Hippocampus and basal ganglia as potential sentinel sites for ischemic pathology after resuscitated cardiac arrest.

AIMS OF THE STUDY: Neurological impairment after resuscitated cardiac arrest (CA) remains a significant unmet medical need. Brain ischemia associated with CA and subsequent reperfusion is evident as two fundamentally different types of damage on neuropathological examination: frank necrosis (involving all cell types) and selective eosinophilic neuronal death (SEND). These types of damage are not only dissimilar in micromorphology, but also differently detectable with clinical brain imaging methods. In a previous study, SEND was reported in most patients surviving the initial CA. This study was undertaken to further characterize and map SEND in an expanded dataset.

METHODS: A cohort of 46 cases was included from an observational study on targeted temperature management (TTM) of resuscitated CA. Six brain and brain stem regions and 21 subregions were examined, and SEND severity was tested for correlation with time to ROSC. Representativity of all regions vis-à-vis global SEND was assessed, to investigate whether any particular region could be used as a "sentinel site" for overall damage.

RESULTS: The thalamus, the CA4 subregion of the hippocampus and the Purkinje cell layer of the cerebellum were the most severely affected subregions. Involvement of the hippocampus, cerebellum, cortex or basal ganglia indicated presence of SEND in other regions. There was a significant correlation between time to ROSC and SEND.

CONCLUSION: There are regional differences in SEND distribution. Cases free of SEND in the hippocampus or basal ganglia are unlikely to have significant SEND in other regions, suggesting that these regions could be used as "sentinel sites" for global SEND in future studies.