Predicting and preventing pre‐eclampsia—where to next?

Aris T PapageorghiouUltrasound 2008 · DOI · OpenAlex

Abstract

Pre-eclampsia complicates about 2% of pregnancies. It accounts for at least 50 000 maternal deaths per year worldwide1, 2 and is the second most common cause of maternal death in the developing world. The incidence of the disease varies according to the population studied: in Latin America rates are much higher (8–10%), and here the disease is the leading cause of maternal death3. In developed countries it is perinatal rather than maternal mortality that is more frequently the problem, and pre-eclampsia is the commonest cause of iatrogenic prematurity. In addition, pre-eclampsia frequently coexists with intrauterine growth restriction (IUGR) and placental abruption, other important causes of adverse perinatal outcome. There have been many studies investigating whether the disease can be prevented. In order to achieve this, one needs to be able to identify firstly those women that are at highest risk of the disease (and therefore most likely to benefit from prophylactic therapy), and secondly a suitable agent that can prevent pre-eclampsia. Recent progress towards both of these goals has been made. Antenatal care originally developed with the aim of detecting pregnancy complications at an early stage. It is known that maternal factors such as primigravidity, black race, previous pre-eclampsia, pre-existing hypertension or vascular disease increase the chances of developing pre-eclampsia. In daily practice, we assess the risk of disease using multiple tests (e.g. history, examination and investigations). When doing so, it is important to take into account interdependence or overlap of information between tests, allowing the relative importance of different factors in prediction to be assessed. Previous studies4-6 have shown that using factors in the maternal history as a predictor has a detection rate of about 45% for a screen-positive rate of 10–25%. In other words, more than half the women who develop pre-eclampsia have no risk factors in their history, and many women with such risk factors do not develop the disease. During physiological pregnancy, the blood supply to the intervillous space increases. This increase is due mainly to invading trophoblast, which causes vascular remodeling of the uterine spiral arteries by invading the media and endothelium and by replacing the muscular and elastic tissue of the arterial wall. Histological examination of the placenta has shown that this remodelling is incomplete in pregnancies affected with early onset pre-eclampsia and IUGR7-9. Assessment of impedance to uterine artery blood flow using Doppler ultrasound, first described by Campbell et al.10, gives a summary measure of the state of trophoblast invasion. Early studies showed that impedance to flow in the uterine arteries decreases with gestation in normal pregnancies, while in pregnancies with established pre-eclampsia or IUGR the impedance is increased11-13. These observations led to a number of screening studies in the second trimester, which assessed if it is possible to predict those pregnancies destined to have complications of impaired placentation14. These studies showed that women with increased impedance to uterine artery blood flow have an increased risk of developing pre-eclampsia, with a likelihood ratio (LR) of about 6, while those women in whom Doppler results are normal have a low risk (LR of approximately, 0.5). The detection rate for pre-eclampsia using Doppler is about 40% for a 5% screen-positive rate, and detection can be further increased by using Doppler in combination with maternal history4-6. Although about 2% of women are affected by pre-eclampsia, severe pre-eclampsia requiring delivery before 34 weeks of gestation occurs in about 0.5% of women. It is this severe end of the spectrum that is associated with IUGR and increased risk of perinatal mortality and morbidity15, 16. One very consistent and important feature of uterine artery Doppler screening studies is that the test is better at predicting this severe disease than it is at predicting less severe disease. While an increased mean uterine artery pulsatility index (PI) (defined as PI > 95th percentile) will identify 40% of women who develop pre-eclampsia, it will predict 80% of those with severe pre-eclampsia requiring delivery before 34 weeks17. Another way of assessing disease severity is by the coexistence of IUGR. At all gestational ages, the sensitivity of uterine artery Doppler in predicting pre-eclampsia with IUGR is much higher (70%) than it is in predicting pre-eclampsia without IUGR (25%) for a 5% screen-positive rate17. A recent paper in this Journal confirmed that pre-eclampsia requiring early delivery is more likely to be associated with IUGR than is less severe pre-eclampsia with delivery at term18. Thus, to summarize, uterine artery Doppler in the second trimester is relatively poor at screening for late-onset or mild disease, but good at screening for disease that is severe. This is not surprising: a study comparing placental lesions in pregnancies affected with pre-eclampsia to those with normal outcome showed not only that there is an increased amount of placental disease in pre-eclampsia, but also that the higher the rates of decidual arteriolopathy, central infarction and hypermaturity of villi within the pre-eclamptic group, the earlier the gestational age at delivery19. Over the last 5–10 years, interest in uterine artery Doppler assessment in predicting pre-eclampsia has shifted from the second to the first trimester. Although the detection rates from these studies are lower compared with those from second-trimester screening20, they confirm that the sensitivity for predicting severe or early-onset disease is much higher than is that for mild or late-onset disease. In the largest study to date6, the detection rate for pre-eclampsia before 34 weeks was 82%, while it was 30% for late disease, for a 10% screen-positive rate. Importantly, correlation and interdependence with maternal variables was investigated and, while this showed that combined screening using maternal variables and uterine artery Doppler resulted in improved detection for late disease (from 30% to 50%), taking maternal factors into account for pre-eclampsia before 34 weeks did not improve the performance of uterine artery Doppler (detection rate, 82%). There is now overwhelming evidence to show that abnormal uterine artery Doppler in the first trimester of pregnancy is associated with abnormalities in trophoblast invasion. Studies have shown that the proportion of decidual vessels with endovascular trophoblast invasion is significantly higher in pregnancies with low uterine artery resistance index (RI) compared to those with high RI21. Normal placental development relies on stringent regulation of processes such as proliferation, motility and apoptosis. One possible underlying mechanism for reduced trophoblast invasion was suggested in a recent study by Whitley et al.22, in which extravillous trophoblasts (EVTs) derived from women at just 11 weeks of gestation were subjected to apoptotic stimuli. EVTs from women with high RI were significantly more susceptible to apoptosis than were EVTs from pregnancies with normal RIs. It is certainly possible that an increase of apoptosis in the endovascular extravillous trophoblast could result in reduced spiral artery remodeling. Whether this is a primary placental problem or under the control of a maternal serum factor (often described as factor 'X') is unknown, but in-vitro studies have shown that maternal serum from women with abnormal uterine artery Doppler inhibits trophoblast invasion23. To summarize, abnormal uterine artery Doppler is a very good predictor of early- but not of late-onset pre-eclampsia, and abnormal uterine artery Doppler is associated with abnormal trophoblast development. One possible conclusion from this is that late-onset pre-eclampsia is associated less strongly, or not at all, with abnormal trophoblast development. There is some epidemiological evidence suggesting that preterm pre-eclampsia and pre-eclampsia with IUGR at term share similar disease origins, while term pre-eclampsia is associated with normal or even large-for-gestational age fetuses24. It has been suggested that although the syndrome of pre-eclampsia is the final common pathway for both preterm and term pre-eclampsia, the former is associated with defective trophoblast development—which may be detected by uterine artery Doppler—while the latter is caused by atherosclerotic changes in spiral arterioles which have previously undergone normal trophoblast invasion—explaining the normal uterine Doppler in these cases25, 26. In addition, studies have shown not only that placental pathology increases with lower gestational age at the time of delivery for women with pre-eclampsia19 or with pre-eclampsia and coexisting IUGR27, but also that placental morphology in late-onset pre-eclampsia in the absence of IUGR is minimally different when compared with gestational-age-matched controls27. Such differences in etiology could therefore explain the relatively poor prediction of term pre-eclampsia by uterine artery Doppler studies. Many older studies assessed uterine artery Doppler for the detection of hypertensive diseases in general, including gestational or pregnancy-induced hypertension14, and it took many years for screening studies to concentrate on pre-eclampsia. It may be that the time has now come to take things a step further. Uterine artery Doppler studies in the second trimester are a very effective test in detecting preterm pre-eclampsia or pre-eclampsia associated with IUGR. However, reporting screening results for early onset pre-eclampsia and pre-eclampsia at term simultaneously has resulted in a number of reviews suggesting that uterine Doppler is of limited use28, 29. This is unfair criticism of a test that will predict over 80% of women developing severe preterm pre-eclampsia for a low screen-positive rate many weeks before clinical onset of the disease, something that cannot be achieved by the traditional method of noting the maternal history4-6. It may be difficult for researchers to exclude from their reports cases that are uncommon and that have been acquired painstakingly, such as women who develop term pre-eclampsia. However, given findings from previous studies, it appears that these cannot be detected effectively by a test that examines a separate etiological origin. It is up to us in the ultrasound community to ensure that the message is clear when reporting such studies. One of the reasons for a move towards first-trimester screening is that prevention of pre-eclampsia by starting pharmacological intervention in the second trimester has largely failed. The latest promising intervention was the use of antioxidants, which followed a long line of potential candidates, including protein restriction or supplementation, increasing or decreasing salt intake, magnesium, zinc, fish oil, evening primrose oil or calcium supplementation and aspirin administration. However, a number of placebo controlled trials of vitamins C and E (which were recently subject to a meta-analysis) did not demonstrate a reduction in the frequency of pre-eclampsia30. Calcium supplementation, an area of research that often seems to be overlooked, appears to almost halve the risk of pre-eclampsia and may be particularly helpful in women at highest risk or those with a low dietary intake of calcium31. Starting supplementation in the middle trimester of pregnancy may be too late to be fully effective, and supplementation in early pregnancy or even before conception may be of more benefit, with some authors suggesting that efforts should be redirected towards calcium supplementation at a community level32. Aspirin, which has been studied extensively, has been shown to reduce the incidence of pre-eclampsia when compared to placebo, but the reduction is modest33. Although this individual patient meta-analysis showed no clear evidence that the relative risk is reduced more if the therapy is started earlier, a number of small randomized trials have suggested that aspirin therapy may be of particular benefit if started in the first trimester in women at high risk of developing the disease on the basis of history and abnormal first-trimester uterine artery Doppler34-36. It is evidence from such studies and the emerging knowledge that pre-eclampsia is a consequence of first-trimester events that have focused the attention of screening to earlier in gestation; high-risk women identified through first-trimester screening will be the most likely candidates to benefit from pharmacological interventions in future trials. Interest in using maternal serum markers that assess fetoplacental endocrinology, maternal renal function, endothelial dysfunction, abnormal oxidative stress or maternal inflammation is likely to continue, and evaluation of promising single- and multiple-marker models is ongoing. Again, research has moved from the second to the first trimester of pregnancy, and markers that are 'piggy-backed' onto existing screening tests, such as for Down syndrome at 11 to 13+6 weeks, are likely to be most readily accepted37. Estimation of placental volume by three-dimensional (3D) ultrasound has shown some promise, but larger observational studies are needed38. If 3D ultrasound, uterine artery Doppler and maternal serum markers are all measurable manifestations of reduced trophoblast invasion, it is likely that there is some correlation between them, and careful investigation of possible associations between placental volume related to crown–rump length and other factors is necessary. One important question that may be answered soon is: why is it that some women with abnormal uterine artery Doppler, and therefore abnormal trophoblast development, develop severe pre-eclampsia, while others have normal pregnancy outcome? The answer to this is likely to lie in central maternal cardiovascular changes. Maternal echocardiographic studies in low-risk women have shown differences in maternal cardiac function between women who will have normal pregnancy outcome and those developing pre-eclampsia (with or without IUGR) at just 11–14 weeks of gestation39. The ability of uterine artery Doppler to reflect the state of trophoblast development means that it will remain a powerful research tool for the identification of women at high risk, making the exploration of other markers—whether in serum, placenta or heart—a practical proposition. Once more effective early identification of women at high risk is achieved, the evaluation of pharmacological interventions in the first trimester will be possible. In the meantime, the clinical application of uterine artery Doppler, a simple, safe, acceptable and accurate test, should be used to assess women's risk for the subsequent development of preterm pre-eclampsia in order to plan appropriate antenatal care.
BibTeX
@inproceedings{papageorghiou2008predicting,
  author = {Aris T Papageorghiou},
  title = {{Predicting and preventing pre‐eclampsia—where to next?}},
  booktitle = {Ultrasound},
  year = {2008},
  doi = {10.1002/uog.5320},
}

Loading related content...

Figure