Haemin deprivation renders Bacteroides fragilis hypersusceptible to metronidazole and cancels high-level metronidazole resistance.

BACKGROUND: Infections with Bacteroides fragilis are routinely treated with metronidazole, a 5-nitroimidazole antibiotic that is active against most anaerobic microorganisms. Metronidazole has remained a reliable treatment option, but resistance does occur, including in B. fragilis. OBJECTIVES: In this study we tested whether haemin, a growth supplement for B. fragilis in vivo and in vitro, had an influence on the susceptibility of resistant B. fragilis strains to metronidazole. We further tested whether haemin-deprived B. fragilis would be more susceptible to oxygen and oxidative stress. Metronidazole has been described to cause oxidative stress, which we argued would be exacerbated in haemin-deprived B. fragilis because the bacteria harness haemin, and the iron released from it, in antioxidant enzymes such as catalase and superoxide dismutase. METHODS: Haemin was omitted from growth media and the effect on metronidazole susceptibility was monitored in susceptible and resistant B. fragilis strains. Further, haemin-deprived B. fragilis were tested for resistance to aeration and hydrogen peroxide and the capacity for the removal of oxygen. RESULTS: Omission of haemin from the growth medium rendered metronidazole-resistant B. fragilis strains, including an MDR isolate from the UK, highly susceptible to metronidazole. Haemin deprivation further rendered B. fragilis highly susceptible to oxygen, which was further exacerbated in resistant strains. B. fragilis was incapable of scavenging oxygen when haemin was omitted. CONCLUSIONS: We propose that haemin deprivation overrules resistance mechanisms by rendering B. fragilis hypersusceptible to metronidazole due to a compromised antioxidant defence. Monitoring of haemin concentrations is imperative when conducting metronidazole susceptibility testing in B. fragilis.

Frequency and associated factors for carbapenem-non-susceptible Bacteroides fragilis group bacteria colonization in hospitalized patients: Case control study in a university hospital in Turkey.

PURPUSE: The carbapenem-resistant Bacteroides fragilis group (CR-BFG) bacteria have been reported in several countries recently with increasing global attention. The high incidence of CR-BFG isolated from our hospitalized patients has become an important problem. Therefore, we aimed to determine the frequency and associated factors for intestinal colonization by carbapenem-non-susceptible BFG (CNS-BFG) among adult patients hospitalized at intensive care units, neurosurgery and internal medicine wards in our hospital. METHODS: Rectal swabs (n = 1200), collected from 766 patients between February 2014 and March 2015, were inoculated onto kanamycin-vancomycin-leaked blood agar containing 0.125 mg/L meropenem. The isolates were identified by MALDI-TOF MS. Susceptibility testing was performed by agar dilution method. The carbapenemase gene (cfiA) was detected by PCR. Logistic regression analysis was used to evaluate the associated factors for intestinal colonization by CNS-BFG. RESULTS: A total 180 non-duplicate BFG isolates were obtained from 164 patients. Ten different species, including Parabacteroides distasonis (n = 46, 25.6%), and Bacteroides fragilis (n = 30; 16.6%), were identified. Twenty-five percent of the isolates were non-susceptible to meropenem (MIC >2 mg/L). The highest prevalence of meropenem resistant strains (MIC >8 mg/L) was detected among B. fragilis (n = 12), followed by Parabacteroides spp. (n = 4). All but one B. fragilis strains were cfiA gene positive. Hospital admission, increasing Charlson score, use of antibiotics; including carbapenems in past three months, colonization with other accompanying carbapenem-resistant Gram negative bacteria (Enterobacteriaceae, Acinetobacter baumannii and Pseudomonas aeruginosa), and having undergone surgical operations were significantly associated with RCS- BFG colonization. CONCLUSIONS: The high carriage rate of CNS-BFG in hospitalized patients may lead to worse clinical outcomes, such as serious infections and mortality, and deserves attention.

Antibacterial and antibiofilm effects of α-humulene against Bacteroides fragilis.

The rapid increase in antibiotic resistance has prompted the discovery of drugs that reduce antibiotic resistance or new drugs that are an alternative to antibiotics. Plant extracts have health benefits and may also exhibit antibacterial and antibiofilm activities against pathogens. This study determined the antibacterial and antibiofilm effects of α-humulene extracted from plants against enterotoxigenic Bacteroides fragilis, which causes inflammatory bowel disease. The minimum inhibitory concentration and biofilm inhibitory concentration of α-humulene for B. fragilis were 2 µg/mL, and the biofilm eradication concentration was in the range of 8-32 µg/mL. The XTT reduction assay confirmed that the cellular metabolic activity in biofilm rarely occurred at the concentration of 8-16 µg/mL. In addition, biofilm inhibition by α-humulene was also detected via confocal laser scanning microcopy. Quantitative real-time polymerase chain reaction (qPCR) was also used to investigate the effect of α-humulene on the expression of resistance-nodulation-cell division type multidrug efflux pump genes (bmeB1 and bmeB3). According to the results of qPCR, α-humulene significantly reduced the expression of bmeB1 and bmeB3 genes. This study demonstrates the potential therapeutic application of α-humulene for inhibiting the growth of B. fragilis cells and biofilms, and it expands the knowledge about biofilm medicine.

Antimicrobial resistance in the Bacteroides fragilis group in faecal microbiota from healthy Danish children.

The Bacteroides fragilis group constitute a significant portion of the human gut microbiota and comprise a major proportion of anaerobic bacteria isolated in human infections. We established a baseline of antimicrobial susceptibility rates in the B. fragilis group in the intestinal tract of relatively antibiotic-naive healthy Danish children. From 174 faecal samples collected from children attending day care, 359 non-duplicate isolates were screened for antimicrobial susceptibility. Of these, 0.0%, 1.9%, 5.0% and 21.2% of isolates were intermediate-susceptible or resistant to metronidazole, meropenem, piperacillin/tazobactam and clindamycin, respectively. Eighteen additional studies reporting susceptibility rates in the B. fragilis group bacteria were identified by conducting a literature search. Heterogeneity among results from studies of B. fragilis group antimicrobial susceptibility rates in faecal microbiota exists.

A Novel Selective Medium for Isolation of Bacteroides fragilis from Clinical Specimens.

A novel Bacteroides fragilis selective (BFS) medium, consisting of a brain heart infusion agar base supplemented with yeast extract, cysteine hydrochloride, bile salts, vitamin K, hemin, glucose, esculin, ferric ammonium citrate, bromothymol blue, gentamicin, kanamycin, and novobiocin, was evaluated. When BFS agar was tested with a collection of 303 bacteria of different genera, it allowed the growth of B. fragilis as large yellow colonies, with blackening of the medium after 48 h of anaerobic incubation, while the growth of most other anaerobes, facultative anaerobes, and aerobes was inhibited. In a prospective comparison of BFS agar with a routinely used medium (neomycin blood agar) in 1,209 clinical specimens, 60 B. fragilis bacteria were detected on BFS agar while 46 were detected on the routine agar (McNemar's test, P = 0.008). In conclusion, this novel medium may be added to improve the recovery of B. fragilis in clinical specimens and to facilitate surveillance of antimicrobial-resistant strains.

Effect of Fidaxomicin versus Vancomycin on Susceptibility to Intestinal Colonization with Vancomycin-Resistant Enterococci and Klebsiella pneumoniae in Mice.

The use of oral vancomycin or metronidazole for treatment of Clostridium difficile infection (CDI) may promote colonization by health care-associated pathogens due to disruption of the intestinal microbiota. Because the macrocyclic antibiotic fidaxomicin causes less alteration of the intestinal microbiota than vancomycin, we hypothesized that it would not lead to a loss of colonization resistance to vancomycin-resistant enterococci (VRE) and extended-spectrum-ß-lactamase-producing Klebsiella pneumoniae (ESBL-Kp). Mice (8 per group) received orogastric saline, vancomycin, or fidaxomicin daily for 5 days at doses resulting in stool concentrations in mice similar to those measured in humans. The mice were challenged with 10(5) CFU of orogastric VRE or ESBL-Kp on day 2 of treatment and concentrations of the pathogens in stool were monitored. The impact of drug exposure on the microbiome was measured by cultures, real-time PCR for selected anaerobic bacteria, and deep sequencing. In comparison to saline controls, oral vancomycin promoted establishment of high-density colonization by VRE and ESBL-Kp in stool (8 to 10 log10 CFU/g; P < 0.001), whereas fidaxomicin did not (<4 log10 CFU; P > 0.5). Vancomycin treatment resulted in significant reductions in enterococci, Bacteroides spp., and Clostridium leptum, whereas the population of aerobic and facultative Gram-negative bacilli increased; deep-sequencing analysis demonstrated suppression of Firmicutes and expansion of Proteobacteria during vancomycin treatment. Fidaxomicin did not cause significant alteration of the microbiota. In summary, in contrast to vancomycin, fidaxomicin treatment caused minimal disruption of the intestinal microbiota and did not render the microbiota susceptible to VRE and ESBL-Kp colonization.

In Vitro Activity of Ceftolozane-Tazobactam against Anaerobic Organisms Identified during the ASPECT-cIAI Study.

The in vitro activities of ceftolozane-tazobactam, meropenem, and metronidazole were determined against anaerobic organisms isolated from patients with complicated intraabdominal infections (cIAI) in global phase III studies. Ceftolozane-tazobactam activity was highly variable among different species of the Bacteroides fragilis group, with MIC90 values ranging from 2 to 64 µg/ml. More-potent in vitro activity was observed against selected Gram-positive anaerobic organisms; however, small numbers of isolates were available, and, therefore, the clinical significance of these results is unknown. Variable activity of ceftolozane-tazobactam against anaerobic organisms necessitates use in combination with metronidazole for the treatment of cIAI.

Development of EUCAST disk diffusion method for susceptibility testing of the Bacteroides fragilis group isolates.

With the emergence of antibiotic resistance among Bacteroides fragilis group isolates the need of susceptibility testing in routine laboratories is increasing. The aims of the present study were to evaluate the disk diffusion method for susceptibility testing in case of different clinical isolates of Bacteroides spp by comparing zone diameter results with MICs obtained earlier during an Europe-wide antibiotic susceptibility surveillance, and to propose zone diameter breakpoints, which correlate for the EUCAST MIC breakpoints. We tested 381 clinical isolates of the B. fragilis group to amoxicillin/clavulanic acid, cefoxitin, clindamycin, imipenem, metronidazole, moxifloxacin, piperacillin/tazobactam, tigecycline by agar dilution method previously. The inhibition zones of the same antibiotics including meropenem disc were determined by the disc diffusion on Brucella blood agar supplemented with haemin and vitamin K1. Plates were incubated at 37 °C in an anaerobic atmosphere for 24 h. The zone diameters were read at 100% inhibition. In case of discrepant results MICs were determined by gradient test and compared with the inhibition zones on the same plate. We found a good agreement between the inhibition zone diameters and the MICs for imipenem, metronidazole, moxifloxacin and tigecyclin. The inhibition zone diameters of meropenem also separated clearly the isolates, which can be considered wild-type isolates. In case of amoxicillin/clavulanic acid and piperacillin/tazobactam intermediate and susceptible isolates according to the MIC determination, overlap during the zone diameter determination. Isolates with an inhibition zone <23 mm for amoxicillin/clavulanic acid and <25 mm for piperacillin/tazobactam should be retested by a MIC determination method. The 10 µg clindamycin disc clearly separated the resistant and the susceptible population of B. fragilis group strains. In the case of cefoxitin only resistant population could be separated with an inhibition zone <17 mm, intermediate and susceptible isolates overlap. In conclusion, we suggest that disk diffusion can be an option for susceptibility testing of B. fragilis group isolates for most relevant antibiotics in routine laboratories.

Bacteroides thetaiotaomicron and Faecalibacterium prausnitzii influence the production of mucus glycans and the development of goblet cells in the colonic epithelium of a gnotobiotic model rodent.

BACKGROUND: The intestinal mucus layer plays a key role in the maintenance of host-microbiota homeostasis. To document the crosstalk between the host and microbiota, we used gnotobiotic models to study the influence of two major commensal bacteria, Bacteroides thetaiotaomicron and Faecalibacterium prausnitzii, on this intestinal mucus layer. B. thetaiotaomicron is known to use polysaccharides from mucus, but its effect on goblet cells has not been addressed so far. F. prausnitzii is of particular physiological importance because it can be considered as a sensor and a marker of human health. We determined whether B. thetaiotaomicron affected goblet cell differentiation, mucin synthesis and glycosylation in the colonic epithelium. We then investigated how F. prausnitzii influenced the colonic epithelial responses to B. thetaiotaomicron. RESULTS: B. thetaiotaomicron, an acetate producer, increased goblet cell differentiation, expression of mucus-related genes and the ratio of sialylated to sulfated mucins in mono-associated rats. B. thetaiotaomicron, therefore, stimulates the secretory lineage, favoring mucus production. When B. thetaiotaomicron was associated with F. prausnitzii, an acetate consumer and a butyrate producer, the effects on goblet cells and mucin glycosylation were diminished. F. prausnitzii, by attenuating the effects of B. thetaiotaomicron on mucus, may help the epithelium to maintain appropriate proportions of different cell types of the secretory lineage. Using a mucus-producing cell line, we showed that acetate up-regulated KLF4, a transcription factor involved in goblet cell differentiation. CONCLUSIONS: B. thetaiotaomicron and F. prausnitzii, which are metabolically complementary, modulate, in vivo, the intestinal mucus barrier by modifying goblet cells and mucin glycosylation. Our study reveals the importance of the balance between two main commensal bacteria in maintaining colonic epithelial homeostasis via their respective effects on mucus.

Therapeutic equivalence requires pharmaceutical, pharmacokinetic, and pharmacodynamic identities: true bioequivalence of a generic product of intravenous metronidazole.

Animal models of infection have been used to demonstrate the therapeutic failure of "bioequivalent" generic products, but their applicability for this purpose requires the accurate identification of those products that are truly bioequivalent. Here, we present data comparing one intravenous generic product of metronidazole with the innovator product in a neutropenic mouse thigh anaerobic infection model. Simultaneous experiments allowed comparisons (generic versus innovator) of potency and the concentration of the active pharmaceutical ingredient (API), analytical chemistry (liquid chromatography/mass spectrometry [LC/MS]), in vitro susceptibility testing, single-dose serum pharmacokinetics (PK) in infected mice, and in vivo pharmacodynamics (PD) against Bacteroides fragilis ATCC 25825 in synergy with Escherichia coli SIG-1 in the neutropenic mouse thigh anaerobic infection model. The Hill dose-response model followed by curve-fitting analysis was used to calculate and compare primary and secondary PD parameters. The generic and the innovator products were identical in terms of the concentration and potency of the API, chromatographic and spectrographic profiles, MIC and minimal bactericidal concentrations (MBC) (2.0 mg/liter), and mouse PK. We found no differences between products in bacteriostatic doses (BD) (15 to 22 mg/kg of body weight per day) or the doses needed to kill 1 log (1LKD) (21 to 29 mg/kg per day) or 2 logs (2LKD) (28 to 54 mg/kg per day) of B. fragilis under dosing schedules of every 12 h (q12h), q8h, or q6h. The area under the concentration-time curve over 24 h in the steady state divided by the MIC (AUC/MIC ratio) was the best PD index to predict the antibacterial efficacy of metronidazole (adjusted coefficient of determination [AdjR(2)] = 84.6%), and its magnitude to reach bacteriostasis in vivo (56.6 ± 5.17 h) or to kill the first (90.8 ± 9.78 h) and second (155.5 ± 22.2 h) logs was the same for both products. Animal models of infection allow a thorough demonstration of the therapeutic equivalence of generic antimicrobials.

Effect of dosing and dosing frequency on the efficacy of ceftizoxime and the emergence of ceftizoxime resistance during the early development of murine abscesses caused by Bacteroides fragilis and Enterobacter cloacae mixed infection.

The efficacy of beta-lactams is thought to be dependent on the time that the unbound concentrations exceed the MIC (fT>MIC). However, the pharmacokinetic/pharmacodynamic index (PDI) that correlates best to the selection of resistance is not yet clear. The selection of ceftizoxime (CZX)-resistant Enterobacter cloacae mutant strains during the development of murine mixed-infection abscesses was studied to determine the PDI that is important for the emergence of resistance and the PDI value needed for the prevention of resistance. Studies were carried out 24 h after inoculation with Bacteroides fragilis ATCC 23745 and E. cloacae 22491. Six to 1,536 mg of CZX/kg of body weight/day given every 2 h (q2h), q4h, q6h, or q8h was started 30 min before inoculation and continued for 24 h. Resistant mutants were isolated to determine mutant frequencies (MF). The fT>MIC varied from 9 to 98% for E. cloacae, the peak concentration (unbound fraction) was 0.6 to 578 mg/liter, and the area under the concentration-time curve (unbound fraction) (fAUC) was 1.9 to 553 mg.h/liter. The fAUC-to-MIC ratio best explained the in vivo efficacy. CZX-resistant B. fragilis and E. cloacae mutants were isolated from untreated controls at an MF of 10(-5) to 10(-7). The MF of resistant B. fragilis did not increase during therapy. The selection of resistant E. cloacae strains at an MF of 10(-1) to 10(-2) was related to the fT>MIC and the ratio of fAUC to MIC following an inverse U shape. However, the ratio of fAUC to MIC was the stronger driver of resistance. The highest MFs were 0.7 to 0.9 at an fAUC-to-MIC ratio of approximately 250. We conclude that the ratio of fAUC to MIC is the PDI that correlated best to the in vivo efficacy of CZX and probably also to the emergence of resistant E. cloacae mutants. An fAUC-to-MIC ratio of 1,000 was needed to prevent the emergence of this resistance.

Anaerobic sepsis due to multidrug-resistant Bacteroides fragilis: microbiological cure and clinical response with linezolid therapy.

We describe the first reported case of anaerobic sepsis due to Bacteroides fragilis with simultaneous resistance to metronidazole, beta-lactams, beta lactam/beta-lactamase inhibitors, carbapenems, macrolides, and tetracyclines. Microbiological cure and clinical improvement was achieved with linezolid therapy, an agent that may be useful for the treatment of multidrug-resistant anaerobic infections.

Levofloxacin plus metronidazole administered once daily versus moxifloxacin monotherapy against a mixed infection of Escherichia coli and Bacteroides fragilis in an in vitro pharmacodynamic model.

Moxifloxacin has been suggested as an option for monotherapy of intra-abdominal infections. Recent data support the use of a once-daily metronidazole regimen. The purpose of this study was to investigate the activity of levofloxacin (750 mg every 24 h [q24h]) plus metronidazole (1,500 mg q24h) compared with that of moxifloxacin (400 mg q24h) monotherapy in a mixed-infection model. By using an in vitro pharmacodynamic model in duplicate, Escherichia coli and Bacteroides fragilis were exposed to peak concentrations of 8.5 mg of levofloxacin/liter q24h, 32 mg of metronidazole/liter q24h, and 2 mg for moxifloxacin/liter q24h for 24 h. The activities of levofloxacin, metronidazole, moxifloxacin, and levofloxacin plus metronidazole were evaluated against E. coli, B. fragilis, and E. coli plus B. fragilis. The targeted half-lives of levofloxacin, metronidazole, and moxifloxacin were 8, 8, and 12 h, respectively. Time-kill curves were analyzed for time to 3-log killing, slope, and regrowth. Pre- and postexposure MICs were determined. The preexposure levofloxacin, metronidazole, and moxifloxacin MICs for E. coli and B. fragilis were 0.5 and 1, >64 and 0.5, and 1 and 0.25 mg/liter, respectively. Levofloxacin and moxifloxacin achieved a 3-log killing against E. coli and B. fragilis in all experiments, as did metronidazole against B. fragilis. Metronidazole did not decrease the starting inoculum of E. coli. The area under the concentration-time curve/MIC ratios for E. coli and B. fragilis were 171.7 and 85.9, respectively, for levofloxacin and 26 and 103.9, respectively, for moxifloxacin. Levofloxacin plus metronidazole exhibited the fastest rates of killing. The levofloxacin and moxifloxacin MICs for B. fragilis increased 8- to 16-fold after the organism was exposed to moxifloxacin. No other changes in the postexposure MICs were found. Levofloxacin plus metronidazole administered once daily exhibited activity similar to that of moxifloxacin against the mixed E. coli and B. fragilis infection. A once-daily regimen of levofloxacin plus metronidazole looks promising for the treatment of intra-abdominal infections.

Therapeutic efficacy of moxifloxacin, a new quinolone, in the treatment of experimental intra-abdominal abscesses induced by Bacteroides fragilis in mice.

Moxifloxacin, a new quinolone, is effective in vitro against several anaerobic bacteria including Bacteroides fragilis, but its in vivo activity against anaerobic infections is not known. In this study, we evaluated the in vivo activity of moxifloxacin in the treatment of experimentally induced intra-abdominal abscesses (IAA) caused by B. fragilis. For comparison, clindamycin, metronidazole, and levofloxacin were used, and saline for control. Absence of bacteria (sterile) in the abscess pus was required to call it a cure. Mice were intraperitoneally injected with B. fragilis plus sterile rat feces and barium sulfate. Animals were treated with moxifloxacin (40 mg/kg/b.i.d.), clindamycin (75 mg/kg/b.i.d.), levofloxacin (40 mg/kg/b.i.d.) or metronidazole (75 mg/kg/b.i.d.) for 10 days. The cure rate was 12% in controls on saline therapy, 57% on metronidazole, 67% on levofloxacin, 73% on moxifloxacin and 79% on clindamycin. The therapeutic efficacy of moxifloxacin in this B. fragilis infection was not significantly different from that observed with clindamycin. By virtue of its established efficacy on gram-negative aerobic bacteria and the observed in vivo efficacy on B. fragilis, moxifloxacin can be evaluated in the treatment of clinical anaerobic infections.

Effect of moxifloxacin versus imipenem/cilastatin treatment on the mortality of mice infected intravenously with different strains of Bacteroides fragilis and Escherichia coli.

OBJECTIVES: To study the effect of moxifloxacin versus imipenem/cilastatin (hereafter referred to as imipenem) treatment on the mortality of mice infected intravenously with different strains of Bacteroides fragilis and Escherichia coli. METHODS: Groups of 20 mice each were infected intravenously with different strains of B. fragilis [moxifloxacin and imipenem susceptible or resistant, and enterotoxin (ET) positive or negative] and E. coli (moxifloxacin and imipenem susceptible). Twenty-four hours post-infection, intravenous therapy with either moxifloxacin (2.0 mg twice a day) or imipenem (2.4 mg three times a day) was started and continued for 3 days. Control groups were left untreated. Survival rates were recorded at day 7 post-infection. At that time, surviving mice were killed and numbers of bacteria in the liver and kidneys were determined. RESULTS: If compared with untreated animals, mice treated with either moxifloxacin or imipenem showed significantly improved survival (P < 0.001). There was no significant difference (P = 0.97) in the survival rates comparing the two treatment regimens irrespective of the ET positivity or the susceptibility to moxifloxacin or imipenem of the infective B. fragilis strain. However, there was a tendency that B. fragilis was recovered more often from the liver and kidneys of mice infected with ET positive strains. CONCLUSIONS: The data show that moxifloxacin was as efficacious as imipenem in reducing the mortality rate of mice suffering from a severe systemic aerobic/anaerobic infection.

Intra-abdominal anaerobic infections: bacteriology and therapeutic potential of newer antimicrobial carbapenem, fluoroquinolone, and desfluoroquinolone therapeutic agents.

Intra-abdominal infections are biphasic, synergistic processes with early peritonitis and bacteremia due to aerobes and a later abscess component due to anaerobes. Although Bacteroides fragilis is the most commonly recognized pathogen, other anaerobes, including other members of the B. fragilis-group species, are major components of infection. Anaerobic bacteremia is often associated with an intra-abdominal source. New antimicrobial agents with anaerobic activity are in various stages of development for the therapy of intra-abdominal infections. The in vitro activity and the currently available sparse clinical data are reviewed for a new carbapenem (ertapenem), several fluoroquinolones (trovafloxacin, moxifloxacin, and gemifloxacin), and a desfluoroquinolone (BMS-284756).

Antimicrobial resistance and clinical outcome of Bacteroides bacteremia: findings of a multicenter prospective observational trial.

There is debate regarding the correlation between in vitro susceptibility testing and clinical response to therapy for Bacteroides bacteremia. We conducted a prospective multicenter observational study of 128 patients with bacteroides bacteremia. Outcome was correlated with results of in vitro susceptibility testing of Bacteroides isolates recovered from blood and/or nonblood sites, determined with use of 3 end points: mortality at 30 days, clinical response (cure vs. failure), and microbiological response (eradication vs. persistence). The mortality rate among patients who received inactive therapy (45%) was higher than among patients who received active therapy (16%; P=.04). Clinical failure (82%) and microbiological persistence (42%) were higher for patients who received inactive therapy than for patients who received active therapy (22% and 12%, respectively; P=.0002 and.06, respectively). In vitro activity of agents directed at Bacteroides species reliably predicts outcome: the specificity was 97%, and positive predictive value was 82%. Antimicrobial susceptibility testing may be indicated for patients whose blood specimens yield Bacteroides species.