Abstract: SA-PO0683
Optimizing Early AKI Definitions in Preterm Neonates Using the PENUT Trial
Session Information
- Pediatric Nephrology: Transplantation, Hypertension, AKI, Genetics, and Developmental Diseases
November 08, 2025 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Pediatric Nephrology
- 1900 Pediatric Nephrology
Authors
- Starr, Michelle C., Indiana University School of Medicine, Indianapolis, Indiana, United States
- Griffin, Russell, University of Alabama at Birmingham Health System Authority, Birmingham, Alabama, United States
- Harer, Matthew W., University of Wisconsin-Madison, Madison, Wisconsin, United States
- Menon, Shina, Stanford Medicine, Stanford, California, United States
- Gist, Katja M., Children's Hospital Colorado, Aurora, Colorado, United States
- Allegaert, Karel, Katholieke Universiteit Leuven, Leuven, Flanders, Belgium
- Schwartz, Samantha Rose, University of Wisconsin-Madison, Madison, Wisconsin, United States
- Baeta Baptista, Rute, Universidade Nova de Lisboa Medical School, Lisbon, Lisbon, Portugal
- Selewski, David T., Medical University of South Carolina, Charleston, South Carolina, United States
- Askenazi, David J., University of Alabama at Birmingham Health System Authority, Birmingham, Alabama, United States
Background
Acute kidney injury (AKI) is common in preterm neonates. Current definitions are adapted from adults and may not be optimal for preterm neonates. Optimized definitions are essential to improve outcome prediction and guide interventions.
Methods
Using serum creatinine (SCr) from the Preterm Erythropoietin Neuroprotection (PENUT) Trial, a RCT in neonates 24-27 weeks gestation, we evaluated 11 SCr-AKI definitions (Table1). Performance was assessed by ability to predict mortality. Youden’s index (YI) was calculated to find SCr thresholds most predictive of mortality (Day 0-2, 3-7 and 8-14).
Results
This study included 923 infants; AKI incidence varied (6.4% [SCr peak >1.5mg/dL] to 58.4% [SCr >75th% for age]). SCr percentiles had the highest sensitivity for predicting mortality (0.65, 0.5 and 0.38 for 75th%, 90th%, and 95th%, respectively), but uniformly low positive predictive value (Table1).
When evaluating optimal cutoffs, the Day 0-2 fluid-adjusted SCr threshold was 0.87mg/dL. Compared to those below 0.87mg/dL, those with SCr higher than threshold had a higher risk of death (LR+1.83[1.32-2.34,p=.0014]). Infants with an absolute or fluid-adjusted Day 3-7 SCr (1.16mg/dL and 1.05mg/dL) had a higher risk of death (LR+1.85[1.18-2.52,p=.013 and LR+1.76[1.32-2.34,p=.0014] respectively (Table2).
Conclusion
Current neonatal SCr-AKI definitions have limited ability to predict death. Improved definitions may require thresholds at different timepoints (fluid-adjusted or absolute SCr on Days 3-7) plus elements with high specificity (biomarkers, urine output, clinical factors or a combination). Optimizing the definition of neonatal AKI remains a balance between predicting adverse outcomes and timely identification to allow intervention.
Funding
- NIDDK Support