[Identification of anaerobic bacteria isolated from various clinical specimens and determination of antibiotic susceptibilities].

Routine isolation, identification and susceptibility testing of anaerobic bacteria present several difficulties leading to defects in the determination of local susceptibility patterns which will guide empirical treatment protocols. This study was carried out to identify the anaerobic bacteria isolated from various clinical materials obtained from the suspected patients with anaerobic infection and to determine the antibiotic susceptibilities against several antibiotics. One hundred clinical specimens (36 blood, 31 abscess, 12 peritoneal fluid, 7 joint fluid, 7 pleural fluid, 3 biopsies, 3 cerebrospinal fluids and 1 surgical wound) that were examined in our laboratory during March 20-October 30 2007, were included in the study. The specimens were collected and transported under anaerobic conditions and inoculated to conventional aerobic media and to Wilkins Chalgren agar, Schaedler agar and chopped-meat broth for anaerobic isolation. Isolated anaerobic bacteria were identified with API 20A panels (Bio-Merieux, France) via conventional methods and by the help of AN-IDENT Discs (Oxoid, England). Penicillin G, clindamycin, cefoxitin, metronidazole, piperacillin/tazobactam and imipenem susceptibility tests were performed with E- test method. Twenty two anaerobic bacteria were isolated from 14 clinical specimens; 7 of the specimens yielding the growth of more than one type of anaerobic bacteria and 8 specimens yielding both anaerobic and facultative anaerobic bacterial (4 Escherichia coli and 4 Enterococcus spp.) growth. Anaerobic bacteria were isolated in 89 abscess and in 6 peritoneal fluid specimens. The distribution of the anaerobic bacteria identified among these specimens were as follows: Bacteroides fragilis (n = 6), Bacteroides spp. other than B.fragilis (n = 4), Clostridium spp. (n = 2), Fusobacterium necrophorum/nucleatum (n = 1), Prevotella intermedia/disiens (n = 1), Peptococcus niger (n = 2), Peptostreptococcus spp. (n = 5), and Lactobacillus acidophilus/lenseii (n = 1). Beta-lactamase activity was detected only in 2 of the 6 B. fragilis isolates. All of the isolates were susceptible to imipenem and piperacillin/tazobactam. The highest rate of resistance was detected against penicillin G (9/22; 41%). While anaerobic gram-positive cocci (n = 7) were found to be sensitive to all antibiotics, the rate of resistance among anaerobic gram-negative bacilli were 75% (9/12) to penicillin, 33.3% (4/12) to clindamycin, 8.3% (1/12) to metronidazole. Among anaerobic gram-positive bacilli (n = 3), 2 were resistant to metronidazole, one to clindamycin and one to cefoxitin. The results of this first anaerobic antimicrobial susceptibility testing study performed at Konya area in Turkey revealed that penicillin was not appropriate in empirical treatment of anaerobic infections, clindamycin susceptibility should be tested before use, metronidazole and cefoxitin could be used in empirical treatment and imipenem and piperacillin/tazobactam should be saved for the treatment of complicated infections and infections caused by resistant bacteria.