Piperacillin v Carbenicillin in the therapy for serious infections.

One hundred seven patients were treated with either piperacillin (56) or carbenicillin (51) in an open randomized trial of hospitalized patients with pleuropulmonary (40), urinary tract (26), gynecologic (21), skin and soft-tissue (eight), joint (five), bone (three), and miscellaneous other infections (four). Patients with urinary tract infections were given 150 mg/kg/day of piperacillin sodium or 200 mg/kg/day or carbenicillin sodium in divided doses every six hours intravenously. Patients with other infections were given 250 mg/kg/day of piperacillin sodium and 450 mg/kg/day of carbenicillin sodium; 53/56 (95%) patients treated with piperacillin and 45/51 (88%) patients treated with carbenicillin were cured clinically. In general, the drugs were well tolerated. There were, however, more adverse experiences in the groups taking carbenicillin. Of special interest was the finding of liver function test abnormalities in 17/78 (21%) carbenicillin recipients (evaluative and nonevaluative cases). We concluded that piperacillin was effective and safe. It has potential for use in a great variety of infections.

Aztreonam plus clindamycin as therapy for pelvic infections in women.

Aztreonam, a new monobactam antibiotic with specific gram-negative aerobic activity, was used in combination with clindamycin in the treatment of 40 women with pelvic infection, including post-partum endometritis, pelvic inflammatory disease, and post-hysterectomy pelvic cellulitis. Clinical cure was achieved in 87 percent of patients. Failure was related to the limited gram-positive aerobic spectrum of clindamycin. All aerobic gram-negative enteric organisms were sensitive in vitro to less than 0.125 microgram/ml of aztreonam.

Safety and effectiveness of ticarcillin plus clavulanic acid in the treatment of community-acquired acute pyelonephritis in adult women.

An open study was conducted to evaluate the safety and effectiveness of ticarcillin plus clavulanic acid in the treatment of 18 adult women with community-acquired Escherichia coli acute pyelonephritis. Eleven of the 18 patients had a history of urinary tract infections, primarily acute pyelonephritis. Six patients had blood culture results positive for E. coli in addition to positive urine culture results. Ticarcillin plus clavulanic acid was administered as the 3.2 g formulation to 11 patients and as the 3.1 g formulation to the other seven. The mean duration of treatment was 9.2 days. Five of the 18 (28 percent) patients had clinical and bacteriologic cures; there were 11 (69 percent) relapses or reinfections and two (11 percent) clinical failures. Adverse reactions (all reversible) were reported in 11 (61 percent) patients, the most frequent of which was phlebitis. These results should prompt further investigation into the treatment of acute pyelonephritis with ticarcillin plus clavulanic acid versus other antibiotics.

Comparison of ticarcillin plus clavulanic acid with cefoxitin in the treatment of female pelvic infection.

Ninety-three female patients with post-cesarean endometritis, post-hysterectomy pelvic cellulitis, and other miscellaneous moderately severe pelvic soft-tissue infections were treated in a randomized fashion with either ticarcillin plus clavulanic acid or cefoxitin. Of the 47 patients treated with ticarcillin plus clavulanic acid, 38 had clinical cures, four showed improvement, therapy failed in three, and two were nonevaluable, for a failure rate of 6.7 percent. Of the 46 patients treated with cefoxitin, 33 had clinical cures, five showed improvement, therapy failed in seven, and one was nonevaluable, for a failure rate of 15.6 percent. Bacteriologically, the addition of clavulanic acid to ticarcillin was found to broaden the antibacterial spectrum to include some Escherichia coli, most Klebsiella, many coagulase-negative staphylococci, and all isolates of Staphylococcus aureus. Adverse reactions were few, with only one patient having therapy with cefoxitin discontinued because of side effects. It is concluded that ticarcillin plus clavulanic acid is quite suitable for antibiotic therapy of female pelvic soft-tissue infection, based on the (expanded) coverage of both aerobic and anaerobic bacterial species.

Use of single-agent antimicrobial therapy in the treatment of polymicrobial female pelvic infections.

One hundred twenty patients with either postpartum endomyometritis or postgynecologic surgical infections were treated either with ticarcillin, clindamycin, or chloramphenicol. One hundred nine (91%) responded successfully to single-agent antimicrobial therapy. Most of the infections were polymicrobial, involving both aerobic and anaerobic bacteria. Forty patients were treated with ticarcillin, with 90% responding successfully; 48 were treated with chloramphenicol, with 94% responding successfully; and 32 were treated with clindamycin, with 88% responding successfully. Single-agent antimicrobial therapy appears to be appropriate for treating polymicrobial obstetric and gynecologic soft tissue infections.

Cefotaxime in the treatment of staphylococcal infections. Comparison of in vitro and in vivo studies.

Staphylococcus aureus strains are well-established pathogens that may cause mild to serious life-threatening disease. Coagulase-negative staphylococci, particularly Staphylococcus epidermidis, also have a pathogenic role in humans and cause infections primarily associated with prosthetic devices and indwelling catheters, whereas Staphylococcus saprophyticus usually causes urinary tract infections. Cefotaxime is a "third-generation" cephalosporin that is stable to the staphylococcal beta-lactamases. In vitro studies over the last 15 years have shown that this parenteral cephalosporin has remained highly active (MIC90 ranges of < or = 2-8 micrograms/ml) against oxacillin-susceptible staphylococci. Cefotaxime therapy of staphylococcal infections has resulted in clinical cure/improvement rates ranging from 78%-100% and bacteriologic eradication rates ranging from 85%-100% in a wide variety of infections. Contrary to contemporary dogma, this "third-generation" cephalosporin appears to be efficacious against staphylococcal infections from a review of 15 years of clinical experience.

Comparison of the in vitro action and interaction of cefotaxime and desacetylcefotaxime against clinical isolates of Bacteroides spp.

Desacetylcefotaxime (dCTX), the in vivo metabolite of cefotaxime (CTX), possess significant in vitro antimicrobial activity similar to the parent compound against a variety of aerobic and anaerobic bacteria. In vitro susceptibility studies showed that CTX inhibited 86% of 473 strains of the Bacteroides fragilis at a concentration of 32 micrograms/ml while dCTX inhibited 91% of the test isolates at the same concentration. Strains of the B. fragilis species were significantly more susceptible to CTX than were the non-B. fragilis species. Susceptibility testing of CTX and dCTX in a 1:1 ratio produced significantly more inhibitory activity, especially against the non-B. fragilis strains. Synergy studies showed that the interaction of CTX and dCTX was either completely or partially synergistic against 85% of 92 test organisms. The presence of dCTX was also shown to lower the CTX MIC values four-fold or greater in 82% of the synergy studies. Synergy was noted against strains of the B. fragilis group, B. melaninogenicus group, B. bivius, B. disiens, and B. capillosus. Through the use of time-kill kinetics studies, the interaction of CTX and dCTX was shown to be additive at subinhibitory and inhibitory concentrations and suggestedly synergistic at suprainhibitory concentrations against strains of the B. fragilis group. These in vitro studies demonstrate that dCTX increases the inhibitory and bactericidal activity of CTX when tested in combination.

Cross-resistance to beta-lactam-beta-lactamase inhibitor combinations and clindamycin among cefoxitin-resistant and cefoxitin-susceptible strains of the Bacteroides fragilis group.

In the present study the cross-resistance rates of cefoxitin-resistant and cefoxitin-susceptible strains of the Bacteroides fragilis group were compared with regard to beta-lactam-beta-lactamase inhibitor combinations and clindamycin. Piperacillin-tazobactam was the most active agent tested with an overall resistance rate of 0.2% and no resistance among cefoxitin-resistant strains. Ticarcillin-clavulanate was the least active combination with an overall resistance rate of 1.6% but a resistance rate of 13.1% among cefoxitin-resistant strains. For ampicillin-sulbactam, amoxicillin-clavulanate, and clindamycin resistance, rates of cefoxitin-resistant versus cefoxitin-susceptible strains were 9.1% and 0.4%, 8.3% and 0.9%, and 28% and 12.9%, respectively.

Anaerobic susceptibility testing. Slight differences in inoculum size can make a difference in minimum inhibitory concentrations.

In the present study, we compared the actual inoculum density from inoculated broth microdilution wells to the targeted inoculum size (10(5) CFU/well) when the inoculum was prepared using a McFarland nephelometer to achieve the standard density. Three target inoculum sizes (10(5), 5 x 10(5), and 10(6) CFU/well) were used to compare the effect of slight inoculum size increases of both ATCC and clinical strains of anaerobes on MICs of various antimicrobials. Actual colony counts of Bacteroides fragilis, Bacteroides thetaiotaomicron, Eubacterium lentum, and Veillonella parvula ranged from 0.7 x 10(5) to 1.4 x 10(5) CFU/well. As the inoculum size rose above the desired 10(5) CFU/well level, the MICs of certain antimicrobials became elevated. Ceftizoxime, cefotaxime, and ceftriaxone MICs rose 4- to 16-fold with as little as 0.5 log10 increase in inoculum size. Other increases were also noted with E. lentum and Clostridium perfringens, but were primarily between the low and high inoculum sizes. Results with cefoxitin, cefotetan, mezlocillin, and imipenem did not show an appreciable increased inoculum effect. This study demonstrates that the variation in organism size among anaerobes (both between species and within species) does make a difference in actual inoculum size and certain anaerobes may require special adjustment to ensure proper MIC results from susceptibility testing.

Major methodology-dependent discordant susceptibility results for Bacteroides fragilis group isolates but not other anaerobes.

Two standardized susceptibility test methods, a broth microdilution (BMD) and agar dilution (AD) method were performed on a total of 441 clinical isolates of anaerobes with ceftizoxime, cefotaxime, ceftriaxone, cefoxitin, piperacillin, and metronidazole. Against the 339 strains of the Bacteroides fragilis group BMD minimum inhibitory concentration (MIC) values were lower than those from AD testing for all the beta-lactams. Overall for the B. fragilis group and the beta-lactams, the mode MIC values were two- to 64-fold lower, and the MIC50 values two- to eightfold lower. Resistance rates were 11%-28% higher overall with AD results and were higher especially for non-B. fragilis species. For non-Bacteroides anaerobes no major discrepancies were noted for Prevotella species, Peptostreptococcus species, and Viellonella parvula. With Clostridium species and Eubacterium species, some differences were noted with ceftizoxime because of differences in cut-off points. These data illustrate the magnitude of differences in results produced by the two methods using essentially the same test medium for the B. fragilis group. Fortunately, such major discordant results were not widely noted with other groups of anaerobes.

Increased in vitro activity of ceftriaxone by addition of tazobactam against clinical isolates of anaerobes.

A total of 461 clinical strains of anaerobes were tested using a broth microdilution test to determine the activity of the combination of ceftriaxone and tazobactam and other antimicrobials against these isolates. Ceftriaxone was combined with tazobactam in ratios of 1:1, 2:1, 4:1, and 8:1 and twofold dilutions of ceftriaxone in constant concentrations to tazobactam of 2, 4, 8, 16, and 32 micrograms/ml. Against beta-lactamase-producing strains of the Bacteroides fragilis group, B. capillosus, and Prevotella species all combinations of ceftriaxone and tazobactam showed enhanced in vitro activity and were eight- to 2048-fold more active than ceftriaxone alone. By comparison ceftriaxone and tazobactam showed superior or equal activity to ampicillin and sulbactam, piperacillin and tazobactam, amoxicillin and clavulanate, ticarcillin and clavulanate, and metronidazole against these same strains. Against beta-lactamase nonproducing strains of Porphyromonas, Fusobacterium, Clostridium, Eubacterium, Peptostreptococcus, and Veillonella parvula the addition of tazobactam produced no appreciable enhanced ceftriaxone activity. Fixed concentrations of tazobactam at 2 and 4 micrograms/ml appear to be most suitable for susceptibility testing and are within the pharmacologic profile of this inhibitor. Pharmacologic and toxicity studies will be needed to define the role of ceftriaxone and tazobactam in infectious diseases.

Lomefloxacin (SC 47111 or NY-198), a new difluorinated quinolone: comparison of the in vitro activity with other broad spectrum antimicrobials against Enterobacteriaceae, Acinetobacter spp, Aeromonas spp, and Pseudomonas aeruginosa.

Using a broth microdilution method, we compared the in vitro activity of lomefloxacin to other broad spectrum antimicrobials against clinical strains of Enterobacteriaceae, Acinetobacter spp, Aeromonas spp, and P. aeruginosa. Against the Enterobacteriaceae and A. hydrophila, lomefloxacin showed excellent activity with MIC50 values ranging from 0.12-0.5 micrograms/ml and MIC90 values ranging from 0.12-1.0 micrograms/ml. By comparison, lomefloxacin had superior activity to nalidixic acid and difloxacin, comparable activity to norfloxacin and fleroxacin, and slightly less activity than ciprofloxacin. Lomefloxacin was also more active than imipenem, ceftazidime, aztreonam, ticarcillin/clavulanic acid (T-CA), and gentamicin. Lomefloxacin was less active (MIC90, 4 micrograms/ml) against Acinetobacter strains than Enterobacteriaceae strains. Against these same Acinetobacter strains, lomefloxacin was 8- to 64-fold more active than ceftazidime, aztreonam, T-CA, and gentamicin but slightly less active than imipenem. Lomefloxacin was 2- to 8-fold more active than fleroxacin, difloxacin, and nalidixic acid against strains of P. aeruginosa. However, lomefloxacin was less active than ciprofloxacin and norfloxacin. Lomefloxacin had comparable activity to imipenem but was 2- to 16-fold more active than ceftazidime, aztreonam, T-CA, and gentamicin. Thus, lomefloxacin demonstrated comparable activity to other fluorinated quinolones in that it possesses high active against clinical strains of Enterobacteriaceae and had good-to-moderate activity against most strains of Acinetobacter and P. aeruginosa.

In vitro synergy and potentiation between cefotaxime and desacetylcefotaxime against clinical isolates of Bacteroides.

A broth microdilution checkerboard system was used to determine the activity of cefotaxime (CTX) and desacetylcefotaxime (des-CTX), alone and in combination, against 92 clinical isolates of Bacteroides species. Overall, CTX was only marginally more active than the metabolite. Resistance to both compounds was found among B. fragilis group, B. disiens, B. bivius, and B. capillosus. The combination of CTX with des-CTX resulted in a marked reduction of the MICs greater than or equal to fourfold for 82% of the strains tested. In addition, complete or partial synergy was noted with the combination against 85% of the strains tested. The ratio of CTX to des-CTX at the inhibitory endpoints varied widely, but in 75% of the cases it was 2:1 to 1:2 parts CTX to des-CTX.

Comparison of the bactericidal activity of clindamycin and metronidazole against cefoxitin-susceptible and cefoxitin-resistant isolates of the Bacteroides fragilis group.

Time-kill kinetic methodology was used to evaluate the bactericidal activity of cefoxitin, cefotetan, clindamycin, and metronidizole against cefoxitin-susceptible and cefoxitin-resistant isolates of the Bacteroides fragilis group. Overall, metronidazole was the most bactericidal agent, with all isolates being killed with less than or equal to 4 micrograms/ml at 24 hr. Clindamycin was the next most bactericidal agent, with 20 of 26 isolates being killed with less than 16 micrograms/ml. Six isolates with clindamycin MICs greater than or equal to 64 micrograms/ml were not killed at 24 hr, with concentrations as high as 256 micrograms/ml. Cefoxitin and cefotetan were the least bactericidal agents tested. Seven isolates with MICs of greater than or equal to 64 micrograms/ml to each agent demonstrated a lack of killing at 24 hr, with concentrations of the respective agent as high as 256 micrograms/ml. At concentrations with either agent of 32 micrograms/ml, the remaining 19 isolates were killed at 24 hr. Of the six B. fragilis isolates resistant to clindamycin, four were also resistant to both cefoxitin and cefotetan. We conclude that in hospitals with cefoxitin-resistant B. fragilis group isolates, metronidazole would provide appropriate therapy.

In vitro killing of penicillin-susceptible, -intermediate, and -resistant strains of Streptococcus pneumoniae by cefotaxime, ceftriaxone, and ceftizoxime: a comparison of bactericidal and inhibitory activity with achievable CSF levels.

This study assessed total microbial killing of 30 penicillin-susceptible, -intermediate, and -resistant strains of Streptococcus pneumoniae by cefotaxime, ceftriaxone, and ceftizoxime and compared these values with MICs for each strain against each agent as determined by three different methods/media. The results confirm the appropriateness of recent NCCLS recommendations for MIC interpretive criteria for third generation cephalosporins in which < or = 0.25 microgram/ml = susceptible and > or = 2.0 micrograms/ml = resistant when these agents are used to treat pneumococcal meningitis and data from total microbial killing studies suggests that most isolates with MICs of 0.5 and 1.0 mcg/ml would respond to high dose therapy with all three agents. The study also confirmed the recently described two- to four-fold decrease in activity of ceftizoxime against S. pneumoniae as compared with either cefotaxime or ceftriaxone; but noted that current NCCLS MIC interpretive criteria for the therapy of meningitis remain valid for all three agents. Finally, the study found that MICs determined by the E test or by microdilution broth methods using supplemented Todd Hewitt broth predict susceptibility as well as the NCCLS reference method. The actual selection among these agents for the therapy of pneumococcal meningitis should also consider other parameters including protein binding, age groups of clinical use, maximum potency against all clinically relevant pathogens, and cost.

Comparison of the in vitro activity of ciprofloxacin and 24 other antimicrobial agents against clinical strains of Chromobacterium violaceum.

Eleven clinical strains of Chromobacterium violaceum were tested for their susceptibility to 25 antimicrobial agents. Ciprofloxacin was the most active of the compounds tested although norfloxacin and pefloxacin were highly active. No resistance was detected to mezlocillin, piperacillin, apalcillin, imipenem, and aztreonam while a single strain was resistant to ticarcillin. Among the cephalosporin/cephamycin group only cefotetan showed good in vitro activity. Gentamicin was more active than amikacin and tobramycin. Good in vitro activity was also noted for chloramphenicol, doxycycline, and trimethoprim-sulfamethoxazole while C. violaceum strains were highly resistant to rifampin and vancomycin. The bactericidal activity of selected agents was shown to be concentration dependent using time-kill kinetic studies. Addition of clavulanic acid did not increase the activity of ticarcillin and in one case was shown to induce beta-lactamase. High correlation was noted between the broth microdilution and disk diffusion susceptibility tests in predicting the susceptibility patterns of C. violaceum.

Inhibitory and bactericidal activity of selected beta-lactam agents alone and in combination with beta-lactamase inhibitors compared with that of cefoxitin and metronidazole against cefoxitin-susceptible and cefoxitin-resistant isolates of the Bacteroides fragilis group.

The inhibitory activity of five beta-lactam agents, alone and in combination with a beta-lactamase inhibitor, was compared with that of cefoxitin and metronidazole against 300 beta-lactamase producing Bacteroides fragilis group isolates. Each of the beta-lactamase inhibitors significantly potentiated the activity of the respective beta-lactam. In the presence of clavulanate, the MIC90 (minimum inhibitory concentration) values of amoxicillin and ticarcillin were reduced 64-fold and 32-fold, respectively. Similarly, sulbactam enhanced the activity of ampicillin and cefoperazone 16-fold and 8-fold, respectively, whereas tazobactam potentiated the activity of piperacillin 16-fold. Few strains were resistant to the beta-lactam-beta-lactamase inhibitor combinations and were comprised of strains of B. fragilis, B. thetaiotamicron, and B. distasonis. Of the strains, 7% were resistant to cefoxitin, and none to metronidazole. Using time-kill kinetic studies, the bactericidal activity of the various beta-lactam agents, with and without beta-lactamase inhibitors, was determined and compared with that of cefoxitin and metronidazole against cefoxitin-susceptible and cefoxitin-resistant isolates of the B. fragilis group. Overall, metronidazole was the most bactericidal agent with all isolates being killed with less than or equal to 4 micrograms/ml at 24 hr. Ampicillin-sulbactam was the next most bactericidal agent with all isolates being killed with less than or equal to 16/8 micrograms/ml of ampicillin-sulbactam at 24 hr. Amoxicillin-clavulanate and cefoperazone-sulbactam had bactericidal activity similar to that of ampicillin-sulbactam. Piperacillin-tazobactam and ticarcillin-clavulanate were bactericidal at higher concentrations with all isolates killed with 64 micrograms/ml of piperacillin and 128 micrograms/ml of ticarcillin combined with their respective beta-lactamase inhibitors. None of the beta-lactam agents alone was able to kill more than 19 of the 26 isolates. We conclude that beta-lactam agents combined with beta-lactamase inhibitors have both inhibitory and bactericidal activity against cefoxitin-resistant members of the B. fragilis group provided that the concentrations achieved for these combinations are at the upper limits for maximum recommended dosing. Although isolates of the B. fragilis group have been reported to produce unusual beta-lactamases that are refractory to beta-lactamase inhibitors, none of the cefoxitin-resistant isolates tested in this study were resistant to the beta-lactam-beta-lactamase inhibitor combinations.

RO23-6240, a new orally absorbed quinolone: in vitro comparison with other broad-spectrum oral antimicrobial agents and imipenem.

A total of 626 clinical isolates were tested for their susceptibility to RO23-6240 and other broad-spectrum antimicrobial agents. RO23-6240 showed good activity against strains of Enterobacteriaceae, Acinetobacter, and P. aeruginosa. RO23-6240 MIC90s ranged from 0.032 to 4 micrograms/ml for these strains. RO23-6240 also showed good activity against staphylococci, both methicillin-susceptible and -resistant strains. The activity in vitro of RO23-6240 was comparable with that of norfloxacin and ofloxacin, and more active than imipenem, trimethoprim-sulfamethoxazole, cefaclor, and amoxycillin/clavulanate potassium. As with the other quinolones tested, increases in inoculum size produced moderate increases in the MICs and MBCs of RO23-6240.

Differences in the in vitro inhibitory and bactericidal activity of ceftizoxime, cefoxitin, cefotetan, and penicillin G against Bacteroides fragilis group isolates. Comparison of time-kill kinetic studies with MIC values.

Nineteen strains of the Bacteroides fragilis group were used to determine the bactericidal activity of ceftizoxime, cefoxitin, cefotetan, and penicillin G with time-kill kinetics studies. Each antimicrobial agent was tested at subinhibitory (1/2 X MIC), inhibitory (1 X MIC), and suprainhibitory (4 X MIC) concentrations. Penicillin G exhibited virtually no sustained bactericidal activity at any of the antimicrobial concentrations tested. At subinhibitory concentrations, ceftizoxime was considerably more bactericidal than cefoxitin or cefotetan: At 12 hr, ceftizoxime killed 89% of the inoculum, whereas cefoxitin and cefotetan killed 35% and 33% of the inoculum, respectively. At inhibitory concentrations, ceftizoxime was again more bactericidal than cefoxitin and cefotetan: At 12 hr, ceftizoxime killed 90% of the inoculum, whereas cefoxitin and cefotetan killed 78% and 73%, respectively. At suprainhibitory concentrations, all three antimicrobial agents showed comparable bactericidal activity at 12 and 24 hr. Ceftizoxime and cefoxitin had somewhat lower killing rates overall against test strains with high MICs (greater than or equal to 32) versus low MICs (less than or equal to 16). However, at subinhibitory concentrations, ceftizoxime killed the B. fragilis group strains with high or low MIC values more effectively than cefotetan killed strains with low MICs. At the highest antibiotic concentrations tested (4 X MIC), only slight differences were seen in the bactericidal activity of the three compounds, regardless of MICs.

In vitro activity of various antimicrobial agents against Staphylococcus aureus isolates including fluoroquinolone- and oxacillin-resistant strains.

To determine susceptibility to 31 old and new antimicrobials, 44 strains of Staphylococcus aureus, most resistant to oxacillin and ciprofloxacin and isolated in a community hospital, were tested in vitro. For the peptide/peptide-derivative compounds, with the exception of mersacidin, all strains were inhibited by less than or equal to 2 micrograms/ml. Minimum inhibitory concentration (MIC)90 values indicated mupirocin, teicoplanin, and MDL 62211 to be fourfold more active than vancomycin, ramoplanin, and decaplanin. For fluoroquinolones, ciprofloxacin-resistant S. aureus exhibited high-level cross-resistance to ofloxacin, norfloxacin, fleroxacin, enoxacin, and Ro 23-9424. WIN 57253, a new fluorinated naphthyridine, showed good activity against these strains. Among the beta-lactams, the penem-derivative compounds (imipenem, meropenem, FCE 22101, and HRE 664) had the greatest activity, although resistance to each compound was detected among oxacillin-resistant S. aureus. The presence of tazobactam reduced the piperacillin MIC90 fourfold. Oxacillin-susceptible strains were susceptible to cephalosporins/cephamycins, whereas most oxacillin-resistant strains exhibited resistance. This study has shown that certain old and new quinolones and peptide-derivative compounds have good in vitro activity against multiply resistant strains of S. aureus.