This article is more than 5 years old. Information may no longer be current.

pCLIF-SOFA predicts mortality in children with decompensated liver disease

The pediatric chronic liver failure sequential organ failure assessment, or pCLIF-SOFA, performed better than Child-Pugh or pediatric end-stage liver disease assessment in predicting 28-day mortality among pediatric patients with decompensated chronic liver disease, according to recently published data.

“Though widely used, [Child-Pugh score and pediatric end-stage liver disease (PELD) score] do not give points for other organ dysfunction like cardiovascular, renal and pulmonary failure, which are often the cause for short-term mortality,” Rishi Bolia, MD, DM, from the Sanjay Gandhi Postgraduate Institute of Medical Sciences, and colleagues wrote. “A scoring system based on organ failure is a better predictor of short-term mortality than liver specific scores in children with advanced chronic liver disease.”

To evaluate the predictors of 28-day outcomes in children with decompensated chronic liver disease (DCLD), Bolia and colleagues prospectively enrolled 110 pediatric patients admitted to their institution. Etiologies included autoimmune liver disease (20%), Wilson’s disease (17.1%), Budd Chiari Syndrome (16.3%) and biliary atresia (14.5%).

At 28 days, 31 patients died in hospital and six died after discharge. Risk factors for mortality included higher international normalized ratio (HR = 1.17; 95% CI, 1.04-1.31), higher bilirubin (HR = 1.04; 95% CI, 1.01-1.08), lower albumin (HR = 0.46; 95% CI, 0.27-0.77), lower sodium (HR = 0.93; 95% CI, 0.89-0.98), no specific therapy for underlying chronic liver disease (HR = 2; 95% CI, 1.4-2.87) and the presence of organ failure (HR = 3.22; 95% CI, 1.98-10.58).

Multivariate analysis showed that lower sodium (HR = 0.94; 95% CI, 0.895-0.987) and the presence of organ failure (HR = 4.89; 95% CI, 1.47-16.22) correlated independently with poor outcomes.

Eighty-two patients had one or more organ failure at admission. While single organ failure did not affect outcomes, patients with two or more organ failures displayed a linear relationship between the number of organ failures and mortality rates. Pediatric patients with respiratory failure had the highest mortality rates (95.4%), followed by circulatory (92.8%), renal (80%), neurological (66.7%), hepatic (57%) and hematological failures (51.7%).

pCLIF-SOFA, Child-Pugh and PELD scores were higher among nonsurviving patients at admission. pCLIF-SOFA (HR = 2.41; 95% CI, 1.43-3.24), Child-Pugh score (HR = 1.01; 95% CI, 1-1.13) and PELD score (HR = 1.06; 95% CI, 1.01-1.11) correlated with mortality.

The integrated discrimination improvement of using pCLIF-SOFA to predict mortality was 23% (P = .04) compared with PELD and 11% (P = .01) vs. the Child-Pugh score. A pCLIF-SOFA score of 11 identified 28-day mortality with a sensitivity of 94.9% and specificity of 91.5%. A Child-Pugh score of 13 identified 28-day mortality with a sensitivity of 87.2% and specificity of 81% and a PELD score of 24 had a sensitivity of 82.8% and specificity of 80.2%.

“Prognosis is an essential part of the assessment of any disease. Early identification of those ‘at risk’ of poor outcome can lead to early introduction of aggressive therapeutic measures and more judicious prioritization of organ allocation,” the researchers wrote. “We found that the pCLIF-SOFA score performed better in predicting outcome than the Child-Pugh and PELD score. This is because the Child-Pugh and PELD are liver based and do not consider extrahepatic organ failure which is very important in determining outcome in patients with DCLD.” – by Talitha Bennett

Disclosure: The authors report no relevant financial disclosures.