Abstract Background Numerous studies indicated that there exists a relationship between methylenetetrahydrofolate reductase (MTHFR) C667T polymorphism and diabetic nephropathy (DN) susceptibility; nonetheless, available proof reported from individual studies has not been consistent, so we performed an updated meta-analysis to evaluate the relationship between MTHFR C667T variant and DN. Materials and methods Relevant studies published before February 2022 were searched from the electronic databases PubMed, Embase, Scopus, Chinese Biology Medicine and the Chinese National Knowledge Infrastructure. The strength of the association was examined by odds ratio (OR) with 95% confidence interval (CI). Results The findings illustrated that there was a significant relationship between the polymorphism of C677T and DN compared with that to DM controls in allele (OR = 1.59, 95% CI = 1.39–1.82), dominant (OR = 1.76, 95% CI = 1.47–2.11) and recessive (OR = 1.85, 95% CI = 1.56–2.20) models in all populations. Moreover, as compared with the healthy controls, a significant relationship between C677T and DN was found in three genetic comparison models (allele: OR = 1.81, 95% CI = 1.43–2.29; dominant: OR = 2.09, 95% CI = 1.54–2.85; recessive: OR = 2.02, 95% CI = 1.51–2.70). Furthermore, stratifying data by race, diabetes duration and whether in Hardy–Weinberg equilibrium revealed substantially augmented vulnerability to DN in all subgroups. Conclusion The current meta-analysis highlighted conclusive results for the robust association between C677T polymorphisms and DN susceptibility.