IMPORTANCE: Prior research suggests reduced cognitive function after transient ischemic attack (TIA). Whether this is directly related to the TIA, a function of preexisting risk factors, or prior cognitive decline remains unclear. OBJECTIVE: To study if a single, diffusion-weighted image-negative, adjudicated TIA is associated with longitudinal declines in cognition, independent of preexisting risk factors. DESIGN, SETTING, AND PARTICIPANTS: This was a secondary data analysis from the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study, a population-based cohort following up 30 239 Black and White participants for incident cerebrovascular events. The setting consisted of telephone cognitive assessments. Participants were individuals with first-time TIA, first-time stroke, and asymptomatic community control groups with neuroimaging used for adjudication. EXPOSURES: First-time TIA and stroke. MAIN OUTCOMES AND MEASURES: Verbal fluency and memory measures administered biannually. Primary outcome was a composite standardized z score, with secondary outcomes individual test performances. Adjusted segmented regression models characterized pre-event and postevent cognition and annual cognitive change. RESULTS: Included in the study were 356 individuals with first-time TIA (mean [SD] age, 66.6 [8.7]
188 female [53%]) and 965 individuals with first-time stroke (mean [SD] age, 66.8 [8.2]
494 male [51%]). A total of 14 882 individuals (mean [SD] age, 63.2 [8.6] years
8439 female [57%]) were included in the asymptomatic control group. Overall cognitive composite before index event was lower in the stroke (-0.25
95% CI, -0.32 to -0.17) than TIA (-0.05
95% CI: -0.17 to 0.07
P = .005) and asymptomatic (0
95% CI, -0.03 to 0.03
P <
.001) groups. After the index event, the cognitive composite of the group with stroke significantly declined (-0.14
95% CI, -0.21 to -0.07) compared with that of the group with TIA (0.01
95% CI, -0.10 to 0.12
P = .02) and controls (-0.03
95% CI, -0.05 to -0.01
P = .003). The annual decline after the index event was faster (P = .001) in the group with TIA (-0.05
95% CI, -0.06 to -0.03) than that for asymptomatic controls (-0.02
95% CI, -0.02 to -0.02) but not different from the group with stroke (-0.04
95% CI, -0.05 to -0.03
P = .43). CONCLUSIONS AND RELEVANCE: Results of this cohort study suggest that despite the quick resolution of stroke symptoms in TIA, there was apparently sufficient impact to be associated with long-term cognitive decline. Whether the underlying mechanisms are by direct or secondary injury and/or interaction with concomitant neurodegenerative factors remains to be elucidated.