Antibiotic-Resistant Corynebacterium jeikeium Urinary Tract Infection in a Cat

Introduction

Corynebacterium jeikeium (C. jeikeium) infection has not been reported as a cause of urinary tract infections (UTIs) in humans or animals.¹ In cats, bacterial UTIs most often develop secondary to systemic diseases or as a consequence of another urinary tract disorder.² Greater than 90% of bacterial UTIs in cats are caused by Escherichia coli, Streptococcus spp., Staphylococcus spp., Proteus spp., Klebsiella spp., Enterobacter spp., Pseudomonas spp., or Pasteurella spp.^2,3 A few case reports exist of Corynebacterium urealyticum causing UTIs in cats and dogs.^4–7

Corynebacterium group JK was first identified and characterized in 1979.⁸ The organism has since been renamed Corynebacterium jeikeium. The bacterium has caused various infections in humans, including sepsis, endocarditis, pneumonia, soft-tissue infection, infection of a cerebrospinal fluid shunt, and infection of a tibial fracture.^9–17 Antimicrobial susceptibility testing of C. jeikeium isolates from human clinical samples has shown that in most cases the organism is resistant to multiple antibiotics.^{12,15,18–22} However, C. jeikeium appears to be consistently susceptible to vancomycin, and most isolates are variably susceptible to other medications.^{12,14,15,17–23}

The purpose of this paper is to describe a clinical case of UTI apparently caused by C. jeikeium, an organism that has not been previously reported to cause disease in animals.

Case Report

A 10-year-old, 6.3-kg, castrated male, domestic longhaired cat was presented for evaluation of urinary tract inflammation that had persisted for 1.5 months, despite antibiotic administration. Historically, hematuria, pyuria, and crystalluria were identified in this animal. Fourteen months earlier, the cat was presented for urethral obstruction, and a perineal urethrostomy was performed at that time. Abdominal ultrasonography done prior to surgery revealed cystic calculi, mineralization of the pelvis and parenchyma of the caudal pole of the left kidney, bilateral pyelectasia, a left ureteral calculus, and a right renal calculus. Urethral calculi and cystoliths were retrieved at the time of surgery. Compositional analysis of the calculi identified their content to be 100% calcium oxalate monohydrate. After undergoing surgery, urinary tract inflammation periodically recurred. Aerobic cultures of urine samples obtained via cystocentesis, performed between the time of surgery and the time of referral, yielded either no growth or light growth of Corynebacterium spp.

In the 1.5 months prior to presentation, four urinalyses were performed by the referring veterinarian. Three of these samples were obtained via cystocentesis, and one was a “free catch” sample. In all four samples, there were varying degrees of pyuria, hematuria, proteinuria, and small numbers of epithelial cells. The pH of the samples ranged from 6 to 6.5. Calcium oxalate crystals were identified in one sample. Bacteriuria was present in the last three urinalyses. Contamination of the sample was presumed to have occurred, and antibiotic susceptibility testing was not performed. Courses of orbifloxacin, enrofloxacin, and cephalexin were prescribed by the referring veterinarian, and the cat was referred for further evaluation.

At the time of presentation, the owner reported that the cat had normal urinary habits. Hematuria had not been observed. The cat had been fed a prescription diet^a for 14 months. On physical examination, the cat was bright, alert, responsive, and in good body condition. The urinary bladder was moderately distended, and the left kidney was misshapen and smaller than the right kidney on abdominal palpation. The perineal urethrostomy site appeared healthy.

Abnormalities identified on the complete blood count included mild leukopenia (white blood cells [WBCs] 4.57 × 10³/μL; reference range 4.79 to 12.52 × 10³/μL), an increased number of reactive lymphocytes (0.091 × 10³/μL; reference range 0 to 0.025 × 10³/μL), and mild toxic changes in the neutrophils. A serum biochemical profile revealed mildly increased blood urea nitrogen (43 mg/dL; reference range 15 to 35 mg/dL) and mild hypercholesterolemia (301 mg/dL; reference range 42 to 265 mg/dL). Serum creatinine concentration was normal (1.9 mg/dL; reference range 0.5 to 2.3 mg/dL). Urine obtained by cystocentesis was cloudy and had a pH of 7 and a specific gravity of 1.030. Dipstick analysis showed 2+ protein and 3+ blood. Protein measured by sulfosalicylic acid precipitation was 1+. Sediment examination revealed hematuria (>100 red blood cells [RBCs] per high-powered field [hpf]), pyuria (>100 WBCs per hpf), bacteriuria (3+), and 5 to 10 epithelial cells per hpf.

Abdominal ultrasonography showed sediment within the urinary bladder, bilateral nephroliths and mild pyelectasia, mineralization of the left renal pelvis, a cyst within the cranial cortex of the right kidney, and calcification in the parenchyma of the right kidney. The left kidney was 42.7 mm in length, and the right kidney was 48.3 mm.

An aerobic urine culture yielded moderate growth of Corynebacterium spp., and a diagnosis of UTI with possible pyelonephritis was made. The specific Corynebacterium organism was not identified, and antibiotic susceptibility testing was not done at this time. The Clinical Laboratory Standards Institute has not published interpretive criteria or quality control standards for antibiotic susceptibility testing of Corynebacterium spp.²⁴ It was presumed that the organism cultured was C. urealyticum, since it is the most commonly reported isolate of Corynebacterium spp. from the urine of small animals. The antibiotic susceptibility patterns of Corynebacterium spp. are notoriously unpredictable, and some animal isolates have been susceptible to multiple antibiotics.^4,6

Historically, this cat became anorexic after administration of many antimicrobial medications, but fluoroquinolones had been administered without notable side effects. For this reason and because of its once-daily administration, marbofloxacin^b was prescribed (2 mg/kg per os [PO] q 24 hours) for 2 weeks. Two weeks later, while the cat was still on marbofloxacin, Corynebacterium spp. were again cultured from urine obtained via cystocentesis. Administration of marbofloxacin was continued, but the dose was increased to 4 mg/kg PO q 24 hours. A second urine culture 3 weeks later, while the cat was being given marbofloxacin, yielded heavy growth of Corynebacterium spp. Antibiotic therapy was empirically changed to azithromycin,^c primarily because of its once-daily dosing (8 mg/kg PO q 24 hours for 5 days, then q 48 hours for 21 days).

Urine culture 3 weeks later was again positive for Corynebacterium spp. Amoxicillin-clavulanate^d (10 mg/kg PO q 8 hours) was started, and a sample of the bacteria was submitted for partial 16S ribosomal ribonucleic acid (rRNA) gene sequencing. The organism was determined to be Corynebacterium jeikeium.^e Ribosomal RNA genes are typically highly conserved among species, and sequencing is often considered more accurate than routine morphological and biochemical techniques for the identification of microorganisms. In addition, agar disk diffusion (i.e., Kirby-Bauer technique) susceptibility testing was attempted. At 24 hours, large zones of inhibition were present for chloramphenicol, gentamicin, vancomycin, tetracycline, amikacin, tobramycin, and nitrofurantoin; however, given that no clinical guidelines for interpretation of susceptibility testing of Corynebacterium spp. have been established, the accuracy of the susceptibility interpretation was uncertain. Because no zone of inhibition was present for amoxicillin-clavulanate and the cat had developed anorexia, the drug was discontinued. Nitrofurantoin was then prescribed (8 mg/kg PO q 8 hours for 21 days) because of the in vitro susceptibility of the bacteria to the drug and because a large percentage of the drug remains unchanged as it is excreted into the urine.

Urinalysis via cystocentesis and urine culture were performed 5 weeks later, after antibiotics had been stopped for 14 days. Hematuria (RBCs >100 per hpf), pyuria (WBCs >100 per hpf), and bacteriuria (3+) persisted. Urine specific gravity was 1.029, and pH was 6. Epithelial cells (0 to 5 per hpf) and trace protein were present. Culture yielded heavy growth of Corynebacterium spp. Treatment with vancomycin was not pursued, because the cat was mildly azotemic at the time of presentation, and treatment with vancomycin is usually reserved for use in life-threatening infections. Treatment with meropenem^f (8 mg/kg subcutaneously q 12 hours) was instituted and continued for 25 days.²⁵ This drug was chosen because of its broad spectrum of activity and its utility in treating infections caused by organisms resistant to more commonly used antimicrobials.

One week after completing meropenem, urine was obtained via cystocentesis for urinalysis and urine culture. Specific gravity of the urine was 1.032, and the pH was 6. Hematuria (3+ blood, RBCs >500 per hpf), pyuria (WBCs >100 per hpf), and proteinuria (1+ dipstick, 1+ sulfosalicylic acid precipitation) were present. Moderate numbers of amorphous crystals were noted upon sediment examination, but bacteria were not found. When urine was cultured in thioglycollate broth, growth of Corynebacterium spp. occurred again. Based on a lack of clinical signs, the cost of treatments, and risks of side effects if vancomycin were to be used for treatment, the owner declined further treatments. It is possible that longer therapy with meropenem may have eliminated the infection in this cat, as the last culture that was performed yielded growth only in thioglycollate broth.

Discussion

The term “diphtheroid” refers to any gram-positive bacillus that appears similar to C. diphtheriae on a Gram stain. The term “coryneform” is often used by microbial taxonomists to describe these organisms instead of the term “diphtheroid.” Several genera of bacteria, including Corynebacterium, are considered to be coryneform. Since they are common normal flora of the skin and mucous membranes, sample contamination is often presumed to have occurred when microbial culture of clinical specimens yields the growth of coryneform.¹ Some coryneform have been shown to be opportunistic pathogens, however.^2,26 Coryneform are considered clinically relevant under the following conditions: when they are isolated from a typically sterile site; when they are the predominant organism found in an adequately collected clinical sample that yields mixed microbial growth; when multiple clinical specimens are positive for the same bacteria; when coryneform bacteria are seen on a Gram stain of the sample, and there is an associated leukocyte reaction; or when other organisms recovered from the sample are of low pathogenicity.¹

The cat described in this case had persistent laboratory evidence of UTI but showed no clinical signs. Cats that have a perineal urethrostomy are at increased risk for developing UTIs, and the sediment within the urinary bladder and the presence of nephroliths may have contributed to the persistence of the UTI in this animal. The source of the C. jeikeium in this cat was not identified, but it is noteworthy that this organism has been cultured from human urine and has the ability to adhere to Teflon-coated cathethers.^27,28 In this case, the decision was made to start antibiotic therapy in order to treat the infection and avoid potential complications (e.g., pyelonephritis).²⁹

The case reported here demonstrated that C. jeikeium may be associated with UTIs in cats. Corynebacterium jeikeium should not be dismissed as contamination if it is the only organism cultured from urine samples obtained from cats with evidence of UTI, and especially if it has been cultured on multiple occasions from appropriately collected urine samples. The organism isolated from this animal was resistant to treatment with multiple antibiotics commonly used to treat UTIs in animals, including nitrofurantoin after susceptibility testing indicated the drug might be effective against the bacterium. The antimicrobial-resistant nature of C. jeikeium has been documented in human infections, but the mechanism of antimicrobial resistance is unknown.²⁶ Based on the results of this case, it appears that even with appropriate culture and susceptibility testing, infection with C. jeikeium may be difficult to treat. The lack of an antibiotic sensitivity profile hindered the management of this case. Some laboratories do not routinely perform sensitivity profiles for Corynebacterium spp., because these bacteria are presumed to be contaminants, and there are no published standards for the interpretation of these profiles. Specific requests for antibiotic sensitivity testing may aid case management when Corynebacterium spp. are identified as potential pathogens.

Conclusion

Corynebacterium jeikeium was cultured repeatedly from a 10-year-old cat with laboratory evidence of UTI. The C. jeikeium isolated in this case was resistant to multiple antibiotics in vivo, and multiple antibiotics failed to eradicate the infection. Based on this case, C. jeikeium may be associated with UTI in cats and is not always a contaminant when cultured from the urine. Additional case studies are necessary to better describe the pathogenic potential and antibiotic susceptibility patterns of C. jeikeium isolated from animals.