Vestibulodynia is a localized pain condition of the vulvar vestibule and is defined as vulvar discomfort, most often described as burning pain, occurring in the absence of relevant visible findings or a clinically identifiable disorder.1 This condition affects approximately 8–15% of women.2–4 Although vestibulodynia is thought to be a leading cause of dyspareunia,5 this condition has been traditionally associated only with premenopausal women. Dyspareunia has been poorly described in menopause, perhaps as a result of assumptions that pain relates simply to tissue aging and hormonal changes.
Addressing dyspareunia and sexual concerns is important throughout a woman's lifetime. Studies indicate that women continue to stay sexually active through the menopause transition and share concerns of dyspareunia and sexual dysfunction similar to their younger counterparts.6,7 Because vulvovaginal hypoestrogenism contributes to poor lubrication and difficult arousal,7–9 many menopausal women treat these symptoms with estrogen supplementation or artificial lubricants, yet when dyspareunia persists, vestibulodynia in this population is often not considered. Persistence of vulvar pain after estrogen support in menopausal women merits evaluation for causes beyond atrophy, including vestibulodynia.
Our vulvar research group has analyzed 109 (88 premenopausal and 21 postmenopausal) archived surgical tissues from patients with vestibulodynia who underwent a vestibulectomy between 2002 and 2008. We recently reported on the histopathologic profile of the 88 premenopausal women with primary and secondary vestibulodynia.10 Primary patients were those who had always noted pain with penetration and secondary patients first had no dyspareunia but then acquired the symptom. Both types of vestibulodynia had histologic features of neurogenic inflammation, which is characterized by abundant lymphocytes, mast cells, and neural hyperplasia or hypertrophy.10 However, there were also significant differences between these two subtypes. Vestibulectomy specimens from women with premenopausal primary vestibulodynia had statistically more neural hyperplasia than did tissue from patients with premenopausal secondary vestibulodynia who in turn had greater lymphocytic infiltration.
In this retrospective analysis, we now describe the histopathologic findings in 21 postmenopausal women with vestibulodynia. We compare these findings with the 88 premenopausal cases from our prior report.10 By comparing these three groups (premenopausal primary, premenopausal secondary, postmenopausal), our objective was to assess whether premenopausal and postmenopausal vestibulodynia have different histologic features that may provide insights into the underlying pathophysiology of this common and difficult condition. Our hypothesis was that vestibule tissues from menopausal women with secondary vestibulodynia would be characterized by neurogenic inflammation similar to premenopausal women with secondary vestibulodynia.
MATERIALS AND METHODS
Using an institutional review board-approved protocol (#3613), we identified 109 archived vestibulectomy specimens from surgeries performed at Oregon Health & Science University from 2002 to 2008. Twenty-one cases were from postmenopausal women diagnosed with secondary vestibulodynia. Eighty-eight cases were from premenopausal women (42 primary and 46 secondary) who have already been described.10 All archived tissue specimens were from patients seen in the Oregon Health & Science University Program in Vulvar Health who met Friedrich criteria for vestibulodynia and had no other identifiable reason for dyspareunia (eg, vulvar dermatosis, infectious vaginitis).11 All patients had reported severe or moderate dyspareunia and had undergone a superficial, localized vestibulectomy.12 Chart review was performed independently by two of the authors (M.F.G. and C.M.L.) to verify patient age, parity, postmenopausal status, primary compared with secondary vestibulodynia classification, hormone therapy (type, delivery, and dose), duration of symptoms, and duration since menopause. Menopause was defined as amenorrhea for at least 12 months or surgical removal of both ovaries. All 109 tissues were stained and analyzed together.
All specimens were obtained from superficial localized vestibulectomies in which the hymen, and occasionally 1 cm of distal vagina, was used for closure.12 Routine paraffin-embedded histologic sections (3–5 micrometers were stained for hematoxylin and eosin to grade chronic inflammation (lymphocytes).10,13 Serial sections were also immunostained for S100 (nerves), CD117 (mast cells), presence of progesterone receptor, and presence of estrogen receptor α.10,13 Notably, estrogen receptor β and epidermal nerve twig staining with protein gene product 9.5 were not informative in our prior study13 and were therefore not used in this analysis. Staining quality and specificity were confirmed using appropriate controls (eg, nerves for S100 and breast tissue for hormone receptors.
Histologic features were scored by two pathologists blinded to clinical and demographic data. Scoring was performed as described in our prior studies where this method was shown to be highly reproducible.10,13 There was good to excellent agreement between pathologists scoring histologic sections for inflammation, neural hyperplasia, and hormone receptors (kappa statistic ranged from .63 to greater than .8). Briefly, chronic inflammation (lymphocytes) was scored as none (0), mild (1+), moderate (2+), or severe (3+). S100-positive nerves were scored as none (0), mild (1+), moderate (2+), or severe (3+) hypertrophy or hyperplasia. Estrogen and progesterone receptor nuclear immunostaining was scored similar to routine breast pathology using a 0–3+ scale: 0=negative; 1+=less than 10%; 2+=11–50%; 3+=51–100% of basal cells. CD117-positive mast cells were quantified as the average number of mast cells per high-power microscopic field (40× objective).
Statistical analyses of the data were performed with SAS statistical software. Patients' demographic and clinical characteristics among study groups were compared using χ2 for parity and GLIMMIX (generalized linear mixed model) procedure for age and duration of symptoms to accommodate nonnormal distribution.14 Median and interquartile range were presented because age, years of symptoms, and years since menopause were skewed (Table 1). As a result of the small sample size, both proportional odds logistic regression models15 for four-category ordinal outcome and binary exact logistic regression models16 for collapsed two-category outcome were used to examine the histopathologic outcome. The proportional odds logistic regression models take advantage of the ordinal nature of the outcomes and are more powerful than binary logistic regression for a given sample size. Results from proportional odds logistic regression models are presented. Both unadjusted and adjusted proportional odds ratios (ORs) and 95% confidence intervals (CIs) with consideration of parity and natural logistic logarithmic transformed duration of symptoms were also estimated (Table 2). Data for mast cell count were normally distributed. The difference of mast cells was determined among groups using analysis of variance; mean and standard error of mean were presented. Two-sided P values were reported; P<.05 was considered statistically significant.
Our sample size had the power to detect a difference based on numbers from our prior premenopausal study.10,13 Sample size analyses used 80% power with a significance level of .05 for all calculations. A 1:2 allocation ratio was used to compare the available 21 postmenopausal samples with 42 premenopausal primary vestibulodynia samples and 46 premenopausal secondary vestibulodynia samples. Neuroinflammation is the hallmark of tissue changes in vestibulodynia and represents our main outcome. Based on the finding of moderate to strong staining for neural hyperplasia and inflammation in 52–76% of premenopausal groups, our sample size can detect a 30–33.5% difference in neural hyperplasia and inflammation using the two-sided z test. Based on rates of positive staining for estrogen receptor α and progesterone receptor in 70–88% of premenopausal tissues, our sample size can detect a 27–32% difference between the premenopausal primary or secondary group and postmenopausal group using the one-sided z test with pooled variance. According to the mean of mast cell counts of 26 or 28 for premenopausal groups, the same sample achieves the proposed study power and significance level to detect a mean difference of 8 between one of the premenopausal groups and the postmenopausal group with a standard deviation of 10 and the two-sided t test.
All 21 postmenopausal patients had secondary vestibulodynia. Demographic information including median age, years since menopause, symptom duration, and parity is listed in Table 1. Seventy-one percent (15/21) of women reported vestibular dyspareunia with a drop in estrogen, although in some patients, this was before menopause. Fifty-two percent of the postmenopausal patients (11/21) developed dyspareunia as one of their initial symptoms at the time of menopause transition. Two had noted onset of dyspareunia when postpartum; it never resolved and they presented during menopause. Two had developed premenopausal dyspareunia and noted worsening dyspareunia with menopause. Six patients (29%) had symptoms that seemed independent in timing from the loss of estrogen. At the time of surgery, 17 of 21 of these patients were using systemic hormone therapy, oral or transdermal, and one was using local estrogen, totaling 86% (18/21) on supplemental estrogen. Ten women (48%) were using systemic progestins for endometrial protection. Three patients were not using any hormones as a result of breast cancer or personal choice (14%).
Histologic features and mast cell counts among groups are presented in Table 2, Figure 1, and Figure 2. Multiple histologic features were different in the postmenopausal patients compared with our previously described premenopausal series.10 Compared with premenopausal primary and secondary biopsies, tissues from postmenopausal patients showed significantly more chronic inflammation (Table 2) (unadjusted OR 9.0, 95% CI 2.8–33.3; adjusted OR for parity and duration of symptoms 9.1, 95% CI 2.6–31.9; unadjusted OR 6.2, 95% CI 1.9–20.0; adjusted OR 6.6, 95% CI 2.0–21.9, respectively) and greater mast cell infiltration (Fig. 2). Mast cell counts were significantly higher in the menopausal group than the primary premenopausal (36 compared with 28, P=.005) or secondary premenopausal (36 compared with 26, P<.001) group. Results remained the same after adjusting for parity and duration of symptoms (P=.01 and P=.03, respectively). There was less neural hyperplasia and progesterone receptor expression in postmenopausal biopsies compared with primary cases and less progesterone receptor but similar neural hyperplasia compared with premenopausal secondary vestibulodynia (Fig. 1). When the results were analyzed excluding the women not using estrogen supplementation (n=3), there was no change in the results. The similar histologic features of women using estrogen (n=18) and women not using estrogen (n=3) are graphically displayed in Figure 3.
We report the histopathologic features of postmenopausal vestibulodynia. Compared with premenopausal primary and secondary vestibular biopsies, postmenopausal tissues had more lymphocytes and mast cells (Figs. 1 and 2, respectively). There was less neural hyperplasia in postmenopausal biopsies compared with primary premenopausal vestibulodynia (Table 2). Premenopausal and postmenopausal vestibulodynia share histologic features of neurogenic inflammation but differ in degree.
Although most research in this field has focused on the premenopausal population, dyspareunia is also an important consideration in postmenopausal women. A recent literature review reports the prevalence of postmenopausal dyspareunia ranging from 2% to 29%17 and another study found that vulvodynia remains significant at 8% from adulthood until age 70 years.18 Kao et al19 reported that 95% of menopausal women with dyspareunia also had localized provoked pain in the vestibule despite the use of hormone supplements in 31% of patients. Because postmenopausal women continue to stay sexually active, dyspareunia, and specifically vestibulodynia, are of clinical and scientific importance.
Marked neurogenic inflammation is a frequently identified histologic characteristic of vestibulodynia.10,13,20–23 We have previously reported significant histologic differences in patients with premenopausal primary vestibulodynia compared with premenopausal secondary vestibulodynia.10 In this retrospective analysis of postmenopausal tissue specimens, neurogenic inflammation was again seen. However, this tissue showed more severe chronic inflammation (lymphocytes) than either premenopausal group and less neural hyperplasia than their premenopausal primary vestibulodynia counterparts. The duration of symptoms did not correlate with increased inflammation, because the group with the least inflammation was the group with the longer duration of symptoms (primary premenopausal). Estrogen receptor expression was not statistically different between all groups, perhaps as a result of varied supplemental hormone regimens used by patients in all groups.
Falling estrogen levels may trigger the development of vestibulodynia. Although the mechanism of vestibulodynia remains unclear, understanding the change in hormonal environment as it correlates to the development of these histologic patterns may help researchers understand the pathophysiology of subtypes of vestibulodynia. The majority (71%) of the patients in our menopausal cohort noted the onset or worsening of pain related to loss of estrogen, either at the time of the menopause transition or with prior postpartum estrogen loss. Twenty-two percent of the premenopausal secondary sufferers reported the advent of pain in the postpartum period. We suspect there is a relationship between hormonal signaling in the vestibule of these women and the neurogenic inflammation characteristic of vestibulodynia. Menopausal estrogen supplementation achieves only low levels of estrogen when compared with the range of physiologic circulating estradiol in premenopausal women. The low estrogen milieu was associated with the greatest lymphocyte and mast cell infiltrates of all groups, perhaps indicating that these cell types respond most strongly to low estrogen levels, stimulating a neural hyperplastic response in the vestibule. When estrogen supplements do not alleviate symptoms of dyspareunia, they may be too late or inadequate to stop the mechanism of tissue changes. The inflammatory response may continue to stimulate hyperplasia and hypertrophy of the nerves, leading to vestibulodynia.
This hypothesis is supported by published studies by Smith and colleagues,24 who reported the association of low estrogen levels and nerve proliferation in vaginas of rats and in the upper genital tracts of mice.25 At the nadir of estrogen levels in estrus cycles, nerves proliferate, and then these nerves retreat when estrogen is highest. This pattern repeats with cyclic plasticity. If the human vestibule shows a similar programming, this may be a phenomenon that explains menopausal vestibulodynia. Despite the common description of estrogen supplementation as “estrogen replacement,” it is meant to be low-dose rather than “replacement” therapy, and tissue levels in the vestibule may be not achieve a threshold that prevents or treats vestibule sensitivity. No one has published histologic analyses of pre-estrogen and postestrogen vestibule biopsies in cases of menopausal dyspareunia, especially because estrogen therapy corrects dyspareunia in many postmenopausal women. However, there may be an estrogen threshold that varies from individual to individual for the presumed nerve changes to recede. As Kao suggests, the common concept of vulvovaginal atrophy is not an adequate construct to explain the findings of vestibular pain.17,26 In 95% of her postmenopausal cohort with vestibular pain and dyspareunia, women used a variety of estrogen supplements and atrophy varied but vestibular pain persisted.19
This study describes the histologic profile of vestibulodynia in a menopausal population and includes representative patients in categories of premenopausal primary vestibulodynia, premenopausal secondary vestibulodynia, and now menopausal vestibulodynia. One limitation of this retrospective study is the ability to draw conclusions from these comparisons of small subgroups. The ability to perform an ad hoc power analysis is limited because no single histologic feature defines vestibulodynia. Another limitation is lack of uniformity of clinical histories and hormone use. Only three women were not on estrogen, but when our data were analyzed excluding this group, results did not vary. The study is descriptive and cannot prove causation.
A diagnosis of vestibulodynia should be considered in menopausal as well as premenopausal women presenting with dyspareunia. Vestibulodynia is a condition that affects women of all ages. Although supporting the menopausal patient with adequate estrogen either locally or systemically is an important clinical consideration, it may not be effective in alleviating vestibular dyspareunia. A better understanding of the relationship between hormone signaling and the pathogenesis of neurogenic inflammation is needed.
1. Moyal-Barracco M, Lynch PJ. 2003 ISSVD terminology and classification of vulvodynia: a historical perspective. J Reprod Med 2004;49:772–7.
2. Harlow BL, Stewart E. A population-based assessment of chronic unexplained vulvar pain: have we underestimated the prevalence of vulvodynia. J Am Med Womens Assoc 2003;58:82–8.
3. Goetsch MF. Vulvar vestibulitis: prevalence and historic features in a general gynecologic practice population. Am J Obstet Gynecol 1991;164:1609–14; discussion 1614–6.
4. Reed BD, Haefner HK, Harlow SD, Gorenflo DW, Sen A. Reliability and Validity of Self-reported symptoms for Predicting vulvodynia. Obstet Gynecol 2006;108:906–13.
5. Harlow BL, Wise LA, Stewart EG. Prevalence and predictors of chronic lower genital tract discomfort. Am J Obstet Gynecol 2001;185:545–50.
6. Shifren JL, Monz BU, Russo PA, Segreti A, Johannes CB. Sexual problems and distress in United States women: prevalence and correlates. Obstet Gynecol 2008;112:970–8.
7. Laumann EO, Paik A, Rosen RC. Sexual dysfunction in the United States: prevalence and predictors. JAMA 1999;281:537–44.
8. Dennerstein L, Alexander JL, Kotz K. The menopause and sexual functioning: a review of the population-based studies. Annu Rev Sex Res 2003;14:64–82.
9. Avis NE, Stellato R, Crawford S, Johannes C, Longcope C. Is there an association between menopause status and sexual functioning? Menopause 2000;7:297–309.
10. Leclair CM, Goetsch MF, Korcheva VB, Anderson R, Peters D, Morgan TK. Differences in primary compared with secondary vestibulodynia by Immunohistochemistry. Obstet Gynecol 2011;117:1307–13.
11. Baggish MS, Miklos JR. Vulvar pain syndrome: a review. Obstet Gynecol Surv 1995;50:618–27.
12. Goetsch MF. Simplified surgical revision of the vulvar vestibule for vulvar vestibulitis. Am J Obstet Gynecol 1996;174:1701–5; discussion 1705–7.
13. Goetsch MF, Morgan TK, Korcheva VB, Li H, Peters D, Leclair CM. Histologic and receptor analysis of primary and secondary vestibulodynia and controls: a prospective study. Am J Obstet Gynecol 2010;202:614.e1–8.
14. Breslow NE, Clayton DG. Approximate inference in generalized linear mixed model. J Am Stat Assoc 1993;88:9–12.
15. Whitehead J. Sample size calculations for ordered categorical data. Stat Med 1993;12:2257–71. Erratum in Stat Med 1994;13:871.
16. Hirji KF. Exact analysis of discrete data. Boca Raton (FL): Chapman & Hall; 2006.
17. Kao A, Binik YM, Kapuscinski A, Khalife S. Dyspareunia in postmenopausal women: a critical review. Pain Res Manag 2008;13:243–54.
18. Reed BD, Harlow SD, Sen A, Legocki LJ, Edwards RM, Arato N, et al.. Prevalence and demographic characteristics of vulvodynia in a population-based sample. Am J Obstet Gynecol 2012;206:170.e1–9.
19. Kao A, Binik YM, Amsel R, Funaro D, Deroux N, Khalife S. Biopsychosocial predictors of postmenopausal dyspareunia: the role of steroid hormones, vulvovaginal atrophy, cognitive-emotional factors, and dyadic adjustment. J Sex Med 2012;9:2066–76.
20. Tympanidis P, Terenghi G, Dowd P. Increased innervation of the vulval vestibule in patients with vulvodynia. Br J Dermatol 2003;148:1021–7.
21. Weström L, Willen R. Vestibular nerve fiber proliferation in vulvar vestibulitis syndrome. Obstet Gynecol 1998;91:572–6.
22. Bohm-Starke N, Hilleges M, Falconer C, Rylander E. Increased intraepithelial innervation in women with vulvar vestibulitis syndrome. Gynecol Obstet Invest 1998;46:256–60.
23. Pyka RE, Wilkinson E, Friedrich E Jr, Croker B. The histopathology of vulvar vestibulitis syndrome. Int J Gynecol Pathol 1988;7:249–57.
24. Ting AY, Blacklock AD, Smith PG. Estrogen regulates vaginal sensory and autonomic nerve density in the rat. Biol Reprod 2004;71:1397–404.
25. Zoubina EV, Smith PG. Sympathetic hyperinnervation of the uterus in the estrogen receptor alpha knock-out mouse. Neuroscience 2001;103:237–44.
© 2013 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
26. Kao A, Binik YM, Amsel R, Funaro D, Leroux N, Khalife S. Challenging atrophied oerspectives on postmenopausal dyspareunia: a systematic description and synthesis of pain characteristics. J Sex Mar Ther 2012;38:128–50.