Exploring the neural correlates of balance and exercise in dementia

Bajwa, Rupinder Kaur (2023) Exploring the neural correlates of balance and exercise in dementia. PhD thesis, University of Nottingham.

[thumbnail of Rupinder Kaur Bajwa_20149592_Thesis_Final.pdf]
Preview
PDF (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution.
Download (5MB) | Preview

Abstract

Balance deficits in dementia are linked to increased falls risk, leading to injury, fear of falling, reduced activity, and loss of independence. Exercise based interventions can be useful in reducing falls risk as well as improving balance in older adults with cognitive impairments. The neural mechanisms involved in balance and exercise in this population are not well understood. The National Institute for Health Research (NIHR) funded Promoting Activity Independence and Stability in Early Dementia (PrAISED2) Randomised Controlled Trial (RCT), is trialling a tailored exercise programme aimed to promote activity, independence, and stability in older adults with Mild Cognitive Impairment (MCI) and mild dementia. This trial presented a unique opportunity to pilot a novel Virtual Reality (VR) based balance task and investigate the neural correlates of balance in older adults with MCI and dementia. To do this, I used functional Magnetic Resonance Imaging (fMRI) a non-invasive neuroimaging technique which detects haemodynamic changes associated with neural activity at rest or during experimental tasks.

The aims of this PhD were 1) conduct a systematic review of the effects of exercise on fMRI outcomes in older adults with MCI and dementia 2) pilot a novel virtual reality-based balance fMRI task in both healthy adults and older adults with MCI or dementia 3) explore activation in response to task conditions and the relationship with balance performance in people with dementia 4) explore the relationship between resting state functional connectivity and balance performance.

Through the systematic review, I identified 12 papers from 6 studies that met the inclusion criteria. Intervention duration ranged from 21-24 weeks and included aerobic training, walking, dancing, and mind-body exercises. No study that included people with dementia was found. Exercise interventions appeared to decrease task-related connectivity and activity during motor, memory, attention, and inhibition task but increased connectivity of the dorsal attention network (DAN), hippocampus and posterior cingulate at rest.

I then recruited healthy young adults aged 18-35 to take part in a pilot fMRI study of the VR balance task. Additionally, I piloted this task in a subset of participants recruited from the PrAISED RCT, who were all older adults aged 65 and over with a diagnosis of MCI or dementia. Both groups completed the MRI tolerability questionnaire and provided feedback on task experience. Both healthy volunteers and older adults with dementia scored overall scanner experience as 4/5 for comfortableness. In both groups, I conducted exploratory whole-brain analyses exploring activation in response to each task condition (walking, obstacle navigation and postural instability) and differences in activation between the conditions. Healthy young adults displayed activation in the cerebellum, visual and motor areas. Older adults with cognitive impairments displayed activation in visual and motor cortices across the task conditions.

In the pilot study with the PrAISED participants, I also explored the relationship between task-related activation in response to each condition (walking, obstacle navigation and postural instability) with performance on static and dynamic balance assessments. Static and dynamic balance performance was associated with activation in motor regions during walking and instability conditions and the anterior cingulate cortex during the obstacle avoidance condition.

In addition to the task fMRI sequence, the participants also underwent a resting state fMRI scan. For the resting state fMRI data, I used a data-driven approach to identify common resting state networks. I explored the relationship between both intra and inter network connectivity with balance performance. Intra network connectivity of the limbic network may be associated with poorer dynamic balance performance whilst inter network connectivity between the visual network and sensorimotor network may be associated with improved dynamic balance performance.

Exercise can alter neural activity and connectivity in people with memory problems, however, future work needs to include people with more advanced dementia. Furthermore, future work should explore the optimal intensity and duration of exercise interventions to be of benefit to the patient. The work presented in this thesis has shown that participants with memory problems can engage with a VR-based task and scanning procedures in this population are well tolerated. VR based balance tasks are a promising technique to be able to improve our knowledge of the neural mechanisms involved in balance dysfunction in dementia, however further work is needed to ensure that the tasks are accessible to people with more severe cognitive impairments and functional limitations. Potential associations of intra and inter network functional connectivity with dynamic balance performance were noted, however, these did not reach statistical significance. Further investigation in larger samples and study designs with participants with differing severity of cognitive impairments is warranted to explore these interactions further.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Dineen, Robert
Van der Wardt, Veronika
Harwood, Rowan
Keywords: Dementia, Exercise, Physical activity, Balance, Postural control, Neuroimaging, fMRI
Subjects: QS-QZ Preclinical sciences (NLM Classification) > QT Physiology
Faculties/Schools: UK Campuses > Faculty of Medicine and Health Sciences > School of Medicine
Item ID: 73612
Depositing User: Bajwa, Rupinder
Date Deposited: 31 Jul 2023 04:41
Last Modified: 31 Jul 2023 04:41
URI: https://eprints.nottingham.ac.uk/id/eprint/73612

Actions (Archive Staff Only)

Edit View Edit View