How strong is the association between maternal thrombophilia and adverse pregnancy outcome? A systematic review.

OBJECTIVE: To determine whether inherited and acquired thrombophilias are associated with adverse obstetric complications. STUDY
DESIGN: A systematic review; studies where women with adverse obstetric complications were tested for one or more acquired and inherited thrombophilias were included. MAIN OUTCOME MEASURES: Prevalence of thrombophilia in women with severe pre-eclampsia/eclampsia, severe placental abruption, intrauterine growth restriction or unexplained stillbirth.
RESULTS: Compared with controls, placental abruption was more often associated with homozygous and heterozygous factor V Leiden mutation, heterozygous G20210A prothrombin gene mutation, homocysteinaemia, activated protein C resistance or anticardiolipin IgG antibodies. Women with pre-eclampsia/eclampsia were more likely to have heterozygous factor V Leiden mutation, heterozygous G20210A prothrombin gene mutation, homozygous MTHFR C677T mutation, protein C deficiency, protein S deficiency or activated protein C resistance compared with controls. Unexplained stillbirth, when compared with controls, was more often associated with heterozygous factor V Leiden mutation, protein S deficiency, activated protein C resistance, anticardiolipin IgG antibodies or lupus anticoagulant. Women with intrauterine growth restriction had a higher prevalence of heterozygous G20210A prothrombin gene mutation, homozygous MTHFR C677T gene mutation, protein S deficiency or anticardiolipin IgG antibodies than controls. There was wide heterogeneity in the prevalence of thrombophilia between the studies.
CONCLUSIONS: Women with adverse pregnancy outcome are more likely to have a positive thrombophilia screen but studies published so far are too small to adequately assess the true size of this association. Screening for thrombophilia should not become standard practice until clear evidence emerges that thromboprophylaxis during pregnancy improves perinatal outcome. Further research into the link between the observed association, causality and heterogeneity is required.