Elevated serum creatinine over the first week of life and mortality risk in extremely preterm neonates: a secondary analysis of the Preterm Erythropoietin Neuroprotection Trial (PENUT)

Authors:

Abstract
The neonatal-modified Kidney Disease: Improving Global Outcomes acute kidney injury (KDIGO-AKI) serum creatinine (SCr) definition is widely used to characterize neonatal AKI. In extremely preterm neonates born < 28 weeks' gestational age (GA), this definition has limitations. We evaluated whether GA- and postnatal age-specific SCr thresholds during the first postnatal week could identify additional neonates at risk for mortality. In a secondary analysis of prospectively collected data from the Preterm Erythropoietin Neuroprotection Trial, daily 95th percentile SCr thresholds were derived over the first week of life for two GA groups (24-25 and 26-27 weeks) using neonates without KDIGO-AKI. Neonates were classified sequentially into three cohorts: KDIGO-AKI; 95th-AKI, defined by at least one SCr exceeding the daily 95th percentile thresholds in the absence of KDIGO-AKI; and controls with no AKI. Time-varying Cox proportional hazards models assessed associations with in-hospital mortality. Among 918 extremely preterm neonates, 106 (12%) had KDIGO-AKI, 72 (8%) had 95th-AKI, and 740 (81%) had no AKI in the first postnatal week. In-hospital mortality was higher with 95th-AKI (18%) or KDIGO-AKI (18%) than with no AKI (9%) (p = 0.003). After adjusting for confounding factors, 95th-AKI was associated with an increased in-hospital mortality (adjusted HR 2.16, 95% CI 1.12-4.18), but KDIGO-AKI was not (adjusted HR 0.99, 95% CI 0.49-2.02). Extremely preterm neonates without KDIGO-AKI but with elevated SCr in the first postnatal week had an increased risk of in-hospital mortality. GA- and postnatal age-specific SCr thresholds may improve early detection of at-risk neonates. ClinicalTrials.gov Identifier: NCT01378273.

This publication has 38 references indexed in Scilit: