MRI Shows Early Brain Aging in Patients With Poor Heart Health

brain health; brain aging: brain health; brain agingAccording to new research, using MRI with a machine learning model can detect signs of premature brain aging in patients in their 30s who are in less-than-optimal heart health. “A neuroimaging-based biomarker termed the brain age is thought to reflect variability in the brain's aging process and predict longevity,” study authors wrote in The Lancet Healthy Longevity (2022; https://doi.org/10.1016/S2666-7568(22)00167-2). “Using Insight 46, a unique narrow-age birth cohort, we aimed to examine potential drivers and correlates of brain age.” Noting the connection between aging and substantial interindividual effects on function, morbidity, and mortality, the researchers wrote that a reliable cross-sectional metric that can quantify this variability—a measure of biological age—would be “valuable both for clinical practice and research into longevity and aging health.” Such a metric, they said, could facilitate the monitoring of age-related changes beyond those captured by disease-specific risk factors by incorporating mechanisms of decline due to both disease and typical aging. Likewise, the metric could aid in the detection of patients who are aging more rapidly than expected before the onset of clinical manifestations, as well as being able to detect traits related to delayed aging, cognitive maintenance, and longevity. For the study, the research team—which included investigators from UCL Queen Square Institute of Neurology—sought to examine the potential links between older brain age and poorer cardiovascular health. The group used an MRI-based machine learning algorithm to evaluate the brain age of 456 British patients born in a single week in 1946. Participants had 24 prospective waves of data collection to date, including MRI and amyloid PET imaging at approximately 70 years old. Using MRI data from a previously defined selection of this cohort, the researchers derived brain-predicted age from an established machine-learning model, trained on 2,001 healthy adults between the ages of 18 and 90. Subtracting from chronological age at the time of assessment provided the brain-predicted age difference (brain-PAD), with the investigators testing associations with data from early life, midlife, and late life, as well as rates of MRI-derived brain atrophy. The mean brain-predicted age among participants was 67.9 years. Overall, female sex was associated with a 5.4 years younger brain-PAD than male sex. An increase in brain-PAD was also linked to increased cardiovascular risk at age 36 and age 69, increased cerebrovascular disease burden, lower cognitive performance, and increased serum neurofilament light concentration. Higher brain-PAD was also connected to future hippocampal atrophy over a subsequent 2-year period. Such findings “support brain-PAD as an integrative summary metric of brain health,” the authors wrote, “reflecting multiple contributions to pathological brain aging, and which might have prognostic utility.” Research has previously shown that measures of the brain's biological age relate to disease diagnosis, prognosis, and health outcomes, noted James Cole, PhD, Associate Professor of Neuroimage Analysis at University College London, and a co-author of the study. With this research, “we wanted to see what historical and contemporary measures related to brain age, using the unique Insight 46 study dataset,” he stated, adding that the team hypothesized that measures of physiological and neurological health would relate to brain age, particularly cardiovascular health and neurofilaments. The machine learning brain-age model Cole and his colleagues relied on uses a “very large training dataset” of brain MRI scans from healthy people to identify the statistical pattern between brain scans and age. “Once the model is trained in this way, it is tested in other people, to determine whether chronological age can be accurately predicted from brain MRI scans,” he continued. “Our model has previously shown to be highly accurate, so it can be used to estimate biological brain age in other people using their brain MRI scans. In instances in which brain age appears older than chronological health, “this suggests people have poorer brain health than would be expected at that age,” Cole said. “The model works well because the MRI scans are very good at detecting brain atrophy, which is something that happens to everyone with aging, though the atrophy occurs at different rates in different people. And the model is able to account for that.” Ultimately, brain MRI has the potential to be used as a screening tool to detect whether patients who have other risk factors for poorer aging health are also at risk for cognitive decline and neurodegenerative diseases, Cole noted, “as brain changes often occur many years before cognitive decline is noticeable.” And, while saying that MRI is unlikely to be used in outwardly healthy individuals in their 30s, Cole stated that the study's results “suggest that maintaining a healthy heart throughout life should have a positive impact on your brain health as you get older.” Mark McGraw is a contributing writer.