To survive a myocardial infarction but then face the potential for the development of dementia is unsettling to say the least. In a study in this issue of Circulation, Sundbøll and colleagues test this association in a Danish nationwide patient registry. Administrative records are used to track incident dementia over ≤35 years of follow-up for 314 911 one-year survivors of myocardial infarction (MI) and 1 573 193 controls without MI. Although all-cause dementia did not differ between the groups over the follow-up period, vascular dementia was more frequent in individuals with an MI, particularly in those with a stroke during follow-up or who required coronary artery bypass grafting (CABG) surgery.
When interpreting what these results mean for the survivor of an MI, however, it is important to consider how these conditions were defined and what the likely explanation might be for the observed associations. In this study, individuals with MI and without MI were clearly 2 different groups of individuals: although matched to the MI cases for sex, birth year, and calendar year, it is not surprising that those people without an MI had lower vascular risk with higher levels of educational attainment and socioeconomic status than did their counterparts who did experience an MI. Although attempts were made to adjust for these observed differences, it is likely that these groups remain substantively different.
As a result of these differences, it remains unclear whether the increased risk is because of the MI or the underlying risk that led to the MI in the first place. This question has been addressed in individuals undergoing CABG surgery: although cognitive decline was noted to be high in earlier studies among individuals undergoing CABG surgery, with 42% reportedly experiencing cognitive decline at 5 years, studies that included a control group of individuals with similar coronary artery disease but who did not undergo CABG found that decline did not differ at 6 years between the CABG group and their nonsurgical coronary artery disease counterparts. Cognitive decline was steeper, however, in both of these coronary artery disease groups than in the group without vascular disease. Thus, if one could study patients with identical vascular risk factors, with similar coronary artery disease, but with some experiencing MI and others not, it is likely that cognitive decline and dementia rates might be similar in these groups. Evaluation of these risk factors in population-based studies supports this finding, with higher rates of dementia in individuals with many of the same vascular risk factors that are known to be important for MI risk: hypertension, diabetes mellitus, smoking, and high cholesterol.[5,6] A recent systematic review considered 24 cohort studies and found an elevated risk of cognitive impairment or dementia (odds ratio, 1.45) in individuals with coronary heart disease, including but not limited to MI.
Another consideration in the interpretation of these results involves the determination of the dementia diagnoses: because dementia diagnoses were obtained from medical records, it is likely that clinicians making these diagnoses were biased based on knowledge of that individual’s vascular (and MI) history. Thus, an individual with new-onset cognitive decline and a history of MI is much more likely to be labeled as having vascular dementia than would someone without these risk factors but with similar decline. In fact, having no vascular history but having new dementia symptoms would probably make it more likely that someone would be diagnosed with Alzheimer’s disease (AD). This might explain the finding in this study that AD is actually less common in this study in individuals with prior MI, with vascular dementia more common in the survivors of an MI.
The association in question is challenging to study, given these clear differences in the types of people who do and do not get MIs and the previously described possibility of bias in identifying and defining dementia cases. Yet there is an important message here: patients who experience MI are diagnosed with vascular dementia at a higher rate than the general population. Furthermore, in real clinical practice, patients with MI are different than those without MI, just as seen in this study, and patients with MI are more likely to be labeled as having vascular dementia than are individuals without an MI. Therefore, this message and its implications need to be understood by clinicians caring for patients recovering from MI. Screening for cognitive impairment may be especially important in individuals with advanced enough coronary artery disease to experience an MI, and certainly risk factor management will be critical.
In this study, individuals experiencing stroke during follow-up had an especially strong association between MI and vascular dementia, with an odds ratio of 4.48, compared with 1.30 for MI in individuals without stroke. This could still reflect differences in the extent of vascular risk, with those individuals with both MI and stroke likely having the greatest extent of underlying vascular risk, thus increasing the risk for vascular dementia, or could reflect ascertainment bias with a further increased likelihood of being given a label of vascular dementia with that history. This is especially likely because cerebrovascular disease, particularly with a clear temporal relation to the onset of dementia, is part of the NINDS-AIREN (National Institute of Neurological Disorders and Stroke and the Association Internationale pour la Recherche et l’Enseignement en Neurosciences) diagnostic criteria for vascular dementia and is a specific exclusionary criterion for AD. Despite these concerns, however, this emphasizes an important aspect of care after an MI and suggests that it may be especially important to prevent stroke in this high-risk population. At 1 year after an MI, stroke is estimated to occur in 21.4 individuals per 1000 MIs and is predicted by individual characteristics (diabetes mellitus, hypertension, age, nonwhite race) as well as characteristics of the MI (anterior location and prior MI or atrial fibrillation), with similar risk factors and rates found in other studies. Although stroke may not be the direct cause of the increased risk of vascular dementia in all survivors of an MI, there is evidence that cognition declines more steeply after a stroke, and thus it is especially important in a vulnerable individual to avoid this complication after an MI.
Vascular dementia is rarely a pure entity, with increasing evidence for overlap between vascular and AD etiologies of dementia. Dementia risk is increased in individuals with elevated vascular risk, and this study adds important data demonstrating potentially higher risk in individuals with MI, particularly those who experience a stroke or require CABG surgery during follow-up after an MI. Treatment after MI that focuses on control and prevention of vascular risk factors is likely to decrease chances of stroke and may even decrease chances of dementia. Survival after MI has improved over the last few decades, which means there is a larger, older portion of the population with elevated vascular risk. Because age is a major risk factor for AD as well, and there is increasing evidence for clinical and even mechanistic overlap between these 2 pathologies (AD and vascular disease),[14,15] it is likely that risk for dementia in this population of survivors of an MI may increase in the coming years. This study reminds us of the importance of vascular risk factor treatment, prevention, and monitoring for cognitive decline in an especially vulnerable portion of the population.