Work

Post-cardiopulmonary bypass colloid osmotic pressure and creatinine-defined acute kidney injury in infants undergoing cardiac surgery: an age-modified association

Year: 2026

Type: article

Abstract: Background Acute kidney injury (AKI) is common after infant cardiac surgery with cardiopulmonary bypass (CPB). Colloid osmotic pressure (COP) changes during CPB in response to hemodilution, albumin administration, ultrafiltration, and hemoconcentration, but its association with postoperative AKI in infants remains incompletely defined. Because renal physiology matures rapidly during the first year of life, the clinical meaning of COP may also vary across infancy. Methods We conducted a single-center retrospective cohort study of 230 infants aged ≤12 months and weighing ≤10 kg who underwent cardiac surgery with CPB between May 2020 and September 2022. The primary exposure was COP measured immediately after CPB, and the primary outcome was creatinine-defined postoperative AKI based on KDIGO thresholds within a prespecified 48-hour postoperative window. Multivariable logistic regression evaluated the covariate-adjusted association between post-CPB COP and AKI. Restricted cubic splines assessed dose-response form, and a generalized additive model (GAM) with a tensor-product smooth of COP and age was used to explore age-related heterogeneity. Time-specific analyses additionally evaluated pre-CPB, intra-CPB, and post-CPB COP. Results AKI occurred in 57 infants (24.8%). After adjustment for covariates, for every 1 mmHg increase in post-CPB COP, the risk of AKI increased by 16% (OR = 1.16; 95% CI: 1.04–1.28; P = 0.004). The dose–response relationship was broadly linear. The GAM curve combining age and COP suggested age-related heterogeneity ( P = 0.022); in an exploratory age-stratified analysis, the estimated odds ratio was highest in the group of infants aged 10–12 months (OR per 1 mmHg increase: 1.79; 95% CI: 1.09–3.64; P = 0.048); however, due to the small number of events in this subgroup, these findings should be interpreted as exploratory. There was no significant association between pre-CPB or during-CPB COP and AKI, whereas post-CPB COP exhibited the strongest time-specific signal. Incorporating COP into the clinical model resulted in only a slight, non-statistically significant improvement in discrimination performance (AUC: 0.701 vs. 0.723; P = 0.324); no significant improvement was observed in NRI, and the overall net benefit was minimal. Conclusions Higher post-CPB COP was associated with creatinine-defined AKI after infant cardiac surgery, with the clearest risk information observed at the end of CPB. COP is better interpreted as a peri-CPB physiologic risk marker that should be considered alongside age and perfusion context, rather than as an independent prediction tool or a fixed therapeutic target.

Source: Frontiers in Pediatrics

DOI: 10.3389/fped.2026.1929104