Empiric Antimicrobial Therapy
Empiric antimicrobial therapy is begun before a specific pathogen has been identified and is based on the presumption of an infection that requires immediate drug treatment. Before initiation of such therapy, accepted practice involves making a clinical diagnosis of microbial infection, obtaining specimens for laboratory analyses, making a microbiologic diagnosis, deciding whether treatment should precede the results of laboratory tests, and, finally, selecting the optimal drug or drugs. A variety of publications provide annually updated lists of antimicrobial drugs of choice for specific pathogens. Such lists can provide a useful guide to empiric therapy based on presumptive microbiologic diagnosis. Tables 51–1 and 51–2 show examples of empiric antimicrobial therapy based on microbiologic etiology.
Table 51–1 Examples of Empiric Antimicrobial Therapy Based on Microbiologic Etiology.a ||Download (.pdf)
Table 51–1 Examples of Empiric Antimicrobial Therapy Based on Microbiologic Etiology.a
|Pathogen||Drug(s) of First Choice||Alternative Drugs|
|Enterococcus spp||Ampicillin +/− gentamicin||Vancomycin +/− gentamicin|
|S aureus or epidermidis|
|Methicillin-susceptible||Nafcillin||Cephalosporin, clindamycin, fluoroquinolone, imipenem|
|Methicillin-resistant||Vancomycin +/− gentamicin +/− rifampin||Daptomycin, doxycycline, fluoroquinolone, linezolid, streptogramins, tigecycline|
|Penicillin-susceptible||Pen G, amoxicillin||Cephalosporin, clindamycin, fluoroquinolone, macrolide, TMP-SMZ|
|Penicillin-resistant||Vancomycin + ceftriaxone or cefotaxime +/− rifampin||Linezolid, streptogramins, third-generation fluoroquinolone|
|N gonorrhoeae||Ceftriaxone, cefixime||Spectinomycin, azithromycin|
|M meningitidis||Penicillin G||Third-generation cephalosporin, chloramphenicol|
|M catarrhalis||Cefuroxime, TMP-SMZ||Amoxicillin-clavulanate, third-generation fluoroquinolone, macrolide|
|C difficile||Metronidazole||Vancomycin, bacitracin|
|C trachomatis||Macrolide or tetracycline||Clindamycin, ofloxacin|
|C pneumoniae||Macrolide or tetracycline||Fluoroquinolone|
|M pneumoniae||Macrolide or tetracycline||Fluoroquinolone|
|T pallidum||Penicillin G||Doxycycline, ceftriaxone, azithromycin|
Table 51–2 Further Examples of Empiric Antimicrobial Therapy Based on Microbiologic Etiology.a ||Download (.pdf)
Table 51–2 Further Examples of Empiric Antimicrobial Therapy Based on Microbiologic Etiology.a
|Pathogen||Drug(s) of First Choice||Alternative Drugs|
|Bacteroides||Metronidazole||Carbapenems, penicillins + beta-lactamase inhibitor, chloramphenicol|
|Campylobacter jejuni||Macrolide||Fluoroquinolone, tetracycline|
|Enterobacter spp||Carbapenem, TMP-SMZ||Aminoglycoside, cefepime, fluoroquinolone, third-generation cephalosporin|
|E coli||Cephalosporin (first and second-generation), TMP-SMZ||Many penicillins +/− beta-lactamase inhibitor, fluoroquinolones, aminoglycosides|
|K pneumoniae||Cephalosporin (first or second-generation), TMP-SMZ||Carbapenems, penicillins + beta-lactamase inhibitor, aminoglycosides, fluoroquinolones|
|P mirabilis||Ampicillin||Cephalosporins, penicillins + beta-lactamase inhibitor, aminoglycosides, TMP-SMZ, fluoroquinolones|
|Proteus-indole positive||Cephalosporin (first or second-generation), TMP-SMZ||Carbapenems, penicillins + beta-lactamase inhibitor, aminoglycosides, fluoroquinolones|
|S typhi||Ceftriaxone or fluoroquinolone||Chloramphenicol, TMP-SMZ, ampicillin|
|Serratia spp||Carbapenem||Aminoglycoside, third-generation cephalosporin, fluoroquinolone, TMP-SMZ|
|Shigella spp||Fluoroquinolone||Azithromycin, TMP-SMZ, ampicillin, ceftriaxone|
Principles of Antimicrobial Therapy
Antimicrobial therapy in established infections is guided by several principles.
The results of susceptibility testing establish the drug sensitivity of the organism. These results usually predict the minimum inhibitory concentrations (MICs) of a drug for comparison with anticipated blood or tissue levels. The 2 most common methods of susceptibility testing are disk diffusion (Kirby-Bauer) and broth dilution. For severe infections caused by certain bacteria (eg, gram-positive cocci, Haemophilus influenzae), a direct test for beta-lactamase is used to aid in the selection of an appropriate antibiotic.
Drug Concentration in Blood
The measurement of drug concentration in the blood may be appropriate when using agents with a low therapeutic index (eg, aminoglycosides, vancomycin) and when investigating poor clinical response to a drug treatment regimen.
Serum Bactericidal Titers
In certain infections in which host defenses may contribute minimally to cure, the estimation of serum bactericidal titers can confirm the appropriateness of choice of drug and dosage. Serial dilutions of serum are incubated with standardized quantities of the pathogen isolated from the patient; killing at a dilution of 1:8 is generally considered satisfactory.
Parenteral therapy is preferred in most cases of serious microbial infections. Chloramphenicol, the fluoroquinolones, and trimethoprim-sulfamethoxazole (TMP-SMZ) may be effective orally.
Monitoring of Therapeutic Response
Therapeutic responses to drug therapy should be monitored clinically and microbiologically to detect the development of resistance or superinfections. The duration of drug therapy required depends on the pathogen (eg, longer courses of therapy are required for infections caused by fungi or mycobacteria), the site of infection (eg, endocarditis and osteomyelitis require longer duration of treatment), and the immunocompetence of the patient.
Clinical Failure of Antimicrobial Therapy
Inadequate clinical or microbiologic response to antimicrobial therapy can result from laboratory testing errors, problems with the drug (eg, incorrect choice, poor tissue penetration, inadequate dose), the patient (poor host defenses, undrained abscesses), or the pathogen (resistance, superinfection).
Factors Influencing Antimicrobial Drug Use
Bactericidal versus Bacteriostatic Actions
Antibiotics classified as bacteriostatic include clindamycin, macrolides, sulfonamides, and tetracyclines. For bacteriostatic drugs, the concentrations that inhibit growth are much lower than those that kill bacteria. Antibiotics classified as bactericidal include the aminoglycosides, beta-lactams, fluoroquinolones, metronidazole, most antimycobacterial agents, streptogramins, and vancomycin. For such drugs, there is little difference between the concentrations that inhibit growth and those that kill bacteria. Bactericidal drugs are preferred for the treatment of endocarditis and meningitis and for most infections in patients with impaired defense mechanisms, especially immunocompromised patients.
Some bactericidal agents (aminoglycosides, fluoroquinolones) cause concentration-dependent killing. Maximizing peak blood levels of such drugs increases the rate and the extent of their bactericidal effects. This is one of the factors responsible for the clinical effectiveness of high-dose, once-daily administration of aminoglycosides. Other bactericidal agents (beta-lactams, vancomycin) cause time-dependent killing. Their killing action is independent of drug concentration and continues only while blood levels are maintained above the minimal bactericidal concentration (MBC).
Inhibition of bacterial growth that continues after antibiotic blood concentrations have fallen to low levels is called the postantibiotic effect (PAE). The mechanisms of PAE are unclear but may reflect the lag time required by bacteria to synthesize new enzymes and cellular components, the possible persistence of antibiotic at the target site, or an enhanced susceptibility of bacteria to phagocytic and other defense mechanisms including postantibiotic leucocyte enhancement. PAE is another factor contributory to the effectiveness of once-daily administration of aminoglycosides and may also contribute to the clinical efficacy of the fluoroquinolones.
Drug Elimination Mechanisms
Changes in hepatic and renal function—and the use of dialysis—can influence the pharmacokinetics of antimicrobials and may necessitate dosage modifications. The major mechanisms of elimination of commonly used antimicrobial drugs are shown in Table 51–3. In anuria (creatinine clearance <5 mL/min), the elimination half-life of drugs that are eliminated by the kidney is markedly increased, usually necessitating major reductions in drug dosage. Erythromycin, clindamycin, chloramphenicol, rifampin, and ketoconazole are notable exceptions, requiring no change in dosage in renal failure. Drugs contraindicated in renal impairment include cidofovir, nalidixic acid, long-acting sulfonamides, and tetracyclines. Dosage adjustment may be needed in patients with hepatic impairment for drugs including amprenavir, chloramphenicol, clindamycin, erythromycin, indinavir, metronidazole, and tigecycline. Dialysis, especially hemodialysis, may markedly decrease the plasma levels of many antimicrobials; supplementary doses of such drugs may be required to reestablish effective plasma levels after these procedures. Drugs that are not removed from the blood by hemodialysis include amphotericin B, cefonicid, cefoperazone, ceftriaxone, erythromycin, nafcillin, tetracyclines, and vancomycin.
Table 51–3 Elimination of Commonly Used Antimicrobial Agents.a ||Download (.pdf)
Table 51–3 Elimination of Commonly Used Antimicrobial Agents.a
|Mode of Elimination||Drugs or Drug Groups|
|Renal||Acyclovir, aminoglycosides, amphotericin B, most cephalosporins, fluconazole, fluoroquinolones, penicillins, sulfonamides, tetracyclines (except doxycycline), TMP-SMZ, vancomycin|
|Hepatic||Amphotericin B, ampicillin, cefoperazone, chloramphenicol, clindamycin, erythromycin, isoniazid, most azoles (not fluconazole), nafcillin, rifampin|
|Hemodialysis||Acyclovir (and most antiviral agents), aminoglycosides, cephalosporins (not cefonicid, cefoperazone, ceftriaxone), penicillins (not nafcillin), sulfonamides|
Pregnancy and the Neonate
Antimicrobial therapy during pregnancy and the neonatal period requires special consideration. Aminoglycosides (eg, gentamicin) may cause neurologic damage. Tetracyclines cause tooth enamel dysplasia and inhibition of bone growth. Sulfonamides, by displacing bilirubin from serum albumin, may cause kernicterus in the neonate. Chloramphenicol may cause gray baby syndrome. Other drugs that should be used with extreme caution during pregnancy include most antiviral and antifungal agents. The fluoroquinolones are not recommended for use in pregnancy or in small children because of possible effects on growing cartilage.
Interactions sometimes occur between antimicrobials and other drugs (see also Chapter 61). Interactions include enhanced nephrotoxicity or ototoxicity when aminoglycosides are given with loop diuretics, vancomycin, or cisplatin. Several drug interactions with sulfonamides are based on competition for plasma protein binding; these include excessive hypoglycemia with sulfonylureas and increased hypoprothrombinemia with warfarin. Disulfiram-like reactions to ethanol occur with metronidazole, with TMP-SMZ, and with several cephalosporins (see Chapter 43). Erythromycin inhibits the hepatic metabolism of a number of drugs, including clozapine, lidocaine, loratadine, phenytoin, quinidine, sildenafil, theophylline, and warfarin. Ketoconazole inhibits the metabolism of caffeine, carbamazepine, cyclosporine, hepatic hydroxylmethylglutaryl coenzyme A (HMG-CoA) reductase inhibitors, methadone, oral contraceptives, phenytoin, sildenafil, verapamil, and zidovudine. Other azole antifungals are weaker inhibitors of drug metabolism.
Rifampin, an inducer of hepatic drug-metabolizing enzymes, decreases the effects of digoxin, ketoconazole, oral contraceptives, propranolol, quinidine, several antiretroviral drugs, and warfarin.
Antimicrobial Drug Combinations
Therapy with multiple antimicrobials may be indicated in several clinical situations.
In severe infections (eg, sepsis, meningitis), combinations of antimicrobial drugs are used empirically to suppress all of the most likely pathogens.
The combined use of drugs is valid when the rapid emergence of resistance impairs the chances for cure. For this reason, combined drug therapy is especially important in the treatment of tuberculosis.
Multiple organisms may be involved in some infections. For example, peritoneal infections may be caused by several pathogens (eg, anaerobes and coliforms); a combination of drugs may be required to achieve coverage. Skin infections are often due to mixed bacterial, fungal, or viral pathogens.
To Achieve Synergistic Effects
The use of a drug combination against a specific pathogen may result in an effect greater than that achieved with a single drug. Examples include the use of penicillins with gentamicin in enterococcal endocarditis, the use of an extended-spectrum penicillin plus an aminoglycoside in Pseudomonas aeruginosa infections, and the combined use of amphotericin B and flucytosine in cryptococcal meningitis. Antibiotic combinations are also commonly used in the management of infections resulting from S epidermidis and penicillin-resistant pneumococci (eg, vancomycin plus rifampin). Several mechanisms, discussed next, may account for synergism.
The combined use of drugs may cause inhibition of 2 or more steps in a metabolic pathway. For example, trimethoprim and sulfamethoxazole (TMP-SMZ) block different steps in the formation of tetrahydrofolic acid.
Blockade of Drug-Inactivating Enzymes
Clavulanic acid, sulbactam, and tazobactam inhibit penicillinases and are often used along with penicillinase-sensitive beta-lactam drugs.
Increased permeability to aminoglycosides after exposure of certain bacteria to cell wall-inhibiting antimicrobials (eg, beta-lactams) is thought to underlie some synergistic effects.
The general principles of antimicrobial chemoprophylaxis can be summarized as follows: (1) Prophylaxis should always be directed toward a specific pathogen; (2) no resistance should develop during the period of drug use; (3) prophylactic drug use should be of limited duration; (4) conventional therapeutic doses should be used; and (5) prophylaxis should be used only in situations of documented drug efficacy.
Nonsurgical prophylaxis includes the prevention of cytomegalovirus (CMV), herpesvirus (HSV) infections, HIV infections in health care workers, influenza, malaria, meningococcal infections, and tuberculosis. In patients with AIDS, prophylactic measures are directed toward prevention of Pneumocystis jiroveci pneumonia (PCP) and toxoplasmosis. Though somewhat less effective, antimicrobial prophylaxis is also commonly used for animal or human bite wounds and chronic bronchitis. Severely leukopenic patients are often given prophylactic antibiotics.
Prophylaxis against postsurgical infections should be limited to procedures that are associated with infection in more than 5% of untreated cases under optimal conditions. Prophylaxis should embody the principles listed previously, with drug selection based on the most likely infecting organism and treatment initiated just before surgery and continued throughout the procedure. A first-generation cephalosporin (eg, cefazolin) is often the prophylactic drug of choice. Cefoxitin or cefotetan may be used for surgical patients at risk for infection caused by anaerobic bacteria. Situations in which surgical prophylaxis is of benefit (or commonly used) include gastrointestinal procedures, vaginal hysterectomy, cesarean section delivery, joint replacement, open fracture surgery, and dental procedures in patients with valvular disease or prostheses.