Early-onset delirium after spontaneous intracerebral hemorrhage

Objective This study aimed at identifying the incidence, predictors, and impact on long-term mortality and dementia of early-onset delirium in a cohort of patients with spontaneous intracerebral hemorrhage. Methods We prospectively recruited consecutive patients in the Prognosis of InTra-Cerebral Hemorrhage (PITCH) cohort and analyzed incidence rate of early-onset delirium (i.e. during the first seven days after intracerebral hemorrhage onset) with a competing risk model. We used a multivariable Fine-Gray model to identify baseline predictors, a Cox regression model to study its impact on the long-term mortality risk, and a Fine-Gray model adjusted for pre-specified confounders to analyze its impact on new-onset dementia. Results The study population consisted of 248 patients (mean age 70 years, 54% males). Early-onset delirium incidence rate was 29.8% (95% confidence interval (CI) 24.3-35.6). Multivariate analysis showed that pre-existing dementia (subhazard ratio (SHR) 2.08, 95%CI 1.32-3.32, p = 0.002), heavy alcohol intake (SHR 1.79, 95%CI 1.13-2.82, p = 0.013), and intracerebral hemorrhage lobar location (SHR 1.56, 95%CI 1.01-2.42, p = 0.049) independently predicted early-onset delirium. Median follow-up was 9.5 years. Early-onset delirium was associated with higher mortality rates during the first five years of follow-up (HR 1.52, 95%CI 1.00-2.31, p = 0.049), but did not predict new-onset dementia (SHR 1.31, 95%CI 0.60-2.87). Conclusion Early-onset delirium is a frequent complication after intracerebral hemorrhage; it is associated with markers of pre-existing brain vulnerability and with higher mortality risk, but not with higher dementia rates during long-term follow-up.