Department of Obstetrics and Gynecology, Sant Joan de Du University Hospital, Barcelona, and b SAMID Network
(Spanish Collaborative Maternal and Child Health Research Network), Institute Carlos III, Madrid, Spain
Abstract
Aim: The potential of uterine artery (UA) Doppler pulsatility
index (PI) and maternal serum placental growth factor (PlGF)
level to predict perinatal outcome was explored in pregnancies complicated by intrauterine fetal growth restriction (IUGR)
or preeclampsia (PE). Methods: This longitudinal, prospective,
and case-controlled study was conducted over a period of 24
months. At-risk pregnancies involving small-for-gestationalage (SGA) fetuses, IUGR, gestational hypertension (GH), or PE
were investigated, analyzing UA Doppler PI findings and maternal PlGF levels determined at the time of diagnosis (third
trimester). Results: UA Doppler PI and maternal serum PlGF
values differed significantly in pregnancies complicated by
IUGR and/or PE (vs. SGA or GH, p < 0.01). In the context of IUGR
or PE, both parameters also differed significantly by perinatal
outcome (adverse vs. normal, p < 0.01), although no predictive
advantage over UA Doppler PI alone was conferred by adding
a PlGF assay. Conclusion: UA Doppler PI and maternal serum
PlGF determinations in the third trimester help identify preg-
Introduction
This study is based on the following research project: FIS PI041637: Use
of VEGF and PlGF levels in maternal serum and UA Doppler pulsatility index to predict intrauterine growth restriction and preeclampsia.
M. Dolores Gomez-Roig
Department of Obstetrics and Gynecology
Sant Joan de Du University Hospital, Pg. Sant Joan de Du, 2
ES08950 Esplugues de Llobregat, Barcelona (Spain)
E-Mail lgomezroig@hsjdbcn.org
Downloaded by:
HINARI Bolivia
198.143.58.33 - 10/20/2015 3:59:12 PM
Key Words
Placental growth factor Uterine artery Doppler pulsatility
index Intrauterine fetal growth restriction Preeclampsia
100
systolic BP 140 mm Hg and/or diastolic BP 90 mm Hg measured twice in 6 h after 10 min of rest with the mother sitting,
and the measure taken in the arm at heart level with negative
proteinuria [2427]. All percentiles of fetal weights were confirmed at birth.
In each instance, the estimated fetal weight was derived from
Hadlocks reference curves, using sonographic measurements of
biparietal diameter, femoral length, and circumferences of the abdomen and head. The gestational age in the first trimester was estimated by fetal crown-rump length (Robinson curves), adjusted
for the date of the last menstrual period, as needed [28].
For an estimated fetal weight <10th percentile, Doppler examinations were used to distinguish SGA from IUGR, based on
the measurements of the umbilical arteries, middle cerebral arteries, and UAs. At least 3 determinations were made at each site,
using the best figure for the final analysis. Pathologic PIs of umbilical (PI >95th percentile) and middle cerebral arteries (PI <5th
percentile) according to the reference curves by Arduini and Rizzo [29] constituted Doppler abnormalities. UA Doppler studies
were considered abnormal at a mean PI >95th percentile bilaterally.
PlGF was measured by enzyme-linked immunosorbent assay
(ELISA) kits (R&D Systems Europe, Ltd., Abingdon, UK). These
assays detect free but not bound growth factors. Venous blood serum was taken within 72 h after diagnosis of SGA, IUGR, GH, or
PE, separated by centrifugation at 1,400 g for 10 min, and stored at
80 C. All samples were collected, handled, and stored under the
same conditions and tested in duplicate. Plate-to-plate variability
was controlled with an independent standard curve on each plate.
Serum from cases and their matched control subjects were simultaneously and blindly run on the same plates. In all kits, the intraassay precision was always <5% and the interassay precision 10%.
The linear regression coefficients of the standard curves were never below 0.99.
Z-scores were calculated from reference tables for the Spanish
population [30]. The clinicians were blinded to PlGF determination and to UA Doppler information during the study. The clinicians had advanced experience in fetal ultrasound.
Exclusion criteria were as follows: (1) medications that possibly
affect fetal growth; (2) genetic variants impacting growth potential
(i.e. chromosomal abnormalities and genetic syndromes); (3) pregestational or gestational diabetes; (4) multiple pregnancies; (5)
infections; (6) medically unattended pregnancies or births; (7)
pregnancies or births at other hospitals (outside of protocol); (8)
no ultrasound before 20 weeks of estimated gestational age; (9)
IUGR/SGA, PE, or GH diagnosed after 37 weeks of gestation, and
(10) hypertension diagnosed in advance of pregnancy or before 20
weeks of gestation.
The following were considered adverse perinatal outcomes: (1)
non-reassuring fetal heart rate before labor; (2) prematurity (<34
weeks) due to PE or IUGR; (3) requiring neonatal unit care, and
(4) maternal complications due to PE or hypertension.
Standard statistical software (SPSS v19; SPSS Inc., Chicago, Ill.,
USA) was used for analysis. For group comparisons, 2 (qualitative
variables) and Students t test (quantitative variables) were applied.
Statistical significance was set to p 0.05.
Our study protocol was approved by the Institutional Review
Board of Barcelona University Hospital. Written informed consent was obtained from each patient before undergoing any studyrelated procedures.
Gomez-Roig/Mazarico/Sabria/Parra/
Oton/Vela
Downloaded by:
HINARI Bolivia
198.143.58.33 - 10/20/2015 3:59:12 PM
placental underperfusion, hypoxia, and ischemia. Accordingly, IUGR and PE are signs of placental decompensation [1]. Uterine artery (UA) Doppler studies reliably
detect inadequate placental perfusion in complicated
pregnancies (SGA, IUGR, GH, and/or PE conditions),
correlating well with maternal/fetal prognosis by the
third trimester [210].
Placental growth factor (PlGF) is a protein belonging
to a family of platelet-derived growth factors implicated
in angiogenesis. The placenta (specifically trophoblast)
is its chief source, so circulating PlGF is a ready index
of placental perfusion. Several studies have found diminished plasma levels of PlGF during the first and second trimesters in pregnancies where GH, PE, SGA, or
IUGR was later manifested [1116]. Studies done during the third trimester are less numerous, and some
have presented contradictory findings in this regard [7,
1720].
For the above reasons, a maternal serum PlGF assay
was added to the UA Doppler pulsatility index (PI) in the
third trimester of at-risk pregnancies (with SGA, IUGR,
GH, and/or PE as complications). Our intent was to assess its impact on differentiating these specific conditions
(IUGR vs. SGA and PE vs. GH) and to predict related
perinatal outcomes.
SGA
(n = 46)
IUGR
(n = 20)
GH
(n = 25)
PE
(n = 42)
59.18
14.29
26.53
39.59 1.92
3,274.91 466.54
52.3 19.3
7.24 0.05
7.28 0.06
67.5
15
17.5
38.57 1.55
2,590 382.71
7.1 2
7.25 0.07
7.29 0.07
28.48
8.66
62.86
36.06 2.50
2,250.62 481.52
3.4 6.8
7.24 0.07
7.29 0.06
61.2
10.5
28.3
37.01 2.40
2,813 367.43
36.21 8.9
7.26 0.04
7.28 0.05
48.98
51.02
47.5
52.5
62.73
37.27
49.1
50.9
p
value
<0.001
19.31
12.39
68.3
35.34 1.32
<0.001
2,250.45 211.5 <0.001
11.12 4.5
<0.001
7.24 0.07
0.639
7.28 0.06
0.072
0.071
45.6
54.4
volving SGA, IUGR, GH, PE, and PE plus IUGR relative to controls
Z-scores
Controls
PE
IUGR
PE plus IUGR
GH
SGA
344
42
20
23
25
46
0.1027
0.6524
0.7695
1.1431
0.3120
0.0444
Total
500
0.1836
p value
<0.01
Mean
Controls
PE
IUGR
PE plus IUGR
GH
SGA
344
42
20
23
25
46
0.8132
1.3251
1.4744
1.6817
0.7462
0.9446
Total
500
1.0077
p value
<0.01
Results
101
Downloaded by:
HINARI Bolivia
198.143.58.33 - 10/20/2015 3:59:12 PM
Mode of delivery, %
Spontaneous delivery
Forceps delivery
Cesarean section
Gestational age at delivery, weeks
Birth weight, g
Birth weight percentile
Arterial pH
Venous pH
Gender of the neonate, %
Male
Female
Controls
(n = 344)
0.6
0.2
0.2
0.4
0.6
1 specificity
0.8
1.0
1.0
Susceptibility
0.8
0.6
0.2
0.2
0.2
0.4
0.6
1 specificity
0.8
1.0
ROC curve
0.4
0.2
0.4
0.6
0.4
0.6
1 specificity
0.8
1.0
ROC curve
1.0
0.8
0.6
0.4
0.2
0
0.2
0.4
0.6
1 specificity
0.8
1.0
102
Gomez-Roig/Mazarico/Sabria/Parra/
Oton/Vela
Downloaded by:
HINARI Bolivia
198.143.58.33 - 10/20/2015 3:59:12 PM
0.8
0.4
1.0
Susceptibility
Susceptibility
0.8
1.0
ROC curve
Susceptibility
ROC curve
Table 4. Positive/negative likelihood ratios (LHR) for a 10% FPR for mean UA Doppler PI and maternal serum
PlGF
Positive LHR
PE and/or IUGR
PE and/or IUGR relative to all complications
Adverse perinatal outcome
Adverse perinatal outcome in complicated pregnancies
Adverse perinatal outcome with PE and/or IUGR
ROC curve
1.0
0.6
PlGF
mean UA
Doppler PI
PlGF
6.91
6.91
8.33
7.29
6.38
6.54
6.3
8.75
7.92
8.04
0.34
0.34
0.18
0.3
0.4
0.38
0.41
0.13
0.23
0.21
0.4
PI
PlGF (inverted)
Baseline
0.2
0
mean UA
Doppler PI
Discussion
0.2
0.4
0.6
1 specificity
0.8
1.0
103
Downloaded by:
HINARI Bolivia
198.143.58.33 - 10/20/2015 3:59:12 PM
Susceptibility
0.8
Negative LHR
cies. Sibiude et al. [17] have demonstrated that PlGF levels in instances of suspected PE or IUGR aid in identifying
cases with worse perinatal outcomes due to eclampsia,
HELLP syndrome, extreme IUGR (<3rd percentile), or
elective delivery at <34 weeks of gestation.
Moreover, another important contribution of our
study is that these markers in the third trimester are also
able to distinguish those pregnancies with GH or SGA at
a higher risk of complications, and not only in pregnancies complicated with IUGR and/or PE.
Interestingly, we found a moderate correlation between UA Doppler PI and PlGF levels. PlGF contributes
to uterine vascular remodeling during pregnancy [32],
which in turn accounts for placental vascular impedance.
Given these results, we corroborated that UA Doppler
and PlGF contributed independently to the prediction of
adverse outcome.
Comparing the results of the ROC curves, the sensitivities, the likelihood ratios and the positive predictive
value of PlGF and mean UA Doppler PI, our results
showed that adding PlGF determination to mean UA
Doppler PI conferred no benefit in terms of predicting
adverse perinatal outcome. However, a recent publication [7] assessing other biomarkers of PE and IUGR includes PlGF as a ratio of sFlt-1:PlGF. More research is
clearly needed to confirm the utility of PlGF alone or
combined with other prognosticators in this setting.
Acknowledgments
We would like to thank the Gynecology and Obstetrics Service
at Sant Joan de Du University Hospital, Barcelona, Spain, the
Fund for Health Research of the Spanish Social Security Service for
financing part of this project (Exp. PI041637), and the Spanish
Collaborative Maternal and Child Health Research Network.
References
104
Gomez-Roig/Mazarico/Sabria/Parra/
Oton/Vela
Downloaded by:
HINARI Bolivia
198.143.58.33 - 10/20/2015 3:59:12 PM
105
Downloaded by:
HINARI Bolivia
198.143.58.33 - 10/20/2015 3:59:12 PM
22 Turan OM, Turan S, Gungor S, Berg C, Moyano D, Gembruch U, Nicolaides KH, Harman
CR, Baschat AA: Progression of Doppler abnormalities in intrauterine growth restriction.
Ultrasound Obstet Gynecol 2008;32:160167.
23 Alfirevic Z, Stampalija T, Gyte GML: Fetal and
umbilical Doppler ultrasound in high-risk
pregnancies. Cochrane Database Syst Rev
2013;11:CD007529.
24 Helewa M, Burrows R, Smith J, Williams K,
Brains P, Rabkin SW: Report of the Canadian
Hypertension Society Consensus Conference:
1. Definitions, evaluation and classification of
hypertensive disorders in pregnancy. CMAJ
1997;157:715725.
25 Duley L: Pre-eclampsia and hypertension.
Clin Evid 2002;7:12961309.