Review of Burn Resuscitation: Is Plasmalyte® a Comparable Alternative to Ringer's Lactate?

Ringer's lactate has been the most widely used fluid for burn resuscitation for decades. Plasmalyte® (PL), a newer balanced crystalloid, is gaining popularity for use in the critically ill, including patients with burns. This popularity is partly due to the fact that PL theoretically offers a favorable metabolic profile, but may also be attributed to its relatively lower cost. Patients who are critically ill with large burns receive enormous volumes of fluids, especially during the resuscitation period. The choice of balanced crystalloid solution used is likely to have an impact on the metabolic status of patients and their overall outcomes. The choice of fluid for burn resuscitation has been one of the most researched topics in burn care and various types of fluids have been superseded based on research findings. This narrative review examines the evidence guiding fluid management in burns and explores the data supporting the use of balanced crystalloid solutions, in particular PL for burn resuscitation. Our literature search revealed only one study that focused on a direct comparison between PL and standard Ringer's Lactate for burn resuscitation. Based on the limited literature on the use of PL in burns, it is difficult to draw meaningful conclusions. Further research, into the suitability of PL for use in burns, is needed before formulary changes are instituted widely.

CXR-Net: A Multitask Deep Learning Network for Explainable and Accurate Diagnosis of COVID-19 Pneumonia From Chest X-Ray Images.

Accurate and rapid detection of COVID-19 pneumonia is crucial for optimal patient treatment. Chest X-Ray (CXR) is the first-line imaging technique for COVID-19 pneumonia diagnosis as it is fast, cheap and easily accessible. Currently, many deep learning (DL) models have been proposed to detect COVID-19 pneumonia from CXR images. Unfortunately, these deep classifiers lack the transparency in interpreting findings, which may limit their applications in clinical practice. The existing explanation methods produce either too noisy or imprecise results, and hence are unsuitable for diagnostic purposes. In this work, we propose a novel explainable CXR deep neural Network (CXR-Net) for accurate COVID-19 pneumonia detection with an enhanced pixel-level visual explanation using CXR images. An Encoder-Decoder-Encoder architecture is proposed, in which an extra encoder is added after the encoder-decoder structure to ensure the model can be trained on category samples. The method has been evaluated on real world CXR datasets from both public and private sources, including healthy, bacterial pneumonia, viral pneumonia and COVID-19 pneumonia cases. The results demonstrate that the proposed method can achieve a satisfactory accuracy and provide fine-resolution activation maps for visual explanation in the lung disease detection. Compared to current state-of-the-art visual explanation methods, the proposed method can provide more detailed, high-resolution, visual explanation for the classification results. It can be deployed in various computing environments, including cloud, CPU and GPU environments. It has a great potential to be used in clinical practice for COVID-19 pneumonia diagnosis.

Differential attainment and recruitment to Intensive Care Medicine Training in the UK, 2018-2020.

BACKGROUND: Differences exist among doctors in examination performance, clinical and academic career progression, and prevalence of performance assessment by professional regulatory bodies. Some of these differences have been reported in relation to individual characteristics. The purpose of this study is to establish whether any specific individual characteristics are associated with performance in selection for entry into specialty training in Intensive Care in the United Kingdom. METHODS: We evaluated data of 509 candidates from the national recruitment rounds of 2018/19 and 2019/20. The outcome evaluated was "success at interview". Variables reaching statistical significance at univariate logistic regression analysis were fed in the multivariable analysis to identify independent predictors of success, with additional exploratory analyses performed, where indicated. RESULTS: The candidates' median age was 31.5 (interquartile range, IQR 30-33.7) years, 324 (63.7%) were male, 256 (50.3%) not married/in civil partnership, 6 (1.2%) pregnant. The majority (316, 62.1%) were White British, 99 (19.5%) of Asian background, other ethnicities represented less than 20% of the sample. Of the 509 candidates, 155 (30.5%) were Atheist, 140 (27.5%) Christian; most were heterosexual (440, 86.4%); 432 (84.9%) reported no disability, while 4 (0.8%) had a minor and 1 (0.2%) had a major disability; 432 (84.9%) candidates held a UK medical degree; 77 (15.1%) a non-UK degree. At univariate logistic regression analysis (LRA) multiple factors were found to be associated with a lower likelihood of success, the strongest being an international medical graduate (IMG, holding a non-UK medical degree); others were increasing age, male gender, being married, Asian or mixed ethnicity, specific religious beliefs (Buddhism, Islam and Hinduism). After feeding all factors significant at univariate analysis, the only two retained as independent predictors at multivariable regression were Asian ethnicity and holding a non-UK degree. Asian UK graduates success rate was 92.7%, comparable to the national average of 92.3%, the Asian IMGs success rate was significantly lower, at 45.5%. CONCLUSIONS: As the imbalances seen within the candidates of Asian background are explained by considering the country of primary medical training, the variations in performance is likely to reflect differences in training systems and understanding of the UK NHS.

Epidemiology of Neonatal Acute Respiratory Distress Syndrome: Prospective, Multicenter, International Cohort Study.

OBJECTIVES: Age-specific definitions for acute respiratory distress syndrome (ARDS) are available, including a specific definition for neonates (the "Montreux definition"). The epidemiology of neonatal ARDS is unknown. The objective of this study was to describe the epidemiology, clinical course, treatment, and outcomes of neonatal ARDS. DESIGN: Prospective, international, observational, cohort study. SETTING: Fifteen academic neonatal ICUs. PATIENTS: Consecutive sample of neonates of any gestational age admitted to participating sites who met the neonatal ARDS Montreux definition criteria. MEASUREMENTS AND MAIN RESULTS: Neonatal ARDS was classified as direct or indirect, infectious or noninfectious, and perinatal (≤ 72 hr after birth) or late in onset. Primary outcomes were: 1) survival at 30 days from diagnosis, 2) inhospital survival, and 3) extracorporeal membrane oxygenation (ECMO)-free survival at 30 days from diagnosis. Secondary outcomes included respiratory complications and common neonatal extrapulmonary morbidities. A total of 239 neonates met criteria for the diagnosis of neonatal ARDS. The median prevalence was 1.5% of neonatal ICU admissions with male/female ratio of 1.5. Respiratory treatments were similar across gestational ages. Direct neonatal ARDS (51.5% of neonates) was more common in term neonates and the perinatal period. Indirect neonatal ARDS was often triggered by an infection and was more common in preterm neonates. Thirty-day, inhospital, and 30-day ECMO-free survival were 83.3%, 76.2%, and 79.5%, respectively. Direct neonatal ARDS was associated with better survival outcomes than indirect neonatal ARDS. Direct and noninfectious neonatal ARDS were associated with the poorest respiratory outcomes at 36 and 40 weeks' postmenstrual age. Gestational age was not associated with any primary outcome on multivariate analyses. CONCLUSIONS: Prevalence and survival of neonatal ARDS are similar to those of pediatric ARDS. The neonatal ARDS subtypes used in the current definition may be associated with distinct clinical outcomes and a different distribution for term and preterm neonates.

Electroanalytical point-of-care detection of gold standard and emerging cardiac biomarkers for stratification and monitoring in intensive care medicine - a review.

Determination of specific cardiac biomarkers (CBs) during the diagnosis and management of adverse cardiovascular events such as acute myocardial infarction (AMI) has become commonplace in emergency department (ED), cardiology and many other ward settings. Cardiac troponins (cTnT and cTnI) and natriuretic peptides (BNP and NT-pro-BNP) are the preferred biomarkers in clinical practice for the diagnostic workup of AMI, acute coronary syndrome (ACS) and other types of myocardial ischaemia and heart failure (HF), while the roles and possible clinical applications of several other potential biomarkers continue to be evaluated and are the subject of several comprehensive reviews. The requirement for rapid, repeated testing of a small number of CBs in ED and cardiology patients has led to the development of point-of-care (PoC) technology to circumvent the need for remote and lengthy testing procedures in the hospital pathology laboratories. Electroanalytical sensing platforms have the potential to meet these requirements. This review aims firstly to reflect on the potential benefits of rapid CB testing in critically ill patients, a very distinct cohort of patients with deranged baseline levels of CBs. We summarise their source and clinical relevance and are the first to report the required analytical ranges for such technology to be of value in this patient cohort. Secondly, we review the current electrochemical approaches, including its sub-variants such as photoelectrochemical and electrochemiluminescence, for the determination  of important CBs highlighting the various strategies used, namely the use of micro- and nanomaterials, to maximise the sensitivities and selectivities of such approaches. Finally, we consider the challenges that must be overcome to allow for the commercialisation of this technology and transition into intensive care medicine.

Perspectives From a Regional Plastic Surgery Centre on Evidence for the Purported Link Between SGLT2 Inhibitors and Fournier's Gangrene.

Introduction: The recent report issued by the MHRA indicating an association of Sodium glucose linked transporter type 2 (SGLT2) Inhibitors with the contraction of Fournier's Gangrene (FG), has been drawn with insufficient supporting evidence and without an adequately powered study to make any meaningful assertions or recommendations. We aimed to look specifically at the currently available dataset used to link SGLT2 Inhibitors to FG and highlight what conclusions or inferences can meaningfully be made, in particular the power of any study that would be required to make sensible conclusions. Methods: World literature review of SGLT2 Inhibitors and FG was performed. With a subsequent 10-year review of cases of FG seen in a regional burns and plastics centre. Data was collected retrospectively from the coding department at Whiston Hospital for all patients with necrotising fasciitis. An electronic document management system was used to identify patients with FG specifically as well as their diabetes state and medication history. Results: Seventy-eight patients were admitted with FG, of whom 32 had diabetes mellitus (DM). Of those with DM none was taking an SGLT2 Inhibitor, 17 patients were taking metformin, a further nine patients were taking a second line medication and 14 required insulin injections. Discussions: DM is a known major risk factor for FG, which is clearly observed in our patient cohort. The risk of patients with DM developing FG is irrespective of the medication patients are taking. The current articles and reports published have little ground to claim an association between SGLT2 Inhibitors and FG and are missing the crucial message that needs to be conveyed to the public: that DM is a major risk factor for FG and patients suffering with diabetes need to be extra vigilant.

Toward the Rapid Diagnosis of Sepsis: Detecting Interleukin-6 in Blood Plasma Using Functionalized Screen-Printed Electrodes with a Thermal Detection Methodology.

This paper reports the detection of the inflammatory and sepsis-related biomarker, interleukin-6 (IL-6), in human blood plasma using functionalized screen-printed electrodes (SPEs) in conjunction with a thermal detection methodology, termed heat-transfer method (HTM). SPEs are functionalized with antibodies specific for IL-6 through electrodeposition of a diazonium linking group and N'-ethylcarbodiimide hydrochloride (EDC) coupling, which was tracked through the use of cyclic voltammetry and Raman spectroscopy. The functionalized SPEs are mounted inside an additively manufactured flow cell and connected to the HTM device. We demonstrate the ability to detect IL-6 at clinically relevant concentrations in PBS buffer (pH = 7.4) with no significant interference from the similarly sized sepsis-related biomarker procalcitonin (PCT). The limit of detection (3σ) of the system is calculated to correspond to 3.4 ± 0.2 pg mL-1 with a working range spanning the physiologically relevant concentration levels in both healthy individuals and patients with sepsis, indicating the sensitivity of the sensor is suitable for the application. Further experiments helped provide a proof-of-application through the detection of IL-6 in blood plasma with no significant interference observed from PCT or the constituents of the medium. Due to the selectivity, sensitivity, straightforward operation, and low cost of production, this sensor platform has the potential for use as a traffic light sensor for the multidetection of inflammatory biomarkers for the diagnosis of sepsis and other conditions in which the rapid testing of blood biomarkers has vital clinical application.

Evaluating the Possibility of Translating Technological Advances in Non-Invasive Continuous Lactate Monitoring into Critical Care.

Lactate is widely measured in critically ill patients as a robust indicator of patient deterioration and response to treatment. Plasma concentrations represent a balance between lactate production and clearance. Analysis has typically been performed with the aim of detecting tissue hypoxia. However, there is a diverse range of processes unrelated to increased anaerobic metabolism that result in the accumulation of lactate, complicating clinical interpretation. Further, lactate levels can change rapidly over short spaces of time, and even subtle changes can reflect a profound change in the patient's condition. Hence, there is a significant need for frequent lactate monitoring in critical care. Lactate monitoring is commonplace in sports performance monitoring, given the elevation of lactate during anaerobic exercise. The desire to continuously monitor lactate in athletes has led to the development of various technological approaches for non-invasive, continuous lactate measurements. This review aims firstly to reflect on the potential benefits of non-invasive continuous monitoring technology within the critical care setting. Secondly, we review the current devices used to measure lactate non-invasively outside of this setting and consider the challenges that must be overcome to allow for the translation of this technology into intensive care medicine. This review will be of interest to those developing continuous monitoring sensors, opening up a new field of research.

Immature platelet indices alongside procalcitonin for sensitive and specific identification of bacteremia in the intensive care unit.

Hematological markers that can be rapidly analyzed and regularly monitored during a patient's stay on ICU, and that can identify bacterial causes of sepsis are being extensively sought. The significance of platelets in early immunological responses provides justification for assessing their usefulness in the identification of bacteremia amongst sepsis patients. In this preliminary study, the full blood count, including the platelet count by impedance (PLT-I), Immature Platelet Fraction (IPF%) and absolute immature platelet count (AIPC), were analyzed in eighty-two sepsis patients daily over the first 5 days stay on ICU. C-Reactive Protein (CRP), procalcitonin (PCT), and lactate were also analyzed daily. Blood cultures confirmed or excluded the presence of bacteremia. PCT provided the earliest indicator of bacteremia, with significant differences between the two cohorts on day 1. The change in IPF% and AIPC from day 1 to day 2 (Δ IPF% and Δ AIPC) provided the most accurate indication; A combination of Δ IPF% and day 2 PCT, provided a positive predictive value and negative predictive value of 100% and 96.10%, respectively. These data provide strong justification for larger multi-center validation studies to confirm the usefulness of these platelet indices during the assessment of sepsis on the ICU.

Methodological challenges in European ethics approvals for a genetic epidemiology study in critically ill patients: the GenOSept experience.

BACKGROUND: During the set-up phase of an international study of genetic influences on outcomes from sepsis, we aimed to characterise potential differences in ethics approval processes and outcomes in participating European countries. METHODS: Between 2005 and 2007 of the FP6-funded international Genetics Of Sepsis and Septic Shock (GenOSept) project, we asked national coordinators to complete a structured survey of research ethic committee (REC) approval structures and processes in their countries, and linked these data to outcomes. Survey findings were reconfirmed or modified in 2017. RESULTS: Eighteen countries participated in the study, recruiting 2257 patients from 160 ICUs. National practices differed widely in terms of composition of RECs, procedures and duration of the ethics approval process. Eight (44.4%) countries used a single centralised process for approval, seven (38.9%) required approval by an ethics committee in each participating hospital, and three (16.7%) required both. Outcomes of the application process differed widely between countries because of differences in national legislation, and differed within countries because of interpretation of the ethics of conducting research in patients lacking capacity. The RECs in four countries had no lay representation. The median time from submission to final decision was 1.5 (interquartile range 1-7) months; in nine (50%) approval was received within 1 month; six took over 6 months, and in one 24 months; had all countries been able to match the most efficient approvals processes, an additional 74 months of country or institution-level recruitment would have been available. In three countries, rejection of the application by some local RECs resulted in loss of centres; and one country rejected the application outright. CONCLUSIONS: The potential benefits of the single application portal offered by the European Clinical Trials Regulation will not be realised without harmonisation of research ethics committee practices as well as national legislation.

Porcine vs bovine surfactant therapy for preterm neonates with RDS: systematic review with biological plausibility and pragmatic meta-analysis of respiratory outcomes.

BACKGROUND: Bovine surfactants are known to be clinically equivalent but it is unclear if porcine or bovine surfactants at their licensed dose should be preferred to treat respiratory distress syndrome in preterm neonates. METHODS: We performed a comprehensive review of biochemical and pharmacological features of surfactants to understand the biological plausibility of any clinical effect. We then performed a pragmatic meta-analysis comparing internationally marketed porcine and bovine surfactants for mortality and respiratory outcomes. Search for randomised controlled trials with no language/year restrictions and excluding "grey" literature, unpublished or non-peer reviewed reports was conducted, following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and the most recent methodological recommendations. RESULTS: Sixteen articles were included in the review and 14 in the meta-analysis (1491 neonates). 200 mg/kg poractant-α (a porcine surfactant) was associated with lower BPD/mortality (OR 0.632[95%CI:0.494, 0.809];p < 0.001),BPD (OR 0.688[95%CI:0.512, 0.925];p = 0.013), retreatment (OR 0.313[95%CI:0.187, 0.522];p < 0.0001), airleaks (OR 0.505[95%CI:0.308, 0.827];p = 0.006) and lung haemorrhage (OR 0.624[95%CI:0.388, 1];p = 0.051). Gestational age is associated with effect size for BPD (coefficient: 0.308 [95%CI:0.063, 0.554];p = 0.014) and surfactant retreatment (coefficient: -0.311 [95%CI:-0.595, - 0.028];p = 0.031). CONCLUSION: 200 mg/kg poractant-α is associated with better respiratory outcomes compared to bovine surfactants at their licensed dose. The effect of poractant-α on BPD and surfactant retreatment is greater at lowest and highest gestational ages, respectively. TRIAL REGISTRATION: PROSPERO n.42017075251 .

Derivation and validation of a prognostic model for postoperative risk stratification of critically ill patients with faecal peritonitis.

BACKGROUND: Prognostic scores and models of illness severity are useful both clinically and for research. The aim of this study was to develop two prognostic models for the prediction of long-term (6 months) and 28-day mortality of postoperative critically ill patients with faecal peritonitis (FP). METHODS: Patients admitted to intensive care units with faecal peritonitis and recruited to the European GenOSept study were divided into a derivation and a geographical validation subset; patients subsequently recruited to the UK GAinS study were used for temporal validation. Using all 50 clinical and laboratory variables available on day 1 of critical care admission, Cox proportional hazards regression was fitted to select variables for inclusion in two prognostic models, using stepwise selection and nonparametric bootstrapping sampling techniques. Using Area under the receiver operating characteristic curve (AuROC) analysis, the performance of the models was compared to SOFA and APACHE II. RESULTS: Five variables (age, SOFA score, lowest temperature, highest heart rate, haematocrit) were entered into the prognostic models. The discriminatory performance of the 6-month prognostic model yielded an AuROC 0.81 (95% CI 0.76-0.86), 0.73 (95% CI 0.69-0.78) and 0.76 (95% CI 0.69-0.83) for the derivation, geographic and temporal external validation cohorts, respectively. The 28-day prognostic tool yielded an AuROC 0.82 (95% CI 0.77-0.88), 0.75 (95% CI 0.69-0.80) and 0.79 (95% CI 0.71-0.87) for the same cohorts. These AuROCs appeared consistently superior to those obtained with the SOFA and APACHE II scores alone. CONCLUSIONS: The two prognostic models developed for 6-month and 28-day mortality prediction in critically ill septic patients with FP, in the postoperative phase, enhanced the day one SOFA score's predictive utility by adding a few key variables: age, lowest recorded temperature, highest recorded heart rate and haematocrit. External validation of their predictive capability in larger cohorts is needed, before introduction of the proposed scores into clinical practice to inform decision making and the design of clinical trials.

Relationship between transcutaneous bilirubin and circulating unbound bilirubin in jaundiced neonates.

BACKGROUND: Transcutaneous bilirubin (TcB) consists of the skin-deposited bilirubin. Free bilirubin represents the protein-unbound bilirubin (UB) that is able to pass into the tissues. We aimed to describe the relationship UB-TcB and study the passage of UB into the skin. METHODS: We prospectively enrolled 194 neonates and we measured TcB, UB, serum bilirubin and albumin. Multiple sites TcB measurement was performed, bilirubin-albumin equilibrium constant and plasma bilirubin avidity (PBA) were calculated. RESULT: TcB has a similar correlation with UB and TSB. There is a quadratic relationship between UB and TcB (R2=0.48; p<0.001), remaining significant (ß for UB2=-0.8; p<0.001. ß for UB=1.1; p<0.001) after adjustment for gestational age, birth weight, postnatal age and albumin (Adj-R2=0.72). UB contributes to the skin bilirubin deposition, as there are significant correlations between albumin and TcB (r=-0.202; p=0.01) and between PBA and ΔTcB (r=0.323; p=0.017). CONCLUSION: TcB assay does not seem to directly replace UB measurement. However, TcB and UB are linked by a quadratic relationship: UB contributes to the skin bilirubin deposition but it is not the only bilirubin species measured by transcutaneous bilirubinometry.

Association between trends in clinical variables and outcome in intensive care patients with faecal peritonitis: analysis of the GenOSept cohort.

INTRODUCTION: Patients admitted to intensive care following surgery for faecal peritonitis present particular challenges in terms of clinical management and risk assessment. Collaborating surgical and intensive care teams need shared perspectives on prognosis. We aimed to determine the relationship between dynamic assessment of trends in selected variables and outcomes. METHODS: We analysed trends in physiological and laboratory variables during the first week of intensive care unit (ICU) stay in 977 patients at 102 centres across 16 European countries. The primary outcome was 6-month mortality. Secondary endpoints were ICU, hospital and 28-day mortality. For each trend, Cox proportional hazards (PH) regression analyses, adjusted for age and sex, were performed for each endpoint. RESULTS: Trends over the first 7 days of the ICU stay independently associated with 6-month mortality were worsening thrombocytopaenia (mortality: hazard ratio (HR) = 1.02; 95% confidence interval (CI), 1.01 to 1.03; P < 0.001) and renal function (total daily urine output: HR =1.02; 95% CI, 1.01 to 1.03; P < 0.001; Sequential Organ Failure Assessment (SOFA) renal subscore: HR = 0.87; 95% CI, 0.75 to 0.99; P = 0.047), maximum bilirubin level (HR = 0.99; 95% CI, 0.99 to 0.99; P = 0.02) and Glasgow Coma Scale (GCS) SOFA subscore (HR = 0.81; 95% CI, 0.68 to 0.98; P = 0.028). Changes in renal function (total daily urine output and renal component of the SOFA score), GCS component of the SOFA score, total SOFA score and worsening thrombocytopaenia were also independently associated with secondary outcomes (ICU, hospital and 28-day mortality). We detected the same pattern when we analysed trends on days 2, 3 and 5. Dynamic trends in all other measured laboratory and physiological variables, and in radiological findings, changes in respiratory support, renal replacement therapy and inotrope and/or vasopressor requirements failed to be retained as independently associated with outcome in multivariate analysis. CONCLUSIONS: Only deterioration in renal function, thrombocytopaenia and SOFA score over the first 2, 3, 5 and 7 days of the ICU stay were consistently associated with mortality at all endpoints. These findings may help to inform clinical decision making in patients with this common cause of critical illness.

Patients with faecal peritonitis admitted to European intensive care units: an epidemiological survey of the GenOSept cohort.

INTRODUCTION: Faecal peritonitis (FP) is a common cause of sepsis and admission to the intensive care unit (ICU). The Genetics of Sepsis and Septic Shock in Europe (GenOSept) project is investigating the influence of genetic variation on the host response and outcomes in a large cohort of patients with sepsis admitted to ICUs across Europe. Here we report an epidemiological survey of the subset of patients with FP. OBJECTIVES: To define the clinical characteristics, outcomes and risk factors for mortality in patients with FP admitted to ICUs across Europe. METHODS: Data was extracted from electronic case report forms. Phenotypic data was recorded using a detailed, quality-assured clinical database. The primary outcome measure was 6-month mortality. Patients were followed for 6 months. Kaplan-Meier analysis was used to determine mortality rates. Cox proportional hazards regression analysis was employed to identify independent risk factors for mortality. RESULTS: Data for 977 FP patients admitted to 102 centres across 16 countries between 29 September 2005 and 5 January 2011 was extracted. The median age was 69.2 years (IQR 58.3-77.1), with a male preponderance (54.3%). The most common causes of FP were perforated diverticular disease (32.1%) and surgical anastomotic breakdown (31.1%). The ICU mortality rate at 28 days was 19.1%, increasing to 31.6% at 6 months. The cause of FP, pre-existing co-morbidities and time from estimated onset of symptoms to surgery did not impact on survival. The strongest independent risk factors associated with an increased rate of death at 6 months included age, higher APACHE II score, acute renal and cardiovascular dysfunction within 1 week of admission to ICU, hypothermia, lower haematocrit and bradycardia on day 1 of ICU stay. CONCLUSIONS: In this large cohort of patients admitted to European ICUs with FP the 6 month mortality was 31.6%. The most consistent predictors of mortality across all time points were increased age, development of acute renal dysfunction during the first week of admission, lower haematocrit and hypothermia on day 1 of ICU admission.

The use of the Berlin definition for acute respiratory distress syndrome during infancy and early childhood: multicenter evaluation and expert consensus.

PURPOSE: A new acute respiratory distress syndrome (ARDS) definition has been recently issued: the so-called Berlin definition (BD) has some characteristics that could make it suitable for pediatrics. The European Society for Pediatric Neonatal Intensive Care (ESPNIC) Respiratory Section started a project to evaluate BD validity in early childhood. A secondary aim was reaching a consensus on clinical tools (risk factors list and illustrative radiographs) to help the application of BD. METHODS: This was an international, multicenter, retrospective study enrolling 221 children [aged greater than 30 days and less than 18 months; median age 6 (range 2-13) months], admitted to seven European pediatric intensive care units (PICU) with acute lung injury (ALI) or ARDS diagnosed with the earlier definition. RESULTS: Patients were categorized according to the two definitions, as follows: ALI, 36; ARDS, 185 (for the American-European Consensus Conference (AECC) definition); mild, 36; moderate, 97; severe ARDS, 88 (for BD). Mortality (13.9 % for mild ARDS; 11.3 % for moderate ARDS; 25 % for severe ARDS, p = 0.04) and the composite outcome extracorporeal membrane oxygenation (ECMO)/mortality (13.9 % for mild ARDS; 11.3 % for moderate ARDS; 28.4 % for severe ARDS, p < 0.01) were different across the BD classes, whereas they were similar using the previous definition. Mortality [HR 2.7 (95 % CI 1.1-7.1)] and ECMO/mortality [HR 3 (95 % CI 1.1-7.9)] were increased only for the severe ARDS class and remained significant after adjustment for confounding factors. PICU stay was not different across severity classes, irrespective of the definition used. There was significant concordance between raters evaluating radiographs [ICC 0.6 (95 % CI 0.2-0.8)] and risk factors [ICC 0.92 (95 % CI 0.8-0.97)]. CONCLUSIONS: BD validity for children is similar to that already reported in adults and mainly due to the introduction of a "severe ARDS" category. We provided clinical tools to use BD for clinical practice, research, and health services planning in pediatric critical care.