The activity of combinations of vancomycin (2 or 10 ,ug/ml), gentamicin (0.3 ,ug/ ml), and rifampin (0.03 [xg/ml) against methicillin-resistant, coagulase-negative staphylococcal isolates was determined by the time-kill method. Combinations of rifampin with gentamicin or with vancomycin 2 ,ug/ml demonstrated enhanced killing against 13 of 17 and 13 of 25 strains, respectively. However, rifampinresistant strains were selected with the latter combination in the remaining 12 of 25 studies.

A Vibrio vulnificus hemolysin (VVH) was purified by two steps of hydrophobic column chromatography on Phenyl-Sepharose HP. The first chromatography was carried out at pH 6.0. In this pH condition, VVH efficiently bound to the column, but the hemolysin fraction eluted was accompanied with colored substance(s). To eliminate this colored substance, the second chromatography was carried out at pH 9.8 in the presence of 1% 3-[(3-cholamidopropyl) dimethylammonio]-1-propanesulfonate (CHAPS), a zwitterionic detergent. Homogeneity of the hemolysin thus obtained was shown by polyacrylamide gel electrophoresis. The specific activity increased 33, 600 times and the yield was 35%. The method is simple and useful to supply enough VVH for study of the role of the hemolysin in the infection by V. vulnificus or on the mechanism of action of the hemolysin.

The reliabilities of five in vitro susceptibility tests (agar dilution, broth microdilution, automated MS-2, Kirby-Bauer disk diffusion, and ability to grow on methicillin-containing agar) to predict the susceptibility of 204 coagulase-negative staphylococcal isolates to penicillinase-resistant semisynthetic penicillins were compared. There was wide variation in susceptibility, with results ranging from 86.3% susceptible by MS-2 to 38.2% by growth on methicillin-containing agar. The results of the broth dilution techniques, including the MS-2, were significantly different (P < 0.02) from the remaining tests. Nafcillin disks were less effective (P < 0.02) than oxacillin disks in predicting resistance. Kirby-Bauer oxacillin disks and the ability to grow on methicillin-containing agar were the most reliable predictors of resistance. The MS-2 did not reliably predict resistance.

No abstract