During the last few decades, and in parallel with increasing resistance to multiple antibiotics, enterococci have become one of the leading pathogens that cause nosocomial infections. High-level gentamicin resistant (HLGR) enterococci have become frequent. Thus, there are compelling reasons to control the transmission of enterococci with HLGR. Many different typing methods have been used for epidemiological typing of enterococci. Pulsed-field gel electrophoresis (PFGE) has been shown to be the most discriminating typing method and is currently considered the "gold standard" for typing of enterococci. However, PFGE is an expensive method and remains time-consuming, which may be of critical importance when comparing data obtained from numerous isolates.
The aim of this thesis was to characterize clinical enterococcal isolates from patients admitted to Swedish hospitals, with special focus on HLGR strains and their genetic relatedness. Our purpose was also to develop a faster PFGE protocol for typing of enterococci, as well as to investigate if the Phene Plate (PhP) system can be used as a rapid screening method for detection of genetically related isolates of enterococci. If this could be achieved, it would be possible to minimize the number of isolates subjected to PFGE typing, which would save time and money.
In paper I, we performed a thorough investigation of eight different parameters of importance for the separation of DNA fragments by PFGE. This resulted in a modified PFGE protocol for typing of enterococci with much enhanced resolution. HLGR E. faecalis isolates obtained from patients admitted to eight Swedish ICUs during 1996 and 1998 (paper II), and isolates obtained from patients with bacteremia in the County of Östergötland during the period 1994-2001 (paper III) were characterized by our modified PFGE method. We found that the majority (69%) of isolates from ICUs in the eastern and central parts of southern Sweden, belonged to one dominating cluster, and the same cluster was also found in blood isolates from Östergötland. In nearly all cases, HLGR was due to the presence of the aac(6 ')Ie-aph(2 '')la gene situated on a Tn5281-like transposon (paper II). In the County of Östergötland, HLGR E. faecalis was first isolated from blood cultures in 1996, and the first blood isolates of HLGR E. faecium were found in 1999. During the study period, only 4 HLGR E. faecium isolates were observed, and all of them showed unique PFGE patterns.
To evaluate the efficiency of the gentamicin disk diffusion method for detection of HLGR in clinical isolates of enterococci, all enterococcal blood isolates from paper III were studied with a 30-µg gentamicin disk as recommended by SRGA, and the method was found to have 100% sensitivity and specificity when compared with PCR.
A "biochemical fingerprinting" method (PhP) was compared with PFGE for epidemiological characterization of enterococci. In earlier studies of the PhP method, enterococci were collected mainly from the environment or from normal human flora. Our study indicates that PhP may be a useful screening method for clinical E. faecalis isolates, with a relatively high concordance with PFGE, but that it is less useful for E. faecium since the concordance for E. faecium was low. To fully evaluate PhP as a tool for epidemiological characterization of enterococci from clinical settings, and to address questions regarding the validity of PFGE, further studies using additional genotyping methods, preferably newer typing systems based on DNA sequencing such as MLST, are warranted.
Linköping: Linköping Universitet , 2005. , 80 p.