Epidemiology of neonatal acute kidney injury after cardiac surgery without cardiopulmonary bypass

Abstract Background The purpose of this Neonatal and Pediatric Heart and Renal Outcomes Network (NEPHRON) study was to describe the epidemiology and outcomes of CS-AKI after cardiac surgery without cardiopulmonary bypass (non-CPB). Methods We performed a retrospective study of neonates (≤30 days) who underwent non-CPB cardiac surgery at 22 centers affiliated with the Pediatric Cardiac Critical Care Consortium. CS-AKI was defined using the modified Kidney Disease: Improving Global Outcomes (KDIGO) serum creatinine (SCr) and urine output criteria from postoperative day (POD) 0-6. CS-AKI by SCr was further subclassified into transient (resolved by POD3) and persistent/late (≥3 days). Multivariable regression analyses were used to determine risk factors for CS-AKI, and associations with outcomes of ventilation hours and cardiac intensive care unit length of stay (LOS). Results 582 neonates [median age at surgery: 9 days (IQR:5-15 days), 25% functional single ventricle] were included. CS-AKI occurred in 38.3%: rate and severity varied across centers. Aggregate daily CS-AKI prevalence peaked on POD1 (17.1%). No stage of CS-AKI was associated with ventilation hours or LOS. Persistent/late CS-AKI occurred in 48 (8%). Prostaglandin use and single ventricle surgery were associated with persistent/late CS-AKI. Higher baseline SCr, but not persistent/late CS-AKI was associated with longer ventilation duration and ICU LOS after adjusting for confounders. Conclusions KDIGO-defined CS-AKI occurred commonly in neonates undergoing non-CPB cardiac surgery. However, most CS-AKI was transient, and no CS-AKI classification was associated with worse outcomes. Further work is needed to determine the CS-AKI definition that best associates with outcomes in this cohort.