Early diagnosis of invasive aspergillosis

Abstract

Sir—In his Feb 5 commentary on invasive aspergillosis, David Denning refers to a study of Maertens and colleagues, in which a prospective screening for circulating galactomannan by a sandwich ELISA (Platelia Aspergillus, Sanofi Diagnostics Pasteur, Marnes la Coquette, France) for patients with haematological disorders at high risk for invasive aspergillosis was assessed. The manufacturer of this assay recommends that the optical density (OD) index is calculated for the samples by dividing the OD value of each serum by the OD of a control serum (threshold serum) and to consider an index of 1·5 and above a positive test, an index of 1–1·5 a grey zone result, and an index of less than 1 a negative test. Maertens and colleagues deviated from the manufacturer’s recommendation and defined an index of 1·0 and more as positive and an index of less than 1·0 as negative. Moreover, a result was considered true positive in this study when two consecutive samples for that patient tested positive, including retesting of the first sample. This approach was shown to be very sensitive and specific compared with the gold standard (necropsy and culture from necropsy specimens). In Denning’s commentary there is the statement concerning the study of Maertens and colleagues, that “a positive result is defined as positive samples collected on two different days and having OD values of more than 1·0”. Denning has confused OD and OD index. The correct statement should be: . . . and having OD indices of 1·0 and more. This difference is clinically important. The OD of the threshold serum is in the range 0·3–0·8. This means, for example, that an OD of 0·9 can correspond to an OD index of more than 1·0—ie, an OD of 0·9 can represent a positive result.