Objectives: To study the associations of pre-stroke cognitive performance with mortality after first-ever stroke or transient ischaemic attack (TIA).
Design: A prospective cohort study.
Setting and participants: In participants having first-ever stroke or TIA during up to 14 years of post-test follow-up (n=155), we investigated the associations of pre-stroke variables and cognitive test results with post-stroke survival. The study is based on those participants of the Uppsala Longitudinal Study of Adult Men who performed cognitive function tests at approximately age 70 (n=919).
Primary outcome measures: Mortality after first-ever stroke or TIA related to pre-stroke executive performance.
Results: Eighty-four (54%) of the first-ever stroke/TIA patients died under a median follow-up of 2.5 years after the event. In Cox proportional hazard analyses adjusting for age, education, social group and traditional stroke risk factors, poor performance in Trail Making Test (TMT)-A was related to mortality (HR 1.88 per SD, 95% CI 1.31 to 2.71, p=0.001). The risk of mortality was approximately threefold higher in the highest tertile compared with the lowest tertile (HR TMT-A= 2.90 per SD, 95% CI 1.24 to 6.77, p=0.014). A similar pattern was seen for TMT-B, but Mini-Mental State Examination results were not related to risk of post-stroke mortality.
Conclusion: Executive performance measured by TMT-A and -B before stroke was independently associated with long-term risk of mortality, after first-ever stroke or TIA in a population-based study of older men.