Nadir Oxygen Delivery During Pediatric Cardiopulmonary Bypass and Postoperative Acute Kidney Injury: A Pilot Cohort...

Background: Among infants undergoing cardiac surgery, AKI may affect as many as 40% of this population and carries an unfavorable prognosis. A range of contributors has been implicated, including early age, cyanotic physiology, and extended time on cardiopulmonary bypass (CPB). Nadir-indexed oxygen delivery (DO₂) is an important determinant for early detection of hypoperfusion and anaerobic metabolism during CPB. This study aimed to explore the relationship between nadir DO₂ during pediatric CPB and postoperative acute kidney injury. Methods: Between 1 October 2024 and 1 December 2024, we enrolled 40 children who underwent cardiac surgery with CPB [median age 6 months (IQR 4-8), median weight 5.5 kg (IQR 5-7 kg)] into an observational cohort. DO₂ was tracked intraoperatively for every patient, and pre- as well as intraoperative variables were examined for independent links to AKI. Postoperative outcomes were then compared between patients who did and did not develop AKI. Results: In our patient population (n = 40), 16 patients (40%) developed AKI according to the pRIFLE criteria (75% risk; 18.7% injury; 6.2% failure; no patient was in the loss or end-stage renal disease categories). Lower nadir DO₂ values were associated with postoperative AKI in this exploratory pilot cohort. ROC analysis identified an exploratory nadir DO₂ threshold of 360 mL/min/m² (sensitivity 70%, specificity 80%) associated with postoperative AKI. Because this threshold was both derived and evaluated within the same cohort without validation, it should be regarded only as a preliminary, hypothesis-generating observation with no implication of clinical applicability and requires validation in larger independent cohorts. Conclusions: In this exploratory pilot cohort, lower nadir DO₂ during cardiopulmonary bypass was independently associated with postoperative AKI. An exploratory ROC-derived threshold of 360 mL/min/m² was identified; however, this threshold should be considered hypothesis-generating rather than clinically validated. The consistency of findings across both dichotomous and continuous DO₂ analyses strengthens the robustness of the observed association, although larger prospective studies are required to externally validate this threshold, which at present carries no implication of clinical applicability.