We identified consistent patterns of diffuse anoxic brain injury that are identifiable on MRI performed within 14 days of ischaemic insult. There is considerable overlap between the patterns of cortical injury and the deep nuclei. DWI is superior to FLAIR when imaging is performed within 5 day, outside that time-window, the two methods are equivocal for diagnosing DABI. FLAIR is typically superior to TSE-T2 sequences for diagnosis, although we did note that brainstem evaluation with T2 was frequently equivalent to FLAIR. Given the lack of isolated brainstem injury in these cases, this is probably of limited clinical utility. The ADC map was of limited utility for assessing the basal ganglia or cerebellum, however on imaging performed within 5 days, it was useful in assessing the cerebral cortex.

While the primary aim of this study was not to assess clinical outcome, we note that the high mortality rate of the disorder makes it difficult to identify patterns predictive of a good clinical outcome. Previous studies, in concordance with our own findings, have noted a poor clinical outcome for most patterns of anoxic brain injury [1, 3, 4].  Some studies [1, 3] noted improved outcomes for patients with a watershed pattern of injury. Given the small numbers of patients in our study (n=2) with this pattern, no conclusion can be drawn regarding improved outcome in that cohort.

In summary, our experience is that anoxic brain injury follows a predictable pattern on MRI and that it is important for radiologists to be cognisant of, the timing of imaging and its relationship to the utility of certain sequences and the most commonly affected areas and their appearance on MRI.