Abstract
Objective
The objective of this article is to review the literature relating migraine, cardiovascular disease, and stroke during pregnancy in order to better define the relationship between migraines and vascular disease.
Methods
We conducted a systematic review of the literature using Medline and Cochrane Review with the following search terms: migraine AND pregnancy and vascular disease OR myocardial infarction OR heart disease OR stroke OR cerebrovascular disease OR hypertension in pregnancy. We also reviewed the bibliographies of papers identified in this search to obtain additional relevant studies.
Results
Of the 219 papers obtained with the primary search, we found 17 that were topically relevant. Altogether, there is an increased risk both of gestational hypertension (OR range from 1.23 to 1.68) and preeclampsia (OR range 1.08 to 3.5) in migraineurs compared to nonmigraineurs. In addition, there is an association between an increased risk of ischemic stroke in pregnancy (OR range 7.9 to 30.7), particularly with active migraine. There is also an association between migraine and increased risk of acute myocardial infarction and heart disease (OR 4.9; 95% CI 1.7, 14.2), and thromboembolic events during pregnancy (deep venous thrombosis OR 2.4; 95% CI 1.3, 4.2 and pulmonary embolus OR 3.1; 95% CI 1.7, 5.6).
Conclusion
In this review, we summarized the association between migraine and risk of vascular disease during pregnancy, based on the available literature. Given the limited amount of data, more research on these associations is needed to determine which women with migraine may be at risk while pregnant.
Introduction
Migraine headache is a primary headache disorder characterized by severe attacks of painful headaches with associated autonomic nervous system dysfunction. The World Health Organization has now ranked migraine headaches in the top 20 among all diseases worldwide causing disability, with migraine accounting for 1.3% of years lost because of disability (1). Several studies have linked migraine with other comorbidities that would put migraineurs in particular at higher risk for vascular events. For instance, migraineurs may be more likely to be obese (2,3), have a higher adjusted total cholesterol and lower adjusted high-density lipoprotein (HDL) (4,5), be less likely to exercise regularly (5) and be more likely to smoke compared to nonmigraineurs (5–7). The relationship between migraine and ischemic stroke appears to be strongest in young women under the age of 45, as shown in two different meta-analyses (8,9).
Women have the highest prevalence of migraines between the ages of 20 to 39 years, i.e. childbearing years (10). Despite most migraine sufferers reporting improvement in their headaches during pregnancy, particularly in the third trimester (11–14), research has implicated a diagnosis of migraines as an independent risk factor for several complications during pregnancy including stroke, myocardial infarction (MI), and venous thromboembolism (15,16). While the relationship between migraine and cardiovascular disease during pregnancy has been clearly demonstrated, there still remains a discrepancy in the literature over the strength and significance of this relationship. Therefore, our aim was to review the literature on the relationships among migraine, cardiovascular disease, and stroke during pregnancy, and to identify major gaps in our understanding of these relationships.
Methods
Studies were identified using a search of Medline and Cochrane Library databases from their first availability through March 2014 with keywords “migraine AND pregnancy,” combined with the search terms “vascular disease,” “myocardial infarction,” “heart disease,” “stroke,” “cerebrovascular disease,” “preeclampsia,” gestational hypertension,” or “thromboembolism.” Studies were included if the results focused on cardiovascular events and stroke/transient ischemic attack (TIA) during pregnancy in women with migraine. We included all randomized controlled trials, cohort studies, and case-control studies with more than 30 participants written in English. Studies were further filtered to include only humans. Additionally, studies in the bibliographies of articles identified as relevant from the above search methods were included. Studies that did not address women during pregnancy, did not differentiate migraine from other headache types, focused on migraine auras rather than cerebrovascular ischemic events, or studies related only to treatment of migraines during pregnancy or medication side effects were excluded as they were felt to be out of the scope of this review.
Results
This search revealed 219 articles in Medline, of which 17 met the inclusion criteria (Figure 1). In the Cochrane database, only one relevant article out of two hits was found, and this was already identified in the Medline search. The relationships between migraine during pregnancy and vascular outcomes are shown in Table 1.
Search results for studies evaluating migraine, preeclampsia, gestational hypertension, and cardiovascular disease in pregnant women. Studies investigating the association between migraine during pregnancy and vascular outcomes. Gest htn: gestational hypertension; DVT: deep venous thrombosis; PE: pulmonary embolism; MI: myocardial infarction; OR: odds ratio; CI: confidence interval.
Gestational hypertension and preeclampsia
The relationship between migraine with gestational hypertension and preeclampsia is the most widely studied topic in this demographic. Gestational hypertension is defined as systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥ 90 mm Hg without proteinuria. Preeclampsia has been traditionally defined as new-onset hypertension with proteinuria. However, the newest guidelines from the American College of Obstetrics and Gynecology have removed the requirement of proteinuria if other signs of severe preeclampsia are present (25). Studies have shown that women with migraines are more likely to have gestational hypertension (4,6,16,20) and preeclampsia (3,17,20,21,23) than women without these conditions during pregnancy (Table 1). Direct comparison between women with reported migraine history and those without have shown migraineurs to have a 4.08 mm Hg higher mean third trimester systolic (95% confidence interval (CI) 3.3–4.9) and a 2.39 mm Hg higher diastolic (95% CI 1.8–3.0) blood pressures (3).
The literature remains mixed in identifying migraine alone as an independent risk factor for gestational hypertension. Some studies have found no association between migraine (without other comorbidities) and a higher risk of gestational hypertension (2,3). However, more studies reported an increased risk both of gestational hypertension and preeclampsia in migraineurs compared to nonmigraineurs. For instance, Cripe et al. found that the risk of gestational hypertension increased by 1.42-fold in women with migraine (19), and Bushnell et al. reported an odds ratio (OR) of 2.3 (CI 2.1–2.5) of preeclampsia and/or gestational hypertension in women with migraine (16). In a large prospective study, Facchinetti et al. also showed an increased risk of gestational hypertension and preeclampsia in migraineurs compared to nonmigraineurs with an adjusted OR of 2.85 (1.4–5.81) (6). In addition, the study that documented blood pressures in migraineurs vs nonmigraineurs throughout each trimester also reported a 1.5-fold increased risk of preeclampsia (OR 1.53, 95% CI 1.09–2.16), and an even higher risk in migraineurs who were overweight or obese (OR 6.10, 95% CI 3.83–9.75) (3).
Ischemic stroke
The risk of ischemic stroke in migraineurs was evaluated using the United States (US) Nationwide Inpatient Sample from the Healthcare Cost and Utilization Project of the Agency for Healthcare Research and Quality, which is the largest all-payer inpatient database in the US. An analysis of the primary and secondary discharge codes was performed to identify patients who were admitted during pregnancy with a primary diagnosis of migraine (15,16). When ischemic stroke was the outcome of interest, migraine was strongly associated with migraine (OR 16.9; CI 9.7–29.5; Table 1) (15). A second analysis was performed with migraine occurrence during pregnancy as the primary outcome, and this showed there was a strong association with ischemic stroke (OR 30.7 (CI 17.4–34.1) (15). However, cause and effect cannot be established because these relationships were based on International Classification of Diseases, Ninth Revision (ICD-9) codes within administrative discharge data, and the incidence of migraine (185 per 100,000 deliveries) was much lower than expected in childbearing women. Therefore, these codes only represented women who were admitted and discharged with a diagnosis of active and severe migraine, and did not reflect all women who had a pre-pregnancy history of migraine. A smaller cohort of pregnant women with acute ischemic stroke from the United Kingdom (UK) showed a history of migraines was a statistically significant risk factor for antenatal stroke (22). Forty-one women participating in a prospective study of migraine during pregnancy were referred to neurologists for transient neurological events, of whom 34 were diagnosed with migraine with aura and two had strokes (one related to preeclampsia). Forty-one pregnant women without migraine (controls) were matched to the group with transient neurologic deficits by age, smoking status, and primiparous vs not. After comparison of multiple clinical features and pathologic laboratory tests, there were no differences between the groups (24).
Migraine and other stroke types
The Nationwide Inpatient Sample analyses showed that there was an association between migraine and intracerebral hemorrhage (ICH) during pregnancy, with an OR of 9.1 (95% CI 3.0–27.8) (16) while the UK study showed an OR of 2.6 (95% CI 0.05–35.5) (22). However, there was no association between migraine and cerebral venous sinus thrombosis (CSVT) because of the selection of ICD-9 codes that were rarely used in this database, nor was there an association between migraine and subarachnoid hemorrhage, although the CIs were wide (OR 3.4, 95% CI 0.5–24.1) (16).
Other vascular diseases and migraines during pregnancy
Only one study reported the potential link between migraines and vascular diseases including MI, heart disease, and thromboembolism during pregnancy (Table 1) (16). The Nationwide Inpatient Sample analysis described above reported that migraines were associated with a nearly five-fold increased risk of acute MI and heart disease during pregnancy (OR of 4.9; CI: 1.7–14.2) (16). The same analysis found that migraines were associated with an increased risk of thromboembolism such as pulmonary embolism (OR 3.1, 1.7–5.6), deep vein thrombosis (OR 2.4, 1.3–4.2), and thrombophilia (OR 3.6, 2.1–6.1) during pregnancy (16).
Discussion
Summary of findings
Research has demonstrated significant relationships exist between migraine headaches and cardiovascular disease such as ischemic stroke, MI, and thromboembolism during pregnancy. Several administrative analyses have reported this association including the Nationwide Inpatient Sample in the US (16) and a large prospective cohort study in Northern Italy (6). The mechanism(s) behind this relationship remain unclear; however, there is evidence that migraines are linked with several cardiovascular risk factors such as hypertension, hyperlipidemia, obesity, and smoking. These risk factors do seem to be more prevalent in women with migraine during pregnancy, shown in the administrative analyses (15,16).
Implications
Pregnancy is commonly thought of as a prothrombotic state, with pregnant women being at increased risk of thromboembolic events including both venous and arterial systems (26,27). Estrogen levels rise during pregnancy starting after 100 days and peak by the last month of gestation. Estrogen is responsible for increased procoagulants during pregnancy through stimulation of hepatic synthesis of clotting factors (28,29). Additionally, estrogen increases the circulating cholesterol levels during pregnancy, thereby increasing the risk of hyperlipidemia and increasing risk of vascular disease (28). Near the end of the third trimester, progesterone levels also increase, which increases venous distensibility, allowing for venous stasis in the hyperdynamic state of pregnancy and thereby increasing the risk of venous thromboembolism (28). The mechanism linking venous thromboembolism with migraine is unclear. However, there is evidence that migraine is a risk factor for cerebral thromboembolism outside of pregnancy with an OR of 2.8 (p < 0.01), raising suspicion that migraine can also contribute to a hypercoagulable state independent of pregnancy (30).
One pathophysiologic mechanism that could link migraine, thromboembolism, and ischemic stroke during pregnancy is the presence of a patent foramen ovale (PFO). PFO is associated with cryptogenic stroke, especially in younger adults (31). The mechanism may be via paradoxical embolism from a distal deep venous thrombosis (DVT) through a right to left shunt (i.e. PFO) and embolism to the brain, or via clots originating from the PFO itself (32). PFO is also prevalent in those with migraines with aura (33). Therefore, given the hypercoagulable state of pregnancy increasing the risk of DVT, the relationship between migraine (especially aura) with PFO and ischemic stroke presents the possibility of PFO as the link tying together pregnancy, thromboembolism, migraines, and ischemic stroke in some cases. However, this connection is purely speculative because of the lack of studies with this level of clinical detail on diagnostic testing for stroke etiology.
The pathophysiology of migraines raising the risk of vascular events in general is poorly understood. Female sex hormones have been shown to play a role in migraine frequency, as indicated by fluctuations in migraine severity with menstruation, use of oral contraceptive pills, pregnancy, and menopause (11,14). Additionally, the pro-inflammatory role of prostaglandins in migraines has been demonstrated both in animal and human models, leading to vasodilation both of intra- and extracranial vessels (34). It is possible that the pro-inflammatory responses associated with migraine cause endothelial changes that are further compounded by the vessel wall changes associated with pregnancy, therefore leading to further risk of embolic events.
An important pregnancy complication that involves endothelial dysfunction is preeclampsia—another disorder that may link migraine to stroke, as shown in our review. For example, preeclampsia increases the risk of stroke around the time of delivery, and the clinical manifestations of severe preeclampsia involve headaches and aura/visual scotoma leading to emergent delivery. Administrative data, such as the Nationwide Inpatient Sample, do not distinguish between migraine and severe preeclampsia, since there is a high likelihood of misclassification, i.e. severe preeclampsia could be coded as migraine in addition to preeclampsia (16). Given these limitations, preeclampsia may be a confounder in the relationship between migraine and stroke during pregnancy. However, preeclampsia may also represent a pathophysiologic link between migraine and stroke during pregnancy, because of the association with rising blood pressures in migraineurs throughout pregnancy and especially in the third trimester (3), vasculopathies known to cause stroke, such as reversible vasoconstriction syndrome, and posterior reversible encephalopathy syndrome (35). Imaging to detect these vasculopathies may not be performed because the emphasis in women with preeclampsia who then develop headaches and neurologic symptoms (classified as severe preeclampsia) is on emergent delivery.
Strengths and limitations
Our review has several strengths because of the systematic nature of our review of the literature, the inclusion of several large cohort studies related to the topic, and stringent inclusion/exclusion criteria to ensure relevant peer-reviewed literature.
The most obvious limitation to the study of this topic is the relatively small number of studies and the rarity of cardiovascular events during pregnancy, thereby limiting the ability for studies to have enough cases for statistical analyses. This is likely to be further compounded by the possibility of publication bias. Additionally, despite the relatively large percentage of women of childbearing years afflicted with migraines, there seems to be a paucity of studies evaluating the influence of this disorder on pregnancy complications. Further potential limitations include most studies relying on self-reported migraine history, raising the possibility of misdiagnosis and under-reporting of true migraine history, although a few research groups have been able to confirm self-reported diagnoses of migraine in 81.6% to>87% of women (36,37), thereby lessening this potential confounder.
Potential clinical/public health implications
Given the evidence linking women with a history of migraines who may be potentially predisposed for developing vascular disease such as ischemic stroke, MI, and venous thromboembolism during pregnancy, it may be prudent to consider these women at generally high risk for maternal complications, as well. Other studies have linked migraines with risk of adverse perinatal outcomes. For example, Marozio et al. found that in an Italian cohort prospective study, women with either migraine or tension-type headaches were at a significantly higher risk of preterm delivery and delivery of small for gestation age infants (38). Therefore, there is an association found not only with the risk for the mother in terms of vascular disease, but also for complications for the infant. Because of these associations, it is felt that women with migraine history should be monitored closely throughout their pregnancy, especially watching for rising blood pressures, counseled about cardiovascular risk factors, and educated about stroke and cardiovascular disease, and preeclampsia, especially if she has active migraines throughout pregnancy. The absolute risk is likely to be low, but education and awareness may identify women at particularly high risk who may benefit from maternal-fetal medicine consultation.
Recommendations for future studies
Further research is warranted to understand the underlying mechanism of how migraines contribute to increased risk of vascular disease, particularly in pregnancy. There are limited data available describing these associations, and the proposed pathophysiology is largely speculative. A better understanding of the mechanisms involved in this association could lead to potential treatment and earlier intervention, thereby reducing the health care costs of morbidity and mortality associated with adverse vascular events in this population.
Large prospective cohort studies and/or registries would be beneficial in further describing the association between migraines and vascular events. This would help in determining the potential need for closer monitoring of these women during pregnancy as high-risk obstetric patients.
Clinical implications
Migraine history is associated with a higher risk of gestational hypertension and preeclampsia. Active migraine places women at significantly higher risk for ischemic stroke during pregnancy. Acute myocardial infarction is associated with active migraines during pregnancy. Pulmonary embolism, deep vein thrombosis, and thrombophilia are associated with migraine during pregnancy; however, cerebral venous thrombosis is not found to be an increased risk for migraineurs. Migraineurs, particularly those with known vascular risk factors, may benefit from high-risk obstetric monitoring to prevent morbidity and mortality.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
None declared.