The 20210A Allele of the Prothrombin Gene Is Frequent in Young Women With Unexplained Spinal Cord Infarction

To the Editor:

Spinal cord infarction is a severe event: because almost the entire motor output and sensory input systems are concentrated in a small cross-sectional area, small localized vascular lesions can produce devastating consequences far beyond the damage. Its rarity is classically explained by the fact that the anterior and posterior spinal arteries are not usually involved by atherosclerosis. Most infarctions are caused by ischemia secondary to distant vascular occlusions, and only occasionally by angeiitis or emboli. They are facilitated by the unique arterial system that irrigates the spinal cord and typically occur in a vascular watershed region between the artery of Adamkiewicz arising from the lower aorta and the anterior spinal artery arising from the vertebral arteries.

We studied nine women, ages 22 to 41 years, who had developed an acute anterior spinal artery syndrome. Computed tomography scanning and magnetic resonance imaging showed aspects of spinal cord infarction. A vascular malformation of the spinal cord had been secondarily ruled out by selective spinal angiography in six of them. No criteria for systemic arteritis, aortic thrombosis/dissection, or emboli could be found. Seven of the women were current smokers (15 to 25 cigarettes per day), all were oral-contraceptive users (35 to 50 μg ethinyl estradiol; second generation compounds in five, third generation in three), and they were nonobese women (body mass indexes ranging from 22.3 to 26.4). The patients had no personal history of thromboembolic disease despite pregnancies in six of them. One of the women had known familial history of phlebitis.

After obtaining informed consent, screening for the prothrombin variant caused by a G to A transition at nucleotide 20210 in the 3′ untranslated region of the prothrombin gene (PT 20210A allele) was performed by HindIII cleavage of a 345-bp fragment amplified by polymerase chain reaction (PCR) using a mutagenic primer as previously described by Poort et al.1 Genetic analysis of the FV Leiden Arg 506 to Gln mutation (1691 G → A) was determined by PCR and Mnl I restriction analysis of PCR-amplified genomic factor V DNA fragments according to Bertina et al.2 The common homozygous 677C → T polymorphism in the methylenetetrahydrofolate reductase (MTHFR) gene (MTHFR TT genotype) was performed according to Frosst et al.3 Antithrombin, protein C, and protein S activities were assayed using commercially available kits from Diagnostica Stago (Asnières, France) (Stachrom antithrombin, Staclot Protein C, Staclot Protein S). Lupus anticoagulant was detected according to the recommendations of the Subcommittee on Lupus Anticoagulant/Antiphopholipid Antibody of the Scientific and Standardisation Committee of the International Society on Thrombosis and Haemostasis.4 Anticardiolipin IgG and IgM antibodies and anti–β2-glycoprotein I IgG antibodies were assayed using commercial kits (Quanta Lite ACA IgM-IgG; Inova Inc, San Diego, CA and Varelisa β2GP1; Elias, Freiburg, Germany).

We found that four patients were heterozygous for the PT 20210A allele. One of them, the youngest of all our patients, was also heterozygous for the FV Leiden mutation and had a familial history of venous thrombosis. Our older patient with the PT 20210A allele was also homozygous for the 677C → T polymorphism in the MTHFR gene. These four patients were current smokers and oral-contraceptive users (third-generation pill in the one patient with both positive genetic markers). No biological abnormality was found in the remaining five patients.

By comparison, in the southern region of France in contiguity with the Mediterranean sea, data obtained in 100 age-matched healthy women with no thrombotic history (control group) showed a 1% incidence of the PT 20210A allele, a 2% incidence of the FV Leiden mutation, and a 14% incidence of the MTHFR TT genotype. Local data obtained in 100 age-matched, current smokers and oral-contraceptive user outpatients with history of venous thromboembolism (venous thrombosis group) showed a 4% incidence of the PT 20210A allele, a 9% incidence of the FV mutation, and a 22% incidence of the MTHFR TT genotype. The PT 20210A allele was more frequent in patients than in the control group or than in the venous thrombosis group (Fisher exact test:P < .01).

In conclusion, these data indicate that the PT 20210A allele is frequently found in the rare women with unexplained spinal cord infarction and suspicion of a localized arterial thrombotic episode. Smoking and oral-contraceptive use may be clinical cofactors of the disease. In young individuals who smoke and use oral contraceptives, the impact of the PT 20210A allele on the onset of arterial thrombotic episodes should be studied. The PT 20210 allele has already been described to increase the risk of myocardial infarction among non-Mediterranean women ages 18 to 44 years and the relative risk was higher in smoking carriers.5 It has also previously been shown that factor V Leiden increases the risk of myocardial infarction in young women⁶: the case of our youngest patient favors the view that individuals with combined genetic defects may be at higher risk than those with a single gene defect. This is further strengthened by the case of our oldest patient: the mutation Ala677 → Val in the MTHFR gene has been reported to be a risk factor for arterial disease.7 Young women often smoke and use oral contraceptives; finally, because of the significant frequency of the PT 20210A allele in asymptomatic women, but due to the clinical rarity of spinal cord infarctions, future questions should concern the identification of the spectrum of genetic (acquired) abnormalities that may cosegrate (be associated) with the PT 20210A allele to increase the penetrance of clinical arterial thrombotic manifestations localized to the spinal cord.