Hans Liu, MD

Community-acquired respiratory tract infections (CARTIs) are a reason for seeking medical attention. In 2001, there were 28.4 million office visits in the United States for an acute respiratory tract infection (excluding pharyngitis).¹

Management of CARTIs poses several challenges. According to the World Health Organization (WHO), “for every 100 respiratory infections, only 20% require antibiotic treatment”² —the remaining 80 infections most likely have a viral origin. Thus, antibacterial therapy should be avoided unless a bacterial cause has been confirmed or is deemed likely.

Once that determination has been made, clinicians need to separate patients who can be safely managed as outpatients from those who need to be hospitalized. Disease severity is, of course, an important consideration in this selection process.^3-8

For management of patients who will not be hospitalized, the WHO and the Society for Healthcare Epidemiology of America/Infectious Diseases Society of America (SHEA/IDSA) offer the 3 Ds: administer the correct drug, at the right dose, and for the appropriate duration, to minimize development and spread of resistance.^9,10 A recent consensus conference coordinated by the Primary Care Education Consortium and Texas Academy of Family Physicians elaborated on this mnemonic as it relates to outpatient management of 3 CARTIs: community-acquired pneumonia, acute bacterial exacerbations of chronic bronchitis, and acute bacterial rhinosinusitis. Its recommendations, which have been drawn largely from existing evidence-based guidelines, form the basis for this review.

Drug selection

Treatment with an antibacterial agent will not be medically warranted in the majority of patients with a CARTI. Many of these patients will, however, expect to go home with an antibiotic prescription. The primary care clinician can reduce such expectations and prevent unnecessary reconsultations by briefly addressing four issues: 1) the natural course of the viral illness, 2) the lack of effectiveness of antibiotics, 3) the problem of antibiotic resistance, and 4) the side effects of antibiotics.¹¹

When selecting an antibacterial agent for patients with pneumonia, bronchitis, or rhinosinusitis for which a bacterial cause has been identified or deemed likely, several factors need to be taken into account, including the suspected or identified pathogens, local resistance patterns, previous therapy, patient allergies, and the patient’s ability to tolerate treatment failure. Many of these factors are considered by professional organizations that regularly develop guidelines for CARTIs based on the best available evidence. Perhaps most critical for decision-making in the primary care setting is an understanding of evolving microbiology and resistance patterns.

Common pathogens

There is considerable overlap among pathogens commonly found in CARTIs. Streptococcus pneumoniae and Haemophilus influenzae are most often observed in the outpatient setting.

Community-acquired pneumonia. In outpatients with mild illness, S pneumoniae, Mycoplasma pneumoniae, Chlamydia species (particularly Chlamydia pneumoniae, now called Chlamydophilia pneumoniae), and H influenzae are the most common pathogens.^12,13 In patients younger than 50 years without significant comorbidity, Mycoplasma species are the most common pathogens. Older patients and those with significant comorbidity are more likely infected with S pneumoniae,¹³ a Gram-negative enteric bacillus, Pseudomonas aeruginosa, or Legionella.^5,14

Bacterial bronchitis. In addition to S pneumoniae and H influenzae, Moraxella catarrhalis is a frequent pathogen in bacterial exacerbations of chronic bronchitis.¹⁵P aeruginosa and other Gram-negative bacilli are also seen, especially in patients with a severe acute exacerbation who have a forced expiratory volume in 1 second (FEV₁) of 35% of predicted or less.¹⁶ Infection due to multiple pathogens occurs in a small percentage of all patients with chronic bronchitis, particularly those with severe exacerbations. Fewer than 10% of acute exacerbations are due to an atypical bacterium, usually C pneumoniae. M pneumoniae and Legionella pneumophila are implicated even less frequently.¹⁵

Bacterial rhinosinusitis.S pneumoniae and H influenzae also are frequent causes of acute bacterial rhinosinusitis. Other pathogens commonly seen in this condition include other Streptococcus species, M catarrhalis, oral anaerobes, Staphylococcus aureus in adults, and M catarrhalis, Streptococcus pyogenes, and anaerobes in children.¹⁷

Resistance

Antibiotic resistance is an important consideration in the management of CARTIs. There is little doubt that widespread use of antibiotics leads to in vitro bacterial resistance.^18-20 However, because clinical success has been observed in the presence of pathogens with low-level resistance, there is some debate as to whether low-level antibiotic resistance has a significant effect on clinical outcomes.^18,21-29 Even so, the US Centers for Disease Control and Prevention has determined that people who attend or work at child-care centers and those who recently used antimicrobial agents are at increased risk for infection with drug-resistant S pneumoniae.³⁰ Moreover, the WHO has stated that infection with resistant pathogens prolongs illness and increases the probability of a fatal outcome.³¹

Several surveillance programs that monitor antibiotic resistance patterns—including the Alexander Project³² and Tracking Resistance in the United States Today (TRUST)^33-36 —have confirmed widespread resistance to antibiotics commonly used to treat CARTIs in the United States. β-Lactam resistance due to penicillin-binding protein changes in S pneumoniae has increased significantly over the past decade. Generally, more than 30% of S pneumoniae are now resistant to penicillins and macrolides (including azithromycin and clarithromycin, the ‘advanced’ agents in this group). A smaller number (6%) are resistant to amoxicillin/clavulanate, although this appears to be a result of in vitro test parameters involving primarily strains with high-level β-lactam resistance. Some cephalosporins also show greater activity than penicillin against intermediately susceptible S pneumoniae, but are not effective against highly resistant strains. In contrast, fewer than 1% of all pneumococci are resistant to newer fluoroquinolones (the so-called respiratory fluoroquinolones, such as gatifloxacin, gemifloxacin, levofloxacin, moxifloxacin) and the ketolide telithromycin.

The prevalence of β-lactamase–producing strains of H influenzae appears to have leveled off. Approximately 30% of H influenzae strains are resistant to ampicillin, while fewer than 1% are resistant to amoxicillin/clavulanate, cefuroxime, macrolides, and newer fluoroquinolones.

More than 90% of M catarrhalis isolates produce β-lactamase, thereby conferring resistance to ampicillin and amoxicillin.

Significant geographical variation in resistance has been observed. The prevalence of penicillin-resistant S pneumoniae ranges from 8% in New England to 25% in the South Atlantic, while ampicillin-resistant H influenzae is seen most often in New England (35%) and least often in the Rocky Mountain region (24%).^33,34,36 Significant differences within a community also have been observed.³⁷ Thus, knowledge of local resistance patterns is necessary. This information generally is available from local hospitals, although such data may be more reflective of nosocomial pathogens, or state health departments.

Community-acquired pneumonia

The 2003 guidelines of the IDSA give advanced macrolides and respiratory fluoroquinolones a prominent role in the management of community-acquired pneumonia (TABLE 1).⁵ The IDSA reviewed data from more than 150 clinical trials conducted in adults over 15 years. The IDSA panel acknowledged the increasing in vitro resistance of S pneumoniae to the macrolides, but noted that reports of clinical failure have not paralleled this. The panel also pointed out the significantly lower rates of resistance to the respiratory fluoroquinolones and expressed concern that abuse of these agents could lead to increased resistance by S pneumoniae.

In a previously healthy person who has not taken an antibiotic in the last 3 months, the IDSA recommends a macrolide or doxycycline as first-line therapy, whereas a fluoroquinolone, high-dose amoxicillin/clavulanate, or a macrolide plus high-dose amoxicillin should be used if an antibiotic has been taken during the last 3 months. Patients with a significant comorbidity can be treated with a fluoroquinolone without regard to recent antibiotic use. Alternatively, a macrolide can be used alone in patients who have not taken an antibiotic in 3 months, but otherwise must be used in combination with high-dose amoxicillin. High-dose amoxicillin/clavulanate or cefpodoxime, cefprozil, or cefuroxime can be used in those with a significant comorbidity and recent antibiotic use.

TABLE 1

Initial empiric therapy in outpatients with community-acquired pneumonia

Bacterial bronchitis

A panel of primary care physicians and specialists convened by the Canadian Thoracic Society (CTS) reviewed nearly 400 published articles on acute bacterial exacerbations of chronic bronchitis, including evidence-based reviews such as the Cochrane Database. The 2003 CTS guidelines recommend that treatment be based on the risk for treatment failure (TABLE 2).⁸

Antibacterial treatment is not recommended for patients whose clinical history and symptoms suggest a viral infection (group 0) unless symptoms persist for more than 10 to 14 days. In those cases, bacterial superinfection with M pneumoniae, C pneumoniae, or Bordetella pertussis is possible. Patients with chronic bronchitis but without risk factors for treatment failure (group 1) may be treated with a variety of first-line agents, including azithromycin, clarithromycin, cefuroxime, cefprozil, cefixime, amoxicillin, doxycycline, or trimethoprim/sulfamethoxazole. For patients in group 1 who fail first-line therapy, and as first-line therapy for patients in group 2, a fluoroquinolone or amoxicillin/clavulanate is recommended. Patients in group 3 are more likely to be infected with a Gram-negative pathogen, such as Ps aeruginosa or Enterobacter species, and are least able to tolerate treatment failure. Hence, ciprofloxacin is appropriate in the outpatient setting.

TABLE 2

Initial empiric therapy in outpatients with acute bacterial exacerbations of chronic bronchitis

Bacterial rhinosinusitis

The recommendations for management of acute bacterial rhinosinusitis issued by the Sinus and Allergy Health Partnership (SAHP), a not-for-profit organization created by the American Academy of Otolaryngic Allergy, the American Academy of Otolaryngology-Head and Neck Surgery, and the American Rhinologic Society, are based on a variety of factors. These include rate of spontaneous resolution, pathogen distribution, antibacterial resistance data, the importance of S pneumoniae in intracranial and extrasinus complications, and the ability of a patient to tolerate treatment failure (TABLE 3).¹⁷ The panel reviewed more than 150 published articles on management of children and adults with bacterial rhinosinusitis.

As in the pneumonia guidelines, recent antibiotic use is an important consideration when selecting an antibiotic since resistant pathogens are likely. β-Lactam agents play a major role as initial therapy in both children and adults. This recommendation is consistent with those of Williams et al who reviewed 49 clinical trials involving 13,660 patients. These investigators recommended 7 to 14 days of penicillin or amoxicillin for acute maxillary sinusitis confirmed radiographically or by aspiration.³⁸

The SAHP recommended higher doses of amoxicillin (with or without clavulanate) in patients who have recently taken an antibiotic or who have moderate disease. Fluoroquinolones are recommended as alternatives in patients with mild disease who have not taken an antibiotic in the last 4 to 6 weeks. However, in patients with mild disease who have taken antibiotics recently or who have moderate disease, fluoroquinolones are recommended as first-line therapy. Macrolides are recommended only for patients with a β-lactam allergy since failure rates of 20% to 25% are possible. Lack of improvement or worsening symptoms after 72 hours should prompt reevaluation, may necessitate cultures and/or a CT scan, and should raise the possibility of causal organisms other than S pneumoniae, H influenzae, and M catarrhalis.

TABLE 3

Initial empiric therapy in outpatients with acute bacterial rhinosinusitis

Dose and duration

While each of the three guidelines provides detailed recommendations regarding selection of an antibacterial agent, the dose and duration of therapy generally are not well defined. Fortunately, other sources provide guidance in these 2 areas.

First, an independent international panel of infectious diseases experts, whose goal was to identify ways to improve prescription of antibiotics for lower respiratory tract infections, stressed that an important purpose of therapy is to reduce bacterial load and, in fact, treat to bacteriologic cure.¹⁹ Antibiotic therapy that allows some bacteria to survive increases the risk of early recurrence or relapse and encourages resistance selection. Such therapy is, therefore, inappropriate. The panel concurred with the WHO and others that the likelihood of bacterial persistence increases when antibiotics are prescribed in low doses, especially if given over long periods.^18,39-41 Prolonged low-dose antibiotic therapy, which has been common practice for many infections, is contrary to the WHO Global Strategy for Containment of Antimicrobial Resistance, which notes that single-agent therapy for a short duration is 1 of several actions that can be taken to minimize bacterial resistance.⁴² Shorter courses of antibiotic therapy also are consistent with SHEA/IDSA recommendations.¹⁰

The clinical appropriateness of this recommendation is reinforced by the changes that have occurred during the past decade in the management of selected urinary tract infections (UTIs). Some UTIs that previously had been treated with low-dose antibiotics for 10 to 14 days now are treated with only 1 or a few high doses of a single agent. Other infections for which clinical data support shorter courses of antibiotic therapy include uncomplicated cellulitis,⁴³ ventilator-associated pneumonia,⁴⁴ and meningococcal disease.⁴⁵

Shorter-course antibacterial therapy for CARTIs increasingly has been the focus of clinical trials. The focus is not unreasonable. Many of the antibiotics used for CARTIs are very potent against the pathogens commonly encountered, penetrate infected tissues well, are available in oral formulations, and are generally well tolerated. However, to optimize an antibiotic’s bactericidal potential, it is necessary to base the dosing regimen on its pharmacodynamics. From a pharmacodynamic perspective, there are 2 groups of antibiotics, those with concentration-dependent killing and those with time-dependent killing. For agents with concentration-dependent killing, such as fluoroquinolones, ketolides, and aminoglycosides, the goal is to select a dose that achieves a higher peak concentration and/or a larger area under the plasma concentration curve, with acceptable tolerability. In contrast, antibiotics that rely on time-dependent killing, such as β-lactams, macrolides, azalides, tetracyclines, and some others, require extended durations of concentrations above the MIC90 of the bacterial pathogen(s). Consequently, multiple daily dosing may be preferable.⁴⁶

Dose. A few studies have compared high-dose, short-course therapy with therapy using standard doses and durations (TABLE 4). To assess the impact of highdose, short-course therapy on post-treatment resistant pneumococcal carriage, Schrag compared amoxicillin given either as 90 mg/kg/day for 5 days (high-dose, short-course) or 40 mg/kg/day for 10 days (standard) in 797 children with a respiratory tract infection.⁴⁷ At day 28, nasal carriage of penicillin-resistant S pneumoniae was detected in 24% of the high-dose, shortcourse group and in 32% of the standard group (relative risk, 0.77; P=0.03). Among the pneumococcal carriers, the risk of penicillin-resistant S pneumoniae was significantly lower in the high-dose, short-course group than in the standard therapy group (relative risk, 0.78; P=0.01)

Another study investigated high-dose, short-course therapy with levofloxacin in patients with mild to severe community-acquired pneumonia. Patients received levofloxacin 750 mg/d for 5 days or 500 mg/d for 10 days.⁴⁸ The clinical success rates were 92.4% and 91.1%, respectively, while the bacteriologic eradication rates at 7 to 14 days post-therapy were 93.2% and 92.4%, respectively, thereby demonstrating that high-dose, shortcourse levofloxacin therapy is at least as effective as standard levofloxacin therapy.

Duration. Short-course therapy using standard doses of azithromycin, gemifloxacin, levofloxacin, moxifloxacin, and telithromycin has been investigated in clinical trials of CARTIs (FIGURE). In patients with community-acquired pneumonia, 5 days of therapy with telithromycin was shown to be equivalent to a 7-day course (both using a single daily dose of 800 mg), as well as to clarithromycin 500 mg bid for 10 days.⁴⁹

In studies of acute exacerbations of chronic bronchitis, a 3-day course of azithromycin was equivalent to clarithromycin for 10 days⁵⁰ and gemifloxacin for 5 days was equivalent to a sequential combination of ceftriaxone and cefuroxime axetil for up to 10 days.⁵¹ Levofloxacin for 3 days and azithromycin for 5 days provided equivalent outcomes,⁵² as did levofloxacin for 5 days and amoxicillin/clavulanate for 10 days.⁵² Moxifloxacin for 5 days provided results equivalent to those of 7 days of amoxicillin, clarithromycin, or cefuroxime axetil.⁵³ Five days of telithromycin was shown to be equivalent to 10 days of amoxicillin/clavu-lanate⁵⁴ or cefuroxime axetil.⁵⁵

Studies of acute bacterial rhinosinusitis have demonstrated equivalent results with azithromycin for 3 days and amoxicillin/clavulanate for 10 days.⁵⁰ Similarly, telithromycin for 5 days was equivalent to 10 days of telithromycin, amoxicillin/clavulanate, or cefuroxime axetil.⁵⁶

These clinical trials demonstrate that short-course therapy achieves clinical cure and/or bacteriologic eradication rates that are at least equivalent to those of standard therapy, with no significant difference in safety. Symptomatic improvement is faster and total antibiotic exposure is reduced with short-course therapy.

A significant advantage of short-course antibacterial therapy is improved patient adherence. Adherence is 10% to 20% better with 5-day courses than with 10-day courses,^47,57 and is significantly better with 1 or 2 daily doses than with 3 or more daily doses.^58-60 In fact, a recent market research study showed that patients perceive once-daily, short-course antibiotic treatment to be significantly more effective than longer courses. This may be due to faster improvement of infection-related symptoms.⁶¹ For example, Dunbar et al observed that significantly more patients treated with high-dose, short-course levofloxacin experienced subjective and objective resolution of fever by day 3 compared with those who received standard-dose, short-course levofloxacin.⁴⁸

TABLE 4

Clinical trials of high-dose, short-course antibiotic therapy

TABLE 5 Clinical trials of standard-dose, short-course antibiotic therapy

Summary

Essential questions that need to be answered for every patient who presents with a possible CARTI include : 1) Is antibacterial therapy necessary? 2) If so, what is the best antibiotic and at what dose and for how long should it be administered? Accumulating evidence indicates that some antibiotics when given in high doses for a short duration are as effective and safe as standard therapy for CARTIs. Short-course therapy also promotes patient compliance.

Disclosures:

The authors reported the following financial relationships: Dr Brunton: consultant to Abbott, Ortho-McNeil Pharmaceutical, Inc., and SanofiAventis. Dr Carmichael: consultant to Ortho-McNeil Pharmaceutical, Inc.; on the speakers’ bureaus for Bristol-Myers Squibb Company, Merck & Co., OrthoMcNeil Pharmaceutical, Inc., and Pfizer Inc. Dr Fitzgerald: on the speakers’ bureaus for Boehringer Ingelheim, GlaxoSmithKline, Ortho-McNeil Pharmaceutical, Inc., Pfizer Inc., Sepracor Inc., and 3M. Dr Liu: on the speakers’ bureaus for Aventis Pharmaceuticals, Bayer Pharmaceuticals Corporation, Bristol-Myers Squibb Company, Cobist, GlaxoSmithKline, Merck & Co., Ortho-McNeil Pharmaceutical, Inc., Pfizer Inc., Purdue Pharma, Oscient Pharmaceuticals Corporation, and Wyeth Pharmaceuticals. Dr Varon: on the speakers’ bureau for Ortho-McNeil Pharmaceutical, Inc. Dr. Weiland: consultant to Abbott Laboratories, Ortho-McNeil Pharmaceutical, Inc., and Pfizer Inc.

This supplement to The Journal of Family Practice is supported by a grant from Ortho-McNeil Pharmaceutical, Inc. It was adapted from a consensus conference coordinated by the Primary Care Education Consortium and Texas Academy of Family Physicians and was edited and peer-reviewed by The Journal of Family Practice. © 2005 Quadrant HealthCom Inc. and Primary Care Education Consortium.

Community-acquired respiratory tract infections (CARTIs) are a reason for seeking medical attention. In 2001, there were 28.4 million office visits in the United States for an acute respiratory tract infection (excluding pharyngitis).¹

Management of CARTIs poses several challenges. According to the World Health Organization (WHO), “for every 100 respiratory infections, only 20% require antibiotic treatment”² —the remaining 80 infections most likely have a viral origin. Thus, antibacterial therapy should be avoided unless a bacterial cause has been confirmed or is deemed likely.

Once that determination has been made, clinicians need to separate patients who can be safely managed as outpatients from those who need to be hospitalized. Disease severity is, of course, an important consideration in this selection process.^3-8

For management of patients who will not be hospitalized, the WHO and the Society for Healthcare Epidemiology of America/Infectious Diseases Society of America (SHEA/IDSA) offer the 3 Ds: administer the correct drug, at the right dose, and for the appropriate duration, to minimize development and spread of resistance.^9,10 A recent consensus conference coordinated by the Primary Care Education Consortium and Texas Academy of Family Physicians elaborated on this mnemonic as it relates to outpatient management of 3 CARTIs: community-acquired pneumonia, acute bacterial exacerbations of chronic bronchitis, and acute bacterial rhinosinusitis. Its recommendations, which have been drawn largely from existing evidence-based guidelines, form the basis for this review.

Drug selection

Treatment with an antibacterial agent will not be medically warranted in the majority of patients with a CARTI. Many of these patients will, however, expect to go home with an antibiotic prescription. The primary care clinician can reduce such expectations and prevent unnecessary reconsultations by briefly addressing four issues: 1) the natural course of the viral illness, 2) the lack of effectiveness of antibiotics, 3) the problem of antibiotic resistance, and 4) the side effects of antibiotics.¹¹

When selecting an antibacterial agent for patients with pneumonia, bronchitis, or rhinosinusitis for which a bacterial cause has been identified or deemed likely, several factors need to be taken into account, including the suspected or identified pathogens, local resistance patterns, previous therapy, patient allergies, and the patient’s ability to tolerate treatment failure. Many of these factors are considered by professional organizations that regularly develop guidelines for CARTIs based on the best available evidence. Perhaps most critical for decision-making in the primary care setting is an understanding of evolving microbiology and resistance patterns.

Common pathogens

There is considerable overlap among pathogens commonly found in CARTIs. Streptococcus pneumoniae and Haemophilus influenzae are most often observed in the outpatient setting.

Community-acquired pneumonia. In outpatients with mild illness, S pneumoniae, Mycoplasma pneumoniae, Chlamydia species (particularly Chlamydia pneumoniae, now called Chlamydophilia pneumoniae), and H influenzae are the most common pathogens.^12,13 In patients younger than 50 years without significant comorbidity, Mycoplasma species are the most common pathogens. Older patients and those with significant comorbidity are more likely infected with S pneumoniae,¹³ a Gram-negative enteric bacillus, Pseudomonas aeruginosa, or Legionella.^5,14

Bacterial bronchitis. In addition to S pneumoniae and H influenzae, Moraxella catarrhalis is a frequent pathogen in bacterial exacerbations of chronic bronchitis.¹⁵P aeruginosa and other Gram-negative bacilli are also seen, especially in patients with a severe acute exacerbation who have a forced expiratory volume in 1 second (FEV₁) of 35% of predicted or less.¹⁶ Infection due to multiple pathogens occurs in a small percentage of all patients with chronic bronchitis, particularly those with severe exacerbations. Fewer than 10% of acute exacerbations are due to an atypical bacterium, usually C pneumoniae. M pneumoniae and Legionella pneumophila are implicated even less frequently.¹⁵

Bacterial rhinosinusitis.S pneumoniae and H influenzae also are frequent causes of acute bacterial rhinosinusitis. Other pathogens commonly seen in this condition include other Streptococcus species, M catarrhalis, oral anaerobes, Staphylococcus aureus in adults, and M catarrhalis, Streptococcus pyogenes, and anaerobes in children.¹⁷

Resistance

Antibiotic resistance is an important consideration in the management of CARTIs. There is little doubt that widespread use of antibiotics leads to in vitro bacterial resistance.^18-20 However, because clinical success has been observed in the presence of pathogens with low-level resistance, there is some debate as to whether low-level antibiotic resistance has a significant effect on clinical outcomes.^18,21-29 Even so, the US Centers for Disease Control and Prevention has determined that people who attend or work at child-care centers and those who recently used antimicrobial agents are at increased risk for infection with drug-resistant S pneumoniae.³⁰ Moreover, the WHO has stated that infection with resistant pathogens prolongs illness and increases the probability of a fatal outcome.³¹

Several surveillance programs that monitor antibiotic resistance patterns—including the Alexander Project³² and Tracking Resistance in the United States Today (TRUST)^33-36 —have confirmed widespread resistance to antibiotics commonly used to treat CARTIs in the United States. β-Lactam resistance due to penicillin-binding protein changes in S pneumoniae has increased significantly over the past decade. Generally, more than 30% of S pneumoniae are now resistant to penicillins and macrolides (including azithromycin and clarithromycin, the ‘advanced’ agents in this group). A smaller number (6%) are resistant to amoxicillin/clavulanate, although this appears to be a result of in vitro test parameters involving primarily strains with high-level β-lactam resistance. Some cephalosporins also show greater activity than penicillin against intermediately susceptible S pneumoniae, but are not effective against highly resistant strains. In contrast, fewer than 1% of all pneumococci are resistant to newer fluoroquinolones (the so-called respiratory fluoroquinolones, such as gatifloxacin, gemifloxacin, levofloxacin, moxifloxacin) and the ketolide telithromycin.

The prevalence of β-lactamase–producing strains of H influenzae appears to have leveled off. Approximately 30% of H influenzae strains are resistant to ampicillin, while fewer than 1% are resistant to amoxicillin/clavulanate, cefuroxime, macrolides, and newer fluoroquinolones.

More than 90% of M catarrhalis isolates produce β-lactamase, thereby conferring resistance to ampicillin and amoxicillin.

Significant geographical variation in resistance has been observed. The prevalence of penicillin-resistant S pneumoniae ranges from 8% in New England to 25% in the South Atlantic, while ampicillin-resistant H influenzae is seen most often in New England (35%) and least often in the Rocky Mountain region (24%).^33,34,36 Significant differences within a community also have been observed.³⁷ Thus, knowledge of local resistance patterns is necessary. This information generally is available from local hospitals, although such data may be more reflective of nosocomial pathogens, or state health departments.

Community-acquired pneumonia

The 2003 guidelines of the IDSA give advanced macrolides and respiratory fluoroquinolones a prominent role in the management of community-acquired pneumonia (TABLE 1).⁵ The IDSA reviewed data from more than 150 clinical trials conducted in adults over 15 years. The IDSA panel acknowledged the increasing in vitro resistance of S pneumoniae to the macrolides, but noted that reports of clinical failure have not paralleled this. The panel also pointed out the significantly lower rates of resistance to the respiratory fluoroquinolones and expressed concern that abuse of these agents could lead to increased resistance by S pneumoniae.

In a previously healthy person who has not taken an antibiotic in the last 3 months, the IDSA recommends a macrolide or doxycycline as first-line therapy, whereas a fluoroquinolone, high-dose amoxicillin/clavulanate, or a macrolide plus high-dose amoxicillin should be used if an antibiotic has been taken during the last 3 months. Patients with a significant comorbidity can be treated with a fluoroquinolone without regard to recent antibiotic use. Alternatively, a macrolide can be used alone in patients who have not taken an antibiotic in 3 months, but otherwise must be used in combination with high-dose amoxicillin. High-dose amoxicillin/clavulanate or cefpodoxime, cefprozil, or cefuroxime can be used in those with a significant comorbidity and recent antibiotic use.

TABLE 1

Initial empiric therapy in outpatients with community-acquired pneumonia

Bacterial bronchitis

A panel of primary care physicians and specialists convened by the Canadian Thoracic Society (CTS) reviewed nearly 400 published articles on acute bacterial exacerbations of chronic bronchitis, including evidence-based reviews such as the Cochrane Database. The 2003 CTS guidelines recommend that treatment be based on the risk for treatment failure (TABLE 2).⁸

Antibacterial treatment is not recommended for patients whose clinical history and symptoms suggest a viral infection (group 0) unless symptoms persist for more than 10 to 14 days. In those cases, bacterial superinfection with M pneumoniae, C pneumoniae, or Bordetella pertussis is possible. Patients with chronic bronchitis but without risk factors for treatment failure (group 1) may be treated with a variety of first-line agents, including azithromycin, clarithromycin, cefuroxime, cefprozil, cefixime, amoxicillin, doxycycline, or trimethoprim/sulfamethoxazole. For patients in group 1 who fail first-line therapy, and as first-line therapy for patients in group 2, a fluoroquinolone or amoxicillin/clavulanate is recommended. Patients in group 3 are more likely to be infected with a Gram-negative pathogen, such as Ps aeruginosa or Enterobacter species, and are least able to tolerate treatment failure. Hence, ciprofloxacin is appropriate in the outpatient setting.

TABLE 2

Initial empiric therapy in outpatients with acute bacterial exacerbations of chronic bronchitis

Bacterial rhinosinusitis

The recommendations for management of acute bacterial rhinosinusitis issued by the Sinus and Allergy Health Partnership (SAHP), a not-for-profit organization created by the American Academy of Otolaryngic Allergy, the American Academy of Otolaryngology-Head and Neck Surgery, and the American Rhinologic Society, are based on a variety of factors. These include rate of spontaneous resolution, pathogen distribution, antibacterial resistance data, the importance of S pneumoniae in intracranial and extrasinus complications, and the ability of a patient to tolerate treatment failure (TABLE 3).¹⁷ The panel reviewed more than 150 published articles on management of children and adults with bacterial rhinosinusitis.

As in the pneumonia guidelines, recent antibiotic use is an important consideration when selecting an antibiotic since resistant pathogens are likely. β-Lactam agents play a major role as initial therapy in both children and adults. This recommendation is consistent with those of Williams et al who reviewed 49 clinical trials involving 13,660 patients. These investigators recommended 7 to 14 days of penicillin or amoxicillin for acute maxillary sinusitis confirmed radiographically or by aspiration.³⁸

The SAHP recommended higher doses of amoxicillin (with or without clavulanate) in patients who have recently taken an antibiotic or who have moderate disease. Fluoroquinolones are recommended as alternatives in patients with mild disease who have not taken an antibiotic in the last 4 to 6 weeks. However, in patients with mild disease who have taken antibiotics recently or who have moderate disease, fluoroquinolones are recommended as first-line therapy. Macrolides are recommended only for patients with a β-lactam allergy since failure rates of 20% to 25% are possible. Lack of improvement or worsening symptoms after 72 hours should prompt reevaluation, may necessitate cultures and/or a CT scan, and should raise the possibility of causal organisms other than S pneumoniae, H influenzae, and M catarrhalis.

TABLE 3

Initial empiric therapy in outpatients with acute bacterial rhinosinusitis

Dose and duration

While each of the three guidelines provides detailed recommendations regarding selection of an antibacterial agent, the dose and duration of therapy generally are not well defined. Fortunately, other sources provide guidance in these 2 areas.

First, an independent international panel of infectious diseases experts, whose goal was to identify ways to improve prescription of antibiotics for lower respiratory tract infections, stressed that an important purpose of therapy is to reduce bacterial load and, in fact, treat to bacteriologic cure.¹⁹ Antibiotic therapy that allows some bacteria to survive increases the risk of early recurrence or relapse and encourages resistance selection. Such therapy is, therefore, inappropriate. The panel concurred with the WHO and others that the likelihood of bacterial persistence increases when antibiotics are prescribed in low doses, especially if given over long periods.^18,39-41 Prolonged low-dose antibiotic therapy, which has been common practice for many infections, is contrary to the WHO Global Strategy for Containment of Antimicrobial Resistance, which notes that single-agent therapy for a short duration is 1 of several actions that can be taken to minimize bacterial resistance.⁴² Shorter courses of antibiotic therapy also are consistent with SHEA/IDSA recommendations.¹⁰

The clinical appropriateness of this recommendation is reinforced by the changes that have occurred during the past decade in the management of selected urinary tract infections (UTIs). Some UTIs that previously had been treated with low-dose antibiotics for 10 to 14 days now are treated with only 1 or a few high doses of a single agent. Other infections for which clinical data support shorter courses of antibiotic therapy include uncomplicated cellulitis,⁴³ ventilator-associated pneumonia,⁴⁴ and meningococcal disease.⁴⁵

Shorter-course antibacterial therapy for CARTIs increasingly has been the focus of clinical trials. The focus is not unreasonable. Many of the antibiotics used for CARTIs are very potent against the pathogens commonly encountered, penetrate infected tissues well, are available in oral formulations, and are generally well tolerated. However, to optimize an antibiotic’s bactericidal potential, it is necessary to base the dosing regimen on its pharmacodynamics. From a pharmacodynamic perspective, there are 2 groups of antibiotics, those with concentration-dependent killing and those with time-dependent killing. For agents with concentration-dependent killing, such as fluoroquinolones, ketolides, and aminoglycosides, the goal is to select a dose that achieves a higher peak concentration and/or a larger area under the plasma concentration curve, with acceptable tolerability. In contrast, antibiotics that rely on time-dependent killing, such as β-lactams, macrolides, azalides, tetracyclines, and some others, require extended durations of concentrations above the MIC90 of the bacterial pathogen(s). Consequently, multiple daily dosing may be preferable.⁴⁶

Dose. A few studies have compared high-dose, short-course therapy with therapy using standard doses and durations (TABLE 4). To assess the impact of highdose, short-course therapy on post-treatment resistant pneumococcal carriage, Schrag compared amoxicillin given either as 90 mg/kg/day for 5 days (high-dose, short-course) or 40 mg/kg/day for 10 days (standard) in 797 children with a respiratory tract infection.⁴⁷ At day 28, nasal carriage of penicillin-resistant S pneumoniae was detected in 24% of the high-dose, shortcourse group and in 32% of the standard group (relative risk, 0.77; P=0.03). Among the pneumococcal carriers, the risk of penicillin-resistant S pneumoniae was significantly lower in the high-dose, short-course group than in the standard therapy group (relative risk, 0.78; P=0.01)

Another study investigated high-dose, short-course therapy with levofloxacin in patients with mild to severe community-acquired pneumonia. Patients received levofloxacin 750 mg/d for 5 days or 500 mg/d for 10 days.⁴⁸ The clinical success rates were 92.4% and 91.1%, respectively, while the bacteriologic eradication rates at 7 to 14 days post-therapy were 93.2% and 92.4%, respectively, thereby demonstrating that high-dose, shortcourse levofloxacin therapy is at least as effective as standard levofloxacin therapy.

Duration. Short-course therapy using standard doses of azithromycin, gemifloxacin, levofloxacin, moxifloxacin, and telithromycin has been investigated in clinical trials of CARTIs (FIGURE). In patients with community-acquired pneumonia, 5 days of therapy with telithromycin was shown to be equivalent to a 7-day course (both using a single daily dose of 800 mg), as well as to clarithromycin 500 mg bid for 10 days.⁴⁹

In studies of acute exacerbations of chronic bronchitis, a 3-day course of azithromycin was equivalent to clarithromycin for 10 days⁵⁰ and gemifloxacin for 5 days was equivalent to a sequential combination of ceftriaxone and cefuroxime axetil for up to 10 days.⁵¹ Levofloxacin for 3 days and azithromycin for 5 days provided equivalent outcomes,⁵² as did levofloxacin for 5 days and amoxicillin/clavulanate for 10 days.⁵² Moxifloxacin for 5 days provided results equivalent to those of 7 days of amoxicillin, clarithromycin, or cefuroxime axetil.⁵³ Five days of telithromycin was shown to be equivalent to 10 days of amoxicillin/clavu-lanate⁵⁴ or cefuroxime axetil.⁵⁵

Studies of acute bacterial rhinosinusitis have demonstrated equivalent results with azithromycin for 3 days and amoxicillin/clavulanate for 10 days.⁵⁰ Similarly, telithromycin for 5 days was equivalent to 10 days of telithromycin, amoxicillin/clavulanate, or cefuroxime axetil.⁵⁶

These clinical trials demonstrate that short-course therapy achieves clinical cure and/or bacteriologic eradication rates that are at least equivalent to those of standard therapy, with no significant difference in safety. Symptomatic improvement is faster and total antibiotic exposure is reduced with short-course therapy.

A significant advantage of short-course antibacterial therapy is improved patient adherence. Adherence is 10% to 20% better with 5-day courses than with 10-day courses,^47,57 and is significantly better with 1 or 2 daily doses than with 3 or more daily doses.^58-60 In fact, a recent market research study showed that patients perceive once-daily, short-course antibiotic treatment to be significantly more effective than longer courses. This may be due to faster improvement of infection-related symptoms.⁶¹ For example, Dunbar et al observed that significantly more patients treated with high-dose, short-course levofloxacin experienced subjective and objective resolution of fever by day 3 compared with those who received standard-dose, short-course levofloxacin.⁴⁸

TABLE 4

Clinical trials of high-dose, short-course antibiotic therapy

TABLE 5 Clinical trials of standard-dose, short-course antibiotic therapy

Summary

Essential questions that need to be answered for every patient who presents with a possible CARTI include : 1) Is antibacterial therapy necessary? 2) If so, what is the best antibiotic and at what dose and for how long should it be administered? Accumulating evidence indicates that some antibiotics when given in high doses for a short duration are as effective and safe as standard therapy for CARTIs. Short-course therapy also promotes patient compliance.

Disclosures:

The authors reported the following financial relationships: Dr Brunton: consultant to Abbott, Ortho-McNeil Pharmaceutical, Inc., and SanofiAventis. Dr Carmichael: consultant to Ortho-McNeil Pharmaceutical, Inc.; on the speakers’ bureaus for Bristol-Myers Squibb Company, Merck & Co., OrthoMcNeil Pharmaceutical, Inc., and Pfizer Inc. Dr Fitzgerald: on the speakers’ bureaus for Boehringer Ingelheim, GlaxoSmithKline, Ortho-McNeil Pharmaceutical, Inc., Pfizer Inc., Sepracor Inc., and 3M. Dr Liu: on the speakers’ bureaus for Aventis Pharmaceuticals, Bayer Pharmaceuticals Corporation, Bristol-Myers Squibb Company, Cobist, GlaxoSmithKline, Merck & Co., Ortho-McNeil Pharmaceutical, Inc., Pfizer Inc., Purdue Pharma, Oscient Pharmaceuticals Corporation, and Wyeth Pharmaceuticals. Dr Varon: on the speakers’ bureau for Ortho-McNeil Pharmaceutical, Inc. Dr. Weiland: consultant to Abbott Laboratories, Ortho-McNeil Pharmaceutical, Inc., and Pfizer Inc.

This supplement to The Journal of Family Practice is supported by a grant from Ortho-McNeil Pharmaceutical, Inc. It was adapted from a consensus conference coordinated by the Primary Care Education Consortium and Texas Academy of Family Physicians and was edited and peer-reviewed by The Journal of Family Practice. © 2005 Quadrant HealthCom Inc. and Primary Care Education Consortium.

Community-acquired respiratory tract infections (CARTIs) are a reason for seeking medical attention. In 2001, there were 28.4 million office visits in the United States for an acute respiratory tract infection (excluding pharyngitis).¹

Management of CARTIs poses several challenges. According to the World Health Organization (WHO), “for every 100 respiratory infections, only 20% require antibiotic treatment”² —the remaining 80 infections most likely have a viral origin. Thus, antibacterial therapy should be avoided unless a bacterial cause has been confirmed or is deemed likely.

Once that determination has been made, clinicians need to separate patients who can be safely managed as outpatients from those who need to be hospitalized. Disease severity is, of course, an important consideration in this selection process.^3-8

For management of patients who will not be hospitalized, the WHO and the Society for Healthcare Epidemiology of America/Infectious Diseases Society of America (SHEA/IDSA) offer the 3 Ds: administer the correct drug, at the right dose, and for the appropriate duration, to minimize development and spread of resistance.^9,10 A recent consensus conference coordinated by the Primary Care Education Consortium and Texas Academy of Family Physicians elaborated on this mnemonic as it relates to outpatient management of 3 CARTIs: community-acquired pneumonia, acute bacterial exacerbations of chronic bronchitis, and acute bacterial rhinosinusitis. Its recommendations, which have been drawn largely from existing evidence-based guidelines, form the basis for this review.

Drug selection

Treatment with an antibacterial agent will not be medically warranted in the majority of patients with a CARTI. Many of these patients will, however, expect to go home with an antibiotic prescription. The primary care clinician can reduce such expectations and prevent unnecessary reconsultations by briefly addressing four issues: 1) the natural course of the viral illness, 2) the lack of effectiveness of antibiotics, 3) the problem of antibiotic resistance, and 4) the side effects of antibiotics.¹¹

When selecting an antibacterial agent for patients with pneumonia, bronchitis, or rhinosinusitis for which a bacterial cause has been identified or deemed likely, several factors need to be taken into account, including the suspected or identified pathogens, local resistance patterns, previous therapy, patient allergies, and the patient’s ability to tolerate treatment failure. Many of these factors are considered by professional organizations that regularly develop guidelines for CARTIs based on the best available evidence. Perhaps most critical for decision-making in the primary care setting is an understanding of evolving microbiology and resistance patterns.

Common pathogens

There is considerable overlap among pathogens commonly found in CARTIs. Streptococcus pneumoniae and Haemophilus influenzae are most often observed in the outpatient setting.

Community-acquired pneumonia. In outpatients with mild illness, S pneumoniae, Mycoplasma pneumoniae, Chlamydia species (particularly Chlamydia pneumoniae, now called Chlamydophilia pneumoniae), and H influenzae are the most common pathogens.^12,13 In patients younger than 50 years without significant comorbidity, Mycoplasma species are the most common pathogens. Older patients and those with significant comorbidity are more likely infected with S pneumoniae,¹³ a Gram-negative enteric bacillus, Pseudomonas aeruginosa, or Legionella.^5,14

Bacterial bronchitis. In addition to S pneumoniae and H influenzae, Moraxella catarrhalis is a frequent pathogen in bacterial exacerbations of chronic bronchitis.¹⁵P aeruginosa and other Gram-negative bacilli are also seen, especially in patients with a severe acute exacerbation who have a forced expiratory volume in 1 second (FEV₁) of 35% of predicted or less.¹⁶ Infection due to multiple pathogens occurs in a small percentage of all patients with chronic bronchitis, particularly those with severe exacerbations. Fewer than 10% of acute exacerbations are due to an atypical bacterium, usually C pneumoniae. M pneumoniae and Legionella pneumophila are implicated even less frequently.¹⁵

Bacterial rhinosinusitis.S pneumoniae and H influenzae also are frequent causes of acute bacterial rhinosinusitis. Other pathogens commonly seen in this condition include other Streptococcus species, M catarrhalis, oral anaerobes, Staphylococcus aureus in adults, and M catarrhalis, Streptococcus pyogenes, and anaerobes in children.¹⁷

Resistance

Antibiotic resistance is an important consideration in the management of CARTIs. There is little doubt that widespread use of antibiotics leads to in vitro bacterial resistance.^18-20 However, because clinical success has been observed in the presence of pathogens with low-level resistance, there is some debate as to whether low-level antibiotic resistance has a significant effect on clinical outcomes.^18,21-29 Even so, the US Centers for Disease Control and Prevention has determined that people who attend or work at child-care centers and those who recently used antimicrobial agents are at increased risk for infection with drug-resistant S pneumoniae.³⁰ Moreover, the WHO has stated that infection with resistant pathogens prolongs illness and increases the probability of a fatal outcome.³¹

Several surveillance programs that monitor antibiotic resistance patterns—including the Alexander Project³² and Tracking Resistance in the United States Today (TRUST)^33-36 —have confirmed widespread resistance to antibiotics commonly used to treat CARTIs in the United States. β-Lactam resistance due to penicillin-binding protein changes in S pneumoniae has increased significantly over the past decade. Generally, more than 30% of S pneumoniae are now resistant to penicillins and macrolides (including azithromycin and clarithromycin, the ‘advanced’ agents in this group). A smaller number (6%) are resistant to amoxicillin/clavulanate, although this appears to be a result of in vitro test parameters involving primarily strains with high-level β-lactam resistance. Some cephalosporins also show greater activity than penicillin against intermediately susceptible S pneumoniae, but are not effective against highly resistant strains. In contrast, fewer than 1% of all pneumococci are resistant to newer fluoroquinolones (the so-called respiratory fluoroquinolones, such as gatifloxacin, gemifloxacin, levofloxacin, moxifloxacin) and the ketolide telithromycin.

The prevalence of β-lactamase–producing strains of H influenzae appears to have leveled off. Approximately 30% of H influenzae strains are resistant to ampicillin, while fewer than 1% are resistant to amoxicillin/clavulanate, cefuroxime, macrolides, and newer fluoroquinolones.

More than 90% of M catarrhalis isolates produce β-lactamase, thereby conferring resistance to ampicillin and amoxicillin.

Significant geographical variation in resistance has been observed. The prevalence of penicillin-resistant S pneumoniae ranges from 8% in New England to 25% in the South Atlantic, while ampicillin-resistant H influenzae is seen most often in New England (35%) and least often in the Rocky Mountain region (24%).^33,34,36 Significant differences within a community also have been observed.³⁷ Thus, knowledge of local resistance patterns is necessary. This information generally is available from local hospitals, although such data may be more reflective of nosocomial pathogens, or state health departments.

Community-acquired pneumonia

The 2003 guidelines of the IDSA give advanced macrolides and respiratory fluoroquinolones a prominent role in the management of community-acquired pneumonia (TABLE 1).⁵ The IDSA reviewed data from more than 150 clinical trials conducted in adults over 15 years. The IDSA panel acknowledged the increasing in vitro resistance of S pneumoniae to the macrolides, but noted that reports of clinical failure have not paralleled this. The panel also pointed out the significantly lower rates of resistance to the respiratory fluoroquinolones and expressed concern that abuse of these agents could lead to increased resistance by S pneumoniae.

In a previously healthy person who has not taken an antibiotic in the last 3 months, the IDSA recommends a macrolide or doxycycline as first-line therapy, whereas a fluoroquinolone, high-dose amoxicillin/clavulanate, or a macrolide plus high-dose amoxicillin should be used if an antibiotic has been taken during the last 3 months. Patients with a significant comorbidity can be treated with a fluoroquinolone without regard to recent antibiotic use. Alternatively, a macrolide can be used alone in patients who have not taken an antibiotic in 3 months, but otherwise must be used in combination with high-dose amoxicillin. High-dose amoxicillin/clavulanate or cefpodoxime, cefprozil, or cefuroxime can be used in those with a significant comorbidity and recent antibiotic use.

TABLE 1

Initial empiric therapy in outpatients with community-acquired pneumonia

Bacterial bronchitis

A panel of primary care physicians and specialists convened by the Canadian Thoracic Society (CTS) reviewed nearly 400 published articles on acute bacterial exacerbations of chronic bronchitis, including evidence-based reviews such as the Cochrane Database. The 2003 CTS guidelines recommend that treatment be based on the risk for treatment failure (TABLE 2).⁸

Antibacterial treatment is not recommended for patients whose clinical history and symptoms suggest a viral infection (group 0) unless symptoms persist for more than 10 to 14 days. In those cases, bacterial superinfection with M pneumoniae, C pneumoniae, or Bordetella pertussis is possible. Patients with chronic bronchitis but without risk factors for treatment failure (group 1) may be treated with a variety of first-line agents, including azithromycin, clarithromycin, cefuroxime, cefprozil, cefixime, amoxicillin, doxycycline, or trimethoprim/sulfamethoxazole. For patients in group 1 who fail first-line therapy, and as first-line therapy for patients in group 2, a fluoroquinolone or amoxicillin/clavulanate is recommended. Patients in group 3 are more likely to be infected with a Gram-negative pathogen, such as Ps aeruginosa or Enterobacter species, and are least able to tolerate treatment failure. Hence, ciprofloxacin is appropriate in the outpatient setting.

TABLE 2

Initial empiric therapy in outpatients with acute bacterial exacerbations of chronic bronchitis

Bacterial rhinosinusitis

The recommendations for management of acute bacterial rhinosinusitis issued by the Sinus and Allergy Health Partnership (SAHP), a not-for-profit organization created by the American Academy of Otolaryngic Allergy, the American Academy of Otolaryngology-Head and Neck Surgery, and the American Rhinologic Society, are based on a variety of factors. These include rate of spontaneous resolution, pathogen distribution, antibacterial resistance data, the importance of S pneumoniae in intracranial and extrasinus complications, and the ability of a patient to tolerate treatment failure (TABLE 3).¹⁷ The panel reviewed more than 150 published articles on management of children and adults with bacterial rhinosinusitis.

As in the pneumonia guidelines, recent antibiotic use is an important consideration when selecting an antibiotic since resistant pathogens are likely. β-Lactam agents play a major role as initial therapy in both children and adults. This recommendation is consistent with those of Williams et al who reviewed 49 clinical trials involving 13,660 patients. These investigators recommended 7 to 14 days of penicillin or amoxicillin for acute maxillary sinusitis confirmed radiographically or by aspiration.³⁸

The SAHP recommended higher doses of amoxicillin (with or without clavulanate) in patients who have recently taken an antibiotic or who have moderate disease. Fluoroquinolones are recommended as alternatives in patients with mild disease who have not taken an antibiotic in the last 4 to 6 weeks. However, in patients with mild disease who have taken antibiotics recently or who have moderate disease, fluoroquinolones are recommended as first-line therapy. Macrolides are recommended only for patients with a β-lactam allergy since failure rates of 20% to 25% are possible. Lack of improvement or worsening symptoms after 72 hours should prompt reevaluation, may necessitate cultures and/or a CT scan, and should raise the possibility of causal organisms other than S pneumoniae, H influenzae, and M catarrhalis.

TABLE 3

Initial empiric therapy in outpatients with acute bacterial rhinosinusitis

Dose and duration

While each of the three guidelines provides detailed recommendations regarding selection of an antibacterial agent, the dose and duration of therapy generally are not well defined. Fortunately, other sources provide guidance in these 2 areas.

First, an independent international panel of infectious diseases experts, whose goal was to identify ways to improve prescription of antibiotics for lower respiratory tract infections, stressed that an important purpose of therapy is to reduce bacterial load and, in fact, treat to bacteriologic cure.¹⁹ Antibiotic therapy that allows some bacteria to survive increases the risk of early recurrence or relapse and encourages resistance selection. Such therapy is, therefore, inappropriate. The panel concurred with the WHO and others that the likelihood of bacterial persistence increases when antibiotics are prescribed in low doses, especially if given over long periods.^18,39-41 Prolonged low-dose antibiotic therapy, which has been common practice for many infections, is contrary to the WHO Global Strategy for Containment of Antimicrobial Resistance, which notes that single-agent therapy for a short duration is 1 of several actions that can be taken to minimize bacterial resistance.⁴² Shorter courses of antibiotic therapy also are consistent with SHEA/IDSA recommendations.¹⁰

The clinical appropriateness of this recommendation is reinforced by the changes that have occurred during the past decade in the management of selected urinary tract infections (UTIs). Some UTIs that previously had been treated with low-dose antibiotics for 10 to 14 days now are treated with only 1 or a few high doses of a single agent. Other infections for which clinical data support shorter courses of antibiotic therapy include uncomplicated cellulitis,⁴³ ventilator-associated pneumonia,⁴⁴ and meningococcal disease.⁴⁵

Shorter-course antibacterial therapy for CARTIs increasingly has been the focus of clinical trials. The focus is not unreasonable. Many of the antibiotics used for CARTIs are very potent against the pathogens commonly encountered, penetrate infected tissues well, are available in oral formulations, and are generally well tolerated. However, to optimize an antibiotic’s bactericidal potential, it is necessary to base the dosing regimen on its pharmacodynamics. From a pharmacodynamic perspective, there are 2 groups of antibiotics, those with concentration-dependent killing and those with time-dependent killing. For agents with concentration-dependent killing, such as fluoroquinolones, ketolides, and aminoglycosides, the goal is to select a dose that achieves a higher peak concentration and/or a larger area under the plasma concentration curve, with acceptable tolerability. In contrast, antibiotics that rely on time-dependent killing, such as β-lactams, macrolides, azalides, tetracyclines, and some others, require extended durations of concentrations above the MIC90 of the bacterial pathogen(s). Consequently, multiple daily dosing may be preferable.⁴⁶

Dose. A few studies have compared high-dose, short-course therapy with therapy using standard doses and durations (TABLE 4). To assess the impact of highdose, short-course therapy on post-treatment resistant pneumococcal carriage, Schrag compared amoxicillin given either as 90 mg/kg/day for 5 days (high-dose, short-course) or 40 mg/kg/day for 10 days (standard) in 797 children with a respiratory tract infection.⁴⁷ At day 28, nasal carriage of penicillin-resistant S pneumoniae was detected in 24% of the high-dose, shortcourse group and in 32% of the standard group (relative risk, 0.77; P=0.03). Among the pneumococcal carriers, the risk of penicillin-resistant S pneumoniae was significantly lower in the high-dose, short-course group than in the standard therapy group (relative risk, 0.78; P=0.01)

Another study investigated high-dose, short-course therapy with levofloxacin in patients with mild to severe community-acquired pneumonia. Patients received levofloxacin 750 mg/d for 5 days or 500 mg/d for 10 days.⁴⁸ The clinical success rates were 92.4% and 91.1%, respectively, while the bacteriologic eradication rates at 7 to 14 days post-therapy were 93.2% and 92.4%, respectively, thereby demonstrating that high-dose, shortcourse levofloxacin therapy is at least as effective as standard levofloxacin therapy.

Duration. Short-course therapy using standard doses of azithromycin, gemifloxacin, levofloxacin, moxifloxacin, and telithromycin has been investigated in clinical trials of CARTIs (FIGURE). In patients with community-acquired pneumonia, 5 days of therapy with telithromycin was shown to be equivalent to a 7-day course (both using a single daily dose of 800 mg), as well as to clarithromycin 500 mg bid for 10 days.⁴⁹

In studies of acute exacerbations of chronic bronchitis, a 3-day course of azithromycin was equivalent to clarithromycin for 10 days⁵⁰ and gemifloxacin for 5 days was equivalent to a sequential combination of ceftriaxone and cefuroxime axetil for up to 10 days.⁵¹ Levofloxacin for 3 days and azithromycin for 5 days provided equivalent outcomes,⁵² as did levofloxacin for 5 days and amoxicillin/clavulanate for 10 days.⁵² Moxifloxacin for 5 days provided results equivalent to those of 7 days of amoxicillin, clarithromycin, or cefuroxime axetil.⁵³ Five days of telithromycin was shown to be equivalent to 10 days of amoxicillin/clavu-lanate⁵⁴ or cefuroxime axetil.⁵⁵

Studies of acute bacterial rhinosinusitis have demonstrated equivalent results with azithromycin for 3 days and amoxicillin/clavulanate for 10 days.⁵⁰ Similarly, telithromycin for 5 days was equivalent to 10 days of telithromycin, amoxicillin/clavulanate, or cefuroxime axetil.⁵⁶

These clinical trials demonstrate that short-course therapy achieves clinical cure and/or bacteriologic eradication rates that are at least equivalent to those of standard therapy, with no significant difference in safety. Symptomatic improvement is faster and total antibiotic exposure is reduced with short-course therapy.

A significant advantage of short-course antibacterial therapy is improved patient adherence. Adherence is 10% to 20% better with 5-day courses than with 10-day courses,^47,57 and is significantly better with 1 or 2 daily doses than with 3 or more daily doses.^58-60 In fact, a recent market research study showed that patients perceive once-daily, short-course antibiotic treatment to be significantly more effective than longer courses. This may be due to faster improvement of infection-related symptoms.⁶¹ For example, Dunbar et al observed that significantly more patients treated with high-dose, short-course levofloxacin experienced subjective and objective resolution of fever by day 3 compared with those who received standard-dose, short-course levofloxacin.⁴⁸

TABLE 4

Clinical trials of high-dose, short-course antibiotic therapy

TABLE 5 Clinical trials of standard-dose, short-course antibiotic therapy

Summary

Essential questions that need to be answered for every patient who presents with a possible CARTI include : 1) Is antibacterial therapy necessary? 2) If so, what is the best antibiotic and at what dose and for how long should it be administered? Accumulating evidence indicates that some antibiotics when given in high doses for a short duration are as effective and safe as standard therapy for CARTIs. Short-course therapy also promotes patient compliance.

Disclosures:

The authors reported the following financial relationships: Dr Brunton: consultant to Abbott, Ortho-McNeil Pharmaceutical, Inc., and SanofiAventis. Dr Carmichael: consultant to Ortho-McNeil Pharmaceutical, Inc.; on the speakers’ bureaus for Bristol-Myers Squibb Company, Merck & Co., OrthoMcNeil Pharmaceutical, Inc., and Pfizer Inc. Dr Fitzgerald: on the speakers’ bureaus for Boehringer Ingelheim, GlaxoSmithKline, Ortho-McNeil Pharmaceutical, Inc., Pfizer Inc., Sepracor Inc., and 3M. Dr Liu: on the speakers’ bureaus for Aventis Pharmaceuticals, Bayer Pharmaceuticals Corporation, Bristol-Myers Squibb Company, Cobist, GlaxoSmithKline, Merck & Co., Ortho-McNeil Pharmaceutical, Inc., Pfizer Inc., Purdue Pharma, Oscient Pharmaceuticals Corporation, and Wyeth Pharmaceuticals. Dr Varon: on the speakers’ bureau for Ortho-McNeil Pharmaceutical, Inc. Dr. Weiland: consultant to Abbott Laboratories, Ortho-McNeil Pharmaceutical, Inc., and Pfizer Inc.

This supplement to The Journal of Family Practice is supported by a grant from Ortho-McNeil Pharmaceutical, Inc. It was adapted from a consensus conference coordinated by the Primary Care Education Consortium and Texas Academy of Family Physicians and was edited and peer-reviewed by The Journal of Family Practice. © 2005 Quadrant HealthCom Inc. and Primary Care Education Consortium.