The Effectiveness of Radial Extracorporeal Shockwave Therapy (rESWT), Sham- rESWT, Standardised Exercise Program or Usual Care for Patients With Plantar Fasciopathy. Study Protocol for a Double-blind, Randomised Sham- Controlled Trial

Plantar fasciopathy is a common cause of plantar heel pain, with reported lifetime prevalence up to 10%. The choice of best practice in these patients is debated. Two randomised studies reported that Radial Extracorporeal Shock Wave Therapy (rESWT) is effective, but a meta-analysis concluded that due to methological limitations the evidence is questioned. There are few studies reporting the effect on exercise programs with high-load strength training, despite widespread use. The usual care at our outpatient clinic is information on the pathogenesis, etiology and prognosis. We give advice on using proper footwear, including foot orthosis and to accept pain to a certain level during activity. To our knowledge there are no other previous trials comparing rESWT, sham-rESWT and exercises to usual care. This trial is designed in order to provide results important for future clinical practice.

The patients who give their informed consent will be randomised into one of the four Groups; rESWT, sham-rESWT, standardized exercise or usual care. A computer generation randomisation schedule with blocks of 8, in a 1:1 ratio, will be performed by an external statistician and electronically concealed. All the patients regardless of group allocation will receive standardized information (oral and written) on the condition and get custom made foot orthosis made by an orthopedic technician at Sophies Minde AS.

The patients are blinded for rESWT/sham-rESWT, whereas blinding of the exercise group and usual care group is not possible. To evaluate the blinding of the rESWT, the patients are asked after the last treatment whether they believe that they have received real rESWT, sham-rESWT or if they do not know.

The patients will have follow-up at 3 months, 6 months and 12 months. The main outcome is change in heel pain (NRS) during activity at 6 months. Patient characteristics, anthropometric data and the duration of symptoms in the different groups will be registered at baseline, and presented as mean values (SD) for continuous variables or numbers (%) for categorical data. Pain and function scores will be presented as mean (SD). In addition to the two- group comparisons with t-tests, we will perform longitudinal data analysis and perform a mixed model analysis to compare differences between groups at follow- up with adjustment for scores at baseline and present mean differences (95% confidence intervals).

We will perform analysis regarding secondary outcome as stated above. In addition we will apply multivariable logistic and linear regression analysis to identify predictive factors as demographics, clinical and ultrasound findings for primary and secondary outcomes. Model building will be done in a way that is appropriate for the given sample sizes, by restricting the number of potential predictive factors and considering shrinkage methods to stabilise predictions.