Migraines During Pregnancy Linked to Stroke and Vascular Diseases: US Population-Based Case-Control Study

The prevalence of migraine in women of childbearing age ranges from 11% to 26%. Two previous population-based reports found that migraine was associated with a 4-fold increased risk of acute myocardial infarction in pregnant women and a 17-fold increased risk of pregnancy-related stroke. Few studies have investigated the prevalence of migraine in pregnant women, and its incidence in this population is largely unknown. This population based case-control study determined the prevalence of migraine during pregnancy and examined the association between migraine and pregnancy-related cardiovascular diseases. Data on 18,345,538 pregnancy-related discharges from hospitals in the United States between 2000 and 2003 were obtained from the Nationwide inpatient sample, from the Healthcare Cost and Utilization Project of the Agency for Healthcare Research and Quality. ICD-9 codes were used to identify migraine, stroke, and other cardiovascular diagnoses. Codes for preeclampsia were separated from the other cardiovascular codes. Of the pregnancy-related discharges, 33,956 had a diagnosis of migraine, a rate of 185 per 100,000 deliveries. Multivariable logistic regression analysis showed that diagnoses independently associated with migraine discharge codes were preeclampsia (adjusted odds ratio [aOR], 2.29; 95% confidence interval [CI], 2.13-2.46), stroke (aOR, 15.05; 95% CI, 8.26-27.4), venous thromboembolism/pulmonary embolus (aOR, 3.23; 95% CI, 2.06-7.07), acute myocardial infarction/heart disease (aOR, 2.11; 95% CI, 1.76-2.54), smoking (aOR, 2.85; 95% CI, 2.53-3.21), and diabetes (aOR, 1.96; 95% CI, 1.64-2.35). The relation between migraine and stroke had the strongest association. There was no association found between a diagnosis of migraine and several nonvascular diagnoses (pneumonia, transfusions, postpartum infection) or intrauterine fetal death. These findings show a strong association between codes for a diagnosis of migraine and codes for cardiovascular events among a subpopulation of pregnant women with active migraine during admission to hospital. These data do not show whether migraine occurred before or after a vascular event and suggest no cause or effect mechanism.