Electroacupuncture for Hypertension in Adults: Study Protocol of a Systematic Review and Meta-Analysis


 Introduction: Hypertension is a common, prevalent cardiovascular disease that requires multidisciplinary treatments. In Asian countries, electroacupuncture (EA) is often used as an adjunct to antihypertensive drugs for lowering blood pressure and relieving related symptoms. However, till now the effects of EA for hypertension are not fully evaluated. This study aims to assess the efficacy and safety of EA in patients with hypertension through a systematic review and meta-analysis of randomized controlled trials.Methods and analysis: Seven electronic databases including Medline, EMBASE, CENTRAL, CNKI, CBM, Wanfang, and VIP Database will be searched from inception to May 30, 2021, to identify clinical trials of EA treating hypertension in adults. Eligible comparisons will be limited to 1) EA versus valid comparators including placebo, antihypertensives, waiting list and 2) EA as an adjunct to antihypertensives versus antihypertensives. Clinical outcomes include blood pressure, response rate, somatic symptom relief, quality of life, and safety from baseline to post-treatment. The Cochrane Risk of Bias assessment tool will be used to evaluate the quality of the included studies and the GRADE system will be employed to summarize the overall quality of evidence. Random-effects model will be performed, where efficacy data will be reported with mean difference (continuous data) or risk ratio (dichotomous data). In addition, heterogeneity and publication bias will be tested with the I2 statistic and the Egger’s test/funnel plot. Ethics and dissemination: As a secondary literature study, this systematic review doesn’t require ethical approval. The study results will be disseminated via peer-review publication or conference presentation. Registration: PROSPERO registration ID CRD42019133937.


Introduction
Hypertension, featured by persistently elevated arterial blood pressure, is the leading cause of major cardiovascular diseases and premature death worldwide 1 . Chronic increase in blood pressure might be asymptomatic yet can cause hypertension-mediated target organ damage which can be life-threatening.
Epidemiology study showed that 31.1% of adults (1.39 billion) worldwide had hypertension in 2010 1 .
Under the 2017 American College of Cardiology(ACC)/ American Heart Association (AHA) guidelines 2 , the prevalence of hypertension increased from 32-45.4% in the U.S. and from 23.2-46.4% in China 3 4 , highlighting the urgent need for early diagnosis and intervention of hypertension. Currently, antihypertensive drugs remain the mainstream for hypertension treatment, yet limited somehow in patients' compliance, medical cost, and drug-related side effects 5 . Non-pharmacological therapies recommended by the ACC/AHA are gaining increased popularity in clinical practice, especially for earlystaged hypertension intervention, relief of related somatic symptoms, and prevention of target organ damages 2 .
Acupuncture is an important non-pharmacological intervention widely used in China for angina pectoris, hypertension, cardiac arrhythmia and other cardiovascular diseases. In clinical practice, electroacupuncture (EA), integrating electric stimulation with classic acupuncture, is well recognized and applied in hypertension management 6 . EA, by using uni ed apparatus and prameters, brings a more standerized and less biased treatment, when compared to manual acupuncture that mostly in uenced by the experience of acupuncturists. Besides, Clinical studies showed that EA helps not only with blood pressure reduction, but also the improvement in its circadian rhythm and daily variations in the hypertensives 7 8 . Researches demonstrated that hypertension is highly associated with immune in ammation 9 10 , and newly released report in Nature has just veri ed low-intensity EA stimulation at ST36 can induce anti-in ammatory effects through vagal-adrenal axis, which provides the potential neuroanatomical bais of EA for hypertension 11 . Other basic science studies supported that EA may trigger the neuroendocrine system and modulate metabolic function to activate neurotransmitter release and exert sympathoinhibitory actions 12 13 . Those mentioned above have provided potential neuroanatomical baises of EA in treating hypertension.
High-quality systematic reviews serve clinicians, policy makers with best reliable evidence, which enables evidence-based decision making and clinical practice for patient care 14 . Yet, to our knowledge, a critical appraisal of clinical studies with a focus on EA for hypertension is lacking till now. Previous systematic reviews on acupuncture for hypertension failed to differentiate the antihypertensive effect of EA from that of manual acupuncture and provided no speci c evidence for EA treating hypertension 15 . Given many studies are statistically underpowered, a pooled analysis is needed 16 . Therefore, the present study is deigned to summarize the results of current trials available and assess the e cacy and safety of EA for adults with hypertension. The study is designed in line with the PRISMA and the Cochrane guidelines 17 18 .

Objectives
The objective of this review is to systematically evaluate the e cacy and safety of EA for adult patients with hypertension compared with sham acupuncture, antihypertensives, and waiting list .

Protocol register
This protocol has been composed under the guidance of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols guidelines, and we used the PRISMA-P checklist when writing our report. This protocol has been previously registered on PROSPERO platform (https://www.crd.york.ac.uk/PROSPERO/ ) with an assigned registration number CRD42019133937.

Eligibility criteria
The PICOS (participant, intervention, comparison, outcome and study design) principle will be employed in this study protocol.

Types of study design
Only paralleled randomized controlled trials(RCTs) will be included. Studies with other designs, unobtainable data, duplicate publications, animal experiments and reviews or case reports will be excluded.

Types of Participants
Adult patients (≥18 years old) clinically diagnosed with essential hypertension, with a systolic blood pressure (SBP) no less than 130 mmHg or a diastolic blood pressure (DBP) no less than 80 mmHg, or both or used antihypertensive drugs will be included.

Types of Interventions
(1) Experimental interventions EA with needle insertion will be considered eligible intervention in this study. EA is de ned as a form of acupuncture which delivers electric stimulation to treat disease through needles inserted at acupoints. EA can be used alone or as an add-on to anti-hypertensive agents. We excluded other methods of stimulating acupoints without needle insertion (such as laser stimulation or transcutaneous electrical stimulation).
(2) Comparator interventions Comparators are as previously reported 19 , including 1) no treatment or waiting list, 2) sham acupuncture, de ned as intervention mimicking 'true' acupuncture, but deviating from at least one perspective regarded essential based on TCM acupuncture theory, such as skin penetration or non-acupoint location, and 3) positive controls like anti-hypertensive agents recommended by o cially released guidelines. Studies comparing different styles of acupuncture therapies (such as EA VS manual acupuncture) or EA of different sessions will be excluded.

Types of Outcomes
(1) Primary outcomes Change in SBP and DBP from baseline from baseline to post-treatment.
(2) Secondary outcomes 1) Response rate: responder is de ned as a patient demonstrating an SBP reduction > 3mmHg from baseline to post-treatment 20 21 .
2) The relief of somatic symptoms such as dizziness, headache, insomnia, etc.   Table 1, which will be appropriately adapted according to syntax-related or lingual requirements of other electronic databases.
To acquire potential unpublished data of ongoing RCTs, clinical trials registries like the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP), clinicaltrials.gov and the Chinese Clinical Trial Register (ChiCTR) will also be searched. References of relevant systematic reviews will be screened, and a manual search of related journals and conference abstracts will be conducted as supplements. Study authors will be contacted via email if needed. All literature citations retrieved will be stored and managed in EndNote X9 software.

Selection of studies and data extraction process
Study selection and data extraction process will be independently conducted and cross-checked by two authors, respectively (Pan Zhang & Yalan Chen). A piloting form will be designed for date extraction, including general information like rst author, publication year, founding source, as well as details of participants, methods, interventions, outcomes, and results. Unreported data will be acquired as possible through contacting authors of included RCTs if necessary. Any discrepancies between the two reviewers will be settled by discussion or the introduction of a third reviewer (Furong Zhang). The ow chart based on PRISMA statement 22 is displayed in Figure 1.

Risk of bias assessment in included studies
The risk of bias (RoB) of each study included will be independently evaluated by two investigators (Yuzhu Qu and JiyaoChen) according to the Cochrane RoB 2 23 assessment tool. According to RoB 2, bias will be evaluated per clinical outcome in ve distinct domains. Within each domain, signaling questions will be answered for further rating "low risk of bias," "some concerns," or "high risk of bias". All these ve rates contribute to an overall risk-of-bias judgment for the result assessed, which helps to stratify metaanalyses. Disagreements between the two authors will be resolved by discussion. If the consensus fails to be reached after discussion, a third reviewer (Furong Zhang) will be consulted for nal decision.

Data mining: characterization of EA for hypertension
Data mining enables to uncover associations between diseases and acupoints 24 . To gure out potential relationship between points selected and essential hypertension, we plan to implement a data mining by collecting and analyzing acupoints selected for treatment from included RCT.

Data synthesis
Continuous and categorical data will be expressed as standard mean differences (SMDs) and risk ratios (RRs) respectively, with 95% con dence intervals (95% CIs). Pair-wise meta-analysis will be performed for direct comparisons with 95%CIs. Chi 2 and I 2 test will be applied to assess for heterogeneity among included studies. I 2 values serves as signs for the degree of statistical heterogeneity. I 2 lower than 50% is considered as homogeneous and a xed model will be used; while I 2 over 50% implies a signi cant heterogeneity. Otherwise, a narrative review will be conducted. Meta-regression will be conducted for screening the in uencing factors of heterogeneity, then sensitivity analysis including leave-one-out and subgroup analysis will be implemented to gure out potential clinical heterogeneity. Besides, subgroup analysis based on hypertension subtypes will be conducted to differentiate bene t degrees of EA when appropriate. Publication bias will be checked with a funnel plot if with over 10 studies included, and the Egger's test will be carried out if it needs to be quanti ed. Statistical analysis will be conducted with Review Manager 5.3 software.

GRADE framework-based quality of the evidence
The GRADE framework will be employed for rating the overall quality of the evidence. In GRADE, highquality evidence is considered as RCTs without important limitations and some observational studies.
Five limitations including study limitations, inconsistent results, indirectness of evidence, imprecision, and publication bias will downgrade evidence quality While three factors can upgrade the quality of evidence like large magnitude of effect; plausible confounding, which would reduce a demonstrated effect; and dose-response gradient. The quality of evidence will be evaluated as high, low or very low quality based on integrative assessment of results.

Patient and public involvement
The protocol is based on a review of relevant studies and does not include original patient data. No patient involved in this protocol.

Discussion
Given that systematic reviews topped of the evidence pyramid, high-quality SRs can help provide best evidence in clinical practice. Till now, no consensus reached on e cacy and safety of EA for hypertension, partially owing to the heterogeneity in the types of interventions which mixed EA with other forms of acupuncture. As is known that EA can bring a more standerized and less biased treatment, by using uni ed apparatus and prameters, yet manual acupuncture might vary from individual acupuncturist's experience. Therefore, in this proposed study, we will focus on the e cacy and safety of EA for hypertension only.
Recent clinical studies demonstrated inconsistent conclusions. Despite the discrepancies in previous studies, a recent randomized controlled trial showed that compared with sham acupuncture, 6-week EA treatment led to a statistically signi cant systolic blood pressure reduction at week 9 but not week 6 25 26 27 . The inconsistency requires a further pooled analysis of current evidence to address the bene ts of EA for patients with hypertension. To our knowledge, a critical appraisal of clinical studies focusing on EA for hypertension is still lacking till now. More importantly, our study highlights the effects of EA for the relief of hypertension-associated somatic symptoms, which distinguishes our study from previous ones.
By conducting this study, we will also identify possible methodological limitations of current clinical studies. Such ndings may contribute to improving the design of sham-controlled RCTs in a more rigorous way, thus minimizing bias, to provide evidence of better con dence on acupuncture treatment for hypertension. The ndings will provide evidence on the effects of EA for hypertension to inform evidencebased practice of nonpharmacologic hypertension management; and also identify possible methodological limitations of current studies to facilitate the design of more rigorous trials on acupuncture treatment for hypertension. Flowchart of study selection.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. PRISMAPchecklist.docx