Clinical Question

Which medication, if any, should be used to prevent headache in children and adolescents with pediatric migraine - amitriptyline, topiramate, or neither?

Inclusion Criteria

• Children and adolescents aged 8 to 17 years
• Diagnosis of migraine with or without aura, or chronic migraine without continuous headache per International Classification of Headache Disorders 2nd Edition (ICHD-II) criteria
• Excluding complicated migraine (hemiplegic, basilar-type, ophthalmoplegic, migrainous infarction)
• PedMIDAS score 11 to 139 (range 0-240; excludes no disability [0-10] and very severe disability [>139])
• Headache frequency ≥4 days from prospective diary over 28-day baseline period
• At randomization: provided diary data on ≥20 of 28 baseline days
• At randomization: ≥15 headache days, each consisting of ≥4 hours continuous headache
• At randomization: ≥50% of headache days were migraine or probable migraine days
• At randomization: ≥4 distinct headache episodes, each lasting ≥4 hours
• Written permission from parent/guardian and appropriate child assent

Exclusion Criteria

• Medical condition that might increase risk with botulinum toxin A exposure (myasthenia gravis, Eaton-Lambert syndrome, amyotrophic lateral sclerosis, any significant disease interfering with neuromuscular function)
• Diagnosis of other primary or secondary headache disorder
• Use of any headache prophylactic medication within 28 days of baseline day 1
• Beck Depression Inventory score >24 at baseline day 1
• Temporomandibular disorder
• Fibromyalgia
• Psychiatric disorders that could interfere with study participation
• Previous exposure at any time to any botulinum toxin serotype
• For women of childbearing potential: positive pregnancy test or not using reliable contraception

Outcomes

Criticisms

• Trial stopped early for futility may have limited ability to detect smaller treatment effects, though multiple sensitivity analyses supported null findings
• Very high placebo response rate (61%) may have made it difficult to demonstrate drug superiority, though this is consistent with pediatric headache literature
• Relatively short 24-week treatment duration may not capture longer-term benefits or harms of these medications
• Dose escalation over 8 weeks followed by only 16-week maintenance may have been insufficient time at therapeutic dose
• No active comparator between the two drugs and placebo - would have benefited from head-to-head comparison in separate powered trial
• Inclusion of both episodic (4-14 days) and chronic (≥15 days) migraine may have created heterogeneous population, though stratification attempted to address this
• Predominantly female (68%) and white (70%) population limits generalizability to more diverse pediatric populations
• Two-thirds of patients had previously tried and failed prophylactic medications, suggesting potentially treatment-refractory population
• No assessment of non-pharmacological interventions or behavioral treatments that might enhance or replace medication
• Missing data rate of 21% in amitriptyline and 22% in topiramate groups (vs 11% placebo) could introduce bias despite intention-to-treat analysis
• Adherence measured by drug levels only in subset of patients, not all participants
• Study powered for 70% drug response vs 50% placebo, but actual rates were much lower, raising questions about initial assumptions
• No long-term follow-up data after weaning period to assess durability of effects or rebound
• Child Depression Inventory and BRIEF assessments done quarterly may have missed psychiatric adverse events between visits
• Exclusion of very severe disability (PedMIDAS >139) limits applicability to most severely affected children

Funding

National Institutes of Health grants U01NS076788 and U01NS077108 from the National Institute of Neurological Disorders and Stroke (NINDS) and Eunice Kennedy Shriver National Institute of Child Health and Human Development