PCR-Based Microarray Enhances Diagnosis of Culture-Negative Biopsied Tissue in Patients with Invasive Mold Infections: Real-World Experience in a Tertiary Medical Center.

A fungal polymerase chain reaction (PCR) amplifies conserved genes across diverse species, combined with the subsequent hybridization of amplicons using a specific oligonucleotide microarray, allowing for the rapid detection of pathogens at the species level. However, the performance of microarrays in diagnosing invasive mold infections (IMI) from infected tissue samples is rarely reported. During the 4-year study period, all biopsied tissue samples from patients with a suspected IMI sent for microarray assays were analyzed. A partial segment of the internal transcribed spacer (ITS) region was amplified by nested PCR after DNA extraction. Amplicons were hybridized with specific probes for a variety of mold species using an in-house oligonucleotide microarray. A total of 80 clinical samples from 74 patients were tested. A diagnosis of an IMI was made in 10 patients (4 proven, 1 probable, 3 possible, 2 clinical suspicion). The PCR/microarray test was positive for three out of four proven IMIs, one probable IMI, and one out of three possible IMIs. Two patients with positive PCR/microarray findings were considered to have clinical suspicion of an IMI, and their responsible physicians initiated antifungal therapy despite the absence of supporting microbiological and histological evidence. Clinical diagnoses were categorized into non-IMI and IMI groups (including proven, probable, possible, and clinical suspicion). The sensitivity and specificity of the microarray in diagnosing the IMIs were 70% and 95.7%, respectively, while the sensitivity and specificity of the culture and histological findings were 10%/96.3% and 40.0%/100%, respectively. PCR-based methods provide supportive microbiological evidence when culture results are inconclusive. The combination of a microarray with fungal culture and histology promotes the precise diagnosis of IMIs in difficult-to-diagnose patients.