Thoughts on testing methods for bacterial isolates

We read with interest the recent case report¹ published in the JAVMA, “Infection with methicillin-resistant Staphylococcus pseudintermedius masquerading as cefoxitin susceptible in a dog.” This report directs attention to the need for proper identification and testing of bacterial isolates from animals. The S pseudintermedius isolate obtained from the dog described in the report was resistant to several B-lactam antimicrobials but was not tested for susceptibility to oxacillin. Disk diffusion testing indicated the isolate was susceptible to cefoxitin, but further testing revealed that it was a strain of methicillin-resistant S pseudintermedius (MRSP) positive for both PBP2a and the mecA gene. The presence of PBP2a is associated with resistance to all B-lactam antimicrobials, including cephalosporins, penicillin derivatives, and carbapenems. Resistance to cefoxitin has been used to detect production of PBP2a by methicillin-resistant Staphylococcus aureus strains and was recommended in previous standards published by the Clinical Laboratory Standards Institute (CLSI)² as a screening test for other coagulase-positive Staphylococcus isolates from animals. However, in this case, the test misidentified the isolate as susceptible to cefoxitin.

The Veterinary Antimicrobial Susceptibility Testing (VAST) subcommittee of the CLSI considered this issue at its 2009 meetings and reached a decision consistent with the recommendation given by Weese et al¹ and by others^3,4 who investigated screening tests for detection of PBP2a among veterinary Staphylococcus isolates. The CLSI VAST subcommittee voted unanimously to remove Table 9D from M31 and add a footnote to Table 2 stating that cefoxitin breakpoints are not predictive of mecA-mediated resistance to methicillin or oxacillin in S pseudintermedius. This is consistent with the previous suggestion^3,4 that cefoxitin testing is not appropriate for detecting methicillin resistance in S pseudintermedius isolates from dogs. The revised standard for interpretation of resistance among Staphylococcus isolates from animals will appear in the next CLSI document M31-S2, which is scheduled to be published in 2010.

At the same meetings, the subcommittee also considered the current CLSI interpretive criteria,² which list the oxacillin minimum inhibitory concentration breakpoint for resistant S aureus and other coagulase-positive staphylococci as ≥ 4 μg/mL and the zone diameter for resistant isolates as ≤ 10 mm. Because these criteria may not identify some MRSP isolates, the subcommittee voted unanimously to revise the current standard to indicate that oxacillin minimum inhibitory concentrations ≥ 0.5 μg/mL and zone diameters ≤ 17 mm are accurate indicators of mecA-mediated resistance in S pseudintermedius isolates. These revised standards also will appear in the next published CLSI document M31-S2.

The report by Weese et al¹ emphasized the need to ensure that diagnostic laboratories are using standardized and validated protocols for testing veterinary bacterial isolates. They also encouraged researchers and laboratory standard organizations to ensure that proper scrutiny is applied to current testing methods and that those methods be continually reassessed in the light of emergence of organisms such as MRSP. The CLSI has been working to achieve those goals, and the VAST subcommittee provides a public standard that is validated and consensus-driven to support the work of veterinary diagnostic laboratories.

On behalf of the CLSI VAST subcommittee

Mark G. Papich, DVM, MS

Vice Chair

College of Veterinary Medicine

North Carolina State University

Raleigh, NC

The report¹ by Weese et al, “Infection with methicillin-resistant Staphylococcus pseudintermedius masquerading as cefoxitin susceptible in a dog,” described an S pseudintermedius isolate that retained in vitro susceptibility to cefoxitin when assessed by means of the Kirby-Bauer disk diffusion method. The isolate was subsequently determined to be methicillin resistant via a PCR assay for the mecA gene and a latex agglutination test for PBP2a. This finding suggested that cefoxitin susceptibility is not a predictor of methicillin susceptibility in canine S pseudintermedius isolates.

Two recent studies^2,3 published in the Journal of Veterinary Diagnostic Investigation confirmed by testing multiple mecA-positive isolates that cefoxitin is an insensitive predictor of methicillin resistance for canine S pseudintermedius isolates and concluded that the cefoxitin test is inappropriate. This is in contrast to human Staphylococcus aureus isolates, for which cefoxitin disk diffusion testing is reportedly a sensitive and specific method for detection of methicillin resistance.^4,5

In the report by Weese et al, gold-standard methods (ie, a PCR assay and latex agglutination testing) were performed to confirm methicillin resistance. However, the authors indicated that these tests may be impractical for some laboratories and suggested that oxacillin Kirby-Bauer disk diffusion and broth microdilution testing be retained for S pseudintermedius isolates. To clarify the matter, it has recently been determined that oxacillin Kirby-Bauer disk diffusion and broth microdilution testing are sensitive^2,3 and specific² when the 2004 Clinical Laboratory Standards Institute (CLSI) breakpoint guidelines are used. Importantly, it was determined that the current 2008 CLSI guidelines for these methods result in an unacceptably high percentage of false-negative results.

The Veterinary Antimicrobial Susceptibility Testing (VAST) subcommittee of the CLSI examined these discrepancies at its 2009 meetings and agreed with the conclusions of Bemis et al² and Schissler et al,³ and the subcommittee voted unanimously to change the standard for interpretation of oxacillin disk diffusion and broth microdilution breakpoints for canine S pseudintermedius isolates. The revised standard will appear in the 2010 CLSI document. Until this publication is available, the VAST subcommittee recommends the use of the interpretive criteria suggested by Bemis et al² and Schissler et al³ for oxacillin disk diffusion testing (zone diameter ≤ 17 mm) and broth microdilution testing (minimum inhibitory concentration ≥ 0.5 μg/mL).⁶ The VAST subcommittee also agreed that the cefoxitin test is inappropriate for detecting methicillin resistance in canine S pseudintermedius isolates and voted unanimously to publish this information in their 2010 document.⁶

I thank the JAVMA and the Journal of Veterinary Diagnostic Investigation for publishing these manuscripts, as well as the CLSI VAST subcommittee for their responsiveness. Additionally, I appreciate the diligent and relevant ongoing research of Drs. Bemis and Weese in the area of veterinary staphylococci. I agree wholeheartedly with the statement by Weese et al that “…veterinary-specific evaluation of laboratory protocols is required, as protocols designed for important human pathogens may not be optimal for related, but different, veterinary pathogens.”

Jennifer Schissler, DVM, MS, DACVD

Louisville Veterinary Specialty and Emergency Services

Louisville, Ky

Questions contents of clotrimazole solution

We applaud the authors of the report¹ “Intranasal infusion of clotrimazole for the treatment of nasal aspergillosis in two cats.” This paper represents an important step forward in the treatment of nasal aspergillosis in cats. However, we would like to highlight an error in the manuscript that we believe has important clinical ramifications. The authors state, incorrectly, that “[t]he clotrimazole solution contains isopropanol and propylene glycol….” Elsewhere in the paper, the authors correctly state that the clotrimazole solution referenced in the footnotes contains polyethylene glycol.

Currently, only three 1% clotrimazole solutions are commercially available in the United States, consisting of products from VET Solutions, Teva Pharmaceuticals, and Taro Pharmaceuticals. The VET Solutions product, marketed by Vetoquinol USA Inc, contains the vehicle propylene glycol; the clotrimazole solutions marketed for human use by Teva Pharmaceuticals and Taro Pharmaceuticals contain polyethylene glycol as the vehicle.

After recently switching brands (from Teva Pharmaceuticals to VET Solutions) in our university hospital, we observed severe, life-threatening adverse effects following instillation of the propylene glycol–based clotrimazole product in the noses or frontal sinuses of four dogs with nasal aspergillosis. These included oral ulceration and nasopharyngeal swelling severe enough to necessitate temporary tracheostomy to maintain airway patency. The edema subsided within 24 to 48 hours without specific treatment, and the ulcerations healed over several weeks. We would anticipate similar reactions if the propylene glycol–based clotrimazole product had been used in cats with nasal aspergillosis. Of additional concern in cats is the development of Heinz body anemia after ingestion of propylene glycol, although whether hematologic abnormalities would occur after nasal instillation of propylene glycol–based clotrimazole is unknown.^2,3

On the other hand, the human formulations of clotrimazole, which use polyethylene glycol as the vehicle, cause minimal adverse effects (minor mucosal irritation) in our experience, and after reverting to the human formulations for the treatment of nasal aspergillosis in dogs, we have not had any episodes of ulceration or nasopharygeal swelling. We suggest that the human formulations are the only 1% clotrimazole solutions currently on the market that should be used for nasal infusion for the treatment of nasal aspergillosis in dogs and cats.

Although no warning against the use of the VET Solutions brand of clotrimazole in treating nasal aspergillosis appears on the label, we have been assured by the company (Vetoquinol) that such a warning will be posted on the product in the future. We thank Drs. Furrow and Groman for moving the treatment of this difficult condition forward and allowing us to notify the veterinary community of the important distinction in clotrimazole products.

Stephen C. Barr, BVSc, PhD, DACVIM

Mark Rishniw,BVSc, PhD, DACVIM

Department of Veterinary Clinical Sciences

College of Veterinary Medicine

Cornell University

Ithaca, NY

Mary Lynch, PharmD

Brian Bohunicky

Companion Animal Hospital Pharmacy

Hospital for Animals

Cornell University

Ithaca, NY

The authors respond:

We appreciate the letter from Dr. Barr et al underscoring important differences between commercially available clotrimazole solutions. Both of the cats in our report were indeed treated with clotrimazole solution in nonaqueous polyethylene glycol (Taro Pharmaceuticals). It is of interest that one cat in a previous report¹ on feline aspergillosis was treated with a propylene glycol–based clotrimazole solution, and no major complications were observed. However, given the risk of hematologic abnormalities and clinically important nasopharyngeal irritation, we concur that a clotrimazole solution containing polyethylene glycol is the more appropriate choice for intranasal infusion. We thank Dr. Barr et al for their interest and clarification.

Eva Furrow, VMD

Department of Veterinary Clinical Sciences

College of Veterinary Medicine

University of Minnesota

Saint Paul, Minn

Reid Groman, DVM, DACVIM, DACVECC

Department of Clinical Studies

School of Veterinary Medicine

University of Pennsylvania

Philadelphia, Pa

Comments on a report involving a canine mandibular mass

I enjoyed the selection of a canine mandibular mass for a recent report¹ in the Pathology in Practice feature, but from the perspective of an oral surgeon, I would like to make some comments and add some clarifications that have important clinical implications, including an important update on differential diagnoses.

The authors stated that, as was the case for this dog, a lack of signs of pain has been reported for animals with ossifying fibromas of the face and jaw bones. However, a lack of signs of pain must not be interpreted as pathognomonic or even of any importance. Veterinary patients are frequently stoic and characteristically do not show signs of pain even with many advanced oral diseases.

The report provides a good example of how early detection of oral masses can be facilitated by appropriate client education. Clients should be educated to regularly perform oral and extraoral facial examinations, both visually and with palpation, and to seek professional care quickly for any abnormalities. Without detection by the client, this mass may have gone unnoticed during an awake physical examination of the dog by the veterinarian, especially if the dog had been uncooperative.

Four months is a long time for a client or veterinarian to monitor a mandibular mass before performing dental radiography to evaluate bony invasion. Delaying full assessment ignores the complexity and functionality of neighboring oral and dental structures that could be compromised by excision of even small but invasive masses.

In this case, results of histologic examination of two needle biopsy specimens were inconclusive. Although often challenging, identifying a large or invasive oral mass is a prerequisite for treatment planning, especially if radical surgery is contemplated.² In this case, the client may have requested circumventing this important step, but this was not mentioned.

Clinical staging by evaluation of regional lymph nodes³ and thoracic radiography is also warranted before radical surgery is performed, but no mention was made of whether these procedures were considered.

In discussing possible differential diagnoses for the mass in this dog, the authors stated that osteosarcomas of the face and jaw bones are rare in dogs. However, recent surveys reveal otherwise. A report⁴ from a private specialty practice indicated that osteosarcoma was the fourth most common tumor in a review of 157 malignant or locally invasive oral tumors in dogs, behind squamous cell carcinoma, fibrosarcoma, and malignant melanoma. Similarly, a survey⁵ of 626 oral lesions submitted for pathologic examination found the number of osteosarcomas was equal to the number of malignant melanomas and surpassed the number of squamous cell carcinomas.

This report demonstrated what veterinary surgeons have recognized for some time: that oral neoplasia can be histologically benign yet biologically malignant.⁶ The diagnostic and treatment challenges of oral masses are compounded by having odontogenic as well as nonodontogenic etiologies.⁷ Referral of any dog with a mass associated with the oral cavity to a specialist (as was done in this case) should always be considered.

Stanley W. Blazejewski III, VMD, DAVDC

Veterinary Referral Center

Malvern, Pa

The authors respond:

We wish to thank Dr. Blazejewski for his comments regarding the clinical aspects of our case. The information given in the report was as complete as could be collected by the authors. Decisions made by the owner during consultations with the referring veterinarian and referral surgeon and the reasoning behind those decisions were not available to the authors. Our intent in submitting this report was to make veterinarians providing dental surgery services and pathologists, particularly those evaluating dental surgical specimens, aware that bone matrix can be produced by tumors and lesions of the jaw, such as fibrous dysplasia, osteodystrophy fibrosa, ossifying epulis, and osteosarcoma. Our report adds ossifying fibroma to this list of differential diagnoses. Ossifying fibroma has not been commonly recognized by the veterinary profession, except for a separate lesion identified as juvenile ossifying fibroma of the incisor region in young horses. Ossifying fibroma of the mandibular and maxillary dental arcades is more common in our practice of pathology than osteosarcoma in dogs. Together, we have had over 80 years of experience in veterinary pathology with one of us having more than 40 years of experience specifically in the histopathologic diagnosis of bone tumors of the skeleton and dental tumors of domestic animals. It is our experience that several submissions of surgical specimens that have had an initial diagnosis of osteosarcoma are actually examples of ossifying fibromas. We have found the initial pathology report of these cases often contains comments that the specimen was considered to be a low-grade spindle cell tumor that had few or no mitotic figures and was therefore considered a low-grade osteosarcoma. It is the keen practitioner who observes that the radiographic appearance is more locally contained and less aggressive in nature than an osteosarcoma and requests a second opinion. In the case presented, the initial diagnosis was probable ossifying fibroma, but this information was kept from the readers to allow them to come up with their own list of differential diagnoses before the final diagnosis was given. Although the reports^1,2 cited by Dr. Blazejewski were not available at the time of submission of this article, care must still be taken when evaluating these tumor statistics to be certain that the diagnosis of osteosarcoma does not include tumors that would more appropriately be categorized as ossifying fibromas.

Molly C. Speltz, DVM, DACVP

Skyline Veterinary Hospital

Fridley, Minn

Roy R. Pool, DVM, PhD

Department of Veterinary Pathobiology

College of Veterinary Medicine and Biomedical Sciences

Texas A&M University

College Station, Tex

David W. Hayden, DVM, PhD, DACVP

Minnesota Veterinary Diagnostic Laboratory

College of Veterinary Medicine

University of Minnesota

Saint Paul, Minn

Working to eliminate cases of pyrethrin toxicosis

I was reminded of a topic that has been bothering me for a long time when I was sifting through emergency clinic faxes on a recent Monday morning. A young adult cat had been seen for treatment of seizures caused by pyrethrin toxicosis. The cat was eventually euthanized because the cost of treatment was more than the family could afford.

I am aghast that these products are still on discount and farm supply store shelves. I have seen toxicoses associated with these products not only in kittens and puppies but also in otherwise healthy adult dogs and cats. I also routinely see fleas on pets that were recently treated. I try to educate my clients on the danger and futility of using these products, but nothing has changed in the past 10 years. Is there anything the AVMA can do to spread the word?

Tracey Stevens, DVM

Genoa, NY

Is the intravenous zinc sulfate dose recommended for dogs with zinc-related cutaneous disorders too high?

A standard textbook recommends IV administration of zinc sulfate at a dose of 10 to 15 mg/kg (4.5 to 6.8 mg/lb) for treatment of dogs with zinc-related cutaneous disorders that do not respond to oral zinc administration.¹ Oral administration of zinc sulfate (10 mg/kg, q 24 h) is safe because the gut regulates absorption, but bypassing the gut increases the risk of toxicosis. The reference for the IV dose advises weekly administration of zinc sulfate at the stated dose, but a footnote states that the author uses a 50 mg/mL solution of zinc sulfate, equivalent to approximately 0.2 mg of elemental zinc/mL, and that panting was the only adverse effect when the injection was given slowly.²

There is a 100-fold mistake in this footnote. The molecular weight of anhydrous zinc sulfate is 161.5 g, and the atomic weight of zinc is 65.4 g. Thus, a 50 mg/mL zinc sulfate solution would contain 20.2 mg of elemental zinc/mL. If the solution did indeed contain only 0.2 mg of elemental zinc/mL, then a 10 to 15 mg/kg dose of zinc sulfate would be equivalent to a 40 to 67 μg/kg (18.2 to 30.5 μg/lb) dose of elemental zinc or approximately 1 mg of elemental zinc for a 25-kg (55-lb) dog. Because the blood volume of dogs is 90 mL/kg, this elemental zinc dose given as a bolus would increase zinc blood concentration transiently by 440 to 740 μg/L. In truth, however, the recommended zinc sulfate dose could increase blood zinc concentration by more than 40 mg/L.

The toxic IV dose of zinc in dogs has not been established, but the median blood zinc concentration was 31 mg/L (range, 5 to 159 mg/L) in 19 dogs with zinc intoxication secondary to coin ingestion, compared with a reference range of 1 to 2 mg/mL.³ Bolus infusion will increase blood zinc concentrations transiently, whereas coin ingestion results in long-term increases, but there must be a risk of toxicosis after bolus infusion of a high dose. In a previous study,⁴ administration of 0.002 mmol (131 μg) of zinc/kg to rats every hour for three hours via the splenic artery caused acute pancreatitis. A referral practice has also reported anuric renal failure developing in a dog with zinc-responsive dermatosis immediately after IV administration of an unstated dose of zinc.⁵

In humans receiving parenteral nutrition solutions IV, 3.2 to 6.5 mg of elemental zinc is infused daily.⁶ This represents 45 to 90 μg of zinc/kg (20.5 to 40.9 μg of zinc/lb) delivered over 24 hours (for a person weighing 70 kg [154 lb]). Dogs maintained plasma zinc concentrations for a week when 70 μg of elemental zinc/kg (31.8 μg of elemental zinc/ lb) was infused daily in an IV nutrition solution.⁷ These doses of zinc are comparable to the lower dose of elemental zinc recommended for treatment of zinc-responsive dermatoses, but with parenteral nutrition solutions, the zinc is delivered over 24 hours and not as a bolus. It is likely, therefore, that the lower dose (50 μg of elemental zinc/kg [22.7 μg of elemental zinc/lb]) can be administered safely to dogs if given over long periods of time, but the higher dose of zinc sulfate (10 to 15 mg/kg) may be toxic.

Richard C. Hill, MA, VetMB, PhD, DACVIM, DACVN

College of Veterinary Medicine

University of Florida

Gainesville, Fla