Heart Risks Leave Distinct Fingerprints on the Aging Brain Across Latino Heritage Groups

The brain’s white matter is the wiring that keeps cognition humming, and a sweeping new study of more than 1,600 older Hispanic/Latino adults shows that the health of that wiring is shaped not only by classic vascular risk factors but also by something researchers have long underappreciated: heritage itself. In an analysis drawn from the SOL-INCA MRI cohort, an ancillary project of the Hispanic Community Health Study/Study of Latinos, investigators found that hypertension and diabetes were each linked to measurable deterioration of white matter integrity, while participants of Cuban and Puerto Rican background carried a heavier burden of brain vascular injury than Mexican participants even after those risk factors were statistically accounted for. The findings, published in GeroScience, add a striking layer of nuance to how scientists understand brain aging in one of the fastest-growing and most diverse populations in the United States.

Cerebral small vessel disease sits at the center of the story. It is the chronic, cumulative damage to the brain’s tiniest blood vessels, and it is a leading contributor to stroke and cognitive decline in older adults. On magnetic resonance imaging, the disease announces itself in two complementary ways. The first is the white matter hyperintensity, or WMH: bright patches that appear on certain MRI sequences where tissue has been damaged by chronic ischemia, blood-brain barrier leakage, and gliosis. The second is subtler and requires diffusion-weighted imaging, which tracks the random motion of water molecules through tissue. In healthy white matter, water diffuses preferentially along the direction of nerve fiber bundles, a property quantified as fractional anisotropy, or FA. When the microstructure of those bundles breaks down, FA falls. A companion measure, free water, estimates the fraction of the signal that comes from unconstrained extracellular fluid, and it rises when inflammation, atrophy, or vascular injury opens up space around the fibers.

The research team, led by Elmira Agah, Charles DeCarli, and Pauline Maillard of the University of California, Davis, together with collaborators across multiple field centers, applied these measures to 1,619 cognitively normal participants, most of them older adults, imaged on standardized 3-Tesla MRI protocols. Restricting the sample to people without cognitive impairment was a deliberate choice: it allows the earliest structural signatures of vascular brain injury to be detected before symptoms emerge, when interventions such as blood pressure control might still alter the trajectory. The participants represented the remarkable heterogeneity of the US Hispanic/Latino population, with heritage groups including Mexican, Cuban, Puerto Rican, Dominican, and Central or South American backgrounds.

The statistical approach was straightforward but rigorous. Linear regression models tested whether each vascular risk factor, namely hypertension, diabetes, total cholesterol of 200 mg/dL or higher, and ever-smoking, was associated with the white matter measures, with adjustments for age, sex, education, and field center. The team then asked whether heritage itself predicted white matter outcomes and, critically, whether the effects of vascular risk factors differed across heritage groups. Multiple comparisons were controlled using false discovery rate methods, a standard safeguard against spurious findings when many tracts and measures are tested simultaneously.

The results on vascular risk were unambiguous. Hypertension was associated with higher free water and greater white matter hyperintensity burden, both statistically robust at p < 0.001, and diabetes showed the same pattern, additionally linked to reduced fractional anisotropy. In other words, both conditions were tied to more visible tissue damage and to microscopic disruption of the fiber bundles themselves. Elevated cholesterol and ever-smoking showed weaker or more conditional associations. These results align with a large body of work, including analyses from the UK Biobank and the Atherosclerosis Risk in Communities study, showing that midlife and late-life vascular risk leaves durable marks on the brain’s structural connectivity. They also reinforce the clinical message that blood pressure and glycemic control are, in effect, brain protection strategies.

The heritage findings are what elevate the study beyond a confirmation of known biology. Compared with Mexican participants, Cuban participants showed significantly higher levels of both free water and white matter hyperintensities, and Puerto Rican participants showed higher WMH burden, after full adjustment for vascular risk factors and demographics. Tract-level analyses, which examined individual fiber pathways such as those connecting distant cortical regions, mirrored the overall pattern, with more impaired measures among Cuban and Puerto Rican participants. Because the models adjusted for the measured risk factors, the differences cannot be explained simply by higher rates of hypertension or diabetes in those groups, although the burden of such factors is known from prior HCHS/SOL work to vary considerably across heritage backgrounds.

There was one notable exception to the pattern of uniform risk effects: smoking. The interaction analysis revealed that smoking was associated with higher free water and lower fractional anisotropy among Cuban and Puerto Rican participants, and with lower FA among Central and South American participants, relative to Mexican participants. This suggests that the same exposure may translate into different degrees of brain injury depending on genetic ancestry, environmental context, dietary patterns, acculturation, or other unmeasured variables that cluster within heritage groups. Disentangling those possibilities is now a central question for follow-up research.

The implications reach well beyond epidemiology. Hispanic/Latino adults experience a high burden of vascular risk factors, and prior studies have documented disparities in hypertension awareness, treatment, and control within this population. Because white matter hyperintensities and microstructural damage predict incident stroke, mild cognitive impairment, dementia, and mortality in longitudinal cohorts, the white matter measures used here function as early warning indicators. If heritage-specific vulnerabilities exist, then prevention programs calibrated for one subgroup may not serve another equally well, and clinical trials of vascular risk reduction in diverse Latino populations should be powered and stratified accordingly. The authors also point toward broader social determinants, including the documented pathways linking racism, xenophobia, discrimination, and structural inequity to cardiometabolic outcomes, as candidate explanations that deserve systematic investigation rather than casual attribution to genetics alone.

Technically, the study showcases how far neuroimaging biomarkers have matured. Free water imaging, originally developed to separate extracellular fluid contamination from true tissue diffusion signal, has proven sensitive to early-stage small vessel disease, detecting alterations in tissue surrounding visible hyperintensities before those regions brighten on conventional scans. Atlas-guided tract reconstruction allowed the team to examine specific white matter pathways comprehensively and automatically, and convolutional neural network segmentation tools now make production-level brain parcellation feasible across large cohorts. Combining these tools with a population-based, multi-heritage cohort design is what gives the study its distinctive power.

The authors are careful about what the data cannot say. This was a cross-sectional analysis, so it captures associations at a single point in time rather than the evolution of white matter injury over years. Residual confounding is possible, and the observed heritage differences, while statistically significant, warrant replication and mechanistic follow-up. Yet the central conclusion stands firmly: white matter integrity differs meaningfully across Hispanic/Latino heritage groups even among cognitively normal older adults with comparable measured risk profiles, and treating this population as a monolith obscures real variation in brain aging. As the demographic weight of the US Latino population grows, studies like SOL-INCA make the case that precision in both research and medicine must extend to heritage, not merely to broad ethnic categories.

Subject of Research: Vascular risk factors and white matter integrity in older Hispanic/Latino adults

Article Title: Association between vascular risk factors and white matter integrity among diverse predominantly older Hispanic/Latino adults from the SOL-INCA MRI study

Article References: Agah, E., Testai, F. D., Daviglus, M., Zhou, H., Isasi, C. R., Gallo, L. C., Talavera, G. A., Agudelo, C., Cordero, C., Sotres-Alvarez, D., Tarraf, W., Gonzalez, H., DeCarli, C., & Maillard, P. (2026). Association between vascular risk factors and white matter integrity among diverse predominantly older Hispanic/Latino adults from the SOL-INCA MRI study. GeroScience. https://doi.org/10.1007/s11357-026-02503-5

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02503-5

Keywords: white matter integrity, cerebral small vessel disease, hypertension, diabetes, Hispanic/Latino health, SOL-INCA, MRI, free water imaging, fractional anisotropy, white matter hyperintensities, brain aging, health disparities