Inflammatory Mediator Profile in Urine and Bladder Wash Fluid of Patients with Interstitial Cystitis

0022-534 7/94/1522-0355$03.00/0 Vol. 152, 355-361, August 1994 Printed in U.S.A.

THE Jrn:JRNAL OF UROLOGY fu"\AERICAN UROLOGICAL AssoCIATION, INC.

Copyright© 1994 by

INFLAMMATORY MEDIATOR PROFILE IN URINE AND BLADDER WASH FLUID OF PATIENTS WITH INTERSTITIAL CYSTITIS DIANE FELSEN,* SUZANNE FRYE, LINDA A. TRIMBLE, TAMARA G. BAVENDAM, C. LOWELL PARSONS, YVONNE SIM AND E. DARRACOTT VAUGHAN, JR. From the Departments of Urology, The New York Hospital-Cornell Medical Center, New York, New York, University of Washington, Seattle, Washington, and Department of Surgery I Urology, University of California at San Diego Medical Center, San Diego, California

ABSTRACT

Interstitial cystitis is a syndrome of urinary urgency, frequency and suprapubic pain. We investigated the role of inflammatory mediators in 96 patients with histories and symptoms consistent with interstitial cystitis, and 13 controls from The New York Hospital-Cornell Medical Center, University of Washington and University of California at San Diego. Patients were classified into either group A (meets all criteria of the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases for inclusion in research studies), group B (meets all of these criteria but without glomerulations) or an "other" group. A small number of group A patients had detectable interleukin-6 in the urine. Urinary concentrations of tumor necrosis factor, prostaglandins E2, D2 and F2a, and thromboxane B2 were not different among either patient groups or controls. Urine specimens contained inhibitors of the bioactivity of interleukin-6 and tumor necrosis factors but no differences between patients or controls were found. No factors chemotactic for human neutrophils were detected in a small patient sample. Bladder wash fluid concentrations of prostaglandins E2, D2 and F2a, and thromboxane were much lower than urinary levels. Bladder wash fluid interleukin-6 and tumor necrosis factor were not detectable. The results suggest that while a small subset of patients may have elevated levels of interleukin-6, the majority of patients do not appear to have elevated levels of inflammatory mediators in the urine or bladder wash fluid. Evaluation of patient bladder tissue may indicate changes not detectable in urine or bladder wash fluid. Alternatively, other etiologies must be considered in those patients. KEY WORDS:

bladder diseases, cystitis, cytokines, eicosanoids

Interstitial cystitis is a syndrome of urinary urgency, frequency and suprapubic pain. A 1987 study estimated that there were approximately 200,000 cases in the United States. However, this number may be an underestimation. The etiology (or etiologies) of interstitial cystitis is unknown. 1 • 2 There are many theories, including increased number or degranulation of mast cells, increased bladder permeability, increases in inflammatory cell populations or mediators and sensory dysfunction. Each of the aforementioned defects may be found in some subset of patients. It is unlikely that there is l underlying explanation for symptoms found in all patients. Since interstitial cystitis is a syndrome, it is diagnosed by symptoms and cystoscopic findings. Reproducible clinical studies have been difficult to achieve. In an effort to provide continuity between studies and to understand interstitial cystitis better, a meeting of the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases (NIDDKD) was convened in 1987. At this meeting criteria were set forth for inclusion of subjects in interstitial cystitis research, including but not limited to a functional bladder capacity of less than 350 ml., Runner's ulcer or a specified number of glomerulations on cystoscopy with the patient under anesthesia, duration of symptoms of greater than 1 year, lack of infection or other known causes of cystitis and absence of carcinoma in situ. 3

Accepted for publication November 19, 1993. Supported in part by Grant DK-42367 from the National Institute of Diabetes and Digestive and Kidney Diseases. *Requests for reprints: Department of Urology, New York Hospital-Cornell University Medical Center, Box 94, 1300 York Ave., New York, New York 10021.

We were interested in the role of inflammatory mediators in interstitial cystitis. To our knowledge there has been little work on mediators to date. Lynes et al examined prostaglandin E2 excretion in urine of patients with interstitial cystitis. 4 In a small number of patients who were studied before the establishment of NIDDKD criteria an increased urinary prostaglandin E2 level was found but only in patients with pyuria, which is not compatible with NIDDKD criteria. Other eicosanoids were not examined. Since the question of mast cell involvement in interstitial cystitis has not been fully resolved, there is an interest in mast cell products in interstitial cystitis" The major prostaglandin metabolite of mast cells is prostaglandin D2. 5 To our knowledge no studies have examined prostaglandin D2 excretion. Recent studies have also examined histamine, another mast cell product, in interstitial cystitis. Yun et al found no difference in urinary histamine levels in normal controls compared with interstitial cystitis patients. 6 However, in interstitial cystitis patients undergoing hydrodistention urine histamine was higher after hydrodistention compared to before distention in a small number. Inflammatory mediators may arise from intrinsic bladder cells or from infiltrating inflammatory cells found in the bladder in interstitial cystitis. A few studies have addressed the nature of inflammatory cells in interstitial cystitis. MacDermott et al found that 8 of 39 patients had severe inflammation in the lamina propria of the bladder. 7 In another study slight changes were noted in the ratio of T:B cells in the bladder and in the proportion of CD4 and CDS cell populations. 8 Others have found changes in T cell and monocyte populations. 9 , 10 Increased inflammation may result in increased inflammatory mediators. Included would be interleukin-6 and tumor necrosis factor. 11 ' 12 Interleukin-6 is a glycoprotein made by

355

356

INFLAMMATORY MEDIATORS IN INTERSTITIAL CYSTITIS

fibroblasts, macrophages and epithelial cells. It is a B cell stimulant and acute phase protein activator, and has antiviral activity. It is found in increased concentration in inflammatory lesions in psoriasis and arthritis. 13 • 14 Tumor necrosis factor is a polypeptide synthesized mainly by macrophages. It is known to stimulate collagenase, angiogenesis, chemotaxis and cachexia.11 We examined the urine and wash fluid of interstitial cystitis patients for several eicosanoids, tumor necrosis factor and interleukin-6. We further examined whether there was an effect of urinary extracts on the biological activity of tumor necrosis factor and interleukin-6. Such findings would be similar to those of Shingleton and Fleischmann, who noted the presence of interleukin-2 inhibitors in urine of interstitial cystitis patients. 15 Whether urine of interstitial cystitis patients is chemotactic for human neutrophils was also studied, since previous studies had indicated that such an activity could be found. 16 These studies were designed to determine if differences in urinary levels of mediators or substances that alter their activity could be detected in interstitial cystitis patients compared with controls. Such differences could help explain the etiology of interstitial cystitis or could be used as tools to diagnose or monitor this disease noninvasively. MATERIALS AND METHODS

Patient selection. Patients were evaluated at the departments of urology of The New York Hospital-Cornell Medical Center (NYH-CUMC), the University of Washington and the University of California, San Diego Medical Center (UCSD). Evaluations were performed by 3 of us (S. F., T. G. B. and C. L. P., respectively). In regard to selection of patients to be studied, at NYHCUMC all patients who presented to 1 of us (S. F.) between September 1989 and September 1992 with a history and voiding symptoms consistent with interstitial cystitis, and who also underwent cystoscopy and hydrodistention while under anesthesia were included in the study group. A group of patients who had been followed at University of Washington (by T. G. B.) and at UCSD (by C. L. P.) were randomly chosen. Patients from the other 2 institutions were included to determine if the findings at NYH-CUMC could be confirmed using patients from other geographic locations. The controls were a group of normal female volunteers with no history of voiding or other urological symptoms. All controls were from the New York area. No further urological evaluation was done on control subjects. Patient evaluation. Each patient studied was evaluated in the same manner. A careful history of voiding symptoms and other medical problems was taken, and a physical examination was performed. Urine was cultured. A cystometrogram was done using normal saline at a fill rate of 50 cc per minute when the patient was awake. At a subsequent visit, each patient underwent cystoscopy while under anesthesia with hydrodistention of the bladder with normal saline using 80 to 100 cm. water pressure for 1 to 2 minutes. After repeat distention, observation was made for glomerulations or Runner's ulcers. Bladder biopsies were then taken. Prominent among the NIDDKD criteria for inclusion of patients in research interstitial cystitis studies and, by definition for research studies, as having a diagnosis of interstitial cystitis were a functional bladder capacity of less than 350 cc and either Runner's ulcer (extremely rare) or glomerulations (at least 10 per quadrant in 3 quadrants). Symptoms of pain associated with the bladder and/or urinary urgency must also be present. Since patients were included in the study based solely on history and symptoms consistent with interstitial cystitis, not all met these criteria. Therefore, patients were separated into group A-patients who met the NIDDKD definition of interstitial cystitis and group B-patients who met all criteria except that during

cystoscopy while under anesthesia there were either no or an insufficient number of glomerulations in the bladder. Patients who did not meet the cystometrogram requirements, regardless of the presence or absence of glomerulations, were categorized as "other." Exclusion criteria were a bladder capacity of more than 350 cc. or a first desire to void of more than 150 cc on an awake cystometrogram; involuntary bladder contractions on cystometrography; duration of symptoms of less than 9 months; absence of nocturia; frequency of less than 8 times per day; cystitis (radiation, tuberculous, bacterial or cyclophosphamide) or vaginitis; uterine, cervical, vaginal or urethral cancer; patient age less than 18 years; benign or malignant bladder tumors; symptomatic urethral diverticulum; active herpes; bladder or lower ureteral calculi, or symptoms relieved by antibiotics, urinary antiseptics or urinary analgesics. Patient specimens. For patients at NYH-CUMC and University of Washington, samples were collected on the day of cystoscopy. Patients were asked to void into and fill a 250 ml. container, and to keep the sample as cold as possible until arrival at NYH-CUMC or University of Washington. Control urine was obtained from patients voiding into a 250 ml. container, as was done with interstitial cystitis patients. Samples from UCSD were from catheterized specimens obtained at cystoscopy with the patient under anesthesia. A bladder wash fluid sample was obtained at cystoscopy at NYH-CUMC, before hydrodistention. Sample preparation. All urine and wash fluid samples were spun at 4C at 300 times gravity before freezing. This step removes any cells from the fluid. Thus, when samples are thawed there is no possibility of releasing intracellular mediators due to the freeze-thaw lysis of cells. For tumor necrosis factor and interleukin-6 measurements, samples were concentrated using Spectrapor dialysis tubing (molecular weight cutoff 6,000 to 8,000) against a slurry of polyethylene glycol. Samples were concentrated approximately 10fold in this step. For measurement of eicosanoids, samples were used without concentration as previously described. 17 Creatinine was measured in voided urine specimens that were not concentrated. The method involves the spectrophotometric determination of creatinine as a picrate derivative. Results are presented as per mg. creatinine when appropriate. Mediator measurement. Tumor necrosis factor: Tumor necrosis factor was measured by enzyme-linked immunosorbent assay (ELISA) and bioassay. The procedure is a sandwich ELISA using plates coated with mouse anti-human tumor necrosis factor antibody. Samples are read at 490/690 nm. for detection of horseradish peroxidase linked samples. This ELISA has been validated for human tumor necrosis factor measurements. 18 The limit of detection of the ELISA is 0.159 ng./ml. Tumor necrosis factor was measured by bioassay using the WEHI clone 13 assay. 19 Tumor necrosis factor samples are incubated with the cells overnight. Then, thiazolyl blue, a dye that is taken up by viable cells, is added. After 4 hours the medium is removed and acidified isopropanol is added, followed by an equal amount of water. Plates are read at 570/690 nm. The limit of detection of the bioassay is 0.012 ng./ml. Samples are assayed in duplicate. Tumor necrosis factor inhibition studies were done by assaying samples in the presence of a known amount of tumor necrosis factor (0.1 ng./ml.) and comparing the ability of tumor necrosis factor to kill WEHI cells in the presence or absence of urine. Interleukin-6: Interleukin-6 was measured by ELISA (using commercially available ELISA kits) and bioassay (using B-9 cells whose growth is dependent on interleukin-6), as described previously. 20 In the latter assay, approximately 5,000 cells per well are cultured in 96 well plates. A standard curve is generated for interleukin-6. Samples are heat-inactivated at 56 degrees for 30 minutes before use and then

357

INFLAMMATORY MEDIATORS IN INTERSTITIAL CYSTITIS

added to the test wells. The ability of samples to stimulate growth of B-9 cells during 4 days is monitored. Cell growth is measured using thiazolyl blue as described for the tumor necrosis factor assay (except that the microtiter plates are centrifuged before removal of the thiazolyl blue-containing media). Interleukin-6 inhibition studies are done using urine or WF samples in the presence of a known amount of interleukin-6. Eicosanoids. The eicosanoids measured were 2,3 dinor thromboxane B2, 13,14-dihydro-15-keto-prostaglandin E2 (prostaglandin E2 metabolite), prostaglandin D2 and prostaglandin F2a. Eicosanoids were measured without extraction by ELISA as previously described. 19 Dinor thromboxane B2 and prostaglandin F2a were measured directly. Prostaglandin E2 metabolite was measured after conversion to bicyclo prostaglandin E2, following derivation in the presence of sodium carbonate overnight at 37C. Prostaglandin D2 was measured after its conversion to the methoxime derivative. Samples were incubated at 60C for 30 minutes in the presence of methyl oximating reagent prepared from methoxyamine hydrochloric acid and anhydrous sodium acetate in aqueous ethanol. All eicosanoids were measured using commercially available antibodies except prostaglandin F2a, which was measured using an antibody prepared in conjunction with Dr. Thomas Lysz (University of Medicine and Dentistry of New Jersey) and used previously. 21 The ELISAs use an acetylcholinesterase-linked eicosanoid, which is then developed using Ellman's reagent. Samples are read at 415 nm. Chemotaxis. Chemotaxis of human neutrophils was measured using a modified Boyden chamber assay. 22 Human polymorphonuclear leukocytes from random donors were prepared using Ficoll-sodium diatrizoate* gradients and counted using a Coulter counter. Polymorphonuclear leukocytes prepared by this method are approximately 98% pure as verified by differential counts. For chemotaxis experiments, 25 µ1. samples were placed in the bottom of the Boyden chamber. They were then covered with a filter and 1 X 105 polymorphonuclear leukocytes were placed in the upper chamber. f-met-leu-phe (f-MLP, 1 X 10- 8 M.) was used as the standard. All samples were done in triplicate. Chemotaxis is measured as the directed migration of neutrophil through the filter using the Optomax program. This program gives a weighted average of the migration from O to 130 µ,. Samples tested for chemotaxis were from freshly voided urine and from frozen specimens. Statistics. Differences between groups were analyzed using analysis of variance followed by means testing using the least significant difference test. RESULTS

Patient demographics. A total of 96 patients and 13 controls was studied. Of the interstitial cystitis patients 69 were from NYH-CUMC, 20 from University of Washington and 7 from UCSD. Controls were all from the New York area. The gender distribution in this study was 92 women and 4 men. All controls were women. Average patient age was 37.9 ± 1.4 years, compared to 30.8 ± 1.8 years for controls. Patients had experienced symptoms for an average of 5. 7 years before diagnosis (range 1 to greater than 10 years). Patient categories. Patients were divided into categories as described. Although all patients were seen due to voiding symptoms and history consistent with interstitial cystitis, only 56.3% of the total patients met all criteria set forth by the NIDDKD to be included in research studies on interstitial cystitis. At NYH-CUMC or University of Washington, 50.0 to 53.6% of the patients were in group A, while 33.3% were in group B (patients who met all criteria except for the presence of glomerulations in the bladder). * Sigma Chemical , St. Louis, Missouri.

Bladder capacity. Table 1 presents the data on bladder capacity of the patients studied, awake and while under anesthesia. Patients in groups A and B have, by definition, a much smaller functional bladder capacity and a much smaller bladder volume at first desire to void than those in the "other" group. All groups had significantly different volumes at first desire to void. Groups A and B patients did not have a significantly different functional capacity but were different from those in the "other" group. Bladder capacity with the patient under anesthesia was also smaller in groups A and B than in the "other" patients. However, groups A and B were not significantly different from each other. The mean bladder capacity in group A was 739.8 ± 35.5 ml. in contrast to 1,260 ± 158.5 ml. in the "other" group. Group A patients from the NYH-CUMC had a total bladder capacity of760.l ± 37.9 ml., which is not different from the mean capacity of all group A patients. The bladder capacity ranged from 250 to 1,350 ml. in group A, 450 to 1,200 ml. in group B and 850 to 1,750 ml. in the "other" group. Infiammatory mediator measurements. Tumor necrosis factor in concentrated urine samples was measured by bioassay and ELISA. Almost all patients and controls were positive for tumor necrosis factor in the bioassay (fig. 1). Group A patients had a mean urinary tumor necrosis factor of 7.3 ± 1.0 pg./mg. creatinine, compared with 16.8 ± 5.4 for group B and 8.5 ± 1.6 for controls. When the same samples were assayed by ELISA, different results were obtained (table 2). The patient samples could be divided into 2 groups, 1 of which was below the limit of detection of the assay. For example 28 of 4 7 group A patients had undetectable levels of tumor necrosis factor. The "other" group had detectable levels that were much higher than those found in the bioassay. The results demonstrate that of the 19 group A patients who were positive for tumor necrosis factor, the mean level was 127.2 ± 46.8 pg.frog. creatinine. Only 2 of 13 controls were positive. Interestingly, several group B patients showed high levels of tumor necrosis factor (4 of 16 had greater than 1,000 pg.frog. creatinine). These results suggest that there is an immunogenic component to tumor necrosis factor that is recognized by the ELISA but is not biologically active in the WEHI cell killing assay. Interleukin-6. In contrast to tumor necrosis factor, which was found in almost all samples, interleukin-6 was detected in only a few samples of concentrated urine by either ELISA or bioassay. Of 7 patients who were positive for interleukin-6 by ELISA 6 were from group A and 1 was from group B, (mean values 6.60 ± 1.21 pg./ml. and 233.2 pg./ml., respectively). On bioassay, all 6 group A patients had detectable interleukin-6 levels, whereas the group B patient did not. This finding suggests that there was an interfering substance in the urine of the group B patient. Thus, a subgroup of patients had significant interleukin-6 levels but the majority did not. Interleukin-6 and tumor necrosis factor were undetectable in urine that was not concentrated before assay. Eicosanoids. Four eicosanoids were measured in the urine: prostaglandin E2 metabolite, dinor thromboxane B2, prostaglandin D2 and prostaglandin F2a. All patients had detectable levels of eicosanoids in the urine. The results are shown in table 3. Analysis of variance showed that there were no statistical differences among the groups. Since prostaglandin D2 is the major metabolite of mast cells, which have been implicated in interstitial cystitis, we were especially interTABLE

1. Bladder capacities of interstitial cystitis

patients and others Bladder Vol. (ml.) At first desire to void Functional (awake) Capacity

Group A (54 pts.) 58.1 168.9 739.8

5.0 10.8 35.5

Group B (32 pts.) 79.5 200.3 813.1

7.5 11.9 38.2

Others (10 pts.) 228.3 468.7 1260

47.0 67.0 158.5

358

INFLAMMATORY MEDIATORS IN INTERSTITIAL CYSTITIS

24 22 ,--,

20

(1)

C C

:;:;

18

0

(]) L

u

16

O>

E

"

O> CL

14

'-.../

LI-

z

12

f-

10 8

6 A

B

OTHER

CONTROL

Fm. 1. Urinary tumor necrosis factor (TNF) was measured in interstitial cystitis patients and controls using WEHI clone 13 bioassay as described. A, group A. B, group B.

ested in its measurement. Furthermore, prostaglandin D2 can be converted to prostaglandin F2a and, therefore, prostaglandin F2a was measured also. However, there were no differences in urinary excretion of prostaglandin D2 or F2a by any group of patients compared to controls. Chemotactic factors. We investigated whether factors that would be chemotactic for human neutrophils might be present in the urine of interstitial cystitis patients. Using the Optomax program, which provides a leukotactic index (a weighted average of migration over a space of 130 µ,.), we noted that the leukotactic index ofrandom neutrophil migration was 16.7. The leukotactic index was increased to 29.55 (176.9%) by the known neutrophil chemotactic peptide f-MLP (fig. 2). Aliquots were tested from urine frozen from 5 group A patients. The chemotactic index was 8.8. Since we were concerned that freezing the samples would cause a loss of chemotactic activity we also assayed chemotaxis in freshly isolated urine from 3 group A and 3 group B patients. Fresh urine from either group A or 2 patients was not chemotactic for human neutrophils (fig. 2, group A patients-leukotactic index 14.47 and group B patients-chemotactic index 17.10). Dilution of the samples did not result in any further chemotactic ability. Bladder wash fluid from 2 patients in group A and 3 in group B also showed no chemotactic ability. Inhibitors of tumor necrosis factor or interleukin-6 bioactivity. In our studies, we observed that the bioassayable levels of interleukin-6 (and a segment of the tumor necrosis factor) were lower than those in the ELISAs, which suggested that there might be inhibitors of interleukin-6 or tumor necrosis factor in the urine being tested. If there were, it would also be of interest if the level were different in urine of interstitial cystitis patients compared to others. The presence of inhibitors was measured by bioassay of urine in the

TABLE

2. Urinary tumor necrosis factor in interstitial cystitis and

other patients as detected by ELISA Group A B Others Controls

Tumor Necrosis Factor (pg./mg. creatinine) 127.2 433.2 208.3 122.6

46.8 129.5 70.0

No. Pts. Detectable/No. Assayed

19/47 16/26 5/9 2/13

presence of a known amount of tumor necrosis factor or interleukin-6. Inhibition of tumor necrosis factor bioactivity was done by adding a 10 µ,l. aliquot of urine to WEHI cells in the presence of a known amount of tumor necrosis factor. The ability of tumor necrosis factor to kill WEHI cells in the presence and absence of urine was compared. Urine from only a small number of patients was able to inhibit tumor necrosis factor killing. Nine group A patients were positive for ability to inhibit tumor necrosis factor. The inhibition ranged from 1.6 to 29.1 %. Urine from 6 control patients also inhibited tumor necrosis factor activity, with a range of 1.8 to 8.7%. One group B patient and 4 "other" group patients were also active. In contrast to the results with tumor necrosis factor, urine from all patients inhibited the activity of interleukin-6 in the B9 bioassay. In this assay, various dilutions of urine are used (compared to the tumor necrosis factor assay in which 1, 10 µ,l. aliquot is tested). The most concentrated samples (1:1 dilution with B9 cells) inhibited interleukin-6 activity by greater than 90% in all samples tested. As the urine concentration was diluted, the ability to inhibit interleukin-6 activity was diluted. In some samples the interleukin-6 inhibitor activity was diluted out at lower concentrations than others but there was no pattern that corresponded to any patient categories. Bladder wash fiuid. To determine if mediators could be released directly from the bladder, the bladder wash fluid obtained during cystoscopy, before hydrodistention, was also assayed. Bladder wash fluid eicosanoid values are shown in table 4. Wash fluid values were strikingly lower than the corresponding urinary eicosanoid levels. There were small differences among the groups. However, the wash fluid from group A did not have the highest concentration of any eicosanoid examined. Bladder wash fluid interleukin-6 and tumor necrosis factor concentration were not detectable (data not shown). DISCUSSION

Although interstitial cystitis may affect as many as 200,000 people in the United States, little is known of the underlying mechanisms of its pathogenesis. Several mechanisms, including changes in mast cells, bladder permeability, sensory dysfunction or inflammatory changes, have been suggested. 1 • 2 We examined various aspects of the synthesis or biological activity of several inflammatory mediators in urine or wash fluid of interstitial cystitis patients, including eicosanoids (prostaglandin E2, thromboxane A2, prostaglandin D2 and prostaglandin F2a), interleukin-6, tumor necrosis factor and chemotactic factors. There were no significant differences between interstitial cystitis patients and others with respect to the aforementioned mediators. Previous literature had suggested that inflammatory findings were characteristic of interstitial cystitis. However, those findings were from patients with classical or ulcerative interstitial cystitis, which is characterized by Runner's ulcer, decreased bladder capacity, fibrosis and severe inflammation. 2 Another different type of interstitial cystitis, referred to as early or nonulcerative interstitial cystitis, is characterized by bladders that are neither shrunken nor fibrotic. Histological changes are not consistent and are much less pronounced. For example, Johanssen and Fall found that patients with nonulcerative interstitial cystitis had sparse inflammatory infiltrates with small mucosal ruptures and suburothelial hemorrhages. 23 Lynes et al found significant histological changes (ulceration, denuded epithelium and submucosal inflammation) only in patients with pyuria or a small bladder capacity. 24 There is no evidence that early interstitial cystitis progresses to the ulcerative form of interstitial cystitis. All but 1 of our patients have nonulcerative interstitial cystitis. Ulcerative interstitial cystitis represents less than 5% of all interstitial cystitis patients.

359

INFLJLMMATORY MEDIATORS IN INTERSTITIAL CYSTITIS TABLE 3. Urinary eicosanoids (pg./mg. creatinine) Dinor Thrornboxane B2

Group

837 ± 76 A 1,030 ± 140 B Other 975 ± 182 Controls 1,036 ± 173 * Measured as bicyclo derivative. t Measured as methoxamine.

13,14 Dihydro-15-Keto Prostaglandin E2* 799 980 1,007 1,312

± 94

± 135 ± 273 ± 239

35

D

GEY'S ~ FMLP

30

~ A ~ A- dil B B [II] B-dil

25

20

15

10

FIG. 2. Effect of urine in patients on human neutrophil chemotaxis. Human neutrophils were prepared and chemotaxis assayed in Boyden chambers as described. Stimuli used were Gey's balanced salt solution (Gey's, random migration), f-MLP (FMLP), bacterial peptide, positive control for chemotaxis, urine from group A patients (A), diluted urine from group A patients (A-dil), urine from group B patients (B) and diluted urine from group B patients (B-dil). Leukotactic index is derived from Optomax computer program which measures neutrophil movement from O to 140 µ,M. (3 to 5 per stimulus).

Although inflammatory changes may not be pronounced in all interstitial cystitis patients, there may still be changes in the cell population in the bladder of some patients. Inflammatory mediators may be produced by either intrinsic bladder cells or infiltrating cells. Klahr et al showed that part of the eicosanoids population found in the urine comes directly from the bladder. 25 Cytokine synthesis by normal bladder tissue has not been well studied. Mast cell infiltration into the bladder was considered to be a hallmark of interstitial cystitis. However, findings have been inconclusive. Hanno et al found detrusor mastocyi;osis in 64% of interstitial cystitis patients and 80% of noninterstitial cystitis patients. 26 However, Christmas and Rode found significantly greater numbers of mast cells in the detrusor in interstitial cystitis compared to controls or patients with bacterial cystitis. 27 In urothelium and submucosa, mast cell numbers were greater in interstitial cystitis and bacterial cystitis compared to controls. Thus, the question of mast cells has not been fully resolved. Although actual mast cell numbers may not be different, it has been suggested that mast cells may be more activated and their contents released more easily in interstitial cystitis patients. The major prostaglandin synthesized by mast cells is prostaglandin D2. We previously described the existence of an enzyme, 11-keto reductase, that can convert prostaglandin D2 to prostaglandin F2a. 28 We examined excretion of prostaglandins D2 and F2a in urine or wash fluid of interstitial cystitis patients. There were no significant differences between interstitial cystitis patients and controls. A recent study on a small number of patients showed in-

Prostaglandin D2t

899 828 850 806

± ± ± ±

105 70 241 90

Prostaglandin F2a

1,068 998 1,030 1,391

± ± ± ±

135 119 149 141

No. Pts.

41 24 10

13

creased urinary histamine in distended bladders of interstitial cystitis patients. 6 Whether these differences reflect mast cell activation cannot be determined to date. Urinary secretion of either prostaglandin E2 or thromboxane A2 was also not different between interstitial cystitis patients and controls. The mast cell population is not the only cell population that may change in interstitial cystitis. For example, infiltrating inflammatory cells can be found in the bladder. T-helper/ inducer cells are the predominant T cell found. B cells and monocyte/macrophages may also be present. Tumor necrosis factor and interleukin-6 are made predominantly by macrophages.11· 12 In our study, a small subset of group A patients had detectable levels of interleukin-6 in the urine. It is possible that in this small subset of patients interleukin-6 may contribute to the pathogenesis of the disease and to the symptomatology. Elevated levels of interleukin-6 have been previously demonstrated in psoriatic lesions and in inflamed joints. 13 ' 14 Perhaps this subset of patients could be amenable to treatments that might use interleukin-6 antagonists or interleukin-6 receptor antagonists. Unfortunately, this group constitutes only a small percentage of group A patients. Low levels of urinary tumor necrosis factor were found in almost all patients. However, there were no significant differences among the groups. It remains to be seen if tissue expression of these cytokines differs and such studies are presently underway. A recent report by Liebert et al examining tissue expression of tumor necrosis factor confirmed that only a small number of samples stained positively for tumor necrosis factor. 29 Although the specific inflammatory mediators that we measured do not appear to be elevated in urine or wash fluid of interstitial cystitis patients, there is evidence that urine of interstitial cystitis patients may modulate the bioactivity of other substances. Shingleton and Fleischmann demonstrated interleukin-2 inhibitors in urine of interstitial cystitis patients using an interleukin-2 dependent CT-6 cell line, which proliferates in the presence of interleukin-2. 15 Elgebaly et al showed that urine of interstitial cystitis patients is chemotactic for elicited rabbit peritoneal neutrophils. 16 We examined whether urine could modulate biological effects of interleukin-6 or tumor necrosis factor, namely the ability of interleukin-6 to support growth of the B-9 cell line and the ability of tumor necrosis factor to kill WEHI clone 13 cells. Other biological activities of these cytokines were not examined. Urine from interstitial cystitis patients substantially inhibited the ability of interleukin-6 to support B-9 growth. This inhibitory ability was diluted out as the urine samples were diluted. Unextracted and concentrated extracts inhibited interleukin-6 bioactivity. However, there was no difference between the effects of urine from interstitial cystitis patients compared to other patients or controls. In contrast to the finding in all patients when interleukin-6 was examined, only urine from a portion of all patients was shown to inhibit tumor necrosis factor bioactivity. Previous work had shown that urine of only interstitial cystitis patients was chemotactic for rabbit peritoneal neutrophils elicited with glycogen. In our study a small number of patients were tested for chemotactic ability. We used human neutrophils obtained from

360

INFLAMMATORY MEDIATORS IN INTERSTITIAL CYSTITIS TABLE

Group A B Others

4. Eicosanoids in bladder wash fluid of interstitial cystitis patients

Prostaglandin E2

Prostaglandin F2a

Prostaglandin D2Methoxamine

Dinor Thromboxane B2

No. Pts.

37.6 ± 11.7 58.6 ± 31.7 15.9 ± 3.2

57.8 ± 18.5 41.2 ± 21.2 25.7 ± 8.9

24.0 ± 3.5 41.9 ± 19.9 17.5 ± 2.9

34.9 ± 7.1 60.5 ± 31.6 22.4 ± 4.6

18 13 5

peripheral blood of random donors. Neutrophil migration in modified Boyden chambers was used as our test of chemotaxis. The standard used was 10-s M. f-MLP, the bacterial chemotactic peptide. f-MLP consistently increased migration by approximately 80%. In no case, using either fresh or frozen urine or wash fluid, were we able to detect chemotactic activity in urine samples done at the same time as f-MLP. It is possible that elicited rabbit peritoneal neutrophils differ in some significant way from human peripheral blood neutrophils such that the urine samples are chemotactic for rabbit neutrophils but not human neutrophils. However, it is clear that looking for chemotactic factors will not be a diagnostic tool for interstitial cystitis. The lack of diagnostic criteria and a definitive etiology has led to major problems in the study of interstitial cystitis. In an effort to understand interstitial cystitis better, as well as to have a more uniform set of criteria to define interstitial cystitis, a meeting was convened in 1987 by the NIDDKD. Criteria were agreed upon by which to include patients in research studies on interstitial cystitis. 3 In this study 67 consecutive patients who presented to 1 of us (S. F.) with a history and symptoms consistent with interstitial cystitis were entered into the study. Six of them were excluded from the NIDDKD criteria by cystometrogram findings. However, the remaining 90% of the patients met all eligibility requirements. At cystoscopy with the patient under anesthesia, only 60% of the patients met the final inclusion criterion for glomerulations. Only 1 patient had a documented Runner ulcer. Thus, 40% of the patients who underwent cystoscopy did not meet the NIDDKD criteria for inclusion in research studies as interstitial cystitis patients. Although patients from University of Washington were chosen at random, the same percentages were found. This finding suggests that the NIDDKD criteria may be too restrictive and that further modifications in the classification of patients presenting with history and symptoms of interstitial cystitis may be warranted. Separation of patients into groups A and B did lead to an interesting finding about bladder capacity. Patients in both groups did not have functional capacities different from each other and both were different from the "other" group patients. However, group A patients had severe glomerulations whereas group B patients had little or none. Therefore, functional bladder capacity does not seem to predict the subsequent appearance of glomerulations when the bladder is distended. Both groups did have significantly smaller bladder capacities with the patient under anesthesia than the "other" group. However, with a mean bladder capacity of greater than 740 ml. these are clearly not low capacity bladders. Of 53 group A patients only 7 had a capacity of 400 ml. or less and 12 had a capacity of 1,000 ml. or more. Volumes for functional capacity and capacity with the patient under anesthesia are similar to those found in Parsons' study of more than 200 patients. 30 The results suggest that most patients who present with interstitial cystitis have the nonulcerative type with bladder capacity nearly equal to that of noninterstitial cystitis patients. The presence of ulcerative interstitial cystitis is rare in this study. In a recent study by Koziol et al on the natural history of interstitial cystitis the mean patient age at the time of survey was 53.8 ± 0. 7 years. 31 The mean age at the first symptom was 42.5 ± 0.8 years. The age of these 374 patients was slightly higher than that found for our group.

In our study almost 100 patients with nonulcerative interstitial cystitis were included. We examined urine and bladder wash fluid for the presence of inflammatory mediators or for substances modulating this activity. The majority of patients were no different from controls. A subset of patients had elevated interleukin-6 in the urine, suggesting that this small subset of patients may have alterations in inflammatory mediators that could one day be amenable to specific therapy. Examination of bladder tissue may indicate changes in inflammatory mediators that are not detectable in urine. Alternatively, other proposed etiologies may account for the syndrome of interstitial cystitis in other subsets of patients. Dr. Lyle Moldawer, CUMC Laboratory of Surgical Metabolism, donated the antibodies used for ELISA and the B-9 cells (developed by Dr. L. Aarden) used for interleukin-6 bioassay. WEHI cells were obtained by Dr. John C. Mathison, Scripps Clinic and Research Foundation. REFERENCES

1. Wein, A. J.: Interstitial cystitis. Sem. Urol., 9: 71, 1991. 2. Sant, G. R.: Interstitial cystitis: pathophysiology, clinical evaluation, and treatment. Urol. Ann., 3: 171, 1989. 3. Gillenwater, J. Y. and Wein, A. J.: 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., 140: 203, 1988. 4. Lynes, W. L., Flynn, S. D., Shortliffe, L. D., Lemmers, M., Zipser, R., Roberts, L. J., II and Stamey, T. A.: Mast cell involvement in interstitial cystitis. J. Urol., 138: 746, 1987. 5. Lewis, R. A., Soter, N. A., Diamond, P. T., Austen, K. F., Oates, J. A. and Roberts, L. J., II: Prostaglandin D2 generation after activation of rat and human mast cells with anti-IgE. J. lmmunol., 129: 1627, 1982. 6. Yun, S. K., Laub, D. J., Weese, D. L., Lad, P. M., Leach, G. E. and Zimmern, P. E.: Stimulated release of urine histamine in interstitial cystitis. J. Urol., 148: 1145, 1992. 7. MacDermott, J.P., Charpied, G. C., Tesluk, H. and Stone, A. R.: Can histological assessment predict the outcome in interstitial cystitis? Brit. J. Urol., 67: 44, 1991. 8. MacDermott, J. P., Miller, C. H., Levy, N. and Stone, A. R.: Cellular immunity in interstitial cystitis. J. Urol., 145: 274, 1991. 9. Harrington, D. S., Fall, M. and Johansson, S. L.: Interstitial cystitis: bladder mucosa lymphocyte immunophenotyping and peripheral blood flow cytometry analysis. J. Urol., 144: 868, 1990. 10. Zhou, Z. Z., Zheng, Y., Monson, F. C., Wein, A. J., Ruggieri, M., Levin, R. M., Hanno, P., McKenna, B. A. W. and Teuscher, C.: lmmunohistochemical analysis of the urinary bladder: a comparative study of the cellular immune response of patients diagnosed as having interstitial cystitis, noninterstitial cystitis voiding dysfunctions, and normal controls. Sem. Urol., 9: 115, 1991. 11. Le, J. and Vilcek, J.: Tumor necrosis factor and interleukin 1: cytokines with multiple overlapping biological activities. Lab. Invest., 56: 234, 1987. 12. Akira, S., Hirano, T., Taga, T. and Kishimoto, T.: Biology of multifunctional cytokines: IL 6 and related molecules (IL 1 and TNF). FASEB J., 4: 2860, 1990. 13. Neuner, P., Urbanski, A., Trautinger, F., Moller, A., Kirnbauer, R., Kapp, A., Schiipf, E., Schwarz, T. and Luger, T. A.: Increased IL-6 production by monocytes and keratinocytes in patients with psoriasis. J. Invest. Dermatol., 97: 27, 1991. 14. Nawata, Y., Eugui, E. M., Lee, S. W. and Allison, A. C.: IL-6 is the principal factor produced by synovia of patients with rheu-

INFLAIVIMATORY MEDIATORS IN INTERSTITIAL CYSTITIS

15. 16.

17.

18.

19.

20.

21.

matoid arthritis that induces B-lymphocytes to secrete immunoglobulins. Ann. N. Y. Acad. Sci., 557: 230, discussion 239, 1989. Shingleton, W. B. and Fleischmann, J.: Urinary interleukin-2 inhibitor and the voiding symptoms in women patients with interstitial cystitis. Sem. Urol., 9: 120, 1991. Elgebaly, S. A., Allam, M. E., Walzak, M. P., Jr., Oselinsky, D., Gillies, C. and Yamase, H.: Urinary neutrophil chemotactic factors in interstitial cystitis patients and a rabbit model of bladder inflammation. J. Urol., 147: 1382, 1992. Perlmutter, A. P., Miller, L., Trimble, L. A., Marion, D. N., Vaughan, E. D., Jr. and Felsen, D.: Toradol, an NSAID used for renal colic, decreases renal perfusion and ureteral pressure in a canine model of unilateral ureteral obstruction. J. Urol., 149: 926, 1993. Barber, A. E., Coyle, S. M., Marano, M.A., Fischer, E., Calvano, S. E., Fong, Y., Moldawer, L. L. and Lowry, S. F.: Glucocorticoid therapy alters hormonal and cytokine responses to endotoxin in man. J. Immunol., 150: 1999, 1993. Espevik, T. and Nissen-Meyer, J.: A highly sensitive cell line, WEHI 164 clone 13, for measuring cytotoxic factor/tumor necrosis factor from human monocytes. J. Immunol. Meth., 95: 99, 1986. Thompson, W. A., Coyle, S., Vanzee K., Oldenburg, H., Trousdale, R., Rogy, M., Felsen, D., Moldawer, L. and Lowry, S. F.: The metabolic effects of platelet-activating factor antagonism in endotoxemic man. Arch. Surg., 129: 72, 1994. Weisman, S. M., Felsen, D. and Vaughan, E. D., Jr.: The effect of sodium intake on renal prostaglandin production. Proc. Soc. Exp. Biol. Med., 181: 357, 1986.

361

22. Wolpe, S. D., Sherry, B., Juers, D., Davatelis, G., Yurt, R. W. and Cerami, A.: Identification and characterization of macrophage inflammatory protein 2. Proc. Natl. Acad. Sci., 86: 612, 1989. 23. Johansson, S. L. and Fall, M.: Clinical features and spectrum of

24. 25. 26. 27. 28. 29.

30. 31.

light microscopic changes in interstitial cystitis. J. Urol., 143: 1118, 1990. Lynes, W. L., Flynn, S. D., Shortliffe, L. D. and Stamey, T. A.: The histology of interstitial cystitis. Amer. J. Surg. Path., 14: 969, 1990. Reyes, A. A. and Klahr, S.: Bladder contributes to eicosanoids excreted in urine. Amer. J. Physiol., 259: F859, 1990. Hanno, P., Levin, R. M., Monson, F. C., Teuscher, C., Zhou, Z. Z., Ruggieri, M., Whitmore, K. and Wein, A. J.: Diagnosis of interstitial cystitis. J. Urol., 143: 278, 1990. Christmas, T. J. and Rode, J.: Characteristics of mast cells in normal bladder, bacterial cystitis and interstitial cystitis. Brit. J. Urol., 68: 473, 1991. Reingold, D. F., Kawasaki, A. and Needleman, P.: A novel prostaglandin 11-keto reductase found in rabbit liver. Biochim. Biophys. Acta, 659: 179, 1981. Liebert, M., Wedemeyer, G., Stein, J. A., Washington, R., Jr., Faerber, G., Flint, A. and Grossman, H. B.: Evidence for urothelial cell activation in interstitial cystitis. J. Urol., 149: 470, 1993. Parsons, C. L.: Interstitial cystitis: clinical manifestations and diagnostic criteria in over 200 cases. Neurourol. Urodynam., 9: 241, 1990. Koziol, J. A., Clark, D. C., Gittes, R. F. and Tan, E. M.: The natural history of interstitial cystitis: a survey of 374 patients. J. Urol., 149: 465, 1993.