Long‐term renal functional outcomes following ureteroureterostomy performed during multi‐organ resection for non‐urothelial cancers

Abstract Objectives To evaluate the long‐term renal function outcomes after ureteroureterostomy (UU) in patients undergoing multi‐organ resection for non‐urothelial cancers. The secondary aim was to examine the length of ureteric defect that can be successfully bridged with UU. Patients and methods We retrospectively reviewed the charts of patients who underwent UU between 1995 and 2012 at our institution. Renal imaging studies performed before and after UU were used to determine whether hydronephrosis was present. Renal function was assessed by comparing estimated glomerular filtration rate (eGFR) before and at the last follow‐up after UU. Results Nineteen patients underwent UU during multi‐organ resection for non‐urothelial cancers. Median follow‐up time was 62 months. Overall, UU had a high success rate, with one patient (5.2%) developing progressive hydronephrosis with a >20% drop in eGFR from baseline due to UU failure. Four additional patients developed progressive hydronephrosis due to cancer recurrence involving the UU. There were no statistically significant differences between pre‐ and post‐UU eGFR in these patient cohort. All patients with a ureteric defect of ≤5 cm underwent successful reconstruction. Conclusions UU maintains long‐term renal function in the majority of patients undergoing multi‐organ resection for non‐urothelial cancers and can be successfully utilized if the resected ureteric length is ≤5 cm.


| INTRODUC TI ON
Resection of a segment of one or both ureters is sometimes necessary to adequately remove advanced pelvic or retroperitoneal cancers. Depending on the length and location of the resected segment, a variety of ureteric reconstruction options are available. 1 If the resected segment of the ureter is mid to distal, then ureteroneocystostomy with or without an elongation procedure (psoas hitch or Boari flap) is simple and highly successful. 2 However, if a relatively short section of the mid or upper ureter is resected, then ureteroureterostomy (UU) may be feasible, particularly in cases in which the distal ureter is viable. Although UU is considered when the ureteric defect is short; the length of ureteric defect that can be bridged with UU including aggressive nephropexy has not been well studied. Some cancers may displace and stretch the ureter allowing a longer ureteric defect to be successfully bridged with UU.
There are very limited reports in the literature regarding renal functional outcomes of UU reconstruction during multi-organ resection for cancer [3][4][5][6][7] (Table 1). Most of the earlier studies include few patients and failure of the UU based on the development of stricture or hydronephrosis is reported in 27%-100% of patients. [3][4][5][6][7] One of the largest series, by Fry et al 8 included two patients that died of urine leakage and sepsis in the postoperative period. 3 Only one of the earlier reports examine renal functional outcomes using estimated glomerular filtration rates based on serum creatinine measurements.
Although there are no standardized or agreed upon definitions of success following UU reconstruction, successful reconstruction would include stable renal function and the absence of progressive hydronephrosis or stricture on subsequent imaging studies.
The aim of our study is to evaluate long-term renal functional outcomes following complex multi-organ resection for non-urothelial cancers with UU reconstruction. A secondary aim was to examine the length of ureteric defect that can be successfully bridged with UU.

| PATIENTS AND ME THODS
A research protocol to evaluate renal function outcomes after a variety of ureteric reconstructions including UU was approved by the Institutional Review Board at the University of Texas MD Anderson Cancer Center. We conducted a retrospective chart review of all patients who underwent UU for a complete ureteric transection between 1995 and 2012. All patients in this study were undergoing complex cancer surgery for non-urothelial cancers. Patients who had other modalities of diversions or had a UU for urothelial cancers were excluded. As these patients underwent complex cancer surgery involving multidisciplinary surgical approach, all patients were counseled pre-operatively about all the available reconstruction options, the risk, and benefits by the treating urologist. However, the decision to perform a UU reconstruction vs other options was made intraoperatively by the treating urologist. In all patients undergoing UU, the proximal and distal ureters were spatulated and anastomosed end to end with absorbable sutures over a ureteric stent.
Nephropexy was utilized as needed. The duration of ureteric stenting was at the discretion of the treating surgeon.
Patients medical records were reviewed in detail for demographic data, oncologic information, and renal function including anastomotic patency, stricture development, and evidence of progressive hydronephrosis on serial imaging including CT scans, MRI's, intravenous pyelography, abdominal ultrasound tests, or obstruction on nuclear renal scan.

| RE SULTS
A total of 32 patients undergoing UU were identified; no follow-up information was available for 8 patients leaving 24 evaluable. Of these, 5 patients were also excluded as they had partial ureteric Mixed population benign and cancer (ovarian, colorectal, or endometrial) (4) Long-term satisfactory 2-21 years; (4) Lost to follow-up after 1 yr; (2)   with Clavien ≥3 complications post-UU (Table 3).

| D ISCUSS I ON
We present the long-term renal functional outcomes using a combination of eGFR and imaging after UU in a series of non-urothelial cancer patients undergoing multi-organ resection. Renal preservation is essential in cancer patients undergoing multi-organ resection as majority require additional systemic therapies. Chronic kidney disease (CKD), a graded and independent risk factor for substantial F I G U R E 1 Baseline and last follow-up eGFR in 24 patients undergoing UU. Baseline GFR is denoted in black and last follow-up GFR in gray | morbidity and death has been found to complicate the treatment of cancer patients. 10 While, all surgical options should be considered to optimize renal preservation in these patients, UU is rarely utilized as a re-constructive option of the ureter, mainly owing to the high rates of complications seen in small studies (Table 1). However, UU is successfully utilized at renal transplant with a meta-analysis demonstrating no difference in the overall rates of complications between UU and uretero neocystostomy, with stricture, obstruction and stone formation being the more common complications associated with UU. 11 In contrast to the other studies, only one patient in our study (5.2%) had failure of the UU as defined by a > 20% drop in eGFR from baseline and progressive hydronephrosis from anastomotic stricture and obstruction of the UU. We found that differences in patients eGFR's before and after UU were not significant; this finding suggests that long-term renal function is maintained in the majority of patients who undergo UU. While not statistically significant, some patients experienced an increase in eGFR following UU, possibly due to improved index kidney function after relief of obstruction in these patients.
Successful UU during multi-organ resection requires a meticulous surgical technique and can be used in select cases. 1 The anastomosis needs to be tension-free and aggressive mobilization of the kidney should be considered if those cases where tension may exist. The distal ureter must be viable with reasonable blood supply. EAU guidelines recommend UU for proximal mid-ureteric injuries involving a ureteric segment of <2-3 cm. 2 The longest ureteric defect to be successfully reconstructed in our series was 5.6 cm and the single patient who ex- Patients undergoing UU during multi-organ resection have cancers with heterogeneous biology and are often undergoing "desperation" surgery. Therefore, our primary goal was not to assess oncologic outcomes, but rather success of the UU reconstruction.
We also included patients who developed cancer recurrence involving the UU with progressive hydronephrosis. Each of these patients developed significant recurrence in the abdomen or pelvis due to aggressive tumor biology and three of these four patients died of progressive malignancy. In order to reduce bias and identify the actual stricture rates associated with UU, we utilized the eGFR values just prior to cancer recurrence in these patients.
The main limitations of our paper are its retrospective nature, limited number of patients and procedures performed at a high volume center which may not be generalizable. However, our report is the largest-to-date of patients undergoing UU during multi-organ resection of cancer (Table 1). An additional strength of our paper is that all patients had baseline and post-UU eGFR information and post-UU imaging with long-term data available for the majority due to ongoing cancer surveillance allowing us to demonstrate durable success in the majority of patients. While we attempted to minimize bias by inclusion of the patients who progressed, this may have also af-

| CON CLUS IONS
Long-term renal function following UU is maintained in the majority of patients undergoing multi-organ resection for non-urothelial