The impact of acute kidney injury (AKI) on brain MRI findings in asphyxiated newborns following therapeutic cooling has not been studied. We hypothesized that hypoxic-ischemic lesion on brain MRI would differ between infants with AKI compared to those without AKI following cooling. The purpose of the study was to determine whether AKI during cooling predicts subsequent brain MRI abnormalities related to hypoxia-ischemia.
Medical records of 88 consecutively cooled infants who had brain MRI were reviewed. All infants had renal function assessed before the start of cooling (baseline); at 24, 48, and 72h through cooling; and then on day 5 or 7 of life. A serum creatinine (SCr) based categorical definition similar to Acute Kidney Injury Network for AKI was used to classify infants according to severity in 3 stages. The MRI images were evaluated for hypoxic-ischemic injury by a neuroradiologist, who was masked to clinical parameters and outcomes. Injuries to both basal ganglia/thalamus and cortex were considered severely abnormal . Multivariate analysis determined if any of the pre-cooling variables (e.g. Apgar score, cord pH and base deficit, neurologic examination, etc) to assess the severity of asphyxia, and the presence of AKI predict the primary outcome of an abnormal MRI.
AKI was found in 34 (39%) of 88 infants with 15, 7, and 12 fulfilling criteria for stage I, II, and III, respectively. Brain MRI abnormalities related to hypoxia-ischemia was present in 50 (59%) infants. In 26 infants (AKI 14, no AKI 12), MRI was severely abnormal.
Abnormal MRI was more frequent in infants from the AKI group (AKI 25 of 34, 73% versus No AKI 25 of 54, 46%, p=0.012, OR 3.2, 95% CI 1.3-8.2). However, severely abnormal MRI was similar between the two groups (p=0.091). Multivariate analysis identified only the AKI (p=0.032, OR 2.9, 95% CI 1.1-7.6), and chest compression for resuscitation to be independently associated with the primary outcome.
Severely abnormal MRI in infants with stage III AKI was not different compared to infants with stage II (stage III AKI 3 of 12, 25% versus stage II AKI 3 of 7, 43%; p= 0.617) or stage I AKI (stage III AKI 3 of 12, 25% versus stage I AKI 8 of 15, 53%; p= 0.238).
AKI is independently associated with the presence of hypoxic-ischemic lesions on post-cooling brain MRI. However, the severity of AKI did not correlate with the severity of brain MRI abnormalities.