Fetal Diagn Ther 2005;20:48–53 Received: September 5, 2003

Accepted: December 24, 2003

DOI: 10.1159/000081369

Prediction of Preeclampsia or Intrauterine

Growth Restriction by Second Trimester Serum
Screening and Uterine Doppler Velocimetry
François Audibert a Yehouda Benchimol a Clarisse Benattar b
Catherine Champagne a René Frydman a
a Service de Gynécologie-Obstétrique, Hôpital Antoine Béclère, Assistance Publique Hôpitaux de Paris et

Université Paris XI, et b Service de Biochimie, Hôpital Antoine Béclère, Assistance Publique Hôpitaux de Paris et
Université Paris XI, Clamart, France

Key Words uterine notch was associated with a significantly higher

·-Fetoprotein W Human chorionic gonadotropin W risk of both preeclampsia and IUGR. The combination of
Intrauterine growth restriction W Preeclampsia W Serum an elevated serum level and the presence of a uterine
markers W Uterine Doppler ultrasound notch had a positive predictive value (PPV) for pre-
eclampsia of 25 and 21% for hCG and AFP, respectively.
The combination of a bilateral notch with a low level of
Abstract hCG or a high level of AFP had a PPV for IUGR of 50 and
Objective: To assess the performance of screening for 43%, respectively. The sensitivity of the different tests
preeclampsia and intrauterine growth restriction by ranged from 2 to 40%. Conclusion: The combination of
combining second trimester maternal serum screening serum markers and abnormal uterine Doppler ultra-
and uterine Doppler ultrasound. Methods: A cohort of sound improves the identification of women at risk for
2,615 women underwent both maternal serum screening subsequent pregnancy complications. These results
(using human chorionic gonadotropin (hCG) and ·-feto- should encourage care providers to perform a uterine
protein (AFP)), and second trimester uterine artery Doppler ultrasound when serum markers are abnormal.
Doppler. The sensitivity, specificity and predictive value However, the sensitivity of these tests is too low to pro-
of different combinations of both tests were compared. vide an efficient generalized screening.
Results: The mean values for hCG and AFP were signifi- Copyright © 2005 S. Karger AG, Basel

cantly higher in women with subsequent preeclampsia

(p ! 0.0003 and p ! 0.03, respectively). Taking into
account obstetrical history, hCG and AFP levels, notch- Introduction
ing on uterine artery Doppler and parity, the adjusted
odds ratios were significantly higher for a high level of Complications of pregnancy such as preeclampsia,
hCG for preeclampsia, intrauterine growth restriction pregnancy-induced hypertension (PIH), intrauterine
(IUGR) and pregnancy-induced hypertension. AFP level growth restriction (IUGR), and abruptio placentae are
11.5 MoM (multiples of the median) was significantly important causes of fetal, maternal and neonatal morbidi-
correlated with subsequent IUGR. The presence of a ty and mortality [1]. Early screening of women at risk is

© 2005 S. Karger AG, Basel Dr. François Audibert

ABC 1015–3837/05/0201–0048$22.00/0 Service de Gynécologie-Obstétrique, Hôpital Sainte-Justine
Fax + 41 61 306 12 34 3175 Côte Sainte-Catherine
E-Mail karger@karger.ch Accessible online at: Montréal H3T1R2 (Canada)
www.karger.com www.karger.com/fdt E-Mail francois.audibert@umontreal.ca
currently based on obstetrical history [2] and uterine Table 1. Frequency of pregnancy complications (n = 2,615)
Doppler ultrasound, generally conducted in the second
Pregnancy complication n Prevalence, %
trimester [3]. Previous reports suggest a correlation be-
tween the levels of maternal serum human chorionic Preeclampsia 51 1.95
gonadotropin (hCG) or ·-fetoprotein (AFP), and subse- PIH 122 4.66
quent pregnancy complications [4–9]. Early preventive HELLP syndrome 3 0.11
treatment with aspirin or other drugs for high-risk women Chronic hypertension 44 1.68
Birthweight ! 10th percentile 230 8.79
would appear to reduce the onset of these complications,
IUFD 11 0.42
confirming the benefits of earlier and more effective Abruptio placenta 7 0.27
screening [10–14].
Routine screening of chromosomal abnormalities is PIH = Pregnancy-induced hypertension; IUFD = intrauterine
currently performed with the double (AFP and hCG) or fetal demise.
the triple (AFP, hCG and uE3) test, with a high uptake in
many countries. We aimed to assess the performance of
early screening for preeclampsia and IUGR by combining
maternal serum screening with uterine Doppler ultra-
sound. elevation of AST (aspartate aminotransferase) 1 70 IU/l or ALT (ala-
nine aminotransferase) 1 60 IU/l; thrombocytopenia ! 100,000 plate-
lets/mm3 [17, 18].
IUGR was defined by a weight below the 10th percentile of our
Material and Methods reference charts [19]. PIH was defined as a systolic blood pressure
6140 mm Hg, or a diastolic blood pressure 690 mm Hg, measured
The study was conducted in a cohort of 2,615 women in whom at least on two occasions, and diagnosed after 20 weeks.
both a double test between 14 and 18 weeks (by maternal serum AFP We used the Amerlite® kit (Ortho Clinical Diagnostics, Issy-les-
and total serum hCG assay), and a uterine Doppler ultrasound Moulineaux, France) for the hCG and AFP assay. Pathological uter-
between 18 and 26 weeks were performed. This cohort was taken ine Doppler ultrasound was defined by the presence a protodiastolic
from an initial database of 4,556 women enrolled for Down’s syn- notch or a pulsatility index above the 1 95th percentile. Placental lat-
drome screening, as previously described [15]. From May 1994 eral location was not recorded in the database.
through April 1998, Down’s syndrome screening was offered to all The statistical analysis used the non-parametric Wilcoxon test to
women registered before 14 weeks in our institution. An ultrasound compare continuous variables, the ¯2 test to compare categorical
was conducted between 10 and 14 weeks to date the pregnancy, and variables and multivariate logistic regression to calculate the ad-
to measure nuchal translucency. This was followed by a maternal justed odds ratios. The statistical software used was Stata 7.0 (Stata
serum marker assay using the double test between 14 and 18 weeks. Corp., College Station, Tex., USA). The significance level was p !
We excluded the following women from the analysis: multiple preg- 0.05.
nancies, women without an ultrasound between 10 and 14 weeks,
women referred for increased nuchal translucency. From this initial
cohort of 4,556 women, 1,941 were excluded for the following rea-
sons: no Doppler ultrasound between 18 and 26 weeks (n = 1,878); Results
delivery ! 24 weeks (n = 55) (including 21 terminations of pregnancy
for chromosomal abnormalities, 8 for major structural anomaly, and The mean age of our population was 30.9 B 4.5 years,
26 spontaneous pregnancy losses). Eight women were lost to follow-
and there were 48.5% nulliparas. The mean (SD) gesta-
up. Thus, the study population consisted of 2,615 women who had
both a double test (between 14 and 18 weeks), a uterine Doppler tional age delivery was 39.7 (1.8) weeks and the mean
ultrasound (between 18 and 26 weeks), who delivered after 24 weeks (SD) birth weight was 3,288 (530) g. The mean values for
and for whom outcome of pregnancy was known. maternal serum hCG and AFP expressed in multiples of
Preeclampsia was defined as a systolic blood pressure 6140 mm the median (MoM) were 1.03 and 1.05 MoM, respec-
Hg or a diastolic pressure 690 mm Hg on two occasions, associated
tively.
with proteinuria 1 0.3 g/24 h or at least ‘two-plus’ protein on the
urine dipstick [16]. Severe preeclampsia was defined by the presence We identified 426 women (16.3%) with at least one of
of at least one of the following criteria: systolic blood pressure the following complications: preeclampsia, IUGR, PIH,
6160 mm Hg or diastolic pressure 6110 mm Hg, after 4 h of bed chronic hypertension, stillbirth or abruptio placenta (ta-
rest; proteinuria 65 g/24 h or at least 3+ protein on the urine ble 1). Among women who developed preeclampsia, 59%
dipstick; oliguria ^400 ml/24 h; cerebral or visual disturbances,
were nulliparous. The prevalence of preeclampsia, IUGR
epigastric pain, pulmonary edema or cyanosis; thrombocytopenia
! 100,000 mm3. HELLP syndrome was defined by the association of: and PIH was higher in nulliparas compared to multiparas
hemolysis (presence of schistocytes on the blood smear, or LDH level (respectively 2.4 vs. 1.6%, p = 0.051; 9.9 vs. 7.7%, p =
1 600 IU/l, haptoglobin level ! 0.4 g/l, total bilirubin 1 20 Ìmol/l); 0.13 and 6.2 vs. 3.4%, p ! 0.001). Among women with a

Screening for Preeclampsia and IUGR Fetal Diagn Ther 2005;20:48–53 49

 Table 2. Logistic regression: independent prediction of the various tests for pregnancy complications

Preeclampsia (n = 51) IUGR (n = 230) PIH (n = 122)

ORa 95% CI ORa 95% CI ORa 95% CI

History of PE, PIH or IUGR 3.37 1.45–7.85 3.19 2.03–5.03 4.91 2.72–8.87
Multiparity 0.81 0.54–1.21 0.81 0.68–0.98 0.59 0.44–0.78
HCG 1 2 MoM 1.82 1.33–2.48 1.10 0.88–1.37 1.35 1.04–1.75
AFP 1 1.5 MoM 1.24 0.74–2.07 1.41 1.07–1.87 0.73 0.43–1.27
Notch (at least one) 2.42 1.68–3.50 1.72 1.38–2.16 1.29 0.95–1.76

PE = Preeclampsia; PIH = pregnancy-induced hypertension; IUGR = intrauterine growth restriction; ORa =

adjusted odds ratio; CI = confidence interval.

Table 3. Performance of different screening tests for predicting preeclampsia

Test Positive Sensitivity Specificity PPV NPV

tests, % % % % %

History of PE, PIH or IUGR 6.45 21.56 93.86 6.71 98.31

Bilateral notch 4.39 21.56 95.94 9.56 98.4
At least 1 notch 12.19 39.21 88.33 6.27 98.65
hCG 1 2 MoM 5.54 15.68 94.65 5.51 98.26
hCG 1 2 MoM and at least 1 notch 0.6 7.84 99.53 25 98.19
History of preeclampsia or bilateral notch
or hCG 1 2.5 MoM 9.04 41.17 91.61 9.13 98.7
AFP 1 1.5 MoM 9.79 19.6 90.4 3.9 98.26
AFP 1 1.5 MoM and at least 1 notch 1.1 7.84 99.02 13.79 98.18
AFP 1 1.5 MoM and bilateral notch 0.53 5.88 99.57 21.43 98.15

PE = Preeclampsia; PIH = pregnancy-induced hypertension; IUGR = intrauterine growth restriction; PPV =

positive predictive value; NPV = negative predictive value; MoM = multiples of the median. Prevalence of pre-
eclampsia = 1.95%.

history of preeclampsia, HELLP syndrome or abruptio Multivariate logistic regression including obstetrical
placenta (n = 50), 10% had a recurrence of preeclampsia, history, hCG and AFP levels, presence of at least one uter-
14% of PIH and 12% of IUGR. Among those with a histo- ine notch and multiparity, is shown in table 2. Taking
ry of IUGR (n = 64), 36% had a recurrence of IUGR. account of these five adjustment variables, the risk of
Total hCG and AFP levels were significantly higher in preeclampsia appeared to be significantly and indepen-
women with preeclampsia than in those without (respec- dently increased in the presence of nulliparity, obstetrical
tively 1.42 vs. 1.03 MoM, p ! 0.0003 for hCG, and 1.15 history, high level of hCG, or uterine artery notching.
vs. 1.05 MoM, p ! 0.03 for AFP). Among women with Similarly, the risk of IUGR was significantly associated
severe IUGR (!5th percentile) but without preeclampsia, with nulliparity, obstetrical history, a high level of AFP,
the AFP level was significantly higher than the rest of the or notching. The risk of PIH was associated with nullipar-
population, with levels of 1.12 vs. 1.05 MoM (p ! 0.03). ity, obstetrical history, and with a high level of hCG.
The AFP level was also significantly higher in cases of Table 3 shows the sensitivity, specificity, positive pre-
intrauterine fetal demise with a mean value of 1.29 vs. dictive value (PPV) and negative predictive value of the
1.05 MoM (p ! 0.03). uterine Doppler ultrasound, serum markers and different
screening combinations for prediction of preeclampsia.

50 Fetal Diagn Ther 2005;20:48–53 Audibert/Benchimol/Benattar/Champagne/

Frydman
 Table 4. Performance of different screening tests for predicting IUGR

Test Positive Sensitivity Specificity PPV NPV

tests, % % % % %

History of PE, PIH or IUGR 6.45 15.69 94.43 21.34 92.09

Bilateral notch 4.39 13.04 96.43 26.08 92
At least 1 notch 12.19 22.6 88.8 16.3 92.24
hCG 1 1.5 MoM 15.87 20 84.52 11.08 91.63
hCG 1 2 MoM 5.54 9.13 94.8 14.48 91.53
hCG ! 0.5 MoM and at least one notch 1.33 3.04 98.82 20 91.35
hCG 1 1.7 MoM and at least one notch 1.45 4.78 98.86 28.94 91.5
hCG ! 0.5 MoM and bilateral notching 0.38 2.17 99.79 50 91.36
AFP 1 1.5 MoM 9.78 13.91 90.6 12.5 91.6
AFP 1 1.5 MoM and at least one notch 1.11 3.91 99.16 31.03 91.45
AFP 1 1.5 MoM and bilateral notching 0.53 2.6 99.66 42.85 91.38

PE = Preeclampsia; PIH = pregnancy-induced hypertension; IUGR = intrauterine growth restriction; PPV =

positive predictive value; NPV = negative predictive value; MoM = multiples of the median. Prevalence of IUGR =
8.79%.

The best sensitivity is obtained with at least one of the plication occurs. The predictive value of uterine Doppler
following risk factors: obstetrical history, bilateral notch- varies according to the prevalence of the different compli-
ing or hCG 12.5 MoM. The best PPV is obtained by asso- cations of pregnancy. The sensitivities and PPVs attribut-
ciating hCG 12 MoM and at least one notch (25%). With ed to the uterine Doppler ultrasound in a low-risk popula-
an AFP level 11.5 MoM in addition to bilateral notching, tion are variable in the literature, ranging from 14 to 77%
1 woman in 5 will develop preeclampsia. and from 7 to 50% respectively for preeclampsia, and
Table 4 shows the sensitivity, specificity, PPV and neg- from 7 to 32% and 10 to 50% for IUGR [20–22].
ative predictive value of the uterine Doppler ultrasound, Many studies have examined a possible correlation
serum markers and some of their combinations for between the hCG level and the occurrence of pregnancy
IUGR. For the prediction of IUGR, Doppler offers a bet- complications, since the introduction of serum screening
ter sensitivity than other tests (22.6% with ‘at least one for Down’s syndrome [5, 6, 23–29]. Most of these articles
notch’). The PPV of a bilateral notch rises from 26 to 43% are case-control studies. The sensitivity and PPVs for
or to 50% when an AFP level 11.5 MoM or an hCG level preeclampsia reported in these studies for an hCG level
!0.5 MoM are respectively associated. 12 MoM are highly variable, ranging from 20 to 69% and
3 to 15% respectively [23–27, 29]. In a large retrospective
study, Walton et al. [9] have examined the association
Discussion between hCG levels and pregnancy outcome in 28,743
women. Despite a higher incidence of stillbirth, placental
This study offers the largest series to date assessing the abnormalities and PIH among women with elevated hCG
test properties of both uterine Doppler ultrasound and levels, the authors did not find any clinically relevant
maternal serum markers for the prediction of pregnancy association between this test and the most frequent preg-
complications in the same group of women. We found a nancy complications. In this study, the results were not
significant correlation between a pathological level of reported as sensitivities, specificities or predictive values.
hCG or AFP on the one hand, the presence of notching on As early as 1,939, a correlation was suggested between a
the uterine Doppler on the other, and the occurrence of high serum hCG concentrations and the onset of pre-
pregnancy complications. The combination of these tests eclampsia [30]. The hypothesis explaining the increase of
improved the predictive value of screening. hCG in preeclampsia or IUGR is an initial drop in pla-
Uterine Doppler ultrasound is currently recommended cental perfusion, at the origin of necrosis of syncytiotro-
in case of an obstetrical history or when a pregnancy com- phoblast cells and of increased cytotrophoblast mitotic

Screening for Preeclampsia and IUGR Fetal Diagn Ther 2005;20:48–53 51

activity. Morssink et al. [29] found a higher number of we selected a high-risk population by eliminating from
placental lesions of the infarctal type and placental isch- our study 1,878 women who did not undergo a uterine
emia in cases of IUGR associated with a high level of Doppler ultrasound between 18 and 26 weeks, since the
hCG. Interestingly, we also found an association between observed prevalence of the various pregnancy complica-
very low levels of hCG (!0.5 MoM) and IUGR. This tions (preeclampsia: 1.95%, IUGR: 8.8%) and the preva-
seemingly contradictory fact could be due to an early lence of bilateral notch (4.4%) or at least one notch
abnormal placental function leading to a very low hCG (12.2%) are similar to those expected in a low-risk popula-
secretion in a subgroup of women. Two studies have tion [20–22].
addressed this issue, yielding conflicting results [31, 32]. Given the results presented above, we believe that in
Further studies are needed to confirm this finding. the absence of chromosomal abnormality or neural tube
Few studies have examined the combination of uterine defect, a high or unusually low level of hCG or a high level
Doppler ultrasound and serum screening to predict pre- of AFP, should lead to propose a uterine Doppler ultra-
eclampsia or IUGR. Most of them suggest an increase in sound. Furthermore, within the scope of screening for
the risk of complications, but do not distinguish between complications of pregnancy, maternal serum markers
the various complications of pregnancy. These studies are should be read individually for serum levels, expressed in
conducted on small numbers of women and in selected MoM and not just as combined risk, as is currently the
populations of pathological serum markers levels [33, 34]. case in Down’s syndrome screening. An abnormal level of
Based on our results, the combination of the three tests a serum marker, once a chromosomal abnormality has
with the highest PPV for preeclampsia and IUGR screen- been ruled out, can be the first sign of a complication of
ing (hCG 12 MoM and at least one notch, or AFP pregnancy. This study, even though carried out in an
11.5 MoM associated with bilateral notching, or hCG unselected population, noted a pregnancy complication
!0.5 MoM and bilateral notching) would enable to identi- rate in 50% of cases who tested positive. However, due to
fy a group of women with a 44% risk of these complica- its poor sensitivity, the test combining second trimester
tions, in our population with a risk of 1.5%. However, the serum markers and uterine Doppler ultrasound cannot be
sensitivity of such a screening would be very low (6.4%). proposed as a generalized screening method for pregnan-
Maternal serum screening is currently performed in cy complications in a low-risk population. By identifying
60–80% of pregnant women in France by routine screen- during the second trimester of pregnancy a group of wom-
ing for chromosomal abnormalities. The use of these en at very high risk of preeclampsia or IUGR, it is now
serum markers, in association with the uterine Doppler possible to offer closer clinical and ultrasound monitor-
ultrasound, represents a screening test which, despite a ing. The predictive value of this combined screening test
poor sensitivity, offers a very high PPV value in a low-risk should now be evaluated during the first trimester. If this
population, multiplying the risk by 2–3 compared to early screening proves to be efficient, prospective con-
Doppler alone. trolled studies of prophylactic treatment (such as aspirin
Although our study included 2,615 women out of an or antioxidants) could be conducted in this high-risk pop-
initial population of 4,556 subjects, we do not believe that ulation.

References

1 National High Blood Pressure Education Pro- 4 Walters BN, Lao T, Smith V, De Swiet M: 7 Williams MA, Hickok DE, Zingheim RW, et
gram Working Group Report on High Blood Alpha-fetoprotein elevation and proteinuric al: Elevated maternal serum ·-fetoprotein lev-
Pressure in Pregnancy. Am J Obstet Gynecol pre-eclampsia. Br J Obstet Gynaecol 1985;92: els and midtrimester placental abnormalities in
1990;163:1691–1712. 341–344. relation to subsequent adverse pregnancy out-
2 Stamilio DM, Sehdev HM, Morgan MA, Pro- 5 Sorensen TK, Williams MA, Zingheim RW, comes. Am J Obstet Gynecol 1992;167:1032–
pert K, Macones GA: Can antenatal clinical Clement SJ, Hickok DE: Elevated second-tri- 1037.
and biochemical markers predict the develop- mester human chorionic gonadotropin and 8 Van Rijn M, van der Schouw YT, Hagenaars
ment of severe preeclampsia? Am J Obstet Gy- subsequent pregnancy-induced hypertension. AM, Visser GH, Christiaens GC: Adverse ob-
necol 2000;182:589–594. Am J Obstet Gynecol 1993;169:834–838. stetric outcome in low- and high-risk pregnan-
3 Campbell S, Pearce JM, Hackett G, Cohen- 6 Gonen R, Perez R, David M, et al: The associa- cies: Predictive value of maternal serum
Overbeek T, Hernandez C: Qualitative assess- tion between unexplained second-trimester screening. Obstet Gynecol 1999;94:929–934.
ment of uteroplacental blood flow: Early maternal serum hCG elevation and pregnancy 9 Walton DL, Norem CT, Schoen EJ, Ray GT,
screening test for high-risk pregnancies. Obstet complications. Obstet Gynecol 1992;80:83– Colby CJ: Second-trimester serum chorionic
Gynecol 1986;68:649–653. 86. gonadotropin concentrations and complica-
tions and outcome of pregnancy. N Engl J Med
1999;341:2033–2038.

52 Fetal Diagn Ther 2005;20:48–53 Audibert/Benchimol/Benattar/Champagne/

Frydman
10 Beaufils M, Uzan S, Donsimoni R, Colau JC: 19 Mamelle N, Munoz F, Martin JL, Laumon B, 27 Luckas M, Hawe J, Meekins J, Neilson J, Wal-
Prevention of pre-eclampsia by early antiplate- Grandjean H: Fetal growth from the AUDI- kinshaw S: Second trimester serum free ß-
let therapy. Lancet 1985;i:840–842. POG study. II. Application for the diagnosis of human chorionic gonadotrophin levels as a
11 Chappell LC, Seed PT, Briley AL, et al: Effect intrauterine growth retardation (in French). J predictor of pre-eclampsia. Acta Obstet Gyne-
of antioxidants on the occurrence of pre- Gynecol Obstet Biol Reprod 1996;25:71–77. col Scand 1998;77:381–384.
eclampsia in women at increased risk: A ran- 20 Harrington K, Cooper D, Lees C, Hecher K, 28 Raty R, Koskinen P, Alanen A, et al: Prediction
domised trial. Lancet 1999;354:810–816. Campbell S: Doppler ultrasound of the uterine of pre-eclampsia with maternal mid-trimester
12 Coomarasamy A, Papaioannou S, Gee H, arteries: The importance of bilateral notching total renin, inhibin A, AFP and free ß-hCG lev-
Khan KS: Aspirin for the prevention of pre- in the prediction of pre-eclampsia, placental els. Prenat Diagn 1999;19:122–127.
eclampsia in women with abnormal uterine abruption or delivery of a small-for-gestational- 29 Morssink LP, de Wolf BT, Kornman LH, et al:
artery Doppler: A meta-analysis. Obstet Gyne- age baby. Ultrasound Obstet Gynecol 1996;7: The relation between serum markers in the sec-
col 2001;98:861–866. 182–188. ond trimester and placental pathology. A study
13 CLASP: A randomised trial of low-dose aspirin 21 Mires GJ, Williams FL, Leslie J, Howie PW: on extremely small for gestational age fetuses.
for the prevention and treatment of pre- Assessment of uterine arterial notching as a Br J Obstet Gynaecol 1996;103:779–783.
eclampsia among 9,364 pregnant women. screening test for adverse pregnancy outcome. 30 Smith G, Smith O: Anterior pituitary like hor-
CLASP (Collaborative Low-Dose Aspirin Am J Obstet Gynecol 1998;179:1317–1323. mone in late pregnancy toxemia: Summary of
Study in Pregnancy) Collaborative Group. 22 Irion O, Masse J, Forest JC, Moutquin JM: results since 1932. Am J Obstet Gynecol 1939;
Lancet 1994;343:619–629. Prediction of pre-eclampsia, low birthweight 36:618–624.
14 McParland P, Pearce JM, Chamberlain GV: for gestation and prematurity by uterine artery 31 Santolaya-Forgas, Burd LI, Burton BK: Clini-
Doppler ultrasound and aspirin in recognition blood flow velocity waveforms analysis in low- cal significance of low levels of second-trimes-
and prevention of pregnancy-induced hyper- risk nulliparous women. Br J Obstet Gynaecol ter maternal serum human chorionic gonado-
tension. Lancet 1990;335:1552–1555. 1998;105:422–429. tropin. Fetal Diagn Ther 1994;20:362–366.
15 Audibert F, Dommergues M, Benattar C, et al: 23 Muller F, Savey L, Le Fiblec B, et al: Maternal 32 Spencer K: Second-trimester prenatal screen-
Screening for Down syndrome using first-tri- serum human chorionic gonadotropin level at ing for Down syndrome and the relationship of
mester ultrasound and second-trimester mater- fifteen weeks is a predictor for preeclampsia. maternal serum biochemical markers to preg-
nal serum markers in a low-risk population: A Am J Obstet Gynecol 1996;175:37–40. nancy complication with adverse outcome.
prospective longitudinal study. Ultrasound 24 Vaillant P, David E, Constant I, et al: Validity Prenat Diagn 2000;20:652–656.
Obstet Gynecol 2001;18:26–31. in nulliparas of increased ß-human chorionic 33 Konchak PS, Bernstein IM, Capeless EL: Uter-
16 Davey DA, MacGillivray I: The classification gonadotrophin at mid-term for predicting preg- ine artery Doppler velocimetry in the detection
and definition of the hypertensive disorders of nancy-induced hypertension complicated with of adverse obstetric outcomes in women with
pregnancy. Am J Obstet Gynecol 1988;158: proteinuria and intrauterine growth retarda- unexplained elevated maternal serum ·-feto-
892–898. tion. Nephron 1996;72:557–563. protein levels. Am J Obstet Gynecol 1995;173:
17 Weinstein L: Syndrome of hemolysis, elevated 25 Ashour AM, Lieberman ES, Haug LE, Repke 1115–1119.
liver enzymes, and low platelet count: A severe JT: The value of elevated second-trimester ß- 34 Palacio M, Jauniaux E, Kingdom J, et al: Peri-
consequence of hypertension in pregnancy. Am human chorionic gonadotropin in predicting natal outcome in pregnancies with a positive
J Obstet Gynecol 1982;142:159–167. development of preeclampsia. Am J Obstet Gy- serum screening for Down’s syndrome due to
18 Sibai BM: The HELLP syndrome (hemolysis, necol 1997;176:438–442. elevated levels of free ß-human chorionic go-
elevated liver enzymes, and low platelets): 26 Pouta AM, Hartikainen AL, Vuolteenaho OJ, nadotropin. Ultrasound Obstet Gynecol 1999;
Much ado about nothing? Am J Obstet Gyne- Ruokonen AO, Laatikainen TJ: Midtrimester 13:58–62.
col 1990;162:311–316. N-terminal proatrial natriuretic peptide, free ß-
hCG, and ·-fetoprotein in predicting pre-
eclampsia. Obstet Gynecol 1998;91:940–944.

Screening for Preeclampsia and IUGR Fetal Diagn Ther 2005;20:48–53 53