| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 R. M. Alden Research Laboratory, Santa Monica, CA 90404, USA
2 UCLA School of Medicine, Los Angeles, CA 90095, USA
3 Anti-Infective Research Laboratory, College of Pharmacy and Health Sciences, Wayne State University, Detroit, MI 48201, USA
Correspondence
Ellie J. C. Goldstein
ejcgmd{at}aol.com
Received 20 August 2007
Accepted 29 November 2007
-haemolysin production. For comparison, 25 strains of outpatient meticillin-susceptible S. aureus (MSSA), 24 strains of healthcare-associated (HA)-MRSA and six historical strains of vancomycin-intermediate S. aureus (VISA) were included. It was found that 21/25 CA-MRSA strains tested were PVL-positive, SSCmec type 4 and agr type 1, whilst 4/25 were PVL-negative, SSCmec type 2 and agr type 2. Two of the agr type 2 strains were negative for -haemolysin but all other strains were positive. Moxifloxacin MIC50/90 values (µg ml–1) were 1/8 for CA-MRSA, 4/32 for HA-MRSA and 
0.03/1 for MSSA and MIC50 of 2 for VISA. The D-test for inducible clindamycin resistance was positive for 3/27 CA-MRSA, 5/14 HA-MRSA and none of the MSSA isolates. In chequerboard studies, fractional inhibitory concentration indices (FICIs) showed that most interactions were additive or indifferent (FICI value >0.5 to 2) as follows: rifampicin 43/52 strains, clindamycin 44/44, SXT 44/47, trimethoprim 41/42 and vancomycin 37/43. The FICI values for doxycycline were 3–6 for 32/34 strains, indicating antagonism, suggesting that it should not be used in combination with moxifloxacin.
Abbreviations: CA, community-associated; FICI, fractional inhibitory concentration index; HA, healthcare-associated; MRSA, meticillin-resistant Staphylococcus aureus; MSSA, meticillin-susceptible Staphylococcus aureus; PVL, Panton–Valentine leukocidin; SCCmec, staphylococcal cassette chromosome mec; SSTI, skin and soft tissue infection; SXT, sulfamethoxazole/trimethoprim; VISA, vancomycin-intermediate Staphylococcus aureus.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Jarraud et al. (2002) noted that CA-MRSA SSTIs are potentially more serious than SSTIs caused by other strains and that virulence factors such as Panton–Valentine leukocidin (PVL) may be a link between phylogenetics and virulence. In addition, Sakoulas et al. (2005) noted that there are ‘sparse data’ about associations of agr type II CA-MRSA with ‘infections with particular infectious syndromes’. Consequently, we selected the first 25 of our 33 CA-MRSA isolates, and determined the virulence genes and expression of PVL, SSCmec type, agr type and -haemolysis production.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Determination of MICs and chequerboard synergy studies were carried out in cation-adjusted Mueller–Hinton broth with moxifloxacin alone and in combination with doxycycline, rifampicin, clindamycin, trimethoprim, sulfamethoxazole/trimethoprim (SXT) and vancomycin. MICs without synergy studies were determined for levofloxacin and erythromycin (CLSI, 2006). Moxifloxacin concentrations were 0.03–4 µg ml–1 plus each of the other agents at four or five clinically appropriate concentrations. The fractional inhibitory concentration index (FICI) was calculated for all combinations (Krogstad & Moellering, 1986). Erythromycin-resistant, clindamycin-susceptible strains were tested for inducible resistance by the D-test. For quality control, we used S. aureus ATCC 29213 and ATCC 43300.
Staphylococcal cassette chromosome mec (SCCmec) type was determined by multiplex PCR methods as described previously by Zhang et al. (2005). The PVL genes, lukS-PV and lukF-PV, were detected by PCR using primers described by Lina et al. (1999). Identification of agr group was accomplished using specific primers. Clinical S. aureus isolates were studied using PCR and agr group-specific primers as follows: agr group I, 5'-CACTTATCATCAAAGAGCC-3' and 5'-CCACTAATTATAGCTGG-3; agr group II, 5'-GTAGAGCCGTATTGATTCC-3' and 5'-TATTTCATCTCTTTAAGG-3'; agr group III, agrC of class III: 5'-CTGCATTTATTAGTGGAATACG-3' and 5'-GTTTCATTTCTTTAAGAG-3'; agr group IV, 5'-GCCGCGAGCTTGGGAG-3' and 5'-GAAGATACGTGGCAAACTGGT-3'. After amplification for 30 cycles (30 s of denaturation at 94 °C, 30 s of annealing at 55 °C and 1 min of extension at 72 °C), the PCR products were resolved by electrophoresis through 1.5 % agarose gels followed by ethidium bromide staining and analysis. S. aureus ATCC 49775 was used as a positive control strain for all PCR runs. Expression of the agr gene cluster was determined by -haemolysin production using a strain of S. aureus (RN4220) that produces a large zone of β-haemolysis without interference by 
- or -haemolysins. Briefly, RN4220 was streaked vertically and test strains were streaked perpendicular to it on sheep blood agar plates. S. aureus strains RN6607 and RN9120 were used as positive and negative controls, respectively. Evidence for enhanced haemolysin production was examined at the intersection of growth of the two organisms (Sakoulas et al., 2003).
|
RESULTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The moxifloxacin MIC50/90 values (µg ml–1) were 1/8 for CA-MRSA, 4/32 for HA-MRSA and 0.03/1 for MSSA, and MIC50 of 2 for VISA. The D-test was positive for 3/27 (11 %) CA-MRSA strains that were resistant to erythromycin but appeared to be susceptible to clindamycin, 5/14 (36 %) of the HA-MRSA strains and none of the MSSA strains.
|
2); however, the FICI values for doxycycline were 3–6 for 32/34 strains, indicating antagonism. Approximately 10 % of moxifloxacin-resistant strains (MIC >0.5 µg ml–1) exhibited a reduction in their MIC to 0.5 µg ml–1 with the addition of clindamycin and SXT and 
25 % with the addition of vancomycin, rifampicin and trimethoprim alone. There was no reduction with the addition of doxycycline.
Of the 25 CA-MRSA strains tested, 21 (84 %) were PVL-positive, SSCmec type 4 and agr type 1, suggesting a dominant geographical clone. Four out of 25 (16 %) were PVL-negative, SSCmec type 2 and agr type 2. Two of these four were negative for -haemolysin production, but all other strains were positive.
|
DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our in vitro data showed that moxifloxacin was two to four times more active than levofloxacin on a weight basis, which might be the result of a single mutation in grlA or gyrA, as postulated by Noguchi et al. (2005). As the time above the MIC is a key parameter in the killing of S. aureus, the protein binding of moxifloxacin versus levofloxacin (30–50 % vs 24–38 %) and the time above the MIC (moxifloxacin has a half-life of 10–14 h vs 7 h for levofloxacin) would give advantage to moxifloxacin (although one should only consider free drug concentrations). Clinically, several studies have suggested the utility of a fluoroquinolone plus rifampicin for the therapy of MSSA infections. Zimmerli et al. (1998) treated 33 patients with ciprofloxacin plus rifampicin for orthopaedic prosthesis infections and had a 100 % cure rate in the experimental group compared with 58 % in the control group (ciprofloxacin plus placebo) (P <0.02). Schrenzel et al. (2004) treated 104 patients with deep-seated S. aureus infections, mostly MSSA, and found that a fluoroquinolone plus rifampicin was effective in 78 % of cases. Our data support these studies and suggest that moxifloxacin plus rifampicin would be effective for MSSA infections.
In our study, 15.2 % of CA-MRSA isolates were susceptible to 0.5 µg moxifloxacin ml–1 and 63.6 % to 1.0 µg moxifloxacin ml–1. It is important to note that there is disparity between Food and Drug Administration and CLSI suggested breakpoints, but, despite in vitro activity, clinically moxifloxacin is not indicated for use in any MRSA infection, and prudent practice would suggest the co-administration of a second agent if MRSA were suspected. In our study, up to 29 % of moxifloxacin-resistant isolates decreased their moxifloxacin MICs to 0.5 µg ml–1 with the addition of a second agent, especially trimethoprim alone.
We found that 100 % of our CA-MRSA isolates were susceptible to doxycycline, SXT, trimethoprim, vancomycin and rifampicin, and 93.9 % to clindamycin, but only 9.1 % to erythromycin. This is similar to the results of Moran et al. (2006), who reported that 95 % of their MRSA isolates were susceptible to clindamycin, 6 % to erythromycin, 60 % to fluoroquinolones, 100 % to SXT and 92 % to tetracyclines. In contrast, Buck et al. (2005) noted that 201/517 SSTI CA-MRSA isolates (39 %) from patients in Minnesota were susceptible to erythromycin and 433/517 (84 %) to ciprofloxacin, whilst susceptibility to other agents was more typical (clindamycin 92 %, rifampicin 99 %, tetracycline 91 % and SXT 99 %). However, they found that, unlike CA-MRSA isolates, only 77 % (44/57) of invasive HA-MRSA isolates were susceptible to clindamycin.
Differences between the susceptibilities of MRSA and MSSA to moxifloxacin were noted by Metzler et al. (2004), who found that moxifloxacin was more potent (MIC90 0.06 mg l–1) than levofloxacin (MIC90 0.25 mg l–1) against MSSA isolates, and this was also true at mutation-prevention concentrations (MPCs) (moxifloxacin 0.25 mg l–1 vs levofloxacin 1 mg l–1); however, MIC90 and MPC90 were higher for both drugs against the MRSA isolates. Metzler et al. (2004) did not look at the comparative resistance prevention of moxifloxacin versus levofloxacin in combination with SXT or rifampicin. In our study, moxifloxacin MIC50/90 values were lowest for MSSA, intermediate for CA-MRSA and highest for HA-MRSA; except for levofloxacin, the activity of the other agents against HA-MRSA was similar to that for CA-MRSA. In contrast, moxifloxacin was active against 23/25 (92 %) of our MSSA isolates at <1 µg ml–1 (range 0.03–4); clindamycin and erythromycin were active against 96 and 76 %, respectively.
For the six VISA strains that we tested, the moxifloxacin MICs were 1–4 µg ml–1. SXT was active against 50 %, doxycycline against 83 % and rifampicin against 33.3 % of strains. All were clindamycin- and erythromycin-resistant. This suggests that the VISA strains are multidrug-resistant and that selection of antimicrobial therapy must be based on individual susceptibility testing.
Our study shows that most interactions with moxifloxacin against MRSA were either additive or indifferent (FICI >0.5 to 2) for rifampicin, clindamycin, SXT, trimethoprim and vancomycin. Of note is the extremely high activity of rifampicin alone against all MRSA and MSSA isolates; future studies to determine whether or not moxifloxacin prevents the emergence of rifampicin-resistant S. aureus would be of interest. The FICI values for doxycycline were 3–6 for 32/34 strains, indicating antagonism and suggesting that this combination should be avoided.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. N. Leonard, G. W. Kaatz, L. R. Rucker, and M. J. Rybak Synergy between gemifloxacin and trimethoprim/sulfamethoxazole against community-associated methicillin-resistant Staphylococcus aureus J. Antimicrob. Chemother., December 1, 2008; 62(6): 1305 - 1310. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
