Differences in Pelvic Floor Morphometry for Women with Persistent Pelvic Pain and Deep Infiltrating Endometriosis-A Cross-Sectional Study

Introduction and Hypothesis This study is aimed at describing the morphometry of the pelvic floor for a large population of women with persistent pelvic pain (PPP) compared with those without pain.

A prospective cross-sectional study was performed between January 2013 and November 2015, recruiting women attending a general gynaecology clinic. Demographic data were collected and translabial four-dimensional ultrasound (4DUS) was performed on all participants. Morphometric assessment of pelvic floor muscles was undertaken at rest, with contraction and Valsalva, and analysed by an assessor blinded to demographic details.

Of the 747 participants, 469 (62.8%) had PPP and 278 (37.2%) reported no pelvic pain. Levator hiatal (LH) area (14.4 vs 18.6 cm2, p = 0.009) and left–right (LR) diameter (3.9 vs 4.1 cm, p = 0.019) were smaller for women with PPP in the Valsalva assessment than in participants with no pain. For women who were currently requiring treatment for pelvic pain (158 out of 747, 21.2%) compared with those not having treatment, there were differences in LH area and LR diameter in all phases of movement; at rest (11.4 vs 12.8 cm2, p = 0.017; 3.6 vs 3.8  cm2, p = 0.038), Valsalva (13.6 vs 16.4 cm2, p = < 0.001, 3.8 vs 4.2  cm2, p = 0.001) and contraction (9.3 vs 10.4 cm2, p = 0.021, 3.3 vs 3.6  cm2, p = 0.006). For women with deep infil- trating endometriosis (DIE), smaller LH areas were seen on rest and Valsalva (adjusted mean 11.89 vs 12.75 cm2, p = 0.015, 15.13 vs 16.98 cm2, p = 0.002) but not contraction (10.05 vs 10.48 cm2, p = 0.161) than in women without endometriosis.

Pelvic floor morphometry differs for women when comparing groups with PPP and no pain, particularly in the group of patients currently receiving treatment for their PPP, as well as the subgroup of DIE even when adjusting for confounding variables such as mode of delivery, age and prolapse. This supports the hypothesis that women with PPP have different PFM characteristics from women without pain symptoms. This study reports on mean measurements with likely overlapping distribution groups; therefore, we are unable to give a diagnostic criterion or measurement that could be used for the ultrasound assessment of women with PPP. Further research could stratify severity using validated assessments and compare clinical examination findings with the results seen on ultrasound.

Biometry · Four-dimensional ultrasound · Morphometry · Pelvic floor · Pelvic pain

Persistent pelvic pain (PPP) in women has a high burden of disease [1–3]. The connection between increased pelvic floor muscle (PFM) tone and PPP has been investigated using numerous tools [4–6]. Unfortunately, synthesis of the litera- ture is difficult owing to design and methodological factors. Despite this, an association is suggested by digital palpation and dynamometry with a strong recommendation to research more objective methods for diagnosis [7]. Three- and four-dimensional ultrasound (3DUS, 4DUS) of the pelvic floor is increasingly utilised in gynaecological Handling Editor: Symphorosa Shing Chee Chan Editor in Chief: Maria A. Bortolini * Erin M. Nesbitt-Hawes [email protected] 1 Discipline of Women’s Health, School of Clinical Medicine, UNSW Sydney, Sydney, NSW, Australia 2 Gynaecological Research and Clinical Evaluation (GRACE) Group, Gynaecology Department, Royal Hospital for Women, Randwick, NSW, Australia 3 Sydney Urodynamic Centres, Sydney, NSW, Australia International Urogynecology Journal investigation, primarily in the diagnosis of obstetric trauma and pelvic organ prolapse [8 –10]. More recently, this has extended to the assessment of pelvic pain such as pelvic gir- dle pain [11], endometriosis [12, 13] and provoked vestibu- lodynia [14, 15]. Data from these small studies suggest that morphometry (shape and dimensions) of the PFM might dif- fer in women with specific pelvic pain conditions (notably, smaller hiatal areas and reduced PFM contraction); however, none of these studies reflects a broader group of women with PPP [11–15, 15]. Pelvic floor physiotherapy is used to treat PPP related to increased PFM tone [16, 17]. Changes in levator hiatal (LH) area and left–right diameter have been recorded when com- paring 3DUS measurements before and after physiotherapy sessions for participants with high-tone pelvic floor dysfunc- tion [18]. In addition to these measurements, we hypothe- sised that increased muscle tone would also create increased bulk, leading to the addition of PFM width measurements for our cohort. We previously reported on nulliparous patients, where obstetric trauma could not bias results. That study reported no difference in 4DUS biometric indices for women with and those without PPP [19]. This study is aimed at reporting on a cohort of women, both nulliparous and multiparous, to identify differences in morphometry between women with and those without PPP.

Study Population This prospective cross-sectional study was conducted between January 2013 and November 2015 with institutional ethical approval (HREC 12/194). The methodology has been previously reported and is summarised here [19]. All women presenting to either a public or private gynaecology clinic were invited to participate, regardless of their clini- cal presentation. Exclusion criteria included current preg- nancy, inability to understand written and spoken English and age lower than 18. The primary outcome of the study was to compare the pelvic floor morphometry of women with and those without PPP. The secondary outcome was to consider the utility of 4DUS as a diagnostic tool for women presenting with PPP. We also performed a subgroup analysis in patients who had a diagnosis of endometriosis to allow comparison with previous literature that has examined this association [12, 16]. Demographic details were recorded by questionnaire filled out by the researchers during interview at the time of patient enrolment, and a dynamic pelvic floor ultrasound was performed for each woman enrolled. Data included the presence of PPP (defined as pain located in the pelvic region that was constant or intermittent, cyclical or non-cyclical pain that persisted for greater than 3 months) [20]. Visual analogue scoring of pain intensity was not performed. Types of PPP experienced were categorised as dysmenorrhoea, non-menstrual pelvic pain, dyspareunia, dysuria and dys- chezia. Current treatment(s) for pain, prior pelvic surgery (including endometriosis and/or hysterectomy) and obstet- ric history were documented, with surgical data including endometriosis subtype (deep or superficial, but not location of lesions) extracted from operation reports where availa- ble. Disease-specific validated questionnaires were not used owing to limited availability at the time of data collection. Pelvic Floor Ultrasound The technique of 3DUS and 4DUS of the pelvic floor is sum- marised here and has been described previously [21]. All pelvic floor ultrasounds were performed by a single opera- tor (gynaecologist with 5 years’ clinical experience at the commencement of the study) following a 3-month period of training with an expert in the field. Inter-observer variability analysis was performed on an initial cohort of 36 partici- pants as reported in a previously published paper that dem- onstrated good to very good correlation [19]. Other studies examining inter-observer repeatability of 3D and 4DUS of the pelvic floor have also shown good to excellent correla- tion for all measurements used in this study [22–24]. A Voluson 730Expert or E8 (GE Medical Ultrasound, Zipf, Austria) and an 8.4-MHz curved array volume probe (acquisition angle 70–85%) was used to capture ultrasound volumes. Participants emptied their bladder and bowel and removed tampons and pessaries prior to the procedure. With participants in the dorsal lithotomy position, the ultra- sound probe was placed on the perineum in the midsagittal plane. The PFMs were assessed at rest and with a functional assessment, including maximal pelvic floor contraction (three repetitions) and maximal Valsalva manoeuvre (three repetitions). Measurements Analysis of the ultrasound volumes was performed offline at a later date (on average 6 months following ultrasound acquisition), with the assessor blinded to demographic and clinical data. Volumes were reviewed and measurements were performed with the PFM at rest, maximal contraction (maximal narrowing of the LH area) and Valsalva (maxi- mal descent of organs). Volumes that were determined to be of insufficient quality for measurements to be per - formed were excluded from analysis. Standard measure- ments for each volume included bladder-neck descent, LH area, pubovisceral-muscle (PVM) length (measured as the external length of the contour of the pubovisceral muscle), International Urogynecology Journal anterior–posterior (AP) and left–right (LR) diameters, and an average of the PVM width or thickness performed at four intervals along the muscle and averaged [23]. Statistical Analysis Statistical analysis was conducted using SPSS version 27. Con- tinuous data are presented as mean values with their respective standard deviations. The normality of the data was assessed using the Shapiro–Wilk test and Q-Q plots. Student's t test and analysis of variance (ANOVA) were used for normally distrib- uted data, whereas the Mann–Whitney U test (Wilcoxon rank- sum test) was used for non-parametric data. Inverse probability weighting and generalised linear models were used to control for confounders. Inverse probability weights were calculated using a logistic regression model including baseline covari- ates. To evaluate covariate balance, we compared weighted and unweighted means and proportions and calculated standard- ised mean differences. Where residual differences remained, the covariates were included in the outcome model. Effective sample sizes after weighting were calculated using the Kish approximation. A significance level of p < 0.05 was considered statistically significant for all tests. An a priori sample size calculation was performed, to detect a difference in the LH area for women with pain using published data for provoked vestibulodynia [15], as no data specific to this field were available. Using a difference of 1.06 cm2 (SD 2.22), power of 80% with a p < 0.05, a mini- mum of 140 patients were required. Owing to an expecta- tion of a wider variation in pelvic floor measurements for a non-pain cohort, incomplete data sets and anticipated mul- tivariate regression for several factors (allowing as a rule of thumb 20–30 patients per factor), we calculated a sample size of 700 women to detect a difference in pelvic muscle morphometry.

Of 830 women approached, 747 (90%) consented to the study and underwent translabial 4DUS. Of these, 469 out of 747 (62.8%) reported PPP of any type. Demographic data are presented in Table 1. The mean age for the study population was 41.2 years (standard deviation [SD]: 13.3 years). Women who described PPP were younger, with a mean age of 36.9 years (SD 10.5 years), vs no PPP 48.5 years (SD:14.2 years). Women experiencing PPP were also more likely to be pre-menopausal (71.1% vs 28.9%) and have had surgery for endometriosis (46.5% versus 11.9%). Most women reported being Australian born (64.5%). When asked about their current non-surgical treatment for pelvic pain, 158 out of 747 women (21.2%) reported current use of analgesic medications (10.1%), physiotherapy (5%), botuli- num toxin injections to the pelvic floor (0.4%), other (0.7%) or a combination of treatments (5%). Among the total number of women in the cohort (N = 747), 14.6% reported prolapse, with participants with prolapse less likely to report PPP symptoms than those with- out prolapse (32.1% vs 67.9%; p < 0.001). Three hundred and eighty (n = 380, 50.9%) women in the study were parous (with 486 individual births). There was a significant difference between women with and those without PPP when comparing those who had ever had a vaginal deliv- ery (including instrumental and those who had both vaginal and caesarean delivery for different births) versus those who only had caesarean deliveries. Women with PPP had a higher Table 1 Demographic characteristics of all women (persistent pelvic pain (PPP) versus no pain) Mode of delivery is reported as n = all that apply, with the cumulative total accounting for more than the number of parous women owing to patients having more than one type of delivery Percentages use the denominator of parous women in that group rather than the group as a whole *Statistically significant BMI body mass index, IQR interquartile range, PPP persistent pelvic pain, SD standard deviation Demographics PPP, N = 469 (62.8%) No pain, N = 278 (37.2%) p value Age (years), mean (SD) 36.9 (10.58) 48.5 (14.29) < 0.001* BMI (kg/m2), mean (SD) 24.9 (5.0) 24.6 (4.5) 0.46 Post-menopausal, % 9.0 36.6 < 0.001* Surgery for endometriosis, % 46.5 11.9 < 0.004* Prolapse, % 9 31.6 < 0.001* Parity, median (IQR) 0 (0–2) 2 (0–2) < 0.001* Parous, n (%) 39.2 70.2 < 0.001* Spontaneous vaginal delivery, n (%) 111/184 (60.3) 147/195 (75.4) 0.002* Caesarean, n (%) 70/184 (38.0) 47/195 (18.7) < 0.001* Forceps, n (%) 34/184 (18.5) 43/195 (22.1) 0.402 Ventouse, n (%) 18/184 (9.8) 13/195 (6.7) 0.262 International Urogynecology Journal frequency of having had only caesarean deliveries 29.3% ver- sus 12.2% (Chi-squared (1) = 17.02, p < 0.001) in the non-pain group. Reasons for the caesarean delivery or the status of PPP prior to pregnancy were not available. Of the 469 women (62.8%) reporting PPP, 345 out of 469 (73.6%) women reported dysmenorrhea, 312 out of 469 (66.5%) non-menstrual pelvic pain, 281 out of 469 (59.9%) dyspareunia, 108 out of 469 (23.0%) dyschezia and 153 out of 469 (32.6%) dysuria. After applying inverse probability weights, the effective sample size was approximately 405 in the PPP group and 160 in the no pain group. Table 2 demonstrates the comparison of unweighted and weighted means. Comparative morphometry data for women with and those without PPP at rest, Valsalva and on contraction are presented in Table 3, with significant differences seen in LH area and LR diameter on Valsalva, as well as PVM width on contraction. To better understand the more severe cohort of PPP, analysis of biometric data was per- formed for women who were currently receiving non-surgical treatment for PPP (Table 4). In this group, significant differ- ences were found in both LH area and LR diameter at rest, on Valsalva and on contraction. Of the 253 out of 747 (33.9%) women who had surgery for endometriosis, 61 out of 253 (24%) had deep infiltrating endometriosis (DIE). After adjusting for prolapse, vaginal delivery, age and BMI, the presence of DIE was signifi- cantly associated with a smaller LH area at rest and Valsalva, although this was not seen for PFM contraction (Table  5). Additionally, a shorter PVM was seen in DIE across all measures of rest, Valsalva and contraction. Dynamic pelvic floor assessments were incomplete for 93 participants, primarily because of technical issues related to Table 2 Unweighted and weighted means and standardised mean difference for persistent pelvic pain (PPP) versus no pain groups BMI body mass index, PPP persistent pelvic pain, SD standard deviation, SMD standardised mean difference PPP (unweighted) No pain (unweighted) SMD PPP (weighted) No pain (weighted) SMD Age (years), mean (SD) 36.9 (10.58) 48.5 (14.29) 0.92 40.4 (13.29) 44.5 (14.02) 0.30 BMI (kg/m2,) mean (SD) 24.9 (5.0) 24.6 (4.5) 0.004 24.7 (4.90) 24.8 (4.570 0.04 Surgery for endometriosis, % 46.5 11.9 0.82 46.4 12 0.82 Prolapse, % 9 31.6 0.59 14.3 25 0.27 Parous, % 39.2 70.2 0.66 39.3 68.5 0.61 Table 3 Biometric measurements of patients with persistent pelvic pain (PPP) versus no pelvic pain (inverse probability weighting was used to adjust for confounding factors of age, parity, vaginal delivery and endometriosis diagnosis) AP anterior–posterior, L–R left–right, PPP persistent pelvic pain, PVM pubovisceral muscle *Statistically significant PPP, mean (SD) No pain, mean (SD) B Standard error 95% CI p value Rest   Levator hiatal area (cm2) 11.8 (3.41) 13.8 (4.70) −0.668 0.3931 (−0.10, 1.43) 0.089   PVM length (cm) 11.0 (1.56) 11.50 (1.72) −0.109 0.1691 (0.22, 0.41) 0.521   AP diameter (cm) 4.7 (0.85) 5.2 (1.23) −0.171 0.1030 (0.03, 2.75) 0.097   L-R diameter (cm) 4.42 (16.10) 3.39 (0.72) 0.839 0.9091 (−0.94, 2.62) 0.356   PVM width (average cm) 1.02 (0.19) 1.05 (0.20) −0.019 0.0205 (−0.06, 0.02) 0.344 Valsalva   Levator hiatal area (cm2) 14.4 (5.73) 18.6 (8.39) −1.673 0.6436 (−2.9, −0.41) 0.009 *   PVM length (cm) 11.6 (1.73) 12.2 (1.96) −0.172 0.1775 (−0.52, 0.18) 0.332   AP diameter (cm) 5.1 (1.14) 8.0 (38.2) −1.588 1.3608 (−4.26, 1.08) 0.243   L–R diameter (cm) 3.9 (0.71) 4.1 (0.97) −0.187 0.0796 (−0.34, −0.03) 0.019 *   PVM width (average cm) 1.1 (1.66) 1.0 (0.16) 0.094 0.0837 (−0.07, 0.26) 0.262 Contraction   Levator hiatal area (cm2) 9.5 (2.97) 11.2 (4.05) −0.517 0.3163 (−1.14, 0.10) 0.102   PVM length (cm) 9.8 (1.37) 10.3 (1.50) −0.234 0.1324 (−0.49, 0.03) 0.077   AP diameter (cm) 4.0 (0.78) 4.3 (0.94) −0.109 0.0813 (−0.27, 0.05) 0.178   L–R diameter (cm) 3.4 (0.65) 3.7 (0.75) −0.088 0.0648 (−0.22, 0.04) 0.177   PVM width (average cm) 1.0 (0.18) 1.1 (0.19) −0.043 0.0179 (−0.08, −0.01) 0.015 * International Urogynecology Journal ultrasound data capture and transfer rather than patient fac- tors. A comparison was performed within the incomplete and complete groups, with no differences in age, BMI, parity and prolapse, although there was a significant difference in endometriosis surgery (Table  6). To account for potential bias, inverse probability weighting was applied.

This is to our knowledge the largest prospective study to assess the pelvic floor using 3/4DUS in a general population presenting for gynaecological care and stratifying by the pres- ence or absence of PPP. This study demonstrates differences in biometric measurements during Valsalva with smaller LH area and LR diameter for women with PPP. In addition, in the group of women requiring current treatment for PPP, differ- ences in LH area and LR diameter were seen across all move- ment parameters of rest, Valsalva and contraction. Smaller LH areas were seen in participants with DIE across Valsalva and contraction manoeuvres as well as AP diameters. The presence of high PFM tone may explain these findings, with reduced movement of the PFMs in women with PPP. This association is supported by a 2023 systematic review and meta-analysis suggesting that stiffness and reduced flexibility might be more common in high-tone PFMs [6]. Previous studies of PFM ultrasound have focused on spe- cific conditions such as pelvic girdle pain [11], endometrio- sis [12, 13] and provoked vestibulodynia [14, 15], with small numbers in each cohort. A case–control study examined 49 women with pelvic girdle pain and 49 age-matched controls, showing a significantly smaller LH area at rest and on con- traction in the pain group [11]. In a separate study with near- identical methodology, 38 women with provoked vestibulo- dynia were compared with 39 asymptomatic controls [14], also reporting a difference between the groups, with smaller LH areas at rest, contraction and Valsalva. Our publication of the nulliparous subgroup failed to show a significant difference in biometric properties of the pelvic floor in women with PPP [19]. The current cohort includes all study participants, adjusted for parity and vagi- nal delivery. This study demonstrates differences in meas- urements in the Valsalva and contraction phases, but not at rest in the PPP group. Given the published literature and our hypothesis, this was unexpected but may point to some

of our study. Pain scales and validated sever - ity scores were not included in the patient questionnaire, and this may result in heterogeneity within the PPP group with the inclusion of self-reported, unscaled pelvic pain. In order to better assess severity within the limitations of the protocol, analysis of the group of women requiring current treatment for their PPP was performed and demonstrated a Table 4 Biometric measurements of patients receiving current treatment for persistent pelvic pain (PPP) versus no treatment for pelvic pain (inverse probability weighting was used to adjust for confounding factors of age, BMI, parity, vaginal delivery and endometriosis diagnosis) AP anterior–posterior, L–R left–right, PPP persistent pelvic pain, PVM pubovisceral muscle *Statistically significant Treatment for PPP, mean (SD) No treatment for PPP, mean (SD) B Standard error 95% CI p value Rest   Levator hiatal area (cm2) 11.4 (3.24) 12.8 (4.11) −0.935 0.3914 (0.17, 1.70) 0.017*   PVM length (cm)) 10.9 (1.61) 11.3 (1.60) −0.165 0.1732 (−0.51, 0.17) 0.340   AP diameter (cm) 4.7 (0.85) 5.0 (1.07) −0.160 0.0957 (−0.35, 0.23) 0.095   L–R diameter (cm) 3.6 (0.58) 3.8 (0.70) −0.144 0.0695 (−0.28, −0.01) 0.038*   PVM width (average cm) 1.0 (0.18) 1.0 (0.19) 0.015 0.0208 (−0.03, 0.06) 0.465 Valsalva   Levator hiatal area (cm2) 13.6 (4.66) 16.4 (7.15) −2.20 0.5901 (−3.36, −1.04) < 0.001*   PVM length (cm) 11.4 (1.72) 11.9 (1.85) −0.319 0.1833 (−0.67, 0.04) 0.082   AP diameter (cm) 5.0 (1.05) 6.6 (28.00) −1.221 0.9001 (−2.99, 0.54) 0.175   L–R diameter (cm) 3.8 (0.63) 4.2 (0.88) −0.254 0.0770 (−0.41, −0.10) <0.001*   PVM width (average cm) 1.0 (0.16) 1.1 (1.6) −0.077 0.0565 (−0.19, 0.03) 0.174 Contraction   Levator hiatal area (cm2) 9.3 (2.82) 10.4 (3.58) −0.705 0.3064 (−1.3, −0.11) 0.021*   PVM length (cm) 9.7 (1.44) 10.1 (1.42) −0.183 0.1402 (−0.46, 0.09) 0.193   AP diameter (cm) 4.0 (0.81) 4.1 (0.86) −0.092 0.0806 (−0.25, 0.07) 0.252   L–R diameter (cm) 3.3 (0.57) 3.6 (0.73) −0.173 0.0633 (−0.30, −0.05) 0.006*   PVM width (average cm) 1.0 (0.16) 1.0 (0.19) −0.026 0.0175 (−0.06, 0.01) 0.141* International Urogynecology Journal reduced hiatal area and LR diameter at rest, Valsalva and contraction, which is more consistent with the findings of other studies. This result suggests a correlation between the severity of the PPP and greater differences in PFM biometric characteristics, possibly due to stiffer, higher-tone PFM. Several studies have assessed nulliparous women with endometriosis and their PFM biometry. The first compared 50 women with endometriosis with 35 women without [12], reporting significant differences for LH areas at rest (12.0 vs 13.2 cm2; p = 0.03) and on Valsalva (14.5 vs 17.3 cm2; p < 0.01) as well as AP diameter on Valsalva but not contrac- tion [12]. In the second, 75 women with DIE were compared with 39 who had ovarian disease [13]. Participants in the DIE group had a reduction in LH area at rest (10.8 vs 11.9 cm2; p = 0.03), contraction (8.8 vs 9.3 cm2; p = 0.03) and Valsalva (12.2 vs 14.2 cm2; p = 0.02) as well as LR diameter in all degrees of movement, compared with the group with ovarian disease [13]. Our findings also suggest a difference for women with DIE with reduced LH area (rest and Valsalva although not contraction), PVM length (rest, Valsalva and contraction) and PVM width (Valsalva). Unlike the above studies, we did not find an association with LR diameter in any of the phases of movement, and it is possible that this relates to the inclu- sion of parous as well as nulliparous women, resulting in a Table 5 Comparison of deep infiltrating endometriosis (DIE) versus no endometriosis among women who had experienced PPP AP anterior–posterior, DIE deep infiltrating endometriosis, L–R left–right, PVM pubovisceral muscle *Statistically significant Endometriosis group Adjusted mean (95% CI) Mean difference vs no endometriosis (B) p value Rest   Levator hiatal area (cm2) No endometriosis (reference) 12.75 (12.48, 13.03) – – DIE 11.89 (11.26, 12.57) −0.070 0.015*   PVM length (cm) No endometriosis (reference) 11.28 (11.15, 11.41) – – DIE 10.83 (10.53, 11.15) −0.040 0.006 *   AP diameter (cm) No endometriosis (reference) 4.98 (4.90, 5.06) – – DIE 4.80 (4.62, 4.99) −0.036 0.071   L–R diameter (cm) No endometriosis (reference) 3.75 (3.70, 3.81) – – DIE 3.63 (3.51, 3.78) −0.031 0.108   PVM width (average cm) No endometriosis (reference) 1.05 (1.03, 1.07) – – DIE 1.04 (1.00, 1.08) −0.010 0.614 Valsalva   Levator hiatal area (cm2) No endometriosis (reference) 16.98 (16.51, 17.47) – – DIE 15.13 (14.06, 16.28) −0.116 0.002*   PVM length (cm) No endometriosis (reference) 11.99 (11.84, 12.13) – – DIE 11.49 (11.13, 11.85) −0.043 0.008*   AP diameter (cm) No endometriosis (reference) 7.72 (7.40, 8.05) – – DIE 7.15 (6.39, 8.00) −0.077 0.190   L–R diameter (cm) No endometriosis (reference) 4.27 (4.20, 4.33) – – DIE 4.11 (3.94, 4.28) −0.038 0.080   PVM width (average cm) No endometriosis (reference) 1.11 (1.07, 1.14) – – DIE 0.96 (0.89, 1.04) −0.143 0.001 * Contraction   Levator hiatal area (cm2) No endometriosis (reference) 10.48 (10.25, 10.71) – – DIE 10.05 (9.48, 10.65 −0.042 0.161   PVM length (cm) No endometriosis (reference) 10.11 (10.00, 10.22) – – DIE 9.80 (9.53, 10.07) −0.031 0.030 *   AP diameter (cm) No endometriosis (reference) 4.19 (4.13, 4.26) – – DIE 4.19 (4.02, 4.36) −0.001 0.958   L–R diameter (cm) No endometriosis (reference) 3.59 (3.53, 3.65) – – DIE 3.49 (3.34, 3.64) −0.029 0.191   PVM width (average cm) No endometriosis (reference) 1.01 (1.00, 1.03) – – DIE 0.98 (0.95, 1.02) −0.030 0.124 International Urogynecology Journal wider range of measurements. Studies associating DIE and smaller LH areas hypothesise that deeply infiltrating disease and inflammation may be more likely to cause both neural and muscular effects, because of either direct injury or increased visceral sensitivity resulting in changes to PFM dimensions [18]. This represents a limitation of our study owing to the absence of information on the specific location of endometrio- sis lesions, thus limiting the capacity for a stratified analysis of morphometry to better address this hypothesis. A difference was demonstrated between the PPP and no pain groups with regard to modes of delivery, with a higher frequency of caesarean delivery in the PPP group. Although this study was not designed to measure this outcome and therefore cannot answer questions such as the presence of PPP prior to pregnancy and delivery, or the reasons for hav- ing caesarean delivery, it documents a correlation that is worthy of future investigation. There is divergent evidence of pain outcomes for women after delivery and the cor - relation with PPP is still not established. One long-term study reported that women with a prior vaginal delivery are more likely to have dyspareunia at 6–11 years postpartum, although general pelvic pain rates were no different [25]. A separate longitudinal study reported reduced new onset pain in the first 3 months postpartum for women who had a caesarean delivery, with a history of pelvic pain being the only risk factor associated with increasing pain scores over time [26]. A link between caesarean delivery and pelvic pain has been reported in a small longitudinal study, and other studies report pelvic girdle pain to be more common after caesarean, both early and at 6 months postnatally [27– 29]. Our hypothesis is that of reverse causation, with pre- existing PPP resulting in the outcome of caesarean delivery. This could be due to higher-tone PFM at baseline resulting in factors that could affect mode of delivery, such as reduced distensibility to allow for vaginal delivery, or that patients may have self-selected caesarean section as their mode of delivery. Further research in this area would be beneficial to further explore and test this hypothesis. There are several limitations of this study. First, pain severity in women with PPP was not assessed, limiting fur- ther stratification. To adjust for this within the study, analysis of the treated subgroup was performed. Second, the sample- size calculation in this study was based on available pub- lished data at the outset of the study. This was in a specific population (provoked vestibulodynia) and resulted in changes in all biometric measures. This range of biometric differences was not supported in this study, where a purposefully broad inclusion criterion was designed for greater external validity. Our decision to use this benchmark instead of conducting a pilot study may have affected sample-size calculations; how- ever, the allowances for contingencies were generous. Third, the evaluation of PFM tone was not clinically assessed. Our aim was to examine the value of an “objective” measure such as ultrasound, as the assessment of tone is variably reported and may not be widely used in a general setting; however, we recognise that a formal PFM assessment via a palpation technique, which would have allowed better understanding of the correlation between high-tone muscles on palpation and biometry on ultrasound. The evolution of 3D/4D ultra- sound as an imaging modality over time has tended to be complementary to clinical examination rather than a replace- ment and this is supported by our findings. Finally, although we report on several ultrasound measurements, our primary analysis was of LH area and LR diameter, consistent with our aims. We did not adjust for multiple comparisons, and this may increase type 1 error. Results should be interpreted with caution, with emphasis placed on effect sizes and confidence intervals rather than on p values alone. This study adds to the literature supporting differences in PFM biometric indices for women with PPP (particularly those requiring current treatment) and DIE when assessed by 3D/4DUS and adds weight to the hypothesis of increased PFM tone in patients with PPP and endometriosis. Although differ- ences were seen, the distribution of measurements is such that there will be overlap between women in the PPP and no pain groups. For this reason, ultrasound measurements alone can- not be used to diagnose a patient with increased PFM tone but may have a role in predicting response to or effect of, treatment such as physiotherapy. Scoring scales have been proposed as an option for women with endometriosis to help to identify those who would benefit from adjuvant physiotherapy [30]. Treat- ment effect with physiotherapy has already been measured using 3DUS and this may have further application in provid- ing an objective measure to determine effect on the musculature over time [18]. Further research that includes validated pain scoring as well as a comparison of 3D/4DUS with digital palpa- tion of high-tone PFM may demonstrate a stronger association between symptomatology and PFM biometry on ultrasound. Appendix 1 Table 6 Comparison of incomplete versus complete ultrasound *Statistically significant Incomplete (n = 93) Complete (n = 654) p value Age, mean (SD) 42.4 (15.75) 41.1 (12.93) 0.45 BMI, mean (SD) 25.4 (5.3) 24.7 (4.7) 0.20 Prolapse, % 14 18 0.55 Pelvic pain, % 62 63 0.93 Endometriosis sur- gery, % 24 35 0.02* Parous, % 45 52 0.24 International Urogynecology Journal

We would like to thank Nancy Briggs for her assistance with the statistical evaluation. We thank the Australasian Gynaecological Endoscopy and Surgery Society (AGES), the Royal Hospital for Women Foundation, Gynaecological Research and Clinical Evaluation (GRACE) grant and the Royal Australian and New Zealand College of Obstetricians and Gynaecologists (RANZCOG) for their support for this study through funding and scholarships. Authors’ Contributions E.M.N.-H.: protocol/project development, eth- ics, data collection and management, data analysis, manuscript writ - ing/editing; H.P.D.: data analysis, manuscript writing/editing; W.L.L.: manuscript writing/editing; J.A.A.: protocol/project development, manuscript writing/editing. Funding Open Access funding enabled and organized by CAUL and its Member Institutions. This study was funded by grants from the AGES, the Royal Hospital for Women Foundation, GRACE grant and a scholarship from the RANZCOG. None of the sources of funding was involved in the design, collection or composition of the manuscript. Declarations Ethical Approval Ethical approval for this study was obtained from the South Eastern Sydney Local Health District Human Research Ethics Committee (HREC 12/194). Informed Consent to Participate Written informed consent was obtained from all subjects before the study. Informed Consent to Publish Written informed consent was obtained from the patient(s) for their anonymised information to be published in this article. Conflicts of interest H.P.D. has received unrestricted educational grants, travel support and lecture fees from GE Healthcare. The other authors have no conflicts of interest to disclose. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

1. Mathias SD, Kuppermann M, Liberman RF, Lipschutz RC, Steege JF. Chronic pelvic pain: prevalence, health-related quality of life, and economic correlates. Obstet Gynecol. 1996;87(3):321–7. 2. Zondervan KT, Yudkin PL, Vessey MP, Jenkinson CP, Dawes MG, Barlow DH, et al. The community prevalence of chronic pelvic pain in women and associated illness behaviour. Br J Gen Pract. 2001;51(468):541. 3. Bachmann GA, Rosen R, Arnold LD, Burd I, Rhoads GG, Lei- blum SR, et al. Chronic vulvar and gynecologic pain: prevalence and characteristics in a self-reported survey. J Reprod Med. 2006;51(1):3. 4. Worman RS, Stafford RE, Cowley D, Prudencio CB, Hodges PW. Evidence for increased tone or overactivity of pelvic floor mus- cles in pelvic health conditions: a systematic review. Am J Obstet Gynecol. 2022;228(6):657-74.e91. 5. Kapurubandara SC, Lowes B, Sansom-Daly UM, Deans R, Abbott JA. A systematic review of diagnostic tests to detect pel- vic floor myofascial pain. Int Urogynecol J Pelvic Floor Dysfunct. 2022;33(9):2379–89. 6. Kadah S, Soh S-E, Morin M, Schneider M, Heron E, Frawley H. Is there a difference in pelvic floor muscle tone between women with and without pelvic pain? A systematic review and meta-analysis. J Sex Med. 2023;20(1):65–96. 7. Kadah S, Soh S-E, Morin M, Schneider M, Ang WC, McPhate L, et al. Are pelvic pain and increased pelvic floor muscle tone asso- ciated in women with persistent noncancer pelvic pain? A system- atic review and meta-analysis. J Sex Med. 2023;;20(9):1206–21. https:// doi. org/ 10. 1093/ jsxmed/ qdad0 89. 8. Blasi I, Fuchs I, D’ Amico R, Vinci V, La Sala GB, Mazza V, et al. Intrapartum translabial three-dimensional ultrasound visualization of levator trauma. Ultrasound Obstet Gynecol. 2011;37(1):88–92. 9. Hoyte L, Schierlitz L, Zou K, Flesh G, Fielding JR. Two- and 3-dimensional MRI comparison of levator ani structure, volume, and integrity in women with stress incontinence and prolapse. Am J Obstet Gynecol. 2001;185(1):11–9. 10. Dietz HP, Steensma AB. The prevalence of major abnormali- ties of the levator ani in urogynaecological patients. BJOG. 2006;113(2):225–30. 11. Stuge B, Saetre K, Braekken IH. The association between pelvic floor muscle function and pelvic girdle pain—a matched case con- trol 3D ultrasound study. Man Ther. 2012;17(2):150–6. 12. Raimondo D, Youssef A, Mabrouk M, Del Forno S, Martelli V, Pilu G, et al. Pelvic floor muscle dysfunction on 3D/4D transper - ineal ultrasound in patients with deep infiltrating endometriosis: a pilot study. Ultrasound Obstet Gynecol. 2017;50(4):527–32. 13. Mabrouk M, Raimondo D, Del Forno S, Baruffini F, Arena A, Benfenati A, et al. Pelvic floor muscle assessment on three-and four-dimensional transperineal ultrasound in women with ovar - ian endometriosis with or without retroperitoneal infiltration: a step towards complete functional assessment. Ultrasound Obstet Gynecol. 2018;52(2):265–8. 14. Thibault-Gagnon S, McLean L, Goldfinger C, Pukall C, Cham- berlain S. Differences in the biometry of the levator hiatus at rest, during contraction, and during Valsalva maneuver between women with and without provoked vestibulodynia assessed by transperineal ultrasound imaging. J Sex Med. 2016;13(2):243–52. 15. Morin M, Bergeron S, Khalife S, Mayrand MH, Binik YM. Mor- phometry of the pelvic floor muscles in women with and with- out provoked vestibulodynia using 4D ultrasound. J Sex Med. 2014;11(3):776–85. 16. Van Reijn-Baggen DA, Han-Geurts IJ, Voorham-van der Zalm PJ, Pelger RC, Hagenaars-van Miert CH, Laan ET. Pelvic floor physical therapy for pelvic floor hypertonicity: a systematic review of treatment efficacy. Sex Med Rev. 2022;10(2):209–30. 17. Schubert R, Song S, Everist R, Nesbitt-Hawes E, Abbott J. The impact of multimodal physiotherapy in an interdisciplinary set- ting for the management of women with persistent pelvic pain and pelvic floor tension myalgia. Eur J Physiother. 2024;26(2):66–71. 18. Volpe LJ, Zugelder M, Kotarinos R, Kotarinos E, Kenton K, Gey- nisman-Tan J. Objective changes in pelvic floor muscle strength and length in women with high-tone pelvic floor dysfunction after pelvic floor physical therapy (RELAX Trial). Urogynecology. 2023. https:// doi. org/ 10. 1097/ spv. 00000 00000 001355. 19. Nesbitt-Hawes E, Dietz H, Abbott J. Morphometry of the nullipa- rous pelvic floor. Ultrasound Obstet Gynecol. 2018;52(5):672–6. International Urogynecology Journal 20. Treede R-D, Rief W, Barke A, Aziz Q, Bennett MI, Benoliel R, et al. Chronic pain as a symptom or a disease: the IASP Clas- sification of Chronic Pain for the International Classification of Diseases (ICD-11). Pain. 2019;160(1):19–27. 21. Dietz H. Ultrasound imaging of the pelvic floor. Part II. Three- dimensional or volume imaging. Ultrasound Obstet Gynecol. 2004;23(6):615–25. 22. Majida M, Braekken IH, Umek W, Bo K, Saltyte Benth J, Ell- strom Engh M. Interobserver repeatability of three- and four- dimensional transperineal ultrasound assessment of pelvic floor muscle anatomy and function. Ultrasound Obstet Gynecol. 2009;33(5):567–73. 23. Dietz HP, Shek C, Clarke B. Biometry of the pubovisceral muscle and levator hiatus by three-dimensional pelvic floor ultrasound. Ultrasound Obstet Gynecol. 2005;25(6):580–5. 24. Tan L, Shek KL, Atan IK, Rojas RG, Dietz HP. The repeatability of sonographic measures of functional pelvic floor anatomy. Int Urogynecol J Pelvic Floor Dysfunct. 2015;26(11):1667–72. 25. Blomquist JL, McDermott K, Handa VL. Pelvic pain and mode of delivery. Am J Obstet Gynecol. 2014;210(5):423.e1-6. 26. Bjelland EK, Owe KM, Pingel R, Kristiansson P, Vangen S, Eber- hard-Gran M. Pelvic pain after childbirth: a longitudinal popula- tion study. Pain. 2016;157(3):710–6. 27. Li W-Y, Liabsuetrakul T, Stray-Pedersen B, Li Y-J, Guo L-J, Qin W-Z. The effects of mode of delivery and time since birth on chronic pelvic pain and health-related quality of life. Int J Gynae- col Obstet. 2014;124(2):139–42. 28. Bjelland EK, Stuge B, Vangen S, Stray-Pedersen B, Eberhard-Gran M. Mode of delivery and persistence of pelvic girdle syndrome 6 months postpartum. Am J Obstet Gynecol. 2013;208(4):298.e1-7. 29. Mukkannavar P, Desai B, Mohanty U, Parvatikar V, Karwa D, Daiwajna S. Pelvic girdle pain after childbirth: the impact of mode of delivery. J Back Musculoskelet Rehabil. 2013;26(3):281–90. 30. Arena A, Degli Esposti E, Cocchi L, Orsini B, Lenzi J, Del Forno S, et al. Three-dimensional ultrasound evaluation of pelvic floor muscle contraction in women affected by deep infiltrating endo- metriosis: application of a quick contraction scale. J Ultrasound Med. 2022;41(12):2973–9. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.