Objective. Treatment of pediatric migraine includes an individually tailored regimen of both nonpharmacologic and pharmacologic measures. The mainstay of symptomatic treatment in children with migraine is intermittent oral or suppository analgesics, but there is no coherent body of evidence on symptomatic treatment of childhood migraine available. The objective of this review is to describe and assess the evidence from randomized and clinical controlled trials concerning the efficacy and tolerability of symptomatic treatment of migraine in children.
Design. Systematic review according to the standards of the Cochrane Collaboration.
Methods. Databases were searched from inception to June 2004. Additional reference checking was performed. Two authors independently selected randomized and controlled trials evaluating the effects of symptomatic treatment in children (<18 years old) with migraine, using headache (HA) clinical improvement as an outcome measure. Two authors assessed trial quality independently by using the Delphi list, and data were extracted from the original reports by using standardized forms. Quantitative and qualitative analysis was conducted according to type of intervention.
Results. A total of 10 trials were included in this review, of which 6 studies were considered to be of high quality. The number of included participants in each trial ranged from 14 to 653, with a total of 1575 patients included in this review. Mean dropout rate was 19.8% (range: 0–39.1%), and the mean age of participants was 11.7 ± 2.2 years (range: 4–18 years).
All studies used HA diaries to assess outcomes. In most studies, a measure of clinical improvement was calculated by using these diaries. Improvement often was regarded as being clinically relevant when the patients' HA declined by ≥50%.
Regarding oral analgesic treatment, the effectiveness of acetaminophen, ibuprofen, and nimesulide were evaluated. When compared with placebo, acetaminophen (relative risk [RR]: 1.5; 95% confidence interval [CI]: 1.0–2.1) and ibuprofen (pooled RR: 1.5; 95% CI: 1.2–1.9) significantly reduced HAs. We conclude that there is moderate evidence that both acetaminophen and ibuprofen are more effective in reduction of symptoms 1 and 2 hours after intake than placebo with minor adverse effects. No clear differences in effect were found between acetaminophen and ibuprofen or nimesulide.
Regarding the nonanalgesic interventions, nasal-spray sumatriptan, oral sumatriptan, oral rizatriptan, oral dihydroergotamine, intravenous prochlorperazine, and ketorolac were evaluated. When compared with placebo, nasal-spray sumatriptan (pooled RR: 1.4; 95% CI: 1.2–1.7) seemed to significantly reduce HAs. We conclude that there is moderate evidence that nasal-spray sumatriptan is more effective in reduction of symptoms than placebo but with significantly more adverse events. No differences in effect were found between oral triptans and placebo. All medications were well tolerated, but significantly more adverse events were reported for nasal-spray sumatriptan compared with placebo.
We also conclude that there is moderate evidence that intravenous prochlorperazine is more effective than intravenous ketorolac in the reduction of symptoms 1 hour after intake. No differences in effect were found between oral dihydroergotamine and placebo.
Conclusions. Acetaminophen, ibuprofen, and nasal-spray sumatriptan are all effective symptomatic pharmacologic treatments for episodes of migraine in children. The new frontier for symptomatic treatment is likely to be the development of triptan agents for use in children. Most treatments have only been evaluated in 1 or 2 studies, which limits the generalizability of the findings.
We strongly recommend performing a large, high-quality randomized, controlled trial evaluating different symptomatic medications compared with each other or to placebo treatment. Favorable high-quality studies should be performed and reported according to the CONSORT statement. Clinical improvement of HA should be used as the primary outcome measure, but quality of life, days missed at school, and satisfaction of child or parents should also be used as an outcome measure in future studies.
Migraine is an important clinical problem in school-aged children, with an estimated prevalence of 2.7% to 10%.1,2 It is characterized by attacks of intense, throbbing, unilateral head pain, often accompanied by nausea, vomiting, photophobia, and phonophobia.3,4
Treatment of pediatric migraine includes an individually tailored regimen of both nonpharmacologic and pharmacologic measures. Nonpharmacologic modalities include lifestyle adjustments (ie, sleep hygiene, dietary adjustment, or exercise program), reassurance, stress management, biofeedback, and other biobehavioral therapies. Pharmacologic interventions include the use of symptomatic medication such as analgesics (acetaminophen, ibuprofen), triptans (sumatriptan), and antiemetics (prochlorperazine) and the use of prophylactic medication. The mainstay of symptomatic treatment in children with migraine is intermittent oral or suppository analgesics, but there is no coherent body of evidence on symptomatic treatment of childhood migraine available.
Good-quality controlled trials preferably summarized in a systematic review form the basis for evidence-based treatment guidelines, which may improve the management and treatment of individual patients. One systematic review on triptans only for pediatric migraine has been performed.5 Based on 4 randomized, controlled trials (RCTs), they found effectiveness with nasal-spray sumatriptan compared with placebo in acute pediatric migraine, whereas oral sumatriptan and rizatriptan were not clearly beneficial.5 To study adverse events, open-label studies were also included.5
The objective of this review was to describe and assess the evidence from RCTs and clinical controlled trials (CCTs) concerning the efficacy and tolerability of symptomatic treatment of migraine in children.
We searched Medline, Embase, PsychInfo, Web of Science, and Cinahl from inception to June 2004 using the terms “migraine,” “headache,” “cephalgia,” “cephalalgia,” “child*,” “infant,” “teenage,” “adolescent,” or “p(a)ediatric” together with the search strategy for identifying clinical trials described by Robinson and Dickerson.6 The Cochrane Controlled Trials Register was searched using the words “migraine,” “headache,” “cephalgia,” “cephalalgia,” “child*,” “infant,” “teenage,” “adolescent,” or “p(a)ediatric.” Additional strategies for identifying trials included searching the reference lists of review articles and included studies.
We selected only RCTs and CCTs including symptomatic pharmacologic medicine used in the management of migraine in children (<18 years old), with criteria designed to distinguish migraine from other types of headache (HA). The use of a specific set of diagnostic criteria (eg, International Headache Society [IHS]3 and Ad Hoc Committee on the Classification of Headache7) was not required, but migraine diagnoses had to be based on at least some of the distinctive features of migraine.3,4 Our outcome measures of interest were: HA intensity, frequency or duration, HA index, and overall improvement. Rescue medication was defined as additional medications (other than study medication) permitted in nonresponders, usually limited to the habitual medications a person uses for their migraine HA. When a protocol permitted the use of rescue medication before outcome measurement (2–6 hours), then the latest outcome assessment not confounded by the use of rescue medication was extracted. No language restriction was applied.
Two authors (L.D. and J.K.J.B.) independently screened titles and abstracts of studies identified by the literature search for eligibility. All potentially relevant studies were retrieved as full papers and then again independently reviewed by 2 authors (L.D. and J.K.J.B.). Any disagreements were resolved through consensus, when possible, or by arbitration of a third author (A.P.V.).
Methodologic Quality and Data Extraction
Two authors (L.D. and J.K.J.B. or A.P.V.) independently assessed the methodologic quality of the included trials using the Delphi list.8 The Delphi list is a generic criteria list developed by international consensus and consists of (1) randomization, (2) adequate allocation concealment, (3) groups similar at baseline, (4) specification of eligibility criteria, (5) blinding of outcome assessor, (6) blinding of care provider, (7) blinding of patient, (8) presentation of point estimates and measures of variability, and (9) intention-to-treat analysis. One extra item was added because it was found to be relevant for these studies: (10) withdrawal/dropout rate (>20% or selective dropout) unlikely to cause bias. The methodologic criteria were scored as yes (1), no (0), or don't know (0).
A quality score of each trial was computed by counting the number of positive scores. Any disagreements were resolved through consensus, when possible, or by arbitration of a third author (J.K.J.B. or A.P.V.).
Data extraction was performed by 1 author (L.D.) and checked by a second author (A.P.V.). Disagreements were resolved by consensus. Extracted information included (if available) demographic data, detailed description of the intervention and control (ie, dose given, study duration), outcome measures, and information on adverse effects.
We calculated relative risks (RRs) with 95% confidence intervals (CIs). We performed an available-case analysis (data reported), and if sufficient data were available, we also performed a worst-case analysis (ie, all dropouts are assumed to be nonresponders). Data are presented as treatment success, indicating that a RR > 1 represents a better outcome for the first mentioned intervention group.
In case of a crossover trial, we ideally would like to restrict our analysis to first-period data only or, in case of a sufficient washout period and no carryover effect, data of both periods could be combined. In this review we analyzed the crossover trials as if they were parallel-group trials, because none of them provided separate data of each treatment period or data concerning a carryover effect, which would otherwise lead to exclusion of most crossover trials.
Statistical pooling was performed in the case of clinical homogeneity. In parallel studies, when >1 comparison from the same study (ie, sumatriptan 5, 10, and 20 mg versus placebo) was used for the statistical pooling of sumatriptan versus placebo, the results from the placebo group were evenly spread out over the 3 comparisons, and the number of patients in the placebo group was divided by 3 to prevent triple counting (R.J.P.M. Scholten, PhD, Dutch Cochrane Centre, verbal communication, 2004).
A qualitative analysis was performed by using a rating system with levels of evidence.9 The evidence was judged to be strong when multiple (≥2) high-quality RCTs produced generally consistent findings. Results were considered consistent if ≥75% of the studies reported similar results on the same outcome measure. It was judged to be moderate when 1 high-quality RCT or multiple (≥2) low-quality RCTs or CCTs produced generally consistent findings. Evidence was considered to be limited when only 1 low-quality RCT or CCT existed and conflicting if the findings of existing trials were inconsistent. No evidence was considered when no RCTs or CCTs were found or when the authors provide no sufficient data for analysis. We arbitrarily regarded trials with methodologic quality scores of ≥6 as of high quality.9
The results of our search strategy are presented in Fig 1. In total, 9 RCTs and 1 CCT are included in this review.
Description of Studies
Full details of the included studies are presented in Table 1.
The number of included participants in each trial ranged from 14 to 653 (mean 158 ± 201 patients), with a total of 1575 patients included in this review. The mean percentage of participants who dropped out from the trials was 19.8% (range: 0–39.1%). Mean age of participants was 11.7 ± 2.2 years (range: 4–18 years). Overall, the percentage of boys was generally the same as girls (mean: 55%; range: 42–82%). Nine of 10 trials used the criteria of the IHS to classify migraine, and 1 trial10 used the criteria of Prensky and Sommer.11 Six studies used a crossover design.
The pharmacologic interventions10,12–20 used for symptomatic treatment could be divided into analgesic treatment and nonanalgesic treatment. Regarding oral analgesic treatment, the effectiveness of acetaminophen, ibuprofen, and nimesulide was evaluated. Regarding the nonanalgesic interventions, nasal-spray sumatriptan, oral sumatriptan, oral rizatriptan, oral dihydroergotamine, intravenous (IV) prochlorperazine, and IV ketorolac were evaluated. Eight studies included a placebo comparison.
All studies used HA diaries to assess outcomes. Using this diary, among others, the frequency, intensity, and duration of HAs were scored on a Likert scale. In most studies a measure of clinical improvement was calculated. Improvement often was regarded as being clinically relevant when the patients' HAs decline by ≥50%. This score is presented as “HA clinical improvement” in Table 1 and is our primary outcome measure. Other outcome measures were HA index, frequency, intensity or duration, rescue medication, and adverse events posttreatment and at follow-up.
Quality scores (with positive items in parentheses) are presented in the “study quality” section of Table 1. The median score for methodologic quality was 6 (range: 3–9). Using a cutoff point of 6 of 10 criteria, 6 studies (60.0%) were considered to be of high quality.10,12,13,16,18–20 Most prevalent methodologic shortcomings were a concealed randomization method (unclear: 80%), the intention-to-treat analysis (unclear: 50%; negative: 20%), and comparability of groups at baseline (unclear: 70%; negative: 10%).
Effectiveness of Symptomatic Treatment
Table 2 gives an overview of effect estimates of the different treatment modalities described below.
Oral Analgesic Treatment
Acetaminophen Versus Placebo
One high-quality study compared 15 mg/kg acetaminophen with placebo.12 HA improvement was significantly higher for acetaminophen, although the use of rescue medication as well as the number of adverse events were not significantly different between the 2 groups. Adverse events mentioned for acetaminophen versus placebo were nausea (2.4% vs 3.7%) and vomiting (2.4% vs 7.4%).
Ibuprofen Versus Placebo
Two studies compared 10 mg/kg ibuprofen with placebo, including 1 high-quality study12 and 1 low-quality study.14 In the pooled analysis, HA improvement was significantly higher for ibuprofen, and the use of rescue medication was significantly higher for placebo. No difference in the number of adverse events was found (vomiting: 4.9% vs 7.4%; nausea: 3.7% for both; gastric pain: 1.2% [ibuprofen only]).12
Acetaminophen Versus Ibuprofen
One high-quality study compared 15 mg/kg acetaminophen with 10 mg/kg ibuprofen and found no significant differences for any outcome measures.12
Acetaminophen Versus Nimesulide
One low-quality study compared 15 mg/kg acetaminophen with 2.5 mg/kg nimesulide and found no significant differences.15 Adverse events (mild abdominal discomfort and nausea) accompanied 9% of the treatments with acetaminophen and 6% of treatments with nimesulide.
We conclude that there is moderate evidence that both acetaminophen and ibuprofen are more effective in reduction of symptoms 1 and 2 hours after intake than placebo, with minor adverse effects. No clear differences in effect were found between acetaminophen and ibuprofen or nimesulide.
Sumatriptan Versus Placebo
Three studies evaluated intranasal sumatriptan.13,16,17 Ahonen et al16 did not present outcomes for the doses separately. At 1 and 2 hours after administration we found significant differences concerning HA improvement in favor of sumatriptan. The use of rescue medication was significantly more for placebo, whereas the number of adverse events was significantly more for sumatriptan. Adverse events mentioned for sumatriptan versus placebo included bad taste (25.7% vs 3.1%), nausea (7.9% vs 8.3%), vomiting (3.9% vs 4.4%), triptan sensation (temperature [warmth], burning/stinging sensations, or paresthesia) (2.4% vs 1.6%), feeling of light-headedness (2.2%, sumatriptan only), and stiffness in the jaw (1.1%, sumatriptan only).
One high-quality study evaluated oral sumatriptan and found no significant differences for any outcome measures.18 Adverse events mentioned for sumatriptan were vomiting (17.4%), nausea (8.7%), dizziness (4.3%), and very strong photophobia (4.3%). For placebo, the adverse events were vomiting (8.7%) and nausea (4.3%).
Rizatriptan Versus Placebo
One low-quality study19 compared oral rizatriptan with placebo and found no significant differences for clinical improvement of HA, rescue medication, or the number of adverse events. Adverse events mentioned for rizatriptan versus placebo were dry mouth (4.7% vs 3.4%), dizziness (4.7% vs 4.8%), asthenia (3.4% vs 2.0%), nausea (2.7% vs 8.2%), and somnolence (2.7% vs 8.2%).
We conclude that there is moderate evidence that nasal-spray sumatriptan is more effective in the reduction of symptoms than placebo, but with significantly more adverse events. No differences in effect were found between oral triptans and placebo.
Dihydroergotamine Versus Placebo
One high-quality study compared oral dihydroergotamine with placebo.20 We found no significant differences for clinical improvement of HA, rescue medication, or the number of adverse events, probably because of statistically low power (smallest treatment group: <25 patients). The 1 adverse event mentioned for dihydroergotamine versus placebo was vomiting (16.7% vs 7.7%).
Prochlorperazine Versus Ketorolac
One high-quality study compared IV prochlorperazine with IV ketorolac.10 After 1 hour, prochlorperazine worked significantly better compared with ketorolac. The number of adverse events was not significantly different between the 2 groups.
We conclude that there is moderate evidence that IV prochlorperazine is more effective than IV ketorolac in the reduction of symptoms 1 hour after intake. No differences in effect were found between oral dihydroergotamine and placebo.
The most effective therapies for migraine are those that can be given quickly at the beginning of an attack and have a rapid onset of action. Based on the available literature, we conclude that over-the-counter analgesics such as acetaminophen and ibuprofen are relatively safe and effective. Most RRs were at ∼1.5, which means that there is 50% more chance (risk) to recover when using, for example, ibuprofen instead of placebo. No clear differences in effect were found between the different analgesics. After finishing our study, another systematic review on treatments of pediatric migraine was published with similar conclusions.21 This review included just 5 controlled trials on pharmacologic treatment, and no statistical pooling resulting in an overall effect estimate was performed.
The new frontier for symptomatic treatment is likely to be the development of triptan agents for use in children. We found moderate evidence that nasal-spray sumatriptan is more effective than placebo, but no differences in effect were found between oral sumatriptan and rizatriptan compared with placebo. It is possible that oral triptans may not be effective in children during acute migraine attacks because of nausea and vomiting. In adults, sumatriptan pharmacokinetic data have been collected after oral administration in patients suffering from migraine: the absorption from an oral tablet was delayed during a migraine attack because of gastric stasis, nausea, and vomiting.22 It is likely that this also takes place in children.
Also, migraine attacks in children tend to be shorter than those in adults and may spontaneously remit within 2 hours. Medications, which would be anticipated to provide maximum benefit close to 2 hours after administration, may not show efficacy for shorter-duration attacks, and results may be comparable to those when using placebos.
Nasal administration offers an advantage over the tablet administration, because it provides faster absorption immediately after dosing, which results in faster initial relief of migraine symptoms. The only disadvantage of the use of nasal-spray sumatriptan compared with the other investigated medications is the adverse events. Although all reported adverse events were minor for all reported medications, the use of nasal-spray sumatriptan resulted in significantly more adverse events than the use of placebo.
Although systematic reviews offer the least biased method of summarizing research literature, our results must be interpreted with consideration of the quality of evidence from which they were obtained. None of the crossover trials provided separate data of each treatment period or data concerning the existence of a carryover effect. Positive results of analgesics and triptans in children and adolescents have frequently emerged from open-label studies, which might be biased.
Most treatments have been evaluated in only 1 or 2 studies, which limits the generalizability of the findings. There are no clinical trials that tested the efficacy of different triptans with each other. Studies comparing the efficacy of triptans to analgesic drugs with known efficacy are also needed.
Clinical improvement of HA and adverse events are 2 outcome measures frequently used in the included trials for symptomatic treatment of migraine attacks. Although most studies described the adverse events as mild and safe, the simple description of the kind and number of adverse events often gives insufficient insight into the severity and appreciation of the adverse events for a child. Outcome measures such as quality of life, satisfaction of child or parents, and days missed at school are not used.
We strongly recommend performing a large high-quality RCT that evaluates different symptomatic medications compared with each other or to placebo treatment. Favorable high-quality studies should be performed and reported according to the CONSORT statement (see www.consort-statement.org). Clinical improvement of HA should be used as the primary outcome measure, but quality of life, days missed at school, and satisfaction of child or parents should also be used as outcome measures in future studies.
There is moderate evidence that analgesics (acetaminophen and ibuprofen) and nasal-spray sumatriptan are more effective than placebo treatment. Based on the available literature, no differences in effect were found between the different analgesics.
We thank the Netherlands Organization for Health Research and Development for funding this research.
- Accepted February 15, 2005.
- Address correspondence to Arianne P. Verhagen, PhD, Department of General Practice, Erasmus Medical Centre, PO Box 1738, 3000 DR, Rotterdam, Netherlands. E-mail:
No conflict of interest declared.
- ↵Robinson KA, Dickerson K. Development of a highly sensitive search strategy for the retrieval of reports of controlled trials using PubMed. Int J Epidemiol.2002;31 :150– 153
- ↵Prensky AL, Sommer D. Diagnosis and treatment of migraine in children. Neurology.1979;29 :506– 510
- ↵Hämäläinen ML, Hoppu K, Valkeila E, Santavuori P. Ibuprofen or acetaminophen for the acute treatment of migraine in children: a double-blind, randomized, placebo-controlled, crossover study. Neurology.1997;48 :103– 107
- ↵Winner P, Rothner AD, Saper J, et al. A randomized, double-blind, placebo-controlled study of sumatriptan nasal spray in the treatment of acute migraine in adolescents. Pediatrics.2000;106 :989– 997
- ↵Soriani S, Battistella PA, Naccarella C, Tozzi E, Fiumana E, Fanaro S. Nimesulide and acetaminophen for the treatment of juvenile migraine: a study for comparison of efficacy, safety, and tolerability. Headache Q.2001;12 :233– 236
- ↵Ahonen K, Hämäläinen ML, Rantala H, Hoppu K. Nasal sumatriptan is effective in treatment of migraine attacks in children: a randomized trial. Neurology.2004;62 :883– 887
- ↵Ueberall MA, Wenzel D. Intranasal sumatriptan for the acute treatment of migraine in children. Neurology.1999;52 :1507– 1510
- ↵Hämäläinen ML, Hoppu K, Santavuori P. Sumatriptan for migraine attacks in children: a randomized placebo-controlled study. Do children with migraine respond to oral sumatriptan differently from adults? Neurology.1997;48 :1100– 1103
- ↵Lewis D, Ashwal S, Hershey A, et al. Practice parameter: pharmacological treatment of migraine headache in children and adolescents: report of the American Academy of Neurology Quality Standards Subcommittee and the Practice Committee of the Child Neurology Society. Neurology.2004;63 :2215– 2224
- ↵Culter NR, Hussey EK, Sramek JJ, et al. Pharmacokinetics of oral sumatriptan in migraine patients during an attack and while pain free. Biol Psychiatry.1992;31 :180A
- Copyright © 2005 by the American Academy of Pediatrics