Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The use of acupuncture in patients with Raynaud’s syndrome: a systematic review and meta-analysis of randomized controlled trials

The use of acupuncture in patients with Raynaud’s syndrome: a systematic review and meta-analysis... Objective: To assess the effectiveness of acupuncture for the treatment of Raynaud’s syndrome by conducting a sys- tematic review and meta-analysis of randomized controlled trials (RCTs). Methods: Studies were identified from English and Chinese databases from their inception to September 2020. The outcomes of interest were remission incidence, number of daily attacks, incidence of positive cold stimulation tests and incidence of cold provocation tests. We conducted meta-analysis and network meta-analysis using meta and gemtc. Results: Six trials (n = 272 participants) were included in the meta-analysis. Pairwise meta-analyses show that acupunc- ture was associated with increased remission incidence (risk ratio (RR) = 1.21, 95% confidence interval (CI) = 1.10 to 1.34), decreased daily number of attacks (weighted mean difference (WMD) = −0.57, 95% CI = −1.14 to −0.01), and increased incidence of positive cold stimulation tests (RR = 1.64, 95% CI = 1.27 to 2.11). There was not enough evidence to associate acupuncture with decreased incidence of positive cold provocation tests. The network meta-analyses did not demonstrate significant results for the effectiveness of any acupuncture treatments (electroacupuncture or manual acupuncture ± moxibustion), compared with controls, in terms of remission incidence or daily number of attacks, pos- sibly due to small sample sizes and a lack of statistical power. Conclusion: The use of acupuncture may be effective for the treatment of Raynaud’s syndrome in terms of increas- ing remission incidence, decreasing daily number of attacks and increasing incidences of positive cold stimulation tests. However, our findings should be interpreted with caution due to small sample sizes, very low quality of evidence and high risk of bias. Future large-scale RCTs are warranted. Keywords acupuncture, meta-analysis, network meta-analysis, Raynaud’s phenomenon, Raynaud’s syndrome Accepted: 9 January 2022 Introduction Faculty of Health Sciences, McMaster University, Hamilton, ON, Raynaud’s syndrome, a phenomenon characterized by Canada color changes in the digits due to exaggerated vasospasms, Faculty of Applied Health Sciences, University of Waterloo, Waterloo, is a prevalent condition that occurs in 3%–5% of the global ON, Canada population. Its onset is typically triggered by cold temper- Corresponding author: atures, emotional stress, or other medical and environmen- Jiawen Deng, Faculty of Health Sciences, McMaster University, 1280 tal factors. During a Raynaud’s episode, commonly Main Street West, Hamilton, ON L8S 4L8, Canada. referred to as an “attack,” the affected digits often undergo Email: dengj35@mcmaster.ca Acupuncture in Medicine, 00(0) 2 Acupuncture in Medicine a three-phase color change: an initial ischemic phase, when a Chinese search strategy: (1) Wanfang Data; (2) Wanfang the digits turn white (pallor); a deoxygenation phase, when Med Online; (3) China National Knowledge Infrastructure the digits turn blue (cyanosis); and a reperfusion phase, (CNKI); (4) Chongqing VIP Information (CQVIP); and (5) when the digits turn red (erythema). SinoMed. Raynaud’s syndrome can be classified clinically as pri- The reference sections of previous reviews identified mary or secondary. Primary Raynaud’s syndrome, the from database searches and clinical trial registrations pub- most common type, is idiopathic, whereas secondary lished before 17 September 2020 in the following trial reg- Raynaud’s syndrome is usually caused by conditions such istries were also hand searched for relevant trials: (1) as autoimmune diseases and cancer, as well as lifestyle ClinicalTrials.gov; (2) World Health Organization (WHO) choices such as smoking and medication use. Some International Clinical Trials Registry Platform; (3) patients with Raynaud’s syndrome, especially secondary, European Union (EU) Clinical Trial Register; and (4) may experience attacks that are frequent and painful, and Chinese Clinical Trial Registry. can lead to digital ulcerations. To mitigate these symp- The search strategy used for the database searches can toms, patients with Raynaud’s syndrome are usually treated be found in Supplemental Table S2/S3. with pharmacotherapies such as calcium channel blockers (which are first-line treatments). However, these may Eligibility criteria cause a variety of different adverse reactions, such as vaso- dilation, gastrointestinal effects, and drug-drug interac- In order to be included in our analysis, parallel RCTs tions. Because of these side effects, many patients with needed to have: (1) recruited patients diagnosed with Raynaud’s syndrome have turned to complementary and Raynaud’s syndrome; (2) used any acupuncture therapy alternative medicine (CAM) to manage their symptoms. compared with an untreated or sham acupuncture group; Acupuncture, a practice that originated from China, has and (3) reported any of our outcomes of interest. recently been identified in the field of CAM as a potential We included studies that may have administered other therapeutic procedure with supportive scientific evidence. therapies concurrently with acupuncture; however, the Recently, there have been several randomized controlled same concurrent therapy needed to have been used in both trials (RCTs) investigating its use in Raynaud’s syndrome; the intervention and control arms (i.e. acupuncture + con- however, these studies have often had low sample sizes and current therapy vs concurrent therapy) to minimize poten- therefore have tended to produce inconclusive results. One tial confounding by inclusion of concurrent therapies. solution to this issue is to conduct a meta-analysis, which enables the pooling of outcome data to increase patient Outcomes sample size and statistical power, thus allowing for a more precise estimate of the treatment effects of an intervention. Our primary outcomes were: (1) incidence of remission, Therefore, we conducted a systematic review and meta- defined according to individual study criteria; and (2) num- analysis of RCTs with the objective of investigating ber of attacks per day after the completion of acupuncture whether the use of acupuncture would result in increased therapy. Our secondary outcomes included: (1) incidence remission rates and decreased daily numbers of attacks in of a positive cold provocation test after the completion of patients with Raynaud’s syndrome. acupuncture—this test involves immersing the affected digits into cold water and observing whether a Raynaud’s attack is triggered using photoplethysmography (a positive Methods test is defined as a successfully triggered attack); and (2) We conducted this systematic review and meta-analysis in incidence of a positive cold stimulation test after the com- accordance with the Preferred Reporting Items for pletion of acupuncture—this test involves immersing the Systematic Reviews and Meta-Analyses (PRISMA) frame- affected digits into cold water and measuring the length of work. The PRISMA checklist for this study is shown in time required for the digits to return to a normal tempera- Supplemental Table S1. ture (a positive test is defined as the restoration of normal temperature before a predefined time point). Study identification Study selection and data extraction We searched the following databases from their inception to 17 September 2020: (1) MEDLINE; (2) EMBASE; (3) Two authors performed title and abstract screening in dupli- Web of Science; (4) Cumulative Index of Nursing and cate based on the aforementioned eligibility criteria. Allied Health Literature (CINAHL); (5) The Cochrane Abstracts deemed to be relevant were then entered into a Library; and (6) Scopus. duplicate full-text screening process. We resolved disagree- We also systematically searched the following Chinese ments by consulting with a senior author (J.D.) to reach databases from their inception to 17 September 2020 using consensus. Acupuncture in Medicine, 00(0) Zhou et al. 3 We carried out data extraction in duplicate using pro- Publication bias spectively developed data extraction sheets. Disagreements We drew funnel plots and used Egger’s regression test to were resolved by consulting the senior author (J.D.) to identify publication bias within the included studies. We review the data. For studies with missing information, we also visually inspected the funnel plots for signs of asym- made attempts to contact the corresponding and/or first metry, as the Egger’s test lacks power to detect publication authors of these studies to obtain unpublished data. bias when there are fewer than 10 studies. Risk of bias Quality of evidence We evaluated the risk of bias of included studies in dupli- Quality of evidence was assessed using the Grading of cate using the Cochrane Collaboration’s revised tool for Recommendations, Assessment, Development and assessing risk of bias (RoB) in randomized trials (RoB 2). Evaluations (GRADE) framework for each outcome. RoB for included studies was rated using the signaling questions and algorithm maps provided by the RoB 2 guid- ance document. Network meta-analysis For the outcome of remission incidence and number of daily Meta-analysis attacks, we encountered several different forms of acupunc- ture, specifically manual acupuncture (MA) with moxibus- We conducted all statistical analyses using R 4.0.0, and we tion and electroacupuncture (EA). As these procedures differ performed random effect meta-analyses using the meta from standalone MA, it was considered that they may result 4.12-0 library. For the outcome of remission incidence and in different treatment effects. To compare the effects of differ- our secondary outcomes, we expressed and pooled treat- ent forms of acupuncture procedure, we conducted random ment effects as risk ratios (RRs) with corresponding 95% effects network meta-analyses (NMAs) using gemtc 0.8-4. confidence intervals (CIs). We also calculated the number Treatment effects in terms of remission incidence were needed to treat (NNT) for these outcomes. For continuous expressed as RRs with 95% credible intervals (CrIs), and outcomes, that is, the number of attacks per day, we treatment effects in terms of the number of daily attacks were expressed and pooled the treatment effect as weighted expressed as WMDs with 95% CrIs. We ranked the treat- mean differences (WMDs) with corresponding 95% CIs. ments in terms of their efficacy using surface under the cumu- lative ranking curve (SUCRA) scores. The heterogeneity associated with the NMAs was examined using I statistics. Heterogeneity assessment The quality of evidence within each network was assessed We assessed the presence of heterogeneity using Cochran’s using the Confidence in Network Meta-Analysis (CINeMA) Q test, and we considered a p < 0.10 as statistically sig- application, based on the GRADE framework. 15 2 nificant. We then quantified heterogeneity using I statis- 15 2 tics. I values were interpreted based on recommendations from the Cochrane Handbook. Results Study selection Meta-regression Our study selection and screening process is shown in We performed meta-regression analyses on several study- Figure 1. In total, we retrieved 4086 abstract entries from level covariates to explore sources of heterogeneity. Our our database search; only 48 abstracts were included in the covariates of interest included: (1) mean age; (2) gender full-text screening. We excluded 42 full-text studies that (percentage of female patients); (3) primary/secondary were duplicate trials, had ineligible comparisons with the Raynaud’s syndrome (percentage of patients with primary use of different concurrent therapies between treatment Raynaud’s syndrome); and (4) length of follow-up. arms, did not report any of our outcomes of interest, or rep- resented observational studies, reviews, editorials or other irrelevant publications. Influence analysis We identified outliers with extreme treatment effects within Study characteristics each meta-analysis using influence analyses and Graphical 16,17 display Of Study Heterogeneity (GOSH) analyses. If We included six RCTs published between 1997 and 2016 8,22–26 we detected outliers, we performed sensitivity analyses with 272 patients in our meta-analysis (see Table 1). 8,25 excluding the outlying studies to examine their effects on Two studies (33%) used verum/traditional MA, two stud- 22,23 the pooled effect size and heterogeneity measures. ies (33%) used MA + moxibustion, and two studies (33%) Acupuncture in Medicine, 00(0) 4 Acupuncture in Medicine Figure 1. PRISMA flowchart for the identification and selection of randomized controlled trials. CINAHL: Cumulative Index of Nursing and Allied Health Literature; CNKI: China National Knowledge Infrastructure; CQVIP: Chongqing VIP Information. 24,26 used EA. Four studies (66%) included patients with pri- providers and outcome assessors including lack of sham 22,24–26 mary Raynaud’s syndrome, one study (17%) included interventions (Figure 2). patients with Raynaud’s syndrome secondary to systemic sclerosis, mixed connective tissue disease or systemic lupus Remission incidence erythematosus, and one study did not report the classifica- tion of Raynaud’s syndrome that was included. The num- Figure 3(a) shows the meta-analysis forest plot for remis- ber of acupuncture sessions varied from 7 to 50, and the sion incidence. Four RCTs (n = 220 patients) were included follow-up period varied from 15 to 91 days. All studies were in this analysis. A majority of the included studies (three deemed to be at high risk of overall bias, mainly due to the studies, 75%) defined remission as a reduction in severity absence of prospectively registered study protocols, unre- of symptoms, increase in cold endurance of the digits, and 23,24,26 ported and/or infeasible blinding of participants, healthcare improvements as observed by nailfold capillaroscopy. Acupuncture in Medicine, 00(0) Zhou et al. 5 Acupuncture in Medicine, 00(0) Table 1. Characteristics of included trials. Study Treatment Concurrent therapy Sample Age Sex Primary Mean disease Follow-up Number of size (mean ± SD (M/F) Raynaud’s duration, months (days) acupuncture or range) syndrome (mean ± SD or treatments (%) range) Appiah None 16 5/11 100 91 7, once every other MA + moxibustion 41.5 ± 10.7 16.1 ± 14.6 et al. day over 2 weeks No treatment 17 5/12 100 91 – 45.5 ± 11.5 11.4 ± 11.1 Wang Metoprolol 100 mg/day 30 26–58 9/21 – 3–36 15 15, once per day MA + moxibustion et al. No treatment 30 24–57 7/23 – 2–34 15 – Hahn MA None 11 1/10 0 – 56 8, once per week 47 ± 12 et al. Off-point sham 8 2/6 0 – 56 – 41 ± 11 acupuncture Yang EA Herbal medicine 30 0/30 100 30 30, once per day 37.70 ± 8.22 4.17 ± 1.26 et al. (safflower injection), cyclophosphamide, buflomedil No treatment 30 0/30 100 30 – 37.97 ± 8.71 4.57 ± 1.77 Ren and MA Reserpine 1 mg/day 30 58 11/19 100 1–60 60 50, once per day Yu with 3-day rest after every 10 treatments No treatment 30 57 13/17 100 1–60 60 – Song EA Herbal medicine 20 33 10/50 100 2–24 30 30, once per day et al. (safflower yellow with 5-day rest after injection), alprostadil every 15 treatments No treatment 20 100 30 – SD: standard deviation; M: male; F: female; MA: manual acupuncture; EA: electroacupuncture. 6 Acupuncture in Medicine Figure 2. Risk of bias 2.0 ratings for included studies. One study (25%) defined remission as a reduction in the Because one study used a different definition of remis- severity of symptoms only. sion compared to the others, we examined the impact of The use of acupuncture was associated with a statisti- excluding this study by conducting an influence analysis. cally significant increase in the incidence of remission We found that excluding this study did not impact the point (RR = 1.21, 95% CI = 1.10 to 1.34; NNT = 6.39). There was estimate of the RR; however, the CI was expanded and a lack of heterogeneity among the included studies crossed the line of no effect (RR = 1.21, 95% CI = 1.00 to (p = 0.84, I = 0%). We did not identify any outliers using 1.46). We suspect this could be due to the decrease in statis- influence and GOSH analyses, nor did we detect the pres- tical power following the removal of the study, as there ence of small study effects using funnel plots (Egger’s test, were no significant changes in heterogeneity following this p = 0.27; see Supplemental Figure S1). Meta-regression study’s exclusion. analyses of study-level covariates revealed no significant correlations between remission incidence and sex Network meta-analysis. Figure 4(a) and (b) show the network (p = 0.34), mean age (p = 0.47), or duration of follow-up and forest diagrams, respectively, for the remission incidence (p = 0.74). We did not complete a meta-regression for pri- NMA. MA + moxibustion (RR = 1.36, 95% CrI = 0.90 to mary/secondary Raynaud’s syndrome, as only studies with 2.11), MA alone (RR = 1.23, 95% CrI = 0.83 to 1.88), and EA primary Raynaud’s syndrome included this outcome. (RR = 1.17, 95% CrI = 0.89 to 1.58) were not associated with Overall, the meta-analysis was based on low-quality evi- a significant increase in remission incidence compared with dence according to GRADE, due to high within-study risk no treatment. According to SUCRA rankings, MA + moxi- of bias. bustion was likely to be the most efficacious in terms of Acupuncture in Medicine, 00(0) Zhou et al. 7 Figure 3. Forest plot comparing the treatment efficacy between control and acupuncture groups: (a) remission incidence, (b) number of daily attacks, (c) incidence of positive cold provocation test, and (d) incidence of positive cold stimulation test. RR: relative risk; MD: mean difference; CI: confidence interval. 8,22 increasing remission incidence (SUCRA 0.776), followed by included in the analysis. The use of acupuncture was MA alone (SUCRA 0.610), EA (SUCRA 0.520), and no associated with a statistically significant decrease in the treatment (SUCRA 0.095). The network experienced low number of daily attacks (WMD = −0.57, 95% CI = −1.14 heterogeneity (I = 7%). Overall, the NMA was based on very to −0.01). There was a lack of heterogeneity among the low-quality evidence due to high within-study risk of bias, included studies (p = 0.89, I = 0%). We were unable to imprecision and incoherence. conduct a meta-regression, outlier analysis or Egger’s regression test due to the low number of included trials reporting this outcome. Overall, the meta-analysis was Number of daily attacks based on very low-quality evidence due to high within- Figure 3(b) shows the meta-analysis forest plot for the study risk of bias and inability to examine publication number of daily attacks. Two RCTs (n = 52 patients) were bias. Acupuncture in Medicine, 00(0) 8 Acupuncture in Medicine Figure 4. Network and forest plots for network meta-analyses. The size of the nodes represents the relative number of patients, and the thickness of the edge represents the relative number of studies comparing the connected treatment nodes in the network meta-analysis. (a) Network diagram for remission incidence. (b) Forest plot for remission incidence. (c) Network diagram for number of daily attacks. (d) Forest plot for number of daily attacks. MA: manual acupuncture; EA: electroacupuncture; SUCRA: surface under the cumulative ranking curve; RR: relative risk; MD: mean difference; CrI: credible interval. Network meta-analysis. Figure 4(c) and (d) show the net- Cold stimulation test work and forest diagrams, respectively, for the number Figure 3(d) shows the meta-analysis forest plot for inci- of daily attacks NMA. Both MA + moxibustion dences of positive cold stimulation tests. Two RCTs (n = 100 (WMD = −0.59, 95% CrI = −1.63 to 0.44) and MA alone 24,26 patients) were included in the analysis. The use of acu- (WMD = −0.59, 95% CrI = −1.63 to 0.44) were not sig- puncture was associated with a significant increase in the nificantly better than no treatment. According to SUCRA incidence of positive tests (RR = 1.64, 95% CI = 1.27 to rankings, MA + moxibustion (SUCRA 0.706) was most 2.11; NNT = 3.13). There was a lack of heterogeneity likely to be the most efficacious in terms of reducing the among the included studies (p = 0.90, I = 0%). We were number of daily attacks, followed by MA alone (SUCRA unable to conduct a meta-regression, outlier analysis or 0.617) and no treatment (SUCRA 0.177). The network Egger’s regression test due to the low number of included experienced low heterogeneity (I = 25%). Overall, the trials. The meta-analysis was based on very low-quality NMA was based on very low-quality evidence due to evidence due to high within-study risk of bias and inability high within-study risk of bias and concerns regarding to examine publication bias. imprecision, publication bias and incoherence. Discussion Cold provocation test In this systematic review and meta-analysis, we investi- Figure 3(c) shows the meta-analysis forest plot for the inci- gated the use of acupuncture versus control in the treatment dence of positive cold provocation tests. Two RCTs (n = 100 24,26 of Raynaud’s syndrome. We found that the use of acupunc- patients) were included in the analysis. There was no ture was associated with increased remission incidence, a significant effect of acupuncture on the incidence of positive decreased number of daily attacks, and an increased num- results, based on our pooled treatment effect (RR = 0.50, ber of positive cold stimulation tests (which indicates an 95% CI = 0.18 to 1.37; NNT = 3.85). There was a lack of het- increased rate of temperature recovery in the digits). erogeneity among the included studies (p = 0.77, I = 0%). According to our results, acupuncture did not significantly We were unable to conduct a meta-regression, outlier analy- decrease the incidence of attacks after cold provocation. sis or Egger’s regression test due to the low number of In addition, we conducted two NMAs for the outcome of included trials. The meta-analysis was based on very low- remission incidence and number of daily attacks, respec- quality evidence due to high within-study risk of bias, tively. Our findings suggested that none of the acupuncture imprecision and inability to examine publication bias. Acupuncture in Medicine, 00(0) Zhou et al. 9 treatments showed statistically significantly effectiveness data extraction process, and helped draft the final manuscript. EH and EZ critically reviewed the draft manuscript and made compared with no treatment. However, our NMAs included important revisions. JD was involved in conceiving and design- a limited number of studies and our pooled estimates had ing the study, performing database searches, conducting all sta- wide 95% CIs. This indicates that, rather than being due to tistical analyses, and drafting the final manuscript. All authors a lack of effect from acupuncture, our result may be more read and approved the final version of the manuscript accepted likely a consequence of insufficient statistical power, since for publication. our pairwise meta-analyses yielded promising results with little to no heterogeneity. Declaration of conflicting interests While the results from our pairwise meta-analyses sup- The authors declared no potential conflicts of interest with respect port the use of acupuncture in patients with Raynaud’s syn- to the research, authorship, and/or publication of this article. drome, it must be noted that these findings were based on six studies with low sample sizes and high RoB. In addi- Funding tion, the GRADE and CINeMA rating showed that the The authors received no financial support for the research, author- quality of evidence provided by our meta-analyses and net- ship, and/or publication of this article. work meta-analyses was very low. These significant limita- tions were due to an overall lack of acupuncture RCTs ORCID iD involving Raynaud’s patients, as well as shortcomings commonly associated with Chinese RCTs, which often Jiawen Deng https://orcid.org/0000-0002-8274-6468 contain poor descriptions of their methodologies due to a lack of Chinese RCT reporting guidelines. Thus, we are Supplemental material only able to offer a weak recommendation for the use of Supplemental material for this article is available online. acupuncture in patients with Raynaud’s syndrome, due to the high uncertainties associated with our findings. References Previous studies have shown that acupuncture can increase 1. Temprano KK. A review of Raynaud’s disease. Mo Med 2016; 113: the blood flow volume in peripheral arteries by stimulating 123–126. the autonomic nervous system, as well as improve local 2. Maundrell A and Proudman SM. Epidemiology of Raynaud’s phe- microcirculation and regulate the concentration of vasodila- nomenon. In: Wigley FM, Herrick AL and Flavahan NA (eds) Raynaud’s phenomenon. New York: Springer, 2015, pp. 21–35. tors such as nitric oxide. Acupuncture can also induce anti- 3. Fardoun MM, Nassif J, Issa K, et al. Raynaud’s phenomenon: a brief nociceptive effects by increasing the release of adenosine; review of the underlying mechanisms. Front Pharmacol 2016; 7: 438. this property may help to relieve pain in patients with 4. Prete M, Fatone MC, Favoino E, et al. Raynaud’s phenomenon: from Raynaud’s syndrome. Because of these benefits of acupunc- molecular pathogenesis to therapy. Autoimmun Rev 2014; 13(6): ture, a previous systematic review has highlighted that 655–667. 5. Russell RP. Side effects of calcium channel blockers. Hypertension acupuncture may be a viable alternative to conventional phar- 1988; 11: II42–II44. macotherapies for treating Raynaud’s syndrome. The results 6. Qureshi NA. How do you treat Raynaud’s syndrome in your practice? of our meta-analysis lend limited support to this viewpoint; Med Acupunct 2019; 31: 193–199. however, further investigations included larger and higher 7. Kawakita K and Okada K. Acupuncture therapy: mechanism of quality RCTs are needed to reinforce our findings. action, efficacy, and safety: a potential intervention for psychogenic disorders? Biopsychosoc Med 2014; 8: 4. 8. Hahn M, Steins A, Möhrle M, et al. Is there a vasospasmolytic effect of acupuncture in patients with secondary Raynaud phenomenon? J Conclusion Dtsch Dermatol Ges 2004; 2(9): 758–762. 9. Lee YH. Strengths and limitations of meta-analysis. Korean J Med To our knowledge, this systematic review is the first knowl- 2019; 94: 391–395. edge synthesis study to investigate the use of acupuncture 10. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for in patients with Raynaud’s syndrome. Our results suggest systematic reviews and meta-analyses: the PRISMA statement. BMJ that acupuncture may be effective for Raynaud’s patients in 2009; 339: b2535. terms of increasing remission incidence, decreasing num- 11. Yang S-S, Park K-M, Kim Y-W, et al. Three-grade classification of photoplethysmography for evaluating the effects of treatment in ber of daily attacks and increasing number of positive cold Raynaud phenomenon. Angiology 2013; 64(8): 609–613. stimulation tests. However, due to the low sample sizes of 12. Kadan M, Erol G, Karabacak K, et al. How can follow-up of patients the included RCTs, very low quality of evidence and high with Raynaud phenomenon be optimized? Med Sci Monit Basic Res RoB, our findings should be taken with caution and vali- 2015; 21: 47–52. dated using larger, higher quality RCTs. 13. Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019; 366: l4898. 14. Higgins JPT, Savović J, Page JM, et al. Revised Cochrane risk-of-bias Contributors tool for randomized trials (RoB 2), 2019, https://drive.google.com/ All authors were involved in article screening, risk of bias anal- file/d/19R9savfPdCHC8XLz2iiMvL_71lPJERWK/view ysis, and data extraction. FZ was involved in designing the 15. Higgins JPT and Green S. Cochrane handbook for systematic reviews methodology of the study, supervised the article screening and of interventions. Hoboken, NJ: Wiley, 2008. Acupuncture in Medicine, 00(0) 10 Acupuncture in Medicine 16. Olkin I, Dahabreh IJ and Trikalinos TA. GOSH—a graphical display 26. Song MY, Liu Y, Guo YZ, et al. Observation of the use of electroacu- of study heterogeneity. Res Synth Methods 2012; 3(3): 214–223. puncture in treating 60 cases of Raynaud’s phenomenon. In: The appli- 17. Viechtbauer W and Cheung MW-L. Outlier and influence diagnostics cation of modern techniques in the treatment of peripheral artery disease for meta-analysis. Res Synth Methods 2010; 1: 112–125. using traditional Chinese medicine. Beijing, China: Chinese Medical 18. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis Doctor Association, Chinese Association of Integrative Medicine, 2016, detected by a simple, graphical test. BMJ 1997; 315: 629–634. pp. 74–81. https://d.wanfangdata.com.cn/conference/9091216 19. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consen- 27. Deng J, Abbas U, Chang O, et al. Antidiabetic and antiosteoporotic sus on rating quality of evidence and strength of recommendations. pharmacotherapies for prevention and treatment of type 2 diabetes- BMJ 2008; 336: 924–926. induced bone disease: protocol for two network meta-analyses. BMJ 20. Daly CH, Neupane B, Beyene J, et al. Empirical evaluation of Open 2020; 10: e034741. SUCRA-based treatment ranks in network meta-analysis: quantifying 28. Takayama S, Watanabe M, Kusuyama H, et al. Evaluation of the effects robustness using Cohen’s kappa. BMJ Open 2019; 9: e024625. of acupuncture on blood flow in humans with ultrasound color Doppler 21. Nikolakopoulou A, Higgins JPT, Papakonstantinou T, et al. CINeMA: imaging. Evid Based Complement Alternat Med 2012; 2012: 513638. an approach for assessing confidence in the results of a network meta- 29. Komori M, Takada K, Tomizawa Y, et al. Microcirculatory responses analysis. PLoS Med 2020; 17(4): e1003082. to acupuncture stimulation and phototherapy. Anesth Analg 2009; 22. Appiah R, Hiller S, Caspary L, et al. Treatment of primary Raynaud’s 108(2): 635–640. syndrome with traditional Chinese acupuncture. J Intern Med 1997; 30. Pan P, Zhang X, Qian H, et al. Effects of electro-acupuncture on 241(2): 119–124. endothelium-derived endothelin-1 and endothelial nitric oxide syn- 23. Wang S, Cai YY and Lu JR. Warm needling therapy for Raynaud’s thase of rats with hypoxia-induced pulmonary hypertension. Exp Biol syndrome in 30 cases. Shanghai J Tradit Chin Med 2003; 37: Med 2010; 235(5): 642–648. 38–39. 31. Goldman N, Chen M, Fujita T, et al. Adenosine A1 receptors medi- 24. Yang LH. Observation of the use of electroacupuncture in the treat- ate local anti-nociceptive effects of acupuncture. Nat Neurosci 2010; ment of Raynaud’s disease. Master’s Thesis, Heilongjiang University 13(7): 883–888. of Chinese Medicine, Harbin, China, 2013. 32. Qureshi NA, Sharma S, Guru R, et al. Raynaud’s disease and clinical 25. Ren TT and Yu BX. Analysis on the clinical effectiveness of acupunc- challenges: a minireview of literature with a spotlight on acupuncture. ture to treat Raynaud disease. Clin J Chin Med 2015; 7: 115–116. J Adv Med Med Res 2019; 29: 1–10. Acupuncture in Medicine, 00(0) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acupuncture in Medicine SAGE

The use of acupuncture in patients with Raynaud’s syndrome: a systematic review and meta-analysis of randomized controlled trials

Download PDF
10 pages

Loading next page...
 
/lp/sage/the-use-of-acupuncture-in-patients-with-raynaud-s-syndrome-a-iF44FKuraU

References (37)

Publisher
SAGE
Copyright
© The Author(s) 2022
ISSN
0964-5284
eISSN
1759-9873
DOI
10.1177/09645284221076504
Publisher site
See Article on Publisher Site

Abstract

Objective: To assess the effectiveness of acupuncture for the treatment of Raynaud’s syndrome by conducting a sys- tematic review and meta-analysis of randomized controlled trials (RCTs). Methods: Studies were identified from English and Chinese databases from their inception to September 2020. The outcomes of interest were remission incidence, number of daily attacks, incidence of positive cold stimulation tests and incidence of cold provocation tests. We conducted meta-analysis and network meta-analysis using meta and gemtc. Results: Six trials (n = 272 participants) were included in the meta-analysis. Pairwise meta-analyses show that acupunc- ture was associated with increased remission incidence (risk ratio (RR) = 1.21, 95% confidence interval (CI) = 1.10 to 1.34), decreased daily number of attacks (weighted mean difference (WMD) = −0.57, 95% CI = −1.14 to −0.01), and increased incidence of positive cold stimulation tests (RR = 1.64, 95% CI = 1.27 to 2.11). There was not enough evidence to associate acupuncture with decreased incidence of positive cold provocation tests. The network meta-analyses did not demonstrate significant results for the effectiveness of any acupuncture treatments (electroacupuncture or manual acupuncture ± moxibustion), compared with controls, in terms of remission incidence or daily number of attacks, pos- sibly due to small sample sizes and a lack of statistical power. Conclusion: The use of acupuncture may be effective for the treatment of Raynaud’s syndrome in terms of increas- ing remission incidence, decreasing daily number of attacks and increasing incidences of positive cold stimulation tests. However, our findings should be interpreted with caution due to small sample sizes, very low quality of evidence and high risk of bias. Future large-scale RCTs are warranted. Keywords acupuncture, meta-analysis, network meta-analysis, Raynaud’s phenomenon, Raynaud’s syndrome Accepted: 9 January 2022 Introduction Faculty of Health Sciences, McMaster University, Hamilton, ON, Raynaud’s syndrome, a phenomenon characterized by Canada color changes in the digits due to exaggerated vasospasms, Faculty of Applied Health Sciences, University of Waterloo, Waterloo, is a prevalent condition that occurs in 3%–5% of the global ON, Canada population. Its onset is typically triggered by cold temper- Corresponding author: atures, emotional stress, or other medical and environmen- Jiawen Deng, Faculty of Health Sciences, McMaster University, 1280 tal factors. During a Raynaud’s episode, commonly Main Street West, Hamilton, ON L8S 4L8, Canada. referred to as an “attack,” the affected digits often undergo Email: dengj35@mcmaster.ca Acupuncture in Medicine, 00(0) 2 Acupuncture in Medicine a three-phase color change: an initial ischemic phase, when a Chinese search strategy: (1) Wanfang Data; (2) Wanfang the digits turn white (pallor); a deoxygenation phase, when Med Online; (3) China National Knowledge Infrastructure the digits turn blue (cyanosis); and a reperfusion phase, (CNKI); (4) Chongqing VIP Information (CQVIP); and (5) when the digits turn red (erythema). SinoMed. Raynaud’s syndrome can be classified clinically as pri- The reference sections of previous reviews identified mary or secondary. Primary Raynaud’s syndrome, the from database searches and clinical trial registrations pub- most common type, is idiopathic, whereas secondary lished before 17 September 2020 in the following trial reg- Raynaud’s syndrome is usually caused by conditions such istries were also hand searched for relevant trials: (1) as autoimmune diseases and cancer, as well as lifestyle ClinicalTrials.gov; (2) World Health Organization (WHO) choices such as smoking and medication use. Some International Clinical Trials Registry Platform; (3) patients with Raynaud’s syndrome, especially secondary, European Union (EU) Clinical Trial Register; and (4) may experience attacks that are frequent and painful, and Chinese Clinical Trial Registry. can lead to digital ulcerations. To mitigate these symp- The search strategy used for the database searches can toms, patients with Raynaud’s syndrome are usually treated be found in Supplemental Table S2/S3. with pharmacotherapies such as calcium channel blockers (which are first-line treatments). However, these may Eligibility criteria cause a variety of different adverse reactions, such as vaso- dilation, gastrointestinal effects, and drug-drug interac- In order to be included in our analysis, parallel RCTs tions. Because of these side effects, many patients with needed to have: (1) recruited patients diagnosed with Raynaud’s syndrome have turned to complementary and Raynaud’s syndrome; (2) used any acupuncture therapy alternative medicine (CAM) to manage their symptoms. compared with an untreated or sham acupuncture group; Acupuncture, a practice that originated from China, has and (3) reported any of our outcomes of interest. recently been identified in the field of CAM as a potential We included studies that may have administered other therapeutic procedure with supportive scientific evidence. therapies concurrently with acupuncture; however, the Recently, there have been several randomized controlled same concurrent therapy needed to have been used in both trials (RCTs) investigating its use in Raynaud’s syndrome; the intervention and control arms (i.e. acupuncture + con- however, these studies have often had low sample sizes and current therapy vs concurrent therapy) to minimize poten- therefore have tended to produce inconclusive results. One tial confounding by inclusion of concurrent therapies. solution to this issue is to conduct a meta-analysis, which enables the pooling of outcome data to increase patient Outcomes sample size and statistical power, thus allowing for a more precise estimate of the treatment effects of an intervention. Our primary outcomes were: (1) incidence of remission, Therefore, we conducted a systematic review and meta- defined according to individual study criteria; and (2) num- analysis of RCTs with the objective of investigating ber of attacks per day after the completion of acupuncture whether the use of acupuncture would result in increased therapy. Our secondary outcomes included: (1) incidence remission rates and decreased daily numbers of attacks in of a positive cold provocation test after the completion of patients with Raynaud’s syndrome. acupuncture—this test involves immersing the affected digits into cold water and observing whether a Raynaud’s attack is triggered using photoplethysmography (a positive Methods test is defined as a successfully triggered attack); and (2) We conducted this systematic review and meta-analysis in incidence of a positive cold stimulation test after the com- accordance with the Preferred Reporting Items for pletion of acupuncture—this test involves immersing the Systematic Reviews and Meta-Analyses (PRISMA) frame- affected digits into cold water and measuring the length of work. The PRISMA checklist for this study is shown in time required for the digits to return to a normal tempera- Supplemental Table S1. ture (a positive test is defined as the restoration of normal temperature before a predefined time point). Study identification Study selection and data extraction We searched the following databases from their inception to 17 September 2020: (1) MEDLINE; (2) EMBASE; (3) Two authors performed title and abstract screening in dupli- Web of Science; (4) Cumulative Index of Nursing and cate based on the aforementioned eligibility criteria. Allied Health Literature (CINAHL); (5) The Cochrane Abstracts deemed to be relevant were then entered into a Library; and (6) Scopus. duplicate full-text screening process. We resolved disagree- We also systematically searched the following Chinese ments by consulting with a senior author (J.D.) to reach databases from their inception to 17 September 2020 using consensus. Acupuncture in Medicine, 00(0) Zhou et al. 3 We carried out data extraction in duplicate using pro- Publication bias spectively developed data extraction sheets. Disagreements We drew funnel plots and used Egger’s regression test to were resolved by consulting the senior author (J.D.) to identify publication bias within the included studies. We review the data. For studies with missing information, we also visually inspected the funnel plots for signs of asym- made attempts to contact the corresponding and/or first metry, as the Egger’s test lacks power to detect publication authors of these studies to obtain unpublished data. bias when there are fewer than 10 studies. Risk of bias Quality of evidence We evaluated the risk of bias of included studies in dupli- Quality of evidence was assessed using the Grading of cate using the Cochrane Collaboration’s revised tool for Recommendations, Assessment, Development and assessing risk of bias (RoB) in randomized trials (RoB 2). Evaluations (GRADE) framework for each outcome. RoB for included studies was rated using the signaling questions and algorithm maps provided by the RoB 2 guid- ance document. Network meta-analysis For the outcome of remission incidence and number of daily Meta-analysis attacks, we encountered several different forms of acupunc- ture, specifically manual acupuncture (MA) with moxibus- We conducted all statistical analyses using R 4.0.0, and we tion and electroacupuncture (EA). As these procedures differ performed random effect meta-analyses using the meta from standalone MA, it was considered that they may result 4.12-0 library. For the outcome of remission incidence and in different treatment effects. To compare the effects of differ- our secondary outcomes, we expressed and pooled treat- ent forms of acupuncture procedure, we conducted random ment effects as risk ratios (RRs) with corresponding 95% effects network meta-analyses (NMAs) using gemtc 0.8-4. confidence intervals (CIs). We also calculated the number Treatment effects in terms of remission incidence were needed to treat (NNT) for these outcomes. For continuous expressed as RRs with 95% credible intervals (CrIs), and outcomes, that is, the number of attacks per day, we treatment effects in terms of the number of daily attacks were expressed and pooled the treatment effect as weighted expressed as WMDs with 95% CrIs. We ranked the treat- mean differences (WMDs) with corresponding 95% CIs. ments in terms of their efficacy using surface under the cumu- lative ranking curve (SUCRA) scores. The heterogeneity associated with the NMAs was examined using I statistics. Heterogeneity assessment The quality of evidence within each network was assessed We assessed the presence of heterogeneity using Cochran’s using the Confidence in Network Meta-Analysis (CINeMA) Q test, and we considered a p < 0.10 as statistically sig- application, based on the GRADE framework. 15 2 nificant. We then quantified heterogeneity using I statis- 15 2 tics. I values were interpreted based on recommendations from the Cochrane Handbook. Results Study selection Meta-regression Our study selection and screening process is shown in We performed meta-regression analyses on several study- Figure 1. In total, we retrieved 4086 abstract entries from level covariates to explore sources of heterogeneity. Our our database search; only 48 abstracts were included in the covariates of interest included: (1) mean age; (2) gender full-text screening. We excluded 42 full-text studies that (percentage of female patients); (3) primary/secondary were duplicate trials, had ineligible comparisons with the Raynaud’s syndrome (percentage of patients with primary use of different concurrent therapies between treatment Raynaud’s syndrome); and (4) length of follow-up. arms, did not report any of our outcomes of interest, or rep- resented observational studies, reviews, editorials or other irrelevant publications. Influence analysis We identified outliers with extreme treatment effects within Study characteristics each meta-analysis using influence analyses and Graphical 16,17 display Of Study Heterogeneity (GOSH) analyses. If We included six RCTs published between 1997 and 2016 8,22–26 we detected outliers, we performed sensitivity analyses with 272 patients in our meta-analysis (see Table 1). 8,25 excluding the outlying studies to examine their effects on Two studies (33%) used verum/traditional MA, two stud- 22,23 the pooled effect size and heterogeneity measures. ies (33%) used MA + moxibustion, and two studies (33%) Acupuncture in Medicine, 00(0) 4 Acupuncture in Medicine Figure 1. PRISMA flowchart for the identification and selection of randomized controlled trials. CINAHL: Cumulative Index of Nursing and Allied Health Literature; CNKI: China National Knowledge Infrastructure; CQVIP: Chongqing VIP Information. 24,26 used EA. Four studies (66%) included patients with pri- providers and outcome assessors including lack of sham 22,24–26 mary Raynaud’s syndrome, one study (17%) included interventions (Figure 2). patients with Raynaud’s syndrome secondary to systemic sclerosis, mixed connective tissue disease or systemic lupus Remission incidence erythematosus, and one study did not report the classifica- tion of Raynaud’s syndrome that was included. The num- Figure 3(a) shows the meta-analysis forest plot for remis- ber of acupuncture sessions varied from 7 to 50, and the sion incidence. Four RCTs (n = 220 patients) were included follow-up period varied from 15 to 91 days. All studies were in this analysis. A majority of the included studies (three deemed to be at high risk of overall bias, mainly due to the studies, 75%) defined remission as a reduction in severity absence of prospectively registered study protocols, unre- of symptoms, increase in cold endurance of the digits, and 23,24,26 ported and/or infeasible blinding of participants, healthcare improvements as observed by nailfold capillaroscopy. Acupuncture in Medicine, 00(0) Zhou et al. 5 Acupuncture in Medicine, 00(0) Table 1. Characteristics of included trials. Study Treatment Concurrent therapy Sample Age Sex Primary Mean disease Follow-up Number of size (mean ± SD (M/F) Raynaud’s duration, months (days) acupuncture or range) syndrome (mean ± SD or treatments (%) range) Appiah None 16 5/11 100 91 7, once every other MA + moxibustion 41.5 ± 10.7 16.1 ± 14.6 et al. day over 2 weeks No treatment 17 5/12 100 91 – 45.5 ± 11.5 11.4 ± 11.1 Wang Metoprolol 100 mg/day 30 26–58 9/21 – 3–36 15 15, once per day MA + moxibustion et al. No treatment 30 24–57 7/23 – 2–34 15 – Hahn MA None 11 1/10 0 – 56 8, once per week 47 ± 12 et al. Off-point sham 8 2/6 0 – 56 – 41 ± 11 acupuncture Yang EA Herbal medicine 30 0/30 100 30 30, once per day 37.70 ± 8.22 4.17 ± 1.26 et al. (safflower injection), cyclophosphamide, buflomedil No treatment 30 0/30 100 30 – 37.97 ± 8.71 4.57 ± 1.77 Ren and MA Reserpine 1 mg/day 30 58 11/19 100 1–60 60 50, once per day Yu with 3-day rest after every 10 treatments No treatment 30 57 13/17 100 1–60 60 – Song EA Herbal medicine 20 33 10/50 100 2–24 30 30, once per day et al. (safflower yellow with 5-day rest after injection), alprostadil every 15 treatments No treatment 20 100 30 – SD: standard deviation; M: male; F: female; MA: manual acupuncture; EA: electroacupuncture. 6 Acupuncture in Medicine Figure 2. Risk of bias 2.0 ratings for included studies. One study (25%) defined remission as a reduction in the Because one study used a different definition of remis- severity of symptoms only. sion compared to the others, we examined the impact of The use of acupuncture was associated with a statisti- excluding this study by conducting an influence analysis. cally significant increase in the incidence of remission We found that excluding this study did not impact the point (RR = 1.21, 95% CI = 1.10 to 1.34; NNT = 6.39). There was estimate of the RR; however, the CI was expanded and a lack of heterogeneity among the included studies crossed the line of no effect (RR = 1.21, 95% CI = 1.00 to (p = 0.84, I = 0%). We did not identify any outliers using 1.46). We suspect this could be due to the decrease in statis- influence and GOSH analyses, nor did we detect the pres- tical power following the removal of the study, as there ence of small study effects using funnel plots (Egger’s test, were no significant changes in heterogeneity following this p = 0.27; see Supplemental Figure S1). Meta-regression study’s exclusion. analyses of study-level covariates revealed no significant correlations between remission incidence and sex Network meta-analysis. Figure 4(a) and (b) show the network (p = 0.34), mean age (p = 0.47), or duration of follow-up and forest diagrams, respectively, for the remission incidence (p = 0.74). We did not complete a meta-regression for pri- NMA. MA + moxibustion (RR = 1.36, 95% CrI = 0.90 to mary/secondary Raynaud’s syndrome, as only studies with 2.11), MA alone (RR = 1.23, 95% CrI = 0.83 to 1.88), and EA primary Raynaud’s syndrome included this outcome. (RR = 1.17, 95% CrI = 0.89 to 1.58) were not associated with Overall, the meta-analysis was based on low-quality evi- a significant increase in remission incidence compared with dence according to GRADE, due to high within-study risk no treatment. According to SUCRA rankings, MA + moxi- of bias. bustion was likely to be the most efficacious in terms of Acupuncture in Medicine, 00(0) Zhou et al. 7 Figure 3. Forest plot comparing the treatment efficacy between control and acupuncture groups: (a) remission incidence, (b) number of daily attacks, (c) incidence of positive cold provocation test, and (d) incidence of positive cold stimulation test. RR: relative risk; MD: mean difference; CI: confidence interval. 8,22 increasing remission incidence (SUCRA 0.776), followed by included in the analysis. The use of acupuncture was MA alone (SUCRA 0.610), EA (SUCRA 0.520), and no associated with a statistically significant decrease in the treatment (SUCRA 0.095). The network experienced low number of daily attacks (WMD = −0.57, 95% CI = −1.14 heterogeneity (I = 7%). Overall, the NMA was based on very to −0.01). There was a lack of heterogeneity among the low-quality evidence due to high within-study risk of bias, included studies (p = 0.89, I = 0%). We were unable to imprecision and incoherence. conduct a meta-regression, outlier analysis or Egger’s regression test due to the low number of included trials reporting this outcome. Overall, the meta-analysis was Number of daily attacks based on very low-quality evidence due to high within- Figure 3(b) shows the meta-analysis forest plot for the study risk of bias and inability to examine publication number of daily attacks. Two RCTs (n = 52 patients) were bias. Acupuncture in Medicine, 00(0) 8 Acupuncture in Medicine Figure 4. Network and forest plots for network meta-analyses. The size of the nodes represents the relative number of patients, and the thickness of the edge represents the relative number of studies comparing the connected treatment nodes in the network meta-analysis. (a) Network diagram for remission incidence. (b) Forest plot for remission incidence. (c) Network diagram for number of daily attacks. (d) Forest plot for number of daily attacks. MA: manual acupuncture; EA: electroacupuncture; SUCRA: surface under the cumulative ranking curve; RR: relative risk; MD: mean difference; CrI: credible interval. Network meta-analysis. Figure 4(c) and (d) show the net- Cold stimulation test work and forest diagrams, respectively, for the number Figure 3(d) shows the meta-analysis forest plot for inci- of daily attacks NMA. Both MA + moxibustion dences of positive cold stimulation tests. Two RCTs (n = 100 (WMD = −0.59, 95% CrI = −1.63 to 0.44) and MA alone 24,26 patients) were included in the analysis. The use of acu- (WMD = −0.59, 95% CrI = −1.63 to 0.44) were not sig- puncture was associated with a significant increase in the nificantly better than no treatment. According to SUCRA incidence of positive tests (RR = 1.64, 95% CI = 1.27 to rankings, MA + moxibustion (SUCRA 0.706) was most 2.11; NNT = 3.13). There was a lack of heterogeneity likely to be the most efficacious in terms of reducing the among the included studies (p = 0.90, I = 0%). We were number of daily attacks, followed by MA alone (SUCRA unable to conduct a meta-regression, outlier analysis or 0.617) and no treatment (SUCRA 0.177). The network Egger’s regression test due to the low number of included experienced low heterogeneity (I = 25%). Overall, the trials. The meta-analysis was based on very low-quality NMA was based on very low-quality evidence due to evidence due to high within-study risk of bias and inability high within-study risk of bias and concerns regarding to examine publication bias. imprecision, publication bias and incoherence. Discussion Cold provocation test In this systematic review and meta-analysis, we investi- Figure 3(c) shows the meta-analysis forest plot for the inci- gated the use of acupuncture versus control in the treatment dence of positive cold provocation tests. Two RCTs (n = 100 24,26 of Raynaud’s syndrome. We found that the use of acupunc- patients) were included in the analysis. There was no ture was associated with increased remission incidence, a significant effect of acupuncture on the incidence of positive decreased number of daily attacks, and an increased num- results, based on our pooled treatment effect (RR = 0.50, ber of positive cold stimulation tests (which indicates an 95% CI = 0.18 to 1.37; NNT = 3.85). There was a lack of het- increased rate of temperature recovery in the digits). erogeneity among the included studies (p = 0.77, I = 0%). According to our results, acupuncture did not significantly We were unable to conduct a meta-regression, outlier analy- decrease the incidence of attacks after cold provocation. sis or Egger’s regression test due to the low number of In addition, we conducted two NMAs for the outcome of included trials. The meta-analysis was based on very low- remission incidence and number of daily attacks, respec- quality evidence due to high within-study risk of bias, tively. Our findings suggested that none of the acupuncture imprecision and inability to examine publication bias. Acupuncture in Medicine, 00(0) Zhou et al. 9 treatments showed statistically significantly effectiveness data extraction process, and helped draft the final manuscript. EH and EZ critically reviewed the draft manuscript and made compared with no treatment. However, our NMAs included important revisions. JD was involved in conceiving and design- a limited number of studies and our pooled estimates had ing the study, performing database searches, conducting all sta- wide 95% CIs. This indicates that, rather than being due to tistical analyses, and drafting the final manuscript. All authors a lack of effect from acupuncture, our result may be more read and approved the final version of the manuscript accepted likely a consequence of insufficient statistical power, since for publication. our pairwise meta-analyses yielded promising results with little to no heterogeneity. Declaration of conflicting interests While the results from our pairwise meta-analyses sup- The authors declared no potential conflicts of interest with respect port the use of acupuncture in patients with Raynaud’s syn- to the research, authorship, and/or publication of this article. drome, it must be noted that these findings were based on six studies with low sample sizes and high RoB. In addi- Funding tion, the GRADE and CINeMA rating showed that the The authors received no financial support for the research, author- quality of evidence provided by our meta-analyses and net- ship, and/or publication of this article. work meta-analyses was very low. These significant limita- tions were due to an overall lack of acupuncture RCTs ORCID iD involving Raynaud’s patients, as well as shortcomings commonly associated with Chinese RCTs, which often Jiawen Deng https://orcid.org/0000-0002-8274-6468 contain poor descriptions of their methodologies due to a lack of Chinese RCT reporting guidelines. Thus, we are Supplemental material only able to offer a weak recommendation for the use of Supplemental material for this article is available online. acupuncture in patients with Raynaud’s syndrome, due to the high uncertainties associated with our findings. References Previous studies have shown that acupuncture can increase 1. Temprano KK. A review of Raynaud’s disease. Mo Med 2016; 113: the blood flow volume in peripheral arteries by stimulating 123–126. the autonomic nervous system, as well as improve local 2. Maundrell A and Proudman SM. Epidemiology of Raynaud’s phe- microcirculation and regulate the concentration of vasodila- nomenon. In: Wigley FM, Herrick AL and Flavahan NA (eds) Raynaud’s phenomenon. New York: Springer, 2015, pp. 21–35. tors such as nitric oxide. Acupuncture can also induce anti- 3. Fardoun MM, Nassif J, Issa K, et al. Raynaud’s phenomenon: a brief nociceptive effects by increasing the release of adenosine; review of the underlying mechanisms. Front Pharmacol 2016; 7: 438. this property may help to relieve pain in patients with 4. Prete M, Fatone MC, Favoino E, et al. Raynaud’s phenomenon: from Raynaud’s syndrome. Because of these benefits of acupunc- molecular pathogenesis to therapy. Autoimmun Rev 2014; 13(6): ture, a previous systematic review has highlighted that 655–667. 5. Russell RP. Side effects of calcium channel blockers. Hypertension acupuncture may be a viable alternative to conventional phar- 1988; 11: II42–II44. macotherapies for treating Raynaud’s syndrome. The results 6. Qureshi NA. How do you treat Raynaud’s syndrome in your practice? of our meta-analysis lend limited support to this viewpoint; Med Acupunct 2019; 31: 193–199. however, further investigations included larger and higher 7. Kawakita K and Okada K. Acupuncture therapy: mechanism of quality RCTs are needed to reinforce our findings. action, efficacy, and safety: a potential intervention for psychogenic disorders? Biopsychosoc Med 2014; 8: 4. 8. Hahn M, Steins A, Möhrle M, et al. Is there a vasospasmolytic effect of acupuncture in patients with secondary Raynaud phenomenon? J Conclusion Dtsch Dermatol Ges 2004; 2(9): 758–762. 9. Lee YH. Strengths and limitations of meta-analysis. Korean J Med To our knowledge, this systematic review is the first knowl- 2019; 94: 391–395. edge synthesis study to investigate the use of acupuncture 10. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for in patients with Raynaud’s syndrome. Our results suggest systematic reviews and meta-analyses: the PRISMA statement. BMJ that acupuncture may be effective for Raynaud’s patients in 2009; 339: b2535. terms of increasing remission incidence, decreasing num- 11. Yang S-S, Park K-M, Kim Y-W, et al. Three-grade classification of photoplethysmography for evaluating the effects of treatment in ber of daily attacks and increasing number of positive cold Raynaud phenomenon. Angiology 2013; 64(8): 609–613. stimulation tests. However, due to the low sample sizes of 12. Kadan M, Erol G, Karabacak K, et al. How can follow-up of patients the included RCTs, very low quality of evidence and high with Raynaud phenomenon be optimized? Med Sci Monit Basic Res RoB, our findings should be taken with caution and vali- 2015; 21: 47–52. dated using larger, higher quality RCTs. 13. Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019; 366: l4898. 14. Higgins JPT, Savović J, Page JM, et al. Revised Cochrane risk-of-bias Contributors tool for randomized trials (RoB 2), 2019, https://drive.google.com/ All authors were involved in article screening, risk of bias anal- file/d/19R9savfPdCHC8XLz2iiMvL_71lPJERWK/view ysis, and data extraction. FZ was involved in designing the 15. Higgins JPT and Green S. Cochrane handbook for systematic reviews methodology of the study, supervised the article screening and of interventions. Hoboken, NJ: Wiley, 2008. Acupuncture in Medicine, 00(0) 10 Acupuncture in Medicine 16. Olkin I, Dahabreh IJ and Trikalinos TA. GOSH—a graphical display 26. Song MY, Liu Y, Guo YZ, et al. Observation of the use of electroacu- of study heterogeneity. Res Synth Methods 2012; 3(3): 214–223. puncture in treating 60 cases of Raynaud’s phenomenon. In: The appli- 17. Viechtbauer W and Cheung MW-L. Outlier and influence diagnostics cation of modern techniques in the treatment of peripheral artery disease for meta-analysis. Res Synth Methods 2010; 1: 112–125. using traditional Chinese medicine. Beijing, China: Chinese Medical 18. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis Doctor Association, Chinese Association of Integrative Medicine, 2016, detected by a simple, graphical test. BMJ 1997; 315: 629–634. pp. 74–81. https://d.wanfangdata.com.cn/conference/9091216 19. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consen- 27. Deng J, Abbas U, Chang O, et al. Antidiabetic and antiosteoporotic sus on rating quality of evidence and strength of recommendations. pharmacotherapies for prevention and treatment of type 2 diabetes- BMJ 2008; 336: 924–926. induced bone disease: protocol for two network meta-analyses. BMJ 20. Daly CH, Neupane B, Beyene J, et al. Empirical evaluation of Open 2020; 10: e034741. SUCRA-based treatment ranks in network meta-analysis: quantifying 28. Takayama S, Watanabe M, Kusuyama H, et al. Evaluation of the effects robustness using Cohen’s kappa. BMJ Open 2019; 9: e024625. of acupuncture on blood flow in humans with ultrasound color Doppler 21. Nikolakopoulou A, Higgins JPT, Papakonstantinou T, et al. CINeMA: imaging. Evid Based Complement Alternat Med 2012; 2012: 513638. an approach for assessing confidence in the results of a network meta- 29. Komori M, Takada K, Tomizawa Y, et al. Microcirculatory responses analysis. PLoS Med 2020; 17(4): e1003082. to acupuncture stimulation and phototherapy. Anesth Analg 2009; 22. Appiah R, Hiller S, Caspary L, et al. Treatment of primary Raynaud’s 108(2): 635–640. syndrome with traditional Chinese acupuncture. J Intern Med 1997; 30. Pan P, Zhang X, Qian H, et al. Effects of electro-acupuncture on 241(2): 119–124. endothelium-derived endothelin-1 and endothelial nitric oxide syn- 23. Wang S, Cai YY and Lu JR. Warm needling therapy for Raynaud’s thase of rats with hypoxia-induced pulmonary hypertension. Exp Biol syndrome in 30 cases. Shanghai J Tradit Chin Med 2003; 37: Med 2010; 235(5): 642–648. 38–39. 31. Goldman N, Chen M, Fujita T, et al. Adenosine A1 receptors medi- 24. Yang LH. Observation of the use of electroacupuncture in the treat- ate local anti-nociceptive effects of acupuncture. Nat Neurosci 2010; ment of Raynaud’s disease. Master’s Thesis, Heilongjiang University 13(7): 883–888. of Chinese Medicine, Harbin, China, 2013. 32. Qureshi NA, Sharma S, Guru R, et al. Raynaud’s disease and clinical 25. Ren TT and Yu BX. Analysis on the clinical effectiveness of acupunc- challenges: a minireview of literature with a spotlight on acupuncture. ture to treat Raynaud disease. Clin J Chin Med 2015; 7: 115–116. J Adv Med Med Res 2019; 29: 1–10. Acupuncture in Medicine, 00(0)

Journal

Acupuncture in Medicine – SAGE

Published: Apr 1, 2023

Keywords: acupuncture; meta-analysis; network meta-analysis; Raynaud’s phenomenon; Raynaud’s syndrome

There are no references for this article.