Open access Original research BMJ Open: first published as 10.1136/bmjopen-2020-045961 on 6 September 2021. Downloaded from Acupuncture-related adverse events: systematic review and meta- analyses of prospective clinical studies Petra Bäumler ,1 Wenyue Zhang,2 Theresa Stübinger,1 Dominik Irnich1 To cite: Bäumler P, Zhang W, ABSTRACT Strengths and limitations of this study Stübinger T, et al. Acupuncture- Objective Overview on risks of acupuncture- related related adverse events: adverse events (AEs). ► First systematic review on acupuncture- related systematic review and Design Systematic review and meta- analyses of meta- analyses of prospective adverse events (AEs) including a risk of bias prospective studies. clinical studies. BMJ Open assessment. Data sources PubMed, Scopus and Embase from 2021;11:e045961. doi:10.1136/ ► First meta- analyses on AEs related to acupuncture. inception date to 15 September 2019. bmjopen-2020-045961 ► Complying with Preferred Reporting Items for Eligibility criteria for selecting studies Prospective Systematic Reviews and Meta- Analyses guidelines. ► Prepublication history and studies assessing AEs caused by needle acupuncture ► Combining studies with heterogeneous AE defini- additional supplemental material in humans as primary outcome published in English or for this paper are available tions but providing respective sensitivity analyses. German. online. To view these files, ► Causality assessment based on descriptions of AEs Data extraction and synthesis Two independent please visit the journal online. as available from the included articles. (http:// dx. doi. org/ 10. 1136/ researchers selected articles, extracted the data and bmjopen- 2020- 045961). assessed study quality. Overall risks and risks for different AE categories were obtained from random effects meta- Received 16 October 2020 analyses. body for therapeutic or preventive purposes. Accepted 29 July 2021 Main outcomes Overall risk of minor AEs and serious It is used worldwide with growing popularity. adverse events (SAEs) per patients and per treatments. In the European Union, acupuncture was Results A total of 7679 publications were identified. identified as the most frequently provided Twenty- two articles reporting on 21 studies were included. method of complementary and alternative Meta- analyses suggest at least one AE occurring in medicine with 80 000 physicians and 16 380 9.31% (95% CI 5.10% to 14.62%, 11 studies) of patients 1 undergoing an acupuncture series and in 7.57% (95% CI non- medical practitioners. In the UK alone, 1.43% to 17.95%, 5 studies) of treatments. Summary 2.3 million traditional acupuncture treat- http://bmjopen.bmj.com/ 2 risk estimates for SAEs were 1.01 (95% CI 0.23 to 2.33, ments are carried each year. In the USA, 11 studies) per 10 000 patients and 7.98 (95% CI 1.39 the number of acupuncturists doubled to 20.00, 14 studies) per one million treatments, for between 2002 and 2012.3 The effectiveness AEs requiring treatment 1.14 (95% CI 0.00 to 7.37, 8 of acupuncture is supported by level 1a studies) per 1000 patients. Heterogeneity was substantial evidence, for example, for chronic musculo- 2 (I >80%). On average, 9.4 AEs occurred in 100 skeletal pain and headache,4–6 postoperative treatments. Half of the AEs were bleeding, pain or flare pain,7 8 postoperative nausea and vomiting,9 at the needle site that are argued to represent intended as well as allergic rhinitis.10 Furthermore, on October 4, 2021 by guest. Protected copyright. © Author(s) (or their acupuncture reaction. AE definitions and assessments employer(s)) 2021. Re- use promising evidence exists for its potential varied largely. permitted under CC BY-NC. No role in the treatment of numerous other commercial re- use. See rights Conclusion Acupuncture can be considered among 11 the safer treatments in medicine. SAEs are rare, and the indications, such as stroke rehabilitation, and permissions. Published by 12 BMJ. most common minor AEs are very mild. AEs requiring depression, aromatase inhibitor-induced 13 14 1Multidisciplinary Pain Centre, medical management are uncommon but necessitate arthralgia, and asthma. Thus, acupunc- Department of Anaesthesiology, medical competence to assure patient safety. Clinical ture offers a non-pharmacological treatment University Hospital LMU Munich, and methodological heterogeneity call for standardised option for various highly prevalent condi- Munich, Germany AE assessments tools, clear criteria for differentiating 2 tions with great disease burden and signif- School of Acupuncture, acupuncture- related AEs from therapeutically desired Moxibustion and Tuina, Beijing icant health economic impact. Long-term reactions, and identification of patient- related risk factors pharmacological treatment of these condi- Rehabilitation Hospital, Beijing for AEs. University of Chinese Medicine, tions is often associated with substantial side PROSPERO registration number CRD42020151930. Beijing, China effects.15 16 Consequently, also risk estimates on acupuncture- related adverse events (AEs) Correspondence to Dr Petra Bäumler; INTRODUCTION are required for evidence-based risk–benefit Petra. Baeumler@ med. uni- Acupuncture describes the insertion of fine considerations that are essential for clinical muenchen. de needles at defined points on the patient’s decision making. Bäumler P, et al. BMJ Open 2021;11:e045961. doi:10.1136/bmjopen-2020-045961 1 Open access BMJ Open: first published as 10.1136/bmjopen-2020-045961 on 6 September 2021. Downloaded from However, uncertainty remains about acupuncture Search strategy safety. AEs related to acupuncture are repeatedly and We searched PubMed, Scopus and Embase for articles controversially discussed in both scientific literature published before 15 September 2019 by applying the and public media. An overview of systematic reviews in following search strategy: 1: acupuncture; 2: “adverse 201717 illustrates that many of the previous reviews on event”; 3: ”adverse events”; 4: “adverse effect”; 5: “adverse the safety of acupuncture just summarised case reports effects”; #1 AND #2; #1 AND #3; #1 AND #4; #1 AND #5. or case series. In turn, those reviews, including studies Additional records were identified from previous reviews that do allow for AE frequency estimation, such as cohort on acupuncture- related AEs.17 “Acupuncture” and studies and large randomised controlled trials (RCTs), “adverse effects” are medical subject headings (MeSH) mostly addressed only certain types of AEs, particular terms. patient groups, restricted acupuncture regimens or certain countries. These data are surely important for Inclusion and exclusion criteria clinical decision making in particular cases but leave the We included articles reporting on prospective studies overall risk of acupuncture- related AEs in the general (cohort studies, RCTs, surveys or surveillances) assessing population obscure. Additionally, debate exists about AEs associated with needle acupuncture involving manual differentiating AEs from therapeutically intended reac- or electrical stimulation combined with or without moxi- tions that are claimed to form part of the acupuncture bustion in humans as their primary outcome. Case reports treatment. For example, international consensus exists and case series were not included. Only articles published that aggravation of symptoms represents an AE, because in English or German were included. Publications on disease burden increases. However, transient worsening assessments of acupuncture point injection therapies or of symptoms followed by long-term improvements can be non- penetrating acupuncture point stimulation, such as interpreted as a so-called healing crisis in complementary laser acupuncture, acupressure or transcutaneous elec- and alternative medicine.18 In contrast, such consensus trical nerve stimulation, were excluded. We also excluded is still missing for local reactions, such as small bleedings articles reporting solely on moxibustion or restricted upon needle withdrawal, needling pain, and flare around acupuncture regimens, such as press- needle, auricular, or the needling site. These are also interpreted as beneficial one- point acupuncture. Trials focusing just on one type signs by acupuncture experts and in standard textbooks of acupuncture- related AE or just on a narrowly defined and have been linked to neurophysiological mechanisms patient population were excluded. of acupuncture. Accordingly, quality and intensity of these events should be considered when classifying them Article selection and data extraction as AE.19–21 Article selection was performed independently by two The last review on prospective studies on AEs related reviewers (WZ and PB, TS and PB, or LM and PB). to acupuncture with high external validity dates back Retrieved records were first screened for eligibility by to 2001,22 did not meta-analytically summarise AE risk abstract. Full texts were obtained for the remaining http://bmjopen.bmj.com/ estimates, and did not assess the quality of included articles. Final decision about eligibility was obtained by studies. In addition, inconsistency and incompleteness consensus of all four reviewers. of reporting in primary studies hampered the drawing Estimates of overall risks and risks for each reported of firm conclusions on acupuncture safety. Since then, type of AEs were extracted as absolute numbers of patients various large- scale clinical trials and nationwide surveys with AE per total number of patients and treatments with on acupuncture