The effects of calf massage in boys with Duchenne muscular dystrophy: a prospective interventional study

Abstract Purpose We explored the effects of standardized calf massage in ambulant boys with Duchenne muscular dystrophy (DMD) using a prospective study design. Materials and Methods Twenty boys completed two study visits, 1 week apart. At both visits, each leg received a 10-min calf massage (intervention) and a 10-min control rest period (placebo) in randomized order. Muscle length of calf and hamstrings and gastrocnemius stiffness were measured by a blinded assessor before and after intervention and placebo. Measures of gait function (timed 10-m walk/run and spatio-temporal gait parameters); gastrocnemius muscle ultrasound findings; participant perception of leg pain, stiffness and effort of walking and general psychological well-being were also collected. Results Consistent significant small increases in muscle length of soleus, gastrocnemius and hamstrings were recorded post-massage, and gastrocnemius stiffness decreased. Small changes in gastrocnemius and soleus length only were also recorded following the control rest period. Gait function and general well-being remained stable throughout. Measurement across both study visits suggested that gains in muscle length may be cumulative with repeated massage. Conclusions Calf massage is safe and associated with benefits to muscle length and stiffness for ambulant boys with DMD. Implications for Rehabilitation In a small sample of boys with Duchenne muscular dystrophy, calf massage was found to be safe, well-tolerated and associated with increased muscle length and decreased stiffness. The use of massage may assist in managing muscle length in boys with Duchenne muscular dystrophy.


Introduction
Duchenne muscular dystrophy (DMD) is an X-linked muscle disorder with an incidence of one in 3500-5000 live male births [1]. Affected boys experience progressive muscle weakness and functional decline in association with muscle fibrosis causing stiffness and shortening. This adversely impacts physical function, including gait performance, participation in physical activities and quality of life. There is no cure for DMD and current treatment options are limited. The lack of muscle extensibility is typically managed with a stretching program, splints or orthotic devices and occasionally orthopaedic surgery [2]. DMD is characterized by the degeneration of skeletal muscle fibres with increased fibrosis and replacement of muscle tissue by fat. Increased muscle fibrosis is specifically associated with increased muscle stiffness [3] and long-term poor functional outcome in DMD [4]. It has been suggested that antifibrotic therapies should be considered in the treatment of DMD and that massage is one modality that may be useful [3]. Three studies have reported that up to 31% of boys with DMD choose to use massage as a complementary therapy with benefits including relaxed muscles and better circulation [5][6][7]. Other studies have found that massage can improve muscle length and range of motion [8][9][10][11]. To date, there has been no objective prospective investigation of the effects of massage therapy in DMD, despite its popularity as an adjunctive treatment in this population.
The aim of this study was to assess the effects of massage to the calf muscle in ambulant boys with DMD. It was hypothesised that massage would increase muscle length and decrease muscle stiffness without adversely affecting gait function or well-being.

Ethics
This study received ethics approval from The Royal Children's Hospital in Melbourne, Australia (HREC34188A). Written informed consent was received from a parent of each participant and where applicable, children also completed a consent form.

Participants
Children attending the neuromuscular outpatient clinic at The Royal Children's Hospital were invited to participate in this study. Inclusion criteria included: genetically confirmed diagnosis of DMD, aged 6 years or older, independently ambulant over distances greater than 10 m and able to tolerate lying prone for more than 10 min at a time.
Exclusion criteria included any calf injury with ongoing signs or symptoms, open wound(s) in the area to be massaged or children who were unable to cooperate with massage and assessment procedures. All participants were asked to abstain from any leg massage in the month preceding study enrolment so as not to confound the baseline assessment findings. No changes were made to any pre-existing physiotherapy programs that individual participants were undertaking.

Study design
This exploratory study used a prospective interventional design to investigate effects of bilateral calf massage on muscle length, muscle stiffness, gait and well-being.
In addition, we used a within-subjects randomized cross-over design with a blinded assessor to compare the effects of a unilateral calf massage (intervention) to a unilateral calf rest period (placebo) on muscle length and stiffness. Cross-over occurred within the one study visit.
Two similar study visits were conducted for each participant approximately 1 week apart (window of 6-10 days). The second visit allowed us to assess if massage effects were cumulative.
Measures of muscle length and stiffness were taken pre-massage, mid-visit (after the first leg was massaged, second leg rested) and post-massage (after both legs had been massaged and rested). The mid-visit measures allowed direct comparison of unilateral calf massage (intervention) to rest (placebo) in individual legs. Pre and post-massage measures allowed overall effects of bilateral calf massage to be determined. Muscle ultrasound, gait and self-perception scales were assessed pre and post-massage; the World Health Organisation-Five Well-Being Index was administered pre-massage only at each visit. Anthropometric measures were performed pre-massage at Visit 1 only. The study design is summarized in Figure 1.

Intervention
Participants lay prone with lower legs supported on a pillow while 10 min of massage was delivered to the calf muscles of each leg according to protocol whilst the other leg rested. The massage intervention comprised: 2 min effleurage (light pressure administered with a flat hand), 3 min petrissage (medium pressure administered in a wringing motion by compressing and lifting the belly of the muscle between the thumb and fingers), 3 min muscle stripping (deep friction administered by thumbs with slow strokes) and finishing with another 2 min effleurage (light pressure as described above). This is consistent with the massage intervention delivered in the 2012 study by Crane and colleagues which demonstrated changes in inflammatory signalling post-massage [12]. The massage intervention was delivered to the entire calf muscle complex including the tendoachilles. Each component of the massage intervention was timed using a stopwatch. All massage therapy was delivered by a single physiotherapist who had undergone additional training in massage therapy (KdV). Every attempt was made to standardize the intensity of massage for each participant at each visit. Between study visits 1 and 2, participants continued with their regular individual physiotherapy programs which may have included stretching, ankle-foot night splint wear and physical activity. Participants did not receive any massage between the two study visits.
Outcome measures Muscle qualities. The primary outcome was change in calf muscle length post-massage measured using a goniometer. Measures were taken of ankle dorsiflexion with knee flexed to 90 , ankle dorsiflexion with knee extended and popliteal angle according to a standard protocol. These measures assessed muscle length of soleus, gastrocnemius and hamstrings, respectively.
Passive muscle stiffness was assessed using a myotonometer (MyotonPRO, Tallinn, Estonia). Myotonometry is a method for objectively assessing and quantifying the compliance of biological soft tissues including muscle. The use of myotonometry has not previously been reported within the DMD population but has the potential to provide clinically relevant information [3].
Myotonometry measurements were performed over the medial head of gastrocnemius with the calf muscle relaxed. A series of five measures with repeatability within 2% was captured for each leg.
Muscle ultrasound assessment was used to quantify fibrosis [13]. We assessed the gastrocnemius of the dominant leg pre and post-massage at one-third of the distance from the popliteal fossa to the medial malleolus, consistent with the area previously assessed by myotonometry. A General Electric (Fairfield, CT) BT12 LOGIQ-e imaging system with a high-frequency (7-13 MHz) broadband linear transducer was used. Machine settings were kept constant throughout the study. A region of interest was manually selected for each image and analysed using a custom-built grayscale analysis program (Qumia) [14]. Three images of the gastrocnemius muscle were taken at each visit, and the average echo intensity calculated.
Gait. Spatio-temporal gait parameters were assessed using an electronic walkway (GAITRite TM , CIR Systems, Inc., Franklin, USA) to record six walks at self-selected preferred barefoot walking speed. The GAITRite has been shown to be a valid and reliable measure of gait in children with movement disorders [15]. A timed function test of running 10 m barefoot at maximal speed was recorded in seconds from a stopwatch. Participant perception. Participants used self-report evaluative scales to rate their perception of leg muscle stiffness, pain and effort of walking on a scale of 0-10 pre-and post-massage. Developed for this study, these scales were based on the Defense and Veterans Pain Rating Scale and included faces, colours and numerical elements [16]. Similar self-report scales have been validated for the evaluation of pain in children; the scales designed for this study have not been validated in children [17]. Subjective psychological well-being was assessed pre-massage at each visit with the World Health Organisation-Five Well-Being Index, 1998 version: a measure that is sensitive and validated for use in children [18].
Anthropometrics. Anthropometric measures including standing height, weight, tibial length and maximal calf circumference were measured to help characterize the participants. Body mass index (BMI) was calculated as kg/m 2 .

Randomization and blinding
Participants rolled a die during the first study visit to randomize massage order-those rolling an odd number had their right leg massaged first; those rolling an even number had their left leg massaged first. All participants received the massage intervention and a control rest period to each leg on both visits. During the second study visit, the order in which the legs were massaged was reversed.
All outcome measures were assessed by a second physiotherapist (KC) who was blinded to the randomization and not present during the massage intervention. All lower legs were concealed in a tubular bandage during goniometry so that potential massagerelated changes in skin colour, temperature, and presence of oil could not be detected.

Data analysis and statistics
Clinical data were collected and managed using REDCap TM electronic data capture tools hosted by Murdoch Children's Research Institute. Data from GAITRite TM , MyotonPRO TM and Qumia ultrasound analysis software were exported into Excel. All data were analysed using StataV R (Version 15) software.
Descriptive statistics were used to describe and characterize the participants. The normality of data distribution was visually assessed and parametric or non-parametric testing applied. Two main analyses were performed. To compare the effects of unilateral calf massage (intervention) to rest (placebo), we calculated difference variables for muscle length and stiffness in each leg immediately before and after each intervention/placebo. Linear regression was then applied to the different variables, clustered by participant, to test the null hypothesis.
To investigate overall treatment effects of bilateral calf massage, we calculated different variables for muscle length, stiffness and ultrasound echo intensity before and after bilateral massage. A similar linear regression was then applied to the different variables, clustered by participant, to test the null hypothesis. Treatment effects on gait and participant perception measures were determined using paired t-tests or Mann-Whitney U tests.
Standard deviation (SD) and ninety-five percent confidence intervals (95% CI) are reported and significance was set at p < 0.05. Cohen's d was calculated for significant differences and interpreted as small (0.2 to <0.5), medium (0.5 to <0.8) or large (!0.8) effect size [19].

Participant characteristics
Twenty male participants were invited to participate and completed the study. They are described in Table 1. Participant age varied from 6 to 14 years with a mean age of 9.3 (SD 2.71) years.
The mean interval between visits 1 and 2 was 7.1 (0.55) days. Participants and families generally tolerated the massage well and reported that they enjoyed participating in this study.

Muscle qualities
We compared immediate muscle responses to two interventions: a unilateral calf massage and a control period of rest. Immediately following each unilateral massage, we detected increased muscle length and decreased stiffness in all muscle groups assessed. These changes were consistent for all muscles at both study visits and were greater at the second visit. Post-rest improvements were seen in gastrocnemius (both study visits) and soleus (first visit only, Table 2).
Following bilateral massage, we recorded consistent small increases in muscle length for soleus, gastrocnemius and hamstrings; and a decrease in calf muscle stiffness post-massage for each study visit (Table 3). Muscle echogenicity increased at Visit 2 only (with small effect size).
Comparison of measures from pre-massage at Visit 1 and postmassage at Visit 2 showed a cumulative effect of two massages on muscle length with medium to large effect sizes (Table 3).

Gait and participant perception
Spatio-temporal gait parameters including gait speed, cadence, step length, base of support width, heel contact time, step length symmetry and step-to-step variability all remained stable throughout the study. The 10-m walk/run time did not change (Supplementary Table S1). Some of the younger participants demonstrated limited understanding of the patient perception scales and did not complete all questions. The 16 participants who completed the scales reported initial low levels of pain, leg stiffness and effort of walking (Visit 1 baseline means 0.9, 1.4 and 1.0, respectively). After massage, they self-reported inconsistent improvements in stiffness (Visit 1 only, p ¼ 0.004, d ¼ 0.68) and pain (Visit 2 only, p ¼ 0.018, d ¼ 0.47). There was no change in perceived effort of walking.
Participant well-being remained stable with World Health Organisation-Five Well-Being Index scores unchanged from Visit 1 (mean score 19.9) to Visit 2 (mean 20.7; p ¼ 0.210).

Discussion
This exploratory study is the first to objectively document effects of massage in ambulatory boys with DMD. Overall the boys  enjoyed participating in the study and no deleterious effects were noted in gait or running function, participant perception or muscle qualities including muscle length and stiffness. The findings of this study indicate that calf massage is a safe and well-tolerated intervention in this population. Calf massage has positive effects on muscle length and calf muscle stiffness as shown by the small but significant increases in calf and hamstring muscle length and reductions in calf muscle stiffness detected in this study. The use of calf massage can therefore be recommended as one strategy for helping to maintain calf and hamstring extensibility in ambulant boys with DMD.
While the changes in muscle length were statistically significant and consistently indicated post-massage gains in muscle length for calf and hamstrings, the changes were small which is consistent with a previous study investigating massage effects on ankle joint flexibility in healthy young adults [20]. Our findings of increased muscle length should be interpreted with caution given the potential measurement error in goniometry, cited to be between 3 and 5 degrees for children with cerebral palsy [21,22]. However, the pathology of fibro-fatty replacement of muscle tissue in DMD is associated with a clinical picture of relentlessly progressive muscle contractures that require consistent stretching efforts to maintain the existing muscle length. In this context, any gain in muscle length is potentially useful.
Massage therapy is typically used repeatedly over time. The cumulative results at the end of the second study visit showed greater increases in muscle length after two sessions of massage than after a single massage. The effect size was also greater. The massaging physiotherapist noted that some of the younger children who were massage-naïve prior to the study demonstrated limited tolerance of stronger and deeper massage techniques (muscle stripping) at Visit 1 only. This tolerance was improved on the second visit. These factors suggest that there may be cumulative benefits with repeated calf massage.
We noted some small increases in calf muscle length following a control period of rest. There are a number of potential explanations for this. It is possible that resting in a prone position has positive effects on calf muscle length. Some children may have been anxious at the start of a study visit, particularly on the first occasion and may not have fully relaxed their muscles when being measured. The repeated measurements of maximal muscle length may conceivably have had a stretching effect. However for a muscle stretch to be effective, it is typically advised that a stretch be sustained for at least 30 s [23] which is much greater than the approximately 5 s required to complete a length measurement. It is also possible that the greatest gains in muscle length do not occur immediately after a massage but that effects on muscle length accumulate with rest.
Based on previous studies and patient report in our own population, the potential benefits of massage are wide-ranging. In addition to improved muscle length, people also report softer muscles, decreased pain and stiffness, improved movement and well-being [5,24,25]. In an effort to capture all potential effects of the calf massage investigated in this study, a number of different outcome measures were used. A previous survey of children with DMD and their families in our centre identified that massage was associated with "softer muscles", something that families considered to be positive [5]. In an attempt to quantify this reported change in muscle compliance, muscle stiffness (the inverse of compliance) at rest was assessed using a myotonometer. Myotonometry has previously been able to detect significant change in local muscle stiffness following massage in a study of adults with low back pain [26]. Obtaining a consistent set of appropriate measures was quick, easy and well tolerated by the participants. Increased muscle stiffness implies increased resistance to change in muscle length and therefore requires increased effort from the agonist muscle to lengthen a stiff antagonist muscle [27]. A reduction in gastrocnemius stiffness could potentially make active dorsiflexion easier during swing phase of the gait cycle, with increased heel strike at initial contact and improved foot clearance in swing, although no changes in gait parameters were observed in this study. The reductions in gastrocnemius stiffness observed in this study are consistent with other research that also reported reduced passive muscle stiffness (as measured using shear wave ultrasound) immediately following calf massage in healthy participants [25].
The physiological mechanisms by which massage may work remains unclear but studies have demonstrated increased blood flow to and temperature of the massaged area [28][29][30][31] and reduced spinal reflex excitability through induced muscle relaxation [32]. Other theories postulate that the massage action may disturb the cross bridges between the actin and myosin filaments within the muscle itself [33]. Effects may be due to a combination of these potential factors. A small but rigorous study of massage for exercise-induced muscle damage in healthy young adults examined quadriceps muscle biopsies and found that markers of cellular stress and inflammatory signalling were reduced following massage [12]. The specific effects of massage in muscles with pathology, as seen in DMD, are unclear.
Ultrasound assessment of the gastrocnemius indicated a trend for the grayscale echogenicity score to increase post-massage; this change was significant at Visit 2 with a small effect size (p ¼ 0.049, d ¼ 0.17). It is unclear if this is a chance finding, or if massage has an immediate effect on muscle echogenicity findings on ultrasound. Massage is known to increase blood flow and decrease oedema; which may affect muscle size and echogenicity.
Some of the younger participants demonstrated limited understanding of the perception rating scales and were unable to reliably self-report. Participants reported very low baseline levels of pain, muscle stiffness and effort of walking and therefore these variables were not useful for detecting massage-related benefits. The stability of all self-reported measures including the World Health Organisation-Five Well-Being Index supports the safety and tolerability of massage for boys with DMD.
The findings from this study are limited by small numbers and short study duration. The young age and apparent presence of learning difficulties in some participants affected participant perception information and may have influenced gait performance. Many variables were statistically tested increasing the likelihood of chance positive findings. Further investigation of the efficacy of massage would benefit from a larger sample size, additional treatment sessions and muscle groups, longer study duration, assessment of long-term effects and possible combination with other treatment techniques such as stretching and wearing night splint ankle-foot orthoses. This would allow more specific clinical recommendations to be made regarding the use of massage for optimal outcomes.

Conclusions
Calf massage is safe and effective for use in ambulant boys with DMD. Massage of both legs in a single session is associated with increased muscle length of calf and hamstring muscles with a reduction in gastrocnemius muscle stiffness. Some cumulative effects on muscle length may be observed with repeated massage over time.