Recurrent cystitis in women represents a heterogeneous group of conditions with complex pathophysiology involving interactions between the urogenital microbiome, local immunity, and neurogenic inflammation. Conventional diagnostic approaches based on urine culture fail to explain persistent symptoms in a substantial proportion of patients, leading to empirical antibiotic overuse and therapeutic failure. We hypothesized that distinct clinical phenotypes exist, characterized by specific patterns of urinary and vaginal microbiota composition, bladder-vaginal correlation, and urinary cytokine profiles. The aim of this study was to develop and validate a phenotypic classification system that integrates microbiological and immunological data to enable personalized therapeutic strategies.

A single-center cross-sectional study included women aged 18–65 years with recurrent cystitis: 65 with frequent episodes (≥3/year) and 115 with long-term remission (>6 months). Controls were healthy volunteers (n=48). Exclusion criteria: pregnancy, recent antibiotics, immunodeficiency, malignancy.

Paired urine and vaginal samples were analyzed by multiplex real-time PCR targeting 16S rRNA for 14 bacterial taxa. Urinary cytokines (MCP-1/CCL2, calprotectin, RANTES/CCL5, MIF, TNF-α, IL-6) were quantified by ELISA.

Bladder-vaginal correlation was assessed using Spearman’s r and Kendall’s τ. Phenotypes were derived by hierarchical cluster analysis based on microbiome composition, correlation patterns, and cytokine profiles.

Three distinct cystitis phenotypes were identified:

Phenotype A – “Acute cystitis following long-term remission” (n=115). Massive urinary overgrowth of Enterobacteriaceae (96%, p<0.001 vs controls). Urinary IL-6 (>100 pg/mL) and calprotectin (>500 ng/mL) markedly elevated; MCP-1 elevated as secondary signal (mean 238 pg/mL). Bladder-vaginal correlation low (r<0.2), indicating autonomous bladder infection. Vaginal microbiome: normocenosis in 44%, dysbiosis in 56%.

Phenotype B – “Persistent / rUTI” (n=65). Moderate Enterobacteriaceae elevation (83%, p<0.001) with significant enrichment of strict anaerobes Lachnobacterium/Clostridium spp. in urine (52%, p<0.001), exceeding vaginal levels (52% vs 24%, p<0.01). Chronic low-grade MCP-1 elevation (mean 82 pg/mL); IL-6 variable. Bladder-vaginal correlation low (r<0.2).

Phenotype C – “Anaerobic / Lymphocytic”. Dominance of bacterial vaginosis-associated microbiota in both biotopes: Gardnerella complex (86% urine, 92% vagina), Atopobium (70%, 83%), Megasphaera complex (66%, 76%), Sneathia/Fusobacterium (50%, 62%). High bladder-vaginal concordance (τ=0.51, p<0.01), reflecting impaired barrier function. Urinary MCP-1 significantly elevated (mean 258 pg/mL) with RANTES.

This study provides an integrated phenotypic classification of recurrent cystitis incorporating paired urinary and vaginal microbiome analysis with cytokine profiling. Three distinct phenotypes were identified with unique microbiome signatures, bladder-vaginal correlation patterns, and cytokine profiles.

Bladder-vaginal correlation is the key differential criterion: low correlation (r<0.2) indicates autonomous bladder colonization (Phenotypes A and B), while high correlation (τ>0.5) signifies barrier dysfunction with vaginal spillover (Phenotype C).

MCP-1 (CCL2) plays context-dependent roles: as a regenerative signal following acute inflammation (Phenotype A), a “monocyte trap” perpetuating bacterial persistence (Phenotype B), and a bridge between innate and adaptive immunity driving lymphocytic inflammation (Phenotype C). The high prevalence of strict anaerobes (Lachnobacterium/Clostridium) in urine of Phenotype B patients (52% detection, exceeding vaginal levels) supports intracellular bacterial persistence within macrophages (“Trojan horse” mechanism). Loss of unique urinary commensals (Tepidimonas, Flavobacterium, Aerococcus) was most pronounced in Phenotype C, indicating replacement of native bladder microbiota by vaginal anaerobes.

Recurrent cystitis comprises three distinct phenotypes:

Phenotype A (acute flare after long remission) – massive Enterobacteriaceae overgrowth, low bladder-vaginal correlation (r<0.2), IL-6/calprotectin dominance.

Phenotype B (persistent / rUTI) – moderate Enterobacteriaceae with Lachnobacterium/Clostridium enrichment in urine, low bladder-vaginal correlation (r<0.2), chronic low-grade MCP-1 elevation.

Phenotype C (anaerobic / lymphocytic) – dominance of bacterial vaginosis-associated microbiota in both biotopes, high bladder-vaginal correlation (τ>0.5), elevated MCP-1 with RANTES.

Bladder-vaginal correlation is the key differential criterion distinguishing autonomous bladder infection (Phenotypes A and B) from barrier dysfunction with vaginal spillover (Phenotype C). Integration of microbiome diagnostics and cytokine profiling into clinical practice has the potential to reduce antibiotic overuse and improve outcomes.

Nickel JC, Shoskes D, Irvine-Bird K. Clinical phenotyping of women with interstitial cystitis/painful bladder syndrome: a key to classification and potentially improved management. J Urol. 2009 Jul;182(1):155-60. doi: 10.1016/j.juro.2009.02.122.
Stapleton AE. The Vaginal Microbiota and Urinary Tract Infection. Microbiol Spectr. 2016 Dec;4(6):10.1128/microbiolspec.UTI-0025-2016. doi: 10.1128/microbiolspec.UTI-0025-2016.
Brubaker L, Wolfe AJ. The female urinary microbiota, urinary health and common urinary disorders. Nat Rev Urol. 2024;21(2):93-105. doi: 10.1038/s41585-023-00812-0.