[
research report
]
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
THOMAS HORSTMANN, MD1,2 • HOLGER M. JUD, MD3 • VANESSA FRÖHLICH, MD4 ANNEGRET MÜNDERMANN, PhD5 • STEFAN GRAU, PhD6
Whole-Body Vibration Versus Eccentric Training or a Wait-and-See Approach for Chronic Achilles Tendinopathy: A Randomized Clinical Trial
C
hronic Achilles affliction and tendinopathy are clinical conditions characterized by pain, swelling, altered joint motion, and limited performance,27 and are frequently observed not only in physically active but also in sedentary persons.23,41 However,
most frequently these conditions develop during recreational physical activity of individuals aged 35 to 45 years,2 with a TTSTUDY DESIGN: Randomized clinical trial. TTOBJECTIVES: To test the hypothesis that
whole-body vibration training results in greater improvements in symptoms and pain, structural changes, and muscle flexibility and strength of the triceps surae muscle-tendon unit than those achieved with eccentric training or with a wait-andsee approach.
TTBACKGROUND: The potential use of vibration
training for the treatment of Achilles tendinopathy has not been explored.
TTMETHODS: Fifty-eight patients (mean age, 46.0
years) with Achilles tendinopathy were randomly assigned to a 12-week intervention using wholebody vibration training, eccentric training, or a wait-and-see approach. Pain, tendon structure and path, and muscle flexibility and strength were assessed at baseline and follow-up, and compared using mixed-factor analyses of variance.
TTRESULTS: Pain improvements at the midsection of the tendon were greater in the vibration- and
higher prevalence in men than in women.23,29 Chronic Achilles tendinopathy can be caused by intrinsic factors (eg, genetic, eccentric-training groups than in the wait-and-see group (mean difference from the vibration-training group, –18.0; 95% confidence interval [CI]: –35.0, –1.1; mean difference from the eccentric-training group, –27.0; 95% CI: –50.9, –3.1). Improvements in pain at the musculotendinous junction were greater in the eccentric-training group than in the other groups (mean difference from the vibrationtraining group, –31.4; 95% CI: –60.7, –2.0; mean difference from the wait-and-see group, –50.2; 95% CI: –82.3, –18.1). Improvements in most participants were achieved in the vibration-training group, followed by the eccentric-training group.
TTCONCLUSION: Vibration training may be an
alternative or a complementary treatment in patients who do not respond well to eccentric training, especially in those with insertional pain.
TTLEVEL OF EVIDENCE: Therapy, level 2b–. J Orthop Sports Phys Ther 2013;43(11):794-803. Epub 9 September 2013. doi:10.2519/jospt.2013.4762
TTKEY WORDS: isokinetic muscle strength, muscle flexibility, pain, tendon structure, ultrasound
metabolic, or endocrine factors, as well as those of age, body size, anatomical misalignment of joint axes, reduced muscle flexibility, or muscle dysfunction or weakness23) and/or extrinsic factors (eg, previous injuries, inappropriate footwear, or environmental factors such as training on hard surfaces or a sudden increase in training loads5,18,23,41). Current therapies for chronic Achilles tendinopathy include prescription of foot orthoses, local or systemic antiinflammatory medication (eg, nonsteroidal antirheumatic agents or cortison) or topical nitroglycerin, stretching, massage therapy, electrotherapy, therapeutic ultrasound, extracorporeal shockwave therapy, or eccentric training.1 Most of these therapies are long-term therapies that are effective in about 3 of 4 patients. 37 Of these, eccentric training generally shows promising results in the treatment of chronic Achilles tendinopathy, reducing pain20 and the necessity of surgical intervention,2,45 and is effective in about 60% of patients.28,44 Despite its common use, to date, there is large variability in training protocols regarding dosage and duration of eccentric training.32 Moreover, it has been argued4 that the mechanism
Medical Park Bad Wiessee St Hubertus, Bad Wiessee, Germany. 2Faculty for Sport and Health Sciences, Technische Universität München, Munich, Germany. 3Department of Orthopaedic Surgery, Krankenhaus Bad Cannstatt, Stuttgart, Germany. 4Berufsgenossenschaftliche Unfallklinik, Tübingen, Germany. 5Osteoarthritis Research Center, University Hospital Basel, Basel, Switzerland; School of Physiotherapy, University of Otago, Dunedin, New Zealand; Division of Sport Science, University of Konstanz, Konstanz, Germany. 6 Department of Sports Medicine, Medical Clinic, University of Tübingen, Tübingen, Germany. This study was approved by the Institutional Ethics Board of the University of Tübingen. The authors certify that they have no affiliations with or financial involvement in any organization or entity with a direct financial interest in the subject matter or materials discussed in the article. Address correspondence to Dr Thomas Horstmann, Medical Park Bad Wiessee GmbH, St Hubertus, Sonnenfeldweg 29, 83707 Bad Wiessee, Germany. E-mail: [email protected] t Copyright ©2013 Journal of Orthopaedic & Sports Physical Therapy® 1
794 | november 2013 | volume 43 | number 11 | journal of orthopaedic & sports physical therapy
43-11 Horstmann.indd 794
10/16/2013 4:54:25 PM
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
of eccentric training that contributes to tendon healing may not be its strengthening effects but rather a modulation of the neurological stretch response. There has been increasing scientific scrutiny of whole-body vibration training in the context of muscle performance training and therapy for a range of medical conditions43,46 and in older adults.24 Improvements in neuromuscular performance with whole-body vibration training have been attributed to more synchronized motor units within muscles6 and increased spatial recruitment via potentiation of the neuromuscular system through muscle spindles, resulting in reflex activation of motor neurons.39 Short-term adaptation processes to whole-body vibration training include performance-enhancing effects on muscle, increased release of growth hormones and testosterone and reduced cortisol levels, and reduced resistance in peripheral vessels, resulting in improved nutrient supply to the muscles.16,17,26 Long-term adaptations include increased isometric and dynamic leg strength8,13 and bone density.47 Though the shortterm effects of whole-body vibration training on lower extremity flexibility and jumping power are controversial,7,11,21,42 in the long term, these parameters show improvements.9,11,42 Further, Hilgers et al15 recently demonstrated that whole-body vibration training increases walking performance in patients with multiple sclerosis, suggesting that vibration training may positively influence neuromotor control. Because the treatment of chronic Achilles tendinopathy aims at increasing ankle flexibility and plantar flexor muscle strength but may also require modulation of the neuromuscular system, wholebody vibration training may be effective in the therapy of this patient group. To date, the potential of using whole-body vibration training as a therapeutic intervention for chronic Achilles tendinopathy has not been explored scientifically. Therefore, the purpose of this study was to test the hypothesis that wholebody vibration training results in greater
TABLE 1
Inclusion and Exclusion Criteria Used in This Longitudinal Study
Inclusion criteria: • Age, 25-55 y • Recurrent complaints in 1 or both Achilles tendons at rest and/or during exercise for the preceding 6 mo • Structural changes of the tendon confirmed via sonographic examination during the initial physical exam • Weekly running mileage less than 50 km (aerobic exercise) • No increase in training intensity during the study • No supplemental treatment of the chronic Achilles tendinopathy during the study • Informed consent Exclusion criteria: • Treatment of the chronic Achilles tendinopathy within the preceding 4 wk, including physiotherapy, physical treatment, self-treatment (eg, heel pads), medication therapy (no steroidal antirheumatics, cortisone) • Complaints in the lower extremity other than those related to the chronic Achilles tendinopathy • Regular intake of medication that might have influence on the outcome • Residence too far from clinic • Pregnancy • Current inflammation of the musculoskeletal system • Fractures of the trained body parts within the past 12 mo • Surgery within the past 6 mo • Current discopathy • Rheumatoid arthritis • Osteoarthritis or arthropathies • Biliary calculus illness or kidney stones
improvements in symptoms and pain, structural changes, and muscle flexibility and strength of the triceps surae muscletendon unit compared to those achieved with established eccentric training or with a wait-and-see approach.
METHODS
R
ecreational runners with chronic Achilles tendinopathy were recruited by advertisement in local newspapers and the University of Tübingen website. Volunteers were initially contacted by telephone to determine their eligibility for participation, based on the inclusion and exclusion criteria listed in TABLE 1. If the volunteer was eligible for this study, an appointment for the initial exam was made. At the first visit, after providing written consent, subjects underwent a clinical examination completed by the lead physician (T.H.), and isokinetic measurements conducted by the junior physician (H.J.). Using a randomization list, the participants were assigned to a vibration-training group, a conventional eccentric-training group,
or a wait-and-see control group (FIGURE 1). Study personnel were blinded to the randomization list and did not know the group assignment of the newly enrolled patients until after their enrollment. Participants completed a 12-week program according to their group assignment, and the clinical exam and isokinetic measurements were repeated after 12 weeks. Both assessors were blinded to the participants’ group assignment. This study was approved prior to subject recruitment by the Institutional Ethics Board of the University of Tübingen.
Patients All patients had chronic Achilles tendinopathy at the midsection of the tendon, and some patients also reported pain at the osseous insertion or the musculotendinous junction. Both Achilles tendons of each participant were examined, but only the more affected side was included in the analysis.
Training Interventions During the 12-week intervention phase, 36 training sessions were scheduled
journal of orthopaedic & sports physical therapy | volume 43 | number 11 | november 2013 | 795
43-11 Horstmann.indd 795
10/16/2013 4:54:26 PM
[
research report
]
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
Volunteers with Achilles tendinopathy, n = 89 Volunteers met at least 1 exclusion criteria, n = 31 Volunteers met all inclusion criteria, n = 58 Volunteers declined to participate, n=0 Volunteers gave informed consent and were randomized into 3 intervention groups, n = 58
Vibration training, n = 23 Lost to follow-up, n = 1 Vibration training (completed final exam), n = 22
Eccentric training, n = 19
Wait and see, n = 16
Lost to follow-up, n = 1 Eccentric training (completed final exam), n = 18
Lost to follow-up, n = 2 Wait and see (completed final exam), n = 14
FIGURE 1. CONSORT flow chart of this longitudinal study.
for each participant in the vibrationtraining and eccentric-training groups. Participants were familiarized with the training equipment in the first training session. Each training session started and ended with a 5- to 10-minute warm-up and cool-down, respectively, on an ergometer or treadmill, and participants were supervised to ensure correct execution of the training exercises. During the 12-week intervention phase, participants were allowed to continue their regular activity schedule without increasing their training workload. A training diary was completed by the participant. The vibration training was conducted on a Galileo Fit (Novotec Medical GmbH, Pforzheim, Germany) and included a warm-up, a training phase, and a cooldown. During the 1-minute warm-up, participants changed consecutively between bipedal stance (neutral stance), unipedal stance, heel rises, and stepping in place, with their knees slightly bent, on the vibration platform set to a vibration frequency of 13 to 18 Hz and at
an amplitude of 0.4 to 0.6 mm. For the training phase, the vibration training was increased from 4 to 5 minutes in the first 4 weeks to 5 to 6 minutes in the second 4 weeks to 6 to 7 minutes in the last 4 weeks, and the vibration frequency and amplitude were increased to 16 to 21 Hz and 0.5 to 0.8 mm, respectively. Though vibration frequencies above 20 Hz are reported to elicit increased muscle performance38 and could have been used in our intervention, patients with Achilles tendinopathy in a pilot study perceived vibration frequencies above 21 Hz as uncomfortable and even painful (unpublished data). Participants intermittently changed between heel rises and heel drops (3 seconds in each position, with a 1-second transition between each position) at the edge of the platform until the onset of fatigue. Participants were allowed to briefly recuperate during heel-rise activities or alternate stretching of the calf muscles. During the 1-minute cool-down, participants relaxed their muscles by statically stretching, alternating between legs, with 1 leg on the
platform at a vibration frequency of 13 to 18 Hz. The eccentric training was conducted using a Reebok Step (Reebok International Ltd, Canton, MA) and included 2 exercises. First, participants stood upright (knee extended), with their entire body weight on the ball of 1 foot and their ankle in plantar flexion at the edge of the step. From this starting position, they slowly lowered their ankle until they reached full dorsiflexion (3-5 seconds).44 Subsequently, they used their contralateral limb and their arms to return to the starting position (1-2 seconds). Participants performed 3 sets of 15 repetitions on each leg.3 To increase training intensity between training periods and to ensure a training effect, participants who completed the 3 sets without any signs of fatigue performed a fourth set, and, if necessary, the load was increased by placing additional weights in a backpack attached to the participant’s back. Instead of completing training intervention sessions, participants in the
796 | november 2013 | volume 43 | number 11 | journal of orthopaedic & sports physical therapy
43-11 Horstmann.indd 796
10/16/2013 4:54:27 PM
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
wait-and-see group filled out an activity or training log to ensure that they maintained their recreational activities. In particular, they noted the time, duration, type, and intensity of their physical activities. After the 12-week control intervention period, participants in this group had the option of receiving a complementary gait analysis and/or participating in eccentric or vibration training.
Clinical Exam Participants were asked about the location of their pain (tendon insertion/ calcaneus, midsection of the tendon [2-6 cm proximal to the insertion], or musculotendinous junction) and the occurrence of their pain (time of first occurrence, sudden or gradual initial onset, or result of trauma). The type, time, and duration of previous treatment interventions were recorded (nonsteroidal antirheumatics, corticoids, and other medication; physical interventions [heat or cold treatment]; physiotherapy [also manual therapy]; foot orthoses; or other interventions). In addition, the type, intensity, duration, and frequency of their typical training regimen for all sports were recorded. The orthopaedic measures assessed in the clinical exam included static rearfoot and midfoot shape (normal, pes cavus, talipes valgus, or flatfoot), static calcaneal alignment (normal, varus, valgus), static forefoot alignment (normal, splayfoot, hallux valgus, or toe deformity), physiological foot correction while standing on the ball of the foot (present, absent), unilateral heel rise (painful, yes/ no), passive anterior translation of the talus (present, absent), range of motion of the ankle joint and the subtalar joint, leg length (leg-length discrepancy, yes/ no), leg alignment (normal, genu recurvatum, genu valgus, genu varum), and pelvic drop (present, absent).
Pain and Pain Impact Assessment The change in symptoms from preintervention to postintervention was assessed using a standard Likert scale (symptoms
completely recovered, much improved, little improved, unchanged, a little worse, or much worse).40 Visual analog scales (VASs) were used to assess the impact of pain on the categories “family and responsibility at home,” “recreation,” “social activities,” and “running training or other physical activities.” Each VAS ranged from “no limitation” (0 mm) to “complete limitation” (100 mm). To assess palpation pain for the Achilles tendon on both legs, the examining physician palpated the Achilles tendons between his thumb and index finger and pinched the tendon for a few seconds with the force that would be necessary to hold a 1-kg weight between the thumb and the index finger.45 Participants were asked to rate their pain on the VAS for the locations musculotendinous junction, midsection of the tendon, and osseous insertion from “no pain” (0 mm) to “excruciating pain” (100 mm).
Sonographic Assessment A sonographic exam of both Achilles tendons was conducted using a SONOLINE Versa Pro (Siemens AG, Munich, Germany). Sonographic images perpendicular to and along the Achilles tendon,40 close to the osseous insertion and 2 cm proximal to the osseous insertion, were examined for abnormalities, including inhomogeneity at tendon insertion, inhomogeneity in tendon path, swelling, bursitis, peritendinosis, tendinitis, and ossification,33 and documented in printouts. We defined a positive diagnosis as presence and a negative diagnosis as absence of any sonographic abnormalities.
Isokinetic Measurements Force measurements were taken with a computerized dynamometer training system (IsoMed 2000; D. & R. Ferstl GmbH, Hernau, Germany). Subjects wore the same standardized footwear (Air Pegasus; Nike, Inc, Beaverton, OR) for both the baseline and follow-up examination. The dynamometer was adjusted to assess ankle dorsiflexion/plantar flexion, with the axis set to 10 cm above the plinth
and the participant’s lower leg secured with a strap to immobilize the fully extended knee while the patient was in a supine position. The dynamometer was adjusted horizontally to the participant’s size by aligning the dynamometer axis to the ankle joint axis, and the participant’s foot was secured to the adapter. These settings were recorded to ensure that they were the same at baseline and at followup. Participants placed their extended arms along their trunk and held on to the cot. Participants completed a training trial to familiarize themselves with the setup and movements. Each assessment for both legs comprised 3 tests, with 30-second predefined resting periods between the passive assessment, the concentric assessment at 60°/s, and the concentric/ eccentric assessment at 20°/s. 34,35 First, the flexibility of the calf muscle, including the Achilles tendon, was measured by moving the foot passively through its range of motion and recording the corresponding passive resistive torques (Nm) at 0°, 5°, 10°, 15°, 20°, and 25° of ankle dorsiflexion. All participants tolerated the dorsiflexion range of motion and reached 25° of ankle dorsiflexion. The relaxed state of the muscle was verified by manual testing. Subsequently, the participants concentrically plantar flexed and dorsiflexed their foot at maximum effort for 3 consecutive cycles at 60°/s from 20° of plantar flexion to 35° of dorsiflexion. Finally, the participants actively plantar flexed their foot (concentric plantar flexion contraction) and resisted dorsiflexion via the dynamometer (eccentric plantar flexion contraction) at 20°/s34,35 from 20° of plantar flexion to 35° of dorsiflexion with maximum effort for 3 consecutive cycles. The velocity of 20°/s was selected on the basis of our pilot study showing that patients with Achilles tendinopathy were able to produce the largest torque in eccentric plantar flexion contraction at this velocity (data not shown). The torque limit was set to 300 Nm and could not be reached by the partici-
journal of orthopaedic & sports physical therapy | volume 43 | number 11 | november 2013 | 797
43-11 Horstmann.indd 797
10/16/2013 4:54:28 PM
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
[ pants. Data were sampled at 1000 Hz and filtered using a sixth-order Butterworth filter with a cutoff frequency of 200 Hz. The maximum torque values (Nm) from filtered signals specific to each concentric plantar flexion, concentric dorsiflexion, and eccentric plantar flexion were extracted. The average of 3 contractions performed in each motion was computed and used to represent the participant’s performance.
Statistical Analysis All statistical tests were carried out in SPSS Statistics Version 19.0.0 (IBM Corporation, Armonk, NY). The normal distribution of continuous parameters was tested using Shapiro-Wilk tests. Changes in palpation pain, VAS pain assessment, and isokinetic force measurements between groups and between baseline and follow-up for normally distributed parameters were detected using repeated-measures analysis of variance, with group as the between-patient factor and time as the within-patient factor, and Scheffé post hoc tests. For nonnormally distributed parameters, all data were converted to ranks, and analyses of variance on ranks were used to detect differences between groups and between baseline and follow-up, as done for normally distributed data. Differences in sonographic assessments between groups were detected using Fisher exact tests, where clinical findings were categorized into “worsening or no changes” or “improvement” for each parameter. All tests were performed using “last score carried forward” as the statistical imputation method for handling missing data. Based on results of previous studies on a 12-week eccentric training in patients with Achilles tendinopathy, 3,45 we estimated that 15 patients were required to detect a significant difference in muscle strength and pain between the intervention and the control groups with a power of 80%, using an effect size of 0.5. The significance level for all statistical tests was set a priori to α = .05.
research report
]
Demographic Parameters of the Vibration-Training, Eccentric- Training, and Wait-and-See Groups
TABLE 2
Vibration Training (n = 23)
Eccentric Training (n = 19)
13/10
10/9
9/7
Age, y
46.0 6.9 (32-55)
45.7 8.5 (25-55)
44.4 7.7 (27-53)
Height, cm
175.8 7.4 (162-184)
Male/female, n
Wait and See (n = 16)
175.1 9.0 (162-195)
173.3 8.9 (160-191)
Body mass, kg
79.2 16.5 (50-111)
74.5 10.3 (62-107)
79.6 15.9 (58-110)
BMI, kg/m2
25.6 3.7 (18.6-32.1)
24.8 2.7 (20.8-29.4)
25.7 4.3 (20.3-33.8)
Abbreviation: BMI, body mass index. *Values are mean SD (range) unless otherwise indicated.
RESULTS Demographic and Clinical Parameters
P
atients in all 3 groups did not differ significantly in age, height, mass, or body mass index (TABLE 2). In addition, none of the parameters describing their clinical and training history differed significantly between groups. On average, patients had been training for 13.2 years and experienced pain for 3.7 years. Participants in the vibration group participated in 2.2 training sessions per week, and participants in the eccentric group completed 3.0 training sessions per week (P
research report
]
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
THOMAS HORSTMANN, MD1,2 • HOLGER M. JUD, MD3 • VANESSA FRÖHLICH, MD4 ANNEGRET MÜNDERMANN, PhD5 • STEFAN GRAU, PhD6
Whole-Body Vibration Versus Eccentric Training or a Wait-and-See Approach for Chronic Achilles Tendinopathy: A Randomized Clinical Trial
C
hronic Achilles affliction and tendinopathy are clinical conditions characterized by pain, swelling, altered joint motion, and limited performance,27 and are frequently observed not only in physically active but also in sedentary persons.23,41 However,
most frequently these conditions develop during recreational physical activity of individuals aged 35 to 45 years,2 with a TTSTUDY DESIGN: Randomized clinical trial. TTOBJECTIVES: To test the hypothesis that
whole-body vibration training results in greater improvements in symptoms and pain, structural changes, and muscle flexibility and strength of the triceps surae muscle-tendon unit than those achieved with eccentric training or with a wait-andsee approach.
TTBACKGROUND: The potential use of vibration
training for the treatment of Achilles tendinopathy has not been explored.
TTMETHODS: Fifty-eight patients (mean age, 46.0
years) with Achilles tendinopathy were randomly assigned to a 12-week intervention using wholebody vibration training, eccentric training, or a wait-and-see approach. Pain, tendon structure and path, and muscle flexibility and strength were assessed at baseline and follow-up, and compared using mixed-factor analyses of variance.
TTRESULTS: Pain improvements at the midsection of the tendon were greater in the vibration- and
higher prevalence in men than in women.23,29 Chronic Achilles tendinopathy can be caused by intrinsic factors (eg, genetic, eccentric-training groups than in the wait-and-see group (mean difference from the vibration-training group, –18.0; 95% confidence interval [CI]: –35.0, –1.1; mean difference from the eccentric-training group, –27.0; 95% CI: –50.9, –3.1). Improvements in pain at the musculotendinous junction were greater in the eccentric-training group than in the other groups (mean difference from the vibrationtraining group, –31.4; 95% CI: –60.7, –2.0; mean difference from the wait-and-see group, –50.2; 95% CI: –82.3, –18.1). Improvements in most participants were achieved in the vibration-training group, followed by the eccentric-training group.
TTCONCLUSION: Vibration training may be an
alternative or a complementary treatment in patients who do not respond well to eccentric training, especially in those with insertional pain.
TTLEVEL OF EVIDENCE: Therapy, level 2b–. J Orthop Sports Phys Ther 2013;43(11):794-803. Epub 9 September 2013. doi:10.2519/jospt.2013.4762
TTKEY WORDS: isokinetic muscle strength, muscle flexibility, pain, tendon structure, ultrasound
metabolic, or endocrine factors, as well as those of age, body size, anatomical misalignment of joint axes, reduced muscle flexibility, or muscle dysfunction or weakness23) and/or extrinsic factors (eg, previous injuries, inappropriate footwear, or environmental factors such as training on hard surfaces or a sudden increase in training loads5,18,23,41). Current therapies for chronic Achilles tendinopathy include prescription of foot orthoses, local or systemic antiinflammatory medication (eg, nonsteroidal antirheumatic agents or cortison) or topical nitroglycerin, stretching, massage therapy, electrotherapy, therapeutic ultrasound, extracorporeal shockwave therapy, or eccentric training.1 Most of these therapies are long-term therapies that are effective in about 3 of 4 patients. 37 Of these, eccentric training generally shows promising results in the treatment of chronic Achilles tendinopathy, reducing pain20 and the necessity of surgical intervention,2,45 and is effective in about 60% of patients.28,44 Despite its common use, to date, there is large variability in training protocols regarding dosage and duration of eccentric training.32 Moreover, it has been argued4 that the mechanism
Medical Park Bad Wiessee St Hubertus, Bad Wiessee, Germany. 2Faculty for Sport and Health Sciences, Technische Universität München, Munich, Germany. 3Department of Orthopaedic Surgery, Krankenhaus Bad Cannstatt, Stuttgart, Germany. 4Berufsgenossenschaftliche Unfallklinik, Tübingen, Germany. 5Osteoarthritis Research Center, University Hospital Basel, Basel, Switzerland; School of Physiotherapy, University of Otago, Dunedin, New Zealand; Division of Sport Science, University of Konstanz, Konstanz, Germany. 6 Department of Sports Medicine, Medical Clinic, University of Tübingen, Tübingen, Germany. This study was approved by the Institutional Ethics Board of the University of Tübingen. The authors certify that they have no affiliations with or financial involvement in any organization or entity with a direct financial interest in the subject matter or materials discussed in the article. Address correspondence to Dr Thomas Horstmann, Medical Park Bad Wiessee GmbH, St Hubertus, Sonnenfeldweg 29, 83707 Bad Wiessee, Germany. E-mail: [email protected] t Copyright ©2013 Journal of Orthopaedic & Sports Physical Therapy® 1
794 | november 2013 | volume 43 | number 11 | journal of orthopaedic & sports physical therapy
43-11 Horstmann.indd 794
10/16/2013 4:54:25 PM
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
of eccentric training that contributes to tendon healing may not be its strengthening effects but rather a modulation of the neurological stretch response. There has been increasing scientific scrutiny of whole-body vibration training in the context of muscle performance training and therapy for a range of medical conditions43,46 and in older adults.24 Improvements in neuromuscular performance with whole-body vibration training have been attributed to more synchronized motor units within muscles6 and increased spatial recruitment via potentiation of the neuromuscular system through muscle spindles, resulting in reflex activation of motor neurons.39 Short-term adaptation processes to whole-body vibration training include performance-enhancing effects on muscle, increased release of growth hormones and testosterone and reduced cortisol levels, and reduced resistance in peripheral vessels, resulting in improved nutrient supply to the muscles.16,17,26 Long-term adaptations include increased isometric and dynamic leg strength8,13 and bone density.47 Though the shortterm effects of whole-body vibration training on lower extremity flexibility and jumping power are controversial,7,11,21,42 in the long term, these parameters show improvements.9,11,42 Further, Hilgers et al15 recently demonstrated that whole-body vibration training increases walking performance in patients with multiple sclerosis, suggesting that vibration training may positively influence neuromotor control. Because the treatment of chronic Achilles tendinopathy aims at increasing ankle flexibility and plantar flexor muscle strength but may also require modulation of the neuromuscular system, wholebody vibration training may be effective in the therapy of this patient group. To date, the potential of using whole-body vibration training as a therapeutic intervention for chronic Achilles tendinopathy has not been explored scientifically. Therefore, the purpose of this study was to test the hypothesis that wholebody vibration training results in greater
TABLE 1
Inclusion and Exclusion Criteria Used in This Longitudinal Study
Inclusion criteria: • Age, 25-55 y • Recurrent complaints in 1 or both Achilles tendons at rest and/or during exercise for the preceding 6 mo • Structural changes of the tendon confirmed via sonographic examination during the initial physical exam • Weekly running mileage less than 50 km (aerobic exercise) • No increase in training intensity during the study • No supplemental treatment of the chronic Achilles tendinopathy during the study • Informed consent Exclusion criteria: • Treatment of the chronic Achilles tendinopathy within the preceding 4 wk, including physiotherapy, physical treatment, self-treatment (eg, heel pads), medication therapy (no steroidal antirheumatics, cortisone) • Complaints in the lower extremity other than those related to the chronic Achilles tendinopathy • Regular intake of medication that might have influence on the outcome • Residence too far from clinic • Pregnancy • Current inflammation of the musculoskeletal system • Fractures of the trained body parts within the past 12 mo • Surgery within the past 6 mo • Current discopathy • Rheumatoid arthritis • Osteoarthritis or arthropathies • Biliary calculus illness or kidney stones
improvements in symptoms and pain, structural changes, and muscle flexibility and strength of the triceps surae muscletendon unit compared to those achieved with established eccentric training or with a wait-and-see approach.
METHODS
R
ecreational runners with chronic Achilles tendinopathy were recruited by advertisement in local newspapers and the University of Tübingen website. Volunteers were initially contacted by telephone to determine their eligibility for participation, based on the inclusion and exclusion criteria listed in TABLE 1. If the volunteer was eligible for this study, an appointment for the initial exam was made. At the first visit, after providing written consent, subjects underwent a clinical examination completed by the lead physician (T.H.), and isokinetic measurements conducted by the junior physician (H.J.). Using a randomization list, the participants were assigned to a vibration-training group, a conventional eccentric-training group,
or a wait-and-see control group (FIGURE 1). Study personnel were blinded to the randomization list and did not know the group assignment of the newly enrolled patients until after their enrollment. Participants completed a 12-week program according to their group assignment, and the clinical exam and isokinetic measurements were repeated after 12 weeks. Both assessors were blinded to the participants’ group assignment. This study was approved prior to subject recruitment by the Institutional Ethics Board of the University of Tübingen.
Patients All patients had chronic Achilles tendinopathy at the midsection of the tendon, and some patients also reported pain at the osseous insertion or the musculotendinous junction. Both Achilles tendons of each participant were examined, but only the more affected side was included in the analysis.
Training Interventions During the 12-week intervention phase, 36 training sessions were scheduled
journal of orthopaedic & sports physical therapy | volume 43 | number 11 | november 2013 | 795
43-11 Horstmann.indd 795
10/16/2013 4:54:26 PM
[
research report
]
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
Volunteers with Achilles tendinopathy, n = 89 Volunteers met at least 1 exclusion criteria, n = 31 Volunteers met all inclusion criteria, n = 58 Volunteers declined to participate, n=0 Volunteers gave informed consent and were randomized into 3 intervention groups, n = 58
Vibration training, n = 23 Lost to follow-up, n = 1 Vibration training (completed final exam), n = 22
Eccentric training, n = 19
Wait and see, n = 16
Lost to follow-up, n = 1 Eccentric training (completed final exam), n = 18
Lost to follow-up, n = 2 Wait and see (completed final exam), n = 14
FIGURE 1. CONSORT flow chart of this longitudinal study.
for each participant in the vibrationtraining and eccentric-training groups. Participants were familiarized with the training equipment in the first training session. Each training session started and ended with a 5- to 10-minute warm-up and cool-down, respectively, on an ergometer or treadmill, and participants were supervised to ensure correct execution of the training exercises. During the 12-week intervention phase, participants were allowed to continue their regular activity schedule without increasing their training workload. A training diary was completed by the participant. The vibration training was conducted on a Galileo Fit (Novotec Medical GmbH, Pforzheim, Germany) and included a warm-up, a training phase, and a cooldown. During the 1-minute warm-up, participants changed consecutively between bipedal stance (neutral stance), unipedal stance, heel rises, and stepping in place, with their knees slightly bent, on the vibration platform set to a vibration frequency of 13 to 18 Hz and at
an amplitude of 0.4 to 0.6 mm. For the training phase, the vibration training was increased from 4 to 5 minutes in the first 4 weeks to 5 to 6 minutes in the second 4 weeks to 6 to 7 minutes in the last 4 weeks, and the vibration frequency and amplitude were increased to 16 to 21 Hz and 0.5 to 0.8 mm, respectively. Though vibration frequencies above 20 Hz are reported to elicit increased muscle performance38 and could have been used in our intervention, patients with Achilles tendinopathy in a pilot study perceived vibration frequencies above 21 Hz as uncomfortable and even painful (unpublished data). Participants intermittently changed between heel rises and heel drops (3 seconds in each position, with a 1-second transition between each position) at the edge of the platform until the onset of fatigue. Participants were allowed to briefly recuperate during heel-rise activities or alternate stretching of the calf muscles. During the 1-minute cool-down, participants relaxed their muscles by statically stretching, alternating between legs, with 1 leg on the
platform at a vibration frequency of 13 to 18 Hz. The eccentric training was conducted using a Reebok Step (Reebok International Ltd, Canton, MA) and included 2 exercises. First, participants stood upright (knee extended), with their entire body weight on the ball of 1 foot and their ankle in plantar flexion at the edge of the step. From this starting position, they slowly lowered their ankle until they reached full dorsiflexion (3-5 seconds).44 Subsequently, they used their contralateral limb and their arms to return to the starting position (1-2 seconds). Participants performed 3 sets of 15 repetitions on each leg.3 To increase training intensity between training periods and to ensure a training effect, participants who completed the 3 sets without any signs of fatigue performed a fourth set, and, if necessary, the load was increased by placing additional weights in a backpack attached to the participant’s back. Instead of completing training intervention sessions, participants in the
796 | november 2013 | volume 43 | number 11 | journal of orthopaedic & sports physical therapy
43-11 Horstmann.indd 796
10/16/2013 4:54:27 PM
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
wait-and-see group filled out an activity or training log to ensure that they maintained their recreational activities. In particular, they noted the time, duration, type, and intensity of their physical activities. After the 12-week control intervention period, participants in this group had the option of receiving a complementary gait analysis and/or participating in eccentric or vibration training.
Clinical Exam Participants were asked about the location of their pain (tendon insertion/ calcaneus, midsection of the tendon [2-6 cm proximal to the insertion], or musculotendinous junction) and the occurrence of their pain (time of first occurrence, sudden or gradual initial onset, or result of trauma). The type, time, and duration of previous treatment interventions were recorded (nonsteroidal antirheumatics, corticoids, and other medication; physical interventions [heat or cold treatment]; physiotherapy [also manual therapy]; foot orthoses; or other interventions). In addition, the type, intensity, duration, and frequency of their typical training regimen for all sports were recorded. The orthopaedic measures assessed in the clinical exam included static rearfoot and midfoot shape (normal, pes cavus, talipes valgus, or flatfoot), static calcaneal alignment (normal, varus, valgus), static forefoot alignment (normal, splayfoot, hallux valgus, or toe deformity), physiological foot correction while standing on the ball of the foot (present, absent), unilateral heel rise (painful, yes/ no), passive anterior translation of the talus (present, absent), range of motion of the ankle joint and the subtalar joint, leg length (leg-length discrepancy, yes/ no), leg alignment (normal, genu recurvatum, genu valgus, genu varum), and pelvic drop (present, absent).
Pain and Pain Impact Assessment The change in symptoms from preintervention to postintervention was assessed using a standard Likert scale (symptoms
completely recovered, much improved, little improved, unchanged, a little worse, or much worse).40 Visual analog scales (VASs) were used to assess the impact of pain on the categories “family and responsibility at home,” “recreation,” “social activities,” and “running training or other physical activities.” Each VAS ranged from “no limitation” (0 mm) to “complete limitation” (100 mm). To assess palpation pain for the Achilles tendon on both legs, the examining physician palpated the Achilles tendons between his thumb and index finger and pinched the tendon for a few seconds with the force that would be necessary to hold a 1-kg weight between the thumb and the index finger.45 Participants were asked to rate their pain on the VAS for the locations musculotendinous junction, midsection of the tendon, and osseous insertion from “no pain” (0 mm) to “excruciating pain” (100 mm).
Sonographic Assessment A sonographic exam of both Achilles tendons was conducted using a SONOLINE Versa Pro (Siemens AG, Munich, Germany). Sonographic images perpendicular to and along the Achilles tendon,40 close to the osseous insertion and 2 cm proximal to the osseous insertion, were examined for abnormalities, including inhomogeneity at tendon insertion, inhomogeneity in tendon path, swelling, bursitis, peritendinosis, tendinitis, and ossification,33 and documented in printouts. We defined a positive diagnosis as presence and a negative diagnosis as absence of any sonographic abnormalities.
Isokinetic Measurements Force measurements were taken with a computerized dynamometer training system (IsoMed 2000; D. & R. Ferstl GmbH, Hernau, Germany). Subjects wore the same standardized footwear (Air Pegasus; Nike, Inc, Beaverton, OR) for both the baseline and follow-up examination. The dynamometer was adjusted to assess ankle dorsiflexion/plantar flexion, with the axis set to 10 cm above the plinth
and the participant’s lower leg secured with a strap to immobilize the fully extended knee while the patient was in a supine position. The dynamometer was adjusted horizontally to the participant’s size by aligning the dynamometer axis to the ankle joint axis, and the participant’s foot was secured to the adapter. These settings were recorded to ensure that they were the same at baseline and at followup. Participants placed their extended arms along their trunk and held on to the cot. Participants completed a training trial to familiarize themselves with the setup and movements. Each assessment for both legs comprised 3 tests, with 30-second predefined resting periods between the passive assessment, the concentric assessment at 60°/s, and the concentric/ eccentric assessment at 20°/s. 34,35 First, the flexibility of the calf muscle, including the Achilles tendon, was measured by moving the foot passively through its range of motion and recording the corresponding passive resistive torques (Nm) at 0°, 5°, 10°, 15°, 20°, and 25° of ankle dorsiflexion. All participants tolerated the dorsiflexion range of motion and reached 25° of ankle dorsiflexion. The relaxed state of the muscle was verified by manual testing. Subsequently, the participants concentrically plantar flexed and dorsiflexed their foot at maximum effort for 3 consecutive cycles at 60°/s from 20° of plantar flexion to 35° of dorsiflexion. Finally, the participants actively plantar flexed their foot (concentric plantar flexion contraction) and resisted dorsiflexion via the dynamometer (eccentric plantar flexion contraction) at 20°/s34,35 from 20° of plantar flexion to 35° of dorsiflexion with maximum effort for 3 consecutive cycles. The velocity of 20°/s was selected on the basis of our pilot study showing that patients with Achilles tendinopathy were able to produce the largest torque in eccentric plantar flexion contraction at this velocity (data not shown). The torque limit was set to 300 Nm and could not be reached by the partici-
journal of orthopaedic & sports physical therapy | volume 43 | number 11 | november 2013 | 797
43-11 Horstmann.indd 797
10/16/2013 4:54:28 PM
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at UCSF Lib & CKM/RSCS Mgmt on August 29, 2014. For personal use only. No other uses without permission. Copyright © 2013 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
[ pants. Data were sampled at 1000 Hz and filtered using a sixth-order Butterworth filter with a cutoff frequency of 200 Hz. The maximum torque values (Nm) from filtered signals specific to each concentric plantar flexion, concentric dorsiflexion, and eccentric plantar flexion were extracted. The average of 3 contractions performed in each motion was computed and used to represent the participant’s performance.
Statistical Analysis All statistical tests were carried out in SPSS Statistics Version 19.0.0 (IBM Corporation, Armonk, NY). The normal distribution of continuous parameters was tested using Shapiro-Wilk tests. Changes in palpation pain, VAS pain assessment, and isokinetic force measurements between groups and between baseline and follow-up for normally distributed parameters were detected using repeated-measures analysis of variance, with group as the between-patient factor and time as the within-patient factor, and Scheffé post hoc tests. For nonnormally distributed parameters, all data were converted to ranks, and analyses of variance on ranks were used to detect differences between groups and between baseline and follow-up, as done for normally distributed data. Differences in sonographic assessments between groups were detected using Fisher exact tests, where clinical findings were categorized into “worsening or no changes” or “improvement” for each parameter. All tests were performed using “last score carried forward” as the statistical imputation method for handling missing data. Based on results of previous studies on a 12-week eccentric training in patients with Achilles tendinopathy, 3,45 we estimated that 15 patients were required to detect a significant difference in muscle strength and pain between the intervention and the control groups with a power of 80%, using an effect size of 0.5. The significance level for all statistical tests was set a priori to α = .05.
research report
]
Demographic Parameters of the Vibration-Training, Eccentric- Training, and Wait-and-See Groups
TABLE 2
Vibration Training (n = 23)
Eccentric Training (n = 19)
13/10
10/9
9/7
Age, y
46.0 6.9 (32-55)
45.7 8.5 (25-55)
44.4 7.7 (27-53)
Height, cm
175.8 7.4 (162-184)
Male/female, n
Wait and See (n = 16)
175.1 9.0 (162-195)
173.3 8.9 (160-191)
Body mass, kg
79.2 16.5 (50-111)
74.5 10.3 (62-107)
79.6 15.9 (58-110)
BMI, kg/m2
25.6 3.7 (18.6-32.1)
24.8 2.7 (20.8-29.4)
25.7 4.3 (20.3-33.8)
Abbreviation: BMI, body mass index. *Values are mean SD (range) unless otherwise indicated.
RESULTS Demographic and Clinical Parameters
P
atients in all 3 groups did not differ significantly in age, height, mass, or body mass index (TABLE 2). In addition, none of the parameters describing their clinical and training history differed significantly between groups. On average, patients had been training for 13.2 years and experienced pain for 3.7 years. Participants in the vibration group participated in 2.2 training sessions per week, and participants in the eccentric group completed 3.0 training sessions per week (P
