The association between retinopathy in type 2 diabetes [diabetic retinopathy (DR)] and the C677T polymorphism in the methylenetetrahydrofolate reductase (MTHFR) gene has been investigated in several case-control studies. These studies rendered contradictory results, some indicating that the polymorphism is associated with the risk of developing DR whereas others concluded there is no association. To shed light on these inconclusive findings, a meta-analysis of all available studies relating the C677T polymorphism to the risk of developing DR was conducted. Four out of five identified studies included populations of East Asian descent, and only one involved samples from European descent (Caucasians). Overall, the meta-analysis suggested large heterogeneity between studies (p = 0.08, I(2) = 52%) and marginal association between C677T transition and the risk of developing DR: random effects odds ratio (OR) = 1.39 [95% CI (1.05, 1.83)]. The sensitivity analysis [exclusion of one East Asian study with the controls not in Hardy-Weinberg equilibrium (HWE)] showed no heterogeneity (p = 0.25, I(2) = 27%) and no significant association: fixed effects OR = 1.22 [95% CI (0.99, 1.51)] and random effects OR = 1.24 [95% CI (0.96, 1.60)]. The sub-group analysis for the East Asian population produced a significant association: fixed effects OR = 1.48 [95% CI (1.20, 1.83)] and random effects OR = 1.52 [95% CI (1.14, 2.03)]. However, sensitivity analysis in East Asians revealed that the association is marginal: fixed effects OR = 1.33 [95% CI (1.04, 1.70)] and random effects OR = 1.36 [95% CI (1.01, 1.83)]. There is a source of bias in the selected studies: the largest studies failed to show association while the smallest study claimed an association. The above findings reinforce the need for larger and more rigourous studies in this area.