Recurrent thrombosis in patients with antiphospholipid antibodies and an initial venous or arterial thromboembolic event: A systematic review and meta-analysis

Background Patients with antiphospholipid antibodies (aPL) and thromboembolism (TE) are at risk for recurrent TE. Few studies, however, distinguish patients based on the initial event. Objectives We performed a systematic review and meta-analysis to investigate patients with aPL and venous TE (VTE), provoked or unprovoked, and patients with arterial TE (ATE). Patients/Methods We conducted searches in PubMed, CINAHL, Cochrane, and EMBASE. Inclusion criteria were prospective trials or cohort studies investigating patients with aPL and ATE or VTE. Excluded studies did not provide estimated recurrence rates, did not specify whether the incident event was ATE or VTE, included patients with multiple events, or included <10 patients. Two-year summary proportions were estimated using a random effects model. Results Ten studies described patients with VTE, 2 with ATE, and 5 with VTE or ATE. The 2-year proportion for recurrent TE in patients with VTE who were taking anticoagulant therapy was 0.054 (95% confidence interval [CI], 0.037-0.079); the 2-year proportion for patients not taking anticoagulant therapy was 0.178 (95% CI, 0.150-0.209). Most studies did not distinguish whether VTE were provoked or unprovoked. The 2-year proportion for recurrent TE in patients with ATE who were taking anticoagulant therapy was 0.220 (95% CI, 0.149-0.311); the 2-year proportion for patients taking antiplatelet therapy was 0.216 (95% CI, 0.177-0.261). Conclusions Patients with aPL and ATE may benefit from a different antithrombotic approach than patients with aPL and VTE. Prospective studies with well-defined cohorts with aPL and TE are necessary to determine optimal antithrombotic strategies.