Clinical Citations: Is there a role for PCR in the evaluation of invasive aspergillosis?

April 1, 2006

High-resolution CT (HRCT) and galactomannan enzyme-linked immunosorbent assay (ELISA) are valuable in the diagnosis of invasive aspergillosis; however, both of these methods have limitations. Although the role of polymerase chain reaction (PCR) testing has not been established yet, the results of a study in the Netherlands are encouraging.

High-resolution CT (HRCT) and galactomannan enzyme-linked immunosorbent assay (ELISA) are valuable in the diagnosis of invasive aspergillosis; however, both of these methods have limitations. Although the role of polymerase chain reaction (PCR) testing has not been established yet, the results of a study in the Netherlands are encouraging.

White and colleagues retrospectively studied 203 patients who were at risk for invasive fungal infection; most of the patients had hematologic malignancy. PCR testing had been performed in all patients, and galactomannan ELISA had been done in 116 of them. Fungal infections were classified on the basis of European Organi- sation for Research and Treatment of Cancer (EORTC) and Mycoses Study Group (MSG) criteria.

PCR had a sensitivity of up to 92.3% and a specificity of up to 94.6%, and it had 76.7% agreement with galactomannan ELISA. PCR also had good agreement with HRCT results, although HRCT was not done in all of the patients.

The authors concluded that negative PCR results can be used to rule out invasive aspergillosis and can be used to limit the use of empiric antifungal therapy. They say it can be particularly useful when used in conjunction with antigen testing.

In an accompanying editorial, Donnelly notes that PCR was not included in the EORTC and MSG criteria for the diagnosis of invasive aspergillosis for a number of reasons, including the absence of standardization and the absence of a commercially available system. Although progress in this area has been slow, the study by White and associates represents a noteworthy advance.

Donnelly points out that physician compliance was low in some aspects of this study. For example, compliance was only about 50% with respect to the requirement that blood samples be submitted twice weekly for galactomannan ELISA testing. This may have resulted in underdiagnosis of invasive aspergillosis. Compliance with HRCT recommendations was also low; HRCT was done in only 37% of patients, despite the recommendation that it be done in all patients in whom invasive aspergillosis is suspected.

Diagnostic features of invasive aspergillosis are presented in the Table.