Inconsistencies in classification by experts of cardiotocograms and subsequent clinical decision

Artificial intelligence based cardiotocogram assessment during labor

OBJECTIVE

To assess whether artificial intelligence, inspired by clinical decision-making procedures in delivery rooms, can correctly interpret cardiotocographic tracings and distinguish between normal and pathological events.

STUDY DESIGN

A method based on artificial intelligence was developed to determine whether a cardiotocogram shows a normal response of the fetal heart rate to uterine activity (UA). For a given fetus and given the UA and previous FHR, the method predicts a fetal heart rate response, under the assumption that the fetus is still in good condition and based on how that specific fetus has responded so far. We hypothesize that this method, when having only learned from fetuses born in good condition, is incapable of predicting the response of a compromised fetus or an episode of transient fetal distress. The (in)capability of the method to predict the fetal heart rate response would then yield a method that can help to assess fetal condition when the obstetrician is in doubt. Cardiotocographic data of 678 deliveries during labor were selected based on a healthy outcome just after birth. The method was trained on the cardiotocographic data of 548 fetuses of this group to learn their heart rate response. Subsequently it was evaluated on 87 fetuses, by assessing whether the method was able to predict their heart rate responses. The remaining 43 cardiotocograms were segment-by-segment annotated by three experienced gynecologists, indicating normal, suspicious, and pathological segments, while having access to the full recording and neonatal outcome. This future knowledge makes the expert annotations of a quality that is unachievable during live interpretation.

RESULTS

The comparison between abnormalities detected by the method (only using past and present input) and the annotated CTG segments by gynecologists (also looking at future input) yields an area under the curve of 0.96 for the distinction between normal and pathological events in majority-voted annotations.

CONCLUSION

The developed method can distinguish between normal and pathological events in near real-time, with a performance close to the agreement between three gynecologists with access to the entire CTG tracing and fetal outcome. The method has a strong potential to support clinicians in assessing fetal condition in clinical practice.

Inter- and intraobserver agreement of antenatal cardiotocography assessments by maternity care professionals: A prospective study

OBJECTIVE

In the Netherlands, antenatal cardiotocography (aCTG) to assess fetal well-being is performed in obstetrician-led care. An innovative initiative was started to evaluate whether aCTG for specific indications-reduced fetal movements, external cephalic version, or postdate pregnancy-is feasible in non-obstetrician-led care settings by independent primary care midwives. Quality assessment is essential when reorganizing and shifting tasks and responsibilities. Therefore, we aimed to assess the inter- and intraobserver agreement for aCTG assessments between and within four professional groups involved in Dutch maternity care regarding the overall classification and assessment of the various components of aCTG.

METHOD

This was a prospective study among 47 Dutch primary care midwives, hospital-based midwives, residents, and obstetricians. Ten aCTG traces were assessed twice at a 1 month interval. To ensure a representative sample, we used two different sets of 10 aCTG traces each. We calculated the degree of agreement using the proportions of agreement.

RESULTS

The proportions of agreement for interobserver agreement on the classification of aCTG between and within the four professional groups varied from 0.82 to 0.94. The proportions of agreement for each professional group were slightly higher for intraobserver (0.86-0.94) than for interobserver agreement. For the various aCTG components, the proportions of agreement for interobserver agreement varied from 0.64 (presence of contractions) to 0.98 (baseline heart frequency).

CONCLUSION

The proportion of agreement levels between and within the maternity care professionals in the classification of aCTG traces among healthy women were comparable. This means that these professional groups are equally well able to classify aCTGs in healthy pregnant women.

Antenatal cardiotocography in dutch primary midwife-led care: Maternal and perinatal outcomes and serious adverse events. A prospective observational cohort study

PROBLEM

It is yet unknown whether shifting antenatal cardiotocography (aCTG) from obstetrician-led to midwife-led care leads to a safe reduction in referrals.

BACKGROUND

ACTG is used to assess fetal well-being. In the Netherlands, the procedure has until now been performed as part of obstetrician-led care. Developments in E-health facilitates the performance of aCTG outside the hospital in midwife-led care, hereby increasing continuity of care.

AIM

To evaluate 1) process outcomes of implementing aCTG for specific indications in primary midwife-led care; 2) maternal and perinatal outcomes of pregnant women receiving aCTG in midwife-led care; 3) serious adverse events (with outcomes, causes, avoidability, and potential prevention strategies) that have occurred during the innovation project 'aCTG in midwife-led care'.

METHODS

Prospective observational cohort study and a case series study of serious adverse events.

FINDINGS

A total of 1584 pregnant women with a specific aCTG indication were included in this cohort study for whom 1795 aCTGs were performed in midwife-led care. 1591 aCTGs(89.7%) were classified as reassuring. Referral to obstetrician-led care occurred for 234 women(13.0%) after an aCTG in midwife-led care of whom 202(86%) were referred back. Severe neonatal morbidity occurred in 27 neonates (1.7%). In the 5736 aCTGs included in the case series study, one case with a serious neonatal outcome was assessed as a serious adverse event attributable to human factors.

DISCUSSION

ACTGs performed in midwife-led care increased continuity of care. In this innovation project, maternal and perinatal outcomes were in the expected range for women in midwife-led care.

Optimizing fetal health prediction: Ensemble modeling with fusion of feature selection and extraction techniques for cardiotocography data

Cardiotocography (CTG) captured the fetal heart rate and the timing of uterine contractions. Throughout pregnancy, CTG intelligent categorization is crucial for monitoring fetal health and preserving proper fetal growth and development. Since CTG provides information on the fetal heartbeat and uterus contractions, which helps determine if the fetus is pathologic or not, obstetricians frequently use it to evaluate a child's physical health during pregnancy. In the past, obstetricians have artificially analyzed CTG data, which is time-consuming and inaccurate. So, developing a fetal health categorization model is crucial as it may help to speed up the diagnosis and treatment and conserve medical resources. The CTG dataset is used in this study. To diagnose the illness, 7 machine learning models are employed, as well as ensemble strategies including voting and stacking classifiers. In order to choose and extract the most significant and critical attributes from the dataset, Feature Selection (FS) techniques like ANOVA and Chi-square, as well as Feature Extraction (FE) strategies like Principal Component Analysis (PCA) and Independent Component Analysis (ICA), are being used. We used the Synthetic Minority Oversampling Technique (SMOTE) approach to balance the dataset because it is unbalanced. In order to forecast the illness, the top 5 models are selected, and these 5 models are used in ensemble methods such as voting and stacking classifiers. The utilization of Stacking Classifiers (SC), which involve Adaboost and Random Forest (RF) as meta-classifiers for disease detection. The performance of the proposed SC with meta-classifier as RF model, which incorporates Chi-square with PCA, outperformed all other state-of-the-art models, achieving scores of 98.79%,98.88%,98.69%,96.32%, and 98.77% for accuracy, precision, recall, specificity, and f1-score respectively.

CTG interpretation templates affect residents' decision making

OBJECTIVE

To study whether a revision of CTG guidelines and educational program influenced the perceived need for intervention by residents in obstetrics and gynecology. A secondary aim was to study the sensitivity and specificity of the classification pathological after classification by residents using two different guidelines in identifying neonates with acidemia.

STUDY DESIGN

Cardiotocograms, CTGs, from 223 neonates with acidemia at birth (cord blood pH < 7.05 at vaginal birth or second stage cesarean, or pH < 7.10 at first stage cesarean) were included, as well as 223 CTGs from neonates with cord blood pH ≥ 7.15. Two separate groups of residents, who each were educated in and had clinical experience only from either of the two different guidelines, SWE09 and SWE17, classified the patterns according to the at the time current template and judged whether the patterns indicated an intervention. Sensitivity, specificity, and agreement were calculated.

RESULTS

Residents using SWE09 found indication to intervene in a higher proportion of neonates with acidemia (84.8%) than residents using SWE17 (75.8%; p = 0.002), as well as in cases without acidemia (29.6% vs 22.4%; p = 0.038). Among residents using SWE09 the perceived need for intervention had a sensitivity of 85% and a specificity of 70% to identify acidemia. With SWE17 the corresponding rates were 76% and 78%. The sensitivity to identify neonates with acidemia by classification pathological was 91% with SWE09 and 72% with SWE17. The specificity was 53% and 76% respectively. The agreement rate between perception of indication to intervene and classification pathological using the SWE09 was κ 0.73, moderate, and with the SWE17 κ 0.77, moderate. The agreement on subjective perception of necessity to intervene between users of the two templates was weak to moderate, κ 0.60, and on classification pathological weak, κ 0.47.

CONCLUSION

The perceived need for intervention by residents interpreting CTGs was significantly affected by the guidelines in use. The difference in decisions were less pronounced than the difference in classification. The sensitivity for both perceived need for intervention and for classification pathological to identify acidosis was higher with SWE09, and the specificity higher with SWE17, when assessed by the two comparable groups of residents.

Ambulatory antenatal fetal electrocardiography in high-risk pregnancies (AMBER): protocol for a pilot prospective cohort study

INTRODUCTION

Fetal heart rate (FHR) monitoring is a vital aspect of fetal well-being assessment, and the current method of computerised cardiotocography (cCTG) is limited to the hospital setting. Non-invasive fetal ECG (NIFECG) has the ability to produce FHR patterns through R wave detection while eliminating confusion with maternal heart rate, but is presently limited to research use. Femom is a novel wireless NIFECG device that is designed to be placed without professional assistance, while connecting to mobile applications. It has the ability to achieve home FHR monitoring thereby allowing more frequent monitoring, earlier detection of deterioration, while reducing hospital attendances. This study aims to assess the feasibility, reliability, and accuracy of femom (NIFECG) by comparing its outputs to cCTG monitoring.

METHODS AND ANALYSIS

This is a single-centred, prospective pilot study, taking place in a tertiary maternity unit. Women with a singleton pregnancy over 28⁺⁰ weeks' gestation who require antenatal cCTG monitoring for any clinical indication are eligible for recruitment. Concurrent NIFECG and cCTG monitoring will take place for up to 60 min. NIFECG signals will be postprocessed to produce FHR outputs such as baseline FHR and short-term variation (STV). Signal acceptance criteria is set as <50% of signal loss for the trace duration. Correlation, precision and accuracy studies will be performed to compare the STV and baseline FHR values produced by both devices. The impact of maternal and fetal characteristics on the effectiveness of both devices will be investigated. Other non-invasive electrophysiological assessment parameters will be assessed for its correlation with the STV, ultrasound assessments and maternal and fetal risk factors.

ETHICS AND DISSEMINATION

Approval has been obtained from South-East Scotland Research Ethics Committee 02 and MHRA. The results of this study will be published in peer-reviewed journals, and presented at international conferences.

TRIAL REGISTRATION NUMBER

NCT04941534.

Hazard and cumulative incidence of umbilical cord metabolic acidemia at birth in fetuses experiencing the second stage of labor and pathologic intrapartum fetal heart rate requiring expedited delivery

PURPOSE

The aim of the study was to determine the cause-specific hazard (CSH) and the cumulative incidence function (CIF) for umbilical cord metabolic acidemia at birth (MA; pH < 7.0 and/or BE [Formula: see text] - 12 mmol/L) at delivery in patients experiencing the 2nd stage of labor (2STG), stratified for both FIGO-2015 pathologic intrapartum cardiotocography requiring expedited delivery (CTG_RED) and duration of 2nd stage of labor.

METHODS

3459 pregnancies experiencing the 2nd stage of labor and delivering at the Division of Obstetrics and Prenatal Medicine, IRCCS Sant'Orsola-Malpighi Hospital, Bologna (Italy), were identified between 2018 and 2019. Survival analysis was used to assess CSH and CIF for MA, stratified for FIGO-2015 pathologic CTG and relevant covariates.

RESULTS

FIGO-2015 pathological CTG with expedited operative delivery or urgent cesarean section within 10 or 20 min from diagnosis, respectively occurred in 282/3459 (8.20%). The rate of MA at delivery was 3.32% (115/3459). The spline of CSH for MA showed a direct correlation with the duration of 2STG always presenting higher values and greater slope in the presence of pathologic CTG, with plateau between 60 and 120 min and rapid increase after 120 min. The CIF at 180 min in the 2STG was 2.67% for nonpathological and 10.63% for pathological CTG_RED. Nulliparity, pathological CTG, and meconium-stained amniotic fluid resulted significant predictors of MA in our multivariable model.

CONCLUSION

The risk for MA increases moderately across the 2STG with nonpathological CTG and quadruples with pathological CTG_RED. Adjustment for other predictors of MA including meconium-stained amniotic fluid and nulliparity reveals a significant hazard increase for MA associated with pathologic CTG_RED.

Clinical practice guidelines for intrapartum cardiotocography interpretation: A systematic review

BACKGROUND

Clinical practice guidelines on intrapartum cardiotocography (CTG) interpretation provide structured tools to detect fetal hypoxia. Despite frequent use of different guidelines, little is known about their comparable consistency. We sought to appraise guidelines relevant to intrapartum CTG interpretation and summarise consensus and non-consensus recommendations.

AIMS

To compare existing intrapartum CTG interpretation guidelines.

MATERIALS AND METHODS

We searched PubMed, CINAHL, Cochrane, Embase, guideline databases and websites of guideline development institutions using terms 'cardiotocography', 'electronic fetal/foetal monitoring', and 'guideline' or equivalent term. The search was restricted to English-language articles published between January 1980 and January 2023 and excluded animal studies. The initial search yielded 2128 articles with 1253 unique citations. Guidelines were included if they: used English as the reporting language; included CTG interpretation criteria or guidelines as a primary objective; were published or updated since 1980; and were the most recently updated publications when multiple versions were identified.

RESULTS

Nineteen studies were considered for full review, and 13 met inclusion criteria. Two reviewers independently assessed guideline quality using the AGREE II instrument, and synthesised consensus and non-consensus recommendations using the content analysis approach. Most guidelines employed a three-tier interpretation framework. There were significant differences between the guidelines for relative importance of key CTG features such as accelerations, decelerations and variability, with respect to the outcome of fetal hypoxia.

CONCLUSIONS

There are significant differences between key intrapartum CTG interpretation guidelines currently being used. Greater consistency is needed across CTG interpretation guidelines to improve the quality of data, clinical governance, monitoring of outcomes, and to support future developments.

Comparison of Four Intrapartum Cardiotocography Classifications for Predicting Neonatal Acidemia at Birth

Objective

To compare diagnostic values of four intrapartum cardiotocography (CTG) classifications in predicting neonatal acidemia at birth.

Methods

Retrospective case-control study. Forty-three CTG traces with an umbilical artery pH < 7.00 (study group) and 43 traces with a pH ≥ 7.00 (control group) were analyzed. Inclusion criteria were singleton pregnancy, cephalic presentation, admission to labour ward during active phase of first stage of labour, and gestational age 37⁺⁰ to 41⁺⁶ weeks. Exclusion criteria were suspected intrauterine growth restriction, oligohydramnios, polyhydramnios, pregestational or gestational insulin-dependent diabetes mellitus, and preeclampsia. Last 30-60 minutes before delivery of CTG traces was classified retrospectively according to four classification systems-International Federation of Gynecology and Obstetrics (FIGO), Royal College of Obstetricians and Gynaecologists (RCOG), National Institute of Child Health and Human Development (NICHD), and the 5-tier system by Parer and Ikeda. Predictive value of each classification for neonatal acidemia was assessed using receiver operating characteristics (ROC) analysis.

Results

FIGO, RCOG, and NICHD classifications predicted neonatal acidemia with areas under the ROC curves (AUC) of 0.73, 95% confidence interval (CI) 0.63-0.84; 0.72, 95% CI 0.60-0.83; and 0.69, 95% CI 0.57-0.80, respectively. The five-tier system by Parer and Ikeda had significantly better predictive value with an AUC of 0.96, 95% CI 0.91-1.00.

Conclusions

The 5-tier classification system proposed by Parer and Ikeda for assessing CTG in labour was superior to FIGO, RCOG, and NICHD intrapartum CTG classifications in predicting severe neonatal acidemia at birth.

Computerized analysis of cardiotocograms in clinical practice and the SisPorto® system thirty-two years after: technological, physiopathological and clinical studies

OBJECTIVES

The objective of this study is to present the why, what and how about computerized analysis of cardiotocograms (cCTG) and the SisPorto system for cCTG.

CONTENT

A narrative review about cCTG and the SisPorto system for cCTG is presented. The meta-analysis of randomized controlled trials (RCT) performed so far have evidenced that cCGT compared to traditional CTG analysis may save time spent in hospital for women, in the antepartum period, and is objective with at least equivalent results in maternal and perinatal outcomes, both in the ante and intrapartum periods. The SisPorto system for cCTG closely follows the FIGO guidelines for fetal monitoring. It may be used both in the ante and intrapartum periods, alone or connected to a central monitoring station, with simultaneous monitoring of fetal and maternal signals, not only in singletons but also in twins. It has been assessed in technical, physiopathological and clinical studies, namely in one large multicentric international RCT during labor and two meta-analysis.

SUMMARY AND OUTLOOK

There is evidence that cCTG may be useful in clinical practice with advantages compared to traditional CTG analysis, although without clear impact on the decrease of preventable maternal and perinatal mortality and morbidity. More studies are warranted, namely on technical improvements and assessment in larger studies in a wider range of clinical settings.

Approaches to Preventing Intrapartum Fetal Injury

Electronic fetal monitoring (EFM) was introduced into obstetric practice in 1970 as a test to identify early deterioration of fetal acid-base balance in the expectation that prompt intervention ("rescue") would reduce neonatal morbidity and mortality. Clinical trials using a variety of visual or computer-based classifications and algorithms for intervention have failed repeatedly to demonstrate improved immediate or long-term outcomes with this technique, which has, however, contributed to an increased rate of operative deliveries (deemed "unnecessary"). In this review, we discuss the limitations of current classifications of FHR patterns and management guidelines based on them. We argue that these clinical and computer-based formulations pay too much attention to the detection of systemic fetal acidosis/hypoxia and too little attention not only to the pathophysiology of FHR patterns but to the provenance of fetal neurological injury and to the relationship of intrapartum injury to the condition of the newborn. Although they do not reliably predict fetal acidosis, FHR patterns, properly interpreted in the context of the clinical circumstances, do reliably identify fetal neurological integrity (behavior) and are a biomarker of fetal neurological injury (separate from asphyxia). They provide insight into the mechanisms and trajectory (evolution) of any hypoxic or ischemic threat to the fetus and have particular promise in signaling preventive measures (1) to enhance the outcome, (2) to reduce the frequency of "abnormal" FHR patterns that require urgent intervention, and (3) to inform the decision to provide neuroprotection to the newborn.

A deep learning mixed-data type approach for the classification of FHR signals

The Cardiotocography (CTG) is a widely diffused monitoring practice, used in Ob-Gyn Clinic to assess the fetal well-being through the analysis of the Fetal Heart Rate (FHR) and the Uterine contraction signals. Due to the complex dynamics regulating the Fetal Heart Rate, a reliable visual interpretation of the signal is almost impossible and results in significant subjective inter and intra-observer variability. Also, the introduction of few parameters obtained from computer analysis did not solve the problem of a robust antenatal diagnosis. Hence, during the last decade, computer aided diagnosis systems, based on artificial intelligence (AI) machine learning techniques have been developed to assist medical decisions. The present work proposes a hybrid approach based on a neural architecture that receives heterogeneous data in input (a set of quantitative parameters and images) for classifying healthy and pathological fetuses. The quantitative regressors, which are known to represent different aspects of the correct development of the fetus, and thus are related to the fetal healthy status, are combined with features implicitly extracted from various representations of the FHR signal (images), in order to improve the classification performance. This is achieved by setting a neural model with two connected branches, consisting respectively of a Multi-Layer Perceptron (MLP) and a Convolutional Neural Network (CNN). The neural architecture was trained on a huge and balanced set of clinical data (14.000 CTG tracings, 7000 healthy and 7000 pathological) recorded during ambulatory non stress tests at the University Hospital Federico II, Napoli, Italy. After hyperparameters tuning and training, the neural network proposed has reached an overall accuracy of 80.1%, which is a promising result, as it has been obtained on a huge dataset.

New Markers for Placental Dysfunction at Term - Potential for More

The remaining placental reserve capacity at term plays a decisive role in the perinatal morbidity of mother and child. Considering advances made in the field of fetal monitoring, the routine examination methods currently used at term or late term may be insufficient to detect subclinical placental dysfunction (PD). The aim of this study is to offer an up-to-date, narrative review of the literature in the context of detecting PD at term using complementary ultrasound markers and biomarkers. Parameters of fetomaternal Doppler ultrasound and fetal cardiac function, as well as (anti-)angiogenic factors in maternal serum are potential PD markers. These may help identify patients that may benefit from an elective, early induction of labor at term, thereby potentially reducing morbidity and mortality. However, their value in terms of the optimal date of delivery must first be determined in randomized controlled trials on a large number of cases.

Non-linear Methods Predominant in Fetal Heart Rate Analysis: A Systematic Review

The analysis of fetal heart rate variability has served as a scientific and diagnostic tool to quantify cardiac activity fluctuations, being good indicators of fetal well-being. Many mathematical analyses were proposed to evaluate fetal heart rate variability. We focused on non-linear analysis based on concepts of chaos, fractality, and complexity: entropies, compression, fractal analysis, and wavelets. These methods have been successfully applied in the signal processing phase and increase knowledge about cardiovascular dynamics in healthy and pathological fetuses. This review summarizes those methods and investigates how non-linear measures are related to each paper's research objectives. Of the 388 articles obtained in the PubMed/Medline database and of the 421 articles in the Web of Science database, 270 articles were included in the review after all exclusion criteria were applied. While approximate entropy is the most used method in classification papers, in signal processing, the most used non-linear method was Daubechies wavelets. The top five primary research objectives covered by the selected papers were detection of signal processing, hypoxia, maturation or gestational age, intrauterine growth restriction, and fetal distress. This review shows that non-linear indices can be used to assess numerous prenatal conditions. However, they are not yet applied in clinical practice due to some critical concerns. Some studies show that the combination of several linear and non-linear indices would be ideal for improving the analysis of the fetus's well-being. Future studies should narrow the research question so a meta-analysis could be performed, probing the indices' performance.

Prelabor and intrapartum Doppler ultrasound to predict fetal compromise

According to current estimates, over 20% of the 4 million neonatal deaths occurring globally every year are related to intrapartum hypoxic complications that happen as a result of uterine contractions against a background of inadequate placental function. Most of such intrapartum complications occur among apparently uncomplicated term pregnancies. Available evidence suggests that current risk-assessment strategies do not adequately identify many of the fetuses vulnerable to periods of intermittent hypoxia that characterize human labor. In this review, we discuss the data available on Doppler ultrasound for the evaluation of placental function before and during labor in appropriately grown fetuses; we also discuss the current strategies for ultrasound-based risk stratification, the physiology of intrapartum compromise, and the potential future treatments to prevent fetal distress in labor and reduce perinatal complications related to birth asphyxia.

Prediction of non-reassuring fetal status and umbilical artery acidosis by the maternal characteristic and ultrasound prior to induction of labor

Objective

To investigate the predictive value of pre-induction digital examination, sonographic measurements and parity for the prediction of non-reassuring fetal status and cord arterial pH < 7.2 prior to the induction of labor (IOL).

Method

This was a prospective observational study, including 384 term pregnancies undergoing IOL. Before the IOL, the Bishop score (BS) by digital examination, sonographic Doppler parameters and the estimated fetal weight (EFW) was assessed. The fetal cord arterial was sampled to measure the pH at delivery. Multivariate logistic regression analysis was performed to identify independent predictors of non-reassuring fetal status and low cord arterial pH.

Results

Forty four cases (11.5%) had non-reassuring fetal status, and 76 cases (19.8%) had fetal cord arterial pH < 7.2. In the non-reassuring fetal status group, the incidence of cord arterial pH < 7.2 was significantly higher than that in the normal fetal heart rate group (χ² = 6.401, p = 0.011). Multivariate analysis indicated that significant independent predictors of non-reassuring fetal status were nulliparity (adjusted odds ratio [AOR]: 3.746, p = 0.003), EFW < 10^th percentile (AOR: 3.764, p = 0.003) and cerebroplacental ratio (CPR) < 10^th centile (AOR:4.755, p < 0.001). In the prediction of non-reassuring fetal status, the performance of the combination of nulliparity and EFW < 10th percentile was improved by the addition of CPR < 10th percentile (AUC: 0.681, (95%CI: 0.636 to 0.742) vs 0.756, (95%CI:0.713 to 0.795)), but the difference was not significant (DeLong test: z = 1.039, p = 0.053).. None of the above variables were predictors of cord arterial pH < 7.2.

Conclusion

The risk of fetal acidosis has increased in cases of non-reassuring fetal status. Nulliparity, small for gestational age and CPR < 10th centile are independent predictors for non-reassuring fetal status in term fetuses, though the addition of CPR < 10th centile could not significantly improve the screening accuracy.

Clinical monitoring of late fetal growth restriction

Late fetal growth restriction (FGR) poses its own challenges in respect of diagnosis, surveillance and delivery timing. Perinatal morbidity is relatively rare, and mortality extremely unusual, but given that late FGR is much more frequent than early FGR, the burden on neonatal services must not be underestimated. Doppler findings are more subtle than in early FGR, and growth rate rather than absolute fetal size may be important in defining the condition. Though umbilical artery Doppler changes form the basis for triggering delivery: reversed end diastolic flow at 32 weeks, absent at 34 weeks and raised PI at 36 weeks, other modalities of monitoring - for example cardiotocography and cerebral Doppler - are important in surveillance and timing follow up of the condition.

An effective digitization method for CTG paper report with binary background grids taken by smartphone

BACKGROUND AND OBJECTIVE

Cardiotocography (CTG) is the most popular prenatal diagnostic examination, which includes continuous monitoring of foetal heart rate (FHR, bpm) and uterine contraction (UC, mmHg) signals. Compared with CTG paper reports, digitized reports have better storage, transmission and retrieval capabilities, in addition to being able to assess foetal health. However, most of the existing digitization methods extract signals from paper reports with colour background grids, and they cannot extract signals completely from paper reports with binary background grids, which are widely used in clinical CTG monitoring. Moreover, the existing digitization algorithms often neglect the image distortion caused by the imaging equipment.

METHODS

To overcome the above drawbacks, a digitization method for CTG paper reports with binary background grids taken by smartphones is proposed in this paper. In the stage of removing the grid background, a region merger based on super-pixels and an improved binary line mask removal are designed. Then, signal extraction is performed separately according to the different states of the image column. Through a projection map used to synchronize the signal, the distortion effect of the mobile phone is removed.

RESULTS

The experimental results show that the average correlation coefficient (ρ) between the recovery signal obtained by the proposed method and the reference signal is 0.9855±0.0108 for FHR and 0.9866 ± 0.1020 for UC, and the root mean square errors (RMSE) of FHR and UC processed by the proposed method are 1.0366 ± 0.4953 and 2.0355 ± 1.0246, and the mean absolute errors (MAE) of FHR and UC processed by the proposed method are 0.8735 ± 0.0684 and 1.4991 ± 0.2837, which are higher than those of the existing digitization methods. Compared with clinical signals, no significant difference is found in the feature of digitization CTG.

CONCLUSION

The proposed digitization method is a promising useful tool to realize the electronization of CTG signal.

Intrapartum Doppler ultrasound: where are we now?

Intrapartum hypoxic events most commonly occur in low-risk pregnancies with appropriately grown fetuses. Continuous intrapartum monitoring by means of cardiotocography has not demonstrated a reduction in the frequency of adverse perinatal outcome but has been linked with an increase in the caesarean section rate, particularly among women considered at low risk. Available data from the literature suggests that abnormalities in the uterine artery Doppler and in the ratio between fetal cerebral and umbilical Doppler (i.e. cerebroplacental ratio [CPR]) are associated with conditions of subclinical placental function occurring in fetuses who have failed to achieve their growth potential regardless of their actual size. In this review we summarize the available evidence on the use of intrapartum Doppler ultrasound for the fetal surveillance during labor and the identification of the fetuses at risk of intrapartum distress.

Systematic Review of Intrapartum Fetal Heart Rate Spectral Analysis and an Application in the Detection of Fetal Acidemia

It is fundamental to diagnose fetal acidemia as early as possible, allowing adequate obstetrical interventions to prevent brain damage or perinatal death. The visual analysis of cardiotocography traces has been complemented by computerized methods in order to overcome some of its limitations in the screening of fetal hypoxia/acidemia. Spectral analysis has been proposed by several studies exploring fetal heart rate recordings while referring to a great variety of frequency bands for integrating the power spectrum. In this paper, the main goal was to systematically review the spectral bands reported in intrapartum fetal heart rate studies and to evaluate their performance in detecting fetal acidemia/hypoxia. A total of 176 articles were reviewed, from MEDLINE, and 26 were included for the extraction of frequency bands and other relevant methodological information. An open-access fetal heart rate database was used, with recordings of the last half an hour of labor of 246 fetuses. Four different umbilical artery pH cutoffs were considered for fetuses' classification into acidemic or non-acidemic: 7.05, 7.10, 7.15, and 7.20. The area under the receiver operating characteristic curve (AUROC) was used to quantify the frequency bands' ability to distinguish acidemic fetuses. Bands referring to low frequencies, mainly associated with neural sympathetic activity, were the best at detecting acidemic fetuses, with the more severe definition (pH ≤ 7.05) attaining the highest values for the AUROC. This study shows that the power spectrum analysis of the fetal heart rate is a simple and powerful tool that may become an adjunctive method to CTG, helping healthcare professionals to accurately identify fetuses at risk of intrapartum hypoxia and to implement timely obstetrical interventions to reduce the incidence of related adverse perinatal outcomes.

Formal training in cardiotocograph interpretation of healthcare practitioners improves interpretation: A prospective descriptive analytical study in a resource constrained setting

OBJECTIVE

To evaluate the effect of structured training on the cardiotocograph (CTG) interpretation skills of healthcare workers (HCWs) involved in obstetric care in a recourse constrained setting.

METHOD

In all, 218 HCWs completed pre- and post-intervention questionnaires. The intervention involved structured teaching on technical aspects of CTG monitoring, fetal physiology, and adaptation to stress, and a holistic approach to interpretation. Descriptive statistics were used to summarize demographic characteristics, frequencies, and percentages for categorical data.

RESULTS

Only 26.2% of participants felt that their pre-graduate training was adequate; this was reflected by the poor baseline knowledge in CTG interpretation (56% misinterpreted) and lack of knowledge regarding technical aspects of CTG monitoring in 48.8% of participants. The training session led to a 65% improvement in technical aspects and interpretation. Those HCWs not receiving continuous training showed a more significant improvement (P  < 0.001) and there was a 100% improvement in knowledge regarding the required documentation before commencing monitoring. Most (99.5%) HCWs acknowledged an improvement in knowledge, and 96.2% would participate in similar mandatory education sessions.

CONCLUSION

These findings highlight the fact that training in CTG monitoring is warranted and desired by HCWs. It also supports the implementation of structured CTG education in resource-constrained settings.

A randomized controlled trial studying the effect of maternal hyperoxygenation on fetal heart rate in suspected fetal distress

OBJECTIVE

To investigate the effect of maternal hyperoxygenation on fetal heart rate (FHR) when applied for suspected fetal distress during the second stage of term labor.

APPROACH

A single-center randomized controlled trial was conducted in a tertiary care hospital in The Netherlands. Participants were included during the second stage of labor in case of an intermediary or abnormal FHR pattern. Patients were randomized to receive either 100% oxygen at 10 l/min until delivery, or conventional care without additional oxygen. The primary outcome was the change in FHR pattern before and after the onset of the study, measured as the change in depth and duration of FHR decelerations. Secondary outcome measures were features based on phase-rectified signal averaging (PRSA), baseline assignability, and deceleration characteristics of the FHR pattern.

MAIN RESULTS

Between March 2016 and April 2018, 117 women were included. The FHR pattern could be analyzed for 71 participants, the other 46 women delivered before the end of the post time-frame. A 2.3% reduction in depth and duration of FHR decelerations was found after maternal hyperoxygenation, compared to a 10% increase in the control group (p = 0.24). Maternal hyperoxygenation had a significantly positive effect on PRSA metrics, with a decrease in PRSA-acceleration capacity (p = 0.03) and PRSA-deceleration capacity (p = 0.02) in the intervention group compared to the control group.

SIGNIFICANCE

The difference in depth and duration of decelerations after the start of the study was not significantly different between both study groups. A statistically significant positive effect on PRSA-deceleration capacity and PRSA-acceleration capacity was found after maternal hyperoxygenation, which might be associated with a positive effect on neonatal outcome.

Intrapartum zigzag pattern of fetal heart rate is an early sign of fetal hypoxia: A large obstetric retrospective cohort study

Introduction

The aim of the present study was to identify possible associations of fetal heart rate (FHR) patterns during the last 2 hours of labor with fetal asphyxia expressed by umbilical artery acidemia at birth and with neonatal complications in a large obstetric cohort.

Material and methods

Cardiotocographic recordings from 4988 singleton term childbirths over 1 year were evaluated retrospectively and blinded to the pregnancy and neonatal outcomes in a university teaching hospital in Helsinki, Finland. Umbilical artery pH, base excess and pO₂, low Apgar scores at 5 minutes, need for intubation and resuscitation, early neonatal hypoglycemia, and neonatal encephalopathy were used as outcome variables. According to the severity of the neonatal complications at birth, the cohort was divided into three groups: no complications (Group 1), moderate complications (Group 2) and severe complications (Group 3).

Results

Of the 4988 deliveries, the ZigZag pattern (FHR baseline amplitude changes of >25 bpm with a duration of 2‐30 minutes) occurred in 11.7%, late decelerations in 41.0%, bradycardia episodes in 52.9%, reduced variability in 36.7%, tachycardia episodes in 13.9% and uterine tachysystole in 4.6%. No case of saltatory pattern (baseline amplitude changes of >25 bpm with a duration of >30 minutes) was observed. The presence of the ZigZag pattern or late decelerations, or both, was associated with cord blood acidemia (odds ratio [OR] 3.3, 95% confidence interval [CI] 2.3‐4.7) and severe neonatal complications (Group 3) (OR 3.3, 95% CI 2.4‐4.9). In contrast, no significant associations existed between the other FHR patterns and severe neonatal complications. ZigZag pattern preceded late decelerations in 88.7% of the cases. A normal FHR preceded the ZigZag pattern in 90.4% of the cases, whereas after ZigZag episodes, a normal FHR pattern was observed in only 0.9%.

Conclusions

ZigZag pattern and late decelerations during the last 2 hours of labor are significantly associated with cord blood acidemia at birth and neonatal complications. The ZigZag pattern precedes late decelerations, and the fact that normal FHR pattern precedes the ZigZag pattern in the majority of the cases suggests that the ZigZag pattern is an early sign of fetal hypoxia, which emphasizes its clinical importance.

Uterine Activity Impacts Fetal and Neonatal Outcomes

Uterine activity impacts fetal and neonatal outcomes. The value of analysis of contraction frequency and the duration of the resting interval were underappreciated until the last two decades. Misconceptions about electronic fetal monitoring and the cesarean section rate may be related to the lack of early research on the significant impact of abnormal uterine activity.

Prelabor screening at term using the cerebroplacental ratio and placental growth factor: a pragmatic randomized open-label phase 2 trial

BACKGROUND AND OBJECTIVE

In some women placental function may not be adequate to meet fetal growth requirements in late pregnancy or the additional demands during labor, thus predisposing these infants to intrapartum fetal compromise and subsequent serious morbidity and mortality. The objective of this study was to determine if the introduction of a prelabor screening test at term combining the cerebroplacental ratio and maternal placental growth factor level would result in a reduction in a composite of adverse outcomes.

STUDY DESIGN

Single-site, nonblinded, randomized controlled trial conducted at a tertiary hospital in Brisbane, Australia. Eligible women were randomized to either receive the screening test performed between 37-38 weeks or routine obstetric care. Screen-positive women were offered induction of labor. The primary outcome was a composite of emergency cesarean delivery for nonreassuring fetal status (fetal distress) or severe neonatal acidosis or low Apgar score or stillbirth or neonatal death.

RESULTS

Women were recruited and randomized (n = 501) between April 2017 and January 2019. Sixty-three of 249 subjects (25.3%) in the screened group compared to 56 of 252 (22.2%) in the control group experienced the primary outcome (relative risk = 1.14 [95% confidence interval, 0.83-1.56]; P = .418). Women who screened positive were more likely to require operative delivery for fetal distress, have meconium-stained liquor, have pathologic fetal heart rate abnormalities, and have infants with lower birthweight compared to women that screened negative.

CONCLUSION

The introduction of this test did not result in improvements in intrapartum intervention rates or neonatal outcomes. However, it did show discriminatory potential, and future research should focus on refining the thresholds used.

Safety and efficacy of sildenafil citrate to reduce operative birth for intrapartum fetal compromise at term: a phase 2 randomized controlled trial

BACKGROUND AND OBJECTIVE

Sildenafil citrate is a vasodilator used in erectile dysfunction and pulmonary hypertension. We tested whether it reduces emergency operative births for fetal compromise and improves fetal or uteroplacental perfusion in labor in a phase 2 double-blind randomized controlled trial.

STUDY DESIGN

Women at term in early labor or undergoing scheduled induction of labor at Mater Mother's Hospital, Brisbane, Australia, were randomly allocated 50 mg of sildenafil citrate orally 8 hourly up to 150 mg or placebo. Intrapartum fetal monitoring followed Royal Australian and New Zealand College of Obstetricians and Gynaecologists guidelines. Primary outcomes were (1) emergency operative birth (by cesarean delivery or instrumental vaginal birth) for intrapartum fetal compromise and (2) mean indices of fetal and uteroplacental perfusion using Doppler ultrasound. Analysis was by intention-to-treat.

TRIAL REGISTRATION NUMBER

ANZCTRN12615000319572 RESULTS: Between September 2015 and January 2019, 300 women were randomized equally to sildenafil citrate or placebo. Sildenafil citrate reduced the risk of emergency operative birth by 51% (18% vs 36.7%; relative risk, 0.49, 95% confidence interval, 0.33-0.73, P=.0004, number needed to treat = 5 [3-11]). There was no difference in indices of fetal and uteroplacental perfusion, but these were ascertained in only 71 women. Sildenafil citrate reduced the risk of meconium-stained liquor or pathologic fetal heart rate patterns by 43% (25.3% vs 44.7%; relative risk, 0.57, 95% confidence interval, 0.41-0.79, P=.0005), but its effects on fetal scalp sampling rates (2.0% vs 6.7%; relative risk, 0.30, 95% confidence interval, 0.08-1.07, P=.06) and adverse neonatal outcome (20.7% vs 21.3%; relative risk, 0.97, 95% confidence interval, 0.62-1.50, P=.89) were inconclusive. Only 3.6% of maternal levels of sildenafil citrate or its metabolite were detected in cord blood. No differences in maternal adverse events were seen.

CONCLUSION

Sildenafil citrate reduced operative birth for intrapartum fetal compromise, but much larger phase 3 trials of its effects on mother and child are needed before it can be routinely recommended.

Near-infrared spectroscopy of the placenta for monitoring fetal oxygenation during labour

Although being the golden standard for intrapartum fetal surveillance, cardiotocography (CTG) has been shown to have poor specificity for detecting fetal acidosis. Non-invasive near-infrared-spectroscopy (NIRS) monitoring of placental oxygenation during labour has not been studied yet. The objective of the study was to determine whether changes in placental NIRS values during labour could identify intrapartum fetal hypoxia and resulting acidosis. We included 43 healthy women in active stage of labour at term. CTG and NIRS parameters in groups with vs. without neonatal umbilical artery pH ≤ 7.20 were compared using Mann-Whitney-U. Receiver-operating-characteristics (ROC) curves were used to estimate predictive value of CTG and NIRS parameters for neonatal pH ≤ 7.20. A computer-based statistical classification was also performed to further evaluate predictive values of CTG and NIRS for neonatal acidosis. Ten (23%) neonates were born with umbilical artery pH ≤ 7.20. Compared to group with pH > 7.20, fetal acidosis was associated with more episodes of placental NIRS deoxygenation (9 (range 2-37) vs. 2 (range 0-65); p<0.001), higher velocity of placental NIRS deoxygenation (2.31 (range 0-22) vs. 1 (range 0-49) %/s; p = 0.03), more decelerations on CTG (25 (range 3-91) vs. 10 (range 10-60); p = 0.02), and more prolonged decelerations on CTG (2 (range 0-4) vs. 1 (range 0-3); p = 0.04). Number of placental deoxygenations had the highest prognostic value for fetal/neonatal acidosis (area under the ROC curve 0.85 (95% confidence interval 0.70-0.99). Computer-based classification also identified number of placental deoxygenations as the most accurate classifier, with 25% false positive and 93% true positive rate in the training dataset, with 100% accuracy when applied to the testing dataset. Placental deoxygenations during labour measured by NIRS are associated with fetal/neonatal acidosis. Predictive value of placental NIRS for neonatal acidosis was superior to that of CTG.

Complexity of Cardiotocographic Signals as A Predictor of Labor

Prediction of labor is of extreme importance in obstetric care to allow for preventive measures, assuring that both baby and mother have the best possible care. In this work, the authors studied how important nonlinear parameters (entropy and compression) can be as labor predictors. Linear features retrieved from the SisPorto system for cardiotocogram analysis and nonlinear measures were used to predict labor in a dataset of 1072 antepartum tracings, at between 30 and 35 weeks of gestation. Two groups were defined: Group A—fetuses whose traces date was less than one or two weeks before labor, and Group B—fetuses whose traces date was at least one or two weeks before labor. Results suggest that, compared with linear features such as decelerations and variability indices, compression improves labor prediction both within one (C-Statistics of 0.728) and two weeks (C-Statistics of 0.704). Moreover, the correlation between compression and long-term variability was significantly different in groups A and B, denoting that compression and heart rate variability look at different information associated with whether the fetus is closer to or further from labor onset. Nonlinear measures, compression in particular, may be useful in improving labor prediction as a complement to other fetal heart rate features.

Saltatory Pattern of Fetal Heart Rate during Labor Is a Sign of Fetal Hypoxia

BACKGROUND

While late decelerations and major bradycardia episodes in intrapartum cardiotocography (CTG) recordings are known to correlate with fetal distress,little is known of the importance of the saltatory pattern.

OBJECTIVE

The aim of the study was to examine whether the fetal heart rate (FHR) saltatory pattern in intrapartum CTG registration is associated with fetal hypoxia during the last 2 h of labor.

DESIGN

The study group consisted of CTG recordings from 194 births with a 1-min Apgar score of <8 (birth weight 3,614 ± 512 g; gestational age 40.6 ± 0.7 weeks). The comparison group included 51 infants with a 1-min Apgar score of ≥9 (birth weight 3,624 ± 400 g; gestational age 40.5 ± 0.4 weeks). FHR patterns were evaluated blindly by 2 experienced perinatologists. The pH, base excess (BE), pO2 and erythropoietin (EPO) were measured from umbilical cord blood at birth as outcome variables.

RESULTS

Saltatory pattern occurred in 31/194 (16.0%) of the study group and in 1/51 (2.0%) of the comparison group. Umbilical artery pH, BE, and pO2 were lower and umbilical vein (UV) EPO higher in the study group than in the comparison group. In the study group, UV EPO level was significantly higher in cases where the saltatory pattern was present (median 241 mU/mL, 95% CI 39.4-16,484), than in those without the saltatory pattern (median 39.4 mU/mL, 95% CI 11-282) (p < 0.0001, for difference). In the study group, no differences in EPO levels were found in cases where episodes of bradycardia, tachycardia, reduced variability, or uterine tachysystole were present or absent. In the study group, saltatory pattern preceded late decelerations in 82.8%.

CONCLUSION

Saltatory pattern in an intrapartum FHR recording is an early sign of fetal hypoxia.

The physiology of intrapartum fetal compromise at term

Uterine contractions in labor result in a 60% reduction in uteroplacental perfusion, causing transient fetal and placental hypoxia. A healthy term fetus with a normally developed placenta is able to accommodate this transient hypoxia by activation of the peripheral chemoreflex, resulting in a reduction in oxygen consumption and a centralization of oxygenated blood to critical organs, namely the heart, brain, and adrenals. Providing there is adequate time for placental and fetal reperfusion between contractions, these fetuses will be able to withstand prolonged periods of intermittent hypoxia and avoid severe hypoxic injury. However, there exists a cohort of fetuses in whom abnormal placental development in the first half of pregnancy results in failure of endovascular invasion of the spiral arteries by the cytotrophoblastic cells and inadequate placental angiogenesis. This produces a high-resistance, low-flow circulation predisposing to hypoperfusion, hypoxia, reperfusion injury, and oxidative stress within the placenta. Furthermore, this renders the placenta susceptible to fluctuations and reduction in uteroplacental perfusion in response to external compression and stimuli (as occurs in labor), further reducing fetal capillary perfusion, placing the fetus at risk of inadequate gas/nutrient exchange. This placental dysfunction predisposes the fetus to intrapartum fetal compromise. In the absence of a rare catastrophic event, intrapartum fetal compromise occurs as a gradual process when there is an inability of the fetal heart to respond to the peripheral chemoreflex to maintain cardiac output. This may arise as a consequence of placental dysfunction reducing pre-labor myocardial glycogen stores necessary for anaerobic metabolism or due to an inadequate placental perfusion between contractions to restore fetal oxygen and nutrient exchange. If the hypoxic insult is severe enough and long enough, profound multiorgan injury and even death may occur. This review provides a detailed synopsis of the events that can result in placental dysfunction, how this may predispose to intrapartum fetal hypoxia, and what protective mechanisms are in place to avoid hypoxic injury.

Use of Machine Learning Algorithms for Prediction of Fetal Risk using Cardiotocographic Data

Background

A major contributor to under-five mortality is the death of children in the 1^st month of life. Intrapartum complications are one of the major causes of perinatal mortality. Fetal cardiotocograph (CTGs) can be used as a monitoring tool to identify high-risk women during labor.

Aim

The objective of this study was to study the precision of machine learning algorithm techniques on CTG data in identifying high-risk fetuses.

Methods

CTG data of 2126 pregnant women were obtained from the University of California Irvine Machine Learning Repository. Ten different machine learning classification models were trained using CTG data. Sensitivity, precision, and F1 score for each class and overall accuracy of each model were obtained to predict normal, suspect, and pathological fetal states. Model with best performance on specified metrics was then identified.

Results

Determined by obstetricians' interpretation of CTGs as gold standard, 70% of them were normal, 20% were suspect, and 10% had a pathological fetal state. On training data, the classification models generated by XGBoost, decision tree, and random forest had high precision (>96%) to predict the suspect and pathological state of the fetus based on the CTG tracings. However, on testing data, XGBoost model had the highest precision to predict a pathological fetal state (>92%).

Conclusion

The classification model developed using XGBoost technique had the highest prediction accuracy for an adverse fetal outcome. Lay health-care workers in low- and middle-income countries can use this model to triage pregnant women in remote areas for early referral and further management.

Significant reduction in umbilical artery metabolic acidosis after implementation of intrapartum ST waveform analysis of the fetal electrocardiogram

BACKGROUND

Although the evidence regarding the benefit of using ST waveform analysis of the fetal electrocardiogram is conflicting, ST waveform analysis is considered as adjunct to identify fetuses at risk for asphyxia in our center. Most randomized controlled trials and meta-analyses have not shown a significant decrease in umbilical metabolic acidosis, while some observational studies have shown a gradual decrease of this outcome over a longer period of time. Observational studies can give more insight into the effect of implementation of the ST technology in daily clinical practice.

OBJECTIVE

To evaluate the change in frequency of perinatal intervention and adverse neonatal outcome after the implementation of ST waveform analysis of the fetal electrocardiogram from 2000 to 2013.

STUDY DESIGN

This retrospective longitudinal study was conducted in a tertiary referral center. A total of 19,664 medium- and high-risk singleton pregnancies with fetuses in cephalic presentation, a gestational age of ≥36 weeks, and the intention to deliver vaginally were included. ST waveform analysis of the fetal electrocardiogram was implemented in the year 2000 and by 2010 all deliveries were monitored using this technology. Data were collected on the following perinatal outcomes: fetal blood sampling, mode of delivery, umbilical cord blood gases, Apgar scores, neonatal encephalopathy, and perinatal death. Longitudinal trend analysis was used to detect changes over time in all deliveries monitored by cardiotocography either alone or in adjunct to ST waveform analysis of the fetal electrocardiogram. Logistic regression was used to correct for possible confounders.

RESULTS

The umbilical artery metabolic acidosis rate declined from 2.5% (average rate of 2000 + 2001 + 2002) to 0.4% (average of 2011 + 2012 + 2013) (P < .001), which represents an 84% decrease. This decrease largely occurred between 2006 and 2008, during the Dutch randomized trial on fetal electrocardiogram ST waveform analysis. At this time, approximately 20% of deliveries were monitored using this method. Furthermore, there were significant reductions in fetal blood sampling rate (P < .001). Overall cesarean and vaginal instrumental deliveries decreased significantly (P < .001), but not for fetal distress. There were no changes in the Apgar scores. The incidence of neonatal encephalopathy was significantly lower in the second part of the study (odds ratio 0.39, 95% confidence interval 0.17-0.89).

CONCLUSION

There was an 84% decrease in the incidence of umbilical artery metabolic acidosis in all deliveries between 2000 and 2013. The neonatal encephalopathy rate, fetal blood sampling rate, and the total number of cesarean and vaginal instrumental deliveries also decreased.

Intrapartum cardiotocogram monitoring between obstetricians and computer analysis

Purpose: To investigate the accuracy of computer analysis and its features to be used as a fu fetal heart rate (FHR) interpretation method in clinical settings.Methods: The Trium CTG Online® was used as the computer analysis software. Twenty-six cases of intrapartum FHR tracings (total time, 6900 min) were randomly selected from third-trimester pregnancies. Three obstetricians blinded to the patients' clinical information traced the decelerations, variability, and baseline cardiotocogram (CTG) data. Three obstetrician observer individually interpreted the data and only the waveforms they interpreted were adopted. The agreement between the deceleration and baseline, variability, and level of five-tier system was estimated. Weighted kappa (κ) statistics were used to assess reliability.Results: Based on the observers and Trium's classification, κ was 0.78 and the strength of agreement level was substantial. The obstetricians and Trium mostly agreed on the variability and baseline data. However, κ of each deceleration was approximately 0.65 (0.63-0.66), with substantial strength of agreement.Conclusion: Based on the obstetricians and Trium's interpretation, the latter was found to be excellent for FHR interpretation. However, it was difficult for Trium to interpret specific waveform patterns. Therefore, clinical staff should understand these characteristics to more sensitively evaluate the fetal well-being.

On the prediction of foetal acidaemia: A spectral analysis-based approach

A computational analysis of physiological systems has been used to support the understanding of how these systems work, and in the case of foetal heart rate, many different approaches have been developed in the last decades. Our objective was to apply a new method of classification, which is based on spectral analysis, in foetal heart rate (FHR) traces to predict foetal acidosis diagnosed with umbilical arterial blood pH ≤ 7.05. Fast Fourier transform was applied to a real database for the classification approach. To evaluate the models, sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve were used. Sensitivity equal to 1, specificity equal to 0.85 and an area under the ROC curve of 0.94 were found. In addition, when the definition of metabolic acidosis of umbilical arterial blood pH ≤ 7.05 and base excess ≤ -10 mmol/L was used, the proposed methodology obtained sensitivity = 1, specificity = 0.97 and area under the ROC curve = 0.98. The proposed methodology relies exclusively on the spectral frequency decomposition of the FHR signal. After further successful validation in more datasets, this approach can be incorporated easily in clinical practice due to its simple implementation. Likewise, the incorporation of this novel technique in an intrapartum monitoring station should be straightforward, thus enabling the assistance of labour professionals in the anticipated detection of acidaemia.

Relative versus absolute rises in T/QRS ratio by ST analysis of fetal electrocardiograms in labour: A case-control pilot study

INTRODUCTION

The additional value of ST analysis during labour is uncertain. In ST analysis, a T/QRS baseline value is calculated from the fetal electrocardiogram and successive T/QRS ratios are compared to this baseline. However, variation in the orientation of the electrical heart axis between fetuses may yield different T/QRS baseline values. In case of a higher T/QRS baseline value more ST events are encountered, although not always related to perinatal outcome. We hypothesised that we can partly correct for this effect by analysing T/QRS rises as a percentage from baseline (relative ST analysis). This study aimed to explore whether relative ST analysis has better diagnostic value for cord acidaemia compared to conventional ST analysis, where predefined fixed T/QRS ratios are used.

METHODS AND MATERIALS

A case-control study was performed in 20 term human fetuses during labour; 10 cases (umbilical cord artery pH <7.05 at birth, defining acidaemia) and 10 controls (pH >7.20) were included. The fetal electrocardiogram was recorded using a STAN monitor. We electronically extracted all T/QRS values, baseline and episodic ST events from the STAN monitor and calculated the relative T/QRS changes. The cut-off for relative ST events was determined in a receiver operator characteristic (ROC) curve at optimal specificity for cord acidaemia. Parameters of interest were area under the curve (AUC) of the ROC curve for relative ST events and test performance of both conventional and relative ST analysis.

RESULTS

Relative ST analysis showed an AUC of 0.99. The optimal cut-off value for relative T/QRS rise was determined at 0.70. Relative vs conventional (absolute) ST analysis showed a specificity of 100% vs 40% (p = 0.031); sensitivity 90% vs 90%; positive likelihood ratio infinity vs 1.5; negative likelihood ratio 0.10 vs 0.25, respectively.

CONCLUSION

Relative ST analysis seems to be a promising method to detect impending fetal acidaemia during labour. Further studies are required to determine the diagnostic accuracy.

Computerized analysis of cardiotocograms and ST signals is associated with significant reductions in hypoxic-ischemic encephalopathy and cesarean delivery: an observational study in 38,466 deliveries

BACKGROUND

Intrapartum cardiotocography is widely used in high-resource countries and remains at the center of fetal monitoring and the decision to intervene, but there is ample evidence of poor reliability in visual interpretation as well as limited accuracy in identifying fetal hypoxia. Combined monitoring of intrapartum cardiotocography and ST segment signals was developed to increase specificity, but analysis relies heavily on intrapartum cardiotocography interpretation and is therefore also affected by the previously referred problems. Computerized analysis was developed to overcome these limitations, aiding in the quantification of parameters that are difficult to evaluate visually, such as variability, integrating the complex guidelines of combined intrapartum cardiotocography and ST analysis, and using visual and sound alerts to prompt health care professionals to reevaluate features associated with fetal hypoxia.

OBJECTIVE

The objective of the study was to evaluate the effect of introducing a central fetal monitoring system with computerized analysis of intrapartum cardiotocography and ST signals into the labor ward of a tertiary care university hospital in which all women are continuously monitored with intrapartum cardiotocography. The incidence of adverse perinatal outcomes and intervention rates was evaluated over time.

STUDY DESIGN

In this retrospective cohort study, yearly rates of hypoxic-ischemic encephalopathy, instrumental vaginal delivery, overall cesarean delivery, and urgent cesarean delivery were obtained from the hospital's clinical databases. The rates occurring in the period from January 2001 to December 2003, before the introduction of the central monitoring system with computerized analysis of intrapartum cardiotocography and ST signals (Omniview-SisPorto), were compared with those occurring from January 2004 to December 2014, after the introduction of the system. All rates were calculated with 95% confidence intervals.

RESULTS

A total of 38,466 deliveries occurred during this period. After introduction of the system, there was a significant decrease in the number of hypoxic-ischemic encephalopathy cases per 1000 births (5.3%, 95% confidence interval [4.0-7.0] vs 2.2%, 95% confidence interval [1.7-2.8]; relative risk, 0.42, 95% confidence interval [0.29-0.61]), overall cesarean delivery rates (29.9%, 95% confidence interval [28.9-30.8] vs 28.3%, 95% confidence interval [27.8-28.8]; relative risk, 0.96, 95% confidence interval [0.92-0.99]), and urgent cesarean deliveries (21.6%, 95% confidence interval [20.7-22.4] vs 19.2%, 95% confidence interval [18.8-19.7]; relative risk, 0.91, 95% confidence interval [0.87-0.95]). The instrumental vaginal delivery rate increased (19.5%, 95% confidence interval [18.7-20.3] vs 21.4%, 95% confidence interval [21.0-21.9; relative risk, 1.07, 95% confidence interval 1.02-1.13].

CONCLUSION

Introduction of computerized analysis of intrapartum cardiotocography and ST signals in a tertiary care hospital was associated with a significant reduction in the incidence of hypoxic-ischemic encephalopathy and a modest reduction in cesarean deliveries.

Maternal hemoglobin level and its relation to fetal distress, mode of delivery, and short-term neonatal outcome: a retrospective cohort study

Aim: We aimed to investigate if the risk of fetal distress during term labor is related to the intrapartum maternal hemoglobin (Hb) level. Second, we investigated the relation between mode of delivery, reason for instrumental delivery and short-term neonatal outcome and maternal Hb. Third, we aimed to identify factors influencing intrapartum maternal Hb level.Methods: A retrospective cohort study was performed in a tertiary hospital in The Netherlands, including data from women who gave birth between 2009 and 2016. To determine whether the likelihood of fetal distress to occur was dependent on intrapartum Hb, multivariate regression models were run with intrapartum Hb as the main independent variable of interest. Hb was used as a continuous value. We repeated this procedure for the likelihood of instrumental vaginal delivery (IVD), cesarean section (CS), 5-min Apgar score < 7, and umbilical cord arterial pH ≤ 7.05 to occur. Also, we identified factors influencing intrapartum Hb level using linear regression analysis.Results: Data of 9144 patients were analyzed. Intrapartum Hb did not contribute to the prediction of the likelihood of fetal distress, IVD for nonprogressive labor, CS for fetal condition, 5-min Apgar score < 7, and pHa ≤ 7.05. However, there was a unique statistically significant contribution of Hb to the prediction of the likelihood of IVD for any reason and IVD for fetal distress and CS for any reason and CS for nonprogressive labor. IVD for fetal distress was related to a higher intrapartum Hb level, whereas CS for nonprogressive labor was related to a lower intrapartum Hb level. Intrapartum Hb level was influenced by maternal age, ethnicity, parity, fetal sex, and birth weight.Conclusions: The risk of fetal distress and adverse neonatal outcome is not related to intrapartum Hb levels. However, our data suggest that mode of delivery is dependent on intrapartum Hb, as shown in a large tertiary population. We recommend further investigating this relation in a large prospective study.

Prelabor screening for intrapartum fetal compromise in low-risk pregnancies at term: cerebroplacental ratio and placental growth factor

OBJECTIVES

To determine the screening performance of low fetal cerebroplacental ratio (CPR), a marker of fetal adaptation to suboptimal growth, and maternal placental growth factor (PlGF) level, both in isolation and in combination, for the prediction of Cesarean section (CS) for intrapartum fetal compromise (IFC) and composite adverse neonatal outcome (CANO).

METHODS

This was a prospective cohort study in low-risk women with uncomplicated singleton pregnancy from 36 weeks' gestation to delivery. CPR and PlGF were assessed fortnightly and intrapartum and neonatal outcomes were recorded. CPR and PlGF values from the final assessment for each woman were corrected for gestational age and assessed for screening performance, firstly as continuous variables and then as binary predictors.

RESULTS

Of the 264 women who consented to participate in the study, 207 were included in the final analysis. Seven pregnancies required CS for IFC and 38 had CANO. Pregnancies delivered by CS for IFC had lower CPR and PlGF centiles than those in all other pregnancies. Pregnancies with CANO had a lower PlGF centile. The greatest areas under the receiver-operating characteristics curves (AUCs) for the prediction of CS for IFC (0.92; 95% CI, 0.86-0.97) and CANO (0.64; 95% CI, 0.54-0.74) were achieved by a combination of CPR 20^th and PlGF 33^rd centile thresholds. This produced sensitivities, specificities and positive likelihood ratios for the prediction of CS for IFC of 100%, 86% and 7.14, respectively, and 34.2%, 87.0% and 2.63, respectively, for the prediction of CANO. There was no statistical difference in the AUC for CS for IFC between the combined model and when CPR was used alone, or for CANO between the combined model and CPR or PlGF in isolation.

CONCLUSIONS

This pilot proof-of-concept study describes the screening performance of CPR and maternal PlGF level for CS for IFC in low-risk women from 36 weeks' gestation. It was found that CPR and maternal PlGF improved the overall predictive utility for CS for IFC, as well as that for CANO. However, given the lack of significant difference between the combined model and its individual components, it is debatable whether the combined model is a superior screening test. Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.

Fetal heart rate classification in routine use: Do your prefer a 3-tier or a 5-tier classification?

OBJECTIVE

To assess the current use of a five-tier fetal heart rate (FHR) classification system (National College of French Obstetricians and Gynecologists, CNGOF, 2007) and of a three-tier system (Federation International of Gynecology and Obstetrics, FIGO, 2015).

MATERIALS AND METHODS

This was a single-center prospective study conducted in April 2016. Midwives were asked to classify FHR hourly during their patients' labors according to two classification systems (CNGOF and FIGO). For each system the midwives rated from 0 to 10 the following elements after delivery: ease of FHR classification, the memorization of the classification, access to routine use, and help with the decision of a second-line examination. Finally, they had to choose which classification system seemed most helpful in their clinical practice.

RESULTS

Forty-six patients were included in the study. The median score for the ease of FHR classification according to the CNGOF system was 7, versus 8 according to the FIGO system (p<0.05). The median score for the ease of remembering the classification was 4 for CNGOF versus 8 for FIGO (p<0.05). The FIGO classification system was considered the easiest to use in 76% of cases and the CNGOF system was the most helpful in 61% of cases. The CNGOF system was seen as a help in deciding on a second-line examination in 70% of cases and the FIGO was a help in 63% of cases.

CONCLUSION

The three-tier FIGO classification system seemed easier to use but the five-tier CNGOF classification system was more helpful. The choice of which system to use should be discussed within each medical team.

Central Fetal Monitoring With and Without Computer Analysis: A Randomized Controlled Trial

OBJECTIVE

To evaluate whether intrapartum fetal monitoring with computer analysis and real-time alerts decreases the rate of newborn metabolic acidosis or obstetric intervention when compared with visual analysis.

METHODS

A randomized clinical trial carried out in five hospitals in the United Kingdom evaluated women with singleton, vertex fetuses of 36 weeks of gestation or greater during labor. Continuous central fetal monitoring by computer analysis and online alerts (experimental arm) was compared with visual analysis (control arm). Fetal blood sampling and electrocardiographic ST waveform analysis were available in both arms. The primary outcome was incidence of newborn metabolic acidosis (pH less than 7.05 and base deficit greater than 12 mmol/L). Prespecified secondary outcomes included operative delivery, use of fetal blood sampling, low 5-minute Apgar score, neonatal intensive care unit admission, hypoxic-ischemic encephalopathy, and perinatal death. A sample size of 3,660 per group (N=7,320) was planned to be able to detect a reduction in the rate of metabolic acidosis from 2.8% to 1.8% (two-tailed α of 0.05 with 80% power).

RESULTS

From August 2011 through July 2014, 32,306 women were assessed for eligibility and 7,730 were randomized: 3,961 to computer analysis and online alerts, and 3,769 to visual analysis. Baseline characteristics were similar in both groups. Metabolic acidosis occurred in 16 participants (0.40%) in the experimental arm and 22 participants (0.58%) in the control arm (relative risk 0.69 [0.36-1.31]). No statistically significant differences were found in the incidence of secondary outcomes.

CONCLUSION

Compared with visual analysis, computer analysis of fetal monitoring signals with real-time alerts did not significantly reduce the rate of metabolic acidosis or obstetric intervention. A lower-than-expected rate of newborn metabolic acidosis was observed in both arms of the trial.

CLINICAL TRIAL REGISTRATION

ISRCTN Registry, http://www.isrctn.com, ISRCTN42314164.

A cross-sectional comparison of three guidelines for intrapartum cardiotocography

OBJECTIVE

To compare the cardiotocography classification systems outlined by the International Federation of Gynecology and Obstetrics (FIGO) in 2015 and the UK National Institute for Health and Care Excellence (NICE) in 2007 and 2014.

METHODS

A cross-sectional observational study of cardiotocography practices at a UK hospital was conducted among labor ward staff (n=21) from November 1 to November 31, 2015. All observers classified ten cardiotocography traces according to the three guidelines using a bespoke form. Outcome measures included interobserver agreement (κ values), percentage agreement, intervention rate, and perceived ease of use.

RESULTS

The κ values were 0.38 (FIGO 2015), 0.37 (NICE 2007), and 0.34 (NICE 2014). The percentage agreement was identical across the three systems for both normal cardiotocography results (100.0%) and for intermediate or suspicious results (80.9%). By contrast, the percentage agreement for abnormal or pathological findings was 47.6% for NICE 2014, 76.2% for FIGO 2015, and 91.0% for NICE 2007 guidelines. Among 210 observations, intervention was deemed necessary for 48 (22.9%) for FIGO 2015, 29 (13.8%) for NICE 2014, and 56 (26.7%) for NICE 2007 guidelines. The FIGO 2015 system was considered the easiest to use by 13 (61.9%) observers.

CONCLUSION

Interobserver agreement of cardiotocography classification is suboptimal. The FIGO 2015 system offered favorable agreement scores, perceived ease of use, and a moderate intervention rate.

Reliability in cardiotocography interpretation - impact of extended on-site education in addition to web-based learning: an observational study

INTRODUCTION

Previous studies have shown poor reproducibility in cardiotocography (CTG) interpretation. Studies evaluating the Swedish web-based CTG-education program have not proven to increase accurate CTG assessments. The aim of this study was to evaluate whether an extended education can improve inter- and intra-observer reliability in CTG interpretation.

MATERIAL AND METHODS

Six obstetricians from two different departments interpreted 106 CTG tracings on two occasions. Both departments used a Swedish national web-based CTG education and test for training. One department had, in addition, an extended education program consisting of on-site lectures and oral examinations. Inter- and intra-observer agreements were calculated by simple or weighted kappa (κ) values for the five parameters assessed on CTG.

RESULTS

In both departments inter-observer and intra-observer κ showed moderate to excellent agreement (ranges for κ 0.41-0.76 and 0.65-0.93, respectively). Obstetricians at the department with extended CTG education had better inter-observer reliability for variability and accelerations. This was also the case for intra-observer reliability with the addition of baseline frequency. Both inter- and intra-observer agreement increased from moderate to substantial in both departments when decelerations were dichotomized into harmless (including early and simple variable decelerations) or hypoxic (including late, severe variable, prolonged and combined decelerations) (κ 0.63-0.78) compared with the current sub-classification of decelerations (κ 0.42-0.65).

CONCLUSIONS

Agreement in CTG interpretation was better than expected in both departments, especially when divided into harmless/hypoxic changes. Combination of different learning methods (web-based, on-site lectures and case discussion) might result in a better CTG interpretation agreement compared with web-based learning solely.

Fetal Heart Rate Analysis for Automatic Detection of Perinatal Hypoxia Using Normalized Compression Distance and Machine Learning

Accurate identification of Perinatal Hypoxia from visual inspection of Fetal Heart Rate (FHR) has been shown to have limitations. An automated signal processing method for this purpose needs to deal with time series of different lengths, recording interruptions, and poor quality signal conditions. We propose a new method, robust to those issues, for automated detection of perinatal hypoxia by analyzing the FHR during labor. Our system consists of several stages: (a) time series segmentation; (b) feature extraction from FHR signals, including raw time series, moments, and usual heart rate variability indices; (c) similarity calculation with Normalized Compression Distance, which is the key element for dealing with FHR time series; and (d) a simple classification algorithm for providing the hypoxia detection. We analyzed the proposed system using a database with 32 fetal records (15 controls). Time and frequency domain and moment features had similar performance identifying fetuses with hypoxia. The final system, using the third central moment of the FHR, yielded 92% sensitivity and 85% specificity at 3 h before delivery. Best predictions were obtained in time intervals more distant from delivery, i.e., 4-3 h and 3-2 h.

Cardiotocography and the evolution into computerised cardiotocography in the management of intrauterine growth restriction

Timely recognition and appropriate management of high-risk pregnancies, such as intrauterine growth restriction (IUGR), are of paramount importance for every obstetrician. After the initial screening of IUGR fetuses through sonographic fetometry and Doppler, the focus is shifted to the appropriate monitoring and timing of delivery. This can, especially in cases of early IUGR, become a very difficult task. At this point, cardiotocography (CTG) is introduced as a major tool in the day-to-day monitoring of the antenatal well-being of the IUGR fetus. Since the first introduction of CTG up to the nowadays widely spreading implementation of computerised CTG in the clinical practice, there has been great progress in the recording of the fetal heart rate, as well as its interpretation. Focus of this review is to offer an understanding of the evolution of CTG from its early development to modern computerised methods and to provide an insight as to where the future of CTG is leading, especially in the monitoring of IUGR.