Preoperative serum albumin as a risk marker for acute kidney injury following pediatric congenital heart surgery
Q D Liu, T Y Gao, J Liu, Y Liu, Y Jin, J P LiuAbstract
Background
Congenital heart disease (CHD) is a common abnormality, with postoperative acute kidney injury (AKI) being a frequent complication. Although routine laboratory tests inform clinical decision-making, indiscriminate testing provides limited risk discrimination and may lead to iatrogenic anemia, underscoring the necessity for accessible predictive biomarkers. Serum albumin (Alb), a routinely measured parameter, serves as an indicator of nutritional status and inflammation and has been demonstrated associated with an increased risk of AKI across various surgical populations. Children with CHD are particularly susceptible to malnutrition and growth challenges. However, the prognostic significance of preoperative Alb levels in predicting AKI within this cohort remains uncertain. This study aims to assess whether Alb levels can predict postoperative AKI in children with CHD.
Methods
This single-center retrospective study included CHD patients (<14 years) undergoing cardiopulmonary bypass (CPB) surgery from 2019 to 2024 with complete preoperative Alb data. The primary outcome was AKI incedence (pediatric Risk, Injury, Failure, Loss, and End-stage renal disease, pRIFLE criteria). Alb was analyzed continuously via multivariable logistic regression with restricted cubic splines (RCS) to assess non-linear association, and categorically by quartile. Subgroup analyses [stratified by age, the society of thoracic surgeons and european association for cardiothoracic surgery congenital heart surgery (STAT) mortality category, preterm and incision type] and sensitivity analyses excluding extreme or missing values were performed. Models adjusted for key confounders such as demographics, STAT, laboratory tests and surgical information.
Results
Among 15182 pediatric patients, AKI incidence was 28.45%. AKI patients had significantly lower Alb versus non-AKI (41.90 vs. 43.20 g/L). Neonates showed markedly lower Alb (32.45 g/L) compared to older children. Multivariable logistic regression demonstrated that each 1 g/L decrease in Alb increased AKI risk by 6% (OR: 0.94, 95% CI: 0.93-0.96). RCS revealed a significant non-linear relationship (P<0.001). Compared with the highest Alb quartile (>44.90 g/L), the lowest quartile (<40.70 g/L) had significantly elevated AKI risk (OR: 1.43, 95% CI: 1.27-1.62). Stratification around the RCS inflection point confirmed a negative correlation on the left (OR: 0.90, 95% CI: 0.88–0.93). Subgroup analyses showed that this association pronounced in neonates (OR: 0.88, 95% CI: 0.80–0.98) and median sternotomy patients (OR: 0.93, 95% CI: 0.91–0.94), with P for interaction = 0.001. Sensitivity analyses confirmed robustness. No significant multicollinearity was detected.
Conclusion
A non-linear relationship was observed between Alb and the risk of AKI, with lower albumin levels conferring a higher risk. This association was particularly pronounced in neonates and in patients undergoing median sternotomy.RCS of the association between Alb and AKI.Main, stratified and sensitivity analysis