Nosocomial outbreak of linezolid-resistant Enterococcus faecalis infection in a tertiary care hospital

Available online at www.sciencedirect.com

Diagnostic Microbiology and Infectious Disease 65 (2009) 175 – 179 www.elsevier.com/locate/diagmicrobio

Case Reports

Nosocomial outbreak of linezolid-resistant Enterococcus faecalis infection in a tertiary care hospital Rosa Gómez-Gila,1 , Maria Pilar Romero-Gómeza,⁎,1 , Africa García-Ariasa , M. Gallego Ubedab , M. Sota Busseloc , Ramón Cisternac , Avelino Gutiérrez-Altésa , Jesus Mingorancea a

Servicio de Microbiología y Parasitología and Unidad de Investigación, Hospital Universitario La Paz, 28046 Madrid, Spain b Servicio de Farmacia Hospitalaria, Hospital Universitario La Paz, 28046 Madrid, Spain c Servicio de Microbiología y Parasitología, Hospital de Basurto, 28046 Bilbao, Spain Received 24 March 2009; accepted 15 June 2009

Abstract We describe 12 cases of linezolid-resistant Enterococcus faecalis. The present study was done in 2 wards of Hospital Universitario La Paz in Madrid, Spain. The 2 wards involved were the intensive care unit (ICU) and reanimation unit. Twelve clinical strains of E. faecalis reported by the clinical laboratory as linezolid resistant based on MICs determined by E-test (AB Biodisk, Solna, Sweden) were collected between September 2005 and October 2006. The MIC of linezolid for all the resistant isolates was N128 μg/mL. The isolates were analyzed for the presence of the G2576T mutation by polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) and pyrosequencing. Pyrosequencing showed that the first isolate had G and T at position 2576 in a 1:1 ratio, whereas the remaining ones had a wild type to mutant ratio of 1:3. PCR-RFLP showed that the mutations were in alleles 1, 3, and 4. The 12 isolates under investigation came from different patients but were indistinguishable by pulsed-field gel electrophoresis (n = 7) and repetitive extragenic palindromic sequence (REP)-PCR (n = 12). This is the first report of a clonal outbreak of linezolid-resistant E. faecalis in Spain. To prevent or minimize the emergence of resistance, we should use linezolid strictly after the therapeutic indications, courses of treatment should be kept as short as possible, and risk factors for resistance development should be considered before starting. In addition, we suggest that susceptibility testing of clinically significant Gram-positive pathogens should be done in all cases of treatment failure, and, depending on the local epidemiology of each ICU, it might be advisable to do it before starting treatment with linezolid. © 2009 Elsevier Inc. All rights reserved. Keywords: Linezolid; Enterococcus faecalis

1. Introduction Enterococci are among the most important nosocomial pathogens. They are a part of the intestinal flora of healthy individuals and are normally not attributed high virulence, but in the hospital environment, they are one of the most frequent causes of endocarditis, bacteremia, and urinary and wound infections. The most commonly isolated species are Enterococcus faecalis (80–90%) and Enterococcus faecium (5–10%). The intrinsic resistance of ⁎ Corresponding author. Tel.: +34-91-727-72-48; fax: +34-91-72773-72. E-mail address: [email protected] (M.P. Romero-Gómez). 1 Both authors participated equally in this work. 0732-8893/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.diagmicrobio.2009.06.010

the enterococci to a large number of antibiotics and their great capacity to develop new resistances have contributed to the growing importance of these bacteria in the hospital environment. The emergence of multiresistant Gram-positive bacteria has prompted research for new types of antibiotics. A product of these efforts has been the development of linezolid, a synthetic antibiotic, which is the first commercially available drug of the group of the oxazolidinones. Linezolid inhibits bacterial protein synthesis by binding to the ribosomal peptidyl transferase center, in the 50S subunit of the ribosome, and interfering with the initiation of translation. It is active against most Gram-positive bacteria, is well absorbed when administered orally, facilitating the conversion from intravenous to oral therapy, and is generally well

176

R. Gómez-Gil et al. / Diagnostic Microbiology and Infectious Disease 65 (2009) 175–179

Table 1 Antibiotic susceptibility profiles of 1 linezolid-resistant isolate and a control strain (ATCC 29212) Antimicrobial agent

Penicillin

Ampicillin

E. faecalis ATCC 29212 MIC (μg/mL) E. faecalis linezolid Resistant MIC (μg/mL)/ interpretative standard

1

≤0.5

1/S

≤0.5/S

Vancomycin

Teicoplanin ≤1

2 ≤1/S

≤1/S

Linezolid 2 N32/R

Gentamicin 500 ≤500 N500/R

MIC = minimum inhibitory concentration; S = susceptible; R = resistant.

tolerated. Clinical trials have confirmed its effectiveness in the treatment of serious infections of the skin, soft tissues, and the lower respiratory tract (Plosker and Figgitt, 2005; Zhanel et al., 2001). Linezolid provides high rates of clinical cure and microbiologic success in complicated infections due to Enterococcus spp. However, several instances of emergence of resistance during linezolid treatment have been reported in clinical isolates of enterococci (E. faecium and E. faecalis) (Dibo et al., 2004; Halle et al., 2004; Johnson et al., 2002; Lobritz et al., 2003; Marshall et al., 2002; Ruggero et al., 2003) and staphylococci (Staphylococcus aureus and Staphylococcus epidermidis) (Meka et al., 2004; Tsiodras et al., 2001; Wilson et al., 2003). The LEADER study has documented between 3% and 5% linezolid resistance in Enterococcus (Kloss et al., 1999; Prystowsky et al., 2001). The G2576T mutation has been found to be related to resistance to linezolid both in vitro and in vivo in several Gram-positive bacteria, including E. faecalis and E. faecium. Other mutations, such as G2528U and the G2505A, have also been shown to cause linezolid resistance in enterococci in vitro (Lobritz et al., 2003; Multnick et al., 2003). E. faecalis contains 4 copies of the 23S rRNA gene, and several groups have reported a direct correlation between the number of alleles containing the G2576T mutation and the level of linezolid resistance (Marshall et al., 2002; Ruggero et al., 2003). In this report, we describe 12 cases of linezolid-resistant E. faecalis detected in our institution between September 2005 and October 2006.

REA. In the following year, 2003, the consumption of linezolid was multiplied by 13.5 in the ICU and by 3.3 in the REA. The increase in the use of linezolid continued until 2005. Since then, it has stabilized at an average of 1561 DDDs per year in the ICU (39 times the DDD in 2002) and 943 DDDs per year in the REA (14 times the DDD in 2002). 2.2. Strain selection Twelve clinical strains of E. faecalis reported by the clinical laboratory as linezolid resistant based on MICs determined by E-test (AB Biodisk, Solna, Sweden) were collected between December 2005 and October 2006. The strains were recovered from several sources: 3 from blood, 2 from catheters, 1 from a surgical wound, 1 from skin, and from various body fluids (1 from pleural fluid, 1 from surgical drainage, and 3 from urine). Strain clonality was analyzed by pulsed-field gel electrophoresis (PFGE) of SmaI-

2. Materials and methods 2.1. Setting The present study was done in 2 wards of Hospital Universitario La Paz in Madrid, Spain. This is a teaching hospital that attends a population of 787 000 people and has 1.328 beds. The 2 wards involved were the intensive care unit (ICU) and reanimation unit (REA). The REA has 11 beds, and the ICU has 20 beds. The use of linezolid in both wards began in the year 2002, with use requiring prior approval by the Clinical Pharmacology Department of the hospital. During the first year, the defined daily dose (DDD) was 40 in the ICU and 66 in the

Fig. 1. Restriction analysis of E. faecalis 23S rRNA genes. The mutation G2576T creates a new NheI site, so mutant alleles can be easily differentiated from wild-type alleles that are not cut by this enzyme. The 4 alleles of the 23S rRNA gene were independently amplified as described by Werner et al. (2004). The amplicons were digested with NheI and analyzed by agarose gel electrophoresis. Lanes 1 to 4: alleles 1 to 4 of a wild-type strain (ATCC 29212). Lanes 5 to 8: alleles 1 to 4 of a linezolid-resistant clinical isolate (patient 3, Table 2). M = 1-kb molecular weight marker; m = 100-bp molecular weight marker.

R. Gómez-Gil et al. / Diagnostic Microbiology and Infectious Disease 65 (2009) 175–179

Streptomycin 1000 ≤1000 N1000/R

Erythromycin 1 N2/R

Clindamycin N2 N2/R

Levofloxacin

Quinupristin/ dalfopristin

1

2

N4/R

N2/R

digested DNA as previously described (Saeedi et al., 2002) by REP-PCR fingerprinting using a Diversilab system (bioMérieux Vitek, , Marcy-l'Etoile, France) (Healy et al., 2005).

177

Trimethoprim– sulfamethoxazole

Chloramphenicol

Rifampin

≤1/19

≤8

≤0.5

N2/38 /R

N8/R

≤0.5/S

Among the 12 patients whose records were examined, the median age was 41 years (range, 21–82 years). Seven of the patients were men and 5 women. We defined a case as any patient who received care at the ICU and had at least 1 sample of blood, catheter, urine, surgical wound, skin, or body fluids growing linezolidresistant E. faecalis. 2.3. Identification and antimicrobial susceptibility testing Identification of E. faecalis was performed according to standard bacteriologic procedures; including Gram stain reactions, colony morphology, and catalase production. Biochemical identification tests were carried out by Vitek 2 GPI card performed using the Vitek 2 System (bioMérieux Vitek). Automated susceptibility testing was performed using the AST-P516 cards by Vitek 2® (bioMérieux Vitek) according to manufacturer recommendations. MICs of vancomycin, teicoplanin, and linezolid were confirmed by using E-test® (AB Biodisk). E-test linezolid strips with a concentration gradient corresponding to 0.016 to 256 μg/mL were used with Mueller–Hinton agar as described by the manufacturer (AB Biodisk, Piscataway, NJ). MICs were determined using the Clinical and Laboratory Standards Institute guidelines (National Committee for Clinical Laboratory Standards, 2005). E. faecalis ATCC 29212 was used as a quality control. 2.4. Detection of the G2576T mutation Isolates were analyzed for the presence of the G2576T mutation in the V domain of the 23S rRNA gene by PCRRFLP and pyrosequencing. PCR amplification and restriction enzyme digestion were performed using primers and conditions previously described (Bonora et al., 2006). Primers for the separate amplification of all 4 23S alleles of E. faecalis have been described by Werner et al. (2004). Analysis of the 2576G/T polymorphism by pyrosequencing was done according to Sinclair et al. (2003).

Fig. 2. Pyrosequencing analysis of the G2576 position of 23S rRNA genes of E. faecalis isolates. Sequencing starts at position 2577 on the antisense strand, so the complementary sequence is read. (A) Wild-type clinical isolate sensitive to linezolid; all the alleles contain a C (G in the sense strand). (B) A linezolid-resistant isolate with a 1:3 ratio of wild-type and mutant alleles. (C) The first isolate (patient 1, Table 2) obtained had a 1:1 ratio of wild-type and mutant alleles.

3. Results 3.1. Clinical associations of the selected isolates Twelve linezolid-resistant E. faecalis isolated in cultures from blood samples (n = 3), catheter (n = 2), surgical

178

R. Gómez-Gil et al. / Diagnostic Microbiology and Infectious Disease 65 (2009) 175–179

Table 2 Clinical associations of E. faecalis isolates carrying the G2576T mutation Patient

Ward

Source

Collection date

Previous therapy

Time of treatment with linezolid (days)

1 2 3 4 5 6 7 8 9 10 11 12

ICU REA REA ICU ICU REA ICU ICU ICU ICU ICU REA

Blood Surgical wound Catheter Urine Urine Surgical drainage Skin Blood Urine Blood Catheter Pleural fluid

09/08/2005 11/03/2005 12/05/2005 12/21/2005 02/21/2006 02/27/2006 03/01/2006 06/23/2006 07/07/2006 07/21/2006 09/05/2006 10/10/2006

LND LND LND + TE + LEV + MP + MZ + FL LND LND + LEV + FOS PTC + TS LND + CO + MP + VO + AF LND + VA + MP + CO + FL LND + XL + LEV LND + XL + VA + CO + TO + MP LEV + TX LND + LEV + MP

8 4 15 14 6 0 10 8 8 4 0 5

LND = linezolid; PTC = piperacillin + tazobactam; MP = meropenem; XL = amoxicillin–clavulanic; MZ = metronidazole; LEV = levofloxacin; TS = cotrimoxazole; TX = ceftriaxone; TO = tobramycin; AF = Anfotericina B; FOS = fosfomycin; CO = colimycin; VA = vancomycin; TE = teicoplanin; FL = fluconazole; VO = voriconazole.

wound (n = 1), skin (n = 1), and various body fluids (1 pleural fluid, 1 surgical drainage, and 3 urine) were identified in our laboratory in the period between September 2005 and October 2006. All the isolates were from patients admitted to ICUs of our hospital (8 from ICU and 4 from REA). The mean time from admission to acquisition of E. faecalis was 40 days (range, 1–90 days). MIC values of linezolid were N128 μg/L for the all isolates. The resistance phenotypes were similar for the 12 isolates. All of them were resistant to levofloxacin, ciprofloxacin, clindamycin, erythromycin, and quinupristin–dalfopristin and susceptible to ampicillin, vancomycin, teicoplanin, and rifampin (Table 1). The presence of the most common linezolid resistance mutation, G2576T, was analyzed by PCR-RFLP of individual alleles in 7 isolates (Fig. 1). They possessed the G2576T mutation in alleles 1, 3, and 4. All the isolates were further analyzed by pyrosequencing, showing that the first isolate had G and T at position 2576 in a 1:1 ratio, whereas the remaining ones had a 1:3 ratio (Fig. 2), consistent with the PCR-RFLP results. The 12 isolates under investigation came from different patients but were indistinguishable by PFGE (n = 7) and REP-PCR (n = 12). Eight of them had been isolated from patients that had been previously treated with various combinations of antimicrobial agents, including broadspectrum cephems (ceftriaxone or ceftazidime), imipenem, and colistin. Ten of our patients had received linezolid with a median time of 7 days (3–13 days), whereas the other 2 patients had not been exposed to linezolid previously. The clinical associations of the 12 E. faecalis isolates and a brief analytic description of each of them are reported in Table 2.

4. Discussion Linezolid is the first member in clinical use of a new class of antibiotics, the oxazolidinones, approved for clinical use in 2000. It is active against the majority of clinically

important Gram-positive cocci, including multidrug-resistant Staphylococcus spp., Enterococcus spp., and streptococci. Its use is restricted to hospitals, and it is recommended for treating infections produced by Gram-positive bacteria (including nosocomial pneumonia, infections of the skin, and soft tissue infections). This is the first report of a clonal outbreak of linezolidresistant E. faecalis in Spain. The MIC of linezolid for all the resistant isolates was N128 μg/mL, where the susceptibility breakpoint is 4 μg/L. Prelicensing studies showed low spontaneous mutation rates to linezolid resistance in enterococci (10−9 to 10−11), suggesting that linezolid resistance would be slow to develop. In some reported cases, linezolid resistance emerged in enterococci during therapy (Auckland et al., 2002; Johnson et al., 2002), but emergence of resistance without prior exposure to linezolid has been reported too (Rahim et al., 2003). A review of the clinical literature reveals at least 2 other cases of linezolidresistant Enterococcus sp., in which the patients had not received linezolid, suggesting that resistance may already be present in the enterococcal population, or that there had been a nosocomial spread of resistance (Dibo et al., 2004; Halle et al., 2004; Herrero et al., 2002; Johnson et al., 2002; Kainer et al., 2007; Lobritz et al., 2003; Marshall et al., 2002; Ruggero et al., 2003). In our study, 10 patients had received linezolid treatment before (3–13 days; median, 7 days), whereas the other 2 patients had not been exposed to linezolid. Previously reported patient factors that might predispose to the development of linezolid resistance include indwelling intravascular devices, underdosage, immunosuppression, and long courses of linezolid therapy (20–40 days). We revised clinical, epidemiologic, and microbiologic parameters. In our patients, the risk factors for linezolid-resistant infections were prolonged hospitalization, serious underlying diseases, and multiple cycles of antibiotic treatments. Although it was not possible to trace the exact origin of the outbreak and its subsequent spread, a common denominator

R. Gómez-Gil et al. / Diagnostic Microbiology and Infectious Disease 65 (2009) 175–179

among all patients (12 of 12) was the admission to the ICU or REA, suggesting that in most cases, the microorganism had probably been acquired in the ICUs. Clonal analysis data supported this view. All the isolates carried the G2576T mutation in the 23S rRNA gene and shared the same allelic profile, antibiotic susceptibility phenotype, and molecular fingerprint profiles. Interestingly, in the first isolate, only 2 alleles contained the mutation, whereas in all the others, there were already 3 mutated alleles, although there were no differences in MIC values between the isolates. This suggests that we detected the outbreak early on, when the mutation was still spreading in the genome. The isolates were recovered during an extended period, with nearly 2 years between the first and the last one. The outbreak was controlled after infection control measures were taken to enhance adherence to isolation and contact precautions. The clinical significance of the isolates is difficult to assess because of the critical conditions of the patients. One of them, found on the skin of a patient, was a colonizer, but in the 11 remaining patients, a role in infection was reasonably clear. Emergence of linezolid-resistant Gram-positive cocci is undoubtedly an important issue. To prevent or minimize the emergence of resistance, we should use linezolid strictly after the therapeutic indications, we should keep courses of treatment as short as possible, and we should consider risk factors for resistance development before starting. In addition, we suggest that susceptibility testing of clinically significant Gram-positive pathogens should be done in all cases of treatment failure and, depending on the local epidemiology of each ICU, before starting linezolid treatment. Acknowledgments J.M. was a recipient of a Ramón y Cajal fellowship financed by the European Social Fund and the Ministerio de Educación y Ciencia. References Auckland C, Teare L, Cooke F, et al (2002) Linezolid-resistant enterococci: report of the first isolates in the United Kingdom. J Antimicrob Chemother 50:743–746. Bonora MG, Solbiati M, Stepan E, Zorzi A, Luzzani A, Catania MR, Fontana R (2006) Emergence of linezolid resistance in the vancomycinresistant Enterococcus faecium multilocus sequence typing C1 epidemic lineage. J Clin Microbiol 44:1153–1155. Dibo I, Pillai SK, Gold HS, et al (2004) Linezolid-resistant Enterococcus faecalis isolated from a cord blood transplant recipient. J Clin Microbiol 42:1843–1845. Halle E, Padberg J, Rosseau S, et al (2004) Linezolid-resistant Enterococcus faecium and Enterococcus faecalis isolated from a septic patient: report of first isolates in Germany. Infection 32:182–183. Healy M, Huong J, Bittner T, Lising M, Frye S, Raza S, Schrock R, Manry J, Renwick A, Nieto R, Woods C, Versalovic J, Lupski JR (2005) Microbial DNA typing by automated repetitive-sequence–based PCR. J Clin Microbiol 43:199–207.

179

Herrero IA, Issa NC, Patel R (2002) Nosocomial spread of linezolidresistant, vancomycin-resistant Enterococcus faecium. N Engl J Med 346:867–869. Johnson AP, Tysall L, Stockdale MV, et al (2002) Emerging linezolid resistant Enterococcus faecalis and Enterococcus faecium isolated from two Austrian patients in the same intensive care unit. Eur J Clin Microbiol Infect Dis 21:751–754. Kainer MA, Devasia RA, Jones TF, Simmons BP, Melton K, Chow S, Broyles J, Moore KL, Craig AS, Schaffner W (2007) Response to emerging infection leading to outbreak of linezolid-resistant enterococci. Emerg Infect Dis 13:1024–1030. Kloss P, Xiong L, Shinabarger D, Mankin AS (1999) Resistance mutations in 23 S rRNA identify the site of action of the protein synthesis inhibitor linezolid in the ribosomal peptidyl transferase center. J Mol Biol 294:93–101. Lobritz M, Hutton-Thomas R, Marshall S, et al (2003) Recombination proficiency influences frequency and locus of mutational resistance to linezolid in Enterococcus faecalis. Antimicrob Agents Chemother 47:3318–3320. Marshall SH, Donskey CJ, Hutton-Thomas R, et al (2002) Gene dosage and linezolid resistance in Enterococcus faecium and Enterococcus faecalis. Antimicrob Agents Chemother 46:3334–3336. Meka VG, Pillai SK, Sakoulas G, et al (2004) Linezolid resistance in sequential Staphylococcus aureus isolates associated with a T2500A mutation in the 23S rRNA gene and loss of a single copy of rRNA. J Infect Dis 190:311–317. Multnick AH, Enne V, Jones RN (2003) Linezolid resistance since 2001: SENTRY Antimicrobial Surveillance Program. Ann Pharmacother 37:769–774. National Committee for Clinical Laboratory Standards (2006) Performance standards for antimicrobial susceptibility testing. Document M100-S16. Wayne, PA: NCCLS. Plosker GL, Figgitt DP (2005) Linezolid: a pharmacoeconomic review of its use in serious Gram-positive infections. Pharmacoeconomics 23:945–964. Prystowsky J, Siddiqui F, Chosay J, et al (2001) Resistance to linezolid: characterization of mutations in rRNA and comparison of their occurrences in vancomycin-resistant enterococci. Antimicrob Agents Chemother 45:2154–2156. Rahim S, Pillai SK, Gold HS, Venkataraman L, Inglima K, Press RA (2003) Linezolid-resistant, vancomycin-resistant Enterococcus faecium infection in patients without prior exposure to linezolid. Clin Infect Dis 36:146–148. Ruggero KA, Schroeder LK, Schreckenberger PC, et al (2003) Nosocomial superinfections due to linezolid-resistant Enterococcus faecalis: evidence for a gene dosage effect on linezolid MICs. Diag Microb Infect Dis 47:511–513. Saeedi B, Hällgren A, Jonasson J, Nilsson LE, Hanberger H, Isaksson B (2002) Modified pulse-field gel electrophoresis protocol for typing enterococci. APMIS 110:869–874. Sinclair A, Arnold C, Woodford N (2003) Rapid detection and estimation by pyrosequencing of 23S rRNA genes with a single nucleotide polymorphism conferring linezolid resistance in Enterococci. Antimicrob Agents Chemother 47:3620–3622. Tsiodras S, Gold HS, Sakoulas G, et al (2001) Linezolid resistance in a clinical isolate of Staphylococcus aureus. Lancet 358:207–208. Werner G, Strommenger B, Klare I, Witte W (2004) Molecular detection of linezolid resistance in Enterococcus faecium and Enterococcus faecalis by use of 5′ nuclease real-time PCR compared to a modified classical approach. J Clin Microbiol 42:5327–5331. Wilson P, Andrews JA, Charlesworth R, et al (2003) Linezolid resistance in clinical isolates of Staphylococcus aureus. J Antimicrob Chemother 51:186–188. Zhanel G, Schroeder C, Vercaigne L, et al (2001) A critical review of oxazolidinones: an alternative or replacement for glycopeptides and streptogramins? Can J Infect Dis 12:6.