Oliguria and Acute Kidney Injury in Critically Ill Children: Implications for Diagnosis and Outcomes

Objectives: Consensus definitions for acute kidney injury are based on changes in serum creatinine and urine output. Although the creatinine criteria have been widely applied, the contribution of the urine output criteria remains poorly understood. We evaluated these criteria individually and collectively to determine their impact on the diagnosis and outcome of severe acute kidney injury. Design and Setting: Post hoc analysis of Assessment of Worldwide Acute Kidney Injury, Renal Angina, and Epidemiology studya prospective international observational multicenter study. Patients: Critically ill children enrolled in Assessment of Worldwide Acute Kidney Injury, Renal Angina and, Epidemiology database. Measurement: To assess the differential impact of creatinine and urine output criteria on severe acute kidney injury (Kidney Disease: Improving Global Outcomes stage = 2). Patients were divided into four cohorts: no-severe acute kidney injury, severe acute kidney injury by creatinine criteria only, severe acute kidney injury by urine output criteria only, and severe acute kidney injury by both creatinine and urine output criteria. Results: Severe acute kidney injury occurred in 496 of 3,318 children (14.9%); 343 (69.2%) were creatinine criteria only, 90 (18.1%) were urine output criteria only, and 63 (12.7%) were both creatinine and urine output criteria. Twenty-eight-day mortality for creatinine criteria only and urine output criteria only patients was similar (6.7% vs 7.8%) and higher than those without severe acute kidney injury (2.9%; p < 0.01). Both creatinine and urine output criteria patients had higher mortality than creatinine criteria only and urine output criteria only patients (38.1%; p < 0.001). Compared with patients without severe acute kidney injury, the relative risk of receiving dialysis increased from 9.1 (95% CI, 3.9-21.2) in creatinine criteria only, to 28.2 (95% CI, 11.8-67.7) in urine output criteria only, to 165.7 (95% CI, 86.3-318.2) in both creatinine and urine output criteria (p < 0.01). Conclusions: Nearly one in five critically ill children with acute kidney injury do not experience increase in serum creatinine. These acute kidney injury events, which are only identified by urine output criteria, are associated with comparably poor outcomes as those diagnosed by changes in creatinine. Children meeting both criteria had worse outcomes than those meeting only one. We suggest oliguria represents a risk factor for poorer outcomes among children who develop acute kidney injury. Application of both the creatinine and urine output criteria leads to a more comprehensive epidemiologic assessment of acute kidney injury and identifies a subset of children with acute kidney injury who are at higher risk for morbidity and mortality.