The trinem sanfetrinem (GV 104326) was administered as the oral hexetil prodrug GV 118819X in two dose levels to six healthy volunteers. A single dose equivalent to 125 mg of sanfetrinem was administered, followed 6 weeks later by a single dose equivalent to 500 mg of sanfetrinem. The concentrations of the drug in plasma, cantharidin-induced inflammatory fluid, and urine were measured with a microbiological assay. The stability of sanfetrinem was studied in serum and inflammatory fluid. The mean peak concentrations in plasma of 0.77 and 2.47 microg/ml were attained at 1.1 and 2.0 h after the 125- and 500-mg doses, respectively. Mean peak concentrations in inflammatory exudate of 0.26 and 0.86 microg/ml were attained at 2.80 and 2.67 h after the 125- and 500-mg doses, respectively. The mean terminal elimination half-lives in plasma were 1.33 and 1.97 h for the 125- and 500-mg doses, respectively. The half-lives in the inflammatory fluid were 1.66 and 1.74 h for the 125- and 500-mg doses, respectively. The overall penetration of the drug into the inflammatory fluid was 51.4 and 47.0% for the 125- and 500-mg doses, respectively. Mean urine recovery was greater following 500 mg (24.15%) than after 125 mg (18.4%) of sanfetrinem. Sanfetrinem was relatively unstable in the inflammatory exudate in vitro (half-life, 5.5 h), and this could explain the poor penetration of the drug in the inflammatory exudate observed in this study.

Time-kill kinetics of BAY 12-8039 were studied at two inocula against three strains each of Bacteroides fragilis, Escherichia coli, Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pyogenes. The postantibiotic effects of BAY 12-8039 were studied on three strains each of E. coli, S. aureus, H. influenzae, Streptococcus pyogenes, and Streptococcus pneumoniae. The pharmacodynamic data demonstrated that BAY 12-8039 has marked activity against gram-positive and gram-negative organisms (under both anaerobic and aerobic conditions) and anaerobes. BAY 12-8039 also exhibited a postantibiotic effect of >1 h for all strains except one E. coli strain.

The in vitro activity of BAY 12-8039, a new fluoroquinolone, was studied in comparison with those of ciprofloxacin, trovafloxacin (CP 99,219), cefpodoxime, and amoxicillin-clavulanate against gram-negative, gram-positive, and anaerobic bacteria. Its activity against mycobacteria and chlamydia was also investigated. BAY 12-8039 was active against members of the family Enterobacteriaceae (MIC at which 90% of strains tested were inhibited [MIC90S] < or = 1 microgram/ml, except for Serratia spp. MIC90 2 microgram/ml), Neisseria spp. (MIC90S, 0.015 microgram/ml), Haemophilus influenzae (MIC90, 0.03 microgram/ml), and Moraxella catarrhalis (MIC90, 0.12 micrgram/ml), and these results were comparable to those obtained for ciprofloxacin and trovafloxacin. Against Pseudomonas aeruginosa, the quinolones were more active than the beta-lactam agents but BAY 12-8039 was less active than ciprofloxacin. Strains of Stenotrophomonas maltophilia were fourfold more susceptible to BAY 12-8039 and trovafloxacin (MIC90S, 2 micrograms/ml) than to ciprofloxacin. BAY 12-8039 was as active as trovafloxacin but more active than ciprofloxacin against Streptococcus pneumoniae (MIC90, 0.25 microgram/ml) and methicillin-susceptible Staphylococcus auerus (MIC90S, 0.12 micrograms/ml). The activity of BAY 12-8039 against methicillin-resistant S. aureus (MIC90, 2 micrograms/ml) was lower than that against methicillin-susceptible strains. BAY 12-8039 was active against anaerobes (MIC90S < or = 2 micrograms/ml), being three- to fourfold more active against Bacteroides fragilis, Prevotella spp., and Clostridium difficile than was ciprofloxacin. Against Mycobacterium tuberculosis, BAY 12-8039 exhibited activity comparable to that of rifampin (MICs < or = 0.5 micrograms/ml). Against Chlamydia trachomatis and Chlamydia pneumoniae BAY 12-8039 was more active (MICs < or = 0.12 microgram/ml) than either ciprofloxacin or erythromycin and exhibited a greater lethal effect than either to these two agents. The protein binding of BAY 12-8039 was determined at 1 and 5 micrograms/ml as 30 and 26.4%, respectively. The presence of human serum (at 20 or 70%) had no marked effect on the in vitro activity of BAY 12-8039.

GV104326 is a novel tricyclic beta-lactam (a trinem or, formerly, tribactam). The in vitro activity of GV104326 was compared with those of cefuroxime, cefixime, amoxicillin, amoxicillin-clavulanic acid, cefpirome, and ciprofloxacin. GV104326 had in vitro activity generally similar to that of cefixime against members of the family Enterobacteriaceae (MIC at which 90% of the isolates are inhibited [MIC90], < or = 2 micrograms/ml), with cefuroxime and amoxicillin-clavulanic acid being 8- to 32-fold less active and with cefpirome being 4- to 8-fold more active against members of this family. The trinem had no activity against Pseudomonas aeruginosa or Stenotrophomonas maltophilia (MIC90, > 128 micrograms/ml) but was the most active agent against Acinetobacter calcoaceticus. GV104326 was particularly active against gram-positive cocci. Ninety percent of methicillin-susceptible Staphylococcus aureus strains were susceptible to 0.03 microgram of GV104326 per ml, making it the most active agent studied. Enterococci and Lancefield group A and B streptococci were generally equally or somewhat more susceptible to GV104326 than they were to amoxicillin. Streptococcus pneumoniae strains were highly susceptible to GV104326, and those strains which showed decreased susceptibility to penicillin were generally twofold more susceptible to the trinem than to amoxicillin. Haemophilus influenzae and Moraxella catarrhalis were highly susceptible to GV104326 (MIC90s, 0.12 and 0.03 microgram/ml, respectively). The anaerobes Clostridium perfringens, Bacteroides fragilis, and Peptostreptococcus spp. were more susceptible to the trinems (formerly tribactams) than to the other agents studied.

A single 200-mg oral dose of trovafloxacin (CP-99,219) was given to each of eight healthy male volunteers, and the concentrations of the drug were measured in plasma, cantharides-induced inflammatory fluid, and urine over the subsequent 36 h. The mean maximum concentration observed in plasma was 2.9 micrograms/ml at a mean time of 0.75 h postdose. The mean maximum concentration observed in inflammatory fluid was 1.2 micrograms/ml at 4.0 h postdose. The mean elimination half-life in plasma was 7.8 h. The overall penetration into inflammatory fluid was 64%, as assessed by determining the ratio of the area under the concentration-time curves. Recovery of the dose in urine within the first 36 h postdose was 5.0% of the administered dose. Our results indicate that trovafloxacin, at a dosage of 200 mg once or twice daily, should be adequate for the treatment of systemic infections caused by most common bacterial pathogens.

Two levofloxacin administration regimens were used for six healthy male volunteers. They received either 500 mg of levofloxacin orally every 12 h for five doses or 500 mg every 24 h for three doses, and then 6 weeks later they received the other course. The concentrations of the drug in plasma, cantharidin-induced inflammatory fluid, and urine were measured with a microbiological assay following administration of the final dose. Mean peak concentrations in plasma of 9.3 and 6.6 micrograms/ml were attained 1.1 and 1.2 h after the 12- and 24-h regimens, respectively. Mean peak concentrations is inflammatory fluid of 6.8 and 4.3 micrograms/ml were attained at 2.3 and 3.7 h, respectively. The average steady-state concentrations were 5.0 and 2.2 micrograms/ml in plasma and 4.7 and 2.3 micrograms/ml in inflammatory fluid, respectively. The mean terminal elimination half-lives for plasma were 7.9 and 8.0 h for the two regimens, respectively, and the same values were noted for inflammatory fluid. The overall penetration into inflammatory fluid ranged from 88 to 101% with the 12-h regimen and 83 to 112% with the 24-h regimen. Mean urinary recoveries were 87 and 86% over the corresponding interval of the 12- and 24-h regimens, respectively. These results suggest that administration of levofloxacin once and twice daily should be efficacious for infections caused by the majority of pathogens.

A single 400-mg oral dose of grepafloxacin (OPC-17116) was given to each of six healthy male volunteers, and the concentrations of the drug in plasma, cantharides-induced inflammatory fluid, and urine were measured over the subsequent 12 h. The mean peak concentration in plasma of 1.5 micrograms/ml was attained at a mean time of 2.0 h postdose. The mean peak concentration in inflammatory fluid of 1.1 micrograms/ml was attained at a mean time of 4.8 h postdose. The mean elimination half-life in plasma was 5.2 h, and that in inflammatory fluid was 12.7 h. The overall penetration into inflammatory fluid was 180.6% (or 133% if one aberrant result from one volunteer is excluded). Recovery of the drug in urine during the first 24 h postdose was 8.3% of the administered dose. Our results indicate that a once- or twice-daily dosage of grepafloxacin should be adequate to treat systemic infections caused by most bacterial pathogens.

A single 1-g or 2-g intravenous dose of the cephalosporin FK-037 was given over 30 min in a cross-over-designed study, to each of six healthy male volunteers, and the concentrations of the drug were measured in plasma and cantharides-induced blister fluid over the subsequent 12 h. Urine was collected over 24 h. After a washout period of 6 weeks, during which the blisters healed, the study was repeated at the other dose level. Following the 1-g dose, the mean peak concentration in plasma was 83.8 micrograms/ml, and after the 2-g dose it was 142.6 micrograms/ml. The mean peak concentrations in the inflammatory fluid were 37.9 and 63.3 micrograms/ml, respectively. The mean elimination half-lives from plasma and inflammatory fluid were 2.0 and 2.5 h, respectively, after 1 g and 2.0 h and 3.7 h, respectively, after 2 g. The amounts of penetration into inflammatory fluid (as assessed by ratios of areas under the concentration-time curves) were 109.9 and 110.5% following doses of 1 and 2 g, respectively. The proportions of the administered drug recovered in the urine by 24 h were 87.6 and 85.7%, respectively. Our results indicate that FK-037 should prove to be efficacious in the treatment of a wide range of systemic infections.

The in vitro activities of two glycylcyclines, CL 329,998 and CL 331,002 (two new semisynthetic tetracyclines), were evaluated in comparison with those of tetracycline and other available oral antimicrobial agents. A total of 523 recent clinical isolates were studied, including strains resistant to tetracycline. Members of the family Enterobacteriaceae were generally > or = 16-fold more susceptible to the glycylcyclines than to tetracycline (although less difference was seen with Proteus spp.). Pseudomonas aeruginosa was modestly susceptible to both new compounds (MIC for 90% of strains tested [MIC90], 16 micrograms/ml). Tetracycline- and methicillin-susceptible and -resistant strains of Staphylococcus aureus were all susceptible to the glycylcyclines (MIC90 < or = 1 microgram/ml). Streptococci (including Streptococcus pneumoniae) and Enterococcus faecalis and Enterococcus faecium displayed a bimodal distribution of susceptibility to tetracycline yet were uniformly susceptible to the glycylcyclines (MIC90 < or = 0.25 microgram/ml). The glycylcyclines were highly potent against Neisseria, Moraxella, Haemophilus, and Bacteroides spp. (MIC90 < or = 0.5 microgram/ml). Strains of Chlamydia spp. (three C. trachomatis strains and one C. pneumoniae strain) were inhibited by < or = 0.25 microgram of CL 329,998 or CL 331,002 per ml. Two strains of Mycoplasma pneumoniae were inhibited by < or = 0.12 microgram of CL 331,002 per ml and by 1 microgram of CL 329,998 per ml. Mycobacterium tuberculosis and Mycobacterium avium were resistant to the two glycylcyclines (MIC > or = 8 micrograms/ml). These results indicate that the two glycylcyclines have potent in vitro activities against a wide range of clinically important pathogenic bacteria.

A single 400-mg oral dose of sparfloxacin was given to each of six healthy male volunteers, and the concentrations of the drug were measured in plasma, cantharides-induced inflammatory fluid, and urine over the subsequent 52 h. The mean peak concentration in plasma of 1.6 micrograms/ml was attained at a mean time of 2.7 h postdose. The mean peak concentration in inflammatory fluid of 1.3 micrograms/ml was attained at a mean time of 5 h postdose. The mean elimination half-life in plasma was 17.6 h, and that in inflammatory fluid was 19.7 h. The overall penetration into inflammatory fluid was 117%. Urinary recovery within the first 52 h postdose was 8.8% of the administered dose. Our results indicate that a once-daily dosage of sparfloxacin should be adequate to treat systemic infections caused by most common bacterial pathogens.

The in vitro activity of L-627, a new parenterally administered carbapenem, was compared with those of imipenem, meropenem, FCE 22101 (a penem), ceftazidime, and ceftriaxone. L-627 was active against members of the family Enterobacteriaceae (MIC for 90% of strains tested [MIC90] ranging from 0.03 to 4 micrograms/ml). L-627 displayed activity equal to that of meropenem against Pseudomonas aeruginosa (MIC90, 2 micrograms/ml), although, as with other carbapenems, the antipseudomonal activity was reduced against D2-deficient strains. Staphylococci and streptococci were susceptible (MIC90 of 1.0 micrograms/ml for Staphylococcus aureus and 0.015 micrograms/ml for group A streptococci). L-627 also had activity against anaerobic bacteria (MIC90, 2.0 micrograms/ml for Bacteroides fragilis). Neisseria gonorrhoeae and Neisseria meningitidis were highly susceptible (MIC90, 0.06 micrograms/ml), and against the common respiratory pathogens (Haemophilus influenzae, Streptococcus pneumoniae, and Moraxella catarrhalis), the MIC90s were less than or equal to 2.0 micrograms/ml. The protein binding of L-627 ranged from 13.8 to 22%, depending on the concentration. The presence of human serum had little effect on the MIC or MBC of L-627. These results suggest that L-627 merits further study in the treatment of infections caused by a wide range of pathogens.

Eight healthy volunteers received a 1,000-mg single oral dose of 2085P which consisted of 800 mg of pivampicillin and 200 mg of brobactam. Concentrations of ampicillin and brobactam in plasma, inflammatory fluid, and urine were measured over the subsequent 24 h. Pivampicillin and brobactam were moderately rapidly absorbed. The mean (standard deviation) maximum concentration in plasma (Cmax) of ampicillin was 8.2 (1.9) micrograms/ml, and that of brobactam was 2.1 (2.0) micrograms/ml at mean times of 1.9 (0.5) and 2.3 (0.8) h, respectively. The elimination half-lives in plasma were 1.8 (0.5) and 1.6 (2.0) h, respectively. Both agents penetrated the experimentally induced inflammatory fluid, reaching a mean maximum at 3 h. The Cmax of ampicillin was 6.8 (2.3) micrograms/ml, and that of brobactam was 1.0 (0.4) micrograms/ml. The penetration (derived by comparing the area under the concentration-time curve from 0 h to infinity for inflammatory fluid with that for plasma) was 97.3% (26.0%) for ampicillin and 81% (22.3%) for brobactam. The 24-h urinary recovery was 54.2% (16.6%) of the administered dose for ampicillin and 40.2% (11.4%) for brobactam. These data suggest that this combination of beta-lactam and inhibitor should be efficacious in treating infections caused by ampicillin-resistant pathogens.

The pharmacokinetics of tazobactam (500 mg) administered intravenously alone were compared with the pharmacokinetics of tazobactam coadministered with piperacillin (4 g), and the penetration into an inflammatory exudate in six healthy males was studied. Piperacillin influenced the pharmacokinetics of tazobactam. The mean levels of tazobactam in plasma at 4 h were 0.6 microgram/ml when it was given alone and 1.2 micrograms/ml when it was given with piperacillin (P = 0.0003). The mean total clearances of tazobactam were 203.5 and 134.2 ml/min (P = 0.035) when it was given alone and with piperacillin, respectively There were no significant differences in the elimination half lives, areas under the concentration-time curve from 0 h to infinity, or volumes of distribution. Inflammatory exudate penetration was rapid, and the mean maximum levels of tazobactam attained were 6.4 and 11.3 micrograms/ml when it was given alone or with piperacillin, respectively (P less than 0.06). The mean percent penetration of tazobactam and the area under the concentration-time curve from 0 h to infinity in inflammatory exudate were greater when tazobactam was given with piperacillin. The mean 24-h urinary recoveries of tazobactam were 63.7% +/- 7.9% when it was given alone and 56.8% +/- 2.7% when it was given with piperacillin. The explanation for the differences in the pharmacokinetics of tazobactam when it was administered alone compared with those when it was given with piperacillin was unclear.

The in vitro activity of GR69153, a new catechol-substituted cephalosporin, was compared with those of ceftazidime, imipenem, meropenem, and ceftriaxone against 604 recent clinical isolates and other strains with known mechanisms of resistance. The MICs of GR69153 for 90% of the members of the family Enterobacteriaceae tested were less than 0.5 micrograms/ml, with the exceptions of those for Serratia spp. (4 micrograms/ml), Citrobacter spp. (2 micrograms/ml), and Enterobacter spp. (8 micrograms/ml). Ninety percent of Pseudomonas aeruginosa isolates were susceptible to less than or equal to 1 microgram of GR69153 per ml. With the exception of methicillin-resistant strains, 90% of Staphylococcus aureus isolates were susceptible to less than or equal to 2 micrograms/ml, and GR69153 was four- to eightfold more active than ceftazidime and ceftriaxone against these strains. Isolates of Haemophilus influenzae, Branhamella catarrhalis, Neisseria spp., and Streptococcus pneumoniae (penicillin susceptible) were highly susceptible (MIC for 90% of the strains, less than or equal to 0.12 micrograms/ml). GR69153 was stable to hydrolysis by the TEM-1 and TEM-5, SHV-1 and SHV-2, and K1 beta-lactamases, but some susceptibility to hydrolysis by the TEM-3, TEM-9, and P99 enzymes was observed. The protein-binding activity of GR69153 was 74.5 to 66.8%, depending on the concentration, and serum had little effect upon activity.

The pharmacokinetics and penetration into a cantharidine-induced inflammatory exudate of meropenem was studied in six volunteers following a single 1-g intravenous dose. Concentrations in plasma, urine, and the inflammatory exudate were determined by a microbiological assay. The mean elimination half-life of meropenem in plasma was 1.1 h, with the concentration in plasma declining from a mean of 23.6 micrograms/ml at 1 h to 0.7 micrograms/ml at 6 h. The inflammatory fluid penetration was rapid (time to maximum concentration of drug in serum, 0.75 h), and the penetration was 111%. The recovery of meropenem in urine at 24 h was 65.4% of the administered dose.

The pharmacokinetics of the cephalosporin ceftibuten were determined after the fifth and final dose of 200 mg given every 12 h. Concentrations in plasma and cantharidin-induced inflammatory fluid were determined by a microbiological assay. Samples for three volunteers were assayed by a high-performance liquid chromatographic procedure to determine levels for both cis and trans ceftibuten. The mean peak level of ceftibuten in serum was 10.9 micrograms/ml at a mean time of 1.8 h after administration, and the mean elimination half-life from plasma was 2.5 h. Penetration into the inflammatory fluid was good, the mean peak level being 9.2 micrograms/ml at a mean time of 3.7 h. The mean percent penetration into the inflammatory fluid was 113.4%. High-performance liquid chromatography analysis showed that the mean peak level of the trans isomer was 5.7% that of the cis isomer. This study suggests that twice-daily doses of ceftibuten should be sufficient to treat urinary or systemic infections caused by susceptible pathogens.

The bronchial mucosal concentrations of lomefloxacin were determined for specimens obtained by fiber-optic bronchoscopy and compared with simultaneous concentrations in serum. The 23 patients studied were given an oral dose of 400 mg once daily for 4 days to achieve steady-state levels. The median concentration in serum was 2.5 micrograms/ml (range, 1.0 to 5 micrograms/ml), and the median bronchial mucosal concentration was 5.0 micrograms/g (range, 0.7 to 18.6 micrograms/g). The median percent penetration was 177% (range, 69 to 541%). The concentrations in serum and mucosa exceeded the MIC for 90% of strains of organisms causing bronchial infections but not sufficiently to recommend lomefloxacin for the routine treatment of pneumococcal infections.

The in vitro activity of Bay v 3522, a new aminobenzothiazol cephem, was compared with those of other oral beta-lactams. Bay v 3522 displayed high activity against Staphylococcus spp. (MICs for 90% of strains tested [MIC90S], 0.5 micrograms/ml), Streptococcus pneumoniae (MIC90, 0.06 micrograms/ml), and Haemophilus influenzae and Branhamella catarrhalis (MIC90S, 2 micrograms/ml). There was limited activity against members of the family Enterobacteriaceae, with the MIC90S being between 4 and greater than 128 micrograms/ml. The stability of Bay v 3522 to hydrolysis by the SHV-1 and TEM-1 enzymes was intermediate to those of cephalexin (least hydrolyzed) and cefaclor, but it was markedly more stable than amoxicillin. There was high affinity to the chromosomally mediated P99 enzyme. The protein binding of Bay v 3522 was 45%. The primary target of Bay v 3522 was penicillin-binding protein 3.

The pharmacokinetics of the quinolone lomefloxacin were determined following a single 400-mg oral dose given to each of six male volunteers. Concentrations in serum, urine, and cantharidin-induced inflammatory fluid were determined by a microbiological assay. Samples from two volunteers were also assayed by high-performance liquid chromatography. The mean peak level in serum, 4.7 micrograms/ml, was attained within 1 h of administration. The mean elimination half-life from serum was 7 h. Inflammatory fluid was penetrated rapidly, with a mean peak level of 3.5 micrograms/ml occurring after 2.7 h. The mean recovery of lomefloxacin from urine over 48 h was 76% of the administered dose. There was a minor peak on the high-performance liquid chromatography trace, suggesting a small amount of unidentified metabolite. This was present only in urine; no detectable metabolites were found in serum. This study suggests that either once-daily or twice-daily dosage of lomefloxacin should be sufficient to treat urinary or systemic infections, respectively, caused by susceptible pathogens.

The in vitro activity of PD 127,391, a dihalogenated quinolone, was compared with those of ofloxacin, ciprofloxacin, nalidixic acid, gentamicin, and cefuroxime against 525 recent isolates and well-characterized antimicrobial agent-resistant strains. The MICs of PD 127,391 against 90% of members of the family Enterobacteriaceae, Bacteroides fragilis, Haemophilus influenzae, Neisseria sp., and Streptococcus pneumoniae were less than or equal to 0.12 microgram/ml. Some 90% of Pseudomonas aeruginosa and staphylococci were susceptible to 0.25 micrograms of PD 127,391 per ml. Against most strains, PD 127,391 was 2- to 8-fold more active than ciprofloxacin, but it was 64-fold more active than ciprofloxacin against B. fragilis. Strains of members of the family Enterobacteriaceae which were resistant to nalidixic acid were less susceptible to all of the quinolones tested, including PD 127,391. The MIC and minimum lethal concentration of PD 127,391 against three strains of Chlamydia trachomatis were each 0.06 microgram/ml, and the MIC against 90% of 21 strains of Mycobacterium tuberculosis was 1 microgram/ml. PD 127,391 was less active at pH 5, its maximal activity being at pH 7 to 8. The presence of urine at pH 5.9 decreased the bactericidal activity. The protein binding of PD 127,391 was 2 to 7%, and serum had little effect on activity.

The in vitro activity of lomefloxacin (SC-47111; NY-198), a new difluorinated quinolone, was compared with those of ofloxacin, ciprofloxacin, fleroxacin, amoxicillin, cefuroxime, and trimethoprim against 585 recent clinical isolates and other strains with known mechanisms of resistance. The MICs of lomefloxacin against 90% of the members of the family Enterobacteriaceae, Pseudomonas aeruginosa, and staphylococci were between 0.25 and 4 micrograms/ml. Ninety percent of Neisseria sp. and Haemophilus influenzae were susceptible to less than or equal to 0.06 micrograms/ml, and streptococci (including Streptococcus pyogenes, Streptococcus pneumoniae, and enterococci) and Bacteroides fragilis were susceptible to 8 micrograms/ml. Lomefloxacin was comparable in activity to fleroxacin and ofloxacin, but it was less active than ciprofloxacin. There was cross-resistance between the quinolone group of antimicrobial agents. The protein binding of lomefloxacin was 15.4%, and serum had little effect on the activity of the compound. However, urine at pH 5.0 decreased the activity by two- to eightfold compared with that at pH 7.0

The pharmacokinetics of the macrolide roxithromycin (RU 28965) were studied during and after administration of 150 mg every 12 h for 3 days (five doses) in six volunteers. The concentrations in serum, blister fluid, and urine were measured. Mean levels in serum taken at 1.5 h after the morning dose increased from 4.4 micrograms/ml on day 1 to 5.9 micrograms/ml on day 2 and 7.4 micrograms/ml on day 3. The mean serum and blister fluid elimination half-lives on day 3 were 13.2 and 12.5 h, respectively. Roxithromycin penetrated blister fluid well; the mean percent penetration (as measured by the ratio of areas under the curve) was 85%. After the final dose, a mean of 10.5% of that dose was recovered in 12 h as microbiologically active compound.

The in vitro activity of CGP 31608, a semisynthetic penem derivative, was compared with that of Sch 34343, imipenem, cefoxitin, cefuroxime, and ceftazidime and other beta-lactams, when appropriate, against 628 recent isolates and other beta-lactam-resistant strains. The MICs of CGP 31608 against 90% of the members of the family Enterobacteriaceae, Pseudomonas aeruginosa, Haemophilus influenzae, Neisseria spp., Bacteroides spp., Clostridium spp., staphylococci, and Streptococcus pneumoniae were between 0.25 and 8 micrograms/ml. The susceptibility of beta-lactamase-producing strains and known porin mutants of the Enterobacteriaceae suggests that CGP 31608 is resistant to many important beta-lactamases (including the mutationally derepressed chromosomal enzymes) and is not excluded from the bacterial cell in strains expressing these known porin mutations. Generally, CGP 31608 was less active than imipenem, Sch 34343, and the cephalosporins, except against Pseudomonas aeruginosa. The activity of CGP 31608 against Staphylococcus aureus (including methicillin-resistant strains) was greater than that of the cephalosporins. The major target site in Escherichia coli K-12 for CGP 31608 was penicillin-binding protein 2. The serum protein binding of 5 micrograms of CGP 31608 per ml was 14%, and serum had little effect on activity.

Ro 23-6240 is a new difluorinated quinolone antimicrobial agent. Its in vitro activity against a wide range of bacteria was compared with those of other quinolones and beta-lactams. Generally, members of the family Enterobacteriaceae were inhibited by low concentrations of Ro 23-6240 (MIC90 [MIC for 90% of isolates tested], less than or equal to 1 microgram/ml). Ninety percent of Staphylococcus aureus (including methicillin-resistant strains) and Neisseria gonorrhoeae isolates were inhibited by 0.5 microgram/ml. Pseudomonas aeruginosa (MIC90, 2 micrograms/ml) and Bacteroides fragilis (MIC90, 4 micrograms/ml) showed intermediate susceptibility, and Streptococcus pneumoniae (MIC90, 8 micrograms/ml) was less susceptible. Strains resistant to nalidixic acid were less susceptible to all the quinolones tested. The protein binding of Ro 23-6240 (5 micrograms/ml) was 27%.

The activity of two iminomethoxy aminothiazoly cephalosporins, Ro 15-8074 and Ro 19-5247, was compared with that of other beta-lactams against a total of 491 bacterial strains. Both were highly active (MIC for 90% of the strains tested [MIC 90], less than or equal to 2 micrograms/ml) against the majority of the members of the family Enterobacteriaceae, Haemophilus influenzae, Neisseria spp., and Streptococcus pneumoniae, being at least 16-fold more active than cephalexin and 8-fold more active than cefuroxime. There was no activity against Pseudomonas aeruginosa and poor activity against Morganella morganii (in the case of Ro 15-8074), Enterobacter sp., and Citrobacter sp. Staphylococcus aureus was moderately susceptible to Ro 19-5247 (MIC90, 8 micrograms/ml), but Ro 15-8074 was eightfold less active. The protein binding of the two compounds at 5 micrograms/ml was 9.1% for Ro 15-8074 and 69.9% for Ro 19-5247. The major target site for the two cephalosporins was PBP 3.