Aspergillus Galactomannan Lateral Flow Assay for Rapid Diagnosis of Invasive Aspergillosis in Bronchoalveolar Lavage

Author:

Martin Hoenigl 1, Sharon Reed 2, Sanjay Mehta 2, Nancy Law 2, Saima Aslam 1, Randy Taplitz 2, Jeffrey Jenks 2

Author address:

1 University of California, San Diego, San Diego, CA
2 UCSD, San Diego, CA

Full conference title:

ID week 2019

Date: 27 October 2019

Abstract:

Background :

Early diagnosis and treatment of invasive pulmonary aspergillosis (IPA) remains the most important factor to reduce mortality. Diagnosis remains a challenge, however, due to unspecific clinical presentation and radiological findings. Only very recently rapid tests for IPA have been developed. The objective of this study was to evaluate the performance of the new CE-marked Aspergillus Galactomannan Lateral Flow Assay (LFA; IMMY, Oklahoma, USA; Figure) for IPA in patients with and without hematological malignancies.

Methods :

The Aspergillus Galactomannan LFA was retrospectively performed according to the manufacturer’s instructions in 106 previously frozen bronchoalveolar lavage fluid (BAL) samples from 106 patients at risk for IPA (23% with underlying hematological malignancies). Samples were collected between September 2016 and September 2018 at the University of California San Diego. Performance of the LFA was compared to Galactomannan, BAL culture and the Aspergillus-specific LFD (another rapid test for IPA). IPA was classified according to revised EORTC/MSG criteria.

Results : Overall, 22 patients met criteria of probable or proven IPA, 9 possible IPA, while 75 patients did not fulfill criteria of IPA. Sensitivity of the Apergilllus Galactomanann LFA for probable/proven IPA was 77% (17/22). Sensitivity was similar to BAL GM (77% with a cutoff of 1.0 ODI), but higher compared to the Aspergillus-specific LFD (59%), and BAL culture (23%). The LFA resulted negative in 7/9 cases with possible IPA and 47/73 cases without IPA (overall specificity 66%, 54/82). The less than perfect specificity was driven particularly by non-neutropenic patients (specificity 62%, 43/69), while specificity was 85% among patients with underlying hematological malignancies. Lower specificity among non-neutropenic patients was also observed for the BAL GM (overall 77%, non-neutropenic patients 72%), the Aspergillus-specific LFD (overall 70%, non-neutropenic patients 67%) and BAL culture (overall 90%, non-neutropenic 88%).

Conclusion :

Our study indicates that the LFA may be useful for rapid diagnosis of IPA in BALF when IPA is clinically suspected. The lower specificity in non-neutropenic patients may be explained by limited applicability of the EORTC/MSG criteria in those patients.