Goodman & Gilman's: The Pharmacological Basis of Therapeutics, 12e

Chapter 48: General Principles of Antimicrobial Therapy

Tawanda Gumbo

SCIENTIFIC BASIS OF ANTIMICROBIAL CHEMOTHERAPY

An important revolution in human understanding of nature was the germ theory of disease based on the work of Louis Pasteur and Robert Koch, which linked specific microorganisms to specific diseases (Koch, 1876; Pasteur, 1861). Modern chemotherapy is predicated on this idea, radical in its time. The germ theory developed considerably in the 20th century, with identification and characterization of many microbial pathogens and their pathogenic mechanisms and the introduction of antimicrobial drugs. With the use of these drugs came issues of appropriate regimens, drug resistance, drug interactions, and toxicity.

Although the antimicrobials are a large group of diverse structures with myriad mechanisms of actions against bacteria, viruses, fungi, and parasites, we can, nonetheless, develop generalizations about important issues surrounding the use of antimicrobial agents. This chapter reviews the general classes of antimicrobial drugs, their mechanisms of action, mechanisms of resistance, and patterns of kill by different classes of the drugs. It also discusses principles for the selection of an appropriate antibiotic, dose, dose schedule, and type of antibiotic therapy. The pharmacological properties and uses of individual classes of antimicrobials are discussed in Chapters 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59.

Classes and Actions of Antimicrobial Agents. Microorganisms of medical importance fall into four categories: bacteria, viruses, fungi, and parasites. The first broad classification of antibiotics follows this classification closely, so that we have (1) antibacterial, (2) antiviral, (3) antifungal, and (4) antiparasitic agents. Within each of these major categories, drugs are further categorized by their biochemical properties. Antimicrobial molecules should be viewed as ligands whose receptors are microbial proteins. The term pharmacophore, first introduced by Ehrlich, defines that active chemical moiety of the drug that binds to the microbial receptor. The microbial proteins targeted by the antibiotic are essential components of biochemical reactions in the microbes, and interference with these physiological pathways kills the microorganisms. To the extent that an antimicrobial agent targets a protein that is not widely expressed in other bacteria, the drug will be relatively selective in its antimicrobial effect. The biochemical processes commonly inhibited include cell wall synthesis in bacteria and fungi, cell membrane synthesis, synthesis of 30s and 50s ribosomal subunits, nucleic acid metabolism, function of topoisomerases, viral proteases, viral integrases, viral envelope fusion proteins, folate synthesis in parasites, and parasitic chemical detoxification processes. Classification of an antibiotic is based on:

the class and spectrum of microorganisms it kills
the biochemical pathway it interferes with
the chemical structure of its pharmacophore

Because antimicrobial agents are ligands that bind to their targets to produce effects, the relationship between drug concentration and effect on a population of organisms is still modeled using the standard Hill-type curve for receptor and agonist (Chapters 2 and 3), characterized by three parameters: the inhibitory concentration 50, or IC₅₀ (also termed EC₅₀), a measure of the antimicrobial agent's potency; the maximal effect, E_max; and H, the slope of the curve, or Hill factor. In antimicrobial therapy, the relationship is often expressed as an inhibitory sigmoid E_max model, to take into account the control bacterial population without treatment (E_con) as a fourth parameter (Equation 48–1 and Figure 48–1), where E is effect as measured by microbial burden.

(Equation 48–1)

Figure 48–1.

Inhibitory sigmoid E_max curve. CFU, colony-forming units.

View Full Size||Download Slide (.ppt)

THE PHARMACOKINETIC BASISOF ANTIMICROBIAL THERAPY

Listen

Penetration of Antimicrobial Agents into Anatomic Compartments. In many infections, the pathogen causes disease not in the whole body, but in specific organs. Moreover, only specific pathological compartments may be infected within these organs. Antibiotics are often administered far away from these sites of infection. To be effective, each antibiotic has to get to where the pathogen is, to penetrate into the infected compartment. Therefore, in choosing an antimicrobial agent for therapy, a crucial consideration is whether the drug can penetrate to the site of infection.

For example, the antibiotic levofloxacin achieves skin tissue/plasma peak concentration ratio of 1.4, epithelial lining fluid to plasma ratio of 2.8, and urine to plasma ratios of 67 (Chow et al., 2002; Conte et al., 2006; Wagenlehner et al., 2006). In a clinical trial on patients treated with levofloxacin, the two most important factors in predicting successful clinical and microbiological outcomes in the patients were a sufficient maximal plasma concentration of the drug (C_Pmax), which needed to achieve a level of 12 times the minimum inhibitory concentration (MIC) (C_Pmax/MIC ≥12), and the site of infection. The failure rate of therapy was 0% in patients with urinary tract infections, 3% in patients with pulmonary infections, and 16% in patients with skin and soft tissue infections (Preston et al., 1998). Clearly, the poorer the penetration into the anatomical compartment, the higher the likelihood of failure.

The penetration of a drug into an anatomical compartment depends on the physical barriers that the molecule must traverse, the chemical properties of the drug, and the presence of multidrug transporters. The physical barriers are usually due to layers of epithelial and endothelial cells, and the type of junctions formed between these cells. Penetration across this physical barrier generally correlates with the octanol-water partition coefficient of the antimicrobial agent, a measure of the hydrophilicity or hydrophobicity of a chemical. Hydrophobic molecules get concentrated in the bi-lipid cell membrane bi-layer, whereas hydrophilic molecules tend to concentrate in the blood, the cytosol, and other aqueous compartments. Thus, the higher the octanol-water partition coefficient (P), the greater the likelihood that an antimicrobial agent will cross physical barriers erected by layers of cells. Conversely, the more charged a molecule is, and the larger it is, the poorer its penetration across membranes and other physical barriers (see Figure 2–3).

Another barrier is due to membrane transporters, which actively export drugs from the cellular or tissue compartment back into the blood (Chapter 5). A well-known example is the P-glycoprotein. Although the octanol-water partition coefficient would favor lipophilic molecules to transverse across cell barriers, P-glycoprotein exports structurally unrelated amphiphilic and lipophilic molecules of 3-4 kDa, reducing their effective penetration. Examples of antimicrobial agents that are P-glycoprotein substrates include HIV protease inhibitors, the antiparasitic agent ivermectin, the antibacterial agent telithromycin, and the antifungal agent itraconazole.

The central nervous system (CNS) is guarded by the blood-brain barrier. The movement of antibiotics across the blood-brain barrier is restricted by tight junctions that connect endothelial cells of cerebral micro-vessels to one another in the brain parenchyma, as well as by protein transporters (Miller et al., 2008). Antimicrobial agents that are polar at physiological pH generally penetrate poorly; some, such as penicillin G, are actively transported out of the cerebrospinal fluid (CSF) and achieve CSF concentrations of only 0.5-5% of that achieved in plasma. However, the integrity of the blood-brain barrier is diminished during active bacterial infection; tight junctions in cerebral capillaries open, leading to a marked increase in the penetration of even polar drugs (Quagliarello and Scheld, 1997). As the infection is eradicated and the inflammatory reaction subsides, penetration diminishes to normal. Because this may occur while viable microorganisms persist in the CSF, drug dosage should not be reduced as the patient improves.

Drug penetration into the eye is especially pertinent in the treatment of endophthalmitis (infection and inflammation of ocular cavity and surrounding tissue) and infections of the retina. There is generally poor penetration of drug from plasma to this compartment, so that the standard therapy is direct instillation of antibiotics into the ocular cavity. In fungal endophthalmitis, however, systemic administration of amphotericin B and triazole is recommended (Pappas et al., 2009) because these agents, especially the triazoles, have sufficient penetration ratios into the vitreous space.

In patients with pulmonary infections such as pneumonia, drugs must penetrate into the epithelial lining fluid, where the pathogens are found. Among antibacterial agents, many β-lactam antibiotics have poor epithelial lining fluid-to-plasma ratios (0.1-0.4:1); macrolides have ratios of 2-40:1, fluoroquinolones have ratios ≥1:1; pyrazinamide has ratios of ~20:1; isoniazid, >1:1; and linezolid, 2.4-4.2:1 (Kiem and Schentag, 2008).

Other important compartments requiring special drug penetration are endocardial vegetations and the biofilm formed by bacteria and fungi on prosthetic devices such as artificial heart valves, long-dwelling intravascular catheters, artificial hips, and devices for internal fixation of bone fractures. Bacterial and fungal biofilms are colonies of slowly growing cells that are enclosed within an exopolymer matrix. The exopolysaccharide is negatively charged, which restricts positively charged antibiotics from reaching their target. This physical barrier restricts the diffusion of antimicrobial molecules and sometimes binds them (Lewis, 2001). To be effective against infections in these compartments, antibiotics have to be able to penetrate the biofilm and endothelial barriers.

Pharmacokinetic Compartments. Once an antibiotic has penetrated to the site of infection, it may be subjected to processes of elimination and distribution that differ from those in the blood. These sites where the concentration-time profiles differ from each other are considered separate pharmacokinetic compartments, thus, the human body is viewed as multicompartmental. The concentration of antibiotic within each compartment is assumed to be homogeneous. If two compartments have similar concentration profiles, then they may be considered a single compartment. Antibiotic concentrations can be analyzed using any number of such compartments, with the best number of compartments chosen based on the least number of compartments that can adequately explain the findings. The model is also defined as open or not open; an open model is one in which the drug is eliminated out of the body from the compartment (e.g., kidneys). The order of the process must also be specified (Chapter 2): a first-order process is directly correlated to concentration of drug D, or [D]¹, as opposed to zero order, which is independent of [D] and reflects a process that is saturated at ambient levels of D (such as the elimination of ethanol; Chapter 23).

Suppose a patient has pneumonia with the pathogen in the lung epithelial lining fluid (ELF). The patient ingests an antibiotic that is absorbed via the GI tract (g) into blood or central compartment (compartment 1), as a first-order input. In this process, the transfer constant from the GI tract to central compartment is termed the absorption constant and is designated k_a. The antibiotic in the central compartment is then delivered to the lungs where it penetrates into the ELF (compartment 2). However, it also penetrates into other tissues of the body peripheral to the site of infection, termed the peripheral compartment (compartment 3). Thus, we have four compartments (including g), each with its own concentration-time profile, as shown in Figure 48–2. The penetration of drug from compartment 1 to 2 is based on the penetration factors discussed earlier and is defined by the transfer constant k₁₂. However, the drug also redistributes from compartment 2 back to 1, defined by transfer constant k₂₁. A similar process between the blood and peripheral tissues leads to transfer constants k₁₃ and k₃₁. The drug may also be lost from the body (i.e., open system) via the lungs and other peripheral tissues (e.g., kidneys or liver) at a rate proportional to the concentration.

Antibiotic concentrations within each compartment change with time. The changes in the amount of antibiotic in each compartment with time are described using standard differential equations (Gibaldi and Perrier, 1975). If X is the amount of antibiotic in a compartment, SCL the drug clearance, and V_c the volume of central compartment, then equations for absorption compartment (Equation 48–2), central compartment (Equation 48–3), site of infection or compartment 2 (Equation 48–4), and peripheral compartment (Equation 48–5) are as follows:

(Equation 48–2)

(Equation 48–3)

(Equation 48–4)

(Equation 48–5)

Such models have been used in conjunction with population pharmacokinetics to describe and model a plethora of antimicrobials used to treat bacteria, fungi, viruses, and parasites (Hope et al., 2007; Tarning et al., 2008; Wilkins et al., 2008; Zhou et al., 1999).

In recent years, semi-mechanistic models have related pathogen response to drug concentrations within these pharmacokinetic compartments in preclinical disease models and in patients (Gumbo et al., 2006; Jumbe et al., 2003; Neumann et al., 1998; Pukrittayakamee et al., 2003; Talal et al., 2006). Pathogens within one or more of these compartments are exposed to dynamic drug concentrations within the compartments, as described by Equations 48–2 to 48–5. A pathogen population (N) may be described as consisting of at least two subpopulations, drug-susceptible (N_S) and drug-resistant (N_R) organisms. The organisms are either killed, inhibited, or continued to grow in response to the antibiotic concentration. For some pathogens such as fungi and bacteria, it is assumed they are in logarithmic growth phase in the absence of drug, exhibiting an exponential density-limited growth rate so that they will reach a stationary phase at a maximal pathogen density, POPMAX. E is a logistic carrying function, so that

(Equation 48–6)

where N is the pathogen burden in a particular compartment. For replications rates of parasites and viruses, unique models may be specified. The drug will affect the growth rate either independent of kill or via killing the pathogen. The drug effect that is independent of kill is through a saturable Michaelis-Menten-type kinetic event (L).

(Equation 48–7)

where H and IC₅₀ are as described in Equation 48–1. Microbial kill is based on the concentrations within the compartment, and is modeled as a sigmoid E_max effect model M:

(Equation 48–8)

where K_kmax is the maximal kill rate. Because the microbial density (N) is in fact a balance between growth (maximal growth rate = K_gmax) and microbial kill, the change in pathogen density as a function of time is described by Equation 48–9:

(Equation 48–9)

The changes in drug-susceptible subpopulation of pathogen with time and the drug-resistant subpopulation are described by:

(Equation 48–10)

(Equation 48–11)

This can be made more complex and predictive adding parameters describing the effect of inoculum, delay in microbial effect, or splitting the drug-resistant population to smaller subpopulations based on molecular mechanism of resistance (Bulitta et al., 2009).

Figure 48–2.

Diagrammatic depiction of a multi-compartment model.

View Full Size||Download Slide (.ppt)

Population Pharmacokinetics and Variability in Drug Response. The framework for the models central to population pharmacokinetics evolved from a series of publications by Lewis Sheiner (Sheiner et al., 1977). What are population pharmacokinetics? Consider a simple example. When multiple patients are treated with the same dose of a drug, each patient will achieve pharmacokinetic parameters that differ from others. This is termed between-patient variability. Even when the same dose is administered to the same patient on two separate occasions, the patient may achieve a different concentration-time profile of the drug between the two occasions. This is termed inter-occasion or within-patient variability. The variability is reflected at the level of the compartmental pharmacokinetic parameters such as k_a, k₁₂, k₂₁, SCL, V_c, and so on. Even when a recommended dose is administered, the drug may fail to reach a therapeutic concentration in some patients. In other patients, the drug may reach high and toxic concentrations. Such variability could be due to factors that can be explained, such as genetic variability. In addition, anthropometric measures such as weight, height, and age also lead to variability. Furthermore, patients may have comorbid conditions such as renal and liver dysfunction, which may lead to variability. Drug interactions are an important source of variability with potentially dangerous consequences. These interactions usually occur when one drug inhibits or induces uptake or clearance mechanisms affecting another drug (i.e., modulation of the activities of xenobiotic metabolic enzymes, drug transporters, and excretion mechanisms; Chapters 5 and 6).

Even when such factors are accounted for, there remains residual variability due to computational noise, assay variability, and unexplainable factors. The common practice of using an "average" value of data or "naive pooling," has implications of smoothing out data and failing to recognize subgroups of patients at risk for therapeutic failure or increased toxicity of antibiotic. Thus the variability itself, the extent of such variability, and any factors that may explain part of the variability of drug response are more important to understand than measures of central tendency and dispersion for a hypothetical average patient. Such variability reflects the fact that pathogens within patients given the same dose of antibiotic will be exposed to different antibiotic concentrations from patient to patient, leading to effective kill in some and resistance emergence in others. Knowledge of covariates associated with pharmacokinetic variability leads to better dose adjustments, or switching therapy from one antibiotic to another, or changing concomitant medications.

IMPACT OF SUSCEPTIBILITY TESTING ON SUCCESS OF ANTIMICROBIAL AGENTS

Listen

The microbiology laboratory plays a central role in the decision to choose a particular antimicrobial agent over others. First, identification and isolation of the culprit organism takes place when the patients' specimens are sent to the microbiology laboratory. Once the microbial species causing the disease has been identified, a rational choice of the class of antibiotics likely to work in the patient can be made. The microbiology laboratory then plays a second role, which is to perform susceptibility testing. This step is crucial in narrowing down the list of possible antimicrobials that could be used.

Millions of individuals across the globe get infected by many different isolates of the same species of pathogen. Evolutionary processes cause each isolate to be slightly different from the next, so that each will have a unique susceptibility to antimicrobial agents. As the microorganisms divide within the patient, they may undergo further evolution between the time of infection and when the disease is diagnosed. As an example, a relatively narrow range of variants are transmitted when patients become infected with HIV. However, HIV replication has poor fidelity and has replication rates that result in up to 10¹⁰ viral particles each day. Moreover, viral recombination occurs commonly. Over many months, under immune pressure, numerous variants arise. The emergence of a quasi species that harbors a mutation associated with drug resistance to at least one drug is high, based on probability factors in the context of high replication rates and the massive numbers of viral particles. Therefore, we expect that there will be a wide distribution of concentrations of antimicrobial agents that can kill pathogens. Often, this distribution is Gaussian, with a skew that depends on where the patient lives. These factors will affect the shape of the inhibitory sigmoid E_max model curve described by Equation 48–1.

With changes in susceptibility, the sigmoid E_max curve shifts in one of two basic ways. The first is a shift to the right, an increase in IC₅₀, as shown in Figure 48–3A. This means that much higher concentrations than before are now needed to show specific effect. Susceptibility tests for bacteria, fungi, parasites, and viruses have been developed to determine whether these shifts have occurred at a sufficient magnitude to warrant higher doses of drug to achieve particular effect. The change in IC₅₀ may become so large that it is not possible to overcome the concentration deficit by increasing the antimicrobial dose without causing toxicity to the patient. At that stage, the organism is now "resistant" to the particular antibiotic. A second possible change in the curve is decrease in E_max (Figure 48–3B), such that increasing the dose of the antimicrobial agent beyond a certain point will achieve no further effect; i.e., changes in the microbe are such that eradication of the microbe by the particular drug can never be achieved. This occurs because the available target proteins have been reduced or the microbe has developed an alternative pathway to overcome the biochemical inhibition. As an example, maraviroc is an allosteric, noncompetitive antagonist that binds to the CCR5 receptor of patient's CD4 cells to deny HIV entry into the cell. Viral resistance occurs by a mechanism that involves HIV adapting to use of the maraviroc-bound CCR5, which results in decrease of E_max in phenotypic susceptibility assays (Hirsch et al., 2008).

Figure 48–3.

Changes in sigmoid E_max model with increases in drug resistance. Increase in resistance may show changes in IC₅₀ (panel A: the IC₅₀ increases from 70 (orange line) to 100 (green line), to 140 (blue line)) or decrease in E_max (panel B: efficacy decreases from full response (orange line) to 70% (green line))

View Full Size||Download Slide (.ppt)

Bacteria. For bacteria, dilution tests employ antibiotics in serially diluted concentrations on solid agar or in broth medium that contains a culture of the test microorganism. The lowest concentration of the agent that prevents visible growth after 18-24 hours of incubation is known as the minimum inhibitory concentration (MIC).

Automated systems also use a broth-dilution method. The optical density of a broth culture of the clinical isolate incubated in the presence of drug is determined. If the density of the culture exceeds a threshold optical density, then growth has occurred at that concentration of drug. The MIC is the concentration at which the optical density remains below the threshold.

The disk-diffusion technique provides only qualitative or semi-quantitative information on antimicrobial susceptibility. The test is performed by applying commercially available filter-paper disks impregnated with a specific amount of the drug onto an agar surface, over which a culture of the microorganism has been streaked. After 18-24 hours of incubation, the size of the clear zone of inhibition around the disk is measured. The diameter of the zone depends on the activity of the drug against the test strain. Standardized values for zone sizes for each bacterial species and antibiotic permit classification of the clinical isolate as either resistant or susceptible. A variant of the disk diffusion tests is the Epsilometer test, or E-test. A rectangular test strip impregnated with changing concentrations of antimicrobial agent, usually across 15 dilutions, is placed on an agar plate that has a heavy inoculum of test organism. The drug concentrations are printed along this long test strip. The cultures are then incubated under favorable conditions for 24 hours, 48 hours, or 5 days, depending on the test organism. There is no growth with higher concentrations and heavy microbial growth where there is lower drug concentration, so that a clear elliptical zone is formed that bisects the test strip at the MIC. This test has the virtue of determining an actual MIC value, rather than the dichotomous categorization of "susceptible" or "resistant." There are test strips for hundreds of antibacterial agents as well as for some antifungal agents active against Candida species.

Sometimes, a biochemical or immunological reaction that leads to a color change is used to detect susceptibility. As an example, methicillin resistance in Staphylococcus aureus (methicillin-resistant S. aureus, or MRSA) occurs because of acquisition of a low-affinity penicillin-binding protein 2' (2a), or PBP2' (PBP2a). The PBP2' is extracted from isolates of the MRSA, and supernatant mixed with a latex reagent with monoclonal antibody against PBP2'. Visible clumping within 3 minutes indicates presence of PBP2', indicating MRSA.

Fungi. For fungi that are yeasts (i.e., Candida), susceptibility testing methods are similar to those used for bacteria. However, the definitions of MIC differ based on drug and the type of yeast, so there are cutoff points of 50% decrease in turbidity compared to controls at 24 hours, or 80% at 48 hours, or total clearance of the turbidity. Susceptibility tests and MICs for triazoles have been extensively shown to correlate with clinical outcomes.

Standardized tests for echinocandin antifungals and amphotericin B-based compounds are available. However, the correlation of these tests with clinical outcomes is still weak. Susceptibility tests for molds have also been developed, but the clinical correlations are still being examined. Different terminology from MICs has been adopted when evaluating echinocandins against molds because the fungal burden can not be measured by counting discrete cells in molds, given that hyphae will break up into unpredictable numbers of discrete fungi when under antifungal pressure. Furthermore, echinocandins often do not completely inhibit mold growth, but instead cause damage reflected by morphological changes in hyphae. Thus, the minimum effective concentration (MEC) for echinocandins is the lowest drug concentration at which short, stubby, and highly branched hyphae are observed on microscopic examination.

Viruses. In HIV phenotypic assays, the patient's HIV-RNA is extracted from plasma, and genes for targets of antiretroviral drugs such as reverse transcriptase and protease are amplified. The genes are then inserted into a standard HIV vector that lacks analogous gene sequences to produce a recombinant virus, which is co-incubated with drug of interest in a mammalian cell viability assay (Hanna and D'Aquila, 2001; Hertogs et al., 1998; Petropoulos et al., 2000). Growth is compared to a standardized wild-type control virus. For HIV reverse transcriptase, e.g., <4-fold increase in IC₅₀ is defined as "sensitive," 4- to 10-fold increase in IC₅₀ is "intermediate," and >10-fold increase is "resistant" (Hanna and D'Aquila, 2001). Further use has been made of the viral IC₅₀ to establish the inhibitory quotient (IQ). The IQ is the ratio of plasma concentration of antiviral drug to the IC₅₀. The phenotypic IQ is the ratio of plasma trough concentration to the IC₅₀.

Genotypic tests are now a standard part of HIV care in many parts of the world, although the tests are rarely used in clinical practice in resource-poor settings. The simplest tests measure presence of mutations associated with loss of function of a drug. This loss of function means that the organism is "resistant" to drug and the drug should not be used to treat that patient. The absence of mutation means the drug has an increased likelihood of working against the pathogen. There are some situations with antiretroviral therapy, however, when presence of such mutations may paradoxically lead to choice of the drug to be part of combination therapy. A famous example is that of the reverse transcriptase inhibitors zidovudine (AZT) and lamivudine (3TC) in the treatment of HIV-1, in which an M184V mutation that causes resistance to 3TC increases viral sensitivity to AZT up to 10-fold. Therefore, although presence of a mutation in a genotype assay may strongly argue for exclusion of some drugs in a regimen, the interpretation can be complex. Indeed, this is one of the limitations of this method in that the true meaning of a mutation may be unclear and interpreted differently by healthcare practitioners.

Parasites. Susceptibility testing for parasites, especially malaria, has been performed in the laboratory. The tests are similar to the broth tests for bacteria, fungi, and viruses. Plasmodium species in the patient's blood are cultured ex vivo in the presence of different dilutions of antimalarial drug (Stepniewska et al., 2007). A sigmoid E_max curve for effect versus drug concentration is used to identify IC₅₀ and E_max. These susceptibility tests are usually field tests at sentinel sites that are used to determine if there is drug resistance in a particular area or not, and they are not used as a routine test for clinical decision making for individual patients. In general, susceptibility tests for parasitic infections are not standardized. These tests are primarily used in the research setting and not for individualization of therapy.

BASIS FOR SELECTION OF DOSE AND DOSING SCHEDULE

Listen

Even though susceptibility testing is central to decision making, it does not completely predict patient response. Unlike in susceptibility tests in the laboratory where drug concentrations are static by design, microorganisms in patients are exposed to dynamic concentrations of drug. In addition, the antibiotics are prescribed at a certain schedule (e.g., three times a day) so that there is a periodicity in the fluctuations of drug at the site of infection. Therefore, the microbe is exposed to a particular shape of the concentration-time curve. Harry Eagle performed studies on penicillin and discovered that the shape of the concentration-time profile was an important determinant of the efficacy of the antibiotic (Eagle et al., 1950). This important observation was forgotten until William Craig and colleagues rediscovered it and performed systematic studies on several classes of antibiotics (Craig, 1998, 2007; Vogelman et al., 1988), initiating the era of antimicrobial pharmacokinetics-pharmacodynamics (PK/PD) (Ambrose et al., 2007; Craig, 1998).

The first basic lesson is based on susceptibility (MIC or EC₉₀) of the organism to the antimicrobial agent. The response of the microbe to a fixed dose of antibiotic differs based on susceptibility of the organism. As an example, vancomycin resistance is said to be present when MIC >2.0 mg/L. When patients with MRSA infection were treated with vancomycin, the success rate was 61% in patients infected with isolates that had an MIC of 0.5 mg/L, 28% for MIC of 1.0 mg/L, and only 11% for MIC of 2.0 mg/L (Moise-Broder et al., 2004). As expected, outcomes were poorer with increasing MIC. This is not a surprise because the IC₅₀ shifts to the right with decrease in susceptibility (Figure 48–3A). The important point is that in determining therapeutic outcomes, it is important to index drug exposure to MIC.

Second, dose itself is a poor measure of drug exposure, given between-patient and within-patient pharmacokinetic variability. Rather, actual drug concentration achieved at site of infection is the important measure. The shape of the relationship between non-protein-bound antibiotic concentration (exposure) versus microbial kill is the inhibitory sigmoid E_max curve of Figure 48–1 (Craig, 2007; Craig and Kunin, 1976). Maximal kill is actually on an asymptote, so that non-protein-bound antimicrobial exposures associated with 80-90% of E_max are termed "optimal" concentrations. The optimal dose of the antibiotic for a patient is the dose that achieves IC₈₀ to IC₉₀ exposures at the site of infection. This exposure can often be easily identified in preclinical models and directly applied to patient populations, provided interspecies differences in protein binding and pharmacokinetic variability are taken into account.

Third, optimal microbial kill by the antibiotic may be best achieved by maximizing certain shapes of the concentration-time curve. Certain dose schedules maximize antimicrobial effect. As an example, consider an antibiotic with a serum t_1/2 of 3 hours that is being used to treat a bloodstream infection by a pathogen with an MIC of 0.5 mg/L, administered with a dosing interval of 24 hours (that is, a once-daily schedule). Figure 48–4A depicts the concentration-time curve of the antibiotic, with definitions of peak concentration (C_Pmax), area under the curve (AUC), and the fraction of the dosing interval for which the drug concentration remains above the MIC (T > MIC), as shown. The AUC is a measure of the total concentration of drug and is calculated by taking an integral between two time points, 0-24 hours (AUC_0-24) in this case. Now, if one were to change the dosing schedule of the same antibiotic amount by splitting it into three equal doses administered at 0, 8, and 16 hours, the shape of the concentration-time curves changes to that shown in Figure 48–4B. Because the same cumulative dose has been given for the dosing interval of 24 hours, the AUC_0-24 will be similar whether it was given once a day or three times a day. For the same pathogen, therefore, the change in dose schedule does not change the AUC_0-24/MIC (or AUC_0-24/EC₉₀). However, the C_Pmax will decrease by a third when the total dose is split into thirds and administered more frequently (Figure 48–4B). Thus, when a dose is fractionated and administered more frequently, the C_Pmax/MIC ratio decreases. In contrast, the time that the drug concentration persists above MIC (T > MIC) will increase with the more frequent dosing schedule, despite the same cumulative dose being administered. Which of the three indices (AUC/MIC, or C_Pmax/MIC or T>MIC) is the most important to the outcome being assessed (i.e., microbial kill)? A common simple approach to the answer is to determine which of these patterns best approximates a perfect inhibitory sigmoid E_max curve (based on various statistical assessments of goodness of fit).

Figure 48–4.

Effect of different dose schedules on shape of concentration-time curve. The total AUC for the fractionated dose in curve B is determined by adding AUC_0-8h, AUC_8-16h and AUC _16-24h, which adds up to the same AUC_0-24h in curve A. The time that the drug concentration exceeds MIC in curve B is also determined by adding up T₁>MIC, T₂>MIC, and T₃>MIC, which results in a fraction greater than that for curve A.

View Full Size||Download Slide (.ppt)

Some classes of antimicrobial agents kill best when concentration persists above MIC for longer durations of the dosing interval while kill is decreased by reducing the time above MIC. Indeed, increasing the drug concentration beyond 4-6 times the MIC does not increase microbial kill. Two good examples are β-lactam antibacterials (e.g., penicillin) and the antifungal agent 5-flourocytosine (5-FC) (Ambrose et al., 2007; Andes and van Ogtrop, 2000). In fact, there are usually good biochemical explanations for this pattern for the drugs. The clinical implication, however, is that a drug optimized by T > MIC should be dosed more frequently, or if possible should have its t_1/2 prolonged by other drugs, so that drug concentrations persist above MIC (or EC₉₅) as long as possible. Thus the effectiveness of penicillin is enhanced when it is given as a continuous infusion. HIV protease inhibitors are often "boosted" with ritonavir. This "boosting" inhibits the metabolism of the protease inhibitors by CYPs 3A4 and 2D6, thereby prolonging time above EC₉₅.

However, the peak concentration is what matters for some antimicrobial agents. Persistence of concentration above MIC has less relevance for these drugs, meaning that these drugs can be dosed more intermittently. Aminoglycosides are a prime example of this class as they used to be given three times a day but are highly effective when given once a day. These C_Pmax/MIC-linked drugs are often able to be administered less frequently due to their long duration of post-antibiotic effect (PAE). In other words, effect continues long after antibiotic concentrations decline below the MIC. An example of such a drug is rifampin (Gumbo et al., 2007a). The entry of rifampin into Mycobacterium tuberculosis increases with increased concentration in the bacillus microenvironment, likely because of a saturable transport process. Once inside the bacteria, the drug's macrocyclic ring binds the β-subunit of DNA-dependent RNA polymerase (rpoB) to form a very stable drug-enzyme complex within 10 minutes, a process not enhanced by longer incubation of drug and enzyme (Harshey and Ramakrishnan, 1976; Wehrli et al., 1968). The PAE of the rifampin is long and concentration dependent (Gumbo et al., 2007a). This means that administering combined doses on a more intermittent basis (i.e., once a day) will maximize effect for these drugs. Toxicity generated by intermittent dosing may preclude such use for some drugs such as rifampin, whereas for others (i.e., aminoglycosides) such infrequent dosing may actually decrease toxicity.

There is a third group of drugs for which dosing schedule has no effect on efficacy, but where it is the cumulative dose that matters. Thus, it is more the total concentration (AUC) to MIC ratio that matters and not the time that concentration persists above a certain threshold. Antibacterial agents such as daptomycin fall into this class (Louie et al., 2001). These agents also have a good PAE. The AUC/IC₅₀ explains why the nucleoside analogue reverse transcriptase inhibitors tenofovir and emtricitabine have been combined into one pill, administered once a day for the treatment of AIDS.

The shape of concentration-time curve that optimizes resistance suppression is often different from that which optimizes microbial kill. In many instances, the drug exposure associated with resistance suppression is much higher than for optimal kill. It should actually be this higher exposure that should be achieved by each dose in patients for optimal effect, rather than the EC₈₀ as discussed earlier. However, this is often precluded by increased drug toxicity when doses are increased. Second, although the relationship between kill and exposure is based on the inhibitory sigmoid E_max model, experimental work with preclinical models demonstrates that this model does not apply to resistance suppression (Gumbo, 2007b; Tam et al., 2007).

The optimal dose should be designed to achieve a high probability of exceeding the EC₈₀ microbial PK/PD index, or index associated with suppression of resistance, given the population pharmacokinetic variability and the MIC distribution of clinical microbe isolates. The population pharmacokinetic variability enables integration of pharmacogenetics, anthropometric measures, and residual variability into the decision to choose optimal dose (Gumbo, 2008). Once that has been achieved, the dose schedule is chosen according to whether efficacy is driven by AUC/MIC (or AUC/EC₉₅), C_Pmax/MIC, or T > MIC. Duration of therapy is then chosen, based on best available evidence.

TYPES AND GOALS OFANTIMICROBIAL THERAPY

Listen

When an infection occurs, the numbers of microorganisms are often small in the beginning. Pathogen burden eventually increases with replication of the organism. Sometimes, the immune system manages to eliminate the infection before it causes any further damage. In other instances, the organism is not entirely eliminated but hides inside the patient's own cells and become dormant, only to reactivate when immune function is compromised in the future. In some patients, the organism may overcome immune defenses and then cause disease. In a subset of such patients, the disease is self-limited. For example, many viral upper respiratory infections are self-limited and should not be treated with antimicrobial agents. Other diseases, however, require antimicrobial therapy. In these patients, therapy should be discontinued after resolution of disease. In special cases where the immune or anatomical defect that led to the infection is still present, suppressive or "maintenance" therapy may be required. A useful way to organize the types and goals of antimicrobial therapy is to consider where along the disease progression timetable therapy is initiated (Figure 48–5); therapy can be prophylactic, preemptive, empirical, definitive, or suppressive.

Figure 48–5.

Aantimicrobial therapy-disease progression timeline. Stages of disease progression are below the horizontal arrow; categories of antimicrobial therapy are above the arrow.

View Full Size||Download Slide (.ppt)

Prophylactic Therapy. Prophylaxis involves treating patients who are not yet infected or have not yet developed disease. The goal of prophylaxis is to prevent infection in some patients or to prevent development of a potentially dangerous disease in those who already have evidence of infection. Ideally, a single, effective, nontoxic drug is successful in preventing infection by a specific microorganism or eradicating an early infection. The main principle behind prophylaxis is targeted therapy. However, prophylaxis to prevent colonization or infection by any or all microorganisms present in the environment of a patient often fails.

Prophylaxis is used in immunosuppressed patients such as those with HIV-AIDS or are post-transplantation and on anti-rejection medications. The efficacy of prophylaxis in these patients is based on excellent evidence (Anonymous, 2000; DHHS Panel on Guidelines for the Prevention and Treatment of Opportunistic Infections in HIV-Infected Adults and Adolescents, 2008). In these groups of patients, specific antiparasitic, antibacterial, antiviral, and antifungal therapy is administered based on the well-defined pattern of pathogens that are major causes of morbidity during immunosuppression. A risk-benefit analysis is used to determine choice and duration of prophylaxis. Prophylaxis of opportunistic infections in patients with AIDS is started when the CD4 count falls below 200 cells per mm³. In post-transplant patients, prophylaxis depends on time since the transplant procedure, which is related to intensity of use and type of immunosuppressive therapy. Prophylaxis should be discontinued in patients who are doing well at certain time points, such as 1 year post-transplant. In AIDS patients, prophylaxis is discontinued when the CD4 count climbs above 200 cells/mm³. Infections for which prophylaxis is given include Pneumocystis jiroveci, Mycobacterium avium-intracellulare, Toxoplasma gondii, Candida species, Aspergillus species, Cytomegalovirus, and other Herpesviridae. In general, lower doses of prophylactic agent are given compared to when the same drug is used for treatment.

Chemoprophylaxis is also used to prevent wound infections after various surgical procedures. Wound infection results when a critical number of bacteria are present in the wound at the time of closure. Antimicrobial agents directed against the invading microorganisms may reduce the number of viable bacteria below the critical level and thus prevent infection. Several factors are important for the effective and judicious use of antibiotics for surgical prophylaxis. First, antimicrobial activity must be present at the wound site at the time of its closure. Thus, infusion of the first antimicrobial dose should begin within 60 minutes before surgical incision and should be discontinued within 24 hours of the end of surgery (Bratzler and Houck, 2004). Second, the antibiotic must be active against the most likely contaminating microorganisms for that type of surgery. A number of studies indicate that chemoprophylaxis can be justified in dirty or contaminated surgical procedures (e.g., resection of the colon), where the incidence of wound infections is high. These include <10% of all surgical procedures. In clean surgical procedures, which account for ~75% of the total, the expected incidence of wound infection is <5%, and antibiotics should not be used routinely. When the surgery involves insertion of a prosthetic implant (e.g., prosthetic valve, vascular graft, prosthetic joint), cardiac surgery, or neurosurgical procedures, the complications of infection are so drastic that most authorities currently agree to chemoprophylaxis for these indications.

Patients at the highest risk for infective endocarditis for which prophylaxis is recommended fall into four groups (Wilson et al., 2007):

those with a prosthetic material used for heart valve repair or replacement
previous infective endocarditis
congenital heart disease such as unrepaired cyanotic heart disease, or within 6 months of repair of the heart disease with prosthetic material, or those with residual defects adjacent to prosthetic material
postcardiac transplant patients with heart valve defects

Chemoprophylaxis is reasonable in these patients when undergoing dental procedures if there is manipulation of gingival tissue or periapical region of teeth, or perforation of oral mucosa, but not for other dental procedures. Recommended therapy is a single dose of oral amoxicillin 30 minutes to 1 hour before the procedure or intravenous ampicillin or ceftriaxone in those unable to take oral medication. A macrolide or clindamycin may be administered for patients who are allergic to β-lactam agents. Therapy may be administered no more than 2 hours after the procedure for patients who failed to receive the prophylaxis prior to the procedure (Wilson et al., 2007). Prophylaxis is also reasonable for procedures that will involve infected skin and soft tissues as well as infected respiratory tract, but not in routine genitourinary and GI tract procedures. If the organism causing the infection is known, then the prophylactic antibiotic for patients undergoing these procedures should be tailored toward that particular organism.

Prophylaxis may be used to protect healthy persons from acquisition of or invasion by specific microorganisms to which they are exposed. This is termed post-exposure prophylaxis. Successful examples of this practice include rifampin administration to prevent meningococcal meningitis in people who are in close contact with a case, prevention of gonorrhea or syphilis after contact with an infected person, and macrolides after contact with confirmed cases of pertussis. Post-exposure prophylaxis is recommended in those patients inadvertently exposed to HIV infection. Currently, combination therapy from one of the classes of antiretrovirals, nucleoside reverse transcriptase inhibitors, nucleotide reverse transcriptase inhibitors, non-nucleoside reverse transcriptase inhibitors, protease inhibitors, and fusion inhibitors is administered for 4 weeks (Panlilio et al., 2005). For influenza, the neuraminidase inhibitor oseltamivir has been recommended for prevention of influenza A and B in healthy adults and children with close contact of laboratory-confirmed cases (Hayden and Pavia, 2006).

Mother-to-child transmission of HIV and syphilis are important public health problems for which specific chemotherapeutic regimens have been devised, based on locality. Anti-retroviral therapy is administered for HIV prophylaxis during the pregnancy and peripartum periods. Prophylactic therapy for syphilis during pregnancy is effective in reducing neonatal death and infant neurological, auditory, and bone malformations.

Pre-emptive therapy. Pre-emptive therapy is used as a substitute for universal prophylaxis and as early targeted therapy in high-risk patients who already have a laboratory or other test indicating that an asymptomatic patient has become infected. The principle is that delivery of therapy prior to development of symptoms (presymptomatic) aborts impending disease, and the therapy is for a short and defined duration (Singh, 2001). This has been applied in the clinic to therapy for cytomegalovirus (CMV) after both hematopoietic stem cell transplants and after solid organ transplantation. It is unclear whether this method is superior to keeping all at-risk patients on ganciclovir. Recent evidence in liver transplant patients suggest this approach may be as efficacious as universal prophylaxis while using far less antiviral medications (Gerna et al., 2008; Singh et al., 2008).

Empirical Therapy in the Symptomatic Patient. Once a patient is symptomatic, should the patient be treated immediately? The first consideration in selecting an antimicrobial is to determine if the drug is indicated. The reflex action to associate fever with treatable infections and prescribe antimicrobial therapy without further evaluation is irrational and potentially dangerous.

The diagnosis may be masked if therapy is started and appropriate cultures are not obtained. Antimicrobial agents are potentially toxic and may promote selection of resistant microorganisms. For some diseases, the cost of waiting a few days is low. These patients can wait for microbiological evidence of infection without empirical treatment. In a second group of patients, the risks of waiting are high, based either on the patient's immune status or other known risk factors for poor outcome with therapy delay. Initiation of optimal empirical antimicrobial therapy should rely on the clinical presentation, which may suggest the specific microorganism, and knowledge of the microorganisms most likely to cause specific infections in a given host. In addition, simple and rapid laboratory techniques are available for the examination of infected tissues.

The most valuable and time-tested method for immediate identification of bacteria is examination of the infected secretion or body fluid with Gram stain. In malaria-endemic areas, or in travelers returning from such an area, a simple thick and thin blood smear may mean the difference between a patient's receiving appropriate therapy and surviving or death while on wrong therapy for presumed bacterial infection. Such tests help to narrow the list of potential pathogens and permit more rational selection of initial antibiotic therapy. Similarly, neutropenic patients with fever have high risks of mortality, and, when febrile, they are presumed to have either a bacterial or fungal infection; thus a broad-spectrum combination of antibacterial and antifungal agents that cover common infections encountered in granulocytopenic patients are given. Performance of cultures is still mandatory with a view to modify antimicrobial therapy with culture results.

Definitive Therapy with Known Pathogen. Once a pathogen has been isolated and susceptibilities results are available, therapy should be streamlined to a narrow targeted antibiotic. Monotherapy is preferred to decrease the risk of antimicrobial toxicity and selection of antimicrobial-resistant pathogens. Proper antimicrobial doses and dose schedules are crucial to maximizing efficacy and minimizing toxicity. In addition, the duration of therapy should be as short as is necessary. The practice of keeping a patient indefinitely on antimicrobial therapy without a particular reason is discouraged. In fact, both experimental and clinical evidence have shown that unnecessarily prolonged therapies lead to the emergence of resistance.

Combination therapy is an exception, rather than a rule. Once a pathogen has been isolated, there should be no reason to use multiple antibiotics, except when evidence overwhelmingly suggests otherwise. Using two antimicrobial agents where one is required leads to increased toxicity and unnecessary damage to the patient's otherwise protective fungal and bacterial flora. This is true, even when intuition suggests use of two agents. As an example, Cosgrove and colleagues recently demonstrated increased nephrotoxicity of low-dose gentamicin administered only for 4 days as "synergistic therapy" with vancomycin or an antistaphylococcal penicillin for S. aureus bacteremia and endocarditis, without improving efficacy (Cosgrove et al., 2009).

However, there are special circumstances where evidence is unequivocal in favor of combination therapy. The principles behind such antimicrobial use include:

preventing resistance to monotherapy
accelerating the rapidity of microbial kill
enhancing therapeutic efficacy by use of synergistic interactions or enhancing kill by a drug based on a mutation generated by resistance to another drug
paradoxically, reducing toxicity (i.e., when full efficacy of a standard antibacterial agent can only be achieved at doses that are toxic to the patient, and a second drug is co-administered to exert additive effects)

Clinical situations for which combination therapy is advised are discussed in the relevant chapters but include antiretroviral therapy for AIDS, antiviral therapy for hepatitis B and C, the treatment of tuberculosis, Mycobacterium avium-intracellulare and leprosy, fixed-dose combinations of antimalarial drugs, the treatment of Cryptococcus neoformans with flucytosine and amphotericin B, during empirical therapy for patients with febrile neutropenia, and advanced AIDS with fever. The combination of a sulfonamide and an inhibitor of dihydrofolate reductase, such as trimethoprim, is synergistic owing to the blocking of sequential steps in microbial folate synthesis. A fixed combination of sulfamethoxazole and trimethoprim is active against organisms that may be resistant to sulfonamides alone, is effective for many infections, and is rarely given as its separate components.

Post-treatment suppressive therapy. In some patients, after the initial disease is controlled by the antimicrobial agent, therapy is continued at a lower dose. This is because in these patients the infection is not completely eradicated and the immunological or anatomical defect that led to the original infection is still present. This is common in AIDS patients and post-transplant patients, for example. The goal is more as secondary prophylaxis. Nevertheless, risks of toxicity from long durations of the therapy are still real. In this group of patients, the suppressive therapy is eventually discontinued if the patient's immune system improves.

MECHANISMS OF RESISTANCETO ANTIMICROBIAL AGENTS

Listen

Antimicrobial agents were viewed as miracle cures when first introduced into clinical practice. However, it became evident rather soon after the discovery of penicillin that resistance developed quickly, terminating the miracle. This serious development is ever present with each new antimicrobial agent and threatens the end of the antimicrobial era. Today, every major class of antibiotic is associated with the emergence of significant resistance. Two major factors are associated with emergence of antibiotic resistance: evolution and clinical/ environmental practices. A species that is subjected to pressure, chemical or otherwise, that threatens its extinction often evolves mechanisms to survive under that stress. Pathogens will evolve to develop resistance to the chemical warfare to which we subject them. This evolution is mostly aided by poor therapeutic practices by healthcare workers, as well as indiscriminant use of antibiotics for agricultural and animal husbandry purposes. Poor clinical practices that fail to incorporate the pharmacological properties of antimicrobials amplify the speed of development of drug resistance.

Antimicrobial resistance can develop at any one or more of steps in the processes by which a drug reaches and combines with its target. Thus, resistance development may develop due to:

reduced entry of antibiotic into pathogen
enhanced export of antibiotic by efflux pumps
release of microbial enzymes that destroy theantibiotic
alteration of microbial proteins that transform pro-drugs to the effective moieties
alteration of target proteins
development of alternative pathways to those inhibited by the antibiotic

Mechanisms by which such resistance develops can include acquisition of genetic elements that code for the resistant mechanism, mutations that develop under antibiotic pressure, or constitutive induction.

Resistance Due to Reduced Entry of Drug into Pathogen. The outer membrane of gram-negative bacteria is a permeable barrier that excludes large polar molecules from entering the cell. Small polar molecules, including many antibiotics, enter the cell through protein channels called porins. Absence of, mutation in, or loss of a favored porin channel can slow the rate of drug entry into a cell or prevent entry altogether, effectively reducing drug concentration at the target site. If the target is intracellular and the drug requires active transport across the cell membrane, a mutation or phenotypic change that slows or abolishes this transport mechanism can confer resistance. As an example, Trypanosoma brucei is treated with suramin and pentamidine during early stages, but with melarsoprol and eflornithine when CNS disease (sleeping sickness) is present. Melarsoprol is actively taken up by trypanosome P2 protein transporter. When the parasite either lacks the P2 transporter, or has a mutant form, resistance to melarsoprol and cross resistance to pentamidine occur due to reduced uptake (Ouellette, 2001).

Resistance Due to Drug Efflux. Microorganisms can overexpress efflux pumps and then expel antibiotics to which the microbes would otherwise be susceptible. There are five major systems of efflux pumps that are relevant to antimicrobial agents:

the multidrug and toxic compound extruder (MATE)
the major facilitator superfamily (MFS) transporters
the small multidrug resistance (SMR) system
the resistance nodulation division (RND) exporters
ATP binding cassette (ABC) transporters

Efflux pumps are a prominent mechanism of resistance for parasites, bacteria, and fungi. One of the tragic consequences of resistance emergence has been the development of drug resistance by Plasmodium falciparum. Drug resistance to most antimalarial drugs, specifically chloroquine, quinine, mefloquine, halofantrine, lumefantrine, and the artemether-lumefantrine combination is mediated by an ABC transporter encoded by Plasmodium falciparum multidrug resistance gene 1 (Pfmdr1) (Happi et al., 2009). Point mutations in the Pfmdr1 gene lead to drug resistance and failure of chemotherapy. Drug efflux sometimes works in tandem with chromosomal resistance, as is seen in Streptococcus pneumoniae, and perhaps, Myobacterium tuberculosis. In these situations, induction of efflux pumps occurs early, which increases the MIC only modestly. However, this MIC increase is enough to allow further microbial replication and an increased mutation frequency, which enable the development of resistance via more robust chromosomal mutations (Gumbo et al., 2007b; Jumbe et al., 2006).

Resistance Due to Destruction of Antibiotic. Drug inactivation is a common mechanism of drug resistance. Bacterial resistance to aminoglycosides and to β-lactam antibiotics usually is due to production of an aminoglycoside-modifying enzyme or β-lactamase, respectively.

Resistance Due to Reduced Affinity of Drug to Altered Target Structure. A common consequence of either single point or multiple point mutations is change in amino acid composition and conformation of target protein. This change leads to a reduced affinity of drug for its target, or of a prodrug for the enzyme that converts the prodrug to active drug. Such alterations may be due to mutation of the natural target (e.g., fluoroquinolone resistance), target modification (e.g., ribosomal protection type of resistance to macrolides and tetracyclines), or acquisition of a resistant form of the native, susceptible target (e.g., staphylococcal methicillin resistance caused by production of a low-affinity penicillin-binding protein) (Hooper, 2002; Lim and Strynadka, 2002; Nakajima, 1999). Similarly, in HIV resistance mutations associated with reduced affinity are encountered in protease inhibitors, integrase inhibitors, fusion inhibitors, and non-nucleoside reverse transcriptase inhibitors (Nijhuis et al., 2009). Similarly, benzimidazoles are used against myriad worms and protozoa and work by binding to the parasite's tubulin; point mutations in the β-tubulin gene lead to modification of the tubulin and drug resistance (Ouellette, 2001).

Incorporation of Drug. An uncommon situation occurs when an organism not only becomes resistant to an antimicrobial agent but subsequently starts requiring it for growth. Enterococcus, which easily develops vancomycin resistance, can, after prolonged exposure to the antibiotic, develop vancomycin-requiring strains.

Resistance Due to Enhanced Excision of Incorporated Drug. Nucleoside reverse transcriptase inhibitors such as zidovudine are 2′-deoxyribonucleoside analogs that are converted to their 5′-triphosphate form and compete with natural nucleotides. These drugs are incorporated into the viral DNA chain and cause chain termination. When resistance emerges via mutations at a variety of points in the reverse transcriptase gene, phosphorolytic excision of the incorporated chain-terminating nucleoside analog is enhanced (Arion et al., 1998).

Hetero-resistance and Viral Quasi Species. Hetero-resistance is said to be present when a subset of the total microbial population is resistant, despite the total population being considered susceptible on testing (Falagas et al., 2008; Rinder, 2001). In a way, this should not be a surprise given that chromosomal mutations are a stochastic process and there is a baseline mutation rate for each gene. Therefore, a subclone that has alterations in genes associated with drug resistance is expected to reflect the normal mutation rates and occur at between 10⁻⁶ and 10⁻⁵ colonies. In bacteria, hetero-resistance has been described especially for vancomycin in S. aureus, vancomycin in Enterococcus faecium, colistin in Acinetobacter baumannii-calcoaceticus, rifampin, isoniazid, and streptomycin in M. tuberculosis, and penicillin in S. pneumoniae (Falagas et al., 2008; Rinder, 2001). Increased therapeutic failures and mortality have been reported in patients with hetero-resistant staphylococci and M. tuberculosis (Falagas et al., 2008; Hofmann-Thiel et al., 2009). For fungi, hetero-resistance leading to clinical failure has been described for fluconazole in Cryptococcus neoformans and Candida albicans (Marr et al., 2001; Mondon et al., 1999).

Viral replication is more error prone than replication in bacteria and fungi. Viral evolution under drug and immune pressure occurs relatively easily, commonly resulting in variants or quasi species that may contain drug-resistant subpopulations. This is not often termed hetero-resistance, but the principle is the same as described for bacteria and fungi: A virus may be considered susceptible to a drug because either phenotypic or genotypic tests reveal "lack" of resistance, when there is a resistant subpopulation just below the limit of assay detection. These minority quasi species that are resistant to antiretroviral agents have been associated with failure of antiretroviral therapy (Metzner et al., 2009).

EVOLUTIONARY BASISOF RESISTANCE EMERGENCE

Listen

Development of Resistance via Mutation Selection. Mutation and antibiotic selection of the resistant mutant are the molecular basis for resistance for many bacteria, viruses, and fungi. Mutations may occur in the gene encoding (1) the target protein, altering its structure so that it no longer binds the drug; (2) a protein involved in drug transport; (3) a protein important for drug activation or inactivation; or (4) in a regulatory gene or promoter affecting expression of the target, a transport protein, or an inactivating enzyme. Mutations are not caused by drug exposure per se. They are random events that confer a survival advantage when drug is present. Any large population of drug susceptible bacteria is likely to contain rare mutants that are only slightly less susceptible than the parent. However, suboptimal dosing strategies lead to selective kill of the more susceptible population, which leaves the resistant isolates to flourish.

In some instances, a single-step mutation results in a high degree of resistance. Examples include M. tuberculosis katG Ser315 mutations that cause resistance to isoniazid; the M814V mutation in the reverse transcriptase gene of HIV-1, causing resistance to lamivudine; a cytochrome-b gene mutation causing resistance to the antimalaria drug atovaquone; and Candida albicans fks1 Ser645 mutations causing resistance to echinocandins.

In other circumstances, however, it is the sequential acquisition of more than one mutation that leads to clinically significant resistance. As an example, the combination of pyrimethamine and sulfadoxine inhibits Plasmodium falciparum's folate biosynthetic pathway via inhibition of dihydrofolate reductase (DHFR) by the pyrimethamine and inhibition of dihydropteroate synthetase (DHPS) by sulfadoxine. Clinically meaningful resistance occurs when there is a single point mutation in the DHPS gene accompanied by at least a double mutation in the DHFR gene.

Hypermutable Phenotypes. The ability to protect genetic information from disintegrating and also to be flexible enough to allow genetic changes that lead to adaptation to the environment is essential to all living things. This is accomplished principally by the insertion of the correct base pair by DNA polymerase III, proofreading by the polymerase, and postreplicative repair. The development of a defect in one of these repair mechanisms leads to a high degree of mutations in many genes; such isolates are termed mutator (Mut) phenotypes and may include mutations in genes causing antibiotic resistance (Giraud et al., 2002). This second-order selection of hypermutable (mutator) alleles based on alterations in DNA repair genes has been implicated in the emergence of multidrug resistant strains of M. tuberculosis Beijing genotype (Rad et al., 2003).

Resistance by External Acquisition of Genetic Elements. Drug resistance may be acquired by mutation and selection, with passage of the trait vertically to daughter cells. For mutation and selection to be successful in generating resistance, the mutation cannot be lethal and should not appreciably alter virulence. For the trait to be passed on, the original mutant or its progeny also must disseminate and replicate; otherwise, the mutation will be lost until it is "rediscovered" by some other mutant arising from within a wild-type population.

Drug resistance more commonly is acquired by horizontal transfer of resistance determinants from a donor cell, often of another bacterial species, by transduction, transformation, or conjugation. Resistance acquired by horizontal transfer can disseminate rapidly and widely either by clonal spread of the resistant strain or by subsequent transfers to other susceptible recipient strains. Horizontal transfer of resistance offers several advantages over mutation selection. Lethal mutation of an essential gene is avoided; the level of resistance often is higher than that produced by mutation, which tends to yield incremental changes; the gene, which still can be transmitted vertically, can be mobilized and rapidly amplified within a population by transfer to susceptible cells; and the resistance gene can be eliminated when it no longer offers a selective advantage.

Horizontal Gene Transfer. Horizontal transfer of resistance genes is greatly facilitated by and is largely dependent on mobile genetic elements. Mobile genetic elements include plasmids and transducing phages. Other mobile elements—transposable elements, integrons, and gene cassettes—also participate in the process. Transposable elements are of three general types: insertion sequences, transposons, and transposable phages. Only insertion sequences and transposons are important for resistance.

Insertion sequences (Mahillon and Chandler, 1998) are short segments of DNA encoding enzymatic functions (e.g., transposase and resolvase) for site-specific recombination with inverted repeat sequences at either end. They can copy themselves and insert themselves into a chromosome or a plasmid. Insertion sequences do not encode resistance, but they function as sites for integration of other resistance-encoding elements (e.g., plasmids or transposons). Transposons are insertion sequences that also code for other functions, one of which can be drug resistance. Because transposons move between chromosome and plasmid, the resistance gene can "hitchhike" with a transferable element out of the host and into a recipient. Transposons are mobile elements that excise and integrate in the bacterial genomic or plasmid DNA (i.e., from plasmid to plasmid, from plasmid to chromosome, or from chromosome to plasmid). Integrons (Fluit and Schmitz, 2004) are not formally mobile and do not copy themselves, but they encode an integrase and provide a specific site into which mobile gene cassettes integrate. Gene cassettes encode resistance determinants, usually lacking a promoter, with a downstream repeat sequence. The integrase recognizes this repeat sequence and directs insertion of the cassette into position behind a strong promoter that is present on the integron. Integrons may be located within transposons or in plasmids, and therefore may be mobilizable, or located on the chromosome.

Transduction is acquisition of bacterial DNA from a phage (a virus that propagates in bacteria) that has incorporated DNA from a previous host bacterium within its outer protein coat. If the DNA includes a gene for drug resistance, the newly infected bacterial cell may acquire resistance. Transduction is particularly important in the transfer of antibiotic resistance among strains of S. aureus. Transformation is the uptake and incorporation into the host genome by homologous recombination of free DNA released into the environment by other bacterial cells. Transformation is the molecular basis of penicillin resistance in pneumococci and Neisseria (Spratt, 1994). Conjugation, as the name implies, is gene transfer by direct cell-to-cell contact through a sex pilus or bridge. This complex and fascinating mechanism for the spread of antibiotic resistance is extremely important because multiple resistance genes can be transferred in a single event. The transferable genetic material consists of two different sets of plasmid-encoded genes that may be on the same or different plasmids. One set encodes the actual resistance; the second encodes genes necessary for the bacterial conjugation process. Conjugation with genetic exchange between nonpathogenic and pathogenic microorganisms probably occurs in the GI tracts of humans and animals. The efficiency of transfer is low; however, antibiotics can exert a powerful selective pressure to allow emergence of the resistant strain. Genetic transfer by conjugation is common among gram-negative bacilli, and resistance is conferred on a susceptible cell as a single event. Enterococci also contain a broad range of host-range conjugative plasmids that are involved in the transfer and spread of resistance genes among gram-positive organisms.

CLINICAL SUMMARY

Listen

Antimicrobial agents work by targeting specific biochemical properties of pathogens, and therefore have a narrow spectrum of organisms that they can kill. Infection is usually in a particular anatomical compartment. Important determinants of success of antimicrobial therapy include proper selection of antimicrobial therapy based on microbiology results and susceptibility testing, knowledge of drug penetration into the infected compartment, and knowledge of compartmental pharmacokinetics. The proper dose and dosing schedule are chosen by integrating microbial pharmacokinetic-pharmacodynamic information, expected pharmacokinetic variability, and the minimum inhibitory concentration of the pathogen. The goals of therapy should be clear. Prophylaxis, pre-emptive therapy, empirical therapy, and definitive therapy should have treatment goals and duration of therapy clearly spelled out in the beginning, based on proper evidence. The general rule is monotherapy, except in select situations where combination therapy has been shown to be superior. Poor dosing strategies lead to catastrophic outcomes such as drug-resistant pathogens and untoward toxicity to the patients.

BIBLIOGRAPHY

Listen

Ambrose PG, Bhavnani SM, Rubino CM et al.. Pharmacokinetics-pharmacodynamics of antimicrobial therapy: It's not just for mice anymore. Clin Infect Dis, 2007, 44:79–86. [PubMed: 17143821]

Andes D, van Ogtrop M. In vivo characterization of the pharmacodynamics of flucytosine in a neutropenic murine disseminated candidiasis model. Antimicrob Agents Chemother, 2000, 44:938–942. [PubMed: 10722494]

Anonymous. Guidelines for preventing opportunistic infections among hematopoietic stem cell transplant recipients. MMWR Recomm Rep, 2000, 49:1–7.

Arion D, Kaushik N, McCormick S et al.. Phenotypic mechanism of HIV-1 resistance to 3′-azido-3′-deoxythymidine (AZT): Increased polymerization processivity and enhanced sensitivity to pyrophosphate of the mutant viral reverse transcriptase. Biochemistry, 1998, 37:15908–15917. [PubMed: 9843396]

Bratzler DW, Houck PM. Antimicrobial prophylaxis for surgery: An advisory statement from the National Surgical Infection Prevention Project. Clin Infect Dis, 2004, 38: 1706–1715. [PubMed: 15227616]

Bulitta JB, Ly NS, Yang JC et al.. Development and qualification of a pharmacodynamic model for the pronounced inoculum effect of ceftazidime against Pseudomonas aeruginosa. Antimicrob Agents Chemother, 2009, 53: 46–56. [PubMed: 18852268]

Chow AT, Chen A, Lattime H et al.. Penetration of levofloxacin into skin tissue after oral administration of multiple 750 mg once-daily doses. J Clin Pharm Ther, 2002, 27:143–150. [PubMed: 11975700]

Conte JE Jr, Golden JA, McIver M, Zurlinden E. Intrapulmonary pharmacokinetics and pharmacodynamics of high-dose levofloxacin in healthy volunteer subjects. Int J Antimicrob Agents, 2006, 28:114–121. [PubMed: 16837169]

Cosgrove SE, Vigliani GA, Fowler VG Jr et al.. Initial low-dose gentamicin for Staphylococcus aureus bacteremia and endocarditis is nephrotoxic. Clin Infect Dis, 2009, 48:713–721. [PubMed: 19207079]

Craig WA. Pharmacokinetic/pharmacodynamic parameters: Rationale for antibacterial dosing of mice and men. Clin Infect Dis, 1998, 26:1–10. [PubMed: 9455502]

Craig WA. Pharmacodynamics of antimicrobials: General concepts and applications. In: Antimicrobial Pharmacodynamics in Theory and Practice, 2nd ed. (Nightangle CH, Ambrose PG, Drusano G.L, Murakawa T, eds.), Informa Healthcare USA, New York, 2007, pp. 1–19.

Craig WA, Kunin CM. Significance of serum protein and tissue binding of antimicrobial agents. Annu Rev Med, 1976, 27: 287–300. [PubMed: 180874]

DHHS Panel on Guidelines for the Prevention and Treatment of Opportunistic Infections in HIV-Infected Adults and Adolescents. Guidelines for Prevention and Treatment of Opportunistic Infections in HIV-Infected Adults and Adolescents. Available at: http://aidsinfo.nih.gov/contentfiles/Adult_OI.pdf. Accessed February 15, 2009.

Eagle H, Fleishman R, Musselman AD. Effect of schedule of administration on the therapeutic efficacy of penicillin; importance of the aggregate time penicillin remains at effectively bactericidal levels Am J Med, 1950, 9:280–299. [PubMed: 14771084]

Falagas ME, Makris GC, Dimopoulos G, Matthaiou DK. Heteroresistance: A concern of increasing clinical significance? Clin Microbiol Infect, 2008, 14:101–104. [PubMed: 18093235]

Fluit AC, Schmitz FJ. Resistance integrons and super-integrons. Clin Microbiol Infect, 2004, 10:272–288. [PubMed: 15059115]

Gerna G, Lilleri D, Callegaro A et al.. Prophylaxis followed by preemptive therapy versus preemptive therapy for prevention of human cytomegalovirus disease in pediatric patients undergoing liver transplantation. Transplantation, 2008, 86:163–166. [PubMed: 18622294]

Gibaldi M, Perrier D. Pharmacokinetics, Marcel Dekker, New York, 1975.

Giraud A, Matic I, Radman M et al.. Mutator bacteria as a risk factor in treatment of infectious diseases. Antimicrob Agents Chemother, 2002, 46:863–865. [PubMed: 11850274]

Gumbo T. Integrating pharmacokinetics, pharmacodynamics and pharmacogenomics to predict outcomes in antibacterial therapy. Curr Opin Drug Discov Devel, 2008, 11:32–42. [PubMed: 18175265]

Gumbo T, Drusano GL, Liu W et al.. Anidulafungin pharmacokinetics and microbial response in neutropenic mice with disseminated candidiasis. Antimicrob Agents Chemother, 2006, 50:3695–3700. [PubMed: 16954319]

Gumbo T, Louie A, Deziel MR et al.. Concentration-dependent Mycobacterium tuberculosis killing and prevention of resistance by rifampin. Antimicrob Agents Chemother, 2007a, 51: 3781–3788.

Gumbo T, Louie A, Liu W et al.. Isoniazid bactericidal activity and resistance emergence: Integrating pharmacodynamics and pharmacogenomics to predict efficacy in different ethnic populations. Antimicrob Agents Chemother, 2007b, 51: 2329–2336.

Hanna GJ, Aquila RT. Clinical use of genotypic and phenotypic drug resistance testing to monitor antiretroviral chemotherapy. Clin Infect Dis, 2001, 32:774–782. [PubMed: 11229846]

Happi CT, Gbotosho GO, Folarin OA et al.. Selection of Plasmodium falciparum multidrug resistance gene 1 alleles in asexual stages and gametocytes by artemether-lumefantrine in Nigerian children with uncomplicated falciparum malaria. Antimicrob Agents Chemother, 2009, 53:888–895. [PubMed: 19075074]

Harshey RM, Ramakrishnan T. Purification and properties of DNA-dependent RNA polymerase from Mycobacterium tuberculosis H37RV. Biochim Biophys Acta, 1976, 432:49–59. [PubMed: 816387]

Hayden FG, Pavia AT. Antiviral management of seasonal and pandemic influenza. J Infect Dis, 2006, 194 (suppl 2):S119–S126. [PubMed: 17163384]

Hertogs K, Bethune MP, Miller V et al.. A rapid method for simultaneous detection of phenotypic resistance to inhibitors of protease and reverse transcriptase in recombinant human immunodeficiency virus type 1 isolates from patients treated with antiretroviral drugs. Antimicrob Agents Chemother, 1998, 42:269–276. [PubMed: 9527771]

Hirsch MS, Gunthard HF, Schapiro JM et al.. Antiretroviral drug resistance testing in adult HIV-1 infection: 2008 recommendations of an International AIDS Society-USA panel. Clin Infect Dis, 2008, 47:266–285. [PubMed: 18549313]

Hofmann-Thiel S, Ingen J, Feldmann K et al.. Mechanisms of heteroresistance to isoniazid and rifampin of Mycobacterium tuberculosis in Tashkent, Uzbekistan. Eur Respir J, 2009, 33:368–374. [PubMed: 18829680]

Hooper DC. Fluoroquinolone resistance among Gram-positive cocci. Lancet Infect Dis, 2002, 2:530–538. [PubMed: 12206969]

Hope WW, Seibel NL, Schwartz CL et al.. Population pharmacokinetics of micafungin in pediatric patients and implications for antifungal dosing. Antimicrob Agents Chemother, 2007, 51:3714–3719. [PubMed: 17638696]

Jumbe N, Louie A, Leary R et al.. Application of a mathematical model to prevent in vivo amplification of antibiotic-resistant bacterial populations during therapy. J Clin Invest, 2003, 112:275–285. [PubMed: 12865415]

Jumbe NL, Louie A, Miller MH et al.. Quinolone efflux pumps play a central role in emergence of fluoroquinolone resistance in Streptococcus pneumoniae. Antimicrob Agents Chemother, 2006, 50:310–317. [PubMed: 16377702]

Kiem S, Schentag JJ. Interpretation of antibiotic concentration ratios measured in epithelial lining fluid. Antimicrob Agents Chemother, 2008, 52:24–36. [PubMed: 17846133]

Koch R. Die aetiologie der Milzbrand-Krankheit, begründet auf die Entwicklungsgeschichte des Bacillus Anthracis. Beiträge zur Biologie der Pflanzen, 1876, 2:277–310.

Lewis K. Riddle of biofilm resistance. Antimicrob Agents Chemother, 2001, 45:999–1007. [PubMed: 11257008]

Lim D, Strynadka NC. Structural basis for the beta lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus. Nat Struct Biol, 2002, 9:870–876. [PubMed: 12389036]

Louie A, Kaw P, Liu W et al.. Pharmacodynamics of daptomycin in a murine thigh model of Staphylococcus aureus infection. Antimicrob Agents Chemother, 2001, 45:845–851. [PubMed: 11181370]

Mahillon J, Chandler M. Insertion sequences. Microbiol Mol Biol Rev, 1998, 62:725–774. [PubMed: 9729608]

Marr KA, Lyons CN, Ha K et al.. Inducible azole resistance associated with a heterogeneous phenotype in Candida albicans. Antimicrob Agents Chemother, 2001, 45:52–59. [PubMed: 11120944]

Metzner KJ, Giulieri SG, Knoepfel SA et al.. Minority quasispecies of drug-resistant HIV-1 that lead to early therapy failure in treatment-naive and -adherent patients. Clin Infect Dis, 2009, 48:239–247. [PubMed: 19086910]

Miller DS, Bauer B, Hartz AM. Modulation of P-glycoprotein at the blood-brain barrier: Opportunities to improve central nervous system pharmacotherapy. Pharmacol Rev, 2008, 60:196–209. [PubMed: 18560012]

Moise-Broder PA, Sakoulas G, Eliopoulos GM et al.. Accessory gene regulator group II polymorphism in methicillin-resistant Staphylococcus aureus is predictive of failure of vancomycin therapy. Clin Infect Dis, 2004, 38:1700–1705. [PubMed: 15227615]

Mondon P, Petter R, Amalfitano G et al.. Heteroresistance to fluconazole and voriconazole in Cryptococcus neoformans. Antimicrob Agents Chemother, 1999, 43:1856–1861. [PubMed: 10428902]

Nakajima Y. Mechanisms of bacterial resistance to macrolide antibiotics. J Infect Chemother, 1999, 5:61–74. [PubMed: 11810493]

Neumann AU, Lam NP, Dahari H et al.. Hepatitis C viral dynamics in vivo and the antiviral efficacy of interferon-alpha therapy. Science, 1998, 282:103–107. [PubMed: 9756471]

Nijhuis M, Maarseveen NM, Boucher CA. Antiviral resistance and impact on viral replication capacity: Evolution of viruses under antiviral pressure occurs in three phases. Handb Exp Pharmacol, 2009, 189:299–320. [PubMed: 19048205]

Ouellette M. Biochemical and molecular mechanisms of drug resistance in parasites. Trop Med Int Health, 2001, 6: 874–882. [PubMed: 11703841]

Panlilio AL, Cardo DM, Grohskopf LA et al.. Updated U.S. Public Health Service guidelines for the management of occupational exposures to HIV and recommendations for postexposure prophylaxis. MMWR Recomm Rep, 2005, 54:1–17. [PubMed: 16195697]

Pappas PG, Kauffman CA, Andes D et al.. Clinical practice guidelines for the management of candidiasis: 2009 update by the Infectious Diseases Society of America. Clin Infect Dis, 2009, 48:503–535. [PubMed: 19191635]

Pasteur L. Mémoire sur les corpuscles organisés qui existent dans l'atmosphére. Examen de la doctrine des générations spontanées Annales des sciences naturelles, 1861, 16: 5–98.

Petropoulos CJ, Parkin NT, Limoli KL et al.. A novel phenotypic drug susceptibility assay for human immunodeficiency virus type 1. Antimicrob Agents Chemother, 2000, 44: 920–928. [PubMed: 10722492]

Preston SL, Drusano GL, Berman AL et al.. Pharmacodynamics of levofloxacin: A new paradigm for early clinical trials. JAMA[JAMA and JAMA Network Journals Full Text], 1998, 279:125–129. [PubMed: 9440662]

Pukrittayakamee S, Wanwimolruk S, Stepniewska K et al.. Quinine pharmacokinetic-pharmacodynamic relationships in uncomplicated falciparum malaria. Antimicrob Agents Chemother, 2003, 47:3458–3463. [PubMed: 14576102]

Quagliarello VJ, Scheld WM. Treatment of bacterial meningitis. N Engl J Med, 1997, 336:708–716. [PubMed: 9041103]

Rad ME, Bifani P, Martin C et al.. Mutations in putative mutator genes of Mycobacterium tuberculosis strains of the W-Beijing family. Emerg Infect Dis, 2003, 9:838–845. [PubMed: 12890325]

Rinder H. Hetero-resistance: An under-recognised confounder in diagnosis and therapy? J Med Microbiol, 2001, 50: 1018–1020. [PubMed: 11761184]

Sheiner LB, Rosenberg B, Marathe VV. Estimation of population characteristics of pharmacokinetic parameters from routine clinical data. J Pharmacokinet Biopharm, 1977, 5:445–479. [PubMed: 925881]

Singh N. Preemptive therapy versus universal prophylaxis with ganciclovir for cytomegalovirus in solid organ transplant recipients. Clin Infect Dis, 2001, 32:742–751. [PubMed: 11229841]

Singh N, Wannstedt C, Keyes L et al.. Valganciclovir as preemptive therapy for cytomegalovirus in cytomegalovirus-seronegative liver transplant recipients of cytomegalovirus-seropositive donor allografts. Liver Transpl, 2008, 14:240–244. [PubMed: 18236404]

Spratt BG. Resistance to antibiotics mediated by target alterations. Science, 1994, 264:388–393. [PubMed: 8153626]

Stepniewska K, Chotivanich K, Brockman A et al.. Overestimating resistance in field testing of malaria parasites: Simple methods for estimating high EC50 values using a Bayesian approach. Malar J, 2007, 6:4.

Talal AH, Ribeiro RM, Powers KA et al.. Pharmacodynamics of PEG-IFN alpha differentiate HIV/HCV coinfected sustained virological responders from nonresponders. Hepatology, 2006, 43:943–953. [PubMed: 16761329]

Tam VH, Louie A, Deziel MR et al.. The relationship between quinolone exposures and resistance amplification is characterized by an inverted U: A new paradigm for optimizing pharmacodynamics to counterselect resistance. Antimicrob Agents Chemother, 2007, 51:744–747. [PubMed: 17116679]

Tarning J, Ashley EA, Lindegardh N et al.. Population pharmacokinetics of piperaquine after two different treatment regimens with dihydroartemisinin-piperaquine in patients with Plasmodium falciparum malaria in Thailand. Antimicrob Agents Chemother, 2008, 52:1052–1061. [PubMed: 18180343]

Vogelman B, Gudmundsson S, Leggett J et al.. Correlation of antimicrobial pharmacokinetic parameters with therapeutic efficacy in an animal model. J Infect Dis, 1988, 158: 831–847. [PubMed: 3139779]

Wagenlehner FM, Kinzig-Schippers M, Sorgel F et al.. Concentrations in plasma, urinary excretion and bactericidal activity of levofloxacin (500 mg) versus ciprofloxacin (500 mg) in healthy volunteers receiving a single oral dose. Int J Antimicrob Agents, 2006, 28:551–519. [PubMed: 17101261]

Wehrli W, Knusel F, Schmid K, Staehelin M. Interaction of rifamycin with bacterial RNA polymerase. Proc Natl Acad Sci U S A, 1968, 61:667–673. [PubMed: 4879400]

Wilkins JJ, Savic RM, Karlsson MO et al.. Population pharmacokinetics of rifampin in pulmonary tuberculosis patients, including a semimechanistic model to describe variable absorption. Antimicrob Agents Chemother, 2008, 52: 2138–2148. [PubMed: 18391026]

Wilson W, Taubert KA, Gewitz M et al.. Prevention of infective endocarditis: Guidelines from the American Heart Association; a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation, 2007, 116:1736–1754. [PubMed: 17446442]

Zhou XJ, Sheiner LB, D'Aquila RT et al.. Population pharmacokinetics of nevirapine, zidovudine, and didanosine in human immunodeficiency virus-infected patients. The National Institute of Allergy and Infectious Diseases AIDS Clinical Trials Group Protocol 241 Investigators. Antimicrob Agents Chemother, 1999, 43:121–128. [PubMed: 9869576]

Pop-up div Successfully Displayed

This div only appears when the trigger link is hovered over. Otherwise it is hidden from view.

Chapter 48: General Principles of Antimicrobial Therapy

Send Email

Citation

Jump to a Section

SCIENTIFIC BASIS OF ANTIMICROBIAL CHEMOTHERAPY

Figure 48–1.

THE PHARMACOKINETIC BASISOF ANTIMICROBIAL THERAPY

Figure 48–2.

IMPACT OF SUSCEPTIBILITY TESTING ON SUCCESS OF ANTIMICROBIAL AGENTS

Figure 48–3.

BASIS FOR SELECTION OF DOSE AND DOSING SCHEDULE

Figure 48–4.

TYPES AND GOALS OFANTIMICROBIAL THERAPY

Figure 48–5.

MECHANISMS OF RESISTANCETO ANTIMICROBIAL AGENTS

EVOLUTIONARY BASISOF RESISTANCE EMERGENCE

CLINICAL SUMMARY

BIBLIOGRAPHY

Pop-up div Successfully Displayed