Identifying Inter-Individual Differences in Cognitive Decline Using the Brain Connectome in Osteoporosis.

Abstract

Background: Osseous structures have been recognized as an endocrine organ that bidirectionally interacts with the brain. Osteoporosis (OP) is a systemic endocrine disorder linked to neurodegenerative disorders. This bone-brain axis interdependence highlights the necessity of cognitive monitoring in OP management to detect early neurodegeneration markers, particularly given individual variability in brain reserve that may predispose patients to accelerated cognitive decline.

Purpose: To investigate the individual differences in functional connectome and its association with cognitive ability in OP.

Study design: Longitudinal human study.

Subjects: A total of 31 OP patients (Age: 56.7 ± 13.2, 17 Male) and 31 healthy controls (HC, age: 55.1 ± 11.3, 15 male).

Field strength/sequence: 3 T, gradient-echo EPI sequence, MP2RAGE sequence.

Assessment: Individual identification analyses were performed to investigate the individual-specific pattern of brain functional connectome in both OP and HC by leveraging longitudinal test-retest fMRI data to map individual variabilities in brain functional connectomes. Cognitive abilities were assessed using the Montreal Cognitive Assessment (MoCA) and Mini-Mental State Examination (MMSE).

Statistical tests: Two-sample t tests, support vector regression, permutation tests, and Bonferroni correction. A p < 0.05 was considered statistically significant.

Results: Significant inter-individual variability (SD_OP = 3.27, SD_HC = 2.35) and robust intra-individual stability (P_OP = 0.17, P_HC = 0.59) in cognitive performance for both OP and HC groups were observed. In addition, functional connectivity profiles could reliably identify individuals across sessions (SuccessRate, SR_OP = 85%, SR_HC = 92%). The support vector regression model revealed that connectivity profiles could predict cognitive ability both within (r_MoCA-OP = 0.63, r_MMSE-OP = 0.54, r_MoCA-HC = 0.58, r_MMSE-HC = 0.61) and between sessions (r_MoCA-OP = 0.47, r_MMSE-OP = 0.41, r_MoCA-HC = 0.53; r_MMSE-HC = 0.54), with the medial-frontal and default-mode networks emerging as the most predictive contributors.

Conclusion: These findings underscore the potential of resting-state functional connectomes characterizing individual variability for cognitive ability in OP patients.

Evidence level: 4.

Technical efficacy: Stage 2.