Mini-Review

Fetal Diagn Ther 2018;44:81–84 Received: March 27, 2018

Accepted after revision: May 14, 2018
DOI: 10.1159/000490047 Published online: July 12, 2018

Antenatal Care of Preeclampsia:

From the Inverted Pyramid to the
Arrow Model?
Tullio Ghi a Andrea Dall’Asta a Herbert Valensise b
     

a Department
of Medicine and Surgery, Obstetrics and Gynaecology Unit, University of Parma, Parma, Italy;
b Obstetrics
and Gynaecology Unit, University of Rome Tor Vergata, Rome, Italy

Keywords Much clinical interest has been raised in the last few
Hypertensive disorder of the pregnancy · Pyramid model · months by the publication of the ASPRE trial [1]. In this
Antenatal screening · Pregnancy care large multicentric randomised controlled trial, the first-
trimester screening of preterm preeclampsia based on
maternal factors (history and mean arterial pressure),
Abstract uterine Doppler findings, and biochemistry has proven to
The recent demonstration of the effectiveness of low-dose as- be valuable. In the screen-positive group, the administra-
pirin administered from the first trimester in the prevention of tion of 150 mg aspirin compared with the placebo has
preeclampsia will probably lead to establishing and radicating yielded a 62% reduction in the rate of preterm preeclamp-
the “inverted pyramid” screening model for preeclampsia. sia, although no differences in terms of maternal and
Such a multiparametric approach for the screening of pre- perinatal events have been demonstrated due to insuffi-
eclampsia in the first trimester, albeit highly sensitive in iden- cient power of the study. This extraordinary large trial
tifying early-onset disease, is poor at screening the forms of (almost 26,000 women enrolled) has confirmed that a
preeclampsia occurring close to term. Late-onset preeclamp- multiparametric assessment at 11–13 weeks is able to de-
sia is 3 to 6 times more common than early-onset preeclamp- tect, at the screen-positive rate of 10%, the vast majority
sia and currently represents the major determinant of mater- of women who will develop preeclampsia before 37 weeks,
nal morbidity related to hypertensive disorders of the preg- and virtually all those who will be affected before 34
nancy. On this ground, we discuss our idea to construct a weeks. The excellent performance of the screening com-
second “screening checkpoint” in the third trimester with the bining maternal factors, uterine Doppler, and placental
aim of reassessing the risk of preeclampsia of those women biochemistry in identifying those pregnancies destined to
who screened negative in the first trimester. If implemented, preterm preeclampsia had been previously reported by
the sequential screening model we propose would convert the same and other groups [2–6]. With the ASPRE study,
the “inverted pyramid model” into an “arrow model” for the for the first time, it was demonstrated that the early
antenatal care of preeclampsia. © 2018 S. Karger AG, Basel screening does make sense because the onset of the dis-

© 2018 S. Karger AG, Basel Prof. Tullio Ghi, MD, PhD

Obstetrics and Gynaecology Unit, University of Parma
Via Antonio Gramsci 14
E-Mail karger@karger.com
IT–43126 Parma (Italy)
www.karger.com/fdt E-Mail tullio.ghi @ unipr.it
 Color version available online
1st trimester

2nd trimester

Fig. 1. Shifting from the inverted pyramid

to the double arrow model for the antenatal 3rd trimester
care of preeclampsia.

ease can be prevented in positive cases through the ad- at identifying the forms of disease occurring close to
ministration of aspirin. In principle, albeit not yet proven, term. At the same screen-positive rate, the detection rate
this is expected to have a dramatically favourable impact of late preeclampsia is in fact below 50%, and this modest
on maternal and perinatal health as the occurrence of pre- performance of the screening has been also confirmed by
term preeclampsia is notoriously associated with high the ASPRE trial, which has identified only 40% of preg-
risk of perinatal and maternal morbidity and mortality nancies who have developed preeclampsia beyond 37
[7–10]. On this ground, the scientific community has weeks. This is biologically plausible since the pathophys-
welcomed this study as a major advance in pregnancy iology of the late forms of preeclampsia seems complete-
care and the wide implementation of the first-trimester ly different from the early forms as recently highlighted
screening of preeclampsia in clinical practice is endorsed by some authors [14, 15]. While in the preterm pre-
by many experts [11]. eclampsia, the abnormal placentation seems the trigger
Very recently, the performance of first-trimester factor of the disease and its effects are amenable to be
screening for preeclampsia combining maternal factors detected at 11–13 weeks, in the late-onset form, the pla-
with biomarkers was shown to be far superior to that of cental dysfunction seems to depend upon the villous
the model currently recommended by NICE guidelines overcrowding which takes place in the third trimester
and based on maternal characteristics and medical his- and does not give any clue at the time of the first-trimes-
tory [12]. ter screening [14]. Moreover, a different cardiovascular
This will probably lead to establishing and radicating profile has been shown at maternal echocardiography
for preeclampsia the “inverted pyramid” screening mod- among normotensive asymptomatic patients who will
el as originally envisaged in this journal by one of the au- develop preeclampsia at late gestation rather than prior
thors of the ASPRE trial [13]. This model is based on the to 34 weeks [15]. These findings further support the hy-
background concept that a massive and efficient screen- pothesis of a different pathophysiology for early and late
ing of the population early in gestation is able to select the preeclampsia.
exceedingly small number of high-risk women who will Therefore, we should bear in mind that the “inverted
be worthy to be followed up later in gestation. In other pyramid model” cannot pick up the vast majority of
words, regarding the antenatal care of preeclampsia, it women who will develop term preeclampsia. This is not
seems better to allocate more resources in the early stage a small problem as late-onset preeclampsia is 3 to 6 times
of pregnancy rather than close to term and this becomes more common compared with the early-onset type and
particularly true if an early intervention on the screen- available data have shown that, although outcomes for
positive cases as demonstrated by the ASPRE [1] can de- the mother and baby are more severe with preterm pre-
crease in fact the occurrence of preterm preeclampsia and eclampsia, the contribution of term preeclampsia to ma-
the number of pregnancies who will need special care and ternal morbidity is substantial [10]. Furthermore, in a re-
surveillance. cent series, the risk of emerging major complications
However, we would like to work out further the fasci- such as peripartum cardiomyopathy has been reported to
nating idea of the inverted pyramid model for the ante- be significantly more likely in women with late pre-
natal care of preeclampsia. As consistently reported eclampsia compared with early forms [16].
across the various studies cited above, the multiparamet- Because of that, if we want to optimise the antenatal
ric screening of preeclampsia in the first trimester is poor care of preeclampsia, the inverted pyramid model based

82 Fetal Diagn Ther 2018;44:81–84 Ghi/Dall’Asta/Valensise

DOI: 10.1159/000490047
on first-trimester screening does not seem to meet all the The work in progress concept is to transform the an-
requirements. Using this approach and giving aspirin to tenatal care of preeclampsia from an inverted pyramid
the screen-positive cases may certainly reduce the inci- model to a double arrow model (Fig. 1); in other words, a
dence of preterm preeclampsia and its related complica- sequential screening of preeclampsia where the primary
tions, but this has little effect on late preeclampsia and its assessment is repeated close to term in all women who
associated maternal risks. Indeed, according to the in- were screen negative in the first trimester with the aim of
verted pyramid model, most of the pregnancies who will identifying the vast majority of cases who will develop
develop term preeclampsia would be assigned to the ap- term preeclampsia. This approach would add back to the
parently low-risk group, which falls outside of the pyra- antenatal care a significant number of women who are at
mid. high risk of term preeclampsia and would be left out of
On this ground, some leading authors in this field have the vertex in a pyramid model. The clinical usefulness and
felt the need to remodulate the antenatal care of pre- the cost-effectiveness of this new type of screening will
eclampsia in order to pick up and possibly to prevent also hopefully be clarified by the ongoing study [19].
the cases occurring late in gestation who might be over- We are aware that this “arrow model” to date cannot
looked by the first-trimester screening. The emerging be recommended in clinical practice because the accuracy
idea is to construct a second checkpoint in the third tri- of the third-trimester screening test of preeclampsia
mester with the aim of reassessing the risk of preeclamp- should be validated on a large population and we still do
sia of those women who were labelled as screen negative not know if pravastatin or any other intervention can be
at first trimester. Some preliminary data have shown that offered to the screen-positive cases to prevent the onset
at 35 to 37 weeks, a multiparametric screening based on of late preeclampsia or its complications. Meanwhile, be-
maternal factors, uterine Doppler, and angiogenic mark- fore this ideal double arrow is built, it is better to keep our
ers may identify the vast majority (85%) of women who eyes open also outside of the pyramid.
will develop late preeclampsia [17]. Recent data from a Finally, in the attempt of optimizing the antenatal care
longitudinal proteomics study have demonstrated that el- of preeclampsia, beside the issue of screening and preven-
evated MMP-7 early in gestation (8–22 weeks) and low tion, the continuing need for appropriate recognition and
PlGF later in gestation (after 22 weeks) are good predic- management of preeclampsia should be emphasized. The
tors for the subsequent development of late-onset pre- fullPIERS model has been shown to improve the predic-
eclampsia, suggesting that the optimal identification of tion of adverse maternal outcomes once preeclampsia is
patients at risk may involve a two-step diagnostic process diagnosed and can favourably impact on patient care
[18]. On this basis, a new large RCT which will evaluate [20].
the role of pravastatin in preventing preeclampsia among
women found to be at high risk of preeclampsia during
Disclosure Statement
the third trimester is on its way and will contribute to
clarify if and how this new approach may affect the epi- All the authors state no financial disclosures nor conflict of in-
demiology of the late forms of the disease [19]. terest related to this work.

References
  1 Rolnik DL, Wright D, Poon LC, O’Gorman N,   3 Poon LC, Stratieva V, Piras S, et al: Hyperten-   6 Crovetto F, Figueras F, Triunfo S, Crispi F,
Syngelaki A, de Paco Matallana C, Akolekar sive disorders in pregnancy: combined Rodriguez-Sureda V, Dominguez C, Llurba E,
R, Cicero S, Janga D, Singh M, Molina FS, Per- screening by uterine artery Doppler, blood Gratacós E: First trimester screening for early
sico N, Jani JC, Plasencia W, Papaioannou G, pressure and serum PAPP-A at 11–13 weeks. and late preeclampsia based on maternal
Tenenbaum-Gavish K, Meiri H, Gizurarson Prenat Diagn 2010;30:216–223. characteristics, biophysical parameters, and
S, Maclagan K, Nicolaides KH: Aspirin versus   4 Scazzocchio E, Figueras F, Crispi F, et al: Per- angiogenic factors. Prenat Diagn 2015; 35:
placebo in pregnancies at high risk for pre- formance of a first-trimester screening of pre- 183–191.
term preeclampsia. N Engl J Med 2017; 377: eclampsia in a routine care low-risk setting.   7 van Esch JJA, van Heijst AF, de Haan AFJ, van
613–622. Am J Obstet Gynecol 2013;208:203.e1–e10. der Heijden OWH: Early-onset preeclampsia
  2 Poon LC, Kametas NA, Maiz N, et al: First-   5 Akolekar R, Syngelaki A, Poon L, Wright D, is associated with perinatal mortality and se-
trimester prediction of hypertensive disor- Nicolaides KH: Competing risks model in vere neonatal morbidity. J Matern Fetal Neo-
ders in pregnancy. Hypertension 2009; 53: early screening for preeclampsia by biophysi- natal Med 2017;30:2789–2794.
812–818. cal and biochemical markers. Fetal Diagn
Ther 2013;33:8–15.

Antenatal Screening for Preeclampsia Fetal Diagn Ther 2018;44:81–84 83

DOI: 10.1159/000490047
  8 Ni Y, Cheng W: Comparison of indications of 12 Tan MY, Wright D, Syngelaki A, Akolekar R, 18 Erez O, Romero R, Maymon E, Chaem-
pregnancy termination and prognosis of Cicero S, Janga D, Singh M, Greco E, Wright saithong P, Done B, Pacora P, Panaitescu B,
mothers and neonates in early- and late-onset A, Maclagan K, Poon LC, Nicolaides KH: Chaiworapongsa T, Hassan SS, Tarca AL: The
preeclampsia. Hypertens Pregnancy 2016;35: Comparison of diagnostic accuracy of early prediction of late-onset preeclampsia: results
315–322. screening for pre-eclampsia by NICE guide- from a longitudinal proteomics study. PLoS
 9 Lisonkova S, Joseph KS: Incidence of pre- lines and a method combining maternal fac- One 2017;12:e0181468.
eclampsia: risk factors and outcomes associ- tors and biomarkers: results of SPREE. Ultra- 19 Randomised controlled trial with pravastatin
ated with early- versus late-onset disease. Am sound Obstet Gynecol 2018;51:743–750. versus placebo for prevention of preeclampsia
J Obstet Gynecol 2013;209:544.e1–544.e12 13 Nicolaides KH: Turning the pyramid of pre- (STATIN trial). Trial registration EudraCT
10 Lisonkova S, Sabr Y, Mayer C, Young C, Skoll natal care. Fetal Diagn Ther 2011;29:183–196. Number: 2015-005130-22; ISRCTN number:
A, Joseph KS: Maternal morbidity associated 14 Redman CW, Staff AC: Preeclampsia, bio- ISRCTN17787139.
with early-onset and late-onset preeclampsia. markers, syncytiotrophoblast stress, and pla- 20 Von Dadelszen P, Payne B, Li J, Ansermino
Obstet Gynecol 2014;124:771–781. cental capacity. Am J Obstet Gynecol 2015; JM, Broughton Pipkin F, Côté AM, Douglas
11 Rolnik DL, O’Gorman N, Roberge S, Bujold 213(4 suppl):S9.e1, S9-11. MJ, Gruslin A, Hutcheon JA, Joseph KS, Kyle
E, Hyett J, Uzan S, Beaufils M, da Silva Costa 15 Valensise H, Vasapollo B, Gagliardi G, No- PM, Lee T, Loughna P, Menzies JM, Merialdi
F: Early screening and prevention of preterm velli GP: Early and late preeclampsia: two dif- M, Millman AL, Moore MP, Moutquin JM,
pre-eclampsia with aspirin: time for clinical ferent maternal hemodynamic states in the Ouellet AB, Smith GN, Walker JJ, Walley KR,
implementation. Ultrasound Obstet Gynecol latent phase of the disease. Hypertension Walters BN, Widmer M, Lee SK, Russell JA,
2017;50:551–556. 2008;52:873–880. Magee LA; PIERS Study Group: Prediction of
16 Arany Z, Elkayam U: Peripartum Cardiomy- adverse maternal outcomes in pre-eclampsia:
opathy. Circulation 2016;133:1397–1409. development and validation of the fullPIERS
17 Andrietti S, Silva M, Wright A, Wright D, model. Lancet 2011;377:219–227.
Nicolaides KH: Competing-risks model in
screening for pre-eclampsia by maternal fac-
tors and biomarkers at 35–37 weeks’ gestation.
Ultrasound Obstet Gynecol 2016;48:72–79.

84 Fetal Diagn Ther 2018;44:81–84 Ghi/Dall’Asta/Valensise

DOI: 10.1159/000490047