Body mass index is the most useful predictive factor for the onset of nonalcoholic fatty liver disease: a community-based retrospective longitudinal cohort study.

BACKGROUND: Nonalcoholic fatty liver disease (NAFLD) can progress to advanced liver disease and non-liver-related diseases. To prevent NAFLD onset, clinicians must be able to easily identify high-risk NAFLD patients so that intervention can begin at an earlier stage. We sought to identify the predictive factors for NAFLD onset.
METHODS: In a community-based, longitudinal design, the records of 6,403 Japanese subjects were reviewed to identify those meeting the criteria for NAFLD onset. Univariate and multivariate logistic regression analyses were used to identify predictive factors for NAFLD onset. The accuracy of different models was evaluated according to their areas under the receiver operating characteristic curves. Comparative risk analysis was performed using the Kaplan-Meier method.
RESULTS: Multivariate analysis of 400 subjects who met the criteria for the onset of NAFLD during the observation period confirmed that body mass index (BMI) at baseline was the most useful predictive factor for NAFLD onset in both sexes. Cutoff levels of BMI for NAFLD onset were estimated at 23 kg/m2 for men and 22.2 kg/m2 for women. The cumulative onset rate of NAFLD was significantly higher in the high BMI group than in the low BMI group in both sexes (P < 0.001).
CONCLUSION: BMI was confirmed as the most useful predictive factor for NAFLD onset in both sexes; its cutoff levels were similar to those recommended by the World Health Organization for helping to prevent metabolic disease. An accurate BMI cutoff level will enable clinicians to identify subjects at risk for NAFLD onset.