Abstract: In the assessment of community-acquired pneumonia, an effort should be made to identify the causal pathogen, since this may permit more focused treatment. However, diagnostic testing should not delay appropriate empiric therapy. The selection of empiric therapy can be guided by a patient stratification system that is based on the severity of illness and underlying risk factors for specific pathogens. For example, outpatients who do not have underlying cardiopulmonary disease or other risk factors can be given azithromycin, clarithromycin, or doxycycline. Higher-risk outpatients should be given a ß-lactam antibiotic plus azithromycin, clarithromycin, or doxycycline, or monotherapy with a fluoroquinolone. If the patient fails to respond to therapy, it may be necessary to do bronchoscopy; CT of the chest; or serologic testing for Legionella species, Mycoplasma pneumoniae, viruses, or other pathogens. (J Respir Dis. 2006;27(2):54-67)
Pneumonia, an acute inflammation of the lung parenchyma, is a common and potentially serious disease. In the United States, the estimated annual incidence of pneumonia is 12 cases per 1000 population, and the estimated costs exceed $20 billion.1 Case rates for community- acquired pneumonia (CAP) have been estimated to be 258 per 100,000 population. This rate rises significantly, however, to 962 per 100,000 population among those aged 65 and older.2
From 1974 to 1994, the rate of pneumonia and influenza increased by 59%, partly a reflection of an aging population more susceptible to these diseases and more likely to die of them.3 Increased incidences of CAP have been observed among patients who have coexisting illnesses, such as chronic obstructive pulmonary disease (COPD), diabetes mellitus, renal insufficiency, congestive heart failure (CHF), and chronic liver disease.
The epidemiology and treatment of CAP have evolved. Developments include an increased number of newly identified or previously unrecognized pathogens, such as Chlamydia pneumoniae and Hantavirus; the increasing incidence of community-acquired methicillin-resistant Staphylococcus aureus (MRSA) infections; the increased risk of "global" pathogens, such as severe acute respiratory syndrome-associated coronavirus and the avian influenza virus strain H5N1; and the risk of exposure to pathogens such as anthrax that result from bioterrorism. The development of new methods of microbial detection, such as polymerase chain reaction, and the development of a number of new antimicrobials, such as ß-lactams, macrolides, fluoroquinolones, oxazolidinones, and streptogramins, are rapidly evolving as well.
Concurrent with these advances, however, has been the evolution of bacterial resistance mechanisms. In addition to MRSA, increased resistance is being identified among Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis, and a number of Gram-negative organisms.
In this article, we will review current practices in the management of CAP and examine ways to increase the effectiveness of diagnosis and therapy.
ETIOLOGY OF CAP
The microbiology of CAP varies with age and severity of disease. In addition, comorbidities and risk factors contribute to the variation among the potential pathogens.
A number of studies have found S pneumoniae to be the most frequently isolated organism in cases of CAP in which the offending pathogen was identified.4-9H influenzae and respiratory viruses are also common. Cases of CAP caused by atypical pathogens, such as Legionella species, Mycoplasma pneumoniae, and C pneumoniae, are frequent but varied. This variation may reflect differences in the population, location, and laboratory techniques used.
Table 1 lists 6 studies that examined causes of CAP in hospitalized patients.4-9 Five of these studies found the pathogens for outpatients and inpatients to be similar.4-7,9 In one ambulatory treatment series, in which serology alone was used to determine the cause among a relatively younger population without comorbidities, atypical pathogens accounted for nearly 50% of cases.10 In addition, various studies have estimated that the proportion of patients hospitalized for CAP who were infected with more than 1 pathogen ranged from 3% to almost 40%.6,11
The microbiology of severe CAP in patients admitted to the ICU is similar to that of non-ICU patients. S pneumoniae is the most frequently isolated pathogen, and it is increasingly common with advancing age. Two thirds of all cases of bacteremic pneumonia are caused by pneumococci, with an estimated mortality rate of 6% to 20%.
H influenzae and Gram-negative pathogens are also frequently isolated in cases of CAP.12 In patients with underlying diseases, such as COPD, Gram-negative infections tend to occur. Pseudomonas aeruginosa is rarely found in immunocompetent patients who do not have structural lung disease. Between 30% and 60% of cases do not yield an identifiable pathogen.13
Atypical pathogens, such as C pneumoniae and M pneumoniae, occur frequently among younger patients. Legionella species is most prevalent among 35- to 49-year-olds who have epidemiologic risk factors, including exposure to spas, recent travel and an overnight stay outside the home, renal or hepat- ic failure, diabetes, and systemic malignancy.14-16
The American Thoracic Society (ATS) published guidelines in 1993 and 2001 to assist the clinician in treating outpatients and inpatients with CAP and to provide criteria for stratifying patients according to severity of disease.17 The guidelines also list underlying risk factors for specific pathogens (Table 2) and contain treatment recommendations for all patient strata and indications for hospitalization.
The ATS guidelines emphasize the value of proper diagnostic assessment. However, because of the difficulty in identifying specific pathogens, the ATS recommends an initial empiric approach to therapy. Starting appropriate treatment should not be delayed, and the results of the initial workup should guide further management.
In 2000, the Infectious Diseases Society of America (IDSA) proposed guidelines that may lead to more efficient care of patients with CAP.18 The IDSA guidelines recommend an initial chest radiograph for all patients in whom CAP is suspected. The chest radiograph is particularly important in the assessment of hospitalized patients.
The main distinction between the ATS and IDSA guidelines is that the IDSA puts greater emphasis on establishing the cause of CAP. According to the IDSA, identifying the specific cause not only helps guide treatment but also provides data for following CAP patterns in a given community.
The IDSA guidelines call for diagnostic studies, including sputum Gram stain, which makes the approach more pathogen-directed. When the pathogen is unknown, empiric therapy should be based on 6 factors:
Severity of disease.
Previous antibiotic therapy.
The ATS guidelines describe the spectrum of etiologic agents and the initial approach to therapy using patient stratification.17 This approach separates patients into categories that are based on place of therapy (outpatient setting, hospital ward, or ICU), coexisting cardiopulmonary disease (COPD, CHF), and modifying factors (risk factors for drug-resistant S pneumoniae [DRSP] and P aeruginosa and other Gram-negative organisms). It is important to note that patients at risk for HIV infection are excluded from these patient categories.
This approach to the choice of empiric therapy takes into account the severity of illness (reflected in the location of therapy) and the presence of risk factors that may predispose the patients to specific pathogens.
Outpatients with no cardiopulmonary disease and no modifying factors: The most common pathogens include S pneumoniae, M pneumoniae, C pneumoniae, and respiratory viruses. Other pathogens include Legionella species, Mycobacterium tuberculosis, and endemic fungi.
Outpatients with cardiopulmonary disease and/or one or more modifying factors: The presence of comorbidities and risk factors changes the likely pathogens. Although pneumococcus remains the most common pathogen, the probability of DRSP and other pathogens is increased and should be considered when choosing antibiotics.
Pneumococcal resistance has increased to a point that is clinically significant in the following classes of antibiotics: ß-lactams (penicillins, cephalosporins, and carbapenems); macrolides (erythromycin, azithromycin, and clarithromycin); lincosamines (clindamycin); tetracyclines and folate inhibitors (trimethoprim-sulfamethoxazole); and fluoroquinolones. Current information about the effectiveness of treatment of patients with DRSP infection is mixed. Several studies suggest that systemic, nonmeningeal infections with DRSP strains respond well to usual doses of ß-lactam antibiotics.19,20 However, other studies have found that the use of ß-lactam antibiotics for such infections is associated with an increased likelihood of mortality, especially among the elderly and patients with underlying diseases.21,22
If the patient is from a nursing home and has bronchiectasis, the possibility of infection with a Gram-negative organism (such as Escherichia coli,Klebsiella species, or P aeruginosa)should be considered. If the patient has poor dentition, impaired consciousness, or a swallowing disorder, aspiration of anaerobes should be considered. Less common pathogens include M catarrhalis,Legionella species, Mycobacterium species, and endemic fungi.
Inpatients not in the ICU: Hospitalized patients usually have risk factors for DRSP or enteric Gram-negative organisms, and they may have underlying cardiopulmonary disease; these factors influence the likely pathogens. These patients are at risk for infection with pneumococci (including DRSP), H influenzae, atypical pathogens, enteric Gram-negative organisms (such as Enterobacteriaceae), and polymicrobial bacterial flora (including anaerobes if risk of aspiration is present). The incidence of Gram-negative infection is generally not high for hospitalized patients with CAP, but it rises among patients admitted to the ICU.
The possibility of tuberculosis should be considered in patients who are from countries in which this disease is endemic, in alcoholic patients, and in nursing home residents.Endemic fungal infections are also possible in specific epidemiologic settings, such as coccidioidomycosis in the southwestern United States and histoplasmosis in the eastern United States.
In a hospitalized patient without cardiopulmonary disease or modifying risk factors, the most likely pathogens are S pneumoniae,H influenzae,M pneumoniae,C pneumoniae, respiratory viruses, and Legionella species. Up to 40% of hospitalized patients may have a polymicrobial infection (bacterial and atypical pathogens).
ICU patients: The pathogens most frequently identified in patients with severe pneumonia are S pneumoniae (including DRSP), Legionella species, H influenzae, enteric Gram-negative bacilli, S aureus, M pneumoniae, respiratory viruses, and certain miscellaneous pathogens (such as C pneumoniae, M tuberculosis, and endemic fungi). P aeruginosa should be considered only when specific risk factors are present (Table 2). Risk factors for S aureus infection include diabetes, renal failure, and recent influenza virus infection.
One should consider the diagnosis of pneumonia in any patient who has a productive cough, with or without dyspnea, or pleuritic chest pain. These acute respiratory symptoms may be accompanied by fe-ver, abnormal breath sounds, and crackles on auscultation.
For patients with symptoms and examination findings that suggest pneumonia, chest radiographs (standard posteroanterior and lateral views) are valuable. In some instances, radiographic findings may suggest specific causes, such as tuberculosis or lung abscess. The presence of pleural effusions or multilobar pneumonia on chest radiographs can help evaluate the severity of illness.
Once the diagnosis of CAP is made, it is important to try to establish a specific cause with appropriate testing. However, diagnostic testing should not delay appropriate empiric therapy. As noted above, the specific cause may not be identified in more than 50% of cases, and a significant number of patients have polymicrobial infection. These factors diminish the value of focused therapy guided by diagnostic testing.
As advocated by the ATS and IDSA guidelines, a Gram stain and culture of properly collected expectorated sputum may be helpful. A good sputum sample should have fewer than 10 squamous epithelial cells and more than 25 neutrophils per low-power field. When using sputum testing to determine initial therapy, it is important to remember its limitations; these include the potential for inadequate sampling, subjective interpreta-tion, and the inability to detect atypical pathogens.
Sputum cultures are known to have low sensitivity and specificity, but they may be particularly helpful when DRSP or other resistant pathogens (such as S aureus)are suspected. Although the overall yield of blood cultures is only about 11%, with severe CAP it is recommended that 2 sets of blood cultures be drawn before the initiation of antibiotic therapy. Pneumococcal urinary antigen assay is a rapid, simple test, recently approved by the FDA, for augmenting standard diagnostic methods of sputum Gram stain and blood cultures.23-25
Routine laboratory tests, such as complete blood cell counts, serum electrolyte levels, liver and renal function, and oxygen saturation, may help define the severity of disease and influence decisions about the appropriate place of treatment. An initial investigation should also include HIV serologic testing and M tuberculosis or fungal cultures based on the patient's epidemiologic profile. If there is epidemiologic evidence of Legionnaires disease, the preferred diagnostic tests are the urinary antigen assay and a culture of respiratory secretions.16
The decision to hospitalize
Risk factors associated with a complicated course or with death have been identified by several studies.1,26-28 These factors include the following:
Age 65 years and older.
Presence of coexisting illnesses, such as malignancy, COPD, chronic renal failure, CHF, chronic liver disease, or cerebrovascular disease.
Specific physical findings, such as respiration rate greater than 30 breaths per minute, systolic blood pressure less than 90 mm Hg, heart rate greater than 125 beats per minute, temperature lower than 35°C (95°F) or higher than 40°C (104°F), or decreased level of consciousness.
Specific laboratory findings, such as white blood cell count lower than 4 3 109/L or higher than 30 3 109/L, absolute neutrophil count lower than 1 3 109/L, PaO2 less than 60 mm Hg, evidence of abnormal renal function with serum creatinine level greater than 1.2 mg/dL or blood urea nitrogen level greater than 20 mg/dL, multilobar pneumonia or pleural effusion on a chest radiograph, hematocrit value of less than 30%, or arterial pH less than 7.35.
These risk factors have been incorporated into the prediction rules developed by the British Thoracic Society Research Committee26 and the Pneumonia Patient Outcome Research Team.27 The resulting protocols classify patients as at high or low risk for death and can be used to determine the need for admission.
When multiple factors are present in the patient, hospitalization should be considered. This does not mean long-term care; hospitalization permits closer observation of the patient's response to therapy, which can then be continued in an outpatient setting.
Other factors should be considered in the decision to hospitalize. These include the availability of outpatient services; the ability to maintain oral intake; cognitive impairment or a history of substance abuse; and social considerations, such as the absence of a responsible caregiver within a stable environment.
The ATS and IDSA guidelines provide the physician with a rational approach to the initial antimicrobial management of CAP, based on the assessment of severity of disease and underlying risk factors. The stratification of patients provides a framework for selecting empiric therapy. The initiation of therapy should not be delayed, and excessively broad antibiotic therapy should be avoided if it is not needed.
The use of these guidelines has been shown to help physicians decrease mortality among patients with CAP admitted to the hospital.29 Therapies for outpatients with no cardiopulmonary disease or modifying risk factors include an advanced-generation macrolide or doxycycline (Table 3). Outpatients with cardiopulmonary disease and/or 1 or more modifying risk factors can be given a ß-lactam plus an advanced-generation macrolide or doxycycline, or monotherapy with an antipneumococcal fluoroquinolone (Table 4).
Treatment recommendations for hospitalized non-ICU patients are more nuanced, differentiating between patients who have cardiopulmonary disease and/or modifying risk factors and those who do not (Table 5). Treatment recommendations for ICU patients depend on whether they are at risk for infection with P aeruginosa (Table 6).
The ATS and IDSA guidelines list multiple alternative therapies, with no particular order of preference. The alternative regimens are safe and effective, and it is not clear whether there is any advantage in choosing one over another.
Antipseudomonal agents are not recommended for routine use unless the patient has risk factors for P aeruginosa infection. Vancomycin also should have a limited role in empiric therapy. It should be used only in patients who have pneumonia caused by DRSP (associated with a high level of resistance) and in whom other therapies are failing or who have meningitis, and in patients with severe pneumonia who reside in a nurs-ing home known to harbor MRSA.
The decision to switch from intravenous to oral antibiotic therapy is based on clinical response (Table 7). Once the patient has stabilized clinically (usually within 3 days), he or she can be switched to oral therapy. Generally, patients who have CAP should be treated for 7 to 10 days; those with atypical CAP should be treated for 10 to 14 days. The severity of illness, coexisting illness, bacteremia, and hospital course should also be considered when determining the duration of therapy.
When a patient fails to respond, examination of a lower respiratory tract secretion obtained by bronchoscopy may help identify unusual organisms, such as M tuberculosis, Pneumocystis jiroveci, and drug-resistant pathogens. CT may provide useful information, such as the presence of cavitation, the degree of parenchymal involvement, andthe presence of pleural fluid.
Pulmonary embolism with infarction should be considered in patients at risk, such as those who are critically ill or immobile, or those receiving mechanical ventilation. Serologic testing for Legionella species, M pneumoniae, viruses, or other unusual pathogens should also be considered. For a seriously ill patient who is not responding to therapy and for whom extensive diagnostic evaluation has not been helpful, an open lung biopsy can be considered.
Important prevention strategies include influenza and pneumococcal vaccination and smoking cessation. The pneumococcal and influenza vaccines have been shown to be safe and effective, and they should be given to patients at risk for CAP when appropriate.
All immunocompetent patients aged 65 and older should receive the pneumococcal vaccine; it contains the purified capsular polysaccharide from 23 serotypes that cause up to 90% of invasive pneumococcal infections in the Unit- ed States. Patients aged 64 and younger should be vaccinated if they have chronic illnesses, such as CHF, COPD, diabetes mellitus, alcoholism, liver cirrhosis, or functional or anatomic asplenia.
Although the effectiveness of the pneumococcal vaccine is uncertain among immunosuppressed patients, it is recommended that they be immunized. This group includes persons with HIV infection, leukemia, lymphoma, Hodgkin disease, multiple myeloma, or chronic renal failure, and persons receiving immunosuppressive therapy.
A single revaccination is indicated in patients 65 years and older who received the vaccine more than 5 years previously and who were younger than 65 when first vaccinated. In immunocompromised patients, revaccination after 5 years is indicated.
Persons at increased risk for influenza complications and those who can transmit the illness to high-risk persons should receive the influenza vaccine yearly (from September through November). Populations at increased risk include persons 65 years and older; residents of nursing homes and chronic-care facilities; persons with chronic cardiopulmonary disease; patients who required frequent medical care because of conditions such as diabetes mellitus, renal failure, or immunosuppression; persons hospitalized within the past year; pregnant women whose second or third trimester falls during influenza season; and persons with HIV infection.
Cigarette smoking is a risk factor for pneumonia. Smoking cessation is therefore an important prevention strategy that should be incorporated into all comprehensive treatment plans.
1. Marrie TJ. Community-acquired pneumonia.
Clin Infect Dis.
2. Centers for Disease Control and Prevention. Premature deaths, monthly mortality and monthly physician contacts: United States.
3. Ely EW. Pneumonia in the elderly: diagnostic and therapeutic challenges.
4. Levy M, Dromer F, Brion N, et al. Community-acquired pneumonia. Importance of initial noninvasive bacteriologic and radiographic investigations.
5. Woodhead MA, Macfarlane JT, McCracken JS, et al. Prospective study of the aetiology and outcome of pneumonia in the community.
6. Fang GD, Fine M, Orloff J, et al. New and emerging etiologies for community-acquired pneumonia with implications for therapy. A prospective multicenter study of 359 cases.
7. Lim I, Shaw DR, Stanley DP, et al. A prospective hospital study of the aetiology of community-acquired pneumonia.
Med J Aust.
8. Porath A, Schlaeffer F, Lieberman D. The epidemiology of community-acquired pneumonia among hospitalized adults.
9. Bohte R, van Furth R, van den Broek PJ. Aetiology of community-acquired pneumonia: a prospective study among adults requiring admission to hospital.
10. Marrie TJ, Peeling RW, Fine MJ, et al. Ambulatory patients with community-acquired pneumonia: the frequency of atypical agents and clinical course.
Am J Med.
11. Lieberman D, Schlaeffer F, Boldur I, et al. Multiple pathogens in adult patients admitted with community-acquired pneumonia: a one year prospective study of 346 consecutive patients.
12. Torres A, Serra-Batlles J, Ferrer A, et al. Severe community-acquired pneumonia. Epidemiology and prognostic factors.
Am Rev Respir Dis.
13. Niederman MS, Bass JB Jr, Campbell GD, et al. Guidelines for the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, and initial antimicrobial therapy. American Thoracic Society. Medical Section of the American Lung Association.
Am Rev Respir Dis.
14. Marston BJ, Lipman HB, Breiman RF. Surveillance for Legionnaires' disease. Risk factors for morbidity and mortality.
Arch Intern Med.
15. Benin AL, Benson RF, Arnold KE, et al. An outbreak of travel-associated Legionnaires disease and Pontiac fever: the need for enhanced surveillance of travel-associated legionellosis in the United States.
J Infect Dis.
16. Helbig JH, Uldum SA, Bernander S, et al. Clinical utility of urinary antigen detection for diagnosis of community-acquired, travel-associated, and nosocomial legionnaires' disease.
J Clin Microbiol.
17. American Thoracic Society. Guidelines for the management of adults with community-acquired pneumonia. Diagnosis, assessment of severity, antimicrobial therapy, and prevention. American Thoracic Society.
Am J Respir Crit Care Med.
18. Bartlett JG, Dowell SF, Mandell LA, et al. Practice guidelines for the management of community-acquired pneumonia in adults. Infectious Diseases Society of America.
Clin Infect Dis.
19. Yu VL, Chiou CC, Feldman C, et al. An international prospective study of pneumococcal bacteremia: correlation with in vitro resistance, antibiotics administered, and clinical outcome.
Clin Infect Dis.
20. Moroney JF, Fiore AE, Harrison LH, et al. Clinical outcomes of bacteremic pneumococcal pneumonia in the era of antibiotic resistance.
Clin Infect Dis.
21. Turett GS, Blum S, Fazal BA, et al. Penicillin resistance and other predictors of mortality in pneumococcal bacteremia in a population with high human immunodeficiency virus seroprevalence.
Clin Infect Dis.
22. Feikin DR, Schuchat A, Kolczak M, et al. Mortality from invasive pneumococcal pneumonia in the era of antibiotic resistance, 1995-1997.
Am J Public Health.
23. Murdoch DR, Laing RT, Mills GD, et al. Evaluation of a rapid immunochromatographic test for detection of
antigen in urine samples from adults with community-acquired pneumonia.
J Clin Microbiol.
24. Farina C, Arosio M, Vailati F, et al. Urinary detection of
antigen for diagnosis of pneumonia.
25. Gutierrez F, Masia M, Rodriguez JC, et al. Evaluation of the immunochromatographic Binax NOW assay for detection of
urinary antigen in a prospective study of community-acquired pneumonia in Spain.
Clin Infect Dis.
26. Neill AM, Martin IR, Weir R, et al. Community acquired pneumonia: aetiology and usefulness of severity criteria on admission.
27. Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia.
N Engl J Med.
28. Marrie TJ, Wu L. Factors influencing in-hospital mortality in community-acquired pneumonia: a prospective study of patients not initially admitted to the ICU.
29. Gordon GS, Throop D, Berberian L, et al. Validation of the therapeutic recommendations of the American Thoracic Society (ATS) guidelines for community acquired pneumonia in hospitalized patients.