Objective. To evaluate the prevalence and outcome of acute kidney injury in paediatric intensive care units using the modified RIFLE score (pRIFLE).
Design. Historical cohort study.
Setting. A paediatric intensive care unit in a regional Hong Kong hospital.
Patients. All paediatric patients aged 1 month to 18 years admitted to a local paediatric intensive care unit in the years 2005 to 2007.
Main outcome measures. For every paediatric intensive care unit admission, acute kidney injury was classified according to the pRIFLE criteria (“R” for risk, “I” for injury, “F” for failure, “L” for loss, and “E” for end-stage). Prevalence and outcome of acute kidney injury were therefore categorised according to the pRIFLE staging.
Results. A total of 140 such patient admissions constituted the study population. The point prevalence of acute kidney injury in these patients on admission was 46% (n=59), whilst 56% (n=78) endured acute kidney injury at some time during their paediatric intensive care unit stay. Worsening of pRIFLE grading during their intensive care unit admission was observed in 20% of the patients who had no acute kidney injury on admission, in 30% of those who had an initial “R” grade, and in 40% of those who had an initial “I” grade of acute kidney injury. Overall mortality in this cohort was 12%, which was significantly higher among patients with acute kidney injury. Having acute kidney injury of grade “F” on admission to the paediatric intensive care unit was an independent predictor of mortality (hazard ratio=5.94; 95% confidence interval, 1.06-33.36; P=0.043).
Conclusion. Among critically ill paediatric patients, the pRIFLE score serves as a suitable classification of acute kidney injury when stratified according to clinical severity. It also provides prognostic information on mortality and renal outcomes.