J Rehabil Med 2019; 51: 201–208 ORIGINAL REPORT RADIAL EXTRACORPOREAL SHOCK WAVE DEVICE APPEARANCE DOES NOT JRM INFLUENCE CLINICAL OUTCOMES: A RANDOMIZED CONTROLLED TRIAL Antoni MORRAL, PT, PhD1, Gerard URRÚTIA, MD, PhD2, Ignasi GICH, MD, PhD2, Reme RUIZ, PT3 and Xavier BONFILL, MD, PhD2 From the 1Blanquerna School of Health Sciences, Ramon Llull University, 2Iberoamerican Cochrane Centre, Institut d’Investigació Biomèdica Sant Pau (IIB Sant Pau), CIBERESP, Universitat Autònoma de Barcelona (UAB), Barcelona and 3Salut i Esport Rehabilitation Journal of Rehabilitation Medicine Center, Santa Perpètua de Mogoda, Spain. Objective: To determine whether the appearance of LAY ABSTRACT a radial extracorporeal shock wave device affects The context and environment in which treatment is ad- clinical outcomes in chronic plantar fasciitis. ministered is highly relevant. Context variables are per- Study design: Randomized controlled parallel asses- ceived and interpreted by patients and can generate po- sor-blinded clinical trial. sitive or negative expectations. Such expectations may Material and methods: A total of 135 patients were influence the therapeutic outcome. This study assessed assigned to 3 groups: group I, standard radial ex- whether a contextual element, such as the external ap- tracorporeal shock wave device; group II, standard pearance of a shock wave device, influenced clinical out- radial extracorporeal shock wave device modified comes in patients with chronic plantar fasciitis. Three to give a more sophisticated appearance; group III, shock wave devices were compared: a standard device; standard radial extracorporeal shock wave device a sophisticated device; and an austere device. The only modified to give a more austere appearance. The difference between the devices was their external ap- radial extracorporeal shock waves emitted by the 3 pearance. The shockwaves emitted by the 3 devices devices were identical. Primary outcome was foot were identical. No differences were found between the function, measured with the Foot Function Index. 3 devices for any of the variables assessed. Health pro- Secondary outcomes were pain at different times, fessionals and future research into the therapeutic en- measured with a visual analogue scale, and plantar counter context should focus more on patient–therapist JRM fascia thickness, measured with ultrasound. interactions than on the appearance of devices. Results: All variables decreased significantly from baseline assessment, in all 3 groups and at all time- points: 1, 2, 4 and 14 months after the last session should be appropriately classified as a “fasciosis” or (p < 0.001). There were no significant differences “fasciopathy” (3). between groups for any of the variables assessed. The causes of plantar fasciitis are multifactorial. It Conclusion: Device appearance had no statistically responds well to multiple conservative therapies. The significant influence on clinical outcomes in patients prognosis is favourable, with resolution of symptoms Journal of Rehabilitation Medicine with chronic plantar fasciitis treated with radial ex- in approximately 80% of cases (4). However, there is tracorporeal shock wave therapy. no evidence-based consensus on the most effective Key words: placebo effect; nocebo effect; plantar fasciitis; treatment. Systematic reviews have shown radial shock wave, therapeutic context. extracorporeal shock waves (rESW) to be effective Accepted Dec 18, 2018; Epub ahead of print Jan 22, 2019 and safe for the treatment of chronic plantar fasciitis (5–7). The biological effects of rESW are tissue rege- J Rehabil Med 2019; 51: 201–208 neration, angiogenesis and analgesia (8, 9), produced Correspondence address: Antoni Morral, Blanquerna School of Health by a mechanotransduction mechanism by which cells Sciences, Ramon Llull University, 08025 Barcelona, Spain. E-mail: an- tonimf@blanquerna.url.edu recognize and respond to mechanical stimulation (10). There are many variables, including the placebo effect, which can explain the clinical improvement P lantar fasciitis is the most common cause of inferior heel pain. Typical symptoms include pain with the first weight-bearing step in the morning. Diagnosis of experienced by patients when they receive a phar- macological treatment or physical treatment, such as rESW (11). The placebo and nocebo responses plantar fasciitis can be made clinically (1). The inci- refer to changes in patients’ symptoms attributable dence and prevalence of the condition are not fully to their participation in a therapeutic encounter and known; however, it is estimated that, in the general its rituals, symbols and interactions (12). However, population, 1 in 10 people will experience plantar other elements, such as patient–therapist relationship, fasciitis at least once in their life (2). psychosocial factors, and patients’ hopes, beliefs and JRM Although the suffix “-itis” implies an inflammatory fears, must also be considered (13, 14). condition, mounting evidence indicates that this foot The context and environment in which treatment is disorder is associated with degenerative changes and administered is highly relevant. Context variables are This is an open access article under the CC BY-NC license. www.medicaljournals.se/jrm Journal Compilation © 2019 Foundation of Rehabilitation Information. ISSN 1650-1977 doi: 10.2340/16501977-2516 202 A. Morral et al. perceived and interpreted by patients and can generate JRM Table I. Inclusion and exclusion criteria for patients with chronic plantar fasciitis enrolled in the present study positive or negative expectations (15). A placebo re- Inclusion criteria sponse occurs when a medical treatment and its context Age over 18 years. trigger specific expectations about a positive thera- Able to understand the explanations about the potential benefits and risks of participating in the study. peutic outcome. Pre-existing optimistic expectations Diagnosis of chronic plantar fasciitis. Tested by clinical examination, meeting can amplify the positive effects of treatments (placebo at least 2 criteria: Journal of Rehabilitation Medicine • Pain on palpation of the proximal insertion of the plantar fascia (area of ​​ effects), but negative expectations can also worsen the medial tuberosity of the calcaneus). treatment in the form of side-effects or a decrease in • Pain with the first weight-bearing step in the morning and/or after a period of time sitting. treatment-typical improvements (nocebo effects) (16). Duration of symptoms equal or superior to 6 months at the time of Studies have shown that modifying the shape, colour enrolment. Have received conservative therapy without success, at least 2 treatments. or price of a placebo drug can produce different pla- The treatments may have been performed in isolation, in combination, or cebo responses (17, 18), yet there are few studies that consecutively. Non-pharmacological treatments explore and identify the contextual elements involved • Physical therapy, such as ice, heat, iontophoresis or ultrasound. in placebo and nocebo effects; this is highly relevant, • Physiotherapy, massage or stretching. • Orthosis. since the therapeutic effect of many treatments is inse- • Heel pads or bandages. parable from the context in which they are administered. • Immobilization. However, there are other phenomena that can explain • Night splints. Pharmacological treatments patients’ clinical improvement, such as the Hawthorne • Topical application of analgesic or anti-inflammatory gels. effect, natural fluctuations in diseases, and regression to • Analgesics or oral anti-inflammatories. • Local anaesthetic injection. the mean (19–21). The placebo effect can be overesti- • Local corticosteroid injection. mated if these are not taken into account (22). Minimum period of time after receiving other treatments: This randomized controlled trial (RCT) aimed to as- • 6 weeks from the last cortisone injection. • 4 weeks from the last iontophoresis session. sess whether a contextual element, such as the external Exclusion criteria appearance of the rESW device, would influence clini- Neurological or vascular insufficiency in the painful heel. cal outcomes in patients with chronic plantar fasciitis. Any inflammation at the level of the ankle. History of rheumatic disease, collagenosis or metabolic diseases. JRM History of hypo- or hyperthyroidism. Malignant disease with or without metastases. METHODS Paget’s disease or calcaneal fat-pad atrophy. Osteomyelitis (acute, subacute or chronic). Study design History of calcaneus fracture. Patients on immunosuppressive therapy. This randomized controlled parallel assessor-blinded clinical Patients on corticosteroid treatment longer than 3 months. trial was conducted in a private rehabilitation and physioth- Patients with diabetes mellitus or cardiac or respiratory disease. erapy centre, “Salut i Esport”, in Santa Perpètua de Mogoda, Patients on anticoagulant drugs. Barcelona province, Spain. Patients were recruited from this Previous surgery on the painful heel. Journal of Rehabilitation Medicine Previous treatments with shock wave devices. centre and other private centres (n = 122) and from public cen- Significant abnormalities in liver function. tres (n = 13). A total of 135 patients were enrolled in the study Pregnancy. during 2014–15. Infection in the treated area. History or documented evidence of sciatica, peripheral neuropathy, such as nerve compression or tarsal tunnel syndrome. Study participants History or documented evidence of systemic inflammatory disease, such as rheumatoid arthritis, ankylosing spondylitis, aseptic bone necrosis, or Patients were referred to the lead investigator and considered for Reiter’s syndrome. participation in the study according to the inclusion and exclu- History or documented proof of labour compensation or litigation for medical conditions. sion criteria summarized in Table I. A total of 135 participants Patients who, in the opinion of the researcher, may be inappropriate for were recruited and 128 completed the study (Fig. 1). inclusion in the study or would not comply with the study requirements. After providing written informed consent, patients with Neurological deficits: sensory or motor deficits, abnormal tendon reflexes. chronic plantar fasciitis (more than 6 months’ duration) were Bilateral plantar fasciitis. randomly assigned to the following 3 groups: • Group I: standard device: therapy applied with a standard rESW device (n = 45). European safety regulations for health equipment (Registra- • Group II: sophisticated device: therapy applied with a modi- tion: EN-60601-1, Class I. Type BF IP40. 93/42 EEC) were fied standard rESW device designed to give a more attractive, used. Two of the 3 devices (corresponding to groups II and III), high-tech appearance, with a larger external component which were initially standard, were modified to change only (n = 45). the external part of the device without affecting the safety of • Group III: austere device: therapy applied with a modified the treatment or the technical performance of the equipment. standard rESW device designed to give a more austere, The same therapeutic radial extracorporeal shock waves were unattractive, low-tech appearance, with a smaller external used in all cases. component (n = 45). The study was approved by the Ethics and Research Com- JRM Three Swiss Doloclast rESW devices (EMS Electro Medical mission of Blanquerna School of Health Science, Ramon Llull Systems, Nyon, Switzerland), which comply with all of the University (Barcelona, Spain). This study was carried out in www.medicaljournals.se/jrm Comparison of 3 devices on clinical outcomes in plantar fasciitis 203 JRM accordance with the World Medical Association Total number of possible participants Declaration of Helsinki (23), according to good (n=234) clinical practice, and registered with ClinicalTri- als.gov (identifier NCT02608723). Excluded (n=99) Did not meet eligibility criteria (n=80) Did not agree to participate in the study (n=19)! Intervention All patients followed the same schedule of visits Journal of Rehabilitation Medicine in an identical setting. During the first visit, pa- Randomized (n=135) tients were given an information document about the study, and the lead researcher explained the study aims with the wording: “the 3 devices are Assigned to the Assigned to the Assigned to the austere effective, all 3 are good, but we want to know standard device group sophisticated device group device group (n=45) which of them is best”. Information on plantar (n=45) (n=45) fascia and gastrocnemius muscle stretching exercises was also provided and explained. The patients then underwent a baseline evaluation: All patients All patients All patients completed the completed the completed the they were asked to complete the Foot Function study protocol study protocol study protocol Index (FFI) questionnaire, a visual analogue scale (VAS) was used to assess pain, and a foot ultrasound was performed. Each patient was then Lost to follow-up in the Lost to follow-up in the Lost to follow-up in taken to another room containing all 3 devices, standard device group sophisticated device group austere device group where they were to be treated. The area of the tre- (n=3) (n=0) (n=4) atment room was 10 m2, it contained a stretcher, 2 chairs and the 3 devices, located next to each other (Figs 2 and 3). A second researcher explai- ned the treatment procedure for 1 min and how the 3 rESW devices worked. He fired 32 shock Analysed in the standard Analysed in the Analysed in the austere waves (4 s) into the air using each device, so that device group during sophisticated device group device group the patient could hear and see the operation of (n=42) (n=45) (n=41) the devices. The envelope assigning the device JRM Fig. 1. Flowchart of participants. to that individual was then opened. The patient looked at the assigned device for 1 min, and was then asked to remove their shoes and lie on the stretcher in a prone position for the first session. All patients received 3 sessions, 1 per week. Patient interaction with the de- vices in the second and third session was the same as in the first session: the researcher explained the operation of the 3 devices for 1 min, then the patient observed the assigned device for 1 Journal of Rehabilitation Medicine min. In each session 2,500 radial extracorporeal shock waves were applied at a frequency of 8 Hz. The applied energy was adjusted to the maximum discomfort the patient could tolerate, resulting in an individual air pressure of between 2.0 and 4.0 bar and, thus, a positive energy flux density of between 0.10 and 0.18 mJ/mm2. No local anaesthesia or analgesic drugs were given during the sessions. The patients were followed-up at 1, 2, 4 and 14 months after the last radial extracorporeal shock wave therapy (rESWT) session. Fig. 2. The devices used in the present study: (1) sophisticated device; (2) austere device; (3) standard device. JRM Fig. 3. Close-up of the devices: (1) sophisticated device; (2) austere device; (3) standard device. J Rehabil Med 51, 2019 204 A. Morral et al. Outcome measures JRM envelopes, ensuring the concealment of assignments. The pa- tients and the therapist were not blinded to the treatment due to The primary outcome was foot function, measured using the the nature of the study, as they knew which of the 3 devices was FFI, a self-administered questionnaire that quantifies the impact being applied, but the evaluator and data analyst were blinded. of pathology on foot function in terms of pain, disability and activity restriction (24), and is one of the most-used assessment tools in clinical trials on plantar fasciitis treatments (25, 26). Statistical analysis The index consists of 23 items that assess foot function on a Descriptive values are expressed as means and standard devia- Journal of Rehabilitation Medicine continuous scale of 100 points, where a lower score indicates tions (SD) for quantitative variables, and relative frequency for better function. This variable was assessed at baseline and at categorical variables. 4 follow-up sessions: at 1, 2, 4 and 14 months after the last As the main study analysis, a 2-way analysis of variance rESWT session. (ANOVA) was performed, with groups (3 levels: (groups I, II Secondary variables included: (i) pain assessment for “pain and III)) and time (5 levels (baseline and follow-ups) of repeated with the first weight-bearing step in the morning” and “pain measures) as independent variables, and clinical variables (foot during the day”, measured with a 10-point VAS at baseline and functionality, morning pain, pain during the day, and plantar at 4 follow-ups: at 1, 2, 4 and 14 months after the last rESWT fascia thickness) as dependent variables. session; (ii) plantar fascia thickness of the affected foot, mea- For secondary analysis with categorical variables a χ2 test sured using ultrasound (Echo Blaster EXT-128) at baseline and was performed. Finally, in the case of quantitative variables, at 2 time-points: 4 and 14 months after the last rESWT session. the inferential analysis was carried out with analysis of va- Complications, adverse effects during and after treatment were riance (ANOVA); the Kruskal–Wallis test was performed as recorded. Information was also collected on discomfort during a non-parametric test alternative to ANOVA (failure to meet the application of rESWT (using a VAS scale). assumptions or for ordinal variables). Baseline patient data collected included age, sex, body mass The sample size was calculated using GRANMO Software index (BMI) (kg/m2), duration of symptoms (months), time (Version 7.12, IMIM-Hospital del Mar, Barcelona, Spain), with spent standing per day (h), exercise level (1, no activity; 2, less the assumption that an ANOVA would be performed among the than 5 h a week; 3, between 5 and 10 h a week; 4, more than 10 3 groups and 5 time-points. Accepting a p-value below 0.05 and h a week), and level of education (1, no education; 2, compul- a statistical power (beta-1 risk) above 0.8 in a bilateral contrast, sory education (primary and secondary school); 3, high-school 45 patients were required in each group to detect a minimum studies or vocational training; and 4, university-level education). difference of 15 points between 2 groups, assuming that there were 3 groups and a standard deviation (SD) of 20. A maximum rate of 15% loss to follow-up was estimated. Randomization and blinding JRM Calculations were performed using IBM-SPSS (Version 22, Randomization was performed through a specific syntax for this IBM, Armonk, NY, USA). study, in IBM-SPSS language (V22), in blocks and perfectly balanced by a statistician from the Ibero-American Cochrane Centre, Clinical Epidemiology and Public Health Service of the RESULTS Hospital de la Santa Creu i Sant Pau. The randomization num- bers were placed in sealed opaque envelopes, thus concealing Baseline patient data allocation from both the patients and the therapist until treatment A total of 135 patients participated in the study; 45 in started. The envelopes were sequentially ordered from 1 to 135 each group. Mean age was 49 years, and 48% of the Journal of Rehabilitation Medicine and stored in a locked cabinet. Both the lead researcher (who assessed patients for eligibility) and the researcher responsible patients were women. Mean duration of symptoms was for delivering the treatment had no access to the content of the 14 months. Table II shows the baseline patient data. Table II. Baseline data Group I: standard device Group II: sophisticated device Group III: austere (n = 45) (n = 45) device (n = 45) Mean (SD) Mean (SD) Mean (SD) p-value Age, years 48.27 (9.96) 52.51 (12.28) 49.31 (11.14) 0.054* Body mass index, kg/m2 27.61 (3.31) 27.66 (3.72) 28.35 (5.25) 0.642 Duration of symptoms, months 13.27 (9.26) 14.98 (12.72) 13.53 (9.09) 0.708 Hours spent standing per day 7.67 (3.53) 8.80 (3.67) 8.33 (4.09) 0.362 Sex, n (%) Female 33.3 60 51.1 0.036* Male 66.7 40 48.9 Exercise, n (%) No activity 44.4 44.4 57.8 0.407 Low activity 35.6 42.2 28.9 Moderate activity 15.6 13.3 8.6 Intense activity 4.4 0 4.4 Educational level, n (%) 0.166 No education 2.2 8.9 6.7 Compulsory education 35.6 44.4 31.1 High-school education or vocational training 33.3 28.9 33.3 University education 28.9 17.8 28.9 JRM *At baseline, significant differences were observed between groups for the distribution of the variable sex (p = 0.036), and for the variable age, the difference was close to significance (p = 0.054). Thus, the entire analysis was corrected for the sex and age of the patients, using a co-variance analysis (ANCOVA). SD: standard deviation. www.medicaljournals.se/jrm Comparison of 3 devices on clinical outcomes in plantar fasciitis 205 significant decrease from baseline assessment in all 3 JRM 70 groups. No significant differences were found among 60 groups (main effect) (p = 0.599) (Fig. 6). 50 Influence of device appearance on plantar fascia FFI- Foot Function Index thickness Journal of Rehabilitation Medicine 40 There was no statistically significant interaction ef- 30 fect between group and time (p = 0.402) for plantar fascia thickness. There was a significant time effect 20 (p < 0.001); i.e. a significant decrease from baseline assessment in all 3 groups. There were no significant 10 differences between groups (main effect) (p = 0.800) (Fig. 7). 0 Basal 1 month 2 months 4 months 14 months Influence of device appearance on adverse effects Group I- Standard device Time Group II- Sophisticated Device Group III- Austere device No serious side-effects or complications were observed Fig. 4. Foot Function Index (FFI) over time, by device group. Time in any group. Adverse effects were increased heel pain factor p < 0.001; device factor p=0.863; device-time interaction factor p = 0.611. **The whole analysis was adjusted, correcting for age and in 19 patients and headache in 9 patients. These effects sex of the patients, by means of a co-variance analysis (ANCOVA). appeared after application of rESWT and resolved The adjustment showed no change in significance with respect to the completely in 4 days without requiring treatment. No initial result. other adverse events were observed and no significant differences were found among groups (p = 0.473). In this clinical trial there was no loss to follow-up Regarding the discomfort perceived by patients during after the first, second and fourth months after inter- the application of rESWT, there were no significant vention, but there were 7 patients lost to follow-up differences between groups (p = 0.660). JRM after the 14th month due to lack of interest or time constraints; hence 128 patients completed the study and were analysed. DISCUSSION Broadly, the placebo and nocebo effects consist of an Influence of device appearance on foot function improvement or worsening of symptoms that is attribu- The data showed no significant interaction effect be­ table to the context of the patient-therapist relationship. tween treatment group and time for the FFI variable Journal of Rehabilitation Medicine (F (6.857, 428.554)= 0.064; p = 0.863). However, there was a 10.00 significant time effect (F (3.428, 428.554)=209.31; p < 0.001); 9.00 that is, a significant decrease from baseline in all 3 groups. No significant differences were found among 8.00 groups (main effect) (F (2,125)=0.196; p = 0.611). (Fig. 4). 7.00 VAS - Visual Analogue Scale Influence of device appearance on pain with the first 6.00 weight-bearing step in the morning 5.00 For the VAS variable there was no significant inte- 4.00 raction effect between group and time (p = 0.910). 3.00 There was a significant time effect (p < 0.001), with a significant decrease from baseline assessment in all 3 2.00 groups. No significant differences were found among 1.00 groups (main effect) (p = 0.623) (Fig. 5). 0.00 Basal 1 month 2 months 4 months 14 months Influence of device appearance on pain during the day Group I- Standard device Time Group II- Sophisticated Device Group III- Austere device For the VAS variable there was no significant inte- Fig. 5. Visual analogue scale (VAS) assessment of pain with the first weight-bearing step in the morning, over time and by device group. JRM raction effect between group and time (p = 0.853). Time factor p < 0.001; device factor p = 0.910; device–time interaction There was a significant time effect (p < 0.001), with a factor p = 0.623. J Rehabil Med 51, 2019 206 A. Morral et al. or austere device). Furthermore, all patients improved JRM 10.00 9.00 significantly over time in relation to the baseline assess- ment. The improvement was observed equally in all 3 8.00 groups and at all assessment time-points: 1, 2, 4 and 14 7.00 months after the last rESWT session, with no statisti- cally significant differences among groups over time. VAS- Visual Analogue Scale Journal of Rehabilitation Medicine 6.00 That is, the clinical outcomes were independent of the device used. Although the appearance of the device 5.00 neither improved nor worsened treatment outcomes, as 4.00 shown in this study, small differences were observed between the austere group and the other groups. The 3.00 austere group obtained, in most evaluations, the worst 2.00 clinical outcomes. Despite an extensive neuroscience base that supports 1.00 the placebo effect, there is a lack of clinical research 0.00 that explores, in a healthcare setting, the context and Basal 1 month 2 months 4 months 14 months placebo responses that accompany the overall thera- Group I- Standard device Group II- Sophisticated Device Fig. 6. Visual analogue scale (VAS) assessment of pain during the day, Group III- Austere device peutic intervention. Most existing knowledge about the over time and by device group. Time factor p < 0.001; device factor placebo effect has come from studies in basic science p = 0.853; device–time interaction factor p = 0.599. and clinical trials that are far removed from usual clini- cal practice. However, it has been found that placebo 8.00 responses are greater under these conditions than in 7.00 clinical trials conducted in real healthcare settings (21). 6.00 Few clinical trials have analysed treatment context Thickness in mm 5.00 factors, such as the appearance of a physical agent 4.00 and its possible influence on patient recovery, making JRM 3.00 it difficult to compare the results of the current study 2.00 1.00 with those obtained in similar studies. In a study con- 0.00 ducted by Dawes et al. (30), 2 identical hearing aids Basal 4 months 14 months were compared: 1 described as “new” and the other as Group I- Standard device Time Group II- Sophisticated Device Group III- Austere device “conventional”. Approximately 75% of the participants Fig. 7. Fascia thickness over time and by device group. Time factor preferred the new device and reported that the hearing p < 0.001; device factor p = 0.402; device–time interaction factor quality it provided was superior to the conventional Journal of Rehabilitation Medicine p = 0.800. one (30). Another clinical trial, with real patients, investigated the influence of different verbal infor- Therapeutic context includes a multitude of signals mation combined with a real analgesic drug (31) and inherent to any intervention, which are perceived and reported similar results to the present study, showing no interpreted by patients and generate positive or nega- differences in pain reduction. Furthermore, in a cros- tive expectations (12). Placebo and nocebo effects can sover clinical trial, it was shown that the same person be generated from expectations (27–29); therefore it is may respond differently to different types of placebo important to assess contextual factors that can generate (placebo tablet or sham acupuncture). The response expectations and their impact on clinical outcomes. to placebo is a complex phenomenon that has many This RCT was designed to evaluate the influence of variables and goes beyond patient characteristics. This the appearance of a physical agent; in this case a rESW could explain the difficulty of detecting a pattern for device, on clinical outcomes in plantar fasciitis. people “responding to placebo” (32, 33). It was found that, in the treatment of chronic plantar The magnitude of the placebo effect depends on fasciitis with rESW, the appearance of the device did numerous design factors. A 2015 meta-analysis found not significantly affect the clinical results. None of that the type of active drug (opioid or non-steroidal the 5 clinical variables analysed (foot function, pain anti-inflammatory drug (NSAID)), the scheduled with the first weight-bearing step in the morning, pain follow-up visits (number of planned face-to-face vi- during the day, plantar fascia thickness, and adverse sits), and randomization ratio (probability of receiving effects) showed statistically significant differences bet- a placebo treatment, 1:4 or 1:1) were predictive of the JRM ween the 3 groups: patients had similar results regard- magnitude of the placebo response, thereby supporting less of the treatment device (standard, sophisticated, the expectancy hypothesis. Exploratory models sho- www.medicaljournals.se/jrm Comparison of 3 devices on clinical outcomes in plantar fasciitis 207 wed that baseline pain intensity, age, washout length, was a group without treatment. Another limitation of JRM and discontinuation due to adverse events accounted this study was the presence of the 3 devices in the same for approximately 10% of placebo response variability, room, since in a real healthcare setting only a single and patients’ perception of treatment allocation and device would be available. This could increase patients’ expectations toward treatment efficacy was strongly expectations and thus generate placebo or nocebo re- predictive of RCT outcomes (34). Nevertheless, most sponses. This aspect might have been relevant for the Journal of Rehabilitation Medicine of the placebo effect is due to individual factors that external validity of the study, if differences had been are not yet well identified; hence, there is a growing found between the groups. interest in determining the individual factors that can The current study may have some practical conse- predict the placebo response (35). quences. The results could encourage the industry to Patient expectations have been identified as one of manufacture rESW devices with a more austere design, the major components contributing to placebo respon- which would allow lower production costs. Therefore, ses. Patients requiring therapeutic interventions are equipment and treatment sessions would be cheaper, exposed to stimuli in the clinical setting that trigger potentially increasing the number of patients who could specific expectations about treatment and outcomes. access this therapy. These stimuli include the nature of the treatment it- Creative experimental efforts are needed to rigo- self, such as surgery, drugs, injections or therapeutic rously assess the clinical significance of placebo and devices. They also include the characteristics of the nocebo effects and investigate the individual elements clinician and the relationship formed with the patient, that may contribute to a therapeutic benefit. Future as well as the clinician’s confidence in the therapy research should place more emphasis on patient–thera- and explanation of the treatment. The wider treatment pist interactions rather than the diagnostic and thera- context, such as the reputation of the facility and status peutic tools used. The placebo effect must be isolated of the clinic, may also affect treatment outcome ex- from other phenomena present in clinical trials, such as pectations. These factors all operate psychologically the Hawthorne effect, natural fluctuations in diseases, and can enhance or diminish treatment efficacy (36). and regression to the mean (20–22). These phenomena The design of the current RCT allowed us to isolate can cause confusion and lead to clinical improvements JRM and analyse the external appearance of the device as a being wrongly attributed to a placebo effect. contextual element of the placebo and nocebo respon- ses. The strict inclusion and exclusion criteria allowed Conclusion appropriate selection of patients with chronic plantar fasciitis, excluding patients with heel pain caused by In patients with chronic plantar fasciitis treated with other pathologies. The main clinical outcome, foot rESW therapy, the appearance of the device did not functionality, was assessed through the FFI, which has influence clinical outcomes: function, pain with the Journal of Rehabilitation Medicine been proven as a validated and reliable instrument, and first weight-bearing step in the morning, pain during is one of the most-used tools in clinical trials on the the day, fascia thickness, and adverse effects. As device effectiveness of plantar fasciitis treatment interventions appearance did not affect treatment outcomes, it should (25, 26); the quality of this evidence has been rated not currently be considered as one of the contextual as high (grade A recommendation by the American elements that generate placebo and nocebo responses. Physical Therapy Association) (37). A strong statistical Further research is needed to identify the contextual va- power was reported; thus we can reasonably dismiss riables, including patient-therapist interactions, which a type II error (or beta type error). In contrast to other influence such responses. studies, this study was carried out in a real healthcare setting, with patients rather than healthy volunteers. REFERENCES Other similar studies have used small samples and short 1. Petraglia F, Ramazzina I, Costantino C. Plantar fasciitis in follow-up periods (18, 36, 38), and were therefore less athletes: diagnostic and treatment strategies. A systematic reflective of real clinical practice. review. Muscles Ligaments Tendons J 2017; 7: 107–118. 2. De Maio M, Paine R, Mangine RE, Drez D Jr. Plantar fasciitis. Orthopedics 1993; 16: 1153–1163. Limitation 3. Lemont H, Ammirati KM, Usen N. Plantar fasciitis: a de- generative process (fasciosis) without inflammation. J Am One limitation was the wording of the study. It was Podiatr Med Assoc 2003; 93: 234–237. decided not to use the words “placebo effect” with the 4. Hawke F, Burns J, Radford JA, du Toit V. Custom-made participants and this was not included in the informed foot orthoses for the treatment of foot pain. Cochrane Database Syst Rev 2008; 16: CD006801. JRM consent form, because it could have confused the par- 5. Lou J, Wang S, Liu S, Xing G. Effectiveness of extracor- ticipants and made them mistakenly believe that there poreal shock wave therapy without local anesthesia in J Rehabil Med 51, 2019 208 A. Morral et al. patients with recalcitrant plantar fasciitis: a meta-analysis of biases. J Clin Epidemiol 2011; 64: 1223–1229. JRM of randomized controlled trials. Am J Phys Med Rehabil 22. Morral A, Urrutia G, Bonfill X. Placebo effect and thera- 2017; 96: 529–534. peutic context: a challenge in clinical research. Med Clin 6. Sun J, Gao F, Wang Y, Sun W, Jiang B, Li Z. Extracorporeal (Barc) 2017; 7: 26–31. shock wave therapy is effective in treating chronic plantar 23. General Assembly of the World Medical Association. World fasciitis: a meta-analysis of RCTs. Medicine (Baltimore) Medical Association Declaration of Helsinki: ethical prin- 2017; 96: e6621. ciples for medical research involving human subjects. J 7. Speed C. A systematic review of shockwave therapies Am Coll Dent 2014; 81: 14–18. in soft tissue conditions: focusing on the evidence. Br J 24. Budiman-Mak E, Conrad KJ, Roach KE. The Foot Function Journal of Rehabilitation Medicine Sports Med 2014; 48: 1538–1542. Index: a measure of foot pain and disability. J Clin Epide- 8. Mattyasovszky SG, Langendorf EK, Ritz U, Schmitz C, miol 1991; 44: 561–570. Schmidtmann I, Nowak TE, et al. Exposure to radial ex- 25. Lohrer H, Nauck T, Dorn-Lange NV, Schöll J, Vester JC. tracorporeal shock waves modulates viability and gene Comparison of radial versus focused extracorporeal shock expression of human skeletal muscle cells: a controlled in waves in plantar fasciitis using functional measures. Foot vitro study. J Orthop Surg Res 2018; 6: 13: 75. Ankle Int 2010; 31: 1–9. 9. Imamura M, Alamino S, Hsing WT, Alfieri FM, Schmitz C, 26. Rompe JD, Cacchio A, Weil L Jr, Furia JP, Haist J, Reiners V Battistella LR. Radial extracorporeal shock wave therapy et al. Plantar fascia-specific stretching versus radial shock- for disabling pain due to severe primary knee osteo- wave therapy as initial treatment of plantar fasciopathy. J arthritis. J Rehabil Med 2017; 19: 54–62. Bone Joint Surg Am 2010; 92: 2514–2522. 10. D Agostino MC, Frairia R, Romeo P, Amelio E, Berta L, 27. Colloca L, Miller FG. How placebo responses are formed: Bosco V, et al. Extracorporeal shockwaves as regenerative a learning perspective. Philos Trans R Soc Lond B Biol Sci therapy in orthopedic traumatology: a narrative review 2011; 27: 1859–1869. from basic research to clinical practice. J Biol Regul Ho- 28. Stewart-Williams S, Podd J. The placebo effect: dissolving meost Agents 2016; 30: 323–332. the expectancy versus conditioning debate. Psychol Bull 11. Finniss DG, Kaptchuk TJ, Miller F, Benedetti F. Biological, 2004; 130: 324–340. clinical, and ethical advances of placebo effects. Lancet 29. Di Blasi Z, Harkness E, Ernst E, Georgiou A, Kleijnen J. 2010; 20: 686–695. Influence of context effects on health outcomes: a syste- 12. Jubb J, Bensing JM. The sweetest pill to swallow: how pa- matic review. Lancet 2001; 10: 757–762. tient neurobiology can be harnessed to maximise placebo 30. Dawes P, Hopkins R, Munro KJ. Placebo effects in hearing- effects. Neurosci Biobehav Rev 2013; 37: 2709–2720. aid trials are reliable. Int J Audiol 2013; 52: 472–477. 13. Barrett B, Rakel D, Chewning B, Marchand L, Rabago D, 31. Oktay C, Eken C, Goksu E, Dora B. Contribution of verbal Brown R et al. Rationale and methods for a trial assessing suggestion to the therapeutic efficacy of an analgesic placebo, echinacea, and doctor-patient interaction in the agent for acute primary headache. Cephalalgia 2015; common cold. Explore (NY) 2007; 3: 561–572. 35: 579–584. JRM 14. Di Blasi Z, Harkness E, Ernst E, Georgiou A, Kleijnen J. 32. Hall KT, Loscalzo J, Kaptchuk TJ. Genetics and the placebo Influence of context effects on health outcomes: a syste- effect: the placebome. Trends Mol Med 2015; 21: 285–294. matic review. Lancet 2001; 10: 757–762. 33. Marostica E, Russu A, Yang S, De Nicolao G, Zamuner 15. Colloca L, Miller FG. How placebo responses are formed: S, Beerahee M. Population model of longitudinal FEV1 a learning perspective. Philos Trans R Soc Lond B Biol Sc data in asthmatics: meta-analysis and predictability of 2011; 27: 1859–1869. placebo response. J Pharmacokinet Pharmacodyn 2014; 16. Rief W, Petrie KJ. Can psychological expectation models 41: 553–569. be adapted for placebo research? Front Psychol 2016; 7: 34. Vase L, Vollert J, Finnerup NB, Miao X, Atkinson G, Marshall 1876. eCollection 2016. S, et al. Predictors of the placebo analgesia response in 17. de Craen AJ, Roos PJ, Leonard de Vries A, Kleijnen J. Ef- randomized controlled trials of chronic pain: a meta- ana- Journal of Rehabilitation Medicine fect of colour of drugs: systematic review of perceived lysis of the individual data from nine industrially sponsored effect of drugs and of their effectiveness. BMJ 1996; 313: trials. Pain 2015; 156: 1795–1802. 1624–1626. 35. Häuser W, Bartram-Wunn E, Bartram C, Reinecke H, Tölle 18. Espay AJ, Norris MM, Eliassen JC, Dwivedi A, Smith MS, T. Systematic review: placebo response in drug trials of Banks C et al. Placebo effect of medication cost in Parkin- fibromyalgia syndrome and painful peripheral diabetic son disease: a randomized double-blind study. Neurology neuropathy-magnitude and patient-related predictors. 2015; 24: 794–802. Pain 2011; 152: 1709–1717. 19. Rossettini G, Carlino E, Testa M. Clinical relevance of 36. Schwarz KA, Pfister R, Büchel C. Rethinking explicit expec- contextual factors as triggers of placebo and nocebo ef- tations: connecting placebos, social cognition, and con- fects in musculoskeletal pain. BMC Musculoskelet Disord textual perception. Trends Cogn Sci 2016; 20: 469–480. 2018; 22: 27. 37. Martin RL, Davenport TE, Reischl SF, McPoil TG, Matheson 20. Zou K, Wong J, Abdullah N, Chen X, Smith T, Doherty M, JW, Wukich DK et al. American Physical Therapy Associa- et al. Examination of overall treatment effect and the pro- tion. Heel pain-plantar fasciitis: revision 2014. J Orthop portion attributable to contextual effect in osteoarthritis: Sports Phys Ther 2014; 44: A1–A33. meta-analysis of randomised controlled trials. Ann Rheum 38. Weimer K, Colloca L, Enck P. Age and sex as moderators Dis 2016; 75: 1964–1970. of the placebo response – an evaluation of systematic 21. Hróbjartsson A, Kaptchuk TJ, Miller FG. Placebo effect reviews and meta-analyses across medicine. Gerontology studies are susceptible to response bias and to other types 2015; 61: 97–108. JRM www.medicaljournals.se/jrm