Migraine occurs in three in 20 adults and three in every four sufferers are female. People who have 15 or more days of headache in a month, with eight or more of those days being migraine, have chronic migraine. People with fewer than 15 days of headache in a month have episodic migraine. We included trials that compared botulinum toxin treatment with placebo injections of salt water, different doses of botulinum toxin, or other oral treatments to prevent migraine. We collected information for the following outcomes: number of migraine days in a month (our preferred measure); migraine severity; use of medications for migraine symptoms; disease-rating scales; quality-of-life scales; side effects; and cost effectiveness of treatment.
Trial characteristics
We found 28 clinical trials involving 4190 participants. Their average age was 42 years and eight in 10 were female. It is likely that we found all relevant trials published before December 2017. Trials were short, the longest lasting nine months. Around half the participants had chronic migraine symptoms and half episodic. Trial doses ranged from 6 to 300 units. The dose recommended for chronic migraine in the UK and USA is 155-195 units. Sixteen trials, involving 8 in 10 participants, were funded by botulinum toxin manufacturers.
Key results
Disappointingly, there was not enough detail in the trial reports about many important measures of disease for us to study them.
People with chronic migraine treated with the recommended dose of botulinum toxin had two fewer migraine days in a month than people treated with placebo. Six trials in both chronic and episodic migraine also reported the number of migraine attacks per month. Botulinum toxin was not proven to be better than placebo at reducing the number of attacks suffered per month. Botulinum toxin may reduce the severity of migraines but we need larger trials to have confidence in this result.
Three trials also compared botulinum toxin (at least 100 units) with oral treatments (sodium valproate and topiramate). There was no difference in the improvement in number of days with migraine; these data came from one trial. Botulinum toxin was no better or worse than oral treatments at reducing the scores on a migraine disability questionnaire (Migraine Disability Assessment) for people with chronic migraine. As all the results for comparison with oral treatments came from a few small trials it is likely that further large trials would change these results and so we cannot be confident in them.
Of the participants treated with botulinum toxin, 60 in 100 reported side effects (most common was drooping eyelid or muscle weakness), which was a little higher than the number receiving placebo (47 in 100). No difference was seen in the risk of side effects between botulinum toxin and oral treatments. Participants from two small trials were nearly four times less likely to stop their treatment if they were given botulinum toxin than if they had oral treatments. Information about side effects was reported for 8 in 10 trial participants.
Quality of the evidence
We rated the quality of the evidence from trials using four levels: very low, low, moderate or high. Very low-quality evidence means that we are very uncertain about the results. High-quality evidence means that we are very confident in the results. The results for the change in migraine days for people with chronic migraine and the number of side effects experienced were based on moderate-quality evidence. All other results discussed in this summary were low or very low-quality evidence, so the true effect is likely to be different to these results.
Authors' conclusions:
In chronic migraine, botulinum toxin type A may reduce the number of migraine days per month by 2 days compared with placebo treatment. Non-serious adverse events were
probably experienced by 60/100 participants in the treated group
compared with 47/100 in the placebo group. For people with episodic migraine, we
remain uncertain whether or not this treatment is effective because the
quality of this limited evidence is very low. Better reporting of outcome measures in published trials would provide a more complete evidence base on which to draw conclusions.