BACKGROUND & AIMS Limited evidence supports a role for physical activity (PA) in nonalcoholic fatty liver disease (NAFLD) independent of visceral obesity or/and insulin resistance. We investigated the association between PA and NAFLD while considering visceral adipose tissue (VAT) and insulin resistance in a large general population.

METHODS Nonalcoholic fatty liver disease was diagnosed based on typical ultrasonographic findings. A detailed PA questionnaire included type, frequency, duration and length of time that the subject engaged in PA. PA was determined by a metabolic equivalent. VAT was evaluated by computed tomography taken at the umbilicus level.

RESULTS A total of 3718 subjects were enrolled in the analysis. After adjusting for age, gender, body mass index, smoking, hypertension, diabetes, soft drink and coffee consumption, the total PA was inversely associated with NAFLD [4th quartile (highest activities), odds ratio (OR) 0.68, 95% confidence interval (CI) 0.54-0.85; 3rd quartile, OR 0.74, 95% CI 0.59-0.93 vs. 1st quartile (lowest activities), P for trend <0.001]. After further adjusting for VAT or insulin resistance or both (P for trend = 0.027, 0.001 and 0.040 respectively), this relationship was slightly attenuated but remained statistically significant. The leisure-time PA and weighted PA also had inverse associations with NAFLD independent of the VAT or insulin resistance or both.

CONCLUSIONS This study showed an inverse association between various types of PA and the prevalence of NAFLD in a dose-dependent manner that was independent of visceral obesity and insulin resistance. This finding suggests a beneficial association between PA and NAFLD.