Elevated tissue levels of interleukin-1β and tumor necrosis factor-α in vulvar vestibulitis

Elevated Tissue Levels of Interleukin-l/3 and Tumor Necrosis Factor-a in Vulvar Vestibulitis DAVID

C. FOSTER, MD, MPH, AND JEFFREY D. HASDAY,

Objective: To compare levels of two inflammatory cytokines, interleukin-16 (IL-lp) and tumor necrosis factor-a (TNF-al, in selected regions of the vulva, vestibule, and vagina in women with vulvar vestibulitis and in asymptomatic controls. Methods: Selective samplings of surgical specimens from 12 women undergoing perineoplasty for vulvar vestibulitis and ten pain-free subjects undergoing posterior vaginal repair were prepared into tissue homogenates and analyzed for concentrations of IL-lp and TNF-a. Interleukin-l/I and TNF-a concentrations were measured by sandwich enzymelinked immunosorbent assay. The results were reported after adjustment for total tissue protein concentration. Results: Median tissue levels of IL-l/3 and TNF-a were elevated 2.3-fold and 1.8-fold, respectively, in women with vulvar vestibulitis relative to pain-free women. Median IL-lfi tissue levels were increased significantly from 1.3 pg/mg to 3.0 pg/mg total protein in women with vulvar vestibulitis compared to pain-free women. Median TNF-a tissue levels were increased from 83 pg/mg to 148 pg/mg total protein in women with vulvar vestibulitis compared to pain-free women. Analysis by selected anatomic site of women with vulvar vestibulitis revealed a significant 2.2fold higher median level of TNFa at the vulvar site compared to the vestibule. Conclusion: Concentrations of IL-10 and TNF-a were elevated in women with vulvar vestibulitis relative to those in asymptomatic controls. This elevation in inflammatory cytokines with vulvar vestibulitis varied according to anatomic site and was, paradoxically, lowest in the area of highest hyperalgesia, the vulvar vestibule. Inflammatory cytokine elevation may contribute to the pathophysiology of mucocutaneous hyperalgesia. (Obstet GynecoI 1997;89:291-6. Copyright 0 2997 by The American College of Obstetricians and Gynecologists.)

From the Department of Obstetrics and Gynecology, The \ohns Hopkir~ Utliversity School qf Medicine, Baltimore; ad Cytokine Core Laboratory, Utkxrsity qf Maryland School of Medicine, Baltimore, Maryland. We gratefully ackrrozchfge the help ofi3ezlerly Smith, BS, uho assisted with chemical arld equipmet~f procrrremcrzt fiir this study, and Robert Kuman, MD, &~ reviewed histopathology:

VOL. 89, N0.2,

FEBRUARY

1997

MD

syndrome is a condition characterized by dyspareunia or pain with insertion of a tampon, associated with one or more painful foci of the vulvar vestibule and no identifiable source of pain such as herpes, candidiasis, or pemphigoid. Peckham et al’ studied 67 women with vulvar vestibulitis, which they termed “focal vulvitis.” Because vestibular inflammation was “rarely seen with the naked eye,“ the authors stated that clinical demonstration of vestibular hypersensitivity and allodynia were “essential” for diagnosis. Exquisite sensitivity to touch was characteristically focal, reproduced with remarkable precision by independent observers, and separated from nontender areas of the vulva and vagina by millimeters. In more than 80% of patients, painful foci were located on the mucosa of the vestibule between 3 and 9 o’clock, inclusive, and limited by line of keratinization (Hart’s line) and the hymeneal ring. The remaining 20% demonstrated focal areas of tenderness within the defined limits of the vestibule but extending toward the urethral orifice. The exact prevalence of vulvar vestibulitis is unknown, and the etiology, which has been considered multifactorial, remains elusive. Proposed causes of pain include sexual trauma, psychiatric illness, human papillomavirus (HPV) infection, postherpetic neuralgia, urinary metabolites, hormonal deficiency, obstructed major vestibular glands, inflamed minor vestibular glands, and neurogenie pain2-” The regulation of inflammatory response at the skin and mucosal level is a function of an interaction of inflammatory cells, cytokines, and neurokines and may be of significance in vulvar vestibulitis. Two cytokines, interleukin-1 (IL-l) and tumor necrosis factor (TNF), directly promote acute and chronic inflammatory conditions and interact with histamine, bradykinin, and substance P in the development of neurogenic inflammation. Two subtypes of IL-l, (Yand /3, are produced by a myriad of cells including fibroblasts, keratinocytes, hepatomas, endothelial cells, astroglia cells, synovial cells, and B lymphocytes.‘” Interleukin-la! is bound to

Vulvar

vestibulitis

0029-7844/97/$17.00 PI1 SOO29-7844(96)00447-4

291

the inflammatory cell membrane, whereas IL-l/3 is found in the extracellular milieu. Recent work demonstrated the induction of cutaneous hyperalgesia by local injection of IL-l/3, which is associated with the release of substance I’ from peripheral nerve C fibers.” Produced primarily by macrophages, TNF-a also mediates the inflammatory process and appears to synergize the effects of IL-lp. Tumor necrosis factor-a provides a positive feedback loop for the production of additional IL-lp and induces release of other cytokines. The inflammatory cascade mediated by TNF-(w may persist for prolonged periods, often becoming dissociated from the inciting event of invasion, injury or infection. The individual response to TNF-CY depends upon tissue levels, duration of exposure, and the presence of synergistic cytokines such as IL-lp.” The purpose of this study was to compare levels of two inflammatory cytokines, IL-l/3 and TNF-u, in selected regions of the vulva, vestibule, and vagina in women with vulvar vestibulitis relative to levels in asymptomatic controls. These cytokines may contribute to the pathophysiologic basis of the local hyperalgesia of vulvar vestibulitis. We hypothesized that tissue concentrations of inflammatory cytokines would be elevated significantly in women with vulvar vestibulitis, compared to pain-free controls, and would differ within the group of vulvar vestibulitis patients as a function of the anatomic site of sampling within the lower genital tract.

Methods We acquired surgical specimens from two groups: women undergoing perineoplasty for vulvar vestibulitis and women undergoing rectocele repair for nonpain-related indications. Women with vulvar vestibulitis met the following criteria: 1) described at least 1 year‘s duration of vulvar burning with insertional dyspareunia or pain with tampon insertion; 2) physical examination demonstrated exquisite tenderness to light touch (allodynia), limited to the vulvar vestibule between Hart’s line and the hymeneal tissue and independently confirmed by a second examiner (pain response to cotton-tip applicator was recorded using a visual analog scale of 1 to 10); 3) did not demonstrate any other specific neuropathology; 4) preoperative laboratory testing did not identify atrophic vaginitis, dermatitis, or pathogens such as fungus or herpes; and 5) a 3to 6-month trial of conservative medical therapy failed to correct the painful condition. All perineoplasty procedures were performed by one of the authors (DCF) and followed the operative technique as described by Woodruff and Parmley.’ The controls were consecutive women undergoing

292

Foster and Hasday

qtokiffes

if7 Vulzv7u Vestihlr~fhs

Figure 1. Selected tissue samples from the lower genital tract, which were analyzed for interleukin-lp and tumor necrosis factor-a.

vaginal surgery over a 6-month period who 1) provided no recent history of vulvar pain, dyspareunia, or pruritus; 2) had a normal neurologic exam (with particular attention paid to the vaginal and perineal area); and 3) required a perineal procedure unrelated to correcting pain or dyspareunia. The pain-free control group with symptomatic rectocele underwent a posterior vaginal repair using standard techniques. Photographs of the vulvar vestibule, the surgical margin highlighted with skin marker, were routinely taken of each subject undergoing perineoplasty. Standard informed consent for operative procedures and photo documentation was acquired from all subjects. All surgical specimens were submitted for hematoxylin and eosin processing. A staff pathologist, blinded to the results of the study, quantified the inflammation (minimal, mild, moderate, or severe), located the regions of greatest inflammation, and reported the predominant inflammatory cell type. Within 20 minutes after resection of perineal tissue, the following three samplings were submitted from regions in the perineal area (Figure 1): a vulvar portion between the surgical margin and the edge of keratinization (Hart’s line), a vestibular portion midway between Hart’s line and the carunculae hymenales, and a hymeneal portion from the junction of hymen and vagina. Specimens were collected from identical areas of vulva, vestibule, and hymen in cases and controls, and were weighed and cooled to -70C for later processing into tissue homogenates for sandwich enzymelinked immunosorbent assays (ELISAs) of IL-l/3 and

TNF-a and standard assay for total protein content. Specimens were thawed, placed in 1.5-mL microcentrifuge tubes with 0.8 mL buffered saline (pH 7.2-7.4) containing 1 mmol/ L phenylmethylsulfonyl fluoride, 4 mmol/ L ethylenediaminetetra-acetic acid (EDTA) and 1 pg/mL leupeptin. Tissues were sonicated four times using a microprobe with an ultrasonicator at 15-second bursts. Triton Xl00 (8 FL) (Sigma, St. Louis, MO) was added; the tube was vortexed vigorously for 2 minutes and centrifuged at 4C for 30 minutes. Supernatants were collected and stored at -70C until analyzed by ELISA. The methods for homogenization and ELISA are part of a larger study in the process of completion. Preliminary assays for validation of technique demonstrate good recovery of a recombinant cytokine spike after homogenization. Interleukin-l/3 and TNF-cu were measured by sandwich ELISA using commercial recombinant standard and monoclonal antibody pairs. Polystyrene plates were coated with capture antibody (mAb-M-420B, for IL-l; Endogen, Cambridge, MA, and mAb-5817508 for TNF; Medgenix, Fleurus, Belgium) in phosphobuffered saline overnight at 25C. The plates were washed four times with 50 mmol/L Tris 0.2% Tween-20 (Sigma, St. Louis, MO), pH 7.0-7.5, and then blocked for 90 minutes at 25C in a 1:l mixture of 2.5% casein and buffered saline containing 4% bovine serum albumin and 0.01% Thimerosal (Sigma, St. Louis, MO), pH 7.2-7.4, assay buffer. The plates were washed four times and 50 PL of assay buffer was added to each well along with 50 PL of sample of recombinant standard (RIL-lbetad for IL-l@; Endogen, Cambridge, MA, and Medgenix 5817510 for TNF-a; Medgenix, Fleurus, Belgium) prepared in assay buffer and incubated at 37C for 2 hours. The plates were washed four times and 100 PL of biotinylated detecting antibody (M-421B for IL-lp; Endogen, Cambridge, MA, and Medgenix 5817503 for TNF-(r; Medgenix, Fleurus, Belgium) in assay buffer was added and incubated for 1 hour at 25C. After the plate was washed four times, strepavidin-peroxidase in assay buffer was added and incubated at 25C for 30 minutes. The plate was washed four times and 100 PL of substrate, commercially prepared (TMB; Dako, Carpintaria, CA) was added and incubated at 25C for approximately 30 minutes. The reaction was stopped with 100 FL of 2N HCl, and A450 was read on a microplate reader (Dynatech, Chantilly, VA). A curve was fit to the standards using the software Deltagraph for Macintosh (Deltapoint, Monterey, CA) and cytokine concentration in each sample was calculated from the standard curve equation. The ELISA of tissue homogenates for IL-1B detected both precursor and mature forms of IL-l/3 and had a sensitivity of 3.125 pg/mL and a coefficient of variation of 17.05%. The ELISA for TNF-cr measured only free (not bound to

VOL.

89, NO.

2, FEBRUARY

1997

Table

1.

Comparison of Casesand Controls

Patient

variable

Age (y) (median) Caucasian (%) Parity (median) IL-l/3 [median (quartiles)] pg / mg tissue protein Hymen Vestibule Vulva TNF-a [median (quartiles)] pg/mg tissue protein Hymen Vestibule Vulva IL-lp = interleukin-16; significant. * Wilcoxon.

Asymptomatic controls (n = 10)

Vulvar vestibulitis (n = 12)

Statistical analysis* P

33 12 (100) 1 3.0 (1.1-6.5)

3 1.3 (0.9-3.5)

.02

3.4 2.2 4.8 148

1.3 0.9 1.6 83

(0.9-2.8) (0.7-2.1) (0.9-2.8) (41-187)

NS NS NS .07

65 (18-120) 50 (36-92) 106 (54-208)

NS NS NS

(1-8.9) (0.8-3.3) (1.2-6.9) (41-280)

168 (47-268) 80 (41-168) 172 (71-314) TNF-(U = tumor

48 6

(60)

necrosis

factor-e;

NS = not

receptor) TNF-a! and had a sensitivity of 31.25 pg/mL and a coefficient of variation of 16.63%. Total protein was measured using commercial reagents and was calculated by interpolating from a standard curve constructed using bovine serum albumin. The total protein assay had a sensitivity of 50 kg/mL. Tissue homogenate concentrations of IL-lp and TNF-a were reported as picograms per milligram of total protein to adjust for variations in mass of tissue sampled and amount of tissue proteins extracted during processing. Adjusted values were grouped by case and control and by anatomic site of sampling (vulva, vestibule, or hymen). Data were presented as median concentration (2 lower and upper quartiles). Medians, lower and upper quartiles, skewness, kurtosis, Wilcoxon rank-sum test for two samples, and Fisher exact test were calculated using STATA software (Stata Corp, College Station, TX). Statistical significance was established at the P = .05 level for a two-tailed distribution. Power analysis to estimate adequate subject numbers could not be calculated because of a lack of preexisting reports of vulvar tissue levels of IL-lp and TNF-(I.

Results Characteristics of cases and controls are given in Table 1. In women with vulvar vestibulitis the range of duration of vulvar pain was 2-9 years. On physical examination, all women with vulvar vestibulitis had hyperalgesia in response to cotton-tip applicator, graded more than 5 of 10 on the visual analogue scale; the median was 8 of 10, which greatly diminished, vaginally, above the level of the carunculae hymenales (0 of 10 on the visual analogue scale) and external to

Foster and Hasday

Cytokines

in Vulmr

Vesfibditis

293

Hart’s line (1 of 10 on the visual analogue scale). All subjects with vulvar vestibulitis had cultures negative for mycoplasma, Ureaplasma, herpes, and fungus. All had maturation indices without parabasal cells, indicating adequate estrogen milieu. For the purpose of identifying potential confounding of cytokine levels by age, race, or parity, multivariate linear regression models were created with IL-lfi or TNF+ as dependent variables. Age, race, and parity did not predict significantly variation of cytokine levels. All controls underwent a posterior vaginal repair for symptomatic rectocele, and none complained of painful intercourse nor had evidence of hyperalgesia in response to cotton-tip applicator on physical examination (0 of 10 on the visual analogue scale). All postmenopausal vulvar vestibulitis subjects and asymptomatic controls were on estrogen replacement therapy before the surgical procedure. The median levels of IL-lp and TNF-a were 2.3-fold and l&fold higher, respectively, in the women with vulvar vestibulitis (Table 1). Interleukin-l/3 tissue levels were significantly higher in vulvar vestibulitis patients (Wilcoxon rank sum; P = .02), and TNF-a tissue levels approached statistical significance in vulvar vestibulitis patients (Wilcoxon rank sum; P = .07). Medians and Wilcoxon rank-sum testing for significance were reported because skewness / kurtosis tests for normality demonstrated significant skewness. Tissue levels of IL-l/3 to TNF-cu, sampled from the same subject and anatomic site, correlated with high statistical significance (r = .75; P < .OOl). Median tissue homogenate levels of IL-lp and TNF-a were grouped according to sampling site within vestibulitis subjects and asymptomatic controls. There was a 2.2-fold higher concentration of TNF-a at the external vulva relative to the vestibule, a statistically significant difference (Wilcoxon rank sum; P = .05). Interleukin-l/3 levels were also 2.2-fold higher at the external vulva relative to the level at the vestibule, but this difference was not statistically significant. Lymphocytic infiltration of both B- and T-cell types were found in tissue samples; however, the concordance between highest cytokine concen+ration and greatest lymphocytic infiltration was only six of 12. Tissue levels of IL-16 and TNF-a were grouped by one of two initial complaints: 1) “insertional pain with intercourse or tampon use” or 2) “vulvar burning.” Seventy-five percent (9/12) of women with vulvar vestibulitis reported initial symptoms that corresponded to the region of the lower genital tract with highest cytokine levels. The highest levels of IL-lb and TNF-cu were found at the hymen in six of 9 (67%) subjects who complained of “insertional pain with intercourse or tampon use.” The highest levels of IL-l/3 and TNF-a were found at the external vulvar site in

294

Foster and Hasday

Cyfakines

iu Vuhar

Vrstibulitis

three of 3 (100%) women who complained of “vulvar burning.” Grouping of cytokine level by initial complaint, however, did not achieve statistical significance (Fisher exact; P = .22).

Discussion The final common neural pathway for the cytokineneurokine-chronic pain state seems to involve the unmyelinated nociceptor, the C fiber. Tissue inflammation via chemical intermediates such as interleukin-lp, tumor necrosis factor, potassium, serotonin, bradykinin, histamine, prostaglandin (PG), and substance I’ sensitize C fibers. A-a fibers and C fibers, which are normally subresponsive to mechanical stimulation, revert to mechanically “sensitive” afferents when an “inflammatory soup” of bradykinin, PG, or histamine is injected in proximity to the nerve fiber.r3 Another study” demonstrated the induction of cutaneous hyperalgesia and neural responsiveness to thermal and mechanical stimuli by peripheral injection of IL-l/X Tissues of urogenital sinus origin, embryologically identical to the vulvar vestibule, have been the subject of studies on the association of substance P and neurogenie inflammation. Mechanical irritation by a transurethral catheter can produce a neurogenic inflammatory response through the local release of vasoactive neuropeptides, substance I’, and calcitonin gene-related peptide, all mediated via afferent C fibers.” In another studyI on the lower urinary tract, a significantly higher number of substance P-positive nerve fibers were demonstrated in bladder submucosa of patients with interstitial cystitis compared to normal controls. These substance P-positive nerve fibers in the submucosa of the bladder were found to be associated closely with mast cells. Earlier researchI suggests that the tissues of urogenital sinus origin are innervated by a preponderance of free nerve endings, in contrast to the differentiated afferent nerves found in the skin of the external vulva. Recently, combined immunocytochemical and electron microscopic studies’ of the vulvar vestibule have shown neural fibers containing substance I’ associated closely with neuroendocrine cells containing serotonin and chromogranin. It is premature to state that the cytokine-neurokine-C fiber connection is, definitively, the mechanism for chronic pain states, such as vulvar vestibulitis, although the present data are compatible with this premise. Significant elevation of inflammatory cytokines may be an anticipated finding in the chronic “inflammatory” condition of vulvar vestibulitis. However, several unanticipated findings need further discussion. Given the well-localized hyperalgesia of the vestibule, the significantly higher level of TNF-a at the external vulva

Obstetrics b Gynecology

compared to the vulvar vestibule is a paradox. The low tissue concentrations of IL-l/3 and TNF-a at the vestibule suggest that cytokines are not likely to be the “final common pathway” to hyperalgesia. Instead, cytokines may increase, regionally, in response to trauma or inflammation, and act upon the unique neural organization of the vulvar vestibule secondarily through the release of neurokinins such as substance P.9 A second unanticipated finding is the 50% concordance of cytokine elevation and inflammatory cell infiltrate. Others have reported” that elevated TNF-(Y levels may persist after clearance of an inflammatory process. Furthermore, increased IL-1p tissue levels may be produced by noninflammatory cells such as fibroblasts, keratinocytes, and endothelial cells.‘” Of course, the unanticipated findings could be based upon artifactual differences in cytokine levels at the three sampling sites as well. However, the highly significant correlation between the two independent assays for IL-lp and TNF-o, performed at each of the sampling sites, supports our cytokine measurement. A number of issues in design of this report are open to criticism. As noted in Table 1, the control group subjects were not matched to the vulvar vestibulitis subjects with respect to age, race, and parity. Before the study was begun, two sources for asymptomatic “control” tissue were identified: women undergoing posterior vaginal repair for symptomatic rectocele and obstetric patients undergoing episiotomy repair. Although the obstetric patients were comparable in age to vulvar vestibulitis patients, the high levels of uromodulin in pregnant women may suppress IL-l/3 action, thereby critically confounding the study.17 On the other hand, women undergoing posterior vaginal repair were generally older than vulvar vestibulitis patients, and a number of studies’s-21 have reported increased levels of IL-lb in monocytes, fibroblasts, and urine (presumably from renal mesangial cells) in older human subjects. Because we hypothesized that the asymptomatic “control” group would have lower cytokine levels than the vulvar vestibulitis group, we selected the older “control” group rather than the obstetric patients. Potential confounding by this older, albeit estrogenized, “control” group would work against our hypothesis and would be therefore less prone to criticism. One study22 reported that Caucasian race and nulliparous status are significantly more common in vulvar vestibulitis cases than in controls. To date, neither a difference in inflammatory responsiveness nor any other physiologic basis has been identified for such increased risk of vulvar vestibulitis in Caucasian or nulliparous subjects. In our study, linear regression models failed to identify a significant effect of age, race, or parity on cytokine variability. In addition to shortcomings with respect to

VOL.

89, NO. 2, FEBRUARY

1997

selection of a control group, a second weakness of our study involved the lack of control of preoperative medical treatments. All vulvar vestibulitis subjects underwent a course of oral desipramine and topical lidoCaine before surgery. Both of these drugs have an effect on level of inflammation. Recently, two studies23,24have reported an anti-inflammatory effect of topical lidocaine in ulcerative colitis in humans and colitis induced experimentally in a rat model. In the current study, dosage and timing of oral tricyclic medications and topical lidocaine were not controlled and in fact varied widely. Overall, the study could be improved with larger sample size, control of presurgical medications, and a more representative control group. It is our intention here not to propose a single pathogenesis of vulvar vestibulitis but to suggest that cytokines may mediate the progression of local inflammation to hyperalgesia and chronic pain. Elevation of inflammatory cytokines may result from a number of sources including invasion by microorganisms or parasites, hypersensitivity reactions, invasion by neoplasm, or tissue damage by chemical or physical trauma. Our findings of elevated inflammatory cytokines in vulvar vestibulitis have several implications. This report confirms that vulvar vestibulitis is not purely a psychogenic disorder. Second, the finding of elevated inflammatory cytokines may promote a therapeutic approach to vulvar vestibulitis, one that minimizes additional iatrogenic physical and chemical trauma. Finally, these findings may suggest new therapeutic modes for vulvar vestibulitis such as inhibition of cytokine synthesis, binding of soluble cytokines, or antagonism of the cytokine receptor.25

References 1. Peckham BM, Maki DG, Patterson JJ, Hafez GR. Focal vulvitis: A characteristic syndrome and cause of dyspareunia. Am J Obstet Gynecol 1986;154:855-64. 2. McKay M. Vulvodynia: A multifactorial problem. Arch Dermatol 1989;125:256-62. 3. Lynch PJ. Vulvodynia: A syndrome of unexplained vulvar pain, psychologic disability and sexual dysfunction. J Reprod Med 1986;31:773-80. 4. Stewart DE, Whelan Cl, Fong IW, Tessler KM. Psychosocial aspects of chronic, clinically unconfirmed vulvovaginitis. Obstet Gynecol 1990;76:852-6. 5. Byrne MA, Walker MM, Leonard JL, Pryce D, Taylor-Robinson D. Recognizing covert disease in women with chronic vulva1 symptoms attending an STD clinic: Value of detailed examination including colposcopy. Genitourin Med 1989;65:46-9. M, Bolinsky D. Pain during sex 6. Sarrel PM, Steege JF, Maltzer response due to occlusion of the Bartholin’s gland duct. Obstet Gynecol 1983;62:261-4. 7. Woodruff JD, l’armley TH. Infection of the minor vestibular gland. Obstet

Gynecol

1983;62:609-12.

Foster and Hasday

Cytokines

in Vulvar

Vestibulitis

295

8. McKay M. Dysesthetic amitriptyline. J Reprod 9. Warner TF, Tomic S, complexes in the minor

“essential” vulvodynia, treatment with Med 1993;38:9-13. Chang CK. Neuroendocrine cell-axonal vrstibular gland. J Reprod Mcd 1996;41:

397-402.

10. Kimball ES. Involvement of cytokiner in neurogenic inflammation. In: Kimball ES, ed. Cytokines and intlammation. Boca Raton, Florida: CRC Press Inc.; 1991:169989. 11. Fukuoka H, Kawatani M, Hisamitsu T, Tnkeshige C. Cutaneous hyperalgesia induced by peripheral injection of intrrleukin-l/3 in the rat. Brain Res 1994;657:133-40. 12. Tracey KJ. In: Thompson A, ed. The cytokine handbook. San Diego: Academic Press Inc., 1994:289-304. 13. Meyer RA, Davis KD, Cohen RH, Treede RD, Campbell JN. Mechanically insensitive afferents (MIAs) in cutaneous nerves of the monkey. Brain Res 1997;561:252-61. 1-f. Abelli L, Conte B, Somma V, Parlani M, Geppetti P, Maggi CA. Mechanical irritation induces neurogenic inflammation in the rat urethra. J Ural 1991;146:1624-6, 15. Pang X, Marchand J, Sant GR, Krram RM, Theoharides TC. Increased number of substance P posittve nerve fibers in interstitial cystitis. Br J Ural 1995;75:744-50. 16. Krantz KE. The anatomy and physiology of the vulva and vagina and the anatomy of the urethra and bladder. In: Philipp EE, Barnes J, Newton M, eds. Scientific foundations of obstetrics and gynecology. Chicago: Year Book Pub., 1977. 17. Muchmore AV, Decker JM. Uromodulin: A unique 85kilodalton immunosuppressive glycoprotein isolated from urine of pregnant women. Science 1985;229:479-81. 18. Fagiolo ti. Cossarizza A, Scala E, Fanales-Belasio E, Ortolani C, Coazi E, et al. Increased cytokine production in mononuclear cells of healthy elderly people. Em J Immunol 1993;23:237558. 19. Liao Z, Tu JH. Small CB, Schnipper SM. Rosenstreich DL. Increased urine interleukin-1 levels in aging. Gerontology 1993;39: 79-27.

296

Foster and Hasday

Cyfokirws in Vzdmr Vestibzflitis

20. Kumar S, Vinci JM, Millis AJ, Saglioni C. Expression of interleukin-l alpha and beta in early passage fibroblasts from aging individuals. Exp Gerontol 1993;28:505-13. 21. Riancho JA, Zarrabeitia MT, Amado JA, Olmos JM. Age-related difference in cytokine secretion. Gerontology 1994;40:8-12. 22. Foster DC, Woodruff JD. Case-control study of vulvar vestibulitis syndrome. J Womens Health 1995;6:677-80. 23. Bjorck S, Dahlstrom A, Johansson L, Ahlman H. Treatment of the mucosa with local anaesthetics in ulcerative colitis. Agents Actions 1992;ChO-72. 24. McCafferty D, Sharkey KA, Wallace JL. Beneficial effects of local or systemic lidocaine in experimental colitis. Am J Physiol 1994; G560-7. 25. Geoffrey

R, Pugh-Humphreys P, Thomson AW. Cytokines and their receptors as potential therapeutic targets. In: Thompson A, ed. The cytokine handbook. San Diego: Academic Press Inc., 1994:52S66.

Address reprint requests to: David C. Fosfer, MD, MPH Department oj Obstetrics and Gynecology Utziuersify of Rochester Medico1 Center 601 Elnmood Averzrte, Bos 668 Rochester, NY 14642

Copyright 0 1997 by The American College of Obstetricians Gynecologists. Published by Elsevier Science Inc.

and

Obstetrics b Gyrwcology