Tuina on knee pain and functional decline of lower limbs for patients with mild-to-moderate knee osteoarthritis in Shanghai: protocol for a multicentre, assessor-blinded, randomised controlled trial

Hypothesis

We will conduct this RCT to hypothesise that the application of Tuina as a complementary therapy would reduce pain levels, improve the quality of life and enhance physical function in patients with KOA. The objectives of this study are to validate the efficacy of Tuina in relieving pain and improving physical function in individuals with mild-to-moderate KOA compared with celecoxib.

Study design and settings

This study is a 12-week, multicentre RCT conducted in eight public hospitals at various levels (Shanghai Municipal Hospital of TCM, Yueyang Hospital of Integrated Traditional Chinese and Western Medicine, Shanghai General Hospital, Shanghai Jing’an District Hospital of TCM, Shanghai Jiading District Hospital of TCM, Shanghai Qingpu District Hospital of TCM, Shanghai Jing’an District Central Hospital, Shanghai Jiangning community health service centre) from February 2023 to December 2025.

Patients who meet the inclusion criteria will be recruited and randomly assigned equally into one of two groups: the Tuina group (TG) or the celecoxib Group (CG). Participants in TG will receive Tuina treatment two times a week for 4 weeks, each treatment lasts for 30 min. In CG, participants will be required to take 200 mg of celecoxib orally daily for 4 weeks. At the end of the intervention, an 8-week follow-up will be conducted, and relevant outcome measures will be assessed at 8, 12 weeks of the follow-up phase. All participants will complete the assessments by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), 36-item Short-Form Health Survey (SF-36), Timed Up and Go test (TUG), Short Physical Performance Battery (SPPB) and Gait analysis (detailed trial process seen in figure 1 and table 1). All patients will be completely informed of the potential benefits, potential risks and responsibilities, those who will sign the informed consents once they are included. Participants will receive brief reports at the end of the study, which demonstrate the changes before and after the intervention.

Figure 1

Study flow chart. SF-36, 36-item Short-Form Health Survey; SPPB, Short Physical Performance Battery; TUG, Timed Up and Go test; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index.

Participants

The trial will recruit patients who had symptomatic, radiologically confirmed KOA according to criteria from the American College of Rheumatology.45 To precisely screen the target patients, the following criteria were developed for this clinical trial.

Inclusion criteria

All patients with mild-to-moderate KOA will be assessed to see if they meet the following inclusion criteria:

1. Men or non-pregnant women.

2. Aged ≥40 years and ≤70 years.

3. Meeting the diagnostic criteria for OA of the knee.45

4. Mild-to-moderate radiographic tibiofemoral OA (Kellgren and Lawrence grades I–II) in at least one knee.46 47

5. WOMAC pain subscore ≥3 on at least one of five questions.48

6. Unilateral knee involvement.

7. Understand Chinese and speak Mandarin Chinese.

8. No previous knee surgery and no accompanying other knee problems.

9. Volunteer to participate in this study and provide written informed consent.

Exclusion criteria

Participants who meet any of the following criteria will be excluded from this trial:

1. History of trauma and surgery on knee.

2. History of other treatment in last 6 months.

3. Have tumours, tuberculosis, osteomyelitis and other diseases in the knee.

4. Have severe liver and kidney dysfunction, severe cardiovascular disease, diabetes mellitus and mental illness.

5. Suffer physical pain caused by other diseases.

6. Non-steroidal drug allergy.

7. Alcohol and drug abuse.

8. Voluntary withdrawal.

9. Current participation in another study.

Recruitment

We will recruit participants from eight public hospitals by recruitment posters. Patients with KOA who show interest in this trial could make a call or scan the QR code on the poster to get in touch with our trial contact person and will be informed to have a face-to-face interview in any of eight public hospitals. On meeting, the inclusion criteria and expressing their willingness to participate in this study, participants will be requested to provide written informed consent and undergo a comprehensive baseline assessment. Patients who have previously participated in trial design consultations will not be recruited as study participants.

Sample size calculation

The calculation of sample size was based on the hypothetical changes in the outcome measurement, WOMAC pain subscale, at week 4. According to our previous clinical trial,43 we formulate expected effects with WOMAC pain subscale as the primary outcome. We hypothesised that a 2-point improvement in WOMAC pain subscale occurred in the TG. Group sample sizes of 142 achieve 90% power to detect non-inferiority using a one-sided, two-sample t-test with the significance level (alpha) of the test is 0.025 through PASS software (V.15.0.5).49 According to the conservative principle and the ability of each centre to enrol patients, considering a drop-out rate of about 20%, the study will require 360 patients in total.

Randomisation and blinding

This trial adopts stratified block randomisation and will be completed by an independent statistical expert team. Eligible patients will be randomly allocated on a 1:1 basis to either TG or CG by a computer-generated random sequence in R (R Project for Statistical Computing, cloud.r-project.org). Randomisation will be stratified among the eight public hospitals and the random block design will be adopted in which the block sizes will be changed randomly. Random allocation hiding will be implemented through random envelopes. Two copies of the random envelope will be given to researchers and each hospital. The randomised envelope of group information stored with the study researchers shall not be opened during the duration of the study.

Due to significant differences in intervention, this trial will be an open one in that researchers and participants will not implement the blind method. However, group concealment can still be implemented through blinding outcome evaluators, data monitors and statistical analysts. Data monitors who will perform data entry, query and data management and statistical analysts will be blinded as to the allocated intervention until all statistical analyses are completed. Each part of this trial will remain separate and independent.

Intervention

Before the trial officially starts, all the physicians and therapists involved in this study will be trained in a standardised operating procedure to acquire a full understanding of this trial. We do not recommend that patients continue to receive additional interventions during this trial. Medications will be permitted if necessary, and the type and dosage of medications will be recorded timely. Participants will not bear the treatment costs during this study.

Celecoxib group

Participants in CG will take 200 mg of oral celecoxib (Pfizer, USA.) daily for 4 weeks. After randomisation, the patients will be required to make weekly visits to the hospital in order to collect their medications for the upcoming week and will be reminded to take their medications by daily text message or email. The whole process will be carried out under the guidance of physicians and medication records will be completed by regular telemonitoring.

Tuina group

Tuina treatment will be performed in the outpatient department of each pilot. The Tuina doctors involved in this study must hold the Chinese medicine practitioner licence from the Ministry of Health of the People’s Republic of China. To minimise the gap of the manipulation caused by the different doctors, we will select those who have more than 5 years of experience for clinical experience and familiarise them with the treatment process. They will be required to strictly adhere to established procedures. Participants in TG will receive 30 min Tuina treatment two times a week for 4 weeks. The interval between treatments should be at least 24 hours. The Tuina treatment procedure comprises three steps, which involve manipulative techniques through massages on both sides of the lower limbs and passive knee joints mobilisation. Details of the Tuina process are depicted and explained in figure 2.

Figure 2

Process and details of Tuina therapy for the knee: Step 1: (A): Roll* back and forth (the forearm rotates to roll the back of the hand) on the posterior thigh, popliteal fossa and posterior leg with the back of the hand for 5 min. (B): Knead and Press* on Weizhong (BL40)** and Chengshan (BL57) for 30 s each. Step 2: (C): Roll back and forth on the anterior thigh with the back of the hand for 5 min. (D): Knead and Press on Heding (EX-LE2), Neixiyan (EX-LE4), Dubi (ST35), Yanglingquan (GB34), Yinlingquan (SP9) and Xuehai (SP10) for 30 s each. Step 3: (E, F): Press the knee joint with one hand and hold the ankle joint with the other hand to passively flex and extend the lower limbs for 1 min. *Frequency: Rolling 80–120 times/min; Kneading and Pressing 60–80 times/min. **All selected acupoints are present on both sides.

Outcome measurements

Based on the study schedule, all outcome measurements will be collected at different time points as the trial progresses.

Primary outcome

The primary outcome of this study is the change in WOMAC pain subscale (standardised range, 0–10 cm; 0 cm=asymptomatic, 10 cm=extreme symptoms) from week 4 to baseline. WOMAC pain subscale will be measured at baseline, 4 weeks of treatment, 8, 12 weeks of follow-up. WOMAC is a self-reported Visual Analogue Scale specifically designed to assess OA symptoms, knee pain in five states will be assessed through five items. Participants will rate these five items according to their actual situation and will get a total score to reflect the pain level of the knee joint. The higher the score, the worse the knee pain and the worse the physical condition.50

Secondary outcomes

1. WOMAC total score and WOMAC stiffness and function subscale

   • WOMAC stiffness subscale which contains 2 items will be used to measure the stiffness of the patient’s knee joint and the WOMAC function subscale also contains 17 items to assess knee joint physical function. The scores of each item in both scales are from 0 to 10 and will be assessed at baseline, 4 weeks of treatment, 8, 12 weeks of follow-up. WOMAC total score ranges between 0 and 240 (higher scores indicate more severe symptoms). The reliability, validity and sensitivity of the knee joint assessment will be evaluated objectively through these scores.

2. SF-36

   • SF-36 covers eight health concepts in separate scales: physical functioning, role limitations due to physical functioning, bodily pain, role limitations due to physical health problems, general mental health, role limitations due to personal or emotional problems, social functioning, energy/fatigue or vitality and general health perceptions. The eight scales are also scored to produce two summary component measures of physical and mental health and will be assessed at baseline, 4 weeks of treatment, 8, 12 weeks of follow-up.51–53

3. TUG test

   • The functional status of the knee joint will be measured at the end of the 4-week intervention through the TUG test. The test is performed so that the patient is properly positioned on the test stand. The patient sits in a steady chair, and after declaring readiness, the task is to get up, cover a distance of 3 m and return to the chair. The tests were performed in triplicate and the mean value was calculated. The normal value for a healthy adult is 10 s or less.

4. SPPB

   • The SPPB is a performance-based test that includes three components: walking velocity, time for five repeated chair stands and a balance test. A score of 0–4 points is assigned to each test (4=best), and scores are summed to yield an overall score ranging from 0 to 12 (12=best). This test is used to evaluate lower extremity function and will be measured after 4 weeks of treatment.

5. Gait analysis

   • Gait analysis parameters are associated with increased odds of KOA onset and progression.54 In this trial, 28 retroreflective markers will be affixed to the pelvis and bilateral lower extremities. Kinematic and kinetic data will be captured with a 16-camera motion analysis system (Vicon Vero V.2.2, Vicon, UK, 100 Hz) as participants walked over an embedded 6- m pressure distribution measurement platform (FDM-3, Zebris, Germany). Participants will complete overground gait analysis and these data will be recorded and held consistent within individuals across trials. Ten non-invasive wireless EMG electrodes (Noraxon Ultium, Noraxon, USA) will be used to collect sEMG data from five muscles, including rectus femoris, vastus medialis, vastus lateralis, biceps femoris and gastrocnemius in both lower limbs of participants.55–57 Additionally, three-dimensional (3D) kinematic and kinetic gait data will be calculated. Ground reaction force data will be tested by four-channel force plates (AMTI BP400600, AMTI, USA, 1000 Hz). Maximal knee compressive force, external knee adduction moment, AP shear force, vertical ground reaction forces, 3D moments and forces of hip, knee and ankle joints will all be collected and analysed at the same time. Moreover, gait kinematics of the lower limbs also include total time of one gait cycle (s), velocity (cm/s), step length (cm), knee joint angles (°) during walking. All data will be processed using Visual 3D, and gait analysis will provide us with functional descriptions of knee joints by comparing the objective differences in gait data before and after the 4-week intervention.

Drop-outs and adverse events

Considering the age of patients with KOA, it is possible that they have other chronic diseases. We will truthfully record the standard medication content of relevant diseases instead of discontinuing the drug medication during the course of the trial. Participants in all groups will continue taking their individual routine dosage from baseline to follow-up. If their conditions change during the study, they will be free to consult our therapist from eight hospitals to make adjustments based on actual conditions. For better implementation of standard medical therapy, researchers will try to carry out proper health education for all patients and supervise them in recording their daily medication dosage.

Any adverse events including unfavourable or unintended signs, symptoms or diseases due to oral administration of celecoxib need to be reported by participants and will be recorded during this trial. Participants who refuse to continue taking celecoxib medication could be considered a drop-out in the study.

Tuina therapy is a non-invasive, safe, manual treatment and does not rely on other equipment, but we will not completely rule out the occurrence of adverse events. Common adverse events were skin breakage, bruising due to local subcutaneous haemorrhage and soft tissue injury. The therapists must promptly report any adverse events, which will then be recorded.

Statistical analysis

The data will be analysed by professional statisticians who are blinded to group assignment and patients’ basic information. The principles of intention-to-treat analysis will be used. A full analysis set will include all qualified participants who have received at least one intervention and at least one outcome measurement. A per-protocol set will also be used to analyse those who completed the study without major protocol violations as an auxiliary set. Missing data will be completed with the last data recorded for each of these variables through last-observation-carried-forward.

To examine the baseline characteristics of the sample, a descriptive analysis will be conducted. The Kolmogorov-Smirnov test with Lilliefors corrections will be used to determine if quantitative variables follow a normal distribution. Initial homogeneity between groups will be assessed using analysis of variance or Kruskal-Wallis test based on data distribution normality. Continuous data will be presented as mean±SDs for normally distributed data or median and IQR for skewed data.

Changes in repeat outcomes over time will be compared by generalised estimated equation. The outcomes measured at baseline and after the 4-week intervention will be compared using a t-test or Mann-Whitney U non-parametric test, depending on whether they follow a normal distribution. The association between changes in WOMAC scale scores and physical parameters (TUG, SPPB, Gait data) will be examined through correlation analysis using Spearman or Pearson correlation coefficients for two continuous variables to help identify the uniformity between patient self-report and changes in objective measures.

All data will be analysed by using the statistical software SPSS (V.24.0). Values of p<0.05 (two sided) are considered to have statistical significance. Data analysis will start after the completion of the follow-up of the last participant.

Data management

The raw data will be captured using case report forms (CRFs, paper) at the designated time. Two data administrators, who are not affiliated with the research team and blinded to group allocation, will independently receive the completed CRFs and enter the raw data into an Excel database (Microsoft, Redmond, Washington, USA). Simultaneously, they will input real-time data into China Clinical Trial Registry (http://www.chictr.org.cn). Both administrators must undergo comprehensive training in data monitoring. To enhance data authenticity, we have established an independent data safety monitoring board (DSMB) responsible for reviewing and monitoring trial data. The DSMB comprises experts in Tuina therapy, orthopaedics, methodology and statistics who ensure the overall quality and completeness of the collected information. Committee members will conduct spot checks on raw data during meetings with evaluators to ensure adherence to protocol guidelines. They possess authority to terminate the study prematurely if necessary. On completion of the study period, all collected data will be locked and researchers will not have access to modify it thereafter. All CRFs will be securely preserved for a minimum of 5 years following article publication. In case reviewers have any concerns regarding our trial’s validity or reliability, they can contact the corresponding author for access to raw data.

Quality control

To ensure consistent procedures at all the hospitals before the official start of the trial, a trial kick-off meeting was scheduled, and all staff (except statisticians) were trained on the participants and content of the trial, patient recruitment, treatment scheduling, the precautions of patient–physician communication and outcome assessment. Uniform and standardised operating procedures are developed and provided to all hospital staff. After undergoing centralised training, the Tuina doctors in each subcentre will undergo assessment and be required to strictly adhere to the established Tuina process and time. Additionally, they are expected to provide appropriate knee joint health guidance verbally during treatment sessions. Dedicated personnel will be stationed at the subcentre hospitals to establish regular contact with the participants via telephone or email, ensuring timely medication intake, providing health guidance and maintaining accurate records of all medications and dosages administered throughout the study period. If the patient proactively discontinues medication due to significant symptom improvement, they will provide feedback to the attending physician through the dedicated personnel, signifying the completion of intervention. The established plan will be used to evaluate primary and secondary outcomes, record relevant data and proceed to the follow-up phase. Comprehensive review of completed cases and clinical procedures by independent monitors was required. Once they find problems or serious adverse events during the intervention, they can raise direct objections and even discontinue this trial.

Patient and public involvement

Prior to the design, the principal investigators consulted with patients who had KOA in clinical departments. The number and duration of treatments were designed from clinical experience and patient feedback.

Ethics and dissemination

This study protocol has been approved by the Ethics Committee of Shanghai Municipal Hospital of Traditional Chinese Medicine (2023SHL-KY-16-01, 2023SHL-KY-16-02), and any revisions or corrections of this protocol will be discussed by the Ethics Committee of Shanghai Municipal Hospital of Traditional Chinese Medicine. Changes of detailed information will be kept. The trial will not officially begin until written informed consent has been collected from all patients. After the study completed and the data analysed, results of this trial will be presented as articles in peer-reviewed journals or at academic conferences.

Trial status

At the time of submission, recruitment for the trial has been started. The first participant was included on 5 August 2023. The study is scheduled to end in December 2025.