Abstract: Objective The aim of this study was to investigate prediction of nonalcoholic steatohepatitis (NASH) by ultrasonic attenuation coefficient (UAC) or controlled attenuation parameter (CAP) and liver stiffness measurement (LSM) combination in patients with nonalcoholic fatty liver diseases (NAFLD). Methods 87 patients with NAFLD were recruited in our hospital between June 2020 and June 2023, and all underwent liver biopsies and ultrasonography for UAC detection or Fibrotouch for CAP and LSM detection. Receiver operating characteristic (ROC) was applied to predict existence of NASH. Results Histopathological examination showed simple fatty liver(SFL) in 61 cases and NASH in 26 cases; the UAC, CAP and LSM in patients with NASH were (0.9±0.1)dB/cm·MHz, (327.4±30.6)dB/mand (11.6±1.6)kPa, all significantly greater than [(0.6±0.1)dB/cm·MHz, (266.3±21.5)dB/m and (6.7±0.4)kPa, respectively, P＜0.05] in patients with SFL; the area under ROC (AUC) was 0.9, with sensitivity (Se) of 84.0% and specificity (Sp) of 87.6%, when CAP of 315.2 dB/m and LSM of 10.3 kPa were combined as cut-off-value (COV) in predicting existence of NASH, much superior to that by UAC (COV was 0.9 dB/cm·MHz, AUC was 0.8, Se was 80.0% and Sp was 72.4%, P<0.05). Conclusion Combination of CAP and LSM in predicting occurrence of NASH in patients with NAFLD has a satisfactory diagnostic performance, which might help to screen in clinical practice.

Key words: Nonalcoholic steatohepatitis, Ultrasonic attenuation coefficient, Controlled attenuation parameter, Liver stiffness measurement, Diagnosis