Correspondence

Drug Therapy of Migraine

N Engl J Med 1994; 330:1015-1016April 7, 1994

Article

To the Editor:

We were surprised to see no mention of corticosteroids in Dr. Welch's otherwise informative survey of current migraine therapy (Nov. 11 issue)1. Corticosteroids have been recommended for the treatment of prolonged migraine2,3 and are commonly prescribed for that indication in the community at large. A survey of the members of the American Association for the Study of Headache found that 20 percent of responding physicians used corticosteroids as first-line agents in patients with prolonged migraine and 19 percent used them as second-line therapy4. Although, as noted by Dr. Welch and his colleagues,5 the efficacy of corticosteroids has not been evaluated in a randomized trial, their widespread use warranted mention in his review.

Amnon Lahad, M.D.
University of Washington, Seattle, WA 98195

Jaime B. Friedman, M.D.
Sand Point Internists, Seattle, WA 98105

5 References

1
Welch KMA. Drug therapy of migrane. N Engl J Med 1993;329:1476-1483
Full Text | Web of Science | Medline

2
Gallagher RM. Emergency treatment of intractable migraine. Headache 1986;26:74-75
CrossRef | Web of Science | Medline

3
Tollison CD, Kunkel RS, eds. Headache diagnosis and interdisciplinary treatment. Baltimore: Williams & Wilkins, 1993.

4
Couch JR Jr, Diamond S. Status migrainosus: causative and therapeutic aspects. Headache 1983;23:94-101
CrossRef | Web of Science | Medline

5
Olesen J, Tfelt-Hansen P, Welch KMA, eds. The headaches. New York: Raven Press, 1993.

To the Editor:

Welch suggests that ketorolac “can be used for emergency treatment of severe migraine attacks complicated by vomiting, although the drug is less effective than other parenteral antimigraine preparations.” This recommendation was based on a study of patients who had no response to their usual medication1. Ketorolac is a nonsteroidal antiinflammatory agent that has serious side effects. In a worldwide survey, the incidence of serious side effects was 1 per 10,000 treated patients; the most common reactions were gastrointestinal disorders (22 percent), blood dyscrasias (20 percent), and renal impairment (13 percent)2,3. As of June 1993, 97 deaths had been reported among patients given ketorolac, and marketing of the drug has been suspended in France and Germany. In our opinion the risk-benefit ratio of ketorolac is unfavorable, arguing against its use in the treatment of migraine.

Gilles Mignot, M.D.
Christophe Kopp, M.D.
Prescrire International, 75527 Paris, France

3 References

1
Klapper JA, Stanton JS. Ketorolac versus DHE and metoclopramide in the treatment of migraine headaches. Headache 1991;31:523-524
CrossRef | Web of Science | Medline

2
Ketorolac: new restrictions on dose and duration of treatmentCurr Probl 1993;19:5-6

3
Ketorolac tromethaminePrescr Intl 1993;2:104-106

Author/Editor Response

Dr. Welch replies:

To the Editor: Drs. Lahad and Friedman take me to task for not recommending corticosteroids in my review. My reasons can be explained by a critical appraisal of the same literature they quote in favor of doing so. I assume that by “prolonged migraine” they mean intractable migraine, a clinical state that is itself poorly defined. The reference they provide, apart from the textbook, is one by Gallagher,1 who reported that among a group of patients who had headache that lasted more than 36 hours and were treated with 8 mg of dexamethasone parenterally in addition to meperidine and promethazine, 72 percent had no or mild headache when contacted 24 hours later. This was a very short report of an open study based on a retrospective chart review. The study groups had unequal numbers of patients and were not matched for age or sex, and the results were not subjected to statistical analysis. There was no measure of the severity of headache before treatment, although this was used as an outcome measure 24 hours later, and the mode of dexamethasone administration was not stated. There were also no details about the questionnaire administered over the telephone 24 hours after treatment, and it is doubtful that all the patients were questioned precisely 24 hours after treatment. Thus, the literature offers no support for the use of corticosteroids as a first- or second-line treatment for migraine. To quote Edmeads's pithy comment,2 this treatment for migraine “must be viewed in the light of the traditional role of corticosteroids as the pharmacological last rite for neurological disease.”

With respect to the use of parenteral ketorolac for a severe migraine attack, I stated that the drug “can be used for emergency treatment,” and then went on to comment about its efficacy, as quoted by Drs. Mignot and Kopp. I am confident that a critical clinician would not interpret this as a recommendation. They are correct in stating that regulatory authorities in Germany and France have suspended the license for the injectable form of ketorolac pending further study. A recent post-marketing surveillance study of 20,000 patients demonstrated that “the safety profile of injectable ketorolac was no different from that seen in clinical trials which formed the basis of the approval of ketorolac for marketing”3. An interim analysis of these data appears in the package insert for the drug in preparations purchased in the United States.

K.M.A. Welch, M.D.
Henry Ford Hospital, Detroit, MI 48202

3 References