Pre-eclampsia is associated with a number of short- and long-term perinatal and neonatal complications, including death. These are mostly related to birth weight and gestational age at delivery, and therefore are most relevant to severe or early onset pre-eclamptic toxaemia. Currently, little information is available on the optimal antenatal testing modality to be used for pre-eclampsia. Significant limitations are associated with fetal movement counts and the biophysical profile. Evidence is accumulating, however, to support the incorporation of umbilical artery and venous Doppler velocimetry in the evaluation of such fetuses, especially in cases of associated placental insufficiency. Pre-eclampsia might confer some survival advantage to small gestational age infants and prematurely born infants compared with infants born after spontaneous preterm labour. The degree of intrauterine growth restriction also has a negative effect on early morbidity. Longer term outcomes for prematurely born infants are dependent on gestational age, and it is unclear whether the survival advantage conferred by pre-eclampsia translates into better long-term neurodevelopmental outcomes. Abnormal umbilical artery flows might predict poorer cognitive outcomes, although evidence for this is not strong.
Introduction
Pre-eclampsia is associated with significant pathological changes of the maternal and fetal vasculature as well as the placenta, including decidual arteriolopathy, infarcts, ischaemic changes and abruption. Therefore, it is not surprising that perinatal outcomes are affected by this syndrome, particularly in the context of severe disease. The ultimate goal of fetal testing is to identify fetuses at risk of complications, including death and long-term neurological insults, and to prevent these complications.
Although various tools exist to assess fetal well-being, their use needs to be dictated by the underlying disease and its severity, gestational age, treatment approach, and the expected outcome. This requires an understanding of the physiological principles behind testing modalities, and how these can be affected in pre-eclampsia. In addition, the predictive values of these assessments must be kept in mind.
We will review short- and long-term perinatal and neonatal outcomes associated with pre-eclampsia, and examine the basis of the various testing modalities currently in clinical use. The interpretation and usefulness of each modality will be examined to suggest a clinically relevant approach for the evaluation of the fetus in the context of pre-eclampsia.
Perinatal outcomes: what are we trying to prevent?
Antenatal testing should target the potential complications associated with the pathology of interest. In the case of pre-eclampsia, those complications are largely dictated by the severity of the disease, the presence of fetal growth restriction, and the gestational age at presentation.
Pre-eclampsia is commonly associated with placental lesions. The underlying vascular manifestations, and the presence of oxidative stress and endothelial damage, affect the uteroplacental circulation and can lead to fetal growth restriction with underlying hypoxia and acidosis. The incidence of this complication varies depending on the population. A recent review reporting perinatal complications from women with pre-eclampsia seen in a tertiary-care centre over a 7-year period, reported the incidence of fetal growth restriction as 28%. Similar numbers were reported in preterm pre-eclampsia and, interestingly, no difference was found in fetal growth restriction between groups with pre-eclampsia at less than 29 weeks compared with 29–32 weeks. The risk, however, decreased significantly after 32 weeks (9%), probably reflecting a more benign process. A study examining perinatal outcomes in expectant management of pre-eclampsia at less than 27 weeks gestation also revealed a much higher incidence of fetal growth restriction (35%) in association with earlier gestational age, highlighting the fact again that early onset pre-eclampsia is of much more consequence for perinatal outcome than late preterm or term pre-eclampsia. Although fetal growth restriction clearly complicates pre-eclampsia, fetal surveillance must be evaluated in the context of early compared with late pre-eclampsia.
As pre-eclampsia is characterised by vascular damage, it is not surprising that the incidence of placental abruption is increased. In a case-control study examining the role of expectant management, a total of 239 women with pre-eclampsia were followed between 24 and 33 weeks. Although numbers were small for each reported gestational age epoch, an important trend was observed in the incidence of abruption. The rate of this complication varied from 3% at 29–30 weeks to 10–11% at 24 and 31 weeks, respectively.
Given the presence of underlying hypoxia in pre-eclampsia, and the frequent associations with fetal growth restriction, the incidence of fetal distress before or during labour is also increased. This is partly related to the minimal fetal reserves available to face a stress such as labour. This is supported by several studies in which levels of markers of chronic hypoxia (e.g.erythropoietin and nucleated red blood cells) in cord blood of fetuses born to women with pre-eclampsia were elevated. One such case-control study examined levels of erythropoietin in the amniotic fluid of fetuses of women with hypertension (including pre-eclampsia) 48 h before birth, and then re-measured these from the umbilical cord at birth. Both amniotic fluid and cord blood levels of erythropoietin were increased in fetuses of mothers with pre-eclampsia, suggesting the presence of chronic hypoxia in these pregnancies. Similarly, another group reported that infants born to women with pre-eclampsia displayed higher levels of nucleated red blood cells in their cord blood compared with control participants. Not surprisingly, this was particularly important in the presence of low-birth-weight infants and infants with fetal growth restriction. Taken together, these data suggest that fetal surveillance in pre-eclampsia must target both chronic and acute hypoxia.
Finally, a most obvious complication that requires prediction and prevention in pre-eclampsia is intrauterine fetal death (IUFD). The risk of IUFD varies widely depending on the population, severity of disease, and the presence of co-morbid factors. For instance, reported perinatal mortality rates vary from 47 to 370 per 1000. An increase in risk from mild to severe pre-eclampsia has also been reported. The country in which the woman is being treated also influences the incidence of IUFD. For instance, in the UK, 6% of children born to women who have had eclampsia die, compared with 27% of those born to women with eclampsia in low- and middle-income countries. For women with pre-eclampsia, infant mortality is three times higher in low- and middle-income countries than in high-income countries. The underlying causes of IUFDs related to pre-eclampsia include acute and chronic hypoxia, placental insufficiency, fetal growth restriction, and placental abruption.
Taken together, these complications and their underlying pathophysiologies suggest that fetal surveillance in pre-eclampsia must include an evaluation of fetal size and growth (evaluating placental function), amniotic fluid volume (a function of placental transport and renal perfusion), heart rate parameters, and cardiovascular parameters.
Perinatal outcomes: what are we trying to prevent?
Antenatal testing should target the potential complications associated with the pathology of interest. In the case of pre-eclampsia, those complications are largely dictated by the severity of the disease, the presence of fetal growth restriction, and the gestational age at presentation.
Pre-eclampsia is commonly associated with placental lesions. The underlying vascular manifestations, and the presence of oxidative stress and endothelial damage, affect the uteroplacental circulation and can lead to fetal growth restriction with underlying hypoxia and acidosis. The incidence of this complication varies depending on the population. A recent review reporting perinatal complications from women with pre-eclampsia seen in a tertiary-care centre over a 7-year period, reported the incidence of fetal growth restriction as 28%. Similar numbers were reported in preterm pre-eclampsia and, interestingly, no difference was found in fetal growth restriction between groups with pre-eclampsia at less than 29 weeks compared with 29–32 weeks. The risk, however, decreased significantly after 32 weeks (9%), probably reflecting a more benign process. A study examining perinatal outcomes in expectant management of pre-eclampsia at less than 27 weeks gestation also revealed a much higher incidence of fetal growth restriction (35%) in association with earlier gestational age, highlighting the fact again that early onset pre-eclampsia is of much more consequence for perinatal outcome than late preterm or term pre-eclampsia. Although fetal growth restriction clearly complicates pre-eclampsia, fetal surveillance must be evaluated in the context of early compared with late pre-eclampsia.
As pre-eclampsia is characterised by vascular damage, it is not surprising that the incidence of placental abruption is increased. In a case-control study examining the role of expectant management, a total of 239 women with pre-eclampsia were followed between 24 and 33 weeks. Although numbers were small for each reported gestational age epoch, an important trend was observed in the incidence of abruption. The rate of this complication varied from 3% at 29–30 weeks to 10–11% at 24 and 31 weeks, respectively.
Given the presence of underlying hypoxia in pre-eclampsia, and the frequent associations with fetal growth restriction, the incidence of fetal distress before or during labour is also increased. This is partly related to the minimal fetal reserves available to face a stress such as labour. This is supported by several studies in which levels of markers of chronic hypoxia (e.g.erythropoietin and nucleated red blood cells) in cord blood of fetuses born to women with pre-eclampsia were elevated. One such case-control study examined levels of erythropoietin in the amniotic fluid of fetuses of women with hypertension (including pre-eclampsia) 48 h before birth, and then re-measured these from the umbilical cord at birth. Both amniotic fluid and cord blood levels of erythropoietin were increased in fetuses of mothers with pre-eclampsia, suggesting the presence of chronic hypoxia in these pregnancies. Similarly, another group reported that infants born to women with pre-eclampsia displayed higher levels of nucleated red blood cells in their cord blood compared with control participants. Not surprisingly, this was particularly important in the presence of low-birth-weight infants and infants with fetal growth restriction. Taken together, these data suggest that fetal surveillance in pre-eclampsia must target both chronic and acute hypoxia.
Finally, a most obvious complication that requires prediction and prevention in pre-eclampsia is intrauterine fetal death (IUFD). The risk of IUFD varies widely depending on the population, severity of disease, and the presence of co-morbid factors. For instance, reported perinatal mortality rates vary from 47 to 370 per 1000. An increase in risk from mild to severe pre-eclampsia has also been reported. The country in which the woman is being treated also influences the incidence of IUFD. For instance, in the UK, 6% of children born to women who have had eclampsia die, compared with 27% of those born to women with eclampsia in low- and middle-income countries. For women with pre-eclampsia, infant mortality is three times higher in low- and middle-income countries than in high-income countries. The underlying causes of IUFDs related to pre-eclampsia include acute and chronic hypoxia, placental insufficiency, fetal growth restriction, and placental abruption.
Taken together, these complications and their underlying pathophysiologies suggest that fetal surveillance in pre-eclampsia must include an evaluation of fetal size and growth (evaluating placental function), amniotic fluid volume (a function of placental transport and renal perfusion), heart rate parameters, and cardiovascular parameters.
Indications for fetal surveillance
Although hypertensive disorders of pregnancy are one of the most common indications for fetal surveillance, little evidence is available to support the use of specific modes of fetal assessments in this context. Furthermore, the timing and frequency of testing have not been adequately evaluated; therefore, the lack of available information precludes us from making strong evidence-based recommendations about timing and frequency of assessments.
On the basis of the premise that distinct and observable fetal responses occur in reaction to hypoxia and acidosis, and on the fact that pre-eclampsia can be associated with such severe oxygenation abnormalities and with fetal death, it would seem reasonable to conclude that antenatal fetal surveillance is indicated in this population. Yet, the usefulness of such assessments needs to be examined in the context of the specific fetal risks and of the predictive value of the tests being used. On this basis, most authorities believe that mild hypertension without superimposed pre-eclampsia does not constitute an indication for fetal surveillance.
With pre-eclampsia, it is generally recommended that some form of antenatal surveillance is established. Unfortunately, no randomised-control trials have been carried out to indicate which testing modality to use. The Society of Obstetricians and Gynaecologists of Canada in its guideline on diagnosis, evaluation and management of the hypertensive disorders of pregnancy recommends serial fetal surveillance, and specifically advises on the use of umbilical artery velocimetry. In the USA, the report from The National High Blood Pressure Education Program on high blood pressure in pregnancy recommended that women with ‘mild’ pre-eclampsia before term should be followed with daily fetal movement assessment, weekly non-stress tests and cardiotocographs, biophysical profiles, or all three. If fetal growth restriction or decreased amniotic fluid volume are present, testing is recommended twice weekly. The guideline for the management of hypertensive disorders of pregnancy published by The Society of Obstetric Medicine of Australia and New Zealand for the management of hypertensive disorders of pregnancy in 2008, suggested ultrasound for fetal growth, amniotic fluid volume, and umbilical artery Doppler be carried out at the onset of disease and repeated every 2–3 weeks with the addition of cardiotocographs twice weekly. If fetal growth restriction is also present, ultrasound for amniotic fluid volume or umbilical artery Doppler twice weekly was also recommended. In cases of severe early onset pre-eclampsia managed expectantly, Sibai and Barton have suggested that daily fetal cardiotocographs or ultrasound surveillance may be useful.
Therefore, the available data and expert opinions suggest that mild and severe pre-eclampsia are indications for fetal testing. In the mild form of the disease, combinations of fetal growth evaluations every 3–4 weeks and cardiotocographs are generally suggested. More severe forms of the disease likely warrants much closer surveillance, which includes umbilical artery Doppler evaluations.
Specific testing modalities
Fetal movement count
In general, fetal movements are perceived by the mother from 18–20 weeks in primigravidae and at 16 weeks for multiparous women. Fetal activity is influenced by maturation of the central nervous system, sleep status, amniotic fluid volume and maternal ingestion of drugs and sedatives. In addition, fetal compromise is also associated with decreased fetal movement. This change in fetal activity has been reported to precede fetal demise by 1–3 days. As such, the maternal perception of fetal movements has been used as a form of surveillance and has been suggested in the context of pre-eclampsia. At this time, however, information is lacking on the use of this specific testing modality in pre-eclampsia. As pre-eclampsia is a high-risk condition, information can be extrapolated from other more general studies. The best available data was summarised in a recent systematic Cochrane review, in which four trials, involving 71,370 women, were included (2,716 individually randomised women; 68,654 women randomised in 33 clusters). Unfortunately, none of the studies compared the effects of fetal movement counting selectively or routinely with no counting on perinatal outcome; as such, the reviewers could not confirm nor refute the effectiveness of fetal movement counting as a method of fetal surveillance. Although this simple and inexpensive method may be helpful, it is important to keep in mind a previous review of two randomised trials, which found that routine counting led to frequent hospital admissions and increased caesarean sections in response to decreased fetal movements.
Non-stress test, cardiotocographs and contraction stress test
The Contraction Stress Test was the first method of fetal heart rate evaluation introduced. It relied on the anticipated late decelerations that a compromised fetus would display in reaction to a contraction, which presumably would partially reduce uteroplacental perfusion. The rate of IUFD reported to occur within 1 week of a negative Contraction Stress Test is 0.04%. This modality, however, is limited by its false-positive rate of up to 30%, and by the various contraindications, such as prematurity and placenta praevia. In most centres it has gradually been replaced by the Non-stress Test and the cardiotocograph, which evaluates fetal heart rate variation and the presence of accelerations as well as decelerations, presumably reflecting the underlying fetal status. The Non-stress Test is reported to have a negative predictive value of 99% for fetal and neonatal death within 1 week of testing. Yet, four blinded randomised trials evaluating the Non-stress Test and cardiotocograph showed an association with an increasedtrend in perinatal deaths (OR 2.85; 95% CI 0.99 to 7.12), thus raising questions about advisability of its use.
The above evidence has led most experts to suggest that the Non-stress Test and cardiocograph is restricted to high-risk groups only, which obviously includes women with pre-eclampsia. Evidence, however, is currently lacking to either support this practice or guide the practitioner on the frequency of testing. Most authorities suggest that women with mild pre-eclampsia undergo weekly Non-stress Test and cardiotocograph, whereas those with severe early onset pre-eclampsia (in particular in the presence of fetal growth restriction or oligohydramnios) be followed with twice weekly Non-stress Test. Nevertheless, in cases of fetal growth restriction, the authors of a recent Cochrane review concluded that ‘there is almost no evidence to date to indicate an optimal antenatal surveillance method for infants identified with impaired growth. This highlights the urgent need for research in this area.
Biophysical profile
The biophysical profile (BPP) developed in the 1980s by Manning et al. relies on the premise that fetal hypoxia results in short- and long-term alterations in variables that can be measured directly by ultrasound and by fetal heart rate monitoring (Non-stress Test and cardiotocograph). In particular, fetal hypoxia is associated with marked changes in central-nervous system performance, thus leading to reduced activity, breathing movements and tone. In the longer term, decreased fetal oxygenation leads to redistribution of renal blood flow to the central nervous system as an adaptive and protective mechanism. This results in reduced amniotic fluid volume. Unfortunately, even though the biophysical profile has been used clinically for decades, evidence is currently lacking to support this practice in high-risk pregnancies. It is indeed surprising that, so far, fewer than 3,000 women have been included in randomised-controlled trials evaluating the usefulness of this form of antenatal testing.
Data from the Pre-eclampsia Integrated Estimate of RiSk (PIERS) database suggest that the BPP has limited, if any, role in the fetal assessment for pregnancies complicated by pre-eclampsia. The authors assessed whether the addition of ultrasound components of the BPP to a Non-stress Test and cardiotocograph improved the prediction of perinatal mortality, fetal acidaemia or the PIERS study combined adverse perinatal outcome defined as one or more of the following: stillbirth, neonatal death, bronchopulonary dysplasia (BPD), intraventricular hemorrhage (IVH Grade III or IV) cystic periventricular leukomalacia (CPL), NEC, hypoxic ischemic encephalopathy (HIE), retinopathy of prematurity (ROP Grade 3–5). Data consisted of information from 89 women in the PIERS database with a full BPP (Non-stress Test, cardiotocograph and ultrasound components) carried out within 72 h of delivery. Odds ratios, sensitivity, specificity and likelihood ratios were calculated for each test to assess the strength of its association with the outcome. A BPP ≤ 4/10 was considered abnormal. In total, 18 (20.2%) women had babies with combined adverse perinatal outcomes, and 13 (14.6%) had fetal acidaemia. Eighteen (20.2%) women had a non-reassuring fetal heart rate; of these seven (38.8%) suffered one or more adverse outcomes and six (33.3%) had acidemia. Five (5.6%) women had an abnormal BPP; of these, two (40.0%) suffered one or more adverse outcomes and two (40.0%) had acidaemia. No perinatal deaths occurred in any group. The investigators concluded that no evidence was available to suggest that the addition of the ultrasound components of the BPP improved the prediction of neonatal outcomes for women with pre-eclampsia, and that larger studies are required to further evaluate this test.
To date, no randomised trial has assessed the use of BPP in pre-eclampsia. Particular concerns have been raised by another recent report on the use of the BPP in 48 preterm fetuses with growth restriction. In this group, seven women had induced delivery because of pre-eclampsia and four were induced because of haemolysis, elevated liver enzymes and low platelet count syndrome. Results indicated that, among this population, the BPP alone did not predict fetal hypoxia and death. For instance, in 10 fetuses with a normal BPP (score of 8), non-reassuring fetal heart rate monitoring was reported. In this group, three fetuses died, and six fetuses had acidaemia.
Similarly, in a prospective evaluation of 56 fetuses with fetal growth restriction (of which 12 required induced delivery because of severe pre-eclamptic toxaemia), the BPP alone was not predictive of academia at birth (as defined by cord pH <7.2). Used as a stand-alone testing modality, its sensitivity was only of 53% with a false positive rate of 26% and a false negative rate of 47%. The investigators identified particular limitations of the BPP, including the high rate of non-reactive Non-stress Tests, which result in the need for increased testing and the limited prediction of longitudinal progression. Furthermore, in a prospective observational study of 113 pregnancies complicated by intrauterine growth restriction due to placental insufficiency (of which 13 women also had a diagnosis of severe pre-eclamptic toxaemia), the BPP did not predict neurodevelopmental outcome at 2 years of age. Finally, a recent Cochrane review of the use of BPP in high-risk pregnancies in general concluded that ‘there is currently insufficient evidence from randomized control trials to support the use of BPP as a test of fetal well being in high risk pregnancies’.
Taken together, these data suggest that BPP alone is unlikely to provide reliable evidence of fetal status. It is possible that, in women with pre-eclampsia, and in particular those with concomitant intrauterine growth restriction, at least two to three testing modalities need to be considered. As such, most authorities recommend a combination of fetal movement count, Non-stress Test, BPP and umbilical artery Doppler weekly for women with mild disease and twice weekly or more frequently for those with severe disease, including intrauterine growth restriction.
Doppler velocimetry
The antenatal surveillance of women at risk or with pre-eclampsia lends itself to the incorporation of the evaluation of the uteroplacental and the fetal circulation. Pre-eclampsia is thought to occur in two phases: the first one consisting of failure of trophoblast invasion and subsequent spiral artery remodelling, which normally occur in the first and early second trimester. This results in increased resistance in the uteroplacental circulation, which can be detected by the presence of notching in the uterine artery Doppler waveform at 20–22 weeks. More recently, data have also suggested that increased resistance indices (pulsatility index and resistive index) in the uterine artery in the first trimester can be a maker for the development of pre-eclampsia and fetal growth restriction. Because of the ability of Doppler velocimetry to evaluate non-invasively the uterine and placental vasculature, this tool has also been used in the assessment of fetuses from high-risk pregnancies, including fetal growth restriction and pre-eclampsia (particularly early onset disease) ( Fig. 1 ). This is based on the fact that both conditions are characterised by uteroplacental insufficiency such that, ultimately, changes in the fetal compartment would be expected to occur. Although most studies report surveillance of the growth-restricted fetus without discriminating for the presence of underlying pre-eclampsia, some evidence suggests that this syndrome may not alter fetal responses in a way that would invalidate the use of such testing. A recent study reported on the outcomes of 31 normotensive growth-restricted fetuses, 31 growth-restricted fetuses with pre-eclampsia and 120 control fetuses. Indices evaluated included umbilical artery, middle cerebral artery, ductus venosus pulsatility index, aortic isthmus blood flow index, E–A ratios, left modified myocardial performance index, and combined cardiac output. The results indicated no difference in uteroplacental and cardiovascular indices between normotensive growth-restricted fetuses and those associated with pre-eclampsia. Therefore, it would seem appropriate to extrapolate the published medical literature on fetal growth restricton, and findings described below to the growth-restricted fetus of a women with pre-eclampsia.
The initial reports of Doppler evaluations focused on the assessment of fetal blood flow in the umbilical artery. The waveforms obtained reflect the downstream resistance, the fetal cardiac contraction force, density of the blood, and the vessel wall elasticity.
Of all indices used clinically to evaluate the umbilical arterial flow, the systolic–diastolic ratio has been the most reliable. During normal development, resistance decreases with advancing gestation, thereby responding to the increased fetal needs. As such, the values of systolic–diastolic decrease from the first to the third trimester, and importantly, the waveform, consistently displays the presence of end diastolic flow. However, in the presence of uteroplacental insufficiency, such as can be seen in early onset and severe pre-eclampsia (and certainly in pre-eclampsia with fetal growth restriction), this forward flow in end diastole is gradually lost. This occurs in the context of significant abnormalities in the uteroplacental circulation, as it is estimated that abnormally high systolic–diastolic ratios are seen when about60–70% of the placental vascular tree is no longer functional.
A recent systematic review of the use of umbilical Doppler in high-risk pregnancies, including pre-eclampsia and fetal growth restriction, examined a total of 18 studies, which included 10,156 women. Results indicated that umbilical Doppler assessment was associated with a significant decrease in perinatal mortality (RR 0.71, 95% CI 0.52 to 0.98). The rate of caesarean section was decreased in the Doppler group (RR 0.90, 95% CI 0.84 to 0.97), as was the number of inductions of labour (average RR 0.89, 95% CI 0.80 to 0.99). Taken together, these data suggested that Doppler interrogation of the umbilical artery contributed to the accurate identification of the fetus at risk.
Although not all high-risk pregnancies included suffered from pre-eclampsia, the effects on perinatal mortality is also evident in that population, although, numbers are much smaller. It is important, however, to highlight that these results apply to women with severe, early onset disease, or both, particularly with fetal growth restriction, as the underlying pathology of interest is uteroplacental insufficiency. The investigators concluded appropriately that Doppler studies of the umbilical artery should be incorporated into protocols for fetal monitoring in high-risk pregnancies thought to be at risk of placental insufficiency.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
