Get PulmCCM’s Weekly Email Update
Stay up-to-date in pulmonary and critical care. No spam.
Community-acquired pneumonia (CAP) is "ordinary" pneumonia, usually (but not always) caused by one of a short list of pathogens susceptible to common antibiotics. Pneumonia remains one of the main reasons for hospital admissions, and causes an estimated 3.5 million deaths yearly, including more than 50,000 in the U.S. Catching pneumonia also increases the risk for early death down the road, even in previously healthy young people. Richard Wunderink and Grant Waterer's review in the February 6 2014 New England Journal of Medicine brings us up to speed on the latest in diagnosis and treatment of community-acquired pneumonia.
See also: Pneumonia: PulmCCM Review & Update
Diagnosis of Community-Acquired Pneumonia
In patients with no underlying heart or lung disease, the diagnosis of pneumonia is usually straightforward: a new or changed infiltrate on chest radiograph, new or worsened respiratory symptoms (cough, dyspnea, chest pain), and fever, chills, or leukocytosis.
Concomitant cardiopulmonary disease (congestive heart failure, COPD, pulmonary fibrosis, lung cancer, etc.) producing abnormal lung imaging may complicate the diagnosis of pneumonia substantially. Elderly patients' only pneumonia symptom may be confusion. Radiologists may disagree 10% of the time in diagnosing pneumonia from chest films.
Once pneumonia is diagnosed, 3 decisions must quickly be made:
- Which antibiotic(s) to prescribe?
- Where should the patient be treated: at home, on the medical ward, or in the ICU?
- How much testing should you perform, to look for the pneumonia's causes?
Antibiotic Treatment of Community-Acquired Pneumonia
The primary concern when selecting antibiotics for community acquired pneumonia is ensuring coverage of Streptococcus pneumoniae and atypical organisms including Mycoplasma, Chlamydophila, and Legionella. Treatment should always be individualized, but published recommendations by the IDSA and ATS have suggested the following:
|Outpatients||Usual Pathogens||Empiric Antibiotics Recommended|
|<60 yrs old, no comorbidities||Streptococcus pneumoniae
|Macrolide (e.g., azithromycin) or doxycycline|
|>65 yrs old, or comorbidities, or antibiotics within 90 days||Above, or:Gram negative bacilli
|Respiratory fluoroquinolone (e.g., levofloxacin, moxifloxacin), or Beta-lactam (e.g., amoxicillin-clavulanate) + macrolide|
|Inpatients||Usual pathogens||Empiric Antibiotics Recommended|
|Not severely ill (i.e., medical ward)||Above, or polymicrobial infections||Respiratory fluoroquinolone (moxi 400 mg daily or levo 750 mg daily), or IV beta lactam (e.g., cefotaxime, ceftriaxone, ampicillin-sulbactam, ertapenem) + macrolide|
|Severely ill (i.e., in ICU)||Above, or Legionella||IV beta lactam (above) + either azithromycin or respiratory fluoroquinolone; for penicillin allergic: respiratory FQ + aztreonam If Pseudomonas suspected: piperacillin-tazobactam, cefepime, imipenem, or meropenem + ciprofloxacin or levofloxacin; can substitute aminoglycoside + azithromycin for the FQIf MRSA possible / suspected, add vancomycin or linezolid|
Who Should Get Broad-Spectrum Antibiotics for Pneumonia?
Overall, patients with pneumonia and certain risk factors (hospitalized or receiving antibiotics within 90 days, nursing home residence, hemodialysis, home wound care, immunosuppression, a family member with a multidrug resistant organism, poor functional status or other factors) are more likely to have pneumonia due to drug-resistant bacteria. They may fail standard therapy for community acquired pneumonia, and have been termed as having healthcare associated pneumonia (HCAP). Until recently, many experts have advised that most such patients be treated with two anti-Pseudomonas agents and coverage for MRSA.
However, the concept of HCAP, and the implication that all such patients should initially be treated with broad spectrum antibiotics, are considered controversial today, because the scheme leads to overuse of broad agents. At some centers, applying the HCAP criteria results in 50% of patients with community-acquired pneumonia being treated with broad-spectrum antibiotics. Most such patients are culture-negative and there seems to be a relatively low rate of antibiotic-resistant bacteria. There is also some suggestion that broad-spectrum antibiotics are associated with worse outcomes in pneumonia diagnosed as HCAP.
This has created confusion and controversy, since inappropriate early antibiotics in people with true HCAP with MDR pathogens (i.e., treating them with traditional antibiotics for community acquired pneumonia) is also associated with worse outcomes.
These authors argue that with the exception of MRSA, the risk of MDR pathogens increases significantly only when 3 or more of the above risk factors are present, implying that traditional CAP treatment is appropriate for most patients. If previous MRSA infection / colonization, hemodialysis, or heart failure (considered MRSA-specific risk factors) and another pneumonia-specific risk factor are present, the risk justifies MRSA coverage, but not dual coverage for Pseudomonas. Your local prevalence of MDR bacteria (in the hospital and the community if known) should also influence the choice of whether to use broad coverage. Whew. You need an ID consult just to get through this stuff.
Community-acquired pneumonia due to MRSA can occur in healthy people with no healthcare contacts. If MRSA is known or suspected (such as in a severe cavitary pneumonia), linezolid or clindamycin added to vancomycin can be given to suppress the endotoxin production that increases pneumonia severity.
Timing of Antibiotics for Pneumonia
Antibiotics should be given as soon as possible for people with pneumonia. Quality metrics defining a 6-hour window were based on weak evidence and have resulted in excessive antibiotic use in EDs, with adverse effects like C. difficile and no observed improvement in mortality. (Improved overall care may have been responsible for the better observed outcomes, not just time to antibiotic dose.) Patients with pneumonia and hypotension should get antibiotics within one hour of presentation; stronger evidence supports lifesaving benefits from early antibiotic administration in severe sepsis, of which pneumonia is a leading cause.
Diagnosing Pneumonia: How Much Testing Is Needed?
No diagnostic testing beyond chest radiograph is necessary for most people with community acquired pneumonia, especially outpatients or those hospitalized with less severe disease. Among people with more severe pneumonia, there is disagreement among experts as to how much diagnostic testing is indicated:
- Blood culture: These authors recommend blood culture only for patients admitted to the ICU or with healthcare associated pneumonia, or who have cirrhosis or lack a spleen.
- Sputum culture: For intubated patients, or those with productive cough with severe pneumonia or HCAP or severe COPD or structural lung disease (e.g. bronchiectasis).
- During influenza season, testing for influenza is indicated for all patients with pneumonia, with treatment with oseltamivir if flu testing is positive (or empirically pending influenza test results).
- Urinary pneumococcal and Legionella antigen testing is recommended for people with severe or health care associated pneumonia.
- Pleural fluid should be always cultured, if it is collected.
One main advantage of diagnostic testing is that (in the case of healthcare associated pneumonia) it encourages safe de-escalation of therapy to traditional agents for community-acquired pneumonia, when cultures are negative.
How Long to Treat Pneumonia? 5 to 7 Days (Even in ICU)
If a patient is not immunosuppressed, community-acquired pneumonia should be treated with antibiotics for 5 to 7 days, according to the IDSA and ATS. Their advice applies even to severely ill patients being treated for pneumonia in the ICU.
Whom Should Be Admitted for Pneumonia Treatment?
There is wide variation in admission rates for community acquired pneumonia; objective scoring systems were created to help standardize decision making and risk stratification. The Pneumonia Severity Index (PSI) is validated but requires a computer to calculate. The CURB 65 score is easier, with 1 point each for:
- Uremia (BUN ≥20 mg/dL)
- Respiratory rate ≥30 / minute,
- Blood pressure <90 mm Hg systolic or ≤60 mm Hg diastolic
- ≥65 years of age
Patients with a CURB-65 score ≥3 should be hospitalized. CURB-65 has not been as well validated as PSI, and notably does not include oxygen saturation. As with all scoring systems, PSI and CURB-65 provide guidance but are not absolute; admission decisions may also include factors like lack of social support, debility, degree of hypoxemia and its chronicity, or nausea.
Where to Treat Pneumonia: Medical Ward or ICU?
The decision of whether to admit a patient with community acquired pneumonia to the ICU might have a major impact on outcomes. Patients initially admitted to the medical floor, then transferred to the ICU within 48 hours have had greater mortality than those with more severe disease (intubated or on pressors) admitted directly to the ICU on admission.
The IDSA and ATS have suggested that any patient with 3 of the following 9 criteria be considered for ICU admission: confusion, BUN ≥20 mg/dL, respiratory rate ≥ 30/min, multilobar pneumonia, hypoxemia with PaO2/FiO2 < 250, platelets < 100,000 mm³, hypotension (SBP<90 mm Hg), hypothermia < 36º C, or white blood cell count < 4,000/mm³.
However, admitting every patient with 3 IDSA/ATS criteria to the ICU results in a high proportion of observation-only patients who never need mechanical ventilation or vasopressors; this might strain resources in already crowded ICUs.
The IDSA/ATS severity criteria might better be applied in the emergency department. In one quality improvement study, patients "flagged" in the ED with 3 criteria and given greater attention had less mortality (6% vs 23%) and fewer ICU transfers (15% vs 32%). This was achieved without markedly increasing admissions from the ED to the ICU.
Regarding antibiotics, ß-lactam + macrolide combinations are associated with better outcomes than beta-lactams + fluoroquinolones, in observational studies of patients with severe pneumonia admitted to ICUs. (Macrolides and fluoroquinolones have similar bacterial coverage, but macrolides' anti-inflammatory effects are hypothesized to provide additional benefit.)
Richard G. Wunderink and Grant W. Waterer. Community-Acquired Pneumonia. N Engl J Med 2014;370:543-51. DOI: 10.1056/NEJMcp1214869
Pneumonia Severity Index Calculator. December 2003. Agency for Healthcare Research and Quality. Rockville, MD.
Infectious Diseases Society of America/American Thoracic Society Consensus Guidelines on the Management of Community-Acquired Pneumonia in Adults. Clinical Infectious Diseases 2007; 44:S27–72.