Abstract WP427: Cognitive Function of Ten-year Stroke Survivors Compared to Non-stroke Individuals: the Lund Stroke Register

Abstract

Introduction: Post-stroke cognitive impairment (PSCI) has a considerable impact on patients and society. However, the characteristics and prevalence of long-term PSCI may be influenced by assessment methods and selection bias. We therefore used two cognitive screening methods to assess PSCI in ten-year stroke survivors, made comparisons with non-stroke individuals, and compared these screening methods. Methods: The Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment (MoCA) were administered and compared in a population-based consecutive sample of ten-year stroke survivors. In addition, age- and sex-matched non-stroke controls were assessed with the MMSE. Regression analyses adjusting for education compared the stroke survivors’ MMSE performance with the controls. Moderate/severe cognitive impairment, approximating to dementia, was defined using MMSE<24 and MoCA<20 as cut-offs. To detect those with mild cognitive impairment, alternative cut-offs of MMSE<27 and MoCA<25 were also used. Results: In total, 127 of 145 stroke survivors participated. The total MMSE-scores were similar for stroke survivors (median 27) and 354 controls (median 27; p =0.922); as well as proportions with MMSE<24 (23% vs. 17%; p =0.175) or MMSE<27 (47% vs. 49%; p =0.671). After adjustment for education, stroke survivors showed an increased risk for moderate/severe cognitive impairment defined by MMSE<24 (OR=1.82; p =0.036). Executive dysfunction was seen in 42% of the stroke survivors vs. 16% of the controls as evaluated by MMSE ( p <0.001). According to MoCA, moderate/severe cognitive impairment (MoCA<20) was observed in 28% of the stroke survivors; any degree of cognitive impairment (MoCA<25) was seen in 61%; and 45-61% displayed executive function deficits. Conclusions: PSCI including executive dysfunction is common among ten-year stroke survivors, who have an increased risk of moderate/severe cognitive impairment compared to non-stroke controls. The prevalence of long-term PSCI may have been previously underestimated, and MoCA may be more suitable for post-stroke cognitive assessment.