Liver stiffness and fibrosis-4 alone better predict liver events compared with aspartate aminotransferase to platelet ratio index in a cohort of human immunodeficiency virus and hepatitis C virus co-infected patients from ANRS CO13 HEPAVIH cohort

International audienceObjectives HIV/hepatitis C virus (HCV) co-infection leads to major complications, and noninvasive markers developed to stage liver fibrosis could be used as prognostic markers. We aimed to compare the performances of liver stiffness (LS), fibrosis-4 (FIB-4), and aspartate aminotransferase to platelet ratio index (APRI) to predict liver-related events in HIV/HCV co-infected patients.Patients and methods HIV/HCV co-infected patients from the ANRS CO13 HEPAVIH cohort were included if they had LS, FIB-4, and APRI measurements done in a window of 3 months. Primary outcome was the time between inclusion and occurrence of a liver-related event. Univariable and multivariable Fine and Gray models were performed. Predictive performances were compared by the area under the receiver operating characteristic (AUROC) differences after correction of optimistic by bootstrap samples. Best cutoffs to predict liver-related events were estimated by sensitivity and specificity maximization.Results A total of 998 patients were included. Overall, 70.7% were men. Their median age was 46.8 years. According to LS value, 204 (20.4%) patients had cirrhosis. Overall, 39 patients experienced at least one liver-related event. In univariable analysis, LS AUROC curve was significantly superior to FIB-4 and APRI AUROC curves, being 87.9, 78.2, and 75.0%, respectively. After adjustment on age, CD4 levels, and insulin resistance, no differences were observed. The best cutoffs to identify patients at low or high risk of liver-related events were below 8.5, 1.00, and 0.35 and above 16.5, 4.00, and 1.75 for LS, FIB-4, and APRI, respectively.Conclusion To predict HCV-related events, APRI had lower performance than LS and FIB-4. FIB-4 is as good as LS to predict HCV-related events, suggesting that it can be used for the management of HIV/HCV co-infected patients and replace LS