The present systematic review and meta-analysis was conducted to investigate the effects of soy intake on liver enzymes, lipid profile, anthropometry indices, and oxidative stress in non-alcoholic fatty liver disease (NAFLD). A systematic search was undertaken in PubMed, Embase, Scopus, Web of Science, and Cochrane Library covering up to 10 January 2020. A fixed-effect or random-effects models were applied to pool mean difference (MD) and its 95 % confidence intervals (CI). Four clinical trials comprising 234 participants were included in the meta-analysis. Compared to the controls, alanine aminotransferase (ALT) levels (MD=-7. 53, 95% CI=[-11. 98,-3. 08], P=0. 001, I2=0. 0 %), body weight (MD=-0. 77, 95 % CI=[-1. 38,-0. 16], P=0. 01, I2=36. 9%), and the concentration of serum Malondialdehyde (MDA) (MD=-0. 75, 95% CI=[-1. 29,-0. 21], P=0. 007, I2=63. 6%) were significantly changed following soy intake. Lipid profile was not significantly affected by soy intake. Moreover, no evidence of a significant publication bias was found. The present study suggests lowering effects for soy intake on ALT levels, body weight, and MDA in nonalcoholic liver patients. Therefore, further large-scale and well-designed clinical trials are needed to find conclusive findings.