Clinical Epidemiology and Ageing

Non-invasive fibrosis tests to predict complications in compensated post-hepatitis C cirrhosis.

Zarski J-P, David-Tchouda S, Trocme C, Margier J, Vilotitch A, Hilleret M-N, Cagnot C, Boursier V, Ziol M, Sutton A, Layese R, Audureau E, Roudot-Thoraval F, Nahon P Clin Res Hepatol Gastroenterol. 2020;44(4):524-531.

BACKGROUND AND AIMS: Markers predicting complications of post-hepatitis C cirrhosis are needed. We asked whether changes in noninvasive markers of fibrosis can predict liver-related complications.

METHODS: This was a case-controlled study using a prospective national cohort (ANRS-CO12-CIRVIR) of 1323 HCV-infected patients with compensated cirrhosis: 97 patients who developed liver-related complications such as hepatocellular carcinoma or hepatic decompensation (cases) matched in age, sex and follow-up duration were compared with 257 patients without complications (controls). Actitest/Fibrotest™, Inflameter/Fibrometer™, ELF™ and Fibroscan™ were performed at baseline and yearly. Samples based on Propensity score matching were built and mixed linear models performed. Outcomes in a sustained virological response (SVR) negative population and a SVR-positive population were also described.

RESULTS: At baseline, all characteristics of patients were similar between the groups. All fibrosis tests were statistically higher for cases compared to controls, Fibroscan™ excepted: Fibrotest™: 0.83±0.13 vs. 0.77±0.16; Fibrometer™: 0.93±0.07 vs. 0.90±0.11; ELF™: 11.4±1.0 vs. 11.0±1.2 (P<0.02). The mean follow-up was 5.7±1.9 years. Over a 3-year period, the significant difference in fibrosis marker values between cases and controls remained constant; with a trend toward a decrease in inflammation markers in controls, independent of SVR status.

CONCLUSIONS: Baseline noninvasive serum fibrosis and inflammation markers were significantly higher in patients developing a complication than in controls. During the follow-up only inflammatory markers decreased in controls, but not in cases, and thus could potentially be used to predict the occurrence of complications in cirrhotic patients.

DOI: 10.1016/j.clinre.2019.11.005