Nonalcoholic Fatty Liver Disease Incidence and Remission and Their Predictors During 7 Years of Follow-up Among Finns

CONTEXT: The incidence and remission of nonalcoholic fatty liver disease (NAFLD) are sparsely studied outside Asia. OBJECTIVE: This prospective study aimed to investigate NAFLD incidence and remission, and their predictors among a general Finnish population. METHODS: The applied cohort included 1260 repeatedly studied middle-aged participants with data on liver ultrasound and no excessive alcohol intake. Hepatic steatosis was assessed by liver ultrasound with a 7.2-year study interval. Comprehensive data on health parameters and lifestyle factors were available. RESULTS: At baseline, 1079 participants did not have NAFLD, and during the study period 198 of them developed NAFLD. Of the 181 participants with NAFLD at baseline, 40 achieved NAFLD remission. Taking multicollinearity into account, key predictors for incident NAFLD were baseline age (odds ratio 1.07; 95% CI, 1.02-1.13; P = .009), waist circumference (WC) (2.77, 1.91-4.01 per 1 SD; P < .001), and triglycerides (2.31, 1.53-3.51 per 1 SD; P < .001) and alanine aminotransferase (ALAT) (1.90, 1.20-3.00 per 1 SD; P = .006) concentrations as well as body mass index (BMI) change (4.12, 3.02-5.63 per 1 SD; P < .001). Predictors of NAFLD remission were baseline aspartate aminotransferase (ASAT) concentration (0.23, 0.08-0.67 per 1 SD; P = .007) and WC change (0.38, 0.25-0.59 per 1 SD; P < .001). CONCLUSION: During follow-up, NAFLD developed for every fifth participant without NAFLD at baseline, and one-fifth of those with NAFLD at baseline had achieved NAFLD remission. NAFLD became more prevalent during the follow-up period. From a clinical perspective, key factors predicting NAFLD incidence and remission were BMI and WC change independent of their baseline level.