Link to Pubmed [PMID] – 25650012
Neurocrit Care 2015 Oct;23(2):243-52
BACKGROUND: To determine the prevalence, type, and significance of brain damage in critically ill patients with a primary non-neurological diagnosis developing acute brain dysfunction.
METHODS: This retrospective cohort study was performed at the Johns Hopkins University School of Medicine, an academic tertiary care hospital. Medical records were reviewed of 479 consecutive ICU patients who underwent brain magnetic resonance imaging (MRI) over a 2-year period. Patients were selected for analysis if MRI was obtained to evaluate an acute onset of brain dysfunction (altered mental status, seizures, and/or focal neurological deficit). Subjects with a history of a central nervous system disorder were excluded. The principal clinical endpoint was Glasgow Outcome Scale (GOS) assessed at discharge. MRI-defined brain abnormalities were classified according to type and location. Factors associated with MRI-defined abnormalities were assessed in uni- and multivariable models.
RESULTS: 146 patients met inclusion criteria (mean age 54 ± 7 years). Brain damage was detected in 130 patients (89 %). The most prevalent lesions were white matter hyperintensities (104/146, 71 %) and acute cerebral infarcts (59/146, 40 %). In a multivariable model, lesions on brain MRI were independently associated with unfavorable outcome (GOS1-3 in 71 % of patients with lesions vs. 44 % in those without, p = 0.007). No adverse events occurred in relation to transport and MRI scanning.
CONCLUSIONS: In critically ill patients without known neurological disease who have acute brain dysfunction, MRI reveals an unexpectedly high burden of underlying brain damage, which is associated with unfavorable outcome. The results indicate that brain damage could be an important and under-recognized factor contributing to critical illness brain dysfunction.