Lakshminarayan Kamakshi, Norby Faye L, Hong Ching-Ping, Folsom Aaron R, Rosamond Wayne D, Gottesman Rebecca F, Koton Silvia, Pike James Russell, Lutsey Pamela L
Division of Epidemiology & Community Health, School of Public Health, University of Minnesota, Minneapolis.
Department of Epidemiology, University of North Carolina, Chapel Hill.
Neurology. 2025 Sep 23;105(6):e214020. doi: 10.1212/WNL.0000000000214020. Epub 2025 Sep 4.
There are inconsistent data on the long-term risk of dementia after TIA. We used the multidecade Atherosclerosis Risk in Communities (ARIC) study to examine long-term risk of incident dementia and other post-TIA outcomes.
The ARIC study recruited participants from 4 US communities-Minneapolis, MN; Jackson, MS; Forsyth County, NC; and Washington County, MD-during the years 1987-1989. Participants have been followed through in-person visits, phone calls, and hospital and death registry surveillance. Our exposure was the first hospitalized TIA event, ascertained between 1987 and 2020. We report on post-TIA incident events ascertained between 1987 and 2020, including adjudicated dementia, stroke, cardiovascular disease (CVD) events, and mortality. We used Cox proportional hazard regression modeling with TIA as a time-dependent exposure to estimate associations with outcome events. Models were adjusted for demographics and vascular risk factors.
The ARIC study enrolled 15,792 participants (27% Black) aged 45-64 years. Our analytic sample included 13,721 participants (mean age 54 years, 55.5% women) who were without history of hospitalized TIAs or the outcomes of interest at baseline. There were 536 hospitalized TIAs over a median of 29 years. Incident dementia risk almost doubled after TIA (hazard ratio [HR] 1.96; 95% CI 1.66-2.31). Participants who had a TIA were at high risk of stroke (8.14; 6.87-9.64) and CVD (1.52; 1.28-1.82) compared with those without a TIA. They were also at an elevated risk of all-cause mortality (HR 1.55; 95% CI 1.39-1.73) and CVD mortality (1.82; 1.48-2.22). A sensitivity analysis restricted to participants with TIAs but without stroke showed an elevated risk of dementia (HR 1.67; 95% CI 1.37-2.03).
We report an increased risk of dementia after TIA in our multidecade cohort. The elevated risk was present after elimination of stroke as an intermediary variable. Consistent with other studies, we found an increased risk of stroke, CVD, and mortality after TIA. The clinical implications of these results are that in addition to vascular risk factor modification, clinicians should consider incorporating cognitive screening into post-TIA follow-up care.
关于短暂性脑缺血发作（TIA）后发生痴呆的长期风险，数据并不一致。我们利用开展了数十年的社区动脉粥样硬化风险（ARIC）研究，来探究发生痴呆及其他TIA后转归的长期风险。
ARIC研究于1987 - 1989年从美国4个社区招募参与者，分别是明尼苏达州明尼阿波利斯市、密西西比州杰克逊市、北卡罗来纳州福赛斯县和马里兰州华盛顿县。通过面对面访视、电话随访以及医院和死亡登记监测对参与者进行随访。我们确定的暴露因素是1987年至2020年间首次住院的TIA事件。我们报告了1987年至2020年间确定的TIA后发生的事件，包括经判定的痴呆、中风、心血管疾病（CVD）事件和死亡率。我们使用Cox比例风险回归模型，将TIA作为时间依赖性暴露因素，以估计与转归事件的关联。模型针对人口统计学和血管危险因素进行了调整。
ARIC研究纳入了15792名年龄在45 - 64岁的参与者（27%为黑人）。我们的分析样本包括13721名参与者（平均年龄54岁，55.5%为女性），他们在基线时没有住院TIA病史或感兴趣的转归情况。在中位29年的时间里共发生了536次住院TIA。TIA后发生痴呆的风险几乎增加了一倍（风险比[HR] 1.96；95%置信区间1.66 - 2.31）。与未发生TIA的参与者相比，发生TIA的参与者发生中风的风险较高（8.14；6.87 - 9.64），发生CVD的风险也较高（1.52；1.28 - 1.82）。他们全因死亡风险（HR 1.55；95%置信区间1.39 - 1.73）和CVD死亡风险（1.82；1.48 - 2.22）也有所升高。一项仅限于有TIA但无中风的参与者的敏感性分析显示，痴呆风险升高（HR 1.67；95%置信区间1.37 - 2.03）。
我们报告了在我们这个数十年队列中，TIA后痴呆风险增加。在排除中风作为中间变量后，风险仍然升高。与其他研究一致，我们发现TIA后中风、CVD和死亡风险增加。这些结果的临床意义在于，除了调整血管危险因素外，临床医生应考虑在TIA后的随访护理中纳入认知筛查。
