Pharmacological Approaches to Managing Migraine and Associated Comorbidities—Clinical Considerations for Monotherapy Versus Polytherapy


  • Stephen D. Silberstein MD,

  • David Dodick MD,

  • Fred Freitag DO,

  • Starr H. Pearlman PhD,

  • Steven R. Hahn MD,

  • Ann I. Scher PhD,

  • Richard B. Lipton MD

  • From the Thomas Jefferson University, Philadelphia, PA, USA (Dr. Silberstein); Mayo Clinic, Scottsdale, AZ, USA (Dr. Dodick); Diamond Headache Center, Chicago, IL, USA (Dr. Freitag); Savannah, GA, USA (Dr. Pearlman); Albert Einstein School of Medicine, Bronx, NY, USA (Drs. Hahn and Lipton); and Uniformed Services University, Bethesda, MD, USA (Dr. Scher).

Address all correspondence to Dr. Stephen D. Silberstein, Jefferson Headache Center, 111 South 11th Street, Suite 8130, Philadelphia, PA 19107.


Comorbidity is defined as an illness that occurs more frequently in association with a specific disorder than would be found as a coincidental association in the general population. Conditions that are frequently comorbid with migraine include depression, anxiety, stroke, epilepsy, sleep disorders, and other pain disorders. In addition, many common illnesses occur concomitantly (at the same time) with migraine and influence the treatment choice. Migraine management, and especially migraine prevention, can be challenging when patients have comorbid or concomitant illnesses. The objectives of this initiative are to review the literature on managing patients who have migraine and common comorbidities, present additional clinical approaches for care of these difficult patients, and evaluate the areas in which research is needed to establish evidence-based guidelines for the management of migraine with associated comorbid conditions.

Migraine is a disorder that is present in approximately 13% of the general population and is associated with both comorbid and concomitant illnesses that influence treatment strategy.1,2 Comorbidity is defined as an illness that occurs more frequently in association with a specific disorder than that would be found as a coincidental association in the general population.3 Concomitant (coexistent) illnesses occur at the same time in the same person at the rate that would be expected by chance.

The common illnesses that are associated with migraine and influence its management include comorbid conditions such as depression, anxiety disorders, epilepsy, sleep disorders, and stroke and concomitant illnesses such as hypertension and obesity (Table 1). Both comorbid and concomitant medical conditions impact migraine treatment. Understanding how to design treatment plans that address migraine in patients with other medical conditions is the focus of this review.

Table 1.—. Comorbid and Concomitant Conditions Associated with Migraine in Clinical Care or Population Studies
 • Heart attack/angina
 • Hypertension or hypotension
 • Stroke
 • Raynaud's syndrome
 • Patent foramen ovale
 • Epilepsy
 • Essential tremor
Mood disorders
 • Depression
 • Mania
 • Anxiety
 • Panic
GI disorders
 • Ulcer disease
 • Colitis
 • IBS
 • Snoring/sleep apnea/sleep disorder
 • Other chronic pain disorders

The biological mechanisms that underlie comorbid conditions and migraine are not understood, and the relationship may be unidirectional or bidirectional, depending on the associated illness.4 One example of a bidirectional relationship is that between migraine and blood pressure. Headache-specific medications may lower blood pressure, and elevated blood pressure might theoretically aggravate headache. Selected medical conditions may be comorbid because they share a genetic factor that increases the risk of both conditions (for example, the MTHFR C677T variant associated both with migraine and with ischemic stroke).5 Diagnosing and treating migraine with associated comorbidities require follow-up and monitoring of all the disorders, and extra time may be necessary to achieve control of both illnesses.

For the last 10 years, recommendations have focused on using one drug to treat migraine and associated comorbid conditions whenever possible.6,7 For example, when a patient has both migraine and depression, past practice has been to use a single medication, such as a tricyclic antidepressant (TCA) or a serotonin reuptake inhibitor (SSRI), to treat both illnesses.

This approach appears to offer the prospect of simplifying management, reducing costs, minimizing potential side effects, and eliminating potential drug interactions. However, the scientific rationale for using a “two-for-one” approach in migraine has not been prospectively tested in controlled trials. Recent studies suggest that using multiple medications (polypharmacy) may confer therapeutic advantages for patients with migraine and other conditions, such as depression or anxiety.8

This review systematically evaluates benefits and limitations of pharmacological monotherapy versus polytherapy for migraine prevention and common associated comorbidities. Individual authors each addressed a specific area of interest, which have been combined into this single comprehensive review. Contributing authors participated in the following areas: Background, introduction, epidemiology, and discussion (S. Silberstein, R. Lipton, and S. Pearlman), depression (S. Hahn), obesity (R. Lipton), stroke and hypertension (D. Dodick), women's issues (F. Freitag; eg, pregnancy, menses, menopause, etc), and epilepsy (A. Scher). This review also identifies specific areas in which research is needed to establish evidence-based treatment recommendations for patients with migraine and comorbid illnesses (all participants).


An illness is comorbid with migraine if it occurs at a higher incidence in migraine patients compared with the general population. For example, patients with depression had a significantly higher chance of developing migraine (9.3%) within 2 years compared with subjects without depression (2.8%; Fig. 1).9 Similarly, subjects with migraine had a significantly higher risk of developing depression (10.5%; Fig. 1) compared with nonmigraine subjects (2.0%; Fig. 1) or those with headaches of the nonmigraine type.9

Figure 1.—.

(a) Presence of major depression at baseline was a risk factor for developing migraine; (b) Presence of migraine at baseline was a risk factor for developing depression by the 2-year follow-up period.

The prevalence of disorders that are comorbid with migraine varies across studies. The heterogeneity of the patient populations studied may account for the differences among epidemiological rates. Additionally, variability in definitions and diagnostic criteria for selected illnesses differs across studies. For example, definitions for depression can vary based on meeting diagnostic criteria versus those with selected symptoms of depression.

The associated comorbidities can be life threatening or severely disabling if not treated appropriately (eg, exacerbation of depression or stroke). Therefore, improving awareness, diagnosis, and treatment of migraine and its associated comorbidities is critical to improve patient care and reduce the burden of illness on health systems, families, and work forces.


Migraine treatment includes the use of acute and preventive medications, along with nonpharmacological/behavioral approaches. Several preventive medications have proven helpful in migraine patients (Table 2). Previous migraine treatment guidelines review the established clinical efficacy of these medications for migraine prevention.6,7 Recommendations to use a “two-for-one” treatment approach for migraine suggested that one medication may meet the therapeutic need of patients with migraine and an associated comorbid condition.6 However, this approach has not been prospectively studied and there are potential limitations to using a single medication to treat 2 separate illnesses such as:

Table 2.—. Primary Medication Groups for Preventive Migraine Treatment
Migraine prevention is recommended when patients have high attack frequency (≥2 attacks per week) or severely disabling attacks. Medications should be chosen from one of these tested classes of agents and based on the drug's side efficacy profile and adverse-event risk and the presence of coexistent and comorbid conditions. Clinical studies summarizing the clinical evidence of individual agents within these classes of medications is reviewed elsewhere.6,7
 • Anticonvulsants
 • Antidepressants/serotonin antagonists
 •β-Adrenergic blockers
 • Calcium channel antagonists
 • Others (including riboflavin, minerals, herbs, and botulinum toxin A)
  • • Risk of treating only one condition: Giving a single medication may not treat 2 different conditions optimally. Although one of the 2 conditions may be adequately treated with a specific medication, the second illness may require a different medication or a different dose or dosing regimen. Therefore, the patient is at risk of the second illness not being adequately treated.
  • • Risk of choosing suboptimal medication: In an effort to use a single medication to treat 2 conditions, the physician may select a second- or third-tier choice for either or both, and this may not provide adequate treatment for either.
  • • Risk of poor tolerability when a third comorbid or coexistent condition is present: If a patient has angina and migraine, a β-Blocker may prove helpful, but it may be poorly tolerated (or even contraindicated) when another comorbid or concomitant condition, such as depression or asthma, is present.
  • • Treatment timelines: Managing 2 separate illnesses may require different timelines for assessing efficacy. A change in migraine status may take 4 or 6 weeks before improvement in attack frequency or severity is measurable, and up to 6 months before improvement is maximal. However, other conditions may respond over a much shorter timeline (eg, hypertension may respond to initiation or a change in treatment within 1 to 2 weeks).


Patients with multiple treatment needs may be optimally treated with a “therapeutic independence” approach, wherein each condition is treated separately. This requires the physician to assess which medication is best suited for each illness independently, and may actually require multiple physicians or specialists working together to provide optimal treatment recommendations. There are 3 separate patient scenarios that the physician is faced with when assessing a patient with migraine and a comorbid condition:

  • 1The treatment-naïve patient: This patient is first seen in the clinic and is not receiving therapy for migraine or any other condition. The challenge for the physician is to ensure accurate diagnosis and treatment for each condition. Referral to an appropriate specialist may be required to help manage one or more of the conditions present.
  • 2The migraine patient: This patient has already been diagnosed with migraine and subsequently presents with a concomitant or comorbid illness. The challenge is to assess whether migraine management is adequate, and then whether to add an additional medication or increase the dose of the existing medication to treat the comorbid or coexistent illness.
  • 3The comorbid migraine patient: This patient is under medical care for a different condition and then presents with migraine as either a comorbid or coexistent illness, or as a result of treatment for the primary condition (eg, the development or worsening of headache in a patient being treated with nifedipine for hypertension).

Designing a management plan for patients with migraine and coexistent conditions is complex and requires sophisticated assessment of the patient's needs. Several challenges exist when developing a management plan for these patients:

  • • Dosing: Many conditions fluctuate in their level of severity, and adjusting a single medication to meet the dosing requirements of 2 separate illnesses may be difficult. Here are 3 different possible dosing scenarios for patients:
  • 1A single dose of medication may adequately manage both conditions. In this scenario, the first condition is well controlled at a lower dose than required for treatment of the second condition, but the patient tolerates the dose requir- ed to treat the second condition. In this scenario, the response of the first condition is incidental to the optimal treatment of the second, and one medication may successfully treat both conditions. For example, a single dose of a sustained-release β-Blocker may be sufficient for a patient with angina and migraine. However, β-Blockers are no longer recommended as first-line treatment for hypertension,10,11 so the combination is no longer first-line.
  • 2The maximum tolerated dose may prove adequate for one condition, but not adequate for the second. In this case, an additional or alternative treatment plan would be needed to assure that the second condition also would be adequately treated.

Each condition may be optimally treated using different doses of the same medication. Therefore, prescribing different medications for each condition may enable therapeutic adjustments based on the status of each illness (low dose of a long-acting β-Blocker may be sufficient for migraine control but not hypertension). Usually, medicine is started with a minimally effective dose and then titrated over weeks or months. This may prove helpful for one condition, but the associated disorder may require a higher dose or different titration timeline. If a dose proves subtherapeutic for either one of the conditions, the patient may be at risk, quality-of-life may be affected, disability may persist, and the patient may prematurely discontinue the medication. Conversely, increasing the dose of a single medication to achieve a therapeutic effect for both conditions may lead to more side effects in some patients than using a different medication for each condition in lower dosages. Additional studies that assess the risk-benefits of using one versus several medications in specific concomitant illnesses, such as hypertension and migraine, are needed to determine optimal treatment strategies.

  • • Drug compatibility: Avoiding drug interactions or increased adverse events is of primary concern when using polytherapy. For example, when a patient is undergoing treatment for depression, adding a β-Blocker for migraine management may exacerbate the depression. On the other hand, patients with depression and migraine may be optimally managed with an SSRI for depression and an anticonvulsant for migraine prevention. However, some anticonvulsants can aggravate depression, and this can be clinically significant. The occurrence of a low-probability adverse effect can be significant for a specific patient, even if the prevalence is low.
  • • Drug tolerability profile: Some medications are associated with adverse events that may prove helpful in migraine or other conditions. For example, for patients with sleep disorders or insomnia not secondary to depression, a sedating TCA may be the optimal choice. For obese or underweight patients, selected anticonvulsants may cause weight loss or weight gain.
  • • Two-for-one: For selected conditions, treating 2 conditions with a single medication may be effective, but careful monitoring is needed. For example, patients with migraine and depression may respond to a tricyclic antidepressant (TCAs). However, appropriate management of depression often requires higher doses of a TCA than a migraine patient may tolerate because of adverse events. Migraineurs who also have hypertension may benefit from calcium-channel blockers, although the evidence for efficacy of calcium-channel blockers in migraine is minimal.6 Migraineurs with epilepsy may achieve control of both conditions with antiepileptic drugs, but the dose required to control seizures may be higher than the dose that is tolerated and required to control migraine. The risk is that one of the conditions is managed appropriately and the second condition requires a higher dose or additional therapy. Therapeutic independence may be needed if monotherapy fails.

Approaching patients who have multiple medical needs and illnesses can be challenging, but managing illnesses that are codependent is largely unexplored. The choice of whether to treat migraine first and follow the coexistent disorder or vice versa or treat both simultaneously with different medications is a complex dilemma for which evidence-based guidelines currently do not exist. Assessing the impact of one illness on the other is also unexplored. This is confounded with the clinical concern that that treating one condition can in fact exacerbate another comorbid condition, so there is a need to monitor this possibility and modify treatment (eg, reduce the dose or discontinue a drug) if one condition appears to be deteriorating despite improvement in the other. Importantly, clinical monitoring of the severity of comorbid conditions when treating migraine needs to be a priority when providing care for these complex patients.


Psychiatric Comorbidities.— Psychiatric disorders that are comorbid with migraine include depression, anxiety, and bipolar disorder.2,9,12–17 Approximately 25% of patients in general medical clinics meet criteria for a current DSM diagnosis of mood, anxiety, substance abuse, and related disorders. In one study of 1000 patients in 4 general medical or family practices, 26% had a current DSM disorder and an additional 13% had subthreshold but probably clinically significant mental disorders (Fig. 2).18,19 One-third of patients attending primary care clinics for other health concerns meet diagnostic criteria for migraine.20 Thus, it is highly likely that a significant number of patients surveyed within primary care settings will have migraine and a coexisting or comorbid psychiatric condition.

Figure 2.—.

Prevalence of Psychiatric Symptoms

Treatment for each of these psychiatric disorders is well defined using newly published guidelines.18,21 For depression, the primary recommendation is to use SSRIs as first-line therapy, followed by TCAs as a second-tier medication.18 However, the selection of first-line therapy may need to change or a second medication may need to be added to manage migraine.

There are no randomized, double-blind trials that establish treatment efficacy for patients with migraine and depression or other psychiatric disorder. However, a lower than recommended TCA dose is usually ineffective in treating depression but may be helpful as a migraine preventive. It is not advisable to use a reduced dose of TCA for maintenance once remission has been achieved; the dose required for remission is the maintenance dose. Consequently, the hope for a TCA agent that can effectively treat migraine and depression concurrently without encountering the side effects or other problems associated with maximum recommended antidepressant dose TCA therapy remains unfulfilled.

SSRIs have not been proven consistently effective for migraine.6 A recommended dose of SSRI combined with a low dose of TCA is a treatment paradigm not tested in clinical studies of patients with migraine and comorbid depression. One study, which examined the use of full-dose amitriptyline versus citalopram alone or in combination in patients with the “triple comorbidity” of migraine, tension-type headache, and depression, confirmed that both drugs used alone were equally efficacious in treating depression, but amitriptyline was more effective in reducing headache frequency.8 Patients with inadequate responses to one drug were given combination therapy and demonstrated improvement in both depression and headache frequency.8 Another possible treatment combination may be an SSRI to treat depression combined with an anticonvulsant to treat migraine.

Thus, the first-line drugs that are best for depression may not be the best choice for migraine. However, for some migraineurs with depression for whom a TCA is an acceptable or the best antidepressant, a single agent may be the optimal treatment for both disorders. Treatment of depression as per recommended guidelines may result in a secondary improvement in migraine for some patients. If neither condition improves, migraine and depression should be treated independently by adding an anticonvulsant, calcium channel blocker or other migraine preventive. Although there is some evidence that β-Blockers do not cause depression,22,23 this is controversial. Less is known about the effect of β-Blockers on patients with existing depression.

Depressive disorders for which pharmacotherapy and psychotherapy have proven efficacy (major depressive disorder, dysthymic disorder, and major depression in partial remission or recurrence) should be treated with these modalities when they are diagnosed. There is no evidence that justifies delaying the treatment of depression to assess the impact of treating migraine on a depressive disorder of this severity unless the mood disorder is believed to be due to a situational adjustment reaction specifically caused by the patient's headache frequency or severity. In this case, treating migraine first (with careful observation for exacerbation of the depression in the meantime) may be desirable.

If the indications for initiating treatment for depression are primary, and the indications for migraine prevention are minimal, one can treat the depression. Once the depression has responded, one can observe the migraine frequency and impairment. However, there is no evidence to suggest that migraineurs who clearly need preventive medication (eg, those who have 3 or more disabling headache days per month) will improve to the point where prevention is no longer needed because treatment for depression was successful. One can take precautions about starting 2 medications at the same time (eg, a staggered start to differentiate potential treatment emergent adverse effects) and simultaneously initiate treatment for both conditions.

The emergence of depressive symptoms in patients with established migraine who do not need preventive medication may be evidence that migraine control is suboptimal. The issue is whether to treat the emerging depression, intensify migraine management with preventive therapy, or do both simultaneously. The decision should be guided by the severity of the depression. Personal and family history may also impact the decision-making process.

Migraine treatment may also be confounded when the “sick role” has become an important part of the way in which patients deal with the world.24 Although the prevalence of sick role-based coping in migraine specifically is not known, we do know that virtually any illness can be elaborated for a sick-role coping strategy.25-27

Migraine and Hypertension/Stroke.— Migraine is an independent risk factor for stroke, and the risk varies according to gender, age, the presence of aura, and the presence of other stroke risk factors (eg, smoking, use of oral contraceptives; Table 3).32 The risk of stroke in women under age 45 is very low, estimated between 5 and 10 per 100,000 woman-years. Stroke risk is increased in female migraineurs <45 years of age, but absolute risk still remains low: 17 to 19 in 100,000. While the most consistent associations between migraine and stroke have been in younger women (age < 45 years) with aura, studies have reported elevated stroke risks in older women and older men who have migraine with aura.33–35

Table 3.—. Migraine as an Independent and Synergistic Stroke Risk Factor
Risk FactorOdds Ratio
  1. Bousser, 2004;28 Chang et al, 1999;29 Etminan et al, 2004;30 Schwartz et al, 1998.31

Migraine2 to 3
Migraine with Aura 6
Smoking 2
Migraine + smoking10
OCP use0.66
Migraine + OCP (low dose) 2
Migraine + smoking + OCP34

The relationship between migraine and other cardiovascular events is uncertain. In the Women's Health and Physicians Health Study, 39,876 women aged 45 years and older were followed for 6 years and 22,071 men aged 40 to 84 years were followed for 12 years.36 Migraine was not associated with increased risk of nonfatal myocardial infarction, fatal coronary heart disease, angina, or coronary revascularization procedures. Lack of association was also reported in 3 other prospective studies.37–39 However, none of these studies provided risk estimates for the presumptive higher risk population (ie, younger migraineurs with aura). A recent reanalysis of the Women's Health Study40 suggests that the lack of association seen in the earlier study was due to short follow-up. The new analysis followed participants for an additional 6 years and also presented results separately for the women with and without aura. Results showed that women with current migraine with aura at baseline were at increased risk of incident ischemic stroke, myocardial infarction, coronary revascularization, angina, and ischemic CVD death compared to nonmigraineurs. No increased risk was found for the women with migraine without aura.

The prevalence of hypertension is estimated to be about 50 million people in the United States.41 An association between hypertension and migraine has not been clearly established, and the prevalence of hypertension in migraineurs is likely to parallel that in the general population. There is a continuous and consistent relationship between blood pressure and risk of cardiovascular disease, which is independent of other risk factors. Each increment of 20/10 mmHg doubles the risk of cardiovascular disease across the entire blood pressure range, starting from 115/75 mmHg. This strongly supports the need for blood-pressure specific medications for patients with hypertension.

Thiazide-type diuretics, either alone or in combination with other drug classes, are recommended for initial drug therapy for most people with hypertension. Most patients will require 2 or more antihypertensive drugs to achieve target blood pressure. If blood pressure is >20/10 mmHg above goal, therapy with 2 agents should be initiated, one of which usually is a thiazide-type diuretic. Thiazide-type diuretics also are useful in slowing demineralization in osteoporosis; however, they should be used cautiously when a patient has gout or a history of significant hyponatremia due to their potential for worsening these conditions. β-Blockers are useful in the treatment of atrial tachyarrhythmias and atrial fibrillation, migraine, thyrotoxicosis (short-term), essential tremor, or perioperative hypertension. They are no longer considered a first-line drug for hypertension.10,11β-Blockers should generally be avoided in patients with asthma, depression, or second- or third-degree heart block because of their potential to exacerbate these conditions. Calcium channel blockers are useful for patients with Raynaud's syndrome and certain arrhythmias. α-Blockers may be useful for patients with prostatism. Angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers have renal protective effects in diabetes but are contraindicated in women who are pregnant or likely to become pregnant. ACE inhibitors should not be used by individuals with a history of angioedema because of their potential to precipitate urticarial reactions and angioedema. Aldosterone antagonists and potassium-sparing diuretics may play a unique role in treating congestive heart failure, but they can cause hyperkalemia and are less effective than other antihypertensive agents.

Some medications are useful for both migraine and hypertension. β-Blockers can be used to treat both conditions and there are minimal or limited drug interactions. β-Blockers are cost-effective and can enhance efficacy when added to other drugs used to treat hypertension, but due to their adverse event profile they are no longer first-line therapy.10,11 A number of β-Blockers including atenolol, metoprolol, nadolol, timolol, and propranolol have been shown to be effective for migraine prevention.6,7 The efficacy of carvedilol has not been established for migraine, and many other β-Blockers used for hypertension have no documented efficacy for migraine prevention. Differences among β-Blockers regarding intrinsic sympathomimetic activity, receptor selectivity, pK profiles, and slow-release formulations may influence efficacy for migraine prevention, but comparative studies are not available.

Poorly controlled hypertension is a serious condition with a potentially catastrophic outcome, and should therefore be treated with priority. In general, a lower dose of a drug can be used to manage migraine than to treat hypertension. For many patients with both conditions, treating hypertension with an appropriate dose may be sufficient to significantly reduce migraine frequency or severity. However, in some patients, dosing challenges may limit the use of β-Blockers for both migraine and hypertension. Hypertension and migraine need to be evaluated as separate illnesses and may require therapeutic independence. A thiazide diuretic is a frequent and often necessary treatment for hypertension but not for migraine.

Migraine and Obesity.— Migraine and obesity are highly prevalent conditions in the United States. Migraine prevalence is 18.2% among women and 6.5% among men,1 and obesity prevalence is about 30%.42,43 The prevalence of migraine in obese and morbidly obese patients is 15.1% and 15.6%, respectively.44 To evaluate the relationship between migraine and obesity, standard definitions are required. Body mass index (BMI) is a standard measure defined as weight in kilograms divided by body surface area in meters squared. A BMI of <25 kg/m2 is considered normal; 25 to 29.9 kg/m2 is considered overweight, >30 to 34.9 kg/m2 is considered obese, and >35 kg/m2 is considered morbidly obese.

Obesity does not appear to be a risk factor for episodic migraine, although it is a risk factor for chronic daily headache (CDH).45 In a longitudinal study, obese headache sufferers had a 5-fold increased risk of developing CDH in comparison with normal-weight control subjects.45 Among migraine sufferers, obesity is associated with an increased risk of attack frequency, pain intensity, and disability.44

Both modifiable and nonmodifiable risk factors for CDH exist.45,46 Nonmodifiable factors include female gender, low socioeconomic status, or previous head injury. Modifiable risk factors include stressful events, snoring, and obesity. Patients who are overweight and at increased risk of developing CDH have several management steps available to them:

  • 1Achieving and maintaining a normal BMI is recommended for migraine sufferers for reasons of general health. Based on cross-sectional studies, this may also reduce the risk of developing CDH.
  • 2If migraine preventive treatment is indicated, the goals are to:
    • • Improve the patient's overall general health.
    • • Reduce the risk of the patient developing CDH. Because obesity is a risk factor for CDH, preventive migraine medication should be strongly considered for obese migraine sufferers to reduce attack frequency.
      • • Avoid preventive medications that cause weight gain, which may have unintended negative consequences that partially offset their therapeutic benefits.
        • • Use preventive medications that are weight-neutral or promote weight loss in patients who are overweight, especially when patients have other coexistent illnesses that may benefit from weight loss (eg, diabetes, hypertension).
          • • Follow-up studies are needed to assess the influence of obesity and weight loss on migraine prognosis.

Migraine and Epilepsy.— The lifetime prevalence of epilepsy in the general population is about 1% at age 20 and 3.4% at age 80.47 While migraine is prevalent in about 13% of the general population, the prevalence of migraine in individuals with epilepsy was 24%.48–50 Among the first-degree relatives of these individuals with epilepsy, coexisting migraine was more likely for those relatives who had epilepsy than for those relatives who did not have epilepsy (24% vs 13%, adjusted RR = 2.4 [2.0 to 2.9]).48–50 In this study, the co-occurrence of migraine and epilepsy was stronger when epilepsy was due to trauma, associated with a positive family history of epilepsy, or associated with partial versus generalized onset seizures. A recent population-based study of Icelandic children provides additional data related to the relationship between migraine and epilepsy.51 Cases of first unprovoked seizure and newly diagnosed epilepsy were identified through an active surveillance system, with subsequent review by study neurologists to classify seizures. Age and gender-matched controls were selected from the population registry, and migraine with (MA) or without (MO) aura was diagnosed in both cases and controls using a structured interview. Results showed an association between migraine overall and developing a first unprovoked seizure (OR = 3.7 [1.6 to 8.3]), with a higher risk in females than males, partial as compared to generalized seizures, and higher for epilepsy rather than a single unprovoked seizures. Secondary analysis showed that the increased risk for first unprovoked seizure was confined to migraine with aura (MA: OR = 8.2 [2.3 to 28.9]; MO: OR = 1.4 [0.5 to 4.0]), and this increased risk associated with MA was evident across seizure types and causes.

To some extent, this evident comorbidity between epilepsy and migraine might be an artifact of diagnostic uncertainty, as there are shared characteristics of seizures and the migraine aura. As with other comorbidities, the mechanism might also be unidirectional (eg, postseizure headaches) or due to shared nongenetic (eg, head trauma) or genetic mechanisms.52,53

When managing patients with migraine and epilepsy, one must carefully consider how selected medications may affect the other condition.

  • • Drugs that lower seizure threshold should generally be avoided (eg, TCAs, SSRIs, and selected neuroleptics), especially if seizure control is less than optimal.
  • • A single dose of an anticonvulsant may prove helpful to treat both conditions. However, because dosing requirements may change, careful monitoring of improvements in each condition is needed.54
  • • As with many preventive treatments, appropriate testing periods are required and careful dose titration is needed to control both disorders.
  • • Selection of antiepileptic medications should be based on the type of epilepsy and patient characteristics. For example, when treating young women, one should avoid using divalproex sodium as first-line seizure control medicine because of its teratogenic risk.

Specific studies that assess the efficacy of treating comorbid epilepsy and migraine are needed. Future studies may need to assess changes in plasma folate cystine levels or ratios of folate/homocysteine plasma levels—emerging risk factors for migraine with aura that are potentially influenced by antiepileptic patents. The effect on migraine of ketogenic-type diets used to treat epilepsy is still unknown.

Managing Women's Issues and Migraine.—

Migraine and Menstruation.— Migraine affects approximately one quarter of women of childbearing potential.1 Hormone fluctuations, specifically falling estrogen levels, exacerbate migraine frequency.55-57 Falling estradiol levels, either naturally or with hormone therapy, may trigger a host of endogenous responses, all of which may potentially lead to an increase in migraine frequency. For example, changes in estradiol levels may cause changes in metabolism (affecting drug metabolism) or decrease serotonin production.58 It has been suggested that a decrease in sensitivity and serotonin receptor production occurs during menstruation.59 A decrease in natural endorphins and possible changes in dopamine metabolism may also occur, which may influence the vulnerability for migraine attacks.

Defining menstrual migraine versus migraine associated with menses is an ongoing classification debate. The most recent criteria recently published in the literature60 define pure menstrual migraine without aura as meeting the same diagnostic criteria as migraine without aura, but with attacks occurring exclusively between day –2 and day +3 of the menstrual cycle in at least 2 out of 3 cycles and at no other times of the cycle. Menstrually associated migraine occurs between day –2 and day +3 of the menstrual cycle in at least 2 out of 3 menstrual cycles and additionally may occur at other times. Clinical characteristics of menstrually associated migraine include:

  • • Slowly developing pain intensity.
  • • Progression to severe intensity.
  • • Long duration (up to 72 hours).
  • • High recurrence rate.
  • • Greater work-related disability compared with nonmenstrually related migraine.
  • • More likely than nonmenstrual migraine to occur in a patient on effective preventive medications.

Migraine During Pregnancy.— Because an estimated 50% of pregnancies are unplanned, exposure to medications potentially puts the fetus at risk.61 Approximately 10% of women will have their first migraine attack during pregnancy.61 Among those who already have a diagnosis of migraine less than 10% experience an exacerbation, but approximately 60% to 70% have less frequent migraines, particularly in the second and third trimesters.

Migraine During Perimenopause and Menopause.— Perimenopause, the transition from monthly menses to menopause, usually lasts 4 or more years. Menopause occurs between ages 43 and 57 and is recognized as the cessation of menstruation for 12 months. During the perimenopausal period, there is significant hormonal variation, and migraine may worsen in response to the fluctuating hormonal milieu. In menopause, with no ovarian follicles nor ovarian estradiol and progesterone secretion, the hormonal state is equivalent to premenarche.62 For this reason, menopause may be associated with a decline in migraine frequency in some patients.

Treatment Approaches to Migraine and Women with Hormonal Considerations.— Hormone fluctuations, whether due to pregnancy, oral contraceptives, hormone replacement, menopause, or other biological conditions, can exacerbate migraine. The increased incidence of migraine associated with menses makes for the consideration of “comorbidity” from an epidemiological view.63 The comorbid factors, which include pregnancy, perimenopause, estrogen replacement therapy, and oral contraceptives, may make treatment more complicated. One must consider therapeutic approaches for migraine as well as the comorbid situation. For example, oral contraceptives may prove effective for their stated purpose but have no effect on menstrually related migraine, or they may exacerbate or produce an improvement in migraine. A different formulation may provoke a different response by the migraine. However, the formulation that may best manage migraine may not be best suited for a given woman. Therefore, it may be necessary to optimize migraine management separately at the menses, and use migraine-specific medications that provide high reliability and tolerability for acute attacks. If the menstrually related attacks are frequent, severe, or poorly responsive to acute medications, then perimenstrual preventive strategies may be required in addition to long-term daily preventive treatment.

Migraine Treatment during Menstruation.— Treatment of migraine associated with menses differs from routine migraine management because of the former's predictability. Acute, preventive, and nonpharmacological treatment approaches have all proven effective. The extent to which therapies have been tested for migraine associated with menses varies. Acute treatments include nonsteroidal anti-inflammatory drugs (NSAIDs), triptans, ergotamine, dihydroergotamine, neuroleptics, opioid analgesics, and intravenous magnesium.64

Because of the predictability and disability associated with menstrually associated and pure menstrual migraine (migraine only during menses), preventive therapies have also proven helpful. These include the perimenstrual use of NSAIDs, triptans, ergotamine, dihydroergotamine, corticosteroids, magnesium salts, and exogenous hormone therapy. Conventional preventive drugs may also be effective. Adjustments can be made to oral contraceptives, including changing the brand, dose, or type, or eliminating the placebo week for 3 to 4 months. Some women benefit from continuous hormone therapy. Other possible preventive medications for menstrual migraine may include dopamine receptor agonists, antiestrogen agents, and GnRH analogs.65

Ongoing debate focuses on the possible role oral contraceptives play in exacerbating migraine. Four double-blind, placebo-controlled studies reported no differences in migraine incidence in the treatment versus the placebo groups.66 There is substantial variability in response to oral contraceptives: 3% to 35% of women with migraine improve with oral contraceptives, 18% to 50% show exacerbation, and another 39% to 65% report no change.67 However, symptoms that may require cessation of oral contraceptive therapy include the development of new or persistent headache, new onset of migraine with aura, increased frequency or intensity of attacks, and unusual or prolonged aura symptoms.

Migraine Treatment during Pregnancy.— A drug-free holiday during pregnancy may be desirable for many women with migraine, but this is not always practical. Frequent debilitating attacks of migraine during pregnancy, especially if associated with vomiting, may pose a special situation. In this circumstance, a risk-benefit assessment needs to be considered, weighing the effects of the migraine illness on the mother and developing fetus versus the potential effects of the medication on development. For a select group who need to take migraine medications during pregnancy, this may be the optimal approach. Others are able to manage with nonpharmacological approaches, such as rest, exercise, cognitive behavior therapy, hydration, diet, and prenatal vitamins and minerals. For those who need pharmacotherapy, the majority of migraine medications fall into FDA Category “B” or “C” and have accumulated safety data during pregnancy to a variable degree. There are also limited animal data. The class “D” rating suggests that well-defined risks are associated with such agents. List of classes for common acute and preventive medications is shown in Table 4.

Table 4.—. List of Medications for Migraine
DrugFDA Pregnancy Category68
  1. *Use during the third trimester should be avoided: Evidence of inhibition of labor, prolonged length of pregnancy, and decreased amniotic fluid. Some evidence of premature closure of the ductus arteriosus leading to pulmonary hypertension.

  2. #High doses associated with infertility, spontaneous abortion, or low birth weight.

Analgesics and NSAIDs
Opiates, barbiturates, 
 narcotics, sedatives 
Serotonergic agonists 
Miscellaneous medications
 Divalproex sodiumD
Calcium channel blockersC
Tricyclic antidepressants

Migraine Treatment during Perimenopause and Menopause.— Hormone fluctuations during perimenopause often lead to changes in migraine status. Treatment of such a variable condition can be challenging, and several different approaches may need to be tested to find one that works. Many women experience menopausal symptoms (hot flushes, etc) and require hormone replacement, which may, in turn, exacerbate headaches.69 For these women, adjustments can be made to help improve headache control, including:

  • • Reducing the dose of estrogen.
  • • Changing the type of estrogen from conjugated estrogens to pure estradiol.
  • • Using an oral contraceptive continuously to provide stable hormone levels.
  • • Switching to a selective estrogen receptor modulator.
  • • Changing or eliminating dose of progesterone.
  • • Switching delivery systems from oral to vaginal or transdermal.

Interestingly, women with surgical menopause often experience a deterioration of their migraine condition, therefore surgery is not recommended for women with hormonally related migraine.70


This round table discussion reviewed the limitations and strengths of different treatment approaches to managing migraine and commonly associated comorbidities. The main limitations to using a single medication to treat 2 separate illnesses include the risk of treating neither condition adequately, treating only one condition adequately while leaving the patient exposed to the risk of suboptimal treatment of a comorbid condition, or pushing a single medication dosage sufficient to manage both conditions but with unacceptable side effects. Therapeutic independence, wherein each medical condition is treated separately, may be most helpful for some patients. Such an approach requires health care providers to assess each patient's treatment priorities and manage the most disabling condition optimally. Specific attention needs to be given to each condition regarding dosing, drug compatibility, and tolerability.

As medical care becomes more sophisticated and a better understanding of the biology underlying migraine and other conditions is elucidated and improved, targeted pharmacotherapy will follow. Studies are needed to specifically study the unidirectional and bidirectional effects of illnesses that are comorbid or concomitant with migraine. For example, if a patient has a disorder such as pain, controlling the pain syndrome is likely to have an effect on other disorders, such as migraine or depression. Alternatively, will improving the status of migraine have a positive impact on depression—or vice versa?

Studies that evaluate how to manage the patient with behavioral illnesses and other illnesses that coexist with migraine are clearly needed. Behavioral and cognitive treatment approaches alone have been proven effective in migraine,7 but combining this approach with pharmacotherapy to improve management of coexisting conditions has not been well studied. Ultimately, the goal of treatment for these patients is the optimum management of migraine as well as the associated comorbidities, which hopefully will lead to meaningful improvements in quality of life.

Conflict of Interest:  Disclosures and Rationale for Funding: Funding for this project was provided by Ortho-McNeil Neurologics, Inc. All authors of this paper serve as scientific consultants or receive funding from Ortho-McNeil Neurologics, Inc., other pharmaceutical companies, nonprofit organizations (AAN, AHS), and/or NIH research funding grants. The purpose for funding this project was to provide a literature review that will help increase awareness of the complexities of managing migraine in patients who have comorbid conditions, improve care of these patients and also identify areas needed for future research in order to establish evidence-based guidelines for the management of migraine patients with associated comorbid conditions.