Subcutaneous ureteral bypass device placement with intraoperative ultrasound guidance, with or without microsurgical ureterotomy, in 24 cats

Abstract

Objectives

The aim of this study was to describe a method of subcutaneous ureteral bypass (SUB) device placement with intraoperative ultrasound guidance, with or without microsurgical ureterotomy, for the treatment of benign ureteral obstruction(s) in cats. A secondary aim was to describe the complications and outcomes in our population, and compare the two groups with and without ureterotomy.

Methods

The medical records of cats with benign ureteral obstruction(s) treated with SUB device placement with intraoperative ultrasound guidance between April 2013 and June 2018 were reviewed.

Results

Twenty-four cats with 30 obstructed ureters had a SUB device placement with intraoperative ultrasound guidance in 26 surgeries. A microsurgical ureterotomy was performed in 14/26 surgeries. Median age was 10.07 years (range 4–16.6). Eleven of 24 cats (46%) had a previous history of chronic kidney disease (CKD). All cats, including two patients with minimal pelvic dilation (⩽4 mm), had a successful intrapelvic placement of the SUB device. Median survival time was 1555 days (4.25 years); this was not significantly different between the two groups (P = 0.4494). Stone analysis and bacterial culture, where available for review, revealed calcium oxalate in 12/12 and a negative culture in 6/7. The ureterotomy significantly prolonged the procedure duration (180 vs 125 mins) without significantly decreasing the short- and long-term complications (P = 0.1588 and P = 0.2921, respectively), or the survival time (P = 0.8437).

Conclusions and relevance

SUB device placement with intraoperative ultrasound guidance is an effective alternative for the treatment of ureteral obstruction. Ultrasound guidance may be a more accessible option when a trained veterinarian does not have access to fluoroscopy. Microsurgical ureterotomy did not show any advantage and prolonged the anesthesia. Our median survival time emphasizes a good outcome of SUB device placement, even in cats with a previous history of CKD.

Introduction

Benign ureteral obstructions have become increasingly prevalent in our feline population. While the pathogenesis of the increasing prevalence of upper urinary tract stones in cats ¹ is still not well understood, ² the decompression of the obstruction, regardless of its origin, should be performed rapidly to avoid secondary kidney damage.^3,4 Subcutaneous ureteral bypass (SUB) device placement is one of the recommended interventional treatments for ureteral obstructions in cats.^3–5 This procedure was first described in 2011, ⁶ and has recently been reported in a large group of cats. ⁷

The current standard surgical technique described by Berent at al uses fluoroscopic guidance, ⁷ limiting the use of this technique to larger, specialty hospitals. A previous study described SUB device placement using blind placement without fluoroscopic guidance in 13/19 cats. ⁸

The current standard technique also leaves in place the ureterolith at the origin of the obstruction, while its removal via microsurgery could, theoretically, remove a potential site of persistent infection and revert to the natural pathway in the event of a SUB device obstruction. A study comparing the outcome of ureterotomy vs ureteral stenting found a greater decrease in blood urea nitrogen (BUN) and creatinine, and a greater survival at discharge for the ureteral stenting group. ⁹

The aim of this study was to describe the method of SUB device placement with intraoperative ultrasound guidance, with and without microsurgical ureterotomy, for the treatment of benign ureteral obstruction(s) in cats. A secondary aim was to describe the complications and outcomes in our population comparing those with and without ureterotomy.

We hypothesized that placement of a SUB device with intraoperative ultrasound guidance would be an effective alternative for the treatment of benign ureteral obstructions in cats. We hypothesized that a microsurgical ureterotomy would significantly increase the procedure duration without improving the outcome.

Materials and methods

Case selection

The medical records of cats with benign ureteral obstructions treated with SUB device placement at Cummings School of Veterinary Medicine, Tufts University, between April 2013 and June 2018 were retrospectively reviewed. Patient follow-up continued until June 2019. A ureteral obstruction was highly suspected/diagnosed based on history, physical examination, previous (if available) and current bloodwork, and ultrasonographic images. Case information included signalment (age at SUB device placement, sex, breed), previous history of chronic kidney disease (CKD), creatinine (on admission, preoperatively, 24 h and 48 h postoperatively, on discharge, and at 1 and 3 months postoperatively), size of pelvis and ureteral dilation, urine culture and sensitivity (UCS) via cystocentesis at admission and via intraoperative pyelocentesis, ureterolith analysis and culture, side of SUB device placement, procedure duration, number of postoperative hospitalization days, and complications [classified as previously described: ³ intraoperative; postoperative (<1 week after surgery) in the short-term (1 week to 1 month after surgery) or long-term (>1 month after surgery)]. The cats were also distributed to two different subgroups, with or without ureterotomy.

SUB device placement

SUB device placement was performed in accordance with the Norfolk Vet surgical guide. ¹⁰ Deviating from the original description, the nephrostomy and cystostomy catheters were placed with intraoperative ultrasound instead of fluoroscopic guidance. The ultrasound was performed by a radiology resident or faculty, or by a trained internal medicine resident. The sterile environment was maintained with a sterile sheet covering the entire ultrasound probe placed directly against the kidney capsule and the bladder wall. Contact was maintained with sterile saline.

When the pelvis was particularly small in size, visualization of the pelvis on ultrasound was better in transverse from the lateral pole than in sagittal from the caudal pole. Thus, the nephrostomy catheter was placed in the pelvis from the lateral border of the kidney rather than from the caudal pole. When the patient had no cardiovascular contraindication, a bolus of 5 ml/kg crystalloid was also given intravenously to improve pelvis visualization and placement accuracy. An 18 G over-the-needle catheter was inserted in the caudal pole of the kidney until the tip of the needle was visualized in the middle of the renal pelvis (Figure 1). A 0.035” metallic J-tip guidewire was then gently advanced through the 18 G catheter and coiled inside the renal pelvis, sometimes along a calice (Figure 2). The 18 G catheter was then carefully removed over the guidewire and the 6.5 F locking-loop pigtail nephrostomy catheter was advanced over the guidewire into the renal pelvis. The guidewire and the stylet of the nephrostomy catheter were removed and the string pulled to lock the coil in place. When the pelvis was particularly small, the stylet was withdrawn once visualized in the pelvis, to keep only the nephrostomy catheter visible. The soft nephrostomy catheter tip was then gently pushed and coiled inside the pelvis/calices, until the radiopaque marker was no longer visible outside of the kidney. To confirm the placement and rule out leakage, an intraoperative bubble study was performed with a syringe connected to an 18 G catheter inserted directly into the end of the nephrostomy catheter (Figure 3). The insertion site of the nephrostomy catheter was directly inspected during the bubble study to rule out leakage before the Dacron cuff was glued to the capsule. The cystostomy catheter was placed in accordance with the surgical guide. The cystostomy catheter placement was also controlled with ultrasound guidance to avoid any contact with the trigone. Once both catheters were attached to the subcutaneous port, a final intraoperative flushing of the SUB device was performed to rule out leakage and confirm proper active and passive draining of both catheters. The subcutaneous port was finally secured to the ventral body wall prior to closure. Abdominal radiographs were taken at the end of the surgery to assure proper placement and rule out any kinking of the catheters.

Figure 1.

The 18 G over-the-needle catheter was inserted in the caudal pole of the kidney until the tip of the needle was visualized in the middle of the renal pelvis (arrow)

Figure 2.

The 0.035” metallic J-tip guidewire coming from the caudal pole was coiled inside the renal pelvis (arrows). Note the J-tip artifact in the renal pelvis (asterisk)

Figure 3.

An intraoperative bubble study was performed with a syringe connected to an 18 G catheter inserted directly at the end of the nephrostomy catheter. The three-dimensional locking loop (thick arrow) can never be fully visualized on the same plane. Note the pelvic dilation and the bubbles in suspension (thin arrows), confirming the adequate placement

Microsurgical ureterotomy

When the patients with ureteroliths were stable enough (based on degree of azotemia, presence of electrolyte derangements or existing comorbidities) and our ophthalmologist was available, the ureteroliths were removed during the same procedure. The ureteroliths were located manually and with intraoperative ultrasound guidance. A partial thickness incision with a #64 Beaver blade was made longitudinally for 3–5 mm. The incision included serosal and muscular layers cranial to the ureterolith and was advanced deeper to involve a small (1–2 mm) portion of mucosa at the level of the ureterolith. The incision was extended using Westcott or Vannas scissors. The ureteroliths were removed with tying forceps and submitted for analysis and culture.

The ureterotomy incision was closed using a simple interrupted suture pattern with 9-0, monofilament Vicryl suture, with spatula needles. The goal was to close only the muscular and serosal layers to prevent future stricture formation, but this could not always be confirmed.

Follow-up

After discharge, the owners were advised to follow-up with a complete blood count, BUN, creatinine, electrolytes, urinalysis with UCS, urinary tract ultrasound with flushing of the SUB device scheduled at 1–2 weeks, 4 weeks, 3, 6, 9 and 12 months postsurgery and then every 3–6 months. The flushing of the SUB device was performed in accordance with the surgical guide provided by Norfolk Vet, ⁵ generally without, or with a mild, sedation.

Statistical analysis

Descriptive statistics were calculated in the usual way. We used a χ² test (or Fisher’s exact test for small cell frequencies) to compare values for categorical variables. For continuous variables, we used a t-test for the comparisons if the outcomes were normally distributed or a Wilcoxon rank sum test if not. We considered a P value ⩽0.05 to be statistically significant. For multivariable models we used general linear models for continuous outcomes and logistic models for categorical/binary outcomes. For survival outcomes, Kaplan–Meier product limit procedures were used with a log-rank test to compare the survival curves with Cox proportional hazards multivariable models for more in-depth analysis of these outcomes. We used SAS statistical software (version 9.4; SAS Institute) for all analysis.

Results

Cats

Twenty-four cats met the inclusion criteria (Table 1). Twenty cats (83%) presented with unilateral and four (17%) presented with bilateral obstructions. Of the 20 cats with unilateral obstructions, two (10%) had a right ureteral obstruction initially, while a left ureteral obstruction was diagnosed on follow-up ultrasound 57 and 232 days later, respectively. Thus, a total of 30 obstructed ureters were corrected during 26 surgeries in 24 cats. The origin of the obstruction was one or multiple ureteroliths in 24 ureters (80%), a suspected ureteral stricture in five ureters (17%) and a suspected blood clot in one ureter (3%). A ureterotomy was performed during the same surgery in 14 (54%) of these 26 surgeries. Median age at the time of (first) SUB device placement was 10.1 years (range 4–16.6). Seventeen (71%) were spayed female cats and seven (29%) were castrated male cats. There were 16 domestic shorthairs (67%) and one cat of each of the following breeds: domestic mediumhair, domestic longhair, Siamese, Persian, Chartreux, Himalayan, Ragdoll and Savannah. Eleven of 24 cats (46%) had a previous history of elevated kidney values with low urine specific gravity. Median packed cell volume (PCV) was 30% (range 12–43) prior to surgery and three patients required a transfusion preoperatively. The median BUN and creatinine on admission were 87 mg/dl (range 19–151) and 7.1 mg/dl (range 1–20), respectively. The median pelvic and ureteral dilation were 16 mm (range 1.5–35) and 4 mm (range 1.3–9), respectively. Two patients had a pelvis dilation ⩽4 mm. UCS collected by cystocentesis on admission was positive in 2/22 cases, isolating Staphylococcus epidermidis in one case and Escherichia coli in the other.

Table 1.

Number of cats, obstructed ureters, surgeries and subcutaneous ureteral bypass (SUB) device placements, with or without ureterotomy

Cats	Underwent only one surgery	22
	Underwent a second surgery (for the contralateral kidney)	2
	Total	24
Obstructed ureters	Right	10
	Left	12
	Bilateral (corrected at the same time)	4
	Total	30*
Surgeries		26
SUB device placement	With ureterotomy	14
	Without ureterotomy	16
	Total	30*

^*Every ureteral obstruction was corrected with the placement of a SUB device and none of the ureters had more than one surgery, so the number of SUB devices equal the total number of obstructed ureters

No significant difference in SUB device side, age at the time of SUB device placement, sex, breed, history of CKD, preoperative pelvic and ureteral dilation was found between the two groups. The only difference was the median creatinine on admission was significantly higher for patients without ureterotomy (10.9 mg/dl vs 5.3 mg/dl; P = 0.022).

Surgery and postoperative care

The median procedure time for the group without and with ureterotomy was 125 mins (range 60–180) and 180 mins (range 100–258), respectively. The median procedure time (time from the beginning to the closure of the laparatomy) was significantly shorter for the group without ureterotomy (P = 0.0053), even though 3/4 simultaneous bilateral SUB device placements were in this group. The median postoperative PCV was 28% (range 13–40), with three patients requiring a blood transfusion after surgery. Of the 12 surgeries without ureterotomy, three patients had a stricture causing the obstruction, three patients had too many calculi, three patients were judged to be too unstable and the ophthalmologist was not available for the three other surgeries.

Review of the medical records revealed that UCS was performed intraoperatively in 12/26 surgeries. Five patients had a culture from the pyelocentesis, three patients had a culture of the ureterolith, and four patients had a culture from the pyelocentesis and of the ureterolith. Three UCSs were positive in three different patients (two from the pyelocentesis and one from the ureterolith), and all isolated an E coli. Interestingly, the UCS obtained via cystocentesis on admission was negative in 2/3 patients. Of the 12 ureteroliths submitted for analyses, all were 100% calcium oxalate monohydrate.

Twenty-three of the 24 cats survived to hospital discharge (96%). The median time of hospitalization after surgery was 4 days (range 1–10) without any significant difference between the two groups (P = 0.5427). The median creatinine on admission, prior to surgery, 24 h postoperatively and at discharge was 7.1 mg/dl, 4.6 mg/dl, 3.45 mg/dl and 2 mg/dl, respectively. The median creatinine between the two groups was different preoperatively (P = 0.02) but not different 24 h and 48 h postoperatively (P = 0.2533 and P = 0.6944, respectively), at discharge (P = 0.3418) and at the 1 month and 3 month follow-up (P = 0.0799 and P = 0.3338, respectively).

Complications

The complications are summarized in Table 2. No difference regarding intraoperative (P = 0.9163), postoperative (P = 0.2475), short-term (P = 0.1339) and long-term complications (P = 0.2921) was found between our two groups.

Table 2.

Complications observed intra- and postoperatively associated with subcutaneous ureteral bypass (SUB) device placement, with or without ureterotomy

Intraoperative (7%)	Postoperative (38%), <1 week after surgery	Short term (30%), 1 week to 1 month after surgery	Long term (52%), >1 month after surgery
• Connection between pelvis and subcapsular space (n = 1/28)	• Transient worsening of kidney values (n = 4/26) • Severe postobstructive diuresis (n = 2/26) • Hematuria (n = 1/26) • Severe anemia (n = 3/26) • Severe coagulopathy and neurologic signs (n = 1/26) • Seroma around the subcutaneous port (n = 1/26)	• UTI (n = 2/21) • Pollakiuria/dysuria without evidence of UTI (n = 2/21) • Difficulty flushing SUB device (n = 1/23) • Subcapsular effusion (n = 1/23) • Persistent seroma around the subcutaneous port (n = 1/23)	• Chronic UTI (n = 4/21) • Pollakiuria/dysuria without evidence of UTI (n = 4/21) • Kink of the tubing at the level of the abdominal wall (n = 2/26)
• Kink of the nephrostomy catheter at the level of the abdominal wall (n = 1/28)
			• Non-patent tubing without evidence of kink (n = 2/26)
			• Perirenal effusion (n = 1/26) • Subcapsular hematoma in the cranial pole of the kidney (n = 1/26) • Displacement of the subcutaneous port (n = 1/26)

UTI = urinary tract infection

Intraoperative complications

Intraoperative complications were described in 2/28 (7%) decompressed ureters available for retrospective evaluation. The first cat had a marked pelvic dilation before surgery (16 mm) and, during the same procedure, the ureterotomy was prioritized before the SUB device was placed, which resulted in renal pelvis decompression and failure to easily and precisely place the nephrostomy catheter within the pelvis. The catheter tip was first noted within the cranial subcapsular region with fluid accumulation around it following intraoperative flushing of the SUB device. The catheter was then pulled back into the pelvis, confirming the location with another flushing of the SUB device. However, a connection between pelvis and subcapsular space was made. The second cat was found to have a kink of the right nephrostomy catheter on postoperative radiographs. The patient was brought directly back to surgery and the kink at the level of the abdominal wall was corrected once the subcutaneous port was moved more cranially. Neither of the two SUB devices placed in a pelvis with minimal dilation (⩽4 mm) had complications.

Postoperative complications

Postoperative complications were described in 10/26 surgeries (38%). The patient that developed hematuria directly after surgery had a small bladder incision performed to remove a cystolith, which was the cause of the hematuria. One cat in the group without ureterotomy that had persistent azotemia became coagulopathic, and required multiple transfusions of packed red blood cells and fresh frozen plasma; the owner elected for euthanasia of the cat 4 days postsurgery owing to deterioration and development of neurologic signs. The cat with the previously described pelvis–subcapsular space communication was one of the patients with transient worsening of kidney values. On recheck ultrasound, subcapsular effusion was present, but no retroperitoneal effusion was seen. Because the patient was clinically unstable, peritoneal dialysis was elected. The exchanges were performed without complication and discontinued after 3 days, when the patient’s kidney values started to improve without dialysate exchanges. The patient was discharged 2 days later with a creatinine almost normalized (1.7 mg/dl; reference <1.6 mg/dl).

Short-term complications

Short-term complications were described in seven of the 21 patients, with 23 decompressed ureters available for retrospective evaluation (30%). The case that had a difficult to flush SUB device was found to have a clog in the cystostomy catheter dislodged 1 month later during surgery after intensive flush of the cystostomy catheter while pinching the nephrostomy catheter. The persistent seroma seen in one patient 4 weeks after surgery was suspected to be secondary to inflammation as it had resolved at the next recheck. The case previously described with a communication between the pelvis and subcapsular space had a persistent subcapsular fluid accumulation without peritoneal effusion.

Long-term complications

Long-term complications were described in 11/21 (52%) patients available for retrospective evaluation. The kinks seen in one nephrostomy and one cystostomy catheter were both at the level of the abdominal wall and were diagnosed 4 and 10 months after SUB device placement, respectively. The suspected non-patent tubing without evidence of kink (n = 2/26) were seen 2 and 3 months after surgery, respectively. The displacement of the subcutaneous port (n = 1/26) required surgical revision 4 months after surgery. Of the four patients with a kink or suspected non-patent tubing, three were from the ureterotomy group. These three cases were not sent back to surgery because they had stable kidney values, and no (n = 2/3) or only mild (n = 1/3) pelvis dilation (2.5 mm) without ureteral dilation. The subcapsular fluid accumulation described previously in the patient with a communication between the pelvis and subcapsular space had resolved on recheck ultrasound 2 months after surgery.

Survival

Four cats were lost to follow-up after discharge from surgery. The median follow-up time of the remaining 20 patients was 881 days (Table 3). The overall median survival time of the 20 cats was 1555 days (range 4–1860) (Figure 4). Seven of these 20 patients had been euthanized by the time of writing (35%). No parameters including signalment, previous history of CKD, BUN on admission, creatinine (on admission, 24 h and 48 h postsurgery, at discharge, and 1 and 3 months postsurgery) had a significant impact on survival. No significant difference was found in survival between our two groups (P = 0.4494) (Figure 5).

Table 3.

Cats with the date and age at their subcutaneous ureteral bypass (SUB) device placement, the number of days of follow-up when data collection was stopped and if they were alive

Cat	Date of (first) SUB	Age at (first) SUB (years)	Alive on 25 June 2019	Survival time until 25 June 2019 (days)
1	12 September 2013	10	N	1555
2	4 October 2013	10	N	1412
3	22 May 2014	8	Y	1860
4	3 June 2014	13	N	1618
5	16 April 2015	5	Y	1531
6	7 May 2015	11	N	956
7	29 May 2015	13	Lost to follow-up
8	18 June 2015	9	Y	1468
9	9 October 2015	12	N	4
10	23 December 2015	11	N	1022
11	2 June 2016	16	N	540
12	21 October 2016	13	Lost to follow-up
13	16 November 2016	12	Y	951
14	20 December 2016	11	Y	917
15	2 March 2017	10	Y	845
16	12 May 2017	14	Y	774
17	28 September 2017	7	Lost to follow-up
18	27 November 2017	9	Y	575
19	6 February 2018	14	Lost to follow-up
20	26 February 2018	4	Y	484
21	27 February 2018	9	Y	483
22	21 May 2018	4	Y	400
23	8 June 2018	16	Y	382
24	15 June 2018	8	Y	375

Median follow-up time was 881 days. Median survival to death of the seven cats that died was 1022 days

N = no; Y = yes

Figure 4.

Kaplan–Meier survival curve for our entire cat population, with or without ureterotomy. Median survival time was 1555 days (4.25 years)

Figure 5.

Kaplan–Meier survival curve for cats with (dotted line) and without (continuous line) ureterotomy. No significant difference was found in survival between the two groups, with or without ureterotomy

All cats but one euthanized 4 days after surgery were euthanized during long-term follow-up, with a minimal survival time of 540 days and a median survival to death of 1022 days (range 4–1618). The cause of death was known in 6/7 patients and all of them had a patent device at the time of euthanasia. One patient developed worsening azotemia and was euthanized 4 days postsurgery owing to coagulopathy and neurologic signs. Two patients with documented pyelonephritis and poor response to antimicrobial therapy were euthanized owing to worsening azotemia. One patient developed worsening azotemia caused by a ureteral obstruction of the contralateral kidney and was euthanized after surgical correction was declined. One patient was euthanized due to congestive heart failure. The final patient was diagnosed with carcinomatosis and transitional cell carcinoma on necropsy.

Discussion

This study presents the successful placement of 30 SUB devices during 26 surgeries for the correction of benign ureteral obstructions in 24 cats. The only patient that was not discharged was euthanized because of persistent azotemia and coagulopathy, despite having a patent device. Intraoperative ultrasound allowed guidance for the successful placement of a nephrostomy catheter in a pelvis as small as 1.5 mm.

Ultrasound guidance has been used in our institution because of logistical difficulties accessing the fluoroscopy unit shared with the large animal clinic. Compared with fluoroscopy, ultrasound guidance has important limitations. Antegrade ureteropyelography is required to definitely confirm the ureteral obstruction, especially in cases of missed or suspected strictures on preoperative ultrasound. In our study, the marked improvement of azotemia following surgery in all but one of the cases was the final confirmation of the suspected diagnosis. For cases with minimal pelvic dilation (⩽4 mm), the contrast and the magnification of the fluoroscopy unit offer better guidance for guidewire and nephrostomy catheter placement. Otherwise, the help of a trained ultrasonographer is needed for successful device placement using ultrasound guidance. Fluoroscopy is also able to highlight a kink before the abdominal closure, which would have avoided a surgical correction in one of our patients. However, ultrasound has the advantage of being more available in specialized clinics and is safer in terms of personnel and patient exposure to radiation but does require individuals having proper training. Recent studies have described the ultrasound evaluation of the renal pelvis in cats with ureteral obstruction.^11–13 Our study showed that ultrasound guidance allowed proper placement in two minimally dilated pelvises (⩽4 mm), where the main difficulty was more the safe insertion of the nephrostomy catheter in the limited space rather than the guidewire or catheter visualization. Facing a marked increase of ureteral obstruction in our feline population, ultrasound guidance may be a successful alternative, between the ideal fluoroscopic method limited to a small number of specialty hospitals and the previously described blinded method, ⁸ which, theoretically, comes with substantially more risk, especially in situations with smaller pelvic dilation. Regardless of the technique used, proper training is required to optimize the outcome of the procedure.

In the past, at our institution, the removal of the ureterolith with microsurgery was thought to be a way to remove a potential site of persistent infection and offer an alternative pathway in the event of a SUB device obstruction. However, this study revealed that only one case had a positive ureterolith culture. Three of the four cases with a non-patent tubing and previous ureterotomy were not sent back for surgical correction as the kidney values were stable and the renal pelvis was not, or only mildly, dilated. However, we are not able to prove fully that the stable kidney values were due to a patent ureter as one patient had a normal contralateral kidney on ultrasound and one patient had a bilateral SUB device with the contralateral SUB device flushing well. We are also not able to prove fully that ureteral patency was restored as no contrast study was performed. The one attempt at flushing contrast through a SUB device did not achieve contrast agent actually passing antegrade down the ureter but instead drained the path of least resistance back through the nephrostomy catheter. The ureterotomy generally made the surgical procedure more complex, prolonged the time of anesthesia significantly and failed to bring any advantage in the general short- and long-term complication rates. Because of these results, a ureterotomy is no longer routinely performed as part of SUB device at our institution and is reserved for unique situations. The median creatinine between our two groups was significantly different preoperatively. This difference is thought to have been a selection bias: ureterotomy was less likely to be performed in the less stable patients with the highest creatinine. The median creatinine between the two groups was not significantly different postoperatively and at the short- and long-term follow-ups. This supports results of prior studies,^7,14 which reported that the preoperative creatinine has no predictive value on the outcome of the surgery.

The analysis and culture of the ureteroliths confirmed that stones in the upper urinary tract of cats continue to be almost always sterile calcium oxalate monohydrate. Interestingly, two cases with ureteroliths had a negative UCS from the cystocentesis collected on admission but grew an E coli in the upper urinary tract (one from the ureterolith and one from the pyelocentesis culture). No prior antimicrobial therapy was initiated when the cystocentesis was collected, but both were on antibiotics at the time of surgery, when the pyelocentesis and ureterotomy were performed. The patients’ cultures from the cystocentesis on admission could have been a false negative, but both cases demonstrated an inactive sediment on urinalysis. The cultures from the upper urinary tract could have been a contamination but both ureterotomy and pyelocentesis were collected in the most sterile manner during surgery.

The most important intraoperative complication in this study could have been avoided if the ureterotomy had been done after the SUB device placement: the pelvis would have been dilated and the parenchyma under pressure to allow an easier placement of the nephrostomy catheter. Interestingly, this communication between pelvis and subcapsular space was not observed on the ultrasound recheck 2 months postoperatively.

The cats with pollakiuria and dysuria had the earlier SUB devices, with a straight cystostomy catheter suspected to cause more bladder irritation. This complication is subjectively less frequent since the replacement of this piece with a locking loop catheter. Finally, the reported UTI in this patient population was based on the presence of pyuria and bacteria on urinalysis and culture results. These findings are now seen as non-clinical bacteriuria if the patient does not present with clinical signs and if the azotemia remains stable. The clinical impression is that these patients can have a persistent bacteriuria for years without having any complications and are also sometimes able to clear the bacteriuria on their own.

The median survival time of 1555 days (4.25 years) in our study was longer than the 827 days previously reported. ⁷ This difference is unlikely to be related to the different guidance for the SUB device placement and may be related to the smaller number of cases. Theoretically, it could be related to a younger cat population or a population with less advanced CKD. However, our cat population with a median age of 10.07 years was not younger than the 9 years previously reported. ⁷ Also, the median creatinine of 2.3 mg/dl (range 1.3–3.2) at 3 months postoperatively was not clinically different than the 2.6 mg/dl previously reported. ⁷ Our results support the findings of previous studies, which have shown that rapid correction of ureteral obstruction is often associated with a good prognosis,^7,14–16 even in cases with previous history of CKD, as seen in almost half of our patients.

Our study has three major limitations. The first limitation is related to its retrospective nature. Because we are a university hospital, different internal medicine residents and one boarded internal medicine specialist managed the cases. However, the follow-up schedule was generally consistent. The second limitation related to our small number of cases may have led to the absence of significant differences. The third limitation was a selection bias for cats receiving ureterotomies that were anesthetically more stable with a lower creatinine.

Conclusions

SUB device placement with intraoperative ultrasound guidance could be an effective alternative for the correction of ureteral obstruction if performed by those trained in the technique. Facing a marked recognition of benign ureteral obstruction in our feline population, this procedure could be a more accessible option that avoids the risks of placing the device blindly when a trained veterinarian does not have access to fluoroscopy. However, microsurgical ureterotomy significantly increased the procedure duration without showing any advantage in improving the outcome and is thus not recommended as part of the standard protocol. Finally, the median survival time of 1555 days (4.25 years) in our population should emphasize the good outcome of this procedure in cases of ureteral obstruction, even in cats with previously diagnosed CKD.