Discussion

To the best of our knowledge, our study represents the largest cohort to date examining the cystoscopic findings of female patients with rUTI. Major urologic societal guidelines regarding the utility of cystoscopy in this patient population currently lack robust data and are based solely on expert opinion, particularly for patient populations with high-risk features that fulfill the criteria for complicated rUTI. Our analysis supports prior observations that the diagnostic yield of cystoscopy in the index rUTI patient population is not only limited, but that the diagnostic yield in patients of advanced age and with complicated rUTI is minimal as well. Additionally, the study demonstrates that cystoscopy is not without risk as clinically insignificant findings can lead to further evaluation with increased risk exposures.

The diagnostic yield of cystoscopy in female patients with rUTI in the literature varies from 0 to 7.6%, with cohort sizes ranging from 15 to 379. ^{Reference Pat, Steffens and Witte20–Reference Santoni, Ng and Skews25} Only 1.3% (6/467) of cystoscopies in this cohort demonstrated significant findings that altered management. It is important to point out that significant findings among all patients were exclusively in the cohort of patients with high-risk features/complicated rUTI. Though all six of these patients fulfilled the criteria for complicated rUTI, these six represented only 3.1% (6/192) of all patients with complicated rUTI (Table 2). The complicated rUTI patients would be the only group undergoing cystoscopy had current guidelines been strictly adhered to. There were no significant findings on cystoscopy among patients with uncomplicated rUTI.

Prior studies excluded certain high-risk features which can translate to lower diagnostic yields. Pagano et al. excluded patients with hematuria in the presence of UTI. ^{Reference Pagano, Barbalat and Theofanides24} Despite our limited exclusions, our combined rate is lower (1.3%, 6/467) than that of Pagano et al (3.8%). ^{Reference Pagano, Barbalat and Theofanides24} While two life-threatening conditions were identified on cystoscopy (both carcinoma), these findings made up just 0.42% of all patients (2/476) and 1.0% of complicated rUTI patients (2/192). This is consistent with the rate of life-threatening cystoscopic findings in rUTI patients by Santoni et al. and Pat et al. of 0.15% and 0.26%, respectively. ^{Reference Pat, Steffens and Witte20,Reference Santoni, Ng and Skews25} There is no clear consensus as to the threshold for a positive yield that should warrant cystoscopy, although a yield of at least 5% has been previously suggested. ^{Reference Santoni, Ng and Skews25} In our study, the number of patients with complicated rUTI with positive findings was 6 out of 192 patients (3.1%). In other words, the number of patients with complicated rUTI needed to undergo cystoscopy to identify one positive finding was 32, demonstrating relatively low yield.

It is important to note that the criteria regarding which high-risk factors qualify a rUTI patient as “complicated” varies between medical societies. We included patients that fit all three criteria from the joint 2019 AUA/SUFU/CUA guideline: (1) history of anatomic or functional abnormalities of the urinary tract, (2) immunocompromising conditions, and (3) MDR organisms. Patients with signs of pyelonephritis and other systemic infections were also included in our analysis, in line with the ACR’s criteria. ^{Reference Venkatesan, Oto and Allen12} We also included all pregnant patients in accordance with the 2022 EAU Guideline. ^{Reference Bonkat, Bartoletti and Bruyere11} Of the 19 high-risk features detailed by the ACR, there are six that we did not consider in our criteria for complicated rUTI (elevated serum creatinine, asymptomatic bacteriuria, severe diabetes mellitus, childhood UTI, analgesic abuse, and urinary incontinence) as we felt they did not align with AUA/CUA/SUFU and EAU guidelines. ^{Reference Anger, Lee and Ackerman10–Reference Venkatesan, Oto and Allen12}

While the EAU guideline recommends against the use of cystoscopy in patients without high-risk features under 40 years of age, they do not offer recommendations for the management of patients of advanced age. ^{Reference Bonkat, Bartoletti and Bruyere11} Notably, the median age of our cohort (64.0 years) is over two decades older than the EAU’s cutoff. ^{Reference Bonkat, Bartoletti and Bruyere11} The median age of patients with high-risk features and significant findings on cystoscopy was 58 years. These findings suggest that the EAU guideline’s recommendation for all patients with high-risk features to undergo cystoscopic evaluation regardless of age may be of limited clinical benefit, and the cutoff age recommending against cystoscopy may have the potential to be increased. ^{Reference Bonkat, Bartoletti and Bruyere11} Our diagnostic yield for patients over the age of 55 years old is consistent with the 3.4% yield demonstrated by Dokubo et al. ^{Reference Dokubo, Reeves, Cashman and Gnanapragasam21} Our findings support the AUA’s expert opinion guideline of discouraging cystoscopy in the index patient, but importantly suggest that the AUA/CUA/SUFU’s, ACR’s, and EAU’s recommendations to consider cystoscopy in patients of advanced age or with high-risk features may be of low diagnostic yield. ^{Reference Anger, Lee and Ackerman10–Reference Venkatesan, Oto and Allen12} Future work may also focus on patient menopausal status in addition to age. Further stratification of patients by risk factors including sexual activity or use of spermicides may also be of clinical significance. ^{Reference Medina and Castillo-Pino1,Reference Geerlings4}

In addition to the low yield for cystoscopy in this patient population, we demonstrated an increased exposure to risks via continued diagnostic workup prompted by findings on cystoscopy. The risks of complications from cystoscopy, including discomfort and post-procedural UTI, have been previously described. ^{Reference Sihra, Malde, Greenwell, Pakzad, Kujawa and Sinclair17–Reference Rao, Rao and Reddy19} However, we demonstrated escalating diagnostic evaluation of three patients that necessitated general anesthesia to tolerate mucosal biopsy. All three of these patients ultimately yielded benign biopsy findings. This highlights the need for shared decision-making between providers and this patient population when weighing the benefits and risks of undergoing cystoscopy for the evaluation of rUTI. A recent systematic review of existing clinical practice guidelines for the assessment and treatment of rUTI found that only five of the eight identified guidelines took patients’ perspectives into account when developing their recommendations and none of the guidelines publicly reported those perspectives. ^{Reference Pat, Witte and Steffens35} Additionally, optional imaging demonstrates a significant cost burden to this patient population with Gaitonde et al. calculating the cost of initial evaluation for patients with rUTI being $730 and $390 with and without optional imaging, respectively. ^{Reference Gaitonde, Malik and Zimmern36} Patients must be informed of the diagnostic yield in the face of the potential risks and costs of pursuing a diagnostic workup in order to achieve appropriate shared decision-making.

Our study is subject to several limitations. It was conducted at a single center in a retrospective manner with heterogeneity among patients regarding prior treatments and management among several treating physicians, with updated guidelines and treatments throughout the observed period of study. Timeframe from first UTI to cystoscopy and types and duration of initial management, including but not limited to vaginal estrogen, were not analyzed in this study. Possible adverse events after diagnostic cystoscopy were also outside the scope of this study and should be examined as a point of further work. Additionally, the criteria for high-risk features for patients with rUTI vary among societal recommendations, and our criteria attempted to combine several guidelines in order to conduct our analysis. We did not correlate imaging data to cystoscopy findings to compare their utilities; however, of the six significant cystoscopic findings documented in our cohort, none would have warranted cystoscopy if preoperative imaging had been initially performed. Another limitation is that although we deemed cystitis cystica as a benign finding, recent preclinical and retrospective studies have suggested it may pose a risk factor in developing rUTIs. ^{Reference Ligon, Wang and DeJong37,Reference Chiu, Zhang and Sutcliffe38} Certain research has also identified inflammatory lesions as a possible bacterial reservoir managed with electrofulguration; however, the identification of these lesions and further management was not included as significant findings in our study. ^{Reference Chavez, Christie and Zimmern39,Reference Ma, Chavez, Christie and Zimmern40} The literature demonstrates a higher yield of cystoscopic findings in male rUTI patients, and so the findings in our female-only cohort should not be generalized to men. ^{Reference Anger, Lee and Ackerman10–Reference Venkatesan, Oto and Allen12,Reference Dokubo, Reeves, Cashman and Gnanapragasam21}

Prospective, randomized studies evaluating the correlation of high-risk features to the probability of obtaining significant cystoscopic findings are needed, but due to the relatively low diagnostic yield, they will require extremely large sample sizes in order to demonstrate statistical significance. Additionally, patients may request additional evaluation after extensive antibiotics or persistent organisms in subsequent cultures. Further research could examine patient satisfaction or patient-recorded outcomes and whether patients request imaging measures for reassurance.

Societal guidelines recommend that cystoscopy should not be obtained in the index rUTI patient, but do not specify their utility in patients >40 years and/or with high-risk features for complicated rUTI and/or urothelial malignancy. These recommendations are based on expert opinion and lack the support of robust evidence. In the largest retrospective study of female rUTI patients undergoing cystoscopy to date, cystoscopy demonstrated no clinical benefit and a low diagnostic yield in patients with uncomplicated and complicated rUTI, respectively. Our findings suggest that cystoscopy may have limited benefit with an increased risk exposure in complicated rUTI patients over the age of 50 and support the recommendation to avoid cystoscopy for rUTI patients without risk factors. Physicians should hold patient-oriented discussions with complicated rUTI patients regarding the likely yield of clinical benefit and risk exposure if considering further cystoscopic management. Further research regarding the utility of cystoscopy for patients with rUTI is needed with increasingly robust methodologic design.