Abstract
Acinetobacter baumannii has emerged over the last decade as a significant opportunistic pathogen. Although it is generally associated with benign colonization of hospitalized patients, it is responsible for about 10 % of nosocomial infection in intensive care unit (ICU) patients, causing a wide range of infections such as bacteraemia, nosocomial pneumonia, urinary tract infection, secondary meningitis and burn and wound infections (Bergogne-Berezin, 2001; Joshi, 1998; Levin et al., 2003; Poirel et al., 1999). In Acinetobacter-associated nosocomial infection, the major problem encountered by ICU clinicians relates to the readily transferable antimicrobial resistance expressed by this organism (Bergogne-Berezin, 2001; Joshi, 1998). In addition to intrinsic resistance, A. baumannii has the ability to acquire resistance to many major classes of antibiotics including newer ß-lactams (Perilli et al., 1996). The presence of resistance plasmids (R-plasmids) is a significant feature of this organism, and plasmid profiling has been proposed as a method of epidemiological typing for Acinetobacter (Joshi, 1998; Perilli et al., 1996). Although A. baumannii colonizes hospitalized patients, approximately 30 % of isolates are associated with frank infection in ICU patients and, in this setting, tend to demonstrate variable susceptibility profiles (Dy et al., 1999). Nevertheless, despite the rising clinical importance of A. baumannii compared with other nosocomial pathogens in developing countries, this organism has been widely overlooked. In this study, five isolates of A. baumannii isolated in 1998 from exudates of post-surgical wound sepsis from our teaching hospital were examined. The isolates were found to harbour transferable extended-spectrum ß-lactamase (ESBL)-encoding resistance determinants on a 66.1 kb conjugative plasmid.
The isolates were identified (Von Graevenitz, 1995) and biotyped according to the method of Bouvet & Grimont (1987). Escherichia coli DH5α was used as the recipient strain for bacterial conjugation experiments. The size of the test plasmid was estimated using E. coli NCTC 50192 carrying plasmids of 154, 66.1, 37.6 and 7.7 kb.
Antimicrobial susceptibility testing of clinical isolates of A. baumannii, E. coli transconjugants and their plasmid-cured derivatives was first performed by the disc-diffusion technique (Hi-Media). The MIC of the isolates was determined according to NCCLS guidelines (NCCLS, 1997). E. coli ATCC 25922 was included as a control strain for susceptibility testing. The antibiotics tested are listed in Table 1. ESBL production was confirmed by using a double-disc synergy (DDS) test, which was performed by placing a cetazidime disc (30 µg) or a cefotaxime disc (30 µg) 30 mm away from a disc containing amoxycillin/clavulanate (60/10 µg), as described previously (Jarlier et al., 1988). ESBL production was considered positive when an enhanced zone of inhibition was visible between the ß-lactam- and ß-lactamase inhibitor-containing disks.