Cognitive dysfunction and multiple sclerosis: developing a rationale for considering the efficacy of exercise training

Introduction

Multiple sclerosis (MS) is both an inflammatory and neurodegenerative disease of the central nervous system (CNS) with a prevalence of 1 per 1000 persons in the United States.^1,2 This disease is initially characterized by periods of inflammation that result in multi-focal demyelination and transection of axons in the CNS,^3,4 and eventually results in axonal loss and gray matter atrophy.^5,6 Gray matter atrophy, in particular, is important for the accumulation of cognitive impairment over time in MS.^7–9 Indeed, cognitive impairment is a prevalent, disabling, and poorly managed consequence of MS. This brief review will develop a rationale for considering the efficacy of exercise training as a means of ameliorating cognitive dysfunction in MS. Such a review is unique because it focuses on exercise training and cognitive impairment in MS – a nascent, but promising line of inquiry with potential profound implications for clinical practice and research.

Why focus on cognitive impairment in MS?

Cognitive impairment occurs in between 43% and 65% of patients with MS,^10–12 and a common complaint in this population is slowed mental processing speed and impaired memory.^11,13 Patients with MS typically present with poor performance on neuropsychological tests that emphasize mental processing speed and episodic memory.^11,14,15 There are further deficits in executive function, but with a lower frequency.^16,17 Cognitive impairment influences quality of life, instrumental activities of daily living, and results in unemployment.^18,19 The life-altering effects of cognitive impairment in persons with MS often occur without the presentation of physical disability.^20–22

There has been significant effort directed toward developing efficacious pharmacological and non-pharmacological strategies for managing cognitive impairment in MS. As of yet, there is no FDA-approved treatment or evidence-based rehabilitation strategy for managing cognitive dysfunction in this population.^23,24 As an example, one randomized, double-blind, placebo-controlled trial involving 151 persons with MS who underwent a 29-day oral dose of 30 mg of L-amphetamine sulfate (or placebo) reported no significant improvements on subjective ratings of cognition or information processing speed. ²⁵ Another randomized, double-blind, placebo-controlled trial involving 54 persons with MS who received memantine, an effective treatment for Alzheimer’s disease, reported no statistically significant results from any of the cognitive assessments. ²⁶ Other research has reported ambivalent effects of donepezil on memory in MS. ²⁷ Collectively, the prevalence, impact, and poor management of cognitive impairment underscore the importance of identifying other promising approaches for managing this manifestation of MS.

Exercise training, cognition, and gerontology: Is there enough evidence to encourage similar research in MS?

Cognitive impairment is a common and disabling part of the aging process, ²⁸ as it is in MS. But, there is a wealth of evidence that aerobic fitness, physical activity, and exercise training are associated with better cognitive function in older adults.^29–31 Before reviewing this literature, we provide working definitions of aerobic fitness, physical activity, and exercise training ³² for clarity and consistency in terminology throughout the remainder of this review. Aerobic fitness is a characteristic of a person that reflects one’s capacity for and history of engaging in regular physical activity or aerobic exercise training. Physical activity generally describes a behavior that reflects both bodily movement and energy expenditure, whereas exercise training is a type of planned, structured, and repetitive physical activity that is done in order to improve health-related aspects of fitness. All three have been associated with better cognitive function in older adults.

The initial research on exercise training and cognition involved cross-sectional studies of the associations between aerobic fitness and physical activity with cognitive function in older adults. This literature has been summarized in a literature review ³³ and a meta-analysis, ³⁴ with both indicating that adults who had higher aerobic fitness performed better on cognitive tasks than adults with lower fitness. The cross-sectional nature of this early research presented major limitations for interpreting direction of causality, but the data did provide the preliminary backdrop for subsequent randomized controlled trials (RCTs) examining aerobic exercise training effects on cognitive function in older adults.^35,36 Again, we do not review single studies because the overall effect of exercise training on cognition already has been summarized in literature reviews^30,37 and three meta-analyses.^29,38,39 The meta-analyses are discussed, rather than literature reviews, because of the quantitative nature of the papers. The first meta-analysis reported a mean overall effect size of 0.478 favoring exercise training versus the effect size for the control condition of 0.164, and large effects were observed for both aerobic and combined aerobic and resistance exercise training, exercise training sessions lasting 30 or more minutes in duration, interventions lasting six or more months in duration, and samples that mostly included women. ²⁹ Importantly, the effect of exercise training on cognition was smallest for simple reaction time, moderate for information processing speed, attention, and learning and memory, and largest for executive control tasks. ²⁹ The second meta-analysis reported that exercise training (consisting of a variety of aerobic, strength, and mobility-related exercises) was associated with improved cognitive performance in older adults with documented cognitive impairment, with an average effect size of 0.59. ³⁸ The third meta-analysis reported modest, but significant improvements in cognition with exercise training among older adults, and the improvements were similar across domains of attention and processing speed, memory, and executive functioning. ³⁹ Overall, this body of research supports the position that aerobic fitness, physical activity, and exercise training are beneficially associated with cognitive function in older adults.

The beneficial changes in cognition associated with aerobic fitness and exercise training have coincided with alterations in brain cortical structure and function, based on magnetic resonance imaging (MRI) protocols.^40,41 For example, one study ⁴⁰ examined the association between aerobic fitness (estimated VO_2max from the Rockport 1-mile walk protocol ⁴² ) and gray and white matter density maps using MRI and voxel-based morphometry (VBM) in a sample of 55 older adults. Aerobic fitness was associated with the amelioration of age-related declines in brain gray and white matter density. ⁴⁰ Importantly, aerobic fitness displayed the greatest sparing of anterior white matter tracts and gray matter density in prefrontal and temporal cortices; those regions and tracts are critical in cognition.^7–9 Another study examined the effect of aerobic exercise training on brain volume in a RCT of 59 older adults. ⁴¹ Participants were randomized into either a 6-month-long aerobic exercise training intervention or stretching-and-toning control group, and brain volume was assessed with MRI and VBM for both baseline and follow-up. ⁴¹ Older adults who completed the aerobic exercise training intervention, but not the control condition, had statistically significant increases in brain volume in both gray and white matter regions, particularly in the prefrontal and temporal cortices. ⁴¹

Collectively, the evidence indicates that exercise training improves cognitive function, particularly information processing speed, learning and memory, and executive functions, perhaps through alterations in brain structure and function, among older adults. Such cognitive domains, and the associated alterations in brain structure and function, ³¹ correspond with the cognitive impairments seen in persons with MS. We further note that MS might mimic an accelerated aging process, and many of the other consequences of exercise training in MS mimic those seen in older adults (for example, quality of life, mobility, and motor function); the later observation will be discussed in a subsequent section of this brief review. Such observations, along with the degree and consequences of cognitive impairment, likely motivated the initial interest in exercise training, cognition, and MS.

Exercise training, cognition, and MS: what is the current state of evidence?

Exercise training might represent a behavioral approach for managing cognitive dysfunction in persons with MS, but this area is nascent and has not been aggressively pursued. To date, we are aware of two studies that examined aerobic fitness as a correlate of cognitive function;^43,44 one study that investigated physical activity and its association with perceived cognitive function; ⁴⁵ and one RCT that examined the effect of exercise training on cognition. ⁴⁶ We describe these studies in detail below because of the minimal number and lack of an existing literature review or meta-analysisin MS, unlike the wealth of evidence for older adults.

As done in gerontology, our team initially examined the association between aerobic fitness, cognitive function, and brain activation using functional MRI (fMRI) in a small sample of 24 females diagnosed with relapsing–remitting MS. ⁴³ Aerobic fitness was assessed using a maximal incremental exercise test on a cycle ergometer with analysis of expired gases using a metabolic cart for the measurement of peak oxygen consumption (VO_2peak). Cognitive status was assessed using a battery of neuropsychological tests including the Kaufman Brief Intelligence Test ⁴⁷ and Rao’s Brief Repeatable Battery of neuropsychological tests ⁴⁸ including the Selective Reminding Test, 10/36 Spatial Recall Test, Symbol Digit Modalities Test (SDMT), Paced Auditory Serial Addition Test (PASAT), and aword list generation task. Cognition was further assessed using an experimental version of the PASAT presented in the visual modality during fMRI. Collectively, aerobic fitness measured by VO_2peak was significantly correlated with PASAT and its visual analog during fMRI. Aerobic fitness was further associated with differential activation of the right inferior frontal gyrus and middle frontal gyrus (regions of the cerebral cortex) and anterior cingulate cortex.

This initial study was followed by an examination of the relationship between aerobic fitness, cognition, and white matter integrity and gray matter volume in a sample of 21 persons with relapsing–remitting MS. ⁴⁴ Aerobic fitness and cognitive function were measured as in our aforementioned study, and white matter integrity and gray matter volume were measured using structural MRI with optimized VBM. Aerobic fitness measured as VO_2peak was associated with better performance on a composite measure of information processing speed (for example, PASAT and SDMT) as well as lesion load volume, greater gray matter volume, and white matter integrity. The initial studies largely parallel the findings from gerontology^{29–31,37,40,41} whereby aerobic fitness was beneficially associated with cognitive performance, brain activation, and white and gray matter in the brain.

As there is evidence from the field of gerontology that physical activity might moderate the decline in cognitive functioning that occurs with increasing age, we next examined the association between physical activity and perceived cognitive impairment in persons with relapsing–remitting MS. ⁴⁵ The sample for this study included 82 persons with relapsing–remitting MS who completed a battery of questionnaires that included the Perceived Deficits Questionnaire (PDQ) and then wore an accelerometer for 7 days and completed the Godin Leisure-Time Exercise Questionnaire (GLTEQ), a validated, self-report measure in persons with MS. Physical activity measured by either an accelerometer or the GLTEQ was associated with overall PDQ scores, even after controlling for clinical and demographic factors. Admittedly, the PDQ is not an ideal measure of self-reported cognitive deficits, as scores have been associated with both performance on neuropsychological tests and mental health status. ⁴⁹ The association between physical activity and cognition has been recently confirmed in two unpublished studies using neuropsychological measures of information processing speed (that is, SDMT and PASAT) in persons with MS.^50,51 Such results are promising and consistent with findings from gerontology and, when combined with the aforementioned research on aerobic fitness, support the need for a RCT of exercise training or physical activity as behavioral approaches for managing cognitive impairment in persons with MS.

We are aware of only one published study that has examined the effects of two modes of exercise training on cognition as a primary outcome measure in persons with MS. ⁴⁶ That study implemented a 6-month intervention whereby 57 persons with MS were randomized into one of three conditions of yoga, aerobic exercise, or a wait-list control. The yoga condition consisted of a 90-min session, once per week, with encouragement of home practice. The aerobic exercise condition consisted of very light or moderate intensity, group exercise on a stationary bicycle once per week, and this was supplemented by the encouragement of home exercise and periodic option of exercising on a Swiss ball. Cognition was assessed before and after the 6-month intervention period using a battery of neuropsychological tests, including the Stroop Color and Word Test ⁵² and PASAT. Neither the yoga group nor the aerobic exercise group had statistically significant improvements in any of the cognitive measures compared with the wait-list control group. Such null results might be expected given the minimal intensity (for example, 2–3 on the Borg Rate of Perceived Exertion (RPE) scale) of supervised aerobic exercise performed one day per week, poor participation in the weekly classes for both experimental conditions (that is, 65% adherence rate), and unsupervised home-based exercise prescription. This study and its methodology would not be on par with the rigor of research on exercise training and cognition in older adults.

Exercise training and other consequences in gerontology and MS: is there enough similarity in results to support additional examinations of exercise and cognition in MS?

As there is limited evidence on exercise training and cognition in MS for direct comparison with gerontology, we believe it prudent that research is briefly presented on the comparability of other consequences of exercise training between older adults and persons with MS, notably mobility, motor function, and quality of life. This would lay the foundation for supporting the likelihood of a beneficial effect of exercise training on cognition in MS as seen in gerontology. Regarding mobility, one meta-analysis of 19 RCTs involving 30 exercise training interventions in older adults reported effects sizes of 0.26 and 0.21 for walking speed and endurance, respectively. ⁵³ We reported in a meta-analysisof 22 exercise training interventions an overall effect size of 0.24 for improvements in quantitative measures of walking mobility in persons with MS. ⁵⁴ With motor function, researchers in gerontology have reported that cognitive and motor function are inter-related, perhaps because of combined effects of physical activity and exercise behavior, ⁵⁵ and we recently speculated on a similar possibility in MS. ⁵⁶ Regarding quality of life, one meta-analysis of 36 exercise training interventions in older adults reported an effects size of 0.24 for psychological well-being. ⁵⁷ We reported in a meta-analysis of 13 exercise training interventions an overall effect size of 0.23 for improvements in quality of life among persons with MS. ⁵⁸ The similarity of beneficial effects for exercise training on walking mobility, motor function, and quality of life in older adults and persons with MS strengthens the rationale for considering exercise training as an approach for improving cognition in persons with MS.

Where should future research begin?

As initially undertaken in gerontology, there already exist cross-sectional data indicating that aerobic fitness and physical activity are associated with better cognitive function in persons with MS. This pilot work suggests that the field is now ready for a well-designed RCT of exercise training and cognition in MS. Perhaps the best launching pad for designing such a RCT in MS would derive from a meta-analysis identifying exercise training parameters that maximized cognitive benefits in older adults. ²⁹ That meta-analysis identified modes of both aerobic and aerobic plus resistance training, programs that exceed 6 months in duration, and training sessions lasting between 30 and 45 min in duration as resulting in the largest improvements in cognition among older adults. There was no moderator analysis reported for the intensity and frequency of training, but many studies of aerobic exercise have used moderate-intensity exercise (for example, an intensity between 50% and 75% heart rate reserve) performed on three or more days per week.^41,59 The majority of studies have further included structured or ‘supervised’ exercise rather than home-based exercise for maximizing and verifying compliance with the training regimen, and the control condition has often included a supervised, minimal exercise attention control for capturing possible confounding effects of social interaction and attention with the active treatment arm. We believe, therefore, that an excellent RCT might involve a 6-month period of moderate-intensity, aerobic exercise training (that is, walking or cycling) performed for upwards of 45 min on 3–4 days per week using a supervised exercise leader. The ideal control condition would include a credible stimulus (that is, minimal exercise stretching and toning) delivered by an exercise leader on the same frequency and duration over a 6-month period. Such a design would be on par with the scientific rigor of research on exercise training and cognition in aging, and would provide a strong test of exercise training effects on cognition in MS.

An alternative approach, not initially undertaken in gerontology, would involve starting with additional cross-sectional research examining the association between different domains of fitness with cognitive function in MS. Such an endeavor might consider the possible independent or additive contributions of aerobic capacity, muscle strength, and balance for explaining variance in neuropsychological performance on tests of mental processing speed, episodic memory, and executive functions. This approach would then set the stage for selecting a single or combination of types or modes of exercise training for inclusion in a RCT that is designed for improving cognitive function in MS.

Conclusion: the time is ripe!

Clearly, cognitive impairment is a prevalent, disabling, and poorly managed consequence of MS. The body of research on exercise, cognition, and brain MRI in gerontology is sufficiently strong to encourage systematical examination of exercise training and cognition in MS. There is existing evidence that better aerobic fitness is associated with cognitive capacity and brain structure and function in MS. The one published study of exercise training was not positive for improving cognitive function in MS, but this could be attributed to problems with the experimental design and exercise stimuli. Collectively, the time is ripe for a research agenda on fitness, exercise training, physical activity, and cognitive function in MS. We believe that a future RCT should adopt research methods and practices from gerontology when examining exercise training and cognition in MS. This will maximize the likelihood of success and assist with generating a body of knowledge on exercise training and cognition that can ultimately have a demonstrable impact on the lives of persons with MS.