Pediatric CLIF-SOFA score is the best predictor of 28-day mortality in children with decompensated chronic liver disease.

BACKGROUND & AIMS: Early identification of children with decompensated chronic liver disease (DCLD) at risk of short-term mortality helps improve outcome. We aimed to evaluate the predictors of outcome and role of Child-Pugh, pediatric end-stage liver disease (PELD) and pediatric chronic liver failure sequential organ failure assessment (pCLIF-SOFA) score for prognosticating 28-day mortality in children with DCLD.

METHODS: DCLD children were prospectively evaluated with a clinico-laboratory proforma and followed for 28 days to determine outcome. Child-Pugh, PELD and pCLIF-SOFA were calculated at admission. Univariate and multivariate analysis was performed to identify the best predictors of outcome.

RESULTS: A total of 110 children (74 boys, 96 [4-204] months) were enrolled and 37 (33.6%) died at 28 days. Significant risk factors for mortality were a higher international normalized ratio (hazard ratio [HR] 1.17; 95% CI 1.04-1.31; p <0.001) and bilirubin (HR 1.04; 95% CI 1.01-1.08; p <0.001), lower albumin (HR 0.46; 95% CI 0.27-0.77; p = 0.03) and sodium (HR 0.93; 95% CI 0.89-0.98; p = 0.01), absence of treatable etiology (HR 2.00; 95% CI 1.40-2.87; p = 0.001) and presence of organ failure (HR 3.22; 95% CI 1.98-10.58; p <0.001). Organ failure and serum sodium were independent predictors of poor outcome on multivariate analysis. pCLIF-SOFA (16 [9-22] vs. 9 [5-15]), Child-Pugh (11 [9-15] vs. 10 [8-14]) and PELD (22.2 [7.5-45.3] vs. 15.3 [4.5-23.9]) scores were significantly higher in non-survivors. The area under the curve was 0.977 for pCLIF-SOFA, 0.815 for Child-Pugh score, and 0.741 for PELD score. A pCLIF-SOFA score of ≥11 identified 28-day mortality with a sensitivity and specificity of 94.9% and 91.5%, respectively.

CONCLUSION: Thirty-four percent of children with DCLD have a poor short-term outcome. Organ failure and low serum sodium are independent predictors of outcome. pCLIF-SOFA performs better than Child-Pugh and PELD in prognostication of 28-day mortality. Our study supports the use of scores based on organ failure in prognosticating children with DCLD.

LAY SUMMARY: The ability to predict the course of a disease is an important part of the assessment, enabling timely interventions that improve outcomes. We evaluated the outcome (death vs. survival) and compared three different scoring systems for their ability to predict mortality within 28 days in children with decompensated chronic liver disease (DCLD). One-third of children with DCLD died within 28 days and the pediatric chronic liver failure sequential organ failure assessment score, which considers the main organ systems of the body (lungs, liver, brain, kidney, blood and cardiac) fared better for identification of children with a poor outcome than the Child-Pugh and pediatric end-stage liver disease score which comprise of only liver-related parameters. Our study supports the use of scores based on organ failure in prognosticating children with DCLD.