NIH Public Access

Author Manuscript
J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.
Published in final edited form as:
NIH-PA Author Manuscript

J Ultrasound Med. 2013 September ; 32(9): 1593–1600. doi:10.7863/ultra.32.9.1593.

The Efficiency of First-Trimester Uterine Artery Doppler,

ADAM12, PAPP-A and Maternal Characteristics in the Prediction
of Pre-Eclampsia
Katherine R. GOETZINGER, M.D., M.S.C.I., Yan ZHONG, M.D., Alison G. CAHILL, M.D.,
M.S.C.I., Linda ODIBO, R.N., M.N., George A. MACONES, M.D., M.S.C.E., and Anthony O.
ODIBO, M.D., M.S.C.E.
Department of Obstetrics & Gynecology, Washington University in St. Louis

Abstract
Objective—To estimate the efficiency of first-trimester uterine artery Doppler, A-disintegrin and
metalloprotease 12 (ADAM12), pregnancy-associated plasma protein A (PAPP-A) and maternal
NIH-PA Author Manuscript

characteristics in the prediction of pre-eclampsia.

Methods—This is a prospective cohort study of patients presenting for first-trimester aneuploidy
screening between 11-14 weeks’ gestation. Maternal serum ADAM12 and PAPP-A levels were
measured by immunoassay, and mean uterine artery Doppler pulsatility indices (PI) were
calculated. Outcomes of interest included pre-eclampsia, early pre-eclampsia, defined as requiring
delivery at <34 weeks’ gestation, and gestational hypertension. Logistic regression analysis was
used to model the prediction of pre-eclampsia using ADAM12 multiples of the median (MoM),
PAPP-A MoM, and uterine artery Doppler PI MoM, either individually or in combination.
Sensitivity, specificity, and area under the receiver-operating characteristic curves (AUC) were
used to compare the screening efficiency of the models using non-parametric U-statistics.
Results—Of 578 patients with complete outcome data, there were 54 (9.3%) cases of
preeclampsia and 13 (2.2%) cases of early pre-eclampsia. Median ADAM12 levels were
significantly lower in patients who developed pre-eclampsia compared to those who did not. (0.81
v. 1.01 MoMs; p<0.04) For a fixed false positive rate (FPR) of 10%, ADAM12, PAPP-A, and
uterine artery Doppler in combination with maternal characteristics identified 50%, 48%, and 52%
of patients who developed pre-eclampsia, respectively. Combining these first-trimester parameters
did not improve the predictive efficiency of the models.
NIH-PA Author Manuscript

Conclusion—First-trimester ADAM12, PAPP-A, and uterine artery Doppler are not sufficiently
predictive of pre-eclampsia. Combinations of these parameters do not further improve their
screening efficiency.

Keywords
ADAM12; PAPP-A; placental dysfunction; pre-eclampsia; uterine artery Doppler

Introduction
Pre-eclampsia affects 5-8% of pregnancies and remains a significant contributor to perinatal
morbidity and mortality worldwide.1,2 Abnormal invasion of the placental trophoblast into

Corresponding Author: Katherine R. Goetzinger, M.D., M.S.C.I. Department of Obstetrics & Gynecology Washington University
School of Medicine 4911 Barnes-Jewish Hospital Plaza Campus Box 8064 St. Louis, MO 63110 Phone: 314-362-8895; Fax:
314-747-1720 goetzingerk@wudosis.wustl.edu.
 GOETZINGER et al. Page 2

the maternal spiral arterioles, as early as the first trimester, is the proposed pathophysiologic
mechanism for the development of this disorder. This abnormal placentation results in a
high resistance uteroplacental circulation bed and persistent placental underperfusion,
NIH-PA Author Manuscript

leading to the phenotype of pre-eclampsia.

The use of biochemical and ultrasound parameters to detect these pathologic changes and
identify patients at high risk for developing pre-eclampsia has been an area of intense
research focus over recent years. Prior studies have demonstrated an association between
low levels of pregnancy-associated plasma protein A (PAPP-A), a serum analyte routinely
measured as part of first-trimester aneuploidy screening, and the subsequent development of
pre-eclampsia later in pregnancy. However, PAPP-A alone has demonstrated only a modest
predictive efficiency for adverse pregnancy outcomes, including pre-eclampsia.3-6 More
recent studies have evaluated the association between low levels of a novel serum analyte, A
Disintegrin and Metalloprotease 12 (ADAM12), and pre-eclampsia and have produced
conflicting results.7-9 These studies mainly have been limited to case-control designs
performed in populations with a low prevalence of preeclampsia, and, therefore, may be
subject to bias. Additionally, these studies have primarily been conducted in low risk
populations, thereby precluding a rigorous evaluation of maternal risk factors which may
increase the screening efficiency of these biomarkers. Finally, uterine artery Doppler studies
have been proposed as a screening tool to detect increased vascular resistance in patients
destined to develop pre-eclampsia, with reported sensitivites ranging from 7-80% when
NIH-PA Author Manuscript

performed in the first trimester.10-15

The objective of this study was to estimate the efficiency of first-trimester ADAM12, PAPP-
A, uterine artery Doppler and maternal characteristics, both individually and in combination,
in the prediction of pre-eclampsia. We hypothesize that a combination of these first-
trimester parameters, in addition to maternal clinical risk factors, may allow for more
accurate identification of women at risk. Although there is currently no known effective
strategy for the prevention of pre-eclampsia, the ability to predict patients at high risk of
developing this disorder not only may provide an opportunity to impact pregnancy
management but may also provide an opportunity to identify an enriched population of
patients who can serve as target subjects for future intervention studies.

Materials and Methods

This was a prospective cohort study of patients presenting to Washington University
Medical Center from 2008-2010 for first-trimester aneuploidy screening. Women with
singleton pregnancies between 11 and 14 weeks’ gestation were eligible for inclusion.
Exclusion criteria were known aneuploidy and major congenital malformations. Institutional
NIH-PA Author Manuscript

review board approval was obtained, and all patients provided written, informed consent.

All consecutive eligible patients were approached for participation in the study at the time of
ultrasound exam. Standard of care for first-trimester aneuploidy screening at our institution
includes a measurement of fetal crown-rump length to confirm pregnancy dating (within ±7
days from menstrual dating), nuchal translucency measurement, and serum measurement of
PAPP-A and free β-human chorionic gonadotropin (β-hCG). Serum measurements of both
PAPP-A and free β-hCG are routinely performed by Perkin Elmer laboratories (Melville,
NY) for all samples collected at our institution. Patients who consented to study
participation provided an additional 10ml of blood which was used to measure ADAM12
concentration. Maternal blood was collected into non-heparinized tubes and centrifuged at
1500 g for 15 minutes. Maternal serum was then extracted and stored at -80° C until
analyzed. A 25μl aliquot of maternal serum was used to determine ADAM12 concentration
through a time-resolved fluorescent immuno-assay, in which the concentration of ADAM12

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 3

was directly proportional to the fluorescence measured at 615nm. (DELFIA/AutoDELFIA

ADAM12 research kit, PerkinElmer Life and Analytical Sciences, Turku, Finland) All
serum analyte levels were converted into multiples of the median (MoMs), adjusted for
NIH-PA Author Manuscript

gestational age, for analysis.

Patients who consented to the study also underwent bilateral uterine artery Doppler
assessment. This assessment was performed using a transabdominal approach with color
flow mapping. A mid-sagittal view of the uterus was obtained and the cervical canal
identified. The transducer was then rotated until the paracervical vessels were identified.
Each uterine artery was then isolated, and the pulsatility index (PI) measured and averaged.
These measurements were also converted into MoMs, adjusted for gestational age. All
participating sonographers were certified by the Fetal Medicine Foundation for first-
trimester Doppler measurements.

Maternal demographics, past medical history, and obstetrical history were obtained through
a detailed questionnaire routinely administered at the time of all initial ultrasound exams in
our unit. Delivery outcome information was obtained through electronic medical record
review by a dedicated nurse coordinator. Patients who delivered outside of our institution
signed a consent for release of medical records at the time of study enrollment.

The primary outcome for this study was pre-eclampsia defined as systolic blood pressure
NIH-PA Author Manuscript

>140 mmHg or diastolic blood pressure >90 mmHg on at least two occasions separated by
at least 4 hours in the presence of proteinuria (≥0.3 grams in a 24 hour specimen or ≥1+
protein on urine dipstick) after 20 weeks’ gestation. Secondary outcomes included early pre-
eclampsia, defined as pre-eclampsia requiring delivery <34 weeks, and gestational
hypertension (HTN), defined as blood pressure >140/90 mmHg in the absence of proteinuria
after 20 weeks’ gestation.1,16

Baseline maternal characteristics as well as ADAM12, PAPP-A, and uterine artery PI MoMs
were compared between patients who developed pre-eclampsia and those who did not.
Categorical variables were compared using chi-square tests, and continuous variables were
compared using the independent sample t-test and Mann-Whitney U test, as appropriate.
Normality of distribution was evaluated using the Kolmogorov-Smirnov test. Logistic
regression was then used to model the prediction of pre-eclampsia incorporating various
combinations of first-trimester parameters as well as maternal factors identified as
significant in the univariate analysis. There was no evidence of collinearity between the
first-trimester parameters and maternal characteristics of interest, thereby justifying their
inclusion together in the various models. Receiver-operating characteristic curves (ROC)
were generated for each model, and the area under the curve (AUC) was compared between
NIH-PA Author Manuscript

each model using non-parametric U statistics. Sensitivity and specificity at both 10% and
20% fixed false positive rates (FPR) were also calculated for each model. P-values <0.05
were considered statistically significant. All statistical analyses were performed using
STATA 12, Special Edition (College Station, TX).

Since the focus of this study was on the development of a highly sensitive prediction model
for pre-eclampsia, the precision of our sample size estimates was based on the ½ width of
the 95% confidence interval and the incidence of pre-eclampsia in our patient population.
With these assumptions, our study was powered to produce a prediction model with a 70%
(95% CI 57-83) sensitivity for pre-eclampsia and a 90% sensitivity (95% CI 75-100) for
early preeclampsia.

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 4

Results
Of 618 patients enrolled, 13 were lost to follow up and 3 withdrew from the study before
NIH-PA Author Manuscript

completion. After excluding patients who underwent spontaneous abortion (n=8), elective
abortion (n=1), and those with incomplete outcome information (n=15), 578 patients were
available for analysis, comprising our final study cohort. Of these patients, 54 (9.3%)
developed pre-eclampsia, 13 (2.2%) developed early pre-eclampsia, and 55 (9.5%)
developed gestational HTN. The mean gestational age at the time of ultrasound and study
enrollment was 12.1 ± 0.6 weeks. Compared to patients who did not develop pre-eclampsia,
patients who went on to develop pre-eclampsia were more likely to be African American,
have a higher pre-pregnancy body mass index (BMI), and have a higher incidence of both
chronic HTN and pre-gestational diabetes. (Table 1)

Patients who developed pre-eclampsia had significantly lower ADAM12 levels (0.81 vs.
1.01; p=0.04) and PAPP-A levels (0.88 vs. 1.18, p<0.001) compared to controls; however,
there was no significant difference in uterine artery Doppler PI levels in pre-eclamptic
patients compared to controls (1.00 vs. 0.99, p=0.77). There was no significant difference in
ADAM12, PAPP-A, or uterine artery Doppler PI levels when comparing patients who
developed early preeclampsia or gestational HTN compared to those who did not. (Table 2)

Individually, ADAM12, PAPP-A, and uterine artery Dopplers were not sufficiently
NIH-PA Author Manuscript

predictive of pre-eclampsia, resulting in AUCs ranging from 0.49-0.64. In fact, maternal

characteristics alone (AUC 0.78, 95% CI 0.71-0.75) were significantly more predictive of
preeclampsia than any first-trimester parameter alone. (ADAM12 alone, p<0.001; PAPP-A
alone, p=0.01; uterine artery Doppler alone, p<0.001) Combining these first-trimester
parameters with maternal characteristics did improve the predictive efficiency of the models.
The best overall predictive efficiency was observed with the combination of maternal
characteristics, ADAM12, PAPP-A and uterine artery Doppler with an AUC of 0.79 (95%
CI 0.71-0.87) and sensitivities of 50% and 75% at 10% and 20% FPR, respectively.
However, this predictive efficiency still was not statistically different from the model
containing maternal characteristics alone (p=0.77). (Table 3) The significant independent
predictors of pre-eclampsia in this model included PAPPA, BMI, history of chronic
hypertension and history of pre-gestational diabetes.

Similar results were observed for the prediction of early pre-eclampsia and gestational HTN.
ADAM12, PAPP-A and uterine artery Doppler alone were not predictive for the
development of early pre-eclampsia or gestational HTN. Combining these parameters with
maternal risk factors did improve the predictive efficiencies; however, models containing
the first-trimester parameters did not differ significantly from those containing maternal
NIH-PA Author Manuscript

characteristics alone. The best overall predictive efficiency for early pre-eclampsia was
observed with the combination of maternal characteristics, ADAM12 and PAPP-A with an
AUC of 0.78 (95% CI 0.63-0.93) and sensitivities of 54% and 62% at 10% and 20% FPR,
respectively. (Table 4) The best overall predictive efficiency for gestational HTN was
observed with the combination of maternal characteristics, ADAM12, PAPP-A and uterine
artery Doppler with an AUC of 0.66 (95% CI 0.58-0.74) and sensitivities of 15% and 28% at
10% and 20% FPR, respectively. (Table 5) The only significant independent predictor of
early pre-eclampsia in this model was a maternal history of chronic hypertension.

In order to evaluate the predictive indices of our model, we created two hypothetical
populations, one with a low prevalence of disease (1%) and one with a high prevalence of
disease (10%). Using the calculated sensitivities of our models at a 10% FPR, positive
(PPV) and negative predictive values (NPV) were estimated for our models containing
maternal characteristics alone as well as the combination of maternal characteristics with

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 5

ADAM12 levels, PAPP-A levels, and uterine artery Doppler. For the outcome of pre-
eclampsia, the PPV and NPV were 4.8% and 99.4%, respectively, in the low prevalence
population and 35.7% and 94.2%, respectively, in the high prevalence population for
NIH-PA Author Manuscript

maternal characteristics alone. For the outcome of early pre-eclampsia, the PPV and NPV
were 5.3% and 99.5% in the low prevalence population and 37.9% and 94.7% in the high
prevalence population for maternal characteristics alone. Finally, for the outcome of
gestational hypertension, the PPV and NPV were 1.5% and 99.0% in the low prevalence
population and 14.3% and 90.5% in the high prevalence population for maternal
characteristics alone. When evaluating the models containing the combination of maternal
characteristics and first-trimester markers of placental dysfunction, the predictive indices
were identical to the models containing maternal characteristics alone.

Discussion
Findings from our prospective cohort demonstrate that both ADAM12 and PAPP-A levels
are significantly reduced in patients who develop pre-eclampsia. Similar to PAPP-A,
ADAM12 is a protease for insulin growth factor (IGF) binding proteins. Low levels of this
analyte reflect an increased amount of IGF in the bound state which is then unavailable to
promote placental growth and development, making this finding biologically plausible.17,18
However, despite these associations, the predictive efficiency of ADAM12 and PAPP-A was
overall modest and not sufficient for clinical use. In fact, our findings suggest that maternal
NIH-PA Author Manuscript

characteristics alone actually demonstrate superior test performance characteristics for the
prediction of pre-eclampsia which are not enhanced by the addition of these first-trimester
markers, either individually or in combination.

Prior studies have been conflicting regarding the role of ADAM12 in the prediction of pre-
eclampsia. Both Laigaard et al. and Spencer et al. demonstrated decreased serum
concentrations of ADAM12 in the first-trimester in women who went on to develop
preeclampsia.7,8 Consistent with our results, Spencer et al. also demonstrated only a modest
predictive efficiency of ADAM12 for pre-eclampsia with an AUC of 0.694 for ADAM12
alone and an AUC of 0.714 when ADAM12 and PAPP-A were combined.8 Alternatively,
Poon et al. found that first-trimester ADAM12 levels were not significantly lower in patients
who developed pre-eclampsia compared to controls.9 They attributed these negative findings
to a priori adjustment of ADAM12 levels for race and maternal weight. When these
maternal characteristics were evaluated in our population, we found no relationship between
ADAM12 and maternal race. While there was a modest correlation between ADAM12
levels and maternal weight, we adjusted for this factor in our prediction model as an
established maternal clinical risk factor for pre-eclampsia.
NIH-PA Author Manuscript

Our study did not demonstrate any significant difference in uterine artery PI measurements
between pre-eclamptic and control patients. Given that maximal trophoblast invasion occurs
during the first-trimester, it would seem justified that enhanced vascular resistance in the
uterine arteries would be detectable at this early stage of gestation in patients with impaired
placentation.19 Recently, Parra-Cordero et al. demonstrated a significant increase in first-
trimester uterine artery Doppler PI in patients who developed both early-onset and late-onset
pre-eclampsia compared to controls. In this study, the sensitivity of maternal history plus
uterine artery Doppler for detecting early-onset and late-onset pre-eclampsia were 43.8%
and 28.3%, respectively, at a 10% FPR.20 While our study did not demonstrate a significant
difference in absolute uterine artery PI values, the sensitivity of maternal characteristics and
uterine artery Doppler were similar to that observed by Parra-Cordero et al. at 54% and
52%, respectively, at a 10% FPR. Again, this finding suggests that the predictive value of
these first-trimester parameters may be primarily driven by the contribution of maternal
characteristics alone. Additionally, Poon et al. demonstrated that first-trimester uterine

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 6

artery Doppler studies were significantly increased in patients who developed hypertensive
disorders later in pregnancy.21 Differences between these studies and our study may be due
to the varying criteria which was used to define an abnormal uterine artery Doppler PI. Both
NIH-PA Author Manuscript

of the above-mentioned studies measured the right and left uterine artery PI and used the
lowest recorded value in their analysis. Alternatively, our study used the average value of
the right and left uterine artery PI measurements, a technique which has been previously
described.12,14 We believe that this average value is more representative of overall placental
perfusion. This discrepancy between studies further supports the claim that first-trimester
uterine artery Doppler may not be a reliable predictor of pre-eclampsia, as we would have
expected to observe more consistent results across studies regardless of the criteria used to
define an abnormal value. Furthermore, differences in study populations as well as
variability in the maternal factors included in the prediction models may also account for
some of the discrepancy observed between studies.

Strengths of our study include its prospective cohort design and low loss to follow up rate.
This study design allowed us to ensure that our unaffected patients were derived from the
same population as our affected patients, thereby eliminating the bias which is often
introduced in case-control studies through the process by which controls patients are
selected. Additionally, given that our study was performed in a tertiary referral center, there
was adequate representation of maternal co-morbidities which could then be rigorously
evaluated as maternal risk factors for the development of pre-eclampsia. Our study is not
NIH-PA Author Manuscript

without limitations. Despite our large cohort, the small number of early pre-eclampsia cases
(n=13) left us under-powered to thoroughly evaluate this outcome. Additionally, while we
did adjust for gestational age, we did not perform a priori adjustment of ADAM12 levels for
other maternal characteristics. However, data in the literature as well as in our own
population has been conflicting as to which factors truly have a significant effect on this
analyte. Given that these proposed adjustment factors are also well-established risk factors
for pre-eclampsia, we instead took the approach of controlling for these factors in our
prediction model in order to provide a more individualized risk assessment based on
maternal characteristics and history. Furthermore, our medical center is a high-risk, tertiary
referral center which likely contributes to the high prevalence of preeclampsia in our study
population; however, this high prevalence of pre-eclampsia should not affect the overall
accuracy of our prediction model, given that sensitivity and specificity are independent of
prevalence. Finally, the methods used in this study are aimed at population-based screening
for hypertensive disorders in pregnancy. These risk factors may still play a significant role in
individualized patient risk assessment and counseling.

In conclusion, our study demonstrates that while first-trimester ADAM12 and PAPP-A
levels are significantly reduced in patients who develop pre-eclampsia, the predictive
NIH-PA Author Manuscript

efficiency of these first-trimester serum markers as well as uterine artery Doppler studies,
both individually and in combination, are not superior to that achieved by maternal
characteristics alone. Continued investigation into novel serum and ultrasound markers of
placental dysfunction is warranted in order to more accurately identify this high risk
population.

Acknowledgments
1. This paper was presented as an oral presentation at the American Institute of Ultrasound in Medicine 2012
Annual Convention in Phoenix, Arizona on March 31, 2012 and was the winner of the New Investigator Award in
Clinical Ultrasound.

2. Dr. Goetzinger is supported by a training grant from the Eunice Kennedy Shriver National Institute of Child
Health and Human Development (5 T32 HD055172) and from a NIH/NCRR/NCATS Washington University ICTS

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 7

grant (UL1 RR024992). The contents of this publication are solely the responsibility of the authors and do not
necessarily represent the official view of the NCRR, NIH, or NCATS.
NIH-PA Author Manuscript

References
1. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin #33. American
College of Obstetricians and Gynecologists; Washington DC: 2002. Diagnosis of management of
preeclampsia and eclampsia.
2. Confidential Enquiry into Maternal and Child Health (CEMACH). England, Wales and Northern
Ireland. CEMACH; London: 2008. Perinatal Mortality.
3. Dugoff L, Hobbins JC, Malone FD, et al. First-trimester maternal serum PAPP-A and free beta
subunit human chorionic gonadotropin concentrations and nuchal translucency are associated with
obstetric complications: A population based screening study (The FASTER Trial). Am J Obstet
Gynecol. 2004; 191:1446–1451. [PubMed: 15507981]
4. Smith GCS, Stenhouse EJ, Crossley JA, et al. 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. [PubMed: 11932314]
5. Ong CYT, Liao AW, Spencer K, et al. First trimester maternal serum free β human chorionic
gonadotrophin and pregnancy associated plasma protein A as predictors of pregnancy
complications. BJOG. 2000; 107:1265–1270. [PubMed: 11028579]
6. Goetzinger KR, Singla A, Gerkowicz S, Dicke JM, Gray DL, Odibo AO. Predicting the risk of pre-
eclampsia between 11 and 13 weeks’ gestation by combining maternal characteristics and serum
NIH-PA Author Manuscript

analytes, PAPP-A and free β-hCG. Prenat Diagn. 2010; 30:1138–1142. [PubMed: 20936638]
7. Laigaard J, Sorensen T, Placing S, et al. Reduction of the disintegrin and metalloprotease ADAM12
in preeclampsia. Obstet Gynecol. 2005; 106:144–149. [PubMed: 15994630]
8. Spencer K, Cowans NJ, Stamatopoulou A. ADAM12s in maternal serum as a potential marker of
pre-eclampsia. Prenat Diagn. 2008; 28:212–216. [PubMed: 18264967]
9. Poon LCY, Chelemen T, Granvillano O, Pandeva I, Nicolaides KH. First-trimester maternal serum a
disintegrin and metalloprotease (ADAM12) and adverse pregnancy outcome. Obstet Gynecol. 2008;
112:1082–1090. [PubMed: 18978109]
10. 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. [PubMed: 11844193]
11. Parra M, Rodrigo R, Barja P, et al. Screening test for preeclampsia through assessment of
uteroplacental blood flow and biochemical markers of oxidative stress and endothelial
dysfunction. Am J Obset Gynecol. 2005; 193:1486–1491.
12. Gomez O, Martinez JM, Figueras F, et al. 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. [PubMed: 16184511]
13. Vainio M, Kujansuu E, Koivisto AM, Maenpaa J. Bilateral notching of uterine arteries at 12-14
NIH-PA Author Manuscript

weeks of gestation for prediction of hypertensive disorders of pregnancy. Acta Obstet Gynecol
Scand. 2005; 84:1062–1067. [PubMed: 16232173]
14. Pilalis A, Souka AP, Antsaklis P, et al. 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. [PubMed: 17221926]
15. Melchiorre K, Wormald B, Leslie K, Bhide A, Thilaganathan B. First-trimester uterine artery
Doppler indices in term and preterm pre-eclampsia. Ultrasound Obstet Gynecol. 2008; 32:133–
137. [PubMed: 18615872]
16. Report of the National High Blood Pressure Education Program Working Group on High Blood
Pressure in Pregnancy. Am J Obstet Gynecol. 2000; 183:S1–S22.
17. Loechel F, Fox JW, Murphy G, Albrechsten R, Wewer UM. ADAM12-S cleaves IGFBP-3 and
IGFBP-5 and is inhibited by TIMP-3. Biochem Biophys res Commun. 2000; 278:511–515.
[PubMed: 11095942]

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 8

18. Shi Z, Xu W, Loechel F, Wewer UM, Murphy LJ. ADAM12, a disintegrin metalloprotease,
interacts with insulin-like growth factor-binding protein-3. J Biol Chem. 2000; 275:18574–18580.
[PubMed: 10849447]
NIH-PA Author Manuscript

19. Caniggia I, Winter J, Lye SJ, Post M. Oxygen and placental development during the first trimester:
implications for the pathophysiology of pre-eclampsia. Placenta. 2000; 21:S25–S30. [PubMed:
10831118]
20. Parra-Cordero M, Rodrigo R, Barja P, et al. Prediction of early and late pre-eclampsia from
maternal characteristics, uterine artery Doppler and markers of vasculogenesis during the first-
trimester of pregnancy. Ultrasound Obstet Gynecol. [published online ahead of print July 17,
2012]. Doi: 10.1002/uog.12264.
21. Poon LCY, Staboulidou I, Maiz N, Plasencia W, Nicolaides KH. Hypertensive disorders in
pregnancy: screening by uterine artery Doppler at 11-13 weeks. Ultrasound Obstet Gynecol. 2009;
34:142–148. [PubMed: 19644947]
NIH-PA Author Manuscript
NIH-PA Author Manuscript

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 9

Table 1
Baseline maternal characteristics and obstetrical history
NIH-PA Author Manuscript

Pre-Eclampsia Control
p-value
(n=54) (n=518)

Maternal Age (years)* 30.2 ± 6.1 31.3 ± 5.8 0.23

Race

Caucasian 42.6% 57.4%

African American 48.1% 28.1%

0.03
Hispanic 0% 2.6%

Asian 5.6% 9.0%

Other 3.7% 2.9%

Body Mass Index (kg/m2)* 34.0 ± 9.2 28.1 ± 7.4 <0.001

Nulliparity 20.4% 27.5% 0.26

Tobacco Use 13.2% 8.7% 0.27

Chronic Hypertension 38.9% 6.7% <0.001

Pre-gestational Diabetes 22.2% 5.6% <0.001

NIH-PA Author Manuscript

*
Data expressed as mean ± standard deviation
NIH-PA Author Manuscript

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 10

Table 2
Comparison of First-Trimester Marker Levels Between Study Groups
NIH-PA Author Manuscript

Pre-Eclampsia

Pre-Eclampsia Control
Marker p-value
(n=54) (n=518)
0.81 1.01
ADAM12 MoM 0.04
(0.61-1.30) (0.74-1.36)

0.88 1.18
PAPP-A MoM <0.001
(0.62-1.27) (0.60-1.63)

Uterine Artery 1.00 0.99

0.77
Doppler PI MoM (0.85-1.22) (0.79-1.23)

Early Pre-Eclampsia

Early Pre-Eclampsia Control

Marker p-value
(n=13) (n=565)
0.82 1.00
ADAM12 MoM 0.11
(0.57-1.21) (0.73-1.35)

1.02 1.14
PAPP-A MoM 0.55
(0.73-1.53) (0.76-1.58)

Uterine Artery 1.00 1.00

0.22
NIH-PA Author Manuscript

Doppler PI MoM (0.79-1.22) (0.89-1.43)

Gestational HTN

Gestational HTN Control

Marker p-value
(n=55) (n=523)
0.81 1.01
ADAM12 MoM 0.07
(0.63-1.22) (0.73-1.35)

1.18 1.13
PAPP-A MoM 0.51
(0.82-1.73) (0.76-1.58)

Uterine Artery 0.96 1.01

0.10
Doppler PI MoM (0.78-1.13) (0.80-1.24)

*
Data expressed as medians and interquartile ranges
*
HTN=hypertension; MoM=multiples of the median
NIH-PA Author Manuscript

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 11

Table 3
Predictive Efficiency of First-Trimester Parameters for Pre-Eclampsia
NIH-PA Author Manuscript

Marker AUC Sensitivity Sensitivity

(95% CI) 10% FPR 20% FPR

Maternal Characteristics Alone 0.78

50% 62%
(0.71-0.85)

ADAM12 Alone 0.58

12% 30%
(0.50-0.67)

PAPP-A Alone 0.64

18% 32%
(0.57-0.72)

Uterine Artery Doppler Alone 0.49

5% 16%
(0.41-0.56)

ADAM12 +PAPP-A + Uterine Artery 0.64

22% 42%
Doppler Alone (0.57-0.72)

Maternal Characteristics + ADAM12 0.78

50% 62%
(0.70-0.85)

Maternal Characteristics + PAPP-A 0.79

48% 64%
(0.72-0.86)

Maternal Characteristics + Uterine 0.77

52% 64%
Artery Doppler (0.69-0.86)
NIH-PA Author Manuscript

Maternal Characteristics + ADAM12 + 0.79

50% 70%
PAPP-A (0.71-0.86)

Maternal Characteristics + ADAM12 + 0.78

52% 62%
Uterine Artery Doppler (0.70-0.86)

Maternal Characteristics + ADAM12 + 0.79

50% 75%
PAPP-A + Uterine Artery Doppler (0.71-0.87)

*
Maternal characteristics: African American race, body mass index, history of chronic hypertension, history of pre-gestational diabetes
NIH-PA Author Manuscript

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 12

Table 4
Predictive Efficiency of First-Trimester Parameters for Early Pre-Eclampsia
NIH-PA Author Manuscript

Marker AUC Sensitivity Sensitivity

(95% CI) 10% FPR 20% FPR

Maternal Characteristics Alone 0.71

55% 58%
(0.54-0.88)

ADAM12 Alone 0.63

22% 30%
(0.46-0.80)

PAPP-A Alone 0.55

16% 22%
(0.38-0.71)

Uterine Artery Doppler Alone 0.60

10% 38%
(0.44-0.75)

ADAM12 +PAPP-A + Uterine Artery 0.65

35% 46%
Doppler Alone (0.48-0.82)

Maternal Characteristics + ADAM12 0.77

54% 62%
(0.620.93)

Maternal Characteristics + PAPP-A 0.75

54% 54%
(0.60-0.90)

Maternal Characteristics + Uterine 0.71

54% 62%
Artery Doppler (0.53-0.90)
NIH-PA Author Manuscript

Maternal Characteristics + ADAM12 + 0.78

54% 62%
PAPP-A (0.63-0.93)

Maternal Characteristics + ADAM12 + 0.77

54% 62%
Uterine Artery Doppler (0.61-0.93)

Maternal Characteristics + ADAM12 + 0.77

54% 62%
PAPP-A + Uterine Artery Doppler (0.62-0.93)

*
Maternal characteristics: African American race and history of chronic hypertension
NIH-PA Author Manuscript

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.

 GOETZINGER et al. Page 13

Table 5
Predictive Efficiency of First-Trimester Parameters for Gestational Hypertension
NIH-PA Author Manuscript

Marker AUC Sensitivity Sensitivity

(95% CI) 10% FPR 20% FPR

Maternal Characteristics Alone 0.62

15% 22%
(0.55-0.70)

ADAM12 Alone 0.57

15% 32%
(0.49-0.66)

PAPP-A Alone 0.53

20% 26%
(0.44-0.61)

Uterine Artery Doppler Alone 0.57

16% 22%
(0.49-0.64)

ADAM12 +PAPP-A + Uterine Artery 0.59

12% 20%
Doppler Alone (0.52-0.66)

Maternal Characteristics + ADAM12 0.62

12% 36%
(0.53-0.72)

Maternal Characteristics + PAPP-A 0.63

12% 28%
(0.55-0.71)

Maternal Characteristics + Uterine 0.64

8% 30%
Artery Doppler (0.56-0.71)
NIH-PA Author Manuscript

Maternal Characteristics + ADAM12 + 0.63

14% 42%
PAPP-A (0.54-0.73)

Maternal Characteristics + ADAM12 + 0.65

10% 28%
Uterine Artery Doppler (0.57-0.73)

Maternal Characteristics + ADAM12 + 0.66

15% 28%
PAPP-A + Uterine Artery Doppler (0.58-0.74)

*
Maternal characteristics: African American race and body mass index
NIH-PA Author Manuscript

J Ultrasound Med. Author manuscript; available in PMC 2014 September 01.