Background

Neonatal acute kidney injury (AKI) is common and is associated with poor outcomes. New criteria for the diagnosis of AKI were introduced based on the increase in serum creatinine (SCr) levels and/or reduction of urine output (UOP). Yet, there is no generally accepted opinion so far, which criteria (whether SCr, UOP, or their combination) are the most appropriate to diagnose neonatal AKI.

Methods

The retrospective study included 195 prematurely born neonates who fulfilled all inclusion criteria (with at least two SCr measurements). In all the neonates included in the study, AKI was diagnosed using three different definitions: (1) SCr criteria (an increase in SCr values of ≥0.3 mg/dl), (2) UOP criteria (UOP < 1.5 ml/kg/h), and (3) SCr + UOP criteria.

Results

Out of all of the patients the study included, 85 (44%) were diagnosed with AKI. The neonates who had AKI had a significantly lower gestational age, birth weight, and Apgar score, longer duration of mechanical ventilation, and a higher mortality rate. SCr + UOP criteria showed higher sensitivity for prediction of death compared to SCr or UOP alone (p = 0.0008, 95% CI 0.040-0.154, and p = 0.0038, 95% CI 0.024-0.125, respectively). If only SCr or only UOP criterion are used, they fail to identify AKI in 61 and 67%, respectively. AKI was an independent risk factor for death (OR 7.4875; CI 3.1887-17.5816).

Conclusions

Similar to other studies, our data showed that neonates with AKI have worse outcome. Neonatal AKI defined based on SCr + UOP criteria is a better predictor of death than neonatal AKI defined based only on the SCr or UOP criteria. Also, by using SCr + UOP criteria for diagnosing neonatal AKI, more patients with AKI are recruited than when only one of those criteria is used.