Ultrasound Obstet Gynecol 2020; 55: 50–57

Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/uog.21867.

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly cited.

Impact of new definitions of pre-eclampsia on incidence

and performance of first-trimester screening
N. KHAN1 , W. ANDRADE1 , H. DE CASTRO1 , A. WRIGHT2 , D. WRIGHT2 and K. H. NICOLAIDES1
1 Fetal Medicine Research Institute, King’s College Hospital, London, UK; 2 Institute of Health Research, University of Exeter, Exeter, UK

K E Y W O R D S: first trimester; pre-eclampsia; screening; small-for-gestational age

CONTRIBUTION Society for the Study of Hypertension in Pregnancy

What are the novel findings of this work? (ISSHP-old), which defines PE as the presence of both
The traditional definition of pre-eclampsia (PE), based hypertension and proteinuria. We reviewed the records
on the development of hypertension and proteinuria, has of pregnancies with GH, and those cases with high
been revised to include cases without proteinuria but with creatinine or liver enzymes or low platelet count were
evidence of renal, hepatic or hematological dysfunction. reclassified as having PE, according to the new criteria
The new definitions of PE resulted in, first, an increase of ISSHP (ISSHP-new) and the new criteria of the
in pregnancies classified as having PE but the additional American College of Obstetricians and Gynecologists
cases had milder disease, and, second, a non-significant (ACOG). The groups of PE according to the traditional
decrease in the performance of first-trimester screening and new criteria were compared for, first, gestational
for PE. age at delivery, birth-weight percentile and incidence
of a small-for-gestational-age (SGA) neonate with birth
What are the clinical implications of this work? weight < 10th percentile and perinatal death, and, second,
The new definitions of PE will inevitably be adopted the predictive performance for preterm PE of the
by professional organizations and will be incorporated competing-risks model based on the combination of
into routine clinical practice. In terms of first-trimester maternal risk factors, uterine artery pulsatility index,
assessment of risk for development of PE, the new mean arterial pressure and serum placental growth factor
definitions of PE would have a minimal impact on the at 11–13 weeks’ gestation (triple test).
performance of screening. Results According to ISSHP-old, 1870 (2.8%) cases
had PE, 2182 (3.3%) had GH and 62 912 (94.0%)
had no PE or GH. The incidence of PE according
ABSTRACT to ACOG was 3.0% (2029/66 964) and ISSHP-new
was 3.4% (2301/66 964). Median gestational age at
Objective The traditional definition of pre-eclampsia
delivery in the extra cases of PE according to ACOG
(PE) is based on the development of hypertension and
(difference, 1.3 weeks; 95% CI, 0.71–1.71 weeks) and in
proteinuria. This has been revised recently to include
the extra cases of PE according to ISSHP-new (difference,
cases without proteinuria but with evidence of renal,
1.5 weeks; 95% CI, 1.29–1.71 weeks) was higher than
hepatic or hematological dysfunction. The aim of this
in cases with PE according to ISSHP-old (38.4 weeks).
study was to examine the impact of new definitions of
The incidence of a SGA neonate in the extra cases of
PE on, first, the incidence and severity of the disease and,
PE according to ACOG (relative risk, 0.57; 95% CI,
second, the performance of the competing-risks model for
0.42–0.79) and in the extra cases of PE according to
first-trimester assessment of risk for PE.
ISSHP-new (relative risk, 0.52; 95% CI, 0.42–0.65) was
Methods This was a retrospective study of 66 964 lower than in the cases of PE according to ISSHP-old
singleton pregnancies that were classified as having (33.64%). In first-trimester screening for preterm PE by
PE, gestational hypertension (GH) or no PE or GH, the triple test, the detection rate, at a 10% false-positive
according to the traditional criteria of the International rate, was 75.9% (95% CI, 70.8–80.6%) for ISSHP-old,

Correspondence to: Prof. K. H. Nicolaides, Harris Birthright Research Centre for Fetal Medicine, Fetal Medicine Research Institute, King’s
College Hospital, 16–20 Windsor Walk, Denmark Hill, London SE5 8BB, UK (e-mail: kypros@fetalmedicine.com)
Accepted: 3 September 2019

© 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd ORIGINAL PAPER
on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.
Impact of new pre-eclampsia definition 51

74.3% (95% CI, 69.2–79.0%) for ACOG and 74.0% last 48 671 pregnancies, we also assessed the risk for
(95% CI, 68.9–78.6%) for ISSHP-new. subsequent PE through measurements of left and right
UtA-PI by transabdominal color Doppler ultrasound
Conclusions The new definitions of PE resulted in,
for calculation of the mean PI25 , MAP by validated
first, an increase in pregnancies classified as having
automated devices and standardized protocol26 and
PE but the additional cases had milder disease, and,
serum concentration of PlGF (DELFIA Xpress system,
second, a non-significant decrease in the performance of
PerkinElmer Life and Analytical Sciences, Waltham, USA
first-trimester screening for PE. © 2019 The Authors.
or BRAHMS KRYPTOR analyzer, Thermo Fisher Scien-
Ultrasound in Obstetrics & Gynecology published by
tific, Hennigsdorf, Germany). The women gave written
John Wiley & Sons Ltd on behalf of the International
informed consent to participate in studies for prediction
Society of Ultrasound in Obstetrics and Gynecology.
of pregnancy complications, which were approved by the
relevant research ethics committee in each participating
INTRODUCTION hospital.
Data on serum creatinine, transaminases, platelet count,
The traditional definition of pre-eclampsia (PE), 24-h urine protein and protein-to-creatinine ratio were
according to the International Society for the Study obtained from the computerized laboratory records of
of Hypertension in Pregnancy (ISSHP), is new onset the two participating hospitals, and data on dipstick
of hypertension (blood pressure ≥ 140 mmHg systolic protein testing and pregnancy outcome were obtained
or ≥ 90 mmHg diastolic) at ≥ 20 weeks’ gestation and from the computerized maternity records which were
proteinuria (≥ 300 mg/24 h or protein-to-creatinine ratio available throughout the period of 2011 to 2018.
> 30 mg/mmol or ≥ 2+ on dipstick testing)1 . This has The inclusion criteria for this study were singleton
been revised recently by the ISSHP2 and the American pregnancy undergoing first-trimester assessment and
College of Obstetricians and Gynecologists (ACOG)3 to subsequent delivery in our hospitals of a morphologically
include cases without proteinuria but with evidence of normal liveborn or stillborn neonate at > 24 weeks’
renal, hepatic or hematological dysfunction. gestation. We excluded pregnancies with aneuploidy and
In previous studies investigating the value of screening
major fetal abnormality and those ending in termination,
for PE by a combination of maternal demographic
miscarriage or fetal death before 24 weeks.
characteristics and medical history with biomarkers in
the first, second and third trimesters in singleton and twin
pregnancies, the outcome measure was PE as defined orig- Diagnosis of pre-eclampsia
inally by the ISSHP1,4–22 . In the first trimester, screening
has been achieved successfully by a combination of mater- Pregnancies were classified as having PE or gestational
nal demographic characteristics and medical history, hypertension (GH) according to the traditional criteria of
uterine artery pulsatility index (UtA-PI), mean arterial the ISSHP (ISSHP-old), which defines PE as the presence
pressure (MAP) and serum placental growth factor (PlGF) of both hypertension and proteinuria1 . We reviewed
through application of the competing-risks approach4–8 . the records of cases with GH, and those with high
In women identified by this approach as being at high risk creatinine or liver enzymes or low platelet count were
of preterm PE, with delivery at < 37 weeks’ gestation, reclassified as having PE according to the new criteria
administration of aspirin (150 mg/day) from 11–14 weeks of ISSHP (ISSHP-new)2 and the criteria of ACOG3 . The
until 36 weeks reduces the incidence of early PE by about criteria for PE in the absence of proteinuria according
90% and that of preterm PE by > 60%23 . to ISSHP-new include serum creatinine ≥ 90 μmol/L,
The objective of this study was to examine the impact alkaline or aspartate transaminase > 40 IU/L and platelet
of the new definitions of PE on, first, the incidence and count < 150 000/μL, neurological complications (includ-
severity of the disease, and, second, the performance of ing altered mental status, blindness, stroke, clonus, severe
the competing-risks model for first-trimester assessment headaches and persistent visual scotomata) and utero-
of risk for PE. placental dysfunction (such as fetal growth restriction,
abnormal umbilical artery Doppler waveform analysis or
stillbirth)2 . We did not provide the criteria for diagnosis
METHODS of fetal growth restriction and acknowledge that there is
Study population controversy as to whether this should be included in the
diagnostic criteria for PE in the absence of proteinuria or
This was a retrospective study of 66 964 singleton abnormal creatinine, transaminases or platelet count2 . For
pregnancies that delivered at King’s College Hospital, the purpose of this study, we did not include neurological
London or Medway Maritime Hospital, Gillingham, UK complications and uteroplacental dysfunction for diag-
between January 2011 and June 2018. All women had a nosis of PE according to ISSHP-new. The criteria for PE
routine examination at 11 + 0 to 13 + 6 weeks’ gestation, in the absence of proteinuria according to ACOG include
which included recording of maternal characteristics serum creatinine > 97 μmol/L, transaminases more than
and medical history4 , and ultrasound examination for twice the upper limit of normal (≥ 65 IU/L for our labora-
assessment of gestational age from fetal crown–rump tory) and platelet count < 100 000/μL, pulmonary edema
length24 and diagnosis of fetal abnormalities. In the or new-onset headache unresponsive to acetaminophen

© 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd Ultrasound Obstet Gynecol 2020; 55: 50–57.
on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.
52 Khan et al.

and not accounted for by alternative diagnoses or visual Pregnancy outcome in the cases of PE and GH
disturbances3 . For the purpose of this study, we included according to ISSHP-old and in the extra cases of PE
only quantitative measures of renal, hepatic or hema- according to ACOG and ISSHP-new is summarized
tological dysfunction for diagnosis of PE according to in Table 2 and illustrated in Figure 1. In PE according
ACOG3 . to ACOG or ISSHP-new (excluding PE according to
ISSHP-old), compared with PE according to ISSHP-old,
Statistical analysis there was higher median gestational age at delivery and
median birth-weight percentile and a lower incidence of
Data were summarized according to outcome (no PE or a SGA neonate with birth weight < 10th percentile, but
GH, GH or PE according to ISSHP-old, PE according to there was no significant difference in the incidence of
ACOG and PE according to ISSHP-new) as median and perinatal death.
interquartile range (IQR) for continuous variables and as In 48 671 pregnancies, MAP, UtA-PI and PlGF were
number and percentage for categorical variables. measured and these included 1171 with PE according
We compared median gestational age at delivery, to ISSHP-old, 1272 with PE according to ACOG, 1457
median birth-weight percentile27 , incidence of a small- with PE according to ISSHP-new, and 1472 with GH and
for-gestational-age (SGA) neonate with birth weight 46 028 without PE or GH according to ISSHP-old. The
< 10th percentile and incidence of perinatal death in demographic characteristics of pregnancies according
cases classified according to ISSHP-old as having PE, to the three definitions of PE were similar and in all
GH and no PE or GH, with those in the extra cases three UtA-PI and MAP were higher and PlGF was lower
of PE according to ACOG and in the extra cases of than in normal pregnancies (Table 3). The detection
PE according to ISSHP-new. Differences in median rates of all PE, preterm PE and early PE, at a 10%
gestational age at delivery and birth-weight percentile false-positive rate, and areas under the receiver–operating
were calculated between pregnancies with PE according characteristics curve in screening by maternal risk factors
to ISSHP-old and each other group; 95% CI for these and combination of maternal risk factors, MAP, UtA-PI
differences were calculated via bootstrapping. Relative and PlGF (triple test) at 11–13 weeks’ gestation, for
risks of SGA and perinatal death were calculated between each of the three definitions of PE, are shown in
pregnancies with PE according to ISSHP-old and each Table 4. There was a non-significant decrease in the
other group; 95% CI for the relative risks were produced. performance of screening when PE was defined according
In the subset of 48 671 pregnancies with measurements to the criteria of ACOG3 or ISSHP-new2 , compared to
of MAP, UtA-PI and PlGF, patient-specific risks of delivery ISSHP-old1 .
with PE at < 34, < 37 and < 41 + 3 weeks’ gestation were
calculated using the competing-risks model to combine
the prior distribution of the gestational age at delivery
DISCUSSION
with PE, obtained from maternal factors, with multiples
of the median (MoM) values of MAP, UtA-PI and Main findings
PlGF5,28–30 . The performance of screening for early PE,
preterm PE and all PE by maternal factors alone and This study has demonstrated that replacement of the
by the triple test was assessed. The area under the traditional definition of the ISSHP for PE, which requires
receiver–operating characteristics curve and detection the development of both hypertension and proteinuria1 ,
rate, at a 10% false-positive rate, for each of the three with new definitions that include cases without protein-
definitions of PE (ISSHP-old, ACOG and ISSHP-new) was uria but with evidence of renal, hepatic or hematological
examined. dysfunction2,3 , results in, first, an increase in the incidence
The statistical software package R, along with the of PE by about 21% (from 2.8% to 3.4%) in the case of the
library PropCIs, was used for data analyses31,32 . new criteria of the ISSHP2 and 7% (from 2.8% to 3.0%) in
the case of the criteria of ACOG3 , second, gestational age
at delivery, birth-weight percentile and incidence of a SGA
RESULTS
neonate in the additional cases of PE were similar to those
The maternal and pregnancy characteristics of the study in cases with GH and less severe than those in cases with
population of 66 964 pregnancies are summarized in PE as defined by the previous criteria of the ISSHP1 , third,
Table 1. According to ISSHP-old, 1870 (2.8%) cases had the incidence of perinatal death was low and there were
PE, 2182 (3.3%) had GH and 62 912 (94.0%) had no PE no significant differences between the groups of PE, and,
or GH. The incidence of PE according to ACOG was 3.0% fourth, the performance of first-trimester screening for PE
(2029/66 964) and ISSHP-new was 3.4% (2301/66 964). by maternal factors or a combination of maternal factors,
In the extra cases that were classified as having PE MAP, UtA-PI and PlGF, was similar when the condition
according to the new definitions, compared to those with was defined by ISSHP-old1 , ISSHP-new2 or ACOG3 , but
PE according to ISSHP-old, body mass index was lower there was a non-significant trend for lower detection
and there were higher incidences of no PE in a previous rate, at a 10% false-positive rate, for PE according to
pregnancy, high serum creatinine, high transaminases and ISSHP-new or ACOG compared with PE according to
low platelet count (Table 1). ISSHP-old.

© 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd Ultrasound Obstet Gynecol 2020; 55: 50–57.
on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.
Impact of new pre-eclampsia definition 53

Table 1 Maternal and pregnancy characteristics of study population of 66 964 singleton pregnancies, according to diagnosis

Extra cases of PE according

Diagnosis according to ISSHP-old to new criteria*
PE No PE or GH GH ACOG ISSHP-new
Variable (n = 1870) (n = 62 912) (n = 2182) (n = 159) (n = 431)

Maternal age (years) 31.0 (26.3–35.0) 31.3 (27.0–34.9) 32.0 (27.7–35.7) 33.3 (29.9–36.2) 32.9 (29.2–36.1)
Body mass index (kg/m2 ) 28.1 (24.3–33.4) 24.6 (22.0–28.5) 27.7 (24.1–32.4) 26.6 (23.4–31.9) 26.4 (23.6–30.8)
Gestational age (weeks) 12.7 (12.3–13.1) 12.8 (12.4–13.2) 12.7 (12.3–13.1) 12.7 (12.2–13.1) 12.7 (12.3–13.1)
Racial origin
White 1150 (61.50) 46 403 (73.76) 1517 (69.52) 110 (69.18) 287 (66.59)
Black 570 (30.48) 10 298 (16.37) 476 (21.81) 38 (23.90) 112 (25.99)
South Asian 92 (4.92) 2960 (4.70) 89 (4.08) 3 (1.89) 13 (3.02)
East Asian 22 (1.18) 1409 (2.24) 31 (1.42) 1 (0.63) 5 (1.16)
Mixed 36 (1.93) 1842 (2.93) 69 (3.16) 7 (4.40) 14 (3.25)
Chronic hypertension 196 (10.48) 666 (1.06) 0 (0) 0 (0) 0 (0)
Diabetes mellitus 47 (2.51) 698 (1.11) 32 (1.47) 1 (0.63) 4 (0.93)
SLE/APS 5 (0.27) 157 (0.25) 5 (0.23) 1 (0.63) 1 (0.23)
Smoker 105 (5.61) 5047 (8.02) 109 (5.00) 7 (4.40) 13 (3.02)
Family history of PE 134 (7.17) 2299 (3.65) 118 (5.41) 5 (3.14) 22 (5.10)
Method of conception
Natural 1765 (94.39) 60 589 (96.31) 2099 (96.20) 149 (93.71) 411 (95.36)
In-vitro fertilization 89 (4.76) 1807 (2.87) 69 (3.16) 10 (6.29) 18 (4.18)
Use of ovulation drugs 16 (0.86) 516 (0.82) 14 (0.64) 0 (0) 2 (0.46)
Parity
Nulliparous 1176 (62.89) 28 868 (45.89) 1310 (60.04) 104 (65.41) 259 (60.09)
Parous with no previous PE 471 (25.19) 32 922 (52.33) 690 (31.62) 47 (29.56) 138 (32.02)
Parous with previous PE 223 (11.93) 1122 (1.78) 182 (8.34) 8 (5.03) 34 (7.89)
Interpregnancy interval (years) 3.6 (2.1–6.5) 2.9 (1.9–4.8) 3.4 (2.2–5.8) 3.5 (2.6–5.7) 3.5 (2.2–6.1)
Serum creatinine
≥ 90 μmol/L 90 (4.8) — — 35 (22.0) 58 (13.5)
> 97 μmol/L 62 (3.3) — — 33 (20.8) 33 (7.7)
Serum transaminases
> 40 IU/L 260 (14.0) — — 121 (76.1) 218 (50.6)
≥ 65 IU/L 122 (6.6) — — 112 (70.4) 112 (26.0)
Platelet count
< 150 000/μL 260 (14.0) — — 43 (27.0) 213 (49.4)
< 100 000/μL 55 (3.0) — — 26 (16.4) 26 (6.0)

Data are given as median (interquartile range) or n (%). *Excluding cases of pre-eclampsia (PE) according to ISSHP-old. ACOG, American
College of Obstetricians and Gynecologists; APS, antiphospholipid syndrome; GH, gestational hypertension; ISSHP-old/-new, old/new
criteria of International Society for the Study of Hypertension in Pregnancy; SLE, systemic lupus erythematosus.

Comparison with results of previous studies proteinuria (defined as ≥ 2+ on dipstick testing) and 864
without hypertensive disorders. When the new criteria
A study from Australia compared pregnancy outcome of the ISSHP were applied to the 272 women with GH
in 958 women with PE according to ISSHP-old, 357 or chronic hypertension, 76 (27.9%) were reclassified as
with non-proteinuric PE, classified as hypertension with PE. The case–control nature of the study does not allow
renal, hepatic or hematological dysfunction, and 1192 conclusions to be drawn concerning the impact of the new
with GH33 . Proteinuric PE, compared to non-proteinuric definitions of PE on the incidence of the disease. However,
PE, was associated with a higher incidence of severe their finding that with ISSHP-new a high proportion
hypertension (39% vs 30%), preterm birth (39% vs 30%) of cases that would have been considered previously
and perinatal death (2.5% vs 0.6%); the respective rates in to have GH are reclassified as PE is consistent with
GH were 9%, 11% and 0.7% and the authors concluded our results.
that non-proteinuric PE has a poorer outcome for women A study from Japan examined 308 women diagnosed
and their fetus than does GH but is a more benign with hypertensive disorders of pregnancy at a tertiary
condition than is proteinuric PE33 . center and divided them, according to the new criteria
Kallela et al. evaluated the impact of new definitions of the ISSHP (including neurological complications and
of PE on the incidence of the disease using the data uteroplacental dysfunction), into three groups: PE with
of the Finnish Pre-eclampsia Consortium (FINNPEC) proteinuria (n = 218), PE without proteinuria (n = 45)
case–control cohort, which was recruited to study the and GH (n = 45); in 69% (31/45) of cases of PE without
genetic background of PE and fetal growth34 . The cohort proteinuria, there was uteroplacental dysfunction35 .
included 1379 women with PE according to the old ISSHP Applying the ISSHP-new criteria increased the number
criteria, 272 with GH or chronic hypertension without of pregnant women diagnosed as having PE by 15%;

© 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd Ultrasound Obstet Gynecol 2020; 55: 50–57.
on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.
54 Khan et al.

Table 2 Pregnancy outcome in 66 964 singleton pregnancies, according to diagnosis

Diagnosis according to ISSHP-old Extra cases of PE according to new criteria*

Outcome PE No PE or GH GH ACOG ISSHP-new

GA at birth (weeks)
Median 38.43 40.00 39.80 39.71 39.93
Difference Reference 1.57 (1.52–1.71) 1.37 (1.29–1.57) 1.29 (0.71–1.71) 1.50 (1.29–1.71)
Birth-weight percentile
Median 28.42 49.25 43.10 42.28 43.39
Difference Reference 20.83 (18.22–24.38) 14.68 (11.18–19.26) 13.86 (3.94–25.27) 14.97 (9.70–22.00)
SGA neonate
% 33.64 11.58 20.26 19.50 17.63
Relative risk Reference 0.34 (0.32–0.37) 0.60 (0.54–0.67) 0.57 (0.42–0.79) 0.52 (0.42–0.65)
Perinatal death
% 1.34 0.39 0.41 1.26 0.46
Relative risk Reference 0.29 (0.20–0.44) 0.31 (0.15–0.65) 0.94 (0.25–3.51) 0.34 (0.09–1.31)

Values in parentheses are 95% CI. *Excluding cases of pre-eclampsia (PE) according to ISSHP-old. ACOG, American College of
Obstetricians and Gynecologists; GA, gestational age; GH, gestational hypertension; ISSHP-old/-new, old/new criteria of International
Society for the Study of Hypertension in Pregnancy; SGA, small-for-gestational age.

PE (ISSHP-old)

No PE or GH (ISSHP-old)

GH (ISSHP-old)

PE (ACOG)

PE (ISSHP-new)

24 28 32 36 40 44 0.01 1 10 50 90 99 99.9
Gestational age at delivery (weeks) Birth-weight percentile

Figure 1 Distribution of gestational age at delivery and birth-weight percentile in pregnancies with pre-eclampsia (PE), those with no PE or
gestational hypertension (GH) and those with GH, according to old criteria of International Society for the Study of Hypertension in
Pregnancy (ISSHP-old), and in extra cases of PE according to new criteria of American College of Obstetricians and Gynecologists (ACOG)
and ISSHP (ISSHP-new).

the cases with and without proteinuria had a similar of fetal growth restriction in the definition of ISSHP-new,
incidence of maternal complications, composite neonatal which would have inevitably increased adverse maternal
complications, gestational age at delivery and birth weight and neonatal outcomes; in PE, the incidence of a
and these outcomes were worse than in the group with SGA neonate is inversely related to gestational age at
GH. The main difference from our study is the inclusion delivery36 .

© 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd Ultrasound Obstet Gynecol 2020; 55: 50–57.
on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.
Impact of new pre-eclampsia definition 55

Table 3 Characteristics of study population of 48 671 singleton pregnancies with available data on uterine artery pulsatility index (UtA-PI),
mean arterial pressure (MAP) and serum placental growth factor (PlGF), according to diagnosis

Diagnosis according to ISSHP-old* Diagnosis of PE according to new criteria

Variable No PE or GH (n = 46 028) PE (n = 1171) ACOG (n = 1272) ISSHP-new (n = 1457)

Maternal age (year) 31.4 (27.1–35.0) 31.3 (26.7–35.1) 31.4 (26.9–35.2) 31.7 (27.1–35.2)
Body mass index (kg/m2 ) 24.6 (22.0–28.4) 28.2 (24.1–33.6) 28.1 (24.0–33.4) 27.8 (23.9–33.0)
Gestational age (weeks) 89.1 (86.3–92.0) 88.9 (86.3–91.8) 88.9 (86.3–91.8) 88.9 (86.3–91.8)
Racial origin
White 33 381 (72.52) 710 (60.63) 784 (61.64) 900 (61.77)
Black 8044 (17.48) 368 (31.43) 390 (30.66) 444 (30.47)
South Asian 2182 (4.74) 55 (4.70) 56 (4.40) 64 (4.39)
East Asian 1072 (2.33) 15 (1.28) 16 (1.26) 18 (1.24)
Mixed 1349 (2.93) 23 (1.96) 26 (2.04) 31 (2.13)
Medical history
Chronic hypertension 502 (1.09) 146 (12.47) 146 (11.48) 146 (10.02)
Diabetes mellitus Type 1 172 (0.37) 15 (1.28) 15 (1.18) 15 (1.03)
Diabetes mellitus Type 2 306 (0.66) 21 (1.79) 21 (1.65) 21 (1.44)
SLE/APS 114 (0.25) 4 (0.34) 5 (0.39) 5 (0.34)
Smoker 3716 (8.07) 59 (5.04) 65 (5.11) 69 (4.74)
Family history of PE 1687 (3.67) 83 (7.09) 87 (6.84) 99 (6.79)
Method of conception
Natural 44 368 (96.39) 1111 (94.88) 1204 (94.65) 1385 (95.06)
In-vitro fertilization 1275 (2.77) 52 (4.44) 60 (4.72) 63 (4.32)
Use of ovulation drugs 385 (0.84) 8 (0.68) 8 (0.63) 9 (0.62)
Parity
Nulliparous 21 216 (46.09) 697 (59.52) 763 (59.98) 862 (59.16)
Parous with no previous PE 23 907 (51.94) 318 (27.16) 347 (27.28) 412 (28.28)
Parous with previous PE 905 (1.97) 156 (13.32) 162 (12.74) 183 (12.56)
Interpregnancy interval (years) 3.0 (1.9–4.9) 3.7 (2.1–6.5) 3.7 (2.1–6.4) 3.7 (2.1–6.5)
UtA-PI MoM 1.000 [0.998–1.002] 1.123 [1.104–1.143] 1.113 [1.094–1.131] 1.100 [1.084–1.118]
MAP MoM 1.000 [0.999–1.001] 1.046 [1.042–1.051] 1.046 [1.042–1.050] 1.047 [1.043–1.051]
PlGF MoM 1.000 [0.996–1.004] 0.785 [0.765–0.807] 0.796 [0.776–0.816] 0.810 [0.791–0.829]

Data are given as median (interquartile range), n (%) or mean [95% CI]. *1472 pregnancies were diagnosed with gestational hypertension
(GH) according to ISSHP-old. ACOG, American College of Obstetricians and Gynecologists; APS, antiphospholipid syndrome;
ISSHP-old/-new, old/new criteria of International Society for the Study of Hypertension in Pregnancy; MoM, multiples of the median;
PE, pre-eclampsia; SLE, systemic lupus erythematosus.

Table 4 Detection rate (DR), at 10% false-positive rate, and area under receiver–operating characteristics curve (AUC) for all pre-eclampsia
(PE), preterm PE and early PE, in screening by maternal risk factors and combination of maternal risk factors, mean arterial pressure, uterine
artery pulsatility index and placental growth factor (triple test) at 11–13 weeks’ gestation, for three definitions of PE

Definition of PE
ISSHP-old ACOG ISSHP-new
Method of DR (n/N AUC DR (n/N AUC DR (n/N AUC
screening (%; 95% CI)) (95% CI) (%; 95% CI)) (95% CI) (%; 95% CI)) (95% CI)

Maternal factors
All PE 459/1171 0.762 487/1272 0.7556 532/1457 0.7441
(39.2; 36.4–42.1) (0.7486–0.7754) (38.3; 35.6–41.0) (0.7424–0.7688) (36.5; 34.0–39.0) (0.7316–0.7567)
PE < 37 weeks 137/316 0.7811 142/327 0.7803 146/334 0.7791
(43.4; 37.8–49.0) (0.7567–0.8054) (43.4; 38.0–49.0) (0.7563–0.8043) (43.7; 38.3–49.2) (0.7551–0.8031)
PE < 34 weeks 57/129 0.7881 57/131 0.7852 59/134 0.7869
(44.2; 35.5–53.2) (0.7512–0.8251) (43.5; 34.9–52.4) (0.7485–0.8219) (44.0; 35.5–52.9) (0.7507–0.823)
Triple test
All PE 644/1171 0.8205 667/1272 0.8109 714/1457 0.7982
(55.0; 52.1–57.9) (0.8079–0.8332) (52.4; 49.7–55.2) (0.7984–0.8233) (49.0; 46.4–51.6) (0.7863–0.8101)
PE < 37 weeks 240/316 0.9132 243/327 0.9067 247/334 0.9039
(75.9; 70.8–80.6) (0.8989–0.9274) (74.3; 69.2–79.0) (0.892–0.9215) (74.0; 68.9–78.6) (0.8886–0.9192)
PE < 34 weeks 112/129 0.9441 112/131 0.9377 114/134 0.9372
(86.8; 79.7–92.1) (0.9283–0.9599) (85.5; 78.3–91.0) (0.9195– 0.9559) (85.1; 77.9–90.6) (0.9192–0.9552)

ACOG, American College of Obstetricians and Gynecologists; ISSHP-old/-new, old/new criteria of International Society for the Study of
Hypertension in Pregnancy.

© 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd Ultrasound Obstet Gynecol 2020; 55: 50–57.
on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.
56 Khan et al.

Implications for clinical practice REFERENCES

1. Brown MA, Lindheimer MD, de Swiet M, Van Assche A, Moutquin JM. The
The new definitions of PE will inevitably be adopted by classification and diagnosis of the hypertensive disorders of pregnancy: Statement
from the international society for the study of hypertension in pregnancy (ISSHP).
professional organizations and will be incorporated into Hypertens Pregnancy 2001; 20: IX–XIV.
routine clinical practice. We have shown that, in terms of 2. Brown MA, Magee LA, Kenny LC, Karumanchi SA, McCarthy FP, Saito S, Hall DR,
Warren CE, Adoyi G, Ishaku S; International Society for the Study of Hypertension
first-trimester assessment of risk for development of early, in Pregnancy (ISSHP). The hypertensive disorders of pregnancy: ISSHP classification,
preterm and all PE, the new definitions of PE would have diagnosis & management recommendations for international practice. Pregnancy
Hypertens 2018; 13: 291–310.
a minimal impact on the performance of screening. 3. ACOG Practice Bulletin No. 202: Gestational Hypertension and Preeclampsia. Obstet
Gynecol 2019; 133: e1–25.
4. Wright D, Syngelaki A, Akolekar R, Poon LC, Nicolaides KH. Competing risks
Strengths and limitations model in screening for preeclampsia by maternal characteristics and medical history.
Am J Obstet Gynecol 2015; 213: 62.e1–10.
5. O’Gorman N, Wright D, Syngelaki A, Akolekar R, Wright A, Poon LC, Nicolaides
The main strengths of the study are, first, examination of KH. Competing risks model in screening for preeclampsia by maternal factors
a large unselected population with prospective recording and biomarkers at 11–13 weeks’ gestation. Am J Obstet Gynecol 2016; 214:
103.e1–12.
of maternal demographic characteristics and medical 6. O’Gorman N, Wright D, Poon LC, Rolnik DL, Syngelaki A, Wright A, Akolekar R,
history, and, second, first-trimester measurements of Cicero S, Janga D, Jani J, Molina FS, de Paco Matallana C, Papantoniou N,
Persico N, Plasencia W, Singh M, Nicolaides KH. Accuracy of competing-risks model
MAP, UtA-PI and PlGF and application of the competing- in screening for pre-eclampsia by maternal factors and biomarkers at 11–13 weeks’
risks model for prediction of PE allowing assessment gestation. Ultrasound Obstet Gynecol 2017; 49: 751–755.
7. Tan MY, Wright D, Syngelaki A, Akolekar R, Cicero S, Janga D, Singh M, Greco E,
of the impact of the new definitions on prediction Wright A, Maclagan K, Poon LC, Nicolaides KH. Comparison of diagnostic accuracy
of PE. of early screening for pre-eclampsia by NICE guidelines and a method combining
maternal factors and biomarkers: results of SPREE. Ultrasound Obstet Gynecol
The main limitations of the study relate to the retro- 2018; 51: 743–750.
spective assessment of renal, hepatic and hematological 8. Wright D, Tan MY, O’Gorman N, Poon LC, Syngelaki A, Wright A, Nicolaides KH.
Predictive performance of the competing risk model in screening for preeclampsia.
dysfunction based on information recorded in hospital Am J Obstet Gynecol 2019; 220: 199.e1–13.
computerized systems and exclusion of neurological 9. Wright D, Gallo DM, Gil Pugliese S, Casanova C, Nicolaides KH. Contingent
screening for preterm pre-eclampsia. Ultrasound Obstet Gynecol 2016; 47:
complications from the new definitions of PE because 554–559.
these are often subjective and their recording would 10. Wright A, Wright D, Syngelaki A, Georgantis A, Nicolaides KH. Two-stage screening
for preterm preeclampsia at 11–13 weeks’ gestation. Am J Obstet Gynecol 2019;
not have been consistent. Accurate assessment of such 220: 197.e1–11.
symptoms will be achieved only in prospective studies 11. Gallo DM, Wright D, Casanova C, Campanero M, Nicolaides KH. Competing
risks model in screening for preeclampsia by maternal factors and biomarkers at
that incorporate these in the diagnosis of the disease. 19–24 weeks’ gestation. Am J Obstet Gynecol 2016; 214: 619.e1–17.
Another limitation is that, in the classification of cases 12. Litwinska M, Wright D, Efeturk T, Ceccacci I, Nicolaides KH. Pro-
posed clinical management of pregnancies after combined screening for
according to ISSHP-new, we did not include the definition pre-eclampsia at 19–24 weeks’ gestation. Ultrasound Obstet Gynecol 2017; 50:
of fetal growth restriction, because this is controversial 367–372.
13. Litwinska M, Syngelaki A, Wright A, Wright D, Nicolaides KH. Management of
and not clearly defined; for example, is the condition pregnancies after combined screening for pre-eclampsia at 19–24 weeks’ gestation.
based on ultrasonographic estimated fetal weight < 10th Ultrasound Obstet Gynecol 2018; 52: 365–372.
or < 5th or < 3rd percentile and is it in addition necessary 14. Tsiakkas A, Saiid Y, Wright A, Wright D, Nicolaides KH. Competing risks model
in screening for preeclampsia by maternal factors and biomarkers at 30–34 weeks’
to have abnormal Doppler indices in the uterine arteries gestation. Am J Obstet Gynecol 2016; 215: 87.e1–17.
or umbilical arteries or fetal middle cerebral arteries, 15. Wright D, Dragan I, Syngelaki A, Akolekar R, Nicolaides KH. Pro-
posed clinical management of pregnancies after combined screening for
or is the diagnosis made retrospectively based on low pre-eclampsia at 30–34 weeks’ gestation. Ultrasound Obstet Gynecol 2017; 49:
birth weight? Another limitation of the study is that 194–200.
16. Panaitescu AM, Wright D, Militelo A, Akolekar R, Nicolaides KH. Pro-
neonatal outcome was confined to gestational age at posed clinical management of pregnancies after combined screening for
delivery, birth weight and perinatal death and that we pre-eclampsia at 35–37 weeks’ gestation. Ultrasound Obstet Gynecol 2017; 50:
383–387.
did not provide information on maternal complications. 17. Panaitescu A, Ciobanu A, Syngelaki A, Wright A, Wright D, Nicolaides KH. Screening
As in the case of neurological symptoms, adequate for pre-eclampsia at 35–37 weeks’ gestation. Ultrasound Obstet Gynecol 2018; 52:
501–506.
examination of detailed maternal and neonatal compli- 18. Ciobanu A, Wright A, Panaitescu A, Syngelaki A, Wright D, Nicolaides KH.
cations requires prospective investigation of predefined Prediction of imminent preeclampsia at 35–37 weeks gestation. Am J Obstet Gynecol
2019; 220: 584.e1–11.
outcome measures. 19. Francisco C, Wright D, Benkő Z, Syngelaki A, Nicolaides KH. Hidden high rate
of pre-eclampsia in twin compared to singleton pregnancies. Ultrasound Obstet
Gynecol 2017; 50: 88–92.
20. Francisco C, Wright D, Benkő Z, Syngelaki A, Nicolaides KH. Competing-risks
Conclusions model in screening for pre-eclampsia in twin pregnancy by maternal characteristics
and medical history. Ultrasound Obstet Gynecol 2017; 50: 501–506.
The new definitions of PE result in, first, an increase in 21. Benkő Z, Chaveeva P, de Paco Matallana C, Zingler E, Wright A, Wright D,
Nicolaides KH. Validation of competing-risks model in screening for pre-eclampsia
pregnancies classified as having PE but the additional in twin pregnancy by maternal factors. Ultrasound Obstet Gynecol 2019; 53:
cases have milder disease, and, second, a non-significant 649–654.
22. Benkő Z, Chaveeva P, de Paco Matallana C, Zingler E, Wright D, Nicolaides KH.
decrease in the performance of first-trimester screening Revised competing-risks model in screening for pre-eclampsia in twin pregnancy by
for PE. maternal characteristics and medical history. Ultrasound Obstet Gynecol 2019; 54:
617–624.
23. Rolnik DL, Wright D, Poon LC, O’Gorman N, Syngelaki A, de Paco Matallana C,
Akolekar R, Cicero S, Janga D, Singh M, Molina FS, Persico N, Jani JC, Plasencia W,
ACKNOWLEDGMENT Papaioannou G, Tenenbaum-Gavish K, Meiri H, Gizurarson S, Maclagan K,
Nicolaides KH. Aspirin versus placebo in pregnancies at high risk for preterm
preeclampsia. N Engl J Med 2017; 377: 613–622.
This study was supported by a grant from The Fetal 24. Robinson HP, Fleming JE. A critical evaluation of sonar crown rump length
Medicine Foundation (Charity No: 1037116). measurements. Br J Obstet Gynaecol 1975; 82: 702–710.

© 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd Ultrasound Obstet Gynecol 2020; 55: 50–57.
on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.
Impact of new pre-eclampsia definition 57

25. Plasencia W, Maiz N, Bonino S, Kaihura C, Nicolaides KH. Uterine artery Doppler 31. R Development Core Team. R. A language and environment for statistical computing.
at 11 + 0 to 13 + 6 weeks in the prediction of pre-eclampsia. Ultrasound Obstet R Foundation for Statistical Computing, Vienna, Austria, 2011. http://www.R-
Gynecol 2007; 30: 742–749. project.org/.
26. Poon LC, Zymeri NA, Zamprakou A, Syngelaki A, Nicolaides KH. Protocol for 32. Ralph Scherer. PropCIs: Various Confidence Interval Methods for Proportions.
measurement of mean arterial pressure at 11–13 weeks’ gestation. Fetal Diagn Ther R package version. 0.3-0, 2018. https://CRAN.R-project.org/package=PropCIs.
2012; 31: 42–48. 33. Homer CS, Brown MA, Mangos G, Davis GK. Non-proteinuric pre-eclampsia: a
27. Nicolaides KH, Wright D, Syngelaki A, Wright A, Akolekar R. Fetal Medicine novel risk indicator in women with gestational hypertension. J Hypertens 2008; 26:
Foundation fetal and neonatal population weight charts. Ultrasound Obstet Gynecol 295–302.
2018; 52: 44–51. 34. Kallela J, Jääskeläinen T, Kortelainen E, Heinonen S, Kajantie E, Kere J, Kivinen K,
28. Wright A, Wright D, Ispas A, Poon LC, Nicolaides KH. Mean arterial pressure in the Pouta A, Laivuori H. The diagnosis of pre-eclampsia using two revised classifications
three trimesters of pregnancy: effects of maternal characteristics and medical history. in the Finnish Pre-eclampsia Consortium (FINNPEC) cohort. BMC Pregnancy
Ultrasound Obstet Gynecol 2015; 45: 698–706. Childbirth 2016; 1: 221.
29. Tayyar A, Guerra L, Wright A, Wright D, Nicolaides KH. Uterine artery pulsatility 35. Tochio A, Obata S, Saigusa Y, Shindo R, Miyagi E, Aoki S. Does pre-eclampsia
index in the three trimesters of pregnancy: effects of maternal characteristics and without proteinuria lead to different pregnancy outcomes than pre-eclampsia with
medical history. Ultrasound Obstet Gynecol 2015; 45: 689–697. proteinuria? J Obstet Gynaecol Res 2019; 45: 1578–1583.
30. Tsiakkas A, Duvdevani N, Wright A, Wright D, Nicolaides KH. Serum 36. Yu CK, Khouri O, Onwudiwe N, Spiliopoulos Y, Nicolaides KH; Fetal Medicine
placental growth factor in the three trimesters of pregnancy: effects of maternal Foundation Second-Trimester Screening Group. Prediction of pre-eclampsia by
characteristics and medical history. Ultrasound Obstet Gynecol 2015; 45: uterine artery Doppler imaging: relationship to gestational age at delivery and
591–598. small-for-gestational age. Ultrasound Obstet Gynecol 2008; 31: 310–313.

© 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd Ultrasound Obstet Gynecol 2020; 55: 50–57.
on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.