Systematic review on efficacy of anticoagulation and antithrombotics in patients with congenital heart diseases

Abstract

Background: Advance in cardiac intervention improved the survival of patients with congenital heart diseases (CHD). However, they may have propensity of thromboembolism and the use of antithrombotic agents was generally based on small studies and consensus opinion.

Objective: To systematically review the current literature on the efficacy and safety of various antithrombotic agents in patients with CHD.

Methods: Studies published in English during the period 1990 – 2012 were identified using keyword search from PubMed, Medline, EMBase, and Cochrane Library. Additional search from reference sections of the articles and clinical trial registry was performed. Data extracted included: type of studies, number of patients, follow-up period during which the patients were on the antithrombotic agents, number of thromboembolic (TE) events, and all, major and minor bleeding events. Event rate as the proportion of events of the patients and event per 100 patients-year were obtained for respective antithrombotic agent in each study. Composite event rate and event per 100 patients-year were estimated after weighting.

Results: Forty studies consisted of 5144 patients were reviewed. Observation period of 8916.6 years was available in 25 studies. Diagnostic categories included: Fontan operation 15, systemic-to-pulmonary artery shunt 7, mechanical valve 8, atrial septal defect occlusion device 2, cyanotic heart 1, mixed 7. Antithrombotic prophylaxis was not used in 13 studies, warfarin in 26, aspirin alone in 22, combined aspirin and dipyridamole in 2. Clopidogrel with concomitant antithrombotic agents was reported in 5 studies. Overall composite TE event rate was 3.9% (95% CI 2.3 – 5.4%) and that of all bleeding rate was 2.8% (95% CI 0 – 5.5%), with 1.4% (95% CI 0.0 – 2.6%) for major and 2.2% (95% CI 0.0 – 4.3%) for minor bleeding. Composite TE rate for no prophylaxis (9.6%; 05% CI 3.7 – 15.5%) was significantly greater than that of warfarin (1.7%; 95% CI 0.1 – 3.3%) and aspirin (1.3%; 95% CI 0.0 – 3.0%). Both TE and all bleeding rate showed no difference between warfarin and aspirin, while major bleeding tended to be higher in warfarin than aspirin(0.9% vs 0.0%, p=0.06). Fontan patients had overall TE rate of 2.7% (95% CI 0.1 – 5.4%). Patients with no prophylaxis (10.2%; 95% CI 9.2 – 18%) had significantly greater TE rate than warfarin (1.4%; 95% CI 0.0 – 0.4%) or aspirin (1.2%; 95% CI 0.0 – 3.0%). All bleeding rate in Fontan patients was 0.5% (95% CI 0.0 – 4.3%). Both TE ad bleeding rates showed no difference between warfarin and aspirin. Overall TE rate for shunt was 7.2% (95% CI 3.7 – 14.3%), being similar between aspirin group and no antithrombotic group. Patients with mechanical valves had TE rate of 7.3% (95% CI 2.9 – 11.6%) and all bleeding rate of 7.2% (95% CI 4.2 – 10.2%). There was no statistical difference between warfarin and APA group. Patients with ASD occlusion device has TE rate of 0.1% (95% CI 0.0 – 0.2%). No bleeding event was reported in the studies.

Conclusion: Patients with congenital heart diseases were at risk of developing thromboembolism which justified the use of anti-thrombotic prophylaxis. Further studies relating the thromboembolic risk profile of patients with CHD to the efficacy of anti-thrombotic agents might help in selection of anti-thrombotic agents.