Relationship between various hepatic function scores and the formation of esophageal varices in patients with HIV/hepatitis C virus co-infection due to contaminated blood products for hemophilia.

Abstract

It is reportedly difficult to accurately assess the liver reserve capacity of patients with HIV/hepatitis C virus (HCV) co-infection through contaminated blood products by the Child-Pugh (CP) classification. Therefore, we investigated a clinically applicable scoring system in determining the risk of esophageal varices in HIV/HCV co-infected patients, known as latent portal hypertension leading to esophageal varices. Forty-three patients with HIV/HCV co-infection underwent clinical examinations, including endoscopy and assessment of hepatic reserve, in our department between 2009 and 2017. Child-Pugh score, the recently developed albumin-bilirubin (ALBI) grade, and the albumin-indocyanine green evaluation (ALICE) were compared to evaluate their diagnostic accuracy for the detection of esophageal varices using the area under the receiver operating characteristic curve (AUROC). The patients were all male hemophiliacs and were positive for both HIV and HCV antibodies, with a median age of 45years (range, 29-66years). Thirty-seven patients (84.1%) were classified as CP A at the examination. The comparison of AUROCs showed a superior diagnostic accuracy for ALICE (AUROC = 0.814) to detect esophageal varices. The positive prediction rate was the highest with ALICE if -2.325 was set, and the negative prediction rate was the highest with ALBI if -2.575 was set. The ALICE showed the highest accuracy compared to other two scores. The ALICE score was found to be the most valuable system for portal hypertension in HIV/HCV co-infected hemophilia patients. Because of its high specificity, ALICE for secondary surveillance could be used after other markers such as the aspartate aminotransferase to platelet ratio index and Fibrosis-4 index.