Improvement of a disk diffusion method for antibiotic susceptibility testing of anaerobic bacteria. French recommendations revisited for 2020.

The disk diffusion test is very popular but for anaerobes the main pitfalls arise from the significant variation of diameters for an individual MIC and the weak correlation observed between the MIC's values and diameters zone that generates many major and very major errors. AIMS OF THE STUDY: without any change in the methodology and revisiting only new diameter breakpoints, we try to improve the previous French recommendations and therefore decrease number of errors by introducing recent EUCAST concepts such as ECOFF and ATU Zone. METHOD: MIC determination by agar dilution was done on 100 anaerobes against 6 antibiotics. Clinical categorization was based on EUCAST Breakpoints. Disk-diffusion method was realized on the same Brucella blood agar incubated in an anaerobic chamber. 550 categorizations by both methods could be done as amoxicillin was not tested on the 50 B. fragilis group. As anaerobic infections are severe and treated by antibiotics at higher dosage, we focus on resistance breakpoint to avoid mainly very major errors (VME). Distribution of inhibition zones for each MIC allow us to fix the zone diameter breakpoints. These results were matched to a large data on distribution of zone diameters for each antibiotic collected from two French hospitals from 1990 to 2005. As example for metronidazole and the B. fragilis group, we calculated the cut-off diameter (15 mm) from a wild type population, at a time when there was no resistance to this antibiotic and observed that it was identical to the diameter breakpoint for susceptibility. RESULTS: For an individual value of MIC, the distribution of diameters is wider for anaerobes especially for clindamycin and metronidazole. Using a 15 mm breakpoint for these two antibiotics limits dramatically the number of very major errors but slowly increases the number of major errors that could be overcome by MIC determination if inhibition zone is less than 15 mm. ATU zones (Area of technical uncertainty) were introduced for amoxicillin-clavulanate (17-20 mm), piperacillin-tazobactam (17-20 mm), imipenem (18-23 mm). Categorization inside the ATU requires MIC determination. Finally, out of 550 determinations, VME were observed in 1.45% of cases, an acceptable rate. CONCLUSION: in combination with introduction of ATU zones disk diffusion method allows to detect resistant strains with little MIC determinations and very major errors.

Antibiotic susceptibility of probiotic strains: Is it reasonable to combine probiotics with antibiotics?

OBJECTIVE: The main goal of this study was to determine the in vitro susceptibility of strains collected from marketed probiotics to antibiotics used to treat community-acquired infections. METHODS: The minimum inhibitory concentrations (MICs) of 16 antibiotics were determined using a gradient strip (E test) or the agar dilution method for fidaxomicin. RESULTS: The probiotics demonstrated various antibiotic patterns. Bacterial probiotics are generally susceptible to most prescribed antibiotics orally administered, whereas yeast probiotics, such as Saccharomyces boulardii, are resistant. CONCLUSION: Special attention must be paid to co-prescriptions of antibiotics and probiotics to ensure that the probiotic strain is not susceptible.

Controlled delivery of a new broad spectrum antibacterial agent against colitis: In vitro and in vivo performance.

Coated pellets and mini-tablets were prepared containing a new broad spectrum antibacterial agent: CIN-102, a well-defined, synergistic blend of trans-cinnamaldehyde, trans-2-methoxycinnamaldehyde, cinnamyl acetate, linalool, ß-caryophyllene, cineol and benzyl benzoate. The aim was to provide a new treatment method for colitis, especially for Inflammatory Bowel Disease (IBD) patients. Since the simple oral gavage of CIN-102 was not able to reduce the pathogenic bacteria involved in colitis (rat model), the drug was incorporated into multiparticulates. The idea was to minimize undesired drug release in the upper gastrointestinal tract and to control CIN-102 release in the colon, in order to optimize the resulting antibiotic concentration at the site of action. A particular challenge was the fact that CIN-102 is a volatile hydrophobic liquid. Pellet cores were prepared by extrusion-spheronization and coated with polymer blends, which are sensitive to colonic bacterial enzymes. Mini-tablets were prepared by direct compression. The release of the main compound of CIN-102 (cinnamaldehyde, 86.7% w/w) was monitored in vitro. Optimized coated pellets and mini-tablets were also tested in vivo: in seven-week-old, male mice suffering from dextran sodium sulfate induced colitis. Importantly, both types of multiparticulates were able: (i) to significantly reduce the number of luminal and mucosal enterobacteria in the mice (the levels of which are increased in the disease state), and (ii) to improve the clinical course of the intestinal inflammation (decrease in the percentages of mice with bloody stools and diarrhea). Thus, the proposed coated pellets and matrix mini-tablets allowing for controlled CIN-102 release show a promising potential for new treatment methods of colitis.

Widespread implementation of EUCAST breakpoints for antibacterial susceptibility testing in Europe.

The European Committee on Antimicrobial Susceptibility Testing (EUCAST) was established to harmonise clinical antimicrobial breakpoints and to define breakpoints for new agents in Europe. Data from the European Antimicrobial Resistance Surveillance Network (EARS-Net) external quality assessment (EQA) exercises from 2009 to 2012, from the United Kingdom External Quality Assessment Scheme (UK NEQAS) from November 2009 to March 2013 and data collected by EUCAST through a questionnaire in the first quarter of 2013 were analysed to investigate implementation of EUCAST guidelines in Europe. A rapid change to use of EUCAST breakpoints was observed over time. Figures for implementation of EUCAST breakpoints at the end of the studied period were 61.2% from EARSNet data and 73.2% from UK NEQAS data. Responses to the EUCAST questionnaire indicated that EUCAST breakpoints were used by over 50% of laboratories in 18 countries, by 10 to 50% of laboratories in eight countries and by less than 10% in seven countries. The EUCAST disk diffusion method was used by more than 50% of laboratories in 12 countries, by 10 to 50% of laboratories in ten countries and by less than 10% in eleven countries. EUCAST guidelines implementation is essential to ensure consistent clinical reporting of antimicrobial susceptibility results and antimicrobial resistance surveillance.

In vivo efficacy of microbiota-sensitive coatings for colon targeting: a promising tool for IBD therapy.

The first proof of concept in vivo for a new type of microbiota-sensitive film coatings allowing for colon targeting is presented. The efficacy of these polysaccharide barriers to optimize drug release for the treatment of inflammation is demonstrated in an experimental colitis model with Wister rats. 5-Aminosalicylic acid (5-ASA) pellets were prepared by extrusion-spheronization and coated with Nutriose:ethylcellulose (EC) 1:4 or peas starch:ethylcellulose 1:2 blends. The pellets were mixed with standard chow, and the daily drug dose was 150mg/kg. For reasons of comparison, also commercially available Pentasa pellets and placebo pellets were studied. At day 3 after the beginning of the treatment, colitis was induced by intrarectal administration of trinitrobenzene sulfonic acid (TNBS). Animals were sacrificed on day 6. Macroscopic and histological evaluations of colitis were performed blindly. In addition, inflammatory markers were evaluated using ELISA and real-time PCR. Rats receiving TNBS and placebo pellets developed a severe colitis in the distal half of the colon. 5-ASA administered in the form of Pentasa pellets reduced macroscopic inflammation by only 5%. In contrast, the colon lesions were much less severe upon treatment with Nutriose:EC- and peas starch:EC-coated pellets: The macroscopic score was reduced by 25 and 24%, respectively. Decreases of 37 and 38% of the histological lesions confirmed the efficacy of these new colon targeting systems. Also, inflammatory markers (MPO, IL-1ß mRNA, TNF mRNA) were significantly decreased in rats receiving Nutriose:EC- and peas starch:EC-coated pellets compared to Pentasa pellets. Furthermore, real-time PCR analysis indicated increased activation of the target receptor PPAR-γ and the HMGCS2 gene in rats upon administration of 5-ASA loaded Nutriose:EC- and peas starch:EC pellets compared to the commercial product. Also, HPLC-MS/MS analysis of plasma samples demonstrated that the level of the main metabolite of the drug (N-acetyl-5-ASA) was much lower upon administration of Nutriose:EC or peas starch:EC coated pellets compared to Pentasa pellets, indicating that undesired premature drug release in the upper gastrointestinal tract was more effectively hindered. In addition to the rat study, in vivo imaging of transgenic mice expressing the luciferase gene evidenced much more pronounced PPAR-γ activation upon 5-ASA administration in the form of Nutriose:EC-coated pellets versus Pentasa pellets. All these results clearly demonstrate the superiority of these microbiota-sensitive polysaccharide-based film coatings for colon targeting in vivo.

Anti-anaerobic activity of a new ß-lactamase inhibitor NXL104 in combination with ß-lactams and metronidazole.

NXL104 is a new ß-lactamase inhibitor (BLI) that inhibits class A and class C ß-lactamase enzymes. In this study, the activity of NXL104 in combination with the third-generation cephalosporins ceftazidime (CAZ) and ceftriaxone (CRO) or with piperacillin (PIP) was evaluated against 316 anaerobic bacteria. Minimum inhibitory concentrations (MICs) were determined using an agar dilution method. The BLIs NXL104 or tazobactam (TAZ) were added to the ß-lactams at a fixed concentration of 4 mg/L. A triple combination of NXL104 with an 8:1 ratio of CAZ and metronidazole (MTZ) was also tested. The activities of CAZ, CRO and PIP in combination with NXL104 were enhanced against many of the bacteria. MIC(50) values (MIC for 50% of the organisms) for CAZ+NXL104 were 8-16-fold lower than those of CAZ against Gram-negative anaerobes. Antibiotic resistance rates against all anaerobic strains were: CAZ, 37.7%; CRO, 31%; CAZ+NXL104, 15.2%; CRO+NXL104, 5.4%; and MTZ, 4.1%. No resistant strains could be observed with PIP+TAZ, PIP+NXL104 or the triple combination MTZ+CAZ+NXL104. In conclusion, the triple combination of MTZ+CAZ+NXL104 demonstrated potent antibacterial activity against anaerobes representing most clinical species. It appears appropriate for the treatment of polymicrobial infections, since CAZ+NXL104 also exhibits potent activity against ß-lactamase-producing Enterobacteriaceae and Pseudomonas aeruginosa. It is currently being tested in phase 2 clinical trials for the treatment of complicated intra-abdominal infections.

MALDI-TOF MS Andromas strategy for the routine identification of bacteria, mycobacteria, yeasts, Aspergillus spp. and positive blood cultures.

All organisms usually isolated in our laboratory are now routinely identified by matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) using the Andromas software. The aim of this study was to describe the use of this strategy in a routine clinical microbiology laboratory. The microorganisms identified included bacteria, mycobacteria, yeasts and Aspergillus spp. isolated on solid media or extracted directly from blood cultures. MALDI-TOF MS was performed on 2665 bacteria isolated on solid media, corresponding to all bacteria isolated during this period except Escherichia coli grown on chromogenic media. All acquisitions were performed without extraction. After a single acquisition, 93.1% of bacteria grown on solid media were correctly identified. When the first acquisition was not contributory, a second acquisition was performed either the same day or the next day. After two acquisitions, the rate of bacteria identified increased to 99.2%. The failures reported on 21 strains were due to an unknown profile attributed to new species (9) or an insufficient quality of the spectrum (12). MALDI-TOF MS has been applied to 162 positive blood cultures. The identification rate was 91.4%. All mycobacteria isolated during this period (22) were correctly identified by MALDI-TOF MS without any extraction. For 96.3% and 92.2% of yeasts and Aspergillus spp., respectively, the identification was obtained with a single acquisition. After a second acquisition, the overall identification rate was 98.8% for yeasts (160/162) and 98.4% (63/64) for Aspergillus spp. In conclusion, the MALDI-TOF MS strategy used in this work allows a rapid and efficient identification of all microorganisms isolated routinely.

[Activity of doripenem against anaerobic bacteria]. / Activité du doripénème sur les bactéries anaérobies strictes.

AIMS OF THE STUDY: This study examines the activity of doripenem, a new carbapenem compound compared with amoxicillin-clavulanic acid, piperacillin+tazobactam, imipenem, clindamycin and metronidazole against 316 anaerobes. METHODS: Inoculum preparation and agar dilution method were performed according to the CLSI method for anaerobes (M11A7). RESULTS: At a concentration of 4µg/ml doripenem and imipenem (IMP) inhibited 122 (96 %) and 126 (99 %) strains of the Bacteroides fragilis group, respectively. In contrast, doripenem appeared more potent than IMP against Gram-positive anaerobes inhibiting at the same concentration of 4µg/ml 145/145 strains (100 %) versus 115/145 for IMP (79.3 %). Against 316 anaerobic strains, the carbapenem doripenem had an MIC(50) of 0.25µg/ml and an MIC(90) of 2µg/ml. Results were similar to those for imipenem (MIC(50) of 0.125µg/ml and MIC(90) of 4µg/ml). If we consider the resistant breakpoints of the two carbapenems as defined by EUCAST, the resistance rate for doripenem (MIC>4µg/ml) 1.6 % is similar to that of imipenem (MIC>8µg/ml) 1.3 %. CONCLUSION: Thus independently of the PK/PD parameters the two carbapenems demonstrated very close activity; doripenem was more potent on Gram-positive anaerobes and slightly less potent against Gram-negative anaerobes mainly the B. fragilis group. Further clinical studies are needed to assess its usefulness in patients.

Anaerobic bacteria and antibiotics: What kind of unexpected resistance could I find in my laboratory tomorrow?

The purpose of this article is to set out some important considerations on the main emerging antibiotic resistance patterns among anaerobic bacteria. The first point concerns the Bacteroides fragilis group and its resistance to the combination of ß-lactam+ß-lactamase inhibitor. When there is overproduction of cephalosporinase, it results in increased resistance to the ß-lactams while maintaining susceptibility to ß-lactams/ß-lactamase inhibitor combinations. However, if another resistance mechanism is added, such as a loss of porin, resistances to ß-lactam+ß-lactamase inhibitor combinations may occur. The second point is resistance to metronidazole occurring due to nim genes. PCR detection of nim genes alone is not sufficient for predicting resistance to metronidazole; actual MIC determinations are required. Therefore, it can be assumed that other resistance mechanisms can also be involved. Although metronidazole resistance remains rare for the B. fragilis group, it has nevertheless been detected worldwide and also been observed spreading to other species. In some cases where there is only a decreased susceptibility, clinical failures may occur. The last point concerns resistance of Clostridium species to glycopeptides and lipopeptides. Low levels of resistance have been detected with these antibiotics. Van genes have been detected not only in clostridia but also in other species. In conclusion, antibiotic resistance involves different mechanisms and affects many anaerobic species and is spreading worldwide. This demonstrates the need to continue with antibiotic resistance testing and surveys in anaerobic bacteria.

Impact of infusion method on amikacin serum levels in humans.

Aminoglycosides are broad-spectrum antibiotics with peak-dependent bactericidal activity, administered by gravity infusion or for more accuracy by electronic pump infusion. The aim of this study was to assess the difference between the two systems and its pharmacokinetic impact. Twenty-four patients hospitalised for community-acquired pulmonary infections received amikacin by IV route over 1 h with a targeted peak concentration of 35 mg/L. They were randomly distributed into two groups, one receiving infusion through a pump system, the other by gravity. Amikacin serum levels were determined at the end of infusion and 24 h later. C(max) values were significantly lower with gravity than pump (40.2 +/- 12.3 vs. 50.6 +/- 17.6 mg/L, respectively; p = 0.04). Elimination half-life time, volume of distribution and clearance did not differ significantly from one group to the other. The percentage of patients who failed to achieve the targeted peak concentration was significantly higher with gravity than pump (41.7% vs. 16.7%, respectively; p < 0.001). Improving infusion flow-rate provides better control over amikacin C(max). This study underlines the fact that infusion device characteristics should be added to the physiopathological information of a patient if we are to make a better estimation of pharmacokinetic parameters.

In vitro activity of secnidazole against Atopobium vaginae, an anaerobic pathogen involved in bacterial vaginosis.

Bacterial vaginosis is a polymicrobial syndrome. The most important marker for bacterial vaginosis is the presence of Gardnerella vaginalis and Atopobium vaginae. In this study, the in vitro susceptibilities to metronidazole and secnidazole of 16 strains of A. vaginae were tested with the agar dilution method. We observed an MIC range for metronidazole of 4-64 mg/L (MIC(50), 8 mg/L; MIC(90), 32 mg/L) and an MIC range for secnidazole of 4-128 mg/L (MIC(50), 16 mg/L; MIC(90), 64 mg/L). According to these findings, we can conclude that the activity of secnidazole is similar to that of metronidazole.

Efficacy and tolerance of rifampicin-linezolid compared with rifampicin-cotrimoxazole combinations in prolonged oral therapy for bone and joint infections.

Both linezolid and cotrimoxazole are antibiotics that are well suited for oral therapy of bone and joint infections (BJI) caused by otherwise resistant Gram-positive cocci (GPC) (resistance to fluoroquinolones, maccolides, betalactamines). However, in this context, no data are currently available regarding the safety and tolerance of these antibiotics in combination with rifampicin. The objective of this study was to compare the efficacy and safety of a combination of rifampicin and linezolid (RLC) with those of a combination of rifampicin and cotrimoxazole (RCC) in the treatment of BJI. Between February 2002 and December 2006, 56 adult patients (RLC, n = 28; RCC, n = 28), including 36 with infected orthopaedic devices (RLC, n = 18; RCC, n = 18) and 20 with chronic osteomyelitis (RLC, n = 10; RCC, n = 10), were found to be eligible for inclusion in this study. Patients who discontinued antibiotic therapy within 4 weeks of commencing treatment were considered to represent cases of treatment failure and were excluded. Rates of occurrence of adverse effects were similar in the two groups, at 42.9% in the RLC group and 46.4% in the RCC group (p = 1.00), and led to treatment discontinuation in four (14.3%) RLC and six (21.4%) RCC patients. Cure rates were found to be similar in the two groups (RLC, 89.3%, RCC, 78.6%; p = 0.47). Prolonged oral RLC and RCC therapy were found to be equally effective in treating patients with BJI caused by resistant GPC, including patients with infected orthopaedic devices. However, the lower cost of cotrimoxazole compared with linezolid renders RCC an attractive treatment alternative to RLC. Further larger clinical studies are warranted to confirm these preliminary results.

[Pneumonia in a traveller coming back from Asia]. / Pneumonie chez un voyageur au retour d'Asie.

A case of Salmonella paratyphi A infection was diagnosed late in a patient treated for febrile pneumonia after his returning from India. This case was remarkable in two aspects: first, it illustrated the reemergence of S.paratyphi A infections in people having traveled to India, with increasing fluoroquinolone resistance, and second the difficulty of diagnosing this disease, since the patient was initially treated for pneumonia and flu-like syndrome. Salmonella typhi or paratyphi infections should be evoked in case of persistent fever in patients having traveled to endemic areas, even if digestive signs are absent. Furthermore, choosing an empiric antibiotic treatment with fluoroquinolones could lead to treatment failure if the patient traveled in a country where fluoroquinolone resistance is high, as in Asia and especially in India.

[Monitoring of antibiotic resistance of gram negative anaerobes]. / Surveillance de la résistance aux antibiotiques des anaérobies stricts à Gram négatif.

MATERIAL AND METHOD: Using an agar reference method (Norma M11-A5, National Committee for Clinical and Laboratory Standards) the minimal inhibitory concentrations of nine antibiotics were determined for 376 anaerobic strains. The following strains were investigated: 254 Bacteroides fragilis group (including 143 B. fragilis), 122 other gram-negative anaerobes (Bacteroides spp., Prevotella, Fusobacterium, Porphyromonas, Suterella, Desulfomonas, Veillonella). RESULTS: In the B. fragilis group resistance rates were: coamoxyclav 2.8%, ticarcillin 27.5%, ticarcillin-clavulanic acid 1.9%, piperacillin-tazobactam 1.9%, cefoxitin 6.2%, imipenem 0.8%, clindamycin 28.3%, respectively. Based on previous studies, resistance to imipenem remained low in 2003 and was only observed for B. fragilis. Resistance to clindamycin was maintained around 25%. No metronidazole resistance was observed, but decreased susceptibility was found for B. fragilis, B. merdae and Prevotella, as in 4.3% of gram-negative anaerobes. DISCUSSION: This study confirms the high resistance rate of gram-negative anaerobes to clindamycin, the efficient activity of imipenem, beta-lactam/beta-lactamase inhibitor combinations and metronidazole. However, reduced metronidazole susceptibility seems to be increasing.

European surveillance study on antimicrobial susceptibility of Gram-positive anaerobic cocci.

Gram-positive anaerobic cocci (GPAC) are a heterogeneous group of microorganisms frequently isolated from local and systemic infections. In this study, the antimicrobial susceptibilities of clinical strains isolated in 10 European countries were investigated. After identification of 299 GPAC to species level, the minimum inhibitory concentrations of penicillin, imipenem, clindamycin, metronidazole, vancomycin and linezolid were determined by the agar dilution method according to the Clinical and Laboratory Standards Institute. The majority of isolates were identified as Finegoldia magna and Parvimonas micra (formerly Peptostreptococcus micros), isolated from skin and soft tissue infections. All isolates were susceptible to imipenem, metronidazole, vancomycin and linezolid. Twenty-one isolates (7%) were resistant to penicillin (n=13) and/or to clindamycin (n=12). Four isolates were resistant to both agents. The majority of resistant isolates were identified as F. magna and originated from blood, abscesses and soft tissue infections.

Antimicrobial susceptibilities and clinical sources of Dialister species.

Seventy-four strains representing the four species of the genus Dialister were isolated from various clinical samples. Dialister pneumosintes and Dialister micraerophilus were the two mainly encountered species. Fifty-five isolates were tested against 14 antimicrobial agents. Decreased susceptibilities to piperacillin, metronidazole, macrolides, fluoroquinolones, and rifampin were demonstrated. The clinical impact of these decreased susceptibilities remains to be investigated but should prompt microbiologists to perform antimicrobial susceptibility testing for clinically important Dialister spp.

[Evaluation of the in vitro activity of two betalactams on the oxidative metabolism of polymorphonuclear neutrophils]. / Evaluation in vitro de l'activité de deux bêtalactamines sur le métabolisme oxydatif de granulocytes neutrophiles.

STUDY AIMS: The aim was to evaluate the in vitro effects of amoxicillin and its combination with clavulanic acid, two beta-lactams intravenously injected, on the oxidative metabolism of polymorphonuclear neutrophils. These cells play the major role in the "respiratory burst" as they produce superoxide anion to kill the infectious agent. An activation of this process by the injected antibiotics could enhance the bactericidal action or explain some of adverse effects. MATERIALS AND METHODS: Two models were used to estimate the O(2)(-) amounts produced in the presence of the antimicrobial agents. In the cellular model, O(2)(-) was generated by neutrophils artificially stimulated or not (separated by a gradient centrifugation through Histopaque 1077). In the acellular model, O(2)(-) was produced by the xanthine-xanthine oxidase system. O(2)(-) was measured by spectrophotometry using the ferricytochrome C reduction. RESULTS: The O(2)(-) production by polymorphonuclear neutrophils was increased when both antibiotics were added to the reaction mixture. A significant activation of the cell oxidative metabolism was observed with amoxicillin using various stimulating agents, that was higher without stimulation and lower when amoxicillin and clavulanic acid were associated. CONCLUSION: Amoxicillin could either activate polymorphonuclear neutrophils NADPH-oxidase or cause its activation by a membrane effect, or interfere with the zymosan activation way. It could then be supposed that this antimicrobial agent intensified the bactericidal effects.

In vitro activity against anaerobes of retapamulin, a new topical antibiotic for treatment of skin infections.

OBJECTIVES: Retapamulin is the first agent of the pleuromutilin class formulated as a topical antibacterial for treating skin infections. The aim of this study was to determine the antimicrobial activity of retapamulin by determining the minimal inhibitory concentration (MIC) values of this new drug and comparators against a wide range of anaerobic bacteria of human origin. METHODS: The in vitro activity of retapamulin and six comparators (amoxicillin, amoxicillin/clavulanic acid, ceftriaxone, imipenem, clindamycin and metronidazole) was evaluated against 232 anaerobic clinical isolates. MICs were determined by the CLSI reference agar dilution method (M11-A6). RESULTS: Ceftriaxone, clindamycin and amoxicillin/clavulanic acid resistance rates were 54%, 42% and 9.6%, respectively, within the Bacteroides fragilis group. Despite high resistance rates to various antibiotics, retapamulin inhibited 37/52 (71%) strains of the B. fragilis group and 85/87 (98%) of the other Gram-negative bacilli at a concentration of 2 mg/L or less. All the investigated strains of Clostridium perfringens were inhibited by 1 mg/L retapamulin. Three strains of C. difficile and one strain of C. clostridioforme demonstrated decreased susceptibility to retapamulin. Based on inhibitory concentrations, retapamulin was more active than clindamycin, metronidazole and ceftriaxone against Propionibacterium acnes and anaerobic Gram-positive cocci, as all isolates were inhibited by