Community-Acquired Pneumonia (CAP)

Updated: Jun 16, 2017
  • Author: Stephanie L Baer, MD; Chief Editor: Michael Stuart Bronze, MD  more...
  • Print

Practice Essentials

Community-acquired pneumonia (CAP) is one of the most common infectious diseases and is an important cause of mortality and morbidity worldwide. Typical bacterial pathogens that cause CAP include Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis (see images below). However, with the advent of novel diagnostic technologies, viral respiratory tract infections are being identified as common etiologies of CAP. The most common viral pathogens recovered from hospitalized patients admitted with CAP include human rhinovirus and influenza. [1]

Gram stain showing Streptococcus pneumoniae. Gram stain showing Streptococcus pneumoniae.
Gram stain showing Haemophilus influenzae. Gram stain showing Haemophilus influenzae.

Presentation and pathogens in typical community-acquired pneumonia

The term “typical” CAP refers to a bacterial pneumonia caused by pathogens such as S pneumoniae, H influenzae, and M catarrhalis. Patients with typical CAP classically present with fever, a productive cough with purulent sputum, dyspnea, and pleuritic chest pain. Characteristic pulmonary findings on physical examination include the following:

  • Tachypnea
  • Rales heard over the involved lobe or segment
  • Increased tactile fremitus, bronchial breath sounds, and egophony may be present if consolidation has occurred.
  • Decreased tactile fremitus and dullness on chest percussion may result from parapneumonic effusion or empyema.

Epidemiologic data may provide clues to the specific pathogen causing CAP, as follows:

  • The most common overall pathogen is S pneumoniae
  • Underlying chronic obstructive pulmonary disease (COPD) [2] : H influenzae or M catarrhalis
  • Recent influenza infection [3] : Staphylococcus aureus
  • Alcoholic patient presenting with “currant jelly” sputum : Klebsiella pneumoniae [2]

Atypical community-acquired pneumonia

The clinical presentation of so-called “atypical” CAP is often subacute and frequently indolent. In addition, patients with atypical CAP may present with more subtle pulmonary findings, nonlobar infiltrates on radiography, and various extrapulmonary manifestations (eg, diarrhea, otalgia). Atypical CAP pathogens include the following:

  • Mycoplasma pneumoniae
  • Chlamydophila ( Chlamydia) pneumoniae
  • Legionnaires disease ( Legionella pneumophila)
  • Respiratory viruses, including the following:
  • Influenza A and B
  • Rhinovirus
  • Respiratory syncytial virus
  • Human metapneumovirus
  • Adenovirus 4 and 7
  • Parainfluenza virus
  • Other rare viral pneumonias include the following:
  • Coxsackievirus
  • Echovirus
  • Coronavirus (MERS-CoV, SARS)
  • Hantavirus
  • Epstein-Barr virus
  • Cytomegalovirus
  • Herpes simplex virus
  • Human herpesvirus 6
  • Varicella-zoster virus
  • Psittacosis ( Chlamydophila psittaci)
  • Q fever ( Coxiella burnetii)
  • Tularemia ( Francisella tularensis)Endemic fungi (causing subacute or chronic pneumonia), as follows:
  • Histoplasma capsulatum
  • Cryptococcus neoformans neoformans and neoformans gattii
  • Coccidioides immitis
  • Mycobacteria: Mycobacteria tuberculosis and nontuberculous mycobacteria (uncommon)

Extrapulmonary signs and symptoms seen in some forms of atypical CAP may include the following:

  • Mental confusion
  • Prominent headaches
  • Myalgias
  • Ear pain
  • Abdominal pain
  • Diarrhea
  • Rash (Horder spots in psittacosis; erythema multiforme in Mycoplasma pneumonia)
  • Nonexudative pharyngitis
  • Hemoptysis
  • Splenomegaly
  • Relative bradycardia

While historical clues and physical examination findings may suggest a causative pathogen, the clinical signs and symptoms of CAP are not sufficiently specific to reliably differentiate the exact etiologic agent. [4] Therefore, additional testing remains necessary to identify the pathogen and to optimize therapy in CAP.


Standard diagnostic studies for CAP include the following:

  • Chest radiography
  • Sputum Gram stain and/or culture
  • Blood cultures

Other laboratory tests

Depending on the perceived severity of illness and suspected etiology, additional workup may be warranted, including the following:

  • Complete blood cell (CBC) count with differential
  • Serum sodium level
  • Serum blood urea nitrogen (BUN) and creatinine levels
  • Serum transaminase levels
  • Serum phosphorus level
  • Lactic acid level
  • Creatine phosphokinase (CPK)
  • C-reactive protein (CRP)
  • Lactate dehydrogenase (LDH)
  • Procalcitonin
  • Cold agglutinin titers
  • Urinary antigen testing for Legionella species and S pneumoniae
  • Serologic studies for M pneumoniae, C pneumoniae, Bordetella pertussis, C burnetii
  • Molecular diagnostics, ie, polymerase chain reaction (PCR) testing

Chest radiography

Obtain chest radiographs in all patients with suspected CAP to evaluate for an infiltrate compatible with the presentation of CAP and to help exclude conditions that may mimic CAP (ie, lung cancer, pulmonary emboli). [5, 6] Patients who present very early with CAP may have negative findings on chest radiography. In these patients, repeat chest radiography within 24 hours may be beneficial. CT scanning may also be necessary in immunocompromised patients who present with symptoms that suggest CAP in whom chest radiography findings are negative. Serial chest radiography can be used to observe the progression of CAP; however, radiographic improvement may lag behind clinical improvement.

Hospital admission

Multiple scoring systems are available to assess the severity of CAP and to assist in deciding whether a patient should be hospitalized or admitted to the intensive care unit (ICU). The CURB-65 (confusion, uremia, respiratory rate, low blood pressure, age >65 years) and the Pneumonia Severity Index (PSI) are currently recommended by the 2007 Infectious Diseases Society of America/American Thoracic Society Consensus Guidelines. [7] Patients with CURB-65 scores of 2 or more or PSI class IV-V may necessitate hospitalization or more intensive in-home services. ICU is recommended for any patient who requires mechanical ventilation or vasopressors. ICU admission should also be considered in patients with 3 or more minor risk factors, including respiratory rate of 30 or more, PaO2/FiO2<250, multilobar infiltrates, confusion, uremia, leukopenia, thrombocytopenia, hypothermia, and hypotension requiring aggressive fluid resuscitation.

Proposed scoring systems may also be helpful in certain populations to predict the severity of CAP. The SMART-COP score emphasizes the ability to predict the need for ventilator or vasopressor support and includes systolic blood pressure, multilobar infiltrates, serum albumin levels, respiratory rate, tachycardia, confusion, oxygenation, and pH level. The A-DROP (age, dehydration, respiratory failure, orientation, systolic blood pressure) is also a severity score. Recently, an expanded CURB-65 has been shown to improve prediction of 30-day mortality. It includes LDH, thrombocytopenia, and serum albumin, along with the traditional CURB-65, and has been shown to have better prediction accuracy. [8]

Antibiotic Therapy

Adequate antimicrobial therapy for CAP includes coverage for S pneumoniae and atypical bacterial pathogens. Outpatient treatment for CAP in patients with no comorbidities and no risk factors for drug-resistant S pneumoniae frequently includes the following: [7]

  •  A macrolide (azithromycin, clarithromycin, or erythromycin)
  •  Doxycycline

Treatment in patients with comorbidities such as chronic heart, lung, liver, or renal disease; diabetes mellitus; alcoholism; malignancy; asplenia; immunosuppression; prior antibiotics within 90 days; or other risk factors for drug-resistant infection includes the following:

  • Respiratory fluoroquinolones (moxifloxacin, levofloxacin)
  • Beta-lactam (high-dose amoxicillin 1 g 3 times/day) or amoxicillin/clavulanate (2 g twice daily), or ceftriaxone, cefpodoxime, or cefuroxime (500 mg twice daily) plus a macrolide or doxycycline

In regions with high rates of macrolide-resistant S pneumoniae, consider a nonmacrolide alternative.

For hospitalized patients, therapy consists of the following:

  • Beta-lactams (ceftriaxone or cefotaxime) plus a macrolide or
  • Respiratory fluoroquinolone

Recent studies have suggested that a beta-lactam alone may be noninferior to a beta-lactam/macrolide combination or fluoroquinolone therapy in hospitalized patients. [9]

Therapy in ICU patients includes the following:

  • Beta-lactam (ceftriaxone, cefotaxime, or ampicillin/sulbactam) plus either a macrolide or respiratory fluoroquinolone
  • For patients with penicillin allergy, a respiratory fluoroquinolone and aztreonam

 If Pseudomonas is suspected, therapy is as follows:

  • Anti-pneumococcal and anti-pseudomonal beta-lactam (piperacillin/tazobactam, cefepime, carbapenem [imipenem, meropenem, or doripenem]) plus ciprofloxacin or levofloxacin or
  • Beta-lactam (as above) plus aminoglycoside and azithromycin or aminoglycoside and fluoroquinolone
  • For patients with penicillin allergy, aztreonam instead of the beta-lactam in the regimen listed above

If methicillin-resistant S aureus (MRSA) is suspected, vancomycin, linezolid, or ceftaroline should be added.

Rapid initiation of therapy is important for improved outcomes in CAP, although blanket measures to hasten treatment are not without potential negative consequences. Quality-improvement efforts aimed at the administration of antibiotics within a certain time period have contributed to increased inappropriate antibiotic use and increased incidence of Clostridium difficile colitis. Nevertheless, in patients with signs of severe CAP or sepsis, antibiotics should be given within the first hour of hypotension onset to reduce mortality. [10] Cultures of respiratory specimens, blood, and pleural fluid; PCR of respiratory samples; or antigen tests should be used to target therapy whenever possible. Inpatient CAP therapy usually consists of intravenous antibiotics followed by transition to an oral course of therapy. [11, 12, 13, 14] Patients who are severely ill or who are unable to tolerate or absorb oral medications may require a longer duration of parenteral therapy before switching to an oral antibiotic. [15]

Mild to moderately ill patients with CAP may be treated entirely via the oral route, on either an inpatient or outpatient basis. The duration of therapy for uncomplicated CAP is usually 5-7 days. [7, 10] Patients should be afebrile for 48-72 hours and have no signs of instability before antibiotic therapy is stopped. The duration of therapy may need to be increased if the initial empiric therapy has no activity against the specific pathogen.

Immunocompromised hosts who present with CAP are treated in the same manner as otherwise healthy hosts but may require a longer duration of therapy. Investigations into pathogens associated with compromised hosts may need to be pursued.



Community-acquired pneumonia (CAP) is one of the most common infectious diseases addressed by clinicians and is an important cause of mortality and morbidity worldwide.

Numerous pathogens can cause CAP. Typical bacterial pathogens that cause CAP include S pneumoniae, H influenzae, and M catarrhalis. Numerous other organisms can cause CAP in the appropriate clinical setting. Furthermore, the so-called “atypical CAP” pathogens are actually common causes of CAP and were originally classified as atypical because they are not readily detectable on Gram stain or cultivatable on standard bacteriologic media.

CAP is usually acquired via inhalation or aspiration of a pathogenic organism.

Aspiration pneumonia is commonly caused by various pathogens (eg, aerobic/anaerobic oral organisms).

Severe community-acquired pneumonia

Severe CAP frequently develops in individuals with comorbid factors such as underlying cardiopulmonary disease, diminished splenic function, and/or heightened pathogenic virulence. Even in young and/or healthy hosts, severe CAP can develop if the causative pathogen is sufficiently virulent. For example, influenza, severe acute respiratory syndrome (SARS), Hantavirus pulmonary syndrome (HPS), and Legionnaires disease may present as severe CAP. [16, 17, 18, 19]

Patients with severe CAP should have the benefit of an infectious disease specialist to assist in the underlying cause of their condition.

Complications associated with community-acquired pneumonia

Complications in CAP depend on the infecting pathogen and the patient’s baseline health. For example, various organisms can cause empyema, including S pneumoniae, K pneumoniae (classically occurring in patients with chronic alcoholism), group A Streptococcus, and S aureus. Cavitation is not a typical feature of pneumococcal pneumonia but is relatively common in K pneumoniae infections.

On occasion, myocardial infarction can be precipitated by community-acquired pneumonia (CAP).

Patients with CAP who have impaired splenic function may develop overwhelming pneumococcal sepsis, potentially leading to death within 12-24 hours, regardless of the antimicrobial regimen used.

Morbidity and mortality

Morbidity and mortality associated with CAP are most common in elderly patients and immunocompromised hosts.

Other factors that can predispose to morbidity and mortality in individuals with CAP include significant comorbidities such as an increased respiratory rate, hypotension, fever, multilobar involvement, anemia, and hypoxia. [20]

For more information, see the following:


Etiology of Community-Acquired Pneumonia

The definitive microbiologic etiology is determined in only 38%-63% of patients who develop community-acquired pneumonia (CAP), depending on the patient population and diagnostic testing used. [1, 21] Organisms have been traditionally classified as “typical” or “atypical” CAP pathogens depending on their ability to be detected on Gram stain or standard bacterial cultures.

Typical community-acquired pneumonia pathogens

Typical bacterial pathogens that cause CAP include S pneumoniae, H influenza, and M catarrhalis (Gram stains shown below). The frequency with which CAP is attributable to one of these pathogens varies according to epidemiologic factors (eg, seasonality, patient demographics, exposure history) and the diagnostic testing used. In the past, these organisms had been reported to account for most CAP cases. [22] However, with improvement in diagnostic techniques allowing for better identification of viruses and fastidious bacteria, our understanding of the etiologic agents involved in the development of CAP has evolved. As a result, a smaller percentage of CAP cases are now being attributed to these typical bacterial pathogens.

Gram stain showing Streptococcus pneumoniae. Gram stain showing Streptococcus pneumoniae.
Gram stain showing Haemophilus influenzae. Gram stain showing Haemophilus influenzae.

S pneumoniae remains the most common bacterial agent responsible for CAP. The incidence of S pneumoniae pneumonia varies according to the population studied. A 2015 study of 267 patients with CAP in Norway reported that S pneumoniae accounted for 30% of cases; this represented 48.5% of the cases in which an organism was identified. [21] A separate study of adults with CAP in the United States identified S pneumoniae as the etiologic agent in only 5% of total cases of CAP (13.5% of cases with an identified pathogen). [1]

S aureus has not traditionally been considered a typical cause of CAP in otherwise healthy hosts. However, S aureus is well known to cause potentially severe CAP after influenza infection. [3] In addition, community-acquired methicillin-resistant S aureus (MRSA) has increasingly been associated with multilobar necrotizing CAP, including in previously healthy individuals.

Importantly, K pneumoniae and Pseudomonas aeruginosa are not typical causes of CAP in otherwise healthy hosts. K pneumoniae CAP occurs primarily in individuals with chronic alcoholism or diabetes mellitus. P aeruginosa is a cause of CAP in patients with bronchiectasis or cystic fibrosis.

In certain patients admitted to the ICU, the microbial etiology of pneumonia may be complex. In a study by Cilloniz et al, 11% of cases were polymicrobial. The most frequently identified pathogens in polymicrobial infections were S pneumoniae, respiratory viruses, and P aeruginosa. Chronic respiratory disease and acute respiratory distress syndrome (ARDS) criteria were independent predictors of a polymicrobial infection. [23]

Other gram-negative pathogens (eg, Enterobacter species, Serratia species, Stenotrophomonas maltophilia, Burkholderia cepacia) rarely cause CAP in patients without underlying lung disease or immunosuppression.

Atypical community-acquired pneumonia pathogens

Atypical bacterial pneumonias can be differentiated into those caused by zoonotic or nonzoonotic atypical pathogens.

Zoonotic atypical CAP pathogens include Chlamydophila (Chlamydia) psittaci (psittacosis), F tularensis (tularemia), and C burnetii (Q fever).

Nonzoonotic atypical CAP pathogens include Legionella species (Legionnaires disease), M pneumoniae, and C pneumoniae. [24]

Respiratory viruses are another important cause of atypical CAP. While certain viruses may be zoonotically transmitted (eg, Hantavirus and avian influenza), most are transmitted person-to-person.


Epidemiology of Community-Acquired Pneumonia

Prevalence in the United States

The number of annual community-acquired pneumonia (CAP) cases is difficult to estimate. One study, in which 46,237 elderly patients were monitored over a 3-year period, showed the rate of CAP among those aged 65-69 years to be 18.2 cases per 1000 person-years. Among person older than 85 years, the rate was 52.3 cases per 1000 person-years. Estimates based on these data suggested that, annually, 1 of 20 persons older than 85 years develop CAP. The investigators also estimated that approximately 915,900 cases of CAP occur among elderly persons annually in the United States. [25, 26, 27]

In a more recent population-based surveillance study of CAP among adults presenting at one of 5 hospitals in Chicago and Nashville, the annual incidence of CAP was estimated to be 24.8 cases (95% CI; 23.5-26.1) per 10,000 adults. [1] The median age of patients presenting with CAP was 57 years (interquartile range, 46-71 years). Similar to the prior studies, the incidence of CAP increased with age.

The prevalence and cost of CAP in 2011 in the United States Department of Veterans Affairs Hospitals was examined, including 35,380 episodes of CAP among 7,824,850 veterans, or 452 cases per 100,000 person-years. The estimated excess cost was $750,170,631, with over half incurred in patients older than 65 years. [28]

International prevalence

A retrospective study using a nationwide claims database to determine the incidence of CAP in the Netherlands identified 195,372 cases of CAP between 2008 and 2011. [29] This represented an average incidence of 295 cases per 100,000 population per year. The authors concluded that the mean annual cost of CAP in this population was 178 million euros, with a disproportionate amount (76%) of the cost being incurred by people older than 50 years.


Advanced age is associated not only with a higher incidence of CAP but also with more severe disease, greater need for hospitalization, and higher mortality. [7, 30]

CAP encountered in the ambulatory setting is more common among young adults and is usually due to atypical CAP pathogens (eg, Mycoplasma pneumoniae). [31]


Prognosis of Community-Acquired Pneumonia

Negative prognostic factors in community-acquired pneumonia (CAP) include preexisting lung disease, underlying cardiac disease, poor splenic function, advanced age, multilobar involvement, and delayed initiation of appropriate antimicrobial therapy. [32]


Extrapulmonary Findings in Atypical Community-Acquired Pneumonia

Atypical community-acquired pneumonia (CAP) has classically been associated with more extrapulmonary manifestations than typical bacterial CAP. However, there can be considerable overlap in the clinical presentation of CAP due to various pathogens such that a definitive microbiologic diagnosis cannot be made based on signs and symptoms alone. Certain constellations of findings in the setting of appropriate historical clues may suggest an increased likelihood of a specific pathogen, thus warranting a targeted investigation for that organism. A detailed review of all potential extrapulmonary findings in CAP is beyond the scope of this article; however, some classic associations are included below:

M pneumoniae CAP is associated with the following findings: [33]

  • Headache, fever, malaise, sore throat in young adult with insidious onset of cough
  • Erythema multiforme major (Stevens-Johnson syndrome)
  • Cardiac conduction abnormalities
  • Hemolytic anemia and cold-agglutinin syndrome
  • Neurologic abnormalities, including aseptic meningitis or meningoencephalitis, Guillain-Barre syndrome, transverse myelitis
  • Associated with epidemic outbreaks, eg, in schools or military barracks

Legionella pneumophila CAP (Legionnaires disease) is associated with the following findings: [34]

  • Gastrointestinal and neurologic symptoms in the setting of pneumonia
  • Positive history of waterborne exposure
  • Relative bradycardia during febrile episode
  • Hyponatremia, hypophosphatemia, elevated creatine phosphokinase (CPK) level, elevated ferritin level, myoglobinuria
  • Leukocytosis with relative lymphopenia
  • Unresponsive to beta-lactam antibiotics
A case of Legionnaires disease from the Philadelph A case of Legionnaires disease from the Philadelphia outbreak, showing characteristics of relative bradycardia and extrapulmonary involvement.
This graph outlines a case of Legionella pneumonia This graph outlines a case of Legionella pneumonia, showing characteristics of bradycardia and extrapulmonary involvement. Also shown is an initial lack of response to beta-lactam antibiotics, followed by effective treatment with doxycycline.

C burnetii CAP (Q fever) is associated with the following findings: [35, 36]

  • Acute infection
  • Severe retrobulbar headache, myalgias, fever, rigors, nonproductive cough
  • Elevated levels of transaminases and thrombocytopenia
  • Maculopapular or purpuric rash
  • Zoonotic exposure (goats, sheep, cattle most common)

Community-acquired pneumonia and shock

With the exception of CAP due to particularly virulent organisms (eg, MRSA, Hantavirus, severe acute respiratory syndrome [SARS], Legionella), CAP does not typically present with shock in otherwise healthy hosts. Therefore, in addition to considering the possibility of CAP due to a hypervirulent pathogen, patients who present with fever, dyspnea, leukocytosis, pulmonary infiltrates, and shock in the absence of conditions associated with hyposplenism should be evaluated for imitators of pneumonia, such as acute myocardial infarction or acute pulmonary embolism.

Conditions that predispose to severe CAP should be considered in patients presenting with CAP and shock in the absence of one of the aforementioned cardiopulmonary diseases. The following disorders and therapies have been associated with severe CAP:


Patient History

Patients with community-acquired pneumonia (CAP) due to typical bacterial CAP pathogens typically present with fever, dyspnea, and productive cough, often with pleuritic chest pain.

The clinical presentation of atypical CAP is more often subacute and associated with extrapulmonary manifestations that may provide a clue to the etiology.

Zoonotic infection

Contact with the appropriate zoonotic vector or its by-product (eg, milk, urine, feces, placenta) is needed to develop a zoonotic CAP. A history of occupational exposure to livestock (eg, farmers, veterinarians) or close contact with a parturient animal should be sought in patients with suspected Q fever. Psittacosis is preceded by recent contact with birds infected with C psittaci. Occupations and avocations associated with increased risk include poultry farming, pet shops, veterinary clinics, and ownership of pet birds (classically of the psittacine, or parrot, family). Hantavirus is transmitted via exposure to wild rodents, specifically to aerosolized rodent urine or feces; thus, queries as to whether a patient presenting with severe CAP works or recreates in a setting conducive to rodent exposure (eg, farms, ranches, forests) is warranted.


Physical Examination

Purulent sputum is characteristic of pneumonia caused by bacterial community-acquired pneumonia (CAP) pathogens and is not usually a feature of pneumonia caused by atypical pathogens, with the exception of Legionnaires disease. Blood-tinged sputum may be found in patients with pneumococcal pneumonia, Klebsiella pneumonia, or Legionella pneumonia.

Rales are heard over the involved lobe or segment. Consolidation may be accompanied by an increase in tactile fremitus, bronchial breathing, and egophony.

Legionella pneumonia, Q fever, and psittacosis are atypical pneumonias that may present with signs of consolidation. Consolidation is not a typical feature of pneumonia caused by M pneumoniae or C pneumoniae. [24]

Be wary when a patient presents with severe CAP, with or without hypotension or shock. In these patients, be sure to exclude an underlying immunocompromise, asplenia, or acute pulmonary or cardiac event that could explain the severity of the CAP.

Pleural effusion

Pleural effusion, if large enough, may be detectable on physical examination. In addition, patients with pleural effusion also have decreased tactile fremitus and dullness on chest percussion.


On physical examination, empyema has the same findings as pleural effusion.


Differential Diagnoses of Community-Acquired Pneumonia

Aside from those mentioned above, the differential diagnoses to consider in the diagnosis of community-acquired pneumonia (CAP) include the following:

  • Acute bronchitis
  • Acute exacerbation of chronic bronchitis
  • Aspiration pneumonitis
  • Myocardial infarction
  • Congestive heart failure and pulmonary edema
  • Pulmonary fibrosis
  • Sarcoidosis
  • SLE pneumonitis
  • Pulmonary drug hypersensitivity reactions (nitrofurantoin)
  • Drug-induced pulmonary disease
  • Cryptogenic organizing pneumonia
  • Pulmonary embolus or infarction
  • Bronchogenic carcinomas
  • Radiation pneumonitis
  • Granulomatosis with polyangiitis (Wegener granulomatosis)
  • Lymphomas
  • Tracheobronchitis

Sputum Studies and Blood Culture

Send sputum samples from patients with community-acquired pneumonia (CAP) for Gram stain and/or culture. Keep in mind that many patients, especially elderly persons, are not able to produce an adequate suitable sputum sample.

Sputum Gram stain is reliable and diagnostic if performed on a well-collected specimen without many squamous epithelial cells (saliva contamination) and if a predominant organism is present. Gram stain shows few or no predominant organisms in patients with atypical CAP.

Obtain blood cultures from all patients with CAP upon admission because some bacterial pathogens, such as S pneumoniae and H influenzae, are frequently associated with positive blood cultures. M catarrhalis bacteremia is rare.


Studies in HIV-Positive Patients with Community-Acquired Pneumonia

The differential diagnoses of community-acquired pneumonia (CAP) in patients with human immunodeficiency virus (HIV) infection is broader than in HIV-negative patients. The patient’s immunologic status, as reflected by the CD4 count, the clinical course, and the chest radiographic appearance, provides clues to the most likely etiologic organism.

Patients with HIV infection and a normal or slightly decreased CD4 count with focal infiltrates have approximately the same pathogen distribution as otherwise healthy hosts (eg, S pneumoniae is most common) and thus warrant the same diagnostic strategies as the general population. Pneumocystis (carinii) jiroveci pneumonia (PJP) should be suspected in patients with a CD4 count of less than 200 cells/µL presenting with gradually progressive dyspnea, nonfocal infiltrates on chest radiography, nonproductive cough, and hypoxemia. A definitive diagnosis of PJP requires visualization of the P jirovecii cysts (ie, using special stains such as Giemsa or methenamine silver); bronchoscopy with bronchoalveolar lavage (BAL) is often necessary to obtain an adequate specimen.

Individuals with HIV infection are also at increased risk for active pulmonary tuberculosis; the radiographic appearance of tuberculosis in HIV-infected patients varies, so tuberculosis should be considered in all patients with HIV and a pulmonary infiltrate. Patients with HIV and CAP should therefore be placed on airborne isolation (unless other etiology identified) until tuberculosis is excluded with three negative findings on acid-fast bacillus (AFB) smears/sputum cultures.

Additional uncommon CAP pathogens, such as histoplasmosis, coccidioidomycosis, and cryptococcosis, should also be considered in HIV-positive patients who present with pulmonary infiltrates, with risk stratified according to potential exposure history. Antigen testing (urine antigen for histoplasmosis; serum antigen for cryptococcosis) may be helpful in these cases.


Other Laboratory Tests for Community-Acquired Pneumonia

Several nonspecific laboratory tests are often performed during the workup of community-acquired pneumonia (CAP), particularly if atypical CAP is suspected.

Serum transaminase, serum sodium, serum ferritin, serum phosphorus, and creatine phosphokinase (CPK) levels may provide evidence supporting a particular pathogen, such as Legionella. C-reactive protein (CRP) levels and procalcitonin may help predict the likelihood of a bacterial origin for CAP. Lactic acid, white blood cell (WBC) count, blood urea nitrogen, and creatinine may be used in categorizing the severity of illness.

Cold agglutinin titers of 1:32 or greater may support a diagnosis of M pneumoniae, however this test is neither sensitive nor specific.

Procalcitonin and C-reactive protein levels

A prospective study of 75 children with CAP, including 37 who met the criteria for presumed pneumococcal CAP (P-CAP), found procalcitonin (PCT) and CRP levels to be reliable markers of P-CAP. [37] High levels of these markers were strongly associated with P-CAP. Whereas a PCT value of 0.5 ng/mL or lower ruled out P-CAP in more than 90% of cases, a value of 1.5 ng/mL or higher in association with a positive pneumococcal urinary antigen test result, yielded a diagnostic probability of almost 80% for P-CAP. [38]

Agglutinin levels

Cold-agglutinin antibodies (immunoglobulin M [IgM] autoantibodies directed against the erythrocyte) develop in 50%-75% of patients with M pneumoniae infection, peaking approximately 2 weeks into the illness. However, this finding is nonspecific, as low-titer cold-agglutinin elevations occur in various viral and neoplastic illnesses. Furthermore, a negative cold agglutinin titer finding does not exclude Mycoplasma infection.

Direct fluorescent antibody testing

Direct fluorescent antibody (DFA) testing of the sputum can be performed to try to assist with the diagnosis of atypical CAP pathogens, including Legionella, P jiroveci, and Chlamydophila; however, the utility of this test is hampered by suboptimal sensitivity and is relatively difficult to perform. A DFA stain showing Legionella infection is seen in the image below.

Sputum direct fluorescent antibody stain showing L Sputum direct fluorescent antibody stain showing Legionella infection.


Serology can be useful in identifying fastidious organisms, particularly when dealing with pathogens with potential epidemiologic implications (ie, epidemic or biohazard situations), such as B pertussis, M pneumoniae, L pneumophila, or C burnetii. Serologic diagnosis is based on a 4-fold or greater increase in titers between acute- and convalescent-phase serum specimens so is generally not useful in the acute-care setting. A significantly elevated IgM titer (typically present in the acute phase of infection) may help support a diagnosis, and some experts have suggested that combining IgM and nucleic acid amplification testing may be the optimal method for diagnosing M pneumoniae infection. [39]

Urinary antigen test

Pneumococcal urinary antigen testing is a useful non–culture-based test for diagnosing pneumococcal infection that has a reported sensitivity of 50%-80% and specificity of more than 90%. [40] Availability of the assay is the main perceived barrier to its routine use. [40]

Urinary antigen testing is considered the first-line diagnostic test for L pneumophila. The sensitivity of the test ranges from 55%-99%, with improved sensitivity paralleling disease severity. [34] The urinary antigen test is applicable only for L pneumophila serogroup type I, which accounts for approximately 80% of infections. The urinary antigen test remains positive for Legionella for long periods but may be negative within the first 48 hours of infection.

Polymerase chain reaction

Polymerase chain reaction (PCR) has emerged as an important diagnostic tool for determining the etiology of CAP, particularly with regard to respiratory viruses and fastidious organisms, including Legionella, Mycoplasma, and Chlamydophila. [2] PCR is a very sensitive and specific method for isolating these pathogens. The increasing commercial availability of various PCR assays (including multiplex) will allow for increased implementation in the clinical setting. However, interpreting the isolation of organisms known to colonize the upper airway or those that may be associated with protracted shedding, such as rhinovirus, remains a challenge. Quantitative PCR methods have shown some promise in improving interpretability of such findings. [41]

A 2016 study in the United Kingdom assessed the use of comprehensive molecular testing of a single lower respiratory tract specimen to detect a pathogen in CAP. [42] The study compared (1) a combination of real-time multiplex PCR assays targeting 26 different pathogens, including bacteria and viruses, with (2) culture-based diagnostic testing of the same specimens. Quantification of bacterial load was performed for 8 common bacteria, including S pneumoniae, H influenzae, and M catarrhalis. A pathogen was detected in 87% of patients with CAP using the multiplex assay, whereas culture achieved a diagnosis in only 39% of cases. Viruses were detected in 30% of cases. H influenzae and S pneumoniae were the most commonly detected bacteria. Pathogen detection was not significantly decreased when a quantitative threshold for detection of ≥105 CFU/mL for all bacterial loads was applied, although the mean bacterial load was lower in patients who had received prior antibiotics than in those who had not. Interestingly, of the 268 patients who received antibiotics prior to testing, 77.6% had a positive bacterial PCR finding, but only 32% were culture-positive. The authors concluded that comprehensive molecular testing on a lower respiratory tract specimen has the potential to positively affect targeted antibiotic therapy in most patients with CAP.


Chest Radiography and CT Scanning

Chest radiography

Obtain chest radiographs in all patients with suspected community-acquired pneumonia (CAP) to exclude conditions that mimic CAP and to confirm the presence of an infiltrate compatible with the presentation of CAP. [5, 6] (See the images below.)

Chest radiograph in a patient with HIV infection, Chest radiograph in a patient with HIV infection, bilateral perihilar infiltrates, and Pneumocystis jiroveci pneumonia.
Chest radiograph in a patient with HIV infection a Chest radiograph in a patient with HIV infection and focal infiltrates due to tuberculosis.

Chest radiography findings may be negative in patients who present very early with CAP. In these patients, repeat chest radiography within 24 hours is recommended.

Chest radiography may assist with the differentiation of viral pneumonias from nonviral pneumonias. Viral pneumonias tend to display few or no infiltrates on chest radiography, but, when infiltrates are present, they are almost always bilateral, perihilar, symmetric, and interstitial.

Bacterial pneumonias have a predominantly focal segmental or lobar distribution, with or without pleural effusions. Atypical bacterial pathogens have variable radiographic findings, ranging from focal segmental to bilateral interstitial disease. P jiroveci (PJP) pneumonia typically manifests as bilateral patchy interstitial infiltrates. Of note, radiographic findings alone are not reliable for differentiating specific etiologies of CAP.

Chest radiographic findings should be negative in patients with asthma or exacerbation of chronic bronchitis who do not have CAP.

The infiltrates observed with congestive heart failure (CHF) appear as increased interstitial markings and show vascular redistribution to the upper lobes. Patients with preexisting heart failure usually have cardiomegaly.

Rapid cavitation is not a typical feature of CAP.

Community-acquired methicillin-resistant S aureus (CA-MRSA) CAP presents as a fulminant CAP with rapid cavitation and necrotizing pneumonia caused by CA-MRSA (SCC mec IV) with the PVL gene, which sometimes occurs after influenza.

Serial chest radiography can be used to observe the progression of CAP. Chest radiographic findings worsen rapidly and require a significant period to improve. It is important to note that clinical resolution occurs long before radiologic resolution.

CT scanning

Obtain a computed tomography (CT) scan of the chest when an underlying bronchogenic carcinoma is suggested or if any abnormalities are not consistent with the diagnosis of pneumonia. Immunocompromised patients may benefit from further narrowing of differential diagnoses with chest CT scanning.


Fine-Needle Aspiration, Transtracheal Aspiration, and Bronchoscopy With Bronchoalveolar Lavage

Diagnostic bronchoscopy with bronchoalveolar lavage (BAL) may be useful in patients with community-acquired pneumonia (CAP) when Pneumocystis, mycobacteria, or fungal pathogens are likely. Diagnostic bronchoscopy with BAL with or without biopsy may also be useful in unresolving CAP that does not respond to appropriate therapy.

Transthoracic fine-needle aspiration (FNA) of the infiltrate can be performed and is most useful in determining the cause of nodules or non–infection-associated infiltrates that are not responding to antibiotic treatment.

Transtracheal aspiration (TTA) is a potentially hazardous procedure and offers no additional diagnostic information in patients with CAP.


Histologic Findings

Lung sections with typical bacterial pneumonias show the progression from red hepatization to white hepatization during the resolution process. The lung is repaired or scarred after bacterial pneumonia is complete and the infectious process resolves.


Hospital Care in Community-Acquired Pneumonia

Although patients with mild community-acquired pneumonia (CAP) may be treated in an ambulatory setting, patients with CAP who are moderately to severely ill should be hospitalized. Patients with severe CAP require admission to an intensive care unit (ICU). Oxygen and/or ventilatory support may be required. [18, 43, 44, 45, 46, 47]

Because the severity of CAP is frequently due to underlying chronic disease, provide evidence-based sepsis management while administering antibiotics for CAP.

Patients admitted with severe CAP and hypotension or shock are often hypotensive because of sepsis.

If no acute cardiopulmonary explanation can be found (eg, exacerbation of severe underlying lung disease, exacerbation of preexisting CHF), patients with shock likely have diminished or absent splenic function or immunocompromise.


Pharmacologic Therapy

There is no optimal therapy for community-acquired pneumonia (CAP). Most experts feel that coverage should be divided against typical and atypical CAP pathogens. [48] Excellent practice guidelines have been promulgated by the Infectious Diseases Society of America (IDSA) and the American Thoracic Society (ATS). These resources provide evidence-based guidelines for the treatment of outpatients, inpatients, and ICU patients with CAP. [7]

Adequate therapy for CAP includes coverage for S pneumoniae and atypical bacterial pathogens. Treatment options for CAP in outpatients with no comorbidities and no risk factors for drug-resistant S pneumoniae include the following: [7]

  • A macrolide (azithromycin, clarithromycin or erythromycin)
  • Doxycycline

Treatment options in patients with comorbidities such as chronic heart, lung, liver or renal disease; diabetes mellitus; alcoholism; malignancy; asplenia; immunosuppression; prior antibiotics within 90 days; or other risk for drug resistant infection include the following:

  • Respiratory fluoroquinolones (moxifloxacin, levofloxacin)
  • Beta-lactam (high-dose amoxicillin 1 g three times/day) or amoxicillin/clavulanate (2 g twice a day), or ceftriaxone, cefpodoxime, or cefuroxime (500 mg twice daily) plus a macrolide or doxycycline

In regions with high rates of macrolide-resistant S pneumoniae, consider a nonmacrolide alternative.

In hospitalized patients, therapy consists of the following:

  • Beta-lactam (ceftriaxone) plus a macrolide or
  • Respiratory fluoroquinolone

Recent studies have suggested that the use of a beta-lactam alone may be noninferior to a beta-lactam/macrolide combination or fluoroquinolone therapy in hospitalized patients. [9]

Therapy in ICU patients includes the following:

  • Beta-lactam (ceftriaxone, cefotaxime, or ampicillin/sulbactam) plus either a macrolide or respiratory fluoroquinolone
  • For patients with penicillin allergy, a respiratory fluoroquinolone and aztreonam

Patients who are severely ill or unable to tolerate or absorb oral medications require a longer duration of intravenous therapy before switching to an oral antibiotic. [15] If Pseudomonas is suspected, an antipseudomonal beta-lactam (piperacillin/tazobactam, cefepime, imipenem, or meropenem) plus ciprofloxacin or levofloxacin is recommended. If MRSA is suspected, vancomycin or linezolid is recommended. For patients with penicillin allergy, aztreonam is used instead of the beta-lactam in the regimen listed above.

Mild to moderately ill patients with CAP may be treated entirely via the oral route, on either an inpatient or outpatient basis. Patients receiving oral antibiotics may be admitted for hospital services (eg, pulmonary toilet and additional diagnostic tests) that are not obtainable on an outpatient basis.

If the patient is switched to an oral regimen and is doing well, earlier discharge from the hospital is possible. The oral therapy regimen can be completed at home. Optimal intravenous-to-oral switch therapy consists of a single agent that has an appropriate spectrum, has excellent bioavailability, is well tolerated, has a low resistance potential, and is relatively inexpensive.

The duration of therapy for uncomplicated CAP is approximately 5-7 days. [7, 10] Patients should be afebrile 48-72 hours and have no signs of instability before antibiotic therapy is stopped. The duration of therapy may need to be longer if initial empiric therapy did not have activity against the detected pathogen. Very healthy young adults and children may be treated for shorter periods.

The use of systemic corticosteroids in patients with CAP may reduce the length of time until clinical stability, reduce hospital length of stay, reduce the need for mechanical ventilation, and reduce the incidence of adult respiratory distress syndrome (ARDS). [49, 50] Recent clinical trials have also shown a possible overall reduction in mortality, although these later results remain in doubt. [51, 52] Furthermore, there is insufficient data or agreement on the dose or duration of the steroid therapy. Adverse effects noted in the studies have included hyperglycemia. [49] There are current ongoing double-blind, randomized, controlled clinical trials to examine the short- and long-term effects of corticosteroid use in the management of CAP. [50]

Comorbid conditions

Comorbid conditions do not affect the selection of antimicrobial therapy. The addition and/or change of antibiotics based on the severity of illness and/or comorbidities makes little sense. Antimicrobial therapy is directed against the pathogen rather than against the comorbid factors. Comorbidities are an important prognostic factor and contribute to the severity index. [32]


The severity of CAP may be estimated with scoring systems that consider age, end organ damage manifested as uremia and/or confusion, respiratory rate, and blood pressure (CURB-65).

Other scoring systems include the Pneumonia Severity Index (PSI), SMART-COP, A-DROP, and expanded CURB-65 (see Hospital admission). Prognosis is significantly affected by underlying conditions of the lungs, heart, and spleen.

Appropriate spectrum

In otherwise healthy hosts with CAP, therapy does not need to cover S aureus, Klebsiella species, or P aeruginosa. S aureus coverage should be included in patients with influenza who have focal infiltrates.

Most antibiotics used to treat community-acquired aspiration pneumonia (eg, beta-lactam/beta-lactamase inhibitor) are highly effective against oral anaerobes. Metronidazole and clindamycin are unnecessary unless anaerobic lung infection is suspected. For aerobic lung abscesses, clindamycin or moxifloxacin is preferable. [53, 54, 55, 56] Coverage should include typical (S pneumoniae, H influenzae, M catarrhalis) and atypical (Legionella and Mycoplasma species, C pneumoniae) pathogens.


Monotherapy coverage of typical and atypical pathogens in CAP is preferred over double-drug therapy; monotherapy is less expensive than double-drug regimens, while being as effective. [57]

Preferred monotherapy for CAP includes doxycycline or a respiratory quinolone. This is the most cost-effective way to optimally treat CAP. It is well-tolerated in oral and intravenous forms. It is ideal for intravenous-to-oral switch monotherapy in terms of patient compliance, safety, and cost. [58, 59]

Penicillin resistance

Most penicillin-resistant S pneumoniae infections may also be treated with beta-lactams. Alternately, doxycycline or respiratory quinolones may be used. Vancomycin is rarely, if ever, needed.

High-level penicillin-resistant S pneumoniae (MIC 6 µg/mL) strains are a rare cause of CAP, although they remain susceptible to ceftriaxone.

Proton-pump inhibitors

Avoid using proton-pump inhibitors (PPIs) in combination with respiratory quinolones for CAP drug therapy. The PPI should be discontinued or replaced with a histamine-2 (H2) blocker for the duration of therapy. However, there is conflicting evidence as to the safety of using PPIs and H2 blockers. [60, 61, 62, 63]

Eurich et al concluded that acid suppressants substantially increase the risk of recurrent pneumonia in high-risk elderly patients. In a cohort of elderly patients who had previously been hospitalized for pneumonia, the investigators studied PPI and H2 blocker use during 5.4 years of follow-up, matching 248 patients with recurrent pneumonia with 2476 controls. [60] Patients in the study who were currently using PPI/H2 blocker had a higher rate of recurrent pneumonia than did nonusers (12% vs 8%, respectively).

In contrast, a population-based, nested case-control study by Dublin et al concluded that PPIs and H2 blockers do not increase the risk of pneumonia in older adults. [62]

Compared with H2 blockers, PPIs were associated with an increased risk of C difficile colitis. [64]


Outpatient Care in Community-Acquired Pneumonia

Monitor patients with mild community-acquired pneumonia (CAP) who are being treated on an outpatient basis to ensure compliance with their medications and improvement. After 1 week, a repeat visit is advisable. If the patient is improving and parapneumonic complications are not evident, posttherapy chest radiography is unnecessary. [65, 66, 67]

The diet in patients with CAP is as tolerated. Guide activity with common sense.



Pneumococcal vaccines have been shown to have some efficacy in preventing vaccine-strain pneumococcal pneumonia, bacteremia, and invasive disease. They have not been shown to prevent community-acquired pneumonia (CAP) of all kinds. [68, 69] Annual influenza vaccination has been shown to decrease pneumonia diagnoses, hospitalizations, and cardiac events in certain populations. [70, 71, 72, 73]

Annual influenza vaccination is recommended in all persons older than 6 months. The options for vaccination include the trivalent inactivated vaccine, the quadrivalent inactivated vaccine, the live attenuated intranasal quadrivalent vaccine, the trivalent inactivated cell culture vaccine, and the trivalent inactivated recombinant vaccine. The live attenuated nasal spray is approved for patients aged 2-49 years who are not pregnant, children receiving aspirin therapy, immunosuppressed persons, children with asthma in the past 12 months, individuals receiving antiviral medications, and caretakers of immunocompromised persons. Persons aged 18 years or older with egg allergy are eligible for the recombinant vaccine, which should be administered by a physician experienced in the management of allergic conditions. People with severe allergy to the flu vaccine should not receive it, and persons with history of Guillain-Barré syndrome should consult with their physician prior to being vaccinated. [74]

Two pneumococcal vaccines are approved in the United States; 13-valent polysaccharide conjugate vaccine (PCV13; Prevnar 13) is approved for children aged 6 weeks to 17 years and adults aged 50 years or older. The 23-valent pneumococcal polysaccharide vaccine (PPSV23; Pneumovax 23) is approved for adults aged 65 years or older and persons aged 2 years or older who are at increased risk for pneumococcal disease.

On February 1, 2016, the Advisory Committee on Immunization Practices (ACIP) published updated recommendations for pneumococcal vaccination in adults. The committee now recommends routine use of PCV13 in addition to PPSV23 for adults older than 65 years. It is recommended that an initial dose of PCV13 be given, followed by PPSV23 at least one year later. If PPSV23 has already been given, it should be followed by PCV13 a year later. A second dose of PPSV23 is not needed in immunocompetent hosts. If PPSV23 is inadvertently given prior to 1 year after PCV13, it does not need to be repeated.

In adults aged 19 years and older with immunocompromising conditions (eg, HIV, cancer, renal disease), functional or anatomic asplenia, cerebrospinal fluid leaks, or cochlear implants, it is recommended that they receive PCV13, followed by PPSV23 at least 8 weeks later. In patients who have previously received PPSV23 vaccine, administer 1 dose of PCV13 at least 1 year after the last PPSV23 dose. A second dose of PPSV23 is recommended 5 years after the first dose in persons in this category aged 19-64 years. If the first dose of PPSV23 was administered prior to age 65 years, an additional dose should be administered at age 65 years or at least 5 years after the prior dose. Subsequent doses of PPSV23 should not be given sooner than 8 weeks after a dose of PCV13 and 5 years after the most recent dose of PPSV23. [75]

Nonleukopenic compromised hosts, such as those with rheumatoid arthritis, SLE, or alcoholism, may not develop an antibody response to the pneumococcal vaccine and may therefore remain susceptible to pneumococcal pneumonia. The same is true concerning the use of the Haemophilus vaccine.


Patient Instructions

Remind patients with community-acquired pneumonia (CAP) to comply with the medication even after they experience clinical improvement. Except in patients with heart failure, adequate hydration and preservation of the cough reflex during the convalescent period are important.