No abstract

Blood and urine concentrations of a new aminoglycoside, tobramycin, were compared with those of gentamicin after intravenous and intramuscular administration in healthy adult volunteers. Serum concentrations for the two antibiotics were the same after an intravenous infusion of 100 mg for the first haur (about 4.75 p.g per milliliter), during the steady state attained by infusing 30 mg per hour for the next 2 haurs (about 3.85 p.g per milliliter), and during the elimination phase after the antibiotics were stopped. After a single intramuscular injection of 100 mg, the blood level curves were equal and the peak concentrations at 20 to 45 minutes were the same as those obtained after intravenous infusion of the same dose for 1 hour. Between 80% and 90% of each of the two antibiotics was recovered in the urine within 24 hours. Plasma clearance was not significantly higher than renal clearance, and glomerular filtration seemed to be the main pathway of elimination. The two aminoglycosides had an apparent volume of distribution of about 30% total body weight. Each had a half‐life of 2 hours.

Ampicillin and amoxicillin (a-amino-p-hydroxybenzyl penicillin) were administered orally in 500-mg doses to eight fasting volunteers in a comparative study in which pharmacokinetic techniques were used. The absorption of amoxicillin was significantly better, as demonstrated by a higher mean peak serum concentration of 7.6 Ag/ml as compared to 3.2 ,ug/ml for ampicillin, an average "area under the curve"that was approximately double that of ampicillin, and an 8-hr urinary recovery for amoxicillin of 60% as compared to 34% for ampicillin. Serum half-lives were the same for the two antibiotics, with values of 60.3 (±3.3) min for ampicillin and 61.3 (±5.6) min for amoxicillin. The latter drug gave measurable concentrations in the blood at 8 hr in all of eight volunteers, as compared to only three of eight with ampicillin.Amoxicillin (BRL-2333, a-amino-p-hydroxybenzyl penicillin) is a semisynthetic penicillin that is comparable to ampicillin in antibacterial spectrum and in vitro activity but yields higher concentrations in the blood after equivalent oral doses (3,(16)(17)(18). Neu and Winsheil reported an average peak concentration in serum of 7.6 ,g/ml after oral administration of 500 mg of amoxicillin to volunteers, as compared to 3.8 Aug/ml for ampicillin (17). Considerably higher peak blood concentrations, 10.8 ,g/ml for amoxicillin and 6.3 ,ug/ml for ampicillin, were found with the same doses in another study (3). Further, only a 20% difference in the urinary recovery of the two agents was noted in one of the studies, as compared with 37% in the other (3,17). Peak serum concentrations of ampicillin attained after 500 mg orally as reported in the literature have ranged from 1.5 to 6.3 jAg/ml (2, 3,8,[13][14][15]17). Because of these inconsistencies and the desire to get more complete information regarding the precise pharmacokinetics of these two ampicillins, including the comparative half-lives, the present study was carried out. MATERIALS AND METHODSEight healthy adult male volunteers received two 250-mg capsules of amoxicillin and of commercially available ampicillin (Beecham Pharmaceuticals) in crossover fashion, with an interval of 7 days or more between the two parts of the study. The volunteers reported to the laboratory after an overnight fast and were instructed to empty their bladders. They were then given the antibiotic capsules with 100 ml of water, and were allowed no food during the first 3 hr of the experiment. Blood samples taken at 0, 0.5, 1, 1.5, 2, 3, 4, 5, 6, 7, and 8 hr, and samples of urine collected between 0 to 4 and 4 to 8 hr, were assayed for antibiotic content by use of a previously described agar-well diffusion technique, with Bacillus stubtilis as the test organism (1). Appropriate standards were prepared in pooled human serum and urine on the morning of each study, with the use of antibiotic powders provided by Beecham Research Laboratories. Specimens and standards were frozen at -20 C, and all assays were carried out within a few days of collection of the samples.The serum levels of...

temperature by use of an ultrafiltration method previously described (3). The concentrations of antibiotics used were those commonly found in the blood during therapy: 5 ,g/ml for gentamicin and tobramycin, S and 15 Ajg/ml for kanamycin, and 15 ,ug/ml for streptomycin. Pooled serum from healthy donors containing the antibiotic under study was adjusted to a pH of 7.4 to 7.5 by bubbling 5.0% CO2 through it. Then 15.4 ml of this 97% serum solution was put in the upper of two glass chambers, which were clamped tightly with a single layer of cellophane membrane (Union Carbide Corp., Dialysis Membrane 1-7u) between them. Ultrafiltration was carried out in an incubator at 37 C by the application of a vacuum to the lower chamber, and yielded 0.8 ml of filtrate within 45 min. By removing 1.0 ml from the upper chamber when 0.4 ml of filtrate had been collected, a midpoint specimen was available for antibiotic assay which corresponded to 100% serum because of the loss of water during filtration. The concentrations of antibiotic in this midpoint serum specimen, and in the protein-free ultrafiltrate, were determined by use of an agar well assay method with Bacillus subtilis ATCC 6633 as the test organism (2). Appropriate standard curves for the microbiological assay were prepared by dissolving known amounts of the antibiotic standards in fresh, pooled human serum and in antibiotic-free filtrate obtained by filtration of pooled serum. Impermeability of the cellophane membrane as a cause of apparent protein binding was investigated by ultrafiltration of buffered saline solutions of the antibiotics (6). The antibiotic standards used in these studies and their suppliers were gentamicin (Schering Corp.), tobramycin (Eli Lilly & Co.), kanamycin (Bristol Laboratories), and streptomycin (Canalco). RESULTSPercent binding was calculated by dividing the difference between the antibiotic concentrations in the serum and filtrate specimens by the serum concentration and multiplying by 100 (3). The numbers of separate ultrafiltrations performed and averaged were 9 for gentamicin, 8 for tobramycin, 12 for kanamycin at 5 Ag/ml, 6 for kanamycin at 15 ,ug/ml, and 6 for streptomycin. These ultrafiltration experiments with gentamicin and tobramycin in 100% human serum demonstrated no evidence of protein binding, with mean values of -2.0 and -2.1%, respectively (Table 1). Kanamycin experiments yielded values of +2.8% at 5,ug/ml and -0.7% at 15,ug/ml, whereas streptomycin was 35.4% bound at a serum concentration of 15,ug/ml. Membrane trapping as a cause of apparent protein binding was found to be negligible (retention <3.