Chronic pelvic pain, defined as pain lasting at least 6 months in the pelvis, is a common disorder affecting 24–147/1000 women.1,2 In a review of studies using laparoscopy to evaluate women with chronic pelvic pain, 36% had normal findings, suggesting other possible sources for their pain.3
Interstitial cystitis is a syndrome characterized by urinary urgency, frequency, nocturia, and pain in the pelvis. The prevalence of interstitial cystitis in the United States ranges from 10–67/100,000.4,5 Classically, the diagnosis of interstitial cystitis has been based upon a combination of symptoms and cystoscopic findings. The cystoscopic findings include glomerulations, Hunner's ulcers, or fibrosis.
In 1997, O'Leary et al developed the Interstitial Cystitis Symptom Index and Problem Index.6 This is a validated questionnaire designed to measure lower urinary tract symptoms (urgency, frequency, nocturia, and pain) in patients with interstitial cystitis. A positive score (≥6) on either the symptom or problem index had 90% sensitivity and 95% specificity for the diagnosis of interstitial cystitis. In the study by O'Leary et al, the questionnaire was evaluated in women already diagnosed with interstitial cystitis. The Interstitial Cystitis Symptom Index and Problem Index has not been evaluated as a screening test for interstitial cystitis.
The primary objective of our study was to evaluate the Interstitial Cystitis Symptom Index and Problem Index as a screening tool for interstitial cystitis in women with chronic pelvic pain. Our secondary objectives were to estimate the prevalence of interstitial cystitis in women with chronic pelvic pain and to identify other clinical risk factors for interstitial cystitis. The sample group studied was women with chronic pelvic pain scheduled to undergo diagnostic laparoscopy by their gynecologist at our institution.
MATERIALS AND METHODS
Between March and September 2000, a cross-sectional study that was approved by our institutional review board was conducted. Patients were recruited in the following manner. A letter describing the study was mailed to all general obstetric-gynecology attending physicians at our hospital. Women with chronic pelvic pain scheduled to undergo diagnostic laparoscopy were identified from the ambulatory surgery unit operating room list. The attending physician of the patient was asked if his/her patient could participate in the study. Informed consent was then obtained from the patient.
Women were included if they were 1) over the age of 18 years, 2) diagnosed with chronic pelvic pain for at least 6 months, and 3) scheduled to undergo a diagnostic laparoscopy. Women were excluded if they had a prior diagnosis of interstitial cystitis, urinary tract infection in the past 4 weeks, active genital herpes, untreated chlamydia or gonorrhea, current vaginitis, benign or malignant bladder tumors, gynecologic cancer, urinary tract calculi, urethral diverticulum, or possible cystitis due to chemotherapy, radiation, or tuberculosis.
The first author interviewed eligible women. Demographic data was collected and lower urinary tract and gynecologic symptoms were assessed. In addition, women rated symptoms of chronic pelvic pain, dyspareunia, and dysmenorrhea with a visual analogue scale ranging from 0–10. They were then administered the Interstitial Cystitis Symptom Index and Problem Index. Gonorrhea, chlamydia, and urinary tract infection status was determined by either reviewing the chart for pre-operative test results or, if test results were not available, by obtaining a urine sample in the ambulatory surgery unit and/or a cervical swab (DNA probe) in the operating room. Any positive results excluded the patient.
In the operating room, cystoscopy under anesthesia with hydrodistension was performed, using a 30-degree cystoscope. The bladder was filled to capacity with sterile water at a pressure of 70–80 cm H2O and held for 2 minutes. The bladder was then drained and the volume measured to determine the maximum bladder capacity. The cystoscope was reinserted and the bladder mucosa examined for glomerulations, Hunner's ulcer, and linear cracking. Pictures were taken of each quadrant of the bladder. A biopsy was taken to rule out carcinoma in situ of the bladder.
Diagnostic laparoscopy was performed by the attending physician. Gynecologic findings such as dense adhesions, fibroids, and adnexal disease were recorded. Endometriosis was diagnosed by visual confirmation and/or biopsy and was staged according to the revised American Fertility Society classification.7 Endometriosis lesions involving the bladder were noted.
The diagnosis of interstitial cystitis required the presence of all of the following: 1) urgency, 2) urinary frequency or nocturia, and 3) positive cystoscopic findings. Urinary frequency was defined as eight or more voids while awake, and nocturia as two or more voids each night.8,9 A positive cystoscopic finding was defined as either 1) glomerulations (ten or more per quadrant, noted in three or more quadrants of the bladder) or 2) a Hunner's ulcer.8,9 The senior author, who interpreted the cystoscopy pictures, was blinded to the patient's symptoms and Interstitial Cystitis Symptom Index and Problem Index scores. If a bladder biopsy was positive for carcinoma in situ, the diagnosis of interstitial cystitis was excluded.
Demographic data was summarized using means, medians, and standard deviations or percentages. Women were divided into two groups based on the diagnosis of interstitial cystitis. The two-sample t test was used to compare continuous variables between the two groups, the Wilcoxon rank-sum test for two independent samples was used to compare ordinal data (the Symptom Index score, the Problem Index score, and the visual analogue pain scores for symptoms of chronic pelvic pain, dyspareunia, and dysmenorrhea), and the χ2 test was used for categoric variables. A contingency table analysis was done to compare the diagnosis of interstitial cystitis to the laparoscopic diagnosis. Univariate analysis was performed to determine risk factors for the diagnosis of interstitial cystitis. A receiver operator characteristic curve was used to compare the sensitivity and specificity of the Symptom Index and the Problem Index, and to determine the best cut-off value of the Symptom Index or the Problem Index for the diagnosis of interstitial cystitis. Based on the receiver operator characteristic curve cutoff value for the Symptom Index and Problem Index, patients were divided into two groups: those with a score below and those with a score above the cut-off value. Multivariable analysis was performed to determine independent risk factors for the diagnosis of interstitial cystitis. P values < .05 were considered statistically significant. All statistical methods were performed using the SAS software package (SAS Institute Inc., Cary, NC).
There were 55 eligible patients, and 45 (82%) agreed to participate. None of the ten women who declined to participate in the study had irritative bladder symptoms. The 45 women were patients of 26 different attending physicians. The demographic data for the study was: mean age 35.2 years (range 20–53), mean parity 1.9 (range 0–4), and mean duration of chronic pelvic pain 71 months (range 6–120). All but one were premeno-pausal. Seventy-three percent were white, 16% were Hispanic, 2% were black, and 9% were of other ethnicity. Prior operative and therapeutic intervention for chronic pelvic pain included: nonsteroidal anti-inflammatory drug use in 84%, oral contraceptive use in 58%, depot leuprolide use in 7%, prior diagnostic laparoscopy in 45%, and hysterectomy in 9%. The median visual analogue pain scale score was 6 (range 0–10) for dyspareunia, 7 (range 0–10) for dysmenorrhea, and 8 (range 4–10) for chronic pelvic pain. The median score for the Interstitial Cystitis Symptom Index was 7 (range 0–15) and for the Problem Index was 5 (range 0–14).
In this study, 17 women (38%; 95% confidence interval [CI] 22%, 53%) were diagnosed with interstitial cystitis. Of the 45 women, 30 (67%) had clinical symptoms of urgency and either frequency or nocturia. At cystoscopy, 19 of the 45 women (42%) had positive cystoscopic findings. No Hunner's ulcers were seen. Of the 26 women with negative cystoscopic findings, ten women had at least ten glomerulations per quadrant noted in only two quadrants of the bladder, whereas the remaining 16 women had a normal cystoscopy. The presence of positive cystoscopic findings was associated with the presence of urinary symptoms. Seventeen of 30 women with urinary symptoms had positive cystoscopic findings, whereas only two of 15 women without urinary symptoms had positive cystoscopic findings (χ2 = 6.0, P = .01). These two women both were premenopausal and were diagnosed with stage I endometriosis. There were no cases of carcinoma in situ of the bladder found on the bladder biopsies.
Table 1 compares the results of the laparoscopic findings with the diagnosis of interstitial cystitis. All women diagnosed with endometriosis had either stage I or II disease, and one had bladder implants. As shown in Table 1, there was no relationship between the diagnosis of interstitial cystitis and the laparoscopic findings (χ2 = .35, P = .84): seven of 21 (33%) women with endometriosis had interstitial cystitis, four of ten (40%) women with adhesions had interstitial cystitis, and six of 14 (42%) women with negative laparoscopic findings had interstitial cystitis.
Risk factors for the diagnosis of interstitial cystitis were determined using univariate analysis. There was an association of the diagnosis of interstitial cystitis with the following risk factors: the Interstitial Cystitis Symptom Index score (median score of 9 in women with interstitial cystitis versus 5 in women without interstitial cystitis, P = .008), the Problem Index score (median score of 7 versus 2, P = .02), the visual analogue scale pain score for dyspareunia (median score of 8 versus 4.5, P = .008), and a history of one or more urinary tract infections per year (59% versus 25%, P = .02).
No significant association was seen between the diagnosis of interstitial cystitis and the following demographic factors: age, parity, race, tobacco use, caffeine use, prior infection with sexually transmitted disease, prior pelvic surgery, and medical history of allergies, abuse, fibromyalgia, irritable bowel syndrome, migraine headache, physical vulvar pain, and the visual analogue scale pain scores for chronic pelvic pain and dysmenorrhea.
Due to the similarity in the questions of the Symptom Index and Problem Index, and because the Symptom Index had a stronger association with the diagnosis of interstitial cystitis, a cut-off score was calculated for the Symptom Index, using a receiver operator characteristic curve. As shown in Figure 1, a score of 5 or more on the Symptom Index was found to be the best cut-off score. The area under the receiver operator characteristic curve was 0.73. Of the 17 women diagnosed with interstitial cystitis, 16 had a score ≥5. Of the 28 women not diagnosed with interstitial cystitis, 14 had a score ≥5. Therefore, a score of 5 or more on the Symptom Index had a sensitivity of 94% (95% CI 71%, 99.8%), a specificity of 50% (95% CI 31%, 69%), a positive predictive value of 53% (95% CI 34%, 72%), and a negative predictive value of 93% (95% CI 68%, 99.8%) for the diagnosis of interstitial cystitis. Similarly, the best cut-off score of the dyspareunia pain score was 7, and the area under the receiver operator characteristic curve was also 0.73. A score of 7 or more had a sensitivity of 76% (95% CI 50%, 93%), a specificity of 68% (95% CI 48%, 84%), a positive predictive value of 59% (95% CI 36%, 79%), and a negative predictive value of 83% (95% CI 61%, 95%).
Stepwise backward logistic regression was then performed with the following risk factors: a score of 5 of more for the Interstitial Cystitis Symptom Index, a score of 7 or more for the visual analogue scale pain score for dyspareunia, and a history of one or more urinary tract infections per year. The Interstitial Cystitis Symptom Index score of 5 or more and a dyspareunia score of 7 or more remained as independent risk factors (Table 2). Additionally, for those women who had both risk factors, the probability of being diagnosed with interstitial cystitis was even higher: sensitivity 76% (95% CI 50%, 93%), specificity 82% (95% CI 63%, 94%), positive predictive value 72% (95% CI 47%, 90%), and negative predictive value 85% (95% CI 66%, 96%).
The prevalence of interstitial cystitis in women with chronic pelvic pain has not been well studied. In our study population of women with chronic pelvic pain, the prevalence of interstitial cystitis was 38% (17 of 45). It is unlikely that this high prevalence of interstitial cystitis is a result of selection bias because women with chronic pelvic pain were recruited from general gynecology attending physicians and not from a urogynecology or urology division with a high prevalence of urinary symptoms. Forty-five of the eligible 55 patients participated; ten women declined to participate, and none had urinary symptoms. If these ten women had agreed to participate and ultimately not been diagnosed with interstitial cystitis, our prevalence rate would have been 31% (17 of 55).
At laparoscopy, we found endometriosis in 47%, dense adhesions in 22%, and negative findings in 31%. This is similar to the data reported in a review by Porpora and Gomel: of 1386 women with chronic pelvic pain, 31% had endometriosis, 23% had adhesions, 36% had negative findings, and the remaining 10% had other findings.3 In our study, interstitial cystitis was not associated with the laparoscopic findings. Therefore, it is not possible to suggest that cystoscopy is necessary only if the laparoscopy is negative, or to suggest that cystoscopy is unnecessary if endometriosis or other pelvic pathology is found.
The best way to diagnose interstitial cystitis is controversial. Our study used the strict inclusion criteria of the National Institute of Diabetes, Digestive, and Kidney Diseases (NIDDK). This set of criteria was developed in 1987 to diagnose interstitial cystitis,8 and was later revised in 1988.9 These criteria are stringent, include the use of cystoscopy, and may exclude some patients with interstitial cystitis, but they ensure that interstitial cystitis research focuses on relatively homogeneous patient populations. Alternative methods to diagnosing interstitial cystitis have been described. Parsons et al recommends the use of the potassium sensitivity test, which has the advantage of avoiding cystoscopy but at the expense of a painful office procedure.10 The Interstitial Cystitis Data Base study11 is a large, multicenter, observational study designed to identify common patient characteristics and success of treatment in patients with interstitial cystitis. The Interstitial Cystitis Data Base study set guidelines requiring only frequency or urgency; cystoscopy was not required.
The role of cystoscopy in the diagnosis interstitial cystitis has been challenged by Waxman et al,12 whose study reported a high prevalence of glomerulations noted on cystoscopy with hydrodistension in 20 asymptomatic women undergoing laparoscopic tubal ligation. They found that nine of 20 women (45%) had glomerulations noted in the right, left, and posterior aspects of the bladder. However, in our study we noted an association between the presence of urinary symptoms and positive cystoscopic findings, as we used the NIDDK criteria of ten or more glomerulations per quadrant in three of four quadrants of the bladder. It is unlikely that this association is a result of observer bias because the senior author evaluating the cystoscopic findings was unaware of the urinary symptoms reported by women. Whether the glomerulations noted at cystoscopy with hydrodistension are the source of the patient's pain is unknown. They may be a manifestation of a small-capacity bladder that has been stretched, a deficient glycosaminoglycan layer, upregulation of neuropathic pathways in the pelvis, or other unrecognized systemic changes.
A score of 7 or more on the dyspareunia pain scale was found to be an independent risk factor for the diagnosis of interstitial cystitis. The relationship of dyspareunia to interstitial cystitis has been reported earlier; vaginal pain was reported by 51% of women in the Interstitial Cystitis Data Base Study.13 It has been theorized that the close proximity of the bladder to the vagina may be the reason for this, or the common embryologic endodermal derivation of the vestibule and bladder trigone.14 It is unlikely that the relationship between dyspareunia and interstitial cystitis in this sample of women with chronic pelvic pain is due to the confounding effect of endometriosis because we did not see an association between interstitial cystitis and endometriosis.
We have shown that the Interstitial Cystitis Symptom Index is a useful screening tool to diagnose interstitial cystitis in women with chronic pelvic pain. A score of 5 or more on the Symptom Index had a sensitivity of 94%, a specificity of 50%, a positive predictive value of 53%, and a negative predictive value of 93% for the diagnosis of interstitial cystitis. More importantly, if a woman has both risk factors of an elevated Symptom Index score and elevated dyspareunia score, then the probability of being diagnosed with interstitial cystitis is 72% (odds ratio 14.9, 95% CI 2.8, 90.9). Wide confidence intervals reported in our study are likely the result of a relatively small sample size. Larger studies on women with chronic pelvic pain drawn from the general community are required to validate our findings.
In women with chronic pelvic pain, one should consider using the Interstitial Cystitis Symptom Index and the visual analogue scale for dyspareunia as simple office-based screening tests. If either risk factor is present, cystoscopy with hydrodistension should be considered at the time of laparoscopy. Our study has not only identified risk factors for interstitial cystitis, but has also emphasized the importance of taking a sexual and urinary history in women with chronic pelvic pain. This may help to reduce the number of women with chronic pelvic pain in whom no cause can be identified.
1. Zondervan KT, Yudkin PL, Vessey MP, Dawes MG, Barlow DH, Kennedy SH. Prevalence and incidence of chronic pelvic pain in primary care: Evidence from a national general practice database. Br J Obstet Gynaecol 1999;106:1149–55.
2. Mathias SD, Kuppermann M, Liberman RF, Lipschutz RC, Steege JF. Chronic pelvic pain: Prevalence, health-related quality of life, and economic correlates. Obstet Gynecol 1996;87:321–7.
3. Porpora MG, Gomel V. The role of laparoscopy in the management of pelvic pain in women of reproductive age. Fertil Steril 1997;68:765–79.
4. Speizer FE, Hunter DJ, Curhan SG, Stampfer MJ. Epidemiology of interstitial cystitis: A population based study. J Urol 1999;161:549–53.
5. Held P, Hanno P, Wein AJ. Epidemiology of interstitial cystitis: 2. In: Hanno PM, Staskin DR, Krane RJ, Wein AJ, eds. Interstitial cystitis. New York: Springer-Verlag, 1990:29–48.
6. O'Leary MP, Sant GR, Fowler FJ Jr, Whitmore KE, Spolarich-Kroll J. The interstitial cystitis symptom index and problem index. Urology 1997;49:58–63.
7. American Fertility Society. Revised American Fertility Society classification of endometriosis. Fertil Steril 1985; 43:351–2.
8. Gillenwater JY, Wein AJ. Summary of the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases Workshop on Interstitial Cystitis, National Institutes of Health, Bethesda, Maryland, August 28–29, 1987. J Urol 1988;140:203–6.
9. Wein AJ, Hanno PM, Gillenwater JY. Interstitial cystitis: An introduction to the problem. In: Hanno PM, Staskin DR, Krane RJ, Wein AJ, eds. Interstitial cystitis. New York: Springer-Verlag, 1990:3–15.
10. Parsons CL, Greenberger M, Gabal L, Bidair M, Barme G. The role of urinary potassium in the pathogenesis and diagnosis of interstitial cystitis. J Urol 1998;159:1862–7.
11. Hanno PM, Landis JR, Matthews-Cook Y, Kusek J, Nyberg L Jr. The diagnosis of interstitial cystitis revisited: Lessons learned from the National Institutes of Health Interstitial Cystitis Data Base Study. J Urol 1999;161:553–7.
12. Waxman JA, Sulak PJ, Kuehl TJ. Cystoscopic findings consistent with interstitial cystitis in normal women undergoing tubal ligation. J Urol 1998;160:1663–7.
13. Simon LJ, Landis R, Erickson DR, Nyberg LM, and the ICDB Study Group. The Interstitial Cystitis Data Base Study: Concepts and preliminary baseline descriptive statistics. Urology 1997;49:64–75.
14. Moore KL. The developing human: Clinically oriented embryology. 4th ed. Philadelphia: WB Saunders, 1988.