Aspergillus-PCR in bronchoalveolar lavage for detection of invasive pulmonary aspergillosis in immunocompromised patients
1 Clinic of Pulmonary Medicine and Respiratory Cell Research, University Hospital Basel, Basel, Switzerland
2 Clinic for Hematology, University Hospital Basel, Basel, Switzerland
3 Department of Microbiology, Kanton Hospital Liestal, Liestal, Switzerland
BMC Infectious Diseases 2012, 12:237 doi:10.1186/1471-2334-12-237Published: 2 October 2012
Invasive fungal disease (IFD) is a frequent and serious infectious complication in immunocompromised patients. Culture and cytology in bronchoalveolar lavage (BAL) have a high specificity but low sensitivity for the diagnosis of IFD as assessed by histology. Molecular methods are expected to allow a rapid diagnosis of IFD with a high sensitivity. We evaluated the diagnostic accuracy of conventional nested PCR in the bronchoalveolar fluid to diagnose IFD in severely immunocompromised patients.
Consecutive immunosuppressed patients undergoing bronchoscopy for suspected pulmonary infection in a tertiary care hospital were included. Patients were classified as having “proven”, “probable”, “possible”, and “no” IFD based on definitions of the European Organization for Research and Treatment of Cancer and National Institute of Allergy and Infectious Diseases (EORTC/NIAID) and on clinical grounds. Conventional nested PCR for aspergillus fumigatus, flavus, niger, glaucus, terreus and tomarrii were applied to 2.5 ml bronchoalveolar fluid.
A total of 191 patients were included. Mean age was 51 y, 61% were male. There were 129 patients with hematological conditions, 26 solid organ transplant recipients, 24 auto-immune disorders, and 12 HIV. According to the EORTC/NIAID classification, there were 53 patients with potential IFD: 3 (2%) had proven, 8 (4%) probable, 42 (22%) possible and 138 (72%) no IFD. A total of 111 (58%) of the patients - 10 (90.9%) proven or probable IFD, 32 (76.2%) possible IFD and 69 (50%) “no” IFD) were on anti-fungal therapy at the time of bronchoscopy. Conventional nested PCR for Aspergillus was positive in 55 cases (28.8%). According to these results, sensitivity, specificity, PPV and NPV for “proven” IFD was 0%, 71%, 0%, 98%, respectively and “probable” IFD was 36%, 72%, 7%, 95%, respectively. In 53 (28%) cases there was a strong clinical suspicion of IFD in the chest-x-ray and/or chest-CT irrespective of the EORTC/NIAID classification. However, from those, only 15 (28%) had a positive conventional nested PCR.
In our experience, conventional nested Aspergillus PCR in the BAL seems to be of limited usefulness for detection of invasive fungal disease in immunocompromised patients due to the limited sensitivity and specificity of the method.