Background: Pauci-immune glomerulonephritis (GN) is a renal manifestation of anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV), occurring in 70-100% of cases. Kidney involvement in AAV can lead to chronic kidney disease (CKD), which progresses to end-stage renal disease (ESRD) in approximately 20% of cases. When patients show positive ANCA results accompanied by urinary abnormalities such as hematuria and proteinuria, pauci-immune GN is suspected. However, there have been some cases in which patients with positive ANCA results revealed unexpected pathologies during tissue examination. Objectives: We aimed to investigate and compare the characteristics of patients showing positive ANCA results with pauci-immune GN to those without, determined based on renal tissue examination. Methods: We retrospectively reviewed the electronic medical records of patients who underwent renal biopsy showing positive ANCA results from October 2013 to August 2021 at a single tertiary center in Seoul, South Korea. Logistic regression analysis was performed to identify the variables associated with pauci-immune GN among the patients who underwent renal biopsy and showed positive ANCA results. Results: A total of 216 patients who underwent renal biopsy showing positive ANCA results were included in our analysis. The mean age of the patients was 61.7 ± 13.2 years, with 50.9% of them being female. Among all patients, 153 (70.8%) were diagnosed with pauci-immune GN through renal biopsy, while 63 (29.2%) were diagnosed with other kidney diseases. There was no statistically significant difference in terms of gender between the pauci-immune GN group and the non-pauci-immune GN group. However, the pauci-immune GN group's mean age (65.3 years) was higher than that of the non-pauci-immune GN group (53.0 years). Autoantibody analysis revealed that 92.8% of the pauci-immune GN group tested positive for anti-myeloperoxidase(MPO), whereas only 38.7% in the non-pauci-immune GN group showed positivity (p < 0.001). Furthermore, the pauci-immune GN group showed a median anti-MPO titer of 115 IU/mL, which was significantly higher than the non-pauci-immune GN group's median titer of 12 IU/mL (p < 0.001). Regarding blood chemistry, the pauci-immune GN group showed a higher white blood cell (WBC) count, lower hemoglobin levels, higher platelet count, elevated erythrocyte sedimentation rate (ESR), increased C-reactive protein (CRP), and higher creatinine levels compared to the non-pauci-immune GN group, all of which were statistically significant differences. Urinalysis was determined according to the Birmingham Vasculitis Activity Score (BVAS) criteria, which indicated that the pauci-immune GN group had fewer patients with proteinuria greater than 1+ and more patients with hematuria greater than 10 per high power field compared to the non-pauci-immune GN group. The renal tissue analysis of the non-pauci-immune GN group revealed diverse pathological findings. The most prevalent conditions included IgA nephritis at 20.6%, followed by lupus nephritis at 17.5% of the group. Additional identified pathologies included membranous glomerulonephritis, diabetic nephropathy, chronic nephropathy, and several others. Considering the correlations among blood chemistry levels, a multivariate analysis was conducted to adjust for age, platelet count, CRP, creatinine, and proteinuria levels. The results of the analysis revealed a significant association between anti-MPO positivity (OR: 14.258; 4.963–40.957) and hematuria (OR: 2.857; 1.102–7.407) with the presence of pauci-immune GN observed in renal tissue examination. Conclusion: In the analyzed cohort of patients, approximately 70% of those who underwent renal biopsy and showed positive ANCA results were diagnosed with pauci-immune GN. There was a significant association between anti-MPO positivity and hematuria on microscopy with pauci-immune GN. Our findings emphasize the importance of identifying high-risk antibody profiles and urinalysis associated with pauci-immune GN. REFERENCES: NIL. Acknowledgements: NIL. Disclosure of Interests: None declared.

resulting from in vitro conception or assisted reproduction techniques.Sociodemographic data, ultrasound information and maternal and perinatal outcomes were retrieved for the patient's medical records.Reference values for normal DVIP were between 0.68 and 1.24.Statistical analysis was performed considering significant associations (p < 0.05).Results: 1620 women were taken to first trimester screening study of which 722 met inclusion criteria.424 pregnancies without adverse outcomes and 298 pregnancies with an adverse outcome.89%, 6,5%, 4.5% had normal, high and low DVIP.Statistically significant differences were found between the patients who presented elevated DVIP and miscarriage (p < 0.01) and pre-eclampsia (P < 0.011).Low DVIP values were associated with congenital heart disease (p < 0.01), neonatal coagulopathy (p < 0.01) and chromosomal abnormality (p 0.014).Finally, combined analysis about adverse maternal-perinatal outcomes did not have statistically significant causal associations.Conclusions: Although DVIP is a measurement used in first trimester screening, we did not find statistically significant causal associations between abnormal DVIP and adverse maternalperinatal outcomes.

Birthweight discordance, defined as a difference in the birthweights of twins, is a well-documented phenomenon in twin pregnancies and occurs in 15% to 30% of twin pregnancies depending on the chronicity, degree of discordance and the threshold. It is not clear whether catch-up growth exists within a set of discordant twins, although twins achieve the normal weight of singletons during infancy and up to adolescence. Therefore, the aim of this study is to evaluate the longitudinal patterns of growth in discordant twins. This study was conducted by merging the Korean National Health Insurance (KNHI) claims database and a National Health Screening Program for Infants and Children database. Using the KNHI claims database, we identified all women who delivered twins between January 1, 2007, and December 31, 2010. Only discordant twins were included for analysis. The weight difference was calculated as “(larger twin weight−smaller twin weight) / larger twin weight×100”. The birthweight discordance and weight at age point discordance were defined as a weight difference of 20% or more at birth and at each age point. The linear spline mixed-effect model (LSMM) was selected, based on the differences between the negative of twice the log-likelihood (-2LogL) among several models with different random-effect terms. An alternative LSMM with a subject-specific random intercept and repeated time of trajectories was also applied to the weight discordance measure of each twin. A total of 3,408 twins and 1,704 mothers were included in this study. The mean of birthweight discordance was 28.68%. The gap in weight between the small twins and their large siblings significantly closed to 10% at 1 year of age and then remained at about 10% until 8 years of age. Small twin in discordant twins showed catch-up growth in the first 12 months of life compared to their large siblings but after 1 year of age, no differences in growth velocity were found. Further studies are needed to evaluate the long-term pattern of growth and its effect on health.

a specificity of 86.0%, a positive predictive value of 23.8%, and a negative predictive value of 100%.Conclusions: We have identified four predictors and a model of high neonatal serum magnesium concentrations that can guide maternal magnesium treatment.Our study will allow safer use of magnesium and better prognosis for neonates.

Results: A total of 120 measurements (67 diabetes, 53 controls) were included in this study.The mean gestational age of measurement was 32.7 ± 2.4 weeks in the control group and 34.1 ± 3.4 weeks in the diabetes group.There was no significant difference for Mod-MPI of the left fetal heart between the diabetes and the control group (0.52 [0.47-0.57] vs. 0.50 [0.45-0.53],p = 0.067).Mod-MPI of the right heart was significantly higher in the diabetes group (0.57 [0.50-0.62])compared to the control group (0.52 [0.47-0.57],p = 0.0070).However, there was no significant difference in Mod-MPI of the fetal right or left heart between pregestational diabetes and gestational diabetes group.Conclusions: Mod-MPI of the right fetal heart is significantly higher in the diabetes group and might have a predictive value of fetal cardiac function associated with the adverse perinatal outcome of diabetic pregnancy.EP01.

The objective of this study was to determine the association between maternal pre-eclampsia and discordant twins, especially in relation to small for gestational age (SGA) in smaller twin. This retrospective multicentre cohort study included 2,047 women with dichorionic (DC) twin pregnancies who delivered after 24weeks of gestation, with confirmed both fetal weight, and no congenital anomalies. Growth discordancy was defined as a birthweight difference ≥15%, and SGA is defined as less than 10th percentile of the Korean twin birthweight reference. First, the occurrence of pre-eclampsia was compared between discordant and concordant twins. Second, the study population was divided into 4 groups based on the presence or absence of discordancy and SGA, and the risks of pre-eclampsia among groups were calculated. The prevalence of pre-eclampsia is 14.2% in the discordant twin, which was significantly higher than the 6.3% in the concordant twin (p < 0.001). However, other maternal morbidities including preterm labour, preterm premature rupture of membranes, gestational diabetes were not associated with discordant twins. The degree of discordancy and the risk of pre-eclampsia showed a linear association. (p < 0.001). When compared to concordant twins without SGA, the highest risk of pre-eclampsia was observed in cases of discordant twins with SGA (adjusted OR 3.289; 95% CI 2.15-5.03, p < 0.001), followed by discordant twins without SGA (adjusted OR 2.071, 95% CI 1.37-2.31, p < 0.001). Maternal pre-eclampsia is associated with discordant DC twins, and significant discordance or the presence of SGA may be contributing factors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Most of the previous studies have established the association between Doppler indices and fetal growth restriction (FGR), with a primary focus on umbilical artery (UA), and middle cerebral artery (MCA). This study aims to assess changes in ultrasonographic uteroplacental-fetal biometry and Doppler measurements more comprehensively including umbilical vein (UV) flow, placental and uterine wall thickness between fetuses with and without FGR. This multicentre observational study will be conducted as both retrospective and prospectively. We will enrol singleton pregnant women who have underwent at least one fetal ultrasound scan during the routine antepartum care. Women with fetuses that have chromosomal or structural abnormalities will be excluded. The study will measure U-AID indices, which are including uterine artery, UA, MCA, and UV flow, as well as placental and uterine wall thickness, and estimated fetal body weight during each trimester. The study population will be divided into two groups based on fetal growth, which are the FGR group and the control group. We will compare the U-AID indices between the two groups and evaluate the association between these indices and adverse perinatal outcomes in the FGR group. Additionally, we will develop a predictive model for adverse perinatal outcomes in the FGR group using both clinical risk factors and U-AID indices by multivariable logistic regression analysis. Not applicable. The U-AID study provides a comprehensive evaluation of uteroplacental-fetal biometry and Doppler indices according to fetal growth. This study will provide valuable insights into the potential utility of these indices in predicting and managing adverse perinatal outcomes. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

The objective of this study was to evaluate the risk of adverse neonatal outcome according to timing of delivery in twins with a gestational age of 37 weeks or more and determine the optimal timing of delivery in uncomplicated twin pregnancies. This retrospective study enrolled uncomplicated Korean twin pregnant women who delivered after 37 weeks of gestation from 2015 to 2021. Women diagnosed with hypertension or diabetes mellitus (DM) prior to pregnancy, those experiencing maternal complications such as gestational hypertension or gestational DM, placenta previa or placental abruption, and those with twin birthweight abnormalities (birthweight of any twin below the 10th percentile or discordance of 20% or more) were excluded from the study. The primary outcome was composite neonatal morbidity including the followings: transient tachypnea, respiratory distress syndrome, hypoglycemia, the use of ventilator and nasal cannula within 7 days after birth. Adverse outcomes in twins were analysed using either individual twins or twin pairs as a case. Logistic regression was applied to calculate the odds ratios of adverse outcomes based on the gestational week at the time of delivery. Out of 25,698 women who delivered at or beyond 37 weeks of gestation, 13,502 women were finally included in the study. In the multivariate analysis, when compared to deliveries at 37 weeks of gestation, the composite morbidity was lower for twin deliveries at both 38 weeks (adjusted odds ratio (aOR), 0.697; 95% Confidence interval (CI), 0.637-0.762) and 39 weeks (aOR, 0.526; 95% CI, 0.417-0.663) of gestation. For deliveries after 40 weeks, the risk of composite morbidity was not significantly different from deliveries at GA 37 weeks (aOR, 0.63; 95% CI, 0.389-1.021). In uncomplicated twin pregnancies, delaying the delivery timing even within the full-term periods is suggested, with appropriate fetal surveillance and personalised consideration of both maternal and fetal conditions.

Objectives: MicroRNAs (miRNA) are regulators of cellular processes including (anti-) angiogenesis and differentiation of trophoblast cells.They play a role in the pathophysiology of pre-eclampsia (PE).The aim of this study is to investigate the

Objectives: To provide reference ranges for fetal cardiac parameters by using HeartAssist™, an automated system for measuring such parameters, in 17 large referral centres in South Korea.Methods: This prospective study included normal singleton fetuses between 16.0 and 36.6 weeks of gestational age (GA) from October 2022 to February 2023.Video clips of the fetal hearts were collected, and a total of 34 parameters were measured in six cardiac views.In the four chamber view, we measured cardiac axis; thoracic area and circumference; heart area and circumference; atrial widths and

Background Neuropsychiatric systemic lupus erythematosus (NPSLE) presenting with mood disorder, headache, psychosis, and cognitive impairment appears within 1 year of SLE diagnosis in more than half of cases (1, 2). Most studies have focused on the epidemiology of neuropsychiatric manifestations in patients with established SLE (3). Therefore, diagnosing NPSLE in patients who have visited the hospital with psychiatric symptoms is challenging. Although some studies have conducted anti-nuclear antibody (ANA) screening in psychiatric patients, none have occurred in an Asian population (4, 5). Objectives We aimed to determine the benefits of ANA screening for NPSLE in patients admitted to the department of psychiatry in Korea. Methods We investigated patients admitted to the department of psychiatry who underwent ANA testing between January 2015 and December 2021 at a single tertiary center in Korea. Patients diagnosed with SLE before admission were excluded from this study. Electronic medical records, including ANA titer, extractable nuclear antigen (ENA), complement, brain magnetic resonance imaging, electroencephalogram, and cerebrospinal fluid analysis, were reviewed retrospectively. Diagnosis at psychiatric hospitalization was classified according to the International Classification of Diseases (ICD)-10. Results Throughout the study period, 2523 patients were hospitalized, 1355 of whom underwent ANA testing. Three patients with SLE were excluded. The median age of all patients was 40 (27–58), and 897 (66.2%) were female. Of the 1355 patients, 96 (7.1%) were positive with a titer of ≥1:80, and 61 (4.5%) were positive with a titer of ≥1:160. Among the 17 patients who underwent ENA testing, 1 was positive for anti-Ro and anti-La, eventually diagnosed with Sjogren's syndrome. According to the diagnostic classification of admission, there was no significant difference in the ANA positivity rate (p=0.205). Conclusion There was no difference in the positivity rate of ANA in the general population when testing was performed for screening purposes on patients admitted to the psychiatric department. Additionally, none of the 1355 patients were diagnosed with NPSLE after undergoing ANA screening. Thus, the benefits of performing routine screening appear to be limited. References [1]Arthritis Rheum. 1999 Apr;42(4):599-608[2]Neurol Clin. 1999 Nov;17(4):901-41[3]Semin Arthritis Rheum. 2012 Oct;42(2):179-85[4]Arthritis Care Res (Hoboken). 2022 Mar;74(3):427-432[5]Medicine (Baltimore). 2016 Nov;95(47):e5288 2 Disclosure of Interests None declaredTable 1Baseline characteristics of patients admitted to the department of psychiatry who underwent ANA testingN = 1355Female897 (66.2)Age, years40 (27-58)ANA titer1:8035 (36.5)1:16037 (38.5)1:32012 (12.5)1:6408 (8.3)1:12804 (4.2)ANA fluorescenceNucleolar25 (26.0)Speckled36 (37.5)Discrete speckled11 (11.5)Homogeneous24 (25.0)Diagnosis at hospitalizationOrganic, including symptomatic, mental disorders44 (3.2)Mental and behavioral disorders due to psychoactive substance use47 (3.5)Schizophrenia, schizotypal and delusional disorders324 (23.9)Mood disorders748 (55.2)Neurotic, stress-related and somatoform disorders166 (12.3)Values are n (%), or median (interquartile range); ANA: anti-nuclear antibody

Background Nonsteroidal anti-inflammatory drugs (NSAIDs) are the basis of treatment for patients with ankylosing spondylitis (AS) to reduce pain and inflammation. However, long-term NSAID use may cause deterioration of renal function. Objectives We aimed to investigate the renal safety from long-term NSAID exposure in patients with AS. Methods We extracted clinical information from the electronic medical records of patients with newly diagnosed AS with an observation period of more than one year who were treated at Asan Medical Center from January 1, 2000, to December 31, 2020, excluding patients with chronic kidney disease Stage 4–5, known kidney disease, and malignancy. Renal function was determined by calcuation of the estimated glomerular filtration rate (GFR) using the Modification of Diet in Renal Disease (MDRD) and Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations. Patients were stratified into three groups according to initial GFR: Stage 1 (GFR≥90), Stage 2 (60≤GFR<90), and Stage 3 (30≤GFR<60). The GFR values for every 3-year interval were imputed by the linear interpolation method using the measured values before and after each point, and NSAID use in each interval was estimated as the medication possession rate (MPR). To evaluate the association between NSAID use and annual changes in GFR, linear mixed effects models were used and adjusted for key covariates including age, sex, body mass index, baseline GFR, relative dosage of NSAIDs, comorbidities, and comedications. Results Among 1,838 patients with AS, the mean age was 38.0 (SD, 13.8) years with a median 6.1 years of follow-up. The majority of 1,386 (75.4%) patients belonged to the Stage 1 group, and the number of patients in the Stage 2 and Stage 3 groups was 442 (24.0%) and 10 (0.5%), respectively. In the multivariable models, the mean annual decrease in GFR was significant (model 1: β=-0.684, 95% CI -0.980 to -0.389, p<0.001; model 2: β=-0.387, 95% CI -0.770 to -0.004, p=0.048), and the MPR of NSAID use was significantly associated with an additional decrease in GFR (model 1: for 100% vs. 0% of MPR, β =-0.5, 95% CI, -0.8 to -0.1, p=0.008; model 2: for highest vs. lowest tertile of MPR, β=-0.519, 95% CI -0.906 to -0.132, p=0.009) in the Stage 1 group but not in patients in the Stage 2 group (Table 1). Conclusion This study suggested that long-term use of NSAIDs was associated with a small but significant decrease in GFR in patients with AS with normal initial renal function. References [1]Möller B et al. Chronic NSAID use and long-term decline of renal function in a prospective rheumatoid arthritis cohort study. Ann Rheum Dis. 2015 Apr;74(4):718-23.[2]Curhan GC et al. Lifetime nonnarcotic analgesic use and decline in renal function in women. Arch Intern Med. 2004 Jul 26;164(14):1519-24. Acknowledgements: NIL. Disclosure of Interests None Declared.Table 1Effect of covariates on eGFR declineModel 1†Model 2‡Beta95% CIp-valueBeta95% CIp-valueLBUBLBUBStage 1Intercept-0.684-0.980-0.389<0.001-0.387-0.770-0.0040.048Baseline GFR-0.066-0.080-0.053<0.001-0.066-0.080-0.053<0.001NSAID MPR-0.005-0.008-0.0010.008NSAID MPR tertile1st tertile (<12)0.0302nd tertile (≥12, <96)-0.352-0.7200.0160.0613rd tertile (≥96)-0.519-0.906-0.1320.009Stage 2Intercept-0.297-3.1552.5610.8370.378-2.4733.2290.794Baseline GFR-0.063-0.1440.0170.122-0.064-0.1450.0170.117NSAID MPR-0.010-0.0250.0060.211NSAID MPR tertile1st tertile (<12)<0.0012nd tertile (≥12, <96)-1.257-2.024-0.4890.0013rd tertile (≥96)-1.862-2.676-1.049<0.001†Model 1 consisted of GFR at baseline, NSAID MPR, other covariates and random effect for patient‡Model 2 consisted of GFR at baseline, tertile of NSAID MPR, other covariates and random effect for patient

We investigated the effect of fetal growth restriction on neonatal morbidities by comparing neonatal outcomes of the smaller twin with small for gestational age (SGA) and larger co-twin in discordant twin pregnancies. In this multicentre retrospective study, twin data were collected for five years from ten university hospitals in South Korea. This study included dichorionic twins with birthweight discordancy ≥ 15%. SGA was defined as birthweight < 10%, according to the Korean twin birthweight reference. First, we divided the study population into two groups according to the presence or absence of SGA in the smaller twin. Second, we compared the neonatal morbidities between smaller twin and larger co-twin in each group. For statistical analysis, generalised estimation equation models were used for comparison between twins. A total of 597 women with discordant twin pregnancy were included, of whom 242 (40.5%) and 355 (59.5%) were in the smaller twin with SGA group and the smaller twin without SGA group, respectively. In the smaller twin with SGA group, the development of RDS was significantly decreased and surfactant use was lower in the smaller twin than the larger co-twin (RDS, 12.4% vs. 16.9%, p < 0.05; surfactant use, 9.9% vs. 14.9%, p < 0.05). This difference remained after adjusting for fetal sex and birth order (adjusted OR, 0.73; 95% CI 0.54-0.98). In the smaller twin without SGA group, the overall neonatal morbidities were not different between smaller twin and larger co-twin. Growth restriction in discordant twins might be protective for neonatal RDS in smaller twin. However, the larger co-twin is at a relatively risk for neonatal RDS. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

A spina bifida is a malformation that occurs when the caudal neural tube fails to fuse. We present a case of spina bifida diagnosed on first trimester. A 35-years-old primigravida presented at our hospital due to fetal megacystis on routine ultrasonography on 12 weeks of gestation at local medical centre. The 2.07x1.29x1.40 cm3 sized, anechoic intraabdominal cyst without keyhole sign suggested bladder outlet obstruction syndrome. Chorionic villi sampling results showed normal chromosomes with microarray. Since the fetus was female, the possibility of urethral atresia was raised. Observation of the fetal brain on ultrasound showed decreased intracranial translucency with crash sign, which refers to thalamus displacement to the lowermost position. The size of cerebral aqueduct was diminished considering the gestational age. View of the fetal spine showed a widening of lumbar vertebra in many segments. The overall signs on the first trimester ultrasound indicated spina bifida with myeloschisis. The patient went through abortion at 13 weeks of gestation at local medical centre and the abortus confirmed the diagnosis of spina bifida. Spina bifida is often diagnosed during the second trimester. However, intracranial markers, including intracranial translucency, can be used during the first trimester for screening spina bifida. Thus, the importance of first trimester screening cannot be more emphasised. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

As technologic advances of ultrasound with high resolution, the importance and accuracy of identifying fetal structural abnormalities in first trimester has increased. We aimed to develop an artificial intelligence system called ViewAssist™ that automatically identifies and classifies the fetal structures in the first trimester and to evaluate its performance. Ultrasound images of over 8,000 first trimester fetuses were prospectively collected from Seoul St. Mary's Hospital, Eunpyeong St. Mary's Hospital, Uijeongbu St. Mary's Hospital, and Buchoen St. Mary's Hospital. The collected images were classified by anatomical areas including head, face, neck, chest, heart, abdomen, extremities, and spine, according to the International Society of Ultrasound in Obstetrics and Gynecology guideline. For effective classification of the anatomical areas, we used the Vision Transformer architecture and pre-processed the data by augmenting images while maintaining the aspect ratio. We randomly selected images from the original data, with 90% for the training set and 10% for the validation set. During the deep learning training stage, we used 100 epochs, with a batch size of 64 and a learning rate of 0.0005. We adjusted the hyperparameters at each epoch to minimise the loss function. The classification accuracy of the anatomical areas using ViewAssist™ was 96.75% overall, with the following category-specific accuracy: 100% for head, 98.9% for face, 93.8% for neck, 85.7% for chest, 100% for heart, 94.5% for abdomen, 97.5% for extremities, and 96.2% for spine. Through deep learning algorithms, ViewAssist™ was able to accurately classify the structures of first trimester fetuses. It is expected to be clinically useful in assisting with identifying fetal structures and diagnosing structural abnormalities in the first trimester.

This study investigated capacity of machine learning prediction of pregnancy associated hypertension (PAH), based on early pregnancy data, in Asian women. We used machine learning approach to predict PAH and preterm PAH on retrospective data of 35,004 pregnancies in multicentres of Korea. After randomly dividing the data into training and test sets at a 70:30 ratio, predictive models for PAH were created by three methods; gradient boosting models using variables by 1) prespecified by the American College of Obstetrics and Gynecology (ACOG), 2) the Shapley values (original) and 3) boruta algorithm from the full variables. Performance was evaluated with area under the curve (AUC) and area under the precision-recall curve (auPR), in total and restricted populations after excluding women who received aspirin during pregnancy. The incidence of PAH was 8.11%, with 3.3% of preterm PAH. At baseline, performance of models for PAH with ACOG, original, and boruta variables achieved AUCs of 0.724, 0.8, and 0.782, and auPRs of 0.236, 0.359, and 0.323, respectively. At 11-13 weeks of gestation, performance with original and boruta variables achieved AUCs of 0.821 and 0.816, and auPRs of 0.412 and 0.416. Selected variables by boruta were maternal age, prepregnancy body mass index, MoMs of PAPP-A and hCG, pre-existing hypertension, family history of hypertension, blood pressure at the first trimester, history of PAH, preterm birth, postpartum bleeding, Caesarean section, and the number of live children. At baseline, performance for preterm PAH with ACOG, original, and boruta variables achieved AUCs of 0.727, 0.791, and 0.785, and at 11-13 weeks of gestation, performance with original and boruta variables achieved AUCs of 0.817 and 0.810, respectively. In the restricted population, results were similar. In conclusion, machine learning models in early pregnancy demonstrated moderate to high performances, with the use of routine clinical variables. (Korea Health Industry Development Institute, HI21C1300).