Principles of Antimicrobial Use

INTRODUCTION

Viruses, bacteria, other unicellular organisms, and multicellular organisms in the environment can also live in the human body. This relationship can produce desirable and undesirable responses within the host. In a healthy person, normal bacteria in the gastrointestinal tract have beneficial effects, assisting in the production of vitamins and the breakdown of foods. However, when nonbeneficial, pathogenic organisms enter the body through oral ingestion, inhalation, trauma, surgical procedures, or any body opening, the result is an unwanted host-pathogen response, termed an infection. The responses to an infection can occur either directly through the release of toxins or antigens by the invading organism or indirectly through tissue invasion or generation of an inflammatory response by the body. Many infections develop as a result of noninfectious primary diseases that suppress the patient’s natural immune response. However, on a global scale, parasitic infections are the greatest cause of morbidity and mortality worldwide.

Table 45-1 lists the types of pathogenic organisms that invade the human body and cause unwanted biological responses. Treatment of the problems generated by these invasions is approached in two ways: (1) destruction or removal of invading organisms, and (2) alleviation of symptoms. The availability of drugs for successful eradication of invading organisms varies considerably with the type and location of the organisms within the human host.

TABLE 45–1 Pathological Organisms that Can Live in a Parasitic Invader-Host Relationship in Humans, Listed in Order of Increasing Complexity

P, Prokaryotes (no nuclear membrane); E, eukaryotes (with a nucleus).

The only acellular organisms known to induce infectious diseases in humans are viruses and proteinaceous agents lacking nucleic acids called prions. Other small molecular weight, acellular nucleic acids called viroids can exist in humans but probably do not contribute to disease. Bacteria are unicellular, nonnuclear organisms. Higher orders of size and complexity are found in unicellular, nucleated fungi (including yeast and filamentous forms) and protozoa. Major differences include the addition of a membrane-enclosed nucleus and mitochondria within the cell. More complex fungi are found in the multicellular, nucleated molds (see Chapter 50). Still higher orders of parasitic organisms are helminths (worms), which are estimated to infect 40% to 60% of the world’s population and are a medical problem in both industrialized and developing countries (see Chapter 52).

Viral infections are also a major source of temporary disability and loss of productivity in humans, and some viral infections are fatal. There are still only a few effective

drugs available to halt proliferation of, or eliminate, clinically important viruses, although this situation is improving. Vaccines have provided the most effective protection against viruses; however, prophylactic use of some antiviral drugs has also proven effective in preventing certain viral infections (see Chapter 51).

The development of antimicrobial therapy is considered by many to be one of the most important advances in the history of medicine. Major historical events leading to the development of antimicrobial therapy are listed in Box 45-1. The term antibiotic traditionally refers to substances produced by microorganisms to suppress the growth of other microorganisms. The term antimicrobial agent is broader in meaning, because it encompasses drugs synthesized in the laboratory and those natural antibiotics produced by microorganisms. Some natural antibiotics produced originally by microbial fermentation are now produced by chemical synthesis. Many agents are semisynthetic; that is, the key portion of the compound is produced by microbial fermentation, and various moieties are attached synthetically. Thus the distinction between the terms antibiotics and antimicrobial agents is somewhat blurred and has little meaning today.

The efficacy and relative safety of these drugs has led to their widespread use and overuse, with up to 50% of antimicrobials consumed in the United States considered unnecessary. Because of their ability to alter microbial flora and lead to antibiotic-resistant microorganisms, antimicrobials are fundamentally different from other types of drugs. It is important to always remember that the pathogenicity of a bacterial species may mutate with time as new strains appear, existing strains develop resistance, old strains disappear, and new problems emerge with old strains thought to be relatively benign. This complicates planning drug treatment and requires the clinician to stay informed and use these agents wisely.

This chapter presents an overview of the classes of antimicrobial drugs and the principles involved in selecting antimicrobial therapy. Although this chapter focuses on antibacterial agents, many of the principles discussed also apply to antifungal, antiviral, and antiparasitic drugs.

Antimicrobial agents can be classified into major groups according to the point in the cellular biochemical pathways at which they exert their primary mechanism of action (Fig. 45-1). These are:

FIGURE 45–1 Antimicrobial sites of bactericidal or bacteriostatic action on microorganisms. The five general mechanisms comprise (1) inhibition of synthesis of cell wall, (2) damage to cell membrane, (3) modification of nucleic acid/DNA synthesis, (4) modification of protein synthesis (at ribosomes), and (5) modification of energy metabolism within the cytoplasm (at folate cycle). PABA, Paraaminobenzoic acid; DHFA, dihydrofolic acid; THFA, tetrahydrofolic acid.

Agents that inhibit the synthesis of cell walls include the β-lactams, such as penicillins, cephalosporins, monobactams, and carbapenems, and others, such as vancomycin. Inhibitors of cytoplasmic membranes include the polymyxins and daptomycin. In fungi the cell wall is damaged by the polyene antifungal drugs or by the azoles. Inhibitors of nucleic acid synthesis include the quinolones, which inhibit deoxyribonucleic acid (DNA) gyrase, and the ribonucleic acid (RNA) polymerase inhibitor rifampin. Protein synthesis is inhibited by aminoglycosides, tetracyclines, chloramphenicol, erythromycin, clindamycin, linezolid, and streptogramins. These agents are usually bacteriostatic, except the aminoglycosides and sometimes streptogramins, which can be bactericidal. Finally, folate antagonists such as sulfonamides and trimethoprim interfere with cell metabolism. These antimicrobial drugs are listed by their mechanisms of action in Table 45-2.

TABLE 45–2 Classification of Antimicrobial Agents by Mechanism of Action

DRUG SELECTION FOR INDIVIDUAL PATIENTS

Major factors to be considered when selecting an antimicrobial agent are outlined in Box 45-2. An understanding of these factors assists with rational selection of an appropriate compound. It is important to keep in mind that each factor is important, but their relative significance may vary from case to case.

Is an Antimicrobial Agent Indicated?

Approximately half of antimicrobial agents used in the United States are not necessary. These drugs are often given for viral infections that do not respond to antibiotics, or to treat noninfectious processes mimicking a bacterial infection. Antibiotics are often misprescribed because of culture isolation of an organism that is colonizing an anatomical site and not causing an infection. In general, the clinician should resist beginning antimicrobial therapy unless there is a reasonable probability that a bacterial infection is present. However, the threshold for starting antimicrobial therapy needs to be adjusted depending on the clinical setting. When the downside risk of withholding therapy is great, such as with bacterial meningitis or in clinically unstable patients, therapy should be started without delay, even when the presence of a bacterial infection is uncertain. Another indication for antimicrobials is prophylactic therapy, which is intended to prevent illness in someone at risk of infection.

Identification of the Pathogen

For many infections an attempt should be made to identify the possible pathogen or pathogens before initiating antimicrobial therapy. Specimens for direct inspection and culture should be obtained before therapy is started, because drug therapy can decrease the yield of culture. Gram staining is the fastest, simplest, and most inexpensive method to identify bacteria and fungi. Because a Gram stain can be accomplished quickly, the results can be used to guide the initial antimicrobial choice. Body fluids that are normally sterile should be Gram stained. In addition, stains of wound exudates, sputum, and fecal material can sometimes yield information about the infecting microorganisms.

Empiric versus Directed Therapy

Although every effort should be made to obtain material for laboratory testing, for certain infections including cellulitis, otitis media, and sinusitis, it is often not possible or practical to obtain specimens for Gram stain and culture. In other infections such as pneumonia, the yield of culture material is low. When it is not possible to obtain a specimen, antimicrobial therapy is often started empirically, based on knowledge of the antimicrobial spectrum of an agent and the likely pathogens causing infection at a particular anatomical site.

Empiric therapy is usually broader than therapy directed at a particular pathogen or therapy based on the results of susceptibility testing. However, the use of agents with broad activity disturbs normal bacterial flora to a greater degree than narrow spectrum therapy and may promote the development of antibiotic resistant pathogens. Although empiric therapy is often used early in the course of therapy, the specter of antibiotic resistance and the added costs associated with broad therapy emphasize the importance of narrowing coverage if and when susceptibility testing results become available.

The setting in which an infection occurs should also be considered when selecting empiric therapy. Nosocomial infections, those occurring in the hospital setting, are often caused by antibiotic-resistant bacteria, and the pattern of antibiotic resistance can vary from hospital to hospital. Most hospitals compile an “antibiogram” that lists the susceptibility profiles of common pathogens recovered at that institution. Knowledge of these local resistance patterns is very helpful in directing empiric therapy.

Susceptibility of Infecting Microorganisms

When pathogenic bacteria are isolated in culture, susceptibility to specific antimicrobial agents can be determined. The results of these tests generally are not available until 18 to 48 hours after an initial culture sample has been obtained.

Susceptibility testing is often performed by automated systems based on the broth dilution method in which antibiotics are tested in serial dilutions that encompass the concentrations normally achieved in humans. This method detects the lowest concentration of antimicrobial agent that prevents visible growth after incubation for 18 to 24 hours, referred to as the minimal inhibitory concentration (MIC). The technique is shown in Figure 45-2, A. Deciding whether the results demonstrate that the organisms are susceptible requires an understanding of the pharmacokinetics of the antimicrobial agent, correlations between clinical outcomes and MIC data, and knowledge of the relationship between resistance mechanisms and MICs. Determining the MIC “cutoff” for susceptibility is often not straightforward and is the purview of an international committee. This same test procedure can be extended further to determine the minimal bactericidal concentration (MBC), or minimal concentration that kills 99.9% of cells. In this test samples are removed from the antibiotic-containing tubes in which there is no visible microbial growth and plated on agar that contains no additional antibiotic. The MBC is the lowest concentration of antibiotic in the original tube from which bacteria do not grow on the agar (see Fig. 45-2, B). MBC determinations are no longer used regularly in most clinical laboratories.

FIGURE 45–2 Dilution/agar tests for determination of MIC and MBC for a given drug and microorganism. A, Dilution test: each tube contains 5 × 10⁵ colony-forming units of bacteria, plus antibiotic at the concentration indicated. The MIC is the minimal drug concentration at which no visible growth of bacteria is observed (4 μg/mL in this example). This method has been adapted to automated systems by using a microtiter plate instead of test tubes. B, Dilution/agar test: each tube in A that shows no visible growth is cultured on a section of the new agar plate (no additional antibiotic is added to the agar plate). The MBC is the lowest concentration at which no growth occurs on the agar (32 μg/mL in this example).

Another method for determining bacterial susceptibility to antibiotics is the disk diffusion method, in which disks impregnated with the drugs to be tested are placed on an agar plate freshly inoculated with the bacterial strain in question. After the plates have been incubated for 18 to 24 hours, bacterial growth occurs everywhere except near the disks, where a “zone of inhibition” may be present. If sufficiently large, the zone indicates that the bacterial strain is susceptible to the antibiotic in the disk. This test is simple to perform but is only semiquantitative and not useful for determining the susceptibility of many slow-growing or fastidious organisms; it has been replaced by automated broth dilution systems in many laboratories. The test procedure and typical results are shown in Figure 45-3. A newer and related method is the E-test, which uses a strip containing the antibiotic in a concentration gradient along with a numerical scale instead of a disk. The point where the zone of inhibition touches the strip indicates the MIC (Fig. 45-4).

FIGURE 45–3 Disk diffusion method for testing bacteria for susceptibility to specific antimicrobial agents.

FIGURE 45–4 E-test method for susceptibility testing. The strip impregnated with a gradient of antibiotic is placed on a freshly inoculated agar plate. The point where the border of the zone of inhibition touches the strip defines the MIC (arrow).

Because of the many antimicrobial agents available, it is difficult to routinely test all antimicrobial agents against an isolate. To circumvent this problem, laboratories often use one compound as representative of a class of compounds. It is important to recognize that susceptibility tests require interpretation, are not error-proof, and may fail to identify a resistant subset population.

Need for Bactericidal versus Bacteriostatic Agents

Antimicrobial agents can be bactericidal (i.e., the organisms are killed) or bacteriostatic (i.e., the organisms are prevented from growing) (Fig. 45-5). A given agent may show bactericidal actions under certain conditions but bacteriostatic actions under others, depending on the concentration of drug and the target bacteria. A bacteriostatic agent often is adequate in uncomplicated infections, because the host defenses will help eradicate the microorganism. For example, in pneumonococcal pneumonia, bacteriostatic agents suppress the multiplication of the pneumococci, and the pneumococci are destroyed by interaction with alveolar macrophages and polymorphonuclear leukocytes. For a neutropenic individual, such a bacteriostatic agent might prove ineffective, and a bactericidal agent would be necessary. Thus the status of the host influences whether a bactericidal or bacteriostatic agent is selected.

FIGURE 45–5 Bactericidal versus bacteriostatic antimicrobial agents. A typical culture is started at 10⁵ colony-forming units (CFU) and incubated at 37° C for various times. In the absence of an antimicrobial agent, there is cell growth. With a bacteriostatic agent added, no growth occurs, but neither are the existing cells killed. If the added agent is bactericidal, 99.9% of the cells are killed during the standardized test time.

In addition, because the site of infection influences the ability of certain host defenses to contend effectively with microbes, bactericidal agents are required for management of infections in areas “protected” from host immune responses, such as endocarditic vegetations and cerebrospinal fluid (CSF). In endocarditis, treatment with bacteriostatic antibiotics such as tetracyclines or erythromycin is associated with an unacceptably high failure rate; in contrast, treatment with bactericidal agents such as penicillin is associated with cure rates in excess of 95%. Another example is the need for concentrations of an antimicrobial agent eightfold to tenfold greater than the MBC in the spinal fluid of patients with meningitis to effect a cure. Thus susceptibility is not the only criterion for efficacy.

Pharmacokinetic and Pharmacodynamic Factors

Antimicrobial agents are usually administered by oral (PO), intramuscular (IM), or intravenous (IV) routes. Most agents reach peak serum concentrations 1 to 2 hours after oral administration. However, peak concentrations may be delayed if drugs are ingested with food or in patients with delayed intestinal transit time such as sometimes occurs in diabetics. Peak plasma concentrations are reached in 0.5 to 1 hour after IM injections and 20 to 30 minutes after an IV infusion. Antimicrobial agents vary widely in their oral bioavailability. Some agents including trimethoprim/sulfamethoxazole, fluoroquinolones, rifampin, and metronidazole are almost completely absorbed after oral administration; these drugs can often be used orally even when a severe infection is present. That said, most life-threatening infections are treated, at least initially, with IV agents. Parenteral therapy ensures adequate serum levels, and, for many agents, higher drug levels can be achieved when administered IV.

The amount of antimicrobial agent that reaches the extravascular tissues and fluids where the infection is usually present depends on basic pharmacokinetic principles (see Chapters 2 and Chapter 3). Considerable variation exists among the many antimicrobial agents, and the pharmacokinetic profile of the antimicrobials is an important factor in selecting the proper drug.

Certain antibiotics including fluoroquinolones and aminoglycosides kill bacteria faster at higher concentrations, a property called concentration-dependent killing. These agents also continue to inhibit growth of bacteria for several hours after the concentrations of the drug fall below the MIC in the serum. This is called postantibiotic