Preeclampsia and Cerebrovascular Disease.

Abstract

HomeHypertensionVol. 74, No. 1Preeclampsia and Cerebrovascular Disease Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissionsDownload Articles + Supplements ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toSupplemental MaterialFree AccessReview ArticlePDF/EPUBPreeclampsia and Cerebrovascular DiseaseThe Maternal Brain at Risk Eliza C. Miller Eliza C. MillerEliza C. Miller Correspondence to Eliza C. Miller, Neurological Institute of New York, 710 W 168th St, Room 641, New York, NY 10032. Email E-mail Address: [email protected] From the Department of Neurology, Division of Stroke and Cerebrovascular Disease, Columbia University Irving Medical Center, New York, NY. Search for more papers by this author Originally published6 May 2019https://doi.org/10.1161/HYPERTENSIONAHA.118.11513Hypertension. 2019;74:5–13Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: May 6, 2019: Ahead of Print Preeclampsia, a hypertensive disease affecting 5% to 8% of pregnancies1, is a multisystem disorder, with vascular dysfunction being central to the disease. The heterogeneity of preeclampsia’s signs and symptoms has generated controversy regarding whether the cause of preeclampsia is mainly related to maternal or placental factors (Tables S1 and S2 in the online-only Data Supplement).2 Most researchers agree that placental hypoxia and oxidative stress play an important role in the pathophysiology of preeclampsia, leading to a cascade of downstream effects.3–5The maternal cerebral vasculature is highly vulnerable to adverse effects of preeclampsia. Short- and long-term cerebrovascular complications of preeclampsia include posterior reversible encephalopathy syndrome (PRES), reversible cerebral vasoconstriction syndrome (RCVS), hemorrhagic and ischemic stroke, cerebral small vessel disease, and vascular dementia.6–10 This review summarizes current evidence and recent advances in our understanding of the effects of preeclampsia on cerebrovascular disease in women and outlines gaps in knowledge and directions for future research.Epidemiology of Preeclampsia-Associated Cerebrovascular DiseasePreeclampsia is a leading cause of maternal morbidity and mortality worldwide.11 In black women, preeclampsia has a higher prevalence and is more likely to be associated with maternal complications, including 3-fold higher mortality rates.12,13 Cerebrovascular disease is the leading cause of maternal mortality in women with preeclampsia, with the majority of deaths caused by intracerebral hemorrhage (ICH).14–17 In the United States, maternal stroke accounted for 7.4% of maternal deaths from 2011 to 2014.18 Rates of antepartum and postpartum stroke, highly associated with preeclampsia, increased 47% and 83%, respectively, from 1994 to 1995 to 2006 to 2007 in the United States.19 This has been directly attributed to increasing rates of preeclampsia.19,20 Preeclampsia increases the risk of maternal stroke ≤6-fold.7,21 Risk factors for peripartum stroke in women with preeclampsia include older age, black race, chronic preexisting hypertension, underlying prothrombotic or inflammatory disorders, and infections.22In the longer term, preeclampsia is now recognized by the American Heart Association/American Stroke Association as a sex-specific risk factor for future stroke,23 and guidelines recommend that all women be evaluated for history of preeclampsia as part of routine cardiovascular risk assessment. Controlling for other risk factors, a history of preeclampsia is associated with a 4-fold increase in risk of developing chronic hypertension,24 a 4-fold increase in heart failure,25 a 3-fold increase in type 2 diabetes mellitus,26 and a 2-fold increase in future stroke.25 Women with early-onset preeclampsia, diagnosed before 34 weeks gestation, are at particularly heightened risk.27 The American College of Obstetrics and Gynecology and the American College of Cardiology recently released a joint statement urging collaboration between obstetrics care providers and primary care providers to identify women during pregnancy who are at elevated risk for future cardiovascular disease and tailor preventive treatments accordingly.28 Preeclampsia is still under-recognized as a sex-specific risk factor for future stroke, however, and many women are unaware of their risk.Preeclampsia and the Cerebral Vasculature: Insights From Basic ResearchA detailed discussion of the vascular biology of preeclampsia is beyond the scope of this review and has been reviewed separately.5 Inflammation, oxidative stress, and hypoxia-induced angiogenic factors including sEng (soluble endoglin) and sFlt-1 (soluble fms-like tyrosine kinase-1), a VEGF (vascular endothelial growth factor) inhibitor, all play important roles in the maternal vascular damage seen in preeclampsia.4 Preeclampsia is unique to human pregnancy, but animal models of preeclampsia have been developed, yielding insights into its cerebrovascular effects (Figure 1).Download figureDownload PowerPointFigure 1. Cerebrovascular effects of preeclampsia (PE). PE causes both acute and chronic cerebrovascular disease. In the immediate peripartum period, PE is associated with increased blood-brain barrier permeability, impaired cerebral autoregulation, hypercoagulability, and inflammation, resulting in complications such as ischemic and hemorrhagic stroke, posterior reversible encephalopathy syndrome, reversible cerebral vasoconstriction syndrome, and cerebral venous sinus thrombosis. Long-term PE is associated with cerebral small vessel disease including stroke and vascular dementia, as well as increased carotid intima-media thickness.Cerebrovascular Changes in Normal PregnancyThe cerebral circulation has several features that distinguish it from other vascular beds. Chief among these is the neurovascular unit (NVU), which may be conceptualized as a complex of endothelial cells, smooth muscle cells, pericytes, astrocytes, neurons, and extracellular matrix proteins, having multiple specialized functions.29 These include maintaining the structural integrity of the blood-brain barrier (BBB), which maintains the neuronal microenvironment through endothelial tight junctions, transcytosis, and efflux transporters.30 In addition, several cells of the NVU, in particular astrocytes, pericytes and smooth muscle cells, mediate cerebral blood flow at the microvascular level.30–33 This includes regulation of neurovascular coupling, the process by which cerebral blood flow responds at the capillary level to increased neuronal metabolic demand (also known as functional hyperemia)34, and cerebral autoregulation, the mechanism by which the cerebral vasculature regulates cerebral blood flow in response to rapid changes in systemic mean arterial blood pressure over a limited but wide range (classically, 50–150 mm Hg), thereby maintaining steady-state cerebral perfusion.35–37During normal pregnancy, cerebral arterioles undergo outward remodeling and capillary density increases.38 Despite this remodeling, cerebrovascular resistance has been shown to be unchanged under normotensive conditions in healthy pregnant rats, and autoregulation was actually improved.38,39 This finding corroborates small clinical studies of cerebral autoregulation in normotensive pregnant women showing autoregulation in the high-normal range.40–42 VEGF and PlGF (placental growth factor), a VEGF homologue, both of which increase BBB permeability, are both increased in normal pregnancy; however, physiological increases in sFlt-1 during pregnancy, inhibiting both VEGF and PlGF, may prevent the development of vasogenic cerebral edema under noninflammatory conditions.43 Nevertheless, some evidence suggests that NVU function may be compromised in healthy pregnancy. Cerebral vessels from healthy late-pregnant rats developed a 6-fold increase in BBB permeability after exposure to plasma from healthy pregnant women.44 A recent study demonstrated free hemoglobin in cerebrospinal fluid from normotensive and preeclamptic pregnant women, suggesting abnormal BBB permeability in both groups.45 Pressor-induced acute hypertension caused breakthrough of autoregulation and significant cerebral hyperperfusion in healthy pregnant but not nonpregnant rats.38,39 Efflux transporters at the BBB in the hippocampus were inhibited in late-pregnant rats, leading to an increase in spontaneous seizures.46 However, tight junction protein expression in the BBB was similar in healthy pregnant and nonpregnant rats.38Neurovascular Dysfunction in PreeclampsiaMultiple studies point to profound dysfunction of the NVU in the setting of preeclampsia. Pregnant rats with mechanically induced placental ischemia had impaired cerebral autoregulation in vivo and increased BBB permeability, demonstrating a direct causal effect of placental ischemia on NVU dysfunction47; this effect seemed to be mediated by inflammatory cytokines known to be elevated in women with preeclampsia.48 Rats fed a high-cholesterol diet to induce experimental preeclampsia had impairment of potassium channel-mediated cerebral arteriolar vasodilation and vasoconstriction, critical mechanisms of neurovascular coupling, and cerebral autoregulation.49 Pregnant rats infused with sFlt-1 and sEng developed a preeclampsia-like syndrome and increased BBB permeability in the postpartum period.50 Exposing rat cerebral vessels to serum from women with preeclampsia resulted in an 18-fold increase in BBB permeability; surprisingly, this effect was prevented by VEGF inhibition.44 Placental ischemia rats had evidence of cerebral edema and neuroinflammation 2 months postpartum, suggesting that the deleterious effects of preeclampsia on the NVU persist after delivery.51 The same rats had decreased expression of occludin—a tight junction protein—in the posterior cortex; however, other tight junction proteins in the BBB were unchanged.51 Magnesium—the standard treatment for prevention of eclamptic seizures in women with severe preeclampsia—has been shown to reduce brain edema and neuroinflammation in a rat model of eclampsia.52 Neuronal biomarkers were elevated in serum from women with preeclampsia during pregnancy and up to a year postpartum, suggesting persistent BBB dysfunction.53,54A number of clinical studies have borne out these experiments. Women with preeclampsia, particularly if they have headache, have been shown to have higher baseline cerebral perfusion pressure, lower cerebrovascular resistance, and decreased vasodilation in response to CO2 inhalation, compared with healthy pregnant women.55–57 Other studies showed impaired dynamic cerebral autoregulation in women with preeclampsia.40,58–60 Cerebral perfusion pressure was elevated in women with preeclampsia even when hypertension was treated, and correlation between blood pressure and cerebral perfusion pressure was increased, implying impaired autoregulation.61 A neuropathological study of 7 women who died from eclampsia showed cerebral perivascular microhemorrhages and microinfarcts, fibrinoid necrosis, edema, and arteriolar vasculopathy in several of the cases; the authors speculated that breakthrough of cerebral autoregulation was responsible for the vasculopathy and BBB breakdown.62 In a case series of 28 women with stroke and severe preeclampsia or eclampsia, all of the women had prestroke systolic blood pressure of ≥155 mm Hg, but only 12.5% of the women had prestroke diastolic pressures >110 mm Hg, and only 25% had prestroke mean arterial pressure over 130 mm Hg.63 Of note, 92% of these women had ICH, despite their blood pressure being not necessarily in a severe range (systolic >160 mm Hg or diastolic >110 mm Hg), implying breakthrough of autoregulation below the classic threshold of 150 mm Hg.Interestingly, angiogenic pathways have also been implicated in NVU dysfunction outside of the context of preeclampsia.64 Elevated sFlt-1 levels in serum and cerebrospinal fluid of nonpregnant patients with subarachnoid hemorrhage (SAH) predicted cerebral vasospasm,65 and the monoclonal VEGF inhibitor bevacizumab (an analog to sFlt-1) can cause a preeclampsia-like syndrome of acute hypertension and vasogenic cerebral edema.66,67 Animal studies have also demonstrated that exogenous VEGF increases BBB permeability68 and that this effect is actually inhibited by sFlt-1 during late pregnancy.43 VEGF has recently been shown in an in vitro BBB model to enhance the activation of caveolin-1—a membrane protein involved in BBB transcytosis; in contrast, magnesium decreased caveolin-1 activity and reduced transcellular BBB permeability,69 suggesting that magnesium’s effect on eclampsia prevention could be because of stabilization of the BBB.Immediate Cerebrovascular Complications of PreeclampsiaAcute cerebrovascular disorders, including PRES, RCVS, ischemic and hemorrhagic stroke, and cerebral venous sinus thrombosis (CVST), are dreaded complications of preeclampsia that can result in permanent maternal disability or death. The risk of acute cerebrovascular disease in pregnancies complicated by preeclampsia is as high as 1 in 500 deliveries22; by comparison, the overall risk of pregnancy-related acute cerebrovascular disease is ≈30 per 100 000 deliveries.70PRES and RCVSPRES is a syndrome of vasogenic edema and BBB breakdown, affecting both cortical and subcortical structures and all regions of the brain. There is a predilection for the parietal and occipital lobes, sometimes resulting in visual disturbances or cortical blindness. In severe cases, PRES may result in coma, status epilepticus, and ICH. PRES can be seen in patients with acute hypertension outside of pregnancy, particularly in patients with renal dysfunction or those taking calcineurin inhibitors for immunosuppression. Up to 98% of women with eclampsia have radiological evidence of PRES, leading some to question whether eclampsia constitutes obstetric PRES.71,72 However, preeclampsia-associated PRES may have unique features, including a higher prevalence of headache, less frequently altered mental status, and possibly a better prognosis.8RCVS was first described in postpartum women by Call et al73 but can occur in nonpregnant patients of both sexes. Presenting with a sudden, severe thunderclap headache, RCVS causes vasospasm of the arteries of the circle of Willis and can be associated with ischemic stroke and nonaneurysmal SAH, usually over the cerebral convexities. Although angiographically, RCVS looks similar to primary central nervous system vasculitis, pathologically, there is no inflammatory infiltrate, and the disorder usually has a transient and monophasic course.74 Unlike vasculitis, RCVS is not steroid responsive; in fact, steroids may worsen the course of both PRES and RCVS.75,76Although the term reversible in both PRES and RCVS implies a benign course, both disorders can lead to devastating consequences (Figure 2). Some authors consider both vasculopathies to be on a continuum with preeclampsia because the 3 disorders share multiple features and often overlap.77 PRES and RCVS occur most often in the postpartum period, sometimes with little warning and no signs of preeclampsia during the pregnancy. However, elevated sFlt-1/PlGF ratio—a marker for preeclampsia—was reported in the serum of a woman with postpartum RCVS and PRES who had no hypertension during pregnancy.78Download figureDownload PowerPointFigure 2. Intracerebral hemorrhage in a postpartum woman with preeclampsia. A 32-y-old woman presented 4 d postpartum with hypertension and severe headache, found to have severe preeclampsia. Computed tomography of the head without contrast (A) showed left sided intracerebral hemorrhage with cerebral herniation. Cerebral angiogram (B) showed diffuse segmental vasospasm of the proximal and distal large cerebral arteries. Brain biopsy showed no evidence of vasculitis. She was diagnosed with the reversible cerebral vasoconstriction syndrome.Arterial Ischemic Stroke in PreeclampsiaArterial ischemic stroke (AIS) occurs when occlusion of a cerebral artery results in infarction of the central nervous system. AIS in women with preeclampsia can occur through multiple mechanisms. Peripartum cardiomyopathy is highly associated with preeclampsia79 and can cause acute systolic heart failure and arrhythmias, both of which can lead to cardioembolic AIS, particularly in the setting of preeclampsia-related hypercoagulability. Severe vasospasm from RCVS can cause hypoperfusion distal to the point of spasm, resulting in AIS.80 Preeclampsia-related hypercoagulability can also provoke in situ thrombosis in cerebral vessels, particularly in cases of eclampsia or the HELLP (hemolysis, elevated liver enzymes, low platelets) syndrome.62,81,82Cervical artery dissections (carotid or vertebral) have been reported in association with preeclampsia, which can cause AIS either through occlusion of the cervical vessel or distal embolization of clot from the false lumen at the site of the dissection.83–85 It is not known whether preeclampsia increases the risk of cervical artery dissection, but preeclampsia has a well-established association with spontaneous coronary artery dissection,86 and preeclampsia-associated aortic dissections have also been reported.87,88Migraine, particularly migraine with aura, increases the risk of AIS in pregnancy; epidemiological studies have found an 8- to 30-fold increase in risk of AIS in pregnant migraineurs.89 Migraine has also been associated with preeclampsia in multiple studies: a systematic review found odds ratios for preeclampsia of 1.08 to 3.5 in migraineurs,89 and some authors have speculated that there may be shared pathophysiology of endothelial dysfunction, platelet activation, and altered vasoreactivity.90 However, preeclampsia may also be a confounder in the relationship between migraine and pregnancy-related AIS.Subarachnoid and Intracerebral Hemorrhage in PreeclampsiaHemorrhagic stroke, comprising ICH and SAH, can occur either spontaneously or secondary to rupture of vascular lesions such as aneurysms, arteriovenous malformations (AVMs), or moyamoya vasculopathy. Data conflict regarding whether pregnancy increases the risk of aneurysm or AVM rupture,91–93 and no studies have evaluated the effect of preeclampsia on risk of rupture. Interestingly, VEGF is increased in brain AVMs, and VEGF inhibition with sFlt-1 decreased brain AVM severity in animal studies.94 Theoretically, this could lead to progression of brain AVMs during normal pregnancy; conversely, high serum levels of sFlt-1 in preeclampsia could ameliorate this effect. A case series of 19 pregnancy-associated hemorrhagic strokes showed that none of the women with ruptured vascular lesions had preeclampsia, whereas 46% of the women with primary spontaneous ICH or nonaneurysmal SAH had preeclampsia.95Moyamoya vasculopathy is a progressive steno-occlusive disorder of the terminal internal carotid arteries, leading to the formation of fragile collaterals, which are prone to rupture, especially in the setting of hypertension. Moyamoya-related ICH in association with preeclampsia has been reported.96,97Although secondary ICH or SAH can occur in women with preeclampsia, hemorrhages without an underlying vascular lesion are seen more often, usually in the peripartum and postpartum period.95 In nonpregnant patients, spontaneous ICH is usually related to cerebral small vessel disease due to chronic hypertension or cerebral amyloid angiopathy. Spontaneous ICH can also be seen with coagulopathies. Similarly, the vascular pathophysiology of ICH in preeclampsia may also be due to arteriolar dysfunction, with compromised autoregulation unable to compensate for acute hypertension49; this could be aggravated by preeclampsia-related coagulopathy. Nonaneurysmal SAH, usually over the cerebral convexities, can be a feature of preeclampsia-related RCVS and portends a better prognosis compared with aneurysmal SAH.6 Unfortunately, few studies have provided radiological or pathological details of primary hemorrhagic strokes seen in preeclampsia.62,98CVST in PreeclampsiaNormal pregnancy causes changes in the coagulation system that contribute to an increased risk of venous thrombotic events, including CVST, particularly in the puerperium.21,99,100 These changes include higher production of procoagulant coagulation factors, including factors V, VII, VIII, IX, X, and von Willebrand factor; decrease in protein S levels and acquired activated protein C resistance; and placentally produced plasminogen activator inhibitors, which decrease endogenous tissue plasminogen activator activity.101 Preeclampsia causes hypercoagulability, systemic inflammation, platelet activation, and endothelial injury, all of which predispose to thrombosis and increase the risk of CVST.102,103 Occurring most often postpartum, CVST presents insidiously with headache and is often mistaken for postdural puncture headache or migraine. Clots can propagate rapidly, leading to venous congestion in the adjacent brain tissue with catastrophic consequences including venous infarction, hemorrhage, and increased intracranial pressure. Cesarean section and infections, both of which are more common in women with preeclampsia, increase the risk of postpartum CVST.104Long-Term Effects of PreeclampsiaThe effects of preeclampsia on the maternal brain appear to continue for years past the initial injury. Women with a history of remote preeclampsia had 3× higher odds of having abnormally high carotid intima-media thickness105—a marker of subclinical atherosclerosis highly associated with future stroke. Women with a history of preeclampsia have increased white matter hyperintensities on brain magnetic resonance imaging9,106, a marker of cerebral small vessel disease that is highly associated with stroke and dementia (Figure 3). Prior preeclampsia is an independent risk factor for future stroke in women, particularly in middle age.27,107–109 Multiple studies have suggested an association between preeclampsia and cognitive decline110–112; this association was recently confirmed in a prospective population-based study showing an increased risk of vascular dementia, but not Alzheimer disease, in women with a history of preeclampsia.10 No randomized trials have been conducted for primary prevention of cerebrovascular disease in women with a history of preeclampsia, but a recent prospective cohort study showed a possible benefit of aspirin for stroke risk reduction in this population.113Download figureDownload PowerPointFigure 3. Long-term cerebral small vessel disease after preeclampsia. A 47-y-old woman with diabetes mellitus, hypertension, active tobacco use, and history of preterm preeclampsia presented with left sided weakness and cognitive complaints. Magnetic resonance imaging of the brain with and without contrast showed an acute lacunar infarct in the right internal capsule on diffusion-weighted images (A), patchy subcortical white matter hyperintensities on fluid-attenuated inversion recovery sequences (B and C), and no contrast enhancement (D). Additional studies including lumbar puncture showed no evidence of demyelinating disease or other inflammatory pathology. She was diagnosed with ischemic stroke and white matter changes due to cerebral small vessel disease.Gaps in Knowledge and Future DirectionsMany questions remain regarding the effects of preeclampsia on the maternal brain, and there is an urgent need for more basic, translational and clinical research in this area.Characterizing Molecular Mechanisms of NVU Dysfunction in PreeclampsiaPreeclampsia clearly causes NVU dysfunction, but the cellular mechanisms are not well understood. Although measures of cerebral hemodynamics have been shown to be disrupted in preeclampsia, it remains uncertain whether BBB dysfunction in preeclampsia is the cause, or the effect, of disrupted cerebral autoregulation. In other words, does impaired autoregulation lead to hyperperfusion at blood pressures below the threshold of normal autoregulatory limits, thus overcoming the BBB due to increased cerebral perfusion pressure? Or does increased BBB permeability permit leakage of toxic factors into the brain parenchymal microenvironment, causing neuroinflammation, impairment of neurovascular signaling, and subsequent loss of autoregulation? What is the role of VEGF and VEGF inhibitors in the development of preeclampsia-related BBB dysfunction, and how does neuroinflammation interact with these factors? A complex and as-yet poorly characterized interaction between inflammatory cytokines and BBB and autoregulatory impairment may underlie preeclampsia-related NVU dysfunction, leading to both immediate and long-term cerebrovascular disease.Predicting Cerebrovascular Complications in the Immediate Postpartum PeriodMost maternal strokes occur in the first 2 weeks postpartum, often after women have been discharged home after delivery. While risk factors associated with postpartum stroke have been identified, there are currently no biomarkers or screening tools to predict which women with preeclampsia may be at higher risk for developing this rare but disastrous complication. Future research is needed to develop screening tools to identify which women are at the highest risk of postpartum stroke. The role of genetics and genomics in preeclampsia-associated cerebrovascular complications remains unexplored.Identifying and Preventing Long-Term Sequelae of PreeclampsiaIncreasing evidence supports that preeclampsia is independently associated with long-term cerebrovascular disease in women.10,25,113 More uncertain is whether preeclampsia is an early marker for higher baseline risk in susceptible women or whether preeclampsia itself causes lasting vascular damage in the brain. Regardless of whether preeclampsia is a marker or a precipitant of cerebrovascular disease, there is a vital need to develop early preventive strategies. Not least of these is education of the public and the medical community regarding the need to engage women with their health after a pregnancy complicated by preeclampsia. Preeclampsia is not currently incorporated into cardiovascular risk calculators, and most women are unaware of their increased risk. Stroke is now the third leading cause of death in US women,114 and cerebrovascular disease is increasingly recognized as a major contributor to dementia. Clinical trials are needed to test strategies for the primary and secondary prevention of cerebrovascular disease in women after preeclampsia. Establishing robust cross-disciplinary collaborations between researchers in maternal medicine, cardiovascular disease, and vascular neuroscience will be critical to future efforts to reduce the burden of cerebrovascular disease in women after preeclampsia.Sources of FundingE.C. Miller receives funding from a National Institutes of Health National Center for Advancing Translation Sciences KL2 Mentored Career Development Award (5KL2TR001874-03) and from the Gerstner Family Foundation Louis V. Gerstner Scholars program.DisclosuresE.C. Miller receives personal compensation from medicolegal consulting related to maternal stroke.FootnotesThe online-only Data Supplement is available with this article at https://www.ahajournals.org/doi/suppl/10.1161/HYPERTENSIONAHA.118.11513.Correspondence to Eliza C. Miller, Neurological Institute of New York, 710 W 168th St, Room 641, New York, NY 10032. Email [email protected]edu