1. Specify, the various conditions causing headaches in children and differentiate them by history and physical examination.

2. Understand the indications for various imaging studies.

3. Be conversant with current views on the pathophysiology of migraine and drugs available for migraine prophylaxis and treatment.

4. Understand social and emotional factors which cause or exacerbate headaches in children and utilize behavioral management and psychological support services when appropriate.

Headaches in Children Clinical Approach and Therapeutic Options

Epidemiology Classification

Headache as a is a common although often nonspecific symptom. The most commonly encountered headaches are those which have no obvious cause. Considering the various categories of headache helps the examiner to take a more comprehensive history. If the physician has not succeeded in classifying a recurrent headache from the patient's history, it is unlikely that the examination will provide an answer.

Headache may conveniently be classified as follows

Without Known Anatomic Basis
Nonspecific, chronic recurring ("tension," "muscle contraction," chronic daily)
Nonmigrainous vascular
Vascular Migraine
Vascular Cluster
Toxic/Metabolic
Emotionally determined

With known or presumed anatomic basis

Increased intracranial pressure
Traction
Craniocephalic Inflammation
Various diseases of the head and neck (including paranasal sinuses)

These categories are not mutually exclusive: some conditions fall in more than one category and some patients have more than one type of headache.

It is helpful to use a form when taking the headache history since some details which will later prove to be relevant will otherwise be omitted. A simple form, which generally suffices for the pediatric headache history, is attached. In adults, who often have a headache history of ten or more years, the headache history is more complex and time consuming.

When possible the headache history should be obtained directly from the child and the parent should be asked to "fill in the blanks" only after the patient has answered the examiner's questions to the best of his or her ability. This approach is valuable because children and their parents often present divergent views concerning the frequency and severity of the problem and the extent to which headache is disabling: ie., how much it interferes with school attendance or participation in other age appropriate activities. Parents and children may also have differing views regarding factors which precipitate the child's pain. The headache history has two major sections: the first concerns the relationship of the headache to the patient's day activities and the way in which headaches have varied over time. Chronic, continuous, daily headache, for example, is not consistent with the diagnosis of classic migraine. Headaches which only occur on Sunday night or Monday morning during the school year have obvious implications. In children of divorced parents with shared custody, headaches which occur only in one parent's home or before leaving the primary residence for a weekend with one of the parents suggest a number of possible etiologies. It is important to determine if headaches result in secondary gain, which may serve as a reinforcing factor. This usually takes the form of school avoidance, reduced family conflict, increased parental concern, or a reprieve from an extracurricular activity--such as high-pressure competitive sports--which may consciously or unconsciously provoke anxiety and fear of failure. Psychological factors should be suspected in patients who have previously undergone extensive, unrevealing evaluations for pain and now have chronic, unremitting headache of obscure etiology.

Headaches are less likely to be emotionally determined if (1) they are unrelated to school attendance, (2) they have a well defined onset and termination, (3) have no recognized association with emotional upset or hyperventilation, (4) the patient has not previously had other ill defined, chronic symptoms, (5) multifocal pain is not reported and (6) other family members do not have headaches which are used for secondary gain.

Headaches associated with increased intracranial pressure characteristically awaken patients at night. Headaches which do not awaken the patient but are present when the patient awakens at the usual hour may also be due to increased intracranial pressure but are less specific. When headaches are due to increased intracranial pressure, they are frequently worsened by bending, stooping, straining, or coughing. If headaches have been present for several years, they are not likely to be the result of a progressive, underlying neurological disorder.

Benign intracranial hypertension is less common in children than in adults. In most patients, the etiology is undefined but numerous etiologies are known. These include dural sinus thrombosis, corticosteroid therapy or discontinuation, use of oral contraceptives and hypervitaminosis A. Some cases in young adolescent girls resolve when menses begin. Hormonal factors often appear to play a causative role in adolescents and young adults since the condition has a predilection for obese females. However, a sex preference does not appear to occur in pre-adolescent cases.

The second part of the headache story is the characterization of a representative episode or of multiple episodes if the patient has more than one type of headache. The progression of symptoms during a typical headache episode should always be reviewed. Discrepancies between the description of a typical episode ("When Jimmy gets a headache, he lies down in his room and watches television") and the patient's reported disability ("It hurts too much for me to stay in school.") suggests that secondary gain plays an important role in the patient's complaints. A history of headache combined with pain in other arm or of migrating pain which may affect multiple body parts at different times often suggests a non-organic etiology.

The temporal profile of migraine, cluster, and nonspecific chronic or recurrent headaches will be reviewed in more detail in the presentation. The brief discussion of headache entities below will focus on issues which are important in children and adolescents.

"Traction headaches" refer to pain produced by intracranial lesions by mechanisms other than intracranial pressure. An intact carotid aneurysm in the cavernous sinus, for example, may cause retro-orbital pain. There is probably overlap between headaches due to inflammation and those due to traction on dura or vessels.

A variety of drugs may be associated with headache in children and adolescents; these include corticosteroids, vitamin A (isoretinoin), psychostimulant medications given for ADHD, and sulfasalazine (Azulfadine).

Headaches of metabolic origin usually occur in association with obvious systemic illness. Patients in chronic renal failure, for example, often experience headache. Headaches due to toxic exposure appear to be uncommon. Chronic, low grade exposure to carbon monoxide may produce headache as one of its symptoms.

The terms "muscle contraction headache" and "tension" headache are often used to refer to the same set of patients, whose complaints are chronic and nonspecific and in whom diagnostic studies are unrevealing. The terms imply both that scalp muscles play a role in the cause of these headaches and that emotional "tension" is a contributory factor. Clear evidence that scalp muscle contraction contributes to headache is not found in most patients, however, and chronic headache is not uncommon in children and adolescents who seem well adjusted at home and in school. Chronic sleep deprivation is probably a factor in some adolescents who are very devoted students or get up early to participate in sports. Obstructive sleep apnea, which produces sleep deprivation, should always be considered as a cause of headache which is not otherwise explained; mouth breathing, obesity, daytime hypersomnia, and a history of snoring suggest the diagnosis. Nonspecific, chronic headache is probably the most common variety encountered in pediatric as well as adult neurology practice and is the most difficult to treat. Therapeutic options will be discussed below.

The most common vascular headache is probably headache associated with fever. Headache is also a very common symptom following a seizure, and presumably reflects transient alterations in circulatory dynamics. Headache is a common complaint in children with severe, acute hypertension. MRI scans of patients with hypertensive encephalopathy often show symmetrical, superficial areas of increased signal intensity suggesting ischemia, edema or both. Whether brain edema with dural traction or changes in cerebral blood vessel walls are the cause of head pain in hypertensive patients is apparently unknown. Headache is occasionally a symptom of patients with severe anemia and may occur as the result of compensatory vasodilatation.

Migraine classification for children is identical to that in adults although the incidence of the different subtypes varies by age group. In addition, migraine is probably underdiagnosed in younger children who cannot fully, describe their symptoms. Migraine is estimated to occur in about 4% of school children. In a large Swedish study, the prevalence was 1.4% at 7 years of age and 5 3% at 15 years of age. For this study, Bille defined migraine as recurrent headache with at least 3 of the following 5 features: paroxysmal onset, hemicranial pain, nausea and vomiting, visual aura, family history of recurrent headache. As his criteria imply, many children with otherwise typical migraine do not report that their headaches are lateralized. Prior to 10 years of age, migraine was equally frequent in boys and girls. Beginning at 11 years of age, there was a gradual increase in the female: male ratio. Bille performed his initial study in 1955 and recently reported a 40 year follow-up of 7 3 children with severe migraine and an average age of onset of 6 years. Twenty-three percent of the children were free of migraine by 25 years of age, and the prognosis was more favorable in boys than in girls. However, at age 50, more than half of the subjects still reported migraine attacks. Fifty-two percent of the subjects who became parents had at least one child with recurrent headache.

"Common migraine" is a term which refers to patients with recurrent, lateralized (and, therefore, presumably vascular) headache without the associated neurological manifestations which are part of classic migraine. The headache may be of longer duration than in classic migraine. Nausea and vomiting are variable features. In classic migraine, visual symptomatology precedes or accompanies head pain and occasionally occurs as an isolated symptom in the absence of headache. Many patients with headache complain of nonspecific visual blurring almost as an afterthought. In classic migraine, however, visual symptoms are prominent and may include frank visual loss, scintillating scotomata, or a zigzag pattern described in older literature as fortification spectra. Children who remember, their migraines will sometimes draw their visual auras. Classic migraine episodes seldom last more than a few hours. The typical progression is visual disturbance----pain----vomiting-----sleep. Pain does not usually persist when the patient awakens. Occasional patients are ravenous on awakening. "Complicated migraine" refers to recurrent headaches associated with neurological symptoms not referable to the occipital cortex. Variants include hemiplegic, ophthalmoplegic, ocular (retinal), and basilar artery migraine. All of these forms may begin in childhood and apparently become less common with maturation. "Basilar migraine," a term used to describe patients whose headaches are accompanied by ataxia and other neurological symptoms referable to the posterior fossa, was originally described in adolescent girls by Bickerstaff. Hemiplegic migraine is sometimes familiar but other forms of complicated migraine are almost always sporadic. Migraine with hemisensory alterations (for example the face and hand) has been more common than hemiplegic migraine in my experience. Patients who are interviewed days or weeks after a headache sometimes confuse motor and sensory phenomena.

A number of conditions have been described as migraine variants or equivalents. These include acute confusional states, transient global amnesia; perceptual alterations such as micropsia, macropsia, and telecopsia, cyclic vomiting and benign paroxysmal vertigo. Motion sickness is more common in children with migraine than in the general population. The incidence of epilepsy is increased in patients with migraine, and migraine is more common in those with epilepsy than in the general population. Epileptiform EEG abnormalities are occasionally found in patients with migraine, and children with occipital spike foci sometimes have spells which are consistent with either migraine or epilepsy. Rolandic spikes, which are usually considered the hallmark of "benign epilepsy of childhood" (Rolandic, epilepsy) have been found in 9% of children with migraine. The most specific EEG finding in migraine (but one which is rarely demonstrated) is probably a transient, occipital delta focus, which is present during and for a short time after an episode. I am not aware of any large study which has evaluated serial EEGs in a large group of patients with migraine and other headache types in a blinded fashion with inter-rater reliability criteria.

A variety of pathophysiological mechanisms have been proposed to explain migraine. These include the spreading depression of Leao, alterations in cerebrovascular reactivity, and a disturbance of neurotransmitter metabolism (serotonin and, more recently, dopamine). The vascular hypothesis suggests that vasoconstriction is responsible for the neurological manifestations in complicated migraine and implies that use of vasoconstrictive agents such as ergotamine might place affected individuals at greater danger of an irreversible neurological deficit. The trigeminovascular hypothesis, which will be briefly discussed in the presentation, provides explanations for a number of the characteristic features of migraine. Food allergy has often been suggested as a cause of migraine, but an allergic mechanism has not been convincingly demonstrated. Rigorous studies to demonstrate a key role of food allergy in migraine would be difficult to devise.

Cluster headache is very rare in pediatric neurological practice. Individual headaches resemble migraine in that they may be lateralized. However, pain is more intense than in migraine and does not allow the patient to fall asleep. Ipsilateral lacrimation and nasal stuffiness during the headache are almost pathognomonic of cluster. As the designation implies, headaches may recur over several days or weeks and the patient may then be symptom free for months or years.

All children with headache merit a complete physical examination and a neurological examination. The general physical examination should include blood pressure measurement, gentle percussion over the paranasal sinuses and cervical spine to elicit tenderness, palpation of the scalp, passive movement of the head, and auscultation of the head and neck for bruits. Neuroimaging studies are seldom informative if the history or physical examination does not suggest an intracranial lesion. It can be argued that a normal CT scan of the head has therapeutic value and reduces the total cost of evaluation and treatment in some children. Most parents do not object to initiation of treatment or observation without neuroimaging studies provided that the physician arranges for appropriate follow-up and indicates a willingness to obtain a CT or NM scan at a later date if symptoms do not resolve. Most patients referred to a pediatric neurologist because of headache require more than one visit. It is useful to provide the patient or parent with a headache log to be maintained for the one to two months between the first and second office visits. The log sometimes demonstrates that a pattern of headache which can easily be explained by the patient's daily routine. It may also reveal that headaches are much less frequent and disabling than the patient or parent recalled during the initial visit. Lastly, the headache log permits a more rational decision concerning the need for prophylactic medication.

Therapy
Behavioral and Psychological Therapy

Various forms of relaxation therapy including biofeedback may be useful for properly selected patients. Most physicians are not trained in such therapy and would not be anxious to personally administer it. It is difficult for physicians to become familiar with the quality of services provided by psychologists or other nonmedical personnel having an interest in headache management and to use these services appropriately. Compared to medication, these "hands on" therapies may be relatively expensive. Psychological counseling is obviously indicated when emotional difficulties or family conflict are thought to lie the major factor responsible for the patient's complaints.

Drug Therapy for Treatment and Prevention of Headstche

Drugs and dosages for the treatment of migraine in childhood are given in the accompanying table. For the agents most commonly utilized, treatment or prevention of migraine or other forms of headache in children is not a listed indication. Therefore, it is appropriate for the prescribing physician to explain to patients and parents, who often read the Physician's Desk Reference or "surf the net" after leaving the physician's office, the rationale for using a medication for an "off-label" application. Large, adequately controlled clinical in various age groups have not been performed for some of the commonly used preparations. Drugs which cause vasoconstriction (such as dihydroergotamine and sumatriptan) should generally be avoided in patients with complicated migraine--i.e., in migraine accompanied by a neurological deficit other than a visual aura. In the absence of adequate, prospective, randomized trials, appropriate dosage in children is often inferred from one or two case series. Ergotamine preparations are generally avoided in young children because of the possibility of ischemic injury.

In my experience, cyproheptadine has not been of substantial benefit in preventing migraine and produces excessive fatigue in most children.

It is recommended that the initial close of sumatriptan, especially by the SC route, be administered in a medical facility. Dosage limits should be carefully reviewed with patients and parents whenever vasoactive drugs are prescribed. Sumatriptan is a particularly useful drug for the acute migraine attack because of its availability in multiple dosages and for three routes of administration (SC, oral, nasal). Zolmatriptan has recently become available, and other triptans, which will vary in terms of speed of onset, duration of effect, and relative incidence of various side effects, will be available over the next few years.

Combination headache remedies containing small amounts of butalbital have produced addictive behavior in a small proportion of individuals. Both patients and parents, who may use the patient's medicine for their own chronic headaches, are at risk.

Butorphanol (Stadol), which is available as a nasal spray for migraine, also carries a risk of dependence and addiction. This drug was rescheduled as a controlled substance through the effort of the American Academy of Neurology in conjunction with Dr. Morris Fisher, a professor in the Department of Neurology at Loyola University, Stritch School of Medicine. Dr. Fisher's son committed suicide during treatment for his Stadol addiction. I have had at least one predadolescent patient who used nasal butorphanol excessively and no longer prescribe this drug for migraine in children.

Over-the-counter (OTC) medications are not always adequately exploited before prescription drugs are used. There are financial and psychological advantages in using OTC medications when they are effective. ASA-containing combination medications should probably be avoided in children during febrile illness, but, to my knowledge, it has not been established that use of such drugs for headache in the afebrile patient with migraine or muscle contraction headache carries an unacceptable risk.

Annotated References

AAN Public Relations Efforts Result in Restrictions on Migraine Drug. AAN News 1997-, Vol 10, Issue 8, pp 1-2.
Discussion of restrictions placed on butorphanol (Stadol).

Barry J, von Baeyer CL. Brief cognitive-behavioral group treatment for children's headache. Clin J Pain 1997;13:215-220.
Benefits of a brief course of cognitive-behavioral therapy for children with headache is uncertain.

Bille B. A 40 year follow-up of school children with migraine. Cephalaigia 1997; 17:488-491.
An extraordinarily long term follow-up of pediatric migraine patients in Sweden. Bille's original articles which is referenced in the current study, provided an important insight into the epidemiology and clinical manifestations of childhood migraine.

Hamalainen ML, Hoppu K, Valkeda E, Santavuori P. Ibuprofen or acetaminophen for the acute treatment of migraine in children: double-blind, randomized, placebo-controlled, crossover study. Neurology 1997;48:103-107.
Finnish study demonstrating both drugs are more effective than placebo for severe or moderate migraine in children. Ibuprofen was more effective than acetaminophen,

Linder CL. Treatment of childhood headache with dihydroergotamine mesylate. Headache 1994; 34:578-80.
Dose of DUE in this study ranged from 0.1 to 0.5 mg IV and was effective in 80% of patients when given in conjunction with oral metoclopramide. However, most patients received multiple doses.

Lipton RB. Diagnosis and epidemiology of pediatric migraine. Curr Opin Neurol 1997; 10:231-236.
Discusses differences between pediatric and adult migraine and gaps in our understanding of the condition in children.

Rothner AD. Headaches in Children and Adolescents. Seminars in Pediatric Neurology, Vol 2, No2, 1995.
This multi-authored issue of Seminars in Pediatric Neurology contains useful reviews of multiple issues in pediatric headache including classification, screening and evaluation, childhood migraine, associated neurological disorders, pharmacology of commonly used headache medications

Ryan R, Elkind A, Baker CC, Mullican W, DeBussey S, Asgharaejad M. Sumatriptan nasal spray for the acute treatment of migraine. Results of two clinical studies. Neurology 1997;49:1225-1230.
Current review of sumatriptan nasal spray for migraine.

Winner PK. Headaches in children. When is a complete diagnostic workup indicated? Postgrad Med 1997; 5:85-81,89-90.
Review article.

Young WB. Appropriate use of ergotamine tartrate and dihydroergotamine in the treatment of migraine: current perspectives. Headache 1997; 37 Suppl 1:S42-45.
Current guidelines for the use of ergotamine and dihydroergotamine as promulgated by the Headache and Facial Pain Section of the Quality Standards Subcommittee of the American Academy of Neurology

Ziegler DK, Schwertfeger TL, Murrow RW. Headache. Chap 13 in Joynt RJ and Griggs RC: Clinical Neurology, V012, Lippincott-Raven Publishers, NY, 1997.
General review of neurological aspects of headache with an adult focus. Approx 35 pp.