Etiology of community-acquired pneumonia and diagnostic yields of microbiological methods: a 3-year prospective study in Norway

Background: Despite recent advances in microbiological techniques, the etiology of community-acquired pneumonia (CAP) is still not well described. We applied polymerase chain reaction (PCR) and conventional methods to describe etiology of CAP in hospitalized adults and evaluated their respective diagnostic yields. Methods: 267 CAP patients were enrolled consecutively over our 3-year prospective study. Conventional methods (i.e., bacterial cultures, urinary antigen assays, serology) were combined with nasopharyngeal (NP) and oropharyngeal (OP) swab samples analyzed by real-time quantitative PCR (qPCR) for Streptococcus pneumoniae, and by real-time PCR for Mycoplasma pneumoniae, Chlamydophila pneumoniae, Bordetella pertussis and 12 types of respiratory viruses. Results: Etiology was established in 167 (63%) patients with 69 (26%) patients having >= 1 copathogen. There were 75 (28%) pure bacterial and 41 (15%) pure viral infections, and 51 (19%) viral-bacterial coinfections, resulting in 126 (47%) patients with bacterial and 92 (34%) patients with viral etiology. S. pneumoniae (30%), influenza (15%) and rhinovirus (12%) were most commonly identified, typically with >= 1 copathogen. During winter and spring, viruses were detected more frequently (45%, P=.01) and usually in combination with bacteria (39%). PCR improved diagnostic yield by 8% in 64 cases with complete sampling (and by 15% in all patients); 5% for detection of bacteria; 19% for viruses (P=.04); and 16% for detection of >= 1 copathogen. Etiology was established in 79% of 43 antibiotic-naive patients with complete sampling. S. pneumoniae qPCR positive rate was significantly higher for OP swab compared to NP swab (P<.001). Positive rates for serology were significantly higher than for real-time PCR in detecting B. pertussis (P=.001) and influenza viruses (P<.001). Conclusions: Etiology could be established in 4 out of 5 CAP patients with the aid of PCR, particularly in diagnosing viral infections. S. pneumoniae and viruses were most frequently identified, usually with copathogens. Viral-bacterial coinfections were more common than pure infections during winter and spring; a finding we consider important in the proper management of CAP. When swabbing for qPCR detection of S. pneumoniae in adult CAP, OP appeared superior to NP, but this finding needs further confirmation.