A National Survey on Physician Trainee Participation in Pediatric Interfacility Transport Objectives: To ascertain the national experience regarding which physician trainees are allowed to participate in pediatric interfacility transports and what is considered adequate education and training for physician trainees prior to participating in the transport of children. Design: Self-administered electronic survey. Setting: Pediatric transport teams listed with the American Academy of Pediatrics Section on Transport Medicine. Subjects: Leaders of U.S. pediatric transport teams. Interventions: None. Measurements and Main Results: Forty-four of the 90 U.S. teams surveyed (49%) responded. Thirty-nine (89%) were university hospital-affiliated. Most programs (26/43, 60%) allowed trainees to participate in pediatric transport in some capacity. Mandatory transport rotations were reported for pediatric critical care (PICU) fellows (9/42, 21%), neonatology (neonatal ICU) fellows (6/42, 14%), pediatric emergency medicine fellows (4/41, 10%), emergency medicine residents (3/43, 7%), and pediatric residents (2/43, 5%). Fellow participation was reported by 19 of 28 programs (68%) with PICU fellowships, 12 of 25 programs (48%) with pediatric emergency medicine fellowships, and 10 of 34 programs (29%) with neonatal ICU fellowships. Transport programs with greater than or equal to 1,000 annual incoming transports were more likely to include PICU and pediatric emergency medicine fellows as providers (p = 0.04; 95% CI, 1.04–25.71 and p = 0.02; 95% CI, 1.31–53.75). Most commonly, trainees functioned as medical control physicians (86%), provided minute-to-minute medical direction for critically ill patients (62%), performed intubations (52%), and were code leaders for patients undergoing cardiopulmonary resuscitation during transport (52%). Most transport programs required pediatric residents, PICU, and pediatric emergency medicine fellows to complete a PICU rotation prior to participating in pediatric transports. The majority of transport programs did not use any metrics to determine airway proficiency of physician trainees. Conclusions: There is heterogeneity with regard to the types of physician trainees allowed to participate in pediatric interfacility transports, the roles played by physician trainees during pediatric transport, and the training (or lack thereof) provided to physician trainees prior to their participating in pediatric transports. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/pccmjournal). The authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: eherrup@gmail.com ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Practices, Perceptions, and Attitudes in the Evaluation of Critically Ill Children for Bacteremia: A National Survey Objectives: Sending blood cultures in children at low risk of bacteremia can contribute to a cascade of unnecessary antibiotic exposure, adverse effects, and increased costs. We aimed to describe practice variation, clinician beliefs, and attitudes about blood culture testing in critically ill children. Design: Cross-sectional electronic survey. Setting: Fifteen PICUs enrolled in the Blood Culture Improvement Guidelines and Diagnostic Stewardship for Antibiotic Reduction in Critically Ill Children collaborative, an investigation of blood culture use in critically ill children in the United States. Subjects: PICU clinicians (bedside nurses, resident physicians, fellow physicians, nurse practitioners, physician assistants, and attending physicians). Interventions: None. Measurement and Main Results: Survey items explored typical blood culture practices, attitudes and beliefs about cultures, and potential barriers to changing culture use in a PICU setting. Fifteen of 15 sites participated, with 347 total responses, 15–45 responses per site, and an overall median response rate of 57%. We summarized median proportions and interquartile ranges of respondents who reported certain practices or beliefs: 86% (73–91%) report that cultures are ordered reflexively; 71% (61–77%) do not examine patients before ordering cultures; 90% (86–94%) obtain cultures for any new fever in PICU patients; 33% (19–61%) do not obtain peripheral cultures when an indwelling catheter is in place; and 64% (36–81%) sample multiple (vs single) lumens of central venous catheters for new fever. When asked about barriers to reducing unnecessary cultures, 80% (73–90%) noted fear of missing sepsis. Certain practices (culture source and indication) varied by clinician type. Obtaining surveillance cultures and routinely culturing all possible sources (each lumen of indwelling catheters and peripheral specimens) are positively correlated with baseline blood culture rates. Conclusions: There is variation in blood culture practices in the PICU. Fear and reflexive habits are common drivers of cultures. These practices may contribute to over-testing for bacteremia. Further investigation of how to optimize blood culture use is warranted. Dr. Milstone is a co-senior author. The Bright Star Authorship group members are listed in Appendix 1 (Supplemental Digital Content 1, http://links.lww.com/PCC/B132). Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/pccmjournal). Supported, in part, by grant from the Agency for Healthcare Research and Quality (grant 1R18HS025642). Drs. Woods-Hill’s and Colantuoni’s institution received funding from the National Institutes of Health (NIH) and the Agency for Healthcare Research and Quality (AHRQ). Dr. Woods-Hill received honorarium for giving grand rounds at St Jude Children’s Research Hospital, which was a discussion about the Bright Star project; she received support from a NIH Ruth L. Kirschstein National Research Service Award (T32HL098054-55). Dr. Miller received support for article research from the NIH. Drs. Koontz, Voskertchian, and Milstone received support for article research from the AHRQ. Drs. King’s, Milstone’s, and Xie’s institutions received funding from the AHRQ. The remaining authors have disclosed that they do not have any potential conflicts of interest. The coordinating center for the study was the Johns Hopkins Children’s Center in Baltimore, Maryland. For information regarding this article, E-mail: woodshillc@email.chop.edu ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Predictors of Reopening the Sternum in Children After Cardiac Surgery Objectives: Capillary leak syndrome can be severe in children after open-heart surgery which may hinder sternum closure and described as mediastinal tamponade. Reopening the sternum postoperatively may help maintaining hemodynamics and respiratory function. We looked for predictors that indicate the need for reopening the sternum. Design: A retrospective cohort study. Setting: A single cardiac center experience from 2009 to end of 2015. Patients: All children who required emergent reopening the sternum in the pediatric cardiac ICU after cardiac surgery were grouped as index cases and matched to a control group for age, body weight, cardiac diagnosis, and type of repair (single vs biventricular). Interventions: Emergent reopening the sternum. Measurements and Main Results: With a ratio of two control cases for each index case, variables related to cardiac output (predictors) were collected in a time line of 12, 6 hours, and just before reopening the sternum. Morbidities and mortality were also reviewed. Thirty-three index cases were compared with 63 control cases. Hospital stay and hospital-acquired infections were the same between the groups. Ventilation hours were longer in the index cases. Temperature gap more than 3°C, inotropic score more than 14, and acute kidney injury indicated by doubled blood urea nitrogen and creatinine were higher in the index group 6 hours before reopening the sternum. Mortality was more in the reopening sternum group with higher risk when extracorporeal membrane oxygenation was needed. Conclusions: Low cardiac output after cardiac surgery in children in form of temperature gap more than 3°C, inotropic score more than 14, and acute kidney injury may predict the need of reopening the sternum. Rate of mortality was higher in the reopening sternum group when extracorporeal membrane oxygenation was needed. The authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: shaathg@gmail.com ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Achieving Sustainability in Reducing Unplanned Extubations in a Pediatric Cardiac ICU Objectives: To determine the incidence of unplanned extubations in a pediatric cardiac ICU in order to prove sustainability of our previously implemented quality improvement initiative. Additionally, we sought to identify risk factors associated with unplanned extubations as well as review the overall outcome of this patient population. Design: Retrospective chart review. Setting: Pediatric cardiac ICU at Children’s Hospital of Colorado on the Anschutz Medical Center of the University of Colorado. Patients: Intubated and mechanically ventilated patients in the cardiac ICU from July 2011 to December 2017. Interventions: None. Measurements and Main Results: A total of 2,612 hospitalizations for 2,067 patients were supported with mechanical ventilation. Forty-five patients had 49 episodes of unplanned extubations (four patients > 1 unplanned extubation). The average unplanned extubation rate per 100 ventilator days was 0.4. Patients who had an unplanned extubation were younger (0.09 vs 5.45 mo; p < 0.001), weighed less (unplanned extubation median weight of 3.0 kg [interquartile range, 2.5–4.5 kg] vs control median weight of 6.0 kg [interquartile range, 3.5–13.9 kg]) (p < 0.001), and had a longer length of mechanical ventilation (8 vs 2 d; p < 0.001). Patients who had an unplanned extubation were more likely to require cardiopulmonary resuscitation during their hospital stay (54% vs 18%; p < 0.001) and had a higher likelihood of in-hospital mortality (15% vs 7%; p = 0.001). There was a significant difference in surgical acuity as denoted by The Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery score and patients with an unplanned extubation had a higher Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery category (p = 0.019). Contributing factors associated with unplanned extubation were poor endotracheal tube tape integrity, inadequate tube securement, and/or inadequate sedation. A low rate of unplanned extubation was maintained even in the setting of increasing patient complexity and an increase in patient volume. Conclusions: A low rate of unplanned extubation is sustainable even in the setting of increased patient volume and acuity. Additionally, early identification of patients at higher risk of unplanned extubation may also contribute to decreasing the incidence of unplanned extubation. The authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: nina.censoplano@childrenscolorado.org This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal. ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Prevalence and Outcomes of Pediatric In-Hospital Cardiac Arrest Associated With Pulmonary Hypertension Objectives: In adult in-hospital cardiac arrest, pulmonary hypertension is associated with worse outcomes, but pulmonary hypertension–associated in-hospital cardiac arrest has not been well studied in children. The objective of this study was to determine the prevalence of pulmonary hypertension among children with in-hospital cardiac arrest and its impact on outcomes. Design: Retrospective single-center cohort study. Setting: PICU of a quaternary care, academic children’s hospital. Patients: Children (<18 yr old) receiving greater than or equal to 1 minute of cardiopulmonary resuscitation (cardiopulmonary resuscitation) for an index in-hospital cardiac arrest with an echocardiogram in the 48 hours preceding in-hospital cardiac arrest, excluding those with cyanotic congenital heart disease. Interventions: None. Measurements and Main Results: Of 284 in-hospital cardiac arrest subjects, 57 (20%) had evaluable echocardiograms, which were analyzed by a cardiologist blinded to patient characteristics. Pulmonary hypertension was present in 20 of 57 (35%); nine of 20 (45%) had no prior pulmonary hypertension history. Children with pulmonary hypertension had worse right ventricular systolic function, measured by fractional area change (p = 0.005) and right ventricular global longitudinal strain (p = 0.046); more right ventricular dilation (p = 0.010); and better left ventricular systolic function (p = 0.001). Children with pulmonary hypertension were more likely to have abnormal baseline functional status and a history of chronic lung disease or acyanotic congenital heart disease and less likely to have sepsis or acute kidney injury. Children with pulmonary hypertension were more likely to have an initial rhythm of pulseless electrical activity or asystole and were more frequently treated with inhaled nitric oxide (80% vs 32%; p < 0.001) at the time of cardiopulmonary resuscitation. On multivariable analysis, pulmonary hypertension was not associated with event survival (14/20 [70%] vs 24/37 [65%]; adjusted odds ratio, 1.30 [CI95, 0.25–6.69]; p = 0.77) or survival to discharge (8/20 [40%] vs 10/37 [27%]; adjusted odds ratio, 1.17 [CI95, 0.22–6.44]; p = 0.85). Conclusions: Pulmonary hypertension physiology preceding pediatric in-hospital cardiac arrest may be more common than previously described. Among this cohort with a high frequency of inhaled nitric oxide treatment during cardiopulmonary resuscitation, pulmonary hypertension was not associated with survival outcomes. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/pccmjournal). Supported, in part, by research funds from the Department of Anesthesiology and Critical Care Medicine at the Children’s Hospital of Philadelphia. Dr. Morgan disclosed that this work was internally funded by the Department of Anesthesiology and Critical Care Medicine at the Children’s Hospital of Philadelphia. His institution received funding from National Institutes of Health (NIH) National Institute of Child Health and Human Development (NICHD) and the NIH National Heart, Lung, and Blood Institute (NHLBI), and he is a member of the American Heart Association (AHA) Emergency Cardiovascular Care Committee. Dr. Kilbaugh’s institution received funding from the Department of Defense, NHLBI, National Institute of Neurological Disorders and Stroke, and Mallinckrodt Pharmaceuticals. Dr. McGowan’s institution received funding from Merck and Transonic Systems, and he received funding from Merck. Dr. Berg’s institution received support from the NICHD and NHLBI. Dr. Sutton’s institution received funding from the NICHD, NHLBI, and Mallinckrodt Pharmaceuticals; he received funding from Zoll Medical (speaking honoraria); he was a member of the 2015 and 2018 AHA Pediatric Advanced Life Support writing group; he is the Chair of the AHA’s Get with the Guidelines-Resuscitation Registry Pediatric Research Task Force; and he disclosed that he is a member of the following subcommittees of the AHA Emergency Cardiovascular Care Committee: Systems of Care, Pediatric Emphasis Group, and Science Review Committee. Dr. Himebauch received funding from the Society of Critical Care Medicine (payment and travel expenses for teaching pediatric bedside ultrasound courses). The remaining authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: morganr1@email.chop.edu ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Changes in Provider Perceptions Regarding Early Mobility in the PICU Objectives: Early mobility in the PICU is safe and feasible. However, PICUs continue to meet barriers to implementing early mobility. PICU providers were surveyed before and after initiating an early mobility protocol to determine perceived barriers and continued challenges in performing early mobility. Design: This single-center prospective study surveyed PICU providers regarding 26 potential barriers to early mobility using a five-point Likert scale. A survey was distributed 1 month prior to and 6 months after beginning an early mobility protocol. Setting: Free-standing academic tertiary care children’s hospital. Subjects: PICU providers of various professions. Interventions: Implementation of PICU-wide early mobility protocol. Measurements and Main Results: Paired pre- and post-early mobility protocol implementation surveys from 97 providers were compared. System-based barriers decreased after implementation of the early mobility protocol, such as lack of guidelines (75–20%; p < 0.01), inadequate training (74–33%; p < 0.01), lack of early mobility orders (72–30%; p < 0.01), and delayed recognition of early mobility candidates (68–35%; p < 0.01). Difficulty coordinating early mobility sessions, although significantly decreased, still remained a concern for 66% of providers in the postsurvey. Lack of resources, specifically staff (85–82%; p = 0.68) and equipment (67–60%; p = 0.36), also remained significant barriers. Presence of an endotracheal tube was a barrier for only 29% of providers’ post-early mobility protocol, compared with 69% prior (p < 0.01). Clinical instability remained a top concern (82–79%; p = 0.63) as well as agitation (74–67%; p = 0.23). Day shift providers, with more early mobility exposure, perceived fewer barriers compared with night shift providers. Ninety percentage of post-early mobility survey participants felt that early mobility positively impacted their patients. Conclusions: Implementation of an early mobility protocol significantly changed provider perceptions regarding barriers to early mobility. Certain factors, such as staff availability, coordination difficulty, equipment shortage, and patient clinical factors, continue to be significant challenges to early mobility in the PICU population. This work was performed at the Monroe Carell Jr. Children’s Hospital at Vanderbilt. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/pccmjournal). Vanderbilt Institute of Translational and Clinical Research provided a $2,000 grant used to purchase Early Mobility equipment for this study. Ms. Hanna received funding from Society of Critical Care Medicine (conference fees). Ms. Hanna and Dr. Betters received funding from Vanderbilt Institute of Translational and Clinical Research (for Early Mobility equipment only). The remaining authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: kristina.betters@vanderbilt.edu ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

The Prognostic Value of Early Amplitude-Integrated Electroencephalography Monitoring After Pediatric Cardiac Arrest Objectives: To assess the ability of amplitude-integrated electroencephalography monitoring within 24 hours of the return of spontaneous circulation to prognosticate neurologic outcomes in children following cardiac arrest Design: Retrospective review of prospectively recorded data. An amplitude-integrated electroencephalography background score was calculated according to background activity during the first 24 hours after return of spontaneous circulation, a higher score correlating with more impaired background activity. The primary endpoint was the neurologic outcome as defined by the Pediatric Cerebral Performance Category at PICU discharge (Pediatric Cerebral Performance Category 1–3: a good neurologic outcome; Pediatric Cerebral Performance Category 4–6: a poor neurologic outcome). Setting: A referral PICU. Patients: Thirty children with a median age of 10 months (2–38 mo) and a male/female sex ratio of 1.3 were included. Interventions: None. Measurements and Main Results: Eighteen patients were assigned to the favorable outcome group and 12 to the unfavorable outcome group. The median time between return of spontaneous circulation and amplitude-integrated electroencephalography initiation was 4 hours (3–9 hr). The amplitude-integrated electroencephalography score within 24 hours after return of spontaneous circulation was significantly higher in the children with poor outcomes compared with those with good outcomes (12 ± 4 vs 25 ± 8; p < 0.001). Background activity during amplitude-integrated electroencephalography monitoring was able to predict poor neurologic outcomes at PICU discharge, with an area under the receiver operating characteristic curve of 0.91 (95% CI, 0.81–1.00). Conclusions: Early amplitude-integrated electroencephalography monitoring may help predict poor neurologic outcomes in children within 24 hours following cardiac arrest. The authors have disclosed that they do not have any potential conflict of interest. This study was conducted at Nantes University Hospital, France. For information regarding this article, E-mail: Pierre.bourgoin@chu-nantes.fr ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Oxygenator Impact on Ceftolozane and Tazobactam in Extracorporeal Membrane Oxygenation Circuits Objectives: To determine the oxygenator impact on alterations of ceftolozane/tazobactam in a contemporary neonatal/pediatric (1/4-inch) and adolescent/adult (3/8-inch) extracorporeal membrane oxygenation circuit including the Quadrox-i oxygenator (Maquet, Wayne, NJ). Design: A 1/4-inch and 3/8-inch, simulated closed-loop extracorporeal membrane oxygenation circuits were prepared with a Quadrox-i pediatric and Quadrox-i adult oxygenator and blood primed. Additionally, 1/4-inch and 3/8-inch circuits were also prepared without an oxygenator in series. A one-time dose of ceftolozane/tazobactam was administered into the circuits and serial preoxygenator and postoxygenator concentrations were obtained at 5 minutes, 1, 2, 3, 4, 5, 6, and 24-hour time points. Ceftolozane/tazobactam was also maintained in a glass vial and samples were taken from the vial at the same time periods for control purposes to assess for spontaneous drug degradation Setting: A free-standing extracorporeal membrane oxygenation circuit. Patients: None. Interventions: Single-dose administration of ceftolozane/tazobactam into closed-loop extracorporeal membrane oxygenation circuits prepared with and without an oxygenator in series with serial preoxygenator, postoxygenator, and reference samples obtained for concentration determination over a 24-hour study period. Measurements and Main Results: For the 1/4-inch circuit, there was approximately 92% ceftolozane and 22–25% tazobactam loss with the oxygenator in series and 19–30% ceftolozane and 31–34% tazobactam loss without an oxygenator in series at 24 hours. For the 3/8-inch circuit, there was approximately 85% ceftolozane and 29% tazobactam loss with the oxygenator in series and 25–27% ceftolozane and 23–26% tazobactam loss without an oxygenator in series at 24 hours. The reference ceftolozane and tazobactam concentrations remained relatively constant during the entire study period demonstrating the drug loss in each size of the extracorporeal membrane oxygenation circuit with or without an oxygenator was not a result of spontaneous drug degradation. Conclusions: This ex vivo investigation demonstrated substantial ceftolozane loss within an extracorporeal membrane oxygenation circuit with an oxygenator in series with both sizes of the Quadrox-i oxygenator at 24 hours and significant ceftolozane loss in the absence of an oxygenator. Tazobactam loss was similar regardless of the presence of an oxygenator. Further evaluations with multiple dose in vitro and in vivo investigations are needed before specific drug dosing recommendations can be made for clinical application with extracorporeal membrane oxygenation. Presented, in part, at the abstract presentation of the 2018 Society of Critical Care Medicine Annual Congress, February 25-28, 2018, San Antonio, TX. Supported, in part, by grant from Merck. Dr. Cies received funding from Merck (who supplied the drug for this study) and Atlantic Diagnostic Laboratories (consultant). He also received funding from Allergan, Melinta, and Thermo Fisher as a consultant (unrelated to this work).The remaining authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: jeffrey.cies@gmail.com ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Effect of Endotracheal Tube Size, Respiratory System Mechanics, and Ventilator Settings on Driving Pressure Objectives: We sought to investigate factors that affect the difference between the peak inspiratory pressure measured at the Y-piece under dynamic flow conditions and plateau pressure measured under zero-flow conditions (resistive pressure) during pressure controlled ventilation across a range of endotracheal tube sizes, respiratory mechanics, and ventilator settings. Design: In vitro study. Setting: Research laboratory. Patients: None. Interventions: An in vitro bench model of the intubated respiratory system during pressure controlled ventilation was used to obtain the difference between peak inspiratory pressure measured at the Y-piece under dynamic flow conditions and plateau pressure measured under zero-flow conditions across a range of endotracheal tubes sizes (3.0–8.0 mm). Measurements were taken at combinations of pressure above positive end-expiratory pressure (10, 15, and 20 cm H2O), airway resistance (no, low, high), respiratory system compliance (ranging from normal to extremely severe), and inspiratory time at constant positive end-expiratory pressure (5 cm H2O). Multiple regression analysis was used to construct models predicting resistive pressure stratified by endotracheal tube size. Measurements and Main Results: On univariate regression analysis, respiratory system compliance (β –1.5; 95% CI, –1.7 to –1.4; p < 0.001), respiratory system resistance (β 1.7; 95% CI, 1.5–2.0; p < 0.001), pressure above positive end-expiratory pressure (β 1.7; 95% CI, 1.4–2.0; p < 0.001), and inspiratory time (β –0.7; 95% CI, –1.0 to –0.4; p < 0.001) were associated with resistive pressure. Multiple linear regression analysis showed the independent association between increasing respiratory system compliance, increasing airway resistance, increasing pressure above positive end-expiratory pressure, and decreasing inspiratory time and resistive pressure across all endotracheal tube sizes. Inspiratory time was the strongest variable associated with a proportional increase in resistive pressure. The contribution of airway resistance became more prominent with increasing endotracheal tube size. Conclusions: Peak inspiratory pressures measured during pressure controlled ventilation overestimated plateau pressure irrespective of endotracheal tube size, especially with decreased inspiratory time or increased airway resistance. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/pccmjournal). The authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: m.c.j.kneyber@umcg.nl ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Nosocomial Infections During Extracorporeal Membrane Oxygenation in Neonatal, Pediatric, and Adult Patients: A Comprehensive Narrative Review Objectives: Extracorporeal membrane oxygenation is increasingly used in critically ill patients with refractory cardiopulmonary failure. Nosocomial infection acquired during extracorporeal membrane oxygenation represents one of the most frequent complications but the available evidence on the risk of infection and its association with outcomes has not been comprehensively analyzed. We performed a narrative review examining the epidemiology of nosocomial infection during extracorporeal membrane oxygenation, association with clinical outcomes, and preventive strategies. Data Sources: We searched PubMed, Web of Science, EMBASE, and the Cochrane Library between 1972 and June 2018. Study Selection: We included any article which detailed nosocomial infection during extracorporeal membrane oxygenation. Articles were excluded if they were not written in English, detailed extracorporeal membrane oxygenation use for infections acquired prior to extracorporeal membrane oxygenation, or used other forms of extracorporeal support such as ventricular assist devices. Data Extraction: Two reviewers independently assessed eligibility and extracted data. We screened 984 abstracts and included 59 articles in the final review. Data Synthesis: The reported risk of nosocomial infection among patients receiving extracorporeal membrane oxygenation ranged from 3.5% to 64% per extracorporeal membrane oxygenation run, while the incidence of infection ranged from 10.1 to 116.2/1,000 extracorporeal membrane oxygenation days. Nosocomial infections during extracorporeal membrane oxygenation were consistently associated with longer duration of extracorporeal membrane oxygenation and, in several large multicenter studies, with increased mortality. Risk factors for nosocomial infection included duration of extracorporeal membrane oxygenation, mechanical and hemorrhagic complications on extracorporeal membrane oxygenation, and use of venoarterial and central extracorporeal membrane oxygenation. Biomarkers had low specificity for infection in this population. Few studies examined strategies on how to prevent nosocomial infection on extracorporeal membrane oxygenation. Conclusions: Nosocomial infections in extracorporeal membrane oxygenation patients are common and associated with worse outcomes. There is substantial variation in the rates of reported infection, and thus, it is possible that some may be preventable. The evidence for current diagnostic, preventive, and therapeutic strategies for infection during extracorporeal membrane oxygenation is limited and requires further investigation. Supplemental digital content is available for this article. Direct URL citationsappear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/pccmjournal). The authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: gmaclaren@iinet.net.au ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies