Pathophysiology

The pathogenesis of pre‐eclampsia originates in the placenta. The disease can occur in the absence of fetal tissue (molar pregnancy) and manifestations of the disease will only resolve following delivery of the placenta. The blueprint for establishing pre‐eclampsia is determined at the outset of pregnancy when placental trophoblast invades the maternal uterine spiral arteries at the time of implantation. In pregnancies destined to be complicated by pre‐eclampsia, transformation of the spiral arteries is impaired, with suboptimal remodelling of small‐capacitance constricted vessels into dilated large‐capacitance conduits. The prevailing theory has been that the subsequent relative placental ischaemia causes release of vasoactive factors into the circulation which then give rise to endothelial‐mediated end‐organ damage and clinical manifestations of the disease. Scientific endeavours to determine these elusive vasoactive factors have largely been responsible for pre‐eclampsia being known as the ‘disease of theories’.

Several candidates have been considered in the role of a key circulating vasoactive factor, including interleukins, tumour necrosis factor (TNF)‐α and components of the angiotensin pathway. Whilst all these elements are subject to modification in pre‐eclamptic pregnancies, it has not been possible to demonstrate that any have an initiating role in the disease process. Pre‐eclampsia is a disease of higher primates only and the lack of a clinically relevant animal model has been a significant research obstacle. The discovery of soluble fms‐like tyrosine kinase (sFlt)‐1 has been particularly exciting because it is the first candidate that has been demonstrated to cause a pre‐eclampsia phenotype in an animal model [5].

sFlt‐1 is variant of vascular endothelial growth factor receptor (VEGFR)‐1, which has an extracellular ligand‐binding domain but lacks the transmembrane and cytoplasmic domains. Circulating sFlt‐1 is therefore able to competitively bind to VEGF and placental growth factor (PGF) and therefore reduce biologically active binding of these factors that usually promote angiogenesis and placentation. Women with pre‐eclampsia have increased circulating levels of sFlt‐1 and reduced circulating free VEGF and PGF. VEGF is important in maintaining normal fenestration of the glomerular endothelium [6,7] and it has been suggested that the early renal manifestations of pre‐eclampsia may be a consequence of the particular sensitivity of the kidney to reduced levels of VEGF. In animals it has been shown that both VEGF and PGF must be reduced to cause a pre‐eclampsia phenotype [5]. In vitro and in vivo studies [8] have shown that the hypoxic placenta produces increased levels of sFlt‐1 and primate studies [9] indicate that this may be sufficient to produce a pre‐eclampsia phenotype. Another factor in this story is endoglin (sEng), a modified form of the transforming growth factor (TGF)‐β coreceptor. sEng is also increased in pre‐eclampsia and has been shown to augment the effect of sFlt‐1 and is particularly associated with hepatic endothelial damage [10]. Importantly, sFlt‐1, PGF and sEng have been shown to be elevated in the serum of women destined to suffer pre‐eclampsia several weeks in advance of clinically evident disease [11].

An intriguing aspect of this hypothesis is its link to a potential explanation as to why smokers have a reduced incidence of pre‐eclampsia. The combustible component of cigarette smoke induces haemoxidase (HO)‐1. This is a stress response gene that has a cellular protective role, particularly against hypoxic injury. HO‐1 degrades haem into biliverdin, carbon monoxide (CO) and free iron. Both biliverdin and CO have been demonstrated to reduce endothelial expression of sFlt‐1 and sEng [12]. Appreciation of the potential role of the HO‐1 pathway has led to the suggestion that pharmacological agents known to have HO‐1 activity might be useful in ameliorating pre‐eclampsia. Statins are widely used outside obstetrics to reduce serum lipids and forthcoming studies will evaluate whether their theoretical potential can be translated safely and usefully into pregnancy.

Defining hypertensive disease in pregnancy

There has always been considerable debate over the most appropriate definition of the hypertensive disorders in pregnancy. It has been recognized that there are benefits in having a broader clinical definition whilst retaining a very tight phenotypic research definition. Hypertension complicates 6–12% of all pregnancies [13], and includes two relatively benign conditions (chronic and gestational hypertension) and the more severe conditions of pre‐eclampsia or eclampsia. Pre‐eclampsia complicates 3–5% of all pregnancies, and is characterized by placental and maternal vascular dysfunction that may lead to adverse outcomes such as severe hypertension, stroke, seizure (eclampsia), renal and hepatic injury, haemorrhage, fetal growth restriction, or even death [14].

The diagnosis of pre‐eclampsia, and hence the prediction of adverse events, is based on traditional but somewhat unreliable and non‐specific clinical markers such as blood pressure, urine protein excretion, and symptoms. For example, more than 20% of women who have eclampsia will fail to meet the common diagnostic criteria of pre‐eclampsia prior to their event, making the prediction of this adverse outcome extremely difficult [15]. Conversely, only 0.7–5.0% of women with classically defined pre‐eclampsia will experience any composite adverse outcomes [16].

For this reason consistency is required both for clinical management and to allow the comparison of outcomes from clinical units/regions. The National Institute for Health and Care Excellence (NICE) Clinical Guideline 107 [17] has defined management pathways for hypertension in pregnancy in the UK. The list that follows outlines the NICE definitions associated with hypertension in pregnancy used in this chapter.

• Gestational hypertension: new hypertension presenting after 20 weeks without significant proteinuria.
  
• Pre‐eclampsia: new hypertension presenting after 20 weeks with significant proteinuria.
  
• Chronic hypertension: hypertension present at the booking visit or before 20 weeks or if the woman is already taking antihypertensive medication when referred to maternity services. It can be primary or secondary in aetiology.
  
• Eclampsia: a convulsive condition associated with pre‐eclampsia.
  
• HELLP syndrome: haemolysis, elevated liver enzymes and low platelet count.
  
• Severe pre‐eclampsia: pre‐eclampsia with severe hypertension and/or with symptoms, and/or biochemical and/or haematological impairment.
  
• Significant proteinuria: defined as a urinary protein/creatinine ratio of greater than 30 mg/mmol or a validated 24‐hour urine collection result showing greater than 300 mg protein.
  
• Mild hypertension: diastolic blood pressure 90–99 mmHg, systolic blood pressure 140–149 mmHg.
  
• Moderate hypertension: diastolic blood pressure 100–109 mmHg, systolic blood pressure 150–159 mmHg.
  
• Severe hypertension: diastolic blood pressure 110 mmHg or greater, systolic blood pressure 160 mmHg or greater.

Measuring blood pressure and proteinuria in pregnancy and pre‐eclampsia

The errors associated with blood pressure measurement have been well described in both non‐pregnant and pregnant populations. Care in taking these measurements will reduce false‐positive and false‐negative results and improve clinical care. Machine/device errors have led to strict validation protocols for automated blood pressure devices in specific populations and clinical settings [18] and the human errors inherent in manual readings have led to guidelines on the measurement of blood pressure with both manual and automated devices in clinical practice [19]. Digit preference (the practice of rounding the final digit of blood pressure to zero) occurs in the vast majority of antenatal measurements and simply taking care to avoid this will limit inaccurate diagnoses. Using a standard bladder in a sphygmomanometer cuff will systematically undercuff 25% of an average antenatal population. Having large cuffs available and using them will prevent the over‐diagnosis of hypertension [20]. Keeping the rate of deflation to 2–3 mmHg/s will prevent over‐diagnosis of diastolic hypertension, as will using Korotkoff 5, which is now universally recommended for diagnosing diastolic hypertension. Korotkoff 4 (the muffling of the sound) is less reproducible, and randomized controlled trials confirmed that it is safe to abandon it, except in those rare situations when the blood pressure approaches zero [21,22].

The reliable detection of proteinuria is essential in differentiating those pregnancies with pre‐eclampsia from those with gestational or chronic hypertension and, in the process, identifying those pregnancies most prone to adverse outcome. The measurement of significant proteinuria, traditionally 300 mg excretion in a 24‐hour period, is also prone to collection and measurement error. The collection of 24‐hour urine samples is not practical as a routine test and so urine dipstick screening is employed as a first‐line screening test with secondary tests employed to confirm positive dipstick diagnoses. Visual dipstick reading is unreliable [23] but the use of automated dipstick readers significantly improves the accuracy of dipstick testing and as such is recommended by NICE for use in pregnancy [24]. NICE also recommends that quantification of proteinuria should follow diagnosis. There are two methods that NICE supports. The first is the 24‐hour urine protein estimation and this requires that an assessment of sample completeness is undertaken, with measurement of creatinine excretion being the most common. NICE also supports the use of the protein/creatinine ratio test. This test is done on a ‘spot’ urine sample and is therefore much quicker. This test has been shown in numerous studies to be comparable to the 24‐hour urine protein estimation [25]. The threshold for defining significant proteinuria by this test is 30 mg protein/mmol creatinine.

Risk assessment and risk reduction

There have been attempts to screen the antenatal population for pre‐eclampsia over the past 60 years, with over 100 potential biochemical, biophysical or epidemiological candidate tests. Despite not yet having a single universal test to apply, it is still possible to advise women regarding their risk of pre‐eclampsia from their clinical history and some investigations.

NICE guidelines for routine antenatal care [26] emphasize that a woman’s risk of pre‐eclampsia should be evaluated. Several risk factors for pre‐eclampsia are known and these have been incorporated into the NICE recommendations [27,28]. Table 7.1 outlines risk factors that should be identified at booking to identify women at risk of pre‐eclampsia. Many of the risk factors listed in this table are modifiable and may lead to a reduction in risk either prior to or between pregnancies.

Individual risk is not a simple numerical addition. A family history of pre‐eclampsia in a first‐degree relative is significant and two relatives even more so, whilst exposure over time to paternal antigen through increased periods of cohabitation and non‐barrier contraception can reduce risk as can prior miscarriage or termination of pregnancy. Pre‐eclampsia is more common at the extremes of reproductive age and is increased further following in vitro fertilization (IVF) treatment, particularly with donor sperm. Other factors often associated with increasing age, such as obesity, gestational and pre‐gestational diabetes, and any disease affecting the cardiovascular system are potent risk factors for pre‐eclampsia. The relative risk for pre‐eclampsia for some of these risk factors is shown in Table 7.2 [28].

Clearly, from the relative risks quoted, the majority of women who are high risk will still not develop pre‐eclampsia whilst a considerable number of cases will arise de novo in the ‘low‐risk’ population. Identifying women at risk will allow increase in surveillance and use of prophylactic therapies can be considered. If adequate preventive measures become available, then these screening tests will become increasingly important. Tests that might be employed to screen the population (high or low risk) for pre‐eclampsia centre on the identification of poor placental function, which is an almost universal prerequisite for the clinical condition. Doppler assessment of the maternal uterine circulation is considered to be a promising test. This test when ‘positive’ demonstrates the high resistance in the uterine arteries as well as a ‘notch’ apparent within the Doppler waveform. These two features have been used in isolation and combination to screen low‐ and high‐risk populations. Early studies suggested that approximately one in five women who have an abnormal Doppler at 20 weeks’ gestation will develop pre‐eclampsia [29], and at 24 weeks’ gestation the prediction value is greater. In 2008, NICE recommended that uterine artery Doppler screening should not be employed universally for low‐risk women [26]. More recently NICE Clinical Guideline 107 recommended that this test should not be universally employed in high‐risk women based on the relatively poor quality of the studies performed to date. However, it did recognize its potential and made a research recommendation regarding its use in the management of high‐risk women.

No other biophysical test other than accurate measurement of blood pressure in the first trimester has either any clinical application or is practical enough to employ in clinical practice. Numerous haematological and biochemical markers have been used to both predict and evaluate pre‐eclampsia. For example, in women who have chronic hypertension the measurement of uric acid and platelets can help in determining those who suffer superimposed pre‐eclampsia; again the tests lack sensitivity and specificity [30]. Furthermore, very few of these markers have been independently evaluated for their ability to separately predict the timing or severity of specific adverse outcomes such as placental abruption, severe hypertension, neurological injury and fetal growth restriction. The reason for this is that the biomarkers previously studied were mostly generic indicators of vascular activation and dysfunction, which arise late in the pre‐eclamptic disease process and which are not specific to pre‐eclampsia or even to pregnancy.

As outlined previously, recent advances have identified a class of pregnancy‐specific angiogenic and anti‐angiogenic factors (e.g. PlGF, VEGF) that are produced by the placenta and which closely correlate with the preclinical and clinical stages of pre‐eclampsia []. Assays of these markers are currently under assessment as tools to predict and/or diagnose pre‐eclampsia prior to the onset of clinical disease and significant morbidity. The FASTER trial of 2003 demonstrated that, when measured as part of the quadruple aneuploidy screen at 15–18 weeks’ gestation, the odds ratios for the development of pre‐eclampsia when inhibin‐A and β‐human chorionic gonadotrophin (hCG) levels are above the 95th centile were 3.42 (95% CI 2.7 and 4.3) and 2.20 (95% CI 1.7 and 2.9), respectively [37].

In 2008, the Society of Obstetricians and Gynaecologists of Canada Genetics Committee, following systematic review, suggested that abnormal uterine artery Doppler in combination with an elevated α‐fetoprotein (AFP), hCG and inhibin‐A, or decreased pregnancy‐associated plasma protein A (PAPP‐A), identify a group of women at increased risk of intrauterine growth restriction and pre‐eclampsia. They also stated that multiple maternal serum screening markers at present should not be used for population‐based screening as false‐positive rates are high, sensitivities are low and no protocols have shown improved outcome [38].

Screening is important to focus resources on high‐risk women as well as to identify those in whom prophylactic therapies might have some benefit. Aspirin and calcium have been found to have a beneficial effect whilst other agents, most recently antioxidants, have not proven useful. NICE Clinical Guideline 107 recommends low‐dose aspirin therapy (75 mg/day) for all high‐risk women from 12 weeks’ gestation. Antiplatelet agents were associated with statistically significant reductions in the risk of pre‐eclampsia in moderate‐risk women and in high‐risk women (moderate‐risk women: 25 studies, N = 28 469, RR 0.86, 95% CI 0.79–0.95; high‐risk women: 18 studies, N = 4121, RR 0.75, 95% CI 0.66–0.85).

A meta‐analysis using individual‐patient data from 32 217 women and their 32 819 babies found a statistically significant reduction in risk of developing pre‐eclampsia (RR 0.90, 95% CI 0.84–0.97). The data from this study suggest that one case of pre‐eclampsia would be prevented for every 114 women treated with antiplatelet agents. In addition to the 10% reduction in pre‐eclampsia in high‐risk women receiving antiplatelet agents, there was a 10% reduction in preterm birth. No particular subgroup of women in the high‐risk group was substantially more or less likely to benefit from antiplatelet agents. There was no statistically significant difference between women who started treatment before 20 weeks (RR 0.87, 95% CI 0.79–0.96) and those who started treatment after 20 weeks (RR 0.95, 95% CI 0.85–1.06; P = 0.24). Of importance, there were no statistically significant differences between women receiving antiplatelet agents and those receiving placebo in the incidence of potential adverse effects such as antepartum haemorrhage, placental abruption or postpartum haemorrhage, but there was a reduction in the risk of preterm birth before 37 weeks (RR 0.93, 95% CI 0.89–0.98) [39].

Trials of calcium to prevent pre‐eclampsia are more controversial. There is good evidence that in areas where the dietary intake of calcium is low, calcium supplementation reduces the risk of pre‐eclampsia but this is also influenced by prior risk status. In studies conducted where dietary calcium intake is normal, supplementation was not found to be of benefit. No other intervention can be recommended, including magnesium, folic acid, antioxidants (vitamins C and E), fish oils or bed rest. Diet or lifestyle changes may be beneficial for general health and weight loss may reduce the prior risk of hypertensive disease but modifications such as a low‐salt diet have no proven benefit.

Chronic hypertension

Women with chronic hypertension should receive pre‐pregnancy care. This should aim to determine the severity and cause of the hypertension; review potentially teratogenic medications such as angiotensin‐converting enzyme (ACE) inhibitors, angiotensin receptor blockers (three times the risk of congenital abnormality) and diuretics; inform women of the risk associated with pregnancy and of prophylactic strategies (all should receive low‐dose aspirin in pregnancy); and to assess comorbidities such as renal impairment, obesity or coexistent diabetes.

The main risk is of superimposed pre‐eclampsia, but even in its absence the perinatal mortality is increased. Drugs appropriate for treating hypertension in pregnancy include methyldopa, labetalol, nifedipine and hydralazine. Safety data on other antihypertensives are lacking but there are several where no association with congenital abnormality has been established and so they can be used when clinically indicated.

Blood pressure control should be tailored to the individual. Where the chronic hypertension is secondary to other disease, then the care should be multidisciplinary with the appropriate physician aiming to keep blood pressure below 140/90 mmHg and often at lower limits. When the chronic hypertension is uncomplicated (usually essential) the target should be 150–155/80–100 mmHg [17].

There is a recognized risk of fetal growth restriction (FGR) in this group and so serial fetal biometry is recommended and women should be seen with increased frequency to maintain blood pressure control and to screen for pre‐eclampsia. Delivery should be for either fetal indications or for poor hypertension control once corticosteroids for fetal lung maturity have been given if less than 34 weeks’ gestation.

At term, NICE recommends delivery after 37 weeks when agreed with the individual, so long as blood pressure control is maintained. Following delivery blood pressure should be maintained below 140/90 mmHg and medication should be reviewed and optimized for both blood pressure control and breastfeeding.