Staphylococcus pseudintermedius can be ...

4 downloads 0 Views 365KB Size Report
Greko C, Moodley A, Kania SA, Frank LA, Bemis DA, Franco A,. Iurescia M ... Weese JS, Fitzgerald JR, Rossano A, Guardabassi L (2010) Clonal spread of ...
Staphylococcus pseudintermedius can be misdiagnosed as Staphylococcus aureus in humans with dog bite wounds S. Börjesson, E. Gómez-Sanz, K. Ekström, C. Torres & U. Grönlund

European Journal of Clinical Microbiology & Infectious Diseases ISSN 0934-9723 Eur J Clin Microbiol Infect Dis DOI 10.1007/s10096-014-2300-y

1 23

Your article is protected by copyright and all rights are held exclusively by SpringerVerlag Berlin Heidelberg. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”.

1 23

Author's personal copy Eur J Clin Microbiol Infect Dis DOI 10.1007/s10096-014-2300-y

ARTICLE

Staphylococcus pseudintermedius can be misdiagnosed as Staphylococcus aureus in humans with dog bite wounds S. Börjesson & E. Gómez-Sanz & K. Ekström & C. Torres & U. Grönlund

Received: 7 July 2014 / Accepted: 7 December 2014 # Springer-Verlag Berlin Heidelberg 2014

Abstract The purpose of this study was to investigate whether S. pseudintermedius is misdiagnosed as S. aureus by clinical laboratories when isolated from humans with dog bite wounds. In addition, we attempted to determine whether S. pseudintermedius isolates related to dog bite wounds share phenotypic and genotypic traits. S. pseudintermedius was identified by PCR targeting the nuc gene. Isolates were tested for antibiotic susceptibility using VetMIC GP-mo microdilution panels. The occurrence of genes encoding leukocidins, exfoliatins, pyrogenic toxin superantigens and enterotoxins was determined by PCR. The relatedness of S. pseudintermedius isolates was investigated using Multi Locus Sequence Typing (MLST). Out of 101 isolates defined as S. aureus by human clinical microbiology laboratories, 13 isolates were re-identified as S. pseudintermedius and one isolate was confirmed to carry the mecA gene, i.e. methicillin-resistant (MRSP). The MRSP isolate was also defined as multi-resistant. Two methicillin-susceptible S. pseudintermedius isolates were also multi-resistant and five were susceptible to all antibiotics tested. With the exception of three S. pseudintermedius isolates belonging to multi locus sequence type (MLST) 158, all the isolates belonged to unique S. Börjesson (*) : K. Ekström : U. Grönlund Department of Animal Health and Antimicrobial Strategies, National Veterinary Institute (SVA), SE751 89 Uppsala, Sweden e-mail: [email protected] E. Gómez-Sanz : C. Torres Department of Biochemistry and Molecular Biology, University of La Rioja, Logroño, Spain E. Gómez-Sanz Environmental Genomics and Systems Biology Research Group, Institute of Natural Resource Sciences, Zurich University of Applied Sciences (ZHAW), Wädenswil, Switzerland U. Grönlund Swedish University of Agriculture Sciences, Uppsala, Sweden

STs. All isolates contained lukS/F-I, siet and se-int, and expA were identified in two isolates and expB and seccanine-sel in one isolate respectively. S. pseudintermedius is frequently misdiagnosed as S. aureus from humans with dog bite wounds showing that it can act as an opportunistic pathogen in humans. No common phenotypic and genotypic traits shared by the S. pseudintermedius isolates could be identified.

Introduction Staphylococcus pseudintermedius, previously typed as Staphylococcus intermedius, is a coagulase-positive staphylococci (CoPS) generally considered part of the normal bacterial flora of healthy dogs [1, 2]. Carriage has been shown to vary between 22 and 69 % in healthy dogs [3–6]. Besides being a commensal, it is also an opportunistic pathogen and a common cause of canine pyoderma [7]. Furthermore, it can cause abscesses, otitis externa, wound infections and urinary tract infections in dogs. S. pseudintermedius has also been isolated from and can cause infections in other animals [7–10]. Increasing reports of antibiotic-resistant S. pseudintermedius, especially the emergence and rapid spread of methicillinresistant S. pseudintermedius (MRSP), are therefore a cause for concern as this limits the treatment options [2, 11], though MRSP isolates are generally susceptible to antibiotics of last resort in human medicine, such as vancomycin, linezolid, and quinupristin/dalfopristin [2, 11, 12]. S. pseudintermedius is not typically considered a zoonotic bacterium, but there are several reports of infections in humans [13–19]. The degree of severity of human infections varies greatly, for example, S. pseudintermedius has been reported to cause otitis externa, bacteraemia, cardiomyopathy and endocarditis. In Sweden, cases of human MRSP infections were first described in 2011 and occurred as an outbreak involving four human patients at a university hospital [19].

Author's personal copy Eur J Clin Microbiol Infect Dis

This outbreak is of particular interest because the infections appear to be hospital acquired without any known dog–human contact. However, carriage of S. pseudintermedius in healthy humans seems to be quite uncommon, even in higher risk groups such as dog owners and veterinary staff [20, 21]. Due to the high phenotypic similarities between S. aureus and S. pseudintermedius, it may be possible that the occurrence of S. pseudintermedius as the causative agent in human infections is being underestimated. There are also examples of S. pseudintermedius being misdiagnosed as S. aureus by clinical microbiology laboratories [22, 23]. The aim of the present study was to investigate whether S. pseudintermedius is being misdiagnosed as S. aureus in humans with infected dog bite wounds at human clinical laboratories. All isolates of S. pseudintermedius were subjected to antibiotic-susceptibility testing, molecular characterisation and the occurrence of toxin-encoding genes.

Material and methods Bacterial isolates Human clinical microbiology laboratories in Sweden were invited to participate in the study and submit isolates defined as S. aureus from humans with infected dog bite wounds to the Section of Antibiotics, at the National Veterinary Institute (SVA), Sweden. Suspected methicillin-resistant S. aureus (MRSA) was not included in the study. Twenty-two laboratories, in seven Swedish counties, participated in and contributed to the study, with a total of 101 clinical isolates being defined as S. aureus during 2010 and 2011 to the SVA. All isolates were transported using COPAN Transystem® specimen swabs (Copan Diagnostics Inc., CA, USA). Upon arrival, isolates were seeded on horse blood agar plates (SVA, Uppsala, Sweden) and after overnight incubation they were stored at −70 ° C until further analysis. Phenotypic characterisation and susceptibility testing The 101 clinical isolates were investigated for haemolysis on cattle blood agar plates (SVA, Uppsala, Sweden), DNAase activity was tested using DNA-agar plates (SVA, Uppsala, Sweden) with the addition of 1 M HCl after overnight incubation, and maltose and trehalose fermentation using maltose and trehalose purple agar (SVA, Uppsala, Sweden). Susceptibility to the antibiotic classes aminoglycosides, β-lactams, chloramphenicol, fluoroquinolones, fusidic acid, lincosamides, macrolides, tetracyclines, and trimethoprim were determined by minimum inhibitory concentrations (MIC) using VetMIC GP-mo v.2 microdilution panels (SVA, Uppsala, Sweden). The breakpoints for the antibiotics used were those recommended in CLSI documents [24], except for

fusidic acid, where the European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoint for S. aureus was used. If resistant to three or more antibiotic classes, isolates were defined as multi-resistant. MRSA CCUG35601, S. aureus CCUG15915 and S. pseudintermedius CCUG49543 were used as controls. Molecular characterisation All isolates were subjected to conventional PCR for species confirmation using the nuc-gene and for occurrence of the mecA gene [25, 26], with MRSA CCUG35601, S. intermedius CCUG6520 and S. pseudintermedius CCUG49543 used as positive controls. All isolates identified as S. pseudintemerdius were subjected to spa-typing and multi locus sequence typing (MLST) according to the scheme http://pubmlst.org/ spseudintermedius/. S. pseudintermedius isolates were tested by PCR, as previously described, for the presence of the leukocidin gene lukS/F-I, and the exfoliative genes siet, expA and expB [27–30]. In addition, isolates were tested by PCR for the presence of sea, seb, sec, sed, see, seg, seh, sei, sej, sek, sel, sem, sen, seo, sep, seq, ser, and seu enterotoxin genes, as well as for se-int [27, 31]. The sec gene was further characterised for the sec canine variant (seccanine) [32] by sequencing and comparison with GenBank accession number U91526 [31].

Results Phenotypic characterisation and PCR species confirmation of human clinical isolates Of the 101 isolates identified as S. aureus in human clinical laboratories based on traditional phenotypic tests, primarily DNAse activity, 13 isolates (13 %) were described as S. pseudintermedius based on a multiplex PCR targeting nuc, 86 isolates (85 %) were S. aureus and two isolates (2 %) were defined as most likely being coagulase negative staphylococci. The S. pseudintermedius isolates were submitted by six different laboratories, with six of the isolates (AB410, AB561, AB563, AB564, AB764, and AB778) submitted by the same laboratory. One S. pseudintermedius isolate was also confirmed by PCR to carry the mecA gene, being MRSP. The 13S. pseudintermedius isolates all tested positive for DNAse, threhalose fermentation, α- and β-haemolysis on cattle blood agar plates and tested negative for maltose fermentation. As for the S. aureus isolates, one tested negative for DNAase, one tested negative for maltose fermentation and six tested negative for trehalose fermentation. The remaining isolates of S. aureus showed a large variation in haemolysis on

Author's personal copy Eur J Clin Microbiol Infect Dis

cattle blood agar plates: 21 had double zone haemolysis, four lacked haemolysis and the remaining 61 isolates had different variations of haemolysis. Susceptibility testing of S. pseudintermedius isolates

Discussion

Of the 13 human clinical isolates, six isolates were above or equal to the CLSI breakpoint for oxacillin (≥0.5 mg/L), five exhibited MIC of 0.5 mg/L and one with 2 mg/μL (the MRSP isolate), all being resistant to penicillin (Table 1). The MRSP isolate was also defined as being multidrug-resistant, because it was resistant to clindamycin, erythromycin, gentamicin, kanamycin, tetracycline and trimethoprim. Of the remaining isolates, two were multidrug-resistant, showing resistance to clindamycin, erythromycin, and kanamycin, while one was also resistant to fusidic acid. Five isolates were susceptible to all the antibiotics tested, three were tetracycline-resistant and two fusidic acid-resistant (Table 1). Molecular characterisation None of the S. pseudintermedius isolates were typeable using spa-typing, but all isolates were typeable by MLST. Eleven sequence types (ST) were identified among the 13 human clinical isolates, with three belonging to ST158 (Table 1). All S. pseudintermedius isolates contained the genes lukS/ F-I, siet and se-int (Table 1). The expA gene was identified in two (15 %) isolates, while expB occurred in one isolate (8 %). The seccanine +sel genes were detected in one isolate (8 %). Three isolates hailing from the same human clinical microbiology laboratory had identical MLST, and presence of the Table 1

same toxin genes and antibiotic susceptibility pattern (Table 1).

S. pseudintermedius can be misdiagnosed as S. aureus in humans with infected dog bite wounds, with 13 % of the clinical isolates defined by clinical laboratories as S. aureus actually being S. pseudintermedius. That S. pseudintermedius can be the causative agent in infected dog bites has also been described earlier [13]. It is likely that this misdiagnosis is due to the standard laboratory practices in human clinical laboratories relying mainly on phenotypic tests to identify S. aureus, and then primarily DNAse activity (personal communication). However, because S. pseudintermedius is a CoPS, it will also be positive in the DNAse test. Furthermore, S. pseudintermedius and S. aureus share many phenotypic traits [22], but can generally be separated through maltose fermentation under aerobic conditions on agar plates, where S. pseudintermedius is generally negative and S. aureus is generally positive. However, the maltose test is not completely accurate, as demonstrated in this study where one of the S. aureus isolates was negative in this test. To be able to differentiate between them more accurately phenotypically, the pyrrolidonyl arylamidase test can be applied [22]. The reason why human clinical laboratories mainly rely on the DNAse test to identify S. aureus is probably due to it being quick and low cost in terms of labour. Furthermore, S. aureus is considered to be part of normal human microbiota and the CoPS normally associated with human infection [33]. As shown in this study,

Characterisation of the 13S. pseudintermedius isolates from humans with infected dog bite wounds

Strain ID

MLST

Leukocidins

Exfoliatins

Enterotoxins

Antimicrobial resistance pattern

AB410 AB508 AB539 AB561a AB563a AB564a AB590 AB632

117 119 122 158 158 158 149 129

lukS/F-I lukS/F-I lukS/F-I lukS/F-I lukS/F-I lukS/F-I lukS/F-I lukS/F-I

siet, expA siet siet, expA siet siet siet siet, expB siet

se-int se-int, seccanine-sel se-int se-int se-int se-int se-int se-int

Cl, Em, Fu, Km Cl, Em, Km

AB633 AB747b AB764 AB778 AB795

191 118 192 190 145

lukS/F-I lukS/F-I lukS/F-I lukS/F-I lukS/F-I

siet siet siet siet siet

se-int se-int se-int se-int se-int

Tc Ox, Cl, Em, Gm, Km, Tc, Tm Tc

Em erythromycin, Cl clindamycin, Fu fusidic acid, Gm gentamicin, Km kanamycin, Tc tetracycline, Tm trimethoprim a

Submitted from the same clinical microbiology laboratory

b

Verified as MSRP using conventional PCR

Tc Fu

Fu

Author's personal copy Eur J Clin Microbiol Infect Dis

this may lead to S. pseudintermedius being classified as S. aureus. It can also lead to other CoPS such as S. delphini, S. hyicus, S. intermedius, and S. schleiferi also being misdiagnosed [22, 26]. Clinical microbiology laboratories should therefore not rely solely on phenotypic tests and should consider confirmation using molecular methods, at least in specific cases such as dog bites. Some clinical laboratories in Sweden are aware that S. pseudintermedius can be the causative factor for infection in dog-bite wounds and that additional testing of isolates should be performed in these cases (Bo Söderkvist, Örebro University Hospital, personal communication). It is likely that the number of misdiagnosed cases will fall in future and that more human clinical cases of S. pseudintermedius will be identified due to the growing use of MALDI-TOF mass spectrometry for species identification. If this is the case and based on the results of this study, there may be a need to re-evaluate the zoonotic potential of S. pseudintermedius and other CoPS. The misidentification of S. pseudintermedius as S. aureus i s a ca us e f or c once r n bec aus e S . aureus a nd S. pseudintermedius can have different MIC values and interpretation criteria for disc diffusion antibiotic susceptibility testing, which can lead to non-accurate antibiograms. The most significant example is the determination of methicillin resistance, where interpretation criteria and MICs defined for S. aureus can lead to MRSP isolates being defined as susceptible. Cefoxitin is generally used as an indicator to identify MRSA, but for S. pseudintermedius cefoxitin is not predictive and oxacillin is usually recommended instead [24]. The European Committee on Antimicrobial Susceptibility Testing (EUCAST) recently suggested a breakpoint of