Diagnostic Microbiology and Infectious Disease 57 (2007) 413 – 418 www.elsevier.com/locate/diagmicrobio

Antimicrobial susceptibility studies

In vitro activity of lysostaphin, mupirocin, and tea tree oil against clinical methicillin-resistant Staphylococcus aureus Kerry L. LaPlante4 Department of Pharmacy Practice, University of Rhode Island, Veterans Affairs Medical Center, Providence, RI 02908, USA Received 26 July 2006; accepted 20 September 2006

Abstract Colonization of methicillin-resistant Staphylococcus aureus (MRSA) commonly leads to infection by the same strain. We examined the activity of lysostaphin, mupirocin, and tea tree oil against clinical MRSA (n = 98) isolates. MIC50 (range) were as follows: lysostaphin, 0.125 mg/L (0.125–0.25); mupirocin, 0.5 mg/L (0.19–1024); tea tree oil, 1024 mg/L (512–2048). High- and low-level mupirocin resistance was noted in 9.2% of our MRSA isolates. Time kill results indicate MRSA activity at 24 h was lysostaphin = gentamicin = vancomycin ( P V .001) N mupirocin N tea tree oil ( P z .05). Checkerboard testing indicated a synergistic relationship between lysostaphin and mupirocin in combination with gentamicin. Antagonism was observed with the combination of vancomycin and tea tree oil; time kill studies confirmed this result. Decolonization options are limited and resistance to mupirocin exists. Lysostaphin and tea tree oil may offer additional therapeutic options for the decolonization of MRSA where current treatment alternatives are limited. D 2007 Elsevier Inc. All rights reserved. Keywords: Lysostaphin; Mupirocin; Tea tree oil; Gentamicin; Vancomycin methicillin-resistant Staphylococcus aureus; Colonize; Decolonize

1. Introduction Methicillin-resistant Staphylococcus aureus (MRSA) is now a public health threat, and little is know regarding effective decolonization therapies and emergence of resistance. Today, infections due to MRSA exist at alarming rates in our hospitals and in our communities, and delineation between community- and hospital-associated MRSA is becoming difficult (Rybak and LaPlante, 2005). Patients who develop these infections have risk factors that include harboring MRSA bacteria in mucosal and epithelial surfaces (Davis et al., 2004; Ellis et al., 2004). When MRSA bacteria are identified in patient’s nares, there is a 10-fold likelihood that these patients will develop an infection from the exact same bacteria strain identified in their nares

A portion of this work was presented at the 45rd Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, DC, December 2005). 4 Tel.: +1-401-273-7100x2339; fax: +1-401-457-3305. E-mail address: [email protected]. 0732-8893/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.diagmicrobio.2006.09.007

(Davis et al., 2004). It is therefore suggested that this bacteria be cleared from the nares of patients before surgery or hospitalization (Dombros et al., 2005; Shrestha et al., 2006; von Eiff et al., 2001; Wertheim et al., 2004). In addition, several outbreaks of community-associated MRSA (CA-MRSA), which are rapid and aggressive colonizers, may further require investigations into additional decolonization options (Ellis et al., 2004; Rybak, 2005). Several agents have been individually evaluated for effectiveness in staphylococcus nasal decolonization. They include: Lysostaphin (cream), a natural enzyme derived from Staphylococcus simulans, which breaks down the S. aureus cell wall (Climo et al., 2001); ! Mupirocin, an antibiotic produced from the bacteria Pseudomonas fluorescens (Cookson, 1998); ! Tea tree oil (terpinen-4-ol, active agent), a product derived from a native Australian plant Melaleuca alternifolia (Carson and Riley, 1995); ! Gentamicin and vancomycin are commonly used antibiotics that demonstrate activity against staphylococci.

!

414

K.L. LaPlante / Diagnostic Microbiology and Infectious Disease 57 (2007) 413 – 418

Most of these compounds have been applied topically to the interior nares of patients or used in oral formulations to decolonize patients who are colonized with drug-resistant S. aureus bacteria (Caelli et al., 2000; Doebbeling et al., 1994; Kokai-Kun et al., 2003; Maraha et al., 2002). To date, there has been no side-by-side comparison in killing activity (defined as, 99.9% kill) of these compounds against MRSA. It is therefore unknown which agent demonstrates the best in vitro activity. In addition, rates of resistance to these compounds against S. aureus are currently unknown. The assessment of killing activity, resistance development, and combination therapy (synergistic versus antagonistic activity) is important when evaluating potential decolonizing options for patients. It is therefore the intent of this study to identify which compound(s) demonstrate good in vitro activity against MRSA bacteria.

2. Materials and methods 2.1. Bacterial isolates Clinical MRSA (n = 98) bacteria were obtained from patients at the Veterans Affairs Medical Center (VAMC) in Providence, RI (June 2004 through December 2005). The bacterial cultures were obtained from blood, nares, sputum, tissue, and urine samples and were provided by Janet Ovalles, Clinical Microbiologist, at the VAMC. Before use, all bacteria were stored in 20% glycerol and frozen at 80 8C. 2.2. Antimicrobial agents Lysostaphin, gentamicin, and vancomycin analytical powder was commercially purchased from Sigma-Aldrich (St. Louis, MO). Lysostaphin powder was stored at 4 8C, and fresh solutions were prepared daily in 0.05 mol/L TrisHCl–0.145 mol/L NaCl. Mupirocin analytical powder was obtained from SmithKline Beecham Pharmaceuticals (Epsom, United Kingdom) and tea tree oil (terpinen-4-ol) was obtained from Plant Life (San Clemente, CA). Mueller– Hinton broth (Difco Laboratories, Sparks, MD) supplemented with 25 mg/L calcium and 12.5 mg/L magnesium were used. Colony counts were determined using tryptic soy agar (TSA, Difco, Becton Dickinson) plates. 2.3. Susceptibility testing MICs and resistance testing were determined in duplicate using serial dilution methodologies described in documents published by the Clinical and Laboratory Standards Institute (CLSI), formerly the National Committee for Clinical Laboratory Standards (2003, 2004). Initial inoculum was 5  105 CFU/mL and subculture volume was 0.01 mL to ensure the accurate determination of the killing end point (Pearson et al., 1980; Taylor et al., 1983). Mupirocin resistance was determined using previously published recommendations defined as susceptible (b 4 mg/L), lowlevel resistance (8–256 mg/L), and high-level resistance ( N512 mg/L) (Kresken et al., 2004).

2.4. Time kill study Time kill experiments with randomly selected MRSA (n = 4) isolates were evaluated. Each time kill experiment was preformed in triplicate. All antimicrobial agents were tested at 4 and 8 times their respective MIC with a starting inoculum of 5  105 CFU/mL using methodologies previously described (LaPlante and Rybak, 2004). Sample aliquots (0.1 mL) were removed from cultures at 0, 4, and 24 h. Antimicrobial carryover was accounted for by serial dilution (10- to 10 000-fold) of plated samples with normal saline. This methodology has a lower limit of detection of 2.0 log10 CFU/mL (Rybak et al., 2005). Growth control wells for each organism were prepared without antibiotic and run in parallel to the antibiotic test wells. For antimicrobial agents not evaluated in combination, bactericidal activity (99.9% kill) was defined as a z 3 log10 CFU/mL reduction in colony count from the initial inoculum at 24 h. Bacteriostatic activity was defined as a b 3 log10 CFU/mL reduction in colony count from the initial inoculum at 24 h, whereas inactive was defined as no observed reductions in the initial inocula. Time to 99.9% kill was determined by linear regression of the sample points if r 2 N .95 or by visual inspection. 2.5. Checkerboard screening test Next, we used the mean fractional inhibitory concentration (FIC) index (AFIC) to screen for potential relationships between the different antimicrobial agents (den Hollander et al., 1998). We randomly selected 2 MRSA isolates to test in combination. For each agent in combination, the AFIC was calculated for each antimicrobial combination as the sum of the individual FICs (Bonapace et al., 2002). The nature of interaction between 2 antibiotics (synergy, indifference, or antagonism) was determined based on FIC index: V 0.5 was considered synergistic, N 0.5 to V 4 was considered indifferent, and N4 was considered antagonistic. Combinations that demonstrated an FIC index of V 0.5 (synergistic) or N4 (antagonistic) were further evaluated in a time kill study as previously described (Rybak et al., 2005). Each organism was tested against each antimicrobial agent alone and in combination. All antimicrobial agents that demonstrated synergy or antagonism using the AFIC method were tested at 2 times their respective MIC. Synergy was defined as an increase in kill of z 2 log10 CFU/mL by combination of antimicrobials versus the most active single agent of that combination at 24 h. Indifference was defined as a 1–2 log10 CFU/mL increase in kill in comparison to the most active single agent. Combinations that resulted in z 1 log10 bacterial growth in comparison to the least active single agent were considered to represent antagonism. The lower limit of detection for this methodology is 2.0 log10 CFU/mL (Rybak et al., 2005). 2.6. Resistance Development of resistance was evaluated at the 24-h time point (time kill) for each of the randomly selected

K.L. LaPlante / Diagnostic Microbiology and Infectious Disease 57 (2007) 413 – 418

415

post hoc test for multiple comparisons. A P value of b.05 indicates statistical significance.

3. Results

Fig. 1. Methicillin-resistant S. aureus (n = 4, average) at (A) 4 MIC and (B) the respective 8 MIC.

isolates. One hundred microliters of samples from the 24-h time points were plated on TSA containing 4-fold the MIC of the respective antibiotic to assess the development of resistance or an increase in MIC. Plates were examined for growth after 24 and 48 h of incubation at 35 8C. When isolates grew on antibiotic containing TSA containing, MIC testing was conducted using CLSI guidelines as previously stated. 2.7. Statistical analysis All statistical analyses were performed using SPSS statistical software (release 14; SPSS, Chicago, IL). After 24 h of exposure to each compound, the bacteria were counted and time to 99.9% kill was compared between groups using 1-way analysis of variance followed by Tukey

Of the 98 clinical MRSA isolates evaluated, 13.3% were derived from blood, 21.4% from nares, 26.5% from tissue, 25.5% from sputum, 8.2% from urine, and 5.1% were from other sites (i.e., ear swab, catheter site, etc.). Three isolates from nasal cultures and 2 isolates from tissue demonstrated high-level mupirocin MIC of N 1024 mg/L. Two isolates from urine specimens and 2 isolates from tissue demonstrated lowlevel MIC of 8–32 mg/L. The remaining isolates demonstrated MIC ranging between 0.094 and 2 mg/L. High-level (5.1%) and low-level (4.1%) mupirocin resistance was noted in 9.2% of our MRSA isolates. The MIC50 of all isolates tested against mupirocin was 0.25 mg/L. For other compounds the MIC50 and MIC range were as follows: gentamicin, 0.5 mg/L (0.125–1); lysostaphin, 0.125 mg/L (0.125–0.25); tea tree oil, 1024 mg/L (512–2048); and vancomycin, 1 mg/L (0.25–1). In a time kill study conducted over 24 h we evaluated 4 MRSA isolates in triplicate at 4 and 8 times the respective MIC. MIC for the isolates tested are as follows: gentamicin, b0.5 mg/L; lysostaphin, b0.25 mg/L; mupirocin, b2 mg/L; tea tree oil, b1024 mg/L; and vancomycin, b1 mg/L. Mupirocin, gentamicin, and vancomycin demonstrated significant in vitro activity at 24 h with the time kill study results at 4 the respective MIC (Fig. 1A). The inoculum decrease for mupirocin was 1.95 F 0.014 log10 CFU/mL, gentamicin 3.345 F 0.072, and vancomycin 3.06 F 0.295. Lysostaphin demonstrated significant bactericidal activity at 4 h but regrew by 24 h with a 2- to 4-fold MIC increase. Initial kill of MRSA was seen with tea tree oil; however, regrowth was observed at 24 h. The MIC for the isolates at the 24-h time point had a 2-fold increase in MIC. At 8 the respective MIC (Fig. 1B), lysostaphin, mupirocin, gentamicin, tea tree oil, and vancomycin demonstrated significant in vitro activity at 24 h for the time kill study results. At 24 h, lysostaphin (decrease, 3.08 F 0.18 log10 CFU/mL), gentamicin (3.26 F 0.09), and vancomycin (3.01 F 0.29) demonstrated bactericidal activity

Table 1 The AFIC calculated for each antimicrobial combination MRSA

Drug

Drug combination

Result

MRSA

Drug

Drug combination

Result

L218

lyso lyso lyso lyso mup mup mup vanco vanco gent

tto vanco gent mup tto vanco gent tto gent tto

Indifference Indifference Synergy Indifference Indifference Indifference Synergy Antagonism Indifference Indifference

L53

lyso lyso lyso lyso mup mup mup vanco vanco gent

tto vanco gent mup tto vanco gent tto gent tto

Indifference Indifference Synergy Indifference Indifference Indifference Synergy Indifference Indifference Indifference

lyso = lysostaphin; tto = tea tree oil; vanco = vancomycin; gent = gentamicin; mup = mupirocin.

416

K.L. LaPlante / Diagnostic Microbiology and Infectious Disease 57 (2007) 413 – 418

Fig. 2. Time kill study of each agent alone and in combination against MRSA (n = 2 averaged).

(99.9% kill). Mupirocin (2.05 F 0.12 log10 CFU/mL) and tea tree oil (0.52 F 0.23) demonstrated bacteriostatic activity. The AFIC was calculated for each antimicrobial combination. Table 1 demonstrates the FIC index results. Results from the FIC index testing were used as a screening tool for further investigations into antimicrobial relationships. Using this screening test, synergy was noted for lysostaphin and gentamicin combinations and mupirocin and gentamicin combinations. Antagonism was noted for vancomycin and tea tree oil. Each of these combination regimens were then tested together in a time kill at 2 times the respective isolate MIC (Fig. 2). Each time kill experiment was performed in triplicate. At 4 h, combination of lysostaphin plus gentamicin demonstrated synergy (difference of 2.898 F 0.13 log10 CFU/mL) against the MRSA isolate, but no difference at 24 h because kill was at the limit of detection. In this same time kill, gentamicin significantly ( P = .04) enhanced mupirocin activity, but, overall, the combination was indifferent. Antagonism was noted with vancomycin and tea tree oil combination with a 1.74 F 0.39 log10 CFU/mL increase in bacterial growth when compared to vancomycin alone. Overall, when agents were evaluated at 2 MIC in combination, no resistance or increase in MIC was noted. 4. Discussion Of the entire human population, S. aureus transiently or persistently colonizes 18–38% of human nares (Claassen et al., 2005; Ellis et al., 2004; Jernigan, 2004). Persons who

harbor this bacteria within their anterior nares are at great risk for the development of infection (Davis et al., 2004). These bacteria residing in the nares serve as a reservoir for future infection. This had been demonstrated in surgical candidates, patients undergoing continuous peritoneal dialysis and hemodialysis, and other groups who are at risk for repeated puncture of the skin (Bloom et al., 1996; Doebbeling et al., 1994; Luzar et al., 1990; Mody et al., 2003; Perl et al., 2002). Studies have demonstrated that these patients are typically infected with the same S. aureus previously found in their nares (Cespedes et al., 2005). Most noteworthy is that MRSA colonization is 10 times more likely to cause an infection than the methicillin-susceptible S. aureus (Davis et al., 2004). Therefore, it is suggested that these patients have MRSA cleared from the nares (Davis et al., 2004; Dombros et al., 2005; Shrestha et al., 2006; von Eiff et al., 2001; Wertheim et al., 2004). Several techniques have been developed and evaluated to assist in the removal of MRSA from colonized individuals. These include use of oral antibiotics, antiseptic body washes, and antibiotics topically placed in the nares. For several reasons, including efficacy, resistance development, and minimization of administering oral antibiotic use, topical intranasal therapies have become the preferred method. Nasal decolonization of methicillin-resistant MRSA is recommended in surgical and dialysis patients. It may also be necessary to evaluate this efficacy in persons colonized with CA-MRSA (Laupland and Conly, 2003; Mody et al., 2003). Mupirocin calcium ointment is the topical antibiotic

K.L. LaPlante / Diagnostic Microbiology and Infectious Disease 57 (2007) 413 – 418

of choice in decolonizing patients from MRSA. High- and low-level mupirocin resistance was noted in 9.2% of our MRSA isolates. We also examined the inhibitory activity of mupirocin against several clinical strains of MRSA and found that lysostaphin, mupirocin, and tea tree oil demonstrate good activity at concentrations exceeding 8 times the respective MIC. In clinical settings, concentrations of lysostaphin 0.5% (150 Ag) in a petrolatum-based cream (Kokai-Kun et al., 2003), mupirocin 2%, and tea tree oil 4% (Caelli et al., 2000) have been evaluated. These concentrations exceed concentrations used in this study and therefore exceed the concentrations necessary to inhibit resistance development in this study. In evaluating the activity of potential decolonizing agents against MRSA in an in vitro assay, the order of activity is lysostaphin = gentamicin = vancomycin ( P V .001) N mupirocin N tea tree oil ( P z .05), whereas gentamicin in combination with mupirocin or lysostaphin may offer additional advantages or prevent resistance. These data suggest that resistance to mupirocin exists, and lysostaphin and tea tree oil may offer additional therapeutic options for the decolonization of MRSA where current treatment alternatives are limited. Further in vitro as well as clinical investigations are warranted. Acknowledgments We thank Kia Lor and Deirdre M. Fuller for technical assistance on this project, and Janet Lane and Janet Ovalles, Clinical Microbiologist, Veterans Affairs Medical Center (Providence, RI), for providing the isolates. This research was supported by a Research Proposal Development Grant from the University of Rhode Island’s Council for Research. References Bloom BS, Fendrick AM, Chernew ME, Patel P (1996) Clinical and economic effects of mupirocin calcium on preventing Staphylococcus aureus infection in hemodialysis patients: a decision analysis. Am J Kidney Dis 27:687 – 694. Bonapace CR, Bosso JA, Friedrich LV, White RL (2002) Comparison of methods of interpretation of checkerboard synergy testing. Diagn Microbiol Infect Dis 44:363 – 366. Caelli M, Porteous J, Carson CF, Heller R, Riley TV (2000) Tea tree oil as an alternative topical decolonization agent for methicillin-resistant Staphylococcus aureus. J Hosp Infect 46:236 – 237. Carson CF, Riley TV (1995) Antimicrobial activity of the major components of the essential oil of Melaleuca alternifolia. J Appl Bacteriol 78:264 – 269. Cespedes C, Said-Salim B, Miller M, Lo SH, Kreiswirth BN, Gordon RJ, Vavagiakis P, Klein RS, Lowy FD (2005) The clonality of Staphylococcus aureus nasal carriage. J Infect Dis 191:444 – 452. Claassen M, Nouwen J, Fang Y, Ott A, Verbrugh H, Hofman A, van Belkum A, Uitterlinden A (2005) Staphylococcus aureus nasal carriage is not associated with known polymorphism in the Vitamin D receptor gene. FEMS Immunol Med Microbiol 43:173 – 176. Climo MW, Ehlert K, Archer GL (2001) Mechanism and suppression of lysostaphin resistance in oxacillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother 45:1431 – 1437.

417

Cookson BD (1998) The emergence of mupirocin resistance: a challenge to infection control and antibiotic prescribing practice. J Antimicrob Chemother 41:11 – 18. Davis KA, Stewart JJ, Crouch HK, Florez CE, Hospenthal DR (2004) Methicillin-resistant Staphylococcus aureus (MRSA) nares colonization at hospital admission and its effect on subsequent MRSA infection. Clin Infect Dis 39:776 – 782. den Hollander JG, Mouton JW, Verbrugh HA (1998) Use of pharmacodynamic parameters to predict efficacy of combination therapy by using fractional inhibitory concentration kinetics. Antimicrob Agents Chemother 42:744 – 748. Doebbeling BN, Reagan DR, Pfaller MA, Houston AK, Hollis RJ, Wenzel RP (1994) Long-term efficacy of intranasal mupirocin ointment. A prospective cohort study of Staphylococcus aureus carriage. Arch Intern Med 154:1505 – 1508. Dombros N, Dratwa M, Feriani M, Gokal R, Heimburger O, Krediet R, Plum J, Rodrigues A, Selgas R, Struijk D, Verger C (2005) European best practice guidelines for peritoneal dialysis. 3 Peritoneal access. Nephrol Dial Transplant 20(Suppl 9):ix8 – ix12. Ellis MW, Hospenthal DR, Dooley DP, Gray PJ, Murray CK (2004) Natural history of community-acquired methicillin-resistant Staphylococcus aureus colonization and infection in soldiers. Clin Infect Dis 39:971 – 979. Jernigan JA (2004) Eradicating nasal carriage of staphylococcus aureus as a strategy for preventing health care-associated infections in nonsurgical patients. Curr Infect Dis Rep 6:417 – 419. Kokai-Kun JF, Walsh SM, Chanturiya T, Mond JJ (2003) Lysostaphin cream eradicates Staphylococcus aureus nasal colonization in a cotton rat model. Antimicrob Agents Chemother 47:1589 – 1597. Kresken M, Hafner D, Schmitz FJ, Wichelhaus TA (2004) Prevalence of mupirocin resistance in clinical isolates of Staphylococcus aureus and Staphylococcus epidermidis: results of the Antimicrobial Resistance Surveillance Study of the Paul-Ehrlich-Society for Chemotherapy, 2001. Int J Antimicrob Agents 23:577 – 581. LaPlante KL, Rybak MJ (2004) Clinical glycopeptide-intermediate staphylococci tested against arbekacin, daptomycin, and tigecycline. Diagn Microbiol Infect Dis 50:125 – 130. Laupland KB, Conly JM (2003) Treatment of Staphylococcus aureus colonization and prophylaxis for infection with topical intranasal mupirocin: an evidence-based review. Clin Infect Dis 37:933 – 938. Luzar MA, Coles GA, Faller B, Slingeneyer A, Dah GD, Briat C, Wone C, Knefati Y, Kessler M, Peluso F (1990) Staphylococcus aureus nasal carriage and infection in patients on continuous ambulatory peritoneal dialysis. N Engl J Med 322:505 – 509. Maraha B, van Halteren J, Verzijl JM, Wintermans RG, Buiting AG (2002) Decolonization of methicillin-resistant Staphylococcus aureus using oral vancomycin and topical mupirocin. Clin Microbiol Infect 8:671 – 675. Mody L, Kauffman CA, McNeil SA, Galecki AT, Bradley SF (2003) Mupirocin-based decolonization of Staphylococcus aureus carriers in residents of 2 long-term care facilities: a randomized, double-blind, placebo-controlled trial. Clin Infect Dis 37:1467 – 1474. NCCLS (2003) National Committee for Clinical Laboratory Standards. 2003. Supplemental tables M100-S13. NCCLS (2004) National Committee for Clinical Laboratory Standards. 2003. Approved standard M7-A6. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Pearson RD, Steigbigel RT, Davis HT, Chapman SW (1980) Method of reliable determination of minimal lethal antibiotic concentrations. Antimicrob Agents Chemother 18:699 – 708. Perl TM, Cullen JJ, Wenzel RP, Zimmerman MB, Pfaller MA, Sheppard D, Twombley J, French PP, Herwaldt LA (2002) Intranasal mupirocin to prevent postoperative Staphylococcus aureus infections. N Engl J Med 346:1871 – 1877. Rybak M, Cha J, Cheung R, Meka M, Kaatz G (2005) Clinical isolates of Staphylococcus aureus from 1987 and 1989 demonstrating heteroge-

418

K.L. LaPlante / Diagnostic Microbiology and Infectious Disease 57 (2007) 413 – 418

neous resistance to vancomycin and teicoplanin. Diagn Microbiol Infect Dis 51:119 – 125. Rybak MJ, LaPlante KL (2005) Community associated methicillin-resistant Staphylococcus aureus: a review. Pharmacotherapy 25:74 – 85. Shrestha NK, Banbury MK, Weber M, Cwynar RE, Lober C, Procop GW, Karafa MT, Gordon SM (2006) Safety of targeted perioperative mupirocin treatment for preventing infections after cardiac surgery. Ann Thorac Surg 81:2183 – 2188. Taylor PC, Schoenknecht FD, Sherris JC, Linner EC (1983) Determination of minimum bactericidal concentrations of oxacillin for Staphylococcus

aureus: influence and significance of technical factors. Antimicrob Agents Chemother 23:142 – 150. von Eiff C, Becker K, Machka K, Stammer H, Peters G (2001) Nasal carriage as a source of Staphylococcus aureus bacteremia. Study Group. N Engl J Med 344:11 – 16. Wertheim HF, Vos MC, Ott A, Voss A, Kluytmans JA, VandenbrouckeGrauls CM, Meester MH, van Keulen PH, Verbrugh HA (2004) Mupirocin prophylaxis against nosocomial Staphylococcus aureus infections in nonsurgical patients: a randomized study. Ann Intern Med 140:419 – 425.