TEXT

Top Abstract Text References

Due to the increasing penicillin resistance among community-acquired Streptococcus pneumoniae isolates (3, 5), as well as the increasing resistance of staphylococci and enterococci to both beta lactams (2, 7) and glycopeptides (6, 10), physicians have sought to establish the efficacy of other antimicrobial agents against these problem pathogens. Newly developed fluoroquinolones such as trovafloxacin, moxifloxacin, and gemifloxacin are potential candidates for the treatment of penicillin-resistant S. pneumoniae infections (1) and may also have utility in the treatment of certain staphylococcal and enterococcal infections.

Gemifloxacin, (R,S)-7-(3-aminomethyl-4-syn-methoxyimino-1-pyrrolidinyl)-1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxo-1,8- naphthyridine-3-carboxylic acid methanesulfonate, exhibits broad-spectrum antibacterial activity (4). Among the fluoroquinolones, gemifloxacin putatively has enhanced activity against staphylococci, streptococci, and enterococci (4). Therefore, we compared the in vitro activity of gemifloxacin against 316 bacteremic isolates of gram-positive cocci with those of ciprofloxacin, grepafloxacin, moxifloxacin, ofloxacin, sparfloxacin, and trovafloxacin in addition to three other respiratation-directed antimicrobial agents (amoxicillin-clavulanic acid, cefuroxime, and azithromycin).

(This work was presented at the 21st International Congress of Chemotherapy, Birmingham, United Kingdom, 4 to 7 July 1999.)

All isolates were obtained from blood cultures of patients at one of three teaching hospitals: Erie County Medical Center, Buffalo, N.Y.; the Henderson Site of Hamilton Health Sciences Corp., Hamilton, Ontario, Canada; or Strong Memorial Hospital, Rochester, N.Y. The microorganisms were detected by BACTEC instrumentation (Becton Dickinson Diagnostic Instrument Systems, Sparks, Md.) at the Henderson and Erie County Medical Center sites and by BacT/Alert (Organon-Teknika, Durham, N.C.) at the Strong Memorial Hospital site. After initial recovery on 5% sheep blood agar, the isolates were preliminarily identified in the participating hospitals' clinical laboratories. Subcultures of the isolates were then transported to the clinical microbiology laboratory of Strong Memorial Hospital for final identification and susceptibility testing. The identity of purported Staphylococcus aureus isolates was confirmed by the tube coagulase test, using rabbit plasma. Coagulase-negative staphylococci were identified to the species level by the use of the Staph-Ident system (Analytab Products, Plainview, N.Y.). S. pneumoniae strains were characterized by bile solubility and optochin susceptibility. Enterococci were identified by the hydrolysis of esculin in the presence of bile and by growth in 6.5% sodium chloride. All enterococcal isolates were identified as Enterococcus faecalis or Enterococcus faecium according to results of biochemical profiles obtained by using the Vitek GPI Identification Card (bioMerieux Vitek Inc., Hazelwood, Mo.) or an API 20 Strep strip (bioMerieux Vitek Inc.).

Antimicrobial agent reference powders used in these studies were as follows: amoxicillin-clavulanic acid (SmithKline Beecham Pharmaceuticals, Collegeville, Pa.), cefuroxime (Glaxo-Wellcome, Research Triangle, N.C.), azithromycin (Pfizer Inc., Groton, Conn.), ciprofloxacin (Bayer Inc., West Haven, Conn.), ofloxacin (R. W. Johnson Pharmaceutical Research Institute, Raritan, N.J.), grepafloxacin (Glaxo-Wellcome), sparfloxacin (Rhone-Poulenc Rorer, Collegeville, Pa.), gemifloxacin (SmithKline Beecham Pharmaceuticals, Harlow, Essex, United Kingdom), moxifloxacin (Bayer Inc.), and trovafloxacin (Pfizer Inc.).

Broth microdilution antimicrobial susceptibility testing was performed in accordance with the National Committee for Clinical Laboratory Standards methodology (8). The reagent powders were dissolved in accordance with the manufacturers' instructions, diluted with Mueller-Hinton broth, and distributed to the wells of microdilution trays. Each tray was inoculated with ~5 × 10⁵ CFU per well to yield a final volume of 0.1 ml per well. The trays were incubated at 35°C for 24 h. Susceptibility testing for staphylococcal and enterococcal isolates was performed in cation-adjusted Mueller-Hinton broth. Cation-adjusted Mueller-Hinton broth with 3 to 5% lysed horse blood was employed for the susceptibility testing of pneumococci. Appropriate quality control strains were included in each run of daily testing. These included S. aureus ATCC 29213, S. pneumoniae ATCC 49619, and E. faecalis ATCC 29212. The recorded MICs of all of the antimicrobial agents were the lowest concentrations that completely inhibited visible growth of the test strain. Antimicrobial agent concentrations that inhibited growth of 50% (MIC₅₀) and 90% (MIC₉₀) of the strains and percentages of organisms susceptible were calculated in accordance with the current National Committee for Clinical Laboratory Standards interpretive breakpoints for amoxicillin-clavulanic acid, cefuroxime, azithromycin, ciprofloxacin, and ofloxacin (9). For all isolates, we used a susceptibility breakpoint of 1 mg/liter for trovafloxacin. For sparfloxacin, a susceptibility breakpoint of 0.5 mg/liter was employed, while 2 mg/liter was the breakpoint used for ofloxacin. In contrast, for pneumococcal and enterococcal isolates, the breakpoint for grepafloxacin susceptibility was 0.5 mg/liter. However, for staphylococcal isolates, the breakpoint for grepafloxacin-susceptible strains was 1 mg/liter. Because no approved susceptibility breakpoints are available for gemifloxacin and moxifloxacin, the percentages of organisms susceptible to these two antimicrobial agents were not recorded.

The phenotypic distribution of the bloodstream isolates was as follows: methicillin-susceptible S. aureus, 42; methicillin-resistant S. aureus (MRSA), 49; penicillin-susceptible S. pneumoniae (PSSP), 22; penicillin-intermediate S. pneumoniae (PISP), 13; penicillin-resistant S. pneumoniae (PRSP), 10; penicillin- and vancomycin-susceptible enterococci (PSVSE), 31 (21 E. faecalis and 10 E. faecium); penicillin-resistant and vancomycin-susceptible enterococci (PRVSE), 29 (11 E. faecalis and 18 E. faecium); penicillin- and vancomycin-resistant enterococci 33 (all E. faecium isolates); methicillin-susceptible Staphylococcus epidermidis (MSSE), 22; methicillin-resistant S. epidermidis (MRSE), 32; Staphylococcus haemolyticus, 10; Staphylococcus hominis, 10; and coagulase-negative Staphylococcus species (CNS), 13.

The susceptibility results, expressed as MIC ranges, MIC₅₀s, MIC₉₀s, and percentages susceptible, are presented in Table 1. Of the drugs tested, gemifloxacin was the most active against PISP, PRSP, MSSE, and coagulase-negative Staphylococcus species, attaining MIC₉₀s of 0.03 mg/liter, while amoxicillin-clavulanic acid was the most potent against PSVSE and PRVSE. Gemifloxacin also proved to be active against MRSE, S. hominis, S. haemolyticus, and PSVSE, with MIC₉₀s of 2 mg/liter. Trovafloxacin and moxifloxacin, in that order the next most potent fluoroquinolones, were fourfold less potent against PSSP, PISP, and PRSP than gemifloxacin and two- to fourfold less active against MSSE, MRSE, S. haemolyticus, and S. hominis than gemifloxacin. However, trovafloxacin exhibited the lowest MIC₉₀s for methicillin-susceptible and -resistant S. aureus, 0.03 and 2 mg/liter, respectively, in comparison with gemifloxacin (0.03 and 8 mg/liter) and moxifloxacin (0.1 and 4 mg/liter). None of the fluoroquinolones exhibited any activity against PRVSE or penicillin- and vancomycin-resistant enterococci.

Among the fluoroquinolones tested, gemifloxacin demonstrated the most potent in vitro activity against commonly encountered PSSP, PISP, and PRSP bloodstream isolates. It also had significant activity against MSSE, MRSE, S. haemolyticus, and S. hominis but was not as active as trovafloxacin against S. aureus isolates. None of the fluoroquinolones tested appears to offer any clinically important activity against penicillin-resistant enterococcal strains. An assessment of gemifloxacin's clinical utility for gram-positive coccus infections must await comparative trails in humans.