Sports Med

DOI 10.1007/s40279-015-0307-x

SYSTEMATIC REVIEW

Aerobic Capacity in Persons with Multiple Sclerosis: A Systematic

Review and Meta-Analysis
Martin Langeskov-Christensen Martin Heine
Gert Kwakkel Ulrik Dalgas

 Springer International Publishing Switzerland 2015

Abstract
Background Aerobic capacity (VO2max) is a strong health
and performance predictor and is regarded as a key physiological measure in the healthy population and in persons
with multiple sclerosis (PwMS). However, no studies have
tried to synthesize the existing knowledge regarding
VO2max in PwMS.
Objectives The objectives of this study were to (1) systematically review the psychometric properties of the
VO2max test; (2) systematically review the literature on
VO2max compared with healthy populations; (3) summarize
correlates of VO2max; and (4) to review and conduct a metaanalysis of longitudinal exercise studies evaluating training-induced effects on VO2max in PwMS.

M. Langeskov-Christensen and M. Heine share first authorship on this

article. Both authors contributed equally to the paper.
M. Langeskov-Christensen (&)
U. Dalgas
Section of Sport Science, Department of Public Health, Aarhus
University, Dalgas Avenue 4, 8000 Aarhus C, Denmark
e-mail: mach@ph.au.dk
M. Heine
Brain Center Rudolf Magnus and Center of Excellence
for Rehabilitation Medicine, University Medical Center Utrecht
and Rehabilitation Center De Hoogstraat, Utrecht,
The Netherlands
G. Kwakkel
Department of Neurorehabilitation, Centre of Rehabilitation
and Rheumatology Reade, Amsterdam, The Netherlands
G. Kwakkel
Department of Rehabilitation Medicine, MOVE Research
Institute Amsterdam, VU University Medical Center,
Amsterdam, The Netherlands

Data Sources and Study Selection A systematic literature

search of six databases (PubMed, EMBASE, Cochrane Library, PEDro, CINAHL and SPORTDiscus) was performed.
To be included, the study had to (1) enrol participants with
definite MS according to defined criteria; (2) assess aerobic
capacity (VO2max) by means of a graded exercise test to
voluntary exhaustion; (3) had undergone peer review; and (4)
be available in English, Danish or Dutch.
Study Appraisal and Synthesis Methods The psychometric properties of the VO2max test in PwMS were reviewed with respect to reliability, validity and
responsiveness. Simple Pearson correlation analysis was
used to assess the relation between key study characteristics and the reported mean VO2max. The methodological
quality of the intervention studies was evaluated using the
original 11-item Physiotherapy Evidence Database
(PEDro) scale. A random coefficient model was used to
summarize individual, weighted, standardized effects of
studies that assessed the effects of exercise on aerobic
capacity in PwMS.
Results A total of 40 studies, covering 165 healthy controls
and 1,137 PwMS, fulfilled the inclusion criteria. VO2max
testing in PwMS can be considered a valid measure of
aerobic capacity, at least in PwMS having low-to-mild disability, and an *10 % change between two tests performed
on separate days can be considered the smallest reliable
change (with 95 % certainty) in VO2max in PwMS. The average body-weight-adjusted VO2max was significantly lower
in PwMS (25.5 5.2 mL
kg-1
min-1) compared with
healthy controls (30.9 5.4 mL
kg-1
min-1). The analysis
of VO2max correlates revealed associations with a variety of
outcomes covering all levels of the International Classification of Functioning, Disability and Health (ICF) model.
The meta-analysis showed that aerobic training in PwMS
may improve VO2max by as much as 3.5 mL
kg-1
min-1.

123

M. Langeskov-Christensen et al.

Conclusions A valid and reliable test can be performed,

in at least ambulant PwMS, by the gold standard wholebody maximal exercise test. Aerobic capacity in PwMS is
impaired compared with healthy people, and is significantly associated with factors on all levels of the ICF
model, including disease severity. Aerobic training can
improve aerobic capacity in PwMS to a degree that is associated with secondary health benefits.

Key Points
Aerobic capacity in persons with multiple sclerosis
(PwMS) is impaired compared with healthy
individuals.
Aerobic capacity is significantly associated with
factors on all levels of the International Classification
of Functioning, Disability and Health model.
Aerobic training can improve aerobic capacity in
PwMS to a degree that is associated with secondary
health benefits.

1 Introduction
Multiple sclerosis (MS) is a chronic disorder of the central
nervous system, being the most frequent neurological disease causing disability in young and middle-aged people
[1]. The exact etiology of MS remains elusive, but epidemiological studies show that it is a multifactorial disease
that is likely caused by a complex interaction between
multiple genes and environmental factors [2].
The clinical complexity and heterogeneity of MS is reflected in the broad range of symptoms in persons with MS
(PwMS). Moreover, the clinical manifestations of MS include motor, sensory, brainstem, cerebellar, visual,
sphincter and cognitive symptoms, which in combination
reduce quality of life (QoL) [3, 4]. The level of disability in
PwMS is most commonly described by the Expanded
Disability Status Scale (EDSS), ranging from a score of 0
(normal neurological examination) to 10 (death by disease)
[5].
In general, PwMS are known to adopt an inactive lifestyle with subsequent deconditioning and concurrent comorbidities similar to those observed when healthy persons
are inactive [6]. Consequently, PwMS exhibit inactivityrelated health problems such as an increased incidence of
osteoporosis, death from cardiovascular diseases (CVD),
depression and fatigue [710]. Aerobic capacity is an important health [11, 12] and performance [13] marker in

123

both healthy subjects and PwMS. Moreover, aerobic capacity is associated with CVD risk [14], better walking
performance [15], improved cognitive processing speed
[16, 17], and a possible prophylactic influence on the
structural decline of brain tissue in PwMS [18], while an
impaired aerobic capacity in healthy individuals is related
to functional limitations that may hinder independent living
[19]. As such, aerobic capacity is regarded as a key
physiological measure in PwMS.
Direct testing of aerobic capacity, using a graded exercise, whole-body (i.e. cycling) protocol with complementary respiratory gas-exchange measurements, is considered
the gold standard when assessing aerobic capacity [20].
Depending on the limiting factors near exhaustion, the
outcome of the test can be defined as peak oxygen uptake
(VO2peak) or maximal oxygen uptake (VO2max) [21].
Although the term VO2peak simply defines the highest value
of VO2 attained during a test, regardless of the subjects
effort and, consequently, does not necessarily represent the
true VO2max, these definitions are used interchangeably in
the current MS literature [22]. In the present systematic
review, aerobic capacity will be described as VO2max rather
than VO2peak.
Despite the widespread use of VO2max testing, no review could be located that summarized the psychometric
properties [23] of the VO2max test in PwMS. Consequently, the validity, reliability and responsiveness of the
test need to be reviewed in these persons. In addition, it is
frequently stated in the literature that VO2max is impaired
in PwMS, despite no studies summarizing the existing
knowledge. Also, as stated previously, a number of crosssectional studies present associations between VO2max and
clinical correlates [16, 24, 25], but so far no studies have
tried to summarize these, although this could lead to new
relevant target areas for MS rehabilitation. Finally, the
VO2max is known to be responsive to aerobic exercise
interventions in most healthy subjects [26], and previous
reviews on exercise and MS reveals that it can also improve in PwMS following aerobic training [27]. Nevertheless, no previous reviews [2729] on aerobic training
have tried to synthesize the existing knowledge regarding
the effects of aerobic training on VO2max in a metaanalysis.
Consequently, the objectives of the present study
were, in PwMS, to (1) systematically review the psychometric properties of gold-standard testing (VO2max
test) of aerobic capacity; (2) systematically review the
literature on aerobic capacity compared with healthy
populations; (3) summarize correlates of VO2max; and (4)
review and conduct a meta-analysis of longitudinal exercise studies evaluating training-induced effects on
aerobic capacity.

Multiple Sclerosis and Aerobic Capacity

2 Methods
2.1 Study Selection
The present systematic review follows the PRISMA
(Preferred Reporting Items for Systematic Reviews and
Meta-Analyses) guidelines on systematic reviews of randomized controlled trials (RCT) [30]. No predefined review
protocol was published. The study was based on a systematic
literature search of six databases (PubMed, EMBASE,
Cochrane Library, PEDro, CINAHL and SPORTDiscus),
which was performed to identify studies on aerobic capacity
in PwMS published before 1 May 2014. For the exact
Medical Subject Heading (MeSH) search terms used in the
various databases, see Table 1.
To be included in the systematic review the study had to:

enrol participants with definite MS according to defined

criteria [31];
assess aerobic capacity (VO2max or VO2peak) by means
of a graded exercise test to voluntary exhaustion;
had undergone peer review;
be available in English, Danish or Dutch.

Reports of case-studies were excluded.

Two authors (ML-C and MH) independently screened
titles and abstracts, read identified full-text articles, extracted study properties from the included articles and
assessed the methodological quality of the included clinical
trials. In case of disagreement, a third author (UD) was
consulted to reach consensus. Corresponding authors were
contacted in case of missing data.
2.2 Psychometric Properties of Maximal Oxygen
Uptake (VO2max) Testing in Persons with Multiple
Sclerosis (PwMS)
The psychometric properties of VO2max testing in PwMS
were reviewed with respect to reliability, responsiveness,

and validity [23]. Validity comprises content, construct,

and criterion validity [23]. In the present review, content
validity was defined as the degree in which the measurement is an adequate reflection of the construct measured, in
this case aerobic capacity. Since VO2max as a measure of
aerobic capacity is already considered the gold standard
[20], content validity was assessed by looking into the
different classical criteria [32, 33] that should be met
during maximal exercise testing for it to be considered a
true test of aerobic capacity. Various definitions of these
criteria exist [34]. However, in general, the primary criterion is considered a plateau in oxygen uptake despite an
increasing workload [32, 34]. The secondary criteria include a post-test blood lactate level C8.0 mmol
L-1, a
respiratory exchange ratio (RER) (VO2/VCO2) exceeding
1.10 or 1.15, a maximal heart rate (HRmax) within 10 % or
10 beats
min-1 of the estimated HRmax (e.g. 220-age or
208-0.7 9 age), and a perceived exertion C17, 18 or 19
on the Borg scale of perceived exertion (range 620) [34].
Construct validity can be defined as the degree to which the
scores of a measure are consistent with hypotheses (for
instance, with regard to internal relationships, relationships
to scores of other instruments, or differences between
relevant groups). Criterion validity can be defined as the
degree to which the scores of a measure are an adequate
reflection of a gold standard. Both construct and criterion
validity were not addressed in the present review since
whole-body testing is already considered the gold standard
for testing aerobic capacity, and these are therefore not
applicable. However, by looking into the reported correlations between VO2max and other outcomes, the relevance
of VO2max beyond the construct of aerobic capacity was
further explored (see Sect. 2.3). Reliability was defined as
the degree to which a measurement is free of measurement
error, while responsiveness was defined as the ability of a
measure to detect change over time [23]. Little measurement error will improve the ability of a measure to detect
change over time and is often referred to as the smallest
detectable change (SDC).

Table 1 Detailed list of retrieved articles and applied search terms in

six different databases

2.3 Aerobic Capacity in PwMS

Database

Articles
retrieved

Search terms (e.g. MeSH)

PubMed
EMBASE
Cochrane
Library
CINAHL
SPORTDiscus

205
950
65

Multiple sclerosis AND (maximal oxygen

consumption OR maximal oxygen uptake
OR cardiopulmonary exercise testing OR
cardiopulmonary exercise test OR
VO2-max OR VO2max OR VO2-peak
OR VO2peak OR breathing gas analysis
OR aerobic capacity OR maximal
aerobic capacity)
Abstract and title: Multiple sclerosis.
Therapy: Fitness training

Data on study characteristics, testing procedures, correlates

of VO2max, and VO2max were extracted from all included
cross-sectional and longitudinal studies in which sufficient
data were available. Baseline VO2max measurements of the
longitudinal studies were extracted and used as cross-sectional data. Subsequently, descriptive statistics were used
to obtain a mean value and standard deviation (SD) of
participants characteristics (e.g. EDSS, age, sex). When
absolute values for VO2max were provided, these were
corrected for the mean body weight of the sample (i.e.
L
min-1 ? mL
kg-1
min-1). Simple Pearson correlation

PEDro

39
16
67

MeSH Medical Subject Headings

123

M. Langeskov-Christensen et al.

analysis was used to assess the association between key

study characteristics and the reported mean VO2max. Strong
associations were defined as coefficients beyond 0.70,
whereas coefficients between 0.30 and 0.70 were classified
as moderate to substantial, and correlation coefficients \0.30 were classified as weak associations. A
p value B0.05 was considered statistically significant.
2.4 Study Appraisal
The methodological quality of the longitudinal intervention
studies was evaluated using the original 11-item Physiotherapy Evidence Database (PEDro) scale [35], by scores
derived from the PEDro database [36]. If no PEDro score
was present in the PEDro database, two authors (MH and
ML-C) independently scored the respective study and
discussed disagreements to reach a consensus. The first
item of the PEDro scale, regarding the description of the
studys inclusion criteria, is considered related to the external validity of the study rather than the internal validity
(i.e. methodological quality) [35, 36]. Hence, the first item
is omitted from the PEDro score which therefore ranges
from 0 to 10.
2.5 Effects of Aerobic Training on Aerobic Capacity
in PwMS
Studies that used a randomized or non-randomized design
to assess the longitudinal effects of exercise on aerobic
capacity, in comparison to a control group, and which reported or provided sufficient data, were included in a metaanalysis. Non-randomized clinical trials were only used if
there were no reported baseline differences in aerobic capacity between the experimental and control groups. Exercise was defined according to the definition by Caspersen
et al. [37], who stated that exercise is a subset of physical
activity that is planned, structured and repetitive and has as
a final or an intermediate objective the improvement or
maintenance of physical fitness. Exercise can be considered aerobic training if it includes many dynamic muscle
contractions against low loads for extended periods of time,
with sufficient progression, and with a minimum intensity
of 60 % of HRmax or equivalent on the Borg scale, VO2
scale or percentage of heart-rate reserve scale. Meta-analytical procedures complying with the Meta-Analysis of
Observational Studies in Epidemiology (MOOSE) framework [38] were applied to evaluate possible effects of
(aerobic) exercise on aerobic capacity in PwMS. The metaanalysis was conducted using Review Manager, version 5.2
(http://tech.cochrane.org/Revman). When studies reported
more than one exercise intervention in comparison with a
control group, the exercise interventions were merged into
a single exercise group [39]. If inadequate data on VO2max

123

were provided, authors were contacted for additional data

to the meta-analysis, and this was provided in two cases
[40, 41]. The standardized mean difference (SMD) was
computed as the mean change from before to after the
intervention of the exercise group minus the mean change
of the control group divided by the pre-intervention pooled
SD and adjusted for sample size. A positive SMD indicates
a beneficial improvement in aerobic capacity in the exercise group when compared with the control group. The
SMD was computed using a random effects model and is
reported as Hedges g, which adjusts for sample size differences across studies. The underlying assumption of the
random effects model is that samples are drawn from
populations with different effect sizes (ES) and that the true
effect differs between studies. Furthermore, based on the
assumption of random effects, we computed a 95 % confidence interval (CI) around the SMD and tested for
heterogeneity of the studies included in the meta-analysis.
Heterogeneity was present if the I2 statistic was [50 %.
The strength of the SMD was interpreted as small, moderate, or large, referring to 0.2 to \0.5 and C0.5 to \0.8,
and [0.80, respectively [42]. To improve clinical interpretability, the SMD was then re-expressed into VO2max by
multiplying the SMD with the SD of the best powered
observational study [43]. By using the best powered study,
it is assumed that the SD of this study best represents the
typical among-person variation and therefore provides a
more precise estimate of the change in VO2max. The effect
of methodological quality on the SMD was assessed by
performing a sensitivity analysis in which non-randomized
trials and studies with a high risk of bias were excluded.

3 Results
3.1 Study Selection
As presented in Fig. 1, the search yielded 1,342 hits; after
removal of duplicates, 1,211 publications remained. A total
of 30 studies that fulfilled the inclusion criteria were
identified. The reference lists of these 30 publications as
well as the authors personal databases were checked for
further relevant publications that were not captured by the
search. This identified ten additional studies (for details see
Fig. 1). These 40 studies were then categorized to serve
one or more of the different objectives of the present review: (1) related to the psychometric properties of VO2max
testing in PwMS (number of studies, n = 2); (2) crosssectional studies reporting VO2max and baseline VO2max
measurements of longitudinal studies (n = 40); and (3)
longitudinal studies reporting the effects of exercise on
VO2max in PwMS (n = 17). In addition, all included studies
were searched for correlates of VO2max.

Multiple Sclerosis and Aerobic Capacity

Fig. 1 PRISMA flowchart of search results and study selection. The included studies could serve multiple objectives. MS multiple sclerosis, VO2
oxygen uptake

3.2 Psychometric Properties of the VO2max Test

in PwMS
Two studies that specifically studied the validity of VO2max
testing in PwMS were identified [22, 44]. Both studies
concluded that VO2max can be considered a valid measure
of aerobic capacity, at least in PwMS having low-tomoderate disability (EDSS B4.0). Overall, 3740 % of
PwMS were able to reach the primary criterion of a VO2
plateau, while 2395 % were able to reach the secondary
criteria. In PwMS having low-to-moderate disability
(EDSS B6.0), this was not different from healthy controls

[22]. In contrast to Langeskov-Christensen et al. [22],

Heine and colleagues [44] found a significant relation between the ability to attain the criteria for VO2max testing
and the level of disability. This was best illustrated by a
decreased ability, for PwMS in the most disabled study
group (EDSS 4.56.0), to reach an HRmax close to their
estimated HRmax.
The reliability of VO2max testing in PwMS was assessed
in one study [22]. This study revealed an apparently random variation around the zero line for the day-to-day
variability of VO2max (limits of agreement -238 to
201 mL
min-1), and concluded that an *10 % change in

123

M. Langeskov-Christensen et al.

two tests performed on separate days can be considered as

the smallest detectable difference (with 95 % certainty) in
VO2max in PwMS. In addition to the two studies that
specifically studied the psychometric properties of VO2max
testing in PwMS, 17 other studies reported data on various
criteria for VO2max testing in PwMS, but without aiming to
specifically determine the validity or reliability of the test
(Table 2). The average RER value across the studies using
leg ergometry was 1.17 0.09 (n = 6), the average lactate
level was 8.2 1.8 (n = 2), and the average perceived
exertion on the Borg Scale was 17.1 1.8 (n = 7) (see
Table 2 for details).
3.3 Aerobic Capacity in PwMS
Table 3 shows the cross-sectional studies (n = 23) that
included aerobic capacity as an outcome measure. Baseline
measurements of intervention studies that included this
outcome (n = 17) are presented in Table 4. In total, the
studies included 165 healthy controls and 1,029 PwMS
(originally 1,137 but aerobic capacity was not reported for
108 PwMS) with an average mean EDSS of 2.9 1.1
(EDSS were provided for 891 PwMS). Seven studies only
included females. From the studies including both males
and females, the MS gender distribution (684 subjects) was
calculated as (mean SD) 73 % 18 % female. Three
studies applied upper-body testing, whereas the remaining
37 studies applied whole-body testing (leg ergometer cycling/combined armleg ergometer). Figure 2 illustrates
the body-weight-adjusted VO2max in PwMS compared with
healthy control groups (control) when possible. One study
[45] did not report the body-weight-adjusted VO2max or the
mean subject weight, while another study did not report
any VO2max data (despite the fact that the test was performed) [18], and these were consequently not included in
Fig. 2. In eight cases, data from the subject sample were
reported in more than one paper. Nine studies incorporated
control, and eight of these used whole-body testing.
However, one study did not report the control VO2max data
[46], one study did not report any VO2max data [18], and
two papers reported on data from the same subject sample
[15, 16]. In the studies using whole-body testing, the
average body-weight-adjusted VO2max of PwMS was
25.5 5.2 mL
kg-1
min-1 (range 16.9 6.336.7
10.7 mL
kg-1
min-1), whereas the average upper body
VO2max was 10.2 0.6 mL
kg-1
min-1 (range 9.6
4.011 2.0 mL
kg-1
min-1). The average whole-body
VO2max of the control was 30.9 5.4 mL
kg-1
min-1
(range 25 5.038.4 6.8 mL
kg-1
min-1), while the
control reached 14.3 mL
kg-1
min-1 in the study using
upper-body testing. A significant difference in VO2max
between PwMS and controls was found in the study using
upper-body testing [47], in four studies using whole-body

123

testing [15, 16, 22, 48] and in a comparison of the average

whole-body VO2max in PwMS and controls in all included
studies (p = 0.04) (Fig. 2).
3.4 Correlates of VO2max in PwMS
In the current review, a linear regression of the mean
aerobic capacity in PwMS of the individual studies on
EDSS and age both demonstrated moderate but significant
negative correlations (Fig. 3). The slope of the correlation between VO2max and EDSS indicated that an increase
of one point in EDSS will lower VO2max by
2.6 mL
kg-1
min-1. Furthermore, the aerobic capacity in
PwMS in relation to the percentage of men in the study
sample revealed a significant positive correlation (Fig. 3).
Within the context of the International Classification of
Functioning, Disability and Health (ICF) model [49],
Fig. 4 summarizes the studies reporting correlations/associations between aerobic capacity in PwMS and different
parameters. The two parameters most frequently associated
with VO2max were neurological disability (EDSS) (n = 8)
and fatigue (n = 4), which both produced weak-to-strong
negative correlations (range -0.25 to -0.77). Other parameters associated with VO2max were QoL, walking (timed
25-foot walk, and six-minute walk test), cognitive function
(cognitive processing speed and Paced Auditory Serial
Addition Test), duration of illness, body weight, plaque
volume in brain, vitamin D serum levels, Sickness Impact
Profile (SIP), Profile of Mood States (POMS), accelerometer counts, Godin leisure time exercise questionnaire
(GLTEQ), and measures of muscle strength.
3.5 Effect of Aerobic Training on VO2max in PwMS
Table 4 shows the intervention studies (n = 17) that included aerobic capacity as an outcome measure. Fourteen
studies (11 unique trials) used a randomized controlled
design (RCT), whereas three studies (two trials) used a
non-randomized controlled design (NRCT). Exercise duration ranged from 3 to 26 weeks, with sessions 25 times
a week, and each session lasting 1545 mins. All studies
used an aerobic training regime, and cycling was the preferred mode of exercise. Other exercise modes were
treadmill walking, rowing, and aqua aerobics. Together,
these studies comprised 330 patients undergoing aerobic
training versus 163 in the control condition.
The average PEDro score of the RCTs was 5.5 1.5
(range 38; see Table 4). Seven of these intervention
studies reported or provided sufficient data to be included
in a meta-analysis [40, 41, 45, 46, 5052]. One study used
upper-body testing, while all other studies applied bicycle
ergometry [52]. Together, these seven studies comprised
125 PwMS in the experimental condition versus 89 in the

Multiple Sclerosis and Aerobic Capacity

Table 2 Data on VO2plateau, RER, HRmax, lactate and RPE reported in the included studies (n = 19)
Testing protocola

Criteria used (achieved/mean SD/range)

Heine et al. [44]

Increment (M): 25 W ? 15 W/min

VO2plateau (37.5 %)

Heine et al. [79]

Increment (F): 25 W ? 10 W/min

RER [1.10 (69.6 %/1.17 0.13)

References

HRmax (48.2 %/90.6 10.9 % HRmax)

RPE C18 (23.2 %/16.3 0.2)
Langeskov-Christensen et al. [22]

Increment: 22.580 W ? 11.2542.5 W/1.5 min

VO2plateau (40 %)
RER [1.15 (95 %/1.27 0.10)
HRmax (65 %/171.1 16.8)
RPE C17 (95 %/17.7 0.8)
Lactate (75 %/9.5 2.7)

Skjerbaek et al. [52]

Feltham et al. [55]
Bansi et al. [54]

Increment: 0 W ? 10 W/min (arm ergometer)

HRmax (125.4 13.5)

Increment: 0 W ? 25 W/2 min

RER (1.11 0.2)

HRmax (continuous 147 7; interval 155 8)

WU: 0 W/3 min

Increment: WU ? 510 W/min

HRmax [EWG 126.8 (119.2134.4); ELG 134.1

(124.1144.0)]
RPE [8.5 (10-point scale)
[EWG 5.9 (4.96.8); ELG 6.3 (5.76.8)]

Motl and Fernhall [48]

WU: 0 W/5 min

RER [1.10 (1.18 0.09)

Increment: 0 W ? 15 W/min

HRmax (151.48 24.51)

Continuous protocol

RPE C17 (17.00 1.84)

Petruzzello and Motl [80]

WU: 0 W/5 min

RER [1.10 (1.2 0.1)

Petruzzello et al. [81]

Increment: 0 W ? 15 W/min

HRmax (156.8 20.6)

Continuous protocol

RPE C17 (16.9 1.6)

Kuspinar et al. [82]

Increment: 10 W ? 10 W/min

HRmax (170.9 16.2)

RPE (19.3 1.0)

Morrison et al. [66]

WU: 0 W/3 min

HRmax (158.6 12.6)

RER (1.2 0.1)

Increment: WU ? 520 W/min
Continuous protocol
Konecny et al. [74]

Increment: 0 W ? 20 W/2 min

HRmax (140.8 23.8)

RER (1.0 0.1)

Bjarnadottir et al. [40]

Koseoglu et al. [47]

Increment (M) = 20 W ? 20 W/min

RPE

Increment (F) = 15 W ? 15 W/min

(EG 17.2; C 16.6)

Increment: 25 W ? NR/3 min (arm ergometer)

HRmax (122.78 24.2)

RER (0.97 0.09)

Rasova et al. [41]

Increment: NR

RPE (untreated 14.62; treated 14.13)

Continuous protocol
Romberg et al. [43]

WU: 30 W/4 min

HRmax (men 159 22; women 152 20)

Increment: WU ? 1025 W/2 min

RPE (men 19.3 1.0; women 19.5 1.0)

Petajan et al. [51]

WU: 1540 W
Increment: WU ? 1040 W/1 or 2 min

HRmax (EG 172 4; C 180 3)

Ponichtera-Mulcare et al. [65]

2 9 5 min submaximal with 5 min rest

in between, followed by maximal stage
with braking force increments every 3060 s

HRmax (LEG 155 24)

Increment: 0 ? 12.5 W/3 min (arm ergometer)

HRmax (125.7 15.7)

Foglio et al. [69]

Lactate (LEG 6.9)

C control group, EG exercise group, ELG ergometer land group, EWG ergometer water group, F females, HRmax maximal heart rate, LEG leg
ergometry, M males, NR not reported, RER respiratory exchange ratio, RPE rating of perceived exertionBorg Scale, SD standard deviation,
VO2plateau plateau in oxygen uptake, WU warming up
a

Whole-body testing applied (e.g. cycle ergometer) unless stated otherwise

123

123
56

40

Heine et al. [44]

Heine et al. [79]

Langeskov-Christensen et al. [22]

62
MS: 31
HC: 31
11

48
MS: 32
HC: 16
(women)
42
AP: 21
IP: 21
25
(women)
49
Study 1: 25
Study 2: 24
(women)
36
MS: 21
HC: 15
(women)

Sandroff and Motl [16]

Sandroff et al. [15]

Motl and Fernhall [48]

Waschbisch et al. [24]

Petruzzello and Motl [80]

Petruzzello et al. [81]

Motl and Goldman [25]

Prakash et al. [18]

Mller et al. [84]

16

Skjerbaek et al. [83]

MS: 20
HC: 20

Sample size (n)

References

Table 3 Cross-sectional studies measuring VO2max in PwMS

2.2 (06)

Study 1: NR
Study 2: 2.4 1.6
(self-report version of
the Kurtzke EDSS)

NR

AP: 1.44 1.0

IP: 1.21 0.9

Median: 1.5 (06)

(self-report version of
the Kurtzke EDSS)

3.3 0.9

Median: 2 (05)
(self-report version of
the Kurtzke EDSS)

3.0 1.0

2.6 1.6

2.8 1.6

EDSS (mean SD/range)a

RR

Study 1: RR
Study 2: RR

RR

RR

RR

RR, PP, SP

RR, SP

RR, PP, SP

RR, SP

NR

MS course

RER [1.10
HRmax
RPE C18

HRmax
RPE C17
RER [1.10
HRmax
RPE C17
NR

RER [1.10
21.7 6.0

NR

HRmax
RPE
RER [1.10
HRmax
RPE C17

RER [1.10
HRmax
RPE C17
VO2plateau
RER [1.15

RER [1.15
HRmax
RPE C17
Lactate
NR

VO2plateau
RER [1.10
HRmax
RPE C18
VO2plateau

Criteria used

(L
min-1):
1.6 0.4
Study 1: 21.4 6.2
Study 2: 21.7 5.9

MS: 22.08 5.81

HC: 25.00 5.01
(L
min-1):
MS: 1.55 0.42
HC: 1.77 0.34
AP: 34.69 8.36
IP: 27.31 6.09

2.2 0.4

31.3 4.5
(L
min-1):

HC: 38.4 6.8

(L
min-1):
MS: 2.40 0.57
HC: 2.84 0.63
(mL
min-1):
2426 575
MS: 23.5 6.4
HC: 29.3 8.8

MS: 32.0 8.6

21.7 7.2
(L
min-1):
1.65 0.54

VO2max (mean SD)a

[mL
kg-1
min-1]

M. Langeskov-Christensen et al.

59

24

Kuspinar et al. [82]

Morrison et al. [66]

24
HFG: 12
LFG: 12
(women)
40
MS: 25
HC: 15
28
AT: 15
C: 13
92
Men: 33
Women: 59

Prakash et al. [17]

Koseoglu et al. [47]

20
MS: 10
HC: 10
16

Ponichtera-Mulcare et al. [65]

5.3 2

1.7 1.1

3.1 0.9

Men: 3.0 1.2

Women: 2.2 0.9

AT: 2.0 1.4

C: 2.5 0.8

4.38 2.6

2.61 1.76

3.0 1.2

Median: 2.75 (03.0)

Median: 1.5 (03.5)

EDSS (mean SD/range)a

NR

NR

NR

NR

RR, PP, SP

NR

RR

RR, PP, SP

NR

Definite MS, CIS

MS course

VO2plateau

(L
min-1):
MS: 2.60 0.64b
HC: 2.71 0.61b
11 2
(L
min-1):
0.7112 0.1516
(arm ergometer)

NR

NR

HRmax
RPE

NR

NR

RER [1.10
HRmax
RPE C18

NR

HRmax
RER
RPE
NR

Criteria used

30.4 8.2 (combined

arm-leg ergometer)

Men: 27.0 5.2

Women: 21.7 5.5
(L
min-1):
Men: 2.2 0.5
Women: 1.5 0.4

AT: 33.1 7.1

C: 28.9 7.8
(baseline values)

MS: 10.06 4.7

HC: 14.27 1.6
(arm ergometer)

HC: 25.7 5.3

(L
min-1):
MS: 1.5 0.5
HC: 1.8 0.6
20.6 5.9
(mL
min-1):
1,432.9 474.4
HFG: 25.41 4.82
LFG: 16.68 2.13

27.6 7.3
(L
min-1):
1.9 0.6
MS: 22.9 6.2

VO2max (mean SD)a

[mL
kg-1
min-1]

AP active PwMS, AT aerobic training, C control, CIS clinically isolated syndrome, EDSS Expanded Disability Status Scale, HC healthy control group, HFG high-fit group, HRmax maximal heart rate, IP inactive
PwMS, LFG low-fit group, MS multiple sclerosis, NR not reported, PwMS persons with multiple sclerosis, PP primary progressive, RER respiratory exchange ratio, RPE rating of perceived exertionBorg Scale,
RR relapsing-remitting, SD standard deviation, SP secondary progressive, VO2max maximal oxygen consumption, VO2plateau plateau in oxygen uptake
a
Unless stated otherwise
b
VO2max and/or SD estimated from figure

Foglio et al. [69]

White et al. [86]

Romberg et al. [72] (baseline results

from Romberg et al. [43])

Gold et al. [85] (results from

Schulz et al.) [46]

35

Konecny et al. [74]

MS: 12
HC: 12

Sample size (n)

References

Table 3 continued

Multiple Sclerosis and Aerobic Capacity

123

Sample
size (n)

42
ARM: 10
ROW: 11
BIC: 11
C: 10

11
AT: 6
C: 5

52
ELG: 28
EWG: 24

21
ATC: 12
ATI: 9

20
CET: 10
C: 10

11
AT ? NER
crossover

References

Briken et al.
[45]

Skjerbaek
et al. [52]

Bansi et al.
[53, 54]

123

Feltham et al.
[55]
Collet et al.
[56]

Golzari et al.
[50]

Rampello
et al. [57]

6.58.0
(inclusion)

RCT
(RoB:
8)

RCT
(RoB:
5)

RCT
(RoB:
3)

RCT
(RoB:
5)

CET:
2.14 1.06
C:
1.95 1.06
AT: 3.5 (16)
NER: 3.25
(1.56)
[median
(range)]

ELG: 4.7
(95 % CI
4.15.3)
EWG: 4.6
(95 % CI
4.05.2)
NR

ARM:
5.2 0.9
ROW:
4.7 0.8
BIC:
5.0 0.8
C: 4.9 0.9

RCT
(RoB:
7)

RCT
(RoB:
7)

EDSS
(mean
SD/range)a

Design

NR

RR

RR, PP, SP

NR

PP, SP

PP, SP

MS course

12

810

Duration
(weeks)

NR. 24
sessions
during
8 weeks

10 sessions
in total

23

Frequency
(days/
week)

Table 4 Studies measuring (the effects of aerobic training on) VO2max in PwMS

60 % of VO2max

NR

ATC: 45 % of peak
power
ATI: 90 % of peak
power

70 % of HRmax/
60 % of VO2max

6575 % of VO2max

Aerobic threshold
and/or 120130 %
of aerobic
threshold

Intensity

AT: 17.1 7.0

NER: 16.8 6.5

Ergometer bicycling
Duration: 40 min

Aerobic exercises
Duration: 20 min

ATC:
20.18 1.56
ATI:
22.89 1.80
(L
min-1):
ATC:
1.47 0.12
ATI:
1.70 0.14
CET:
33.84 13.08
C: 32.00 4.99

ATC:
23.59 2.29
ATI:
23.59 2.64
(L
min-1):
ATC:
1.69 0.17
ATI:
1.75 0.20
CET:
32.07 12.05
C:
35.00 15.78
AT: 20.0 6.6
NER: 16.9 6.1

EWG: 18.8
(16.321.2)
ELG: 19.7
(17.222.3)

(L
min-1):
ARM:
1,388 411
ROW:
1,312 416
BIC:
1,697 636
C:
1,254 1,041
(mL
min-1):
AT: 950 247
C: 874 365

(L
min-1):
ARM:
1,352 431
ROW:
1,306 422
BIC:
1,490 528
C:
1,377 1,145
(mL
min-1):
AT: 642 209
C: 872 386
(ARM)
EWG: 17.1
(14.619.6)
ELG: 18.1
(15.720.5)

VO2max
(mean SD)a
[mL
kg-1
min-1]
(post)

VO2max
(mean SD)a
[mL
kg-1
min-1]
(pre)

ATC ? ATI: ergometer bicycling

Fitness equipment designed for

persons with impaired lower-body
function,
arm ergometry and arm/leg
ergometry. Interval training
Duration: * 25 min
Ergometer bicycling/aquatic bike
Duration: 30 min

ARM, ROW or BIC.

Duration: 1545 min

Training regime

M. Langeskov-Christensen et al.

95
AT: 36
PT: 24
PT ? AT: 19
C: 16

28
AT: 15
C: 13
HC: 20
91
CET: 45
C: 46

Rasova et al.
[41]

Schulz et al.
[46]
Heesen et al.
[87]

46
AT: 21
C: 25

AT: 2.21
PT: 4.10
PT ? AT:
3.42
C: 2.31

AT:
2.0 1.4
C: 2.5 0.8
CET: 2.0
(15.5)
C: 2.5 (15.5)
[median
(range)]
AT:
4.6 1.2
C: 4.5 1.9

NRCT

RCT
(RoB:
5)

RCT
(RoB:
6)

RCT
(RoB:
5)

NRCT

AMB:
1.9 0.6
SEMI:
5.8 0.8
C: 3.3 0.9
AT:
3.8 0.3
C: 2.9 0.3

AT: 2.1
C: 1.8

RCT
(RoB:
6)

RCT
(RoB:
3)

EDSS
(mean
SD/range)a

Design

NR

15

24

34

RR, CP, B

NR

26

NR

NR

RR, PP and
SP

Duration
(weeks)

NR
(PP
excluded)

MS course

Frequency
(days/
week)

60 % of VO2max

5560 % of VO2max

Aerobic threshold
Approx. 60 % of
VO2max

NR

Maximal intensity
of 75 % of peak
power

60 % of VO2max

55 % of VO2max

Intensity

AT: 24.2 1.1

C: 25.8 1.3

AT: 29.4 1.3

C: 26.4 1.4

(L
min-1):
AMB: 2.05
SEMI: 1.35
C: 1.45

(L
min-1):
AMB: 1.73
SEMI: 1.27
C: 1.59
Arm/leg ergometer bicycling
Duration: 30 min

Arm and leg ergometry

Duration: 40 min

NR

NR

AT:
28.89 6.90
PT:
24.18 5.69
PT ? AT:
28.59 5.16
C: 28.14 8.57
AT: 36.5 7.6
C: 31.3 6.9

AT: 31.3
C: 22.9

VO2max
(mean SD)a
[mL
kg-1
min-1]
(post)

AT: 22.7
C: 22.3

NR

AT: 27.3
C: 23.4
(L
min-1):
AT: 1.9
C: 1.7
AT:
28.76 8.48
PT:
22.47 6.14
PT ? AT:
28.00 6.63
C: 29.92 8.33
AT: 33.1 7.1
C: 28.9 7.8

VO2max
(mean SD)a
[mL
kg-1
min-1]
(pre)

Ergometer bicycling
Duration: 30 min

Progressive home-based program.

Strength training and aerobic
exercise (aquatic training or other
preferred type)

Ergometer bicycling
Interval training Duration: 30 min

Ergometer bicycling
Duration: initial phase 210 min
Duration was increased progressively
until study end, where it was
1030 min

Ergometer bicycling (1520 min),

progressive resistance training,
stretching
Duration: 60 min

Training regime

AMB ambulatory, ARM arm ergometry, AT aerobic training, ATC continuous aerobic training, ATI interval aerobic training, B benign, BIC bicycle ergometry, C control, CET combined exercise training, CP
chronic progressive, EDSS: Expanded Disability Status Scale, ELG ergometer land group, EWG ergometer water group, HC healthy control group, HRmax maximal heart rate, MS multiple sclerosis, NER
neurological rehabilitation, NR not reported, NRCT non-randomized controlled trial, PP primary progressive, post post-intervention values, pre baseline values, PT physiotherapy, PwMS persons with
multiple sclerosis, RCT randomized controlled trial, RoB risk of bias by means of the PEDro scale, ROW rowing, RR relapsing-remitting, SEMI semi-ambulatory, SP secondary progressive, VO2max maximal
oxygen consumption
a
Unless stated otherwise

Petajan et al.
[51]

PonichteraMulcare
et al. [59]
Rodgers et al.
[60]

Mostert and
Kesselring
[58]

26
AT: 13
C: 13
23
AMB: 11
SEMI: 8
C: 4

16
AT: 6
C:10

Bjarnadottir
et al. [40]

Romberg et al.
[43]

Sample
size (n)

References

Table 4 continued

Multiple Sclerosis and Aerobic Capacity

123

M. Langeskov-Christensen et al.

Fig. 2 Column barchart illustrating absolute measures (mL
kg-1
min-1)

of VO2max in PwMS (MS group) compared with healthy controls
(CON group) when possible. Only baseline measures from intervention studies are reported. a Mean of different MS groups (e.g.
training group ? control group). b Absolute measures converted

(L
min-1 ? mL
kg-1
min-1) using reported mean subject weight.

c Only women. d Standard error of the mean between MS groups
shown. PwMS persons with multiple sclerosis, VO2max maximal
oxygen uptake, Asterisk indicates significant difference between
groups (p \ 0.05)

non-training control condition. A heterogeneous (I2 = 76 %,

p \ 0.01) significant effect was found in favor of the
experimental condition (Fig. 5; SMD = 0.63, 95 % CI
0.001.26, Z = 1.96, p = 0.05). Subsequently, a sensitivity analysis was performed to assess the effect of risk of
bias by excluding the studies using a non-randomized
controlled design or a PEDro score \5. A heterogeneous
(I2 = 74 %, p \ 0.01), significant effect was found in
favor of the experimental condition (SMD = 0.92, 95 %
CI 0.161.68, Z = 2.36, p = 0.02). Figure 6 shows the
percentage change in longitudinal studies (n = 10) that

assessed the effects of aerobic training on VO2max relative

to the SDC (8.1 %), as determined by Langeskov-Christensen et al. [22]. Eight of these ten studies showed improvements beyond 8.1 %.
The best powered observational study included in the
present review was Romberg et al. [43] (n = 95). Given
the SMD of 0.63 (95 % CI 0.001.26) found in favor of the
experimental group, and given the SD in the observational
study by Romberg et al. (5.2 for men, 5.5 for women), the
expected change in VO2max following an experimental
condition can be re-expressed as 3.3 mL
kg-1
min-1

123

Multiple Sclerosis and Aerobic Capacity

[56] and Feltham et al. [55] studied the effects of a continuous, interval, and combined aerobic training regime.
VO2max was reported for a subset of PwMS undergoing
continuous and interval aerobic training [55]. When combined, the total group showed a significant improvement in
VO2max following the intervention phase, with no difference between continuous or interval training. Rampello
et al. [57] studied the effects of aerobic training versus a
neurological rehabilitation program in a randomized
crossover trial. VO2max improved only after aerobic training but not following neurological rehabilitation. Romberg
et al. [43] studied the effects of 3 weeks of aquatic training
followed by 21 weeks of the participants preferred mode
of aerobic exercise. They found no group-by-time effect on
VO2max. Mostert and Kesselring [58] could not demonstrate
an improvement in VO2max following a short-term exercise
intervention consisting of five aerobic training sessions
during a 3-week timeframe. However, they did demonstrate a significant rightward shift of the aerobic threshold
in the exercise group. Finally, Ponichtera-Mulcare et al.
[59] and Rogers et al. [60] reported on the effects of
a 6-month aerobic training regime in ambulant PwMS
(EDSS 1.04.5) and semi-ambulant PwMS (EDSS
5.06.5). The ambulant participants improved their VO2max
by an average of 19 %, whereas the semi-ambulant participants improved their VO2max by an average of 7 %.

4 Discussion
Fig. 3 Aerobic capacity (whole-body VO2max: mL
kg-1
min-1) in
relation to the (a) EDSS, (b) age (years), and (c) sex distribution
(% men) in studies reporting these measures. Each dot represents the
mean value of aerobic capacity in a single study. If the aerobic
capacity was reported in L
min-1, it was corrected for the mean
sample body weight. VO2max maximal oxygen uptake, EDSS
Expanded Disability Status Score, Asterisk indicates significant
correlation

(95 % CI 0.06.6) for men and 3.5 mL
kg-1
min-1 (95 %

CI 0.06.9) for women.
Ten other studies assessed the effects of exercise versus
either a control condition or against a different exercise
modality [43, 50, 5360]. Bansi et al. [54] compared the
effects of cycling endurance training over land or in temperature regulated water (28 C). They hypothesized that
by training in water, symptom increment due to heat sensitivity may be limited, allowing for better effectiveness of
aquatic endurance training compared with over-land endurance training. Although both groups improved VO2max
(1.7 mL
kg-1
min-1) during the intervention phase, there
was no group-by-time interaction effect, indicating a differential effect of water-based training versus over-land
training. Published in two separate papers, Collett et al.

The present review provides a comprehensive overview of

the aerobic capacity in PwMS. The current body of
knowledge on the (1) psychometric properties of the gold
standard VO2max test in PwMS, (2) aerobic capacity of
PwMS, (3) effects of exercise on aerobic capacity in
PwMS, and (4) associations between aerobic capacity and
other outcomes in PwMS, was critically appraised. In
summary, the present review shows that, in general, the
gold standard VO2max test is valid and can be used reliably
in PwMS, that the VO2max was significantly lower in
PwMS compared with healthy controls, and that aerobic
training seems to be an effective tool in improving aerobic
capacity. Moreover, favorable associations between aerobic capacity and outcomes on all levels of the ICF model
have been reported, thereby suggesting that improving
aerobic capacity may induce improvements in MS-related
symptoms as well as provide secondary health benefits.
4.1 Testing of VO2max in PwMS
Although whole-body exercise testing of maximal aerobic
capacity, in terms of VO2max, is considered the gold standard, little research has been undertaken to study the

123

M. Langeskov-Christensen et al.

Fig. 4 Overview, in context of the ICF model, of studies reporting

significant correlations/associations between aerobic capacity
(VO2max) in PwMS and different parameters. EDSS Expanded
Disability Status Scale, FSS Fatigue Severity Scale, GLTEQ Godin
Leisure-Time Exercise Questionnaire, ICF International Classification of Functioning, Disability and Health, MFIS Modified Fatigue

Impact Scale, PASAT Paced Auditory Serial Addition Test, POMS

Profile of Mood States, PwMS persons with multiple sclerosis, SF-36
Short-Form Health Survey, SIP Sickness Impact Profile, T25FW
timed 25-foot walk, VO2max maximal oxygen uptake, 6MWT sixminute walk test

Fig. 5 Meta-analysis of aerobic training (experimental) versus a nontraining control condition (control) on VO2max in PwMS. A significant
effect (p = 0.05) in favor of aerobic training was found with a
standardized mean difference of 0.63 (95 % CI 0.001.26). CI

confidence interval, df degrees of freedom, IV inverse variance,

PwMS persons with multiple sclerosis, SD standard deviation, Std
standardized, VO2max maximal oxygen uptake

psychometric properties of this test in PwMS. A detailed

examination of studies that include VO2max measurements
in PwMS (Table 2) shows that the classical criteria are
often inadequately reported. The inconsistency in the literature to measure and/or report on the criteria may rely on
the uncertainty of the ability of PwMS to attain these criteria. Consequently, the term VO2peak is regularly used.

This is problematic since an accurate and valid VO2max

value is of considerable physiological importance when
analyzing the health of the implicated subjects in the present review. However, based on two studies [22, 44],
testing of VO2max can be considered valid in ambulant
PwMS (EDSS B4.0) and possibly in PwMS with higher
levels of disability. When comparing VO2max results, it is

123

Multiple Sclerosis and Aerobic Capacity

Fig. 6 Change (%) in VO2max
following aerobic training in
PwMS. The dashed line
represents the smallest reliable
change of 8.1 %, as identified
by Langeskov-Christensen et al.
[22]. PwMS persons with
multiple sclerosis, VO2max
maximal oxygen uptake

crucial to consider the applied test protocols since the

outcome may vary as a consequence of, for example, test
duration, test equipment (i.e. leg, arm, or combined arm/leg
ergometer) or watt increase. Although the watt increase
varied in the two studies (Table 2), test duration and
equipment were similar and caused voluntary exhaustion
within the recommended 812 min [61]. Furthermore, in
low to mild disabled PwMS, the occurrence of a VO2
plateau, the primary criterion for a true maximal exercise
test, was found equally frequent to healthy people. In addition, the secondary criteria seem to also be well-attained
(Table 2). However, a relatively low heart rate at voluntary
exhaustion, in the absence of heart-rate regulation
medication, may be an indicator of a symptom-limited test
result [44]. It should also be considered that MS-related
alterations in body function, for instance changes in energy
substrate use [62] or autonomic dysfunction [63], may
compromise the validity of some of the secondary criteria
in PwMS. One study assessed the reliability of VO2max
testing in PwMS [22]. Langeskov-Christensen and colleagues showed that a [8.1 % change in VO2max indicates
a significant improvement. No research was available
which studied the clinically relevant difference of such a
change in VO2max in PwMS on, for instance, fatigue, risk of
comorbidity, disease progression or health-related QoL.
Consequently, future studies evaluating the smallest
clinically relevant change in different disability subgroups
are warranted. The associations that have been identified
between VO2max and various other outcomes (Fig. 4) offer
opportunities to assess clinically meaningful changes related to changes in VO2max. The psychometric properties of
VO2max testing have so far been studied in ambulant PwMS
(EDSS B6.0) using whole-body testing, but recently the
potential benefits of aerobic training have also been studied
in PwMS with more severe levels (EDSS [6.0) of disability [52]. In these PwMS, whole-body testing is most
likely not feasible due to leg impairment, making studies

evaluating the psychometric properties of upper-body

testing (in severely disabled PwMS) warranted.
4.2 Aerobic Capacity in PwMS
It is frequently stated in the literature that the VO2max is
impaired in PwMS [3, 64], despite no studies summarizing
the existing knowledge. The cross-sectional results of the
present review, covering VO2max measurements from 40
studies on PwMS, reveal that healthy controls achieve
significantly higher VO2max values compared with PwMS.
This is demonstrated by the comparison of the average
whole-body as well as upper-body VO2max in PwMS and
the control from all included studies (Fig. 2). Furthermore,
five individual MS studies incorporating controls [15, 16,
22, 47, 48] also found a significant difference in VO2max
between PwMS and the control. Two studies [65, 66]
found no difference between PwMS and controls, but one
of the studies included sedentary controls to make fitness
levels as equivalent as possible between the study groups
[66], while the second study included a small study sample
(PwMS = 10) with a mild mean disability level (EDSS
1.7 1.1) and a control selected on fitness parameters
(i.e. exercise participation level, lifestyle) to match the
PwMS as closely as possible [65]. Nonetheless, by
matching subjects with respect to activity level, these
studies indicate that the lower VO2max generally seen in
PwMS is probably related to a lower daily activity level,
which is frequently reported in PwMS [67], rather than
non-reversible tissue damage. In accordance, cardiovascular adaption in PwMS at maximal and submaximal
levels of aerobic intensity has been shown to be analogous
to non-diseased adults [55]. Furthermore, the difference in
aerobic capacity may also be related to a lowering of the
HRmax in PwMS since the VO2max is closely connected
with HRmax [68]. Previous studies have reported significantly lower HRmax in PwMS versus healthy controls

123

M. Langeskov-Christensen et al.

[48, 58]. In contrast, it has been demonstrated that autonomic test results during a simple cycle test [63] and HR
measured at 25, 50, 75, and 100 % of VO2max [66] did not
differ between PwMS and healthy controls. In some
PwMS, a non-linear heart response is observed, possibly
due to deconditioning [55]. Interestingly, this has been
suggested to improve from aerobic training [55], relating
abnormal heart rate in PwMS to deconditioning rather than
autonomic dysfunction. Foglio et al. [69] has stated respiratory dysfunction as the cause of reduction in exercise
capacity in PwMS. However, the study employed arm
ergometry, which is not necessarily reflective of wholebody VO2max testing because of the small proportion of
muscle mass involved [70]. Furthermore, the high respiratory reserve [58] and ventilation at the time of VO2max
[22] reported elsewhere demonstrated that ventilation is
unlikely the limiting factor for PwMS.
VO2max is a strong health predictor, and data from
otherwise healthy men and women show that individuals
with a VO2max of 27.7 mL
kg-1
min-1 or more have a
substantially lower rate of all-cause mortality and cardiovascular events compared with those with a VO2max of
less than 27.7 mL
kg-1
min-1 [71]. The current systematic review established an average body-weight-adjusted
VO2max of 25.5 5.2 mL
kg-1
min-1 in PwMS, which is
clearly below this threshold, strongly emphasizing that
interventions aiming at improving the VO2max should be
encouraged for PwMS.
4.3 Correlates of VO2max in PwMS
The variety of whole-body VO2max values observed in
the reported MS studies (range 16.9 6.336.7 10.7
mL
kg-1
min-1) may be related to differences in EDSS,
age and sex distribution of the experimental groups since
the present study found that VO2max correlated significantly
with these factors (Fig. 3). Earlier individual MS studies
have reported similar findings with weak to moderate
correlations (range -0.25 to -0.58) between VO2max and
EDSS [22, 44, 47, 59, 65, 7274], while no previous MS
studies examining the association with age and sex distribution were found (Fig. 4). However, since the EDSS increases with disease duration, which is associated with
aging, and the VO2max drops expectedly with age [75], it
still seems unclear to what extent the EDSS may influence
the VO2max in PwMS. The present data did not allow a
multivariate regression analysis to clarify this. Moreover,
the EDSS is not an ideal neurologic rating scale and has
been criticized for expressing a medium reliability, nonlinearity, poor sensitivity and a lack of psychometric input
[76, 77], which may further complicate the interpretation of
the presented linear correlation. Nonetheless, evidence in
the literature indicates that aging does not represent a

123

contraindication to aerobic training in elderly healthy

people and in elderly patients with different kinds of comorbidities [75]. This suggests that PwMS across the disability spectrum may achieve the beneficial effects of
aerobic training on the physiologic decline of cardiovascular performance occurring with age in the same way as
the healthy population. Furthermore, sex distribution correlated significantly with VO2max, which may influence the
average VO2max in PwMS in this review, given the high
proportion of women in the included studies. In accordance, the mean VO2max in the studies only including
women was 23.9 4.5 mL
kg-1
min-1, suggesting that
the mean VO2max in PwMS with an even sex distribution
may be near the all-cause mortality and cardiovascular
events threshold value of 27.7 mL
kg-1
min-1 reported
in the healthy population [71]. In light of this, the slope of
the correlation between VO2max and EDSS in the current study, indicating that VO2max will decrease by
2.6 mL
kg-1
min-1 per one point increase of EDSS, suggests that disease progression is likely to considerably increase the risk of all-cause mortality and cardiovascular
events in PwMS.
The analysis of correlates revealed associations with a
variety of outcomes covering all levels of the ICF model
(Fig. 4). Although the causality and precise underlying
mechanisms are not fully understood, this implies that the
VO2max is an important health and performance marker in
PwMS. In addition, this map may help identify new areas
that can be affected by aerobic training or other interventions that improve VO2max in MS.
4.4 Effect of Aerobic Training on VO2max in PwMS
The present review shows that maximal aerobic capacity,
in terms of VO2max, is indeed compromised in PwMS.
Fortunately, VO2max can be improved by aerobic training
programs including sufficient intensity, progression and
duration to stress the cardiopulmonary system. Seventeen
trials that studied the effect of aerobic training on VO2max
in PwMS were identified in the present review. A metaanalysis, including the seven studies that provided sufficient data for meta-analysis, showed that aerobic training
in PwMS may improve VO2max by as much as
3.5 mL
kg-1
min-1. However, it should be noted that
aerobic training was used primarily in ambulant PwMS
(EDSS B6.0) and little is known regarding the feasibility
and effects of aerobic training in PwMS that have a higher
disability level. Nonetheless, the clinical relevance of such
a change is illustrated in a recent paper by Kodama et al.
[71] who showed that a change of one metabolic equivalent
(MET), which equals 3.5 mL
kg-1
min-1 VO2, was associated with a 1.48 difference in the odds of developing
CVD. However, the heterogeneous moderate effect found

Multiple Sclerosis and Aerobic Capacity

in the present study (SMD = 0.63, 95 % CI 0.001.26)

indicates a varied response to exercise between studies.
Unfortunately, insufficient data were available to perform a
sensitivity analysis based on the duration, frequency and/or
intensity of the aerobic training regime. However, it is
likely that the heterogeneity and moderate effect are partly
explained by responders and non-responders to aerobic
training on an individual level [26]. A second important
consideration is the level of disability and whether this
affects the potential benefits of exercise therapy in PwMS.
For example, Ponichtera-Mulcare et al. divided their
PwMS into two groups based on EDSS score (EDSS B4.5
vs. EDSS 56.5). Following 24 weeks of combined arm
leg ergometry, the group with a low level of disability
improved their VO2max by 19 %, whereas the group with a
moderate level of disability improved their VO2max by 7 %
[59]. This suggests that although deconditioning seems
larger in the moderate disabled group, the level of improvement was smaller, which may also argue for early
treatment. However, the contrary has also been shown, as
in the study by Mostert and Kesselring [58], where the
more disabled PwMS tended to improve more. In addition
to the level of disability, intraindividual genetic susceptibility to improve cardiovascular function may also differ
and may even result in deterioration in VO2max in some
[26]. Hence, the imminent challenge is to identify responders from non-responders, and to provide an optimal
and effective treatment that lasts beyond the intervention
phase by means of incorporating an active lifestyle. The
latter means that health behavioral techniques are incorporated into exercise therapeutic programs to improve
physical activity behavior [78].
4.5 Future Directions
Aerobic training to improve aerobic capacity in PwMS is a
viable therapeutic agent to improve MS-related symptoms
and provide health benefits. Furthermore, increasing the
physical activity level may also improve VO2max, given the
frequently reported inactivity in PwMS. Future research
should further expand our knowledge on the optimal
(personalized) content of aerobic training for PwMS, the
application of behavior-changing models to increase the
physical activity level in PwMS, and how aerobic capacity
is related to MS-related symptoms and disease progression.
Such research can enlarge our confidence in the added
value of aerobic training for PwMS as the present review
shows large heterogeneity in the effects of such training in
PwMS. Moreover, the limited knowledge on the psychometric properties of VO2max testing in PwMS warrants
future studies, including studies that determine the required
change in VO2max by which beneficial improvements in the
associated outcomes can be expected, as well as the

psychometric properties of upper-body testing (in severely

disabled PwMS). Finally, the relation between EDSS and
aerobic capacity, shown in the present review, suggest a
rationale for preventive interventions related to physical
activity behavior to maintain or reduce the decline in cardiovascular health with disease progression.
5 Conclusions
Aerobic capacity in PwMS is impaired compared with
healthy individuals. A valid, reliable, and responsive test
can be performed in at least ambulant PwMS (EDSS B6.0)
by using the gold standard whole-body maximal exercise
test. Aerobic capacity is associated with factors on all
levels of the ICF model, including disease progression. The
current literature shows that aerobic training can improve
aerobic capacity to a level where it likely reduces secondary health risks.
Acknowledgments Martin Langeskov-Christensen, Martin Heine,
Gert Kwakkel and Ulrik Dalgas have no potential conflicts of interest
that are directly relevant to the content of this review. The study was
not funded by any external source.

References
1. Koch-Henriksen N, Sorensen PS. The changing demographic
pattern of multiple sclerosis epidemiology. Lancet Neurol.
2010;9(5):52032.
2. Compston A, Coles A. Multiple sclerosis. Lancet. 2008;
372(9648):150217.
3. Motl RW, Pilutti LA. The benefits of exercise training in multiple
sclerosis. Nat Rev Neurol. 2012;8(9):48797.
4. Confavreux C, Vukusic S. Natural history of multiple sclerosis: a
unifying concept. Brain. 2006;129(Pt 3):60616.
5. Kurtzke JF. Rating neurologic impairment in multiple sclerosis:
an Expanded Disability Status Scale (EDSS). Neurology.
1983;33(11):144452.
6. Marrie RA, Horwitz RI. Emerging effects of comorbidities on
multiple sclerosis. Lancet Neurol. 2010;9(8):8208.
7. Ehde DM, Bombardier CH. Depression in persons with multiple
sclerosis. Phys Med Rehabil Clin N Am. 2005;16(2):43748 (ix).
8. Fisk JD, Pontefract A, Ritvo PG, et al. The impact of fatigue on
patients with multiple sclerosis. Can J Neurol Sci. 1994;21(1):
914.
9. Dalgas U, Ingemann-Hansen T, Stenager E. Physical exercise and
MS: recommendations. Int MS J. 2009;16(1):511.
10. Wens I, Dalgas U, Stenager E, et al. Risk factors related to cardiovascular diseases and the metabolic syndrome in multiple
sclerosis: a systematic review. Mult Scler. 2013;19(12):155664.
11. Carnethon MR, Gulati M, Greenland P. Prevalence and cardiovascular disease correlates of low cardiorespiratory fitness in
adolescents and adults. JAMA. 2005;294(23):29818.
12. Eriksen L, Curtis T, Gronbaek M, et al. The association between
physical activity, cardiorespiratory fitness and self-rated health.
Prev Med. 2013;57(6):9002.
13. Bassett DR Jr, Howley ET. Limiting factors for maximum oxygen uptake and determinants of endurance performance. Med Sci
Sports Exerc. 2000;32(1):7084.

123

M. Langeskov-Christensen et al.
14. Marrie RA, Hanwell H. General health issues in multiple sclerosis: comorbidities, secondary conditions, and health behaviors.
Continuum (Minneap Minn). 2013;19(4 Multiple Sclerosis):
104657.
15. Sandroff BM, Sosnoff JJ, Motl RW. Physical fitness, walking
performance, and gait in multiple sclerosis. J Neurol Sci.
2013;328(12):706.
16. Sandroff BM, Motl RW. Fitness and cognitive processing speed
in persons with multiple sclerosis: a cross-sectional investigation.
J Clin Exp Neuropsychol. 2012;34(10):104152.
17. Prakash RS, Snook EM, Erickson KI, et al. Cardiorespiratory
fitness: a predictor of cortical plasticity in multiple sclerosis.
Neuroimage. 2007;34(3):123844.
18. Prakash RS, Snook EM, Motl RW, et al. Aerobic fitness is associated with gray matter volume and white matter integrity in
multiple sclerosis. Brain Res. 2010;1341:4151.
19. Cress ME, Meyer M. Maximal voluntary and functional performance levels needed for independence in adults aged 65 to 97
years. Phys Ther. 2003;83(1):3748.
20. Stickland MK, Butcher SJ, Marciniuk DD, et al. Assessing exercise limitation using cardiopulmonary exercise testing. Pulm
Med. 2012;2012:824091.
21. Whipp B. The peak versus maximum oxygen uptake issue. CPX
International Inc.; 2010. p. 19.
22. Langeskov-Christensen M, Langeskov-Christensen D, Overgaard
K, et al. Validity and reliability of VO-max measurements in
persons with multiple sclerosis. J Neurol Sci. 2014;342(12):
7987.
23. Mokkink LB, Terwee CB, Patrick DL, et al. The COSMIN study
reached international consensus on taxonomy, terminology, and
definitions of measurement properties for health-related patientreported outcomes. J Clin Epidemiol. 2010;63(7):73745.
24. Waschbisch A, Wenny I, Tallner A, et al. Physical activity in
multiple sclerosis: a comparative study of vitamin D, brainderived neurotrophic factor and regulatory T cell populations. Eur
Neurol. 2012;68(2):1228.
25. Motl RW, Goldman M. Physical inactivity, neurological disability, and cardiorespiratory fitness in multiple sclerosis. Acta
Neurol Scand. 2011;123(2):98104.
26. Bouchard C, An P, Rice T, et al. Familial aggregation of VO(2max)
response to exercise training: results from the HERITAGE family
study. J Appl Physiol (1985). 1999;87(3):10038.
27. Dalgas U, Stenager E, Ingemann-Hansen T. Multiple sclerosis
and physical exercise: recommendations for the application of
resistance-, endurance- and combined training. Mult Scler.
2008;14(1):3553.
28. Petajan JH, White AT. Recommendations for physical activity in
patients with multiple sclerosis. Sports Med. 1999;27(3):17991.
29. Latimer-Cheung AE, Pilutti LA, Hicks AL, et al. Effects of exercise training on fitness, mobility, fatigue, and health-related
quality of life among adults with multiple sclerosis: a systematic
review to inform guideline development. Arch Phys Med Rehabil. 2013;94(9):180028.
30. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement
for reporting systematic reviews and meta-analyses of studies that
evaluate healthcare interventions: explanation and elaboration.
BMJ. 2009;339:b2700.
31. Polman CH, Reingold SC, Banwell B, et al. Diagnostic criteria
for multiple sclerosis: 2010 revisions to the McDonald criteria.
Ann Neurol. 2011;69(2):292302.
32. Howley ET, Bassett DR Jr, Welch HG. Criteria for maximal
oxygen uptake: review and commentary. Med Sci Sports Exerc.
1995;27(9):1292301.
33. Duncan GE, Howley ET, Johnson BN. Applicability of VO2max
criteria: discontinuous versus continuous protocols. Med Sci
Sports Exerc. 1997;29(2):2738.

123

34. Midgley AW, McNaughton LR, Polman R, et al. Criteria for determination of maximal oxygen uptake: a brief critique and recommendations for future research. Sports Med. 2007;37(12):
101928.
35. Maher CG, Sherrington C, Herbert RD, et al. Reliability of the
PEDro scale for rating quality of randomized controlled trials.
Phys Ther. 2003;83(8):71321.
36. Herbert R, Moseley A, Sherrington C. PEDro: a database of
randomised controlled trials in physiotherapy. Health Inf Manag.
1998;28(4):1868.
37. Caspersen CJ, Powell KE, Christenson GM. Physical activity,
exercise, and physical fitness: definitions and distinctions for
health-related research. Public Health Rep. 1985;100(2):12631.
38. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting.
Meta-analysis Of Observational Studies in Epidemiology
(MOOSE) group. JAMA. 2000;283(15):200812.
39. Higgins J, Green S, editors. Cochrane handbook for systematic
reviews of interventions. Version 5.1.0 [updated March 2011].
The Cochrane Collaboration. Available at: www.cochranehandbook.org. Accessed 24 Nov 2014
40. Bjarnadottir OH, Konradsdottir AD, Reynisdottir K, et al. Multiple sclerosis and brief moderate exercise. A randomised study.
Mult Scler. 2007;13(6):77682.
41. Rasova K, Havrdova E, Brandejsky P, et al. Comparison of the
influence of different rehabilitation programmes on clinical,
spirometric and spiroergometric parameters in patients with
multiple sclerosis. Mult Scler. 2006;12(2):22734.
42. Cohen J. Statistical power analysis for the behavioral sciences.
New York: Academic Press; 1988.
43. Romberg A, Virtanen A, Ruutiainen J, et al. Effects of a 6-month
exercise program on patients with multiple sclerosis: a randomized study. Neurology. 2004;63(11):20348.
44. Heine M, Hoogervorst EL, Hacking HG, et al. Validity of maximal exercise testing in patients with low to moderate multiple
sclerosis. Phys Ther. 2014;94(8):116875.
45. Briken S, Gold SM, Patra S, et al. Effects of exercise on fitness
and cognition in progressive MS: a randomized, controlled pilot
trial. Mult Scler. 2014;20(3):38290.
46. Schulz KH, Gold SM, Witte J, et al. Impact of aerobic training on
immune-endocrine parameters, neurotrophic factors, quality of
life and coordinative function in multiple sclerosis. J Neurol Sci.
2004;225(12):118.
47. Koseoglu BF, Gokkaya NK, Ergun U, et al. Cardiopulmonary and
metabolic functions, aerobic capacity, fatigue and quality of life
in patients with multiple sclerosis. Acta Neurol Scand.
2006;114(4):2617.
48. Motl RW, Fernhall B. Accurate prediction of cardiorespiratory
fitness using cycle ergometry in minimally disabled persons with
relapsing-remitting multiple sclerosis. Arch Phys Med Rehabil.
2012;93(3):4905.
49. Jette AM. Toward a common language for function, disability,
and health. Phys Ther. 2006;86(5):72634.
50. Golzari Z, Shabkhiz F, Soudi S, et al. Combined exercise training
reduces IFN-(gamma) and IL-17 levels in the plasma and the
supernatant of peripheral blood mononuclear cells in women
with multiple sclerosis. Int Immunopharmacol. 2010;10(11):
14159.
51. Petajan JH, Gappmaier E, White AT, et al. Impact of aerobic
training on fitness and quality of life in multiple sclerosis. Ann
Neurol. 1996;39(4):43241.
52. Skjerbaek AG, Naesby M, Lutzen K, et al. Endurance training is
feasible in severely disabled patients with progressive multiple
sclerosis. Mult Scler. 2014;20(5):62730.
53. Bansi J, Bloch W, Gamper U, et al. Endurance training in
MS: short-term immune responses and their relation to

Multiple Sclerosis and Aerobic Capacity

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

cardiorespiratory fitness, health-related quality of life, and fatigue. J Neurol. 2013;260(12):29933001.

Bansi J, Bloch W, Gamper U, et al. Training in MS: influence of
two different endurance training protocols (aquatic versus overland) on cytokine and neurotrophin concentrations during three
week randomized controlled trial. Mult Scler. 2013;19(5):
61321.
Feltham MG, Collett J, Izadi H, et al. Cardiovascular adaptation
in people with multiple sclerosis following a twelve week exercise programme suggest deconditioning rather than autonomic
dysfunction caused by the disease. Results from a randomized
controlled trial. Eur J Phys Rehabil Med. 2013;49(6):76574.
Collett J, Dawes H, Meaney A, et al. Exercise for multiple
sclerosis: a single-blind randomized trial comparing three exercise intensities. Mult Scler. 2011;17(5):594603.
Rampello A, Franceschini M, Piepoli M, et al. Effect of aerobic
training on walking capacity and maximal exercise tolerance in
patients with multiple sclerosis: a randomized crossover controlled study. Phys Ther. 2007;87(5):54555.
Mostert S, Kesselring J. Effects of a short-term exercise training
program on aerobic fitness, fatigue, health perception and activity
level of subjects with multiple sclerosis. Mult Scler. 2002;8(2):
1618.
Ponichtera-Mulcare JA, Mathews T, Barrett PJ, et al. Change in
aerobic fitness of patients with multiple sclerosis during a
6-month training program. Sports Med Train Rehabil. 1997;
7(34):26572.
Rodgers MM, Mulcare JA, King DL, et al. Gait characteristics of
individuals with multiple sclerosis before and after a 6-month
aerobic training program. J Rehabil Res Dev. 1999;36(3):1838.
Buchfuhrer MJ, Hansen JE, Robinson TE, et al. Optimizing the
exercise protocol for cardiopulmonary assessment. J Appl Physiol
Respir Environ Exerc Physiol. 1983;55(5):155864.
Mahler A, Steiniger J, Bock M, et al. Is metabolic flexibility
altered in multiple sclerosis patients? PLoS One. 2012;7(8):
e43675.
Hale LA, Nukada H, Du Plessis LJ, et al. Clinical screening of
autonomic dysfunction in multiple sclerosis. Physiother Res Int.
2009;14(1):4255.
Gallien P, Nicolas B, Robineau S, et al. Physical training and
multiple sclerosis. Ann Readapt Med Phys. 2007;50(6):3736
(36972).
Ponichtera-Mulcare JA, Mathews T, Glaser RM, et al. Maximal
aerobic exercise of individuals with multiple sclerosis using three
modes of ergometry. Clin Kinesiol. 1995;49(1):413.
Morrison EH, Cooper DM, White LJ, et al. Ratings of perceived
exertion during aerobic exercise in multiple sclerosis. Arch Phys
Med Rehabil. 2008;89(8):15704.
Ng AV, Kent-Braun JA. Quantitation of lower physical activity
in persons with multiple sclerosis. Med Sci Sports Exerc.
1997;29(4):51723.
Christou DD, Seals DR. Decreased maximal heart rate with aging
is related to reduced {beta}-adrenergic responsiveness but is
largely explained by a reduction in intrinsic heart rate. J Appl
Physiol (1985). 2008;105(1):249.
Foglio K, Clini E, Facchetti D, et al. Respiratory muscle function
and exercise capacity in multiple sclerosis. Eur Respir J.
1994;7(1):238.

70. Taylor HL, Buskirk E, Henschel A. Maximal oxygen intake as an

objective measure of cardio-respiratory performance. J Appl
Physiol. 1955;8(1):7380.
71. Kodama S, Saito K, Tanaka S, et al. Cardiorespiratory fitness as a
quantitative predictor of all-cause mortality and cardiovascular
events in healthy men and women: a meta-analysis. JAMA.
2009;301(19):202435.
72. Romberg A, Virtanen A, Aunola S, et al. Exercise capacity,
disability and leisure physical activity of subjects with multiple
sclerosis. Mult Scler. 2004;10(2):2128.
73. Motl RW, Goldman MD. Physical inactivity, neurological disability, and cardiorespiratory fitness in multiple sclerosis. Mult
Scler. 2010;16(8):1019.
74. Konecny L, Pospisil P, Dufek M, et al. Functional impairment in
multiple sclerosis. Scr Med Fac Med Univ Brun Masarykianae.
2007;80(5):22532.
75. Vigorito C, Giallauria F. Effects of exercise on cardiovascular
performance in the elderly. Front Physiol. 2014;5:51.
76. Hobart J, Freeman J, Thompson A. Kurtzke scales revisited: the
application of psychometric methods to clinical intuition. Brain.
2000;123(Pt 5):102740.
77. Sharrack B, Hughes RA, Soudain S, et al. The psychometric
properties of clinical rating scales used in multiple sclerosis.
Brain. 1999;122(Pt 1):14159.
78. Geidl W, Semrau J, Pfeifer K. Health behaviour change theories:
contributions to an ICF-based behavioural exercise therapy for
individuals with chronic diseases. Disabil Rehabil. 2014;36(24):
2091100.
79. Heine M, Verschuren O, Kwakkel G. Validity of oxygen uptake
efficiency slope in patients with multiple sclerosis. J Rehabil
Med. 2014;46(7):65661.
80. Petruzzello SJ, Motl RW. Acute moderate-intensity cycling exercise is associated with reduced fatigue in persons with multiple
sclerosis. Ment Health Phys Act. 2011;4(1):14.
81. Petruzzello SJ, Snook EM, Gliottoni RC, et al. Anxiety and mood
changes associated with acute cycling in persons with multiple
sclerosis. Anxiety Stress Coping. 2009;22(3):297307.
82. Kuspinar A, Andersen RE, Teng SY, et al. Predicting exercise
capacity through submaximal fitness tests in persons with multiple sclerosis. Arch Phys Med Rehabil. 2010;91(9):14107.
83. Skjerbaek AG, Mller AB, Jensen E, et al. Heat sensitive persons
with multiple sclerosis are more tolerant to resistance exercise
than to endurance exercise. Mult Scler. 2013;19(7):93240.
84. Mller AB, Bibby BM, Skjerbaek AG, et al. Validity and variability of the 5-repetition sit-to-stand test in patients with multiple
sclerosis. Disabil Rehabil. 2012;34(26):22518.
85. Gold SM, Schulz KH, Hartmann S, et al. Basal serum levels and
reactivity of nerve growth factor and brain-derived neurotrophic
factor to standardized acute exercise in multiple sclerosis and
controls. J Neuroimmunol. 2003;138(12):99105.
86. White AT, Wilson TE, Davis SL, et al. Effect of precooling on
physical performance in multiple sclerosis. Mult Scler. 2000;6(3):
17680.
87. Heesen C, Gold SM, Hartmann S, et al. Endocrine and cytokine
responses to standardized physical stress in multiple sclerosis.
Brain Behav Immun. 2003;17(6):47381.

123