Abstract
Background: Post-stroke cognitive impairment (PSCI) is common and contributes substantially to disability, yet prospective hospital-based data in Vietnam remain limited.
Objective: To estimate the 6-month post-stroke dementia prevalence and identify associated factors among patients with stroke.
Methods: This prospective study consecutively enrolled inpatients with ischemic or hemorrhagic stroke without pre-stroke dementia (IQCODE < 3.6), aged 60–80 years with ≥5 years of education. Demographic, clinical, vascular risk-factor, and imaging data were collected during hospitalization. At 6 months, cognition was assessed using the VnCA battery and dementia was diagnosed according to VASCOG criteria. Domain z-scores (memory, executive function, and language) were calculated and compared using the Wilcoxon rank-sum test with Holm adjustment. Multivariable Firth logistic regression assessed associations of age, education, sex, and admission NIHSS with dementia; adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were reported.
Results: Among 223 patients followed, 44 completed cognitive assessment at 6 months. The median age was 67 (IQR 63–70) years; 54.5% were men; the median admission NIHSS was 4.0 (IQR 1.25–7.0). The 6-month post-stroke dementia prevalence was 45.5% (20/44). The dementia group had more severe stroke (median NIHSS 6.0 vs 3.0). Domain z-scores were markedly lower in the dementia group, particularly executive function (−1.17 vs 0.19), language (−1.08 vs −0.30), and memory (−2.86 vs −1.37) (Holm-adjusted p < 0.01). In multivariable Firth logistic regression, admission NIHSS was independently associated with dementia (aOR 1.23 per 1-point increase; 95% CI 1.04–1.55; p=0.0095).
Conclusions: Six-month post-stroke dementia was frequent in this hospital-based cohort. Stroke severity (NIHSS) emerged as a key predictor, supporting systematic cognitive screening and longitudinal follow-up with particular attention to executive and language deficits.