Catherine M. Oliphant and Karl Madaras-Kelly
LEARNING OBJECTIVES
Upon completion of the chapter, the reader will be able to:
1. Recognize that antimicrobial resistance is an inevitable consequence of antimicrobial therapy.
2. Describe how antimicrobials differ from other drug classes in terms of their effects on individual patients as well as on society as a whole.
3. Identify two guiding principles to consider when treating patients with antimicrobials, and apply these principles in patient care.
4. Differentiate between microbial colonization and infection based on patient history, physical examination, and laboratory and culture results.
5. Evaluate and apply at least six major drug-specific considerations when selecting antimicrobial therapy.
6. Evaluate and apply at least seven major patient-specific considerations when selecting antimicrobial therapy.
7. Select empirical antimicrobial therapy based on spectrum-of-activity considerations that provide a measured response proportional to the severity of illness. Provide a rationale for why a measured response in antimicrobial selection is appropriate.
8. Identify and apply five major principles of patient education and monitoring response to antimicrobial therapy.
9. Identify two common causes of patients failing to improve while on antimicrobials, and recognize other less common but potential reasons for antimicrobial failure.
KEY CONCEPTS
An inevitable consequence of exposing microbes to antimicrobials is that some organisms will develop resistance to the antimicrobial.
Antimicrobials are different from other classes of pharmaceuticals because they exert their action on bacteria infecting the host as opposed to acting directly on the host.
Two guiding principles to consider when treating patients with antimicrobials are (a) make the diagnosis and (b) do no harm!
Only bacteria that cause disease should be targeted with antimicrobial therapy, and colonizing flora should be left intact whenever possible.
Bacterial cultures should be obtained prior to antimicrobial therapy in patients with a systemic inflammatory response, risk factors for antimicrobial resistance, or infections where diagnosis or antimicrobial susceptibility is uncertain.
Drug-specific considerations in antimicrobial selection include the spectrum of activity, effects on nontargeted microbial flora, appropriate dose, pharmacokinetic and pharmacodynamic properties, adverse-effect and drug-interaction profile, and cost.
Empirical therapy should be based on patient- and antimicrobial-specific factors such as the anatomic location of the infection, the likely pathogens associated with the presentation, the potential for adverse effects, and the antimicrobial spectrum of activity.
Key patient-specific considerations in antimicrobial selection include recent previous antimicrobial exposures, identification of the anatomic location of infection through physical examination and diagnostic imaging, history of drug allergies, organ dysfunction that may affect drug clearance, immunosuppression, pregnancy, and compliance.
Patient education, de-escalation of antimicrobial therapy based on culture results, monitoring for clinical response and adverse effects, and appropriate duration of therapy are important treatment components.
Inadequate diagnosis resulting in poor initial antimicrobial selection, poor source control, or the development of a new infection with a resistant organism are relatively common causes of antimicrobial failure.
The discovery of antimicrobials is among the greatest medical achievements of the 20th century. Prior to the antimicrobial era, patients who contracted common infectious diseases developed significant morbidity or perished. The discovery of penicillin in 1927, followed by the subsequent discovery of other antimicrobials, contributed to a significant decline in infectious disease-related mortality during the next five decades. However, since 1980, infectious diseases related mortality in the United States has begun to increase, in part owing to increases in antimicrobial resistance.
The discovery of virtually every new class of antimicrobials has occurred in response to the development of bacterial resistance and loss of clinical effectiveness to existing antimicrobials. 
An inevitable consequence of exposing microbes to antimicrobials is that some organisms will develop resistance to the antimicrobial. Today, there are dozens of antimicrobial classes and hundreds of antimicrobials available for clinical use. However, in many cases, differences in mechanisms of action between antimicrobials are minor, and the microbiologic properties of the agents are similar. Antimicrobials are different from other classes of pharmaceuticals because they exert their action on bacteria infecting the host as opposed to acting directly on the host. Because use of an antimicrobial in one patient affects not only that patient but also other patients if they become infected with resistant bacteria, correct selection, use, and monitoring of clinical response are paramount.
There are two guiding principles to consider when treating patients with antimicrobials: (a) make the correct diagnosis and (b) do no harm! Patients with infections frequently present with signs and symptoms that are nonspecific for infection and may be confused with other noninfectious disease. Not only is it important to determine if a disease process is of infectious origin, but it is also important to determine the specific causative pathogen of the infection. Antimicrobials vary in their ability to inhibit or kill different species of bacteria, or their spectrum of activity. Antimicrobials that kill many different species of bacteria are called broad-spectrum antimicrobials, whereas antimicrobials that kill only a few different species of bacteria are called narrow-spectrum antimicrobials. One might argue that treating everybody with very broad antimicrobial coverage will increase the likelihood that a patient will get better without making a definitive diagnosis. However, counter to this argument is the principle of “Do no harm!” Very broad antimicrobial coverage does increase the likelihood of empirically targeting a causative pathogen; unfortunately, the development of secondary infections caused by selection of antimicrobial-resistant nontargeted pathogens is a common problem. In addition, adverse events are thought to complicate up to 10% of all antimicrobial therapy, and for select agents, the adverse-event rates are similar to classical high-risk medications such as warfarin, digoxin, or insulin.1 Therefore, the overall goal of antimicrobial therapy should be to cure the patient’s infection; limit harm by minimizing patient risk for adverse effects, including secondary infections; and limit societal risk from antimicrobial-resistant bacteria.
EPIDEMIOLOGY AND ETIOLOGY
Infectious disease–related illnesses, particularly respiratory tract infections, are among the most common reasons patients seek medical care.2 Approximately two-thirds of outpatient antimicrobial use is prescribed for respiratory tract infections, and the Centers for Disease Control and Prevention (CDC) estimate that one-third to one-half of all outpatient antimicrobials are used inappropriately to treat nonbacterial processes.3 However, recent trends in prescribing suggest a modest reduction in antimicrobial use for these infections, suggesting an increased recognition of the negative consequences of antimicrobial use.4 Prescription of antimicrobials in hospitalized patients is also common because up to one-half of all patients receive at least one antimicrobial during hospitalization. In addition, the CDC estimates that almost 2 million nosocomial or hospital-acquired infections and 90,000 related deaths occur annually.5 Generally nosocomial infections tend be associated with more antimicrobial-resistant strains of bacteria. In recent years, there has been a shift in the etiology of some community-acquired infections. Increasingly, infections caused by antimicrobial-resistant pathogens, traditionally nosocomial in origin, are being identified in ambulatory care settings. Reasons for this change include an aging populace, improvement in the management of chronic comorbid conditions including immunosuppressive conditions, and increases in outpatient management of more debilitated patients. The majority of infections caused by antimicrobial-resistant pathogens in the ambulatory care setting have had recent exposure to some aspect of the health care system, therefore are defined as health care–associated infections. The converging bacterial etiologies and increasing resistance in all health care environments emphasize the need to “make the diagnosis.”
PATHOPHYSIOLOGY
Normal Flora and Endogenous Infection
Many areas of the human body are colonized with bacteria—this is known as normal flora. Infections often arise from one’s own normal flora (also called an endogenous infection). Endogenous infection may occur when there are alterations in the normal flora (e.g., recent antimicrobial use may allow for overgrowth of other normal flora) or disruption of host defenses (e.g., a break or entry in the skin). Knowing what organisms reside where can help to guide empirical antimicrobial therapy (Fig. 69–1). In addition, it is beneficial to know what anatomic sites are normally sterile. These include the cerebrospinal fluid, blood, and urine.
Determining Colonization Versus Infection
Infection refers to the presence of bacteria that are causing disease (e.g., the organisms are found in normally sterile anatomic sites or in nonsterile sites with signs/symptoms of infection). Colonization refers to the presence of bacteria that are not causing disease. 
Only bacteria that cause disease should be targeted with antimicrobial therapy, and nondisease-producing colonizing flora should be left intact. It is important to differentiate infection from colonization because antimicrobial therapy targeting bacterial colonization is inappropriate and may lead to the development of resistant bacteria.
Exogenously Acquired Bacterial Infections
Infections acquired from an external source are referred to as exogenous infections. These infections may occur as a result of human-to-human transmission, contact with exogenous bacterial populations in the environment, and animal contact. Resistant pathogens such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus spp. (VRE) may colonize hospitalized patients or patients who access the health care system frequently. It is key to know which patients have acquired these organisms because patients generally become colonized prior to developing infection, and colonized patients should be placed in isolation (per infection-control policies) to minimize transmission to other patients.
Contrasting Bacterial Virulence and Resistance
Virulence refers to the pathogenicity or disease severity produced by an organism. Many bacteria may produce toxins or possess growth characteristics that contribute to their pathogenicity. Some virulence factors allow the organism to avoid the immune response of the host and cause significant disease. Virulence and resistance are different microbial characteristics. For example, Streptococcus pyogenes, a common cause of skin infections, produces toxins that can cause severe disease, yet it is very susceptible to penicillin. Enterococcus faecium is a highly resistant organism but is frequently a colonizing flora that causes disease primarily in the immunocompromised.
FIGURE 69–1. Normal flora and concentrations of bacteria (organisms per milliliter).
CLINICAL PRESENTATION AND DIAGNOSIS
Physical Examination
Findings on physical examination, along with the clinical presentation, can help to provide the anatomic location of the infection. Once the anatomic site is identified, the most probable pathogens associated with disease can be determined based on likely endogenous or exogenous flora. Fever often accompanies infection and is defined as a rise in body temperature above the normal 37°C (98.6°F). Oral and axillary temperatures may underestimate core temperature by at least 0.6°C (1°F), whereas rectal temperatures best approximate core temperatures. Fever is a host response to bacterial toxins. However, bacterial infections are not the sole cause of fever. Fever also may be caused by other infections (e.g., fungal or viral), medications (e.g., penicillins, cephalosporins, salicylates, and phenytoin), trauma, or other medical conditions (e.g., autoimmune disease, malignancy, pulmonary embolism, and hyperthyroidism). Some patients with infections may present with hypothermia (e.g., patients with overwhelming infection). Elderly patients may be afebrile, as may those with localized infections (e.g., urinary tract infection).6 For others, fever may be the only indication of infection. For example, neutropenic patients may not have the ability to mount normal immune responses to infection (e.g., infiltrate on chest x-ray, pyuria on urinalysis, and erythema or induration around catheter site), and the only finding may be fever.
Imaging Studies
Imaging studies also may help to identify anatomic localization of the infection. These studies usually are performed in conjunction with other tests to establish or rule out the presence of an infection. X-rays are performed commonly to establish the diagnosis of pneumonia, as well as the severity of disease (single versus multilobe involvement). CT scans are a type of x-ray that produces a three-dimensional image of the combination of soft tissue, bone, and blood vessels. In contrast, MRI use electromagnetic radio waves to produce two- or three-dimensional images of soft tissue and blood vessels with less detail of bony structures. MRI produce more detailed images of soft tissues and organs than CT scans, whereas CT scans produce more detailed images of bones.
Nonmicrobiologic Laboratory Studies
Common nonmicrobiological laboratory tests include the white blood cell count (WBC) and differential, erythrocyte sedimentation rate (ESR), and determination of the C-reactive protein level. In most cases, the WBC count is elevated in response to infection, but it may be decreased owing to overwhelming or long-standing infection. The differential is the percentage of each type of WBC (Table 69–1). In response to infection, neutrophils leave the bloodstream and enter the tissue to “fight” against the offending pathogens (i.e., leukocytosis). During an infection, immature neutrophils (e.g., bands) are released at an increased rate to help fight infection, leading to what is known as a bandemia or left shift. Therefore, a WBC count differential is key to determining if an infection is present. It is important to note that some patients may present with a normal total WBC with a left shift (e.g., the elderly). ESR and C-reactive protein (CRP) are nonspecific markers of inflammation. They increase as a result of the acute-phase reactant response, which is a response to inflammatory stimuli such as infection or tissue injury. These tests may be used as markers of infectious disease response because they are elevated when the disease is acutely active and usually fall in response to successful treatment. Clinicians may use these tests to monitor a patient’s response to therapy in osteomyelitis and infective endocarditis. These tests should not be used to diagnose infection because they may be elevated in noninfectious inflammatory conditions (e.g., rheumatoid arthritis, polymyalgia rheumatica, and temporal arteritis).
Patient Encounter 1
HPI: A 72-year-old man with a history of congestive heart failure, diabetes, hypertension, and hyperlipidemia presents to the local emergency room with complaints of increasing shortness of breath, cough productive of yellow-green sputum, chest pain, fever, and malaise. He was hospitalized 12 days ago for urosepsis, for which he received 10 days of levofloxacin.
PMH: Congestive heart failure, diabetes mellitus × 22 years, hypertension, hyperlipidemia
• Chronic renal insufficiency: baseline SCr 1.8 mg/dL (159 μmol/L)
FH: Father died of a myocardial infarction at age 75.
• Mother died of complications after cerebrovascular accident (CVA) at age 87
SH:
• Retired accountant
• Alcohol: two to three drinks per day
Allergies: NKDA
Meds:
• Enalapril 10 mg twice daily
• Metoprolol XL 25 mg daily
• Furosemide 40 mg daily
• Insulin glargine (Lantus) 30 units SC at bedtime
• Glipizide 10 mg daily
• Atorvastatin 20 mg daily
• Aspirin 81 mg daily
• Home oxygen at 2 L/min
What information in the history supports an infectious etiology?
Is this patient at risk for resistant pathogens? Why?
Table 69–1 WBC and Differential
Microbiologic Studies
Microbiologic studies that allow for direct examination of a specimen (e.g., sputum, blood, or urine) also may aid in a presumptive diagnosis and give an indication of the characteristics of the infecting organism. Generally, microbial cultures are obtained with a Gram stain of the cultured material.
A Gram stain of collected specimens can give rapid information that can be applied immediately to patient care. A Gram stain is performed to identify if bacteria are present and to determine morphologic characteristics of bacteria (such as gram-positive or gram-negative or shape—cocci, bacilli). Certain specimens do not stain well or at all and must be identified by alternative staining techniques (Mycoplasmaspp., Legionella spp., Mycobacteriumspp.). Figure 69–2 identifies bacterial pathogens as classified by Gram stain and morphologic characteristics. The presence of WBCs on a Gram stain indicates inflammation and suggests that the identified bacteria are pathogenic. The Gram stain may be useful in judging a sputum specimen’s adequacy. For example, the presence of epithelial cells on sputum Gram stain suggests that the specimen is either poorly collected or contaminated. A poor specimen can give misleading information regarding the underlying pathogen and is a waste of laboratory personnel time and patient cost.
Clinical Presentation of Antimicrobial Regimen Selection
• Review of symptoms consistent with an infectious etiology?
• Signs and symptoms may be nonspecific (e.g., fever) or specific.
• Specific signs and symptoms are beyond the scope of this chapter (see disease state-specific chapters for these findings).
Patient History
• History of present illness
• Comorbidities
• Current medications
• Allergies
• Previous antibiotic exposure (may provide clues as to colonization or infection with new specific pathogens or pathogens that may be resistant to certain antimicrobials)
• Previous hospitalization or health care utilization (also a key determinant in selecting therapy because the patient may be at risk for specific pathogens and/or resistant pathogens)
• Travel history
• Social history
• Pet/animal exposure
• Occupational exposure
• Environmental exposure
Physical Findings
• Findings consistent with an infectious etiology?
• Vital signs
• Body system abnormalities (e.g., rales, altered mental status, localized inflammation, erythema, warmth, edema, pain, and pus)
Diagnostic Imaging
• Radiographs (x-rays)
• CT scans
• MRI
• Labeled leukocyte scans
Nonmicrobiologic Laboratory Studies
• White blood cell count (WBC) with differential
• Erythrocyte sedimentation rate (ESR)
• C-reactive protein
Microbiologic Studies
• Gram stain
• Culture and susceptibility testing
FIGURE 69–2. Important bacterial pathogens classified according to Gram stain and morphologic characteristics. (From Rybak MJ, Aeschlimann JR. Laboratory tests to direct antimicrobial pharmacotherapy. In: In DiPiro JT, Talbert RL, Yee GC, et al., eds. Pharmacotherapy: A Pathophysiologic Approach. 6th ed. New York: McGraw-Hill; 2005:1894.)
FIGURE 69–3. Macrotube minimal inhibitory concentration (MIC) determination. The growth control (C), 0.5 mg/dL, and 1 mg/dL tubes are visibly turgid, indicating bacterial growth. The MIC is read as the first clear test tube (2 mg/dL). (From Rybak MJ, Aeschlimann JR. Laboratory tests to direct antimicrobial pharmacotherapy. In: In DiPiro JT, Talbert RL, Yee GC, et al., eds. Pharmacotherapy: A Pathophysiologic Approach. 6th ed. New York: McGraw-Hill; 2005:1897.)
Culture and susceptibility testing provides additional information to the clinician to guide appropriate therapy. Specimens are placed in or on culture media that provide the proper growth conditions. Once the bacteria grow on culture media, they can be identified through a variety of biochemical tests. Once a pathogen is identified, susceptibility tests can be performed to various antimicrobial agents. The minimum inhibitory concentration (MIC) is a standard susceptibility test. The MIC is the lowest concentration of antimicrobial that inhibits visible bacterial growth after approximately 24 hours (Fig. 69–3). Breakpoint and MIC values determine if the organism is susceptible (S), intermediate (I), or resistant (R) to an antimicrobial. The breakpoint is the concentration of the antimicrobial that can be achieved in the serum after a normal or standard dose of that antimicrobial. If the MIC is below the breakpoint, the organism is considered to be susceptible to that agent. If the MIC is above the breakpoint, the organism is said to be resistant. Reported culture and susceptibility results may not provide MIC values but report the S, I, and R results.
In general, bacterial cultures should be obtained prior to initiating antimicrobial therapy in patients with a systemic inflammatory response, risk factors for antimicrobial resistance, or infections where diagnosis or antimicrobial susceptibility is uncertain. The decision to culture depends on the sensitivity and specificity of the physical findings, diagnostic examination findings, and whether or not the pathogens are readily predictable. Culture and susceptibility testing usually is not warranted in a young, otherwise healthy woman who presents with signs and symptoms consistent with a urinary tract infection (UTI) because the primary pathogen, Escherichia coli, is readily predictable. Cultures and susceptibility testing are routine for sterile-site specimens (e.g., blood and spinal fluid), as well as for material presumed to be infected (e.g., material obtained from joints and abscesses). Cultures need to be interpreted with caution. Poor specimen collection technique and processing speed can result in misleading information and inappropriate use of antimicrobials.
Patient Encounter 2: Review of Symptoms, Physical Examination, and Laboratory Data
ROS: Patient with malaise, wheezing, dyspnea, cough, chest pain, and chills. No reports of emesis or diarrhea but with decreased appetite.
PE:
• VS: BP 160/88, P 84, RR 28, T 39.1 °C (102. 4°F), O2 sat 86%, Ht 6 ft, 0 in. (183 cm), Wt 80 kg (176 lb)
• HEENT: Dry mucous membranes
• Chest: Rales and rhonchi R greater than L; diminished breath sounds RML and RLL
• CV: Tachycardic with regular rhythm; normal heart sounds
Labs:
• WBC 18.8 × 103/mm3 (18.8 × 109/L), segs 80%, bands 10%, lymphs 10%
• SCr 2. 5 mg/dL (221 μmol/L)
• Glucose 322 mg/dL (17.9 mmol/L)
• Sputum Gram stain: less than 10 epithelial cells, greater than 25 WBCs, predominance of gram-negative bacilli
CXR: Pulmonary infiltrate right middle and lower lobes of right lung
What findings on physical examination are suggestive of an infectious process?
What laboratory findings and/or diagnostic studies have been performed to help establish the presence of an infection?
Are the findings of these laboratory and diagnostic studies suggestive of an infection?
What is your working diagnosis based on this patient encounter?
TREATMENT
General Approach to Treatment, Including Nonantimicrobial Treatment
While selection of antimicrobial therapy may be a major consideration in treating infectious diseases, it may not be the only therapeutic intervention. Other important therapies may include adequate hydration, ventilatory support, and other supportive medications. In addition, antimicrobials are unlikely to be effective if the process or source that leads to the infection is not controlled. Source control refers to this process and may involve removal of prosthetic materials such as catheters and infected tissue or drainage of an abscess. Source control considerations should be a fundamental component of any infectious diseases treatment. It is also important to recognize that there may be many different antimicrobial regimens that may cure the patient. While the following therapy sections provide factors to consider when selecting antimicrobial regimens, an excellent and more in-depth resource for selecting antimicrobial regimens for a variety of infectious diseases is the Infectious Diseases Society of America Guidelines.7
Antimicrobial Considerations in Selecting Therapy
Drug-specific considerations in antimicrobial selection include spectrum of activity, effects on nontargeted microbial flora, appropriate dose, pharmacokinetic and pharmacodynamic properties, adverse-effect and drug-interaction profile, and cost (Table 69-2).
Spectrum of Activity and Effects on Nontargeted Flora
Most initial antimicrobial therapy is empirical because cultures usually have not had sufficient time to identify a pathogen. 
Empirical therapy should be based on patient-and antimicrobial-specific factors such as the anatomic location of the infection, the likely pathogens associated with the presentation, the potential for adverse effects in a given patient, and the antimicrobial spectrum of activity. Prompt initiation of appropriate therapy is paramount in hospitalized patients who are critically ill. Patients who receive initial antimicrobial therapy that provides coverage against the causative pathogen survive at twice the rate of patients who do not receive adequate therapy initially.8Empirical selection of antimicrobial spectrum of activity should be related to the severity of the illness. Generally, acutely ill patients may require broader-spectrum antimicrobial coverage, whereas less ill patients may be managed initially with narrow-spectrum therapy. While a detailed description of antimicrobial pathogen-specific spectrum of activity is beyond the scope of this chapter, this information can be obtained readily from a number of sources.9,10
Table 69–2 Considerations for Selecting Antimicrobial Regimens
Collateral damage is defined as the development of resistance occurring in a patient’s nontargeted antimicrobial flora that can cause secondary infections. For example, clindamycin is an excellent antimicrobial for treating streptococcal infections. However, many antimicrobials can treat this relatively susceptible pathogen. Clindamycin also readily selects for resistance in a nontargeted organism that may be present in the intestinal tract, namely, Clostridium difficile. Collateral damage is manifested because clindamycin is considered to be a major risk factor for C. difficile–associated diarrhea.11 If several different antimicrobials possess activity against a targeted pathogen, the antimicrobial that is least likely to be associated with collateral damage may be preferred.
Single Versus Combination Therapy
A common subject of debate involves the need to provide similar bacterial coverage with two antimicrobials for serious infections. Proponents state that double coverage may be synergistic, prevent the emergence of resistance, and improve outcome. However, there are few clinical examples in the literature to support these assertions. Examples where double coverage is considered superior are limited to infections associated with large bacterial inocula and in species that are known to readily develop resistance such as active tuberculosis or enterococcal endocarditis.12,13 A study of patients with Pseudomonas aeruginosa infections, an intrinsically resistant organism, demonstrated that empirical double coverage with two antipseudomonal antimicrobials improved survival.14 The analysis found that combination therapy increased the likelihood of appropriate empirical coverage; however, once organism susceptibilities were known, there was no difference in outcome between double coverage and monotherapy. Double antimicrobial coverage with similar spectra of activity may be beneficial for selected infections associated with high bacterial loads or for initial empirical coverage of critically ill patients in whom antimicrobial-resistant organisms are suspected. Monotherapy usually is satisfactory once antimicrobial susceptibilities are established.
Antimicrobial Dose
Clinicians should be aware that dosage regimens with the same drug maybe different depending on the infectious process. For example, ciprofloxacin, a fluoroquinolone, has various dosage regimens based on site of infection. The dosing for uncomplicated UTIs is 250 mg twice daily for 3 days. For complicated UTIs, the dose is 500 mg twice daily for 7 to 14 days. Severe complicated pneumonia requires a dosage regimen of 750 mg twice daily for 7 to 14 days. Clinicians are encouraged to use dosing regimens designed for treatment of the specific diagnosed infection because they have demonstrated proven efficacy and are most likely to minimize harm.
Pharmacokinetic Properties
Pharmacokinetic properties of an antimicrobial may be important in antimicrobial regimens. Pharmacokinetics refers to a mathematical method of describing a patient’s drug exposure in vivo in terms of absorption, distribution, metabolism, and elimination. Bioavailability refers to the amount of antimicrobial that is absorbed orally relative to an equivalent dose administered intravenously. Drug-related factors that may affect oral bioavailability include the salt formulation of the antimicrobial, the dosage form, and the stability of the drug in the gastrointestinal tract. Frequently, absorption may be affected by gastrointestinal tract blood flow. All patients that manifest systemic signs of infection such as hypotension or hypoperfusion should receive intravenous antimicrobials to ensure drug delivery. In almost all cases where patients have a functioning gastrointestinal tract and are not hypotensive, antimicrobials with almost complete bioavailability (greater than 80%) such as the fluoroquinolones, fluconazole, and linezolid may be given orally. With antimicrobials with modest bioavailability (e.g., many β-lactams), the decision to choose an oral product will depend more on the severity of the illness and the anatomic location of the infection. In sequestered infections, where higher systemic concentrations of antimicrobial may be necessary to reach the infected source (e.g., meningitis) or for antimicrobials with poor bioavailability, intravenous formulations should be used.
Several points regarding how the antimicrobial distributes into tissue are worth mentioning. First, only antimicrobials not bound to albumin are biologically active. Protein binding is likely clinically irrelevant in antimicrobials with low or intermediate protein binding. However, highly protein bound antimicrobials (greater than 50%) also may not be able to penetrate sequestered compartments, such as cerebral spinal fluid, resulting in insufficient concentration to inhibit bacteria. Second, some drugs may not achieve sufficient concentrations in specific compartments based on distribution characteristics. For example, Legionella pneumophilia is a nonenteric gram-negative organism that causes severe pneumonia. The organism is known to survive and reside inside pulmonary macrophages. Treatment with an antibiotic that works by inhibiting bacterial cell wall synthesis, such as a cephalosporin, will be ineffective because it only distributes into extracellular host tissues. However, macrolide or fluoroquinolone antimicrobials, which concentrate in human pulmonary macrophages, are highly effective against pneumonia caused by this organism.
Many antimicrobials undergo some degree of metabolism once ingested. Metabolism may occur via hepatic, renal, or nonorgan-specific enzymatic processes. The route of elimination of the metabolic pathway may be exploited for infections associated with tissues related to the metabolic pathways. For example, many fluoroquinolone antimicrobials are metabolized only in part and undergo renal elimination. Urinary concentrations of active drug are many times those achieved in the systemic circulation, making several of these agents good choices for UTIs.
Pharmacodynamic Properties
Pharmacodynamics describes the relationship between drug exposure and pharmacologic effect of antibacterial activity or human toxicology. Antimicrobials generally are categorized based on their concentration-related effects on bacteria. Concentration-dependent pharmacodynamic activity occurs where higher drug concentrations are associated with greater rates and extents of bacterial killing. Concentration-dependent antimicrobial activity is maximized when peak antimicrobial concentrations are high. In contrast, concentration-independent (or time-dependent) activity refers to a minimal increase in the rate or extent of bacterial killing with an increase in antimicrobial dose. Concentration-independent antimicrobial activity is maximized when these antimicrobials are dosed to maintain blood and/or tissue concentrations above the MIC in a time-dependent manner. Fluoroquinolones, aminoglycosides, and metronidazole are examples of antimicrobials that exhibit concentration-dependent activity, whereas β-lactam and glycopeptide antimicrobials exhibit concentration-independent activity. Pharmacodynamic properties have been optimized to develop new dosing strategies for older antimicrobials. Examples include single-daily-dose aminoglycoside or β-lactam therapy administered by continuous infusion. The product labeling for many new antimicrobials takes pharmacodynamic properties into account.
Antimicrobials also can be classified as possessing bactericidal or bacteriostatic activity in vitro. Bactericidal antibiotics generally kill at least 99.9% (3 log reduction) of a bacterial population, whereas bacteriostatic antibiotics possess antimicrobial activity but reduce bacterial load by less than 3 logs. Clinically, bactericidal antibiotics may be necessary to achieve success in infections such as endocarditis or meningitis. A full discussion of the application of antimicrobial pharmacodynamics is beyond the scope of this chapter, but excellent sources of information are available.15
Adverse-Effect and Drug-Interaction Properties
A major concern when selecting antimicrobial regimens should be the propensity for the regimen to cause adverse effects and the potential for interaction with other drugs. Patients may possess characteristics or risk factors that increase their likelihood of developing an adverse event, emphasizing the need to obtain a good patient medical history. In general, if several different antimicrobial options are available, antimicrobials with a low propensity to cause specific adverse events should be selected, particularly for patients with risk factors for a particular complication. Risk factors for adverse events may include the coadministration of other drugs that are associated with a similar type of adverse event. For example, coadministration of the known nephrotoxin gentamicin with vancomycin increases the risk for nephrotoxicity compared with administration of either drug alone.16 Other drug interactions may predispose the patient to dose-related toxicity through inhibition of drug metabolism. For example, erythromycin has the potential to prolong cardiac QT intervals in a dose-dependent manner, potentially increasing the risk for sudden cardiac death. A cohort study of patients taking oral erythromycin found that patients with concomitantly prescribed medications that inhibited the metabolism of erythromycin exhibited a fivefold increase in cardiac death versus controls.17
Antimicrobial Cost
A final consideration in selecting antimicrobial therapy relates to cost. It is important to remember that the most inexpensive antimicrobial is not necessarily the most cost-effective antimicrobial. Antimicrobial costs constitute a relatively small portion of the overall cost of care. Frequently, regulatory studies are not designed to identify differences in hospital length of stay, less common adverse events, monitoring costs, collateral damage, or antimicrobial-specific resistance issues, all of which may contribute to medical costs. Careful consideration of antimicrobial microbiologic, pharmacologic, and patient-related factors such as compliance and a variety of clinical outcomes is necessary to establish the cost versus benefit of an antimicrobial in a given patient. If there is no difference or a small difference in these factors, the least costly antimicrobial may be the best choice.
Patient Considerations in Antimicrobial Selection
Key patient-specific considerations in antimicrobial selection include recent previous antimicrobial exposures, identification of the anatomic location of infection through physical examination and diagnostic imaging, history of drug allergies, pregnancy or breast-feeding status, organ dysfunction that may affect drug clearance, immunosuppression, compliance, and the severity of illness (see Table 69–2).
Host Factors
Host factors can help to ensure selection of the most appropriate antimicrobial agent. Age is an important factor in antimicrobial selection. With regard to dose and interval, renal and hepatic function varies with age. Populations with diminished renal function include neonates and the elderly. Hepatic function in the neonate is not fully developed, and drugs that are metabolized or eliminated by this route may produce adverse effects. For example, sulfonamides and ceftriaxone may compete with bilirubin for binding sites and may result in hyperbilirubinemia and kernicterus. Gastric acidity also depends on age; the elderly and children younger than 3 years of age tend to be achlorhydric. Drugs that need an acidic environment (e.g., ketoconazole) are not well absorbed, and those whose absorption is enhanced in an alkaline environment will have increased concentrations (e.g., penicillin G).
Disruption of host defenses owing to IV catheters, indwelling Foley catheters, burns, trauma, surgery, and increased gastric pH (secondary to antacids, H2 blockers, and proton pump inhibitors) may place patients at higher risk for infection. Breaks in and entry into the skin provide a route for infection because the natural barrier of the skin is disrupted. Increased gastric pH can allow for bacterial overgrowth and has been associated with an increased risk of pneumonia.18
Recognizing the presumed site of infection and most common pathogens associated with the infectious source should guide antimicrobial choice, dose, and route of administration. For example, community-acquired pneumonia is caused most commonly by S. pneumoniae, E. coli is the primary cause of uncomplicated UTIs, and staphylococci and streptococci are implicated most frequently in skin and skin-structure infections (e.g., cellulitis).
Patients with a history of recent antimicrobial use may have altered normal flora or harbor resistant organisms. If a patient develops a new infection while on therapy, fails therapy, or has received antimicrobials recently, it is prudent to prescribe a different class of antimicrobial because resistance is likely. Previous hospitalization or health care utilization (e.g., residing in a nursing home, hemodialysis, and outpatient antimicrobial therapy) are risk factors for the acquisition of nosocomial pathogens, which are often resistant organisms.
Antimicrobial allergies are some of the most common drug-related allergies reported and have significant potential to cause adverse events. In particular, penicillin-related allergy is common and can be problematic because there is an approximately 4% cross-reactivity with cephalosporins as well as carbapenems.19,20 In general, a patient’s medical history should be reviewed to determine the offending β-lactam and nature of the allergic reaction. In some cases, patients with mild or nonimmunologic reactions may receive a β-lactam antimicrobial with low cross-reactive potential. However, patients with a history of physical findings consistent with IgE-mediated reactions such as anaphylaxis, urticaria, or bronchospasm should not be administered any type of β-lactam antimicrobial, including cephalosporins, unless there are no other alternatives. Administration of potentially cross-reactive agents in this situation should occur only under controlled conditions, and some patients may need to undergo desensitization. If the specific medical history relating to a reported allergy cannot be obtained, the patient should be assumed to have had an IgE-mediated reaction and should be managed in a similar manner.
Renal and/or hepatic function should be considered in every patient prior to initiation of antimicrobial therapy. In general, most antimicrobials undergo renal elimination and exhibit decreased clearance with diminished renal function, and dosing adjustments are found readily in the literature.21 In contrast, dosing adjustments for antimicrobials that are not eliminated renally are less well documented. Failure to adjust the antimicrobial dose or interval may result in drug accumulation and an increase in adverse effects.
Concomitant administration of other medications may influence the selection of the antimicrobial, dose, and monitoring. Medications that are commonly associated with drug interactions include, but are not limited to, warfarin, rifampin, phenytoin, digoxin, theophylline, multivalent cations (e.g., calcium, magnesium, and zinc), and sucralfate. Drug interactions between antimicrobials and other medications may occur via the cytochrome P-450 system, protein-binding displacement, and alteration of vitamin K-producing bacteria. Interactions may result in increased concentrations of one or both agents, increasing the risk of adverse effects or additive toxicity. A key consideration in selecting antimicrobial regimens starts with obtaining a good patient medical and drug history, recognizing drug-specific adverse-event characteristics, and anticipating potential problems proactively. If it is necessary to use an antimicrobial with a relatively high incidence of adverse effects, informing patients of the risks and benefits of therapy, as well as what to do if an adverse effect occurs, may improve patient compliance and may facilitate patient safety.
Patient Encounter 3: Empirical Selection of Antibiotics
Based on the information presented, select an empirical antimicrobial regimen for this patient. Your plan should include
(a) a tentative infectious diagnosis or source, including likely pathogens or resistant organisms;
(b) a specific antimicrobial(s) regimen, including drug(s), dose, and route of administration;
(c) description of any ancillary treatments; and
(d) a rationale for your empirical antimicrobial selection based on drug- and patient-specific considerations.
Antimicrobial agents must be used with caution in pregnant and nursing women. Some agents pose potential threats to the fetus or infant (e.g., quinolones, tetracyclines, and sulfonamides). For some agents, avoidance during a specific trimester of pregnancy is warranted (e.g., trimethoprim/sulfamethoxazole). Pharmacokinetic variables also are altered during pregnancy. Both the clearance and volume of distribution are increased during pregnancy. As a result, increased dosages and/or more frequent administration of certain drugs may be required to achieve adequate concentrations. This information can be obtained from a number of sources.22,23
Adherence is essential to ensure efficacy of a particular agent. Patients may stop taking their antibiotics once the symptoms subside and save them for a “future” infection. If the patient does not complete the course of therapy, the infection may not be eradicated, and resistance may emerge. Self-medication of saved antibiotics may be inappropriate and harmful and may select for resistant organisms. Poor patient adherence may be due to adverse effects, tolerability, cost, and lack of patient education.
OUTCOME EVALUATION
Figure 69–4 provides an overview of patient- and antimicrobial agent-specific factors to consider when selecting an antimicrobial regimen. It further delineates monitoring of therapy and actions to take depending on the patient’s response to therapy. The duration of therapy depends on patient response and type of infection being treated.
Modifying Empirical Therapy Based on Cultures and Clinical Response
If a successful clinical response occurs and culture results are available, therapy should be de-escalated. De-escalation refers to decreasing antimicrobial regimen spectrum of activity to provide coverage against specific antimicrobial-sensitive pathogens recovered from culture. The purpose of de-escalation therapy is to minimize the likelihood of secondary infections owing to antimicrobial-resistant organisms. In cases where a specific organism is recovered that has a known preferred agent of choice, therapy might be changed to that specific agent. For example, antistaphylococcal penicillins are considered to be the agents of choice for methicillin-susceptible S. aureus owing to their bactericidal activity and narrow-spectrum activity and may be preferable to other antibiotic regimens. In other cases, empirical coverage might be discontinued if a specific suspected pathogen is excluded by culture or an alternative, noninfectious diagnosis is established. In addition, intravenous antimicrobials frequently are more expensive than oral therapy. Therefore, it is desirable to convert therapy to oral antimicrobials with a comparable antimicrobial spectrum or specific pathogen sensitivity as soon as the patient improves clinically.24
Failure of Antimicrobial Therapy
While many infections respond readily to antimicrobials, some infections do not. A relatively common question when a patient’s condition fails to improve relates to whether the antimicrobial therapy has failed? Changing antimicrobials generally is one of the easiest interventions relative to other options. However, it is important to remember that antimicrobial therapy comprises only a portion of the overall disease treatment, and there may be many factors that contribute to a lack of improvement. In general, inadequate diagnosis resulting in poor initial antimicrobial or other nonantibiotic drug selection, poor source control, or the development of a new infection with a resistant organism are relatively common causes of antimicrobial failure. An infection-related diagnosis may be difficult to establish and generally has two components: (a) differentiating infection from noninfection-related disease and (b) providing adequate empirical spectrum of activity if the cause is infectious. Failure of improvement in a patient’s condition should warrant broadening the differential diagnosis to include noninfection-related causes, as well as considering other potential infectious sources and/or pathogenic organisms. Another common cause of failure is poor source control. A diagnostic search for unknown sources of infection and removal of indwelling devices in the infected environment or surgical drainage of abscesses should be undertaken if the patient’s condition is not improving. Less common but still frequent causes of therapeutic failure include the development of secondary infections. In this case, the patient generally improves, but then develops a new infection caused by an antimicrobial resistant pathogen and relapses. The emergence of resistance to a targeted pathogen while on antimicrobial therapy can be associated with clinical failure but usually is limited to tuberculosis, pseudomonads, or other gram-negative enterics. Drug- and patient-specific factors such as appropriate dosing, patient compliance, and drug interactions can be associated with therapeutic failure and also should be considered. A common assumption is that the correct diagnosis was made, but the patient was not treated long enough with antimicrobials. There are certain types of infections (e.g., endocarditis or osteomyelitis) where the standard of care is to treat for prolonged periods of time (i.e., weeks or months). However, the optimal duration of therapy for many infectious diseases is somewhat subjective. Recently, studies of several infectious processes have suggested that shorter durations of therapy can result in similar clinical outcomes as longer durations of therapy, frequently with fewer complications or secondary infections.25–27 The general trend has been to treat these disease processes with shortened courses of antibiotic therapy. In this period of extensive antimicrobial resistance, clinicians should keep abreast of changing recommendations emphasizing shorter durations of therapy.
FIGURE 69–4. Approach to selection of antimicrobial therapy.
Patient Encounter 4: Patient Care and Monitoring
Update: The patient was admitted to the hospital with a presumptive diagnosis of health care-associated pneumonia (based on the recent hospitalization). He received IV hydration with normal saline, 5 L oxygen via face mask, an insulin infusion to control his glucose, and empirical antimicrobial therapy with piperacillin-tazobactam 3.375 g IV every 6 hours and vancomycin 1 g IV every 48 hours. All other medications are continued with the exception of the diabetes medications.
After 48 hours of therapy, the following parameters are obtained:
PE
• VS: BP 145/82, P 77, RR 22, T 37.9°C (100.2°F), O2 sat 92% on 4 L
• Repeat CXR; Increased fluid density in bases, infiltrate unchanged, rales and rhonchi unchanged
Labs:
• WBC 13.2 × 103/mm3 (13.2 × 109/L)
• SCr 1.9 mg/dL (168 μmol/L)
• Glucose 181 mg/dL (10.0 mmol/L)
• Urine and blood cultures × 2: Negative
• Sputum culture: 3+ P. aeruginosa
Sensitivity Report:
Cefepime sensitive (MIC = 1)
Ceftazidime sensitive (MIC = 1)
Piperacillin/taz sensitive (MIC = 4)
Imipenem resistant (MIC greater than 64)
Gentamicin sensitive (MIC = 1)
Amikacin sensitive (MIC = 0.5)
Ciprofloxacin resistant (MIC greater than 4)
What information suggests improvement in the patient’s condition?
Do any of the antimicrobial doses need to be adjusted for changes in organ function?
Should antimicrobial therapy be modified based on the culture results?
Can the antimicrobial therapy be converted from IV to oral therapy?
Patient Care and Monitoring
After selection and initiation of antimicrobial regimen, there are a number of additional patient care and monitoring considerations that should be addressed to improve the likelihood of a successful outcome. Patient education, de-escalation of antimicrobial therapy based on culture results, monitoring for clinical response and adverse effects, and appropriate duration of therapy are important.
Patient Education
Provide patient education with regard to appropriate use of antimicrobials (e.g., dose, interval), adverse effects, and drug interactions (which may play a role in therapy failure and increased toxicity).
Therapeutic Monitoring
Monitor for therapeutic response by assessing efficacy and toxicity of the antimicrobial regimen.
• Clinical presentation/physical findings
• Efficacy: Assess vital signs (monitor for return to normal or lack of altered findings; e.g., fever), physical examination findings, patient’s subjective impression
• Toxicity: Monitor and assess for adverse effects and evaluate antimicrobial serum concentrations when appropriate to minimize toxicity and improve outcomes
• Diagnostic imaging: Diagnostic testing is disease state-dependent
• Laboratory data
• Monitor and assess laboratory data: WBC with differential (goal is a reduction in WBC if elevated initially and resolution of left shift), renal and/or hepatic function (consider need for dosage adjustments), other labs as indicated (e.g., ESR, CRP)
• Follow-up on culture and susceptibility reports with subsequent de-escalation of therapy, if possible
• Reculture of specimens is not performed routinely except in few cases (e.g., endocarditis) or where a secondary infection is suspected because data may be misleading and lead to the addition of broader or more powerful antimicrobials
CONCLUSION
Antimicrobial regimen selection is a complex process involving the integration of a multitude of factors. The guiding principles to make the diagnosis and do no harm must be considered when choosing an antimicrobial for a given patient. In summary, when infection is suspected, rapid and accurate diagnosis should be followed by early intervention that includes admi nistration of appropriately dosed antibiotics with appropriate empirical spectrums. De-escalation to suitable narrow-spectrum antibiotics if susceptibilities are known should occur as soon as possible, and therapy should be stopped as soon as the patient is cured. These fundamental actions improve infectious disease outcomes and minimize collateral damage and adverse effects.
Abbreviations Introdued in This Chapter
Self-assessment questions and answers are available at http://www.mhpharmacotherapy.com/pp.html.
REFERENCES
1. Shehab N, Patel PR, Srinivasan A, Budnitz DS. Emergency department visits for antibiotic-associated adverse events. Clin Infect Dis 2008;47:735–743.
2. National Vital Statistics Reports 2005;(53)17.
3. Centers for Disease Control Antimicrobial Resistance. http://www.cdc.gov/drugresistance/community/faqs.htm.
4. Vanderweil SG, Pelletier AJ, Hamedani AG, Gonzales R, Metlay JP, Camargo CA Jr. Declining antibiotic prescriptions for upper respiratory infections, 1993–2004. Acad Emerg Med 2007;14:366–369.
5. Burke J. Infection control—A problem for patient safety. N Engl J Med 2003;348:651–655.
6. Mackowiak PA. Temperature regulation and the pathogenesis of fever. In: Mandell GL, Bennett JE, Dolin R. Mandell, Douglas, Bennett’s, eds. Principles and Practice of Infectious Diseases. 5th ed. Philadelphia: Churchill Livingstone, 2000:604–620.
7. Practice Guidelines from the Infectious Diseases Society of America. http://www.journals.uchicago.edu/IDSA/guidelines/.
8. Kollef MH, Sherman G, Ward S, Fraser VJ. Inadequate antimicrobial treatment of infections: A risk factor for hospital mortality among critically ill patients. Chest 1999;115:462–474.
9. Antimicrobial spectra. In: Gilbert DN, Moellering RC, Eliopoulos GM, Sande MA, eds. The Sanford Guide to Antimicrobial Therapy 2008. 38th ed. Sperryville, VA: Antimicrobial Inc.
10. John Hopkins University Division of Infectious Diseases Antibiotic Guide. http://www.hopkins-abxguide.org/.
11. Johnson S, Samore MH, Farrow AF, et al. Epidemics of diarrhea caused by a clindamycin resistant strain of Clostridium difficile in four hospitals. N Engl J Med 1999;341:1645–1651.
12. Peloquin CA. Tuberculosis. In: DiPiro JT, Talbert RL, Yee GC, et al., eds. Pharmacotherapy: A Pathophysiological Approach. 6th ed. New York: McGraw-Hill, 2005:2015–2334.
13. Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis: diagnosis, antimicrobial therapy, and management of complications: a statement for healthcare professionals from the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease, Council on Cardiovascular Disease in the Young, and the Councils on Clinical Cardiology, Stroke, and Cardiovascular Surgery and Anesthesia, American Heart Association: endorsed by the Infectious Diseases Society of America. Circulation 2005;111:394–434.
14. Chamot E, Boffi E, Amari E, et al. Effectiveness of combination antimicrobial therapy for Pseudomonas aeruginosa bacteremia. Antimicrob Agents Chemother 2003;47:2756–2764.
15. Craig WA. Pharmacodynamics of antimicrobials. In: Nightingale CH, Murakawa T, Ambrose PG, eds. Antimicrobial Pharmacodynamics in Theory and Clinical Practice. 1st ed. New York: Marcel Dekker, 2002:1–22.
16. Rybak MJ. The pharmacokinetic and pharmacodynamic properties of vancomycin. Clin Infect Dis 2006;42(Suppl 1):S35–S39.
17. Ray WA, Murray KT, Meredith S, et al. Oral erythromycin and the risk of sudden death from cardiac causes. N Engl J Med 2004;351:1089–1096.
18. Laheij RJ, Sturkenboom MC, Hassing R, et al. Risk of community-acquired pneumonia and use of gastric acid-suppressing drugs. JAMA 2004;292:1955–1960.
19. Gruchalla RS, Pirmohamed M. Clinical practice. Antibiotic allergy. N Engl J Med 2006;354:601–609.
20. Weiss ME, Adkinson NF. B-lactam Allergy. In: Mandell GL, Bennett JE, Dolin R, eds. Principles and Practice of Infectious Diseases. 6th ed. Philadelphia, PA: Churchill Livingstone, 2005.
21. Frye RF, Matzke GR. Drug therapy individualization for patients with renal insufficiency. In: DiPiro JT, Talbert RL, Yee GC, et al., eds. Pharmacotherapy: A Pathophysiological Approach. 6th ed. New York: McGraw-Hill, 2005:919–933.
22. Briggs GG, Freeman RK, Yaffe SJ. Drugs in Pregnancy and Lactation. 7th ed. Philadelphia: Lippincott Williams & Wilkins, 2005.
23. Hale TW. Medications and Mothers’ Milk. 11th ed. Amarillo, TX: Pharmasoft, 2004.
24. Press RA. The use of fluoroquinolones as antiinfective transition-therapy agents in community-acquired pneumonia. Pharmacotherapy 2001;21(7 pt 2):100S–104S.
25. Goff DA. Short-duration therapy for respiratory tract infections. Ann Pharmacother 2004;38(Suppl 9):S19–S23.
26. Chastre J, Wolff M, Fagon JY, et al. For the PneumA Trial Group. Comparison of 8 vs 15 days of antibiotic therapy for ventilator-associated pneumonia in adults: A randomized trial. JAMA 2003;290:2588–2598.
27. Kollef MH, Napolitano LM, Solomkin JS, Wunderink RG, Bae IG, Fowler VG, Balk RA, Stevens DL, Rahal JJ, Shorr AF, Linden PK, Micek ST. Healthcare-associated infection (HAI): A critical appraisal of the emerging threat-proceedings of the HAI Summit. Clin Infect Dis 2008 Oct 1;2 (Suppl 47):S55–S99.