Greater overall cerebral small vessel disease burden was associated with poorer cognition and daily function across the full range of diagnoses recruited into Canada's national dementia cohort, according to a cross-sectional analysis of baseline data from 958 participants published in Alzheimer's & Dementia (2026; 22(6):e71546). Dylan Guan, Zahinoor Ismail, Eric Smith, Aravind Ganesh and colleagues at the University of Calgary analysed the Comprehensive Assessment of Neurodegeneration and Dementia (COMPASS-ND) study, the signature clinical study of the Canadian Consortium on Neurodegeneration in Aging, which enrols at 32 centres across Canada and spans cognitively normal volunteers, subjective cognitive decline, mild cognitive impairment, Alzheimer's disease, dementia of mixed etiology, Lewy body disease and frontotemporal lobar degeneration. Of the 958 analysed, 19.4 percent were cognitively normal, 14.9 percent had subjective cognitive decline, 40.3 percent had mild cognitive impairment and 25.4 percent had dementia.

Brain MRI scans were visually rated for the standard markers of vascular injury: lacunes, cerebral microbleeds, white matter hyperintensities graded by the Fazekas scale, and enlarged perivascular spaces. These were added into a cumulative small vessel disease score, with two alternative versions also tested, including one proposed as a non-diagnostic staging measure of cerebral amyloid angiopathy burden. Across the cohort, lacunes were present in 142 participants (14.8 percent) and microbleeds in 190 (19.8 percent), and the average total enlarged perivascular space score was 5.2, with a standard deviation of 1.4. The vascular imaging protocol was designed into COMPASS-ND from the outset; the cohort itself was planned around 2310 deeply phenotyped participants.

Higher small vessel disease scores were associated with higher Hachinski ischemic scores, poorer performance on the Montreal Cognitive Assessment, worse Clinical Dementia Rating Sum of Boxes, and lower composite neuropsychological battery z-scores. The associations were strongest for executive function, attention and learning. Among the individual markers, white matter hyperintensity burden showed the strongest and most consistent link to lower cognitive performance, with a larger effect for periventricular than for deep white matter. An earlier conference abstract from the same group, published in January 2026 and preliminary, reported that in a slightly different sample of 972 participants a higher global score carried adjusted odds of 1.14 (95 percent confidence interval 1.02 to 1.26) of sitting in a more severe cognitive diagnostic category.

The authors state the boundary of their own work plainly: the design is cross-sectional, a single snapshot at study entry, and they write that small vessel disease burden "may further contribute" to poorer cognition rather than asserting that it causes it. Visual rating scales are coarser than volumetric or continuous measures, though the authors argue additive visual scores remain the more practical option in clinical settings, and their findings held across two different additive scores. Diagnoses in the cohort are clinical, not confirmed at autopsy.