As of August 2026, the American College of Cardiology now recommends screening older adults with cardiovascular disease for cognitive impairment, which necessitates the availability of tools that are capable of measuring change over time, rather than identifying impairment at a single visit, and are sensitive to cognitive changes more subtle than moderate or severely progressed dementia.
The Lancet Commission on dementia prevention estimates that around 45% of dementia cases worldwide are potentially preventable through modifiable risk factors, several of which fall within routine cardiovascular care, including hypertension, diabetes, obesity, smoking, and elevated cholesterol.
Other estimates attribute a quarter of dementia cases to cardiovascular risk factors alone. The relationship is visible in practice, where one in three patients presenting at cardiology clinics have some degree of cognitive impairment, and up to half of those cases go unrecognized.
Those exposures are widespread and concentrate with age. Hypertension affects 58.6% of adults aged 45 to 64 and 76.5% of those 65 and older, and around 42% of US adults have obesity. On current projections, hypertension and obesity will each affect more than 180 million US adults by 2050. This article examines what that requires in cardiac populations.
Article Highlights
As of August 2026, the American College of Cardiology recommends cognitive screening for older adults with cardiovascular disease, while acknowledging that strategies suited to high-volume practice have yet to be developed.
- Cognitive impairment affects up to 81% of patients with heart failure, and up to half of cases in cardiology clinics go unrecognized.
- In one survey of cardiologists, 83% believed insufficient attention was paid to cognitive impairment, while 3% performed structured screening.
- Across two decades of CABG trials, 3.1% reported any cognitive assessment, and tasks used before and after surgery showed no difference at any postoperative timepoint.
- The review proposes six criteria for cognitive outcomes suited to widespread implementation.
Cognitive impairment is common in cardiac populations and hard to detect
Elevated cognitive risk is a shared comorbidity across cardiac conditions rather than concentrated in one. Cognitive impairment affects up to 81% of patients with heart failure. Atrial fibrillation carries a 39% increased risk of memory or thinking problems, coronary heart disease a 27% higher risk of incident dementia, and as many as half of patients experience cognitive decline after myocardial infarction.
The mechanisms of this heart-brain connection are shared rather than incidental: reduced cerebral perfusion secondary to endothelial dysfunction and arterial stiffness, inflammaging, and cerebrovascular injury spanning large and small vessel disease. Heart failure is itself a risk factor for Alzheimer's disease, and cognitive decline proceeds more rapidly after a heart failure diagnosis than without one.
Detection can be most challenging at the mild end of impairment. Mild cognitive impairment is often underrecognized in specialty care, and patients frequently present unaware of any declines in their own cognitive function, appearing superficially intact across a brief encounter, particularly where a spouse supplies the missing detail.
Why is cognitive impairment missed in cardiology practice?
These challenges are reflected in practice: In a survey of cardiologists drawn from working groups on geriatric cardiology and heart failure, 83% believed that insufficient attention was paid to cognitive impairment, though just 3% performed structured cognitive screening. A more recent survey of the Hellenic Heart Failure Clinics Network found that 80% of respondents were aware of the high prevalence of cognitive decline while just 14% screened systematically, citing lack of time and unfamiliarity with assessment tools as the principal barriers.
Cognitive impairment affects outcomes cardiology is already accountable for. It complicates medication adherence, shared decision making, and postprocedural care planning, and is associated with higher rates of hospitalization and mortality. Among 565 patients in an Australia-wide heart failure study, death or readmission within 30 days of discharge was twice as likely among those with at least mild cognitive impairment.
Death or readmission within 30 days of discharge was twice as likely among heart failure patients with at least mild cognitive impairment.
Huynh et al., International Journal of Cardiology
The ACC statement recommending cognitive screening acknowledges that screening strategies suited to high-volume cardiac practice have yet to be developed, citing workflow and resource barriers to implementation. What screening needs to accomplish also varies with where in the cardiac pathway the question arises.
Where does cognitive change surface in cardiac care?
The neuropsychological profile of vascular cognitive impairment (VCI) differs from that of Alzheimer's disease. Diagnostic criteria for VCI no longer require memory impairment, and deficits in executive function and processing speed are the more characteristic findings.
Assessment of this profile, coupled with the fact that cognitive decline in cardiac populations is frequently subtle and occurs over long periods of time, is poorly matched to the cognitive instruments most widely available. The ACC statement notes that outpatient screening tools such as the Mini-Cog, MMSE, MoCA, and SLUMS demonstrate variable sensitivity, particularly for vascular dementia, and that slowed processing speed and impaired problem-solving are difficult to assess within a brief office visit.
Matching an instrument to that profile is further complicated by its variability. Impairment attributable to sporadic cerebral small vessel disease has been reported across all major cognitive domains rather than confined to a dysexecutive pattern, and the deficits observed in VCI differ between patients and with lesion location.
Where deficits are distributed unpredictably, an instrument evaluating a single domain will detect impairment in some patients and miss it in others. Assessment spanning multiple domains is more likely to register impairment wherever it emerges, and establishes a broader baseline against which subsequent change can be interpreted.
The purpose of cognitive assessment also differs by clinical context:
| Clinical context | Presentation | Question to be resolved |
|---|---|---|
| Perioperative |
|
Whether observed change is consistent with expected recovery, attributable to delirium, or represents the onset of a longer trajectory in a patient at elevated baseline risk |
| Medication management |
|
Whether symptoms are plausibly attributable to the treatment regimen, the underlying condition, or an unrelated process, and whether subsequent adjustment yields measurable benefit |
| Longitudinal care |
|
Whether the observed pattern is more consistent with disease progression, medication effects, or age-related change, and at what point diagnostic referral is indicated |
Why do many cognitive instruments miss change after cardiac surgery?
Creyos scientists and researchers at Sunnybrook Health Sciences Centre, the University of Toronto, Columbia, Cornell, and Western University examined two decades of coronary artery bypass graft (CABG) randomized trials. Of 2,284 trials, only 3.1% reported any cognitive assessment. The remaining 96.9% of the trial literature contained no cognitive outcome.
What was measured, and how
Approaches varied widely among trials that did assess cognition. The 71 trials used 145 distinct cognitive tasks, more than 100 of which appeared only once, frequently as components of a unique battery. Trail Making Test Part B was the most common instrument, administered in 56.3% of trials, followed by Part A in 53.5%. Screening measures such as the MMSE and MoCA appeared in a third of trials, and 18.3% reported a screening measure alone.
Of the seven tasks that were used both preoperatively and postoperatively in at least three trials (Controlled Oral Word Association, Digit Span forward and backward, Digit Symbol Substitution, and Trail Making Test Parts A and B), none demonstrated a difference between baseline and any postoperative timepoint, including assessments performed within the first week, when cognitive change would be most expected.
Attrition compounds the challenges with interpretation. Weighted by study size, attrition across follow-up timepoints averaged 30%, ranging from 0% to 62%. Trials employing detailed batteries reported higher attrition than those relying on brief screens, supporting the need for assessment strategies with a patient-friendly experience that are resilient to attrition.
The MMSE and MoCA were developed to detect the emergence of dementia rather than to characterize cognitive performance, and both are subject to floor and ceiling effects that limit sensitivity to mild impairment. Detailed neuropsychological batteries offer greater sensitivity but require trained administrators, carry substantial cost, and can be difficult to implement across multicentre and multilingual settings.
Six criteria for a feasible cognitive assessment in cardiac populations
The review proposes six criteria for cognitive outcomes suited to widespread implementation:
- Availability in multiple languages;
- Standardized administration without trained neuropsychometrists;
- Capacity to detect change between assessments;
- Reduced vulnerability to attrition;
- Coverage of key cognitive domains or patient-oriented outcomes; and
- Time and cost efficiency for participants and sites.
Brief screens are efficient and well tolerated but often constrained in range. Detailed batteries are comprehensive but poorly scalable. Informant-reported scales are broadly accessible but do not measure objective cognitive function directly. Online assessments, including Creyos, meet the remaining criteria: administered at home or in clinic without trained personnel, they cover multiple domains with fewer ceiling effects and support repeat administration against a patient's own baseline.
Whichever instrument is selected, the design of its use determines whether change becomes visible. That requires a baseline captured before the period of interest, whether preoperatively, at initiation of a cardiac regimen, or at entry into longitudinal follow-up; repeat assessment at intervals matched to the clinical question; and interpretation against the patient's own prior performance, anchored to a normative sample sufficient to make the baseline itself interpretable.
The lessons of the review apply beyond CABG, extending to cardiovascular and surgical trials more broadly, and the same logic governs routine care.
Feasibility in cardiac settings is testable. In a pilot study of patients undergoing open heart surgery at two Canadian tertiary care centres, 94% reported no barriers to completing the Creyos battery remotely, and patients demonstrated impaired performance relative to healthy norms both before and after surgery, most notably on tasks requiring verbal processing.
Closing the measurement gap
The review concludes that an update to consensus recommendations on cognitive outcomes is warranted, identifying measures that better serve patients undergoing cardiac surgery and that clinicians and trialists can agree on. The ACC statement reaches a parallel conclusion from the practice side, calling for pragmatic screening strategies feasible in high-volume settings.
Both are questions of measurement before they are questions of guidelines. If affected cognition is part of the patient experience, it should be part of the evidence collected.
Written by Sydni Paleczny, Staff Scientist at Creyos
Sydni earned her MSc in Neurosciences at Western University under Dr. Adrian Owen. Her research explores neuropsychological outcomes after cardiac surgery, with interests in cognitive neuroscience, critical care, and brain health. At Creyos, she supports scientific validity, health technology, and ongoing research.