Acupuncture vs Standard Pharmacological Therapy for Migraine Prevention Report Writing

Background—Standard pharmacological treatment of migraine has many shortcomings. Acupuncture is becoming a more  widely used therapy for the prevention and treatment of migraine, but its effectiveness is still in question when compared to  the pharmacological treatments even though very few of these have Class A and B evidence for migraine prevention. This is a  systematic review of data from existing randomized trials that compare the effectiveness of acupuncture treatment with conventional migraine preventative medications. 

Methods—Custom-designed strategy was used for searching Pubmed (includes MEDLINE), Scopus (includes EMBASE).  The inclusion criteria were English language and randomized trials. No date restriction was utilized. We included randomized  trials and randomized controlled trials in adult patients that compared the clinical effects of acupuncture with a standard  migraine preventive medication in patients with a diagnosis of chronic or episodic migraine with or without aura. We excluded  letters and studies on acupuncture for headaches other than migraine. Two reviewers checked eligibility; extracted information  on patients, interventions, methods, and results; and assessed the quality of the acupuncture intervention based on the American  Academy of Neurology Classification of evidence matrix for therapeutic trials. The present review was not registered. 

Results—Out of the 706 search results, 7 clinical trials, with a total of 1430 participants, met inclusion criteria for trials  comparing the effectiveness of acupuncture to standard pharmacologic treatment. Several of the studies showed acupuncture to  be more effective than standard pharmacological treatments for migraine prevention; however, methodological heterogeneity  precluded aggregation of these data. 

Conclusions—There is growing evidence that acupuncture is just as effective and has fewer side effects than many of the  standard pharmaceutical agents that are currently used. However, the heterogeneity of the existing studies limits the effective  comparison and analysis. 

Key words: migraine, acupuncture, preventive medications, review 

INTRODUCTION 

Acupuncture is becoming a more widely accepted  form of integrative medicine for the prevention and treat ment of migraine in the West. Its mechanism of action  does not have a definite explanation, but some studies suggest that acupuncture may have anti-inflammatory  action via release of neuropeptides from nerve endings,  including calcitonin gene-related peptide (CGRP).1,2 Other theories suggest that acupuncture may exert anal gesic effects by the hypothalamic-pituitary-adrenal axis and the endogenous opioid system which are important  mediators of the stress response to pain.3 

In acupuncture research, true acupuncture is often  compared with sham acupuncture. There are many dif ferent types of sham acupuncture intervention; these  include lack of skin penetration by the needle, shallow  penetration of the needle, insertion at points that are not  traditional acupuncture points, or not achieving “deqi”  which is an expected needling response (subjective sen sation of local warmth, paresthesia tenderness) that is  considered an integral element of the healing process.  The 2016 Cochrane review by Linde is one of the most  recent and comprehensive investigations that concluded  that acupuncture is more effective than no prophylactic  treatment, more effective than sham acupuncture, and  as effective as pharmaceutical intervention in reducing  the frequency of headaches in patients with episodic  migraine.4 The authors found that acupuncture is at  least as effective as, or possibly more effective than, prophylactic drug treatment, and has fewer adverse effects.  Since 2016, several new studies appear to support the  effectiveness of acupuncture in migraine prevention.  This article is a focused, systematic review of high quality, randomized controlled trials that study the effective ness of acupuncture for the treatment of episodic, or  chronic migraine compared to the effectiveness of stan dard pharmacological therapies for the treatment of  migraine. In addition, we examined and summarized  point selection, treatment period, total number of treat ment sessions, and the frequency of treatments. 

METHODS 

Inclusion Criteria.—We included randomized con trolled trials in adults that were published in English.  The last search was conducted on May 15, 2019. Subjects  had a diagnosis of chronic or episodic migraine with or  without aura. Studies focusing on migraine but includ ed patients with other types of headache were included.  Treatments involved needle insertion at acupuncture  points, pain points, or trigger points. Studies involving  laser stimulation and transcutaneous electrical stimula tion (electroacupuncture) at acupuncture points were not  excluded. In electroacupuncture, a small electric current  is passed between pairs of acupuncture needles and cre ates a stronger stimulation than using the needle alone. In  clinical practice, licensed acupuncturists typically offer both the standard acupuncture and electroacupuncture  to patients. Some trial control interventions included no  treatment or routine care, sham intervention (either no  skin penetration or insertion at points that are not traditional acupuncture points), and other treatments (drugs).  The primary outcomes differed across studies; the main  outcome for the majority of studies was reduction of  migraine days per month. Custom search strategies (see  Appendix) were written by 2 Stanford University Information Services Librarians for Pubmed (includes MED LINE) and Scopus (includes EMBASE). The present  review was not registered. 

DATA COLLECTION AND ANALYSIS Selection of Studies.—One reviewer screened all  abstracts identified by the search strategy. Custom  search strategies yielded a total of 706 articles. A  total of 158 duplicates removed. Based on abstract/ title, an additional 318 irrelevant items were removed.  A total of 230 articles were available for full text  screening. One reviewer extracted seven articles for  review (Fig. 1). Clearly irrelevant articles and dupli cates were excluded. Full texts of relevant articles  were reviewed by 2 reviewers for eligibility. Orig inal authors did not need to be contacted. Data  extraction included review of headache classifications  used, number of randomized patients, reporting of  acupuncture protocol in accordance with STRICTA  (Standards for Reporting Interventions in Clinical  Trials of Acupuncture),5 duration of treatment, and  follow-up periods, number of drop outs, treatment  characteristics (total number of sessions, frequency,  duration of each session, point selection). Extract ed data were summarized in narrative form by each  reviewer and then compared. Disagreements were  resolved by discussion.

RESULTS 

Custom search strategies yielded a total of 706  articles. A total of 158 duplicates removed. Based on abstract/title, an additional 318 irrelevant items were  removed. A total of 230 articles were available for full  text screening. Seven articles were extracted for review  (Fig. 1).

We set out to aggregate these data, but methodological heterogeneity precluded us from doing so.  Specifically, the comparison groups differed drasti cally among the studies. Several studies used an acu puncture sham, whereas most of the studies used an  active comparator; therefore, we could not estimate a  meaningful relative improvement of the treatment vs  the comparator. In addition, the primary outcomes dif fered substantially among studies (disability, headache  days, and percent responders). Other design elements  differed substantially as well. For example, the follow up periods differed, the use of double dummy designs.  Because of the aforementioned differences, a systematic review without meta-analysis is warranted. The following is a summary of the evidence for  the efficacy of acupuncture for migraine prevention  and treatment. General information regarding safety  and tolerability is included. Main outcome measures include reduction of headache days, disability, percent  responders (Table 1). 

Facco et al6.—A total of 82 out of 100 patients com pleted this randomized controlled trial that studied the  effects after a total of 20 sessions of acupuncture vs  valproic acid 600 mg/day, over 10 weeks. The primary  end point was the Midas Index (MI). The study fol lowed STRICTA. At 3 months, MI improved in both  groups but pain intensity was better in the valproic acid  group. At 6 months, pain intensity, Pain Relief score  (PRS), and rizatriptan intake were better in the acu puncture group. The rate of adverse events was 47.8%  in the valproic acid group and 0% in the acupunc ture group. 

Wang et al7.—In this multicenter trial, treatment  with true acupuncture (12 total sessions over 4 weeks)  plus placebo was compared to treatment with sham  acupuncture plus flunarizine. Sham acupuncture protocol involved using 5 acupoints near the elbow and knee  that were defined as unrelated to headache (Table 2).  A total of 140 patients with more than 2 migraine attacks within 4 weeks were recruited for the study; 120  patients completed the study. The primary end point  was the proportion of responders with at least a 50%  reduction in migraine days. At 3 months, the patients  in the acupuncture group had better responder rates  (a reduction of migraine days by at least 50%) and  fewer migraine days compared with the control group  (P < .05). 

Streng et al8.—There was failure to meet recruitment  target because of high drop out in the metoprolol group.  Only exploratory analyses could be performed, no con firmatory interpretation, and the results of the trial  must be interpreted with caution. Acupuncture (8-15  sessions over 12 weeks) was similarly effective and bet ter tolerated than metoprolol. The improvements per sisted during follow-up in both groups, with a tendency  to favor acupuncture. The population was somewhat  biased given that 50% of participants had received acu puncture treatment for migraine before and 95% of the  patients in the acupuncture group expected a definite  improvement. Only a minority of patients received a  daily dose of metoprolol exceeding 100 mg. 

Diener et al9.—This was a multicenter trial that  compared treatment with true acupuncture vs sham  acupuncture vs standard therapy (metoprolol, flu narizine, or valproic acid) in patients who had 2-6  migraine attacks per month. Patients underwent a  total of 10 true acupuncture or sham acupuncture  sessions over a period of 6 weeks. Sham acupuncture  was performed using points that were not traditional  acupuncture points. Primary outcome showed a mean  reduction of 2.3 headache days (95% CI 1.9-2.7) in the  true acupuncture group, 1.5 days (1.1-2.0) in the sham  acupuncture group, and 2.1 days (1.5-2.7) in the standard therapy group. These differences were statistically  significant compared with the baseline (P  <  .0001),  but not across the treatment groups (P  =  .09). The  authors concluded that the treatment outcomes for  migraine do not differ between patients treated with  sham acupuncture, true acupuncture, or standard therapy. All treatments were effective. After 26 weeks, acu puncture treatment over 6 weeks had a similar efficacy  compared with 24 weeks of continuous treatment with  standard drug therapy. 

Allais et al10.—This single center trial showed that  acupuncture in episodic migraine patients is more effective than flunarizine in reducing the frequency and  intensity of attacks after 2 and 4 months of treatment  (T1a 2.95 ± 0.39 vs T1f 4.10 ± 0.42, 95% CI, 0.02-2.28.  T2a 2.30 ± 0.20 vs T2f 2.93 ± 0.24 95% CI, 0.02-1.24).  But there was no difference in frequency of attacks at  6 months (T3a 2.05 ± 0.22 vs T3f 2.32 ± 0.27; 95% CI,  ?0.41-0.95). A total of 8 sessions of acupuncture were  performed over 8 weeks and then 4 sessions were per formed over 16 weeks. Acupuncture reduced progressively and significantly the amount of analgesics taken  for migraine relief and it showed greater effectiveness  than flunarizine after the first 2 months of treatment.  A good clinical result was maintained by acupuncture  in the last 4 months of therapy with a very low frequency  of therapeutic sessions (1 session per month). There  were significantly lower number of side effects in the  acupuncture group (sedation after treatment 10%,  local pain 8%) than the flunarizine group (drowsiness  35%, weight gain 22%, depression 7%). 

Yang et al11.—Conformance to STRICTA was fol lowed. This was a single-center study showing that the  mean number of headache days decreased by 10.7 days  in 1 month for chronic migraine patients in the acu puncture group (24 sessions over 12 weeks) and by 7.9  days per month for patients in the topiramate group  (mean dose was 84.0 mg/day). These significant differ ences were also true for patients who were overusing  acute headache medication. Acupuncture is similarly  effective or more effective than prophylactic drug treat ment with less side effects in migraine prophylaxis. Side  effects were reported in 6% of the acupuncture group.  The side effects identified in the study included local  pain, ecchymosis, and local paresthesia during the ses sion (Of note, local paresthesia is typically not consid ered an adverse event in acupuncture treatment because  is usually considered part of “deqi” which describes the  expected, subjective sensations experienced by patients  during acupuncture.) Adverse side effects were reported  by 66% of the patients in the topiramate group (pares thesia 48.4%, difficulty with memory 36.3%, dyspep sia 36.3%, fatigue 24.2%, dizziness 21.2%, somnolence  18.1%, nausea 12.1%). Three patients (9%) withdrew  because of intolerable adverse side effects. 

Naderinabi et al12.—In this single-center, random ized controlled study, 162 patients with the diagno sis of chronic migraine, were randomly allocated to 3 groups: (1) acupuncture (30 sessions over 60 days),  (2) botulinum toxin-A group receiving PREEMPT  155 units (3) control group on sodium valproate  500  mg/day. A total of 150 patients completed the  study while 12 patients dropped out because of low  compliance with the survey but not affected by  severe adverse effects. Reduction of pain intensity by  VAS was the primary outcome. During the 3-month  study, the pain intensity significantly decreased  in all 3 groups (P  =  .0001) with greatest reduction  in the acupuncture group (P  =  .0001). The rate of  side effects was significantly lower in the acupunc ture group. 

Summary of Trials.—We reviewed a total of 7 ran domized controlled trials, 2 of which studied patients  with chronic migraine, whereas the other 5 studies  recruited patients with episodic migraine. Rates of  adverse events were consistently higher in groups using  standard pharmacological interventions. Allais et al  found that acupuncture is more effective in reducing  number of attacks in first 4 months but no difference  at 6 months. Wang et al. reported that the acupuncture  plus placebo groups had more people who had at least  a 50% reduction in headache days compared to the  group who had sham acupuncture and flunarizine  (P < .05). Yang et al found that the acupuncture group  had a greater decrease in headache days per month  than the topiramate group. Diener et al did not find a  mean reduction of headache days that was statistically  significant across study groups which included acu puncture, beta blocker, valproic acid, and flunarizine.  Streng et al was only able to perform exploratory analysis because of the high dropout rate in the metoprolol  group. Facco et al found that in both the valproic acid  group and the acupuncture group, the MI improved at  3 and 6 months (P < .0001); pain intensity was better  at 3 months for the valproic acid group (P < .0001), but  pain intensity and PRS were better in the acupuncture  group at 6 months (P = .02). Finally, Naderinabi et al  showed that patients in the acupuncture group had the  greatest reduction in pain severity when compared to  patients receiving botulinum toxin A and patients on  sodium valproate. 

In terms of acupuncture treatment protocol, treat ment period ranged from 4 to 24 weeks. The total num ber of acupuncture treatment sessions ranged from 8 to 24 sessions. The average number of treatments  per week ranged from 0.5 to 3.5 sessions per week.  Duration of each session mostly ranged from 20 to  30 minutes. 

DISCUSSION 

An analysis from the American Migraine  Prevalence and Prevention (AMPP) Study showed that  the vast majority of episodic migraine patients were  not able to obtain minimal pharmacologic management.13 There is growing evidence that acupuncture is  just as effective and has fewer side effects than many  of the standard pharmaceutical agents that are cur rently used. In clinical practice, patients frequently  ask for guidance on whether or not to try acupuncture  and about how frequent and how long they need to  try it for. However, we still have very little understand ing of the mechanism by which acupuncture works  and the data are still limited. Since the publication of  the systematic review in 2016 by Linde et al, several  new studies have shown that acupuncture is effective  in the prevention of migraine when compared to the  standard pharmacological treatment.4 We conducted a  focused review of existing randomized controlled tri als to study the effectiveness of acupuncture treatment.  We also examined the duration and frequency of treat ments. Our objective was to use the available data to  formulate a practical summary of results for clinicians  and their patients. 

The heterogeneity and variability of the study pro tocols bring up the larger question of how the study  of acupuncture can be standardized. According to tra ditional Chinese medicine, migraine is a complex dis ease and a variety of syndrome diagnoses are possible.8 Traditionally, treatment is individualized and is not a  uniform intervention that can be standardized easily.4,8 Acupoint selection is extremely variable, as can be seen  even in the small collection of studies that we exam ined. Of these 7 studies, 4 used “part standardization”  which involves the use of a set of obligatory points and  additional points. The additional points are chosen  depending on different syndromes present in the indi vidual patient. Therefore, the absence of standardized  approaches to treatment makes it difficult to reproduce  studies. Future studies are needed to confirm whether  there is a real difference between individualized and standardized acupuncture. The most recent revision  of STRICTA (Standard for Reporting Interventions  in Clinical Trials of Acupuncture) is intended as a  guideline to improve reporting of acupuncture trials  and facilitate their interpretation and replication. Of  the studies we reviewed, only the studies by Yang et al  and Facco et al explicitly mention conformance to  STRICTA. 

Both the studies from Wang et al and Diener et al  employed the use of sham acupuncture. Numerous  studies have shown that sham acupuncture is just  as effective as true acupuncture for treatment of  migraine. In the 2009 review by Linde et al, pooled  analyses of 14 studies did not show a statistically significant superiority for true acupuncture compared  with sham acupuncture in migraine prophylaxis. But  any intervention involving skin penetration cannot  be considered an inert placebo.8,14 Sham acupuncture  may still induce a wide range of peripheral, segmental,  and central physiological responses to an unpredictable degree.11 Also, acupuncture-specific non-needle  and/or needle components may be retained in sham  treatments.15 Given these reasons, it may be more  practical, in future studies, to compare acupuncture  with standard therapies rather than comparing it to  sham acupuncture. 

The tolerability and safety of a treatment are key  factors in maximizing compliance. Several large sur veys have shown that acupuncture is well tolerated and  that serious complications are rare events.8 In several  of the studies that we reviewed, a greater number of  patients in the standard therapy group had to withdraw  because of adverse effects. Reported side effects of acu puncture included mild bleeding at site of needle inser tion, ecchymosis, local pain, sedation, and fatigue. In  all of the studies, acupuncture consistently had a lower  rate of side effects compared to beta blockers, valproic  acid, topiramate, and flunarizine.