Accuracy of non-invasive liver stiffness measurement and steatosis quantification in patients with severe and morbid obesity.

Abstract

Vibration controlled transient elastography (VCTE) and controlled attenuation parameter (CAP™) have shown reliable performance predicting fibrosis and steatosis in normal- to overweight patients but have not been validated in severe to morbid obesity. This study aimed at determining the accuracy of VCTE, CAP™ and the composite score FibroScan-AST (FAST) in patients with a body mass index (BMI) of ≥35 kg/m2. Patients scheduled for bariatric-metabolic surgery underwent preoperative VCTE/CAP™ measurement, and intraoperative liver biopsy. The feasibility and accuracy of VCTE, CAP™ and the composite score FAST were retrospectively analysed to evaluate fibrosis, steatosis and active fibrotic non-alcoholic steatohepatitis [NASH + non-alcoholic fatty liver disease (NAFLD) activity score ≥4 + fibrosis grade ≥2] using per protocol (PP) and intent to diagnose (ITD) calculation. In total, 170 patients (median BMI 44.4 kg/m2) were included in the study. Liver biopsy showed NASH, simple steatosis, and normal livers in 60.6% (n=103), 28.8% (n=49), and 10.6% (n=18), respectively. VCTE and CAP™ delivered reliable results in 90.6% (n=154/170) and 90.5% (n=134/148). The AUC (PP) of VCTE, CAP™, and FAST were 0.687 (≥F2), 0.786 (≥F3), 0.703 (≥S2), 0.738 (S3), and 0.780 (active fibrotic NASH). The AUC increased to 0.742 (≥F2), 0.842 (≥F3), 0.712 (≥S2), 0.780 (S3), and 0.836 (active fibrotic NASH) in patients below the median BMI of 44.4 kg/m2. VCTE, CAP™ and FAST show acceptable accuracy for the detection of fibrosis, steatosis and NASH in a real-life cohort of patients with obesity. Accuracy improves in patients with a BMI <44.4 kg/m2.