Neurocognitive impairment across the continuum of critical illness: exploration of acute insults, functional risk factors, and clinical monitoring tools.
Abstract
Critically ill patients of all ages suffer from high burden of neurocognitive impairment during (i.e. delirium) and following (i.e. long term cognitive impairment) critical illness that is associated with worse patient and healthy system outcomes. Ischemia has emerged as a plausible mechanism given the high prevalence of hypotension and shock, ischemic injury on neuroimaging, and impairment of cerebral autoregulation in these patients. However, the burden of ischemic insults during critical illness and mechanisms responsible for these insults are poorly described. Furthermore, while baseline impairment in cerebrovascular function can render patients more vulnerable to ischemia, such baseline functional assessments in patients with high risk of critical illness have not been considered. Finally, the operational limitations of existing cognitive batteries preclude routine linkage of ischemic insults and baseline impairment in cerebrovascular function with neurocognitive outcomes. In this work we carried out three studies to address these knowledge gaps. In the first study, we showed that critically ill patients with respiratory failure or shock experience deviations in cerebral blood flow velocity consistent with ischemia or hyperemia for 17-24% of the observation time. These deviations occurred irrespective of the state of cerebral autoregulation and were not explained by concurrent changes in blood pressure or CO2. These deviations represent a plausible ischemic insult that may explain high prevalence of ischemic injury in previous neuroimaging and histopathologic studies, and warrants further research to understand the underlying mechanism and link with neurocognitive outcomes. In the second study, we showed that hemodialysis patients have baseline impairment in cerebrovascular function prior to onset of critical illness, which may render them more vulnerable to ischemic injury during critical illness as a result of perturbation in cerebral blood flow shown in our first study. In our third study, we optimized an existing comprehensive web-based cognitive battery for monitoring cognitive outcomes in ICU patients, which should enable future linkage of ischemic insults and baseline impairment in cerebrovascular function from our first two studies with neurocognitive outcomes, as well enable routine clinical monitoring of cognitive recovery in ICU survivors.