Study design and setting

We followed the design, conduct and reporting of clinical trials on KOA to report our trial protocols. The study protocol was registered with the China Clinical Trial Registry (registration number: ChiCTR2300077308). The study will follow the rigorous standards outlined in the Declaration of Helsinki and the Standard Protocol Items: Recommendations for Interventional Trials.30 This study protocol was approved by the Ethics Committee of Shuguang Hospital, Shanghai University of Traditional Chinese Medicine (2023-1357-124-01). We plan to conduct a single-centre, randomised, parallel, controlled trial at the Shuguang Hospital, Shanghai University of Traditional Chinese Medicine, which is a tertiary hospital. The study duration will be 24 weeks. Eligible and consenting 106 patients with KOA will be randomly allocated in a 1:1 ratio using a computer-generated random number to the experimental group (n=53) or control group (n=53) to receive treatment for 8 weeks. Participants will be assessed at baseline and at 4, 8, 12 and 24 weeks post-intervention to determine whether better long-term outcomes are achieved with TCM massage and exercise than with exercise alone. A flowchart of the study process is shown in figure 1. The timelines for patient enrolment, intervention and assessment are presented in table 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Trial design of the research procedure.

Recruitment and informed consent

All participants will be recruited from the hospital and community. The trial will be designed with an open recruitment advertisement to recruit patients online or offline (eg, WeChat public website). Orthopaedic, TCM and rehabilitation physicians will strictly determine whether patients are eligible for this study based on the inclusion and exclusion criteria, and screening of participants will be conducted by a regular non-participant in the study. If the patients who are interested and volunteered to participate in this trial meet the inclusion criteria, they will sign a written informed consent form before the start of the study. All participants will receive nominal compensation (including a gift voucher for a future massage). Participants will be fully informed about the study, including its procedures, benefits and potential risks, except for information about other in-group interventions. They are free to withdraw from the study at any time without penalty or any negative impact on their future medical treatments.

Study population

Randomization and blinding

The study will be randomised and parallel-controlled, with a separation of efficacy evaluators, patients and statisticians, and will not be blinded to the patients or physicians. Indicator evaluators and statisticians will be unaware of the subgroups, and therapists will be unaware of the outcome evaluation. The assessor will be blinded to follow-up outcome assessment. A random number table will be generated by a statistician using statistical software and distributed to third-party researchers for allocation concealment. Third-party personnel will place the numerical tables in opaque envelopes, each marked with a time and signature, and hand them uniformly to the study coordinator to ensure adequate randomisation concealment. The study coordinators will contact potential participants through telephone to inquire about their willingness to participate in the clinical trial. Individuals who meet the inclusion criteria will be invited to the biomechanics laboratory for baseline testing, where subjective scale scores, objective physical function data and biomechanical data will be recorded. Each participant’s sex and severity of condition will be collected. After completion of baseline testing, the participants will be allocated to the test and control groups according to random numbers on a 1:1 basis. Stratified randomised subgroups will be formed based on the order of random allocation using a block randomisation grouping method with block lengths set at two or four to ensure that patients in each group are matched for sex and severity of KOA.

Intervention

All patients will receive health education that includes awareness on KOA disease and treatment goals, explanations of issues related to joint loading and dosage of physical activity/exercise, encouragement on how to make healthy lifestyle changes by adopting behaviours that may have a beneficial effect on their condition or at least prevent worsening it, management of unexpected high-risk events, and advice on how to control their weight and maintain a normal weight.33 Details of the health education programme are displayed in online supplemental appendix 1.

Experimental group

Patients in the experimental group will undergo an exercise programme and TCM massage. Considering the patients’ clinical symptoms and level of functioning, the exercise programme will be individualised to meet their specific needs. The exercise programme will be based on the American College of Sports Medicine guidelines on exercise for older adults, which encourages patients with KOA to be actively involved in exercise.34 The exercise goal is 50%–70% increasing in the heart rate reserve. Participants will engage in a total of 24 group training sessions of 20–30 min exercise three times per week for 8 weeks. Prescribed exercise programmes will include aerobic walking and resistance training of the affected limb.10 Information on pain level, local vs central sensitisation, range of motion and weight-bearing tolerance within the effective range of motion of the knee will be obtained from the patient’s past and present medical history and clinical physical assessment. This information will be used in the baseline setup of the exercise; the physical therapist selects a starting position (figure 2) and weight resistance that will be matched to the patient’s ability to perform two sets of 10 repetitions. To achieve the desired dose of exercise, there will be constant communication between the physiotherapists and the patient, and perceived exercise intensity (rate of perceived exertion) will be assessed at the completion of each exercise. The principles of self-reduced stress will be applied to ensure that the exercise is comfortable for the patient and is performed with little or no pain. The exercise logs include attendance, number of exercises, duration of each exercise, number of repetitions and number of sets. At the end of the exercise intervention period, the patients will be advised to maintain their normal daily activities.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Resistance training—guiding the patient to produce the correct quadriceps power pattern.

After the exercise programme is conducted, TCM massage treatment will be performed by licensed doctors from Shuguang Hospital with at least 3 years of clinical experience who have undergone specific training in tissue manipulation before the start of the trial. Prior to each treatment, the practitioner will check the patient’s current signs and adjust the massage intensity accordingly. After skin disinfection, the physician will press and relax the affected limbs (figure 3). The doctor will then apply superficial-to-deep pressure to the acupuncture points around the knee joint with the thumb to stimulate the acupuncture points and induce a localised sensation of soreness, distention or irradiation in patients who feel comfortable with the applied pressure. Each acupoint, including Liangqiu (ST34), Xuehai (SP10), Heding (EX-LE2), Dubi (ST35), Neixiyan (EX-LE4), Yinlingquan (SP9), Yanglingquan (GB34), Zusanli (ST36), Weizhong (BL40), Chengshan (BL57), Yingu (KI10), Ququan (LR8) and ashi point, will be pressed for 30 s (table 2). The treatments will be administered every alternate day for a week. The treatment sessions will last for approximately 30 min, and the entire intervention will last for 8 weeks.35

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Release massage techniques for the lower extremities.

View this table:

• View inline
• View popup

Table 2

Location of acupoints

Concomitant medications

All patients will be advised to avoid using any other treatments, such as painkillers, glucocorticoids, other physical therapies, etc, for KOA during this period as much as possible through the informed consent form. However, if the patient had intolerable pain and the outcome was not assessed within the next 48 hours, the use of painkillers will be permitted. During the study, any other therapies used by patients to relieve KOA pain must be known to the researcher, who will record the therapy details, including the name of the treatment, dosage and duration.

At the first visit, participants will provide information on any other illnesses they currently have and the corresponding medications they use, all of which will be recorded in case report forms. All treatments administered will be recorded, including the name of the drug, its dose, the duration of treatment, and the start and end of dosing times.

Follow-up and outcomes

Knee pain, stiffness, active function, muscle recruitment and proprioception during walking, as well as patient satisfaction and treatment safety, will be measured at 4, 8, 12 and 24 weeks after randomisation of all participants.

The following measures will be taken to retain participants as much as possible: free medical consultation and guidance advice for other diseases without compromising KOA treatment; a daily schedule of optional treatment times and giving patients a 1-week notice to schedule their treatment for the week according to their schedule; adoption of a micro-appointment system with micro-text and phone calls to remind participants of upcoming treatments; and obtaining permission. The telephone numbers of the next of kin will also be recorded to facilitate follow-up.

Outcome assessments will be performed by a commissioner not involved in the intervention process and completed by the patients in a separate space in the outpatient department at each participating site. All outcome assessors will be trained in interviews and measurements before the start of the study and will follow a standard protocol.

Primary outcome

Secondary outcomes

The timed "up and go"

The timed up-and-go test is a valuable indicator of balance and general motor ability in older people and is a quick, meaningful and objective measure of functional mobility.39 It is widely used in many other neurological and neuromuscular disorders as well as in several other disease states; it has excellent internal reliability and very high retest reliability.40 It measures the time taken for the patient to stand up from the armchair, walk a distance of 6 m, turn around, walk back to the chair and then sit down, with more time taken indicating poor mobility.41

Emg data

EMG data will be collected from the affected limb using a Noraxon Ultrium surface EMG telemetry device (EMG software version: MyoResearch XP Basic edit on 6.59, EMG data acquisition frequency of 2000 Hz) manufactured by NORAXON (USA). The target electrodes will be mounted in the order of the MyoResearch software target electrode apparatus, defining odd numbers as left and even numbers as right, and labelling the target muscles according to numbers 1–6: medial femoral (ie, medial quadriceps), lateral femoral (ie, lateral quadriceps), biceps femoral (ie, lateral popliteal), semitendinosus and semimembranosus (ie, medial popliteal), medial gastrocnemius and lateral gastrocnemius muscles. The EMG signal from the gait test will be normalised to maximal voluntary contraction. Record the limb dynamics and muscle co-contraction in the gait of the patient while walking.42

Knee proprioception

Knee proprioceptive testing will be performed using an isometric muscle strength assessment training system (BIODE MEDICAL SYSTEMS, USA; Model: System 4 Pro).43 (1) Muscle force acquisition: the test protocol will be set for unilateral isometric centripetal exercise with an angular velocity of 60°/s at low speed and a cycle termination of 5 times as a reflection of the maximum muscle force of the muscles measured. The peak moments, peak moment ratios of the anterolateral homonymous muscle groups, and peak moment ratios of the ipsilateral flexors and extensors will be selected. The peak moment is the maximum moment output generated during muscle contraction. (2) Positional sensory acquisition/positional sensory perception: passive flexion of the knee will be done at a constant angular velocity starting from the fully extended knee position at 45° and 60° for 10 s to allow the patient to fully perceive. Active angular error: after returning to knee extension, patients will actively perform a knee flexion movement when they believe that the target angle has been reached; the movement will then be stopped, and the difference between the actual and target angles will be measured. The system automatically calculates the difference between the target angle and current angle for the active angle reproduction (AAR) value. The AAR will be used to evaluate knee position awareness, with a smaller AAR representing better position awareness.44

Treatment satisfaction

Patient satisfaction with rehabilitation will be collected at 4, 8, 12 and 24 weeks using a 20-cm Visual Analogue Scale anchored with ‘no satisfaction’ (0 cm) and ‘complete satisfaction’ (20 cm).45

Safety

To describe the distribution of adverse events, adverse reactions, significant adverse events and serious adverse events in participants and compare the incidence using the χ² test or Fisher’s exact probability method, we will tabulate the details of adverse events during treatment and analyse whether the adverse event was related to the treatment. Reporting of serious adverse events: all serious adverse events, whether related to the intervention or not, will result in immediate withdrawal of the patient from the trial, prompt treatment and reporting to the relevant unit within 24 hours.

Monitoring

To avoid atypical findings, each questionnaire will be reviewed before data entry and double-checked to be error-free before storing and backing up. Scale assessors will be professionals who are familiar with the study’s guiding phrases and entries and understand the meaning of all entries, to ensure correct interpretation. Problems identified during the assessment will be promptly addressed and resolved by the researcher. The scale will be administered in a quiet, comfortable environment. The purpose and precautions of the assessment will be explained to the participants before the assessment begins, and the assessment will be conducted entirely voluntarily. After completing the assessment, the researcher will check and supply missing items to ensure that the information is complete and accurate. The ethics committee and sponsor require that all relevant information should be recorded in the case files.

Patient and public involvement

Patients and the public are not involved in the development of the research questions or study design. They are not involved in recruitment, implementation or reporting of the study. The results of the study will be disseminated to study participants via social media.

Sample size

The effectiveness of TCM massage in the treatment of KOA was reported in a previous study26 to be approximately 60%. However, the proposed combination of TCM massage and exercise in this study is expected to increase the effectiveness by 25% (85% effectiveness). A completely randomised design estimated that 48 observations would be required in each group (P1=60%, P2=85%, α=0.05, β=0.20), and if 10% shedding was expected, 53 cases would need to be enrolled in each group, resulting in a total of 106 cases in two groups.

Statistical analysis

Excel 2019 (Microsoft) and Statistical Package for the Social Sciences (SPSS) Statistics V.25.0 (SPSS IBM) will be used to analyse the data. First, we will perform normality and variance χ² tests on the obtained data to determine the appropriateness of subsequent analysis. Continuous data will be expressed as mean±SD and IQR. For data that satisfied the normal distribution and variance χ² test, we will use t-tests to compare the differences in means between the groups. If the data are not normally distributed, non-parametric tests will be used. Categorical data will be expressed as percentages (n %), and baseline analysis will be performed using the χ² test or Fisher’s exact test.

The initial data analysis will be performed according to the intention-to-treat principle. Data of randomised participants, regardless of the actual treatment received, will be coded according to their allocation status. In preliminary analyses, to deal with repeated measures and account for inter-individual variability, we will use linear two-way analysis of variance (ANOVA) comparing outcomes at each time point between the two intervention groups, with time as a categorical variable to analyse the interaction between time and intervention at all time points except baseline, with age-stratified and sex-stratified factors as fixed effects, and with missing data supplemented with random interpolation. Secondary analyses will be performed using a conforming protocol analysis, including only participants who successfully completed the trial according to the intervention plan and did not receive any other treatment in the previous 4 weeks. During the analysis, we adjusted the p values for two-by-two post hoc comparisons using the Bonferroni method to control for statistical error rates. Additionally, Pearson’s correlation coefficients will be calculated to assess the strength of the association between secondary and primary outcomes.

Study duration

The trial recruitment period is 12 months, and it commenced in September 2023. The final participant was recruited in September 2024.