The proper choice of antibiotic for Clostridium perfringens infections in patients allergic to penicillin is not dear; the usual recommendations and recent in vitro studies disagree. We tested the susceptibility of 57 strains ofC. perfringens to eight penicillins, seven cephalosporins, two tetracyclines, clindamycin, chloramphenicol, and rifampin by the agar dilution method. All strains were inhibited by (per milliliter) 4 ,ug or less of any of the penicillins, chloramphenicol, or clindamycin and 8 ,mg or less of any of the cephalosporins tested. Penicillin G and amoxicillin inhibited all strains at 0.12 ,g or less per ml. Only 54% of the strains were inhibited by 1 ,ug of tetracycline per ml. Penicillin G remains the drug of first choice for infections with C. perfringens; it need not be added to a regimen containing a penicillinase-resistant penicillin given parenterally in high doses. The cephalosporins should be considered as alternative drugs for penicillin-allergic patients. Clindamycin and chloramphenicol are also effective. Tetracyclines cannot be depended upon in clostridial infections without in vitro testing, which is impracticable for initial empirical therapy.Penicillin G is the antibiotic of first choice for treating cellulitis, sepsis, and myonecrosis caused by Clostridium perfringens (1, 2,8,11,18). Although erythromycin and tetracycline are the alternative drugs usually recommended (1, 2,8,11,18), 10 to 30% of C. perfringens are resistant to tetracycline, and a few strains are also resistant to erythromycin (6,8,9,13). Cephalosporins and penicillinase-resistant and other semisynthetic penicillins are used for contaminated wounds after severe trauma and for suspected sepsis in patients with cancer, who are also at a higher risk for clostridial infections (3,4,14). Because of both scanty information about many of these agents and the need for a safe antibiotic for penicillin-allergic patients, we studied the activity of 20 antibiotics against strains of C. perfringens isolated from human infections or feces. Susceptibility testing. Agar dilution tests were done by the method of Sutter and Washington (15).
MATERIALSAntibiotics were diluted in twofold steps above and below 1 pg/ml in Brucella agar (Pfizer) supplemented with 5% laked sheep blood and vitamin K1 (10 ug/ml). The inoculum was adjusted to a no. 1 MacFarland turbidity standard (3 x 108 colonyforming units) and then diluted with prereduced Carey-Blair buffer, so that each spot delivered by 695 on May 9, 2018 by guest http://aac.asm.org/ Downloaded from