It may be that patients have better control of cerebrovascular risk factors after the event, protecting against future impairment.
Although it makes sense that a transient ischemic attack (TIA) could increase an individual’s chances of developing vascular dementia, new data from the Framingham Heart Study (FHS) indicate otherwise: these patients don’t seem to be at greater risk when compared with the general population.
For the investigators, the findings came as somewhat of a surprise.
“We thought that with TIA being in the spectrum of cerebrovascular disease that we’d probably see some signal, although maybe not as strong as after a full-blown stroke,” Vasileios-Arsenios Lioutas, MD (Beth Israel Deaconess Medical Center, Boston, MA), the study’s lead author, told TCTMD.
“So initially we thought maybe it’s a chance finding. Then we tried to approach it in many different ways, [including an] analysis for competing risk of death,” he added. That analysis also showed no difference in dementia incidence between people who did or did not have a TIA after age 60.
While not conclusive, the researchers have settled on a two-part hypothesis that may explain the findings. The first is that while TIA may be a risk factor for dementia, it could be a weak one.
“For quite a few people, TIA may function as a bit of an early warning [that] makes them pay more attention to lifestyle, improve blood pressure control a bit, take medications such as aspirin that prevent cerebrovascular disease, and kind of counter-balance that somewhat high, but not exceedingly high risk, to the point that the long-term risk of dementia becomes similar to that of people without a TIA,” Lioutas said.
The hypothesis that changes made after the TIA bring these patients in line with the general population in terms of their dementia risk is a tantalizing one, but it’s also difficult to prove, at least as far as the Framingham participants are concerned. The research group did, however, see evidence that patients who had a TIA reduced their frequency of smoking from 18% before the event to 11% over the follow-up (P = 0.025), had an increase in anticoagulant use from 3% to 18% (P = 0.0005), and a marginal increase in aspirin use, from 46% to 61% (P = 0.052) compared with the controls. These data are based on 80 matched pairs.
Philip Barber, MBChB, MD (University of Calgary, Canada), who was not involved in the study, said while the idea that some risk factors might have been modified enough to reduce the risk of dementia in the TIA group is interesting, these types of large cohort studies are difficult to interpret.
In an email, Barber said there are multiple potential confounders and biases, with ascertainment and selection bias being most relevant to this paper.
“The authors say what is not a TIA, but they don’t explicitly state how the TIAs were selected,” Barber added. “The outcome of dementia is problematic for a relatively small study like this. I would consider cognitive impairment more important to individuals because symptoms can be troublesome even if they do not meet criteria for dementia, and they are more frequent than dementia.”
To TCTMD, Lioutas said lack of information around the types and causes of TIAs that people had is a limitation.
“A TIA due to carotid stenosis is not the same thing as the TIA due to small-vessel disease,” he noted. “That is one nuance that we have to really take into account when interpreting our findings. We don’t have that etiologic classification subtyping available. Anything to do with TIA [has an] inherent limitation or uncertainty as to what is really a TIA and who’s making this determination, [but] I would say that at least for Framingham, it’s probably as good as it can get.”
Mining Framingham for Clues
For the analysis, published last week in the Journal of the American Heart Association, the researchers included 297 participants who had a TIA at age 60 or older (mean age 73 years; 53% women) and 1,485 controls without TIA. All were free of dementia at the time of TIA or matching and were part of the FHS, either through the original or the offspring cohort.
TIA was defined as an episode of rapid onset focal neurological dysfunction attributed to focal cerebral ischemia that resolved within 24 hours.
Compared with matched controls, patients who experienced a TIA had higher systolic BP and were more likely to have hypertension, CAD, and atrial fibrillation.
With up to 20 years of follow-up (median 8.69 years), dementia was diagnosed in 19% of those with a TIA and 24% of matched controls (P = 0.63). Adjusting for stroke occurrence following the TIA did not change the association.
In a sensitivity analysis to examine the change in vascular risk factors and antithrombotic medication use before and after TIA, which involved a 5-year horizon that was adjusted for age and sex, rates of dementia were 8% in the TIA group and 7.5% in controls (P = 0.53).
Other study limitations include FHS itself: cohorts are predominantly white, of European descent, and tend to be largely classified as middle-class and well educated.
The researchers say while the findings “challenge our initial hypothesis and diverge from certain existing literature studies that have proposed persistent cognitive decline following TIA,” other studies suggest have suggested that the cognitive impairment that results from a TIA may be transient and “potentially part of an acute confusional state or delirium.”
As for any message for TIA patients in these data, Lioutas urged caution.
“Let’s not forget that the main concern after a TIA is a stroke, and there is definitely risk of stroke after a TIA that is well established. So, people should adhere to all prescribed medications, especially early on [since they] have been proven to reduce the risk of having a stroke,” he said. “ There is maybe some evidence that if one adheres to antihypertensives, lipid-lowering agents, antiplatelets, and lifestyle modifications [that] in addition to reducing the risk of stroke [it may] prevent them from having more longer-term consequences such as dementia, although this is not a hundred percent proven.”