Prophylaxis of migraine headache

The case

A 40-year-old woman presented with an eight-year history of migraine headaches. The headaches occurred once or twice per week, but usually responded well to a triptan taken soon after onset. Sometimes the headache returned the next day, necessitating a second dose. Occasionally the headaches did not respond well to the triptan; they caused the patient to miss work about once per month. The patient’s headache diary shows that she takes a triptan nine days per month. Her physician suggests a prophylactic medication, given that she is at risk for medication overuse headache and experiencing substantial disability despite taking the medication. The patient’s medical history includes asthma that is well controlled. She generally sleeps well and has no history of clinically significant mood disturbance. Physical examination is normal except for obesity. Which prophylactic medication should be tried first?

Migraine headache is a common, disabling condition. The diagnostic criteria for migraine headache 1 according to the International Headache Society are shown in Box 1. When migraine episodes are frequent, treatment can be challenging. Prophylactic therapy for migraine remains one of the more difficult aspects. Although valid randomized controlled trials exist to aid decision-making, all of the medications used in treatment have incomplete efficacy, and most produce adverse effects.

• Download figure
• Open in new tab
• Download powerpoint

Box 1: International Headache Society criteria for migraine headache 1

In this review, we discuss when to consider prophylaxis for the patient with migraine, and provide a systematic review of the evidence available from randomized controlled trials on prophylactic agents. Existing guidelines on this topic are out of date 2 or do not include a systematic review of the literature. 3

Medications used to treat this condition can be divided into two broad categories: symptomatic or acute-care medications to treat individual migraine episodes, and prophylactic or preventive medications, which are used to reduce headache frequency.

Symptomatic migraine therapy alone, although helpful for many patients, is not adequate treatment for all. Patients who have frequent episodes of migraine may sustain substantial disability despite appropriate symptomatic therapy. When symptomatic medications are used too frequently, they can result in increased frequency of headache and medication overuse headache. 1^,4 The Canadian Headache Outpatient Registry and Database Project found that 21% of patients with a diagnosis of migraine who were referred to headache specialists had symptomatic medication overuse. 5

When prophylaxis should be started is a matter of clinical judgement. Evidence on which to base this decision is lacking, and many medications are available from which to choose. Prophylactic medications are no substitute for attention to patient lifestyle and avoidance of migraine triggers. All patients for whom migraine prophylactic agents are considered should be educated in migraine triggers and lifestyle-related factors. Common headache triggers include caffeine withdrawal, alcohol, sunlight, menstruation and changes in barometric pressure. Lifestyle factors, including stress, meal-skipping, obesity, and erratic sleep and work schedules, can precipitate migraine.

When should prophylaxis be considered?

Prophylactic therapy should be considered for patients whose migraine episodes have a substantial impact on their lives (i.e., causing them to miss workdays and family activities) despite appropriate use of symptomatic medications, or whose frequency of episodes is such that reliance on acute medications alone puts them at risk for medication overuse headache. 3 According to the criteria of the International Headache Society, 1 medication overuse headache is defined as headache present on more than 15 days per month, which has developed or markedly worsened during medication overuse. The headache associated with medication overuse is variable, with features of both migraine and tension-type headache (i.e., bilateral location, pain that ranges from mild to severe, and associated nausea and light sensitivity). Medication overuse occurs when acute treatments are used frequently and regularly, i.e., 15 days or more per month for simple analgesics and 10 days or more per month for triptans, opioids, and combination analgesics.

Some guidelines have made consensus-based recommendations that prophylactic treatment should be considered when a patient has three or more severe migraine episodes per month that fail to respond adequately to symptomatic treatment with medication. 6 Although episode frequency is helpful in determining the need for prophylactic therapy, the decision to discuss prophylaxis with the patient should be individualized, and all aspects of the patient’s migraine syndrome, including the risk of symptomatic medication overuse, need to be considered. According to estimates, about 25% of all migraine sufferers should be offered prophylactic therapy, 7 and prophylactic medications are thought to be underutilized. 5

When migraine prophylactic therapy is started, one of three outcomes can be anticipated. The patient may show substantial benefit, which is usually defined as a reduction in frequency of headaches of 50% or more. The frequency can be assessed by having the patient keep a diary of the headaches. Although the primary effect of prophylactic medications is a reduction in frequency, some patients report that they also experience reduced intensity of their headaches and improved response to symptomatic medication while taking a prophylactic medication. 8

The second possibility is that the patient may develop intolerable adverse effects (i.e., somnolence, nausea, cognitive slowing or weight gain), that may necessitate stopping therapy. The third possible outcome is that the medication may show insufficient efficacy. Most prophylactic medications show substantial efficacy within one month after initiation of treatment, and their therapeutic effect may continue to increase for several months after that point. 9^–11 If the patient shows no benefit after two months of therapy at the target dose, prophylactic treatment should be stopped and, if indicated, another medication tried.

Overuse of symptomatic medications for headache is considered by headache specialists to reduce the effectiveness of migraine prophylaxis. Avoidance of medication overuse is recommended to improve the likelihood of success when starting prophylactic therapy. 1^,6 Several clinical trials suggest that one particular prophylactic medication, topiramate, can be effective in reducing the frequency of migraine in the presence of medication overuse, and results in a reduction in use of symptomatic medication. 12^,13

Two and a half per cent of migraine patients per year progress to a very frequent headache pattern, either as a result of medication overuse, or because of an apparent progression to chronic migraine. 14 Chronic migraine is defined as migraine that occurs on 15 or more days per month for more than three months. Patients with an intermediate headache frequency of six to nine days per month are at greater risk for further progression to chronic migraine. The risk is even greater in patients who have headaches on 10 to 14 days per month. 15 Reduction in frequency of migraine through use of prophylactic therapy may prevent or delay progression to chronic migraine, although this effect is not proven. 16

Methods

We performed a systematic review of prospective double-blind, randomized controlled trials of medications for prevention of migraine episodes. Study participants in the trials were eligible if they were adults and met International Headache Society 1 or Ad Hoc 17 criteria for the diagnosis of migraine headache. We excluded trials of patients with chronic daily headache, chronic tension-type headache or chronic migraine. Because of the large number of different pharmacologic agents assessed over the past 60 years for the prophylaxis of migraine, we limited our search to agents that are commonly used in clinical practice in North America and Europe.

We searched Ovid MEDLINE (1950 to April 2008) and EMBASE (1980 to April 2008) using a search strategy that combined the subject search with a highly sensitive search string for randomized controlled trials. The subject search string used a combination of controlled vocabulary and free-text terms. In addition, we searched the Cochrane Collaboration Library for systematic reviews of agents used for migraine prophylaxis. We used Cochrane systematic reviews to summarize trial data if similar inclusion criteria and methodology were used in the review. The titles and abstracts of studies identified by the literature search were screened for eligibility by two independent reviewers (TP and WB). Papers that we could not exclude with certainty on the basis of the information contained in the title or abstract were retrieved in full for screening. Papers that passed the initial screening process were retrieved and the full text was reviewed independently by two reviewers (TP and WJD). We performed meta-analyses of selected groups of studies.

Our search yielded 883 abstracts and three Cochrane systematic reviews. After examination of the abstracts and systematic reviews, only 59 studies and one Cochrane systematic review met our inclusion criteria and were included. Most of the studies used the International Headache Society 1 or AdHoc 17 criteria for the diagnosis of migraine. Most participants were women. Pregnant or lactating women were excluded. Most studies specifically excluded people who had chronic daily headache or used medication on more than 10 to 15 days per month. To be included, participants in most of the studies had to have experienced between two and eight migraine episodes per month and not have taken other prophylactic treatment for migraine during the study period. The use of symptomatic therapy for migraine episodes was permitted.

The most consistently reported outcome among the studies was the responder rate, which is defined as the percentage of patients with a 50% or greater reduction in frequency of migraine compared with baseline. The duration of trials varied from 8 to 26 weeks, with most trials using a four-week prospective baseline followed by a 12-week treatment phase.

Pharmacologic prophylactic therapies

The results of our review of studies for quality of evidence, impression of efficacy and frequency of adverse effects of medications for prophylaxis of migraine are summarized in Table 1. We used the criteria of the United States Preventive Task Force to grade the quality of evidence. 76 The grading of the evidence may not be consistent with other guidelines or reviews that use different rating systems. An informational aid to choosing prophylactic therapy is provided in Table 2. Its guidance is based on expert opinion and analysis of the evidence from randomized controlled trials on treatment efficacy and adverse effects.

How should prophylaxis be initiated?

The choice between prophylactic treatments is often based on adverse effects. Medications with the fewest adverse effects included propranolol, lisinopril, candesartan, riboflavin, coenzyme Q10 and butterbur. Such medications should be initiated in patients who have a history of poor tolerance to adverse effects of medication. Clinical experience among headache specialists suggests that the antiepileptics, antihypertensives and antidepressants have greater clinical effect than vitamins, minerals and herbal agents (which generally have limited efficacy but very low adverse-effect profiles). Unfortunately, given the limited number of head-to-head trials comparing different prophylactic regimens, evidence-based statements about the relative efficacy of medications cannot be made. Migraine prophylactic medications should be started at a low dosage and gradually increased over several weeks to the target dosage to avoid adverse effects. Patients should be cautioned that an adequate trial of medication takes eight to twelve weeks, and that more than one agent may need to be tried.

When should prophylaxis be stopped?

Little evidence exists to suggest how long successful migraine prophylaxis should be continued. Recent studies suggest that most patients relapse to some extent after cessation of prophylactic therapy. One study found that 75% of patients developed increased frequency of migraine episodes when successful prophylaxis was stopped. 80 Athough the time to relapse was highly variable from patient to patient, it occurred an average of six months after cessation of prophylaxis.

Another study, which randomized patients to placebo or continued topiramate therapy after 26 weeks of topiramate prophylaxis, found that within one month, patients on placebo had deteriorated substantially in terms of migraine frequency compared with those who continued topiramate therapy. 81 However, even after 26 weeks on placebo, headache frequency had not increased to the frequency at baseline (i.e., before starting topiramate prophylaxis).

These findings suggest that in many patients migraine prophylactics should be stopped at some point to see if they are still needed. Data from placebo-controlled crossover trials of valproate 20 and flunarizine 43 show that, after short-term prophylactic therapy (usually three months), the benefits begin to wane within four weeks of stopping prophylaxis.

Existing guidelines either do not address discontinuation, 2 or they recommend tapering the prophylactic medication after three to six months, if headaches are well controlled, 1 or after “several” months. 6 Given the evidence that the effects of prophylactic medications begin to wane quickly after prophylaxis is stopped, continuing prophylaxis for much longer is prudent for patients with difficult migraine that has caused substantial disability in the past. The ultimate decision depends on the benefit experienced by the patient and whether substantial adverse effects are experienced.

Follow-up every three to six months is important for patients taking prophylaxis, for if the initial benefits of the prophylactic medication eventually disappear, little apparent purpose remains in continuing it. Diary-based documentation of headache frequency can be helpful in making such decisions. Generally, when the decision to stop prophylaxis is made, the dosage of medication is tapered over a two- to four-week period.

When should a patient be referred to a headache specialist?

The generalist should refer a patient whenever questions about the diagnosis, treatment or response exceed the clinical comfort of dealing with uncertainty. In particular, referral is expected when the phenomenology of the headache does not fit a recognized pattern. Treatment-related dilemmas leading to referral may arise when the patient fails to respond to usual therapy, when medication overuse contributes to the headache burden, when regular use of opiates is considered or occurs, or when patients’ comorbidities, intolerance or preferences limit pharmacologic treatment.

Gaps in knowledge

Many aspects of migraine prophylactic therapy are not well understood. In particular, the mechanisms by which established prophylactic medications reduce frequency of migraine episodes are not known, and this lack of information has hampered development of more effective medications. Because few head-to-head medication-comparison trials have been done, little evidence exists to guide the choice of medication based on relative efficacy. For patients with refractory migraine, no evidence exists to guide decision-making in the use of combinations of prophylactic medications.

The case revisited

After discussion, the patient was given topiramate 25 mg to take at bedtime, with a plan to increase the dosage by 25 mg every week, up to a total of 100 mg per day. Prophylaxis was indicated because the patient’s migraines caused her substantial disability, and her use of triptan was approaching levels that put her at risk for medication overuse headache. Treament with a β-blocker was contraindicated by her history of asthma. She had no sleep disturbance or mood disorder to support use of a tricyclic, which might have promoted further weight gain. Topiramate, supported by good evidence of efficacy for migraine prophylaxis, might assist her with weight loss.