Cross-validated prediction model for severe adverse neonatal outcomes in a term, non-anomalous, singleton cohort

Assessment of the cerebroplacental ratio and uterine arteries in low-risk pregnancies in early labour for the prediction of obstetric and neonatal outcomes

BACKGROUND

The evidence-based management of human labor includes the antepartum identification of patients at risk for intrapartum hypoxia. However, available evidence has shown that most of the hypoxic-related complications occur among pregnancies classified at low-risk for intrapartum hypoxia, thus suggesting that the current strategy to identify the pregnancies at risk for intrapartum fetal hypoxia has limited accuracy.

OBJECTIVE

To evaluate the role of the combined assessment of the cerebroplacental ratio (CPR) and uterine arteries (UtA) Doppler in the prediction of obstetric intervention (OI) for suspected intrapartum fetal compromise (IFC) within a cohort of low-risk singleton term pregnancies in early labor.

METHODS

Prospective multicentre observational study conducted across four tertiary Maternity Units between January 2016 and September 2019. Low-risk term pregnancies with spontaneous onset of labor were included. A two-step multivariable model was developed to assess the risk of OI for suspected IFC. The baseline model included antenatal and intrapartum characteristics, while the combined model included antenatal and intrapartum characteristics plus Doppler anomalies such as CPR MoM < 10th percentile and mean UtA Doppler PI MoM ≥ 95th percentile. Predictive performance was determined by receiver-operating characteristics curve analysis.

RESULTS

804 women were included. At logistic regression analysis, CPR MoM < 10th percentile (aOR 1.269, 95 % CI 1.188-1.356, P < 0.001), mean UtA PI MoM ≥ 95th percentile (aOR 1.012, 95 % CI 1.001-1.022, P = 0.04) were independently associated with OI for suspected IFC. At ROC curve analysis, the combined model including antenatal characteristics plus abnormal CPR and mean UtA PI yielded an AUC of 0.78, 95 %CI(0.71-0.85), p < 0.001, which was significantly higher than the baseline model (AUC 0.61, 95 %CI(0.54-0.69), p = 0.007) (p < 0.001). The combined model was associated with a 0.78 (95 % CI 0.67-0.89) sensitivity, 0.68 (95 % CI 0.65-0.72) specificity, 0.15 (95 % CI 0.11-0.19) PPV, and 0.98 (0.96-0.99) NPV, 2.48 (95 % CI 2.07-2.97) LR + and 0.32 (95 % CI 0.19-0.53) LR- for OI due to suspected IFC.

CONCLUSIONS

A predictive model including antenatal and intrapartum characteristics combined with abnormal CPR and mean UtA PI has a good capacity to rule out and a moderate capacity to rule in OI due to IFC, albeit with poor predictive value.

Assessing and Validating a Model of Study Completion for a Prospective Cohort of Healthy Newborns

OBJECTIVES

To identify potentially modifiable or actionable factors related to study completion among healthy mother-infant dyads participating in prospective research.

PATIENTS/METHODS

We conducted a secondary analysis of completion data from a prospective study on newborn jaundice in the first week of life at a tertiary-care hospital in Philadelphia, PA, from 2015 to 2019. Participation in the original study involved enrollment before newborn discharge and subsequent follow-up for a jaundice assessment between 2 and 6 days of life. For this study, our primary outcome was completion of all study procedures. Associations between predictor variables and the outcome were assessed using bivariate and multivariable analyses. We fit a predictive model of study completion using logistic regression and validated the model using 5-fold cross-validation.

RESULTS

Of 501 mother-infant dyads enrolled in the original study, 304 completed the study. Median maternal age was 28 years and 81.8% of mothers delivered via vaginal birth. Study completion was associated with colocation of the study visit with the initial well-child visit (adjusted odds ratio [aOR], 2.99, 95% confidence interval [CI], 2.01-4.46) and provision of an alternate phone number by the participant (aOR, 1.99; 95% CI, 1.34-2.96). The cross-validated model performed similarly to our final predictive model and had an average area under the receiver operating characteristic curve of 0.67 (range, 0.59-0.72), with a sensitivity of 68% and specificity of 60%.

CONCLUSIONS

Findings demonstrate the importance of communication and patient-centric approaches for recruitment and retention in newborn research. Future work should incorporate these approaches while continuing to evaluate study retention strategies.

Probabilistic ratiocination of hepatocellular carcinoma after resection: evaluation of expected to be promising approaches

Background

Precise prediction of survival after treatment is of great importance for patients with diseases with high mortality. RNA sequencing data and deep learning (DL) methods are expected to become promising approaches in the development of prediction models in the future. We aimed to evaluate the optimal covariates and methodology for patients with hepatocellular carcinoma (HCC) undergoing surgical resection.

Methods

The Cox proportional hazards regression model and the DL approach were used to develop prediction models incorporating clinical, genetic, and combined clinical and genetic variables for survival prediction in patients with HCC after resection. A total of 1,114 patients and 184 patients were enrolled in the present study from 2,163 and 601 patients from Eastern Hepatobiliary Surgery Hospital and Renji Hospital, respectively. The models were internally validated through random sampling and externally validated in clinical cohorts. Between-model comparisons were carried out in terms of the integrated discrimination improvement and net reclassification index.

Results

The Cox and DL clinical models were developed by adopting 7 independent prognostic factors (total bilirubin, prothrombin time, tumor size, tumor number, lymph node metastasis, and vascular invasion) and 22 clinical factors, respectively. Both the Cox clinical model and the DL clinical model showed excellent performances in the derivation [area under the curve (AUC): 0.75 vs. 0.77] and validation (AUC: 0.83 vs. 0.80) sets. The derived Cox genetic model with 6 significant prognostic genes was not as effective as the DL approach involving 686 genes. A combined clinical and genetic approach modified the performances of both the Cox and DL models. The integrated discrimination improvement and net reclassification index of the DL clinical model were generally better than those of the Cox clinical model.

Conclusions

Our Cox clinical model sufficiently provided precise survival prediction in patients with HCC after resection. It may serve as an accurate and cost-effective tool for predicting survival in such patients.

Factors Associated with Increased Risk of Early Severe Neonatal Morbidity in Late Preterm and Early Term Infants

Although the risk of neonatal mortality is generally low for late preterm and early term infants, they are still significantly predisposed to severe neonatal morbidity (SNM) despite being born at relatively advanced gestations. In this study, we investigated maternal and intrapartum risk factors for early SNM in late preterm and early term infants. This was a retrospective cohort study of non-anomalous, singleton infants (34⁺⁰-38⁺⁶ gestational weeks) born at the Mater Mother's Hospital in Brisbane, Australia from January 2015 to May 2020. Early SNM was defined as a composite of any of the following severe neonatal outcome indicators: admission to neonatal intensive care unit (NICU) in conjunction with an Apgar score <4 at 5 min, severe respiratory distress, severe neonatal acidosis (cord pH < 7.0 or base excess <-12 mmol/L). Multivariable binomial logistic regression analyses using generalized estimating equations (GEE) were used to identify risk factors. Of the total infants born at 34⁺⁰-38⁺⁶ gestational weeks, 5.7% had at least one component of the composite outcome. For late preterm infants, pre-existing diabetes mellitus, instrumental birth and emergency caesarean birth for non-reassuring fetal status were associated with increased odds for early SNM, whilst for early term infants, pre-existing and gestational diabetes mellitus, antepartum hemorrhage, instrumental, emergency caesarean and elective caesarean birth were significant risk factors. In conclusion, we identified several risk factors contributing to early SNM in late preterm and early term cohort. Our results suggest that predicted probability of early SNM decreased as gestation increased.

Prediction of severe adverse neonatal outcomes at the second stage of labour using machine learning: a retrospective cohort study

OBJECTIVE

To create a personalised machine learning model for prediction of severe adverse neonatal outcomes (SANO) during the second stage of labour.

DESIGN

Retrospective Electronic-Medical-Record (EMR) -based study.

POPULATION

A cohort of 73 868 singleton, term deliveries that reached the second stage of labour, including 1346 (1.8%) deliveries with SANO.

METHODS

A gradient boosting model was created, analysing 21 million data points from antepartum features (e.g. gravidity and parity) gathered at admission to the delivery unit, and intrapartum data (e.g. cervical dilatation and effacement) gathered during the first stage of labour. Deliveries were allocated to high-risk and low-risk groups based on the Youden index to maximise sensitivity and specificity.

MAIN OUTCOME MEASURES

SANO was defined as either umbilical cord pH levels ≤7.1 or 1-minute or 5-minute Apgar score ≤7.

RESULTS

The model for prediction of SANO yielded an area under the receiver operating curve (AUC) of 0.761 (95% CI 0.748-0.774). A third of the cohort (33.5%, n = 24 721) were allocated to a high-risk group for SANO, which captured up to 72.1% of these cases (odds ratio 5.3, 95% CI 4.7-6.0; high-risk versus low-risk groups).

CONCLUSIONS

Data acquired throughout the first stage of labour can be used to predict SANO during the second stage of labour using a machine learning model. Stratifying parturients at the beginning of the second stage of labour in a 'time out' session, can direct a personalised approach to management of this challenging aspect of labour, as well as improve allocation of staff and resources.

TWEETABLE ABSTRACT

Personalised prediction score for severe adverse neonatal outcomes in labour using machine learning model.

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.