This analysis used participants from the population-based Rotterdam Study who were dementia-free at enrolment between 2007 and 2012. All included participants underwent carotid ultrasound; those with intima–media thickness >2.5 mm received additional carotid MRI. The cohort was followed for incident dementia through January 2022. Imaging and clinical data were used to classify carotid status and examine associations with cognitive outcomes.
The analytic sample comprised 4,267 participants with a mean age of 67.5 years; 55.5% were women. At baseline, carotid findings were categorized as: plaque without stenosis (n = 483, 11.3%), 1–49% stenosis (n = 989, 23.2%), 50–99% stenosis (n = 107, 2.5%), and occlusion (n = 15, 0.4%). These categories provided the exposure groups for subsequent time-to-event and longitudinal cognitive analyses.
Participants underwent repeated cognitive assessments over the follow-up period and were monitored for incident dementia until January 2022. Mean follow-up time was 9.7 years. The primary clinical endpoint for time-to-event models was all-cause dementia. For longitudinal cognitive change, a general cognitive factor (g-factor) was derived and modelled to capture change over time using multivariable linear mixed-effects methods.
During follow-up, 391 participants developed dementia. Using multivariable Cox proportional hazards models, the study found an elevated dementia risk in participants with carotid plaque without stenosis compared with those without carotid atherosclerosis (hazard ratio [HR] 1.39, 95% CI 1.05–1.83). Carotid occlusion showed a markedly higher risk (HR 4.61, 95% CI 1.82–11.66), although occlusion was rare in the cohort.
In contrast, stenosis categories did not show statistically significant associations with dementia after adjustment: 1–49% stenosis had HR 1.08 (95% CI 0.84–1.39) and 50–99% stenosis had HR 1.18 (95% CI 0.70–1.99). Thus, moderate and high-grade stenosis (50–99%) was not associated with a clear increase in dementia risk in this population-based sample.
Multivariable linear mixed-effects models assessing change in the g-factor over time indicated that neither carotid plaques nor stenosis categories affected the rate of cognitive decline. The study reports no meaningful differences in longitudinal cognitive trajectories between participants with and without plaque or stenosis when modelled across repeated assessments.
The authors examined whether specific plaque characteristics modified the observed associations. Reported results did not demonstrate consistent differences in dementia risk or cognitive decline by plaque characteristics. No additional details on individual plaque features or their separate effect sizes beyond this summary were provided in the source abstract.
The pattern of results—higher dementia risk with carotid plaque without stenosis and with occlusion, but not with 50–99% stenosis—led the authors to interpret findings as consistent with detrimental effects of generalized atherosclerotic disease on dementia risk, and with severe haemodynamic impairment in the case of occlusion. The absence of a clear association for high-grade but asymptomatic stenosis may help explain inconsistent cognitive benefit results from carotid revascularisation trials, suggesting that underlying systemic atherosclerosis rather than focal stenosis could drive dementia risk in some cases.
The Rotterdam Study had approval from the Medical Ethics Committee of Erasmus MC and from the Dutch Ministry of Health, Welfare and Sport under the Population Screening Act. Participant consent procedures and data protection registrations are reported. The authors declared no competing interests. Requests for access to anonymised data follow the Rotterdam Study management team protocol; data are not publicly available due to privacy and consent restrictions.
Conclusion
In this large, prospective, population-based cohort with nearly a decade of mean follow-up, presence of carotid plaque without stenosis and carotid occlusion were associated with increased risk of all-cause dementia, while 1–49% and 50–99% asymptomatic stenosis were not associated with statistically significant increases in dementia risk. Carotid plaques and stenosis were not associated with measurable differences in longitudinal cognitive decline as assessed by the g-factor. These findings are presented by the investigators as supporting a role for generalized atherosclerotic disease and for severe haemodynamic compromise in dementia risk rather than a consistent effect of asymptomatic stenosis alone.