Skip Navigation

Epidemiological Profile of Linezolid-Resistant Coagulase-Negative Staphylococci

  1. Brian A. Potoski1,3,4,
  2. Jennifer Adams3,
  3. Lloyd Clarke4,
  4. Kathleen Shutt3,
  5. Peter K. Linden2,
  6. Carla Baxter5,
  7. A. William Pasculle5,
  8. Blair Capitano1,3,4,
  9. Anton Y. Peleg3,
  10. David L. Paterson3,4, and
  11. Dora Szabo3,a
  1. 1Departments of Pharmacy and Therapeutics, Pittsburgh, Pennsylvania
  2. 2Departments of Critical Care Medicine, Pittsburgh, Pennsylvania
  3. 3Departments of Division of Infectious Diseases, Pittsburgh, Pennsylvania
  4. 4Departments of Antibiotic Management Program, Pittsburgh, Pennsylvania
  5. 5Departments of Clinical Microbiology Laboratory, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
  1. Reprints or correspondence: Dr. David L. Paterson, Div. of Infectious Diseases, Ste. 3A, Falk Medical Bldg., 3601 5th Ave., Pittsburgh, PA 15213 (patersond{at}dom.pitt.edu).
  • a Present affiliation: Semmelweis University, Budapest, Hungary.

Abstract

Background. Surveillance studies have shown that <0.1% of coagulase-negative staphylococci are linezolid resistant; however, at our institution, 4% of such organisms were found to be resistant. We investigated the risk factors for and the epidemiological profile of linezolid-resistant coagulase-negative staphylococci.

Methods. Susceptibility testing and pulsed-field gel electrophoresis were performed to analyze the genetic relatedness of both linezolid-resistant and linezolid-susceptible isolates. Clinical data were retrieved from medical records, and a case-case-control study was performed to identify unique risk factors for linezolid resistance.

Results. Isolates recovered from 25 patients with linezolid-resistant coagulase-negative staphylococci were examined; all but 1 of the isolates were identified as Staphylococcus epidermidis, and all but 1 had a minimum inhibitory concentration of linezolid of >256 µg/mL. Pulsed-field gel electrophoresis showed that 21 (84%) of 25 linezolid-resistant isolates exhibited genetic relatedness, whereas linezolid-susceptible isolates were of diverse clones. Unique, independent predictors of linezolid resistance included receipt of linezolid in the 3 months preceding isolation of the coagulase-negative staphylococci (odds ratio, 20.6; 95% confidence interval, 5.8–73.0).

Conclusion. Linezolid-resistant coagulase-negative staphylococci have emerged at our institution and are predominately of a single clone. We believe that the most likely scenario to explain this emergence is that person-to-person spread of linezolid-resistant coagulase-negative staphylococci led to establishment of skin colonization with the strain. Subsequent use of linezolid was followed by selection of the linezolid-resistant strain, which then became the dominant skin flora. The potential for a parallel scenario involving clonal dissemination followed by selection of linezolid-resistant methicillin-resistant Staphylococcus aureus is a real possibility.

Linezolid is an oxazolidinone antimicrobial agent that is typically used for the treatment of infections due to vancomycin-resistant Enterococcus faecium, for the treatment of either pneumonia or skin or osteoarticular infection due to methicillin-resistant Staphylococcus aureus (MRSA), or for the treatment of heteroresistant vancomycin-intermediate S. aureus. Resistance of vancomycin-resistant E. faecium to linezolid has been widely reported, especially (but not exclusively) in the United States [16]. In contrast, at the present time, reports of resistance of MRSA to linezolid have been limited to case reports [5, 713]. Surveillance programs have reviewed the linezolid susceptibility of >10,000 isolates of coagulase-negative staphylococci during the past 5 years, with <0.1% of isolates determined to be linezolid resistant [5, 1420].

Linezolid susceptibility testing was introduced on a routine basis in our institution (University of Pittsburgh Medical Center, Pittsburgh, PA) in late January 2005. From February through October 2005, 74 (4.4%) of 1680 isolates of coagulase-negative staphylococci at our institution were found to be linezolid resistant. For this reason, we commenced an epidemiological investigation of linezolid resistance in coagulase-negative staphylococci. Furthermore, such an investigation was believed to be important, because it might shed light on the potential for the spread of linezolid-resistant S. aureus.

Methods

Bacterial isolates. Routine linezolid susceptibility testing of all gram-positive cocci in the clinical microbiology laboratory of the University of Pittsburgh Medical Center began in late January 2005. This testing was performed using disk diffusion methods consistent with Clinical and Laboratory Standards Institute recommendations [21]. In August 2005, a review of the findings revealed that linezolid-resistant coagulase-negative staphylococci were being detected at rates well above those that had been previously reported. From this point in time, we prospectively saved linezolid-resistant isolates and retrieved blood culture isolates of organisms with linezolid resistance from February through August 2005. The microbiological and subsequent epidemiological analyses were approved as a quality-improvement study by the hospital's Total Quality Council. Linezolid-susceptible coagulase-negative staphylococci were also identified during the same period and were used for comparison. The study period concluded at the end of October 2005.

In vitro susceptibility testing. The MICs of linezolid, daptomycin, and vancomycin for the coagulase-negative staphylococci available for this analysis were determined using E test (AB Biodisk). Speciation was performed using Staph API strips (bioMérieux).

PFGE. Two sets of organisms were assessed by PFGE: 25 isolates of linezolid-resistant coagulase-negative staphylococci and a contemporaneous, randomly selected sample of 30 isolates of linezolid-susceptible coagulase-negative staphylococci. PFGE was performed using methods described elsewhere [22]. Digestion of the genomic DNA was performed using SmaI (New England Biolabs); the restriction fragments were separated by PFGE performed using a temperature-controlled CHEF DR III system (Bio-Rad). After undergoing staining with ethidium bromide, the fragments were visualized using a Gel Doc 2000 system (Bio-Rad). For PFGE pattern analysis, Bionumerics software, version 4.0 (Applied Maths), with use of the unweighted pair group method with arithmetic mean and the dice setting for clustering analysis, was applied. Genetic relatedness of the isolates was determined using the criteria of Tenover et al. [23].

Microbiological definitions. We defined bloodstream infection as growth of coagulase-negative staphylococci in 2 of 2 blood culture sets. A contaminant was defined as a positive result for only 1 of 2 blood culture sets. An indeterminate result occurred when only 1 blood culture set was collected and had positive results.

Institutional use of antibiotics. Pharmacy records were reviewed to determine the aggregate use of linezolid within the hospital and within particular intensive care units (ICUs) of the hospital. The defined daily dose (DDD) of linezolid was 1200 mg. Data were expressed as the DDDs per 100 patient-days of bed occupancy in the hospital.

Case-case-control study. A case-case-control study design was performed according to the methods of Harris et al. [24] and Kaye et al. [25]. This study design involves 2 separate case-control analyses. In the first analysis, patients with linezolid-resistant coagulase-negative staphylococci (known as “resistant case patients”) were compared with a group of control patients who were randomly selected from hospitalized patients receiving care from the same health care services and at similar times as the resistant case patients (i.e., both case patients and control patients were from identical source populations [24, 25]). For each resistant case patient, 4 control patients were evaluated. In the second analysis, 50 patients with linezolid-susceptible coagulase-negative staphylococci identified in clinical cultures (known as “susceptible case patients”) were randomly selected from the same time period as resistant case patients and were compared with the same control group, as described above (with the exception that any patient who was identified as a resistant case patient could not be included in the study as a control patient).

The benefits of this study methodology are that the 2 risk models can be compared against a source population. Therefore, the independent predictors for resistant case patients are the unique risk factors for linezolid resistance. This methodology, compared with a “traditional” case-control study, is particularly important in determining whether previous use of linezolid is a potential risk factor for linezolid resistance. Traditional case-control studies are likely to falsely overemphasize previous linezolid therapy as a risk factor.

Statistics. Potential risk factors for linezolid resistance in coagulase-negative staphylococci were evaluated using SAS software, version 8.2 (SAS Institute). All categorical variables were examined using the SAS procedure Genmod. Continuous variables were examined using the nonparametric Wilcoxon test. All factors were eligible for entry in the multivariable model. Unconditional multivariable stepwise logistic regression was performed using the SAS procedure Logistic, with the entry criteria set to 0.15 and the stay criteria set to 0.075.

Results

Overview. From February through October 2005, 74 (4.4%) of 1680 isolates of coagulase-negative staphylococci at our institution were found to be linezolid resistant. After elimination of multiple isolates recovered from individual patients and exclusion of isolates that were not available for verification of linezolid resistance, 25 patients were identified as having linezolid-resistant coagulase-negative staphylococci (table 1). Nineteen (76%) of these 25 patients had used linezolid during the 3 months preceding isolation of the linezolid-resistant coagulase-negative Staphylococcus organism. The majority of isolates (22 of 25 isolates) recovered from these patients were obtained from blood cultures. Five of the patients had bloodstream infections; 14 had blood culture contaminants; and 3 had just a single set of blood cultures collected, making interpretation of the significance of these isolates impossible.

Table 1
Table 1

Characteristics of patients with linezolid-resistant coagulase-negative staphylococci.

Microbiological and in vitro susceptibility testing. All but 1 of the linezolid-resistant isolates was identified as Staphylococcus epidermidis. The remaining isolate was Staphylococcus lugdunensis. The MIC of linezolid was >256 µg/mL for all but 1 isolate (table 2). The MICs of vancomycin were 1.5–3 µg/mL, and those of daptomycin were 0.38–1 µg/mL. of 30 randomly selected linezolid-susceptible isolates, 13 (43%) of 30 were identified as S. epidermidis, 8 (27%) of 30 were Staphylococcus hominis, 3 (10%) of 30 were Staphylococcus capitis, and the remaining 6 isolates comprised 6 different species.

Table 2
Table 2

Microbiological data for each isolate recovered from patients with linezolid-resistant coagulase-negative staphylococci.

PFGE. Isolates recovered from 21 (84%) of 25 patients with linezolid-resistant isolates had some genetic relatedness, whereas isolates recovered from 4 patients were genetically distinct (table 2). Fifteen of the 21 patients who had isolates with genetic relatedness had been accommodated in ICU A at some time before the linezolid-resistant isolate was recovered. Thirteen of these patients had also spent some time in ward C, which serves as a non-ICU inpatient area for patients after their stay in ICU A.

Two patients had multiple isolates recovered. Patient 1 had 2 isolates recovered 1 week apart; these 2 isolates were indistinguishable. However, a third isolate from this patient, which was recovered 4 months after recovery of the first isolate, had 2–3 fragment differences, compared with the original isolate. Patient 13 had 2 isolates recovered 2 weeks apart. These isolates were indistinguishable from each other. In contrast, none of the 30 isolates of linezolid-susceptible coagulase-negative staphylococci shared genetic relatedness.

Institutional use of antibiotics. Linezolid use at our institution has steadily increased over the past 5 years. The numbers of DDDs of linezolid per 100 patient-days were 0.61 (in 2001), 1.05 (in 2002), 1.11 (in 2003), 1.27 (in 2004), and 1.74 (in 2005). During 2005, use of linezolid in ICU A was 13.6 DDDs per 100 patient-days.

Case-case-control study. The first analysis of the case-case-control study involved the 25 patients with confirmed linezolid-resistant coagulase-negative staphylococci (i.e., resistant case patients) and 100 control patients from the same source population. Univariate analysis revealed that the following risk factors were significantly more likely to be associated with case patients with linezolid-resistant coagulase-negative staphylococci than with control patients: location in ICU A, location in ward C, presence of acute renal failure necessitating continuous renal replacement therapy, receipt of a sold-organ transplant, surgery in the 3 months preceding isolation of the coagulase-negative staphylococci, receipt of total parenteral nutrition, recent receipt of piperacillin-tazobactam, recent receipt of metronidazole, and recent receipt of linezolid (table 3).

Table 3
Table 3

Risk factors for linezolid-resistant coagulase-negative staphylococci, according to univariate analysis.

Multivariate analysis revealed that recent receipt of linezolid (OR, 20.6; 95% CI, 5.8–73.0) and location in ward C (OR, 12.4; 95% CI, 3.4–45.5) were independent predictors of isolation of a linezolid-resistant coagulase-negative staphylococcus by culture. In the second analysis of the case-case-control study, which compared patients who had linezolid-susceptible coagulase-negative staphylococci with control patients, the only variable that differentiated case patients from control patients in multivariate analysis was receipt of linezolid during the 3 months before isolation of the coagulase-negative staphylococci. This variable was significantly less likely to be associated with patients with linezolid-susceptible coagulase-negative staphylococci (OR, 0.09; 95% CI, 0.01–0.57).

Discussion

An annual appraisal of susceptibility of gram-positive organisms to linezolid is conducted by the manufacturer of the antibiotic [14]. In this surveillance program, numerous institutions worldwide submit 200 gram-positive strains to a reference laboratory. In the most recently published surveillance study, which was performed in 2002, a total of 870 strains of coagulase-negative staphylococci were evaluated. Just 1 linezolid-resistant isolate was found (MIC, ⩾16 µg/mL) [14]. An alternative surveillance program (LEADER 2004) assessed 496 strains of coagulase-negative staphylococci from 50 hospital laboratories in the United States [26]. Only 1 linezolid-nonsusceptible strain (MIC, 32 µg/mL) was found. However, in 2005, at our institution, 4% of all isolates of coagulase-negative staphylococci were found to be linezolid resistant when we commenced routine susceptibility testing. Therefore, we conducted an investigation into the epidemiological profile of linezolid-resistant coagulase-negative staphylococci at our institution.

The major findings of the present study were as follows: (1) the molecular epidemiological profile revealed that the majority of patients (21 [84%] of 25 patients) had isolates that exhibited some genetic relatedness, (2) previous linezolid use was an independent predictor of linezolid-resistance in coagulase-negative staphylococci (the majority of patients—19 [76%] of 25 patients—with linezolid-resistant isolates had previously used linezolid), and (3) the majority of patients with linezolid-resistant coagulase-negative staphylococci had been accommodated in a single ward (ward C) and its “feeder” ICU (ICU A). Historically, these wards have had higher rates of vancomycin-resistant enterococci than have other wards in the hospital. Our hospital has relatively high rates of linezolid usage overall (1.7 DDDs per 100 patient-days, compared with a mean of 0.4 DDDs per 100 patient-days in previously studied US hospitals [14]), and linezolid use has been particularly heavy in this ICU (13.6 DDDs per 100 patient-days). Previous studies of the molecular epidemiological profile of coagulase-negative staphylococci (which, presumably, are linezolid susceptible) have confirmed that clonal spread of coagulase-negative staphylococci occurs within hospitals [27].

One potential explanation, therefore, is that patient-to-patient transmission of linezolid-resistant strains has occurred, with establishment of linezolid-resistant coagulase-negative staphylococci as part of the skin flora. Heavy use of linezolid in these patients may have created substantial selection pressure in favor of the linezolid-resistant isolates. An alternative explanation is that selection for de novo resistance in a prevalent nosocomial clone of linezolid-susceptible coagulase-negative staphylococci has occurred. This is not implausible among a population who have a long duration of hospitalization and are receiving an extraordinary amount of linezolid (almost 10-fold the amount received by the general hospital population). We acknowledge that there are some limitations in confirming our hypotheses, because the study was performed in a single center and assessed a small cohort of 25 patients. This limits our ability to draw extensive conclusions from the sample. Furthermore, we were unable to determine whether patients were colonized with the resistant strain before linezolid exposure or after linezolid exposure. Indeed, the origin of the linezolid-resistant clone is unknown and may have even been introduced from outside our institution.

What is the clinical significance of linezolid resistance in coagulase-negative staphylococci? The majority of isolates in the present investigation were blood culture contaminants. Only 5 patients had true bacteremia. At our institution, vancomycin is the standard treatment for bacteremia due to coagulase-negative staphylococci. A number of isolates had MICs of vancomycin of 3 µg/mL, according to the E test. At the present time, 4 µg/mL is the upper limit of the range considered to denote susceptibility in coagulase-negative staphylococci [21]. It would be likely, therefore, that vancomycin would be active even for these strains. An occasional patient is intolerant of vancomycin, so an alternative agent, such as linezolid or daptomycin, would be necessary for that patient. Given the advent of linezolid resistance in coagulase-negative staphylococci, physicians should ensure that linezolid susceptibility is confirmed by their clinical microbiology laboratories if linezolid is to be used.

Mupirocin resistance has been found to be successfully transferred from coagulase-negative staphylococci to S. aureus [28]. In an individual patient, mupirocin-resistant MRSA and S. epidermidis harbored identical plasmids that carried both the resistance gene and the genes for conjugative DNA transfer. These plasmids could be transferred in vitro from both clinical isolates to a reference S. aureus strain [28]. This raises the specter that linezolid resistance could be transferred from linezolid-resistant coagulase-negative staphylococci to S. aureus. However, to our knowledge, genes encoding linezolid resistance have never been found on transferable genetic elements. We are currently investigating the mechanisms of linezolid resistance in these strains, and we will investigate whether they are chromosomally encoded or plasmid encoded.

We believe that the major significance of our description of the epidemiological profile of linezolid-resistant coagulase-negative staphylococci is the concern that clonal spread and linezolid selection pressure could occur with MRSA. Use of linezolid for treatment of MRSA-associated pneumonia is increasing, because it has produced successful results, compared with use of vancomycin [29]. At the present time, few cases of linezolid-resistant MRSA have been reported. However, just as linezolid-resistant coagulase-negative staphylococci were virtually nonexistent in surveillance studies as recently as 2002 [14], we have demonstrated that dissemination and selection of linezolid-resistant strains can occur. This has also recently been noted in a university hospital in New Jersey [30]. In that setting, as in our institution, some patients developed infections with linezolid-resistant coagulase-negative staphylococci in the absence of exposure to linezolid [30]. We are currently contemplating performing additional studies of the selection of linezolid-resistant skin flora, including S. aureus, in patients treated with linezolid. In addition, we have bolstered infection-control and antibiotic-management activities in ICU A and ward C. Understanding the development and epidemiological profile of linezolid resistance is likely to be crucial to controlling this emerging problem. Linezolid-resistant coagulase-negative staphylococci may be a harbinger of linezolid-resistant MRSA or heteroresistant vancomycin-intermediate S. aureus.

Acknowledgments

We thank Julie Klukan and Melissa Saul for their assistance with data collection.

Potential conflicts of interest. B.A.P. is on the speakers' bureau of Wyeth Pharmaceuticals; P.K.L. has received recent research funding from Pfizer, is a consultant to Cubist, and is on the speakers' bureaus of Pfizer, Cubist, and Merck; B.C. is on the speakers' bureau of Wyeth Pharmaceuticals; and D.L.P. has received recent research funding from AstraZeneca, Pfizer, Merck, and Elan and is on the speakers' bureaus of Roche, AstraZeneca, Pfizer, Cubist, Merck, and Elan. All other authors: no conflicts.

  • Received December 22, 2005.
  • Accepted March 20, 2006.

References

| Table of Contents