Effect of a 24-month physical activity program on brain changes in older adults at risk of Alzheimer's disease: the AIBL active trial

Highlights

• Results of a 24-month physical activity intervention in adults at risk of cognitive decline.

• Minutes of physical activity per week was different between groups only at 6 months.

• No intervention effect on primary outcome measure “white matter hyperintensities.”

• Cognitive status or beta amyloid status did not affect outcome.

• Secondary outcome measure “hippocampal volume loss” was not affected by intervention.

Abstract

White matter hyperintensities (WMHs) are a risk factor for cognitive decline. Physical activity (PA) is associated with lower WMH. Whether long-term exposure to PA programs has beneficial effects on WMH progression in older adults with memory complaints and comorbid conditions has had limited exploration. This study explored whether a 24-month moderate-intensity PA intervention can delay the progression of WMH and hippocampus loss in older adults at risk for cognitive decline. Data acquired on magnetic resonance imaging were used to measure the progression of WMH and hippocampus loss. The results of this study showed no effect of intervention on either the primary outcome measure “WMH” or the secondary outcome measure “hippocampal volume.” In addition, neither beta amyloid status nor the adherence to the intervention had any effect on the outcome. In this cohort of subjective memory complaints and mild cognitive impairment participants with vascular risk factors, there was no effect of long-term moderate-intensity PA on WMH or hippocampal loss.

Introduction

Delaying the onset of dementia would reduce health care costs and benefit the quality of life of people at risk of dementia. Studies exploring the benefit of modifiable interventions such as physical exercise, cognitive activities, and diet have raised the hypothesis that these factors may enhance brain plasticity (Livingston et al., 2017). Several hypotheses have emerged from animal studies to explain the cognitive improvement effects of physical activity (PA). These include neurogenesis (van Praag et al., 2005), increased cerebral blood flow, reduced apoptosis (Kim et al., 2010), and reduced brain beta amyloid burden (Stranahan et al., 2012).

In human studies, it has been shown that PA provides health benefits essential for maintaining cognition and functional fitness in older age (Lautenschlager et al., 2012). After education and smoking, physical inactivity is considered to be the third largest modifiable contributor (around 17%) to dementia cases (Norton et al., 2014). Epidemiological studies have also demonstrated that PA is a protective factor against cognitive impairment in the oldest old (Middleton et al., 2010, Sumic et al., 2007). However, not all studies have shown similar results. In cognitively normal, PA has only been shown to be beneficial in improving reaction times in apolipoprotein E (APOE) ε4 carriers (Deeny et al., 2008). Another study (Sturman et al., 2005) has also suggested that without additional cognitively stimulating activities, PA does not protect against cognitive decline. Alzheimer's disease (AD) and vascular dementia have both been linked to vascular risk factors (VRFs) (Knopman and Roberts, 2010) such as prevalence of diabetes and cardiovascular disease. PA can decrease the effect of VRFs that are associated with cognitive decline in older adults (Uemura et al., 2012). Only a few randomized controlled trials (RCTs) have reported a positive effect of PA on cognitive function in older adults with healthy cognition, subjective memory complaints (SMCs), mild cognitive impairment (MCI), and dementia (Gates et al., 2013, Lautenschlager et al., 2008). No studies have reported on the effects of a long-term PA program on cognition in older adults with MCI or SMC who also have least one VRF.

Neuroimaging studies have reported mixed effects on brain volumes due to PA. Hippocampal atrophy is considered a marker of neurodegeneration (Jack et al., 2011). An increase of hippocampal volume by 2% was found in a randomized trial of 1-year aerobic training in 120 cognitively healthy older participants (Erickson et al., 2011). A recent trial using high-field imaging showed that a 24-month program of moderate-intensity PA in 26 participants was associated with larger hippocampal volumes (Rosano et al., 2017). An observational study conducted over 10 years showed higher rates of exercise were associated with larger superior frontal lobes and a reduced rate of medial temporal lobe atrophy (Bugg and Head, 2011). Furthermore, other observational studies have shown that the association of higher fitness levels and reduced brain atrophy are seen in AD but not in normal aging (Burns et al., 2008). By contrast, a recent cohort of cognitively normal participants followed over 10 years showed that those with low PA at baseline were at higher risk of lower brain volumes and 1.5 times higher risk of having a diagnosis of dementia at follow-up (Tan et al., 2017).

White matter hyperintensity (WMH) is another neuroimaging marker that is associated with cognitive decline, dementia, and older age (DeCarli et al., 1995, Prins and Scheltens, 2015). A recent study (Sudre et al., 2017) demonstrated that the rates of accumulation of WMH were higher in AD than cognitively normal (controls accumulated 5.8% per year compared with 14.7% per year for AD participants). Studies have also demonstrated an association between WMH and APOE ε4 status in participants with preclinical AD (Kandel et al., 2016).

In patients with AD (Nelson et al., 2007, Richard et al., 2010), progression of WMH over a two-year period has been shown to be delayed in individuals receiving a vascular care intervention that included physical exercise and medication. Other longitudinal studies have failed to find an effect of PA with WMH progression (Podewils et al., 2007). Multiple cross-sectional studies exploring the effects of PA on WMH have found no significant association (Burzynska et al., 2014, Fleischman et al., 2015, Rosano et al., 2010, Tseng et al., 2013). However, other studies have shown that higher intensity of PA was associated with reduced WMH volume (Wirth et al., 2014). A recent review article (Torres et al., 2015) of twelve studies examining the relationship between global measures of WMH and PA found mixed results with six studies reporting no association.

The main purpose and novel aspect of the present study was to conduct a rigorous RCT to determine whether PA over a 24-month period can delay the progression of WMH as measured by magnetic resonance imaging (MRI) in older adults with SMC or MCI and VRFs. A secondary objective was to investigate the effects of PA on progression of hippocampal volume loss. Finally, a third objective was to explore possible interaction of baseline beta amyloid measured using positron emission tomography (PET) imaging on MRI measures. Our study is the largest study exploring the effect of 24-month PA intervention on neuroimaging markers in community-dwelling older adults with MCI/SMC who have at least one VRF (Australia New Zealand Clinical Trial Registry, ACTRN, 12611000612910).