This study is a multicentre randomised controlled clinical trial study conducted in four clinical centres (Figure 1). The Wuxi Hospital of Traditional Chinese Medicine is the main centre. The Yueyang Hospital of Integrative Medicine affiliated with the Shanghai University of Traditional Chinese Medicine, Wuxi Rehabilitation Hospital and the Wuxi Xinwu District Hospital of Traditional Chinese Medicine are the three subcentres. Patients were recruited through advertisements at each centre. Patients who met the inclusion criteria were randomly assigned to either the TCE group or the Tuina combined with TCE group at a 1:1 ratio. The trial consisted of a four-week intervention period and subsequent three- and six-month follow-up periods. Data were collected before treatment, after 4 weeks of treatment, and at 3 months and 6 months of follow-up. The outcome assessment indicators included the Oswestry Disability Index (ODI), Short Form of Quality of Life (SF-36) Scale, Short-Form McGill Pain Questionnaire (SF-MPQ) Scale and gait analysis, which was used only for efficacy evaluations before and after 4 weeks of treatment.

The study was approved by the Ethics Committee of the Wuxi Hospital of Traditional Chinese Medicine (STHZG2023021301), and it was registered in the China Registered Clinical Trial Registration Centre (ChiCTR2300077361). The study strictly followed the criteria outlined in the Declaration of Helsinki, the Consolidated Standards of Reporting Trials (COMSORT) and the Standardised Program Item: Recommendations for Interventional Trials (SPIRIT) guidelines (25).

Subjects were recruited at each subcentre through a variety of channels, including printed advertisements, flyer distribution, telephone inquiries, and public announcements on the hospital’s WeChat public website. Each subject participating in the trial must meet the inclusion criteria. Subjects who meet any of the exclusion criteria subject will be excluded. The principal investigator will not request that the subject sign an informed consent form until the subject has had sufficient opportunity to understand the nature of the trial and to agree to participate in it.

The random sequence was generated by a statistician independent of the study through a central randomisation system. The random numbers and group assignments were provided to the independent researcher through the central randomisation system. This researcher was not involved in any other aspect of the trial. The physician only obtained the grouping information from this independent researcher by telephone before the treatment. The central randomisation system had strict limitations on the permissions granted, and only the supervisory staff had access to these documents. This procedure ensured adequate concealment of the randomisation. The study was not blinded to the patients and therapists, but it was blinded to the evaluators. This evaluator was independent of this study. The TCEs therapists were separated from the Tuina therapists, and neither group was aware of the grouping of the patients. Patients were not allowed to interact with the therapist beyond general counseling about their health status at each session.

The trial was designed as a superiority trial with ODI as the primary observation index. The previous study demonstrated that the mean and standard deviation of ODI after 4 weeks of intervention in the TCEs group were 19.73 and 1.94, respectively, whereas the mean and standard deviation of ODI after 4 weeks of intervention in the Tuina combined with TCE group were 14.53 and 2.60, respectively. The sample size was calculated using the following formula:

The sample size for each group was 74. Because this trial was a superiority trial, the sample size of each group was the same, and the clinical trial was set up in two groups (Tuina with TCE group and TCE group), combined with the relevant regulations. To evaluate the efficacy and safety of this study fully, the total number of cases was planned to be 166 cases, considering the potential for adverse effects. The Wuxi Hospital of Traditional Chinese Medicine was responsible for this project and was the main centre where the case study was undertaken. In other subcentres, the number of cases can be adjusted according to the situation, but no fewer than 30 cases should be in each group in each centre. If enrolment cannot be completed in any centre due to insufficient suitable subjects, the cases were distributed among the centres.

This study is a multicentre clinical trial. To ensure the consistency and standardisation of treatment modalities across all clinical centres, therapists from the main centre conducted training sessions for therapists from the subcentres. The objective of these sessions was to standardise the Tuina treatment protocols and the TCE movement, and to treat patients in accordance with the interventions outlined in the study protocol. Prior to intervening with patients, therapists at subcentres must undergo training and evaluation at the main centre. They are not qualified to intervene until they have met the training requirements. Furthermore, to mitigate bias in the treatment of patients by different therapists, only one Tuina therapist and one TCE therapist were appointed to conduct the intervention at each clinical centre. TCE therapists must have at least 10 years of experience teaching TCEs.

To ensure that each subject can master the movement essentials of TCEs, a TCE treatment manual was issued to the subjects before the intervention. A comprehensive introduction to the movement essentials and precautions of TCEs through offline lectures and online video teaching mode followed. The therapist maintained a treatment log for each subject and recorded the subject’s progress and any issue that arise. If a subject has not completed the prescribed treatment within the allotted time, the therapist will contact the subject by phone to ensure that the subject has met the standard treatment volume.

The TCE group comprised four movements, selected from Baduanjin, Taiji and Wuqinxi, with minor modifications to accommodate the patients’ varying learning and exercise abilities. The program primarily encompassed the fourth movement of Baduanjin, the Yunshou of Taiji, Huju and Luben from Wuqinxi. All movements were supervised by qualified TCE instructors. Firstly, patients performed warm-up exercises for about 5 min to avoid muscle strain during exercise. Then, the TCEs were performed in the order of A–D (Figure 2), and each movement was repeated 3 times, with 4 movements as 1 set, for a total of 2 sets of exercise, which should last 15 min or more. Finally, a 5-min stretching exercise was performed. Treatments were performed 3 times a week for 4 weeks.

The Tuina procedure was divided into localised muscle release and lumbar joint adjustment, and the whole process lasted for 20–30 min (Figure 3). The procedure started with gentle rubbing and kneading on both sides of the patient’s spine and buttocks using the fingertip surface for 10–15 min, and then pointing and pressing the local pain points or acupoints (e.g., BL23, BL24, BL25, BL40, and GB30) for 5 min. Following the completion of local muscle release, joint adjustments were performed utilising a lumbar blique-pulling manipulation and posterior lumbar extension. The lumbar blique-pulling manipulation was comparable with chiropractic manipulation (27), where an audible “click” indicated the successful completion of the procedure. Finally, lumbar back extension was performed. The patient was initially placed in the prone position, and the therapist pressed one hand on the patient’s spinal pain and slightly lifted the patient’s lower limbs with the other hand. The therapist then waited for the patient’s lumbar muscles to relax. Then, the therapist quickly lifted the patient’s lower limbs up to 30°–40° and then placed them down immediately. This maneuver can be done 3–5 times for about 5 min. Following Tuina, the patient should be permitted to rest in bed for approximately 10–15 min. The training movements and methods employed by the TCEs were identical to those utilised in the control group. Treatments were performed 3 times a week for 4 weeks.

The efficacy evaluation in this study concentrated on the primary outcome indicator, namely the recovery of dysfunction in patients before, after and during the follow-up period. In addition, several secondary indicators were evaluated, including quality of life, pain and gait analysis. The data collection at each time point was conducted by independent researchers who were not privy to the specific subgroups and were not involved in other aspects of the study. The specific time points for data collection are presented in Figure 4.

The ODI scale, which consists of 10 items in 3 dimensions, namely pain, individual functioning and overall personal functioning, was used as a comparison of efficacy before and after treatment and between groups (28). The maximum score was 50; the higher the score, the higher the degree of dysfunction of the patients.

The SF-36 Scale evaluates health in eight dimensions, including physical functioning, general mental health and social functioning, and higher scores indicate better health (29). Each dimension was rated with a maximum of 100 points and a minimum of 0 points.

The SF-MPQ Scale includes three parts: pain rating index (PRI), (VAS) and present pain intensity (PPI) (30). The PRI assesses the patient’s emotional and sensual experiences of pain. The VAS provides a straightforward visual representation of the patient’s pain level. The PPI determines the intensity of the patient’s current pain on a six-point scale (0–5).

Gait analysis is the process of collecting patient information using a wearable 3D gait analyser. Gait analysis encompasses a range of patient data, including step length, cadence, stride length and double support phase. This information is utilised to assess the patient’s motor function and musculoskeletal rehabilitation status (31).

After all the relevant items were confirmed by the patients themselves, the total score reflecting the degree of low back and leg pain in each period was calculated cumulatively by a specialised statistician, and the results were used for pre- and post-treatment comparisons as well as comparisons between groups.

The treatment responses of all patients with lumbar intervertebral disc herniation who received Tuina in conjunction with TCEs were monitored and documented throughout the course of the study. Adverse events during the treatment period were recorded and analysed to ascertain whether it was related to the treatment. Adverse events included those that occurred during the clinical trial and resulted in hospitalisation, prolonged hospitalisation, disability, impairment of the ability to work, life-threatening injury or death. In case of any serious adverse event, whether related to the intervention, the trial should be terminated immediately and appropriate rescue treatment initiated without delay. Such incidents must also be reported within 24 h to the responsible unit, the ethics committee, the sponsoring organisation and the relevant personnel of the drug regulatory authority. The relevant information must be recorded in the case file.

Prior to the implementation of the trial, ensuring that all study personnel received a uniform, centralised training was necessary to achieve an accurate understanding and mastery of the various criteria for cases (diagnosis, inclusion, exclusion, exclusion, shedding and termination criteria), the technical protocols of the treatment, the observational indicators and the scoring criteria. The case report form (CRF) was completed in its entirety and in accordance with the prescribed specifications, with no arbitrary alterations. Furthermore, observations of the same subject were recorded by the same individual whenever feasible.

A paper-based case record form was employed for data collection. The healthcare professionals proactively cooperated with the patients to complete the assessment of each outcome indicator and tolerantly answered the patients’ questions related to each assessment scale. Feedback was provided to patients promptly following each outcome assessment. The sub-centers used a uniform case record form for data collection. The case record forms were then reviewed by the subject leader and research assistant before submission to the data manager for data entry and management. The data were entered and managed by the data managers using an Excel database. To ensure the accuracy of the data, double-entry assessment and proofreading was conducted independently by the two data managers. In the event of uncertainty regarding the data entry on the case record form, the data administrator will complete the question-and-answer form and forward a query to the individual responsible for the subject matter through the research assistant. The individual responsible for the subject matter should provide a response as soon as possible. The data administrator will then modify, confirm and enter the data in accordance with the response provided by the individual in charge of the subject matter. Once the database has been confirmed to be error-free, it will be locked by the subject person in charge, the data manager and the statistical analyst. The data collected from subjects will not be made public without their consent.

The data will be subjected to analysis by our team of professional statisticians, who will employ IBM SPSS Statistics V.27.0 for this purpose. To assess the normality of the data, the Kolmogorov–Smirnov test was used. The quantitative results were expressed in terms of the mean and standard deviation. The purpose of the primary and secondary outcome indicators was to assess the difference in efficacy between the two groups. Hypothesis testing was employed to this end. The chi-square test was used to ascertain the dissimilarities in baseline data between the two groups. The analysis of variance for repeated measures was utilised to examine the alterations in efficacy between the two groups before, during and after the intervention, as well as after the follow-up period. For confounding factors that were challenging to regulate or were not regulated prior to treatment, such as an imbalance between the groups prior to treatment, were treated as covariates, and logistic regression was employed to ascertain the variability of efficacy between groups and to eliminate the influence of these factors on efficacy. The data collection and data summarisation were meticulously examined to minimise data anomalies. In the event of missing data, the data were assumed to be missing at random, and multiple interpolation would be employed for analysis. Subjects who were excluded from the trial will be excluded from the data set. The level of statistical significance was set at 5% (p < 0.05), and 95% confidence intervals were employed as effect estimates.