Technical advance

Development and optimization of quantitative PCR for the diagnosis of invasive aspergillosis with bronchoalveolar lavage fluid

Prasanna D Khot¹ , Daisy L Ko¹ , Robert C Hackman^1,2 and David N Fredricks^1,2

¹  Fred Hutchinson Cancer Research Center, Seattle, WA, USA

²  University of Washington, Seattle, WA, USA

author email corresponding author email

BMC Infectious Diseases 2008, 8:73doi:10.1186/1471-2334-8-73

Published:	29 May 2008

Abstract

Background

The diagnosis of invasive pulmonary aspergillosis (IPA) remains challenging. Culture and histopathological examination of bronchoalveolar lavage (BAL) fluid are useful but have suboptimal sensitivity and in the case of culture may require several days for fungal growth to be evident. Detection of Aspergillus DNA in BAL fluid by quantitative PCR (qPCR) offers the potential for earlier diagnosis and higher sensitivity. It is important to adopt quality control measures in PCR assays to address false positives and negatives which can hinder accurate evaluation of diagnostic performance.

Methods

BAL fluid from 94 episodes of pneumonia in 81 patients was analyzed. Thirteen episodes were categorized as proven or probable IPA using Mycoses Study Group criteria. The pellet and the supernatant fractions of the BAL were separately assayed. A successful extraction was confirmed with a human 18S rRNA gene qPCR. Inhibition in each qPCR was measured using an exogenous DNA based internal amplification control (IAC). The presence of DNA from pathogens in the Aspergillus genus was detected using qPCR targeting fungal 18S rRNA gene.

Results

Human 18S rRNA gene qPCR confirmed successful DNA extraction of all samples. IAC detected some degree of initial inhibition in 11 samples. When culture was used to diagnose IPA, the sensitivity and specificity were 84.5% and 100% respectively. Receiver-operating characteristic analysis of qPCR showed that a cutoff of 13 fg of Aspergillus genomic DNA generated a sensitivity, specificity, positive and negative predictive value of 77%, 88%, 50%, 96% respectively. BAL pellet and supernatant analyzed together resulted in sensitivity and specificity similar to BAL pellet alone. Some patients did not meet standard criteria for IPA, but had consistently high levels of Aspergillus DNA in BAL fluid by qPCR.

Conclusion

The Aspergillus qPCR assay detected Aspergillus DNA in 76.9% of subjects with proven or probable IPA when the concentrated BAL fluid pellet fraction was used for diagnosis. There was no benefit from analyzing the BAL supernatant fraction. Use of both extraction and amplification controls provided optimal quality control for interpreting qPCR results and therefore may increase our understanding of the true potential of qPCR for the diagnosis of IPA.