Multiple Urinary Cytokine Levels of Bacterial Cystitis

Vol. 157,1980-1985, May 1997 Pnnkd in U S A

MULTIPLE URINARY CYTOKINE LEVELS OF BACTERIAL CYSTITIS RAMIN DAVIDOFF, RON YAMAGUCHI, GARY E. LEACH, EUNHEE PARK

AND

PRAMOD M. LAD*

From the DepaTtment of Urology and Regional Reseamh Laboratory, Kaiser Foundation Hospitals, LQS Angeles, California

ABSTRACT

Purpose: We have examined urinary cytokine levels to define the inflammatory response in patients with bacterial cystitis or microhematuria relative to normal subjects. Cytokines examined include interleukin-Ip (IL-lp),&la, tumor necrosis factor (TNFa),IL-6 and IL-4. Unique features of this study include a) a simultaneous study of several relevant cytokines b) a study of the inflammatory response a t both low and high counts of bacterial infection and c) an assessment of whether microhematuria without bacterial cystitis or pyuria is associated with cytokine elevation compared to normals. Materials and Methods: Enzyme immunoassays were utilized for each cytokine. Patients studied include those with bacterial cystitis (n = 49), patients with microhematuria (n = 11),and normal subjects (n = 36). Cytokine levels were also determined for patients with low count bacterial cystitis (1,000-50,OOO organisms; n = 15)and compared to high count bacterial cystitis (>100,000 organisms; n = 34) and normal subjects. Statistical analysis was carried out using the Kruskal-Wallis test followed by pairwise testing with Newman-Keuls test. Results: a) The means for normal, microhematuria and bacterial cystitis groups were significantly different (p <0.05)for IGlP, IL-la, TNFa and IL-6, but not for IL-4. b) Except for IL-4, all cytokines were found to be significantly elevated in low count bacterial cystitis compared to normals. No statistically significant difference was observed between low and high count bacterial cystitis groups for any of the cytokines tested. Conclusions: a) Significant and similar inflammatory responses are present in both low and high count bacterial cystitis groups as compared with the normal group. b) IL6 and TNFa are significantly elevated in patients with microhematuria compared to normals. c) The potential clinical utility of the assays lies in identifylng the specific cytokines elevated, understanding the pathways that give rise to their production, and in defining potential virulence factors that may produce significant inflammation a t low count bacterial infections. Key WORDS:cytokines, cystitis, urinary tract

Analysis of urinary cytokines and mediators of inflammation is emerging as an important area of urologic research. Previous important findings in cytokine research include a) elevation in I G l p in bacterial cystitis but not in interstitial cystitis (IC)l, b) the increase in urinary histamine metabolites and tryptase in IC with or without hydrodistention2-4, c) the use of IGF and T N F a in monitoring the therapeutic response to BCG6.7, d) the potential increase in IG2 in bladder cancee.9 and most recently the report that IGS may be elevated in vesicoureteral reflux disease.1° These results suggest that cytokines may be elevated in bladder diseases such as IC, bladder cancer, and bacterial cystitis and may play a key role in defining the disease type and severity.ll-13 IG6 has also been reported to be elevated in pyelonephritis,l4 in bladder cancer15 and also in renal transplant recipients.16 Exclusion of bacterial cystitis is an important element in defining cytokine levels in other complex disease entities such as IC and bladder cancer. A limitation in realizing this goal is that the analysis of cytokine elevations in bacterial cystitis has been camed out on a limited basis and the relative value of one cytokine determination in comparison to the others is unclear. Recent approaches have explored model systems in which deliberate colonizationof the bladder has been used to assess cytokine responses.17.18 We have therefore chosen a panel of cytokines to elucidate

infection

inflammatory pathways in bacterial cystitis. The choice of the cytokines was based on current findings that are briefly discussed here. For example, IL-lpl and IL-618 were included because both cytokines are signiilcantly elevated in bacterial cystitis, although simultaneous elevations in both cytokines have not been critically examined. Studies with bladder epithelial cell lines have shown that IL-6 may be produced from bladder epithelial cell lines particularly after stimulation with various hormones,’g suggesting that the bladder epithelium is capable of producing IL-6. Furthermore, the role of IG6 in IC remains unclear. Elevation of IL-6 in IC has been previously reported,20 but not confirmed. The studies presented here are therefore potentially important in assessing the role of IL-6 as a differential marker between IC and bacterial cystitis in the future. IL4, like IL-6, is associated predominantly with Blymphocyte growth. However, IL-4 is much more cell-line restricted and is released predominantly by mast cells, and plays a role in lymphocyte development.21 The role of IL-4 in bladder diseases is not known. The role of I G l a is particularly intriguing because it is a membrane associated cytokine.22 IL-lahas been shown to regulate ICAM-1expression.23 ICAM-1 is an adhesion molecule that could in part contribute to bacterial infection as well as to other cellular interactions. IL-la may be amplified in and possibly released from bladder cell lines.24 A recent Accepted for publication December 23, 1996. uesta for reprints Department of Urology and Regional Re- report has described elevation in IL-la in bacterial infections among the bacteriuric institutionalized elderly.25 se:m%boratory, 1515 Vermont Ave. (Basement),Kaiser Founda. tion Hospitals, Los Angeles, CA 90027. In contrast to IL-la, TNFa is released during the last 1980

1981

MULTIPLE URINARY CYTOKINE LEVELS

stages of macrophage differentiation. Thus, IL-la and TNFa present the initial and terminal steps of a complex macrophage cascade. TNFa has also been proposed as a potentially useful marker in the therapeutic monitoring of BCG treatment.2” Finally, IL-2 couid not be evaluated in this panel because the limit of detection for this assay precluded the determination of a normal range. Our studies assess the specific cytokine elevations in bacterial cystitis as well as microhematuria, and examine their association with low count bacterial infections. MATERIALS AND METHODS

Clean catch urine samples were collected from patients in the urology clinic and the hospital laboratory. Urine samples were subjected to a dipstick analysis, and urine cultures were obtained. Urine cultures showing greater than 100,000 organisms were classified as high count bacterial cystitis. Counts in the range of 1,000-50,000 were classified as low count bacterial cystitis. None of the patients had infections in the range of 50,000-100,000 organisms. For these studies, the bacterial cystitis group was defined only on the basis of positive urine culture and number of organisms. No attempt was made to further subdivide the bacterial cystitis group using other parameters such as red blood cells. Normal urine samples were collected from the Employee Health Center, with IRB informed consent being obtained to alert patients to the purpose of the study. Normal subjects are defined as those whose urine samples were negative on dipstick, urinalysis, urine culture, and individuals who had no known urologic history. The microhematuria group included patients who were urine culture negative but were dipstick positive for hemoglobin or had 10-25 red blood cells on urinalysis. Microhematuria was observed as the initial clinical presentation in this group, and the clinical workup for microhematuria is in progress in these patients. For some patients the reasons for the microhematuria remain unknown (table 1). Gender and age distribution were similar between the normal subjects and the bacterial cystitis subjects. Subjects were randomly recruited into both groups, thus no bias was built into the research design. The sample size was constrained by three factors: a) logistics of doing multiple cytokine measurements b) costs of the assays themselves and c) the limited availability of the patients at least in some categories in our clinics. The patients were evaluated as they presented with symptoms and were thus not undergoing a n y treatment. Assays for cytokines. Enzyme immunoassay ( E m R & D Systems, Minneapolis, MN) was used to measure the urinary cytokines.27.28 The assay methodology employs antibody bound to a solid surface that traps the cytokine and a second horseradish peroxidase linked monoclonal antibody to detect the bound cytokine. Color development following the horseradish peroxidase enzymatic reaction was monitored using a Biotek EIA reader. A standard curve was constructed using purified cytokine supplied in the kit. Validation of assays. An obvious concern is that artifacts

TABLE1. Summary ~ _ _ _ _ _ _

Patients

of diagnosis and cytokine elevation in Datients microhematuria _ _ _ _ _ ______

Diaenosis

Cvtokine Elevated

Simple renal cyst T N F a , IL-6 2 Renal cyst TNFa, IL-6 3 Renal cyst None 4 Grade I1 TCC* &la, TNFa, IL-6 5 IgA nephropathy** TNFa 6 Urethral stricture None 7-10 Unknown None 11 Unknown TNFa, IL-6 * Transitional-cell carcinoma; urine collected prior to BCG. ** IL6 not assayed. 1

be avoided with respect to urine collection and storage. Loss of urinary cytokines results from sample handling or storage. False positive cultures could result from inadequate collection procedures. The samples were collected in a clean catch manner, centrifuged at 1000 g for 15 minutes and stored at -7OC prior to assay. The samples were used directly after thawing and then discarded. Parameters for sample clarification such as centrifugation and filtration and addition of bacteriostatic agents were studied for I L l p and TNFa assays. A study was carried out on the influence of temperature and storage time on cytokine stability in urine. Cytokine values were expressed both as the urinary cytokine levels or as the cytokindcreatinine ratio. No difference was noted in the results regardless of the form in which they were expressed. The possibility of protein-cytokine interactions in urine was assessed by diluting the urine and evaluating the linearity of the dilution curve. No problems were noted for the cytokines studied here. Statistical analysis. The data was analyzed using a nonparametric method (Kruskal-Wallis test).The statistical program Epistat was used for the analysis. The basic research designs employed were as follows: a) Evaluation of the mean % SEM for each cytokine level between the normal, bacterial cystitis, and microhematuria groups. Each group was compared using the Kruskal-Wallis test followed by a pairwise analysis using Newman-Keuls test. b) Evaluation of the mean 2 SEM for each cytokine level in normal, patients with bacterial cystitis of low (1,000-50,OOOorganisms) and high count (>100,000 organisms). Each group was compared using the Kruskal-Wallis test followed by a pairwise analysis using Newman-Keuls test. c) The urine culture (>100,000 organisms) was used as the gold standard for ”true” infection. The definitions used for calculations of specificity and sensitivity were as follows: True positive: urine culture and cytokine assay positive; true negative: urine culture and cytokine assay negative; false positive: urine culture negative and cytokine assay positive; and false negative: urine culture positive and cytokine assay negative. Continuous sensitivityspecificity values were derived at various cutoff levels for each cytokine, a method which is preferred over a fixed cutoff for a normal range, particularly when multiple parameters are being studied. Assay specificity in the cytokine assays. Since I L l a is structurally related to IG1/3, the ligand directed specificity of the monaclonal antibodies used for detection of these two cytokines was investigated. No overlap in detection was noted suggesting that the monoclonal antibodies utilized were highly restricted to each antigen. RESULTS

Cytokine evalwtion in bacterial cystitis. The cytokine levels in normal, bacterial cystitis and microhematuria groups are shown in fig. 1,A-E. Table 2 shows the means 2 SEM for each cytokine in the three different groups. This table also shows the statistical analysis for the difference among the three groups using a non-parametric test (Kruskal-Wallis) followed by a pairwise comparison for each group, carried out for each cytokine, IGlP (p = 0.0005),ILla (p = 0.005), T N F a (p = 0.05) and IL-6 (p = 0.01) are significantly different among normal, bacterial cystitis, and microhematuria groups, using the Kruskal-Wallis test. I L 4 is not significantly different among the three groups. Pairwise testing allows us to assess which cytokine is best at distinguishing the different pairs using Newman-Keuls test. At a p value of 0.0001, IL6, IG1/3, TNFa and &la were found to be different between normal and bacterial cystitis groups. Comparison of the normal to microhematuria groups show that at a p of lo-’, only IG6 and TNFa show a significant difference between these groups, being elevated in the microhematuria group. The main finding is that ILlfi, &la,

MULTIPLE URINARY CYTOKINE LEVELS

1982

I

I

N

N

B

C

H

B

N

C

H

B

C

H

FIG.1. Urinary cytokine levels in normal (N), patients with bacterial cystitis (BC)of both high (7100,000 organisms) and low count (1,OOO50,000 organisms), and microhematuria (H). Values are grouped according to clinical status and represent the ratio of cytokine to creatinine. TABLE2. Summary

of

statistics of normal

(N),patients with bacterial cystitis (BC)of both low (1,O0050,000 organisms) and high count (H)

(>1OO,OOO organisms) and microhematuria group

CytokindCreatinine Cytokine

IGlg

&la

TNFa

IL6

IL4

Mean ? SEM (pglmg) 0.04 2 0.01 1.67 2 0.28 0.03 C 0.01 0.95 2 0.37 0.56 2 0.07 Normal (n = 36) 0.04 2 0.00 10.50 2 1.39 4.86 2 1.43 19.90 2 5.41 53.76 C 11.74 Bacterial cystitis In = 49) 0.04 2 0.01 5.38 2 1.50 0.42 2 0.14 0.61 2 0.53 0.27 C 0.08 hiicrohematuria (n = 11) Kruskal-Wallis (pvalue)* 0.0005 0.005 0.05 0.01 Not significant Newman-Keuls (N, BC) (N, BC) None (N, BC) (N, BC) significantly (N, H) (BC, H) (N, H) (BC, H) Werent pairs * The Kruskal-Wallis test was used to assess if the normal, bacterial cystitis and microhematuria groups were significantly different. Each cytokine was analyzed separately for all three groups. The pairwiise analysis followed by the Kruskal-Wallis test was done using the Newman-Keuls test. The mediana (pg/mg) for each cytokine in the diRerent p u p s were I L l p (0.42). IL-la (0.13). TNFa (O.M)), IL-6 (1.17). IL-4 (0.03) for normal group; I L l S (14.91). IGla (2.10). TNFa (0.59). IL6 (7.72), IL-4 (0.04) for bacterial cystitis group; I L l p (0.27),ILla (0.08). TNFa (0.27), IL6 (3.24), 1L4 (0.03) for miemhematuna p u p .

TNFa, and IL-6 can distinguish normals from the bacterial this study, we assessed sensitivitylspecificity pairs at differcystitis group, while only 1L-6 and T N F a can distinguish the ent cutoff values for the normal range for each cytokine. The normal and microhematuria groups with high level of significance (p = Cytokine evaluation in the low (1,00040,000 organisms) and high count (>100,000organisms) bacterial cystitis groups. A similar analysis was carried out between normal, low and high count bacterial cystitis groups for all the cytokines. The results are shown in fig. 2, A-E and table 3. The cytokines which show significant elevations for these three groups at a p of 0.05, were IL1& I L l a , TNFa and IG6. IL-4 showed no difference among the three groups. Several cytokines (ILlS, IL-la, TNFa and IL-6) were significantly elevated in both low and high count bacterial cystitis groups when compared to the normal group. There were significant differences between the normal and low count as well as high count bacterial cystitis groups for the following cytokines: T N F a (p = lo-'), I L 6 (p = lo-'), I L l p (p = and I L l a (p = 0.05). None of the cytokines could distinguish between the low count and high count bacterial cystitis groups. Specificity and sensitivity tradeoffsfor the cytokines analysis. Increasing the sensitivity of an assay occurs a t the expense of specificity. Since multiple markers are involved in

analysis was carried out for all cytokines and is shown in detail in fig.3 for IL-lp. The results allow us to assess the highest level of sensitivity which can be gained for any assay without compromising specificity. An optimal cutoff value for I G l p is 0.5 pg/mg, where a sensitivity of 85% can be gained at a specificity level of 90%. The data on patient based specificity and sensitivity is not meant to provide a definitive evaluation of the application of these assays t o a clinical setting. Rather, a provisional assessment of possible cutoffs for "normality" for each cytokine and its implication for the tradeoff between specificity and sensitivity is the issue. This approach is referred to as receiver operating characteristic (ROC) analysis and is frequently used to assess operational features of laboratory analysis. Pyuria us. cytokine elevation. The correlation of cytokine elevation in the pyuria is shown in table 4. A simple hypothesis would be that cytokines are merely more elevated and expensive measures of pyuria. This is clearly not the case. Cytokines may be elevated in the absence of pyuria. This is observed for TNFa and JL-6. By contrast, IL-lp is predominantly elevated where leukocytes are present (table 4).

MULTIPLE URINARY CYTOKINE LEVELS

N

Low

N

Low

High

N

Low

N

High

Low

1983

N

€Ugh

Low

High

High

FIG. 2. Urinary cytokine levels in normal (N), low count (Low), and high count (High) bacterial cystitis. Values are grouped according to clinical status and represent the ratio of cytokine to creatinine.

CytakindCreatinine Cytokine ~

ILlS

TNFa

&la

IL-4

I M

~

Mean 2 SEM (pg/mg) 1.67 ? 0.28 0.04 2 0.01 0.03 2 0.01 0.95 2 0.37 0.56 -2 0.07 Normal (n = 36) 10.21 2 2.59 0.06 -c_ 0.01 4.87 2 2.90 35.04 2 13.46 35.60 2 16.13 Low count (n = 15) 10.62 '+ 1.66 0.04 2 0.01 4.86 2 1.65 13.67 lr 4.84 60.58 2 15.03 High count (n = 34) Kruskal-Wallis (D value)* 0.0001 0.05 o.Ooo1 o.Ooo1 Not significant Newman-Keuls (N,High) (N,High) (N, High) None (N, High) simificantlv (N,Low) (N, LOW) (N, Law) (N,Law) different pairs * The Kruskal-Wallis test was used to assess if the normal, low and high count bacterial cystitia groups were significantly Werent. Each +kine was ~ using the Newmaa-Keuls test. analyzed separately for all three groupe. The pairwise analysis followed by the Kruskal-Wallis test w a done The medians (pg/mg) for each cytokine in the Werent groups were I L l p (0.42).I L l a (0.131, TNFa (O.oO), IG6(1.17), IG4 (0.03)for normal group; I L l p (11.27), IL-la (2.01),TNFa (0.86),IL-6 (6.14),IG4 (0.04) for low count bacterial cystitis group; I L l p (25.32). L l a (2.10), TNFa (0.58), IL-6 (7.92), IL4(0.04) for high count bacterial cystitis.

DISCUSSION

Of particular interest is the previous finding that IL-1P is not elevated in IC but is elevated in bacterial cystitis.' Since IL-la is elevated along with IGlP in bacterial cystitis, a similar finding would be expected in IC and studies to test this hypothesis are in progress. In contrast to IL-lP, IL-6 is reported to be elevated in IC2O and is elevated in microhematuria in the absence of bacterial infection. These results imply that IL-6 may be elevated by multiple mechanisms in a variety of urologic diseases. By contrast histamine and tryptase are elevated in IC but not in bacterial cystitis. " F a and IL-2 have been shown to be markedly elevated in bladder cancer patients and may be further increased after BCG therapy. The unique feature of our study is that we assessed multiple relevant urinary cytokines simultaneously in patients with low and high count bacterial cystitis, as well as patients with microhematuria compared to normals. Among the cytokines studied IL-lP, TNFa and IL-la and IL-6 are differentially elevated in the bacterial cystitis group but not in normals. The predominant pattern noted is a simultaneous elevation in several cytokines, including IL-lP, IL-la, TNFa and IL-6, but not in IL-4. Overall, the results imply that a strong inflammatory response accompanies bacterial cystitis involving multiple cytokines, possibly derived from multiple cell types. The fact that TNFa is elevated along with IL-lp implies a potent inflammatory response involving

A i70

A 60

i

40

v)

20

lo> 00

2

4

6

8

b"

10

IL-1p I ueatinine (pglrng)

FIG.3. Sensitivity and specificity tradeoff curve for IL-lP. Sensitivities and specificities for IL-18 were calculated at multiple cutoff values of I L l p as indicated in the Materials and Methods section.

macrophages andlor neutrophils. While the correlation of pyuria with cytokine elevation was not the primary goal of this study, pyuria was generally associated with higher levels of IG1& TNFa and IL-la. However elevated cytokines

1984

MULTIPLE URINARY CYTOKINE LEVELS

TALSLE 4. Correlation ofpyuria with cytokine elevation 'k of Elevated Cybkine*

IL1p

ILln

TNFn

IL6

WBC

Presence 89.6 60.7 72.4 82.1 Absence 33.3 16.7 50 58.3 p value" O.OOO9 0.027 0.31 0.232 * 'k of patients wrth bacterial cystitis of low and high count who showed elevation of cytokine. ** Chi-square test.

were noted in bacterial cystitis even in the absence of pyuria. Thus the relationship of pyuria to cytokine elevation is complex and requires further study. Larger study groups would be needed to stratify patients by leukocyte level. Thus the precise correlation between leukocytes and cytokines remains to be delineated. IL-6 elevation is observed in the majority of patients with elevated I L l p . Whether this represents solely leukocyte production of IL-6 or release from bladder epithelial cells as well is unclear. The elevation of I L l a is surprising given the prevailing view that this is a membrane bound cytokine. The release of IL-la suggests that the membrane bound form of I L l a may be further modified prior to release into urine. Defining a significant bacterial count is a difficult goal. The number widely used to define a clinically significant infection (>100,000 organisms) is a n arbitrary one. Our results suggest that there is a significant inflammatory component in low count bacterial infections in that several cytokines (IL18, I L l a , T N F a and IL-6) are elevated in low count bacterial infections relative to normal subjects. These observations are significant in that they imply that dismissal of low count infections as insignificant without assessing the inflammatory response may not be judicious. These results also confirm our previous findings' concerning IL-lp which suggested that low count infections with certain gram negative strains (particularly Pseudomonas and Klebsiella) can elicit remarkably strong inflammatory responses. Also certain strains of E. coli are considerably more potent at provoking inflammation than others. Thus, bacterial count alone is a poor predictor of severity of the associated inflammatory responses. Symptomatic patients would be treated regardless of bacterial count. However, the cytokine assays may provide some utility in assessing inflammation in non-routine situations such as very low bacterial counts in asymptomatic individuals. Two questions approached in our studies are whether microhematuria without infection or pyuria is associated with an inflammatory response and whether the cytokines involved are similar or different from those seen in bacterial cystitis. The results are somewhat surprising. Neither I L l p nor I L l a levels are significantly elevated in the microhemat u n a group relative to normal group. By contrast, I L 6 and T N F a are elevated in microhematuria. The underlying pathology which contributes to the elevation in IL6 and T N F a in the absence of a bacterial cystitis or pyuria is unclear. Studies are in progress to establish a link between these cytokines and the diseases which present clinically with microhematuria including certain genitourinary malignancies, stone disease, and medical renal diseases among others. Whatever the underlying pathology, it is clear that the pattern of cytokine elevation in microhematuria is different from that in bacterial cystitis. Our results indicate that both the low and high count bacterial cystitis are associated with significant inflammation since I L l p , I L l a , TNFa and IL-6 are elevated compared to normals. The pattern of cytokine elevation in low count bacterial cystitis is similar to that in high count bacterial cystitis. Furthermore, IL-6 and TNFa are significantly elevated in patients with microhematuria.

Other cytokines, in addition to the cytokines studied here, are also of potential interest. These include IL-1 receptor antagonist and I L 1 receptor. In particular, IL-1 receptor antagonist may in part explain why inflammation is pronounced in some individuals relative to others. Also the high degree of inflammation may be associated with IL-1 receptor shedding. These hypotheses need to be evaluated. Elevations in particular cytokines may serve to distinguish I bladder disease and are certainly useful in defining the in- , flammatory response. Thus, cytokine assays may have a 1 promising future in the characterization of bladder diseases. REFERENCES

1. Martins, S. M., Darlin, D. J., Lad, P. M. and Zimmern, P. E.: Interleukin-lp: A clinically relevant urinary marker. J . Urol., 151: 1198, 1994. 2. 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. 3, El-Mansoury, M., Boucher, W., Sant, G. R. and Theoharides, T. C.: Increased urine histamine and methylhistamine in interstitial cystitis. J . Urol., 152 350, 1994. 4. Boucher, W., El-Mansoury, M., Pang, X., Sant, G . R. and Theoharides, T. C.: Elevated mast cell tryptase in the urine of patients with interstitial cystitis. Br. J. Urol., 7 6 94, 1995. 5. De Boer,E. C., De Jong, W. H., Steerenberg, P. A., Aarden, L. A., Tetteroo, E., De Groot, E. R., Van der Meijden, A. P. M., Vegt, P. D. J., Debruyne, F. M. J. and Ruitenberg, E. J.: Induction of urinary interleukin-1 (IL-1). IL-2, IL-6, and tumor necrosis factor during intravesical immunotherapy with bacillus Calmette-Guerin in superficial bladder cancer. Cancer Immunol. Immunother., 34: 306, 1992. 6. Jackson, A. M., Alexandroff, A. B., Prescott, S. and James, K.: Production of urinary tumor necrosis factors and soluble tumor necrosis factor receptors in bladder cancer patients aRer bacillus Calmette-Guerin immunotherapy. Cancer Immunol. Immunother., 40:119, 1995. 7. Schamhart,D. H. J., Kurth, K. H., de Reijke, T. M. and Vleeming, R.: BCG treatment and the importance of a n inflammatory response. Urol. Res., 2 0 199, 1992. 8. Bohle, A,, Nowc, C., Ulmer, A. J., Musehold, J., Gerdes, J., Hofstetter, A. G. and Flad, H. D.: Elevations of cytokmes interleukin-1, interleukin-2 and tumor necrosis factor in the urine of patients after intravesical Bacillus Calmette-Guerin immunotherapy. J. Urol., 144: 59, 1990. 9. Ha&, E. O., Catalona, W. J. and Ratliff, T. L.: Detection of interleukin-2 in the urine of patients with superficial bladder tumors after treatment with intravesical BCG. J. Urol., 136 970,1986. 10. Haraoka, M., Senoh, K, Ogata, N., Furukawa, M., Matusumoto, T. and Kumazawa, J.: Elevated interleukin-8 levels in the urine of children with renal scarring and/or vesicoureteral reflux. J. Urol., 155 678, 1996. 11. Ratliff, T. L., Hudson, M. A. and Catalona, W. J.: Strategy for improving therapy of superficial bladder cancer. World J. Urol., 9 95, 1991. 12. De Reijke, T. M., De Boer, E. C., Kurth, K H. and Schamhart, D. H. J.: Urinary cytokine during intravesical Bacillus Calmette-Guerin therapy for superficial bladder cancer: processing, stability and prognostic value. J. Urol., 155: 477, 19%. 13. Fleischman, J. D., Toossi, Z., Ellner, J. J., Wentworth, D. B., Ratliff, T. L. and Imbembo, A. L.: Urinary interleukins in patients receiving intravesical Bacillus Calmette-Guerin therapy for superficial bladder cancer. Cancer, 64: 1447, 1989. 14. Hedges, S., Stenqvist, K., Lidin-Janson, G., Martinell, J., Sandberg, T. and Svanborg, C.: Comparison of urine and serum concentrations of interleukin-6 in women with acute pyelonephritis or asymptomatic bacteriuria. J. Infect. Dis., 166: 653,1992. 15. Seguchi, T., Yokokawa, K., Sugao, H., Nakano, E., Sonoda, T. and Okuyama, T.: Interleukin-6 activity in urine and serum in patients with bladder carcinoma. J. Urol., 148: 791, 1992. 16. v a n Oers, M. H. J., Van Der Heyden, A. A. P. A. M. and Aarden, L. A.: Interleukin 6 1IL61 in serum and urine of renal transplant recipients. Clin. Exp. Immunol., 71: 314, 1988.

MULTIPLE URINARY CYTOKINE LEVELS

1986

on bladder cancer cell lines: evidence for the role of L l a . 17. Hedges, S., Anderson, P., LidinJanson, G., De Man, P. and Svanborg, C.: Interleukin-6 response to deliberate colonization Immunol. Lett., 10:117, 1994. of the human urinary tract with gramnegative bacteria. In- 24. Hayashi, 0.. h h i , M., % h e , M., Hanezawa. IL and fect. Immun., 6 9 421,1991. Kitagawa, R.: Detection of interleubin-1 activity in human bladder cancer cell lines. J. Urol., 16k 760,1994. 18. De Man, P., Van h t e n , C., Aarden, L.. Engbeq, I., Linder, H. and Svanborg Eden, C.: Interleukin-6 induced at mucosal sur- 26. Nicolle, L E., BRmka, J., Orr, P., Wilkins, J. and Harding, G. K M.: Urinary immunoreactive interleubin-la and faces by gram-negative bacterial infection. Infect.Immun., 67: interleukin-6 in bacteriuric institutionalized elderly mbjd. 3383, 1989. J. Urol., 149: 1049,1993. 19. Esuvaranathan, 11,Alexandroff, A B., McIntyre, M.,Jackson, A. M., Prescott, S., Chisholm, G. D. and James, K: 26. Balbay, D., Ozen, H., ozkardea,H., Barut, A,Bahhsloglu, M., Tasar, C. and Remzi. D.: Detection of urinary interleukin-2. Interleukin-6 production by bladder tumors is upregulated by interleukin-2 receptor, and tumor neemsia factor levels in BCG immunotherapy. J. Urol., 154: 572, 1995. patients witb superficial bladder tumors after intravesical 20. Lotz, M., Villiger, P., Hugli, T., Koziol, J. and Zuraw, B. L.: BCG immunotherapy. Urology, 43:187,1994. Interleukin-6 and interstitial cystitis. J. Urol., 162 869,1994. 21. Vilcek, J. and Le,J.: Immunology of Cytokines: An Introduction. 27. Grassi, J.. Roberge, C. J., h b e r t , Y.,Pradelles, P. and Poubelle, P. E.: Determination of IL.1 alpha, IL1 beta and IL2 in biologIn: The Cytokine Handbook. Edited by A. Thornson. London: ical media using specificity enzyme immunometric assays. Academic Press, chapt. 1, pp. 1-20, 1995. Immunol. Rev., 119: 125, 1991. 22. Dinarello, C. A.: The interleukin-1 family: 10 years of discovery. 28. Kimbal, E. S.,Pidceral, S. F., Oppenheim, J. J. and -0, FASEB. J., 8: 1314,1994. J. L:Interleukin 1activity in normal human urine. J. Immu23. Alexandroff, A. B., Jackson, k M., Esuvaranathan. K, Preseott, nol., 13s:256,1984. S.and James, R: Autocrine regulation of ICAM-1 expression