Preeclampsia and Hypertensive Disorders

Gestational Hypertension

• Systolic BP ≥140 mm Hg but less than 160 mm Hg or

• Diastolic BP ≥90 mm Hg but less than 110 mm Hg and

• These pressures must be observed on at least two occasions 4 hours apart but no more than 7 days apart.

Severe Hypertension

Severe hypertension refers to sustained elevations in systolic BP to at least 160 mm Hg and/or in diastolic BP to at least 110 mm Hg for at least 4 hours or once if the patient received intravenous (IV) antihypertensive medications prior to the 4-hour period. In the most recent ACOG practice Bulletin, women who have severe GH are considered to have PE with severe features.

Proteinuria

Proteinuria may also be present before pregnancy, or it may be newly diagnosed during pregnancy. The definition of proteinuria is the same no matter when it occurs:

• Greater than 0.3 g in a 24-hour urine collection or protein/creatinine (P/C) ratio greater than 0.3. If it is not possible to measure 24-hour protein or P/C ratio, proteinuria can be defined as a dipstick measurement of at least 1+ on two occasions.

• Protein excretion in the urine increases in normal pregnancy from approximately 5 mg/dL in the first and second trimesters to 15 mg/dL in the third trimester. These low levels are not detected by dipstick. The concentration of urinary protein is influenced by contamination with vaginal secretions, blood, bacteria, or amniotic fluid. It also varies with urine specific gravity and pH, exercise, and posture. In addition, it is influenced by accuracy of the collection method, and total volume of urine collected.

• Proteinuria usually appears after hypertension in the course of the disease process, but in some women, it may appear before hypertension.

Edema

Edema is defined as excessive weight gain (>4 lb [1.8 kg] in 1 week) in the second or third trimester, and it may be the first sign of the potential development of PE. However, 39% of patients with eclampsia do not have edema.

Preeclampsia and Eclampsia

PE is GH plus proteinuria or presence of symptoms consistent with PE. Box 38.1 lists criteria for diagnosis of GH and PE. The ACOG Task Force on Hypertension in Pregnancy Group classifies PE based on whether it is has severe features. The term mild PE has been removed from the ACOG classification system and should not be used in clinical practice.

It is recognized that some women with GH may have undiagnosed chronic hypertension, whereas others will subsequently progress to develop the clinical syndrome of PE. In general, the likelihood of progression to PE depends on gestational age at time of diagnosis, with higher rates if the onset of hypertension is before 35 weeks’ gestation ( Fig. 38.1 ).

Chronic Hypertension

Chronic hypertension is defined as hypertension present prior to pregnancy or that is diagnosed before 20 weeks of gestation. Hypertension that persists for more than 3 months postpartum is also classified as chronic hypertension.

Gestational Hypertension

GH is the most frequent cause of hypertension during pregnancy. The incidence ranges between 6% and 29% in nulliparous women and between 2% and 4% in multiparous women. The incidence is markedly increased in patients with multiple gestations. In general, most cases of GH develop at or beyond 37 weeks’ gestation, and thus the overall pregnancy outcome is usually similar to that seen in women with normotensive pregnancies ( Table 38.3 ). However, women with mild GH have higher rates of induction of labor and progression to PE.

Atypical Preeclampsia

The criteria for atypical PE include gestational proteinuria or FGR plus one or more of the following symptoms of PE: hemolysis, thrombocytopenia, elevated liver enzymes, early signs and symptoms of PE-eclampsia earlier than 20 weeks, and late postpartum PE-eclampsia (>48 hours postpartum).

Capillary Leak Syndrome: Facial Edema, Ascites and Pulmonary Edema, and Gestational Proteinuria

Hypertension is considered to be the hallmark for the diagnosis of PE. However, in some patients with PE, the disease may manifest as either a capillary leak (proteinuria, facial and vulvar edema, ascites, pulmonary edema); excessive weight gain, particularly during the second and early third trimester; or a spectrum of abnormal hemostasis with multiple-organ dysfunction. These women usually present with clinical manifestations of atypical PE, such as proteinuria with or without facial edema, vulvar edema ( Fig. 38.4 ), excessive weight gain (>4 lb/wk), ascites, or pulmonary edema in association with abnormalities in laboratory values or presence of symptoms but without hypertension. Therefore we recommend that women with capillary leak syndrome with or without hypertension be evaluated for platelet, liver enzyme, and renal abnormalities. Those with symptoms such as new onset of unrelenting severe headache, severe visual disturbances, or abnormal blood tests should be considered to have possible PE.

Risk Factors for Preeclampsia

Several factors have been identified with increased risk for PE ( Box 38.2 ). In general, PE is considered a disease of primigravid women. The risk increases in those who have limited sperm exposure with the same partner before conception. The protective effects of long-term sperm exposure with the same partner might provide an explanation for the high risk for PE in women younger than 20 years old. A previous abortion (spontaneous or induced) or a previous normal pregnancy with the same partner is associated with a lower risk for PE. However, this protective effect is lost with a change of partner or with prolonged interval between pregnancies. In addition, both Scandinavian and US studies have confirmed the importance of paternal factors as a contributor to PE as well as an interpregnancy interval greater than 7 years.

Pregnancies conceived by assisted reproductive technology have been associated with an increased risk for PE. Among the factors cited are a greater proportion of women older than 40 years, infertile women during their first gestation, multifetal gestation, obese women with polycystic ovary syndrome, and women who become pregnant with donated gametes or embryos. The use of donated gametes can influence the maternal-fetal immune interaction. In addition, infertile women with recurrent miscarriage are also reported to be at increased risk for PE.

Obesity reflected as increased body mass index (BMI) heightens the risk for PE. The worldwide increase in obesity is thus likely to lead to a rise in the frequency of PE. Obesity has a strong link to insulin resistance, which is also a risk factor for PE. The exact mechanism by which obesity or insulin resistance is associated with PE is not well understood.

Earlier studies found an overall higher rate of thrombophilia in women with PE compared with controls. Recently, a number of reports have failed to reproduce these findings. The disparity in results may reflect the heterogeneity of the women being studied. In the largest series of preeclamptic women with thrombophilia, women had increasing risk for very early onset, severe disease (delivery before 28 weeks) compared with those who did not have thrombophilia.

Pathophysiology

The etiology of PE remains unknown. Many theories have been suggested, but most of them have not withstood the test of time. Some of the theories still under consideration are listed in Box 38.3 .

During normal pregnancy, impressive physiologic changes occur in the uteroplacental vasculature and in the cardiovascular system. These changes are most likely induced by the interaction of the fetal (parental) allograft with maternal tissue. The development of mutual immunologic tolerance in the first trimester is thought to lead to important morphologic and biochemical changes in the systemic and uteroplacental maternal circulation.

Uterine Vascular Changes

The human placenta receives its blood supply from numerous uteroplacental arteries that are developed by the action of migratory interstitial and endovascular trophoblasts into the walls of the spiral arterioles . This transforms the uteroplacental arterial bed into a low-resistance, low-pressure, high-flow system. The conversion of the spiral arterioles of the nonpregnant uterus into the uteroplacental arteries has been termed physiologic changes . In a normal pregnancy, these trophoblast-induced vascular changes extend all the way from the intervillous space to the origin of the spiral arterioles that represent the radial arteries in the inner one third of the myometrium. It is suggested that these vascular changes are effected in two stages, “the conversion of the decidual segments of the spiral arterioles by a wave of endovascular trophoblast migration in the first trimester and the myometrial segments by a subsequent wave in the second trimester.” This process is reportedly associated with extensive fibrinoid formation and degeneration of the muscular layer in the arterial wall. These vascular changes result in the conversion of about 100 to 150 spiral arterioles into distended, tortuous, and funnel-shaped vessels that communicate through multiple openings into the intervillous space.

In contrast, pregnancies complicated by PE or by FGR demonstrate inadequate maternal vascular response to placentation. In these pregnancies, the previously mentioned vascular changes are usually found only in the decidual segments of the uteroplacental arteries. Hence the myometrial segments of the spiral arterioles continue to exhibit their characteristic musculoelastic architecture, thereby leaving them responsive to hormonal influences. In additionally, the number of well-developed arterioles is smaller than that found in normotensive pregnancies.

It has been postulated that this defective vascular response to placentation is due to inhibition of the second wave of endovascular trophoblast migration that normally occurs from approximately 18 weeks’ gestation onward. These pathologic changes may have the effect of curtailing the increased blood supply required by the fetoplacental unit in the later stages of pregnancy and may correlate with the decreased uteroplacental blood flow seen in most cases of PE . Frusca and associates studied placental bed biopsy specimens obtained during cesarean delivery from normal pregnancies ( n = 14), preeclamptic pregnancies ( n = 24), and chronic hypertensive pregnancies only ( n = 5). Biopsy specimens from the preeclamptic group demonstrated abnormal vascular changes in every case, and 18 had acute atherosclerotic changes. In contrast, 13 of the 14 specimens from normotensive pregnancies had normal vascular physiologic changes. In addition, they found that the mean birthweight was significantly lower in the group with atherosclerosis than it was in the other group without such findings. It is important to note that these vascular changes may also be demonstrated in a significant proportion of normotensive pregnancies complicated by FGR. Meekins and associates demonstrated that endovascular trophoblast invasion is not an all-or-none phenomenon in normal and preeclamptic pregnancies. These authors observed that morphologic features found in one spiral artery may not be representative of all vessels in a placental bed.

Vascular Endothelial Activation and Inflammation

The mechanism by which placental ischemia leads to the clinical syndrome of PE is thought to be related to the production of placental factors that enter the maternal circulation and result in endothelial cell dysfunction. Soluble fms-like tyrosine kinase 1 (sFlt-1) is a protein produced by the placenta. It acts by binding to the receptor-binding domains of vascular endothelial growth factor (VEGF), and it also binds to placental-like growth factor (PLGF). Increased levels of this protein in the maternal circulation result in reduced levels of free VEGF and free PLGF, with resultant endothelial cell dysfunction.

Maternal serum and placental levels of sFlt-1 are increased in pregnancies complicated by PE values above seen during normal pregnancies. Maynard and coworkers demonstrated that soluble placenta-derived VEGF receptor (sFlt-1)—an antagonist of VEGF and PLGF—is unregulated in PE, which leads to increased systemic levels of sFlt-1 that fall after delivery. Increased circulating sFlt-1 in PE is associated with decreased circulating levels of free VEGF and PLGF and results in endothelial dysfunction. The magnitude of increase in sFlt levels correlates with disease severity, which lends further support to VEGF–soluble Flt balance and represents one of the final common pathophysiologic pathways.

First trimester PLGF levels are decreased in future preeclamptic pregnancies and in pregnancies complicated by FGR, whereas sFlt levels do not differ from controls. Again, these data are compatible with decidual angiogenic growth factors, in particular PLGF, as being essential for early placental development (PLGF is low in both FGR and PE), with a later involvement of sFlt as a fetal rescue signal steering the maternal response (i.e., the degree of maternal systemic hypertension). This hypothesis is supported by Levine and colleagues, who demonstrated that during the last 2 months of pregnancy in normotensive controls, the level of sFlt-1 increased and the level of PLGF decreased.

Levine and associates investigated urinary PLGF levels in pregnant women with and without PE and found that among normotensive pregnant women, urinary PLGF increased during the first two trimesters, peaked at 29 to 32 weeks, and decreased thereafter. Among women who ultimately developed PE, the pattern of urinary PLGF was similar, but levels were significantly reduced beginning at 25 to 28 weeks. Particularly large differences were seen among those who subsequently developed early-onset PE and in those who delivered SGA infants. A similar study suggested that urinary angiogenic factors can identify women with severe PE.

During the past decade, our understanding of the molecular basis for the pathophysiologic abnormalities in PE has reached an unprece­dented level. Clear appreciation currently exists for the role of cell adhesion molecules and angiogenic proteins and for activation of the inflammatory system in the pathogenesis of microvascular dysfunction in women with PE. Evidence also suggests an exaggerated inflammatory response (abnormal cytokine production and neutrophil activation) in women with the clinical findings of PE. However, this enhanced inflammatory response is absent before the development of PE.

Recent studies have confirmed increased levels of angiotensin II autoantibodies, increased exosomes production, and activation of the complement system in women with PE, suggesting an important role in the pathophysiology of this disorder. Endothelial dysfunction and inappropriate endothelial cell activation associated with alterations in nitric oxide levels in PE explains most typical clinical manifestations, including the increased endothelial cell permeability and increased platelet aggregation.

Genetics and Genetic Imprinting

According to the genetic conflict theory, fetal genes are selected to increase the transfer of nutrients to the fetus, whereas maternal genes are selected to limit transfer in excess of some optimal level. The phenomenon of genomic imprinting means that a similar conflict exists within fetal cells between genes that are maternally derived and those that are paternally derived. The conflict hypothesis suggests that placental factors (fetal genes) act to increase maternal BP, whereas maternal factors act to reduce BP. Endothelial cell dysfunction may have evolved as a fetal rescue strategy to increase nonplacental resistance when the uteroplacental blood supply is inadequate.

Nilsson and associates published a model that suggests a heritability estimate of 31% for PE and 20% for GH. It is unlikely that one major PE gene will be found because such a gene would be selected against through evolution, unless it also carried a major reproductive advantage. It is more likely that a rapidly growing number of susceptibility genes will be uncovered and that many of these will be found to interact with the maternal cardiovascular-hemostatic system or in the regulation of maternal inflammatory responses. These loci segregate with different populations, and it should be noted that these loci only explain a relatively small percentage of the overall cases of PE. In addition, although these linkage studies indicate maternal susceptibility, they do not exclude the additional involvement of fetal genes. Another important consideration regarding the genetics of PE is the confounding effect of the so-called fetal origins of adult disease hypothesis, which suggests that a hostile intrauterine environment for a female fetus would form the basis for the insulin resistance syndrome with its associated endothelial dysfunction and, as such, that it would lead to an increased risk for PE.

Epigenetic features and imprinting are also involved in the pathogenesis of PE. Further evidence of the role of imprinting was recently suggested by Oudejans and van Dijk and Nafee and associates.

Changes in Prostanoids

Several investigators have described levels of the various prostaglandins and their metabolites throughout pregnancy. They have measured the concentrations of these substances in plasma, serum, amniotic fluid, placental tissues, urine, and cord blood. The data have been inconsistent, which reflects differences in methodology. During pregnancy, prostanoid production increases in both maternal and fetoplacental tissues. Prostacyclin is produced by the vascular endothelium and in the renal cortex. It is a potent vasodilator and inhibitor of platelet aggregation. Thromboxane A ₂ (TXA ₂ ) is produced by the platelets and trophoblasts; it is a potent vasoconstrictor and platelet aggregator. Hence, these eicosanoids have opposite effects and play a major role in regulating vascular tone and vascular blood flow. An imbalance in prostanoid production or catabolism has been suggested as being responsible for the pathophysiologic changes in PE. However, the precise role by which prostaglandins are involved in the etiology of PE remains unclear.

Lipid Peroxide, Free Radicals, and Antioxidants

Evidence is accumulating that lipid peroxides and free radicals may be important in the pathogenesis of PE. Superoxide ions may be cytotoxic to the cell by changing the characteristics of the cellular membrane and producing membrane lipid peroxidation. Elevated plasma concentrations of free radical oxidation products precede the development of PE. In addition, some studies reported lower serum antioxidant activity in patients with PE than in those with normotensive pregnancies.

Much of the controversy about oxidative stress is related to the nonspecificity of the markers. A recent study by Moretti and associates measured oxidative stress “online” in exhaled breath (not subjective to in vitro artifacts) and confirmed greater oxidative stress in women with PE compared with nonpregnant controls and those who had uncomplicated pregnancies.

Diagnosis of Preeclampsia

PE is a clinical syndrome that embraces a wide spectrum of signs and symptoms that have been clinically observed to develop alone or in combination. Elevated BP is the traditional hallmark for diagnosis of the disease. The diagnosis of PE and the severity of the disease process are generally based on maternal BP. Many factors may influence the measurement of BP, including the accuracy of the equipment used, the size of the sphygmomanometer cuff, duration of the rest period before recording, posture of the patient, and the Korotkoff phase used (phase IV or phase V for diastolic BP measurement). It is recommended that all BP values be recorded with the woman in a sitting position for ambulatory patients or in a semireclining position for hospitalized patients. The right arm should be used consistently, and the arm should be in an approximately horizontal position at heart level. For diastolic BP measurements, both phases—muffling sound and disappearance sound—should be recorded. This is very important because the level measured at phase IV is approximately 5 to 10 mm Hg higher than that measured at phase V. A rise in BP has been used by several authors as a criterion for the diagnosis of hypertension in pregnancy. This definition is usually unreliable because a gradual increase in BP from the second to third trimester is seen in most normotensive pregnancies. Villar and Sibai prospectively studied BP changes during the course of pregnancy in 700 young primigravidas and found that 137 patients (19.6%) had PE. The sensitivity and positive predictive values (PPVs) for PE of a threshold increase in diastolic BP of at least 15 mm Hg on two occasions were 39% and 32%, respectively. The respective values for a threshold increase in systolic pressures of at least 30 mm Hg were 22% and 33%.

Three recent studies from New Zealand, the United States, and Turkey investigated pregnancy outcomes in women with a rise in diastolic BP of more than 15 mm Hg but an absolute diastolic level less than 90 mm Hg compared with gravidas who remained normotensive. The New Zealand report and a Turkish study included women with elevated BPs without proteinuria, whereas the American investigation included women with an increased diastolic pressure by 15 mm Hg or more plus proteinuria (≥300 mg/24 h). Overall, pregnancy outcomes were similar among women who remained normotensive and those who demonstrated a rise in diastolic pressure of 15 mm Hg or higher but did not reach 90 mm Hg. The use of a specific rise in BP over baseline as a diagnostic criterion is principally influenced by two factors: gestational age at time of first observation and frequency of BP measurements. Thus a 15 mm Hg rise in diastolic BP or a 30 mm Hg rise in systolic pressure are unreliable to diagnose PE.

Prediction of Preeclampsia

A review of the world literature reveals that more than 100 clinical, biophysical, and biochemical tests have been recommended to predict or identify the patient at risk for future development of PE. The results of the pooled data for the various tests and the lack of agreement among serial tests suggest that none of these clinical tests is sufficiently reliable for use as a screening test in clinical practice.

Numerous biochemical markers have been proposed to predict which women are destined to develop PE. These biochemical markers were generally chosen on the basis of specific pathophysiologic abnormalities that have been reported in association with PE. Thus these markers have included markers of placental dysfunction, endothelial and coagulation activation, angiogenesis, and markers of systemic inflammation. However, the results of various studies to evaluate the reliability of these markers in predicting PE have been inconsistent, and many of these markers suffer from poor specificity and predictive values that are too low for routine use in clinical practice.

During the past decade, several prospective and nested case-control studies have found that certain maternal risk factors, biophysical clinical factors, and serum biomarkers obtained in the first trimester are associated with subsequent development of hypertensive disorders of pregnancy (HDPs), GH, or PE. These studies evaluated the use of these factors or markers alone or in combination, and they provided detection rates for various subtypes of hypertension and PE with a false-positive rate (FPR) of either 5% or 10%. Overall, neither the maternal factors nor the serum biomarkers, either alone or combined, had an adequate detection rate for either all HDPs or GH or PE developing at 37 weeks of gestation or later. In the same studies, using maternal factors and mean arterial pressure (MAP) in the first trimester, the detection rate for PE before 34 weeks was 73%, and for PE before 37 weeks, it was 60% with an FPR of 10%. Using data from the Fetal Medicine Foundation, the use of combined maternal factors and biophysical and biochemical markers increased the detection rate to 95% for PE that required delivery before 34 weeks of gestation and 77% for PE that required delivery at before 37 weeks of gestation with an FPR of 10%. However, the PPV for such a screen remained less than 10%. In addition, these studies were conducted in a heterogeneous group of women at various risks for HDPs and PE. It has been suggested that the performance of the FMF algorithm for predicting PE is superior to the risk factors proposed by ACOG. In contrast, a recent study by Giguère and colleagues evaluated combined maternal factors and serum markers measured in the first trimester in 7929 women who were at very low risk for GH (2.7%) and PE (1.8%). In those with PE, the incidence was 0.2% at less than 34 weeks and 1.2% at less than 37 weeks of gestation. They found that a clinical model that included maternal risk factors, BMI, and MAP had a detection rate of 54% and a PPV of 3% with an FPR of 10% for PE at less than 37 weeks of gestation, whereas a full model that also included serum biomarkers had a detection rate of 39% and a PPV of 2% for PE at less than 37 weeks of gestation. Similar studies revealed that the Fetal Medicine Foundation algorithm has lower sensitivity and PPV for PE when tested in US populations.

Based on the results of this study and other reports in recent years, it is clear that evaluation of maternal clinical factors and other biophysical and biomarkers measured in the first trimester is useful only for the prediction of those who will ultimately progress to PE that will require delivery prior to 34 weeks of gestation. However, given the poor PPV for PE before 34 weeks and the poor detection rates for all cases of GH and PE, the clinical indications for a PE screening test in the first trimester remain unclear. A major concern of screening for PE without evidence of benefit is the unintended consequences of those identified as at risk, particularly for a test with a very high FPR of cases with a very low PPV. Some authors and commercial entities have recommended that those who screen positive should have more frequent visits, more maternal and fetal testing, bedrest, and other interventions that could be potentially harmful to both mother and infant, especially when incorrectly labeled at risk by a test with a high FPR. In addition, even the most reliable predictive test will not have clinical utility unless effective preventative approaches and therapeutic interventions are available.

Currently, no prospective studies or randomized trials have evaluated the benefits and risks of first trimester screening for prediction of PE. Until then, the use of such tests for screening should remain investigational.

Doppler ultrasound is a useful method to assess uterine artery blood flow velocity in the second trimester. An abnormal uterine artery velocity waveform is characterized by a high resistance index or by the presence of an early diastolic notch (unilateral or bilateral). Pregnancies complicated by abnormal uterine artery Doppler findings in the second trimester are associated with more than a sixfold increase in the rate of PE . However, the sensitivity of an abnormal uterine artery Doppler for predicting PE ranges from 20% to 60% with a PPV of 6% to 40%. Current data do not support Doppler studies for routine screening of pregnant women for PE. The ACOG Task Force report on Hypertension in Pregnancy and the ACOG Practice Bulletin recommend only using risk factors for identifying women considered at increased risk for PE.

Prevention of Preeclampsia

Numerous clinical trials describe the use of various methods to prevent or reduce the incidence of PE. ^, Because the etiology of the disease is unknown, these interventions have been used in an attempt to correct theoretic abnormalities in PE. A detailed review of these trials is beyond the scope of this chapter; however, the results of these studies have been the subject of several recent systemic reviews. In short, randomized trials have evaluated protein or salt restriction; zinc, magnesium, fish oil, or vitamin C or E supplementation; the use of diuretics and other antihypertensive agents; and the use of heparin to prevent PE in women with various risk factors. These trials have had limited sample sizes, and results have revealed minimal to no benefit. Some of the methods studied are summarized in Box 38.4 .

Laboratory Abnormalities in Preeclampsia

Women with PE may exhibit a symptom complex that ranges from minimal BP elevation to derangements of multiple organ systems. The renal, hematologic, and hepatic systems are most likely to be involved.