Inhibitory and Bactericidal Activities of Daptomycin, Vancomycin, and Teicoplanin against Methicillin-Resistant Staphylococcus aureus Isolates Collected from 1985 to 2007

The inhibitory and bactericidal activities of daptomycin, vancomycin,and teicoplanin against a collection of 479 methicillin-resistantStaphylococcus aureus isolates were assessed. The isolates werecollected from U.S. and European hospitals from 1985 to 2007and were primarily from blood and abscess cultures. The MICsand minimum bactericidal concentrations (MBCs) of the threeagents were determined, and the MBC/MIC ratios were calculatedto determine the presence or absence of tolerance. Tolerancewas defined as an MBC/MIC ratio of $≥$ 32 or an MBC/MIC ratio of $≥$ 16 when the MBC was greater than or equal to the breakpointfor resistance. Tolerance to vancomycin and teicoplanin wasobserved in 6.1% and 18.8% of the strains, respectively. Toleranceto daptomycin was not observed.

INTRODUCTION

Top

INTRODUCTION

Although vancomycin and teicoplanin are the standard therapiesfor staphylococcal bacteremia, tolerance to vancomycin and teicoplaninhas been demonstrated in both coagulase-negative staphylococciand Staphylococcus aureus as well as in various Streptococcusspecies (2, 3, 7, 10, 13, 15, 20, 21, 23, 25). Daptomycin, alipopeptide antibiotic, has been demonstrated to have rapidbactericidal activity against gram-positive bacteria, includingmethicillin-resistant Staphylococcus aureus (MRSA), and toleranceto this drug has not been demonstrated (2, 9, 10, 19, 21, 24,26, 28).

The issue of antibiotic tolerance is a complicated one. Somestudies have suggested that infections caused by tolerant strainsmay be more difficult to treat, especially when they cause complicatedinfections such as endocarditis, meningitis, or osteomyelitisor cause infections in immunocompromised patients (7, 8, 14,15, 16, 18, 20, 22, 23, 25). Other investigators' expert analysesdo not agree that there is proof of a correlation between tolerantstrains and treatment failures or that bactericidal activityis required for the treatment of serious MRSA infections (17,25, 26, 27, 28). Controversy concerning the appropriate methodsfor the determination of tolerance in clinical isolates andin the practicality of testing isolates for tolerance in theclinical laboratory also exists.

This study looked at MRSA isolates obtained primarily from bloodand abscess cultures collected between 1985 and 2007. The mainpurpose of the study was to determine the in vitro inhibitoryand bactericidal activities and the level of tolerance to thethree drugs observed by standardized MIC and minimum bactericidalconcentration (MBC) tests (4, 5, 19).

(This study was presented in part at the 47th Interscience Conferenceon Antimicrobial Agents and Chemotherapy, Chicago, IL, 17 to20 September 2007.)

MATERIALS AND METHODS

Top

MATERIALS AND METHODS

Microorganisms. A collection of 479 MRSA isolates, primarily from blood andabscess sources of infection, collected from U.S. and Europeanhospitals between 1985 and 2007 were tested. The isolates wereunique and nonconsecutive and were collected for a variety ofstudies. All of the strains were susceptible to daptomycin,vancomycin, and teicoplanin and resistant to oxacillin by previousMIC testing. Vancomycin-intermediate and -resistant strainswere excluded from the test set.

Susceptibility testing. MICs were determined by broth microdilution, in accordance withthe guidelines of the Clinical and Laboratory Standards Institute(formerly the National Committee for Clinical and LaboratoryStandards) (5, 6). Daptomycin (Cubist Pharmaceuticals), vancomycin(Sigma-Aldrich), and teicoplanin (Molcan Corp) were tested atdilution ranges of 0.015 to 512 µg/ml. Standard cation-adjustedMueller Hinton Broth (CAMHB) was used for vancomycin and teicoplaninMIC testing. CAMHB containing 50 mg/liter of calcium and 11mg/liter of magnesium was used for daptomycin MIC testing. Quantitativecolony counts were performed from the growth control well ofeach microdilution panel.

MBCs were determined in accordance with the guidelines of theClinical and Laboratory Standards Institute (4) and the workof Peterson and Shanholtzer (19). The entire volume (0.1 ml)of the MIC well and the wells with 4 dilutions above the MICwere spread across the center of a blood agar plate and allowedto dry for 20 min. After the plates had dried, a sterile spreadingrod was used to evenly disperse the inoculum over the entiresurface of the plate, which was then incubated at 35 to 37°Cfor 24 to 48 h. The MBCs were recorded as the lowest dilutionthat produced a $≥$ 99.9% reduction in growth ( $≥$ 3-log₁₀ reductionin CFU/ml) in comparison to the growth of the control.

RESULTS

Top

RESULTS

Daptomycin was the most potent antibiotic evaluated, as determinedfrom the MIC₅₀ and the MIC₉₀ (0.5 and 0.5 µg/ml, respectively);vancomycin and teicoplanin had MIC₅₀s of 1 and 1 µg/ml,respectively, and MIC₉₀s of 0.5 and 1 µg/ml, respectively(Table 1). Daptomycin was also highly bactericidal, as determinedfrom its MBC₅₀ and MBC₉₀ (0.5 and 1 µg/ml), followed byvancomycin and teicoplanin, for which the MBC₅₀s were 1 and2 µg/ml, respectively, and the MBC₉₀s were 1 and 32 µg/ml,respectively (Table 1).

View this table:
[in this window]
[in a new window]

TABLE 1. MIC₅₀s, MIC₉₀s, MBC₅₀s, MBC₉₀s, and percent tolerance for 479 isolates collected from 1985 to 2007

Tolerance, defined as an MBC/MIC ratio of $≥$ 32 or an MBC/MIC ratioof $≥$ 16 with an MBC in the resistant range, was evaluated (2,13). The overall tolerance rates for vancomycin and teicoplaninwere 6.1% (29/479 isolates) and 18.8% (90/479 isolates), respectively.No tolerance was observed for daptomycin (Tables 1 and 2). VancomycinMBC/MIC ratios were 1 for 84.5% of the strains and $≤$ 2 for 92.5%of the strains, while the teicoplanin MBC/MIC ratios were 1for 49.3% of the strains and $≤$ 2 for 74.3% of the strains (Table2). Daptomycin MBC/MIC ratios were 1 for 87.5% of the strainsand $≤$ 2 for 100% of the strains (Table 2).

View this table:
[in this window]
[in a new window]

TABLE 2. MBC-to-MIC ratios for daptomycin, vancomycin, and teicoplanin across all study years

The MICs for vancomycin and teicoplanin were not predictiveof tolerance in this study, similar to the findings of a studyof strains from 30 bacteremic patients reported by Sakoulaset al. (23). The incidence of tolerance at each MIC increasedas the MIC increased (Tables 3 and 4). Among the isolates inthe group of 29 vancomycin-tolerant isolates, 24 (82.8%) werealso tolerant to teicoplanin (Table 5). The highest daptomycinMIC in this group was 1 µg/ml (one isolate), and two isolateshad MBC/MIC ratios of 2 (data not shown).

View this table:
[in this window]
[in a new window]

TABLE 3. Evidence of vancomycin tolerance at each MIC

View this table:
[in this window]
[in a new window]

TABLE 4. Evidence of teicoplanin tolerance at each MIC

View this table:
[in this window]
[in a new window]

TABLE 5. Comparative MBC/MIC ratios of 29 vancomycin-tolerant strains

Twenty-four of 90 teicoplanin-tolerant isolates (26.7%) werealso vancomycin tolerant (Table 6). Among the isolates in thisgroup, 12 (15.4%) isolates had daptomycin MICs of 1 µg/ml,while the other 78 strains had daptomycin MICs of $≤$ 0.5 µg/ml.Twelve of the strains tolerant to teicoplanin had daptomycinMBC/MIC ratios of only 2 (data not shown).

View this table:
[in this window]
[in a new window]

TABLE 6. Comparative MBC/MIC ratios of 90 teicoplanin-tolerant strains

DISCUSSION

Top

DISCUSSION

Since tolerance to daptomycin was not exhibited in vitro inthis study or in other studies, it has been suggested that daptomycinremains an alternative to standard therapy for bacteremia, endocarditis,and other life-threatening infections caused by S. aureus (2,8, 9, 10, 14, 26, 27). For vancomycin, studies looking at toleranceby the use of MBC or time-kill studies have shown a varietyof results. The rates of tolerance to vancomycin have rangedfrom 0% (12) to as high as 47% (15). Biedenbach et al. (2) reporteda 3.2% rate of tolerance to vancomycin and a 31.6% rate of toleranceto teicoplanin for 76 MRSA isolates from a collection of SENTRYAntimicrobial Surveillance Program strains collected from eightmedical centers in the Asia-Pacific region, while Jones (13)reported a rate of tolerance to vancomycin of 15% for 105 wild-typeMRSA isolates collected from medial centers across six continentsfor inclusion in the 1997-2003 SENTRY Antimicrobial SurveillanceProgram database. There are fewer data in the literature ontolerance to teicoplanin; however, isolates have been shownto exhibit more tolerance to teicoplanin than to vancomycinin previous studies (2, 15).

The issue of tolerance on the basis of bactericidal test resultsremains complicated, since technical factors have been knownto affect the results and tolerance measured in vitro may notalways translate to tolerance in vivo (4, 17, 21). The two typesof tests used to determine the presence of tolerance are theMBC and the time-kill assays. Both tests use the same definitionof bactericidal activity: a reduction of the CFU by 99.9% ( $≥$ 3-log₁₀-unitreduction in the numbers of CFU/ml) from the starting inoculum.Isolates are considered tolerant to antimicrobial agents thatare known to be bactericidal but that do not show killing atan MBC/MIC ratio of $≥$ 32 or an MBC/MIC ratio of $≥$ 16 with an MBCabove the MIC resistance breakpoint (4, 9, 13, 15, 18, 19).May et al. (15) found the two methods to be comparable. TheMBC test has been standardized, can be performed in the clinicallaboratory, and, if it is done correctly, has been shown tobe reproducible (4, 11, 19). Because of the relative ease ofperformance, the MBC test can be one of the tools that clinicianscan use to make treatment decisions for individual patients(11, 19).

Conclusions. Daptomycin was more potent in vitro than either vancomycin orteicoplanin against MRSA according to its MIC₉₀ and was morebactericidal according to its MBC₉₀ and MBC/MIC ratios. A totalof 6.1% (29/479) and 18.8% (90/479) of the strains tested exhibitedtolerance to vancomycin and teicoplanin, while tolerance todaptomycin was not observed for any of the 479 isolates. Twenty-four(5%) of all strains were tolerant to both vancomycin and teicoplanin.There have been a variety of rates of vancomycin and teicoplanintolerance reported in the literature; one explanation for thevariability might be that there has been a wide variety in theorganism sets that were studied. In addition there has beenvariability in the MBC and/or time-kill methods used to determinetolerance rates. Finally, the MICs of vancomycin and teicoplaninwere not predictive of tolerance; however, the percentage ofstrains exhibiting tolerance increased as the MICs of both vancomycinand teicoplanin increased in this study.

ACKNOWLEDGMENTS

Funding for this work was provided by Cubist Pharmaceuticals,Inc.

FOOTNOTES

* Corresponding author. Mailing address: Clinical Microbiology Institute, 9725 SW Commerce Circle, Suite A-1, Wilsonville, OR 97070. Phone: (503) 682-3232. Fax: (503) 682-2065. E-mail: mtrac{at}clinmicroinst.com

Published ahead of print on 17 February 2009.

REFERENCES

Top