ABSTRACT
BACKGROUND: Factor V Leiden deficiency is associated with an increased risk of initial venous thromboembolism. The prevalence of Factor V Leiden deficiency varies with ethnicity and age of onset of initial venous thromboembolism. If Factor V Leiden deficiency predicts recurrent venous thromboembolism, putting affected patients on extended anticoagulation therapy may be beneficial. This study evaluated the cost effectiveness of testing patients for Factor V Leiden deficiency after initial venous thromboembolism, taking into account costs and complications associated with recurrent venous thromboembolism, compared with ongoing anticoagulation therapy.
POPULATION STUDIED: This study was a decision analysis that assumed a base case of a 35-year-old woman with initial venous thromboembolism. Subpopulations for sensitivity analyses were based on ethnicity, prevalence of Factor V Leiden deficiency, age at onset, precipitating factors for venous thromboembolism, length of therapeutic intervention, and morbidities attributed to anticoagulation. The risk for recurrent venous thromboembolism in patients homozygous for Factor V Leiden deficiency is high; this study focused on heterozygotes, a population whose recurrence risk is less well defined.
STUDY DESIGN AND VALIDITY: This decision analysis used sound methods and the sensitivity analyses were appropriate. It is unclear whether a systematic literature review was performed. A pivotal factor was the assumption of an increased risk of recurrence in patients with Factor V Leiden deficiency as compared with patients without the deficiency. The authors based this assumption on an 8-year study that showed an increased risk of recurrence of 2.4 (95% confidence interval [CI], 1.3–4.5, n = 41). All of the recurrences were detected in the first 3 years. However, other studies, of somewhat shorter duration, demonstrated risk ratios of 1.1 (95% CI, 0.7–1.6, n = 112) within 4 years1 and 0.5 (95% CI, 0.1–1.8, n = 37) within 2 years.2
OUTCOMES MEASURED: The primary outcomes reported were the risk of recurrence of deep venous thrombosis, morbidity associated with therapeutic intervention, and the cost effectiveness of 3 different treatment strategies.
RESULTS: All Factor V Leiden-deficient patients were assumed to have a 7.4% per-year risk of recurrence. Various models were constructed based on the duration of that increased risk. The base case assumed a 0% recurrent deep venous thrombosis risk after 3 years; the modified-base case strategy assumed that patients returned to the population average of 2.3% per year after 3 years; and the constant rate model assumed a continued 7.4% per-year risk of recurrent deep venous thrombosis. In all models, testing and treating for life was most expensive. The base and modified-base models demonstrated testing and treatment for 3 years were the most cost effective. If a patient population has a risk of major hemorrhage of more than 1.9% per year, a low prevalence of Factor V Leiden deficiency, a clear precipitant for venous thromboembolism, or a recurrence risk for venous thromboembolism from Factor V Leiden deficiency of less than 1.9, then testing is not indicated.
If the assumptions made in this study are true, then patients at low risk for long-term anticoagulation should be tested for Factor V Leiden deficiency, and if positive, treated for 3 years, pending longer-term studies. However, studies have not clearly defined an increased risk for recurrent venous thromboembolism in patients with Factor V Leiden deficiency. Until the true relative risk is ascertained, routine screening of patients with initial idiopathic venous thromboembolism for Factor V Leiden deficiency should not be used to determine length of anticoagulation.