2199. The Etiology of Community-Acquired Pneumonia with Attention to the Role of Normal Respiratory Flora

Abstract

BackgroundIntensive studies have failed to identify an etiologic agent in >50% of patients (patients) who are hospitalized for community-acquired pneumonia (CAP). Gram stain and culture of sputum samples frequently yield “normal respiratory flora” (NRF). We hypothesized that careful study might (1) increase the yield of recognized pathogens; and (2) show, in some patients, an etiologic role for NRF.MethodsWe studied a convenience sample of adults hospitalized for CAP at a VA Medical Center if they met four criteria: (1) clinical syndrome consistent with pneumonia; (2) newly recognized pulmonary infiltrate; (3) sputum with > 10 WBC per epithelial cell; and (4) < 18 hours antibiotic treatment. For quantification of bacteria, sputum was liquefied in 2% N-acetyl cysteine and diluted serially. Other studies in nearly all patients included blood cultures, urine for pneumococcal (Spn) and Legionella antigen, procalcitonin, B-natriuretic protein and PCR for 13 respiratory viruses, Mycoplasma and Chlamydia. >106 bacteria/mL and a consistent Gram stain indicated a bacterial cause, positive viral PCR indicated a viral cause, and both indicated coinfection.Results119 patients met study criteria. Recognized bacterial pathogens alone were identified in 47 (40%) cases led by Spn 17 (14%), Haemophilus 17 (14%) and S. aureus 6 (5%). A virus alone was identified in 17 (15%) and coinfection in 11 (9%). We applied these same criteria for NRF. NRF alone were found in 22 (19%) patients with S. mitis predominating. NRF and a respiratory virus were coinfecting in 10 (8%) patients. In total, with the inclusion of NRF, an etiologic agent was found in 95% of patients.ConclusionOur high yield is attributable to selection criteria. With a good-quality sputum and absent prolonged antibiotics, a bacterial cause for CAP was found in 59% of patients, a viral cause in 15%, and coinfection in 17%. Bacterial CAP due to recognized pathogen follows microaspiration of colonizing bacteria from the upper airways. Aspiration of a sufficient inoculum of so-called NRF, especially in older adults or those with damaged clearance mechanisms, might well do the same. Careful microbiologic study of patients who are able to provide a valid sputum sample before prolonged antibiotics enables a microbiologic diagnosis in nearly all cases and shows a potential etiologic role for NRF in about 20%. Disclosures All authors: No reported disclosures.