Non-thyroidal illness syndrome and its relationship with mortality risk in critically ill children.

Introduction: Non-thyroidal illness syndrome (NTIS) is considered to be associated with adverse outcomes in critically ill children.The hypothesis that thyroid hormones and inflammatory markers are associated with increased prediction of mortality risk scores is tested in this paper. Methods: A prospective observational study was set up in a pediatric intensive care unit (PICU). One hundred and three patients were included. NTIS was defined as a low free triiodothyronine (FT3) value for the patient's age. Thyroid hormones levels and inflammatory markers were determined at admission: FT3, FT4 (free thyroxine), TSH (thyroid-stimulating hormone), rT3 (reverse triiodothyronine), CRP (C-reactive protein) and PCT (Procalcitonin). They were compared between children with a pediatric risk of mortality score PRISM-III >75th percentile (group A, n= 25) and the rest (group B, n = 78). Results: A FT4 value lower than 16.6 pmol/L showed an area under the curve (AUC) of 0.655 (0.56-0.78, p = 0.02), with 76% sensitivity and 61.5% specificity to detect a high risk of mortality. A multiple regression analysis revealed that a FT4 lower than 16.6 pmol/L [OR: 4.92 (1.60-18.19), p = 0.009] and having NTIS [OR: 6.04 (1.45-27.93), p = 0.016] could predict a high risk of mortality. Conclusions: In unselected critically ill children, FT4 and FT3 values at admission could be used as a good predictor of a high mortality risk. We have not achieved a predictive model that combines hormones with inflammatory markers.

Non-dipping blood pressure pattern in pediatricians during on-duty.

INTRODUCTION: People with a reduced nighttime dip in blood pressure have an increased cardiovascular risk. Our objective was to describe the different patterns in blood pressure (BP) among pediatricians who work in long on-duty shifts in relation with sex, medical rank and sleeping time. METHODS: Descriptive, cross-sectional, two-center study. On duty pediatric Resident physicians and pediatric Consultants were recruited between January 2018 and December 2021. RESULTS: Fifty-one physicians were included in the study (78.4% female, 66.7% Resident physicians). Resident physicians had a higher night/day ratio (0.91 vs 0.85; p<0.001) and a shorter nighttime period (3.87 vs 5.41, p<0.001) than Consultants. Physicians sleeping less than 5h had a higher night/day ratio (0.91 vs 0.87, p=0.014). Being a Resident showed a ∼4.5-fold increased risk of having a non-dipping BP pattern compared to Consultants. CONCLUSION: We found a potential link between both being a Resident and, probably, having shorter sleeping time, and the non-dipping BP pattern in physicians during prolonged shifts.

Predicting the Risk of Mortality in Children using a Fuzzy-Probabilistic Hybrid Model.

Introduction: The mortality risk in children admitted to Pediatric Intensive Care Units (PICU) is usually estimated by means of validated scales, which only include objective data among their items. Human perceptions may also add relevant information to prognosticate the risk of death, and the tool to use this subjective data is fuzzy logic. The objective of our study was to develop a mathematical model to predict mortality risk based on the subjective perception of PICU staff and to evaluate its accuracy compared to validated scales. Methods: A prospective observational study in two PICUs (one in Spain and another in Latvia) was performed. Children were consecutively included regardless of the cause of admission along a two-year period. A fuzzy set program was developed for the PICU staff to record the subjective assessment of the patients' mortality risk expressed through a short range and a long range, both between 0% and 100%. Pediatric Index of Mortality 2 (PIM2) and Therapeutic Intervention Scoring System 28 (TISS28) were also prospectively calculated for each patient. Subjective and objective predictions were compared using the logistic regression analysis. To assess the prognostication ability of the models a stratified B-random K-fold cross-validation was performed. Results: Five hundred ninety-nine patients were included, 308 in Spain (293 survivors, 15 nonsurvivors) and 291 in Latvia (282 survivors, 9 nonsurvivors). The best logistic classification model for subjective information was the one based on MID (midpoint of the short range), whereas objective information was the one based on PIM2. Mortality estimation performance was 86.3% for PIM2, 92.6% for MID, and the combination of MID and PIM2 reached 96.4%. Conclusions: Subjective assessment was as useful as validated scales to estimate the risk of mortality. A hybrid model including fuzzy information and probabilistic scales (PIM2) seems to increase the accuracy of prognosticating mortality in PICU.

SARS-CoV-2 Spike Protein in Intestinal Cells of a Patient with Coronavirus Disease 2019 Multisystem Inflammatory Syndrome.

A previously healthy 12-year-old boy had severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-related multisystem inflammatory syndrome (MIS-C) that was rapidly fatal. Autopsy revealed the presence of a large intracardiac thrombus. SARS-CoV-2 spike protein was detected in intestinal cells, supporting the hypothesis that viral presence in the gut may be related to the immunologic response of MIS-C.

Effectiveness of steroids versus placebo in preventing upper airway obstruction after extubation in critically ill children: rationale and design of a multicentric, double-blind, randomized study.

BACKGROUND: Post-extubation upper airway obstruction (UAO) is a frequent complication causing stridor and respiratory distress, which occasionally require reintubation, thereby increasing morbidity and mortality rates. Contradictory results have been obtained in studies assessing the effectiveness of steroids in preventing post-extubation UAO, and the available evidence is limited. We designed a multicentric randomized, placebo-controlled study to explore the effectiveness of dexamethasone in preventing post-extubation UAO in children. METHODS: A multicentric, prospective, double-blind, randomized, placebo-controlled, phase IV clinical trial has been designed. The sample will include pediatric patients who are between 1 month and 16 years of age and who have been intubated for more than 48 h. Patients who have airway disorders or who have received steroids within the previous seven days will be excluded. Patients will be randomly assigned to receive either placebo or a therapy with dexamethasone 0.25 mg/kg every 6 h to be started 6 to 12 h prior to extubation (to a total of four doses). Randomization will be performed at a 1:1 ratio. Follow-up of patients will be carried out for 48 h after extubation. The main objective of this study is to access the reduction in the incidence of moderate to severe UAO symptoms following extubation. Secondary objectives include assessing the decrease in the incidence of reintubation, evaluating the use of additional therapies for UAO, and monitoring potential side effects of dexamethasone. DISCUSSION: The results of this study will contribute to the existing evidence on prophylaxis for post-extubation airway obstruction. TRIAL REGISTRATION: EudraCT identifier: 2009-016596-30. Registered on May 11, 2010.

[Validation of a Therapeutic Intervention Scoring System (TISS-28) in critically ill children]. / Validación de la escala simplificada de puntuación de intervenciones terapéuticas (TISS-28) en niños críticamente enfermos.

INTRODUCTION: There are numerous scales in intensive care units that are used to quantify the severity of patients. Most of them are very useful, although sometimes laborious to complete, thus limiting their use in usual practice. One of these scales, the Therapeutic Intervention Scoring System (TISS 76), has been validated in adult and paediatric units. Its simplified and updated version, the Simplified Therapeutic Intervention Scoring System (TISS 28), has not yet been validated in paediatric units. The aim of this study is to validate TISS 28, in order to have a simple and rapid scale. MATERIAL AND METHOD: A prospective non-interventional observational study was conducted in a Paediatric Intensive Care Unit (PICU) of a university hospital. Data were collected from 935 consecutive patients admitted to the PICU over a 3-year period. These included the values of TISS 76 and TISS 28 during the first 4days of admission and the subsequent outcome of the patients. RESULTS: The mean values of TISS 76 and TISS 28 for the first day of admission were 18.27 and 18.02, respectively. Values were higher in patients who had sequelae or died (17.58 versus 27.23 and 37.44, respectively for TISS 76 (P<.01); and 17.51 versus 23.80 and 33.44, respectively for TISS 28 (P<.01). A very good correlation was found between TISS 76 and TISS 28, with Pearson correlation and intraclass correlation coefficients> 0.9 (except for the 2nd day). The correlation equation for the overall 4 days was: TISS76=- 1.74+1.05×TISS28. TISS 28 was able to explain 82.4% of variability of TISS 76. The area under the curve with a confidence interval (CI) of 95% for the first day was 0.80 (0.73-0.87) for TISS 76, and 0.76 (0.67-0.84) for TISS 28. CONCLUSIONS: On observing the results obtained, TISS 28 can replace TISS 26 in our PICU, without worsening the information provided. Being a reliable scale and easier to apply, its practical application could be useful.

Comparison of the effect of three different protein content enteral diets on serum levels of proteins, nitrogen balance, and energy expenditure in critically ill infants: study protocol for a randomized controlled trial.

BACKGROUND: Nutritional support is essential in the care of critically ill children since malnutrition in this population is associated with increased morbidity and mortality. Injury in patients admitted to pediatric intensive care units (PICU) results in a catabolic state and augmented protein breakdown, leading to a negative protein balance. Current recommendations about protein prescription in the PICU are fundamentally based on expert opinions, and the minimum threshold is 1.5 g/kg per day of protein, although protein needs could be higher in certain subgroups of patients. The main objectives of the present study are to examine whether the administration of a protein-enriched infant formula increases the serum levels of total proteins, albumin, prealbumin, transferrin, and retinol and improves nitrogen balance and to analyze the effect of the high-protein diet on energy expenditure. A secondary objective is to register possible secondary effects of the protein-enriched diet. METHODS: A multicenter prospective randomized controlled trial (RCT) will be performed in three hospitals. Patients meeting inclusion criteria will be randomly allocated to one of three enteral feeding formulae with different protein contents. Blood and urine test, nitrogen balance assessment, and energy expenditure testing by indirect calorimetry will be performed at the beginning of the nutrition regimen and at 24 h, 72 h and 5-7 days after initiation. The sample size for this trial is estimated to be 90 participants (about 30 participants in each group). The data analysis will be by intention to treat. DISCUSSION: This RCT will provide new data about the amount of protein needed to improve levels of serum protein and nitrogen balance, a surrogate of protein balance, in critically ill infants receiving enteral nutrition. TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT03901742 . Registered April 1, 2019 - Retrospectively registered.

Non-invasive ventilation practices in children across Europe.

OBJECTIVES: To describe the diversity in practice in non-invasive ventilation (NIV) in European pediatric intensive care units (PICUs). WORKING HYPOTHESIS: No information about the use of NIV in Pediatrics across Europe is currently available, and there might be a wide variability regarding the approach. STUDY DESIGN: Cross-sectional electronic survey. METHODOLOGY: The survey was distributed to the ESPNIC mailing list and to researchers in different European centers. RESULTS: One hundred one units from 23 countries participated. All respondent units used NIV. Almost all PICUs considered NIV as initial respiratory support (99.1%), after extubation (95.5% prophylactically, 99.1% therapeutically), and 77.5% as part of palliative care. Overall NIV use outside the PICUs was 15.5% on the ward, 20% in the emergency department, and 36.4% during transport. Regarding respiratory failure cause, NIV was delivered in pneumonia (97.3%), bronchiolitis (94.6%), bronchospasm (75.2%), acute pulmonary edema (84.1%), upper airway obstruction (76.1%), and in acute respiratory distress syndrome (91% if mild, 53.1% if moderate, and 5.3% if severe). NIV use in asthma was less frequent in Northern European units in comparison to Central and Southern European PICUs (P = 0.007). Only 47.7% of the participants had a written protocol about NIV use. Bilevel NIV was applied mostly through an oronasal mask (44.4%), and continuous positive airway pressure through nasal cannulae (39.8%). If bilevel NIV was required, 62.3% reported choosing pressure support (vs assisted pressure-controlled ventilation) in infants; and 74.5% in older children. CONCLUSIONS: The present study shows that NIV is a widespread technique in European PICUs. Practice across Europe is variable.

Helmet Versus Nasal-Prong CPAP in Infants With Acute Bronchiolitis.

BACKGROUND: Nasal prongs are frequently used to deliver noninvasive CPAP in bronchiolitis, especially in the youngest children. A helmet interface is an alternative that might be comparable to nasal prongs. We sought to compare these interfaces. METHODS: We performed a prospective, randomized, crossover, single-center study in an 8-bed multidisciplinary pediatric ICU in a university hospital. Infants age <3 months who were consecutively admitted to the pediatric ICU during a bronchiolitis epidemic season and fulfilled inclusion criteria were recruited. Subjects were randomly allocated to receive CPAP via a helmet or nasal prongs for 60 min. The subjects were then placed on the other CPAP system for another 60-min period (helmet then nasal prongs [H-NP] or nasal prongs then helmet [NP-H]). Measurements were taken at 30, 60, 90, and 120 min. Failure was defined as the need for further respiratory support. RESULTS: Sixteen subjects were included, with 9 in the H-NP group and 7 in the NP-H group. CPAP significantly reduced respiratory distress, showing no differences between the H-NP and NP-H groups in terms of improving the Modified Wood's Clinical Asthma Score from 4.8 ± 1 to 3 ± 0.9 and 2.7 ± 1.7 points at 60 min and 120 min in the H-NP group, respectively, and from 4.2 ± 0.9 to 2.8 ± 0.9 and to 2.9 ± 0.9 at 60 min and 120 min, respectively, in the NP-H group. Sedatives were used in only 3 subjects (2 in the NP-H group, P = .77). The failure rate was similar in both groups (3 of 9 subjects vs 3 of 7 subjects, P = .70). No significant differences were seen for heart rate, breathing frequency, FIO2 , or transcutaneous oxygen saturation response. CONCLUSIONS: Our results suggest that CPAP delivered by nasal prongs and CPAP delivered by helmet are similar in terms of efficacy in young infants with acute bronchiolitis.

Receiver operating characteristic curve generalization for non-monotone relationships.

The receiver operating characteristic curve is a popular graphical method frequently used in order to study the diagnostic capacity of continuous markers. It represents in a plot true-positive rates against the false-positive ones. Both the practical and theoretical aspects of the receiver operating characteristic curve have been extensively studied. Conventionally, it is assumed that the considered marker has a monotone relationship with the studied characteristic; i.e., the upper (lower) values of the (bio)marker are associated with a higher probability of a positive result. However, there exist real situations where both the lower and the upper values of the marker are associated with higher probability of a positive result. We propose a receiver operating characteristic curve generalization, [Formula: see text], useful in this context. All pairs of possible cut-off points, one for the lower and another one for the upper marker values, are taken into account and the best of them are selected. The natural empirical estimator for the [Formula: see text] curve is considered and its uniform consistency and asymptotic distribution are derived. Finally, two real-world applications are studied.

Severe subcutaneous emphysema and pneumomediastinum secondary to noninvasive ventilation support in status asthmaticus.

A 12-year-old male with status asthmaticus developed subcutaneous emphysema and pneumomediastinum. He was transferred to our unit, where he received noninvasive ventilation (NIV). This respiratory support technique is not an absolute contraindication in these cases. After 2 h on NIV, he worsened sharply and the subcutaneous emphysema got bigger suddenly. He needed invasive ventilation for 5 days. Final outcome was satisfactory. This case illustrates that it is mandatory to keep a high level of vigilance when using NIV in patients with air leaks.

[High levels of atrial natriuretic peptide and copeptin and mortality risk]. / Asociación de valores elevados de péptido natriurético auricular y copeptina con riesgo de mortalidad.

OBJECTIVE: To determine whether high levels of mid-regional pro-atrial natriuretic peptide (MR-proANP), copeptin, and procalcitonin (PCT) plasma concentrations are associated with increased mortality risk. METHODS: Prospective observational study including 254 critically ill children. MR-proANP, copeptin and PCT were compared between children with high (Group A; n=33) and low (Group B; n=221) mortality risk, and between patients with failure of more than 1 organ (Group 1; n=71) and less than 2 (Group 2; n=183). RESULTS: Median (range) of MR-proANP, copeptin, and PCT levels in group A vs B were, respectively: 209.4 (30.5-1415.8) vs. 75.0 (14.6-867.2) pmol/L (P<.001); 104.4 (7.4-460.9) vs. 26.6 (0.00-613.1) pmol/L (P<.001), and 7.8 (0.3-552.0) vs. 0.3 (0.02-107.0) ng/mL (P<.001). The area under the curve (AUC) for the differentiation of group A and B was 0.764 (95% CI: 0.674-0.854) for MR-proANP; 0.735 (0.642-0.827) for copeptin, and 0.842 (0.744-0.941) for PCT, with no statistical differences. The AUCs for the differentiation of group 1 and 2 were: 0.837 (0.784-0.891) for MR-proANP, 0.735 (0.666-0.804) for copeptin, and 0.804 (0.715-0.892) for PCT, with statistical differences between MR-proANP and copeptin, P=.01. CONCLUSIONS: High levels of MR-proANP, copeptin and PCT were associated with increased mortality risk scores. MR-proANP showed a higher association than copeptin with number of organs in failure.

Vitamin D deficiency at pediatric intensive care admission / Deficiência de vitamina D em internações na unidade de terapia intensiva pediátrica

OBJECTIVE: to assess whether 25hydroxivitaminD or 25(OH)vitD deficiency has a high prevalence at pediatric intensive care unit (PICU) admission, and whether it is associated with increased prediction of mortality risk scores. METHOD: prospective observational study comparing 25(OH)vitD levels measured in 156 patients during the 12 hours after critical care admission with the 25(OH)vitD levels of 289 healthy children. 25(OH)vitD levels were also compared between PICU patients with pediatric risk of mortality III (PRISM III) or pediatric index of mortality 2 (PIM 2) > p75 [(group A; n = 33) vs. the others (group B; n = 123)]. Vitamin D deficiency was defined as < 20 ng/mL levels. RESULTS: median (p25-p75) 25(OH)vitD level was 26.0 ng/mL (19.2-35.8) in PICU patients vs. 30.5 ng/mL (23.2-38.6) in healthy children (p = 0.007). The prevalence of 25(OH)vitD < 20 ng/mL was 29.5% (95% CI: 22.0-37.0) vs. 15.6% (95% CI: 12.2-20.0) (p = 0.01). Pediatric intensive care patients presented an odds ratio (OR) for hypovitaminosis D of 2.26 (CI 95%: 1.41-3.61). 25(OH)vitD levels were 25.4 ng/mL (CI 95%: 15.5-36.0) in group A vs. 26.6 ng/mL (CI 95%: 19.3-35.5) in group B (p = 0.800). CONCLUSIONS: hypovitaminosis D incidence was high in PICU patients. Hypovitaminosis D was not associated with higher prediction of risk mortality scores. .

OBJETIVO: avaliar se a deficiência da 25-hidroxivitamina D, ou 25 (OH) vitD, tem prevalência elevada em internações na unidade de terapia intensiva pediátrica, e se estaria relacionada à previsão de escores de risco de mortalidade. MÉTODO: estudo observacional prospectivo comparando níveis de 25 (OH) vitD de 156 pacientes, mensurados nas primeiras 12 horas da internação em terapia intensiva, com níveis de 25 (OH) vitD de 289 crianças saudáveis. Os níveis de 25 (OH) vitD também foram comparados entre pacientes na UTIP com escore PRISM III ou PIM 2 > p75 (Grupo A; n = 33), e o restante, (Grupo B; n = 123). A deficiência de vitamina D foi definida como níveis < 20 ng/mL. RESULTADOS: o nível médio (p25-p75) de 25 (OH) vitD foi 26,0 ng/mL (19,2-35,8) em pacientes internados na UTIP, em comparação a 30,5 ng/mL (23,2-38,6) em crianças saudáveis (p = 0,007). A prevalência de 25 (OH) vitD < 20 ng/mL foi de 29,5% (IC 95%, 22,0-37,0), em comparação a 15,6% (IC 95%,12,2-20,0) (p = 0,01). Os pacientes em terapia intensiva pediátrica apresentaram uma razão de chance (RC) para hipovitaminose D de 2,26 (IC 95%, 1,41-3,61). Os níveis de 25 (OH) vitD foram 25,4 ng/mL (IC 95%, 15,5-36,0) no grupo A, em comparação a 26,6 ng/mL (IC 95%, 19,3-35,5) no grupo B (p = 0,800). CONCLUSÕES: a incidência de hipovitaminose D foi elevada em pacientes em terapia intensiva pediátrica, mas não foi associada à maior previsão de escores de risco de mortalidade. .

Vitamin D deficiency at pediatric intensive care admission.

OBJECTIVE: to assess whether 25hydroxivitaminD or 25(OH)vitD deficiency has a high prevalence at pediatric intensive care unit (PICU) admission, and whether it is associated with increased prediction of mortality risk scores. METHOD: prospective observational study comparing 25(OH)vitD levels measured in 156 patients during the 12 hours after critical care admission with the 25(OH)vitD levels of 289 healthy children. 25(OH)vitD levels were also compared between PICU patients with pediatric risk of mortality III (PRISM III) or pediatric index of mortality 2 (PIM 2) > p75 [(group A; n = 33) vs. the others (group B; n = 123)]. Vitamin D deficiency was defined as < 20 ng/mL levels. RESULTS: median (p25-p75) 25(OH)vitD level was 26.0 ng/mL (19.2-35.8) in PICU patients vs. 30.5 ng/mL (23.2-38.6) in healthy children (p = 0.007). The prevalence of 25(OH)vitD < 20 ng/mL was 29.5% (95% CI: 22.0-37.0) vs. 15.6% (95% CI: 12.2-20.0) (p = 0.01). Pediatric intensive care patients presented an odds ratio (OR) for hypovitaminosis D of 2.26 (CI 95%: 1.41-3.61). 25(OH)vitD levels were 25.4 ng/mL (CI 95%: 15.5-36.0) in group A vs. 26.6 ng/mL (CI 95%: 19.3-35.5) in group B (p = 0.800). CONCLUSIONS: hypovitaminosis D incidence was high in PICU patients. Hypovitaminosis D was not associated with higher prediction of risk mortality scores.

Pro-adrenomedullin, pro-endothelin-1, procalcitonin, C-reactive protein and mortality risk in critically ill children: a prospective study.

INTRODUCTION: We tested the hypothesis that higher mid-regional pro-adrenomedullin (MR-proADM), carboxy-terminal pro-endothelin-1 (CT-proET-1), procalcitonin (PCT) and C-reactive protein (CRP) plasma concentrations would be associated with increased prediction of mortality risk scores. METHODS: Prospective observational study set in two pediatric intensive care units (PICUs). Two-hundred-thirty-eight patients were included. MR-proADM, CT-proET-1, PCT and CRP levels were compared between children with PRISM III and PIM 2 > p75 (Group A; n = 33) and the rest (Group B; n = 205). RESULTS: Median (range) MR-proADM levels were 1.39 nmol/L (0.52-12.67) in group A versus 0.54 (0.15-3.85) in group B (P < 0.001). CT-proET-1 levels were 172 pmol/L (27-500) versus 58 (4-447) (P < 0.001). PCT levels were 7.77 ng/mL (0.34-552.00) versus 0.28 (0.02-107.00) (P < 0.001). CRP levels were 6.23 mg/dL (0.08-28.25) versus 1.30 mg/dL (0.00-42.09) (P = 0.210). The area under the ROC curve (AUC) for the differentiation of group A and B was 0.87 (95% CI:0.81-0.821) for MR-proADM, 0.86 (95% CI:0.79-0.92) for CT-proET-1 and 0.84 (95% CI:0.74-0.94) for PCT. A MR-proADM > 0.79 nmol/L had 93% sensitivity and 76% specificity to differentiate groups, whereas a CT-proET-1 > 123 pmol/L had 77% sensitivity and 84% specificity, and a PCT concentration > 2.05 ng/mL had 80% sensitivity and specificity. CONCLUSIONS: In critically ill children, high levels of MR-proADM, CT-proET-1 and PCT were associated with increased prediction of mortality risk scores. MR-proADM, CT-proET-1 and PCT concentrations higher than 0.80 nmol/L, 123 pmol/L and 2 ng/mL, respectively, could be used by clinicians to identify critically ill children at higher prediction of risk death scores.

Predicting non-invasive ventilation failure in children from the SpO2/FiO2 (SF) ratio.

PURPOSE: Our objective was to assess whether SpO2/FiO2 (SF) ratio could be a useful NIV outcome predictor in children with acute respiratory failure (ARF) and tried to develop a predictive model of NIV failure. METHODS: Prospective, observational, multicenter study. Episodes of ARF-fulfilling inclusion criteria from 15 January 2010 to 14 January 2011 were treated with NIV according to a pre-established protocol. Clinical variables were collected at baseline and at 1, 2, 6, 12 and 24 h. Failure criterion was the need for endotracheal intubation. Failures were considered as "early" if occurring ≤6 h after NIV initiation, "intermediate" if occurring between 6 and 24 h, and "late" if occurring after 24 h. Variables with a p < 0.1 in univariate analysis corrected by age were included in multivariate analysis. Models were calculated based on multivariate analysis. RESULTS: During the study period, 390 episodes were included. NIV success rate was 81.3 %. Among ARF causes, failure occurred most frequently in ARDS episodes. The failure predictive model for the whole sample included SF ratio at 1 h, age and PRISM III-24 (area under the curve AUC of 0.755). For early NIV failures, SF ratio at 1 h was the only variable within model (AUC 0.748). The analysis of intermediate NIV failures identified 3 variables independently linked to NIV outcome: PRISM III-24, RR decrease at 6 h, and SF ratio at 6 h (AUC 0.895). No model was identified for late NIV failure. CONCLUSIONS: SF ratio is a reliable predictor of early NIV failure in children.