Neonatal intensive care admission for term neonates and subsequent childhood mortality: a retrospective linkage study.

BACKGROUND: Neonatal intensive care unit (NICU) admission among term neonates is a rare event. The aim of this study was to study the association of the NICU admission of term neonates on the risk of long-term childhood mortality. METHODS: A single-center case-control retrospective study between 2005 and 2019, including all in-hospital ≥ 37 weeks' gestation singleton live-born neonates. The center perinatal database was linked with the birth and death certificate registries of the Israeli Ministry of Internal Affairs. The primary aim of the study was to study the association between NICU admission and childhood mortality throughout a 15-year follow-up period. RESULTS: During the study period, 206,509 births were registered; 192,527 (93.22%) term neonates were included in the study; 5292 (2.75%) were admitted to NICU. Throughout the follow-up period, the mortality risk for term neonates admitted to the NICU remained elevated; hazard ratio (HR), 19.72 [14.66, 26.53], (p < 0.001). For all term neonates, the mortality rate was 0.16% (n = 311); 47.9% (n = 149) of those had records of a NICU admission. The mortality rate by time points (ratio1:10,0000 births) related to the age at death during the follow-up period was as follows: 29, up to 7 days; 20, 7-28 days; 37, 28 days to 6 months; 21, 6 months to 1 year; 19, 1-2 years; 9, 2-3 years; 10, 3-4 years; and 27, 4 years and more. Following the exclusion of congenital malformations and chromosomal abnormalities, NICU admission remained the most significant risk factor associated with mortality of the study population, HRs, 364.4 [145.3; 913.3] for mortality in the first 7 days of life; 19.6 [12.1; 32.0] for mortality from 28 days through 6 months of life and remained markedly elevated after age 4 years; HR, 7.1 [3.0; 17.0]. The mortality risk related to the NICU admission event, adjusted for admission diagnoses remained significant; HR = 8.21 [5.43; 12.4]. CONCLUSIONS: NICU admission for term neonates is a pondering event for the risk of long-term childhood mortality. This group of term neonates may benefit from focused health care.

Normal labor curve in twin gestation.

BACKGROUND: Failure to progress is one of the leading indications for cesarean delivery in trials of labor in twin gestations. However, assessment of labor progression in twin labors is managed according to singleton labor curves. OBJECTIVE: This study aimed to establish a partogram for twin deliveries that reflects normal and abnormal labor progression and customized labor curves for different subgroups of twin labors. STUDY DESIGN: This was a multicenter, retrospective cohort analysis of twin deliveries that were recorded in 3 tertiary medical centers between 2003 and 2017. Eligible parturients were those with twin gestations at ≥34 weeks' gestation with cephalic presentation of the presenting twin and ≥2 cervical examinations during labor. Exclusion criteria were elective cesarean delivery without a trial of labor, major fetal anomalies, and fetal demise. The study group comprised twin gestations, whereas singleton gestations comprised the control group. Statistical analysis was performed using Python 3.7.3 and SPSS, version 27. Categorical variables were analyzed using chi-square tests. Student t test and Mann-Whitney U test were applied to analyze the differences in continuous variables, as appropriate. RESULTS: A total of 1375 twin deliveries and 142,659 singleton deliveries met the inclusion criteria. Duration of the active phase of labor was significantly longer in twin labors than in singleton labors in both nulliparous and multiparous parturients; the 95th percentile duration was 2 hours longer in nulliparous twin labors and >3.5 hours longer in multiparous twin labors than in singleton labors. The cervical dilation progression rate was significantly slower in twin deliveries than in singleton deliveries with a mean rate in twin deliveries of 1.89 cm/h (95th percentile, 0.51 cm/h) and a mean rate of 2.48 cm/h (95th percentile, 0.73 cm/h) in singleton deliveries (P<.001). In addition, epidural use further slowed labor progression in twin deliveries. The second stage of labor was also markedly longer in twin deliveries, both in nulliparous and multiparous women (95th percentile, 3.04 vs 2.83 hours, P=.002). CONCLUSION: Twin labors are characterized by a slower progression of the active phase and second stage of labor compared with singleton labors in nulliparous and multiparous parturients. Epidural analgesia further slows labor progression in twin labors. Implementation of these findings in clinical management might lower cesarean delivery rates among cases with protracted labor in twin gestations.

Prediction model for obstetric anal sphincter injury using machine learning.

INTRODUCTION AND HYPOTHESIS: Obstetric anal sphincter injury (OASI) is a complication with substantial maternal morbidity. The aim of this study was to develop a machine learning model that would allow a personalized prediction algorithm for OASI, based on maternal and fetal variables collected at admission to labor. MATERIALS AND METHODS: We performed a retrospective cohort study at a tertiary university hospital. Included were term deliveries (live, singleton, vertex). A comparison was made between women diagnosed with OASI and those without such injury. For formation of a machine learning-based model, a gradient boosting machine learning algorithm was implemented. Evaluation of the performance model was achieved using the area under the receiver-operating characteristic curve (AUC). RESULTS: Our cohort comprised 98,463 deliveries, of which 323 (0.3%) were diagnosed with OASI. Applying a machine learning model to data recorded during admission to labor allowed for individualized risk assessment with an AUC of 0.756 (95% CI 0.732-0.780). According to this model, a lower number of previous births, fewer pregnancies, decreased maternal weight and advanced gestational week elevated the risk for OASI. With regard to parity, women with one previous delivery had approximately 1/3 of the risk for OASI compared to nulliparous women (OR = 0.3 (0.23-0.39), p < 0.001), and women with two previous deliveries had 1/3 of the risk compared to women with one previous delivery (OR = 0.35 (0.21-0.60), p < 0.001). CONCLUSION: Our machine learning-based model stratified births to high or low risk for OASI, making it an applicable tool for personalized decision-making upon admission to labor.

Transporting an Artificial Intelligence Model to Predict Emergency Cesarean Delivery: Overcoming Challenges Posed by Interfacility Variation.

Research using artificial intelligence (AI) in medicine is expected to significantly influence the practice of medicine and the delivery of health care in the near future. However, for successful deployment, the results must be transported across health care facilities. We present a cross-facilities application of an AI model that predicts the need for an emergency caesarean during birth. The transported model showed benefit; however, there can be challenges associated with interfacility variation in reporting practices.

Prediction of vaginal birth after cesarean deliveries using machine learning.

BACKGROUND: Efforts to reduce cesarean delivery rates to 12-15% have been undertaken worldwide. Special focus has been directed towards parturients who undergo a trial of labor after cesarean delivery to reduce the burden of repeated cesarean deliveries. Complication rates are lowest when a vaginal birth is achieved and highest when an unplanned cesarean delivery is performed, which emphasizes the need to assess, in advance, the likelihood of a successful vaginal birth after cesarean delivery. Vaginal birth after cesarean delivery calculators have been developed in different populations; however, some limitations to their implementation into clinical practice have been described. Machine-learning methods enable investigation of large-scale datasets with input combinations that traditional statistical analysis tools have difficulty processing. OBJECTIVE: The aim of this study was to evaluate the feasibility of using machine-learning methods to predict a successful vaginal birth after cesarean delivery. STUDY DESIGN: The electronic medical records of singleton, term labors during a 12-year period in a tertiary referral center were analyzed. With the use of gradient boosting, models that incorporated multiple maternal and fetal features were created to predict successful vaginal birth in parturients who undergo a trial of labor after cesarean delivery. One model was created to provide a personalized risk score for vaginal birth after cesarean delivery with the use of features that are available as early as the first antenatal visit; a second model was created that reassesses this score after features are added that are available only in proximity to delivery. RESULTS: A cohort of 9888 parturients with 1 previous cesarean delivery was identified, of which 75.6% of parturients (n=7473) attempted a trial of labor, with a success rate of 88%. A machine-learning-based model to predict when vaginal delivery would be successful was developed. When features that are available at the first antenatal visit are used, the model showed a receiver operating characteristic curve with area under the curve of 0.745 (95% confidence interval, 0.728-0.762) that increased to 0.793 (95% confidence interval, 0.778-0.808) when features that are available in proximity to the delivery process were added. Additionally, for the later model, a risk stratification tool was built to allocate parturients into low-, medium-, and high-risk groups for failed trial of labor after cesarean delivery. The low- and medium-risk groups (42.4% and 25.6% of parturients, respectively) showed a success rate of 97.3% and 90.9%, respectively. The high-risk group (32.1%) had a vaginal delivery success rate of 73.3%. Application of the model to a cohort of parturients who elected a repeat cesarean delivery (n=2145) demonstrated that 31% of these parturients would have been allocated to the low- and medium-risk groups had a trial of labor been attempted. CONCLUSION: Trial of labor after cesarean delivery is safe for most parturients. Success rates are high, even in a population with high rates of trial of labor after cesarean delivery. Application of a machine-learning algorithm to assign a personalized risk score for a successful vaginal birth after cesarean delivery may help in decision-making and contribute to a reduction in cesarean delivery rates. Parturient allocation to risk groups may help delivery process management.

Real-time data analysis using a machine learning model significantly improves prediction of successful vaginal deliveries.

BACKGROUND: The process of childbirth is one of the most crucial events in the future health and development of the offspring. The vulnerability of parturients and fetuses during the delivery process led to the development of intrapartum monitoring methods and to the emergence of alternative methods of delivery. However, current monitoring methods fail to accurately discriminate between cases in which intervention is unnecessary, partly contributing to the high rates of cesarean deliveries worldwide. Machine learning methods are applied in various medical fields to create personalized prediction models. These methods are used to analyze abundant, complex data with intricate associations to aid in decision making. Initial attempts to predict vaginal delivery vs cesarean deliveries using machine learning tools did not utilize the vast amount of data recorded during labor. The data recorded during labor represent the dynamic process of labor and therefore may be invaluable for dynamic prediction of vaginal delivery. OBJECTIVE: We aimed to create a personalized machine learning-based prediction model to predict successful vaginal deliveries using real-time data acquired during the first stage of labor. STUDY DESIGN: Electronic medical records of labor occurring during a 12-year period in a tertiary referral center were explored and labeled. Four different models were created using input from multiple maternal and fetal parameters. Initial risk assessments for vaginal delivery were calculated using data available at the time of admission to the delivery unit, followed by models incorporating cervical examination data and fetal heart rate data, and finally, a model that integrates additional data available during the first stage of labor was created. RESULTS: A total of 94,480 cases in which a trial of labor was attempted were identified. Based on approximately 180 million data points from the first stage of labor, machine learning models were developed to predict successful vaginal deliveries. A model using data available at the time of admission to the delivery unit yielded an area under the curve of 0.817 (95% confidence interval, 0.811-0.823). Models that used real-time data increased prediction accuracy. A model that includes real-time cervical examination data had an initial area under the curve of 0.819 (95% confidence interval, 0.813-0.825) at first examination, which increased to an area under the curve of 0.917 (95% confidence interval, 0.913-0.921) by the end of the first stage. Adding the real-time fetal heart monitor data provided an area under the curve of 0.824 (95% confidence interval, 0.818-0.830) at first examination, which increased to an area under the curve of 0.928 (95% confidence interval, 0.924-0.932) by the end of the first stage. Finally, adding additional real-time data increased the area under the curve initially to 0.833 (95% confidence interval, 0.827-0.838) at the first cervical examination and up to 0.932 (95% confidence interval, 0.928-0.935) by the end of the first stage. CONCLUSION: Real-time data acquired throughout the process of labor significantly increased the prediction accuracy for vaginal delivery using machine learning models. These models enable translation and quantification of the data gathered in the delivery unit into a clinical tool that yields a reliable personalized risk score and helps avoid unnecessary interventions.

Cervical dilation at the time of epidural catheter insertion is not associated with the degree of prolongation of the first or second stages of labor, or the rate of instrumental vaginal delivery.

INTRODUCTION: Epidural analgesia (EA) is an established option for efficient intrapartum analgesia. Meta-analyses have shown that EA differentially affects the first stage of labor but prolongs the second. The question of EA timing remains open. We aimed to investigate whether EA prolongs delivery in total and whether the EA administration timing vis-à-vis cervical dilation at catheter insertion is associated with a modulation of its effects on the duration of the first and second stages, as well as the rate of instrumental vaginal delivery in primiparas and multiparas. MATERIAL AND METHODS: A retrospective electronic medical records-based study of 18 870 singleton term deliveries occurring in our institution from 2003 to 2015. Cervical dilation was determined within a half-hour of EA administration. We examined whether cervical dilation at EA administration correlated with the duration of the first and/or second stage, with the rate of prolonged second stage, and with the rate of interventional delivery. The study group was stratified to 10 subgroups defined by 1-cm intervals of cervical dilation at EA administration. Logistic regression modeling was applied to analyze the association between EA timing and rate of instrumental delivery while controlling for possible confounders. RESULTS: In primiparas, receiving EA correlated with longer medians of active first stage (+51 minutes; P < .001) and second stage (+55 minutes; P < .001). In multiparas, median increases in active first stage (+43 minutes; P < .001) and second stage (+8 minutes; P < .001) were noted. The timing of EA, vis-à-vis cervical dilation (1-10 cm) was not associated with a substantial modulation of these effects. Logistic regression showed that cervical dilation at EA was not associated with a higher instrumental vaginal delivery rate. CONCLUSIONS: Epidural analgesia prolonged the first and second stages of labor vs no epidural. Having EA was associated with a higher instrumental delivery rate but not with higher rates of maternal or neonatal complications, in primi- and multiparas. Importantly, the timing of EA, vis-à-vis cervical dilation, was not associated with substantial changes in the duration of labor stages or the instrumental delivery rate. Thus, EA may be offered early in the first stage of labor.

Unplanned Cesarean Delivery in the Second Stage of Labor Holds Higher Odds of Complications than in the First Stage, while Similar in Primiparas and Multiparas.

BACKGROUND: While endeavors to reduce cesarean delivery (CD) rates are given priority worldwide, it is important to evaluate if these efforts place parturients and neonates at risk. CD performed in the second stage of labor carries higher risks of maternal and fetal complications and is a more challenging surgical procedure than that performed in the first stage or before labor. In a population with a low CD rate, we sought to evaluate the rate of maternal and fetal complications associated with unplanned CD (UCD) performed in the second vs. the first stage of labor, in primiparas and multiparas, as well as the risk factors leading to and the complications associated with UCD in the second stage of labor in this low-CD rate setting. METHODS: This was a retrospective, electronic medical record-based study of 7,635 term and preterm singletons born via UCD in the period 2003-2015. Maternal and neonatal background and outcome parameters were compared between groups. Logistic regression modeling was applied to adjust for clinically and statistically significant risk factors. RESULTS: UCD was more likely to be performed in the second stage of labor in mothers delivering larger fetuses (head circumference and body weight ≥90 centile) and those with persistent occiput posterior (POP) presentation. UCD in the second stage was strongly associated with serious maternal complications (excessive hemorrhage and fever) compared to UCD performed in the first stage, in both primiparas and multiparas. CONCLUSIONS: UCD performed in the second stage of labor, while less frequent than first-stage UCD, is more likely with larger neonates and POP presentation, and is associated with a higher rate of maternal complications in primiparas and multiparas. Complication rates in our low-CD-rate population did not exceed those reported in the literature from high-CD-rate areas.

A Prediction Model of Autism Spectrum Diagnosis from Well-Baby Electronic Data Using Machine Learning.

Early detection of autism spectrum disorder (ASD) is crucial for timely intervention, yet diagnosis typically occurs after age three. This study aimed to develop a machine learning model to predict ASD diagnosis using infants' electronic health records obtained through a national screening program and evaluate its accuracy. A retrospective cohort study analyzed health records of 780,610 children, including 1163 with ASD diagnoses. Data encompassed birth parameters, growth metrics, developmental milestones, and familial and post-natal variables from routine wellness visits within the first two years. Using a gradient boosting model with 3-fold cross-validation, 100 parameters predicted ASD diagnosis with an average area under the ROC curve of 0.86 (SD < 0.002). Feature importance was quantified using the Shapley Additive explanation tool. The model identified a high-risk group with a 4.3-fold higher ASD incidence (0.006) compared to the cohort (0.001). Key predictors included failing six milestones in language, social, and fine motor domains during the second year, male gender, parental developmental concerns, non-nursing, older maternal age, lower gestational age, and atypical growth percentiles. Machine learning algorithms capitalizing on preventative care electronic health records can facilitate ASD screening considering complex relations between familial and birth factors, post-natal growth, developmental parameters, and parent concern.

Machine Learning in Electronic Health Records: Identifying High-Risk Obstetric Patients Pre and During Labor.

Our goal is to apply artificial intelligence (AI) and statistical analysis to understand the relationship between various factors and outcomes during pregnancy and labor and delivery, in order to personalize birth management and reduce complications for both mothers and newborns. We use a structured electronic health records database with data from approximately 130,000 births to train, test and validate our models. We apply machine learning (ML) methods to predict various obstetrical outcomes before and during labor, with the aim of improving patient care management in the delivery ward. Using a large cohort of data (∼180 million data points), we then demonstrated that ML models can predict successful vaginal delivery, in the general population as well as a sub-cohort of women attempting trial of labor after a cesarean delivery. The real-time dynamic model showed increasing rates of accuracy as the delivery process progressed and more data became available for analysis. Additionally, we developed a cross-facilities application of an AI model that predicts the need for an unplanned cesarean delivery, illuminating the challenges associated with inter-facility variation in reporting practices. Overall, these studies combine novel technologies with currently available data to predict and assist safe deliveries for mothers and babies, both locally and globally.

Early developmental milestone clusters of autistic children based on electronic health records.

Autistic children vary in symptoms, co-morbidities, and response to interventions. This study aimed to identify clusters of autistic children with a distinct pattern of attaining early developmental milestones (EDMs). The clustering of 5836 autistic children was based on the attainment of 43 gross motor, fine motor, language, and social developmental milestones during the first 3 years of life as recorded in baby wellness visits. K-means cluster analysis detected four EDM clusters: mild (n = 1686); moderate (n = 1691); severe (n = 2265); and global (n = 194). The most prominent cluster differences were in the language domain. The global cluster showed earlier and greater developmental delay across domains, unique early gross motor delays, and more were born preterm via cesarean section. The severe cluster had poor language development prominently in the second year of life, and later fine motor delays. Moderate cluster had mainly language delays in the third year of life. The mild cluster mostly passed milestones. EDM clusters differed demographically, with higher socioeconomic status in mild cluster and lowest in global cluster. However, the severe cluster had more immigrant and non-Jewish mothers followed by the moderate cluster. The rates of parental concerns and provider developmental referrals were significantly higher in the global, followed by the severe, moderate, and mild EDM clusters. Autistic children's language and motor delay in the first 3 years can be grouped by common magnitude and onset profiles as distinct groups that may link to specific etiologies (like prematurity or genetics) and specific intervention programs. Early autism screening should be tailored to these different developmental profiles.

Predicting autism traits from baby wellness records: A machine learning approach.

LAY ABSTRACT: Timely identification of autism spectrum conditions is a necessity to enable children to receive the most benefit from early interventions. Emerging technological advancements provide avenues for detecting subtle, early indicators of autism from routinely collected health information. This study tested a model that provides a likelihood score for autism diagnosis from baby wellness visit records collected during the first 2 years of life. It included records of 591,989 non-autistic children and 12,846 children with autism. The model identified two-thirds of the autism spectrum condition group (boys 63% and girls 66%). Sex-specific models had several predictive features in common. These included language development, fine motor skills, and social milestones from visits at 12-24 months, mother's age, and lower initial growth but higher last growth measurements. Parental concerns about development or hearing impairment were other predictors. The models differed in other growth measurements and birth parameters. These models can support the detection of early signs of autism in girls and boys by using information routinely recorded during the first 2 years of life.

Maternal hybrid immunity and risk of infant COVID-19 hospitalizations: national case-control study in Israel.

Hybrid immunity, acquired through vaccination followed or preceded by a COVID-19 infection, elicits robust antibody augmentation. We hypothesize that maternal hybrid immunity will provide greater infant protection than other forms of COVID-19 immunity in the first 6 months of life. We conducted a case-control study in Israel, enrolling 661 infants up to 6 months of age, hospitalized with COVID-19 (cases) and 59,460 age-matched non-hospitalized infants (controls) between August 24, 2021, and March 15, 2022. Infants were grouped by maternal immunity status at delivery: Naïve (never vaccinated or tested positive, reference group), Hybrid-immunity (vaccinated and tested positive), Natural-immunity (tested positive before or during the study period), Full-vaccination (two-shot regimen plus 1 booster), and Partial-vaccination (less than full three shot regimen). Applying Cox proportional hazards models to estimate the hazard ratios, which was then converted to percent vaccine effectiveness, and using the Naïve group as the reference, maternal hybrid-immunity provided the highest protection (84% [95% CI 75-90]), followed by full-vaccination (66% [95% CI 56-74]), natural-immunity (56% [95% CI 39-68]), and partial-vaccination (29% [95% CI 15-41]). Maternal hybrid-immunity was associated with a reduced risk of infant hospitalization for Covid-19, as compared to natural-immunity, regardless of exposure timing or sequence. These findings emphasize the benefits of vaccinating previously infected individuals during pregnancy to reduce COVID-19 hospitalizations in early infancy.

Partogram of Grandmultiparous Parturients: A Multicenter Cohort Study.

Sparse and conflicting data exist regarding the normal partogram of grand-multiparous (GMP, defined as parity of 6+) parturients. Customized partograms may potentially lower cesarean delivery rates for protraction disorders in this population. In this study, we aim to construct a normal labor curve of GMP women and compare it to the multiparous (MP, defined as parity of 2-5) partogram. We conducted a multicenter retrospective cohort analysis of deliveries between the years 2003 and 2019. Eligible parturients were the trials of labor of singletons ≥37 + 0 weeks in cephalic presentation with ≥2 documented cervical examinations during labor. Exclusion criteria were elective cesarean delivery without a trial of labor, preterm labor, major fetal anomalies, and fetal demise. GMP comprised the study group while the MP counterparts were the control group. A total of 78,292 deliveries met the inclusion criteria, comprising 10,532 GMP and 67,760 MP parturients. Our data revealed that during the first stage of labor, cervical dilation progressed at similar rates in MPs and GMPs, while head descent was a few minutes faster in GMPs compared to MPs, regardless of epidural anesthesia. The second stage of labor was faster in GMPs compared to MPs; the 95th percentile of the second stage duration of GMPs (48 min duration) was 43 min less than that of MPs (91 min duration). These findings remained similar among deliveries with and without epidural analgesia or labor induction. We conclude that GMPs' and MPs' cervical dilation progression in the active phase of labor was similar, and the second stage of labor was shorter in GMPs, regardless of epidural use. Thus, GMPs' uterus function during labor corresponds, and possibly surpasses, that of MPs. These findings indicate that health providers can use the standard partogram of the active phase of labor when caring for GMP parturients.

Maternal third dose of BNT162b2 mRNA vaccine and risk of infant COVID-19 hospitalization.

Infants are at a higher risk of Coronavirus Disease 2019 (COVID-19)-related hospitalizations compared to older children. In this study, we investigated the effect of the recommended third maternal dose of BNT162b2 COVID-19 vaccine during pregnancy on rates of infant COVID-19-related hospitalizations. We conducted a nationwide cohort study of all live-born infants delivered in Israel between 24 August 2021 and 15 March 2022 to estimate the effectiveness of the third booster dose versus the second dose against infant COVID-19-related hospitalizations. Data were analyzed for the overall study period, and the Delta and Omicron periods were analyzed separately. Cox proportional hazard regression models estimated hazard ratios and 95% confidence intervals (CIs) for infant hospitalizations according to maternal vaccination status at delivery. Among 48,868 live-born infants included in the analysis, rates of COVID-19 hospitalization were 0.4%, 0.6% and 0.7% in the third-dose, second-dose and unvaccinated groups, respectively. Compared to the second dose, the third dose was associated with reduced infant hospitalization with estimated effectiveness of 53% (95% CI: 36-65%). Greater protection was associated with a shorter interval between vaccination and delivery. A third maternal dose during pregnancy reduced the risk of infant hospitalization for COVID-19 during the first 4 months of life, supporting clinical and public health guidance for maternal booster vaccination to prevent infant COVID-19 hospitalization.

Chorioamnionitis and its association with neonatal and maternal adverse outcomes in women with and without epidural analgesia administration.

OBJECTIVE: To evaluate neonatal fever and adverse maternal and neonatal outcomes in febrile laboring women and assess whether the time interval between epidural analgesia (EA) administration and chorioamnionitis is associated with these complications. METHODS: A retrospective cohort study at a university affiliated medical center between 2003 and 2015. Included were women who underwent term vaginal delivery attempt and diagnosed with chorioamnionitis. The primary outcomes compared between febrile women with and without EA were neonatal fever and adverse neonatal and maternal outcomes. The association between time from EA to fever (<6, 6-12, >12 h) and maternal and neonatal complications was also assessed. RESULTS: During the study period, 1,933 women with chorioamnionitis were assessed. Of them, 1,810 (93.6%) received EA prior to fever and 123 (6.4%) febrile parturients did not receive EA. Neonatal fever and other neonatal adverse outcomes were similar in the EA vs. non-EA group (2.2% vs. 0.8% and 2.7% vs. 4.9% (NS)), except for transient tachypnea of the newborn rates which were lower in the EA group (1.4% vs. 4.1%, p = 0.043). Maternal complications were similar, besides for higher rates of instrumental deliveries found in the EA group (24.0% vs. 5.7%, p < 0.001). Time between EA and fever onset was not associated with neonatal complications in logistic regression analysis. CONCLUSION: Neonatal and maternal outcomes are similar in febrile laboring women with and without EA. The time interval between EA and onset of fever is not associated with increased rates of neonatal fever or adverse outcomes and should not affect the management of labor.

Primary risk stratification for neonatal jaundice among term neonates using machine learning algorithm.

BACKGROUND: Neonatal jaundice occurs in approximately 60% of term newborns. Although risk factors for neonatal jaundice have been studied, all the suggested strategies are based on various newborn tests for bilirubin levels. We aim to stratify neonates into risk groups for clinically significant neonatal jaundice using a combined data analysis approach, without serum bilirubin evaluation. STUDY DESIGN: Term (gestational week 37-42) neonates born in a single medical center, 2005-2018 were identified. Anonymized data were analyzed using machine learning. Thresholds for stratification into risk groups were established. Associations were evaluated statistically using neonates with and without clinically significant neonatal jaundice from the study population. RESULTS: A total of 147,667 consecutive term live neonates were included. The machine learning diagnostic ability to evaluate the risk for neonatal jaundice was 0.748; 95% CI 0.743-0.754 (AUC). The most important factors were (in order of importance) maternal blood type, maternal age, gestational age at delivery, estimated birth weight, parity, CBC at admission, and maternal blood pressure at admission. Neonates were then stratified by risk: 61% (n = 90,140) were classed as low-risk, 39% (n = 57,527) as higher-risk. Prevalence of jaundice was 4.14% in the full cohort, and 1.47% and 8.29% in the low- and high-risk cohorts, respectively; OR 6.06 (CI: 5.7-6.45) for neonatal jaundice in high-risk group. CONCLUSION: A population tailored "first step" screening policy using machine learning model presents potential of neonatal jaundice risk stratification for term neonates. Future development and validation of this computational model are warranted.

Risk factors for uterine incision extension during cesarean delivery.

BACKGROUND: Unintentional extension of uterine incision is a known complication during cesarean delivery estimated to occur in 4-8% of cesarean deliveries. The aim of this study was to examine risk factors associated with unintentional uterine incision extension and to assess which of them are independent risk factors for this condition. STUDY DESIGN: We conducted a retrospective cohort study at a large public university tertiary referral center between 2003 and 2017. Included were women who underwent cesarean delivery during this time period. Demographic, medical, obstetrical and surgical data were collected. The primary outcome was the presence of uterine incision extension during cesarean delivery. Secondary outcomes included detection of risk factors associated with uterine incision extension. A multivariate analysis was additionally performed to identify general and labor related risk factors for unintentional extension of uterine incision among patients that underwent cesarean delivery during second stage of labor. RESULTS: During the study period, 25,879 cesarean deliveries performed in our medical center were assessed. Out of them, 731 (2.8%) cases of unintended uterine incision extension were identified. Women in this group had high rates of full cervical dilatation and increased maternal hemorrhage. Assessment of incision extension direction revealed that two-thirds of extensions were lateral, mostly unilateral. Median size of the extension was 2.7 ± 1.2 cm.Independent parameters associated with unintended uterine incision extension included nulliparity, vertex presentation, epidural anesthesia and cesarean section indication. Further analysis including cesarean deliveries performed during the second stage of labor revealed 397 (15.3%) cesarean deliveries in which incision extension was noted and 2205 (84.7%) cesarean deliveries without incision extension. Following multivariate analysis performed in women who underwent cesarean delivery during second stage of labor, two independent parameters associated with unintended uterine incision extension remained significant - past cesarean delivery and failed vacuum attempt. CONCLUSIONS: Vacuum extraction attempt and previous cesarean delivery are independent risk factors for uterine incision extension in women undergoing cesarean delivery during the second stage of labor. We also showed the majority of these extensions to be lateral.

Prediction of neonatal subgaleal hemorrhage using first stage of labor data: A machine-learning based model.

BACKGROUND: Subgaleal hemorrhage (SGH) is a rare neonatal condition, mainly associated with instrumental delivery, mainly vacuum extractor (VE). The aim of this study was to develop a machine learning model that would allow a personalized prediction algorithm for Subgaleal hemorrhage (SGH) following vacuum extraction (VE), based on maternal and fetal variables collected during the first stage of labor. MATERIALS AND METHODS: A retrospective cohort study on data from a university affiliated hospital, recorded between January 2013 and February 2017. Balanced random forest algorithm was used to develop a machine learning model to predict personalized risk of the neonate developing SGH, in the eventuality that vacuum extraction was used during delivery. RESULTS: During the study period, 35,552 term, singleton spontaneous or induced trials of labor deliveries were included in this study. Neonatal SGH following vacuum extraction (SGH-VE) occurred in 109 cases (0.3%). Two machine learning models were developed: a proof of concept model (model A), based on a cohort limited to the (n=2955) instances of vacuum extraction, and the clinical support model (model B), based on all spontaneous or induced trials of labor (n=35,552). The models stratified parturients into high- and low-risk groups for development of SGH-VE. Model A showed a 2-fold increase in the high-risk group of parturients compared to the low risk group (OR=2.76, CI 95% 1.85-4.11). In model B, a 4-fold increase in the odds of SGH was observed in the high-risk group of parturients compared to the low risk group (OR=4.2, CI 2.2-8.1), while identifying 90.8% (99/109) of the SGH cases. CONCLUSIONS: Our machine learning-based model stratified births to high or low risk for SGH, making it an applicable tool for personalized decision-making during labor regarding the application of VE. This model may contribute to improved neonatal outcomes.

NICU Admission for Term Neonates in a Large Single-Center Population: A Comprehensive Assessment of Risk Factors Using a Tandem Analysis Approach.

Objective: Neonatal intensive care unit (NICU) admission among term neonates is associated with significant morbidity and mortality, as well as high healthcare costs. A comprehensive NICU admission risk assessment using an integrated statistical approach for this rare admission event may be used to build a risk calculation algorithm for this group of neonates prior to delivery. Methods: A single-center case−control retrospective study was conducted between August 2005 and December 2019, including in-hospital singleton live born neonates, born at ≥37 weeks' gestation. Analyses included univariate and multivariable models combined with the machine learning gradient-boosting model (GBM). The primary aim of the study was to identify and quantify risk factors and causes of NICU admission of term neonates. Results: During the study period, 206,509 births were registered at the Shaare Zedek Medical Center. After applying the study exclusion criteria, 192,527 term neonates were included in the study; 5292 (2.75%) were admitted to the NICU. The NICU admission risk was significantly higher (ORs [95%CIs]) for offspring of nulliparous women (1.19 [1.07, 1.33]), those with diabetes mellitus or hypertensive complications of pregnancy (2.52 [2.09, 3.03] and 1.28 [1.02, 1.60] respectively), and for those born during the 37th week of gestation (2.99 [2.63, 3.41]; p < 0.001 for all), adjusted for congenital malformations and genetic syndromes. A GBM to predict NICU admission applied to data prior to delivery showed an area under the receiver operating characteristic curve of 0.750 (95%CI 0.743−0.757) and classified 27% as high risk and 73% as low risk. This risk stratification was significantly associated with adverse maternal and neonatal outcomes. Conclusion: The present study identified NICU admission risk factors for term neonates; along with the machine learning ranking of the risk factors, the highly predictive model may serve as a basis for individual risk calculation algorithm prior to delivery. We suggest that in the future, this type of planning of the delivery will serve different health systems, in both high- and low-resource environments, along with the NICU admission or transfer policy.