Trial design
This will be a randomised, double-blind, parallel positive controlled, multicenter clinical trial, each participant will sign a informed consent form (ICF) before the research was performed. A total of eleven medical institutions are involved in the study, subjects will be enrolled at eleven hospitals, including Longhua Hospital Affiliated to Shanghai University of TCM, The first Affiliated hospital of Guangzhou University of TCM, Zhengzhou Central Hospital, Suzhou Hospital of TCM, Luoyang Orthopaedics-Traumatological Hospital, Xiangyang First People's Hospital, Liaoning Hospital of TCM, The Second Hospital of Nanjing Medical University, The Fourth Central Hospital of Tianjing, Changchun University of TCM, Shandong University of TCM. Longhua Hospital Affiliated to Shanghai University of TCM take charge of the total clinical scheme design. The study protocol has been approved by the ethics committee Shanghai University of TCM on the use of human subjects for research (approval number: 2016LCSY097), each participating center has conducted ethical filing and agreed with the ethical approval of the main centre hospital.
The study phases are shown in Fig. 1. A total of 2000 patients with KOA will be recruited and randomly allocated into experimental group (n=1500) or control group (n=500), each patient will undergo a 2-week treatment with herbal patches for one session per day, a flow chart of trial participation is provided in Fig. 2. Efficacy and safety data will be collected throughout the whole study.
Inclusion criteria
The following inclusion criteria should be met:
1) Participants who have symptomatic KOA, diagnosis of KOA was based on criteria developed by the American College of Rheumatology (ACR) in 1986 [17];
2) Standard of TCM disease and syndrome diagnosis[18, 19];
3) Symptomatic KOA with a pain of at least 30 mm on a 100-mm visual analog scale (VAS);
4) The patient who were 20 years of age;
5) All the patients signed the ICF before study begins.
Moreover, if the patient has osteoarthritis in both knees, we chose the more severe side of the knee-joint. If the pain scores of two knees are the same, the researchers chose one side of the knee-joint for intervention according to the research demands.
Exclusion criteria
1) There is a history of trauma or surgery in the knee joint in the last six months before the trial began;
2) Arthroscopy and intra-articular injection was performed in the last three months before the trial began or hormone therapy was used in the first month of screening or participants who have underwent knee arthroplasty;
3) Subjects are undergoing or have participated in other clinical trials in the last three months;
4) Participants with other knee joint diseases such as chondromalacia patellae, rheumatic arthritis or rheumatoid factor positive (RF>40) will be excluded from the trial;
5) Patients has mental disorder or severe diseases and complications such as severe diabetes, serious liver and kidney disease, malignant tumors, infectious diseases or complications affect the joints;
6) Plaster allergy and pregnant and lactating women were not considered for this study.
Intervention
All patients will be randomly divided into Tong-luo Qu-yu plaster group (experimental group) and Qi-zheng Xiao-tong plaster (control group), The experimental group and the control group received Tong-luo Qu-yu plaster and Qi-zheng Xiao-tong plaster respectively. Tong-luo Qu-yu plaster is a tape-type patch, which is made by Henan Lingrui Pharmaceutical Ltd, the validity period of Tong-luo Qu-yu plaster is 24 months. Qi-zheng Xiao-tong plaster is a medicated plaster made by Tibet Qizheng Tibetan Medicine Ltd, it is valid for 36 months, Two kinds of plasters are identical to the tested formulation in terms of texture, size, color, and odor. Both groups use the plaster according to the instructions given by nurses, all participants will receive a conventional conservative treatment lasted for 14 days as two courses, daily 1 time. Patients will have three follow-up visits, the clinicians, subjects, investigators and assessors are masked to treatment allocation. In the process of trial, patients are not allowed to use other TCMs. The subjects were in severe pain with VAS scores above 80 mm, the researchers can be allowed to give celebrex for relieving pain within two daily dose, if patients have other accompanying disease that have to take the medicine for treatment, the principle is to not affect the evaluation of the trial drugs.
Safety assessment
Adverse events (AEs) concerning clinical symptoms and signs as well as laboratory tests will be documented during clinical trials. Skin irritation will be recorded by Berger Bowman scoring system [20], subjective symptoms including itching, pain, burning sensation and the skin lesions manifested as erythema, papules, edema, blisters, erosions, skin ulcer, and so on will be recorded after 1 and 2 weeks of the treatment. The drug safety monitoring is conducted by blood routine examination (BRE), urine routine test (URT), liver function tests (LFTs) including serum glutamic oxaloacetic transaminase (AST), serum glutamic pyruvic transaminase (ALT), serum total protein (TP), serum alkaline phosphatase (ALP), serum total bilirubin (TBIL), kidney function tests (KFT) including blood urea nitrogen (BUN) and serum creatinine between the start and end of the trial. erythrocyte sedimentation rate (ESR), antistreptococcus hemolysin (ASO), rheumatoid factor (RF), electrocardiogram (ECG) and X-ray examination is also detected. Serious adverse events (SAE) will be reported to the local drug administration authorities in 24 hours.
Outcome assessment
Assessment standards is based on the Chinese traditional medicine new drug clinical research guiding principle and standards for diagnosis and curative effect of Chinese medical symptom [18, 19].
Primary Outcome
The Western Ontario and McMaster universities osteoarthritis index (WOMAC) as objective indicators of efficacy is the primary efficacy endpoint of the study, which is a widely used, proprietary set of standardized questionnaires used by health professionals to evaluate the condition of patients with osteoarthritis of the knee and hip, including pain, stiffness, and physical functioning of the joints [21]. The WOMAC measures 5 items for pain (score range 0–20), 2 for stiffness (score range 0–8), and 17 for functional limitation (score range 0–68) to assess the severity of the arthritis and its therapeutic effect according to the symptom and sign of the patients. It can fully reflect the basic situation of osteoarthritis [22]. The primary outcome is the improvement of the total WOMAC scores, which will be measured during the assessment visits (baseline and 1-, 2-week follow-up).
Secondary outcome assessment
The secondary outcome is the changes in TCM syndrome quantitative score [19], visual analog scale (VAS) scores [23], onset time of drug for pain relief between baseline and the end of treatment. The VAS scores range from 0 mm to 100 mm, which is widely used for the clinical evaluation on the degree of pain. Onset time of drug for pain relief is determined by the time that was taken in the VAS scores fell at least 10 mm at the first time.
Sample size estimation
Our study was designed as a non-inferiority trial, sample size calculations are based on the primary outcome measurement. First, minimal clinically important differences for WOMAC scales in KOA are estimated as 15.50 points from previous studies [24]. Second, we assume that square deviation of WOMAC score is 318.88 based on previous literature [25]. Considering a power of 80 %, and a alpha value of 2.5 % (two-tailed), the sample size will be calculated using the following formula:
n=(4/3 (u_α (+u)_β )σ^2)/(∆-δ)^2 (u_α=1.6449, u_β=1.2816)
Thus we obtained the sample size of 1600 patients for this trial, allowing for a conservative 20% dropout rate, the total sample size was set to be 2000 patients (1500 in Tong-luo Qu-yu plaster group).
Randomization assignment
This study is designed as a randomized, double-blind, parallel positive drug control, multi-center clinical trial. A total of 2000 eligible participants will be randomized (3:1) using A stratified-block randomized method based on the disease and the center, select the appropriate length to one of the two treatment groups: experimental group (Tong-luo Qu-yu group) and control group (Qi-zheng Xiao-tong group). SAS statistical software 9.2 (SAS Institute, Cary, NC, USA) will be used to generate a randomization scheme based on the PROC PLAN function, which will be used to link the patient to a treatment arm and will specify a unique medication number for the first package of study drug to be dispensed to the patient.
Drug management
During the study, one experimental drug administrator was set up to be responsible for drug save, delivery, recording and recycle. The whole process of drug coding is written by a blind coder and forge a document. The drug boxes and emergency envelope contains the corresponding drug number are randomly divided into each center according to the central number of random stratification. Emergency unblinding can only happen when there was a serious adverse event in the study, the subject will exit the trial and the investigator should record detailed reason for withdrawing from the trial in the case report form.
Statistical analysis
The statisticians are responsible for the statistical analysis plan with the main researchers, SAS 9.2 statistical software was adopted for all statistical analyses. Data sets including full analysis set (FAS), per-protocol set (PPS), safety set (SS) are analyzed, actual subjects, shedding cases, excluding cases, demographic and characteristics of cases, efficacy and safety analyses will be conducted according to the intention-to-treat (ITT) principle.
Categorical data including frequency tables or percentages and continuous variable data including presented as mean±standard deviation, median, superiorand inferior quartiles, minimum value, maximum value will be used to describe characteristics of patients in both groups. Primary outcome will be compared between both groups, categorical data will be conducted by performing a chi-squared test or Fisher's exact test, continuous data fitted normal distribution is performed by t tests or the variance test method. If the data doesn't satisfy the normal distribution or homogeneity of variance, it will be analysed using a Wilcoxon rank sum test or Wilcoxon symbols test to compare the two treatment arms. A two-sided P value of ≤ 0.05 or ≤ 0.01 is considered to be statistically significant.