Managing routine and difficult urinary tract infections in dogs (Proceedings)

Urinary tract infection (UTI) occurs when bacteria colonize portions of the urinary tract that are normally sterile (i.e., kidney, ureter, bladder, and proximal urethra). UTI result from abrogation of one or more natural defense mechanisms that allow bacteria to ascend from the perineum to the urethra, and then to the bladder. In some instances, bacteria will ascend the ureters and enter the kidneys and cause pyelonephritis.

Bacteria causing UTIs often ascend following contamination with the dog's own fecal flora or from ascent of organisms from the skin. Enteric bacteria, such as the coliforms (i.e., E. coli, Klebsiella spp., Enterobacter spp., and Serratia spp.) and Enterococci, are normal inhabitants of the lower digestive tract and are present in feces in relatively high numbers-usually > 1x105 cfu/gram. Some strains of E. coli possess virulence factors that enhance attachment to the uroepithelium.

Organisms associated with dermatoses and perivulvar inflammation (e.g., coagulase-positive Staphylococci and Pseudomonas aeruginosa) are less frequently isolated from UTIs than organisms of fecal origin. UTI rarely results from the hematogenous spread of organisms to the urinary tract. E. coli is the most commonly isolated organism as the cause for UTI in both dogs and cats. E. coli, Staphylococcus spp., and Proteus spp. account for most cases of UTI in primary care practices. Enterococcus spp. are isolated in dogs and cats with UTI with increasing frequency at referral centers and have limited treatment options at times because of resistance patterns of this organism.

Diagnosis

UTIs are definitively diagnosed following isolation of bacteria using quantitative culture of urine and reported as colony-forming units/ml (cfu/ml). We discourage any use of qualitative culture methods that do not report cfu/ml as the number of isolated organisms are a major factor when determining the likelihood that a true UTI exists. The vast majority of UTIs are associated with one organism. Isolation of multiple types of organisms suggests the possibility that contamination of the urine sample occurred during collection. Interpretation of culture results from cystocentesis-acquired urine is the most straightforward as no growth should be identified, though low-magnitude contamination with organisms from the skin is possible (< 1,000 cfu/ml).

If urine is acquired via catheterization or midstream free catch, contamination with normal flora from the distal urethra or genital tract is likely to be isolated during urine culture, even of healthy animals. Large quantitative bacterial growth often occurs from voided urine samples collected from healthy female dogs. Consequently, we do not recommend culture of voided urine specimens in dogs because the degree of bacterial contamination can be large and it becomes impossible to know from where the organisms arose. Urine should ideally be cultured as soon as possible following collection since bacteria continue to divide in urine at room temperature. It is recommended that urine be plated to culture media within 30 minutes of collection or stored at 4°C until plated.

In-house culture as surveillance for occult UTI in Cushing's disease, diabetes mellitus, or CRF (especially in cats) may be particularly useful as there is no urgency for culture results. In-house culture could still performed while awaiting culture results in those displaying lower urinary tract clinical signs after starting empirical antimicrobial therapy. Positive culture plates may be forwarded to external laboratories for organism identification and antimicrobial susceptibility testing.  An in-house culture setup would include some basic supplies, such as quantitative loops and culture media, and an incubator.

A very convenient approach is to use a dip-paddle culture system, such as the Uricult™ (Vetlab Supply; Palmetto Bay, FL). Other than the incubator, this system is completely self-contained and provides quantitative information. The density of quantitative growth is compared to provided charts to estimate the cfu/ml. Different growth media on the paddles allows the organisms to be identified as either Gram-positive or Gram-negative but species identification and organisms susceptibility are not provided. In-house urine culture may represent a new revenue “stream” for veterinary clinics.

Antimicrobal therapy

Antimicrobial therapy is the mainstay of treatment for UTI. The urinary concentration of the antimicrobial agent that is achieved is the most important factor in determining the likelihood of bacteriological cure in uncomplicated UTI.  Major anatomical, metabolic, or functional abnormalities within the urinary system can make it impossible to effect or sustain long-term sterility within the urinary tract regardless of the antimicrobial agent that is chosen for therapy. An ideal antimicrobial agent would be excreted extensively by the kidneys (GFR and or tubular secretion) to achieve a high urinary concentration (ug/ml).

Further increase in the urinary concentration of this agent would be achieved following normal tubular function and resultant high urine specific gravity or osmolality.  A drug with a high likelihood toward the organism's susceptibility should be chosen to start therapy; the drug may need to be changed following return of susceptibility testing from the microbiology laboratory. In UTI seated deeply within tissues (kidney, prostate, very thickened bladder wall), the concentration of the antibacterial agent that can be achieved in the plasma and tissues is more important than that achieved in the urine.

The chosen drug should be readily available, labeled for veterinary use for the intended species, and be reasonably affordable. The preferred drug would be one that is given orally once a day for a short number of days in order to increase owner compliance in giving all the prescribed medication. This drug should be one that is unlikely to induce resistance of host bowel organisms, unlikely to be associated with adverse reactions, and works in all urinary pH.  Ideally, your practice has categorized its antimicrobial formulary into three groups as recommended by the American College of Veterinary Internal Medicine (ACVIM) 2005 consensus statement on antimicrobial use.

Empirical therapy

A patient with lower urinary tract signs and a likely UTI can be in considerable discomfort; therefore, antibacterial treatment should not be withheld pending susceptibility results. Findings from urinalysis can suggest the likelihood of a true UTI (pyuria, hematuria, bacteriuria). Pain relief with drugs such as buprenorphine or tramadol can be considered while awaiting results of a urine culture when the diagnosis of UTI is questionable.

When sediment examination is performed, the finding of excess white blood cells in combination with bacterial organisms highly suggests that a true UTI exists. When the bacteria are identified as rod-shaped organisms, it is reasonable to choose some antimicrobial agent with gram-negative coverage such as a fluoroquinolone, clavulanate- amoxiciliin, or trimethoprim (ormetoprim)-sulfa product. If sediment examination reveals cocci, then it is likely that the infection is due to a gram-positive organism-notably Enterococcus spp.

In that case, the animal should be put on amoxicillin or amoxicillin-clavulanate. If there are cocci present and the urinalysis reveals alkaline pH, it is likely that the infection is caused by a Staphylococcus spp. (because of urease production). If Staphylococcus spp. are suspected, it is more prudent to use amoxicillin-clavulanate because of the common production of beta-lactamase by Staphylococci. Empirical therapy should NOT be prescribed for patients with chronic or repetitive UTI with a history of extensive antimicrobial use. Young cats with urinary urgency have bacterial UTI very infrequently, so treatment with pain-relieving drugs can be given while awaiting urine culture results.

Interpretation of antimicrobial susceptibility reports and selection of agents

The drugs chosen for susceptibility testing by individual laboratories is variable, and the methodology used- Kirby-Bauer disc or broth microdilution-is also subject to laboratory preference. The method of susceptibility utilized by a specific laboratory has direct implications on how laboratory results are reported. If Kirby-Bauer disc diffusion is used, then a simple interpretation of “S” (sensitive), “I” (intermediate), or “R” (resistant) is given. If broth microdilution is used, then an interpretive value “S/I/R” is given plus they may additionally report a minimum inhibitory concentration (MIC) value for each drug. When laboratories follow appropriate performance guidelines and interpretive criteria, the values of both methods agree with each other ≥ 96% of the time for a given bacterial isolate.

How long should antibacterial treatment be given?

Ten to 21 days of an appropriate antibacterial agent for treatment of an uncomplicated lower UTI is often recommended. At least 30 to 60 days of antimicrobial therapy is usually needed to sterilize the upper urinary tract (kidneys and ureters) – sometimes long term bacteriological cure is not possible. Antibacterial treatment for sexually intact males with UTI is given for at least 30 days – longer courses are often necessary. These guidelines for duration of treatment are based on conventional wisdom and experience over the years, but surprisingly little data exists to support this. Ultimately, antimicrobial agents should be given for as long as is necessary to effect a bacteriologically sterile urine during administration of the medication and for a protracted time following discontinuation of the treatment.

Single dose and 3-day antibacterial dosing regimens have been effective in some human populations with UTI. Trimethoprim-sulfadiazine at one time was commonly used in short-course treatment protocols for women with UTI. Fluroquinolones are often prescribed to treat uncomplicated UTI for as little as 3 days in women. Why then have short-term antibacterial protocols for treatment of UTI in dogs not been recommended?  Conventional wisdom has stated that dogs don't have as many early mucosal infections with their UTI as do their human counterparts – if true, this would mean that dogs would have more deep tissue invasion making it more difficult to treat.

The effect of single or 3 day doses of either amikacin or trimethoprim-sulfa on experimentally induced uropathogenic E. coli UTI in dogs has been studied. Large quantitative bacterial growth was present in 2 of 4 dogs 14 days after a single oral dose of trimethoprim-sulfa (30 mg/kg) was administered; very low quantitative growth existed in the other 2 dogs. A single dose of amikacin (20 mg/kg SQ) was associated with large quantitative bacterial growth in 3 of 4 dogs 14 days after treatment; urine was sterile in one remaining dog. When dogs in this model of UTI were treated with amikacin at 10 mg/kg SQ BID for 3 days, large quantitative bacterial growth persisted 14 days after treatment in 6 of 8 dogs; 1 of the remaining dogs had sterile urine and another had low quantitative growth.

In dogs treated with trimethoprim sulfa at 15 mg/kg BID for 3 days, 2 of 8 dogs had high quantitative bacterial growth, 2 had low count quantitative bacterial growth, and 4 were sterile 14 days following treatment. Interestingly, all the dogs with sterile urine after the 3-day course of trimethoprim-sulfa were female dogs – all those with any degree of quantitative growth were males. It appears that in this severe model of experimentally induced UTI in dogs that a 3 day course of trimethoprim-sulfa resulted in sterile urine two weeks following treatment in females (Rogers JAVMA 1988). Males used in this study were sexually-intact; results may be different in castrated male dogs but this has not been studied.                                                                              

In a recent prospective clinical study, treatment of uncomplicated bacterial UTI in dogs was compared between a high-dose short duration of enrofloxacin and a standard duration regimen of amoxicillin-clavulanate in an interim analysis (Irom S. J Vet Int Med 2011).  Exclusion criteria included those with a persistent UTI, frequently recurrent UTI, uncontrolled systemic illness, and recent administration of antimicrobials or glucocorticosteroids. Enrofloxacin was administered at 18-20mg/kg orally once daily for 3 consecutive days and amoxicillin-clavulante was administered at 13.75-25mg/kg orally twice daily for 14 days.  

Both treatment groups had urinalyses and urine cultures submitted on day 0, 10, and 21. Urine culture results were compared between day 10 for the enrofloxacin treated dogs and day 21 for the amoxicillin-clavulanate treated dogs – comparing bacteriological results 7 days after finishing treatment for both groups. Thirty-six dogs were analyzed in this interim report.

Bacteriological cure was achieved in 15 dogs (83%) treated with enrofloxacin and 14 dogs (78%) treated with amoxicillin-clavulante, respectively. These data suggest that the high-dose, short-duration enrofloxacin protocol was equally effective to the standard protocol  of 14 days of treatment with amoxicillin-clavulanate in treating uncomplicated canine UTI in this sample patient population and may represent a viable alternative therapeutic regimen for similar patients.  Urinary levels of enrofloxacin and ciprofloxacin were measured at 2, 8  and 24 hours in 6 normal dogs administered 20 mg/kg of enrofloxacin as a single oral dose. The urinary concentration of enrofloxacin at 8 hours post administration was from approximately 70 to 165 ug/ml and from  195 to 435 ug/ml for ciprofloxacin at the same time (Irom S OSU Master's 2010).

The package insert label for Baytril® reports urinary levels of enrofloxacin at 43 ug/ml at 2 hours and at 55 ug/ml at 8 hours following a single oral dose at 2.5 mg/kg in the 2 dogs reported. The insert label does not report the levels of ciprofloxacin achieved. It appears that a single dose of enrofloxacin at 20 mg/kg achieves high levels of urinary enrofloxacin and very high levels of ciprofloxacin.  Future decision making for likely urinary susceptibility to enrofloxacin should take into account the high levels of urinary ciprofloxacin generated following metabolism of enrofloxacin in addition to that of enrofloxacin.

Recurrent UTI: Reinfection or relapse?

Reinfection is defined as another clinical episode caused by an organism different from the one previously involved. This organism may be an entirely different genus and species, or it may be the same organism but a different biotype, which is the case in 50% of recurrent UTI. This situation represents a new infection that typically occurs weeks to months after discontinuation of drug therapy for a previous UTI. Multiple new infections suggest that the animal's host defense mechanisms are not operating properly. A search for predisposing factors should be undertaken, including anatomical defects, urolithiasis, urine retention (e.g., neurologic dysfunction), and neoplasia. In some instances, dogs with reinfections will have moderate to severe recession of the vulva and overlying skin folds.

Vulvar recession seems to be a risk factor for recurrent UTI in dogs, but many dogs without UTI also have vulvar recession. The types and numbers of organisms in the vulvar area likely favor increased ascent of bacteria, and a recessed vulva may serve as a barrier to complete emptying of the bladder, which can contribute to incontinence or ascending infection because of a “wicking” effect. Vulvoplasty or episioplasty can dramatically reduce recurrence of UTI in affected dogs. Primary care veterinarians, internists, and surgeons frequently overlook this risk factor.

Relapsing infection is another clinical episode of UTI caused by the same organism and implies persistence of an organism that was never eradicated. Relapse suggests that the infection is deep-seated in the tissues or that the organism is resistant to the chosen antimicrobial. Clinical signs tend to occur soon after discontinuation of treatment, usually within days to one week. Persistent UTI is a variant of relapsing infection in which bacterial cultures remain positive with the same organism during antimicrobial treatment. In this instance, the organism has not been eradicated, even transiently. Persistent infection occurs in approximately 2% of all recurrent UTI and implies severe abrogation of local host defenses or that the organism is highly resistant to the administered antimicrobial drug.

A search for predisposing factors should be undertaken to exclude: pyelonephritis; obstructive nephropathy; urolithiasis; chronic bladder wall changes allowing sequestration of bacteria; anatomical defects; polypoid cystitis; urine retention; and reinoculation of the organism from prostatic or uterine disease. A unique form of relapsing UTI is caused by Corynebacterium urealyticum, in which encrustations of urinary tissue and struvite prevent eradication of the organism with medical treatment.

By definition, relapsing UTI means that the organism has never been fully eradicated from the urinary tract because they are inaccessible, therapeutic antimicrobial concentrations are not achieved in the urinary tract, or that the organisms are highly resistant to the chosen antimicrobial. Long-term therapy with an appropriate antimicrobial for 30 to 60 days or longer may be necessary.

Susceptibility testing, preferably with MIC, should be performed to ensure selection of an antibiotic that is likely to be effective. A change in the antibiotic used to one that achieves greater tissue penetration may be necessary (e.g., fluoroquinolones). Predisposing anatomical factors (e.g., urolithiasis, polypoid cystitis, urachal remnants) that allow sequestration of bacteria must be identified and eliminated. Culture of urine while the animal is receiving antimicrobials is advocated as an in vivo method of susceptibility testing.

Successful treatment is defined as sterile urine during and after medication administration. Resolution of clinical signs, such as hematuria, proteinuria, and microscopic bacteriuria, can be misleading as these may resolve transiently because of reduced activity of the UTI without eradication. Quantitative urine cultures are recommended at five to seven days, one month, and three months after medication has been discontinued to ensure sterility of the urinary tract in patients with initial UTI. For those with recurrent UTI, quantitative culture of urine during treatment can be quite helpful.

Culture of urine three to five days after beginning treatment documents effective eradication of the organism in the urine, identifies rapid emergence of resistance if present, and rules out persistent infection. Culture of urine three days before the end of treatment will rule out development of a superinfection (i.e., new organism). Culture of urine seven to 10 days after treatment has ended rules out rapid relapse, whereas cultures at one, two, three, six, and 12 months are performed to identify reinfection. This regimen of culture is most useful for difficult cases in which UTI frequently recurs or relapses.

Selected reading

Ambrose et al. In: Lorian V, ed. Antibiotics in laboratory medicine, 5th ed. Philadelphia, Pa: Lippincott Williams & Wilkins, 2005;1-144.

Barsanti, JA. In: Greene CE, ed. Infectious diseases of the dog and cat, 3rd ed. Philadelphia, Pa: Elsevier Saunders, 2006;935-961.

Cohn LA, Gary AT, Fales WH, Madsen RW. Trends in fluoroquinolone resistance of bacteria isolated from canine urinary tracts. J Vet Diagn Invest 2003;15:338-43.

Chew DJ, DiBartola SP, Schenck PA. Cystitis and Urethritis-infections: Urinary Tract Infection. In: Canine and Feline Nephrology and Urology, 2nd ed. Philadelphia, Pa: Elsevier, 2010;240-271.

Irom S, Westropp J, Chew D, Daniels J: Interim evaluation of the efficacy and safety of a high dose short duration enrofloxacin treatment regimen for urinary tract infection in dogs. J Vet Int Med 25:723, 2011.

Morley PS, Apley MD, Besser TE, Burney DP, Fedorka-Cray PJ, Papich MG, Traub-Dargatz JL, Weese JS: ACVIM Consensus Statement : Antimicrobial Drug Use in Veterinary Medicine. J Vet Intern Med 2005;19:617–629.