Assess associations between cultured urinary microbiota and lower urinary tract symptoms profiles (symptotypes) in ambulatory, adult women with acute, uncomplicated UTI symptoms.

This is a subanalysis of clinical and urine culture data obtained in an IRB-approved clinical trial that included adult women presenting for UTI symptoms evaluation. Women who responded “yes” to “do you feel you have a UTI” and agreed to urethral catheterization were included in the study. Women currently on antibiotics, pregnant, had an indwelling catheter, performing intermittent self-catheterization, or whose clinician deemed she needed empiric treatment at enrollment were excluded. Following verbal and written consent, participants answered the validated Urinary Tract Infection Symptom Assessment (UTISA) questionnaire to describe symptoms at the time of the visit. Participants then contributed catheterized urine specimens, which we assessed by expanded quantitative urine culture (EQUC). Principal component analysis (PCA) of UTISA scores clustered participants into symptotypes based on severity (sev) and degree of bother (both) of UTI symptoms and subsequent biplot analysis identified symptoms responsible for variation. We then compared EQUC results between symptotypes. These analyses included all EQUC results, including negative and non-uropathogen cultures.

Of the 225 participants recruited, the average age was 66 years. PCA of UTISA scores resulted in 6 distinct symptotypes (Figure): A (N=87), B (N= 42); C (N=34), D (N=26), E (N=23) and F (N=37). Biplot analysis revealed UTISA components responsible for variation between symptotypes: urgency severity/degree of bother and incomplete emptying severity/degree of bother, abdominal pain severity, frequency severity/degree of bother and overall symptom severity, and dysuria severity/degree of bother. Some symptotypes (A, C, E) complained of severe, mild, or moderate symptom severity/degree of bother for each UTISA category. The other symptotypes (B, D, F) had increased severity/degree of bother for frequency and urgency, incomplete emptying, and frequency, respectively, compared to other categories. Comparison of EQUC results across symptotypes showed that no single microbe was unique to or always present within samples grouped by a single symptotype. EQUC detected Escherichia coli in 48% of participants (N=110) and detected E. coli monocultures in only 13% of participants (N=29). Interestingly, EQUC detected a similar prevalence of E. coli in all symptotypes (p=0.31); however, the frequency of some non-E.coli microbiota, including uropathogens and non-uropathogens, varied by symptotype.

These data suggest that the current definition of UTI may encompass multiple, distinct disease states, and that E. coli alone may not be the underlying cause of those states or any individual UTI symptoms. Instead, the structure of the overall microbial community, including the presence of non-E. coli uropathogens and non-uropathogenic organisms, may contribute to development certain UTI symptoms.

This UTI population contained 6 distinct symptotypes characterized by severity and degree of bother of certain UTISA components. The structure of the overall microbial community appears to differ across symptotypes; however, it is possible that host genetic differences may account for this difference in UTI symptoms. Overall, results of these analyses suggest that the composition of the urinary microbiome may play an important role in UTI pathology.