Core stability in women with and without chronic pelvic pain

Background Physical activity (PA) and core stability levels are unknown for women with musculoskeletal chronic pelvic pain (CPP) compared to healthy women. At this time, there is no validated treatment method aimed specifically at CPP.

To elucidate the PA habits and core stability of women with and without CPP. We hypothesized that aerobic and resistance PA (aPA, rPA) and core stability would differ significantly. Design Prospective, cross-sectional study. Wilcoxon rank sum tests, simple and multiple linear regressions, and Pearson chi-square test were used for analysis. Setting Primary and specialty care outpatient physical medicine and rehabilitation, female pelvic medicine and reconstructive surgery, and gynecology clinics associated with a large tertiary center. Patients We recruited 91 women aged 18–70 years with and without CPP (defined by American College of Obstetricians and Gynecologists). Interventions Not applicable. Main Outcome Measures Physical activity, core stability scores.

Women with CPP demonstrated lower overall core stability (p = .04). Having CPP predicted a 2-point decrease in overall core score (p < .01). Women with CPP demonstrated significantly worse scores for the supine bridge (p < .01) and prone bridge (p = .03). There were no significant differences in aPA or rPA between women with and without CPP. Pelvic floor muscle strength did not have a significant effect on core scores in women with CPP before or after adjusting for age, parity, and body mass index.

This study reports a novel finding that women with CPP had significantly reduced core scores (suggestive of weaker core stability) but similar levels of physical activity (aPA and rPA) compared to women without CPP. Future directions will include larger, multicenter, randomized controlled trials to investigate the association between pelvic floor muscles and core stability as well as therapeutic interventions to include specific exercises targeted at improving overall core stability in women with CPP.

Chronic pelvic pain (CPP) is defined by the American College of Obstetrics and Gynecology (ACOG) as noncyclical, persistent pain in pelvic-related structures lasting at least 6 months.1 Even though CPP occurs in 5.7%–26.6% of women,2 evidence-based therapy is limited, and pain often goes unresolved.1, 3, 4 Symptoms of CPP may be visceral, musculoskeletal, or neurological.4 Many women with CPP experience comorbid psychologic disorders (sleep, anxiety, and depression) and overlapping pain syndromes.4 For example, 60%–70% of women with CPP have endometriosis or interstitial cystitis/bladder pain syndrome and at least 20% of women have irritable bowel syndrome.5 Although visceral overlapping syndromes are more commonly reported, emerging evidence demonstrates a high prevalence of overlapping musculoskeletal disorders in women with CPP including pelvic floor myofascial pain and chronic low back pain.6 Treatment for CPP is generally long term, multimodal, and aimed at addressing underlying causes and comorbidities. Multimodal treatment of CPP may include medications, injections, cognitive behavioral therapy, and acupuncture, but the first line therapy for musculoskeletal CPP is pelvic floor physical therapy (PT).3-5, 7 As a treatment modality, PT typically includes myofascial strategies for pain management, muscle strengthening/relaxation, and guided return to aerobic and resistance physical activity (aPA and rPA). aPA and rPA have been prescribed as first-line nonpharmacologic treatment for other chronic pain syndromes, such as chronic low back pain and fibromyalgia, but outside of PT the use of a prescribed aPA/rPA program for CPP treatment has not been studied to date.8, 9 Current U.S. guidelines recommend healthy adults perform 150–300 minutes of moderate-intensity or 75–150 minutes of vigorous-intensity aPA per week, or an equivalent combination of the two; it is also suggested that adults perform moderate or vigorous muscle-strengthening activities (including upper body, core, and lower body) at least 2 days per week.10 Current evidence shows that only 28% of adults in the United States meet combined aPA and rPA guidelines.11 A 2020 study revealed that up to 52% of adults meet aPA requirements whereas only 29%–35% complete weekly strengthening exercise.11 Other chronic pain populations, including individuals with chronic low back pain and fibromyalgia, have been demonstrated to have lower aPA levels than the general population.12, 13 At the time of design of this study, neither aPA or rPA had been formally studied in women with CPP. However, our now published retrospective data demonstrated that 49.6% of women with CPP were sedentary or performed only light activity and 57.5% did not participate in rPA.14 Additionally, women with CPP may be especially at risk for core weakness given the important contribution of the pelvic floor muscles (PFM) to the core musculature group (diaphragm, transversus abdominus, lumbar multifidi, and pelvic floor).15 At this time, the relationship between CPP, core stability, apA, and rPA has not been studied. In this prospective study, we aimed to compare core stability (using the “Core Score”) and aPA/rPA (using the Godin Leisure Time Exercise Questionnaire [GLTEQ]) in women with and without CPP. Our secondary aim was to assess the relationship between CPP patients' pain and PFM physical examinations to their core stability and PA levels. We hypothesized that women with CPP would have worse core stability and reduced PA levels compared to healthy controls.

Study design, participants, inclusion/exclusion criteria For this prospective cross-sectional study, women were recruited from outpatient physical medicine and rehabilitation, female pelvic medicine and reconstructive surgery, and gynecology clinics. We included English-speaking women ages 18–70 years who presented with CPP as defined by ACOG,1 as well as women controls without CPP or lower urinary tract symptoms. Exclusion criteria were current pregnancy, neurological diseases (ie, multiple sclerosis, movement disorders, neurodegenerative disorders, etc.), abdominal or pelvic surgery in the last 8 weeks, current lower or upper extremity injuries, or current low back pain. All participants were compensated with a free parking pass ($5) at the conclusion of their study visit. Before beginning study procedures, this project was approved by the institutional review board (IRB 215904051322) and all participants signed an informed consent; control women were recruited using the same methods and exclusion criteria for a separate, but similar study that was occurring in tandem (IRB 214646042221). Study measures Data were collected at a single in-person study visit during the participant's clinic appointment. Our study included collection of baseline demographic information (including medical and surgical histories), physical activity levels, and core stability. The primary outcome measure in this study was core stability. Core stability was measured using the Core Stability Score, a validated assessment consisting of four exercises in which the participant is asked to perform 5-second holds to determine their core stability. Exercises include a single-leg balance into a squat, a hip bridge with leg raises, a side plank with a leg raise, and a prone plank with leg raises (Figure 1). Each exercise is given a score of 0–3 based on the quality of the move and ability to hold it. A score of 0–1 indicates poor core stability, and a score of 2–3 indicates good core stability. A total Core Stability Score is calculated out of 12 by summing the 4 individual scores together; a higher score indicates greater core stability.16 The secondary outcome measure was aPA and rPA. The GLTEQ is a validated measure of aPA and asks how often people perform strenuous, moderate, or mild exercise for at least 15 minutes across a 7-day period; scores are calculated by multiplying strenuous value by 9, moderate value by 5, and mild value by 3. These individual scores are then summed; a score of at least 24 is considered sufficiently active and a score <24 is considered insufficiently active.17-19 To our knowledge, there is no validated measure for rPA. To that end, a novel Resistance Activity Questionnaire (RAQ) was used to determine how many days in the last week participants performed lower body, upper body, and core exercises; scores were calculated by summing the number of resistance exercise occurrences per week. Based on World Health Organization guidelines,10 RAQ scores were assessed as inadequate (any regional score of <2 and a total body score of <6), or adequate (all regional scores ≥2 and a total score ≥6). Gold-standard measures for physical activity including VO2 max testing and actigraphy/accelerometry (for aerobic PA), and surface electromyography (sEMG) and wearable sensors (such as inertial measurement units and force/pressure sensors) were cost prohibitive for this unfunded study. In addition, we used data from participants' clinical appointments from the same date as the study visit to assess (1) their level and location of pain via a Body Pain Diagram and Pain Numeric Rating Score,20 (2) anxiety levels via the Generalized Anxiety Disorder 7-item (GAD-7),21 depression levels via the Patient Health Questionnaire (PHQ-9),22 and physical disability via the Pain Disability Index.6 Finally, a retrospective chart review was completed for women with CPP to determine the presence versus absence of myofascial pain with palpation of external and internal PFM during routine clinical PFM examination as well as PFM strength. The pelvic floor physical examination used in the physical medicine and rehabilitation pelvic pain clinic includes manual palpation (single gloved digit) of the PFM including the internal/deep layer of the PFM (levator ani complex and obturator internus) as previously described and validated by Meister et al.,23 and strength was also assessed with a manual (single gloved digit) examination for intravaginal pelvic floor strength via the Modified Oxford Score (0 to 5)24 Regarding the retrospective PFM data, this was obtained from clinic visits ranging from October 29, 2021 to December 6, 2022 at either the time of or prior to the study visit. The PFM exam always occurred on the same day as or before the day of the core stability exam (never after). Statistical analysis Our sample size estimate and power calculation were based on a significance level of α = .05, an anticipated 1:2 ratio of participants with and without CPP, and an anticipated 0.5 probability of healthy controls being active. We aimed for a sample size of 84 to achieve 80% power in order to accurately detect differences conferred by a CPP diagnosis, but recruitment continued for a handful of participants beyond the 84 estimated by the a priori power and sample size analysis to ensure that our study was adequately powered and to minimize the potential loss of information from non-evaluable participants. Given the observational study design and that no criteria were met to justify excluding these additional participants, we opted to retain these additional participants in our analytic sample, resulting in a final sample size of 91 participants. Descriptive statistics were calculated for demographic data. Frequencies and percentages are reported for categorical variables. Means and SDs or medians and interquartile ranges are reported for continuous variables as appropriate. Pearson chi-square tests for independence are used to compare categorical variables and independent, two-sample t-tests are used to compare normally distributed continuous variables. Fisher's exact tests are used in place of chi-square tests where expected frequencies are <5 (Table 1). Component and composite core stability scores were described overall and stratified by CPP diagnosis. Means and standard deviations were reported. Independent, two-sample t-tests compared component and composite core stability scores between women with and without a CPP diagnosis (Table 2). Simple and multiple linear regression models estimated the unadjusted and adjusted effects of a CPP diagnosis on composite core stability scores, respectively. Linear regression models used an identity link and a normal distribution. Estimated mean differences for the effect of a CPP diagnosis in reference to no CPP diagnosis were reported, along with corresponding Wald 95% confidence intervals (CIs) and type 3 Wald chi-square p values. The multiple linear regression model adjusted for patient age, body mass index (BMI), and number of previous deliveries (Table 3). Medians and interquartile ranges were reported to described GLTEQ item and total scores overall and stratified by CPP diagnosis. Wilcoxon rank-sum tests compared GLTEQ item and total scores between women with and without a CPP diagnosis. GLTEQ total scores were calculated as the sum of strenuous activity frequencies multiplied by a factor of 9, moderate activity frequencies multiplied by a factor of 5, and mild activity frequencies multiplied by a factor of 3 (Table 4). Medians and interquartile ranges were reported to describe RAQ item and total scores overall and stratified by CPP diagnosis. Wilcoxon rank-sum tests compared RAQ item and total scores between women with and without a CPP diagnosis. RAQ total scores were calculated as a simple, arithmetic sum of the upper body, lower body, and core component scores (Table 4). Frequencies and column percentages were reported to illustrate the frequency distribution of dichotomized GLTEQ total scores both overall and stratified by CPP diagnosis. GLTEQ total scores were dichotomized as either insufficiently active/sedentary (<24) or sufficiently active (≥24). Univariable and multivariable binary logistic regression models estimated the unadjusted and adjusted effects of a CPP diagnosis on the logit of insufficiently active/sedentary GLTEQ total scores, respectively. Logistic regression models featured a logit link and a binary distribution. Odds ratio estimates were reported for the effect of a CPP diagnosis in reference to no CPP diagnosis on dichotomized GLTEQ total scores, along with corresponding Wald 95% CIs and type 3 Wald chi-square p values. The multivariable model adjusted for patient age, BMI, and number of previous deliveries (Table 5). Simple and multiple linear regression models estimated the unadjusted and adjusted effects of pelvic floor strength on composite core stability scores, respectively. Linear regression models used an identity link and a normal distribution. Estimated mean differences for the effect of a one-unit increase in pelvic floor strength score were reported, along with corresponding Wald 95% CIs and type 3 Wald chi-square p values. The multiple linear regression model adjusted for patient age, BMI, and number of previous deliveries (Appendix A Table A1). | Overall n = 91 | CPP n = 42 | No CPP n = 49 | ||||| |---|---|---|---|---|---|---|---| | Demographic | n | Mean (SD) | n | Mean (SD) | n | Mean (SD) | p | | Age (y) | 91 | 42 (12.9) | 42 | 41 (13.3) | 49 | 43 (12.6) | .51 | | Body mass index (kg/m2) | 91 | 30 (8.0) | 42 | 28 (8.7) | 49 | 32 (7.0) | .05* | | Number of previous deliveries | 90 | 1 (1.0) | 41 | 1 (1.0) | 49 | 1 (1.0) | .17 | | Demographic n (%) | Overall n = 91 (%) | CPP n = 42 (%) | No CPP n = 49 (%) | p value | |---|---|---|---|---| | Race | <.01*a | ||| | White/Caucasian | 45 (49) | 28 (67) | 17 (35) | | | Hispanic/Latina | 12 (13) | 7 (17) | 5 (10) | | | Black/African American | 28 (31) | 6 (14) | 22 (45) | | | Asian | 3 (3) | 0 (0) | 3 (6) | | | Other | 3 (3) | 1 (2) | 2 (4) | | Clinical characteristics n (%) | Overall n = 91 (%) | CPP n = 42 (%) | No CPP n = 49 (%)) | p value | |---|---|---|---|---| | Anxiety | 46 (51) | 29 (69) | 17 (35) | <.01* | | Abdominal/pelvic surgeries | 36 (40) | 18 (43) | 18 (37) | .49 | | Back pain or injury in last 6 months | 14 (15) | 13 (31) | 1 (2) | <.01* | | Depression | 29 (32) | 18 (43) | 11 (22) | .04* | | Dysmenorrhea | 25 (27) | 21 (50) | 4 (8) | <.01* | | Constipation | 22 (24) | 19 (45) | 3 (6) | <.01* | | Irritable bowel syndrome | 13 (14) | 11 (26) | 2 (4) | <.01* | | Endometriosis | 11 (12) | 8 (19) | 3 (6) | .06a | | Interstitial cystitis/painful bladder syndrome | 6 (7) | 6 (14) | 0 (0) | .01*a | | Sexually active | 63 (69) | 29 (69) | 34 (69) | .85 | | Pain with sexual activity | 33 (36) | 31 (74) | 2 (4) | <.01* | | Ever received formal core strengthening instruction | 39 (43) | 20 (48) | 19 (39) | .34 | | Currently receiving formal core strengthening instruction | 13 (14) | 10 (24) | 3 (6) | .02* | | MSK diagnosis n (%) | Overall n = 91 (%) | CPP n = 42 (%) | No CPP n = 49 (%) | p value | |---|---|---|---|---| | Shoulder pain | 15 (16) | 11 (26)b | 4 (8) | .02* | | Wrist pain | 19 (21) | 14 (33) | 5 (10) | .01* | | Low back pain | 39 (43) | 27 (64)b | 12 (24) | <.01* | | Low back arthritis | 7 (8) | 6 (14)b | 1 (2) | .04* | | Pelvic pain | 42 (46) | 40 (95) | 2 (4) | <.01* | | Sacroiliac joint pain | 11 (12) | 10 (24)c | 1 (2) | <.01* | | Hip pain | 20 (22) | 17 (40) | 3 (6) | <.01* | | Knee arthritis | 13 (14) | 11 (26) | 2 (4) | .01* | - Abbreviations: CPP, chronic pelvic pain; MSK, musculoskeletal. - a Fisher's exact test p value. - b Missing 1. - c Missing 3. - * Significant at α = .05 level. | Core stability score mean (SD) | Overall (n = 91) | CPP (n = 42) | No CPP (n = 49) | p value | |---|---|---|---|---| | Single-leg squat | 2.16 (0.86) | 2.19 (0.77) | 2.14 (0.94) | .79 | | Supine bridge | 2.15 (0.71) | 1.88 (0.71) | 2.39 (0.64) | <.01* | | Side bridge | 1.40 (0.84) | 1.26 (0.96) | 1.51 (0.71) | .17 | | Prone bridge | 1.92 (0.92) | 1.69 (0.90) | 2.12 (0.90) | .03* | | Composite score | 7.64 (2.58) | 7.02 (2.65) | 8.16 (2.43) | .04* | - Abbreviation: CPP, chronic pelvic pain. - * Significant at α = .05 level. | Predictor effect | Adjusted mean effect (95% CI)as | p value | |---|---|---| | CPP vs. no CPP (REF) | −1.97 (−2.77, −1.16) | <.01* | - Abbreviations: CI, confidence interval; CPP, chronic pelvic pain. - a One observation excluded due to missing number of previous deliveries data. - * Significant at α = .05 level. | GLTEQ score median (IQR) | Overall (n = 91) | CPP (n = 42) | No CPP (n = 49) | p value | |---|---|---|---|---| | Strenuousa | 0.00 (0.00–2.00) | 0.00 (0.00–2.00) | 0.00 (0.00–2.00) | .92 | | Moderate | 2.00 (0.00–3.50) | 2.50 (1.00–3.50) | 2.00 (0.00–3.00) | .41 | | Mild | 3.00 (1.00–5.00) | 3.00 (1.00–5.00) | 3.00 (0.00–4.00) | .56 | | Totala | 27.00 (15.00–44.00) | 30.00 (17.00–40.00) | 21.00 (11.00–46.00) | .64 | | RAQ score median (IQR) | Overall (n = 91) | CPP (n = 42) | No CPP (n = 49) | p value | |---|---|---|---|---| | Upper body | 0.00 (0.00–2.00) | 0.00 (0.00–2.00) | 0.00 (0.00–2.00) | .43 | | Lower body | 1.00 (0.00–3.00) | 1.00 (0.00–2.00) | 1.00 (0.00–3.00) | .87 | | Core | 1.00 (0.00–3.00) | 1.00 (0.00–3.00) | 0.00 (0.00–2.00) | .21 | | Total | 3.00 (0.00–6.00) | 3.00 (0.00–6.00) | 2.00 (0.00–6.00) | .62 | - Abbreviations: CPP, chronic pelvic pain; GLTEQ, Godin Leisure Time Exercise Questionnaire; IQR, interquartile range; RAQ, (Resistance Activity Questionnaire. - a One participant with CPP missing GLTEQ strenuous and total score data. | Dichotomized GLTEQ total score | Overall (n = 91) | CPP (n = 42)a | No CPP (n = 49) | Unadjusted OR (95% CI)b | p value | Adjusted OR (95% CI)c | p value | |---|---|---|---|---|---|---|---| | Insufficiently active/sedentary (<24) | 40 (44.4) | 14 (34.2) | 26 (53.1) | 0.46 (0.20, 1.08) | .07 | 0.53 (0.21, 1.37) | .19 | | Sufficiently active (≥24) | 50 (55.6) | 27 (65.8) | 23 (46.94) | - Abbreviations: CPP, chronic pelvic pain; GLTEQ, Godin Leisure Time Exercise Questionnaire; OR, odds ratio. - a One participant with chronic pelvic pain missing dichotomized GLTEQ total score data. - b Estimated effect is CPP vs. no CPP (REF). - c Two observations excluded due to missing number of previous deliveries or dichotomized GLTEQ total score data.

Our study cohort consisted of 91 women aged 18–70 years, 42 with CPP and 49 without CPP. Women without CPP had a mean age of 43 years and BMI of 32; women with CPP had a mean age of 42 years and BMI of 28. This difference in BMI was statistically significant (p = .05). Women with CPP were also more likely to have overlapping pain diagnoses and mood disorders including history of back pain, constipation, anxiety (all p ≤ .01), and depression (p = .04). There was no difference between groups regarding having ever received formal core strength training, but women with CPP were more likely to currently be receiving formal training (p = .02) (Table 1). Women with CPP were significantly more likely to have the following musculoskeletal conditions: shoulder pain, wrist pain, low back pain or arthritis, pelvic pain, sacroiliac joint pain, hip pain, or knee arthritis (all p < .04) (Table 1). Women with CPP were found to perform significantly worse at the supine bridge, on average receiving a score of 1.88 compared to 2.14 in women without CPP (p < .01). Similarly, women with CPP averaged a score of 1.69 on the prone bridge compared to 2.12 in women without CPP (p = .03). These contributed to the differences seen in overall core stability scores, where women with CPP averaged a score of 7.02 compared to 8.16 in women without CPP (p = .04) (Table 2). Women with CPP had significantly lower composite core stability scores compared to healthy women (p = .03) and this effect was strengthened after adjusting for age, BMI, and number of previous deliveries (p < .01). On average, composite core scores were 2 points lower for women with a CPP diagnosis compared to healthy women without CPP (Table 3). There was no significant difference in aPA or rPA levels between women with and without CPP based on GLTEQ and RAQ (Table 4). When GLTEQ was dichotomized to insufficiently or sufficiently active, we found that a larger portion of the CPP cohort was sufficiently active, but the odds ratio was not quite statistically significant (p = .07) (Table 5). The 42 women with CPP clinically underwent a PFM exam where 15% had external tenderness, 15% had a positive Q-tip test, and 83% had internal tenderness. They demonstrated a mean pelvic floor strength of 3/5 (n = 34). Thirty of the women with CPP (out of 42) had their PFM exam on the same day as their core stability exam. Pelvic floor strength did not demonstrate a significant effect on composite core stability scores among women with CPP in unadjusted (p = .71) or adjusted (p = .94) analyses. On average, composite core stability scores increased by 0.16 units per each one-unit increase in pelvic floor strength score among women with CPP (mean difference 0.16; 95% CI: −0.68, 1.00). On average, composite core stability scores decreased by 0.02 units per each 1-unit increase in pelvic floor strength score among women with CPP, after adjusting for patient age, BMI, and number of previous deliveries (Mean Difference −0.02; −95% CI: 0.59, 0.55). However, these effects were not statistically significant (Appendix A. Table A1). On a scale of 1–10, the average pain score for women with CPP was 4; their average pain score over the last week prior to clinical appointments was 5. Finally, in the CPP group, results from the GAD-7 demonstrated out of 32 respondents, 12 had minimal to no anxiety, 12 had mild anxiety, and 8 had moderate to severe anxiety; the PHQ-9 showed 18 had no signs of depression, 11 had some level of depression, and 3 had major depression. The Pain Disability Index demonstrated the highest levels of disability with sexual behavior; total scores ranged from 0 to 56 out of 70 in patients with CPP.

Our prospective, cross-sectional study confirmed that women with CPP had significantly worse overall core stability scores than women without CPP. The core stability tests demonstrated specific weakness with the supine and prone bridge in women with CPP, despite the CPP group being more likely to be currently participating in in formal core training at the time of the study. Friedrich et al.'s (2017)16 original core score study was performed in 12 healthy females and 11 males with a mean age of 31.7 years where a majority reported moderate or vigorous PA ≥3 days a week with core strengthening at least once a week. Their mean composite scores were 9.54 (SD, 1.90; range, 4–12) and the lowest composite score was 4. In contrast, our population is older (mean 42 years, SD 12.9) and reported less activity, with no reports of “strenuous activity,” although most women with or without CPP were sufficiently active. Our control group of women without CPP demonstrated a mean core score of 8.16 (SD 2.43), which does fall within an SD of the reported mean of the original standard core score group in healthy participants. Our population with CPP had a mean core score of 7.02 (SD 2.65) which does not fall in the range of the original study group, further highlighting the reduced core stability of women with CPP. It should be noted that our study is the only one to date that has used the core score in the CPP population, so we have no others to compare to. With that being said, we do believe it may prove to be a helpful objective measure for future clinical studies in women with CPP. Contrary to our hypothesis that women with CPP would have decreased aPA and rPA, we found that women with and without CPP had similar levels of PA. This was supported by additional data from our study team that has been recently published but was not available at the time of this study's design.14 In fact, although ultimately not statistically significant, the GLTEQ did demonstrate that CPP women may be more likely to be sufficiently active, but there was still a sizable percentage of women with CPP (45%) who were insufficiently active relative to national guidelines.10 The RAQ showed that like national survey data, neither group of women met guidelines for rPA (Table 4).10, 11 Altogether, this supports further investigation of aPA and rPA as a treatment modality for CPP. Emerging evidence suggests that exercise-induced analgesia may provide pain relief for many chronic pain disorders with studies suggesting that regular exercise results in activation of nascent opioids and serotonin, resulting in pain relief.25, 26 Recent evidence demonstrates support for exercise-induced analgesia in CPP-related syndromes including primary dysmenorrhea,27 low back pain,28, 29 and fibromyalgia.29, 30 A larger study powered towards better understanding PA levels via use of more objective gold-standard measurements (ie, VO2 max testing, accelerometry/actigraphy, sEMG, and wearable sensors such as inertial measurement units or force/pressure sensors) in women with CPP may further elucidate our understanding of how aPA and rPA participation influences CPP. Although the women in this study demonstrated similar rPA and aPA levels, they did have some significant differences in their clinical characteristics. The CPP group had a significantly lower BMI, possibly related to having a slightly higher likelihood of being sufficiently active with aPA, though this could also be multifactorial. Other studies investigating disorders like CPP, including endometriosis31 and interstitial cystitis,32 have also found that afflicted women have lower BMIs compared to heathy control, indicating this may be a characteristic of these populations. Alternatively, it should be noted that our study did not inquire about any history of or current eating disorders or body dysmorphia. A 2023 study on women with genitopelvic pain and penetration disorders demonstrated elevated levels of disordered eating (86% of their study population) as well as sexual trauma,33 and a 2018 ACOG bulletin identifies pelvic pain as a gynecological concern for adolescents and young adult women with eating disorders presenting for gynecologic care.34 Our study team observed that there appeared to be a subset of participants who were relatively young, had low-normal BMIs, and often reported a history of activities such as distance running or cycling. Although no conclusions can be drawn here, further investigation of this apparent subset of CPP patients is warranted considering the comorbidities associated with eating disorders, such as osteoporosis, that may affect our understanding of their unique disease experience. The women with CPP were also more likely to have overlapping pain diagnoses and mood disorders, which may be expected given the prevalence of such overlapping syndromes with CPP. It is also particularly worth noting the high prevalence of wrist, shoulder, back, and hip pain found in our patient population with CPP compared to healthy controls (Table 1), as this highlights the overlapping musculoskeletal pain disorders clearly present in the population. The most common comorbidities (hip and back pain) affected 40% and 64% of our cohort with CPP, respectively; considering the overlapping muscle groups between the hips, low back, and pelvic floor, the results support the importance of screening for overlapping musculoskeletal disorders in CPP and including treatment of any overlapping diagnoses in the PT prescription. In particular, therapeutic core stability exercises have been extensively studied in nonspecific low back pain. A recent systemic review analyzed five moderate-quality studies with similar core stabilization treatments and measures of pain, disability, and function. They concluded that core stabilization exercises are overall an effective therapeutic option for nonspecific low back pain, consistently resulting in lower pain ratings, improved muscular function, and improved quality of life.35 Given the clear relationship between back pain and CPP shown here, the core stability data presented here provide a possible starting point for further investigation of the muscle groups and core stability exercises that could be targeted in women with CPP. To our knowledge, this is the first prospective study to objectively assess core stability and aPA/rPA levels in women with and without CPP. The study exceeded its a priori sample size calculation to ensure it would be adequately powered, but this resulted in more balanced study groups, which boosted statistical power. The use of the CSS is a strength as it is an objective measure of core stability and was administered by examiners who were trained in scoring the maneuvers, and a prior study showed substantial inter-rater reliability.16 The remainder of outcomes assessed (GLTEQ, RAQ, GAD-7, PHQ-9) were patient-reported outcomes and are therefore limited due to recall bias. Ideally, objective measures of PA (actigraphy/accelerometry, VO2 max testing, sEMG, and wearable sensors) would be used but were cost prohibitive for this study. Additionally, the reporting of pelvic floor examination findings is a strength in that it better defined the group with CPP but is limited in that it was collected retrospectively via chart review and there was variation in components of the physical examination depending on examiner, which led to incomplete data. Nonetheless, the data from this secondary outcome were considered of value in that they may inform future prospective studies in the assessment of PFM strength/function and core stability in women with CPP. Furthermore, the data were available only in the group with CPP as they were not collected in healthy controls as part of their routine clinical care, so comparison between women with CPP and healthy controls could not be completed but would be important in future studies. This study had a diverse participant population but was limited by our design as a single visit collected from a single center, which reduces the heterogeneity and generalizability. Finally, the study was not powered to answer questions about core stability and PFM strength, but this should be addressed in a future study.

This study reports a novel finding that women with CPP had significantly reduced CSS (suggestive of weaker core stability) but similar levels of physical activity (aPA and rPA) compared to women without CPP. To better understand these relationships between CPP, core stability, aPA, and rPA, a large, prospective multicenter trial using gold-standard measures of PA (actigraphy and VO2 max testing) would help to validate these findings and better characterize the exercise habits of women with CPP compared to women without CPP. Furthermore, future randomized controlled trials to assess the impact of core stability and aPA/rPA programs on CPP are needed to improve our understanding of these as treatment options for women with CPP. DISCLOSURE None of the authors have a conflict of interest to disclose. APPENDIX A | Predictor effect | Unadjusted mean effect (95% CI)a | p value | Adjusted mean effect (95% CI)b | p value | |---|---|---|---|---| | 1-unit increase in pelvic floor strength score | 0.16 (−0.68, 1.00) | .71 | −0.02 (−0.59, 0.55) | .94 | - Abbreviation: CI, confidence interval. - a 8 observations excluded due to missing pelvic floor strength score data. - b 9 observations excluded due to missing pelvic floor strength score or number of previous deliveries data. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available from the corresponding author upon reasonable request.