Risk for venous thrombosis related to antiphospholipid antibodies in systemic lupus erythematosus--a meta-analysis.

OBJECTIVE

To describe the relative risk for venous thrombosis (VT) associated with antiphospholipid antibodies (aPL) in systemic lupus erythematosus (SLE).

DESIGN

Systematic review and meta-analysis of 26 articles that examined the association between aPL and VT in SLE.

SETTING

Mostly secondary and tertiary referral centres.

PATIENTS

2249 patients with SLE, 1120 tested for LA (lupus anticoagulant) and 1563 tested for aCL (anticardiolipin antibodies).

MAIN OUTCOME MEASURES

A summary of study characteristics and a critical appraisal of study quality were done. Two statistical combinations of 18 primary studies that examined the association of VT and LA and of 14 studies that examined the association of VT and aCL were performed to estimate the risk for VT associated with aPL.

RESULTS

The odds ratios of the risk of VT related to the LA summarized from 18 studies were 5.61 [95% CI; 3.80-8.27] overall, 6.32 [CI; 3.71-10.78] for deep venous thrombosis and pulmonary embolism, 11.6 [3.65-36.91] for recurrent venous thrombosis after the first event. The odds ratios of the risk of VT related to aCL summarized from 14 studies were 2.17 [95% CI; 1.51-3.11] overall, 2.50 [CI; 1.51-4.14] for deep venous thrombosis and pulmonary embolism, 3.91 [1.14-13.38] for recurrent venous thrombosis after the first event.

CONCLUSIONS

Patients with SLE and LA are at approximately six times greater risk for VT than patients without LA, whereas patients with SLE and aCL are approximately two times greater risk for VT than patients without aCL. We have identified important methodologic limitations and differences in study characteristics. Other risk factors for VT have not been thoroughly evaluated in these studies. Further studies are needed that provide an accurate estimate of the absolute risk for aPL related VT.