ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Mar. 1979, p. 337-341 0066-4804/79/03-0337/05$02.00/0

Vol. 15, No. 3



Department ofMedicine, Louisiana State University School of Medicine, New Orleans, Louisiana 70112 Received for publication 5 October 1978

Thirty-one adult patients with infections due to anaerobic bacteria were treated with cefamandole. BacterQides fragilis group (17) and Bacteroides melaninogenicus (13) were the most frequent anaerobes isolated. Duration of therapy varied from 2 to 49 days. Results were judged satisfactory in 26 cases, and unsatisfactory in 1 case. Four cases could not be evaluated. Adverse reactions occurred in 16 patients and included positive direct Coombs' test without hemolysis, transient liver function abnormalities, phlebitis, reversible neutropenia, fever, eosinophilia, and toxic epidermal necrolysis. The more significant reactions were associated with prolonged therapy. None was lethal. These data suggest that cefamandole is effective in treatment of most anaerobic infections. abscess (9). The specimens were immediately injected into oxygen-free vials (Port-a-Cul; Baltimore Biological Laboratory [BBL], Cockeysville, Md.) and transported to the laboratory. When specimens could not be aspirated, specifically in cases of postpartum, postabortal, or intrauterine device-related endometritis, the endocervical pus was collected with a sterile swab and transported to the laboratory in oxygen-free tubes (BBL). Specimens for anaerobic culture were inoculated onto prereduced blood agar and kanamycin-vancomycin blood agar plates and into Schaedler broth. All anaerobic media were supplemented with hemin (5 ,ug/ml) and menadione (0.5 jig/ml) and incubated in an anaerobic chamber (National Appliance Company, Hollywood, Fla.) containing an anaerobic gas mixture of 85% N, 10% H2, and 5% CO2. Anaerobic bacteria were identified by colony morphology, reaction to Gram stain, growth on selective media, gas chromaPatients. Thirty-one patients with anaerobic infec- tography of fermentative products, and biochemical tions who had been hospitalized at Charity Hospital reactions. The criteria used for these procedures are of New Orleans between July 1977 and August 1978 those outlined by the Anaerobic Laboratory at Virwere admitted to the study. Patients were included ginia Polytechnic Institute (8). Biochemical reactions only if they (i) were over 13 years of age, (ii) were not were carried out by means of the Minitek Miniaturized pregnant, (iii) had no history of allergy to penicillin or Differentiation System (BBL) (16). Aerobic bacteria cephalosporin antibiotics, and (iv) had not received were isolated and identified by standard laboratory any antibiotics once culture specimens were taken. techniques (11). Susceptibility tests. Antibiotic susceptibilities of Informed consent was obtained from each patient anaerobic bacteria were determined by using the folbefore cefamandole therapy was started. In all but three patients anaerobic organisms were lowing agar-dilution procedure. After each organism isolated from culture material. In the three patients in had been isolated and checked for purity, it was inocwhom anaerobes were not recovered, the pus obtained ulated onto a prereduced blood agar plate and incusmelled foul, its Gram stain revealed a variety of bated at 37°C for 48 h under anaerobic conditions. On pleomorphic gram-negative organisms, and its culture the day before testing, serial dilutions of the antibiotic did not grow any aerobic gram-negative organism. The to be tested were added to tubes of melted Brucella clinical diagnoses in these three patients were lung agar (cooled to 48°C) containing hemin (5 i.g/ml), menadione (0.5 ,g/ml), and 5% laked sheep blood. The abscess, salpingitis, and postabortal endometritis. Isolation and identification of bacteria. Most contents of each tube were mixed and poured into specimens were collected by transtracheal aspiration square petri dishes with 13-mm grids. After drying at (15) or by aspiration of material directly from an room temperature for 30 min, the plates were stored 337

Cefamandole nafate, a new parenteral cephalosporin, has a broader spectrum of activity against both aerobic and anaerobic bacteria than cephalothin or cefazolin (9, 10). In vitro it is active against a wide range of gram-positive and gram-negative aerobic and anaerobic bacteria (5, 13). On the basis of these reports, we evaluated cefamandole in the treatment of anaerobic infections. (The results presented here have been reported, in part, in abstract form [R. N. Greenberg, M. C. Scalcini, C. V. Sanders, and A. C. Lewis, Program Abstr. Intersci. Conf. Antimicrob. Agents Chemother. 18th, Atlanta, Ga., Abstr. no. 360, 1978].) MATERIALS AND METHODS



in the anaerobic chamber for 24 h before inoculation. The final concentration of antibiotics ranged from 0.5 to 64 yg/ml. The inoculum was prepared by transferring the organisms from blood agar plates to Schaedler broth, and the turbidity was adjusted to a McFarland no. 1 standard (2). Plates containing antibiotic were inoculated with approximately 104 colony-forming units with a Steers replicator (17) and incubated for 48 h at 37°C in an anaerobic chamber. All plates were run in duplicate with the appropriate growth controls included. The minimum inhibitory concentration (MIC) was considered to be the lowest concentration of antibiotic yielding no growth, a barely visible haze, or one discrete colony (4). Antibiotic susceptibilities for aerobic bacteria were determined by the Kirby-Bauer disk diffusion technique (3). Assay of antibiotic level. Serum, empyema fluid, and urine were collected from one patient and assayed for cefamandole. Serum, urine, and empyema fluid concentrates were measured by the Bacillus subtilis cup plate method (7). Therapeutic regimen and evaluation of response. Most patients were initially treated with intravenous (i.v.) cefamandole in a dose of 2 g over 10 to 20 min every 4 h. In life-threatening infection, a 2-g dose was administered every 3 h. Doses were adjusted in patients with renal failure as follows: 2 g every 6 h (serum creatinine, 1.5 to 2 mg/100 ml), 1.5 g every 6 h (serum creatinine, 2 to 4 mg/100 ml), 1 g every 6 h (serum creatinine, 4 to 10 mg/100 ml), and 0.5 g every 8 h (serum creatinine greater than 10 mg/100 ml). Duration of therapy ranged from 2 to 49 days. Tobramycin therapy was started in 18 patients at the beginning of cefamandole therapy. However, only six of those patients received 5 or more days of concurrent tobramycin therapy; those patients were also infected with aerobic gram-negative rods. Adjunctive surgical drainage was required in seven patients. Patients were followed up for 1 to 6 months after therapy was stopped. There were no recurrent infections. Results of therapy were judged as satisfactory if the patient's infection resolved or responded after 5 or more days of treatment, as unsatisfactory if the patient did not improve within 3 days, and as not able to be evaluated if (i) the patient died from a noninfectious cause during the study (two cases) or (ii) the patient's thoracic empyema was not adequately drained (two cases). Side effects to cefamandole were documented. RESULTS Types of infections. Cases included lung abscess (10 patients), postpartum, postabortal, or intrauterine device-related endometritis (7 patients), aspiration pneumonia (3 patients), pharyngeal abscess (3 patients), salpingitis (3 patients), lung abscess with empyema (2 patients), ruptured appendix with peritonitis (2 patients), and ruptured tubo-ovarian abscess with peritonitis (1 patient). Aerobic bacteria were isolated in 24 infections. Bacteroides fragilis (subsp. fragilis or distasonis) was isolated in two cases each of aspiration pneumonia, lung


abscess, ruptured appendix, thoracic empyema, and salpingitis, and in one case of endometritis. Bacteriology. Twenty-nine species of anaerobic bacteria and 11 species of aerobic bacteria were isolated. Table 1 lists the most common isolates. B. fragilis (subsp. fragilis or distasonis) was the most common anaerobe isolated. The Neisseria gonorrhoeae isolates were recovered from the endocervical cultures from four patients with endometritis. It was impossible to determine whether the anaerobes isolated from endocervical pus were infective pathogens or members of the normal flora. Thus, all of these isolates were studied. Antibiotic susceptibilities. Seventy-four of 89 (83%) anaerobes were inhibited by c32 jig of cefamandole per ml, and 80 of 89 (90%) were inhibited by s64 yg/ml (Table 2). It should be noted that only 8 of 17 isolates (47%) of B. fragilis group were susceptible to '32 ,tg/ml and 11 of 17 (65%) to 64




13 9

0 1 2

0 1 2


VOL. 15, 1979


tient was studied. He suffered from a thoracic empyema and had serum, urine, and empyema fluid measured for cefamandole. He was receiving 2 g i.v. over 10 to 20 min every 3 h. A serum level 15 min after the infusion was 96 ,ug/ml, a serum level 3 h after the infusion was 3 ,tg/ml, a spot urine level was 10,300 ,ug/ml, and an empyema level was undetectable. Clinical response. Twenty-six patients responded satisfactorily and one unsatisfactorily to therapy, and four could not be evaluated (Table 3). The single patient in whom treatment failed had an aspiration pneumonitis. The culture of his transtracheal aspirate yielded a Bacteroides melaninogenicus (MIC = 1 ,ug/ml), an Enterobacter aerogenes (MIC c 30 ,ug/ml), and a B. fragilis (MIC > 64 ug/ml). The patients who could not be evaluated included: one patient with a lung abscess, who died of pulmonary embolus 2 days after therapy was begun; one patient with a lung abscess, who died of gastric hemorrhage after 6 days of therapy; and two patients with thoracic empyemas that were not

adequately drained during the study. B. fragilis groups were isolated in 11 patients; the response was considered satisfactory in 8 and unsatisfactory in 1, and could not be evaluated in 2. Table 4 lists the diagnosis of each of these patients and their responses. Tolerance and toxicity. Fifteen patients tolerated cefamandole without any problems. The remaining 16 patients experienced a total of 24 drug-related side effects, and therapy with the drug had to be discontinued in 7 patients. Table 5 lists the side effects. Most of those reactions were minor and none was lethal or irreversible. There were no hemolytic episodes due to the development of a positive direct Coombs' test result. Of those patients in whom cefamandole therapy was discontinued, three had a reversible neutropenia on days 28, 34, and 40 of therapy; TABLE 3. Response to therapy Response Infection

Lung abscess Endometritis Aspiration pneumonia Pharyngeal abscess Salpingitis Ruptured appendix with peritonitis Lung abscess with empyema Ruptured tuboovarian abscess

No. of cases Satisfac- Unsatis- Not evaltory factory uated

10 7 3

8 7 2

3 3 2

3 3 2



2 1




TABLE 4. Response to cefamandole in cases in which B. fragilis group was isolated Response No. of cases Satisfac- Unsatis- Not evaltory factory uated


Lung abscess with empyema Ruptured appendix Aspiration pneumonia Endometritis Lung abscess Salpingitis



2 2

2 1

2 2 1

2 2 1


TABLE 5. Adverse reaction to cefamandole (16 patients) Abnormality

No. of pa-


Positive direct Coombs' test result 6 Transient mild elevation in SGOTa and/or 6 alkaline phosphatase Phlebitis 6 Reversible neutropenia 3 Drug fever with eosinophilia 1 Eosinophilia 1 Toxic epidermnal necrolysis 1 a SGOT, Serum glutamic oxalacetic transaminase.

one had drug-related fever and eosinophilia on day 30; one had a drug-related eosinophilia on day 35; one displayed toxic epidermal necrolysis, possibly related to cefamandole, on day 17; and one had severe phlebitis on day 19 (see Table 6).

DISCUSSION Our results further support previous observations that cefamandole may be an effective antibiotic in the treatment of anaerobic infections (18; D. G. Russell, L. R. Levine, and E. McCain, Program Abstr. Intersci. Conf. Antimicrob. Agents Chemother. 17th, New York, N.Y., Abstr. no. 174, 1977). In 26 of 27 (96%) patients, the drug was clearly effective in treating the infection; 4 patients could not be adequately evaluated. The patient in whom treatment failed had aspiration pneumonitis that worsened despite 4 days of cefamandole administration. From a transtracheal aspirate were isolated E. aerogenes, B. melaninogenicus, and B. fragilis; the first two were susceptible to cefamandole, but the B. fr-agilis had an MIC > 64 ,ug/ml. Defining resistance to cefamandole is difficult. Levels in serum after i.v. infusion of 2 g of cefamandole for 10 to 30 min are between 147 and 240 ,ug/ml; however, the half-life after i.v. administration is about 0.7 h because the drug is rapidly cleared by the kidneys. At 4 h after




TABLE 6. Adverse reactions compared with duration of therapy Type of re-

action actlon No. of patients

Duration of therapy (days)


Range 17-40 6-49 2-18

Severea 7 29.0 Not severe 9 24.4 None 15 7.6 a Therapy discontinued at onset of reaction.

the 2-g dose, drug levels are barely demonstrable (14). In our study a 2-g dose was given over 10 to 20 min, usually every 4 h. Serum levels were measured in one patient and ranged from 96,ug/ ml at 15 min after the dose -to 3 ,ug/ml at 3 h after the dose. Grose et al. (6) have maintained mean serum levels of 25 to 30 Itg/ml by using a loading dose of 2 g i.v. followed by a 2-g i.v. dose over 6 h. Hence, serum concentrations of cefamandole vary greatly depending on dose, duration of administration, and the time blood levels are measured. During the study we isolated 89 anaerobes, of which 74 (83%) were inhibited by 32 Ag of cefamandole per ml and 80 (90%) by 64 yg/ml. Anaerobes resistant to 64 yg/ml included six B. fragilis group, one Bacteriodes pneumosintes, one Bifidobacterium breve, and one Eubacterium contortum; these bacteria were isolated from five patients. Except for one B. fragilis (isolated from the treatment failure) and the B. breve (isolated from a patient who could not be evaluated), the remainder of these bacteria were in three patients (lung abscess, salpingitis, and ruptured appendix with peritonitis), who clearly responded. Hence, we feel that future studies should include susceptibility data at higher drug concentrations than 64 ug/ml. From our data we can only report that patients with infections caused by anaerobes susceptible to 64 ,ug of cefamandole per ml, and some patients with infections caused by anaerobes resistant at this level, responded at a dose of 2 g i.v. given every 4 h. On the other hand, this is not a recommendation that the drug be given in such high dosages. The findings of Stone et al. (18) show the relative efficacy in anaerobic infection at smaller dosages (4 to 8 g/day). Rather, we feel there is a need to have MIC susceptibility data as complete as possible in drug efficacy studies. We considered surgical drainage of loculated purulent material to be an important part of therapy. Surgical drainage was performed in seven patients, and in two instances (two patients with thoracic empyemas) the drainage was inadequate. Neither of the patients responded to appropriate trials of chloramphenicol or clindamycin. One died, and the other re-

covered once drainage was established. In the surviving patient, we measured the empyema level of cefamand'ole while he received 2 g i.v. every 3 h. Interestingly, despite high peak serum levels, cefamandole was not detectable in the empyema fluid, and viable anaerobes were recovered from it. Whether the drug was undetectable because of inadequate penetration or inactivation of the antibiotic by the pus is not known. The incidence of significant adverse reactions, which included three cases of reversible neutropenia, one case of drug-related fever with eosinophilia, one case of eosinophilia, and one case of toxic epidermal necrolysis (1, 19), was higher in our study than incidences reported in other trials with cefamandole (12, 15). This finding is probably related to the large daily dose coupled with the long duration of treatment (Table 6); the mean duration of therapy at onset of these reactions was 28 days (17- to 40-day range). None of the side effects resulted in death. Overall, cefamandole appeared effective in treating most of our patients. It should be noted, however, the dosage used was 12 to 16 g per day i.v.. Although none of the drug's toxicities was irreversible or lethal, a significant number of adverse reactions were encountered with prolonged usage at this high dose level. Our in vitro data substantiate earlier reports (9, 10) that most anaerobes (excluding the B. fragilis group) are susceptible to cefamandole at c32 ,ug/ml. We believe cefamandole has a place in the treatment of anaerobic infection, but should not be regarded as a first-line drug. Its use should be undertaken with caution because strains of resistant B. fragilis group are not uncommon. ACKNOWLEDGMENTS We appreciate the cooperation of the Charity Hospital of New Orleans house staff, the technical assistance of K. McGarry and J. Mitchell, and the editorial assistance of C. Chapman, V. Howard, and L. Kutash. This work was supported by a grant from Eli Lilly and Co. LITERATURE CITED 1. Bailey, G., J. M. Rosenbaum, and B. Anderson. 1965.

Toxic epidermal necrolysis. J. Am. Med. Assoc. 191: 107-110. 2. Bailey, W. R., and E. G. Scott (ed.). 1970. Diagnostic microbiology, 3rd ed., p. 368. C.V. Mosby, St. Louis. 3. Bauer, A. W., W. M. M. Kirby, J. C. Sherris, and M. Turek. 1966. Antibiotic susceptibility testing by a standardized single disc method. Am. J. Clin. Pathol. 45: 493-496. 4. Ericsson, H. M., and J. C. Sherris. 1971. Antibiotic sensitivity testing: report of an intemational collaborative study. Acta Pathol. Microbiol. Scand. Sect. B Suppl. 217:1-90. 5. Ernst, E. C., S. Berger, M. Barza, N. V. Jacobus, and F. P. Tally. 1976. Activity of cefamandole and other cephalosporins against aerobic and anaerobic bacteria. Antimicrob. Agents Chemother. 9:852-855.

VOL. 15, 1979


6. Grose, W. E., G. P. Bodey, and D. Stewart. 1976. 7.

8. 9.

10. 11.


Observations in man on some pharmacologic features of cefamandole. Clin. Pharmacol. Ther. 20:579-584. Grove, D., and W. A. Randall. 1955. Assay methods for antibiotics. In Antibiotic monograph no. 2. Medical Encyclopedia, Inc., New York. Holdeman, L. V., and W. E. C. Moore. 1972. Anaerobe laboratory manual. Anaerobe Laboratory, Virginia Polytechnic Institute and State University, Blacksburg. Jones, R. N., and P. C. Fuchs. 1976. Comparison of in vitro antimicrobial activity of cefamandole and cefazolin with cephalothin against over 8,000 clinical bacterial isolates. Antimicrob. Agents Chemother. 9:1066-1069. Kaiser, G. V., M. Gorman, and J. A. Webber. 1978. Cefamandole-a review of chemistry and microbiology. J. Infect. Dis. 137(Suppl.):S10-S16. Lennette, E. HI, E. H. Spaulding, and J. P. Truant (ed.). 1974. Manual of clinical microbiology. American Society for Microbiology, Washington, D.C. Liu, C., D. R. Hinthorn, G. R. Hodges, W. Rosett, P. Gerjarusak, D. Dworzack, and J. Harms. 1978. Clinical evaluations of the efficacy and safety of cefamandole nafate, p. 811-812. In W. Siegenthaler and R. Luthy (ed.), Current chemotherapy: Proceedings of the 10th International Congress of Chemotherapy. American Society for Microbiology, Washington, D.C.


13. Meyers, B. R., and S. Z. Hirschman. 1978. Antibacterial activity of cefamandole in vitro. J. Infect. Dis.

137(Suppl.):S25-S31. 14. Neu, H. C. 1978. Comparison of the pharmacokinetics of






cefamandole and other cephalosporin compounds. J. Infect. Dis. 137(Suppl.):S80-S87. Perkins, R. L., R. J. Fass, J. F. Warner, and R. B. Prior. 1978. Cefamandole nafate evaluation in 74 patients, p. 815-817. In W. Siegenthaler and R. Luthy (ed.), Current chemotherapy: Proceedings of the 10th International Congress of Chemotherapy. American Society for Microbiology, Washington, D.C. Starge, M. D., F. S. Thompson, S. E. Phillips, G. L. Lombard, and V. R. Dowell, Jr. 1976. Modification of the Minitek Miniaturized Differentiation System for characterization of anaerobic bacteria. J. Clin. Microbiol. 3:291-301. Steers, E. E., E. L. Foltz, and B. S. Graves. 1959. An inocula replicating apparatus for routine testing of bacterial susceptibility to antibiotics. Antibiot. Chemother. 9:307-311. Stone, H. H., B. S. Guest, C. E. Geheber, and L. D. Kolb. 1978. Cefamandole in treatment of peritonitis. J. Infect. Dis. 137(Suppl.):S103-S109. Zak, F. G., M. J. Fellner, and A. J. Geller. 1964. Toxic epidermnal necrolysis (Lyell). Am. J. Med. 37:140-146.

Cefamandole therapy in anaerobic infections.

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Mar. 1979, p. 337-341 0066-4804/79/03-0337/05$02.00/0 Vol. 15, No. 3 Cefamandole Therapy in Anaerobic Infecti...
754KB Sizes 0 Downloads 0 Views