Pelvic floor myoelectric excitation and force output across menstrual-cycle phases

Abstract

Objective

To determine whether menstrual-cycle (MC) phase is associated with changes in pelvic floor muscle (PFM) function, focusing on myoelectric excitation (intravaginal sEMG) at rest and during standardized contraction tasks, and on mechanical strength assessed by dynamometry and digital palpation.

Methods

Prospective observational, within-subject study in 23 healthy nulliparous eumenorrheic women (18–35 years) not using hormonal contraceptives. Each participant was assessed in early follicular (EFP), ovulatory (OP), and mid-luteal (MLP) phases, confirmed using urinary luteinizing hormone (LH) testing, basal temperature, and transvaginal ultrasonography. Outcomes included Modified Oxford Scale (MOS), Pelvibex^® dynamometry (passive, active, net force), and intravaginal sEMG (MVC plus Glazer protocol: baseline, phasic, tonic, endurance, baseline post-endurance). Friedman tests with Holm-adjusted post hoc comparisons and Spearman correlations were applied.

Results

Voluntary contraction sEMG (MVC, phasic, tonic, endurance) showed no significant phase differences (p > 0.05). Resting sEMG differed by phase: baseline average and peak amplitude was lower in OP vs. EFP (pHolm = 0.012; pHolm = 0.018, respectively). Strength outcomes (MOS and dynamometric measures) did not differ across phases (all p > 0.05). sEMG exhibited strong within-phase coherence and cross-phase stability, while baseline myoelectric excitation correlated inversely with force measures.

Conclusions

MC phase was associated with subtle modulation of baseline PFM myoelectric excitation (most consistently lower at ovulation), without detectable phase-related changes during voluntary contractions or in mechanical strength. Baseline sEMG amplitude showed inverse associations with active force, suggesting that higher residual excitation may be associated with reduced mechanical strength. Overall, electrophysiological and force-based outcomes should be interpreted cautiously.

Introduction

The pelvic floor (PF) is a complex musculoaponeurotic system located in the inferior pelvis. It comprises skeletal muscles (most prominently the levator ani and coccygeus) together with connective tissue and fascial components. This system provides baseline tone that supports and maintains the position of the pelvic organs, ensures continence via sphincter control, contributes to sexual function, and promotes lumbopelvic stability during movement [1, 2]. Pelvic floor muscle (PFM) tone reflects the interaction between active contractile elements and passive connective-tissue structures. The active component arises from muscle fibers that sustain a resting level of activation, whereas the passive component derives from connective tissue that provides elastic recoil and structural support [3]. Together, these elements maintain a pre-tensioned state that underpins PF function [4]. Rather than functioning as a simple tonic contraction between two fixed skeletal segments, PFM are anchored through a connective-tissue network, including the perineal membrane, the perineal body (tendinous perineal centre), and the tendinous arch of the levator ani, which enables efficient load transfer and functional closure [5]. This structural complexity underscores the importance of visceral support and highlights the risk of PF dysfunctions when baseline tone, connective-tissue tension, or muscle strength decreases [6, 7].

Female sex-related differences due to the menstrual cycle (MC) are associated with cyclical fluctuations in endogenous sex hormones, primarily estrogen and progesterone [8]. The MC is divided into a follicular phase and luteal phase, separated by ovulation [8]. During the follicular phase, estrogen concentrations rise progressively to a preovulatory peak triggered by the luteinizing hormone (LH) surge [9, 10] accompanied by a transient increase rise in testosterone that serves as a substrate for estradiol synthesis [11]. After ovulation, estrogen declines and progesterone increases, with both hormones remaining relatively elevated in the mid-luteal phase before falling toward menstruation. These hormones not only control reproductive functions and sexual characteristics, but also impact various biological systems [12, 13]. Accordingly, hormonal variation across the MC has been proposed to influence aspects of women’s physical and physiological performance, potentially contributing to fluctuations in perceived energy, mood, and physical capacity [14–17], however, recent syntheses indicate that phase-related effects on exercise performance are generally trivial at the group level and highly variable between individuals [18].

Physiological evidence also suggests that neuromuscular and connective-tissue function may be sensitive to estrogen and progesterone, which could modulate tissue elasticity, muscle tone, and neuromuscular activation across the MC [19, 20]. In this context, estrogen appears to exert anabolic actions in skeletal muscle [21] and may support gains in muscle mass and responsiveness to strength training [22, 23]. The effects of menstrual hormonal fluctuations on connective tissue have also been investigated [24], including studies assessing anterior cruciate ligament laxity [25] and the plantar fascia properties [26]; nevertheless, findings remain inconsistent. Conversely, in chronic low-estrogen states (e.g., menopause or experimental ovariectomy), more durable alterations have been described in both muscle and connective tissue [27, 28]. In postmenopausal women, estrogen status has been linked to neuromuscular function (e.g., strength and power), although findings are heterogeneous, and estradiol replacement has been associated with differences in tendon collagen turnover and biomechanical properties compared with non-users [29]; consistent with this, ovariectomized animal models show declines in muscle size/function that are attenuated by estradiol replacement [30].

At the PF level, findings remain mixed. Menopausal age has been linked to lower clinically graded contractility, whereas ultrasound-based functional markers may not differ between pre- and postmenopausal women, suggesting that estrogen deprivation may not independently predict contractility [31]. PFM training improves strength in postmenopausal women, with some evidence of greater gains in those not receiving hormone replacement therapy, implying that the hormonal milieu could modulate training adaptation [32]. Across the MC, data on PFM strength and neuromuscular control are still limited and heterogeneous. Surface electromyography (sEMG) findings indicate that maximal PFM activation may not differ substantially between MC phases, whereas resting activity appears to fluctuate (typically lower around the follicular/ovulatory phases and higher during the luteal phase) suggesting a potential hormonal modulation of baseline activity [33]. More recent transperineal ultrasound data collected during high-load tasks have reported subphase-related differences in PF tissue distensibility, implying that hormonal status may influence the mechanical response of the PF under demanding conditions even when intra-abdominal pressure is comparable [34].

In a prior study, we assessed PF strength across different phases of the MC using dynamometry and observed no significant differences [35]. However, dynamometry reflects global force output and may not capture phase-related differences in myoelectric activity (i.e., sEMG amplitude) at rest and during voluntary contractions, or other support-related mechanisms contributing to pelvic support. Therefore, the primary aim of the present study was to examine whether MC-related hormonal variation is associated with changes in PF function across distinct phases, focusing on PFM myoelectric activity (resting and contraction tasks) and on force output assessed by dynamometry and digital palpation. As a secondary aim, we examined the associations between force outcomes and complementary measures to provide a more comprehensive characterization of PF function.

Materials and methods

Design

This was a prospective, cross-sectional observational study that assessed the activity of the PFM during three different phases of the menstrual cycle in eumenorrheic women, using digital palpation, dynamometry and electromyography (sEMG). The study adhered to the STROBE guidelines for observational studies [36].

Participants

The inclusion criteria were healthy nulliparous women aged 18–35 years; no use of hormonal contraceptives within the previous 6 months; and regular menstrual cycles, defined as a cycle-length variation of < 3 days over the previous 4 months. Participants were excluded if they had a menstrual disorder, were pregnant, had urgent urinary incontinence, active or recurrent genitourinary infections, polycystyc ovary syndrome, phobias or conditions preventing vaginal probe insertion, a history of urogynecological surgery, or degenerative neurological diseases or if they were unable to complete testing on the prespecified cycle days.

Before the beginning of the experimental protocol, participants were fully informed of the experimental procedures and signed an informed written consent to participate in the investigation. The study was approved by the Camilo José Cela University Research Ethics Committee (04_23_CISP). All the research protocols were in accordance with the latest version of the Declaration of Helsinki.

Instrumentation and data collection

Verification of menstrual cycle phase and data collection

After enrolment, participants completed one trial in each of the following 3 phases of their individually characterized menstrual cycle: (i) early follicular (EFP); (ii) ovulatory (OP); and (iii) mid-luteal (MLP) phase. These 3 phases were chosen because they represent menses, peak serum oestrogen concentration and peak serum progesterone concentration, respectively [37].Each participant’s regularity and length of the menstrual cycle (27 ± 2 days, interindividual range from 24 to 31 days) were monitored for 4 months prior to the investigation using a mobile application (Mycalendar, Period-tracker, USA). Each phase of the menstrual cycle was individually confirmed based on specific physiological markers characteristic of that stage. The EF phase was assessed beginning the day after the initial appearance of menstrual bleeding. The OP phase was identified the day following a surge in urinary luteinizing hormone (LH) [25], as detected by test strips (One Step Ovulation LH Test Strip; CVS Corporation, US), and further validated by a rise in basal tympanic temperature (HDT8208C model, Nursal Ear Thermometer, China). For the MLP, the trial was scheduled on a day corresponding to 70% of the individual’s cycle duration—for example, day 20 in a 28-day cycle.

A preliminary gynecological evaluation was performed to exclude any criteria and confirm the MC phase using transvaginal ultrasonography. This method is preferred for examining ovarian volume, antral follicles, and endometrium thickness. This multi-method approach was used to ensure that assessments were performed in the intended phases despite inter-individual variability in cycle length and ovulation timing. The starting menstruation phase was randomized and counterbalanced. To achieve this the first phase of the investigated menstrual cycle was randomly assigned, and a similar number of participants started in EFP (8 participants), OP (8 participants) and MLP (7 participants).

Participants provided their anthropometric information (including age, weight, height, body mass index - BMI, and percentage of body fat) and medical history (including common diseases, urinary infections, and gynecological and obstetric history) through an online platform.

Pelvic floor muscle assessment

Each participant, with an empty bladder, was positioned supine on the stretcher with a pillow under the head. The knees were gently flexed, supported by a roller under the knees, and the lumbar spine was in a physiological position. All measurements were conducted, after a 10-minute rest period in the supine position, by the same physiotherapist with more than 10 years of clinical experience in women’s health.

The Modified Oxford Scale (MOS) was utilized to assess PFM contraction via internal vaginal palpation [38], whereby one or two fingers were inserted into the vaginal canal and participants were instructed to perform a maximal inward-and-upward voluntary contraction to evaluate vaginal closing pressure and overall contractile capacity. Contraction intensity was graded from 0 (no palpable activity) to 5 (strong contraction), reflecting both vaginal wall constriction and lift. If the contraction was inadequate, biofeedback was provided to refine technique while avoiding accessory contractions. The strength of PFMs was assessed using a clinical dynamometer (Pelvibex^®) and measured in newtons (N), reproduced from the previous study [35]. The device included an adjustable vaginal speculum with an inductive displacement sensor. Measurements of baseline strength (passive force) and contraction strength (maximal voluntary contraction -MVC-) were taken, and two consecutive measurements were recorded with a 30-second rest period between them. The strength of PFM contraction was calculated as the difference between contraction and baseline strength [39]. To elicit the MVC, participants were instructed as follows: “Squeeze your anus as if you were trying to hold back an intense urge to urinate or pass gas and draw it inward and upward as forcefully as possible, without bearing down” [40].

For the sEMG measurement, we utilized the mDurance^® system (Granada, Spain) together with the the Perisize 4^® vaginal probe, following the same setup described in a previous study [41]. The vaginal probe was inserted by an experienced physiotherapist using conductive gel, with the electrodes placed in the 3 and 9 o’clock positions, and a reference electrode was positioned on the anterior superior iliac spine in accordance with SENIAM recommendations [42]. The women performed the MVC of the PFM twice, with 30-second relaxation period between attempts, and completed the five activities in the Glazer protocol [43]: one minute rest (baseline), where patients were instructed to feel the PF in resting position; five 2 s phasic contractions, with a 10 s rest between (phasic); five 10 s tonic contraction, with a 10 s rest in between (tonic); 1 min endurance contraction (endurance) and 1 min rest (baseline post-endurance). The order of Glazer protocol activities was counterbalanced randomly to minimize sequence and order effects.

Peak amplitude was the highest value of the two MVC, in microvolts (µV), which was the reference value set as 100% for sEMG. Glazer protocol measurements were taken, and values ​​were represented in µV and percentages, according to MVC data. Peak and average amplitude results for each activity were compared to the peak and average amplitude of the MVC respectively.

Statistical analysis

The sample size was estimated a priori using G*Power (version 3.1.9.6, Kiel University, Kiel, Germany) for a within-subject (repeated-measures) design comparing the three menstrual-cycle phases. The primary outcome for this calculation was the peak VMC of PF sEMG amplitude. Assuming a two-tailed α of 0.05, desired power of 0.99, and a large expected within-subject effect (equivalent to Cohen’s d = 0.80), the required sample was 23 participants. This target sample size was also consistent with previous studies using comparable pelvic floor or sEMG-related outcomes and repeated assessments in women [44–46].

For data analysis, researchers used IBM SPSS Statistics, version 26 (IBM Corp, New York, NY, USA). The data were described using mean and standard deviation, with a 95% confidence interval (95% CI) and additionally as median (P25–P75, 25th–75th percentile). Normality was tested with Shapiro–Wilk tests. The Friedman test was used to compare dynamometric values and sEMG measurements across different menstrual phases. When pairwise comparisons were required, Wilcoxon signed-rank tests were conducted for the three planned contrasts (follicular vs. ovulation, follicular vs. luteal, and ovulation vs. luteal), and p-values were adjusted for multiple testing using the Holm–Bonferroni procedure. Holm correction was applied by ordering the three p-values from smallest to largest and sequentially comparing each to α/(m − i + 1), with m = 3 comparisons. Spearman’s correlation coefficient was calculated to assess relationships among quantitative variables. Potential order and sequence effects arising from the counterbalanced Glazer protocol were tested using the Kruskal–Wallis test. The confidence level was set at 95%, with statistical significance defined as p ≤ 0.05. The confidence level was set at 95%, with statistical significance defined as p ≤ 0.05.

Results

Twenty-seven women were initially recruited; however, four were excluded because they did not complete all three assessment sessions. The final sample comprised 23 women (mean age 25.09 ± 4.83 years) with a mean BMI of 21.86 ± 3.07 kg/m², consistent with a healthy weight range.

Across menstrual-cycle phases, average sEMG amplitude during voluntary contractions did not differ for MVC, phasic, tonic, or endurance tasks (p > 0.05) (Table 1). During resting recordings, however, a phase effect was observed for baseline average amplitude (χ²(2) = 10.174; p = 0.006), with lower values in OP compared with EFP (pHolm = 0.012), while EFP–MLP and OP–MLP comparisons were not significant (pHolm = 0.330 and 0.146, respectively). Outcomes were not influenced by the order of menstrual-cycle testing sessions or by the order of tasks within the Glazer protocol (all p > 0.05).

Table 1.

Average sEMG amplitude (µV) across menstrual-cycle phases

Average amplitude (µV)		EFP	OP	MLP	χ²(2)	p	pHolm (EFP–OP)	pHolm (EFP–MLP)	pHolm (OP–MLP)
MVC (2 × 5 s)	Mean (SD)	57.50 ± 25.37	58.18 ± 46.23	63.27 ± 47.24	2.435	0.296	0.424	0.879	0.249
	Median (P25–P75)	48.23 (32.84–78.18)	43.07 (31.73–67.89)	50.87 (37.21–78.84)
Baseline (rest 60 s)	Mean (SD)	12.26 ± 6.30	9.59 ± 6.13	11.12 ± 5.49	10.174	0.006*	0.012†	0.330	0.146
	Median (P25–P75)	11.66 (6.96–17.54)	6.71 (4.54–15.36)	9.15 (6.80–16.17)
Phasic (5 × 2 s)	Mean (SD)	59.54 ± 37.17	50.98 ± 31.47	55.19 ± 34.57	1.652	0.438	0.093	0.534	0.534
	Median (P25–P75)	48.07 (35.08–71.87)	47.39 (33.66–57.37)	44.03 (36.91–71.86)
Tonic (5 × 10 s)	Mean (SD)	61.26 ± 43.63	56.27 ± 41.61	59.42 ± 44.83	2.696	0.260	0.636	0.738	0.636
	Median (P25–P75)	53.27 (35.84–75.52)	42.11 (32.10–67.62)	45.58 (34.97–81.47)
Endurance (1 × 60 s)	Mean (SD)	54.20 ± 46.30	47.08 ± 31.75	52.54 ± 36.77	4.957	0.084	0.574	0.584	0.093
	Median (P25–P75)	44.95 (29.26–59.57)	37.81 (27.24–57.36)	42.63 (33.38–63.44)
Baseline post-endurance (rest 60 s)	Mean (SD)	13.21 ± 4.83	11.55 ± 5.47	12.68 ± 6.35	0.348	0.840	0.099	0.858	0.858
	Median (P25–P75)	12.82 (9.12–16.97)	12.37 (6.61–15.33)	13.62 (7.07–18.76)

p values from Friedman test; post-hoc pairwise comparisons by Wilcoxon signed-rank test with Holm adjustment. * p < 0.05 (Friedman). † pHolm < 0.05. s seconds, SD Standard deviation, P25–P75, 25th–75th percentile, EFP Early follicular phase, OP Ovulatory phase, MLP Mid luteal phase, MVC maximum voluntary contraction

Regarding peak sEMG amplitude, no significant phase effects were found for phasic, tonic, or endurance tasks (p > 0.05) (Table 2). In contrast, resting peak amplitude showed a significant phase effect at baseline (χ²(2) = 8.435; p = 0.015), with lower peaks in OP vs. EFP (pHolm = 0.018). Additionally, baseline post-endurance peak amplitude showed no overall phase effect (χ²(2) = 2.696; p = 0.260), but pairwise analysis indicated lower OP vs. EFP (pHolm = 0.039). For MVC peak amplitude, the overall Friedman test was not significant (χ²(2) = 5.478; p = 0.065), although the OP–MLP comparison remained significant after Holm correction (pHolm = 0.021) (Fig. 1).

Table 2.

Peak sEMG amplitude (µV) across menstrual-cycle phases

Peak amplitude (µV)		EFP	OP	MLP	χ²(2)	p	pHolm (EFP–OP)	pHolm (EFP–MLP)	pHolm (OP–MLP)
MVC (2 × 5 s)	Mean (SD)	93.38 ± 40.17	83.82 ± 60.92	97.24 ± 64.87	5.478	0.065	0.084	0.715	0.021†
	Median (P25–P75)	77.70 (59.55–119.68)	72.22 (47.49–100.76)	79.25 (56.78–120.21)
Baseline (rest 60 s)	Mean (SD)	20.37 ± 10.51	15.88 ± 10.22	18.48 ± 8.77	8.435	0.015*	0.018†	0.287	0.118
	Median (P25–P75)	18.59 (11.36–31.02)	11.62 (7.38–25.63)	15.28 (11.62–27.98)
Phasic (5 × 2 s)	Mean (SD)	90.46 ± 57.03	77.89 ± 48.74	83.83 ± 51.44	4.174	0.124	0.057	0.242	0.465
	Median (P25–P75)	74.11 (54.19–105.83)	69.03 (50.40–89.49)	68.96 (52.26–109.48)
Tonic (5 × 10 s)	Mean (SD)	100.13 ± 66.37	91.90 ± 65.57	96.93 ± 68.14	4.261	0.119	0.219	0.951	0.520
	Median (P25–P75)	82.59 (62.79–120.64)	75.43 (51.49–122.51)	76.55 (61.33–123.72)
Endurance (1 × 60 s)	Mean (SD)	98.70 ± 51.61	94.13 ± 65.34	104.47 ± 74.31	1.652	0.438	0.602	0.602	0.342
	Median (P25–P75)	85.06 (58.94–115.47)	74.01 (56.52–111.83)	83.23 (64.55–127.79)
Baseline post-endurance (rest 60 s)	Mean (SD)	24.04 ± 8.50	20.27 ± 8.79	23.13 ± 11.10	2.696	0.260	0.039†	0.903	0.249
	Median (P25–P75)	23.71 (16.42–29.11)	21.71 (11.91–26.42)	25.26 (11.11–33.03)

p values from Friedman test; post-hoc pairwise comparisons by Wilcoxon signed-rank test with Holm adjustment. * p < 0.05 (Friedman). † pHolm < 0.05. s seconds, SD Standard deviation; P25–P75, 25th–75th percentile, EFP Early follicular phase, OP Ovulatory phase, MLP Mid luteal phase, MVC maximum voluntary contraction

Fig. 1.

Representation of PFM measurements according to the MC phase, dynamometry measurements with values ​​in newtons (N) and electromyography measurements with values ​​in microvolts (µV)

Regarding normalized sEMG expressed as percentage, average amplitude differed across phases only for the 60-s resting baseline condition (χ²(2) = 6.870; p = 0.032), with lower myoelectric excitation in the OP compared with the EFP (pHolm = 0.042) and showing a tendency toward lower values compared with the MLP (pHolm = 0.073). No phase differences were observed for phasic, tonic, endurance, or baseline post-endurance average amplitude (all p > 0.05). Peak values showed no significant phase differences across tasks (p > 0.05), although a trend was observed for phasic peak amplitude (χ²(2) = 5.478; p = 0.065) (Table 3).

Table 3.

Percentages of average and peak amplitudes across menstrual-cycle phases

Average amplitude (%)		EFP	OP	MLP	p
Baseline (rest 60 s)	Mean (SD)	24.33 ± 15.66	21.61 ± 15.49	23.56 ± 14.96	0.032*
	Median (P25–P75)	20.93 (13.45–33.37)	15.42 (11.28–30.40)	18.87 (16.19–29.78)
Phasic (5 × 2 s)	Mean (SD)	104.07 ± 40.27	92.73 ± 12.93	99.32 ± 53.47	0.119
	Median (P25–P75)	96.01 (89.11–103.79)	90.12 (85.27–99.63)	89.24 (82.41–100.50)
Tonic (5 × 10 s)	Mean (SD)	105.17 ± 49.80	98.15 ± 14.75	93.74 ± 11.83	0.499
	Median (P25–P75)	92.45 (86.43–104.06)	96.15 (89.03–106.27)	92.45 (86.04–98.50)
Endurance (1 × 60 s)	Mean (SD)	93.63 ± 56.84	83.59 ± 14.74	84.70 ± 17.70	0.183
	Median (P25–P75)	86.22 (72.83–94.05)	82.10 (75.86–88.62)	85.19 (67.91–96.80)
Baseline post-endurance (rest 60 s)	Mean (SD)	26.58 ± 14.54	24.07 ± 12.20	24.46 ± 14.13	0.438
	Median (P25–P75)	24.21 (17.41–29.54)	20.50 (16.94–30.79)	23.80 (13.90–33.55)

Peak amplitude (%)		EFP	OP	MLP	p
Baseline (rest 60 s)	Mean (SD)	24.46 ± 14.45	23.03 ± 14.46	24.24 ± 15.20	0.568
	Median (P25–P75)	23.24 (13.58–33.23)	18.82 (12.07–32.88)	21.43 (16.43–29.73)
Phasic (5 × 2 s)	Mean (SD)	95.15 ± 23.38	97.14 ± 20.11	84.49 ± 10.37	0.065
	Median (P25–P75)	89.89 (82.88–97.10)	93.25 (82.46–109.64)	85.39 (78.17–92.04)
Tonic (5 × 10 s)	Mean (SD)	103.94 ± 28.73	112.62 ± 32.46	98.90 ± 13.49	0.183
	Median (P25–P75)	99.67 (92.03–110.94)	104.93 (96.54–115.63)	97.55 (91.24–103.13)
Endurance (1 × 60 s)	Mean (SD)	104.20 ± 18.41	117.48 ± 42.21	106.88 ± 22.35	0.568
	Median (P25–P75)	102.19 (95.48–112.14)	101.21 (95.84–122.43)	100.26 (89.80–117.83)
Baseline post-endurance (rest 60 s)	Mean (SD)	28.63 ± 12.47	28.21 ± 11.14	27.53 ± 15.09	0.676
	Median (P25–P75)	25.59 (19.45–37.46)	28.24 (19.29–34.00)	25.76 (16.10–35.56)

p values from Friedman test. * p < 0.05 (Friedman). Values are expressed as a normalised ratio (%), calculated relative to the MVC (reference) obtained in the supine position. s, seconds, SD Standard deviation, P25–P75, 25th–75th percentile, EFP Early follicular phase, OP Ovulatory phase, MLP Mid luteal phase, MVC maximum voluntary contraction

Strength outcomes did not differ across menstrual-cycle phases (Table 4) (Fig. 1). No phase-related differences were observed in MOS scores (p = 0.449) or in dynamometric force measures (p > 0.05).

Table 4.

PFM strength across menstrual-cycle phases

		EFP	OP	MLP	p
Strength (Modified Oxford Scale)	Mean (SD)	3.70 (0.47)	3.70 (0.56)	3.61 (0.58)	0.449
	Median (P25–P75)	4.00 (3.00–4.00)	4.00 (3.00–4.00)	4.00 (3.00–4.00)
Baseline strength (N) 1st measurement	Mean (SD)	4.85 (0.10)	4.90 (0.17)	4.86 (0.29)	0.363
	Median (P25–P75)	4.90 (4.80–4.90)	4.90 (4.80–5.00)	4.90 (4.70–4.90)
Baseline strength (N) 2nd measurement	Mean (SD)	4.93 (0.15)	4.89 (0.20)	4.93 (0.43)	0.841
	Median (P25–P75)	4.90 (4.80–5.00)	4.90 (4.80–5.00)	4.90 (4.70–5.00)
Contraction strength (N) 1st measurement	Mean (SD)	5.29 (0.36)	5.38 (0.62)	5.43 (0.70)	0.694
	Median (P25–P75)	5.20 (5.00–5.50)	5.20 (5.00–5.60)	5.20 (5.00–5.70)
Contraction strength (N) 2nd measurement	Mean (SD)	5.43 (0.44)	5.45 (0.54)	5.48 (0.68)	0.492
	Median (P25–P75)	5.40 (5.10–5.70)	5.30 (5.10–5.80)	5.20 (5.00–5.60)
Strength (N) 1st measurement	Mean (SD)	0.44 (0.36)	0.48 (0.63)	0.57 (0.71)	0.920
	Median (P25–P75)	0.40 (0.20–0.70)	0.30 (0.00–0.60)	0.30 (0.10–1.00)
Strength (N) 2nd measurement	Mean (SD)	0.50 (0.49)	0.56 (0.56)	0.55 (0.67)	0.711
	Median (P25–P75)	0.30 (0.10–0.80)	0.30 (0.00–1.00)	0.30 (0.10–0.70)

p values from Friedman test, SD Standard deviation, P25–P75, 25th–75th percentile, EFP Early follicular phase, OP Ovulatory phase, MLP Mid luteal phase, N Newtons

PF sEMG outcomes showed high internal consistency and cross-phase stability: within each phase, MVC sEMG was strongly associated with phasic, tonic, and endurance contractions (EFP: rs = 0.932–0.956; OP: rs = 0.960–0.977; MLP: rs = 0.920–0.987; all p < 0.001), and baseline and post-endurance resting activity were strongly correlated (rs = 0.613–0.850, p ≤ 0.002). MOS scores correlated positively with contraction sEMG in selected comparisons (e.g., EFP MVC peak: rs = 0.641, p = 0.001; MLP phasic mean: rs = 0.721, p < 0.001), but negatively with % resting activity (baseline mean: EFP rs = − 0.513, p = 0.012; OP rs = − 0.698, p < 0.001; MLP rs = − 0.755, p < 0.001; baseline post-endurance mean: EFP rs = − 0.513, p = 0.012; OP rs = − 0.769, p < 0.001; MLP rs = − 0.584, p = 0.003). Active dynamometry and net voluntary force (active–passive) were negatively associated with baseline and post-endurance resting sEMG (e.g., rs = − 0.775, p < 0.001; rs = − 0.637, p = 0.001).

Discussion

This study examined whether MC–related hormonal variation is associated with phase-dependent changes in PF function, focusing first on PFM myoelectric excitation (sEMG amplitude) at rest and during voluntary contractions and, secondly, on force output assessed by dynamometry and digital palpation. Overall, mechanical strength outcomes remained stable across phases, with no significant differences in MOS scores or dynamometric measures between the phases. In contrast, sEMG outcomes suggested subtle, task-dependent modulation of PFM myoelectric excitation, most clearly under resting conditions and, to a lesser extent, in peak contraction measures, where OP values tended to be lower.

These findings align with prior PF evidence suggesting that MC-related modulation may be more apparent in resting (baseline) myoelectrical activity than in maximal voluntary activation. Micussi et al. [34] reported phase-related differences in resting PFM sEMG alongside no meaningful changes in MVC across the same MC phases, supporting the view that background excitation may be more sensitive to hormonal variation than peak voluntary output. Notably, they observed higher baseline activity in the MLP and lower values in the EFP; however, the starting phase was not randomized, and the apparent rise in baseline tone in the MLP could partly reflect learning, habituation, or improved task familiarity across sessions rather than endocrine effects per se. Consistent with this, Micussi also reported stable MVC outcomes across phases. In our study, resting excitation was lowest in OP relative to EFP, a directionality that differs from some reports and highlights that resting sEMG may be shaped by behavioural and methodological influences (e.g., relaxation capacity, anticipatory co-activation, and subtle acquisition-related variability) in addition to endocrine status. Beyond PF studies, menstrual-phase modulation of sEMG-derived outcomes has been reported in other muscle groups, although effects are typically small and muscle- and task-specific [44–47].

Beyond phase comparisons, the correlational structure supports high intrapersonal stability and suggests at least two partially distinct components of myoelectric behavior. MVC-normalized sEMG was strongly associated with phasic, tonic, and endurance outcomes within each phase, indicating a coherent “dynamic excitation” profile across contraction modalities. Baseline myoelectric activity was also strongly correlated within phases, consistent with a partly distinct “basal/residual” component. Notably, participants tended to preserve their relative level of excitation across phases, suggesting that stable individual factors (e.g., learned recruitment strategies, coordination, or functional anatomy) account for substantial variance in sEMG amplitude. Interpretation of amplitude differences requires caution. sEMG amplitude is not a direct measure of motor unit recruitment, neural drive, or mechanical efficiency, and it is shaped by motor unit discharge behavior, action potential properties, and recording geometry [48–50]. Accordingly, lower resting or peak amplitudes in OP should be interpreted descriptively as reduced global myoelectric excitation under specific conditions, without inferring “greater efficiency” or “less fibre recruitment.” Given the placement sensitivity of sEMG recordings, the inherent noisiness of amplitude measures, and the between-day variability reported for intravaginal probes [51], small phase-related differences should be considered hypothesis-generating.

Force outcomes were stable across EFP, OP, and MLP. MOS grades nor dynamometric measures differed by phase, indicating preserved mechanical performance despite MC-related endocrine variation. These findings align with prior PF work reporting minimal phase-related differences in PFM strength across the cycle [52]. Reliability-focused evidence also supports the use of repeated dynamometric assessments across different MC periods, suggesting that the absence of phase differences is unlikely to be attributable solely to measurement instability [53]. At the broader neuromuscular level, systematic reviews and meta-analyses indicate that any MC-related effects on maximal strength and exercise performance are generally small at the group level and highly variable between individuals [18, 19]. This broader context strengthens the plausibility of a “stable force, subtle excitation modulation” pattern, where any hormonal effects may be expressed more in motor control or baseline excitation than in net mechanical output in healthy women performing standardized tasks.

Combining sEMG with dynamometry and digital palpation illustrates that electrophysiological and mechanical assessments capture complementary, non-redundant dimensions of PF function. MOS grades correlated positively with contraction sEMG in several comparisons, supporting partial convergence between clinical grading and myoelectric excitation during voluntary tasks, consistent with previous reports of associations between digital palpation and sEMG across female reproductive states [54]. Nevertheless, MOS is an ordinal measure with documented inter-rater limitations, which likely attenuates the magnitude and consistency of correlations with instrumented outcomes [55]. Dynamometry indicated that passive tissue distension was consistent across all three phases, suggesting that the passive “pre-tension” state assessed by the device remained stable throughout the cycle. In contrast, baseline sEMG activity was lower during OP, a finding compatible with reduced background myoelectric excitation under resting conditions. Interpreted jointly, these results suggest that the passive resistance recorded during speculum-branch opening may be influenced more strongly by connective-tissue behaviour than by active muscle activation at rest. While some work in other anatomical regions suggest increased connective-tissue laxity around ovulation [26, 27], it remains unclear whether analogous patterns occur in PF connective tissues; thus, any mechanistic interpretation should be considered provisional. One plausible interpretation is that MC-related hormonal variation may shift the balance between passive and active contributors to support, such that passive elements maintain pre-tension while baseline myoelectric excitation decreases during OP—without measurable changes in net force output.

The identification of hormonal receptors in the levator ani muscle, urogenital ligaments, round ligament, and nuclei of smooth muscle cells in females suggests that the PF is influenced by estrogen. These receptors appear mainly in connective-tissue cells within the muscle, though also found in striated muscle cells, implies that connective-tissue acts like a glue, binding different muscles together and aiding their attachment to the pelvis [56]. Consistent with this, experimental work has reported that estrogen/estradiol supplementation can alter the mechanical properties of pelvic ligaments (e.g., increasing uterosacral ligament stiffness while decreasing round ligament stiffness) [57], indicating that ovarian hormones can modulate passive pelvic support tissues, at least under certain conditions.

Finally, baseline sEMG amplitude showed inverse associations with active dynamometry and net voluntary force (active–passive), indicating that higher basal/residual excitation does not translate into greater mechanical output. This pattern is compatible with less selective motor control or reduced down-regulation (i.e., elevated background activation with a smaller effective increment during effort) and aligns with clinical observations that heightened tone can coexist with altered contractility [4]. Together, the findings support a conceptual separation between a stable “dynamic excitation” profile during voluntary tasks and a partly independent basal/residual component that relates inversely to force-based performance.

No single instrument can measure all aspects of muscle tone comprehensively. Various tools can evaluate specific components such as contractility, stiffness, or elasticity [58]. In our study, it was shown that tissue resistance was the same in the phases of the cycle, but it cannot be stated that the opening of the hiatus was the same. We cannot confirm any changes in hiatal function or closure, as ultrasound measurements were not performed [5]. It would be interesting to carry out a complementary analysis with ultrasound and elastography.

It is important to acknowledge certain limitations that may affect the interpretation and generalizability of our findings. First, the relatively small sample size may have reduced statistical power. Additionally, the inclusion of perceived exertion scale would have allowed us to determine whether subjective effort remained consistent throughout the MC, allowing for comparison with the obtained values. Finally, assessing potential changes in the urogenital hiatus and levator ani hiatus via transperineal ultrasound could have facilitated the identification of structural variations associated with hormonal fluctuations, thereby providing a more comprehensive understanding of the neuromuscular and morphological adaptations and highlighting directions for future research.

Conclusions

Surface EMG outcomes indicated subtle, condition-specific modulation of resting PFM myoelectric excitation across the menstrual cycle, whereas sEMG during voluntary contractions and force output remained stable. Baseline sEMG amplitude showed inverse associations with active force, suggesting that higher resting/residual excitation may be associated with a smaller mechanical output. sEMG measures also demonstrated high within-phase coherence and cross-phase stability, indicating largely preserved individual excitation profiles.

Overall, these findings suggest that menstrual-cycle–related modulation may primarily affect baseline myoelectric activity rather than voluntary force capacity; given the small effects and the inherent variability of amplitude-based sEMG, results should be interpreted cautiously and considered hypothesis-generating.

Abbreviatons

BMI

Body Mass Index

CI

Confidence Interval

EFP

Early Follicular Phase

sEMG

surface Electromyography

MC

Menstrual Cycle

MLP

Mid-Luteal Phase

MOS

Modified Oxford Scale

MVC

Maximal Voluntary Contraction

N

Newtons

OP

Ovulatory Phase

PF

Pelvic Floor

PFM

Pelvic Floor Muscle

SD

Standard Deviation

µV

Microvolts

Authors’ contributions

COM developed the concept and design. Statistical analysis: COM and ESR-L ran the model analysis. COM, ESR-L, MVd-D and BL contributed to acquisition and interpretation of data. COM, ESR-L, ANM, MTG and AI contributed to drafting of the manuscript. All authors contributed to critical review of the manuscript and content. COM is the guarantor.

Funding

This research project, with the acronym CiSP, was financed with funds from the X Research Call of the University Camilo José Cela.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of University Camilo José Cela (Code 04_23_CISP, 12 April 2023; Madrid, Spain). All participants signed an informed written consent to participate in the investigation.

Competing interests

The authors declare no competing interests.