Vaginal matrix metalloproteinase-9 (MMP-9) as a potential early predictor of preterm birth.

OBJECTIVES: To evaluate the differences in vaginal matrix metalloproteinases (MMP) and tissue inhibitors of metalloproteinases (TIMPs) in pregnant patients with a history of prior preterm birth compared with controls. METHODS: A prospective cohort pilot study recruited patients during prenatal care with history of prior spontaneous preterm birth (high-risk group) or no history of preterm birth (low-risk/controls). Inclusion criteria were singleton gestation at 11-16 weeks and between 18 and 55 years of age. Exclusion criteria were diabetes mellitus, hypertension, diseases affecting the immune response or acute vaginitis. A vaginal wash was performed at time of enrollment, and patients were followed through delivery. Samples were analyzed using semi-quantitative analysis of MMPS and TIMPS. The study was approved by the IRB and a p-value <0.05 was considered significant. RESULTS: A total of 48 pregnant patients were recruited: 16 with a history of preterm birth (high-risk group) and 32 with no history of preterm birth (low-risk group/controls). Groups were similar in age, race, BMI, and delivery mode. The high-risk group had more multiparous women (100 vs. 68.8 %; p=0.02), a greater preterm birth rate (31.2 vs. 6.3 %; p=0.02), and a lower birth weight (2,885 ± 898 g vs. 3,480 ± 473 g; p=0.02). Levels of vaginal MMP-9 were greater in high-risk patients than low-risk patients (74.9 % ± 27.0 vs. 49.4 % ± 31.1; p=0.01). When dividing the cohort into patients that had a spontaneous preterm birth (7/48, 14.6 %) vs. those with a term delivery (41/48, 85.4 %), the vaginal MMP-9 remained elevated in the cohort that experienced a preterm birth (85.46 %+19.79 vs. 53.20 %+31.47; p=0.01). There were no differences in the other MMPS and in TIMPs between high and low-risk groups. CONCLUSIONS: There was an increase in vaginal MMP-9 during early pregnancy in those at high risk for preterm birth and in those who delivered preterm, regardless of prior pregnancy outcome. Vaginal MMP-9 may have potential as a marker of increased risk of preterm birth.

Missingness patterns in a comprehensive instrument identifying psychosocial and substance use risk in antenatal care.

BACKGROUND: Psychosocial vulnerabilities (e.g. inadequate social support, financial insecurity, stress) and substance use elevate risks for adverse perinatal outcomes and maternal mental health morbidities. However, various barriers, including paucity of validated, simple and usable comprehensive instruments, impede execution of the recommendations to screen for such vulnerabilities in the first antenatal care visit. The current study presents findings from a newly implemented self-report tool created to overcome screening barriers in outpatient antenatal clinics. METHODS: This was a retrospective chart-review of 904 women who completed the Profile for Maternal & Obstetric Treatment Effectiveness (PROMOTE) during their first antenatal visit between June and December 2019. The PROMOTE includes the 4-item NIDA Quick Screen and 15 additional items that each assess a different psychosocial vulnerability. Statistical analysis included evaluation of missing data, and exploration of missing data patterns using univariate correlations and hierarchical clustering. RESULTS: Three quarters of women (70.0%) had no missing items. In the entire sample, all but four PROMOTE items (opioid use, planned pregnancy, educational level, and financial state) had < 5% missing values, suggesting good acceptability and feasibility. Several respondent-related characteristics such as lower education, less family support, and greater stress were associated with greater likelihood of missing items. Instrument-related characteristics associated with missing values were completing the PROMOTE in Spanish or question positioning at the end of the instrument. CONCLUSIONS AND IMPLICATIONS: Conducting a comprehensive screening of theoretically and clinically meaningful vulnerabilities in an outpatient setting is feasible. Study findings will inform modifications of the PROMOTE and subsequent digitisation.

Applying machine learning methods to psychosocial screening data to improve identification of prenatal depression: Implications for clinical practice and research.

We utilized machine learning (ML) methods on data from the PROMOTE, a novel psychosocial screening tool, to quantify risk for prenatal depression for individual patients and identify contributing factors that impart greater risk for depression. Random forest algorithms were used to predict likelihood for being at high risk for prenatal depression (Edinburgh Postnatal Depression Scale; EPDS ≥ 13 and/or positive self-injury item) using data from 1715 patients who completed the PROMOTE. Performance matrices were calculated to assess the ability of the PROMOTE to accurately classify patients. Probability for depression was calculated for individual patients. Finally, recursive feature elimination was used to evaluate the importance of each PROMOTE item in the classification of depression risk. PROMOTE data were successfully used to predict depression with acceptable performance matrices (accuracy = 0.80; sensitivity = 0.75; specificity = 0.81; positive predictive value = 0.79; negative predictive value = 0.97). Perceived stress, emotional problems, family support, age, major life events, partner support, unplanned pregnancy, current employment, lifetime abuse, and financial state were the most important PROMOTE items in the classification of depression risk. Results affirm the value of the PROMOTE as a psychosocial screening tool for prenatal depression and the benefit of using it in conjunction with ML methods. Using such methods can help detect underreported outcomes and identify what in patients' lives makes them more vulnerable, thus paving the way for effective individually tailored precision medicine.

BOOST ENSEMBLE LEARNING FOR CLASSIFICATION OF CTG SIGNALS.

During the process of childbirth, fetal distress caused by hypoxia can lead to various abnormalities. Cardiotocography (CTG), which consists of continuous recording of the fetal heart rate (FHR) and uterine contractions (UC), is routinely used for classifying the fetuses as hypoxic or non-hypoxic. In practice, we face highly imbalanced data, where the hypoxic fetuses are significantly underrepresented. We propose to address this problem by boost ensemble learning, where for learning, we use the distribution of classification error over the dataset. We then iteratively select the most informative majority data samples according to this distribution. In our work, in addition to addressing the imbalanced problem, we also experimented with features that are not commonly used in obstetrics. We extracted a large number of statistical features of fetal heart tracings and uterine activity signals and used only the most informative ones. For classification, we implemented several methods: Random Forest, AdaBoost, k-Nearest Neighbors, Support Vector Machine, and Decision Trees. The paper provides a comparison in the performance of these methods on fetal heart rate tracings available from a public database. Our results show that most applied methods improved their performances considerably when boost ensemble was used.

UNSUPERVISED CLUSTERING AND ANALYSIS OF CONTRACTION-DEPENDENT FETAL HEART RATE SEGMENTS.

The computer-aided interpretation of fetal heart rate (FHR) and uterine contraction (UC) has not been developed well enough for wide use in delivery rooms. The main challenges still lie in the lack of unclear and nonstandard labels for cardiotocography (CTG) recordings, and the timely prediction of fetal state during monitoring. Rather than traditional supervised approaches to FHR classification, this paper demonstrates a way to understand the UC-dependent FHR responses in an unsupervised manner. In this work, we provide a complete method for FHR-UC segment clustering and analysis via the Gaussian process latent variable model, and density-based spatial clustering. We map the UC-dependent FHR segments into a space with a visual dimension and propose a trajectory-based FHR interpretation method. Three metrics of FHR trajectory are defined and an open-access CTG database is used for testing the proposed method.

IMPROVING PHASE-RECTIFIED SIGNAL AVERAGING FOR FETAL HEART RATE ANALYSIS.

Low umbilical artery pH is a marker for neonatal acidosis and is associated with an increased risk for neonatal complications. The phase-rectified signal averaging (PRSA) features have demonstrated superior discriminatory or diagnostic ability and good interpretability in many biomedical applications including fetal heart rate analysis. However, the performance of PRSA method is sensitive to values of the selected parameters which are usually either chosen based on a grid search or empirically in the literature. In this paper, we examine PRSA method through the lens of dynamical systems theory and reveal the intrinsic connection between state space reconstruction and PRSA. From this perspective, we then introduce a new feature that can better characterize dynamical systems comparing with PRSA. Our experimental results on an open-access intrapartum Cardiotocography database demonstrate that the proposed feature outperforms state-of-the-art PRSA features in pH-based fetal heart rate analysis.

Cardiotocography analysis by empirical dynamic modeling and Gaussian processes.

Introduction: During labor, fetal heart rate (FHR) and uterine activity (UA) can be continuously monitored using Cardiotocography (CTG). This is the most widely adopted approach for electronic fetal monitoring in hospitals. Both FHR and UA recordings are evaluated by obstetricians for assessing fetal well-being. Due to the complex and noisy nature of these recordings, the evaluation by obstetricians suffers from high interobserver and intraobserver variability. Machine learning is a field that has seen unprecedented advances in the past two decades and many efforts have been made in computerized analysis of CTG using machine learning methods. However, in the literature, the focus is often only on FHR signals unlike in evaluations performed by obstetricians where the UA signals are also taken into account. Methods: Machine learning is a field that has seen unprecedented advances in the past two decades and many efforts have been made in computerized analysis of CTG using machine learning methods. However, in the literature, the focus is often only on FHR signals unlike in evaluations performed by obstetricians where the UA signals are also taken into account. In this paper, we propose to model intrapartum CTG recordings from a dynamical system perspective using empirical dynamic modeling with Gaussian processes, which is a Bayesian nonparametric approach for estimation of functions. Results and Discussion: In the context of our paper, Gaussian processes are capable for simultaneous estimation of the dimensionality of attractor manifolds and reconstructing of attractor manifolds from time series data. This capacity of Gaussian processes allows for revealing causal relationships between the studied time series. Experimental results on real CTG recordings show that FHR and UA signals are causally related. More importantly, this causal relationship and estimated attractor manifolds can be exploited for several important applications in computerized analysis of CTG recordings including estimating missing FHR samples, recovering burst errors in FHR tracings and characterizing the interactions between FHR and UA signals.

The impacts of the COVID-19 pandemic on birth satisfaction in a prospective cohort of 2,341 U.S. women.

PROBLEM: Birth satisfaction is an important health outcome that is related to postpartum mood, infant caretaking, and future pregnancy intention. BACKGROUND: The COVID-19 pandemic profoundly affected antenatal care and intrapartum practices that may reduce birth satisfaction. AIM: To investigate the extent to which pandemic-related factors predicted lower birth satisfaction. METHODS: 2341 women who were recruited prenatally in April-May 2020 and reported a live birth between April-October 2020 were included in the current analysis. Hierarchical linear regression to predict birth satisfaction from well-established predictors of birth satisfaction (step 1) and from pandemic-related factors (step 2) was conducted. Additionally, the indirect associations of pandemic-related stress with birth satisfaction were investigated. FINDINGS: The first step of the regression explained 35% of variance in birth satisfaction. In the second step, pandemic-related factors explained an additional 3% of variance in birth satisfaction. Maternal stress about feeling unprepared for birth due to the pandemic and restrictions on companions during birth independently predicted lower birth satisfaction beyond the non-pandemic variables. Pandemic-related unpreparedness stress was associated with more medicalized birth and greater incongruence with birth preference, thus also indirectly influencing birth satisfaction through a mediation process. DISCUSSION: Well-established contributors to birth satisfaction remained potent during the pandemic. In addition, maternal stress and restriction on accompaniment to birth were associated with a small but significant reduction in birth satisfaction. CONCLUSION: Study findings suggest that helping women set flexible and reasonable expectations for birth and allowing at least one intrapartum support person can improve birth satisfaction.

Psychological measures of patient's decision-making for prenatal genetic testing.

OBJECTIVE: To use a questionnaire to determine the levels of maternal decision-related distress, clarity of the pros and cons, and certainty when considering prenatal genetic diagnostic testing; and to assess the relationship between these constructs and patient characteristics. METHOD: Cross-sectional study. Voluntary, anonymous questionnaires distributed 2017-2019 to women referred for invasive prenatal genetic testing. Excluded: English or Spanish illiterate. Maternal characteristics were collected. Questions evaluated distress, decisional certainty, and decisional clarity on a 5-point Likert scale (range: 0 = low/uncertain/unclear to 4 = high/certain/clear). Analysis: non-parametric Kruskal-Wallis, correlation statistics, and ANOVA. RESULTS: Forty-four female patients completed it. Most were married, white, Catholic, and multiparous. 58% had already made a testing decision. Patients expressed low distress levels (mean 1.18 ± 0.80) and expressed high decisional certainty (mean 3.28 ± 0.76) and clarity (mean 3.30 ± 0.99). Decisional certainty and clarity were positively correlated (r = 0.47, p < .01), whereas distress was negatively correlated with decisional certainty (r = -0.8136, p < .0005) and decisional clarity (r = -0.49, p = .007). No significant differences by religion or parity. Greater distress (p < .05) and less decisional clarity (p = .07) occurred between those still debating testing vs those who had decided. CONCLUSIONS: Higher maternal distress scores were associated with lower decisional certainty and decisional clarity in women considering prenatal genetic testing.

IDENTIFICATION OF UTERINE CONTRACTIONS BY AN ENSEMBLE OF GAUSSIAN PROCESSES.

Identifying uterine contractions with the aid of machine learning methods is necessary vis-á-vis their use in combination with fetal heart rates and other clinical data for the assessment of a fetus wellbeing. In this paper, we study contraction identification by processing noisy signals due to uterine activities. We propose a complete four-step method where we address the imbalanced classification problem with an ensemble Gaussian process classifier, where the Gaussian process latent variable model is used as a decision-maker. The results of both simulation and real data show promising performance compared to existing methods.

CLASS-IMBALANCED CLASSIFIERS USING ENSEMBLES OF GAUSSIAN PROCESSES AND GAUSSIAN PROCESS LATENT VARIABLE MODELS.

Classification with imbalanced data is a common and challenging problem in many practical machine learning problems. Ensemble learning is a popular solution where the results from multiple base classifiers are synthesized to reduce the effect of a possibly skewed distribution of the training set. In this paper, binary classifiers based on Gaussian processes are chosen as bases for inferring the predictive distributions of test latent variables. We apply a Gaussian process latent variable model where the outputs of the Gaussian processes are used for making the final decision. The tests of the new method in both synthetic and real data sets show improved performance over standard approaches.

Adverse Perinatal Outcomes Predicted by Prenatal Maternal Stress Among U.S. Women at the COVID-19 Pandemic Onset.

BACKGROUND: High stress prenatally contributes to poor maternal and infant well-being. The coronavirus disease 2019 (COVID-19) pandemic has created substantial stress for pregnant women. PURPOSE: To understand whether stress experienced by women pregnant at the beginning of the pandemic was associated with a greater prevalence of adverse perinatal outcomes. METHODS: Pregnant women across the USA aged ≥18 years old enrolled in a prospective cohort study during the pandemic onset (T1) in April-May 2020. This report focuses on the 1,367 participants who gave birth prior to July-August 2020 (T2). Hierarchical logistic regression models predicted preterm birth, small for gestational age infants, and unplanned operative delivery from T1 stress, sociodemographic, and medical factors. RESULTS: After controlling for sociodemographic and medical factors, preterm birth was predicted by high prenatal maternal stress, delivering an infant small for gestational age was predicted by interpersonal violence and by stress related to being unprepared for birth due to the pandemic, and unplanned cesarean or operative vaginal delivery was predicted by prenatal appointment alterations, experiencing a major stressful life event, and by stress related to being unprepared for birth due to the pandemic. Independent of these associations, African American women were more likely than other groups to deliver preterm. CONCLUSION: Pregnant women who are experiencing high stress during the COVID-19 pandemic are at risk of poorer perinatal outcomes. A longitudinal investigation is critical to determine whether prenatal maternal stress and resulting outcomes have longer-term consequences for the health and well-being of children born in the midst of the current pandemic.

Bayesian Nonparametric Dimensionality Reduction of Categorical Data for Predicting Severity of COVID-19 in Pregnant Women.

The coronavirus disease (COVID-19) has rapidly spread throughout the world and while pregnant women present the same adverse outcome rates, they are underrepresented in clinical research. We collected clinical data of 155 test-positive COVID-19 pregnant women at Stony Brook University Hospital. Many of these collected data are of multivariate categorical type, where the number of possible outcomes grows exponentially as the dimension of data increases. We modeled the data within the unsupervised Bayesian framework and mapped them into a lower dimensional space using latent Gaussian processes. The latent features in the lower dimensional space were further used for predicting if a pregnant woman would be admitted to a hospital due to COVID-19 or would remain with mild symptoms. We compared the prediction accuracy with the dummy/one-hot encoding of categorical data and found that the latent Gaussian process had better accuracy.

Unsupervised Detection of Anomalies in Fetal Heart Rate Tracings using Phase Space Reconstruction.

Detection of anomalies in time series is still a challenging problem. In this paper, we provide a new approach to unsupervised detection of anomalies in time series based on the concept of phase space reconstruction and manifolds. We propose a rotation-insensitive metric for quantifying the similarity of manifolds and a method that uses it for estimating the probability of an outlier. The proposed method does not rely on any features and can be used for signals with variable lengths. We tested it on both synthetic signals and real fetal heart rate tracings. The method has promising performance and can be used for interpreting the severity of fetal asphyxia.

Vulnerability and resilience to pandemic-related stress among U.S. women pregnant at the start of the COVID-19 pandemic.

RATIONALE: Women pregnant during the COVID-19 pandemic are experiencing moderate to high levels of emotional distress, which has previously been shown to be attributable to two types of pandemic-related pregnancy stress: stress associated with feeling unprepared for birth due to the pandemic (Preparedness Stress) and stress related to fears of perinatal COVID-19 infection (Perinatal Infection Stress). OBJECTIVE: Given the well-documented harms associated with elevated prenatal stress and the critical importance of developing appropriately targeted interventions, we investigated factors predictive of pandemic-related pregnancy stress. METHOD: Between April 25 and May 15, 2020, 4,451 pregnant women in the U.S. were recruited via social media to complete an online questionnaire that included sociodemographic, medical, and COVID-19 situational factors, as well as the Pandemic-Related Pregnancy Stress Scale (PREPS). Binary logistic regression was used to calculate odds ratios for high stress. RESULTS: Nearly 30% of participants reported high Preparedness Stress; a similar proportion reported high Perinatal Infection Stress. Abuse history, chronic illness, income loss due to the pandemic, perceived risk of having had COVID-19, alterations to prenatal appointments, high-risk pregnancy, and being a woman of color were associated with greater levels of one or both types of stress. Access to outdoor space, older age, and engagement in healthy behaviors were protective against stress. CONCLUSIONS: Practices that may alleviate pandemic-related stress such as minimizing disruptions to prenatal care, ensuring access to outdoor space, and motivating engagement in health behaviors are of vital importance. Particular attention is needed for more vulnerable populations including women of color, women with a history of abuse, and those with high-risk pregnancy. Research focused on the short and longer-term impact of pandemic-related pregnancy stress on maternal mental and physical health, perinatal outcomes, and child development is critical to identify these effects and marshal appropriate resources to reduce them.

Identification of factors associated with delayed treatment of obstetric hypertensive emergencies.

BACKGROUND: Obstetric hypertensive emergency is defined as having systolic blood pressure ≥160 mm Hg or diastolic blood pressure ≥110 mm Hg, confirmed 15 minutes apart. The American College of Obstetricians and Gynecologists recommends that acute-onset, severe hypertension be treated with first line-therapy (intravenous labetalol, intravenous hydralazine or oral nifedipine) within 60 minutes to reduce risk of maternal morbidity and death. OBJECTIVE: Our objective was to identify barriers that lead to delayed treatment of obstetric hypertensive emergency. STUDY DESIGN: A retrospective cohort study was performed that compared women who were treated appropriately within 60 minutes vs those with delay in first-line therapy. We identified 604 patients with discharge diagnoses of chronic hypertension, gestational hypertension, or preeclampsia using International Classification of Diseases-10 codes and obstetric antihypertensive usage in a pharmacy database at 1 academic institution from January 2017 through June 2018. Of these, 267 women (44.2%) experienced obstetric hypertensive emergency in the intrapartum period or within 2 days of delivery; the results from 213 women were used for analysis. We evaluated maternal characteristics, presenting symptoms and circumstances, timing of hypertensive emergency, gestational age at presentation, and administered medications. Chi square, Fisher's exact, Wilcoxon rank-sum, and sample t-tests were used to compare the 2 groups. Univariable logistic regression was applied to determine predictors of delayed treatment. Multivariable regression model was also performed; C-statistic and Hosmer and Lemeshow goodness-of-fit test were used to assess the model fit. A result was considered statistically significant at P<.05. RESULTS: Of the 213 women, 110 (51.6%) had delayed treatment vs 103 (48.4%) who were treated within 60 minutes. Patients who had delayed treatment were 3.2 times more likely to have an initial blood pressure in the nonsevere range vs those who had timely treatment (odds ratio, 3.24; 95% confidence interval, 1.85-5.68). Timeliness of treatment was associated with presence or absence of preeclampsia symptoms; patients without preeclampsia symptoms were 2.7 times more likely to have delayed treatment (odds ratio, 2.68; 95% confidence interval, 1.50-4.80). Patients with hypertensive emergencies that occurred overnight between 10 pm and 6 am were 2.7 times more likely to have delayed treatment vs those emergencies that occurred between 6 am and 10 pm (odds ratio, 2.72; 95% confidence interval, 1.27-5.83). Delayed treatment also had an association with race, with white patients being 1.8 times more likely to have delayed treatment (odds ratio, 1.79; 95% confidence interval, 1.04-3.08). Patients who were treated at <60 minutes had a lower gestational age at presentation vs those with delayed treatment (34.6±5 vs 36.6±4 weeks, respectively; P<.001). For every 1-week increase in gestational age at presentation, there was a 9% increase in the likelihood of delayed treatment (odds ratio, 1.11; 95% confidence interval, 1.04-1.19). Another factor that was associated with delay of treatment was having a complaint of labor symptoms, which made patients 2.2 times as likely to experience treatment delay (odds ratio, 2.17; 95% confidence interval, 1.07-4.41). CONCLUSION: Initial blood pressure in the nonsevere range, absence of preeclampsia symptoms, presentation overnight, white race, having complaint of labor symptoms, and increasing gestational age at presentation are barriers that lead to a delay in the treatment of obstetric hypertensive emergency. Quality improvement initiatives that target these barriers should be instituted to improve timely treatment.

Estimation of Consecutively Missed Samples in Fetal Heart Rate Recordings.

During labor, fetal heart rate (FHR) is monitored externally using Doppler ultrasound. This is done continuously, but for various reasons (e.g., fetal or maternal movements) the system does not record any samples for varying periods of time. In many settings, it would be quite beneficial to estimate the missing samples. In this paper, we propose a (deep) Gaussian process-based approach for estimation of consecutively missing samples in FHR recordings. The method relies on similarities in the state space and on exploiting the concept of attractor manifolds. The proposed approach was tested on a short segment of real FHR recordings. The experimental results indicate that the proposed approach is able to provide more reliable results in comparison to several interpolation methods that are commonly applied for processing of FHR signals.

Sacral osteomyelitis after laparoscopic sacral colpopexy performed after a recent dental extraction: a case report.

BACKGROUND: Sacral osteomyelitis and subsequent discitis is a rare complication after laparoscopic sacral colpopexy to repair apical vaginal prolapse. CASE: We present a patient who developed Bacteroides fragilis sacral osteomyelitis and discitis after laparoscopic sacrocolpopexy with synthetic monofilament mesh and sacral titanium coil fixation. The patient had undergone dental extraction of 3 infected teeth approximately 2 weeks before sacrocolpopexy for stage IV apical vaginal prolapse. Computed tomography and magnetic resonance imaging (MRI) confirmed sacral osteomyelitis and discitis along with Bacteroides fragilis bacteremia approximately a week and a half after the original surgery. The patient was followed up with serial MRIs of the spine which revealed degeneration at the sacral promontory. The patient underwent successful removal of the entire mesh and sacral titanium coils with resolution of her symptoms. Follow-up MRI of the spine revealed resolution of her sacral osteomyelitis. CONCLUSIONS: Sacral osteomyelitis is a rare complication after sacrocolpopexy for pelvic organ prolapse repair. There should be a high index of suspicion for patients presenting with disproportionate low back pain and vague symptoms after surgery. Recent oral surgery may increase the risk of bacteremia and subsequent infectious morbidity after sacrocolpopexy with the use of synthetic mesh for prolapse repair.