Acupuncture therapy for stable angina pectoris: a systematic review and meta-analysis

Background Stable angina pectoris (SAP) is a common cardiovascular disease, which brings health burden to society. Acupuncture therapy is effective in improving SAP as adjunctive therapy, nevertheless, there were controversies on the effect of acupuncture on disease-affected meridian (DAM), sham acupoints and nonaffected meridian (NAM). This study evaluated the effect of acupuncture on DAM as adjunctive therapy, and the difference in acupuncture on DAM, sham acupoints, and nonaffected meridian NAM. Methods Thirteen online databases were searched from inception to December 2, 2019. Risk of bias and quality of evidence for outcomes were respectively assessed by the Cochrane risk of bias assessment tool and the GRADE approach. RevMan 5.3 was adopted to conduct meta-analysis. Results Thirteen RCTs including 1026 participants were included in this study. Acupuncture therapy could be effective in the improvement of angina severity, ECG results, 6-MWT, SAQ results (physical limitation, angina stability, angina frequency, treatment satisfaction, disease perception), anxiety, HRV (LF, LF/HF), ET, CRP, and reduction of nitroglycerin intake. Acupuncture on DAM could be more beneficial than acupuncture on sham acupoints in aspects of angina severity, 6-MWT, SAQ results (angina stability, angina frequency, treatment satisfaction, disease perception). Compared to acupuncture in NAM, acupuncture on DAM could be more effective in SAQ results (angina stability, angina frequency, treatment satisfaction). Moreover, acupuncture on DAM did not increase the risk of dropout and adverse effect. Conclusions Acupuncture on DAM could effectively improving SAP, and it could be more effective compared with acupuncture on sham acupoints or NAM. However, these conclusions were limited by low-quality of most of included studies. study to assess (a) the effect and safety of various types of acupuncture therapies on DAM as adjunctive therapy in treating SAP, (b) the difference of therapeutic results in acupuncture therapies on DAM, sham acupoints and NAM. Type of outcomes measures: the primary outcome was frequency of angina attacks, and the secondary outcomes included severity of angina, the number of participants with changes in electrocardiogram (ECG), 6-minute walk test (6-MWT), nitroglycerin intake, Seattle Angina Questionnaire (SAQ), intensity of anxiety, intensity of depression, heart rate variability (HRV), endothelin (ET), C-reactive protein (CRP), brain natriuretic peptide (BNP), retention of treatment, adverse events. MD, CI p=0.85). pooled from NAM group in angina stability (n=328; MD, 11.05; 95% CI 7.06 15.04; p<0.00001; heterogeneity: X 2 =1.70, p=0.64, I 2 angina frequency (n=328; MD, to p=0.0001; 2 p=0.82,

6 When outcomes were measured by different tools, data were summarized as SMDs; when outcomes were measured by same tool, data were summarized as MDs. Heterogeneity was tested by I 2 statistic. If I 2 statistic > 50%, data analyses were conducted using random-effects model; if I 2 statistic ≤ 50%, fixed-effect model was adopted to perform data analyses [27]. Sensitivity analyses, which were conducted to explore the potential sources of heterogeneity, were performed by excluding studies with high risk of bias from meta-analysis. On account of the difference in therapeutic effect caused by different types of acupuncture therapies, subgroup analysis was carried out when the number of studies was sufficient. Reporting bias was assessed by funnel plots, if the number of included studies was more than 10 [27].

Selected Studies
A total of 14660 articles were initially identified by online database searches, from which 6455 were duplicates, 7366 were excluded through reading titles and abstracts, 839 were reserved for further assessment. After screening by full-texts, 15 articles were included in this study, of which 4 articles [30][31][32][33] belonged to 2 studies. Flow diagram for study selection was showed in Figure 1.
For other bias, eleven studies were judged to be at 'low risk of bias', and two studies [20,32,33] were assessed as 'high risk of bias'. Table 2 presented risk of bias summary.

Table 2. Risk of bias of included studies.
: low risk of bias; : high risk of bias; : unclear risk of bias.

Effects of intervention
Quality of evidence for outcomes was presented in Table 3. Figures for following meta-analyses were showed in additional file 2.

Intensity of depression
The pooled analysis showed that there was no difference between DAM group and wait list group  [40] reported that there was no difference between DAM group and wait list group in adverse effect. In the study of Lan [19], a total of four (14%) participants reported adverse effects, and all participants recovered without treatment. In the studies of Li et al. [30,31] and Jing [18], no participants reported adverse effect.

Heterogeneity
Considering the high heterogeneity for some outcomes, sensitivity analysis and subgroup analysis were conducted. Heterogeneity was reduced by the removal of study assessed as 'high risk of bias' or subgroup analysis based on different type of acupuncture therapies in most of meta-analysis in this study, but not all meta-analysis for outcomes. In fact, there were clinical heterogeneity caused by stimulation modes, duration, dosages, choice of acupoints, etc. In addition, basic therapies had differences. In ten studies, participants all received antianginal western medications, and the other studies adopted Chinese medicine as basic therapy, namely Danshen dripping pills. On account of the number of included studies, we cannot conduct subgroup analysis on all influence factors and sensitivity analysis in all outcomes.

Reporting Bias
Considering the number of included studies was insufficient, funnel plot was not generated by RevMan 5.3.

Discussion
This systematic review and meta-analysis included 13 studies involving 1026 participants with SAP. In summary, acupuncture on DAM had certain differences from no acupuncture treatment, acupuncture on sham acupoints, and acupuncture on NAM.
For primary outcome, there was no evidence that acupuncture therapies on DAM was more effective than no acupuncture treatment, acupuncture on sham acupoints, and acupuncture on NAM. However, in subgroup analysis, acupoint application on DAM could effectively reduce frequency of angina attacks as adjunctive therapy. The quality of these outcomes was low and very low, thus, we could not come to a firm conclusion. In this study, 15 percent of selected studies did not report random sequence generation, 77 percent of selected studies did not describe the approach of allocation concealment, 54 percent of selected studies did not mention blinding of participants, 46 percent of selected studies did not mention blinding of outcome assessment, 15 percent of selected studies did not report completeness of outcome data, 62 percent of selected studies did not publish study protocols. In short, seven (54%) studies were rated as 'high risk of bias', five (38%) studies were judged as 'unclear risk of bias', and only one (8%) study was assessed as 'low risk of bias'. Because of quality of outcomes and small number of studies, the conclusions in our study was limited.
Acupuncture has physiologic analgesic effects [41], and our study indicated that acupuncture on DAM could effectively improve SAP. Nevertheless, the mechanisms have not yet comprehensively explained. SAP is relevant to the mismatch between demand and supply of myocardial oxygen [42], which is caused by stable coronary artery plaque when myocardial oxygen consumption increase [43]. Previous studies [44][45][46][47] stated that acupuncture on DAM can decrease oxygen demand, infarct size, ventricular arrhythmias, left ventricle dysfunction, and so on. Acupuncture on DAM also has the improvements of myocardial ischemia syndromes in regulating autophagy [48], the expression level of adenosine receptor [49], myocardial enzyme [48], myocardial remodeling [50], microvessel formation[51].
Although we searched international general healthcare databases, regional databases, subject-

Declarations Ethics approval and consent to participate
Not applicable.

Consent for publication
Not applicable.