Journal of the Peripheral Nervous System 20:347–362 (2015)

REVIEW
Systematic review of exercise for
Charcot-Marie-Tooth disease

Amy D. Sman1,2,3,† , Daniel Hackett4,† , Maria Fiatarone Singh4,5,6 , Ché Fornusek4 ,

Manoj P. Menezes1,7 , and Joshua Burns1,2,3
1
Institute for Neuroscience and Muscle Research, The Children’s Hospital at Westmead, Westmead, Australia; 2 Discipline of
Physiotherapy, Faculty of Health Sciences, The University of Sydney, Lidcombe, Australia; 3 Paediatric Gait Analysis Service of
New South Wales, Sydney Children’s Hospitals Network (Randwick and Westmead), Sydney, Australia; 4 Discipline of
Exercise and Sport Science, Faculty of Health Sciences, The University of Sydney, Lidcombe, Australia; 5 Sydney Medical
School & Charles Perkins Centre, The University of Sydney, Camperdown, Australia; 6 Hebrew Senior Life, and Jean Meyer
USDA Human Nutrition Research Center on Aging, Tufts University, Boston, MA, 02111, USA; and 7 T.Y. Nelson Department of
Neurology & Neurosurgery, The Children’s Hospital at Westmead, Westmead, Australia

Abstract Charcot-Marie-Tooth disease (CMT) is a slowly progressive hereditary degen-

erative disease and one of the most common neuromuscular disorders. Exercise may be
beneficial to maintain strength and function for people with CMT, however, no comprehen-
sive evaluation of the benefits and risks of exercise have been conducted. A systematic
review was completed searching numerous electronic databases from earliest records to
February 2015. Studies of any design including participants of any age with confirmed diag-
nosis of CMT that investigated the effects of exercise were eligible for inclusion. Of 13,301
articles identified following removal of duplicates, 11 articles including 9 unique studies met
the criteria. Methodological quality of studies was moderate, sample sizes were small, and
interventions and outcome measures used varied widely. Although the majority of the stud-
ies identified changes in one or more outcome measurements across exercise modalities,
the majority were non-significant, possibly due to Type II errors. Significant effects described
included improvements in strength, functional activities, and physiological adaptations fol-
lowing exercise. Despite many studies showing changes in strength and function following
exercise, findings of this review should be met with caution due to the few studies avail-
able and moderate quality of evidence. Well-powered studies, harmonisation of outcome
measures, and clearly described interventions across studies would improve the quality and
comparability of the evidence base. The optimal exercise modality and intensity for people
with CMT as well as the long-term safety of exercise remain unclear.

Key words: Charcot-Marie-Tooth disease, exercise, hereditary motor and sensory neuropa-
thy, strength, systematic review

Introduction
It is well established that there are many potential
Address correspondence to: Amy D. Sman, Institute for Neuro- health benefits from exercise in healthy individuals,
science and Muscle Research, The Children’s Hospital at West- including reduced risk of obesity, osteoporosis,
mead, Locked Bag 4001, Westmead, NSW 2145, Australia.
Tel: +(61)2-9845-3004; Fax +(61)2-9845-1317; E-mail: amy.sman@ heart disease, and diabetes (Murphy et al., 2010).
sydney.edu.au Exercise for people with Charcot-Marie-Tooth disease
† These authors contributed equally to this work. (CMT) may be beneficial to maintain strength and

© 2015 Peripheral Nerve Society 347

Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

functional range of motion, using muscle strength- databases (Appendix S1, Supporting Information). Ref-
ening exercises, stretches, and functional exercises erence lists of all retrieved papers were hand searched
(Pareyson and Marchesi, 2009). Low levels of physical for additional relevant manuscripts.
activity are known to increase body weight, muscle
loss, musculoskeletal pain, and reduce functional Eligibility criteria
capacity (Tremblay et al., 2010). This is of major con- Studies of human participants of any sample size
cern for people with CMT and other neuromuscular were eligible, and there were no age or language
disorders because it can increase the risk factors restrictions. Participants should have a clinical or genet-
for co-morbidities such as metabolic syndrome and ically confirmed diagnosis of CMT. Studies must have
cardiovascular disease (Aitkens et al., 2005). However, investigated the effects of exposure to any form of
participating in exercise can be challenging for people exercise to be eligible for inclusion. All study designs
with CMT due to the increased energy demands with were included, with the exception of narrative and
activities such as walking, possibly due to abnormal systematic reviews. Although reviews were excluded,
gait mechanics (Menotti et al., 2011). There are con- the reference lists from these systematic reviews
cerns that exercise causes overwork weakness in were examined for any additional relevant references.
people with neuromuscular disease and the existence Unpublished data and data from studies with no
of overwork weakness in people with CMT has been full-length publication were excluded.
the subject of ongoing debate because of contradic- Studies could be included if they used validated
tory results (Vinci et al., 2003; van Pomeren et al., outcome measures to determine the effect of exercise
2009; Videler et al., 2010; Arthur-Farraj et al., 2012; exposure at baseline and during follow-up or on com-
Piscosquito et al., 2014). pletion of the exercise program. Outcome measures
To date, there are no uniform guidelines on could include: muscle strength, aerobic capacity, bal-
exercise for people with CMT or other neuromuscular ance, psychological status or quality of life, physiolog-
disorders (Cup et al., 2007). The American College ical adaptations (e.g., muscle fibre increases, changes
of Sports Medicine (ACSM) has formulated minimal in heart rate response during submaximal exercise), or
requirements of exercise prescription for healthy functional ability. Functional ability (either observed or
adults (Garber et al., 2011), but these may not be reported) could include tasks of mobility such as walk-
optimal or safe for people with CMT (Ramdharry et al., ing, stair climbing, sit-to-stand, lift, and reach, or inde-
2014). Several systematic reviews have synthesised pendence in activities of daily living such as washing,
the available evidence of exercise in general neuro- dressing, and preparing food.
muscular disorders, reporting numerous benefits (Neill Safety of the intervention could be assessed
et al., 2006; Cup et al., 2007; Hall et al., 2008; Dibble using outcomes such as: (1) disease deterioration
et al., 2009; Clauw, 2014). Only one recent review for example, an increase in disability or decrease of
(Anziska and Sternberg, 2013) provided an overview strength and/or function; (2) development or increase
of exercise studies conducted in CMT, however, no in pain sufficient to require the use, or increased use,
methodology for the selection of studies was included, of analgesics.
nor were the studies critiqued. Information regarding
safe levels of exercise, such as the optimal amount, Study selection and data extraction
frequency, timing, and type of exercise to optimise Following the deletion of duplicates, titles, and
function may assist in the development of exercise abstracts from the search results were screened by
guidelines. The aim of this paper was to systematically two authors (A. S/.D. H.) using the predetermined
review the available evidence regarding the efficacy eligibility criteria. Full-text articles were retrieved for
and safety of exercise for individuals with CMT. the remaining articles and independently reviewed by
two authors (A. S./D. H.) for inclusion. Data extracted
included study design, sample size, and participant
Materials and Methods characteristics, intervention details and outcome mea-
Identification of studies sures. A description of the intervention details included
in the studies can be found in Table 1.
Search strategy
Electronic database searches were performed in Quality assessment
AMED (via OvidSP), CINAHL (via EBSCO), Cochrane, Methodological quality of the included studies was
EMBASE, MEDLINE (via OvidSP), Scopus, SPORTDis- assessed using a modified Downs and Black (1998)
cus (via EBSCO), and Web of Science from earliest checklist recommended by the Cochrane Handbook
record to February 2015. The search strategy was for Systematic Reviews of Interventions (Reeves et al.,
developed for MEDLINE and modified for use in other 2011). The tool consists of 27 items rated as No = 0,

348
 Table 1. Details of exercise prescription of included studies
Load, number of repetitions, sets and
Study n Study design Intervention progression Frequency and duration
Sman et al.

Strengthening exercises

Burns et al. (2009a) 1 Case report Ankle dorsiflexion exercise 60–80%, 1RM, 2–3 sets, 10 reps (4 s per rep), 1-min 3×/week for 12 week
rest between sets
Chetlin et al. (2004a; 20 Pre-post test Exercises: (1) elbow flexion, (2) 40–50% of MVIC (knee flexion/extension), 20–30% 3×/week for 12 week
2004b) elbow extension, (3) knee MVIC (elbow flexion/extension), 100% of MVIC hand
flexion, (4) knee extension, (5) grip, 3 sets, 4–10 reps for exercise 1–4, 3 sets, 4
hand grip reps for exercise 5, 1-min rest between sets
Lindeman et al. 62 (CMT Randomised controlled Exercises: (1) knee extension, (2) Week: 1–8: 60% 1RM, 3 sets, 25 reps; Week: 9–16: 70% 3×/week for 24 week, 30 min
(1995; 1999) n = 29) trial knee flexion, (3) hip extension, 1RM, 3 sets, 15 reps; Week: 18–24: 80% 1RM, 3 sets,
(4) hip abduction 10 reps, 1-min rest between sets
Ramdharry et al. 32 Randomised controlled Hip flexor exercise 40% of MVIC to 60% of MVIC, 2 sets, 8–12 reps 4×/week for 16 week
(2014) trial (pilot)
Smith et al. (2006) 18 Pre-post test Exercises: (1) elbow flexion, (2) 40–50% of MVIC (knee flexion/extension), 20–30% 3×/week for 12 week
elbow extension, (3) knee MVIC (elbow flexion/extension), 100% of MVIC hand
flexion, (4) knee extension, (5) grip, 3 sets, 4–10 reps for exercise 1–4, 3 sets, 4
hand grip reps for exercise 5, 1-min rest between sets

Aerobic exercises

El Mhandi et al. 8 Pre-post test Cycling 5 min warm up, 6 × 5-min bouts (4 min at 40% P max 3×/week for 24 week, 45 min
(2008) and 1 min at 80% P max ), 10 min cool down at
self-selected power output (i.e. 50–60 W),
progressive increase (10 W increments) in exercise

349
intensity at 80% P max to maintain heart rates within
70–90% MHR
Florence and 12 (CMT Pre-post test Cycling 70% of VO2max , 6 × 5 min bouts, 2 min rest between 3×/week for 12 week, 45 min
Hagberg (1984) n = 2) each bout

Combination programs

Maggi et al. (2011) 8 Pre-post test Exercise program of: (1) (1) Treadmill (30 min) – walking at 40% maximal load 2×/week for 8 week, 90 min
treadmill, (2) stretching, (3) reach from initial test, with 10% increases in
respiratory rehabilitation, (4) subsequent sessions up to attainment of 70%; (2)
proprioceptive exercises respiratory rehabilitation (25 min) – positive
expiratory pressure-bottle and the expiration with
glottis open in lateral position; (3) proprioceptive
exercise (25 min) – Perfetti method. Balance training
exercise carried on by calculating bars of increasing
difficulties in instruments used and in tasks required
Matjacic and Zupan 16 Randomised controlled Exercise program of: (1) passive (1) Passive stretching (10 min); (2) muscle 6×/week for 2 week, 40 min
(2006) trial stretching, (2) muscle strengthening program (10 min); (3) balance training
strengthening, (3) dynamic during standing (20 min). Transfer a ball from one
balance training during hand to the other positioned in the height of the
standing and stepping using a eyes and the knees; turning the body to the left and
mechanical apparatus. Control right with gaze following the direction of turning;
group underwent the same controlled weight transfer during parallel and
program, however, balance tandem stance, simulating push-off, and weight
training was managed by a acceptance while standing in tandem stance
physiotherapist instead of the
apparatus

1RM, one repetition maximum; MHR, maximum heart rate; MVIC, maximal voluntary isometric contraction; P max , power output maximum; VO2max , maximal oxygen uptake; rep/s, repetition/repetitions.
Journal of the Peripheral Nervous System 20:347–362 (2015)
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

Unable to determine = 0, and Yes = 1 and includes cri-

96.3
59.3
59.3
70.4

44.4
55.6

63.0
66.7
77.8
%
teria such as: clear description of the aims, interven-
tions, outcome measurements and participants, rep-

Score

26/27

21/27
16/27
16/27

12/27
15/27

17/27
18/27
19/27
Total
resentativeness of participant groups, appropriateness
of statistical analyses, and correct reporting. The final

1¶

1¶
27

0
0

0
0
0

1
0
item (number 27) relating to statistical power was mod-

0||

0||
1¶
26
ified (from the original score of 0–5 to No = 0, Unable

0
0

1
1
to determine = 0, and Yes = 1) to be consistent with

1¶
25

0
0
0

0
0
0

1
0
Confounding§
the scoring used for the other items (Appendix S2).

0||

0||
1¶

1¶
1¶
24
Some items of the checklist, such as participant blind-

0
0

0
ing, did not apply to all study types. It seemed inap-

1¶
23

1
0
0

1
0
1

1
1
propriate to penalise a study for not meeting a criterion

0||

0||

0||

0||
0||
1¶

1¶
22
which was not applicable to their study design, hence

1
Internal validity
in these cases the maximum score for the question

0||
0||
1¶

1¶
1¶
21

1
1
1
was awarded (MacLehose et al., 2000). A percentage

20
score was generated to facilitate comparison of quality

1
1
1
1

1
1
1

1
1

*Reporting category includes items such as, study aims, reported outcomes, patient characteristics, confounders, adverse events, and loss to follow-up.
between studies.

0||

0||
19

1
1
1

1
1
The details of the quality assessment can be found

Table 2. Results of the modified Downs and Black methodological quality assessment (Downs and Black, 1998)

1¶
18
in Table 2. Two reviewers (A. S./D. H.) independently

1
1
1

1
1
1

1
1
Bias‡
assessed the eligibility of studies, extracted data, and

1¶

1¶
17

1
1

1
1
1

1
1
assessed methodological quality. Discrepancies were

1¶

1¶
16
solved using consensus. Where no consensus could

1
0

0
0
0

1
1
be reached, a third author (J. B.) assessed the quality

0||
0||
1¶
15

0
0
0

1
0
1
to make a final decision.

0||
1¶

1¶
14

1
0

0
0

0
1
13

1
1
1
1

1
1
0

1
1
validity†
External

Results
0||
0||

0||

0||
1¶

1¶
12

1
Description of included studies

§Internal validity – confounding includes items such as study selection, randomisation, and study power.
0||
0||

0||
1¶

1¶
11

0
0
The initial electronic database search resulted in
1¶

1¶
a total of 13,682 articles, leaving 13,301 articles after
10

0
1

1
1
1

1
1
the removal of duplicates. No additional articles were
0||
0||
1¶
09

1
0
0

0
1
identified following a hand search of reference lists.
Upon completion of the title and abstract screening,
08

1
1
1
0

1
0
0

0
0

‡Internal validity – bias includes items such as blinding, follow-up, and compliance.

¶Criteria did not apply to study type, therefore scored as 1 (MacLehose et al., 2000).
35 were selected for possible inclusion in the review
1¶

1¶
07

1
1

1
1
1

1
1

and full-text articles were retrieved. Following the

Reporting*

screening of the full text, 11 articles including nine

06

1
1
1
1

1
1
1

1
1

†External validity includes questions regarding to the study population.

unique studies met the inclusion criteria and were
05

0
0
0
0

1
0
0

1
0

included in the review (Florence and Hagberg, 1984;

04

Lindeman et al., 1995; 1999; Chetlin et al., 2004a;

1
1
1
1

1
1
1

1
1

2004b; Matjacic and Zupan, 2006; Smith et al., 2006; El

03

1
1
1
1

1
1
0

0
1

Mhandi et al., 2008; Burns et al., 2009a; Maggi et al.,

02

2011; Ramdharry et al., 2014). A flow diagram of the

1
1
1
1

1
1
1

1
1

search history and selection process is presented in

01

1
1
1
1

1
1
1

1
1

Figure 1.
2004a; 2004b

Of the nine unique studies included, three studies

1995, 1999

were randomised controlled trials and were described

1, criteria met; 0, criteria not met.
2009

2008

2006

2006
1984

2014
Year

2011

||Unable to determine, scored 0.

in four articles (Lindeman et al., 1995; 1999; Mat-

jacic and Zupan, 2006; Ramdharry et al., 2014), five
studies were quasi-experimental (i.e., pre-post testing)
Ramdharry et al.

and were described in six articles (Florence and Hag-

El Mhandi et al.

Lindeman et al.
Florence and

Matjacic and
Chetlin et al.

berg, 1984; Chetlin et al., 2004a; 2004b; Smith et al.,

Maggi et al.

Smith et al.
Burns et al.

Hagberg

2006; El Mhandi et al., 2008; Maggi et al., 2011), and

Zupan
Author

one study was a case report (Burns et al., 2009a).

Exercise interventions involved resistance training in

350
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

Identification
Records identified through database Additional records identified
searching through other sources
(n=13,682) (n=0)

Records after duplicates removed

(n=13,301 )
Screening

Records title/abstract
screened Records excluded
(n=13,301 ) (n=13,266)
Eligibility

Full-text articles assessed Full-text articles excluded

for eligibility (n=24)
(n=35) • Review paper (n=12)
• Letter to editor (n=1)
• Abstract (n=3)
• No confirmed CMT (n=1)
• Other (n=7)
Include

Studies included in review

(n=9, reported in 11
articles)

Figure 1. Flow diagram.

five studies (Lindeman et al., 1995; 1999; Chetlin there were 52% male and 48% female participants
et al., 2004a; 2004b; Smith et al., 2006; Burns et al., (Florence and Hagberg, 1984; Lindeman et al., 1995;
2009a; Ramdharry et al., 2014), aerobic training in 1999; Chetlin et al., 2004a; 2004b; Matjacic and Zupan,
two studies (Florence and Hagberg, 1984; El Mhandi 2006; El Mhandi et al., 2008; Burns et al., 2009a;
et al., 2008), and a combination program (involving Maggi et al., 2011; Ramdharry et al., 2014). The most
a combination of either aerobic exercise, resistance common type of CMT was CMT1A (60 participants).
exercise, stretching, or balance exercises) in two stud- The CMT type for three participants was unknown;
ies (Matjacic and Zupan, 2006; Maggi et al., 2011) while the CMT type for two participants was not
(Table 1). Resistance training interventions ranged reported (Table 3).
from 12 to 24 weeks in duration and were con- Eight articles described the effects of exercise
ducted 3–4 days/week. Intensities ranged between on muscle strength from six studies (Lindeman et al.,
20%–100% of the maximum voluntary isometric con- 1995; 1999; Chetlin et al., 2004a; 2004b; El Mhandi
traction (MVIC) and 60%–80% of the one repeti- et al., 2008; Burns et al., 2009a; Maggi et al., 2011;
tion maximum (1RM), targeting exercises for both the Ramdharry et al., 2014), nine articles examined func-
upper and lower body. The aerobic training interven- tional activities from five studies (Lindeman et al.,
tions ranged from 12 to 24 weeks in duration, on 1995; Chetlin et al., 2004a; 2004b; Matjacic and Zupan,
3 days/week at intensities between ∼70% and 90% 2006; Smith et al., 2006; El Mhandi et al., 2008;
maximum heart rate (MHR). Burns et al., 2009a; Maggi et al., 2011; Ramdharry
Out of the nine studies included in this review, a et al., 2014), and nine articles reported other physi-
total of 134 participants with CMT with an average ological outcomes from seven studies (Florence and
age of 38 years participated. In the eight studies Hagberg, 1984; Lindeman et al., 1995; 1999; Chetlin
where gender distribution was reported in the paper, et al., 2004a; 2004b; Smith et al., 2006; El Mhandi

351
 Sman et al.

Table 3. Characteristics of the CMT participants in the included studies

CMT Male Female

Reference subjects (n) (%) (%) CMT type Age (years) Height (cm) Weight (kg)

Burns et al. (2009a) 1 0 100 Unknown 15 170.0 64.0

Chetlin et al. (2004a; 20 45 55 CMT1A F: 44.0 ± 7.0 F: 164.0 ± 10.0 F: 84.0 ± 25.0
2004b)
CMT2 (n = 2) M: 46.0 ± 11.0 M: 176.0 ± 10.0 M: 91.0 ± 22.0
El Mhandi et al. (2008) 8 100 0 CMT1A (n = 4) CMT1A: 32.0 ± 6.0 175.0 ± 6.0 69.6 ± 12.1
CMT2 (n = 4) CMT2: 34.0 ± 11.0
Florence and Hagberg 2 87.5 12.5 NR 25.5 ± 9.2 NR 53.8 ± 11.0
(1984)
Lindeman et al. (1995; 29 44.80 55.20 CMT1 (n = 21) Control: 38.0 ± 11.0 NR NR
1999)
CMT2 (n = 6) Exercise: 35.0 ± 10.0
Unknown (n = 2)

352
Maggi et al. (2011) 8 37.5 62.5 CMT1A (n = 4); CMT1 (n = 2); 49.3 ± 17.4 167.0 ± 10.0 67.7 ± 14.6
CMTX (n = 1); CMT1B (n = 1)
Matjacic and Zupan (2006) 16 44 56 CMT1 Control: 37.6 ± 13.5 NR NR
Exercise: 38.5 ± 13.7
Ramdharry et al. (2014) 32 53.1 46.9 Group A: CMT1 (n = 12, 11 Group A: 42±15 NR NR
CMT1A); CMT2 (n = 5); CMT Group B: 46±15
intermediate (n = 1)
Group B: CMT1A (n = 5); CMT 2
(n = 3); CMT X (n = 4); CMT
intermediate (n = 1); HSN-1
(n = 1)
Smith et al. (2006) 18 NR NR CMT1A PRT + CR: 43.0 ± 8.0 PRT + CR: 169.0 ± 10.0 PRT + CR: 94.1 ± 25.0
CMT 1A PRT: 46.0 ± 10.0 PRT: 170.0 ± 10.0 PRT: 83.4 ± 21.2

CR, creatine; CMT, Charcot-Marie-Tooth disease; F, female; M, male; NR, not reported; PRT, progressive resistance training.
Journal of the Peripheral Nervous System 20:347–362 (2015)
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

et al., 2008; Maggi et al., 2011; Ramdharry et al., 2014). (isometric) contraction of the left hip flexors (p < 0.05)
Four out of the five studies reporting on adverse (Ramdharry et al., 2014), isokinetic knee torque exten-
events described measures to assess adverse events sion (p = 0.01) (Lindeman et al., 1995), and maximal
which included completing diaries and/or question- voluntary contraction for knee extension (p = 0.03) (Lin-
naires where the subjects would rate perceived exer- deman et al., 1999). The aerobic training intervention
tion, soreness, and fatigue experienced immediately study reported a significant improvement in isokinetic
after training and reported that no adverse events had knee extension (p = 0.03) and isometric knee flexion
been observed (Lindeman et al., 1995; Chetlin et al., (p = 0.003) following 24 weeks of training (El Mhandi
2004a; 2004b; El Mhandi et al., 2008; Burns et al., et al., 2008). Eight weeks of a combined exercise
2009a; Ramdharry et al., 2014). Compliance (i.e., num- intervention (Maggi et al., 2011) showed no significant
ber of completed sessions throughout the interven- improvements in strength measures using the Medical
tion) was high (>80%) in the five studies where this Research Council (MRC) scale.
was assessed (Lindeman et al., 1995; Chetlin et al.,
2004a; 2004b; El Mhandi et al., 2008; Burns et al., Functional activities
2009a; Ramdharry et al., 2014). Functional activities were described in nine arti-
cles of eight studies (Lindeman et al., 1995; Chetlin
Quality assessment et al., 2004a; 2004b; Matjacic and Zupan, 2006; Smith
The majority of the nine studies in this review were et al., 2006; El Mhandi et al., 2008; Burns et al., 2009a;
of moderate quality with studies on average meeting Maggi et al., 2011; Ramdharry et al., 2014) (Table 5).
66% of the criteria applicable to study type. One study All studies found positive changes in one or more
(Maggi et al., 2011) met less than 50% of the crite- of the functional outcome measurements, with six
ria, six studies (Florence and Hagberg, 1984; Chetlin studies finding significant changes (Lindeman et al.,
et al., 2004a; 2004b; Matjacic and Zupan, 2006; Smith 1995; Chetlin et al., 2004a; Matjacic and Zupan, 2006;
et al., 2006; El Mhandi et al., 2008; Ramdharry et al., Smith et al., 2006; El Mhandi et al., 2008; Maggi et al.,
2014) met between 51% and 74%, and two studies 2011). In one study, (Smith et al., 2006) the time to
met between 75% and 100% (Lindeman et al., 1995; rise from a chair significantly improved for both the
1999; Burns et al., 2009a) of the criteria (Table 2). treatment and control groups (both p < 0.05). A sim-
Seven out of the nine included studies (Florence and ilar study (Chetlin et al., 2004a; 2004b) also found
Hagberg, 1984; Chetlin et al., 2004a; 2004b; Matjacic significant improvements in the chair rise, supine
and Zupan, 2006; Smith et al., 2006; El Mhandi et al., rise, and stair climb (all p < 0.001) but only when
2008; Burns et al., 2009a; Maggi et al., 2011) did not the data for the groups was pooled. This study
address confounders (Q5). Other criteria that were also reported moderate inverse correlations between
commonly unfulfilled were loss to follow-up (Q9), rep- chair and supine rise times and strength normalised
resentation of the entire population (Q11, 12), blinding for lean mass in the lower (r = −0.56, p < 0.02) and
of the intervention (Q14), recruitment of subjects in upper (r = −0.66, p < 0.01) body muscles (Chetlin et al.,
different groups over the same period of time (Q22), 2004a). Lindeman et al. (1995) reported an improve-
adequate adjustment for confounders in the analyses ment in walking speed over 6 m (comfortably) after a
(Q25), and sufficient power to detect a clinically impor- 24-week resistance training program. This same study
tant effect (Q27). also reported that 46% of participants in the resis-
tance training group, and none in the control group
Muscle strength (who continued with normal everyday activities), doc-
Six studies reported changes in strength (Linde- umented an increased ability to perform activities
man et al., 1995; 1999; Chetlin et al., 2004a; 2004b; and 93% in the intervention group believed they had
El Mhandi et al., 2008; Burns et al., 2009a; Maggi derived benefit from the exercise, based on responses
et al., 2011; Ramdharry et al., 2014) (Table 4). Sig- from the WOMAC (Western Ontario and McMas-
nificant improvements in strength following a resis- ter University Osteoarthritis Index) and PET (Problem
tance training intervention were found for only 7 Elicitation Technique) Questionnaire (Lindeman et al.,
out of 70 strength measures (10%), reported in 1995).
four articles of three studies (Lindeman et al., 1995; Numerous improvements in functional
1999; Chetlin et al., 2004b; Ramdharry et al., 2014). time-scored activities were found following one of the
Improvements in four of these strength measures two aerobic training interventions (El Mhandi et al.,
(left knee extension, right knee flexion, left hand-grip, 2008). These included: descending stair (p = 0.002),
right hand-grip; p < 0.05) were reported in the study ascending stair (p = 0.005), rising from supine posi-
by Chetlin et al. (2004b). Significant improvements tion (p = 0.007), walking 6 m (p = 0.003), and walking
for strength were also found for maximal voluntary 50 m (p = 0.006). Rising from a chair (p = 0.08) did not

353
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

Table 4. Outcomes of muscle strength described in the included articles

Pre, Post, Change,

Reference Mode Outcome measure mean mean mean

Strengthening exercises
Burns et al. PRT (n = 1) Muscle strength
(2009a) Left dorsiflexion (N) 27.3 42.7 15.4
Right dorsiflexion (N) 23.3 40.0 16.7
Left 1RM plantarflexion (N) 267.0 319.3 52.3
Right 1RM plantarflexion (N) 286.0 329.3 43.3
Left 1RM inversion (N) 92.3 136.3 44.0
Right 1RM inversion (N) 138.3 143.3 5.0
Left 1RM eversion (N) 63.0 62.7 −0.3
Right 1RM eversion (N) 74.7 59.7 −15
Chetlin, et al. PRT; females; (n = 11) Left elbow extension (kg) 8.7 ± 1.5 9.7 ± 1.5 1.0
(2004a) Right elbow extension (kg) 8.7 ± 2.1 9.3 ± 1.1 0.6
Left elbow flexion (kg) 16.1 ± 3.7 16.3 ± 2.9 0.2
Right elbow flexion (kg) 15.0 ± 4.2 15.0 ± 2.7 0.0
Left knee extension (kg) 21.6 ± 8.8 26.0 ± 5.2 4.4
Right knee extension (kg) 22.9 ± 9.6 25.0 ± 8.0 2.1
Left knee flexion (kg) 12.6 ± 5.6 14.0 ± 2.4 1.4
Right knee flexion (kg) 12.0 ± 5.7 15.5 ± 3.2 3.5
Left ankle dorsiflexion (kg) 3.6 ± 3.0 3.7 ± 3.6 0.1
Right ankle dorsiflexion (kg) 4.0 ± 5.4 4.1 ± 3.0 0.1
PRT; males (n = 9) Left elbow extension (kg) 13.4 ± 6.4 12.9 ± 4.9 −0.5
Right elbow extension (kg) 13.2 ± 4.4 13.9 ± 4.0 0.7
Left elbow flexion (kg) 20.3 ± 3.7 0.5
Right elbow flexion (kg) 20.7 ± 5.8 0.4
Left knee extension (kg) 35.8 ± 9.8 3.3
Right knee extension (kg) 34.0 ± 12.6 2.0
Left knee flexion (kg) 14.7 ± 6.7 16.6 ± 6.7 1.9
Right knee flexion (kg) 15.2 ± 7.4 17.1 ± 5.9 1.9
Left ankle dorsiflexion (kg) 4.1 ± 3.0 3.6 ± 4.4 −0.5
Right ankle dorsiflexion (kg) 3.5 ± 4.3 3.5 ± 4.3 0.0
Chetlin, et al. PRT + CR (n = 10) Maximal voluntary isometric testing
(2004b) Left elbow extension (kg) 11.8 ± 6.5 11.8 ± 4.6 0.0
Right elbow extension (kg) 11.9 ± 4.9 12.2 ± 4.5 0.3
Left elbow flexion (kg) 18.3 ± 4.7 19.0 ± 3.9 0.7
Right elbow flexion (kg) 17.7 ± 5.3 18.1 ± 5.6 0.4
Left knee extension (kg) 30.3 ± 10.9 33.6 ± 9.7 3.3
Right knee extension (kg) 31.9 ± 13.0 34.3 ± 10.8 2.4
Left knee flexion (kg) 15.2 ± 5.2 17.3 ± 4.3 2.1
Right knee flexion (kg) 15.5 ± 6.6 18.0 ± 3.8 2.5
Left hand-grip (kg) 21.9 ± 5.4 23.2 ± 7.3 1.3
Right hand-grip (kg) 23.9 ± 7.3 25.8 ± 7.1 1.9
PRT (n = 10) Left elbow extension (kg) 10.5 ± 6.5 11.2 ± 2.1 0.7
Right elbow extension (kg) 9.7 ± 2.8 10.9 ± 2.3 1.2
Left elbow flexion (kg) 17.1 ± 4.6 17.1 ± 3.9 0.0
Right elbow flexion (kg) 17.0 ± 6.8 16.8 ± 5.4 −0.2
Left knee extension (kg) 24.1 ± 9.0 27.4 ± 7.8 3.3
Right knee extension (kg) 23.2 ± 10.0 25.6 ± 8.9 2.4
Left knee flexion (kg) 12.7 ± 7.2 13.7 ± 4.2 1.0
Right knee flexion (kg) 12.1 ± 6.7 14.7 ± 4.8 2.6
Left hand-grip (kg) 20.6 ± 9.0 22.4 ± 7.7 1.8
Right hand-grip (kg) 18.0 ± 9.1 19.6 ± 7.4 1.6
Combined Left elbow extension (kg) 11.2 ± 5.0 11.5 ± 3.7 0.3
(PRT+CR/PRT) (n = 20) Right elbow extension (kg) 10.9 ± 4.2 11.6 ± 3.7 0.7
Left elbow flexion (kg) 17.8 ± 4.5 18.2 ± 3.9 0.4
Right elbow flexion (kg) 17.4 ± 5.8 17.6 ± 5.4 0.2
Left knee extension (kg) 27.5 ± 10.3 30.8 ± 9.2 3.3*
Right knee extension (kg) 28.0 ± 12.2 30.5 ± 10.7 2.5

354
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

Table 4. Continued

Pre, Post, Change,

Reference Mode Outcome measure mean mean mean

Left knee flexion (kg) 14.1 ± 6.1 15.7 ± 4.5 1.6

Right knee flexion (kg) 14.0 ± 6.7 16.5 ± 4.5 2.5*
Left hand-grip (kg) 21.3 ± 7.0 22.8 ± 7.2 1.5*
Right hand-grip (kg) 21.3 ± 8.5 23.1 ± 7.7 1.8*
Lindeman, et al. PRT (n = 13) Isokinetic knee torque extension (Nm) 91.0 ± 31.0 12.4 ± 15.4*
(1995) Isokinetic knee torque flexion (Nm) 41.0 ± 16.0 5.0 ± 7.3
Maximal Voluntary Contraction (isometric) Nm 16.6 ± 19.7
Endurance test at 80% MVC (s) 1.8 ± 17.4
Control (n = 13) Isokinetic knee torque extension (Nm) −5.3 ± 17.3
Isokinetic knee torque flexion (Nm) −1.1 ± 7.8
Maximal Voluntary Contraction (isometric) Nm 4.0 ± 16.9
Endurance test at 80% MVC (s) 1.5 ± 11.5
Lindeman, et al. PRT (n = 15) Knee extension, MVC (Nm) 110.0 ± 47.0 22.0 ± 25.0*
(1999) Knee extension, endurance (s) 44.0 ± 23.0 0.70 ± 17.3
Control; (n = 13) Knee extension, MVC (Nm) 110.0 ± 47.0 3.0 ± 21.0
Knee extension, endurance (s) 44.0 ± 23.0 −3.3 ± 12.4
Ramdharry, et al. (n = 32) Hip flexor MVC (Nm/kg) left hip 1.14 ± 0.44 1.19 ± 0.45 0.05*
(2014) Hip flexor MVC (Nm/kg) right hip 1.17 ± 0.45 1.21 ± 0.43 0.04
Aerobic exercises
El Mhandi, et al. (n = 8) Isokinetic muscle strength (Nm)
(2008) Isometric knee extension (0∘ .s_1) 185.0 ± 62.0 188.0 ± 69.0 3.0
Isokinetic knee extension (120∘ .s_1) 116.0 ± 41.0 128.0 ± 42.0 12.0*
Isometric knee flexion (0∘ .s_1) 121.0 ± 34.0 126.0 ± 28.0 5.0
Isokinetic knee flexion (120∘ .s_1) 77.0 ± 14.0 87.0 ± 17.0 10.0*
Combination
programs
Maggi, et al. EX (n = 8) MRC plantar flexion right foot 4.6 ± 0.6 4.7 ± 0.6 0.1
(2011) MRC plantar flexion left foot 4.5 ± 0.7 4.5 ± 0.7 0.03
MRC plantar flexion right foot 2.9 ± 1.9 3.1 ± 2.0 0.1
MRC plantar flexion left foot 2.3 ± 2.0 2.5 ± 2.0 0.3

Change scores without complete pre and/or post means is presented for Lindeman et al. (1995; 1999) as the data could not be obtained.
1RM, one repetition maximum; CR, creatine, EX, exercise group, MVC, maximal voluntary contraction, MRC, medical research council scale;
PRT, progressive resistance training group.
*p < 0.05.

significantly improve in this study (El Mhandi et al., 1995; Chetlin et al., 2004b; Smith et al., 2006; El
2008). Mhandi et al., 2008) (Table 6). Significant increases in
Two studies (Matjacic and Zupan, 2006; Maggi myosin heavy chain (MHC) 1 were reported follow-
et al., 2011) that investigated interventions using a ing resistance training (p = 0.045) (Smith et al., 2006).
combination program also reported several improve- Another study found a significant increase in the size
ments. Maggi et al. (2011) reported a significant of type 1 muscle fibres (p = 0.03) when combining the
improvement for right ankle range of motion (p < 0.03), treatment group (resistance training and supplement)
left ankle range of motion (p < 0.03), and time to walk with the control group (resistance training) (Chetlin
6 m (p < 0.02). Matjacic and Zupan (2006) reported et al., 2004b). Chetlin et al. (2004b) also reported sev-
a significant improvement for the timed up and go eral significant correlations between the size of type 2×
test (p = 0.04), 10-m walk (p = 0.01), and Berg Balance muscle fibre and measures of strength (r = 0.70–0.84,
(p = 0.01), however, the control group also showed p < 0.05). Following resistance training, Lindeman et al.
a significant improvement for the Berg Balance (1995 found that myoglobin increased to a greater
(p = 0.01). extent in participants with CMT compared to partici-
pants with myotonic dystrophy (p = 0.02).
Physiological measurements A significant improvement in aerobic capacity
Nine articles from seven studies reported phys- (p = 0.02) following an aerobic training intervention
iological changes after exercise (Florence and Hag- was reported in the study conducted by El Mhandi
berg, 1984; Lindeman et al., 1995; 1999; Chetlin et al., et al. (2008). This same study documented that par-
2004a; 2004b; Smith et al., 2006; El Mhandi et al., ticipants reported significantly less pain and fatigue
2008; Maggi et al., 2011; Ramdharry et al., 2014), with after a training session (acute) at the end of the
four studies finding significant change (Lindeman et al., intervention compared to the beginning (p < 0.001).

355
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

Table 5. Outcomes of functional activities described in the included articles

Pre, Post, Change,

Reference Mode Outcome measure mean mean mean

Strengthening exercises
Burns et al. (2009a) PRT (n = 1) Balance, BOT score 20.0 18.0 −2.0
(0–37 points)
Power, long jump (cm) 68.0 79.0 11.0
Endurance, 6 min walk 415.0 410.0 −5.0
test (m)
Walking ability
Speed (cm/s) 105.6 112.8 7.2
Cadence (steps/min) 107.9 111.4 3.5
Step time left (s) 0.6 0.5 0.1
Step time right (s) 0.6 0.5 0.1
Step length left (cm) 60.8 62.1 1.3
Step length right (cm) 56.6 59.6 3.0
Stride length left (cm) 118.4 121.8 3.4
Stride length right (cm) 117.3 123.5 6.2
Chetlin et al. (2004a) PRT females (n = 11) Lift and reach (s) 20.9 ± 5.0 17.7 ± 3.5 −3.2
PRT males (n = 9) Lift and reach (s) 20.2 ± 3.1 17.6 ± 3.4 −2.6
PRT females (n = 11) Chair raise (s) 1.3 ± 0.2 1.0 ± 0.3 −0.3
PRT males (n = 9) Chair raise (s) 1.3 ± 0.2 1.1 ± 0.2 −0.2
PRT females (n = 11) Supine rise (s) 2.1 ± 1.2 1.6 ± 0.7 −0.5
PRT males (n = 9) Supine rise (s) 2.1 ± 0.6 1.5 ± 0.5 −0.6
PRT females (n = 11) Stair climb (s) 15.1 ± 10.2 12.3 ± 8.8 −2.8
PRT males (n = 9) Stair climb (s) 13.6 ± 12.6 11.7 ± 9.8 −1.9
Chetlin et al. (2004b) PRT + CR (n = 10) Lift and reach (s) 18.3 ± 1.6 16.4 ± 1.0 −1.9
PRT (n = 10) Lift and reach (s) 22.9 ± 4.8 18.9 ± 4.0 −2.0
Combined (n = 20) Lift and reach (s) 20.6 ± 4.2 17.6 ± 3.3 −3.0*
PRT + CR (n = 10) Chair raise (s) 1.2 ± 0.2 1.0 ± 0.2 −0.2
PRT (n = 10) Chair raise (s) 1.4 ± 0.3 1.1 ± 0.3 −0.3
Combined (n = 20) Chair raise (s) 1.3 ± 0.3 1.0 ± 0.2 −0.3*
PRT + CR (n = 10) Supine rise (s) 1.8 ± 0.3 1.4 ± 0.3 −0.4
PRT (n = 10) Supine rise (s) 2.4 ± 1.3 1.7 ± 0.8 −0.7
Combined (n = 20) Supine rise (s) 2.1 ± 1.0 1.6 ± 0.6 −0.5*
PRT+CR (n = 10) Stair climb (s) 10.7 ± 7.2 8.9 ± 5.0 −1.8
PRT (n = 10) Stair climb (s) 18.1 ± 13.3 15.2 ± 11.2 −2.9
Combined (n = 20) Stair climb (s) 14.4 ± 11.1 12.1 ± 9.0 −2.3*
Lindeman et al. (1995) PRT (n = 13) Descending stairs (s) 0.7 ± 1.7
Climbing stairs (s) 0.7 ± 1.4
Standing up from a 0.2 ± 0.5
chair (s)
Standing up from lying 0.3 ± 0.6
supine (s)
Walking 6 m 1.0 ± 0.5*
(comfortably) (s)
Walking 50 m (fast) (s) 2.2 ± 2.8
Control (n = 13) Descending stairs (s) −0.09 ± 1.3
Climbing stairs (s) −0.01 ± 1.2
Standing up from a 0.05 ± 0.3
chair (s)
Standing up from lying 0.1 ± 0.5
supine (s)
Walking 6 m 0.3 ± 0.7
(comfortably) (s)
Walking 50 m (fast) (s) 0.3 ± 2.9
Ramdharry et al. (2014) Training (n = 32) 6-min timed walk (m) 342 ± −67 350 ± −68 8
Walking speed 10 m 1.18 ± 0.24 1.19 ± 0.24 0.01
(self-selected) (m/s)
Walking speed 10 m 1.48 ± 0.33 1.47 ± 0.36 −0.01
(maximum) (m/s)
CMTES (median) 10.5 11 0.5
Walk-12 (median) 34 33 −1
Phone FITT FDI 55.4 45.7 −9.7
(median)

356
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

Table 5. Continued

Pre, Post, Change,

Reference Mode Outcome measure mean mean mean

Control (n = 32) 6-min timed walk (m) 352 ± −73 347 ± −75 −5
Walking speed 10 m 1.18 ± 0.21 1.19 ± 0.23 0.01
(self-selected) (m/s)
Walking speed 10 m 1.48 ± 0.32 1.44 ± 0.32 −0.04
(maximum) (m/s)
CMTES (median) 10 12 2
Walk-12 (median) 36.5 36 −0.5
Phone FITT FDI 49.4 48.5 −0.9
(median)
Smith et al. (2006) PRT+CR (n = 10) Chair rise-time (s) 1.4 ± 0.0 1.1 ± 0.0 0.2*
PRT (n = 8) Chair rise-time (s) 1.1 ± 0.0 1.0 ± 0.0 0.1*
Aerobic exercises
El Mhandi et al. (2008) (n = 8) Functional time-scored
activity (s)
Descending stair 5.5 ± 0.9 4.1 ± 0.7 −1.4*
Ascending stair 5.8 ± 0.4 4.7 ± 0.5 −1.1*
Rising from chair 1.7 ± 0.6 1.2 ± 0.5 −0.5
Rising from supine 3.1 ± 0.5 2.1 ± 0.7 −1.0*
position
Walking 6 m 5.4 ± 0.6 4.5 ± 0.6 −1.0*
Walking 50 m 28.5 ± 4.2 26.8 ± 2.9 −1.7*
Combination programs
Maggi et al. (2011) EX (n = 8) Dorsal flexion right −2.9 ± 4.5 −0.3 ± 5.7 2.6*
angle ROM (deg)
Dorsal flexion left angle −2.1 ± 6.0 0.5 ± 5.8 2.6*
ROM (deg)
Tinetti Scale 11.5 ± 3.6 13.4 ± 3.8 1.9
Physical performance 7.5 ± 3.3 9.5 ± 2.8 2.0
battery
CMTNS 13.5 ± 3.1 13.5 ± 3.1 0
Time to walk 6 m (s) 9.1 ± 2.4 7.4 ± 1.5 1.7*
Peak of slope 12.9 ± 3.8 14.5 ± 2.3 1.6
Peak of speed 5.0 ± 1.7 5.7 ± 1.6 0.7
Matjacic and Zupan (2006) EX (n = 8) Berg Balance (0–56 43.5 ± 10.4 46.9 ± 8.7 3.4 ± 2.7*
scale)
Control (n = 8) Berg Balance (0–56 43.0 ± 8.9 45.1 ± 8.5 2.1 ± 1.6*
scale)
EX (n = 8) Up and go (s) 13.6 ± 5.0 12.0 ± 3.9 −1.6 ± 1.6*
Control (n = 8) Up and go (s) 15.1 ± 7.0 13.5 ± 4.9 −1.6 ± 2.8
EX (n = 8) 10 m walk (s) 13.7 ± 5.7 11.5 ± 3.9 −2.1 ± 1.7*
Control (n = 8) 10 m walk (s) 14.2 ± 6.7 13.1 ± 5.1 −1.1 ± 1.9

Change scores without complete pre and/or post means is presented for Lindeman et al. (1995) as the data could not be obtained.
CR, creatine; EX, exercise group; ROM, range of motion; CMTNS, Charcot-Marie-Tooth Neuropathy Score; CMTES, Charcot-Marie-Tooth
Examination Score; Phone FITT FDI, frequency, duration, and intensity score for the Phone-FITT scale; PRT, progressive resistance training
group; m, metre; s, seconds.
*p < 0.05.

In the study conducted by Florence and Hagberg neuromuscular diseases (∼ −16% to 21%). Maggi
(1984), two out of eight participants with neuromus- et al. (2011, Ramdharry et al. (2014) found no significant
cular disease had CMT. Lower and non-significant improvements for any physiological outcomes.
improvement in VO2max was found for the CMT par-
ticipants compared to participants with other neuro- Discussion
muscular diseases (∼7% vs. ∼23%) following 12 week The effects of exercise for people with CMT remain
aerobic exercise intervention Florence and Hagberg, unclear after this systematic review of the published
1984). Additionally, heart rates during cycling at two literature due to the small number of studies, predom-
submaximal workloads (established at baseline and inance of non-robust study designs, and likelihood of
kept the same at post-training) following the exercise Type II errors. Although several significant effects were
intervention increased for participants with CMT (∼ found within the included studies, approximately 60%
14%), and decreased for the participants with other of these were obtained with cohorts of 10 participants

357
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

Table 6. Outcomes physiological measurements described in the included articles

Reference Mode Outcome measure Pre, mean Post, mean Change, mean

Strengthening exercises
Chetlin et al. (2004a) Females PRT (n = 11) FFM (kg) 46.6 ± 8.9 46.3 ± 8.7 −0.3
Males PRT (n = 9) FFM (kg) 56.0 ± 11.4 56.1 ± 11.2 0.1
Females PRT (n = 11) BMI (kg/m2 ) 31.0 ± 8.5
Males PRT (n = 9) BMI (kg/m2 ) 29.1 ± 5.7
Females PRT (n = 11) Percent body fat (%) 37.0 ± 8.5
Males PRT (n = 9) Percent body fat (%) 44.0 ± 14.1
Chetlin et al. (2004b) PRT+CR (n = 10) Type 1 fibre 51.2 ± 12.3 56.6 ± 13.0 5.4
PRT (n = 10) Type 1 fibre 44.6 ± 21.6 55.2 ± 23.7 10.6
Combined (n = 20) Type 1 fibre 48.2 ± 14.2 55.4 ± 14.8 7.2*
PRT+CR (n = 10) Type 2a fibre 48.4 ± 14.3 48.1 ± 7.4 −0.3
PRT (n = 10) Type 2a Fibre 45.6 ± 6.0 51.7 ± 9.5 6.1
Combined (n = 20) Type 2a fibre 47.6 ± 12.0 49.1 ± 7.4 1.5
PRT+CR (n = 10) Type 2x fibre 43.6 ± 12.5 52.3 ± 19.1 8.7
PRT (n = 10) Type 2x fibre 31.0 ± 1.8 43.6 ± 3.5 12.6
Combined (n = 20) Type 2x fibre 40.5 ± 12.1 50.1 ± 16.7 9.6
Lindeman et al. (1995) PRT (n = 15) Serum myoglobin (ng/l) 93.0 ± 51.0/12.0 ± 135.0 113.0 ± 70.0/15.0 ± 96.0 20.0/3.0
initial value/increase
value (due to test)
Control (n = 13) Serum myoglobin (ng/l) 99.0 ± 75.0/7.0 ± 16.0 94.0 ± 54.0/25.0 ± 70.0 −5.0/18.0
initial value/increase
value (due to test)
MyD (n = 31) Serum myoglobin (ng/l) 235.0 ± 125.0/31.0 ± 39.0
initial value/increase
value (due to test)
CMT (n = 28) Serum myoglobin (ng/l) 96.0 ± 62.0/10.0 ± 27.0 p = 0.02*
initial value/increase
value (due to test)
Lindeman et al. (1999) PRT (n = 15) RMS (μv)
VM 172.0 ± 90.0 29.0 ± 59.0
RF 228.0 ± 101.0 51.0 ± 92.0
VL 254.0 ± 124.0 53.0 ± 121.0
BF 78.0 ± 45.0 11.0 ± 45.0
EXT 218.0 ± 96.0 44.0 ± 72.0
TER20-80 (Nm/μv)
VM 0.9 ± 0.6 −0.0 ± 0.0
RF 0.6 ± 0.2 −0.1 ± 0.2
VL 0.5 ± 0.2 −0.1 ± 0.2
BF 2.1 ± 1.3 −0.2 ± 0.4
EXT 0.7 ± 0.3 −0.1 ± 0.2
Slope (Hz/s)
VM −0.1 ± 0.2 0.1 ± 0.2
RF −0.3 ± 0.2 0.1 ± 0.2
VL −0.2 ± 0.2 0.1 ± 0.2
BF −0.1 ± 0.2 0.1 ± 0.2
EXT −0.2 ± 0.2 0.1 ± 0.2
Control (n = 13) RMS (μv)
VM 172.0 ± 90.0 −17.0 ± 78.0
RF 228.0 ± 101.0 −5.0 ± 80.0
VL 254.0 ± 124.0 32.0 ± 72.0
BF 78.0 ± 45.0 −2.0 ± 30.0
EXT 218.0 ± 96.0 3.0 ± 55.0
TER20-80 (Nm/μv )
VM 0.9 ± 0.6 −0.1 ± 0.5
RF 0.6 ± 0.2 −0.0 ± 0.2
VL 0.5 ± 0.2 −0.1 ± 0.1
BF 2.1 ± 1.3 −0.2 ± 0.7
EXT 0.7 ± 0.3 −0.1 ± 0.2
Slope (Hz/s)
VM −0.1 ± 0.2 0.0 ± 0.0
RF −0.3 ± 0.2 −0.0 ± 0.0
VL −0.2 ± 0.2 −0.0 ± 0.0
BF −0.1 ± 0.2 0.1 ± 0.4
EXT −0.2 ± 0.2 −0.0 ± 0.0
Ramdharry et al. (2014) Training (n = 32) Borg Perceived 11 11 0
Exertion scale
(median)
Fatigue Severity Scale 29.5 27 −2.5
(median)
Modified PCI 1.93 ± 0.59 1.89 ± 0.59 −0.4
(beats/min/m)
Control (n = 32) Borg scale (median) 12 12 0
Fatigue Severity Scale 29 31 2
(median)
Modified PCI 1.89 ± 0.51 1.96 ± 0.57 0.7
(beats/min/m)

358
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

Table 6. Continued
Reference Mode Outcome measure Pre, mean Post, mean Change, mean

Smith et al. (2006) PRT+CR (n = 10) MHC 1 (%) VL 42.4 ± 4.6 10.0
PRT (n = 8) MHC 1 (%) VL 46.5 ± 5.1 20.0*
PRT+CR (n = 10) MHC 2a (%) VL 43.1 ± 7.1 46.7 ± 5.0 3.6
PRT (n = 8) MHC 2a (%)VL 54.0 ± 9.5 24.9 ± 6.2 −29.1
PRT+CR (n = 10) MHC 2x VL 9.1 ± 3.2 9.6 ± 5.9 0.5
PRT (n = 8) MHC 2x VL 9.4 ± 3.3 8.8 ± 3.7 −0.6
Aerobic exercises
El Mhandi et al. (2008) (n = 8) Cardiorespiratory cycle 36.4 ± 4.6 39.9 ± 4.7 3.5*
test. VO2 peak
(ml kg−1 min−1 )
P max , W 188.6 ± 38.8 210.1 ± 48.8 21.5
HRmax , beats/min 185 ± 11 189 ± 9 4.0
Lactatemax , mmol/l 10.4 ± 2.0 11.5 ± 1.4 1.1
End 80% MVIC (s) 29.0 ± 7.0 30.0 ± 7.0 1.0
Fatigue test (50% MVIC) −8.9 ± 7.2 −6.9 ± 5.5 2.0
Repetitions to fatigue 25.0 ± 16.0 32.0 ± 14.0 7.0
following fatigue
protocol
Mean blood CK (UI) 197.0 ± 78.0
(rest)
Mean blood CK (UI/L) 277.0 ± 86.0 212.0 ± 98.0 −65.0
(24 h after Ex)
VAS Pain and fatigue less pronounced
at the end of the training session
then in the beginning (p < 0.001).
Last 12 training sessions less
painful and fatiguing compared
to first 12 sessions (p < 0.01)
Florence and Hagberg Other NMD (n = 6) VO2 max 1.90 ± 0.52 2.33 ± 0.19 0.44
(1984) CMT only (n = 2) VO2 max 1.51 ± 1.38 1.61 ± 1.41 0.1 ± 0.03
Control (n = 4) VO2 max 1.73 ± 0.18 1.75 ± 0.15 0
Other NMD (n = 6) HR, at submax −16.0 ± 8.6%
workload #1
CMT only (n = 2) HR, at submax 14.0 ± 8.5%
workload #1
Other NMD (n = 6) HR, at submax −21.2 ± 8.6%
workload #2
CMT only (n = 2) HR, at submax 14.0 ± 1.4%
workload #2
Combination programs
Maggi et al. (2011) EX (n = 8) HRmax 132.6 ± 24.5 132.1 ± 32.4 0.5
Peak of VO2 1.31 ± 0.94 1.2 ± 1.1 −0.1
METS 1.0 ± 0.3 0.8 ± 0.4 −0.2
Peak of METS 5.8 ± 2.7 4.7 ± 3.3 −1.1
Respiratory Borg Scale 5.5 ± 0.9 5.1 ± 0.8 −0.4
Fatigue Borg Scale 7.6 ± 0.7 7.1 ± 0.8 −0.5

Change scores without complete pre and/or post means is presented for Chetlin et al. (2004a), Lindeman et al. (1995), Smith et al. (2006) and Florence and Hagberg
(1984) as the data could not be obtained.
BF, biceps femoris; BMI, body mass index; CK, creatine kinase; CR, creatine; EX, exercise group; EXT, mean of the extensor muscles; FFM, fat-free mass; HR, heart
rate; METS, metabolic equivalents (1 MET, 3.5 ml kg−1 min−1 ); MHC, myosin heavy chain; MVC, maximal voluntary contraction; MVIC, maximal voluntary isometric
contraction; NMD, neuromuscular disease; PCI, physiological cost index; PRT, progressive resistance training group; RF, rectus femoris; RMS, root mean square;
TER20-80 , Torque–EMG ratios at 20%, 40%, 60%, and 80% MVC; VAS, visual analogue scale; VL, Vastus Lateralis; VM, Vastus Medialis; VO2 , oxygen uptake.
*p < 0.05.
†Average pre- and post-scores.

or less. Owing to the low number of randomised con- fibre size. Similarly, aerobic training led to favourable
trolled trials or comparisons with control groups, and changes in some measures of strength and functional
the limited evidence of pre-post differences, the true activities, as well as an increase in aerobic capacity.
efficacy and safety of exercise for people with CMT Combined exercise intervention studies found positive
remains unclear. However, it appears that exercise in changes in ankle flexibility, balance, agility, and mobil-
CMT may be effective in improving some components ity. Compliance to exercise was high and studies that
of health and fitness without harmful effects in the assessed adverse events reported that none occurred
short-term. during the study period.
The majority of studies included in this review The data from this review show that 70 statistical
investigated resistance training interventions. This tests for strength measures were used and only 7
mode of exercise was found to result in positive were significant at p < 0.05. Four of these significant
changes in strength, functional activities, and muscle increases for strength were from the study by Chetlin

359
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

et al. (2004b) of a combined group, where in one group aerobic capacity during maximal exercise. The find-
exercise was combined with creatine. There was a ings may be explained by the fact that only two CMT
trend for non-significant improvements in strength of participants were included. The results of this study
approximately 5% in all studies, but it is unknown should therefore be interpreted with caution until fur-
whether a 5% change would be clinically meaningful ther research studies with larger sample sizes are con-
for people with CMT. ducted.
In studies of neuromuscular disease, recruiting Although the aerobic exercise interventions used
enough participants to sufficiently power a study and by studies in this review were consistent with rec-
see meaningful changes is challenging (Abresch et al., ommendations for improving cardiorespiratory fitness
2012). The 5% improvement in strength from this (Garber et al., 2011), the majority of interventions
review was calculated on averages of data, but in order were below the recommended exercise guidelines for
to detect a 5% change in strength that is statistically weight loss and maintenance of healthy body weight
significant, it would require a very large sample size. (Donnelly et al., 2009). Little data were presented in
For example, a sample size of approximately over 100 the included studies to adequately evaluate the poten-
subjects would be required to detect a 5% improve- tial of exercise to cause muscle damage in CMT.
ment in knee flexion strength based on the data from Although no adverse events were reported in the stud-
Chetlin et al. (2004b). ies included in this review, it would seem unlikely that
Only one of the studies included in this review none would experience some sort of difficulty with
conducted a sample size calculation to ensure that exercise. Only 50% of the included studies used spe-
their study had enough power to detect a clinically cific measures to assess adverse events and details of
important effect (Ramdharry et al., 2014). Several stud- the tools used were not well described. To be able to
ies combined participants from various groups (Chetlin adequately understand how people with CMT respond
et al., 2004a; 2004b) or combined participants with to exercise, more detailed measures and longitudinal
various disorders (Florence and Hagberg, 1984) to studies need to be conducted to help identify difficul-
detect change. Grouping various neuromuscular dis- ties or barriers encountered.
A limitation to this review is that the majority
eases together in particular raises concerns because
of outcomes extracted constitute pre-post changes
potential differences in severity, rate of progression,
within the intervention groups of uncontrolled trials
and disease type could affect exercise response. The
and pre-post differences within groups do not allow for
small changes found in the studies included in the
causal inference. Differences between pre-post results
review may also be because of the age of participants
may reflect a host of non-specific treatment effects and
included in the studies and their stage of disease pro-
within this context non-significant differences do not
gression. Optimal benefits of exercise are likely to be
necessarily mean non-efficacy. Other limitations of the
achieved in the early stages of the disease when mus-
current review include the moderate quality of the stud-
cle is relatively preserved, however, with the exception
ies. Despite our extensive search of the literature, we
of one pilot study involving one participant (Burns et al.,
found only three randomised controlled trials, with the
2009a), all included trials recruited adult participants majority of studies using a quasi-experimental study
with an average age of 38 years across the included design (i.e., without a control group). Finally, the large
studies. Research including younger participants may variability between studies (e.g., interventions, out-
provide valuable additional information in regards to the come measures) made comparison of results difficult
risks and benefits of exercise across the lifespan. and unsuitable for pooling. Therefore, a meta-analysis
Some increase in aerobic capacity was found in was not performed.
CMT participants following aerobic training interven-
tions. However, in one study (Florence and Hagberg,
1984), the increase in VO2 max was less compared to
a group of participants with other neuromuscular dis- Conclusion
eases. In addition, submaximal heart rate responses Although benefits appear to be gained from
at two initial workloads increased for the participants exercise in strength and function in some studies,
with CMT following the exercise intervention. This find- most outcomes reported were not statistically signifi-
ing is unusual because training adaptations following cant. The optimal exercise modality and intensity for
aerobic exercise usually result in a decreased submax- people with CMT, the clinical relevance of the changes
imal heart rate due to an increased stroke volume (Yil- observed and the safety of exercise is still unclear.
maz et al., 2013). Therefore, cardiac output remains the The findings of this review should be met with caution
same when exercising at the initial submaximal work- due to the small number of randomised controlled
load, but leads to a greater cardiac output and thus trials included, which reduces the quality of evidence

360
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

available for this review. Future adequately-powered Florence JM, Hagberg JM (1984). Effect of training on the
randomised controlled trials including detailed adverse exercise responses of neuromuscular disease patients. Med
event reporting are recommended to investigate the Sci Sports Exerc 16:460–465.
Garber CE, Blissmer B, Deschenes MR, Franklin BA, Lamonte
risks and benefits of exercise for people with CMT MJ, Lee IM, Nieman DC, Swain DP (2011). American Col-
across the lifespan. Recruiting large samples in rarer lege of Sports Medicine position stand. Quantity and qual-
diseases, such as CMT, is challenging, therefore ity of exercise for developing and maintaining cardiorespira-
studies including well-described interventions, and tory, musculoskeletal, and neuromotor fitness in apparently
more specifically harmonisation of outcome mea- healthy adults: guidance for prescribing exercise. Med Sci
sures across studies, may improve the quality and Sports Exerc 43:1334–1359.
comparability of the evidence. These trials would ulti- Hall J, Swinkels A, Briddon J, McCabe CS (2008). Does aquatic
exercise relieve pain in adults with neurologic or muscu-
mately facilitate the development of uniform exercise loskeletal disease? A systematic review and meta-analysis
guidelines for people with CMT. of randomized controlled trials. Arch Phys Med Rehabil
89:873–883.
Lindeman E, Leffers P, Spaans F, Drukker J, Reulen J, Kerckhoffs
References M, Koke A (1995). Strength training in patients with myotonic
Abresch RT, Carter GT, Han JJ, McDonald CM (2012). Exercise dystrophy and hereditary motor and sensory neuropathy: a
in neuromuscular diseases. Phys Med Rehabil Clin N Am randomized clinical trial. Arch Phys Med Rehabil 76:612–620.
23:653–673. Lindeman E, Spaans F, Reulen J, Leffers P, Drukker J (1999).
Aitkens S, Kilmer DD, Wright NC, McCrory MA (2005). Metabolic Progressive resistance training in neuromuscular patients.
syndrome in neuromuscular disease. Arch Phys Med Rehabil Effects on force and surface EMG. J Electromyogr Kinesiol
86:1030–1036. 9:379–384.
Anziska Y, Sternberg A (2013). Exercise in neuromuscular dis- MacLehose R, Reeves B, Harvey I, Sheldon T, Russell I (2000). A
ease. Muscle Nerve 48:3–20. systematic review of comparison of effect sizes derived from
Arthur-Farraj PJ, Murphy SM, Laura M, Lunn MP, Manji H, randomised and non-randomised studies. Health Technol
Blake J, Ramdharry G, Fox Z, Reilly MM (2012). Hand Assess 4:154.
weakness in Charcot-Marie-Tooth disease 1X. Neuromuscul Maggi G, Monti Bragadin M, Padua L, Fiorina E, Bellone E,
Disord 22:622–626. Grandis M, Reni L, Bennicelli A, Grosso M, Saporiti R,
Burns J, Raymond J, Ouvrier R (2009a). Feasibility of foot Scorsone D, Zuccarino R, Crimi E, Schenone A (2011).
and ankle strength training in childhood Charcot-Marie-Tooth Outcome measures and rehabilitation treatment in patients
disease. Neuromuscul Disord 19:818–821. affected by Charcot-Marie-Tooth neuropathy: a pilot study.
Chetlin RD, Gutmann L, Tarnopolsky M, Ullrich IH, Yeater RA Am J Phys Med Rehabil 90:628–637.
(2004a). Resistance training effectiveness in patients with Matjacic Z, Zupan A (2006). Effects of dynamic balance training
Charcot-Marie-Tooth disease: recommendations for exercise during standing and stepping in patients with hereditary
prescription. Arch Phys Med Rehabil 85:1217–1223. sensory motor neuropathy. Disabil Rehabil 28:1455–1459.
Chetlin RD, Gutmann L, Tarnopolsky MA, Ullrich IH, Yeater RA Menotti F, Felici F, Damiani A, Mangiola F, Vannicelli R, Macaluso
(2004b). Resistance training exercise and creatine in patients A (2011). Charcot-Marie-Tooth 1A patients with low level of
with Charcot-Marie-Tooth disease. Muscle Nerve 30:69–76. impairment have a higher energy cost of walking than healthy
Clauw DJ (2014). Fibromyalgia: a clinical review. JAMA individuals. Neuromuscul Disord 21:52–57.
311:1547–1555. Murphy MH, McNeilly AM, Murtagh EM (2010). Session 1: public
Cup EH, Pieterse AJ, Ten Broek-Pastoor JM, Munneke M, van health nutrition: physical activity prescription for public health.
Engelen BG, Hendricks HT, van der Wilt GJ, Oostendorp RA Proc Nutr Soc 69:178–184.
(2007). Exercise therapy and other types of physical ther- Neill J, Belan I, Ried K (2006). Effectiveness of non-
apy for patients with neuromuscular diseases: a systematic pharmacological interventions for fatigue in adults with
review. Arch Phys Med Rehabil 88:1452–1464. multiple sclerosis, rheumatoid arthritis, or systemic lupus ery-
Dibble LE, Addison O, Papa E (2009). The effects of exercise thematosus: a systematic review. J Adv Nurs 56:617–635.
on balance in persons with Parkinson’s disease: a systematic Pareyson D, Marchesi C (2009). Diagnosis, natural history, and
review across the disability spectrum. J Neurol Phys Ther management of Charcot-Marie-Tooth disease. Lancet Neurol
33:14–26. 8:654–667.
Donnelly JE, Blair SN, Jakicic JM, Manore MM, Rankin JW, Piscosquito G, Reilly MM, Schenone A, Fabrizi GM, Cavallaro
Smith BK (2009). American College of Sports Medicine Posi- T, Santoro L, Vita G, Quattrone A, Padua L, Gemignani F,
tion Stand. Appropriate physical activity intervention strate- Visioli F, Laura M, Calabrese D, Hughes RA, Radice D, Solari
gies for weight loss and prevention of weight regain for A, Pareyson D (2014). Is overwork weakness relevant in
adults. Med Sci Sports Exerc 41:459–471. Charcot-Marie-Tooth disease? J Neurol Neurosurg Psychiatry
Downs SH, Black N (1998). The feasibility of creating a check- 85:1354–1358.
list for the assessment of the methodological quality both of van Pomeren M, Selles RW, van Ginneken BT, Schreuders TA,
randomised and non-randomised studies of health care inter- Janssen WG, Stam HJ (2009). The hypothesis of overwork
ventions. J Epidemiol Community Health 52:377–384. weakness in Charcot-Marie-Tooth: a critical evaluation. J
El Mhandi L, Millet GY, Calmels P, Richard A, Oullion R, Rehabil Med 41:32–34.
Gautheron V, Feasson L (2008). Benefits of interval-training Ramdharry GM, Pollard A, Anderson C, Laurá M, Murphy
on fatigue and functional capacities in Charcot-Marie-Tooth SM, Dudziec M, Dewar EL, Hutton E, Grant R, Reilly
disease. Muscle Nerve 37:601–610. MM (2014). A pilot study of proximal strength training

361
Sman et al. Journal of the Peripheral Nervous System 20:347–362 (2015)

in Charcot-Marie-Tooth disease. J Peripher Nerv Syst Tremblay MS, Colley RC, Saunders TJ, Healy GN, Owen N (2010).
19:328–332. Physiological and health implications of a sedentary lifestyle.
Reeves BC, Deeks JJ, Higgins JPT, Wells GA (2011). Chapter Appl Physiol Nutr Metab 35:725–740.
13: including non-randomized studies. In: Cochrane Handbook Videler AJ, Beelen A, Nollet F (2010). Verifying the hypothesis of
for Systematic Reviews of Interventions. Chapter 13.5.2.3. overwork weakness in Charcot-Marie-Tooth. J Rehabil Med
Version 5.1.0 [updated March 2011]. Higgins J, Green S 42:380; author reply 380–381.
(Eds). The Cochrane Collaboration, London, United Kingdom. Vinci P, Esposito C, Perelli SL, Antenor JA, Thomas FP (2003).
Available at: www.cochrane-handbook.org. Accessed April 2, Overwork weakness in Charcot-Marie-Tooth disease. Arch
2015. Phys Med Rehabil 84:825–827.
Smith CA, Chetlin RD, Gutmann L, Yeater RA, Alway SE (2006). Yilmaz DC, Buyukakilli B, Gurgul S, Rencuzogullari I (2013). Adap-
Effects of exercise and creatine on myosin heavy chain tation of heart to training: a comparative study using echocar-
isoform composition in patients with Charcot-Marie-Tooth diography & impedance cardiography in male & female ath-
disease. Muscle Nerve 34:586–594. letes. Indian J Med Res 137:1111–1120.

Supporting Information
Additional Supporting Information may be found in the online version of this article.

362