Staphylococcus epidermidis--hospital epidemiology and the detection of methicillin resistance.

Infections in immunocompromised patients and in patients with indwelling prosthetic devices are often caused by hospital strains of Staphylococcus epidermidis resistant to methicillin. Tests for the detection of methicillin resistance, indicating resistance to all beta-lactam antibiotics, were evaluated in order to define a suitable screening test. A broth tube breakpoint test with a large inoculum, 10(7) colony forming units (cfu), gave the highest recovery of resistant strains. False resistance due to hyperproduction of beta-lactamase was excluded. The results correlated completely with the detection of the resistance gene, mecA, by the polymerase chain reaction. In 2/3 of the resistant strains tested the expression of the methicillin resistance was heterogeneous, only one cell in 10(2) to 10(4) expressed the resistance within 72 h in both. In broth screening tests an inoculum of at least 10(6) cfu therefore was required to detect all resistant strains within 24 h. Using agar dilution, 48 h incubation must be considered. In disc diffusion tests reliable results were obtained after only 16 h of incubation when discs containing cephradine 5 and 30 micrograms, oxacillin 1 microgram or cephalexin 30 micrograms were used, and the first disc is recommended for routine work. The epidemiology of S. epidermidis strains resistant to ciprofloxacin and/or gentamicin was studied in an isolation unit for patients undergoing bone marrow transplantation. Antibiograms and plasmids were used for typing and 31 such strains were found. Of 54 staff members 10 were colonized in the nares only, two in the nares and perineum and one in the nares and stool. In ambient air and on the clothes of staff a few of the strains predominated quantitatively. These strains colonized the skin of some of the patients who seemed to be the main dispersers. Possible routes of cross-infection were indirect contact transfer via the hands and clothes of staff (82% of the clothes were contaminated), and direct as well as indirect airborne transmission. To study the effects of chlorhexidine on skin bacteria, ten nurses washed one arm with chlorhexidine-detergent every morning for 3 weeks; the other arm served as control. The depression of the normal skin flora did not lead to a colonization with more antibiotic-resistant hospital strains. During the wash period the counts of antibiotic-resistant S. epidermidis on the treated arms were significantly reduced compared with the control arms, as also were the number of different strains.(ABSTRACT TRUNCATED AT 400 WORDS)