Background: Previous studies support an association between elevated total homocysteine (tHcy) levels and increased all-cause mortality. However, few prospective studies have examined this association in hypertensive patients, and/or tested any effect modification by the methylene tetrahydrofolate reductase (MTHFR) C677T genotype. Methods: This was a post hoc analysis of the China Stroke Primary Prevention Trial. SerumtHcy and folate were measured at baseline. Individual MTHFR C677T genotype (CC, CT, and TT) was determined. Evidence for death included death certificates or home visits. Cumulative hazards of all-causemortality by tHcy quartiles were estimated using the Kaplan-Meier method, and group differences were compared by log-rank tests. Hazard ratios (HRs) and 95% confidence intervals were estimated by Cox proportional-hazard regression models, adjusting for age, sex, baseline folate, vitamin B12, blood pressure, body mass index, smoking and alcohol drinking status, study center, total cholesterol, triglycerides, high-density lipoprotein cholesterol, fasting glucose, creatinine, and treatment group. Potential effect modification by the MTHFR genotype on the relationship between tHcy and all-cause mortality was tested. Results: The analyses included 20,424 hypertensive patients (41% males) without a history of myocardial infarction or stroke. Baseline mean age (SD) was 60 +/- 7.5 years and mean (SD) serum tHcy was 14.5 +/- 8.4 mu mol/L. After a mean follow-up period of 4.5 years, there were 612 (3%) all-cause deaths. Kaplan-Meier survival curves revealed a graded relationship between tHcy quartiles and all-cause mortality. The HRs, using the lowest quartile as the reference, were 1.2, 1.2, and 1.5 in Q2, Q3, and Q4, respectively. A linear trend test, using natural log-transformed tHcy, resulted in an HR of 1.5 (95% confidence interval 1.2-1.9, P<.001) after adjustment for lifestyle and health-related variables. Whereas the MTHFR genotype alone had little effect on mortality, it significantly modified the tHcy-mortality association, which was much stronger in the CC/CT genotype than in the TT genotype (P for interaction < 0.05). Conclusions: Among Chinese hypertensive patients without cardiovascular comorbidities, elevated tHcy was a significant risk marker for death from all causes, and the association was subject to effect modification by MTHFR genotypes. If confirmed that tHcy and MTHFR genotypes may serve as useful biomarkers for mortality risk assessment and targeted intervention.