Bacteroides species

Monograph
Tables
What's New
Reviews
History

MICROBIOLOGY

Bacteroides spp are non-spore forming gram-negative bacilli that are part of the human resident flora. Microbiologically, they are distinguished from other genera by growth in 20% bile. At present, the Bacteroides fragilis group consists of ten species: B. fragilis (the most frequent isolate), B. distasonis, B. thetaiotaomicron, B. vulgatus, B. ovatus, B. eggerrthii, B. merdae, B. stercoris, B. uniformis, and B. caccae. Since 1990, many organisms previously designated as Bacteroides have been reclassified (see chapter on Anaerobes other than Bacteroides). Bacteroides, the predominant genus in the human intestine, are important in numerous metabolic activities and may provide some level of protection from invasive pathogens. All 10 species are usually isolated from the colon, although infections caused by or associated with them can include virtually any organ.

Isolation and identification of Bacteroides spp pose a hurdle to many clinical laboratories; most only identify the genus. Other laboratories may identify B. fragilis species, and lump all others into the B. fragilis group. Further, if more than three anaerobic organisms are isolated in a clinical specimen, many laboratories do not perform further identification. Those laboratories that do identifyBacteroides frequently do not perform susceptibility testing. Lack of identification and susceptibility testing of anaerobes can be attributed, in part, to its expense and the lack of timeliness in providing relevant information to the clinician (48). Thus, choice of therapy for infections that may involve Bacteroides is empiric.

EPIDEMIOLOGY

Bacteroides fragilis are endogenous organisms of the GI tract. Spread of strains among patients is not known, although this topic has not been well studied. Thus, infections due to this organism are most likely caused by endogenous strains.

CLINICAL MANIFESTAIONS

The hallmarks of nearly any infection involving Bacteroides spp include abscess formation, and they are frequent isolates in polymicrobial infections. Typical sites of polymicrobial infections involving Bacteroides include the abdomen and pelvis, perirectal, skin and soft tissue, and solid organs. Although isolation of Bacteroides spp as the sole pathogen can occur, it is unusual. Infections where single organism isolation is most commonly associated with Bacteroides include endocarditis, meningitis, septic arthritis and osteomyelitis (36).

LABORATORY DIAGNOSIS

Bacteroides fragilis may be isolated as a single agent, such as in blood cultures, or more typically from mixed infections. The organism is aerotolerant, but requires an anaerobic environment to propagate. Simple identification from blood cultures include Gram stain and growth on blood agar and Bacteroides-bile-esculin (BBE) agar for isolation and presumptive identification ofBacteroides fragilis group (as well as Bilophila wadsworthia). B. fragilis will appear as dark colonies with brown-black halos on BBE agar due to the hydrolysis of esculin. B. fragilis can be further presumptively identified by resistance to kanamycin, vancomycin and colistin, using a disk test, and will grow in 20% bile, produce catalase (most strains), and is variably indole positive. Many laboratories will confirm the identification using a rapid identification kit or individual fermentation reactions (20).

PATHOGENESIS

Multiple virulence factors have been implicated in the pathogenesis of this organism. They include the capsular polysaccharide (which inhibits opsonophagocytosis and promotes abscess formation), pili and fimbriae (promotes adherence), and production of a number of different enzymes (hyaluronidase, hemolysin, peroxidase, collagenase, protease, heparinase, and neuraminidase). In addition, superoxide dismutase and catalase also considered virulence factors. These enzymes defend B. fragilis against oxygen radicals and increase aerotolerance (14,30,57,82,114,119).

SUSCEPTIBILITY IN VITRO AND IN VIVO

In general, Bacteroides spp grow well on enriched media in an anaerobic environment. Aero tolerance is greater in this genus than more fastidious anaerobes. Historically, susceptibility testing of Bacteroides has utilized a variety of media and methods. The National Committee for Clinical Laboratory Standards (NCCLS) recommends the agar dilution method as the reference standard (76). The testing medium recommended is Brucella agar supplemented with lysed sheep blood, Vit K1, and hemin, which supports the growth of virtually all anaerobes. This method is generally time consuming and expensive to perform, leading to its infrequent use. The alternative method recommended is the broth micro dilution technique (76). In addition to the NCCLS approved methods, the FDA has also approved use of the E-Test gradient method, which is a more rapid but somewhat expensive method (21,94). Currently, the NCCLS recommends surveillance testing on an annual basis to monitor susceptibility patterns as a guide for empiric antibiotic choice. Susceptibility testing for individual patients, isolates, should be performed when there is known resistance of a particular species, failure of a usual regimen, a pivotal role of the antimicrobial agent in determining outcome, and the need for long term therapy. In particular, susceptibility testing is indicated for isolates from brain abscess, meningitis, osteomyelitis, joint infections, infection of prosthetic devices or vascular grafts, and refractory bacteremia (76).

For the purpose of this chapter, in vitro data for each of the antibiotics were chosen from publications using NCCLS agar dilution or broth microdilution methodology. Intermediate and susceptible break points for each antibiotic with activity against Bacteroides were set by the NCCLS (76). The intermediate range was established because of the difficulty in reading end points and the clustering of MICs at breakpoint concentrations. With maximum dosages of appropriate antibiotics, organisms with susceptible or intermediate endpoints are generally amenable to therapy. Susceptibility among the different species of the B. fragilis group can vary significantly. Table 1 contains susceptibility results for the most frequently isolated species from various recent publications.

Single Drugs

Penicillins: Penicillin hydrolyzing activity in B. fragilis group members was first recognized in 1968 (87). Greater than 90% of clinical isolates produce beta-lactamases that are predominantly active against cephalosporins, have high activity, are cell associated, and are produced constitutively (3,25,81,116). Thus most Bacteroides isolates are resistant to penicillin G and ampicillin, but may remain susceptible to cephamycins, and extended spectrum penicillins. Table 1 illustrates the comparative activity of ticarcillin, and piperacillin. Ticarcillin is moderately active against B. fragilis group members, while piperacillin demonstrates significantly greater activity. However, recent data has demonstrated steadily decreasing activity of piperacillin over the previous decade, noted by increasing MIC50 and MIC90 values (103). Susceptibility may range from 30-90%, depending upon the species tested (50), with B. thetaiotaomicron and B. distasonis the most resistant.

Cephalosporins: Three cephamycins (cefoxitin, cefotetan, cefmetazole) and two extended spectrum cephalosporins (ceftizoxime, moxalactam) have demonstrated generally good activity against B. fragilis group members, although marked differences among species can be noted for some agents. Cefoxitin is the most potent in this class, with 90% of strains below the susceptible or indeterminate breakpoint, although resistance has been noted to occur in clusters (24). Its activity is similar against all members of the B. fragilis group. The two other cephamycins, cefotetan and cefmetazole, have similar activity compared with cefoxitin against B. fragilis. However, their activity against the non-fragilis members of the B. fragilis group is significantly lower, and species dependent (Table 1) (23). Ceftizoxime has been shown to have activity against most Bacteroides spp. Early studies demonstrated activity similar to that of cefoxitin, but more recent data suggest resistance to this agent is increasing (52,80,104,123). Moxalactam also has excellent in vitro activity, but is little used due to serious side effects. Broad spectrum cephalosporins, including cefaclor, cefuroxime, cefotaxime, cetazidime, cefpodoxime, cefepime have generally poor activity against the B. fragilis group, and are generally not considered as effective against these organisms (106).

Carbapenems: The carbapenems, imipenem and meropenem, are resistant to hydrolysis by a number of beta-lactamases, including those of Bacteroides spp. Thus, both agents demonstrate excellent activity against all species within the B. fragilis group. MIC90 values for both imipenem and meropenem are 2 μg/ml or less (50,88,104). Overall, the susceptibility of B. fragilis group members to the carbapenems is 99.6 % in the United States. Most resistant strains produce a zinc metalloenzyme beta-lactamase (58). A significantly higher level of resistance has been reported from Japan, which is attributed to greater usage of imipenem in that country (5). Resistance to carbapenems also confers cross-resistance to all beta-lactam agents, including the beta-lactam/beta-lactamase inhibitor combinations.

Clindamycin: Historically, B. fragilis group members have been highly susceptible to clindamycin, and its predecessor lincomycin (23,39). Throughout the 1970's, clindamycin demonstrated the greatest activity with MIC90 of 4 μg/ml or less for most clinical isolates (109). However, by the late 1970's, isolates with high-level resistance to clindamycin emerged (96). In 1979, three laboratories reported the identification of a clindamycin resistance gene on transferable plasmids (97). Further, these genes were found to share >95% homology among their coding sequences (90). Subsequently, resistance among Bacteroides spp has increased in both the United States, and worldwide. In one report, resistance at four Chicago area medical centers was > 25% (52). High-level clindamycin resistance is also reported worldwide (6). Among the different species, B. fragilis, B. vulgatus, are the most susceptible (MIC50 = 1 and 0.25 μg/ml, respectively), although the MIC90 for both species was >256 (50). A national survey has confirmed a statistically significant increase in resistance among all members of the B. fragilis group from 1990 to 1996 (103), with a predicted 15.1% and 8.18% annual increase for B. fragilis and other members of the B.fragilis group, respectively.

Chloramphenicol: Chloramphenicol is very active against all members of the B. fragilis group. Resistance has been reported rarely, and is due to either inactivation by a nitroreductase or chloramphenicol acetyltransferase, and resistance transfer has been reported (10,51,95). Use of this drug for clinical infections involving anaerobes is generally limited, primarily due to concerns about toxicity.

Aminoglycosides: Bacteroides are uniformly resistant to aminoglycosides. This is likely due to the failure to transport the amino glycosides into bacterial cells and their inability to reach the ribosomal target site.

Metronidazole: Nitroimidazoles, including metronidazole, have a high degree of activity against Bacteroides, with MIC90 values < 1 μg/ml (73). They are bactericidal and penetrate well into abscesses frequently making them the ideal candidates in therapy. Resistance to metronidazole is extremely rare and has been limited primarily to strains isolated in France (92). The resistance determinant has been localized to two different transferable plasmids (91), suggesting that resistance may increase over time.

Fluoroquinolones: The activity of several fluoroquinolones have been evaluated against anaerobes including the B. fragilis group. Ciprofloxacin, ofloxacin, levofloxacin, sparfloxacin, and grepafloxacin have generally fair to poor activity against members of the B. fragilis group with MIC90 values that range from 2 μg/ml -32 μg/ml, although they demonstrate greater activity against nonBacteroides anaerobes (47,53). Trovafloxacin has significantly greater activity against all members of the B. fragilis group (MIC90 0.39μg/ml - 2 μg/ml) but has very limited clinical indications due to toxicity. Moxifloxacin has similar activity to that of trovafloxacin, and is currently in clinical trials. (1,56). Investigational agents, including DU6859a and BM5284756 also appear very active in vitro, and await results of clinical trials (1,56,126,127).

Tetracyclines: Once a mainstay in therapy of anaerobic infections in the 1960's, more than 80% of Bacteroides isolates are now resistant (35). Tetracycline derivatives, the glycylcyclines, demonstrate excellent in vitro activity, but are still in early investigational studies (46,125).

Combination Drugs

Penicillins Plus Beta-lactamase Inhibitors: Most of the beta-lactamases produced by B. fragilis group members are inhibited by sulbactam, clavulanic acid, and tazobactam. As a result, nearly all isolates of the B. fragilis group are susceptible to ampicillin/sulbactam, ticarcillin/clavulanate, and piperacillin/ tazobactam. All three combinations demonstrate very potent activity against members of the B. fragilis group, although strains of B. distasonis that do not produce beta-lactamases have somewhat higher MIC values (Table 1) (61,124). Susceptibility of Bacteroides to these agents has not changed significantly from 1990-1996 (103).

ANTIMICROBIAL THERAPY

General

The principles of treatment for infection involving members of the B. fragilis group are a) making the environment such that anaerobic bacteria find it difficult to proliferate, b) checking the spread of anaerobic bacteria into healthy tissues, c) neutralizing the toxins of anaerobes, and d) supportive care (37). Control of the local environment includes debridement of necrotic tissue, drainage of pus collections, and improvement in oxygenation. For intra-abdominal abscesses, this can be performed via surgery or percutaneously. Failure of a patient to improve within 48 hours after initial drainage is an indication for repeat radiographic scanning and drainage, or laparotomy. When a drain has been placed, criteria for removal of the catheter includes clinical resolution of sepsis, minimal drainage from catheter, and radiographic evidence of abscess resolution.

Antimicrobial therapy is an important adjunct to removal of infected tissues. Most infections involving B. fragilis are polymicrobial infections, involving aerobes and anaerobes. Thus, antibiotic therapy usually requires activity against B. fragilis, along with enteric gram -negative and/or gram-positive organisms, depending upon the site of isolation. In general, maximum dosages of antimicrobial agents recommended the manufacturers are suggested for infections involving B. fragilis group members (76).

Intra-abdominal and Pelvic Infections

B. fragilis is isolated in 70% of cases of intra-abdominal infection involving abscesses (7). Early on, animal models demonstrated the importance of anaerobes, particularly B. fragilis, in intra-abdominal infection and abscess formation (83). The first clinical trial in humans was reported in 1973, comparing cephalothin/kanamycin to clindamycin/kanamycin (117). The infectious complication rate due to anaerobes noted in the cephalothin/kanamycin group was 21% compared with 2% in the clindamycin/kanamycin group. This study established establishing that treatment of the anaerobe was important, and clindamycin/aminoglycoside became the "gold standard" to which other regimens were compared. Treatment of facultative bacteria is also important (19,31). Interested readers are referred to an excellent review for a summary of comparative clinical trials (49). Of particular note are the studies that do not require an aminoglycoside in combination with clindamycin, and the use of monotherapy (69,74).

Drugs of Choice: Table 2 lists suitable agents determined to be effective in clinical trials, either a monotherapy or in combination. Monotherapy with a carbapenem (i mipenem, 500 mg i.v. q6h;meropenem 1g i.v. q8h), beta-lactam/beta-lactamase inhibitor combinations (ampicillin/sulbactam 3g i.v. q6h; ticarcillin/clavulanate 3.1g i.v.q 8h; piperacillin/tazobactam 3.75 g i.v. q6h), ormetronidazole (500 mg i.v. or p.o. q6h) combined with either an aminoglycoside or a third generation cephalosporin could be considered the most active. Clindamycin (900 mg i.v. q8h)/aminoglycoside, piperacillin (3g i.v. q6h)/aminoglycoside, cefoxitin (2g i.v.q6h), cefotetan (2g i.v. q12h), and ceftizoxime (2g i.v. q8h) might now be considered second line therapy due to concerns about resistance among aerobic and anaerobic organisms. The principle behind all agents listed as monotherapy or in combination is to treat B. fragilis, other anaerobes, andEnterobacteraciae. However, the choice of antimicrobial therapy may require modification depending upon the clinical setting. For example, a patient that develops an intra-abdominal infection while in hospital may be colonized with more resistant gram-negative organisms, such as Pseudomonas or resistant Enterobacter spp.

Antimicrobial therapy for pelvic infections involving Bacteroides spp is similar to that of intra-abdominal therapy. Polymicrobial infections that include B. fragilis group organisms involving the female genital tract the pelvis include soft tissue infection following episiotomy, postpartum endometritis, pelvic abscess, tuboovarian abscess, and acute pelvic inflammatory disease. Most agents listed in table 2 have demonstrated similar clinical efficacy against B. fragilis as part of a polymicrobial infection in this setting (67,111,113). Failures associated with these regimens are usually due toEnterococcus. A specific regimen is recommended for pelvic inflammatory disease. Current recommendations from the Centers for Disease Control and Prevention include cefoxitin (2g i.v. q6h) or cefotetan (2g i.v. q12h) plus doxycycline (100 mg p.o. bid), or clindamycin (900mg i.v. q8h) plus gentamicin followed by doxycycline (70). These regimens are recommended to treat C. trachomatisand N. gonorrhoeae in addition to B. fragilis and other anaerobes.

Bacteremia

B. fragilis is the most common anaerobe isolated from blood cultures, although the frequency of positive cultures is decreasing overall (11,28,122). In virtually all cases, isolation of a member of the B. fragilis group in blood indicates underlying infection (122), and is associated with 60% mortality if untreated (18). The source of bacteremia is most commonly intra-abdominal, female genital tract, or soft tissue (38,41,85).

Drug of Choice: Empiric antimicrobial therapy of Bacteroides bacteremia includes parenteral agents that are effective against B. fragilis and aerobic organisms, depending upon the source of bacteremia (Table 2). Specific choices of an agent against Bacteroides should be guided by local patterns of resistance, or obtaining the susceptibility results for the isolate. Bacteroides strains demonstrating higher or resistant MICs were correlated with treatment failure and a trend towards increased mortality (78,102). Identification and drainage of the source of bacteremia remains a mainstay of nonantibiotic therapy.

Skin and Soft Tissue

B. fragilis is not considered part of the normal skin flora. However, these organisms are an important cause of disease when soft tissues are damaged and in diabetes. The most serious infections involving B. fragilis include synergistic necrotizing fasciitis and Fournier's gangrene. Other sites of infection include wound infections following abdominal or pelvic surgery, pilonidal cysts, decubitus wounds, and diabetic foot ulcers (44,107). The role of anaerobes in an uncomplicated diabetic foot infection has been debated, with only 13% of these infections associated with anaerobes (mostly gram-positive) in one study (4,42).

Drug of Choice: Treatment of B. fragilis in serious soft tissue infection requires parenteral therapy resembling that used for abdominal infections (Table 2). Thus, empiric coverage for gram-negative and gram-positive anaerobes, plus enteric gram-negative organisms is required. Empiric coverage for S. aureus and Streptococci may be considered in more serious infections until culture results are available. Ampicillin/sulbactam (3g i.v. q6h)+ aminoglycoside, ticarcillin/clavulanate (3.1g i.v. q8h), piperacillin/tazobactam (3.75g i.v. q8h), i mipenem (500mg i.v. q6h) or meropenem (1g i.v. q8h), or a combination of metronidazole (500 mg i.v. or p.o. q6h)/aminoglycoside (if Staphylococci or Streptococci are not suspected) may be suitable for empiric therapy. Patients who have been hospitalized may have resistant gram-negative aerobic bacteria as well, with an antimicrobial choice modified accordingly. Changes in the empirical regimen should be guided by culture results. For less serious soft tissue infections, such as the diabetic foot, parenteral or oral agents can be used. Currently, recommendations for empirical parenteral therapy include beta-lactam/beta-lactamase inhibitor combinations (ampicillin/sulbactam 3g i.v. q6h; ticarcillin/clavulanate 3.1g i.v.q 8h; piperacillin/tazobactam 3.75 g i.v. q6h), clindamycin (900 mg i.v. q8h) plus a third generation cephalosporin or fluoroquinolone, cefoxitin (2g i.v. q6h) or ceftizoxime (2g i.v. q8h) or metronidazole (500 mg i.v. or p.o. q6h) plus a fluoroquinolone (ciprofloxacin 750mg p.o. bid, ofloxacin 200 mg p.o. bid) (42). Therapy may also require coverage for Pseudomonas, depending upon culture results. Examples of effective oral therapy include clindamycin (300mg p.o qid) or metronidazole (250-500 mg p.o. q.i.d.) in combination with a fluoroquinolone, or monotherapy with amoxicillin/clavulanate (500mg p.o t.i.d. or 875mg p.o. b.i.d.), or trovafloxacin (200 mg p.o. q.d.) when Bacteroidesspp are involved (42,62,86).

Special Situations

Bacteroides spp can be the sole cause of infection in certain rare clinical situations requiring special consideration. Endocarditis, osteomyelitis, septic arthritis, and meningitis are four such examples. In addition, brain abscesses caused by B. fragilis alone, or in a mixed infection require special consideration. In all circumstances, antibiotic susceptibility of B. fragilis should be tested to determine the most appropriate antibiotic.

Endocarditis, Pericarditis, and Vascular Graft Infections

Bacteroides fragilis is one of the most common causes of anaerobic infective endocarditis, although overall quite rare. It is usually associated with large valvular vegetations and peripheral embolization (38). Thrombophlebitis and congestive heart failure are common complications as well (73). Mortality is reported to be as high as 46% (33).

Drug of Choice: Bactericidal antimicrobials are indicated in treatment of anaerobic infective endocarditis. In experimental endocarditis due to B. fragilis, metronidazole alone or in combination withclindamycin were superior to clindamycin or cefoxitin alone (45). The potent activity of carbapenems, and possibly beta-lactam/beta-lactamase inhibitors may also prove effective. In one report of one patient, i mipenem was administered, followed by a combination of ampicillin/clavulanate and metronidazole with a successful outcome (60). B. fragilis is an extremely rare cause of pericarditis, mycotic aneurysm, and vascular graft infection. All three clinical situations are associated with a high mortality, and no specific data on antimicrobial efficacy is available (65,99,101). This author suggests choosing the most potent antimicrobial that is bactericidal (such as metronidazole) in conjunction with appropriate surgical intervention.

Meningitis

B. fragilis is a rare cause of meningitis in the absence of brain abscess and is found most commonly in neonates as a result of congenital malformations, necrotizing enterocolitis, bowel perforation, or shunts. Anaerobes are so rare a cause of meningitis that clinical laboratories do not routinely culture cerebrospinal fluid for them. Thus, the only clue may be positive blood cultures (32).

Drug of Choice: Therapy of B. fragilis meningitis requires bactericidal therapy with metronidazole (500mg i.v. or p.o. q6h). Because of limited or unknown penetration of beta-lactam/beta-lactamase inhibitors and i mipenem, these agents cannot be recommended (84,120,121). Further, case reports of B. fragilis meningitis treated with chloramphenicol resulted in death or failure that required a change to metronidazole. Chloramphenicol failure is likely due to its bacteriostatic activity against B. fragilis. Some authors have measured bactericidal activity during therapy as a guide (32).M eropenem has been recently approved for treatment of meningitis. Efficacy against B. fragilis in the setting of meningitis is unknown.

Septic Arthritis

In the largest review of anaerobic septic arthritis, 20 out of 180 cases were caused by B. fragilis (34). Of the Bacteroides cases, most were from hematogenous spread, with the knee as the most common involved joint. One recent report described B. fragilis septic arthritis of the shoulder in a patient with Sickle cell disease (77). Prosthetic joints appear more susceptible to anaerobic infection, primarily with Peptostreptococci followed by B. fragilis (13,72).

Drug of Choice: Therapy for joint infections is directed at the B. fragilis group species and its susceptibility. Given the limited clinical experience, metronidazole (500mg i.v. or p.o q6h), carbapenems (i mipenem 500mg i.v. q6h; meropenem 1g i.v. q8h), and beta-lactam/beta-lactamase inhibitor combinations (ampicillin/sulbactam 3g i.v. q6h; ticarcillin/clavulanate 3.1g i.v.q 8h;piperacillin/tazobactam 3.75 g i.v. q6h) are suitable empiric choices until results can be obtained from the laboratory. Duration of therapy is usually 4-6 weeks. Prosthetic joints usually require removal, followed by 4-6 weeks of antibiotic therapy. One case of B. fragilis septic arthritis in a patient with sickle cell disease was treated successfully with four weeks of metronidazole and chloramphenicol, while a second responded to metronidazole and ticarcillin/clavulanic acid (64,77).

Brain Abscess

B. fragilis is a frequently isolated pathogen as a component of a mixed infection in brain abscesses. Previous studies have reported Bacteroides to be isolated as frequently as 20-60% from culture, associated with a number of other pathogens (particularly Streptococci) (12,16,110). This high percentage represents the inclusion of organisms formerly classified as Bacteroides.

Drug of Choice: Historically, treatment of brain abscess has included penicillin (20-24 million units/d) and chloramphenicol (1-1.5 g i.v. q6h) as empiric therapy (110). However, while penicillin remains a mainstay of therapy, chloramphenicol has been largely replaced with metronidazole. Metronidazole (500mg i.v. or p.o q6h) provides several advantages, including its bactericidal activity and excellent concentration in brain abscess pus (2,59,100,110). It is not affected by steroids, and the superior outcomes may reflect degradation of chloramphenicol in pus. No randomized trials have been conducted to compare these two agents. Patients suspected to have Staphylococci as a component of the abscess should be treated with nafcillin in place of penicillin, along with metronidazole. Alternatives for penicillin allergic patients include cefotaxime (3 g i.v. q8h) or ceftriazone (2-4 g / day). Duration of therapy, when surgery is performed, is dependent upon clinical and radiologic response. Generally, 4-6 weeks of therapy is required. When surgical drainage is not performed, longer duration may be attempted (6-8 weeks). Although some patients respond to medical management alone, most require surgical therapy. Aspiration and drainage by steriotaxic C.T. guidance may suffice in conjunction with antibiotics for stable patients without progressive neurologic signs (93).

Osteomyelitis

The site of infection is important when B. fragilis is suspected as part of a polymicrobial aerobic/anaerobic osteomyelitis. Trauma and direct spread from contiguous foci are the principle indicative factors for development of anaerobic osteomyelitis. B. fragilis is most frequently associated with osteomyelitis when peripheral vascular disease secondary to diabetes is present (4). Using careful culture techniques, 70% of diabetic associated osteomyelitis cases are polymicrobial, with B. fragilis as the second most common anaerobe isolated (128). One recent report describes vertebral osteomyelitis complicated by rupture of the aorta and a second vertebral osteomyelitis following a presumed bacteremia secondary to anal dilatation procedure (17,27).

Drug of Choice: Treatment requires identification of infecting organism from bone, and susceptibility testing of B. fragilis group isolates, when possible (13). Empiric therapy should be based upon knowledge of resistance of the Bacteroides spp, and coinfecting organisms. Agents that may be suitable as monotherapy, or in combination, include cefoxitin (2g i.v. q6h), clindamycin (600-900 mg i.v. q8h), metronidazole (500 mg q6-8h), beta-lactam/beta-lactamase inhibitors (ampicillin/sulbactam 3g i.v. q6h; ticarcillin/clavulanate 3.1g i.v. q 8h; piperacillin/tazobactam 3.75 g i.v. q6h) and carbapenems (i mipenem, 500 mg i.v. q6h; meropenem 1g i.v. q8h). Moxifloxacin may be an alternative therapeutic choice, although data are not yet available. Therapy should continue for 4-6 weeks, although prolonged oral therapy including amoxicillin/clavulanate (500mg p.o. t.i.d. or 875 mg p.o. b.i.d.) may be required when vascular insufficiency is not corrected.

Underlying Diseases

Patients with intra-abdominal infections due to leakage of bowel contents into the peritoneum (as a result of perforation, erosion from cancer, post-surgical complications, or complication of endoscopy) are highly susceptible to this organism (>85%). Again, these are almost exclusively mixed infections. Patients with diabetes are also susceptible to mixed infections that include B. fragilis. More serious infections in this setting include necrotizing fasciitis and synergistic anaerobic infections. Typically less life threatening, but possibly limb threatening infections include diabetic foot ulcers. In the latter case, the role of B. fragilis in the infectious process is not well defined (42).

ADJUNCTIVE THERAPY

The standard approach to infections involving B. fragilis typically includes both modalities of debridement and antibiotics. Hyperbaric oxygen treatment has been advocated as an adjunct for more serious infections involving B. fragilis group and other anaerobes. In animal experiments, hyperbaric oxygen has been reported to improve survival in mixed aerobic and anaerobic infections (118). However, despite its use, data supporting its role in the treatment of B. fragilis infections is not conclusive (55).

ENDPOINTS FOR MONITORING THERAPY

The high degree of association of abscess and B. fragilis infections typically requires follow-up radiographic imaging to monitor resolution of infectious process and the absence of new fluid collections (such as in intra-abdominal infections). Clinical follow-up would be the same as for most infections with resolution of fever and leukocytosis as sentinel indicators.

VACCINES

There are no vaccines currently available.

PREVENTION

Prophylaxis for gastrointestinal surgical procedures is important in prevention of mixed infections including B. frgailis. Current recommendations include oral neomycin and erythromycin base (1g each at 1pm, 2pm and 11pm the day before an 8am operation) as first line agents for elective procedures, with cefoxitin (1 g IV) or cefazolin (1 g IV) + metronidazole (0.5g IV) as second line agents (68). Other general measures for prevention of infections involving B. fragilis include good skin care in diabetic and bed ridden patients to prevent the development of pressure ulcerations.

CAVEATS AND COMMENTS

The role of B. fragilis has been well established for many types of infection. However, as a component of a polymicrobial infection, it is difficult to determine if the organism is always playing a dominant role. In the most serious infections involving B. fragilis, there appears to be good correlation between susceptibility test results and clinical response. However, it is more difficult to correlate specific susceptibility results with clinical outcomes in individual cases. The clinical outcome may be difficult to assess because of many influencing factors. Some patients may respond without antimicrobial therapy, or to inappropriate chemotherapy, with or without surgical intervention. The effects of surgical drainage, debridement, and other procedures are also important. Improper surgical management may result in poor clinical outcome even when antimicrobial therapy is adequate, and proper surgical management may result in a good outcome, despite inappropriate antimicrobial therapy (75). Recently, the correlation of poor microbiologic outcomes and a strong trend towards increased mortality was reported for patients with Bacteroides spp bacteremia resistant to antibiotics of which they were treated (78). This sentinel study now establishes the importance of antimicrobial susceptibility testing in the empiric and specific treatment of infections involving this group of organisms.

REFERENCES

1. Ackermann G, Schaumann R, Pless B, Claros MC, Goldstein EJ, Rodloff AC. Comparative activity of moxifloxacin in vitro against obligately anaerobic bacteria. Eur J Clin Microbiol Infect Dis 2000; 19(3):228-232. [PubMed]

2. Alderson D, Strong AJ, Ingham HR, Selkon JB. Fifteen year review of the mortality of brain abscess. Neurosurgery 1981; 8:1-6. [PubMed]

3. Anderson JD, Sykes RB. Characterization of a beta-lactamase obtained from a strain of Bacteroides fragilis resistant to beta-lactamase antibiotics. J Med Microbiol 1973; 6:201-206. [PubMed]

4. Bamberger DM, Daus GP, Gerding DN. Osteomyelitis in the feet of diabetic patients. Am J Med 1987; 83:653-660. [PubMed]

5. Bandoh K, Ueno K, Watanabe K, Kato N. Susceptibility patterns and resistance to imipenem in the Bacteroides fragilis group species in Japan: A 4-year study. Clin Infect Dis 1993; 16:S382- S386. [PubMed]

6. Baquero F, Reig M. Resistance of anaerobic bacteria to antimicrobial agents in Spain. Eur J Clin Microbiol Infect Dis 1992; 11:1016-1020. [PubMed]

7. Bennion RS, Thompson JE, Baron EJ, Finegold SM. Gangrenous and perforated appendicitis with peritonitis: treatment and bacteriology. Clin Ther 1990; 12:S31-S44. [PubMed]

8. Birolini D, Moraes MF, de Souza O. Aztreonam plus clindamycin vs tobramycin plus clindamycin for the treatment of intra-abdominal infections. Rev Infect Dis 1985; 7:S724- S728. [PubMed]

9. Brismar B, Malmborg AS, Tunevall G, Wretlind B, Bergman L, Mentzing LO, Nystrom PO, Kihlstrom E, Backstrand B, Skau T, Kasholm-Tengve B, Sjoberg L, Olsson-Liljequist B, Tally FP, Gatenbeck L, Eklund AE, Nord CE. Piperacillin-tazobactam versus imipenem-cilastatin for treatment of intra-abdominal infections. Antimicrob Agents Chemother 1992; 36(12):2766- 2773. [PubMed]

10. Britz ML, Wilkinson RG. Chloramphenicol acetyl-transferase of Bacteroides fragilis. Antimicrob Agents Chemother 1978; 14:105-111. [PubMed]

11. Brook I. Anaerobic bacterial bacteremia: 12-year experience in two military hospitals. J Infect Dis 1989; 160:1071-1075. [PubMed]

12. Brook I. Aerobic and anaerobic bacteriology of intra-cranial abscesses. Pediatr Neurol 1992; 8:210-214. [PubMed]

13. Brook I, Frazier EH. Anaerobic osteomyelitis and arthritis in a military hospital: A 10-year e experience. Am J Med 1993; 94:21-28. [PubMed]

14. Brook I, Myhal ML. Adherence of Bacteroides fragilis group species. Infect Immun 1991; 59(2):742-744. [PubMed]

15. Bryan LE, Kowand SK, Van Den Elzen HM. Mechanism of aminoglycoside antibiotic resistance in anaerobic bacteria, Clostridium perfringens and Bacteroides fragilis. Antimicrob Agents Chemother 1979; 15:7-13. [PubMed]

16. Chaudhry R, Dhawan B, Laxmi BV, Mehta VS. The microbial spectrum of brain abscess with special reference to anaerobic bacteria. Br J Neurosurg 1998; 12(2):127-130. [PubMed]

17. Chazan B, Strahilevitz J, Millgram MA, Kaufmann S, Raz R. Bacteroides fragilis vertebral osteomyelitis secondary to anal dilatation. Spine 2001; 26(16):E377-E378. [PubMed]

18. Chow AW, Guze LB. Bacteroidaceae bacteremia: Clinical experiences with 112 patients. Medicine 1974; 53:93-123. [PubMed]

19. Chow AW, Marshall JR, Guze LB. A double-blind comparison of clindamycin with penicillin plus chloramphenicol in treatment of septic abortion. J Infect Dis 1977; 135:S35-S39. [PubMed]

20. Citron DM. Rapid Identification of Anaerobes in the Clinical Laboratory. Anaerobe 1999; 5:109-113.

21. Citron DM, Ostavari A, Karlsson A, Goldstein EJ. Evaluation of the epsilometer (E-test) for susceptibility testing of anaerobic bacteria. Journal of Clinical Microbiology 1991; 29:2197- 2203. [PubMed]

22. Condon RE, Walker AP, Sirinek KR, White PW, Fabian TC, Nichols RL, Wilson SE. Meropenem versus tobramycin plus clindamycin for treatment of intra-abdominal infections: results of a prospective, randomized, double-blind clinical trial. Clin Infect Dis 1995; 21(3):544-550. [PubMed]

23. Cuchural GJ, Tally FP, Jacobus NV, Aldridge K, Cleary T, Finegold SM, Hill G, Iannini P, O'Keefe JP, Pierson CL, Crook DW, Russo TA, Hecht DW. Susceptibility of Bacteroides fragilisgroup in the United States: Analysis by site of isolation. Antimicrob Agents Chemother 1988; 32(5):717-722. [PubMed]

24. Cuchural GJ, Tally FP, Jacobus NV, Marsh PK, Mayhew JW. Cefoxitin inactivation by Bacteroides fragilis. Antimicrob Agents Chemother 1983; 3:936-940. [PubMed]

25. Darland G, Birnbaum J. Ceoxitin resistance to beta-lactamase: A major factor for susceptibility of Bacteroides fragilis to the antibiotic. Antimicrob Agents Chemother 1975; 9:725-734. [PubMed]

26. de Groot HG, Hustinx PA, Lampe AS, Oosterwijk WM. Comparison of imipenem/cilastatin with the combination of aztreonam and clindamycin in the treatment of intra-abdominal infections. J Antimicrob Chemother 1993; 32:491-500. [PubMed]

27. Doita M, Marui T, Kurosaka M, Yoshiya S, Tsuji Y, Okita Y, et al. Contained rupture of the aneurysm of common iliac artery associated with pyogenic vertebral spondylitis. Spine 2001; 26(13):E303-E307. [PubMed]

28. Dorsher CW, Rosenblatt JE, Wilson WR, Ilstrup DM. Anaerobic bacteremia: Decreasing rate over a 15-year period. Rev Infect Dis 1991; 13:633-636. [PubMed]

29. Drusano GL, Warren JW, Saah AJ, Caplan ES, Tenney JH, Hansen S, Granados J, Standiford HC, Miller EH. A prospective randomized controlled trial of cefoxitin versus clindamycin- aminoglycoside in mixed anaerobic-aerobic infections. Surg Gynecol Obstet 1982; 154:715-720. [PubMed]

30. Duerden BI. Virulence factors in anaerobes. Clin Infect Dis 1994; 18 Suppl 4:S253-S259. [PubMed]

31. el-Sefi TA, el-Awadi HM, Shehata MI, Al-Hindi MA. The place of antibiotics in the prevention of post-appendicectomy sepsis: a prospective study of 400 cases. Int Surg 1986; 71:18-21. [PubMed]

32. Feder HM. Bacteroides fragilis meningitis. Rev Infect Dis 1987; 9(4):783-786. [PubMed]

33. Felner JM, Dowell VR, Jr. Anaerobic bacterial endocarditis. N Engl J Med 1970; 283:1188-1192. [PubMed]

34. Finegold SM. Bone and Joint Infections. In: Finegold SM, editor. Anaerobic Bacteria in Human Disease. New York: Academic Press, 1977: 433-453.

35. Finegold SM. Therapy and Prognosis in Anaerobic Infections. In: Finegold SM, editor. Anaerobic Bacteria in Human Disease. New York: Academic Press, 1977: 534-562.

36. Finegold SM. General Aspects of Anaerobic Infection. In: Finegold SM, George WL, editors. Anaerobic Infections in Humans. San Diego: Academic Press, 1989: 137-149.

37. Finegold SM. Therapy of Anaerobic Infections. In: Finegold SM, George WL, editors. Anaerobic Infections in Humans. Orlando, FL: Academic Press, 1989: 793-818.

38. Finegold SM, George WL, Mulligan ME. Anaerobic infections. Part I. DM. Dis Month 1985; 31:1-77. [PubMed]

39. Finegold SM, Harada NE, Miller LG. Lincomycin: Activity against anaerobes and effect on normal human fecal flora. Antimicrob Agents Chemother 1966; 1965:659-667. [PubMed]

40. Fink MP, Helsmoortel CM, Arous EJ, Doern GV, Moriarty KP, Fairchild PG, Townsend PL. Comparison of the safety and efficacy of parenteral ticarcillin/clavulanate and clindamycin/gentamicin in serious intra-abdominal infections. J Antimicrob Chemother 1989; 24:147-156. [PubMed]

41. Galpin JE, Chow AW, Bayer AS, Guze LB. Sepsis associated with decubitus ulcers. Am J Med 1976; 61:346-350. [PubMed]

42. Gerding DN. Foot infections in diabetic patients: The role of anaerobes. Clin Infect Dis 1995; 20:S283-S288. [PubMed]

43. Gilstrap LC, Maier RC, Gibbs RS, Connor KD, St.Clair PJ. Piperacillin versus clindamycin plus gentamicin for pelvic infections. Obstet Gynecol 1984; 64:762-765. [PubMed]

44. Giuliano A, Lewis F Jr., Hadley K, Blaisdell FW. Bacteriology of necrotizing fasciitis. Am J Surg 1977; 134:52-56. [PubMed]

45. Goldman PL, Durack DT, Petersdorf RG. Effect of antibiotics on the prevention of experimental Bacteriodes fragilis endocarditis. Antimicrob Agents Chemother 1978; 14:755-760. [PubMed]

46. Goldstein EJ, Citron DM, Merriam CV, Warren Y, Tyrrell K. Comparative in vitro activities of GAR-936 against aerobic and anaerobic animal and human bite wound pathogens. Antimicrob Agents Chemother 2000; 44(10):2747-2751. [PubMed]

47. Goldstein EJ. Patterns of susceptibility to flouoroquinolones among anaerobic bacterial isolates in the United States. Clin. Infect. Dis. 1993; 16:S377-381. [PubMed]

48. Goldstein EJ, Citron DM, Goldman RJ, Claros MC, Hunt Gerardo S. United States national hospital survey of anaerobic culture and susceptibility methods, II. Anaerobe 1995; 1:309-314.

49. Gorbach SL. Antibiotic treatment of anaerobic infections. Clin Infect Dis 1994; 18:S305-S310. [PubMed]

50. Hecht DW, Lederer L. Effect of choice of medium on the results of in vitro susceptibility testing of eight antibiotics against the Bacteroides fragilis group. Clin.Infect.Dis. 1995; 20:S346-349. [PubMed]

51. Hecht DW, Malamy MH, Tally FP. Mechanisms of resistance and resistance transfer in anaerobic bacteria. In: Finegold SM, George WL, editors. Anaerobic Infections in Humans. Orlando, FL: Academic Press, 1989.

52. Hecht DW, Osmolski JR, O'Keefe JP. Variation in the susceptibility of Bacteroides fragilis group isolates from six Chicago Hospitals. Clin Infect Dis 1993; 16:S357-360. [PubMed]

53. Hecht DW, Wexler HM. In vitro susceptibility of anaerobes to quinolones in the United States. Clin Infect Dis 1997; 23:S2-S8. [PubMed]

54. Hemsell DL, Martens MG, Faro S, Gall S, McGregor JA. A multicenter study comparing intravenous meropenem with clindamycin plus gentamicin for the treatment of acute gynecologic and obstetric pelvic infections in hospitalized women. Clin Infect Dis 1997; 24:S222-S230. [PubMed]

55. Hirn M, Niinikoski J, Lehtonen O-P. Effect of hyperbaric oxygen and surgery on experimental multimicrobial gas gangrene. Eur Surg Res 1993; 25:265-269. [PubMed]

56. Hoellman DB, Kelly LM, Jacobs MR, Appelbaum PC. Comparative antianaerobic activity of BMS 284756. Antimicrob Agents Chemother 2001; 45(2):589-592. [PubMed]

57. Hofstad T. Pathogenicity of anaerobic gram-negative rods: possible mechanisms. Rev Infect Dis 1984; 6(2):189-199. [PubMed]

58. Hurlbut S, Cuchural GJ, Tally FP. Imipenem resistance in Bacteroides distasonis mediated by a novel beta-lactamase. Antimicrob Agents Chemother 1990; 34(1):117-120. [PubMed]

59. Ingham HR, Selkon JB, Roxby CM. Bacteriological study of otogenic cerebral abscesses:Chemotherapeutic role of metronidazole. Br Med J 1977; 2:991-993.

60. Jackson RT, Dopp AC. Bacteroides fragilis endocarditis. South Med J 1988; 81:781-782. [PubMed]

61. Jacobs MR, Spangler SK, Appelbaum PC. Beta-Lactamase production, beta-lactam sensitivity and resistance to synergy with clavulanate of 737 Bacteroides fragilis group organisms from thirty-three US centers. J Antimicrob Chemother 1990; 26:361-370. [PubMed]

62. Karchmer AW, Gibbons GW. Foot infections in diabetes: Evaluation and management. In: Remington JS, Swartz MN, editors. Curr CliniTop in Infect Dis. Boston: Blackwell Scientific, 1994: 7-10. [PubMed]

63. Kempf P, Bauernfeind A, Muller A, Blum J. Meropenem monotherapy versus cefotaxime plus metronidazole combination treatment for serious intra-abdominal infections. Infection 1996; 24(6):473-479. [PubMed]

64. Konstantopoulos K, Avlami A, Demarongona K, Sideris G, Rekoumi L, Stefanou J, Voskaridou E, Loukopoulos D. Bacteroides fragilis arthritis in a sickle cell-thalassaemia patient. Scand J Infect Dis 1994; 26:495-497. [PubMed]

65. Lam S, Greenberg R, Bank S. An unusual presentation of colon cancer: purulent pericarditis and cardiac tamponade due to Bacteroides fragilis. Am J Gastroenterol 1995; 90(9):1518-1520. [PubMed]

66. Larsen JW, Gabel-Hughes K, Kreter B. Efficacy and tolerability of imipenem-cilastatin versus clindamycin+gentamicin for serious pelvic infections. Clin Ther 1992; 14:90-96. [PubMed]

67. Martens MG, Faro S, Hammill HA, Smith D, Riddle G, Maccato M. Ampicillin/sulbactam versus clindamycin in the treatment of postpartum endomyometritis. South Med J 1990; 83(4):408-413. [PubMed]

68. Medical Letter and Loyola Surgical Prophylactic Orders. Antimicrobials for Surgical Prophylaxis. 2001; 43.

69. Meller JL, Reyes HM, Loeff DS, Federer L, Hall JR. One-drug versus two-drug antibiotic therapy in pediatric perforated appendicitis: A prospective randomized study. Surgery 1991; 110:764-768. [PubMed]

70. Morbidity and Mortality Weekly Report. 1998 Guidelines for Treatment of Sexually Transmitted Diseases. 1998; 47:RR-1. Epidemiology Program Office, Centers for Disease Control, Public Health Service, U.S. Department of Health and Human Services , Atlanta, GA. [[PubMed]

71. Najem AZ, Kaminski ZC, Spillert CR, Lazaro EJ. Comparative study of parenteral piperacillin and cefoxitin in the treatment of surgical infections of the abdomen. Surgery 1983; 157:423- 425. [PubMed]

72. Nakata MM, Lewis RP. Anaerobic bacteria in bone and joint infections. Rev Infect Dis 1984; 6:165-170. [PubMed]

73. Nastro LJ, Finegold SM. Endocarditis due to anaerobic gram-negative bacilli. Am J Med 1973; 54:482-496. [PubMed]

74. Nathens AB, Rotstein OD. Antimicrobial therapy for intra-abdominal infection. Am J Surg 1996; 172:S1-S6. [PubMed]

75. National Committee for Clinical Laboratory Standards. Methods for Antimicrobial Susceptibility Testing of Anaerobic Bacteria, 4th ed. 1997; Approved standard M11-A4. Villanova, Pa.

76. National Committee for Clinical Laboratory Standards. Methods for Antimicrobial Susceptibility Testing of Anaerobic Bacteria, 5th Edition. 2001; Approved Standard M11-A4. Villanova, Pa.

77. Nehra A, Mehta RN, Encarnacion C. Bacteroides fragilis arthritis in a patient with sickle cell disease. Am J Med 2001; 111(2):161. [PubMed]

78. Nguyen MH, Yu VL, Morris AJ, McDermott L, Wagener MW, Harrell L, Snydman DR. Antimicrobial resistance and clinical outcome of Bacteroides bacteremia: findings of a multicenter prospective observational trial. Clin Infect Dis 2000 Jun; 30: 870-6. [PubMed]

79. Nichols RL, Smith JW, Klein DB, Trunkey DD, Cooper RH, Adinolfi MF, Mills J. Risk of infection after penetrating abdominal trauma. N Engl J Med 1984; 311:1065-1070. [PubMed]

80. O'Keefe JP, Venezio FR, DiVincenzo CA, Shatzer KL. Activity of newer beta-lactam agents against clinical isolates of Bacteroides fragilis and other Bacteroides species. Antimicrob Agents Chemother 1987; 31:2002-2004. [PubMed]

81. Olsson B, Nord CE, Wadstrom T. Formation of beta-lactamase in Bacteroides fragilis: Cell- bound and extracellular activity. Antimicrob Agents Chemother 1976; 9:727-735. [PubMed]

82. Onderdonk AB, Kasper DL, Cisneros RL, Bartlett JG. The capsular polysaccharide of Bacteroides fragilis as a virulence factor: comparison of the pathogenic potential of encapsulated and unencapsulated strains. J Infect Dis 1977; 136(1):82-89. [PubMed]

83. Onderdonk AB, Weinstein WM, Sullivan NM, Bartlett JG, Gorbach SL. Experimental intra- abdominal abscesses in rats:quantitative bacteriology of infected animals. Infect Immun 1974; 10:1256-1259. [PubMed]

84. Patey O, Breuil J, Fisch A, Burnat C, Vincent-Ballereau F, Dublanchet A. Bacteroides fragilis meningitis: report of two cases. Rev Infect Dis 1990; 12(2):364-365. [PubMed]

85. Peraino VA, Cross SA, Goldstein EJ. Incidence and clinical significance of anaerobic bacteremia in a community hospital. Clin Infect Dis 1993; 16:S288-S291. [PubMed]

86. Peterson LR, Lissack LM, Canter K, Fasching CE, Clabots C, Gerding DN. Therapy of lower extremity infections with ciprofloxacin in patients with diabetes mellitus, peripheral vascular disease, or both. Am J Med 1989; 86:801-808. [PubMed]

87. Pinkus G, Veo G, Braude AI. Bacteroides penicillinase. J Bacteriol 1968; 96:1437-1438. [PubMed]

88. Pitkin D, Sheikh W, Wilson S, Hemsell D, Nichols R, Nadler, Dowzicky M. Comparison of the activity of meropenem with that of other agents in the treatment of intra-abdominal, obstetric/gynecologic, and skin and soft tissue infections. Clin Infect Dis 1995; 20:s372-s375. [PubMed]

89. Polk HC, Fink MP, Laverdiere M, Wilson SE, Garber GE, Barie P, Hebert JC, Cheadle WG. Prospective randomized study of piperacillin/tazobactam therapy of surgically treated intra- abdominal infection. Am Surg 1993; 59:598-605. [PubMed]

90. Rasmussen J, Odelson D, Macrina F. Complete nucleotide and transcription of ermF, a macrolide-lincosamide-sterptogramin B resistance determinant from Bacteroides fragilis. J Bacteriol 1986; 168:523-533. [PubMed]

91. Reysset G, Haggoud A, Su WJ, Sebald M. Genetic and molecular analysis of pIP417 and pIP419: Bacteroides plasmids encoding 5-nitroimidazole resistance. Plasmid 1992; 27:181- 190. [PubMed]

92. Reyssett G, Su W-J, Sebald M. Genetics of 5-nitroimidazole resistance in Bacteroides, p494- 504. In: Sebald M, editor. Genetics and Molecular Biology of Anaerobic Bacteria. New York: Springer-Verlag, 1993.

93. Rosenblatt J. Antimicrobic susceptibility of anaerobic bacteria. In: Finegold SM, George WL, editors. San Diego, CA: Academic Press, 1989: 715-727.

94. Rosenblatt J, Gustafson DR. Evaluation of the E-test for susceptibility testing of anaerobic bacteria. Diagn Microbiol Infect Dis 1995; 22:279-284. [PubMed]

95. Rotimi V, Duerden B, Hafiz S. Transferable plasmid-mediated antibiotic resistance in Bacteroides. J Med Microbiol 1981; 14:359-370. [PubMed]

96. Salaki JS, Black R, Tally FP, Kislak JW. Bacteroides fragilis resistant to clindamycin. Am J Med 1976; 60:426-428. [PubMed]

97. Salyers AA, Shoemaker NB, Guthrie EP. Recent advances in Bacteroides genetics. CRC Crit Rev Microbiol 1987; 14:49-71. [PubMed]

98. Scandinavian Study Group. Imipenem-cilastatin versus gentamicin-clindamycin for treatment of serious bacterial infections. Lancet 1984; 1:868-871. [PubMed]

99. Sioson P, Brown RB. Successful medical treatment of an infected prosthetic aortic graft. West J Med 1993; 158:301-303. [PubMed]

100. Sjolin J, Eriksson N, Arneborn P, Cars O. Penetration of ceftizoxime and desacetylcefotaxime into brain abscesses in humans. Antimicrob Agents Chemother 1991; 35:2606-2610. [PubMed]

101. Skiest DJ, Steiner D, Werner M, Garner JG. Anaerobic pericarditis: Case report and review. Clin Infect Dis 1994; 19:435-440. [PubMed]

102. Snydman DR, Cuchural GJ, McDermott L, Gill M. Correlation of various in vitro testing methods with clinical outcomes in patients with Bacteroides fragilis group infections treated with cefoxitin: a retrospective analysis. Antimicrob Agents Chemother 1992; 36(3):540-544. [PubMed]

103. Snydman DR, Jacobus NV, McDermott LA, Supran S, Cuchural GJ, Finegold S, Harrell L, Hecht DW, Iannini P, Jenkins S, Pierson C, Rihs J, Gorbach SL. Multicenter study of in vitrosusceptibility of the Bacteroides fragilis group, 1995 to 1996, with comparison of resistance trends from 1990 to 1996. Antimicrob Agents Chemother 1999; 43(10):2417-2422. [PubMed]

104. Snydman DR, McDermott L, Cuchural GJ, Hecht DW, Iannini P, Harrell LJ, Jenkins SG, O’Keefe JP, Pierson CL, Rihs JD, Yu VL, Finegold SM, Gorbach SL. Analysis of trends in antimicrobial resistance patterns among clinical isolates of Bacteroides fragilis group species from 1990 to 1994. Clin.Infect.Dis. 1996; 23: S54-S65. [PubMed]

105. Solomkin JS, Dellinger EP, Christou NV, Busuttil RW. Results of a multicenter trial comparing imipenem cilastatin to tobramycin-clindamycin for intra-abdominal infections. Ann Surg 1990; 212:581-591. [PubMed]

106. Spangler SK, Jacobs MR, Appelbaum PC. Activity of WY-49605 compared with those of amoxicillin, amoxicillin-clavulanate, imipenem, ciprofloxacin, cefaclor, cefpodoxime, cefuroxime, clindamycin, and metronidazole against 384 anaerobic bacteria. Antimicrob Agents Chemother 1994; 38:2599-2604. [PubMed]

107. Stone HH, Martin JD, Jr. Synergistic necrotizing cellulitis. Ann Surg 1972; 175:702. [PubMed]

108. Study Group of Intraabdominal Infections. A randomized controlled trial of ampicillin plus sulbactam vs gentamicin plus clindamycin in the treatment of intra-abdominal infections. Rev Infect Dis 1986;8:S583-8. [PubMed]

109. Sutter VL. In vitro susceptibility of anaerobes: Comparison of clindamycin and other antimicrobial agents. J Infect Dis 1977; 135:S7-S12. [PubMed]

110. Swartz MN. Central nervous system infections. In: Finegold SM, George WL, editors. Anaerobic Infections in Humans. San Diego: Academic Press, 1989: 155-212.

111. Sweet RL. Role of cephamycins in obstetrics and gynecology. J Reprod Med 1990; 35:S1064- S1069. [PubMed]

112. Sweet RL, Robbie MO, Ohm-Smith M, Hadley WK. Comparative study of piperacillin versus cefoxitin in the treatment of obstetrics and gynecologic infections. Am J Obstet Gynecol 1983; 145:342-349. [PubMed]

113. Sweet RL, Roy S, Faro S, O'Brien WF, Sanfilippo JS, Seidlin M. The Piperacillin-Tazobactam Study Group. Piperacillin and tazobactam versus clindamycin and gentamicin in the treatment of hospitalized women with pelvic inflammation. Obstet Gynecol 1994; 83:280-286. [PubMed]

114. Tally FP, Goldin BR, Jacobus NV, Gorbach SL. Superoxide dismutase in anaerobic bacteria of clinical significance. Infect Immun 1977; 16:20-25. [PubMed]

115. Tally FP, McGowan K, Kellum JM, Gorbach SL, O'Donnell TF. A randomized comparison of cefoxitin with or without amikacin and clindamycin plus amikacin in surgical sepsis. Ann Surg 1981; 193:318-323. [PubMed]

116. Tally FP, O'Keefe JP, Sullivan NM, Gorbach SL. Inactivation of cephalosporins by Bacteroides. Antimicrob Agents Chemother 1979; 16:565-571. [PubMed]

117. Thadepalli H, Gorbach SL, Broido PW, Norsen J, Nyhus L. Abdominal trauma, anaerobes, and antibiotics. Surg Gynecol Obstet 1973; 137:270-276. [PubMed]

118. Thom SR, Lauermann MW, Hart GB. Intermittent hyperbaric oxygen therapy for reduction of mortality in experimental polymicrobial sepsis. J Infect Dis 1986; 154(3):504-510. [PubMed]

119. Tzianabos AO, Kasper DL, Onderdonk AB. Structure and function of Bacteroides fragilis capsular polysaccharides: relationship to induction and prevention of abscesses. Clin Infect Dis 1995; 20 Suppl 2:S132-S140 [PubMed].

120. Unhanand M, Mustafa MM, McCracken GH, Nelson JD. Gram-negative enteric bacillary meningitis: a twenty-one year experience. J Pediatr 1993; 122:15-21. [PubMed]

121. Warner JF, Perkins RL, Cordero L. Metronidazole therapy of anaerobic bacteremia, meningitis and brain abscess. Arch Intern Med 1979; 139:167-169. [PubMed]

122. Weinstein MP, Reller LB, Murphy JR, Lichtenstein KA. The clinical significance of positive blood cultures: A comprehensive analysis of 500 episodes of bacteremia and fungemia in adults. I. Laboratory and epidemiologic observations. Rev Infect Dis 1983; 5:35-53. [PubMed]

123. Wexler HM, Finegold SM. Media- and method-dependent variation in MIC values of ceftizoxime for clinical isolates of the Bacteroides fragilis group. Clin Therapeutics 1990; 12 (suppl C):13-24. [PubMed]

124. Wexler HM, Molitoris E, Finegold SM. Effect of beta-lactamase inihibitors on the activities of various beta-lactam agents against anaerobic bacteria. Antimicrob Agents Chemother 1991; 35:1219-1224. [PubMed]

125. Wexler HM, Molitoris E, Finegold SM. In vitro activities of two new glycylcyclines, N,N- dimethylglycylamido derivatives of minocycline and 6-demethyl-6-deoxytetracycline, against 339 strains of anaerobic bacteria. Antimicrob Agents Chemother 1994; 38:2513-2515. [PubMed]

126. Wexler HM, Molitoris E, Reeves D, Finegold SM. In vitro activity of clinafloxacin (CI-960) and PD 131628-2 against anaerobic bacteria. J Antimicrob Chemother 1994; 34:579-584. [PubMed]

127. Wexler HM, Molitoris E, Reeves D, Finegold SM. In Vitro activity of DU-6859a against anaerobic bacteria. Antimicrob.Agents Chemother 1994; 38:2504-2509. [PubMed]

128. Wheat LJ, Allen SD, Henry M, Kernek CB, Siders JA, Kuebler T, Fineberg N, Norton J. Diabetic foot infections:Bacteriologic analysis. Arch Intern Med 1986; 146:1935-1940. [PubMed]

129. Williams RR, Hotchkin D. Aztreonam plus clindamycin versus tobramycin plus clindamycin in the treatment of intra-abdominal infections. Rev Infect Dis 1991; 13:S629-S633. [PubMed]

Tables

Table 1. Comparative activity of antimicrobial agents against B. fragilis group species*

		B. fragilis			B. distasonis			B. thetaiotamicron
Antibiotic	NCCLS breakpointsH (S, I)	MIC50	MIC90	Range	MIC50	MIC90	Range	MIC50	MIC90	Range
Ticarcillin	(32, 64)	32	128	NA	128	128	NA	64	128	NA
Piperacillin	(32, 64)	16	256	NA	64	128	NA	32	128	NA
Cefoxitin	(16, 32)	8	32	2 - 256	16	32	4 - 256	8	16	1 - 256
Cefotetan	(16, 32)	8	64	2 - 256	128	256	4 - 256	64	256	1 - 256
Ceftizoxime	(16, 32)	64	128	4 - 256	32	256	1 - 256	64	256	1 - 256
Imipenem	(4, 8)	0.5	0.5	0.015 - 4	0.06	0.25	0.015-0.25	0.5	1	0.03 - 4
Meropenem	(4, 8)	0.5	0.5	0.015 - 4	0.06	0.25	0.015-0.25	0.5	1	0.03 - 4
Amox /Clavulanate	(4, 8)	1	4	0.25 - 4	2	32	0.125-32	1	2	0.125 - 4
Ampicillin/Sulbactam	(8, 16)	2	8	NA	8	32	NA	4	16	NA
Piperacillin/Tazobactam	(32, 64)	0.5	4	NA	8	32	NA	8	16	NA
Ticarcillin/Clavulanate	(32, 64)	1	8	NA	16	64	NA	4	16	NA
Clindamycin	(2, 4)	0.5	128	0.06-128	2	128	0.06-128	4	128	0.06-128
Chloram - phenicol&	(8, 16)	1	8	NA	1	8	NA	1	8	NA
Metronidazole	(8, 16)	0.5	0.5	0.25 - 1	0.5	0.5	0.25 - 1	0.5	1	0.25 - 1
Trovafloxacin	(2, 4)	0.25	2	0.25 - 2	0.5	1	0.125 - 1	0.5	2	0.25 - 4

	B. ovatus			B. vulgatus
Antibiotic	MIC50	MIC90	Range	MIC50	MIC90	Range
Ticarcillin	64	128	NA	32	128	NA
Piperacillin	32	128	NA	64	128	NA
Cefoxitin	8	16	2 - 64	4	16	2 - 64
Cefotetan	64	128	2 - 128	4	128	2 - 256
Ceftizoxime	32	256	4 - 256	16	128	1 - 256
Imipenem	0.125	0.25	0.015-0.5	0.25	0.5	0.015-0.5
Meropenem	0.125	0.25	0.015-0.5	0.25	0.5	0.015-0.5
Amox /Clavulanate	2	8	0.25 - 16	2	8	0.25 - 8
Ampicillin/Sulbactam	4	16	NA	4	16	NA
Piperacillin/Tazobactam	8	16	NA	8	16	NA
Ticarcillin/Clavulanate	4	16	NA	2	8	NA
Clindamycin	2	128	0.06-128	0.06	128	0.03-128
Chloram - phenicol&	1	8	NA	1	8	NA
Metronidazole	0.5	1	0.25 - 1	0.5	0.5	0.25 - 1
Trovafloxacin	1	2	0.125 - 2	0.25	4	0.125 - 4

* Data based upon various published studies; H, S, I; Susceptible and intermediate breakpoints in μg/ml; I, Not available; &, Drug tested only at screening dilutions

Table 2. Comparative antimicrobial trials for treatment of intra-abdominal and pelvic infections.

Intra-abdominal
Cefoxitin + amino glycoside vs. clindamycin + amino glycoside	(29,79,115)
Ceftizoxime vs. cefoxitin	(7)
Piperacillin vs. cefoxitin	(71)
Ampicillin/sulbactam vs. clindamycin + amino glycoside	(108)
Imipenem vs. clindamycin + amino glycoside	(98,105)
Aztreonam + clindamycin vs. clindamycin + amino glycoside	(8,129)
Aztreonam + clindamycin vs. imipenem	(26)
Ticarcillin/clavulanate vs. clindamycin amino glycoside	(40)
Piperacillin/tazobactam vs. imipenem	(9)
Piperacillin/tazobactam vs. clindamycin + amino glycoside	(89)
Meropenem vs. clindamycin + tobramycin	(22)
Meropenem vs. cefotaxime + metronidazole	(63)
Pelvic
Piperacillin vs. cefoxitin	(112)
Piperacillin vs. clindamycin + gentamicin	(43)
Piperacillint/tazobactam vs. clindamycin + gentamicin	(113)
Ampicillin/sulbactam vs. clindamycin	(67)
Imipenem vs. clindamycin gentamicin	(66)
Meropenem vs. clindamycin gentamicin	(54)