The effects of metformin administration on liver enzymes and body composition in non-diabetic patients with non-alcoholic fatty liver disease and/or non-alcoholic steatohepatitis: An up-to date systematic review and meta-analysis of randomized controlled trials

[Display omitted] Non-alcoholic fatty liver disease (NAFLD) is the most common liver disease worldwide. One treatment is the use of metformin but its efficacy remains to be established. The present systematic review and meta-analysis aimed to provide a more robust examination of the evidence for the effectiveness of metformin for treating non-diabetic NAFLD patients. An extensive literature search was undertaken using online databases (PubMed, Embase, Scopus, Web of Science and Cochrane Library) to detect randomized controlled trials (RCTs) investigating the effect of metformin administration on liver enzymes and body composition in non-diabetic NAFLD patients up to 10 December 2019. A random-effects or fixed-effect models were performed to pool weighted mean difference (WMD) and 95% confidence intervals (CI). Six RCTs involving 307 individuals were included to the present meta-analysis. Compared to controls, metformin significantly reduced body mass index (BMI) (WMD: -0.77 kg/m2, 95 % CI = [-1.46, -0.07], P = 0.03, I2 = 0.0 %) and serum aspartate aminotransferase (AST) (WMD: -5.94 U/L, 95 % CI = [-11.51, -0.38], P = 0.03, I2 = 67.6 %). Also, body weight (WMD: -2.70 kg, 95 % CI = [-5.49, 0.09], P = 0.05, I2 = 33.7%) was marginally significant and serum alanine transaminase (ALT) (WMD: -5.04 U/L, 95 % CI = [-13.92, 3.84], P = 0.26, I2 = 60.9 %) was not statistically significant affected by metformin administration. There was no evidence of publication bias. In summary, the present study emphasizes the clinical importance of metformin administration for improving liver function and body composition in non-diabetic NAFLD patients. Moreover, the further large-scale and well-designed RCTs are required to confirm these findings.