Susceptibility of Eikenella corrodens to Ten Cephalosporins

The susceptibility of 24 strains of Eikenella corrodens was determined, by the agar dilution technique, to 10 cephalosporins, as well as to clindamycin, penicillin and dicloxacillin. All strains were uniformly very susceptible to penicillin G and cefoxitin and resistant to clindamycin and dicloxacillin. Cefazolin showed good activity. Cephalexin, cephradine, and cefadroxil showed poor activity, and cefamandole's activity was relatively poor. Cephalothin, cephapirin, cefaclor, and cephaloridine showed moderate activity, with some strains relatively resistant.

The susceptibility of 24 strains of Eikenella corrodens was determined, by the agar dilution technique, to 10 cephalosporins, as well as to clindamycin, penicillin and dicloxacillin. All strains were uniformly very susceptible to penicillin G and cefoxitin and resistant to clindamycin and dicloxacillin. Cefazolin showed good activity. Cephalexin, cephradine, and cefadroxil showed poor activity, and cefamandole's activity was relatively poor. Cephalothin, cephapirin, cefaclor, and cephaloridine showed moderate activity, with some strains relatively resistant.
The in vitro susceptibility of E. corrodens to cephalothin has been reported as variable (2), and failure of clinical response has been noted (2). E. corrodens has not been tested previously against any of the other cephalosporins despite their widespread usage in types of infections where E. corrodens is a potential pathogen.
We report the susceptibility of 24 strains of E. corrodens to 10 cephalosporins as well as to clindamycin, penicillin, and dicloxacillin.
All strains tested were recovered from clinical specimens obtained between 1975 and 1978. Isolates were confirmed as E. corrodens according to standard criteria (2, 7). All strains had typical corroding colonial morphology, were oxidase positive, and reduced nitrate to nitrite.
The sources of the isolates were as follows: abscesses, 8; transtracheal aspirates, 5; human bite wounds, 5; wounds, 4; blood, 1; and sputum, 1. How many patients had been on cephalosporin therapy prior to culture is not known. Laboratory standard powders of the following antimicrobial agents were kindly supplied as follows: cephalothin, cefazolin, cephalexin, cephaloridine, cefaclor, cefamandole, and penicillin G by Eli Lilly & Co., Indianapolis, Ind. Agar dilution susceptibility testing was done by the following techniques.
Strains of E. corrodens were taken from frozen stock culture and grown on Brucella agar supplemented with 5% sheep blood, hemin (5 ,ug/ml), and vitamin K1 (10 ,ug/ml) in 5 to 10% C02 for 48 h. To assure purity, typical colonies were picked and transferred to another Brucella agar mixture and reincubated in 5 to 10% C02 at 370C for 48 h. Growth was transferred to tubes containing Mueller-Hinton broth plus hemin (5 ,ug/ml) and incubated overnight in 5 to 10% C02.
Turbidity of growth was adjusted to one-half McFarland Standard no. 1, corresponding to approximately 10" colony-forming units/ml. Mueller-Hinton agar plates supplemented with 5% sheep blood and the various antimicrobial concentrations, to yield 128 to 0.0625 ,Lg/ml, were inoculated by using a Steers replicator. Control plates containing no antibiotics were inoculated before and after each series of antibiotic containing plates. All plates were incubated in 5 to 10% C02 at 370C for 48 h and then examined. The activity of the 13 antimicrobial agents tested against the 24 strains of E. corrodens is shown in Table 1. E. corrodens was uniformly very susceptible to penicillin G and cefoxitin and resistant to dicloxacillin and clindamycin. Cefazolin showed good activity at achievable peak serum levels. Susceptibility was variable to the other agents. Cephalothin, cephapirin, cefaclor, and cephaloridine showed moderate activity with some strains relatively resistent. Cephalexin, cephradine, and cefadroxil showed poor activity; cefamandole's activity was relatively poor.
The cephalosporins have gained widespread clinical usage. They are frequently used as initial empiric therapy for a variety of situations including orally contaminated wounds and abscesses (3,5,12) where they are directed against the streptococci, staphylococci, and gram-negative rods which may be present. E. corrodens may also be a potential pathogen in these same situations. However, due to its unusual antimicrobial susceptibility pattern, such therapy may not prove adequate against E. corrodens. E. corrodens has been tested previously against cephalothin and found to have variable in vitro susceptibility (2). Brooks et al. (2) note that some of their patients developed infections with Eikenella while receiving cephalothin therapy providing "clinical confirmation of the in vitro resistance." Since then several newer cephalosporins have come into widespread clinical usage and others are being developed.
Our data are in agreement with prior reports that E. corrodens is susceptible to penicillin but resistant to dicloxacillin and clindamycin (2,11,13). On a weight basis, only cefoxitin, a cephamycin, showed comparable or better activity than penicillin against the strains tested. At achievable peak serum levels, cefazolin showed good activity, whereas cephalexin, cephradine, and cefadroxil showed poor activity. Cephalothin, cephapirin, cefaclor, and cephaloridine were active at the upper limits of achievable peak serum levels.
Penicillin (or ampicillin) remains the drug of choice in infections caused by E. corrodens. Cefoxitin is more active than penicillin G on a weight basis and will apparently be clinically effective. Two of our isolates were from patients with pneumonia who were treated with cefoxitin. In both cases E. corrodens was recovered by transtracheal aspiration prior to therapy; both patients had an excellent clinical response (unpublished data). In situations where other cephalosporins are to be used for infection with E. corrodens, susceptibility testing of the involved strain should be performed.