Several studies have recently suggested that bacterial cystitis is a potential aetiological factor in the development of refractory detrusor overactive (refractory DO). One proposed mechanism for this has been release of inflammatory cytokines, from the bladder which then sensitise suburothelial afferent nerves, leading to enhanced signalling and urgency [1]. Funding was obtained to determine the concentration of urinary cytokines in a small proportion of women who were enrolled in a RCT of antibiotics in refractory DO. This study aimed to examine the relationship between antibiotic treatment and concentration of cytokines in urine from women with refractory DO.

Women with refractory DO were recruited to a double-blind, placebo controlled, randomized trial. Inclusion criteria were: age> 50 years, urodynamically proven DO that was refractory to treatment. Exclusion criteria were known kidney dysfunction or voiding dysfunction. Recruits underwent a 2.5 week washout period to establish a baseline (0 weeks), and were then randomised into 6 weeks of antibiotic or placebo (2:1 ratio). Antibiotic treatment comprised 2 weeks each of Norfloxacin, Augmentin Duo and Nitrofurantoin.  
Midstream urine (MSU) samples were collected at baseline (0 weeks), then at weeks 2, 4, and 6 (during antibiotic/ placebo treatment), and also at 1 month after the conclusion of the 6 week antibiotic (or placebo) treatment period and after 6 months. This represents a total of 6 MSU samples per patient. Half of the MSU sample was sent to the Hospital Microbiology department for culture. A diagnosis of UTI was determined by the presence of a single bacterial species (>106 CFU/L). “Any bacteriuria” was also reported, when more than one bacterial species was isolated, with or without epithelial cells, with or without pyuria. “Any bacteriuria” has been included in this study as culture independent methods have suggested that mixed growth of organisms may be a precursor to later UTI in refractory DO [2]. 
The remaining urine was centrifuged to remove exfoliated urothelial cells before being frozen at -80oC. Cytokines (IL-1α, IL-4, IL-6, IL-8, IL-10, CXCL10 (IP-10), TNF-α, IL-1 receptor antagonist and MCP-1) were measured using Elisa kits as directed by the manufacturer. These cytokines have been seen to be important components of the host response to bacterial cystitis [3]. As part of the interim analysis for futility testing, the scientist on our team, who works at a separate University, was able to access treatment codes thus the concentration of cytokines was compared between antibiotic and placebo groups (Mann Whitney test).

Thus far cytokines have been measured in 12 women undergoing antibiotic treatment (median age 65.5 years, range 53-79) and 7 women undertaking placebo (median age 64.5 years, range 57-80). In the antibiotic treated women, UTI was diagnosed in approximately 10% of MSU samples (Table 1). In contrast UTI was diagnosed in just over 20% of placebo treated women (Table 1). Any bacteriuria was more common than UTI in both antibiotic and placebo treated women (Table 1).
At each time point the concentration of IL-6, IL-8 and CXCL-10 was consistently, but not significantly, lower in the antibiotic treated women than in the placebo group (Figure 1A-C). There was no particular pattern in the concentrations of the remaining cytokines (IL-1α, IL-4, IL-10, TNF-α, IL-1 receptor antagonist and MCP-1) at the different time points in either the antibiotic or placebo treatment. 
In patients where bacteriuria was seen (combined UTI and any bacteriuria) the concentration of IL-6 (P 0.0003), IL-8 (P0.0035), CXCL-10 (P 0.0014) was significantly lower if the patient was in the antibiotic treatment group (Figure 1D-F). The same trend was observed in those with classical UTI (IL-6, P 0.045, IL-8 P 0.006, CXCL-10 P 0.004). There were no differences in the concentrations of the remaining cytokines (IL-1α, IL-4, IL-10, TNF-α, IL-1 receptor antagonist and MCP-1) in the antibiotic or placebo treatment in bacteriuric women.

The results from this study suggest that treatment with antibiotics not only reduces the incidence of bacteriuria (both UTI and “any bacteriuria”) but also reduces the concentration of cytokines released in response to bacteriuria. In the antibiotic treated patients, three of the nine cytokines measured were very significantly reduced. These three cytokines (IL-6, IL-8 and CXCL-10) are all associated with activation of the innate immune response. Interleukin 6 stimulates the production of neutrophils while IL8 and CXCL-10 are chemotactic for neutrophils and macrophages respectively.  We hypothesise that antibiotic therapy decreases the innate immune response, leading to decreased tissue damage in response to infection and less stimulation of afferent nerves and urgency.

Antibiotics may inhibit the cytokine release that occurs in response to bacteriuria. Research groups in three different centres have recently published that approximately half of patients with refractory DO have some form of recurrent bacteriuria.  Cytokines associated with innate immunity are reduced by antibiotic therapy, which may reduce the activation of afferent nerves and also inhibit the sensation of urgency. This has important implications for our understanding of the aetiology of refractory DO in women who do not respond to pharmacotherapy.

De Groat, W. C., & Yoshimura, N. (2009). Afferent nerve regulation of bladder function in health and disease. Handbook of experimental pharmacology, 194, 91-138.
Chen et al., (2018). The urinary microbiome in patients with refractory urge incontinence and recurrent urinary tract infection. International Urogynecology Journal, 29(12): 1775–1782
Sundac et al., (2016) Protein-based profiling of the immune response to uroprathogenic Escherichia coli in adult patients immediately following hospital admission for acute cystitis. Pathogens and Disease. 74. ftw062.