Asanguineous Cardiopulmonary Bypass in Infants: Impact on Postoperative Mortality and Morbidity

Abstract Background We routinely start cardiopulmonary bypass (CPB) for pediatric congenital heart surgery without homologous blood, due to circuit miniaturization, and blood-saving measures. Blood transfusion is applied if hemoglobin concentration falls under 8 g/dL, or it is postponed to after coming off bypass or after operation. How this strategy impacts on postoperative mortality and morbidity, in infants weighing ≤ 7 kg? Methods Six-hundred fifteen open-heart procedures performed from January 2014 to June 2018 were selected. One-hundred sixty-three patients (26.5%) were transfused on CPB (group 1), while 452 (73.5%) patients were not transfused on CPB (group 2). Operative risk and complexity were similar in both groups. Postoperative mortality and morbidity were compared. Multiple logistic regression was used to detect factors independently associated with outcome. Results Observed mortality in nontransfused group (0.7% = 3/452) was significantly lower than expected (4.2% = 19/452): p = 0.0007, and much lower than in transfused group (6.7% = 11/163): p < 0.0001. CPB transfusion (p = 0.001) was independently associated with mortality, either acting as the sole factor or in combination with the Society of Thoracic Surgeons morbidity score (p = 0.013). Patients not transfused during CPB required less frequently vasoactive inotropic drugs (p = 0.011) and duration of their mechanical ventilation was shorter (93 ± 134 hours) than for transfused patients (142 ± 170 hours): p = 0.0003. CPB transfusion was an independent determinant factor for morbidity (p = 0.05), together with body weight (p < 0.0001), vasoactive inotropic score (p < 0.0001), CPB duration (p = 0.001), and postoperative transfusion (p = 0.009). Conclusion The strategy of transfusion-free CPB course, feasible in most patients ≤ 7kg, was associated with improved outcome. Asanguineous priming of CPB circuit should become standard, even in neonates and infants.