Botulinum Toxin Type A as a Migraine Preventive Treatment


Address all correspondence to Dr. Stephen Jenkins, Allergan, Inc, 2525 Dupont Drive, Irvine, CA 92715.


Objective.—To assess the safety and efficacy of botulinum toxin type A (BOTOX; Allergan, Inc) in the prevention of migraine.

Background.—Current migraine preventive therapies are often unsatisfactory because of their limited efficacy, adverse effects, and drug interactions. Botulinum toxin type A injections often reduce the pain associated with conditions such as cervical dystonia, achalasia, rectal fissures, and myofascial pain syndrome. An open-label, noncontrolled study of botulinum toxin type A suggested benefits for patients with migraine.

Design and Methods.—This was a double-blind, vehicle-controlled study of 123 subjects with a history of two to eight moderate-to-severe migraine attacks per month, with or without aura. Participants were randomized to receive single administrations of vehicle or botulinum toxin type A, 25 U or 75 U, injected into multiple sites of pericranial muscles at the same visit. During a 1-month baseline period and for 3 months following injection, subjects kept daily diaries in which they recorded migraine frequency, migraine severity, and the occurrence of migraine-associated symptoms.

Results.—Compared with vehicle treatment, subjects in the 25-U botulinum toxin type A treatment group showed significantly fewer migraine attacks per month, a reduced maximum severity of migraines, a reduced number of days using acute migraine medications, and reduced incidence of migraine-associated vomiting. Both the 25-U and 75-U botulinum toxin type A groups were significantly better than the vehicle group on subject global assessment. Botulinum toxin A treatment was well tolerated, with only the 75-U treatment group exhibiting a significantly higher rate of treatment-related adverse events than vehicle.

Conclusions.—Pericranial injection of botulinum toxin type A, 25 U, was found to be a safe treatment that significantly reduced migraine frequency, migraine severity, acute medication usage, and associated vomiting.


botulinum toxin type A

Botulinum toxin type A (BTX-A) has been used clinically for a number of disorders believed to be due to overactive striated or smooth muscles, such as cervical dystonia, 1 blepharospasm, 2 spasticity, 3,4 rectal sphincter spasm with fissure, 5 and achalasia. 6 Benefits have also been reported for hyperhidrosis, 7 showing that BTX-A can inhibit overactive nonmotor as well as motor peripheral acetylcholine neurons. In addition to reducing muscle hyperactivity and spasm, BTX-A treatment often reduces the pain associated with cervical dystonia, achalasia, and rectal fissures. Preliminary evidence suggests that it may also be beneficial in the treatment of chronic low back pain associated with muscle spasm. 8

A recent open-label, noncontrolled study reported that treatment of patients with BTX-A markedly reduced migraine frequency and symptoms. 9 Our study attempted to confirm these findings by examining the safety and efficacy of BTX-A in the prevention of migraine, using a double-blind, vehicle-controlled design.


Study Protocol and Treatment.

This was a double-blind, randomized, parallel-group study of subjects recruited from 12 headache centers across the United States. The protocol required a 1-month baseline, an injection visit, three monthly postinjection visits, and completion of a daily headache diary. Eligible subjects were randomized to one of three groups: vehicle or BTX-A, 25 U or 75 U. At the injection visit, qualified subjects received symmetrical injections into the frontalis, temporalis, and glabellar (corrugator and procerus) muscles. Figure 1 illustrates the injection sites and units injected per muscle.

Figure 1.—.

Injection sites. Subjects received injections (10 U or 30 U) into four frontalis sites, two temporalis sites (6 U or 18 U), and five glabellar sites (corrugator: four sites, 6 U or 18 U; procerus: one site, 3 U or 9 U). The lower doses were administered to the subjects receiving a total dose of 25 U BTX-A and the higher doses were administered to the subjects receiving a total dose of 75 U BTX-A.

For 1 month prior to and 3 months following treatment, subjects recorded the following headache parameters in diaries provided by the investigators: the occurrence of migraine and nonmigraine headaches, the start and stop times of migraines, migraine severity, the presence of migraine aura, migraine-associated symptoms, and acute migraine medications or treatments used.

This study was conducted in compliance with institutional review board regulations, informed consent regulations, the Declaration of Helsinki, and the International Headache Society (IHS) guidelines for studies of the prevention of migraine.

Participant Characteristics.

Participants were men or women, aged 18 to 65 years, who had a history of IHS-defined migraine, with or without aura. Subjects were eligible for this study if they had experienced an average of two to eight moderate-to-severe migraines per month over the previous 3 months and if they recorded two to eight such migraines during the 1-month baseline period.

Participants had to have received their first diagnosis of migraine before the age of 50 years and had to be able to distinguish migraine from nonmigraine headaches. In addition, participants were required to have a stable frequency and severity of migraines and no unstable medical conditions. Migraines had to be relieved to an acceptable level by acute migraine therapy, and if subjects were taking concurrent prophylactic medications for migraine, the doses had to have been stable for at least 3 months immediately prior to enrollment.

Key exclusionary criteria included the following: (1) any medical condition or the use of any agent that may have put the participant at risk with exposure to BTX-A (eg, neuromuscular disorders, aminoglycoside antibiotics, curarelike agents), (2) a history of complicated migraine (eg, migrainous infarction, hemiplegic migraine), (3) typical migraine pain localized predominantly to the occipital or suboccipital region of the cranium, (4) planned or actual pregnancy or lactation, (5) known allergy or sensitivity to the study medication or its components, (6) injection of anesthetic or steroid into the muscles to be injected in the month immediately prior to enrollment, (7) more than 15 headache days per month, or (8) symptomatic medication overuse. Participants could voluntarily leave the study at any time. The investigator could elect to discontinue participants at any time for reasons unrelated to the study. Those who did not complete the study for administrative reasons (eg, failure to adhere to the visit schedule) were dismissed from the study.

Outcome Measures

Efficacy Variables.

All efficacy variables were recorded in the subjects' daily headache diaries and were assessed monthly at 0, 30, 60, and 90 days postinjection. The primary efficacy variable was the change from baseline in the frequency (number per month) of moderate-to-severe migraines. Other efficacy variables assessed were the change from baseline in the frequency of migraines of any severity, the percentage of subjects with at least a 50% decrease in the frequency of any migraines, the percentage of subjects with a decrease of two or more migraines per month, maximum migraine severity (on a scale of 0 to 3), days with acute migraine medication use, and percentage of subjects with migraine-associated symptoms. Subject response to treatment was rated on a 9-point Global Assessment Scale (from −4 = very marked worsening to+4 = complete abolition of signs and symptoms) and recorded at monthly visits on a case report form.

Safety Measures.

Any adverse event that a subject reported during the study was recorded by the investigator, graded for severity (mild, moderate, or severe), and assessed for its relationship to study treatment (none, possible, probable, or definite). A serious adverse event was defined as one that was fatal, life threatening, permanently disabling, or required admission to hospital.

Laboratory Parameters.

Standard blood chemistry, electrolyte panel, and hematology analyses were performed once during the 30-day preinjection phase and once at month 3 postinjection.

Statistical Analyses.

All analyses were performed on data from the intent-to-treat population. For variables with ordered response categories and variables with a continuous response range, simultaneous comparisons among all three treatments were carried out with the Kruskal-Wallis test. In addition, comparisons between pairs of treatment groups were done with Wilcoxon rank sum tests. If there were significant baseline differences among treatments, a baseline covariate was included in an analysis of covariance of the variable. The Kruskal-Wallis and Wilcoxon rank sum tests were carried out by analysis of variance (or covariance) of the ranked response variables.

For variables with posttreatment data, the treatment comparisons were done on change from baseline data. For binomial data, comparisons between treatment groups were done with chi-square tests, with 2 × 3 for comparisons of all three groups and 2 × 2 for comparisons of pairs of groups. If the expected value of any cell in a 2 × 2 table was less than 5, the affected distribution was tested with a Fisher 2 × 2 exact test.


Population Characteristics.

Subject Demographics and Disposition.

A total of 123 subjects (105 women, 18 men) participated in this study (vehicle, n = 41; 25 U BTX-A, n = 42; 75 U BTX-A, n = 40). Of these 123 participants, 122 completed the study. Most participants were white (95%). Others included Hispanics (3%), African Americans (1%), and Asian Americans (1%). Their weights ranged from 44 to 132 kg (mean, 71 kg), their heights ranged from 1.5 to 2.0 m (mean, 1.7 m), and their ages ranged from 22 to 63 years. The mean age for each treatment group was as follows: vehicle, 46.8 years; 25 U BTX-A, 42.8 years; and 75 U BTX-A, 42.4 years. The only statistically significant difference in any of the demographics examined was age: participants in the vehicle group had a higher mean age than those in the other two groups (P = .022).

Migraine History.

The mean migraine frequencies per month in the vehicle, 25-U BTX-A, and 75-U BTX-A treatment groups were 4.8, 4.3, and 4.0, respectively, and the mean headache durations were 35.9, 32.9, and 32.2 hours, respectively. There were no statistically significant differences among groups in migraine severity, distribution (unilateral versus bilateral), type of pain, or effect of physical activity.

All but one subject in the study characterized their migraine pain as moderate to severe in intensity, and all participants experienced at least two of the four symptoms typically associated with a migraine attack (ie, nausea, vomiting, photophobia, phonophobia). There was a statistically significant difference among groups in time since onset of migraines (P< .001), with a greater mean time since onset in the vehicle (27.4 years) and 25-U BTX-A (23.4 years) groups than in the 75-U BTX-A group (16.9 years).

Efficacy Variables.

Baseline frequencies of moderate-to-severe migraines were comparable for the groups (vehicle, 4.4; 25 U BTX-A, 4.45; 75 U BTX-A, 3.95; P = .226). There was a significantly greater reduction in the number of moderate-to-severe migraines in the 25-U BTX-A group than in the vehicle group at months 2 (vehicle, −0.37; 25 U BTX-A, −1.57; P = .008) and 3 (vehicle, −0.98; 25 U BTX-A, −1.88; P = .042) postinjection (Figure 2).

Figure 2.—.

Mean decrease from baseline in the number of moderate-to-severe migraines per month. Asterisks indicate that the 25-U BTX-A group was significantly different from the vehicle group at 2 and 3 months postinjection (Pleqslant R: less-than-or-eq, slant.042).

Baseline frequencies of migraines of any severity were significantly lower in the 75-U BTX-A treatment group (4.40) than in the 25-U BTX-A (5.48) or vehicle (5.20) groups (Pleqslant R: less-than-or-eq, slant.046). There was a significantly greater reduction in the number of migraines of any severity in the 25-U BTX-A group than in the vehicle group at month 3 (vehicle, −0.90; 25 U BTX-A, −2.12; P = .014) and a tendency toward a greater reduction of migraines at month 2 (vehicle, −0.37; 25 U BTX-A, −1.55; P = .072).

Significantly more subjects reported at least two fewer migraines of any severity at month 3 in the 25-U BTX-A group than in the vehicle group (P = .011) (Figure 3). Similarly, a significantly greater number of subjects in the 25-U BTX-A than in the vehicle group had a decrease in migraine frequency of at least 50% at month 3 (vehicle, 10 [24%] of 41; 25 U BTX-A, 19 [45%] of 42; P = .046).

Figure 3.—.

Percentage of subjects with at least a decrease of two headaches in the frequency of their migraines. The asterisk indicates that the 25-U BTX-A group was significantly different from the vehicle group at month 3 (P = .046).

At baseline, maximum migraine severity scores (0-to-3 scale), numbers of subjects who experienced migraine-associated vomiting, and days with acute migraine medication were comparable among treatment groups. There was a significantly greater reduction in the mean severity of migraines in the 25-U BTX-A group than in the vehicle group at months 1 and 2 (Pleqslant R: less-than-or-eq, slant.029; Figure 4). Significantly fewer subjects experienced migraine-associated vomiting in the 25-U BTX-A group compared with the vehicle group at month 3 (vehicle, 13 of 41; 25 U BTX-A, 4 of 42; P = .012). There was a significantly greater decrease in the number of days on which subjects used acute migraine medication in the 25-U BTX-A group than in the vehicle group at month 2 (vehicle, −0.76; 25 U BTX-A, −2.45; P = .028). Subject Global Assessment scores were significantly better in both BTX-A treatment groups than in the vehicle group at month 2 (vehicle, 0.46; 25 U BTX-A, 1.19; 75 U BTX-A, 1.25; Pleqslant R: less-than-or-eq, slant.041).

Figure 4.—.

Mean decrease from baseline in the maximum severity of migraines (rated on a 0-to-3 scale). Asterisks indicate that the 25-U BTX-A group was significantly different from the vehicle group at 1 and 2 months postinjection (Pleqslant R: less-than-or-eq, slant.029).

Adverse Events.

All treatments were generally well tolerated and there were no serious treatment-related adverse events. The incidence of treatment-related adverse events in the 25-U BTX-A treatment group was not significantly different from the vehicle group, but there was a higher incidence in the 75-U BTX-A group than the vehicle group (50% versus 24%; P = .017). All treatment-related adverse events were transient and included blepharoptosis (vehicle, 0 cases; 25 U BTX-A, 6 cases; 75 U BTX-A, 7 cases), diplopia (vehicle, 0 cases; 25 U BTX-A, 0 cases; 75 U BTX-A, 2 cases), and injection site weakness—an expected drug effect (vehicle, 1 case; 25 U BTX-A, 4 cases; 75 U BTX-A, 5 cases).

There were no clinically significant changes in any laboratory parameters.


In our study, injection of a total dose of BTX-A, 25 U, into three pericranial muscle groups was significantly superior to vehicle in reducing migraine frequency and severity, the use of acute migraine medication, and migraine-associated vomiting. The beneficial effects of BTX-A were observed mainly at 2 and 3 months posttreatment, as has been reported for other preventive migraine treatments. 10 This is consistent with reports that botulinum toxin requires up to 3 weeks to achieve its maximum effect. 11 Improvement produced by BTX-A continues through month 3, suggesting continued improvement beyond 3 months. The 75-U dose of BTX-A did not perform as well as the 25-U dose in our study, perhaps due to the lower frequency of migraines at baseline in this treatment group. Botulinum toxin type A treatment is safe for migraine: only the 75-U BTX-A group exhibited significantly more treatment-related adverse effects than the vehicle group.

The potential efficacy of BTX-A in the treatment of migraine raises interesting questions regarding the role of muscle contraction in the pathophysiology of migraine 12 and the mechanism of action of botulinum toxin treatment in pain disorders in general. The simplest explanation is as follows: pericranial muscle contractions somehow contribute to the migraine triggering process and BTX-A reduces migraine frequency and severity by reducing these contractions. However, since botulinum toxins are general exocytotic inhibitors, 13 the possibility exists that botulinum toxin inhibits pain pathways through some, as yet unknown, effect on the sensory system.

These preliminary results suggest that treatment with BTX-A may be useful in the prevention of migraine. Further studies are needed to determine optimal dosing and injection sites, optimal patient characteristics, and functional benefits that affect patient quality of life.

Acknowledgments:  Financial support was provided by Allergan, Inc.

The BOTOX® Migraine Clinical Research Group: Sheena Aurora, MD and K.M.A. Welch, MD (Henry Ford Hospital, Detroit, Mich); Roger Cady, MD and Dale Carter, MD (Headache Care Center, Springfield, Mo); Jack Klapper, MD (Colorado Neurology and Headache Center, Denver, Colo); David Kudrow, MD (California Medical Clinic for Headache, Encino, Calif); Elizabeth Loder, MD (Spaulding Rehabilitation Hospital, Boston, Mass); Joseph Nicolas, MD and Nabih Ramadan, MD (Cincinnati Headache Center, Cincinnati, Ohio); Egilius Spierings, MD (Boston Clinical Research Center, Wellesley Hills, Mass); Thomas Ward, MD (Dartmouth-Hitchcock Medical Center, Lebanon, NH); Paul Winner, DO (Premiere Research Institute, West Palm Beach, Fla); and Ron DeGryse, MS, Nina Eadie, MBA, and Eric Chun, BS, CCRA (Allergan, Inc, Irvine, Calif).