In house method for rapid identification of fungi from fungus-positive bottles by MALDI-TOF mass spectrometry in patients with bloodstream infection

Abstract

Objective. To present the results of using in-house method for rapid identification of fungi from funguspositive bottles with routine conventional culture-based identification by matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) in patients with bloodstream infection. Materials and Methods. Prospective study was performed from 2016 to 2019 at the National Research Center for Hematology, Moscow. During the study period, all blood cultures (BC) bottles obtained from hematological patients were incubated in the BACTEC FX system (Becton Dickinson, USA). Positive BC bottles were examined by Gram stain. In house method was used after Gram stain was positive for yeast cells or hyphae. For that, BC media was transfer from fungus-positive bottles into tube. In house method included series section steps consisted from centrifugation and extraction of fungal proteins by adding of sodium dodecyl sulfate. Routine conventional culture-based identification on Sabouraud with chloramphenicol agar (bioMerieux, France) for yeasts and on Sabouraud dextrose agar (Oxoid, UK) for molds was used simultaneously with the in house method. Results. During the study period, 16 fungus-positive bottles were obtained from which were isolated in monoculture 14 (87.5%) Candida spp.: C. parapsilosis (n = 5), C. tropicalis (n = 4), C. albicans (n = 3), C. krusei (n = 1), C. guilliermondii (n = 1), one (6.3%) Rhodotorula mucilaginosa and one (6.3%) Fusarium dimerum. The in house method resulted in 75% (12⁄16) and 68.8% (11⁄16) identification rate at the genus and species level of fungi, respectively. The identification of fungi to species level was confirmed by conventional culture-based method in all cases. The median time from the start of vial incubation in BACTEC FX system to identification of fungi by in house method was less than conventional culture-based identification: 36 hrs 20 min vs 55 hrs 31 min (p = 0.028). Conclusions. A high rate of correct direct species identification and significant reduction in time to verification of fungi from fungus-positive bottles by in house method were obtained. The proposed in house method should be recommended for use in real microbiology practice to reduce the time for submitting results of identification to clinical units.