Clinical features and predictors of outcome in acute hepatitis A and hepatitis E virus hepatitis on cirrhosis

Authors


  • Re-use of this article is permitted in accordance with the Creative Commons Deed, Attribution 2.5, which does not permit commercial exploitation

Correspondence
Y. Radha Krishna, Department of Gastroenterology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Raebareli Road, Lucknow 226014, India
Tel: +97 0488 3149
Fax: +97 (0)40 2784 0980
e-mail: radhayellapu@yahoo.co.in

Abstract

Background and Objectives: Acute hepatitis A and E are recognized triggers of hepatic decompensation in patients with cirrhosis, particularly from the Indian subcontinent. However, the resulting acute-on-chronic liver failure (ACLF) has not been well characterized and no large studies are available. Our study aimed to evaluate the clinical profile and predictors of 3-month mortality in patients with this distinctive form of liver failure.

Methods: ACLF was diagnosed in patients with acute hepatitis A or E [abrupt rise in serum bilirubin and/or alanine aminotransferase with positive immunoglobulin M anti-hepatitis A virus (HAV)/anti-hepatitis E virus (HEV)] presenting with clinical evidence of liver failure (significant ascites and/or hepatic encephalopathy) and clinical, biochemical, endoscopic (oesophageal varices at least grade II in size), ultrasonographical (presence of nodular irregular liver with porto-systemic collaterals) or histological evidence of cirrhosis. Clinical and laboratory profile were evaluated, predictors of 3-month mortality were determined using univariate and multivariate logistic regression and a prognostic model was constructed. Receiver-operating curves were plotted to measure performance of the present prognostic model, model for end-stage liver disease (MELD) score and Child–Turcotte–Pugh (CTP) score.

Results: ACLF occurred in 121 (3.75%) of 3220 patients (mean age 36.3±18.0 years; M:F 85:36) with liver cirrhosis admitted from January 2000 to June 2006. It was due to HEV in 80 (61.1%), HAV in 33 (27.2%) and both in 8 (6.1%). The underlying liver cirrhosis was due to HBV (37), alcohol (17), Wilson's disease (8), HCV (5), autoimmune (6), Budd-Chiari syndrome (2), haemochromatosis (2) and was cryptogenic in the rest (42). Common presentations were jaundice (100%), ascites (78%) and hepatic encephalopathy (55%). Mean (SD) CTP score was 11.4±1.6 and mean MELD score was 28.6±9.06. Three-month mortality was 54 (44.6%). Complications seen were sepsis in 42 (31.8%), renal failure in 45 (34%), spontaneous bacterial peritonitis in 27 (20.5%), UGI bleeding in 15(11%) and hyponatraemia in 50 (41.3%). On univariate analysis, ascites, hepatic encephalopathy, renal failure, GI bleeding, total bilirubin, hyponatraemia and coagulopathy were significant predictors of mortality. Multivariate analysis revealed grades 3 and 4 HE [odds ratio (OR 32.1)], hyponatraemia (OR 9.2) and renal failure (OR 16.8) as significant predictors of 3-month mortality and a prognostic model using these predictors was constructed. Areas under the curve for the present predicted prognostic model, MELD, and CTP were 0.952, 0.941 and 0.636 respectively.

Conclusions: ACLF due to hepatitis A or E super infection results in significant short-term mortality. The predictors of ominous outcome include grades 3 and 4 encephalopathy, hyponatraemia and renal failure. Present prognostic model and MELD scoring system were better predictors of 3-month outcome than CTP score in these patients. Early recognition of those with dismal prognosis may permit timely use of liver replacement/supportive therapies.

Ancillary