Key points are not available for this paper at this time.
Abstract Background Delirium affects a quarter of patients after acute stroke and predicts poorer outcomes. Our aim was to determine whether either qualitative assessment or quantitative assessments of the regional atrophy obtained from routinely performed computed tomography (CT) brain imaging could identify those most at risk of developing delirium. Methods We recruited 95 patients with acute stroke (age ≥65) over one year. Follow-up for delirium and cognition was performed at 1, 3, 5, 7, 14, 21, 28 days, 4 months and 12 months. All participants underwent routine CT brain (Toshiba 64-slice or 128-slice scanner). White matter disease and atrophy were rated qualitatively (mild, moderate, severe). Atrophy in multiple specific areas was measured quantitatively. Results Twenty-six (27%) developed delirium during the 12 months of follow-up. On univariable analysis, delirium was associated with increasing age, being female, less independent in pre-stroke activities of daily living, pre-existing cognitive impairment, increasing stroke severity, having had a total anterior circulation stroke and global cerebral atrophy on brain CT. Multivariable analysis demonstrated that only global cerebral atrophy, being female and having a more severe stroke predicted delirium. This model accounted for between 38% and 55% of the variance in delirium. For quantitative CT analysis, on univariable analysis, delirium was associated with atrophy in several specific brain areas. On multivariable analysis, only NIHSS (for every one point increase OR 1.23, 95% CI 1.06-1.43; p=0.006)) and cistern ambiens ratio (OR 1.41, 95% CI 1.48-4.96; p=0.028) were significantly associated. This model accounted for between 35.1% and 51.2% of delirium variance. Conclusion Clinical variables together with either qualitative atrophy assessment or cistern ambiens ratio on routine CT brain could identify stroke patients most at risk of delirium and to stratify patients in clinical trials of delirium prevention and treatment.
Barugh et al. (Wed,) studied this question.