Screening for pre-eclampsia by maternal factors and biomarkers at 11-13 weeks' gestation

Preeclampsia pathogenesis and prediction - where are we now: the focus on the role of galectins and miRNAs

Preeclampsia is a complex, progressive multisystem hypertensive disorder during pregnancy that significantly contributes to increased maternal and perinatal morbidity and mortality. Two screening algorithms are in clinical use for detecting preeclampsia: first-trimester screening, which has been developed and validated for predicting early-onset preeclampsia but is less effective for late-onset disease; and the sFlt-1:PlGF biomarker ratio (soluble tyrosine kinase and placental growth factor) used in suspected cases of preeclampsia. This ratio has a high negative predictive value, allowing for the reliable exclusion of the disease. Both of these screening tests have not met expectations. This review attempts to summarize the current knowledge on the pathogenesis and prediction of preeclampsia and to draw attention to novel biomarkers with a focus on microRNAs and galectins. Although these molecules belong to two distinct biological classes, they functionally converge in regulating placental and immune pathways. Ample evidence supports their involvement in the molecular mechanisms underlying preeclampsia. Based on current knowledge, galectin-13, C19MC members, and miRNA-210 are associated with the trophoblast/placenta and conditions of placental ischemia or hypoxia. Their levels differ significantly in pregnant women at risk of preeclampsia as early as the late first and early second trimester, making them potential markers for predicting preeclampsia.

Temporal validation of machine learning models for pre-eclampsia prediction using routinely collected maternal characteristics: A validation study

BACKGROUND

Pre-eclampsia (PE) contributes to more than one-fourth of all maternal deaths and half a million newborn deaths worldwide every year. Early screening and interventions can reduce PE incidence and related complications. We aim to 1) temporally validate three existing models (two machine learning (ML) and one logistic regression) developed in the same region and 2) compare the performances of the validated ML models with the logistic regression model in PE prediction. This work addresses a gap in the literature by undertaking a comprehensive evaluation of existing risk prediction models, which is an important step to advancing this field.

METHODS

We obtained a dataset of routinely collected antenatal data from three maternity hospitals in South-East Melbourne, Australia, extracted between July 2021 and December 2022. We temporally validated three existing models: extreme gradient boosting (XGBoost, 'model 1'), random forest ('model 2') ML models, and a logistic regression model ('model 3'). Area under the receiver-operating characteristic (ROC) curve (AUC) was evaluated discrimination performance, and calibration was assessed. The AUCs were compared using the 'bootstrapping' test.

RESULTS

The temporal evaluation dataset consisted of 12,549 singleton pregnancies, of which 431 (3.43 %, 95 % confidence interval (CI) 3.13-3.77) developed PE. The characteristics of the temporal evaluation dataset were similar to the original development dataset. The XGBoost 'model 1' and the logistic regression 'model 3' exhibited similar discrimination performance with an AUC of 0.75 (95 % CI 0.73-0.78) and 0.76 (95 % CI 0.74-0.78), respectively. The random forest 'model 2' showed a discrimination performance of AUC 0.71 (95 % CI 0.69-0.74). Model 3 showed perfect calibration performance with a slope of 1.02 (95 % CI 0.92-1.12). Models 1 and 2 showed a calibration slope of 1.15 (95 % CI 1.03-1.28) and 0.62 (95 % CI 0.54-0.70), respectively. Compared to the original development models, the temporally validated models 1 and 3 showed stable discrimination performance, whereas model 2 showed significantly lower discrimination performance. Models 1 and 3 showed better clinical net benefits between 3 % and 22 % threshold probabilities than default strategies.

CONCLUSIONS

During temporal validation of PE prediction models, logistic regression and XGBoost models exhibited stable prediction performance; however, both ML models did not outperform the logistic regression model. To facilitate insights into interpretability and deployment, the logistic regression model could be integrated into routine practice as a first-step in a two-stage screening approach to identify a higher-risk woman for further second stage screening with a more accurate test.

Effect of aspirin on biomarker profile in women at high risk for preeclampsia

BACKGROUND

There is limited evidence in the literature regarding the temporal changes of preeclampsia-related biomarkers during pregnancy in high-risk women who develop preeclampsia despite the administration of aspirin prophylaxis.

OBJECTIVE

This study aimed to compare the temporal changes in mean arterial pressure, uterine artery pulsatility index, placental growth factor, and soluble fms-like tyrosine kinase-1 across gestation in women identified as having high risk for preterm preeclampsia receiving aspirin prophylaxis and low-risk women without aspirin treatment.

STUDY DESIGN

This was a prospective longitudinal nested case-control study of 2007 women with singleton pregnancies who participated in the first-trimester screen-and-prevent program for preeclampsia at the Prince of Wales Hospital, Hong Kong Special Administrative Region, China, between January 2020 and May 2023. The risk of developing preterm preeclampsia was determined using the Fetal Medicine Foundation triple test (maternal factors, mean arterial pressure, uterine artery pulsatility index, and placental growth factor). High-risk women (adjusted risk ≥1:100) were administered a daily dose of aspirin at either 100 or 160 mg according to maternal weight, starting before 16 weeks until 36 weeks or until delivery or the onset of preeclampsia before 36 weeks. Low-risk women were matched according to maternal age, weight, and the date of the scan. The participants were followed up at 12 to 15+6, 20 to 24+6, and 30 to 37+6 weeks to measure mean arterial pressure, uterine artery pulsatility index, placental growth factor, and soluble fms-like tyrosine kinase-1 at each visit. The level of biomarker was expressed as multiple of the median. Log10 transformation was applied to fit the data to a Gaussian distribution before statistical analysis. A linear mixed-effects analysis was performed to compare the longitudinal changes of these biomarkers across gestation between the study groups.

RESULTS

Our study involved 403 low-risk women without preeclampsia, 1471 high-risk women without preeclampsia, and 133 high-risk women who developed preeclampsia. The low-risk group had significantly lower estimated marginal mean log10 mean arterial pressure multiple of the median, log10 uterine artery pulsatility index multiple of the median, and log10 soluble fms-like tyrosine kinase-1 multiple of the median, and higher estimated marginal mean log10 placental growth factor multiple of the median across gestation compared with the high-risk groups (P<.001). Among high-risk women, those who developed preeclampsia exhibited a significantly higher estimated marginal mean log10 mean arterial pressure multiple of the median (0.06378 vs 0.02985; P<.001), log10 uterine artery pulsatility index multiple of the median (0.08651 vs 0.02226; P<.001), and log10 soluble fms-like tyrosine kinase-1 multiple of the median (0.13204 vs 0.01234; P<.001), and lower estimated marginal mean log10 placental growth factor multiple of the median (-0.33504 vs -0.16388; P<.001) across gestation compared with those without preeclampsia. In the individual gestational time point analysis, compared with high-risk women without preeclampsia, those who developed preeclampsia exhibited higher log10 mean arterial pressure multiple of the median in all 3 trimesters, higher log10 uterine artery pulsatility index multiple of the median and lower log10 placental growth factor multiple of the median in the second and third trimesters, and higher log10 soluble fms-like tyrosine kinase-1 multiple of the median in the third trimester.

CONCLUSION

This study demonstrated that high-risk women who developed preeclampsia consistently exhibited high mean arterial pressure levels from the first trimester that remained unchanged during pregnancy, high uterine artery pulsatility index levels and low placental growth factor levels starting from the second trimester, and high soluble fms-like tyrosine kinase-1 levels in the third trimester compared with those who did not develop preeclampsia despite the administration of low-dose aspirin. These findings underscore the role of these biomarkers in further risk stratification for the development of preeclampsia among high-risk women following aspirin administration.

Expression and Clinical Significance of Irisin in Serum and Placenta Tissues of Pregnant Women with Severe Preeclampsia

Objective

Preeclampsia (PE) is a serious pregnancy-specific disorder that poses significant risks to maternal and fetal health, with severe preeclampsia (SPE) being a particularly life-threatening complication. The objective of this study is to investigate the effects and clinical significance of irisin in pregnant women with severe preeclampsia (SPE). Irisin levels in the serum and placental tissues of healthy pregnant women and those with early- and late-onset SPE were measured and compared.

Methods

A total of 70 pregnant women treated at our hospital from January to November 2023 were selected for this study. The participants were divided into three groups: 20 women with early-onset severe preeclampsia (ES-PE group), 20 women with late-onset severe preeclampsia (LS-PE group), and 30 healthy pregnant women (control group). Fasting peripheral blood samples (5 mL) were collected from each participant, and placental tissues were obtained after delivery. Irisin levels in serum were measured using enzyme-linked immunosorbent assays (ELISA) with a commercial kit, and irisin expression in placental tissues was assessed by immunohistochemistry (IHC) with a rabbit anti-irisin antibody. The modes of delivery were also recorded.

Results

The concentrations of irisin in both serum and placental tissues were significantly higher among pregnant women in the control group compared to the ES-PE and LS-PE groups. There was a significant difference between the control group and the ES-PE and LS-PE groups in the mode of delivery. Additionally, a significant positive correlation was identified between the serum irisin concentration and its differential expression in placental tissues, while there was a significant negative correlation between irisin levels in both serum and placental tissue and systolic and/or diastolic blood pressure.

Conclusion

Reduced serum and placental irisin levels in pregnant women with SPE were associated with the onset and progression of SPE and may serve as a potential biological marker for SPE screening.

Prediction of Preeclampsia in Twins Using First Trimester: cffDNA Fraction, PlGF, and MAP

OBJECTIVES

To evaluate cell-free fetal DNA fraction (cffDNAF) as a first-trimester screening marker for preeclampsia necessitating delivery before 37 weeks' gestation in twin pregnancies alone and combined with other bio-markers.

METHODS

Women with two live fetuses were enrolled in the first trimester, and evaluated for cffDNAF as a first trimester preeclampsia marker alone, and with placental growth factor (PlGF), mean arterial pressure (MAP), and uterine artery pulsatility index (UtA-PI).

RESULTS

There were 20 affected women; the cffDNAF was 9.0% (IQR: 8.4%-10.3%) in the affected, compared to 14% (IQR: 11%-16%) in 163 unaffected cases (p < 0.001). The AUROC for cffDNAF was 0.73 (95% CI: 0.61-0.85, p < 0.001), PlGF had an AUROC of 0.71 (0.59-0.83, p = 0.001), MAP had AUROC of 0.61 (0.50-0.72, p = 0.053) whereas UtA-PI had AUROC of 0.54 (0.39-0.69, p > 0.05). Combining all three biomarkers yielded an AUROC of 0.89 (0.78-0.98), with a sensitivity of 81%, specificity of 90%, negative predictive value (NPV) of 97.5%, and positive predictive value (PPV) of 50.7 UtA-PI did not contribute to the AUROC.

CONCLUSION

In twin pregnancies low first trimester cffDNAF effectively screens for preeclampsia necessitating delivery before 37 weeks' gestation, which is augmented with PlGF and MAP.

Comparison of first trimester preeclampsia combined screening performances with various approaches in the Indonesian population

INTRODUCTION

This study aimed to compare Fetal Medicine Foundation (FMF), Indonesian Maternal and Children Health Handbook (MCH-HB), and Indonesian Prenatal Institute (IPI) models for predicting PE.

MATERIALS/SUBJECTS AND METHODS

Maternal risk factors, biophysical, and biochemical markers were recorded from women screened for PE at 11-14 gestational weeks. The receiving operator curve (ROC) analysis was used to compare the detection rate (DR) among prediction models.

RESULTS

For all PE at a 10% false-positive rate (FPR), FMF had a DR 62.9%; MCH-HB had a DR 50.0%; IPI had a DR 66.9%. For early-onset PE, at 10% FPR FMF had a DR 80.3%; MCH-HB had a DR 71.4%; IPI had a DR 81.5%. For preterm PE at 10% FPR, FMF had a DR 70.2%; MCH-HB had a DR 38.5%; IPI had a DR 66.9%.

DISCUSSION

IPI algorithm is comparable to FMF and outperforms MCH-HB algorithm for all, early-onset, and preterm PE screening.

Improved Prediction Accuracy for Late-Onset Preeclampsia Using cfRNA Profiles: A Comparative Study of Marker Selection Strategies

Background: Late-onset pre-eclampsia (LO-PE) remains difficult to predict because placental angiogenic markers perform poorly once maternal cardiometabolic factors dominate. Methods: We reanalyzed a publicly available cell-free RNA (cfRNA) cohort (12 EO-PE, 12 LO-PE, and 24 matched controls). After RNA-seq normalization, we derived LO-PE candidate genes using (i) differential expression and (ii) elastic-net feature selection. Predictive accuracy was assessed with nested Monte-Carlo cross-validation (10 × 70/30 outer splits; 5-fold inner grid-search for λ). Results: The best LO-PE elastic-net model achieved a mean ± SD AUROC of 0.88 ± 0.08 and F1 of 0.73 ± 0.17-substantially higher than an EO-derived baseline applied to the same samples (AUROC ≈ 0.69). Enrichment analysis highlighted immune-tolerance and metabolic pathways; three genes (HLA-G, IL17RB, and KLRC4) recurred across >50% of cross-validation repeats. Conclusions: Plasma cfRNA signatures can outperform existing EO-based screens for LO-PE and nominate biologically plausible markers of immune and metabolic dysregulation. Because the present dataset is small (n = 48) and underpowered for single-gene claims, external validation in larger, multicenter cohorts is essential before clinical translation.

Increase of augmentation index (AIx@75): a promising tool for screening hypertensive pregnancy disorders

BACKGROUND

Screening tools in the first trimester of pregnancy for hypertensive pregnancy disorders need to be determined.

OBJECTIVES

To compare cardiovascular parameters between pregnant (PG) and non-pregnant women (NPG) and to evaluate the sensitivity and specificity of arterial stiffness indices in screening for hypertensive pregnancy disorders and their possible association with the mean uterine artery pulsatility index (MUA-PI).

METHODS

This study included 77 pregnant women (11-13.6 gestational weeks) and 77 age-matched non-pregnant women. Cardiovascular parameters were non-invasively measured using Mobil- O-Graph®, a cuff-based oscillometric device. The Doppler Ultrasonographic was used to evaluate the MUA-PI.

RESULTS

Augmentation index (AIx@75) was significantly higher in PG compared to NPG. ROC curve of AIx@75 showed area under curve (AUC): 0.7303, Sensitivity: 74.03% and Specificity: 64.94% and Cutoff: 22.50%. The systolic volume index was lower and the heart rate was higher in PG compared to NPG. Of the 77 pregnant women, 12 had an unfavorable outcome with hypertensive changes. Central systolic blood pressure (109.1 ± 8.84mmHg) and AIx@75 (31.97 ± 5.47%) were significantly higher in the group of pregnant women with outcome compared to the group without outcome (103.0 ± 8.53mmHg and 26.80 ± 8.71%). ROC curve showed better performance of the AIx@75 [AUC: 0.7179, Sensitivity: 83.33% and Specificity: 60.00%, Cutoff: 27.67%] compared to MUA-PI [AUC: 0.5098, Sensitivity: 8.333% and Specificity 98.44%].

CONCLUSIONS

AIx@75 was significantly higher in PG compared to NPG. We compared the AIx@75 of PG with and without outcomes. ROC curve analysis showed that this index could discriminate between PG with and without an outcome. Differently, the MUA-PI did not differ between PG with and without outcome, suggesting the superiority of AIx@75 in relation to MUA-PI as a method of screening in the first trimester for hypertensive disease of pregnancy. AIx@75 did not assotiate with MUA-PI. Prospective studies will be needed to confirm these findings.

External validation and comparison of Fetal Medicine Foundation competing-risks model for small-for-gestational-age neonate in the first trimester: multicenter cohort study

OBJECTIVES

To examine the predictive performance of the Fetal Medicine Foundation (FMF) competing-risks model for the first-trimester prediction of a small-for-gestational-age (SGA) neonate in a large, independent, unselected European cohort and to compare the competing-risks algorithm with previously published logistic-regression models.

METHODS

This was a retrospective, non-interventional, multicenter cohort study including 35 170 women with a singleton pregnancy who underwent a first-trimester ultrasound assessment between 11 + 0 and 13 + 6 weeks' gestation. We used the default FMF competing-risks model for the prediction of SGA combining maternal factors, uterine artery pulsatility index (UtA-PI), pregnancy-associated plasma protein-A (PAPP-A) and placental growth factor (PlGF) to obtain risks for different cut-offs of birth-weight percentile and gestational age at delivery. We examined the predictive performance in terms of discrimination and calibration and compared it with the published data on the model's development population and with published logistic-regression equations.

RESULTS

At a 10% false-positive rate, maternal factors and UtA-PI predicted 42.2% and 51.5% of SGA < 10th percentile delivered < 37 weeks and < 32 weeks, respectively. The respective values for SGA < 3rd percentile were 44.7% and 51.7%. Also at a 10% false-positive rate, maternal factors, UtA-PI and PAPP-A predicted 42.2% and 51.5% of SGA < 10th percentile delivered < 37 weeks and < 32 weeks, respectively. The respective values for SGA < 3rd percentile were 46.2% and 51.7%. At a 10% false-positive rate, maternal factors, UtA-PI, PAPP-A and PlGF predicted 47.6% and 66.7% of SGA < 10th percentile delivered < 37 weeks and < 32 weeks, respectively. The respective values for SGA < 3rd percentile were 50.0% and 69.0%. These data were similar to those reported in the original model's development study and substantially better than those calculated using pre-existing logistic-regression models (McNemar's test, P < 0.001). The FMF competing-risks model was well calibrated.

CONCLUSIONS

The FMF competing-risks model for the first-trimester prediction of SGA is reproducible in an independent, unselected low-risk cohort and superior to logistic-regression approaches. © 2025 International Society of Ultrasound in Obstetrics and Gynecology.

Exploring new predictors for hypertensive disorders of pregnancy

The best performing predicting Bayesian algorithm for preeclampsia, endorsed by FIGO, identifies high-risk women at first trimester screening who benefits of a closer monitoring and possibly preventive measures. Unfortunately, the most frequent late term and term preeclampsia are less efficiently predicted. This algorithm is based on statistical assumptions at odds with the physiopathology: preeclampsia is a disease and not a syndrome, as we know it is, and the contingent time-based criteria according to which all pregnancies if not terminated by nature should develop this "disease". In addition to this, we know that gestational hypertension might cause in fifty percent of cases severe outcome, comparable to preeclampsia. The very definition of preeclampsia as proteinuric hypertension is now extended to hypertension associated with other end-organ damage, including fetal growth restriction (FGR), this latter condition proceeding, in early onset cases, hypertension. Predicting phenotypes of hypertensive Disorders of pregnancy (HDP) could better help clinical practice. This study reports exploratory observations in women resulted at high and low risk at first trimester screening followed up at second and third trimester, to term. The co-variates interrogated were sFlt1/PlGF ratio, the uterine arteries PI, the systemic vascular resistances (SVR), maternal total body water and visceral fat. Women were classified as HDP-AGA, HDP-FGR, normotensive-FGR and uneventful pregnancies (controls). We performed a longitudinal Bayesian multivariate mixed-effects model corrected both for pre-gestational BMI and trimester of analysis. The sFlt-1/PlGF ratio and SVR confirmed their significant difference in HDP-AGA, in normotensive FGR, and HDP-FGR along the three trimesters from controls, but with different strength along the three trimesters. The bioimpedance analysis of total body water and visceral fat confirmed the association of these co-factors with women who will develop HDP-AGA. The strength of longitudinal changes observed, even on a limited number of cases, provide evidence that Bayesian algorithms applied at screening tests at different gestational ages, should be based on co-variates significantly associated either with HDP-FGR or with HDP-AGA provided that the main causative co-factors involved are adopted by predictive models aimed at these distinct diseases.

Formative research to optimize pre-eclampsia risk-screening and prevention (PEARLS): study protocol

BACKGROUND

Pre-eclampsia is a leading cause of maternal and neonatal mortality, affecting nearly 5% of pregnant women worldwide. Accurate and timely risk-screening of pregnant women is essential to start preventive therapies as early as possible, including low-dose aspirin and calcium supplementation. In the formative phase for the "Preventing pre-eclampsia: Evaluating AspiRin Low-dose regimens following risk Screening" (PEARLS) trial, we aim to validate and implement a pre-eclampsia risk-screening algorithm, and validate an artificial intelligence (AI) ultrasound for gestational age estimation. In the trial phase, we will compare different daily aspirin doses (75 mg v 150 mg) for pre-eclampsia prevention and postpartum bleeding. This study protocol outlines the mixed-methods formative phase of PEARLS, which will identify challenges and the feasibility of implementing these activities in participating facilities in Ghana, Kenya, and South Africa.

METHODS

We will employ qualitative and quantitative methods to identify factors that may influence trial implementation. In-depth interviews and focus group discussions with policy stakeholders, research midwives, health workers, and pregnant women will explore the barriers, facilitators, and acceptability of pre-eclampsia risk screening, AI ultrasound, and aspirin uptake. A cross-sectional survey of antenatal care and maternity health workers will assess current clinical practices around pre-eclampsia and willingness to participate in the trial activities. Data will be analyzed using thematic analysis and triangulated across sources and participant groups. The findings will inform trial design and help optimize implementation.

DISCUSSION

The research will provide critical insights into the feasibility of pre-eclampsia risk screening and AI ultrasound for gestational age estimation in resource-limited settings. By identifying factors that can influence implementation of pre-eclampsia prevention and care pathways, the findings will inform successful execution of the PEARLS trial, and post-research scale-up activities. This, in turn, can help reduce the prevalence of pre-eclampsia, and improve maternal and newborn outcomes in high-burden settings.

TRIAL REGISTRATION

PACTR202403785563823 || pactr.samrc.ac.za (Date of registration: 12 March 2024).

Advancing preeclampsia prediction: a tailored machine learning pipeline integrating resampling and ensemble models for handling imbalanced medical data

BACKGROUND

Constructing a predictive model is challenging in imbalanced medical dataset (such as preeclampsia), particularly when employing ensemble machine learning algorithms.

OBJECTIVE

This study aims to develop a robust pipeline that enhances the predictive performance of ensemble machine learning models for the early prediction of preeclampsia in an imbalanced dataset.

METHODS

Our research establishes a comprehensive pipeline optimized for early preeclampsia prediction in imbalanced medical datasets. We gathered electronic health records from pregnant women at the People's Hospital of Guangxi from 2015 to 2020, with additional external validation using three public datasets. This extensive data collection facilitated the systematic assessment of various resampling techniques, varied minority-to-majority ratios, and ensemble machine learning algorithms through a structured evaluation process. We analyzed 4,608 combinations of model settings against performance metrics such as G-mean, MCC, AP, and AUC to determine the most effective configurations. Advanced statistical analyses including OLS regression, ANOVA, and Kruskal-Wallis tests were utilized to fine-tune these settings, enhancing model performance and robustness for clinical application.

RESULTS

Our analysis confirmed the significant impact of systematic sequential optimization of variables on the predictive performance of our models. The most effective configuration utilized the Inverse Weighted Gaussian Mixture Model for resampling, combined with Gradient Boosting Decision Trees algorithm, and an optimized minority-to-majority ratio of 0.09, achieving a Geometric Mean of 0.6694 (95% confidence interval: 0.5855-0.7557). This configuration significantly outperformed the baseline across all evaluated metrics, demonstrating substantial improvements in model performance.

CONCLUSIONS

This study establishes a robust pipeline that significantly enhances the predictive performance of models for preeclampsia within imbalanced datasets. Our findings underscore the importance of a strategic approach to variable optimization in medical diagnostics, offering potential for broad application in various medical contexts where class imbalance is a concern.

Placental growth factor as a predictive marker of preeclampsia in twin pregnancy

OBJECTIVES

Placental growth factor (PlGF) has been reported as a good biomaker for the prediction of preeclampsia occurring in the short term in singleton pregnancies, in women presenting with clinical suspicion of preeclampsia. This study aims to evaluate the predictive value of the PlGF in twin pregnancies.

METHODS

Twin pregnancies with clinically suspected preeclampsia (24 weeks 0 days-36 weeks 6 days of gestation) were enrolled in this study. The threshold of PlGF for predicting preeclampsia was determined on the basis of a receiver-operating characteristic curve to predict preeclampsia and the short-term occurrence of preeclampsia.

RESULTS

Within 1 week, 2 weeks, and 4 weeks of testing respectively, a cutoff value of 215 pg/mL for PlGF to predict preeclamsia in twin pregnancies suspected to have preeclampsia has a specificity of 100 %[51.7 %, 100 %], 100 %[62.9 %, 100 %], 93.8 %[667.6 %, 99.7 %], and a negative predictive value of 100 %[94.8 %, 100 %], 100 %[95.0 %, 100 %], and 98.9 %[93.0 %, 99.9 %].

CONCLUSIONS

A cutoff value of 215 pg/mL for PlGF is a useful tool to exclude the development of preeclampsia within 4 weeks of measurement.

Molecular Biomarkers for Timely and Personalized Prediction of Maternal-Fetal Health Risk

Molecular biomarker profiling is an emerging field in maternal-fetal health with the potential to transform early detection and prediction of placental dysfunction. By analysing a range of biomarkers in maternal blood, researchers and clinicians can gain crucial insights into placental health, enabling timely interventions to enhance fetal and maternal outcomes. Placental structural function is vital for fetal growth and development, and disruptions can lead to serious pregnancy complications like preeclampsia. While conventional methods such as ultrasound and Doppler velocimetry offer valuable information on fetal growth and blood flow, they have limitations in predicting placental dysfunction before clinical signs manifest. In contrast, molecular biomarker profiling can provide a more comprehensive assessment by measuring proteins, metabolites, and microRNAs (miRNAs) in maternal blood, reflecting the placenta's endocrine and metabolic functions. This approach offers a deeper understanding of placental health and function, aiding in early detection and prediction of complications. Challenges in developing molecular biomarker profiling include pinpointing specific molecular changes in the placenta linked to pathologies, timing predictions of conditions before clinical onset, and understanding how placental dysfunction affects maternal metabolism. Validating specific biomarkers and integrating them effectively into clinical practice requires further research. This review underscores the significance of molecular biomarker profiling as a powerful tool for early detection and prediction of placental dysfunction in maternal-fetal health. Through an exploration of biomarker analysis, we delve into how a deeper understanding of placental health can potentially improve outcomes for both mother and baby. Furthermore, we address the critical need to validate the utility of biomarkers and effectively integrate them into clinical practice.

First-Trimester PlGF and PAPP-A and the Risk of Placenta-Mediated Complications: PREDICTION Prospective Study

OBJECTIVES

This study aimed to estimate the association between low first-trimester maternal serum PlGF (placental growth factor) and PAPP-A (pregnancy-associated plasma protein A) and the risk of placenta-mediated complications.

METHODS

We performed a secondary analysis of the PREDICTION study, including nulliparous participants recruited at 11 to 14 weeks of pregnancy. First-trimester PlGF and PAPP-A levels were reported in multiples of the median (MoM) adjusted for maternal characteristics and gestational age. Participants were stratified into 4 groups based on absence/presence of low (<0.4 MoM) PlGF and PAPP-A values. A composite of adverse pregnancy outcomes (including preeclampsia, fetal growth restriction, fetal death, and placental abruption) was calculated for deliveries occurring before 34 weeks, before 37 weeks, and at or after 37 weeks.

RESULTS

Out of the 7262 participants, 86 (1.2%) experienced the composite outcome before 37 weeks of gestation, including 35 (0.4%) before 34 weeks. The combination of low PAPP-A and low PlGF levels was associated with the greatest risk of adverse outcomes before 37 weeks (21%) and before 34 weeks (12%) compared with low PlGF alone (7% and 3%), low PAPP-A alone (2% and 1%), or neither marker (1% and 0.4%, respectively; P < 0.001). For preterm preeclampsia specifically, the combination of low PAPP-A and low PlGF was also associated with a greater risk (12%) compared with low PlGF alone (6%), low PAPP-A alone (0.5%), or neither marker (0.7%; P < 0.001).

CONCLUSIONS

The combination of low PAPP-A and low PlGF levels is associated with a very high risk for adverse outcomes before 34 and 37 weeks. An isolated low PAPP-A should not be considered a risk factor for adverse pregnancy outcomes.

Prediction of pre-eclampsia using maternal hemodynamic parameters at 12 + 0 to 15 + 6 weeks

OBJECTIVES

To compare the maternal hemodynamic profile at 12 + 0 to 15 + 6 weeks' gestation in women who subsequently developed pre-eclampsia (PE) and those who did not, and to assess the screening performance of maternal hemodynamic parameters for PE in combination with the Fetal Medicine Foundation (FMF) triple test, including maternal factors (MF), mean arterial pressure (MAP), uterine artery pulsatility index and placental growth factor.

METHODS

This was a prospective case-control study involving Chinese women with a singleton pregnancy who underwent preterm PE screening at 11 + 0 to 13 + 6 weeks' gestation using the FMF triple test, between February 2020 and February 2023. Women identified as being at high risk (≥ 1:100) for preterm PE by the FMF triple test were matched 1:1 with women identified as low risk (< 1:100) for maternal age ± 3 years, maternal weight ± 5 kg and date of screening ± 14 days. Two-dimensional transthoracic echocardiography was performed at 12 + 0 to 15 + 6 weeks to evaluate maternal hemodynamic parameters (heart rate (HR), stroke volume (SV), cardiac output (CO) and systemic vascular resistance (SVR)). Maternal hemodynamic parameters were expressed as multiples of the median (MoM) values, determined by linear regression models to adjust for gestational age (GA) and MF. The distribution of log10 MoM values of maternal hemodynamic parameters in cases of PE and unaffected pregnancies, and the association between these hemodynamic parameters and GA at delivery, were assessed. The risks of preterm PE (delivery before 37 weeks) and any-onset PE (delivery at any time) were reassessed using Bayes' theorem after maternal hemodynamic parameters were added to the FMF triple test. The screening performance for preterm PE and any-onset PE was determined by the area under the receiver-operating-characteristics curve (AUC) and detection rate at a 10% fixed false-positive rate (FPR). Differences in AUC (ΔAUC) were assessed using DeLong's test.

RESULTS

A total of 743 cases were analyzed, of whom 39 (5.2%) subsequently developed PE, including 29 (3.9%) cases of preterm PE and 10 (1.3%) cases of term PE. Mean log10 SVR MoM was significantly higher in cases of preterm PE and any-onset PE compared with unaffected pregnancies. Mean log10 SV MoM and log10 CO MoM were significantly lower in cases of preterm PE and any-onset PE compared with unaffected pregnancies. Mean log10 HR MoM was not significantly different between the study groups. Mean log10 CO MoM and log10 SVR MoM were not significantly correlated with GA at delivery in preterm PE and any-onset PE. For the prediction of preterm PE and any-onset PE, adding CO or SVR or replacing MAP with CO and SVR in the FMF triple test achieved an identical or greater AUC compared with the FMF triple test, but ΔAUC was not significantly different. In addition, adding CO or SVR or replacing MAP by CO and SVR in the FMF triple test did not improve the detection rate for preterm PE and any-onset PE at a fixed FPR of 10%.

CONCLUSIONS

Women with preterm PE or any-onset PE exhibited increased SVR and decreased CO before the clinical manifestations of PE became apparent. These changes may serve as early indicators of cardiovascular maladaptation. However, assessment of maternal hemodynamics at 12 + 0 to 15 + 6 weeks does not enhance the screening performance for preterm PE and any-onset PE of these parameters. The FMF triple test remains superior to other biomarker combinations for predicting PE. © 2025 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Prenatal diagnosis of vasa praevia in routine clinical practice: Prevention of stillbirths and impact on perinatal outcomes

BACKGROUND

Vasa praevia (VP) is defined as the presence of unsupported fetal blood vessels in close proximity of the internal os of the cervix. There is robust evidence from observational cohort studies and meta-analysis that prenatal diagnosis of VP is associated with excellent perinatal outcomes. We have previously proposed a two-stage strategy for prenatal diagnosis that can be implemented in routine clinical practice leading to effective prenatal diagnosis and prevention of fetal and neonatal mortality and morbidity.

OBJECTIVES

To demonstrate the feasibility and effectiveness of a two-stage screening strategy for prenatal diagnosis of VP in routine clinical practice and to estimate the potential impact on prevention of stillbirths and perinatal deaths.

STUDY DESIGN

This was an observational retrospective cohort study carried out at the Medway Fetal and Maternal Medicine Centre between January 2010 and June 2022. We examined the feasibility and effectiveness of this policy in terms of identification of a high-risk cohort and prenatal diagnosis of VP through routine 11-13 and 20-22 weeks' ultrasound assessments based on the two-stage protocol. We also examined the impact on maternal, neonatal and perinatal outcomes in pregnancies with a confirmed diagnosis of VP. Absolute risks (95 %) were calculated based on rates of events in the two groups. Logistic regression analysis was used to estimate independent contribution from maternal and pregnancy characteristics in prediction of VP.

RESULTS

The study population of 53,648 singleton pregnancies included 45 pregnancies with VP (0.83 per 1,000 pregnancies or an incidence of 1 in 1,192 pregnancies). VP was suspected in 56 cases and were resolved in 11 cases (19.6 %), thus leaving 45 pregnancies with a confirmed diagnosis of VP. The main findings that predicted VP included a low-lying placenta at 20-22 weeks', placenta praevia, bilobed placenta and a velamentous cord insertion. In our study population, pregnancies with a prenatal diagnosis of VP had a livebirth rate of 100 % and an intact perinatal survival rate of 97.8 %.

CONCLUSION

Our study demonstrates that effective prenatal diagnosis of pregnancies with VP can be achieved in routine clinical practice with good perinatal outcomes.

Kidney disease and reproductive health

Understanding the relationship between reproductive health and kidney function is important to provide holistic care for people living with kidney disease. Chronic kidney disease (CKD) has negative impacts on both male and female fertility owing to factors including inflammation, hormonal dysregulation, reduced ovarian reserve, reduced sperm quality and sexual dysfunction. However, pregnancy is achievable for most cisgender women with kidney disease, including kidney transplant recipients and patients on dialysis. CKD in pregnancy is associated with health risks to the mother and child, including increased risk of progression of kidney disease, hypertensive complications of pregnancy, and neonatal complications including fetal growth restriction, preterm birth and stillbirth. However, with appropriate pre-pregnancy counselling, fertility assessment and support, health optimization, and evidence-based antenatal care, the majority of patients will achieve a good outcome. Medication safety should be reviewed before and during pregnancy and lactation, weighing the risk of disease flare against potential adverse effects on the offspring. Important areas for further research include the optimal timing of delivery and the short- and long-term cardiovascular and renal impacts of pregnancy in patients with CKD, as well as long-term kidney and cardiovascular outcomes in their offspring.

Preterm preeclampsia screening and prevention: a comprehensive approach to implementation in a real-world setting

BACKGROUND

Preeclampsia significantly impacts maternal and perinatal health. Early screening using advanced models and primary prevention with low-dose acetylsalicylic acid for high-risk populations is crucial to reduce the disease's incidence. This study assesses the feasibility of implementing preterm preeclampsia screening and prevention by leveraging information from our current aneuploidy screening program in a real-world setting with geographic separation clinical site and laboratory analysis site.

METHODS

A prospective cohort study involved pregnant individuals undergoing nuchal translucency scans between 11 and 14 weeks. Risk for preterm preeclampsia was assessed using the Fetal Medicine Foundation algorithm, which includes maternal risk factors, uterine artery Doppler, mean arterial pressure and serum markers (Placental growth factor, PlGF and Pregnancy-associated plasma protein-A, PAPP-A). High-risk patients were offered low-dose acetylsalicylic acid prophylaxis. Feasibility outcomes, such as recruitment rates, protocol adherence, operational impact, integration with existing workflows, screening performance and pregnancy outcomes, were evaluated.

RESULTS

Out of 974 participants, 15.6% were deemed high-risk for preterm preeclampsia. The study achieved high recruitment (82.1%) and adherence rates, with 95.4% of high-risk patients prescribed low-dose acetylsalicylic acid. Screening performance, adjusted for low-dose acetylsalicylic acid use, showed a detection rate of 88.9-90% (FPR 13.0% and 12.7%) for preterm preeclampsia. High-risk group for preeclampsia had higher incidences of adverse outcomes, including preterm preeclampsia (7.5 vs 0.4%; p < 0.001), preterm delivery (21.2 vs 6.2%; p < 0.001), low birth weight (23.3 vs 5.6%; p < 0.001) and birthweight < 10th percentile (11% vs 5.6%; p = 0.015) compared to low-risk group. The integration of preeclampsia screening had a minimal effect on the time required for aneuploidy screening, with results obtained within a rapid turnaround time.

CONCLUSIONS

The study confirms the feasibility of integrating comprehensive preeclampsia screening into clinical practice, notwithstanding geographic separation between laboratory and clinical settings. It underscores the need for broader adoption and enhanced infrastructure to optimize patient care and outcomes across diverse healthcare settings.

TRIAL REGISTRATION

Clinical trial: NCT04412681 (2020-06-02).

Role of routine uterine artery Doppler at 18-22 and 24-28 weeks' gestation following routine first-trimester screening for pre-eclampsia

OBJECTIVES

To assess the performance of mean uterine artery pulsatility index (UtA-PI) at 18-22 and 24-28 weeks of gestation in the prediction of pre-eclampsia (PE) and small-for-gestational age (SGA), and its role in reassessing the risk of PE and SGA in pregnancies screened for PE in the first trimester.

METHODS

This was a retrospective observational cohort study of 4464 women with singleton pregnancy screened routinely for PE in the first trimester, using the Gaussian algorithm, from March 2019 to May 2021, and who underwent UtA-PI assessment at 18-22 gestational weeks. Women were categorized as low risk or high risk based on the risk index obtained after first-trimester screening for PE. In high-risk patients, UtA-PI was also assessed at 24-28 weeks of gestation. Sensitivity, specificity, positive predictive value, negative predictive value (NPV), positive likelihood ratio, negative likelihood ratio and area under the receiver-operating-characteristics curve were calculated to assess the performance of UtA-PI at 18-22 and 24-28 weeks in predicting PE and SGA in the high-risk group. In all participants, different UtA-PI percentiles at 18-22 or 24-28 weeks, or their combination, were analyzed to explore their role in reassessing the risk of PE and SGA following first-trimester PE screening.

RESULTS

The performance of UtA-PI at 18-22 and 24-28 weeks in the high-risk group was good for predicting preterm PE and preterm SGA, and excellent for predicting early-onset PE and early-onset SGA, with an NPV of > 97% for all outcomes. In the low-risk group, UtA-PI ≥ 95th percentile at 18-22 weeks' gestation identified a subgroup of pregnancies with a significantly higher risk of preterm SGA compared to the low-risk group. In the high-risk group, UtA-PI < 60th percentile at 18-22 weeks' gestation, UtA-PI < 85th percentile at 24-28 weeks' gestation, and UtA-PI < 85th percentile at 24-28 weeks' gestation in women with UtA-PI ≥ 60th percentile at 18-22 weeks, identified subgroups of pregnancies with a risk of PE and SGA comparable to that of the low-risk group.

CONCLUSIONS

The performance of UtA-PI at 18-22 and 24-28 gestational weeks in high-risk pregnancies identified during first-trimester screening for PE is similar to that in the general population. The risk of PE and SGA in a high-risk cohort can be reassessed by measuring UtA-PI at 18-22 weeks, 24-28 weeks or both, allowing adjustment of follow-up, particularly de-escalation of care. © 2024 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Risk Assessment for Preeclampsia in the Preconception Period Based on Maternal Clinical History via Machine Learning Methods

Objective: This study was aimed to identify the most effective machine learning (ML) algorithm for predicting preeclampsia based on sociodemographic and obstetric factors during the preconception period. Methods: Data from pregnant women admitted to the obstetric clinic during their first trimester were analyzed, focusing on maternal age, body mass index (BMI), smoking status, history of diabetes mellitus, gestational diabetes mellitus, and mean arterial pressure. The women were grouped by whether they had a preeclampsia diagnosis and by whether they had one or two live births. Predictive models were then developed using five commonly applied ML algorithms. Results: The study included 100 mothers divided into four groups: 22 nulliparous mothers with preeclampsia, 25 nulliparous mothers without preeclampsia, 28 parous mothers with preeclampsia, and 25 parous mothers without preeclampsia. Analysis showed that maternal BMI and family history of diabetes mellitus were the most significant predictive variables. Among the predictive models, the extreme gradient boosting (XGB) classifier demonstrated the highest accuracy, achieving 70% and 72.7% in the respective groups. Conclusions: A predictive model utilizing an ML algorithm based on maternal sociodemographic data and obstetric history could serve as an early detection tool for preeclampsia.

Derivation and Validation of Prediction of Preterm Preeclampsia Using Machine Learning Algorithms

OBJECTIVE

 This study aimed to develop machine learning (ML) models for predicting preterm preeclampsia using the information available before 23 weeks gestation.

STUDY DESIGN

 This was a secondary analysis of the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be (nuMoM2b) cohort. We considered 131 features available before 23 weeks including maternal demographics, obstetrics and family history, social determinants of health, physical activity, nutrition, and early second-trimester ultrasound. Our primary outcome was preterm preeclampsia before 37 weeks. The dataset was randomly split into a training set (70%) and a validation set (30%). ML models using glmnet, multilayer perceptron, random forest, XGBoost (extreme gradient boosting), and LightGBM models were developed. Using the ML approach that achieved the best area under the curve (AUC), we developed the final model. Further feature selection was conducted from the top 25 important features based on SHapley Additive exPlanations (SHAP) values. The performance of the final model was assessed using the validation dataset.

RESULTS

 Of 9,467 individuals, 219 (2.3%) had preterm preeclampsia. The AUC of the XGBoost model was the highest (AUC = 0.749 [95% confidence interval (95% CI), 0.736-0.762]) compared with other models. Therefore, XGBoost was used to develop models using fewer variables. The XGBoost model with the eight features (in order of importance: mean uterine artery pulsatility index in the early second trimester, chronic hypertension, pregestational diabetes, uterine artery notch, systolic and diastolic blood pressure in the first trimester, body mass index, and maternal age) was chosen as the final model as it had an AUC of 0.741 (95% CI, 0.730-0.752) which was not inferior to the original model (p = 0.58). The final model in the validation dataset had an AUC of 0.779 (95% CI, 0.722-0.831). An online application of the final model was developed ( https://kawakita.shinyapps.io/Preterm_preeclampsia/ ).

CONCLUSION

 ML algorithms using information available before 23 weeks can accurately predict preterm preeclampsia before 37 weeks.

KEY POINTS

· Prediction models using uterine artery Doppler have not been adopted in the US.. · We developed a model using an ML algorithm.. · An online application of the final model was developed.. · ML algorithms using information available before 23 weeks can accurately predict preterm preeclampsia before 37 weeks..

Preeclampsia and its prediction: traditional versus contemporary predictive methods

OBJECTIVE

Preeclampsia (PE) poses a significant threat to maternal and perinatal health, so its early prediction, prevention, and management are of paramount importance to mitigate adverse pregnancy outcomes. This article provides a brief review spanning epidemiology, etiology, pathophysiology, and risk factors associated with PE, mainly discussing the emerging role of Artificial Intelligence (AI) deep learning (DL) technology in predicting PE, to advance the understanding of PE and foster the clinical application of early prediction methods.

METHODS

Our narrative review comprehensively examines the PE epidemiology, etiology, pathophysiology, risk factors and predictive approaches, including traditional models and AI deep learning technology.

RESULTS

Preeclampsia involves a wide range of biological and biochemical risk factors, among which poor uterine artery remodeling, excessive immune response, endothelial dysfunction, and imbalanced angiogenesis play important roles. Traditional PE prediction models exhibit significant limitations in sensitivity and specificity, particularly in predicting late-onset PE, with detection rates ranging from only 30% to 50%. AI models have exhibited a notable level of predictive accuracy and value across various populations and datasets, achieving detection rates of approximately 70%. Particularly, they have shown superior predictive capabilities for late-onset PE, thereby presenting novel opportunities for early screening and management of the condition.

CONCLUSION

AI DL technology holds promise in revolutionizing the prediction and management of PE. AI-based approaches offer a pathway toward more effective risk assessment methods by addressing the shortcomings of traditional prediction models. Ongoing research efforts should focus on expanding databases and validating the performance of AI in diverse populations, leading to the development of more sophisticated prediction models with improved accuracy.

Prediction of pre-eclampsia with machine learning approaches: Leveraging important information from routinely collected data

BACKGROUND

Globally, pre-eclampsia (PE) is a leading cause of maternal and perinatal morbidity and mortality. PE prediction using routinely collected data has the advantage of being widely applicable, particularly in low-resource settings. Early intervention for high-risk women might reduce PE incidence and related complications. We aimed to replicate our machine learning (ML) published work predicting another maternal condition (gestational diabetes) to (1) predict PE using routine health data, (2) identify the optimal ML model, and (3) compare it with logistic regression approach.

METHODS

Data were from a large health service network with 48,250 singleton pregnancies between January 2016 and June 2021. Supervised ML models were employed. Maternal clinical and medical characteristics were the feature variables (predictors), and a 70/30 data split was used for training and testing the model. Predictive performance was assessed using area under the curve (AUC) and calibration plots. Shapley value analysis assessed the contribution of feature variables.

RESULTS

The random forest approach provided excellent discrimination with an AUC of 0.84 (95% CI: 0.82-0.86) and highest prediction accuracy (0.79); however, the calibration curve (slope of 1.21, 95% CI 1.13-1.30) was acceptable only for a threshold of 0.3 or less. The next best approach was extreme gradient boosting, which provided an AUC of 0.77 (95% CI: 0.76-0.79) and well-calibrated (slope of 0.93, 95% CI 0.85-1.01). Logistic regression provided good discrimination performance with an AUC of 0.75 (95% CI: 0.74-0.76) and perfect calibration. Nulliparous, pre-pregnancy body mass index, previous pregnancy with prior PE, maternal age, family history of hypertension, and pre-existing hypertension and diabetes were the top-ranked features in Shapley value analysis.

CONCLUSION

Two ML models created the highest-performing prediction using routinely collected data to identify women at high risk of PE, with acceptable discrimination. However, to confirm this result and also examine model generalisability, external validation studies are needed in other settings, utilising standardised prognostic factors.

Antenatal screening - The roles of medical and family history, routine tests, and examination findings

Routine antenatal care includes history, examination, and several standard laboratory tests. Other than the original objectives, the generated data is seldom utilised for screening for adverse obstetric and perinatal outcomes. Although new approaches and sophisticated tests improve prediction of complications such as pre-eclampsia, these may not be available globally. Maternal age, race/ethnicity, anthropometry, and method of conception can influence the occurrence of pregnancy complications. The importance of medical and obstetric history is well documented but often ignored. Routine test results including blood picture, hepatitis B and rubella serology, and sexually transmitted diseases, have additional health implications. The awareness of, and the ability to utilise, available antenatal data and tests in obstetric management will enhance individualised obstetric risk assessment thus facilitating the targeting of high-risk gravidae for further management, including the use of specific and technology-driven tests where available, and close monitoring and treatment, in a cost-effective manner.

Redesigning antenatal care: Prospective use of an implementation framework to establish a population-based multidisciplinary first-trimester screening, assessment and prevention service

BACKGROUND

Australian rates of adverse obstetric outcomes have improved little despite guidelines recommending history-based screening and intervention. The first trimester provides a unique opportunity to predict and prevent complications, yet population-based screening has failed to be translated into broad clinical practice.

AIMS

This study aimed to redesign antenatal care within an Australian public healthcare centre to align with evidence-based maternity care, including population-based first-trimester screening with early initiation of preventative strategies in high-risk pregnancies.

METHODS

A five-phase action-process model, sharing key elements with implementation science theory, was used to explore barriers to change in antenatal care, co-design a novel service with consumers and establish a population-based antenatal pathway commencing with a multidisciplinary first-trimester screening, assessment and planning visit.

RESULTS

The case for change and associated barriers were defined from the perspective of antenatal care stakeholders. Key needs of each group were established, and solutions were created using co-design methodology, allowing the team to create a novel approach to antenatal care which directly addressed identified barriers. Implementation of the service was associated with a fall in the median gestation at first specialist maternity care provider visit from 20 to 13 weeks.

CONCLUSIONS

This study confirms the feasibility of establishing a comprehensive first-trimester screening program within a public Australian healthcare setting and highlights a co-design process which places individualised assessment at the forefront of antenatal care. This framework may be applicable to most public maternity settings in Australia, with expansion aimed at providing equity of care, including in rural and remote settings.

High Serum Adrenomedullin and Mid-Regional Pro-Atrial Natriuretic Peptide Concentrations in Early Pregnancy Predict the Development of Gestational Hypertension

OBJECTIVES

Adrenomedullin (AM) and natriuretic peptide levels are elevated in pre-eclampsia. The aim of the present study was to determine AM and natriuretic peptide concentrations before 20 weeks of pregnancy in women who later developed gestational hypertension and in normal pregnancies.

METHODS

95 pregnant Caucasian women were included in the study. Gestational hypertension (GH) was diagnosed in 18 patients. The control group consisted of 41 patients with normal pregnancies (non-GH). Blood samples were taken during the first trimester of pregnancy.

RESULTS

Analysis of NT-proBNP showed no significant differences between the group of patients who later developed GH and those with normal pregnancies. Patients who developed GH later in pregnancy had higher levels of both MR-proANP (p < 0.001) and adrenomedullin (p < 0.001). Higher levels of MR-proANP were found in the GH with pre-eclampsia group compared with the GH without pre-eclampsia group. Higher levels of AM (p < 0.05) and MR-proANP (p < 0.005) correlated with the risk of preterm birth.

CONCLUSIONS

(1) Plasma adrenomedullin and MR-proANP concentrations were higher before the 20th week of pregnancy in women who later developed GH; (2) NT-proBNP concentrations did not differ between women with pregnancy-induced hypertension and normal pregnancies; (3) MR-proANP concentrations were highest in patients who developed pre-eclampsia in advanced pregnancy; and (4) there was a correlation between higher plasma adrenomedullin, MR-proANP concentrations before the 20th week of pregnancy, and the risk of preterm birth.

Continuous Risk Assessment of Late and Term Preeclampsia Throughout Pregnancy: A Retrospective Cohort Study

Background and Objectives: To evaluate the diagnostic accuracy of widely available biomarkers longitudinally measured throughout pregnancy to predict all and term (delivery at ≥37 weeks) preeclampsia (PE). Materials and Methods: This is a longitudinal retrospective study performed at Hospital Universitario de Torrejón (Madrid, Spain) and Shterev Hospital (Sofia, Bulgaria) between August 2017 and December 2022. All pregnant women with singleton pregnancies and non-malformed live fetuses attending their routine ultrasound examination and first-trimester screening for preterm PE at 11 + 0 to 13 + 6 weeks' gestation at the participating centers were invited to participate in a larger study for the prediction of pregnancy complications. The dataset was divided into two subsets to develop and validate a joint model of time-to-event outcome and longitudinal data, and to evaluate how the area under the receiving operating characteristic curve (AUROC) evolved with time. Results: 4056 pregnancies were included in the training set (59 all PE, 40 term PE) and 944 in the validation set (23 all PE, 20 term PE). For the joint model and all PE, the AUROC was 0.84 (95% CI 0.73 to 0.94) and the detection rate (DR) for a 10% screening positive rate (SPR) was 56.5 (95% CI 34.5 to 76.8). For term PE, AUROC was 0.80 (95% CI 0.69 to 0.91), and DR for a 10% SPR was 55.0 (95% CI 31.5 to 76.9). The AUROC using only information from the first trimester was 0.50 (95% CI 0.37 to 0.64) and it increased to 0.84 (0.73 to 0.94) when using all information available. Conclusions: Routinely measuring MAP and UtA-PI throughout pregnancy may improve the predictive prediction power for all and term-PE.

Clinical and sonographic risk factors for developing pre-eclampsia refractory to aspirin prophylaxis

OBJECTIVE

Identify risk factors for development of preeclampsia refractory to aspirin prophylaxis in women at high-risk of preeclampsia.

MATERIAL AND METHODS

A retrospective cohort study analyzed 206 women identified as high-risk for preeclampsia through first-trimester screening and prescribed aspirin prophylaxis. We compared maternal characteristics, medical history, biochemical markers, and uterine artery Doppler indices between those with and without preeclampsia.

RESULTS

Women with preeclampsia had significantly higher rates of chronic hypertension (54.3% vs. 8.2%), higher first-trimester mean arterial pressure (MAP, 109.6 vs. 95.4 mmHg), and higher body mass index (BMI, 27.6 vs. 24.9) compared to controls. Second-trimester MAP and mean uterine artery pulsatility index (UtA-PI) were also significantly elevated in the preeclampsia group (103.3 mmHg and 1.39, respectively) compared to controls (89.7 mmHg and 1.05). ROC curve analysis identified an optimal second trimester UtA-PI cut-off of 1.36 for predicting preeclampsia, with sensitivity of 49% and specificity of 87.1%. When using a cut-off value of 0.77 for the second-to-first trimester UtA-PI ratio, the sensitivity and specificity were 60% and 90.6%, respectively.

CONCLUSION

Chronic hypertension, high first and second trimester MAP, higher BMI, and elevated second trimester UtA-PI are associated with preeclampsia despite aspirin prophylaxis. Evaluating second trimester UtA-PI or the ratio of second to first trimester UtA-PI may be a promising tool for identifying women who do not respond to aspirin.

Implementation of First-Trimester Screening and Prevention of Preeclampsia: A Stepped Wedge Cluster-Randomized Trial in Asia

BACKGROUND

This trial aimed to assess the efficacy, acceptability, and safety of a first-trimester screen-and-prevent strategy for preterm preeclampsia in Asia.

METHODS

Between August 1, 2019, and February 28, 2022, this multicenter stepped wedge cluster randomized trial included maternity/diagnostic units from 10 regions in Asia. The trial started with a period where all recruiting centers provided routine antenatal care without study-related intervention. At regular 6-week intervals, one cluster was randomized to transit from nonintervention phase to intervention phase. In the intervention phase, women underwent first-trimester screening for preterm preeclampsia using a Bayes theorem-based triple-test. High-risk women, with adjusted risk for preterm preeclampsia ≥1 in 100, received low-dose aspirin from <16 weeks until 36 weeks.

RESULTS

Overall, 88.04% (42 897 of 48 725) of women agreed to undergo first-trimester screening for preterm preeclampsia. Among those identified as high-risk in the intervention phase, 82.39% (2919 of 3543) received aspirin prophylaxis. There was no significant difference in the incidence of preterm preeclampsia between the intervention and non-intervention phases (adjusted odds ratio [aOR], 1.59 [95% CI, 0.91-2.77]). However, among high-risk women in the intervention phase, aspirin prophylaxis was significantly associated with a 41% reduction in the incidence of preterm preeclampsia (aOR, 0.59 [95% CI, 0.37-0.92]). In addition, it correlated with 54%, 55%, and 64% reduction in the incidence of preeclampsia with delivery at <34 weeks (aOR, 0.46 [95% CI, 0.23-0.93]), spontaneous preterm birth <34 weeks (aOR, 0.45 [95% CI, 0.22-0.92]), and perinatal death (aOR, 0.34 [95% CI, 0.12-0.91]), respectively. There was no significant between-group difference in the incidence of aspirin-related severe adverse events.

CONCLUSIONS

The implementation of the screen-and-prevent strategy for preterm preeclampsia is not associated with a significant reduction in the incidence of preterm preeclampsia. However, low-dose aspirin effectively reduces the incidence of preterm preeclampsia by 41% among high-risk women. The screen-and-prevent strategy for preterm preeclampsia is highly accepted by a diverse group of women from various ethnic backgrounds beyond the original population where the strategy was developed. These findings underpin the importance of the widespread implementation of the screen-and-prevent strategy for preterm preeclampsia on a global scale.

REGISTRATION

URL: https://www.clinicaltrials.gov; Unique identifier: NCT03941886.

Prediction of Fetal Death in Preterm Preeclampsia Using Fetal Sex, Placental Growth Factor and Gestational Age

BACKGROUND/OBJECTIVES

Preeclampsia (PE) is a systemic disease that affects 4.6% of pregnancies. Despite the existence of a first-trimester screening for the prediction of preterm PE, no consensus exists regarding neither the right moment to end the pregnancy nor the appropriate variables to estimate the prognosis. The objective of this study was to obtain a prediction model for perinatal death in patients with preterm PE, useful for clinical practice.

METHODS

Singleton pregnant women with PE and preterm delivery were included in an observational retrospective study. Multiple maternal and fetal variables were collected, and several multivariable logistic regression analyses were applied to construct models to predict perinatal death, selecting the most accurate and reproducible according to the highest area under the curve (AUC) and the lowest Akaike Information Criteria (AIC).

RESULTS

A group of 148 pregnant women were included, and 18 perinatal deaths were registered. Univariable logistic regression selected as statistically significant variables the following: gestational age (GA) at admission, fetal sex, poor response to antihypertensive drugs, PlGF, umbilical artery (UA) pulsatility index (PI), cerebroplacental ratio (CPR), and absent/reversed ductus venosus (DV). The multivariable model, including all these parameters, presented an AUC of 0.95 and an AIC of 76.5. However, a model including only GA and fetal sex presented a similar accuracy with the highest simplicity (AUC 0.93, AIC 67.6). Finally, in fetuses with a similar GA, fetal death became dependent on PlGF and fetal sex, underlying the role of fetal sex in all circumstances.

CONCLUSIONS

Female fetal sex and low PlGF are notorious predictors of perinatal death in preterm PE, only surpassed by early GA at birth.

First-Trimester Preeclampsia-Induced Disturbance in Maternal Blood Serum Proteome: A Pilot Study

Preeclampsia (PE) is a complex and multifaceted obstetric syndrome characterized by several distinct molecular subtypes. It complicates up to 5% of pregnancies and significantly contributes to maternal and newborn morbidity, thereby diminishing the long-term quality of life for affected women. Due to the widespread dissatisfaction with the effectiveness of existing approaches for assessing PE risk, there is a pressing need for ongoing research to identify newer, more accurate predictors. This study aimed to investigate early changes in the maternal serum proteome and associated signaling pathways. The levels of 125 maternal serum proteins at 11-13 weeks of gestation were quantified using liquid chromatography-multiple reaction monitoring mass spectrometry (LC-MRM MS) with the BAK-125 kit. Ten serum proteins emerged as potential early markers for PE: Apolipoprotein M (APOM), Complement C1q subcomponent subunit B (C1QB), Lysozyme (LYZ), Prothrombin (F2), Albumin (ALB), Zinc-alpha-2-glycoprotein (AZGP1), Tenascin-X (TNXB), Alpha-1-antitrypsin (SERPINA1), Attractin (ATRN), and Apolipoprotein A-IV (APOA4). Notably, nine of these proteins have previously been associated with PE in prior research, underscoring the consistency and reliability of our findings. These proteins play key roles in critical molecular processes, including complement and coagulation cascades, platelet activation, and insulin-like growth factor pathways. To improve the early prediction of PE, a highly effective Support Vector Machine (SVM) model was developed, analyzing 19 maternal serum proteins from the first trimester. This model achieved an area under the curve (AUC) of 0.91, with 87% sensitivity and 95% specificity, and a hazard ratio (HR) of 13.5 (4.6-40.8) with p < 0.001. These findings demonstrate that serum protein-based SVM models possess significantly higher predictive power compared to the routine first-trimester screening test, highlighting their superior utility in the early detection and risk stratification of PE.

First trimester risk of preeclampsia and rate of spontaneous birth in patients without preeclampsia

BACKGROUND

First-trimester screening for preeclampsia using a combination of maternal risk factors and mean arterial pressure, uterine artery pulsatility index, and placental growth factor, as proposed by the Fetal Medicine Foundation, provides effective prediction of preterm preeclampsia. Placental dysfunction is a potential precursor of spontaneous birth.

OBJECTIVE

The objective of this study was to examine if the estimated risk of preeclampsia is associated with the gestational age at onset of spontaneous delivery in the absence of preeclampsia.

STUDY DESIGN

This was a secondary analysis of the data from the Screening programme for pre-eclampsia trial in which there was a comparison of the performance of first-trimester screening for preterm preeclampsia using the Fetal Medicine Foundation model vs a traditional history-based risk scoring system. A subgroup of women from the trial with spontaneous onset of delivery (labor with intact membranes or preterm prelabor rupture of membranes) was included in this study and was arbitrarily divided into 3 groups according to the risk for preterm preeclampsia as determined by the Fetal Medicine Foundation model at 11 to 13 weeks' gestation as follows: group 1 low risk (˂1/100); group 2 intermediate risk (1/50 to 1/100); and group 3 high risk (˃1/50). A survival analysis was carried out using a Kaplan-Meier estimator and a Cox regression analysis with stratification by the 3 preeclampsia risk groups. Occurrence of spontaneous birth in the study groups was compared using log-rank tests and hazard ratios.

RESULTS

The study population comprised 10,820 cases with delivery after spontaneous onset of labor among the 16,451 cases who participated in the Screening programme for pre-eclampsia trial. There were 9795 cases in group 1, 583 in group 2, and 442 in group 3. The gestational age at delivery was <28, <32, <35, <37, and <40 weeks in 0.29%, 0.64%, 1.68%, 4.52%, and 44.97% of cases, respectively, in group 1; 0.69%, 1.71%, 3.26%, 7.72%, and 55.23% of cases, respectively, in group 2; and 0.45%, 1.81%, 5.66%, 13.80%, and 63.12% of cases, respectively, in group 3. The curve profile of gestational age at spontaneous birth in the 3 study groups was significantly different overall and in pairwise comparisons (P values <.001). The Cox regression analysis showed that risks increased for spontaneous birth by 18% when the intermediate-risk group was compared with the low-risk group (P˂.001) and by 41% when the high-risk group was compared with the low-risk group (P˂.001).

CONCLUSION

In this study that investigated birth after spontaneous onset of labor in women without preeclampsia, there were 2 major findings. First, the duration of pregnancy decreased with increasing first-trimester risk for preeclampsia. Second, in the high-risk group, when compared with the low-risk group, the risk for spontaneous birth was 4 times higher at a gestational age of 24 to 26 weeks, 3 times higher at 28 to 32 weeks, and 2 times higher at 34 to 39 weeks. These differences present major clinical implications for antepartum counselling, monitoring, and interventions in these pregnancies.

Validation of the first-trimester machine learning model for predicting pre-eclampsia in an Asian population

OBJECTIVES

To evaluate the performance of an artificial intelligence (AI) and machine learning (ML) model for first-trimester screening for pre-eclampsia in a large Asian population.

METHODS

This was a secondary analysis of a multicenter prospective cohort study in 10 935 participants with singleton pregnancies attending for routine pregnancy care at 11-13+6 weeks of gestation in seven regions in Asia between December 2016 and June 2018. We applied the AI+ML model for the first-trimester prediction of preterm pre-eclampsia (<37 weeks), term pre-eclampsia (≥37 weeks), and any pre-eclampsia, which was derived and tested in a cohort of pregnant participants in the UK (Model 1). This model comprises maternal factors with measurements of mean arterial pressure, uterine artery pulsatility index, and serum placental growth factor (PlGF). The model was further retrained with adjustments for analyzers used for biochemical testing (Model 2). Discrimination was assessed by area under the receiver operating characteristic curve (AUC). The Delong test was used to compare the AUC of Model 1, Model 2, and the Fetal Medicine Foundation (FMF) competing risk model.

RESULTS

The predictive performance of Model 1 was significantly lower than that of the FMF competing risk model in the prediction of preterm pre-eclampsia (0.82, 95% confidence interval [CI] 0.77-0.87 vs. 0.86, 95% CI 0.811-0.91, P = 0.019), term pre-eclampsia (0.75, 95% CI 0.71-0.80 vs. 0.79, 95% CI 0.75-0.83, P = 0.006), and any pre-eclampsia (0.78, 95% CI 0.74-0.81 vs. 0.82, 95% CI 0.79-0.84, P < 0.001). Following the retraining of the data with adjustments for the PlGF analyzers, the performance of Model 2 for predicting preterm pre-eclampsia, term pre-eclampsia, and any pre-eclampsia was improved with the AUC values increased to 0.84 (95% CI 0.80-0.89), 0.77 (95% CI 0.73-0.81), and 0.80 (95% CI 0.76-0.83), respectively. There were no differences in AUCs between Model 2 and the FMF competing risk model in the prediction of preterm pre-eclampsia (P = 0.135) and term pre-eclampsia (P = 0.084). However, Model 2 was inferior to the FMF competing risk model in predicting any pre-eclampsia (P = 0.024).

CONCLUSION

This study has demonstrated that following adjustment for the biochemical marker analyzers, the predictive performance of the AI+ML prediction model for pre-eclampsia in the first trimester was comparable to that of the FMF competing risk model in an Asian population.

Hypertensive Disorders of Pregnancy: Diagnosis, Management and Timing of Birth

Hypertensive disorders of pregnancy are significant contributors to maternal and perinatal morbidity and mortality. The definition, classification, and management of these disorders have evolved over time. Notably, the disease classification enables caretakers to manage the disease as well as safeguard maternal and fetal health. The approach and management for pregnancies with gestational and chronic hypertension or pre-eclampsia with or without severe features should be adequately elucidated to mitigate adverse perinatal outcomes. This review aimed to present the most recent definition and classification of hypertensive disorders of pregnancy to address their management, determine the optimal timing of birth, and establish short- and long-term follow-up protocols following parturition.

First-trimester predictive models for adverse pregnancy outcomes-a base for implementation of strategies to prevent cardiovascular disease development

Introduction

This study aimed to establish efficient, cost-effective, and early predictive models for adverse pregnancy outcomes based on the combinations of a minimum number of miRNA biomarkers, whose altered expression was observed in specific pregnancy-related complications and selected maternal clinical characteristics.

Methods

This retrospective study included singleton pregnancies with gestational hypertension (GH, n = 83), preeclampsia (PE, n = 66), HELLP syndrome (n = 14), fetal growth restriction (FGR, n = 82), small for gestational age (SGA, n = 37), gestational diabetes mellitus (GDM, n = 121), preterm birth in the absence of other complications (n = 106), late miscarriage (n = 34), stillbirth (n = 24), and 80 normal term pregnancies. MiRNA gene expression profiling was performed on the whole peripheral venous blood samples collected between 10 and 13 weeks of gestation using real-time reverse transcription polymerase chain reaction (RT-PCR).

Results

Most pregnancies with adverse outcomes were identified using the proposed approach (the combinations of selected miRNAs and appropriate maternal clinical characteristics) (GH, 69.88%; PE, 83.33%; HELLP, 92.86%; FGR, 73.17%; SGA, 81.08%; GDM on therapy, 89.47%; and late miscarriage, 84.85%). In the case of stillbirth, no addition of maternal clinical characteristics to the predictive model was necessary because a high detection rate was achieved by a combination of miRNA biomarkers only [91.67% cases at 10.0% false positive rate (FPR)].

Conclusion

The proposed models based on the combinations of selected cardiovascular disease-associated miRNAs and maternal clinical variables have a high predictive potential for identifying women at increased risk of adverse pregnancy outcomes; this can be incorporated into routine first-trimester screening programs. Preventive programs can be initiated based on these models to lower cardiovascular risk and prevent the development of metabolic/cardiovascular/cerebrovascular diseases because timely implementation of beneficial lifestyle strategies may reverse the dysregulation of miRNAs maintaining and controlling the cardiovascular system.

MRI assessed placental volume and adverse pregnancy outcomes: Secondary analysis of prospective cohort study

INTRODUCTION

Our goal was to evaluate the potential utility of magnetic resonance imaging (MRI) placental volume as an assessment of placental insufficiency.

METHODS

Secondary analysis of a prospective cohort undergoing serial placental MRIs at two academic tertiary care centers. The population included 316 participants undergoing MRI up to three times throughout gestation. MRI was used to calculate placental volume in milliliters (ml). Placental-mediated adverse pregnancy outcome (cAPO) included preeclampsia with severe features, abnormal antenatal surveillance, and perinatal mortality. Serial measurements were grouped as time point 1 (TP1) <22 weeks, TP2 22 0/7-29 6/7 weeks, and TP3 ≥30 weeks. Mixed effects models compared change in placental volume across gestation between cAPO groups. Association between cAPO and placental volume was determined using logistic regression at each TP with discrimination evaluated using area under receiver operator curve (AUC). Placental volume was then added to known clinical predictive variables and evaluated with test characteristics and calibration.

RESULTS

59 (18.7 %) of 316 participants developed cAPO. Placental volume growth across gestation was slower in the cAPO group (p < 0.001). Placental volume was lower in the cAPO group at all time points, and alone was moderately predictive of cAPO at TP3 (AUC 0.756). Adding placental volume to clinical variables had moderate discrimination at all time points, with strongest test characteristics at TP3 (AUC 0.792) with sensitivity of 77.5 % and specificity of 75.3 % at a predicted probability cutoff of 15 %.

DISCUSSION

MRI placental volume warrants further study for assessment of placental insufficiency, particularly later in gestation.

Maternal angiogenic factor disruptions prior to clinical diagnosis of preeclampsia: insights from the REVAMP study

Preeclampsia is characterized by impaired angiogenesis and assessment of angiogenic factors can play a crucial role in the early diagnosis of preeclampsia. The current study reports the levels of angiogenic factors longitudinally from early pregnancy in women with preeclampsia and in the subtypes of preeclampsia, to identify their role in early prediction of preeclampsia. A total of 1154 women with singleton pregnancies were recruited in early pregnancy from 2 hospitals. Blood samples were collected, plasma samples were separated and stored at four time points across gestation: V1 = 11-14 weeks, V2 = 18-22 weeks, V3 = 26-28 weeks, and V4 = at delivery. The current study includes a total of 108 women developed preeclampsia (PE), and 216 matched controls. Angiogenic factors were estimated using commercially available ELISA kits. Receiver operating characteristic (ROC) curves were used to evaluate the potential diagnostic value in the prediction of PE. Lower levels of VEGF, PlGF, and higher levels of sEng and sEng/PlGF ratio (p < 0.05 for all) predate clinical diagnosis in women with preeclampsia. sEng levels and sEng/PlGF ratio showed significant correlation with odds of preeclampsia at all the timepoints. This study identifies a cut off of 33.5 for sFlt-1/PlGF and 25.9 for sEng/PlGF for prediction of early onset preeclampsia. This study reports various angiogenic factors serially across gestation in a general population to identify women at risk of developing preeclampsia and its subtypes. The study also reports a potential biomarker and a pragmatic window for estimation of angiogenic markers to identify women at risk.

Placental Origins of Preeclampsia: Insights from Multi-Omic Studies

Preeclampsia (PE) is a major cause of maternal and neonatal morbidity and mortality worldwide, with the placenta playing a central role in disease pathophysiology. This review synthesizes recent advancements in understanding the molecular mechanisms underlying PE, focusing on placental genes, proteins, and genetic variants identified through multi-omic approaches. Transcriptomic studies in bulk placental tissue have identified many dysregulated genes in the PE placenta, including the PE signature gene, Fms-like tyrosine kinase 1 (FLT1). Emerging single-cell level transcriptomic data have revealed key cell types and molecular signatures implicated in placental dysfunction and PE. However, the considerable variability among studies underscores the need for standardized methodologies and larger sample sizes to enhance the reproducibility of results. Proteomic profiling of PE placentas has identified numerous PE-associated proteins, offering insights into potential biomarkers and pathways implicated in PE pathogenesis. Despite significant progress, challenges such as inconsistencies in study findings and lack of validation persist. Recent fetal genome-wide association studies have identified multiple genetic loci associated with PE, with ongoing efforts to elucidate their impact on placental gene expression and function. Future directions include the integration of multi-omic data, validation of findings in diverse PE populations and clinical subtypes, and the development of analytical approaches and experimental models to study the complex interplay of placental and maternal factors in PE etiology. These insights hold promise for improving risk prediction, diagnosis, and management of PE, ultimately reducing its burden on maternal and neonatal health.

Machine learning: a new era for cardiovascular pregnancy physiology and cardio-obstetrics research

The maternal cardiovascular system undergoes functional and structural adaptations during pregnancy and postpartum to support increased metabolic demands of offspring and placental growth, labor, and delivery, as well as recovery from childbirth. Thus, pregnancy imposes physiological stress upon the maternal cardiovascular system, and in the absence of an appropriate response it imparts potential risks for cardiovascular complications and adverse outcomes. The proportion of pregnancy-related maternal deaths from cardiovascular events has been steadily increasing, contributing to high rates of maternal mortality. Despite advances in cardiovascular physiology research, there is still no comprehensive understanding of maternal cardiovascular adaptations in healthy pregnancies. Furthermore, current approaches for the prognosis of cardiovascular complications during pregnancy are limited. Machine learning (ML) offers new and effective tools for investigating mechanisms involved in pregnancy-related cardiovascular complications as well as the development of potential therapies. The main goal of this review is to summarize existing research that uses ML to understand mechanisms of cardiovascular physiology during pregnancy and develop prediction models for clinical application in pregnant patients. We also provide an overview of ML platforms that can be used to comprehensively understand cardiovascular adaptations to pregnancy and discuss the interpretability of ML outcomes, the consequences of model bias, and the importance of ethical consideration in ML use.

Prevention of Pregnancy Complications Using a Multimodal Lifestyle, Screening, and Medical Model

Prevention of pregnancy complications related to the "great obstetrical syndromes" (preeclampsia, fetal growth restriction, spontaneous preterm labor, and stillbirth) is a global research and clinical management priority. These syndromes share many common pathophysiological mechanisms that may contribute to altered placental development and function. The resulting adverse pregnancy outcomes are associated with increased maternal and perinatal morbidity and mortality and increased post-partum risk of cardiometabolic disease. Maternal nutritional and environmental factors are known to play a significant role in altering bidirectional communication between fetal-derived trophoblast cells and maternal decidual cells and contribute to abnormal placentation. As a result, lifestyle-based interventions have increasingly been recommended before, during, and after pregnancy, in order to reduce maternal and perinatal morbidity and mortality and decrease long-term risk. Antenatal screening strategies have been developed following extensive studies in diverse populations. Multivariate preeclampsia screening using a combination of maternal, biophysical, and serum biochemical markers is recommended at 11-14 weeks' gestation and can be performed at the same time as the first-trimester ultrasound and blood tests. Women identified as high-risk can be offered prophylactic low dose aspirin and monitored with angiogenic factor assessment from 22 weeks' gestation, in combination with clinical assessment, serum biochemistry, and ultrasound. Lifestyle factors can be reassessed during counseling related to antenatal screening interventions. The integration of lifestyle interventions, pregnancy screening, and medical management represents a conceptual advance in pregnancy care that has the potential to significantly reduce pregnancy complications and associated later life cardiometabolic adverse outcomes.

Comparison of different methods of first-trimester screening for preterm pre-eclampsia: cohort study

OBJECTIVE

To compare the predictive performance of three different mathematical models for first-trimester screening of pre-eclampsia (PE), which combine maternal risk factors with mean arterial pressure (MAP), uterine artery pulsatility index (UtA-PI) and serum placental growth factor (PlGF), and two risk-scoring systems.

METHODS

This was a prospective cohort study performed in eight fetal medicine units in five different regions of Spain between September 2017 and December 2019. All pregnant women with singleton pregnancy and a non-malformed live fetus attending their routine ultrasound examination at 11 + 0 to 13 + 6 weeks' gestation were invited to participate in the study. Maternal characteristics and medical history were recorded and measurements of MAP, UtA-PI, serum PlGF and pregnancy-associated plasma protein-A (PAPP-A) were converted into multiples of the median (MoM). Risks for term PE, preterm PE (< 37 weeks' gestation) and early PE (< 34 weeks' gestation) were calculated according to the FMF competing-risks model, the Crovetto et al. logistic regression model and the Serra et al. Gaussian model. PE classification was also performed based on the recommendations of the National Institute for Health and Care Excellence (NICE) and the American College of Obstetricians and Gynecologists (ACOG). We estimated detection rates (DR) with their 95% CIs at a fixed 10% screen-positive rate (SPR), as well as the area under the receiver-operating-characteristics curve (AUC) for preterm PE, early PE and all PE for the three mathematical models. For the scoring systems, we calculated DR and SPR. Risk calibration was also assessed.

RESULTS

The study population comprised 10 110 singleton pregnancies, including 32 (0.3%) that developed early PE, 72 (0.7%) that developed preterm PE and 230 (2.3%) with any PE. At a fixed 10% SPR, the FMF, Crovetto et al. and Serra et al. models detected 82.7% (95% CI, 69.6-95.8%), 73.8% (95% CI, 58.7-88.9%) and 79.8% (95% CI, 66.1-93.5%) of early PE; 72.7% (95% CI, 62.9-82.6%), 69.2% (95% CI, 58.8-79.6%) and 74.1% (95% CI, 64.2-83.9%) of preterm PE; and 55.1% (95% CI, 48.8-61.4%), 47.1% (95% CI, 40.6-53.5%) and 53.9% (95% CI, 47.4-60.4%) of all PE, respectively. The best correlation between predicted and observed cases was achieved by the FMF model, with an AUC of 0.911 (95% CI, 0.879-0.943), a slope of 0.983 (95% CI, 0.846-1.120) and an intercept of 0.154 (95% CI, -0.091 to 0.397). The NICE criteria identified 46.7% (95% CI, 35.3-58.0%) of preterm PE at 11% SPR and ACOG criteria identified 65.9% (95% CI, 55.4-76.4%) of preterm PE at 33.8% SPR.

CONCLUSIONS

The best performance of screening for preterm PE is achieved by mathematical models that combine maternal factors with MAP, UtA-PI and PlGF, as compared to risk-scoring systems such as those of NICE and ACOG. While all three algorithms show similar results in terms of overall prediction, the FMF model showed the best performance at an individual level. © 2024 International Society of Ultrasound in Obstetrics and Gynecology.

Machine Learning Algorithms Versus Classical Regression Models in Pre-Eclampsia Prediction: A Systematic Review

PURPOSE OF REVIEW

Machine learning (ML) approaches are an emerging alternative for healthcare risk prediction. We aimed to synthesise the literature on ML and classical regression studies exploring potential prognostic factors and to compare prediction performance for pre-eclampsia.

RECENT FINDINGS

From 9382 studies retrieved, 82 were included. Sixty-six publications exclusively reported eighty-four classical regression models to predict variable timing of onset of pre-eclampsia. Another six publications reported purely ML algorithms, whilst another 10 publications reported ML algorithms and classical regression models in the same sample with 8 of 10 findings that ML algorithms outperformed classical regression models. The most frequent prognostic factors were age, pre-pregnancy body mass index, chronic medical conditions, parity, prior history of pre-eclampsia, mean arterial pressure, uterine artery pulsatility index, placental growth factor, and pregnancy-associated plasma protein A. Top performing ML algorithms were random forest (area under the curve (AUC) = 0.94, 95% confidence interval (CI) 0.91-0.96) and extreme gradient boosting (AUC = 0.92, 95% CI 0.90-0.94). The competing risk model had similar performance (AUC = 0.92, 95% CI 0.91-0.92) compared with a neural network. Calibration performance was not reported in the majority of publications. ML algorithms had better performance compared to classical regression models in pre-eclampsia prediction. Random forest and boosting-type algorithms had the best prediction performance. Further research should focus on comparing ML algorithms to classical regression models using the same samples and evaluation metrics to gain insight into their performance. External validation of ML algorithms is warranted to gain insights into their generalisability.

Prospective Validation of First-Trimester Screening for Preterm Preeclampsia in Nulliparous Women (PREDICTION Study)

BACKGROUND

Fetal Medicine Foundation (FMF) studies suggest that preterm preeclampsia can be predicted in the first trimester by combining biophysical, biochemical, and ultrasound markers and prevented using aspirin. We aimed to evaluate the FMF preterm preeclampsia screening test in nulliparous women.

METHODS

We conducted a prospective multicenter cohort study of nulliparous women recruited at 11 to 14 weeks. Maternal characteristics, mean arterial blood pressure, PAPP-A (pregnancy-associated plasma protein A), PlGF (placental growth factor) in maternal blood, and uterine artery pulsatility index were collected at recruitment. The risk of preterm preeclampsia was calculated by a third party blinded to pregnancy outcomes. Receiver operating characteristic curves were used to estimate the detection rate (sensitivity) and the false-positive rate (1-specificity) for preterm (<37 weeks) and for early-onset (<34 weeks) preeclampsia according to the FMF screening test and according to the American College of Obstetricians and Gynecologists criteria.

RESULTS

We recruited 7554 participants including 7325 (97%) who remained eligible after 20 weeks of which 65 (0.9%) developed preterm preeclampsia, and 22 (0.3%) developed early-onset preeclampsia. Using the FMF algorithm (cutoff of ≥1 in 110 for preterm preeclampsia), the detection rate was 63.1% for preterm preeclampsia and 77.3% for early-onset preeclampsia at a false-positive rate of 15.8%. Using the American College of Obstetricians and Gynecologists criteria, the equivalent detection rates would have been 61.5% and 59.1%, respectively, for a false-positive rate of 34.3%.

CONCLUSIONS

The first-trimester FMF preeclampsia screening test predicts two-thirds of preterm preeclampsia and three-quarters of early-onset preeclampsia in nulliparous women, with a false-positive rate of ≈16%.

REGISTRATION

URL: https://www.clinicaltrials.gov; Unique identifier: NCT02189148.

The role of the PLGF in the prediction of the outcome in pregnancies with a small for gestational age fetus

PURPOSE

To explore the value of measuring maternal serum PLGF in the prediction of the outcome of small for gestational age fetuses (SGA).

METHODS

Singleton pregnancies referred with suspicion of SGA in the third trimester were included if they had: no indication for nor signs of imminent delivery, fetal abdominal circumference (AC) at or below the 10th centile and/or estimated fetal weight (EFW) at or below the 10th centile and/or umbilical artery pulsatility index (Umb-PI) at or above the 90th centile for gestation. Women with pre-eclampsia at presentation were excluded. Maternal blood was drawn at the first (index) visit and analyzed retrospectively.

RESULTS

Fifty-one fetuses were examined. Multiple regression analysis showed that family history of microsomia, index EFW and PLGF were significant predictors of the birthweight centile; index femur length centile and PLGF were significant predictors of pre-eclampsia; PLGF and index systolic blood pressure were significant predictors of iatrogenic preterm delivery < 37 weeks, whereas PLGF and index EFW were significant predictors of birthweight ≤ 5th centile and admission to the neonatal intensive care unit. For all outcomes, the addition of maternal-fetal parameters did not improve the prediction compared to PLGF alone. Using a cutoff of 0.3 MoM for PLGF would identify 94.1% of the pregnancies with iatrogenic preterm delivery and/or intra-uterine death and all of the cases that developed pre-eclampsia, for a screen positive rate of 54.9%. Women with PLGF ≤ 0.3 MoM had a poor fetal/maternal outcome (iatrogenic preterm delivery, pre-eclampsia, intra-uterine death) in 61.5% of cases.

CONCLUSION

In pregnancies complicated by SGA, PLGF identifies a very high-risk group that may benefit from intense surveillance.

Characteristics of the Basel Postpartum Hypertension Cohort (Basel-PPHT Cohort): An Interim Analysis

Postpartum hypertension (PPHT) is hypertension that persists or develops after delivery and is a frequent cause of readmission, affecting 10% of pregnancies. This interim analysis aims to describe the cohort and to determine the feasibility and acceptance of a home-based telemonitoring management strategy (HBTMS) in PPHT patients. Enrollment at the University Hospital Basel began during the 2020 SARS-CoV-2 pandemic. Maternity-ward patients were screened for preexisting hypertension, hypertensive disorders of pregnancy, and de novo PPHT. In this pragmatic non-randomized prospective trial, the participants chose the HBTMS or standard of care (SOC), which consisted of outpatient hypertension clinic appointments. The HBTMS was a smartphone application or a programmed spreadsheet to report blood pressure (BP), followed by telephone consultations. Three months postpartum, the participants underwent a 24 h BP measurement and a blood, biomarker, and urine analysis. A total of 311 participants were enrolled between 06/20 and 08/23. The mean age was 34 (±5.3) years. The current pregnancy history demonstrated the following (≥1 diagnosis possible): 10% had preexisting hypertension, 27.3% gestational hypertension, 53% preeclampsia (PE), 0.3% eclampsia, 6% HELLP (hemolysis, elevated liver enzymes, and low platelets), and 18.3% de novo PPHT. A family history of cardiovascular disease and PE was reported in 49.5% and 7.5%, respectively. In total, 23.3% were high-risk for PE. A total of 68.5% delivered via c-section, the mean hospitalization was 6.3 days (±3.9), and newborn intrauterine growth restriction occurred in 21%. A total of 99% of the participants chose the HBTMS. This analysis demonstrated that the HBTMS was accepted. This is vital in the immediate postpartum period and pertinent when the exposure of hospital visits should be avoided.

Early-pregnancy N-terminal pro-brain natriuretic peptide level is inversely associated with hypertensive disorders of pregnancy diagnosed after 35 weeks of gestation

Hypertensive disorders of pregnancy (HDP) are among the major causes of high maternal and fetal/neonatal morbidity and mortality rates. Patients with HDP have significantly elevated N-terminal pro-brain natriuretic peptide (NT-proBNP) levels at diagnosis; however, the NT-proBNP levels during early pregnancy are largely unknown. This study aimed to validate the association between HDP and NT-proBNP levels. This retrospective study evaluated 103 pregnant women who developed HDP diagnosed after 35 weeks of gestation and 667 who did not. The HDP group had significantly lower early-pregnancy NT-proBNP levels than the without HDP group. However, the two groups did not significantly differ in terms of the late-pregnancy NT-proBNP levels. After adjusting for confounding factors such as age, body mass index, parity, and blood pressure levels, high early-pregnancy NT-proBNP levels were associated with a lower HDP risk. Early-pregnancy NT-proBNP levels ≥ 60.5 pg/mL had a negative predictive value of 97.0% for ruling out HDP, with a sensitivity of 87.4% and specificity of 62.5%. In conclusion, elevated early-pregnancy NT-proBNP levels were associated with a lower HDP risk. Moreover, a cutoff point of ≥ 60.5 pg/mL for early-pregnancy NT-proBNP levels had a high negative predictive value and sensitivity for ruling out HDP. These findings can provide new clinical implications.