This is a single-center, two-armed, parallel-group RCT that will be conducted at Beijing Hospital of Traditional Chinese Medicine Affiliated to Capital Medical University. The protocol has been registered on the Chinese Clinical Trial Registry (No. ChiCTR2100043005) and will be conducted following the Declaration of Helsinki. The study will be reported following the SPIRIT guidelines (Additional file 1). Figure 1 shows the flow diagram of the trial.
Study setting, recruitment, and ethics
The recruitment will be conducted at Beijing Hospital of Traditional Chinese Medicine. The study has been approved by the medical ethical review committee of Beijing Hospital of Traditional Chinese Medicine (No. 2020BL02-057-02). KOA patients diagnosed according to the American College of Rheumatology (ACR) criteria  will be recruited from outpatients in the pain department and the massage department, meanwhile, posters will be posted on the we-media and the outpatient hall. According to the principle of fair distribution of burden and benefit, relevant information will be released through subject recruitment announcement to attract the attention of interested subjects. Researchers will screen and qualify the subjects for the study according to the inclusion criteria and exclusion criteria. Before signing the written informed consent, the researcher needs to explain the study in detail with potential patients, for example, the study purpose, procedures, as well as the potential risks and benefits associated with participation in the study. The written informed consent will be provided by all eligible participants. The confidentiality of patient records will be protected. Data will be processed in a "data anonymous" manner, omitting information that can identify individual subjects. If the results of the trial are published, the identity information of the subjects shall remain confidential.
1. Age 40–80 years, male or female
2. Diagnosis of knee osteoarthritis according to the ACR criteria
3. Symptoms have been present for more than 6 months
4. Radiologic confirmation of knee osteoarthritis (Kellgren-Lawrence grade  II to Ⅳ in the recent 6 months)
5. An average knee pain severity of 4 or more out of 10 on an 11-point numerical rating scale (NRS)  over the past week
6. Signed written informed consent
1. History of knee surgery or waiting for surgery (knee replacement or knee arthroscopy)
2. Knee pain caused by other diseases (such as autoimmune diseases, infection, malignant tumours, trauma, fracture, severe effusion of joint cavity, joint bodies, etc).
3. History of arthroscopy within 1 year or intraarticular injection within 6 months
4. History of receiving acupotomy or acupuncture treatment within 3 months
5. Severe acute/chronic organic or mental diseases
6. Pregnant and lactating women
7. Coagulation disorders (such as hemophilia, etc.)
8. Participation in another clinical study in the past 3 months
9. Fear to acupotomy
Randomization and allocation concealment
Eligible patients will be randomly assigned to the acupotomy group (A group), or the diclofenac diethylamine emulgel group (D group) by using block randomization in a 1:1 ratio. The randomization sequence will be generated by independent research staff with SPSS 20.0 software and will be put into the sealed and opaque brown envelopes with a serial number affixed to ensure the concealment of the random allocation scheme.
The nature of acupotomy means that patients and acupotomy doctor cannot be blinded during the treatment. Outcome assessors and statisticians will be blinded to group assignments. In this study, acupotomy doctor, outcome assessors, and statisticians will be independent.
Patients in both groups will be prescribed diclofenac diethylamine emulgel (Voltalin, Beijing Novartis Pharma Ltd) of 3–4 times per day. A proper amount of diclofenac diethylamine emulgel will be applied according to the size of the knee pain area, then gentle rub will be needed to make it penetrate the skin. The course of treatment will be 4 weeks. During the study period, loxoprofen sodium tablets (Loxonin, Dachi Sankyo Pharma Ltd) will be provided as rescue medication in case of intolerable knee pain, with the usage documented by the outcome assessors. Besides, all other interventions, such as acupuncture, moxibustion, physical therapy, surgical procedures, and intra-articular injections for KOA, will be forbidden. Other medications that do not influence pain will be allowed.
The patients in the acupotomy group will receive acupotomy one time per week for 4 weeks. A sterilized, disposable needle-scalpel (Hanzhang No.4, length: 60 mm, diameter: 0.6 mm, Beijing Hanzhang Medical Device Co., Ltd) will be used. To keep the consistency and standardization of the operation, the acupotomy will be performed by an experienced doctor who has the Chinese medicine practitioner license and has been qualified for at least 15 years according to the basic operating procedures of acupotomy in the Principles of Acupotomy. The appropriate body position will be chosen depending on the location of the operation, such as supine, prone or side lying position. The medial collateral ligaments, lateral collateral ligaments, hamstring muscle, quadriceps femoris, iliotibial band, popliteus muscle, anserine bursa, subpatellar fat pad, and the peripatellar area will be touched, and pressed, then the spots with induration and cordlike tissue will be identified as positive points. Generally, 5–7 positive points will be selected in light of the specific pain sites, functional limitations, and imaging examination results of the patients' joints. The selected points will be marked with gentian violet, and then the surrounding skin will be disinfected with iodophor. After sterile skin preparation, a needle-scalpel will penetrate the skin and gradually go deeper. When feeling a sense of toughness or resistance under the needle-scalpel, the manipulation of acupotomy is performed. After treatment completion, the doctor will check for any abnormality and bleeding, then apply the Band-Aid to the pinhole.
If only one knee meets the ACR criteria and Kellgren-Lawrence grade II to Ⅳ, then this knee will be evaluated. When both knees are in accordance with the inclusion criteria (ACR and Kellgren–Lawrence grade II to Ⅳ), the more painful knee will be chosen for assessment. Figure 2 shows the treatment and outcome measurement schedules.
Primary outcome measurement
The primary outcome will be the response rate. The response rate is the percentage of patients with improvement in average pain NRS of at least 2 points  and in the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC)  function subscale score of at least 6 points  at week 4 compared with baseline. The NRS is a digital scale using “0” to “10” to indicate the severity of the average pain over the last week, with “0” representing painless and “10” representing the most severe pain. The WOMAC function subscale (Likert version 3.1) consists of 17 questions (each scored from 0 to 4) relating to the difficulty experienced due to knee osteoarthritis, with higher scores referring to poorer knee function.
Secondary outcome measurements
The NRS and WOMAC pain subscale will be used to measure the average pain over the last week at weeks 0, 4, 8, and 24. WOMAC pain subscale ranges from 0 to 20. It has five items assessing the severity of pain in five different conditions, with higher scores indicating worse pain.
The WOMAC function subscale will be used to measure the physical function at weeks 0, 4, 8, and 24. The WOMAC function subscale ranges from 0 to 68 and includes 17 items, with lower scores indicating better function.
The WOMAC stiffness subscale will be used to measure the degree of stiffness of the knee at weeks 0, 4, 8, and 24. The WOMAC stiffness subscale ranges from 0 to 8 and includes two items, with higher scores indicating more stiffness.
The 12-item Short Form Health Survey (SF-12)  will be used to assess the quality of life at weeks 0, 4, 8, and 20. SF-12 includes the mental and physical domains. Each domain ranges from 0 to 100, with higher scores indicating a better quality of life.
The use of rescue medicine (loxoprofen sodium tablets) will also be recorded throughout the trial.
Laboratory outcome measurement
Joint fluid and venous serum will be collected at week 0 and week 4, and the levels of tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β) and matrix metalloproteinase-3 (MMP-3) will be tested.
All adverse events will be recorded throughout the trial by researchers using a record book, which consists of the time, severity and duration of the adverse event, the disposition taken and the prognosis of the event. Common treatment-related adverse events include bleeding, subcutaneous hematoma, infection, continuous post-needling pain, dizziness, syncope, etc.
Data collection, management and monitoring
The Case Report Form (CRF) will be completed in paper form by a data entry operator (DEO) and then entered into an Excel spreadsheet by two electronic data interchange (EDI) administrators independently to ensure the accuracy of the data. The researchers will keep all trial data, including signed informed consent, CRF, operation procedure records, etc. The results of this study may be published in peer-review medical journal, but we will not disclose the privacy of patients. The research supervision department, the ethics committee of the hospital, and its relevant personnel will inspect the trial per year.
Before the inclusion of patients, all the researchers will be trained on the implementation program, including scale evaluation, data measurement, and the CRF filling, etc. A special quality examiner will be set up to check the quality of the research regularly. The CRF should be strictly managed, and the researchers should truthfully, meticulously record as required to ensure that the contents of the CRF are reliable and completed. Regular follow-up will be conducted to ensure the compliance of patients during the study, and the drop-out rate should be controlled within 20%.
The primary outcome in this study is the response rate, and the allocation ratio between the two groups is 1:1. A bilateral difference test will be used to calculate the sample size. According to previous review, the response rate of the diclofenac diethylamine emulgel for KOA is assumed to be 60% . The response rate in the acupotomy group is assumed to be 80% based on preliminary clinical experience. The sample size of each group is 80 calculated by PASS 11.0.7 (NCSS, LLC, USA) software with a two-sided significance level of 5% and power of 80%. Considering a 20% drop-out rate, the final sample size of each group will be 100, with a total of 200.
SPSS 19.0 software will be used. For all statistical tests, P < 0.05 will be considered to be statistically significant. Continuous data will be presented as mean ± standard deviation (M ± SD), while dichotomous data will be presented as a percentage. The independent sample t-test or χ2 test will be used to compare demographic data and other baseline indicators to measure the equilibria between the two groups.
Covariance analysis will be used for continuous data and logistic regression analysis will be used for dichotomous data, which will be adjusted according to baseline and dosages of loxoprofen sodium. The results will be presented as mean difference or odds ratio, providing a 95% confidence interval (CI). Intention to treat analysis (ITT) will be carried out for all the randomized patients. The last observation carried forward (LOCF) method and the maximum likelihood regression will be used for the missing data of primary outcome.
The per-protocol (PP) set will include only those who complete at least three sessions and all planned follow-ups, and have no major protocol violations (taking other drugs during the trial, etc.). The ITT set will be used primary analysis, and the PP set will be used as in the sensitivity analysis.