1. ¹Section of Experimental Haematology, Division of Cancer Science and Molecular Pathology, University of Glasgow, UK
2. ²Department of Haematology, Lothian University Hospitals NHS Trust, Edinburgh, UK
3. ³Department of Haematology, North Glasgow University Hospitals NHS Trust, Glasgow, UK
4. ⁴Department of Medical Microbiology, North Glasgow University Hospitals NHS Trust, Glasgow, UK

Correspondence: Dr BL Jones, Department of Medical Microbiology, Lister Building, Glasgow Royal Infirmary, Castle Street, Glasgow, G4 OSF, UK. E-mail: b.l.jones@clinmed.gla.ac.uk

Received 1 June 2004; Accepted 13 September 2004; Published online 10 January 2005.

Abstract

A blinded prospective study was performed to determine whether screening of whole blood using a real-time, panfungal polymerase chain reaction (PCR) technique could predict the development of invasive fungal infection (IFI) in immunocompromised haemato-oncology patients. In all, 78 patients (125 treatment episodes) were screened twice weekly by real-time panfungal PCR using LightCycler™ technology. IFI was documented in 19 treatment episodes (five proven, three probable and 11 possible), and in 12, PCR was sequentially positive. PCR positivity occurred in: 4/5 proven; 2/3 probable; 6/11 possible; and 29/106 with no IFI. In 8/12 with IFI and sequentially positive PCR results, PCR positivity occurred before (median 19.5 days) and in 4/12 (median 10.5 days) after the initiation of empirical antifungal therapy. Based on sequential positive results for proven/probable IFI sensitivity, specificity, positive predictive value and negative predictive value were 75, 70, 15 and 98%, respectively. Real-time panfungal PCR is a sensitive tool for the early diagnosis of IFI in immunocompromised haemato-oncology patients. It may be most useful as a screening method in high-risk patients, either to direct early pre-emptive antifungal therapy or to determine when empirical antifungal therapy can be withheld in patients with antibiotic--resistant neutropenic fever. However, these strategies require further assessment in comparative clinical trials.