Background and purpose: Aphasia is an important post-stroke sequela. We estimated the prevalence and main determinants of post-stroke aphasia in the prefecture of Arcadia, Greece.
Method: Prospective population-based study of Arcadia residents diagnosed with first ever stroke. within a 2 year period. Associations of aphasia with potential predictors were analysed by logistic regression in: (i) the entire cohort and (ii) the sub-cohort of patients who were alive 28 days post-stroke. Multivariate regression models were adjusted for left hemispheric stroke localization and modified Rankin Scale 28 days post-stroke (MRS-1mo).
Results: Of 555 subjects, 126 (22.7%) suffered from aphasia. When only the 405 survivors were considered, 77 (19.0%) suffered from aphasia. In adjusted multivariate models of the entire cohort, factors conferring significantly higher risk for aphasia included female gender, diabetes mellitus (borderline significance) and heart disease. In adjusted multivariate models of survivors, only diabetes was associated with significantly higher risk for aphasia.
Conclusions: Female gender, diabetes and heart disease were independent prospective predictors of aphasia. The study offers a quantitative estimate of the public health problem of post-stroke aphasia in Greece and suggests that the role of diabetes in post-stroke aphasia may be more important than previously appreciated.