To explore the potential role of language ability in delirium in patients with intracerebral hemorrhage (ICH).

Non-pharmacologic prevention and management of delirium emphasizes the use of verbal reorientation. However, the effectiveness of such strategies is uncertain, and their use is limited in patients with communication barriers. We sought to gain insights into the potential impact of aphasia on delirium via patients with ICH, who experience a high incidence of both delirium and language dysfunction.

We performed a single-center cohort study on consecutive non-traumatic ICH patients admitted from 2018-2019. Aphasia was classified according to initial recorded NIH Stroke Scale language subscores. Delirium was diagnosed according to DSM-5 criteria and categorized as persistent or resolved at hospital discharge. We used multivariable logistic regression models adjusted for ICH severity to determine associations between aphasia, incident delirium, persistent delirium, and discharge to inpatient rehabilitation vs. a skilled nursing facility (SNF). We also performed sensitivity analyses in which we substituted non-English-speaking status as a potential delirium risk factor.

Among 474 patients with a mean age of 70.4 years (SD 15.3), 59% (n=282) experienced delirium and 44% (n=210) had aphasia. In univariate analyses, delirium occurred more frequently in aphasic than non-aphasic patients (83% vs. 41%, p<0.001) and was also more likely to be persistent at hospital discharge (56% vs. 35%, p=0.002). In multivariable models, aphasia remained associated with incident delirium (OR 3.3, 95% CI 1.9-5.9) but not persistent delirium (OR 1.2, 95% CI 0.60-2.6) or SNF discharge (OR 1.0, 95% CI 0.47-2.0). In sensitivity analyses, non-English-speaking status was not associated with delirium incidence, persistent delirium, or discharge location.

Aphasia may be associated with delirium incidence, but further studies are needed to determine whether this reflects the independent effects of language dysfunction. However, language barriers do not appear to influence delirium resolution.