Exercise and cognition in multiple sclerosis: The importance of acute exercise for developing better interventions

Accepted Manuscript Title: Exercise and Cognition in Multiple Sclerosis: The Importance of Acute Exercise for Developing Better Interventions Author: Brian M. Sandroff PhD PII: DOI: Reference:

S0149-7634(15)30181-0 http://dx.doi.org/doi:10.1016/j.neubiorev.2015.10.012 NBR 2296

To appear in: Received date: Revised date: Accepted date:

17-9-2015 19-10-2015 31-10-2015

Please cite this article as: Sandroff, B.M.,Exercise and Cognition in Multiple Sclerosis: The Importance of Acute Exercise for Developing Better Interventions, Neuroscience and Biobehavioral Reviews (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.10.012 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Exercise and Cognition in Multiple Sclerosis: The Importance of Acute Exercise for Developing Better Interventions

Kessler Foundation, West Orange, New Jersey

us

cr

1

ip t

Brian M. Sandroff, PhD1

an

Correspondence to Brian M. Sandroff, 300 Executive Drive, Suite 70, West Orange, NJ, 07052,

te

d

M

USA, phone 1-973-324-8388, fax 1-973-324-8362, e-mail [email protected]

Abstract

Ac ce p

Cognitive dysfunction is highly prevalent, disabling, and poorly-managed in persons with multiple sclerosis (MS). Exercise training represents a promising approach for managing this clinical symptom of the disease. However, results from early randomized controlled trials of exercise on cognition in MS are equivocal, perhaps due to methodological concerns. This underscores the importance of considering the well-established literature in the general population that documents robust, beneficial effects of exercise training on cognition across the lifespan. The development of such successful interventions is based on examinations of fitness, physical activity, and acute exercise effects on cognition. Applying such an evidencebased approach in MS serves as a way of better informing exercise training interventions for improving cognition in this population. To that end, this paper provides a focused, updated review on the evidence describing exercise effects on cognition in MS, and develops a rationale and framework for examining acute exercise on cognitive outcomes in this population. This will

1 Page 1 of 39

provide keen insight for better developing exercise interventions for managing cognitive

ip t

impairment in MS.

cr

Keywords: Exercise; Cognition; Multiple Sclerosis; Acute Exercise; Executive Functioning; Walking

an

Exercise training is a promising approach for managing cognitive impairment in MS However, research on exercise and cognition in MS is still in its infancy It is important to develop better exercise protocols to improve cognition in MS One way to do this involves examining the acute effects of exercise on cognition This has been done in the general population, and should be applied in MS research

M

• • • • •

us

Highlights

1. Introduction

d

Multiple sclerosis (MS) can be described as a non-traumatic, immune-mediated and

te

neurodegenerative disease of the central nervous system (CNS). This disease is initially

Ac ce p

characterized by multi-focal areas of inflammation resulting in demyelination in the CNS [1]. The disease process eventually causes irreversible damage (i.e., neurodegeneration) of both grey and white matter in the brain, resulting in the accumulation of both physical and cognitive disability [1].

Cognitive impairment is highly prevalent in MS. Upwards of 50% of patients demonstrate cognitive impairment based on performance on neuropsychological tests [2,3], which primarily manifests as slowed cognitive processing speed, impaired learning and memory, and executive dysfunction, but less so as impairment in intellectual functions and language skills [2-4].

2 Page 2 of 39

Impairment in those cognitive domains might be more pronounced in persons with progressive presentations of the disease compared with relapsing-remitting MS [5]. Cognitive impairment is a highly disabling consequence of MS and has been associated

ip t

with many negative health outcomes. For example, MS-related cognitive dysfunction is

cr

associated with depression [6]. Cognitive impairment and decline have further been associated with unemployment and loss of employment [7,8], loss of driving abilities [9], and reduced

us

social functioning [5] in persons with MS. There is additional, longitudinal evidence that

an

decreased performance on a battery of neuropsychological tests is associated with loss of independence, and increased need for substantial assistance with activities of daily living [10].

M

Cognitive impairment is poorly managed in persons with MS. There is limited support for disease-modifying (e.g., interferon-beta-1a, interferon-beta-1b, glatiramer acetate,

d

natalizumab) and symptomatic (e.g., L-amphetamine sulfate, modafinil, donepezil)

te

pharmacological treatments for mitigating cognitive impairment, and this is largely based on

Ac ce p

methodological concerns in several randomized controlled trials (RCTs) [11]. There further is a growing body of literature describing overall conflicting evidence for the efficacy of cognitive rehabilitation interventions on cognitive dysfunction in this population [11]. Indeed, one Cochrane review and meta-analysis highlighted the heterogeneity of cognitive rehabilitation intervention characteristics for improving various domains of cognition, leading to overall small and inconsistent effects of cognitive rehabilitation on cognition in MS [12]. Since that meta-analysis, there have been recent, promising data supporting cognitive rehabilitation for improving several domains of cognition and brain structure and function in MS [13]. For example, several targeted interventions have reported improvements in learning and

3 Page 3 of 39

memory [14,15] and concomitant increases in hippocampal activation [16,17] and functional connectivity [15]. Alternatively, there is less convincing evidence supporting targeted cognitive rehabilitation interventions on cognitive processing speed and executive

ip t

functioning in MS [See 18 for review]. Indeed, one review highlighted the complete lack of

cr

published cognitive rehabilitation interventions directly targeting slowed cognitive processing speed in MS, despite this being the most commonly-impaired cognitive domain [11]. In

us

addition, this review outlined the lack of convincing evidence for targeted cognitive

an

rehabilitation interventions for executive dysfunction in MS [11]. However, there seems to be emerging evidence supporting tailored cognitive remediation for improving MS-related

M

executive dysfunction [19], in addition to improving neural activation [20] and functional connectivity [21] in regions and networks associated with executive functioning. However,

d

cognitive rehabilitation is difficult to apply outside of clinical settings (i.e., low ecological

te

validity) and does not generally result in physical health benefits beyond improved cognition

Ac ce p

[18]. This underscores the consideration of more ecologically valid behavioral approaches for managing cognitive impairment in MS that can result in many physical health benefits. One such approach involves exercise training [22]. Exercise training is defined as planned, structured, repetitive physical activity behavior aimed at improving or maintaining an individual’s physical fitness (i.e., characteristics of an individual’s capacity to perform physical work) [23]. Though this behavior has been identified as a promising approach for managing cognitive dysfunction in MS, this area has been understudied. One review paper highlighted null results of early RCTs of exercise training on cognition in MS, and concluded that higher-quality research was needed [22]. Based on this

4 Page 4 of 39

concern, that review paper outlined a conceptual approach for better examining exercise training effects on cognition in patients with MS [22]. This mainly involved considering the wellestablished literature involving exercise training effects on cognition in older adults [24] as a

ip t

foundation for designing well-informed RCTs in MS. Conceptually, by carefully evaluating the

cr

wealth of evidence supporting exercise for improving cognition in older adults, researchers could systematically identify features of an exercise training intervention for optimally

us

improving cognition in persons with MS, thus better planning future RCTs. Since the publication

an

of that review, there have been several investigations involving exercise, physical fitness (as a cross-sectional surrogate for exercise training), physical activity (i.e., behavior reflecting

M

skeletal muscle contraction resulting in increased energy expenditure beyond resting levels) and cognition in MS; however, there are still no seminal RCTs to provide convincing evidence

d

for the effects of exercise training on cognition in MS (Table 1). Therefore, it is necessary to

te

continue to inform the development of such a potential RCT by delineating features and

Ac ce p

mechanisms of a future optimal exercise training intervention for improving cognition in MS. To that end, this paper provides a focused, updated review on the evidence of beneficial effects of exercise, physical fitness, and physical activity on cognition in MS, and develops a rationale and framework for examining the effects of acute exercise on cognition in this population. This is largely based on literature in the general population across the lifespan that describes acute exercise benefits on cognitive functioning. Although not a systematic, quantitative review, such a paper will provide keen insight for better developing exercise interventions for managing cognitive impairment in MS. This is timely and important given the prevalence, burden, and lack of effective treatments of this aspect of the disease.

5 Page 5 of 39

2. An Update on Exercise and Cognition in MS There is emerging interest by researchers and clinicians in the effects of exercise (i.e., planned behavior aimed at improving an individual’s fitness), fitness (i.e., characteristics of an

ip t

individual, reflecting cardiorespiratory capacity, muscular strength, balance, body composition,

cr

etc.) and physical activity (i.e., skeletal muscle contraction resulting in increased energy

expenditure beyond resting levels) on cognition in individuals with MS. To that end, this section

us

of the review provides a brief history of the evidence of exercise training, fitness, and physical

an

activity effects on cognitive function in MS, based on publication date. Of note, studies that were published prior to 2011 are included as critical background information for the systematic

M

development of an optimal exercise training intervention on cognition in MS, as proposed by a previous review [22], though the primary focus is on more recent research. A summary of the

te

MS is presented in Table 1.

d

reviewed evidence of exercise, physical fitness, and physical activity effects on cognition in

Ac ce p

Two initial RCTs examined the effects of exercise training on cognition in persons with MS. The first RCT involved a six-month intervention consisting of a weekly yoga class, weekly aerobic exercise class, or wait-list control condition in 57 individuals with mild MS disability (mean Expanded Disability Status Scale (EDSS) score = 2.9) [25]. Participants were randomized to the yoga, aerobic exercise, or control group. The yoga intervention consisted of a 90-minute session of Iyengar yoga, adapted for persons with MS, taking place once per week. The aerobic exercise intervention entailed group exercise on a stationary bicycle at the intensity of 2-3 on the Borg Rating of Perceived Exertion (RPE) scale (i.e., minimal exertion); sessions took place once per week, and home exercise was encouraged. Periodically, participants were given the

6 Page 6 of 39

option of exercising on a Swiss ball rather than cycling. Cognition was measured prior to and following the six-month period using a battery of neuropsychological tests that mainly captured various aspects of attention, cognitive processing speed, and executive function. Following the

ip t

completion of the intervention, neither the yoga group nor the exercise group demonstrated

cr

significant improvements in any cognitive measure (e.g., Stroop Color-Word Interference Test, Paced Auditory Serial Addition Test; (PASAT)) compared with the wait-list control group.

us

Importantly, there are some methodological concerns with that study. One concern is

an

that the aerobic exercise intervention was conducted at an extremely light intensity (e.g., 2-3 on the Borg RPE scale) as well as occurring only once per week, such that the maximal benefits

M

of exercise training on cognition could not be realized. In fact, the aerobic exercise stimulus was not frequent nor intense enough to meet public health guidelines for accruing general health

d

benefits (i.e., resulting in less than 150 minutes of moderate-to-vigorous physical activity per

te

week) [26]. Participation in the weekly classes of either experimental condition was quite low

Ac ce p

(65%), further curtailing the potentially beneficial effects of aerobic exercise or yoga on cognition in persons with MS. Moreover, the home exercise that was encouraged to participants was unsupervised, potentially compromising the possible benefits of the aerobic exercise intervention on cognition. This methodological concern was compounded by the absence of an intervention effect on physical fitness. Physical fitness is a key factor in evaluating the success of any exercise training intervention (i.e., manipulation check), as by definition, a goal of exercise training always involves the improvement or maintenance of physical fitness [27]. This lack of a manipulation check further brings into question the effectiveness of the exercise stimuli for optimally benefitting cognitive function in MS.

7 Page 7 of 39

The other early RCT of exercise training on cognition in MS involved 95 persons with mild MS disability (median EDSS = 2.0) who were randomized into either a six-month long exercise training intervention or wait-list control group [28]. The exercise training intervention

ip t

consisted of five resistance training sessions and five aerobic training sessions in the first 3

cr

weeks of the intervention, followed by 23 weeks of a prescribed home exercise regimen. This part of the program, prescribed by physiotherapists, entailed mostly resistance exercise (i.e.,

us

suggested to take place 3-4 times per week) with additional sessions of aerobic exercise (i.e.,

an

suggested to take place once per week). Participants were encouraged via phone to adhere to the home exercise program on four separate occasions over the six-month intervention.

M

Cognition was measured prior to and following the six-month period using only the PASAT. Accordingly, cognitive processing speed did not significantly improve in persons with MS who

d

underwent the exercise intervention.

te

One similar concern about the results of this RCT [28] is that physical function did not

Ac ce p

significantly improve in the exercise group over the six-month period (ES = −0.19), indicating that the exercise prescription was ineffective. Additionally, physical function was assessed through a self-reported measure of quality of life (MSQOL-54), rather than with an objective performance measure (i.e., cardiorespiratory fitness). Given that this exercise stimulus largely consisted of resistance training, a major study limitation is that there was no measure of muscular strength that was administered prior to and following the intervention. A possible factor for the lack of improvement of physical function (and possibly cognitive processing speed) was the under-regulation of the exercise training program from weeks 4-26 (i.e., only

8 Page 8 of 39

four instances of contact between physiotherapists and participants during a 23-week long intervention). As results from the first two RCTs of exercise training on cognition in MS [25,28] were

ip t

disappointing, perhaps due to methodological concerns, several studies then examined the

cr

cross-sectional effects of physical fitness (a presumed surrogate for exercise training) on

cognition (i.e., primarily cognitive processing speed) in MS to better inform subsequent RCTs.

us

Briefly, the first cross-sectional study reported that better cardiorespiratory fitness was

an

significantly correlated with higher scores on the PASAT (i.e., faster cognitive processing speed), but not scores on neuropsychological measures of learning and memory, in 24 persons with

M

mild MS disability (mean EDSS = 2.6) [29]. Additionally, better cardiorespiratory fitness was associated with greater activation of the right inferior frontal gyrus and middle frontal gyrus,

d

and less activation of the anterior cingulate cortex during an fMRI task. This pattern of neural

te

activation in persons with MS is similar to the pattern of activation observed during successful

Ac ce p

completion of an executive control task in high fit older adults and healthy young adults [30]. A later cross-sectional study examined the relationship between cardiorespiratory fitness, structural MRI outcomes, and cognition in a sample of 21 persons with mild MS disability (mean EDSS = 2.2) and 15 healthy controls [31]. Briefly, there were significant differences in cardiorespiratory fitness, measures of cognitive processing speed (i.e., Symbol Digit Modalities Test (SDMT) and PASAT scores), and MRI outcomes (i.e., grey matter volume based on voxel-based morphometry, and white matter integrity based on diffusion-tensor imaging, respectively) between persons with MS and healthy controls (i.e., d > 0.67). In the MS subsample, better cardiorespiratory fitness was associated with better performance on a

9 Page 9 of 39

composite measure of cognitive processing speed, but not on tests of learning and memory. This is consistent with results from the previous cross-sectional study of fitness and cognition in MS [29]. A novel result from this study was that in the MS subsample, better cardiorespiratory

ip t

fitness was associated with less lesion-load volume, greater grey matter volume, and greater

cr

white matter integrity (i.e., greater fractional anisotropy in the left posterior thalamic radiation and right anterior corona radiata). The collective results from that study suggest that

us

cardiorespiratory fitness, cognitive processing speed, and brain measures of white and grey

an

matter integrity are interrelated in persons with mild MS disability [31].

The third, more recent cross-sectional study examined the relationships among multiple

M

domains of physical fitness (e.g., cardiorespiratory fitness, muscle strength asymmetry, and balance) and cognitive processing speed in 31 persons with mild MS disability (median Patient-

d

Determined Disease Steps (PDDS) score = 2) and 31 controls matched by age, sex, height, and

te

weight [32]. Cardiorespiratory fitness was measured as peak oxygen consumption (VO2peak)

Ac ce p

using an incremental exercise test to exhaustion on a cycle ergometer; lower limb muscle strength asymmetry (i.e., relative difference in the strength of a particular muscle group between strong and weak limbs) was measured using an isokinetic dynamometer; balance was measured using static posturography. Cognitive processing speed was expressed as a composite z-score based on performance on the SDMT and PASAT. Briefly, there were significant differences in measures of fitness and cognition between persons with MS and matched controls that were moderate-to-large in magnitude (i.e., d > 0.46). In the MS subsample, worse cardiorespiratory fitness (r = .44), worse balance (i.e., larger postural sway area) (r = −.52), and greater knee extensor asymmetry (r = −.39) were significantly associated with slower cognitive

10 Page 10 of 39

processing speed. Additional noteworthy results from that study were that those three domains of fitness accounted for differences in cognitive processing speed between persons with MS and controls, and explained a statistically significant amount of variance in cognitive processing

ip t

speed (R2 = .39) in the MS subsample [32].

cr

Collectively, the cross-sectional results from studies of physical fitness and cognition in MS seemingly indicate that physical fitness is associated with cognitive processing speed, but

us

not learning and memory in persons with mild MS disability [29,31,32]. This suggests that

an

enhancing cardiorespiratory fitness, balance, and muscular strength might represent avenues for improving slowed cognitive processing speed in this population. Importantly, results from

M

cross-sectional studies of fitness and cognition, described above, are not consistent with the null results from previous RCTs of exercise training and cognition in MS. Despite the null results

d

of the early RCTs, the cross-sectional research on fitness and cognition seems to suggest

te

potential benefits of aerobic exercise, and perhaps resistance and balance exercise, on

Ac ce p

cognitive processing speed in persons with MS. One recent RCT recognized the importance of physical fitness and examined the effects of a supervised, aerobic exercise training intervention on fitness and cognitive function in 42 persons with progressive MS who had moderate disability (mean EDSS = 4.95) [33]. Participants were randomly assigned to one of three aerobic exercise training conditions (i.e., arm ergometry, rowing, cycle ergometry) or a wait-list control condition. All participants underwent an incremental exercise test to exhaustion to determine cardiorespiratory fitness at baseline and again at follow-up. Further, all participants underwent a battery of neuropsychological tests for measuring cognitive processing speed (SDMT), verbal learning and memory (Verbal

11 Page 11 of 39

Learning and Memory Test; VLMT), attention (Test Battery of Attention), executive function (Achievement Testing System), and verbal fluency (Regensburg Verbal Fluency Test) at baseline and follow-up. The exercise training sessions took place 2-3 times per week for 15-45 minutes

ip t

per session over 8-10 weeks; the exercise intensities per session were prescribed based on

demonstrated during the baseline incremental exercise test.

cr

percentage of anaerobic threshold (i.e., between 120-130% of anaerobic threshold)

us

Persons who underwent cycle ergometer training, but not rowing training, arm

an

ergometer training, or wait-list control conditions demonstrated significant improvements in cardiorespiratory fitness. Those who underwent cycle ergometer training further

M

demonstrated concomitant improvements in verbal memory and alertness, but not in cognitive processing speed, executive function, or verbal fluency compared with the wait-list control

d

group. Those changes in cardiorespiratory fitness explained ~16% of variance in changes in

te

verbal memory and alertness within the cycle ergometry condition [33].

Ac ce p

The results from the recent exercise training RCT are promising, as that study reported improvements in physical fitness following the cycle ergometer training, along with significant improvements in verbal learning and memory in persons with progressive MS and moderate disability. This further has been replicated in a case study involving two memory-impaired persons with MS [34]. However, those results still were not in-line with those from crosssectional studies of fitness and cognition in MS, such that there were not improvements in cognitive processing speed following either exercise training intervention. This might be attributable to disability status, as the sample from the recent RCT had substantially worse disability compared with the cross-sectional samples. Indeed, there have been two published

12 Page 12 of 39

studies on free-living physical activity behavior (i.e., behavior reflecting contraction of the skeletal muscles resulting in increased energy expenditure beyond resting levels) and

moderating factor of the physical activity/cognition relationship [35,36].

ip t

cognitive processing speed in persons with MS that have examined disability as a potential

cr

One prospective study examined physical activity and cognition in 78 persons with mild and moderate MS disability [35]. At baseline, physical activity was measured objectively using

us

accelerometers, and was expressed as steps/day. After a six-month period without

an

intervention, participants underwent neuropsychological assessments of cognitive processing speed, consisting of both oral and written versions of the SDMT, and the PASAT. After

M

controlling for age, sex, and self-reported disability status in the overall sample, steps/day was significantly associated with cognitive processing speed based on oral and written SDMT scores

d

(prs = .25-.29); steps/day was not significantly associated with performance on the PASAT (prs =

te

.12). Importantly, the researchers split the overall sample into two groups of mild and

Ac ce p

moderate disability based on scores from a self-reported EDSS (SR-EDSS) in order to examine the role of disability status as a potential moderator of the association between steps/day and cognitive processing speed. After controlling for age, steps/day was associated with cognitive processing speed (based on oral and written SDMT scores) in persons with mild (SR-EDSS < 4.0; prs = .35-.36), but not moderate (SR-EDSS ≥ 4.0; prs = .24-.27) MS disability [35]. Another study used a RCT design to examine the effects of a six-month, Internet-based behavioral physical activity intervention on cognition in 76 persons with mild and moderate MS disability [36]. All participants completed the International Physical Activity Questionnaire (IPAQ; a self-report measure of physical activity that is valid in persons with MS) and underwent

13 Page 13 of 39

the oral version of the SDMT at baseline and follow-up. Participants were randomized to either the physical activity intervention or wait-list control condition. The intervention condition involved visiting a study website that was based on social cognitive theory; wearing a

ip t

pedometer, completing a log-book, and using computerized software for guiding goal-setting

cr

and attainment; and participating in one-on-one video chat sessions with a behavior-change coach.

us

Following the six-month period, persons with mild disability (PDDS = 0-2) in the

an

intervention condition reported large increases in physical activity (i.e., d = 1.63), whereas those with moderate disability (PDDS = 3-6) in the intervention condition reported a small

M

increase in physical activity relative to baseline (d = 0.24). The most important result from this study was that persons who underwent the intervention condition with mild MS disability

d

demonstrated a clinically meaningful improvement in SDMT scores (~6 point increase), whereas

te

those with moderate MS disability who underwent the intervention condition demonstrated

Ac ce p

minimal change (~1 point decrease) in SDMT scores. There were no changes in physical activity or cognitive processing speed over the six-month period for those who were randomized to the wait-list control condition. Further, among persons in the intervention condition, changes in average steps/day from the pedometer (i.e., objectively-measured physical activity) were associated with changes in SDMT scores in persons with mild (ρ = .52), but not moderate (ρ = .09) MS disability [36]. Following the publication of the poorly-designed early RCTs [25,28], there seems to be an emerging pattern of results from cross-sectional studies of fitness, a recent RCT and case study involving aerobic exercise training, and studies of physical activity and cognitive function.

14 Page 14 of 39

That pattern of results suggests that different domains of exercise training and fitness might be associated with different domains of cognitive function. Those associations further might depend on disability status. To clarify that previous research, a cross-sectional study was

ip t

subsequently completed that directly examined multiple domains of physical fitness (i.e.,

cr

cardiorespiratory fitness and muscular strength) and multiple domains of cognition (i.e.,

cognitive processing speed and learning and memory) in a relatively large sample of persons

us

with mild, moderate, and severe MS disability (i.e., N=62) [37]. Briefly, cardiorespiratory fitness

an

was measured as VO2peak from an incremental exercise test to exhaustion on a recumbent stepper, and lower limb muscle strength was expressed as peak torque of knee extensors and

M

flexors, as well as asymmetry scores based on a maximal isometric muscle strength protocol on an isokinetic dynamometer. Participants also underwent the Brief International Cognitive

d

Assessment in Multiple Sclerosis neuropsychological battery, which consists of the oral version

te

of the SDMT as a measure of cognitive processing speed, the California Verbal Learning Test-2

Ac ce p

(CVLT-2) as a measure of verbal learning and memory, and the Brief Visuospatial Memory TestRevised (BVMT-R) as a measure of visual learning and memory. In the overall sample, measures of cardiorespiratory fitness and lower-limb muscle strength were associated with cognitive processing speed (r = .35-.41), but not learning and memory (r < .19). Similar to previous examinations of physical activity and cognition [35,36], disability status moderated those associations such that cardiorespiratory fitness and muscle strength were associated with cognitive processing speed in persons with mild (EDSS 0-3.5; r = .39-.53), but not moderate (EDSS 4.0-5.5) or severe (EDSS 6.0-6.5) MS disability (r < .21). Further, a post-hoc stepwise linear regression analysis indicated that cardiorespiratory fitness,

15 Page 15 of 39

but not muscular strength outcomes, independently explained a statistically significant portion of variance (i.e., 17%) in SDMT scores. Results from this study provided direct, preliminary evidence of a selective association of physical fitness and cognitive function domains in persons

ip t

with MS. Namely, cardiorespiratory fitness, and to a lesser extent lower limb muscle strength, is

cr

associated with the specific domain of cognitive processing speed, particularly among those with mild disability. There further is preliminary evidence that body composition (i.e., another

us

component of physical fitness) might not affect cognitive processing speed or learning and

an

memory in persons with MS across the disability spectrum [38]. Collectively, this supports the design and implementation of an aerobic exercise training intervention, in particular, for

M

improving cognitive processing speed, but not learning and memory, in persons with mild MS disability.

d

Interestingly, there is emerging evidence that suggests a link between fitness and speed-

te

related aspects of executive functioning in MS. For example, one longitudinal study involving a

Ac ce p

12-week telephone-based health promotion program reported that improvements in physical fitness (based on several fitness domains) were associated with improvements in executive function based on Trail-Making Test performance (d = 0.58) in 88 persons with MS (disability status not reported) [39]. Another study reported that better cardiorespiratory fitness is significantly and moderately associated with better executive control (i.e., shorter reaction time on a modified Eriksen flanker task) (ρ > 0.48) in 28 persons with mild MS disability [40]. Results from that study further suggested that improving cardiorespiratory fitness might even have the potential for persons with mild MS to improve performance on the modified flanker task to the level of a healthy control [40]. To that end, an optimal aerobic exercise training intervention

16 Page 16 of 39

might aim to improve aspects of executive functioning, in addition to cognitive processing speed, in persons with mild MS. Overall, the research examining exercise training, fitness, physical activity and cognition

ip t

in MS is clearly in its infancy. The disappointing results from early training studies, along with

cr

the growing pattern of results from recent studies of exercise, fitness, physical activity, and cognition in MS highlight the importance of again considering the well-established literature in

us

the general population across the lifespan [41] for informing better RCTs of exercise training on

an

cognition in MS. That body of literature documents robust, beneficial effects of exercise training on executive functioning and the neural substrates serving those executive functions

M

across the lifespan (i.e., in healthy children, young adults and older adults) [42]. The delineation of exercise stimuli for such successful interventions is, in part, based on examinations of the

d

effects of acute bouts of aerobic exercise on cognitive function [43-45].

te

3. The Importance of Acute Exercise Effects on Cognition

Ac ce p

There is a growing body of literature examining the effects of single (i.e., acute) bouts of exercise on cognitive functioning in the general population across the lifespan [43-45]. This paradigm is important for identifying the ideal modality and intensity of an exercise stimulus for improving cognitive function for inclusion in a training study. This is because presumably, the acute effects of exercise on cognitive functioning will be additive during chronic exercise training, much like the cumulative benefits of single bouts of aerobic exercise on cardiorespiratory fitness [46]. Further, the examination of acute exercise can provide critical information about the biobehavioral mechanisms as to how exercise improves cognition [47]. This section first highlights reviews and meta-analyses that have summarized the beneficial

17 Page 17 of 39

effects of acute bouts of exercise on cognition in the general population across the lifespan. Based on such evidence, a recent pilot study of acute exercise on cognition in persons with MS is then described; that study was completed to inform the development of an optimal

ip t

exercise stimulus for improving cognition in this population. Future directions regarding the

3.1 Acute Exercise Effects on Cognition in the General Population

cr

examination of acute exercise effects on cognition in MS are then proposed.

us

To date, the evidence describing the effects of acute exercise on cognitive function has

an

been summarized in an initial comprehensive review [43], and the overall effects have been quantified in two subsequent meta-analyses [44,45]. The initial review grouped studies

M

involving the acute effects of exercise on cognition into three different categories: (a) studies examining cognitive performance following maximal exercise testing; (b) studies examining

d

exercise-induced arousal on cognitive performance (i.e., studies that used the Yerkes-Dodson

te

Law (inverted-U hypothesis) as a framework); and (c) studies examining the effects of relatively

Ac ce p

long, submaximal aerobic bouts of exercise on cognition [43]. Based on the experimental design of the existing published acute exercise research on cognition in MS, this section only discusses the results from that third grouping. Briefly, a majority of the reviewed studies, mostly involving healthy college students as participants, did report significant improvements in cognition following acute bouts of aerobic exercise (i.e., ~73%) [43]. However, a small minority of the studies included in this classification did not report positive associations between acute aerobic exercise and cognitive function (i.e., ~13%) [43]. Nevertheless, that review summarized that exercise, when performed as treadmill exercise or cycle ergometry, exerted a selective, beneficial effect on cognition, when performed at a moderate intensity for fewer than 90

18 Page 18 of 39

minutes. The largest effects of acute aerobic exercise were for complex problem solving and executive functioning, in particular, with additional benefits on tasks involving speed of decision-making; that review then described that acute aerobic exercise did not seem to affect

ip t

perceptual tasks (i.e., stimulus identification).

cr

The overall effects of acute exercise on cognition have been quantitatively summarized by two meta-analyses [44,45]. The first meta-analysis examined 40 total studies yielding 235 total

us

effects involving the acute effects of aerobic exercise on cognitive functioning in healthy adults

an

[44]. That meta-analysis separately examined the effects of aerobic exercise on cognition both during and following exercise. Each study that was included in the meta-analysis involved

M

repeated-measures, within-subjects designs.

Overall, when cognitive performance was measured during acute aerobic exercise, there

d

were small, negative effects on cognition (i.e., ∆ = −0.14); those effects were stronger (i.e.,

te

more negative) for processing speed tasks. This suggests that processing speed decreases

Ac ce p

during acute aerobic exercise in healthy adults. One moderator of the acute exercise/cognition association was the time at which cognitive performance was assessed during exercise. There were mostly negative effects on cognition when measured during the first 20 minutes of aerobic exercise, whereas there were mostly positive effects on cognition when measured after at least 20 minutes of aerobic exercise. This suggests that cognitive performance seems to worsen during the first twenty minutes of an aerobic exercise bout, but is seemingly enhanced after this point (i.e., rebound effect). An additional moderator included the modality of aerobic exercise, with more positive effects occurring for cycle ergometry and more negative effects for treadmill exercise. This meta-analysis did not examine the negative effects of treadmill exercise

19 Page 19 of 39

on cognitive function based on walking or running exercise. Of note, since the publication of that meta-analysis, there is evidence that suggests that cognitive performance might not change during acute aerobic exercise in preadolescent children [48].

ip t

On the other hand, when cognition was measured following acute bouts of aerobic

cr

exercise, there were overall positive effects on cognitive functioning that were small in

magnitude (i.e., ∆ = 0.20). One moderator of the acute exercise/cognition association (when

us

cognition was measured after exercise) was the cognitive domain that was examined. When

an

measuring cognitive performance following acute aerobic exercise, there were larger effects on learning and memory and smaller effects on processing speed tasks. In general, the effects of

M

acute exercise on cognition were larger following cycle ergometry, and smaller following treadmill exercise.

d

One limitation of that meta-analysis was the overall small number of studies and effects that

te

were included, based on the overall lack of evidence for acute aerobic exercise on cognition at

Ac ce p

the time of publication [45]. To that end, a more recent meta-analysis provided an updated summary of the effects of acute exercise (including both aerobic and resistance exercise) on cognition in the general population, across the lifespan. That meta-analysis [36] examined 79 studies that yielded 1034 effects—a volume of literature that was substantially larger than the number of studies and effects included in the previous meta-analysis [44]. An important advantage of the more recent meta-analysis is that it included both healthy persons and persons with physical and cognitive limitations. Consistent with the previous meta-analysis, the more recent meta-analysis assessed the overall effects of acute exercise on cognition separately based on the timing of the measurement of cognitive performance. The authors

20 Page 20 of 39

quantified the effects of exercise on cognition when measured (a) during exercise; (b) immediately following exercise (i.e., within 1 minute of exercise completion); and (c) after a delay period of at least 1 minute following exercise.

ip t

Overall, there were small, positive effects of acute exercise on cognition (i.e., g = 0.10)

cr

regardless of the timing of cognitive measurement. When cognition was measured during

exercise, there were small, positive effects of acute exercise on cognitive functioning (g = 0.10).

us

Importantly, that meta-analysis identified the largest effects of acute exercise on tasks of

an

executive function during exercise (d = 0.26), as well as larger effects when cognition was measured after at least 20 minutes of exercise. This suggests that during an acute bout of

M

exercise, executive function is seemingly enhanced, in particular, after at least 20 minutes of exercise. This is not fully consistent with results from the previous meta-analysis [44] describing

d

largely negative effects on cognitive performance during acute exercise.

te

When cognition was measured immediately after exercise, there too were small,

Ac ce p

positive effects on cognitive functioning (g = 0.11), with the largest effects on domains of attention (d = 0.42), crystallized intelligence (d = 0.27), and executive function (d = 0.19). This suggests that immediately following acute exercise, those cognitive functions are seemingly enhanced. Importantly, intensity of exercise was identified as a significant moderator when cognition was measured immediately following exercise, such that there were larger effects for light (d = 0.17) and moderate (d = 0.12) intensity exercise, but there were no effects identified for hard, very hard, or maximal intensity exercise. This suggests that cognitive performance is enhanced immediately following light and moderate intensity exercise, in particular.

21 Page 21 of 39

When cognition was measured after a delay following exercise, there were small, positive effects on cognitive functioning (g = 0.10), with the largest effects on crystallized intelligence (d = 0.28) and executive functioning tasks (d = 0.17). This suggests that acute

ip t

exercise seemingly benefits crystallized intelligence and executive functions after a delay period

cr

following an exercise bout. Exercise intensity was identified as a moderator when cognition was measured after a delay period, such that there were positive effects on cognition when exercise

us

was performed at a light, moderate, hard, very hard, or maximal intensity (d > 0.20).

an

Interestingly, there was an apparent dose-response effect, such that the largest effects of acute exercise on cognition following a delay period were for the most intense exercise stimuli,

M

whereas the smallest effects were for exercise stimuli that were performed at a very light intensity.

d

That meta-analysis also examined the overall effects of acute exercise on cognition across

te

different age groups. Acute exercise was significantly associated with cognition in children (d =

Ac ce p

.17), younger adults (d = .07), middle-aged adults (d = .18), and older adults (d = .18), independent of the timing of cognitive assessment. Importantly, when examining cognitive domain as a moderator in the overall sample, there were significant effects of acute exercise on executive function tasks including Stroop color-word interference (d = 0.25), digit-symbol substitution (d = 0.17), incompatible reaction time (d = 0.29), decision-making (d = 0.30), and verbal fluency (d = 0.31). There further were large effects identified for visual short-term memory (d = 0.23) and free-recall (d = 0.49) [45]. Individual studies have demonstrated that complex neurocognitive tests of executive control (e.g., modified flanker tasks) are particularly sensitive to acute aerobic exercise, as highlighted

22 Page 22 of 39

in a comprehensive review paper [41]. Neither meta-analysis on the overall effects of acute exercise on cognition included those tests as a moderator, when examining the effects of acute aerobic exercise on executive functioning [44,45]. However, another recent, but smaller meta-

ip t

analysis that evaluated the effects of acute exercise on executive function task performance did

cr

examine task complexity as a moderator variable [49]. For example, the authors hypothesized that increasing task complexity (i.e., complex central executive tasks, including inhibitory

us

control based on modified flanker task performance) would be more sensitive to the acute

an

effects of aerobic exercise than simpler recall/basic attention tasks. The results from that metaanalysis indicated that indeed, there were larger effects on complex central executive tasks (g =

M

0.77) versus simpler recall/attention tests (g = 0.31). Further, on complex central executive tasks, the largest effects of acute aerobic exercise seemed to be for measures of speed, rather

d

than measures of accuracy in studies involving adults [49].

te

3.2 Acute Exercise Effects on Cognition in MS

Ac ce p

To date, there is only one published study on the acute effects of exercise on cognitive functioning in persons with MS. That pilot study examined the acute effects of different modalities of exercise on executive control relative to a quiet rest control condition in 24 persons with mild MS disability (median EDSS = 3.0) using a completely within-subjects, repeated measures design [50] (Table 1). Executive control was measured using a modified flanker task, as has been done in the general population [49]. All participants completed 5 testing sessions that were each separated by 7 days. Participants initially completed a baseline session that consisted of repeated cognitive testing to minimize potential practice effects during sessions 2-5. Sessions 2-5 consisted of 20 minutes of quiet rest, moderate intensity

23 Page 23 of 39

treadmill walking exercise, moderate intensity cycle ergometer exercise, and guided yoga that was adapted for persons with MS. The order of exercise and rest sessions was randomized and counterbalanced across participants such that the sample size of 24 allowed for one complete

ip t

replication of the exercise conditions to minimize potential order effects of exercise conditions

cr

on executive control. To examine the acute effects of each condition on executive control, the modified flanker task was administered immediately before and after each condition.

us

There were general pre-to-post improvements in reaction time (regardless of

an

congruency), but not accuracy, on the modified flanker task for all three exercise modalities compared with quiet rest. Those effects were large in magnitude (all ηp2 > .24). Of note, there

M

were additional, selective improvements in reaction time based on congruency (i.e., pre-to-post reductions in the cost of interfering stimuli; interference control) on the modified flanker task

d

for treadmill walking (ηp2 = .17), but not cycle ergometry (ηp2 < .01) or guided yoga (ηp2 < .03),

te

compared with quiet rest. The primary results from that pilot study seemingly support treadmill

Ac ce p

walking as the ideal modality of acute exercise for selectively improving speed-related aspects of executive control (i.e., interference control of reaction time) in persons with MS [50]. This finding mirrors the meta-analytic evidence in the general population [49], and provides critical information regarding the optimal exercise stimulus (i.e., treadmill walking) for possibly improving cognition over time in MS. 3.3 Future Directions Regarding Acute Exercise Effects on Cognition in MS Collectively, the existing literature on exercise and cognition in MS is still clearly in its infancy. To date, cross-sectional investigations of fitness and cognition seemingly favor the development of an aerobic exercise training intervention for improving cognitive processing

24 Page 24 of 39

speed and perhaps aspects of executive functioning in persons with mild MS disability. There further is preliminary evidence that treadmill walking might be the exercise modality that exerts the greatest beneficial effects on speed-related aspects of executive control in this

ip t

population, based on the examination of acute exercise [50]. Indeed, this is consistent with the

cr

acute exercise literature in the general population that suggests that after a single bout of

aerobic exercise, speed-related aspects of executive control might be improved. This too is in-

us

line with the exercise training literature in the general population that describes robust,

an

beneficial effects of chronic aerobic exercise on executive function [42]. The seemingly beneficial effects of acute exercise on cognition in MS complement the cross-sectional

M

associations of fitness and cognition; this advances the field closer towards the development of an optimal exercise training intervention on cognitive processing speed and executive

d

functioning in this population. However, these data are preliminary; there is much work to be

te

done for fully-optimizing such an exercise training intervention for improving cognition in

Ac ce p

persons with MS. A possible starting point involves continued examination of the acute effects of exercise on cognition.

The first important issue is that the optimal intensity of treadmill walking exercise for selectively improving executive control in persons with mild MS is unknown. This would involve examining the dose-dependent effects of acute bouts of varying intensities of treadmill walking exercise on executive control. Such an investigation would represent the next important step in delineating the optimal exercise stimulus, or stimuli, for improving this cognitive function in persons with MS, given that modality and intensity are two key elements of any exercise

25 Page 25 of 39

prescription [26]. This is necessary prior to designing any RCT of exercise training on cognitive functioning in MS. Further, the potential mechanisms for how treadmill walking exercise improves

ip t

cognition in MS have not been directly tested. Indeed, MS researchers have speculated that

cr

treadmill walking exercise is essentially an inhibitory control task, and requires more

attentional resources than other modalities of exercise (e.g., stationary cycling) [50] consistent

us

with data in the general population [44,51]. An extension of such speculation is that with

an

chronic treadmill walking exercise, the acute reductions in the cost of interfering stimuli on reaction time would be additive and cumulative over time (i.e., compensation). To that end,

M

future research should consider examining cognitive performance during acute bouts of aerobic exercise in persons with MS to test the hypothesis of whether treadmill walking exercise

d

actually requires more attentional resources than other exercise modalities, and whether or not

te

this varies based on cardiorespiratory fitness. This approach has been undertaken in the

Ac ce p

general population [45]. One study involving healthy adults described that an acute bout of aerobic exercise might modulate multiple stages of information processing (i.e., from earlystage sensory processing to target categorization) based on event-related potential indices (e.g., P300) [52]. The authors suggested that perhaps a single bout of aerobic exercise can result in the improved ability (i.e., indexed by decreased P300 latency) to classify distractor stimuli in the visual domain [52]. By extension, this mechanism might explain the acute reductions in the cost of interfering stimuli on reaction time for treadmill walking exercise relative to quiet rest that have been previously described in persons with MS [50]. Another potential mechanism for how exercise improves cognition involves increased cerebral blood

26 Page 26 of 39

perfusion. Perhaps a single bout of aerobic exercise increases cerebral blood perfusion in brain regions that are important for facilitating certain cognitive processes. Those increases further might accumulate over time with chronic aerobic exercise training, resulting in meaningful

ip t

improvements in cognition. Indeed, there is preliminary evidence in the general population that

cr

an acute bout of aerobic exercise increases white-matter cerebral blood perfusion [53]. There are no such data in MS. As such, future research endeavors should consider examining the

us

acute effects of exercise on cerebral blood perfusion as a mechanism for improved cognitive

an

performance in persons with MS. Alternatively, it is plausible that other neurobiological mechanisms might explain the beneficial effects of acute exercise on cognition in MS. For

M

example, perhaps an acute bout of treadmill walking exercise results in increases in dopamine transport within neural networks that are important for executive functioning

d

(e.g., fronto-striatal network; central executive network), as has been postulated in the

te

general population [54]. Indeed, future acute and chronic exercise studies in MS might

Ac ce p

consider including neuroimaging measures of resting-state functional connectivity in such networks, as has been done in the cognitive rehabilitation literature in MS [15]. Future research endeavors might also consider evaluating possible genetic influences (e.g., polymorphisms on genes associated with dopamine transport [54]) on the effects of both acute and chronic exercise on cognition in this population. Importantly, in the general population, a majority of acute exercise studies on cognition examine mechanisms of changes in cognitive performance related to task variables and/or individual characteristics [47]. Using this literature as a framework, acute exercise studies in MS can address several important questions related to task variables for planning a highly-

27 Page 27 of 39

developed exercise training intervention for improving cognition. This includes examining whether the acute effects of exercise generalize to other aspects of executive functioning (i.e., response inhibition) beyond interference control, as has been reported in the general

ip t

population [55]. Further, it is unknown whether a single bout of continuous aerobic exercise or

cr

high-intensity interval exercise is more beneficial for cognitive functioning in MS. Other

researchers have posed that perhaps aerobic exercise plus cognitive training/rehabilitation

us

might have a synergistic effect on cognitive functioning in MS compared with aerobic exercise

an

or cognitive rehabilitation alone [18]. An examination of the acute effects of those behaviors on cognition might provide a basis for addressing that hypothesis. Regarding individual

M

characteristics, an important issue is that it is unknown whether the acute effects of exercise have beneficial effects on cognition in fully-ambulatory, but cognitively-impaired persons with

d

MS, as those are the individuals who have the greatest need for intervention. This too might

te

depend on the domain of cognition that is impaired. For example, perhaps acute aerobic

Ac ce p

exercise benefits memory, but not executive control, in memory-impaired persons with MS, analogous to the results of a recent case study involving exercise training effects on cognition in MS [34].

Finally, the examination of acute exercise effects on cognition has implications for persons with progressive MS with more severe ambulatory disability. This is important given that ambulatory impairment is a hallmark feature of MS [56], and is a major barrier to engaging in exercise behavior [57]. Although the majority of data in MS seemingly support an exercise intervention for improving cognition in persons with mild disability (i.e., fullyambulatory persons with MS), there is preliminary evidence that chronic aerobic exercise can

28 Page 28 of 39

improve verbal memory and alertness in persons with progressive MS [33]. Those results are particularly exciting given the equivocal effects of disease modifying therapies on reducing brain atrophy in persons with substantial MS ambulatory disability [58]. To that end, there

ip t

currently are no published acute exercise studies examining cognition in this cohort of persons

cr

with MS. Studies of acute exercise effects on cognition in these persons can provide critical information regarding additional features of an optimal exercise training intervention for

us

improving cognition in persons with moderate-to-severe ambulatory impairment. For

an

example, perhaps there are adaptive modalities of aerobic exercise (i.e., recumbent stepping [59]) that might result in acute improvements in cognition in those with substantial ambulatory

M

disability. 4. Conclusions

d

Exercise training represents a promising behavioral approach for managing cognitive

te

dysfunction in persons with MS. To overcome possible methodological limitations of early RCTs,

Ac ce p

there has been interest in examining the effects of physical activity, fitness, and acute exercise on cognition for better delineating an optimal exercise training intervention for improving cognition over time in this population. This line of research has begun to not only identify features of such an optimal exercise training intervention, but also elucidate potential mechanisms of how exercise can improve cognition in this population. Future work should consider the potential impact of acute exercise research for better understanding chronic exercise training as an approach for improving cognition in those with MS who need it most. This is of particular importance, given that cognitive impairment is common, debilitating, and poorly-managed in persons with MS.

29 Page 29 of 39

5. Acknowledgements This work was not funded by an external grant. The author reports no conflicts of

ip t

interest and thanks Robert W. Motl for providing comments on an earlier draft of this

an

References

us

cr

manuscript.

M

1. Trapp, B.D. & Nave, K. (2008). Multiple sclerosis: An immune or neurodegenerative disorder? Annual Review of Neuroscience, 31, 247-269. 2. Chiaravalloti, N.D. & DeLuca, J. (2008). Cognitive impairment in multiple sclerosis. The Lancet Neurology, 7, 1139-1151.

te

d

3. Benedict, R.H.B. & Zivadinov, R. (2011). Risk factors for and management of cognitive dysfunction in multiple sclerosis. Nature Reviews Neuroscience, 7(6), 332-342.

Ac ce p

4. Prakash, R.S., Snook, E.M., Lewis, J.L., Motl, R.W., & Kramer, A.F. (2008). Cognitive impairments in relapsing-remitting multiple sclerosis: a meta-analysis. Multiple Sclerosis Journal, 14, 1250-1261. 5. Amato, M.P., Zipoli, V., & Portaccio, E. (2006). Multiple sclerosis-related cognitive changes: a review of cross-sectional and longitudinal studies. Journal of the Neurological Sciences, 245, 41-46. 6. Arnett, P.A., Barwick, F.H., & Beeney, J.E. (2008). Depression in multiple sclerosis: review and theoretical proposal. Journal of the International Neuropsychological Society, 14(5): 691-724. 7. Morrow, S.A., Drake, A., Zivadinov, R., Munschauer, F., Weinstock-Guttman, B., & Benedict, R.H.B. (2010). Predicting loss of employment over three years in multiple sclerosis: clinically meaningful cognitive decline. The Clinical Neuropsychologist, 24, 1131-1145. 8. Benedict, R.H.B., Wahlig, E., Bakshi, R., Fishman, I., Munschauer, F., Zivadinov, R., & Weinstock-Guttman, B. (2005). Predicting quality of life in multiple sclerosis: accounting 30 Page 30 of 39

for physical disability, fatigue, cognition, mood disorder, personality, and behavior change. Journal of the Neurological Sciences, 231(1-2), 29-34. 9. Schultheis, M.T., Garay, E., & DeLuca, J. (2001). The influence of cognitive impairment on driving performance in multiple sclerosis. Neurology, 56(8), 1089-1094.

cr

ip t

10. Amato, M.P., Ponziani, G., Siracusa, G., & Sorbi, S. (2001). Cognitive dysfunction in earlyonset multiple sclerosis: a reappraisal after 10 years. Archives of Neurology, 58, 16021606.

us

11. Amato, M.P., Langdon, D., Montalban, X., Benedict, R.H.B., DeLuca, J., Krupp, L.B., Thompson, A.J., et al. (2013). Treatment of cognitive impairment in multiple sclerosis: position paper. Journal of Neurology, 260, 1452-1468.

an

12. Rosti-Otajarvi, E.M. & Hamalainen, P.I. (2014). Neuropsychological rehabilitation for multiple sclerosis. Cochrane Database Systematic Reviews, 2, CD009131.

M

13. Chiaravalloti, N.D., Genova, H.M., & DeLuca, J. (2015). Cognitive rehabilitation in multiple sclerosis: the role of plasticity. Frontiers in Neurology, 6(67), 1-10.

d

14. Chiaravalloti, N.D., Moore, N.B., Nikelshpur, O.M., & DeLuca, J. (2013). An RCT to treat learning impairment in multiple sclerosis: the MEMREHAB trial. Neurology, 81, 20662072.

Ac ce p

te

15. Bonavita, S., Sacco, R., Della Corte, M., Esposito, S., Sparaco, M., d’Ambrosio, A., et al. (2015). Computer-aided cognitive rehabilitation improves cognitive performances and induces brain functional connectivity changes in relapsing remitting multiple sclerosis patients: an exploratory study. Journal of Neurology, 262(1), 91-100. 16. Chiaravalloti, N.D., Wylie, G., Leavitt, V., & DeLuca, J. (2012). Increased cerebral activation after behavioral treatment for memory deficits in MS. Journal of Neurology, 259, 1337-1346. 17. Dobryakova, E., Wylie, G.R., DeLuca, J., & Chiaravalloti, N.D. (2014). A pilot study examining functional brain activity 6 months after memory retraining in MS: the MEMREHAB trial. Brain Imaging & Behavior, 8(3), 403-406. 18. Motl, R.W., Sandroff, B.M., & DeLuca, J. (in press). Exercise training and cognitive rehabilitation: a symbiotic approach for rehabilitating walking and cognitive functions in multiple sclerosis? Neurorehabilitation and Neural Repair. 19. Mattioli, F., Stampatori, C., Bellomi, F., Danni, M., Compagnucci, L., Uccelli, A., et al. (2015). A RCT comparing specific intensive cognitive training to aspecific psychological intervention in RRMS: the SMICT study. Frontiers in Neurology, 5(278), 1-8. 31 Page 31 of 39

20. Cerasa, A., Gioia, M.C., Valentino, P., Nistico, R., Chiriaco, C., Pirritano, D., et al. (2013). Computer-assisted cognitive rehabilitation of attention deficits for multiple sclerosis: a randomized trial with fMRI correlates. Neurorehabilitation and Neural Repair, 27(4), 284-295.

cr

ip t

21. Parisi, L., Rocca, M.A., Mattioli, F., Copetti, M., Capra, R., Valsasina, P., et al. (2014). Changes of brain resting state functional connectivity predict the persistence of cognitive rehabilitation effects in patients with multiple sclerosis. Multiple Sclerosis Journal, 20(6), 686-694.

us

22. Motl, R.W., Sandroff, B.M., & Benedict, R.H.B. (2011). Cognitive dysfunction and multiple sclerosis: developing a rationale for considering the efficacy of exercise training. Multiple Sclerosis Journal, 17(9), 1034-1040.

M

an

23. Bouchard, C., Shephard, R.J., & Stephens, T. (1994). Physical activity, fitness, and health: international proceedings and consensus statement. Champaign, IL: Human Kinetics Publishers. 24. Colcombe, S. & Kramer, A.F. (2003). Fitness effects on the cognitive function of older adults: a meta-analytic study. Psychological Science, 14(2), 125-130.

te

d

25. Oken, B.S., Kishiyama, S., Zajdel, D., Bourdette, D., Carlsen, J., Haas, M., Hugos, C., et al. (2004). Randomized controlled trial of yoga and exercise in multiple sclerosis. Neurology, 62, 2058-2064.

Ac ce p

26. American College of Sports Medicine. (2013). ACSM’s resource manual for guidelines for exercise testing and prescription, seventh edition. Philadelphia, PA: Lippincott Williams & Wilkins. 27. Motl, R.W., Pilutti, L.A., & Sandroff, B.M. (2013). The importance of physical fitness in multiple sclerosis. Journal of Novel Physiotherapies, 3(2), 141-147. 28. Romberg, A., Virtanen, A., & Ruutiainen, J. (2005). Long-term exercise improves functional impairment but not quality of life in multiple sclerosis. Journal of Neurology, 252, 839-845. 29. Prakash, R.S., Snook, E.M., Erickson, K.I., Colcombe, S.J., Voss, M.W., Motl, R.W., & Kramer, A.F. (2007). Cardiorespiratory fitness: A predictor of cortical plasticity in multiple sclerosis. NeuroImage, 34, 1238-1244. 30. Colcombe, S., Kramer, A.F., Erickson, K.I., Scalf, P., McAuley, E., Cohen, N.J., Webb, A., et al. (2004). Cardiovascular fitness, cortical plasticity, and aging. Proceedings from the National Academy of Sciences, 101(9), 3316-3321. 32 Page 32 of 39

31. Prakash, R.S., Snook, E.M., Motl, R.W., & Kramer, A.F. (2010). Aerobic fitness is associated with gray matter volume and white matter integrity in multiple sclerosis. Brain Research, 1341, 41-51.

ip t

32. Sandroff, B.M. & Motl, R.W. (2012). Fitness and cognitive processing speed in persons with multiple sclerosis: a cross-sectional investigation. Journal of Clinical and Experimental Neuropsychology, 34(10), 1041-52.

us

cr

33. Briken, S., Gold, S.M., Patra, S., Vettorazzi, E., Harbs, D., Tallner, A., Ketels, G., et al. (2014). Effects of exercise on fitness and cognition in progressive MS: a randomized, controlled pilot trial. Multiple Sclerosis Journal, 20(3), 382-390.

an

34. Leavitt, V.M., Cirnigliaro, C., Cohen, A., Farag, A., Brooks, M., Wecht, J.M., Wylie, G.R., et al. (2014). Aerobic exercise increases hippocampal volume and improves memory in multiple sclerosis: preliminary findings. Neurocase, 20(6), 695-697.

M

35. Sandroff, B.M., Pilutti, L.A., Dlugonski, D., & Motl, R.W. (2013). Physical activity and information processing speed in persons with multiple sclerosis: a prospective study. Mental Health and Physical Activity, 6(3), 205-211.

te

d

36. Sandroff, B.M., Klaren, R.E., Pilutti, L.A., Dlugonski, D., Benedict, R.H.B., & Motl, R.W. (2014). Randomized controlled trial of physical activity, cognition, and walking in multiple sclerosis. Journal of Neurology, 261(2), 363-372.

Ac ce p

37. Sandroff, B.M., Pilutti, L.A., Benedict, R.H.B., & Motl, R.W. (2015). Association between physical fitness and cognitive function in multiple sclerosis: does disability status matter? Neurorehabilitation and Neural Repair, 29(3), 214-223. 38. Sandroff, B.M., Hubbard, E.A., Pilutti, L.A., & Motl, R.W. (2015). No association between body composition and cognition in ambulatory persons with multiple sclerosis: a brief report. Journal of Rehabilitation Research and Development, 53(2), 301-308. 39. Beier, M., Bombardier, C.H., Hartoonian, N., Motl, R.W., & Kraft, G.H. (2014). Improved physical fitness correlates with improved cognition in multiple sclerosis. Archives of Physical Medicine and Rehabilitation, 95(7), 1328-1334. 40. Sandroff, B.M., Hillman, C.H., & Motl, R.W. (2015). Aerobic fitness is associated with inhibitory control in persons with multiple sclerosis. Archives of Clinical Neuropsychology, 30(4), 329-340. 41. Hillman, C.H., Erickson, K.I., & Kramer, A.F. (2008). Be smart, exercise your heart: exercise effects on brain and cognition. Nature Reviews Neuroscience, 9, 58-65.

33 Page 33 of 39

42. Voss, M.W., Nagamatsu, L.S., Liu-Ambrose, T., & Kramer, A.F. (2011). Exercise, brain, and cognition across the life span. Journal of Applied Physiology, 111, 1505-1513. 43. Tomporowski, P.D. (2003). Effects of acute bouts of exercise on cognition. Acta Psychologica, 112, 297-234.

ip t

44. Lambourne K. & Tomporowski, P. (2010). The effect of exercise-induced arousal on cognitive task performance: a meta-regression analysis. Brain Research, 1341, 12-24.

cr

45. Chang, Y.K., Labban, J.D., Gapin, J.I., & Etnier, J.L. (2012). The effects of acute exercise on cognitive performance: a meta-analysis. Brain Research, 1453, 87-101.

an

us

46. Haskell, W.L. (1994). Health consequences of physical activity: understanding and challenges regarding dose-response. Medicine & Science in Sports and Exercise, 26, 649660.

M

47. Pesce, C. (2009). An integrated approach to the effect of acute and chronic exercise on cognition: the linked role of individual and task constraints. In: McMorris, T., Tomporowski, P.D., & Audriffen, M. (Eds.), Exercise and Cognitive Function: John Wiley and Sons, Hoboken, NJ, pp. 213-226.

te

d

48. Drollette, E.S., Shishido, T., Pontifex, M.B., & Hillman, C.H. (2012). Maintenance of cognitive control during and after walking in preadolescent children. Medicine & Science in Sports and Exercise, 44(10), 2017-2024.

Ac ce p

49. McMorris, T. & Hale, B.J. (2012). Differential effects of differing intensities of acute exercise on speed and accuracy of cognition: a meta-analytical investigation. Brain and Cognition, 80, 338-351. 50. Sandroff, B.M., Hillman, C.H., Benedict, R.H.B., & Motl, R.W. (2015). Acute effects of walking, cycling, and yoga exercise on cognition in persons with relapsing-remitting multiple sclerosis without impaired cognitive processing speed. Journal of Clinical and Experimental Neuropsychology, 37(2), 209-219. 51. Voelcker-Rehage, C. & Niemann, C. (2013). Structural and functional brain changes related to different types of physical activity across the lifespan. Neuroscience and Biobehavioral Reviews, 37, 2268-2295. 52. Bullock, T., Cecotti, H., & Giesbrecht, B. (2015). Multiple stages of information processing are modulated during acute bouts of exercise. Neuroscience, 307, 138-150. 53. MacIntosh, B.J., Crane, D.E., Sage, M.D., Saeed Rajab, A., Donahue, M.J., McIlroy, W.E., & Middleton, L.E. (2014). Impact of a single bout of aerobic exercise on regional brain perfusion and activation responses in healthy young adults. PLoS ONE, 9(1), e85163. 34 Page 34 of 39

54. Berse, T., Rolfes, K., Barenberg, J., Dutke, S., Kuhlenbaumer, G., Volker, K., et al. (2015). Acute physical exercise imporves shifting in adolescents at school: evidence for a dopaminergic contribution. Frontiers in Behavioral Neuroscience, 9(96), 1-9.

ip t

55. Kamijo K., Nishihira, Y., Hatta, A., Kaneda, T., Wasaka, T., Kida, T., & Kuroiwa, K. (2004). Differential influences of exercise intensity on information processing in the central nervous system. European Journal of Applied Physiology, 92, 305-311.

us

cr

56. Goldman, M.D., Marrie, R.A., & Cohen, J.A. (2008). Evaluation of the six-minute walk in multiple sclerosis subjects and healthy controls. Multiple Sclerosis Journal, 14, 383390.

an

57. Motl, R.W. (2010). Physical activity and irreversible disability in multiple sclerosis. Exercise & Sports Science Reviews, 38(4), 186-191.

M

58. PRISMS (Prevention of Relapses and Disability by Interferon beta-1a Subcutaneously in Multiple Sclerosis) Study Group. (1998). Randomised double-blind placebocontrolled study of interferon beta-1a in relapsing/remitting multiple sclerosis. Lancet, 352(9139), 1498-1504.

Ac ce p

te

d

59. Gappmaier, E. (2012). The submaximal clinical exercise tolerance test (SXTT) to establish safe exercise prescription parameters for patients with chronic disease and disability. Cardiopulmonary Physical Therapy Journal, 23(2), 19-29.

Table 1 Summary of evidence of exercise, physical fitness, and physical activity effects on cognition in persons with multiple sclerosis Study Category Exercise Stimuli Fitness Cognitive Outcomes ‡

Results

35 Page 35 of 39

(RCT)

Characteristics

Outcomes †

Mode: Iyengar yoga Frequency: 1x/week; 26 weeks Intensity: Minimal Duration: 90 minutes/session

N/A

Mode: Cycle ergometry Frequency: 1x/week; 26 weeks Intensity: Light Duration: 90 minutes/session

Sandroff & Motl, 2012 [32]

PF

cr N/A

LM Test(s): Weschler Memory Scales III Logical Memory CPS Test(s): PASAT

CRF

CPS Test(s): SDMT

↑CRF associa

EF Test(s): WCST LM Test(s): SRT, SPART

↑CRF associa activation dur

te

(Crosssectional)

Ac ce p

Prakash et al., 2010 [31]

us

PF

EF Test(s): Stroop Color-Word

an

Prakash et al., 2007 [29]

(RCT)

Mode: Home-based cycle ergometry/resistance training Frequency: 3-4x/week; 26 weeks Intensity: N/A Duration: N/A N/A

M

ET

N/A

CRF

(Crosssectional)

PF

N/A

(Crosssectional)

↔ on any cog compared wit

CPS Test(s): PASAT

d

Romberg et al., 2005 [28]

ATT Test(s): Covert orienting of spatial attention task, modified Useful Field of View Task

ip t

Oken et al., 2004 [25]

(Design) * ET

↔ on PASAT control group

CRF not asso cognitive dom ↑CRF associa

VF Test(s): WLG CPS Test(s): SDMT, PASAT

↑CRF associa matter volum integrity

LM Test(s): SRT, SPART VF Test(s): WLG CRF

CRF not asso cognitive dom ↑CRF associa

CPS Test(s): SDMT, PASAT

LLMS

↓LLMS asym with ↑CPS

Balance Briken et al., 2014 [33]

ET (RCT)

Mode: Arm ergometry Frequency: 2-3x/week; 8-10 weeks

CRF

↑balance asso Cycle ergome ↑alertness sub of Attention c

ATT Test(s): Test Battery of Attention

36 Page 36 of 39

CPS Test(s): SDMT

Intensity: Moderate Duration: 15-45 minutes/session

EF Test(s): Achievement Testing System

Mode: Rowing Frequency: 2-3x/week; 8-10 weeks Intensity: Moderate Duration: 15-45 minutes/session

ip t

cr us

an

(Case study)

VF Test(s): Regensburg Verbal Fluency Test

CRF

M

ET

LM Test(s): VLMT

CPS Test(s): SDMT, PASAT EF Test(s): Stroop, Digit Span

↑CRF compa participant

↑CVLT-2, BV with control p

LM Test(s): CVLT-2, BVMT-R

↑Hippocampa state function compared wit

Ac ce p

te

d

Leavitt et al., 2014 [34]

Mode: Cycle ergometry Frequency: 2-3x/week; 8-10 weeks Intensity: Moderate Duration: 15-45 minutes/session Mode: Cycle ergometry Frequency: 3x/week; 12 weeks Intensity: Moderate-to-vigorous Duration: 30 minutes/session

control group ↔ on all othe measures com group

Sandroff et al., 2013 [35]

Sandroff et al., 2014 [36]

PA

N/A

N/A

CPS Test(s): SDMT, PASAT

↔ on all othe measures Mild MS disa ↑Accelerome steps/day asso over 26-week

CPS Test(s): SDMT

Moderate MS Steps/day not SDMT Mild MS disa ↑Self-report P measured step with control g

(PrePost)

PA

(RCT)

Mode: Free-living walking Frequency: N/A Intensity: N/A Duration: N/A

N/A

↑SDMT comp group

Moderate MS 37 Page 37 of 39

↔Self-report measured step with control g

N/A

(Crosssectional)

CRF

CPS Test(s): SDMT

LLMS

LM Test(s): CVLT-2, BVMT-R

ip t

PF

cr

Sandroff et al., 2015 [37]

↔SDMT com group Mild MS disa ↑CRF and ↑L with ↑SDMT

PF not associ

(PrePost) PF

N/A

M d

PF

Ac ce p

Beier et al., 2014 [39]

PF N/A (Crosssectional)

te

Sandroff et al., 2015 [38]

an

us

Moderate MS PF not associ cognitive outc

Sandroff et al., 2015 [40]

Sandroff et al., 2015 [50]

N/A

Body fat

CPS Test(s): SDMT

Lean body mass

LM Test(s): CVLT-2, BVMT-R

BMD CRF LLMS

CPS Test(s): PASAT

CRF

Severe MS di PF not associ cognitive outc Body fat, lean not associated outcome

↑PF associate 12-week perio

EF Test(s): TMT-A, TMT-B EF Test(s): Modified flanker task

↑CRF associa modified-flan

EF Test(s): Modified flanker task

High-fit perso similar RT on controls Pre-to-post ↓R rest for all 3 e

(Crosssectional)

AE

(Within subjects, repeated measures )

Mode: Treadmill walking Frequency: N/A Intensity: Moderate Duration: 20 minutes

N/A

Treadmill wa in interferenc relative to qui

Mode: Cycle ergometry Frequency: N/A Intensity: Moderate Duration: 20 minutes

Cycle ergome interference c to quiet rest

Mode: Hatha yoga 38 Page 38 of 39

Yoga: ↔ in i of RT relative

Ac ce p

te

d

M

an

us

cr

ip t

Frequency: N/A Intensity: Light-to-moderate Duration: 20 minutes Note: *AE=acute exercise; ET=exercise training; PA=physical activity; PF=physical fitness; RCT=randomized controlled trial. †CRF=cardiorespiratory fitness; LLMS=lower limb muscular strength; BMD=bone mineral density. ‡ATT=attention; BVMT-R=Brief Visuospatial Memory Test-Revised; CPS=cognitive processing speed; CVLT-2=California Verbal Learning Test-2; EF=executive function; LM=learning and memory; PASAT=Paced Auditory Serial Addition Test; SDMT=Symbol Digit Modalities Test; SPART=Spatial Recall Test; SRT=Selective Reminding Task; TMT=Trail-Making Test; VF=verbal fluency; VLMT=Verbal Learning and Memory Test; WCST=Wisconsin Card Sorting Task; WLG=Word List Generation. ‽ ↑=increased; ↓=decreased; ↔=no significant change.

39 Page 39 of 39