Transurethral cystoscopy in dogs with recurrent urinary tract infections: Retrospective study (2011‐2018)

Abstract Background Urinary tract infections (UTIs) are common in female dogs and recurrent infections often require investigation by transurethral cystoscopy. Hypothesis/Objectives Describe the findings of transurethral cystoscopy in dogs presented for recurrent urinary tract infections (RUTI). Animals Fifty‐three client‐owned dogs with RUTI were included in the study. Methods Retrospective study. Data collected from medical records included signalment, clinical findings, bladder wall culture, cystoscopic, and histopathologic findings. UTI was defined as: presence of compatible clinical signs and at least 2 out of 3 of the following criteria: (1) pyuria, (2) positive urine culture, (3) resolution of clinical signs with antibiotic treatment. Recurrence of UTI was defined as at least 2 episodes of UTI within 6 months or at least 3 or more in 1 year. Results The mean age at presentation was 3.8 years with a majority of female dogs (48/53), 40/48 of which were spayed. Main breeds were Labrador (10/53), Australian Shepherd (4/53), and Miniature Schnauzer (3/53). A hooded vulva was noted in 33/48 of females. Transurethral cystoscopy showed anomalies in 45/53 of cases: mucosal edema (19/53), vestibulovaginal septal remnant (15/48), lymphoid follicles (8/53), short urethra (6/53), and ectopic ureter (5/53). Urine culture at the time of cystoscopy was positive in 13/49. Bladder wall edema and ulceration were the most common findings on histopathology (25/39). Conclusion and Clinical Importance RUTI occurred more frequently in spayed female dogs. Transurethral cystoscopy is useful in the diagnosis and treatment of anomalies in dogs with RUTIs.


| Definition of UTI
A UTI was defined as: presence of compatible clinical signs (ie, hematuria, pollakiuria, dysuria, stranguria) with concurrent evidence of bacterial cystitis based on at least 2 out of 3 of the following criteria: (1) pyuria (defined as >5 WBCs per high power field) 23 ; (2) positive urine culture; (3) resolution of clinical signs with antibiotic treatment.
Recurrence of UTI was defined as at least 2 episodes of UTIs within a 6-month period or at least ≥3 in 1 year.

| Case selection and data collection
Medical records of all dogs presented for RUTI that underwent transurethral cystoscopy at the Veterinary Teaching Hospital of the University of Montreal (CHUV) from January 1, 2011 to December 31, 2018 were retrospectively reviewed. Data collected from medical records included signalment, clinical presentation, physical examination findings, clinicopathologic/histopathologic, and cystoscopic findings. Age was approximated to the closest half year at the time of the first visit.
Breeds were recorded and breeds with 3 or more (5%) individuals were compared to the reference population. Cases were excluded if they presented with subclinical bacteriuria or if they underwent a percutaneous cystolithotomy. The presence of a hooded vulva was defined on physical examination by a board-certified internist. It was subjectively defined as the presence of a skin fold over the dorsal aspect of the vulva with less than approximately 50% of the vulva being visualized without any manipulation of the surrounding skin.

| Cystoscopic procedure
Transurethral cystoscopy was routinely performed by 1 of 2 board-certified internists (MD or CV). Dogs were anesthetized and placed in dorsal recumbency. Vulva, preputial, and perineal regions were clipped, aseptically prepared and a sterile drape was applied. A flexible cystoscope was used in male dogs (Storz© flexible 11278AUI/CE 0123 for cystoscopy before   August 2018 then Storz© flexible Flex XC 11278/VSU #series SN25921, KARL STORZ Endoscopy Canada Ltd., Mississauga, Canada) and rigid cystoscopy was performed in female dogs (Wolf© rigid 8616.401/10.5 Fr used for female weighing less than 15 kg or Wolf© rigid 8642.403/14 Fr for heavier females >15 kg). The entire lower urogenital tract was systematically evaluated (prepuce, clitoral fossa, vestibule, urethra, bladder, ureteral papilla, vagina). When the urethra seemed subjectively short in regard to the dog size, the urethral length was measured from the proximal urethra (urethrovesicular junction) to the distal urethra (urethrovestibular junction).
It was considered short when the length was inferior to 7 cm. We chose this length based on previous studies that measured mean normal urethral length of 8 cm. 18,24 Anatomic anomalies were noted and corrected during the same procedure (eg, ectopic ureters, vestibulovaginal septal remnant, urethral septum) using a Holmium:YAG or diode laser.

| Bacterial culture and histopathologic findings
Urine samples for bacterial culture were obtained through the working channel of the cystoscope immediately upon entering the bladder.
Bladder wall samples for histopathology were obtained at the end of the procedure using sterile cup biopsy forceps (Endoscopy Support Services 5 Fr x 65 cm, Brewster, NY) introduced through the working channel of the cystoscope. The area of sampling was chosen by the operator where the bladder wall was noted to be macroscopically abnormal while avoiding bladder trigone and the ureterovesical junction. First administration of peri-operative antibiotics (Ampicillin 22 mg/kg q90min) was given at the time of introduction of the cystoscope.

| Control population
In order to compare the study population to the overall population of dogs presented to the CHUV during the same study period, medical records of all dogs presented to the hospital were collected. Breed, sex, neuter status, and age at presentation were retrospectively obtained. The original visit was the only visit considered. Age was approximated to the closest half year at the time of the first visit. For dogs <3 months of age, 0 was assigned.

| Statistical analyses
Descriptive statistics were used for signalment, clinical data, laboratory, and cystoscopic findings. All statistical analyses were performed with a commercial software program (SAS v.9.3. Cary, NC). An exact chi-square test was used to compare prevalence of breed and sex distribution between the 2 study groups. An exact chi-square test was also performed in order to evaluate a potential association between the administration of antibiotics before the procedure and the results of bladder wall culture. An unequal variance t test was used to compare the mean age between the 2 populations. A P value of less than .05 was considered statistically significant for all comparisons.

| Study and reference population
A total of 53 dogs met the inclusion criteria and were included in the study population. This population included 48 (91%) female dogs, 40 of which (83%) were spayed females and 5 (10%) male dogs, 2 (40%) of which were neutered. The mean age at presentation of the study population was 3.64 ± 3.14 years with a median age of The mean age was significantly lower in the study population than in the control population (P = .002). The reproductive status was also significantly different, with proportionately more intact females noted in the study population (P < .001). Labrador Retrievers (P = .05), Australian Shepherds (P = .01), and Miniature Schnauzers (P = .01) were overrepresented in the study population. RUTI were present in 0.25% (53/21683) of the general population during our study period.
A total of 137 episodes of suspected UTI were reported in the medical file for the entire study population before presentation; each dog had at least 2 or 3 episodes depending on the time of recurrence (more or less than 6 months). Descriptions of each episode are described in Table 1.
Ureterovesical junction stenosis was diagnosed in 3 dogs, 1 was bilateral, 1 was unilateral, and 1 was unilateral and associated with EU. In dogs with EU (n = 5), 4/5 were unilateral and 4/5 were noted to also have a vestibulovaginal septal remnant, all of them were presented with a T A B L E 1 RUTI was defined by as at least 2 out of 3 criteria (1) compatible clinical signs; (2)   Two cases of urethral stricture were reported but with variable severity.
The first dog with urethral stricture was a neutered male. The stricture prevented the cystoscope from passing beyond the distal pelvic urethra.
A voiding cystourethrogram was performed and revealed a 1.5 cm long stricture that required balloon dilatation in order to advance the cystoscope beyond the stricture. The second dog did not require urethral dilatation (see Figure 1).
In dogs with anomalies, 12/45 (27%) had nonspecific findings such as mucosal edema, prominent lymphoid follicles, or both of them.
Of the study population, 28
Ulceration and mucosal edema were most commonly reported in 24/38 dogs (63%). No relevant anomalies were found in 11/38 dogs.
Other findings included transitional cell carcinoma (1/38), Single bladder stone One of the main nonspecific findings was prominent lymphoid follicles. Lymphoid follicles are not detected by radiographs nor CT scan. 18 Though the clinical relevance of this finding remains unknown, cystoscopy is superior to other imaging modalities in their identification. Dogs with EU commonly have concurrent VVSR in 80% (4/5) which is similar to previous reports (83%). 19 We observed a high prevalence of hooded vulva (69%); however, episioplasty was performed in only 21%. Episioplasty reduces UTIs in 84% to 100% of dogs, 29,30 likely by reducing bacterial population in the perivulvar region and therefore in the vulva and vestibule. We chose not to systematically perform episioplasty in dogs that were obese, had concurrent anomalies or were young intact females as these conditions might have predisposed to bacteria in the region. Alternatively, weight loss and allowing a female to come into heat is recommended. 31 Episioplasty surgery was considered if UTIs continued to recur despite passing a heat cycle and undergoing correction of cystoscopically identified anomalies.
Our most frequent finding on histopathology was ulceration and mucosal edema (63%). We believe this finding could be related to chronic inflammation from the RUTI and filling the bladder with saline during the procedure that might alter the bladder epithelium. Further studies are required to differentiate these 2 hypotheses. If lesions are induced by filling the bladder with saline, biopsy sampling should be performed at the beginning of the procedure in order to minimize these changes. This can however limit examination because of bleeding and prolong the procedure time and thus anesthesia time. Biopsies were performed at the end of the procedure at our institution. Histopathology was nonspecific except for 1 case of transitional cell carcinoma.
The most common bacterial isolate was E. coli. E. coli is commonly cultured from urine of dogs with UTIs with a reported prevalence of 30% to 50% of urine cultures submitted. 13,19 The most prevalent bacteria after E. coli was Staphylococcus and Enterococcus as in 3 retrospective case studies. 2,27,32 In our culture results, a single pathogen was present in 92% of dogs, with 2 pathogens isolated in only 1 urine sample. Sensitivity testing revealed that 7 out of the 12 positive urine cultures were resistant to at least 1 commonly administered oral antibiotic. In 2 dogs, urine culture was negative; however, bladder mucosal biopsy was positive which is in disagreement with previous reports. 13,32,33 This discrepancy might be explained by the invasion of superficial bladder epithelial cells and submucosa by bacteria, 34 even if the isolated bacteria in our cases were not E. coli, which has been previously described in host cells with type-1 fimbriae-mediated invasion. Even if this was present in a small number of dogs, we recommend urine culture and bladder mucosal biopsy culture in dogs undergoing cystoscopy for RUTI.
This study is limited by its retrospective nature and the small number of cases. The diagnosis and treatment of UTIs initially done by referring veterinarians was not standardized. It was not possible for us to differentiate relapse and reinfection as urine culture was not consistently repeated for every episode of UTI and clonality testing was not performed. 35 Unfortunately, outcome and long-term follow-up were unable to be assessed given the retrospective nature of the study.
In conclusion, RUTI affect mainly spayed females (83%) with a mean age of 3.78 years. Labrador Retrievers, Australian Shepherds, and Miniature Schnauzers were over-represented in our study. Transurethral cystoscopy is an effective, safe, and minimally invasive procedure to investigate RUTI. Anomalies were found in the majority of dogs (85%-45/53) with a correction performed at the time of cystoscopy in 53% (28/43).