ABSTRACT
BACKGROUND: Until 1995, there were no cases of vancomycin resistant enterococcus (VRE) identified at our university hospital. From May 1995 to August 1996, we investigated a cluster of 10 cases of phenotypic class Van B Enterococcus faecium. METHODS: Patients were matched with controls who were on the same unit for at least 7 days prior to the case developing VRE. Control patients were age and sex matched if possible, and had duration of hospitalization at least as long as the number of days it took the patient to become VRE positive. We analyzed 16 independent risk factors using Epi-info version 6. Environmental cultures were obtained in the MICU where 5 of the patients were located. All 10 patient isolates and environmental isolates were analyzed by pulsed field gel electrophoresis (PFGE). RESULTS: PFGE confirmed the genetic relatedness of all 10 patient isolates and environmental isolates. The VRE-positive group was more likely to be immunosuppressed and to have exposure to 3 physicians. In the MICU, significant, P < 0.05) risk factors for VRE were higher Apache scores, location adjacent to a VRE case, duration of vancomycin and amino-glycoside use, duration of invasive catheter use, and diarrhea. Among the VRE-positive environmental cultures was a blood pressure cuff wash that was used on several patients. CONCLUSION: We hypothesize that a VRE strain was introduced into our hospital environment and was spread by personnel or contaminated equipment. As a consequence of this study, a hospital-wide VRE policy was implemented.
Subject(s)
Anti-Bacterial Agents/therapeutic use , Disease Outbreaks , Enterococcus faecium/drug effects , Gram-Positive Bacterial Infections/epidemiology , Vancomycin/therapeutic use , Adolescent , Adult , Aged , DNA, Bacterial/analysis , Drug Resistance, Microbial , Enterococcus faecium/genetics , Enterococcus faecium/isolation & purification , Female , Gram-Positive Bacterial Infections/drug therapy , Gram-Positive Bacterial Infections/microbiology , Hospitals, University , Humans , Male , Middle Aged , Prevalence , Retrospective Studies , Utah/epidemiologyABSTRACT
An outbreak of methicillin-resistant Staphylococcus aureus infections at the University of Utah Health Sciences Center occurred over a 7-month period. While the isolates phenotypically appeared to be similar in gross morphology and have similar Vitek antibiotic susceptibility patterns, two additional methods of strain characterization were evaluated to enhance the epidemiological investigation: pulsed-field gel electrophoresis and gas chromatography with the MIDI Sherlock system. Sherlock uses gas chromatography to qualitatively and quantitatively analyze the cellular fatty acid composition of organisms and creates two-dimensional plots based on principal-component analysis to define groups of closely related organisms. All isolates were also evaluated by digesting their chromosomal DNAs with the low-frequency-cutting enzyme SmaI and separating the restriction fragments by contour-clamped homogeneous electric field gel electrophoresis. Sample preparation for this pulsed-field gel electrophoresis included a novel cell lysis procedure involving achromopeptidase, greatly reducing the turnaround time. Isolates tested were recovered from the following: 45 suspected outbreak patients, 6 hospitalized patients believed to be unrelated to the outbreak, 6 patients from outside the hospital, and one health care practitioner implicated in the outbreak. Of 45 phenotypically similar suspect strains, 43 clustered tightly on the Sherlock two-dimensional plot. All outbreak patient isolates were also identical by pulsed-field gel electrophoresis with the exception of the same two outliers identified by Sherlock. In this epidemiologic investigation, we found an excellent correlation between the Sherlock and pulsed-field gel electrophoresis results for strain characterization of methicillin-resistant S. aureus.
