Moderately elevated plasma homocysteine levels have been established as an independent risk factor for atherosclerosis and its complications, including cerebrovascular disease. A common mutation (C677T) in the gene encoding for the enzyme methylenetetrahydrofolate reductase (MTHFR) has been linked to increased plasma homocysteine levels in homozygous carriers, particularly in the presence of low folate levels. However, the results of most of the previous studies suggest that the C677T MTHFR mutation is not a significant risk factor for arterial disease. This discrepancy might, at least partly, be due to the fact that plasma homocysteine levels are influenced by several other factors, including age, gender, renal function, and vitamin status. We investigated the relation between plasma homocysteine levels, the C677T MTHFR mutation, and these other factors in a population of 96 patients with transient ischemic attacks or minor strokes and in 96 age- and sex-matched healthy control subjects. We further tested the value of a multivariate model for the prediction of plasma homocysteine levels under particular consideration of the MTHFR mutation status. In the patients, plasma homocysteine levels were significantly higher than in the healthy control subjects. With regard to the MTHFR mutation, the distribution of the C/C, C/T, and T/T genotypes was not significantly different between patients and healthy control subjects. Univariate (linear regression) analysis revealed significant (positive) correlations between plasma homocysteine levels on the one hand and age and creatinine on the other, the latter particularly in subjects with creatinine levels in the upper quartile. Significant (negative) correlations were found between plasma homocysteine levels, vitamin B12, and folate levels. However, these relations could much better be expressed by means of a multiplicative regression model. T/T subjects exhibited slightly higher homocysteine levels than C/C and C/T subjects; however, the differences between the 3 genotypes were not significant. Multivariate (stepwise regression) analysis revealed age, vitamin B12 levels, folate levels, and creatinine levels as significant independent variables influencing plasma homocysteine levels, whereas the MTHFR mutation status and gender were removed from the model. Considering all 192 subjects, only 28.8% of the variance of plasma homocysteine levels could be accounted for by the model. However, in homozygous carriers of the MTHFR mutation, the predictive power of the model is very high, explaining 76.1% of the variance of plasma homocysteine levels. According to our results, the C677T mutation does not constitute a major risk factor for transient ischemic attack or minor stroke, even under consideration of other possibly confounding factors that are known to affect plasma homocysteine levels. However, it is possible to predict plasma homocysteine levels in homozygous carriers of the mutation with high accuracy. The knowledge of the MTHFR mutation status may therefore help to identify subjects at high risk for hyperhomocysteinemia.