Objectives: Invasive fungal infection (IFI) is a major cause of morbidity and mortality in severely immunocompromised patients and is difficult to diagnose. The significance of molecular methods for diagnosis of IFI is still controversial. In a subset of patients treated within the AmBiLoad Trial, samples were investigated prospectively by a nested Aspergillus PCR assay to re-evaluate the significance of PCR in this setting.
Patients and methods: In the randomized, prospective multicenter AmBiLoad trial, patients with proven or probable IFI were randomized to receive liposomal amphotericin B (L-AMB) 3 or 10 mg/kg QD for 14 d followed by L-AMB 3 mg/kg QD. From 91 patients, 459 serial samples (98% blood samples) were investigated by a nested PCR assay for Aspergillus DNA. All samples were investigated in our laboratory with a previously described nested and a quantitative PCR assay. As required by the study protocol, serial Aspergillus antigen galactomannan was performed. IFI was defined according to modified EORTC/MSG 2002 criteria as applied in the AmBiLoad trial.
Results: Seven and 52 patients had proven and probable IFI according to modified EORTC/MSG criteria, respectively. The median number of samples investigated per patient was 4. Seventy percent of samples were obtained during treatment with antifungal study medication. Forty-three samples gave positive PCR results. Patients with an unfavorable outcome had a significantly higher rate of positive PCR results (48% versus 21%).
Conclusions: The sensitivity of Aspergillus PCR testing is limited during antifungal therapy. The tendency for persistently positive PCR results to indicate a poor prognosis has to be confirmed in further studies.