Criteria for microbiological diagnosis
| Organism | Diagnostic criteria |
| All bacteria | Isolation from blood processed in Bactec 9240 system or from pleural fluid, bronchoscopic aspirates, or post mortem lung aspirates |
| S pneumoniae, H influenzae orM catarrhalis infection | (a) Isolation from washed and diluted sputum in significant numbers by semiquantitative culture or |
| (b) for S pneumoniae: • Pneumococcal antigen (PCA) detected in urine (BINAX-NOW kit, results read at 60 minutes) or in sputum by countercurrent immunoelectrophoresis (CIE) or | |
| (c) Serological criteria for S pneumoniae: • ⩾3-fold rise in antibody titre against C-polysaccharide (CPS) or ⩾2-fold rise against pneumolysin (PLY), pneumococcal surface antigen A (PsaA)22 or PLY-, PsaA- or CPS-specific immune complexes detected23 | |
| (d) Serological criteria forH influenzae or M catarrhalis: • ⩾3-fold rise in antibody titre24 | |
| Infection with other bacteria including Gram negative enterobacteria (GNEB) andStaph aureus | The predominant organism in the sputum Gram stain in addition to isolation from washed and diluted sputum in significant numbers by semiquantitative culture |
| Atypical and viral pathogens | |
| C pneumoniae | (a) at least 3-fold rise in IgG or IgA antibodies or (b) presence of IgM antibody, using EIA kit (Labsystems, Helsinki, Finland) |
| L pneumophila | Isolation of organism from respiratory samples or Legionellaantigen detected in urine by Biotest kit25 26 or 4-fold or greater rise in immunofluorescent antibody titre to ⩾64 using formalised yolk sac antigen to L pneumophilaserogroup 1 |
| M pneumoniae, Chlamydia spp,C burnetti, influenza A and B, respiratory syncytial virus (RSV) and adenoviruses | 4-fold or greater antibody rise or a single titre of ⩾128, by complement fixation test |
Serum samples were tested at the National Public Health Institute, Department in Oulu, Finland for antibody responses toC pneumoniae, H influenzae, M catarrhalis, andS pneumoniae. Urine antigen testing was performed in conjunction with the Central Public Health Laboratory, Colindale, London, UK.
The criteria used to define infection in the 1982 BTS study were followed, updated for new investigations and pathogens.3