Optimizing hemostasis to prevent bleeding and reduce transfusion requirements in neonatal and infant cardiac surgery with cardiopulmonary bypass: a systematic review and network meta-analysis.

BACKGROUND

Effective hemostatic management remained a critical challenge in neonatal and infant cardiac surgery with cardiopulmonary bypass (CPB), and a systematic review and network meta-analysis (NMA) had not been conducted to evaluate optimal strategy selection for this vulnerable population.

METHODS

We systematically searched PubMed, Web of Science, Cochrane Library, and Embase from inception to December 3, 2025. We use bivariate analysis and NMA with random effects. We use the surface beneath the cumulative ranking curve (SUCRA) to display the order of interventions.

RESULTS

Of the 8,280 records screened, 27 studies involving 4,556 patients were included. Aprotinin (APR) was identified as the optimal hemostatic strategy for reducing 24-hour blood loss (SUCRA=99.15%), platelet (PLT) (SUCRA=82.19%) and fresh frozen plasma (FFP) (SUCRA=84.82%) transfusion requirements, re-sternotomy for hemostasis (SUCRA=72.85%), and risks of thrombosis (SUCRA=79.13%) and renal dysfunction (SUCRA=72.61%). ε-aminocaproic acid (EACA) was found to be the most effective strategy for reducing red blood cell (RBC) transfusion needs (SUCRA=69.04%). Tranexamic acid (TXA) emerged as the optimal intervention for reducing mortality (SUCRA=84.77%).

CONCLUSION

Antifibrinolytic agents may demonstrated significant hemostatic efficacy and a generally acceptable safety profile in neonatal and infant cardiac surgery with CPB. Among these agents, while APR showed the highest efficacy, its use is constrained by regulatory restrictions and evidence limitations; therefore, it should be used with caution. TXA remains a practical first-line alternative. In the available network, blood component therapies did not emerge as consistently superior strategies compared with other hemostatic interventions. For 24-hour blood loss, APR was associated with lower blood loss than FFP, FC, and PCC. However, because the relevant comparisons were largely against active comparators rather than placebo or no hemostatic intervention, the efficacy of blood component therapies relative to no blood component therapy remains uncertain. Future large-scale RCTs are needed to further validate these findings.