Objective: To characterise MRI-derived structural age-related brain health in Parkinson’s disease (PD) patients as a function of cognitive ability.

Background: Normal ageing produces a stereotypical trajectory of structural brain changes, which can be quantified using machine learning techniques such as Relevance Voxel Machine (RVoxM) regression. Recent RVoxM work suggests an accelerated brain ageing in Alzheimer’s disease. Other neurodegenerative diseases, such as PD, may also exacerbate this age-related trajectory and produce premature brain atrophy.

Method: The study comprised of 212 PD patients (45 – 86 years; M = 70 years; at study entry PD-N = 96, PD-MCI = 85 and PDD = 31) who were assessed and scanned at least once over an 11-year period, totalling 571 participant sessions (PD-N = 226, PD-MCI = 172, and PDD = 50). Patients received a MR-T1-weighted 3D SPGR image and MDS Level II neuropsychological assessment to determine cognitive status. Global cognitive ability was expressed as an aggregate z score derived from averaging performance from tests conducted in four cognitive domains. An RVoxM model, trained on T1 scans from 101 healthy non-PD control participants (19 – 87 years; M = 55 years; 10-fold cross validation), was applied to the PD cohort and used to estimate age at each MRI scan. The difference between estimated age relative to chronological age is referred to as the age-difference score, which was assessed across cognitive abilities within PD using Bayesian regression models; 95% uncertainty interval were used for statistical significance.

Results: Compared to the PD-N group, the mean age-difference scores of the PD-MCI group was 3.8 years (95% uncertainty interval [2.2, 5.4]) greater and the PDD group was 6.2 years [3.6, 8.6] greater, when sex and years of education where accounted for. There was no evidence of a difference between the PD-MCI and PDD groups (2.4 years [-0.05, 4.7]). PD symptom duration (but not years of education, sex, motor impairment or levodopa equivalent dose scores) was associated with the age-difference score. Across all PD patients, worsening global cognitive ability was associated with an increasing age-difference score.

Conclusion: Evidence from our study suggests that an acceleration of age-related brain atrophy may be one contributory factor for worsening cognitive ability in PD.

« Back to 2019 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/association-of-mri-derived-estimated-brain-age-and-cognitive-ability-in-parkinsons-disease/