Intrapartum non‐invasive electrophysiological monitoring: A prospective observational study

Abstract

Introduction Doppler ultrasound cardiotocography is a non‐invasive alternative that, despite its poor specificity, is often first choice for intrapartum monitoring. Doppler ultrasound suffers from signal loss due to fetal movements and is negatively correlated with maternal body mass index (BMI). Reported accuracy of fetal heart rate monitoring by Doppler ultrasound varies between 10.6 and 14.3 bpm and reliability between 62.4% and 73%. The fetal scalp electrode (FSE) is considered the reference standard for fetal monitoring but can only be applied after membranes have ruptured with sufficient cervical dilatation and is sometimes contra‐indicated. A non‐invasive alternative that overcomes the shortcomings of Doppler ultrasound, providing reliable information on fetal heart rate, could be the answer. Non‐invasive fetal electrocardiography (NI‐fECG) uses a wireless electrode patch on the maternal abdomen to obtain both fetal and maternal heart rate signals as well as an electrohysterogram. We aimed to validate a wireless NI‐fECG device for intrapartum monitoring in term singleton pregnancies, by comparison with the FSE. Material and methods We performed a multicenter cross‐sectional observational study at labor wards of 6 hospitals located in the Netherlands, Belgium, and Spain. Laboring women with a healthy singleton fetus in cephalic presentation and gestational age between 36 and 42 weeks were included. Participants received an abdominal electrode patch and FSE after written informed consent. Accuracy, reliability, and success rate of fetal heart rate readings were determined, using FSE as reference standard. Analysis was performed for the total population and measurement period as well as separated by labor stage and BMI class (≤30 and >30 kg/m2). Results We included a total of 125 women. Simultaneous registrations with NI‐fECG and FSE were available in 103 women. Overall accuracy is −1.46 bpm and overall reliability 86.84%. Overall success rate of the NI‐fECG is around 90% for the total population as well as for both BMI subgroups. Success rate dropped to 63% during second stage of labor, similar results are found when looking at the separate BMI groups. Conclusions Performance measures of the NI‐fECG device are good in the overall group and the separate BMI groups. Compared with Doppler ultrasound performance measures from the literature, NI‐fECG is a more accurate alternative. Especially, when women have a higher BMI, NI‐fECG performs well, resembling FSE performance measures.

Non-invasive Fetal Electrocardiography for Intrapartum Cardiotocography

“…Because of this limitation, we cannot apply our method to public datasets such as the Physionet Non-invasive Fetal ECG Database, as in this dataset "electrode positioning was varied in order to improve SNR" (27). Yet, we can compare our method to results from (6) and (12) where similar devices are tested on similar datasets. Both devices from these studies (i.e., Monica Healthcare AN24 and Nemo Fetal Monitoring System) use transabdominal electrodes to record a multi-channel fetal ECG and use proprietary signal processing methods to extract the FHR from these recordings.…”

“…As evaluation metrics, we opt to express the performance in FHR estimation in terms of success rate, reliability, and accuracy. Here, success rate is defined as the percentage of time the method can provide a FHR estimation (6,12). Reliability is expressed in terms of positive percent agreement (PPA) which is defined as the percentage of FHR values provided by the method that are within a 10% margin from a valid simultaneous FHR from the FSE (6).…”

“…The average accuracy and 95% CI were determined by taking the mean of the distribution and the 2.5 and 97.5% centile of the distribution, respectively. From the 10,000 bootstrap samples, also the mean standard deviation was determined, which was subsequently used to calculate limits of agreement as mean accuracy ± 2 × mean standard deviation (12).…”

“…Over the past decades, extensive research has focused on noninvasive fetal electrophysiological recordings for measurement of the CTG (10). Other than Doppler ultrasound, these electrophysiological recordings are hardly affected by movement and body mass (11,12). However, the recordings are corrupted by many electrical interferences, of which the maternal heart is the dominant source.…”

“…one of the authors is co-founder). The use of these solutions in clinical practice is still fairly limited, mainly due to the relatively poor performance during second stage of labor (6,12). This paper proposes a practical solution to solve the problems that limit the application of non-invasive fetal electrocardiography-based cardiotocography, especially during second stage of labor.…”