104 - Acute kidney injury in neonates with hypoxic ischemic encephalopathy based on serum creatinine decline compared to KDIGO criteria

Poster Number: 104
Publication Number: 104.125

Haejun C. Ahn, Stanford University School of Medicine, Palo Alto, CA, United States; Adam Frymoyer, Stanford University School of Medicine, San Carlos, CA, United States; Derek B. Boothroyd, Stanford University, Palo Alto, CA, United States; Scott Sutherland, Lucile Packard Children's Hospital Stanford, Stanford, CA, United States; Valerie Y. Chock, Stanford University School of Medicine, Sunnyvale, CA, United States

Background: Neonates with hypoxic ischemic encephalopathy (HIE) receiving therapeutic hypothermia after birth asphyxia are at risk for developing acute kidney injury (AKI).  The Kidney Disease Improving Global Outcomes (KDIGO) criteria are widely accepted as the consensus definition for AKI and identify events based primarily on a rise in serum creatinine (SCr).  This definition, however, has been challenging to apply in neonates given the physiologic decline in SCr seen during the first week of life.  Gupta et al. have proposed alternative neonatal criteria which incorporate the rate of SCr decline.

Objective: To compare the rate of AKI based on KDIGO and Gupta criteria in neonates with HIE receiving therapeutic hypothermia and to examine associations with mortality and morbidities.

Design/Methods: A retrospective chart review was performed of neonates with moderate to severe HIE who underwent therapeutic hypothermia from 2008-2017 in a quaternary neonatal intensive care unit. Neonates were assessed for AKI in the first 7 days after birth by KDIGO criteria (SCr rise 0.3 mg/dL in 48 hr, 1.5x increase in SCr, or urine output (UOP)≤1ml/kg/hr over 24 hr) and by Gupta criteria (SCr decline ≤33%, ≤40%, or ≤46% from birth on days 3, 5, and 7, respectively). Mortality, length of stay among survivors, duration of respiratory support among survivors, MRI severity of injury, and Bayley Scales of Infant Development-III scores among available survivors were compared between groups.

Results: Among 151 neonates, 48 (32%) met AKI-KIDGO criteria, 42 (28%) additional neonates met AKI-Gupta criteria, and 61 (40%) did not meet criteria for either definition (Fig. 1). Demographic characteristics were similar between groups (Table 1). Neonates who only met AKI-Gupta criteria had a longer length of stay compared to neonates who did not meet either AKI criteria (p=0.0037). Neonates who met AKI-KDIGO criteria also had a longer length of stay but additionally had a longer duration of respiratory support, higher MRI severity of injury, and worse Bayley motor scores compared to neonates who did not meet either AKI criteria; there was no significant difference in outcomes between AKI-KDIGO and AKI-Gupta (Table 2). Conclusion(s): Neonates with KDIGO-defined AKI have significantly worse outcomes than neonates without AKI. However, KDIGO may miss neonates with impaired kidney function in the first week after HIE. Incorporating the rate of SCr decline into the neonatal AKI definition may increase identification of clinically relevant kidney injury.
HA_CV.pdf
Table 1. Demographics and Perinatal Characteristics