Pharmacologic acute & preventive treatment for migraine in children and adolescents

Evidence Profile & Guideline Recommendations

Guideline Title	Acute treatment of migraine in children and adolescents1	Pharmacologic treatment for pediatric migraine prevention2
Release Date	August 14, 2019
Prior version	Practice parameter: pharmacological treatment of migraine headache in children and adolescents from American Academy of Neurology (AAN) and Child Neurology Society, 2004
Developer	AAN and American Headache Society
Funding Source	AAN
Target population	Children and adolescents with migraine
Major recommendations and quality/evidence ratings	Take a detailed history and make a specific headache diagnosis (B). Counsel patients to treat acute migraine early in the attack (B). Use ibuprofen to treat pain in children/adolescents; in adolescents consider sumatriptan/naproxen oral, zolmitriptan nasal, sumatriptan nasal, rizatriptan ODT, or almotriptan oral (B). If one triptan is ineffective, try another or a non-steroidal anti-inflammatory drug-triptan combination (B). If headache pain spikes quickly try a non-oral triptan (C). If nausea/vomiting try non-oral triptan or add anti-emetic (B-C). Avoid triptans in patients with cardiovascular disease and accessory conduction pathway disorders (A). Patients with migraine with aura may use triptans (B).	Advise patients and families on lifestyle factors, migraine triggers, and avoidance of acute medication overuse (B). Consider preventive treatments in children/adolescents with frequent or disabling headaches or medication overuse (B). Discuss with patients/families whether to use a preventive medication, since placebo was as effective as the studied medication in many trials (B). Discuss the evidence for and side effects of amitriptyline combined with cognitive behavioral therapy, topiramate, and propranolol for migraine prevention (B). When relevant, discuss teratogenic effects of topiramate and valproate, and advise patients to use effective birth control methods and take folate (A). Divalproex, onabotulinumtoxinA, amitriptyline alone, nimodipine, and flunarizine are not clearly better than placebo (insufficient evidence). Continue to monitor the effectiveness of preventive treatments over time (A-B).

Summary of the Clinical Problem:

Migraine affects 9.1% of children and adolescents³, causing significant impairment in quality of life⁴ and school performance⁵. Acute treatments are appropriate for all patients in order to decrease symptoms and disability from individual attacks. Preventive treatments aim to reduce the frequency and severity of migraine attacks. Within each of these categories, there are three different classes of treatment modalities: Self-management interventions & Lifestyle modifications, Non-pharmacologic, and Pharmacologic. The 2004 treatment guidelines highlighted that there was little evidence to guide the choice of pharmacologic therapy, hence these guidelines were undertaken to incorporate more recent trials and provide an update on pharmacologic recommendations.

Characteristics of the Guideline Source & Evidence Base:

The guidelines were created following the 2011 American Academy of Neurology (AAN) guideline development process manual as amended, in compliance with the National Academy of Medicine (formerly Institute of Medicine) Standards for Systematic Reviews. The AAN limits the participation of authors with substantial conflicts of interest. A multi-disciplinary panel of authors reviewed the literature published between 12/1/03 and 8/25/17, and graded the evidence, including an assessment of confidence in the evidence. Of note, the systematic review excluded intravenous therapies, as well as nonpharmacologic therapies unless paired with pharmacologic. The panel then wrote recommendations, incorporating both the strength of the evidence and other factors including principles of care, balance of risks and benefits, cost and availability, and patient preference. The initial study protocol and final draft of the guidelines were posted for public comment, and the manuscript was peer reviewed before publication.

Guideline Rating	Acute Guideline	Preventive Guideline
Establishing transparency	Good	Good
Management of Conflict of Interest	Good	Good
Guideline group composition	Good	Good
Guideline-systematic review intersection	Fair: Many of the recommendations expanded beyond systematic review	Fair: Many of the recommendations expanded beyond systematic review
Establishing evidence foundations and rating strength for each of the guideline recommendations	Fair: Some of the recommendations reflect clinical practice, but do not have strong base of evidence in pediatrics.	Fair: Some of the recommendations reflect clinical practice, but do not have strong base of evidence in pediatrics.
Articulation of recommendations	Good	Fair: many of the recommendations are not easily evaluable
External review	Good	Good
Updating	Fair: No specific date given	Fair: No specific date given
Implementation issues	Good	Fair: Nuanced counseling, including discussion of placebo, may be difficult to implement. Some evidence-based therapies (CBT) not available in all locations.

Discussion:

Despite 15 years passing between the 2004 practice parameter and these recently released guidelines, there were only 21 additional studies on pharmacologic treatments which met the bar for inclusion. Several of these trials were parallel design studies which had a very high rate of response to placebo, and failed to demonstrate additional benefit of the studied treatment. Since all designs were weighted equally in the meta-analysis, acute medications which were tested only with novel study designs (placebo lead-in or crossover) fared better in the meta-analysis. Both acute and preventive medications which had been tested solely or partly in parallel design trials fared worse. As a result, even some medications, namely rizatriptan and topiramate, which have met the bar for FDA-approval as migraine therapies in children/adolescents received lukewarm recommendations in this guideline. This highlights the need for further well designed and executed studies for the treatment of pediatric migraine. There are several pediatric and adolescent studies planned or ongoing to test medications and neuromodulatory devices which have received FDA-approval or clearance in adult migraine; we hope these investigators will consider trial design and patient selection carefully in order to demonstrate efficacy if it truly exists.

Given that very few studies of pharmacologic therapies were ultimately included in the full literature review, we were surprised by the decision to exclude non-pharmacologic therapies and nutraceuticals. We hope that these will be included in a separate guideline in the near future.

The treatment recommendations in the guidelines (summarized in the table) go beyond the scope of the initial clinical questions and focused literature review to incorporate practical aspects of care. There are pros and cons to this approach. For example, the inclusion of recommendations on lifestyle modifications appropriately highlights the importance of treating the entire patient. However, because the literature review focused on pharmacologic therapies, the recommendations to counsel about lifestyle factors was not held to the same standard of evidence. While we think of lifestyle modifications as likely having a favorable risk to benefit ratio, adherence to lifestyle changes can be difficult, which can lead to guilt and frustration in patients. In the next guidelines it would be ideal if recommendations about lifestyle habits could be truly evidence-based. In another example, the acute guidelines recommend inclusion of an anti-emetic to address nausea/vomiting, and recommend the use of combination therapies in patients who do not respond to monotherapy. These suggestions reflect our clinical practice, but neither of these recommendations are actually backed by clear evidence in pediatric migraine. These discrepancies highlight the need for rigorous studies to examine every aspect of our treatment recommendations – lifestyle modifications, non-pharmaceutic treatments, and combinations or series of treatments over time.

Overall, we appreciate that the authors have completed this large undertaking so that our field can reflect on the scope of the current evidence. Across multiple trials and retrospective studies, approximately two-thirds of pediatric migraine patients improve with the studied therapy. For these patients our current standard of care (as outlined in the guideline recommendations) is probably sufficient. However, for the remaining third, which translates to 3% of all children, we must do better to prevent suffering and disability. For these children, we hope that the guidelines are seen as a call to action to continue and expand pediatric migraine research.