Cervical cancer is the fourth leading cause of death worldwide and one of the leading causes of cancer-related morbidity in women. The wide implementation of pelvic radiotherapy as standard-of-care for localized disease in 1999 has improved survival, and we are now able to assess the long-term urological effects. It was our clinical impression that survivors of cervical cancer with urological complications following radiotherapy required numerous repeated urological procedures. To better quantify this hypothesis, we examined the rate of radiation-induced urological complication requiring procedural intervention (RUCPI) at our institution.

We performed a single-institution retrospective chart review of all cervical cancer patients who underwent primary radiotherapy between 1998 and 2012. We included all patients with total radiation dose >50 Gy and followup duration longer than 6 month. RUCPI was defined as a urological complication requiring urological procedure following radiotherapy. Data regarding all procedures of the urinary tract were collected including those performed in the operative room and clinic settings, and those performed by interventional radiology. We compared the study cohort of patients with RUCPI vs those with no procedure. Patients undergoing a urological procedure not clearly attributable to radiation (ie history of urological procedure prior to radiotherapy, oncologic recurrence) were excluded from further analysis. Multivariate analysis was used to identify predictors of RUCPI.

Out of 378 patients seen by gynecology for cervical cancer, a total of 134 patients with FIGO stage 1A2-4B cervical cancer met inclusion criteria. Of these, 26 patients underwent urological procedures, 18 of whom experienced RUCPI (14.3%). The mean followup duration for the entire cohort was 63 (24.5–88.0) months. The cohort received a mean total radiation dose of 82.0 (±7.4) Gy. The most common complications were ureteral stricture and radiation cystitis (table 1). Those with RUCPI differed from those without procedures in smoking status, type of sensitizing chemotherapy, and duration of followup. However, in multivariate analysis, only smoking status at the time of diagnosis was predictive of RUCPI (OR 3.44 [95% CI 1.34–19.26], p=0.03). Stage was not a significant predictor of RUCPI.

To our knowledge, this was the first retrospective series to show the rate of urological complications requiring procedural intervention in patients receiving primary radiotherapy for cervical cancer and to quantify the number of procedures required to manage these complications.

We identified a high rate (14.3%) of RUCPI in patients undergoing radiotherapy for cervical cancer, which is consistent with recent literature of 10% and 15% in a claims-based study and a Surveillance, Epidemiology, and End Results-Medicare database study assessing Clavien III–IV complications.^1,2 It is worth noting, however, that although the risk of urological complications continued to increase for 25 years,³ the majority of gynecologic oncology literature does not include late complications. Focused studies reporting on only ureteral stricture rates following radiotherapy for cervical cancer (2.5%–3%) date back to the 1990s and report only severe complications requiring major open operations.^3,4 Most recently the EMBRACE trial showed an actuarial risk of ureteral stricture of 2.0% at 5 years.⁵ However, its median followup duration of 34 months limited its capture of late complications as seen in our series.

Our most interesting finding lies in the sheer number of procedures endured by those with RUCPI. Over the median followup duration of 7.3 months for those with RUCPI, a total of 253 procedures were performed, representing 14.4 procedures per patient with RUCPI. In patients with ureteral stricture, the average number of procedures was even greater at 26.4 procedures per patient.

Ureteral stenting was the most common procedure in those with ureteral stricture, reflecting the repetitive nature of ureteral stenting in otherwise healthy patients (table 2). Thus, the consequences of ureteral stenting in patients with incidentally discovered hydronephrosis must be carefully considered given the repetitive nature of these procedures. Traditional ureteral stenting with 3–6 ureteral stents per year is thought to incur estimated medical charges of $18,218–$36,437 per patient, in addition to the cost of missed work and travel.⁶ Additionally, cervical cancer survivors with chronic ureteral stents have higher rates of urinary tract infections, lower urinary tract symptoms, and pain.⁷ Unfortunately, the alternatives are either renal failure or reconstructive surgery. Neither are desirable outcomes as reconstruction of irradiated ureters has a high failure rate with significant morbidity in 17% and re-intervention in 28%.^8,9

Smoking at the time of diagnosis was the only predictor of RUCPI after adjusting for age, stage, radiation dose and type of chemotherapy used. This is consistent with prior data showing that smoking at the time of radiation was predictive of urological complications.¹⁰ As such, patients should be counseled on the importance of smoking cessation prior to radiotherapy to mitigate the risk of urological complications.

Limitations to this study include the small cohort size of patients managed at a single institution, loss to followup and relatively short followup duration. Further, dose of radiation had no predictive ability in our analysis due to tight range of radiotherapy dosing at our institution during this time period. As such, the lack of dose-effect should not be misinterpreted.

In conclusion, our study shows that multiple procedures may be necessary to manage the sequelae of complications following radiotherapy for cervical cancer, highlighting the complexity of the management of cervical cancer survivors with urological complications.