|
Clinical Pharmacology Mark G. Papich, DVM, MS Diplomate American College of Veterinary Clinical Pharmacology Professor of Clinical Pharmacology, College of Veterinary Medicine North Carolina State University, Raleigh, North Carolina Strategies for Antibacterial Therapy in Small Animals INTRODUCTION Antibiotic therapy has made many advances that has given veterinary medicine a large number of effective drugs and provided pharmacokinetic and pharmacodynamic information to guide dosing. New approaches to bacterial identification and susceptibility testing have helped to provide information for the most appropriate drug selection. This presentation will review the current concepts that guide antibiotic therapy in veterinary medicine and provide important strategies for effective dosing. BACTERIAL SUSCEPTIBILITY Most bacteria that cause infections come from the following list: Staphylococcus intermedius, (and occasionally other staphylococci) Escherichia coli, Klebsiella pneumoniae, Pasteurella multocida, beta-hemolytic streptococci, Pseudomonas aeruginosa, Proteus mirabilis (and occasionally indole-positive Proteus), Enterobacter spp and Enterococcus spp. If the bacteria are accurately identified, antibiotic selection is simplified because the susceptibility pattern of many organisms is predictable. For example, if the bacteria is likely to be Pasteurella, Streptococcus, or Actinomyces, susceptibility is expected to penicillin or an aminopenicillin such as ampicillin, amoxicillin, or amoxicillin-clavulanic acid (Clavamox). Usually Susceptible Bacteria Staphylococcus isolated from small animals is most likely to be S. intermedius rather than S. aureus. S. intermedius will usually have a predictable susceptibility to ß-lactamase resistant ß-lactam antibiotics such as amoxicillin combined with a ?-lactamase inhibitor (Clavamox), or first-generation cephalosporin such as cephalexin or cefadroxil, or the third-generation cephalosporin, cefpodoxime (Simplicef). Staphylococcus also is susceptible to oxacillin and dicloxacillin but these are not used as commonly in small animal medicine. Reports of studies on S. intermedius have shown that, despite frequent use of the above mentioned drugs in small animals, the incidence of resistance has not increased (Lloyd, et al, 1996). Most staphylococci are also sensitive to fluoroquinolones. The majority of staphylococci are sensitive to lincosamides (clindamycin, lincomycin), trimethoprim-sulfonamides, or erythromycin, but resistance can occur in as high as 25% of the cases. If the bacteria is an anaerobe (for example, Clostridium, Fusobacterium, Prevotella, Actinomyces, or Porphyromonas) predictable results can be attained by administering a penicillin, chloramphenicol, metronidazole, clindamycin, amoxicillin-clavulanic acid, or one of the second-generation cephalosporins such as cefotetan or cefoxitin. Metronidazole is consistently highly active against anaerobes including B. fragilis. The activity of first-generation cephalosporins, trimethoprim-sulfonamides/ormetoprim-sulfonamides, or fluoroquinolones for an anaerobic infection is unpredictable. If the anaerobe is from the Bacteroides fragilis group, resistance may be more of a problem because they produce a beta-lactamase that may inactivate 1st generation cephalosporins and ampicillin/amoxicillin. Some of these Bacteroides may also be resistant to clindamycin. More resistant strains of Bacteroides have been observed in recent years (Jang et al 1997). Problem, or Resistant Bacteria If the organism is Pseudomonas aeruginosa, Enterobacter, Klebsiella, Escherichia coli, or Proteus, resistance to many common antibiotics is possible and a susceptibility test is advised. For example, a report showed that among nonenteric E. coli, only 23% were sensitive to a 1st generation cephalosporin and less than half were sensitive to ampicillin. In the same study, 13%, and 23% were intermediate or resistant to enrofloxacin, and orbifloxacin, respectively (Oluoch, et al 2001). In urinary tract infections (Torres et al, 2005) half of the E. coli were resistant to cephalexin, and only 22% were sensitive to enrofloxacin. Based on these data as well as other studies, for initial therapy we usually expect the gram-negative enteric bacteria to be susceptible to fluoroquinolones and aminoglycosides. An extended-spectrum cephalosporin (second- or third-generation cephalosporin) usually is active against enteric-gram negative bacteria, but will not be active against Pseudomonas aeruginosa. If the organism is a Pseudomonas aeruginosa, inherent resistance against many drugs is common, but it may be susceptible to fluoroquinolones, aminoglycosides, or extended-spectrum penicillin such as ticarcillin or piperacillin. In one published study, the in vivo activity was examined in 23 strains of Pseudomonas: 19 Ps. aeruginosa, 3 Ps. fluorescens and one Pseudomonas spp. The most effective antibiotics were tobramycin (100% susceptible), marbofloxacin (91.3%) and ceftazidime (91.3%). Ticarcillin and gentamicin, showed good activity (86 and 65.2% respectively). Lower susceptibility was found with enrofloxacin (52.1%) (Martin Barrasa et al, 2000). Isolates of Pseudomonas aeruginosa from otitis media showed that 97% were susceptible to ceftazidime, and 81% to carbenicillin (Colombini et al 2000). Fewer were susceptible to enrofloxacin (51%) and gentamicin (68%). In a study that isolated Pseudomonas aeruginosa from the skin and ears of dogs, the pattern of resistance is similar (Petersen et al, 2002). There were no trends identified, and most isolates were susceptible to ciprofloxacin, piperacillin, ticarcillin, amikacin, and gentamicin (enrofloxacin was not tested). However, isolates from the ears tended to be more resistant than isolates from the skin, with lower susceptibility to topical drugs such as gentamicin. When administering a fluoroquinolone to treat Pseudomonas aeruginosa the high-end of the dose range is suggested. Of the currently available fluoroquinolones, (human or veterinary drugs) ciprofloxacin is the most active against Pseudomonas aeruginosa. BACTERIAL SUSCEPTIBILITY TESTING Bacterial susceptibility to drugs has traditionally been tested with the agar-disk-diffusion test (ADD), also known as the Kirby-Bauer test. With this test, paper disks impregnated with the drug are placed on an agar plate and the drug diffuses into the agar. Activity of the drug against the bacteria correlates with the zone of bacterial inhibition around the disk. The inoculation variables must be well controlled and the test must be performed according to strict procedural guidelines (Lorian, 1996). The precise incubation time (usually 18 to 24 hours), selection and preparation of the agar, and interfering compounds should be known. The ADD test results are qualitative (that is, it determines only resistant vs sensitive) rather than providing quantitative information. If this test is performed using standardized procedures, it is valuable, even though it may sometimes overestimate the degree of susceptibility. MIC Determination It is becoming more common for laboratories to directly measure the minimum inhibitory concentration (MIC) of an organism with an antimicrobial dilution test. The test is usually performed by inoculating the wells of a plate with the bacterial culture and dilutions of antibiotics are arranged across the rows. The MIC can be directly determined by observing the lowest concentration required to inhibit bacterial growth. In some laboratories other methods to measure the MIC are being used such as the E-test (epsilometer test) by AB Biodisk. The E-test is a quantitative technique which measures the MIC by direct measurement of bacterial growth along a concentration gradient of the antibiotic contained in a test strip. Resistance and susceptibility are determined by comparing the organism's MIC to the drug's breakpoint as established by the Clinical and Laboratory Standards Institute (CLSI) - formerly known as the National Committee for Clinical Laboratory Standards (NCCLS) (NCCLS, 2002 & 2004). After a laboratory determines an MIC, it may use the CLSI "SIR" classification for breakpoints (S, susceptible; I, intermediate, or R, resistant). In everyday practice, if the MIC for the bacterial isolate falls in the susceptible category, there is a greater likelihood of successful treatment (cure) than if the isolate were classified as resistant. It does not assure success; drug failure is still possible owing to other drug or patient factors (for example, immune status, immaturity, or severe illness that compromises the action of antibacterial drugs), and interactions. If the MIC is in the resistant category, bacteriologic failure is more likely because of specific resistance mechanisms or inadequate drug concentrations in the patient. However, a patient with a competent immune system may sometimes eradicate an infection even when the isolate is resistant to the drug in the MIC test. The intermediate category is intended as a buffer zone between susceptible and resistant strains. This category reflects the possibility of error when an isolate has an MIC that borders between susceptible and resistant. The intermediate category is not intended to mean "moderately susceptible." If the MIC value is in the intermediate category, therapy with this drug at the usual standard dosage is discouraged because there is a good likelihood that drug concentrations may be inadequate for a cure. However, successful therapy is possible when drug concentrates at certain sites - in urine, or as the result of topical therapy, for example - or at doses higher than the minimum effective dose listed on the label. For example, fluoroquinolone antimicrobials have been approved with a dose range that allows increases in doses when susceptibility testing identifies an organism in the Intermediate range of susceptibility. In these cases higher drug concentrations make a cure possible if the clinician is able to safely increase the dose above the minimum labeled dose. (For example, in the case of enrofloxacin in dogs, this would be equivalent to a dose of 10 to 20 mg/kg/day, rather than the minimum dose of 5 mg/kg/day.) MIC tests are more quantitative than an ADD test, but must be performed according to strict guidelines (NCCLS 2004; Lorian, 1996). In some cases, even when the breakpoint is below the "susceptible" range, the organism is resistant in vivo. Examples include cephalosporins for treating oxacillin-resistant staphylococci, and ampicillin for treating ?-lactamase producing staphylococci. PENETRATION TO THE SITE OF INFECTION For most tissues, antibiotic drug concentrations in the serum or plasma approximate the drug concentration in the extracellular space (interstitial fluid). This is because there is no barrier that impedes drug diffusion from the vascular compartment to extracellular tissue fluid (Nix et al, 1991). There is really no such thing as "good penetration" and "poor penetration" when referring to most drugs in most tissues. Pores (fenestrations) or microchannels in the endothelium of capillaries are large enough to allow drug molecules to pass through unless the drug is restricted by protein binding in the blood. Tissues lacking pores or channels may inhibit penetration of some drugs (discussed below). Diffusion Into Tissues Diffusion of most antibiotics from plasma to tissues is limited by tissue blood flow, rather than drug lipid solubility. This has been called perfusion-rate limited drug diffusion. If adequate drug concentrations can be achieved in plasma, it is unlikely that a barrier in the tissue will prevent drug diffusion to the site of infection as long as the tissue has an adequate blood supply. Rapid equilibration between the extracellular fluid and plasma is possible because of high surface area:volume ratio (high SA:V). That is, the surface area of the capillaries is high relative to the volume into which the drug diffuses. Drug diffusion into an abscess or granulation tissue is sometimes a problem because in these conditions drug penetration relies on simple diffusion and the site of infection lacks adequate blood supply. In an abscess, there may not be a physical barrier to diffusion - that is, there is no impenetrable membrane - but low drug concentrations are attained in the abscess or drug concentrations are slow to accumulate because in a cavitated lesion there is low surface area to volume ratio (low S:V ratio). In some tissues a lipid membrane (such as tight junctions on capillaries) presents a barrier to drug diffusion. This has been called permeability-rate limited drug diffusion. In these instances, a drug must be sufficiently lipid-soluble, or be actively carried across the membrane in order to reach effective concentrations in tissues. These tissues include: the central nervous system, eye, and prostate. A functional membrane pump (p-glycoprotein) also contributes to the barrier. There also is a barrier between plasma and bronchial epithelium (blood:bronchus barrier). This limits drug concentrations of some drugs in the bronchial secretions and epithelial fluid of the airways. Lipophilic drugs may be more likely to diffuse through the blood-bronchus barrier and reach effective drug concentrations in bronchial secretions. Intracellular Infections Most bacterial infections are located extracellular, and a cure can be achieved with adequate drug concentrations in the extracellular (interstitial) space rather than intracellular space. Intracellular infections present another problem. For drugs to reach intracellular sites, they must be carried into the cell or diffuse passively. Generally, lipid-soluble drugs are best able to diffuse through the cell membrane for intracellular infections. Examples of drugs that accumulate in leukocytes, fibroblasts, macrophages, and other cells are fluoroquinolones, lincosamides (clindamycin, lincomycin), macrolides (erythromycin, clarithromycin), and the azalides (azithromycin) (Pasqual, 1995). ?-lactam antibiotics and aminoglycosides do not reach effective concentrations within cells. Intracellular organisms such as Brucella, Chlamydia, Rickettsia, Bartonella and Mycobacteria are examples of intracellular pathogens. Staphylococci may in some cases become resistant to treatment because of intracellular survival. Fluoroquinolones and tetracyclines such as doxycycline are frequently administered to treat Rickettsia and Ehrlichia infections. There is good evidence for efficacy of doxycycline or fluoroquinolones (enrofloxacin is the only one tested) for treating Rickettsia, but only doxycycline should be considered for its efficacy for treating canine ehrlichiosis. LOCAL FACTORS THAT AFFECT ANTIBIOTIC EFFECTIVENESS Local tissue factors may decrease antimicrobial effectiveness. For example, pus and necrotic debris may bind and inactivate vancomycin or aminoglycoside antibiotics (gentamicin or amikacin), causing them to be ineffective. Cellular material also can decrease the activity of topical agents such as polymyxin B. Foreign material in a wound (such as material surgically implanted) can protect bacteria from antibiotics and phagocytosis by forming a biofilm (glycocalyx) at the site of infection (Habash & Reid, 1999; Smith 2005). Cellular debris and infected tissue can inhibit the action of trimethoprim-sulfonamide combinations through the secretion of thymidine and PABA, both known to be inhibitors of the action of these drugs. This may explain why trimethoprim-sulfonamide combinations have not been effective in some infected tissues. Cations can adversely affect the activity of antimicrobials at the site of infection. Two important drug groups diminished in activity by cations such as Mg++, Al+3, Fe+3, and Ca++ are fluoroquinolones and aminoglycosides. (Cations such as magnesium, iron, and aluminum also can inhibit oral absorption of fluoroquinolones.) An acidic environment of infected tissue may decrease the effectiveness of clindamycin, erythromycin, fluoroquinolones, and aminoglycosides. Penicillins and tetracycline activity is not affected as much by tissue pH, but hemoglobin at the site of infection will decrease the activity of these drugs. An anaerobic environment decreases the effectiveness of aminoglycosides because oxygen is necessary for drug penetration into bacteria. As mentioned previously, an adequate blood flow is necessary to deliver an antibiotic to the site of infection. Effective antibacterial drug concentrations may not be attained in tissues that are poorly vascularized (eg, extremities during shock, sequestered bone fragments, and endocardial valves). PHARMACOKINETIC-PHARMACODYNAMIC (PK-PD) OPTIMIZATION OF DOSES To achieve a cure, the drug concentration in plasma, serum, or tissue fluid should be maintained above the minimum inhibitory concentration (MIC), or some multiple of the MIC, for at least a portion of the dose interval. Antibacterial dosage regimens are based on this assumption, but drugs vary with respect to the peak concentration and the time above the MIC that is needed for a clinical cure. Pharmacokinetic-pharmacodynamic (PK-PD) relationships of antibiotics attempt to explain how these factors can correlate with clinical outcome (Nicolau et al. 1995, Hyatt et al. 1995). Shown on Figure 1 are some terms used to describe the shape of the plasma concentration vs time profile. The CMAX is simply the maximum plasma concentration attained during a dosing interval. The CMAX is related to the MIC by the CMAX:MIC ratio. The AUC is the total area-under-the-curve. The AUC for a 24 hour period is related to the MIC value by the AUC:MIC ratio. Also shown in Figure 1, is the relationship of time to MIC measured in hours (T > MIC). Antibiotics can be bactericidal, bacteriostatic, depending on the drug and the organism. For some drug, certain organisms the drugs are bacteriostatic, and against other organisms they are bactericidal. For a drug that is bactericidal, it may be either concentration-dependent in its action, or time-dependent. If concentration-dependent, one should administer a high enough dose to maximize the CMAX: MIC ratio. If time-dependent, the drug should be administered frequently enough to maximize the T > MIC. For bacteriostatic drugs, the drug concentration should be kept above the MIC at the site of action for as long as possible during the dosing interval. Examples of how these relationships affect drug regimens are described below: Aminoglycosides Aminoglycosides (eg, gentamicin, or amikacin) are concentration-dependent bactericidal drugs, therefore the higher the drug concentration, the greater the bactericidal effect. An optimal bactericidal effect occurs if a high enough dose is administered to produce a peak of 8-10x the MIC. This can be accomplished by administering a single dose once daily. This regimen is at least as effective, and perhaps less nephrotoxic, than lower doses administered more frequently (Freeman et al, 1997). Our current regimens in small animals employ this strategy. The single daily dose is based on the drug's volume of distribution (calculated using the area method). A once daily dose for gentamicin is 5-8 mg/kg for cats, and 10-14 mg/kg for dogs, once daily. An appropriate dose for amikacin is 10-15 mg/kg for cats and 15-30 mg/kg for dogs once daily. The efficacy of these regimens has not been tested for conditions encountered in veterinary medicine, but the relationships are supported by experimental evidence. These regimens assume some competency of the immune system. If the animal is immunocompromised, one may consider a more frequent interval for administration. In animals with decreased renal function, longer intervals may be considered. Fluoroquinolones For the fluoroquinolone antimicrobials, as reviewed by Hyatt et al (1995), Dudley (1991), and recently by Wright et al (2000) and Papich & Riviere (2001) investigators have shown that either the peak plasma concentration above bacterial minimum inhibitory concentration (MIC), also known as the CMAX:MIC ratio, or the total AUC above the MIC (also known as the AUC:MIC ratio), may predict clinical cure in studies of laboratory animals, and in a limited number human clinical studies. There are no published studies involving dogs or cats that indicate which of these parameters is the best predictor of clinical cure, or what the respective target ratios might be. Therefore, the optimum value for these surrogate markers has not been determined for infections in dogs or cats. However, derived from other studies, a CMAX:MIC of 8-10, or a AUC:MIC of greater than 125 have been associated with a cure. As reviewed by Wright et al (2000), for some clinical situations AUC:MIC ratios as low as 30-55 for a clinical cure, since the study in which 125 was cited involved critically ill human patients. This difference may also be organism specific. Sensitive bacteria from small animals might be expected to have an MIC for fluoroquinolones in the range of 0.125 mg/mL, (+/- one dilution) (Pirro et al 1999). Using this value for MIC, the administration of the lowest label dose of any of the currently available fluoroquinolones usually meets the goal of a CMAX:MIC ratio or a AUC:MIC ratio in the range cited above. To take advantage of the flexible dosing for fluoroquinolones, low doses of fluoroquinolones have been administered to treat susceptible organisms with low MIC, such as E. coli or Pasteurella. But, for bacteria with a higher MIC, (for example gram-positive cocci) a slightly larger dose can be used. To achieve the necessary peak concentration for a bacteria such as Pseudomonas aeruginosa, that usually has the highest MIC among susceptible bacteria, the highest dose in a range is recommended. Bacteria such as streptococci and anaerobes are more resistant and even at high doses, a sufficient peak concentration or AUC:MIC ratio will be difficult to achieve. Beta-lactam antibiotics ?-lactam antibiotics such as penicillins, potentiated-aminopenicillins, and cephalosporins are slowly bactericidal. Their concentration should be kept above the MIC throughout most of the dosing interval (long T>MIC) for the optimal bactericidal effect (Turnidge 1998). Dosage regimens for the ?-lactam antibiotics should consider these pharmacodynamic relationships. Therefore, for treating a gram-negative infection, especially a serious one, some regimens for penicillins and cephalosporins require administration 3 to 4 times per day. Some long-acting formulations have been developed to prolong plasma concentrations. Some of the third-generation cephalosporins have long half-lives and less frequent regimens have been used for some of these drugs (for example cefpodoxime proxetil, cefotaxime and ceftiofur). (However, the long half-life for ceftriaxone in people does not occur in animals because of differences in drug protein binding.) Gram-positive organisms are more susceptible to the ?-lactams than are gram-negative bacteria. Additionally, since the MICs are lower for gram-positive bacteria, and antibacterial effects occur at concentrations below the MIC (post antibiotic effect or PAE), longer dose intervals may be possible for infections caused by gram-positive as compared to gram-negative bacteria. For example, cephalexin or amoxicillin-clavulanate have been used successfully to treat staphylococcal infections when administered only once daily (although twice-daily administration is recommended to obtain maximum response). Cefpodoxime proxetil (Simplicef) is effective for once-daily administration, which is due to both high activity (low MIC values) and a longer half-life compared to other cephalosporins. Bacteriostatic drugs The drugs such as tetracyclines, macrolides (erythromycin and derivatives), sulfonamides, lincosamides (lincomycin and clindamycin), and chloramphenicol derivatives act in a bacteriostatic manner against most bacteria. However, against susceptible gram-positive bacteria, the macrolides appear to be bactericidal and can demonstrate a post-antibiotic effect. Chloramphenicol also can produce a bactericidal effect if the organism is very susceptible. These drugs are most effective when the drug concentrations are maintained above the MIC throughout the dosing interval. In this way, they act in a time-dependent manner. Even in situations in which macrolides act in a bactericidal manner, their action is still time-dependent because the bactericidal action is so slow. Most of the bacteriostatic drugs must be administered frequently to achieve this goal. However, a property of some of these drugs is that they persistent in tissues for a prolonged time, which allows infrequent dosing intervals. The cattle drug tilmicosin (Micotil) attains drug concentrations in lungs for at least three days for treating susceptible Pasteurella bacteria. The macrolide derivative azithromycin (Zithromax) has shown tissue half-lives as long as 70-90 hours in cats and dogs, permitting infrequent dosing. Tissue concentrations of trimethoprim-sulfonamides persist long enough to allow once-daily dosing for many infections. Most published dosage regimens are designed to take the pharmacokinetic properties of these drugs into account. SUGGESTED EMPIRICAL TREATMENT BASED ON TISSUE SITE On the following table is a list that includes some (but not all) possible choices for common infections encountered in veterinary medicine. In this list the "first choice" is a drug with a high likelihoood of success, low expense and few risks. If the first choice has not been effective, or if patient factors preclude using the first choice (eg, allergy) the alternate choice should be considered.
+ Fluoroquinolone = enrofloxacin, difloxacin, marbofloxacin or orbifloxacin # Extended spectrum cephalosporin = 2nd - or 3rd-generation drugs (eg, cefotetan, cefotaxime, cefpodoxime). REFERENCES CITED Colombini S, Merchant RS, and Hosgood G. Microbial flora and antimicrobial susceptibility patterns from dogs with otitis media. Veterinary Dermatology 11: 235-239, 2000. Freeman CD, Nicolau DP, Belliveau PP, Nightingale CH: Once-daily dosing of aminoglycosides: review and recommendations for clinical practice. J Antimicrob Chemother 39: 677, 1997. Habash M, & Reid G (1999) Microbial biofilms: their development and significance for medical device-related infections. J Clin Pharmacol 39: 887-898. Hyatt JM, McKinnon PS, Zimmer GS, Schentag JJ: The importance of pharmacokinetic/pharmacodynamic surrogate markers to outcome. Clin Pharmacokinet 28: 143-160, 1995. Jang SS, Breher JE, Dabaco LA, & Hirsh DC (1997) Organisms isolated from dogs and cats with anaerobic infections and susceptibility to selected antimicrobial agents. J Am Vet Med Assoc 210: 1610-1614. Lloyd DH, Lamport AI, Feeney C: Sensitivity to antibiotics amongst cutaneous and mucosal isolates of canine pathogenic staphylococci in the UK, 1980-1996. Vet Derm 7: 171-175, 1996. Lode, H., Borner, K., Koeppe, P. Pharmacodynamics of Fluoroquinolones. Clinical Infectious Diseases. 1998; 27: 33-9. Lorian V: Antibiotics in Laboratory Medicine 4th Ed. Williams & Wilkins, 1996. Martin Barrasa JL, Lupiola Gomez P, Gonzalez Lama Z, et al. Antibacterial susceptibility patterns of Pseudomonas strains isolated from chronic canine otitis externa. J Vet Med B Infect Dis Vet Public Health 2000;47:191-6. Meinen JB, McClure JT, Rosin E: Pharmacokinetics of enrofloxacin in clinically normal dogs and mice and drug pharmacodynamics in neutropenic mice with Escherichia coli and staphylococcal infections. Am J Vet Res 56: 1219-1224, 1995. NCCLS (2004): Performance standards for antimicrobial susceptibility testing; fourteenth informational supplement. M100-S14, 2004; 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087. NCCLS (2002): Performance standards for antimicrobial disk and dilution susceptibility tests for bacteria isolated from animals; approved standards - second edition. M31-A2, 2002; 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087. Nicolau DP, Quintiliani R, Nightingale CH: Antibiotic kinetics and dynamics for the clinician. Med Clinics North America 79: 477-495, 1995. Nix DE, Goodwin SD, Peloquin CA, et al. Antibiotic tissue penetration and its relevance: Impact of tissue penetration on infection response. Antimicrob Agents Chemother 35: 1953-1959, 1991. Oluoch AO, Kim C-H, Weisiger RM, et al. Nonenteric Escherichia coli isolates from dogs: 674 cases (1990-1998). J Am Vet Med Assoc 218: 381-384, 2001. Papich MG, Riviere JE: Fluoroquinolones. In: Adams HR (editor) Veterinary Pharmacology and Therapeutics, 8th Edition, Iowa State University Press, Ames, Iowa, 2001. Pascual A: Uptake and intracellular activity of antimicrobial agents in phagocytic cells. Rev Med Microbiol 6: 228-235, 1995. Petersen AD, Walker RD, Bowman MM, Schott HC, Rosser EJ. Frequency of isolation and antimicrobial susceptibility patterns of Staphylococcus intermedius and Pseudomonas aeruginosa isolates from canine skin and ear samples over a 6 year period (1992-1997). J Am Anim Hosp Assoc 38: 407-413, 2002. Pirro, F., Edingloh, M., Schmeer, N., (1999) Bactericidal and inhibitory activity of enrofloxacin and other fluoroquinolones in small animal pathogens. Compendium on Continuing Education for the Practicing Veterinarian 21 (Supplement 10). Smith AW. Biofilms and antibiotic therapy: Is there a role for combating resistance by the use of novel drug delivery systems? Advanced Drug Delivery Reviews 57: 1539-1550, 2005. Torres SM, Diaz SF, Nogueira SA, Jessen C, Polzin DJ, Gilbert SM, and Horne LK. Frequency of urinary tract infection among dogs with pruritic disorders receiving long-term glucocorticoid treatment. J Am Vet Med Assoc 227: 239-243, 2005. Turnidge JD. The pharmacodynamics of ?-lactams. Clinical Infectious Diseases. 1998; 27: 10-22. Wright DH, Brown GH, Peterson ML, Rotschafer JC. Application of fluoroquinolone pharmacodynamics. Journal of Antimicrobial Chemotherapy 46: 669-683, 2000. Update on New Drug for Antimicrobial Therapy INTRODUCTION Treatment of common infections in small animals has been reported to provide guidelines and established regimens. Drug manufacturers have produced several important drugs to treat the most common infections encountered in small animals. However, the drugs and approaches to therapy are more limited when the infection is more refractory, resistant, or is associated with another complicating factor. Susceptibility of the most common isolates has been documented well enough to make sound judgments and empirical antimicrobial drug choices. However, when the patient has a refractory and/or resistant infection, or is seriously ill with an infection, other strategies and drugs may be necessary. As with many new treatments, there are few veterinary clinical studies to support a recommended use and dose and many of these details have been extrapolated from human medicine. ARE THERE ANY NEW DRUGS AVAILABLE (OR NEW INFORMATION ON OLD DRUGS)? Cephalosporins We are familiar with the cephalosporins commonly referred to as the 1st -generation cephalosporins represented by the oral drugs cephalexin (Keflex) and cefadroxil (Cefa-Tabs, Cefa-Drops), and the injectable drug cefazolin. These drugs have a spectrum of activity that includes staphylococci, streptococci, and many of the enteric gram-negative bacilli. However, resistance among gram-negative bacteria develops easily, primarily from synthesis of ?-lactamase enzymes that can hydrolyze these drugs. Extended-spectrum cephalosporins include cephalosporins from the 2nd, 3rd, and 4th generation. Traditionally, in veterinary medicine the use of extended-spectrum cephalosporins has been reserved for treatment of bacterial infections that are resistant to other drugs. The bacteria often identified in these resistance problems have been Escherichia coli, Klebsiella pneumoniae, Enterobacter species, Proteus species (especially indole-positive), and Pseudomonas aeruginosa. However, there are approved extended-spectrum cephalosporins registered for routine use available for once-daily treatment in dogs. New developments are discussed below. Of the 2nd-generation cephalosporins, the ones used most often in veterinary medicine are cefoxitin and cefotetan. Their use has been valuable for treating organisms resistant to the 1st generation cephalosporins or in cases in which there are anaerobic bacteria present. Anaerobic bacteria such as those of the Bacteroides fragilis group can become resistant by synthesizing a cephalosporinase enzyme. Cefoxitin and cefotetan, which are in the cephamycin group, are resistant to this enzyme and may be active against these bacteria. Therefore, these drugs may be valuable for some cases such as septic peritonitis that may have a mixed population of anaerobic bacteria and gram-negative bacilli. The 3rd -generation cephalosporins are the most active of the cephalosporins against gram negative bacteria, especially enteric bacteria that are resistant to other cephalosporins. Most drugs in this group (exceptions discussed below) are intended for IV or IM administration. For convenience, some have been administered to animals SC. But one should be warned that the IM or SC administration of these drugs could be irritating and painful. One of the most frequently administered drugs in this group is cefotaxime (Claforan) because of its potency and activity against most enteric gram-negative bacteria and some streptococci. Compared to other cephalosporins, ceftazidime is the most active against Pseudomonas, against which all of the other cephalosporins, except cefoperazone, have little or no activity. Doses have been derived for ceftazidime in dogs based on pharmacokinetic data (Moore et al, 2000). Doses vary depending on the indication, but 30 mg/kg every 8 hours IM, SC, or IV will maintain effective concentrations for many infections. Since the drugs mentioned are all injectable, there has been a need for an oral extended-spectrum cephalosporin. Of the human drugs available, cefixime (Suprax) has been used in dogs because it is one of the few 3rd -generation cephalosporins that can be administered orally. Doses have been derived from pharmacokinetic studies (Bialer et al, 1987; Lavy et al 1995). The doses have ranged from 5 to 10 mg/kg twice daily orally. Another oral 3rd-generation cephalosporin is cefpodoxime proxetil. This drug has been used off-label in its human form (Vantin) by veterinarians for several years, but now there is a veterinary formulation (Simplicef). This drug has a longer half-life than most other cephalosporins. To treat the indications for which it is registered in dogs (skin infections), it can be given once a day orally at a dose of 5-10 mg/kg. Its absorption from oral administration is good (63%) compared to other oral third generation cephalosporins and it is excreted mostly in the urine with a half-life of 5.6-6 hours. For more serious or refractory infections against bacteria that may have high MIC values (off-label indications), or for treating immunosuppressed animals with gram-negative infections, this author suggests that veterinarians consider twice a day frequency instead of the labeled frequency of once-a-day in order to maintain the drug concentration above the MIC for a longer period. Since cefpodoxime is excreted in the urine, it may have efficacy against urinary tract infections, but clinical results have not been reported. Cefpodoxime proxetil is not registered for use in cats. Other cephalosporins have been used safely in cats, with doses extrapolated from dogs, but there is no data reported for cefpodoxime. Cefpodoxime is more active than many other third-generation cephalosporins against Staphylococcus. However, it is not active against Pseudomonas aeruginosa, Enterococcus, or methicillin-resistant Staphylococcus. One should be aware that the break-point for susceptibility is lower than for other third-generations cephalosporins. Therefore, it is possible for a bacterial isolate to be sensitive to cefotaxime or ceftazidime (breakpoint 8 µg/mL) but resistant to cefpodoxime (breakpoint 2 µg/mL) (NCCLS 2002). Specific disks are suggested for testing bacterial isolates, rather than relying on the results from other cephalosporins. The most recent development in this class is the 4th generation cephalosporins. The first 4th generation cephalosporin is cefepime (Maxipime). It is unique from other cephalosporins because of its broad spectrum of activity that includes gram positive cocci, enteric gram negative bacilli, and Pseudomonas. It has the advantage of activity against some extended-spectrum ?-lactamase (ESBL) producing strains of Klebsiella and E. coli that have become resistant to many other ?-lactam drugs and fluoroquinolones. Except for one investigation in dogs, adult horses, and foals, the use of cefepime has been limited in veterinary medicine (Gardner, et al. 2001). Carbapenems The carbapenems are beta-lactam antibiotics that include imipenem-cilastatin sodium (Primaxin), meropenem (Merrem), and most recently, ertapenem (Invanz). Imipenem is administered with cilastatin to decrease renal tubular metabolism. Cilastatin does not affect the antibacterial activity. Imipenem has become a valuable antibiotic because it has a broad spectrum that includes many bacteria resistant to other drugs (Edwards & Betts, 2000). Imipenem is not active against methicillin-resistant staphylococci or resistant strains of Enterococcus faecium. The high activity of imipenem is attributed to its stability against most of the ?-lactamases (including ESBL) and ability to penetrate porin channels that usually exclude other drugs (Livermore 2001). The carbapenems are more rapidly bactericidal than the cephalosporins and less likely to induce release of endotoxin in an animal from gram-negative sepsis. Resistance to carbapenems has been extremely rare in veterinary medicine. Some disadvantages of imipenem are the inconvenience of administration, short shelf-life after reconstitution, and high cost. It must be diluted in fluids prior to administration. A common dose for small animals is 10 mg/kg q8h or 5 mg/kg q6h. This dose must be given by constant rate infusion over 30-60 minutes, but it has been administered subcutaneously. One of the adverse effects caused from imipenem therapy is seizures. Another problem is the risk of renal injury, which should be minimized by the addition of cilastatin (Barker et al, 2003). Meropenem, one of the newest of the carbapenem class of drugs has antibacterial activity approximately equal to, or greater than imipenem. Other characteristics are similar to imipenem. Its advantage over imipenem is that it is more soluble and can be administered in less fluid volume and more rapidly. For example, small volumes can be administered subcutaneously with almost complete absorption. There also is a lower incidence of adverse effects to the central nervous system, such as seizures (Edwards & Betts, 2000). Based on pharmacokinetic experiments in our laboratory (Bidgood & Papich, 2002), the recommended dose for Enterobactericeae and other sensitive organisms is 8.5 mg/kg SC every 12hr, or 24 mg/kg IV every 12 hr. For infections caused by Pseudomonas aeruginosa, or other similar organisms that may have MIC values as high as 1.0 mcg/mL: 12 mg/kg q8h, SC, or 25 mg/kg q8h, IV. For sensitive organisms in the urinary tract, 8 mg/kg, SC, every 12 hours can be used. In our experience, these doses have been well-tolerated except for slight hair loss over some of the SC dosing sites. Ertapenem is the newest drug in this class. It has a longer half-life in people and can be administered once a day. Experiments are underway in animals to determine the optimum dosing. Ertapenem has good activity against most gram-negative organisms, except Pseudomonas aeruginosa. Fluoroquinonlones The fluoroquinolones approved in the U.S. for animals include enrofloxacin, marbofloxacin, difloxacin, and orbifloxacin. In the U.S. all of these drugs are approved for dogs; orbifloxacin, marbofloxacin, and enrofloxacin are approved for cats. Enrofloxacin 100 mg/mL injection and danofloxacin (A180) injection are approved for cattle. A topical formulation of enrofloxacin and silver sulfadiazine (Baytril Otic) is registered for treating otitis in dogs. There are several other fluoroquinolones approved for use in human medicine (ciprofloxacin, lomefloxacin, enoxacin, ofloxacin), but their used has been limited in veterinary medicine, except for ciprofloxacin. The mechanisms of action and important pharmacological properties have been reviewed elsewhere (Papich & Riviere, 2001). These drugs have as their advantages: (1) spectrum of activity that includes most gram-negative bacteria and many gram-positive bacteria, including staphylococci, (2) oral and injectable administration, and (3) good safety profile. Important deficiencies in the spectrum of activity include gram-positive cocci, especially enterococci (Enterococcus faecalis and Enterococcus faecium), and anaerobic bacteria. The newest generations of fluoroquinolones (referred to by some authors as the 3rd-generation fluoroquinolones) include trovafloxacin, grepafloxacin, gatifloxacin, gemifloxacin, and moxifloxacin. Two of these, trovafloxacin and grepafloxacin, have already been discontinued for use in people because of adverse effects (abnormal cardiac rhythms and hepatic injury). The new generation of fluoroquinolones, with substitutions at the C-8 position, (C-8 methoxy for example) have as their advantage a broader spectrum that includes anaerobic bacteria and gram-positive cocci. The difference in spectrum of activity is largely caused by increased activity against the DNA-gyrase of gram-positive bacteria, rather than activity against Topoisomerase IV, which is the target in gram-positive bacteria for the older quinolones (Pestova et al, 2000; Hooper, 2000). Premafloxacin, a veterinary 3rd generation quinolone is the only one of this class for which reports are available on its potential in veterinary medicine (Watts et al, 1997), but this drug will not be available. Pradofloxacin has been evaluated in dogs and cats, but the experience thus far is limited to a few research abstracts (de Jong et al, 2004; Stephan et al, 2004). It was more active than other fluoroquinolones against bacterial isolates from dogs and cats (de Jong et al 2004). At a dose of 3 mg/kg orally it was effective for treatment of urinary tract infections in dogs (Stephan et al, 2004). Moxifloxacin (Avelox), is a human drug of this group and has been used on a limited basis for treatment of infections in dogs and cats caused by bacteria that have been refractory to other drugs. Of the currently available fluoroquinolones, all have a similar spectrum of activity, but they may vary in potency. Against some gram-negative bacilli, especially Pseudomonas aeruginosa, the human drug ciprofloxacin is more active than veterinary quinolones. Enrofloxacin in small animals is metabolized to ciprofloxacin, which may account for 10-20% of the total quinolone maximum plasma concentrations (CMAX) and as much as 35% of the total AUC (Cester et al 1996). Orbifloxacin and marbofloxacin have little or no active metabolites, but they are well-absorbed and achieve higher plasma concentrations after equivalent doses compared to enrofloxacin and difloxacin. Generally, all of the veterinary fluoroquinolones attain good concentrations in tissues, with tissue:plasma concentration ratios approaching, or greater than 1.0 (Bidgood & Papich, 2004). Fluoroquinolones have had a good safety record after administration to animals. Central nervous system effects, such as seizures, may occur at high doses but are rare. In young animals, especially dogs and foals, arthropathy of the developing cartilage is possible, leading to joint injury and lameness. Recently, blindness in cats caused by fluoroquinolones has attracted attention. The labeled dose for enrofloxacin (Baytril) use in cats in the U.S. was recently changed by the drug manufacturer (Bayer Corporation). Because of dose-related ocular toxicosis, the dose in cats should not exceed 5 mg/kg/day. The mechanism for the toxicity is not understood, but degenerative lesions in the retina have been identified. In studies performed by the manufacturer, there were no adverse effects observed in cats treated with 5 mg/kg/day of enrofloxacin. However, the administration of enrofloxacin at 20 mg/kg or greater, caused salivation, vomiting, and depression. At doses of 20 mg/kg or greater, there were mild to severe fundic lesions on ophthalmologic examination including changes in the fundus and retinal degeneration. There was also abnormal electroretinograms, including blindness. Ocular problems have not been reported in other species. Besides enrofloxacin, the other fluoroquinolones registered for use in cats are orbifloxacin(Orbax) and marbofloxacin (Zeniquin). The current approved dose of orbifloxacin for cats is 2.5-7.5 mg/kg/day. In a published study, (Kay-Mugford et al, 2001) orbifloxacin oral liquid was administered to cats at 0, 15, 45, and 75 mg/kg for at least 30 days (8 cats/group). This represents 6, 18, and 30 x the lowest label dosage. No ocular lesions were observed in any cats treated with 15 mg/kg. At the higher doses, (18 and 30 x dose) there was tapetal hyperreflectivity in the area centralis and minimal photoreceptor degeneration. When marbofloxacin was administered to cats at 5.55, 16.7, and 28 mg/kg, representing 2, 6, and 10x the lowest label dose, for 6 weeks there were no ocular lesions in cats (manufacturer's data). At 55.5 mg/kg (10 x the lowest label dose) for 14 days there were also no lesions from marbofloxacin. Ciprofloxacin use in animals: We encourage veterinarians to consider veterinary-labeled fluoroquinolones in their patients first because safety and efficacy data have been specifically derived for animals before FDA approval. Ciprofloxacin is a human drug, not registered for animals. However, it can be used legally by veterinarians, as long as it is not administered to food animals. The use would be considered extra-label and subject to other extra-label restrictions (eg, a veterinarian-client-patient-relationship - VCPR - need to be established). Since ciprofloxacin became available in a generic formulation, there has been interest in its use in animals because of a difference in cost. In cats, it was safe when administered at 100 mg/kg without producing ocular toxicity (Schluter 1987). When cats were given ciprofloxacin orally oral absorption is low 22-33% and would not be effective for gram-positive bacteria even at 10 mg/kg (Albarellos et al, 2004). At 10 mg/kg every 12 hours, it was able to reach therapeutic targets against susceptible gram-negative bacteria. In dogs, oral absorption of ciprofloxacin has been reported in only a few limited studies. Estimates derived from independent studies (Abadía et al, 1994; Abadía et al, 1995; Walker et al 1990), indicates that oral absorption may approach 74 to 97%. However, in the only crossover study reported, (Nakamura et al, 1990) the oral absorption in dogs was only 42%. Other fluoroquinolones registered for animals have near complete bioavailability. The potentially low ciprofloxacin oral availability for dogs suggests that doses should be higher than the doses currently used for drugs such as enrofloxacin, marbofloxacin, or orbifloxacin. Injectable ciprofloxacin is available in a human formulation, usually 10 mg/mL (in sterile water) or 2 mg/mL (premixed with 5% dextrose). Clinical use of this formulation has not been reported in animals. For humans, it is recommended to dilute the concentrated form to 1-2 mg/mL prior to IV use with an intravenous solution and infuse the final solution over 60 minutes. Administration protocols have not been evaluated in dogs or cats. Do not infuse concurrently with other medications (for example in a piggy-back) because inactivation may occur. Solutions of 0.5 to 2 mg/mL retain potency up to 14 days when stored. Aminoglycosides: Despite their drop in popularity in favor of fluoroquinolones and extended-spectrum cephalosporins, aminoglycosides such as gentamicin and amikacin are still valuable drugs. They are rapidly bactericidal and therefore are useful for treating acute, life-threatening gram-negative infections. The draw-back for more frequent use of these drugs is the potential for nephrotoxicity (unlikely with short-term therapy) and need for injections. Because aminoglycosides (eg, gentamicin, or amikacin) are concentration-dependent bactericidal drugs, the higher the drug concentration, the greater the bactericidal effect. An optimal bactericidal effect occurs if a high enough dose is administered to produce a peak of 8-10x the MIC. This can be accomplished by administering a single dose once daily. This regimen is at least as effective, and perhaps less nephrotoxic, than lower doses administered more frequently (Freeman et al, 1997). Our current regimens in small animals employ this strategy and have been calculated for most domestic animals. A once daily dose for gentamicin is 5-8 mg/kg for cats, and 10-14 mg/kg for dogs, once daily. An appropriate dose for amikacin is 10-15 mg/kg for cats and 15-30 mg/kg for dogs once daily. Macrolide Antibiotics Erythromycin is an effective drug that has been available for many years. However, it has disadvantages, which include a narrow antibacterial spectrum, adverse gastrointestinal effects (nausea and vomiting), poor oral absorption, short half-life, and need for frequent dosing intervals. In people, it has recently been revealed that erythromycin may be associated with serious cardiac arrhythmias, which presents another potential complication. There are now new derivatives of this macrolide drug that are designed to improve therapy and produce fewer adverse reactions. Azithromycin (Zithromax) is the first drug in the class of azalides. (Lode et al, 1996). Azalides are derived from erythromycin and these drugs share a similar mechanism of action. (Erythromycin is a 14 member ring, and azithromycin has a 15 member ring structure.) The important difference between azithromycin and erythromycin is better oral absorption, it is better tolerated, has a much longer half-life (especially in tissues), and has a slightly broader spectrum of activity. The primary pharmacokinetic difference between azithromycin and erythromycin is the long half-life and high concentration in tissues. The tissue concentrations of azithromycin can be as much as 100 x serum concentrations and the concentrations in leukocytes may be 200x the concentrations in serum. The other new derivative from this class of drugs is clarithromycin (Biaxin). This drug has a similar spectrum of activity as azithromycin, but more active against some organisms. However, clarithromycin has a shorter half-life and must be administered more frequently than azithromycin. Azithromycn is active against gram-positive aerobic bacteria (staphylococci and streptococci) and anaerobes. The activity of azithromycin against staphylococci is not superior to erythromycin, but it has activity against intracellular organisms such as Chlamydia, and Toxoplasma. In people, it is used to treat infections caused by Bartonella, and has been used for the same purpose in dogs and cats. (Clinical efficacy against these pathogens has not been confirmed, however.) It is also active against mycobacteria and Mycoplasma. There are only some limited published clinical reports on the use of azithromycin in dogs, cats, and horses, but the use is increasing. Because of the long half-life and persistence of drug in tissues, the regimen employed in people is to administer a dose once daily for 3 to 5 days, at which time effective drug concentrations are expected in tissues for up to 10 days. There has been one preliminary study in which azithromycin was administered to dogs with pyoderma at a dose of either 10 mg/kg on day 1, followed by 5 mg/kg on days 2 through 5, or alternatively, 5 mg/kg given two days per week for 3 weeks (Holm et al, 2004). In this limited study, the response was equal statistically to cephalexin at 22 mg/kg twice daily, but there were only 20 dogs in the study. Other suggested doses based on unpublished data in dogs suggests 5-10 mg/kg once daily, orally for 1 to 5 days. In cats, doses of 5 mg/kg once daily, or every other day, orally for 1 to 5 days have been used. It is available in a 250 mg capsule, tablets, and an oral suspension. Chloramphenicol Chloramphenicol is banned from use in food animals. It is still used in horses and small animals, but is more difficult to obtain because of few human drug sources. It is no longer used in people because of the risk of bone marrow injury. Chloramphenicol is still used safely in dogs and cats, although cats can be susceptible to bone marrow suppression with use lasting longer than 14 days. Some veterinarians have substituted the cattle drug, florfenicol (Nuflor) for chloramphenicol. It is pharmamcologically similar and somewhat more active microbiologically. Florfenicol is only available in an injectable formulation; therefore oral administration is difficult unless placed in another vehicle. Pharmacokinetic experiments used the 300 mg/ml solution administered IV, oral, IM and SC to dogs and a 100 mg/ml solution administered to cats IV, IM, and oral. Absorption and clearance in dogs were not very suitable for clinical use (poor absorption and rapid clearance). However, in cats, based on the results it is predicted that a dose of 22 mg/kg every 12 hours orally would produce plasma concentrations sufficient to be inhibitory for a range of susceptible bacteria. New Drugs: Oxazolidinones Linezolid (Zyvox): Linezolid is the first in the class of oxazolidinones to be used in medicine. It is currently being used in people to treat vancomycin resistant gram-positive infections caused by enterococci and streptococci. Linezolid inhibits protein synthesis by binding to the bacterial ribosome. It has activity against staphylocci and enterococci. Linezolid is absorbed orally and also is administered IV. Linezolid is available in 600 mg tablets ($53 per tablet!), oral suspension, and injection. Streptogramins Two streptogramin compounds are currently marketed in a combination of 30:70 quinupristin:dalfopristin called Synercid. These compounds were approved specifically for treating infections caused by Enterococcus faecium, and Staphylococcus aureus that are resistant to penicillins and vancomycin. Quinupristin and dalfopristin act in a synergistic manner to inhibit protein synthesis. In addition to Enterococcus faecium, and Staphylococcus, streptogramins also have activity against Mycoplasma and Clostridium. The clinical use of Synercid has not been reported in veterinary medicine. The use of this combination has been limited to human hospitals in which nosocomial organisms can cause serious resistance problems. It must be administered IV through a large central vein. Adverse effects have been common. Current cost of treatment is very expensive, for example, $3,000 per treatment. Daptomycin Daptomycin (Cubicin) is a new cyclic lipopeptide antibiotic. It is only used for staphylococcal, streptococcal, and enterococcal infections that are resistant to other drugs. It can only be given IV (dose in humans is 4 mg/kg IV once daily). The spectrum includes methicillin-susceptible Staphylococcus aureus, methicillin-resistant S. aureus, vancomycin-resistant S. aureus, penicillin-resistant Streptococcus pneumoniae, and ampicillin- and vancomycin-resistant enterococci. There are no reports of its use in veterinary medicine. Cost for four day treatment in people is $480.00. Ketolides Telithromycin (Ketek) is the first of a new class of drugs called ketolides. They have some similarities to the older macrolide compounds (eg. erythromycin), with slightly improved spectrum and tolerability. Ketolides are semisynthetic derivatives of macrolides and block protein synthesis at a similar target as for macrolides. Telithromycin is available as an oral tablets that is used in people primarily for respiratory tract infections and sinus infections, for example those caused by Streptococcus pneumoniae or Haemophilus influenzae. It is also active against Staphylococcus aureus and intracellular pathogens such as Chlamydia. It has a long half-life of 10-13 hours in people with high tissue and leukocyte concentrations. It is given at a dose of 10 mg/kg once daily orally. The cost for a 10-day course in people is $114. There are no reported veterinary uses. APPROACHES FOR MANAGING DIFFICULT INFECTIONS In addition to new drugs used in veterinary medicine, new approaches have been used as well. These have not been rigorously evaluated for efficacy, long term safety, or effect on emergence of resistance. However, veterinarians are often presented with few alternatives. Some of the approaches to recurrent infections have been to either administer long-term, low dose antibiotics on a chronic regimen, or, to administer regular therapeutic doses intermittently (eg, once per week) on a long-term basis. This approach has been used to treat recurrent urinary tract and skin infections, especially in immunosuppressed patients. This approach has not been evaluated rigorously. In one preliminary report, the investigators used weekend therapy (two days per week at 15 mg/kg q12h orally) of cephalexin to prevent relapses of recurrent idiopathic superficial or deep pyoderma in dogs. This approach was effective for preventing relapses and did not produce resistance. An additional approach used for recurrent urinary tract infections has been to administer urinary antiseptics such as methenamine (methenamine hippurate, or methenamine mandelate) on a long-term basis. This drug is metabolized to formaldehyde in acidic urine and may inhibit growth of organisms in the urine. It has inhibited growth of bacteria in canine urine when administered at a dose of approximately 50 mg/kg. Long-term effects have not been evaluated. Urine should be maintained at an acid pH for methenamine to produce an optimum effect. Another human drug used in women, fosfomycin, also has been used intermittently to decrease recurrent infections caused by E. coli. In one report management strategies of persistent urinary tract infections and re-infections were evaluated in 100 dogs (Seguin et al, 2003). In that study, almost an equal number of dogs with persistent urinary tract infections had failure to clear the original infection as did those with new infections. Strategies included conventional antibiotic therapy, log-term low dose antibiotics, or administration of methenamine or urine acidifier. When correction of the underlying disorder was not possible, more dogs responded to low dose, long term antibiotic therapy than the other regimens. Methenamine was somewhat effective, but was only evaluated in a small number of animals. Treating infections that have not responded to other drugs is challenging and sometimes frustrating. The first step it to try to assess why the patient isn't responding. Is it: 1) immunosuppression caused by drugs or some disease, 2) some other concurrent disease (eg, cancer or fungal infection) that isn't responding to antibiotics, 3) is this a re-infection after the original infection has been cleared, 4) is this a relapse of the previous infection because the original one wasn't cleared, or 5) is this a bacteria that is resistant to multiple drugs, or 6) has the problem bacteria been correctly identified? Treating resistant gram-negative bacteria: The most common bacteria to develop resistance in veterinary small animal medicine are the gram-negative bacilli, especially the enteric isolates. If the organism is Pseudomonas aeruginosa, Enterobacter, Klebsiella, Escherichia coli, or Proteus, resistance to many common antibiotics is possible and a susceptibility test is advised. After a susceptibility report is available, one may find that the only drugs to which a gram-negative bacilli is sensitive are extended-spectrum cephalosporins and carbapenems. The injectable cephalospsorins most often used are cefotaxime and ceftazidime, although individual veterinary hospitals have utilized others in this group. These drugs are expensive, injectable (except for the oral exceptions cited earlier), and must be administered frequently. The carbapenems have been valuable for treatment of resistant gram-negative bacteria. These drugs include imipenem, meropenem, and ertapenem. All three have activity against the enteric gram-negative bacilli. More specific details are discussed in the section on treating Pseudomonas infections. Pseudomonas aeruginosa Of the 8-lactam antibiotics, a few are designated as anti-Pseudomonas antibiotics. Those with activity against this organism include the ureidopenicillins (mezlocillin, azlocillin, piperacillin) and the carboxylic derivatives of penicillin (carbencillin, ticarcillin). Ticarcillin is more active than carbenicillin against gram-negative bacteria, but whether this relates to clinical differences is not clear. These derivatives are available as sodium salts for injection; there are no orally-effective formulations in this class, except indanyl carbenicillin (Geocillin, Geopen) which is poorly absorbed and not useful for systemic infections. These drugs are more expensive than the more-commonly used penicillins, and must be administered frequently (eg, at least 4 times daily) to be effective. Ticarcillin is available in combination with the beta-lactamase inhibitor clavulanic acid (Timentin). Ticarcillin also has been used in compounded topical formulations applied to the ear canal for treatment of otitis externa caused by Pseudomonas. Because these drugs degrade quickly after reconstitution, observe the storage recommendations on the package insert to preserve the drug's potency. Of the cephalosporins, only the 3rd-generation cephalosporins, ceftazidime (Fortaz, Tazidime), cefoperazone (Cefobid), or cefepime (Maxipime), a 4th-generation cephalosporin, have predictable activity against Pseudomonas aeruginosa. Ceftazidime has greater activity than cefoperazone and is the one used most often in veterinary medicine. These drugs must all be injected, and are usually given IV, although SC, and IM routes have been used. As with the penicillins, frequent administration is necessary. The 8-lactam antibiotics with greatest activity against Pseudomonas aeruginosa are the carbapenems. The drugs in this class are imipenem-cilastatin, and meropenem. Ertapenem is a new addition to the class of carbapenems but it does not have anti-Pseudomonas activity. Aminoglycosides include gentamicin, amikacin, and tobramycin. They are active against most Pseudomonas aeruginosa strains. Amikacin and tobramycin are more active than gentamicin, and resistance is less likely to these drugs (Petersen et al, 2002). The aminoglycosides are limited to topical and injectable administration. They have been administered once-daily for systemic infections IV, IM, or SC. There are two important disadvantages to systemic use of aminoglycosides: (1) Treatment usually must extend for at least two weeks or longer. Risk of nephrotoxicosis is greater with longer duration of treatment. (2) Activity of aminoglycosides is diminished in the presence of pus and cellular debris (Konig et al 1998). This may decrease their usefulness for the treatment of wound and ear infections caused by Pseudomonas aeruginosa. Fluoroquinolones are active against Pseudomonas aeruginosa, but usually MIC values are higher than against other gram-negative organisms. Subsequently, when administering a fluoroquinolone to treat Pseudomonas aeruginosa the high-end of the dose range is suggested. Of the currently available fluoroquinolones, (human or veterinary drugs) ciprofloxacin is the most active against Pseudomonas aeruginosa. Treatment of Resistant Gram-Positive Bacteria Resistant Staphylococcus: Staphylococcal resistance can be caused by altered penicillin-binding proteins (for example the resistance carried by the gene mecA). These are known as methicillin-resistant staphylococci or MRSA (Gortel et al, 1999; Deresinski 2005). Oxacillin is now used more commonly than methicillin as the marker for this type of resistance, and resistance to oxacillin is equivalent to methicillin-resistance. This pattern of resistance is uncommon among veterinary isolates (Petersen et al, 2002; Normand et al, 2000; Prescott et al, 2002; Lloyd et al, 1996). Oxacillin- or methicillin-resistance is found more often among Staphylococcus aureus than Staphylococcus intermedius, (Tomlin et al. 1999; Gortel et al, 1999). If staphylococci are resistant to oxacillin or methicillin, they should be considered resistant to all other ?-lactams, including cephalosporins and amoxicillin-clavulanate (eg, Clavamox), regardless of the susceptibility test result. Adding a ?-lactamase inhibitor will not overcome methicillin resistance. These bacteria should be tested for susceptibility to clindamycin or a fluoroquinolone. In some instances the only drug that is active for treatment will be a glycopeptide such as vancomycin (Vancocin). Resistant Enterococcus Enterococci are gram-positive cocci that have emerged as important causes of infections, especially those that are nosocomial. The most common species identified are Enterococcus faecalis and E. faecium. Enterococcus faecalis is more common, but E. faecium is usually the more resistant. Wild-strain enterococci may still be sensitive to penicillin G and ampicillin, or amoxicillin. However, the enterococci have an inherent resistance to cephalosporins and fluoroquinolones. These strains also are usually resistant to trimethoprim-sulfonamide combinations, clindamycin, and erythromycin. Susceptibility test results for cephalosporins, beta-lactamase resistant penicillins (eg, oxacillin), trimethoprim-sulfonamide combinations, and clindamycin can give misleading results. Even if isolates are shown to be susceptible to a fluoroquinolone, this class of drugs may not be a good alternative for treatment. In human medicine frequent use of fluoroquinolones and cephalosporins (both of which have poor activity against enterococci), has been attributed to emergence of a higher rate of enterococcal infections. Evidence to document this trend is limited in veterinary medicine, but one study from a veterinary teaching hospital indicated increased rate of enterococcal urinary tract infections (Prescott, et al, 2002). Treatment of Enterococcus is frustrating because there are so few drug choices. If the Enterococcus isolated is sensitive to penicillins, administer amoxicillin or ampicillin at the high-end of the dose range. When possible, combine an aminoglycoside with a beta-lactam antibiotic for treating serious infections. Each drug alone is poorly bactericidal against enterococci when used alone. Occasionally, one of the carbapenems (imipenem-cilastatin) or an extended-spectrum penicillin (eg, piperacillin) can be considered for treatment of E. faecalis (but not E. faecium). When enterococci are present in wound infections, peritoneal infections, and body cavity infections (eg, peritonitis), the organism may exist with other bacteria such as gram-negative bacilli, or anaerobic bacteria. In these cases, there is evidence that treatment should be aimed at the anaerobe, and/or gram-negative bacilli and not directed at the enterococcus. Treatment cures are possible if the other organisms are eliminated without specific therapy for enterococcus (Bartlett et al 1978). Often, the only effective drug will be a glycopeptide. Of the glycopeptides, vancomycin is the only one used in veterinary medicine. Vancomycin (Vancocin) has been given as an IV infusion administered over 30 to 60 minutes. It is not absorbed orally and is too painful when injected IM. The dose to maintain concentrations within the therapeutic range, and avoid toxicity is 15 mg/kg, q6h, IV. For successful therapy of serious infections, an aminoglycoside such as gentamicin or amikacin should be administered with vancomycin. REFERENCES CITED Abadía AR, Aramayona JJ, Muñoz MJ, et al. Disposition of ciprofloxacin following intravenous administration in dogs. J Veterinary Pharmacology and Therapeutics 17: 384-388, 1994. Abadía AR, Aramayona JJ, Muñoz MJ, et al. Ciprofloxacin pharmacokinetics in dogs following oral administration. Journal Veterinary Medicine A. 42: 505-511, 1995. Albarellos GA, Kreil VE, and Landoni MF. Pharmacokinetics of ciprofloxacin after single intravenous and repeat oral administration to cats. J Veterinary Pharmacology Therapeutics 27: 155-162, 2004. Barker CW, Zhang W, Sanchez S, et al. Pharmacokinetics of imipenem in dogs. American Journal of Veterinary Research 64: 694-699, 2003. Bialer M, Wu WH, Look AM, Silber BM, and Yacobi A. Pharmacokinetics of cefixime after oral and intravenous doses in dogs: Bioavailability assessment for a drug showing nonlinear serum protein binding. Research Communications in Chemical Pathology and Pharmacology 56: 21-32, 1987. Bidgood T, Papich MG. Plasma pharmacokinetics and tissue fluid concentrations of meropenem after intravenous and subcutaneous administration in dogs. American Journal of Veterinary Research . 2002;63(12):1622-1628. Bidgood T, Papich MG. Pharmacokinetics of enrofloxacin and marbofloxacin and tissue concentrations in dogs. (2004. pending) Carlotti DN, Jasmin P, Gardey L, & Sanquer A. Evalation of cephalexin intermittent therapy (weekend therapy) in the control of recurrent idiopathic pyoderma in dogs: a randomized, double-blinded, placebo-controlled study. Veterinary Dermatology 15 (Supplement 1): 8, 2004 [abstract 13]. Cester CC, Schneider M, Toutain P-L: Comparative kinetics of two orally administered fluoroquinolones in dog: Enrofloxacin versus marbofloxacin. Revue Méd Vét 147: 703-716, 1996. Cole, L.K., Kwochka, K.W., Kowalaski, J.J., and Hillier, A. Microbial flora and antimicrobial susceptibility patterns of isolated pathogens from the horizontal ear canal and middle ear in dogs with otitis media. Journal of the American Veterinary Medical Association 1998; 212: 534-538. deJong A, Stephan B, Friederichs S. Antibacterial activity of pradofloxacin against canine and feline pathogens isolated from clinical cases. [abstract] AAVM, Ottawa, Canada, June 2004. Deresinski S. Methicillin-resistant Staphylococcus aureus: an evoluationary, epidemiologic, and therapeutic odyssey. Clin Infect Dis 40: 562-573, 2005. Edwards JR, Betts MJ: Carbapenems: the pinnacle of the Beta-lactam antibiotics or room for improvement? J Antimicrob Chemother 2000; 45: 1-4. Everett, M.J., Jin, Y.F., Ricci, V., Piddock, L.J.V. Contributions of individual mechanisms to fluoroquinolone resistance in 36 Escherichia coli strains isolated from humans and animals. Antimicrobial Agents and Chemotherapy 1996; 40: 2380-2386. Freeman CD, Nicolau DP, Belliveau PP, Nightingale CH: Once-daily dosing of aminoglycosides: review and recommendations for clinical practice. J Antimicrob Chemother 39: 677, 1997. Gardner, S.Y., Papich, M.G. Comparison of cefepime pharmacokinetics in neonatal foals and adult dogs. Journal of Veterinary Pharmacology and Therapeutics 24: 187-192, 2001. Gortel K, Campbell KL, Kakoma I, Whittem T, Schaeffer DJ, Weisiger RM. Methicillin resistance among staphylococci isolated from dogs. Am J Vet Res 60: 1526-1530, 1999. Holm K, Rosenbaum M, and Byrne K. A double-blinded pilot study on the use of azithromycin vs cephalexin for canine pyoderma. Veterinary Dermatology 15: 199, 2004 [abstract]. Hooper DC. Mechanisms of action and resistance of older and newer fluoroquinolones. Clin Infect Dis 31(Suppl 2): S24-S28, 2000. Ihrke PJ, Papich MG, DeManuelle TC: The use of fluoroquinolones in veterinary dermatology. Vet Dermatol 1999; 10: 193-204. Kay-Mugford PA, Ramsey DT, Dubielzig RR, et al. Ocular effects of orally administered orbifloxacin in cats. American College of Veterinary Ophthalmologists 32nd Annual Meeting. (abstract) October 9-13, 2001. Konig C, Simmen HP, Blaser J. Bacterial concentrations in pus and infected peritoneal fluid - implication of bactericidal activity of antibiotics. J Antimicrob Chemother 42: 227-232, 1998. Lavy E, Ziv G, Aroch I, Glickman A: Clinical pharmacologic aspects of cefixime in dogs. Am J Vet Res 56: 633-638, 1995. Livermore DM. Of Pseudomonas, porins, pumps, and carbapenems. J Antimicrob Chemother 47: 247-250, 2001. Lode H, Borner K, Koeppe P, and Schaberg T: Azithromycin: review of key chemical, pharmacokinetic, and microbiological features. J Antimicrob Chemother 37 (Suppl C): 1-8, 1996. Lloyd DH, Lamport AI, Feeney C. Sensitivity to antibiotics amongst cutaneous and mucosal isolates of canine pathogenic staphylococci in the UK, 1980-96. Vet Dermatol 7: 171-175, 1996. Moore KW, Trepanier LA, Lautzenhiser SJ, et al. Pharmacokinetics of ceftazidime in dogs following subcutaneous administration and continuous infusion and the association with in vitro susceptibility of Pseudomonas aeruginosa. Am J Vet Res 2000; 61: 1204-1208. NCCLS. NCCLS. Performance Standards for Antimicrobial Disk and Dilution Susceptibility Tests for Bacteria Isolated from Animals; Approved Standard - Second Edition. NCCLS document M31-A2. NCCLS, 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087, 2002. Nakamura S, Kurobe N, Ohue T, Hashimoto M, Shimizu M. Pharmacokinetics of a novel quinolone, AT-4140, in animals. Antimicrob Agents Chemother 34: 89-93, 1990. Normand EH, Gibson NR, Taylor DJ, et al. Trends of antimicrobial resistance in bacterial isolates from a small animal referral hospital. Vet Record 146: 151-155, 2000. Papich MG, & Riviere JE. Fluoroquinolone antimicrobial drugs, Chapter 45. In H.R. Adams (ed) Veterinary Pharmacology and Therapeutics, 8th Edition. Ames Iowa, Iowa State University Press. 2001; Page 898-917. Periti P, Mazzei T. New criteria for selecting the proper antimicrobial chemotherapy for severe sepsis and septic shock. Int J Antimicrob Agents 1999; 12: 97-105. Pestova E, Millichap JJ, Noskin GA, Peterson LR. Intracellular targets of moxifloxacin: a comparison with other fluoroquinolones. J Antimicrob Chemother 45: 583-590, 2000. Petersen AD, Walker RD, Bowman MM, Schott HC, Rosser EJ. Frequency of isolation and antimicrobial susceptibility patterns of Staphylococcus intermedius and Pseudomonas aeruginosa isolates from canine skin and ear samples over a 6 year period (1992-1997). J Am Anim Hosp Assoc 38: 407-413, 2002. Petersen SW, Rosin E: In vitro antibacterial activity of cefoxitin and cefotetan and pharmacokinetics in dogs. Am J Vet Res 54: 1496-1499, 1993. Prescott JF, Hanna WJB, Reid-Smith R, and Drost K. Antimicrobial drug use and resistance in dogs. Canadian Veterinary Journal 43: 107-116, 2002. Schluter, Ciprofloxacin: Review of potential toxicologic effects. American Journal of Medicine 82 (suppl 4A): 91-93, 1987. Seguin MA, Vaden SL, Altier C, Stone E, and Levine JF. Persistent urinary tract infections and reinfections in 100 dogs (1989-1999). Journal of Veterinary Internal Medicine 17: 622-631, 2003. Smith, K.E., Besser, J.M., Hedberg, C.W., et al. Quinolone-resistant Campylobacter jejuni infections in Minnesota, 1992-1998. New England Journal of Medicine 1999; 340: 1525-1532. Stephan B, Roy O, Skowronski V, Edingloh M, and Greife H. Clinical efficacy of pradofloxacin in the treatment of canine urinary tract infections. [abstract] AAVM, Ottawa, Canada, June 2004. Threlfall EJ, Frost JA, Ward LR, and Rowe R: Epidemic in cattle of S typhimurium DT 104 with chromosomally-integrated multiple drug resistance. Vet Rec 134: 577, 1995. Tomlin J, Pead MJ, Lloyd DH, Howell S, et al. Methicillin-resistant Staphylococcus aureus infections in 11 dogs. Veterinary Record 144: 60-64, 1999. Walker AL, Jang S, Hirsh DC. Bacteria associated with pyothorax of dogs and cats: 98 cases (1989-1998). J Am Vet Med Assoc 2000; 216: 359-363. Walker RD, Stein GE, Hauptman JG, et al. Serum and tissue cage fluid concentrations of ciprofloxacin after oral administration of the drug to healthy dogs. American Journal of Veterinary Research 51: 896-900, 1990. Watts JL, Salmon SA, Sanchez MS, Yancey RJ. In vitro activity of Premafloxacin, a new extended-spectrum fluoroquinolone, against pathogens of veterinary importance. Antimicrob Agents Chemother 41: 1190-1192, 1997. Sorting Out the Choices Among Nonsteroidal Anti-Inflammatory Drugs (Nsaid) PHARMACOLOGY OF NSAID The pharmacologic action of the nonsteroidal anti-inflammatory drugs (NSAID) have been reviewed in recent articles (Vane and Botting, 1995; Papich, 2000). There have been several new developments in the NSAIDs in the last few years, including new drugs for use in dogs. The change in our understanding of NSAIDs happened when it was learned that there are two isoenzymes of cyclo-oxygenase (prostaglandin endoperoxide synthase) that are responsible for synthesis of prostaglandins. Prostaglandin synthase-1 (COX-1) is usually a constitutive enzyme expressed in tissues (Meade et al 1994). Prostaglandins, prostacyclin, and thromboxane synthesized by this enzyme are responsible for normal physiological functions. Prostaglandin synthase-2 (COX-2), on the other hand, is inducible and synthesized by macrophages and inflammatory cells after stimulation by cytokines and other mediators of inflammation. In some tissues, COX-2 may be constitutive. The perception that COX-2 is a bad enzyme, and COX-1 is a good enzyme is probably overly simplistic as we now understand that there is some overlap in the functions of these isoforms. Nevertheless, the target of recently-developed NSAID has been COX-2, with the goal of producing analgesia and suppressing inflammation without inhibiting physiologically important prostanoids. In some instances, the mechanism of action may not be entirely known. For example, carprofen appears to be a COX-1 sparing drug, (Ricketts et al 1998) but there is not agreement among investigators on whether or not it also inhibits COX-2 in vivo. Although there is evidence for inhibitory effects on the enzyme cyclooxygenase in some models, carprofen did not show an in vivo anti-prostaglandin effect in dogs (McKellar et al 1994), which may explain the low rate of gastrointestinal adverse effects at approved doses. In one recent study, the investigators were unable to show that carprofen inhibited either COX-1 or COX-2 (Bryant et al, 2003). Selectivity of COX-2 vs COX-1 is often expressed as the COX-1/COX-2 inhibitory ratio. The higher the value above 1.0, the more specific the drug is for COX-2 compared to COX-1. Drugs with high ratios also are referred to as "COX-1 sparing". When one examines the drugs registered for veterinary medicine, there is disagreement in the literature with respect to the selectivity for the COX-2 enzyme. For example, deracoxib is considered a highly selective COX-2 inhibitor based on an assay performed in purified enzymes (Gierse et al 2002). In this study, the COX-1/COX-2 ratio was 1275; much higher than other drugs tested. But when tested in canine whole blood and compared to other NSAIDs, deracoxib had a ratio of only 12. In this study, carprofen had a ratio of 6-7, and firoxoxib (the newest NSAID for dogs) had a ratio of 384-427. (McCann et al, 2004). Some of the confusion regarding understanding the action of the veterinary NSAID is that in vitro studies to examine their relative effects on COX-1 vs COX-2 have varied in their techniques and the cell system used. For example, in a study using canine enzyme systems, carprofen had a COX-1/COX-2 ratio of 129 (Ricketts et al 1998). In another study, using cell lines of another species (sheep and rodent) the ratio was 1.0 (Vane and Botting, 1995), and in a study using canine macrophages, the ratio was 1.75 (Kay-Mugford et al 2000). Yet another study on carprofen showed a ratio of 5.3 and that it was 1,000 times less potent in whole blood than in cell culture (Wilson et al, 2004). This emphasizes the effect of protein binding on in vitro assays. There also have been conflicting results when other drugs have been examined. The ratios for etodolac, another NSAID approved for dogs, has a COX-1/COX-2 ratio of 8.1 in humans, but 0.52-0.53 in dogs. Another study with etodolac showed that the selectivity for COX-2 was 10 times greater in people than dogs (Gierse et al 2002; Glaser 1995). Dr. Vane, a preeminent expert on COX inhibition, concludes that, Athe inhibitory activity of a drug for COX-1 to its inhibitory activity for COX-2 can vary according to whether tests are done on pure enzymes, cell homogenates, intact cells, or with the types of cells used.@(Vane and Botting, 1995). According to Dr. Lees, one of the leading investigators of NSAIDs in veterinary medicine, there are several unexplored questions to be answered for veterinary drugs (Lees, 2003). Whether or not in vitro measurements of COX-1 vs COX-2 inhibition predict in vivo response and safety has been debated. Sometimes the in vitro results are supported by in vivo assays, such as in one study comparing meloxicam to aspirin in dogs. In this study, meloxicam, which is a somewhat selective COX-2 inhibitor using a whole-blood assay, also had a sparing effect on gastrointestinal prostaglandins (COX-1 mediated) compared to aspirin (Jones et al 2002). Meloxicam also was a potent inhibitor of lipopolysaccharide (LPS)-induced prostaglandin synthesis (COX-2 mediated). These findings are consistent with COX-2 inhibition and COX-1 sparing effects of meloxicam, but lack of such specificity for aspirin. However, in a follow-up study by the same laboratory they compared carprofen, deracoxib, and etodolac (Sessions et al, 2005). All three drugs failed to inhibit prostaglandins in the stomach mucosa, and thromboxane in platelets, consistent with a COX-1 sparing effect. All three drugs produced the same degree of COX-1 sparing, despite a wide range in COX-1/COX-2 inhibitory ratios among these drugs. In the same study, etodolac did not suppress the COX-2-mediated product PGE2 in a blood assay compared to carprofen and deracoxib on days 3 and 10 of treatment. Carprofen and deracoxib did not differ in their in vivo effects on either COX-1 or COX-2 inhibition, despite large differences for in vitro COX-1/COX-2 ratios. These results cited above also were inconsistent with another study in dogs in which carprofen and etodolac were equally effective for reducing pain scores in experimentally-treated dogs (Borer et al, 2003), and both were more effective than meloxicam. Although, for most NSAIDS we assume that prostaglandin inhibition is the most important mechanism of action, there may be other mechanisms B some not fully understood B that also may explain the actions of these drugs. For example, some NSAIDs, including salicylates, have been suggested to also inhibit nuclear factor kappa-B (NF-?B). NF-?B is an important promoter for inflammatory mediators. Veterinary drugs, such as carprofen and others also may act through inhibition of the activation of NF-?B (Bryant et al, 2003). IS THERE AN ADVANTAGE FOR COX-2 INHIBITORS? Recently, drugs for human use have become available that are highly COX-2 specific, celecoxib (Celebrex), valdecoxib (Bextra, now discontinued), and rofecoxib (Vioxx, now discontinued) (FitzGerald & Patrono, 2001). These are often referred to as the COXIBs and they were among the top-selling prescription drugs of any category in human medicine. Deracoxib was the first veterinary drug in this group; the next one approved was firocoxib (Previcox) by Merial Ltd. Other COX-2 specific drugs may follow in veterinary medicine. Firocoxib (Previcox) is more specific for COX-2 than deracoxib with a COX-1/COX-2 ratio of 384-427 compared to deracoxib with a ratio of 12 according to one study (McCann et al, 2004). This drug will be marketed in the summer of 2005. In efficacy studies, firocoxib was compared to etodolac and carprofen and was shown in some measurements to have better improvement in lameness scores. In a safety assessment, firocoxib, caprofen, and etodolac all were similar with respect to incidence of vomiting, anorexia in dogs. But with respect to diarrhea there was a lower incidence with firocoxib compared to carprofen and etodolac and less melena compared to etodolac (data available from drug sponsor). Evaluations of these newer drugs in people show that they are not necessarily more effective than older drugs, but they may be safer for the gastrointestinal tract (Peterson and Cryer 1999) during short-term evaluations. In veterinary studies, there is no convincing evidence that drugs with higher COX-1/COX-2 ratios produce fewer gastrointestinal or renal adverse effects than drugs with low ratios. In human medicine, the studies demonstrating safety in people have been criticized (Malhotra et al, 2004). Some skeptics have proposed that selective COX-2 inhibitors may not be appropriate for all patients because COX-2 enzyme products may be involved in actions other than inflammation. For example, COX-2 products may be biologically important for angiogenesis, renal function, regulation of bone resorption, reproductive function, and healing of gastroduodenal ulcers (Wolfe et al 1999). COX-2 selective drugs also may cause a higher risk of cardiovascular problems in people because it preserves COX-1 which may promote platelet aggregation and vasoconstriction (Mukherjee et al, 2001). This is the reason that the popular drug rofecoxib (Vioxx) has been voluntarily taken off the market, and soon followed by valdecoxib (Bextra). There has been serious concerns expressed about the events that lead up this withdrawal. Some experts believe that the high COX-2 selectivity of this drug led to this increased risk (Topol 2004). DUAL INHIBITORS There have been older drugs promoted to be "dual inhibitors" of arachidonic acid metabolites, but none were commercially successful. Dual inhibitor drugs effectively inhibit both cyclo-oxygenase (COX) and lipoxygenase (LOX). Therefore, they inhibit synthesis of both inflammatory prostaglandins (PG) and leukotrienes (LT). Interest in a dual inhibitor has focused on the potential benefits in inhibiting LOX, which may include higher GI safety, and greater analgesic efficacy (Trang et al, 2004). Lipoxygenase metabolites are involved in hyperalgesia, and inflammatory responses (Bertolini et al, 2001). Older drugs thought to have dual inhibitor capability were benoxaprofen and ketoprofen. Benoxaprofen was taken off the market, and the evidence for ketoprofen acting as a dual inhibitor is weak. A new drug being evaluated in people and dogs is licofelone, which is a true dual inhibitor, but not yet on the market. Licofelone may have greater gastrointestinal safety than other NSAIDs (Moreau et al, 2005). Corticosteroids have been shown to be dual inhibitors in some studies because they inhibit Phosphodiesterase A2 the enzyme that forms arachidonic acid from cell membranes. However, corticosteroid inhibition of both LT and PG may not be clinically relevant. The only drug approved in Europe and the U.S. that acts as a dual inhibitor in animals is tepoxalin (Zubrin). The metabolite is active, but only acts as a COX inhibitor. The COX inhibitor functions are more specific for COX-1 than COX-2, although this was not a canine-specific assay (data from Schering-Plough). Using dogs in an in vivo study Agnello et al, (2005) showed that tepoxalin administration to dogs inhibited both COX isoforms as well as the LOX activity. Despite being a non-selective COX inhibitor, tepoxalin has a good gastrointestinal safety profile that matches other more selective COX-2 inhibitors. Tepoxalin has been shown to be effective in dogs with osteoarthritis and showed GI safety at several times the label dose. The only question remaining about tepoxalin is the duration of the LOX inhibitory effect. As shown in the accompanying table, the half-life for the LOX inhibitor parent drug is much shorter than the metabolite, which has little LOX inhibition. The other question remaining to be answered for tepoxalin is the contribution of anti-LOX action on the overall therapeutic effect. Studies in osteoarthritis in dogs (the registered indication for tepoxalin) have not revealed whether or not it is the COX or the LOX inhibition (or possibly some other mechanism) that is responsible for a favorable clinical effect. Whether or not the dual inhibition action of tepoxalin will be effective for other inflammatory diseases (for example, respiratory disease, dermatitis) has not been reported. PHARMACOKINETIC FEATURES For most of the NSAID there is adequate pharmacokinetic data for dogs, and some for cats. Most of the traditional drugs in this group are weak acids that are highly protein bound and most of them have a small volume of distribution (some new drugs are an exception to this standard). These drugs are excreted at varying rates, depending on the metabolic pathway and extent of enterohepatic circulation. There are tremendous species differences in drug elimination among the NSAIDs. For some drugs the enterohepatic cycling may increase the risk of toxicosis because the local effects of the drug may be focused on the intestinal mucosa through repeated cycling in the biliary system. Although the drug distribution, half-life, and clearance, have been characterized for most NSAIDs used in animals, this information has not always been of use for predicting safe and effective dosage regimens. For example, NSAIDs such as ibuprofen and indomethacin easily cause toxicity in dogs even though they have short half-lives. On the other hand, naproxen and piroxicam have long half-lives of 74 hours and 40 hours, respectively, but have been used safely when dosed carefully. It should be noted that among the small animal NSAIDs, half-lives do not correlate with the frequency of administration. Most currently-used NSAIDs are given once a day, but half-lives vary widely. See table below:
An important feature of the NSAID pharmacokinetics is that anti-inflammatory and analgesic effects persist longer than the plasma half-lives would predict. In dogs, several NSAID have half-lives of 24 hours or less, (aspirin carprofen, 8 hours ; phenylbutazone 6 hours; flunixin: 3.7 hours; meloxicam: 10-24 hours; etodolac, 8-12 hours), but have been administered once every 24 hours with effective results (Mathews 1996). One explanation for the long duration of effect is the high protein binding. The tissue protein binding (for example the protein in an inflamed site) may serve as a reservoir for the drug after it has been eliminated from the plasma. Thus, the NSAID may persist in inflamed sites longer than the plasma. ADVERSE EFFECTS OF NSAID Gastrointestinal Toxicity Among the adverse reactions caused by NSAID, gastrointestinal problems are the most frequent reason to discontinue NSAID therapy or consider alternative treatment. The FDA=s Freedom of Information (FOI) Summary for all the approved veterinary drugs provides the documentation of tests to determine a drug's safety. These summaries also list GI effects as important adverse events. In animals, gastrointestinal (GI) effects range from mild gastritis and vomiting, to severe GI ulceration, bleeding and even deaths. These effects are identified on the drug sponsor's data from FOI summaries published by the FDA and are often observed in target animal safety studies when high doses were administered. Gastrointestinal adverse events also have been documented for the past three decades in the veterinary literature. Gastrointestinal toxicity is caused by two mechanisms: direct irritation of the drug on the GI mucosa and the result of prostaglandin inhibition (Wolfe et al 1999). Direct irritation occurs because the acidic NSAID become more lipophilic in the acid milieu of the stomach and diffuse into the gastric mucosa where they cause injury. Prostaglandins have a cytoprotective effect on the GI mucosa and inhibition of these compounds results in decreased cytoprotection, diminished blood flow, decreased synthesis of protective mucus, and inhibition of mucosal cell turnover and repair. In the GI tract of healthy dogs, COX-1 is the primary COX enzyme that produces prostaglandins (primarily PGE2) (Wilson et al, 2004). An examination of published reports of GI toxicity from administration of NSAID in animals indicates that the most serious problems are caused from doses that are higher than recommended, but toxicity also has been observed from relatively mild doses in susceptible individuals. Some factors may increase the risk of GI toxicosis, including concurrent corticosteroids, and other gastrointestinal diseases. The most recently-approved NSAID in the United States for dogs are carprofen, etodolac, meloxicam, deracoxib, firocoxib, and tepoxalin. A few other drugs are approved in Canada and Europe (tolfenamic acid, and ketoprofen for example). For the newer veterinary-registered NSAIDs, the gastrointestinal safety profile, in comparison with older drugs, has contributed to their popularity in veterinary medicine. However, there is no evidence in the published literature using controlled clinical trials to show that one is safer or more effective than another. In a study in which carprofen, meloxicam, and ketoprofen were compared in dogs after endoscopic evaluation after 7 and 28 days of administration, there was no statistical difference among the drugs with respect to development of gastroduodenal lesions (Forsyth et al 1998). In a study that compared the gastrointestinal effects of recommended doses of carprofen, etodolac, and aspirin on the canine stomach and duodenum for 28 days, etodolac and carprofen produced significantly fewer lesions than aspirin (Reimer et al 1999). Lesion scores in the carprofen and etodolac-treated groups were no different than administration of placebo. The putative explanation for this degree of safety of carprofen, etodolac, deracoxib, firocoxib, and meloxicam is that these drugs have preferential inhibitory action for COX-2 than COX-1 (high COX-1: COX-2 ratio). Perhaps a more accurate description of these drugs is that they have a COX-1 sparing effect (Peterson and Cryer 1999). However, as discussed earlier, COX-1:COX-2 ratios many not necessarily correlate with GI safety, and the calculated ratios may vary from study to study, and from species to species. Some drugs may lose their COX-2 selectivity at high doses (Wolfe et al 1999). The dose-dependence was shown for etodolac. At the label dose it was safe, but at higher doses, (2.7 x dose) produced gastrointestinal lesions, and, at the high dose, (5.3 x dose), caused death. At high doses, meloxicam also has demonstrated some GI toxicity. The sponsors of this drug in Europe recommended reducing the original approved dose from 0.2 mg/kg to 0.1 mg/kg because of some initial gastrointestinal problems (Forsyth et al 1998). Renal toxicity: In the kidney, prostaglandins play an important role to modulate the tone of blood vessels and regulating salt and water balance. Renal injury caused by NSAID has been described in people, and horses, but has not been well documented in small animals. Reported cases of toxicity occurred when high doses were used or when there were other complicating factors. Renal injury occurs as a result of inhibition of renal prostaglandin synthesis. (Brown, 1989). In animals that have decreased renal perfusion caused by dehydration, anesthesia, shock, or pre-existing renal disease, this leads to renal ischemia (Mathews, 1996). Additional information is needed with regard to the safety of currently available COX-2 inhibitors on the kidney. Some of the prostaglandins that play an important role in salt and water regulation and hemodynamics in the kidney are synthesized by COX-2 enzymes (Rossat et al 1999). Constitutive COX-2 is found in various sections of the kidney and administration of drugs that are selective for COX-2, may adversely affect the kidney during in some situations. Administration of a specific COX-2 inhibitor to salt-depleted people decreased renal blood flow, glomerular filtration rate, and electrolyte excretion (Rossat et al 1999). Of the currently-available NSAIDs, carprofen's effect on renal function has been the most extensively studied. Because carprofen is registered for use in perioperative situations in an injectable formulation (Rimadyl injectable 50 mg/mL) investigations were performed to determine if there was any evidence of renal toxicity, particularly during conditions of anesthesia. In one study, carprofen, ketorolac and ketoprofen were examined in healthy dogs undergoing surgery, but without IV fluid administration. There were minor increases in renal tubular epithelial cells on urine sediment, but overall carprofen had no adverse effects on renal function (Lobetti et al, 2000). Some ketorolac and ketoprofen-treated dogs had transient azotemia. In other studies, administration of carprofen to anesthetized healthy dogs had no adverse effects on renal function (Ko et al 2000; Böstrom et al 2002). Renal effects from deracoxib were reported by the manufacturer. At high doses, there is a dose-dependent effect on renal tubules. It is well-tolerated in most dogs up to 10 mg/kg for 6 months, but there is a potential for a dose-dependent renal tubular degeneration/regeneration at doses of 6 mg/kg or higher. (Clinically approved dose for long-term treatment is 1-2 mg/kg per day.) Tepoxalin was evaluated in anesthetized, healthy, normotensive, normovolemic dogs at a dose of 10 mg/kg (currently registered dose) using renal scintigraphy. There were no adverse effects on renal function detected (Gogny et al, 2004). Sensitivity of NSAID in Cats The toxic effects of salicylates in cats are well documented. Cats are susceptible because of slow clearance and dose-dependent elimination. Affected cats may have hyperthermia, respiratory alkalosis, metabolic acidosis, methemoglobinemia, hemorrhagic gastritis, and kidney and liver injury. Cats also are prone to acetaminophen toxicosis because of their deficiency in drug metabolizing enzymes. Cases of acetaminophen toxicity in cats also have been well-documented. Treatment of acetaminophen toxicity consists of measures to replenish compounds that can conjugate the metabolites of acetaminophen and increase clearance, such as acetylcysteine (Hjelle and Grauer, 1986) S-adenosyl methonine (SAME). Despite the sensitivity in cats to some of the NSAID, there are still drugs in this group have been used safely. Aspirin has been used at doses of 10 mg/kg every other day. There are also reports of the safe use of ketoprofen (registered in Canada) at a dose of 1 mg/kg/day x 4 days and flunixin meglumine (1 mg/kg once) in cats for short-term treatment. Meloxicam has been used in Europe and Canada in cats. In the United States meloxicam is registered for single use at 0.3 mg/kg. The label instructions carefully warn not to administer more than one dose. When cats were administered high doses (5x dose) vomiting and other gastrointestinal problems were reported. With repeated doses (9 days) of 0.3 mg/kg per day to cats, inflamed GI mucosa and ulceration was observed. Despite these precautions, many veterinarians have administered meloxicam to cats for multiple doses. Some regimens recommend meloxicam in cats at 0.2 mg/kg initially, followed by 0.1 mg/kg per day. If a favorable response is seen in the first few days, increase the dose interval to once every 48 to 72 hours, and/or the dose lowered to 0.05 mg/kg and as low as 0.025 mg/kg. We have used meloxicam at NCSU at a dose of 0.1 mg/kg per day initially, followed by progressively lowering the dose and increasing the dose interval. Use of carprofen in cats has been discouraged because of reports of gastroduodenal toxicosis when it was administered according to canine dose rates. Tepoxalin has not been tested clinically in cats, even though pharmacokinetic studies showed that both the parent drug and metabolite would allow for safe dosing at 10 mg/kg. However, at high doses it has produced adverse effects and a safe dose for routine therapeutic use has not been identified. There is one report of use of firocoxib in cats (McCann et al, 2005). In this report, cats were given doses of 0.75 to 3 mg/kg (single dose) and it was effective for attenuating experimentally-induced fever. Hepatic Safety As pointed out in a recent review, any NSAID has the potential for causing hepatic injury (Lee 2003). The author states that NSAIDs as a class have been associated with considerable hepatotoxicity. Hepatic toxicity caused by NSAID can be either idiosyncratic (unpredictable, non-dose related) or intrinsic (predictable and dose-related) (Tolman 1998; Bjorkman 1998). Toxicity to acetaminophen and aspirin are intrinsic; reactions to other drugs tend to be idiosyncratic and unpredictable. Administration of NSAID to animals with hepatic disease has been questioned because of the role of the liver in metabolizing these drugs, but there is no evidence that prior hepatic disease predisposes a patient to NSAID-induced liver injury. Drug enzyme systems are remarkably preserved in hepatic disease and pre-existing hepatic disease is not a contraindication for administration of an NSAID. Patients with liver disease may be more prone to gastrointestinal ulceration, and there is concern that administration of NSAID could increase the risk of this complication. Carprofen was approved by FDA in October 1996 for relief of pain and inflammation in dogs. Before this approval, it was registered for treatment of dogs in Europe (Zenecarp), and was evaluated in clinical trials. In studies in dogs with arthritis, it was effective and had a low incidence of adverse effects (Vasseur et al 1995). In longer long-term studies in which carprofen was administered from 2 weeks to 5 years, the incidence of adverse reactions was only 1.3%. Vomiting, diarrhea, anorexia, and lethargy were the most common adverse reactions documented. Attention has focused on the hepatic toxicity caused by carprofen because of a report in the published literature (MacPhail et al 1998). In this report, 21 dogs were described in which carprofen was associated with acute, idiosyncratic hepatotoxicosis. Affected dogs had diminished appetites, vomited, and were icteric with elevations in hepatic enzymes and bilirubin. Dogs received the usual recommended dose and developed signs an average of 19 days after therapy was initiated. No predisposing conditions were identified. Most dogs recovered without further consequences. Many of the dogs in that report were Labrador Retrievers, but there is no follow-up evidence to show that this breed of dogs has increased risk of carprofen hepatotoxicity. Among the other drugs, the newest drug firocoxib caused fatty liver changes in young dogs when administered at high doses (manufacturer's data). Other NSAIDs used in veterinary medicine also have potential for causing liver injury. Idiosyncratic reactions are rare (1/1,000 to 1/10,000 patients). But, any unexplained increase in hepatic enzymes or bilirubin 7 to 90 days after initiating NSAID administration should be investigated. Effects on articular cartilage: Chronic therapy with some NSAID may worsen cartilage degeneration in arthritic animals. In experimental models of arthritis in dogs, lesions were worse in arthritic joints of animals treated with NSAID compared to joints not treated (Palmoski & Brandt 1982). Aspirin, indomethacin, ibuprofen, and naproxen have caused increased cartilage degeneration in arthritic joints, presumably owing to decreased synthesis of glycosaminoglycans in unstable joints. In some studies, high NSAID doses were needed to produce lesions. For example 120 mg/kg/day aspirin was used in one study involving dogs. Therefore, it is not known if NSAID administered at the usual recommended doses also are associated with clinically significant joint cartilage degradation. Although some NSAID, such as piroxicam, diclofenac, and tiaprofenic acid have been suggested to have a "chondroprotective" effect by preserving synthesis of glycosaminoglycan in arthritic joints, this has not been demonstrated in clinical patients. Carprofen and meloxicam appear to lack adverse effects on the cartilage (Benton et al 1997; Rainsford et al 1999). Indomethacin on the other hand, had significant inhibitory effects on proteoglycan synthesis (Rainsford et al 1999). Effects of carprofen on cultured osteoarthritic canine cartilage cells were examined. At the concentrations achieved clinically in articular cartilage, carprofen increased the rate of glycosaminoglycan synthesis and only at high concentrations was the synthesis inhibited. When effects of meloxicam were examined, it also lacked adverse effects on articular cartilage in vitro as demonstrated by a lack of inhibition on proteoglycan synthesis. This study with meloxicam also demonstrated that analgesia in canine arthritic joints can occur without suppression of inflammation. Often NSAID are administered with chondroprotective agents such as polysulfated glycosaminoglycan, or glucosamine/chondroitin sulfate. If there is a positive synergistic effect from these compounds, it has not been reported. Likewise, it is not established whether or not chondroprotective compounds can protect from NSAID-induced injury to joints. CLINICAL DRUG SELECTION For acute pain, such as perioperative use, veterinarians have administered some of the injectable NSAIDs with good results. Drugs used in these instances include ketoprofen, flunixin meglumine, carprofen, meloxicam, tolfenamic acid (Tolfedine, available outside the U.S.), and ketorolac tromethamine (Toradol). These drugs have been used for short-term of one or two days to decrease fever and decrease pain from surgery or trauma. Preoperative injections of carprofen to dogs were shown to be beneficial to decrease post-operative pain in dogs after ovariohysterectomy (Lascelles et al 1998). Meloxicam has been evaluated in two published studies for perioperative use and was shown to be superior to butorphanol in some of the pain assessments that were measured. Carprofen and meloxicam are the only registered injectable NSAID for dogs in the U.S. Carprofen has been shown to lack significant renal effects when administered to anesthetized healthy dogs and is safe in the perioperative period. Oral NSAID also may be used for acute treatment of myositis, arthritis, and post-operative pain, or they may be administered chronically for osteoarthritis. Drugs that have been administered in the U.S. to small animals in these cases include aspirin, phenylbutazone, ketoprofen, carprofen, etodolac, piroxicam, naproxen, and meclofenamic acid. The most recently approved drugs are carprofen, etodolac, meloxicam, firocoxib, tepoxalin, and deracoxib. Doses are listed in the accompanying table. For long-term use there are no controlled studies that compare which is the most effective. When drugs are compared to one another, it is difficult, using subjective measurements, to demonstrate differences between these drugs for reducing pain in animals. In summary, there are several choices for treating dogs with osteoarthritis with NSAID. Like people, there may be greater differences among individuals in their response than there are differences among the drugs. When selecting drugs, veterinarians sometimes select aspirin as an initial drug because it is inexpensive and familiar to most pet owners. However, gastrointestinal problems can be common with aspirin especially at high doses. Other veterinarians prefer to select a drug that is approved by the FDA such as the newer etodolac, carprofen, meloxicam, tepoxalin, or deracoxib for treatment in dogs. Many veterinarians use a rotating schedule of two or more drugs to identify which drug is better tolerated and effective in each patient. Outside the U.S. ketoprofen and tolfenamic acid have been used. REFERENCES CITED AND ADDITIONAL READING Benton HP, Vasseur PB, Broderick-Villa GA, Koolpe M: Effect of carprofen on sulfated glycosaminoglycan metabolism, protein synthesis, and prostaglandin release by cultured osteoarthritic canine chondrocytes. Am J Vet Res 58: 286-292, 1997. Bertolini A, Ottani A, and Sandrini M. Dual acting anti-inflammatory drugs: a reappraisal. Pharmacological Research 44: 437-450, 2001. Bjorkman D: Nonsteroidal anti-inflammatory drug-associated toxicity of the liver, lower gastrointestinal tract, and esophagus. Am J Med 105 (Suppl. 5A): 17S-21S, 1998. Borer LR, Peel JE, Seewald W, Schawalder P, Spreng DE. Effect of carprofen, etodolac, meloxicam, or butorphanol in dogs with induced acute synovitis. Am J Vet Res 64: 1429-1437, 2003. Boström IM, Nyman GC. Lord PF, et al. Effects of carprofen on renal function and results of serum biochemical and hematologic analyses in anesthetized dogs that had low blood pressure during anesthesia. Am J Vet Res 63: 712-721, 2002. Brown SA: Renal effects of nonsteroidal anti-inflammatory drugs. Current Veterinary Therapy X. Kirk RW (ed): Philadelphia: WB Saunders Co., 1989, 1158-1161. Bryant CE, Farnfield BA, Janicke HJ. Evaluation of the ability of carprofen and flunixin meglumine to inhibit activation of nuclear factor kappa B. Am J Vet Res 64: 211-215, 2003. FitzGerald GA, Patrono C. The Coxibs, selective inhibitors of cyclooxygenase-2. New Engl J of Medicine 345: 433-442, 2001. Forsyth SF, Guilford WG, Haslett SJ, and Godfrey J: Endoscopy of the gastroduodenal mucosa after carprofen, meloxicam and ketoprofen administration in dogs. J Small Anim Pract 39: 421-424, 1998. Gierse JK, Staten NR, Casperson GF, et al. Cloning, expression, and selective inhibition of canine cyclooxygenase-1 and cyclooxygenase-2. Veterinary Therapeutics 3: 270-280, 2002. Glaser KB: Cyclooxygenase selectivity and NSAIDs: cyclooxygenase-2 selectivity of etodolac (Lodine) Inflammopharmacology 3: 335-345, 1995. Gogny M, Fusellier M, Delpong V, Marescaux L, Desfontis J-C. Renal scintigraphy. Application to the study of the renal tolerability of tepoxalin (Zubrin) in the anesthetized dog. (presented at the Worldwide Zubrin Symposium, Athens Greece, October 2003). Hjelle JJ, Grauer GF: Acetaminophen-induced toxicosis in dogs and cats. J Am Vet Med Assoc 188: 742-746, 1986. Jones CJ, Streppa HK, Harmon BG, Budsberg SC. In vivo effects of meloxicam and aspirin on blood, gastric mucosal, and synovial fluid prostanoid synthesis in dogs. Am J Vet Res 63: 1527-1531, 2002. Kay-Mugford P, Benn SJ, LaMarre J, Conlon P: In vitro effects of nonsteroidal anti-inflammatory drugs on cyclooxygenase activity in dogs. Am J Vet Res 61: 802-810, 2000. Ko JCH, Miyabiyashi T, Mandsager RE, Heaton-Jones TG, Mauragis DF: Renal effects of carprofen administered to healthy dogs anesthetized with propofol and isoflurane. Am J Vet Med Assoc 217: 346-349, 2000. Konturek JW, Dembinski A, Stoll R. et al. Mucosal adaptation to aspirin induced gastric damage in humans. Studies on blood flow, gastric mucosal growth, and neutrophil activation. GUT 35: 1197-1204, 1994. Laneuville O, Breuer DK, DeWitt DL, et al: Differential inhibition of human prostaglandin endoperoxide synthases-1 and -2 by nonsteroidal anti-inflammatory drugs. J Pharm Exp Ther 271: 927-934, 1994. Lascelles BDX, Cripps PJ, Jones A, and Waterman-Pearson AE. Efficacy and kinetics of carprofen, administered preoperatively or postoperatively, for the prevention of pain in dogs undergoing ovariohysterectomy. Veterinary Surgery 27: 568-582, 1998. Lee WM. Drug induced hepatotoxicity. New Engl J of Med 349: 474-485, 2003. Lees P. Pharmacology of drugs used to treat osteoarthritis in veterinary practice. Inflammopharmacology 11: 385-399, 2003. Lobetti RG, Joubert KE,. Effect of administration of nonsteroidal anti-inflammatory drugs before durgery an renal function in clinically normal dogs. Am J Vet Res 61: 1501-1506, 2000. MacPhail CM, Lappin MR, Meyer DJ, et al: Hepatocellular toxicosis associated with administration of carprofen in 21 dogs. J Am Vet Med Assoc 212: 1895-1901, 1998. Malhotra S, Shafiq N, and Pandhi P. COX-2 inhibitors: A CLASS act, or just VIGORously promoted. Medscape General Medicine 6(1), 2004. (www.medscape.com) Mathews KA: Nonsteroidal anti-inflammatory analgesics in pain management in dogs and cats. Can Vet J 37: 539-545, 1996. McCann ME, Andersen DR, Zhang D, Brideau C, Black WC, Hanson PD, and Hickey GJ. In vitro effects and in vivo efficacy of a novel cyclo-oxygenase-2 inhibitor in dogs with experimentally induced synovitis. Am J Vet Res 65: 503-512, 2004. McCann ME, Rickes EL, Hora DF, Cunningham PK, Zhang D, Brideau C, Black WC, & Hickey GJ. In vitro effects an din vivo efficacy of a novel cyclooxygenase-2 inhibitor in cats with lipopolysaccharide-induced pyrexia. Am J Vet Res 66: 1278-1284, 2005. McKellar QA, Delatour P, Lees P: Stereospecific pharmacodynamics and pharmacokinetics of carprofen in the dog. J Vet Pharmacol Therap 17: 447-454, 1994. Meade EA, Smith WL, DeWitt DL: Pharmacology of prostaglandin endoperoxide synthase isozymes-1 and -2. Annals NY Academy of Sciences 714: 136-142, 1994. Moreau M, Daminet S, Martel-Pelletier J, Fernandes J, and Pelletier J-P. Superiority of the gastrointestinal safety profile of licofelone over rofecoxib, a COX-2 selective inhibitor in dogs. J Vet Pharmacol Therap 28: 81-86, 2005. Mukherjee D, Nissen SE, Topol EJ. Risk of cardiovascular events associated with selective COX-2 inhibitors. J American Medical Assoc 286: 954-959, 2001. Papich MG: Table of common drugs: Approximate dosages. In Bonagura JD (ed): Current Veterinary Therapy XIII, 2000; WB Saunders Co., Philadelphia. pp. 1241-1264. Papich MG: Pharmacologic considerations for opiate analgesic and nonsteroidal anti-inflammatory drugs. Veterinary Clinics of North America (Small Animal) 30: 815-837, 2000. Peterson WL, Cryer B: COX-1-sparing NSAIDs B is the enthusiasm justified? J Am Med Assoc 282: 1961-1963, 1999. (See also pages 1921 and 1929 of this issue.) Rainsford KD, Sherry TM, Clinderine P, Delaney K: Effects of the NSAIDs meloxicam and indomethacin on cartilage proteoglycan synthesis and joint responses to calcium pyrophosphase crystals in dogs. Vet Res Comm 23: 101-113, 1999. Reimer ME, Johnston SA, Leib MS, et al: The gastrointestinal effects of buffered aspirin, carprofen, and etodolac in healthy dogs. J Vet Intern Med 13: 472-477, 1999. Ricketts AP, Lundy KM, Seibel SB: Evaluation of selective inhibition of canine cyclooxygenase 1 and 2 by carprofen and other nonsteroidal anti-inflammatory drugs. Am J Vet Res 59: 1441-1446, 1998. Rossat J, Maillard M, Nussberger JU, et al: Renal effects of selective cyclooxygenase-2 inhibition in normotensive salt-depleted subjects. Clin Pharmacol Ther 66: 76-84, 1999. Sessions JK, Reynolds LR, and Budsberg SC. In vivo effects of carprofen, deracoxib, and etodolac on prostanoids production in blood, gastric mucosa, and synovial fluid in dogos with chronic osteoarthritis. Am J Vet Res 66: 812-817, 2005. Tolman KG: Hepatotoxicity of non-narcotic analgesics. Am J Med 105 (Suppl 1B: 13S-17S, 1998. Topol EJ. Failing the public health -- rofecoxib, Merck and the FDA. New England Journal of Medicine. 351: 1707-1709, 2004. Vane JR, Botting RM: New insights into the mode of action of anti-inflammatory drugs. Inflamm Res 44: 1-10, 1995. Vasseur PB, Johnson AL, Budsberg SC, et al: Randomized, controlled trial of the efficacy of carprofen, a nonsteroidal antiinflammatory drug, in the treatment of osteoarthritis in dogs. J Am Vet Med Assoc 206: 807-811, 1995. Wilson JE, Chandrasekharan NV, Westover KD, Eager KB, and Simmons DL. Determination of expression of cyclooxygenase-1 and -2 isozymes in canine tissues and their differential sensitivity to nonsteroidal anti-inflammatory drugs. Am J Vet Res 65: 810-818, 2004. Wolfe MM, Lichtenstein DR, Singh G: Gastrointestinal toxicity of nonsteroidal antiinflammatory drugs. New Engl J Med 340: 1888-1899, 1999. Cyclosporine: Its Clinical Use in Small Animals Cyclosporine is a fat-soluble, cyclic polypeptide fungal product with potent immunosuppressive activity. It has been an important drug used in humans, primarily to produce immunosuppression in organ transplant patients. This drug binds to a specific cellular receptor on calcineurin and inhibits the T-cell receptor-activated signal transduction pathway. Particularly important are its effects to suppresses interleukin-2 (IL-2) and other cytokines, and blocks proliferation of activated T-lymphocytes. The action of cyclosporine is more specific for T-cells as compared to B-cells. One important advantage in comparison to other immunosuppressive drugs is that it does not cause significant myelosuppression or suppress nonspecific immunity. Clinical Use Cyclosporine has been used for a number of diseases in veterinary medicine. Many of these diseases have been dermatologic, as reviewed in a recent paper by Robson & Burton, (2003). In dogs, when used in the treatment of perianal fistulas, (Mathews et al 1997; Griffiths et al 1999) an 85% healing rate was found in one study (Mathews, et al 1997) (2.5-6 mg/kg/day); in sebaceous adenitits, good response was reported in one case (Carothers et al1991) and in other cases as long as treatment was continued (Linek et al, 2005). It has been effective for idiopathic sterile nodular panniculitis in which excellent results were seen in 2 reported cases which were followed for 6 months following discontinuance of the cyclosporine (Guaguere 2000). It may be effective for treating dogs with granulomatous meningoencephalitis although some of these dogs received higher doses than used for atopy (Adamo et al, 2004). There are unpublished accounts of successful treatment with cyclosporine in dogs with inflammatory bowel disease. Immune-mediated diseases: Cyclosporine has been used for treatment of a variety of immune-mediated diseases, that include immune-mediated hemolytic anemia (IMHA), inflammatory bowel disease (IBD), immune-mediated polyarthritis, and aplastic anemia (AA), as well as others. For these diseases, the true efficacy has not been measured on the basis of controlled clinical studies. Results are primarily anecdotal. It is generally accepted that the dose should be in the range of at least 10 mg/kg/day, and perhaps twice a day to produce clinical effects. Trough blood cyclosporine concentrations should be at least 600 ng/mL. For immune-mediated skin diseases, in small pilot studies results for treating pemphigus foliaceous in dogs have been disappointing (Rosenkrantz et al, 1989). It did not help any patients with mycosis fungoides. In a study in which 5 dogs with pemphigus foliaceus were treated, there was little benefit (Olivry et al, 2003). The dogs with pemphigus foliaceus were initially treated with the "atopy dose" of 5 mg/kg/day. If there was no response the dose was increased to 10 mg/kg/day. At the end of the trial, it was concluded that at 5-10 mg/kg/day cyclosporine was unable to produce complete remission in any of the 5 dogs. In a case study of immune-mediated polyarthritis in dogs (Clements et al, 2004), dogs were treated with prednisolone and various other immune-modifying drugs. Three dogs treated with cyclosporine at 5 mg/kg per day did not respond. Atopic dermatitis: In people it has been used successfully for treatment of atopic dermatitis (Camp et al 1993). Because of this efficacy, the use of cyclosporine for treating canine atopic dermatitis has been investigated in dogs (Marsella & Olivry, 2001). In a pilot study, cyclosporine was effective in 13/14 dogs with atopic dermatitis (Fontaine & Olivry, 2001) . In another trial of 30 dogs treated with either cyclosporine (5 mg/kg/day) or prednisolone (0.5 mg/kg/day), the efficacy of cyc |