The prevalence of staphylococcal resistance to penicillinase-resistant penicillins. A retrospective and prospective national surveillance trial of isolates from 40 medical centers.

In a retrospective survey of resistance to penicillinase-resistant penicillins (PRPs) in 152,076 clinical staphylococcal strains isolated in 40 United States Hospitals in 1985 and 1986, rates of resistance to oxacillin were found to be 11 and 13%, respectively, among Staphylococcus aureus isolates. The rates were approximately four times higher among coagulase-negative staphylococcal strains. In a prospective study of 1,408 wound or bacteremia isolates from the participant hospitals, oxacillin and methicillin agar screening, disk diffusion, and broth microdilution testing were conducted at a single reference laboratory. These tests yielded PRP resistance rates of 15% among S. aureus, 75% among S. epidermidis, and 48% among other coagulase-negative strains. No major changes in the distribution of resistance rates among hospitals or by hospital type were observed. Dilution susceptibility testing of several antimicrobial agents against PRP-resistant isolates and species-matched susceptible isolates from the same hospital showed that teicoplanin and vancomycin were the most active drugs (100% of S. aureus isolates were susceptible). Teicoplanin and vancomycin disk diffusion testing of all PRP-resistant staphylococcal strains also showed that these isolates were susceptible to the glycopeptides. However, agar dilution screening and broth microdilution tests revealed that several coagulase-negative strains, predominantly S. haemolyticus, had teicoplanin MICs greater than or equal to 8 micrograms/ml. S. haemolyticus isolates represented a very small number of the total stains tested. Teicoplanin and vancomycin were also the most active drugs when tested against older (1962-82) clinical PRP-resistant S. aureus strains from the reference laboratory collection. The methods found to be superior in detecting PRP-resistant strains were the oxacillin 6 micrograms/ml agar screening test in 4% NaCl-supplemented Mueller-Hinton agar and the 1 microgram oxacillin disk test. By reference laboratory standards, participant laboratories were incorrect in only 2.3% of species identifications and 4.5% of oxacillin-susceptibility determinations, indicating acceptable contemporary agreement and accuracy.