Trial design
The study was a single-blind, double-arm, parallel-design RCT and was randomized into a TA group and a SA group two groups in a 1:1 ratio. The following is a flow chart (Figure 1).
Participants
Recruitment advertisements were disseminated via social media and Outpatient departments in the hospital. Interested patients could consult us by provided phone number or Wechat account. A research assistant would explain the research contents to patients and get the informed consent form signed; Subsequently, the patients would be assigned to undergo carotid ultrasound examination for free. After the screening, Eligible patients would be included according to ultrasound results and demographic information (including age, gender, BMI, blood pressure, whist line, smoking habit, comorbidity and family history). According to SAGER guideline [10], the results and data analysis on gender are not involved in this trail, gender distribution will be reported in order to avoid selective bias only.
Eligibility criteria: Carotid ultrasound examinations (at least one side IMT≥1.0 mm, and both side IMT≤1.5 mm) were required for eligibility before inclusion. To be considered eligible, patients had to comply with the following: Age: 20 -70 years old; Volunteer to take part in the study and signed informed consent form; Good understanding of the trail; Without any other treatment for atherosclerosis in the past 2 weeks (i.e., asprin, plavix, fibrates, statins etc.).
Exclusion criteria
Risk of 10 years’ cardio-cerebrovascular disease≥20% [11]; History of acute cerebrovascular events, or severe trauma or major surgery, other serious diseases; Mental disorders; Skin diseases and bleeding disorders; Fear of acupuncture; Pregnant women or lactating women; Severe diabetes, skin irritation, or acute skin lesion.
Randomization
Patients who meet the inclusion criteria will be randomly assigned to TA group and SA group in a 1:1 ratio.
Sequence generation: Randomization was conducted by SPSS software (20.0 ver. IBM SPSS Statistics, IBM Crop, Somers, New York, USA). The allocation sequence was computer-generated with RV. UNIFORM (0,100), set seed=20190901.
Implementation: The random numbers and random cards were produced and stored by Guangdong Provincial Hospital of Traditional Chinese Medicine, Acupuncture Research Centre. Each randomly generated serial number was put into an opaque envelope. These envelopes were allocated to corresponding patients by researchers according to sequential order after inclusion.
Blinding
Considering the specificity of the acupuncture treatment, this trial cannot blind the acupuncturists. Each stage including recruitment, intervention, sample collection, ultrasound examination and follow-up, will be carried out independently by different staffs. Patients, data collectors, statistical analysts and ultrasound physicians were all blinded in this trial.
Ethics
Patients were recruited in Guangdong Provincial Hospital of Traditional Chinese Medicine with the approval of the local Clinical Ethics Committee (YF2019-071-01). Researchers have made sure every patient has a well-grounded understanding of the research contents and get informed consent before inclusion. All collected data from patients including photographs, hard-copy reports and spreadsheets are locked with limited access. The trial was registered at http://www.chictr.org.cn (NO.ChiCTR1900025551,31/08/2019).
Interventions
Patients were randomly divided into two groups: TA group and SA group. Each group’s acupuncture points were the same. The selection and location of acupoints were consistent with the WHO acupuncture point positioning criteria (As shown in the table and figure of Supplement 3 eAppendix 2) [12]. The interventions were implemented by 4 experienced and trained acupuncturists under the Clinical Research Acupuncture Operation Standard (STRICTA) [13]. Patients in both groups received a total of 24 treatments twice per week for 12 weeks.
TA group
Before treatment, patients were asked to have a rest in a warm and comfortable environment. The acupuncturist sterilized his hands and patient’s skin with 75% alcohol. Next, the acupuncturist located acupuncture points and pasted customized foam pads (1*1*1 cm3 in size) with a hollow tube on the skin. Hwato brand disposable sterile needles (size 0.25*25 mm) were used to pass the tube and then penetrate into skin through the foam pad in the order of Neiguan (PC6), Yanglingquan (GB34), Renying (ST9), Baihui (DU20) and Yintang (EX-HN3). The depth was between 0.5-1.0 cm according to patients’ body shape. The needles were retained for 30 minutes. Finally, the needles were removed without bleeding.
SA group
The environment demand, acupoints, procedures, retention time and acupuncturists were the same as TA group. After sterilization, the acupuncturists covered the points with the same foam pads. Disposable sterile blunt needles (0.25*25 mm) that could not pierce the skin were applied in SA group (As shown in the table and figure of Supplement 3 eAppendix 2). Although blunt needles were inserted through the hollow tube and the foam pad and touched the patient’s skin, they could not puncture the skin surface. Therefore, the needles only produced a tingling sensation which was similar to true acupuncture.
Usual care
Patients in both groups received the healthy diet and lifestyle education during treatment periods according to 2020 Guidelines for primary prevention of cardiovascular disease in China [14]. Patients were allowed to receive treatments that were not related to atherosclerosis or other cardiovascular diseases and recorded in CRFs truthfully.
Outcomes
Primary outcomes
The primary outcome was ultrafast pulse wave velocity (ufPWV), which is a shock wave caused by blood flow along the aorta to peripheral arteries during the contraction and dilatation phases of cardiac ejection. The primary outcome was measured by French Supersonic Aixplorer ultrasonic diagnostic instrument with the SL10-2 probe, frequency 2-10 Hz, vascular PWV mode. The sonographer selected 1-1.5 cm below the carotid bifurcation of the common carotid artery as measurement site. The indicators include the pulse wave velocity at begin-systole (BS, m/s) and at end-systole (ES, m/s), as well as the standard deviation (Δ±). If the standard deviation is more than 20% of the corresponding speed, the operation is deemed invalid, according to UFPWV operating guideline from Supersonic Aixplorer Company (details in supplement 4). Both sides carotid arteries were measured at least three times and then the researcher recorded the average PWV value by Excel.
Secondary outcomes
Carotid intima media thickness (cIMT): cIMT is a repeatable and convenient indicator of atherosclerosis as well as a vital indicator for predicting brain and cardiovascular events. According to Chinese Guideline for Ultrasound Examination of Blood Vessels, we defined IMT≥1.5 mm as local plaque formation. Intima - media thickened (at least one side IMT≥1.0 mm, and both side IMT≤1.5 mm) were required for eligibility before inclusion. Patients underwent IMT examinations before the first treatment and then at 4 weeks, 8 weeks and 12weeks.
Lipid level: lipid level indicators included cholestenone (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL), low-density lipoprotein cholesterol (LDL) and apolipoprotein (Apoa). The tests will be carried out by a third party, namely, the Department of Examination of Guangdong Provincial Hospital of Chinese Medicine.
Blood platelet (PLT) and fibrinogen (FIB): Blood platelet (PLT) and fibrinogen plays a vital role in endothelial impairment and plaque formation. Researches suggested that high PLT level patients were more likely to suffer from atherosclerosis or stroke. Fibrinogen is a Blood coagulation functional protein synthesized mainly by liver cells and is the most abundant coagulation factor in plasma. Fibrinogen is an important substrate and key step in thrombosis, and can stabilize the thrombosis structure. High fibrinogen is an important risk factor for various thrombotic diseases including atherosclerosis
Demographic and clinical baseline information at baseline
From patient record: age, gender, family history of cardiovascular disease, co-morbidity, smoking habit and all medication.
Measurements: BMI, blood pressure, waistline.
Sample size
As no clinical studies related to this study were found during the review of previous studies so far, at least 40 participants are required with a 1:1 ratio for two groups. Considering the maximum 15% dropout rate, 46 participants will be included.
Statistical analysis
ITT principle was applied to all participants who have received at least one treatment. As uncompleted patients, missing data was imputed using the last-observation-carried-forward method. Data statistics were runed with SPSS 20.0 software (IBM SPSS Statistics, IBM Crop, Somers, New York, USA) with a set of 95% confidence interval and 2-sided significance level of less than 0.05.
For baseline data, the consistency between two groups will be compared by T test or Mann-Whitney U test. Continuous variables (age, BMI, blood pressure, waistline) were presented by mean and standard deviation if meet normal distribution, otherwise would be presented by mean (median, range) and categorical variables (gender, family history, co-morbidity, smoking habit and medication) were presented by composition ratio. For primary and secondary outcome, t test was used if continuous variables conformed to the normal distribution and uniform variance. Otherwise, we used corrected t-test or two-sample nonparametric tests. Categorical variables were analysed by chi-square test or Fisher’s exact test. When P is less than 0.05, it will be defined as having a significant difference. The same approach was used for changes in PWV and cIMT at weeks 4, 8, and 12.