Secondary Logo

Journal Logo


Shock Wave Therapy for Chronic Achilles Tendon Pain

A Randomized Placebo-controlled Trial

Costa, M L FRCS (TR AND ORTH)*; Shepstone, L PHD; Donell, S T MD; Thomas, T L FRCS

Author Information
Clinical Orthopaedics and Related Research: November 2005 - Volume 440 - Issue - p 199-204
doi: 10.1097/01.blo.0000180451.03425.48
  • Free


Shock waves are high-energy sound waves. Shock wave therapy has been used for treatment of urinary tract stones for 30 years. During the last decade, interest in orthopaedic use of shock waves has increased. In Europe and parts of the United States, it has become a recognized treatment modality for certain soft tissue conditions, including plantar fasciitis, tennis elbow, and rotator cuff tendinopathy.3,4,7,8,10-13,15-17,20 However, shock wave therapy in the UK has been restricted to a few enthusiastic centers. The Cochrane Review of 2001 suggested that there was little evidence to support the use of any therapy for treatment of chronic Achilles pain.9

Given the encouraging, if equivocal results in other tendinopathies, we primarily sought to determine whether shock wave therapy relieves chronic Achilles tendon pain. Secondarily, we wondered if shock wave therapy improves a patient’s general health and lower limb function or alters a patient’s range-of-motion (ROM) at the ankle. What happens to the patient’s pain during the year after treatment? Does shock wave therapy cause tendon rupture or other complications?


We completed a randomized double-blind controlled study in which patients and the investigators recording the outcome measures were blinded to their allocated treatment (Fig 1). Patients were randomized to the treatment group or control group using computer-generated random numbers produced before the start of the trial. These were sealed in sequentially numbered opaque envelopes. The next envelope was opened after each patient consented to participation. All patients were referred by other physicians. They were enrolled from April to November 2001 and were followed up for a minimum of 1 year. Twenty-one patients were recruited in Norwich, and 28 were recruited in Colchester. All patients had tenderness exacerbated by dorsiflexion of the ankle. No patients were excluded because of their clinical findings. Three patients had pain at the insertion of the tendon. The remaining patients had midsubstance fusiform swelling consistent with underlying tendinosis. The eligibility criteria were kept as broad as possible to facilitate generalization of the results to other patient groups. Patients were included regardless of previous treatment or of any underlying degenerative joint disease. The only exclusion criteria were conditions that generally are considered contraindications to any form of shock wave therapy. The inclusion criteria were: older than 18 years with Achilles tendon pain present for at least 4 months. The exclusion criteria were: pregnancy, local malignancy, coagulopathy, or a pacemaker.

Fig 1.
Fig 1.:
The design of the trial with the numbers of initial patients and numbers of final patients in the two groups is shown.

An early trial using shock wave therapy for plantar fasciitis indicated a 50% reduction in visual analog pain scores.13 In the absence of previous comparable work on the Achilles tendon, this result was used to determine the sample size of the trial. With a power of 90%, and a significance of 5%, this equated to 20 patients in each group. We aimed to recruit 48 patients with chronic Achilles tendon pain, allowing for potential loss to followup. The local research ethics and research governance committees at the study hospitals approved the trial.

On the day of the trial, each patient had a detailed examination of their Achilles tendon and hindfoot to confirm the diagnosis and to exclude other conditions. Each patient then was given an information sheet and asked for informed consent to participate. Of the 50 patients invited, one declined because his pain had resolved almost entirely at that stage. This left 49 patients who consented to enter the trial.

The baseline characteristics of the treatment and the control groups were the same for gender, duration of pain before treatment, participation in impact sports, and the number of patients working manual jobs (Table 1). However, there was a 10-year difference in age, with the treatment group being a mean 10 years younger.

Table 1
Table 1:
Baseline Patient Characteristics

All but three patients had previous treatment for their Achilles pain. Most patients tried more than one therapy without success (Table 2). Pain scores were similar for pain at rest, pain on walking, and pain during sport participation, although the standard deviations were wide. As expected, the mean pain scores were progressively worse from rest to walking, with a peak during sport participation. Some patients had no pain when at rest, but most described at least an ache, particularly after having exercised during the day.

Table 2
Table 2:
Outcome Measures

Ranges of motion at the ankle before treatment were similar in the treatment and control groups, with a mean dorsiflexion of approximately 10° and plantar flexion of approximately 45°. There was much variation among individuals. Calf and tendon diameters also were similar.

For functional limitation, only 13 patients could maintain tiptoe for more than 10 seconds, reflecting the severity of the Achilles pain. Half of the patients in each group were able to do a tiptoe jump.

The system used in our study was the Storz Modulith® SLK (Storz AG, Kreuzlingen, Switzerland), an electromagnetic shock wave generator. It has a fine focal area 4 mm in diameter. The head of the shock wave unit contains a high-resolution ultrasound probe. This was used to delineate the area of maximum degeneration in the tendon, which also was correlated with the area of maximum tenderness as described by the patient. The fine focal area then could be directed accurately onto the tissue of interest. The energy flux density used was titrated to a maximum of 0.2 mJ/mm2 according to individual pain tolerance. This maximum level accounts for the dose-dependent effect of the shock wave.14 In a histologic study, Steinbach et al suggested that 0.3 mJ/mm2 is the lower threshold of severe vascular damage.22 No anesthetic was used. Each patient had 1500 shocks at each visit, and the shock wave treatment was repeated each month for 3 months. In the absence of evidence to guide the number and frequency of treatment, this protocol was based on the experience of the senior author (TLT).

Each therapy session was done by the same two fully trained radiographers employed by the company supplying the machine (Impact Medical Ltd, London, UK). The patient was asked to lie prone on a couch, and the head of the shock wave probe was lowered onto the patient’s tendon. A thin layer of ultrasound gel was applied over the skin. As previously described, the treatment area then was identified using the patient’s tenderness and ultrasound as a guide.

The control group had the same treatment given by the same radiographers with the same machine at the same settings. However, bubble wrap covered in an opaque cloth was interposed between the head of the machine and the tendon. This created an air gap between the tendon and the head that dissipated the shock waves completely. The technique was used by Rompe et al to produce a placebo shock wave.13

Immediately after randomization, the patients completed a 100-mm visual analog pain scale (VAS) for pain at rest, pain during walking, and pain during sport participation (100 = worst pain imaginable, 0 = no pain at all). They also completed a baseline set of questionnaires. The first was the Functional Index of Lower Limb Activity (FIL), which is a validated lower limb-specific activity questionnaire.19 The other was the EuroQol (EQol) generalized health status questionnaire.1 The EQol is subdivided into two parts: the EQ-5D, based on five daily activities of living, and the health score, which is a summary score of general health.

Clinical assessments were made at the same time by one of four physiotherapists who were blind to the treatment allocation. The patients knelt on a chair and were examined with the knee at 90° flexion. Measurements included the maximum calf circumference and maximum tendon diameter measured with tape measure, ankle ROM measured with a goniometer, and the ability to do a single-leg heel raise and a single-leg tiptoe jump. Range of motion at the ankle was used as an indirect measurement of calf unit length. Increased dorsiflexion suggested increased length, and reduced dorsiflexion suggested reduced length. Maximum calf diameter reflected the size of the gastrocnemius and soleus muscles that power the calf unit. These measurements did not assess the quality of the muscle, but could be used to detect gross muscle wasting. The tendon diameter measurements were used to monitor the response of the tendon. Typically, patients with chronic Achilles tendon pain present with fusiform swelling reflecting underlying tendinosis. The single-leg heel raises and jumps were used as a guide to calf unit function. This was affected by pain. At each treatment session, complications, including local bruising and palpable tendon defects were documented.

The VAS, questionnaires, and clinical assessment were repeated for each patient before each monthly treatment. They then were repeated at the primary outcome point 3 months after the first application of shock wave therapy (ie, 1 month after the final treatment). Finally, the VAS and questionnaires were sent to the patients 1 year after enrollment. The primary outcome measure was a change in VAS score for pain on walking between the baseline and 3-month scores. The secondary outcome measures at 3 months included changes in VAS for pain at rest and for pain during sporting activity. We also recorded changes in ankle ROM, calf muscle circumference, tendon diameter, FIL, and EQol.

The VAS, FIL, EQol data, and changes in clinical assessments were analyzed using a predetermined two-sample t test or Mann-Whitney U-test as appropriate. Confidence intervals accompanied all hypothesis testing. If patients did not supply data at 3 months, the data set from their last attendance was used for comparison with the baseline. A p value of 0.05 was considered significant.


During the 1-year study, six patients were lost to followup. Nine patients did not attend followup at 3 months. Two patients sustained tendon injuries but continued to be followed up for 1 year. One patient said that his pain had completely disappeared after the first two treatments and he did not want to take time off work. The other patients did not give a reason for not attending. This may reflect the fact that no treatment session was offered at this time. For these nine patients, outcome analysis was completed using the last set of data provided by each patient. All of the patients attended for at least two treatment sessions.

Neither the VAS pain scores for pain at rest or during sport participation showed any difference between the groups (Table 2). The baseline score for pain during walking was 55 for both groups. The score after the intervention was 34 points in the treatment group and 50 in the control group (p = 0.127; CI −4.7-36.2).

There were no differences between the groups in ROM at the ankle or differences in the FIL or EQol scores (Table 2).

When both groups were combined, the 1-year followup data showed a mean pain on walking score of 30 points. Only 13 of the 41 patients who provided 1-year pain scores were pain free. Similarly, there was little change in the mean pain scores at rest, which were 35 pretreatment, 37 at 1 month, 41 at 2 months, 33 at 3 months, and 22 at 1 year. In contrast, there was a reduction in pain during sport participation during the year. The results were 65 pretreatment, 62 at 1 month, 57 at 2 months, and 47 at 3 months. The 1-year mean declined to 31 points.

During the trial, two patients sustained a rupture of the Achilles tendon. Both patients were in the treatment group and were 62 years and 65 years of age, respectively. Both ruptures occurred within 2 weeks of their first treatment. The first patient stumbled down a step in her back garden, and the second injured herself stepping out of the back of an ambulance. Each patient elected nonoperative treatment for their injury, and both made an unremarkable recovery. One of the patients had complete resolution of symptoms at the end of her rehabilitation. The other patient had persistent weakness on the affected side and had aching related to activity.

There were no minor complications. In particular, there was no local bruising after application of the shock waves. The majority of patients reported aching in the calf in the 48 hours after treatment, although this was divided evenly between the treatment and control groups.


The mechanism of action of shock waves is not fully understood. Some authors have suggested that it has a centrally modulated effect on nociception via stimulation of peripheral nerves (the gating theory).18 In vitro studies have confirmed that energy densities as low as 0.12 mJ/mm2 produce intracellular damage resulting in increased membrane permeability.23 This may cause nerve depolarization. It also has been shown to affect other cells such as neutrophils.6 Local cell damage leads to a phased, cytokine-mediated increased vascularity.5 There also is histologic evidence of the direct effect of shock waves on vascular endothelium.21 The most recent review combines much of this work, suggesting that there may be a short-term analgesic effect followed by a longer-term microtrauma-mediated healing response.11

Much of the early clinical work investigating the use of shock wave therapy came from centers in Europe, and little of it relates directly to the Achilles tendon. There is a small case series in the German literature in which it is reported that 70% of 31 patients with resistant Achilles pain had some benefit after treatment with shock waves.24 However, a mixed case series in which it is reported that less than ½ of 20 patients with Achilles tendinosis had improvement with shock wave treatment.2

Our primary outcome showed no treatment effect from shock wave therapy. This interpretation is supported by lack of differences in the secondary outcomes. However, the confidence intervals associated with the outcome measures are wide and could include a clinically relevant treatment effect. This raises the possibility of a Type II error (ie, the sample size was too small to detect a meaningful treatment effect). In retrospect, the assumptions on which the sample size calculation was based were flawed. The anticipated treatment effect was derived from a study of plantar fasciitis done by advocates of shock wave therapy.13 The 50% difference in outcomes between shock waves and placebo they found was perhaps unrealistically extrapolated to this trial. Based on our results, 170 patients would be required to provide 80% power. It could be argued that the trial should be expanded to increase the sample size.

Rupture of the Achilles tendon is a serious injury requiring prolonged rehabilitation. Both of the ruptures during the trial occurred in patients in the treatment group, and both ruptures occurred within a few days of the first application of shock waves. However, the patients were older than 60 years and had advanced tendinosis. Both injuries could have occurred by chance in these patients. Although the ethics committee allowed our trial to continue after the report of these complications, caution is advocated for treating older patients with shock wave therapy.

The 1-year followup data showed only a small reduction in pain at rest and during walking. This suggests that the symptoms associated with painful tendinosis are chronic, as all of the patients previously had tried other treatment regimes without success. The 1-year data for pain during sport participation did show a reduction with time. However, this may reflect the tendency for patients to reduce their sporting activity in response to chronic pain.

Results of our study do not provide any evidence for use of shock wave therapy for treatment of chronic Achilles tendon pain. However, the confidence intervals do include a potentially clinically relevant treatment effect. Complications in the treatment group included two tendon ruptures, suggesting caution in treating older patients with shock wave therapy.


We thank Kate MacMillan and Debbie Halliday in Norwich and Chris Carr and Sophie Florence in Colchester for help with the clinical assessments; Pauline Ewart in Colchester and Kathleen Ambler, Adele Cooper, and Xanthe Scott in Norwich for administrative support; and Clare Darrah, Research Coordinator at The Institute of Orthopaedics in Norwich, for proofreading the original protocol for the study.


1. Brooks R: EuroQol: The current state of play. Health Policy (New York) 37:53-72, 1996.
2. Brunner W, Thuringer R, Ascher G, et al: Extracorporeal shock wave therapy in orthopaedics: Results of 443 cases three months after treatment. Orthop Prax 33:461-464, 1997.
3. Crowther MA, Bannister GC, Huma H, Rooker GD: A prospective randomised study to compare extracorporeal shock-wave therapy and injection of steroid for the treatment of tennis elbow. J Bone Joint Surg 84B:678-679, 2002.
4. Haake M, Buch M, Schoellner C, et al: Extracorporeal shock wave therapy for plantar fasciitis: Randomised controlled multicentre study. BMJ 327:75-79, 2003.
5. Haupt G: Use of extracorporeal shock waves in the treatment of pseudarthosis, tendinopathy and other orthopaedic diseases. J Urol 158:4-11, 1997.
6. Holmes RP, Yeaman LD, Taylor RG, McCullough DL: Altered neutrophil permeability following shock wave exposure in vitro. J Urol 147:733-737, 1992.
7. Loew M, Jurgowski W, Mau HC, Thomsen M: Treatment of calcifying tendinitis of rotator cuff by extracorporeal shock waves: A preliminary report. J Shoulder Elbow Surg 4:101-106, 1995.
8. Mandalia V, Thomas TL, Chari R: The analgesic effect of extracorporeal shock wave therapy (ESWT) on tennis elbow, golfer’s elbow and plantar fasciitis: A preliminary report. J Orthop Med 24:13-16, 2002.
9. McLauchlan GJ, Handoll HH: Interventions for treating acute and chronic Achilles tendinitis. Cochrane Database Syst Rev 2:CD000232, 2001.
10. Ogden JA, Alvarez R, Levitt R, Cross GL, Marlow M: Shock wave therapy for chronic proximal plantar fasciitis. Clin Orthop Relat Res 387:47-59, 2001.
11. Ogden JA, Alvarez RG, Levitt R, Marlow M: Shock wave therapy (Orthotripsy) in musculoskeletal disorders. Clin Orthop Relat Res 387:22-40, 2001.
12. Rompe JD, Hopf C, Kullmer K, et al: Low-energy extracorporeal shock wave therapy for persistent tennis elbow. Int Orthop 20:23-27, 1996.
13. Rompe JD, Hopf C, Nafe B, Burger R: Low-energy extracorporeal shock wave therapy for the painful heel: A prospective controlled single-blind study. Arch Orthop Trauma Surg 115:75-79, 1996.
14. Rompe JD, Kirkpatrick CJ, Kullmer K, Schwitalle M, Krischek O: Dose-related effects of shock waves on rabbit tendo Achillis: A sonographic and histological study. J Bone Joint Surg 80B:546-552, 1998.
15. Rompe JD, Kullmer K, Riehle HM, et al: Effectiveness of low-energy extracorporeal shock waves for chronic plantar fasciitis. Foot Ankle Surg 2:215-221, 1996.
16. Rompe JD, Rumler F, Hopf C, Nafe B, Heine J: Extracorporeal shock wave therapy for calcifying tendinitis of the shoulder. Clin Orthop Relat Res 321:196-201, 1995.
17. Rompe JD, Schoellner C, Nafe B: Evaluation of low-energy extracorporeal shock-wave application for treatment of chronic plantar fasciitis. J Bone Joint Surg 84A:335-341, 2002.
18. Russo S, Gigliotti S, de Durante C: Shock Wave Therapy in Orthopaedics. Naples, Giuseppe de Nicola Editore 1-6, 1996.
19. Salen BA, Spangfort EV, Nygren AL, Nordemar R: The Disability Rating Index: An instrument for the assessment of disability in clinical settings. J Clin Epidemiol 47:1423-1434, 1994.
20. Schmitt J, Haake M, Tosch A, et al: Low-energy shock-wave treatment (ESWT) for tendinitis of the supraspinatus: A prospective, randomised study. J Bone Joint Surg 83B:873-876, 2001.
21. Seidl M, Steinbach P, Worle K, Hofstadter F: Induction of stress fibres and intercellular gaps in human vascular endothelium by shock-waves. Ultrasonics 32:397-400, 1994.
22. Steinbach P, Hofstadter F, Nicolai H, et al: Determination of the energy-dependent extent of vascular damage caused by high-energy shock waves in an umbilical cord model. Urol Res 21:279-282, 1993.
23. Steinbach P, Hofstadter F, Nicolai H, Rossler W, Wieland W: In vitro investigations on cellular damage induced by high energy shock waves. Ultrasound Med Biol 18:691-699, 1992.
24. von Perlick L, Boxberg W, Glebel G: Hochenergetische stoβwellenhandlung der therapieresisteten Achillodynie. Orthop Prax 34:558-561, 1998.
© 2005 Lippincott Williams & Wilkins, Inc.