Journal of Antimicrobial Chemotherapy (1996) 37, Paul H. Edelstein' J, Martha A. C. Edelstein', K. H. Lehr f.

Transcription

1 Journal of Antimicrobial Chemotherapy (1996) 37, In-vitro activity of levofloxacin against clinical isolates of Legionella spp, its pharmacokinetics in guinea pigs, and use in experimental Legionella pneumophila pneumonia Paul H. Edelstein' J, Martha A. C. Edelstein', K. H. Lehr f and Jianjun Ren' 'Departments of Pathology and Laboratory Medicine and b Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA , USA; 'Department of Corporate Kinetics and Metabolism, Hoechst AG, Frankfurt 65926, Germany The activities of levofloxacin and ofloxacin against 22 clinical legionella isolates was determined by microbroth dilution susceptibility testing. Growth inhibition of two Legionella pneumophila strains grown in guinea pig alveolar macrophages by levofloxacin, ofloxacin, or erythromycin was also determined. The drug concentrations required to inhibit 90% of strains tested was mg/l for levofloxacin or ofloxacin, and was mg/l for ciprofloxacin. BYEa broth significantly inhibited the activities of all three drugs tested, as judged by the susceptibility of control Escherichia coli strains. Levofloxacin (0.25 mg/l) reduced bacterial counts of two L. pneumophila strains grown in guinea pig alveolar macrophages by 1 log 10, but regrowth occurred over a 3 day period; levofloxacin (1 mg/l) reduced bacterial counts by 2-3 logio cfu/ml. Levofloxacin was significantly more active than erythromycin, and as active as ofloxacin or ciprofloxacin in this assay. Pharmacokinetic and therapy studies of levofloxacin and ofloxacin were performed in guinea pigs with L. pneumophila pneumonia. For the pharmacokinetic study, levofloxacin was given (lomg/kg) by the intraperitoneal route to infected guinea pigs; mean peak plasma and lung concentrations were 3.4 mg/l and 1 4 /Jg/g, respectively, at 0.5 h and 2.6 mg/l and 0.6/jg/g at 1 h. The terminal half-life phase of elimination from plasma and lung was c. 1 h. All 15 infected guinea pigs treated with levofloxacin (10 mg/kg/day given ip once daily) for 5 days survived for 9 days after antimicrobial therapy, as did all 14 guinea pigs treated with the same dose of ofloxacin. None of 13 animals treated with saline survived. Levofloxacin is effective against L. pneumophila in vitro and in a guinea pig model of legionnaire's disease. Levofloxacin should be evaluated as a treatment of human legionnaires' disease. Introduction Levofloxacin, the levorotary component of the racemic mixture comprising ofloxacin, is more active against most bacteria than is ofloxacin (Davis & Bryson, 1994). Otherwise the pharmacokinetics and other properties of the drug appear to be similar to ofloxacin. Since ofloxacin is known to be active in vitro and in vivo against Legionella pneumophila, it is reasonable to assume that the same would hold for levofloxacin (Saito et al,, 1985; DeForges et al., 1986; Bertrand et al., 1987; Fitzgeorge, Featherstone & Baskerville, 1988; Leroy et al., 1989; Mouton et al., 1990; Rajagopalan-Levasseur et al., 1990; /96/ $12.00/0 % 1996 The British Society for Antimicrobial Chemotherapy

2 118 P. H. Edelstein et al. Hardy, 1991; Jacobs, 1991; Kitsukawa, Hara & Saito, 1991; Wynckel et al., 1991; Gaja, 1992). We demonstrate that levofloxacin is as or more active than ofloxacin against Legionella spp. in vitro, in L. pneumophila infected alveolar macrophages, and in a guinea pig model of L. pneumophila pneumonia. Bacterial strains and growth conditions Methods All legionellae studied were low-passage clinical isolates. These strains were identical to those used in previous studies, and comprised two strains each of Legionella dumoffii, Legionella longbeachae, Legionella micdadei, one strain of Legionella bozemanii and 15 strains of L. pneumophila (Edelstein & Edelstein, 1989; Edelstein et al., 1990). Staphylococcus aureus ATCC and Escherichia coli, ATCC were used as control organisms for susceptibility testing. To obtain inocula for susceptibility testing, legionellae were grown on locally made MOPS-buffered charcoal yeast extract medium supplemented with 0.1% a-ketoglutarate (BCYEa), and other organisms were grown on commercial tryptic soy agar containing 5% sheep blood (Edelstein & Edelstein, 1993). All cultures were incubated at 35 C in humidified air for h, depending on organism and growth rate. Antimicrobials Standard powders of levofloxacin and ofloxacin were obtained from Hoescht Roussel-Uclaf, Romainville, France. Erythromycin base and ciprofloxacin were obtained from Abbott Laboratories, North Chicago, Illinois, and Miles Laboratories, West Haven, Connecticut, respectively. To prepare levofloxacin for injection, the standard powder was dissolved in sterile saline for injection, USP. The drug concentration used was 3.5 mg/ml for both the pharmacokinetic and treatment studies. Levofloxacin solution for injection was prepared daily and used within 1 h. Ofloxacin for injection was obtained from McNeil Pharmaceuticals, Springhouse, Pennsylvania. Antimicrobial susceptibility testing Microbroth dilution susceptibility testing was performed using BYEa broth (Legionella) or Mueller-Hinton broth (other bacteria), with a final volume of 200/JL and final bacterial concentration of 5 x 10 s cfu/ml (Edelstein, 1985). Otherwise the microbroth method was performed exactly as described previously for a macrobroth method (Edelstein & Edelstein, 1989). All testing was done in duplicate; when discrepancies occurred, the geometric mean value was used as the MIC. Antimicrobial susceptibility testing with erythromycin for the bacteria used in this study has been reported previously (Edelstein & Edelstein, 1989). Growth inhibition in alveolar macrophages Guinea pig pulmonary alveolar macrophages were harvested and purified as described previously (Edelstein & Edelstein, 1989). The final concentration of macrophages was

3 Levofloxacin for Legionella 119 approximately 10 5 cells/well. Incubation conditions for all macrophage studies were 5% CO 2 in air at 37 C. L. pneumophila strains F889 and F2111 grown overnight on BCYEa agar were used to infect the macrophages. Approximately 10 4 bacteria were added to each well. Bacteria were incubated with macrophages for 1 h in a shaking incubator, and then for 1 day in stationary culture, as described previously (Edelstein & Edelstein, 1989). One set of replicate wells was washed (500/iL) three times with tissue culture medium, then sonicated at low energy to release intracellular bacteria, which were quantified using BCYEa agar. Antimicrobials were then added to the washed non-sonicated wells; several wells without antimicrobial served as growth controls. The infected tissue cultures were then incubated for 2 days, after which supernatant samples were taken for quantitative culture. The antimicrobials were then removed by washing, and the experiment continued for a further 4 days, with daily quantification of L. pneumophila in well supernatants. All experiments were carried out in duplicate or triplicate, and quantitative plating was done in duplicate. All wells were examined microscopically daily to detect macrophage infection and count approximate numbers of macrophages in the wells. In this system there is no extracellular growth of L. pneumophila, so all increases in supernatant bacteria are the result of intracellular growth. Guinea pig pneumonia model Hartley strain male guinea pigs, c. 320 g, were used for the pneumonia model, as previously described (Edelstein, Calarco & Yasui, 1984). Animals were observed for illness 1 week before infection; in the case of the animals used for the treatment study, temperatures and weights were obtained during the preinfection period. The guinea pigs were infected with L. pneumophila serogroup 1, strain F889, administered intratracheally. About 4 x 10 6 cfu were administered for both the pharmacokinetic and treatment studies. All studies with animals were approved by the University of Pennsylvania Institutional Animal Use and Care Committee. Pharmacokinetic study Levofloxacin plasma and lung concentrations were measured in guinea pigs with L. pneumophila pneumonia as described previously (Edelstein el al., 1990). The drug was given in a single ip dose (10 mg/kg in c. 0.9 ml) to guinea pigs 1 day after infection; the mean guinea pig weight was 345 g. At timed intervals after drug injection, anaesthetized animals in groups of two to four were exsanguinated by removal of heart blood under direct vision. The lungs were then removed, rinsed in sterile saline to remove adherent blood, blotted dry on gauze, and weighed. Heart blood was collected with a syringe and needle, then transferred immediately to heparinised tubes (Vacutainer, Becton-Dickinson, Rutherford, New Jersey). The heparinised blood was refrigerated immediately afterwards at 5 C. Within 2-12 h, the plasma was separated from the cellular blood components by centrifugation at 5000 at 5 C for lomin. Before separation of the plasma, a small amount of the heparinised blood was set aside and stored at 5 C for later determination of haemoglobin concentration. One lung was homogenized in a known volume of phosphate buffered saline, ph 7, 0.01 M. Lung homogenates were then centrifuged at 14,000 g at room temperature (21 C C to 23 C) for 5 min. A small amount of lung homogenate supernatant was set aside and stored at 5 C C

4 120 P. H. Edelstein et at. for later determination of haemoglobin concentration. Measured volumes of blood plasma and lung homogenate supernatants were transferred to microcentrifuge tubes and stored at 20 C. At a later time, acetonitrile was added in equal volume to the fluids to be analysed, the suspension vortex mixed, and then centrifuged at 14,000 # for 1 min. The supernatant was transferred to another tube and frozen at 70 C until analysed by HPLC. Negative controls included guinea pig plasma and lung homogenate supernatant that had been similarly collected from normal guinea pigs given identical anaesthesia but no antimicrobial. Haemoglobin assay Haemoglobin determinations were performed on blood and lung homogenate specimens to determine the degree of blood contamination of the lungs. Haemoglobin content was determined using a commercial colorimetric assay (Sigma Chemical, catalog #525-A, St Louis, Missouri), according to the manufacturer's directions. Drug assay The acetonitrile deactivated specimens (0.2 ml) were mixed with a ciprofloxacin internal standard (0.05 ml) and acetonitrile (0.2 ml). The specimens were then centrifuged, and the supernatants added to 0.2 M phosphoric acid, adjusted to ph 1.85 with tetraethylammonium hydroxide (0.15 ml), and chloroform (2 ml). A portion (20 fil) of the resulting aqueous supernatant was injected into the chromatograph. Levofloxacin was quantified in the samples by high-pressure liquid chromatography on a Nucleosil C18 column (125 x 4.6 mm, 5 /im particle size, Macherey-Nagel, Diiren, Germany). The mobile phase consisted of 220 parts 0.2 M phosphoric acid, adjusted to ph 1.85 with tetraethylammonium hydroxide, seven parts tetrahydrofurane, and four parts acetonitrile. The flow rate was 1.5mL/min, and column temperature 40 C. Detection was by a 821-FP fluorescence detector (Jasco, GroB-Umstadt, Germany) at AEX = 298 nm, and AEm = 458 nm. A standardisation curve for peak-height ratio was determined using different concentrations of levofloxacin contained in normal guinea pig plasma and lung homogenate supernatant, and was found to be linear over the concentration range of mg/l, and mg/l, respectively. The lower limit of detection was 0.1 and 0.05 mg/l for plasma and lung respectively. Animal treatment study Guinea pigs surviving surgery were randomised into three treatment groups 1 day after infection. Starting on that day, treatment was given once daily in the morning for 5 days. One group of animals received levofloxacin (10 mg/kg in 1.0 ml), another 14 animals received the same dose of ofloxacin, and the last group of 13 animals received 1.0 ml of normal saline. All dosing was by the ip route. Animal weights were measured periodically during the 14 day post-infection observation period. Necropsies and quantitative lung cultures were performed on all animals that died. All animals surviving for 14 days post-infection were killed with pentobarbital. Necropsies, lung histopathology, and quantitative lung cultures were performed on the seven lowest weight survivors from each treatment group (Edelstein et al., 1984).

5 Levofloxacin for Legionella 121 Broth dilution susceptibility Results All 22 legionella strains tested by the agar dilution method were susceptible to concentrations of levofloxacin and ofloxacin readily achievable in human serum (Davis & Bryson, 1994) (Table I). The average MIC, the MIC required to inhibit 50% of strains, the MIC required to inhibit 90% of strains, and the MIC range of levofloxacin were 0.018,,, and to mg/l respectively. Ofloxacin MICs were generally one two-fold dilution higher. The levofloxacin and ofloxacin MICs for L. pneumophila strains F889 and F2111 were and mg/l, respectively. Ciprofloxacin showed similar activity to levofloxacin. All three drugs tested were variably inhibited by BYEa broth, depending on the test bacterium. There were no significant differences in the MICs for the control S. aureus strain between BYEa and Mueller-Hinton broths. The MICs of all three antimicrobial agents tested for E. coli however were two two-fold dilutions lower in Mueller-Hinton broth than in BYEa broth. Antimicrobial inhibition of intracellular growth Both L. pneumophila serogroup 1 strains grown in guinea pig alveolar macrophages were significantly inhibited by levofloxacin, ofloxacin, and erythromycin (Figure 1). Levofloxacin and ofloxacin were significantly more inhibitory than Table I. MICs (mg/l) for 22 Legionella spp. of three antimicrobial agents Strain Identity Levofloxacin Ofloxacin Ciprofloxacin < \ (Los Angeles 1) L. micdadei L. longbeachae SG L. micdadei L. dumoffii SG 1 (Knoxville) L. bozemanii SG 2 L. dumoffii SG ( S < ) (Philadelphia 1)

6 122 P. H. Edelsteln et al (b) / 2 / 4 ^^ (d) 5 / / 2 / 4 6 i 8 * 3 i -r' 3 J _^--, i. i 1 \ 7. v 1 i 1 I 1, Time after infection (days) Figure 1. Growth of L. pneumophila serogroup 1 strains F889 (a) and F2111 (b) in guinea pig alveolar macrophages after infection. All points represent the mean of triplicate wells counted in duplicate; error bars represent 95% confidence intervals, which unless shown were smaller than the height of the symbol representing the mean. The lower limit of detection of the assay is log l0 1.5cfu/mL; undetectable bacterial counts arc shown graphically as logio 1.0 cfu/ml. (c), (d) Represent the same experiment, to show all results more clearly, (a), (b). Control; +, levofloxacin 0 25 mg/l, D, levofloxacin 1 mg/l; *, erythromycin 1 mg/l; 0, ofloxacin 0.25 mg/l; O, oftoxacin I mg/l (c), (d), Control; 0. levofloxacin 0 25 mg/l;, levofloxacin I mg/l; *, ciprofloxacin 0.25 mg/l; V, ciprofloxacin 1 mg/l. erythromycin 0.25 mg/l of levofloxacin or ofloxacin were more active than 1.0 mg/l of erythromycin. In addition both 0.25 and 1 mg/l of levofloxacin or ofloxacin caused a more prolonged post-antibiotic effect than erythromycin. Levofloxacin showed similar activity to ofloxacin and was slightly less active than ciprofloxacin in these assays. In none of the macrophage experiments did macrophage monolayers become detached before visible onset of infection with L. pneumophila, indicating the lack of macrophage toxicity by the antibiotics tested. Pharmacokinetic study Pharmacokinetic data are shown in Figure 2. None of the negative control samples contained measurable drug. The levofloxacin plasma and lung terminal phase (/} phase) half-lives of elimination were calculated to be c. 1.0 h. The lung homogenate supernatants contained an average of 6% of the haemoglobin concentration of the corresponding blood specimens, with a range of 5% to 16%, and 99% confidence interval of 6.4% to 6.4%. Such a low level of blood contamination of the lungs confirms that the measured lung levofloxacin concentrations were due to lung accumulation of drug rather than an artefact of blood contamination during processing.

7 Levofloxacin for Legionella p Tfc 1.00 r I I i I I I I I I I, I I I, I I I I I r X r : Time after injection (h) Figure 2. Mean plasma (a) and lung (b) levofloxacin concentrations in guinea pigs with L. pneumophila pneumonia. Animals were given a single 10 mg/kg dose administered by the intraperitoneal route at time 0 h. Three animals were sampled at each time point The vertical bars represent the ranges for each time point. The dashed horizontal lines represent the lower limit of detection of the assays; data points below the limit of detection are plotted arbitrarily below the detection limit. The curved dashed lines show the theoretical decay curves. Therapy in guinea pigs All guinea pigs treated with levofloxacin (n = 15) or ofloxacin (n = 14) survived, compared with 100% deaths in the 13 guinea pigs who received saline alone (Figure 3). All lung cultures and necropsy results of all saline treated animals were diagnostic of L. pneumophila pneumonia; the mean concentration of L. pneumophila was logio 9.5 cfu/g of lung, with a range of logm 9.4 to 10.2 cfu/g. Three of the seven lungs examined from the ofloxacin treatment group survivors were positive for L. pneumophila; these contained logio 2.4, 3.2, and 2.2 cfu/g. Two of the seven lungs examined from the levofloxacin treatment group contained L. pneumophila; these contained log,o 2.8 and 3.5 cfu/g. Both animal weights and temperatures were identical for the levofloxacin and ofloxacin treatment groups (Figure 3; temperature data not shown). No significant differences in lung histopathology were observed for the levofloxacin and ofloxacin treatment group lungs examined (Table II). Infiltration by macrophages and lymphocytes characterised the consolidation observed in some lungs. Some fibrosis, but almost no necrosis or infiltration by polymorphonuclear leucocytes, was seen in the consolidated areas. Discussion Levofloxacin, like ofloxacin, was very active against Legionella spp. in vitro. In addition,

8 124 (b) ^ r i ii 11 i i i i Time after infection (days) Figure 3. Percentage survival (a) and weights (g) (b) of guinea pigs with L. pneumophila pneumonia after infection. Animals were treated on post-infection days 1 to 5 Numbers of animals in each group were saline (. 13), levofloxacin (O, 15), and ofloxacin (A, 14) levofloxacin was active against intracellular L. pneumophila, as judged by its activity in infected macrophages, and in a guinea pig model of L. pneumophila pneumonia. Levofloxacin was about twice as active as ofloxacin in vitro. Other investigators have made similar findings, using different methods (Fujimoto & Mitsuhashi, 1990; Kitsukawa el al., 1991). Our method of determining antimicrobial agent MICs for Legionella spp. may have overestimated the actual MICs, as judged by higher MICs for the control E. coli, but not for the 5. aureus strain. All known methods of testing the susceptibility of Legionella spp. bacteria in vitro share this problem, although the meaning of a 'true' MIC is unclear (Edelstein, 1995). Levofloxacin, ofloxacin, and ciprofloxacin were more active against intracellular L. pneumophila than was erythromycin, for the two strains studied. Previous studies by us and others have shown that erythromycin is only inhibitory in this and other Table II. Number of guinea pigs with lung consolidation in the two treatment groups % Consolidation Ofloxacin Levofloxacin >20

9 Levofloxacin for Legionella 125 macrophage systems, even at concentrations as high as 5 mg/l (Horwitz & Silverstein, 1983; Vilde, Dournon & Rajagopalan, 1986; Edelstein & Edelstein, 1991). In contrast, most fluoroquinolone antimicrobial agents tend to be cidal against L. pneumophila in macrophages, as was observed in these experiments for all three agents studied at a concentration of 1 mg/l (Vilde et al., 1986; Havlichek, Saravolatz & Pohlod, 1987; Edelstein & Edelstein, 1989; Edelstein et al., 1990; Edelstein & Edelstein, 1991; Kitsukawa et al., 1991). In addition, a relatively long period was required for regrowth of L. pneumophila after removal of drug (1 mg/l) for all compounds studied except erythromycin. Such a prolonged 'post-antibiotic' effect has been regularly observed for fluoroquinolone antimicrobial agents tested using these methods (Edelstein & Edelstein, 1989; Edelstein et al., 1990; Rajagopalan-Levasseur et al., 1990). Both levofloxacin and ofloxacin were slightly less active than ciprofloxacin in this assay, although the meaning of such differences is uncertain. Kitusawa et al. (1991) reported similar findings in another macrophage model. The pharmacokinetic profile of plasma levofloxacin in the guinea pig is different from that in man (Davis & Bryson, 1994). Levofloxacin administered to guinea pigs was cleared much more rapidly than in man; in man the serum half-life of terminal elimination (/? phase) is about 4-7 h, in contrast to the 1 h half-life measured in our study. This implies that levofloxacin treatment of legionnaires' disease in man may be more effective than in guinea pigs, based on pharmacokinetic considerations only. Levofloxacin was very effective for the treatment of experimental Legionnaires' disease, even when given only once daily. It appeared to be as active as ofloxacin in the animal model. Our results with ofloxacin therapy in the animal model are similar to those of Saito et al. (1985) and Gaja (1992), who found ofloxacin to be more active than erythromycin or ciprofloxacin. There appear to be very good correlations of drug effectiveness in this animal model and clinical response in the treatment of human legionnaires' disease (Edelstein, 1995). In fact, ofloxacin has been reported to be active in the treatment of human Legionnaires' disease (Bertrand et al., 1987; Leroy et al., 1989; Mouton et al., 1990; Wynckel et al., 1991). Our results suggest that levofloxacin will be equally effective. Acknowledgements Funded in part by Hoescht Roussel-Uclaf, Romainville, France. Zunxuan Chen provided expert technical assistance. References Bertrand, A., Janbon, F., Despaux, E., Jonquet, O. & Reynes, J. (1987). L'ofloxacine (RU 43280). Etude clinique. Pathologie Biologie 35, Davis, R. & Bryson, H. M. (1994). Levofloxacin. A review of its antibacterial activity, pharmacokinetic and therapeutic efficacy. Drugs 47, DeForges, L., Fournet, M. P., Soussy, C. J. & Duval, J. (1986). Etude de l'activitite antibacterienne in vitro de dix quinolones sur vingt souches de Legionella pneumophila. Pathologie Biologie 34, Edelstein, P. H. (1985). Legionnaires' Disease Laboratory Manual, 3rd edn. National Technical Information Service, Springfield, VA. Edelstein, P. H. (1995). Antimicrobial therapy of Legionnaires' disease A review. Clinical Infectious Diseases. Edelstein, P. H., Calarco, K. & Yasui, V. K. (1984). Antimicrobial therapy of experimentally

10 126 P. H. Edelstein et al. induced Legionnaires' disease in guinea pigs. American Review of Respiratory Diseases 130, Edelstein, P. H. & Edelstein, M. A. C. (1989). WIN is bactericidal for Legionella pneumophila grown in alveolar macrophages. Antimicrobial Agents and Chemotherapy 33, Edelstein, P. H. & Edelstein, M. A. C. (1991). In vitro activity of azithromycin against clinical isolates of Legionella species. Antimicrobial Agents and Chemotherapy 35, Edelstein, P. H. & Edelstein, M. A. C. (1993). Comparison of three buffers used in the formulation of buffered charcoal yeast extract medium. Journal of Clinical Microbiology 31, Edelstein, P. H., Edelstein, M. A. C, Weidenfeld, J. & Dorr, M. B. (1990). In vitro activity of sparfloxacin (CI-978; AT-4140) for clinical Legionella isolates, pharmacokinetics in guinea pigs, and use to treat guinea pigs with L. pneumophila pneumonia. Antimicrobial Agents and Chemotherapy 34, Fitzgeorge, R B., Featherstone, A. S. & Baskerville, A. (1988). The effect of ofloxacin on the intracellular growth of Legionella pneumophila in guinea pig alveolar phagocytes. Journal of Antimicrobial Chemotherapy 22, Suppl. C, Fujimoto, T. & Mitsuhashi S. (1990). In vitro antibacterial activity of DR-3355, the S-(-)-isomer of ofloxacin. Chemotherapy 36, Gaja, M. (1992). In vitro and in vivo antibacterial activity of four newly developed quinolone agents against Legionella infection. Chemotherapy 40, Hardy, D. J. (1991). In vitro activity of temafloxacin against gram-negative bacteria: an overview. American Journal of Medicine 91, Suppl. 6A, 19S-23. Havlichek, D., Saravolatz, L. & Pohlod, D. (1987). Effect of quinolones and other antimicrobial agents on cell-associated Legionella pneumophila. Antimicrobial Agents and Chemotherapy 31, Horwitz, M. A. & Silverstein, S. C. (1983). Intracellular multiplication of Legionnaires' disease bacteria (Legionella pneumophila) in human monocytes is reversibly inhibited by erythromycin and rifampin. Journal of Clinical Investigation 71, Jacobs, M. R. (1991). Evaluation of the bactericidal activity of temafloxacin. American Journal of Medicine 91, Suppl. 6A, 31S-4. Kitsukawa, K., Hara, J. & Saito, A. (1991). Inhibition of Legionella pneumophila in guinea pig peritoneal macrophages by new quinolone, macrolide and other antimicrobial agents Journal of Antimicrobial Chemotherapy 27, Leroy, O., Beuscart, C, Chidiac, C, Sivery, B., Senneville, E., Vincent du Laurier, M. et al. (1989). Traitement des pneumonies dues aux legionelles, mycoplasmes, Chlamydiae, et rickettsies par l'ofloxacine. Pathologie Biologie 37, 1137^40. Mouton, Y., Leroy, O., Beuscart, C, Sivery, B., Senneville, E., Chidiac, C. et al. (1990). Efficacy of intravenous ofloxacin: a French multicentre trial in 185 patients. Journal of Antimicrobial Chemotherapy 26, Suppl. D, Rajagopalan-Levasseur, P., Dournon, E., Dameron, G., Vilde, J. L. & Pocidalo, J. J. (1990). Comparative postantibacterial activities of perfloxacin, ciprofloxacin, and ofloxacin against intracellular multiplication of Legionella pneumophila serogroup 1. Antimicrobial Agents and Chemotherapy 34, Saito, A., Sawatari, K.., Fukuda, Y., Nagasawa, M., Koga, H., Tomonaga, A. et al. (1985). Susceptibility of Legionetla pneumophila to ofloxacin in vitro and in experimental Legionella pneumonia in guinea pigs. Antimicrobial Agents and Chemotherapy 28, Vilde, J. L., Dournon, E. & Rajagopalan, P. (1986). Inhibition of Legionella pneumophila multiplication within human macrophages by antimicrobial agents. Antimicrobial Agents and Chemotherapy 30, Wynckel, A., Toupance, O., Melin, J. P., David, C, Lavaud, S., Wong, T. et al. (1991). Traitement des legionelloses par ofloxacine chez le transplante renal. Absence d'interference avec la ciclosporine A. Presse Medicate 20, (Received 27 June 1995; accepted 30 August 1995)