Nonrespiratory pediatric logistic organ dysfunction-2 score is a good predictor of mortality in children with acute respiratory failure

Abstract

Objective: Multiple organ dysfunction, not respiratory failure, is the major cause of death in children with acute lung injury or acute respiratory distress syndrome. This study was undertaken to estimate the predictive value of death of the nonrespiratory Pediatric Logistic Organ Dysfunction-2 in children with acute respiratory failure.

Design: Analysis of the database of the recently published Pediatric Logistic Organ Dysfunction-2.

Setting: Nine multidisciplinary, tertiary-care PICU of university-affiliated hospitals in France and Belgium.

Patients: All consecutive children (excluding neonates) admitted to these PICUs (June 2006 to October 2007) and having acute respiratory failure.

Intervention: None.

Measurements and main results: We prospectively collected data on variables considered for the Pediatric Logistic Organ Dysfunction-2 score during PICU stay up to eight time points: days 1, 2, 5, 8, 12, 16, and 18, plus PICU discharge. For each variable considered in the Pediatric Logistic Organ Dysfunction-2 score, the most abnormal value observed during time points was collected. Outcome was vital status at PICU discharge. We used areas under receiver operating characteristic curve to estimate the discrimination and Hosmer-Lemeshow goodness-of-fit tests to estimate calibration of the Pediatric Logistic Organ Dysfunction-2 and the nonrespiratory Pediatric Logistic Organ Dysfunction-2 scores, with correction for the optimism bias using a bootstrap resampling method. We included 1,572 consecutive patients (median age, 20.6 months; interquartile range, 5.5-80.2; mortality rate, 9.5%). Discrimination of the Pediatric Logistic Organ Dysfunction-2 and the nonrespiratory Pediatric Logistic Organ Dysfunction-2 were excellent (areas under receiver operating characteristic curve = 0.93 and 0.92, respectively) and calibration (chi-square test for goodness-of-fit = 5.8, p = 0.45 and 7.6, p = 0.27, respectively) was good. The four nonrespiratory organ dysfunctions were closely related to the risk of mortality (p< 0.001).

Conclusion: Our study demonstrates that the nonrespiratory Pediatric Logistic Organ Dysfunction-2 score of the entire PICU stay is highly predictive of death in children with acute respiratory failure of whom 94.3% were invasively ventilated. The nonrespiratory Pediatric Logistic Organ Dysfunction-2 score could represent the nonrespiratory organ failure definition tool whose development was recommended in the international expert recommendations on pediatric acute respiratory distress syndrome.