記住我
ORIGINAL ARTICLE
Year : 2023 | Volume
: 34
| Issue : 2 | Page : 93-98
Does bacterial motility influence the extent of bladder wall involvement in women with recurrent urinary tract infections?
Rahul S Patel1, Alana L Christie2, Philippe E Zimmern3
1 Division of Urology, University of Texas McGovern Medical School at Houston, Houston, Texas, USA
2 Simmons Comprehensive Cancer Center Biostatistics, Dallas, Texas, USA
3 Department of Urology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
Correspondence Address:
Philippe E Zimmern
University of Texas Southwestern Medical Center, Dallas, Texas
USA
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/UROS.UROS_74_22
Purpose: Because bacterial motility has been implicated as a potential virulence factor, we compared the motile and nonmotile properties of bacteria isolated from urine cultures in women with recurrent urinary tract infections (RUTI) to test if such motility properties are linked to the extent of bladder wall infection on cystoscopy. Materials and Methods: After IRB approval, a retrospective review of women with antibiotic-recalcitrant symptomatic RUTI (≥3 Urinary tract infections/year) who underwent electrofulguration (EF) for cystitis was conducted. Pre-operative office cystoscopy staged cystitis as stage 1: trigone (trigonitis), stage 2: stage 1 and bladder base, stage 3: stage 2 and one or both lateral walls, and stage 4: the whole bladder (pancystitis). Positive urine cultures before EF for the type of bacteria (motile vs. non-motile) were compared among stages. Results: From 2006 to 2020, 139 women with RUTI met the study criteria. The median age was 67 with 95% Caucasian. Those with stage 4 cystitis were older (P = 0.0009) and less sexually active (P = 0.038). Patients with a higher stage were associated with a higher number of motile organisms (P = 0.0056), with an increased presence of non-E. Coli motile organisms (P < 0.0001) such as Proteus (P = 0.0024), Pseudomonas (P = 0.0062), Enterobacter (P = 0.020) and Citrobacter (P = 0.0067). Conclusion: The comparison of bacterial motility in women with RUTI at four different stages of bladder surface involvement identified unique characteristics and bacterial strains associated with each stage. RUTI women with higher stages of cystitis had more unique types of motile urinary bacterial strains, possibly suggesting a role in the spread of cystitis beyond the trigone.
Keywords: Bacterial motility, chronic cystitis, postmenopausal women, recurrent urinary tract infections
Urinary tract infection (UTI) is a bothersome condition affecting women with an estimated annual incidence of 12.6% and a lifetime likelihood of approximately 60%.[1],[2] The annual costs for the evaluation and management of UTIs in the United States are approximately $2 billion.[3] The most widely accepted theory behind the pathogenesis of UTIs is that bacteria ascend the urethra - presumably with flagella motion or mechanical means (i.e., catheter) – through the bladder neck and attach to the base of the bladder (trigone) through pili.[4],[5] Thus, recurrent UTIs (RUTI) through an ascending mechanism would logically lead to a chronic infection of just the trigone – also termed trigonitis.[6] However, we have noticed that many RUTI patients have bladder wall involvement past the trigone on cystoscopy, with some having lesions involving the lateral walls of the bladder or even the entire bladder wall (pancystitis). This led to the development of a simple classification of cystitis in 4 stages, which has been useful to better characterize the severity of the infection process and its effect on RUTI management.
But why are some patients having visible lesions of chronic bladder wall infection manifesting beyond their trigone? Several explanations are plausible. First, chronic urinary retention with urine stagnating in the bladder could encourage the bathing of other bladder walls than the trigone with bacteria-laden urine. However, retention is not always related to UTI or UTI evolution.[7] Second, in women with a long history of UTI or very frequent episodes of UTI, there could be a higher propensity for planktonic bacteria to anchor at new bladder wall sites beyond the trigone. However, there are some women who have had UTIs for many years or decades, but only have chronic trigonitis on cystoscopy.[8],[9]
In the uropathogenic Escherichia coli (UPEC) model reported by Hultgren and colleagues, it has been shown that once UPEC attaches to the bladder using type 1 pili[5],[10] UPEC becomes internalized and rapidly replicates to form intracellular bacterial communities (IBCs).[11] These IBCs eventually disassociate following surface exfoliation of bladder epithelium. This process exposes the deeper bladder layers to bacterial attachment, thus encouraging the development of quiescent intracellular reservoirs (QIR) for further recurrent infections.[12],[13],[14] The presence of tissue-resident bacteria has now been confirmed in bladder biopsies of women with RUTIs.[15],[16] It is thought that the filamentous morphology used to flux out and invade neighboring cells is flagella-mediated.[17] However, unlike the swimming motility used to ascend the urethra (which requires a liquid medium, i.e., urine), this is thought to resemble a swarming motility used to move across surfaces, i.e., the bladder epithelium.[18] Some bacteria have both the ability to anchor themselves with pili or to express flagella to allow motility. What triggers the change from attachment/pili to motility/flagella is not known in the urinary environment.
To see if the more advanced stages of cystitis had more frequently associated bacteria with motile properties, we reviewed our dataset in women with antibiotic-recalcitrant RUTIs who elected to undergo fulguration of areas of chronic cystitis to improve their UTI condition.
Materials and MethodsAfter Institutional Review Board approval, (University of Texas Southwestern Medical Center IRB: STU 082010-016), we reviewed a prospectively maintained database of nonneurogenic women who underwent electrofulguration (EF) for antibiotic-recalcitrant RUTIs and had proven cystitis on office cystoscopy during their RUTI evaluation. According to a recent review on RUTI definition, RUTI was defined as at least three culture-proven symptomatic UTIs within one year (or 2 within 6 months).[19] All patients had completed an upper (renal ultrasound) and lower urinary tract (pelvic examination, noninvasive flow, post-void residual, standing voiding cystourethrogram, and office flexible cystoscopy) evaluation for uncomplicated RUTI and only had findings of cystitis on cystoscopy as a plausible explanation for their recurrent UTI pattern. Exclusion criteria included patients with catheter-related etiologies of UTI such as clean intermittent catheterization and indwelling catheters. Also excluded were those with no retrievable urine cultures within 3 months of their first EF.
Data acquisition was conducted by a neutral investigator not involved in the care of these patients and gathered from an electronic medical record (EPIC) and database. Demographic data included age, body mass index (BMI), race, parity, and past medical history including diabetic status, smoking status, menopausal status, prior hysterectomy, hormonal replacement status, and sexual activity.
All positive urine cultures (>105 colony forming units/ml) before the patient's first EF were reviewed. Based on the known properties of the infecting organism, strains were categorized into either motile or nonmotile groups.[20],[21],[22],[23],[24] Next, depending on the presence of motile and/or nonmotile strains, each patient was categorized as predominantly motile or nonmotile. In cases where the patient had a history of both motile and nonmotile urinary strains, the number of urine cultures was used to categorize. Therefore, if the patient had 50% or more motile urinary cultures, they were categorized as predominantly motile. In addition, the number of non-E. Coli motile organisms and the presence of more than one motile organism were collected.
Pre-EF office cystoscopy was used to categorize the patients based on the extent of bladder surface involvement. Pre-operative office cystoscopy staged cystitis as stage 1: trigone (trigonitis)[6], stage 2: stage 1 and bladder base, stage 3: stage 2 and one or both lateral walls (starting lateral to the ureteric orifice), and stage 4: the whole bladder (pancystitis) [Figure 1]. All cystoscopies were performed by a single FPMRS-trained physician and documented in the patient's chart with photographs [Figure 2].
Figure 2: Trigonitis with no lesions observed beyond the inter-ureteric ridge (stage 1) (a) Cystitis cystica lesions extending to bladder base (stage 2) (b)Descriptive statistics were provided as medians and interquartile ranges for continuous values and frequencies and percentages for categorical values. Overall association of cystitis stage with patient data was tested using Fisher's exact test for categorical values. We tested for ordinal associations between cystitis stage and categorical urine culture data using Mantel-Haenszel Chi-square tests. Differences in means by cystitis stage for the continuous patient and urine characteristics were tested using linear regression F-tests. Tukey-adjusted post hoc t-tests were used to determine which pairs of stages were significantly different than each other for significant results from the linear regression F-test. For factors that were found to be significantly associated with stage on univariate analysis, we tested if the factor remained significant when controlling for age using logistic regression. In the cases where an organism was not observed in all four stages, we used exact logistic regression to obtain model estimates. All analyses were completed at the 0.05 significance level using SAS 9.4 (SAS Institute Inc., Cary, NC, USA).
ResultsFrom 2006 to 2020, 139/230 patients met all study criteria. Of the 91 excluded, 84 had cultures done at outside institutions with different culture methodologies, thus precluding their inclusion in our dataset, and 7 had catheter-related RUTIs. Patient demographics are presented in [Table 1]. The majority of patients were Caucasian (95%) with a median age of 67 years and median BMI of 27.6, and 17% had a diagnosis of type 2 diabetes.
Also [Table 1] presents the breakdown of stages of bladder involvement. Of the total 139 patients, 57 (41%) were Stage 1, 16 (12%) Stage 2, 31 (22%) Stage 3, and 35 (25%) Stage 4. There was a significant difference in age (P = 0.0009) and parity (P = 0.019) by stage. Post hoc comparisons showed that those with Stage 1 were significantly younger than those with Stage 4 (P = 0.0008). Women with Stage 4 were significantly less sexually active (P = 0.038). There were no other significant differences in patient characteristics among the different stages of bladder involvement.
The specific bacterial strains found in each stage are presented in [Table 2]. The most common bacterial strain isolated in all stages was E. coli. Among the non-motile strains, the most commonly cultured bacterial strains were Klebsiella Pneumoniae and Enterococcus Faecalis. There was no significant difference in overall motility. Having a higher stage was associated with having multiple organisms (P = 0.039). The number of positive cultures was also associated with stage (P = 0.0002), with post hoc comparisons showing significant differences between Stage 1 and Stage 4 (P = 0.0008). Similar results were also seen when looking at the association between number of cultures with motile organisms (P = 0.0014) and number of cultures with non-motile organisms (P = 0.031). A higher number of motile organisms seen across cultures was significantly associated with a higher stage (P = 0.0056), with an increased presence of non- E. coli motile organisms (P < 0.0001). These included significantly more isolated strains of Proteus Mirabilis (P = 0.0024), Pseudomonas Aeruginosa (P = 0.0062), Enterobacter Cloacae (P = 0.020), and Citrobacter (P = 0.0067) in higher stages. Of the nonmotile organisms, patients with Klebsiella Pneumoniae had significantly higher stage (P = 0.0062).
After adjusting for age in logistic regression [Table 3], we found that there was no longer a significant association between stage and patients who have had multiple organisms, who have had motile and non-motile organisms, who have had Enterobacter infections, or who have had Citrobacter infections. Stage was still significantly associated with history of multiple motile organisms (P = 0.0027), presence of a non E. coli motile organism (P = 0.0005), presence of Proteus (P = 0.020), presence of Pseudomonas (P = 0.017), and presence of Klebsiella (P = 0.040). In all cases, Stage 4 versus Stage 1 showed a significant increase in odds of the presence of the given factor. For example, patients with Stage 4 had 25.8 times the odds of having a non E. coli motile organism as compared to patients with Stage 1 when controlling for age.
DiscussionAs observed on office cystoscopy, some women with RUTIs exhibit different levels of bladder wall involvement, ranging from just the trigone (stage 1) to the whole bladder (stage 4).[6] Some fare well after EF, likely due to the permanent removal of these chronic reservoirs of infection within the bladder wall. Our prior long-term EF series showed that women with cystitis localized to the trigone had 82% resolution based on post-EF cystoscopy.[9] However, some women have more diffuse lesions spreading beyond the trigone to the bladder base, lateral walls, and sometimes the entire bladder.
Our follow-up series in those who required repeat EF revealed that women with cystitis beyond the trigone were significantly more likely to have endoscopic failure.[25] Furthermore, some secondary lesions appeared after an initial EF and were new, as they had not been previously documented, suggesting unique bacterial properties of persistence and motility. Therefore, an understanding of the factors that specifically lead to the spread of cystitis is essential to assist treatment in this difficult subset of women. The purpose of this study was to examine the association between bacterial motility properties with stages of bladder wall involvement. While overall motility was similar, we did observe more numerous and unique types of motile bacteria at higher stages of bladder wall infection suggesting that motility may influence the spread of cystitis past the trigone.
The mouse model to study UPEC is well suited to an acute UTI event. However, it does not fully replicate the clinical situation of chronic RUTIs. E. coli was found to be the predominant strain amongst all stages and the mechanisms behind E. coli motility might be more complex. A recent study examining different strains of E. coli showed that nonpathogenic E. coli and UPEC strains differed in surface motility with regards to appendage usage (fimbriae vs. flagella) and other metabolic requirements such as glucose requirements.[26] In the present study, we did not differentiate between the different strains of E. coli by sequencing methodology.
Excluding UPEC, common organisms responsible for RUTI included Proteus mirabilis, Enterococcus sp., Klebsiella pneumoniae, Pseudomonas aeruginosa, and Staphylococcus spp. consistent with the distribution of other series.[27] Interestingly, we observed a preponderance of the motile bacteria such as Proteus and Pseudomonas at higher stages of cystitis. In vitro, Pseudomonas aeruginosa exhibits twitching characterized by pilus retraction and Proteus mirabilis exhibits swarming characterized by multicellular flagella-mediated movements.[28] Therefore, while E. coli motility may be more complex, motility such as swarming and twitching may represent a virulence factor to explain the cystitis spread of other common bacteria involved in RUTI.
The strengths of this study are the stages of bladder wall involvement based on photographic cystoscopy documentation and a detailed operative note report on the extent of cystitis. Furthermore, all studied strains were those identified before EF for antibiotic-recalcitrant RUTIs. Complicated RUTI patients were excluded from the study and only those with a completely negative upper and lower urinary tract evaluation with positive cystoscopy findings for chronic cystitis were included in the study. This study's limitations include a retrospective analysis of a prospectively acquired dataset at a single tertiary care center, which limits generalizability. In addition, due to our referral pattern, we were not able to successfully retrieve all documented occurrences of positive urinary cultures before EF and had to eliminate data obtained from outside laboratories to maintain consistency of data acquisition and reporting. These urine cultures represent a single data point in time, although the process to develop bladder surface inflammatory lesions may be much longer. Furthermore, while we were able to control for age, we could not account for all potential host characteristics related to the susceptibility of developing a UTI. Finally, there was no predictable progression of stages of cystitis from one to four. Some patients were in stage one whereas others started off at stage four. Therefore, we were not able to rely on the progression of bladder wall involvement when interpreting our data.
ConclusionIn this series of nonneurogenic, uncomplicated, postmenopausal women with cystoscopy-proven cystitis scheduled for an EF procedure, we observed that those with higher stages of visible bladder inflammation had more unique types of motile and nonmotile urinary bacterial strains compared to those with lower stages. A better understanding of the motility properties of these strains is needed to design effective preventive approaches. Office cystoscopy should be encouraged in women with recurrent UTIs to detect lesions at an early stage.
Acknowledgment
The authors are very grateful for the review of this article by Bonnie Prokesch, MD and Larry Reitzer, PhD, and for their insightful comments and suggestions.
Financial support and sponsorship
Nil
Conflicts of interest
Dr. Philippe E. Zimmern, an editorial board member at Urological Science, had no role in the peer review process of or decision to publish this article. The other authors declared no conflicts of interest in writing this paper.
References
留言 (0)