Monitoring Contractility of Junctional Zone Endometrium across Menstrual Cycle Using the ElectroUteroGraph (EUG): A Clinical Evaluation

Abstract

(1) Background: Abnormal uterine contractility for nonpregnant women has been associated with gynecological pathologies and infertility. The objective of this study was to evaluate the ability of a novel monitoring technique to assess the contractility of the nongravid uterus using a simple, standardized, direct, in vivo methodology during the different phases of the menstrual cycle. (2) Methods: Twenty-six healthy women of reproductive age (28–48 years) were recruited. An ElectroUteroGraph (EUG) was used to measure the electrical activity from the contractility of the junctional zone endometrium (JZE) across the menstrual cycle. Derived recordings were separated into the early proliferative (EP) (n = 10), late proliferative (LP) (n = 31), early luteal (EL) (n = 27), and late luteal (LL) (n = 12) phases of the menstrual cycle. EUG recordings were performed by inserting a flexible electrode array into the endometrial cavity. (3) Results: Waveforms that were measured from the electrode closer to the fundus (1 cm distance) were processed. The Root-Mean-Square (RMS) Voltage Amplitude (VJZE-RMS) (in μV) and the Mean Frequency (fJZE-mean) (in cycles/min) of the JZE’s electrical activity, as direct indicators of the intensity and frequency changes in the JZE’s contractions, were extracted from the recorded waveforms. There was a trend in the median values of the VJZE-RMS decreasing from the EP to the LP phase (247–158 μV). During the EL phase, an upward trend was observed (158–374 μV, p < 0.05), reaching its highest value during the LL phase (374–477 μV, p < 0.05) when compared to the LP phase. The fJZE-mean showed the opposite trend, increasing from 2.5 cycles/min during the EP phase to 2.96 cycles/min during the LP phase. During the EL phase, a downward trend was observed (2.96–2.37 cycles/min), continuing to fall to 1.33 cycles/min, in the LL phase, with p < 0.05 when compared to the previous three phases. (4) Conclusions: The novel in vivo monitoring technique has shown clinically, for the first time, significant electrical activity differences in the different sub-phases of the menstrual cycle, recorded in a safe and painless way.

1. Introduction

The junctional zone endometrium (JZE) is the layer between the endometrium and myometrium. It is described as a compaction of smooth muscle fibers surrounded by little extracellular matrix. The JZE is better visualized and evaluated using T2-weighted Magnetic Resonance Imaging (MRI) and 3D Transvaginal Ultrasound Scanning (TVUS) [1]. Current scientific data provide evidence that the JZE is an important part of the nongravid uterus anatomy, structure, and functionality. During the late proliferative and luteal phases, the myometrium is in relaxation and the JZE contracts regularly [1,2]. The JZE thickness and contractility alterations have been associated with myometrial and sub-endometrial pathologies like submucosal fibroids and adenomyosis [3]. Additionally, when the JZE exceeds 12 mm in thickness, this is considered to be a bad prognostic factor for the implantation potential [2]. The JZE’s contractility frequency is adjusted by the ovarian hormones’ cyclic activity and is also altered by pathologies adjacent to it, such as fibroids and polyps. Hence, the appearance, thickness, and contractility of the JZE can be used as markers for the diagnosis and prognosis of normal and abnormal uterine functioning, the implantation potential, early pregnancy loss, and probably for the early detection of endometrial cancer [1,2,3].

Currently, the uterine JZE contractility has only been monitored within the research context via TVUS and MRI. MRI evaluation of the JZE allows for better imaging acquisition and has a higher diagnostic accuracy with lower interobserver variation than TVUS. However, both techniques are complex and require costly machinery and highly experienced operators, whilst ultrasound lacks reproducibility [1,2,3]. Hence, a simple, reliable, low-cost tool for the objective and accurate assessment of uterine contractility is missing [4].

As in every muscle, the contraction of uterine muscles is generated due to the propagation of electrical biosignals, which potentially play a significant role with regards to fertility and various benign and malignant conditions [5]. Currently, there is no standard method for directly recording electrical biosignals from inside the endometrium without affecting the measurand. In an in vivo attempt using intrauterine electrodes, the administration of sedatives was necessary, especially to nulliparous women, to allow them to withstand the potential pain and discomfort of the procedure [6]. Furthermore, due to the invasiveness of the procedure, there is an increased probability of secondary contractility effects due to the introduced mechanical pressure that is applied to the uterus from the inserted electrodes, affecting the normal uterine peristaltic activity and leading to measurement artefacts. Alternative methods that involve placing sensors on the abdomen have shown promising results for pregnant uteri [7]. However, when using the same technique for nonpregnant uteri, even though fine peristaltic movements with contractility variations across the menstrual cycle have been observed, no significant differences outside menstruation have been noted [8,9,10].

Recently, we developed the ElectroUteroGraph (EUG), a methodology with technical requirements for the direct recording of electrical biosignals from the endometrial cavities of nonpregnant women using a low-profile intrauterine electrode array [11]. The procedure mimics the procedure of embryo transfer, whereby an outer-sheath catheter with a stiff guide wire is used for the placement of the electrode array. Additionally, the electrode placement and the measurement technique are simple, standardized, and pain-free. In this article, we focus on our clinical findings measuring the JZE’s electrical activity across the menstrual cycle for parous nonpregnant women using the EUG device.

2. Materials and Methods

2.1. Participants

The study was advertised to all patients presenting in a gynecological clinic in Aretaeio Hospital, Nicosia, through leaflets. Forty-four patients showed interest in participating in the study and twenty-six met the inclusion criteria. The inclusion criteria for the participants were healthy, nonpregnant, sexually active, adult women at the reproductive stage, experiencing regular menstrual bleeding (21–32 days), without any vulva or vaginal infections, with normal cervical smear test results within at least 1 year of the EUG test, with a visibly normal cervical os upon examination and with normal uterine and ovarian morphologies in TVUS prior to the EUG test. If the participants had any previous uterine pathologies, they were included in the study only if they had been successfully treated. Additionally, all participants had to be free from any gynecological or other general health problems during the EUG test and not be receiving any form of medication. If any of the inclusion criteria were not met in the follow-up EUG test, the participants were excluded from the study. Participants who were not able to reattend the follow-up sessions were also excluded from the study. Twenty-six participants were registered for the 2020–2021 period. The participants’ demographics, past uterine pathologies/treatment, and information regarding their menstrual cycles and derived EUG recordings are presented in

Table 1. Demographic and gynecological/clinical characteristics of the healthy, parous participants.

Healthy Parous Women	(n = 26)
Age	28–48
BMI	17.3–32
Number of children	1–4
Menstrual cycle duration (days)	21–32
Period duration (days)	2–8
History of Treated Past Pathologies
Abortion, D&C	1
Laparoscopic surgery for Endometrioma & Endometriosis	1
Hysteroscopic Polypectomy	1
Myomectomy	1
Caesarean Section	1

Number of participants is shown with the value of “n”. Values are displayed as minimum–maximum.

. Grouping was not necessary in this study. The study is reported according to the STARD criteria.

2.2. Ethical Approval

The aims of the direct in vivo measurements of the endometrial-cavity research program were presented to all recruited women that satisfied the study’s inclusion criteria. An EUG video and written information were provided to the participant prior to the procedure. Ample time for explanation and discussion was given to the participant, several days before signing the consent form to undergo an EUG recording. The EUG clinical study and the associated procedures were approved by the Cyprus National Bioethics Committee (CNBC), registered as “ΕΕΒΚ/ΕΠ/2019/70”, 23/01/2020.

2.3. Equipment

For the electrical measurements inside the endometrial cavity, a custom-designed and -manufactured, flexible, low-profile, electrode array was used, with gold-plated electrode pads and dimensions of 400 mm (length) × 1.5 mm (width) × 0.25 mm (thickness). The signals of the electrode array were recorded using a commercially available Electromyography (EMG) Device [11]. For the TVUS assessment and measurements, the E8 Voluson (GE Healthcare, Chicago, IL, USA) ultrasound (US) device with abdominal and transvaginal probes was used.

2.4. Placement Procedure

The patients were placed in the lithotomy position, where they remained for the whole 10 min of the procedure, and they were advised not to move. A vaginal examination was performed using an open-sided Pederson speculum. The cervical os was then carefully cleaned with a sterile cotton swab and immersed in antiseptic solution. A disposable stiff outer-sheath Wallace catheter (Wallace^® Classic) with a hard-guided curved probe and a rounded bulbed tip was used to facilitate insertion into the cervix and progress towards the uterine fundus (CooperSurgical Inc., Trumbull, CT, USA), as is performed in embryo transfer, after in vitro fertilization (IVF). The hard-guided probe was then removed, and the gas-sterilized EUG electrode array was inserted into the catheter until it reached the fundus. The sheath catheter was pulled back whilst maintaining the electrode array in place and in direct contact with the uterine walls. The physician verified that the tip of the electrode array was abutting the fundus by comparing the length of the electrode array inserted into the endometrial cavity with the length from the fundus to the cervix, derived from a previous TVUS examination, as well as from a 7 cm fixed marking labeled on the electrode. Furthermore, when the electrode was outside the endometrial cavity, the EMG device monitor detected a high signal impedance (>50 kΩ) and stopped recording from that channel, ensuring proper electrode placement. The Pederson speculum was then removed, and the electrode array terminal was connected to the recording device. The EUG recording was not started for three minutes to allow possible mechanically induced contractions to subside. The participants were questioned to verify that they did not feel any uterine contractions/cramps or any discomfort before initializing the EUG recording. No pain, discomfort, or bleeding were observed before starting the procedure. The electrode was compacted between the anterior and posterior uterine walls and therefore remained stable throughout the procedure. Artefacts were minimized by asking patients to remain immobile during the recordings. Movement artefacts were subsequently easily distinguished. After the end of the recording, the electrode array was removed from the uterine cavity. Figure 1 shows a TVUS recording after the electrode array placement inside the uterine cavity, with the mapping between the electrode array and its US image. No adverse events were identified or reported by the participants during or after the procedure.

2.5. Measurement Protocols and Signal Processing

Quantitative variables were measured in the following way. In this study, the menstrual cycle was separated into four distinct phases. The early proliferative (EP) phase, which corresponds to days 1–7, the late proliferative (LP) phase, which corresponds to days 8–14, the early luteal (EL) phase, which corresponds to days 15–21, and the late luteal (LL) phase, which corresponds to days 22–28 of the menstrual cycle. Attention and signal processing were focused on the electrical activity derived by the monopolar signal of the electrode 1 cm from the fundal region, as a recent study found that placing the embryo within one or two centimeters of the fundus significantly increased the pregnancy rate [12]. Recordings were separated into 5 min segments, with a maximum of 2 segments per recording. The electrical signal was lowpass-filtered, at 0.1 Hz, which cuts off noise signals corresponding to more than 6 cycles per minute, thereby limiting signals to frequency bands corresponding to the previously observed frequency of uterine contractions across the menstrual cycle [4,13]. The following outcomes were extracted and measured from the filtered recordings: the Root-Mean-Square (RMS) Voltage Amplitude (VJZE-RMS) and the Mean Frequency (fJZE-mean) of the electrical activity recorded from within the endometrium, as direct indicators of the intensity and frequency changes in the uterine contractions.

2.6. Statistical Analysis

The Kruskal–Wallis test was used for identifying whether there were statistically significant differences among the four compared phases of the menstrual cycle (EP, LP, EL, and LL) for each of the associated features (VJZE-RMS, fJZE-mean). Then, the Dunn–Sidak approach was performed for post hoc analysis. p < 0.05 was chosen to show the statistical significance. All results are displayed as median values and interquartile ranges (IQRs) in parentheses.

3. Results

Regarding the VJZE-RMS, there is a clear trend in its median values, decreasing from the early proliferative phase to the late proliferative phase (from (247(91) μV to 158(252) μV)), where it reached its minimum value. During the early luteal phase, an upward trend was observed (from (158(252) μV to 374(397) μV, p < 0.05)), reaching its highest value during the late luteal phase (from (374(397) μV to 477(304) μV, p < 0.05, when compared to the late proliferative phase). The fJZE-mean clearly shows exactly the opposite trend, increasing from 2.5(0.54) cycles/min during the early proliferative phase to 2.96(0.97) cycles/min during the late proliferative phase, reaching its maximum value. During the early luteal phase, a downward trend was observed (from 2.96(0.97) cycles/min to 2.37(1.05) cycles/min), continuing to fall to its lowest value of 1.33(0.55) cycles/min in the late luteal phase, with p < 0.05 when compared to the previous three phases. Boxplots of the derived features across the menstrual cycle are shown in Figure 2.

The VJZE-RMS divided by a factor of ten closely follows the Intrauterine Pressure (IUP) values, creating an electrically derived IUP index. The same situation applies between the uterine contractility shown both with an IUP device and with US and the fJZE-mean, without the need to multiply or divide with a constant factor, for three of the four phases of the menstrual cycle, creating an electrically derived uterine contractility index. The comparison bar plots are shown in Figure 3.

The EUG demonstrated the detection of changes in the peristaltic activity among the phases of the menstrual cycle that coincide with the hormonal changes of estrogen and progesterone. A qualitative representation comparing the measured electrical activity using the EUG and the change in the hormonal levels across the menstrual cycle is shown in Figure 4.

Complications

Twenty-one women experienced no pain during the ten minutes of the procedure and scored zero on the visual pain score. Five patients complained of light menstrual-like pain for one–two minutes after the insertion of the electrode, scoring two on the visual pain score for the first two minutes and then zero for the rest of the procedure.

4. Discussion

In this paper, we test the hypothesis that the proposed EUG can be used for monitoring the uterine contractility across the menstrual cycle by measuring the electrical activity of the JZE in healthy, parous, nonpregnant women. To summarize, waveforms closer to the fundus were processed. The RMS Voltage Amplitude (VJZE-RMS) (in μV) and the Mean Frequency (fJZE-mean) (in cycles/min) of the JZE’s electrical activity, as direct indicators of the intensity and frequency changes in the JZE’s contractions, were extracted from the recorded waveforms. There was a trend in the median values of the VJZE-RMS decreasing from the EP to the LP phase. During the EL phase, an upward trend was observed, reaching its highest value during the LL phase when compared to the LP phase. The fJZE-mean showed the opposite trend, increasing from the EP to the LP phase. During the EL phase, a downward trend was observed, which continued to fall in the LL phase, when compared to the previous three phases.

In a nongravid uterus, the innermost layer of the myometrium, the JZE, is the main source of peristaltic activity across the menstrual cycle. The endometrial stem/progenitor cells lie in the basalis layer of the endometrium next to the myometrium and the junctional zone. Myometrial cells are niche cells, regulating the activities of endometrial mesenchymal stem-like cells (eMSCs) with self-renewal activity, influenced by cyclic estrogens and progesterone secretion [14]. The uterus derives its innervation from the lumbar plexus. The lumbar plexus is formed from the nerve roots L1–L4 and gives rise to the iliohypogastric, ilioinguinal, genitofemoral, femoral, and obturator nerves. As such, these nerves are important to consider when describing the contractility, the neurophysiology, and possibly the pain pathology of the uterus [15]. Previous studies have determined that the endometrial waves are initiated in the sub-endometrial myometrium, but the link between the innervation of the uterus and this contractility mechanism is still not clear [16,17]. The estrogens affect the central nervous system and, together with progesterone, play a role in neurotransmission and act as neuromodulators in the uterus. The estrogens accelerate the uterine contractility, while the progesterone decreases the uterine contractility, creating a quiescent state for assisting successful embryo placement [16,18]. Based on the anatomy and neurophysiology described, the placement position of the EUG was important, as it enabled the capture of the most representative signals. By only having a few single layers of cells between the electrode and the JZE, the signals were interpreted as contractions with an origin in the JZE, confirming its ability to assist in the monitoring of the contractive uterus. Future research could subsequently focus on the EUG’s ability to assist in decision making in, for instance, monitoring the optimal day for embryo transfer and/or assisting in personalized hormonal treatments during IVF.

In parous, healthy women with normal uterine and ovarian sonographic characteristics, the fJZE-mean showed statistically significant lower values for the LL phase, compared to the other three phases of the menstrual cycle. The fJZE-mean measures the frequency content of the JZE’s electrical activity, which is closely related to the uterine contractility. The results agree with the previously observed contractility trends of Bulletin et al. (2000), who successfully demonstrated the emphasis on the hormonal dependence in uterine contractility [8]. The EUG also successfully distinguished the different phases of the menstrual cycle that were associated with the JZE’s contractility. The VJZE-RMS, which correlates with the uterine contraction intensity, is grounded in the logical hypothesis that more synchronized depolarization in a larger volume of muscular tissue leads to higher voltage amplitudes. Statistically significant lower values for the LP phase of the menstrual cycle, in comparison to the EL and LL phases, align with findings in previous literature [8]. When directly comparing our results with those of Bulletti et al., 2000, we noted that the VJZE-RMS, divided by a factor of ten, closely mirrors the Intrauterine Pressure (IUP) values, establishing an electrically derived IUP index. Similarly, the electrically derived uterine contractility index aligns with the uterine contractility measured with an IUP device and ultrasound (US) for three of the four menstrual cycle phases, without the need for constant multiplication or division. Figure 3 presents the comparison bar plots.

For the nonpregnant uterus, only the JZE is responsible for the uterine contractions, while the two outer layers of the myometrium are responsible for the contractions during labor [4,19,20]. The contractile activity during menstruation, when the muscles are attempting to expel the endometrial tissue, resembles the behavior of the two outer layers during parturition [21]. Using external electrodes in a feasibility study for nonpregnant women, statistically significant differences were only found between menstruation and at least one of the LP, EL, and LL phases, but not among the non-menstruating phases [10]. Hence, it can be argued that external electrodes can only track the difference in contractility between the inner one-third of the myometrium and the outer two-thirds, which have significantly different characteristics. Regarding the contractility characteristics of the JZE, it is considered to have a finer and more wavelike contractility pattern, which is called peristalsis and more closely resembles the peristaltic activity of bowel movements [2].

The peristaltic activity of the uterine muscles during the menstrual cycle has also been correlated with the hormonal levels for nonpregnant women. After menstruation, the estrogen levels start to increase, showing a peak during days 8–12. The observed peak agrees with the peak of the uterine contractions, which, according to the literature, plays a significant role in the facilitation of sperm transport [2,8,22]. On the contrary, after ovulation, the estrogen levels show a downward trend and, at the same time, the progesterone levels show an upward trend, with a peak around seven days after ovulation. Furthermore, after ovulation and until the end of the menstrual cycle, the number of intrauterine contractions decreases compared to the periovulatory period, showing that progesterone inhibits the excessive peristalsis of the uterus, even though a second smaller estrogen peak is observed on around the seventh day after ovulation. Hence, it is obvious that progesterone dominates the peristaltic activity of the nonpregnant uterus during the luteal phase, creating a quiescent environment, which facilitates easier embryo implantation to the endometrium [2,8,22]. The EUG demonstrated the detection of changes in the peristaltic activity among the phases of the menstrual cycle that coincide with the hormonal changes of estrogen and progesterone. Furthermore, in a study measuring the electrical activity of extra-corporeal-perfused human uteri, the frequency of the rhythmic electrical activity was increased with the administration of a dose of 17β-estradiol and was reduced with the administration of a high dose of progesterone [23]. These results further support our observations regarding the frequency changes in the JZE’s electrical activity and provide an additional electrochemical validation of our results.

When comparing the frequency- and amplitude-related features, the amplitude-related features are dependent on the impedance of the recording area. The impedance can possibly vary depending on the volume and ionic concentrations of the fluid inside the uterine cavity and the conductivity of the endometrial tissue, which might be subject to changes from participant to participant and during the menstrual cycle. This interparticipant variability is assumed to be the most likely reason for the high variability in the VJZE-RMS. The frequency-related features are more robust to electrical measurements. However, the variability is still considerable, due to interparticipant changes. It is important to note that the clear interparticipant variability trends and statistically significant differences across the menstrual cycle observed regarding the frequency and voltage amplitude of the electrical activity agree with the previously aforementioned findings [8]. In future studies, we would like to investigate how obesity and menstruation change the EUG features, as, in this study, we had only one participant with a BMI > 30 and only one recording during menstruation. Furthermore, we would like to investigate whether there is a difference in the EUG values between parous and nulliparous women. However, this study would require more strict and complicated selection criteria to minimize the risk of recruiting nulliparous women with infertility problems.

Uterine peristalsis is an important function of nonpregnant women, for which deviations from normal could increase the risk of gynecological pathologies and infertility. The existing tools for recording uterine peristalsis are intricate, costly, and lack reproducibility. This study highlights that the EUG presents a standardized, minimally invasive, safe, and pain-free procedure, offering reproducible electrical measurements from the endometrial cavity [11]. The clinical assessment in this article focuses on the EUG’s ability to monitor the uterine contractility across the menstrual cycle in healthy, parous, nonpregnant women. An analysis of the recordings indicates that the measured electrical activity aligns with previously observed patterns of uterine contractions throughout the menstrual cycle found in the literature [24,25,26]. When comparing the JZE peristalsis measurements taken using TVUS and MRI, the EUG measurements are independent of the operator and directly contact the JZE. Recognizing the significance of uterine contractility in nonpregnant women, the EUG shows promise for assessing the uterine physiology. Further prospective studies in multi-center settings will enhance our understanding of the clinical applications of the EUG.

Limitations

The present study has some limitations that need to be taken into consideration. First, the sample size comprised only 26 healthy reproductive-age women, potentially limiting the generalizability of the findings. Replicating the study with a more extensive and diverse population would enhance the external validity and robustness of the results. Second, while the research identifies differences in the electrical activity across the menstrual cycle sub-phases, it falls short of establishing direct links to specific conditions. Future investigations could delve deeper into the clinical relevance and implications of these variations in uterine electrical activity. Third, the insertion of a flexible electrode array into the endometrial cavity for EUG recordings raises concerns about invasiveness. Despite the positive feedback received from the participants for this painless procedure, evaluating the safety and acceptability of this procedure is important for its potential adoption. Additionally, as a single-center study, the research’s applicability to diverse settings may be limited. Conducting replications in multiple centers would strengthen the generalizability and reliability of the findings. To overcome these limitations, subsequent investigations should emphasize larger and more diverse cohorts, establish clinical correlations, incorporate long-term monitoring, and thoroughly assess the acceptability of the EUG procedure across different populations.

5. Conclusions

Research demonstrates that the EUG can effectively record JZE contractility throughout the menstrual cycle in a safe, reliable, and painless manner. The derived contractility from the EUG aligns with the existing literature exploring electrical, mechanical, and chemical changes across the menstrual cycle. Our study on the nongravid uterus extends the myocytes’ synchronicity model, traditionally applied to the gravid uterus. The proximity of the EUG electrodes to the JZE enables the direct recording and monitoring of the uterine electrical activity, capturing signals undetectable with abdominal electrode arrays. Importantly, our findings reveal, for the first time, electrical activity differences in various sub-phases of the menstrual cycle. Consequently, the innovative EUG tool holds promise for assessing the uterine physiology.