. 2, Table 4). There was no significance for background pattern (Hellström-Westas), continuity, lower border, and bandwidth (all Burdjalov). Third day of life: On day 3, we found no difference between aEEG parameters, but the absolute number in the group of deceased infants was small (n = 2). For the details for all 3 days of life refer to Table 4. The Burdjalov total score significantly differed between groups on day 2 (p = 0.0284, day 1 and 3 not significant) and the adapted Burdjalov total score on day 2 (p = 0.0183) and day 3 (p = 0.0472). However, the range is large and overlaps between groups, but the maximum total score is lower in the group of deceased infants (Table 4, Fig. 3).Fig. 3 The Burdjalov total score and the adapted Burdjalov total score (median) in the three different outcome groups during the first 3 days of life

) and day 3 (p = 0.0472). However, the range is large and overlaps between groups, but the maximum total score is lower in the group of deceased infants (Table 4, Fig. 3).Fig. 3 The Burdjalov total score and the adapted Burdjalov total score (median) in the three different outcome groups during the first 3 days of life the Burdjalov total score significantly differed between outcome groups on day 2 (p = 0.0284, Kruskal-Wallis test) and the adapted Burdjalov total score on day 2 (p = 0.0183, Kruskal-Wallis test) and day 3 (p = 0.0472, Kruskal-Wallis test), see also Table 4 Multiple regression analyses revealed sleep-wake cycling (Hellström-Westas) on day 3 of life (p = 0.0059) and background on day 3 (Hellström-Westas; p = 0.0212) as independent predictors for MDI whereas no independent predictor for PDI was found (multiple regression analyses).

the Burdjalov total score significantly differed between outcome groups on day 2 (p = 0.0284, Kruskal-Wallis test) and the adapted Burdjalov total score on day 2 (p = 0.0183, Kruskal-Wallis test) and day 3 (p = 0.0472, Kruskal-Wallis test), see also Table 4 Multiple regression analyses revealed sleep-wake cycling (Hellström-Westas) on day 3 of life (p = 0.0059) and background on day 3 (Hellström-Westas; p = 0.0212) as independent predictors for MDI whereas no independent predictor for PDI was found (multiple regression analyses). Discussion Continuous monitoring of brain function in very immature infants remains exceptional in clinical routine, as recording, assessment and interpretation of preterm EEG and aEEG are time consuming and appear sometimes conflicting with principles of minimal handling. To establish easy algorithms for aEEG interpretation, several authors have reported maturational changes of electrocortical activity according to both postnatal and gestational age. Recent studies have pointed out a correlation of early aEEG parameters with short- and long-term neurological outcome [3, 14, 18, 25, 27, 29]. In this study, we refer to and compare the classifications by Hellström-Westas et al. [11] and the aEEG maturation score introduced by Burdjalov et al. in a single center cohort of prematurely born infants ≤32 weeks gestation [6]. We showed that there is a correlation in both classifications between the absence of cycling on the second day of life and the risk of death. Also, we demonstrated a correlation of background pattern and cyclicity (both HW) with long-term mental outcome.

enter cohort of prematurely born infants ≤32 weeks gestation [6]. We showed that there is a correlation in both classifications between the absence of cycling on the second day of life and the risk of death. Also, we demonstrated a correlation of background pattern and cyclicity (both HW) with long-term mental outcome. It has previously been shown, that continuity within the first 72 h of life, especially the presence of pathological background pattern correlates with adverse short-term outcome (IVH °III/IV) in infants between 25 and 32 weeks gestational age [25]. The presence of low voltage pattern and the lack of sleep-wake cycling during the first 72 h of life were associated with death in preterm infants ≤1500 g or ≤32 weeks [3]. The present results from our cohort are in wide agreement with these findings. Infants who died during treatment on the NICU rarely showed physiological patterns in aEEG recordings within the first 72 h of life. The Burdjalov total score and the adapted Burdjalov total score did not add additional information despite the fact that significant differences between scores were found on day 2 (usual Burdjalov total score) or on days 2 and 3 (adapted Burdjalov total score), as the ranges between the different outcome groups were strongly overlapping.

score and the adapted Burdjalov total score did not add additional information despite the fact that significant differences between scores were found on day 2 (usual Burdjalov total score) or on days 2 and 3 (adapted Burdjalov total score), as the ranges between the different outcome groups were strongly overlapping. There is strong evidence that neuropsychological deficits can be predicted within the first 72 h of life. Sleep-wake cycling and a combined score of background activity, cycling and seizure activity during the first 2 weeks of life correlate with Bayley results at 3 years of age, cerebral palsy and death in preterm infants <30 weeks gestational age [14]. Prolonged interburst intervals and burst suppression pattern are strong predictors of poor neurodevelopmental outcome at 2 years corrected age in preterm infants between 22 and 30 gestational weeks [29]. Burdjalov total scores from aEEGs obtained within the first 6 weeks of life in infants ≤30 gestational weeks were associated with perinatal factors that are known to predict adverse neurodevelopmental outcome [18]. On the other hand, the rate of brain wave maturation between 28 and 36 weeks gestational age in preterm infants <28 weeks does not predict neurological development at 18 to 22 months corrected age [27].

ional weeks were associated with perinatal factors that are known to predict adverse neurodevelopmental outcome [18]. On the other hand, the rate of brain wave maturation between 28 and 36 weeks gestational age in preterm infants <28 weeks does not predict neurological development at 18 to 22 months corrected age [27]. The majority of studies on prediction of outcome by aEEG revealed, that the presence of pathological patterns predicts poor outcome (short- and long-term). Our findings are in agreement with the fact that prediction of favorable outcome seems to be more complex. Of those infants who died during their stay in the NICU, only very few showed physiological aEEG patterns. Thus, the presence of physiological patterns seems to be a predictor of survival. Only the classification by Hellström-Westas included independent predictors for long-term outcome: Cycling and background on day 3 were associated with MDI at 24 months corrected age.

ICU, only very few showed physiological aEEG patterns. Thus, the presence of physiological patterns seems to be a predictor of survival. Only the classification by Hellström-Westas included independent predictors for long-term outcome: Cycling and background on day 3 were associated with MDI at 24 months corrected age. We conclude that both systems are valuable tools for the assessment of aEEG tracings in preterm infants. They are easy to apply and make bedside evaluation possible, as they can be interpreted “at a glance”. One main difference between the two classifications is that the Burdjalov score is primarily designed to describe the physiological maturation of electrocortical activity and only indirectly provides measures for pathological patterns. The classification by Hellström-Westas on the other hand, is designed to distinguish pathological and physiological patterns rather than to describe maturational changes over time. In our study, the presence of cycling was helpful to predict survival. Background pattern and cycling (both HW) were helpful to predict long-term mental outcome. As the results of this study are preliminary data, prospective studies are needed to further evaluate the potential contribution of early aEEG recordings to decision-making in very sick preterm infants. Abbreviations aEEGAmplitude-integrated EEG BGBackground pattern BWBandwidth EEGElectroencephalography HWHellström-Westas IVHIntraventricular hemorrhage LBAAmplitude of lower border MDIMental development index PDIPsychomotor development index SEMStandard error of the mean SWCSleep-wake cycling

We conclude that both systems are valuable tools for the assessment of aEEG tracings in preterm infants. They are easy to apply and make bedside evaluation possible, as they can be interpreted “at a glance”. One main difference between the two classifications is that the Burdjalov score is primarily designed to describe the physiological maturation of electrocortical activity and only indirectly provides measures for pathological patterns. The classification by Hellström-Westas on the other hand, is designed to distinguish pathological and physiological patterns rather than to describe maturational changes over time. In our study, the presence of cycling was helpful to predict survival. Background pattern and cycling (both HW) were helpful to predict long-term mental outcome. As the results of this study are preliminary data, prospective studies are needed to further evaluate the potential contribution of early aEEG recordings to decision-making in very sick preterm infants. Abbreviations aEEGAmplitude-integrated EEG BGBackground pattern BWBandwidth EEGElectroencephalography HWHellström-Westas IVHIntraventricular hemorrhage LBAAmplitude of lower border MDIMental development index PDIPsychomotor development index SEMStandard error of the mean SWCSleep-wake cycling Authors’ contributions NB, UFM, and HM designed the study and wrote the paper. NB and HM analyzed the aEEGs using different aEEG classifications. BH and TS collected the aEEG recordings. JH and FD organized and performed the Bayley testing. CW realized the statistical tests.

SEMStandard error of the mean SWCSleep-wake cycling Authors’ contributions NB, UFM, and HM designed the study and wrote the paper. NB and HM analyzed the aEEGs using different aEEG classifications. BH and TS collected the aEEG recordings. JH and FD organized and performed the Bayley testing. CW realized the statistical tests. Compliance with ethical standards The study was approved by the local ethics committee and in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Conflict of interest The authors declare that they have no conflict of interest. No funds were received for this study.

Introduction Inherited NeuroMetabolic Diseases (iNMDs) represent a particularly important group of rare diseases being constituted by genetic metabolic disorders that may impact on the brain from birth (sometimes already in utero) and during the whole developmental period causing mental retardation and/or different neurological signs and symptoms, may be progressive and may ultimately end with irreversible consequences and even in early death [2]. Today, as a result of medical research, major progress has been made. The scenario of the ‘personalised medicine’, based on the premise ‘the right treatment for the right patient at the right time’, is proposing the use of the latest advancements in biological knowledge and technology, such as the completion of the human genome mapping, to improve understanding of inter-individual variability, in order to include them in the medicine developmental process. The research based on ‘omics’ technologies, including pharmacogenetics/pharmacogenomics (PGt/PGx), is producing a very high amount of data. In fact, these approaches have been demonstrated to be useful both in pharmacological treatment and in diagnosis. However, the number of curative interventions remains low and this low number of approved drugs is strictly connected to the difficulties characterising the research in this field that can be summarised as the following.

Today, as a result of medical research, major progress has been made. The scenario of the ‘personalised medicine’, based on the premise ‘the right treatment for the right patient at the right time’, is proposing the use of the latest advancements in biological knowledge and technology, such as the completion of the human genome mapping, to improve understanding of inter-individual variability, in order to include them in the medicine developmental process. The research based on ‘omics’ technologies, including pharmacogenetics/pharmacogenomics (PGt/PGx), is producing a very high amount of data. In fact, these approaches have been demonstrated to be useful both in pharmacological treatment and in diagnosis. However, the number of curative interventions remains low and this low number of approved drugs is strictly connected to the difficulties characterising the research in this field that can be summarised as the following. First of all, when dealing with children, clinical research raises many scientific and ethical concerns, and special provisions should apply [6]. Due to developmental and cognitive limitations and legal incapacity, a child is legally unable to provide informed consent to participate in research that should be done by a legal representative [28]. On the other hand, children are entitled to receive appropriate information about their health and their participation in clinical studies that imply that their views can be sought and taken into account. Challenges and obstacles in accessing the views of children and young people have been identified by the Paediatric Committee (PDCO) and include access, language/ability to express oneself, ethnic and cultural differences, chronological age and maturity, ill-informed or preconceived notions about a child’s level of understanding and maturity [26]. The call for more medical research involving children, as enlightened by the European Directive 2001/20/EC [30] and the Paediatric Regulation [32, 33], has also raised the question whether the law strikes an appropriate balance between the need for clinical trials and the interests and rights of the child.

turity [26]. The call for more medical research involving children, as enlightened by the European Directive 2001/20/EC [30] and the Paediatric Regulation [32, 33], has also raised the question whether the law strikes an appropriate balance between the need for clinical trials and the interests and rights of the child. Secondly, when paediatric patients are affected by rare diseases, difficulties double [5] as children should be considered ‘twice orphan’, and randomised controlled trials, considered as the standard in research design, are even more unfeasible due to the smaller number of patients. It becomes ethically problematic to propose a control arm (in which the investigational product/approach is not provided to a segment of the population) for a study aimed to establish the efficacy of a new product/approach. The difficulties to propose a control arm are also true for prenatal and neonatal screenings, which represent a fundamental step in the diagnosis of genetic diseases. Screening programs for genetic diseases have multiplied in the last 50 years [50]. Research in this field is often observational with either historical control data or control through comparisons with similar populations without screening [3].

gs, which represent a fundamental step in the diagnosis of genetic diseases. Screening programs for genetic diseases have multiplied in the last 50 years [50]. Research in this field is often observational with either historical control data or control through comparisons with similar populations without screening [3]. More in general, clinical research on genetic diseases raises additional scientific and ethical concerns, including specific ethical procedures for genetic research, collection, storage and access to genetic materials, aims of the use of genetic information, time of archiving genetic material in biobanks, informed consent, special issues dealing with confidentiality and paediatrics. Genetic testing guides the prevention, clinical management and drug treatment based on genetic susceptibilities [4]. Carriers/patients must consider disclosure of information to others and weigh the right to privacy against common interests [17, 39]. In addition, knowledge of one’s carrier status for recessive genetic diseases is useful primarily in making reproductive decisions. Such decisions are within the private domain of the young adults who are dating, mating and forming new families. The privacy of these decisions may be compromised when parents know the carrier status of their children.

carrier status for recessive genetic diseases is useful primarily in making reproductive decisions. Such decisions are within the private domain of the young adults who are dating, mating and forming new families. The privacy of these decisions may be compromised when parents know the carrier status of their children. Another aspect to be considered is the use of advanced therapy medicinal products (ATMPs), including gene therapy, and the so-called personalised medicine. Today, for this type of diseases, this represents the most innovative research. For example, when a product is autologous (meaning that it ‘originates from the same patient’), who is the owner of the product? The company or the patient? Finally, the neurological impairment characterising patients affected by iNMDs can lead to mental disability and make both paediatric and adult patients vulnerable (persons relatively or absolutely incapable of protecting their own interests and therefore not capable of giving adequately consent) unable to give their consent or assent. The current legislation [28] defines these patients ‘incapacitated subjects’. In order to perform well-conducted research in this field, many regulatory/ethical and legal provisions need to be followed by researchers, not limited to good clinical practice [25], to considering all these aspects. For this reason, this work aims the following:To identify the relevant provisions to perform well-conducted research from the ethical point of view in the field of iNMDs, taking into account the issues listed above

be followed by researchers, not limited to good clinical practice [25], to considering all these aspects. For this reason, this work aims the following:To identify the relevant provisions to perform well-conducted research from the ethical point of view in the field of iNMDs, taking into account the issues listed above To evaluate if the existing provisions cover the ethical issues related to:Rare diseases: clinical studies require ad hoc methodological approaches which should minimise the number of patients and ensure the adequateness of results and the statistical power at the same time Children and patients with neurological symptoms: vulnerable patients requiring ad hoc protection Genetic/inherited diseases: specific ethical procedures for genetic research, collection, storage and access to genetic materials, aims of the use of genetic information, time of archiving genetic material in biobanks and informed consent All these conditions share the important ethical issue concerning the need for studying and making available drugs notwithstanding the scarce economic return coming from their development. Methods In order to reach the aim of this work, the following actions were undertaken:To search for the current ethical provisions applicable in Europe to the iNMD research

All these conditions share the important ethical issue concerning the need for studying and making available drugs notwithstanding the scarce economic return coming from their development. Methods In order to reach the aim of this work, the following actions were undertaken:To search for the current ethical provisions applicable in Europe to the iNMD research The actual European regulatory/ethical and legal framework was reviewed. Other relevant international guidelines and texts were taken into account, as well. The following sources were consulted: EudraLex Volume 10—Clinical trials guidelines; International Conference on Harmonisation (ICH) guidelines; World Health Organization (WHO)—Council for International Organizations of Medical Sciences (CIOMS) guidelines; EC Public Health publications; Council of Europe treaties; World Medical Association publications; UNESCO Declarations; and European Medicines Agency (EMA). Then, the applicability of retrieved documents in the iNMD field was analysed by considering the specific features of iNMDs, i.e.:The rarity of conditions The presence of paediatric patients The neurological impairment that often compromises their capability to provide the consent or assent The genetic nature of the disease 2. To review relevant data from the literature An ad hoc bibliographic search was performed in PubMed (Fig. 1). The publications discussing ethical issues related to research involving iNMD patients and the application of the existing provisions in the field were considered compliant with the search.Fig. 1 Bibliographic search flow chart

2. To review relevant data from the literature An ad hoc bibliographic search was performed in PubMed (Fig. 1). The publications discussing ethical issues related to research involving iNMD patients and the application of the existing provisions in the field were considered compliant with the search.Fig. 1 Bibliographic search flow chart The following search strategy was used: (“Brain Diseases, Metabolic, Inborn”[Mesh]) AND “Ethics, Clinical”[Mesh] OR (“Brain Diseases, Metabolic, Inborn”[Mesh]) AND “Ethics, Research”[Mesh]. Only publications in English were considered. Even in this second step, the rarity of conditions, the inheritance and familiar implications, the presence of paediatric patients and the manifestation of neurologic symptoms which can make paediatric and adult patients unable to provide the consent were focused on.3. To find relevant results and achievements from European projects We also looked for data relevant to the aims of this work from already existing EU-funded projects on paediatric and rare disease research. The Community Research and Development Information Service (CORDIS) and Consumers, Health, Agriculture and Food Executive Agency (CHAFEA) were searched by using the following keywords: (paediatric OR rare or genetic) AND disease AND clinical AND research AND ethics. Public websites were consulted, where existing.

arch. The Community Research and Development Information Service (CORDIS) and Consumers, Health, Agriculture and Food Executive Agency (CHAFEA) were searched by using the following keywords: (paediatric OR rare or genetic) AND disease AND clinical AND research AND ethics. Public websites were consulted, where existing. Results The current ethical provisions applicable in Europe to inherited neurometabolic diseases The actual European regulatory/ethical and legal framework offers document ruling, in general, the clinical research including ethical provisions applicable to the iNMD research: Regulation (EU) 536/2014 [28]; Directive 2005/28/EC [20]; ICH Guideline for Good Clinical Practice [25]; EMA Reflection paper on ethical and GCP aspects of clinical trials of medicinal products for human use conducted outside of the EU/EEA and submitted in marketing authorization applications [27]; and European Commission guidelines on good clinical practice specific to advanced therapy medicinal products [22]. Issues related to both paediatrics and vulnerability and disability were found in almost all documents, mainly dealing with informed consent and ethics committee expertise/composition (Table 1). In contrast, no specific ethical issue on ATMPs is available.Table 1 Regulatory/ethical/legal provisions applicable in the European Union, the research on inherited neurometabolic diseases

lity were found in almost all documents, mainly dealing with informed consent and ethics committee expertise/composition (Table 1). In contrast, no specific ethical issue on ATMPs is available.Table 1 Regulatory/ethical/legal provisions applicable in the European Union, the research on inherited neurometabolic diseases Document Relevant topics Specific provisions Paediatric issues Rare diseases Inherited/genetic issues Mental disability/vulnerability Regulation (EU) 536/2014 Authorisation and conduct of trials, ethics committees, informed consent and assent process, vulnerability, minors, data protection and confidentiality, protocol Art. 10, 32, 35 Recitals 19, 27 Recitals 9, 10 Art. 10, 28, 29, 31, 35 Recitals 15, 19, 27 Directive 2005/28/EC Conduct of the trial, ethics committee n.s. n.s. n.s. n.s. ICH Topic E 6 (R1) Guideline Conduct of trials, informed consent, clinical trial protocol, ethics committee, vulnerability Par. 4.8.12 n.s. n.s. Par. 1.61, 3.1.1, 4.8.12 EMA Reflection paper on clinical trials conducted outside EU/EEA Conduct of multi-national trials, informed consent and assent process, ethics committee, confidentiality vulnerability, design of clinical trials 4.2, 4.3, 4.5, 5 5.1, 5.2 4.2, 4.5 EC guidelines on advanced therapy medicinal products Conduct of trials with advanced therapies, clinical trial protocol n.s. n.s. n.s. n.s. EU Charter of Fundamental rights Children’s rights Art. 24 n.s. n.s. n.s. Declaration of Helsinki Conduct of human research, vulnerability, risk/benefit, ethics committee, informed consent, privacy and confidentiality 19, 20, 28, 29 n.s. n.s. 19, 20, 28, 29, 30 Oviedo Convention Informed consent, subjects unable to give the consent, persons who have a mental disorder, subject’s rights Art. 12 Art. 6, 7, 17 Additional Protocol to the Oviedo Convention Risk/benefit, ethics committee, information and consent, privacy and confidentiality, vulnerability Art. 15 n.s. n.s. Art. 15 Recommendation Rec(2006)4 Biological samples handling, information and consent, privacy and confidentiality n.s. n.s. n.s. n.s. CIOMS-WHO guidelines 2002 Informed consent, clinical trial protocol, ethics committee, benefit/risk, vulnerability, privacy and confidentiality, secondary use of data Guideline 14 n.s. Guidelines 4, 5, 8, 18 Guidelines 4, 9, 13, 15 Directive 95/46/EC Data protection and confidentiality, informed consent, subjects unable to give the consent, subject’s rights (art. 1-8; 10-34) n.s. n.s. n.s. n.s.

ommittee, benefit/risk, vulnerability, privacy and confidentiality, secondary use of data Guideline 14 n.s. Guidelines 4, 5, 8, 18 Guidelines 4, 9, 13, 15 Directive 95/46/EC Data protection and confidentiality, informed consent, subjects unable to give the consent, subject’s rights (art. 1-8; 10-34) n.s. n.s. n.s. n.s. Directive 2001/83/EC Data protection and retention, confidentiality (art. 21, 5.2.c) n.s. n.s. n.s. n.s. Recommendation No. R (97) 5 Data protection and confidentiality, informed consent, subjects unable to give the consent, subject’s rights n.s. n.s. 4.7 12.2 Convention of 28 January 1981 Data protection and retention n.s. n.s. n.s. n.s. UNESCO Declaration on Human Genetic Data, 2003 Informed consent, subjects unable to give the consent, biological samples, ethics committee, data protection and confidentiality n.s. n.s. Whole document focused on genetic issues n.s. Recommendation No. R (92) 3 Informed consent, minors, persons suffering from mental disorders, data protection and handling, confidentiality Principle 5 n.s. Principle 5 ICH Topic E 11 Design and conduct of paediatric trials, assent, information, ethics committee Whole document focused on paediatric issues n.s. n.s. 2.6.3 Ethical Recommendations, 2008 Design and conduct of paediatric trials, assent, information, ethics committee, risk/benefit, data protection, disclosure of genetic findings 6.1, 9.1 9.1, 18 Paediatric Regulation (EC) 1901-1902/2006 Balance between the development of well-studied medicines and ethical concerns for enrolling young patients (Recital 7) n.s. n.s. n.s. n.s. not specified

Directive 2001/83/EC Data protection and retention, confidentiality (art. 21, 5.2.c) n.s. n.s. n.s. n.s. Recommendation No. R (97) 5 Data protection and confidentiality, informed consent, subjects unable to give the consent, subject’s rights n.s. n.s. 4.7 12.2 Convention of 28 January 1981 Data protection and retention n.s. n.s. n.s. n.s. UNESCO Declaration on Human Genetic Data, 2003 Informed consent, subjects unable to give the consent, biological samples, ethics committee, data protection and confidentiality n.s. n.s. Whole document focused on genetic issues n.s. Recommendation No. R (92) 3 Informed consent, minors, persons suffering from mental disorders, data protection and handling, confidentiality Principle 5 n.s. Principle 5 ICH Topic E 11 Design and conduct of paediatric trials, assent, information, ethics committee Whole document focused on paediatric issues n.s. n.s. 2.6.3 Ethical Recommendations, 2008 Design and conduct of paediatric trials, assent, information, ethics committee, risk/benefit, data protection, disclosure of genetic findings 6.1, 9.1 9.1, 18 Paediatric Regulation (EC) 1901-1902/2006 Balance between the development of well-studied medicines and ethical concerns for enrolling young patients (Recital 7) n.s. n.s. n.s. n.s. not specified Rare diseases are only cited in the new EU regulation [28] and EMA Reflection Paper [27], but no provision is established.

Directive 2001/83/EC Data protection and retention, confidentiality (art. 21, 5.2.c) n.s. n.s. n.s. n.s. Recommendation No. R (97) 5 Data protection and confidentiality, informed consent, subjects unable to give the consent, subject’s rights n.s. n.s. 4.7 12.2 Convention of 28 January 1981 Data protection and retention n.s. n.s. n.s. n.s. UNESCO Declaration on Human Genetic Data, 2003 Informed consent, subjects unable to give the consent, biological samples, ethics committee, data protection and confidentiality n.s. n.s. Whole document focused on genetic issues n.s. Recommendation No. R (92) 3 Informed consent, minors, persons suffering from mental disorders, data protection and handling, confidentiality Principle 5 n.s. Principle 5 ICH Topic E 11 Design and conduct of paediatric trials, assent, information, ethics committee Whole document focused on paediatric issues n.s. n.s. 2.6.3 Ethical Recommendations, 2008 Design and conduct of paediatric trials, assent, information, ethics committee, risk/benefit, data protection, disclosure of genetic findings 6.1, 9.1 9.1, 18 Paediatric Regulation (EC) 1901-1902/2006 Balance between the development of well-studied medicines and ethical concerns for enrolling young patients (Recital 7) n.s. n.s. n.s. n.s. not specified Rare diseases are only cited in the new EU regulation [28] and EMA Reflection Paper [27], but no provision is established. Among the other relevant international guidelines/documents and texts [9, 11, 12, 15, 16, 53], CIOMS-WHO guidelines [9] add specific details on the justification for enrolling ‘special populations’ like children and vulnerable patients, as well as on the informed consent and assent process. The recommendation is to seek a separate informed consent for the analysis of genetic material, if the genetic research is not a necessary part of the main clinical study. In addition, the use of results of genetic tests and familial genetic information and the precautions to prevent disclosure of the results of a subject’s genetic tests to immediate family relatives or to others are taken into account. The need for medical and psychological support for children and parents is also mentioned.

dition, the use of results of genetic tests and familial genetic information and the precautions to prevent disclosure of the results of a subject’s genetic tests to immediate family relatives or to others are taken into account. The need for medical and psychological support for children and parents is also mentioned. Further ethical provisions on data protection and confidentiality [10, 14, 29, 31] apply in the field of clinical research. However, no special requirements are available for issues taken into account in this work (i.e. paediatric and vulnerable patients, rare and genetic diseases). Council of Europe Committee of Ministers (CoE) recommendation on the protection of medical data [14] specifies rules to allow a clinical study even though personal data are not fully anonymised. Specific documents deal with genetic tests and PGt/PGx research [13, 52]. They recommend (and do not make mandatory) that ad hoc consent is to be sought for genetic tests, thus meaning that a consent separate from the main study is appropriate in the case of clinical trials with genetic sub-studies. The consent should include information on subsequent processing, use and storage and indicate that the subject has the right to decide whether or not to be informed of the results [52].

ests, thus meaning that a consent separate from the main study is appropriate in the case of clinical trials with genetic sub-studies. The consent should include information on subsequent processing, use and storage and indicate that the subject has the right to decide whether or not to be informed of the results [52]. Further guidance covering ethical aspects in the paediatric field can be found [21, 24, 32, 33]. In particular, ICH topic E11 [24] adds special restrictions when planning a paediatric clinical trial in ‘more vulnerable populations’ and handicapped or institutionalised paediatric populations. Rare diseases are mentioned in the EC Recommendations [21], as well as genetic issues, for which disclosure in clinical trials of genetic findings is recognised as a possible risk, thus requiring expert counselling in an adequate setting. The need for justifying alternative designs and/or analyses and agreeing these applications with competent authorities is mentioned. Table 1 summarises the provisions applicable in the iNMD field. The application of ethical provisions retrieved in the literature We searched the literature in order to find the possible existing information on the application of ethical rules in iNMD clinical research and the current debate on the ethical issues (Table 2).Table 2 Field of application, topics dealt and outcomes resulting from the bibliographic search

s retrieved in the literature We searched the literature in order to find the possible existing information on the application of ethical rules in iNMD clinical research and the current debate on the ethical issues (Table 2).Table 2 Field of application, topics dealt and outcomes resulting from the bibliographic search Source Fields Relevant topics Recommendations/suggestions Ross, 2010 [47] Newborn genetic screening - Use of stored samples for research - Parents’ informed consent - To seek for the consent from parents for research on stored sample storage and research Simopoulos, 2009 [50] Newborn genetic screening - Appropriateness of the investigation - To protect individuals identified by genetic screening against the psychological and social hazards Glantz et al., 2008 [38] Biobanks - Use of stored samples for research - Informed consent for secondary use - Owner of stored samples - To establish clear rules on the use of samples in medical research and genetic privacy when information is used by companies - To provide participants the right to withdraw the consent Botkin, 2005 [3] Newborn genetic screening - Use of stored samples for research - Appropriateness of the investigation - To apply an approach to evaluate screening tests as rigorous as the approach for drugs (phases I to IV) Sheela et al., 2005 [49] Trials/experimental uses - Parents’ informed consent - Vulnerable status - Approval from the ethics committee - To provide full and appropriate information to parents - To offer adequate time for the parents to absorb information Gelsinger, 2002 [35] Trials/experimental uses - Informed consent - Potential benefits, risks and discomforts - To provide full and appropriate information - To carefully review paediatric protocols Merz et al., 2002 [43] Genetic research - Use of stored samples for research - Parents’ informed consent - Rights of research participants and children - To address issues on the control of research results and the sharing of benefits before the research is performed - To involve research participants in decision-making - To develop policies for protecting the interests of subjects Clague, Thomas, 2002 [7] Newborn genetic screening - Secondary use of samples for research - Parents’ informed consent for secondary use - Right to privacy and anonymised data - Appropriateness of the investigation - To undertake long-term storage, to allow re-evaluation of apparently erroneous results - To perform anonymously research Pschera, 2000 [44] Trials/experime

Secondary use of samples for research - Parents’ informed consent for secondary use - Right to privacy and anonymised data - Appropriateness of the investigation - To undertake long-term storage, to allow re-evaluation of apparently erroneous results - To perform anonymously research Pschera, 2000 [44] Trials/experime ntal uses - Appropriateness of the investigation - To enhance the methodology for a safe and broad application of experimental approaches Fox,. 2000 [34] Trials/experimental uses - Informed consent for gene therapy trials - Disclosure of conflicts of interest - To provide full and appropriate information - To disclosure conflicts of interest During, 1996 [18] Trials/experimental uses - Informed consent - Expertise in ethics committees - To provide full and appropriate information - To foresee specific expertise in ethics committee on gene therapy The literature search produced 50 records. The abstract/full text of two of them were not available. Eleven resulted compliant, i.e. dealing with ethical issues of research involving iNMD patients [3, 7, 18, 34, 35, 38, 43, 44, 47, 49, 50]. As shown in Table 3, these publications are set in several fields.Table 3 Field of application, topics dealt and outcomes resulting from the European projects

m were not available. Eleven resulted compliant, i.e. dealing with ethical issues of research involving iNMD patients [3, 7, 18, 34, 35, 38, 43, 44, 47, 49, 50]. As shown in Table 3, these publications are set in several fields.Table 3 Field of application, topics dealt and outcomes resulting from the European projects Project Fields Relevant topics Outcomes BIONET [1] Genetic research - Ethical governance and regulation of research in China and EU No relevant public information COB [8] Genetic research - Cultural diversity and harmonisation of governance in Europe No relevant public information—recommendations for policy makers not publicly available Ethical, Legal and Social aspects of Brain Research [19] Neurological disorders - Ethical, legal and social aspects of brain research - Need to define ethical and legal standards specifically applicable to brain research InNerMeD-I-Network [40] Rare inherited, paediatric, neurological diseases - Ethical issues on clinical research, informed consent, vulnerable patients, regulatory, ethics committees - Recommendations for researchers, ethics committees and patients to conduct studies involving iNMD patients with the highest ethical standards LEUKOTREAT [41] Rare inherited, paediatric, neurological diseases - Informed consent - Ethical impacts of the proposed therapeutic challenges - Data sharing - Identification of patient’ expectations towards research - A charter on data sharing within the project database - Recommendations on data processing, participant information - Template of information sheet for patients and families RD-CONNECT [45] Rare diseases - Informed consent - Data protection - Data sharing - Registries - Biobanks - Guidelines and standards for informed consent procedures, including essential elements to be dealt with and templates of prospective, retrospective and broad consent - Recommendations for researchers on the informed consent with specific information on data sharing - Appropriate time to think and ask questions for participants - To consult patients/participant representatives on the quality of information - Guidelines for effective and ethical data and sample sharing [42] RESPECT [46] Paediatrics - Participation and empowerment of children in clinical trials, expectations and needs - Informed consent/assent - Main barriers to participation of children in clinical trials identified - Recommendations for investigators, sponsors, patient organisations and policy makers SCPE network [48] Neurological paedia

iatrics - Participation and empowerment of children in clinical trials, expectations and needs - Informed consent/assent - Main barriers to participation of children in clinical trials identified - Recommendations for investigators, sponsors, patient organisations and policy makers SCPE network [48] Neurological paedia tric diseases - Informed consent No relevant public information TEDDY [51] Paediatrics - Informed consent/assent process - Minimal risks/burden - Emergency situations - Placebo use - Confidentiality and privacy, right to information - Contents of documents to be submitted to ethics committee - Compensation for damage - To achieve agreed common definitions and procedures in the European Union with reference to genetic tests and studies - To adopt special measures, taking into account developmental needs of children at different ages and individual variations between children of the same age Five out of 11 publications were laid down in the context of trials/experimental uses: three of them were focused on gene therapy [7, 18, 34], one showed results from the experimental treatment of an extemporaneous formulation administered to a paediatric patient [35] and one discussed about in utero stem cell transplantation [44]. Four publications were laid down in the context of newborn genetic screening programs [3, 7, 47, 50]. Finally, one publication was about biobanks [38], and another one was related to genetic research [49]. More in detail, different topics were examined as follows.

Five out of 11 publications were laid down in the context of trials/experimental uses: three of them were focused on gene therapy [7, 18, 34], one showed results from the experimental treatment of an extemporaneous formulation administered to a paediatric patient [35] and one discussed about in utero stem cell transplantation [44]. Four publications were laid down in the context of newborn genetic screening programs [3, 7, 47, 50]. Finally, one publication was about biobanks [38], and another one was related to genetic research [49]. More in detail, different topics were examined as follows. Five publications dealt with the use of stored samples for research purposes [3, 7, 38, 43, 47]. Specifically, Glantz and colleagues considered the issue on the ownership of the stored samples as well as rights of (paediatric) participants [38]. Eight publications dealt with informed consent issues [7, 18, 34, 35, 38, 43, 47, 49]. Three discussed about the consent from parents/legal guardians in the case of paediatric participants [7, 47, 49]. None mentioned the assent and the involvement of minors in the informed consent/assent procedure (in the case of neonatal screening [7, 47], we did not expect to find any information on this). In addition, three publications explicitly considered the need for participant/legal guardian consent for the secondary use of data and samples [7, 38, 47]. The right to privacy and confidentiality was mentioned in three publications [7, 38, 50]. Three publications specifically discussed the interests and rights of children as research participants [7, 38, 43].

In addition, three publications explicitly considered the need for participant/legal guardian consent for the secondary use of data and samples [7, 38, 47]. The right to privacy and confidentiality was mentioned in three publications [7, 38, 50]. Three publications specifically discussed the interests and rights of children as research participants [7, 38, 43]. Finally, four publications argued about appropriateness of the experimental approach to be followed in order to protect individuals [3, 7, 44, 50]. Other topics were also discussed. Sheela and colleagues considered the need for the ethics committee’s approval [49]. During highlighted the need for specific expertise in ethics committee assessing gene therapy trials [18]. Ross claimed the need for trained paediatricians [47]. Fox examined the need for disclosing conflicts of interest [34]. Seven publications (7/11) were about paediatric patients [3, 7, 38, 43, 47, 49, 50], most of them dealing with research in newborn screenings [3, 7, 47, 50]. Five publications faced ethical aspects related to genetic/inherited features of the diseases affecting research participants [3, 7, 47, 49, 50]. Two publications took into account the rarity of conditions [7, 50]. Only one publication discussed about the vulnerable status of an iNMD patient due to the frequently neurodegenerative nature of disease and about the potential impacts on unaffected family members [49]. No publication faced the balance between the development of well-studied and appropriate medicines in this field and the ethical concerns for enrolling young patients.