LETTER
Candida auris, an emerging multidrug-resistant yeast associated with a high mortality rate, has been increasingly reported outside the United States to cause outbreaks in hospital settings (1). Although this organism is rare in the United States, its prevalence may be underestimated because of unreliable identification (2–4). The CDC has recently recommended that health care facilities place patients with C. auris colonization or infection in single rooms (2). Therefore, it is imperative for clinical microbiology laboratories to accurately identify this organism to aid in preventing health care-associated outbreaks.
Since a majority of U.S. clinical microbiology laboratories do not have experience in identifying C. auris, the CDC has recently prepared a panel of C. auris and related species to assist clinical microbiology laboratories with implementing and validating methods to identify this organism. The panel includes 10 isolates of C. auris, 3 of C. duobushaemulonii, 2 of C. haemulonii, 2 of Saccharomyces, and 1 each of Kodamaea ohmeri, Candida krusei, and C. lusitaniae. We tested C. auris isolates, as well as phylogenetically closely related C. haemulonii and C. duobushaemulonii isolates, on four commercial biochemical identification platforms and two matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) systems (Tables 1 and 2). The identities of the 15 isolates tested were all confirmed correctly to the species level by sequencing of the internal transcribed spacer and D1/D2 regions.
TABLE 1.
Biochemical identification
| Isolate no. | Species tested | Identification according to: |
|||
|---|---|---|---|---|---|
| API 20C AUXa | BD Phoenixb | Vitek-2c | MicroScand | ||
| 1 | C. auris | R. glutinis | C. catenulata | C. haemulonii | C. famata |
| 2 | C. auris | R. glutinis | C. haemulonii | C. haemulonii | C. famata |
| 3 | C. auris | R. glutinis | C. haemulonii | C. haemulonii | C. famata |
| 4 | C. auris | R. glutinis | C. haemulonii | C. haemulonii | C. lusitaniae |
| 5 | C. auris | R. glutinis | C. haemulonii | C. haemulonii | C. guilliermondii |
| 6 | C. auris | R. glutinis | C. haemulonii | C. haemulonii | C. famata |
| 7 | C. auris | R. glutinis | C. haemulonii | C. haemulonii | C. guilliermondii |
| 8 | C. auris | R. glutinis | C. haemulonii | C. haemulonii | C. parapsilosis |
| 9 | C. auris | R. glutinis | C. haemulonii | C. haemulonii | C. guilliermondii |
| 10 | C. auris | R. glutinis | C. haemulonii | C. haemulonii | C. guilliermondii |
| 11 | C. duobushaemulonii | R. glutinis | C. parapsilosis | C. haemulonii | C. guilliermondii |
| 12 | C. duobushaemulonii | R. glutinis | C. parapsilosis | C. haemulonii | C. guilliermondii |
| 13 | C. haemulonii | R. glutinis | C. haemulonii | C. haemulonii/K. ohmeri | C. catenulata |
| 14 | C. duobushaemulonii | R. glutinis | C. parapsilosis | C. haemulonii | C. parapsilosis |
| 15 | C. haemulonii | R. glutinis | None | C. haemulonii/K. ohmeri | C. parapsilosis |
Identification at 48 and 72 h of incubation; API 20C AUX does not have C. auris, C. haemulonii, or C.duobushaemulonii in its library.
C. haemulonii is in the BD Phoenix library, but C. auris and C. duobushaemulonii are not.
The Vitek-2 library has C. haemulonii but not C. auris or C. duobushaemulonii.
MicroScan does not have C. auris, C. haemulonii, or C. duobushaemulonii in its library.
TABLE 2.
Identification by MALDI-TOF MS
| Isolate no. | Species tested | Bruker Biotyper |
Vitek-MS |
|||
|---|---|---|---|---|---|---|
| FDA librarya direct on-plate extraction (score) | RUOb library |
Direct extraction (% identity) |
||||
| Direct on-plate extraction (score) | Full-tube extraction (score) | IVDc library | RUO libraryd | |||
| 1 | C. auris | No IDe | C. auris (1.75) | C. auris (2.19) | No ID | C. auris (99) |
| 2 | C. auris | No ID | No ID | C. auris (2.02) | No ID | C. auris (98) |
| 3 | C. auris | No ID | C. auris (1.80) | C. auris (2.04) | No ID | C. auris (99) |
| 4 | C. auris | No ID | C. auris (1.73) | C. auris (2.10) | No ID | C. auris (99) |
| 5 | C. auris | No ID | No ID | C. auris (1.95) | No ID | C. auris (99) |
| 6 | C. auris | No ID | No ID | C. auris (1.97) | No ID | C. auris (87) |
| 7 | C. auris | No ID | No ID | C. auris (2.05) | No ID | C. auris (99) |
| 8 | C. auris | No ID | C. auris (1.73) | C. auris (1.99) | No ID | C. auris (99) |
| 9 | C. auris | No ID | No ID | C. auris (2.00) | No ID | C. auris (92) |
| 10 | C. auris | No ID | C. auris (1.75) | C. auris (1.92) | No ID | C. auris (99) |
| 11 | C. duobushaemulonii | C. duobushaemulonii (2.29) | C. duobushaemulonii (2.29) | Not tested | No ID | No ID |
| 12 | C. duobushaemulonii | C. duobushaemulonii (2.19) | C. duobushaemulonii (2.19) | Not tested | No ID | No ID |
| 13 | C. haemulonii | C. haemulonii (2.29) | C. haemulonii (2.29) | Not tested | C. haemulonii (99) | C. haemulonii (99) |
| 14 | C. duobushaemulonii | C. duobushaemulonii (2.27) | C. duobushaemulonii (2.27) | Not tested | No ID | C. duobushaemulonii (94) |
| 15 | C. haemulonii | C. haemulonii (2.21) | C. haemulonii (2.21) | Not tested | C. haemulonii (99) | C. haemulonii (99) |
The Bruker FDA library contains C. haemulonii (12 entries) and C. duobushaemulonii (7 entries).
The Bruker RUO library has C. auris (three entries).
IVD, in vitro diagnostic. The Vitek MS IVD library is an FDA-approved library, and it contains C. haemulonii.
The Vitek MS RUO library has C. auris and C. duobushaemulonii.
ID, identification.
All C. auris isolates were misidentified as Rhodotorula glutinis by API 20C AUX (bioMérieux, Marcy l'Etoile, France), as C. haemulonii (except one as C. catenulata) by BD Phoenix (BD Diagnostics, Sparks, MD), as C. haemulonii by Vitek-2 (bioMérieux), and as C. famata, C. lusitaniae, C. guilliermondii, or C. parapsilosis by MicroScan (Beckman Coulter, Pasadena, CA) (Table 1). Because of the lack of C. auris entries in the FDA-approved libraries, it was unidentified by both the Bruker Biotyper (Bruker, Billerica, MA) and Vitek-MS (bioMérieux) MALDI-TOF MS systems when queried on the FDA-approved libraries (Table 2). Incorporation of a research-use-only (RUO) library containing C. auris rendered correct identification of this organism by both MALDI-TOF MS systems. In the Vitek MS system, all were identified correctly by the direct extraction method. However, it requires the full-tube extraction method for the Bruker MS system because the direct on-plate extraction method resulted in low-score matches for 50% of the C. auris isolates and left another 50% unidentified. Although it was not tested in our study, a partial extraction method that is less laborious than full-tube extraction has been reported to achieve excellent identification by the Bruker MS system (5, 6).
In summary, C. auris cannot be reliably identified by standard biochemical identification platforms/kits primarily because of a lack of the organism in their databases. Identification of Rhodotorula glutinis by API 20C AUX should trigger further investigation if colonies on culture plates are not pink with a negative urease reaction. Likewise, identification of C. haemulonii by BD Phoenix and Vitek-2 requires further testing by DNA sequencing to rule out C. auris. MicroScan users may find this especially challenging, since C. auris was misidentified as several different Candida species. Particular attention needs to be paid to C. famata, since it has microscopic features very similar to those of C. auris (e.g., no pseudohypha production). While C. auris is not identified by both the Bruker and Vitek MALDI-TOF MS systems with FDA-approved libraries, it can be reliably identified by both MALDI-TOF MS systems with an RUO library with C. auris entries (Table 2). Importantly, however, Bruker MALDI-TOF MS users must be aware that C. auris could potentially be unidentified if the direct on-plate extraction method is used; thus, the full-tube extraction method or possibly a partial extraction method should be applied for reliable identification.
REFERENCES
- 1.Pan American Health Organization/World Health Organization. 3 October 2016. Epidemiological alert: Candida auris outbreaks in health care services. Regional Office for the Americas of the World Health Organization, Washington, DC: http://www.paho.org/hq/index.php?option=com_docman&task=doc_view&Itemid=270&gid=36354&lang=en. [Google Scholar]
- 2.Centers for Disease Control and Prevention. 2016. Clinical alert to U.S. healthcare facilities—June 2016: global emergence of invasive infections caused by the multidrug-resistant yeast Candida auris. Centers for Disease Control and Prevention, Atlanta, GA: http://www.cdc.gov/fungal/diseases/candidiasis/candida-auris-alert.html. [Google Scholar]
- 3.Chowdhary A, Sharma C, Duggal S, Agarwal K, Prakash A, Singh PK, Jain S, Kathuria S, Randhawa HS, Hagen F, Meis JF. 2013. New clonal strain of Candida auris, Delhi, India. Emerg Infect Dis 19:1670–1673. doi: 10.3201/eid1910.130393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kathuria S, Singh PK, Sharma C, Prakash A, Masih A, Kumar A, Meis JF, Chowdhary A. 2015. Multidrug-resistant Candida auris misidentified as Candida haemulonii: characterization by matrix-assisted laser desorption ionization–time of flight mass spectrometry and DNA sequencing and its antifungal susceptibility profile variability by Vitek 2, CLSI broth microdilution, and Etest method. J Clin Microbiol 53:1823–1830. doi: 10.1128/JCM.00367-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Fraser M, Brown Z, Houldsworth M, Borman AM, Johnson EM. 2016. Rapid identification of 6328 isolates of pathogenic yeasts using MALDI-ToF MS and a simplified, rapid extraction procedure that is compatible with the Bruker Biotyper platform and database. Med Mycol 54:80–88. doi: 10.1093/mmy/myv085. [DOI] [PubMed] [Google Scholar]
- 6.Borman AM, Szekely A, Johnson EM. 2016. Comparative pathogenicity of United Kingdom isolates of the emerging pathogen Candida auris and other key pathogenic Candida species. mSphere 1(4):e00189-16. doi: 10.1128/mSphere.00189-16. [DOI] [PMC free article] [PubMed] [Google Scholar]