Pre-eclampsia and intrauterine growth restriction are responsible for significant maternal and fetal morbidity and mortality worldwide. Identifying pregnancies at highest risk for their development would allow increased surveillance in individual pregnancies and also allow therapeutic trials to decrease their incidences in the future. To date, multiple attempts to develop a screening test for these disorders have met with limited success. Proposed screening methods have included maternal serum biochemical parameters as well as ultrasonographic markers. Uterine artery Doppler, direct evaluation of the spiral arteries using colour and spectral Doppler, three-dimensional placental volume analysis and, most recently, three-dimensional power Doppler angiography have all been suggested. Although an adequate screening method remains elusive, advances in ultrasound technology have improved our ability to observe the pathophysiologic changes that occur with these conditions early in pregnancy, bringing us closer to a reproducible screening model.
Background
Successful pregnancy relies on the adequate transfer of oxygen, nutrients and wastes between maternal and fetal compartments. Critical to this is the trophoblast-mediated development of the uteroplacental circulation space (UPCS) in early pregnancy. During UPCS formation, migrating extravascular trophoblast cells invade the walls of the spiral arteries changing them from small-calibre, high resistance vessels into wide-calibre, low-resistance vessels that deliver blood to the intervillous space at low pressure. When complete, the UPCS includes spiral arteries and decidua on the maternal side, placental villi on the fetal side and, in between the two, the intervillous space where oxygen and nutrient exchange occurs.
UPCS development is thought to occur in two stages. The first stage takes place between 8 and 10 weeks of gestation and begins with endovascular plugging of the spiral arteries by trophoblast cells. This is followed by trophoblast invasion and destruction of the musculoelastic media of the intradecidual segments of the spiral arteries. The second stage occurs between 14 and 16 weeks, and involves trophoblast invasion of the spiral artery segments in the inner third of the myometrium. Although the precise trigger for trophoblast invasion is unclear, oxygen tension and blood flow haemodynamics are proposed to play a role in light of trophoblast selectivity for arteries but not veins.
Inadequate modification of the spiral arteries, leading to underdevelopment of the UPCS, results in increased placental resistance and has been implicated in the pathophysiology of pre-eclampsia and intrauterine growth restriction (IUGR).
With this in mind, researchers have sought to use ultrasonography to study the early signs of increased placental resistance, hoping to identify pregnancies at highest risk for the development of these pathologies. A potential screening method would ideally identify at-risk women before the second phase of placental development when therapeutic intervention might still be possible. This is supported by studies showing decreased rates of pre-eclampsia and IUGR related to aspirin treatment before 16 weeks of gestation, whereas its use in later pregnancy has not been shown to decrease these outcomes.
In this review, we will discuss ultrasonographic techniques, direct as well as indirect, that have been proposed to screen for pre-eclampsia and IUGR.
Uterine artery Doppler velocimetry
A discussion of sonographic screening methods for pre-eclampsia and IUGR would not be complete without mention of uterine artery spectral Doppler. Increased resistance at the utero-placental interface causes increased resistance upstream in the uterine arteries that can be evaluated by their Doppler insonation. In this way, uterine artery Doppler provides an indirect assessment of increased resistance in the UPCS. Abnormally high indices of resistance have been correlated with inadequate trophoblast modification of the spiral arteries on histopathologic examination. This fact, along with the ease of application and good inter- and intra-observer variability, made uterine artery Doppler a desirable method for potential screening. Studies have evaluated elevated pulsatility index, elevated resistance index, elevated systolic and diastolic ratio and presence of a diastolic notch as high-risk markers. Unfortunately, actual detection rates using the technique have been disappointing and thus limited its utility.
In the first trimester, a uterine artery pulsatility index above the 95 th centile was reported to have a sensitivity of 24–27% for detection of pre-eclampsia and 11.7–23% for detection of IUGR. The sensitivities for both entities improve when the focus is on identifying only those women with severe disease (such as those who require delivery before 32 weeks) with improvement to 60% for pre-eclampsia and 27.8% for IUGR in one study. As abnormally high indices of resistance in the first trimester have been correlated with similar findings in the second trimester, the hope was that first-trimester identification of women at high risk for developing pre-eclampsia and IUGR might allow time for therapeutic intervention before the second phase of placental development. In practice, however, the low sensitivity has limited the feasibility of using the technique as a screening method.
Second-trimester uterine artery Doppler has the benefit of improved detection rates compared with first trimester, but may be identifying pregnancies at a point in which therapeutic intervention would no longer be possible. In a multi-centre trial in London, UK, pulsatility index above the 95th centile at 23 weeks had a sensitivity of 69% for pre-eclampsia with IUGR, 24% for pre-eclampsia without IUGR, 13% for IUGR without pre-eclampsia, 41% for pre-eclampsia irrespective of IUGR and 16% for IUGR irrespective of pre-eclampsia. Again, sensitivities improved when the focus was on women requiring delivery before 32 weeks. The overall detection rates remain low, however, and a recent Cochrane review reported no benefit to the use of second-trimester uterine artery Doppler for screening for hypertensive disorders in a low-risk population.
The relatively poor detection rates of uterine artery Doppler may be explained by the indirect assessment the method provides. Similar to umbilical artery Doppler, which only reveals significant abnormalities after two-thirds of small placental villous vessels are obliterated, uterine artery Doppler may become abnormal only after both phases of placental development have failed to properly occur and high levels of resistance have resulted. In order to identify women before resistance has reached a potential point of no return, a more direct means of evaluating the crucial process of spiral artery modification might be necessary. This could be possible through direct evaluation of the UPCS.
Uterine artery Doppler velocimetry
A discussion of sonographic screening methods for pre-eclampsia and IUGR would not be complete without mention of uterine artery spectral Doppler. Increased resistance at the utero-placental interface causes increased resistance upstream in the uterine arteries that can be evaluated by their Doppler insonation. In this way, uterine artery Doppler provides an indirect assessment of increased resistance in the UPCS. Abnormally high indices of resistance have been correlated with inadequate trophoblast modification of the spiral arteries on histopathologic examination. This fact, along with the ease of application and good inter- and intra-observer variability, made uterine artery Doppler a desirable method for potential screening. Studies have evaluated elevated pulsatility index, elevated resistance index, elevated systolic and diastolic ratio and presence of a diastolic notch as high-risk markers. Unfortunately, actual detection rates using the technique have been disappointing and thus limited its utility.
In the first trimester, a uterine artery pulsatility index above the 95 th centile was reported to have a sensitivity of 24–27% for detection of pre-eclampsia and 11.7–23% for detection of IUGR. The sensitivities for both entities improve when the focus is on identifying only those women with severe disease (such as those who require delivery before 32 weeks) with improvement to 60% for pre-eclampsia and 27.8% for IUGR in one study. As abnormally high indices of resistance in the first trimester have been correlated with similar findings in the second trimester, the hope was that first-trimester identification of women at high risk for developing pre-eclampsia and IUGR might allow time for therapeutic intervention before the second phase of placental development. In practice, however, the low sensitivity has limited the feasibility of using the technique as a screening method.
Second-trimester uterine artery Doppler has the benefit of improved detection rates compared with first trimester, but may be identifying pregnancies at a point in which therapeutic intervention would no longer be possible. In a multi-centre trial in London, UK, pulsatility index above the 95th centile at 23 weeks had a sensitivity of 69% for pre-eclampsia with IUGR, 24% for pre-eclampsia without IUGR, 13% for IUGR without pre-eclampsia, 41% for pre-eclampsia irrespective of IUGR and 16% for IUGR irrespective of pre-eclampsia. Again, sensitivities improved when the focus was on women requiring delivery before 32 weeks. The overall detection rates remain low, however, and a recent Cochrane review reported no benefit to the use of second-trimester uterine artery Doppler for screening for hypertensive disorders in a low-risk population.
The relatively poor detection rates of uterine artery Doppler may be explained by the indirect assessment the method provides. Similar to umbilical artery Doppler, which only reveals significant abnormalities after two-thirds of small placental villous vessels are obliterated, uterine artery Doppler may become abnormal only after both phases of placental development have failed to properly occur and high levels of resistance have resulted. In order to identify women before resistance has reached a potential point of no return, a more direct means of evaluating the crucial process of spiral artery modification might be necessary. This could be possible through direct evaluation of the UPCS.
Colour and spectral Doppler of the uteroplacental circulation space
Colour and spectral Doppler of the UPCS, including the distal spiral arteries, is feasible beginning in the first trimester ( Fig. 1 ). As this is where placental formation and vessel modification occurs, it is a logical place for evaluation of the changes that potentially precede pre-eclampsia and IUGR. In the following section, we will discuss the use of colour and spectral Doppler of the UPCS vessels in early through late gestation.
Early gestation
Several studies have evaluated colour and spectral Doppler of the developing utero-placental vessels at less than 12 weeks. In 2004, Makikallio et al. longitudinally investigated 16 normal singleton pregnancies to determine the evolution of spectral Doppler parameters of the uterine, arcuate, radial and spiral arteries between 5 and 10 weeks. Spiral artery impedance (measured through spiral artery pulsatility index and peak systolic velocity) decreased, and blood flow increased, as early as 5–7 weeks of gestation, whereas the uterine and arcuate artery haemodynamics remained unchanged until after 8 weeks. This delay may be explained by the upstream effects of placental development occurring only after significant downstream events.
The same investigators then evaluated women after successful in-vitro fertilisation cycles. Doppler parameters in the first trimester were compared between 10 women who later developed pre-eclampsia, preterm labour, or both, and 31 women with subsequently uncomplicated pregnancies. Although an increased resistance index was seen in the intraplacental blood vessels at 8 weeks in women who developed pre-eclampsia, spiral artery Doppler parameters did not differ between the groups at any point over the study period. The investigators reported good intra- and inter-observer correlation.
In another study, Ozkaya et al. evaluated Doppler of the UPCS in 105 pregnant women between 6 and 12 weeks of gestation. Spiral artery pulsatility index and resistance index were higher in the group with an adverse pregnancy outcome (spontaneous abortion [ n = 3], missed abortion [ n = 6], preterm labour [ n = 2], IUGR [ n = 3], pre-eclampsia [ n = 1] and abruption [ n = 1]). This finding, however, was not statistically significant.
Although these studies were limited by low numbers of participants and heterogeneous study groups, it can be concluded that the early spiral artery changes visible by colour and spectral Doppler are probably insufficient to allow the use of this method to screen for pre-eclampsia and IUGR.
Late first trimester and second trimester
Although changes in haemodynamics before 12 weeks might be insufficient to predict pre-eclampsia and IUGR, evaluation in the second trimester (after the first phase of placental development is completed) might prove more sensitive.
Hung et al. evaluated colour and spectral Doppler of the spiral arteries at 13–19 weeks and 20–25 weeks. Their two-part study involved establishment of spiral artery waveform nomograms in 175 uncomplicated singleton pregnancies followed by the prospective analysis of 305 consecutive singleton pregnancies who were followed for pregnancy outcome. The investigators found that using the 50th centile cut-off of the spiral artery pulsatility index values resulted in a sensitivity for the detection of pregnancy-induced hypertension (PIH) and IUGR of 52 and 50%, respectively, between 13 and 19 weeks gestational age and 56 and 57%, respectively, between 20 and 25 weeks. They concluded that measurement of utero-placental blood velocity waveforms in the second trimester is not sensitive enough to be used as a screening tool for PIH and IUGR in a low-risk population.
More recently, Deurloo et al. carried out spiral artery flow velocity measurements using colour Doppler between 11w0d to 13w6d, 14w0d to 17w6d and 18–24 weeks in 108 women with singleton gestations of whom four developed pre-eclampsia and 10 developed IUGR. In uncomplicated pregnancies, multilevel modelling showed that the pulsatality index, resistance index and systolic and diastolic ratio of the spiral arteries decreased progressively with increasing gestational age. No difference was found in any of the Doppler parameters in uncomplicated pregnancies compared with those complicated by pre-eclampsia and IUGR. This study was limited, however, by poor intra-observer variability.
Third trimester
Finally, two investigators evaluated colour and spectral Doppler of the spiral arteries in the third trimester of pregnancy. In the first study, spiral artery resistance index carried out between 28 and 41 weeks gestation correlated significantly with uterine artery resistance index and was superior to the latter in identifying 20 out of 43 women with PIH, IUGR, or both, who subsequently had adverse outcomes. In the second study, which included 85 women with normal pregnancies and 16 women with pregnancies complicated by IUGR between 27 and 38 weeks, the investigators found a decreased number of utero-placental vessels in the UPCS of pregnancies complicated by growth restriction, but pulsatility index of the detected vessels did not differ significantly from those in the normal pregnancies.
Analysis of the current data on direct spectral Doppler of the spiral arteries throughout pregnancy leads to the inevitable conclusion that, although changes in spiral artery blood flow are identifiable by Doppler, suboptimal reproducibility of the technique leads to poor results. This may relate to either subjective difficulties carrying out the measurements, suboptimal ultrasound technology to adequately image spiral arteries in some women, or more likely to a significant variance in flow between different spiral arteries sampled in the same woman. The latter hypothesis is supported sonographically by reports of decreased resistance index and pulsatility index in central compared with peripheral spiral arteries, and pathologically by a recent review explaining that the degree and extent of spiral artery remodelling in pre-eclampsia and IUGR varies according to the area of placental bed sampled.
Three-dimensional ultrasound modalities
Placental volume and placental quotient
With the introduction of three-dimensional ultrasound technology, an interest arose in using three-dimensional placental volume for potential screening. In a prospective study by Hafner et al., placental volume and placental quotient (placental quotient = placental volume/crown-rump length) in the first trimester correlated with uterine artery Doppler in the second trimester. Furthermore, using a placental quotient ≤ 10th centile and a uterine artery pulsatility index of ≥ 90th centile, the sensitivity for detection of IUGR was 25.0% and 20.2%; for women with pre-eclampsia without small gestational fetuses it was 20.0% and 28%, and for women with PIH and pre-eclampsia with small gestational fetuses it was 30.8% and 46.1%, respectively.
Again this method was limited by the use of an indirect approach, now using two indirect measurements of placental vascular development (i.e. placental size and upstream uterine artery Doppler effects). The limited sensitivity in detection of pre-eclampsia and IUGR was thus not surprising and the search for more direct evaluation methods continued.
Three-dimensional power Doppler angiography of the uteroplacental circulation space
The development of three-dimensional power Doppler (3DPD) angiography has allowed a more direct evaluation of utero-placental blood flow. Power Doppler has the benefit of displaying the magnitude of colour flow output rather than the Doppler frequency signal, thus providing increased sensitivity to low flows and velocities ( Fig. 2 ). When used with very low pulse repetition frequencies (between 0.6 and 0.9), and in combination with three-dimensional technology, which allows sampling of multiple spiral arteries in one sample, it allows complex evaluation of the extremely low-velocity flow that occurs in the UPCS ( Fig. 3 ).
