Objective Obesity and sarcopenia are known to be closely related to nonalcoholic fatty liver disease (NAFLD). We attempted to explore the combined influence of fat and muscle tissue on NAFLD by using visceral fat area to appendicular muscle mass ratio (VAR) as a novel parameter. Material and methods In this cross-sectional study, a total of 3255 adults (1399 men and 1856 women) coming for a health examination were enrolled. NAFLD was diagnosed using ultrasound and VAR was measured by bioelectrical impedance analyzer. Results The prevalence of NAFLD was 46.5% in men and 26.6% in women. VAR differed significantly between subjects with and without NAFLD (4.27 vs. 3.26 in men, 7.89 vs. 5.01 in women, respectively, p < .001). Logistic regression analysis determined VAR as a risk factor for NAFLD, and the multivariable-adjusted odds ratios in the highest VAR quartile was 9.57 (95%CI: 5.98–15.30) for men and 12.37 (95%CI: 6.37–24.05) for women. From the receiver operating characteristic analysis, the area under the curve was 0.767 and 0.834, with the suitable cut-off VAR value of 3.469 and 6.357 for men and women, respectively. To control the influence of obesity, all subjects were stratified according to their BMI. For each BMI group, individuals with VAR above the cut-off value had significant higher prevalence and risk of NAFLD, with odds ratios ranging from 1.76 to 4.75. Conclusions Increased VAR is strongly associated with higher risk of NAFLD in both sexes independent of obesity and can serve as a screening reference for NAFLD.