Antithrombotic strategies and outcomes in neonates and infants with cardiac shunts: a systematic review and meta-analysis.

Background: Cardiac shunt thrombosis in neonates and infants remains a concern for shunt failure and mortality. The optimal strategy for thromboprophylaxis remains unknown. Objectives: This systematic review aims to characterize antithrombotic strategies and outcomes in neonates and infants with a cardiac shunt. Methods: MEDLINE, Embase, and Cochrane CENTRAL were searched from inception through July 2024 for studies reporting shunt thrombosis prevalence among infants who received a cardiac shunt. We estimated the pooled prevalence of shunt thrombosis using random-effects meta-analysis. In the subgroup analysis, we evaluated the effects of shunt type and antithrombotic strategies on shunt thrombosis prevalence. Results: A total of 39 studies (29 retrospective, 10 prospective) were included, totaling 4735 patients. The most common shunt type was the modified Blalock-Taussig (n = 2224, 47%). Mortality related to shunt thrombosis occurred in 102 (26.2%) patients with shunt thrombosis. The most common antithrombotic agents in the acute postoperative setting were unfractionated heparin (UFH; n = 1452, 30.7%) and aspirin (n = 1413, 29.3%). The pooled prevalence of shunt thrombosis was 8.4% (95% CI, 6.5%-10.4%) and varied among antithrombotic agents: aspirin 7.4% (95% CI, 4.0%-11.4%), UFH 3.8% (95% CI, 0%-12.3%), or UFH followed by aspirin 6.3% (95% CI, 3.6%-9.4%). Conclusions: This systematic review of nearly 5000 neonates and infants reveals a high rate of mortality associated with shunt thrombosis. Collaborative prospective studies are warranted to evaluate antithrombotic regimen-outcome relationships and prognostic factors for shunt thrombosis and bleeding outcomes in these children.