A prospective observational study of urinary cytokines and inflammatory response in patients with Overactive Bladder Syndrome

Overactive bladder syndrome (OAB) is characterized by urinary urgency, with or without urgency urinary incontinence, usually with increased daytime frequency and nocturia, if there is no proven infection or other obvious pathology [1]. The diagnosis is contingent on exclusion of urinary tract infection (UTI), usually achieved by bedside urinary dipstick testing and/or midstream culture analyses using Kass criterion for exclusion of infection [2]. Recently, dipstick and culture testing have been discredited due to poor sensitivity and ambiguous and misleading culture data [3, 4]. This raises the possibility that the OAB may need to be revaluated.

We have published data from fresh urine microscopy [5], routine culture and spun urinary sediment culture [3, 6]. We reported significant differences in the urinary white cell and urothelial cell counts and in the results of spun sediment culture between OAB patients and normal controls, over a period of 12-months. Like others [4, 7], we found that the dipsticks and routine culture failed to discriminate between the OAB patients and normal controls. However, other surrogate markers of infection and inflammation: Spun sediment cultures, microscopic pyuria (WBC), and urinary uroepithelial cell (EPC) counts, revealed significant between group differences [5]. These data suggest that standard screening for UTI in the assessment of patients with OAB may be misleading when excluding infection.

Given the implications of these findings, we set out to validate our observations by comparing those three surrogate infection markers against the behaviour of two key urinary cytokines namely IL6 and Lactoferrin, in patients with OAB. We have already published on ATP in this context [8]. We chose IL-6 because it is a well-established inflammatory marker, and it has been reported to rise in association with increased bacterial loads in UTI. Lactoferrin acts to sequester iron and thereby deprive offending microbes of vital nutrition. Increases in urinary lactoferrin have been reported in association with UTI.

To explore the relationship between urinary cytokines and key urinary markers of bacterial cystitis; pyuria [9], and urinary spun sediment culture [10, 11] in women with OAB symptoms comparing them to normal controls, over a period of twelve months.

We conducted a prospective, blinded, observational cohort study of female outpatients with OAB symptoms, and asymptomatic control subjects. The study groups included patients who described OAB symptoms as defined by the ICS definition [12], including urinary urgency, increased day time frequency, nocturia with or without urge incontinence. Healthy female adults matched for age and menopausal status, with no urinary symptoms were recruited as controls. All participants provided written, informed consent. Women who were pregnant or planning a pregnancy were not eligible for inclusion. We did not include patients with structural disease of the urinary tract and other systemic diseases.

Study participants attended two monthly study visits over a one year. Patients and control subjects provided midstream urine (MSU) samples using the clean-catch method. All samples were subject to analysis by blinded researchers. We assessed the patient’s symptoms, urinary cytology, microbiology and urinary immune response. Symptoms were assessed using standardised validated questionnaires including the ICIQ-FLUTS, a pain score and an urgency score [6]. Aliquots of spun urine were frozen at − 80 °C for cytokine analysis. Methods are discussed in detail below.

The methods of the assessment of pyuria and the microbiological assessment using the spun urinary sediment culture have been reported [3, 4, 7, 9, 10]. We used unspun fresh urine microscopy to obtain pyuria. Enhanced urinary sediment cultures were performed for quantitative and qualitative microbial analysis [3]. All specimens were subject to urinary dipsticks analysis and routine NHS culture with or without sensitivity where a positive culture was declared if ≥ 10⁵ cfu/ml. of a pure culture of a known urinary pathogen.

Fresh urine samples were centrifuged and the supernatant was removed and aliquots frozen at -80 °C. For each cytokine analysis a separate aliquot was used and the urine samples only underwent one freeze / thaw cycle to ensure cytokine stability.

The results of the MSU cultures and dipsticks tests are reported in Tables 1 and 2. There were significant differences between the OAB patients and the normal controls in that the OAB patients were more likely to demonstrate a positive test.

The linear mixed-effects models procedure, used to analyse the longitudinal data, explored the relationship between urinary IL-6 / Lactoferrin and urgency score, pain score, LUTS score, pyuria and microbial growth on enhanced sediment culture. In the first model the independent covariate was the group (0 or 1 / control or patient), the dependant variable IL-6 / Lactoferrin and the repeated variables were indexed on visit number.

A second model examined the pooled data from 144 OAB patient visits and 132 control visits.

The dependant variable was log IL-6 / Lactoferrin and the independent covariates were the group number (0/1; controls/patients), urgency score, pain score, LUTS score, log pyuria and log total microbial growth.

There was a significant difference in urinary IL-6 between patients and controls (Table 3) and similar differences were found with Lactoferrin Table 4.

Table 5 shows the results of the mixed model repeated measures analysis applied to IL-6; it can be seen that pyuria count (log₁₀ wbc µl⁻¹) and however total bacterial growth (log₁₀ CFU ml⁻¹) proved a significant predictor of IL-6 however the symptoms scores did not.

Mean urinary IL-6 was found to be significantly higher in patients when compared with controls across each visit (Fig. 1).

The linear mixed-effects models procedure was similarly used to explore the relationship between urinary Lactoferrin and urgency score, pain score, LUTS score, pyuria and microbial growth on enhanced sediment culture. In the first model the independent covariate was the group (0 or 1 / control or patient), the dependant variable Lactoferrin, and the repeated variables were indexed on visit number. There was a significant difference in urinary Lactoferrin between patients and controls (Table 4).

A multiple mixed-effects model analysis was performed where the dependant variable was log Lactoferrin and the independent covariates were the group number (0/1), LUTS score, urgency score, pain score, log total microbial growth and log pyuria. Table 6 shows the results for each of these parameters. The predictors of Lactoferrin proved to be the LUTS symptoms, urgency and pyuria count (log₁₀ wbc µl⁻¹) however total bacterial growth (log₁₀ CFU ml⁻¹) did not.

Mean urinary Lactoferrin was found to be significantly higher in the patient group when compared with controls across each visit (Fig. 2).

This prospective, blinded comparative study addressed the role of inflammation in the urine of patients with overactive bladder symptoms, expressed as an increase in urinary cytokines. We also measured bacterial colonisation, and markers of infection, by pathogens in this group. The data demonstrate evidence of urinary tract inflammation and this was associated with increased evidence of microbial infection in women with OAB syndrome.

These data should be interpreted with caution due to the following limitations. We followed patients over a year whilst they were attending a centre for treatment. We made no attempt to measure the effect of treatment, whether by antimuscarinics or antibiotics; the study did not have that scope. Our purpose was to test for evidence of inflammation, ascertain whether there were any correlations with markers of infection and whether the findings were consistent and persisted over a twelve month period. Treatment effects in similar patients have been studied elsewhere [18, 19] but RCT data are still outstanding. In the context of this study; intervention studies that examine treatment effects on the variables we scrutinised are still necessary. Our measures are indirect markers of inflammation, we did not study the bladder histology or cystoscopic appearances. However, these patients had undergone normal cystoscopy and renal tract ultrasound prior to their diagnosis of OAB. The measures that we used to diagnose UTI -microscopy of fresh unspun urine and sediment culture are well validated. We emphasise that the gold standard criteria for the diagnosis or exclusion of UTI using MSU culture has been shown to be incompetent at this function. Enhanced culture studies and 16 s-rRNA genomics have shown that Kass’ key assumption of a sterile normal bladder was incorrect and that his culture methods are overlooking a substantial number of infections [3, 10]. Determining causation is impossible given such a diverse microbiome with extensive overlap between patients and controls [3, 10, 20] thus we are limited to observations of association.

There is a growing body of evidence, from cross-sectional data, that bacterial infection may play a role in LUTS and OAB [6]. In this study, the patient group and controls were matched for age, BMI and menopause status and given the sample size, statistical power was maintained throughout. Inflammatory disease process in OAB patients compared to controls was clearly evident when we compared the microscopic pyuria, urinary IL-6 and Lactoferrin between the two groups. On the other hand urinary dipstick tests and MSU cultures, which are known to be insensitive [3], nevertheless differed between the two study groups. We noted a significant between groups difference in microbial abundance when we used a validated enhanced culture [10].

Urinary IL-6 association with pyuria and microbial growth did not come as a surprise; IL-6 acts as a pro-inflammatory agent attracting white cells into the tissues [21]. Increased urinary IL-6 secretion has been demonstrated in prospective studies of acute UTI [22]. Bacteria will trigger an innate immune response by parasitising uroepithelial cells. Like other mucosal surfaces, the lining of the urinary tract carries receptors capable of recognising invading microbes by binding to receptors to “Pathogen associated molecular patterns” (PAMPS). Of the various of these immune surveillance molecules the Toll-like receptor (TLR) family is the best characterised through studies of bacterial adherence and TLR4 signalling. In vivo studies, of a bacterial challenge to urothelial cells, have shown a rapid cytokine response with production of interleukin-1beta (IL-1beta), IL-6, and IL-8 (CXCL8). Type 1-fimbriated E. coli were found to activate IL-6 and IL-8 production more efficiently than the non-fimbriated[23]. IL-6 has been detected in the urine of a majority of patients with positive cultures, but it was only detected in the serum of symptomatic patients [24]. It has been argued that the cytokine response during UTI could have local and systemic components; epithelial cells are thought to be responsible for the local cytokine reaction [25]. Studies have shown that IL-6 production increases neutrophil migration and activation of IL8 [26, 27]. Uroepithelial cells also respond, not just to bacteria, but they can be stimulated by cytokines: IL-1alpha and TNFalpha induce secretion, by urothelial cells of IL-6 and IL-8 and the upregulation of mRNAs for IL-1alpha, IL-1beta, IL-6 and IL-8 [28]. There is a complex regulation of the production of IL-6 by membrane-bound and soluble receptors, which are modulated by numerous influences. Strains of Escherichia coli (E.coli) have developed strategies to evade immune response by suppressing cytokine reactions to infection [29].The studies discussed in this section have addressed acute UTI and its influence on urinary IL-6 secretion [22]; Our study takes a further step by implying a role for IL-6 in chronic disease.

Our data showed that urinary Lactoferrin was associated with pyuria and symptoms. Lactoferrin’s primary role is to sequester free iron, so it inconveniences bacteria by removing a substrate essential for bacterial growth [30]. Lactoferrin has other antimicrobial properties, a key one is by binding to the lipopolysaccharide (LPS) of the bacterial wall and then generating peroxides; these oxidise membrane molecules causing bacterial lysis [30]. Lactoferrin will stimulate phagocytes. It has been reported to interfere with proton translocation through the cell membrane, disrupting microbial chemistry and neutralising cell attachment capacity [31, 32]. Specifically, Lactoferrin interacts with membrane-LPS E. coli, removing LPS from the membrane [33]. The susceptibility of the bacteria to Lactoferrin is thought to be dependent on their growth phase since bacteria are more open to killing by Lactoferrin when in early phase [32]. Lactoferrin has also been shown to modulate the activity of known antibacterial agents such as lysozymes and antibiotics [34]. It is thought that the dominant mechanism of action is via its iron binding properties and interaction with LPS on Gram-negative bacteria [35].

The data from this study were collected by application of a blinded, prospective, comparative protocol. The analysis of various measures that reflect different perspectives is called consilience and it strengthens the validity of the observations because of their coherence. This work has demonstrated strong evidence of inflammation of the urothelium associated with OAB and there is good evidence that infection is playing a significant part. For many years UTI has been excluded in OAB patients by deploying tests that are insensitive. Our data present a compelling case for us to re-evaluate our understanding of OAB, detrusor instability and detrusor hyperreflexia.