Routinely collected antenatal data for longitudinal prediction of preeclampsia in nulliparous women: a population-based study

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.

Preeclampsia and eclampsia: Enhanced detection and treatment for morbidity reduction

Preeclampsia is a life-threatening complication that develops in 2-8% of pregnancies. It is characterized by elevated blood pressure after 20 weeks of gestation and may progress to multiorgan dysfunction, leading to severe maternal and fetal morbidity and mortality. The only definitive treatment is delivery, and efforts are focused on early risk prediction, surveillance, and severity mitigation. Anesthesiologists, as part of the interdisciplinary team, should evaluate patients early in labor in order to optimize cardiovascular, pulmonary, and coagulation status. Neuraxial techniques are safe in the absence of coagulopathy and aid avoidance of general anesthesia, which is associated with high risk in these patients. This review aims to provide anaesthesiologists with a comprehensive update on the latest strategies and evidence-based practices for managing preeclampsia, with an emphasis on perioperative care.

Preeclampsia Prediction Using Machine Learning and Polygenic Risk Scores From Clinical and Genetic Risk Factors in Early and Late Pregnancies

BACKGROUND

Preeclampsia, a pregnancy-specific condition associated with new-onset hypertension after 20-weeks gestation, is a leading cause of maternal and neonatal morbidity and mortality. Predictive tools to understand which individuals are most at risk are needed.

METHODS

We identified a cohort of N=1125 pregnant individuals who delivered between May 2015 and May 2022 at Mass General Brigham Hospitals with available electronic health record data and linked genetic data. Using clinical electronic health record data and systolic blood pressure polygenic risk scores derived from a large genome-wide association study, we developed machine learning (XGBoost) and logistic regression models to predict preeclampsia risk.

RESULTS

Pregnant individuals with a systolic blood pressure polygenic risk score in the top quartile had higher blood pressures throughout pregnancy compared with patients within the lowest quartile systolic blood pressure polygenic risk score. In the first trimester, the most predictive model was XGBoost, with an area under the curve of 0.74. In late pregnancy, with data obtained up to the delivery admission, the best-performing model was XGBoost using clinical variables, which achieved an area under the curve of 0.91. Adding the systolic blood pressure polygenic risk score to the models did not improve the performance significantly based on De Long test comparing the area under the curve of models with and without the polygenic score.

CONCLUSIONS

Integrating clinical factors into predictive models can inform personalized preeclampsia risk and achieve higher predictive power than the current practice. In the future, personalized tools can be implemented to identify high-risk patients for preventative therapies and timely intervention to improve adverse maternal and neonatal outcomes.

An Interpretable Longitudinal Preeclampsia Risk Prediction Using Machine Learning

Background

Preeclampsia is a pregnancy-specific disease characterized by new onset hypertension after 20 weeks of gestation that affects 2-8% of all pregnancies and contributes to up to 26% of maternal deaths. Despite extensive clinical research, current predictive tools fail to identify up to 66% of patients who will develop preeclampsia. We sought to develop a tool to longitudinally predict preeclampsia risk.

Methods

In this retrospective model development and validation study, we examined a large cohort of patients who delivered at six community and two tertiary care hospitals in the New England region between 02/2015 and 06/2023. We used sociodemographic, clinical diagnoses, family history, laboratory, and vital signs data. We developed eight datasets at 14, 20, 24, 28, 32, 36, 39 weeks gestation and at the hospital admission for delivery. We created linear regression, random forest, xgboost, and deep neural networks to develop multiple models and compared their performance. We used Shapley values to investigate the global and local explainability of the models and the relationships between the predictive variables.

Findings

Our study population (N=120,752) had an incidence of preeclampsia of 5.7% (N=6,920). The performance of the models as measured using the area under the curve, AUC, was in the range 0.73-0.91, which was externally validated. The relationships between some of the variables were complex and non-linear; in addition, the relative significance of the predictors varied over the pregnancy. Compared to the current standard of care for preeclampsia risk stratification in the first trimester, our model would allow 48.6% more at-risk patients to be identified.

Interpretation

Our novel preeclampsia prediction tool would allow clinicians to identify patients at risk early and provide personalized predictions, as well as longitudinal predictions throughout pregnancy.

Funding

National Institutes of Health, Anesthesia Patient Safety Foundation.

RESEARCH IN CONTEXT

Evidence before this study: Current tools for the prediction of preeclampsia are lacking as they fail to identify up to 66% of the patients who develop preeclampsia. We searched PubMed, MEDLINE, and the Web of Science from database inception to May 1, 2023, using the keywords "deep learning", "machine learning", "preeclampsia", "artificial intelligence", "pregnancy complications", and "predictive models". We identified 13 studies that employed machine learning to develop prediction models for preeclampsia risk based on clinical variables. Among these studies, six included biomarkers such as serum placental growth factor, pregnancy-associated plasma protein A, and uterine artery pulsatility index, which are not routinely available in our clinical practice; two studies were in diverse cohorts of more than 100 000 patients, and two studies developed longitudinal predictions using medical records data. However, most studies have limited depth, concerns about data leakage, overfitting, or lack of generalizability.Added value of this study: We developed a comprehensive longitudinal predictive tool based on routine clinical data that can be used throughout pregnancy to predict the risk of preeclampsia. We tested multiple types of predictive models, including machine learning and deep learning models, and demonstrated high predictive power. We investigated the changes over different time points of individual and group variables and found previously known and novel relationships between variables such as red blood cell count and preeclampsia risk.Implications of all the available evidence: Longitudinal prediction of preeclampsia using machine learning can be achieved with high performance. Implementation of an accurate predictive tool within the electronic health records can aid clinical care and identify patients at heightened risk who would benefit from aspirin prophylaxis, increased surveillance, early diagnosis, and escalation in care. These results highlight the potential of using artificial intelligence in clinical decision support, with the ultimate goal of reducing iatrogenic preterm birth and improving perinatal care.

Prediction of Preeclampsia from Clinical and Genetic Risk Factors in Early and Late Pregnancy Using Machine Learning and Polygenic Risk Scores

Background

Preeclampsia, a pregnancy-specific condition associated with new-onset hypertension after 20 weeks gestation, is a leading cause of maternal and neonatal morbidity and mortality. Predictive tools to understand which individuals are most at risk are needed.

Methods

We identified a cohort of N=1,125 pregnant individuals who delivered between 05/2015-05/2022 at Mass General Brigham hospitals with available electronic health record (EHR) data and linked genetic data. Using clinical EHR data and systolic blood pressure polygenic risk scores (SBP PRS) derived from a large genome-wide association study, we developed machine learning (xgboost) and linear regression models to predict preeclampsia risk.

Results

Pregnant individuals with an SBP PRS in the top quartile had higher blood pressures throughout pregnancy compared to patients within the lowest quartile SBP PRS. In the first trimester, the most predictive model was xgboost, with an area under the curve (AUC) of 0.73. Adding the SBP PRS to the models improved the performance only of the linear regression model from AUC 0.70 to 0.71; the predictive power of other models remained unchanged. In late pregnancy, with data obtained up to the delivery admission, the best performing model was xgboost using clinical variables, which achieved an AUC of 0.91.

Conclusions

Integrating clinical and genetic factors into predictive models can inform personalized preeclampsia risk and achieve higher predictive power than the current practice. In the future, personalized tools can be implemented in clinical practice to identify high-risk patients for preventative therapies and timely intervention to improve adverse maternal and neonatal outcomes.

The Stockholm-Gotland perinatal cohort-A population-based cohort including longitudinal data throughout pregnancy and the postpartum period

BACKGROUND

Register-based reproductive and perinatal databases rarely contain detailed information from medical records or repeated measurements throughout pregnancy and delivery. This lack of enriched pregnancy and birth data led to the initiation of the Swedish Stockholm-Gotland Perinatal Cohort (SGPC).

OBJECTIVES

To describe the strengths of the SGPC, as well as the unique research questions that can be addressed using this cohort.

POPULATION

The SGPC is a prospectively collected, population-based cohort that includes all births (from 22 completed gestational weeks onwards) between 1 January 2008 and 15 June 2020 in the Stockholm and Gotland regions of Sweden (N 335,153 singleton and N 11,025 multiple pregnancies).

DESIGN

Descriptive study.

METHODS

The SGPC is based on the electronic medical records of women and their infants. The medical record system is used for all antenatal clinic visits and admissions, delivery and neonatal admissions, as well as postpartum clinical visits. SGPC has been further enriched with data linkages to 10 Swedish National Health Care and Quality Registers.

PRELIMINARY RESULTS

In contrast to other reproductive and perinatal databases available in Sweden, including the Medical Birth Register and the Pregnancy Register, SGPC contains highly detailed medical record data, including time-varying serial measurements for physiological parameters throughout pregnancy, delivery, and postpartum, for both mother and infant. These strengths have enabled studies that were previously inconceivable; the effects of serial measurements of pregnancy weight gain, changes in haemoglobin counts and blood pressure during pregnancy, fetal weight estimations by ultrasound, duration of stages and phases of labour, cervical dilatation and oxytocin use during delivery, and constructing reference curves for umbilical cord pH.

CONCLUSIONS

The SGPC-with its rich content, repeated measurements and linkages to numerous health care and quality registers-is a unique cohort that enables high-quality perinatal studies that would otherwise not be possible.

Development and evaluation of a nomogram for adverse outcomes of preeclampsia in Chinese pregnant women

Objective

Preeclampsia, the main cause of maternal and perinatal deaths, is associated with several maternal complications and adverse perinatal outcomes. Some prediction models are uesd to evaluate adverse pregnancy outcomes. However, some of the current prediction models are mainly carried out in developed countries, and many problems are still exist. We, thus, developed and validated a nomogram to predict the risk of adverse pregnancy outcomes of preeclampsia in Chinese pregnant women.

Methods

The clinical data of 720 pregnant women with preeclampsia in seven medical institutions in Chongqing from January 1, 2010, to December 31, 2020, were analyzed retrospectively. The patients were divided into two groups: 180 cases (25%) with adverse outcomes and 540 cases (75%) without adverse outcomes. The indicators were identified via univariate analysis. Logistic regression analysis was used to establish the prediction model, which was displayed by a nomogram. The performance of the nomogram was evaluated in terms of the area under the receiver operating characteristic (ROC) curve, calibration, and clinical utility.

Results

Univariate analysis showed that 24 indicators were significantly different (P < 0.05). Logistic regression analysis showed that gestational age, 24 h urine protein qualitative, and TT were significantly different (P < 0.05). The area under the ROC curve was 0.781 (95% CI 0.737–0.825) in training set and 0.777 (95% CI 0.689–0.865) in test set. The calibration curve of the nomogram showed good agreement between prediction and observation. The analysis of the clinical decision curve showed that the nomogram is of practical significance.

Conclusion

Our study identified gestational age, 24 h urine protein qualitative, and TT as risk factors for adverse outcomes of preeclampsia in pregnant women, and constructed a nomogram that can easily predict and evaluate the risk of adverse pregnancy outcomes in women with preeclampsia.

Association between systolic blood pressure course and outcomes after stroke thrombectomy

Background

Systolic blood pressure (SBP) after endovascular thrombectomy (EVT) for large artery occlusive stroke is dynamic, requiring adaptable early prediction tools for improving outcomes. We investigated if post-EVT SBP course was associated with outcomes.

Methods

EVT-treated patients who had a stroke at Karolinska University Hospital, Stockholm, Sweden, were included in the study during 12 February 2018–11 February 2020. SBP was recorded during the first 24 hours after EVT. Primary outcome was functional independence defined by a Modified Rankin Scale score of 0–2 at 3 months. Secondary outcomes were death by 3 months, symptomatic intracranial haemorrhage and any intracranial haemorrhage. Patients with favourable outcomes were used as a reference SBP course in mixed linear effects models and compared with SBP courses of patients with unfavourable outcomes using the empirical best linear unbiased predictor, measuring deviations from the reference SBP course using the random effects. We tested model predictive stability for SBP measurements of only 18, 12 or 6 hours after EVT.

Results

374 patients were registered, with mean age 71, median NIHSS score of 15, and 53.2% men. Deviating from a linear SBP course starting at 130 mm Hg and decreasing to 123 mm Hg at 24 hours after EVT was associated with lower chances of functional independence (adjusted OR 0.53, 95% CI 0.29 to 0.88, for reaching either 99 or 147 mm Hg at 24 hours after EVT). All SBP course models for the remaining outcomes did not show statistical significance. Functional independence models showed stable predictive values for all time periods.

Conclusion

Deviating from a linear SBP course was associated with lower chances of 3-month functional independence.