Prevalence and Clinical Features of Migraine

Migraine is an episodic headache disorder with a 1-year prevalence of approximately 18% in women, 6% in men, and 4% in children. Three methodologically identical studies were conducted during a 15-year period with samples of more than 20,000 in the American Migraine Study I (AMS-I); 30,000 in the American Migraine Study II (AMS-II); and 160,000 in the American Migraine Prevalence and Prevention Study (AMPPS) . They demonstrated that migraine prevalence has been stable in the United States over more than a decade. In all three studies, the average female-to-male migraine prevalence ratio was around 2.8, with a peak of 3.3 between 40 and 45 years. The ratio remained above 2.0 even after the age of menopause.

Migraine attacks occur in 6% of women and 4% of men in their teenage years, but reach the peak between ages 30 and 39 with 24% women and 7% men affected. The prevalence remains high in people in their 40s and 50s, but gradually decreases to 5% in women and 2% in men when people enter their 60s . Clearly, migraine is most common in the childbearing and child-rearing years (ages of 25–55), so pre-pregnancy, pregnancy, and postpartum issues are important considerations for health care providers.

The Ad Hoc Committee on the Classification of Headache described headache of migraine type as recurrent attacks of headache, widely varied in intensity, frequency, and duration. The attacks are commonly unilateral in onset; are usually associated with anorexia and sometimes with nausea and vomiting; some are preceded by, or associated with, conspicuous sensory, motor, and mood disturbances; and are often familial . Migraines can be divided into two major subtypes: migraine with aura (MA) and migraine without aura (MO). MO is the most common type, accounting for approximately 80% of cases. The diagnosis of MO requires at least five lifetime attacks, lasting 4–72 h, with at least two of four pain features and at least one of two sets of associated symptoms ( Table 2.1 ). MA is characterized by the presence of reversible focal neurologic symptoms that usually develop over 5–20 min and last for <60 min. Visual aura is overwhelmingly the most common ( Table 2.2 ) . The aura usually precedes the headache, although in many instances the aura can occur during or after the headache. More than in any other headache disorder, migraine sufferers identify triggers. Stress is the trigger most commonly listed by patients. Dietary factors also are frequently reported, although few have been scientifically validated. Oversleeping and sleep deprivation are commonly recognized triggers. Hormonal headaches are triggered by variations in female estrogen levels and possibly other hormonal factors. Noise, bright lights, and fumes are commonly identified migraine triggers. Physical exertion can cause exercise-induced migraine .

Effect of Pregnancy on Migraine

There are a number of important issues to be addressed by the physician when a woman with migraine becomes pregnant. Below are some common questions that migraine patients might raise.

• •
  

  I’m pregnant … what’s going to happen to my migraine?

  
  
• •
  

  I’m pregnant … is migraine going to harm my baby?

  
  
• •
  

  What’s going to happen to my migraine after I have the baby?

In answer to the first question, migraine in women is influenced by hormonal changes throughout the life cycle: menarche, menstruation, oral contraceptive use, pregnancy, menopause, and hormonal replacement therapy. Somerville’s work in the early 1970s suggested that falling estrogen levels were the provocative factor in menstrual migraine . Studies examining the natural history of migraines in pregnancy have suggested that migraine generally improves during pregnancy. Studies conducted after 1988 all used the IHS classification of migraine, improving the standardization of headache diagnosis. However, the majority of these studies is retrospective, the patients having been evaluated several years after pregnancy or, at best, in the postpartum period . For example, in the Collaborative Perinatal Project, Chen and Leviton retrospectively reviewed 55,000 pregnancies in the United States. Only 2% of these women had self-reported migraine. Of the 484 cases analyzed, 17% had a complete remission and another 62% showed some improvement with pregnancy. Granella et al. found that headache improvement during pregnancy occurred in only 43% of patients who had MA, compared with 72% of patients who had MO . A large Norwegian population-based study examined the effect of pregnancy and parity on headache prevalence . This study found that although the headache prevalence was lower among pregnant than among nonpregnant women, the association between headache and pregnancy was significant for nulliparous (one who has never given birth; OR = 0.5, 95% CI = 0.4–0.7), but not for primiparous (one child only) and multiparous (several children) women (OR = 0.8, 95% CI = 0.7–1.0). This was evident for both migraine and nonmigrainous headache.

A limited number of prospective studies has been conducted on the pattern of migraine during pregnancy . Sances et al. prospectively investigated the course of migraine during pregnancy and postpartum in pregnant women attending an obstetrics and gynecology department for a routine first trimester antenatal checkup. Forty-nine migraineurs (2 with MA and 47 with MO) who had experienced at least one attack during the 3 months preceding pregnancy were identified, enrolled in the study, and given a headache diary. Examinations were performed at the end of the second and third trimesters and 1 month after delivery. Migraine improved in 46.8% of the 47 MO sufferers during the first trimester, in 83.0% during the second and in 87.2% during the third, while complete remission was attained by 10.6%, 53.2%, and 78.7% of the women, respectively. Although the headache of migraine is frequently better with the stable, high estrogen levels of pregnancy, aura may occur more frequently or for the first time during pregnancy .

Recurrence during the postpartum period is common . Sances et al. found that migraine recurred during the first week after childbirth in 34.0% of the women and during the first month in 55.3%. Recurrence during the postpartum period is significantly less frequent in women who breast-feed . The mechanism underlying this finding is unknown. Rather than bottle feeding representing a risk factor, it could be hypothesized that breast-feeding exerts a protective action, as breast-feeding increases the levels of the antinociceptive hormones vasopressin and oxytocin and contributes to successful natural bonding . Women with menstrual migraine are most likely to experience postpartum headaches. Falling estrogen levels after delivery are blamed for this phenomenon, although lactation may delay or decrease the rate of estrogen lowering, thus decreasing the postpartum headache.

Effect of Migraine on Pregnancy

Migraine does not seem to directly affect the outcome of pregnancy . Wainscott and Volans et al. found that the incidence of miscarriage, toxemia, congenital anomalies, and stillbirth was not increased in a sample of 777 migraine sufferers compared with the national averages or controls. A retrospective case–control study of over 38,000 deliveries showed that mean gestational age and birth weight in babies born to women with migraine did not differ from the general population, and the proportions of low birth weight and preterm births were the same . However, migraine appears to be associated with several maternal pregnancy complications. Women with migraine during pregnancy may be at increased risk of gestational hypertension and pre-eclampsia that may appear earlier and be more severe . A large population-based study found that migraineurs, particularly those with aura, have a higher cardiovascular risk profile than individuals without migraine. Women with migraine were more likely to have been diagnosed with gestational hypertension, even after taking into account age and number of pregnancies . In a case–controlled study of pregnancy discharges, inpatient coding for peripartum migraine was associated with ischemic stroke. The results from this study also indicated associations between peripartum migraine and heart disease, venous thromboembolism, and cardiovascular risk factors such as hypertension, diabetes, and smoking . However, due to the confounding presence of gestational hypertension, pre-eclampsia, and eclampsia, this study could not determine the independent effect of migraine on vascular disorders. A recent 3-year nationwide population-based study in Taiwan revealed that women with migraines were at increased risk of having low birth weight infants, preterm birth, pre-eclampsia, and delivery by cesarean section, compared with unaffected mothers; however, given the large size of the sample used in this study, overpower might be a concern, giving small differences statistical significance which may not actually be clinically significant. In addition, the database did not include complete information regarding medications taken during pregnancy, the confounding role of medications in the relationship between migraines, and adverse birth outcomes were not assessed.

Tension-Type Headaches

Tension-type headache (TTH) is the most common type of primary headache, with 1-year prevalence of approximately 40% in the United States . In contrast to migraine, the main pain features of TTH are bilateral location, nonpulsating quality, mild-to-moderate intensity, and lack of aggravation by routine physical activity. The pain is not accompanied by nausea, although just one of photo- or phonophobia does not exclude the diagnosis.

The International Classification of Headache Disorders-2 (ICHD-2) distinguishes three main subtypes of TTH ( Table 2.3 ):

• 1.
  

  Infrequent episodic TTH, with headache episodes less than 1 day a month

  
  
• 2.
  

  Frequent episodic TTH, with headache episodes 1–14 days a month

  
  
• 3.
  

  Chronic TTH, with headaches 15 or more days a month

In a recent population-based study using the ICHD-2 criteria to classify TTH subtypes, a Danish twin registry found that the 1-year-period prevalences of infrequent episodic TTH, frequent episodic TTH, and chronic TTH were 63.5%, 21.6%, and 0.9%, respectively .

TTH affects women slightly more often than men. The age of onset is usually between 20 and 30 years, and peak prevalence is between the ages of 30 and 39 for both sexes . Clinical experience would suggest that TTH is less likely to improve during pregnancy because it is not hormonally mediated. However, TTH has not been extensively studied in pregnant women. One study noted that 48% of migraine patients improved during pregnancy compared with only 28% of patients with TTH .

Cluster Headache

The prevalence of cluster headache (CH) in reproductive-aged women is 7.5 of 100,000 women with a male to female ratio 7.5 to 1–2.5 to 1 . A rise in female prevalence of cluster headache may perhaps be due to changes in women’s lifestyles as they have entered the workforce. We are now slowly starting to see pregnant patients with CH. It has been shown to have a close association with both smoking and alcohol ingestion and sleep apnea, and because comorbidities can affect the outcome of pregnancy, these should be addressed .

Hormonal Influences on the Central Nervous System and Migraine During Pregnancy

During pregnancy, hormonal changes occur that affect the nervous system. The placenta becomes the primary site of production of gonadal steroids. Other minor sources of steroids include the ovary, adrenal gland, and brain. Serum levels of estradiol/estriol and progesterone begin to rise in the mother during the sixth to eighth weeks of pregnancy as the placenta begins to produce steroids. They continue to increase to their highest levels gradually, representing a 30- to 40-fold increase of estradiol and a 20-fold increase of progesterone during the third trimester .

Estradiol/estriol and progesterone directly affect the central nervous system (CNS). Pregnancy hormones passively diffuse across the blood–brain barrier (BBB) and act on the neurons and glia. Estrogen receptors (ER) and progesterone receptors (PR) are expressed in various regions of the brain and implicated in many different neuronal functions. As listed in Table 2.4 , researchers demonstrated ERα and ERβ receptors in the TNC, brain stem nuclei, and many other regions of the brain .

The action of estrogen at the molecular level is carried out via nongenomic and genomic mechanisms. Through the nongenomic mechanism, estrogen uncouples μ-opioid and γ-aminobutyric acid type B (GABA _B ) receptors from their effector systems, whereas allopregnanolone increases the opening time of GABA _A receptors . Through the genomic mechanism, gene expression levels can be changed and thus neuron function is modulated.

In other studies, PR, including PRA, PRB, and 7-transmembrane progesterone receptor β isoform (7-TMPR), were expressed at different levels in many regions throughout the brain . Their levels may vary between regions and change in response to stimulation and different hormonal states, such as the phases of the menstrual cycle, pregnancy, and postpartum period. Estrogen induces PRA isoform expression, whereas progesterone does not affect the expression of any PR isoforms .

The fact that migraine improves during pregnancy is at least partially attributed to the lack of fluctuation of estrogen levels during pregnancy. Numerous clinical and animal studies support the generally accepted concept that fluctuating estrogen levels may trigger migraine. Either a sharp increase or a drop in estrogen level was shown to be associated with migraine attacks, while little evidence suggests that progesterone can independently modulate or trigger migraine .

In human genetic studies, there are controversial reports on the ER and PR linkage to migraine. One study reported PR linkage with migraine-associated vertigo . Another study found that PROGINS allele, PR marker, interacted synergistically with the ESR 1 594A allele, which alone was reported to be linked to migraine, to increase the risk of migraine by a factor of three . A recent study investigated the role of common polymorphisms in the ER and PR genes in the risk for migraine in a Spanish population. Surprisingly, it did not support a major contribution of ER1 and PR to the pathogenesis of migraine . However, many other physiologic and pharmacologic studies on animals and humans do suggest at least a modulatory effect of these neurohormones on migraine, through interactions with almost all of the major neurotransmitter systems, including serotonin, opioid, glutamatergic, and GABAergic as well as the sympathetic and nonsympathetic systems .

The serotonergic system modulates the functions of many regions in CNS. The dysfunction of this system leads to many disorders including mood disorders, neuro-degenerative disease, and neurodevelopment disorders . Migraineurs have been demonstrated to have low plasma serotonin interictally, with a 60% surge during attacks . In the brain, serotonin is synthesized by tryptophan hydroxylase type 2 (TPH2) . Levels of serotonin at the synapse are primarily determined by the serotonin reuptake transporter (SERT), which was linked to migraine in recent genetic studies . Estrogen was also found to increase TPH expression in the monkey, whereas the addition of progesterone had no significant effect on this enzyme . Estrogen and/or progesterone also modulate levels of SERT, and monoamine oxidases A and B likely through ER, PR, and NF-KappaB .

Following the earlier studies, a functional study of dorsal raphe neurons showed that during pregnancy a graded augmentation and diminution in 5-HT neuron firing rate occurs. This mirrors that of circulating progesterone concentrations. Such firing rate change in dorsal raphe neurons seems to be independent of LC activity during pregnancy . These studies suggest that estrogen and progesterone may modulate the serotonin system through distinct mechanisms and participate in different physiologic functions. Regardless of the estrogen/progesterone-related 5-HT changes in raphe neurons, higher serotonin levels during pregnancy have been noted for years. This might explain the pleasant emotion and fewer headaches often observed during pregnancy.

Both estrogen and progesterone are implicated as the major gonadal hormones that modulate migraine severity during pregnancy. Increased levels of progesterone may cause changes in nervous system function important to pain-related mechanisms. These include increased release of endorphins , altered neuropeptide expression level and vascular sensitivity to them , decreased sensitization to glutamate or decreased α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor binding , and increased activity of GABAergic neurons .

ER and neuropeptides, including pituitary adenylate cyclase-activating polypeptide (PACAP), substance P (SP), calcitonin-gene-related-peptide (CGRP), have been found in human TNC and brain stem nuclei, including the LC, PAG, and raphe magnus (NRM) . CGRP is the most studied among them. It is the most potent peptide vasodilator known and can function in the transmission of pain. 5-HT1B/D agonists, such as sumatriptan, increase intracellular calcium, which causes decrease in CGRP promoter activity . 17β – estradiol and progesterone cause vasorelaxation at high nonphysiologic concentrations via rapid nongenomic pathways, but they may also manifest their actions via slow genomic pathways .

The expression of CGRP receptors is augmented during pregnancy, probably through ER _α . In rats, pregnancy had a trophic effect on trigeminal perivascular innervation. The increase of CGRP in trigeminal innervation during pregnancy may be more related to nociception than in control of resting cerebral blood flow . Receptor antagonists such as telcagepant, which is currently in phase III trials, have promise in limiting the effects of CGRP . Although it is hoped that this new drug will be effective in migraine, its safety and availability for use in pregnancy will not be known for years.

There are few studies in the literature that can explain why some women experience new onset of migraine or exacerbation of migraine during pregnancy. Perhaps the imbalance between estrogen and progesterone is associated with such headache, as animal studies found that estrogen may facilitate neuronal excitation while progesterone apposes it . Future studies on the linkage between PR allele or neuropeptide allele and pregnancy onset of migraine might prove worth the effort.

Tension-Type Headache and Cluster Headache

TTH is less studied than migraine. Clinical studies showed less improvement of TTH during pregnancy than migraine . The hypothetical pathophysiologic mechanism of tension headache shares some common features with migraine, including involvement of the trigeminal system and central sensitization. Chronic TTH (CTTH) can be associated with referred pain from active trigger points (TrPs) in the posterior cervical, head, and shoulder muscles mediated through the spinal cord and the brain stem TNC. TrPs may represent the primary hyperalgesic zones responsible for the development of central sensitization in CTTH . Pregnancy-related changes in endorphins and GABAergic neurotransmitter systems may apply to the attenuation of headache observed in pregnant patients with CTTH.

CH is rare in pregnancy. The acute attack of pain may be regarded as a manifestation of the trigeminal-autonomic reflex. The clinical feature of circannual and circadian periodicity, the neuroendocrine changes, and the functional imaging studies implicate the hypothalamic region as fundamental to CH . A population-based questionnaire study was performed among 224 female CH patients, and the possible effect of hormonal influences on CH attacks studied. It was found that menstruation, use of oral contraceptives, pregnancy, and menopause had a much smaller influence on CH attacks than in migraine .

Nonpharmacologic Treatments

Because of the uncertain safety of fetal exposure to most pharmaceuticals, many women with problematic headaches and migraine turn to nonpharmacologic treatments in pregnancy. For many acute attacks, rest, reassurance, and ice packs will be sufficient . The American Academy of Neurology guidelines list pregnancy, planned pregnancy, or nursing as some of the characteristics in which nonpharmacologic treatment options may be chosen, among others. These therapies can also be adjunctive to medication to improve outcome. They are categorized as physical therapies and behavioral treatments. Although some behavioral therapies are recommended with Grade A evidence to support their efficacy, evidence-based recommendations are not available for hypnosis, acupuncture, transcutaneous electrical nerve stimulation, chiropractic or osteopathic cervical manipulation, occlusal adjustment, and hyperbaric oxygen as preventive or acute therapy for migraine . It is important to note that herbal remedies are not listed in this category, as they may indeed have pharmacologic effects and side effects which may or may not be known.

Complementary and Alternative Medicine in Pregnancy

In a recent survey of providers in Australia, the majority of obstetricians and midwives had a positive view of complementary and alternative medicine (CAM) including massage, acupuncture, vitamins, yoga, meditation, and hypnosis for a variety of ailments in pregnancy. However, 72% of obstetricians thought there was a need for more evidence . Patients often turn to CAM independently, with or without health care provider input. A survey of pregnant women in Germany found that approximately half used CAM during pregnancy; headache was not addressed. Of those who consulted a provider, midwives were the main prescribers of CAM therapies. CAM often included herbal treatments that were not prescribed by or reported to their health care provider. Because of this reason, the authors concluded that the practice of using CAM in pregnancy and delivery should be used with caution, and providers should rely on those treatments backed by some positive evidence .

Physical Therapies

Physical treatments for headache include chiropractic manipulation and/or mobilization of the cervical spine (CMM), physical therapy (PT), acupuncture, massage, traction, craniosacral therapy, and TrP therapy. PT is often used in the treatment of TTH. It includes improvement of posture, relaxation, exercise programs, hot and cold packs, ultrasound, and electrical stimulation . Patients are more likely to benefit from exercise and more active PT strategies than from passive techniques . Numerous case reports , series , and reviews indicate high utilization of CMM (up to 12 million people annually) and good results in many instances of headache and neck and back pain . CMM may be associated with an initial increase in headache intensity if delivered too vigorously. It is better tolerated when gentle muscle stretching and manual cervical traction precedes CMM and it is slowly advanced to include strengthening and aerobic conditioning . Sometimes anesthetic blocks or neurolytic procedures also are used for temporary pain relief during therapy, but this has not been addressed in pregnancy. It is presumed that cervical techniques exert their effect on headaches by activating descending pain inhibitory systems from the brain stem/periaqueductal gray and/or activation of β-endorphin-driven endogenous anti-nociception , thus minimizing chronic nociceptive input into the CNS .

Due to the nature of these treatments, sham controls are difficult to produce in a scientific research setting, limiting conclusions about safety and efficacy, or providing contradictory results. In one study, researchers found no supporting evidence in eight randomized controlled trials for the efficacy of CMM in the treatment of headache . For the treatment of TTH, another review found no rigorous evidence that manual therapies (CMM, connective tissue manipulation, soft tissue massage, traction, or craniosacral techniques) provided any positive effect . Another systematic review in 2005 showed evidence of better efficacy of PT compared to CMM (which had possible minor benefit), acupuncture, or massage in frequent TTH. They found some benefit in migraine patients treated with a combination of PT, thermal biofeedback, relaxation training, and exercise, but found CMM to be probably more effective in TTH than in migraine . More recent studies published in 2010 indicate that CMM is effective for migraine and cervicogenic headache, but inconclusive for TTH; and for neck pain, massage was most effective , and that cervicogenic headache patients seem to fare slightly better with CMM than they do with light massage .

A meta-analysis of eight studies involving acupuncture in the treatment of TTH concluded that it had limited efficacy when compared to sham controls , whereas a recent Cochrane review of eleven trials including over 2300 patients with episodic and/or chronic TTH revealed there may be relevant short-term benefits of adding acupuncture to routine care, and they found significant efficacy over sham interventions . Acupuncture was effective for pregnancy-associated nausea and for dental pain in a review of this modality in pregnancy, but the data were equivocal or contradictory as pertains to chronic pain, back pain, and headache, despite activation of endogenous opioid mechanisms . A randomized controlled trial (unblinded and with no sham procedure) looking at medical management versus medical management plus acupuncture for chronic daily headache revealed that the latter resulted in improvements in health-related quality of life and less suffering due to headaches . More recent reviews of acupuncture in pregnancy corroborate the benefit in treating nausea, with some effectiveness in managing back and pelvic pain and assisting in labor. Headache was not mentioned . Reflexology, on the other hand, has not demonstrated effectiveness in treating any medical condition including headache . A controlled trial of craniosacral therapy for migraine is currently underway .

Most chiropractors agree that CMM is safe for pregnant patients with neck or back pain, although a majority avoids manipulation in the presence of comorbidity such as pre-eclampsia, hypertension, blood-clotting disorder, or ectopic pregnancy; and some feel it may be inappropriate in patients with gestational diabetes or in the case of multiple pregnancy . Patients undergoing spinal manipulation do need to have adequate informed consent as to possible adverse events which, although are rare (5–10 per one million CMMs) can be catastrophic. Carotid and/or vertebral artery dissection , spinal cord injury, cervical nerve root damage, phrenic nerve injury, epidural hematoma, ruptured disk, and vertebral fractures can follow cervical manipulation. Patients with a known connective tissue disorder such as Ehlers–Danlos and Marfan syndrome may be more susceptible to arterial dissection; manipulation-induced vertebral fractures can occur more commonly in patients with underlying weakened vertebral bodies due to angiomata or osteoporosis. Because screening cervical radiologic tests are contraindicated in pregnancy, these patients may be at higher risk for spinal complications . Fortunately, most pregnant women are young with relatively healthy bones.

Behavioral Therapies

Pregnancy is an ideal time for the patient with problematic headaches to explore behavioral therapies such as cognitive behavioral therapy, relaxation training, biofeedback, and hypnosis. This is also a time when discussions about lifestyle changes in the patient’s diet, sleep, and exercise habits is especially appropriate. Trigger factors should be re-analyzed and avoided when possible. Bed rest or reduced work and home responsibilities may be helpful for some . For the few that suffer from lactation headache, simply putting the baby to the breast, resulting in triggering the milk-ejection reflex, may relieve the symptoms .

Cognitive behavioral therapy can teach patients to identify thoughts and beliefs that generate stress and aggravate headache, and address coping strategies. It is probably most effective in patients with psychologic issues or problems such as chronic work stress, mood disorders, or adjustment disorders . The exact efficacy of this modality is difficult to estimate, but is associated with very little harm if the patient is open to it. In addition, it can open the door for more intimate psychotherapy if appropriate.

Biofeedback is a method for learned control of physiologic responses of the body. These include both the voluntary system and the involuntary or autonomic system. The patient can thus learn to influence the skeletal muscles, heart rate, vascular responses (frequently indirectly measured as temperature), and sympathetic discharges (measured by the electrical galvanic skin response) to emotional or physical stimuli. With practice, biofeedback training and relaxation training results in a less excitable basal state, which may result in fewer headaches . In pregnant women with recurring headache, stress management and biofeedback have been shown to decrease headache frequency by 50–79%, with lasting benefit for a year . There are no recent studies proving benefit of hypnosis in headache treatment due to a lack of available sham procedures , but some patients may respond to this as a form of relaxation therapy.

Supplements and Herbs

Supplements and herbs are often misunderstood by the lay public, and the possible risks associated with them are underestimated. Therefore, patients should be adequately informed, especially those who are planning pregnancy. During patient encounters, health care providers should ask if patients are taking any of these treatments that may have not been prescribed. It is not unusual for patients to seek alternative advice. One recent observational study found that 82% of health food store clerks asked to recommend a treatment for migraine and nausea in pregnancy did so readily. Their investigation revealed that 5% of the recommended products were specifically contraindicated in pregnancy; for most others the safety was unknown .

Most supplements and herbs should be avoided in pregnancy. The exceptions are oral magnesium and riboflavin (vitamin B2). Magnesium is approved by the FDA for migraine prophylaxis during pregnancy. Riboflavin has shown efficacy in decreasing frequency of migraine . There is no danger of toxicity associated with large doses of riboflavin. It is readily excreted in the urine and absorption by the digestive tract is relatively low. Pregnant patients should avoid butterbur root ( Petasites hybridus ) which can contain pyrrolizidine alkaloid metabolites that are toxic and carcinogenic and have been associated with Budd–Chiari syndrome (occlusion of the hepatic veins) . Feverfew may be oxytocic and may be teratogenic, based on rat studies . Co-Q 10 is not recommended in pregnancy or lactation . Caffeine may be used safely if overuse is avoided .

Off-the-Shelf Analgesics and Pharmaceuticals

The safest simple off-the-shelf analgesic for acute headache treatment in pregnancy and the postpartum period is acetaminophen, despite transient effects on the uterus and platelet function . The nonsteroidal anti-inflammatory drugs (NSAIDs), ibuprofen and naproxen, should be avoided while the patient is attempting to conceive due to interference with implantation of the embryo. NSAIDs also may cause premature closure of the ductus arteriosus and fetal pulmonary hypertension and so are contraindicated in the third trimester, although they are compatible with breast-feeding . Aspirin should be avoided in the third trimester as it may inhibit uterine contractions, increase bleeding of mother or baby, and also cause narrowing of the ductus arteriosus . Phosphorylated carbohydrate solution (Emetrol), doxylamine (Unisom), or pyridoxine (Vitamin B6) can safely be used for nausea .

Prescription Medications for Acute Treatment

The goal of pharmacotherapy in pregnancy is to use the fewest and safest drugs at the lowest dose possible to achieve improvement in comfort and function, and avoid dehydration. Often the first line of treatment in a patient with a severe headache with vomiting is intravenous fluids and an intravenous antiemetic . For the acute headache, prescription medications may be necessary for severe headache attacks. Most opioid analgesics are safe when used judiciously; they are rated as B unless use is prolonged or at term, in which case they have a D rating due to neonatal dependence . Propoxyphene and codeine are category C drugs. Propoxyphene has been loosely associated with various malformations in early case reports . Codeine has been associated with cleft palate and other midline malformations, such as inguinal hernias . The sedating phenothiazines (category C) chlorpromazine, prochlorperazine, and promethazine, or metaclopromide (category B) may also be used. For a prolonged attack, prednisone (category B) or dexamethasone (category C) can break the headache cycle . Barbiturates and benzodiazepines should be limited or avoided altogether due to their association with medication overuse headache, neonatal withdrawal syndromes, and possible fetal neurodevelopmental problems in the case of barbiturates .

Ergotamine and dihydroergotamine are to be strictly avoided, as they are oxytocic and teratogenic. Increased uterine tone and impaired placental blood flow can result in spontaneous abortion or fetal distress. Congenital defects have been reported including intestinal atresia, neurodevelopmental disorders, and Mobius syndrome .

The use of triptans has been studied extensively mainly through pregnancy registries organized and run by the pharmaceutical industry. The GlaxoSmithKline (GSK) registry is the longest and largest database to date, listing 829 pregnancies exposed to sumatriptan as of October 2008 . The total number of major birth defects (MBDs) associated with sumatriptan exposure in the first trimester in this registry was 4.6%, excluding spontaneous abortion, induced abortion, and fetal deaths in which no defects were reported. Similarly, 4.8% of those fetuses exposed during the second trimester were born with MBDs, and none of 12 exposed in the third trimester, with an average of 4.7% across all trimesters. For comparison, data from general population studies report the incidence of MBDs as 1.6–2.2% (without versus with genetic defects) in a study in 1989, 2.1–2.7% (at 1 week versus 6 years) in 2007, and 3.4% in migraineurs in 1978. This suggests there is no substantial increase in MBD risk (when defined as greater than a two- to threefold increase) following exposure to sumatriptan thus far. Of all the MBDs recorded, ventricular septal defects were noted somewhat more commonly, comprising 0.9% of all first trimester exposures in the GSK registry, and 1.1% among pregnancy outcomes in the Swedish Medical Birth Register.

Researchers of the Norwegian Mother and Child Cohort Study, a major database of almost 70,000 primaparous women, found MBDs in 3.0% of the triptan-exposed group, 5.9% of the migraine control group, and 5.0% in the nonmigraine control group . Further, they found that medical and obstetric complications were somewhat higher in the triptan-exposed group. Compared to nonmigraine controls, women who took triptan medication during pregnancy were more likely to experience emesis, fever, hypertension in the first trimester, pre-eclampsia/eclampsia, folate-deficiency anemia, hospitalization, and vaginal bleeding. Compared to migraine controls they suffered more often from folate-deficiency anemia and vaginal bleeding.

Triptan use late in pregnancy was associated with a slight increase in risk of atonic uterus and hemorrhage during labor. This could be due to serotonin effects. Serotonin stimulates myometrial contractions via 5-HT2B receptors suggesting a possible link between severe migraine and impaired uterine contractility. It is also known that serotonin has relaxant effect on porcine myometrial muscles via 5-HT7 receptors, at which triptans have an agonistic effect. At the conclusion of the Norwegian publication it was stated: “While it is important to exert caution when using any medication during pregnancy, this study indicates that migraineurs can continue an already established triptan therapy or start using triptans during pregnancy without any major risk of adverse pregnancy outcomes .” However, most experts agree at this time that the total reported number of exposures to triptans in pregnancy is still too low to definitively discern the safety of these medications or to recommend their use.

Prescription Medications for Preventive Treatment

If migraine headaches become excessively frequent (3–4 prolonged, severe attacks per month) or disabling, preventive treatment may rarely be indicated, for the reasons already stated earlier. None of the medications commonly used for migraine prevention have a category A or B rating for use in pregnancy. Propranolol (C) is the most commonly used β-blocker, but there are reports of intrauterine growth retardation, hypoglycemia, bradycardia, and respiratory depression associated with this drug . It is probably safe to use in lactation . Pindolol is a nonselective β-blocker which has a category B rating , but unfortunately, perhaps due to its intrinsic sympathetic activity, it has no demonstrated efficacy in migraine prophylaxis . Labetalol (C) has also been proposed for use during pregnancy due to its unique selective α-1 and nonselective β1 and 2 adrenergic antagonist effects. It is the most commonly used medication for treating hypertension in pregnancy, as it decreases systemic vascular resistance with little change in cardiac output. Therefore, it does not impair placental blood flow, and is associated with good fetal outcome . It has not been associated with any problems in lactation .

Selective serotonin reuptake inhibitors (SSRIs) may be considered with caution in a pregnant migraineur, especially with comorbid depression, as most are category B or C . The exception is paroxetine (D), which has been linked to an increased risk of congenital heart defects . Fluoxetine has both an FDA rating of B and a TERIS rating of N . Its safety in lactation is unknown, but is associated with no significant levels in breast-fed infants and is rated as L2 by Hale . The tricyclic antidepressants amitriptyline and nortriptyline are both category D; doxepin, desipramine, and protriptyline are category C and are probably safe in lactation . The anticonvulsants may be considered with comorbid migraine and epilepsy. Gabapentin and topiramate are most often suggested for use in pregnancy with an FDA category C rating . There are reports of craniofacial abnormalities and hypospadias associated with topiramate exposure, and gabapentin may cause bone malformations . These drugs have unknown safety in lactation with L2 and L3 Hale ratings, respectively . Valproic acid (D) is associated with significant fetal defects, especially neural tube defects, and is to be avoided before and during pregnancy. Ironically, it is not associated with any problems in lactation .

In order to minimize infant exposure to drugs during breast-feeding, the patient can be educated as to the timing of administration (take directly after feeding) and to pump and discard the first milk after administration of their medication . Medications that may be used safely while breast-feeding include acetaminophen, moderate amounts of caffeine, opioid analgesics, and most NSAIDs. Those that should be avoided include bromocriptine, ergotamine, and lithium. Eletriptan has a Hale L2 rating whilst the other triptans are L3 and may be used with caution. The benzodiazepines, phenobarbital, most antidepressants, neuroleptics, aspirin, and ketarolac also may be used with caution .

Treatment of Nonmigraine Primary Headache Types in Pregnancy

The management of TTH in pregnancy is very similar to migraine treatment. Acetaminophen and NSAIDs with nonpharmacologic strategies are used most often. Preventive treatment with prescription medication is rarely required.

For CH in pregnancy, 100% high-flow oxygen via facemask is safe and can abort up to 70% of attacks. As discussed earlier, there is limited data concerning the use of triptans in pregnancy. Prednisone is safe to use in pregnancy and lactation and can stop or attenuate the cluster bout. Verapamil, a long-standing preventive for CH, has an FDA category C rating, but is compatible with breast-feeding . Intranasal lidocaine (4% solution) may be tried as well as intranasal capsaicin: no fetotoxic properties known in animals but there is no human data. Greater occipital nerve blockade can also be used in pregnancy. Lithium, methysergide, and valproic acid are contraindicated .

Trigeminal neuralgia frequently responds to carbamazepine. Recently, the rate of MBDs of infants born to patients with epilepsy and carbamazepine monotherapy (3.0%) was reported to be close to that of controls (2.1%) , and general IQ scores of exposed children of epileptic mothers were not reduced compared to controls in a systematic review and meta-analysis . However, the FDA category for this drug is category D and many malformations have been seen in mice pups when exposed during organogenesis . The potential risks probably usually outweigh the benefit for nonepileptic pregnant patients, but it is safe for lactating mothers to use postnatally.

Inadvertent Exposure

In the event of an inadvertent drug exposure during pregnancy, the health care provider should document all details of the exposure (drug, dosage, duration, and gestational age), the FDA and TERIS rating of the drug, the patient’s health and family history of congenital or genetic disorders. In the case of the triptans, the appropriate pregnancy registry should be contacted before delivery of the baby. The obstetrician should be notified, and if the fetus was exposed during embryogenesis, high-resolution ultrasound may be performed to assess particular organ systems. The patient and her significant other should be educated as to the risks, including cognitive or neurodevelopmental delay that may not be predicted prenatally .