Randomized, placebo-controlled, double-blind clinical trial evaluating the treatment of plantar fasciitis with an extracoporeal shockwave therapy (eswt) device: a north american confirmatory study

Randomized, Placebo-Controlled, Double-Blind ClinicalTrial Evaluating the Treatment of Plantar Fasciitis with anExtracorporeal Shockwave Therapy (ESWT) Device:A North American Confirmatory Study Patricia Kudo,1 Katie Dainty,1 Michael Clarfield,2 Larry Coughlin,3 Pauline Lavoie,4 Constance Lebrun1 1Fowler Kennedy Sport Medicine Clinic, 3M Centre University of Western Ontario, London, Ontario N6A 3K7, Canada 2The Sports Medicine Specialists, 150 Eglinton Avenue East, Toronto, Ontario M4P 1E8, Canada 3West Island Orthopedic & Sports Medicine Center, 3881 Boulevard St-Jean Dollard-des-Ormeaux,Quebec H9G 2V1, Canada 4Action Sport Physio, 1451 rue Montarville, St-Bruno, Quebec J3V 3T6, Canada Received 2 December 2004; accepted 16 May 2005 Published online 18 November 2005 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jor.20008 ABSTRACT: Despite numerous publications and clinical trials, the results of treatment ofrecalcitrant chronic plantar fasciitis with extracorporeal shockwave therapy (ESWT) still remainequivocal as to whether or not this treatment provides relief from the pain associated with thiscondition. The objective of this study was to determine whether extracorporeal shock wave therapycan safely and effectively relieve the pain associated with chronic plantar fasciitis compared toplacebo treatment, as demonstrated by pain with walking in the morning. This was set in amulticenter, randomized, placebo-controlled, double-blind, confirmatory clinical study undertakenin four outpatient orthopedic clinics. The patients, 114 adult subjects with chronic plantar fasciitis,recalcitrant to conservative therapies for at least 6 months, were randomized to two groups.
Treatment consisted of approximately 3,800 total shock waves (Æ10) reaching an approximated totalenergy delivery of 1,300 mJ/mm2 (EDþ) in a single session versus placebo treatment. This studydemonstrated a statistically significant difference between treatment groups in the change frombaseline to 3 months in the primary efficacy outcome of pain during the first few minutes of walkingmeasured by a visual analog scale. There was also a statistically significant difference betweentreatments in the number of participants whose changes in Visual Analog Scale scores met the studydefinition of success at both 6 weeks and 3 months posttreatment; and between treatment groups inthe change from baseline to 3 months posttreatment in the Roles and Maudsley Score. The results ofthis study confirm that ESWT administered with the Dornier Epos Ultra is a safe and effectivetreatment for recalcitrant plantar fasciitis. ß 2005 Orthopaedic Research Society. Published byWiley Periodicals, Inc. J Orthop Res 24:115–123, 2006Keywords: plantar fasciitis; shock wave therapy; heel pain; ESWT tripsy (ESWL) has been well established for over20 years for the treatment of urologic conditions,1 In the past 20 years extracorporeal shock waves and more recently, there has been significant have been used to safely and effectively treat a interest in orthopedic applications such as non- number of medical conditions. Shock wave litho- union fractures and several types of tendonopa-thies. Despite numerous publications and clinical Correspondence to: Katie Dainty (Telephone: 416-480-6100, trials, one orthopedic application of ESWT, which still remains highly equivocal, is the treatment of ß 2005 Orthopaedic Research Society. Published by Wiley Periodicals,Inc.
JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2006 Plantar fasciitis is defined as a tensile overload Two ESWT devices have now gained approval of the plantar fascia at its origin on the medial from the Food and Drug Administration for the tubercle of the calcaneus.15 The plantar fascia is a treatment of recalcitrant plantar fasciitis in thick fibrous tissue on the bottom of the foot that the United States; however, the evidence is still protects sensitive plantar structures such as divergent. The significant differences in the nerves, vessels, muscles, and tendons, and in results of the various studies may be explained addition, is responsible for maintaining the plan- by a number of factors including technical differ- tar arch. The symptoms usually start as a dull ences (machine design, shock intensity and fre- intermittent pain that most often progresses to a quency, and the use of different forms of placebo sharp persistent pain. The patient typically suffers treatment), as well as differences in subject pain with the first steps in the morning or after populations, severity of disease, and study design.
period of prolonged sitting. This pain is aggravated This highlights the need for further investigation by continuous weight bearing, and becomes pro- using solid randomized prospective and confirma- gressively more severe. Its onset is insidious, and tory clinical trials. To further enhance the results not always associated with a specific incident or shown in the first study using the Dornier Epos trauma. Standard care at present is conservative Ultra, the present study was designed as a con- treatment, but about 10% of patients fail to firmatory evidence trial to assess the safety and respond or heal spontaneously.3 This extremely effectiveness of the Dornier Epos Ultra in the painful condition has been reported to effect up to treatment of pain associated with chronic plantar 20% of the general population over their lifetime,4 and is responsible for approximately 1 millionpatient visits per year in the United States.5 In a review of the current published literature on the use of shockwave therapy for the treatmentof plantar fasciitis, several clinical trials were This was a multicenter, randomized, placebo-con-trolled, prospective, double-blind, confirmatory clinical found. Among a plethora of nonrandomized pub- study with two groups: one group receiving ESWT with lications, there are only six placebo-controlled the Epos Ultra (Active group) and a Control group trials.6–11 all of which have reported extremely receiving placebo treatment. The objective of the study variable results. A meta-analysis done by Ogden was to determine whether ESWT could safely and et al. in 200212 found that those published studies effectively relieve the pain associated with chronic that fulfilled the criteria for acceptable methodol- plantar fasciitis compared to placebo treatment with a ogy with sufficient duration did show that directed single high-energy treatment, as demonstrated by relief application of shockwaves to the origin of the of pain with the first few minutes of walking in the plantar fascia is a safe and effective nonsurgical method for treating chronic, recalcitrant heel pain The initial sample size calculation was based on the syndrome.4 However, recent studies such as those primary efficacy outcome, defined as the differencebetween the Active Epos treatment and the Placebo by Buchbinder et al.,7 Haake et al.,8 and Speed treatment measured by the change from baseline to et al.10 have reported no statistically significant 3 months in the Visual Analog Scale (VAS) score for pain differences in the degree of improvement between while walking for the first few minutes in the morning ( p ¼ 0.05). The expected effect size of the primary Of those trials that reported a positive outcome, outcome was estimated from the treatment difference shock wave therapy for the treatment of plantar and standard deviation of 1.4 and 3.0, respectively, fasciitis was shown to be most efficacious with found in the original pivotal U.S. clinical study11 should a single therapy session. A pivotal study approv- be #13. The calculation was done using Statistical ed by the Food and Drug Administration in Solutions nQuery Advisor1 Release 3.0 software, and 2002, showed that the Dornier Epos Ultra shock- was adjusted by 15% to account for attrition rates.
wave device could safely produce clinical improve- Secondary efficacy outcomes included change scores forthe American Orthopedic Foot and Ankle Society ment in chronic plantar fasciitis using a single (AOFAS) ankle–hindfoot scale score16 (pain and range therapeutic session.11 The Active treatment of motion domains), the Roles and Maudsley Score2 group in this trial reported 56% success and the (a four-point patient self-assessment of pain and limita- control group reported 47% success at 3 months tions of activity), the SF 12 Global Health Rating posttreatment. Other publications from all over Scale,14 and pain on palpation (point of tenderness) as the world have shown success rates as high as measured with a pressure threshold meter (PTM, Pain Diagnostics and Thermography, Great Neck, NY). A JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2006 CONFIRMATORY RCT OF ESWT FOR PLANTAR FASCIITIS primary safety analysis was also done comparing the tion. Absence of a calcaneal fracture, bony abnormality, incidence of adverse events between groups at the time or other pathology (i.e., tumors) was confirmed with a of treatment, and during follow-up. All follow-up data lateral radiograph prior to treatment.
was measured by independent research investigators ateach site, all of whom were blinded to the randomization The randomization scheme was generated by BiostatInternational, Inc., Tampa, Florida. Sealed, opaque, tamper-proof envelopes containing individual randomi-zation assignments were provided to each investiga- The study was conducted at four centers throughout tional site prior to the beginning of the study. Subjects Canada. Each site obtained approval from an institu- were randomized by the treating investigator just prior tional ethics review board review board prior to begin- to the beginning of treatment. The first subject was ning the study. An Investigational Testing Authorization randomized in November 2000, and the last subject was from the Therapeutic Products Programme Division of Health Canada was also granted. Subjects were recruit-ed through outpatient clinics at each of the study sites.
All coinvestigators were primary care, sport medicinephysicians or orthopedic specialists and were trained on All procedures were performed in outpatient settings treatment with the Dornier Epos Ultra extracorporeal using a single treatment method with the Dornier shockwave system prior to the study. All potential Epos Ultra extracorporeal shockwave therapy system subjects were assessed according to the inclusion and (Dornier MedTech Systems, GmbH, Germany). The exclusion criteria (Fig. 1) in the study protocol and subjects were placed either prone (44.7% of subjects) or signed informed consent prior to their baseline evalua- on their side (55.3% of subjects) on the examination table JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2006 with the study foot placed in a supported position. Choice 3 months posttreatment. Unblinding occurred at the of position was based on patient comfort. Prior to shock 3-month visit. Subjects who received Active treatment wave exposure, the area of pain was marked with an X continued in the study and were evaluated at 6 and on the skin to assist in focusing the delivery of the shock 12 months posttreatment. Subjects who originally waves, and all study subjects, including the Placebo received placebo treatment and whose symptoms were group, were given a medial calcaneal nerve block using still significant according to specified inclusion criteria 5 mL of 1% Xylocaine, 15–20 min prior to the procedure.
were offered to ‘‘crossover,’’ and receive Active treat- The therapy head was coupled tangentially on the ment with the Epos Ultra after their 3 month follow-up medial aspect of the foot, and ultrasound localization visit. Subjects originally randomized to the Placebo was used for positioning of the focal area.
group who elected not to cross over at 3 months were The Active treatment session was performed using discontinued from study follow-up. All subjects were the energy levels indicated in Table 1. The energy given a pain medication diary with instructions during parameter was 0.36 mJ/mm2 (EDþ), which is equivalent screening and at each follow-up visit. Entries were to 0.64 mJ/mm2 (ED). Shock wave frequency began at 60 made by the subject for any alternative medication shocks/min, and was increased in increments of 30 taken between follow-up visits (i.e., Tylenol for a shocks/min. During treatment, the frequency of release headache). All subjects underwent a physical examina- of the shock waves began at 60 shocks/min at level 1, and tion including a pressure threshold measurement and was increased by one level of 30 shocks/min at each were asked to assess their pain using a VAS for various energy level until 240 shocks/min were reached at level activities of daily living, and complete the Roles and 7. Fifty (Æ10) shocks were delivered at levels 1–6 as the Maudsley Pain questionnaire, the AOFAS ankle–hind- frequency was being increased. Approximately 3,500 foot scale, and the SF-12 Global Health Rating Scale (Æ10) shock waves were administered at level 7 to reach before treatment and at follow-up visits. Adverse events an approximated total energy delivery of 1,300 mJ/mm2 were evaluated by the type, nature, severity, and (EDþ) or 2,330 mJ/mm2 (ED) (3,800 total shocks).
intensity during treatment and at each follow-up visit.
The Placebo group received the identical treatment The last follow-up visit for the primary efficacy end- procedure; however, shock waves were prevented from entering the subject’s foot by a thin foam cushion placedon the therapy head with an application of ultrasound gel. The cushion was put in place prior to the subject’sarrival in the treatment room to maintain blinding. A All statistical analyses were performed using the SAS1 new cushion was used with each treatment session.
System (Cary, NC), with a significance level of 0.05 and All treatments were performed according to instruc- on an intention-to-treat basis. The primary analysis tions in the Epos Ultra Operating Manual. Pain method was a two-sample t-test comparing treatment intensity during treatment and immediately posttreat- groups in the changes from baseline to 3 months post- ment were recorded, as well as any adverse events treatment. Statistical testing also included a repeated resulting during the treatment session. After treatment measures analysis of the changes from baseline, testing and at each follow-up visit, blinding was assessed by for treatment and time main effects, and treatment by asking subjects to identify which treatment they interaction effects, with relevant covariates, such as believed they received. All subjects were instructed to baseline VAS score, included in the model. The effect of eliminate athletic activities and pain medication post- missing data on efficacy results was determined prior to therapy until the 6 week follow-up evaluation.
analysis. All follow-up, evaluations were included in theanalysis out to 3 months, prior to treatment unblinding.
Investigational site effects on the changes in pain score at 3 months were tested for significance in a two-way All subjects were evaluated by an independent analysis of variance. To reduce the size of the residual (blinded) investigator at 3–5 days, 6 weeks, and error term used in making inferences on treatment JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2006 CONFIRMATORY RCT OF ESWT FOR PLANTAR FASCIITIS effect at 3 months, analysis of covariance was employed group missed the 6-week visit. One subject in the to investigate linear effects of baseline characteristics, Active group and two subjects in the Placebo for example, pain, age, or weight. In addition to eva- group missed the 3-month follow-up visit. Rea- luating the actual changes in pain score, the proportion sons for withdrawal are included in Figure 2. At of subjects achieving at least 60% improvement in pain 3 months, 53 of the 58 subjects from the Active was compared between treatment groups at 3 months treatment group and 52 of the 56 subjects from using a chi-square test. Proportions of subjects experi- the Placebo group were evaluated (92%).
encing adverse events were also compared betweentreatment groups via Fisher’s Exact tests, whereby the column totals (denominators) were the total number of respect to baseline demographics such as age, gender, height, weight, duration of symptoms,pain on VAS, and characteristics of physicalinspection (Table 2). There were no significant differences between groups in the baseline data forprevious therapies tried.
One hundred fourteen study participants were Significant differences were found between randomly assigned to either the Active treatment groups on outcomes measured during treatment group (58) or the Placebo control group (56). Two including pain and verification of blinding. Forty- participants in the Active group and two partici- six of the 58 participants in the Active group pants in the Placebo group withdrew after the reported pain during treatment compared to five follow-up visit at 3–5 days. Two subjects in the in the Placebo group ( p < 0.0001). There was no Active group withdrew from the study after significant difference between groups with regard the visit at 6 weeks and one subject in the Placebo to pain reported immediately after treatment.
JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2006 Participate in weekly exercise (No. of pts) When comparing verification of blinding data, tently numerically superior to placebo with mar- 34 participants in the Active group believed they ginal statistical significance ( p < 0.10) between had received the ESWT treatment when ques- treatments in these clinical outcomes. Success, tioned immediately posttreatment versus only defined as a score of none or mild on the pain 13 in the Placebo group ( p ¼ 0.0007). Twenty-two portion of the AOFAS ankle–hindfoot scale participants in the Active group (37.9%) and was also numerically superior to placebo with 33 (58.9%) in the Placebo group reported they did marginal statistical significance at 3 months not know whether they had received the treatment or not. This was not statistically significant.
In terms of the Secondary Outcomes measures, With regard to the primary outcome measure, a no significant difference between groups was statistically significant difference ( p ¼ 0.0124) found with the numbers available in any of the was found in the change from baseline to 3 months AOFAS ankle–hindfoot indices (Table 3) or the in the VAS scores of the treated versus Placebo SF-12 Global Health Rating Scale. However, a group (Table 3). In the Active treatment group, the significant difference between groups was deter- mean pain score decreased from 7.5 to 3.9 at 3 mined on the Roles and Maudsley scores ( p ¼ months ( p < 0.0001), resulting in a mean percen- 0.0121) using a Cochran–Mantel–Haenszel mean tage improvement of 49.1%. In the Placebo group, score test and the pain measurement on palpation the mean pain score decreased from 7.9 to 5.3 ( p ¼ 0.0027) using a two-way ANOVA F-test for at 3 months ( p < 0.0001), a mean percentage group effect at 3 months posttreatment (Table 3).
Adverse events (other than pain) reported Clinical success was defined as >60% improve- during treatment or in the first 3–5 days after ment from baseline in VAS scores for pain during treatment were relatively few, and there was no the first few minutes of walking. Table 3 shows significant difference in number of side effects that at 3 months after treatment, there was a reported between groups through 3 months. The statistically significant difference between the adverse events reported were primarily antici- percentage of Active treatment and Placebo treat- pated and included ecchymosis, edema, pain, and ment subjects that met the above definition of a transient parasthesias. There was one report of success. In the Active group, 47% (25 of 53) of the low back pain in the Active group and one of subjects achieved greater than 60% improvement pruritis in the Placebo group. Both were deemed in pain, and in the Placebo group only 23% (12 of unrelated to the study intervention.
52) met the same criteria ( p ¼ 0.0099). Although After 3–5 days and through 3 months post- both Placebo and Active groups also reported treatment, one participant in the Placebo group significant improvement in their pain with normal sustained an accidental injury, which led to activity, leisure/sport activity, and prior to bed, the increased pain in the study foot, and one had gene- improvements in the Active group were consis- ralized spasms in the study foot following activity.
JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2006 CONFIRMATORY RCT OF ESWT FOR PLANTAR FASCIITIS Results for Primary and Secondary Outcome Measures Primary outcome measurePain during the first few minutes of walking scored on VAS Defined by >60% improvement on the primary aDefined as score of none or mild on the pain domain.
bCochran-Mantel-Haenszel mean score test.
The onset of pain and edema during this period visit (Table 3). Statistical analysis is pending for were comparable between treatment groups. One the 6- and 12-month follow-up (Active group) and subject in the Active Group experienced tingling in Crossover safety and efficacy data.
the affected foot at the 6-week follow-up visit. Theevent was coded as anticipated/not serious and resolved by the 3-month visit. One subject in theActive Group experienced peripheral neuritis at This study demonstrated a statistically signifi- the 6-week visit. The event was coded as antici- cant difference between groups in the primary pated/not serious and resolved prior to the 3 month outcome measure of change from baseline to JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2006 3 months after treatment in VAS pain scores in results. Our results are only valid for the ther- the first few minutes of walking (49.1% vs. 33.3%; apeutic variables used in this study. It is difficult p ¼ 0.0124). Although improvement was noted in to compare studies, which use different patient the Placebo group, this phenomenon could simply populations, energy sources, and treatment proto- reflect the spontaneous remission or natural his- cols. It is unclear if the negative results of other tory of plantar fasciitis as a self-limiting condition studies are due to insufficient energy levels, or a sustained placebo effect. Standard treatment possible over treatment, which can produce a lack for plantar fasciitis is conservative, but about 10% of/or negative biologic effect, or inclusion of sub- of patients fail to respond or heal spontaeously.3 jects who might not benefit from ESWT. The Because this represents a significant number of results presented here confirm those of the pre- people, we consider our findings about the effect vious randomized controlled trial performed as of ESWT as an alternative treatment are quite part of the initial study in which some of the same relevant and useful. There were also statistically significant differences between treatments in the ESWT has several advantages and should be number of participants whose changes in VAS considered an effective and safe tool in the treat- scores met the study definition of success and in ment of chronic plantar fasciitis. As an alternative the distribution of Roles and Maudsley pain and to surgery, it is a noninvasive technology, which activity self-assessment scores. This provides has considerably less complications. It has a further evidence that ESWT does offer an addi- relatively short recovery time during which the tional benefit with regard to pain and activity patient can continue with most employment and levels to at least 3 months posttreatment. The activities of daily living, as soon as the day Roles and Maudsley score is considered to be following treatment. Finally, because ESWT can clinically significant for providing patient self- be used utilized earlier in the course of this disease, assessment information,2 which in many cases is it can aid in reducing patient suffering, loss of time moreimportantthanother clinicaloutcomes.Unfor- at work, and health care costs associated with tunately, the study was not powered to show signi- ficant difference in the SF-12 scale, as this wouldhave required an unfeasible amount of patients.
The significant difference in blinding verifica- tion between the groups deserves explanation.
This most likely was influenced by the subject’s Present conservative treatments for plantar judgment about the presence or absence of pain fasciitis include rest, physical therapy, heel during treatment, which incidentally was also cushions, nonsteroidal anti-inflammatory drugs, statistically significant. The presence or absence of corticosteroid injections, taping, orthotics, shoe pain during treatment in either group could be due modifications, nightsplinting, and casting. ESWT to several variables such as differences in subject’s is proposed as an additional conservative treat- pain tolerance or inconsistent adequacy of the ment to be used to avoid surgery, when other available conservative methods have failed. Relief To truly compare a clinical intervention to from pain can be recognized with a single session placebo, as the comparative clinical trials in the compared to traditional conservative therapies past have claimed to do, blinding of subjects and that require multiple applications and for which assessment of the efficacy of the blinding are clear benefits have not been established. Shock necessary to attemp to control the placebo effect.
wave therapy is minimally invasive, has a short Many previous trials of ESWT for plantar fasciitis recuperation period, and reports only minor, tran- did not include blinding or assessment of blinding, sient side effects. Also, shock wave therapy may so it is difficult to compare our results in this area circumvent the need for surgical intervention to others in the literature. It should be noted that and the associated costs, lost time from work, our assessment of the subjects’ blindness to the and complications associated with surgery.
type of treatment is of interest only for evaluating The results of this study confirm that high- our method of blinding, and we can safely conclude energy ESWT, administered with the Dornier that our method of blinding worked as well as Epos Ultra is a safe and effective treatment for patients who have failed previous conservative nonsurgical treatments for chronic plantar fascii- controlled trials published recently with varying tis. The future of our research of ESWT will include JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2006 CONFIRMATORY RCT OF ESWT FOR PLANTAR FASCIITIS further study of optimal dosing, frequency, and placebo-controlled double-blind study. Z Orthop 7. Buchbinder R, Ptasznik R, Gordon J, et al. 2002.
therapy for plantar fasciitis: a randomized con-trolled trial. JAMA 288:1364–1372.
All clinical sites received research grant funding from 8. Haake M, Buch M, Schoellner C, et al. 2003. Extra- Dornier MedTech America, Inc., Atlanta, GA. We grate- corporeal shock wave therapy for plantar fasciitis: fully acknowledge the following investigators for their randomised controlled multicentre trial. BMJ dedication and participation in the clinical trial: Joyce Johansson, MD, Alain Scheldeman, B.Sc. PT., CAT(c), 9. Hammer DS, Adam F, Kreutz A, et al. 2003. Extra- Rory Mullin MSc, ATC, CAT (c), Dac., and Susanne corporeal shock wave therapy (ESWT) in patients Pare, Bsc.PT. A special thank you to the many other with chronic proximal plantar fasciitis: a 2-year colleagues and especially the Research Coordinators, follow-up. Foot Ankle Int 24:823–828.
who could not all be named but without whom the study 10. Speed CA, Nichols D, Wies J, et al. 2003. Extra- corporeal shock wave therapy for plantar fasciitis.
A 11. Theodore GH, Buch M, Amendola A, et al. 2004.
Extracorporeal shock wave therapy for the treat- 1. Chaussy C, Eisenberger F, Wanner K. 1976. The ment of plantar fasciitis. Foot Ankle Int 25:290–297.
use of shock waves for the destruction of renal cal- 12. Ogden JA, Alvarez RG, Marlow M. 2002. Shock- culi without direct contact. Urol Res 181:352–357.
wave therapy for chronic proximal plantar fasciitis: 2. Roles NC, Maudsley RH. 1972. Radial tunnel a meta-analysis. Foot Ankle Int 23:301–308.
syndrome: resistant tennis elbow as a nerve 13. Strash WW, Perez RR. 2002. Extracorporeal shock- entrapment. J Bone Joint Surg Br 54:499–508.
wave therapy for chronic proximal plantar fasciitis.
3. Davis PF, Severud E, Baxter DE. 1994. Painful heel syndrome: results of non-operative treatment. Foot 14. Ware JE, Kosinski M, Keller SE. 1996. A 12-Item short-form health survey: construction of scales 4. Ogden JA, Alvarez R, Levitt R, et al. 2001. Shock and preliminary tests of reliability and validity.
wave therapy for chronic proximal plantar fasciitis.
15. Schon LC. 1993. Plantar fasciitis/heel pain. In 5. Riddle DL, Schappert SM. 2004. Volume of ambu- Pfeffer GB, Frey CC, Anderson RB, et al., editors.
latory care visits and patterns of care for patients Current practice in foot and ankle surgery, Vol. 1.
diagnosed with plantar fasciitis: a national study of New York: McGraw-Hill; pp 243–261.
medical doctors. Foot Ankle Int 25:303–310.
16. Kltaoka HB, Alexander IJ, Adelaar RS, Nunley JA, 6. Abt T, Hopfenmuller W, Mellerowicz H. 2002.
et al. 1994. Clinical rating systems for the ankle- Shock wave therapy for recalcitrant plantar fascii- hindfoot, midfoot, hallux, and lesser toes. Foot tis with heel spur: a prospective randomized JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2006

Source: http://www.healthmedica.ca/main/docs/treatment-of-plantar-fasciitis-with-eswt-high-energy-2005.pdf