Attempts to associate methylenetetrahydrofolate reductase (MTHFR) gene C677T polymorphism with hypertension have been extensively evaluated, but mostly in Caucasians. We therefore meta-analysed this polymorphism in association with hypertension and hypertensionin-pregnancy (HIP) among Chinese subjects. A randomeffects model was applied irrespective of between-study heterogeneity, which was evaluated by subgroup and meta-regression analyses. Study quality was assessed in duplicate. Publication bias was weighed using Egger's test and funnel plot. Data on 20 qualified studies totalling 4461 Chinese subjects were meta-analysed. In overall allelic/genotypic models, carriers of 677T or 677TT were consistently at significantly increased risk of developing hypertension and HIP. No between-study heterogeneity was identified for all genetic models, with the exception for contrast of allele 677T versus 677C in hypertension association studies (P ¼ 0.03), and the dominant contrast in HIP association studies (P ¼ 0.025). Both the subgroup and meta-regression analyses indicated that study design was a potential source of between-study heterogeneity. Funnel plot and Egger's test suggested no evidence of publication bias. Taken together, our results supported the notion that carriers of 677T or 677TT were at significantly increased risk of developing hypertension and HIP, but indicated caution in interpreting this result, because the study design made marginal significant contribution to the heterogeneity.