الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 97 |  |  |
| | | from | 97 | | |
Abstract
Enterococci, traditionally viewed as Gram positive commensal bacteria inhabiting the alimentary canals of humans and animals, are now acknowledged to be organisms capable of causing life-threatening infections in humans, especially in the hospital environment. Enterococci exhibit intrinsic resistance and acquired resistance to many antibiotics. Vancomycin resistant enterococci (VRE) have emerged as important nosocomial pathogens. They are of great clinical concern due to their increased prevalence and their ability to transfer vancomycin resistance to other bacteria (including MRSA).
The present study aimed at monitoring the prevalence of VRE colonization through active surveillance for gastrointestinal colonization with VRE and the identification of the species of the VRE isolated. The study was carried out in two ICUs in the NUH: ICU1 and ICU7. Rectal swabs were collected on admission. Subsequently rectal swabs were collected on weekly basis from all non-colonized patients until VRE colonization was documented, patient discharged or study period ended. Further on monthly basis rectal swabs were collected from all patients in each ICU on the same day as part of point prevalence survey to measure the density of colonization. Species level identification was attempted to 37 isolates using Vitek 2 compact.
The overall prevalence of VRE colonization during the study period (period prevalence) in the studied ICUs was 50.4%: 81% in ICU1 and 35.5% in ICU7. Although the VRE colonization on admission in ICU1 was 24.3%, which is almost similar to that found in ICU7 (21%), the HA-VRE colonization in ICU 1 (56.7%) was fourfold that in the ICU7 (14.4%).
The length of hospital stay was significantly longer for both categories of VRE colonized patients: colonized on admission and those who acquired HA-VRE colonization as compared to non colonized patients in both ICUs. The highest acquisition of VRE colonization was within the first month of ICU stay, in both ICUs. Vancomycin utilization was proved to have a significant association with VRE colonization.
The colonization pressure, calculated as repeated point prevalence once per month throughout the study period, ranged from 50 to 100% in ICU1, with a median of 62.5% While in ICU7, it ranged from 0 to 100% and the median was 31.5%.
Colonization with VRE has been associated with progression to infection. A total of 23(19.8%) of the studied patients acquired HA-VRE infections. In ICU1, out of the 30 colonized patients 9 (30%) acquired HA-VRE infections, and in ICU7 10 (37.03%) out of 27 VRE colonized patients developed HA-VRE infections. The relationship between colonization and infection was statistically significant in both ICUs.
The impact of VRE colonization on VRE–HA infection rates had been evaluated through monitoring the correlation between the colonization pressure and the incidence of HA-VRE infection rates per 1000 patient days per month during the study period. Obvious correlation between colonization and infection was seen in ICU1, while the incidence of VRE infections did not correlate with the colonization pressure in ICU7.
VAP was the most common type of VRE infections encountered in the present study, representing 38.46% and 56.25% and CA-UTI infections were the second most common VRE infections associated with 38.46% and 37.5% of HA-VRE infections in ICU1 and ICU7 respectively. Bacteremia was the least type of VRE infection encountered in both of the studied ICUs: 23.08% in ICU1 and 6.25% in ICU7. Multiple sites infection was encountered in three patients (33.3%) in ICU1 and 2 patients (14.2%) in ICU7.
In the present study, 37 VRE isolates were identified to the species level. E. faecium was the predominant VRE species among both; colonization associated isolates 90% (9/10) and 36.6% (4/11), and HA-VRE infection associated isolates 62.5% (5/8) and 37.5% (3/8) in ICU1 and ICU7 respectively. On the other hand VRE. faecalis represented 18.9% (7/37) of the isolates subjected to species level identification: 6 infection associated isolates and one colonization associated isolates This indicates that E. faecalis was more associated with VRE infections rather than colonization.
In ICU1, all the colonization associated isolates and 84.62% of the infection associated isolates were vancomycin resistant vanA phenotype and in ICU7 63.3% and 75% respectively. While only 3 vancomycin resistant vanB phenotype isolates were identified the 3 were HA-VRE infection associated isolates: 1 (7.69%) and 2 (12.5%) in ICU1 and ICU7 respectively.
Regarding the culture systems, the sensitivity, specificity, PPV and NPV of two culture systems, BEAVAN 6μg/ml and EBVAN15μg/ml enrichment system, were tested in the present study as compared to EBVAN 4μg/ml enrichment system. Both culture systems had specificity and positive predictive values of 100%. The limitation of both systems was in the sensitivity and negative predictive values which were apparently comparable, yet the strains that failed to grow on each of the culture system tested differed: Three (5%) of the high level resistance vanA phenotypes failed to grow on EBVAN15μg/ml while all of the 8 strains that failed to grow on EBVAN15μg/ml were of intermediate level resistance (5) and of low level resistance (3). Thereby utilization of the BEAVAN 6μg/ml could result in missing of high level resistant isolates in contrast EBVAN 15μg can reduce the number of low level and intermediate level resistant van C-containing non E. faecium and E. faecalis enterococci, while still capable of detecting all of the clinically important isolates of vanA phenotype.
The most important parameter of the antimicrobial profiling of enterococci is the delineation of their vancomycin susceptibility pattern. The modified Kirby Bauer disc diffusion technique, was evaluated against the microbroth dilution method. All isolates classified as resistant by disc diffusion were confirmed to be resistant by the microbroth technique (specificity and PPV 100%). The problem was that 40 % of the isolates categorized as sensitive or intermediate by the disc diffusion were found to be resistant. This necessitate that all isolates with sensitive or intermediate zones should be tested by MIC method to confirm that they are true vancomycin sensitive or intermediate isolates. Similar results were encountered with teicoplanin.