Abstract Aim Reflux esophagitis is associated with metabolic dysfunction. Recently, fatty liver has been redefined as metabolic dysfunction‐associated fatty liver disease (MAFLD). We investigated the impact of MAFLD and its subtypes on the incidence of reflux esophagitis. Methods This multicenter, observational cohort study enrolled 9100 consecutive health‐check examinees who underwent esophagogastroduodenoscopy and ultrasonography. All patients were classified into the MAFLD or non‐MAFLD group. Based on the Asian cut‐off value for body mass index (BMI), the MAFLD group was further classified into the lean/normal‐weight (BMI <23 kg/m 2 ) and overweight/obese (BMI ≥23 kg/m 2 ) subgroups. The impact of MAFLD and its subtypes on the cumulative incidence of reflux esophagitis was evaluated using multivariable Cox proportional hazards regression analysis. Results MAFLD was diagnosed in 26.5% (2416/9100) of patients. Multivariable Cox proportional hazards regression analysis indicated that MAFLD (hazard ratio [HR] 1.2183; 95% confidence interval [CI] 1.0954–1.3550; p = 0.0003), hiatal hernia, and aging were independent risk factors for reflux esophagitis. Stratification analysis indicated that cumulative incidence of reflux esophagitis among patients with MAFLD was significantly higher in the lean/normal‐weight than in the overweight/obese group (HR 1.3274; 95% CI 1.0043–1.7547; p = 0.0466). Among various metabolic factors, visceral adiposity was the only independent metabolic risk factor for reflux esophagitis (HR 2.8331; 95% CI 1.0201–7.8691; p = 0.0457) in the lean/normal‐weight MAFLD group. Conclusions MAFLD, in particular lean/normal‐weight MAFLD, is independent risk factor for reflux esophagitis. Furthermore, visceral adiposity was identified as the most strong metabolic risk factor for reflux esophagitis in lean/normal‐weight patients with MAFLD.