Delayed Urethral Obstruction After Uterine Torsion in a Pregnant Dog

Introduction

Most complications of uterine torsion occur acutely and are due to vascular compromise and hypovolemic and/or septic shock. In dogs, the recommended treatment of uterine torsion is ovariohysterectomy and aggressive supportive care.¹ Urethral obstruction can also be life threatening due to severe electrolyte and acid-base disturbances and hemodynamic instability.² The actual incidence of uretheral obstruction is not reported; however, complete urethral obstruction is uncommon in the female dog, most likely because of a relatively short, wide urethra. The purpose of this report is to describe a complete urethral obstruction in a female dog that developed 17 days after partial obstruction of urine outflow caused by torsion of the nongravid uterine horn during pregnancy. The complete obstruction was caused by an exuberant fibroplastic response to ischemic changes. Urethral obstruction is a newly reported complication of uterine torsion in the dog.

Case Report

A 4 yr old, 4.5 kg, pregnant female shih tzu had a transient episode of bloody vulvar discharge approximately 5 wk into gestation. Seventeen days later, the discharge recurred, and one or more contractions were seen without subsequent parturition. The dog was presented to the primary care veterinarian the following day. The dog was painful and dysuric, but able to urinate. Blood analyses revealed azotemia (blood urea nitrogen was 108 mg/dL; reference range, 7–25 mg/dL, creatinine was 4.0 mg/dL; reference range, 0.3–1.4 mg/dL) and leukocytosis (17,000/μL; reference range, 600–16,000/μL). Abdominal radiographs revealed a distended urinary bladder and one large fetus. No signs of fetal death or uterine malposition were evident. At surgery, the nongravid uterine horn was twisted approximately 180 degrees along its long axis. It was lying over the trigone region of the bladder, restricting urine outflow from the bladder. The malpositioned uterine horn and the urinary bladder were markedly hyperemic and had evidence of intramural hemorrhage. The bladder was also extremely dilated. A cesarean section was performed, followed by ovariohysterectomy. The nonviable fetus appeared premature and had a cleft palate. The ovarian pedicles and uterine body were ligated with 3-0 polydioxanone. Postoperatively, the dog had blood-tinged urine and pollakiuria. Both resolved over the next several days. Enrofloxacin (5 mg/kg per os q 12 hr for 1 wk) was dispensed. At a recheck examination for suture removal 16 days postoperatively, the dog appeared to have fully recovered.

On the evening of the seventeenth day after surgery for correction of the uterine torsion, the dog was presented to an emergency clinic because of discomfort, pollakiuria, and owner uncertainty about urine production. The bladder was greatly distended on abdominal palpation and radiographs. The cause of the urinary obstruction was unclear. Abnormalities on a complete blood count (CBC), serum chemistry panel, and urinalysis included a normocytic, normochromic anemia (packed cell volume [PCV] was 26.7%; reference range, 31.0–55.0%), decreased globulins (1.9 g/dL; reference range, 2.3–5.2 g/dL), and pyuria (3+ WBC), respectively. Activated partial thromboplastin time was elevated (115 sec; reference range, 71–102 sec), and the prothrombin time was within the reference range. Cystocentesis was performed to relieve bladder distension. Additional treatments included IV fluids, cefazolin, dexamethasone sodium phosphate, and butorphanol. The dog dribbled small amounts of urine overnight. The following morning, the bladder was small. Attempts at urethral catheterization were blocked approximately 1.5 cm cranial to the urethral papilla. When the bladder reached a moderate size, it was manually expressed, although with resistance that was greater than normal. The dog intermittently dribbled small amounts of urine and was sent home after urinating a more normal amount later in the day; however, emergency cystocentesis was required that evening. The dog was prescribed enrofloxacin as previously described, and the case was referred to the Washington State University Veterinary Teaching Hospital.

On presentation to the Washington State University Veterinary Teaching Hospital 19 days after the ovariohysterectomy for uterine torsion, the dog was bright, alert, and weighed 3.3 kg. Abdominal palpation revealed a moderately distended bladder with a tubular mass dorsocaudal to the bladder neck. A 20 mm × 5 mm firm, nonpainful, tubular structure oriented along the path of the urethra, cranial to the pelvic inlet and caudal to the palpable urinary bladder, was identified on rectal examination. A separate urethra was not palpable. A CBC revealed a mild, normocytic, normochromic anemia (PCV was 33%; reference range, 38–59%). Serum biochemical analyses were within reference ranges. Urine specific gravity was 1.024. Urinalysis showed a moderate amount of hematuria and trace proteinuria with no bacteria or crystals evident on cytology. Abdominal ultrasound identified a homogenous soft tissue structure just cranial to the pelvic inlet in the area of the urethra (Figure 1). The urethra was dilated cranial to the soft tissue mass. No evidence of urinary bladder wall thickening or intraluminal abnormalities was seen.

View Figure

• Download as Powerpoint
• Download Figure

The dog was premedicated with acepromazine^a (0.03 mg/kg IV) and butorphanol^b (0.2 mg/kg IV). General anesthesia was induced with propofol^c (5 mg/kg IV) and maintained with isoflurane^d in 100% oxygen. The urethral papilla was visualized with the aid of a vaginal speculum, and a lubricated 6 Fr Foley urinary catheter^e, a 5 Fr red rubber tube^f, and a 3.5 Fr polypropylene tomcat catheter^g were each sequentially advanced into the urethra. In each case, retrograde catheter advancement was blocked approximately 1.5 cm into the urethra. Manual expression of the bladder failed to produce any urine flow. A diagnosis of complete urethral obstruction was made. A temporary cystostomy tube was placed using a 6 Fr Foley catheter^e. The balloon was distended with 2.5 mL of sterile saline and the catheter was connected to a closed collection system. The dog was hospitalized over the weekend pending further diagnostic imaging. Cefazolin^h (22.7 mg/kg IV q 8 hr), IV fluids (2.7 mL/kg/hr Normosol Rⁱ), and hydromorphone^j (0.09 mg/kg intramuscularly [IM] q 4–6 hr as required) were administered.

Normograde and retrograde contrast cystourethrograms and a retrograde contrast vaginogram were performed using an iodinated contrast agent (150mg/mL)^k and fluoroscopic imaging with the dog under general anesthesia. Approximately 2.5 cm of the proximal pelvic urethra did not distend with contrast (Figures 2A, B). It could not be determined whether this was a mural or extramural filling defect. The tubular structure felt per rectum and seen ultrasonographically was immediately cranial to the level of the urethral obstruction and did not fill with contrast. The owner took the dog home and evacuated the bladder via the cystostomy tube q 6–8 hr while deciding on whether to proceed with surgery.

View Figure

• Download as Powerpoint
• Download Figure

Nine days following cystostomy tube placement (29 days postovariohysterectomy), the dog was presented to an emergency clinic for hematuria. Enrofloxacin^l was prescribed as before for presumptive lower urinary tract infection. Two days later, the dog was presented again to the Washington State University Veterinary Teaching Hospital for surgical correction of the urethral obstruction. The cystostomy tube was patent. Rectal examination findings were unchanged. A CBC revealed a neutrophilia (13,552/μL; reference range, 3,000–7,100/μL), with resolution of the previous anemia (PCV was 40%). Serum biochemical analyses were within reference ranges. Urinalysis found hematuria, mild proteinuria, pyuria (5–8 WBC/high-power field), specific gravity of 1.023, and no evidence of bacteria. Thoracic radiographs were within normal limits. An echocardiogram was performed at this time because profound bradycardia had occurred under anesthesia during placement of the temporary cystostomy tube. Mild pulmonic regurgitation was evident.

Twelve days following cystostomy tube placement (32 days postovariohysterectomy), the dog was premedicated with acepromazine^a (0.03 mg/kg IM), hydromorphone^j (0.1 mg/kg IM), and glycopyrrolate^m (0.01 mg/kg IM). Lactated Ringer's solutionⁿ was administered perioperatively (10 mL/kg/hr IV). General anesthesia was induced and maintained as before. A morphine^o epidural (0.1 mg/kg) was also performed. IV cefazolin^h (22 mg/kg q 90 min during anesthesia) was administered. A ventral midline laparotomy was performed from the umbilicus to the pubis. The bladder was adhered to the right abdominal wall via the cystostomy tube. The left ovarian pedicle was adhered to the left cranial bladder wall and the uterine stump was adhered to the dorsal bladder wall. The uterine stump was of normal size and appearance and was not involved in the urethral obstruction. A firm mass measuring 15 mm in length was palpated in the pelvic canal and appeared to surround the pelvic urethra. The urinary bladder was isolated with moist laparotomy pads. Stay sutures were placed in the apex of the bladder to retract it cranially. A 5 Fr red rubber tube^f was introduced through a small cystotomy. The catheter could be advanced approximately 25 mm into the proximal urethra, but could not be advanced beyond the observed mass. The mass appeared to consist of fibrous and granulomatous tissue. Blunt dissection and bipolar electrocautery were used in an attempt to dissect the mass away from the ventral urethral wall. Despite the dissection, the catheter could not be advanced past the mass. The urethra was transected just cranial to the mass. The mass was excised along with the associated pelvic urethra and cranial vagina. The excised tissue was submitted for histopathology. The red rubber catheter was removed and the cystotomy was closed with a pursestring suture of 3-0 polydioxanone^p. A left paramedian prepubic urethrostomy was performed as previously described.³ The abdominal incision was closed routinely, and anesthetic recovery was uneventful.

IV fluids were administered for 48 hr postoperatively. Voluntary urination through the urethrostomy site was observed 24 hr after surgery. The cystostomy tube was removed 2 days posturethrostomy. At discharge, 6 days after urethrostomy (38 days postovariohysterectomy), occasional involuntary urine dribbling from the stoma was observed. Two years after the urethrostomy, the dog had good urinary function with the exception of an occasional episode of incontinence when stretching.

The histologic diagnosis was severe, multifocal, necrosuppurative urethritis with fibroplasia, fibrosis, and cellulitis (Figure 3). Mass formation was attributable to an exuberant fibroplastic response to areas of necrosis and suppuration secondary to ischemia. Vaginal tissue was histologically normal.

View Figure

• Download as Powerpoint
• Download Figure

Discussion

Uterine torsion is an infrequent disorder in the bitch. It occurs more commonly in the pregnant animal.⁴ Torsion may involve one or both uterine horns. The horns can twist about their own axes or each other, and may include the uterine body.⁵ Disorders associated with canine uterine torsion include dystocia, pyometra, septic shock, peritonitis, coagulopathy, and death.^4–6 No reports of urethral obstruction as a sequela of uterine torsion were found in a review of the veterinary literature.

Causes of urethral obstruction reported in female dogs include: neoplasia (e.g., transitional cell carcinoma, squamous cell carcinoma, leiomyoma, or myxosarcoma); calculi; blood clots; bladder entrapment in perineal, inguinal, or other hernias; suture entrapment during surgery (e.g., ovariohysterectomy); lack of coordination between urethral relaxation and detrusor contraction (reflex dyssynergia); proliferative urethritis; and trauma.⁷ In the case described herein, no evidence of urethral or periurethral neoplasia was seen on histopathology. No evidence of urolithiasis was seen in the urine sediment, ultrasound, urinary contrast radiographs, or histopathology. A blood clot in the urethral lumen would not stimulate the fibroplastic response that produced the mass extending into the periurethral tissues. No evidence of bladder entrapment or external force trauma was identified in the history or with physical examination, ultrasound, contrast radiography, or surgical exploration of the abdomen. Suture entrapment of the urethra was ruled out because contrast studies showed that the lumen of the uterine stump extended cranial to the area of urethral compromise, no association was seen between the sites of uterine stump ligation and urethral obstruction during the urethral surgery, and histopathology did not identify any suture material in the resected tissue. Uterine torsion and/or surgical manipulation could have caused reflex dyssynergia by compromising the innervation of the bladder and urethral sphincter.⁸ The presence of an actual mechanical obstruction in the urethra, the inability of the clinician to express the dog's bladder as the obstruction progressed, and the dog's ability to urinate normally after recovery from the ovariohysterectomy and from the urethrostomy all suggest that any functional obstruction was transient and in addition to the mechanical obstruction. Proliferative or granulomatous urethritis is characterized by proliferation of urethral mucosa and a lymphoplasmacytic or neutrophilic inflammatory infiltrate.^9–12 Granulation tissue, fibrosis, and necrosis predominated in the urethra of the dog described in this report. A resolving proliferative urethritis could not be completely ruled out, but the histologic changes did not support this diagnosis.

The episode of urinary obstruction, which became clinically apparent 17 days after the uterine torsion, was most likely due to a fibroblastic response secondary to urethral ischemia caused by uterine horn torsion. At the time of the cesarian section, the bladder was hyperemic and had intramural hemorrhage, signs that vascularity was compromised. This could be due to pressure on the intramural vessels by the large volume of retained urine and/or compression of vesicular vessels before they entered the bladder. At the same time, the nongravid uterine horn, which was also very congested, was significantly displaced and appeared to be compromising outflow from the bladder. Vascular compromise to this uterine horn was most likely due to obstruction of the uterine artery because there was no luminal distension of the horn. The vaginal artery, a division of the internal pudendal artery, gives off branches to the caudal bladder, urethra, and uterus.¹³ A 180° caudal reflection of a uterine horn could “kink” the distal vaginal artery, compromising the blood supply to all of the aforementioned structures. The weight of the fetus in the normally positioned uterine horn may have augmented the abnormal force exerted on the vasculature by the malpositioned uterine horn. At the time of uterine torsion, the urinary bladder distension seen on preoperative radiographs, the significant azotemia (although dehydration could not be ruled out due to lack of data), the distension and discoloration of the bladder seen at surgery, the position of the twisted uterine horn, and the ability of the dog to urinate normally after recovery from ovariohysterectomy were all consistent with the malpositioned uterine horn physically obstructing outflow from the bladder. This may have been due to compression of the urethral lumen by the malpositioned horn within the limited confines of the pelvic canal. This compression could have caused additional intramural vascular compromise and/or trauma to the urethra.

The urethral branch of the vaginal artery is the main source of the urethra's blood supply. Lack of redundancy in the urethral vasculature could explain persistence of ischemia and the subsequent necrosis, inflammation, and fibroplasia that eventually led to delayed urethral obstruction. Collagen deposition typically starts 3–5 days after injury and peaks about 10–14 days later.¹⁴ This timeframe is consistent with the onset of clinical signs of urethral obstruction 17 days after uterine torsion in this dog. Unlike the urethra, the bladder receives blood from a branch of the umbilical artery (the cranial vesicular artery) in addition to the vaginal artery. This additional blood supply could aid in recovery of the bladder if the caudal vesicular branch of the vaginal artery is compromised.

Clinical signs of urethral obstruction postuterine torsion were reported to have begun acutely in this dog. Luminal narrowing of 60% or more appears to be necessary before dogs show clinical evidence of urethral stricture.¹⁵ Thus, it is likely that on day 17, luminal compromise by the fibroplastic process finally progressed to a point where the dog could no longer effectively urinate. It is also possible that the dog's owners may have overlooked initial signs of dysuria, believing they were part of the normal postoperative course of recovery.

The urethral mass was discovered by rectal palpation, which underlies the importance of this procedure as part of a full physical examination. Management of urethral obstruction depends on the degree of obstruction and the underlying cause. Initial emergency management involves establishing a means of urine egress. Caution is advised when performing cystocentesis in an obstructed bladder, because rupture can occur if the integrity of the bladder wall is compromised.² Because urethral catheterization attempts were unsuccessful in this dog, a temporary percutaneous cystostomy tube was placed to bypass the obstruction and allow further diagnostic imaging.

Options for definitive resolution of irresolvable urethral obstruction include: dilation or stenting; urethral resection and anastomosis; and diversionary techniques. Balloon dilation has been used to relieve urethral strictures in 3 dogs and urethral stents have successfully palliated malignant urethral obstruction in 12 dogs.^16–19 Neither balloon dilation nor urethral stenting was considered a viable option in this dog because of the complete, nondistensible obstruction of the urethra. End-to-end urethral anastomosis was not possible in this case due to resection of the entire distal urethra, including the urethral papilla. Vaginourethroplasty has been successfully used to resect distal urethral lesions, including those that invade the ventral wall of the vagina.²⁰ The radiographic appearance in this dog suggested that vaginourethroplasty might have been accomplished, but the surgeon's intraoperative assessment after resection of the distal urethra and associated vagina and cervix was that tension on the anastomosis would have been excessive in this dog. Excessive tension at an anastomotic site will separate the urethral ends and cause stricture formation.²¹ It is possible that the length of obstructed urethra had increased during the 12 days that passed between the contrast study and the urethrostomy. In the future, replacement urethral segments engineered from cell-seeded, tubularized grafts of extracellular matrix may provide an additional option for treatment when direct anastomosis is not possible.²²

Permanent percutaneous cystostomy tubes have been used to manage urethral obstruction in dogs with transitional cell carcinoma or uterine carcinoma.^23,24 Common complications associated with cystostomy tubes are recurrent or persistent urinary tract infection and peristomal leakage of urine.²⁴ Long-term management of a cystostomy tube is feasible in the dog, especially when a low-profile tube is used; however, it was not considered the best choice for this relatively young animal.²⁴ Complications associated with diversion of the urethra into the colon, including hyperammonemia, electrolyte disturbances, and ascending urinary tract infections, made this technique undesirable.²⁵ Permanent cystostomy was also considered a poor choice because of urinary incontinence and risk of ascending infection.²⁶ Prepubic urethrostomy, the procedure used in the dog reported herein, can provide long-term resolution of urethral obstruction with maintenance of urinary continence.²⁶ Dogs with a prepubic urethrostomy should be monitored for urine scalding of peristomal skin, urinary tract infection, and stomal stricture.²⁶

Conclusion

A rectal examination can provide valuable information about the cause of urinary tract obstruction and should be a routine part of examination in dogs with this condition. Delayed urethral obstruction secondary to ischemic damage and fibroplasia is a potential complication of uterine torsion in the dog. Signs of dysuria after uterine torsion may more commonly be secondary to a urinary tract infection; however, dogs should be monitored closely for evidence of urinary outflow obstruction. Prepubic urethrostomy can provide a satisfactory resolution to nonmalignant distal urethral obstruction in the female dog.