Epidemiology of VRE in Dialysis
THE DESCRIPTIVE EPIDEMIOLOGY OF A VANCOMYCIN-RESISTANT ENTEROCOCCUS (VRE) FAECIUM OUTBREAK IN A DIALYSIS POPULATION
Lior L, Litt M, Hockin J, Kennedy C*, Jolley BA, Garcia M, Gillis G, Humar A, Louvaris J, Turchet S, McCumber T, Izatt S, Cardella C, Campbell I, Brunton J, Dedier H, Conly J
The Laboratory Centre Disease Control, Health Canada, Ottawa, ON, and The Toronto Hospital, University of Toronto, Toronto, ON, Canada
INTRODUCTION
Vancomycin resistant Enterococcus (VRE) is an important emerging pathogen, first identified in 1986 in Europe and the United States in 1988. The U.S. experience from 1989 to 1993 has shown VRE nosocomial infections rose from 0.3% to 7.9%, primarily in intensive care unit patients. The Hospital Infection Control Practices Advisory Committee (HICPAC) issued "Recommendations for Preventing the Spread of Vancomycin Resistance" which were published in Morbidity and Mortality Weekly Report (MMWR) Sept. 22, 1995. Prior to our institution's experience in August, 1995, there had been no reported outbreaks of VRE in Canada.
BACKGROUND
The Toronto Hospital is a 1200 bed, two division, tertiary care hospital with a multi-organ transplant program with one in-patient nephrology unit. There is a haemodialysis unit at each site with a total patient population of 312, as well as a 250 patient continuous ambulatory peritoneal dialysis program with a short stay and training unit.
For two years, the microbiology department had been conducting periodic screening of urine isolates for VRE, and in August, 1995, a resistant Enterococcus faecium was isolated from a routine urine culture of a nephrology in-patient. Isolation precautions were initiated and surveillance cultures of close contacts of the patient were done but no further VRE isolates were found at that time. However, VRE was identified in a urine culture from a second patient on the same unit five weeks later and an outbreak investigation was initiated at that time.
The unit was closed to new admissions and the Laboratory Centre for Disease Control (LCDC) was notified and two field epidemiologists were sent to assist in the investigation and conduct a case-control study. Of the 42 VRE isolates identified between August 5 and November 17, only one was a bacteremia and the rest were colonization.
METHODS
Surveillance:
Patient Surveillance
contacts were identified using admission, discharge and transfer records
peri-rectal swabs or stool cultures, plus urine, wound or exit site cultures as clinically appropriate were obtained from the following:
- in-patients on the affected medical ward
- in-patients transferred within the hospital from the affected unit
- discharged dialysis patients
- all Toronto Hospital haemodialysis outpatients
- all Toronto Hospital peritoneal dialysis outpatients
- patients discharged to other health care institutions
- demographic and clinical variables were obtained by chart review
Staff Surveillance
hand and peri-rectal swabs were obtained on a voluntary basis from the staff of the affected ward
Environment
items such as call bells, telephones, hand-rails, toilet seats, floors, counter tops, computer keyboards, commode chairs and electronic thermometers were swabbed
IDENTIFICATION OF ISOLATES
Bacto Enterococci confirmatory agar (Difco), a sodium azide containing selective medium to which vancomycin 6 mg/L was added was used the first two months
Bacto m Enterococcus agar with 6 mg/L Vancomycin was used following the first two months
plates were incubated up to 72 hours at 35°C
presumed to be Enterococcus species, vancomycin resistant, if gram positive cocci in chains, catalase negative, bile esculin positive and PYR positive
identification to species done by conventional methods [1]
vancomycin resistance was confirmed by vancomycin agar screen with 6 mg/L vancomycin, incubated up to 48 hours at 35°C [2]
susceptibility testing using MicroScan Walkaway System (Micro, Sacremento, Ca.) with Positive MIC 6 panels with manual readings
vancomycin MIC performed at the Mount Sinai Hospital, Toronto by the NCCLS microbroth dilution method (courtesy of D. Low, B. Willey)
Genotypic Analysis
chromosomal DNA was extracted and plugs made using conventional techniques
restriction endonulease digestion performed using 5 maI
inserts loaded onto gels and DNA fragments separated on an agarose gel using the CHEF II system (Biorad) for pulsed field gel electrophoresis (PFGE) [3]
PCR genotyping was performed (C. Rourke) using CDC methods with primers supplied courtesy of G. Tyrell, Halifax [4]
RESULTS
Positive surveillance cultures
~700 41= Enterococcus faecium 1= Enterococcus faecalis
staff 0/ 37
environment 10/140 (call bells, bed rails, toilet seats and bathroom floors)
Case Characteristics
- mean age 63 years (range 29-90)
- chronic dialysis 40/42 (95%)
- limited activity/bedridden 30/42 (71%)
- diarrhea ± incontinence 20/42 (48%)
- C. difficile toxin positive 9/20 (45%)
- received 1 antibiotic 40/42 (95%)
- received ³2 antibiotics 29/42 (70%)
- received vancomycin 17/42 (42%)
MICROBIOLOGY
Outbreak strain (41/42) was identified a sorbitol positive, ampicillin resistant, teicoplanin sensitive Enterococcus faecium
MIC's to vancomycin varied between 4-128 ug/ml (32% < mg/ml)
PFGE identified a single clone (< 2 band difference) of E. faecium (Figure 1) PCR genotyping identified the E. faecium as Van B
CONTROL MEASURES
Isolation Strategies
isolation or cohorting of colonized patients
strict body substance precautions with handwashing, the use of gloves and single use of gowns
closure of the affected unit to new admissions and limiting transfers within the hospital
2% aqueous chlorhexidine for handwashing
70% alcohol-based handwash solution for patient use or for areas without a convenient sink
Environmental Cleaning
scrupulous environmental cleaning, using a phenolic based solution
a second cleaning of bathrooms with a hypochlorite solution
the use of dedicated equipment if possible e.g. stethoscopes, glass thermometers, commode chairs
any equipment removed from the room was to be cleaned with a phenolic solution
removal of any damaged call bell
Education
education directed to staff, patients, families and visitors
use of in-services, newsletters, information sheets and signage
Antimicrobial Control
a restriction of the use of vancomycin for C difficile using HICPAC guidelines
reduced use of vancomycin with nephrology patients
recommendations for the judicious use of antibiotics
Surveillance
active surveillance of all contacts
surveillance of previously VRE positive patients, if started on antibiotics or if diarrhea developed
screening of stools sent for C difficile for VRE
SUMMARY & CONCLUSIONS
The majority of VRE positive patients were ill and debilitated, on multiple antibiotics, with diarrhea, often with incontinence and poor hygiene.
No staff hand carriage was found, however VRE contaminated environmental surfaces and equipment such as call bells emphasizes the need for effective environmental cleaning.
Dialysis patients represent a unique, logistical problem in the control of VRE due to the chronicity of their disease, the frequency of re-admission, immunosuppression, the frequent use of broad spectrum antibiotics and the frequent use of vancomycin based on the convenience of dosing.
An internal flagging system has since been developed to identify in-patients on isolation precautions.
A flagging system, as recommended in the HICPAC guidelines, has been requested to enable the tracking of previously positive colonized patients upon re-admission.
REFERENCES
- Facklam RR, Sahm DF. Enterococcus. In: PR Murray et al (ed). Manual of Clinical Microbiology, 6th ed. ASM Press 1995, pp. 308-314.
- Swenson JM et al. Development of a standardized screening method for detection of vancomycin- resistant Enterococci. J Clin Microbiol 1994;32:1700-1704.
- Persin OH, Smith TF, Tenover FC, White TJ. Diagnostic Molecular Microbiology. ASM Press, Washington, 1993.
- Clark NC, Cooksey RC, Hill B, Swenson JM, Tenover F. Characterization of glycopeptide- resistant Enterococci from U.S. hospitals. Antimicrob Agents Chemother 1993;37:2311-2317.
Back to Posters and Abstracts
