INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Infections caused by Corynebacterium species and other facultative, non-spore-forming, gram-positive bacilli have emerged (3, 6, 13), and most recent studies show an alarming rate of antibiotic resistance among such organisms (6, 9, 12, 13, 15, 18). Resistance to -lactams, clindamycin, erythromycin, azythromycin, ciprofloxacin, and gentamicin is quite frequent, with vancomycin, doxycycline, fusidic acid, and prystinamycin being the agents that are most active in vitro (6, 12, 13, 15, 18).

Ketolides are a new class of macrolide-like antibiotics having a mechanism of action like those of macrolides but with greater in vitro activity against multidrug-resistant gram-positive organisms, including staphylococci, enterococci, pneumococci, and anaerobes (4, 5, 7, 14). In the present study we compared the in vitro activity of the ketolide HMR 3647 (RU 66647) with the activities of 14 other agents (including 8 macrolide-lincosamide-streptogramin antibiotics) against more than 240 isolates of aerobic and facultative non-spore-forming gram-positive bacilli.

(Part of this work has been presented at the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Ontario, Canada, 28 September to 1 October 1997 [16].)

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Bacterial strains. A total of 210 aerobic and facultative, non-spore-forming, gram-positive bacilli isolated from clinical samples in our laboratory were examined. In addition, 37 other clinical isolates, kindly provided by L. Martínez-Martínez (Department of Microbiology, University of Seville, Seville, Spain), were also included. All strains were collected from 1987 to 1996. Prior to testing, strains were subcultured, checked for purity, and reidentified by standard techniques (2), with additional tests (6) performed if necessary.

Antimicrobial agents. HMR 3647 (RU 66647), roxithromycin, azithromycin, clarithromycin, levofloxacin, vancomycin, teicoplanin, and rifapentine were obtained from Hoechst-Marion-Roussel (Romainville, France); lincomycin and benzylpenicillin were from Fluka Chemie AG (Buchs, Switzerland), quinupristin-dalfopristin (30/70) and pristinamycin were from Rhône-Poulenc-Rorer (Collegeville, Pa., and Vitry sur Seine, France, respectively); josamycin was from ICN Biomedicals, Inc. (Aurora, Ohio), erythromycin was from Pierrel (Milan, Italy), and doxycycline was from Sigma Chemical Co. (St. Louis, Mo.).

MIC determinations. The MICs were determined by a standard agar dilution method (11) in Mueller-Hinton agar supplemented with 5% defibrinated sheep blood. The plates were incubated aerobically at 35°C for 24 or 48 h, as required, to determine the MIC endpoint. Breakpoints for susceptibility were those proposed by the National Committee for Clinical Laboratory Standards, as follows: erythromycin, 0.5 µg/ml; clarithromycin, azythromycin, benzylpenicillin, and levofloxacin, 2 µg/ml; doxycycline and vancomycin, 4 µg/ml; and teicoplanin, 8 µg/ml (11). For antibiotics for which the National Committee for Clinical Laboratory Standards has no proposed breakpoint and only for comparison purposes, we used the following breakpoints for susceptibility: HMR 3647 and lincomycin, 0.5 µg/ml; josamycin, roxithromycin, rifapentine, and quinupristin-dalfopristin, 1 µg/ml; and pristinamycin, 2 µg/ml (1, 7, 17). Staphylococcus aureus ATCC 29213, Enterococcus faecalis ATCC 29212, and Escherichia coli ATCC 25922 were used as controls and were tested up to eight times.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

The results of the tests in which the MICs for the clinical isolates were determined are presented in Table 1. HMR 3647 was very active against most organisms tested (MIC at which 50% of isolates are inhibited [MIC₅₀], 0.25 µg/ml) except Corynebacterium striatum, coryneform CDC group I2, and Oerskovia spp. The MIC₅₀ and MIC₉₀ for Corynebacterium jeikeium were 0.12 and >128 µg/ml, respectively. The frequency of resistance to erythromycin and the other macrolides tested as well as to lincomycin was very high, with the activity of clarythromycin being superior to those of the other macrolides tested. The frequency of resistance to benzylpenicillin was high, particularly in Corynebacterium urealyticum, C. jeikeium, Rhodococcus equi, Oerskovia spp., Corynebacterium afermentans, and coryneform CDC group I2. The activities of doxycycline and quinupristin-dalfopristin were also very high, although resistance to the latter was observed mainly in Listeria, R. equi, Erysipelothrix rhusiopathiae, C. striatum, and Oerskovia strains. Levofloxacin was also very active against most strains tested except Corynebacterium amycolatum, C. urealyticum, C. jeikeium, and Oerskovia strains. Rifapentine was active against many species tested, but resistance to this agent was always present in E. rhusiopathiae and was very common in C. striatum, C. urealyticum, and C. amycolatum. Vancomycin, teicoplanin, and pristinamycin were very active against most strains tested, with only E. rhusiopathiae being resistant to glycopeptide antibiotics and some strains of R. equi, Listeria spp. (but no Listeria monocytogenes), and Oerskovia spp. being resistant to pristinamycin. The activity of the ketolide against erythromycin-susceptible and -nonsusceptible strains is represented in Table 2. HMR 3647 was very active against all erythromycin-sensitive strains and 32 of the 34 (94%) erythromycin-intermediate (MICs, 1 to 4 µg/ml) strains. Among the 80 erythromycin-resistant (MICs, 8 µg/ml) strains tested, 44 (55%) were susceptible to HMR 3647 and the MICs for the remaining strains were 1 µg/ml. Such resistance occurred mainly in C. striatum, C. urealyticum, and C. jeikeium strains, of which only 5, 17, and 61% of erythromycin-resistant strains, respectively, were inhibited by a concentration of 0.5 µg HMR 3647 per ml.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The results of this study confirm those of previous publications showing a high incidence of erythromycin-resistant strains among aerobic and facultatively anaerobic, non-spore-forming, gram-positive bacilli (12, 13, 15, 18). Therefore, erythromycin, once considered to be the drug of choice for the treatment of infections caused by these organisms, can no longer be recommended for use in treatment unless an in vitro study confirms the erythromycin susceptibility of the infecting strain. The ketolide was very active against erythromycin-susceptible strains (MIC₉₀ range, 0.015 to 0.25 µg/ml). On the other hand its activity against erythromycin-nonsusceptible strains varied (MIC₉₀ and MIC₅₀ ranges, 0.06 to >128 and 0.03 to 16 µg/ml, respectively). Most erythromycin-intermediate strains (94%) were susceptible to the ketolide, as were 55% of the erythromycin-resistant strains. Resistance to HMR 3647 was particularly common among erythromycin-resistant strains of C. striatum and C. urealyticum. Macrolides other than erythromycin showed activity similar to that obtained with erythromycin, with clarithromycin being much more active due not only to its higher susceptibility breakpoint (2 µg/ml) but also to its greater intrinsic activity. Lincomycin was active against all Arcanobacterium haemolyticum and Arcanobacterium pyogenes strains as well as many strains of E. rhusiopathiae. Pristinamycin was very active against most strains tested, with only some strains of R. equi, Listeria spp., and Oerskovia spp. being resistant to this drug. The in vitro activity of quinupristin-dalfopristin (both drugs are pristinamycin derivatives) was similar to that of pristinamycin, although the MIC of pristinamycin was two to eight times lower than those obtained with the combination derivatives. The combination quinupristin-dalfopristin has recently been tested against L. monocytogenes and C. jeikeium (1, 10, 14), and the MICs were slightly lower than (1, 10) or similar to (14) those that we obtained for C. jeikeium. However, published data for L. monocytogenes indicate that the MIC₉₀s for this strain are between 1 and 2 µg/ml (1, 10, 14), while our results gave slightly higher MIC₉₀s (4 µg/ml). Resistance to macrolide-lincosamide-streptogramin antibiotics was probably due to target modification (erythromycin, josamycin, and lincomycin resistance) in most strains (8). However, resistance from drug inactivation (lincomycin resistance and erythromycin and josamycin sensitivity) probably occurred in C. jeikeium, C. striatum, C. minutissimum, R. equi, and Listeria spp. (8). In addition, resistance to macrolide-lincosamide-streptogramin antibiotics due to other non-well-defined mechanisms may have been present in C. urealyticum, C. minutissimum, R. equi, and Listeria spp.

The data obtained with glycopeptide antibiotics (vancomycin and teicoplanin), doxycycline, and pristinamycin confirm those presented in previously published reports describing the marked activities of these compounds (12, 15). On the other hand, our data also indicate that benzylpenicillin remains very active against many organisms but not against C. urealyticum, C. jeikeium, R. equi, and Oerskovia spp. so that it can be used, alone or in combination with aminoglycosides, for the treatment of infections, including endocarditis, caused by penicillin-susceptible organisms. Data on the in vitro activity of levofloxacin against the organisms studied are scarce, and this antibiotic is very active against many species but not against C. amycolatum, C. urealyticum, and C. jeikeium. Although this study does not compare the activity of levofloxacin with those of other quinolones, the results obtained in this study are very similar to those published previously for the activity of ciprofloxacin against other gram-positive bacilli (9, 13, 15).

Rifapentine was very active against most organisms tested, although resistance in E. rhusiopathiae and isolates of C. urealyticum and C. striatum was quite common. The results with rifapentine are very similar to those previously published for rifampin against coryneform organisms (13, 15).