Validation of machine-learning model for first-trimester prediction of pre-eclampsia using cohort from PREVAL study.

OBJECTIVE: Effective first-trimester screening for pre-eclampsia (PE) can be achieved using a competing-risks model that combines risk factors from the maternal history with multiples of the median (MoM) values of biomarkers. A new model using artificial intelligence through machine-learning methods has been shown to achieve similar screening performance without the need for conversion of raw data of biomarkers into MoM. This study aimed to investigate whether this model can be used across populations without specific adaptations. METHODS: Previously, a machine-learning model derived with the use of a fully connected neural network for first-trimester prediction of early (< 34 weeks), preterm (< 37 weeks) and all PE was developed and tested in a cohort of pregnant women in the UK. The model was based on maternal risk factors and mean arterial blood pressure (MAP), uterine artery pulsatility index (UtA-PI), placental growth factor (PlGF) and pregnancy-associated plasma protein-A (PAPP-A). In this study, the model was applied to a dataset of 10 110 singleton pregnancies examined in Spain who participated in the first-trimester PE validation (PREVAL) study, in which first-trimester screening for PE was carried out using the Fetal Medicine Foundation (FMF) competing-risks model. The performance of screening was assessed by examining the area under the receiver-operating-characteristics curve (AUC) and detection rate (DR) at a 10% screen-positive rate (SPR). These indices were compared with those derived from the application of the FMF competing-risks model. The performance of screening was poor if no adjustment was made for the analyzer used to measure PlGF, which was different in the UK and Spain. Therefore, adjustment for the analyzer used was performed using simple linear regression. RESULTS: The DRs at 10% SPR for early, preterm and all PE with the machine-learning model were 84.4% (95% CI, 67.2-94.7%), 77.8% (95% CI, 66.4-86.7%) and 55.7% (95% CI, 49.0-62.2%), respectively, with the corresponding AUCs of 0.920 (95% CI, 0.864-0.975), 0.913 (95% CI, 0.882-0.944) and 0.846 (95% CI, 0.820-0.872). This performance was achieved with the use of three of the biomarkers (MAP, UtA-PI and PlGF); inclusion of PAPP-A did not provide significant improvement in DR. The machine-learning model had similar performance to that achieved by the FMF competing-risks model (DR at 10% SPR, 82.7% (95% CI, 69.6-95.8%) for early PE, 72.7% (95% CI, 62.9-82.6%) for preterm PE and 55.1% (95% CI, 48.8-61.4%) for all PE) without requiring specific adaptations to the population. CONCLUSIONS: A machine-learning model for first-trimester prediction of PE based on a neural network provides effective screening for PE that can be applied in different populations. However, before doing so, it is essential to make adjustments for the analyzer used for biochemical testing. © 2023 International Society of Ultrasound in Obstetrics and Gynecology.

Comparison of different methods of first-trimester screening for preterm pre-eclampsia: cohort study.

OBJECTIVE: To compare the predictive performance of three different mathematical models for first-trimester screening of pre-eclampsia (PE), which combine maternal risk factors with mean arterial pressure (MAP), uterine artery pulsatility index (UtA-PI) and serum placental growth factor (PlGF), and two risk-scoring systems. METHODS: This was a prospective cohort study performed in eight fetal medicine units in five different regions of Spain between September 2017 and December 2019. All pregnant women with singleton pregnancy and a non-malformed live fetus attending their routine ultrasound examination at 11 + 0 to 13 + 6 weeks' gestation were invited to participate in the study. Maternal characteristics and medical history were recorded and measurements of MAP, UtA-PI, serum PlGF and pregnancy-associated plasma protein-A (PAPP-A) were converted into multiples of the median (MoM). Risks for term PE, preterm PE (< 37 weeks' gestation) and early PE (< 34 weeks' gestation) were calculated according to the FMF competing-risks model, the Crovetto et al. logistic regression model and the Serra et al. Gaussian model. PE classification was also performed based on the recommendations of the National Institute for Health and Care Excellence (NICE) and the American College of Obstetricians and Gynecologists (ACOG). We estimated detection rates (DR) with their 95% CIs at a fixed 10% screen-positive rate (SPR), as well as the area under the receiver-operating-characteristics curve (AUC) for preterm PE, early PE and all PE for the three mathematical models. For the scoring systems, we calculated DR and SPR. Risk calibration was also assessed. RESULTS: The study population comprised 10 110 singleton pregnancies, including 32 (0.3%) that developed early PE, 72 (0.7%) that developed preterm PE and 230 (2.3%) with any PE. At a fixed 10% SPR, the FMF, Crovetto et al. and Serra et al. models detected 82.7% (95% CI, 69.6-95.8%), 73.8% (95% CI, 58.7-88.9%) and 79.8% (95% CI, 66.1-93.5%) of early PE; 72.7% (95% CI, 62.9-82.6%), 69.2% (95% CI, 58.8-79.6%) and 74.1% (95% CI, 64.2-83.9%) of preterm PE; and 55.1% (95% CI, 48.8-61.4%), 47.1% (95% CI, 40.6-53.5%) and 53.9% (95% CI, 47.4-60.4%) of all PE, respectively. The best correlation between predicted and observed cases was achieved by the FMF model, with an AUC of 0.911 (95% CI, 0.879-0.943), a slope of 0.983 (95% CI, 0.846-1.120) and an intercept of 0.154 (95% CI, -0.091 to 0.397). The NICE criteria identified 46.7% (95% CI, 35.3-58.0%) of preterm PE at 11% SPR and ACOG criteria identified 65.9% (95% CI, 55.4-76.4%) of preterm PE at 33.8% SPR. CONCLUSIONS: The best performance of screening for preterm PE is achieved by mathematical models that combine maternal factors with MAP, UtA-PI and PlGF, as compared to risk-scoring systems such as those of NICE and ACOG. While all three algorithms show similar results in terms of overall prediction, the FMF model showed the best performance at an individual level. © 2024 International Society of Ultrasound in Obstetrics and Gynecology.

Incidence of stillbirth: effect of deprivation.

OBJECTIVES: To examine the relationship between the English index of multiple deprivation (IMD) and the incidence of stillbirth and assess whether IMD contributes to the prediction of stillbirth provided by the combination of maternal demographic characteristics and elements of medical history. METHODS: This was a prospective, observational study of 159 125 women with a singleton pregnancy who attended their first routine hospital visit at 11 + 0 to 13 + 6 weeks' gestation in two maternity hospitals in the UK. The inclusion criterion was delivery at ≥ 24 weeks' gestation of a fetus without major abnormality. Participants completed a questionnaire on demographic characteristics and obstetric and medical history. IMD was used as a measure of socioeconomic status, which takes into account income, employment, education, skills and training, health and disability, crime, barriers to housing and services, and living environment. Each neighborhood is ranked according to its level of deprivation relative to that of other areas into one of five equal groups, with Quintile 1 containing the 20% most deprived areas and Quintile 5 containing the 20% least deprived areas. Logistic regression analysis was used to determine whether IMD provided a significant independent contribution to stillbirth after adjustment for known maternal risk factors. RESULTS: The overall incidence of stillbirth was 0.35% (551/159 125), and this was significantly higher in the most deprived compared with the least deprived group (Quintile 1 vs Quintile 5). The odds ratio (OR) in Quintile 1 was 1.57 (95% CI, 1.16-2.14) for any stillbirth, 1.64 (95% CI, 1.20-2.28) for antenatal stillbirth and 1.89 (95% CI, 1.23-2.98) for placental dysfunction-related stillbirth. In Quintile 1 (vs Quintile 5), there was a higher incidence of factors that contribute to stillbirth, including black race, increased body mass index, smoking, chronic hypertension and previous stillbirth. The OR of black (vs white) race was 2.58 (95% CI, 2.14-3.10) for any stillbirth, 2.62 (95% CI, 2.16-3.17) for antenatal stillbirth and 3.34 (95% CI, 2.59-4.28) for placental dysfunction-related stillbirth. Multivariate analysis showed that IMD did not have a significant contribution to the prediction of stillbirth provided by maternal race and other maternal risk factors. In contrast, in black (vs white) women, the risk of any and antenatal stillbirth was 2.4-fold higher and the risk of placental dysfunction-related stillbirth was 2.9-fold higher after adjustment for other maternal risk factors. CONCLUSIONS: The incidence of stillbirth, particularly placental dysfunction-related stillbirth, is higher in women living in the most deprived areas in South East England. However, in screening for stillbirth, inclusion of IMD does not improve the prediction provided by race, other maternal characteristics and elements of medical history. © 2022 International Society of Ultrasound in Obstetrics and Gynecology.

Performance of first-trimester combined screening for preterm pre-eclampsia: findings from cohort of 10 110 pregnancies in Spain.

OBJECTIVE: To evaluate the diagnostic accuracy of the Fetal Medicine Foundation (FMF) competing-risks model, incorporating maternal characteristics, mean arterial pressure (MAP), uterine artery pulsatility index (UtA-PI) and placental growth factor (PlGF) (the 'triple test'), for the prediction at 11-13 weeks' gestation of preterm pre-eclampsia (PE) in a Spanish population. METHODS: This was a prospective cohort study performed in eight fetal medicine units in five different regions of Spain between September 2017 and December 2019. All pregnant women with a singleton pregnancy and a non-malformed live fetus attending a routine ultrasound examination at 11 + 0 to 13 + 6 weeks' gestation were invited to participate. Maternal demographic characteristics and medical history were recorded and MAP, UtA-PI, serum PlGF and pregnancy-associated plasma protein-A (PAPP-A) were measured following standardized protocols. Treatment with aspirin during pregnancy was also recorded. Raw values of biomarkers were converted into multiples of the median (MoM), and audits were performed periodically to provide regular feedback to operators and laboratories. Patient-specific risks for term and preterm PE were calculated according to the FMF competing-risks model, blinded to pregnancy outcome. The performance of screening for PE, taking into account aspirin use, was assessed by calculating the area under the receiver-operating-characteristics curve (AUC) and detection rate (DR) at a 10% fixed screen-positive rate (SPR). Risk calibration of the model was assessed. RESULTS: The study population comprised 10 110 singleton pregnancies, including 72 (0.7%) that developed preterm PE. In the preterm PE group, compared to those without PE, median MAP MoM and UtA-PI MoM were significantly higher, and median serum PlGF MoM and PAPP-A MoM were significantly lower. In women with PE, the deviation from normal in all biomarkers was inversely related to gestational age at delivery. Screening for preterm PE by a combination of maternal characteristics and medical history with MAP, UtA-PI and PlGF had a DR, at 10% SPR, of 72.7% (95% CI, 62.9-82.6%). An alternative strategy of replacing PlGF with PAPP-A in the triple test was associated with poorer screening performance for preterm PE, giving a DR of 66.5% (95% CI, 55.8-77.2%). The calibration plot showed good agreement between predicted risk and observed incidence of preterm PE, with a slope of 0.983 (95% CI, 0.846-1.120) and an intercept of 0.154 (95% CI, -0.091 to 0.397). CONCLUSIONS: The FMF model is effective in predicting preterm PE in the Spanish population at 11-13 weeks' gestation. This method of screening is feasible to implement in routine clinical practice, but it should be accompanied by a robust audit and monitoring system, in order to maintain high-quality screening. © 2023 International Society of Ultrasound in Obstetrics and Gynecology.

SARS-CoV-2-specific antibodies and neutralization capacity in breast milk following infection vs vaccination.

Anti-severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibodies have been found in breast milk following both natural SARS-CoV-2 infection and coronavirus disease 2019 (COVID-19) vaccination. This was a prospective study to evaluate the temporal changes in amount and neutralization capacity of anti-SARS-CoV-2 antibodies in breast milk stimulated by natural infection and by vaccination. Serial breast milk samples were collected from postnatal women who were recruited through convenience sampling. We found a rapid increase in neutralizing SARS-CoV-2-specific antibodies in breast milk from both study groups. Amongst the infection group, the median immunoglobulin A (IgA) level was 16.99 (range, 0-86.56) ng/mL and median binding capacity was 33.65% (range, 0-67.65%), while in the vaccination group these were 30.80 (range, 0-77.40) ng/mL and 23.80% (range, 0-42.80%), respectively. In both groups, both binding capacity and IgA levels decreased progressively over time after peaking. Neutralizing activity had become undetectable by about 150 days after the first dose of the vaccine, but a vaccine booster dose restored secretion of neutralizing IgA, albeit with different levels of response in different individuals. This highlights the importance of the vaccine booster dose in sustaining neutralizing antibody levels in breast milk, which may potentially provide protection for very young children, who cannot receive the COVID-19 vaccine. © 2022 International Society of Ultrasound in Obstetrics and Gynecology.

Prognostic utility of the assessment of diastolic function in patients undergoing cardiac resynchronization therapy.

Conflicting data exist about the relationship between cardiac resynchronization therapy (CRT) and diastolic function. Aims of the study are to assess diastolic patterns in patients undergoing CRT according to the 2016 recommendations of the American Society of Echocardiography/European Association of Cardiovascular Imaging and to evaluate the prognostic value of diastolic dysfunction (DD) in CRT candidates. METHODS AND RESULTS: One-hundred ninety-three patients (age: 67 ± 11 years, QRS width: 167 ± 21 ms) were included in this multicentre prospective study. Mitral filling pattern, mitral tissue Doppler velocity, tricuspid regurgitation velocity, and indexed left atrial volume were used to classify DD from grade I to III. CRT-response, defined as a reduction of left ventricular (LV) end-systolic volume > 15% at 6-month follow-up (FU), occurred in 132 (68%) patients. The primary endpoint was a composite of heart transplantation, LV assisted device implantation, or all-cause death during FU and occurred in 29 (15%) patients. CRT was associated with a degradation of DD in non-responders. At multivariable analysis corrected for clinical variables, QRS duration, mitral regurgitation, CRT-response and LV dyssynchrony, grade I DD was associated with a better outcome (HR 0.37, 95% CI: 0.14-0.96). Non-responders with grade II-III DD had the worse prognosis (HR 4.36, 95%CI: 2.10-9.06). CONCLUSIONS: The evaluation of DD in CRT candidates allows the prognostic stratification of patients, independently from CRT-response.

Heart rate differences between symptomatic and asymptomatic Brugada syndrome patients at night.

OBJECTIVE: Ventricular arrhythmias in Brugada syndrome (BS) mainly occur at rest, especially during nighttime, suggesting that parasympathetic activity at night may play an important role in the arrhythmogenesis of the disease. This study examined and compared the autonomic function of symptomatic and asymptomatic BS patients overnight. APPROACH: We analyzed various heart rate variability (HRV) and heart rate complexity (HRC) markers in a clinical series including 87 BS patients, where 23 were symptomatic. MAIN RESULTS: Statistically significant differences were found in markers MIRR, SDNN, SDANN, [Formula: see text] and SampEn, suggesting that symptomatic patients may be related to lower heart rate variability and complexity values, as well as to greater circadian fluctuations overnight. SIGNIFICANCE: The results provide further evidence for the role of autonomic imbalance in the pathophysiology of BS, highlighting the relevance of nighttime analysis to the unmasking of significant ANS changes. Based on these outcomes, the role of HRV and HRC assessment at night could be a step forward towards the understanding of BS and the risk for the occurrence of symptoms in these patients, with a potential future impact on therapeutic strategies.

Heart rate complexity analysis in Brugada syndrome during physical stress testing.

Symptoms such as ventricular arrhythmias in Brugada syndrome (BS) typically occur at rest, especially during sleep, suggesting that the autonomic nervous system (ANS) function may be relevant in the arrhythmogenesis of the disease. The aim of this work was to assess the ANS response captured by nonlinear heart rate variability (HRV) measures in 69 patients diagnosed with BS, who underwent a standardized physical stress test. Heart rate complexity (HRC) was evaluated by the power-law scaling analysis (ß slope) during rest, exercise, recovery and rest post-recovery, in order to discriminate between symptomatic and asymptomatic BS patients. Symptomatic patients showed a significant reduction in HRC in comparison to asymptomatic subjects, after exertion (p = 0.015); during the whole recovery period (p = 0.023), and in particular within the passive recovery phase (p = 0.025), as well as during rest post-recovery (p = 0.022). Based on these results, symptoms could be associated with a lower ANS complexity during the stress test stages where parasympathetic activity is predominant. Therefore, the proposed HRV indicators could be of help in the risk stratification of asymptomatic patients.

A bond graph model of the cardiovascular system.

The study of the autonomic nervous system (ANS) function has shown to provide useful indicators for risk stratification and early detection on a variety of cardiovascular pathologies. However, data gathered during different tests of the ANS are difficult to analyse, mainly due to the complex mechanisms involved in the autonomic regulation of the cardiovascular system (CVS). Although model-based analysis of ANS data has been already proposed as a way to cope with this complexity, only a few models coupling the main elements involved have been presented in the literature. In this paper, a new model of the CVS, representing the ventricles, the circulatory system and the regulation of the CVS activity by the ANS, is presented. The models of the vascular system and the ventricular activity have been developed using the Bond Graph formalism, as it proposes a unified representation for all energetic domains, facilitating the integration of mechanic and hydraulic phenomena. In order to take into account the electro-mechanical behaviour of both ventricles, an electrophysiologic model of the cardiac action potential, represented by a set of ordinary differential equations, has been integrated. The short-term ANS regulation of heart rate, cardiac contractility and peripheral vasoconstriction is represented by means of continuous transfer functions. These models, represented in different continuous formalisms, are coupled by using a multi-formalism simulation library. Results are presented for two different autonomic tests, namely the Tilt Test and the Valsalva Manoeuvre, by comparing real and simulated signals.