Antimicrobials and UTIs: part 3-building the dosing regimen (Proceedings)

Dosing regimens for antimicrobials should be related to MIC. However, simply achieving the MIC in the patient is not likely to be sufficient for a variety of reasons. This reflects, in part the fact that susceptibility cards used by microbiology laboratories only test in a very limited range. Thus an isolate that is designated as "S" may very well have already implemented the first step toward resistance, resulting in a higher MIC that is still considered "S". Thus, culture procedures largely test for resistance, not susceptibility, failing to provide information about how susceptible an isolate is to the drug of interest.

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 to therapeutic failure. 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. For UTI, there is no question that cystocentesis is the preferred technique, even if an animal is catheterized. Just as absence of growth does not indicate absence of infection, isolation of an organism is not necessarily evidence of infection. Clearly, culture of an organism from a tissue that is normally sterile indicates infection. For the urine, 1000 CFU/ml might be a reasonable threshold. Shipment of urine specimens may complicate colony counts; accordingly, a urine paddle system should be considered. 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 federal 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 tested as a model for other drugs in the class. For some classes of drugs, cross-reactivity can be similar within the class (for example, an organism that is R to one fluorinated quinonolone (including ciprofloxacin) is likely to be R to all). However, the same is not true for others. Amikicacin is often more effective than gentamicin for (hence both are often on a report). Cephalothin serves as a model for all 1st generation cephalosporins but it underestimates efficacy of cefazolin toward Gram negative organisms. The spectrum of 3rd and 4th generation cephalosporins is too variable to allow one to predict the susceptibility of others and as such, multiple drugs are likely to be included. Culture and susceptbility techniques may not accurately reflect resistance that has developed in the infecting organism to drug to which the organism is generally susceptible. These cephalosporins also offer another example of concern: they are susceptible to extended spectrum beta-lactamase (ESBL) that will be produced in vivo but not in vitro and despite an "S" designation, therapeutic failure may occur.

Bactericidal versus bacteriostatic drugs

The term "bactericidal" is so often abused that the distinction from bacteriostatic should be underemphasized. Although it is appropriate for clinicians to reach for a drug that is "cidal" rather than "static", it is not appropriate to assume that the ability of that drug to kill rather than simply inhibit an organisms will enhance therapeutic efficacy. This may be true, but only if concentrations of the drug achieved at the site of infection are sufficient to kill the microbe. The term "bactericidal" is an in vitro definition and is based on killing rates (eg, 99.9% reduction in bacterial inoculum within a 24 hr period of exposure) as well as the proximity of the minimum bactericidal concentration (MBC) of a drug to the MIC. The MBC is determined based on kill curves, or following tube dilution procedures: tubes with no observable growth are inoculated on agar gel. For "bacteriostatic" drugs, growth on the agar plate will occur for several tube dilutions above the MIC, indicating that organisms were not killed. However, bacteriostatic drugs are capable of killing (eg, some organisms are exquisitely sensitive to the effects of selected drugs; some "static" drugs are acculumulated to concentrations that are likely to be cidal [eg, macrolides and lincosamides in phagocytes; urine concentration). However, killing concentrations are generally more likely to be achieved for a cidal drugs compared to a static concentration.

Relationship between MIC, Plasma and tissue Drug Concentrations

The bridging of pharmacodynamic (susceptibility) data with pharmacokinetic data can begin by comparing the MIC established by in vitro testing and the concentrations of drug achieved in the patient at the site of infection when the drug is administered at the labeled dose. A number of investigators have examined several parameters that combine pharmacokinetic (PK) and pharmacodynamic (PD) indices as predictors of clinical antimicrobial efficacy. Among those most commonly sited as potentially useful are the ratio of Cmax (peak PDC) to MIC, defined here as IQ; the area under the inhibitory curve (AUIC; the ratio of area under the PDC versus time curve to the MIC of the infecting organism); and the percent time that PDC are above the MIC [T> MIC]. Based on these relationships, two generally categories of drugs have been described.

Time versus Concentration Dependent Drugs

The relationship between MIC and the magnitude and time course of PDC allows drugs to be categorized as to either concentration-dependent (sometimes referred to as dose dependent) or time-dependent; these definitions are supported by primarily by in vitro but also in vivo studies. Concentration dependent drugs, best represented by the fluoroquinolones and aminoglycosides, are characterized by efficacy best predicted by the magnitude of plasma drug concentration (Cmax) compared to the MIC of the infecting organism (ie, the inhibitory quotient or IQ). For such drugs, the magnitude of the IQ generally should be 8-10 but ideally is higher for more difficult infections (eg, Pseudomonas aeruginosa, or infections caused by multiple organisms. The duration that PDC is above the MIC is not as important as is the IQ; in fact, efficacy may be enhanced by a drug-free period (i.e., a long interval between doses). For concentration dependent drugs, a dose that is too low is particularly detrimental. In a mouse model of E. coli peritonitis, the antibacterial efficacy of ciprofloxacin, but not imipenem, was improved by doubling the dose. As such, concentration-dependent drugs generally can be administered at longer intervals, ie, once a day. However, for the fluorinated quinolones (FQ), efficacy may alsobe time related. Efficacy of FQ also is predicted in vitro by AUIC: a ratio of < 60 renders the drugs bacteriostatic, whereas ≥ 125 results in (slow) killing but also decreases the risk of resistance and ≥ 250 causing more rapid bacterial killing. Thus, resistance might be less likely to develop for FQ characterized by longer half-lives (or for ENR, by the production of an active metabolite). Twice daily administration of an FQ might be indicated for organisms already characterized by low level resistance (see MPC below); however, the once daily dose should be given twice daily in such situations. In contrast to concentration dependent drugs, efficacy of time-dependent drugs (eg, β-lactams) is enhanced if PDC remain above the MIC for the majority (60 to 70%) of the dosing interval; efficacy is best predicted by percent time that PDC are above the MIC [T> MIC]. For such drugs, simply achieving the MIC (IQ = 1) is insufficient because PDC (and certainly tissue concentrations) fall below the MIC immediately. With time-dependent drugs, increasing the IQ also may be beneficial, even though although efficacy clearly is related to T>MIC, simply because increasing the dose will increase time above the MIC. Since drug concentrations decrease by 50% every drug half-life, an IQ of two will result in PDC below the MIC in one half-life. However, the relationship between dose increase and time above the MIC (in terms of half-life) is logarithmic, not linear. To increase the dosing interval by two half-lives, the same dose (designed to simply achieve the MIC) would need to be quadrupled; to add three half-lives onto the dosing interval, the dose would have to be increased 8 fold. For example, for a Staphylococcus intermedius with an MIC for axoxcillin-clavulanic acid of 4 mcg/ml, assume PDC achieve the breakpoint MIC (32 mcg/ml) when a recommended dose is administered (13 mg/kg). In one half-life (1.5 hr), PDC have dropped to 16; in 3 hr, to 8 and in 4.5 hr, to 4 mcg/ml. Another 25% of the dosing interval (1.5 hr) can elapse before the animal should be redosed. If the dose is doubled, this can be extended to 8 hours; the dose would have to be quadrupled to extend to 9 hrs, and increased 8 fold to reach the standard 12 hrs. This is for an organism with a relatively low MIC; the problem is markedly compounded for organisms still considered "S" but with MIC that approach the breakpoint MIC. Adding a 3rd or 4th dose is a much more cost effective method of increasing efficacy (compared to increasing the dose) of time dependent drugs; as such, efficacy of time dependent drugs may be best facilitated by both a dose increase (to assure PDC are sufficiently higher than MIC) and an interval decrease increase. Constant rate infusion or slow release products might be ideal for time dependent drugs, assuming drug release (from slow release products) is sufficiently fast to allow Cmax to surpass the MIC. Efficacy also should be enhanced for time dependent drugs which persist in selected tissues (again, assuming the MIC is surpassed). An example would be drugs which accumulate in active (unbound) form in tissues (ie macrolides) or drugs that accumulate in phagocytes).

Postantibiotic Effect

The post antibiotic effect (PAE) describes the continued inhibition of microbial growth after a short exposure of the organisms to the drug. The impact of the PAE on antimicrobial efficacy can be profound, particularly for concentration-dependent drugs. It is the PAE that allows some of these drugs to be administered at long intervals. The PAE may be absent for some organisms or some patients (e.g., some immunocompromised patients). In general, concentration- dependent drugs appear to exhibit longer PAE; further, the duration of the PAE may vary with the magnitude of the peak PDC (ie, longer with higher PDC) and is enhanced by combination antimicrobial therapy. PAEs vary with each drug and each organism.

Penetrating the site of infection: Interpretation of C&S is based on the assumption that the MIC should be achieved in plasma. Basing MIC interpretation on plasma drug concentrations (PDC) might result in over or under estimation of drug efficacy. For tissues which concentrate the drug (or if the drug can be applied topically), and for drugs which can be concentrated by phagocytes and thus transported to the site of infection, concentrations may markedly exceed PDC, resulting in underestimation of efficacy. In such circumstances, a drug noted as "I", or,under the appropriate circumstances (eg, topical therapy) a drug designated as "R" might actually be effective. The more likely scenario is overestimating drug efficacy because of failure to achieve the anticipated PDC at the site of infection. Most infections occur in extracellular tissues; because most tissues are characterized by fenestrated capillaries, antimicrobials can move unimpeded from plasma into ECF . However, whether or not concentrations in ECF are equivalent to that in plasma is not clear. Much of the data for water soluble drugs (volume of distribution [Vd] generally ≤ 0.3 L/kg) suggests antimicrobial concentrations may be 30% or less of PDC in some tissues, although the method of fluid collection used among the studies (eg, tissue homogenate versus tissue cages versus ICF collection) may impact the results. However, studies do suggest for some tissues, doses may need to be increased several fold to target effective concentrations. Certainly tissues characterized by non-fenestrated capillaries (including the prostate, eye, brain, testicles, cartilage) will be difficult to penetrate. In humans, recommended doses of beta-lactams drugs (water soluble) are increased 5 to 10 fold when treating infections of the central nervous system. For difficult to penetrate tissues, intracellular infections, or perhaps infections characterized by marked inflammatory debris, lipid soluble drugs (Vd generally ≥ 0.6 l/kg) may be more likely to distribute to sites of infection at concentrations necessary to kill organisms. However, note that a Vd > 0.6 L/kg does not necessarily indicate better penetration of a drug; Vd is a theoretical compartment that describes how much tissue dilutes a drug, but provides no indication as to where the drug distributions. In the presence of marked inflammation, use of a drug that accumulates in phagocytes (eg FQs, macrolides, lincosamides) is likely to increase distribution of the drug to the site.

Urine vs PDC

Ideally, a drug that is renally excreted should be selected for treatment of UTIs. Urinary concentrations of such drugs often surpass serum concentrations (up to 300-fold), and as such susceptibility data should be based on urinary rather than plasma drug concentrations. Indeed, drugs that might not be useful for treatment of non-UTIs (because of failure to achieve MIC in blood or tissues) often can be used to treat UTIs (e.g., carbenicillin, nitrofurantoin). In addition, renal elimination may result in bactericidal concentrations of drugs for which only bacteriostatic concentrations can be safely achieved in serum. Several caveats must be recognized, however, when basing antimicrobial selection on renal elimination and anticipation of high urine drug concentrations:

1. If the UTI is associated with infection in the blood (or in the presence of bacteremia), kidney, or prostate, then antimicrobial selection should be based on anticipated plasma (or tissue) drug concentrations and serum breakpoint MIC. Urine concentrations, albeit higher than plasma concentrations, will be helpful, but do not translate to higher concentrations in any tissue other than the urine itself.

2. Exceptions may be made in the presence of decreased renal function. If renal function is sufficiently decreased, urinary drug concentrations also will be decreased. In addition, depending on the safety of the drug, drug doses may need to be decreased.

3. Many antimicrobials are characterized by a short elimination half-life. For renally eliminated drugs, however, plasma elimination half-life may not accurately reflect contact time of drug in the target tissue (i.e., urine). Presumably, drug eliminated in the urine will be in contact longer with the infected tissue (i.e., lower UTI) longer than other tissues, and therefore the basis for a recommendation to use drugs at a shorter interval when treating UTIs compared with other infections may not be relevant for UTIs. However, this also assumes that the patient retains the urine in the bladder. Accordingly, treatment of UTI's might include treating the patient at night such that urine is retained, and consideration for crating the patient during the day for at least 50% of the dosing interval.

Mutant Prevention Concentration.

Perhaps the most important reason to target drug concentrations much higher than the MIC is the concept of the Mutant Selection Window which is the concentration between the MIC obtained on culture and the MPC. Based on natural mutational frequencies, a mutation that leads to a single step in resistance to an antibiotic can be expected in populations whose density is ≥ 107 colony forming units (CFU). Because most C&S procedures are based on 105 CFU, these organisms are not likely to be detected on C&S, yet, if the inoculum is large enough, they will be in the patient. Simply achieving the MIC of the cultured organism will inhibit the growth of all isolates except those whose MIC exceed that reported, including the first step mutants. The larger the inoculum, or the more immunocompromised the patient, the more likely resistant mutants will survive antimicrobial therapy administered at a dose designed to simply achieve the reported MIC. Subsequent growth of these mutant organisms will yield a population still considered susceptible, but with higher MIC than the original (cultured) population. Once this new population of organisms with higher MIC achieves ≥ 107 CFU, a second mutation will yield organisms that are characterized by a high level of resistance. The MIC of this second population of organisms will no longer be achievable at recommended doses. In vitro evidence is supported by in vivo experiences describing this scenario; further, clinical experience suggests this to be true, particularly for animals with chronic UTI, otitis externa or other infections characterized by previous antimicrobial use. A first step mutation has been demonstrated in bacterial prostatitis in humans receiving only three days of ciprofloxacin at doses designed to achieve the MIC; prophylactic doses of ciprofloxacin for urological procedures was associated with the advent of resistant E coli. Not surprising, a novel approach intended to avoid second step mutation and high level resistance has been proposed in the design of dosing regimens in humans. The goal of therapy is to achieve the mutant prevention concentration (MPC), rather than the MIC, of the drug (at the site) in the patient. Achieving concentrations in the mutant selection window should be avoided; the bottom of the window is the MIC reported by culture techniques (based on 105 CFU) whereas the top of the window is the MPC. The MPC is defined as the highest MIC identified in a population (≥107) of susceptible organisms, thus including those organisms that have undergone the first step mutation; alternatively, it is defined as the drug concentration that would require an organism to develop two concurrent mutations in order to grow in the presence of the drug. Whereas the MIC indicates the concentration of drug which inhibits the growth of the organism, the MPC indicates the concentration of drug that would minimize the advent of resistance. To prevent the advent of resistance, the concentration of the drug at the site of infection should be sufficiently high to inhibit the growth of the organisms that have undergone the first step mutation. Interestingly, achieving drug concentrations below the MIC of the infecting organism is not likely to facilitate the advent of resistance because selection pressure will not be applied to the infecting organism. Unfortunately, determining the MPC of an isolate cultured from a patient requires culture techniques based on ≥ 107 organisms, currently cost prohibitive. Insufficient information is available to allow a reasonable prediction of the relationship between the MPC and MIC of an organism, which is likely to vary with the drug and the organism. For veterinary FQs, Wetzstein and coworkers have demonstrated the MPC to MIC ratio to be approximately 6 to 10 for E coli (ATCC 8739) but as high as 23 to 50 for Staphylococcus aureus (ATCC 6538).

Host Factors

The impact of host response to infection can be profound. Problems contributing to therapeutic failure for UTI include but are not limited to: conformational problems (including PU), immunocompromise (including treatment with immunosuppressive drugs or anticancer drugs), diabetes mellitus or hyperadrenocorticism, and infection elsewhere in the urinary tract.

Microbial Factors

Microbial resistance is addressed in another manuscript in this same proceedings. Materials released from microbes facilitate invasion, impair cellular phagocytosis, and damage host tissues. Most staphylococci associated with canine pyoderma produce "slime," a material that facilitates bacterial adhesion to cells. Soluble mediators released by organisms (hemolysin, epidermolytic toxin, leukocidin) may damage host tissues or alter host response. Staphylococcal organisms contain protein A, which impairs antibody response, activates complement, and causes chemotaxis. Nocardia stimulates the formation of calcium-containing "sulfur granules" that impair drug penetration to the organisms Pseudomonas and other gram-negative organisms produce a glycocalix, or biofilm, that protects the organism.. Biofilms are microcolonies of pathogenic and host microbes embedded in a polysaccharide matrix ("slime" or "glycocalyx") produced by the bacteria; dental plaque is the prototypic example. Normal microflora of the skin or mucous membranes in the biofilm are lost with shedding of the skin surface or by the excretion of mucus; new cells and mucus are rapidly colonized by biofilm forming bacteria. Translocation of the normal microflora to otherwise sterile tissues (which can be facilitated by the presence of foreign bodies) may lead to acute infections (again, associated with biofilm) and accompanying inflammatory response. Persistent, chronic bacterial infections may reflect biofilm producing bacteria; persistant inflammation associated with immune complexs contributes to clinical signs. Unfortunately, bacteria growing in biofilms more easily resist antimicrobial killing and immune defenses of the host. In addition to debridement or other methods of cleansing should facilitate antimicrobial penetration; dose modification (increase) may be indcated to compensate for debris.

Drug Factors

In addition to drug characteristics previously addressed (eg, concentration versus time dependent, static versus cidal, drug distribution), pharmaceutical manufacturers have been able to manipulate antimicrobial drugs in a variety of ways such that efficacy and thus bacterial killing is enhanced such that resistance might be reduced. For example, efficacy has been decreased by synthesizing smaller molecules that can penetrate smaller porins (e.g., the extended spectrum penicillins ticarcillin and piperacillin); "protecting" the antibiotic (e.g., with clavulanic acid, which "draws" the attention of the β-lactamase away from the penicillin); modifying the compound so that it is more difficult to destroy (e.g., amikacin, which is a larger and more difficult to reach molecule than gentamicin); and developing lipid-soluble compounds that are more able to achieve effective concentrations at the site of infection (e.g., doxycycline compared with other tetracyclines). However, with each innovative approach to reducing resistance, microbes are able to circumvent the drug in a disconcertingly short time. The use of pro-biotics or pre-biotics to minimize emergence of resistance in the gastrointestinal tract is controversial and requires additional scientific evidence.

In conclusion, increasingly, rationale use of antimicrobial therapy should focus on short term therapy with concentrations sufficiently high to kill the infecting microbes, thus assuring that mutation to resistant organisms does not occur. Design of doses are most appropriately based on culture and susceptibility data that indicates the MIC; the closer the MIC is to the breakpoint MIC, the more important increasing doses (of concentration dependent drugs) and shortening the interval of time dependent drugs. Combination therapy is another method whereby the advent of resistance can be minimized.