Pulmonary exacerbations of cystic fibrosis: Definition, inflammatory markers, and the role of atypical bacteria and respiratory viruses

The underlying hypothesis was that co-infection with respiratory viruses and/or "atypical" bacteria in young adult patients with cystic fibrosis, who already experience chronic pulmonary bacterial colonisation, contributes to pulmonary morbidity. The most likely mechanism would be the precipitation of acute pulmonary exacerbations. Peripheral markers of inflammation were first assessed as end-points. White cell count, plasma viscosity, and c-reactive protein increased in a quadratic fashion with declining pulmonary function. White cell count and c-reactive protein, but not plasma viscosity, demonstrated a fall with intravenous antibiotic treatment which could not be accounted for by improvement in pulmonary function. All three markers were frequently normal at the outset of treatment. In contrast neutrophil elastase was elevated in all patients tested both at the outset and completion of treatment. To define respiratory exacerbation, four approaches were compared. Inflammatory markers were too insensitive. The coefficient of variation of quantitative sputum bacteriology was too high. A combination of change from baseline FEV1 or FVC of 10% or more, or an increase in two or more lower respiratory symptoms, was found to be optimal. With this definition, no permanent loss of lung function was attributable to exacerbations. With conventional culture and serological methods, respiratory viruses were temporally associated with 11 of 373 exacerbations (3%). There was one asymptomatic seroconversion. All diagnoses were made with serology. No atypical bacteria were identified. The rate of seroconversion in the 60 patients who were followed for the full two year period was 0.083/patient/year. Because of concern about the insensitivity of the culture methods, a multiplex PCR method was developed. With combined serology and PCR, 12 of 82 (15%) exacerbations in an 8 month period were associated with viral infection. Infections were no more common in patients with CF than in healthy controls, controls with asthma, and controls with non-CF bronchiectasis. Patients who had viral infections had worse pulmonary function throughout the two year study. Plasma viscosity and CRP, but not white cell count, were higher in exacerbations with viral co-infection, and there was a greater decline in pulmonary function from pre-exacerbation baseline. FEV1 remained depressed following treatment of the exacerbation, but in most patients gradually improved subsequently. Viral infection was associated with a rise in anti-pseudomonal IgG ELISA score, which may predate deterioration in pulmonary function