Objective
The objective of the study was to assess the association between hormone receptor densities, pain nerves, and inflammation in vestibulodynia patients.
Study Design
In a prospective study, tender and nontender biopsies from 10 primary and 10 secondary vestibulodynia patients were compared with biopsies in 4 nontender controls. Hormone receptors were evaluated using immunohistochemistry for estrogen receptor-α and -β, androgen, and progesterone receptors. Inflammation, nerves, and mast cells were assessed histologically. Statistical analysis was by Fisher’s exact test, analysis of variance, paired Student t test, and Wilcoxon rank test.
Results
Tender sites from primary vestibulodynia had increased nerve density compared with secondary and control biopsies ( P = .01). Tender sites in secondary vestibulodynia had more lymphocytes than tender primary sites and control biopsies ( P < .0001). Mast cells were increased in tender sites compared with nontender and controls. There were no differences in hormone receptor expression.
Conclusion
Markers of inflammation differed between primary and secondary vestibulodynia and controls.
Vestibulodynia (V) is increasingly recognized as a relatively common cause of localized vulvar pain and severe sexual difficulty. Its cause remains unclear. Two of the primary theories under investigation include: (1) defective regulation of the inflammatory response in the vestibule and (2) hormonal triggers such as low estrogen or progestin effects of contraceptive steroids.
Both theories aim to correlate the presence of a higher density of epithelial nerves in V. Histologic examinations have revealed that the affected mucosa in V has more large and fine nerves in the stroma and epithelium compared with nontender subjects. Researchers hypothesize that dysregulation of inflammation is involved, allowing a local irritant to generate a self-sustaining abundance of nerves but not a classic inflammatory reaction.
Bohm-Starke et al provided data indicating that V pain is not mediated by classic mechanisms, explaining why treatments that target the cyclooxygenase 2 or nitric oxide systems are not effective. Moreover, the literature is mixed regarding the presence or absence of increased lymphocytic infiltrate in V. Some investigators have found no difference in nontender and tender tissues, whereas others have noted increased lymphocytes in vestibulodynia.
This variation in neural hyperplasia and inflammation may be related to hormone status. Zoubina and Smith have shown that nerve density in rodent endometrium and vaginal tissues increases when estrogen levels are low in estrus cycles, and the nerve density lessens when estrogen levels are highest. Low estrogen states are associated with introital pain in women, both during the postpartum time and in menopause.
Studies on vestibulodynia have focused primarily on premenopausal women. Eva et al reported spotty and reduced presence of estrogen receptor (ER)-α in vestibulodynia compared with nontender controls, offering the explanation that circulating estrogen is ineffective in which the tissues have fewer ERα receptors. Others have not noted these findings. Johannesson et al published data on an array of sex steroid receptors and found that V tissues had increased expression of ERα but not ERβ, androgen receptor (AR), or progesterone receptor (PR).
Few investigators have reported their findings in relation to patient history (ie, primary vs secondary onset of symptoms). Primary V sufferers have noted pain from the first introital touch, whether with attempts at tampon insertion or with sexual debut. Those with secondary V have developed this condition after first having no history of dyspareunia.
Our prospective study is the first to perform a comprehensive histologic and sex steroid receptor analysis comparing primary and secondary V to controls. Studying vestibulodynia with a focus on patients’ histories could elucidate whether there are different pathophysiologic pathways, helping to explain the mixed results reported in the literature.
Our study had 3 objectives: (1) assessment of steroid receptor expression (ERα, ERβ, AR, PR) and nerve density in V; (2) assessment of inflammation in women with V (lymphocytes and mast cells); and (3) comparison of subjects with primary and secondary V. A secondary objective was to compare 2 different stains for nerves, S-100 and PGP9.5, which may have different clinical utility.
Materials and Methods
Subjects
We prospectively recruited premenopausal women from the Program in Vulvar Health or the generalist practice at Oregon Health and Science University using an institutional review board–approved protocol. Inclusion criteria were the following: Subjects were older than 18 years and premenopausal; had severe entry dyspareunia for more than 1 year; and had extreme tenderness with light touch to the vestibule, reversible temporarily with application of aqueous 4% lidocaine to confirm that their pain was localized and superficial.
Subjects agreed to sampling of tender as well as nontender vestibule mucosa, thereby allowing each subject to serve as her own control. Subjects therefore had nontender touch test findings at areas of the vestibule anteriolateral to the level of the urethra.
Women designated as having primary V had noted significant pain from the first attempted vaginal entry, either with tampons or coitus. Secondary sufferers had the onset of pain after having initially been free of localized entry pain for months or years.
General exclusion criteria included pregnancy, menopause, chronic pelvic pain, local vulvar or vaginal infection, dermatoses, scar or neuroma formation, and generalized vulvodynia, based on screening examinations. The control group had no entry dyspareunia.
Subjects and controls were not offered compensation for their participation. All enrollees were planning surgery, either for correction of vestibulodynia or, in the case of controls, for benign gynecologic diagnoses for which a vulvar surgical approach was planned. This strategy was chosen so that biopsies would not cause additional pain during recovery from the procedures. Biopsies were scheduled during the follicular menstrual phase of ovulatory patients. Users of contraceptive hormones were off the active pills for at least 5 days but fewer than 15 days. To assure that patients had not ovulated, all had measurement of serum progesterone, which had to be 3 ng/mL or less on the date of tissue collection. Two surgeons (M.F.G., C.M.L.) recruited the subjects, and the procedure was the superficial localized posterior vestibulectomy described by Goetsch.
Serum studies and tissue biopsies
Serum samples were drawn for 17β-estradiol and progesterone levels on the day of surgery. We used progesterone greater than 3 ng/mL as the definition of ovulation. Biopsies were performed once the women were sedated for surgery. A 6 mm diameter sample of mucosa was removed immediately adjacent to a Bartholin’s duct orifice in all patients, controls, and cases. An additional 3 mm diameter biopsy was removed in V subjects but not controls. It was taken from the nontender right or left vestibule anterolateral to the urethra, an area located within Hart’s line. Tissues were immediately fixed in formalin.
Histology and immunohistochemistry
Formalin fixed biopsies were paraffin embedded, and serial histologic sections (3 μm thick) were cut for hematoxylin and eosin (H&E) staining or immunohistochemical analysis. Immunohistochemistry was performed by the Oregon Health and Science University Pathology Research Core Laboratory supervised by a board-certified pathologist (T.K.M.). Cases and controls were immunostained in duplicate (n = 2 for each) using the following primary antibodies: S100 (DAKO, Carpinteria, CA); PGP 9.5 (DAKO), CD117 (DAKO); AR (DAKO); PR (DAKO); ERα (DAKO), and ERβ (DAKO). Positive antibody labeling was visualized using the Envision system (DAKO) according to the manufacturer’s instructions.
Sections were counterstained with hematoxylin and specificity confirmed using commonly used positive controls (eg, breast for ER and PR; GIST for CD117; prostate for AR, nerve for S100 and PGP 9.5) and appropriate negative controls (isotype-negative control antibody). Staining quality and specificity was confirmed by a pathologist (T.K.M.).
Semiquantitative analysis
The entire study set was scored in duplicate using light microscopy while blinded to phenotype (vestibulodynia [primary vs secondary] vs controls) by an expert gynecologic pathologist (T.K.M.) and an independent pathology resident (V.B.K.). Inflammation was scored in H&E-stained sections as negative, mild, or moderate.
The number of infiltrating mast cells was analyzed for specificity by scoring CD117-immunostained sections using the average count per ×20 objective field counted over the 3 most populated fields. Nerve density was scored by using S100-stained sections and reported as either positive or negative for relative neural hyperplasia. Sections stained for PGP9.5 revealed small nerve twigs in the epidermis of some cases but not others and were therefore scored as either positive or negative. Hormone markers were scored similar to routine breast pathology using a 0-3 scale with a score of 0, negative; 1, less than 10%; 2, less than 50%; and 3, less than 100% of basal cells.
Statistical analysis
Statistical analyses were performed with SAS statistical software (SAS Institute Inc, Cary, NC). A power analysis based on the findings of Eva et al on ERα indicated that we needed 20 women with vestibulodynia and 6 control women. Patients’ characteristics, the densities of mast cells, and immunochemical staining results among control, primary V, and secondary V were compared using analysis of variance for normally distributed continuous variables or Kruskal-Wallis tests for nonnormally distributed continuous variables and Fisher’s exact test for categorical variables.
Continuous variables were described by calculating a mean and noncontinuous variables by median. The differences between tender and nontender sites in vestibulodynia cases in mast cell counts and the results of immunochemical staining were compared using paired Student t tests or Wilcoxon signed rank tests. A kappa statistic was calculated to express reproducibility between pathologists and a kappa statistic greater than 0.80 was defined as excellent. All reported P values are 2 sided, and a value of P < .05 was considered statistically significant.
Results
Twenty-four women met criteria for inclusion. Of others enrolled, 1 control and 1 subject were excluded when serum testing revealed they had ovulated. Another control was excluded because of her age older than 50 years and low estrogen levels, even though she was still bleeding cyclically. Another case was excluded because on the day of surgery, she had no completely nontender vestibule sites for the second nontender biopsy. During processing, 3 nontender site biopsies in cases, 2 from secondary and 1 from a primary case were maloriented and could not provide proper sections. One nontender case biopsy did not have enough tissue for all stains to be done.
Ten women with primary vestibulodynia, 10 with secondary vestibulodynia, and 4 control women were analyzed in this study.
Demographics
Patient characteristics are listed in Table 1 . The mean age was greater in the control group but similar in the primary and secondary groups. The mean duration of dyspareunia was somewhat longer in the primary group.
| Characteristic | Controls(n = 4) | Primary V(n = 10) | Secondary V(n = 10) | P value |
|---|---|---|---|---|
| Age, y; mean ± SD (range) | 40.3 ± 3.7 (36–45) | 28.4 ± 7.1 (20–46) | 30.8 ± 7.5 (21–40) | |
| Symptom duration, y; median (range) | n/a | 6.5 (2–25) | 4.5 (1–20) | .13 |
| Follicular day, median (range) | 9.5 (5–14) | 5.5 (5–13) | 7.0 (5–9) | NS |
| Estrogen level, median (range) | 138.5 (65–253) | 55 (19–192) | 55.5 (19–237) | .13 |
| Nullipara, n (%) | 3 (75) | 7 (70) | 7 (70) | .99 |
| Combined contraceptives, n (%) | 1 (25) | 7 (70) | 7 (70) | |
| Surgical result, n (%) | N/A | Cured 9 (90) | Cured 7 (70) | |
| Improved 1 (10) | Improved 3 (30) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
