PRACTICE GUIDELINES FOR MANAGEMENT OF COMMUNITY-ACQUIRED PNEUMONIA IN ADULTS

PRACTICE GUIDELINES FOR MANAGEMENT OF COMMUNITY-ACQUIRED PNEUMONIA IN ADULTS (Bartlett J, et al. CID 2000;31:347): These are the guidelines of the IDSA that were published in the August 2000 issue of CID. Process: The guideline panel is composed of six members: John Bartlett (Hopkins), Scott Dowell (CDC), Lionel Mandell (McMaster University, Toronto), Thomas File, Jr. (Northeastern Ohio University College of Medicine, Cleveland), Dan Musher (Baylor College of Medicine, Houston), and Michael Fine (University of Pittsburgh). The guidelines also make extensive use of expertise within the society by review from 28 members who are assigned specific content areas for review according to expertise. The guidelines are compiled through meetings, conference calls, and correspondence. The final product is submitted to the Guideline Committee for their review and is then submitted to the Council for final review. Grading system: As is customary for IDSA guidelines, recommendations are categorized by letters A–D indicating the strength of the recommendation, and by Roman numerals I–III indicating the quality of the evidence to support the recommendation. Thus, the recommendation for gram stain of expectorated sputum was rated B-II, indicating moderate strength of the recommendation and evidence from at least one clinical trial to support the recommendation. Site of care: Recommendations are based on the Pneumonia Patient Outcome Research Team (Pneumonia PORT) study in which patients are stratified in to five severity classes based on a point system, which has been shown with extensive validity tests to correlate well with mortality. It is acknowledged that social factors, such as outpatient support mechanisms, need to be considered in this decision. Laboratory tests: All patients with possible pneumonia should have a chest x-ray. Hospitalized patients should have CBC, a chemistry profile, HIV serology (with informed consent), and oxygen saturation assessment. For microbiology studies, there should be two pretreatment blood cultures and gram stain of expectorated sputum; selected patients should have microbiology studies for tuberculosis and for Legionella. The preferred tests for detection of Legionella are the urinary antigen assay (which detects only L. pneumophila serogroup 1 that accounts for 40–70% of legionnaires’ disease cases) and Legionella culture using selective media. No tests for etiologic agent are considered standard for outpatients with CAP. Special Consideration: This section includes a discussion of issues that are considered especially important, including the following: Pneumococcal pneumonia: This organism is the major identifiable pathogen in virtually all studies of CAP. The pneumococcus has shown increased resistance to penicillin and most other drugs in recent years. The preferred drug for strains with an MIC ≤2 μg/mL is penicillin or amoxicillin; for penicillin-resistant strains, the selection should be based on in vitro sensitivity tests with probable activity of cefotaxime, ceftriaxone, fluoroquinolones, or vancomycin. Aspiration pneumonia: This term is used loosely to refer to three distinctly different syndromes: 1) chemical pneumonitis due to chemical insult, such as acid aspiration (“Mendelson’s syndrome”); 2) aspiration of inert fluids (drown) or particulate matter with mechanical obstruction; and 3) infection, usually due to oropharyngeal bacteria, primarily anaerobes in the gingival crevice. Anaerobic bacteria: These organisms have accounted for 10–20% of CAP cases, according to TTA data, but will not be detected in the current era due to the failure to obtain appropriate specimens to establish this diagnosis. The best clues are suspected aspiration, infection in a dependent pulmonary segment, chronic infection, cavitation, and/or putrid discharge. C. pneumoniae: This organism is estimated to cause 5–15% of cases, but is very difficult to prove due to the lack of valid diagnostic tests that are readily available. Legionnaire’s disease: This organism and S. pneumoniae are emphasized as the two pathogens that cause serious, life-threatening pneumonia in previously healthy adults. Hantavirus Pulmonary Syndrome: This is a recently recognized agent of serious disease that is not really a pneumonia, but may present with ARDS. Mycoplasma pneumoniae: This organism is traditionally reviewed as an important pathogen in children and young adults, and diagnostic tests available to most physicians are not good. P. carinii (PCP): This topic was included even though the guidelines do not apply to immunocompromised patients because of its importance in the current era as a presenting condition in those with HIV infection. Influenza: Emphasis is placed on the important role this virus plays in morbidity and mortality, the use of rapid diagnostic tests with a sensitivity of 70–90% (about the same as clinicians), and the availability of four drugs for therapy. Empyema: The importance of pleural fluid pH in decisions regarding drainage is emphasized. Acute bronchitis: This really does not belong in a discussion of pneumonia, but was included because of the importance of distinguishing acute bronchitis and pneumonitis due to the implications for antibiotic abuse. Pneumonia in the context of bioterrorism: This topic was included to further call attention to the importance of considering bioterrorism in cases of pneumonia that are severe, unusual, or epidemic, especially in the non-flu season. The three “category A” organisms that present with pneumonia are inhalation anthrax, pneumonic plague, and tularemia. Treatment: The assumption is that patients will be in two categories with regard to antibiotic selection. One will be patients who have an identified etiologic agent in which the selection of antibiotics is based on established or anticipated in vitro susceptibility tests. There is little controversy here. The second category is empiric treatment according to the following guidelines: Outpatients: Doxycycline, a macrolide, or a fluoroquinolone. Hospitalized patients on a general medical ward: Cephalosporin or betalactam/betalactamase inhibitor plus a macrolide, or a fluoroquinolone alone. Hospitalized patients in intensive care unit: A cephalosporin or betalactam/betalactamase inhibitor plus either a fluoroquinolone or a macrolide. Specific agents: The preferred fluoroquinolones are gatifloxacin, levofloxacin, or moxifloxacin (in no order), preferred cephalosporins for parenteral use are cefotaxime or ceftriaxone, and preferred betalactam/betalactamase inhibitor combinations are ampicillin-sulbactam, or pipercillin-tazobactam. Special considerations: Patients with structural lung disease should have consideration of an antipseudomonal agent plus a fluoroquinolone; for suspected aspiration, the preference is for a fluoroquinolone with or without clindamycin, metronidazole, or a betalactam–betalactamase inhibitor. For outpatients, physicians are reminded that the selection of drugs should take in to account regional patterns of susceptibility of S. pneumoniae. Influenza: Amantadine and rimantadine are active against influenza A. Relenza and oseltamivir are active against influenza A and B. All four must be given within 48 hours of the onset of the influenza symptoms, and they appear equally effective. IV-PO Switch: It is acknowledged that many hospitalized patients could probably be treated with oral agents from the start. Recommendations for switching in those given parenteral drugs are: condition is improving clinically, the patient is hemodynamically stable, is able to ingest drugs, and the GI track is functioning. Performance indicators: Most MCOs and other organizations, including HCFA, now want performance indicators, which are included in many IDSA guidelines, including this one: Blood cultures before antibiotic therapy in hospitalized patients. Initiation of antibiotics within eight hours of hospitalization. Use of culture and/or urinary antigen testing for detection of Legionella in 50% of patients hospitalized in the ICU for enigmatic CAP. Demonstration of infiltrate on chest x-ray in patients with an IDC-9 code diagnosis of CAP. Measurement of blood gases or pulse oximetry before admission or within eight hours of admission. Prevention: The main prevention methods are influenza vaccine and pneumococcal vaccine according to the CDC guidelines. Updates: A mechanism is now established to update these guidelines when new information becomes available that is relevant to the topic. This includes the availability of newly approved drugs, new clinical trial information, new diagnostic tests, etc. It is anticipated that most changes will result from external commentaries, which are highly desired.