Antimicrobial Susceptibility of Obligate Anaerobes Isolated from Military Trauma Patients
Methods: We utilized all initial unique isolates of obligate anaerobes collected as part of the Trauma Infectious Disease Outcome Study (9/09-5/13). Susceptibility testing was performed using CLSI agar dilution method. Tested abx included beta-lactam/beta-lactamase inhibitors (amp/sulb, pip/tazo ), cephalosporin (cefoxitin), carbapenems (ertapenem, imipenem, meropenem), clindamycin, metronidazole, moxifloxacin, tigecycline, and linezolid. Resistance was defined by CLSI criteria except for tige which utilized FDA criteria and no criteria exists for linezolid.
48 patients with 63 unique anaerobic isolates were included ( 15 (24%) B. fragilis, 16 (25%) B. non-fragilis, 11 (17%) Clostridium spp., 12 (19%) Finegoldia and Peptostreptococus spp., and 9 (14%) P. acnes). For Bacteroides spp, the 4 most active agents were metro (90% susceptible), pip/tazo (93%), imipenem (95%) and tigecycline (95%) with an MIC50/90 of 4/8 mcg/ml for linezolid. 100% of Clostridium spp isolates were susceptible to amp/sulb, pip/tazo, ertapenem, meropenem, metro, and tige with a MIC50/90 of 4/16 mcg/ml for linezolid. 100% of Finegoldia, Peptostrepcoccus and P. acnes isolates were susceptible to amp/sulb, pip/tazo, cefoxitin, ertapenem, imipinem, meropenem, and tige with a linezolid MIC90 of 0.5 for P. acnes and 2 for the other isolates. 6(10%) of the isolates were resistant to 3 or more classes(excluding tige and linezolid) of abx and (of those 6 isolates, 5 were Bacteroides spp.), 3 (5%) isolates to 4 or more classes (all Bacteroides spp.). One B. fragilis isolate was resistant to all abx tested except for tige. Of the 45 isolates that had prior exposure to 82 abx with anaerobic activity, 19 (23%) combinations of drug and bacteria demonstrated resistance.
Conclusion: This study reveals high levels of resistance among some obligate anaerobes. Further studies are warranted to evaluate risk factors for resistance and to determine if in vitro susceptibilities correlate with clinical outcomes.
A. Weintrob, None
M. Beckius, None
W. Zera, None
D. Lu, None
W. P. Bradley, None
D. Tribble, None
C. K. Murray, None
E. Rini, None