Staphylococcus argenteus in the Netherlands: not a new arrival

Methicillin-resistant Staphylococcus aureus (MRSA) can be responsible for severe infections and outbreaks. In the Netherlands, an active search and destroy policy that includes submitting MRSA isolates to the national MRSA reference center for molecular characterization has kept incidence of MRSA low (2017: 1.2%)1,2. Typing of S. aureus is performed using multiple-locus variable number tandem repeat analysis (MLVA) and approximately 4,000 MRSA strains are tested annually.Recently, Staphylococcus argenteus, a new clinically relevant and globally spreading species has emerged. Identification is difficult, since both phenotypical and molecular assays misidentify S. argenteus as S. aureus. The ESCMID study group for Staphylococci and Staphylococcal diseases (ESGS) recommends this species, part of the Staphylococcus aureus complex, to be handled as (methicillin-resistant) S. aureus.3Besides a single case report,4 little is known about the epidemiology of S. argenteus in the Netherlands. Here, we present the second and third case of methicillin-resistant S. argenteus (MRSArg) from demographically distinct regions in the Netherlands and demonstrate the presence of MRSArg in our country as early as 2008 based on data from the Dutch national MRSA surveillance.The first case concerns a 31-year-old repatriated patient, originally hospitalized in Australia in 2017. Screening for MRSA carriage resulted in positive cultures of nose, throat, rectum and a foot wound. Identification using VITEK MALDI-TOF (bioMérieux, Marcy-l'Étoile, France) repeatedly showed S. aureus with 99.9% confidence, but MLVA showed that the isolate was non-typable (MLVA-NT). Molecular identification with in-house multiplex PCR 5 yielded a positive mecA, a delayed femA and no SA442 signal. Additional BDmax PCR (BD, Franklin Lakes, New Jersey, USA) confirmed MRSA. The second case concerns a frequent visitor of the Philippines who was screened for MRSA in July 2018 as follow-up on initial MRSA finding (May 2018, MLVA-Complex 45). MRSA was detected in a throat swab using a mecA/mecC and SA442 real-time PCR assay.6 Phenotypical testing with coagulase, DNase, and MALDI-TOF (Bruker, Billerica, Massachusetts, USA) confirmed S. aureus. PCR revealed the presence of the mecA gene, but the SA442 fragment produced an aberrant low signal, while the GeneXpert MRSA assay (Cepheid, Sunnyvale, California, USA) identified the strain as MRSA. This strain was submitted for MLVA, but was MLVA-NT. The isolates from both cases were subjected to WGS. Blasting the five largest contigs (> 100.000 bp) to the NCBI database revealed a >99% sequence identity with the available S. argenteus complete chromosomes (AP018562, CP023076 and CP025023) for all five contigs. Since both S. argenteus strains were MLVA-NT, we retrospectively interrogated the Dutch MRSA surveillance system for other MRSArg isolates amongst all MLVA-NT strains available from 2008 - May 2019 (n=49). Re-identification was done for all MLVA-NT MRSA strains by Bruker MALDI-TOF MS using an updated database (December 2018) that currently contains 4 spectra of S. argenteus. In 37 of the 49 MLVA-NT strains, MALDI-TOF identified 36 S. argenteus and 1 S. aureus as the 2 top hits of both duplicates with a score of >2.0, thereby providing reliable identification. A S. argenteus specific real-time PCR correctly confirmed the MALDI-TOF results (table 1). For 12 strains, identification with MALDI-TOF was not conclusive according to the manufacturers guidelines, since the difference in MALDI score was not >0.3 between the first two organisms reported per identification. Next to S. argenteus, Staphylococcus schweitzeri (n=6), S. aureus (n=5) or both (n=1) were found as top hit in at least one of the two duplicates (table 1). However, since only four spectra of S. argenteus are included in the database and this species is closely related to both S. aureus and S. schweitzeri, it is not surprising that differentiation of these species included in the S. aureus complex using MALDI-TOF is challenging. However, the S. argenteus-specific PCR tested positive for all 12 isolates indicating that these were indeed S. argenteus, resulting in 48 S. argenteus strains found among the 49 MLVA-NT strains (table 1). The first S. argenteus in the Dutch MRSA surveillance dated from 2008 showing that S. argenteus has been present in the Dutch patient population for over a decade. However, 41 of the 48 S. argenteus cases were submitted after 2013. Our cases showed difficulties in detecting S. argenteus and discrimination from S. aureus. Using an updated version of the MALDI-TOF database (Bruker only) and a specific real-time PCR assay, we were able to show that MRSArg has been found in the Netherlands as early as 2008 and has been misidentified as MRSA for many years. We support the proposal of the ESGS of reporting S. argenteus as S. aureus complex to clinicians,3 but underline the need for laboratories to distinguish between the S. aureus and S. argenteus in order to generate more data on the prevalence, epidemiology and the pathogenicity of S. argenteus.7,8 By using the updated MALDI-TOF database and/or the specific real-time PCR, this should be possible for the vast majority of laboratories.