Diagnostic accuracy of liver and spleen stiffness measured by fibroscan® in the prediction of esophageal varices in HCV-related cirrhosis patients treated with oral antivirals

Abstract

IntroductionThe aim of this study was to investigate the accuracy of liver and spleen stiffness measurement by transient elastography for the prediction of gastroesophageal varices in patients with HCV-associated cirrhosis treated with new direct-acting antiviral agents. Patients and methodsThis cross-sectional observational study included patients with compensated HCV-related cirrhosis and sustained virological response after direct-acting antiviral therapy. Patients underwent liver and spleen stiffness measurement, abdominal ultrasound and oesophago-gastroduodenoscopy. Clinical and laboratory data and non-invasive markers such as the liver stiffness–spleen diameter to platelet ratio score, variceal risk index and platelet count to spleen diameter ratio were analyzed. ResultsNinety-seven consecutive patients were included. Liver stiffness measurement (12.2 vs 16; p=0.02), spleen stiffness measurement (39.4 vs 46.05; p=0.04), liver stiffness–spleen diameter to platelet ratio score (1.21 vs 2.02; p=0.008), platelet count to spleen diameter ratio (1102.19 vs 829.7; p=0.04) and variceal risk index (−3.4 vs −1.02; p=0.01) showed significant differences between patients without/with gastroesophageal varices. The best cut-off value to discard the presence of gastroesophageal varices was 12.3kPa for liver stiffness measurement and 27kPa for spleen stiffness measurement. However, diagnostic accuracy was moderate (AUROC: 0.671 and 0.624 respectively). Combining different non-invasive parameters did not significantly improve the overall performance. DiscussionLiver and spleen stiffness measurement showed suboptimal results for non-invasive assessment of gastroesophageal varices in HCV cirrhotic patients treated with direct-acting antiviral agents. Our results suggest that non-invasive methods cannot substitute standard procedures for predicting gastroesophageal varices in this population.