Microstructural Brain Changes Associated With Frailty in Older Adults: Insights From Diffusion Tensor Imaging

Betul Sumbul Sekerci; Bahar Atasoy; Ozge Pasin; Irem Bilgic; Serdar Balsak; Lee Smith; Alpay Alkan; Pinar Soysal

doi:10.1016/j.jamda.2025.105738

Microstructural Brain Changes Associated With Frailty in Older Adults: Insights From Diffusion Tensor Imaging

J Am Med Dir Assoc. 2025 Jul 7:105738. doi: 10.1016/j.jamda.2025.105738. Online ahead of print.

Authors

Betul Sumbul Sekerci¹, Bahar Atasoy², Ozge Pasin³, Irem Bilgic⁴, Serdar Balsak⁵, Lee Smith⁶, Alpay Alkan⁵, Pinar Soysal⁷

Affiliations

¹ Department of Clinical Pharmacy, Faculty of Pharmacy, Bezmialem Vakıf University, Istanbul, Turkey.
² Department of Radiology, Bakırköy Sadi Konuk Research and Training Hospital, Istanbul, Turkey.
³ Department of Biostatistics, Hamidiye Medical Faculty, Health Sciences University, Istanbul, Turkey.
⁴ Department of Geriatric Medicine, Faculty of Medicine, Bezmialem Vakıf University, Fatih, Istanbul, Turkey.
⁵ Department of Radiology, Bezmialem Vakıf University Hospital, Istanbul, Turkey.
⁶ Centre for Health Performance and Wellbeing, Anglia Ruskin University, Cambridge, United Kingdom.
⁷ Department of Geriatric Medicine, Faculty of Medicine, Bezmialem Vakıf University, Fatih, Istanbul, Turkey. Electronic address: psoysal@bezmialem.edu.tr.

PMID: 40639423
DOI: 10.1016/j.jamda.2025.105738

Abstract

Objectives: This study investigates the relationship between frailty status and white matter integrity (WMI) in older adults, focusing on associations with frailty status.

Design: A cross-sectional study.

Setting and participants: 125 older adults aged ≥65 years without dementia were included.

Methods: A comprehensive geriatric assessment was conducted. Frailty status was evaluated by Fried Frailty criteria: unintentional weight loss, exhaustion, low levels of activity, weakness, and slowness: 0 for nonfrail, 1 to 2 for prefrail, and 3 to 5 for frail. Brain magnetic resonance imaging with diffusion tensor imaging (DTI) was performed, and mean diffusivity, fractional anisotropy, axial diffusivity, and radial diffusivity values were calculated in white matter integrity using a region of interest-based method. Multinomial logistic regression analyzed the multivariate effects of clinical variables on the frailty-WMI association, whereas canonical correlation analysis examined multivariate relationships between diffusion tensor imaging measurements and geriatric assessment parameters.

Results: The mean age was 81.76 ± 7.07 years, and 64% were female. Of the participants, 55.2% were frail, 33.6% prefrail, and 11.2% nonfrail. After adjusting for age and sex, WMI impairment varied in the cingulum, inferior longitudinal fasciculus, and forceps major in the left hemisphere among frailty statuses. Regression analysis showed demyelination of the left cingulum [odds ratio (OR) 0.97, 95% CI 0.95-0.99] and left forceps major (OR 0.99, 95% CI 0.98-0.99) was linked to increased frailty risk. For the frail group, DTI variables and malnutrition were significant. An increase in malnutrition status or risk raised the risk of frailty by 18.86 times. Canonical correlation analysis identified the Lawton-Brody IADL and Epworth Sleepiness Scale scores as the strongest predictors, whereas the cingulum showed the highest association among white matter tracts.

Conclusions and implications: Frailty and prefrailty were associated with deteriorated WMI, particularly in the left cingulum and left forceps major. The strong link between frailty and WMI may reflect shared underlying mechanisms, including sleep, functionality, and nutrition. Therefore, a multidimensional approach to frailty is important.

Keywords: DTI; Frailty; microstructural brain changes; prefrailty.