Practice Patterns Regarding Female Reproductive Organ-Sparing and Nerve-Sparing Radical Cystectomy Among Urologic Oncologists in the United States.

Overall, SUO membership included 723 members at the time of the survey, with 52.4% listing bladder cancer as a specific area of interest. Among 168 urologists who responded to our survey (44.3% response rate among SUO members with a specific interest in bladder cancer), 101 participants indicated that they had previously performed a female RC and completed all survey questions about operative approach in premenopausal women and were, therefore, included in our analysis. Overall, the median age of participants was 43 years (range: 31–80 years, interquartile range [IQR]: 38–52), 10 participants (9.9%) were female, 84 (83.2%) were fellowship-trained in urologic oncology, 81 (80.2%) were in academic practice, 92 (91.1%) were involved in training residents, 45 (44.6%) were involved in training urologic oncology fellows, and the median time in practice was 9 years (IQR: 3–18) ( Table 1 ). When asked about female RC volume in the 12 months prior to the survey, 32 participants (31.7%) reported that they had performed at least 10 female RCs in the past 12 months, and the median number of female RCs reported in the 12 months prior to the survey was 5.5 (IQR: 3–10) ( Table 1 ). Overall, 80 participants (79.2%) routinely resected the uterus or cervix, 68 (67.3%) routinely resected the neurovascular bundle, 49 (48.5%) routinely resected the ovaries, and 19 (18.8%) routinely resected a portion of the vagina when performing RC in premenopausal patients with clinically-localized T2 MIBC or NMBIC that had failed intravesical therapy ( Figure 1 ). Furthermore, 18 participants (17.8%) reported that they never spare the uterus or cervix, 21 (20.8%) that they never spare the neurovascular bundle, 6 (5.9%) that they never spare the ovaries, and 1 (1.0%) that they never spare a portion of the vagina when performing female RC in a premenopausal patient ( Figure 1 ). Provider characteristics such as female sex, age, academic practice, urologic oncology fellowship training, and years in practice were not significantly associated with routinely resecting the uterus or cervix, neurovascular bundle, ovaries, or a portion of the vagina during RC in premenopausal patients with clinically-localized T2 MIBC or NMIBC that had failed intravesical therapy ( Table 2 ). Additionally, 72 participants (71.3%) correctly identified the neurovascular bundle as coursing along the lateral walls of the vagina, and provider knowledge of the anatomical location of the neurovascular bundle was not significantly associated with performing nerve-sparing RC in premenopausal patients (relative risk [RR] 0.88, 95% CI: 0.66–1.19). Overall, 71 participants (70.3%) reported that they were less likely to spare the uterus or cervix, 44 (43.6%) were less likely to spare the neurovascular bundle, 70 (69.3%) were less likely to spare the ovaries, and 23 (22.8%) were less likely to spare a portion of the vagina when performing RC in a postmenopausal patient with organ-confined disease. 42 participants (41.6%) reported that the type of urinary diversion impacted whether they performed ROS or nerve-sparing RC in female patients. Among 29 participants who provided free text responses about the impact of urinary diversion type on operative approach, 26 reported an increased likelihood for ROS RC when constructing a neobladder in a female patient. Participants reported that they preferred to spare the uterus/cervix, neurovascular bundle, and vagina to maximize pelvic support and optimize the neurovascular supply of the sphincter to improve urinary continence in patients receiving a neobladder. Some participants preferred to avoid vaginal resection when creating a neobladder to reduce the risk of fistula development. One participant explicitly stated that they were more likely perform a urethrectomy in patients receiving an ileal conduit, which rendered nerve-sparing more difficult, while urethral preservation during neobladder creation facilitated nerve-sparing. 11 participants provided free text responses about their approach to ROS or nerve-sparing RC in premenopausal patients. Reasons for performing a hysterectomy included a lack of patient desire to have more children and preoperative gynecologic issues (e.g., menorrhagia or fibroids). Reasons for resecting the neurovascular bundle included a lack of clarity about the necessity of the neurovascular bundle, with one participant stating that the bundle was “not important for continence or satisfaction/climax.” Alternatively, another participant cited preservation of vaginal lubrication as a rationale for conducting nerve-sparing RC. Reasons for oophorectomy included patient preference, patient age, concomitant endometriosis, and concern about future malignancy, while reasons for sparing the ovaries included preservation of bone and hormonal health. Reasons for vaginal preservation included preservation of vaginal capacity, avoidance of vaginal reconstruction, decreased length of operation and blood loss, and reduction in rates of fistulas or enteroceles. 13 participants provided free text responses about their approach to ROS and nerve-sparing RC in postmenopausal patients. Reasons for performing a hysterectomy cited by providers included patient preference, the patient being past reproductive age, and reduced oncologic risk. Reasons for resecting the neurovascular bundle included a lack of sexual activity among older patients. Reasons for performing oophorectomy included a lack of functional ovaries, prophylactic avoidance of ovarian cysts, and reduced oncologic risk. Reasons for sparing the vagina included patients’ desire to maintain sexual activity and preservation of pelvic floor support.

We developed and piloted a survey based on a literature review and formative interviews with urologic oncologists at Johns Hopkins Hospital as previously described 14 ( Supplementary Figure 1 ). Survey items assessing provider demographics and clinical experience were adapted from a study by Sussman et al. as previously reported. 13 , 14 We also developed survey items that assessed providers’ operative technique when performing female RC. Specifically, participants were asked how frequently they would spare the uterus/cervix, neurovascular bundle, ovaries, and a portion of the vagina when performing a female RC in a premenopausal patient with clinically-localized T2 MIBC or NMIBC that had failed intravesical therapy. Frequency was measured on a 5-point Likert scale (never, rarely [<25%], sometimes [25–50%], most of the time [51–99%], and always), and participants were also invited to provide free text responses about the rationale for their approach. 15 We also asked urologists how their operative approach would change with respect to preserving the uterus/cervix, neurovascular bundle, ovaries, and a portion of the vagina if the patient were a postmenopausal woman. Change in operative technique for a postmenopausal patient was assessed on a 3-point scale (less likely to spare, no change, and more likely to spare), and participants were invited to provide free text responses about their rationale for changes in approach. Additionally, participants were asked if the type of urinary diversion impacted whether or not ROS or nerve-sparing RC was performed in female patients. They were also asked to demonstrate anatomic knowledge of the location of the neurovascular bundle in relation to the vagina ( Supplementary Figure 1 ). Our recruitment methodology was previously published. 14 Briefly, after receiving approval from the Johns Hopkins Institutional Review Board, a link to the survey was sent via e-mail to members of the Society of Urologic Oncology (SUO) in the U.S., and responses were collected electronically from September 2018 through December 2018. Surveys were anonymous; however, individual responses were tracked via a unique respondent identification number. Respondents who indicated that they had previously performed a female RC and who completed all survey questions about operative approach in premenopausal patients were included in our analysis. Descriptive statistics were used to summarize provider characteristics, the frequency of sparing each reproductive organ (i.e. uterus/cervix, ovaries, and a portion of the vagina) or sparing the neurovascular bundle during RC in premenopausal patients with clinically-localized T2 MIBC or NMIBC that had failed intravesical therapy, and changes to ROS and nerve-sparing approaches in postmenopausal patients with a similar disease burden. We also calculated the proportion of participants who affirmed that the type of urinary diversion impacted whether or not ROS or nerve-sparing was performed during female RC and the proportion of participants who correctly identified the anatomical location of the neurovascular bundle on the knowledge item of the survey. A provider was considered to routinely resect a specific female reproductive organ or the neurovascular bundle during RC if they reported that they never, rarely, or sometimes spared the organ (corresponding to sparing ≤50% of the time). We used modified Poisson regression to examine associations between provider characteristics and provider-reported rates of routine resection of each female reproductive organ or the neurovascular bundle during RC in premenopausal patients with clinically-localized T2 MIBC or NMIBC that had failed intravesical therapy. 16 We also used modified Poisson regression to examine the association between provider knowledge of the anatomic location of the neurovascular bundle and routine resection of the neurovascular bundle during RC in premenopausal patients with clinically-localized T2 MIBC or NMIBC that had failed intravesical therapy. We also analyzed participants’ free text responses regarding rationale for approach to ROS and nerve-sparing in premenopausal patients, changes in approach in postmenopausal patients, and the impact of urinary diversion type on operative approach. All statistical analyses were performed using STATA 12.0/MP for Linux (Stata Corp. LP, College Station, TX).

Multiple studies have demonstrated the oncologic safety and functional benefit—including sexual function benefit—of ROS and nerve-sparing RC in select female patients with bladder cancer. 7 – 9 , 11 , 12 , 17 , 18 For example, potential benefits of ROS and nerve-sparing RC in female patients include preservation of fertility, improved urinary control in patients receiving a neobladder, improved pelvic floor support, improved sexual function, and avoidance of surgical menopause. 9 To this end, the AUA guidelines advocate for consideration of ROS and nerve-sparing RC as an oncologically safe option in select patients with organ-confined disease. 2 However, in our national survey of urologic oncologists performing female RC in the U.S., a majority of providers reported that they routinely perform a hysterectomy and do not spare the neurovascular bundle at the time of RC in premenopausal patients with organ-confined disease. Additionally, nearly 50% reported that they routinely perform oophorectomy at the time of RC (with 5.9% reporting that they never spare the ovaries), and nearly 20% reported that they routinely resect a portion of the vagina at the time of RC in premenopausal patients with organ-confined disease. Furthermore, provider characteristics and clinical experience were not predictive of operative approach, indicating these practices are prevalent among urologists of varying backgrounds. Additionally, we identified gaps in provider knowledge related to female ROS and nerve-sparing RC. It is likely that the rates of routine resection of reproductive organs and the neurovascular bundle in female patients with organ-confined disease reported by providers in our study are not optimal, given that published rates of uterine or cervical involvement at the time of RC are <10% and rates of ovarian involvement at the time of RC range from 0% to < 5%. 13 , 19 – 23 Additionally, surgical menopause as a result of oophorectomy at the time of RC in premenopausal patients can have a significant deleterious impact on patients’ overall health, as surgical menopause has been associated with increased all-cause mortality, increased mortality from non-ovarian cancers, reduced bone health, reduced cardiovascular health, and impaired cognitive function. 24 – 28 In a prior survey of SUO members, a majority of urologic oncologists were aware of the increased risk of cardiovascular disease and osteoporosis associated with surgical menopause, but cited oncologic concerns as the reason for performing oophorectomy at the time of RC. 13 However, our current understanding of ovarian cancer pathogenesis indicates that prophylactic salpingectomy without oophorectomy can reduce, although not eliminate, the risk of ovarian cancer and avoids the increased mortality and morbidity associated with surgical menopause. 29 Therefore, given our understanding of ovarian cancer pathogenesis, the low rates of urothelial metastasis to the ovaries, and the negative effects of surgical menopause, salpingectomy at the time of RC may be an alternate and preferable approach to reducing the risk of future ovarian malignancy in premenopausal patients with organ-confined disease. To our knowledge, this is the first study to examine provider-reported practice patterns regarding female ROS and nerve-sparing RC in the U.S. However, this study has several limitations. As a cross-sectional survey, our data is susceptible to responder bias, especially given that not all urologic oncologists perform RC in their practice. However, since those who reported that they had not previously performed a female RC were excluded from our analysis, it is probable that any responder bias in this study would actually favor representation of SUO members who are familiar with performing RC. While we did not identify differences in practice patterns based on provider characteristics such as age or practice setting, approximately 80% of participants in our study were in academic practice and a future larger study including a diverse sample of urologic oncologists is warranted to further explore potential differences in practice patterns among providers. Additionally, we did not ascertain location of practice in our survey, and an examination of geographic trends in female ROS and nerve-sparing RC remains an area for future research. We also did not inquire about differences in surgical decision-making based on a robotic versus open RC approach and did not ask providers about other patient factors, such as obesity and variations in anatomy, that could impact surgical decision-making. Additional research is needed to better characterize factors influencing operative decision-making and barriers to performing ROS and nerve-sparing RC in female patients with organ-confined disease. As part of this effort, there is an ongoing multi-center clinical trial ( NCT05399004 ) that seeks to further characterize the impact of ROS and nerve-sparing RC on sexual function outcomes among female patients. Future efforts should also seek to address knowledge gaps about female ROS and nerve-sparing RC among urologists through training and education, which may be a fruitful area for future quality improvement work by professional organizations such as the SUO and AUA.

In conclusion, our study highlights important gaps in adoption of female ROS and nerve-sparing RC in premenopausal and postmenopausal patients with organ-confined disease among U.S. urologic oncologists. Future efforts should be directed towards improving training in and education about indications for ROS and nerve-sparing RC techniques among urologists to optimize postoperative functional and overall health outcomes among female patients with bladder cancer undergoing RC.

Radical cystectomy (RC) remains the gold standard treatment for patients with high-risk non-muscle-invasive bladder cancer (NMIBC) unresponsive to intravesical therapies and muscle-invasive bladder cancer (MIBC). 1 , 2 In addition to removal of the bladder, the American Urological Association (AUA) guidelines state that standard RC in female patients should also involve removal of the uterus, fallopian tubes, ovaries, and anterior vaginal wall (clinical principle). 2 While the oncologic rationale for this en bloc approach is removal of adjacent organs at risk for local tumor extension, this surgical approach can result in a significant negative impact on female sexual and overall health due to anatomic and endocrine changes such as disruption of the clitoral vascular supply, alterations in vaginal anatomy, damage to the neurovascular bundle, weakening of the pelvic floor, and induction of surgical menopause. 3 – 6 Alternatively, female reproductive organ-sparing (ROS) and nerve-sparing RC techniques, which aim to preserve some or all of the gynecologic organs and neurovascular bundle, have been shown to be oncologically safe in select patients and to improve functional outcomes, including sexual function outcomes. 7 – 12 To that end, the AUA guidelines state that “clinicians should discuss and consider sexual function preserving procedures” for patients with organ-confined disease and absence of bladder neck or urethral involvement (Grade C). 2 While a recent study suggests that urologists are aware of the risks and benefits of preserving the ovaries during female RC, overall practice patterns and factors impacting surgical decision-making regarding ROS (i.e. sparing of the uterus/cervix, ovaries, and/or vagina) and nerve-sparing techniques among urologists performing female RC are not known. 13 We hypothesized that urologists performing female RC in the U.S. do not adequately consider ROS and nerve-sparing techniques when making operative decisions in patients with organ-confined disease. Therefore, we sought to characterize provider-reported practice patterns regarding performance of ROS and nerve-sparing RC in premenopausal and postmenopausal patients with organ-confined disease, as well as factors associated with operative decision-making, among urologic oncologists in the U.S.