Incidence and characteristics of viral community-acquired pneumonia in adults.

BACKGROUND: In adults, viral causes of community-acquired pneumonia (CAP) are poorly characterised. The aims of this study were to characterise the viral aetiology of CAP in adults by using an extensive array of viral diagnostic tests and to compare the characteristics of viral pneumonia with those of pneumococcal pneumonia. METHODS: Adults admitted to Christchurch Hospital over a 1-year period with CAP were included in the study. Microbiological testing methods included blood and sputum cultures, urinary antigen testing for Streptococcus pneumoniae and Legionella pneumophila, antibody detection in paired sera and detection of respiratory viruses in nasopharyngeal swabs by immunofluorescence, culture and PCR. RESULTS: Of 304 patients with CAP, a viral diagnosis was made in 88 (29%), with rhinoviruses and influenza A being the most common. Two or more pathogens were detected in 49 (16%) patients, 45 of whom had mixed viral and bacterial infections. There were no reliable clinical predictors of viral pneumonia, although several variables were independently associated with some aetiologies. The presence of myalgia was associated with pneumonia caused by any respiratory virus (OR 3.62, 95% CI 1.29 to 10.12) and influenza pneumonia (OR 190.72, 95% CI 3.68 to 9891.91). Mixed rhinovirus/pneumococcal infection was associated with severe disease. CONCLUSIONS: Virus-associated CAP is common in adults. Polymicrobial infections involving bacterial and viral pathogens are frequent and may be associated with severe pneumonia.

Comparison of real-time PCR and conventional hemi-nested PCR for the detection of Bordetella pertussis in nasopharyngeal samples.

We directly compared a conventional hemi-nested PCR assay with a real-time (LightCycler) PCR assay for the detection of Bordetella pertussis in nasopharyngeal samples. Of the 152 samples tested, 49 (32%) were positive by first-round conventional PCR, 56 (37%) were positive by the hemi-nested PCR assay, and 39 (26%) were positive by the real-time assay. All samples testing positive with the real-time assay were also positive by the hemi-nested assay (both first- and second-round PCR), and all culture-positive samples tested positive by both PCR assays. These results suggest that the hemi-nested assay is more sensitive than the real-time assay for detecting B. pertussis.