Optimized CAP cut-offs for metabolic dysfunction associated steatotic liver disease in patients living with obesity: a large biopsy-based prospective study
摘要
Controlled Attenuation Parameter (CAP) quantifies hepatic steatosis non-invasively, but current cut-offs limit diagnostic precision, especially in obese populations. We aimed to validate CAP against liver biopsy and propose optimized thresholds tailored for obesity. In this single-centre, cross-sectional, biopsy-validated study, 798 adults undergoing laparoscopic cholecystectomy (n = 629) or sleeve gastrectomy (n = 169) between 2019 and 2024 were included. All underwent Fibro Scan and wedge liver biopsy. Histology was scored by two blinded pathologists. Diagnostic accuracy of CAP was assessed using AUROC, sensitivity, specificity, predictive values, calibration, net reclassification improvement (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA). Newly derived cut-offs were benchmarked against existing ones. Histology revealed S0 in 72.7%, S1 in 19.7%, S2 in 7.0%, and S3 in 0.6%. Standard CAP cut-offs showed suboptimal specificity (49.7% for ≥S1, 55.8% for ≥S2). Newly derived thresholds of 290 dB/m (≥ S1) and 317 dB/m (≥ S2) improved specificity (74.4% and 82.5%) while maintaining good sensitivity (62.7% and 71.7%). NRI and DCA demonstrated superior alignment and higher net benefit compared with all published thresholds. Optimized CAP cut-offs (290 and 317 dB/m) provide better diagnostic precision and clinical utility than current thresholds strengthening MASLD screening and clinical trial eligibility, particularly in obese populations.