A family history can display a synergistic effect of atherogenic and prothrombotic risk in pregnancy

We have assessed the relationship between being a carrier of genetic factor II variant (20210G>A) and myocardial infarction. The factor II variant has been reported in 1–2% of a anticoagulant clinic population in the United Kingdom (1); it increases the level of factor II (2), and is associated with venous thrombotic disease (1). Our patient is a 24-year old healthy primigravidae with normal BMI, blood group 0Rh+ and who, before pregnancy, used combined oral contraceptives for four to five years. Although the family history was negative for venous thromboembolic disease, she had several relatives who had had a myocardial infarction at an early age. Our patient developed in 29 weeks of gestation a thromboembolic complication in the right popliteal vein which extended above the inguinal ligament. Surgical thrombectomy was successfully performed and she was given an intravenous infusion of unfractionated heparin for seven days maintaining an APTT 1.5–2.5 times normal value. Subsequently a subcutaneous injection twice a day at a dosage of 10.000 IE dalteparin was administered for one month thereafter, aiming to reach above 0.6 and not higher than 1.3 IE/ml plasma three hours post injection measured as anti factor Xa activity determined by a chromogenic substrate assay. After improvement and disappearance of clinical symptoms and earliest one month after the thromboembolic event Fragmin treatment was reduced to give a concentration corresponding to 0.1–0.2 IE anti-Xa units before injection and less than 0.45 IE anti-Xa units per ml plasma three hours post injection. This treatment continued until delivery (3). At time of birth the dose was lowered to 2.500 IE dalteparin twice a day and a full-term healthy infant was delivered by vaginal delivery. Postpartum, she was given warfarin for four months. To date, six years later, she has had no further episodes of venous thrombosis. She has not been pregnant and uses other contraceptives than combined oral contraceptives. In a follow-up program we did thrombophilic screen on days 5–8 in the menstrual cycle which showed normal levels of antithrombin, protein C activity, protein S total/free, but the factor V Leiden mutation in heterozygous form. Because of the history of arterial thromboembolic disease in her family, we also investigated various arterial markers: factor VIII 1.23 IE/ml (ref. values 0.50–1.50), prothrombin 1.26 IE/ml (ref. values 0.80–1.40), PAI-1 3 IE/ml (ref. values<15), vWF 1.02 IE/ml (ref. values 0.50–1.50), homocysteine 19 μmol/L (ref. values 7–22), cardiolipin IgG <5 (ref. values<20) and lupus anticoagulants 1.00 E (ref. values<1.23). In a case-controlled study of patients with the Leiden mutation, factor II 20210G>A transition was found more often in women with myocardial infarction (5.1%) than in healthy women (1.6%) (4). We therefore examined our patient further and found the prothombin gene mutation in homozygous form. Venous thromboembolism (VTE), in association with pregnancy, occurs in around 1/1 000 deliveries and is the main cause of maternal mortality. The condition is also associated with considerable morbidity, about 50–70% of those affected develop symptoms of postphlebitic syndrome secondary to chronic venous insufficiency. It is concluded by our group that surgical thrombectomy offers no advantage over anticoagulant treatment alone for the long-term outcome in cases of ileo-femural venous thrombosis. In all cases in our report there was a high degree of re-canalization of the thrombosed iliac veins and contrary to previous reports long-term symptoms after proximal vein thrombosis seem to be relatively mild in both groups (5). This report highlights another inherited thrombophilia which may contribute to the development of venous thrombosis during pregnancy with natural anticoagulants – e.g. antithrombin, protein C, protein S or factor Leiden in symptomatic families. The original description of the prothrombin 20210A allele, by Poort et al. (2), confirmed that a mutation occurs in 6.2% of subjects with the first episode of VTE and in 2.3% of healthy controls, giving an odds ratio for VTE of 2.8. This finding has been supported by other studies (1, 6, 7). It is less common than the factor V Leiden mutation, which occurs in around 5–10% of a healthy population, in 15–20% of those with a first episode of VTE and in 45% of those with a VTE in pregnancy (8). In contrast, the mutation appears to be rare among Chinese and absent among Japanese and Africans (9). The risk of having a pregnancy-related venous thromboembolism in the presence of prothrombin G20210A was similar to that of having factor V Leiden mutation (10). In a prospective evaluation the presence of prothrombin mutation was associated with a significant increased risk of recurrent VTE, particularly among those with co-inherited factor V Leiden (11). In our center, we screen women with a personal history of venous thromboembolism, women with a first degree relative in whom there has been an episode or women with multiple more distant relatives for the prothrombin 20210G → A mutation. Carriers of the prothrombin mutation have been reported to have higher plasma prothrombin levels. This in itself was shown to be a risk factor for venous thrombosis (2). We found high prothrombin levels in our patient. As yet, only a few homozygous cases seem to be due to the very low prevalence (estimated 0.015%). The exact clinical significance of homozygosity for this polymorphism is not understood (12). Arterial thrombosis is rarely associated with these mutations. The prevalence of 0.7% for 20210A allele in the control group increased to 4.3% (p=0.021) among patients with venous thrombosis and to 5.7% (p=0.013) in a selected group with arterial disease (13). Some data suggest that the risk of myocardial infarction is high among carriers of the prothrombin variant, especially among those young females (4). Our case had several relatives who had had a myocardial infarction at an early age, her own mother was operated on by a by-pass operation at the age of 50 and her aunt had a heart attack at the age of 49. It is well recognized that coexistence of factor V Leiden and factor II A20210 mutation significantly increases the rate of recurrent venous thromboembolism (14, 15). Interestingly, Poort et al. reported a family in whom the FV mutation and the prothrombin gene 20210A allele were segregated in family members and observed that thrombosis was found only in persons having both genetic defects (2). Women with a personal or family history of venous or arterial thromboembolism should be screened for the prothrombin mutation in addition to screening other thrombophilias (16). The prothrombin gene variant might contribute to thromboembolic disease in many patients with other prothrombotic defects, such as factor FVR506Q mutation and may influence the age, site, and type of thrombotic manifestation in such patients (17). The risk among women with both the G20210A prothrombin gene mutation and factor V Leiden is disproportionately higher than that among women with only one mutation (18). A novel feature was the calculation of a positive predictive value for each genetic defect, assuming an underlying rate of venous thromboembolism of 0.67 per 1000 pregnancies. The values were 1:5000 for V Leiden, 1:200 for the G20210A prothrombin-gene mutation, and 4.6:100 for the combination of the two defects. These data are invaluable in the management of pregnancies complicated by thrombophilia.