Oxacilin-resistant Coagulase-negative staphylococci (CoNS) bacteremia in a general hospital at São Paulo city, Brasil.

In the last decades, coagulase-negative staphylococci (CoNS), especially Staphylococcus epidermidis have become an important cause of bloodstream infections. In addition, rates of methicillin-resistance among CoNS have increased substantially, leading to the use of glicopeptides for therapy. The objective of this study was to evaluate eleven consecutives clinically relevant cases of oxacillin-resistant CoNS bacteremia in a general hospital localized in São Paulo city, Brazil. Five different species were identified by different phenotypic methods, including S. epidermidis (5), S. haemolyticus (3), S. hominis (1), S. warneri (1) and S. cohnii subsp urealyticus (1). A variety of Pulsed Field Gel Electrophoresis profiles was observed by macrorestriction DNA analysis in S. epidermidis isolates, but two of three S. haemolyticus isolates presented the same profile. These data indicated the heterogeneity of the CoNS isolates, suggesting that horizontal dissemination of these microorganisms in the investigated hospital was not frequent. One S. epidermidis and one S. haemolyticus isolates were resistant to teicoplanin and susceptible to vancomycin. The selective pressure due to the use of teicoplanin in this hospital is relevant.

INTRODUCTION

Coagulase-negative staphylococci (CoNS) are major causes of nosocomial bloodstream infection and responsible for high morbidity and mortality rates, mainly in hospitalized patients (9). Members of the genera Staphylococcus are catalase-positive, gram-positive cocci, coagulase-negative, aerobes and, when present in human infections, can present multiresistant profiles (8,20). These strains may constitute a dangerous reservoir of resistance genes in a hospital (20).

Staphylococci generally present a benign or symbiotic relationship with their host. However, they may become pathogens when entering the host tissue through break of the cutaneous barrier, inoculation by needles or implantation of medical devices (5). It is increasingly important to accurately identify CoNS isolates to the species level in order to determine the clinical significance of these bacteria, the proper epidemiological surveillance, and the management of patients infected with CoNS in case of relapse (15).

A substantial increase in the frequency of oxacillin-resistance (methicillin-resistant) in CoNS isolates has occurred over the last decades (4). Between 50% and 80%, depending on the species, are mec A positive or oxacillin resistant (1,6).

According to the results of the SENTRY antimicrobial surveillance program, carried out with Brazilian bloodstream isolates over a five-year period from 1997 to 2001, the oxacillin susceptibility in Staphylococcus aureus was 68.2% and 19.2% in CoNS (17).

Staphylococcus epidermidis and Staphylococcus haemolyticus are the most frequent species in nosocomial infections, and the frequency of oxacillin resistance is higher in CoNS clinical isolates (3). S. haemolyticus have been reported to show multiple resistance to antimicrobials and quite frequently clinical isolates present with reduced susceptibility or are resistant to teicoplanin (16).

Vancomycin is usually considered the treatment of choice for infections caused by these microorganisms. However, due to the emergence of vancomycin-resistant enterococci (9) and vancomycin-resistant staphylococci (19), reduction in the use of this drug has been recommended (20). A few reports have shown that the mechanism of glycopeptide resistance in S. epidermidis, S. haemolyticus and S. hominis is similar to that described in VISA and hetero-VISA strains (13). The objective of this study was to evaluate eleven consecutive clinically relevant cases of oxacillin-resistant CoNS bacteremia in a general hospital where therapy with the glycopeptide teicoplanin is broadly utilized.

MATERIALS AND METHODS

Bacterial isolates

The study was carried out with eleven consecutive bloodstream CoNS isolates, obtained between June and July 2005 from patients at 9 de Julho Hospital, a 250 beds general hospital localized in the city of São Paulo, Brazil. According to the CDC criteria, these isolates were considered clinically relevant by the National Nosocomial Infections Surveillance Committee of the hospital (12).

Identification

Staphylococci identification was carried out by test for oxidation-fermentation, coagulase (Laborclin, Brazil), catalase, alkaline phosphatase (Sigma-Aldrich, Germany), ornithine (Merck, Germany), urease (Oxoid, UK), PYR (pyrrolidinyl-β-naphthylamide hydrolysis, Probac do Brasil, Brazil), hemolysis in sheep blood agar, acid production from trehalose (Sigma-Aldrich, Germany), mannitol (Nuclear, Brazil), mannose (Vetec, Brazil), sucrose (Reagen, Brazil), maltose (Sigma-Aldrich, Germany), lactose (Difco , USA), cellobiose (Sigma-Aldrich, Germany) and anaerobic growth in thioglicolate (Merck, Germany). Susceptibility to novobiocin (Oxoid, UK), polymyxin B (Oxoid, UK), bacitracin (CECON, Brazil), desferrioxamine (Ciba Geigy, Switzerland) and fosfomycin (Oxoid,UK) was also determined. Isolates were kept frozen at -20°C in Skim Milk (Difco, USA). Bacteria to be tested were suspended in 0,5 ml of saline to a McFarland standard and 50ml were added to each sugar carbohydrate tube. The acid production from carbohydrates was evaluated after 24, 48 and 72hs of incubation at 35–37°C. The final evaluation was at the 7th day. The phenotypic tests were accomplished in parallel with a positive control (S. epidermidis ATCC 12228).

Antimicrobial susceptibility

The isolates were tested for susceptibility to oxacillin, vancomycin and teicoplanin by the agar disk diffusion method with Mueller-Hinton agar plates (Difco, USA) according to Clinical Laboratory and Standards Institute (CLSI 2005) recommendations and confirmed by Etest (AB Biodisk, Sweden). The susceptibility to novobiocin, polymyxin B, bacitracin, desferrioxamine and fosfomycin was determined according to Monsen et al. (10).

PFGE typing

Chromosomal DNA CoNS was prepared in agarose blocks and digested with SmaI (New England BioLabs, USA), as described elsewhere (14). The isolates were run on a 1% agarose gel (Invitrogen, USA) in a CHEF DRIII system (Bio-Rad, USA) under the following conditions: run time, 23 h; temperature, 13°C; voltage, 200 V; initial forward time, 5 s; final forward time, 60 s. The molecular weight markers (New England BioLabs, USA) were run in the first and in the last lane. The gels were stained with ethidium bromide, washed in water, and photographed under UV light by using the Gel Doc 1000 system (Bio-Rad, USA). The gel patterns were read by visual inspection. The isolates were classified as identical if they shared the same band profile, and isolates differing by more than six bands were considered to represent distinct DNA types (22).

RESULTS AND DISCUSSION

A total of eleven CoNS isolates belonging to five different species were identified including S. epidermidis (5 isolates), S. haemolyticus (3 isolates), S. hominis (1 isolate), S. warneri (1 isolate) and S. cohnii subsp urealyticus (1 isolate) (Table 1). All isolates, except the number 20994, were identified to the species level by the conventional method of Kloos and Banermann (7,8). The isolate 20994 could not be identified by the conventional method, so it was identified by Vitek-2 (bioMèrieux, France) as S. cohnii subsp urealyticus.

Table 1

Bloodstream Methicillin resistant Coagulase-negative staphyloccci isolated from patients at Hospital 9 de Julho (June/ July 2005).

Specie	Clinical isolate	Isolation date	Ward	MIC Vanco (μg/ml)	MIC Teico (μg/ml)	PFGE Profile
S. epidermidis	21170	17/07/05	internal medicine	2.0	16.0	“A”
S. epidermidis	21169	11/07/05	ICU	1.5	4.0	“B”
S. epidermidis	21168	20/07/05	oncology	1.5	4.0	“C”
S. epidermidis	20944	29/06/05	ICU	2.0	4.0	“D”
S. epidermidis	21171	01/07/05	oncology	2.0	4.0	“E”
S. haemolyticus	20995	07/07/05	internal medicine	2.0	12.0	“G”
S. haemolyticus	21172	13/07/05	oncology	1.5	4.0	“F”
S. hominis	20947	20/06/05	ICU	0.75	0.75	NC
S. warneri	20993	08/07/05	internal medicine	1.0	2.0	NC
S. haemolyticus	20946	24/06/05	internal medicine	2.0	3.0	“F”
S. cohnii subsp urealyticus	20994	06/07/05	oncology	1.0	3.0	NC

The two species most frequently encountered were S. epidermidis and S. haemolyticus.

The S. epidermidis and S. hominis isolates were identified by the disk diffusion susceptibility test to desferrioxamine, since other species of CoNS are resistant to desferrioxamine. To differentiate S. epidermidis from S. hominis, other phenotypic tests were used, as fermentation of trehalose (negative for S. epidermidis), alkaline phosphatase (positive for S. epidermidis) and growth in thioglicolate (positive for S. epidermidis).

The test for production of urease allowed the differentiate S. haemolyticus (urease negative), from S. epidermidis, S. hominis and S. warneri (urease positive). The test of positive PYR along with the hemolytic properties in sheep blood agar and absence of fermentation of mannose allowed the differentiation of S. haemolyticus, the second most prevalent species.

S. cohnii subsp urealyticus (isolate 20994) was the less common specie. Classical tests of resistance to novobiocin, production of urease and absence of sucrose fermentation confirmed the identification of the specie. A discrepancy in the alkaline phosphatase production was noted: the result was negative in the conventional test and positive in the automated system.

All isolates were methicillin-resistant by the disk diffusion test, with MICs ≥ 256 μg/ml by E-Test. Two isolates (S. epidermidis and S. haemolyticus) presented reduced susceptibility to teicoplanin (Table 1). Strains with this characteristic have been reported by Nunes et al. (13) and may be associated with treatment failures or may become precursors of glycopeptide-resistant strains (18).

S. cohnii subsp urealyticus is an unusual opportunist species that has been found in hospital environment like pediatric ICUs (24), and may constitute a dangerous reservoir of multiple antimicrobial plasmid mediated resistance genes (21,23).

Among the five clinical isolates of S. epidermidis five different patterns of PFGE were observed, indicating absence of clonal dissemination among the patients. The same did not occurr with the three clinical isolates of S. haemolyticus, where two isolates from patients at different wards, more than one month apart, presented the same profile, suggesting nosocomial transmission (Fig. 1).

Figure 1

PFGE profile of SmaI-digested chromosomal DNA of CoNS isolates, obtained from patients in 9 de Julho Hospital in São Paulo city, Brazil. λ lamba ladder DNA markers; lanes 1–5 S. epidermidis; lanes 6,7 and 10: S. haemolyticus; lane 8: S. hominis; lane 9: S. warneri; lane 11: S. cohnii spp urealyticus.

In the last years, the importance of CoNS has been increasing due to their pathogenicity and involvement in human diseases. Their identification species in the clinical laboratories is important but not an easy task, because classical phenotypic tests do not differentiate them from other staphylococci require more time in the identification compared to commercial kits. Many clinical laboratories use automated systems for identification of Staphylococcus spp., although the reliability of results for certain species is not always satisfactory, particularly for species other than S. epidermidis.

Two isolates, one S. epidermidis (isolate 21170) and one S. haemolyticus (isolate 20995) presented high MIC for teicoplanin, but were susceptible to vancomycin (Table 1).

At 9 de Julho hospital, staphylococcal infections, mainly those caused by S. aureus, have been successfully treated with teicoplanin instead of vancomycin, for more than a decade, particulary IV-cateter related infections. In counter part, treatment of central nervous infections and endocarditis with teicoplanin has been less effective, probably due to the occurrence of oxacillin-resistant CoNS also resistant to teicoplanin. A surveillance program of glycopeptide resistance and adequate CoNS specie identification have great importance in determination of risk factors and implementation of nosocomial infection control measures.