Design and registration of the review

This study has been registered on PROSPERO, and the protocol will be based on the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols statement (PRISMA-P).20 The PRISMA-P checklist is shown in online supplemental appendix S1.

Eligibility criteriaTypes of studies

All clinical RCTs of acupuncture treating APPCI will be included without restrictions on the country, and there will be no language limitation.

Reviews, case reports, theory discussions, animal experiments, other diseases, conference summaries and other non-randomised clinical studies will be excluded.

Types of participants

This study will enrol patients undergoing PCI in IHD, regardless of gender, age, occupation, education, severity, etc.

Types of interventions

All types of acupuncture therapies should be treated in the experimental group. Regardless of stimulation approaches, acupoint selection and additional therapies will be included.

The control group will include sham acupuncture, acupuncture on corresponding non-acupoints, drugs and a waiting list. We will also exclude studies comparing acupuncture and other complementary and alternative therapies. If there are uncertain interventions, this study will be excluded.

Types of outcome measuresPrimary outcomes

The main primary outcome is the severity of angina-related clinical evaluation symptoms assessed by validated and reliable scales such as the Seattle angina questionnaire (SAQ) score, the Canadian Cardiovascular Society (CCS) angina score, the Visual Analogue Scale and the dose of nitroglycerin. One of the standardised scales (eg, CCS, SAQ) is more helpful in ensuring consistency of results.

Secondary outcomes

Initial time and incident rate of major adverse cardiovascular events.

Serum inflammatory factors, such as high-sensitivity C-reactive protein, interleukin-6.

Nature and adverse effect rate (relevant symptoms caused by acupuncture).

Data sources and search strategy

Potentially eligible studies published since inception to 8 June 2024 will be retrieved from the following eight electronic databases: PubMed, Embase, Cochrane Library, Web of Science Core Collection, Chinese Biomedical Database, Chinese National Knowledge Infrastructure, Wanfang Database, VIP Database.21

Considering the particularity of the databases, we will develop special search strategies for each. The search strategy for PubMed is shown in table 1, and more is shown in online supplemental appendix S2.

Table 1

Search strategy for PubMed

RCTs on acupuncture treatment in patients undergoing PCI will be searched for independently by two reviewers in those sources.

Data collection and analysisStudy selection and data extraction

Selection of studies: EndNote V.21 will be used to manage the search results from the databases mentioned above.

Two researchers (XZ and YY) will be trained to extract data independently.

In the first step, two reviewers (XZ and YY) will have professional training about the review’s background, purpose and process.

In the second step, the two reviewers will independently select and record by their titles, abstracts and keywords according to inclusion criteria, and ineligible studies will be removed from the trash in EndNote V.21.

In the third step, the two reviewers will independently read the full-text version of the remaining studies to select those that meet the inclusion criteria. The two reviewers will cross-check the election results. Any divergences in the data obtained will be arbitrated by the third reviewer (YR). Each unique study ID will be allocated to each eligible study, comprising the first author’s last name and the year of publication (eg, Zhou 2023).

In the fourth step, the two reviewers will independently extract the data into a self-designed data extraction form. The data extraction form will include basic information (article title, authors, publication date, country), study design (study type, sample size, characteristics of participants, intervention details, duration, outcome measures) and conclusions. Details of acupuncture therapy will be covered according to Revised Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA): Extending the Consolidated Standards of Reporting Trials statement (STRICTA).22

In the fifth step, the two reviewers will cross-check the completed data extraction forms. The third reviewer will resolve any diversities. The entire study selection process for this review will be shown in a PRISMA-P flow chart (figure 1).

Figure 1

Flow chart of the study selection process.

Assessment of risk bias

Two review authors will independently measure the risk of bias in the included studies using the Cochrane risk of bias assessment tool. The assessment of the risk of bias will consist of the following seven items: sequence generation, allocation concealment, participant blinding, the blindness of result evaluators, incomplete outcome data, selective outcome reporting and other biases. The risk level will be categorised as low risk of bias, unclear risk of bias and high risk of bias. If necessary, the third reviewer (YR) will help us judge the consistency. Manager (RevMan) V.5.4 will be used to generate the risk of bias.

When the articles’ data are insufficient or ambiguous, one of the authors will contact the original author to request detailed information about the research by email or telephone or estimate the data. These studies will be excluded if missing data is unavailable or the author cannot be contacted. We will conduct a limited analysis based on available data and discuss the potential impact of missing data.

Measures of treatment effect and assessment of heterogeneity testing

Measures of treatment effect will be analysed by RevMan V.5.4.

Dichotomous data will be expressed as relative risk with 95% CIs, and continuous variables will be described as the mean difference (MD) or standardised MD with 95% CI.

The significance level α will be set at 0.05. I2 will be selected to identify statistical heterogeneity across studies in all analyses based on Cochrane Handbook V.5.4. If there is no significant heterogeneity among the studies (I²<50%, p>0.1), a fixed-effect model will be employed for the analysis. Conversely, if significant heterogeneity is present, subgroup analyses will be conducted to clarify the sources of heterogeneity further. If heterogeneity cannot be resolved, a random-effects model will be used for the analysis.

As potential heterogeneity is unavoidable, we will first consider the sources of clinical heterogeneity and identify covariates. These may include variations in acupuncture interventions (such as electroacupuncture, manual acupuncture and auricular acupuncture). Patients with ACSs and stable CAD are included in our study population. The study population may be a source of heterogeneity. In addition, other covariates that may affect outcome indicators, such as the age of the population, duration of the disease, duration of treatment, and treatment in the control group, are also collected for subgroup analyses.

Sensitivity analysis

We will conduct a sensitivity analysis to determine the robustness and quality of the results. The principal criteria include methodological quality, sample size and data analysis techniques. The meta-analysis will be operated repeatedly. If studies have significant differences in assessment results of risk of bias, we will exclude studies with a high risk of bias from analysis.

Grading the quality of evidence

We will evaluate the quality of evidence according to the Grades of Recommendations Assessment, Development and Evaluation (GRADE) tool.23 Two investigators (XZ and YY), who will have been trained on GRADE, will independently perform a quality assessment of the evidence. Based on GRADE, the quality level will be classified as high, moderate, low or critically low. Any controversy will be handled by the third reviewer (YR).

Assessment of reporting bias

Reporting bias includes publication bias, time lag bias, duplicate publication bias, outcome reporting bias, etc. To avoid reporting bias as much as possible, we will search the most important databases, international general healthcare databases, subject-specific electronic bibliographic databases, citation index databases, dissertations and theses databases, grey literature databases and clinical trials registry platforms. Moreover, more than 10 studies are included in the meta-analysis. In that case, we will evaluate the publication bias by the Egger’'s regression test, and the evaluation will be presented in the form of funnel plots.