Controlled attenuation parameter for the detection and quantification of hepatic steatosis in nonalcoholic fatty liver disease

Abstract

Background and aim: Controlled attenuation parameter (CAP) has been suggested as a noninvasive method for detection and quantification of hepatic steatosis. We aim to study the diagnostic performance of CAP in nonalcoholic fatty liver disease (NAFLD) patients.

Methods: Transient elastography was performed in consecutive NAFLD patients undergoing liver biopsy and non-NAFLD controls. The accuracy of CAP for the detection and quantification of hepatic steatosis was assessed based on histological findings according to the Nonalcoholic Steatohepatitis Clinical Research Network Scoring System.

Results: Data for 101 NAFLD patients (mean age 50.3 ± 11.3 years old, 51.5% male) and 60 non-NAFLD controls were analyzed. CAP was associated with steatosis grade (odds ratio [OR] = 29.16, P < 0.001), body mass index (BMI; OR = 4.34, P < 0.001) and serum triglyceride (OR = 13.59, P = 0.037) on multivariate analysis. The median CAP for steatosis grades S0, S1, S2, and S3 were 184 dB/m, 305 dB/m, 320 dB/m, and 324 dB/m, respectively. The areas under receiver operating characteristics curves (AUROC) for estimation of steatosis grades ≥ S1, S2, and S3 were 0.97, 0.86, and 0.75, respectively. The optimal CAP cutoffs for estimation of steatosis grades ≥ S1, S2, and S3 were 263 dB/m, 281 dB/m, and 283 dB/m, respectively. Among non-obese patients, the AUROC for estimation of steatosis grades ≥ S1 and S2 were 0.99 and 0.99, respectively. Among obese patients, the AUROC for estimation of steatosis grades ≥ S1, S2, and S3 were 0.92, 0.64, and 0.58, respectively.

Conclusions: CAP is excellent for the detection of significant hepatic steatosis. However, its accuracy is impaired by an increased BMI, and it is less accurate to distinguish between the different grades of hepatic steatosis.

Keywords: controlled attenuation parameter; hepatic steatosis; nonalcoholic fatty liver disease; noninvasive assessment.