Prediction of stillbirth from maternal demographic and pregnancy characteristics

Interpretable machine learning to predict adverse perinatal outcomes: examining marginal predictive value of risk factors during pregnancy

BACKGROUND

The timely identification of nulliparas at high risk of adverse fetal and neonatal outcomes during pregnancy is crucial for initiating clinical interventions to prevent perinatal complications. Although machine learning methods have been applied to predict preterm birth and other pregnancy complications, many models do not provide explanations of their predictions, limiting the clinical use of the model.

OBJECTIVE

This study aimed to develop interpretable prediction models for a composite adverse perinatal outcome (stillbirth, neonatal death, estimated Combined Apgar score of <10, or preterm birth) at different points in time during the pregnancy and to evaluate the marginal predictive value of individual predictors in the context of a machine learning model.

STUDY DESIGN

This was a secondary analysis of the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-be data, a prospective cohort study in which 10,038 nulliparous pregnant individuals with singleton pregnancies were enrolled. Here, interpretable prediction models were developed using L1-regularized logistic regression for adverse perinatal outcomes using data available at 3 study visits during the pregnancy (visit 1: 6 0/7 to 13 6/7 weeks of gestation; visit 2: 16 0/7 to 21 6/7 weeks of gestation; visit 3: 22 0/7 to 29 6/7 weeks of gestation). We identified the important predictors for each model using SHapley Additive exPlanations, a model-agnostic method of computing explanations of model predictions, and evaluated the marginal predictive value of each predictor using the DeLong test.

RESULTS

Our interpretable machine learning model had an area under the receiver operating characteristic curves of 0.617 (95% confidence interval, 0.595-0.639; all predictor variables at visit 1), 0.652 (95% confidence interval, 0.631-0.673; all predictor variables at visit 2), and 0.673 (95% confidence interval, 0.651-0.694; all predictor variables at visit 3). For all visits, the placental biomarker inhibin A was a valuable predictor, as including inhibin A resulted in better performance in predicting adverse perinatal outcomes (P<.001, all visits). At visit 1, endoglin was also a valuable predictor (P<.001). At visit 2, free beta human chorionic gonadotropin (P=.001) and uterine artery pulsatility index (P=.023) were also valuable predictors. At visit 3, cervical length was also a valuable predictor (P<.001).

CONCLUSION

Despite various advances in predictive modeling in obstetrics, the accurate prediction of adverse perinatal outcomes remains difficult. Interpretable machine learning can help clinicians understand how predictions are made, but barriers exist to the widespread clinical adoption of machine learning models for adverse perinatal outcomes. A better understanding of the evolution of risk factors for adverse perinatal outcomes throughout pregnancy is necessary for the development of effective interventions.

Prediction of stillbirth low resource setting in Northern Uganda

Background

Women of Afro-Caribbean and Asian origin are more at risk of stillbirths. However, there are limited tools built for risk-prediction models for stillbirth within sub-Saharan Africa. Therefore, we examined the predictors for stillbirth in low resource setting in Northern Uganda.

Methods

Prospective cohort study at St. Mary’s hospital Lacor in Northern Uganda. Using Yamane’s 1967 formula for calculating sample size for cohort studies using finite population size, the required sample size was 379 mothers. We doubled the number (to > 758) to cater for loss to follow up, miscarriages, and clients opting out of the study during the follow-up period. Recruited 1,285 pregnant mothers at 16–24 weeks, excluded those with lethal congenital anomalies diagnosed on ultrasound. Their history, physical findings, blood tests and uterine artery Doppler indices were taken, and the mothers were encouraged to continue with routine prenatal care until the time for delivery. While in the delivery ward, they were followed up in labour until delivery by the research team. The primary outcome was stillbirth 24 + weeks with no signs of life. Built models in RStudio. Since the data was imbalanced with low stillbirth rate, used ROSE package to over-sample stillbirths and under-sample live-births to balance the data. We cross-validated the models with the ROSE-derived data using K (10)-fold cross-validation and obtained the area under curve (AUC) with accuracy, sensitivity and specificity.

Results

The incidence of stillbirth was 2.5%. Predictors of stillbirth were history of abortion (aOR = 3.07, 95% CI 1.11—8.05, p = 0.0243), bilateral end-diastolic notch (aOR = 3.51, 95% CI 1.13—9.92, p = 0.0209), personal history of preeclampsia (aOR = 5.18, 95% CI 0.60—30.66, p = 0.0916), and haemoglobin 9.5 – 12.1 g/dL (aOR = 0.33, 95% CI 0.11—0.93, p = 0.0375). The models’ AUC was 75.0% with 68.1% accuracy, 69.1% sensitivity and 67.1% specificity.

Conclusion

Risk factors for stillbirth include history of abortion and bilateral end-diastolic notch, while haemoglobin of 9.5—12.1 g/dL is protective.

Maternal Race and Stillbirth: Cohort Study and Systematic Review with Meta-Analysis

Accurate identification of independent predictors of stillbirth is needed to define preventive strategies. We aim to examine the independent contribution of maternal race in the risk of stillbirth after adjusting for maternal characteristics and medical history. There are two components to the study: first, prospective screening in 168,966 women with singleton pregnancies coordinated by the Fetal Medicine Foundation (FMF) and second, a systematic review and meta-analysis of studies reporting on race and stillbirth. In the FMF study, logistic regression analysis found that in black women, the risk of stillbirth, after adjustment for confounders, was higher than in white women (odds ratio 1.78, 95% confidence interval 1.50 to 2.11). The risk for other racial groups was not significantly different. The literature search identified 20 studies that provided data on over 6,500,000 pregnancies, but only 10 studies provided risks adjusted for some maternal characteristics; consequently, the majority of these studies did not provide accurate contribution of different racial groups to the prediction of stillbirth. It is concluded that in women of black origin, the risk of stillbirth, after adjustment for confounders, is about twofold higher than in white women. Consequently, closer surveillance should be granted for these women.

External validation of prognostic models to predict stillbirth using International Prediction of Pregnancy Complications (IPPIC) Network database: individual participant data meta-analysis

OBJECTIVE

Stillbirth is a potentially preventable complication of pregnancy. Identifying women at high risk of stillbirth can guide decisions on the need for closer surveillance and timing of delivery in order to prevent fetal death. Prognostic models have been developed to predict the risk of stillbirth, but none has yet been validated externally. In this study, we externally validated published prediction models for stillbirth using individual participant data (IPD) meta-analysis to assess their predictive performance.

METHODS

MEDLINE, EMBASE, DH-DATA and AMED databases were searched from inception to December 2020 to identify studies reporting stillbirth prediction models. Studies that developed or updated prediction models for stillbirth for use at any time during pregnancy were included. IPD from cohorts within the International Prediction of Pregnancy Complications (IPPIC) Network were used to validate externally the identified prediction models whose individual variables were available in the IPD. The risk of bias of the models and cohorts was assessed using the Prediction study Risk Of Bias ASsessment Tool (PROBAST). The discriminative performance of the models was evaluated using the C-statistic, and calibration was assessed using calibration plots, calibration slope and calibration-in-the-large. Performance measures were estimated separately in each cohort, as well as summarized across cohorts using random-effects meta-analysis. Clinical utility was assessed using net benefit.

RESULTS

Seventeen studies reporting the development of 40 prognostic models for stillbirth were identified. None of the models had been previously validated externally, and the full model equation was reported for only one-fifth (20%, 8/40) of the models. External validation was possible for three of these models, using IPD from 19 cohorts (491 201 pregnant women) within the IPPIC Network database. Based on evaluation of the model development studies, all three models had an overall high risk of bias, according to PROBAST. In the IPD meta-analysis, the models had summary C-statistics ranging from 0.53 to 0.65 and summary calibration slopes ranging from 0.40 to 0.88, with risk predictions that were generally too extreme compared with the observed risks. The models had little to no clinical utility, as assessed by net benefit. However, there remained uncertainty in the performance of some models due to small available sample sizes.

CONCLUSIONS

The three validated stillbirth prediction models showed generally poor and uncertain predictive performance in new data, with limited evidence to support their clinical application. The findings suggest methodological shortcomings in their development, including overfitting. Further research is needed to further validate these and other models, identify stronger prognostic factors and develop more robust prediction models. © 2021 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Predicting singleton antepartum stillbirth by the demographic Fetal Medicine Foundation Risk Calculator—A retrospective case-control study

Objective

To assess the risk of singleton intrauterine fetal death (IUFD) in women by the demographic setting of the online Fetal Medicine Foundation (FMF) Stillbirth Risk Calculator.

Methods

Retrospective single-centre case-control study involving 144 women having suffered IUFD and 247 women after delivery of a live-born singleton. Nonparametric receiver operating characteristics (ROC) analyses were performed to predict the prognostic power of the FMF Stillbirth risk score and to generate a cut-off value to discriminate best between the event of IUFD versus live birth.

Results

Women in the IUFD cohort born a significantly higher overall risk with a median FMF risk score of 0.45% (IQR 0.23–0.99) compared to controls [0.23% (IQR 0.21–0.29); p<0.001]. Demographic factors contributing to an increased risk of IUFD in our cohort were maternal obesity (p = 0.002), smoking (p<0.001), chronic hypertension (p = 0.015), antiphospholipid syndrome (p = 0.017), type 2 diabetes (p<0.001), and insulin requirement (p<0.001). ROC analyses showed an area under the curve (AUC) of 0.72 (95% CI 0.67–0.78; p<0.001) for predicting overall IUFD and an AUC of 0.72 (95% CI 0.64–0.80; p<0.001), respectively, for predicting IUFD excluding congenital malformations. The FMF risk score at a cut-off of 0.34% (OR 6.22; 95% CI 3.91–9.89; p<0.001) yielded an 82% specificity and 58% sensitivity in predicting IUFD with a positive and negative predictive value of 0.94% and 99.84%, respectively.

Conclusion

The FMF Stillbirth Risk Calculator based upon maternal demographic and obstetric characteristics only may help identify women at low risk of antepartum stillbirth.

Development and validation of model for prediction of placental dysfunction-related stillbirth from maternal factors, fetal weight and uterine artery Doppler at mid-gestation

OBJECTIVE

To examine the performance of a model combining maternal risk factors, uterine artery pulsatility index (UtA-PI) and estimated fetal weight (EFW) at 19-24 weeks' gestation, for predicting all antepartum stillbirths and those due to impaired placentation, in a training dataset used for development of the model and in a validation dataset.

METHODS

The data for this study were derived from prospective screening for adverse obstetric outcome in women with singleton pregnancy attending for routine pregnancy care at 19 + 0 to 24 + 6 weeks' gestation. The study population was divided into a training dataset used to develop prediction models for placental dysfunction-related antepartum stillbirth and a validation dataset to which the models were then applied. Multivariable logistic regression analysis was used to develop a model based on a combination of maternal risk factors, EFW Z-score and UtA-PI multiples of the normal median. We examined the predictive performance of the model by, first, the ability of the model to discriminate between the stillbirth and live-birth groups, using the area under the receiver-operating-characteristics curve (AUC) and the detection rate (DR) at a fixed false-positive rate (FPR) of 10%, and, second, calibration by measurements of calibration slope and intercept.

RESULTS

The study population of 131 514 pregnancies included 131 037 live births and 477 (0.36%) stillbirths. There are four main findings of this study. First, 92.5% (441/477) of stillbirths were antepartum and 7.5% (36/477) were intrapartum, and 59.2% (261/441) of antepartum stillbirths were observed in association with placental dysfunction and 40.8% (180/441) were unexplained or due to other causes. Second, placental dysfunction accounted for 80.1% (161/201) of antepartum stillbirths at < 32 weeks' gestation, 54.2% (52/96) at 32 + 0 to 36 + 6 weeks and 33.3% (48/144) at ≥ 37 weeks. Third, the risk of placental dysfunction-related antepartum stillbirth increased with increasing maternal weight and decreasing maternal height, was 3-fold higher in black than in white women, was 5.5-fold higher in parous women with previous stillbirth than in those with previous live birth, and was increased in smokers, in women with chronic hypertension and in parous women with a previous pregnancy complicated by pre-eclampsia and/or birth of a small-for-gestational-age baby. Fourth, in screening for placental dysfunction-related antepartum stillbirth by a combination of maternal risk factors, EFW and UtA-PI in the validation dataset, the DR at a 10% FPR was 62.3% (95% CI, 57.2-67.4%) and the AUC was 0.838 (95% CI, 0.799-0.878); these results were consistent with those in the dataset used for developing the algorithm and demonstrate high discrimination between affected and unaffected pregnancies. Similarly, the calibration slope was 1.029 and the intercept was -0.009, demonstrating good agreement between the predicted risk and observed incidence of placental dysfunction-related antepartum stillbirth. The performance of screening was better for placental dysfunction-related antepartum stillbirth at < 37 weeks' gestation compared to at term (DR at a 10% FPR, 69.8% vs 29.2%).

CONCLUSIONS

Screening at mid-gestation by a combination of maternal risk factors, EFW and UtA-PI can predict a high proportion of placental dysfunction-related stillbirths and, in particular, those that occur preterm. Such screening provides poor prediction of unexplained stillbirth or stillbirth due to other causes. © 2021 International Society of Ultrasound in Obstetrics and Gynecology.

Predictive performance for placental dysfunction related stillbirth of the competing risks model for small for gestational age fetuses

OBJECTIVES

First, to examine the predictive performance for placental dysfunction related stillbirths of the competing risks model for small for gestational age (SGA) fetuses based on a combination of maternal risk factors, estimated fetal weight (EFW) and uterine artery pulsatility index (UtA-PI); and second, to compare the performance of this model to that of stillbirth-specific model utilizing the same biomarkers and to the Royal College of Obstetricians and Gynecologists (RCOG) guideline for the investigation and management of the SGA fetus.

DESIGN

Prospective observational study.

SETTING

Two UK maternity hospitals.

POPULATION

131,514 women with singleton pregnancies attending for routine ultrasound examination at 19-24 weeks' gestation.

METHODS

The predictive performance for stillbirth achieved by three models was compared. Main outcome measure Placental dysfunction related stillbirth.

RESULTS

At 10% false positive rate, the competing risks model predicted 59%, 66% and 71% of placental dysfunction related stillbirths, at any gestation, at <37 weeks and at <32 weeks, respectively, which were similar to the respective figures of 62%, 70% and 73% for the stillbirth-specific model. At a screen positive rate of 21.8 %, as defined by the RCOG guideline, the competing risks model predicted 71%, 76% and 79% of placental dysfunction related stillbirths at any gestation, at <37 weeks and at <32 weeks, respectively, and the respective figures for the RCOG guideline were 40%, 44% and 42%.

CONCLUSION

The predictive performance for placental dysfunction related stillbirths by the competing risks model for SGA was similar to the stillbirth-specific model and superior to the RCOG guideline.

Artificial intelligence in obstetrics

This study reviews recent advances on the application of artificial intelligence for the early diagnosis of various maternal-fetal conditions such as preterm birth and abnormal fetal growth. It is found in this study that various machine learning methods have been successfully employed for different kinds of data capture with regard to early diagnosis of maternal-fetal conditions. With the more popular use of artificial intelligence, ethical issues should also be considered accordingly.

Ethnic variation in causes of stillbirth in high income countries: A systematic review and meta-analysis

OBJECTIVE

Inequities in stillbirth rate according to ethnicity persist in high income nations. The objective of the present study is to investigate whether causes of stillbirth differ by ethnicity in high-income nations.

METHODS

The following databases were searched since their inception to 1 February 2021: Medline, Embase, Scopus, CINAHL, Cochrane Library, and Global Health. Cohort, cross-sectional, and retrospective studies were included. Causes of stillbirth were aligned to the International Classification of Disease 10 for Perinatal Mortality (ICD10-PM) and pooled estimates were derived by meta-analysis.

RESULTS

Fifteen reports from three countries (72 555 stillbirths) were included. Seven ethnic groups - "Caucasian" (n = 11 studies), "African" (n = 11 studies), "Hispanic" (n = 7 studies), "Indigenous Australian" (n = 4 studies), "Asian" (n = 2 studies), "South Asian" (n = 2 studies), and "American Indian" (n = 1 study) - were identified. There was an overall paucity of recent, high-quality data for many ethnicities. For those with the greatest amount of data - Caucasian, African, and Hispanic - no major differences in the causes of stillbirth were identified.

CONCLUSION

There is a paucity of high-quality information on causes of stillbirth for many ethnicities. Improving investigation and standardizing classification of stillbirths is needed to assess whether causes of stillbirth differ across more diverse ethnic groups.

Novelelectronic health records applied for prediction of pre-eclampsia: Machine-learning algorithms

OBJECTIVE

To predict risk of pre-eclampsia (PE) in women using machine learning (ML) algorithms, based on electronic health records (EHR) collected at the early second trimester.

STUDY DESIGN

A total of 3759 cases of pregnancy who received antenatal care at Xinhua hospital Chongming branch Affiliated to Shanghai Jiaotong University were included in this retrospective EHR-based study. Thirty-eight candidate clinical parameters routinely available at the first visit in antenatal care were collected by manual chart review. Logistic regression (LR), random forest (RF), support vector machine (SVM) and extreme gradient boosting (XGBoost) were used to construct the prediction model. Features that contributed to the model predictions were identified using XGBoost.

OUTCOME MEASURES

The performance of ML models to predict women at risk of PE was quantified in terms of accuracy, precision, recall, false negative score, f1_score, brier score and the area under the receiver operating curve (auROC).

RESULTS

The XGboost model had the best prediction performance (accuracy = 0.920, precision = 0.447, recall = 0.789, f1_score = 0.571, auROC = 0.955). The most predictive feature of PE development was fasting plasma glucose, followed by mean blood pressure and body mass index. An easy-to-use model that a patient could answer independently still enabled accurate prediction, with auROC of 0.83.

CONCLUSION

risk of PE development can be predicted with excellent discriminative ability using ML algorithms based on EHR collected at the early second trimester. Future studies are needed to assess the real-world clinical utility of the model.

Racial Disparities in Stillbirth Risk Factors among non-Hispanic Black Women and non-Hispanic White Women in the United States

INTRODUCTION

Historically, stillbirth risk factors are more prevalent among non-Hispanic Black women than non-Hispanic White women, including age < 20, lower formal educational attainment, prepregnancy obesity, smoking, hypertension, diabetes, short interpregnancy interval, small for gestational age newborn, late prenatal care, and previous cesarean birth. We examined whether these disparities have changed since 2011 and identified a group of risk factors that differed between Black women and White women when accounting for correlations among variables.

METHODS

In a random sample of 315 stillbirths from the National Center for Health Statistics' 2016 fetal death data, Black women and White women were compared for each risk factor using t-tests or chi-square tests. Variables with p ≤ .20 were analyzed using multivariate analysis of variance.

RESULTS

In this sample, Black women experiencing stillbirth were less likely to have a Bachelor's degree (12.94% vs. 28.49%, p = .04), and more likely to be obese (44.5% vs. 29.1%, p = .01) than White women. Multivariate analysis accounting for correlations among variables showed a group of risk factors that differed between Black women and White women: age < 20, lower education, prepregnancy obesity, hypertension (chronic and pregnancy-associated), nulliparity before stillbirth, and earlier gestation.

CLINICAL IMPLICATIONS

Less formal education, obesity, age <20, hypertension, chronic and pregnancy-associated, nulliparity, and earlier gestation are important to consider in multilevel stillbirth prevention interventions to decrease racial disparity in stillbirth. Respectfully listening to women and taking their concerns seriously is one way nurses and other health care providers can promote equity in health outcomes for childbearing women.

Predictive Model for Late Stillbirth Among Antenatal Hypertensive Women

Objective

To develop a predictive model for late stillbirth among women with hypertensive disorders of pregnancy (HDP) in low- and middle-income countries.

Materials and Methods

Study was part of the WHO newborn birth defect (NBBD) project and included all stillbirths occurring in the facility from November 2015 to December 2020. The age and parity matched subjects with HDP having live birth were taken as controls. All significant predictors were analyzed and a predictive model was developed.

Results

Out of 69,007 deliveries, 1691(24.5/1000) were stillborn. HDP was seen in (390/1691, 23.0%), in 265/390 (67.4%) cases it occurred at or after 28 weeks of gestation and were included as cases. On comparing the cases with controls, the significant factors were estimated fetal weight less than 2000 gms (P < 0.001, OR 10.3), poor antenatal care (p < 0.001, OR-5.9), family history of hypertension (p < 0.018, OR-4.4) and the presence of gestational hypertension (p = 0.001, OR 2.2). The predictive model had sensitivity and specificity of 80.3% and 70.03%, respectively, the receiver operating curve showed the area under the curve(AUC) in the range of good prediction (0.846).

Conclusion

The predictive model could play a potential role in stillbirth prevention in women with HDP in low- and middle-income countries.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13224-021-01561-3.

Impact of Chronic Hypertension and Antihypertensive Treatment on Adverse Perinatal Outcomes: Systematic Review and Meta-Analysis

Background Maternal chronic hypertension is associated with adverse pregnancy outcomes. Previous studies examined the association between either chronic hypertension or antihypertensive treatment and adverse pregnancy outcomes. We aimed to synthesize the evidence on the effect of chronic hypertension/antihypertensive treatment on adverse pregnancy outcomes. Methods and Results Medline/PubMed, EMBASE, and Web of Science were searched; we included observational studies and assessed the effect of race/ethnicity, where possible, following a registered protocol (CRD42019120088). Random-effects meta-analyses were used. A total of 81 studies were identified on chronic hypertension, and a total of 16 studies were identified on antihypertensive treatment. Chronic hypertension was associated with higher odds of preeclampsia (adjusted odd ratio [aOR], 5.43; 95% CI, 3.85-7.65); cesarean section (aOR, 1.87; 95% CI, 1.6-2.16); maternal mortality (aOR, 4.80; 95% CI, 3.04-7.58); preterm birth (aOR, 2.23; 95% CI, 1.96-2.53); stillbirth (aOR, 2.32; 95% CI, 2.22-2.42); and small for gestational age (SGA) (aOR, 1.96; 95% CI, 1.6-2.40). Subgroup analyses indicated that maternal race/ethnicity does not influence the observed associations. Women with chronic hypertension on antihypertensive treatment (versus untreated) had higher odds of SGA (aOR, 1.86; 95% CI, 1.38-2.50). Conclusions Chronic hypertension is associated with adverse pregnancy outcomes, and these associations appear to be independent of maternal race/ethnicity. In women with chronic hypertension, those on treatment had a higher risk of SGA, although the number of studies was limited. This could result from a direct effect of the treatment or because severe hypertension during pregnancy is a risk factor for SGA and women with severe hypertension are more likely to be treated. The effect of antihypertensive treatment on SGA needs to be further tested with large randomized controlled trials.

Synthetic minority oversampling of vital statistics data with generative adversarial networks

Objective

Minority oversampling is a standard approach used for adjusting the ratio between the classes on imbalanced data. However, established methods often provide modest improvements in classification performance when applied to data with extremely imbalanced class distribution and to mixed-type data. This is usual for vital statistics data, in which the outcome incidence dictates the amount of positive observations. In this article, we developed a novel neural network-based oversampling method called actGAN (activation-specific generative adversarial network) that can derive useful synthetic observations in terms of increasing prediction performance in this context.

Materials and Methods

From vital statistics data, the outcome of early stillbirth was chosen to be predicted based on demographics, pregnancy history, and infections. The data contained 363 560 live births and 139 early stillbirths, resulting in class imbalance of 99.96% and 0.04%. The hyperparameters of actGAN and a baseline method SMOTE-NC (Synthetic Minority Over-sampling Technique-Nominal Continuous) were tuned with Bayesian optimization, and both were compared against a cost-sensitive learning-only approach.

Results

While SMOTE-NC provided mixed results, actGAN was able to improve true positive rate at a clinically significant false positive rate and area under the curve from the receiver-operating characteristic curve consistently.

Discussion

Including an activation-specific output layer to a generator network of actGAN enables the addition of information about the underlying data structure, which overperforms the nominal mechanism of SMOTE-NC.

Conclusions

actGAN provides an improvement to the prediction performance for our learning task. Our developed method could be applied to other mixed-type data prediction tasks that are known to be afflicted by class imbalance and limited data availability.

Proposing a machine-learning based method to predict stillbirth before and during delivery and ranking the features: nationwide retrospective cross-sectional study

Background

Stillbirth is defined as fetal loss in pregnancy beyond 28 weeks by WHO. In this study, a machine-learning based method is proposed to predict stillbirth from livebirth and discriminate stillbirth before and during delivery and rank the features.

Method

A two-step stack ensemble classifier is proposed for classifying the instances into stillbirth and livebirth at the first step and then, classifying stillbirth before delivery from stillbirth during the labor at the second step. The proposed SE has two consecutive layers including the same classifiers. The base classifiers in each layer are decision tree, Gradient boosting classifier, logistics regression, random forest and support vector machines which are trained independently and aggregated based on Vote boosting method. Moreover, a new feature ranking method is proposed in this study based on mean decrease accuracy, Gini Index and model coefficients to find high-ranked features.

Results

IMAN registry dataset is used in this study considering all births at or beyond 28th gestational week from 2016/04/01 to 2017/01/01 including 1,415,623 live birth and 5502 stillbirth cases. A combination of maternal demographic features, clinical history, fetal properties, delivery descriptors, environmental features, healthcare service provider descriptors and socio-demographic features are considered. The experimental results show that our proposed SE outperforms the compared classifiers with the average accuracy of 90%, sensitivity of 91%, specificity of 88%. The discrimination of the proposed SE is assessed and the average AUC of ±95%, CI of 90.51% ±1.08 and 90% ±1.12 is obtained on training dataset for model development and test dataset for external validation, respectively. The proposed SE is calibrated using isotopic nonparametric calibration method with the score of 0.07. The process is repeated 10,000 times and AUC of SE classifiers using random different training datasets as null distribution. The obtained p-value to assess the specificity of the proposed SE is 0.0126 which shows the significance of the proposed SE.

Conclusions

Gestational age and fetal height are two most important features for discriminating livebirth from stillbirth. Moreover, hospital, province, delivery main cause, perinatal abnormality, miscarriage number and maternal age are the most important features for classifying stillbirth before and during delivery.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-021-03658-z.

Protocol for the development and validation of a risk prediction model for stillbirths from 35 weeks gestation in Australia

BACKGROUND

Despite advances in the care of women and their babies in the past century, an estimated 1.7 million babies are born still each year throughout the world. A robust method to estimate a pregnant woman's individualized risk of late-pregnancy stillbirth is needed to inform decision-making around the timing of birth to reduce the risk of stillbirth from 35 weeks of gestation in Australia, a high-resource setting.

METHODS

This is a protocol for a cross-sectional study of all late-pregnancy births in Australia (2005-2015) from 35 weeks of gestation including 5188 stillbirths among 3.1 million births at an estimated rate of 1.7 stillbirths per 1000 births. A multivariable logistic regression model will be developed in line with current Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD) guidelines to estimate the gestation-specific probability of stillbirth with prediction intervals. Candidate predictors were identified from systematic reviews and clinical consultation and will be described through univariable regression analysis. To generate a final model, elimination by backward stepwise multivariable logistic regression will be performed. The model will be internally validated using bootstrapping with 1000 repetitions and externally validated using a temporally unique dataset. Overall model performance will be assessed with R², calibration, and discrimination. Calibration will be reported using a calibration plot with 95% confidence intervals (α = 0.05). Discrimination will be measured by the C-statistic and area underneath the receiver-operator curves. Clinical usefulness will be reported as positive and negative predictive values, and a decision curve analysis will be considered.

DISCUSSION

A robust method to predict a pregnant woman's individualized risk of late-pregnancy stillbirth is needed to inform timely, appropriate care to reduce stillbirth. Among existing prediction models designed for obstetric use, few have been subject to internal and external validation and many fail to meet recommended reporting standards. In developing a risk prediction model for late-gestation stillbirth with both providers and pregnant women in mind, we endeavor to develop a validated model for clinical use in Australia that meets current reporting standards.

Predicting risk of stillbirth and preterm pregnancies with machine learning

Modelling the risk of abnormal pregnancy-related outcomes such as stillbirth and preterm birth have been proposed in the past. Commonly they utilize maternal demographic and medical history information as predictors, and they are based on conventional statistical modelling techniques. In this study, we utilize state-of-the-art machine learning methods in the task of predicting early stillbirth, late stillbirth and preterm birth pregnancies. The aim of this experimentation is to discover novel risk models that could be utilized in a clinical setting. A CDC data set of almost sixteen million observations was used conduct feature selection, parameter optimization and verification of proposed models. An additional NYC data set was used for external validation. Algorithms such as logistic regression, artificial neural network and gradient boosting decision tree were used to construct individual classifiers. Ensemble learning strategies of these classifiers were also experimented with. The best performing machine learning models achieved 0.76 AUC for early stillbirth, 0.63 for late stillbirth and 0.64 for preterm birth while using a external NYC test data. The repeatable performance of our models demonstrates robustness that is required in this context. Our proposed novel models provide a solid foundation for risk prediction and could be further improved with the addition of biochemical and/or biophysical markers.

The PLANES study: a protocol for a randomised controlled feasibility study of the placental growth factor (PlGF) blood test-informed care versus standard care alone for women with a small for gestational age fetus at or after 32 + 0 weeks' gestation

Background

Stillbirth remains a major concern across the globe and in some high-resource countries, such as the UK; efforts to reduce the rate have achieved only modest reductions. One third of stillborn babies are small for gestational age (SGA), and these pregnancies are also at risk of neonatal adverse outcomes and lifelong health problems, especially when delivered preterm. Current UK clinical guidance advocates regular monitoring and early term delivery of the SGA fetus; however, the most appropriate regimen for surveillance of these babies remains unclear and often leads to increased intervention for a large number of these women. This pilot trial will determine the feasibility of a large-scale trial refining the risk of adverse pregnancy outcome in SGA pregnancies using biomarkers of placental function sFlt-1/PlGF, identifying and intervening in only those deemed at highest risk of stillbirth.

Methods

PLANES is a randomised controlled feasibility study of women with an SGA fetus that will be conducted at two tertiary care hospitals in the UK. Once identified on ultrasound, women will be randomised into two groups in a 3:1 ratio in favour of sFlt-1/PlGF ratio led management vs standard care. Women with an SGA fetus and a normal sFlt-1/PlGF ratio will have a repeat ultrasound and sFlt-1/PlGF ratio every 2 weeks with planned birth delayed until 40 weeks. In those women with an SGA fetus and an abnormal sFlt-1/PlGF ratio, we will offer birth from 37 weeks or sooner if there are other concerning features on ultrasound. Women assigned to standard care will have an sFlt-1/PlGF ratio taken, but the results will be concealed from the clinical team, and the woman’s pregnancy will be managed as per the local NHS hospital policy. This integrated mixed method study will also involve a health economic analysis and a perspective work package exploring trial feasibility through interviews and questionnaires with participants, their partners, and clinicians.

Discussion

Our aim is to determine feasibility through the assessment of our ability to recruit and retain participants to the study. Results from this pilot study will inform the design of a future large randomised controlled trial that will be adequately powered for adverse pregnancy outcome. Such a study would provide the evidence needed to guide future management of the SGA fetus.

Trial registration

ISRCTN58254381. Registered on 4 July 2019

Can risk prediction models help us individualise stillbirth prevention? A systematic review and critical appraisal of published risk models

BACKGROUND

Stillbirth prevention is an international priority - risk prediction models could individualise care and reduce unnecessary intervention, but their use requires evaluation.

OBJECTIVES

To identify risk prediction models for stillbirth, and assess their potential accuracy and clinical benefit in practice.

SEARCH STRATEGY

MEDLINE, Embase, DH-DATA and AMED databases were searched from inception to June 2019 using terms relevant to stillbirth, perinatal mortality and prediction models. The search was compliant with Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.

SELECTION CRITERIA

Studies developing and/or validating prediction models for risk of stillbirth developed for application during pregnancy.

DATA COLLECTION AND ANALYSIS

Study screening and data extraction were conducted in duplicate, using the CHARMS checklist. Risk of bias was appraised using the PROBAST tool.

RESULTS

The search identified 2751 citations. Fourteen studies reporting development of 69 models were included. Variables consistently included were: ethnicity, body mass index, uterine artery Doppler, pregnancy-associated plasma protein and placental growth factor. For almost all models there were significant concerns about risk of bias. Apparent model performance (i.e. in the development dataset) was highest in models developed for use later in pregnancy and including maternal characteristics, and ultrasound and biochemical variables, but few were internally validated and none were externally validated.

CONCLUSIONS

Almost all models identified were at high risk of bias. There are first-trimester models of possible clinical benefit in early risk stratification; these require validation and clinical evaluation. There were few later pregnancy models but, if validated, these could be most relevant to individualised discussions around timing of birth.

TWEETABLE ABSTRACT

Prediction models using maternal factors, blood tests and ultrasound could individualise stillbirth prevention, but existing models are at high risk of bias.

New predictors of early impaired placentation preceding miscarriage before 10 weeks of gestation in IVF pregnancies: A prospective study

INTRODUCTION

In a recent study of 10,011 pregnant women, 95% of miscarriages occurred before routine ultrasound scan at 11-14 weeks. Our study aimed to identify early first trimester parameters which may predict miscarriage before 10 weeks of gestation for in vitro fertilization (IVF) pregnancies.

METHODS

A cohort of 115 healthy IVF patients with a singleton viable embryo in early first trimester were studied in a tertiary university-affiliated medical center (April 2017-June 2018). Calculations included gestational age (GA); ultrasound evaluation of crown-rump length (CRL), mean gestational sac diameter (GSD) and volume (GSV), mean yolk sac diameter (YSD) and volume (YSV); fetal heart rate (FHR), mean uterine arteries pulsatility index (UtA-PI); and maternal blood placental protein 13 (PP13) levels. Patients were divided into three groups by GA; and early miscarriage versus ongoing pregnancy after GA 10 weeks.

RESULTS

Early fetal loss occurred in 14.8% of patients; miscarriage group had higher discrepancy between calculated and measured GA (P < 0.001), lower GSD and GSV (P = 0.005 and P = 0.02, respectively), significantly different YSD and YSV, and lower GSD/YSD and GSV/YSV ratios (P = 0.001 and P = 0.003, respectively). UtA-PI/CRL ratio was higher in patients with miscarriage at GA 46-48 days and GA >48 days (P = 0.034 and P = 0.026, respectively). PP13/CRL ratio was higher in patients with miscarriage at GA >48 days (P = 0.041).

DISCUSSION

In IVF pregnancies with live embryo at first ultrasound scan, high UtA-PI/CRL and maternal blood PP13/CRL ratios may indicate impaired placentation preceded early pregnancy loss. A larger cohort is needed to further verify these predictions.

Factors contributing to late stillbirth among women with pregnancy hypertension in a developing country

OBJECTIVE

To investigate the factors associated with late stillbirth among women with hypertensive disorders of pregnancy (HDP).

MATERIAL AND METHODS

The clinical details of women with HDP having late stillbirth were compared with controls having livebirth.

RESULTS

Total 208 cases and 288 controls were included in the study. Inadequate antenatal visits (p < 0.001, OR-5.92). birth weight < 2000 gms (p < 0.001, OR 10.3) and BW/PW ratio > 8 contributed significantly (p = 0.0001, OR-5.6) to stillbirth.

CONCLUSION

Poor antenatal care, birth weight below 2000gms and high BW/PW ratio was associated with a higher risk of stillbirth.

Stillbirth risk prediction using machine learning for a large cohort of births from Western Australia, 1980-2015

Quantification of stillbirth risk has potential to support clinical decision-making. Studies that have attempted to quantify stillbirth risk have been hampered by small event rates, a limited range of predictors that typically exclude obstetric history, lack of validation, and restriction to a single classifier (logistic regression). Consequently, predictive performance remains low, and risk quantification has not been adopted into antenatal practice. The study population consisted of all births to women in Western Australia from 1980 to 2015, excluding terminations. After all exclusions there were 947,025 livebirths and 5,788 stillbirths. Predictive models for stillbirth were developed using multiple machine learning classifiers: regularised logistic regression, decision trees based on classification and regression trees, random forest, extreme gradient boosting (XGBoost), and a multilayer perceptron neural network. We applied 10-fold cross-validation using independent data not used to develop the models. Predictors included maternal socio-demographic characteristics, chronic medical conditions, obstetric complications and family history in both the current and previous pregnancy. In this cohort, 66% of stillbirths were observed for multiparous women. The best performing classifier (XGBoost) predicted 45% (95% CI: 43%, 46%) of stillbirths for all women and 45% (95% CI: 43%, 47%) of stillbirths after the inclusion of previous pregnancy history. Almost half of stillbirths could be potentially identified antenatally based on a combination of current pregnancy complications, congenital anomalies, maternal characteristics, and medical history. Greatest sensitivity is achieved with addition of current pregnancy complications. Ensemble classifiers offered marginal improvement for prediction compared to logistic regression.

Maternal characteristics and pregnancy outcomes of women with chronic hypertension: a population-based study

Background We aimed to study the maternal characteristics and obstetric and neonatal outcomes in pregnant mothers with chronic hypertension (CHTN) compared to non-CHTN. Methods The study was a population-based cohort study, and a PEARL-Peristat Study (PPS) for the year of 2017. There were 20,210 total births including 19,762 singleton and 448 multiple births. We excluded multiple gestations from the analysis as they differ in fetal growth, duration of gestation and have a higher rate of obstetric and neonatal complications. We compared the maternal characteristics of mothers with pre-existing HTN with non-hypertensive mothers and studied the obstetric and neonatal outcomes including cesarean section, stillbirths, prematurity, macrosomia and postpartum hemorrhage (PPH). Results We identified 223 births of mothers with essential HTN. The overall prevalence of CHTN in our population was 1.1% (223/20,210). In regard to maternal characteristics, women with CHTN were at or above 35 years of age at the time of delivery 58.9% compared to non-CHTN women 18.7%, P-value <0.001. Pre-existing diabetes was found more in women with CHTN 15.1% compared to non-CHTN women 1.9%, P-value <0.001; while obesity was found in 64% of women with CHTN compared to 32.5% in non-CHTN women, P-value <0.001. Preterm birth was noted in 26% compared to 8% in CHTN compared to non-CHTN women, respectively, P-value <0.001. The rate of stillbirth was similar between the two groups, 0.9% compared to 0.6% in CHTN compared to non-CHTN women, respectively, P-value 0.369. Conclusion Hypertensive mothers have multiple other comorbidities. When compared to the general population, they are older, parous, diabetic and obese with an increased risk of preterm birth and cesarean deliveries. Lifestyle modification, extensive pre-conceptional counseling and multidisciplinary antenatal care are required for such a high-risk group.

The International Federation of Gynecology and Obstetrics (FIGO) initiative on pre-eclampsia: A pragmatic guide for first-trimester screening and prevention

Pre‐eclampsia (PE) is a multisystem disorder that typically affects 2%–5% of pregnant women and is one of the leading causes of maternal and perinatal morbidity and mortality, especially when the condition is of early onset. Globally, 76 000 women and 500 000 babies die each year from this disorder. Furthermore, women in low‐resource countries are at a higher risk of developing PE compared with those in high‐resource countries. Although a complete understanding of the pathogenesis of PE remains unclear, the current theory suggests a two‐stage process. The first stage is caused by shallow invasion of the trophoblast, resulting in inadequate remodeling of the spiral arteries. This is presumed to lead to the second stage, which involves the maternal response to endothelial dysfunction and imbalance between angiogenic and antiangiogenic factors, resulting in the clinical features of the disorder. Accurate prediction and uniform prevention continue to elude us. The quest to effectively predict PE in the first trimester of pregnancy is fueled by the desire to identify women who are at high risk of developing PE, so that necessary measures can be initiated early enough to improve placentation and thus prevent or at least reduce the frequency of its occurrence. Furthermore, identification of an “at risk” group will allow tailored prenatal surveillance to anticipate and recognize the onset of the clinical syndrome and manage it promptly. PE has been previously defined as the onset of hypertension accompanied by significant proteinuria after 20 weeks of gestation. Recently, the definition of PE has been broadened. Now the internationally agreed definition of PE is the one proposed by the International Society for the Study of Hypertension in Pregnancy (ISSHP). According to the ISSHP, PE is defined as systolic blood pressure at ≥140 mm Hg and/or diastolic blood pressure at ≥90 mm Hg on at least two occasions measured 4 hours apart in previously normotensive women and is accompanied by one or more of the following new‐onset conditions at or after 20 weeks of gestation: 1.Proteinuria (i.e. ≥30 mg/mol protein:creatinine ratio; ≥300 mg/24 hour; or ≥2 + dipstick); 2.Evidence of other maternal organ dysfunction, including: acute kidney injury (creatinine ≥90 μmol/L; 1 mg/dL); liver involvement (elevated transaminases, e.g. alanine aminotransferase or aspartate aminotransferase >40 IU/L) with or without right upper quadrant or epigastric abdominal pain; neurological complications (e.g. eclampsia, altered mental status, blindness, stroke, clonus, severe headaches, and persistent visual scotomata); or hematological complications (thrombocytopenia–platelet count <150 000/μL, disseminated intravascular coagulation, hemolysis); or 3.Uteroplacental dysfunction (such as fetal growth restriction, abnormal umbilical artery Doppler waveform analysis, or stillbirth). It is well established that a number of maternal risk factors are associated with the development of PE: advanced maternal age; nulliparity; previous history of PE; short and long interpregnancy interval; use of assisted reproductive technologies; family history of PE; obesity; Afro‐Caribbean and South Asian racial origin; co‐morbid medical conditions including hyperglycemia in pregnancy; pre‐existing chronic hypertension; renal disease; and autoimmune diseases, such as systemic lupus erythematosus and antiphospholipid syndrome. These risk factors have been described by various professional organizations for the identification of women at risk of PE; however, this approach to screening is inadequate for effective prediction of PE. PE can be subclassified into: 1.Early‐onset PE (with delivery at <34+0 weeks of gestation); 2.Preterm PE (with delivery at <37+0 weeks of gestation); 3.Late‐onset PE (with delivery at ≥34+0 weeks of gestation); 4.Term PE (with delivery at ≥37+0 weeks of gestation). These subclassifications are not mutually exclusive. Early‐onset PE is associated with a much higher risk of short‐ and long‐term maternal and perinatal morbidity and mortality. Obstetricians managing women with preterm PE are faced with the challenge of balancing the need to achieve fetal maturation in utero with the risks to the mother and fetus of continuing the pregnancy longer. These risks include progression to eclampsia, development of placental abruption and HELLP (hemolysis, elevated liver enzyme, low platelet) syndrome. On the other hand, preterm delivery is associated with higher infant mortality rates and increased morbidity resulting from small for gestational age (SGA), thrombocytopenia, bronchopulmonary dysplasia, cerebral palsy, and an increased risk of various chronic diseases in adult life, particularly type 2 diabetes, cardiovascular disease, and obesity. Women who have experienced PE may also face additional health problems in later life, as the condition is associated with an increased risk of death from future cardiovascular disease, hypertension, stroke, renal impairment, metabolic syndrome, and diabetes. The life expectancy of women who developed preterm PE is reduced on average by 10 years. There is also significant impact on the infants in the long term, such as increased risks of insulin resistance, diabetes mellitus, coronary artery disease, and hypertension in infants born to pre‐eclamptic women. The International Federation of Gynecology and Obstetrics (FIGO) brought together international experts to discuss and evaluate current knowledge on PE and develop a document to frame the issues and suggest key actions to address the health burden posed by PE. FIGO's objectives, as outlined in this document, are: (1) To raise awareness of the links between PE and poor maternal and perinatal outcomes, as well as to the future health risks to mother and offspring, and demand a clearly defined global health agenda to tackle this issue; and (2) To create a consensus document that provides guidance for the first‐trimester screening and prevention of preterm PE, and to disseminate and encourage its use. Based on high‐quality evidence, the document outlines current global standards for the first‐trimester screening and prevention of preterm PE, which is in line with FIGO good clinical practice advice on first trimester screening and prevention of pre‐eclampsia in singleton pregnancy.1 It provides both the best and the most pragmatic recommendations according to the level of acceptability, feasibility, and ease of implementation that have the potential to produce the most significant impact in different resource settings. Suggestions are provided for a variety of different regional and resource settings based on their financial, human, and infrastructure resources, as well as for research priorities to bridge the current knowledge and evidence gap. To deal with the issue of PE, FIGO recommends the following: Public health focus: There should be greater international attention given to PE and to the links between maternal health and noncommunicable diseases (NCDs) on the Sustainable Developmental Goals agenda. Public health measures to increase awareness, access, affordability, and acceptance of preconception counselling, and prenatal and postnatal services for women of reproductive age should be prioritized. Greater efforts are required to raise awareness of the benefits of early prenatal visits targeted at reproductive‐aged women, particularly in low‐resource countries. Universal screening: All pregnant women should be screened for preterm PE during early pregnancy by the first‐trimester combined test with maternal risk factors and biomarkers as a one‐step procedure. The risk calculator is available free of charge at https://fetalmedicine.org/research/assess/preeclampsia. FIGO encourages all countries and its member associations to adopt and promote strategies to ensure this. The best combined test is one that includes maternal risk factors, measurements of mean arterial pressure (MAP), serum placental growth factor (PLGF), and uterine artery pulsatility index (UTPI). Where it is not possible to measure PLGF and/or UTPI, the baseline screening test should be a combination of maternal risk factors with MAP, and not maternal risk factors alone. If maternal serum pregnancy‐associated plasma protein A (PAPP‐A) is measured for routine first‐trimester screening for fetal aneuploidies, the result can be included for PE risk assessment. Variations to the full combined test would lead to a reduction in the performance screening. A woman is considered high risk when the risk is 1 in 100 or more based on the first‐trimester combined test with maternal risk factors, MAP, PLGF, and UTPI. Contingent screening: Where resources are limited, routine screening for preterm PE by maternal factors and MAP in all pregnancies and reserving measurements of PLGF and UTPI for a subgroup of the population (selected on the basis of the risk derived from screening by maternal factors and MAP) can be considered. Prophylactic measures: Following first‐trimester screening for preterm PE, women identified at high risk should receive aspirin prophylaxis commencing at 11–14+6 weeks of gestation at a dose of ~150 mg to be taken every night until 36 weeks of gestation, when delivery occurs, or when PE is diagnosed. Low‐dose aspirin should not be prescribed to all pregnant women. In women with low calcium intake (<800 mg/d), either calcium replacement (≤1 g elemental calcium/d) or calcium supplementation (1.5–2 g elemental calcium/d) may reduce the burden of both early‐ and late‐onset PE.

How many roads lead to stillbirth rate reduction? A 30-year analysis of risk factors in a Northern Italy University care center

BACKGROUND

Stillbirths affect more than 2.5 million pregnancies worldwide every year and the progress in reducing stillbirth rates is slower than that required by World Health Organization. The aim of the present study was to investigate which factors were associated with stillbirths in a University Hospital in the North of Italy, over a time span of 30 years. The goal was to identify which factors are potentially modifiable to reduce stillbirth rate.

METHODS

Retrospective case-control study (358 stillbirths, 716 livebirths) subdivided into two study periods (1987-2006 and 2007-2017).

RESULTS

The prevalence of conception obtained by assisted reproductive technologies, pregnancy at advanced maternal age, and complications of pregnancy such as preeclampsia, fetal growth restriction (FGR), and other fetal diseases (abnormal fetal conditions including fetal anemia, fetal hydrops, TORCH infections) increased through the years of the study. Despite a rising prevalence, the last 10 years showed a significant reduction in stillbirths associated with preeclampsia and FGR. Similarly, the risk of stillbirth related to abnormal fetal conditions decreased in the second study period and a history of previous stillbirth becomes a nonsignificant risk factor.

CONCLUSIONS

Altogether these results suggest that in pregnancies perceived as "high risk" (i.e. previous stillbirth, preeclampsia, FGR, abnormal fetal conditions) appropriate care and follow-up can indeed lower stillbirth rates. In conclusion, the road to stillbirth prevention passes inevitably through awareness and recognition of risk factors.

The prediction of early preeclampsia: Results from a longitudinal proteomics study

OBJECTIVES

To identify maternal plasma protein markers for early preeclampsia (delivery <34 weeks of gestation) and to determine whether the prediction performance is affected by disease severity and presence of placental lesions consistent with maternal vascular malperfusion (MVM) among cases.

STUDY DESIGN

This longitudinal case-control study included 90 patients with a normal pregnancy and 33 patients with early preeclampsia. Two to six maternal plasma samples were collected throughout gestation from each woman. The abundance of 1,125 proteins was measured using high-affinity aptamer-based proteomic assays, and data were modeled using linear mixed-effects models. After data transformation into multiples of the mean values for gestational age, parsimonious linear discriminant analysis risk models were fit for each gestational-age interval (8-16, 16.1-22, 22.1-28, 28.1-32 weeks). Proteomic profiles of early preeclampsia cases were also compared to those of a combined set of controls and late preeclampsia cases (n = 76) reported previously. Prediction performance was estimated via bootstrap.

RESULTS

We found that 1) multi-protein models at 16.1-22 weeks of gestation predicted early preeclampsia with a sensitivity of 71% at a false-positive rate (FPR) of 10%. High abundance of matrix metalloproteinase-7 and glycoprotein IIbIIIa complex were the most reliable predictors at this gestational age; 2) at 22.1-28 weeks of gestation, lower abundance of placental growth factor (PlGF) and vascular endothelial growth factor A, isoform 121 (VEGF-121), as well as elevated sialic acid binding immunoglobulin-like lectin 6 (siglec-6) and activin-A, were the best predictors of the subsequent development of early preeclampsia (81% sensitivity, FPR = 10%); 3) at 28.1-32 weeks of gestation, the sensitivity of multi-protein models was 85% (FPR = 10%) with the best predictors being activated leukocyte cell adhesion molecule, siglec-6, and VEGF-121; 4) the increase in siglec-6, activin-A, and VEGF-121 at 22.1-28 weeks of gestation differentiated women who subsequently developed early preeclampsia from those who had a normal pregnancy or developed late preeclampsia (sensitivity 77%, FPR = 10%); 5) the sensitivity of risk models was higher for early preeclampsia with placental MVM lesions than for the entire early preeclampsia group (90% versus 71% at 16.1-22 weeks; 87% versus 81% at 22.1-28 weeks; and 90% versus 85% at 28.1-32 weeks, all FPR = 10%); and 6) the sensitivity of prediction models was higher for severe early preeclampsia than for the entire early preeclampsia group (84% versus 71% at 16.1-22 weeks).

CONCLUSION

We have presented herein a catalogue of proteome changes in maternal plasma proteome that precede the diagnosis of preeclampsia and can distinguish among early and late phenotypes. The sensitivity of maternal plasma protein models for early preeclampsia is higher in women with underlying vascular placental disease and in those with a severe phenotype.

Prediction of fetal loss in Chinese pregnant patients with systemic lupus erythematosus: a retrospective cohort study

OBJECTIVE

To develop a predictive model for fetal loss in women with systemic lupus erythematosus (SLE).

DESIGN

A retrospective cohort study.

SETTING

Data were collected in a tertiary medical centre, located in Shanghai, China, from September 2011 to May 2017.

PARTICIPANTS

338 pregnancies with SLE were analysed retrospectively. Cases of multiple pregnancy and those in which artificial abortion was performed for personal reasons were excluded.

PRIMARY OUTCOME MEASURES

Fetal loss was the primary outcome. A stepwise regression to identify the predictors related to the fetal loss and coefficient B of each variable was used to develop a predictive model and make a corresponding risk classification. The Hosmer-Lemeshow test, Omnibus test and area under the receiver-operating characteristic curve (AUC) were used to assess the goodness-of-fit and discrimination of the predictive model. A 10-fold cross validation was used to assess the model for overfitting.

RESULTS

Unplanned pregnancies (OR 2.84, 95% CI 1.12 to 7.22), C₃ hypocomplementemia (OR 5.46, 95% CI 2.30 to 12.97) and 24 hour-urinary protein level (0.3≤protein<1.0 g/24 hours: OR 2.10, 95% CI 0.63 to 6.95; protein≥1.0 g/24 hours: OR 5.89, 95% CI 2.30 to 15.06) were selected by the stepwise regression. The Hosmer-Lemeshow test resulted in p=0.325; the Omnibus test resulted in p<0.001 and the AUC was 0.829 (95% CI 0.744 to 0.91) in the regression model. The corresponding risk score classification was divided into low risk (0-3) and high risk groups (>3), with a sensitivity of 60.5%, a specificity of 93.3%, positive likelihood ratio of 9.03 and negative likelihood ratio of 0.42.

CONCLUSIONS

A predictive model for fetal loss in women with SLE was developed using the timing of conception, C₃ complement and 24 hour-urinary protein level. This model may help clinicians in identifying women with high risk pregnancies, thereby carrying out monitoring or/and interventions for improving fetal outcomes.

Screening and management of the small for gestational age fetus in the UK: A survey of practice

BACKGROUND

Antenatal detection of the small for gestational (SGA) fetus has become an important indicator of quality of antenatal care in the UK. This has been driven by a desire to reduce stillbirth in this at risk group.

METHODS

We conducted a postal survey of 187 NHS consultant units within the UK to determine what the current practice for the detection and subsequent management of the suspected SGA fetus was following the guidance from the Royal College of Obstetricians and Gynaecologists (RCOG) in 2013.

RESULTS

The survey was performed in 3 rounds between 2016 and 2017 with a response rate of 65%. 85% of units assessed risk factors for SGA at booking. 81% of units used a customized symphysis fundal height (SFH) chart to screen for SGA with 95% of them using a cut off of <10th centile to refer for ultrasound assessment. When ultrasound is used to detect SGA, 80% of units used estimated fetal weight (EFW), with 89% of these using a cut off of <10th centile to diagnose SGA. Umbilical artery (UA) Doppler monitoring was undertaken in 97% of management and 94% delivered after 37 weeks. Only 24% of units had a dedicated fetal growth clinic, whilst 48% of units were able to offer computerised CTG to monitor the SGA fetus.

CONCLUSIONS

Overall there is consistency in the screening methods for SGA (customised SFH charts) and identification of suspected SGA (SFH <10th centile, EFW <10th centile, UA monitoring and induction of labour at term). There was a low uptake of computerized CTG to monitor SGA babies and a low number of specialised fetal growth clinics.

Hypertensive disorders in pregnancy and stillbirth rates: a facility-based study in China

Objective

To assess the association between hypertensive disorders in pregnancy and the stillbirth rate.

Methods

We obtained all data from China's National Maternal Near Miss Surveillance System for 2012 to 2016. Associations between hypertensive disorders in pregnancy and stillbirths, stratified by fetus number and gestational age, were assessed using Poisson regression analysis with a robust variance estimator.

Findings

For the period, 6 970 032 births, including 66 494 stillbirths, were reported to the surveillance system. The weighted stillbirth rate in women with a hypertensive disorder in pregnancy was 21.9 per 1000 births. The risk was higher in those who had received few antenatal care visits or who were poorly educated. For singleton pregnancies, the adjusted risk ratio (aRR) for a stillbirth among women with hypertensive disorders in pregnancy compared with normotensive women was 3.1 (95% confidence interval, CI: 2.85-3.37). The aRR for hypertensive disorder subtypes was: 6.66 (95% CI: 5.57-7.96) for superimposed preeclampsia; 4.15 (95% CI: 3.81-4.52) for preeclampsia or eclampsia; 2.32 (95% CI: 1.87-2.88) for chronic hypertension; and 1.21 (95% CI: 1.08-1.36) for gestational hypertension. For multiple pregnancies, the association between stillbirths and hypertensive disorders in pregnancy was not significant, except for superimposed preeclampsia (aRR: 1.95; 95% CI: 1.28-2.97).

Conclusion

To minimize the incidence of stillbirths, more attention should be paid to chronic hypertension and superimposed preeclampsia in singleton pregnancies and to superimposed preeclampsia in multiple pregnancies. Better quality antenatal care and improved guidelines are needed in China.

A stillbirth calculator: Development and internal validation of a clinical prediction model to quantify stillbirth risk

Objective

To generate a clinical prediction tool for stillbirth that combines maternal risk factors to provide an evidence based approach for the identification of women who will benefit most from antenatal testing for stillbirth prevention.

Design

Retrospective cohort study

Setting

Midwestern United States quaternary referral center

Population

Singleton pregnancies undergoing second trimester anatomic survey from 1999–2009. Pregnancies with incomplete follow-up were excluded.

Methods

Candidate predictors were identified from the literature and univariate analysis. Backward stepwise logistic regression with statistical comparison of model discrimination, calibration and clinical performance was used to generate final models for the prediction of stillbirth. Internal validation was performed using bootstrapping with 1,000 repetitions. A stillbirth risk calculator and stillbirth risk score were developed for the prediction of stillbirth at or beyond 32 weeks excluding fetal anomalies and aneuploidy. Statistical and clinical cut-points were identified and the tools compared using the Integrated Discrimination Improvement.

Main outcome measures

Antepartum stillbirth

Results

64,173 women met inclusion criteria. The final stillbirth risk calculator and score included maternal age, black race, nulliparity, body mass index, smoking, chronic hypertension and pre-gestational diabetes. The stillbirth calculator and simple risk score demonstrated modest discrimination but clinically significant performance with no difference in overall performance between the tools [(AUC 0.66 95% CI 0.60–0.72) and (AUC 0.64 95% CI 0.58–0.70), (p = 0.25)].

Conclusion

A stillbirth risk score was developed incorporating maternal risk factors easily ascertained during prenatal care to determine an individual woman’s risk for stillbirth and provide an evidenced based approach to the initiation of antenatal testing for the prediction and prevention of stillbirth.

Prediction of stillbirth from placental growth factor at 19-24 weeks

OBJECTIVES

To investigate whether the addition of maternal serum placental growth factor (PlGF) measured at 19-24 weeks' gestation improves the performance of screening for stillbirth that is achieved by a combination of maternal factors, fetal biometry and uterine artery pulsatility index (UtA-PI) and to evaluate the performance of screening with this model for all stillbirths and those due to impaired placentation and unexplained or other causes.

METHODS

This was a prospective screening study of 70 003 singleton pregnancies including 268 stillbirths, carried out in two phases. The first phase included prospective measurement of UtA-PI and fetal biometry, which were available in all cases. The second phase included prospective measurement of maternal serum PlGF, which was available for 9870 live births and 86 antepartum stillbirths. The values of PlGF obtained from this screening study were simulated in the remaining cases based on bivariate Gaussian distributions, defined by the mean and standard deviations. Multivariable logistic regression analysis was used to determine whether the addition of maternal serum PlGF improved the performance of screening that was achieved by a combination of maternal factors, fetal biometry and UtA-PI.

RESULTS

Significant contribution to the prediction of stillbirth was provided by maternal factor-derived a-priori risk, multiples of the median values of PlGF, UtA-PI and fetal biometry Z-scores. A model combining these variables predicted 58% of all stillbirths and 84% of those due to impaired placentation, at a false-positive rate of 10%. Within the impaired-placentation group, the detection rate of stillbirth < 32 weeks' gestation was higher than that of stillbirth ≥ 37 weeks (97% vs 61%; P < 0.01).

CONCLUSIONS

A high proportion of stillbirths due to impaired placentation can be identified effectively in the second trimester of pregnancy using a combination of maternal factors, fetal biometry, uterine artery Doppler and maternal serum PlGF. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.

Prediction of stillbirth from biochemical and biophysical markers at 11-13 weeks

OBJECTIVES

To develop a model for the prediction of stillbirth that is based on a combination of maternal characteristics and medical history with first-trimester biochemical and biophysical markers and to evaluate the performance of screening with this model for all stillbirths and those due to impaired placentation and unexplained causes.

METHODS

This was a prospective screening study of 76 897 singleton pregnancies, including 76 629 live births and 268 (0.35%) antepartum stillbirths; 157 (59%) were secondary to impaired placentation and 111 (41%) were due to other or unexplained causes. Multivariable logistic regression analysis was used to determine if there was a significant contribution to prediction of stillbirth from the maternal factor-derived a-priori risk, fetal nuchal translucency thickness, ductus venosus pulsatility index for veins (DV-PIV), uterine artery pulsatility index (UtA-PI) and maternal serum free β-human chorionic gonadotropin and pregnancy-associated plasma protein-A (PAPP-A). The significant contributors were used to derive a model for first-trimester prediction of stillbirth.

RESULTS

Significant contribution to prediction of stillbirth was provided by maternal factors, PAPP-A, UtA-PI and DV-PIV. A model combining these variables predicted 40% of all stillbirths and 55% of those due to impaired placentation, at a false-positive rate of 10%. Within the impaired-placentation group, the detection rate of stillbirth < 32 weeks' gestation was higher than that of stillbirth ≥ 37 weeks (64% vs 42%).

CONCLUSIONS

A model based on maternal factors and first-trimester biomarkers can potentially predict more than half of subsequent stillbirths that occur due to impaired placentation. The extent to which such stillbirths could be prevented remains to be determined. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.

Prediction of stillbirth from maternal factors, fetal biometry and uterine artery Doppler at 19-24 weeks

OBJECTIVES

To evaluate the performance of screening for all stillbirths and those due to impaired placentation and unexplained or other causes using a combination of maternal factors, fetal biometry and uterine artery pulsatility index (UtA-PI) at 19-24 weeks' gestation and to compare this performance with that of screening by UtA-PI alone.

METHODS

This was a prospective screening study of 70 003 singleton pregnancies including 69 735 live births and 268 (0.38%) antepartum stillbirths; 159 (59%) were secondary to impaired placentation and 109 (41%) were due to other or unexplained causes. Multivariable logistic regression analysis was used to develop a model for prediction of stillbirth based on a combination of maternal factors, fetal biometry and UtA-PI.

RESULTS

Combined screening predicted 55% of all stillbirths, including 75% of those due to impaired placentation and 23% of those that were unexplained or due to other causes, at a false-positive rate of 10%. Within the impaired placentation group, the detection rate of stillbirth < 32 weeks' gestation was higher than that of stillbirth ≥ 37 weeks (88% vs 46%; P < 0.001). The performance of screening by the combined test was superior to that of selecting the high-risk group on the basis of UtA-PI > 90^th percentile for gestational age, which predicted 48% of all stillbirths, 70% of those due to impaired placentation and 15% of those that were unexplained or due to other causes.

CONCLUSIONS

Second-trimester screening by a combination of UtA-PI with maternal factors and fetal biometry can predict a high proportion of stillbirths and, in particular, those that are due to impaired placentation. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.

Prediction of stillbirth from placental growth factor at 11-13 weeks

OBJECTIVES

To investigate whether the addition of maternal serum placental growth factor (PlGF) measured at 11-13 weeks' gestation improves the performance of screening for stillbirths that is achieved by a combination of maternal factors and first-trimester biomarkers such as maternal serum pregnancy-associated plasma protein-A (PAPP-A), fetal ductus venosus pulsatility index for veins (DV-PIV) and uterine artery pulsatility index (UtA-PI) and to evaluate the performance of screening with this model for all stillbirths and those due to impaired placentation and unexplained causes.

METHODS

This was a prospective screening study of 45 452 singleton pregnancies including 45 225 live births and 227 (0.49%) antepartum stillbirths; 131 (58%) were secondary to impaired placentation and 96 (42%) were due to other or unexplained causes. Multivariable logistic regression analysis was used to determine whether the addition of maternal serum PlGF improved the performance of screening that was achieved by a combination of maternal factors and PAPP-A, DV-PIV and UtA-PI.

RESULTS

Significant contribution to the prediction of stillbirth was provided by maternal factor-derived a-priori risk and multiples of the median values of PlGF, DV-PIV and UtA-PI but not of serum PAPP-A. A model combining these variables predicted 42% of all stillbirths and 61% of those due to impaired placentation, at a false-positive rate of 10%; within the impaired placentation group the detection rate of stillbirth < 32 weeks' gestation was higher than that of stillbirth ≥ 37 weeks (71% vs 46%; P = 0.031).

CONCLUSIONS

A high proportion of stillbirths due to impaired placentation can be identified effectively in the first trimester of pregnancy. Addition of PlGF improves the performance of screening achieved by other maternal factors and biomarkers. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.