Multiple randomized controlled trials fail to show the effectiveness of migraine chemoprophylaxis in children and adolescents

Référence

Locher C, Kossowsky J, Koechlin H, et al. Efficacy, safety, and acceptability of pharmacologic treatments for pediatric migraine prophylaxis: a systematic review and network meta-analysis. JAMA Pediatr 2020;174(4):341-349.

Sommaire

These authors searched several databases and a clinical trials registry to identify randomized trials of pharmacologic approaches to prevent migraine in children and adolescents younger than 18 years. The selection of studies, data extraction, and assessment of risk of bias were all conducted independently by 2 or more authors. Unfortunately, and not surprisingly, the authors found the studies used a variety of outcomes, so they created a hierarchy and selected the highest measure available in the included studies. From highest to lowest, the efficacy measures were: monthly migraine frequency; number of monthly migraine days; number of headache days each month; headache frequency; and indices of headache intensity, frequency, and duration. The authors also assessed treatment safety and tolerability. After pooling the data, they performed a network meta-analysis to combine the direct and indirect effects of the included pharmacologic agents. Ultimately, they included 23 relatively short duration (median 12 weeks) randomized trials with 2217 migraineurs. Overall, the risk of bias in the included studies was assessed as moderate. The pharmacologic agents included anticonvulsants, beta blockers, antidepressants, calcium channel blockers, and various supplements. Independently, many agents demonstrated trends to benefit but had very wide confidence intervals. The authors noted similar trends for safety and tolerability. When they plugged all the data into their network analysis, propranolol and topiramate were the most effective, but with confidence intervals that included the possibility that there was no overall effect. Since multiple agents have been studied only in single trials, this is by no means an answered question.