Little is known about the persistence of colonization with vancomycin-resistant Enterococcus faecium (VRE) in the nononcologic, non-intensive care unit patient. We studied all patients who had VRE isolated on > or =2 occasions of > 1 year apart (Study A) and those who had been "cleared" of VRE colonization after 3 negative stool cultures (Study B). Twelve patients had stored VRE isolates recovered > 1 year apart (Study A), and 58% of paired isolates were genotypically related according to pulsed field gel electrophoresis patterns. In Study B, stool samples were obtained weekly from 21 "cleared" patients for 5 weeks, which revealed that 24% were VRE positive. For these culture-positive patients, 72% of the cultures failed to detect VRE. Recent antibiotic use was significantly more common in the culture-positive patients, as compared with culture-negative patients (P=.003). Colonization with VRE may persist for years, even if the results of intercurrent surveillance stool and index site cultures are negative. Cultures for detection of VRE in stool samples obtained from patients declared "cleared" are insensitive.

Two bacteriologic systems for detecting bacteria in blood were compared; the automated radiometric BACTEC and the conventional method used in our laboratory for many years. BACTEC consisted of two bottles with 30 ml and the conventional method with 50 ml of media for aerobes and anaerobes. The BACTEC bottles were inoculated with 2 to 3 ml and the conventional with 4 to 5 ml of blood at the patient's bedside. Out of the 3,045 blood specimens cultured (804 patients), 262 (117 patients) were positive by one or both methods. The conventional system detected 5% more cultures. The explanation of the differences is discussed. Positive blood cultures were detected by the BACTEC procedure as early as 6 h after the blood collection. In the first 24 h, on the average, 77% of aerobic organisms were detected by the BACTEC as compared to 48% by the conventional system. All anaerobic BACTEC cultures were positive within 4 days, whereas the conventional system detected at that time 74%. At day 4, 67% of fungi were detected by the BACTEC and only 27% by the conventional system. Of the 3,045 blood cultures examined by the BACTEC, 208 were recorded as false positive with growth index readings ranging from 30 to 59.

Two simultaneous Klebsiella pneumoniae outbreaks resistant to multiple antibiotics are reported. The closely related strains were distinguished by comparing the zone diameters obtained by disk diffusion susceptibility testing but not by interpretations of susceptible, resistant, or intermediate. Similar conclusions were made when minimal inhibitory concentrations were considered. The separation was subsequently confirmed by capsular serotyping into types 2 and 30. The data presented suggest that routine antimicrobial susceptibility patterns obtained from disk diffusion and/or minimal inhibitory concentrations may serve as an inexpensive and reliable typing system for epidemiological investigations.