Editorials

Migraine prevention

The choices continue to grow

Papers BMJ USA p 91

This article originally appeared in BMJ USA

In the last 10 years, the management of migraine headache has changed dramatically. The addition of the triptans has been a breakthrough in the treatment of acute migraine headache. The effectiveness of these medications in relieving the severe pain, disability, and other migraine-associated symptoms has reduced the proportion of patients who require prophylaxis for episodic migraine. Even frequently occurring migraines can be effectively treated with acute medications alone.

Clinicians and patients are now faced with the difficult decision of when to start prophylactic drugs. It is estimated that preventive therapies are used by only 3% to 5% of migraineurs. ^{1 2} Circumstances indicating the need for preventive treatment include: a) two or more disabling headaches per month; b) ineffective symptomatic treatment; c) use of abortive medication more than twice per week; and d) migraines with potential neurological sequelae.³

The mechanisms by which prophylactic drugs reduce migraine frequency and severity remain unknown.⁴ Many agents appear to have different pharmacological properties and multiple actions on vascular structures and in the nervous system. Likewise, the relative importance of each action is unknown. Therefore, the development of preventive medications was based initially on clinical observations and then supported by clinical trials. All except methysergide were developed to treat other medical conditions. Their anti-migraine effect was discovered serendipitously.

Currently, there are six established classes of agents used to prevent migraine headache.³ These first-line therapies, which show the greatest clinical efficacy with the fewest adverse effects, are tricyclic antidepressants, -blockers, calcium channel blockers, anti-inflammatory drugs, anti-serotonin agents (methysergide), and anticonvulsants (divalproex sodium). Second-line therapies that are emerging as new, but unproven, medications include the antidepressant venlafaxine, the anticonvulsant drugs gabapentin and topiramate, high-dose riboflavin and magnesium supplementation, and the herbs feverfew and Petasites hybridus rhizoma.⁵

This issue of BMJ USA includes a paper by Schrader et al⁶ (BMJ USA p 91), which reports the results of a randomized, double-blind, placebo-controlled, crossover trial of lisinopril in the prophylactic treatment of migraine. This represents a new preventive drug class. This study followed recent methodology outlined for clinical trials of migraine prophylaxis.⁷ The primary outcome measures were hours with headache, number of days with headache, and number of days with migraine. Results showed a reduction by 20% (95% confidence interval, 5% to 36%), 17% (5% to 30%), and 21% (9% to 34%), respectively. A small subset of patients (14/47, 30%) experienced a greater than 50% reduction in the number of migraine days. Unfortunately, the large confidence intervals make the precise anti-migraine effect unknown and detract from the power of the study. These results are similar to those of other well-designed migraine clinical trials that show a small anti-migraine effect for the entire patient group but a substantial effect for subgroups.

As reviewed by Ramadan et al,⁸ the three most commonly used drugspropranolol, amitriptyline, and verapamilhave not been shown irrefutably to prevent migraine. In this review, most of the randomized trials of preventive anti-migraine drugs lacked scientific robustness. Furthermore, for entire patient groups, the prophylactic effect was not dramatic and did not exceed 50% over placebo. However, individual patients were recognized to have had dramatic responses to various medications. In clinical practice, this is typically what is observed. This finding is not unexpected given the heterogeneity of the migraine population. The best scientific evidence for migraine prevention for a patient group exists for divalproex sodium. Recent well-designed clinical trials have found divalproex sodium efficacious for the prophylaxis of episodic, transformed, and pediatric migraines.

Currently, it is impossible to predict which patient will respond to a particular medication. In clinical practice, decisions are typically based on patient characteristics. Ideally, decisions should be made on scientifically documented efficacy, past treatment response, contraindications, side effects and tolerability, patient preference, cost, and comorbid disease. Silberstein³ reviewed the relative indications and contraindications for some comorbid conditions. Frequently, a good decision point for choosing a prophylactic medication is comorbidity, associated symptoms, and patient preference. For example, in patients with associated mild depression and insomnia, a tricyclic antidepressant is an excellent choice. Similarly, in patients with a history of migraine variability coinciding with major depression, venlafaxine might be indicated. In young male athletes with a tendency toward asthma or erectile dysfunction and borderline hypertension, verapamil or lisinopril probably would be superior to a -blocker. Lastly, for women considering pregnancy (a contraindication to almost all preventive medications, except possibly magnesium supplementation), a more aggressive preventive program could be designed to focus on lifestyle change, trigger and stress management, and an intraoral splint for associated jaw pain and nocturnal clinching. Given the individual variability in drug response and anecdotal evidence of dramatic responses to different drugs, it is reasonable to follow a poor response to one drug with a trial of an alternate drug.

In summary, lisinopril represents a new class of drug in the widening array of prophylactic choices. The effectiveness of any single medication probably depends mostly on patient characteristics, patient preferences, and other illnesses. No single agent or class is better for all patients.

Alfred L Clavel Jr. 

Department of Neurology, Hennepin County Medical Center, Minneapolis, MN, USA (clave003{at}tc.umn.edu)

Footnotes

Papers (BMJ USA p 91)