BackgroundAlthough the yeast Cyberlindnera fabianii (C. fabianii) has been rarely reported in human infections, nosocomial outbreaks caused by this organism have been documented. Here we report a pseudo-outbreak of C. fabianii in a urology department of a Chinese hospital over a two-week period.MethodsThree patients were admitted to the urology department of a tertiary teaching hospital in Beijing, China, from Nov to Dec 2018, for different medical intervention demands. During the period Nov 28 to Dec 5, funguria occurred in these three patients, and two of them had positive urine cultures multiple times. Sequencing of rDNA internal transcribed spacer (ITS) region and MALDI-TOF MS were applied for strain identification. Further, sequencing of rDNA non-transcribed spacer (NTS) region and whole genome sequencing approaches were used for outbreak investigation purpose.ResultsAll the cultured yeast strains were identified as C. fabianii by sequencing of ITS region, and were 100% identical to the C. fabianii type strain CBS 5640T. However, the MALDI-TOF MS system failed to correctly identify this yeast pathogen. Moreover, isolates from these three clustered cases shared 99.91%-100% identical NTS region sequences, which could not rule out the possibility of an outbreak. However, whole genome sequencing results revealed that only two of the C. fabianii cases were genetically-related with a pairwise SNP of 192 nt, whilst the third case had over 26,000 SNPs on its genome, suggesting a different origin. Furthermore, the genomes of the first three case strains were phylogenetically even more diverged when compared to a C. fabianii strain identified from another patient, who was admitted to a general surgical department of the same hospital 7 months later. One of the first three patients eventually passed away due to poor general conditions, one was asymptomatic, and other clinically improved.ConclusionIn conclusion, nosocomial outbreaks caused by emerging and uncommon fungal species are increasingly being reported, hence awareness must be raised. Genotyping with commonly used universal gene targets may have limited discriminatory power in tracing the sources of infection for these organisms, requiring use of whole genome sequencing to confirm outbreak events.