Résumé
The blood culture isolates obtained over the period 1985-1990 in a general teaching hospital were reviewed to determine trends in the prevalence of resistance to antimicrobial drugs. The percentages of Staphylococcus aureus isolates resistant to methicillin increased each year. Resistance among coagulase negative staphylococci also increased in prevalence: by 1990 approximately 50% of such isolates were resistant to methicillin, erythromycin, co-trimoxazole and gentamicin, 24% were resistant to clindamycin, 20% to fucidic acid but only 0.5% to vancomycin. Isolates of Enterobacteriaceae, excluding community-acquired salmonellae, showed increasing prevalence of resistance to beta-lactams, as did Acinetobacter spp isolates to gentamicin, co-trimoxazole and ceftriaxone. The isolates of Pseudomonas aeruginosa were exceptional, having no evident increase in the prevalence of resistance during the period. The rapid increases observed in relation to the other pathogens indicate the need for an antibiotic policy based on continuous surveillance of susceptibility patterns in the hospital.
Sujets)
Antibactériens/usage thérapeutique , Infections bactériennes/sang , Infections communautaires/traitement médicamenteux , Infection croisée/sang , Résistance microbienne aux médicaments , Utilisation médicament , Hôpitaux généraux , Hôpitaux d'enseignement , Humains , Prévention des infections , Tests de sensibilité microbienne , Admission du patient/tendances , PrévalenceRésumé
Results and costs of the first six months experience with BACTEC NR-730 were compared with a series of blood cultures performed by the conventional method previously used. The newer technology detected the growth of 14.1% of significant isolates on the day of receipt of the specimens. The previous method lacked blind subcultures on the day of receipt and therefore detected growth only after overnight incubation. No direct comparison of the sensitivities of the methods was possible, but the percentages of cultures yielding significant isolates were similar for the two methods. With the new method, technicians needed less time for daily screening of blood cultures, fewer subcultures were required and less contamination was observed. The method used to calculate the directly-related variable costs of the two methods is set out. In the particular situation reported, workload and labor costs were such that introduction of BACTEC NR-730 resulted in a saving on variable costs.