Absolutely fucking useless! PREVENT THAT RISK, you blithering idiots!
Estimating Brain Age With MRI Helps Stratify Dementia Risk
A magnetic resonance imaging (MRI)-based measure of brain aging may help predict progression to dementia among patients with subjective cognitive decline (SCD), according to study findings published in Neurology.
Researchers examined whether brain-predicted age difference (brain-PAD), an MRI-derived measure comparing estimated brain age with chronological age, could provide additional prognostic information beyond standard MRI assessment among patients with SCD or mild cognitive impairment (MCI).
The researchers conducted a retrospective cohort study of patients with SCD or MCI who had available Mini-Mental State Examination (MMSE) scores and brain MRI data and at least 1 follow-up assessment within 10 years. They calculated brain age from baseline MRI using 2 pretrained approaches, BrainageR and cNeuro, and calculated brain-PAD by comparing estimated brain age with chronological age. Cox regression models assessed the association between brain-PAD and subsequent dementia while accounting for age, sex, MMSE score, visual MRI measures, and, when available, amyloid status.
MRI-derived brain-PAD predicts progression to dementia in memory clinic patients with SCD or MCI.
The analysis included a total of 799 patients, comprising 412 with SCD and 387 with MCI. Mean age was 63.3 years, 39.4% of participants were women, and the mean MMSE score was 27.4. Amyloid status was available for 679 patients, with amyloid positivity identified in 26.6% of the SCD group and 69.0% of the MCI group. Over a median follow-up period of 3.2 years, 235 patients (29.4%) progressed to dementia, including 39 with SCD and 196 with MCI.
The patients who progressed to dementia were older at baseline, had lower MMSE scores, and had greater abnormalities across visual measures of brain atrophy. Among patients with available amyloid data, 83.3% of those who developed dementia were amyloid-positive compared with 32.4% of nonconverters.
In the overall cohort, a higher BrainageR-derived brain-PAD was associated with increased risk for progression to dementia after adjustment for age, sex, and MMSE score (hazard ratio [HR], 1.06; 95% CI, 1.04-1.08; P <.001). The association remained significant after adjustment for visual MRI measures (HR, 1.04; 95% CI, 1.02-1.06; P =.003) and amyloid status (HR, 1.04; 95% CI, 1.01-1.06; P =.003).
The association differed by baseline cognitive status. Among patients with SCD, brain-PAD remained predictive after adjustment for visual MRI measures (HR, 1.09; 95% CI, 1.04-1.15; P =.003). Adding brain-PAD improved model fit (χ2, 10.17; P =.003) and increased the C-index from 0.77 to 0.79. After amyloid adjustment, brain-PAD remained associated with progression (HR, 1.08; 95% CI, 1.02-1.15; P =.021), although the C-index remained 0.84 with or without brain-PAD.
Among patients with MCI, BrainageR-derived brain-PAD was not independently associated with dementia progression after adjustment for visual MRI findings (HR, 1.01; 95% CI, 0.99-1.04; P =.351). Adding brain-PAD did not improve model discrimination.
The researchers also identified a data-driven brain-PAD threshold of -2.6 years in the SCD group. At this cutoff, the negative predictive value was 0.99 at 2 years and 0.91 at 10 years, while positive predictive values ranged from 0.05 to 0.31. The authors noted that external validation is needed before clinical use.
Study limitations include the relatively young population, which may reduce generalizability to older patients.
“MRI-derived brain-PAD predicts progression to dementia in memory clinic patients with SCD or MCI,” the study authors concluded. They continued, “In particular, brain-PAD provided additional predictive value to visual rating scales in patients with SCD.”
Disclosures: This research was supported by Health~Holland, Stichting Alzheimer Nederland, Stichting Steun Alzheimercentrum Amsterdam, Houbolt Stichting, and the Noaber Foundation. Multiple study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of disclosures.
