Objective : To evaluate the association between patient outcomes and the severity of acute kidney injury (AKI) at renal replacement therapy (RRT) initiation using the pediatric RIFLE criteria (pRIFLE).

Design and setting : Single center, retrospective observational study in a pediatric intensive care unit (PICU).

Patients and methods : Data extraction was performed for the first treatment of RRT in children admitted to the PICU between 2008 and 2018.

Main results : Ninety-four patients required RRT.84% presented with AKI according to the pRIFLE criteria at RRT initiation (10.1% stage “R” (risk), 8.9% “I” (injury), and 81% “F” (failure)).

Mortality was 45.7% with no significant difference between the different degree of AKI according to pRIFLE criteria at RRT initiation. No difference in PICU lengths of stay (LOS), duration of mechanical ventilation, and duration of RRT according to the pRIFLE criteria at RRT initiation. In multivariable logistic regression analysis, non-surgical cardiac disease, an elevated PELOD score and fluid overload at RRT initiation were associated with increased odds of mortality. Increased time spent in stage F (>24h vs early<24H) was associated with longer use of vasoactive support but there was not with mortality, PICU LOS, or duration of mechanical ventilation.

Conclusion : The severity of AKI according to the pRIFLE criteria before RRT initiation could not predict mortality or morbidity. The optimal timing to initiate RRT in children remains unknown and the severity of kidney dysfunction appeared to be important but insufficient by itself to predict the clinical outcome of children requiring RRT.