AHA Scientific Sessions 2026 AbstractNov 7, 2026, 01:29 GMT

Blood protein linked to heart rhythm disorder and dementia

Each doubling of the blood protein osteopontin was linked to hazard ratios of 1.16 for atrial fibrillation and 1.30 for dementia.

Paper by Chang Cui, Tobyn Chiu, Yuekai Ji, +5Lin Yee Chen · Lillehei Heart Institute and Cardiovascular Div

A blood sample with dense and sparse protein ribbons sits between hearts with scattered or orderly shapes and brains with wider or narrower grooves.

Our read

Biomarkers

Studied:: Cohort study of participants
What changed:: Risk was higher with each doubling of osteopontin: hazard ratios of 1.16 for atrial fibrillation and 1.30 for dementia
The catch:: Cardiovascular illness and inflammation could still confound associations despite statistical adjustment
In the record:: The 12th paper on osteopontin with results in people; 93 on record in all.
Why it’s aging research:: It measures an outcome that worsens with age.

Editor of record: Boris Djordjevic · AgingResearch.News

Paper

Participants in the Atherosclerosis Risk in Communities study included 3,964 people assessed for atrial fibrillation, an irregular heartbeat, and 4,075 assessed for dementia. The prospective observational study measured osteopontin, a blood protein, in samples collected in 2011–2013. Researchers recorded 801 new atrial fibrillation cases through 2023 and 982 dementia cases through 2022.

Higher osteopontin levels remained associated with both outcomes after adjustment for inflammation, cardiovascular conditions and other measured risk factors. The hazard ratios compare how often diagnoses occurred over time between people whose protein levels differed by a doubling. The dementia association was stronger in people younger than 75 than in those aged 75 or older.

Why it matters

Dementia and atrial fibrillation affect health in later life. The findings support investigating whether osteopontin contributes to these conditions or serves as a marker of risk.

Caveats

Cardiovascular illness and inflammation could still have influenced both protein levels and later diagnoses despite statistical adjustment. The findings were reported in a conference abstract.