Diagnostic role of polymerase chain reaction in bronchoalveolar lavage fluid for invasive pulmonary aspergillosis in immunocompromised patients – A retrospective cohort study. (June 2019)

• Record Type: Journal Article
• Authors: Hardak, Emilia; Fuchs, Eyal; Leskes, Hannah; Geffen, Yuval; Zuckerman, Tsila; Oren, Ilana
• Is Part Of: International journal of infectious diseases. Volume 83(2019)
• Page Start: 20
• Page End: 25
• Publication Date: 2019-06
• Subjects: PCR -- Bronchoalveolar lavage -- Diagnosis -- Invasive pulmonary aspergillosis; Communicable diseases -- Periodicals; Communicable Diseases -- Periodicals; Communicable diseases; Periodicals; Electronic journals; 616.9
• DOI: 10.1016/j.ijid.2019.03.025
• Languages: English
• ISSNs: 1201-9712

Abstract

Highlights: Early diagnosis of invasive pulmonary aspergillosis is notoriously difficult. We have evaluated the diagnostic role of PCR detection of aspergillus DNA in Bronchoalveolar lavage fluid, in a large cohort of immunocompromised patients. We have found it to have an added value to other mycological criteria and thus a valuable aid for early and accurate diagnosis of IPA. Abstract: Background: This study aimed to evaluate the diagnostic role of PCR detection of Aspergillus DNA in the broncho-alveolar lavage (BAL) fluid in a large cohort of patients suspected to have invasive pulmonary aspergillosis (IPA). Methods: Consecutive immunocompromised patients who underwent bronchoscopy with BAL sampling and PCR detection of Aspergillus DNA for the diagnosis of pulmonary infiltrates were included in the study. Galactomannan (GM) antigen testing in BAL and serum and BAL fungal culture were also performed. Patients were classified as having IPA (proven/probable/possible) or no-IPA according to the EORTC/MSG diagnostic criteria. Results: During 12 years (2005–2016), 1248 bronchoscopies were performed for 1072 patients. 77% had hematological malignancy, of them 40% had AML and 35.6% underwent HSCT. IPA was diagnosed in 531 patients (42.5%), 7-proven, 280-probable and 244-possible. PCR was positive in 266 cases, of them 213 had IPA, indicating a true positive rate of 80% (213/266) and a false positive rate of 20% (53/266). These results establish the diagnostic performance of PCR to have sensitivity of 40%, specificity of 93%, PPV- 80% and NPV-68%. Of 244 patients with possible IPA, 80 had positive PCR. Including PCR in the diagnostic criteria would move 80 cases from the possible group to the probable one. A combination of positive PCR and/or BAL-GM increases sensitivity to 74%, while positivity of both tests elevates PPV to 99.4%. Conclusions: Inclusion PCR for the detection of Aspergillus-DNA in BAL in the mycological criteria of the EORTC/MSG definitions increases the rate and the certainty of IPA diagnosis.