Long-term incidence of ischemic stroke following transient ischemic attack: a nationwide cohort study during 2014-2020

Abstract

Abstract Background The short-term incidence of ischemic stroke after a transient ischemic attack (TIA) is high. However, data on the long-term incidence are sparse but relevant in order to guide preventive strategies. Purpose Estimate 5-year incidence of ischemic stroke after TIA. Methods Using the Danish Stroke Registry, patients with first-time TIA during the period September 2014-December 2020 were identified and matched 1:4 with individuals from the background population, and 1:1 with patients with a first-time ischemic stroke, based on age, sex, and calendar year. The cumulative incidences of ischemic stroke and mortality were estimated by the Aalen-Johansen and Kaplan-Meier estimators, respectively, and compared between groups by use of conditional multivariable Cox regression. Results After matching, we included 23,244 patients with TIA, 92,976 patients from the background population, and 23,244 patients with ischemic stroke (median age 70 years [25th-75th percentile 60-78]: 53% men). Patients with TIA had more comorbidities than the background population, yet less than the ischemic stroke population. The five-year incidence of ischemic stroke following TIA (6.5% [95% CI 6.1-7.0]) was higher than for the background population (2.3% [95% CI 2.2-2.4], hazard ratio (HR) 3.31 [95% CI 3.04-3.61]), but lower than for the control stroke population (10.3% [95% CI 9.8-10.8], HR 0.60 [95% CI 0.55-0.66]). Five-year mortality for patients with TIA (18.0% [95% CI 17.8-19.1]) was higher than for the background population (15.0% [95% CI 14.7-15.3], HR 1.20 [95% CI 1.14-1.26]), but lower than for the control stroke population (29.9% [95%CI 29.2-30.6], HR 0.41 [95% CI 0.39-0.44]) (Figure 1). Conclusion Patients with first-time transient ischemic attach had a 5-year incidence of ischemic stroke of 6.5%. After adjustment for potential confounders, the rate of ischemic stroke was approximately three-fold higher than that of the background population, and 40% lower than that of patients with ischemic stroke.Figure 1.Cumulative incidence of stroke