Successful treatment of Corynebacterium urealyticum encrusting cystitis with systemic and intravesical antimicrobial therapy

Abstract

A 6-year-old Saint Bernard dog was diagnosed with encrusting cystitis caused by Corynebacterium urealyticum. The infection persisted despite the prolonged use of antimicrobials and surgical debridement of the urinary bladder. Resolution occurred following intravenous vancomycin, urine acidification, and intravesical gentamicin. The challenges involved in the treatment of encrusting cystitis are described.

Encrusting cystitis caused by Corynebacterium urealyticum is infrequently reported in veterinary patients. Corynebacterium urealyticum, an aerobic, Gram-positive bacillus, has potent urease activity, leading to the formation of excessive ammonia, urine alkalinization, and ammonium magnesium phosphate (struvite) deposition on the bladder mucosa (1,2). While this organism is not unique in its urease activity (other bacteria such as Proteus mirabilis and Staphylococcus pseudintermedius also produce this enzyme), the development of encrusting lesions within the urinary tract is more commonly associated with C. urealyticum (2,3). The organism strongly adheres to uroepithelial cells and can readily penetrate the bladder mucosa and ascend the ureters to colonize the renal pelvis (3,4). Unlike most pathogens of the urinary tract, C. urealyticum is slow growing; consequently isolates may be missed if incubation is not extended. Observation of urine cultures on enriched media such as blood agar for 72 h or longer may be required (1).

Corynebacterium urealyticum is typically multi-drug resistant making its eradication challenging. Various treatment modalities have been described in human and veterinary patients including antimicrobial therapy, urine acidification, and removal of calcified plaques via cystoscopy or surgery (1,3,5). The patient in this report had a persistent C. urealyticum infection over 19 mo despite the prolonged use of antimicrobials and surgical debridement of bladder encrustations. Resolution was ultimately attained with the combined administration of an intravenous glycopeptide, urine acidifier, and intravesical antimicrobial therapy, which has not previously been reported in veterinary medicine for treatment of encrusting cystitis.

Case description

A 6-year-old castrated male Saint Bernard dog was presented for evaluation of urine leakage from the perineal region and recurrent urinary tract infections (UTI). On physical examination a 2 to 3 mm, circular skin defect was present on the midline, approximately 10 cm ventral to the anus. Urine was observed dripping from this orifice.

Initial diagnostic investigations consisted of a complete blood (cell) count (CBC) and chemistry panel, urinalysis, urine culture with antimicrobial susceptibility testing, and a positive contrast cystourethrogram and fistulogram. The CBC and chemistry panel were unremarkable. Urinalysis revealed a urine pH of 8, specific gravity of 1.027, 8 to 10 neutrophils/high power field (hpf), and 2+ triple phosphate crystals. Large numbers of P. mirabilis, S. pseudintermedius, and Enterococcus spp. were isolated from urine collected via cystocentesis, following routine aerobic bacterial culture and identification (6–8). Amoxicillin (Amoxil; Novopharm, Toronto, Ontario), 22 mg/kg body weight (BW), PO, q12h was prescribed for treatment of the UTI based on in vitro antimicrobial susceptibility tests. Contrast studies revealed that the skin defect communicated with the membranous urethra via a sac-like tract. Based on these findings, a urethroperineal fistula was suspected. An abdominal ultrasound was recommended to further evaluate the urinary tract, followed by surgical resection of the fistula. However, additional diagnostics and treatment were not performed due to the owner’s financial constraints.

Four months later the dog was presented for gross hematuria and pollakiuria. A CBC and chemistry panel were unremarkable aside from a mild hypoalbuminemia (29 g/L, reference interval: 30 to 36 g/L). Fistulectomy was again recommended. The owner consented to surgery, which was subsequently performed. Histopathologic analysis of the excised tissue revealed a tube lined with squamous epithelial cells, with multifocal necrosis and marked, mixed inflammation, confirming the diagnosis of a urethroperineal fistula.

Urine collected via urethral catheterization at the time of surgery was submitted for culture and susceptibility testing. Amoxicillin and clavulanic acid (Clavamox; Pfizer Animal Health, Kirkland, Quebec), 13.75 mg/kg BW, PO, q12h was started while awaiting the results. A slow growing, small Gram-positive bacillus was isolated in large numbers and pure culture after 48-hour incubation on Columbia agar with 5% sheep blood. The isolate was identified by API Coryne V3.0 (BioMérieux) as Corynebacterium urealyticum (98.8% probability and API Profile number 2001004) and by matrix assisted laser desorption-time of flight mass spectrometry with BioTyper™ identification software (MALDI Biotyper: Bruker Daltonics, Milton, Ontario) with an organism best match score value of 1.94 (match score range for genus identification is 1.7 to 1.99; match score range for species identification is 2.0 to 3.0). The isolate was strongly urease positive. In vitro disk diffusion antimicrobial susceptibility testing was performed using Mueller-Hinton agar with 5% lysed horse blood based on CLSI guidelines (8). Although this is not a validated methodology for this particular organism, the C. urealyticum isolate was shown to be resistant to amoxicillin/clavulanic acid, ampicillin, cephalexin, cefovecin, clindamycin, erythromycin, and penicillin but was susceptible to enrofloxacin, chloramphenicol, doxycycline, gentamicin, and amikacin. Amoxicillin/clavulanic acid was replaced with enrofloxacin (Baytril; Bayer, Toronto, Ontario), 5 mg/kg BW, PO, q24h for 4 wk. The owner was advised to return with the dog 5 to 7 d after completion of the antibiotics for a urine culture.

Five weeks later, the surgical site had fully healed without complication, but the patient was experiencing persistent hematuria. Urine was collected via cystocentesis. Urinalysis revealed a urine pH of 9, specific gravity of 1.024, 5 to 6 neutrophils/hpf, 2 to 4 red blood cells/hpf, and 3+ proteinuria. Culture and antimicrobial susceptibility again revealed growth of C. urealyticum, which was now resistant to enrofloxacin but susceptible to chloramphenicol, doxycycline, gentamicin, and amikacin. Treatment with doxycycline (Apo-Doxy; Apotex, Weston, Ontario), 5.5 mg/kg BW, PO, q12h for 6 wk was initiated.

Two weeks after initiation of doxycycline therapy, ultrasonographic evaluation of the urinary system revealed severe thickening of the wall of the urinary bladder, measuring 1.4 cm throughout the circumference. The mucosal surface of the urinary bladder and proximal urethra were hyperechoic and irregular. Hyperechoic foci of various sizes were suspended within the lumen of the bladder and focal corticomedullary mineralization of the kidneys was apparent bilaterally. Urine was collected via cystocentesis and C. urealyticum was again cultured. Based on the persistent infection and the severity of mucosal mineralization, surgical debridement of the urinary bladder was recommended; however, the owner declined surgical therapy. The doxycycline dose was increased to 10 mg/kg BW, PO, q12h and the owner was advised to return the dog in 10 to 14 d for a urinalysis and urine culture.

The dog was presented 2 months later with weight loss of 10 kg, lethargy, severe hematuria, and dysuria. A urinalysis revealed a urine pH of 9, specific gravity of 1.018, 6 to 10 neutrophils/hpf, 4 to 6 red blood cells/hpf, 3+ proteinuria, and 2+ triple phosphate crystals. At this time the owner consented to surgical debridement of the urinary bladder. The patient was premedicated with acepromazine (Atravet; Boehringer Ingelheim Canada, Burlington, Ontario), 0.02 mg/kg BW, IM and hydromorphone (Hydromorphone Hydrochloride; Sandoz Canada, Boucherville, Quebec), 0.05 mg/kg BW, IV. Anesthesia was induced with propofol (Diprivan; AstraZeneca, St. Laurent, Quebec, Canada) 3 mg/kg BW, IV, and maintained with isoflurane (Pharmaceutical Partners of Canada, Richmond Hill, Ontario) in 100% oxygen. During surgery the patient received fentanyl (Fentanyl Citrate; Sandoz Canada), 5 μg/kg BW per hour, as a constant rate infusion. A standard celiotomy and ventral cystotomy were performed. The lumen of the bladder was exposed, revealing copious yellow sludge as well as multifocal black-green plaques on the mucosal surface (Figure 1). A sample of the urinary bladder was obtained for histopathologic analysis, and a swab sample of the bladder mucosa was obtained for culture. The mucosal plaques were debrided with a number 15 blade. The bladder wall was closed in a simple continuous pattern with 3-0 PDS (Ethicon; Johnson & Johnson Medical, Markham, Ontario), and the abdomen was closed routinely. Histologic examination of the bladder mucosa revealed severe, suppurative, necrotizing, ulcerative cystitis with mineralization of the mucosal surface. A Gram stain showed aggregates of short, Gram-positive bacilli. The swab of the bladder mucosa yielded heavy growth of C. urealyticum. Antimicrobial susceptibility testing was performed by an automated broth microdilution method to determine minimal inhibitory concentration (MIC) using cation-adjusted Mueller-Hinton broth containing 5% lysed horse blood (Sensititre System, Trek Diagnostic Systems, companion animal MIC plate, Oakwood Village, Ohio, USA); this being a validated testing methodology for Corynebacterium species (9). Post-operative treatment consisted of the administration of an isotonic crystalloid solution (Plasmalyte-148; Baxter, Mississauga, Ontario) supplemented with potassium chloride (Hospira; Montreal, Quebec), 15 mEq/L at 60 mL/kg BW per day, IV; tramadol (Chiron Compounding Pharmacy; Guelph, Ontario), 4 mg/kg BW, PO, q8h, meloxicam (Metacam; Boehringer Ingelheim Canada), 0.1 mg/kg BW, PO, q24h, and gentamicin (Merck & Co., Kirkland, Quebec), 6 mg/kg BW, IV, q24h for 8 d.

Figure 1.

Intra-operative view of the bladder. A cystotomy reveals multifocal black-green plaques on the mucosal surface.

Serial CBC evaluations and biochemical and urine analyses were performed during the patient’s hospitalization. Seven days after surgery there was a mild, non-regenerative anemia (hematocrit 0.369 L/L; RI: 0.40 to 0.56 L/L, reticulocyte count 19.01 × 10⁹/L), and a mild panhypoproteinemia (albumin 22 g/L and globulin 24 g/L; RI: 25 to 38 g/L). Urine pH was 7 and specific gravity was 1.010. There was pyuria, hematuria, and proteinuria. No microbial growth was noted on extended urine culture 8 days following the cystotomy.

Initially, the patient appeared to improve; however, clinical signs recurred and worsened over subsequent weeks. Repeat urine culture revealed growth of C. urealyticum. Based on in vitro susceptibility testing, chloramphenicol (Vétoquinol, Lavaltrie, Quebec) was administered at a dosage of 25 mg/kg BW, PO, q8h for 7 d. Despite recommendations for follow-up, the patient was not presented again for nearly 4 mo. At that time the dog was experiencing severe lethargy, stranguria, and hematuria. On physical examination the dog had tacky mucous membranes and was mildly tachycardic. The inguinal region was severely urine scalded. A CBC and chemistry panel revealed a moderate lymphopenia and hypoalbuminemia (23 g/L; RI: 30 to 36 g/L). Urine pH was 9. Repeat abdominal ultrasound revealed recurrence of severe, diffuse thickening of the urinary bladder wall with mineralization of the mucosa.

A stored isolate of C. urealyticum previously obtained from the patient was used for vancomycin disk diffusion susceptibility testing. The patient was hospitalized and intravenous fluid therapy was initiated using Lactated Ringer’s Solution (Baxter) with potassium chloride, 15 mEq/L at 80 mL/kg BW per day during the first 24 h to correct dehydration. Vancomycin (Vancocin; Iroko International, Philadelphia, Pennsylvania, USA) was administered for 13 d (15 mg/kg BW diluted in 50 mL of sterile water, IV over 1 h q12h) following receipt of test results showing the organism was susceptible in vitro to vancomycin. Methionine (Lloyd; Brampton, Ontario) 50 mg/kg BW, PO, q8h was also administered to aid in urine acidification. Subsequent hematologic and biochemical analyses revealed a mild, normocytic, hypochromic, non-regenerative anemia (HCT 0.318 L/L) and a mild to moderate hypoalbuminemia (21 g/L). Urine pH 7 d after initiation of vancomycin and methionine was 6.5.

Despite the vancomycin therapy, the patient was still experiencing significant stranguria and hematuria. Additionally, the owner had stated that if resolution of the infection could not be achieved, euthanasia would be elected. Therefore, the decision was made to utilize intravesical gentamicin in addition to the systemic antimicrobial therapy. The day after completion of vancomycin therapy, the patient was sedated with acepromazine (Atravet; Boehringer Ingelheim Canada), 0.01 mg/kg BW, IM and a sterile 10-Fr, 55 cm Foley urinary catheter (Smiths Medical ASD; St. Paul, Minnesota, USA) was placed. Gentamicin (6 mg/kg BW) was infused into the urinary bladder. The catheter was left in place and occluded to prevent the patient from voiding. The patient was monitored closely for straining during the dwell time. The catheter was removed after 6 h and the patient voided the gentamicin immediately thereafter. Treatment was repeated the next day.

Subsequent urine cultures performed 1 and 6 wk following treatment did not reveal any microbial growth. Repeat ultrasound of the urinary bladder 2 mo later revealed complete resolution of the hyperechoic foci and significant improvement of the mucosal irregularity and thickening. One year following treatment the patient remained symptom-free.

Discussion

Factors implicated in the development of encrusting cystitis in human and veterinary patients include previous UTIs, past use of broad-spectrum antibiotics, prolonged bladder catheterization, immunosuppression, bladder trauma, neurologic disease leading to abnormal micturition, and urologic disease (1,3,4,5). The patient in this report had several predisposing factors that likely contributed to the development of C. urealyticum infection. The dog was originally diagnosed with a urethroperineal fistula and secondary UTI with Proteus, Staphylococcus, and Enterococcus spp. The initial urine pH was 8, which presumably created a favorable environment for the invasion and growth of C. urealyticum. In addition, the patient had received broad-spectrum antibiotics during the months preceding referral.

As demonstrated in this case, C. urealyticum is typically resistant to most antimicrobial agents (10,11). In vitro susceptibility testing of the first isolate revealed susceptibility to enrofloxacin; however, subsequent isolates were resistant to fluoroquinolones. This finding is not uncommon; recent studies have shown an increased level of resistance to this class of antimicrobials (4,12). While susceptibility testing is routinely recommended to guide antimicrobial selection, in vitro susceptibility may not correlate with in vivo activity as shown in this case and previous cases in which there was in vitro susceptibility to doxycycline but prolonged use of this antibiotic failed to clear the infection (5,13).

Antimicrobial management of encrusting cystitis in humans typically relies on the systemic administration of glycopeptides, such as teicoplanin or vancomycin, to which the organism is consistently susceptible (1,4,9,14–16). Although vancomycin has been used successfully for the treatment of C. urealyticum encrusting cystitis in canine patients, there is no formulation that has been approved for veterinary patients and reports of its use in veterinary medicine are limited (5,17,18). The use of vancomycin in the treatment of multi-drug resistant Gram-positive pathogens and the development of vancomycin resistant organisms underscore the importance of restricted use of this drug (18). In this case, the severity of the patient’s clinical signs and encrusting lesions, in conjunction with the lack of response to surgical debridement and the limited antimicrobial in vitro susceptibility led to the decision to use vancomycin.

While the prognosis for resolution of C. urealyticum infection is good with appropriate antimicrobial therapy, persistent bacteriuria may occur (3). Urine acidification is often recommended in conjunction with antimicrobial therapy to improve antibiotic efficacy (1,3). Antibiotics such as glycopeptides and tetracyclines have been demonstrated to be less effective at a high urine pH. Thus, the use of urine acidifying agents and inhibitors of urease, such as acetohydroxamic acid, may help to achieve better clinical results (1,19,20). In our case, acetohydroxamic acid was not available, so methionine was utilized to attempt to reduce the urinary pH. Following metabolism of methionine, sulfate is excreted in the urine as sulfuric acid, thereby acidifying it (21). The use of methionine in veterinary patients with C. urealyticum infection has been reported; however, the efficacy of this drug has not been proven (12,22). In our patient, the pH decreased from 9 to 6.5 following the combined use of methionine and vancomycin.

In those patients for which appropriate antimicrobial therapy and urine acidification alone are ineffective, combined local treatment may be warranted (1,3). Cystoscopic or surgical removal of bladder wall encrustations has been reported in human and veterinary patients (1,3,5). Resolution of C. urealyticum encrusting cystitis was described in 2 dogs following surgical debridement (5). However, both dogs also received antimicrobial therapy, with 1 of the dogs receiving vancomycin (5). In the case reported here, surgery and post-operative intravenous antimicrobial therapy were not successful in eradicating the infection. Potential reasons for treatment failure include inadequate bladder debridement, the presence of encrusting lesions in the urethra or ureters or ineffective antimicrobial therapy following surgery.

In conjunction with vancomycin therapy and urine acidification, the patient herein also received intravesical gentamicin. Intravesical therapy offers several benefits compared with other routes of antibiotic administration. These benefits include the ability to achieve high concentrations of antibiotics at the site of the infection while minimizing systemic effects (23–26). Gentamicin is highly cationic and does not easily cross lipid membranes; therefore, this drug in particular may provide high local antibiotic concentrations, with low risk of systemic absorption (23,24,26).

Limitations and possible adverse effects of intravesical therapy include the low permeability of the urothelium, the need for repeated catheterizations, washing out of drug solutions due to periodic voiding of urine, overdistension of the bladder with possible rupture and absorption of toxic concentrations of drugs (25–27). However, intravesical gentamicin therapy has been shown to be safe and effective for complex urologic cases in human medicine (26). Defoor et al (24) reported intravesical gentamicin therapy on 80 human patients, most of whom had neuropathic bladder secondary to myelomeningocele and for whom oral antibiotic administration was unsuccessful. The duration of treatment varied from 3 to 1095 days. No patient had serum gentamicin levels > 0.4 μg/mL (normal trough level 0.2 to 2 μg/mL). Small increases in serum creatinine were seen in 3 patients with chronic renal insufficiency; however, this finding was attributed to progression of disease. No adverse events, clinical toxicities or allergic reactions were noted (24). Wan et al (23) examined intravesical instillation of gentamicin in rats, dogs, and humans. Rat and dog bladder models were used to determine the effects of inflammation, infection, bilateral vesiculoureteral reflux, and elevated bladder pressure on systemic absorption of gentamicin. No significant absorption was noted in the dog model; serum levels of gentamicin were not detectable and serum creatinine levels remained within reference range (23).

In conclusion, this case highlights the challenges involved in the treatment of encrusting cystitis caused by C. urealyticum. Prompt identification of the bacteria, in vitro susceptibility testing, and knowledge of the pathophysiology of this multi-drug resistant microorganism are essential for effective treatment. The use of urinary acidifying agents, in conjunction with appropriate antimicrobial agents, may help to achieve improved treatment outcome. To the authors’ knowledge, this is the first report of the use of intravesical gentamicin sulfate therapy for the eradication of C. urealyticum in a dog. Intravesical antimicrobial therapy may represent a unique adjunct therapy for complex urological cases involving C. urealyticum.