Pain with orgasm in endometriosis: potential etiologic factors and clinical correlates

Abstract

Background

Pelvic pain worsened by orgasm is a poorly understood symptom in patients with endometriosis.

Aim

To assess the prevalence of pelvic pain worsened by orgasm in patients with endometriosis and explore its association with potential etiologic factors, including pelvic floor myalgia, uterine tenderness and adenomyosis, and central nervous system sensitization.

Methods

An analysis was done of a prospective data registry based at a tertiary referral center for endometriosis. Eligible participants were patients aged 18 to 50 years who were referred between January 1, 2018, and December 31, 2019, diagnosed with endometriosis, and subsequently underwent surgery at the center. Clinical features were compared between participants reporting worsening pelvic pain with orgasm and those without worsening pain with orgasm, including patient-reported variables, physical examination findings, and anatomic phenotyping at the time of surgery. Pelvic floor myalgia and uterine tenderness were assessed by palpation on pelvic examination, adenomyosis by ultrasound, and central nervous system sensitization via the Central Sensitization Inventory (range, 0-100).

Outcomes

Outcomes included pelvic or lower abdominal pain in the last 3 months that worsened with orgasm (yes/no).

Results

Among 358 participants with endometriosis, 14% (49/358) reported pain worsened by orgasm while 86% (309/358) did not. Pain with orgasm was significantly associated with pelvic floor myalgia (55% [27/49] vs 35% [109/309]; Cohen’s h = 0.40, P = .01) and higher scores on the Central Sensitization Inventory (mean ± SD, 53.3 ± 17.0 vs 42.7 ± 18.2; Cohen’s d = 0.60, P < .001) but not with uterine tenderness or adenomyosis. Other clinical features associated with pain with orgasm were poorer sexual health (higher scores: deep dyspareunia, Cohen’s h = 0.60; superficial dyspareunia, Cohen’s h = 0.34; and Female Sexual Distress Scale–Revised, Cohen’s d = 0.68; all P < .05) and poorer mental health (higher scores: Patient Health Questionnaire–9, 12.9 ± 6.7 vs 9.1 ± 6.3, Cohen’s d = 0.59, P < .001; Generalized Anxiety Disorder–7, 9.4 ± 5.6 vs 6.8 ± 5.5, Cohen’s d = 0.48, P = .002). Anatomic findings at the time of surgery did not significantly differ between the groups.

Clinical Implications

Interventions targeting pelvic floor myalgia and central nervous system sensitization may help alleviate pain worsened by orgasm in patients with endometriosis.

Strengths and Limitations

A strength is that pain worsened by orgasm was differentiated from dyspareunia. However, pain with orgasm was assessed by only a binary question (yes/no). Also, the study is limited to a single center, and there were limited data on sexual function.

Conclusion

Pelvic pain exacerbated by orgasm in people with endometriosis may be related to concurrent pelvic floor myalgia and central sensitization.

Introduction

Endometriosis is a chronic condition associated with proliferation of ectopic endometrial-like tissue estimated to affect 10% of reproductive-aged women and individuals with a uterus worldwide. Anatomic subtypes include deep infiltrating endometriosis nodules, ovarian endometriosis, and superficial peritoneal endometriosis.¹ Patients with endometriosis often present with significant pain symptoms, including dysmenorrhea (painful menstrual cramps), dyspareunia (typically deep pain during sex), dyschezia (painful bowel movements), and chronic pelvic pain.¹

Pain due to endometriosis is multifactorial, resulting from peripheral and central mechanisms.² Peripheral mechanisms include inflammation and local innervation changes induced by endometriosis lesions, increased pain sensation (hyperalgesia) in the uterus due to coexisting adenomyosis (endometrial-like cells in the myometrium), and local reflexive myofascial pain (eg, pelvic floor myalgia) due to nociceptive factors.³ In the central nervous system, amplification of central nociceptive pathways (central sensitization) results in more widespread pain, hyperalgesia, and/or allodynia.^4,5 Central sensitization can also contribute to development of comorbid pain conditions such as pelvic floor myalgia and be associated with psychological factors including anxiety and depression.^6,7

The female orgasm involves repetitive reflex-mediated contractions of the pelvic floor skeletal muscles.⁸ Thus, pelvic pain with orgasm in endometriosis may be related to pelvic floor myalgia. In addition, pelvic floor contractions with orgasm may be accompanied by rhythmic cervicouterine contractions.^8,9 Therefore, uterine factors (eg, adenomyosis or an otherwise tender uterus) can be another contributor to orgasm pain. Moreover, a report of uncommon periorgasmic symptoms included 3 cases of pain with orgasm that were treated successfully with amitriptyline, which suggests a possible origin in the central nervous system.¹⁰

In this analysis of a prospective registry, we explored the prevalence of pelvic or lower abdominal pain worsening with orgasm in people with endometriosis, by using a binary question (yes/no) from a series of questions about chronic pelvic pain from an endometriosis questionnaire, as well as its associated clinical features. We hypothesized that pain worse with orgasm in endometriosis would be associated with pelvic floor myalgia, uterine factors (tenderness and adenomyosis), and underlying central sensitization.

Method

Study setting

The prospective data registry Endometriosis Pelvic Pain Interdisciplinary Cohort (EPPIC) is based at a tertiary referral center with gynecologists specializing in pelvic pain and endometriosis surgery. EPPIC was designed to examine participant-reported outcomes and factors associated with different types of pelvic pain among people with endometriosis (ClinicalTrials.gov NCT02911090).¹¹ Ethics approval was obtained from the institutional research ethics boards (H22-00586).

Inclusion and exclusion criteria

We included EPPIC-consented participants who were newly or re-referred to the center for an initial baseline visit appointment between January 1, 2018, and December 31, 2019. Inclusion criteria for the study were age |$\ge$|18 and |$\le$|50 years, diagnosis of endometriosis, and completion of prospective surgery for endometriosis at the center after the initial visit. The endometriosis diagnosis inclusion criterion was defined as 1 or more of the following at baseline: (1) previous surgical diagnosis before the initial visit (visual-only diagnosis or histologic confirmation); (2) current ovarian endometriosis cyst informed by endovaginal ultrasound at the initial visit; (3) current deep infiltrating endometriosis nodule informed by palpation during pelvic examination or endovaginal ultrasound at the initial visit; or (4) otherwise clinically suspected endometriosis by the gynecologist, based on history and physical examination at the initial visit in the absence of a previous negative laparoscopy result. In contrast, endometriosis anatomic findings at the prospective surgical procedure done at the center after baseline were utilized for associations with pain with orgasm.

The flow of participants is shown in Figure 1. Participants were excluded if they were postmenopausal (spontaneous or surgical) or had missing data from the participant-reported Central Sensitization Inventory (CSI) questionnaire.

This cohort was derived from a cohort used to assess the CSI and surgical outcomes in endometriosis.¹²

Data collection

The EPPIC registry is compliant with the clinical data standards of the World Endometriosis Research Foundation’s EPHect Project (Endometriosis Phenome and Biobanking Harmonization).¹³ EPPIC prospective registry data on variables were collected at the following time points (Table 1):

• At baseline: Prior to the initial appointment at the center, participants completed a preappointment survey using an online questionnaire as part of the online EPPIC registry.

• At baseline at the first appointment at the center: Physical examination and ultrasound were done, and data were recorded by the clinician on a standardized paper form and then entered into the online EPPIC registry by the research team.

• At prospective surgery at the center: After baseline, surgical data were entered by clinicians on the day of surgery directly into the online EPPIC registry.

Variables

The main outcome of interest was general pelvic or lower abdominal pain worsened by orgasm from the baseline preappointment survey (Table 1). It was assessed by participant responses to a series of questions with branching logic as shown in Figure 2. Specifically, participants were shown a diagram to define the pelvis and lower abdomen. Those who checked “orgasm” in the final question were defined as the study group, having pelvic pain in the last 3 months worsened by orgasm. In the original EPHect questionnaire, this question combined “intercourse or orgasm”; however, we separated the 2 entities when implementing EPHect in the EPPIC registry.¹³ Participants’ comprehension of the questionnaire items after separating the 2 entities was not assessed prior to the study. The comparison group consisted of the following participants, as also shown in Figure 2: those without general pelvic pain (ever), those with pelvic pain but >3 months ago, or those with pelvic pain in the last 3 months but not worse with orgasm. A sensitivity analysis was done by comparison with only the last group (ie, pelvic pain in the last 3 months but not worse with orgasm).

Clinical variables from EPPIC were conceptualized in 6 categories (collected at the time points shown in Table 1). First, demographic variables were age and body mass index. Second, the uterine-related variables were the presence of uterine or cervical tenderness on digital or ultrasound pelvic examination¹⁴ and evidence of uterine adenomyosis on endovaginal ultrasound.¹⁵ The third set, sexual health variables, were severity of deep dyspareunia on a 11-point numeric rating scale (0 = no pain, 10 = worst pain imaginable), severity of superficial dyspareunia (0 = no pain, 10 = worst pain imaginable),¹⁶ and the Female Sexual Distress Scale–Revised (FSDS-R; range, 0-52) with higher scores indicating more sexual distress (Cronbach’s alpha = 0.96).¹⁷ The FSDS-R is a 13-item self-report scale assessing sexually related personal distress.¹⁸ Deep dyspareunia was defined as pelvic pain with deep penetration during sexual activity, while superficial dyspareunia was defined as pain at the vaginal introitus.¹¹ The fourth set was other types of pain variables: severity of dysmenorrhea, dyschezia, chronic pelvic pain, and back pain (all measured on an 11-point numeric rating scale; 0 = no pain, 10 = worst pain imaginable).¹⁶

The fifth set of variables was related to pelvic pain comorbidities and underlying central nervous system sensitization. Central sensitization was assessed with the CSI part A (range, 0-100), with higher scores indicating more symptoms of central sensitization (Cronbach’s alpha = 0.92), which we have validated in the endometriosis population.^6,19,20 CSI part A assesses the frequency of 25 symptoms that are common in central sensitivity syndromes, such as pain beyond the pelvis/abdomen, sensitivity to environmental stimuli, and sleep disturbance, each scored to result in a total ranging from 0 to 100.¹⁹ Pelvic pain comorbidities included the following:

• Pelvic floor myalgia: defined as tenderness on pelvic examination digital palpation of the right or left levator ani muscle^21-23

• Irritable bowel syndrome diagnosed by Rome III criteria²⁴

• Painful bladder syndrome diagnosed with criteria from the American Urological Association or International Continence Society^25,26

• Pelvic girdle pain defined as 1 of the following findings on gynecologist-reported examination: right or left dorsal sacroiliac ligament tenderness, pain on right or left active straight-leg raise, positive Faber test result on right or left, right or left tenderness on posterior pelvic pain provocation test²⁷

• Abdominal wall pain diagnosed by a positive Carnett’s sign in at least 1 of the following regions of the abdomen: right lower quadrant, left lower quadrant, or suprapubic (often related to myofascial trigger points)^16,28,29

• Generalized Anxiety Disorder–7 (GAD-7; range, 0-21), with higher scores indicating more severe anxiety (Cronbach’s alpha = 0.92)³⁰

• Patient Health Questionnaire–9 (PHQ-9; range, 0-27), with higher scores indicating more severe depression (Cronbach’s alpha = 0.90)³¹

• Pain Catastrophizing Scale (PCS; range, 0-52), with higher scores indicating more severe pain-related catastrophic thinking (Cronbach’s alpha = 0.95)—specifically, the constructs of rumination, magnification, and helplessness in the setting of experienced or anticipated pain³²

Findings at the time of prospective surgery at the center were the sixth set of variables (Table 1): visual diagnosis of endometriosis at the time of surgery, divided into the 3 anatomic subtypes; pouch of Douglas obliteration; score on the revised surgical staging system of the American Society for Reproductive Medicine (range, 1 - >40; with higher scores indicating more extensive endometriosis)³³; and histologic confirmation of endometriosis after surgical excision.

Data analysis

The primary analyses based on our hypothesis compared the 2 groups (pain worse vs not worse with orgasm) for the following variables: pelvic floor myalgia, uterine variables (tenderness or adenomyosis), and central sensitization as measured by the CSI score. Secondary analyses were performed on the remaining variables from the 6 conceptual categories. Categorical variables were compared by chi-square test or by Fisher’s exact test if any subgroup sample size was |$\le$|5. For continuous variables, comparisons were made with independent samples t-tests (Welch’s t-test for unequal variances). The Benjamini-Hochberg correction for multiple comparisons was done with a false discovery rate of 0.05.³⁴ All comparison analyses were performed with IBM SPSS Statistics version 27.0.1. Effect sizes were measured with Cohen’s h for categorical variables and Cohen’s d for continuous variables.³⁵ Means are shown with SD. Significance was set at P  |$\le$| .05.

Results

Primary analyses

There were 358 participants in the EPPIC registry who met the study criteria (Figure 1). Socioeconomic and baseline characteristics of the sample are described in Table 2. The average age of participants was 34.3 years (SD, 7) and their average body mass index was 26.5 (SD, 6) kg/m². Sexual orientations were heterosexual (n = 299, 84%), lesbian (n = 13, 4%), bisexual (n = 33, 9%), asexual (n = 1, 0.3%), or none of these categories or preferred not to answer (n = 12, 3%). For gender, 5 participants (1%) indicated nonbinary or gender queer. Most participants in the analysis were in a relationship (n = 286, 80%). Ancestries are listed in Table 2.

Of the 358 participants, 49 (14%) reported general pelvic or lower abdominal pain in the last 3 months that is worsened by orgasm (study group), while 309 (86%) did not (comparison group). In terms of endometriosis diagnosis inclusion criteria, pain with orgasm was present in 12% (21/174) of those with a previous surgical diagnosis before the initial visit, 9% (8/90) with a current ovarian endometrioma, 9% (2/23) with a current deep endometriosis nodule, and 20% (25/126) with otherwise clinically suspected endometriosis based on history/examination (note: categories not mutually exclusive).

Table 3 shows comparisons of clinical features between participants who reported pain worsened by orgasm and those who did not. Participants with pain worsened by orgasm more often had pelvic floor myalgia (P = .01, Cohen’s h = 0.40) and higher scores on the CSI (P < .001, Cohen’s d = 0.60). In contrast, there were no differences in uterine tenderness or ultrasound evidence of adenomyosis between participants who had pain worsened by orgasm and those who did not.

Secondary analyses

Pain worsened by orgasm was associated with significantly worse sexual health variables (Table 3): more severe deep dyspareunia (P < .001, Cohen’s d = 0.60), more severe superficial dyspareunia (P = .04, Cohen’s d = 0.34), and worse sexual distress (FSDS-R; P < .001, Cohen’s d = 0.68).

In addition to deep and superficial dyspareunia, pain worse with orgasm was associated with worse dyschezia (P = .04, Cohen’s d = 0.33) and worse chronic pelvic pain (P < .001, Cohen’s d = 0.57). There were no significant associations with dysmenorrhea or back pain.

Pain worsened by orgasm was significantly associated with depression (PHQ-9; P < .001, Cohen’s d = 0.59) and anxiety symptoms (GAD-7; P = .002, Cohen’s d = 0.48). However, there were no significant differences in irritable bowel syndrome, painful bladder syndrome, pelvic girdle pain, abdominal wall pain, or PCS score.

At the time of surgery, there were no differences in anatomic characteristics of the endometriosis between the groups except for a significant difference in the presence of ovarian endometriosis (P = .048, Cohen’s h = 0.35). There was a higher proportion of ovarian endometriosis in people with no pain with orgasm vs those with pain (52% vs 35%).

When the Benjamini-Hochberg correction for multiple comparisons was applied with a false discovery rate of 0.05, the following variables remained significantly associated with pain worsened by orgasm (Table 3): pelvic floor myalgia, CSI score, deep dyspareunia, sexual distress (FSDS-R), chronic pelvic pain, anxiety (GAD-7), and depression (PHQ-9).

Sensitivity analysis

For the comparison group (n = 309), 86 participants reported no general pelvic pain (ever), 30 had pelvic pain but >3 months ago, and 193 had pelvic pain in the last 3 months but not worse with orgasm. To determine whether these results were affected by the inclusion of participants who did not have any pelvic or lower abdominal pain in the last 3 months, we removed these participants in a sensitivity analysis of the significantly associated variables. Therefore, participants who had pelvic pain in the last 3 months worsened with orgasm were compared with those who had pelvic pain in the last 3 months but reported that their pain did not worsen with orgasm (n = 193; Table A1). The findings remain unchanged: pelvic pain worse with orgasm was significantly associated with pelvic floor myalgia, CSI score, deep dyspareunia, sexual distress (FSDS-R), chronic pelvic pain, anxiety (GAD-7), and depression (PHQ-9).

Discussion

In this study of 358 participants at a tertiary referral center, 14% of participants with endometriosis reported pelvic or lower abdominal pain worsened by orgasm. In line with our hypothesis, we found that pain with orgasm was significantly associated with pelvic floor myalgia and a higher CSI score; in contrast, there was no association with uterine tenderness or adenomyosis. Participants who had pain with orgasm also reported worse pain severity for deep dyspareunia and chronic pelvic pain, as well as higher sexual distress, depression, and anxiety scores, after correction for multiple comparisons. There was some specificity to these associations, as pain with orgasm was not associated with superficial dyspareunia, other musculoskeletal pain conditions (abdominal wall pain or pelvic girdle pain), or visceral pain conditions (irritable bowel syndrome and painful bladder syndrome). These associations remained significant after a sensitivity analysis. No strong correlations with endometriosis anatomy at the time of surgery were observed.

We hypothesized that pelvic pain worsened by orgasm would be associated with pelvic floor myalgia, given the pelvic floor contractions that occur with orgasm.⁹ This may arise from abnormal tone or coordination of the muscles. The relationship with pelvic floor myalgia was supported by the association between pain worse with orgasm and deep dyspareunia and dyschezia that can arise from pelvic floor dysfunction.^22,23,36 Therefore, pelvic floor physiotherapy is likely to be an important treatment modality for management of pain worse with orgasm in endometriosis.³⁷ These results are supported by a recent study showing that among patients with provoked vulvodynia, worse pelvic floor muscle pain correlated with lower scores on the orgasm and pain subscales of the Female Sexual Function Index, while worse vulvar mucosal pain did not; the association with the orgasm subscale score remained significant after adjusting for potential confounders.³⁸

In contrast, uterine factors (tenderness and adenomyosis) were not associated with pain worse with orgasm. This contrasts with a recent case report of putative uterine origin of painful orgasm in endometriosis.³⁹ It may be that cervicouterine contractions occur only in a subset of people with orgasm; thus, a larger sample size may be needed to detect an association with uterine factors. The frequency of uterine contractions during orgasm in the population is not well described.⁴⁰ Regardless, uterine origin of pain worse with orgasm appears to be rarer than that of pelvic floor origin.

The role of central sensitization is supported by higher CSI scores in participants with pelvic pain worsened by orgasm. This is further supported by the associations with chronic pelvic pain, anxiety, and depression. Pelvic floor myalgia itself can also be secondary to central sensitization.⁶ However, other pain comorbidities known to be associated with central sensitization (eg, irritable bowel syndrome and painful bladder syndrome) had no relation to pain with orgasm. These differential results support a specific relationship between orgasm pain and pelvic floor somatic pain rather than visceral pain from the bowel and bladder. Interestingly, there is evidence that orgasm and pain share neurologic pathways in the central nervous system such that orgasm could be considered a form of nonaversive pain where orgasm can reduce pain and vice versa.⁴¹

In terms of sexual health variables, pain worse with orgasm was associated with dyspareunia severity. This relationship may be due to pelvic floor myalgia or underlying central sensitization. In addition, pain during vaginal penetration may prime the nervous system for amplification of pain with orgasm. Hence, one therapeutic strategy could be to target dyspareunia itself, with the potential secondary effect of reducing pain at the time of orgasm. As would be expected, pain with orgasm was significantly associated with greater sexual distress on the FSDS-R.

Note that there were no clear associations between pain worse with orgasm and anatomic findings of endometriosis at the time of surgery. There was an apparent difference with the presence of ovarian endometriosis being associated with less frequent pain with orgasm; however, this was not significant after correction for multiple comparisons, and is likely due to ascertainment bias related to referral pattern, as our center receives fertility referrals of patients with ovarian endometriosis cysts. Surgical findings in endometriosis correlate only partially with pain symptoms, so it is unsurprising that there were no significant differences in this study.¹

A final note should be made about terminology. It is not clear that “dysorgasmia” is synonymous with pelvic pain worsened with orgasm in endometriosis.¹⁰ Dysorgasmia may be more idiopathic, whereas participants in the registry have a known potential etiologic factor (endometriosis). Pain with orgasm was not part of the recently published nosology for female orgasm disorders.⁴² However, there may be overlap with the concept of female orgasmic illness syndrome in the nosology, as this syndrome includes abdominal pain as one of the peripheral aversive symptoms.⁴²

Strengths of the study are inclusion of participants with clinically diagnosed endometriosis who prospectively had surgery done at a tertiary referral center for endometriosis surgery, as well as the use of a normalized database.⁴³ Surgical data were prospectively recorded on the day of surgery by the clinical team and directly entered into a standardized online form, allowing for detailed and accurate phenotyping of endometriosis anatomic findings at the time of surgery. A normalized database allowed comparisons of multidimensional clinical features between a cohort of participants with endometriosis who have pain with orgasm and those who do not. Primary limitations are the small sample size of participants with orgasm-associated pain and the use of a single-item measure with binary responses to describe the independent variable. A binary response to indicate pain worsened by orgasm does not capture characteristics such as timing, quality, and location, and is subject to participants’ interpretations of worse pain. Mildly worsened pain that was considered insignificant may be unreported, therefore masking a result. There were also no data on the frequency of orgasm in the past 3 months or whether orgasms were partnered or nonpartnered; in addition, we had data on sexual distress (FSDS-R) but not sexual function in the registry. We did not perform a pilot study to assess participants’ understanding of the question regarding pain with orgasm, although there was only a mild-moderate correlation between the questions about pain with intercourse and pain with orgasm (correlation, phi = 0.361), which suggests that participants were able to differentiate between the types of pain. A sample from a single tertiary center biases toward more severe symptoms that may not represent the general population with endometriosis.

Conclusions

Pelvic pain worsened with orgasm can occur in endometriosis, with potential etiologic factors including pelvic floor myalgia and underlying central nervous system sensitization. Further research is needed to assess the prevalence, impact, and possible treatments for pain with orgasm in patients with endometriosis.

Acknowledgments

We thank Dr. Christina Williams for her contributions to participant recruitment.

Author contributions

A.D.: data analysis, interpretation of results, and manuscript writing. H.N.: data acquisition and critically reviewing the manuscript. K.N.B.: results interpretation and critically reviewing the manuscript. M.A.B., C.L., and C.A.: data acquisition and critically reviewing the manuscript. N.L.O.: data acquisition and critically reviewing the manuscript. P.J.Y.: conception and design of the study, data acquisition, data analysis, interpretation of results, and manuscript writing. All authors reviewed the results and approved the final version of the manuscript.

Funding

This work received support from the Canadian Institutes of Health Research (grant PJT-156084).

Conflicts of interest

C.A. has financial affiliations with AbbVie and Pfizer. M.A.B. has financial affiliations with AbbVie, Baxter, and Ferring.

Appendix

References