An Overview of Systematic Reviews: Acupuncture in the Treatment of Essential Hypertension

Essential hypertension (EH) is a chronic condition characterized by high blood pressure, tinnitus, and dizziness as the primary clinical symptoms, and accounts for more than 95% of people with hypertension. Hypertension is a leading cause of unnatural death, and chronically high blood pressure can harm the brain, heart, kidneys, and other organs, bringing down quality of life.¹ It has been reported that in 2025, 1.5 billion people worldwide will suffer from hypertension, accounting for one-third of the global population.² The prevalence, treatment, and control rates of hypertension in Chinese adults are 46.9%, 40.7%, and 15.3% respectively, according to the Clinical Application Guidelines for the Treatment of Essential Hypertension with Proprietary Chinese Medicines (2021),with treatment and control rates remaining low compared to those in developed countries;³ the incidence among adults over the age of 80 is as excessive as 90%.⁴ The definition of hypertension varies from country to country; in the USA⁵ and China³, for example.

The prevailing hypertension treatment has a considerable duration and treatment regimen. Western drugs are effective but easily resistant, have toxic side effects on the liver and kidneys when taken for a long time, and are not as effective in controlling blood pressure as they should be, and this can be a incremental waste of medical resources.⁶ As a result, researchers are currently investigating how to improve the therapy procedure. Hypertension is classified as “vertigo” and “headache” in Chinese medicine. Long-term activation of the sympathetic nervous system (SNS) is now linked to the development of EH, according to a growing body of research,⁷ and that acupuncture may produce antihypertensive effects through neuroendocrine regulation,⁸ making it a potentially safer and more effective antihypertensive therapy. In recent years, evidence-based medicine and clinical research on acupuncture have become increasingly intertwined,⁹ and systematic review are the best evidence to guide therapeutic practice from an evidence-based viewpoint. The authors found through a preliminary search that several systematic views exploring the efficacy of acupuncture therapy on EH have been published, but only one review of the methodological quality of these systematic evaluations has been published.¹⁰ The quality of the evidence provided should be updated accordingly. Currently, there are many tools to overview systematic reviews, such as PRISMA 2020, GRADE, AMSTER2, ROBIS. This study utilizes systematic reviews to overview the quality of evidence from current systematic reviews in order to further clarify the efficacy of acupuncture in the treatment of EH and provide ideas and evidence to support better research and treatment of EH in clinical practice, for the for the benefit of patients.

Information and Methods

Inclusion Criteria

Study Population

Patients who met the diagnostic criteria of the country’s hypertension guidelines,^3,5 regardless of gender, age and race.

Interventions

The test group was treated with acupuncture (including acupuncture, electroacupuncture, warm acupuncture, etc.), The selected acupuncture points are the 12 main meridian points, and the control group with relevant non-acupuncture treatments (including sham acupuncture, placebo, western medicine, life interventions, etc.).

Outcome Indicators

At least one pre-reported outcome: effective rate, end-of-life SBP, end-of-life DBP, change in SBP value, change in DBP value, adverse effects.

Type of Study Design

A systematic review based on a randomised controlled trial (RCT), with or without quantitative Meta-analysis. The language was English and Chinese.

Exclusion Criteria

Acupuncture as a non-primary intervention in the treatment group, for examples include Ashi point, ear needle, head needle and ankle acupuncture points other than the 12 main meridians, sham acupuncture, acupressure, moxibustion, Tuina and wet cupping or acupuncture as an intervention in the control group, ② Inclusion of non-RCT literature in the systematic review, ② Inclusion of non-RCT literature in the systematic review ③ duplicate publications, ④ non-systematic reviews/meta-analyses, reviews, conference abstracts, ⑤ studies for which data could not be identified, and ⑥ dissertations.

Search Strategy

Two researchers (Mi Zhou and Dongjun Wang) independently and systematically searched Pubmed, CNKI, WANGFANG, VIP and other Chinese and English databases, all from the date of creation to 13 Oct 2022, for the Chinese keywords “acupuncture”, “electro acupuncture”, “needle warming”, “hypertension”, “systematic review”, in English “hypertension” “acupuncture” “needle warming” “electro acupuncture” “systematic review”, the search was carried out using a combination of subject terms and free words. Using Pubmed as an example, the search strategy is shown in Figure 1. after checking using the ENDNOTE software, two researchers independently screened according to exclusion criteria.

Figure 1 Search strategy for Pubmed.

Study Inclusion and Data Extraction

Two evaluators (Mi Zhou and Yuanyuan Guan) alone included studies that met the requirements according to the inclusion criteria, and extracted data from the final included studies, mainly on study title, authors, year of publication, number of included studies, interventions, controls, original study methodological quality assessment tools, outcome indicators, and conclusions. In case of disagreement, a third person (Dongjun Wang) ruled.

Relevant Assessments

Two evaluators (Mi Zhou and Xuan Sun), with the names of the authors withheld, assessed the quality of reporting of the included studies by PRISMA 2020, the methodological quality of the included studies by AMSTAR 2, the quality of evidence of the included studies by GRADE, and the bias of the included studies by ROBIS, respectively, and in case of disagreement, the third person (Dongjun Wang) ruled.

Study Results

Literature Screening Process and Results

The computer database yielded a total of 256 items, including 79 in Chinese and 177 in English. After duplicating 38, reviewing the titles and abstracts 183 were deleted, leaving 35. Finally, after manual checking and careful reading of the full text, 11 articles met the requirements.(Figure 2).

Figure 2 Identification of studies via databases and registers.

Basic Characteristics of Included Studies

A total of 11 included studies,^6,11–20 of which 3 were in English^6,11,12 and 8 in Chinese,^13–20 with the earliest and most recent published in 2011¹³ and 2018.^19,20 Of the 11^6,11–20 papers, seven^{6,11,12,16,18–20} used the risk of bias tool recommended by the Cochrane system for assessing the methodological quality of original studies and four^13–15,17 used the modified Jadad scale, the number of included studies ranged from 5 to 53, with the main treatment modalities in the treatment group being acupuncture, acupuncture plus Western medicine, electroacupuncture, and electroacupuncture plus Western medicine, and in the control group being western medicine, sham acupuncture plus western medicine, and lifestyle, etc. The specific characteristics of the included studies are detailed in Table 1 below.

Table 1 Basic Characteristics of the Included Studies

PRISMA Statement for Evaluation of Report Quality of Included Studies

The quality of reports from the included systematic reviews was evaluated using the PRISMA 2020, with each entry judged to be normative and correctly reported rated as 1, partially reported as 0.5, and not reported or incorrectly scored as 0; the total score was 27, with scores of 22–27 indicating a more complete report, followed by those of 16–21, and less than 15 indicating a more flawed report.²¹ The scores showed that only one¹² study was more complete, 4 studies were more flawed^13,14,16,20 and the rest were of average quality. The main points lost were in 8, 15, 22, 24, 26, 27, which shows that there were some deficiencies in independent screening by multiple researchers, using GRADE for analysis, registering in advance, stating conflicts of interest, and public access to information. The quality assessment of evaluation reports (based on PRIMSA 2020) are detailed in Table 2 below.

Table 2 Quality Assessment of Evaluation Reports (Based on PRIMSA 2020)

AMSTAR2

Evaluated the methodological quality of systematic reviews through 16 entries.²² 9^{6,11,13–17,19,20} studies were considered very low, they all had more than one critical flaw (entries 2/4/7/9/11/13/15) and multiple non-critical flaws, 1¹⁸ were rated low and 1 was rated medium level.¹² As shown in Table 3.

Table 3 Analysis of Methodological Quality (Based on the AMSTAR2)

Grading of Evidence Evaluation

GRADE system grading criteria generated in 2004 are more appropriate for grading the evidence for the development of Chinese medicine.¹⁰ In this study, the quality of evidence was assessed based on the GRADE system methodology²³ for the 11 outcome indicators reported in the included systematic reviews, with a total of 40 items, with 5 items (13.5%) being low grade and all the rest being very low grade (86.5%), with risk bias (100%) and publication bias (100%) being the main reasons for downgrading. The assessment table (Table 4) shows that the quality of evidence for a few of the above outcome indicators was low, with the remainder downgraded to very low due to poor methodological quality, consistency and precision of the included studies.

Table 4 Level of Evidence Ratings (Based on the GRADE)

Risk of Bias Assessment

ROBIS is a new tool published by the University of Bristol, UK, in 2014 to assess the risk of bias for inclusion in studies,²⁴ and there are 3 main stages to the assessment process, which are assessed in this paper for stages 2 and 3. The first assesses the degree of bias in the inclusion exclusion criteria and shows that all (100%) of the inclusion exclusion criteria are low risk; the second assesses the degree of risk of bias in the retrieval and screening of studies and shows that very few (18.1%) are low risk, the third assesses the degree of bias in data extraction and quality assessment and shows that the majority (90.9%) are low risk and the fourth assesses the degree of bias in data synthesis and presentation of results The fourth item assessed the degree of bias in data synthesis and presentation of results and showed that very few (18.1%) were low risk, and the fifth item assessed the risk of bias in systematic evaluation and showed that few (27.3%) were low risk. It is inferred that there was a degree of bias in the included studies. As shown in Table 5.

Table 5 Suggested Tabular Presentation for ROBIS Results

Adverse Reactions

Only brief descriptions of adverse effects were offered in none of the studies that were included.

Discussion

Main Outcome

A systematic review is a comprehensive overall analysis of primary data and is one of the best sources of evidence for evidence-based clinical practice. It is at a high level in the hierarchy of evidence, and the quality of the systematic review determines whether it can truly provide a scientific and theoretical basis for clinicians and policy makers. The overview of this systematic review concludes that acupuncture has been shown to be effective in improving EH without major adverse effects. However the quality of the included studies was generally low with very few exceptions. The main findings of the studies were as follows:

The methodological quality of all the included systematic reviews on the use of acupuncture in EH was very low, as indicated by AMSTAR 2 scale in the current work. A few^6,12 described searches and grey literature searches. In addition, 2^14,20 systematic reviews did not describe in detail the essential characteristics of the included studies, and one¹⁵ with incomplete basic features, which prevented us from comparing studies at baseline. 3 reviews did not provide a list of excluded literature or did not provide information on excluded literature.^13–15 Only 1¹² mentioned a relevant conflict of interest. Only 1¹² assessed the risk of bias for each included study at the time of Meta-analysis and only 5^{6,11,12,18,20} conducted quantitative analyses that adequately investigated publication bias, all of which compromise the strength of the argument for outcome indicators and suggest that the current original study design of acupuncture therapy for EH is flawed and lacks scientific and standardised methodological guidance. At this point medical researchers should pay more attention to the above issues and improve their work and methodological quality in order to provide high quality findings to inform guideline development and clinical decision making.

In this work, The GRADE evidence rating system was used to rate the quality of individual outcome indicators. Based on the results obtained, the quality of evidence for 35 outcome indicators was very low, 5 were of low quality and there was little variation in the quality of individual indicators, the very low quality of evidence for most indicators indicates that there is variation in study outcomes. Typically, imprecision and publication bias in studies were the main reasons for most downgrades, followed by inconsistency and limitations. Of these, imprecision was mainly attributed to wide confidence intervals (CIs) or inadequate sample sizes for the included studies, as well as poor overlap of CIs. In addition, publication bias was mainly reflected in asymmetric funnel plots, insufficiently narrow confidence intervals, and the inclusion of studies that did not meet the criteria for potentially publication biased studies. On the other hand, inconsistency was reflected in the low overlap of CIs between studies, small p and high heterogeneity (I ² > 50%). Limitations were due to the inclusion of original studies at risk of randomisation, blinding and hidden implementation allocation. The degradation in the quality of evidence for outcome indicators has decreased considerably over the past 5 years, suggesting that researchers are increasingly focusing on randomisation, blinding, allocation concealment, and study design aspects of study design. Therefore, to address the issues of imprecision and publication bias, the author recommends that researchers should estimate sample size, standardise the design and conduct of clinical trials, and enhance the search for grey areas to reduce bias.

The first of the ROBIS bias scenarios, the degree of bias in the inclusion and exclusion criteria, a small number of inclusion criteria were less clear and the results showed that the majority of inclusion and exclusion criteria were low risk. The fourth assessed the degree of bias in the synthesis of data and presentation of results, as most studies were heterogeneous and some did not use sensitivity analysis, the results showed less low risk and more uncertainty, the fifth assessed the risk of bias in the systematic reviews, most studies did not explain the risk of bias in the treatment and few studies discussed the statistical significance and the results showed mostly high risk. It is inferred that there was some degree of bias in the included studies.

Limitations and publication bias were the main reasons for downgrading according to the GRADE quality grading, however, we did not find a steady improvement in the methodological quality of the projects over the last years, mainly because the main purpose of reporting guidelines is to help authors remember all the items to be reported, not to perform an overview.²⁵ The results of this review are consistent with those of reviews in other areas (eg depression),²⁶ where methodological quality and strength of evidence were low or very low. Furthermore, Chun²⁷ reached the same conclusion in their study on the methodological quality and strength of evidence of reviews on acupuncture for COPD. Medical research suggests that there may be substantial methodological problems with acupuncture in the treatment of disease. We must therefore acknowledge that a double-blind approach to acupuncture trials is not feasible. However, the quality of the original studies included in the study is also very important and therefore some items may not be well reflected and covered in the GRADE system and the AMSTAR 2 tool, thus affecting the strength of the evidence. In addition, a number of methodological issues and the quality of outcome indicators are closely related to the quality of the included studies, including the listing and selection of inclusion/exclusion criteria, the description of baseline characteristics and conflicts of interest, the design of study pre-protocols, the assessment of study outcomes, and the interpretation of heterogeneity and publication bias.²⁸ As previously mentioned, although double-blinding of acupuncture trials is not usually feasible, investigators need to combine diagnostic and therapeutic features with modern clinical RCTs.

Limitations

The current methods of measuring hypertension values are mainly divided into in-office blood pressure (OBPM), 24-hour ambulatory blood pressure (ABPM), and home self-measured blood pressure (HBPM).⁵ OBPM is currently the main method of diagnosing hypertension, grading blood pressure levels, and observing the efficacy of antihypertensive therapy in China, but there are cases of incorrect posture of patients, unclear clinic environment, resulting in biased measurement values, and ABPM is regarded as the gold standard for blood pressure monitoring and management, with high predictive value, better reflecting patients’ blood pressure levels and fluctuations, and closely related to target organ damage indicators such as left ventricular hypertrophy, facilitating the detection of nocturnal hypertension and covert hypertension; however, it is more expensive, time-consuming and requires special computer software assistance, and reliable thresholds for diagnosing hypertension and criteria for lowering blood pressure need further exploration. HBPM is not affected by patient stress and avoids white-coat hypertension, but has the disadvantages of an irregular detection method, low accuracy, and inability to obtain nocturnal hypertension. Overall, OBPM dominates the diagnosis of hypertension, but with the intensive research and widespread use of ABPM and HBPM, the status of OBPM has declined and diversified blood pressure measurement methods will play a crucial role in blood pressure diagnosis and management.²⁹ However, only a few of the papers included in this paper clarify the specific measurement of blood pressure values, and according to the descriptions, there is no agreement on the measurement methods and the data are subject to error, which may bias the final conclusions. In the most recent study, different blood pressure measurement modalities yielded different conclusions regarding the specific clinical efficacy of acupuncture in improving hypertension: OBPM, daytime ABPM, and SBP measured by a 6-week acupuncture cycle were significantly reduced and persisted for at least the first 2 months after completion, whereas DBP values detected by nighttime ABPM did not appear to change as a result of acupuncture treatment.³⁰

In addition, acupuncture points, practitioner’s technique, duration of acupuncture, and frequency of treatment in acupuncture treatment affect the outcome of treatment, and only half of the included literature specifically described acupuncture manipulation. The included studies could not be quantitatively fused as each outcome indicator was not standardised; and reporting of final scores on evaluation scales such as PRISMA2020, AMSTAR2, GRADE and ROBIS is highly subjective, and multiple scores can be subjective; thus creating bias. Due to language limitations, only literature published in Chinese and English was included in this study, and the authors of the three English-language publications included were Chinese. Korean and Japanese databases, which also have a strong background in TCM research, were not searched, and the search process could not actually be carried out manually, so there was some selective bias.

Outlook

EH is an important risk factor for cardiovascular disease, which can cause impairment or even failure of the heart, brain, kidney and other organs. However, there is a lack of observations and evidence on endpoint events in acupuncture for EH. Therefore, to confirm the efficacy of acupuncture for EH, it is recommended that future clinical trials should focus on endpoint events such as long-term cardiovascular and cerebrovascular events and target organ damage, in addition to routine observations of blood pressure values.

The current study found that there are few health economics studies related to acupuncture for EH, which lack a standardized body of evidence to provide guidance for evidence-based clinical practice of acupuncture. At present, there are more standardized research methods for health economics, and it is possible to draw on related methodologies to establish a system of economic evaluation methods for acupuncture clinical practice, which will be more conducive to the development of acupuncture clinical practice.

The treatment strategy for EH is still based on daily regulation combined with antihypertensive drugs, but due to drug resistance, drug side effects, et., there is also an increasing clinical interest in TCM, especially acupuncture, for the treatment of EH. This systematic overview also demonstrates that acupuncture for EH is a promising treatment modality, although the quality of reporting, methodological quality, quality of evidence and bias of the included studies are low, but it can be recommended that future investigators adopt uniform clinical reporting standards (CONSORT), and reporting standards for interventions (STRICTA2010), in order to improve the feasibility of clinical trials. A detailed description of the specific acupuncture operations (acupuncture points, needle specifications, doctor’s technique, whether qi is obtained, duration of needle retention, frequency of acupuncture, etc.) should also be given; the selection of acupuncture points should follow the results of the patient’s TCM diagnosis and treatment, and the selection of main acupuncture points and points should be standardised; blood pressure measurement methods should also be standardised in order to achieve a final quantitative score.

Conclusions

Acupuncture currently has some benefit in the treatment of EH, although the research is of poor quality. The quality of relevant literature reporting, methodological quality, evidence quality, and bias are all hoped to improve.