People with cerebral amyloid angiopathy (CAA) are at a significantly higher risk of developing dementia, according to new research set to be presented at the American Stroke Association's International Stroke Conference 2026 in New Orleans. The study found that individuals diagnosed with CAA were four times more likely to develop dementia within five years, regardless of whether they had experienced a stroke.
CAA is characterized by the accumulation of amyloid protein in the brain’s blood vessels, which weakens them and can lead to hemorrhagic (bleeding) or ischemic (clot-caused) strokes. While some amyloid buildup is common as people age and may not cause symptoms, a clinical diagnosis of CAA is made when this accumulation becomes significant enough to damage vessels and affect brain function. In severe cases, the vessel walls may crack, leading to bleeding in the brain. CAA also contributes to cognitive impairment and frequently co-occurs with Alzheimer’s disease.
The research team analyzed health data from over 1.9 million Medicare beneficiaries aged 65 and older between 2016 and 2022. They examined new dementia diagnoses alongside records of ischemic and hemorrhagic strokes among those with CAA. Patients were monitored for changes in their health status over time—categorized as having neither CAA nor stroke, CAA only, stroke only, or both conditions—to assess when dementia developed.
"Our study calculated estimates from a large sample of Medicare patients whether people with CAA are more likely to be newly diagnosed with dementia and to clarify how CAA and stroke - separately and together - relate to new dementia diagnoses."
The findings showed that having CAA substantially increased the likelihood of developing dementia within five years, even more so than experiencing a stroke alone.
"What stood out was that the risk of developing dementia among those with CAA without stroke was similar to those with CAA with stroke, and both conditions had a higher increase in the incidence of dementia when compared to participants with stroke alone. This suggests that non-stroke-related mechanisms are instrumental to dementia risk in CAA," Bruce said. "These results highlight the need to proactively screen for cognitive changes after a diagnosis of CAA and address risk factors to prevent further cognitive decline."
Steven M. Greenberg, M.D., Ph.D., FAHA—a former chair of the International Stroke Conference who authored commentary on cerebral amyloid angiopathy but was not involved in this study—commented: "Diseases of the brain's small blood vessels are major contributors to dementia. This is especially true for CAA, which often occurs together with Alzheimer's disease, making for a potent 1-2 punch. We know there is risk for dementia after any type of stroke, but these results suggest even greater risk for CAA patients." Greenberg is also a professor of neurology at Harvard Medical School in Boston.
The researchers noted several limitations: they relied on administrative diagnosis codes from inpatient and outpatient Medicare claims data rather than direct clinical assessments or imaging studies. "These codes are an imperfect proxy for clinical diagnoses, and misclassifications can occur," Bruce said. The team attempted to address this by using codes shown to accurately capture correct diagnoses in administrative data but acknowledged that access to imaging would have provided more rigorous confirmation.
They concluded that further prospective studies are needed using standardized diagnostic methods for both CAA and stroke.
