Mupirocin Resistance in Isolates of Staphylococcus spp. from Nasal Swabs in a Tertiary Hospital in France

Abstract

Mupirocin is a topical antibiotic largely used to eradicate staphylococcal nasal carriage. Here, we investigated the prevalence of mupirocin-resistant Staphylococcus aureus and coagulase-negative staphylococcal isolates recovered from patients in different wards in a hospital (Lyon, France), which were determined both phenotypically with an Epsilometer test (Etest) and genetically by PCR for mupA and mupB.

TEXT

Staphylococcus aureus, especially methicillin-resistant S. aureus (MRSA), is a leading cause of nosocomial infections. It has been well established that staphylococcal carriers are three times more likely to develop infections than noncarriers (1, 2). Moreover, these infections typically originate from patients' own flora (3). This finding is of major concern, in light of the fact that 30% of the population worldwide is persistently or occasionally colonized by S. aureus, especially in the anterior nares (4). Colonized patients can also be a source of dissemination to other patients and health care workers (5). To limit and reduce this burden, several drugs have been developed in order to eradicate the nasal carriage of methicillin-susceptible S. aureus (MSSA)/MRSA before surgery or, more commonly, once a patient has been identified as an MRSA carrier.

Mupirocin is the most widely used topical antibiotic for nasal application (6). It binds to the isoleucyl-tRNA synthetase (IleRS) required for bacterial protein synthesis and acts as an analog of isoleucine to inhibit the transfer of this amino acid. Since it was first employed in the clinic, mupirocin has been widely used, which has led to the emergence of resistant isolates (7). The following resistance mechanisms have been identified (8): (i) low-level resistance (LMupR) due to mutations in the gene that encodes IleRS that prevent mupirocin binding, corresponding to an MIC ranging from 8 to 256 mg/liter; and (ii) high-level resistance (HMupR) due to the acquisition of a plasmid containing the mupA gene coding for an alternative IleRS that is not targeted by mupirocin, corresponding to an MIC of ≥512 mg/liter (9). More recently, a new plasmid-mediated resistance gene, mupB, which confers a high level of resistance to mupirocin (MIC, 1,024 mg/liter) in S. aureus, has been identified (10).

Several studies have reported the emergence of resistant isolates with various prevalences according to geographical area, time period, and hospital ward (11–13). The aim of the present study was to determine the prevalence of mupirocin-resistant isolates of S. aureus and coagulase-negative staphylococci (CoNS) colonizing the nares of patients hospitalized in intensive care units (ICUs), which is the population primarily targeted for MRSA screening, and those of patients admitted for orthopedic surgery (OS), who are representative of the general inpatient population.

Nasal swabs from both nostrils (one sample per patient) were routinely collected from June to December 2010 from ICU and OS patients at the Northern Hospital Group of the Hospices Civils de Lyon, Lyon, France. Our institution is a 680-bed tertiary care university hospital with medical, surgical, and neonatal intensive care units. Isolation of MRSA-colonized and MRSA-infected patients is performed in all wards, while systematic screening for MRSA nasal carriage is restricted to ICU patients. The patients included in this study were admitted to a 38-bed ICU, which includes 20 beds of medical care and 18 beds of surgical care, or to a 39-bed OS ward with single rooms. The nasal swabs were used to seed chromogenic ChromID medium (bioMérieux, Marcy l'Etoile, France), which supports the growth of all staphylococcal species, using the quadrant technique described previously (14). Identification of the different staphylococcal isolates present (on the basis of colony morphology) was performed using matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) (bioMérieux) (15). The MIC of mupirocin was determined using the Etest technique (Epsilometer test; bioMérieux), according to the manufacturer's instructions. In addition, a genetic screening for the presence of genes conferring resistance to mupirocin and methicillin in all isolates was performed by multiplex PCR for mupA, mecA, and 16S and for mupB and 16S, as described elsewhere (16, 17). Statistical analyses were conducted with the R software, version 2.14.2 (The R Foundation for Statistical Computing, Vienna, Austria), and significance was determined using Z-tailed Fisher's exact test, with a threshold of 0.05.

Three hundred thirty-one nasal swabs were collected from 185 ICU patients and 146 OS patients. A total of 635 isolates belonging to staphylococcal species (n = 317 from ICU patients [50%] and n = 318 from OS patients [50%]) were isolated and identified as S. aureus (n = 85) or CoNS (n = 550) (Table 1). MALDI-TOF MS identification of the isolates revealed that S. epidermidis, S. aureus, and S. haemolyticus were the most prevalent species (comprising 64.1, 13.4, and 10.1% of isolates, respectively) (Table 2) . Species distribution in ICU and OS patients differed significantly only for S. haemolyticus and S. warneri, which were more prevalent in ICU and OS patients, respectively (P < 0.001 for both differences). Out of these 635 staphylococcal isolates, methicillin resistance was detected in 269 (42.3%) isolates, including the following: 8 MRSA isolates, representing 9.4% (8/85) of all S. aureus isolates, and 261 methicillin-resistant CoNS (MRCoNS) isolates, representing 47.5% (261/550) of all CoNS isolates. Compared with the rate of colonization in the OS patients, the rates of methicillin resistance for the S. aureus (P < 0.05) and CoNS strains (P < 0.001) were significantly higher in the ICU patients (Table 1). Among the S. aureus isolates, mupirocin resistance was not detected by phenotypic or genetic tests. Out of 550 CoNS isolates, 12 (2.2%) exhibited LMupR (8 < MIC < 256 mg/liter) and showed negative PCR results for mupA and mupB. Conversely, 31 isolates (5.6%) showed HMupR, with a corresponding MIC of ≥512 mg/liter, and all 31 tested positive by mupA PCR (Table 1). No isolates tested positive for mupB. Thirty-nine out of 43 mupirocin-resistant CoNS (90.7%) were S. epidermidis, and 30 (70%) were also methicillin resistant. Considering S. aureus carriers only (n = 85), 5 patients (1 with MSSA and 4 with MRSA) (6%) were cocolonized with CoNS HMupR. No case of cocolonization with S. aureus and CoNS LMupR was identified in this study.

TABLE 1.

Prevalence of methicillin and mupirocin resistance, including low- and high-level resistance, among S. aureus and coagulase-negative staphylococcal strains found in intensive care units and orthopedic surgery wards

Staphylococcal straina	No. (%) of strains inb:		P valuec	Total no. (%) (n = 635)
	OS patients (n = 318)	ICU patients (n = 317)
S. aureus	36 (11.3)	49 (15.4)	NS	85 (13.4)
MRSA	1 (0.3)	7 (2.2)	<0.05	8 (1.3)
LMupR	0 (0.0)	0 (0.0)	NS	0 (0.0)
HMupR	0 (0.0)	0 (0.0)	NS	0 (0.0)
CoNS	282 (88.7)	268 (84.5)	NS	550 (86.6)
MRCoNS	67 (21.1)	194 (61.2)	<0.001	261 (41.1)
LMupR	1 (0.3)	11 (3.5)	<0.01	12 (1.9)
HMupR	20 (6.3)	11 (3.5)	NS	31 (4.9)

^aMR, methicillin resistant; LMupR, low-level resistance to mupirocin due to a mutation in the ileS gene, corresponding to an MIC of between 8 and 256 mg/liter; HMupR, high-level resistance to mupirocin due to the acquisition of a plasmid containing the mupA gene, corresponding to an MIC of ≥512 mg/liter; CoNS, coagulase-negative staphylococci.

^bOS patients were admitted for orthopedic surgery and are representative of the general inpatient population. ICU, intensive care unit.

^cStatistical analyses were conducted with the R software, version 2.14.2 (The R Foundation For Statistical Computing, Vienna, Austria), and significance was determined using Z-tailed Fisher's exact test, with a threshold of 0.05. NS, nonsignificant.

TABLE 2.

Distribution of staphylococcal species found in intensive care units and orthopedic surgery wards

Staphylococcus species	No. (%) of strains ina:		P valueb	Total no. (%) (n = 635)
	OS patients (n = 318)	ICU patients (n = 317)
S. aureus	36 (11.3)	49 (15.5)	NS	85 (13.4)
S. capitis	9 (2.8)	10 (3.2)	NS	19 (3.0)
S. caprae	1 (0.3)	0 (0)	NS	1 (0.2)
S. carnosus	1 (0.3)	0 (0)	NS	1 (0.2)
S. cohnii	5 (1.6)	1 (0.3)	NS	6 (0.9)
S. delphini	0 (0)	1 (0.3)	NS	1 (0.2)
S. epidermidis	222 (69.8)	185 (58.4)	NS	407 (64.1)
S. haemolyticus	15 (4.7)	49 (15.5)	<0.001	64 (10.1)
S. hominis	5 (1.6)	10 (3.2)	NS	15 (2.4)
S. intermedius	0 (0)	1 (0.3)	NS	1 (0.2)
S. lugdunensis	4 (1.3)	6 (1.9)	NS	10 (1.6)
S. pseudintermedius	0 (0)	1 (0.3)	NS	1 (0.2)
S. saprophyticus	1 (0.3)	1 (0.3)	NS	2 (0.3)
S. schleiferi	1 (0.3)	0 (0)	NS	1 (0.2)
S. sciuri	0 (0)	1 (0.3)	NS	1 (0.2)
S. simulans	1 (0.3)	0 (0)	NS	1 (0.2)
S. warneri	17 (5.3)	2 (0.6)	<0.001	19 (3.0)

^aOS patients were admitted for orthopedic surgery and are representative of the general inpatient population. ICU, intensive care unit.

^bStatistical analyses were conducted with the R software, version 2.14.2 (The R Foundation For Statistical Computing, Vienna, Austria), and significance was determined using Z-tailed Fisher's exact test, with a threshold of 0.05. NS, nonsignificant.

Our study shows that mupirocin resistance is infrequent in S. aureus isolates and remains low in CoNS (6.8%) at our institution. Our results were in line with those of previous studies. Lamy et al. (18) observed no resistance among 235 S. aureus strains isolated from community-acquired skin and soft tissue infections, while Desroches et al. (7) reported mupirocin resistance rates of 2.2% and 10.3% in invasive MRSA and CoNS isolates, respectively, collected from 37 hospitals. Of note, in our institution, mupirocin is not routinely used to decolonize preoperative patients or patients hospitalized in the ICU, which might explain the low rates of resistance observed in our study. Nasal decolonization using mupirocin is proposed only in patients with recurrent S. aureus infection. However, in our study, two-thirds (31/43) of the mupirocin-resistant CoNS exhibited high-level resistance due to the presence of a plasmid containing mupA. These data raise the question of the risk the acquisition of mupirocin resistance by S. aureus isolates in case of cocolonization with mupirocin-resistant CoNS, especially when the bacteria are under mupirocin selective pressure (19, 20). This risk was confirmed by a recent study demonstrating the acquisition of HMupR by CoNS following nasal decolonization with mupirocin, in which 21% (192/936) and 43% (406/936) of patients were found to carry mupirocin-resistant CoNS before treatment and after decolonization, respectively (P < 0.001) (21). Considering these findings, we suggest monitoring the development of resistance in S. aureus when decolonization strategies involving mupirocin are used. This recommendation is further supported by several studies reporting high rates of mupirocin resistance or clonal dissemination of mupirocin-resistant MRSA and CoNS in hospital settings (7). Together, these findings highlight the importance of performing regular sentinel studies in different parts of the world to rapidly identify the emergence and dissemination of mupirocin-resistant staphylococci (7, 11, 21, 22).