Urine neutrophil gelatinase-associated lipocalin predicts kidney support therapy duration and liberation in critically ill children.

Abstract

Background: Early identification of kidney recovery in critically ill children and young adults undergoing continuous kidney support therapy (CKST) is essential to optimize care and minimize complications. Urinary neutrophil gelatinase-associated lipocalin (uNGAL) is a biomarker of acute kidney injury, but its utility in predicting CKST duration and guiding timely liberation remains unclear.

Methods: We retrospectively analyzed urinary uNGAL levels in pediatric intensive care unit (PICU) patients (aged 2 months to 25 years) who received CKST between July 2018 and April 2024. We evaluated two outcomes: (1) the ability of peak uNGAL levels during the first four days of CKST to predict prolonged therapy (>7 days), and (2) the performance of uNGAL measured during KST liberation attempts in predicting successful liberation.

Results: Among 57 patients, early peak uNGAL predicted CKST duration >7 days with good accuracy (AUC-ROC 0.85 [95% CI, 0.73-0.97], optimal cutoff 2600 ng/mL). uNGAL also showed excellent performance in predicting successful KST liberation (AUC-ROC 0.95 [95% CI, 0.89-1.00], optimal cutoff 900 ng/mL).

Conclusions: uNGAL may be useful for predicting prolonged CKST duration and guiding KST liberation in critically ill children. Larger prospective studies are needed to confirm its role in personalized CKST management.

Impact statement: Peak uNGAL levels within the first four days of CKST are strongly associated with treatment duration >7 days, showing good predictive accuracy. uNGAL demonstrates excellent performance in identifying patients likely to achieve successful KST liberation. uNGAL measured during KST may support clinical decision-making by providing timing insights, potentially optimizing treatment duration in critically ill pediatric patients.