Ultrasound Obstet Gynecol 2014; 43: 34–40

Published online 8 December 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/uog.12537

First-trimester screening for early and late

small-for-gestational-age neonates using maternal serum
biochemistry, blood pressure and uterine artery Doppler
F. CROVETTO*†, F. CRISPI*, E. SCAZZOCCHIO‡, I. MERCADE*, E. MELER*‡, F. FIGUERAS*
and E. GRATACOS*
*Maternal–Fetal Medicine Department, ICGON, Hospital Clı́nic, Universitat de Barcelona; Fetal and Perinatal Medicine Research Group,
Institut d’Investigacions Biomediques August Pi i Sunyer (IDIBAPS); and Centro de Investigación Biomédica en Red de Enfermedades Raras
(CIBERER), Barcelona, Spain; †Fondazione Ca’ Granda, Ospedale Maggiore Policlinico, Dipartimento Ostetricia e Ginecologia,
Univerisità degli Studi di Milano, Milan, Italy; ‡Obstetrics, Gynecology and Reproductive Medicine Department, Institut Universitari
Dexeus, Barcelona, Spain

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

ABSTRACT
Objective To assess the effectiveness of first-trimester
screening for early and late small-for-gestational-age
(SGA) neonates using maternal serum biochemistry, blood
pressure and uterine artery Doppler.
Methods This was a prospective study of 4970 women
with a singleton pregnancy who underwent routine first-
trimester screening between 2009 and 2011. A logistic
regression-based predictive model for SGA, defined as
birth weight < 10th percentile, divided into early- or late-
onset based on gestational age at delivery before or
after 34 weeks’ gestation, was constructed. The model
included maternal baseline characteristics: serum levels of
pregnancy-associated plasma protein-A and free β-human
chorionic gonadotropin at 8–12 weeks and blood pressure
and uterine artery Doppler at 11 + 0 to 13 + 6 weeks.
Results The prevalence of early and late SGA was 0.6%
and 7.9%, respectively. Association with pre-eclampsia
was 67% and 8%, respectively. At a false-positive rate of
15%, the detection rate for early SGA was 73%; however
it differed substantially for cases with and without pre-
eclampsia (90% vs 40%). For late SGA, at false-positive
rates of 15 and 50%, detection rates were 32% and 70%,
respectively, and did not substantially differ between cases
with and without pre-eclampsia.
Conclusions First-trimester screening predicts early SGA
mainly because of its strong association with pre-
eclampsia. Although prediction of late SGA was poorer,
at a high false-positive rate it might be considered as
part of a first-trimester strategy to select women requiring
ultrasound assessment in the third trimester. Copyright
 2013 ISUOG. Published by John Wiley & Sons Ltd.
INTRODUCTION
Small-for-gestational-age (SGA) newborns are at
increased risk of mortality and morbidity in the perinatal
period1,2 . The condition is also associated with perma-
nent neurodevelopmental, metabolic and cardiovascular
changes, conferring an increased risk of morbidity and
mortality that persists into adulthood3 – 5 . Antenatal
identification of SGA is of obvious importance for
optimizing the timing of delivery and reducing perinatal
risks6 . Detection of SGA is challenging, and the detection
rates achieved in routine care settings are normally low7 .
Early identification of risk by means of first-trimester
screening may offer an opportunity to select patients for
targeted follow-up, and allow higher detection rates8 .
Previous studies have assessed the performance of
first-trimester screening for SGA by combining different
variables, such as maternal history and biochemical
and biophysical markers9 – 23 . Nevertheless, the reported
detection rate varies widely, from 12 to 73%, which
may partially reflect some heterogeneity in the definition
of SGA used in previous studies9,20 . The existence of
differences in the detection of early-onset vs late-onset
forms of SGA has only rarely been considered20 . However,
there is a growing consensus that early- and late-onset
forms of SGA must be dealt with as separate conditions,
from both clinical and research perspectives24 . Early-onset
SGA is always a severe disease, highly associated with
pre-eclampsia and abnormal placental implantation25 .
Late-onset SGA is rarely associated with pre-eclampsia
but, because it is far more prevalent than the early-onset
clinical form, it is responsible for a large proportion of
perinatal deaths26 . Furthermore, the association of late-
onset SGA with poorer outcome is now well known3,27 ,
consequently the importance of its early detection is

Correspondence to: Prof. E. Gratacos, Department of Maternal-Fetal Medicine, Hospital Clinic, University of Barcelona, Sabino de Arana 1,
08028, Barcelona, Spain (e-mail: egratacos@clinic.ub.es)
Accepted: 29 May 2013

Copyright  2013 ISUOG. Published by John Wiley & Sons Ltd. ORIGINAL PAPER
First-trimester SGA screening
increasingly being recognized6 . Recent studies have found
the detection rate by first-trimester screening to be much
lower for SGA fetuses at term than for those delivering
preterm20,28 .
The aim of this prospective study was to assess the
effectiveness of first-trimester screening for the detection
of early- vs late-onset SGA, based on a combination of
maternal history, maternal mean arterial pressure (MAP),
uterine artery (UtA) Doppler ultrasound and biochemical
markers, assessed under normal care conditions in a
general population of pregnant women.
METHODS
This was a prospective study of women with a singleton
pregnancy who underwent routine first-trimester screen-
ing for aneuploidies in the Department of Maternal–Fetal
Medicine at Hospital Clinic, Barcelona. The local ethics
committee approved the study protocol and each patient
provided written informed consent. Gestational age in
all pregnancies was calculated based on the crown–rump
length (CRL) at first-trimester ultrasound29 . Maternal
characteristics and medical history were recorded and
MAP, UtA Doppler, plasma concentrations of pregnancy-
associated plasma protein-A (PAPP-A) and free β-human
chorionic gonadotropin (β-hCG) were measured in the
first trimester.
Between May 2009 and October 2011 a total of
6283 women were assessed. Of these, 830 (13.2%) were
excluded because of missing outcome data, 236 (3.8%)
because of miscarriage or fetal death before 24 weeks’ ges-
tation, 217 (3.5%) because of fetal abnormalities and 30
(0.5%) because of termination of pregnancy in the absence
of medical indication, leaving 4970 cases for analysis.
The women’s characteristics, including demographic
details and obstetric and medical history, were prospec-
tively recorded at the time of first-trimester screening
(8 + 0 to 13 + 6 weeks’ gestation) and entered in our
database. The following variables were registered:
maternal age at first-trimester screening, ethnicity, parity,
height, weight, smoking status, medical history and
obstetric history. Data on pregnancy and neonatal
outcomes were obtained from the delivery database of
our hospital. Doctors who managed the cases clinically
were blinded to the study parameters obtained during the
first trimester.
Blood samples were drawn from each woman between
8 + 0 and 11 + 6 weeks’ gestation. PAPP-A and free β-hCG
were measured using the AutoDELFIA Xpress Analyzer
(Perkin-Elmer Life Sciences, Turku, Finland). These values
were converted to multiples of the median (MoM), which
were corrected for CRL, maternal age, ethnicity, body
mass index (BMI), smoking and diabetes status, according
to local references and as reported elsewhere30 .
In each patient, transvaginal UtA Doppler assessment
was performed at the time of the first-trimester scan, at
11 + 0 to 13 + 6 weeks, in order to measure the pulsatility
index (PI) in the left and right UtAs, as previously
described31 . Blood pressure (BP) was automatically
Copyright  2013 ISUOG. Published by John Wiley & Sons Ltd.
35
recorded with a calibrated OMRON M6 Comfort
machine (OMRON Corporation, Kyoto, Japan) in our
outpatient clinics. MAP was measured in one arm (right
or left) while the women were seated and after a 5-min
rest. MAP was calculated as: diastolic BP + (systolic
BP − diastolic BP)/3.
SGA was defined as birth weight less than the 10th
centile adjusted by gender and gestational age at delivery
according to local standards32 . The SGA group was then
classified into early-onset SGA (requiring delivery before
34 + 0 weeks) and late-onset SGA (requiring delivery at
or after 34 weeks) groups. All pregnancies complicated
by SGA were assessed with serial ultrasound examina-
tions including Doppler evaluation at regular intervals,
depending on the severity of the disease. Indications for
delivery before 34 weeks were abnormalities in ductus
venosus blood flow, reversed flow in the umbilical artery
or a decelerative/silent pattern in the fetal heart-rate trac-
ing. The control group comprised those neonates with
birth weight appropriate for gestational age, defined as
birth weight ≥ the 10th centile adjusted by gender and
gestational age at delivery according to local standards32 .
Intrauterine growth restriction (IUGR) was defined as
an estimated fetal weight (EFW) below the 10th cen-
tile according to local standards with abnormalities in
Doppler evaluation32 . Pre-eclampsia was defined accord-
ing to the guidelines of the International Society for the
Study of Hypertension in Pregnancy. This requires two
recordings of systolic BP of ≥ 140 mmHg or diastolic
BP ≥ 90 mmHg at least 4 h apart in a previously normoten-
sive woman after 20 weeks of gestation, and proteinuria
of 300 mg or more in 24 h33 . Pre-eclampsia was defined
as early if requiring delivery before 34 weeks, and late if
delivery took place at or after 34 weeks.
Statistical analysis
The Mann–Whitney U-test and Pearson’s χ2 test were
performed to make univariate comparisons of quanti-
tative and qualitative variables, respectively, between
groups. A post-hoc Bonferroni correction was conducted
to maintain a type-I error of 0.05 (P = 0.025). A binomial
distribution model was used to determine the 95% CIs of
proportions.
Logarithmic transformation was performed to normal-
ize mean MAP and UtA-PI. Expected log10 values were
calculated for all cases using linear regression analysis
of unaffected cases and included the following covari-
ables: maternal age at first-trimester ultrasound (years),
ethnicity (white European, Black, South American and
other), maternal height (cm) and weight (kg) at examina-
tion, parity (nulliparous vs parous), smoking status upon
examination (0, 1–9, 10–20 and > 20 cigarettes/day)
and CRL at first-trimester ultrasound (mm). Individual
observed values of PAPP-A and free β-hCG were expressed
as MoM of the expected value.
Logistic regression analysis was used to estimate each
woman’s a priori risk with respect to the following
covariates: maternal age, ethnicity, parity, BMI (kg/m2 ),
Ultrasound Obstet Gynecol 2014; 43: 34–40.
36 Crovetto et al.

smoking (cigarettes/days), previous medical history of Table 1 Baseline characteristics of 4970 women with singleton
chronic hypertension, diabetes mellitus, renal or autoim- pregnancy who underwent routine first-trimester screening between
2009 and 2011, according to study group
mune diseases, congenital and acquired thrombophilic
conditions and previous obstetric history of SGA, Controls Early SGA Late SGA
pre-eclampsia or stillbirth. Characteristic (n = 4549) (n = 30) (n = 391)
Logistic regression analysis was performed to estimate
the individual risks for early and late SGA with Age (years) 32 32 33.6
(28–35.2) (30–36) (29.3–36)*
respect to the following covariates: a priori risk (log- BMI (kg/m2 ) 23.9 24.1 23.8
transformed), log10 (MoM-mean-UtA-PI), log10 (MoM- (22.6–24.7) (22.9–27.4) (22.1–24.7)
MAP), log10 (MoM-PAPP-A) and log10 (MoM-free-β- Ethnicity
hCG). Receiver–operating characteristics (ROC) curves Caucasian 3329 (73.2) 19 (63.3) 321 (82.1)*
were constructed to analyze model performance, which Black 21 (0.5) 2 (6.7)†‡ 1 (0.3)
was expressed as detection rate for different cut-offs of South American 760 (16.7) 5 (16.7) 41 (10.5)*
Other 439 (9.7) 4 (13.3) 28 (7.2)
false-positive rate. Smoking status
In all regression models, stepwise forward algorithms Non-smoker 4230 (93.0) 27 (90.0) 321 (82.1)*
were constructed to select variables at a P-value cut-off of 1–9 cigarettes/day 89 (2.0) 0 20 (5.1)*
0.05. Goodness-of-fit models were assessed by calculating 10–20 cigarettes/day 193 (4.2) 2 (6.7) 40 (10.2)*
Nagelkerke’s R2 . The statistical package IBM SPSS 19.0 > 20 cigarettes/day 37 (0.8) 1 (3.3) 10 (2.6)*
(IBM Corp., Armok, NY, USA) was used to conduct all Medical history
Chronic hypertension 47 (1.0) 4 (13.3)†‡ 8 (2.0)
the statistical analyses and graphs were generated with
Diabetes mellitus 91 (2.0) 0 (0.0) 1 (0.3)*
MedCalc software (MedCalc, Mariakerke, Belgium). Renal disease 6 (0.1) 2 (6.7)†‡ 1 (0.3)
Autoimmune disease 56 (1.2) 1 (3.3) 11 (2.8)*
RESULTS Coagulation disorder 39 (0.9) 0 4 (1.0)
Obstetric history
Among the 4970 pregnancies included in the study, 421 Nulliparous 2664 (58.6) 16 (53.3) 267 (68.3)*
(8.5%) neonates were SGA, including 30 (0.6%) cases Previous SGA 24 (0.5) 2 (6.7)†‡ 3 (0.8)
of early SGA and 391 (7.9%) cases of late SGA. The Previous pre-eclampsia 29 (0.6) 4 (13.3)†‡ 7 (1.8)
Previous stillbirth 54 (1.2) 3 (10.0)†‡ 6 (1.5)
baseline characteristics of the normal pregnancy group
and the early- and late-SGA groups are summarized in Data given as n (%) or median (interquartile range). Statistically
Table 1. Potential predictive variables assessed at the time significant difference between: *controls and late SGA; †controls
of first-trimester evaluation, i.e. MAP, UtA-PI, PAPP-A and early SGA; ‡early and late SGA. BMI, body mass index; SGA,
and free β-hCG, are summarized in Table 2. Perinatal small-for-gestational age.
outcome according to study group is shown in Table 3.
Significant contributions to the prediction of early SGA Log10 (MoM-mean-UtA-PI) was significantly greater in
were provided by Black ethnicity, chronic hypertension, both early- and late-onset SGA than in unaffected
renal disease, previous SGA/pre-eclampsia, MAP and UtA gestations. Log10 (MoM-MAP) was significantly higher
Doppler. Significant contributions to the prediction of in the early-SGA group than in the unaffected group, but
late SGA were provided by maternal age, smoking status, it did not show any significant difference in the late-SGA
parity, diabetes, autoimmune disease and UtA Doppler. group. Log10 (MoM-PAPP-A) and log10 (free-β-hCG) were
The following models best fit the a priori risk for early similar among SGA cases and unaffected pregnancies.
and late SGA (a priori risk = ey /(1 + ey )): The following models best fit the patient-specific a
posteriori risk (a posteriori risk = ey /(1 + ey )):
Early SGA: Y = −5.399 + (2.222 if Black ethnic-
ity) + (1.803 if chronic hypertension) + (2.572 if
Early SGA: Y = −33.788 + (2.390 × a priori early
renal disease) + (1.716 if previous obstetrical his-
risk) + (9.480 × log10 (MoM-mean-UtA-PI)) +
tory); R2 = 8.8%.
(33.412 × log10 (MoM-MAP)); R2 = 21.2%.
Late SGA: Y = −3.872 + (0.047 × maternal age)
− (0.556 if nulliparous) + (0.509 if smoker) Late SGA: Y = −0.256 + (1.000 × a priori late risk)
− (2.169 if diabetes mellitus) + (0.858 if autoim- + (0.262 × log10 (MoM-mean-UtA-PI)); R2 = 4.6%.
mune disease); R2 = 4.3%.
Table 4 provides the values of the logistic regression
The following equation best fits the log10 (MoM-UtA-PI):
models. Figure 1 shows the ROC curves for detection
Y = 0.668 − (0.0011 × maternal age) − (0.0027 × of early-onset SGA (area under the curve (AUC), 0.848
CRL) − (0.0015 × height); R2 = 4.6%. (95% CI, 0.768–0.928)) and late-onset SGA (AUC, 0.652
(95% CI, 0.623–0.680)). The detection rates for early and
The following equation best fits the log10 (MoM-MAP): late SGA for different false-positive rates are presented in
Table 5.
Y = 1.801+ (0.001× maternal age) + (0.001× weight); In the early-onset SGA group 20 women (66.7%)
R2 = 15.0%. were affected by pre-eclampsia and in the late-onset

Copyright  2013 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2014; 43: 34–40.
First-trimester SGA screening 37

Table 2 Potential predictive variables in 4970 women with singleton pregnancy who underwent routine first-trimester screening, according
to study group

Characteristic Controls (n = 4549) Early SGA (n = 30) Late SGA (n = 391)

MAP (mmHg) 78.5 (74.1–83.2) 83.0 (78.0–88.1)*† 78.1 (73.4–83.2)

Mean uterine artery PI 1.67 (1.50–1.74) 2.04 (1.68–2.60)*† 1.69 (1.59–1.90)‡
Maternal serum biochemistry
PAPP-A (MoM) 1.13 (0.99–1.20) 1.15 (0.78–1.40) 1.13 (0.67–1.29)
free β-hCG (MoM) 0.99 (0.90–1.02) 0.99 (0.64–1.16) 0.99 (0.67–1.08)

Data given as median (interquartile range). Statistically significant difference between: *controls and early SGA; †early and late SGA;
‡controls and late SGA. β-hCG, β-human chorionic gonadotropin; MAP, mean arterial pressure; MoM, multiples of the median; PAPP-A,
pregnancy-associated plasma protein-A; PI, pulsatility index; SGA, small-for-gestational age.

Table 3 Maternal and neonatal outcomes in 4970 singleton pregnancies which underwent routine first-trimester screening, according to
study group

Characteristic Controls (n = 4549) Early SGA (n = 30) Late SGA (n = 391)

Gestational hypertension 39 (0.9) 1 (3.3) 8 (2.0)

Pre-eclampsia 82 (1.8) 20 (66.7)*† 32 (8.2)‡
Early 4 (0.1) 20 (66.7)* —
Late 78 (1.7) — 32 (8.2)‡
Gestational diabetes mellitus 226 (5.0) 3 (10.0) 13 (3.3)
Placental abruption 16 (0.4) 2 (6.7)*† 3 (0.8)
Preterm delivery 264 (5.8) 30 (100.0)*† 35 (9.0)‡
Chorioamnionitis 15 (0.3) 1 (3.3)* 1 (0.3)
Intrauterine growth restriction 0 20 (66.7)*† 25 (6.4)‡
Gestational age at delivery (weeks) 40 (38.8–40.7) 32 (30–32)*† 39.5 (38.1–40.7)‡
Birth weight (g) 3320 (3060–3630) 1245 (840–1540)*† 2560 (2330–2740)‡
Birth-weight percentile 49.8 (30.7–71.8) 1.4 (0.13–5.7)*† 6.0 (3.2–8.2)‡
Umbilical artery pH 7.25 (7.20–7.29) 7.18 (7.11–7.26)*† 7.24 (7.19–7.28)‡
Perinatal death 14 (0.3) 2 (6.7)*† 0

Data given as n (%) or median (interquartile range). Statistically significant difference between: *controls and early SGA; †early and
late SGA; ‡controls and late SGA. SGA, small-for-gestational age.

Table 4 Logistic regression model values SGA. It provides evidence that both clinical forms of SGA
can be predicted using a combination of maternal history
Early SGA (n = 30) Late SGA (n = 391)
and biophysical markers assessed in the first trimester of
Characteristic P SE P SE pregnancy. As expected, the statistical models applying
Log10 (a priori risk) < 0.001 0.379 < 0.001 0.274 to the two conditions differed in terms of the variables
Log10 (MoM-mean-UtA-PI)< 0.001 1.629 0.020 0.260 included and the accuracy achieved for each clinical form.
Log10 (MoM-MAP) 0.001 9.640 These results add to the body of evidence suggesting that
MAP, mean arterial pressure; MoM, multiples of the median;
early and late SGA should be considered as two different
PI, pulsatility index; SE, standard error; SGA, small-for-gestational disease entities, with different pathogenetic mechanisms
age; UtA, uterine artery. and clinical characteristics24 . In this study we used a cut-
off of gestational age at birth of 34 weeks to define early
SGA group 32 women (8.2%) developed pre-eclampsia. and late-onset SGA. While we acknowledge that there is
Table 6 illustrates the detection rate for early and late no universally accepted definition of early- and late-onset
SGA with or without pre-eclampsia for different false- SGA, recent studies having used arbitrarily gestational-age
positive rates. A recently developed screening model for cut-offs ranging from 34 to 37 weeks20,24,35 , we decided
pre-eclampsia was applied to the same population for on 34 weeks, as it has been widely used for the distinction
the prediction of SGA34 . While the performance for early between early- and late-onset forms of pre-eclampsia33 .
SGA (AUC, 0.839 (95% CI, 0.756–0.921)) was similar In relation to SGA, this cut-off has been shown to
to that obtained with our SGA model, the performance differentiate well between the main features of early-onset
for late SGA (AUC, 0.551 (95% CI, 0.529–0.581)) was forms of SGA, including the association with abnormal
poorer than that provided by our model. umbilical artery Doppler and pre-eclampsia, which are
not present in late-onset SGA36 .
Concerning early-onset SGA, a combination of mater-
DISCUSSION
nal baseline characteristics, MAP and UtA Doppler
This study evaluated the differential predictive perfor- achieved detection rates ranging from 50 to 73.3% for
mance of first-trimester screening for early- vs late-onset false-positive rates of 5 and 15%, respectively. These

Copyright  2013 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2014; 43: 34–40.
38 Crovetto et al.

(a) 100 Table 5 Diagnostic performance of logistic regression-based

predictive model for early and late small-for-gestational-age (SGA):
detection rate (DR) and positive (LR+) and negative (LR–)
likelihood ratios for fixed false-positive rates (FPR)
80
DR (95% CI) (%) LR + (%) LR– (%)

Early SGA (n = 30)

Sensitivity (%)

60 FPR = 5% 50.0 (32.1–67.9) 10.0 1.9

FPR = 10% 60.0 (42.5–77.5) 6.0 2.2
FPR = 15% 73.3 (57.5–89.1) 4.9 3.2
Late SGA (n = 391)
40 FPR = 5% 14.1 (10.7–17.5) 2.8 1.1
FPR = 10% 23.0 (18.8–27.2) 2.3 1.2
FPR = 15% 32.0 (27.4–36.6) 2.1 1.3
FPR = 40% 62.4 (57.6–67.2) 1.0 1.0
20
FPR = 50% 70.3 (65.8–74.8) 1.4 1.7
FPR = 60% 80.3 (76.3–84.2) 2.0 3.0

0
0 20 40 60 80 100
Table 6 Diagnostic performance of logistic regression-based
1 – Specificity (%)
predictive model for early and late small-for-gestational-age (SGA)
with or without pre-eclampsia: detection rate (DR) and positive
(b) 100 (LR+) and negative (LR–) likelihood ratios for fixed false-positive
rates (FPR)

DR (95% CI) LR + LR−

80
(%) (%) (%)

Early SGA (n = 30)

Pre-eclampsia (n = 20 (66.7%))
Sensitivity (%)

60
FPR = 5% 70.0 (49.9–90.1) 14.0 3.2
FPR = 10% 75.0 (56.0–94.0) 7.5 3.6
FPR = 15% 90.0 (76.9–100.0) 6.0 8.5
40 No pre-eclampsia (n = 10 (33.3%))
FPR = 5% 10.0 (0.0–28.6) 2.0 1.1
FPR = 10% 30.0 (11.4–48.6) 3.0 1.3
FPR = 15% 40.0 (9.65–70.4) 2.7 1.4
20
Late SGA (n = 391)
Pre-eclampsia (n = 32 (8.2%))
FPR = 5% 18.8 (5.3–32.3) 3.8 1.2
0 FPR = 10% 31.3 (15.2–47.4) 3.1 1.3
0 20 40 60 80 100 FPR = 15% 34.3 (17.9–50.7) 2.3 1.3
1 – Specificity (%) No pre-eclampsia (n = 359 (91.8%))
FPR = 5% 13.6 (12.3–14.9) 2.7 1.1
FPR = 10% 22.3 (18.0–26.6) 2.2 1.2
Figure 1 Receiver–operating characteristics curves for the
FPR = 15% 31.8 (27.0–36.6) 2.1 1.2
detection of early-onset (a) and late-onset (b) small-for-
gestational-age fetuses using maternal characteristics, Doppler and
biochemical parameters.
for a false-positive rate of 10%, an estimated detection
rate of 73%. The better performance reported by this
figures are in general higher than those reported in the lit- group could be explained by the inclusion of angiogenic
erature, and reflect the more specific definition we used for factors in their predictive algorithms.
early-onset SGA9,20,28 . However, prediction of early-onset Regarding the prediction of late SGA, our study showed
SGA depended heavily on the presence of pre-eclampsia. a markedly lower predictive value as compared with early-
The prediction performance for isolated early-onset SGA onset disease. Karagiannis et al.20 reported a detection rate
was much lower. A few studies have evaluated the predic- of 46% for a false-positive rate of 10% for SGA (defined
tion of SGA without pre-eclampsia. Direct comparisons as EFW < 5th centile) delivering after 37 weeks’ gestation.
are not possible because SGA in these studies was defined In spite of this, from a public health perspective, the ability
as an EFW below the 5th centile, in contrast to the 10th to select patients at higher risk for late-onset SGA, even
centile used in this study. Papastefanou et al.23 reported if with a high false-positive rate, could be of interest for
a detection rate of 55% for a false-positive rate of 20%, institutions not offering a systematic third-trimester scan.
but SGA was considered as a single category and was Souka et al.37 reported that a contingent strategy using a
defined as an EFW below the 5th centile. In the only study first-trimester prediction model to select patients requiring
considering preterm SGA, defined as fetuses requiring a third-trimester scan would be as good at predicting SGA
delivery before 37 weeks, Karagiannis et al.20 reported, as a universal third-trimester scan strategy. In line with this

Copyright  2013 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2014; 43: 34–40.
First-trimester SGA screening
contention, our findings suggest that first-trimester screen-
ing could select the half of the population that would need
a third-trimester scan to achieve a 70% detection rate for
late-onset SGA. This figure compares favorably with the
reported detection rates of about 50% for routine third-
trimester ultrasonography in low-risk pregnancies6,37 .
In this study we evaluated common biomarkers used
for trisomy 21 screening, PAPP-A and free β-hCG. In
general, data regarding the association of SGA with these
biomarkers is not consistent, some studies showing a
correlation19,20 and others not12,14 . In addition, studies
on the prediction of pre-eclampsia have shown that the
potential added value of PAPP-A measurement when
used in combination with UtA Doppler is virtually
negligible14,18,34 .
One strength of this study is the use of two different
models to predict each of the two clinical forms of
SGA. In the only previous study reporting early and
late forms of SGA separately, Karagiannis et al.20
distinguished between preterm and term SGA, defined
as those delivering before or after 37 weeks’ gestation,
respectively. In this study we distinguished between
early- and late-onset SGA at a cut-off of 34 weeks,
as this has been more widely used in recent studies.
Another important factor is the definition used for
SGA. A substantial proportion of studies used the 5th
centile to define this condition. This results in a higher
sensitivity for any predictive study. However, strong
evidence supports the idea that fetuses and newborns
in the 5–9th centiles also have a substantially increased
risk of abnormal perinatal and fetal cardiovascular
programming38 . Consequently, while compromising the
detection rate of any screening method, using the 10th
centile is probably more relevant for clinical practice and is
in line with institutional recommendations39,40 . Another
strong point of the current study is that the screening
policy was implemented under normal clinical conditions
and by the usual staff involved in patient care. This
allowed evaluation of the effectiveness of the screening
policy in a routine care setting rather in a research setting.
Among the limitations of the study, it must first be
acknowledged that in spite of including a large cohort,
only a small minority of newborns were diagnosed with
early SGA (n = 30). This was a limitation on the creation
of a meaningful algorithm for early SGA in the absence
of pre-eclampsia. This limitation is a consequence of
our inability to perform a reliable a priori sample size
justification at the time of the study design. A second
limitation is that we are reporting a predictive model
without validating its performance, and we are therefore
likely to overestimate its performance41 . A third limitation
is the gestational age at sampling for PAPP-A and free
β-hCG; as their best performance in the prediction of
trisomy 21 is achieved at 9–10 weeks’ gestation, this may
generate conflict in combining strategies for screening for
pre-eclampsia, SGA and aneuploidy42 . In addition, we
acknowledge that the use of a small number of serum
biomarkers may have limited the detection rate of the
algorithm. Further research on the same population is
Copyright  2013 ISUOG. Published by John Wiley & Sons Ltd.
39
required to explore whether the predictive performance
reported here could be improved.
First-trimester integrated screening may be useful in the
prediction of early SGA, which may, at least in part, reflect
its strong association with pre-eclampsia. The prediction
of late SGA is not as good, but even at high false-positive
rates it might be useful as part of a first-trimester strategy
for the selection of women requiring ultrasound screening
in the third trimester, since it may identify pregnancies at
higher risk of late SGA newborns.
ACKNOWLEDGMENTS
We appreciate the contribution of Prof. Luigi Fedele, from
the Università degli Studi di Milano, Italy, for support of
Francesca Crovetto. The study was sponsored by Siemens
Healthcare Diagnostics, Tarrytown, NY, USA.
REFERENCES
1. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in
relation to morbidity and mortality among newborn infants. N
Engl J Med 1999; 340: 1234–1238.
2. Breeze AC, Lees CC. Prediction and perinatal outcomes of
fetal growth restriction. Semin Fetal Neonatal Med 2007; 12:
383–397.
3. Figueras F, Oros D, Cruz-Martinez R, Padilla N, Hernandez-
Andrade E, Botet F, Costas-Moragas C, Gratacos E. Neurobe-
havior in term, small-for-gestational age infants with normal
placental function. Pediatrics 2009; 124: e934–941.
4. Barker DJ, Gluckman PD, Godfrey KM, Harding JE, Owens
JA, Robinson JS. Fetal nutrition and cardiovascular disease in
adult life. Lancet 1993; 341: 938–941.
5. Crispi F, Bijnens B, Figueras F, Bartrons J, Eixarch E, Le
Noble F, Ahmed A, Gratacós E. Fetal growth restriction results
in remodeled and less efficient hearts in children. Circulation
2010; 121: 2427–2436.
6. Lindqvist PG, Molin J. Does antenatal identification of small-
for-gestational age fetuses significantly improve their outcome?
Ultrasound Obstet Gynecol 2005; 25: 258–264.
7. Figueras F, Gardosi J. Intrauterine growth restriction: new
concepts in antenatal surveillance, diagnosis, and management.
Am J Obstet Gynecol 2011; 204: 288–300.
8. Bujold E, Morency AM, Roberge S, Lacasse Y, Forest
JC, Giguère Y. Acetylsalicylic acid for the prevention of
preeclampsia and intra-uterine growth restriction in women
with abnormal uterine artery Doppler: a systematic review and
meta-analysis. J Obstet Gynaecol Can 2009; 31: 818–826.
9. Martin AM, Bindra R, Curcio P, Cicero S, Nicolaides KH.
Screening for pre-eclampsia and fetal growth restriction by
uterine artery Doppler at 11–14 weeks of gestation. Ultrasound
Obstet Gynecol 2001; 18: 583–586.
10. Smith GC, Stenhouse EJ, Crossley JA, Aitken DA, Cameron AD,
Connor JM. Early pregnancy levels of pregnancy-associated
plasma protein a and the risk of intrauterine growth restriction,
premature birth, preeclampsia, and stillbirth. J Clin Endocrinol
Metab 2002; 87: 1762–1767.
11. Krantz D, Goetzl L, Simpson JL, Thom E, Zachary J, Hallahan
TW, Silver R, Pergament E, Platt LD, Filkins K, Johnson A,
Mahoney M, Hogge WA, Wilson RD, Mohide P, Hershey
D, Wapner R; First Trimester Maternal Serum Biochemistry
and Fetal Nuchal Translucency Screening (BUN) Study Group.
Association of extreme first-trimester free human chorionic
gonadotropin-beta, pregnancy-associated plasma protein A,
and nuchal translucency with intrauterine growth restriction
and other adverse pregnancy outcomes. Am J Obstet Gynecol
2004; 191: 1452–1458.
Ultrasound Obstet Gynecol 2014; 43: 34–40.
40
12. Spencer K, Yu CK, Cowans NJ, Otigbah C, Nicolaides
KH. Prediction of pregnancy complications by first-trimester
maternal serum PAPP-A and free beta-hCG and with second-
trimester uterine artery Doppler. Prenat Diagn 2005; 25:
949–953.
13. Dugoff L, Lynch AM, Cioffi-Ragan D, Hobbins JC, Schultz LK,
Malone FD, D’Alton ME; FASTER Trial Research Consortium.
First trimester uterine artery Doppler abnormalities predict
subsequent intrauterine growth restriction. Am J Obstet
Gynecol 2005; 193: 1208–1212.
14. Pilalis A, Souka AP, Antsaklis P, Daskalakis G, Papantoniou
N, Mesogitis S, Antsaklis A. Screening for pre-eclampsia and
fetal growth restriction by uterine artery Doppler and PAPP-A
at 11–14 weeks’ gestation. Ultrasound Obstet Gynecol 2007;
29: 135–140.
15. Spencer K, Cowans NJ, Avgidou K, Molina F, Nicolaides
KH. First-trimester biochemical markers of aneuploidy and
the prediction of small-for-gestational age fetuses. Ultrasound
Obstet Gynecol 2008; 31: 15–19.
16. Poon LC, Zaragoza E, Akolekar R, Anagnostopoulos E,
Nicolaides KH. Maternal serum placental growth factor (PlGF)
in small for gestational age pregnancy at 11(+0) to 13(+6)
weeks of gestation. Prenat Diagn 2008; 28: 1110–1115.
17. Melchiorre K, Leslie K, Prefumo F, Bhide A, Thilaganathan B.
First-trimester uterine artery Doppler indices in the prediction
of small-for-gestational age pregnancy and intrauterine growth
restriction. Ultrasound Obstet Gynecol 2009; 33: 524–529.
18. Law LW, Leung TY, Sahota DS, Chan LW, Fung TY, Lau TK.
Which ultrasound or biochemical markers are independent
predictors of small-for-gestational age? Ultrasound Obstet
Gynecol 2009; 34: 283–287.
19. Poon LC, Karagiannis G, Staboulidou I, Shafiei A, Nicolaides
KH. Reference range of birth weight with gestation and first-
trimester prediction of small-for-gestation neonates. Prenat
Diagn 2011; 31: 58–65.
20. Karagiannis G, Akolekar R, Sarquis R, Wright D, Nicolaides
KH. Prediction of small-for-gestation neonates from biophysical
and biochemical markers at 11–13 weeks. Fetal Diagn Ther
2011; 29: 148–154.
21. Nanda S, Akolekar R, Sodre D, Vaikousi E, Nicolaides KH.
Maternal serum adiponectin at 11–13 weeks of gestation in
pregnancies delivering small for gestation neonates. Fetal Diagn
Ther 2011; 29: 274–279.
22. Hui D, Okun N, Murphy K, Kingdom J, Uleryk E, Shah
PS. Combinations of maternal serum markers to predict
preeclampsia, small for gestational age, and stillbirth: a
systematic review. J Obstet Gynaecol Can 2012; 34: 142–153.
23. Papastefanou I, Souka AP, Pilalis A, Eleftheriades M, Michalitsi
V, Kassanos D. First trimester prediction of small- and large-
for-gestation neonates by an integrated model incorporating
ultrasound parameters, biochemical indices and maternal
characteristics. Acta Obstet Gynecol Scand 2012; 91: 104–111.
24. Crispi F, Llurba E, Domı́nguez C, Martı́n-Gallán P, Cabero L,
Gratacós E. Predictive value of angiogenic factors and uterine
artery Doppler for early- versus late-onset pre-eclampsia and
intrauterine growth restriction. Ultrasound Obstet Gynecol
2008; 31: 303–309.
25. Baschat AA, Cosmi E, Bilardo CM, Wolf H, Berg C, Rigano S,
Germer U, Moyano D, Turan S, Hartung J, Bhide A, Müller
T, Bower S, Nicolaides KH, Thilaganathan B, Gembruch U,
Ferrazzi E, Hecher K, Galan HL, Harman CR. Predictors of
neonatal outcome in early-onset placental dysfunction. Obstet
Gynecol 2007; 109: 253–261.
26. Gardosi J, Kady SM, McGeown P, Francis A, Tonks A.
Classification of stillbirth by relevant condition at death
(ReCoDe): population based cohort study. BMJ 2005; 331:
1113–1117.
27. Cruz-Martinez R, Figueras F, Oros D, Padilla N, Meler E,
Copyright  2013 ISUOG. Published by John Wiley & Sons Ltd.
Crovetto et al.
Hernandez-Andrade E, Gratacos E. Cerebral blood perfusion
and neurobehavioral performance in full-term small-for-
gestational-age fetuses. Am J Obstet Gynecol 2009; 201:
474.e1–7.
28. Poon LC, Syngelaki A, Akolekar R, Lai J, Nicolaides KH.
Combined screening for preeclampsia and small for gestational
age at 11–13 weeks. Fetal Diagn Ther 2013; 33: 16–27.
29. Robinson HP, Sweet EM, Adam AH. The accuracy of
radiological estimates of gestational age using early fetal
crown–rump length measurements by ultrasound as a
basis for comparison. Br J Obstet Gynaecol 1979; 86:
525–528.
30. Casals E, Aibar C, Martinez JM, Borrell A, Soler A, Ojuel J,
Ballesta AM, Fortuny A. First-trimester biochemical markers
for Down syndrome. Prenat Diagn 1999; 19: 8–11.
31. Gomez O, Martinez JM, Figueras F, Del Rı́o M, Borobio
V, Puerto B, Coll O, Cararach V, Vanrell JA. Uterine
artery Doppler at 11–14 weeks of gestation to screen for
hypertensive disorders and associated complications in an
unselected population. Ultrasound Obstet Gynecol 2005; 26:
490–494.
32. Figueras F, Meler E, Iraola A, Eixarch E, Coll O, Figueras
J, Francis A, Gratacos E, Gardosi J. Customized birthweight
standards for a Spanish population. Eur J Obstet Gynecol
Reprod Biol 2008; 136: 20–24.
33. Brown MA, Lindheimer MD, de Swiet M, Van Assche
A, Moutquin JM. The classification and diagnosis of the
hypertensive disorders of pregnancy: statement from the
International Society for the Study of Hypertension in
Pregnancy (ISSHP). Hypertens Pregnancy 2001; 20: IX–XIV.
34. Scazzocchio E, Figueras F, Crispi F, Meler E, Masoller N, Mula
R, Gratacos E. Performance of a first-trimester screening of
preeclampsia in a routine care low-risk setting. Am J Obstet
Gynecol 2013; 208: 203.e1–10.
35. Akolekar R, Syngelaki A, Sarquis R, Zvanca M, Nicolaides
KH. Prediction of early, intermediate and late pre-eclampsia
from maternal factors, biophysical and biochemical markers at
11–13 weeks. Prenat Diagn 2011; 31: 66–74.
36. Crispi F, Domı́nguez C, Llurba E, Martı́n-Gallán P, Cabero L,
Gratacós E. Placental angiogenic growth factors and uterine
artery Doppler findings for characterization of different subsets
in preeclampsia and in isolated intrauterine growth restriction.
Am J Obstet Gynecol 2006; 195: 201–207.
37. Souka AP, Papastefanou I, Pilalis A, Michalitsi V, Kassanos
D. Performance of third-trimester ultrasound for prediction
of small-for-gestational-age neonates and evaluation of contin-
gency screening policies. Ultrasound Obstet Gynecol 2012; 39:
535–542.
38. Crispi F, Figueras F, Cruz-Lemini M, Bartrons J, Bijnens B,
Gratacos E. Cardiovascular programming in children born
small for gestational age and relationship with prenatal signs of
severity. Am J Obstet Gynecol 2012; 207: 121.e1–9.
39. American College of Obstetricians and Gynecologists. ACOG
practice bulletin No 12: clinical management guidelines
for obstetrician-gynecologists. Intrauterine growth restriction.
ACOG: Washington, DC, 2000.
40. Royal College of Obstetricians and Gynaecologists. Green-top
guide No.31. Small-for-Gestatational-Age fetus, Investigation
and Management. Royal College of Obstetricians and Gynecol-
ogists: London, UK, 2002.
41. Altman DG, Vergouwe Y, Royston P, Moons KG. Prognosis
and prognostic research: validating a prognostic model. BMJ
2009; 338: 1432–1435.
42. Borrell A, Casals E, Fortuny A, Farre MT, Gonce A, Sanchez
A, Soler A, Cararach V, Vanrell JA. First-trimester screening
for trisomy 21 combining biochemestry and ultrasound at
individually optimal gestational ages. An interventional study.
Prenat Diagn 2004; 24: 541–545.
Ultrasound Obstet Gynecol 2014; 43: 34–40.