23 - Associations between clinical risk factors, placental pathology and MRI injury patterns in the HEAL Trial

Poster Number: 23
Publication Number: 23.475

Background: Hypoxic ischemic encephalopathy (HIE) is a heterogenous disorder. The distribution of brain injury identified by MRI varies with the underlying cause and timing of the inciting insult. Although clinical risk factors and placental pathology may offer insights into the timing or mechanism of injury, it remains to be determined whether these are associated with the acuity or pattern of brain injury observed on MRI.

Objective: To characterize associations between clinical risk factors, placental pathology and brain injury on MRI.

Design/Methods: This is a posthoc analysis of data from the HEAL trial. Clinical risk factors were collected at the time of initial hospitalization. Placental pathology reports generated as part of clinical care were centrally reviewed by a placental pathologist. Placental abnormalities were characterized as acute only, chronic only, both acute and chronic, or none. MRI scans were collected prospectively using a harmonized protocol and scored centrally. Pattern of injury was characterized as central (involving basal ganglia or thalamus +/- perirolandic cortex), peripheral (bilateral watershed white matter +/- cortex), global (near total involvement of cerebrum), punctate white matter lesions (focal injury to periventricular white matter) or atypical (other). Acuity of injury was restricted to the infants who underwent MRI <= 7 days, with acute brain injury defined by the presence of reduced diffusivity; subacute by MRI signal abnormalities without diffusion restriction; chronic by volume loss. We used Fisher’s exact test (categorical data) or ANOVA (continuous) to determine whether the distribution of injury acuity or pattern differed in the presence of clinical risk factors or placental abnormalities.

Results: MRI data from 414 of 473 infants (88%) were acquired at <= 7 days and included here. 270 also had placental pathology. Among the clinical factors, 10-min APGAR < 5 and delivery room intubation were associated with higher rates of acute brain injury (Table 1). pH and base deficit showed linear relations, with the lowest pH (and largest base deficit) observed in infants with acute or acute + subacute brain injury followed by subacute (only) and then no brain injury. Last, among those with atypical injury, we observed a higher rate of no placental abnormalities (Table 2).

Conclusion(s): Clinical factors reflecting metabolic and respiratory compensation were associated with acuity of brain injury on MRI. While these factors may provide insight posthoc into the timing of injury, they are unlikely to be sufficiently discriminatory for guiding clinical management.