Advances in paediatric migraine

Abstract

Headache is one of the most common complaints of children who present to primary care providers. Although parents are often concerned about sinister pathology, the majority of children with headaches have primary headache syndromes, including migraine and tension-type headaches. Diagnostic criteria for children are currently evolving to better reflect the unique challenges of this group of patients. Advances in migraine pharmacotherapy have been achieved through the understanding of serotonin and its role in migraine pathophysiology. Serotonin agonists, commonly known as ‘triptans’, are the current standard of care in the management of acute migraine in adults. Recent evidence has confirmed that the efficacy of triptans also occurs in children. The present article focuses on recent advances in the areas of epidemiology, diagnostic criteria and pathophysiology of paediatric migraine. In addition, the present article reviews the evidence of management issues, including neuroimaging and the use of triptans in children.

EPIDEMIOLOGY

Generations of children have grown up believing the myth that migraine is an exclusively adult ailment. Although the first record of childhood headache dates back to biblical times (1), it has been only in recent years that epidemiological data have substantiated the plight of children with migraine. Bille (2) used a cohort of 9000 school-aged children in Uppsula, Sweden and found that by the age of seven years, 2.5% had frequent, nonmigrainous headaches, and 1.4% had true migraines. By age 15 years, the prevalence increased to 15.7% and 5.3%, respectively. More recent studies have confirmed the overall prevalence of migraines in children to be 3% to 10% (3,4). When the data were further analyzed by sex, the prevalence in seven-year-old children was found to be similar at 2.9% for boys and 2.5% for girls (5). However, by age 14 years, the prevalence in girls jumped to 14.8% compared with 6.5% in boys, which is comparable to data seen in adult studies (6).

Data for the long term outcome of childhood migraines are beginning to emerge. Bille (7) conducted a 40-year follow-up of his original patients with migraines, and found that by the age of 25 years, 23% of the children with migraine were permanently headache-free. By age 50 years, 46% were permanently headache-free. Data consistently showed sex differences, with men being more likely than women to be migraine-free. A Canadian group (8) reported similar data from a 10-year follow-up study that showed only 18.5% of children with a diagnosis of migraine were headache-free at reassessment. Age and sex at the time of diagnosis and at follow-up were analyzed (9). Boys younger than seven years of age with migraines showed total remission in 26%, partial remission in 37% and persistence in 37% compared with 19%, 37% and 44%, respectively, in girls. In the eight- to 14-year-old group, persistence of headache occurred in 78% of boys and in 73% of girls. These data suggest that migraine is a life-long disorder, with boys being more likely to experience prolonged remissions.

DIAGNOSTIC CRITERIA

An essential prerequisite in the management of migraine is establishing an accurate diagnosis; however, children pose several unique limitations in headache assessment. First, subjective descriptors are limited, especially in very young children. Moreover, the spectrum of migraine symptoms clearly changes over the years. Typically, the preschool-aged child with migraine is less able to describe the headache accurately, but displays an abundance of gastrointestinal-related symptoms than does the adolescent, who more closely resembles the adult profile. These distinctions limit the clinical utility of rigid adult-based criteria. Vahlquist (10) proposed the first criteria to establish a diagnosis for paediatric migraine, requiring the presence of recurring headaches and at least two of the following: nausea, visual aura, family history, or unilateral pain. Prensky and Sommer (11) modified these criteria to require three of the following: nausea or vomiting, abdominal pain, visual aura, family history, unilateral pain or throbbing. These criteria were useful clinically and particularly sensitive for younger patients. In 1988, the International Headache Society (IHS) created the current gold standard for the diagnosis of migraine and other headache disorders (12) (Table 1). The IHS criteria make a single concession for children younger than 15 years of age, in that the duration of the attack is reduced from 4 to 72 h to 2 to 48 h. Data from large clinical trials consistently identified weaknesses in applying the IHS criteria to children. Sechia et al (13) found that only 61% of children with a clinical diagnosis of migraine met IHS criteria. Similar data were obtained by Wober-Bingol et al (4) and Gladstein et al (14), who reported 71% and 69% agreements with the IHS criteria, respectively.

TABLE 1.

International Headache Society criteria for migraine

Migraine without aura	Migraine with aura
At least five attacks fulfilling B-D Headache attacks lasting 2–48 h Headache has at least two of the following: Unilateral location Pulsating quality Moderate to severe intensity Aggravation by routine physical activity During headache, at least one of the following occur: Nausea and/or vomiting Photophobia or phonophobia	At least two attacks fulfilling B At least three of the following: One or more fully reversible aura symptoms indicating focal cortical and/or brain-stem dysfunction At least one aura developing gradually over more than 4 min, or two or more symptoms occurring in succession No auras lasting more than 60 min Headache follows less than 60 min

Proposals to modify the IHS criteria for children have been suggested by several authors. Maytal et al (15) found the IHS criteria to have poor sensitivity (27%), but excellent specificity (92%). They suggested that changes to headache duration, unilateral of pain, could increase sensitivity and make the IHS criteria more useful clinically. Winner (16), using the IHS-R (revised) criteria, improved the sensitivity of diagnosis from 66% to 93% (Table 2). In children younger than 12 years, the differences were more marked, with IHS diagnosis of only 49% compared with 87% for IHS-R.

TABLE 2.

Revised (proposed) International Headache Society criteria for migraine without aura in children

At least five attacks fulfilling B-D Headache attacks lasting 1–48 h Headache has at least two of the following: Bilateral location (frontal and temporal) or unilateral location Pulsating quality Moderate to severe intensity Aggravation by routine physical activity During headache, at least one of the following occur: Nausea and/or vomiting Photophobia and/or phonophobia

From a clinical perspective, diagnostic criteria raise important caveats. If diagnostic criteria are rigid and consequently difficult to apply in the office setting, they are of little clinical value. Large groups of children with headaches not fulfilling the criteria would be left undiagnosed, without proper direction for management. Conversely, research trials demand adherence to strict criteria to ensure that they accurately reflect the case definition. In the interim, clinicians wait to see whether the proposed revisions to the IHS criteria will be accepted.

PATHOPHYSIOLOGY

It is well know that the brain, cranium and the majority of the dura are not pain-sensitive structures. Intracranial structures that are pain-sensitive include the large cerebral vessels, pial vessels and large venous sinuses. Headache pain originates from a complex interaction of neuromediators that include intracranial and extracranial structures, and are processed in central connections in the brain. The pain-sensitive structures are innervated by unmyelinated fibres that arise from peripheral branches of the trigeminal ganglion and the upper cervical nerve roots. This distinction is useful clinically because pain over the front of the face and vertex originates in the trigeminal system, while pain in the occiput or neck is mediated via the cervical roots.

Although migraine has traditionally been viewed as the prototype vascular headache, recent evidence suggests that neurogenic mechanisms play an important role. The current model of migraine focuses on the trigeminovascular system (17). Central mechanisms involve a wave of spreading cortical depression in a raustral to caudal direction at a rate of 2 to 3 mm/min – the same rate reported by Leao (18) in his classic study of cortical depression in 1944. This spreading depression accounts for the aura and focal cerebral injury, leading to sterile inflammation and edema. Significant advances in the understanding of migraine pathophysiology occurred following the discovery of serotonin and the subsequent characterization of its receptor subtypes. The locus ceruleus and the nucleus raphe dorsalis within the brain stem are richly supplied with serotoninergic and noradrenergic pathways that project to the cortex. Serotonin pharmacology has identified at least seven receptor subtypes (19). 5-hydroxytryptamine-1-D (5-HT1D) receptors cause vasoconstriction of cerebral vessels and arteriovenous anastomoses and, therefore, play a central role in abortive therapy of migraine.

Recent evidence has focused on the many vasoactive peptides released within the trigeminal system (20). Calcitonin gene-related peptide (CGRP), neurokinin-A and substance P are powerful vasoactive neuropeptides that contribute to neurogenic sterile inflammation during migraine. The trigeminal nerve releases substance P and CGRP into dural and meningeal blood vessels, which causes the degranulation of mast cells, leading to histamine release. In addition, platelets release serotonin, causing extravasation of plasma proteins and further potentiating the inflammatory cascade leading to pain.

The initiating event in migraine is not yet known, but various triggers have been recognized. These triggers stimulate cortical, thalamic or hypothalamic sites, activating the brain stem nuclei that, through serotoninergic connections to the cortex, produce a neurogenic spread of cortical depression, leading to sterile arteritis. Local tissue injury results in the release of vasoactive neuropeptides, causing inflammation of dural and pial blood vessels, from which nociceptive afferents in the trigeminal nerve mediate pain centrally.

NEUROIMAGING

The history and physical examination are the most important tools in evaluating children with headache. Children who present with symptoms consistent with uncomplicated migraine, and who have a normal neurological examination, do not require any investigations. Bloodwork and electroencephalogram are often ordered as part of the diagnostic ‘work up’, but are not helpful. The use of neuroimaging has increased significantly in recent years due to patient demands for tests as well as the fear of litigation.

Several recent studies have examined the use of neuroimaging in paediatric headache. Children with uncomplicated migraines and chronic daily headache were studied using computed tomography (CT) or magnetic resonance imaging (MRI). The yield in uncomplicated migraine was 3.7% compared with 16.6% in the chronic daily group. It is important to note that the positive imaging studies identified only incidental findings that did not have an impact on management (21). Maytal et al (22) found similar results and calculated the maximal rate of abnormality to be 3.8%. A recent study reviewed the clinical predictors for space-occupying lesions in children presenting with headache (23). A positive correlation between the number of predictors and the risk of space-occupying lesions was noted. Suggested guidelines for neuroimaging are summarized in Table 3. The choice of CT versus MRI has been recently addressed. Children at low risk (uncomplicated migraine and normal neurological examination) do not require imaging, but, if imaging is indicated, MRI has been shown to maximize the quality-adjusted life expectancy at a reasonable cost effective ratio (24).

TABLE 3.

Guidelines for neuroimaging

Nocturnal or early morning headaches

Progressive headache

Abnormal neurological findings on examination

Focal neurological symptoms during headache (except visual aura)

Headache with change in personality

Age less than five years

TREATMENT

Abortive therapy

Successful treatment of migraine begins with preventative measures including careful review of potential triggers and adjustments of sleep cycle irregularities. A headache calendar is useful to accurately record the frequency of headaches, due to a strong recall bias in children (25). Tailoring medication to the severity of the attack leads to the more rational use of drugs. Simple analgesics must be given at the earliest onset of a migraine attack to maximize efficacy. Ibuprofen has been shown to be twice as effective in reducing headache pain as acetaminophen (26), but acetaminophen has a faster onset of headache relief. Codeine and combination analgesics are not recommended for routine use because they have not been well studied in children with migraines and they contribute to rebound headaches. Antiemetics are useful for children who have gastrointestinal disturbances during migraine attacks. In addition to increasing absorption across the gastrointestinal tract, antiemetics are also helpful in migraine therapy by promoting sleep, which decreases serotoninergic firing from the brain stem nuclei.

Triptans:

Treatment of severe migraine has been revolutionized by the introduction of triptans. This class of drug targets serotonin receptors in a highly selective and effective fashion. Sumatriptan, a 5-HT1D agonist, is the first and best-studied triptan, with an overall clinical efficacy of approximately 75% in adults. Although triptans are the standard of care in adult migraine, it must be emphasized that at the time of writing this paper, Health tCanada had approved none of the triptans for paediatric use. Paediatric studies using all three routes (oral, injection and nasal spray) have now been published and show generally similar data. Sumatriptan is effective in migraine with or without aura, but is contraindicated in basilar and hemiplegic migraine. Adverse events include noncardiac chest pain and a bothersome side effect known as a triptan effect. The latter is characterized by facial flushing and tingling of the extremities. Oral sumatriptan was highly efficacious and well tolerated in a group of adolescent patients, with 74% reporting significant headache improvement at 4 h (27). Adolescents are given 25 to 50 mg of sumatriptan at the onset of headache, but vomiting and gastroparesis often limit oral absorption of the drug. Linder (28) used subcutaneous sumatriptan and reported an efficacy of 72% within 30 min of administration. Despite the rapidity and efficacy of the subcutaneous route, most children prefer a less invasive route, such as a nasal spray. In a recent large, randomized, double-blind, placebo controlled study of 12- to 17-year-old adolescents, sumatriptan nasal spray (20 mg) was found to be 66% effective at 2 h, with only taste disturbance being reported as a common adverse event (29). In a one-year tolerability and efficacy study, 437 adolescents were treated for 3272 migraine attacks by using sumatriptan nasal spray. The results revealed the 20 mg nasal spray to be 72% effective in reducing headache, with 40% of the patients being pain-free at 2 h. Although well tolerated in most patients, one patient experienced a facial nerve ischemic event that was thought to be drug-related (30).

Current research is focusing on the use of sumatriptan in children under 12 years of age, and preliminary data indicate similar efficacy and tolerability of sumatriptan in this age group (31,32). Hershey et al (31) treated 10 children between the ages of five and 12 years with nasal sumatriptan, and reported an 82.5% response to the 57 headaches treated. In another small study of children younger than 10 years of age, sumatriptan nasal spray was effective in 85% of children and offered complete headache relief in 65% (32). The second generation triptans, including zolmitriptan, rizatriptan and eletriptan, are also being investigated for clinical efficacy in adolescents. Zolmitriptan 2.5 mg was found to be effective and well tolerated in adolescents, with 80% responding to pain and 66% being pain-free at 2 h (33). Further data are needed before advocating their use in clinical practice.

Status migrainosus

Status migrainosus is defined as a migraine headache that persists for more than 48 h, and it represents a formidable challenge to the clinician in the office or emergency department (34). Sumatriptan is the drug of choice for children who are triptan-naive on presentation with status migrainosus. Children who are refractory to sumatriptan can be treated with intravenous chlorpromazine. Chlorpromazine is believed to block the effects of serotonin and histamine, both of which are responsible for increased vascular permeability (35). Chlorpromazine 10 to 25 mg is given intravenously by slow infusion while hydration status and blood pressure are closely monitored. It is recommended that a saline bolus be given before the chlorpromazine infusion and that blood pressure be rechecked after infusion. Oral chlorpromazine may be given on discharge to reduce the recurrence of headache.

Dihydroergotamine (DHE) is a synthetic ergotamine with affinity for several serotonin receptors, and is useful in treating acute migraine. It has been used in children for the treatment of status migrainosus and for in-patient management of refractory migraine. Linder (36) reported on the use of intravenous DHE in children with status migrainosus. Using a modified Raskin protocol, 80% of patients responded to the protocol with minimal side effects. For children over 12 years of age, premedication consists of oral metaclopramide before administration of DHE. Unlike the Raskin protocol, which calls for intravenous metaclopramide, this study found that oral metaclopramide leads to significantly fewer side effects, such as nausea and extra pyramidal dysfunction. Metaclopramide 0.1 mg/kg intravenously alone is an effective option in status migrainosus.

The use of steroids has been well established in adults with status migrainosus, but data in children are lacking (34). Steroids exert their effects in status migrainosus by reducing central neurogenic inflammation. Dexamethasone can be given intravenously or intramuscularly. Alternatively, a five-day oral course may be given.

Prophylactic therapy

Prophylactic therapy needs to be considered for either frequent migraine attacks or individual attacks refractory to abortive therapy. Clinical factors such as the age of the child, degree of disability and pre-existing medical conditions are important factors that influence the decision to initiate prophylactic therapy. A child experiencing four to six severe migraine attacks per month would be a candidate for prophylaxis. Before establishing a diagnosis of refractory migraine, the management protocol that is being followed by the patient must be reviewed. Common pitfalls include the delay of instituting abortive medication, the lack of antiemetic use, suboptimal dosing and an inappropriate delivery of the medication. The pharmacological treatment of migraine was reviewed and it was found that many of the standard drugs being used in prophylaxis lack proven efficacy in children (37). Since the publication of the article by Igarashi et al (37), controlled studies have been conducted for the medications reviewed below.

Beta blockers:

Traditionally the mainstay of prophylaxis, propranolol was accidentally discovered and is now an effective agent in adult migraine. Data for propranolol use in children are limited. In one study (38), 32 children who were taking propranolol reported a decreased incidence of migraine symptoms without adverse events, while two additional studies reported no significant improvement (39,40). Propranolol is usually started at 10 mg twice daily and titrated clinically thereafter. Beta blocker use is limited by its contraindication in children with asthma or diabetes and its potential for aggravating or causing depression.

Calcium channel blockers:

Flunarazine is thought to be effective in migraine prophylaxis by blocking 5-HT release and inhibiting the contraction of vascular smooth muscle and prostaglandin formation. Flunarazine is one of the best studied prophylactic agents in children, with several reports demonstrating clinical efficacy (41,42). Recently, low dose flunarazine was studied in children younger than 10 years of age and showed a decreased frequency of migraine attacks (43). Adolescents should begin with 5 mg before bedtime and should be monitored for side effects, including constipation, weight gain and hypotension.

Anticonvulsants:

Divalproex sodium was introduced as a migraine prophylaxis in 1988. The mechanism of action is believed to be via gamma-aminobutyric acid (GABA) receptors in the raphe nucleus where it decreases the firing rate of serotonergic cells. Adult data demonstrate a 50% reduction in the frequency of migraine attacks. A recent paediatric study (44) also reported a 50% reduction in migraine attacks for 70.5% of children. In addition, 14% of the children experienced a 75% reduction in migraine frequency (44). The usual starting dose of 250 mg twice daily is less than that used in the treatment of epilepsy, and therapeutic drug monitoring is not necessary. Gastrointestinal upset, dizziness, weight gain and tremor were the most frequent adverse events. Gabapentin is structurally related to GABA receptors, but does not interact with GABA, and its mechanism of action is unknown. Gabapentin has been shown to be an effective and well tolerated prophylactic agent in adults (45), but no paediatric trials have been reported.

Antidepressants:

Tricyclic antidepressants such as amitriptyline have been reported to decrease migraine frequency by as much as 42% in adults, but studies in children are limited. Its effect in prophylaxis appears to be independent of its antidepressive properties. In an open label study, amitriptyline showed a 50% to 80% reduction in migraine frequency over a three-month study period (46). Hershey et al (47) used a standardized dosage for various headache syndromes, including migraine, and demonstrated an overall improvement of 84%, with minimal side effects. For adolescents, a starting dose of 10 mg before bedtime is recommended with clinical titration to 75 mg/day. Serotonin-selective reuptake inhibitors are effective in adults, but no study results are available for children.

Others:

Pizotifen is a serotonin antagonist widely used in Canada and Europe, where it is particularly useful as a first line agent in young children. The drug has antiserotonin, antihistamine and anticholinergic activity. Pizotifen was found to be an effective prophylaxis in 70% of children with migraine and has also demonstrated excellent results in children with abdominal migraine (48). Pizotifen is started at 0.5 mg/day and titrated clinically up to 4 to 6 mg/day. Side effects of weight gain and appetite stimulation are usually not a problem in this age group and are generally avoided by giving a single dose before bedtime.

Nonpharmacological therapy

There has been a global trend toward an increased interest in alternatives to drug therapy in all areas of medicine. While the field of headache is no exception, nondrug therapies are particularly appealing to the parents of migraine sufferers, due to their fears of adverse effects of medications on the growing child. Several behavioural therapies have been studied in children and represent a safe and well tolerated therapy for paediatric migraine. An important limitation of behaviour therapy is that it is more effective in the management of chronic or recurrent migraine than in episodic and severe migraine. Moreover, behavioural therapy is most beneficial when administered by experienced therapists, but these services are often limited to tertiary headache clinics.

Relaxation therapy:

Developed in the 1920s, this method relies primarily on tensing and releasing exercises of large muscle groups. Additional relaxation methods include guided imagery, massage and hypnosis. Two recent reviews demonstrated the efficacy of behavioural therapy compared with drug therapy in migraine (49,50).

Biofeedback:

The objective is similar to relaxation therapy in its attempt to reduce the physiological arousal associated with migraine headaches. Compared with relaxation, biofeedback is a more controlled and focused relaxation technique that uses electronic instruments to measure, process and feed back information to help gain control over physiological processes. Thermal biofeedback involves teaching children to regulate peripheral temperature to treat headaches. In a meta-analysis, Herman et al (50) demonstrated that biofeedback is efficacious in treating paediatric migraine. Biofeedback combined with relaxation therapy reduced the headache index (frequency × intensity) more effectively than a beta-blocker in children with migraine (51).

Homeopathic medicine:

A variety of agents have been used in adult migraine prophylaxis including riboflavin, feverfew and acupuncture, and are reviewed by Mauskop (52). In a systematic review, the clinical effectiveness of feverfew as a prophylactic agent could not be established (53). Data on the results of using these agents are not available in children.

CONCLUSION

The future is bright for children with migraine. Paediatric migraine is beginning to receive international attention in academic and research forums. Ongoing research is focused on improving diagnostic methodologies and understanding the genetic basis of migraine. The treatment of migraine will also improve as new studies explore the use of second generation triptans in children.