Purpose The reported prevalence of Aspergillus fumigatus in CF sputum varies widely from 12 to 57%. While patients with ABPA are routinely treated with antifungals, it is not know whether colonised or sensitised patients would benefit from antifungal treatment. To aid treatment decisions and to monitor response more accurate methods to detect Aspergillus in sputum are needed. This study aimed to identify CF patients with Aspergillus colonisation, using real time PCR, and examine the relationship of colonisation to markers of sensitisation.

Methods 108 adult CF patients provided a sputum sample and a blood sample. Serological tests included total IgE, specific A. fumigatus IgE and specific A. fumigatus IgG performed by Phadia ImmunoCAP^® assay, and A. fumigatus precipitins by counter immunoelectrophoresis. Sputum was homogenised with sputasol and sonication. 10 μl was cultured on sabouraud agar (Oxoid, UK) for 72 h. The remaining sample was used in a commercial real time PCR assay, MycAssay Aspergillus. Patients on antifungal treatment were excluded from serological data analysis.

Results 30% of the 108 sputum samples were positive for Aspergillus species by standard culture whereas 80% were positive for Aspergillus species by PCR. 15 patients were on antifungal therapy of whom 7 were PCR positive. Of the serological tests, only specific IgG correlated to positive PCR. Using a ROC curve, a specific IgG level above 65 mg/l gave 85% sensitivity and 100% specificity for positive PCR. 12 patients met the 2003 consensus minimum criteria for ABPA. All were PCR positive supporting the use of antifungals for ABPA. 38 patients were sensitised to aspergillus (specific IgE >0.4 KUa/l), 28 of these were PCR positive. A group of 32 patients was identified that had a rise in specific IgG and positive PCR but no IgE rise. They may represent ‘aspergillus bronchitis’. All patients with negative serology were PCR negative.

Conclusion Real time PCR can accurately identify CF patients with Aspergillus in their sputum, including those in whom antifungal therapy is inadequate. However, PCR alone cannot distinguish between ABPA, sensitisation and colonisation. Positive PCR correlates to a specific IgG >65 mg/l. A randomised trial of antifungal therapy is required to determine if there is clinical benefit in treating PCR positive patients.