A novel score based on CAP accurately predicts hepatic steatosis in individuals with nonalcoholic fatty liver disease: a derivation and independent validation study

Zi-Ming An,Qiao-Hong Liu,Xin-Jian Ye,Qian Zhang,Hua-Fu Pei,Xin Xin,Jie Yuan,Qian Huang,Kun Liu,Fang Lu,Zhi-Han Yan,Yu Zhao,Yi-Yang Hu,Ming-Hua Zheng,Qin Feng

DOI: https://doi.org/10.14309/ctg.0000000000000680

2024-01-20

Clinical and Translational Gastroenterology

Abstract:Background and Aims: In non-alcoholic fatty liver disease, the diagnostic efficacy of CAP was not very accurate in evaluating liver fat content. The aim of this study was to develop a score, based on CAP and conventional clinical parameters, to improve the diagnostic performance of CAP regarding liver fat content. Methods: A total of 373 participants from two independent Chinese cohorts were included and divided into derivation (n = 191), internal validation (n = 75), and external validation (n = 107) cohorts. Based on the significant difference index between the two-group defined by the MRI-PDFF in derivation cohort, the optimal model (CBST) was screened by the number of parameters and the area under the receiver operating characteristic curve (AUROC). In the internal and external validation cohorts, the AUROC and corresponding 95% confidence intervals (CIs) were used to compare the diagnostic performance of CBST with that of CAP. Results: We constructed the CBST = -14.27962+0.05431*CAP-0.14266*BMI+0.01715*AST. When MRI-PDFF was ≥ 20%, ≥ 10%, and ≥ 5%, the AUROC for CBST was 0.77 (95% CI, 0.70-0.83), 0.89 (95% CI, 0.83-0.94), and 0.93 (95% CI, 0.88-0.98), which was higher than that for CAP respectively. In the internal validation cohort, the AUROC for CBST was 0.80 (95% CI, 0.70-0.90), 0.95 (95% CI, 0.91-1.00), and 0.98 (95% CI, 0.94-1.00). The optimal thresholds of CBST were -0.5345, -1.7404 and -1.9959 for detecting MRI-PDFF ≥ 20%, ≥ 10%, and ≥ 5%. Conclusion: The CBST score can accurately evaluate liver steatosis and is superior to the CAP.

gastroenterology & hepatology