Search

The aim of this study was to investigate the pulmonary vasculature in baseline conditions and after maternal hyperoxygenation in growth restricted fetuses (FGR). A prospective cohort study of singleton pregnancies including 97 FGR and 111 normally grown fetuses was carried out. Ultrasound Doppler of the pulmonary vessels was obtained at 24-37 weeks of gestation and data were acquired before and after oxygen administration. After, Machine Learning (ML) and a computational model were used on the Doppler waveforms to classify individuals and estimate pulmonary vascular resistance (PVR). Our results showed lower mean velocity time integral (VTI) in the main pulmonary and intrapulmonary arteries in baseline conditions in FGR individuals. Delta changes of the main pulmonary artery VTI and intrapulmonary artery pulsatility index before and after hyperoxygenation were significantly greater in FGR when compared with controls. Also, ML identified two clusters: A (including 66% controls and 34% FGR) with similar Doppler traces over time and B (including 33% controls and 67% FGR) with changes after hyperoxygenation. The computational model estimated the ratio of PVR before and after maternal hyperoxygenation which was closer to 1 in cluster A (cluster A 0.98 ± 0.33 vs cluster B 0.78 ± 0.28, p = 0.0156). Doppler ultrasound allows the detection of significant changes in pulmonary vasculature in most FGR at baseline, and distinct responses to hyperoxygenation. Future studies are warranted to assess its potential applicability in the clinical management of FGR., (© 2024. The Author(s).)

Most artificial intelligence (AI) research and innovations have concentrated in high-income countries, where imaging data, IT infrastructures and clinical expertise are plentiful. However, slower progress has been made in limited-resource environments where medical imaging is needed. For example, in Sub-Saharan Africa, the rate of perinatal mortality is very high due to limited access to antenatal screening. In these countries, AI models could be implemented to help clinicians acquire fetal ultrasound planes for the diagnosis of fetal abnormalities. So far, deep learning models have been proposed to identify standard fetal planes, but there is no evidence of their ability to generalise in centres with low resources, i.e. with limited access to high-end ultrasound equipment and ultrasound data. This work investigates for the first time different strategies to reduce the domain-shift effect arising from a fetal plane classification model trained on one clinical centre with high-resource settings and transferred to a new centre with low-resource settings. To that end, a classifier trained with 1792 patients from Spain is first evaluated on a new centre in Denmark in optimal conditions with 1008 patients and is later optimised to reach the same performance in five African centres (Egypt, Algeria, Uganda, Ghana and Malawi) with 25 patients each. The results show that a transfer learning approach for domain adaptation can be a solution to integrate small-size African samples with existing large-scale databases in developed countries. In particular, the model can be re-aligned and optimised to boost the performance on African populations by increasing the recall to [Formula: see text] and at the same time maintaining a high precision across centres. This framework shows promise for building new AI models generalisable across clinical centres with limited data acquired in challenging and heterogeneous conditions and calls for further research to develop new solutions for the usability of AI in countries with fewer resources and, consequently, in higher need of clinical support., (© 2023. The Author(s).)

The identification of factors predisposing to severe COVID-19 in young adults remains partially characterized. Low birth weight (LBW) alters cardiovascular and lung development and predisposes to adult disease. We hypothesized that LBW is a risk factor for severe COVID-19 in non-elderly subjects. We analyzed a prospective cohort of 397 patients (18-70 years) with laboratory-confirmed SARS-CoV-2 infection attended in a tertiary hospital, where 15% required admission to Intensive Care Unit (ICU). Perinatal and current potentially predictive variables were obtained from all patients and LBW was defined as birth weight ≤ 2.500 g. Age (adjusted OR (aOR) 1.04 [1-1.07], P = 0.012), male sex (aOR 3.39 [1.72-6.67], P < 0.001), hypertension (aOR 3.37 [1.69-6.72], P = 0.001), and LBW (aOR 3.61 [1.55-8.43], P = 0.003) independently predicted admission to ICU. The area under the receiver-operating characteristics curve (AUC) of this model was 0.79 [95% CI, 0.74-0.85], with positive and negative predictive values of 29.1% and 97.6% respectively. Results were reproduced in an independent cohort, from a web-based survey in 1822 subjects who self-reported laboratory-positive SARS-CoV-2 infection, where 46 patients (2.5%) needed ICU admission (AUC 0.74 [95% CI 0.68-0.81]). LBW seems to be an independent risk factor for severe COVID-19 in non-elderly adults and might improve the performance of risk stratification algorithms.

Fetal growth restriction defined as the failure to achieve the fetal genetic growth potential is a major cause of perinatal morbidity and mortality. The role of maternal adaptations to placental insufficiency in this disorder is still not fully understood. We aimed to investigate the biological processes and protein-protein interactions involved in late-onset fetal growth restriction in particular. We applied 2D nano LC-MS/MS proteomics analysis on maternal blood samples collected at the time of delivery from 5 singleton pregnancies with late-onset fetal growth restriction and 5 uncomplicated pregnancies. Data were analyzed using R package "limma" and Ingenuity Pathway Analysis. 25 proteins showed significant changes in their relative abundance in late-onset fetal growth restriction (p value < 0.05). Direct protein-protein interactions network demonstrated that Neurogenic locus notch homolog protein 1 (NOTCH1) was the most significant putative upstream regulator of the observed profile. Gene ontology analysis of these proteins revealed the involvement of 14 canonical pathways. The most significant biological processes were efflux of cholesterol, efflux of phospholipids, adhesion of blood cells, fatty acid metabolism and dyslipidemia. Future studies are warranted to validate the potential role of the detected altered proteins as potential therapeutic targets in the late-onset form of fetal growth restriction.

The goal of this study was to evaluate the maturity of current Deep Learning classification techniques for their application in a real maternal-fetal clinical environment. A large dataset of routinely acquired maternal-fetal screening ultrasound images (which will be made publicly available) was collected from two different hospitals by several operators and ultrasound machines. All images were manually labeled by an expert maternal fetal clinician. Images were divided into 6 classes: four of the most widely used fetal anatomical planes (Abdomen, Brain, Femur and Thorax), the mother's cervix (widely used for prematurity screening) and a general category to include any other less common image plane. Fetal brain images were further categorized into the 3 most common fetal brain planes (Trans-thalamic, Trans-cerebellum, Trans-ventricular) to judge fine grain categorization performance. The final dataset is comprised of over 12,400 images from 1,792 patients, making it the largest ultrasound dataset to date. We then evaluated a wide variety of state-of-the-art deep Convolutional Neural Networks on this dataset and analyzed results in depth, comparing the computational models to research technicians, which are the ones currently performing the task daily. Results indicate for the first time that computational models have similar performance compared to humans when classifying common planes in human fetal examination. However, the dataset leaves the door open on future research to further improve results, especially on fine-grained plane categorization.

Microbial invasion of the amniotic cavity (MIAC) is only identified by amniocentesis, an invasive procedure that limits its clinical translation. Here, we aimed to evaluate whether the vaginal metabolome discriminates the presence/absence of MIAC in women with preterm labor (PTL) and intact membranes. We conducted a case-control study in women with symptoms of PTL below 34 weeks who underwent amniocentesis to discard MIAC. MIAC was defined as amniotic fluid positive for microorganisms identified by specific culture media. The cohort included 16 women with MIAC and 16 control (no MIAC). Both groups were matched for age and gestational age at admission. Vaginal fluid samples were collected shortly after amniocentesis. Metabolic profiles were analyzed by nuclear magnetic resonance (NMR) spectroscopy and compared using multivariate and univariate statistical analyses to identify significant differences between the two groups. The vaginal metabolomics profile of MIAC showed higher concentrations of hypoxanthine, proline, choline and acetylcholine and decreased concentrations of phenylalanine, glutamine, isoleucine, leucine and glycerophosphocholine. In conclusion, metabolic changes in the NMR-based vaginal metabolic profile are able to discriminate the presence/absence of MIAC in women with PTL and intact membranes. These metabolic changes might be indicative of enhanced glycolysis triggered by hypoxia conditions as a consequence of bacterial infection, thus explaining the utilization of alternative energy sources in an attempt to replenish glucose.

This study aims to investigate the relation between vaginal microbiota and exposition to intra-amniotic inflammation (IAI). We conducted a prospective cohort study in women with preterm labor <34 weeks who had undergone amniocentesis to rule out IAI. Vaginal samples were collected after amniocentesis. Women with IAI included those with positive amniotic fluid (AF) for a microorganism identified by specific culture media and Sanger sequencing 16S ribosomal RNA gene and/or high AF interleukin (IL)-6 levels. Vaginal microbiota was characterized by 16S ribosomal RNA gene amplicon sequencing. Specific quantitative PCR targeted to Lactobacillus spp. was also performed. Regression models were used to evaluate associations between vaginal microbiota and exposition to IAI. Concerning our results, 64 women were included. We observed an inverse association between AF IL-6 levels and load of Lactobacillus spp. Depletion in Lactobacillus spp. load was significantly associated with an early gestational age at delivery and a short latency to delivery. Microbial-diversity was found to be a risk factor for the subsequent occurrence of clinical chorioamnionitis. To the contrary, higher Lactobacillus spp. load had a protective role. In conclusion, the study identifies reduced bacterial load of Lactobacillus spp. in women exposed to IAI and found microbial-diversity and Lactobacillus spp. depletion to be associated with a worse perinatal outcome.

BACKGROUNDThe aims of this study were to (i) compare the concentrations of two neural injury markers, S100B protein and neuron-specific enolase (NSE), in intrauterine growth-restricted (IUGR) fetuses and in fetuses with appropriate growth-for-gestational-age (AGA), and (ii) investigate potential relationships between concentrations of these markers, Doppler abnormalities, and adverse perinatal or neonatal outcomes.METHODSThis was a case-controlled, cooperative, prospective study among Spanish Maternal and Child Health Network (Retic SAMID) hospitals. At inclusion, biometry for estimated fetal weight and feto-placental Doppler were measured. At the time of delivery, maternal venous blood and fetal umbilical arterial blood samples were collected. S100B and NSE concentrations were determined from these samples.RESULTSIn total, 254 pregnancies were included. Among these, 147 were classified as IUGR and 107 as AGA. There were no differences between the groups in S100B concentrations. However, levels of NSE in maternal and umbilical cord serum differed significantly between these groups (2.31 in AGA vs. 2.51 in IUGR in (P<0.05); and 2.89 in AGA vs. 3.25 in IUGR (P<0.05), respectively). No differences were observed in these neurological markers when stratified by perinatal or neonatal complications.CONCLUSIONAlthough some variations exist in these neurological markers, they did not correlate with perinatal or neonatal complications.

Background: Our aim was to determine the influence of breastfeeding and postnatal nutrition on cardiovascular remodeling induced by fetal growth restriction (FGR)., Methods: A cohort study including 81 children with birthweight <10th centile (FGR) and 121 with adequate fetal growth for gestational age (AGA) was conducted. Cardiovascular endpoints were left ventricular sphericity index (LVSI), carotid intima-media thickness (cIMT), and blood pressure (BP) at 4-5 y of age. The combined effect of FGR and postnatal variables-including breastfeeding, fat dietary intake, and BMI-on cardiovascular endpoints was assessed by linear and robust regressions., Results: FGR was the strongest predictor of cardiovascular remodeling in childhood, leading to lower LVSI and increased cIMT and BP as compared with AGA. Breastfeeding >6 mo (coefficient: 0.0982) and healthy-fat dietary intake (coefficient: -0.0128) showed an independent beneficial effect on LVSI and cIMT, respectively. Overweight/obesity induced an additional increment of 1 SD on cIMT in FGR children (interaction coefficient: 0.0307) when compared with its effect in AGA. BMI increased systolic BP (coefficient: 0.7830) while weight catch-up increased diastolic BP (coefficient: 4.8929)., Conclusions: Postnatal nutrition ameliorates cardiovascular remodeling induced by FGR. Breastfeeding and healthy-fat dietary intake improved while increased BMI worsened cardiovascular endpoints, which opens opportunities for targeted postnatal interventions from early life.

To assess the usefulness of circulating microRNAs (miRNAs) as non-invasive molecular biomarkers for early prediction of preeclampsia, a differential miRNA profiling analysis was performed in first-trimester pooled sera from 31 early preeclampsia patients, requiring delivery before 34 weeks of gestation, and 44 uncomplicated pregnancies using microfluidic arrays. Among a total of 754 miRNAs analyzed, the presence of 63 miRNAs (8%) was consistently documented in the sera from preeclampsia and control samples. Nevertheless, only 15 amplified miRNAs (2%) seemed to be differentially, although modestly, represented (fold change range: 0.4-1.4). After stem loop RT-qPCR from individual samples, the statistical analysis confirmed that none of the most consistent and differentially represented miRNAs (3 overrepresented and 4 underrepresented) were differentially abundant in serum from preeclamptic pregnancies compared with serum from normal pregnancies. Therefore, maternal serum miRNA assessment at first-trimester of pregnancy does not appear to have any predictive value for early preeclampsia.