Introduction
Charcot–Marie–Tooth (CMT) is the most common polygenic peripheral nerve disease, defined by axonal, dysmyelinating, and intermediate phenotypes.Reference Schenone, Nobbio, Monti Bragadin, Ursino and Grandis 1 CMT results in progressive distal muscle weakness, loss of sensation, and deep tendon reflexes,Reference Birouk, Gouider and Le Guern 2 as well as such consequent distal deformities as pes cavus and equinus.Reference Burns, Ryan and Ouvrier 3 The associated weakness and restricted movement results in muscle cramps,Reference Burns, Ryan and Ouvrier 3 imbalance, and gait difficulties.Reference Newman, Walsh and O’Sullivan 4
The estimated incidence of CMT has been recently increased from 1:2,500 to as high as ~1:1,200.Reference Braathen, Sand, Lobato, Høyer and Russell 5 Despite such a high prevalence, there remains a paucity of published studies to inform guidelines for CMT management. Current treatments include such nonsurgical interventions as stretching, bracing, and serial casting, although there is conflicting evidence regarding the long-term efficacy and sustainability of these methods.Reference Rose, Raymond, Refshauge, North and Burns 6 Previous studies have supported the utility of rehabilitation as a treatment for CMT patients.Reference Piscosquito, Reilly and Schenone 7 In particular, a small body of evidence suggests that exercise may be beneficial for long-term strength and functioning in CMT (see Kilmer [2002]Reference Kilmer 8 and Sman et al. [2015]Reference Sman, Hackett, Fiatarone Singh, Fornusek, Menezes and Burns 9 for reviews). In addition, preliminary research has shown that at-home resistance training can result in improvements in both strength and strength endurance.Reference Turna, Baker and Igdoura 10 However, the literature supporting the use of exercise as a treatment for CMT is still limited. Our study aimed to examine differences in strength among exercising and nonexercising individuals from both the CMT1 and CMT2 subtypes.
Methods
The charts of 297 CMT patients at McMaster University Medical Centre in Hamilton, Ontario, were reviewed retrospectively. The diagnosis of CMT was established in these patients in one of two ways: (1) positive genetic testing or (2) positive family history (with genetic testing) along with supportive clinical exam and electrophysiology (i.e., terminal motor latency >150% upper limit of normal; median motor conduction velocity<35 m/s), via either genetic testing, or family history and electrophysiological studies, in keeping with a hereditary dysmyelinating axonopathy in the setting of appropriate clinical dysmorphisms.
Data were collected from all visits and included demographic information, self-reported exercise patterns, and strength testing as recorded in the charts. Strength testing was achieved by standard objective measurement using Biodex (Biodex Medical Systems, Shirley, New York) and Jamar dynamometers (Performance Health, Warrenville, Illinois), according to a standardized procedure.Reference Beenakker, van der Hoeven, Fock and Maurits 11 Hand grip was measured in kilograms (kg), and muscle strength in other groups was measured at highest peak torque (newton-meters, Nm). Since many patients were lost to follow-up, the earliest visit with each patient was used for data analysis. Isometric strength testing scores were recorded for right- and left-hand grip strength, knee extension, elbow flexion, and ankle dorsiflexion. Strength values were converted according to age- and sex-matched normative values, by adjusting participants’ scores based on previously reported differences from baseline.Reference Harbo, Brincks and Andersen 12 Hand grip strength was converted for each participant into a percentage of normal, based on previously reported age- and sex-normative values.Reference Mathiowetz, Kashman, Volland, Weber, Dowe and Rogers 13 , Reference Mathiowetz, Wiemer and Federman 14
The data were analyzed quantitatively via t test in SPSS with a significance value of p<0.05. Participants who did not have a known CMT subtype diagnosis (CMT1 or CMT2) or who did not have sufficient strength data recorded were excluded from the analyses (a total of 112 patients excluded). Pediatric patients (under 18 years of age) were also excluded due to a lack of available normative values. For analysis of CMT1 subtypes, only those participants who had a known genetically diagnosed subtype of CMT1A, CMT1B, or CMT1D were included. In total, 117 CMT1 patients and 128 CMT2 patients were compared. Of these participants, 11 CMT1 and 16 CMT2 patients did exercise and were included in the exercise-positive (EP) group, while 8 CMT1 and 12 CMT2 patients did not exercise and were included in the exercise-negative (EN) group (Table 1). The remaining participants either did not have any exercise patterns recorded in their charts or had exercise that did not meet the inclusion or exclusion criteria.
A number of inclusion and exclusion criteria were employed for comparison of participants who exercised versus those who did not in order to optimize the validity of our results. The inclusion criteria for the EP group included the following: (1) those who reported their exercise routine frequency either as two or more times per week or as two or more types of exercises that met the criteria; (2) in order to meet the criteria, exercise types must have been whole-body training for strength building or cardiovascular exercise (e.g., walking, weights, aquafit). The exclusion criteria for this group were as follows: (1) those who did not indicate whether or not they did exercise; (2) those who only indicated that they did exercise but did not specify the frequency or type of exercise; and (3) those who did not do whole-body training but rather performed targeted exercise (e.g., physiotherapy for knee only).
In the EN group, only those who specifically indicated that they did not exercise were included. The exclusion criteria for this group were as follows: (1) those who did not indicate whether or not they did any exercise; and (2) those who specified a limitation that made it unclear for how long ago they stopped exercising (e.g., stopped due to a cardiac problem). In addition, some participants did not have data recorded for one or more measures, in which case they were only included for the measures that were recorded.
Results
Identical analyses were performed using both the scores adjusted for age and sex differences and the original strength values. Conversion of scores did not alter the statistical significance of our results. Of the charts reviewed, 117 adult participants had CMT1 or a genetically confirmed subtype of CMT1, and 125 had CMT2 or a genetically confirmed subtype of CMT2. Power testing demonstrated that CMT2 participants were stronger than CMT1 participants in right- and left-hand grip, dorsiflexion, and elbow flexion (Figure 1). Knee extension did not reach statistical significance (p=0.1259). CMT1 was further examined based on its subtypes, CMT1A versus CMT1B and CMT1D. CMT1A was weaker (p<0.05) than CMT1B and 1D in knee extension, with other strength values not reaching statistical significance (Figure 2). Unfortunately, the sample size was too small to examine CMT1B and D individually, so that these participants were combined.
Figure 1 Comparison of CMT1 versus CMT2 participants. CMT1 participants were statistically significantly stronger than CMT2 participants in right- and left-hand grip, dorsiflexion, and elbow flexion (p<0.05).
Figure 2 CMT1B and 1D participants were statistically significantly stronger than CMT1A participants in knee extension (p<0.05).
When comparing the data of the participants included in the exercise analysis, the group of 27 participants in the EP group were stronger than the 20 participants in the EN group in elbow flexion and dorsiflexion (p<0.05) (Figure 3). Exercise did not have an impact on hand grip strength.
Figure 3 Comparison of EP versus EN groups. Those who did exercise were statistically significantly stronger than those who did not in dorsiflexion and elbow flexion (p<0.05).
The CMT1 and CMT2 groups were further analyzed based on their exercise patterns. Within CMT1 participants, dorsiflexion remained statistically significantly higher in the EP group compared to those in the EN group (p<0.05) (Figure 4). When comparing only those with CMT2 who did exercise versus those who did not, those who did exercise had statistically stronger strength values in elbow flexion and dorsiflexion (p<0.05) (Figure 5). Exercise did not have an impact on hand grip strength in either group (Table 2).
Figure 4 CMT1 participants in the EP group were statistically significantly stronger in dorsiflexion than those in the EN group (p<0.05).
Figure 5 CMT2 participants who did exercise were statistically significantly stronger than those who did not in dorsiflexion and elbow flexion (p<0.05).
Discussion
Despite currently available treatment methods, individuals with CMT experience significant detriments to their quality of life.Reference Vinci, Serrao and Millul 15 , Reference Pfeiffer, Wicklein, Ratusinski, Schmitt and Kunze 16 In particular, distal weakness results in foot deformities, gait difficulties, and imbalance.Reference Burns, Ryan and Ouvrier 3 , Reference Newman, Walsh and O’Sullivan 4 This provides an obvious target for therapy. However, the benefits of exercise in different CMT subtypes is an area of CMT management that continues to lack evidence.
In our study, we have shown that CMT patients who participate in self-directed whole-body exercise have greater strength values in ankle dorsiflexion and elbow flexion. This was true when all patients were compared, as well as when CMT1 and CMT2 were examined individually (excluding the CMT1 group, where only dorsiflexion remained statistically significant), decreasing the likelihood of CMT subtype imbalance confounding the results. These results are in keeping with the results of existing literature from randomized trials that exercise improves strength in CMT patientsReference Kilmer 8 , Reference Sman, Hackett, Fiatarone Singh, Fornusek, Menezes and Burns 9 and should be included in management plans. We have also shown that exercise does not have an impact on hand grip strength, supporting previous evidence that exercise in CMT does not lead to overwork weakness.Reference Piscosquito, Reilly and Schenone 7 These results suggest that self-directed exercise may be a convenient, sustainable, and effective method of improving strength and decreasing disability in this population. The benefits on dorsiflexion are particularly notable, as distal weakness and the resulting foot deformities contribute heavily to many of the impairments these patients experience.
When comparing strength of CMT1 versus CMT2 subtypes, CMT2 participants were stronger in right- and left-hand grip, ankle dorsiflexion, and elbow flexion. Of the CMT1 subtypes, CMT1A was statistically weaker than CMT1B and 1D in knee extension. Larger sample sizes would have allowed for more accurate comparison of these groups and may have led to more significant results in other muscle groups. Nonetheless, these results are in keeping with existing evidence that the combined axo-glial dysfunction (i.e., dysmyelination and secondary axonal atrophy) in CMT1 may lead to greater loss of strength compared to the strictly axonal degeneration in CMT2,Reference Pareyson, Scaioli and Laurà 17 and that this process may be more pronounced in CMT1A compared to other CMT1 subtypes. It is possible that the benefits gained from exercise may be particularly beneficial in this group; however, this cannot be determined based on our study design. Future studies should aim to more closely compare the effects of exercise on differing CMT subtypes and examine whether there is greater benefit for more severe phenotypes. Unfortunately, there were not enough participants with diagnoses of different CMT2 subtypes (e.g., 2K, 2F, 2I) in order to analyze these individually. Future studies should also be aimed at examining more of these participants in order to have a more conclusive dataset for all CMT subtypes.
Notwithstanding that a causal relation cannot be inferred from this retrospective chart review, the most notable limitation is the possibility that individuals with a milder CMT phenotype, and thus a higher capacity to exercise, preselected themselves to exercise more than their counterparts with a more disabling phenotype. However, the presence of foot drop in participants in both the EP and EN groups suggests that the degree of disability may not have caused significant barriers to an ability to exercise. Given that this is a preliminary study, much work remains to be done to delineate the type of exercise prescription that would best address the needs of the CMT population. Additionally, understanding whether lifelong exercise has the potential to mitigate disease progression and ultimately change the natural history of their disease remains unknown and represents a significant limitation in the clinical management of individuals with CMT.
There are several other notable limitations to our study. A potential confound could include differences in the age and sex distribution of participants between the various groups. This limitation was addressed through adjustment of scores according to previously reported normative data. Importantly, conversion of strength scores did not alter the statistical findings, suggesting that the distribution of age and male-to-female ratios of participants were not significant. Further limitations of the study include the following: only a small number of patients met the inclusion criteria for exercise analysis; the study did not include longitudinal strength testing due to patient attrition; and, despite the use of inclusion and exclusion criteria for exercise, the study lacked a standardized measurement or definition of exercise. Despite these limitations, our study does provide preliminary evidence that (1) self-directed exercise regimes may be beneficial for CMT patients of varying subtypes, (2) that exercise is not detrimental for strength in these patients, and (3) that there is merit in exploring this area in future studies.
To our knowledge, ours is the first study to examine the association between self-directed exercise and strength in different CMT subtypes. Secondary measures were not examined in this study and should be a goal for subsequent research, including whether improvements in strength also result in benefits for gait and balance. At present, we have shown that even self-selected exercise may benefit CMT patients without causing overwork weakness. Given the known consequences of weakness on disability in this population, improving strength may have a dramatic impact on quality of life. The management plans of patients with CMT should be comprehensive and include accessible exercise programs.
Table 1 Age and sex information for each group analyzed
Table 2 Summary of statistical analyses for all groups and strength values
*p<0.05
Acknowledgments
We would like to thank the patients, who continue to help inform research on CMT.
Disclosures
Djurdja Djordjevic, Sabrina Fell, and Steven Baker hereby declare that have no conflicts of interest to disclose.
