Evidence and clinical relevance of maternal-fetal cardiac coupling: A scoping review

Abstract

Background

Researchers have long suspected a mutual interaction between maternal and fetal heart rhythms, referred to as maternal-fetal cardiac coupling (MFCC). While several studies have been published on this phenomenon, they vary in terms of methodologies, populations assessed, and definitions of coupling. Moreover, a clear discussion of the potential clinical implications is often lacking. Subsequently, we perform a scoping review to map the current state of the research in this field and, by doing so, form a foundation for future clinically oriented research on this topic.

Methods

A literature search was performed in PubMed, Embase, and Cochrane. Filters were only set for language (English, Dutch, and German literature were included) and not for year of publication. After screening for the title and the abstract, a full-text evaluation of eligibility followed. All studies on MFCC were included which described coupling between heart rate measurements in both the mother and fetus, regardless of the coupling method used, gestational age, or the maternal or fetal health condition.

Results

23 studies remained after a systematic evaluation of 6,672 studies. Of these, 21 studies found at least occasional instances of MFCC. Methods used to capture MFCC are synchrograms and corresponding phase coherence indices, cross-correlation, joint symbolic dynamics, transfer entropy, bivariate phase rectified signal averaging, and deep coherence. Physiological pathways regulating MFCC are suggested to exist either via the autonomic nervous system or due to the vibroacoustic effect, though neither of these suggested pathways has been verified. The strength and direction of MFCC are found to change with gestational age and with the rate of maternal breathing, while also being further altered in fetuses with cardiac abnormalities and during labor.

Conclusion

From the synthesis of the available literature on MFCC presented in this scoping review, it seems evident that MFCC does indeed exist and may have clinical relevance in tracking fetal well-being and development during pregnancy.

1. Introduction

Although the mother and fetus are physically distinct from each other, their cardiac systems are connected via the placenta to facilitate gas and nutrient exchange for the fetus [1]. Both cardiac systems are constantly adapting in response to external as well as internal stimuli [2].

For example, the mother’s heart rate (HR) is influenced by the environmental temperature and the time of day but also changes in response to her stress levels [2, 3]. Similarly, the fetal HR will be regulated in response to internal triggers, for example, fetal blood oxygen levels [4], as well as external triggers such as lights and sounds sensed through the maternal abdomen [5]. However, since the external environment of the fetus is that of the maternal womb, the fetus also responds to changes in maternal physiology, for example changing maternal stress levels [6]. Moreover, the fetus forms part of the internal environment of the mother, and maternal HR has also been observed to change in response to fetal movement [7]. Researchers have suggested that maternal HR may respond to changes in fetal HR and vice versa–this mutual interaction is referred to as maternal-fetal cardiac coupling (MFCC) [8].

Since Hildebrandt et al. in 1979 first suggested that there may be an interaction between maternal and fetal heartbeats [9], researchers have investigated the potential existence and applications of MFCC [10–12]. Quantifying and understanding the presence, strength, and direction of MFCC is valuable. Not only could assessments of MFCC elucidate gestational cardiac physiology, but such assessments may also offer tools to track fetal development and screen for maternal and fetal complications [13–15].

The potential interaction between maternal and fetal heart rhythms is a complex and not yet clearly defined research field [8, 16]. Although more than 20 research studies have been published on the topic of MFCC, these studies not only employ different methods and study different populations, but also define MFCC differently. Consequently, how to quantify and interpret MFCC remains unclear. Moreover, while clinical relevance is a common aim of research on physiological coupling, results of MFCC analyses are reported without a clear discussion of the potential clinical implications.

Therefore, an exploratory mapping of existing literature–presented in a clinically accessible manner–is a necessary foundation for future clinically motivated research in this field. As MFCC is an area of emerging research, this topic lends itself to a scoping review. A scoping review provides a detailed overview of all research in the field and goes beyond answering a specific question, as is typically the motivation for a systematic review. In this manner, scoping reviews generate findings that help refine research priorities and inform future primary research [17, 18].

With this scoping review, we aim to ascertain the current state of research on MFCC and, in doing so, form a foundation for future clinically oriented research on this topic. To this end, we perform a search of all available research in this field. Thereafter, we synthesize the evolution of the methodologies employed to capture MFCC. Next, we summarize the results to determine whether MFCC exists and, if so, which physiological pathways may regulate MFCC. Finally, we discuss the potential clinical implications of MFCC.

2. Methodology

The methodology for this scoping review followed the framework first suggested by Arksey and O’Malley [17] while incorporating further suggestions and insights from Levac et al. [19], Daudt et al. [20], Munn et al. [18], and Peters et al. [21]. The review was reported per the PRISMA guidelines extension for scoping reviews (PRISMA-ScR) [22]. The protocol for this review was preregistered before the literature search and data extraction on Open Science Framework [23].

2.1 Search strategies and study selection

The search strategy was developed in consultation with a clinical librarian and can be found in S1 File. Searches were carried out on 27 October 2022 in PubMed, Embase, and Cochrane. No date limits or other filters were applied, but the language was limited to English, Dutch, and German, owing to the language proficiency of the primary authors. Search results were downloaded and systematically sorted using Rayyan QCRI, a platform specifically designed to manage the review process (https://rayyan.qcri.org/welcome). This platform was also utilized to automatically identify and eliminate duplicate studies identified across multiple databases. Citations and references of the included studies were further searched to identify more potential studies. Additionally, a search was performed of all the works published by the researchers of the included studies to identify any further work concerning MFCC.

Studies had to meet certain criteria to be eligible for the review. All studies assessing MFCC–regardless of the coupling method used, gestational age (GA), or the maternal-fetal health condition–which incorporated HR measurements from both the mother and fetus, were allowed for this scoping review. Studies measuring only other types of coupling, e.g. coupling between maternal HR and fetal movement, were excluded.

The review process comprised two levels of screening. First, the title and abstract of all the identified literature were screened. Thereafter, a full-text review of studies identified in the first level was carried out to assess eligibility. The review process was carried out independently by two researchers (MB, TN), blinded to each other’s results [19]. After each level of screening, the identified studies were discussed. Disagreement was resolved by discussion. If necessary, an independent researcher was consulted to decide whether an article should be included.

In some cases, research was disseminated first as a conference paper and thereafter as a journal article. In these cases, when everything reported in the conference paper was encompassed in the journal article, the conference paper was excluded. Furthermore, if only a conference abstract was available for a study without an accompanying paper, the abstract was excluded.

3. Results

A total of 6,672 studies were identified by searching the indicated databases. An additional six studies were found through other resources; three were found by searching the references of studies included through the database search, while three were found via searches of publications from researchers active in the field of MFCC. The latter three were either conference papers from technically oriented conference proceedings [24] or articles from journals that are not listed in PubMed, Embase, or Cochrane [8, 12]. After removing the duplicate studies, 4,813 unique studies remained, of which 32 were found eligible for full-text screening. After this full-text assessment, 23 studies were included in this review. Fig 1 shows a flowchart of the selection process. The study characteristics and results are summarized in Table 1.

[Figure omitted. See PDF.]

[Figure omitted. See PDF.]

Hereafter we will elaborate on four aspects of the results reported in the table, namely: the different methodologies that have been used to capture MFCC; the results on the existence and direction of MFCC; the physiological explanations offered for MFCC; and the potential clinical possibilities of MFCC suggested in the included studies.

3.1 MFCC: Methodologies

Broadly, MFCC analyses may be assigned to three groups: synchronization or coordination, describing a fixed relationship between two signals in either phase or time; pattern-matching, where the aim is to see if similar activity occurs in both signals; and modulation, which implies that changes in one signal results in or relates to changes in another [25]. The methodology of earlier studies investigating MFCC focused on finding periods of synchronization with synchrograms and corresponding phase coherence indices [9, 26–31], as well as corresponding patterns between maternal and fetal cardiac activity with cross-correlation [7, 32, 33]. In line with the latter, joint-symbolic dynamics was subsequently used to investigate whether maternal and fetal HR behavior corresponded to each other [8, 34, 35]. In more recent studies, the focus mostly shifted towards methods more closely associated with modulation [6, 10–12, 36]. A summary of these methods is presented here.

3.1.1 Synchrograms and phase coherence index.

Synchrograms are a visual representation of the relative phases of the maternal and fetal heartbeats. The more fixed the relationship between the relative phases of the maternal and fetal heartbeats are, the higher the coherence is between them. When periods of sufficient coherence occur (i.e., where the metric describing coherence exceeds a prespecified threshold), it is determined that phase locking occurs in this period of the signal. The expected ratio between the heartbeats needs to be defined a-priori. Periods where phase locking is detected are reported either as the number of occurrences of these phase locking periods or as their prevalence in the signals (e.g., phase synchronization of two fetal heartbeats to one maternal heartbeat, 2:1, was found in 8% of the signal). Such analyses do not address the potential directionality of MFCC. In other words, it does not say whether the fetal HR affects the maternal HR or the other way around. Additionally, the final study included in Table 1 uses an artificial intelligence method known as Deep coherence [37]. This method is a deep learning implementation of the phase coherence index, where the deep learning model seeks to identify phases of synchronization in correspondence with what would be found with the original method described above, but without any mathematical derivations, pre-processing steps, or signal transformations to the input data [37].

3.1.2 Cross-correlation.

This method assesses the similarity of two signals as a function of the displacement of one signal relative to the other. Therefore, cross-correlation accounts for a possible lag in the relationship between the maternal and fetal heartbeats. Therefore, it is possible to see when a pattern in one signal precedes the pattern in the other which may offer some indication of the directionality of the coupling. A higher cross-correlation value, therefore, implies stronger coupling.

3.1.3 Joint-symbolic dynamics (JSD).

JSD is a processing technique where information in a complex signal is simplified by replacing it with symbols (known as course-graining). In the case of MFCC, for example, each heartbeat may be replaced with a symbol indicating that it is increased (I), constant (C), or decreased (D) to the previous beat. In this way, patterns are detected in the signal, for example, DDD would indicate a sustained decrease in HR. In JSD, both the maternal and fetal HR signals are replaced with such symbols. Thereafter, the overlap between the two signals is measured with for example cross-correlation methods or cross-sample entropy.

3.1.4 Transfer entropy (TE).

TE assesses whether having information about the past activity of signal 1 reduces the information needed to describe the current or future activity in signal 2. The more the past information of signal 1 reduces the uncertainty in describing signal 2, the higher the information flow, and therefore TE, is from signal 1 to signal 2. A higher TE value suggests stronger coupling. TE inherently assumes a direction between the interactions.

3.1.5 Granger causality and partial directed coherence (PDC).

Granger causality operates on a similar principle to TE. If past information from signal 1 is useful in predicting the current state of signal 2, signal 1 is said to cause signal 2. Granger causality therefore inherently presumes a directionality between the information flow of the two signals. PDC, which is said to determine the intensity of information flow, is based on the principle of Granger causality. However, while Granger causality is assessed in the time domain, PDC is calculated using the frequency information of the time series. A higher causality or coherence value indicates stronger coupling.

3.1.6 Bivariate phase rectified signal averaging (BPRSA).

BPRSA assumes that changes in signal 1 (the trigger signal) result in or correspond to changes in signal 2 (the target signal). Anchor points–which are defined as the location of certain signal phases of interest, such as where the HR decelerates–are identified on the trigger signal. A signal segment is isolated around each anchor point which is sufficiently long to capture the expected interactions. All identified signal segments are then aligned and averaged. This process is then repeated in the target signal, using the anchor points identified in the trigger signal. If no relationship exists between the two signals, then this averaging should result in a flatline signal. However, if a relationship indeed exists, there should be an observable response in both averaged-out signals, implying that activity in the trigger signal is in some way influencing the target signal. By specifying the trigger and target signal, a directional relationship is inherently being investigated, yet, changes in both trigger and target signals may be modulated by a tertiary mechanism. Subsequently, an observed relationship does not imply causality.

3.1.7 Methods used in papers included in this scoping review.

Synchrograms and phase coherence index were used to investigate MFCC in 11 (47.9%) of the included studies, cross-correlation was used in three (13.0%) studies, JSD was investigated in three (13.0%) studies, TE in two (8.7%) studies, Granger causality and PDC in three studies (13.0%), and finally BPRSA was used to investigate MFCC in one study (4.4%).

3.2 MFCC: Presence and directionality

Of the included studies, 21 (91.3%) found that MFCC existed, at least, occasionally. The remaining two (8.3%) studies, which used cross-correlation to capture MFCC [7, 32], did not find any evidence of MFCC.

Studies investigating the phase locking between the maternal and fetal cardiac systems using synchrograms and phase coherence indices found occasional periods of synchronization. Using these methodologies, researchers demonstrated how the prevalence of these periods of synchronization increases or decreases under certain conditions such as different maternal respiration rates [27, 28], progressing GA [26, 30], or regular maternal exercise [28]. While epochs of synchronization were present in all recordings regardless of the maternal respiration rate, synchronization was more prevalent at higher rates of respiration. On the other hand, mothers who exercised regularly during pregnancy had lower incidences of MFCC than their less active controls [28]. GA seems to influence the synchronization ratio as it gradually reduces with progressing GA. However, another study could not demonstrate the influence of progressing GA on synchronization. Two studies using cross-correlation did not find MFCC. A third performed their analysis using nighttime recordings–owing to the reduced effect of motion artifacts and external stimuli during this period–and stratified their cross-correlation analysis by sleep stages. This study reported occasional MFCC in the period of wakefulness after sleep onset period [7, 32, 33].

The method of JSD was used in three studies, in each of which MFCC was captured and found to change with progressing GA. MFCC patterns were significantly different between the early- and mid-GA groups as well as between the early- and late-GA groups (16–25 weeks, 26–31 weeks, and 32–41 weeks GA, respectively) [34]. Furthermore, one of these studies compared MFCC in women between their third trimester and during labor, finding stronger MFCC patterns during labor as compared to the third trimester [8]. The third study found altered MFCC patterns in fetuses with cardiac abnormalities in comparison to healthy fetuses. The changes in the MFCC patterns of the complicated pregnancies compared to the healthy ones differed depending on the type of fetal cardiac anomaly [35].

Furthermore, MFCC was investigated with TE, and researchers found MFCC in both directions. We introduce the terminology of MFCC_M→F if information flows from the mother to the fetus and MFCC_F→M for the alternative. The mentioned directionality should not be interpreted as implying causality. MFCC_M→F was found to either stay constant or increase only slightly with progressing GA, while MFCC_F→M was found to reduce with advancing gestation [10, 11]. Studies using PDC or Granger causality similarly found MFCC to be present in both directions with the strength of MFCC_M→F increasing with GA while the strength of MFCC_F→M decreased with progressing GA [12, 15, 36].

3.3 MFCC: Physiological pathways

No studies included in this review described specific investigations into the physiological pathways that are responsible for MFCC. However, some researchers suggest that the maternal heart rhythm mechanically or vibroacoustically stimulates the fetal heart [26–28]. The pulsation of the maternal arteries causes vibrations which may be sensed or heard by the fetus. When the frequency of these vibrations approaches that of the fetal heart rhythm, the fetal heartbeat may become entrained to the maternal heart [9, 38, 39]. Furthermore, researchers suspect that the autonomic nervous system (ANS) serves as a pathway for MFCC. Consequently, small scale studies performed on mice models were used to test this hypothesis. These studies revealed alterations in MFCC under maternal sympathetic or para-sympathetic blockade [40], although no clear conclusions could be drawn as to the role of the ANS in MFCC.

3.4 MFCC: Clinical relevance

Overall, researchers suggest that assessing MFCC may serve as a tool to assess fetal well-being during pregnancy and labor, and to track fetal development. However, three clinical applications of MFCC have been specifically investigated: first, the potential for using MFCC indices to discriminate between normal and abnormal fetuses [15, 30, 35]; second, estimating GA based on MFCC indices in rural or remote setups where ultrasound technology or expertise is not available [14, 31]; and third, using MFCC as an index of prenatal exposure to maternal stress [6].

Three studies specifically investigated abnormal fetuses in comparison to healthy fetuses. Two of these studies investigated a heterogeneous group of pregnancies with fetal cardiac anomalies or fetal cardiac heart rhythm disorders such as fetal bradycardia, fetal tachycardia, or premature atrial contractions. The first study, using JSD, found stronger MFCC patterns for the abnormal cases when compared to pregnancies with healthy fetuses [35]. The second study, using PDC, found decreased MFCC_F→M, while MFCC_M→F was increased compared to healthy fetuses [15]. Finally, one other study using phase locking found significant differences in phase coherence indices between fetuses affected by different types of congenital heart diseases and healthy fetuses [30].

Two other studies showed that incorporating synchronization and phase coherence parameters could improve the estimation of GA with regression models when compared against models using only maternal and fetal HR variability features. When compared against the gold standard of establishing GA from crown-rump length, the best performing model had a mean root mean square error of 2.67 weeks [14, 31].

Finally, one study used BPRSA to investigate MFCC in fetuses with stressed mothers (as assessed with the Perceived Stress Scale index) [6]. Features from the BPRSA analysis were used to develop a fetal stress index (FSI). The FSI was significantly higher in fetuses with stressed mothers compared to controls.

4. Discussion

Although there is heterogeneity in methodologies used and populations assessed in the studies included in this scoping review, it seems that MFCC does indeed exist, both from the mother to her fetus (MFCC_M→F) as well as from the fetus to its mother (MFCC_F→M). Furthermore, there is potential clinical value in assessing MFCC for monitoring fetal well-being and tracking fetal development.

While analyses using cross-correlation did not yield convincing evidence for MFCC [7, 32], phase synchronization, along with its phase coherence index, captured occasional MFCC between the maternal-fetal pair [9, 14, 15, 26–31, 36, 40]. Researchers have also remarked that using cross-correlation to investigate associations between timeseries data (such as maternal and fetal heart rhythms) often leads to an underestimation of the strength of the association [7]. Considering this limitation and further considering that most studies support the existence of MFCC, we conclude that cross-correlation is not an appropriate method for capturing MFCC.

The seemingly intermittent nature of MFCC motivated investigations into the conditions under which MFCC occurs. While regular maternal physical exercise resulted in less synchronization between maternal and fetal HR [28], higher instances of MFCC were found at higher maternal respiration rates [27]. Furthermore, MFCC also varies with progressing pregnancy both in strength and direction. In early pregnancy, the influence is mainly from the fetus to the mother, while in later pregnancy MFCC_M→F is dominant [11, 12, 15, 41].

The etiology of MFCC is currently unknown. Some researchers suggest that this type of coupling may be mechanically or acoustically driven [26–28]. Similar to cardiac rhythms becoming entrained to locomotor actions in cardiac-locomotor coupling (i.e., when the frequency of a rhythmic activity such as walking becomes close to the frequency of the HR, or a fixed factor thereof, and the two synchronize to each other) [42], the fetal heart rhythm may become entrained to the forcing maternal cardiac oscillator, i.e., the maternal pulse waves [9]. Furthermore, the fetal HR changes in reaction to the mechanical energy from the maternal vessels may be enhanced by the fetus’s auditory perception of the frequency range of the pulsating maternal arteries–a phenomenon called the vibroacoustic effect [26–28]. This could explain the increasing strength of MFCC_M→F with GA, as the fetal auditory system is only fully developed at 27 weeks of gestation [43]. The vibroacoustic effect has been observed in adults where a frequency-lock was found in reaction to an external acoustic signal, but only when the frequency was similar to that of the subject’s HR [38]. The same might be happening to the fetal HR in the case of MFCC, although this would rarely be observed as the HR frequencies of the mother and fetus would be too far apart to induce MFCC under normal circumstances.

Subsequently, it stands to reason that a higher incidence of MFCC may be observed at higher maternal HR. This aligns with findings suggesting more periods of MFCC during quicker maternal respiration; increased respiration narrows the maternal interval between successive heartbeats, potentially encouraging the entrainment of the fetal rhythm to that of its mother [27]. Similarly, MFCC was less common in pregnancies where mothers had higher cardiovascular fitness and correspondingly lower resting respiratory rate and HR [28]. However, rather than changes in maternal cardiac rhythm modulating changes in the fetal rhythm (or vice versa), it may be possible that a third system is driving changes in both these systems [33]. Specifically, it is feasible that maternal respiration acts as a common driving force, simultaneously modulating both the maternal and the fetal HR [27]. This modulating effect of maternal respiration is yet to be directly investigated. However, a faster-paced maternal respiratory rate did induce higher instances of MFCC [27]. We propose that the movement of the maternal diaphragm may also play a role here, exhorting a vibroacoustic effect on both the maternal and fetal cardiac system, but this has not been investigated.

The increased strength of MFCC_M→F with progressing GA is likely linked to the maturation of the fetal ANS, reaching maturity around the transition from the second to the third trimester [44]. With gestational progression, the increasingly stable and finely tuned fetal ANS may lead to an enhanced fetal cardiac reaction to maternal input [10, 41, 45]. On the other hand, while respiratory sinus arrhythmia is typically present in the mother, in the fetus it is present in increasing strength from 32 weeks GA onwards [44]. Theoretically, the fetal HR would become more closely coupled with its own respiratory system from this point onward. Yet, this manifests as a decrease in MFCC_F→M with progressing GA rather than a decrease in MFCC_M→F, further highlighting the complexity and dearth of knowledge concerning MFCC.

Additionally, the adrenergic innervation of the uterine wall may play an important role in MFCC_F→M [46, 47]. Fetal movements may stimulate the maternal sympathetic nervous system, resulting in higher maternal HR. Theoretically, this effect would increase with gestational progression as larger fetuses are capable of stronger movements. However, the opposite is observed; MFCC_F→M decreases with progressing pregnancy. This decrease is likely due to the maternal ANS becoming increasingly hypo-responsive to external stimuli, such as fetal movements, during healthily progressing pregnancy [41, 48, 49]. Small-scale studies using mice models also support the hypothesis of the ANS playing a central role in MFCC as indices of MFCC in pregnant mice reveal an antagonistic response to maternal sympathetic or para-sympathetic blockade [40]. However, these animal studies used phase coherence for the assessment of MFCC and were therefore not able to assess directionality.

Still, even though the etiology of MFCC is not yet clear, results do suggest that assessments of this coupling may have clinical relevance. Indices of MFCC are altered in pregnancies with fetuses affected by cardiac arrhythmias or fetal cardiac anomalies compared to healthy fetuses [15, 30, 35]. Furthermore, MFCC parameters have been used to estimate fetal GA fairly successfully against the gold standard [14, 31].

Maternal stress during pregnancy has also been found to affect MFCC [6]. Based on this finding, researchers have developed an FSI (based on MFCC features) to identify infants at risk for altered neurodevelopmental trajectories due to perinatal stress exposure [6]. While no further clinical applications have been investigated, the most common suggestion for clinical applications is tracking fetal neurodevelopment to screen for abnormalities.

The effect of maternal complications on MFCC has yet to be explored. Such analyses are potentially interesting since, as previously discussed, MFCC seems to be affected by autonomic changes, and complications such as hypertensive disorders of pregnancy are associated with dysfunctional autonomic regulation [50]. Furthermore, assessments of maternal-infant cardiac coupling in the immediate postnatal period–preferably in preterm infants where the autonomic behavior is still similar to that of the fetus–may be illuminating. In such a study design, various possible influencing factors could be examined under controlled conditions, for example, changes in maternal respiration rate or HR. Additionally, synchronization under specific (patho)physiological conditions such as fetal behavioral state or fetal hypoxemia should be investigated. The latter might be particularly interesting. While the evolutionary driver behind MFCC is unknown, it may be in some way related to the oxygenation of the fetus; i.e., lower oxygenation levels in the fetal blood could trigger increases in maternal HR to increase gas exchange via the placenta [51]. On the other hand, when maternal oxygen levels decrease, the fetal HR responds by increasing the fetal HR [52].

Several limitations exist that affect the investigation of MFCC. While each of the studies in this scoping review is limited in some ways, there are also inherent difficulties in studying MFCC. First, since time-synchronized maternal and fetal HR are needed, options for measurement technologies to capture MFCC are limited. While magnetocardiography can be used, it is impractical, owing to the size and expense of the equipment, therefore leaving abdominal electrocardiophysiology (ECG) as the pragmatic option. Fetal HR can be difficult to accurately detect from abdominal ECG and signals capturing the electrophysiological activity of the fetal heart are typically weak (i.e., of low signal-to-noise ratio). However, recent advancements in the field of fetal electrocardiography have greatly contributed to solving this problem by providing higher quality fetal signals that enable more accurate MFCC investigation [53]. Second, the majority of methods used to assess the coupling between systems derive from different scientific domains and are not specifically designed to study coupling between physiological systems, which might make them less effective. Third, the studies included in this review reveal that there is no consensus on the definition of MFCC. This is important since the definition of coupling determines the method by which researchers chose to study its potential occurrence; a presumption of fixed phase ratios between the maternal and fetal heartbeats would most likely lead to analysis via synchrograms, while hypothesizing that modulations in one signal lead to or correspond to changes in the other would likely result in a TE or BPRSA analysis.

Lastly, a deep learning approach called deep coherence was recently proposed in the field of MFCC research [37]. Deep learning methods like deep coherence may help to reduce the need for a priori assumptions and processing. However, from this review it is clear that while MFCC does seem to exist, our understanding of MFCC is limited. Therefore, techniques which are not fully explainable to capture MFCC should be used with caution. Rather, it may be beneficial for future research to first directly compare known coupling techniques for the assessment of MFCC to narrow down those which are useful [54]. Furthermore, more research is necessary to probe the pathway behind and nature of MFCC.

5. Conclusion

We conclude that the studies included in this scoping review suggest that MFCC does exist and that its strength and direction change with progressing GA. Although the physiological pathways of MFCC are not yet sufficiently substantiated, assessing MFCC during pregnancy may offer opportunities to assess fetal development and well-being and may potentially aid in detecting fetal (cardiac) abnormalities.

Supporting information

S1 Checklist. PRISMA-ScR checklist.

https://doi.org/10.1371/journal.pone.0287245.s001

(DOCX)

S1 File. Search strategy (Pubmed, Embase).

https://doi.org/10.1371/journal.pone.0287245.s002

(DOCX)

Citation: Nichting TJ, Bester M, Joshi R, Mischi M, van der Ven M, van der Woude DAA, et al. (2023) Evidence and clinical relevance of maternal-fetal cardiac coupling: A scoping review. PLoS ONE 18(7): e0287245. https://doi.org/10.1371/journal.pone.0287245

About the Authors:

Thomas J. Nichting

Contributed equally to this work with: Thomas J. Nichting, Maretha Bester

Roles: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Visualization, Writing – original draft

E-mail: [email protected]

Affiliations: Department of Gynaecology and Obstetrics, Veldhoven, The Netherlands, Department of Electrical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands, Eindhoven MedTech Innovation Centre, Eindhoven, The Netherlands

ORICD: https://orcid.org/0000-0001-9905-4380

Maretha Bester

Contributed equally to this work with: Thomas J. Nichting, Maretha Bester

Roles: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Visualization, Writing – original draft

Affiliations: Department of Electrical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands, Eindhoven MedTech Innovation Centre, Eindhoven, The Netherlands, Department of Patient Care and Monitoring, Philips Research, Eindhoven, The Netherlands

Rohan Joshi

Roles: Supervision, Writing – review & editing

Affiliation: Department of Patient Care and Monitoring, Philips Research, Eindhoven, The Netherlands

Massimo Mischi

Roles: Supervision, Writing – review & editing

Affiliations: Department of Electrical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands, Eindhoven MedTech Innovation Centre, Eindhoven, The Netherlands

Myrthe van der Ven

Roles: Supervision, Writing – review & editing

Affiliations: Department of Gynaecology and Obstetrics, Veldhoven, The Netherlands, Eindhoven MedTech Innovation Centre, Eindhoven, The Netherlands, Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands

Daisy A. A. van der Woude

Roles: Supervision, Writing – review & editing

Affiliations: Department of Gynaecology and Obstetrics, Veldhoven, The Netherlands, Department of Electrical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands, Eindhoven MedTech Innovation Centre, Eindhoven, The Netherlands

S. Guid Oei

Roles: Supervision, Writing – review & editing

Affiliations: Department of Gynaecology and Obstetrics, Veldhoven, The Netherlands, Department of Electrical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands, Eindhoven MedTech Innovation Centre, Eindhoven, The Netherlands

Judith O. E. H. van Laar

Roles: Supervision, Writing – review & editing

Affiliations: Department of Gynaecology and Obstetrics, Veldhoven, The Netherlands, Department of Electrical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands, Eindhoven MedTech Innovation Centre, Eindhoven, The Netherlands

Rik Vullings

Roles: Supervision, Writing – review & editing

Affiliations: Department of Electrical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands, Eindhoven MedTech Innovation Centre, Eindhoven, The Netherlands, Nemo Healthcare, Veldhoven, The Netherlands

ORICD: https://orcid.org/0000-0002-2392-6098

References

1. Maltepe E. and Fisher S. J., “Placenta: The Forgotten Organ,” Annu. Rev. Cell Dev. Biol., vol. 31, no. 1, pp. 523–552, Nov. 2015, pmid:26443191

2. Shaffer F., McCraty R., and Zerr C. L., “A healthy heart is not a metronome: an integrative review of the heart’s anatomy and heart rate variability,” Front. Psychol., vol. 5, Sep. 2014, pmid:25324790

3. Kim H.-G., Cheon E.-J., Bai D.-S., Lee Y. H., and Koo B.-H., “Stress and Heart Rate Variability: A Meta-Analysis and Review of the Literature,” Psychiatry Investig, vol. 15, no. 3, pp. 235–245, Mar. 2018, pmid:29486547

4. Luttkus A. K., Friedmann W., Homm-Luttkus C., and Dudenhausen J. W., “Correlation of fetal oxygen saturation to fetal heart rate patterns, Evaluation of fetal pulse oximetry with two different oxisensors: Correlation of fetal oxygen saturation to fetal heart rate patterns, Evaluation of fetal pulse oximetry with two different oxisensors,” Acta Obstetricia et Gynecologica Scandinavica, vol. 77, no. 3, pp. 307–312, Mar. 1998,

5. Sjöström K., Valentin L., Thelin T., and Maršál K., “Maternal anxiety in late pregnancy: effect on fetal movements and fetal heart rate,” Early Human Development, vol. 67, no. 1–2, pp. 87–100, Apr. 2002, pmid:11893440

6. Lobmaier S. M. et al., “Fetal heart rate variability responsiveness to maternal stress, non-invasively detected from maternal transabdominal ECG,” Arch Gynecol Obstet, vol. 301, no. 2, pp. 405–414, Feb. 2020, pmid:31781889

7. Dipietro J. A., Irizarry R. A., Costigan K. A., and Gurewitsch E. D., “The psychophysiology of the maternal-fetal relationship,” Psychophysiology, vol. 41, no. 4, pp. 510–520, Jul. 2004, pmid:15189474

8. Tepichín-Castro C. A., Ledesma-Ramírez C. I., Peña-Castillo M. Á., Pacheco-López G., Echeverría J. C., and Reyes-Lagos J. J., “Joint symbolic dynamics identifies differences in the maternal-fetal cardiac coupling between nonlaboring and laboring women,” Biomedical Signal Processing and Control, vol. 68, p. 102769, Jul. 2021,

9. Hildebrandt G. and Klein H. R., “Uber die Phasenkoordination von miitterlichem und foetalem Herzrhythmus wiihrend der Schwangerschaft,” Klin. Wochenschr, vol. 57, pp. 87–91.

10. Marzbanrad F., Kimura Y., Palaniswami M., and Khandoker A. H., “Quantifying the Interactions between Maternal and Fetal Heart Rates by Transfer Entropy,” PLoS ONE, vol. 10, no. 12, p. e0145672, Dec. 2015, pmid:26701122

11. Avci R., Escalona-Vargas D., Siegel E. R., Lowery C. L., and Eswaran H., “Coupling Analysis of Fetal and Maternal Heart Rates via Transfer Entropy Using Magnetocardiography,” in 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI: IEEE, Jul. 2018, pp. 1–4. https://doi.org/10.1109/EMBC.2018.8513053 pmid:30440290

12. Khandoker A. H. et al., “Analysis of maternal–fetal heart rate coupling directions with partial directed coherence,” Biomedical Signal Processing and Control, vol. 30, pp. 25–30, Sep. 2016,

13. Lakhno I., “Autonomic imbalance captures maternal and fetal circulatory response to pre-eclampsia,” Clin Hypertens, vol. 23, no. 1, p. 5, Dec. 2017, pmid:28191323

14. Khandoker A. H., Wahbah M., Al Sakaji R., Funamoto K., Krishnan A., and Kimura Y., “Estimating Fetal Age by Fetal Maternal Heart Rate Coupling Parameters,” in 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Montreal, QC, Canada: IEEE, Jul. 2020, pp. 604–607. https://doi.org/10.1109/EMBC44109.2020.9176049 pmid:33018061

15. Khandoker A. H., Schulz S., Al-Angari H. M., Voss A., and Kimura Y., “Alterations in Maternal–Fetal Heart Rate Coupling Strength and Directions in Abnormal Fetuses,” Front. Physiol., vol. 10, p. 482, Apr. 2019, pmid:31105586

16. Schulz S. et al., “Cardiovascular and cardiorespiratory coupling analyses: a review,” Phil. Trans. R. Soc. A., vol. 371, no. 1997, p. pmid:23858490, Aug. 2013.

17. Arksey H. and O’Malley L., “Scoping studies: towards a methodological framework,” International Journal of Social Research Methodology, vol. 8, no. 1, pp. 19–32, Feb. 2005,

18. Munn Z., Peters M. D. J., Stern C., Tufanaru C., McArthur A., and Aromataris E., “Systematic review or scoping review? Guidance for authors when choosing between a systematic or scoping review approach,” BMC Med Res Methodol, vol. 18, no. 1, p. 143, Dec. 2018, pmid:30453902

19. Levac D., Colquhoun H., and O’Brien K. K., “Scoping studies: advancing the methodology,” Implementation Sci, vol. 5, no. 1, p. 69, Dec. 2010, pmid:20854677

20. Daudt H. M., van Mossel C., and Scott S. J., “Enhancing the scoping study methodology: a large, inter-professional team’s experience with Arksey and O’Malley’s framework,” BMC Med Res Methodol, vol. 13, no. 1, p. 48, Dec. 2013, pmid:23522333

21. Peters M. D. J. et al., “Updated methodological guidance for the conduct of scoping reviews,” JBI Evidence Synthesis, vol. 18, no. 10, pp. 2119–2126, Oct. 2020, pmid:33038124

22. McGowan J. et al., “Reporting scoping reviews—PRISMA ScR extension,” Journal of Clinical Epidemiology, vol. 123, pp. 177–179, Jul. 2020, pmid:32229248

23. Bester M. et al., “Maternal-fetal cardiac coupling: a scoping review,” Open Science Framework, May 2021,

24. Khandoker A., Wahbah M., Yoshida C., Kimura Y., and Kasahara Y., “Effect of Anesthesia on Fetal-Maternal Heart Rate Variability and Coupling in Pregnant Mice and Fetuses,” presented at the 2020 Computing in Cardiology Conference, Dec. 2020.

25. Schäfer C., Rosenblum M. G., Abel H.-H., and Kurths J., “Synchronization in the human cardiorespiratory system,” Phys. Rev. E, vol. 60, no. 1, pp. 857–870, Jul. 1999, pmid:11969830

26. Leeuwen P. V., Geue D., Lange S., Cysarz D., Bettermann H., and Grönemeyer D. H., “Is there evidence of fetal-maternal heart rate synchronization?,” BMC Physiology, p. 11, 2003.

27. Van Leeuwen P. et al., “Influence of paced maternal breathing on fetal-maternal heart rate coordination,” Proceedings of the National Academy of Sciences, vol. 106, no. 33, pp. 13661–13666, Aug. 2009, pmid:19597150

28. Van Leeuwen P., Gustafson K. M., Cysarz D., Geue D., May L. E., and Grönemeyer D., “Aerobic Exercise during Pregnancy and Presence of Fetal-Maternal Heart Rate Synchronization,” PLoS ONE, vol. 9, no. 8, p. e106036, Aug. 2014, pmid:25162592

29. Wang Qiangian et al., “Investigating the beat by beat phase synchronization between maternal and fetal heart rates,” in 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Osaka: IEEE, Jul. 2013, pp. 3821–3824. https://doi.org/10.1109/EMBC.2013.6610377 pmid:24110564

30. Alangari H. M., Kimura Y., and Khandoker A. H., “Preliminary Evaluation of Fetal Congenital Heart Defects Changes on Fetal-Maternal Heart Rate Coupling Strength,” in 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI: IEEE, Jul. 2018, pp. 251–254. https://doi.org/10.1109/EMBC.2018.8512272 pmid:30440385

31. Wahbah M., Al Sakaji R., Funamoto K., Krishnan A., Kimura Y., and Khandoker A. H., “Estimating Gestational Age From Maternal-Fetal Heart Rate Coupling Parameters,” IEEE Access, vol. 9, pp. 65369–65379, 2021,

32. DiPietro J. A., Caulfield L. E., Irizarry R. A., Chen P., Merialdi M., and Zavaleta N., “Prenatal development of intrafetal and maternal-fetal synchrony.,” Behavioral Neuroscience, vol. 120, no. 3, pp. 687–701, 2006, pmid:16768620

33. DiPietro J. A. et al., “Fetal heart rate during maternal sleep,” Dev Psychobiol, vol. 63, no. 5, pp. 945–959, Jul. 2021, pmid:33764539

34. Khandoker A. H., “Investigating maternal-fetal heart rate coupling by high resolution joint symbolic dynamics,” p. 4.

35. Khandoker A. H. et al., “Quantification of maternal-fetal cardiac couplings in normal and abnormal pregnancies applying high resolution joint symbolic dynamics,” Mathematical Biosciences and Engineering, vol. 17, no. 1, pp. 802–813, 2020, pmid:31731378

36. Riedl M. et al., “Testing foetal–maternal heart rate synchronization via model-based analyses,” Phil. Trans. R. Soc. A., vol. 367, no. 1892, pp. 1407–1421, Apr. 2009, pmid:19324716

37. Alkhodari M. et al., “Deep learning identifies cardiac coupling between mother and fetus during gestation,” Front. Cardiovasc. Med., vol. 9, p. 926965, Jul. 2022, pmid:35966548

38. Anishchenko V. S., Balanov A. G., Janson N. B., Igosheva N. B., and Bordyugov G. V., “Entrainment between heart rate and weak noninvasive forcing,” Int. J. Bifurcation Chaos, vol. 10, no. 10, pp. 2339–2348, Oct. 2000,

39. Porcaro C. et al., “Fetal auditory responses to external sounds and mother’s heart beat: Detection improved by Independent Component Analysis,” Brain Research, vol. 1101, no. 1, pp. 51–58, Jul. 2006, pmid:16784726

40. Khandoker A. H. et al., “Investigating the effect of cholinergic and adrenergic blocking agents on maternal-fetal heart rates and their interactions in mice fetuses,” Biology Open, vol. 11, no. 4, p. bio058999, Apr. 2022, pmid:35188546

41. Marzbanrad F., Kimura Y., Endo M., Palaniswami M., and Khandoker A. H., “Transfer entropy analysis of maternal and fetal heart rate coupling,” in 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Milan: IEEE, Aug. 2015, pp. 7865–7868. https://doi.org/10.1109/EMBC.2015.7320215 pmid:26738115

42. De Bartolo D. et al., “Effects of cognitive workload on heart and locomotor rhythms coupling,” Neuroscience Letters, vol. 762, p. 136140, Sep. 2021, pmid:34324958

43. Hepper P. G. and Shahidullah B. S., “Development of fetal hearing,” Archives of Disease in Childhood—Fetal and Neonatal Edition, vol. 71, no. 2, pp. F81–F87, Sep. 1994, pmid:7979483

44. Schneider U. et al., “Developmental milestones of the autonomic nervous system revealed via longitudinal monitoring of fetal heart rate variability,” PLoS ONE, vol. 13, no. 7, p. e0200799, Jul. 2018, pmid:30016343

45. Kisilevsky B. S., Muir D. W., and Low J. A., “Maturation of Human Fetal Responses to Vibroacoustic Stimulation,” Society for Research in Child Development, vol. 63, no. 6, pp. 1497–1508, 1992, pmid:1446565

46. DiPietro J. A., Voegtline K. M., Costigan K. A., Aguirre F., Kivlighan K., and Chen P., “Physiological reactivity of pregnant women to evoked fetal startle,” Journal of Psychosomatic Research, vol. 75, no. 4, pp. 321–326, Oct. 2013, pmid:24119937

47. Owman C., Rosenbren E., and Sjoberg N. O., “Adrenergic innervation of the human female reproductive organs: a histochemical and chemical investigation,” Obstet Gynecol, vol. 30, no. 6, pp. 763–773, Dec. 1967. pmid:4294068

48. Entringer S. et al., “Attenuation of maternal psychophysiological stress responses and the maternal cortisol awakening response over the course of human pregnancy,” Stress, vol. 13, no. 3, pp. 258–268, May 2010, pmid:20067400

49. Bester M., Joshi R., Mischi M., van Laar J. O. E. H., and Vullings R., “Longitudinally Tracking Maternal Autonomic Modulation During Normal Pregnancy With Comprehensive Heart Rate Variability Analyses,” Front. Physiol., vol. 13, p. 874684, May 2022, pmid:35615673

50. Moors S. et al., “Heart rate variability in hypertensive pregnancy disorders: a systematic review,” Pregnancy Hypertension, Mar. 2020, pmid:32179490

51. Bekedam D. J., Mulder E. J. H., Snijders R. J. M., and Visser G. H. A., “The effects of maternal hyperoxia on fetal breathing movements, body movements and heart rate variation in growth retarded fetuses,” Early Human Development, vol. 27, no. 3, pp. 223–232, Dec. 1991, pmid:1802673

52. Kutlu T. et al., “Acute fetal heart rate tracing changes secondary to cigarette smoking in third trimester pregnancies,” The Journal of Maternal-Fetal & Neonatal Medicine, vol. 30, no. 12, pp. 1407–1409, Jun. 2017, pmid:27440435

53. Vullings R. and van Laar J. O. E. H., “Non-invasive Fetal Electrocardiography for Intrapartum Cardiotocography,” Front. Pediatr., vol. 8, p. 599049, Dec. 2020, pmid:33363064

54. Rozo A. et al., “Benchmarking Transfer Entropy Methods for the Study of Linear and Nonlinear Cardio-Respiratory Interactions,” Entropy, vol. 23, no. 8, p. 939, Jul. 2021, pmid:34441079