Background Stable COPD patients are colonised if potentially pathogenic organisms are identified on sputum culture. Associations between colonisation and clinical features such as exacerbation frequency and airway inflammation have been suggested. There is however limited data describing the reproducibility of sputum microbiology results in clinically stable COPD patients.
Aims Examine repeatability of sputum microbiology in subjects with stable COPD over time.
Methods Subjects with COPD were enrolled into an observational study and seen at baseline and at stable visits after 3 and 6 months. Sputum was obtained and samples were divided and analysed over time using standard culture, semi-quantitative bacterial count (colony forming units, CFU), PCR for potentially pathogenic organisms [(Haemophilus influenzae (HI), Streptococcus pneumoniae (SP), Staphylococcus aureus (SA), Moraxella catarrhalis (MC)] and quantitative bacterial 16S analysis.
Results 63 subjects provided paired sputum samples; 52 were male with a mean (SD) FEV1 (L) and FEV1/FVC ratio (%) of 1.48(0.54) and 53% (12.8) respectively. 40% were current smokers with an exacerbation frequency of 3 in the preceding year.
Results for standard culture were divided into two groups (culture positive or negative). Results are expressed as Kappa values (95% CI). There was moderate agreement after 3 months, Kappa = 0.48 (0.24 to 0.71); and after 6 months, Kappa = 0.50 (0.25 to 0.76). Individual PCR revealed fair agreement after both time intervals. After 3 months, HI=0.17(–0.08 to 0.43), SA=0.27(–0.03 to 0.56), SP=0.30(0.06 to 0.53), MC=0.19(–0.04 to 0.43). After 6 months, HI=0.09(–0.18 to 0.35), SA=0.10(–0.22 to 0.43), SP=0.37(0.13 to 0.62) and MC= –0.14(–0.4 to 0.11).
Quantitative bacterial analysis demonstrated no differences (mean difference; 95% CI) at 3 or 6 months in bacterial load measured by CFU (–0.18; –0.41 to 0.04, p=0.11 and –0.06; –0.32 to 0.2, p=0.65 respectively) or 16S (–0.03; –0.28 to 0.33, p=0.86 and –0.1; –0.42 to 0.22, p=0.54 respectively).
Discussions These results demonstrate that sputum microbiological assessment in stable COPD is complex. Further longitudinal assessments of sputum microbiology and associations with clinical features are needed.