Long-term adverse neurodevelopmental outcomes of discordant twins delivered at term: A nationwide population-based study.

OBJECTIVE: To evaluate long-term adverse neurodevelopmental outcomes of discordant twins delivered at term. DESIGN: Retrospective cohort study. SETTING: Nationwide (Republic of Korea). POPULATION: All twin children delivered at term between 2007 and 2010. METHODS: The study population was divided into two groups according to inter-twin birthweight discordancy: the 'concordant twin group', twin pairs with inter-twin birthweight discordancy less than 20%; and the 'discordant twin group', twin pairs with inter-twin birthweight discordancy of 20% or more. The risk of long-term adverse neurodevelopmental outcomes was compared between the concordant twin group and the discordant twin group. Long-term adverse neurodevelopmental outcomes between smaller and larger twin children within twin pairs were further analysed. The composite adverse neurodevelopmental outcome was defined as the presence of at least one of the following: motor developmental delay, cognitive developmental delay, autism spectrum disorders/attention deficit hyperactivity disorders, tics/stereotypical behaviour or epileptic/febrile seizure. MAIN OUTCOME MEASURES: Long-term adverse neurodevelopmental outcome. RESULTS: Of 22 468 twin children (11 234 pairs) included, 3412 (15.19%) twin children were discordant. The risk of composite adverse neurodevelopmental outcome was higher in the discordant twin group than in the concordant twin group (adjusted hazard ratio [HR] 1.13, 95% CI 1.03-1.24). The long-term adverse neurodevelopmental outcomes were not significantly different between smaller and larger twin children in discordant twin pairs (adjusted HR 1.01, 95% CI 0.81-1.28). CONCLUSION: In twin pairs delivered at term, an inter-twin birthweight discordancy of 20% or greater was associated with long-term adverse neurodevelopmental outcomes; and long-term adverse neurodevelopmental outcomes were not significantly different in smaller or larger twin children in discordant twin pairs.

Ultrasonographic monitoring of fetal eye growth parameters throughout gestation in the common marmoset (Callithrix jacchus).

The common marmoset (Callithrix jacchus) is considered an ideal species for developing genetically modified nonhuman primates (NHP) models of human disease, particularly eye disease. They have been proposed as a suitable bridge between rodents and other NHP models due to their similar ophthalmological features to humans. Prenatal ultrasonography is an accurate and reliable diagnostic tool for monitoring fetal development and congenital malformation. We monitored fetal eye growth and development using noninvasive ultrasonography in 40 heads of clinically normal fetuses during pregnancy to establish the criteria for studying congenital eye anomalies in marmosets. The coronal, sagittal, and transverse planes were useful to identify the facial structures for any associated abnormalities. For orbital measurements, biorbital distance (BOD), ocular diameter (OD), interorbital distance (IOD), and total axial length (TAL) were measured in the transverse plane and carefully identified for intraorbital structures. As a result, high correlations were observed between delivery-based gestational age (GA) and biparietal diameter (BPD), BOD, OD, and TAL. The correlation assessments based on BOD provide more reliable results for monitoring eye growth and development in normal marmosets than any other parameters since BOD has the highest correlation coefficient according to both delivery-based GA and BPD among ocular measurements. In conclusion, orbital measurements by prenatal ultrasonography provide reliable indicators of marmoset eye growth, and it could offer early diagnostic criteria to facilitate the development of eye disease models and novel therapies such as genome editing technologies in marmosets.

Real-time Classification of Fetal Status Based on Deep Learning and Cardiotocography Data.

This study uses convolutional neural networks (CNNs) and cardiotocography data for the real-time classification of fetal status in the mobile application of a pregnant woman and the computer server of a data expert at the same time (The sensor is connected with the smartphone, which is linked with the web server for the woman and the computer server for the expert). Data came from 5249 (or 4833) cardiotocography traces in Anam Hospital for the mobile application (or the computer server). 150 data cases of 5-minute duration were extracted from each trace with 141,001 final cases for the mobile application and for the computer server alike. The dependent variable was fetal status with two categories (Normal, Abnormal) for the mobile application and three categories (Normal, Middle, Abnormal) for the computer server. The fetal heart rate served as a predictor for the mobile application and the computer server, while uterus contraction for the computer server only. The 1-dimension (or 2-dimension) Resnet CNN was trained for the mobile application (or the computer server) during 800 epochs. The sensitivity, specificity and their harmonic mean of the 1-dimension CNN for the mobile application were 94.9%, 91.2% and 93.0%, respectively. The corresponding statistics of the 2-dimension CNN for the computer server were 98.0%, 99.5% and 98.7%. The average inference time per 1000 images was 6.51 micro-seconds. Deep learning provides an efficient model for the real-time classification of fetal status in the mobile application and the computer server at the same time.

Metabolic Dysfunction-Associated Fatty Liver Disease and Subsequent Development of Adverse Pregnancy Outcomes.

BACKGROUND & AIMS: Recently, metabolic dysfunction-associated fatty liver disease (MAFLD), rather than nonalcoholic fatty liver disease (NAFLD), was proposed to better describe liver disease associated with metabolic dysfunction (MD). In this study, we attempted to investigate the impact of MAFLD on pregnancy complications. METHODS: The current study is a secondary analysis of a multicenter prospective cohort designed to examine the risk of NAFLD during pregnancy. In the first trimester, enrolled pregnant women were evaluated for hepatic steatosis by liver ultrasonography, and blood samples were collected for biochemical measurements. The study population was divided into 3 groups: no NAFLD, hepatic steatosis but without metabolic dysfunction (non-MD NAFLD), and MAFLD. The primary outcome was the subsequent development of adverse pregnancy outcomes, including gestational diabetes mellitus, pregnancy-associated hypertension, preterm birth, and fetal growth abnormalities. RESULTS: The study population consisted of 1744 pregnant women, including 1523 with no NAFLD, 43 with non-MD NAFLD, and 178 with MAFLD. The risk of subsequent development of adverse pregnancy outcomes was higher in MAFLD than in non-MD NAFLD (adjusted odds ratio, 4.03; 95% CI, 1.68-9.67), whereas the risk was not significantly different between no NAFLD and non-MD NAFLD. Among women with no NAFLD, the presence of MD increased the risk of adverse pregnancy outcomes. However, women with MAFLD were at higher risk for adverse pregnancy outcomes than women with no NAFLD without MD or those with no NAFLD with MD. CONCLUSIONS: In pregnant women, MAFLD may be associated with an increased risk of subsequent adverse pregnancy outcomes.

Association between breastfeeding duration and diabetes mellitus in menopausal women: a machine-learning analysis using population-based retrospective study.

BACKGROUND: Breastfeeding resets insulin resistance caused by pregnancy however, studies on the association between breastfeeding and diabetes mellitus (DM) have reported inconsistent results. Therefore, we aimed to investigate the risk of DM according to breastfeeding duration in large-scale population-based retrospective study. In addition, machine-learning prediction models for DM and hemoglobin A1c (HbA1c) were developed to further evaluate this association. METHODS: We used the Korean National Health and Nutrition Examination Surveys database, a nationwide and population-based health survey from 2010 to 2020. We included 15,946 postmenopausal women with a history of delivery, whom we divided into three groups according to the average breastfeeding duration: (1) no breastfeeding, (2) < 12 months breastfeeding, and (3) ≥ 12 months breastfeeding. Prediction models for DM and HbA1c were developed using an artificial neural network, decision tree, logistic regression, Naïve Bayes, random forest, and support vector machine. RESULTS: In total, 2248 (14.1%) women had DM and 14,402 (90.3%) had a history of breastfeeding. The prevalence of DM was the lowest in the < 12 breastfeeding group (no breastfeeding vs. < 12 months breastfeeding vs. ≥ 12 months breastfeeding; 161 [10.4%] vs. 362 [9.0%] vs. 1,725 [16.7%], p < 0.001). HbA1c levels were also the lowest in the < 12 breastfeeding group (HbA1c: no breastfeeding vs. < 12 months breastfeeding vs. ≥ 12 months breastfeeding; 5.9% vs. 5.9% vs. 6.1%, respectively, p < 0.001). After adjustment for covariates, the risk of DM was significantly increased in both, the no breastfeeding (adjusted odds ratio [aOR] 1.29; 95% CI 1.29, 1.62]) and ≥ 12 months of breastfeeding groups (aOR 1.18; 95% CI 1.01, 1.37) compared to that in the < 12 months breastfeeding group. The accuracy and the area under the receiver-operating-characteristic curve of the DM prediction model were 0.93 and 0.95, respectively. The average breastfeeding duration was ranked among the top 15 determinants of DM, which supported the strong association between breastfeeding duration and DM. This association was also observed in a prediction model for HbA1c. CONCLUSIONS: Women who did not breasted had a higher risk of developing DM than those who breastfed for up to 12 months.

Exploring the Utility of remimazolam in cesarean sections under general anesthesia: A preliminary retrospective analysis and Implications for future study.

Background: Remimazolam has recently been introduced as a maintenance agent for general anesthesia. However, the effect of remimazolam on peripartum prognosis has not been reported. Therefore, this study aimed to compare the effects of remimazolam and propofol for uterotonic drugs following cesarean section. Methods: The electronic medical records of 51 adult women who underwent elective cesarean sections by single obstetrician under general anesthesia were collected. Participants were categorized into two groups: the propofol group and the remimazolam group. General anesthesia was maintained by continuous infusion of propofol or remimazolam after delivery. The number of uterotonic drugs administered during the cesarean section, the estimated blood loss (EBL), and length of hospital stay (LOS) after delivery were assessed. Results: Of the 51 patients included in the study, 35 were in the propofol group and 16 in the remimazolam group. In the remimazolam group, five patients (31.3%, 5/16) received more uterotonics than the standard regimen. Conversely, in the propofol group, 19 patients (54.3%, 19/35) were injected with more uterotonics than the standard regimen. Logistic regression analysis showed that abnormal positioning of the placenta (P = 0.079) and not using remimazolam (P = 0.100) were the most relevant factors associated with the increased use of uterotonics. There was no significant difference in EBL between the two groups. The use of remimazolam was clinically relevant with a shorter LOS (P = 0.059). Conclusions: The use of remimazolam as a maintenance agent did not result in significantly higher use of intrapartum uterotonics compared to the use of propofol. These results cannot exclude all adverse effects of remimazolam during cesarean delivery. Further randomized controlled trials must be conducted to obtain high-quality evidence.

Machine learning analysis for the association between breast feeding and metabolic syndrome in women.

This cross-sectional study aimed to develop and validate population-based machine learning models for examining the association between breastfeeding and metabolic syndrome in women. The artificial neural network, the decision tree, logistic regression, the Naïve Bayes, the random forest and the support vector machine were developed and validated to predict metabolic syndrome in women. Data came from 30,204 women, who aged 20 years or more and participated in the Korean National Health and Nutrition Examination Surveys 2010-2019. The dependent variable was metabolic syndrome. The 86 independent variables included demographic/socioeconomic determinants, cardiovascular disease, breastfeeding duration and other medical/obstetric information. The random forest had the best performance in terms of the area under the receiver-operating-characteristic curve, e.g., 90.7%. According to random forest variable importance, the top predictors of metabolic syndrome included body mass index (0.1032), medication for hypertension (0.0552), hypertension (0.0499), cardiovascular disease (0.0453), age (0.0437) and breastfeeding duration (0.0191). Breastfeeding duration is a major predictor of metabolic syndrome for women together with body mass index, diagnosis and medication for hypertension, cardiovascular disease and age.

Machine learning analysis with population data for prepregnancy and perinatal risk factors for the neurodevelopmental delay of offspring.

Neurodevelopmental disorders (NDD) in offspring are associated with a complex combination of pre-and postnatal factors. This study uses machine learning and population data to evaluate the association between prepregnancy or perinatal risk factors and the NDD of offspring. Population-based retrospective cohort data were obtained from Korea National Health Insurance Service claims data for 209,424 singleton offspring and their mothers who gave birth for the first time in 2007. The dependent variables were motor development disorder (MDD), cognitive development disorder (CDD) and combined overall neurodevelopmental disorder (NDD) from offspring. Seventeen independent variables from 2002 to 2007 were included. Random forest variable importance and Shapley Additive Explanation (SHAP) values were calculated to analyze the directions of its associations with the predictors. The random forest with oversampling registered much higher areas under the receiver-operating-characteristic curves than the logistic regression of interaction and non-linearity terms, 79% versus 50% (MDD), 82% versus 52% (CDD) and 74% versus 50% (NDD). Based on random forest variable importance, low socioeconomic status and age at birth were highly ranked. In SHAP values, there was a positive association between NDD and pre- or perinatal outcomes, especially, fetal male sex with growth restriction associated the development of NDD in offspring.

Long-term cardiovascular outcome in women with preeclampsia in Korea: a large population-based cohort study and meta-analysis.

Recent studies reported the long-term cardiovascular risk of preeclampsia. However, only a few studies have investigated the association between preeclampsia and long-term cardiovascular disease in Asian populations, although there could be racial/ethnic differences in the risk of cardiovascular diseases. Therefore, we aimed to evaluate the long-term effects of preeclampsia on cardiovascular disease in an Asian population. This study included 68,658 parous women in the Health Examinees Study (HEXA) cohort of South Korea and compared the risk of long-term cardiovascular disease, including ischemic heart disease and stroke, according to the history of preeclampsia. We also performed a meta-analysis combining current study data with data from existing literature in the Asian population. Among the study population, 3413 (5.23%) women had a history of preeclampsia, and 767 (1.12%) and 404 (0.59%) women developed ischemic heart disease and stroke for 22 years. Women with a history of preeclampsia were at a higher risk for both ischemic heart disease (adjusted hazard ratio 1.66 [1.19-2.04]) and stroke (adjusted hazard ratio 1.48 [1.02-2.16]) than those without. In the meta-analysis, the pooled hazard ratio of ischemic heart disease and stroke were also increased in women with a history of preeclampsia (ischemic heart disease 1.65 [1.51-1.82]; stroke 1.78 [1.52-2.10]).

Ultrasonographic assessment of abnormal fetal growth related to uteroplacental-fetal biometrics and Doppler (U-AID) indices: Protocol for multicenter retrospective cohort study trial.

Fetal growth restriction (FGR) is one of the leading causes of perinatal morbidity and mortality. Many studies have reported an association between FGR and fetal Doppler indices focusing on umbilical artery (UA), middle cerebral artery (MCA), and ductus venosus (DV). The uteroplacental-fetal circulation which affects the fetal growth consists of not only UA, MCA, and DV, but also umbilical vein (UV), placenta and uterus itself. Nevertheless, there is a paucity of large-scale cohort studies that have assessed the association between UV, uterine wall, and placental thickness with perinatal outcomes in FGR, in conjunction with all components of the uteroplacental-fetal circulation. Therefore, this multicenter study will evaluate the association among UV absolute flow, placental thickness, and uterine wall thickness and adverse perinatal outcome in FGR fetuses. This multicenter retrospective cohort study will include singleton pregnant women who undergo at least one routine fetal ultrasound scan during routine antepartum care. Pregnant women with fetuses having structural or chromosomal abnormalities will be excluded. The U-AID indices (UtA, UA, MCA, and UV flow, placental and uterine wall thickness, and estimated fetal body weight) will be measured during each trimester of pregnancy. The study population will be divided into two groups: (1) FGR group (pregnant women with FGR fetuses) and (2) control group (those with normal growth fetus). We will assess the association between U-AID indices and adverse perinatal outcomes in the FGR group and the difference in U-AID indices between the two groups.

Preterm birth and maternal heart disease: A machine learning analysis using the Korean national health insurance database.

BACKGROUND: Maternal heart disease is suspected to affect preterm birth (PTB); however, validated studies on the association between maternal heart disease and PTB are still limited. This study aimed to build a prediction model for PTB using machine learning analysis and nationwide population data, and to investigate the association between various maternal heart diseases and PTB. METHODS: A population-based, retrospective cohort study was conducted using data obtained from the Korea National Health Insurance claims database, that included 174,926 primiparous women aged 25-40 years who delivered in 2017. The random forest variable importance was used to identify the major determinants of PTB and test its associations with maternal heart diseases, i.e., arrhythmia, ischemic heart disease (IHD), cardiomyopathy, congestive heart failure, and congenital heart disease first diagnosed before or during pregnancy. RESULTS: Among the study population, 12,701 women had PTB, and 12,234 women had at least one heart disease. The areas under the receiver-operating-characteristic curves of the random forest with oversampling data were within 88.53 to 95.31. The accuracy range was 89.59 to 95.22. The most critical variables for PTB were socioeconomic status and age. The random forest variable importance indicated the strong associations of PTB with arrhythmia and IHD among the maternal heart diseases. Within the arrhythmia group, atrial fibrillation/flutter was the most significant risk factor for PTB based on the Shapley additive explanation value. CONCLUSIONS: Careful evaluation and management of maternal heart disease during pregnancy would help reduce PTB. Machine learning is an effective prediction model for PTB and the major predictors of PTB included maternal heart disease such as arrhythmia and IHD.

Patient blood management to minimize transfusions during the postpartum period.

Patient blood management is an evidence-based concept that seeks to minimize blood loss by maintaining adequate hemoglobin levels and optimizing hemostasis during surgery. Since the coronavirus disease 2019 pandemic, patient blood management has gained significance due to fewer blood donations and reduced amounts of blood stored for transfusion. Recently, the prevalence of postpartum hemorrhage (PPH), as well as the frequency of PPH-associated transfusions, has steadily increased. Therefore, proper blood transfusion is required to minimize PPH-associated complications while saving the patient's life. Several guidelines have attempted to apply this concept to minimize anemia during pregnancy and bleeding during delivery, prevent bleeding after delivery, and optimize recovery methods from anemia. This study systematically reviewed various guidelines to determine blood loss management in pregnant women.

Association between early preterm birth and maternal exposure to fine particular matter (PM10): A nation-wide population-based cohort study using machine learning.

Although preterm birth (PTB), a birth before 34 weeks of gestation accounts for only less than 3% of total births, it is a critical cause of various perinatal morbidity and mortality. Several studies have been conducted on the association between maternal exposure to PM and PTB, but the results were inconsistent. Moreover, no study has analyzed the risk of PM on PTB among women with cardiovascular diseases, even though those were thought to be highly susceptible to PM considering the cardiovascular effect of PM. Therefore, we aimed to evaluate the effect of PM10 on early PTB according to the period of exposure, using machine learning with data from Korea National Health Insurance Service (KNHI) claims. Furthermore, we conducted subgroup analysis to compare the risk of PM on early PTB among pregnant women with cardiovascular diseases and those without. A total of 149,643 primiparous singleton women aged 25 to 40 years who delivered babies in 2017 were included. Random forest feature importance and SHAP (Shapley additive explanations) value were used to identify the effect of PM10 on early PTB in comparison with other well-known contributing factors of PTB. AUC and accuracy of PTB prediction model using random forest were 0.9988 and 0.9984, respectively. Maternal exposure to PM10 was one of the major predictors of early PTB. PM10 concentration of 5 to 7 months before delivery, the first and early second trimester of pregnancy, ranked high in feature importance. SHAP value showed that higher PM10 concentrations before 5 to 7 months before delivery were associated with an increased risk of early PTB. The probability of early PTB was increased by 7.73%, 10.58%, or 11.11% if a variable PM10 concentration of 5, 6, or 7 months before delivery was included to the prediction model. Furthermore, women with cardiovascular diseases were more susceptible to PM10 concentration in terms of risk for early PTB than those without cardiovascular diseases. Maternal exposure to PM10 has a strong association with early PTB. In addition, in the context of PTB, pregnant women with cardiovascular diseases are a high-risk group of PM10 and the first and early second trimester is a high-risk period of PM10.

Associations of Preterm Birth with Dental and Gastrointestinal Diseases: Machine Learning Analysis Using National Health Insurance Data.

BACKGROUND: This study uses machine learning with large-scale population data to assess the associations of preterm birth (PTB) with dental and gastrointestinal diseases. METHODS: Population-based retrospective cohort data came from Korea National Health Insurance claims for 124,606 primiparous women aged 25-40 and delivered in 2017. The 186 independent variables included demographic/socioeconomic determinants, disease information, and medication history. Machine learning analysis was used to establish the prediction model of PTB. Random forest variable importance was used for identifying major determinants of PTB and testing its associations with dental and gastrointestinal diseases, medication history, and socioeconomic status. RESULTS: The random forest with oversampling data registered an accuracy of 84.03, and the areas under the receiver-operating-characteristic curves with the range of 84.03-84.04. Based on random forest variable importance with oversampling data, PTB has strong associations with socioeconomic status (0.284), age (0.214), year 2014 gastroesophageal reflux disease (GERD) (0.026), year 2015 GERD (0.026), year 2013 GERD (0.024), progesterone (0.024), year 2012 GERD (0.023), year 2011 GERD (0.021), tricyclic antidepressant (0.020) and year 2016 infertility (0.019). For example, the accuracy of the model will decrease by 28.4%, 2.6%, or 1.9% if the values of socioeconomic status, year 2014 GERD, or year 2016 infertility are randomly permutated (or shuffled). CONCLUSION: By using machine learning, we established a valid prediction model for PTB. PTB has strong associations with GERD and infertility. Pregnant women need close surveillance for gastrointestinal and obstetric risks at the same time.

The Skin Antiseptic agents at Vaginal dElivery (SAVE) trial: study protocol for a randomized controlled trial.

BACKGROUND: Cleansing of the vulva and perineum is recommended during preparation for vaginal delivery, and special attention is paid to cleansing before episiotomy because episiotomy is known to increase the risk of perineal wound infection and/or dehiscence. However, the optimal method of perineal cleansing has not been established, including the choice of antiseptic agent. To address this issue, we designed a randomized controlled trial to examine whether skin preparation with chlorhexidine-alcohol is superior to povidone-iodine for the prevention of perineal wound infection after vaginal delivery. METHODS: In this multicenter randomized controlled trial, term pregnant women who plan to deliver vaginally after episiotomy will be enrolled. The participants will be randomly assigned to use antiseptic agents for perineal cleansing (povidone-iodine or chlorhexidine-alcohol). The primary outcome is superficial or deep perineal wound infection within 30 days after vaginal delivery. The secondary outcomes are the length of hospital stay, physician office visits, or hospital readmission for infection-related complications, endometritis, skin irritations, and allergic reactions. DISCUSSION: This study will be the first randomized controlled trial aiming to determine the optimal antiseptic agent for the prevention of perineal wound infections after vaginal delivery. TRIAL REGISTRATION: ClinicalTrials.gov NCT05122169. First submitted date on 8 November 2021. First posted date on 16 November 2021.

Preterm Labor and Preterm-PROM at a Lower Gestational Age Are Associated with a Longer Latency-to-Delivery Even in Patients with the Same Intensity of Intra-Amniotic Inflammation: "Carroll-Model" Revisited.

A previous study by Carroll et al. demonstrated that the time from preterm-PROM to delivery was longer at a lower gestational age (GA) when the membranes rupture, although the presence or absence of intra-amniotic inflammation (IAI) was not examined in that study. However, patients with either preterm labor (PTL) or preterm-PROM at a lower GA had more frequent IAI, which was associated with a shorter amniocentesis-to-delivery (ATD) interval as compared with inflammation-free amniotic fluid (AF). Up to now, there is no information about whether PTL and preterm-PROM at a lower GA are associated with a shorter or longer latency to delivery in cases with the same intensity of IAI. The objective of the study is to examine this issue. AF MMP-8 was measured in 476 singleton early preterm-gestations (21.5 < GA at amniocentesis < 34 wks) with PTL (n = 253) and preterm-PROM (n = 223). Patients were divided into three groups according to GA at amniocentesis (i.e., group-1: <26 wks; group-2: 26−30 wks; group-3: 30−34 wks). IAI was defined as an elevated AF MMP-8 (≥23 ng/mL), and IAI was classified into either mild IAI (AF MMP-8: 23−350 ng/mL) or severe IAI (AF MMP-8 ≥ 350 ng/mL). ATD interval was examined according to GA at amniocentesis in the context of the same intensity of IAI (i.e., inflammation-free AF, mild IAI, and severe IAI) among pregnant women with either PTL or preterm-PROM. IAI was more frequent at a lower GA in cases with PTL (group-1 vs. group-2 vs. group-3; 59.5% vs. 47.4% vs. 25.1%; X2test, p = 0.000034 and linear by linear association [LBLA], p = 0.000008) and in those with preterm-PROM (group-1 vs. group-2 vs. group-3; 69.2% vs. 50.0% vs. 32.0%; X2test, p = 0.000104, and LBLA, p = 0.000019). Of note, cases without IAI at a lower GA had a longer ATD interval in both PTL (Spearman's rank correlation test, γ = −0.360, p = 0.000003) and preterm-PROM (γ = −0.570, p = 0.000001) groups. Moreover, the lower the GA, the longer the ATD interval, even among patients with mild and severe IAI in both PTL (Spearman's rank correlation test; mild IAI, γ = −0.290, p = 0.039; severe IAI, γ = −0.299, p = 0.048) and preterm-PROM (mild IAI, γ = −0.565, p = 0.000013; severe IAI, γ = −0.363, p = 0.015) groups. In conclusion, PTL and preterm-PROM at a lower GA are associated with a longer latency to delivery, even in patients with the same intensity of IAI. This finding suggests that a more intense IAI may be needed for spontaneous preterm birth at a lower GA.

Impact of COVID-19 on pregnant women in South Korea: Focusing on prevalence, severity, and clinical outcomes.

BACKGROUND: In the era of coronavirus disease 2019 (COVID-19) pandemic, there is a paucity of information regarding actual prevalence of COVID-19 in pregnant women compared to non-pregnant women. The purpose of this study was to investigate the prevalence of COVID-19 infection and clinical outcome in pregnant women and non-pregnant women. METHODS: This is a nationwide cross-sectional study in South Korea between January 2020 and February 2021 using the claim database. The primary outcome was the prevalence of COVID-19 in pregnant women, and the secondary outcome was the occurrence of severe COVID-19 illness among infected patients. Severity of COVID-19 was classified into four categories according to WHO ordinal scale. RESULTS: The prevalence of COVID-19 infection was lower in pregnant women than non-pregnant women aged 20-44 (0·02% vs. 0.14%, p < 0.0001). However, among COVID-19 positive women at age 20-44, pregnant women was at higher risk of oxygen therapy after hospitalization (score 4 in WHO ordinal scale: 6.4% vs. 1.6%, p < 0.05). There were no deaths or hospitalized severe disease in pregnant women with COVID-19, although the majority of them (96·2%) were admitted to hospital. On the other hand, 42·3% of non-pregnant women at 20-44 age were admitted to hospital and 0.04% of them died and 0.1% had hospitalized severe disease. CONCLUSIONS: The prevalence of COVID-19 infection in pregnant women was lower than non-pregnant women in Korea, resulting in relatively small cases of fatality. It has implications that public health policy, such as an effective response to COVID-19 and a powerful preemptive strategy for pregnant women, can lower risk of COVID-19 infection and better clinical outcomes in pregnant women with COVID-19.

Nonalcoholic fatty liver disease and early prediction of gestational diabetes mellitus using machine learning methods.

BACKGROUND/AIMS: To develop an early prediction model for gestational diabetes mellitus (GDM) using machine learning and to evaluate whether the inclusion of nonalcoholic fatty liver disease (NAFLD)-associated variables increases the performance of model. METHODS: This prospective cohort study evaluated pregnant women for NAFLD using ultrasound at 10-14 weeks and screened them for GDM at 24-28 weeks of gestation. The clinical variables before 14 weeks were used to develop prediction models for GDM (setting 1, conventional risk factors; setting 2, addition of new risk factors in recent guidelines; setting 3, addition of routine clinical variables; setting 4, addition of NALFD-associated variables, including the presence of NAFLD and laboratory results; and setting 5, top 11 variables identified from a stepwise variable selection method). The predictive models were constructed using machine learning methods, including logistic regression, random forest, support vector machine, and deep neural networks. RESULTS: Among 1,443 women, 86 (6.0%) were diagnosed with GDM. The highest performing prediction model among settings 1-4 was setting 4, which included both clinical and NAFLD-associated variables (area under the receiver operating characteristic curve [AUC] 0.563-0.697 in settings 1-3 vs. 0.740-0.781 in setting 4). Setting 5, with top 11 variables (which included NAFLD and hepatic steatosis index), showed similar predictive power to setting 4 (AUC 0.719-0.819 in setting 5, P=not significant between settings 4 and 5). CONCLUSION: We developed an early prediction model for GDM using machine learning. The inclusion of NAFLDassociated variables significantly improved the performance of GDM prediction. (ClinicalTrials.gov Identifier: NCT02276144).

Development of early prediction model for pregnancy-associated hypertension with graph-based semi-supervised learning.

Clinical guidelines recommend several risk factors to identify women in early pregnancy at high risk of developing pregnancy-associated hypertension. However, these variables result in low predictive accuracy. Here, we developed a prediction model for pregnancy-associated hypertension using graph-based semi-supervised learning. This is a secondary analysis of a prospective study of healthy pregnant women. To develop the prediction model, we compared the prediction performances across five machine learning methods (semi-supervised learning with both labeled and unlabeled data, semi-supervised learning with labeled data only, logistic regression, support vector machine, and random forest) using three different variable sets: [a] variables from clinical guidelines, [b] selected important variables from the feature selection, and [c] all routine variables. Additionally, the proposed prediction model was compared with placental growth factor, a predictive biomarker for pregnancy-associated hypertension. The study population consisted of 1404 women, including 1347 women with complete follow-up (labeled data) and 57 women with incomplete follow-up (unlabeled data). Among the 1347 with complete follow-up, 2.4% (33/1347) developed pregnancy-associated HTN. Graph-based semi-supervised learning using top 11 variables achieved the best average prediction performance (mean area under the curve (AUC) of 0.89 in training set and 0.81 in test set), with higher sensitivity (72.7% vs 45.5% in test set) and similar specificity (80.0% vs 80.5% in test set) compared to risk factors from clinical guidelines. In addition, our proposed model with graph-based SSL had a higher performance than that of placental growth factor for total study population (AUC, 0.71 vs. 0.80, p < 0.001). In conclusion, we could accurately predict the development pregnancy-associated hypertension in early pregnancy through the use of routine clinical variables with the help of graph-based SSL.