Factor V Leiden mutation and pregnancy-related venous thromboembolism: what is the exact risk? Results from a meta-analysis.

The magnitude of the association of factor V Leiden mutation with pregnancy-related venous thrombosis remains unclear. Our objective was to undertake a systematic review and a metaanalysis of the literature to estimate precisely the association of factorV Leiden mutation with the risk of first, or recurrent, pregnancy-related venous thromboembolism. Studies published before October 2005 were identified by Medline((R)). Using both fixed and random effect models, odds ratios (OR) with accompanying 95% confidential intervals (CI) were calculated for the factor V Leiden mutation and the clinical end-point (Yusuf-Peto adaptation of the Mantel-Haenszel, DerSimonian and Laird method). We identified 13 studies including 7 cohorts and 6 casecontrol studies relating to factor V Leiden and pregnancy-related venous thrombosis. The results from the cohorts showed a pooled OR of 4.46 (95% CI, 1.82-10.94; 7,879 pooled women), with no evidence of statistical heterogeneity (p = 0.36), for the risk of a first venous thromboembolism during pregnancy or the postpartum period associated with the factor V Leiden mutation. Case-control studies revealed a higher risk (OR 8.6, 95% CI, 5.85-12.63; 1,433 [corrected] pooled women) with significant heterogeneity (p < 0.005). Because of insufficient data, an analysis for the risk of recurrence could not be performed. Our findings emphasize the fact that limited data are available on this topic. This meta-analysis provides clinicians with an estimate of the average risk of a first thrombosis occurring during pregnancy in women carrying the factor V Leiden to assist the management of such women.