Michel Volcy Gomez MD
Stewart J. Tepper MD
Headache disorders constitute a major public health problem. In the United States, the annual expense due to migraine-related loss of productivity is estimated between $5.6 billion and $17.2 billion. Migraines are responsible for approximately 36 million days of bed rest per year and an additional 21.5 million days of restricted activity. Direct costs to the medical system are also high; the total health care cost for patients with migraine averages of $145 per month, compared with $89 per month for persons who do not have migraine. Persons with migraine use 2.5 times the prescription drugs and 6 times the diagnostic services of persons without migraine. Nationally, direct costs are estimated to be as high as $9.5 billion, with $2.7 billion spent for prescription drugs and $730 million for hospital care. Headache represents the fourth most common cause of emergency department visits.
The International Headache Society (IHS) classifies headache into 13 main groups. The second International Classification of Headache Disorders (ICHD-II) recognizes two basic subgroups: primary headache syndromes and secondary headache syndromes (Table 12-1 and Table 12-2). Primary headaches are those without any intracranial structural pathology. The primary headaches can be divided into three main groups: migraine, tension-type headache, and trigeminal autonomic cephalalgias. The secondary headaches are symptoms of an identifiable disorder.
Essentials of Diagnosis
The prevalence of migraine is age and gender dependent. It is estimated that the 1-year migraine prevalence in Western countries is around 10 to 12%, 6% in men and 15 to 18% in women. Age at onset of migraine is earlier in boys than in girls, but prevalence quickly equalizes until menarche, when female predominance begins. Migraine is two to three times more common in women than in men, with peak prevalence occurring during mid-life in both sexes. Migraine prevalence is higher in whites than in blacks or Asians. Significant associations (comorbidities) have been reported between migraine and certain psychiatric disorders (anxiety, depression), stroke in women under the age of 45 years, and epilepsy.
Successful treatment begins with accurate diagnosis. Well-defined criteria for diagnosing headache published by the IHS have been available for nearly 20 years. The classification scheme has been well suited for research. However, clinically, the criteria are not always used or are misapplied. Many clinicians mistakenly define migraine by a single feature, such as unilaterality or throbbing quality or by the presence or absence of aura. Less than 20% of patients with migraine demonstrate typical aura, a reversible neurologic event lasting longer than 5 minutes but less than 1 hour and followed within 1 hour by the headache.
It has been suggested that the IHS migraine criteria lack accurate sensitivity and specificity. The most sensitive and specific symptom for diagnosis is nausea (sensitivity 82%, specificity 96%), then photophobia (sensitivity 79%, specificity 87%), phonophobia (sensitivity 69%, specificity 87%), and worsening with physical activity (sensitivity 81%, specificity 78%), with approximately 66% of migraine headaches described as unilateral (specificity 78%), and 76% as throbbing (specificity 77%). Recently, osmophobia and taste abnormalities were demonstrated to be very specific (86.7% and 90.2%, respectively) in diagnosing migraine but very insensitive (around 20%).
Table 12-1. Diagnostic Characteristics of Primary Headaches.
Table 12-2. Causes of Secondary Headaches.
Headache experts consider that a stable pattern (>6 months) of episodic, disabling headaches with a normal physical examination is likely to be migraine. Ninety-four percent of patients with a physician diagnosis of nonmigraine primary headache or a new physician diagnosis of migraine at a primary care clinic were determined on the basis of longitudinal diary data to have migraine-type headaches. However, the diagnosis of nonmigraine headache made at a clinic visit is usually inaccurate: 82% of patients with a clinic diagnosis of nonmigraine primary headache were found to have migraine-type headaches when they kept diaries for six attacks, which were reviewed for ICHD diagnoses.
In order to avoid migraine misdiagnosis, it is important to gather information about multiple attacks. Misdiagnosis of migraine is frequently related to (1) a patient's failure to report typical features during a single
encounter, (2) self-assigned diagnoses (that is, the patient reports what he or she thinks the diagnosis is), or (3) multiple headache attack types over time. Location of pain, which is not considered in the IHS criteria, often leads to misdiagnosis; most of what is diagnosed as sinus headache is actually migraine. Patients with migraine manifest a wide variety of headache attack types over time, including migraine (with or without aura) and tension-type headaches, often masking proper diagnosis.
The ICHD-II criteria still require separate diagnosis of every headache type in any patient. However, evidence exists that tension-type headache is different in patients with migraine than in those who never manifest migraine; in patients with migraine, the triptans relieve every headache type, including those that look like tension-type headaches. For persons with only episodic tension headache, the triptans are not effective.
In the revision of the ICHD published in 2004, the term “probable migraine” now replaces “migrainous disorder” to describe headache that is missing one criterion for migraine. Migraine or probable migraine occurring on at least 8 days per month in a patient with at least 15 days of headache per month for at least 3 months is chronic migraine. In the new headache classification, the diagnosis of chronic migraine should not be made when medications for acute headaches are overused. For these patients, the diagnosis of medication overuse headache is more appropriate.
Migraine may be initiated in an upper brainstem migraine generator (periaqueductal gray, dorsal raphe nucleus) that activates trigeminal efferents to the meninges where inflammation and vasodilation occur. This neurogenic sterile meningitis in turn sensitizes pain afferents, also trigeminal, which convey the nociceptive signal back to the lower brainstem (trigeminal nucleus caudalis) where it is integrated. Following the return of the pain signal centrally, there is activation of adjacent autonomic nuclei (eg, nucleus tractus solitarius) to account for nausea, and then an ascent up to thalamus and cortex. The putative migraine generator in the midbrain periaqueductal gray region has been imaged in functional imaging studies. During migraine, compared with a headache-free period, blood flow increased in the contralateral periaqueductal gray/dorsal raphe nucleus region of the brainstem, even after complete headache relief with a triptan. Iron homeostasis abnormalities have also been found in this region, with increased levels of nonheme iron in tissue from patients with episodic and chronic migraine compared with age-matched, headache-free controls. Thus, the periaqueductal gray/dorsal raphe nucleus, a major source of serotonergic input to the brainstem that also projects to the cortex and is an integral part of the endogenous pain modulatory system, may be the site of migraine generation.
There is also evidence that cortical function is different in persons with migraine. Patients with migraine have reduced habituation to different sensory modalities, they perform tasks that require low-level visual processing faster, and the threshold to produce phosphenes with transcranial magnetic stimulation is lower in those with migraine with aura.
The aura of migraine has been hypothesized to reflect a process related to cortical spreading depression. Cortical spreading depression is misnamed; it is a process in which the cortex is activated, with intense hyperemia, and the activation spreads forward at a rate of 3 to 4 mm/min. The wave of neuronal activation is associated with the aura symptoms and is followed by quiescent neurons analogous to postictal neurons. During this quiescent period, there is associated oligemia, without ischemia. Thus, aura is neuronal, not vascular.
Aura, when it occurs, is anatomically linked to the activation of the trigeminovascular system and the migraine generator. One concept that explains how an aura can occur in the absence of pain, or with a lesser headache than migraine, is that the aura reaches a variable threshold, necessary to trigger the generator and associated pain.
The phenomenon of sensitization on primary pain afferents refers to the presence of enhanced responses to stimuli. The meningeal primary afferents become hypersensitive to mechanical stimulation so that those stimuli normally producing little or no activity now evoke much larger responses. Peripheral sensitization is proposed to underlie the throbbing pain of headache. In addition, second-order neurons receiving input from sensitized primary afferents and skin also become sensitized. This alteration of central neuronal response is called central sensitization and provides a physiologic basis for scalp tenderness and other migraine symptoms (allodynia).
Cutaneous allodynia, the experience of touch as painful, is a marker of central sensitization, and can be demonstrated easily in patients with migraine. It has been found that ipsilateral allodynia develops, especially at the headache peak, in almost 75% of persons with migraine who were tested. The presence of allodynia correlates with illness duration and attack frequency. Reduced efficacy of triptan therapy occurs in the presence of cutaneous allodynia, suggesting the need for early treatment in migraine before central sensitization develops.
The objective of treating an acute migraine is to restore normal function by rapidly and consistently alleviating the patient's head pain and the associated symptoms of nausea, vomiting, and sensory phobias without side
effects and without attack recurrence. Several drug options and different formulations are available to treat acute migraine. From the medical point of view, the choice of a specific medication type has depended on individual characteristics such as headache intensity, time-to-peak intensity, speed of onset of action of medication, presence of associated symptoms, degree of incapacitation, and patient response. For patients, the most desired qualities for treatment are pain freedom, speed of onset, no headache recurrence, tolerability, medication availability, cost, and overall satisfaction or well-being after taking the medication.
The nonspecific treatments for migraine attacks are aspirin, acetaminophen (and other simple analgesics), nonsteroidal anti-inflammatory drugs (NSAIDs), neuroleptics, opioids, combination analgesics, and short-acting barbiturates. Triptans and ergots are considered the specific migraine treatments. Monotherapy using serotonin (5-HT) subreceptor agonists (5-HT1B/5-HT1D), especially when administered orally, does not always result in rapid, consistent, and complete relief of all migraine attacks as desired by patients. In addition, the efficacy of some nonspecific migraine agents, such as various NSAIDs, has been shown in many trials and cannot be called into question (Table 12-3).
The paradigm for treating an acute migraine is to start therapy early in its onset. This may prevent the pain from becoming moderate or severe. Migraine progresses on a time and intensity continuum, and more than 70% of untreated, mild migraines eventually become moderate or severe in patients with histories of recurrent disabling migraine. As pain builds, there is increased central trigeminal neuronal sensitivity. In addition, gastroparesis, which is associated with migraine, is thought to impair absorption of oral medication. Thus, the earlier mild migraine pain is treated with an oral triptan, before gastroparesis is established, the more complete the absorption and the greater the probability of response will be.
The occurrence of adverse events associated with trip-tans in general, and central nervous system side effects in particular, may lead to a delay in initiating or even avoiding an otherwise effective treatment. These side effects are lessened by early administration of the triptans, another reason for early treatment. Finally, there is evidence linking a pain-free response (that is, zero pain, as opposed to just pain relief) with a reduced recurrence of headache and therefore a reduced number of treatments in a given attack. Early treatment is linked to higher pain-free response as well.
Drugs in this class are the treatment of choice for acute disabling migraine, in the absence of vascular disease. The decision about which triptan to prescribe is often based on the relative need for speed, the formulation needed, and the formulary availability for a given patient. If oral treatment is to be used, it is clear that while seemingly a homogenous group of drugs, there are significant differences in speed of onset, efficacy, recurrence rates, and tolerability among oral triptans.
Triptans relieve migraine symptoms via three possible mechanisms of action: (1) selective intracranial/ extracerebral meningeal vasoconstriction; (2) presynaptic inhibition of vasoactive inflammatory neuropeptides release in the meninges; (3) inhibition of the return of pain signals to the trigeminal-cervical complex neurons in the brainstem and upper cervical column. Triptans stimulate 5-HT1B receptors, located predominantly on cranial blood vessels, causing vasoconstriction counteracting neurogenic vasodilation. Triptans also stimulate 5-HT1D receptors located on peripheral and central trigeminal nociceptive nerve terminals, inhibiting nociceptive transmission from pain-sensitive meningeal structures.
In addition, triptans block vasoactive and proinflammatory neuropeptide release through peripheral meningeal 5-HT1D-receptor stimulation and may also interact with 5-HT1B/1D/1F receptors on central neurons, although this remains to be defined.
Headache recurrence can occur with any type of urgent antimigraine treatment and is defined as the return of a headache after successful treatment, generally within 24 hours. When recurrence occurs, the recurrence time for the triptans is generally around 12 hours. The incidence of headache recurrence varies among drugs in the triptan class. Migraine recurrence does not appear to be related to initial clinical efficacy but is influenced by the pharmacologic and pharmacokinetic properties of the individual triptans (Table 12-4). The triptans with longer half-life and greater 5-HT1B-receptor potency may have the lowest rates of headache recurrence, although this remains relatively unproven and controversial. One explanation for recurrence might be that the migraine generator is still active in spite of apparent symptomatic relief.
The most common central nervous system treatment-related side effects with triptans are fatigue and somnolence, dizziness, and asthenia. These side effects are rare and may be related to dosage. Eletriptan, rizatriptan, and zolmitriptan have active metabolites, while lipophilicity is lowest for almotriptan and sumatriptan; however, it is not clear that these differences have played out in clear clinical differences in large populations. While population differences are small when triptans are compared, individual preferences for particular triptans can be large.
Triptans are not widely used in clinical practice despite their well-established efficacy. Although the relatively restricted use of triptans may be attributed to several factors, the most important concern has been cardiovascular safety. Triptans all bind to 5-HT1B receptors, causing vasoconstriction, and most people have some 5-HT1B receptors on coronary arteries, although far fewer than on meningeal arteries. The average incidence of coronary vasoconstriction due to triptans is 10 to 20%, generally not clinically significant in the absence of atherosclerotic coronary artery disease. Triptans cannot be easily differentiated with respect to their effects on human isolated coronary arteries; clinical doses all drugs in this class contract arteries to about the same extent, and the US Food and Drug Administration has placed identical template warnings in the prescribing information for all seven triptans contraindicating their use in patients with vascular disease.
Table 12-3. Abortive Medications for Migraine.a
Table 12-4. Pharmacologic Characteristics of Triptans.
It has been suggested that while caution is advisable for patients with two or more cardiac risk factors (ie, hypertension, diabetes, obesity, age over 40 years in men, post-menopausal women, premature family history of vascular disease, or smoking), triptans can be prescribed without concern among low-risk patients, probably the majority of patients with migraine. Triptans should not be withheld from patients who have either no risk factors or only one risk factor.
The incidence of triptan-associated serious cardiovascular adverse events in both clinical trials and clinical practice appears to be extremely low. Patients with a family history of early atherosclerosis or patients with multiple risk factors are more likely to have endothelial changes predisposing to vasospasm, and initial treatment should be monitored. Obviously, triptan therapy should be with held from patients with symptomatic or known obstructive coronary disease, and patients at intermediate or high risk should be evaluated prior to treatment. A functional evaluation to exclude coronary disease would suffice in most cases, with further diagnostic investigations only in those patients with positive findings.
Chest symptoms are a rare adverse effect (1 to 4%) unrelated to coronary vasoconstriction in most patients. Although the etiology of chest symptoms remains to be fully elucidated, they could be related to pulmonary vasoconstriction, esophageal abnormalities (increased amplitude and duration of esophageal contractions, abnormal esophageal motility), triptan-associated reductions in the oxygen stores of skeletal muscles, or heightened sensory sensitivity. Overall, the incidence of these symptoms also appears to be dose-related.
Triptans can be divided into fast onset and slow onset (Table 12-5). The fast onset triptans are sumatriptan, zolmitriptan, rizatriptan, almotriptan, and eletriptan. The slow onset triptans are naratriptan and frovatrip-tan. The slow onset oral triptans take twice as long to work as the fast onset triptans, showing responses at 4 hours comparable to what the fast onset triptans show at 2 hours. Accordingly, the slow onset triptans work on fewer patients, but in general are associated with greater tolerability.
Triptans can be divided by formulation as well. Sumatriptan is available as a subcutaneous injection, nasal spray, and rapid release tablet. Zolmitriptan is available as a conventional tablet, orally dissolvable tablet, and nasal spray. Rizatriptan is available as a conventional and orally dissolvable tablet. The other 4 triptans are available only as conventional tablets.
Table 12-5. Formulations of Fast- and Slow-acting Triptans.
Metabolic degradation pathways can also distinguish triptans, which can predict potential drug-drug interactions (see Table 12-4). Triptans with predominantly monoamine oxidase (MAO) degradation pathways (sumatriptan, rizatriptan, almotriptan, and zolmitriptan) should not be used in patients taking MAO inhibitors. Eletriptan, with predominant cytochrome P450 3A4 degradation, should not be used with potent inhibitors of that system (eg, erythromycin, clarithromycin, fluconazole, ketoconazole, verapamil, ritonovir and other similar AIDS drugs).
There is an interaction between propranolol and rizatriptan in which the propranolol raises the rizatriptan serum level. Thus, only the 5-mg dose of rizatriptan should be used with propranolol. The same interaction does not occur between rizatriptan and other β-blockers, or between other triptans and propranolol.
These medications have a high affinity for 5-HT1A, 5-HT1B, 5-HT1D, and 5-HT1F, 5-HT2A, 5-HT2C, and 5-HT4receptor subtypes, and low affinity for the 5-HT1Ereceptor. In addition, they bind to receptors in the adrenergic and dopaminergic systems. It is likely that the beneficial
effects of ergotamine and dihydroergotamine arise from their agonist properties at 5-HT1B, 5-HT1D, and perhaps 5-HT1F receptors that, as with the triptans, lead to meningeal vasoconstriction and trigeminal inhibition.
The unwanted side effects of ergotamine and dihydroergotamine probably arise from actions at central 5-HT1A receptors (nausea and dysphoria) and dopamine D2 receptors (nausea and vomiting). Both dihydroergotamine and ergotamine can cause vasoconstriction and venoconstriction by stimulating α-adrenergic and 5-HT2A receptors. These peripheral vascular effects are more pronounced with ergotamine than with the trip-tans, since the triptans do not have activity at adrenergic and 5-HT2A receptors. Ergotamine and dihydroergotamine can also constrict coronary blood vessels through actions at coronary artery smooth muscle 5-HT1B and 5-HT2A receptors. The contractile response in coronary arteries is more prolonged with ergots than with triptans.
Ergotamine (orally/rectally, and caffeine combination) may be considered in the treatment of selected patients with moderate to severe migraine. Nausea and vomiting are the most commonly observed short-term adverse effects associated with ergot use. The combination of ergotamine plus metoclopramide reduces the incidence of nausea and vomiting compared with ergotamine alone. Patients with very long attacks or with frequent headache recurrence may be especially suited for treatment with ergotamine because headache recurrence is probably less likely with ergotamine.
Dihydroergotamine SC/IV/IM and nasal spray may be given to patients with nausea and vomiting and can be a reasonable initial treatment choice when the headache is moderate to severe, or in migraine of any severity when nonspecific medication has failed. Parenteral antinauseants are often used with it as adjunctive symptomatic treatments. Clinical opinion suggests that subcutaneous dihydroergotamine is relatively safe and effective when compared with other migraine therapies, and subcutaneous dihydroergotamine has fewer adverse events than when delivered intravenously. Dihydroergotamine can also be used parenterally for menstrual migraine prophylaxis and can be used for cluster headache prevention. Several of the metabolites of ergotamine and dihydroergotamine have biologic activity similar to the parent drugs and are often present in concentrations several times higher. Also, ergotamine and dihydroergotamine are strongly sequestered in tissues, which could contribute to the persistence of their biologic effects.
The decision about whether to use a nonspecific treatment or a selective 5-HT1 B/1 D agonist, the specific therapies for acute migraine, despite clinical practice experience supporting the better effectiveness of the triptans, remains controversial. Clinical trials do not always reflect the favorable clinical experience with triptans, and direct comparisons made between triptans and some nonspecific treatments (such as NSAIDs) do not always favor triptans. Nevertheless, headache-experienced physicians have found that, in general, triptans provide vastly superior efficacy in comparison to nonspecific agents. Combining specific and nonspecific therapies, in at least a subset of patients, provides additional benefit, including greater efficacy and reduced recurrence when NSAIDs are taken with triptans.
Daily scheduled opioids have been used in some instances to treat intractable headache. However, a long-term study of their effectiveness found a lack of benefit in over 75 % of patients with chronic headache who received opioids. Seventy-four percent of those treated either did not show significant improvement or were discontinued from the program for clinical reasons. The relatively low percentage of patients with demonstrated efficacy and unexpectedly high prevalence of misuse (50%) due to dose violations, lost prescriptions, and multisourcing discourage use of opioids in chronic headache patients.
The purpose of migraine prevention is to reduce the frequency of attacks, modify their severity and impact, and improve the efficacy of abortive therapy. The ultimate goal of migraine prophylaxis is to improve sufferers′ quality of life and ameliorate their migraine-related disability. Consequently, migraine preventive strategies should be efficacious, safe, and well tolerated.
Daily migraine preventive therapy is indicated when the following circumstances are present:
Also, daily drug therapy is often recommended in the setting of very frequent headaches or a pattern of increasing attacks over time, with the risk of developing transformation into daily headache with medication overuse.
An effective preventive treatment plan can be achieved by establishing a correct diagnosis; assessing the overall impact of the attacks; creating the plan with the patient in a therapeutic alliance; emphasizing drug dosage, therapeutic effects, and side effects; and describing expectations explicitly. As much as possible, patients should be empowered with their own care.
The preventive medications for migraine can be divided into five major categories:
It is preferable, when possible, to select a particular prophylactic drug based on evidence-based principles, with a balance between therapeutic effects and side effect potential (therapies may lose their effectiveness when poorly tolerated because patients may become non-compliant), and guided by comorbid conditions when present. By carefully selecting preventive medication, potential drug-drug interactions will be minimized and the rate and extent of side effects will be decreased.
β-Blockers, the most widely used prophylactic class of drugs in migraine prevention, are up to 60 to 80% effective in producing at least a 50% reduction in attack frequency. Propranolol, the most frequently used β-blocker, has multiple actions contributing to its effectiveness. First, it inhibits norepinephrine release through a β1-mediated agonist action, reducing central catecholaminergic hyperactivity. Second, propranolol antagonizes 5-HT1A and 5-HT2Breceptors, reducing neuronal excitability. Third, propranolol inhibits nitric oxide production by blocking inducible nitric oxide synthase, through β2-agonist action. Nitric oxide is believed to be the common final pathway for vasodilation in migraine. Fourth, propranolol inhibits excitatory glutamate receptors, decreasing neuronal activity. Finally, propranolol has membrane-stabilizing properties. β-blockers with intrinsic sympathomimetic activity (acebutolol, alprenolol, oxprenolol, pindolol) have not been found to be effective for migraine prevention.
The usual dose of propranolol is 160 mg/d. There are randomized controlled studies establishing the effectiveness for timolol (dose range 20 to 40 mg/d), metoprolol (50 to 100 mg/d), and nadolol (40 to 80 mg/d). Both propranolol and timolol are approved by the US Food and Drug Administration in the prevention of migraine.
The relative efficacy of the different β-blockers has not been clearly established, and most studies show no significant difference between drugs. One trial comparing propranolol to amitriptyline suggested that propranolol is more effective in migraine alone, and amitriptyline is superior for patients with both migraine and tension-type headache.
All β-blockers can produce central nervous system side effects, such as fatigue, sleep disorders, and depression. Another common side effect is decreased exercise tolerance. Less common are orthostatic hypotension, significant bradycardia, and impotence. Congestive heart failure, asthma, and insulin-dependent diabetes are contraindications to the use of nonselective β-blockers.
Amitriptyline, a tricyclic antidepressant (TCA), is a mixed serotonin-norepinephrine re-uptake inhibitor (SNRI). In addition, amitriptyline is a sodium channel blocker. Others SNRI antidepressants (eg, imipramine, venlafaxine, duloxetine) may have a potential role in migraine prevention, although they have not been extensively studied. Side effects are common with TCA use. The anticholinergic antimuscarinic adverse effects are most common; however, overdose adverse effects are related to antihistaminic and α-adrenergic overactivity and cardiac toxicity, and orthostatic hypotension can occur. Anticholinergic side effects include dry mouth, constipation, dizziness, mental confusion, tachycardia, blurred vision, and urinary retention. Antihistaminic activity may be responsible for weight gain. Any antidepressant treatment may change depression to hypomania or frank mania in patients with bipolar disorder. Older patients are more vulnerable to anticholinergic side effects.
The dose range for amitriptyline in migraine prevention is 25 to 100 mg. Consensus also suggests effectiveness for nortriptyline in the same dose range.
Evidence for the use of selective serotonin reuptake inhibitors (SSRIs) is poor. They may be helpful in patients with comorbid depression because of their better tolerability profile, but their efficacy in reducing the frequency of episodic migraines is not established. The most common side effects for SSRIs include insomnia, sweating, and sexual dysfunction. The combination of an SSRI and a TCA can be beneficial in treating refractory depression but may require downward dose adjustment of the TCA because levels may significantly increase.
MAO inhibitors are believed to be effective in migraine prevention by expert consensus opinion, but there are no randomized controlled trials. The difficulties in using MAO inhibitors are the requirement for a special exclusion diet and the need for medication restrictions to avoid tyramine-containing products or adrenergic drugs, both of which can precipitate a so-called “cheese effect” or hypertensive crisis. For this reason, MAO inhibitor use in migraine prevention is infrequent. The most common side effects of MAO inhibitors include insomnia, orthostatic hypotension, constipation, increased perspiration, weight gain, peripheral edema. Less common side effects include inhibition of ejaculation, anorgasmia, or reduced libido. The dose range for phenelzine is 45 to 60 mg/d.
Table 12-6. Preventive Medications for Migraines.
The mechanism of action of these drugs in migraine prevention is uncertain. Probably, they block 5-HT release, interfere with neurovascular inflammation, or interfere with the initiation and propagation of the cortical spreading depression of aura. Certain forms of aura are associated with inherited calcium channelopathies (eg, familial hemiplegic migraine). These rare forms of migraine respond well to calcium channel antagonist prophylaxis. The evidence for their effectiveness in conventional migraine is not robust. Side effects of the calcium channel antagonists depend on the drug and include dizziness and headache (particularly with nifedipine), depression, vasomotor changes, tremor, gastrointestinal complaints (including constipation), peripheral edema, orthostatic hypotension, and bradycardia with other types. The dose range for verapamil inmigraine prevention is 240 to 480 mg/d and for amlodipine, it is 5 to 10 mg/d.
Anticonvulsants or antiepilepsy drugs (AEDs) suppress neuronal activity through different mechanisms. Several neuromodulators possess antimigraine and antipain properties, including valproate, topiramate, gabapentin, levetiracetam, and zonisamide.
Valproate was the first AED approved by the US Food and Drug Administration for migraine prevention. It may raise inhibitory tone in the hyper-excitable migraine brain via GABA. The usual effective dose of divalproex sodium is 500 to 1000 mg/d of the extended release formulation.
The most frequently reported adverse events are nausea (42%), infection (39%), alopecia (31%), tremor (28%), asthenia (25%), dyspepsia (25%), somnolence (25%) as well as occasional weight loss or gain. Valproate has little effect on cognitive functions and rarely causes sedation. On rare occasions, valproate administration is associated with severe adverse reactions, such as hepatitis or pancreatitis, hyperandrogenism, ovarian cysts, and thrombocytopenia. Valproate is severely teratogenic and should not be used as a first line preventive medication in women of child-bearing age.
Topiramate is a neuromodulator agent that was approved by the US Food and Drug Administration for migraine prophylaxis in 2004. Topiramate has a variety of actions that may prevent migraine, including increasing inhibitory GABA activity, blocking calcium channels, and inhibiting carbonic anhydrase.
Topiramate is well tolerated when started at a low dose, usually 15 or 25 mg, and increased weekly up to 100 mg, which is the recommended dose. Topiramate is associated with weight loss (3.3 to 4.1%). It causes transient distal paresthesias in the extremities. Less common side effects are cognitive dysfunction, acute angle-closure glaucoma, both of which are reversible with therapy discontinuation, and kidney stones (1.5%). Very rarely, oligohydrosis and hyperthermia have been described, especially in adolescents.
Gabapentin and pregabalin modulate glutamate and GABA function, as well as regulating intracellular calcium influx (see Chapter 3). Gabapentin has been studied in one large randomized controlled study and was effective in reducing migraine frequency with an average effective dose of 2400 mg/d.
The most common adverse events reported in association with gabapentin are dizziness and drowsiness.
Levetiracetam is a new anticonvulsant with an unknown mechanism of action. It has not proved effective in two small proof of concept trials for episodic migraine prevention, but it shows promise in treatment of chronic daily headache.
Minimally effective doses appear to be 1500 mg and most patients need 2000 to 2500 mg daily, with few adverse events. The most frequent side effects reported (in at least 3% of the patients) have been fatigue or tiredness, somnolence, and dizziness. Rarely, behavioral disturbance develops in patients.
Zonisamide is a sulfonamide derivative, chemically and structurally unrelated to other AEDs. Anecdotally, it has been reported helpful in episodic migraine prevention, but no randomized controlled studies have been completed. Zonisamide has similar mechanism and dosing to topiramate and is sometimes used in patients who responded to but could not tolerate therapy with topiramate. The reported side effects of zonisamide have been paresthesias, fatigue, anxiety, and weight loss. Agitated dysphoria and difficulty concentrating have also been observed.
AEDs have proven effective in migraine prophylaxis. However, in patients responsive to other medications for acute migraine, the AEDs are most cost-effective for those with a high migraine frequency and those with comorbid diseases.
This divalent cation is an essential co-factor in more than 350 enzymes. Magnesium may have a role in migraine prevention because low magnesium in the brain appears to destabilize neuronal membranes, resulting in calcium influx and aura initiation. At least 400 to 600 mg of chelated magnesium is necessary for up to 3 months of supplementation in migraine prevention and may be most appropriate for patients with migraine with aura.
An additional proposed mechanism for migraine brain hyperactivity is a defect in energy generation in the mitochondria. Vitamin
B2 or riboflavin is required for the electron transport chain. Coenzyme Q10, like riboflavin, participates in the electron transport chain in the mitochondria, although it is not a cofactor but rather transfers electrons. Both have been tested in very small randomized controlled trials and have shown efficacy. The dose of riboflavin may be as low as 25 mg/d, and the dose of coenzyme Q10 is 300 mg/d.
The evidence for the antimigraine efficacy of feverfew, a dried leaf preparation of the weed Tanacetum pathenium, is not fully established. Feverfew's side effects include mouth ulceration and a more widespread oral inflammation associated with loss of taste. Feverfew's mechanism of action is uncertain.
Lisinopril is an angiotensin-converting enzyme (ACE) inhibitor frequently used to treat hyper tension and heart failure. It does not have an indication for the prevention of migraine, although it possesses various pharmacologic effects that may be relevant in the pathophysiology of migraine and was found effective in one small randomized controlled trial. It blocks the con version of angiotensin I to angiotensin II, and blocks degradation of bradykinin, enkephalin, and substance P. Lisinopril has a clear potential for migraine prophylaxis because patients with migraine more commonly have the ACE DD gene, which codes for a higher ACE activity. The main side effects are cough, hypotension, and fatigue. A dose of 10 mg/d was tested in migraine patients.
Angiotensin II type 1 receptors (AT1) are presynaptic inhibitors of GABA release. Candesartan, an angiotensin II-receptor blocker, was also tested in one small randomized controlled trial for migraine prevention at a dose of 16 mg and was effective. Side effects were minimal, except for decreased blood pressure.
The anti-serotonin migraine-preventive drugs are potent 5-HT2B- and 5-HT2C-receptor antagonists, and these serotonin receptors are believed excitatory to migraine. Methy-sergide (no longer available in the United States), methylergonovine, cyproheptadine, and pizotifen (not available in the United States) are effective migraine pro phylactic drugs and are 5-HT2B- and 5-HT2C-receptor antagonists. Methysergide breaks down to methy lergonovine, and both have side effects including nausea, vomiting, abdominal pain, and diarrhea. Also frequently reported are leg symptoms (restlessness or pain), dizziness, and drowsiness. The major (albeit rare) complication of methysergide, which may also occur with methylergonovine, is the development of retroperitoneal, pulmonary, or endocardial fibrosis (1/1500 to 1/5000). Because methysergide and methy lergonovine are ergots and can narrow vessels, they are contraindicated for use with triptans. The dose of methylergonovine in migraine prevention is 0.2 mg three times daily.
Cyproheptadine is an old-fashioned antihistamine which is also a 5-HT2 antagonist. Its effectiveness in migraine is established by consensus only. Common side effects are sedation, weight gain, and dry mouth. The dose of cyproheptadine that works in migraine prevention is 4 to 8 mg each night at bedtime, and it is usually used as adjunctive treatment rather than as primary treatment.
Pizotifen is a similar medication to cyproheptadine, licensed for migraine preventive use in the United Kingdom. It can also cause substantial weight gain and drowsiness.
Botulinum neurotoxin type-A (BTX-A) has been approved in the United States for blepharospasm and for forehead wrinkles. BTX-A inhibits the release of the neurotransmitter acetylcholine at the neuromuscular junction, inhibiting striatal muscle con tractions. However, BTX-A can also reduce pain in various pain syndromes independent of its muscle paralyzing effects, probably related to decreasing calcitonin gene-related peptide release and reducing the inflammatory and vasodilating components of migraine.
A typical treatment protocol is to inject 100 units of BTX-A symmetrically into glabellar, frontalis, temporalis, trapezius, and other neck muscles. The side effects are transient and can include frontal weakness, ptosis, and local pain. Injections can be repeated every 3 to 6 months if patients have a beneficial effect, which wears off 3 to 6 months after treatment. The efficacy of BTX-A in migraine prevention is not fully established, and large international randomized controlled trials are underway.
Tizanidine is a centrally acting, presynaptic α2-adrenergic agonist. Its mechanism of action is thought to be through a decrease of norepinephrine re lease from the upper dorsal brainstem. Thus, its effect in headache prevention may be due to its reducing central excitability. One randomized controlled study of tizanidine found efficacy in preventing chronic daily headache. The most frequent side effects described are somnolence, asthenia, dizziness, and dry mouth in less than 10% of the patients. It is so sedating that tiny doses of 2 to 4 mg are necessary initially, with a gradual ascent (range, 2 to 24 mg). Since therapy with tizanidine may rarely result in liver toxicity, patients should have blood levels monitored.
Petasites is an extract from the plant Petasites hybridus (butterbur), which grows throughout Europe and parts of Asia. This compound has been marketed in Germany for migraine and seems to act through calcium channel regulation and inhibition of inflammation. Although the butterbur root is toxic, the Petadolex extraction has been followed by German regulatory
authorities and appears safe. Two randomized controlled studies suggest efficacy in episodic migraine prevention, with the larger study finding an optimal dose of 150mg/d.
Treatment failures can be grouped into the following five categories:
Migraine is optimally managed with migraine-specific abortive medications if there is significant disability. If frequency of migraine is high, preventive agents should be introduced, selecting the medication based on the patient's medical and psychiatric comorbidity.
Headache Classification Subcommittee of the International Headache Society. The International Classification of Headache Disorders: 2nd ed. Cephalalgia.2004;24(suppl 1):9..
Lipton RB et al. Prevalence and burden of migraine in the United States: data from the American Migraine Study II. Headache. 2001;41:646..
Lipton RB et al. Why headache treatment fails. Neurology. 2003;60:1064..
Matchar DB et al; US Headache Consortium. Evidence-Based Guidelines for Migraine Headache in the Primary Care Setting: Pharmacological Management of Acute Attacks. American Academy of Neurology; 2000. Available at: http://www.aan.com/professionals/practice/pdfs/gl0087. pdf Last accessed: September 19, 2005.
Rapoport AM et al. The triptan formulations: how to match patients and products. CNS Drugs. 2003; 17:431..
Silberstein SD et al. Migraine: preventive treatment. Cephalalgia. 2002;22:491..
Silberstein SD et al. Multispecialty consensus on diagnosis and treatment of headache. Neurology. 2000;54:1553..
Essentials of Diagnosis
Tension-type headache is best defined as “not migraine.” It is not unilateral, not severe, not throbbing, not worsened by routine physical activity, and not associated with nausea. Patients with tension-type headache do not usually complain about it in the office.
Because of the lack of positive specific symptoms, investigations to exclude other organic diseases are performed more frequently in tension-type headache than in other headaches. A careful history to uncover coexisting depression, anxiety, and other disorders is also extremely important.
In a clinical 10-year follow-up study of episodic tension-type headache, 75% of patients continued to have episodic headaches, but the chronic form developed in 25%. In those persons with initial chronic tension-type headache, 31% remained chronic, medication overuse headache developed in 21%, and the remaining 48% had reversed to the episodic form with or without prophylactic treatment. Depression, anxiety, and medication overuse were predictors for a poor outcome.
It is important to detail the relation of tension-type headache to migraine. The Spectrum Study has shown that there are really two types of phenotypic episodic tension-type headache. Patients with migraine have attacks of what appear to be episodic tension-type headaches clinically but which respond like migraine to triptans when compared with placebo. It is believed that the attacks of episodic tension-type headaches in patients with migraine are phenotypic episodic tension-type headaches, but are genotypic migraine; in other words, those patients with migraine get low-level migraine attacks that look like tension but behave like migraine.
Patients with episodic tension-type headaches who never get migraines do not respond to triptans any better than placebo. They have different, “pure” episodic tension-type headaches not linked to migraine by continuum.
Chronic tension-type headache is chronic, low-level tension headache. Chronic migraine is migraine or probable migraine on at least 8 days per month in patients with headaches at least 15 days per month. The intensity of the daily symptoms and the associated migraine phenomena distinguish the two disorders, and, as noted, medication overuse headache is characterized by ingestion of acute, abortive medications at least 10 to 15 days per month and can appear clinically like chronic tension-type headache or chronic migraine.
Tension-type headache varies widely in frequency, duration, and severity. In a Danish population-based study, it was found that 59% of persons experiencing tension-type headache had it 1 day a month or less, and 37% had it several times a month. In the total population, 3% had chronic tension-type headache. The tension-type headache male to female ratio is 4:5, indicating that, unlike migraine headache, women are affected only slightly more frequently. In both sexes, the prevalence seems to peak between the ages of 30 and 39 years and then declines with increasing age. The average age of onset of tension-type headache is 25 to 30 years, and the mean tension-type headache duration has been reported to be 10.3 to 19.9 years.
Although tension-type headache is the most prevalent headache and affects 78% of the general population, the substantial societal and individual burden associated with tension-type headache has been overlooked. Most patients with chronic tension-type headache are left with virtually no specific treatment. Daily or near daily headaches constitute a major diagnostic and therapeutic problem, and distinguishing chronic tension-type headache from migraine headache and from medication overuse headache is a substantial diagnostic challenge because management strategies are completely different. Obviously, in medication overuse headache, detoxification is the most important intervention.
Because of its high prevalence, tension-type headache has a greater socioeconomic impact than any other headache type. The direct costs include medical costs and social services; indirect costs stem from lost production in the economy because of morbidity. Indirect costs include reduced quality of life and reduced work capability. Because of the headache attacks, the socioeconomic costs from absenteeism among patients with tension-type headache are quite substantial.
The most prominent clinical finding in patients with tension-type headache is considerably increased pericranial myofascial tissue tenderness to palpation. In addition, it has been demonstrated that the pericranial tenderness is associated with the intensity and the frequency of tension-type headache. It is unknown for certain whether the increased tenderness in tension-type headache is a primary or a secondary phenomenon. It has been suggested that increased presence of muscle activity is a normal protective adaptation to pain, muscle ischemia, and abnormal blood flow. The release of neuropeptides (eg, substance P and calcitonin gene-related peptide) from muscle afferents may play a role in myofascial pain. The mode of action of the various mediators is complex and poorly understood.
The increased myofascial pain sensitivity in tension-type headache could be caused by central factors, such as sensitization of second-order neurons at the level of the spinal dorsal horn/trigeminal nucleus and above. It has been suggested that myofascial tenderness may be the result of a lowered pressure pain threshold, central sensitization, or a combination of both.
Simple analgesics and NSAIDs are used widely to treat acute tension-type headaches. Unfortunately, there is no selective or specific therapy.
As in migraine treatment, simple analgesics, such as aspirin and acetaminophen;, nonopioid analgesics; and antipyretics, such as dipyrone (not available in the United States); NSAIDs; or combination analgesics can be used. Muscle relaxants are not considered to be effective in treating acute episodes of tension-type headache because the few studies done have not shown efficacy, and there is risk of habituation.
Preventive treatment is considered if the patient experiences a headache on more than 15 days each month (ie, chronic tension-type headache) or has very frequent episodic tension-type headache. See Preventive Therapy for migraine for medications that can prevent tension-type headache. Randomized controlled studies show amitriptyline, fluoxetine, and tizanidine are specifically beneficial in patients with tension-type headache.
Ashina M. Neurobiology of chronic tension-type headache. Cephalalgia. 2004;24:161..
Ashina S et al. Current and potential future drug therapies for tension-type headache. Curr Pain Headache Rep. 2003;7:466..
Bandstand L. Central and peripheral sensitization in tension-type headache. Curr Pain Headache Rep. 2003;7:460..
Jensen R. Diagnosis, epidemiology, and impact of tension-type headache. Curr Pain Headache Rep. 2003;7:455..
Trigeminal Autonomic Cephalalgias
The headache syndromes in this category are characterized by trigeminal pain associated with autonomic parasympathetic features that include alterations in sweating, ptosis, miosis, scleral and conjunctival injection, nasal stuffiness, and/or rhinorrhea. The presumed mechanism is V1 trigeminal nerve stimulation that produces activation of cranial parasympathetic reflexes; the presumed pathway involves outflow through the superior salivatory nucleus and the pterygopalatine ganglion.
Essentials of Diagnosis
Cluster headache is the most common trigeminal autonomic cephalalgia. In its primary form, cluster headache occurs in 0.02 to 0.06% of the population and has incidence rates of 9.8/100.000 inhabitants per year (15.6 males, 4.0 females). The male/female ratio ranges from 4:1 to 12:1.
Recent research points to the hypothalamus as the probable generator of cluster headache. The suprachiasmatic nucleus in the hypothalamic gray is a pacemaker region involved in neuroendocrine hypothalamic regulation. Low testosterone levels as well as abnormal secretory circadian rhythms of luteinizing hormone, cortisol, pro-lactin, growth hormone, follicle-stimulating hormone, and thyroid-stimulating hormone have been found during cluster cycles and interictally. In addition, melatonin is chronically reduced during the cluster phase and interictally.
Positron emission tomography and functional magnetic resonance imaging (MRI) have demonstrated increased ipsilateral hypothalamic gray activity during attacks. More recently, the insertion of stimulating electrodes into this hypothalamic gray area in 15 patients with cluster headache in Italy and Belgium inhibited further attacks.
The pain of cluster headache is invariably unilateral, and the side affected is generally consistent for every attack and every cluster period. Very occasionally, the pain may affect the contralateral side, but cluster switching sides during a cluster period is rare enough to warrant a look for secondary causes. Predominantly, the pain is in or around the orbit (92% of cases) and temporal locations (70% of cases) and may radiate into the ipsilateral neck, ear, cheek, jaw, upper and lower teeth, and nose. The pain is described as boring, tearing, stabbing, or burning.
Autonomic features occur in 70 to 90% of patients. During the attack there is at least one autonomic ipsilateral symptom of lacrimation (91 % of cases), conjunctival injection (77%), nasal congestion (75%), ptosis or eyelid swelling (74%), or rhinorrhea (72%). Other autonomic features include forehead and facial sweating and miosis. The autonomic features are short-lived, lasting only for the attack duration. Rarely, a partial Horner syndrome (ptosis or miosis) may persist after an acute attack.
The attack begins abruptly and rapidly intensifies, reaching a peak within 5 to 15 minutes. The attack also ceases suddenly, and the patient is often left feeling exhausted. The untreated attack typically lasts from 15 to 180 minutes (75% of attacks last less than 60 minutes). The attack frequency is around one to eight daily. The daily attacks usually last for 2 to 3 months (the cluster period), and the remission period (absence of attacks) can last for months to years.
In cluster headache, there is a remarkable timing predictability with a circadian or circannual periodicity; there is a nightly periodicity with attacks most frequently awakening the patient 90 minutes after falling asleep,
corresponding to the onset of the first period of rapid eye movement (REM) sleep. Clinically, cluster headache can present in an episodic form or chronic pattern. In the chronic form, the patient continues to have daily attacks without remission and never has 1 month of no headache/year.
In contrast to migraine, oral contraceptive use, menses, menopause, and hormone replacement therapy do not trigger cluster headache in women. Alcoholic beverages and vasodilator medications, such as nitroglycerin, usually trigger an attack during the cluster period. Unlike persons with migraine, patients with cluster headache are agitated and restless and prefer to be upright and to move about. The pain intensity may cause some patients to wail loudly, and others may engage in destructive activities, such as banging their heads or actually attempting suicide due to the severity of pain. Cluster headache is frequently misdiagnosed, mainly in younger patients, as sinus headache or migraine.
Cluster treatment is based on the implementation of immediate, transitional, and prophylactic treatment.
The two most effective abortive treatments are oxygen 100% at a flow rate of 7 to 15 L/min or sumatriptan SC 6 mg. Because of the sudden onset and severity of a cluster headache attack, most oral abortive treatments lack enough speed to be effective. Zolmitriptan (10 mg) was found to be superior to placebo in decreasing pain intensity by two points at 30 minutes, which is too slow for most patients with cluster to tolerate. Sumatriptan (20 mg) and zolmitriptan (5 mg nasal sprays), and dihydroergotamine (1 mg SC and 2 to 4 mg nasal spray) have been recommended as optional abortive therapies. Other alternative abortive therapies studied include 4 to 6% lidocaine, 100 mcL of 0.025% civamide (25 mcg) (capsaicin isomer) nasal drops, and 5 to 10 mg of olanzapine.
Transitional therapy is used to offer pain relief as preventive medications are added. Most commonly, corticosteroids are used, such as prednisone 1 mg/kg for 2 to 3 weeks. Other alternatives are methylergonovine 0.6 mg/d, daily ergotamine 1 to 2 mg, naratriptan 2.5 mg twice daily, or greater occipital nerve block.
Preventive treatment must be used in all patients with cluster. Verapamil, 240 mg to >480 mg, is the preferred treatment. Other prophylactic treatments are lithium, 600 to 1200 mg; valproic acid, 500 to 2500 mg; and topiramate, 25 to 125 mg. Gabapentin has also been suggested as preventive treatment. Usually at least two preventive medications are used together. Optional suggested therapies in refractory cases include methylergonovine, 0.6 mg/day; daily ergotamine; melatonin, 3 to 12 mg; and daily therapy with triptans that have a long half-life, such as naratriptan.
When preventive treatment fails, surgical treatment is suggested for refractory cases. A surgical treatment can be recommended for patients with 100% one-sided pain in the first trigeminal division and without psychiatric disease or history of drug abuse. The most frequently used surgical treatment is radiofrequency trigeminal gangliorhizolysis, effective in up to 70% of patients, with recurrence rates of 20% and failure rates around 30%.
Recently, the implantation of hypothalamic stimulators has shown promise but has not yet been tried in North America.
The natural history of cluster headache is related to the clinical presentation. Episodic cluster headache continues to be episodic in 53.2 to 67.1% of cases, becomes chronic in 2.4 to 12.9%, or can go into a prolonged remission period in 13.6 to 38.7%. Chronic cluster headache becomes episodic in 20 to 32.6% of cases, continues as chronic in 48 to 53.1%, or goes into a prolonged remission period in 12%. The factors that portend a bad prognosis are onset after 30 years of age (especially in females), long cluster periods (lasting longer than 8 weeks) with sporadic attacks in between, more than four associated symptoms, and short remission periods (lasting less than 6 months).
Essentials of Diagnosis
Episodic paroxysmal hemicrania and chronic paroxysmal hemicrania are rare syndromes characterized by headaches of short duration, high attack frequency, and associated autonomic symptoms. Clinically, the paroxysmal hemicrania attacks look like short cluster attacks occurring at higher frequency per day. Unlike cluster, with its male predominance, paroxysmal hemicrania appears to be more common in women with a female to male ratio of 3:1. As with cluster, paroxysmal hemicrania that occurs daily for months with periods of remission is considered the episodic form; patients who do not remit over 1 year for at least 1 month have chronic paroxysmal hemicrania.
The most important feature that distinguishes paroxysmal hemicrania from cluster is the frequency of attacks per day. Persons with paroxysmal hemicrania have more than five attacks per day for >50% of the time. The pain is severe in intensity and, like cluster headache, has been described as boring or clawlike. Normal headache duration is between 2 and 30 minutes. Associated symptoms are characterized by the same autonomic phenomena as cluster headache. Most patients with chronic paroxysmal hemicrania exhibit lacrimation (62%), nasal congestion (42%), conjunctival injection and rhinorrhea (36%), or ptosis (33%).
Paroxysmal hemicrania is one of the rare headache disorders that by definition is totally responsive to indomethacin. The normal starting dose of indomethacin is one 25 mg tablet three times daily for 3 days; this dose can be increased to 2 tablets (50 mg) three times daily if there is no total relief of pain. Most patients respond to 150 mg daily, and the response can be dramatic, with quick and complete dissipation of headache symptoms. A beneficial effect is normally seen within 48 hours after the correct dose is administered. If the patient does not respond to 75 mg three times daily, an alternative diagnosis should be considered. The gastrointestinal side effects normally can be controlled with proton pump inhibitors. Symptoms usually recur within several days after discontinuing indomethacin.
Essentials of Diagnosis
Although the pathophysiology of hemicrania continua is still unknown, contralateral posterior hypothalamus and ipsilateral dorsal rostralpons activation have been demonstrated on functional MRI. If posterior hypothalamic and brainstem activation are considered as markers of trigeminal autonomic headaches and migrainous syndromes, respectively, then the activation pattern demonstrated in hemicrania continua overlaps with trigeminal autonomic headaches and migraine.
The revised IHS classification defines hemicrania continua as a persistent, daily and continuous headache of moderate intensity that is strictly unilateral without side-shift for more than 3 months, without pain-free periods but with exacerbations of severe pain and associated with at least one ipsilateral autonomic symptom. Other clinical characteristics described are the presence of migrainous features during exacerbation periods, icepick-like pains (jabs and jolts or primary stabbing headaches), or a foreign body sensation in the eye. Hemicrania continua can present as a remitting form (11.8% of cases), as a continuous pattern evolving from a remitting variety (35.3%), or as a chronic disorder from the onset (52.9%).
Hemicrania continua responds exclusively to indomethacin treatment. For cases that are refractory to indomethacin therapy, alternatives include other NSAIDs, cyclooxygenase-2 inhibitors, topiramate, lamotrigine, or gabapentin.
Aurora SK. Etiology and pathogenesis of cluster headache. Curr Pain Headache Rep. 2002;6:71..
Bahra A et al. Cluster headache: a prospective clinical study with diagnostic implications. Neurology. 2002;58:354..
Geweke LO. Misdiagnosis of cluster headache. Curr Pain Headache Rep. 2002;6:76..
Matharu MS et al. Trigeminal autonomic cephalgias. J Neurol Neurosurg Psychiatry. 2002;72(Suppl II):ii19..
Rozen TD. Short-lasting headache syndromes and treatment options. Curr Pain Headache Rep. 2004;8:268..
Chronic Daily Headache
Daily or near-daily headache is a widespread problem in clinical practice. The general term of “chronic daily headache” encompasses primary headaches presenting more than 15 days per month and lasting more than 4 hours per day. Chronic daily headache includes chronic migraine, transformed migraine, chronic tension-type headache, new daily persistent headache, and hemicrania continua (Table 12-7). At least 40% of patients who seek medical attention at a specialized headache clinic meet diagnostic criteria for chronic daily headache, of which 80% are women. Of those, 60% suffer transformed migraine, 20% chronic tension-type headache, 20% meet new daily persistent headache criteria, and up to 80% of patients overuse symptomatic medications (medication overuse headache).
Chronic daily headache prevalence in the general population is 4 to 5% (up to 8 to 9% for women). New daily persistent headache is rare (0.1%), chronic migraine is intermediate (1.5 to 2%), and chronic tension-type headache is most common (2.5 to 3%). In contrast to data from specialized clinics, only around 25% of persons with chronic daily headache in the general population overuse analgesics.
Comorbid factors may be significant contributors to the development and maintenance of chronic daily headache, but a causal relationship does not necessarily exist. In addition to medication overuse, the following associations occur commonly in persons with chronic daily headache: hypothyroidism, obesity, snoring, asthma, hypertension, and daily consumption of caffeine.
Essentials of Diagnosis
Patients with chronic headaches and medication overuse headaches are particularly difficult to treat because prophylactic medications lack efficacy in this setting, and because discontinuation of the offending medication can lead to a severe withdrawal headache, resembling an acute migraine attack. Withdrawal headache, rebound headache, or medication overuse headache is often accompanied by vegetative symptoms. In addition, patients often show signs of physical and emotional dependence and psychological involvement. A successful treatment has wide-ranging positive benefits reducing both headache and associated disability.
Many drugs are known to cause medication overuse headache. The diagnosis of medication overuse headache requires that the headache be present for more than 15 days each month, and the headache should develop or markedly worsen during medication overuse. The headache resolves or reverts to its previous pattern within 2 months after discontinuation of medication.
Medication use needs to occur several times a week to cause medication overuse headache. This syndrome is less likely to develop in patients who intersperse consecutive days of medication use (eg, with menstrual migraine) with long drug-free periods. Headache characteristics associated with overuse vary and may be tension-like or migraine-like, frequently waxing and waning and also varying in location, but neck pain is frequent. There is a circadian rhythmicity to medication overuse headache, with morning headache common, since withdrawal occurs through the night.
Screening questions for drug overuse should always be asked of patients with chronic daily headache. The overused medication may provide information about the patient's neuropsychological condition. Type I patients prefer over-the-counter analgesics or nonsedating prescription medication. Their response to treatment is usually good. Type II or “beaten down” patients are more depressed and limited by headache disability. These patients prefer multiple medications and often use opioids. Psychotherapy and antidepressant medication helps their recovery. Type III patients show drug-seeking and compulsive behavior toward opioids and similar habituating medications. Treatment outcome in this group is poor, and an approach for primary substance abuse may be helpful.
The initial approach is to discontinue the overused medications. There are two general outpatient strategies. One approach is to taper the overused medication, gradually establishing effective preventive therapy as the taper progresses. The second strategy is to abruptly discontinue the overused drug when safe, use a bridge or transitional medication to ease the withdrawal temporarily, and rapidly add preventive treatment at the same time. Drugs used for the bridge include NSAIDs, corticosteroids, ergots, and triptans.
Table 12-7. Diagnostic Characteristics of Chronic Daily Headache.
During the washout period, typical withdrawal symptoms can occur (eg, headache exacerbation, nausea, restlessness, and sleep disorder). Typically, the intensity of the headache increases 2 days into the withdrawal period and declines by the end of the week. Once the withdrawal period is over, which may last 1 to 6 weeks without treatment, there is frequently considerable headache improvement, with gradual cessation of daily headache, and reestablishment of an episodic migraine pattern.
Hospital admission should be considered when patients meet the following criteria:
The treatment process can be enhanced and shortened and the patient's symptoms made more tolerable by administering repetitive intravenous dihydroergotamine with an antiemetic, such as metoclopramide. It has been reported that up to 92% of patients became headache-free usually within 2 to 3 days with an average hospital stay of 4 to 7 days. Patients who are not candidates for or do not respond to dihydroergotamine can be treated with repetitive intravenous valproate, ketorolac, other neuroleptics, ondansetron, or corticosteroids. These agents may also supplement repetitive intravenous dihydroergotamine in refractory cases (Table 12-8).
If patients with medication overuse headache do not respond to potentially effective preventive therapies, another trial of previously used preventive medications may be successful after the patient has been detoxified and has recovered from rebound headache. The patient and the treating physician must understand that a given preventive medication may not become fully effective until the overused medication is eliminated and the washout period complete. Patients suffering from drug-induced headache often exhibit depression, low frustration tolerance, and physical and emotional dependency. Hospitalization can provide patient education and behavior modification and can initiate an outpatient program of preventive and immediate therapy.
As noted above, after the offending agents have been withdrawn, the headache pattern usually becomes episodic. This generally takes weeks, although some reports suggest that the change from daily to episodic headache may take as long as 6 months. At least 60% of these patients no longer have daily headache, and approximately 40% still have migraine attacks. There are no literature reports of spontaneous improvement from rebound headache without detoxification. The relapse rate within 6 months after withdrawal therapy is approximately 30% and increases steadily to 50% after 5 years, without close follow-up.
Chronic migraine is the most common type of chronic daily headache in specialty care. Persons with transformed migraine usually report a process of transformation over months or years, and as headache increases in frequency, associated symptoms become less severe and frequent. The process of transformation frequently ends in a pattern of daily or nearly daily headache that resembles chronic tension-type headache, with some attacks of full-blown migraine superimposed. In the original proposal for this disorder, subsequently validated, chronic migraine could occur with or without medication overuse.
Many patients with primary chronic migraine no longer have daily headache after undergoing withdrawal of medication overuse; however, a significant subgroup still has episodic migraine. Nevertheless, regular use of analgesics with an existing history of headache, in particular migraine, is not always the sole factor responsible for the development of chronic daily headache, and complex genetic factors may also play a role. It has also become clear that analgesics per se do not cause the development of daily headache de novo in persons with no previous headache history.
Bigal ME et al. Chronic daily headache: identification of factors associated with induction and transformation. Headache. 2002;42:575..
Bigal ME et al. Transformed migraine and medication overuse in a tertiary headache center-clinical characteristics and treatment outcomes. Cephalalgia.2004;24:483..
Freitag FG et al. US Headache Guidelines Consortium, Section on Inpatient Treatment Chairpersons. Inpatient treatment of headache: an evidence-based assessment. Headache. 2004;44:342..
Krymchantowski AV et al. Out-patient detoxification in chronic migraine: comparison of strategies. Cephalalgia. 2003;23:982..
Levin M. Chronic daily headache and the revised international headache society classification. Curr Pain Headache Rep. 2004;8:59..
Limmroth V et al. Features of medication overuse headache following overuse of different acute headache drugs. Neurology. 2002;59:1011..
Srikiatkhachorn A. Pathophysiology of chronic daily headache. Curr Pain Headache Rep. 2001;5:537..
Table 12-8. Pharmacologic Treatments for Refractory Cases.
Neuroimaging in Primary Headaches
The primary reason to obtain a neuroimaging study in a headache sufferer is to rule out significant and treatable disease. Secondary indications include relieving patient anxiety as well as avoiding litigation concerns responding to a family request. The threshold for investigating headache decreases in patients with red flags or atypical signs. Red flags include sudden severe headache; change in headache pattern, such as rapid increasing headache frequency or severity; changes in
headache characteristics; 100% one-sided headaches; or headache in special populations, such as the elderly and persons with HIV or cancer.
Imaging studies should always be considered in patients with focal neurologic signs or atypical symptoms, such as seizures. The prevalence of neuroimaging abnormalities in patients who have episodic migraine with normal neurologic examination is around 0.18%. In chronic daily headache sufferers without red flags, abnormalities are found in 0.67%. Nonspecific brain abnormalities have been reported in 12 to 46% of migraine sufferers; however, in the population as a whole, there is not enough evidence to demonstrate that migraine is an independent risk factor for cerebrovascular disorders. The potential discovery of serious treatable lesions is about 0.4% in migraine using computed tomography or MRI scans, and the side effects related to the use of these diagnostic tools can be as high as 10% due to iodine reaction and claustrophobia.
The US Headache Consortium Guidelines identified four consensus-based recommendations for imaging:
Evans RW. Diagnostic testing for headache. Med Clin North Am. 2001;85:865..
Frishberg B et al. US Headache Consortium. Evidenced-Based Guidelines in the Primary Care Setting: Neuroimaging in Patients with Nonacute Headache. American Academy of Neurology; 2000. Available at: http://www.aan.com/professionals/practice/pdfs/gl0088. pdf Last accessed: September 19, 2005.
Jamieson DG et al. The role of neuroimaging in headache. J Neu-roimaging. 2002;12:42..