3.1 Literature Search and Screening Results.
A total of 273 related literatures were retrieved, and after layer-by-layer screening, 14 [26-39] literatures were finally included. The specific screening process is shown in Figure 1.
3.2 The Basic Characteristics of the Included Literature.
Among the 14 SRs/MAs included, 5[35-39] SRs/MAs were in English, 9[26-34] SRs/MAs were in Chinese, 13 SRs/MAs were studied in China, and 1[34] SR/MA was studied in South Korea, the publication year was 2009-2022, and the number of RCTs studies was 4-53. In terms of quality assessment of included RCTs, 8[28-30,32,34,37-39] SRs/MAs were assessed by the Cochrane risk of bias tool, and 5[26,27,31,33,36] SRs/MAs were assessed by the Jadad scale was used for evaluation, and 1[35] SR/MA used the Oxford scale. The intervention measures in the treatment group were one of the three methods of simple acupuncture, electro-acupuncture, and warm acupuncture or combined with the relevant treatment recommended by the guidelines (including lifestyle regulation, CAD),the control group was the relevant treatment methods recommended by the guidelines or placebo sham acupuncture.See Table 2 for specific information.
3.3 Results on SRs/MAs Quality Assessment.
3.3.1 Results of the Methodological Quality.
The quality of the included SRs/MAs was assessed by AMSTAR-2, and the results showed that the included 12[26-37] SRs/MAs were of very low quality because none of the included SRs/MAs met the key item 2(none of the included SRs/MAs was a registered protocol);2[38, 39] SRs/MAs were of low quality, and none of the 14 SRs/MAs met key items 7(neither SRs/MAs provided an exclusion list) and item 3(did not explain the reasons for selecting the type of systematic review included in the study design).The methodological quality limitation also included the following items: item 1(1 SR/MA author did not fully describe the PICO elements in the SR),item 8(authors of 2 SRs/MAs did not fully describe essential characteristics of included studies),item 10(2 SRs/MAs did not report funding RCTs/SRs/MAs), item 12(authors of 12 SRs/MAs did not investigate the presence of risk of bias on the total effect),Item 13(4 SRs/MAs authors did not discuss the effect of risk of bias on the total effect of included studies), Item 14(8 SRs/MAs authors did not investigate sources of heterogeneity in results, and/or did not discuss their effect on study results), item 15(authors of 7 SRs/MAs did not test for publication bias, and/or discussed its effect on results), item 16(authors of 2 SRs/MAs did not describe funding sources, and/or statements conflict of interest).The results are summarized in Table 3.
3.3.2 Results of the Risk of Bias Assessment.
Regarding the results of the ROBIS assessment, both Phase 1 and Domain 1 of Phase 2 rated SRs/MAs as having a low risk of bias;In the phase 2, 4 SRs/MAs in Domain 2 were rated as having low risk of bias;8 SRs/MAs in Domain 3 were rated as low risk of bias, 1 SR/MA in Domain 4 was rated as low risk of bias;And all SRs/MAs were rated as high risk of bias in Phase 3. The ROBIS scale evaluation results are shown in Table 4.
3.3.3 Report Quality of the Included SRs/MAs.
17 of the 27 items had a "yes or partial yes" response rate of over 80%, indicating that the report was relatively complete. However, there were also some reported flaws in other projects.The reports for item 5(Methods: Protocol and registration), item 8(Methods: Search), item 15(Risk of bias across studies), and item 24(Funding) were incomplete("Yes or Partial Yes" response rate was less than 50%).The results of the PRISMA checklist assessment are shown in Table 5.
3.3.4 Results of the Quality of the Evidence.
Meta-analysis was performed on the outcome indicators in the study, and the GRADE system was used to evaluate the quality of 86 outcome indicators under different intervention indicators one by one.Of these, 2 were of moderate quality, 23 were of low quality, and 61 were of very low quality. Limitation downgrading due to risk of bias was more common in included studies (n=86), followed by imprecision (n=63), publication bias (n=62), inconsistency (n=43), and indirectness (n=0). See Table 6 for details.
3.3.5 Summary of Results Included.
The outcome measures extracted from the included studies are listed in Table 6.
3.3.5.1 The effective rate in lowering blood pressure.
6[26,27,30-32,38] SRs/MAs reported the effective rate of blood pressure reduction.2[26,27] SRs/MAs reported that acupuncture alone was more effective than CAD in the treatment of EH;4[30-32,38] SRs/MAs reported that acupuncture combined with CAD was more effective than CAD alone in the treatment of EH.
3.3.5.2 The efficacy of improving SBP and DBP.
11[27,29-34,36-39] SRs/MAs reported the efficacy of improving SBP and DBP. 9[27,29,31-33,36-39] SRs/MAs reported that acupuncture alone was more effective in treating EH than CAD or placebo sham acupuncture or no treatment;10[29-34,36-39] SRs/MAs reported that acupuncture combined with lifestyle regulation or CAD was more effective than lifestyle regulation or CAD or placebo sham acupuncture in the treatment of EH.
3.3.5.3 The effect of reducing the magnitude of SBP and DBP.
3[28,35,37] SRs/MAs reported the magnitude of blood pressure reduction of SBP and DBP.3[28,35,37] SRs/MAs reported that acupuncture alone was more effective than CAD or placebo sham acupuncture in the treatment of EH, and reported that the efficacy of acupuncture combined with CAD in the treatment of EH was better than that of CAD alone or placebo sham acupuncture combined with CAD.
3.3.5.4 The effective rate of comprehensive treatment.
3[32-34] SRs/MAs reported the effective rate of comprehensive treatment. 2[32,33] SRs/MAs reported that the efficacy of acupuncture alone in the treatment of EH was better than that of CAD, and reported that acupuncture combined with CAD or behavioral adjustment was better than CAD alone in the treatment of EH.1[34] SR/MA reported that acupuncture combined with TCM decoction Tianma Gouteng Decoction(TMGTD) was more effective than CAD or TMGTD in the treatment of EH.
3.3.5.5 The effective rate of improving symptoms.
2[26,32] SRs/MAs reported the effective rate of symptom improvement. 2[26,32] SRs/MAs reported that the efficacy of acupuncture alone in the treatment of EH was better than that of CAD, and the efficacy of acupuncture combined with CAD in the treatment of EH was better than that of CAD alone.
3.3.5.6 Adverse reactions.
8[28, 31, 32, 35-39] SRs/MAs reported adverse reactions. The main adverse reactions included occasional acupuncture site bleeding, dizziness, headache, cough, nausea, pain, etc., which did not require treatment and resolved spontaneously after rest. Since most studies failed to report adverse reactions in a standard way, quantitative analysis was not performed.