Research Review By Dr. Stephen Burnie©

Date Posted:

April 2009

Study Title:

Acupuncture in Migraine Prevention: A randomized sham controlled study with 6-months post-treatment follow-up


Alecrim-Andrade J et al.

Author's Affiliations:

Departments of Neurology and Epidemiology, University of Barcelona and the State University of Campinas in Brazil

Publication Information:

Clinical Journal of Pain 2008; 24(2): 98-105.

Background Information:

The prevalence of migraine is estimated to be approximately 18% in females and 5% in males and acupuncture has become a common treatment choice for sufferers. Until publication of this study, only the benefits of acupuncture for chronic low back pain and postoperative dental pain had high level and generally accepted evidence.

Although several trials have been performed to test the efficacy of acupuncture for migraine control, several methodological problems made it impossible to reach firm conclusions based on these studies. The purpose of this study was to assess the efficacy and effectiveness of acupuncture for migraine prophylaxis.

Pertinent Results:

This study included 37 migraine patients with and without aura, randomly assigned to sham or real acupuncture groups. 36 patients completed the study and were included in statistical analysis. Both groups were homogenous in almost all characteristics at baseline, except that there were a greater percentage of patients with aura in the sham group.

The authors point out that previous research has shown that the presence or absence of aura does not affect the outcome of treatment. Significant decreases in migraine frequency (? 50%) were found in the acupuncture group in the second month of therapy. In the third and last treatment month, migraine attack frequency decreased in both groups, especially the sham group. This meant that although there were significant decreases in migraines with both groups, the significant difference seen between groups in the second month was no longer present.

There continued to be no statistically significant difference between groups at the short- and long-term follow-ups with regards to the primary outcome measure. Secondary outcomes with significant differences between groups were found in the second month of treatment in three parameters: number of days with migraine per month, total duration of migraine in hours per month, and reduction of ? 40% in frequency of migraine attacks. This statistically significant difference between groups disappeared from the third month follow-up to long-term follow-up.

Within-group analysis revealed that the acupuncture maintained significant improvements in all parameters except for headache severity from the first month follow-up to the long term follow-up. Similar within-group results were found for the sham group, however significant improvements were not detected until the second month follow-up with this group.

No serious adverse events were observed in the 576 sessions performed and 8640 needles inserted. The most common adverse event was ecchymosis, reported in 1 of 8 sessions without significant difference between groups. There was no statistical difference between the groups with regards to perception of whether they were in the sham or acupuncture group: 36.8% in the real acupuncture and 11.8% in the sham group correctly identified their treatment.

Clinical Application & Conclusions:

The primary outcome measure chosen to evaluate the effectiveness of acupuncture for migraines was ? 50% reduction in migraine attack frequency. The real acupuncture group started to show benefits within the first month of treatment, with these results becoming significant between groups in the second month of treatment.

When the threshold was decreased to a ? 40% reduction in migraine frequency, a significant benefit of acupuncture vs. the sham group was seen in the first and second months. Although the sham group only showed slight improvement in the first 2 months of treatment (5.9% and 11.8%, respectively), at the third month of treatment the sham group had a very high improvement. At this point, the significant difference between groups disappeared in all outcome measures, with the 63% of the acupuncture group and 47% of the sham group reaching ? 50% reduction in migraine frequency.

The authors speculate that this large improvement in the sham group may be “due to chance”, however in such a well-controlled study this is an unlikely explanation. A more likely explanation is that the sham group actually experienced a treatment benefit, as they were given minimal acupuncture consisting of superficial insertion that may have had some therapeutic benefit.

It is therefore impossible to tell whether both groups achieved significant improvements in migraine because of acupuncture benefits or due to natural history. As improvements due to natural history do not generally come into play for chronic migraine sufferers over a one year period, this leaves the possibilities that improvements were due to placebo effect or true acupuncture benefit in both groups.

Although this paper suggests a benefit of acupuncture for migraines, without a no-treatment control group (no acupuncture or detuned ultrasound, for example) it is impossible to know the true benefit of acupuncture for these migraine sufferers.

In conclusion, this study provides further justification for the use of acupuncture in chronic migraine sufferers. While it is impossible to elucidate the extent of benefit acupuncture patients experienced compared to placebo effect due to the study design, the dramatic improvements reported in this study provide justification for doing a trial of acupuncture in chronic sufferers.

Study Methods:

Subjects included in this study were 37 Brazilian residents recruited via advertising with migraine (with or without aura) for a period of at least 1 year. Major inclusion criteria were:
  • age between 18 and 50
  • 2 to 6 migraine attacks per month
  • avoidance of prophylactic migraine medications or acupuncture for at least the past 3 months
Randomization and blinding of patients and outcome assessors into 2 groups was performed very well by using opaque numbered and sealed envelopes with block randomization; each envelope contained either the letter “C” or “D” to assign groups, with only the treating acupuncturist knowing the meaning of group assignment.

Patients were randomized into either a sham or real acupuncture group and each patient attended 16 treatment sessions. The medical acupuncturist was instructed to adopt a uniform, neutral attitude toward patients to preserve blinding, and patients were asked to report at the end of the study whether they believed that they received real or sham acupuncture.

The true acupuncture group received 30 minutes of individualized treatment without moxa or electrical stimulation. Acupuncture points were chosen “based on the principles of TCM” by one of 2 experienced acupuncturists with a maximum of 20 needles inserted and manipulated by rotation to produce a De Qi sensation. Unfortunately, a list of acupuncture points used in this study was not published.

The sham group received minimal acupuncture, consisting of superficial insertion in acupuncture points without any physical manipulation. Between 10 and 15 needles were used per session, and the points were selected to be in areas that were not associated with headache treatment in their literature review.

Outcomes were documented with the use of daily diaries for a 6-month period: baseline (1 month), during the treatment period (3 months), short-term follow-up (first month after the last treatment), and long-term follow-up (sixth month after the last treatment).

Details included in the diary were migraine intensity at set points during the day, medication intake (type and doses), vomiting, nausea, and menstruation. The primary outcome was percentage of patients with ? 50% reduction in migraine attack frequency each month compared with the baseline period. Severity was evaluated on a 4-point scale as recommended by the International Headache Society (0-no headache, 1-mild headache, 2-moderate headache that interferes with daily activities, and 3-severe headache making normal daily activities impossible).

Secondary outcome parameters were percentage of patients with ? 40% reduction in migraine attack frequency, number of attacks per month, number of days with migraine per month, total duration of migraines pain per month, mean headache severity in each attack, amount and type of rescue medication, and nausea/vomiting frequency. Within and between group comparisons were made to analyze the data.

Study Strengths / Weaknesses:

This study was very well randomized and blinded compared to most research in the manual therapy field, as all parties were unaware of treatment grouping except for the treatment provider. The percentage of patients able to perceive the group to which they were assigned was not statistically significant, ensuring that patients remained blinded for the duration of the study.

The low drop out rate (only 1 patient) is also quite rare for a study of this duration and adds to the strength of the findings. The use of a sham acupuncture group was very useful, as it potentially allows us to see the true benefit of acupuncture while controlling for natural history and placebo effect. Of course, this only holds true if the sham acupuncture is truly a sham (see next paragraph for further comments on this).

A final advantage of this study design is that the primary outcome measure was an endpoint that is very clinically significant (? 50% reduction in migraine frequency) and not merely a small statistically significant difference that may not have been clinically relevant. Use patient focused outcomes rather than arbitrary improvements that may not be perceptible to the migraine sufferer makes the results more clinically useful.

An obvious weakness of this study is the small sample. A further weakness of this study is that the sham group actually received a degree of acupuncture, as they were given 10 to 15 superficial acupuncture insertions at each treatment.

Although the needles were not inserted deeply, previous studies point to the possibility that even superficial needling at non-acupuncture points may have therapeutic benefit. The “sham” acupuncture in this study demonstrated large improvements in migraine frequency that call into question just how non-therapeutic this “sham” intervention truly was.

The only way to control for this would have been to include a third study group that received no treatment or to use a different form of sham acupuncture that does not involve skin penetration. A second weakness is that each patient received individualized acupuncture points based on the acupuncturists assessment. As such, each patient may have received different acupuncture point treatments.

Without publication of these points or standardization of points to be used, it is difficult to evaluate whether different results would have been found had different acupuncture points been used.

Additional References:

  1. Backer M, et al. Acupuncture in Migraine: investigation of autonomic effects. Clinical Journal of Pain 2008:24(2), 106-115.
  2. Jena S et al. Acupuncture in patients with headache. Cephalalgia 2008; 28: 969-979.