Northwest Women & Children Hospital

Preterm brain injury (PBI) is a prevalent complication in preterm infants, leading to the destruction of critical structural and functional brain connections and placing a significant burden on families. The timely detection of PBI is of paramount importance for the prevention and treatment of the condition. However, the absence of specific clinical manifestations in the early stages of PBI renders it susceptible to misdiagnosis and missed diagnoses. Moreover, once it occurs, there is no specific treatment available. The aim of this study was to develop and validate a machine learning (ML) based interpretable model for the early detection of PBI, as well as the assessment of patient-wide and individual risk factors for this disease.

This study utilized a cohort of premature infants provided by Northwest Women's and Children's Hospital in China, comprising medical records of 650 premature infants, spanning from 2019 to 2021. PBI were identified based on cranial magnetic resonance imaging (MRI). Fourteen machine learning models were employed with stratified 10-fold cross-validation method used to evaluate model performance. The Shapley Additive Explanations (SHAP) method was applied for model interpretation. Feature selection methods were used to determine the final model which was validated on the independent test set. Subsequently, risk factors for the entire cohort and individual patients were assessed.

Among the fourteen machine learning models, the CatBoost model demonstrated the best discriminative ability. Following feature selection, the final model was constructed using seven features, designated as PBIPred (Preterm Brain Injury Predictor). PBIPred exhibited strong performance in both 10-fold cross-validation and independent test set (AUC = 0.8229) for accurately predicting PBI. The screening for risk factors in the cohort and individuals identified the following variables as positive risk factors for PBI: Mechanical ventilation (MV), Weight, Anemia of prematurity (AOP), Respiratory distress syndrome (RDS), Albumin (ALB), and White blood cell (WBC).

The PBIPred webserver and PBIPred tool were developed for clinical diagnosis and large-scale local medical record data prediction. They can be accessed freely at http://pbipred.liaolab.net and https://github.com/chikit2077/PBIPred, respectively.

Are live birth rates (LBRs) per woman following flexible progestin-primed ovarian stimulation (fPPOS) treatment non-inferior to LBRs per woman following the conventional GnRH-antagonist protocol in expected suboptimal responders undergoing freeze-all cycles in assisted reproduction treatment?

In women expected to have a suboptimal response, the 12-month likelihood of live birth with the fPPOS treatment did not achieve the non-inferiority criteria when compared to the standard GnRH antagonist protocol for IVF/ICSI treatment with a freeze-all strategy.

The standard PPOS protocol is effective for ovarian stimulation, where medroxyprogesterone acetate (MPA) is conventionally administered in the early follicular phase for ovulatory suppression. Recent retrospective cohort studies on donor cycles have shown the potential to prevent premature ovulation and maintain oocyte yields by delaying the administration of MPA until the midcycle (referred to as fPPOS), similar to GnRH antagonist injections. With milder pituitary suppression, the fPPOS protocol may be a less costly option for women expected to have a low or suboptimal response if a fresh embryo transfer is not intended.

This was a non-inferiority, open-label randomized controlled trial conducted at a tertiary assisted reproduction center. A total of 484 participants were randomized in the study between July 2020 and June 2023 with a 1:1 allocation.

Infertile women with a predicted suboptimal ovarian response (<40 years old, antral follicle count <10, and basal serum FSH < 12 mIU/ml) were randomly assigned to receive either fPPOS treatment or GnRH antagonist treatment. MPA (10 mg) or GnRH antagonist (0.25 mg) was administered daily once the leading follicle reached 14 mm and continued until the day of trigger. All viable embryos were cryopreserved for subsequent frozen-thawed embryo transfer in both groups. The primary endpoint was the proportion of live births per woman within 12 months post-randomization (with a non-inferiority margin of –12.5%). The analysis was assessed in the per-protocol population.

Twenty-two women withdrew at the beginning of the stimulation phase due to COVID-19. Eight women did not proceed with the assigned frozen embryo transfer, and six switched from the fPPOS to the antagonist protocol. Overall, 449 women were included in the per-protocol analysis, with 216 in the fPPOS group and 233 in the GnRH antagonist group. The LBRs per woman were 44.4% (96/216) for participants in the fPPOS group and 48.9% (114/233) for participants in the GnRH antagonist group [risk ratio (RR) 0.91 (95% CI, 0.74, 1.11), risk difference (RD) –4.5% (95% CI, –13.7, 4.7)], which did not meet the non-inferiority criterion (–12.5%). Oocyte and embryonic parameters were not significantly different between the two groups. Nine women (4.17%) in the fPPOS group experienced a premature luteinizing hormone surge, compared to five women (2.15%) in the antagonist group. Only one woman in the fPPOS group ovulated before oocyte retrieval.

The distinct routes of administration for the medications precluded blinding in this open-label trial, potentially influencing outcome assessments. All participants were recruited in a single center from one country, limiting the generalizability.

While MPA is considered a patient-friendly alternative to antagonists for women undergoing scheduled freeze-all cycles, the GnRH antagonist protocol should still be the preferred treatment for anticipated suboptimal responders in terms of LBR.

This trial was funded by Science and Technology Department of Shaanxi Province, China (2021SF-210). Innovation Team of Shaanxi Provincial Health and Reproductive Medicine Research (2023TD-04); Key Industrial Chain Projects in Shaanxi Province: Research on Assisted Reproductive Technologies and Precision Prevention System for Genetic Diseases Preconception (2023-ZDLSF-48). Science and Technology Department of Shaanxi Province, China (2022SF-564). B.W.M. reports consultancy, travel support and research funding from Merck KGaA and consultancy for Organon and Norgine; owning stock in ObsEva; and holding an NHMRC Investigator Grant (GNT1176437). Other authors declare no conflicts of interest. All other authors have nothing to declare.

Registered at Chinese clinical trial registry (www.chictr.org.cn). Registry Identifier: ChiCTR2000030356.

29 February 2020.

11 March 2020

This research seeks to ascertain the prevalence and determinants of mirror-image dextrocardia in fetuses

With December 2022 as the reference point, we compiled colleted data on pregnant women who carried fetuses with mirror-image dextrocardia in Xi’an, Shaanxi Province: September–October 2022, November 2022, and December 2022–January 2023. An online questionnaire was distributed to 209 pregnant across China who had contracted COVID-19. The case group comprised women whose final menstrual cycle occurred in November 2022 and who had a fetus with mirror-image dextrocardia. Women with a November 2022 final menstrual period and a fetus without this condition made up the control group. To identify the risk factors associated with fetal mirror-image dextrocardia, both univariate and multivariate logistic regression analyses were employed.

A significant difference was noted in the gestational age at COVID-19 infection women with a September to October 2022 and December 2022 to January 2023 final menstrual period who did not bear a fetus with mirror-image dextrocardia, and those with a November 2022 final menstrual period whose fetus exhibited this condition. The univariate and multivariate analyses conducted on pregnant women with a final menstrual period in November 2022 who had contracted COVID-19 revealed significant differences in the presence and duration of fever between those bearing fetuses with mirror-image dextrocardia and those without (P = 0.000).

The findings suggest two critical factors to the increased prevalence of fetal mirror-image dextrocardia: 1) the infection timing which occurs between the 4th and 6th week of pregnancy and 2) the presence of fever and its prolonged duration.