Cefoxitin: Antimicrobial Activity, Susceptibility, Administration and Dosage, Clinical Uses etc.

The cephamycins are often referred to as second-generation cephalosporin antibiotics, although they differ quite markedly from antibiotics such as cefuroxime or cefaclor. A primary attribute of the cephamycins is their resistance to a variety of beta-lactamase types, such as the extended-spectrum beta-lactamase (ESBLs) and their narrower-spectrum parents, TEM and SHV beta-lactamases. 

Structurally, the cephamycins are related to but distinct from cephalosporin C. Both contain a 7-alpha-methoxy group (Komiya et al., 1981; Ayers et al., 1982). Cefotetan possesses an N-methylthiotetrazole side chain (Cohen et al., 1987; Ward and Richards, 1989) (Figure 24.1). These antibiotics act on bacteria in a manner similar to other beta-lactam antibiotics.

ANTIMICROBIAL ACTIVITY

a. Routine susceptibility

The in vitro activity of cefoxitin is summarized in Table 24.2.

Gram-positive aerobic bacteria

Most staphylococci and streptococci, such as Staphylococcus aureus (including those resistant to penicillin G but not methicillin-resistant S. aureus), coagulase-negative staphylococci (except methicillinresistant strains), Streptococcus pyogenes, S. pneumoniae, group B streptococci, and the alpha-hemolytic streptococci (viridans streptococci), are cefoxitin susceptible (Neu, 1974; Wallick and Hendlin, 1974; Fong et al., 1976a; Stapley et al., 1979), although induction of resistance among S. aureus was described in the mid-1970s (Hoeprich and Huston, 1976). The activity of cefoxitin against these organisms is similar to that of cephalexin. Cefotetan is less active than cefoxitin against aerobic Gram-positive bacteria such as S. aureus and the streptococci (Ayers et al., 1982; Wise et al., 1982; Clarke and Zemcov, 1983). Some aerobic Gram-positive bacilli, such as Corynebacterium diphtheriae, are also cefoxitin susceptible.

The cephamycins lack activity against enterococci, as is the case with all cephalosporins. Indeed, Enterococcus faecalis (MIC typically 800 mg/ml) is more resistant to cefoxitin than to most other cephalosporins (Hamilton-Miller, 1974). Listeria monocytogenes is also resistant to the cephamycins (Moellering et al., 1974; Stapley et al., 1979).

Gram-negative aerobic bacteria

A feature of the cephamycins is their stability to ESBLs and to their parent beta-lactamases such as narrower-spectrum TEM and SHV beta-lactamases (Onishi et al., 1974). However, cephamycins will be hydrolyzed by AmpC beta-lactamases and most carbapenemases. Both cefoxitin and cefotetan have rapid bactericidal activity against most Enterobacteriaceae (Vuye et al., 1979; Brook, 1989; Goldstein et al., 1991a). Their stability to TEM and SHV beta-lactamases allows them to retain activity against many of the bacteria which are resistant to firstgeneration cephalosporins (Neu, 1974; Stapley et al., 1979). Thus, they typically have reliable activity against Escherichia coli, Proteus mirabilis, and Klebsiella spp. (Brumfitt et al., 1974; Neu, 1974; Norrby et al., 1976; Jackson et al., 1977; Stapley et al., 1979). Cefotetan is typically some 4- to 8-fold more active than cefoxitin against the Enterobacteriaceae such as E. coli, P. mirabilis, and Klebsiella (Chattopadhyay and Teli, 1982; Wise et al., 1982; Dette et al., 1983; Phillips et al., 1983; Neu, 1986). Cefotetan inhibits many Enterobacter and Citrobacter spp. strains which are cefoxitin resistant. However, when AmpC beta-lactamase production is increased, all cephamycins will be resistant (because they are hydrolyzed by AmpC).

The activity of cephamycins against Salmonella and Shigella spp. is also good, unless these strains produce plasmid-mediated AmpC betalactamases (Moellering et al., 1974; Neu, 1974). The cephamycins have poor activity against Acinetobacter spp., Pseudomonas aeruginosa, and Stenotrophomonas maltophilia (Ayers et al., 1982; Phillips et al., 1983; Wallick and Hendlin, 1974).

Anaerobic bacteria

Historically, the cephamycins had excellent activity against anaerobic bacteria, including the Bacteroides fragilis group. In an early survey in the USA of 750 clinical isolates of the B. fragilis group of anaerobic bacteria, only eight (2%) were cefoxitin resistant (Tally et al., 1983). By the late 1980s/early 1990s, resistance was described by a number of authors (Appelbaum et al., 1990; Aldridge and Stratton, 1991) and at least two US studies found resistance rates of approximately 10% (Cuchural et al., 1990; Appelbaum et al., 1991). Rates of resistance of B. fragilis have continued to rise.

In a recent longitudinal survey in the USA, rates of resistance to cefoxitin were reported to be 10.3% in B. fragilis group, with 5.2% in B. fragilis and 15.7% in non-B. fragilis species (Snydman et al., 2007). The most recent survey from Europe demonstrated a 38% rate of resistance to cefoxitin (Wybo et al., 2007). Resistance to cefoxitin and/or cefotetan is higher among non-B. fragilis species than B. fragilis of the B. fragilis group in some studies (Werner, 1983; Fox and Phillips, 1987; Snydman et al., 2007; Teng et al., 2002). Regional differences in key anaerobes are summarized in Table 24.3.

Other organisms

The cephamycins often have clinically useful activity against rapidly growing mycobacteria (Mycobacterium abscessus, M. chelonae and M. fortuitum). Indeed, cefoxitin susceptibility testing is recommended routinely for rapidly growing mycobacteria to aid in identification and treatment; M. abscessus is susceptible to cefoxitin, approximately half of M. fortuitum isolates are susceptible, whereas M. chelonae is usually resistant (Griffith et al., 2007).

The majority (99%) of M. abscessus isolates are susceptible to cefoxitin, with in vitro breakpoints of MIC less than or equal to 16 mg/ ml (Park et al., 2008). Some strains (o50%) of M. fortuitum and M. chelonae can be inhibited by 16 mg/ml of cefoxitin (Casal and Rodriguez, 1982; Cynamon and Palmer, 1982; Swenson et al., 1985; Finch, 1986). Cefoxitin is inactive against N. asteroides, with most strains needing 50 mg/ml or more for inhibition (Cynamon and Palmer, 1981; Gutmann et al., 1983). Cefoxitin lacks activity against Chlamydia trachomatis (Bowie, 1982).

b. Emerging resistance and cross-resistance

As mentioned above, ESBLs do not hydrolyze cephamycins with the 7 alpha-methoxy groups, such as cefoxitin or cefotetan. However, types of plasmid-mediated ESBLs have appeared in some Enterobacteriaceae, which can hydrolyze cefoxitin and cefotetan (Jacoby and Archer, 1991). Cefoxitin acts as an inducer of AmpC beta-lactamase production. By this mechanism, it is possible that cefoxitin resistance may emerge during treatment of patients (Sanders, 1983; Seeberg et al., 1983; Moritz and Carson, 1986; Turgeon et al., 1994).

The most common mechanism of resistance to penicillins is the chromosomally encoded class A/group 2e cephalosporinase (product of the cepA gene) (Rogers et al., 1993). This enzyme is active against penicillins and first-generation cephalosporins, but not against cephamycin or carbapenem antibiotics. Resistance to cefoxitin is contributed by CfxA, a class A/group 2e cephalosporinase encoded by the cfxA gene, which has been shown to be distantly related to the B. fragilis endogenous cepA (Parker and Smith, 1993). This resistance gene is harbored in a mobilizable transposon (Tn4555) (Ferreira et al., 2007). The CfxA enzyme confers resistance to cephamycins and all other b-lactam antibiotics, except the carbapenems (Parker and Smith, 1993). The B. fragilis group has been shown to rarely produce a class B metallo-blactamase, encoded by the cfiA gene (also known as the ccrA gene) which causes resistance to all beta-lactam antibiotics, including cephamycins and carbapenems (Thompson and Malamy, 1990; Bandoh et al., 1991).

c. In vitro synergy and antagonism

If cefoxitin is used with another beta-lactam antibiotic that is not stable to the AmpC beta-lactamases produced by Gram-negative bacilli, cefoxitin can antagonize the other beta-lactam antibiotic by inducing AmpC production. This antagonism with cefoxitin has been demonstrated in vitro and in animals (Kuck et al., 1981; Goering et al., 1982; Sanders et al., 1982; Miller et al., 1983). Cefoxitin appears to cause antagonism more frequently than many other cephalosporins.

MODE OF DRUG ADMINISTRATION AND DOSAGE

The existing cephamycins can only be administered parenterally. Intravenous administration is preferred over intramuscular injection. Studied intravenous infusion length has been over periods of 3–120 minutes (Goodwin et al., 1974; Geddes et al., 1977; Feldman et al., 1980). More prolonged infusions have not been studied.

a. Adults

The adult dosage of cefoxitin can be varied according to the nature and severity of the infection, from 1 g 8-hourly to 2 g 6-hourly. Dosage as high as 2 g 4-hourly has been used (Geddes et al., 1977; Heseltine et al., 1977).

The main difference between cefotetan and cefoxitin is that the former has a more prolonged serum elimination half-life (approximately 3.5 hours), so that it can be administered intramuscularly or intravenously at 12-hourly intervals (Carver et al., 1989). Generally, a cefotetan dosage of 2 g 12-hourly should be routinely used for moderate and severe systemic infections (Nakagawa et al., 1982; Cox et al., 1983; Nolen et al., 1983). A dosage of 3 g 12-hourly should not be exceeded. b. Newborn infants and children The cefoxitin dose for children ranges from 15 mg/kg 8-hourly to 30 mg/kg 6- to 4-hourly. For the treatment of children aged 3–15 months with severe infections, Feldman et al. (1980) found a dose of 150 mg/kg/day (37.5 mg/kg 6-hourly) to be satisfactory. Cefotetan can be administered at a dose of 30 mg/kg 12-hourly if the patient is over six months of age (Martin et al., 1994). No information is available pertaining to dosing in premature neonates.

PHARMACOKINETICS AND PHARMACODYNAMICS

a. Bioavailability

Both cefotetan and cefoxitin are parenteral antibiotics and are therefore 100% bioavailable. The serum half-life for cefotetan in patients with normal renal function is approximately 3.5 hours and 39 minutes for cefoxitin (Kampf et al., 1981). Binding of cefoxitin to serum proteins is approximately 20% (O’Callaghan, 1975).

b. Drug distribution

If a 2-g dose of cefoxitin is administered intravenously over 3 minutes to adults, the mean peak serum level attained at approximately 5 minutes is 222.6 mg/ml. A concentration of 3.4 mg/ml is still detectable at 3 hours. When the same dose of cefoxitin is given as a 30-minute intravenous infusion, the peak serum level (immediately after the infusion) is lower, but subsequent serum concentrations are slightly more sustained (Goodwin et al., 1974). Doubling the dose of cefoxitin virtually doubles serum concentrations (Brumfitt et al., 1974; Geddes et al., 1977).

Cefoxitin does not penetrate into normal cerebrospinal fluid (CSF) (Geddes et al., 1977). After large parenteral doses to patients with bacterial meningitis, moderate CSF concentrations of the drug are found, but these are not always high enough to inhibit susceptible strains of H. influenzae and S. pneumoniae (Nair et al., 1979; Humbert et al., 1980; Feldman et al., 1982). Cefotetan CSF concentrations are invariably low (Martin et al., 1994).

Parenterally administered cefoxitin penetrates quite well into the peritoneal fluid of surgical patients (Wise et al., 1981), and into normal human interstitial fluid, where concentrations are very similar to serum levels (Gillett and Wise, 1978). It also reaches therapeutically effective levels in pelvic tissue of patients undergoing abdominal hysterectomy (Bawdon et al., 1982). Cefoxitin diffuses poorly into experimental B. fragilis intra-abdominal abscesses in animals, producing a concentration of only 2% of the simultaneous serum level (O’Keefe et al., 1979). Cefoxitin penetrates into normal lung and bone tissue, but these tissue levels are considerably lower than simultaneous serum levels (Summersgill et al., 1982; Perea et al., 1983). The cefoxitin concentration in breast milk of one patient collected 2 hours after an i.v. dose of 1 g was 5.6 mg/ml (Geddes et al., 1977).

c. Clinically important pharmacokinetic and pharmacodynamic features

Like other cephalosporins, the cephamycins are time-dependent killers. However, exploitation of this by use of continuous or extended infusion has not yet been shown to be clinically advantageous. A study of continuous versus intermittent cefoxitin infusion after colorectal surgery showed no significant differences in clinical outcome (Suffoletta et al., 2008).

d. Excretion

Cefoxitin and cefotetan are predominantly excreted unchanged in the urine (Ohkawa et al., 1983). They are excreted by both glomerular filtration and active tubular secretion. Probenecid delays, but does not diminish, its excretion (Vlasses et al., 1980; Arvidsson et al., 1981). About 90% or more of a parenterally administered dose can be recovered from the urine as active unchanged drug during the following 12 hours (Brumfitt et al., 1974). High cefoxitin concentrations are attained in urine; after a 500-mg dose, these are 1000– 3500 mg/ml during the first 3 hours, and 22–350 mg/ml in the succeeding 9 hours (Kosmidis et al., 1973). The renal elimination of cefoxitin is unaffected by concomitant administration of furosemide (frusemide) (Trollfors and Norrby, 1980).

In some subjects, cefoxitin is deacylated in the body to detectable amounts of antibacterially inactive decarbamoyl-cefoxitin. This only occurs to a minor extent, as the metabolite can only be found in the urine after a delay of several hours, and it always accounts for less than 2% of the administered dose (Goodwin et al., 1974; Schrogie et al., 1979).

TOXICITY

Eosinophilia (Heseltine et al., 1977) and rash (McCloskey, 1977) can occur with any of the cephamycins. A small number of patients have developed a positive Coombs’ test during cefoxitin therapy (Heseltine et al., 1977), and rarely hemolytic anemia and pancytopenia have been reported (De Torres, 1983). An autoimmune hemolytic anemia has been observed with cefotetan (Chenoweth et al., 1992). At abnormally high concentrations, cefoxitin may suppress ADP-induced platelet aggregation (Bang and Kammer, 1983). Surprisingly, in a clinical study, Brown et al. (1986) observed clinical bleeding in 8.2% of cefoxitintreated patients. The exact cause for this was not determined.

During initial pharmacologic studies in human volunteers, cefoxitin appeared free from nephrotoxicity (Kosmidis et al., 1973; Brumfitt et al., 1974). Subsequent large-scale clinical studies showed that cefoxitin nephrotoxicity was extremely uncommon (Neu, 1979). It does not aggravate pre-existing renal failure, provided that appropriate doses are given (Trollfors et al., 1978; Trollfors, 1980). In animals, cefoxitin is not nephrotoxic and does not potentiate aminoglycoside nephrotoxicity (Ormrod and Miller, 1981; Viotte et al., 1981). Cefoxitin and cefotetan are category B in pregnancy and are felt to have minimal risk to the breastfeeding infant if used as perioperative prophylaxis in cesarean sections (Roex et al., 1987).

CLINICAL USES OF THE DRUG

a. Surgical chemoprophylaxis

Cefoxitin and cefotetan have long been used as perioperative prophylaxis for colonic surgery. Cephamycin prophylaxis has also been used for penetrating abdominal injuries (Dellinger, 1991; Dellinger et al., 1994) and for prophylaxis of obstetric and gynecologic infections (Counts, 1988). The role of cephamycins as perioperative prophylaxis for colonic surgery has been brought into recent perspective via an industry-sponsored randomized multicenter double-blind trial, which assessed the comparative efficacy and safety of prophylactic use of cefotetan versus ertapenem in patients undergoing elective colorectal surgery. A modified intentionto-treat analysis showed a lower prophylactic antibiotic failure rate in the ertapenem arm compared with the cefotetan arm (failure rates 40.2% versus 50.9%; difference  10.7%; 95% CI–17.1 to  4.1). Of the pathogens that were recovered in postoperative infections, 66.7% in the cefotetan group were resistant to cefotetan, whereas 16.3% in the ertapenem group were resistant to ertapenem.

b. Intra-abdominal infections

As the cephamycins have historically had efficacy against many Gramnegative aerobic bacilli (such as E. coli) and Gram-negative anaerobes (such as B. fragilis), they have been extensively used for intraabdominal infections such as peritonitis. Early experiments with induced intra-abdominal sepsis in animals have shown that cefoxitin was as effective as other drug regimens, such as clindamycin– gentamicin, metronidazole–gentamicin, and imipenem (Bartlett et al., 1981; Joiner et al., 1982; Bartlett et al., 1983; Nichols, 1983; Lau et al., 1986; Tanner et al., 1986). Clinical studies performed more than 30 years ago confirmed this efficacy (Geddes et al., 1977; Nair and Cherubin, 1978; Tally et al., 1979; Gorbach and McGowan, 1981). However, even in comparatively early work, the efficacy was dependent on the Bacteroides spp. strain involved being cefoxitin susceptible (Snydman et al., 1992).

c. Anaerobic infections

In the early years of their use, the cephamycins were successfully used for anaerobic infections in cancer patients (Klastersky et al., 1979), infections of the female genital tract (Rosene et al., 1986; Counts, 1988), and pleuropulmonary infections (Le Frock et al., 1982). The increasing prevalence of cephamycin resistance in anaerobes has mirrored the poorer outcome of patients with Bacteroides bacteremia when treated with cefotetan. Cephamycin resistance was associated with previous use of b-lactams with anti-anerobic therapy (Nguyen et al., 2000). The caveats mentioned above regarding current susceptibilities of anaerobes to the cephamycins bear repeating.

d. Gram-negative aerobic infections

There are a small number of published reports of the use of cephamycins (for example, cefoxitin, cefmetazole, flomoxef) in the treatment of ESBL-producing Gram-negative bacterial infections. These reports number fewer than ten patients in total (Pangon et al., 1989; Siu et al., 1999; Lee et al., 2006). In a group of seven patients in Taiwan, outcomes with flomoxef for infections due to ESBL-producers were equivalent to carbapenems (Lee et al., 2006). However, selection of porin- resistant mutants occurring during therapy, resulting in cefoxitin resistance and relapse of infection, has been described (Pangon et al., 1989). In addition, combined cephamycin and carbapenem resistance in K. pneumoniae has been observed in the setting of widespread cephamycin use in response to an outbreak of infection with ESBL-producing organisms (Bradford et al., 1997).

e. Infections due to Mycobacterium abscessus, Mycobacterium fortuitum, and Mycobacterium chelonae

The cephamycins find a place in the early treatment of rapidly growing mycobacterial infections when parenteral therapy is required. Typically, this is in combination with amikacin or macrolides. Recommendations for initial therapy for M. abscessus infections of skin, soft-tissue, and bone infections include combination therapy of amikacin and highdose cefoxitin (up to 12 g/d intravenously in divided doses) for at least 2 weeks, dependent on a confirmed clinical response. Osteomyelitis may need up to six months of therapy. Oral therapy (for example, with clarithromycin) may be considered after the initial parenteral regimen, depending on susceptibilities (Griffith et al., 2007). Cefoxitin in a dose of 12 g intravenously daily plus 2–4 g oral probenecid has been used for the treatment of nonpulmonary infections by these mycobacteria. For the initial 2–4 weeks, the drug is often combined with amikacin, followed by cefoxitin therapy alone for 10–12 weeks (Wallace et al., 1985; Raad et al., 1991). Pulmonary infection with M. chelonae has been treated successfully by i.v. cefoxitin and oral ciprofloxacin (Singh and Yu, 1992). A series of 19 cases of M. fortuitum endocarditis showed a high mortality rate (85%) regardless of therapy (Olalla et al., 2002).

References

Arvidsson A, Borga˚ O, Kager L, Pieper R (1981). Renal elimination of cefoxitin and effect of probenecid after single and repeated doses. J Antimicrob Chemother 7: 423.
Ayers LW, Jones RN, Barry AL et al. (1982). Cefotetan, a new cephamycin: comparison of in vitro antimicrobial activity with other cephems, betalactamase stability, and preliminary recommendations for disc diffusion testing. Antimicrob Agents Chemother 22: 859.
Bach VT, Roy I, Thadepalli H (1977). Susceptibility of anaerobic bacteria to cefoxitin and related compounds. Antimicrob Agents Chemother 11: 912.
Bandoh K, Muto Y, Watanabe K et al. (1991). Biochemical properties and purification of metallo-beta-lactamase from Bacteroides fragilis. Antimicrob Agents Chemother 35: 371.
Bang NU, Kammer RB (1983). Hematologic complications associated with beta-lactam antibiotics. Rev Infect Dis 5 (Suppl 2): 380.
Bartlett JG, Louie TJ, Gorbach SL, Onderdonk AB (1981). Therapeutic efficacy of 29 antimicrobial regimens in experimental intraabdominal sepsis. Rev Infect Dis 3: 535.
Cuchural Jr GJ, Tally FP, Jacobus NV et al. (1990). Comparative activities of newer beta-lactam agents against members of the Bacteroides fragilis group. Antimicrob Agents Chemother 34: 479.
Cynamon MH, Palmer GS (1981). In vitro susceptibility of Nocardia asteroides to N-formimidoyl thienamycin and several cephalosporins. Antimicrob Agents Chemother 20: 841.
Dellinger EP (1991). Antibiotic prophylaxis in trauma: penetrating abdominal injuries and open fractures. Rev Infect Dis 13 (Suppl 10): 847.
Dellinger EP, Gross PA, Barrett TL et al. (1994). Quality standards for antimicrobial prophylaxis in surgical procedures. Clin Infect Dis 18: 422.
Fox AR, Phillips I (1987). The antibiotic sensitivity of Bacteroides fragilis group in the United Kingdom. J Antimicrob Chemother 20: 477.
Geddes AM, Schnurr LP, Ball AP et al. (1977). Cefoxitin: a hospital study. BMJ 1: 1126.
Gillett AP, Wise R (1978). Penetration of four cephalosporins into tissue fluid in man. Lancet 1: 962.
Goering RV, Sanders CC, Sanders Jr WE (1982). Antagonism of carbenicillin and cefamandole by cefoxitin in treatment of experimental infections in mice. Antimicrob Agents Chemother 21: 963.