Stroke in Pregnancy

Christina Mijalski Sells, MD, MPH; Steven K. Feske, MD


Semin Neurol. 2017;37(6):669-678. 

In This Article

Epidemiology of Stroke in Pregnancy

Several studies describe the epidemiology of stroke during pregnancy and in the postpartum period. Comparison of these studies is complicated by the overall rare occurrence of stroke events, differing age groups used for comparison, and varying definitions of the postpartum period. Referral and selection bias limit conclusions that can be drawn from data derived from hospital-based studies.

Stroke Risk in Women during the Reproductive Years

Stroke among women of childbearing age is an uncommon occurrence, and there are limited data regarding the overall incidence of stroke in this population. Petitti et al[1] retrospectively reviewed hospital admission data from a large health maintenance organization to identify first-time nonpregnancy-related strokes between 1991 and 1994 and observed an annual incidence of 10.7 strokes per 100,000 women. This figure is similar to the calculated incidence of 11.4 strokes per 100,000 from data reported by Kittner et al[2] among women of the same age group in Baltimore and the Washington, DC area. A recent study by Ban et al[3] reviewed primary care electronic medical records in the United Kingdom and found a much higher rate of 25 per 100,000 in this non-pregnant population. However, in this study, the upper age limit extended to 49 years and may have enlisted many women with causes of stroke more commonly seen in older patients, thus accounting for the high rate.

Stroke Risk during Pregnancy

Multiple studies have found an increased risk of ischemic stroke and intracerebral hemorrhage (ICH) during the puerperium. During pregnancy, there appears to be a much smaller increased risk of hemorrhagic stroke and no clear increased risk of ischemic stroke. Kittner et al identified women of childbearing age in the Baltimore and Washington, DC area with any stroke over the 3-year period between1988 and 1991 and did not find an increase in risk of total stroke (i.e., combining ischemic and hemorrhagic events) during the period of pregnancy up to delivery.[2,4] Sharshar et al, representing the French Stroke in Pregnancy Study Group, conducted a population-based study of women from 63 hospitals in the Île-de-France region from 1989 to 1992. This study also did not support an increased annual risk of stroke during pregnancy through 2 weeks postpartum.[5] In the most recent population-based study by Ban et al,[3] during the years 1997 to 2014, women in the United Kingdom between 15 and 49 years were found to have an annual incidence of antepartum stroke of 10.7 per 100,000. This rate is similar to the incidence in the general population of the same age identified in the studies by Kittner and Sharshar.[2,3,5]

Stroke Risk during the Puerperium

While data do not suggest an overall increased risk of stroke during pregnancy, there is good evidence for an increased risk during the postpartum period. Kittner et al extended the period of observation to the early postpartum period and found a significant increase in the relative risk (RR) of both ischemic (RR = 8.7) and hemorrhagic (RR = 28.3) stroke for 6 weeks after delivery. Calculations from data presented in this study suggest that 42% of strokes among women between the ages of 15 and 44 years are associated with pregnancy. However, it is most important to emphasize that the increased stroke risk is largely experienced in the early postpartum period.[2] The French study did not observe the same trend during the postpartum period, which is probably accounted for by differences of definitions, especially the short 2-week postpartum period. The more recent study by Ban et al defined the peripartum period as 2 days before until 1 day after delivery, and the early peripartum as the first 6 weeks after delivery. The authors demonstrated a significant increase in the incidence of total stroke during the peripartum and early postpartum periods. They reported a nine-fold increased risk of ischemic or hemorrhagic stroke during the peripartum period and a three-fold increased risk during the early postpartum period, in keeping with the overall trend reported by Kittner et al.[2]

Ischemic Stroke during Pregnancy

So far, we have considered total stroke, combining ischemic and hemorrhagic events. The picture is clarified by separating these different stroke types. The risk of ischemic stroke does not appear to be increased during pregnancy compared with non-pregnant women of childbearing age.[2,5] Interestingly, the United Kingdom study, reported by Ban et al, found a decreased risk of any stroke during the months of pregnancy, including almost a 3-fold decrease in ischemic stroke incidence compared with non-pregnant women in the same cohort.[3] Liang et al found a slightly higher incidence of ischemic stroke in a Taiwanese population, 13.5 per 100,000, compared with those reported in non-pregnant women of childbearing age, with the majority of events occurring before delivery. This study did not detect the same trend of timing that was seen in the studies from the United States, France, and the United Kingdom. These differences may reflect differences in race, and it is important to note that there is likely referral bias in this tertiary center hospital-based study.[6] A second Taiwanese study by Jeng et al reported an incidence of 8 per 100,000 deliveries when corrected for referral patients, and, like the other studies, did find an increased risk during the puerperium compared with pregnancy.[7] Overall, these population-based studies do not support an increased risk of ischemic stroke in pregnancy beyond the expected incidence among all women of childbearing age.

Ischemic Stroke during the Peripartum and Puerperium

The greatest burden of ischemic stroke in women of childbearing age is borne in the early postpartum period, with a significant increase in RR of 8.7 as reported by Kittner et al.[2] While the French study only followed up patients for 2 weeks postpartum compared with 6, the majority of events were observed during the 2 weeks immediately following delivery.[5] Ban et al reported the highest incidence of ischemic stroke in the peripartum (that is, 2 days before until 1 day after delivery; 60 per 100,000 patient-years) and early postpartum periods (31 per 100,000 patient-years). This study included a longer period of observation, with the late postpartum period defined as weeks 6 to 12 after delivery.[3] Taken together, these population-based studies do not support an increase in risk of ischemic stroke during the antepartum and late postpartum periods, but agree on the increased risk in the early postpartum period, up to 6 weeks after delivery. The Ban study also suggests the added peripartum risk. The Taiwanese study as reported by Jeng et al also supported the conclusion of increased risk of any type of stroke in the puerperium.[7]

Hemorrhagic Stroke during Pregnancy

Many studies have shown an increased incidence of hemorrhagic stroke during both the months of pregnancy and the puerperium. Kittner et al[2] reported a RR among pregnant women of 2.5 compared with their non-pregnant counterparts between 15 and 44 years of age. Sharshar et al[5] found the greatest incidence of hemorrhagic stroke during the third trimester, followed by the postpartum period. Ban et al defined hemorrhage more specifically, distinguishing between subarachnoid hemorrhage (SAH) and ICH. These authors did not find an increased incidence of ICH during pregnancy, but did report an increased incidence of SAH in the peripartum period.[3] Jeng et al also reported separate incidence rates for both ICH and SAH and found that all SAH cases occurred during the third trimester and peripartum periods. ICH was not observed to have this same trend, with cases fairly evenly distributed among the first trimester, second trimester, and peripartum time period.[7]

Hemorrhagic Stroke during the Puerperium

The early postpartum period through 6 weeks after delivery is the most likely time for a woman to experience a pregnancy-related ICH. Kittner et al reported an RR of 28.3 compared with non-pregnant women in that same cohort.[2] In a United Kingdom population, Ban et al observed an increase in the incidence of ICH in the postpartum period, with an incidence of 10.3 per 100,000 years.[3] In addition, Ban et al also reported an increased incidence of SAH in the peripartum period that was not observed among postpartum women.[3] Jeng et al found a higher incidence of SAH in the postpartum period and a similar incidence of ICH compared with the third trimester.[7] As noted above, low numbers of events and referral bias may influence the trends reported in this study.

Race and Incidence of Stroke

There are well-described racial differences in stroke risk, which may reflect racially based predilections to specific pathologies predisposing to ischemic and hemorrhagic stroke or socially based disparities in care. Vladutiu et al reported results from the Reasons for Geographic and Racial Differences in Stroke (REGARDS) Study, with data collected between 2003 and 2007. During this period, 447 strokes occurred. The majority of strokes were classified as ischemic (85%), followed by hemorrhagic (7.6%), with a small percentage (7.4%) reported as undefined by stroke subtype.[8] These data showed a significant interaction between race and parity, such that black women were more likely to have multiparty of five or greater compared with white women. Both white and black women showed a trend toward a higher stroke risk with multiparity (≥5 live births versus 1); however, this finding was not statistically significant when controlled for baseline characteristics and confounders. Although the findings in this study lacked statistical significance, other studies have found positive associations of multiparity and stroke risk; therefore, multiparity may be an important risk factor for stroke. Cheng et al[9] did find a statistically significant increase in risk of ischemic or hemorrhagic stroke in women having their third or fourth pregnancy; however, this study was limited to a Taiwanese population and may lack generalizability.

The population-based studies by Kittner and Sharshar included predominantly Caucasian and African-American women and can be reasonably applied to these populations. Cheng et al aimed to determine the risk of stroke in pregnancy in a predominantly Asian (Taiwanese) population of women between the ages of 18 and 45 years. They found that most strokes occurred during pregnancy and within 6 weeks postpartum, with events evenly distributed between these two periods of observation. This study also suggests that there may be an increased risk of stroke during the first year postpartum, with a possible protective effect for the risk of stroke after 1 year. This study demonstrates the important finding that Asian women may share some of the same risk factors for stroke and are also more likely to suffer a stroke during pregnancy and the puerperium.[9]

Yoshida et al studied a Japanese population treated at national certified teaching hospitals between 2012 and 2013. The incidence of pregnancy-associated strokes reported in this population was similar to other populations, 10.2 per 100,000 deliveries. Yet, the stroke type was overwhelmingly hemorrhagic (73.5%), followed by ischemic (24.5%), with two percent having both. A large proportion (36.9%) of hemorrhagic strokes were associated with aneurysms or arteriovenous malformations (AVMs). Almost 20% of these patients had associated hypertension or Hemolysis, Elevated Liver enzymes, and Low Platelet count (HELLP) syndrome.[10] This study provides additional evidence to support the observation that Asians bear an increased risk of hemorrhagic stroke compared with Caucasians and African Americans.

Table 1 lists multiple epidemiologic studies of stroke in pregnancy. To allow an estimation of incidence, our discussion has focused on the population-based studies by Ban et al, Kittner et al, and Sharshar et al. These three studies demonstrate equal distribution of ischemic and hemorrhagic strokes during the variable periods of observation.

Temporal Trends in the Risk of Stroke, 1979–2014

Kuklina et al reviewed a large national inpatient database in the United States to compare rates of pregnancy-related stroke, including transient ischemic attack (TIA) and cerebral venous thrombosis (CVT). Patients in this study were observed during hospitalizations antenatally or postpartum, from 1994 to 1995 and from 2006 to 2007. The rate of strokes rose across this time interval, for antenatal strokes from 15 to 22 per 100,000 deliveries and for postpartum strokes from 12 to 22 per 100,000 deliveries.[11] Patients included in this study were also found to have a significant increase in the prevalence of hypertension and heart disease across this time interval, likely accounting, at least in part, for the increased event rates over time. The general population, including pregnant patients, appears to have experienced an increased prevalence of vascular risk factors that may confer an increased risk of stroke during all stages of pregnancy. Alternatively, increased surveillance, improved imaging techniques, greater availability of diagnostic studies, and improved reporting systems may contribute to some of the apparent increased incidence rates observed across the years.