Ultrasound Obstet Gynecol 2017; 50: 492–495

Published online 24 August 2017 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/uog.18816

ASPRE trial: performance of screening for preterm

pre-eclampsia
D. L. ROLNIK1 , D. WRIGHT2 , L. C. Y. POON1,3 , A. SYNGELAKI1 , N. O’GORMAN1 ,
C. DE PACO MATALLANA4 , R. AKOLEKAR5 , S. CICERO6 , D. JANGA7 , M. SINGH8 ,
F. S. MOLINA9 , N. PERSICO10 , J. C. JANI11 , W. PLASENCIA12 , G. PAPAIOANNOU13 ,
K. TENENBAUM-GAVISH14 and K. H. NICOLAIDES1
1
King’s College Hospital, London, UK; 2 University of Exeter, Exeter, UK; 3 The Chinese University of Hong Kong, Shatin, Hong Kong;
4
Hospital Clı́nico Universitario Virgen de la Arrixaca, Murcia, Spain; 5 Medway Maritime Hospital, Gillingham, Kent, UK; 6 Homerton
University Hospital, London, UK; 7 North Middlesex University Hospital, London, UK; 8 Southend University Hospital, Essex, UK;
9
Hospital Universitario San Cecilio, Granada, Spain; 10 Ospedale Maggiore Policlinico, Milan, Italy; 11 University Hospital Brugmann,
Université Libre de Bruxelles, Brussels, Belgium; 12 Hospiten Group, Tenerife, Canary Islands, Spain; 13 Attikon University Hospital, Athens,
Greece; 14 Rabin Medical Center, Petach Tikva, Israel

K E Y W O R D S: ASPRE trial; first-trimester screening; mean arterial pressure; placental growth factor; pre-eclampsia;
pregnancy-associated plasma protein-A; pyramid of pregnancy care; uterine artery Doppler

ABSTRACT of term PE and 25 167 (97.6%) without PE. In combined

first-trimester screening for preterm PE with a risk cut-off
Objective To examine the performance of screening
of 1 in 100, the DR was 76.7% (138/180) for preterm PE
for preterm and term pre-eclampsia (PE) in the study
and 43.1% (194/450) for term PE, at screen-positive rate
population participating in the ASPRE (Combined Mul-
of 10.5% (2707/25 797) and FPR of 9.2% (2375/25 797).
timarker Screening and Randomized Patient Treatment
with Aspirin for Evidence-Based Preeclampsia Preven- Conclusion The performance of screening in the ASPRE
tion) trial. study was comparable with that of a study of
approximately 60 000 singleton pregnancies used for
Methods This was a prospective first-trimester multi-
development of the algorithm; in that study, combined
center study on screening for preterm PE in 26 941
screening detected 76.6% of cases of preterm PE and
singleton pregnancies by means of an algorithm that com-
38.3% of term PE at a FPR of 10%. Copyright © 2017
bines maternal factors, mean arterial pressure, uterine
ISUOG. Published by John Wiley & Sons Ltd.
artery pulsatility index and maternal serum pregnancy-
associated plasma protein-A and placental growth factor
at 11–13 weeks’ gestation. Eligible women with an esti- INTRODUCTION
mated risk for preterm PE of > 1 in 100 were invited to
participate in a double-blind trial of aspirin (150 mg per The ASPRE (Combined Multimarker Screening and
day) vs placebo from 11–14 until 36 weeks’ gestation, Randomized Patient Treatment with Aspirin for
which showed that, in the aspirin group, the incidence Evidence-Based Preeclampsia Prevention) trial was a
of preterm PE was reduced by 62%. In the screened prospective first-trimester multicenter study on screen-
population, the detection rates (DRs) and false-positive ing for preterm PE in 26 941 singleton pregnancies by
rates (FPRs) for delivery with PE < 37 and ≥ 37 weeks means of an algorithm that combines maternal factors,
were estimated after adjustment for the effect of aspirin in mean arterial pressure (MAP), uterine artery pulsatility
those receiving this treatment. We excluded 1144 (4.2%) index (UtA-PI), and maternal serum pregnancy-associated
pregnancies because of loss to follow-up or study with- plasma protein-A (PAPP-A) and placental growth factor
drawal (n = 716), miscarriage (n = 243) or termination (PlGF) at 11–13 weeks’ gestation1 . The algorithm was
(n = 185). developed from a study of approximately 60 000 singleton
pregnancies; in that study, combined screening detected
Results The study population of 25 797 pregnancies 76.6% of cases of preterm PE and 38.3% of term PE at a
included 180 (0.7%) cases of preterm PE, 450 (1.7%) false-positive rate (FPR) of 10%2 .

Correspondence to: Prof. K. H. Nicolaides, Fetal Medicine Research Institute, King’s College Hospital, 16–20 Windsor Walk, Denmark Hill,
London SE5 8BB, UK (e-mail: [email protected])
Accepted: 19 July 2017

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. ORIGINAL PAPER
ASPRE trial screening performance 493

In the ASPRE study, eligible women with an estimated in Pregnancy12 . PE was defined as systolic blood
risk for preterm PE of > 1 in 100 were invited to pressure ≥ 140 mmHg and/or diastolic blood pres-
participate in a double-blind trial of aspirin (150 mg sure ≥ 90 mmHg on at least two occasions 4 h apart
per day) vs placebo from 11–14 weeks until 36 weeks’ developing after 20 weeks of gestation in previously
gestation1 . In the aspirin group, the incidence of preterm normotensive women. Hypertension was defined as pro-
PE was reduced by 62%. teinuria ≥ 300 mg in 24 h or two readings of at least
The objective of this study was to report the accuracy of ++ on dipstick analysis of midstream or catheter urine
the previously reported first-trimester model of screening specimens if no 24-h collection was available. PE superim-
for PE2 in the screened population of the ASPRE study. posed on chronic hypertension was defined as significant
The hypothesis was that the performance of screening proteinuria (as defined above) developing after 20 weeks
would be similar to that estimated from the original of gestation in women with known chronic hypertension
model. (history of hypertension before conception or presence of
hypertension at booking visit before 20 weeks’ gestation
METHODS in the absence of trophoblastic disease).
Data on pregnancy outcome were collected from the
Study design and participants hospital maternity records of the women. The obstetric
records of all women with pre-existing or pregnancy-
This was a prospective, multicenter study of singleton
associated hypertension were examined to determine if
pregnancies at 11 + 0 to 13 + 6 weeks’ gestation in women
the condition was PE.
attending routine pregnancy care at one of 13 maternity
hospitals in the UK, Spain, Italy, Belgium, Greece and
Israel1 . Approval for the trial was obtained from the rel- Statistical analysis
evant research ethics committee and competent authority The previously described algorithm was used for the
in each country in which the trial was conducted. calculation of patient-specific risk of delivery with
The eligibility criteria were maternal age ≥ 18 years, no PE < 37 weeks’ gestation2 . Eligible women with an
serious mental illness or learning difficulty and singleton estimated risk for preterm PE of > 1 in 100 were invited
pregnancy with live fetus with no major abnormality to participate in a double-blind trial of aspirin (150 mg
demonstrated on the 11–13-week scan. We excluded per day) vs placebo from 11–14 weeks until 36 weeks’
pregnancies with no follow-up and those ending in gestation1 , which showed that, in the aspirin group, the
termination or miscarriage. incidence of preterm PE was reduced by 62%. In the
The Standards for Reporting Diagnostic Accuracy screened population, the FPRs and detection rates (DRs)
Studies (STARD)3 were adhered to. for delivery with PE < 37 and ≥ 37 weeks were estimated
after adjustment for the effect of aspirin in those receiving
Test methods this treatment.
The index test, or the test for which accuracy has
RESULTS
been evaluated, was the previously reported algorithm
for first-trimester assessment of risk for PE by maternal Participants
factors, MAP, UtA-PI, PAPP-A and PlGF2 . Maternal fac-
tors were recorded4 , MAP was measured by validated A total of 26 941 women with singleton pregnancy
automated devices and standardized protocol5 , transab- underwent screening for PE (Figure 1). For the purpose of
dominal color Doppler ultrasound was used to measure this study, we excluded 1144 (4.2%) pregnancies because
the left and right UtA-PI and the average value was of loss to follow-up (n = 716), miscarriage (n = 243)
recorded6 , serum PAPP-A and PlGF concentrations were or termination (n = 185). The group lost to follow-up
measured by an automated device (PAPP-A and PlGF included 152 high-risk pregnancies that participated in
1-2-3™ kits, DELFIA® Xpress random access platform; the trial but subsequently withdrew consent, of which
PerkinElmer Inc. Wallac Oy, Turku, Finland). All oper- 78 allowed reporting of their screening data; the baseline
ators undertaking the Doppler studies had received the characteristics of the women who withdrew consent were
appropriate Certificate of Competence from The Fetal similar between those assigned to receive aspirin and those
Medicine Foundation. Measured values of MAP, UtA-PI, assigned to receive placebo1 .
PAPP-A and PlGF were expressed as a MoM, adjusting The characteristics of the study population of 25 797
for those characteristics found to provide a substantive pregnancies are shown in Table 1. In this population, the
contribution to the log10 transformed value including the risk for preterm PE was > 1 in 100 in 2707 (10.5%)
maternal factors in the prior model7–10 . and ≤ 1 in 100 in 23 090 (89.5%). In the group with a
The index test was carried out prospectively in con- risk of > 1 in 100, 806 participated in the trial and were
secutive singleton pregnancies at 11 + 0 to 13 + 6 weeks’ assigned to receive placebo, 785 participated in the trial
gestation; gestational age was determined from measure- and were assigned to receive aspirin and 1116 did not
ment of fetal crown–rump length11 . participate in the trial, either because they did not want
The target condition was PE, as defined by the to do so (n = 806) or they did not fulfill the eligibility
International Society for the Study of Hypertension criteria (n = 310) due to hypersensitivity to aspirin, peptic

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2017; 50: 492–495.
494 Rolnik et al.

Screened for preterm PE

(n = 26 941)

Excluded (n = 1144)
• Miscarriage (n = 243)
• TOP (n = 185)
• Withdrew consent (n = 152)
• Lost to follow-up (n = 564)

Study population (n = 25 797)

Risk for preterm PE ≤ 1 in 100 Risk for preterm PE > 1 in 100

(n = 23 090) (n = 2707)

• Preterm PE (n = 42) Assigned to placebo (n = 806)

• Term PE (n = 256) • Preterm PE (n = 35)
• No PE (n = 22 792) • Term PE (n = 59)
• No PE (n = 712)

Assigned to aspirin (n = 785)

• Preterm PE (n = 34*)
• Term PE (n = 53)
• No PE (n = 698)

Did not participate in trial (n = 1116)

• Preterm PE (n = 69)
• Term PE (n = 82)
• No PE (n = 965)

Figure 1 Flowchart summarizing screening for preterm pre-eclampsia (PE), interventions and follow-up in 26 941 singleton pregnancies.
*Adjusted number, derived from 13 observed cases and assuming 62% reduction of preterm PE caused by aspirin. TOP, termination of
pregnancy.

Table 1 Characteristics of study population ulceration or bleeding disorder, treatment with aspirin
within 28 days before screening or participation in another
Characteristic Study population (n = 25 797) drug trial within 28 days before screening.
Maternal age (years) 31.7 (27.7–35.2)
Maternal weight (kg) 66.0 (58.7–76.5)
Test results
Maternal height (cm) 164 (160–169)
Body mass index (kg/m2 ) 24.4 (21.8–28.2)
The incidence of preterm and term PE in the
Gestational age (weeks) 12.7 (12.3–13.1)
Racial origin screen-positive and screen-negative groups is shown in
Caucasian 20 383 (79.0) Figure 1. In the group assigned to receive aspirin, there
Afro-Caribbean 3117 (12.1) were 13 cases of preterm PE and 53 cases of term PE. The
East Asian 517 (2.0) ASPRE trial demonstrated that administration of aspirin,
South Asian 1194 (4.6)
compared with placebo, resulted in a 62% reduction in
Mixed 586 (2.3)
Medical history the incidence of preterm PE but had no significant effect
Chronic hypertension 319 (1.2) on the incidence of term PE. Consequently, the observed
Diabetes mellitus 207 (0.8) number of 13 cases of preterm PE in the aspirin group was
APS/SLE 135 (0.5) adjusted to the expected number of 34 had these patients
Cigarette smoker 2072 (8.0)
not received aspirin (Figure 1).
Family history of pre-eclampsia 851 (3.3)
Mode of conception The study population of 25 797 pregnancies included
Spontaneous 24 868 (96.4) 180 (0.7%) cases of preterm PE, 450 (1.7%) of term
In-vitro fertilization 764 (3.0) PE and 25 167 (97.6%) without PE. In combined
Ovulation drugs 165 (0.6) first-trimester screening for preterm PE with a risk cut-off
Parity
of 1 in 100, the DR was 76.7% (138/180) for preterm PE
Nulliparous 12 181 (47.2)
Parous and 43.1% (194/450) for term PE, at a screen-positive rate
No previous pre-eclampsia 13 097 (50.8) of 10.5% (2707/25 797) and FPR of 9.2% (2375/25 797).
Previous pre-eclampsia 519 (2.0)
No previous SGA 12 767 (49.5)
Previous SGA 849 (3.3) DISCUSSION
Interpregnancy interval (years) 2.8 (1.6–4.8)
Main findings
Data are given as median (interquartile range) or n (%). APS, anti-
phospholipid syndrome; SGA, small-for-gestational-age neonate; This prospective multicenter study demonstrates the
SLE, systemic lupus erythematosus. feasibility of incorporating first-trimester screening for PE

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2017; 50: 492–495.
ASPRE trial screening performance 495

into routine clinical practice. The performance of screen- at 11–13 weeks’ gestation is by far superior to those of
ing for PE at 11–13 weeks by a combination of maternal NICE and ACOG in identifying the group who would
factors and biomarkers is similar to that estimated from benefit from prophylactic use of aspirin.
the original model2 . The estimated DR of screening by
maternal factors, MAP, UtA-PI, PAPP-A and PlGF was
77% for PE < 37 weeks and 43% for PE ≥ 37 weeks at a ACKNOWLEDGMENTS
FPR of 9.2%; the rates in the dataset used for development The study was supported by grants from The Fetal
of the model were 77%, 38% and 10%, respectively2 . Medicine Foundation (Charity No: 1037116) and by
the European Union 7th Framework Programme –
Study limitations FP7-HEALTH-2013-INNOVATION-2 (ASPRE Project
# 601852).
There were two components to the ASPRE study; first,
routine screening of all pregnancies meeting the eligibility
criteria and second, participation of a high proportion REFERENCES
of the screen-positive group in a trial of aspirin vs 1. Rolnik DL, Wright D, Poon LC, O’Gorman N, Syngelaki A, de Paco Matallana C,
placebo1 . The trial demonstrated a beneficial effect of Akolekar R, Cicero S, Janga D, Singh M, Molina FS, Persico N, Jani JC, Plasencia
W, Papaioannou G, Tenenbaum-Gavish K, Meiri H, Gizurarson S, Maclagan K,
aspirin in reducing the rate of preterm PE and therefore Nicolaides KH. Aspirin versus placebo in pregnancies at high risk for preterm
the observed number of cases with preterm PE in the preeclampsia. N Engl J Med 2017; 377: 613–622.
2. Akolekar R, Syngelaki A, Poon L, Wright D, Nicolaides KH. Competing risks model
aspirin group had to be adjusted to take into account in early screening for preeclampsia by biophysical and biochemical markers. Fetal
this beneficial effect. In this respect, this was not a Diagn Ther 2013; 33: 8–15.
3. Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig L, LijmerJG
non-intervention validation study. Moher D, Rennie D, de Vet HCW, Kressel HY, Rifai N, Golub RM, Altman
DG, Hooft L, Korevaar DA, Cohen JF, For the STARD Group. STARD 2015: An
updated list of essential items for reporting diagnostic accuracy studies. BMJ 2015;
Implications for practice 351: h5527.
4. Wright D, Syngelaki A, Akolekar R, Poon LC, Nicolaides KH. Competing risks
model in screening for preeclampsia by maternal characteristics and medical history.
The ASPRE trial demonstrated that, in women with Am J Obstet Gynecol 2015; 213: 62.e1–10.
singleton pregnancy who were identified by means of 5. Poon LC, Zymeri NA, Zamprakou A, Syngelaki A, Nicolaides KH. Protocol for
measurement of mean arterial pressure at 11-13 weeks’ gestation. Fetal Diagn Ther
first-trimester combined screening as being at high risk 2012; 31: 42–48.
for preterm PE, the administration of aspirin at a dose 6. Plasencia W, Maiz N, Bonino S, Kaihura C, Nicolaides KH. Uterine artery Doppler at
11+0 to 13+6 weeks in the prediction of pre-eclampsia. Ultrasound Obstet Gynecol
of 150 mg per day from 11–14 weeks until 36 weeks’ 2007; 30: 742–749.
gestation reduces the incidence of preterm PE by > 60%1 . 7. Wright A, Wright D, Ispas A, Poon LC, Nicolaides KH. Mean arterial pressure in the
three trimesters of pregnancy: effects of maternal characteristics and medical history.
The traditional approach of identifying women at Ultrasound Obstet Gynecol 2015; 45: 698–706.
high risk of PE who could potentially benefit from 8. Tayyar A, Guerra L, Wright A, Wright D, Nicolaides KH. Uterine artery pulsatility
index in the three trimesters of pregnancy: effects of maternal characteristics and
the prophylactic use of aspirin is based on maternal medical history. Ultrasound Obstet Gynecol 2015; 45: 689–697.
characteristics and medical history. In the UK, the 9. Wright D, Silva M, Papadopoulos S, Wright A, Nicolaides KH. Serum pregnancy
associated plasma protein-A in the three trimesters of pregnancy: effects of
National Institute for Health and Care Excellence (NICE) maternal characteristics and medical history. Ultrasound Obstet Gynecol 2015; 46:
recommends the identification of the high-risk group on 42–50.
10. Tsiakkas A, Duvdevani N, Wright A, Wright D, Nicolaides KH. Serum placental
the basis of 10 factors, including maternal characteristics growth factor in the three trimesters of pregnancy: effects of maternal characteristics
and features of the medical and obstetric histories13 . and medical history. Ultrasound Obstet Gynecol 2015; 45: 591–598.
11. Robinson HP, Fleming JE. A critical evaluation of sonar crown rump length
However, the performance of such screening is poor, measurements. Br J Obstet Gynaecol 1975; 82: 702–710.
with a DR of preterm PE of 39% at a FPR of 10%14 . 12. Brown MA, Lindheimer MD, de Swiet M, Van Assche A, Moutquin JM. The
classification and diagnosis of the hypertensive disorders of pregnancy: Statement
In the USA, the American College of Obstetricians and from the international society for the study of hypertension in pregnancy (ISSHP).
Gynecologists (ACOG) recommends the use of aspirin Hypertens Pregnancy 2001; 20: IX–XIV.
13. National Collaborating Centre for Women’s and Children’s Health (UK).
in women with a history of PE in more than one Hypertension in pregnancy: the management of hypertensive disorders during
pregnancy or a history of PE that resulted in delivery pregnancy. London: RCOG Press, 2010.
14. O’ Gorman N, Wright D, Poon LC, Rolnik DL, Syngelaki A, de Alvarado M1,
before 34 weeks’ gestation15 . However, this subgroup Carbone IF, Dutemeyer V, Fiolna M, Frick A, Karagiotis N, Mastrodima S,
constitutes only approximately 0.3% of all pregnancies de Paco Matallana C, Papaioannou G, Pazos A, Plasencia W, Nicolaides KH.
Multicenter screening for pre-eclampsia by maternal factors and biomarkers at 11–13
and includes only 5% of women who develop preterm weeks’ gestation: comparison with NICE guidelines and ACOG recommendations.
PE14 . Our approach to screening with the use of Bayes’ Ultrasound Obstet Gynecol 2017; 49: 756–760.
15. Hypertension in pregnancy: report of the American College of Obstetricians and
theorem to combine the a-priori risk from maternal factors Gynecologists’ Task Force on Hypertension in Pregnancy. Obstet Gynecol 2013;
with biophysical and biochemical measurements obtained 122: 1122–31.

This article has been selected for Journal Club.

A slide presentation, prepared by Dr Fiona Brownfoot,
one of UOG's Editors for Trainees, is available online.
Chinese translation by Dr Yang Fang. Spanish translation by Dr Ruben Dario Fernandez.

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2017; 50: 492–495.
Ultrasound Obstet Gynecol 2017; 50: 492–495
Published online 24 August 2017 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/uog.18816

Ensayo ASPRE: el comportamiento del cribado de preeclampsia pret érmino

RESUMEN
Objetivo Estudiar el comportamiento del cribado de pre-eclampsia pretérmino y a término (PE) en la población de
estudio que participa en el ensayo ASPRE (Cribado combinado basado en evidencia mediante múltiples marcadores y
tratamiento aleatorizado de la paciente con aspirina para la prevención de preeclampsia).
Métodos Se trata de un estudio multicéntrico prospectivo de primer trimestre sobre el cribado de PE pretérmino en
26 941 embarazos con feto único, mediante un algoritmo que combina factores maternos como la presión arterial
promedio, el ı́ndice de pulsatilidad de la arteria uterina y la proteı́na plasmática A del suero materno asociada al
embarazo y el factor de crecimiento de la placenta a las 11-13 semanas de gestación. Se invitó a las mujeres con
posibilidades de ser elegidas por tener un riesgo estimado de PE pretérmino >1 entre 100 a participar en un ensayo
doble ciego de aspirina (150 mg por dı́a) versus un placebo, desde las semanas 11-14 a las 36 semanas de gestación, que
resultó en una reducción de la incidencia de PE prematura de un 62% en el grupo que tomó aspirina. En la población
en la que se hizo el cribado, después del ajuste del efecto de la aspirina en las mujeres que recibieron este tratamiento, se
estimaron las tasas de detección (TD) y las tasas de falsos positivos (TFP) para el parto con PE <37 y >37 semanas. Se
excluyeron 1144 (4,2%) embarazos debido a falta de seguimiento o abandono del estudio (n = 716), aborto (n = 243)
o terminación (n = 185).
Resultados La población estudiada de 25 797 embarazos incluyó 180 (0,7%) casos de PE pretérmino, 450 (1,7%) de
PE a término y 25 167 (97,6%) sin PE. En el cribado combinado del primer trimestre para PE pretérmino con un lı́mite
de riesgo de 1 entre 100, la TD fue del 76,7% (138/180) para PE pretérmino y 43,1% (194/450) para PE a término,
con una tasa positiva del cribado del 10,5% (2707/25797) y una TFP del 9,2% (2375/25797).
Conclusión El comportamiento del cribado en el estudio ASPRE fue comparable con el de un estudio
de aproximadamente 60 000 embarazos con feto único utilizados para el desarrollo del algoritmo; en ese estudio,
el cribado combinado detectó el 76,6% de los casos de PE pretérmino y el 38,3% de los de PE a término con una TFP
del 10%.

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. ORIGINAL PAPER