We have registered this overview under the PROSPERO registry (CRD42020188107). The methodology followed the guidelines stated in the Cochrane Handbook, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist (15), and some high-quality methodological overviews (16, 17).

PubMed, Cochrane Library, Embase, Web of Science, China National Knowledge Infrastructure, Wanfang database, Chinese Biomedical Literature database, and the Chinese Science Journals Database were searched from their inception to December 2020. The following keywords were used in the search: angina pectoris, acupuncture, systematic review, and meta-analysis. The search strategy for PubMed is shown in Table 1.

Two independent reviewers assessed the literature for inclusion and abstracted the data based on details. Any disagreements were resolved by consensus through discussion. The following details were extracted: author(s), publication year, simple size, interventions, outcomes relevant to this overview, quality assessment methods, and the summary estimate of the intervention effects.

Two independent reviewers assessed the methodological quality of the included studies using Assessing the Methodological Quality of Systematic Reviews 2 (AMSTAR-2) (18). According to the information provided in the included studies, each of the 16 standards of AMSTAR-2 is given a rating of “Yes,” “No,” or “Partial Yes.” Any disagreements were resolved by consensus through discussion.

Two independent reviewers assessed the reporting quality of the included studies using the PRISMA checklist (15). According to the information provided in the included studies, each of the items in the PRISMA checklist is given a rating of “Yes,” “No,” or “Partial Yes.” Any disagreements were resolved by consensus through discussion.

Two reviewers independently assessed the evidence quality using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) (19). Evaluation of the evidence quality mainly considers the following aspects: risk of bias, inconsistency, indirectness, imprecision, and publication bias. Disagreements were resolved by discussion or consensus with a third-party reviewer.

The initial database search yielded 128 potentially relevant published studies; 10 publications were retrieved after reviewing the titles and abstracts. Following the review of full texts, seven published SRs/MAs (8–14) met the inclusion criteria. Figure 1 shows details of the flow diagram.

All included studies were published between 2012 and 2019. Six of the studies were published in English; the rest are in Chinese (8). The number of trials ranged from 4 to 25, and the sample size ranged from 254 to 2,058. The intervention methods were mostly acupuncture therapy plus CM in the experimental group and CM and sham acupuncture in the control group. For the assessment of methodological quality, six studies used the Cochrane criteria and the rest used the Jadad scale (9). Details are presented in Table 2.

Among the included studies, two (11, 12) were rated as of moderate quality; the rest were of very low quality. Items 2 and 7 were rated as of particularly low quality. Only two studies (11, 12) established a prior study protocol. No study provided a complete list of excluded studies with reasons or explained the reasons for the selection of study types. Details are shown in Table 3.

With the PRISMA checklist, only one study (11) reported all items in the checklist, while others had reporting deficiencies. The checklist items incomplete reported were mainly Q5 (protocol and registration), Q8 (search), Q16 (additional analyses of methods), and Q23 (additional analyses of results). Details are shown in Table 4.

A total of 18 outcome indicators extracted from the included studies were evaluated; the evidence quality was very low in three, low in three, moderate in eight, and high in four. Risk of bias was the most common downgrading factor, followed by imprecision, inconsistency, publication bias, and indirectness. Details are shown in Table 5.

Through a comprehensive search, seven SRs/MAs on the effectiveness of acupuncture for SAP were identified. According to the results of the AMSTAR-2 assessment, only two of which were regarded as of moderate methodological quality; the rest were of very low methodological quality. Items 2 (protocol registration) and 7 (list of excluded studies) were rated as of particularly low quality. In performing SR/MA, a pre-registration protocol is quite essential, which helps to promote processing transparency and avoid methodological bias (20). In this context, registration platforms, such as PROSPERO, have been highlighted and advocated. Therefore, more attention should be paid to registering the work. Future investigators should register protocols in advance, before the SRs/MAs are carried out. For study selection, all the included studies were selected in duplicate, resolving disagreements by discussion or consensus with a third author. However, none of the studies provided a list of excluded trials, which made it difficult to guarantee the reproducibility of the results (21). Similar to the results of the AMSTAR-2 evaluation, assessment using the PRISMA checklist found that only one study (11) reported the checklist in its entirety, while others had reporting deficiencies. Incomplete reports of checklist items were mainly missed information on protocol and registration, literature search, and additional analyses.

The evidence quality was assessed by the GRADE system. Among the 17 outcome indicators, the evidence quality was very low in three, low in three, moderate in eight, and high in four. Risk of bias was the most common downgrading factor, followed by imprecision, inconsistency, publication bias, and indirectness. Through further analysis, it was found that the limitation in this area was mainly due to the inability of acupuncture therapy to blind the implementer. Thus, there are issues that should be paid attention to. With the development of evidence-based acupuncture, it is hoped that researchers will continue to promote standardization of the techniques and operating procedures of acupuncture in the future in order to provide more high-quality evidence.

Of the included studies, it is worth noting that, although almost all indicated that acupuncture is an effective approach in treating SAP, most authors did not want to draw firm conclusions because of the small sample sizes or the low methodological quality of the included studies. Meanwhile, given that the results assessed in this overview suggest that neither the methodological quality of the included studies nor the evidence quality of the reported outcome indicators is satisfactory, we cannot draw firm conclusions regarding acupuncture for SAP based on the currently published evidence. Therefore, in the future, RCTs with adequate blinding and large simple sizes are still needed to determine whether acupuncture or combination therapies are more effective than other treatments. Furthermore, strict adherence to the Cochrane Handbook for the design and implementation of SRs/MAs needs to be encouraged and advocated in order to be able to provide more convincing evidence.

To the best of our knowledge, this is the first study using the method of overview to systematically appraise and synthesize evidence of acupuncture for SAP. The results of this overview will be beneficial to evidence users and help researchers improve the quality of their studies. However, limitations should be acknowledged, especially the differences involving populations, interventions, comparators, outcomes, and study designs (PICOS). Firstly, all included SRs/MAs mentioned that the included patients were diagnosed with SAP according to national or international standards. However, only one study (12) mentioned the use of the WHO criteria, while other studies did not account for the specific diagnostic criteria used. Moreover, patients were heterogeneous in terms of age, disease duration, and complications. The above heterogeneity from the patient aspect may affect the credibility of the conclusions. Secondly, some studies have looked at the use of acupuncture to combine other interventions on a control group basis, which is easily mixed with the effects of acupuncture. Thus, experimental measures of acupuncture alone or plus a control group may increase credibility. Thirdly, the diversity of the acupuncture treatment protocols has resulted in heterogeneity, including acupoint selection, stimulation methods, and the frequency of treatment. Although the use of some acupoints was concentrated, the overall differences in the acupuncture treatment regimens in the included studies were large. Moreover, the CM used in the control group was different. This may have caused some deviation. Fourthly, the measurement scales used in the RCTs were subjective clinical assessment tools based on the observations of assessors and may not be accurate for the assessment of treatment effects. The use of effective rate or the number of patients with no improvements in ECG also limited the generalization of the results. A fifth limitation is that, although all SRs/MAs were based on RCTs, inadequate reporting of randomization, allocation concealment, blinding, intention-to-treat analysis, and dropout interpretation in most RCTs may have contributed to the risk of bias. Finally, all the included studies in this overview were performed in China, so whether the conclusions of this study can be generalized to other populations remains to be studied further.