Search

The natural history of the carriage of methicillin-resistant Staphylococcus aureus (MRSA) was examined in a 9-year retrospective cohort study of 102 known carriers. The populations studied consisted of patients admitted to a university hospital from 1989 through 1991; a review extending back to January 1983 was conducted. The focuses of the study included the duration of carriage among patients who were known to have carried MRSA previously and who were readmitted to the hospital (36 patients) and the optimal anatomic site for screening (66 patients). Cultures of the nares (sensitivity, 93%; negative predictive value, 95%) were considerably more valuable for the detection of MRSA colonization than were cultures of cutaneous sites of the axilla, groin, and perineum (sensitivity, < or = 39%; negative predictive value, < or = 69%). The estimated half-life of MRSA colonization in this special population of patients was approximately 40 months. Restriction enzyme analysis of plasmid types of paired isolates from the 12 patients with MRSA carriage persisting for > 12 months revealed five instances (42%) in which both isolates were of the same type. In summary, our results indicate that the majority of readmitted carriers harbor MRSA for > 3 years and that, in this population, culture of the anterior nares alone (with culture of wound or sputum, when present) is a valid and efficient method for the detection of persistent MRSA carriage.

Three commercially available systems (API Staph-Trac, API 20GP, and Vitek GPI), used to identify coagulase-negative staphylococci, were evaluated against 277 bloodstream isolates, including 94 isolates of Staphylococcus epidermidis and 183 isolates of other coagulase-negative Staphylococcus species. The conventional method of Kloos and Schleifer served as the reference method. Controls included 14 ATCC type culture strains of coagulase-negative staphylococci. The API Staph-Trac system showed the highest rate of agreement with reference method, correctly identifying 73% of the isolates. The Vitek GPI System had an overall rate of agreement of 67% and the API 20GP system correctly identified 61%. The API Staph-Trac system correctly identified 94% of the isolates of S. epidermidis compared with 64% by both Vitek GPI and API 20GP. The most common error for both Vitek GPI and API 20GP systems was the failure to identify organisms contained within the database of the systems. Because none of the tested commercial identification systems identified "non-epidermidis" coagulase-negative Staphylococcus species with a high degree of accuracy, the systems need to be markedly improved or new systems developed.

A total of 899 urine cultures were evaluated to assess the need for and cost-effectiveness of using a gram-positive coccal selective plate in the initial plating of urine cultures. Of these cultures, 437 were examined retrospectively and 462 were examined prospectively. Urines were quantitatively plated to three media: sheep blood sugar, MacConkey agar, and phenyl ethyl alcohol agar. Of all urine samples in both studies, 52% yielded no growth on any of the three media. Of all 899 urine cultures, there were only 5 cultures (less than 1%) in which a significant count of a gram-positive organism was detected only on the phenyl ethyl alcohol agar plate and not recoverable on the sheep blood agar plate. In each of these five instances, the need for the use of the selective plate occurred when a Proteus mirabilis strain swarmed over an enterococcus. The inclusion of a selective gram-positive coccal medium for initial plating of urine cultures is unnecessary and not cost-effective. When Proteus swarms on sheep blood agar, a sweep should be made with an inoculating loop from the sheep blood agar and streaked to phenyl ethyl alcohol agar or a similar gram-positive coccal selective medium.

The usefulness of a test for slime production as a marker for clinically significant infections with coagulase-negative staphylococci and its implications for therapy were examined. Hospital records were reviewed for 59 patients from each of whom more than one isolate of coagulase-negative staphylococci was obtained. In patients with a prosthetic device, 81% of 59 infectious episodes were due to a slime-positive coagulase-negative staphylococci. In contrast, 22 noninfectious episodes (in which the organisms were contaminants) were equally distributed between episodes due to slime-positive or slime-negative isolates (P = .005). Only 32% of infections caused by slime-positive organisms, in contrast to 100% of infections caused by slime-negative organisms, were improved by treatment with antibiotics alone (P = .02). Prosthetic device removal in addition to antibiotic treatment significantly improved the outcome in patients with infections due to slime-positive organisms when compared with treatment with antibiotics alone (93% vs. 32% improvement; P = .00025).