The alanine/valine (A/V) gene polymorphism of 5, 10-methylenetetrahydrofolate reductase (MTHFR), one of the key enzymes catalyzing remethylation of homocysteine, has been reported and the VV genotype is associated with increased plasma homocysteine levels as a result of the reduced activity and increased thermolability of this enzyme. Although previous studies have suggested that the VV genotype is a risk factor for arterial occlusive disease, whether the VV genotype is a risk factor for venous thrombosis is still controversial. Here we screened 72 Japanese patients with deep venous thrombosis (DVT) and 85 controls for this mutation, and we measured plasma levels of homocysteine to determine whether the thermolabile variant with the VV genotype is a risk factor for DVT in a Japanese population. Of the 72 patients with DVT, 10 (13.9%) were found to be homozygous for the VV genotype, and in 6 (7.0%) of 85, control individuals and the difference was not significant (odds ratio=2.12, 95% CI=0.73-6.16, p=0.19). When we divided the DVT patients into subgroups, with and without predisposition of thrombophilia, including deficiencies of proteins C and S, plasminogen, and lupus anticoagulant, the prevalence of the VV genotype in DVT patients with predisposition was significantly higher than that of the normal controls (odds ratio=5.99, 95% CI=1. 56-22.96, p=0.01). However, the prevalence of the VV genotype in DVT patients without predisposition was not significantly different from that of the normal controls (odds ratio=1.20, 95% CI=0.32-4.47, p=0. 75). The plasma homocysteine levels in patients with DVT (11.6+/-5.2 nmol/ml) was not significantly different from that of the control subjects (11.6+/-3.7 nmol/ml). Individuals with the VV genotype showed higher plasma homocysteine levels (15.4+/-6.9 nmol/ml) than did individuals with the AV genotype (11.2+/-3.7 nmol/ml, p=0.009) or in individuals with the AA genotype (11.1+/-4.2 nmol/ml, p=0.004). Serum folate and vitamin B12 levels were not correlated with the plasma homocysteine levels. In conclusion, even though homozygosity for the VV genotype of the MTHFR gene was associated with higher plasma homocysteine levels, we found no association between plasma levels of homocysteine and DVT or between the genotype of the MTHFR gene and the DVT incidence. However, we found that the VV genotype of the MTHFR gene is a risk factor for DVT only when combined with the predisposition of thrombophilia.