Journal of Clinical and Diagnostic Research (Oct 2018)
Comparative Assessment of Validated Prognostic Models in Estimation of 30 Day Mortality in Alcoholic Hepatitis
Abstract
Introduction: Alcoholic Hepatitis (AH) is a clinical syndrome seen in those with heavy alcohol intake and carries a very high mortality risk despite therapy. Patients with Severe Alcoholic Hepatitis (SAH) are currently advocated therapy with either Corticosteroids (CS) or Pentoxifylline (PTX). Various prognostic scores have been validated to assess for mortality risk. Studies comparing these validated prognostic scores for prediction of early mortality risk in Indian patients with SAH have been scarce. Aim: To assess and compare utility of various prognostic scoring systems in predicting short term (30 day) mortality in patients with SAH. Materials and Methods: This was a prospective comparative study and was approved by in house Ethics Committee of the Osmania general hospital, Hyderabad, Telangana, India. Patients 18 years or older with clinical alcoholic hepatitis, Aspartate Aminotransferase (AST)/Alanine Aminotransferase (ALT) ratio >2 with an AST level 90% specificity for 30 day mortality while sensitivity and the AUROCs of all scores were also increased, ranging from AUROC of 0.933 for UKELD at a value of 65.6 to an AUROC of 1.00 for a MELD score of 25.9. MELD at day seven at a cut-off of 28 showed 100% specificity with a PPV of 100 while MDF at day seven with best cut-off of 59.7 had 100% sensitivity and a Negative Predictive Value (NPV) of 100. However, CTP did not show significant dynamic change at day seven. Conclusion: All evaluated scores at the time of admission and on day seven showed good sensitivity and specificity for assessment of 30 day mortality risk in patients with SAH. UKELD, CTP calculated on day one showed most significance on multivariate cox-regression analysis for 30 day mortality. At higher cut-off values, MDF, MELD, UKELD and GAHS can be used to stratify risk and identify patients with very high risk of 30 day mortality.
Keywords