Abstract Background and aims In-hospital intracerebral hemorrhage (ICH) is understudied despite evidence of its substantial morbidity and mortality. We described the clinical profile and outcomes of in-hospital ICH patients in a provincial cohort. Methods We conducted a retrospective cohort study of adults with in-hospital ICH across acute-care hospitals in Alberta (2018–2022). Cases were identified using ICD-10 codes and validated through chart abstraction. Demographic, clinical, radiographic, laboratory, and outcome variables were collected from administrative data and chart review. We compared survivors and non-survivors; and performed multivariable logistic regression to determine variables associated with in-hospital mortality, including age, NIHSS, prior ICH, antithrombotic use, ICH volume and location. Results Among 222 patients (median age 66 years (IQR 56-76); 45.1% women), in-hospital mortality was 59.1% (n=131) with 45.0% of deaths occurring within 7 days. Compared with survivors, non-survivors had higher NIHSS scores (median 26 vs. 7; p0.001), larger hematoma volumes (median 47.1 vs. 8.3 ml; p0.001), and more frequent intraventricular extension (55.7% vs. 29.7%; p0.001). Rates of comorbidities, prior ICH, alcohol use, liver disease, hematologic conditions, antithrombotic use, and laboratory abnormalities (low platelets and high INR) were similar between groups. Age (OR 1.03; 95% CI 1.00-1.06) higher NIHSS scores (NIHSS 6–15: OR 4.21 1.33–13.35; NIHSS 21–42 OR 19.81 5.94-66.03; and larger hematoma volumes (OR 1.01 per ml 1.00–1.02) were significantly associated with mortality on multivariable analysis. Conclusions In-hospital ICH was associated with high early mortality. In this provincial cohort, older age, higher baseline stroke severity, and larger hematoma volume were associated with in-hospital death. Conflict of interest Mohamad Mehdi: nothing to disclose; Farah Ali: nothing to disclose; Rana Abdalrahman: nothing to disclose; Ameen Alizada: nothing to disclose; Benjamin Agnelli: nothing to disclose; Jessalyn Holodinsky: nothing to disclose; Bijoy Menon: nothing to disclose; Michael Hill: nothing to disclose; Mohammed Almekhlafi: nothing to disclose; Katrina Hannah Ignacio: nothing to disclose.
Mehdi et al. (2026) studied this question.