Migraine-Preventive Medications: Guidelines for Success

Migraine-Preventive Medications: Guidelines for Success

Many factors need to be taken into account when deciding whether to begin preventive medications. The degree of patient disability is the most important. For some patients, disability may be profound, with as few as 2 headaches per month that fail to respond to acute treatment. It also is important to assess duration of headache because a single migraine attack may last as long as 72 hours.

Disability associated with the patient's prodrome and postdrome should be identified because function frequently is reduced during these periods as well. Patients who have unusual migraine syndromes, such as familial hemiplegic migraine (FHM), basilar migraine, or prolonged aura, are candidates for prevention.

Prevention particularly helps those who are unable to successfully treat individual headaches because of contraindications to acute medications or tolerability issues. Patients who have concomitant medical conditions, such as hypertension, anxiety disorder, depression, bipolar disorder, fibromyalgia, and epilepsy, may benefit from the same medications used for migraine prevention.


It is important that the physician and patient have reasonable goals for treatment and define success appropriately (Table 1). It is not reasonable for the physician to expect or promise freedom from headache. Reduction in overall disability and improvement in quality of life are the primary goals.

The patient should be informed of the most common side effects of the medication in a clear and supportive manner so that when the side effects occur, the patient will not discontinue the medication. The physician and patient should recognize that pharmacologic treatments frequently involve trade-offs, and some side effects may be acceptable givøen the reduction in disability.

An appropriate trial period at an adequate dose is necessary to assess efficacy. Medication and substance overuse must be eliminated to provide the best environment for successful prevention. In addition, identification and control of migraine triggers are essential.


Undertreatment leads to a host of consequences that affect quality of life for the patient and overall economic and health care resources. Many of the following consequences can be avoided by appropriate migraine prevention strategies (Table 2).

Frequent headaches often lead to overuse of acute medications and subsequent failure of acute medications. Migraineurs miss more workdays and social functions than the average person. They also tend to utilize expensive and inconvenient emergency room or urgent care services. Direct and indirect costs attributed to migraine disability are estimated to be in excess of $1 billion and $18 billion annually, respectively.3 Furthermore, because of patient and physician frustration, patients with poorly controlled migraines are at risk for primary treatment with narcotic medications when better options exist.

It is believed that patients with poorly controlled intermittent migraine (less than 15 days per month) are at higher risk for the development of chronic migraine (15 days per month or more).2 One mechanism may be related to medication overuse.2,4-7 Another mechanism may be related to kindling effects, in which frequent activation of pain pathways ultimately leads to lower thresholds to initiate pain events and development of greater pain fields.6,7

Subclinical brain lesions have been identified in migraine patients. A Dutch study compared MRI scans of brains of patients with migraine with MRI scans of age-matched controls to assess the prevalence of subclinical brain lesions. None of the participants had a history of stroke and all had normal neurologic examinations at the time of inclusion in the study.

A statistically higher number of infarcts in the posterior circulation, especially the cerebellum, were evident in patients who had migraine with aura compared with controls. The risk of infarct increased with increased headache frequency (defined as 1 headache per month or more). A higher risk of deep white- matter lesions also was seen in women (but not men) with migraine, and this risk increased with increased headache frequency.8 The results of this study are intriguing and need to be confirmed with additional studies.


Migraine-preventive study trials have improved over time. More rigorous study designs and more meaningful end points have been created so that earlier studies with statistically significant findings may not be reasonably compared with more recent studies. Indeed, one review of preventive trials performed before 1997 concluded that most of those studies lacked appropriate scientific rigor.9

Now, the standard of study is randomized, double-blind, and placebo-controlled (RDBPC). The study should have a population size that is large enough and a study period long enough to perform statistical analysis appropriately and assess efficacy. The definition of migraine should be based on International Headache Society criteria.

All the participants randomized in the study should be included in the statistical analysis (the intent-to-treat population), not just those who complete the study (completer study). Completer studies demonstrate artificially high end points and do not take into account dropouts attributed to side effects or lack of drug efficacy.

The primary end points currently in use are reduction in mean monthly migraine frequency and 50% responder rate. A successful preventive medication will reach or exceed a 50% responder rate, which is an arbitrary measure defined as a 50% reduction in headaches in 50% of patients receiving medication.

Placebo response is usually quite high in both acute migraine and preventive studies (approximately 20% to 30%, or higher). Participants should not be using other medications that may interfere with study outcome. Finally, the best-designed trial will use an active comparator arm, which incorporates a known effective migraine preventive medication, to establish not only the effectiveness of the study drug but also how well it compares with a medication already available.


b-Blockers An evidence-based review of trials by the US Headache Consortium10 supported the efficacy of b-blockers in migraine prevention. Both propranolol and timolol are FDA-approved medica- tions for migraine prevention and received the consortium's highest grade for quality of evidence. The effective dose range for propranolol is 60 to 320 mg/d; for timolol, it is 20 to 60 mg/d. Side effects of b-blockers include fatigue, drowsiness, orthostatic hypotension, depression, sexual dysfunction, and exercise intolerance. Atenolol (50 to 200 mg/d), metoprolol (50 to 200 mg/d), and nadolol (20 to 240 mg/d) also were considered effective but had lower quality-of-evidence and efficacy ratings.

Calcium channel blockers Three double-blind placebo-controlled studies showed a superiority of verapamil over placebo (18% to 49% reduction in migraine frequency).11-13 These studies are considered to have low scientific scores and are limited by low numbers of participants and high dropout rates.9,14 Although more effective products than verapamil exist for migraine headache, it still remains the agent of choice for cluster headache treatment.

Angiotensin II receptor blockers Candesartan (Atacand, AstraZeneca) was evaluated in a RDBPC cross-over study with a 4-week placebo run-in period, followed by two 12-week treatment periods separated by 4 weeks of placebo washout.15 Sixty patients were randomly assigned to receive either candesartan 16 mg/d or matching placebo for 12 weeks, followed by a 4-week washout period, with a crossover to the opposite treatment arm (placebo or candesartan) for another 12 weeks.

Analysis was performed on 57 patients who made up the intent-to-treat population. End points were calculated with a nontraditional format that measured total number of headache days, headache hours, migraine days, and migraine hours over the entire 12-week treatment period.

The primary end point was number of total headache days in 12 weeks. Patients receiving candesartan had a mean 13.6 days compared with 18.5 days for patients receiving placebo. The responder rate was 31.6%, demonstrating a 50% reduction in number of headache days.


Tricyclic antidepressants (TCAs) Despite general consensus regarding efficacy, none of the TCAs are FDA-approved for migraine prevention; however, amitriptyline is the best-studied for this indication. The effective dose range is 25 to 150 mg/d. One study demonstrated a 40% reduction in migraine attacks with amitriptyline.16 Side effects include sedation, weight gain, dry mouth, urinary retention, constipation, and orthostatic hypotension. Less rigorous quality of evidence supports efficacy with other tricyclic and polycyclic antidepressants, including desipramine, doxepin, nortriptyline, and protriptyline.10

Selective serotonin reuptake inhibitors (SSRIs) The SSRI fluoxetine (Prozac, Lilly) was evaluated in 2 RDBPC studies and yielded conflicting results.17,18 In the first trial, fluoxetine 20 to 40 mg/d was evaluated over a 10-week period. Thirty-two participants met inclusion criteria, but only 18 participants completed the study.

Among the "completers," a statistically significant improvement in headache score was seen in the patients receiving fluoxetine compared with the patients receiving placebo. Headache score was calculated based on intensity and duration of headache as well as amount of medication used.

In the second trial, fluoxetine 20 to 40 mg/d was studied for 16 weeks in patients with chronic daily headache or migraine. No significant improvement in the actively treated migraine group was seen, although patients with chronic daily headache appeared to benefit. However, the authors attributed the improvement to the mood-modifying properties of the study drug.

At this time, there is insufficient quality of evidence to recommend the routine use of SSRIs for migraine prevention, including fluoxetine, sertraline (Zoloft, Pfizer), paroxetine (Paxil, GlaxoSmithKline), and escitalopram oxalate (Lexapro, Forest).

Serotonin norepinephrine reuptake inhibitors (SNRIs) The SNRI venlafaxine extended-release (Effexor XR, Wyeth) was studied in 114 patients in an open-label retrospective chart review.19 Patients given venlafaxine at a dosage of 37.5 to 300 mg/d demonstrated a reduction in mean number of headaches per month from 16.1 at baseline to 11.1 at the final visit, which occurred 6 months later on average.

The responder rate was 28%. Twenty-seven percent of the patients discontinued treatment because of side effects (18%) or lack of efficacy (9%).

In an RDBPC study by a team of Turkish investigators, venlafaxine dosages of 75 and 150 mg/d were compared with placebo in a 10-week trial.20 The primary end point was number of headache attacks, which was found to be statistically lower for study participants receiving the 150 mg/d regimen compared with participants receiving placebo. The study did not address end points of responder rate or reduction in mean monthly migraine frequency.


Topiramate Topiramate (Topamax, Ortho-McNeil) is the newest medication to gain FDA approval for migraine prevention. It is a sulfamate-substituted monosaccharide. The exact mechanism of migraine prevention is unknown. The drug's known mechanisms of action include inhibition of presynaptic voltage-gated sodium and L-type calcium channels; facilitation of g-aminobutyric acid-A receptor chloride flux; inhibition of postsynaptic AMPA/kainate glutamate receptors; and carbonic anhydrase inhibition.

Two US-completed trials (MIGR-001 and MIGR-002) represent the largest and best-designed RDBPC migraine-preventive studies performed to date.21,22 These studies included 970 persons who participated in a 6-month trial. Participants were randomly assigned to total doses of 50, 100, or 200 mg of topiramate, or to placebo administered in 2 divided doses over the trial period. The study drug was titrated at 25 mg/wk increments until the target dose was reached.

Statistically significant improvements in mean monthly migraine frequency were seen with the 100 and 200 mg/d doses in both studies. Responder rates ranged from 47% in the 200 mg/d group (MIGR-002) to 54% in the 100 mg/d group (MIGR-001). Improvements in frequency were seen as early as 1 month.

The most common side effects observed were paresthesias, fatigue, nausea, and loss of appetite. Weight loss was seen at all doses, although more participants (11%) experienced this effect at the highest study dose (200 mg/d).

Divalproex sodium Divalproex sodium (Depakote, Abbott) and divalproex sodium extended-release (Depakote ER, Abbott) are approved medications for migraine prevention. The extended-release formulation was studied in a 17-week RDBPC study. The authors concluded that a once-daily dose of 500 or 1000 mg (patients received a 500-mg dose if the 1000-mg dose was not tolerated) was more effective than placebo.

A total of 237 patients made up the intent-to-treat population. The primary end point was reduction in 4-week migraine headache rate compared with baseline. Patients treated with active medication saw their average monthly migraine frequency reduced from 4.4 to 3.2 compared with patients receiving placebo, who saw a reduction of 4.2 to 3.6.

Only 30% of participants in the active arm achieved a 50% reduction in headache frequency, compared with 24% in the placebo arm (did not achieve statistical significance). Sixty-eight percent of patients treated with active medication experienced side effects. The rate was similar to that seen in the placebo group (70%). Forty-two percent of the treated patients showed a weight gain of at least 2%.23 Clinicians should remember, however, that divalproex sodium is a pregnancy Category D medication. The classification indicates that the drug is associated with an increased incidence of fetal malformations.

Gabapentin An RDBPC study of gabapentin (Neurontin, Pfizer) 1800 to 2400 mg/d showed the drug to be more effective than placebo.24 The study lasted 12 weeks plus an initial 4-week single-blind placebo baseline period.

Only patients who reached a stable dose of 2400 mg/d were included in the per-protocol analysis. At the end of the study, the median 4-week migraine rates were 2.7 for patients receiving gabapentin (compared with 4.2 at baseline) and 3.5 for patients receiving plabebo (compared with 4.1 at baseline).

When the statistical analysis was revised to intent-to-treat, a responder rate of 36% was achieved among patients receiving active treatment. In comparison, the responder rate among patients receiving placebo was 14%. Side effects occurred in 67% of patients, most commonly dizziness (26%) and somnolence (26%).

Lamotrigine A 3-month RDBPC study of lamotrigine (Lamictal, GlaxoSmithKline) 200 mg/d used an unusual study design.25 It attempted to eliminate the typical high placebo response rate by excluding all patients whose headache frequency, after a 4-week single-blind placebo-only phase, fell below the initial minimum requirement. These placebo-responders were not included in the active double-blind phase and accounted for 30% (33/110) of the total patient group.

The remaining 77 patients were randomly assigned to receive active drug at a full dose of 200 mg/d, but because of a high incidence of rash, the protocol was changed so the drug was titrated over 5 weeks until 200 mg/d was reached. The primary end point of reduction in mean monthly attack frequency in the active medication group was not significantly different from that in the placebo group.

Other agents There is currently insufficient quality of evidence to recommend the routine use of the antiepileptic medications levetiracetam (Keppra, UCB Pharma) and zonisamide (Zonegran, Eisai) for migraine prevention.


Methysergide Methysergide (Sansert, Novartis), an ergot alkaloid, is FDA-approved for migraine prevention10,14 but is no longer available in the United States. It is an antagonist at serotonin 2B and serotonin 2C receptors. The effective dose range is 4 to 8 mg/d, and the range of improvement over placebo has been documented as 14% to 30%.9

Prolonged use of this product is associated with retroperitoneal, pulmonary, and cardiac valve fibrosis. It is recommended that patients have a 3- to 4-week drug holiday for every 6 months of medication use. It is a pregnancy Category X (teratogenic) medication. Patients using methysergide should avoid the concurrent use of triptan medications.


Effective treatment of migraine depends on a multifaceted plan. Medical decision making should be guided by the patient's headache disability and frequency, comorbid medical conditions, current overuse of analgesics or other pain medications, and medication side-effect profile. A primary focus should be on controlling headache triggers, such as stress, caffeine use, smoking, and sedentary lifestyle. *


1. Lipton RB, Diamond S, Reed M, et al. Migraine diagnosis and treatment: results from the American Migraine Study II. Headache. 2001;41:638-645.

2. Silberstein SD, Lipton RB, Goadsby PJ. Headache in Clinical Practice. 2nd ed. London: Martin Dunitz; 2002.

3. Hu XH, Markson LE, Lipton RB, et al. Burden of migraine in the United States: disability and economic costs. Arch Intern Med. 1999;159:813-818.

4. Silberstein SD, Welch KM. Painkiller headache. Neurology. 2002;59:972-974.

5. Mathew NT. Transformed migraine. Cephalalgia. 1993;13(suppl 12):78-83.

6. Srikiatkhachorn A. Chronic daily headache: a scientist's perspective. Headache. 2002;42:532-537.

7. Srikiatkhachorn A, Tarasub N, Govitrapong P. Effect of chronic analgesic exposure on the central serotonin system: a possible mechanism of analgesic abuse headache. Headache. 2000;40:343-350.

8. Kruit MC, van Buchem MA, Hofman PA, et al. Migraine as a risk factor for subclinical brain lesions. JAMA. 2004;291:427-434.

9. Ramadan NM, Schultz LL, Gilkey SJ. Migraine prophylactic drugs: proof of efficacy, utilization and cost. Cephalalgia. 1997;17:73-80.

10. Silberstein SD. Practice parameter: evidence-based guidelines for migraine headache (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology. 2000;55:754-762.

11. Solomon GD, Steel JG, Spaccavento LJ. Verapamil prophylaxis of migraine. A double-blind, placebo-controlled study. JAMA. 1983;250:2500-2502.

12. Markley HG, Cheronis JC, Piepho RW. Verapamil in prophylactic therapy of migraine. Neurology. 1984;34:973-976.

13. Solomon GD, Diamond S, Freitag FG. Verapamil in migraine prophylaxis comparison of dosages [abstract]. Clin Pharmacol Ther. 1987;41:202.

14. Silberstein SD, Lipton RB, Dalessio DJ. Wolff's Headache and Other Head Pain. 7th ed. Oxford, England: Oxford University Press; 2001.

15. Tronvik E, Stovner LJ, Helde G, et al. Prophylactic treatment of migraine with an angiotensin II receptor blocker: a randomized controlled trial. JAMA. 2003;289:65-69.

16. Gomersall JD, Stuart A. Amitriptyline in migraine prophylaxis. Changes in pattern of attacks during a controlled clinical trial. J Neurol Neurosurg Psychiatry. 1973;36:684-690.

17. Adly C, Straumanis J, Chesson A. Fluoxetine prophylaxis of migraine. Headache. 1992;32:101-104.

18. Saper JR, Silberstein SD, Lake AE 3rd, Winters ME. Double-blind trial of fluoxetine: chronic daily headache and migraine. Headache. 1994;34: 497-502.

19. Adelman LC, Adelman JU, Von Seggern R, Mannix LK. Venlafaxine extended release (XR) for the prophylaxis of migraine and tension-type headache: a retrospective study in a clinical practice. Headache. 2000;40:572-580.

20. Ozyalcin SN, Talu GK, Kiziltan E, et al. The efficacy and safety of venlafaxine in the prophylaxis of migraine. Headache. 2005;45:144-152.

21. Silberstein SD, Neto W, Schmitt J, et al. Topiramate in migraine prevention: results of a large controlled trial. Arch Neurol. 2004;61:490-495.

22. Brandes JL, Saper JR, Diamond M, et al. Topiramate for migraine prevention: a randomized controlled trial. JAMA. 2004;291:965-973.

23. Freitag FG, Collins SD, Carlson HA, et al. A randomized trial of divalproex sodium extended-release tablets in migraine prophylaxis. Neurology. 2002;58:1652-1659.

24. Mathew NT, Rapoport A, Saper J, et al. Efficacy of gabapentin in migraine prophylaxis. Headache. 2001;41:119-128.

25. Steiner TJ, Findley LJ, Yuen AW. Lamotrigine versus placebo in the prophylaxis of migraine with and without aura. Cephalalgia. 1997;17:109-112.

Loading comments...
Please Wait 20 seconds or click here to close