A methicillin/aminoglycoside-resistant strain of Staphylococcus aureus (MARS) was likely introduced by transfer of a patient from another hospital. Over the next year, 20 other patients were colonized or infected with MARS of the same phage type, although antibiograms varied. Affected patients usually had serious underlying disease and were in intensive care units. Vancomycin therapy was frequently delayed and MARS may have contributed to the death of some patients. The mode of spread was not definitively delineated, but two nurses were found to be colonized. Institution of isolation procedures was difficult, but the problem gradually waned.

Susceptibility testing showed vancomycin to be the most active agent. Synergy studies showed no consistent effect of combining methicillin with an aminoglycoside. This experience illustrates the problem of MARS spread between hospitals and wards, the need for institution of effective control measures, and consideration of early empiric use of vancomycin.

We retested 2,181 bacteria-antibiotic combinations with the Kirby-Bauer disc-diffusion technique and found interpretive changes with 120 (5.5%). Most changes (101 of 120) were single steps (i.e. from R to I, I to R or S, or S to I). Of the 19 remaining, 10 of them were from R to Sand nine from S to R. These changes were significantly more frequent for combinations with zone diameters clustered near interpretive breakpoints (within 2mm) than for other combinations, and there was a linear relationship between decreased reproducibility and increased clustering near interpretive breakpoints.

Based on an analysis of all susceptibility testing results performed in 1978, combinations most commonly clustered near interpretive breakpoints included: Ampicillin with Klebsiella pneumoniae; erythromycin with enterococci; chloramphenicol with Serratia marcescens; gentamicin with Pseudomonas aeruginosa and enterocci; and tetracycline with Enterobacter spp., Escherichia coli, and K. pneumoniae.

Reusable, corrugated, expiratory limb ventilator tubing that had been in use for 24 hours, were randomly allocated to one of three groups: no treatment (N=36); detergent wash (N = 83); or a detergent wash followed by a 10 minute immersion in a 1:16 dilution of synergized glutaraldehyde-phenate solution which was reused for 30 days. (Between 10 and 22 tubes were tested in each five day interval during this 30-day period.) Tubes were quantitatively and qualitatively cultured. There were significant differences in both the percent of contaminated tubes (no treatment = 92%, detergent wash = 72%, glutaraldehyde-phenate = 0 to 20%) and numbers of microorganisms per tube (no treatment = 3.2 × 106, detergent wash = 1.3 x 104, glutaraldehyde-phenate = 0 to 182) between groups. There was no apparent decrease in glutaraldehyde-phenate's efficacy throughout the 30-day reuse period, and in the final five days of the reuse period it was completely effective.