A prospective randomized clinical superiority trial is performed with a balanced randomization (1: 1). Reporting of the protocol follows the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) statement (33) (checklist uploaded as Additional file). Before the inclusion of the first participant, the trial was registered at ClinicalTrials.gov (NCT03854682).
Participants and recruitment
Individuals referred to the orthopaedic department with plantar heel pain will be examined and screened for potential participation. The recruitment will be performed by a specially trained healthcare personal (SM), who has a special interest in the PF pathology. The recruiting healthcare personal is also trained and experienced in ultrasound scanning of the plantar fascia thickness with regards to the diagnostic evaluation.
Participants will attend the baseline and a link to the questionnaires used will be sent via RedCap (Vanderbilt University, Nashville, TN, USA) to participants’ e-mail addresses for the 4-week, 12-week, 26-week and 52-week follow-ups.
The diagnosis of PF is based on the diagnostic criteria according to the literature (5, 6, 21): pain must be well defined with initiation of pain after rest (first-step pain) and the gait pattern should be changed to relieve the foot. Palpation is performed to identify the area of pain, which should be located at the proximal plantar fascia insertion area and the patient is examined for differential diagnoses (e.g. Tinel´s sign over the tarsal tunnel, fat pad syndrome, anamnesis of bilateral debut of symptoms). Findings are verified with ultrasound, where a plantar fascia thickness of > 4 mm. should be present.
Patients diagnosed with PF, who are older than 18 years, have well-defined plantar heel pain, a pain intensity of 30/100 within the last 7 days (VAS), palpation tenderness at the plantar fascia insertion, and experience initiation pain after rest (first-step pain) are offered participation in the study. The duration of symptoms must be over 9 months. People with systemic diseases, diabetes, pregnant women, previous heel surgery, Tinel's signs over the Tarsal tunnel, have received medical or physiotherapy treatment / cortisone injection within the last 3 months or have a fascia thickening (<4 mm) are excluded. See table 1 for in- and exclusion criteria’s.
Patients are informed orally and in writing and if they are willing to participate, they will sign a declaration of consent before inclusion. After this, randomization will take place. Patients who wants a consideration period will be given a statement of consent and "subjects' rights" and will be contacted by telephone after 3 days.
All patients are informed about the latest knowledge within the area of PF, including the expected prognosis, as well as risks. They will also receive information about the trial’s content, purpose, and time horizon, including a right to withdraw their consent at any time.
It is permitted that patients who have performed self-administered treatment in the form of foot massage, thermotherapy and stretching of the fascia may continue with this, provided this has been done for a minimum of 4 weeks before inclusion.
Shortly after the written consent is signed the patient will be contacted by a research secretary for further allocation. The randomization is performed by a research secretary with a software program (RedCap), which generates the allocation sequence in blocks of 10 patients, where 5 will be randomized to HRST treatment and 5 will be randomized to RFM treatment. The randomization within the blocks of 10 is random and is done without stratification. The randomization is done by the secretary in a separate room before contact.
The nature of the trial means that the group allocation cannot be hidden (blinded) from the patient or the researchers involved. To reduce the risk of bias, the assessor and statistician is blinded and patients are encouraged not to state their assigned treatment in case of contacts.
Based on a standard deviation of 19.7 points, a two-sided significance level of 0.05 and a power of 80%, it will require 32 participants in each group to be able to detect a difference in the clinically relevant 14.1 points in the pain domain of FHSQ-DK (34). To account for
possible drop-outs, we will include a total of 70 participants.
Prior to the study, a pilot study was made in which the patients' preference (n = 24) for treatment type (due to patients' equipoise) (35), was compared to symptom duration and previous treatment. Patients were presented with a short, standardized text and video of the two types of intervention and asked if they would choose the non-surgical or surgical treatment. No significant preference was found compared to duration of symptoms or prior treatment.
All statistical analyses have been planned in cooperation with a statistician assigned to the statistical department of OPEN (Odense, Denmark), that are responsible of data storage in RedCap. The primary analysis will investigate the between-group difference in FHSQ-DK pain. The analyses will be performed by a blinded data analyst (at group level) using a linear mixed effect model with the participant as a random effect and time (baseline, 4, 12, 26 and 52 weeks) and group allocation (HRST or RFM) as fixed effects. The model will include the baseline measurements with the constraint of no group difference at baseline.
An intention-to-treat analysis is planned, where a per protocol analysis could be carried out (as a supplementary analysis), provided that a very varying following is seen with the intervention procedures. Missing outcome data will be imputed using a multivariate model based on a normal distribution.
It is also planned that analyses of the mean values of the secondary continuous outcomes will be by using linear mixed models. Resample-bootstrap will be used in case there are substantial ambiguous data for these outcomes.
The time of when PASS is obtained between the two groups are compared. If a participant changes PASS multiple times (e.g., achieving PASS before 12 weeks, reporting not to have achieved PASS at 26 weeks and then having achieved PASS again at the 52-week follow-up), only time to the first PASS achieved is used in the analysis (36). We will calculate the relative risk of achieving PASS at each follow-up.
In an additional analysis, we will by linear mixed models investigate if anthropometric baseline values e.g. gender, age and length of symptoms influence the primary outcome.
Dropouts will be registered and reasons for this noted.
All participants are provided with a relieving heel insert or they can continue with their own preferred insert. This is permitted, since no difference in the effect of different inserts have been found (37).
Both groups will receive access to video-based patient education to support the adherence to the recommendations during an intervention (38-41).
The possibility of informative control sessions is offered in each group. The training group meets for physical control of the exercise itself and clarification of any relevant question will be answered. The surgical group can call if guidance is needed. If they present any signs of complications, this will be handled by on-call staff and registered and reported cf. regulations.
The treatment is performed percutaneously through plantar access with the patient under local anaesthesia. Pre- and postoperative information is given orally and they have access to a video presentation with information and educational content (imparting knowledge of loading, pain and adaptation) on the procedure and post-operative regime. The treatment is performed in an area of approx. 4 x 5 cm corresponding to the affected area of the fascia and injected with Lidocaine and Adrenaline. 25 independent insertions are made through the skin with 2 mm k-wires. Then a probe is inserted and radiofrequency energy discharge is performed in 2 rounds, partly corresponding to the surface of the fascia plantaris (light contact of tendon tissue with the tip of the probe) and subsequent perforation in the tendon substance itself. Water cooling is performed at 3 drops per second.
Compressive bandages are then applied. The patient is instructed in a sedentary to light activity regimen for the first two days without support on the foot, due to the risk of bleeding. Removal of the large dressing is done by the patient at home after 3-4 days and the inner bandage can be removed after a week. The patient is guided in unloaded activity with venous pump exercises for the first 2 days and then the patient must start with partial loading the following 3-4 days with increased load to full weight bearing after 1 week.
The time of return to work is divided into the degree of strain of standing and walking activity. Light work and weight-bearing activities (approx. 7d.), Moderate (approx. 14 d.), Hard (approx. 4 weeks). The procedure is performed by an experienced chief physician in orthopaedic surgery, who has worked with RFM treatment since 2013.
The non-surgical treatment consists of a specific strength exercise (HSRT), according to the protocol of Rathleff et al. (2015), which consists of a one-legged heel raise with a rolled-up towel under the forefoot (20). The exercise should activate the windlass effect and increase the mechanical stress on the tendon (fascia plantaris) (42). The towel should ensure the maximum comfortable dorsiflexion of the toes. Patients are instructed to perform the exercise on a stair tread, a thick book or similar, so that the heel during the exercise is lowered below the horizontal plane. To ensure maximum effort and avoid postural balance issues, the patient should secure support from e.g. a railing or wall structure. The exercise is performed every other day with as many sets as possible, and as heavy as possible, but not heavier than the patient can perform 8 repetitions per set.
The exercise is modified to be auto-regulated and not fixed, in the effort to improve patients’ adherence with the training (21). Autoregulation refers to individuals self-selecting the exercise dosage (sets) based on their individual circumstances, response to exercise and readiness to train (43).
The load is pro- and regressed from two- to one-leg +/- backpack loading. The exercise is performed as 3s / 2s / 3s concentric, isometric and eccentric, respectively, followed by a 2-minute break. The patient is encouraged to train for a minimum of 3 months and can stop his or her training 4 weeks after a satisfactory symptom state (patient acceptable symptom state) is achieved to emulate clinical practice. Pain during and after exercise is scored according to a numerical 0 - 10 scale in an assigned app (genoptræn.dk). Alternatively, a training diary in paper form can be handed out. Patients are instructed reduce the load accordingly if the pain during or shortly after the exercise exceeds 5 out of 10.
Outcomes / Variables
Follow-up is planned at baseline and 1, 3, 6 and 12 months. Primary end-point is 6 months.
The assessment schedule is found in the SPIRIT figure (Fig. 1). During the screening, the clinical examination and follow-up, we will collect the following data (see below).
The primary outcome is the mean pain domain score of the FHSQ-DK after 6 months (44).
FHSQ-DK consists of 13 questions that are divided into four domains: pain, function, footwear and general foot health. It takes the patient less than 10 minutes to complete. FHSQ scores for each domain range from 0 (worst) to 100 (best). The minimal clinically important
difference of the pain domain is reported to be 14.1 points, 7.4 points for function and 9.2 points for footwear (34). FHSQ-DK is a Danish validated translation of the original questionnaire (44). FHSQ has been reported as the preferred questionnaire for patients with plantar fasciopathy (45).
The selected HRST regimen has shown a larger pain reduction after 3 months, but not after 12 months compared to a stretching group (20). And the HRST regimen may not be sufficient to achieve patient acceptable symptom state in most people with plantar fasciopathy after 12 weeks (21). It can take time to manifest changes to a training regimen and therefore, we aimed for a primary end-point at 6 months.
- The other domains of FHSQ-DK (function, footwear and general foot health)
- Numeric Rating Scale (NRS) for pain during and after training (genoptræn.dk).
- Global Perceived Effect scale (GPE). Overall satisfaction with result and treatment measured by using GPE which is a recommended responder criterion and is rated on a 7-point Likert scale. A subjective assessment of change is made for the areas; pain, symptoms, activity and treatment (46, 47).
- Physical Activity Questionnaire (IPAQ). Physical activity is measured by self-reported participation in sports and leisure activities is measured by IPAQ. The questions relate to the time spent on physical activity during the last 7 days. The patient indicates the type of activities performed and divided into number of days or time in minutes / hours. Assessment is categorised into work, activities that form part of the work at home or in the garden, activities to get from one place to another, and leisure time activities related to relaxation, exercise or sports are assessed (48, 49).
- Patient Acceptable Symptom State (PASS). PASS is used as a measure of the time when the patient's symptoms are acceptable and no further treatment is needed (50).
Ultrasound scan for diagnostic value
All patients will be examined with an ultrasound scan of the plantar fascia thickness at the insertion of calcaneus. The positive predictive value for assessing an ultrasound scan is 0.83-0.90, while the negative predictive value is 0.57-0.89 in a group with symptoms (51-53). To increase the reliability, an average of 3 measurements is made (54). Here a plantar fascia thickness above 4 mm. must be seen around the proximal insertion of calcaneus (55).
Other outcome variables
Adherence is measured as the number of performed training sessions. "Very good" is achieved when at least 75% of the training is completed. "Good adherence" is achieved by completing 50-74%, "moderate" by 25-49% and "bad" by less than 25%.
a. Adverse events
- Number of sessions (% of total) that are completed with self-reported pain > 5 on the NRS 0-10 scale.
- Number of sessions (% of total) that are not completed due to pain or challenges related to the affected leg / foot. The cause for not completing is shortly noted.
Adherence and pain is recorded as an immediate self-reported assessment after the exercise using the NRS scale. The pain score is grouped as 0-2 (safe), 3-5 (acceptable) and> 5 (high pain risk). Thus, the total percentage of completed training sessions with acceptable pain can be assessed and evaluated in relation to plantar-specific adverse effects.
b. Barriers towards training
It is assessed on a Likert scale whether the patient experiences any barriers towards a training protocol before the start of the intervention and is compared with adherence towards the agreed upon training sessions and primary outcome.
Age, sex, weight, height, BMI, right / left side symptoms, symptom duration, number of sick days before and after intervention, level of education, type of job (sedentary vs. active) and consumption of painkillers, smoking and co-morbidity are recorded. Self- administered treatment is registered. Complications and side effects are collected at follow-up or reported by on-call staff in acute cases.
The study complies with the Declaration of Helsinki and the study is approved by the regional science ethics committee and the Danish Data Protection Agency before commencement. The potential risks and benefits of participation are presented to patients and the necessary insurance conditions are ensured before start-up.