Consistency Analysis of Imaging and Histological Diagnosis of Non-alcoholic Fatty Liver Disease

Background Nonalcoholic fatty liver disease (NAFLD) is a common digestive system disease, which is often diagnosed by imaging methods in clinical work. At present, there is a lack of research on the effectiveness evaluation of imaging diagnostic methods with histology as the gold standard. Objective To study the consistency of imaging and histological diagnosis of NAFLD, and to analyze the influencing factors of gallstone disease (GD) complicated with NAFLD. Methods From January 2021 to July 2022, 53 patients with GD who underwent simple cholecystectomy and liver biopsy were selected from the Department of Hepatobiliary surgery, Nanjing Hospital affiliated to Nanjing University of traditional Chinese Medicine. The demographic characteristics, BMI, systolic blood pressure (SBP) and diastolic blood pressure (DBP) of the subjects were collected, and the laboratory indexes, imaging and histological results were collected. The consistency of imaging and histological diagnosis of NAFLD was evaluated by Kappa consistency test. According to the histological diagnostic criteria, the subjects were divided into NAFLD group (n=15) and non-NAFLD group (n=38). Multivariate Logistic regression analysis was used to explore the influencing factors of NAFLD in patients with GD. Results The imaging detection rate of NAFLD in GD patients was 20.7% (11/53), which was lower than that of histology (28.3%) (15/53) (Kappa=0.404, P=0.001). The sensitivity of imaging diagnosis was 60.0% (9/15), the specificity was 94.7% (36/38), and the rate of missed diagnosis was 40.0% (6/15). There were 8 cases of nonalcoholic fatty liver (NAFL), 7 cases of nonalcoholic steatohepatitis (NASH) and no NASH associated cirrhosis in 15 patients with histological manifestation of NAFLD. The proportion of fibrosis in the NASH group was higher than that in the NAFL group. There were significant differences in the levels of BMI, SBP, fasting blood glucose (FPG) and aspartate aminotransferase (AST) between NAFLD group and non-NAFLD group (P<0.05). Multivariate Logistic regression analysis showed that BMI (OR=1.500, 95%CI=1.084-2.075, P=0.014) and FPG (OR=2.163, 95%CI=1.246-3.756, P=0.006) were the influencing factors of GD patients with NAFLD. Conclusion Compared with histological diagnosis, the missed diagnosis rate of imaging diagnosis of NAFLD is higher, and the incidence of NAFLD may be underestimated by imaging diagnosis as a means of epidemiological investigation. High BMI and high FPG are the main risk factors of GD complicated with NAFLD.