This will be a parallel-group, randomised, controlled trial with two groups, using masked assessors to collect outcome data with a 1:1 allocation ratio. The trial was registered on September 17, 2021 with the Chinese Clinical Trial Registry (Registration number ChiCTR2100051275). There are no secondary registrations. It was considered on October 21, 2021 by the Medical Ethics Committee of the General Hospital of the Central Theatre of the Chinese People's Liberation Army (Registration number: [2021]052 − 01) The research is funded by Wuhan FL Medical Science & Technology Ltd. who have no other role. There are no secondary sponsors. There are no additional individuals or committees (groups) overseeing this trial. It started patient recruitment on November 3rd 2021.
Public and scientific queries will be taken by Lixia Huang, MD (email: [email protected]), or via the contact details given on the title page.
This protocol adheres to the SPIRIT statement [4] and contains the relevant items from the World Health Organisation’s Trial Registration Data Set. [5] It also adheres to the relevant parts of the CONSORT statement for pilot and feasibility studies [6] and considers the guidance on reporting protocols for pilot and feasibility studies. [7] The schedule for the trial is shown using the SPIRIT figure (figure one) and CONSORT flow diagram (figure two)
Figure one
Figure two
Recruitment and selection.
Patients will be recruited from out-patients and inpatients attending the department of Orthopaedics at the Central Theatre General Hospital of the Chinese People's Liberation Army, a large class III class general hospital. There will also be general publicity within the community. All patients attending the hospital, and all patients in contact with the research team will be given an information leaflet. This leaflet will explain that the patient should be willing to consider participation in a controlled clinical trial, receiving one of the two treatments and that they will need access to a mobile phone. The first contact will usually be with the clinic doctor.
Any patients interested in participating in the trial will be given further information and explanation by the clinician they are seeing or, more commonly, by a trained researcher. If they want more time to consider participation, a further appointment will be arranged. Otherwise, if they remain interested, then the research doctor will obtain informed consent. The researcher will check clinical eligibility.
To be included, the patient must be an adult aged 18 years or more and must have a clinical diagnosis of osteoarthritis in one knee. [8] If both knees are affected, the primary data will come from the knee considered by the patient to be more severely affected.
A patient may then be excluded if they:
-
have additional knee pathology such as joint dystrophy, infected joint, rheumatoid arthritis, osteonecrosis, cruciate ligament injury or post-traumatic arthritis.
-
have had a hip or knee replacement surgery or tibial osteotomy.
-
have used steroid medication orally over the previous month or injected within the last three months.
-
have a neurological or mental disorder that affects cognitive ability sufficiently to affect participation in the exercises and/or the research.
-
are pregnant, anticipating becoming pregnant, or breastfeeding.
-
have a body mass index of 36kg/m 2 and above.
Sample size.
There are no data available to allow an estimate of the relative effect size of the comparator intervention (pendulum swinging of the leg). Sample sizes between 24 and 50 are recommended for feasibility studies to estimate standard deviations in measures for use in a future sample size calculation for fully powered testing [9][10]. A total sample size of 50 will allow us to estimate a drop-out rate of 80% to within a 95% confidence interval of +/- 11%”, [1, 2]. A sample of 50 will also allow for process and feasibility issues to be explored with an emphasis on the context and mechanism of using the system (data collection and feedback). This sample size is supported but other empirical work. [11]
Registration and random allocation.
Once a patient has been recruited, has given consent, and has passed the selection criteria, the clinician or researcher will register the patient by telephoning the research office where an independent researcher will allocate a patient to their group using a simple 1:1 random allocation process. Before starting, a series of numbered envelopes will be filled with a group allocation using a random number generator. The envelopes will be opaque and sealed. The researcher will give the patient the next registration number and open the pre-filled numbered envelope. The treating researcher will be informed of the allocation.
Data collection.
Data will be collected by researchers who are not informed of a patient’s allocated group. Patients will be asked not to disclose their group. At the end of each research assessment, the person collecting data will be asked to indicate which group they think a patient was in. If the patient has told them, they should give that group.
The following demographic and clinical information will be extracted from the clinical record:
-
gender, age, weight in kilograms, height in centimetres, estimated age of onset
-
Kellgren-Lawrence X-ray grading of the most affected knee [12]
The following clinical data will be collected
-
Pain, measured from 0–10 on a numerical rating scale from 0, “no pain”, to 10, “worst pain”. [13]
-
Oxford Knee score [14][15]
-
Six-minute walking distance, measured in a gym or corridor. [16]
-
30-second Sit-to-Stand test [17]
-
strength of the quadriceps muscle, using a hand-held dynamometer [18]
-
gait kinematics using triaxial accelerometer [19]
The clinical data will be collected at baseline at the hospital, at 12 weeks (end of treatment programme) at the hospital and some data will be collected at 24 weeks by telephone.
In addition, data will be collected on the feasibility and experience of (a) the allocated treatment and (b) the research processes. These data will be recorded by the patient at home, with researchers collecting the recorded data from the patient. They include:
-
Recruitment rate to the study and reasons for non-participation
-
Adverse events
-
recording missing data from each measure at each point for both intervention arms
-
recording how many patients dropped out, when and if a reason is given, why
-
reviewing comments within treatment diaries
-
recording degree of reported adherence to the exercise intervention
-
reviewing the data collected during treatment sessions from the accelerometers or pedometers
Last, data will be collected during the exercise sessions undertaken by each patient. Each patient will be allocated four accelerometers, two for each leg. Two labelled accelerometers are delivered by a research team member to each patient each week while the two accelerometers in use are collected at the same time. Each accelerometer contains a MiniSD card, After collection of the accelerometer from a patient, the MiniSD card is removed by the researcher and the data are downloaded. The data is then cleared from the card which is then replaced in the accelerometers. The following week the two accelerometers will be returned in exchange for the pair used.
The accelerometer is strapped on at the ankle, with a clear sign indicating the position. The therapist will teach the patient how to apply the accelerometers. There is also a video for the patient. When the accelerometers are changed at the end of a week, the patient may report any difficulties, which will be documented. They may also be given further advice or teaching if needed. This will also be documented.
Furthermore, patients in each group will be asked to keep a daily diary.
In the pendulum therapy group, it is anticipated that some patients will find standing on one leg taking the whole-body weight for up to 50 minutes may cause pain or discomfort, and this will be recorded. No other specific adverse events are anticipated, but all patients will be told that they may rest if they experience unusual or troublesome pain, breathlessness, or other symptoms. They should keep a diary of their actual pendulum therapy each day, recording any symptoms they consider notable including symptoms that limit exercise.
In the walking therapy group, it is anticipated that some patients will experience knee pain during or within an hour after waking, and patients will be asked to record this. No other specific adverse events are anticipated, but all patients will be told that they may rest if they experience unusual or troublesome pain, breathlessness, or other symptoms. They should keep a diary of their walking therapy each day, recording any symptoms they consider notable including symptoms that limit exercise.
The data at baseline and 12 weeks will be collected when the patient visits the hospital, by masked researchers who will not be told the patient’s group. Patients will be asked not to disclose their group. Nonetheless, patients are likely to disclose something. Each outcome assessor will be asked to say which group they believed the patient to be in. The data at 24 weeks will be collected by telephone.
Figure one shows the schedule for the study including what data are collected when.
Treatments.
The patient will be allocated to one of two treatments: walking exercise, or pendulum therapy. All patients will be asked to undertake their allocated treatment on at least five days each week. Patients will be free to undertake more of their allocated treatment; those in the walking exercise group will be walking anyway. Patients in the walking exercise group will be asked not to undertake pendulum therapy for the 12 weeks of the trial. All patients will be given a diary to record the date and duration of each session they undertake. They will be asked to complete it each day.
Pendulum therapy.
To undertake this, the patient should stand on a platform 15 cm above the ground, with one leg standing on the edge of the platform as a supporting leg, and the other leg hanging. The patient should hold on to something with the arm on the opposite side to the swinging leg, to maintain balance. The subject should flex the leg at the hip (i.e. move the leg forward) to about 45–60 degrees. The knee should be kept extended. Then the patient should extend the leg at the hip (i.e. move it backwards) until it is extended by 15–20 degrees. The patient will swing their leg at a rate between 30–50 swings per minute.
The researcher will instruct the patient in the exercise and in fitting and activating the accelerometer. The researcher may also show the patient ways to reduce the stress on the leg bearing the whole-body weight.
The target is to undertake 1600–2000 swings (one swing includes forward and back, a whole cycle) each day, and the target swing rate is 40 swings a minute. At the first, instruction session, a researcher assistant will:
-
teach the patient how to fit, activate and use the accelerometer collecting data
-
teach the patient how to swing (setting up, and swinging)
-
observe the rate and any features such as fatigue and pain
-
agree on an initial target rate and session length or count
-
suggest a schedule for progression towards the target rates to be achieved over eight weeks, shown in Table one
-
suggest that, if fatigue or other symptoms limit a session, the patient undertakes two shorter recorded sessions each day
Table one
Patients will be reassured that they may undertake additional pendulum therapy if they wish and that they may walk and do anything else that they wish.
Walking therapy.
Patients in this group will also see a researcher to be assessed and taught how to use a pedometer and advised on walking as an exercise.
In the first session with a researcher, they will be given advice on the known benefits of walking exercise and observed walking for six minutes (possibly as part of the six-minute walking test).
The target will be to undertake 50 minutes of specific walking exercise each day. This can include walking for another purpose, such as visiting a friend or going to and from a shop, if it is a walk and not ‘wandering about a shop’. They also need to wear a pedometer strapped to the ankle. They will be given a similar graded programme of increasing exercise, moderated by the patient’s ability and symptoms. The schedule is also shown in Table one. As with the other group, an option of two shorter episodes will be given.
Patients will be reassured that they may walk as much as they like outside treatment sessions and undertake any other activities. The only exception is that they will be asked not to undertake any pendulum swinging during the first 12 weeks.
Data handling.
The clinical data will be recorded in a Clinical Research File, on paper, completed by researchers and signed by a doctor. The data will be entered into a computer database.
Data from the accelerometers worn during the exercise will be downloaded each week. The data will be stored and sent to data analysts who will analyse the data on a batch basis later.
Information on any reported adverse event will be documented in the patient’s Clinical Record Form and reported to the local principal investigator who will assess its significance and the likelihood it is related to the trial process. The outcome of that judgement will be recorded using the severity categories [20]
-
Grade 1 Mild; asymptomatic or mild symptoms; clinical or diagnostic observations only; no intervention indicated
-
Grade 2 Moderate; minimal, local or noninvasive intervention indicated; limiting age-appropriate instrumental ADL
-
Grade 3 Severe or medically significant but not immediately life-threatening; hospitalization or prolongation of hospitalization indicated; disabling; limiting self-care ADL
-
Grade 4 Life-threatening consequences; urgent intervention indicated.
-
Grade 5 Death related to AE.
The causal link, as determined by the principal investigator, between the treatment and the adverse event will also be graded:
-
Not related
-
Unlikely related
-
Possibly related
-
Probably related
-
Definitely related
Information reported on the interventions by patients when data are being collected will be noted in the clinical record.
All diaries will be collected at the 12-week follow-up appointment. The record of time spent each day will be transferred to the database. Comments will be transferred to the clinical record.
Telephone data at 24 weeks will be transferred to the clinical record and entered in the database.
Data analysis.
Most of the analysis will be descriptive.
The feasibility criteria we have set are:
- missing data from each measure at each point, <10% at baseline and <20% 12 weeks
- the criterion set for 24 weeks is <30%
- patients dropping out, <5% at 12 weeks, <30% at 24 weeks
- achieving target (self-reported and monitored) goals of exercise by six weeks
- 70%
- a further 30% will be within 20% of their target
- self-reported and monitored achievement in the last 6 weeks
- recruitment,
- >70% of eligible patients
- >40% of patients initially approached
- reviewing the data collected during treatment sessions from the accelerometers or pedometers
We will compare the demographic data at baseline, and the clinical data associated with osteoarthritis at baseline and at 12 weeks. We will use a t-test for parametric data and a Mann-Witney U test for non-parametric data. We do not expect a statistically significant difference between the groups.
We will estimate the number of patients needed to determine a difference using the Oxford Knee score at 24 weeks of a clinically meaningful difference of 5.0 points (95% CI 4.4–5.5), and the six-minute walking distance at 12 weeks.