To the Editor:

Genetic variants leading to a persistent hypercoagulable state may predispose to thrombotic events. A recently discovered G to A mutation at position 1691 of the factor V gene (factor V Leiden mutation), occurring in 3% to 5% of the normal Caucasian population, has emerged as a major genetic risk factor of venous thrombosis.1 The role of this mutation for arterial vascular events, in particular cerebrovascular disease, is still under discussion.2-6Another recently described mutation (G to A at position 20210 in the 3′-untranslated region of the prothrombin gene) has been associated with a threefold increased risk of venous thrombosis in heterozygous carriers of the mutation.7 The mutation is as frequent as 1% to 2% in the population and also seems to represent an important genetic risk factor of venous thrombosis. Currently, the role of the prothrombin variation for arterial vascular disease is unclear. In a recent study, 5.1% heterozygous carriers of the mutation among young women with myocardial infarction were found, compared with 1.6% heterozygous carriers among a control population.8The age-adjusted odds-ratio for MI was 4.0 and was particularly high in the presence of other vascular risk factors, eg, smoking. In another recent study, the G20210A prothrombin mutation was found in 5.1% of 98 patients with coronary heart disease as compared with 1.96% among healthy newborns.9 On the other hand, three studies did not find a significantly increased prevalence of the mutation in patients with cerebrovascular disease.10-12 Only in one of these studies were cerebrovascular risk factors detailed,10 and only one study also analyzed the presence of the factor V Leiden mutation in age- and sex-matched control subjects.12Furthermore, patients were highly selected in one of these studies,11 with 30 of 125 patients suffering from arterial dissection as the cause of cerebral infarction. Another study found no increased prevalence of the prothrombin G20210A mutation in patients with cerebrovascular disease13; however, a synergistic role of the prothrombin mutation and the factor V Leiden mutation in the development of thrombosis was observed. To date, no study has evaluated the effect of the factor V Leiden mutation and the G20210A prothrombin mutation in patients with cerebrovascular disease and age- and sex-matched control subjects under particular consideration of other vascular risk factors in patients with and without the prothrombin G20210A mutation.

We investigated the prevalence of the factor V Leiden mutation and the G20210A prothrombin mutation in 96 patients with transient ischemic attacks (TIA) or minor strokes (MS; 58 men and 38 women; age [x ± s], 64.4 ± 13.1 years; range, 28 to 91 years) and in 96 age- (±10 years) and sex-matched control subjects free of clinically manifest vascular disease. Furthermore, we compared history, clinical data, and the prevalence of risk factors for stroke between patients with and without the G20210A prothrombin mutation. All patients underwent cranial computerized tomography, duplex sonography of carotid and vertebral arteries, and a thorough cardiac investigation (including electrocardiography, transthoracic echocardiography, as well as transesophageal echocardiography in selected cases). In addition, all patients were evaluated with respect to the generally accepted risk factors for stroke and any coexisting diseases, in particular coronary heart disease, peripheral arterial disease, and venous thrombosis. Concerning the presumed etiology of stroke, each patient was assigned to one of the following four groups14: lacunar (n = 23; 24.0%); atherothrombotic (n = 33; 34.4%); cardioembolic (n = 24; 25.0%); or undetermined (n = 16; 16.6%). The prevalence of the 20210 AG genotype was equal in patients and in control subjects, with 5 of 96 subjects in each group (patients: 4 men and 1 woman; mean age, 64 years; control subjects: 3 men and 2 women; mean age, 63 years; odds ratio [OR], 1.0; 95% confidence interval [CI], 0.52 to 1.91). The prevalence of the factor V Leiden mutation was 8 of 96 in the patient group (5 men and 3 women; mean age, 60 years) and 5 of 96 in the control group (3 men and 2 women; mean age, 60 years; OR, 1.65; 95% CI, 0.92 to 2.98). One patient (53-year-old man) and none of the control subjects was homozygous for the factor V Leiden mutation. One patient (67-year-old man) and none of the control subjects carried both mutations. This patient had experienced recurrent cerebrovascular events (starting at the age of 60 years) and one myocardial infarction (at the age of 65 years) but had no history of venous thrombosis. Coronary artery disease had become manifest at the age of 45 years (angina pectoris). However, this patient also had multiple other vascular risk factors (hypertension, hypercholesterolemia, overweight, and cigarette smoking).

Patients exhibiting the prothrombin wild-type genotype and patients carrying the G20210A prothrombin mutation were not different with regard to current age, age at the time of the first cerebrovascular event, or age at the time of the first vascular event (cerebral/cardiac/peripheral arterial system or venous). Recurrent cerebrovascular events were reported by 1 of 5 carriers of the 20210 GA genotype and by 4 of 91 carriers of the wild-type (not significant [NS]). The prevalence of ischemic cardiac disease was 3 of 5 in carriers of the GA genotype and 28 of 91 in carriers of the GG genotype (NS). The prevalence of peripheral arterial disease was 1 of 5 in carriers of the GA genotype and 8 of 91 in carriers of the GG genotype (NS). Two of five patients with the GA genotype and 13 of 91 patients with the GG genotype reported a history of deep-vein thrombosis (with and without pulmonary embolism; NS). Carotid artery disease was present in 1 of 5 of the GA patients and in 11 of 91 of the GG patients. The etiology of the cerebrovascular event in the 5 carriers of the GA genotype was classified as lacunar (n = 1), atherothrombotic (n = 3), and cardioembolic (n = 1). This distribution was not significantly different from that in the patients carrying the wild-type.

Comparing the risk factors for stroke, no significant differences were found between the two groups with respect to cardiac disease, hypertension, diabetes mellitus, hypercholesterolemia, hyperfibrinogenemia, alcohol abuse (history), cigarette smoking (recent), history of other drug abuse, and oral contraceptive use.

In conclusion, our data do not suggest a role for the G20210A mutation as a relevant risk factor for TIA/MS in an unselected group of patients. We did not find evidence for an effect of the factor V Leiden mutation or an increased combined occurrence of both mutations in patients with TIA/MS. In comparison with carriers of the wild-type, the carriers of the 20210 GA genotype did not show distinct clinical features or a distinct risk factor profile. The question whether the G20210A prothrombin mutation, alone or in combination with other vascular risk factors, increases the risk of cerebrovascular disease in specific populations (as it could be shown for myocardial infarction8) remains to be answered in further studies.