Timothy C. Hain, MD •Page last modified: March 6, 2021
You may also be interested in our many other pages on migraine on this site
|Migraine prevalence by age (Stewart et al, 1994) . Dashed line are females, solid line, males. This was a very large study.|
As an overview, only about 5% of children younger than 12 have migraine. At puberty, while both genders develop more migraine, the rate of growth is much higher in women. In the figure above, it is clear that migraine increases with age and is far more common in women. The data are not aligned on menarche or menopause. It seems likely that were the female data analyzed this way, there would be a much sharper inflection around 12 years old in women, as well as a second spike at menopause.
Lewis et al. (2004), in a practice parameter for the evaluation of pediatric and adolescent headache, reported a meta-analysis of six prevalence studies involving 13,130 children. Migraine was reported in 1.2% to 3.2% of those aged 3 to 7 years, 4% to 11% of children aged 7 to 11 years, and 8% to 23% of children aged 11 to 15 years or older. There was increasing prevalence with age as well as a shifting from a slight male predominance to a female predominance in adolescents. There were many more children with recurrent headaches than "migraine". Of course, migraine is a diagnosis defined by a commitee (the IHS), and it seems very likely that many of the recurrent headaches have the same mechanism for headaches as the commitee defined patients.
According to Jahn et al (2011), about 5.7% of 10 year olds have "vestibular vertigo", and about 40% of them, vestibular migraine. In other words, they suggest that roughly 2.5% of the 10 year old population has vestibular migraine. We think this is a very high estimate, as only 1% of adults are thought to have vestibular migraine (Neuhauser et al, 2006).
Migraine in children is disruptive of school performance (Arruda, 2012).
The IHS criteria for diagnosis of migraine are found here. There are many disorders, possibly migranous, that don't "fit" the IHS criteria. Some of those that can cause vertigo are listed below.
This is a disorder of uncertain origin, possibly migrainous. It's initials (BPV) are easily confused with those of Benign Paroxysmal Positional Vertigo (BPPV), but it is not caused by the same mechanisms. This disorder consists of spells of vertigo and disequlibrium without hearing loss or tinnitus (Basser, 1964). The majority of reported cases occur between 1 and 4 years of age, but this syndrome seems indistinguishable from benign recurrent vertigo (BRV, see following) in adults which is presently attributed to migraine, or so-called "vestibular Menieres", which is also attributed to migraine. The differential diagnosis includes Menieres disease, vestibular epilepsy, perilymphatic fistula, posterior fossa tumors, and psychogenic disorders.
This is a very disturbing disorder in which persons suddenly develop vomiting, generally without headache or hearing symptoms. It usually responds to migraine prevention medications, as well as some abortives. See this page for more.
According to Silver et al (2008), only ibuprofen and sumatriptan have been shown to be more effective than placebo. On the other hand, there is a reasonable literature suggesting that dopamine blockers (such as metoclopramide) are useful in the pediatric acute migraine care.
In the ER for treatment of migraine in children includes dopamine antagonists such as prochlorperazine, metoclopramide, or chlorpromazine, as well as analgesics such as acetaminophen and ibuprofen or ketorolac (Richer et al, 2010; Walker and Teach, 2008). Intravenous prochlorperazine has been reported effective in acute confusional migraine (Khatri et al, 2009).
According to Papetti et al (2010), almotriptan was approved for treatment by the FDA of adolescent migraine. Nasal sumatriptan and nasal zolmitriptan were approved for acute treatment of adolescent migraine by the EMEA. Effective doses for Sumatriptan NS are reportedly 10 mg for children weighing < 40 kg, and 20 mg for children > 40 kg. As of 2010, no conclusive data was available regarding use of oral or subcutaneous sumatriptan preparations in children or adolescents, but in general, a positive effect is noted. Berenson et al (2010), reported reasonable responses to oral almotriptan (12.5) in adolescents.
DHE, dihydroergotamine, given intravenously has been reported effective in children and adolescents(Kabbouche et al, 2009).
According to Sheridian et al (2012), subanesthetic doses of propofol are effective. We have also heard that IV midazolam can be effective. While these medications would certainly be effective in calming children down, we think that giving general anesthetics to children is too risky to be practical.
Our suggestions for dietary management of migraine in children can be found on this single page migraine-diet handout
As in adults, one starts with behavioral/dietary changes. Diet, good sleep and avoidance of sun was reported by Eidlitz-Markes (2010) as being effective in children younger than 6. Other suggested strategies are to hydrate, and to avoid artificial sweeteners. We do not know of any literature that establishes that this is effective, but on the other hand, lack of evidence is not evidence of lack.
Our suggestions for patients in our clinical practice are confined to diet/sleep. Regarding diet, MSG is often the biggest trigger as it is commonly found in "junk" food such as "Cheetos" and "Pringles" chips. Chocolate can also be a problem. Lots of cheese (such as in pizza) is sometimes a problem. Lots of caffeine (such as in soda) also can be problematic.
In children who are developing their brains and going to school, but who are sometimes old enough to have children of their own, we think it is best to be very conservative about medications. Here we organize treatment of migraine in chilidren in the same categories as treatment of migraine in adults. As an overview, there is little data, and what little there is suggests there is no "magic bullet". Common treatments for prevention of migraine in children, amitriptyline and topiramate, appear to be no different in effect than placebo (Powers et al, 2016). Others that are not clearly better than placebo include sodium valproate, botulinum toxin in jection, nimodipine, and flunarizine. (Szperka et al, 2020). It is possible however that these are effective in subgroups.
Amitriptyline is one of the most widely used agents (Papetti et al, 2010), with dosing similar to used in adults (up to 1 mg/day). According to Powers et al, placebos are equally effective as amitriptyline. According to Eidlitz-Markus et al (2012), amitriptyline (0.26 mg/kg/day) is moderately effective for pediatric migraine. For a 50 kilo teen, 12.5 mg of amitriptyline. This would be a reasonable dose for a similarly sized adult too. This medication has weight gain as a problem and is also sedating. On the other hand, improving sleep might be a benefit too. As in adults, we think it should be used as a last-resort agent rather than first agent.
Battistella (1993) reported that trazodone (1 mg/kg) was effective in pediatric migraine (ages 7-17). They used a 1 month "run in" period, and prescribed it in divided doses, TID. So as an example, 50 mg/day for a 110 lb child. This would be a small dose of trazodone for an adult. Trazodone is generally felt to be a very safe sleep medication for older adults. We would like to see at least one more randomized trial.
Although in adults, other antidepressant agents such as nortriptyline and venlafaxine are highly effective for migraine, no studies of migraine of these agents in children have been reported. In adults, venlafaxine does not have weight issues or sedation issues. Venlafaxine is also somewhat effective for ADD. When venlafaxine is used for psychiatric purposes in children, as in adults, withdrawal syndromes can be serious. (Hosenbocus et al, 2011).
Anticonvulsants such as topiramate, are commonly used to treat migraine in children.
As of 2014, the evidence for anticonvulstant efficacy is weak. However, they do seem more promising than most alternatives. As of 2016, leviteracetam may be the best pick.
Topiramate appears to be a slightly effective drug, only slightly better than placebo in children according to data from Winner et al (2005). Powers et al (2016) found it to be identical to placebo. Winner used a dose of 2 mg/kg/day. For a 110 lb adolescent, this would be about 100 mg/day. As topiramate is associated with a high risk of birth defect, caution is appropriate in adolescents. According to Shamilyn (2013), Topiramate (two RCTs), divalproex (one RCT), and clonidine (one RCT) were no more effective than placebo in preventing migraine. Weight loss, speech disturbance (word finding) and difficulty thinking are potential problems with topiramate. In adults, more than 50 mg/day increases the risk of speech disturbance greatly. Obviously, in a child going to school, a drug that reduces vocabulary could cause difficulties. There is also a tendency to develop kidney stones (Dell'orto VG et al, 2013).
Sodium valproate also has been a slightly effective drug for pediatric migraine in several placebo controlled trials. According to Shamiliyan (2013), sodium valproate demonstrated no significant differences for migraine prevention or migraine-related disability compared with propranolol (two RCTs) or topiramate (one RCT). Tremor and weight gain are problems for sodium valproate. It also is a high risk drug in pregnancy.
Leviteracetam, as of 2010, had seen no placebo controlled trials. In open label trials, responses reported were quite good, with only 10% discontinuing due to side effects.
Only a single randomized control was obtained of gabapentin, reporting good results in 18 year olds. We would not consider 18 year olds much different than adults. Gabapentin is generally a safe medication, though ineffective medication, in adults, and we would expect that it might also be safe and ineffective in children.
Evidence for these working for migraine in children is weak to nonexistent.
According to Eidlitz-Markus et al (2012), propranolol (0.4 mg/kg/day) is moderately effective for pediatric migraine. For a 50 kilo teen, this might work out to about 20 mg of propranolol. For an adult, this would be a very low dose. According to Toldo et al, the data regarding propranolol is "conflicting". Propranolol was compared with valproate as well as behavioral treatment, and 2 studies compared different doses of topiramate; none of these trials showed significant differences (El-Chammas et al, 2013 ). This would suggest either that propranolol is not especially effective, or in all of the trials, by happenstance, all treatments were effective. We favor the first idea.
According to Shamilyan (2013), metoprolol tended to be less effective than stress management in preventing migraine or reducing migraine severity (one RCT). Propranolol had less effect than self-hypnosis on absolute number of migraine attacks (one RCT). These data are difficult to interpret.
Nimodipine (a calcium channel blocker) is reported to be ineffective (Toldo et al, 2012). This would suggest that other calcium channel blockers that work in adults (such as verapamil) may also be ineffective.
The calcium channel and dopamine blocker, flunarizine, is thought to be effective ( Visudtibhan et al, 2004; Papetti, 2010; Evers, 2008). Among the 10 active comparator trials, flunarizine was no better than aspirin, dihydroergotamine, or propranolol. (El-Chammas et al, 2013 ). There are good reasons to avoid this medication, such as it's very long half life, weight gain and drug induced parkinsonism.
Cyproheptadine was associated with a "positive response" at 6 months in 83%. Headache frequency was reduced by 55%. (Lewis et al, 2004). Weight gain can be an issue.
Botox has been reported effective in adolescents in an open label study (Chan et al, 2009). As botox is not especially effective in adults, one would not expect it to be a powerful treatment in children.
According to Powers et al, amitriptyline and topiramate are placebos.
According to Damen et al (2006), nimodipine, clonidine, L-5HTP, trazodone and papaverine showed no effect when compared with placebo.
According to El-Chammis (2013) ineffective drugs for episodic migraine included clonidine, flunarizine, pizotifen, propranolol, and valproate. Unlike Damen et al, El-Chammis reported that Trazodone was found effective for episodic migraine.
According to Toldo et al (2012), the results from the use of propranolol are conflicting, whereas nimodipine and clonidine have been shown to be noneffective.
According to Eiland et al (2007), In clinical trials, natural supplements have shown a lack of efficacy versus placebo for prophylaxis of migraines in children.