To the Editor—Linezolid resistance is increasingly described among the 3 most prevalent clones of linezolid-resistant Staphylococcus epidermidis (LRSE) that are occasionally involved in large outbreaks: sequence type 2 (ST2), ST5, and ST22.Reference Dortet, Glaser and Kassis-Chikhani 1 The resistant phenotype has been related to the occurrence of mutations in genes coding for the V domain of 23SrRNA and ribosomal proteins L3/L4 or cfr acquisition.Reference Dortet, Glaser and Kassis-Chikhani 1 The LRSE-ST22 isolates have been associated with infection and colonization cases in Spain, France, Germany, and, particularly, Greece, and most of these LRSE-ST22 arise after treatment with linezolid.Reference Dortet, Glaser and Kassis-Chikhani 1 – Reference Lozano, Ruiz-García and Gómez-Sanz 7 In this study, we aimed to characterize the first methicillin- and linezolid-resistant S. epidermidis from a patient without previous linezolid exposure in Turkey and to assess its genetic similarity to the close geographical S. epidermidis from Greece.
In October 2016, a hypertensive 70-year-old male attended the emergency service at a hospital in Rize, Turkey, and was hospitalized with syncope and poor general condition (day 1). He was hospitalized in the neurology ward, where his symptoms deteriorated. These symptoms included fever, stiff neck, and confusion, and a diagnosis of clinical meningitis was established. A cerebrospinal fluid sample was collected, which was negative on cultural and microscopic analyses. Antibiotic therapy on day 2 included ceftriaxone, netilmicin, and vancomycin. The patient’s condition deteriorated, and he was transferred to intensive care with room and contact isolation. Multidrug-resistant (MDR) S. epidermidis exhibiting resistance to oxacillin and linezolid was identified in 2 blood samples collected on days 14 and 15. The patient died on day 16 from multiple organ failure. Only a previous hospitalization for blood pressure control was registered 2 years before in the same hospital. Linezolid was never given to the patient, and additional linezolid-resistant gram-positive isolates were not detected before this case or until December 2017.
An LRSE isolate was sent to our laboratory for further characterization. The susceptibility to linezolid and vancomycin was confirmed by broth microdilution, to daptomycin and tedizolid by Etest, and to other 12 antibiotics by disk diffusion. 8 Using PCR and type sequencing, we searched cfr, cfr(B), optrA, mecA, mecC genes, mutations in the 23S-rRNA-V-domain, and genes coding for L3/L4/L22 ribosomal proteins. Clonality was evaluated using pulsed-field gel electrophoresis (PFGE) and multilocus type sequencing (MLST; www.pubmlst.org). Antibiotic resistance stability (linezolid/tedizolid) was assessed after 100 daily passages in antibiotic-free Mueller-Hinton agar. Linezolid dependence was evaluated because it is a possible factor contributing to the emergence of ST22-S. epidermidis in Greece.Reference Karavasilis, Zarkotou and Panopoulou 5 The Turkish LRSE and 2 LRSE-ST22 isolates from a high number of patients under linezolid therapy in 2 Greek regions (Patris and Athens) during 2008–2012Reference Papadimitriou-Olivgeris, Giormezis and Fligou 2 , Reference Liakopoulos, Spiliopoulou and Damani 3 were compared by performing additional experiments not available in those studies: PFGE, ribosomal protein mutations, minimum inhibitory concentration (MIC)-tedizolid.
The Turkish LRSE-ST22 expressed resistance to linezolid (MIC ≥256 mg/L), tedizolid (MIC >32 mg/L), vancomycin (MIC = 4 mg/L), cefoxitin (mecA), and 8 other antibiotics. Linezolid resistance was related to T2504A and C2534T mutations in the 23S-rRNA-V domain and to the amino acid changes L94V, G152D, D159Y in L3 and N158S in L4 proteins (S. epidermidis RP62A numbering). The cfr, cfr(B) and optrA genes were not detected. The high linezolid and tedizolid MIC values were stably maintained after 100 serial passages, suggesting the absence of a biological burden linked to the identified mutations in nonselective contexts. Linezolid dependence was not observed in the conditions tested (Figure S1), suggesting a variable phenotype potentially dependent on previous linezolid exposure, as has been described for some strains.Reference Karavasilis, Zarkotou and Panopoulou 5 The 3 Greek and Turkish isolates presented the same ribosomal mutations, MIC values for linezolid and tedizolid, and the same pulsotype A (Figure S2; Table 1).Reference Ikonomidis, Grapsa, Pavlioglou, Demiri, Batarli and Panopoulou 4
NOTE. BL, blood; CT, catheter; I, infection; C, colonization; MIC, minimum inhibitory concentration; MRSE, methicillin-resistant Staphylococcus epidermidis; Ref, reference. Antibiotics: CIP, ciprofloxacin; CLI, clindamycin; CLO, chloramphenicol; DAP, daptomycin; ERY, erythromycin; FUS, fusidic acid; GEN, gentamicin; KAN, kanamycin; LIN, linezolid; RIF, rifampicin; SXT, cotrimoxazol; TEC, teicoplanin; TED, tedizolid; TET, tetracycline; TIG, tigecycline; TOB, tobramycin; VAN, vancomycin; UK, unknown.
a Numeration according to S. epidermidis RP62A (GenBank no. CP000029.1).
b Mutations among ribosomal protein L22 were not detected.
c Previous linezolid exposure occurred in some patients.
d Results of published strains2,3 obtained in this study.
e PFGE pulsotype “A” (E Petinaki, MD, PhD, written personnel communication, December 2017) was identified in all ST22 isolates (n=12) from reference 3.
This study reports the dissemination of an LRSE-ST22 strain in the Greek-Turkish region at least since 2008. The Turkish LRSE-ST22 is one of the few LRSE infections described without previous linezolid exposure nor related to a hospital outbreak, which could have been facilitated by the stability of both oxazolidinones resistance, allowing its persistence in linezolid nonselective contexts. Linezolid was scarcely used in this Turkish hospital due to the low rate of vancomycin-resistant enterococci (6%) or the absence of vancomycin-resistant Staphylococcus spp. The origin of this strain (eg, transfer from this or other patient’s microbiota, healthcare staff, or environment) remains to be elucidated, but the hypothesis of LRSE-ST22 transmission from the community cannot be ruled out, similarly to methicillin-resistant Staphylococcus aureus. Notably, the limited remaining last-resort therapeutic options (vancomycin and daptomycin) both require parenteral administration (in contrast to oxazolidinones) for the treatment of infections caused by these and other LRSE-ST22 strains.Reference Dortet, Glaser and Kassis-Chikhani 1 Moreover, the high MIC value for vancomycin in the Turkish case (4 mg/L) and the coresistance to teicoplanin observed in other recent LRSE-ST22 strainsReference Ikonomidis, Grapsa, Pavlioglou, Demiri, Batarli and Panopoulou 4 suggests the possibility of poor patient outcomes when glycopeptides are used.
This study demonstrates that identical MDR LRSE-ST22 strains expressing resistance to last-resort antibiotics, including linezolid and tedizolid, are spreading among hospitals in different countries. Additional studies including community sources are crucial to better understanding the factors driving the emergence and transmission routes of these strains and to timely optimization of antimicrobial stewardship.
We would like to thank Dr Iris Spiliopoulou, MD, PhD (School of Medicine, University of Patras; National Reference Laboratory for Staphylococcus, Patras) and Dr Eythymia Petinaki, MD, PhD (Department of Microbiology, University Hospital of Larissa, Larissa) for the LRSE-ST22 strains from Greece. We also thank to Jorge Oliveira, PhD (Faculty of Pharmacy, University of Porto) for his statistical analysis.
Financial support: This work was supported by Fundação para a Ciência e a Tecnologia (FCT)/Ministério da Educação e Ciência (MEC) through national funds and was co-financed by Fundo Europeu de Desenvolvimento Regional (FEDER) under the Partnership Agreement PT2020 (grant no. UID/MULTI/04378/2013–POCI/01/0145/FEDER/007728). A.R.F. was supported by a postdoctoral fellowship from FCT (grant no. SFRH/BPD/96148/2013) through Programa Operacional Capital Humano (POCH).
Potential conflicts of interest: All authors report no conflicts of interest relevant to this article.
To view supplementary material for this article, please visit https://doi.org/10.1017/ice.2018.5