Reducing migraine return with corticosteroids: An extra chance to improve migraine care

Migraine treatment in the emergency department is a scenario patient and physician alike would prefer to avoid. With severe pain, aversion to light and sound, nausea, vomiting, and a desire to achieve sleep, a migraine sufferer would naturally choose less provocative surroundings, if only relief could be had elsewhere. For her emergency physician, the patient with migraine almost never has a life-threatening, but a life-interrupting, episode, one that lasts hours to days and responds inconsistently to interventions. Then, at discharge, after success and relief appear together, the question looms for both patient and physician, will the migraine return? Just as the most common rank of a graduating class (minimum size one) is valedictorian, any given visit to an emergency department for headache is most likely be the first, but it is repeated visits, repetitive, treatmentresistant migraine attacks, that raise the signal that a migraine sufferer is not managing well. One possible tool to delay headache recurrence after a prolonged migraine attack is the class of corticosteroids (denoting any naturally occurring or analog glucocorticoid or mineralocorticoid). In their systematic review in the current issue of Cephalalgia, Woldeamanuel and coauthors (1) searched the entirety of accessible medical literature from the time of production of synthetic corticosteroids to the present to identify relevant data about a common pharmacological treatment of migraine in emergency departments and similar settings. The authors used multiple search strategies of published and unpublished sources, evaluated study quality, and abstracted information to allow for meta-analysis using current methods. The report of this exhaustive search should be of interest to headache physicians because it contains both encouraging and discouraging news. In 60 years’ data, only 25 studies appear to address this clinical scenario, corticosteroid treatment of migraine attack in acute care settings. For the reader’s convenience and further study, these 25 studies are summarized by the authors in Table 1 (1). The identified studies’ sum is greater than the parts; for example, although the conclusion of one individual study (2) was negative in a primary outcome, the secondary outcome in a subset of patients was positive, and leads to classification under ‘favorable outcomes’ for the treatment (1 (Table 1)). The data sets retrieved and reviewed point in the same direction: For the migraine attack that resists other therapy, corticosteroids (most often a single dose of 10mg intravenous dexamethasone, Figure 5 (1)) tend to reduce the recurrence rate and severity of subsequent headaches. Opportunities to expand reviewable data abound: The number of patients for whom data have been captured for this systematic review and analysis (just under 4000) must be a figure many orders of magnitude lower than those exposed over six decades to treatment outside of experimental settings. The conclusion based on available data is promising: Short-term, high-dose corticosteroid use now justifiably retains its ‘‘time-honoured place’’ (3) in the toolkit for the treatment of prolonged migraine attacks—the accumulation of evidence supports current practice. Some of the studies identified may not be widely generalizable because of the adjuvant role of the corticosteroid, with primary treatments that may not be applied in all or most cases (e.g. metoclopramide and diphenhydramine in Friedman et al. (4)). Discouraging for patients and clinicians seeking all the answers today, but encouraging for those keen to tackle today’s questions, such limitations show the work to be done: Increasing the number of attackand-treatment pairs studied, improving reporting and accessibility of data, and comparing routes and particular agents head-to-head would refine subsequent recommendations regarding corticosteroids. The question

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.