Reducing the risk of antimicrobial resistance: Picking the right drug (Proceedings)

Introduction

"Even experienced practitioners may not realize that giving a patient antibiotics affects not just that patient, but also their environment, and all the other people that come into contact with that environment". With that statement Dancer (2001) summarizes the importance of judicious antimicrobial therapy in veterinary (and human) medicine. In addition to the impact in human medicine, the advent of resistance has and is increasingly impacting successful therapy in veterinary patients. In human medicine, antibiotic stewardship has become the focus for reducing resistance. Prudent veterinarian and veterinary practices will implement decision making processes (antimicrobial use paradigms) that minimize the temptation to use antimicrobials as alternative therapies. The age of designing a dosing regimen based on cost and convenience rather than pharmacodynamics and pharmacokinetics must end. Antimicrobial stewardship begins by recognizing the problems and issues leading to antimicrobial resistance, and successfully implementing procedures that reasonably minimize the impact of antimicrobial use in the patient while not forfeiting the likelihood of therapeutic success. The goals of antimicrobial therapy is not simply to eradicate the infection, but also to avoid resistance. The two do not necessarily go in hand unless all infecting microbes are eradicated (DEAD BUGS DON'T MUTATE).

The most important considerations for selection of an antimicrobial are in order of priority: 1. confirming (as much as possible) the need to treat (in other words, do not treat indiscriminately); 2.identifying the target organisms such that the spectrum of drug can be as narrow as possible; matching the drug to the bug, which includes assessing the relative susceptibility of the bug to the chosen drug, and 4. identifying the target site and assessing the likelihood of drug delivery to that site. The 5th through nth steps of antimicrobial selection must take into account the host (including tissue distribution, but also degree of inflammation, and underlying factors that predispose to recurrent infection), and microbial factors (eg, glycocalyx [biofilm], intracellular location) that can negatively impact the selection of the drug and the design of the dosing regimen. Among the more relevant considerations are identifying mechanisms of resistance, the likelihood of achieving bactericidal (DEAD BUGS DON"T MUTATE] concentrations at the site, and time- versus concentration- dependency as it relates to convenience of the dosing interval. Adverse reactions tend to be less important than other classes of drugs because, as a general rule, antimicrobials tend to be safer. Some notable exceptions occur, particularly in cats (eg, fluoroquinolones and retinal degeneration). Cost should be the last consideration that influences selection. Selection should strive for a spectrum that is as narrow as possible, thus avoiding the sequelae of drug use on the normal flora and unnecessary selection pressure.

Bridging Pharmacokinetics And Pharmacodynamics

For complicated infections, or in patients at risk for therapeutic failure or developing resistance, identification of the infecting organism most appropriately should be based on appropriately collected cultures with (ideally) tube dilution susceptibility testing to allow assessment of "how" susceptible the isolate is, in general, to other drugs, but specifically to the drug of choice, such that a dosing regimen can be designed for the bug in the patient. Basing antimicrobial selection on C&S data does not guarantee success, just as failing to use C&S as a basis for selection (or selecting a drug characterized by "R" on the data) does not guarantee failure. The "90-60 rule" implies that approximately 90% of infections treated based on C&S are likely to respond if an "S" drug is selected; yet, up to 60% will respond even if an "R" drug is selected. The most likely situations where the latter is true is if the infection is at a site in which drug much higher than that achieved in the test tube (ie, much higher than the minimum inhibitory concentration (MIC).

To wait or not to wait? Frequently, antimicrobial therapy is begun before cultures are collected. This is particularly true and appropriate in critical patients; therapy also must begin in patients for which clinical signs are evident and are detracting from quality of life (patient or parent). However, should therapy begin and the choice prove to be wrong once C&S data is received, the original C&S data collected before the drug was begun may no longer be relevant to the patient. The use of the drug may change the pattern of resistance versus susceptibility, or may result in higher MIC (see mutant prevention concentration). A reculture may be indicated at that time, if possible. Certainly, dosing regimens with the appropriate drug should take into account the possibility that some level of resistance has developed toward the indicated drug.

Culture And Susceptibility Testing: A Tool To Guide Therapy

Pitfalls of Susceptibility Testing: Although culture and susceptibility data (C&S) can be a powerful tool to guide selection, it nonetheless is an in vitro test applied to in vivo conditions; over-reliance on the information can contribute to therapeutic failure. As early as the collection process, C&S data can be misleading.

To grow or not to grow: It is beyond the scope of this manuscript to delineate proper techniques of specimen culture collection, but C&S data is no more accurate than the sample collection; close adherence to recommended procedures including but not limited to site selection, site preparation and sample handling are critical to proper interpretation. Anaerobic infections are particularly problematic. Obligate anaerobes are exquisitely sensitive to increased oxygen tension and will not survive if exposed to oxygen. No growth may be mistakenly interpreted as lack of anaerobic infection. The presence of a foul smell or abscess, and (less commonly seen) gas or pigment changes support infection with anaerobes and drug selection might include efficacy toward anaerobes. Many organisms are facultative anaerobes, capable of growth in anaerobic environments. Aerobic cultures may yield their growth, but the anaerobic environment in the patient may limit response to antimicrobials (particularly aminoglycosides). Organisms without cell walls (Mycoplasma, L-forms, etc) and others difficult to culture, or slow growing organisms (anaerobes, selected Gram positives, Nocardia , atypical Mycobacteria and others and C&S data may not include MIC.

To treat or not to treat: Just as absence of growth does not indicate absence of infection, isolation of an organism is not necessarily evidence of infection, nor even if infection is present, does the isolated organism represent the infecting organism. Clearly, (properly collected) culture of an organism from a tissue that is normally sterile indicates infection. However, discriminating between normal and infecting flora can be difficult. Pure, vibrant (meaning special media was not needed to coax the growth of the organism) of a large number of colonies are indicators of infection. The isolation of three or more different organisms (including more than one strain of the same organism) may indicated contamination and reculture should be considered. The extent of growth should be strongly considered when deciding to treat. Generally, infection is considered to be present if >107 CFU/ml at the site. For C&S purposes, quantitative cultures can be helpful: the urinary tract is not considered infected until >105 CFU /ml are present whereas only > 103 is indicative of infection in the respiratory tract. Laboratories may indicate "heavy, moderate or light" growth. For multiple organism, that with the greatest growth should be the primary focus of therapy. A call to the diagnostic lab might be prudent before marked financial commitment is put into treating an organism that is not causing infection. This is particularly important if the organisms' presence is unusual (eg, Lactobacillus sp in urine).

To trust or not to trust: The C&S procedures themselves are fraught with potential errors. For practices that provide in- house susceptibility testing, care must be taken to follow guidelines established and published by (or comparable to) the Clinical and Laboratory Standards Institute (CLASI) or comparable standard-setting agency. Materials, including interpretive standards, should be validated by the appropriate agency. Minor changes in pH, temperature, humidity, etc can profoundly affect results. Personnel should be trained specifically in culture techniques and hospitals that provide this service (as do diagnostic labs) should maintain well designed and adequately collected quality control data to validate their procedures (CLASI indicates control organisms). Pitfalls of susceptibility testing also reflect the drugs selected for testing. Not all companies are interested in establishing interpretive criteria and as such, not all drugs are available for testing. Because automated systems can not accommodate and laboratories (nor clients) can not afford to test all potential drugs used to treat an infection, one drug often is tested as a model for other drugs in the class. For example, cephalothin models first generation cephalosporins, even though it is no longer used clinically. Note that it does not represent cefazolin well, the latter being more effective toward Gram negative (especially E coli ) isolates. No single cephalosporin can represent 2nd or beyond generations. Enrofloxacin often represents the fluoroquinolones. In general, cross resistance can be expected among the FQs, although differences in potency do exist (for example, ciprofloxacin is more potent toward Pseudomonas or E coli, but less to Gram positives compared to enrofloxacin). Culture does not take into account active metabolites of some drugs (eg, enrofloxacin converted to ciprofloxacin). Note that if an organism is R to any FQ, FQ should be used only cautiously even if another is "S". Amikicacin is often more effective than gentamicin toward many organisms, but less effective toward Staphylococcus sp. (hence both are often on a report). Note that CLSI interpretive criteria are generated for specific species, and often for specific organisms and specific infections. Human laboratories will use human interpretive criteria, which often are not relevant to animals (eg., ciprofloxacin). Fewer interpretive criteria are available for antibiotics used in animals. A "good" laboratory will note when interpretive criteria should be considered cautiously. A final concern relates to the 3rd and 4th generation (extended spectrum) cephalosporins: they are susceptible to extended spectrum beta-lactamase (ESBL) that tend to be induced in vivo but often missed in vitro. Despite an "S" designation, unless specifically tested for, therapeutic failure may occur. Ceftazidime or cefotaxime is often used as a test for the presence of ESBL; imipenem and clavulanic acid (eg, amoxicillin-clavulanic acid) generally are not susceptible to these enzymes. Laboratories currently are generating methods intended to detect this level of resistance.

Bridging Pharmacokinetics And Pharmacodynamics

Interpreting C&S (pharmacodynamic) data: This is most helpful when considered in the context of the clinical patient. The first step of antimicrobial therapy should focus on comparing what is needed – the MIC of the isolate for the drug of interest – to what is achieved. MIC data is generated from tube dilution procedures which subject the isolate collected from a patient to increasing concentrations of the drug of interest. The concentrations increase two fold, and although the concentrations tested are the same for all drugs, the range tested varies with the drug depending on concentrations of the drug achieved in the patient.

Even if samples are properly collected (a big IF) and cultured (use a laboratory that follows CLASI guidelines), C&S data is inherently deficient because in vitro methods can not mimic in vivo conditions. For example, the isolate is exposed to the constant conditions, including drug concentrations throughout the in vitro incubation period; in vitro methods can not take into account host factors which detract from efficacy. Protein binding is not considered. Among the most problematic concerns is interpretation of MIC. The breakpoint MIC (MICBP), (Table 1) the threshold of susceptibility established by the Clinical Laboratory Standards Institute (CLASI) is based, in part, on peak plasma drug concentrations (Cmax, Table 2) established, ideally, in the species of interest, and tested against pathogens infecting the targeted animals. An isolated considered susceptible is characterized by an MIC below the MICBP (usually 1-3 tube dilutions) whereas a resistant isolate is one whose MIC is equal to or surpasses the MICBP; isolates whose MIC is approaching but not yet equal to the MICBP are considered intermediate (Table 1). Many drugs used by veterinarians are approved for use in humans. Although interpretive MIC data has been determined by CLASI for some drugs in animals, many have not and interpretation may be inappropriate. Ciprofloxacin (CIP) is an excellent example: its oral bioavailability in dogs is 30 to 40% of that in humans, and despite its increased potency compared to enrofloxacin (ENR) toward Gram negative organisms, its potential efficacy (MICBP) is equivalent to or less for many organisms. Susceptibility data also does not take into account active metabolites, again exemplified by ENR, which is metabolized to CIP: both Cmax and area under the curve (AUC) of bioactivity of ENR may increase up to 50% or more by CIP; as such, C&S data may underestimate efficacy. MICBP generally are based on the highest labeled dose, but higher doses might be safely administered for many antibiotics. If recommended doses change, the manufacturer should provide CLASI with updated pharmacokinetic information so that interpretive criteria may change accordingly, and automated systems should incorporate those changes in their methods. Again, ENR offers an example. Originally approved at 2.5 mg/kg, 1 μg/ml (≤ 1= S; ≥ 2 =R) was the MICBP; the current dose is up to 20 mg/kg and new interpretive criteria identify ≥ 4 μg/ml = R. One of the disadvantages of current susceptibility testing is that the concentrations tested are close to the MICBP and thus, does not allow identification of isolates that are very susceptible (that is, MIC are far away from the MICBP). As such, drugs may be chosen based on isolates that have already undergone first step mutations (see below).

Table 1. CLSI interpretive criteria for selected drugs

Concentrations

In the event that patient pharmacodynamic information is not available, population data might be used to guide therapy (. Ideally, population data reflects the MIC of the drug of interested determined in at least 100 isolates of the same organism collected from the same species. A distribution curve of the resulting MIC yields a range, the median (MIC50) and the MIC90. Simplistically, the MIC90might be interpreted as the concentration needed to inhibit 90% of the isolates of the organism; alternatively, there is a 90% probability that the organism infecting the patient will be at that concentration or lower. Population information is available on package insert for new drugs (see figure). This data can be used to select drugs or to desdosing regimen. Selecting a drug based on package insert: Using Proteus as an example, comparison of Cmax to MIC90 reveals a ratio 2:0.125 or 16 for marbofloxacin compared to 1.8:1.8 or 1 for difloxacin, using the low dose for each drug. For E. coli, the numbers are 2:0.06 or 33 for marbofloxacin compared to 16 for difloxacin. For either organism, marbofloxacin offers the best ratio. For E. coli, the low does might be acceptable for marbofloxacin, and potential for difloxacin, although the latter might not be prudent. For Proteus, again, the low dose of marbofloxacin might not be prudent, and difloxacin should not be used to treat Proteus.

Plasma pharmacokinetics following administration of difloxacin tablets and MIC values of difloxacin for bacterial pathogens