Community-Acquired Pneumonia: Empiric Antimicrobial Therapy
ABSTRACT: There are 5 key considerations in selecting antimicrobial therapy for community-acquired pneumonia (CAP): antibiotic activity, tissue penetration, resistance potential, adverse effects, and cost. If the initial empiric antibiotic is chosen carefully, there is no reason to narrow the spectrum of antimicrobial therapy regardless of the cause of the CAP. Properly selected monotherapy is as effective and less problematic (no drug-drug interactions and fewer missed doses) than combination therapy and permits efficient intravenous to oral switch therapy. The duration of empiric antimicrobial therapy for CAP in immunocompetent patients ranges between 7 and 10 days. Patients with underlying cardiopulmonary disease or immunocompromised hosts often require longer courses of treatment of 14 or more days of therapy.
Key words: antibiotic therapy, community-acquired pneumonia, antibiotic resistance
The appropriate therapy for community-acquired pneumonia (CAP) depends on having a correct presumptive diagnosis. If it has been established that the patient has CAP rather than a mimic of CAP, then antimicrobial therapy should be started as soon as CAP is diagnosed.
The empiric therapy of CAP in normal hosts may be approached from several clinical perspectives. First, if the clinician cannot narrow the pathogen range and the pathogen is unknown (excluding Mycobacterium tuberculosis, viral pneumonias, Pneumocystis [carinii] jirovecii, and fungal pneumonias), then coverage should be directed against the common CAP pathogens.1-4 If the clinician can differentiate typical from atypical CAPs, then empiric therapy may be more finely tuned and directed against typical or atypical bacterial CAP pathogens (Table 1). There are 5 key considerations in selecting an antimicrobial therapy: antibiotic activity, tissue penetration, resistance potential, adverse effects, and cost.
(Click to enlarge)
ANTIBIOTIC ACTIVITY
With any treatable bacterial infectious disease, the clinician should consider 5 factors in selecting an antimicrobial for treatment. The first consideration is that the antibiotic selected must exhibit a high degree of activity against the known or presumed pathogen. The antibiotic selected should also have demonstrated clinical efficacy. Some antibiotics are effective in vitro—that is, they appear effective against a given pathogen—but are ineffective in vivo when used clinically. For example, Haemophilus influenzae appears susceptible to penicillin, but penicillin is ineffective against this organism in vivo. If an organism is susceptible in vitro, there is no advantage to selecting an antibiotic with a lower minimum inhibitory concentration against a particular pathogen. If the organism is susceptible, then a lower minimum inhibitory concentration has no clinical relevance if that concentration is within the range of achievable serum and serum/lung concentrations of the selected antibiotic.
The other antibiotic attribute that has no clinical relevance for CAP is whether the antibiotic is bacteriostatic or bactericidal. Bactericidal antibiotics have shown possible benefits in patients with febrile neutropenia, acute bacterial meningitis, or endocarditis. There are exceptions even in these situations, but for practical purposes in treating CAP, the bacteriostatic versus bactericidal profile of an antibiotic is never a therapeutic consideration.1,2,5
TISSUE PENETRATION
The second consideration in selecting an antimicrobial for a treatable infectious disease is its pharmacokinetic attributes—that is, whether the antibiotic as dosed reaches therapeutic concentrations, not only in the serum but also in the target tissue, such as the lungs in the case of CAP. Fortunately, the lungs represent a large, well-vascularized capillary bed that presents no barriers to antibiotic penetration into the lung. In contrast, antibiotic penetration into an abscess and empyemic fluid is difficult.
The considerations are different for patients who have a pulmonary infection or complication.1-3 In addition, for the treatment of Mycoplasma pneumoniae or Chlamydophila (Chlamydia) pneumoniae infection, penetration into the respiratory secretion fluids and surface epithelial cells of the tracheobronchial tree is the key pharmacokinetic consideration after an antimicrobial demonstrates in vitro activity against these organisms.4-6 The treatment of intracellular pathogens, such as those that cause legionnaire’s disease, requires an antibiotic that exhibits activity against the organisms and is able to penetrate the alveolar macrophage to eliminate them.
These situations aside, the achievable serum concentrations are approximately the same as the lung parenchymal concentrations necessary for antibiotics that are used to treat CAP. Some antibiotics, such as daptomycin, that achieve therapeutic concentrations in the lungs may not be effective in treating CAP. Similarly, aminoglycosides achieve therapeutic lung concentrations but are inactivated by local conditions in infected areas of the lung, including local hypoxia, cellular debris, and tissue acidosis. In such situations, although the aminoglycoside lung concentrations are adequate, aminoglycoside activity is greatly reduced by these factors, which may result in therapeutic failure.1,2,6
(Click to enlarge)
RESISTANCE POTENTIAL
The third consideration in selecting an antimicrobial is its resistance potential. This is the most overlooked factor by clinicians who are primarily concerned with treating their patients. However, clinicians should also consider the potential risk of drug resistance associated with the antibiotics they prescribe. It does no good to cure the individual patient while promoting widespread drug resistance in the community through the use of certain “high resistance potential” antibiotics, such as trimethoprim-sulfamethoxazole, macrolides, and ciprofloxacin.
For the treatment of typical CAP, cephalosporins have a “low resistance potential,” which is therefore not a consideration in selecting any of these agents. Macrolides should be avoided, since they are associated with both natural resistance in 25% to 35% cases of Streptococcus pneumoniae infection and acquired resistance related to the drugs’ widespread use. In treating atypical CAPs, the “respiratory quinolones” such as levofloxacin and moxifloxacin, but excluding ciprofloxacin, have a low resistance potential, and their use has not been associated with increased resistance from S pneumoniae or from any other typical or atypical CAP pathogen.2,7-10
ADVERSE EFFECTS
The fourth consideration in selecting an antimicrobial is its safety profile. Adverse reactions may occur with any drug, but the clinician should try to avoid those antibiotics whose adverse effects either are uncommon but severe or are not severe but common. Most of the antibiotics used for the treatment of CAP are fairly safe for most patients. Obviously, patients who are allergic to various antibiotic classes should not be receiving drugs from those classes. Virtually none of the commonly used antibiotics for CAP is neurotoxic, hepatotoxic, or nephrotoxic. Drug-drug interactions are relatively uncommon, as well, and the adverse effects associated with beta-lactam antibiotics in the treatment of typical CAP are largely related to either drug-related fever/rash or Clostridium difficile diarrhea.
Respiratory quinolones are rarely associated with C difficile diarrhea (excluding ciprofloxacin, a common cause of C difficile diarrhea) unless a patient is receiving concomitant proton pump inhibitor (PPI) therapy, such as esomeprazole. The risk of C difficile diarrhea with either PPI therapy alone or respiratory quinolone therapy alone appears to be low, but when these two drugs are combined, the incidence is high. Patients already taking PPIs who are about to receive respiratory quinolone therapy should either discontinue the PPI regimen during antibiotic therapy or switch to H2 blocker therapy, such as famotidine, for the duration of treatment.1,2,11
COST CONSIDERATIONS
The fifth consideration is the cost of therapy. In the hospital, intravenous (IV) antibiotics for CAP are more expensive than their oral counterparts. The cost of IV antibiotics must take into account the acquisition cost of the antibiotic as well as the cost of the IV administration charged by the institution, which is usually $10 per dose. The costs should also take into account a drug’s resistance potential, which has public health as well as economic implications. If a patient requires hepatic or renal monitoring, then the associated costs must be factored into the cost of therapy. Last, if the drug is associated with adverse effects, such as C difficile diarrhea, then the cost of increased length of stay as well as the cost of re-treating the pneumonia with an antibiotic that does not have a high C difficile potential must be considered. In addition, the cost of the treatment of C difficile infection itself must be factored into the total cost of antimicrobial therapy for the patient and the institution.1,2,12
(Click to enlarge)
OTHER THERAPEUTIC CONSIDERATIONS
Aside from an antimicrobial’s bacteriostatic versus bactericidal profile, another notion that causes much clinical confusion is that of the broadness or narrowness of the antimicrobial’s spectrum of coverage. If a patient with CAP begins empiric antimicrobial therapy with a cephalosporin and is subsequently found to have CAP due to S pneumoniae, some clinicians would feel compelled to “narrow the spectrum” and complete treatment with penicillin. Neither clinical experience nor study data suggest that this is a valid concept. There is no difference in outcome in treating S pneumoniae with ceftriaxone versus penicillin. There is no advantage in narrowing the spectrum. A drug selected for empiric therapy, if it has a low resistance potential, may be given indefinitely without a considerable increase in resistance potential. Ceftriaxone is given less often and is less expensive than parenteral penicillin G therapy.
There is simply no rationale for narrowing the spectrum of the initial antibiotic selected if the 5 factors, cited above, are taken into account. The clinical “take home” point is that if you select your initial empiric antibiotic carefully, there will be no reason to narrow or de-escalate antimicrobial therapy regardless of the identified cause of the CAP (Tables 2, 3, 4, 5, and 6).1,2
IV, IV-TO-ORAL, OR ORAL ANTIBIOTIC THERAPY
The other therapeutic consideration that sometimes perplexes clinicians is whether the antibiotic should be administered intravenously or orally. Usually, patients with CAP who are admitted to the hospital undergo empiric treatment with a parenteral antibiotic. After they improve clinically, their therapy is switched to an equivalent oral antibiotic therapy to complete either in or outside the hospital. Initial empiric therapy may be given by mouth, unless the patient is in shock or has an altered gastrointestinal tract or otherwise grossly impaired gastric absorption. In such cases, therapeutic serum/lung concentrations are achieved with most antibiotics given orally within 1 hour.
The antibiotic selected for oral therapy for CAP must have high bioavailability (more than 90%), which means that the serum/lung drug concentration achievable with that oral antibiotic will approximate that of the same dose of the drug administered intravenously. There is no difference in outcome whether the therapy is administered totally intravenously, through an IV-to-oral switch program, or completely orally.
The initial preference for an IV antibiotic is related to severity of the CAP. After the patient improves, unless there are compelling reasons not to, the patient should be switched to an oral equivalent antibiotic as soon as possible and therapy should be completed with that antibiotic. The selection of an oral antibiotic, bioavailability aside, should take into account the same 5 aforementioned principles of antibiotic selection.1,2,13-20
MONOTHERAPY VERSUS COMBINATION ANTIBIOTIC THERAPY
The next therapeutic consideration to be decided is whether a patient should undergo empiric treatment with combination therapy or with monotherapy. In infectious disease in general, there is little rationale supporting combination therapy. Properly selected monotherapy is as effective, less expensive, and less problematic (involving no drug-drug interactions and fewer missed doses, for example) than combination therapy. The rationale behind combination therapy is that it either prevents resistance (which is incorrect), or provides additional coverage (which is occasionally correct), or increases the spectrum of one drug (correct if the physician cannot differentiate typical from atypical pathogens). As mentioned previously, well-selected monotherapy solves all of these problems.
There are no data indicating that combination therapy decreases the emergence of resistance from any of the pulmonary pathogens. Decreased resistance with combination therapy has been shown only with antipseudomonal penicillin and aminoglycoside against Pseudomonas aeruginosa (not a CAP pathogen). Synergy is not necessary against any of the CAP pathogens, thus negating this potential rationale for double-drug therapy.1,2,5,13
Double-drug therapy is commonly used to cover both typical and atypical pathogens. For example, a beta-lactam will be prescribed to cover the typical CAP pathogens and azithromycin will be added to cover the atypical CAP pathogens. If a clinician administers empiric monotherapy with either doxycycline or a respiratory quinolone, double therapy will be unnecessary. Respiratory quinolones are as efficacious against the typical bacterial CAP pathogens as ceftatriaxone. Similarly, doxycycline or a respiratory quinolone is much more efficacious than azithromycin as part of a combination regimen.
Macrolides should not be used as monotherapy because of their associated S pneumoniae resistance and their relative lack of activity against H influenzae as well as their minimal activity against Chlamydophila (Chlamydia) psittaci. As mentioned previously, well-selected empiric monotherapy with doxycycline or a respiratory quinolone provides optimal therapy with minimal resistance potential at a lower cost than with combination therapy, such as ceftriaxone plus azithromycin.
Another advantage to well-chosen monotherapy with doxycycline or a respiratory quinolone is that it makes the switch from IV to oral therapy simple. If the clinician begins empiric double drug therapy with ceftriaxone plus azithromycin and wants to switch to oral therapy (pathogen remains undiagnosed), it is difficult because there is no oral ceftriaxone preparation. Since azithromycin has the limitations cited above, it should not be used as monotherapy unless the patient has either C pneumoniae or M pneumoniae. Therefore, the clinician is forced to choose an oral cephalosporin that is not expensive and to continue with oral azithromycin therapy. This means that initial parenteral double-drug therapy must necessarily be followed by oral double-drug therapy, another disadvantage of double-drug therapy with ceftriaxone plus azithromycin. Since doxycycline and respiratory quinolones demonstrate excellent bioavailability with a low resistance potential and are available in IV and oral formulations, both are ideal agents for the empiric monotherapy of CAP in hospitalized patients (initial IV therapy followed by IV-to-oral therapy) or in outpatients (to complete oral therapy).1,2,4,5,21-25
APPROACH TO APPARENT ANTIBIOTIC FAILURE
Clinicians are often faced with a patient who has failed to respond to seemingly effective antimicrobial therapy. Clinicians should consider several factors when trying to clinically evaluate the potential for therapeutic failure. First, they should determine whether their patient did in fact have typical or atypical CAP or whether he or she presented with a noninfectious mimic of CAP, such as systemic lupus erythematosus pneumonitis, pulmonary drug reaction, congestive heart failure, a pulmonary embolus or infarct, radiation pneumonitis, or bronchogenic carcinoma. If the diagnosis of CAP is secure, then the clinician should consider first whether the patient had received the optimal antimicrobial therapy and, second, whether the pneumonia had sufficient time to respond to the therapy.
Patients and physicians seem to expect nearly immediate resolution of pneumonia. All infectious diseases take time to respond, and this time varies by host defense status and the pathogen involved. All other things being equal, immunocompromised hosts will take longer to respond to treatment than normal hosts with the same infection. Legionnaire’s disease takes longer to respond to appropriate antimicrobial therapy than does pneumococcal pneumonia.
The findings on chest films always lag behind clinical improvement; therefore, the clinician should look to a decreased temperature and peripheral white blood cell count as the first indicators of improvement. If a patient appears to be clinically improving and the chest films do not show worsening but remain approximately the same, then further improvement may be expected. On the other hand, if the patient defervesces and after several days the fever reappears, the clinician should consider either drug-related fever or a complication of pneumonia as a potential explanation.
The patient with drug-related fever will appear “relatively well” and yet will have fever while improving or deteriorating clinically. While on therapy, if the patient with a new fever is not receiving beta-blocker, verapamil, or diltiazem therapy and there is a pulse temperature deficit (relative bradycardia), then the diagnosis of drug-related fever is the most likely explanation. Eosinophils are often on the peripheral smear but eosinophilia is less common. The drug responsible for drug-related fever is most often not the antimicrobial and is often a sulfa-containing diuretic, such as furosemide, or a stool softener, such as docusate. Doxycycline and respiratory quinolones are rare causes of drug-related fever, in contrast to beta-lactam therapy, which is a common cause.
If a patient has empyema, it will be readily recognizable on chest radiographs as a pleural effusion. The patient with empyema will have persistent fever, leukocytosis, and the appearance of pleural effusion on the chest film. Such patients should undergo chest tube drainage in addition to antimicrobial therapy as part of their treatment. The CAP organism most commonly associated with empyema is S pneumoniae.1,2
(Click to enlarge)
DURATION OF THERAPY
The duration of empiric antimicrobial therapy for CAP in immunocompetent patients ranges between 7 and 10 days. Healthy individuals without underlying systemic disorders may receive either IV/oral or oral-only treatment for 7 days. Patients with underlying cardiopulmonary or renal disease may require a longer course of treatment lasting 14 or more days.
Similarly, immunosupressed patients or those undergoing immunomodulating therapies often require longer treatment. As a general rule, patients infected with intracellular pathogens, as in the case of legionnaire’s disease, may require a longer course of antimicrobial therapy to prevent relapse. For this reason, normal and immunocompromised patients with legionnaire’s disease commonly undergo 3 weeks of therapy.
The initiation of therapy is a relatively unimportant consideration once a patient begins treatment with an oral antibiotic. Initial treatment with an oral antibiotic or early transition from an IV to oral antibiotic permits a shorter length of stay and an earlier discharge from the hospital. Oral empiric antimicrobial therapy also permits the treatment of non-severe CAP in the outpatient setting and often prevents by early treatment subsequent admission to a hospital.
Patients who have received treatment for pneumonia should undergo repeat chest radiography. No further radiographs will be necessary once they have shown clinical improvement, unless fever subsequently develops, such as from possible empyema. After that initial improvement, serial chest films are largely unnecessary. Clinicians often forget that with S pneumoniae CAP, some chest film findings may persist for as long as 14 weeks after the infection. Therefore, if a patient clinically improves and defervesces, the clinician need only decide on the duration of therapy (7 to 14 days). This decision is based on the patient’s rate of improvement, the pathogen, and the underlying host factors.1-4,26
(Click to enlarge)
REFERENCES:
1.Cunha BA (ed). Pneumonia Essentials (3rd ed). Sudbury, MA: Jones and Bartlett; 2010.
2.Cunha BA (ed). Antibiotic Essentials (12th ed). Sudbury. MA: Jones and Bartlett; 2013.
3.Cunha BA. Antibiotic pharmacokinetic considerations in pulmonary infections. Semin Respir Infect. 1991;6:168-182.
4.Cunha BA. Empiric antibiotic therapy for community acquired pneumonia: guidelines for the perplexed? Chest. 2004;125:1913-1921.
5.Mandell LA, Wunderink RG, Anzueto A, et al. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community acquired pneumonia in adults. Clin Infect Dis. 2007;1:S27-S72.
6.Cunha BA. Factors in antibiotic formulary selection: pharmacokinetics/pharmacodynamic factors (part II). Pharmacy & Therapeutics. 2003;28:468-470.
7.Rudin JE, Evans TL, Wing EJ. Failure of erythromycin in treatment of Legionella mecdadei pneumonia. Am J Med. 1984;76:318-320.
8.Sabria M, Pedro-Botet ML, Gomez J, et al. Fluoroquinolones vs macrolides in the treatment of Legionnaire’s disease. Chest. 2005;128:1401-1405.
9.Cunha BA. Antibiotic resistance: effective control strategies. Lancet. 2001;357:1307-1308.
10.Cunha BA. Factors in antibiotic formulary selection: antibiotic resistance (part III). Pharmacy & Therapeutics. 2003;28:524-527.
11.Cunha BA. Factors in antibiotic formulary selection: antibiotic adverse effects (part IV). Pharmacy & Therapeutics. 2003;28:594-596.
12.Cunha BA. Factors in antibiotic formulary selection: antibiotic adverse effects (part V). Pharmacy & Therapeutics. 2003;28:662-665.
13.Cunha BA. Community acquired pneumonia. Diagnostic and therapeutic approach. Med Clin North Am. 2001;85:43-77.
14.Quintiliani R, Nightingale CH. Transitional antibiotic therapy. Infect Dis Clin Pract. 1994;3:161-167.
15.Sensakovic JW, Smith LG. Oral antibiotic treatment of infectious diseases. Med Clin North Am. 2001;85:115-123.
16.Cunha BA. Intravenous to oral antibiotic switch therapy. Drugs Today. 2001;37:311-319.
17.Cunha BA. Doxycycline re-visited. Arch Intern Med. 1999;158:1006-1007.
18.Cunha BA. Doxycycline for community acquired pneumonia. Clin Infect Dis. 2003;37:870.
19.Mykietiuk A, Carratal J, Fernandex-Sabe N, et al. Clinical outcomes for hospitalized patients with Legionella pneumonia in the antigenuria era: the influence of levofloxacin therapy. Clin Infect Dis. 2005;15:794-799.
20.Tanaseanu C, Milutinovic S, Calistru Pl, et al. Efficacy and safety of tigecycline versus levofloxacin for community acquired pneumonia. BMC Pulm Med. 2009;9:44.
21.Choudhury G, Mandal P, Singanayagam A, et al. Seven-day antibiotic courses have similar efficacy to prolonged courses in severe community acquired pneumonia—a propensity-adjusted analysis. Clin Microbiol Infect. 2011;17:1852-1858.
22.Teh B, Grayson ML, Johnson PD, Charles PG. Doxycycline vs. macrolides in combination therapy for treatment of community acquired pneumonia. Clin Microbiol Infect. 2012;18:E71-E73.
23.Le Bris-Tomczak A, Bedos JP, Billon C, et al. Antibiotic strategy in severe community acquired pneumococcal pneumonia. Med Mal Infect. 2012;42:226-234.
24.Wilcox MH, Finch R, Wyncoll D, et al. Fluoroquinolones in the treatment of severe community acquired pneumonia. Br J Hosp. 2011:12:S1-S7.
25.Pedreo Botet ML, Yu VL. Treatment strategies for Legionella infection. Expert Opin Pharmacother. 2009;10:1109-1121.
26.Siegel RE, Halpern NA, Almenoff PL, et al. A prospective randomized study of inpatient intravenous antibiotics for community acquired pneumonia: the optimal duration of therapy. Chest. 1996;110:965-971.