Objective
To evaluate the screening accuracy of pregnancy hypertensive disorders by maternal serum biomarkers and uterine artery Doppler in the first trimester.
Study Design
Prospectively enrolled nulliparous women had uterine artery Doppler and serum measured at 11-13 weeks. Maternal characteristics, uterine artery Doppler, and serum placental biomarkers (pregnancy-associated plasma protein-A, Inhibin-A, placental protein 13, A disintegrin and metalloprotease 12, free β-hCG, placental growth factor) were recorded.
Results
Among 893 women, 20 (2.2%) had gestational hypertension developed and 40 (4.5%) had preeclampsia developed, including 9 (1.0%) early-onset preeclampsia and 16 (1.8%) severe preeclampsia. A combined screening model with clinical characteristics, pregnancy-associated plasma protein-A, Inhibin-A, and placental growth factor could detect 75% of early-onset preeclampsia at a 10% false-positive rate. After adjustment for clinical variables, uterine artery Doppler, placental protein 13, and A disintegrin and metalloprotease 12 did not improve the diagnostic accuracy.
Conclusion
A combination of clinical characteristics and first-trimester maternal serum biomarkers (pregnancy-associated plasma protein-A, Inhibin-A, and placental growth factor) provides an accurate screening for early-onset preeclampsia in nulliparous women.
Preeclampsia, a disease affecting about 3% of pregnant women, contributes substantially to maternal and fetal morbidity and mortality worldwide. Estimating each woman’s individual risk would allow appropriate antenatal surveillance, and would also enable to test preventive strategies such as low-dose aspirin in selected high-risk groups. However, the use of prophylactic treatments is likely to be more beneficial when started earlier in pregnancy, ideally before 16 weeks. It would be thus important to develop an effective method of early identification of high-risk groups. Currently, women at risk are mainly identified based on clinical history. Nulliparity is one of the major clinical risk factors for the development of preeclampsia. Other relevant factors include an increased body mass index and other medical conditions such as prepregnancy diabetes, previous preeclampsia, or chronic hypertension. However, screening by maternal history alone will detect only 30% of women who will have preeclampsia develop. The clinical risk-based strategy is not effective for nulliparous women without other risk factors. The addition of maternal serum markers and uterine artery Doppler (uAD) could improve the prediction of the disease in this population. Several serum markers have been suggested for first-trimester screening of preeclampsia, including free β-hCG, pregnancy-associated plasma protein-A (PAPP-A), placental protein 13 (PP13), Inhibin-A, placental growth factor (PlGF), and A disintegrin and metalloprotease 12 (ADAM12). Although uAD screening has been mainly studied in the second trimester of pregnancy, first-trimester Doppler has more recently shown some promise. A systematic review of screening tests for preeclampsia concluded that no single test is yet available to provide a good diagnostic accuracy. A combined screening involving several relevant markers is more likely to provide the best prediction. First-trimester screening would represent a major advantage over a second-trimester approach because it opens prospects for early and more efficient interventions. The aim of this study was to analyze, in a group of nulliparous women, the diagnostic accuracy of several relevant biomarkers for the detection of subsequent preeclampsia in the first trimester of pregnancy, associated with maternal characteristics and uAD.
Materials and Methods
In this prospective cohort study, we recruited 1000 pregnant women at the time of screening for Down syndrome at 11-13 weeks, between November 2006 and June 2008. Multiparous women, multiple gestations, and pregnancies with a major fetal chromosomal or structural anomaly were excluded from further analysis. Written informed consent was obtained from all women before enrolment in the study, and the project was approved by the institutional ethics committee.
At the time of inclusion, women had an interview with a research nurse and answered a standardized questionnaire on maternal age, ethnic origin, smoking during pregnancy, and medical conditions. The maternal weight and height were measured and the body mass index (BMI) was calculated. An ultrasound examination was carried out for diagnosis of major fetal defects, measurement of nuchal translucency thickness, and crown-rump length (CRL), which was used to determined gestational age. At the same visit both uterine arteries were examined by transabdominal Doppler velocimetry using a standardized technique and performed by trained sonographers. Each uterine artery was identified by color flow mapping, at the level of the apparent crossover with the external iliac vessels. This approach was used because it is similar to the well-validated second-trimester technique for uAD. Once ensured that the angle of insonation was less than 50 degrees, pulsed-wave Doppler was used to cover the whole width of the vessel. The signal was updated until at least 6 clear similar consecutive waveforms were obtained. The pulsatility index (PI) was measured from both uterine arteries, and we recorded the mean, lowest, and highest PI of both sides, and the presence of a protodiastolic notch. The lowest PI was suspected to better reflect the placental function than the mean PI, as previously described. The results of first trimester Doppler were not communicated to the physicians in charge of the follow-up and were not recorded in the ultrasound report.
Maternal nonfasting blood samples were collected, immediately centrifuged for 10 minutes at 4000 rpm, divided into 4 aliquots, and frozen at −80°C until used for the analyses. Free β-hCG and PAPP-A were, respectively, measured with fully automated solid-phase, enzyme-labeled chemiluminescent immunometric and solid-phase, 2-site chemiluminescent immunometric assays on the IMMULITE 2000 (Siemens Medical Solutions Diagnostics, Los Angeles, CA). Total coefficients of variation (CV) were 3.8% and 6.0% at 0.4 IU/L and 4.2 IU/L for PAPP-A, and 4.5% and 4.8% at 10 μg/L and 312 μg/L for free β-hCG. Inhibin-A was measured by an automated 2-site chemiluminescent immunometric assay on the Access-2 (Beckman-Coulter, Chaska, MN). Total imprecisions at 51 ng/L and 331 ng/L were 5.9% and 4.9%, respectively.
PP13, PlGF, and ADAM12 in maternal serum were measured with DELFIA kits (AutoDELFIA PP13 RUO 4062-0010, DELFIA Xpress PlGF 6007-0010, and AutoDELFIA ADAM12 RUO 4025-0010, respectively) obtained from Perkin-Elmer Life and Analytical Sciences (Turku, Finland). The total variations of the assays were 6.6%, 4.8%, and 4.3%, respectively. PlGF could only be measured in 531 women, because this test was initially not scheduled and additional serum was not available for women included in the first year of the study.
The laboratory professionals who performed the biochemical assays were blinded to the clinical status of subjects, and the results were not revealed to the clinicians, except for the first-trimester free β-hCG and PAPP-A levels, routinely reported for Down syndrome screening.
Data on pregnancy outcomes were collected from the hospital medical charts. The records of all women with hypertension were reviewed by 2 independent investigators blinded to the first-trimester test results to determine whether the final diagnosis was chronic hypertension, preeclampsia, or gestational hypertension. For quality control, we examined the records of 50 randomly selected women without any hypertensive disorder during pregnancy. The definitions of preeclampsia and gestational hypertension were those of the International Society for the Study of Hypertension in Pregnancy. Preeclampsia was defined as resting blood pressure ≥140/90 mm Hg on 2 occasions at least 4 hours apart and proteinuria ≥0.3 g/d after 20 weeks in previously normotensive women. Gestational hypertension was defined as hypertension in pregnancy without proteinuria. Early-onset preeclampsia was defined as preeclampsia diagnosed before 34 weeks. Severe preeclampsia was defined as blood pressure ≥160/110 mm Hg, proteinuria ≥5 g/d, or the presence of an adverse condition, including maternal symptoms, maternal signs of end-organ dysfunction, abnormal maternal laboratory testing, or fetal compromise, as described elsewhere.
For statistical analysis, the following steps were taken:
- 1
The women were subdivided into 3 groups depending on pregnancy outcome: preeclampsia, gestational hypertension, and unaffected by any hypertensive disorders. Cases of severe preeclampsia and early-onset preeclampsia were separately identified in the preeclampsia group.
- 2
The distributions of uAD indices and biomarker levels were made Gaussian after logarithmic transformation and the normality of the distributions was confirmed by Skewness and Kurtosis tests.
- 3
We used multiple linear regressions in the unaffected group to obtain expected log uAD indices and biomarker levels depending on fetal CRL, maternal weight, ethnicity, and smoking as described by Poon et al. Variables were converted to multiples of the expected median (MoM) for each subject.
- 4
For each outcome group, uAD indices and biomarker levels expressed in log MoM were compared by the Student t test to values of the unaffected group. The diagnostic accuracy was evaluated by receiver-operating characteristics (ROC) curves and area under the ROC curve (AUC).
- 5
The risk of preeclampsia, severe preeclampsia, and early-onset preeclampsia based on a combination of maternal characteristic was determined by multiple logistic regression and these values were then log 10 transformed. Multiple logistic regression analysis was then used to analyze combinations of biomarkers and Doppler indices with maternal characteristics for the prediction of preeclampsia, severe preeclampsia, and early-onset preeclampsia. The performance of screening was then determined by ROC curves. The statistical software packages SPSS 16.0 (SPSS Inc, Chicago, IL) and STATA 10.0 (StataCorp, College Station, TX) were used for data analysis. A P value < .05 was considered significant.
Results
Of 1000 consecutively enrolled women, 107 (10.7%) were excluded from further analysis because of multiparity (n = 48), gestational age outside the inclusion criteria (n = 39), twin pregnancy (n = 2), major fetal defect (n = 2), miscarriage or fetal death before 20 weeks (n = 6), or missing outcome data (n = 10). Among the remaining 893 women, 20 had gestational hypertension (2.2%) develop, and 40 had preeclampsia (4.5%) develop, including 9 early-onset preeclampsia (1.0%) and 31 late preeclampsia (3.5%). Sixteen women had severe preeclampsia (1.8%). The characteristics of women are summarized in Table 1 . Women with preeclampsia or gestational hypertension did not differ from those in the unaffected group for maternal age and smoking status. The BMI at enrollment was significantly higher in women with preeclampsia or gestational hypertension than in unaffected women ( P < .01). Early-onset preeclampsia was more likely to occur in Afro-Caribbean women and in women with a preexisting medical condition ( P < .05).
| Variable | Unaffected | Gestational hypertension | Severe preeclampsia | Early preeclampsia | Late preeclampsia |
|---|---|---|---|---|---|
| (n = 833) | (n = 20) | (n = 16) | (n = 9) | (n = 31) | |
| Age, y, median (IQR) | 29.0 (27.0–32.7) | 26.0 (25.2–31.7) | 30.5 (24.5–35.2) | 32.0 (28.5–34.5) | 30.0 (27.0–32.0) |
| BMI, kg/m 2 , median (IQR) | 22.5 (20.8–24.9) | 26.6 (23.1–30.4) a | 26.6 (21.5–32.6) b | 28.9 (21.8–35.1) b | 27.5 (23.4–30.2) a |
| Ethnicity, n (%) | |||||
| White | 671 (80.6) | 17 (85.0) | 12 (75.0) | 6 (66.7) | 24 (77.4) |
| Afro-Caribbean | 58 (7.0) | 0 | 3 (18.8) | 3 (33.3) b | 4 (12.9) |
| Asian | 14 (1.7) | 0 | 0 | 0 | 0 |
| Hispanic | 23 (2.8) | 2 (10.0) | 1 (6.2) | 0 | 1 (3.2) |
| Mid-Eastern | 27 (3.2) | 0 | 0 | 0 | 0 |
| Other/mixed | 40 (4.8) | 1 (5.0) | 0 | 0 | 2 (6.5) |
| Medical history, n (%) | |||||
| None | 803 (96.4) | 19 (95.0) | 10 (62.5) | 6 (66.7) | 27 (87.1) |
| One or more | 30 (3.6) | 1 (5.0) | 6 (37.5) a | 3 (33.3) b | 4 (12.9) b |
| Chronic hypertension | 8 (1.0) | 0 | 4 (25.0) | 2 (22.2) | 2 (6.5) |
| Diabetes type 1 | 4 (0.5) | 0 | 0 | 0 | 0 |
| Diabetes type 2 | 2 (0.2) | 0 | 0 | 0 | 0 |
| Autoimmune disease | 6 (0.7) | 0 | 0 | 0 | 1 (3.2) |
| Thrombophilia | 6 (0.7) | 0 | 1 (6.2) | 1 (11.1) | 0 |
| Hyperthyroidism | 1 (0.1) | 0 | 1 (6.2) | 0 | 1 (3.2) |
| Renal disease | 3 (0.4) | 1 (5.0) | 2 (12.5) | 0 | 2 (6.5) |
| Smoking during pregnancy, n (%) | 56 (6.7) | 2 (10.0) | 2 (12.5) | 1 (11.1) | 1 (3.2) |
| Medication during pregnancy, n (%) | |||||
| None | 799 (96.1) | 20.0 (100) | 7 (43.8) | 3 (33.3) | 22 (71.0) |
| Antihypertensive | 10 (1.2) | 0 | 7 (43.8) | 4 (44.4) | 8 (25.8) |
| Heparin | 10 (1.2) | 0 | 0 | 0 | 3 (9.7) |
| Low-dose aspirin | 17 (2.0) | 0 | 3 (18.7) | 2 (22.2) | 1 (3.2) |
| Gestational age at delivery, wk, median (IQR) | 39.8 (38.9–40.7) | 39.1 (37.9–40.0) b | 36.1 (34.5–39.0) a | 34.9 (33.9–36.4) a | 38.9 (37.6–39.5) a |
| Birthweight, g, median (IQR) | 3400 (3045–3665) | 3387 (2781–3681) | 2215 (1632–2869) a | 2000 (1465–2400) a | 3035 (2709–3390) a |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
