Abstract
Objectives:
Ureteroscopic ureteral perforations have been reported in up to 6% of cases, with recent studies suggesting a decline to less than 2%. Ureteroscopic perforations are managed with prolonged ureteral stenting of up to 6 weeks based on historical data. We sought to evaluate the time of urothelial healing and duration of ureteral stenting following a ureteroscopic perforation in a porcine model.
Materials and Methods:
Part A: Ureteral perforation using a semirigid ureteroscope was performed in 37 ureters. The ureters were stented using 4.7F × 22 cm stents for 3, 7, 10, or 14 days, and retrograde pyelograms performed after stent removal. Injured ureteral segments were collected for histologic evaluation. Part B: 8 ureters had endoscopic perforation and stenting for 7 days and then survived for 4 weeks for evaluation of urinary extravasation or hydronephrosis and histologic evaluation.
Results:
Part A: At 3 days of ureteral stenting, there was urinary extravasation on retrograde pyelograms and gross defect in all ureters; average creatinine increased (1.55–1.75 mg/dL). Starting at 7 days, no evidence of gross urothelial defects or extravasation, and average creatinine was stable. Histologic evaluation revealed urothelial healing by 7 days with ongoing tissue healing. Granulation tissue predominated in early phase of healing. Part B: With only 7 days of ureteral stenting, no extravasation or hydronephrosis developed a month after stent removal.
Conclusions:
Following ureteroscopic ureteral perforation in a porcine model, the urothelium is functionally intact with 7 days of stenting. These results are sustained without complications for at least 4 weeks after stent removal. While further studies are warranted, these results challenge the current practice of maintaining ureteral stenting for several weeks following ureteral perforation during ureteroscopy.
Introduction
Ureteroscopy (URS)
There is a paucity of data evaluating ureteral regeneration following a complete perforation of the ureter during URS, as well as the optimal length of time for ureteral stent duration to ensure continuity of the ureteral mucosa for urine transport. Therefore, we sought to evaluate the time of urothelial healing after a complete endoscopic ureteral perforation using a porcine model.
Materials and Methods: Part A
The current study was approved by the Institutional Animal Care and Use Committee for the U.S. Air Force 59th Medical Wing Clinical Investigation and Research Support (Lackland, Air Force Base, TX) (FWH20170094A). This facility's animal care and use program is accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care International.
We conducted a study of a complete endoscopic ureteral perforation using 24 female Yorkshire–Landrace crossbred (average age 2.9 months, average weight 54.5 kg) local vendor pigs that were divided into four different treatment arms based on length of time of postinjury ureteral stenting (3, 7, 10, or 14 days). Our primary outcome was to determine the time to urothelial healing by gross evaluation as well as microscopic examination by a veterinarian pathologist. Secondary outcomes assessed immediate and short-term complications to include acute kidney injury, ureteral abnormalities, or urinary extravasation.
Animals were sedated using an intramuscular (IM) injection of Telazol 4.4 mg/kg and ketamine 2.2 mg/kg. Anesthesia was induced with 2% to 4% isoflurane and maintained using 1.5% to 2.5% isoflurane during the procedure. Preanalgesia was provided with buprenorphine 0.01 to 0.05 mg/kg IM and postoperatively ketoprofen, 3 mg/kg for 2 to 3 days.
In anesthetized pigs, and after intravenous administration of weight-based ceftriaxone, we gained laparoscopic access to the abdomen and dissected a small peritoneal window overlying the proximal or midureter in all cases. Dissection of the periureteral tissue was limited to decrease the risk of ischemia but to allow for direct visualization of the ureter during URS and ultimately the perforation site. After a guidewire was introduced in a retrograde manner up the ureter and into the renal pelvis using a flexible cystoscope (Olympus Surgical, Southborough, MA), the semirigid ureteroscope (6.5F; Richard Wolf, Knittlingen, Germany) was advanced into the proximal or midureter, and along an oblique trajectory, a submucosal tunnel and complete perforation of the ureter under direct laparoscopic visualization were performed solely using the ureteroscope as to best simulate perforation during URS (Fig. 1A).

Ureteroscopic injury of porcine ureter.
While maintaining the ureteroscope within the ureter, two sutures were placed laparoscopically, with minimal manipulation of the periureteral tissue, near the proximal and distal limits of the injured ureter for identification of the injured site at the time of ureter collection (Fig. 1B). The ureteroscope was removed leaving the guidewire within the collecting system, and a 4.7F × 22 cm indwelling ureteral stent (Boston Scientific, Quincy, MA) was placed in a retrograde manner from the bladder using direct laparoscopic vision to ensure that the stent was across the injured site and within the renal pelvis. This procedure was repeated on the contralateral ureter of the same pig. The ureteral stents were maintained for 3, 7, 10, or 14 days depending on the treatment arm. The pigs were evaluated daily for any signs of complications.
At the time of tissue collection, the pig was reanesthetized, as described above, and the stents were removed with a flexible cystoscope. A retrograde pyelogram of each ureter was then performed to evaluate for urinary extravasation or ureteral abnormalities. The pigs were then humanely euthanized using IV pentobarbital, 100 mg/kg (euthanasia solution) and in accordance with the 2013 AVMA Guidelines for the Euthanasia of Animals. A midline laparotomy was performed and both ureters were collected at least 6 inches proximal and distal to the injured sites (previously marked with suture). The ureters were collected for gross examination by the surgical team followed by a microscopic evaluation by a veterinarian pathologist. Laboratory data to monitor renal function were collected preoperatively as well as the time of ureter collection.
Ureteral tissue sections were fixed in 10% neutral buffered formalin, processed routinely into 5 μm hematoxylin/eosin (H&E)-stained tissue sections, and evaluated microscopically for evidence of tissue healing by a board-certified veterinarian pathologist. Tissue healing was evaluated microscopically to assess injury severity and the percentage of granulation tissue versus fibrosis replacing normal architecture with the assumption that the higher percentage of fibrosis suggested more advanced healing. One hundred percent fibrosis with no granulation tissue would be considered fully healed. The integrity of the ureter to prevent urine leakage into the abdomen was determined qualitatively by assessing the continuity of the urothelium and muscle layers. If the urothelium was a minimum of one cell layer thick or ulcerated with underlying granulation tissue in the lamina propria, the ureter was considered functionally intact. If the urothelium and the muscle layers were discontinuous, the ureter was considered unhealed to the point of preventing urine leakage.
Materials and Methods: Part B
Based on the initial findings of Part A, a final short-term follow-up study was conducted in four additional pigs (8 ureters) to ensure continued durability of the ureteral urothelium at the injured site following stent removal and to identify any hydronephrosis or short-term complications. The same experiment was performed as described above in this cohort with the ureteral stents remaining in place for 7 days. At that time, under general anesthesia, the ureteral stents were removed and retrograde pyelograms were performed. The pigs then survived (without ureteral stents) for an additional 4 weeks in which a retrograde pyelogram was performed followed by collection of the ureteral segments. As in Part A, the ureteral segments were evaluated grossly and microscopically, and renal function laboratories were collected at each surgery date.
Results: Part A
Three-day stent trial
Five pigs (10 ureters) had ureteral stents in place for the 3-day time period. All five pigs survived the surgical procedure and had expected postoperative courses. All 10 ureteral stents were in appropriate position on fluoroscopic imaging and were removed without complications. On retrograde pyelogram evaluation, there was urinary extravasation noted in all 10 ureters (Fig. 2A). In addition, there was an obvious gross ureteral defect in all ureteral segments. There was significant urine ascites noted in the abdomen at the time of ureteral collection. On average, creatinine values increased from 1.48 to 1.66 mg/dL (ref. range 1–2.07 mg/dL 15 ) (Table 1).

Retrograde pyelogram of injured ureters.
Averaged Animal Creatinine Values
POD, post operative day.
There was evidence of transmural microscopic injury in 10/10 ureters characterized by urothelial damage, moderate to marked inflammation, muscular disruption and granulation tissue. At low magnification, there is full-thickness injury to the tissue layers, including degeneration, necrosis, and loss of urothelium with adjacent granulation tissue, hemorrhage, and inflammatory infiltrate (Fig. 3A). At higher (20 × ) magnification, there is degeneration, necrosis, and loss of the urothelium. Within the lamina propria, there is granulation tissue, hemorrhage, fibrin, and edema with mixed inflammation, predominantly neutrophils (Fig. 4A).

Low-magnification (2 × ) microscopic evaluation of ureteral injury.

Higher magnification (20 × ) of ureteral injury.
Seven-day stent trial
Four pigs (8 ureters) had ureteral stents in place for the 7 days. One pig had both stents migrate distally past the injury site and was excluded in the analysis since the time duration of stent migration was unknown. Of the included pigs, all survived the surgical procedure and had an expected postoperative course. After ureteral stent removal, retrograde pyelogram evaluation of each ureter did not reveal any urinary extravasation, ureteral narrowing, or hydronephrosis. Upon tissue collection, urinary ascites was not identified. Creatinine values decreased from an average of 1.42 to 1.27 mg/dL (Table 1).
There were no gross defects in any of these ureteral segments. The microscopic evaluation of the injured ureters at low magnification (2 × ) revealed that circumferentially there is an intact urothelium with focal hyperplasia. Within the lamina propria, there are focally extensive fibroplasia and granulation tissue extending into the muscularis (Fig. 3B). At higher magnification, there were focal urothelial hyperplasia and the focally extensive granulation tissue and myofibroblasts surrounding the mycotes of the muscularis (Fig. 4B). Almost all of the examined sections had changes within the adventitia such as varying degrees of hemorrhage, edema, perivascular fibrin, vessel thrombosis, and both vessel and generalized tissue necrosis. All ureteral segments were considered functionally intact for urine transport based on retrograde pyelogram analysis.
Ten- and 14-day stent trials
In the 10 days [four pigs (8 ureters) one pig had both stents migrate distally and was excluded] and 14 days [five pigs (9 ureters), 1 stent had migrated distally and that ureter was excluded] groups, all pigs survived to the designated collection date without complications. Similar to the 7-day group, there was no evidence of urinary extravasation, ureteral narrowing, or hydronephrosis on any retrograde pyelograms. On laparotomy, there were no urinary ascites and no gross defects on any ureteral segments. Creatinine values remained the same or decreased in both groups (Table 1).
Microscopic evaluation in the 10-day cohort at low magnification (2 × ) revealed cystic degeneration of the urothelium and inflammatory infiltrates within the lamina propria. In addition, there are fibroplasia and granulation tissue surrounding islands of hyperplastic urothelium (Fig. 3C). At higher magnification, there are continued fibroplasia and granulation tissue within a hyperplastic urothelium along with predominant lymphocytes and histiocytes (Fig. 4C). At both lower magnification (Fig. 3D) and higher magnification (Fig. 4D), the 14-day cohort displayed an intact hyperplastic urothelium and focal muscle fiber loss with replacement of fibrous connective tissue.
Results: Part B
All four pigs (8 ureters) survived to the designated collection date of 4 weeks following ureteral stent removal. There were no adverse events or complications noted during this time frame. On retrograde pyelogram, at time of ureteral stent removal (1 week from injury) and specimen collection (5 weeks from injury), there was no evidence of hydronephrosis, ureteral narrowing, or urinary extravasation (Fig. 5A). Renal function remained stable with average creatinine of 1.1 mg/dL before injury and 1.2 mg/dL 1 month postoperatively (Table 1).

Experiment Part B (ureteral stent maintained for 7 days and survival for 30 days following stent removal)
Microscopically, all sections of ureter evaluated were consistent with normal ureteral tissues in terms of being void of any significant overlaying fibrosis or scar formation (Fig. 5B). The ureteral lumens all appeared patent without stricture. In a few of the sections, there was identification of fibrosis within the adventitia, most likely due to resolved suture/foreign body reaction as it appeared microscopically separate from the muscularis layer.
Discussion
Recently, there has been a significant increase in the use of URS, 16 –18 which is now considered a first-line treatment for ureteral and kidney stones less than 2 cm based on recent guideline updates. 3,4 Despite new technologic advances in ureteroscopes and associated equipment, including ureteral access sheaths, there continues to be some degree of ureteral tissue trauma in almost half of all ureteroscopic procedures. A postureteroscopic lesion score developed by Schoenthaler et al. revealed ureteral tissue “lesions” in over 50% of their cases, with 22% described as submucosal tissue disruption or partial perforation. 19 Similarly, Trexer and Thomas evaluated ureters following ureteral access sheath placement and discovered severe injury in about 13% of cases involving the smooth muscle layers of the ureter. 20
Historically, complete ureteral perforation during URS has occurred in 2% to 6% of cases due to ureteroscopes, basket equipment, laser lithotripsy, ureteral dilation, access sheath placement, or other associated equipment, 6 –10 and while more recent studies suggest a decrease in URS perforation rate, 9 the management of such injuries continues to be ureteral stent placement for 3 to 6 weeks or more, based largely on historical data. 10,11,19,20
Most of the literature regarding ureteral injury healing and stent management is derived from the early experiences of the intubated ureterotomy in the canine model by Davis in 1943. 12 In their study, ureteral smooth muscle was circumferentially regrown in about 90% of the ureter at 6 weeks. Oppenheimer and Hinman excised half of the circumference of canine ureters, placed a polyethylene stent, and again noted almost complete regeneration of smooth muscle at 6 weeks. 21 After endopyelotomy in a porcine model, Kerbel et al. demonstrated that the mucosa heals within the first 2 to 3 weeks, 22 and Rehman et al. determined that by 3 months the ureteral defect had been bridged by functional smooth muscle cells. 14 However, despite these previous studies, it remains unclear at what point the ureter is functionally intact to allow for safe urinary drainage after ureteral stent removal. We sought to determine the appropriate minimal length of ureteral stent duration following endoscopic perforation of the ureter to facilitate tissue healing for an intact urothelium.
A porcine model was chosen given its anatomical similarities to the upper urinary tract of the human. 22 –25 In our study, the 7-day cohort was noted to have a functionally healed ureter with no contrast extravasation or urothelial defect at the time of tissue collection. Therefore, we concluded that the ureteral tissue was healed enough for urine drainage at 7 days. Similar histologic findings of an intact urothelium were seen within the ureteral segments of the 7-, 10-, and 14-day cohorts with granulation tissue decreasing linearly over time with no fibrosis evident. In addition, at 14 days, muscle fiber loss was replaced with fibrous connective tissue. Furthermore, all retrograde pyelograms of the 7-, 10-, and 14-day cohorts were without extravasation or ureteral narrowing. The kidney function of the animals within the 7-, 10-, and 14-day time frames remained stable throughout the experiment.
In contrast, the 3-day ureteral stent duration cohort revealed contrast extravasation on all retrograde pyelograms as well as gross ureteral defects. Histologic examination revealed significant serosal necrosis and transmural disruption of all layers of these ureters. Based on these findings, we concluded that at 3 days, without an intact urothelium noted both grossly and microscopically, removing the stent could lead to continued urinary extravasation, but 7 days may be the minimal stent duration to ensure a functionally intact urothelium within a porcine model.
These results were confirmed by our Part B experimental cohort, which demonstrated continued ureteral integrity for urine transport for 4 weeks following ureteral stent removal. In addition, there was preserved renal function and no extravasation, ureteral narrowing, or hydronephrosis identified on any retrograde pyelogram in this cohort. Histologically, at 1 month following stent removal (5-week postinjury), the tissues revealed mild evidence of fibrosis, but no significant scarring within the urothelium, lamina propria, or muscularis. We do acknowledge the possibility that during the microscopic evaluation of the Part B cohort, we may have “missed” the ureteral defects during the processing of these segments. Grossly, these areas between the marking sutures appeared normal. In addition, all retrograde pyelograms were without extravasation or ureteral narrowing.
To our knowledge, this is the first study specifically evaluating ureteral healing after full-thickness endoscopic ureteral perforation during URS. In a porcine model, regeneration of just the urothelium appears to be sufficient to allow adequate urinary drainage while there is ongoing healing of the lamina propria and muscular layers. Furthermore, these results are maintained for up to 4 weeks following stent removal without the development of urinary extravasation or hydronephrosis. Clinically speaking, this suggests that 7 days of ureteral stenting may be a sufficient duration in which stents may be safely removed with no adverse events in a porcine model.
While there are potential limitations with animal models, most endourologic investigations of the upper urinary tract are conducted within a porcine model due to its anatomic and physiologic similarities to the human urinary system. 14,22 –25 More specifically, Schwalb et al. demonstrated the reproducibility of effective URS using standard-sized endoscopy equipment that would be used during human URS. 24 Furthermore, several studies have specifically evaluated ureteral healing following endoureterotomy procedures 14,22,26 in determining the optimal healing using a ureteral stent within a porcine model. 27,28
Our findings for the optimal duration of ureteral stenting following perforation during URS may be cautiously translated to clinical practice and be less than the standard 3 to 6 weeks of recommendations. While longer term animal and prospective clinical studies are needed to validate our results, decreasing ureteral stent indwelling time may allow for improved patient discomfort, fewer urinary infections, and decreased risks and complications associated with retained stents. Our study challenges the current practice of maintaining a ureteral stent for several weeks following endoscopic perforation and helps to identify the optimal time of ureteral stenting at least within a porcine model.
Conclusions
Following ureteroscopic perforation of a porcine ureter, the urothelium is functionally intact after 7 days. Our findings suggest that within a porcine model, 7 days may be a sufficient duration of ureteral stenting in which stents may be safely removed with no adverse events. While further prospective animal and clinical studies are warranted, these results challenge the current practice of maintaining a ureteral stent for several weeks after a ureteral perforation during URS and may help to guide future endoscopic practices.
Footnotes
Disclaimers
The views expressed are those of the authors and do not reflect the official views or policy of the Department of Defense or its Components.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
No funding was received for this article.
