Iatrogenic injury to the genitourinary tract is a rare, but significant cause of morbidity in patients undergoing gynecologic surgery for benign indications.1–3 In 2012, the American Association of Gynecologic Laparoscopists recommended that routine cystoscopy be performed after all laparoscopic hysterectomies, whereas the American College of Obstetricians and Gynecologists limited the endorsement to prolapse and incontinence procedures.4,5 Currently, no definitive guidelines regarding the role of cystoscopy at the time of hysterectomy for benign indications in the absence of prolapse and incontinence procedures currently exist.
Universal cystoscopy, where all women undergoing hysterectomy undergo cystoscopy, has been advocated because as many as 75% of genitourinary injuries occur in women without identifiable risk factors. At the University of Michigan, implementation of a universal cystoscopy policy at the time of hysterectomy for benign indications performed by benign gynecologists was associated with a significant decrease in delayed postoperative urologic complications (seven injuries, 0.7% 95% CI 0.3–1.2% preimplementation vs two injuries, 0.1% 95% CI 0.0–0.3% postimplementation).6 The authors concluded that a universal cystoscopy policy at the time of hysterectomy is a low-cost intervention that poses minimal risk to patients. Larger studies, however, have not corroborated this finding. A systematic review and meta-analysis that included 79 studies with 41,482 hysterectomies found that universal cystoscopy did not decrease the incidence of delayed genitourinary injuries.7 Risks of cystoscopy include increased rates of urinary tract infection, bladder and urethral trauma, allergic reaction to contrast agents, cost and increased operating room time. Given the conflicting reports in the literature regarding the benefits of cystoscopy in decreasing the incidence of delayed genitourinary injuries, further research into the best application of cystoscopy to hysterectomy for benign indications is needed to inform quality and practice guidelines.8
Routine postprocedure cystoscopy is intended to recognize a genitourinary injury intra-operatively, allowing for immediate repair. If cystoscopy is a sufficiently sensitive screening tool, women who undergo cystoscopy should have a lower risk of a subsequent delayed genitourinary tract injury.9–11 The objective of this study was to estimate the association between cystoscopy at the time of hysterectomy for benign indications and the subsequent occurrence of a delayed genitourinary tract injury. Secondary objectives were to estimate the association between cystoscopy and urinary tract infection, as well as operative time.
A retrospective cohort study of women who underwent hysterectomy for benign indications recorded in the National Surgical Quality Improvement Program was performed. The National Surgical Quality Improvement Program is a national quality database that collects preoperative, intraoperative and postoperative variables related to surgical procedures. Hospitals voluntarily participate in the database, and in exchange for participation, are given data regarding their own procedures to drive quality improvement.12 Data are abstracted by trained clinical reviewers and are audited regularly. For an institution's data to be used in the nationally available file, the interobserver agreement during the audit must be greater than 95% and averages 98% for included sites.13 Within the National Surgical Quality Improvement Program, there is a targeted hysterectomy file that includes patient history, and intraoperative and postoperative variables specific to hysterectomy.14 This study was reviewed by the Institutional Review Board at Northwestern University and was deemed exempt from formal review given the deidentified nature of the data.
Our cohort included all women recorded in both the hysterectomy-specific file and the general National Surgical Quality Improvement Program file from 2015 to 2017, which were linked. Patients who underwent hysterectomy for benign indications were identified by excluding those patients who underwent hysterectomy for cancer using the cancer case variable. Patients who underwent surgery with gynecologic oncologists, maternal–fetal medicine specialists, reproductive endocrinologists, urogynecologists, and physicians categorized as “other” were excluded given the differences in patient populations and guidelines regarding cystoscopy for the various subspecialties.4,15 Hysterectomy with additional procedures performed to treat prolapse and urinary incontinence, as defined by Current Procedural Terminology (CPT) code (Table 1) were also excluded, again given the clear guidelines for cystoscopy in this population owing to the higher rate of urinary tract injury. Hysterectomies with additional procedures performed that were not directed at prolapse or incontinence were included. Data regarding surgical volume or center volume was not available in this data source.
The primary outcome was delayed lower genitourinary tract injury. This was a composite outcome defined as an occurrence of a National Surgical Quality Improvement Program–defined lower genitourinary tract injury as well as reoperation for a repair of the genitourinary tract. The National Surgical Quality Improvement Program collects three genitourinary complications from the medical record: ureteral obstruction, bladder fistula and ureteral fistula. As these complications are collected from direct medical record review, nonsurgical treatment, such as percutaneous nephrostomy tube placement or catheter placement to treat a genitourinary tract injury, are included. The postoperative day on which the complication is recognized is also recorded. Women who experienced a lower genitourinary tract injury on postoperative days 1–30 were defined as having a delayed lower genitourinary tract injury, and women who had the injury diagnosed on postoperative day 0 were not. Furthermore, women who experienced a reoperation for repair of the genitourinary tract were also defined as experiencing a delayed lower genitourinary tract injury. The CPT codes that defined a reoperation for a delayed lower genitourinary tract injury are listed in Table 2. Our exposure of interest was the occurrence of cystoscopy during hysterectomy that was defined by CPT code (Table 1) listed in addition to the hysterectomy CPT code.
The exposure of interest, cystoscopy at the time of hysterectomy, was also compared with additional secondary outcomes of urinary tract infection and operative time. Urinary tract infection was defined by the National Surgical Quality Improvement Program participant user file. Operative time was also defined as in the National Surgical Quality Improvement Program, however, because the National Surgical Quality Improvement Program contains up to 20 additional procedural codes for multiple procedures performed at the time of surgery, this analysis was performed in a subset of patients. To isolate the operative time attributable to cystoscopy alone, patients with only a hysterectomy CPT code (Table 1) and a diagnostic cystoscopy code (CPT 52000) were compared with patients with only a hysterectomy code to eliminate operative time that could be explained by the time required to perform additional procedures.
Associations between cystoscopy and primary and secondary outcomes were estimated using chi-square test, Fisher exact test, Wilcoxon rank sum test, and modified Poisson regression as appropriate. Given the low number of genitourinary injuries, there were insufficient outcomes to perform a multivariable analysis. A stratified analysis was performed by route of hysterectomy, which was defined based on CPT code as open, laparoscopic or robotic assisted laparoscopic, and vaginal or vaginally assisted (Table 1). All P-values were two sided with P<.05 considered significant. STATA version 14.0 was used for all analyses.
Sensitivity analyses were performed to ensure that our results remained unchanged in subgroups. Specifically, a comparison was made between each of the different delayed genitourinary complications to ensure that there were not differences in specific injuries with and without cystoscopy that would be masked by a composite outcome. Furthermore, the association between cystoscopy and the outcomes of delayed lower genitourinary tract injury and urinary tract infection were also performed in the subgroup of women who had only one hysterectomy CPT code recorded compared with those who had one hysterectomy code and a diagnostic cystoscopy CPT code (as described above).
We identified 39,529 women who underwent hysterectomy for benign indications with an obstetrician–gynecologist. Patient characteristics are described in Table 2. Patients who underwent cystoscopy were more likely to be of white and Asian race and less likely to be of black race, compared with patients who did not undergo cystoscopy. Patients who underwent cystoscopy were also 1–2% more likely to have a history of abdominal or pelvic surgery and had higher parity. Additionally, the use of cystoscopy increased through the study period from 21.1% in 2015 to 27.4% in 2017 (P<.001). There were few differences between patients who experienced a delayed lower genitourinary tract injury and those who did not (Table 3). Patients who experienced a delayed injury had larger uteri by approximately 30 grams compared with those who did not.
For the primary outcome, there was no difference in the rate of 30-day delayed lower genitourinary tract injury between patients who underwent cystoscopy and those who did not (Table 4). Patients who did not undergo cystoscopy had a 0.24% rate of 30-day injury, whereas patients who did undergo cystoscopy had a 0.27% injury rate; this difference was neither clinically nor statistically significant. Subsets of the composite outcome were examined to ensure that there was not an association between cystoscopy and delayed lower genitourinary tract injury in certain types of injury. Rates of ureteral obstruction, bladder fistula, ureteral fistula and any reoperation for a lower genitourinary tract injury were all similar with no significant clinical or statistical difference between women who underwent cystoscopy and those who did not. Additionally, the median postoperative day the injury was diagnosed was not statistically different between patients who underwent cystoscopy and those who did not (Table 4).
A stratified analysis was also performed by route of surgery to investigate whether there were differences in the relationship between cystoscopy and delayed lower genitourinary tract injury by route of surgery (Table 5). Surgical approach was open (26%), laparoscopic or robotic (46%), and vaginal or vaginally assisted (28%). The rates of cystoscopy differed by surgical approach with laparoscopic, robotic and vaginal approaches having an approximately 25–32% prevalence of cystoscopy, whereas patients undergoing open surgery had a lower prevalence of cystoscopy (11%, P<.001). For laparoscopic, robotic and vaginal approaches there was no association between cystoscopy and subsequent delayed lower genitourinary tract injury; for both groups, the rate of delayed lower genitourinary tract injury was approximately 0.18–0.25%. However, for open surgery the rate of delayed lower genitourinary tract injury was higher in the patients who had cystoscopy performed. These patients had a 0.78% delayed genitourinary injury rate compared with 0.30% for patients who did not undergo cystoscopy (P=.01).
Patients undergoing cystoscopy had a 2.6% rate of subsequent urinary tract infection, whereas women who did not had a 2.0% rate (RR 1.27 95% CI 1.09–1.47). Patients undergoing cystoscopy also had a longer median operative time. In a subgroup analysis of patients with only two CPT codes recorded (n=6,072), one hysterectomy code and one diagnostic cystoscopy code, compared with patients with only a hysterectomy code recorded (n=25,837), patients who underwent cystoscopy had a median operative time of 132 minutes (interquartile range 100–172) and patients who did not had a median operative time of 115 minutes (interquartile range 86–157) (Table 6). This was a difference of 17 minutes and was statistically significant (P<.001). Additional sensitivity analyses were carried out in this population for both primary and secondary outcomes of delayed lower genitourinary tract injury and urinary tract infection, which are reported in Table 6 and confirmed our primary analysis.
Cystoscopy after hysterectomy for benign indications was not associated with a decrease in delayed 30-day genitourinary injuries compared with no cystoscopy in this study. The rate of delayed lower genitourinary tract injury was 0.27% and 0.24% with and without cystoscopy, respectively, and the results are neither statistically or clinically significant (P=.64). This is consistent with the results of a prior systematic review and meta-analysis that found universal cystoscopy was not associated with a decrease in the rate of delayed lower genitourinary tract injury.7 The incidence of delayed lower genitourinary tract injury in that meta-analysis was 0.07–0.16% which is similar although less frequent than our findings.
Our data also confirms that the sensitivity of cystoscopy as a screening tool for detection and prevention of a delayed lower genitourinary tract injury is not 100%, as the rate of delayed injury remained 0.27% even after cystoscopy. Specifically, intraoperative cystoscopy is likely to be normal in the case of thermal injuries. In the era of increased use of minimally invasive surgery, thermal injuries may become a more prevalent mechanism of injury. The rate of delayed lower genitourinary tract injury in women undergoing laparoscopic or robotic hysterectomy was not different and was without a suggestion of clinical benefit (0.18% vs 0.20%), potentially because these injuries are more likely to be thermal in nature. Interestingly, the one statistically significant finding in this paper was the association between cystoscopy and subsequent delayed lower genitourinary tract injury in women undergoing open surgery. This likely represents a selection bias given the low prevalence of cystoscopy (11%) in the open group. Surgeons likely performed cystoscopy because of an intraoperative concern of lower genitourinary tract injury. However, it is interesting that even after presumably identifying cases at higher risk, cystoscopy did not ultimately prevent these women from experiencing a delayed lower genitourinary tract injury or developing a fistula. Consideration for an intraoperative consultation or prophylactic stent placement may be warranted if there is a suspicion of genitourinary tract injury, even if cystoscopy is normal, because it cannot definitively rule out these conditions. Although cystoscopy is a powerful tool to assess the genitourinary tract, ultimately, the best approach to prevent injury is meticulous retroperitoneal dissection and awareness of the ureter and bladder through the entire procedure.
The policies and surgical practice patterns regarding cystoscopy for these institutions and surgeons are unknown, and, thus, the debate regarding universal compared with selective cystoscopy cannot be resolved with these data. However, approximately 70% of the women in this study did not undergo cystoscopy and had delayed urologic injury rates that were not different from the women who underwent cystoscopy, suggesting that surgeons were able to identify a large population of women in whom delayed urinary tract injury rates were not increased with the omission of cystoscopy. Therefore, these data could be interpreted in favor of selective over universal cystoscopy.
Ultimately, the decision to use cystoscopy in a given hysterectomy is a risk-benefit determination by the surgeon. Although the risks to cystoscopy are minimal, in this study, we did find a statistically significant 27% increase in the risk of urinary tract infections in women who underwent cystoscopy. Urinary tract infections are the fourth most common health care associated infection and urinary tract infection is included in the Medicare list of eight hospital acquired infections for which penalties or nonpayment can be levied against institutions.16,17 Additionally, cystoscopy was associated with a median increase in operative time of 17 minutes. Increases in operative time and cost are the most commonly cited concerns regarding cystoscopy with recent estimates of operating room time, not including surgeon and anesthesiologist fees, at $37 per minute,18 cystoscopy results in an additional $629 per hysterectomy in operating room time alone. The only published cost-effectiveness analysis on this topic suggests that urinary tract injury rates would have to be between 1.5 and 2.0% for universal cystoscopy to be considered cost-effective.19 Although this analysis has been criticized and cystoscopy may be cost effective at lower thresholds of injury; the most recent systematic review and meta-analysis found a contemporary lower genitourinary tract injury rate of 0.3%20 and cost-effectiveness analyses performed using these lower genitourinary tract injury rates are of great interest. Other risks of cystoscopy include a false sense of reassurance for the surgeon and potentially changes in surgeon behavior or technique as a result of knowing cystoscopy will be performed.
This study has the strengths of a large sample size from a large repository of surgical quality data with robust information about postoperative complications including three variables specific for complications related to the urinary tract. Additionally, the data set contains a large amount of information regarding the surgical procedure itself and any reoperations, which were particularly important in defining our exposure and our outcome. Finally, our results were unchanged in various subpopulations and sensitivity analyses. Limitations of this study include that the National Surgical Quality Improvement Program is a hospital-based registry and thus is not nationally representative. Additionally, complications in the National Surgical Quality Improvement Program are only recorded for 30 days postoperatively, and thus any complications beyond this are unknown. However, genitourinary injuries recognized at greater than 30 days after hysterectomy may be more likely to have a normal cystoscopy given the amount of time required for the injury to manifest. Finally, although we had a large sample size, delayed genitourinary injuries were rare, thus, the ability to adjust for confounders or to have sufficient power to detect a small benefit to cystoscopy (smaller than magnitude 0.16% or 0.18% for the sensitivity analysis) was limited; however, as can be seen in Tables 2 and 3, clinical differences to suggest confounding were rare. Although injuries were rare, the 97 injuries among the more than 39,000 women reported in this paper are similar in size to the single systematic review and meta-analysis on this topic and significantly larger than the majority of papers commonly cited. However, even with these limitations, in this contemporary series, the rate of delayed lower genitourinary tract injury after hysterectomy for benign indications was approximately 0.25% in both women who underwent cystoscopy and those who did not. Our data highlight the limitations of cystoscopy in prevention of delayed lower genitourinary tract injury and encourages the use of other strategies, beyond cystoscopy, to improve surgical quality and decrease the rate of delayed urinary tract injury in women undergoing hysterectomy for benign indications.
1. Dowling RA, Corriere JN Jr, Sandler CM. Iatrogenic ureteral injury. J Urol 1986;135:912–5.
2. Ozdemir E, Ozturk U, Celen S, et al. Urinary complications of gynecologic surgery: iatrogenic urinary tract system injuries in obstetrics and gynecology operations. Clin Exp Obstet Gynecol 2011;38:217–20.
3. Sawkar HP, Kim DY, Thum DJ, et al. Frequency of lower urinary tract injury after gastrointestinal surgery in the nationwide inpatient sample database. Am Surg 2014;80:1216–21.
4. The role of cystourethroscopy in the generalist obstetrician-gynecologist practice. ACOG Committee Opinion No. 372. American College of Obstetricians and Gynecologists. Obstet Gynecol 2007;110:221–4.
5. AAGL Advancing Minimally Invasive Gynecology Worldwide. AAGL practice report: practice guidelines for intraoperative cystoscopy in laparoscopic hysterectomy. J Minim Invasive Gynecol 2012;19:407–11.
6. Chi AM, Curran DS, Morgan DM, Fenner DE, Swenson CW. Universal cystoscopy after benign hysterectomy: examining the effects of an institutional policy. Obstet Gynecol 2016;127:369–75.
7. Teeluckdharry B, Gilmour D, Flowerdew G. Urinary tract injury at benign gynecologic surgery and the role of cystoscopy: a systematic review and meta-analysis. Obstet Gynecol 2015;126:1161–9.
8. Peacock LM, Young A, Rogers RG. Universal cystoscopy at the time of benign hysterectomy: a debate. Am J Obstet Gynecol 2018;219:75–7.
9. Gilmour DT, Dwyer PL, Carey MP. Lower urinary tract injury during gynecologic surgery and its detection by intraoperative cystoscopy. Obstetrics Gynecol 1999;94:883–9.
10. Pettit PD, Petrou SP. The value of cystoscopy in major vaginal surgery. Obstet Gynecol 1994;84:318–20.
11. Wu HH, Yang PY, Yeh GP, Chou PH, Hsu JC, Lin KC. The detection of ureteral injuries after hysterectomy. J Minim Invasive Gynecol 2006;13:403–8.
12. American College of Surgeons. American College of Surgeons National Surgical Quality Improvement Program user guide for the 2015 participant use file. Chicago (IL): American College of Surgeons; 2016.
13. Shiloach M, Frencher SK Jr, Steeger JE, Rowell KS, Bartzokis K, Tomeh MG, et al. Toward robust information: data quality and inter-rater reliability in the American College of surgeons national surgical quality improvement program. J Am Coll Surgeons 2010;210:6–16.
14. American College of Surgeons. American College of Surgeons National Surgical Quality Improvement Program user guide for the 2015 procedure targeted participant use file. Chicago (IL): American College of Surgeons; 2016.
15. Barber EL, Rossi EC, Alexander A, Bilimoria K, Simon MA. Benign hysterectomy performed by gynecologic oncologists: is selection bias altering our ability to measure surgical quality? Gynecol Oncol 2018;151:141–4.
16. Centers for Disease Control and Prevention. Urinary tract infection (catheter-associated and non-catheter-associated UTI) and other urinary system infection events. Available at: http://www.cdc.gov/nhsn/pdfs/pscmanual/7psccauticurrent.pdf
. Retrieved November 22, 2018.
17. Centers for Medicare and Medicaid Services. Hospital-acquired condition reduction program. Available at: https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/HAC/Hospital-Acquired-Conditions.html
. Retrieved November 25, 2018.
18. Childers CP, Maggard-Gibbons M. Understanding costs of care in the operating room. JAMA Surg 2018;153:e176233.
19. Visco AG, Taber KH, Weidner AC, Barber MD, Myers ER. Cost-effectiveness of universal cystoscopy to identify ureteral injury at hysterectomy. Obstet Gynecol 2001;97:685–92.
20. Wong JMK, Bortoletto P, Tolentino J, Jung MJ, Milad MP. Urinary tract injury in gynecologic laparoscopy for benign indication: a systematic review. Obstet Gynecol 2018;131:100–8.