Background: AspID (OLM innovations, Newcastle, UK) is a new quantitative assay based on
multiplex real-time PCR for detection of clinically relevant Aspergillus species. The diagnostic
performance of this assay was evaluated in bronchoalveolar lavage fluid (BALF) specimens
from patients at risk of developing invasive pulmonary aspergillosis (IPA).
Material/methods: Thirty-six BALF specimens obtained from 18 patients with IPA and 18 with
no evidence for IPA were studied (Table 1). Patients were classified as having IPA when BALF
galactomannan (GM) determination yielded an optical density index (ODI) >3.0 and patients
had clinical and radiological findings compatible with IPA. Those without IPA had a BALF GM
level <0.5 and no clinical/radiological findings compatible with IPA. Patients with and
without IPA were matched 1:1 regarding underlying diseases and ICU admission. For
detection of Aspergillus DNA, BALFs were extracted using the specific B protocol of the
NucliSENS easyMAG instrument (bioMérieux, Marcy-l´Etoile, France) with an input volume of
400 μl and an elution volume of 40 μl. After the lysis step, 4 μl of internal extraction control
included in the AspID assay was added. Amplification and detection were performed on the
LC 480 II instrument (Roche Diagnostics, Rotkreuz, Switzerland).
Results: With the new AspID, 20 BALF samples were found to be positive and 14 negative.
Two samples (one from a patient with IPA and one from a patient without IPA) showed
inhibition and were excluded from analysis. When AspID results were compared to those
obtained from GM determination, 29 were found to be concordant and 5 discordant (four
AspID-positives in patients without IPA and one AspID-negative in a patient with IPA).
Sensitivity, specificity, positive- and negative likelihood ratio for AspID including 95%
confidence interval (CI) were 94.1% (95%CI 73.3 – 99.9), 76.5% (95% CI 50.1 – 93.2), 4 (95%
CI 1.7 – 9.5) and 0.1 (95% CI 0.01 – 0.5) respectively.
Conclusions: Detection of clinically relevant Aspergillus species in BALF specimens with
AspID PCR reagents seems to be a promising diagnostic approach in patients at risk for IPA. It
may allow early diagnosis and rapid initiation of anti-mold therapy. Four results may have
been false positive. This may be due to contamination during bronchoscopy or during the
laboratory workflow, possibly through Aspergillus species contamination in the airways.
Full conference title:
- ECCMID 27th (2017)