Enantioselective disposition of 2-arylpropionic acid nonsteroidal anti-inflammatory drugs. III. Fenoprofen disposition

The disposition of fenoprofen enantiomers has been studied in nine healthy rabbits. A mean (S.E.M.) of 0.73 (0.07) of R-fenoprofen was inverted to S-fenoprofen and the distribution volumes for bound plus unbound R-fenoprofen and S-fenoprofen were 50.3 (4.5) and 98.5 (5.9) ml/kg, respectively. A model was developed which predicted the area under the S-fenoprofen plasma concentration-time curve after bolus administration of racemic fenoprofen. The mean (S.E.M.) predicted area, 2.1 (0.2) mg X min/ml/kg, was within 94% of the observed area 2.2 (0.2) mg X min/ml/kg. The effect of phenobarbital on the disposition of fenoprofen enantiomers was examined in an additional eight rabbits. During the control study the glucuronidation of R-fenoprofen exceeded the corresponding clearance term for the S-enantiomer by 2.1-fold. The clearances of individual enantiomers to their respective glucuronides increased after phenobarbital pretreatment by a mean 1.6-fold for R- and 2.3-fold for S-fenoprofen. The clearance of S-fenoprofen by processes other than glucuronidation and elimination of unchanged drug in urine was increased by a mean of 2.1-fold after phenobarbital pretreatment but the fractional inversion and the inversion clearance of R- to S-fenoprofen were not affected. These data indicate that on racemic fenoprofen administration the area under the curve for the pharmacologically active S-enantiomer would be reduced by phenobarbital pretreatment.     

Enantioselective disposition of 2-arylpropionic acid nonsteroidal anti-inflammatory drugs. IV. Ketoprofen disposition

The disposition of ketoprofen enantiomers has been studied in 12 rabbits with normal renal function (control) and in 6 of these rabbits with renal dysfunction. In control animals a mean (S.E.M.) of 0.09 (0.01) of R-ketoprofen was inverted to its S-enantiomer. The mean distribution volumes for R- and S-ketoprofen were 114 (7.4) and 294 (76) ml/kg, respectively. The mean clearance of the R-enantiomer from plasma was 292 (40) ml/kg/hr compared to 36 (3.2) ml/kg/hr for the S-enantiomer. This high degree of enantioselectivity was not due to inversion, renal clearance or clearance by glucuronidation, but was the result of an unknown clearance process(es). Comparing the mean parameters of ketoprofen disposition in renal dysfunction with those in the same animals during the control study, the mean fraction of the R-dose inverted increased by 290%, but this effect was not due to decreased clearance of R-ketoprofen renally or by glucuronidation. The distribution volumes of R- and S-ketoprofen were increased by 200 and 183% by renal dysfunction, consistent with an implied decrease in plasma protein binding. The clearances of bound plus unbound R- and S-ketoprofen from plasma remained unaltered by renal dysfunction.     

Oral disposition kinetics of ofloxacin in patients with compensated liver cirrhosis.

The disposition kinetics of ofloxacin, a quinolone antibacterial agent excreted essentially unmodified by the kidney, was studied after single oral administration in 8 patients with compensated liver cirrhosis and in 8 control subjects. Mean elimination half-life and apparent volume of distribution were significantly increased in the cirrhotic group (7.6 vs. 4.9 h and 1.6 vs. 1.2 liters kg-1, respectively). A reduction in the renal clearance of ofloxacin was also observed in the cirrhotic patients, in spite of an apparently normal renal function. These observations indicate that also the pharmacokinetics of unmetabolized drugs may be altered in compensated liver cirrhosis. The serum concentration-time profiles of nearly all subjects exhibited a secondary peak 4-6 h after dosing. This double-peak behavior was interpreted as either enterohepatic circulation or biphasic gastric emptying of ofloxacin.     

Age and propranolol stereoselective disposition in humans.

The apparent oral clearance of S(-)- and R(+)-propranolol as a function of age was evaluated in 53 healthy male volunteers (age range, 21 to 84 years) after a single 40 mg oral dose of the racemic mixture. No significant age-associated change in the total (bound plus unbound) and unbound S(-) and R(+) apparent oral clearance was observed (p greater than 0.05). Stereoselectivity in apparent oral clearance (both total and unbound) remained unaffected by advancing age (p greater than 0.05). The relationship between age and propranolol enantiomer plasma protein binding was also evaluated in 70 subjects, 53 of whom were from this study (age range, 21 to 89 years). Plasma free fractions for S(-)- and R(+)-propranolol were unchanged with increasing age (p greater than 0.05), even though the binding was stereoselective (plasma free fractions for R(+) greater than plasma free fractions for S(-); p less than 0.05). The findings from this relatively large and extensive study indicate that age does not influence the stereoselective disposition of propranolol.     

Enantioselective disposition of 2-arylpropionic acid nonsteroidal anti-inflammatory drugs. I. 2-Phenylpropionic acid disposition

The disposition of 2-phenylpropionic acid (2PPA) enantiomers has been studied in 12 rabbits with normal renal function (control) and in 6 rabbits with renal dysfunction. In control animals a mean (S.E.M.) of 0.23 (0.02) of R-2PPA was inverted to S-2PPA, and the distribution volumes for unbound S- and R-2PPA were 0.30 (0.02) and 0.62 (0.07) liters/kg, respectively, but there was no evidence of stereoselectivity in the glucuronidation of R- or S-2PPA. A model was developed that predicted the fraction of the 2PPA in plasma present as unbound S-2PPA upon infusion of racemic 2PPA to steady state. The mean (S.E.M.) observed fraction was 0.69 (0.01) and the predicted fraction was 0.66 (0.03). Comparing the mean parameters of 2PPA disposition in renal dysfunction with those in the same animals during a control study, the fraction of R-2PPA undergoing inversion increased by 350%, the clearance of unbound S-2PPA decreased to 25% and the unbound clearance of R-2PPA by processes other than inversion decreased to 9% of the control means. These changes are consistent with the effects of renal dysfunction on acyl-glucuronide futile cycles for both R- and S-2PPA. Using the above data it can be calculated that, in an animal with severe renal dysfunction, upon an infusion of racemic 2PPA to steady state the concentration of the unbound S-enantiomer would be 5.6 times that in a control animal.     

Metabolic disposition of prostacyclin in humans

Quantitative analysis of metabolite levels is a useful approach to investigation of the in vivo synthesis of short-lived mediators such as prostacyclin (PGI2). In order to establish the basis for metabolite assays of PGI2, we have studied the fate of radiolabeled PGI2 administered to man. Three healthy male volunteers each received i.v. 11 beta-[3H]PGI2 at 4 ng/kg/min for 24 hr. A gradual increase in plasma radioactivity was observed throughout the infusion period, followed by a biphasic decline postinfusion (T 1/2 alpha, 53 min; T 1/2 beta, 246 min) suggestive of the presence of long-lived metabolites of PGI2 in the circulation. The recovery of radioactivity averaged 82% in urine and, in contrast to other species, only 4% in feces. Direct analysis of urine by high-pressure liquid chromatography revealed the presence of at least 16 compounds and documented their relative abundance. Ten compounds were subsequently identified by gas chromatography-mass spectrometry. All identified metabolites retained the 6-keto-prostaglandin F structure characteristic of PGI2 hydrolysis and were each formed in less than 10% yield from administered PGI2. Of interest was the finding that 6-keto-prostaglandin F1 alpha accounted for 5.9% of systemically administered 11 beta-[3H]PGI2. These results identify urine as the major route of human PGI2 metabolite excretion and also illustrate the utility of direct chromotographic analysis of urine in the elucidation of prostaglandin disposition in humans.     

Plasma dihydroxyphenylglycol and the intraneuronal disposition of norepinephrine in humans.

We examined plasma levels of the sympathetic neurotransmitter norepinephrine (NE) and its deaminated metabolite dihydroxyphenylglycol (DHPG) during supine rest in healthy human subjects and in sympathectomized patients, during physiological (tilt) or pharmacological (yohimbine, clonidine) manipulations known to affect sympathetically mediated NE release, during blockade of neuronal uptake of NE (uptake-1) using desipramine, and during intravenous infusion of NE. Healthy subjects had a mean arteriovenous increment in plasma DHPG in the arm (10%, P less than 0.05), whereas sympathectomized patients had a mean arteriovenous decrement in DHPG in the affected limb (mean decrease 21%, P less than 0.05 compared with healthy subjects). Tilt and yohimbine, which stimulate, and clonidine, which inhibits, release of endogenous NE, produced highly correlated changes in plasma NE and DHPG (r = 0.94). Pretreatment with desipramine abolished DHPG responses to yohimbine while enhancing NE responses. To attain a given increase in plasma DHPG, about a tenfold larger increment in arterial NE was required during NE infusion than during release of endogenous NE. When plasma NE was markedly suppressed after administration of clonidine, plasma DHPG decreased to a plateau level of 700-800 pg/ml. The results indicate that (i) plasma DHPG in humans is derived mainly from sympathetic nerves; (ii) increments in plasma DHPG during stimulation of NE release result from uptake of NE into sympathetic nerve endings and subsequent intraneuronal conversion to DHPG; (iii) plasma DHPG under basal conditions probably is determined mainly by net leakage of NE into the axonal cytoplasm from storage vesicles; and (iv) increments in NE concentrations at neuronal uptake sites can be estimated by simultaneous measurements of DHPG and NE during NE infusion and NE release. Measurement of NE and DHPG provides unique clinical information about sympathetic function.     

Xylose disposition in humans as a function of age

D-xylose disposition was examined in 24 healthy men between 32 and 85 years of age. Xylose was administered as a 5 gm iv infusion and as a 25 gm po solution. Serum xylose concentrations and urinary excretion of intact xylose were determined. There were statistically significant inverse relationships with age for each of the following parameters after intravenous infusion: elimination rate constant (r2 = 0.71); systemic clearance (r2 = 0.66); renal clearance (r2 = 0.66); and nonrenal clearance (r2 = 0.35). Similar inverse relationships were found after oral dosing for the elimination rate constant (r2 = 0.69) and renal clearance (r2 = 0.54). There was no significant age relationship for the apparent volume of distribution or the steady-state volume of distribution. The percentage of the oral and intravenous dose recovered in urine up to 5 hours after dosing was significantly and inversely correlated with age. The implications of the latter finding are discussed with regard to the interpretation of the xylose tolerance test used to assess gastrointestinal absorptive capacity.     

Influence of seven beverages on salicylate disposition in humans

OBJECTIVE: Aspirin administered orally is one of most widely self-prescribed drugs to treat headaches or other pains. The aim of this study was to evaluate whether the influence of different beverages may be used to help in the ingestion of an aspirin tablet on the pharmacokinetic parameters of this drug. METHOD: This study was undertaken in five healthy volunteers. Seven beverages were tested: water, tea, coffee, orange juice, milk, beer and 40 degrees distilled alcohol. After plasma extraction, aspirin and salicylic acid were measured by HPLC with UV detection. The main pharmacokinetic parameters were determined by the compartimental method and drug disposition profiles by the Wagner-Nelson modified method. RESULTS: Elimination was not modified by any of the beverages but absorption was affected. Two opposing effects were observed: 40 degrees alcohol seemed to increase AUC and Cmax. Milk and beer seemed to decrease these parameters. With 40 degrees alcohol and tea, the amount absorbed and the disposition rate were higher. For milk and orange juice, the amount absorbed was lower and the disposition rate was unaffected. For beer, both the amount absorbed and the disposition rate increased. For coffee, both the amount absorbed and disposition rate were not significantly modified. CONCLUSION: The bioavailability of salicylates on the healthy volunteers in this study was significantly affected by concomitant administration of 40 degrees alcohol (spirit), beer and milk. The beverages seem to interfere with aspirin absorption and the drug disposition profile was modified.     

Effects of chronic phenobarbital on verapamil disposition in humans

Very little is known about the effects of hepatic enzyme induction with phenobarbital on the disposition of high clearance drugs in humans. Our study was undertaken to investigate the effect of phenobarbital on both alpha-1 acid glycoprotein concentrations and total and free verapamil and its metabolites. Single oral, single i.v., and multiple oral verapamil administrations were evaluated before and after 21 days of phenobarbital treatment in healthy caucasian male volunteers. Significant changes in the pharmacokinetics of total and free verapamil and its metabolites occurred in a predictable manner. Mean total apparent oral clearance after a single dose of verapamil was increased after phenobarbital treatment (75.1 +/- 49.2 vs. 376.2 +/- 221.8 ml/min/kg, P less than .05). Clearance of free drug increased to a similar magnitude. Mean total verapamil systemic clearance was increased (9.95 +/- 1.3 vs. 18.9 +/- 8.7 ml/min/kg, P less than .05); however free drug clearance was not altered. After multiple oral administration, total verapamil apparent oral clearance was increased after phenobarbital pretreatment (21.2 +/- 9.8 vs. 91.2 +/- 28.5 ml/min/kg, P less than .05). Free drug clearance was increased similarly. Finally, the pharmacokinetic theories derived for hepatic extraction of drugs subject to a high metabolic clearance can be successfully applied to men after liver enzyme induction with phenobarbital.     

Lack of effect of cimetidine on acetaminophen disposition in humans

The effect of cimetidine administration on the disposition of acetaminophen was evaluated in seven men and six women. One gram of acetaminophen was administered to each volunteer after an overnight fast on two occasions in a balanced crossover design with and without cimetidine, 300 mg every 6 hours beginning 50 hours before acetaminophen administration and continuing for 22 hours after. N-Acetylcysteine was administered on both occasions when acetaminophen was ingested to protect against glutathione depletion. Blood samples were collected serially for 12 hours after acetaminophen administration, and total urine volume was collected for 24 hours. Fractional clearances of acetaminophen through renal and metabolic routes (sulfation, glucuronidation, 3-hydroxylation, and glutathione conjugate formation) were not altered by cimetidine administration. Studies in microsomes prepared from two human organ donors indicated that cimetidine inhibited acetaminophen reactive metabolite formation noncompetitively, with Ki values of 0.35 mmol/L and 0.32 mmol/L for the respective livers, which is 5 to 10 times the putative cimetidine concentration required for therapeutic effect.     

Enantioselective disposition of 2-arylpropionic acid nonsteroidal anti-inflammatory drugs. II. 2-Phenylpropionic acid protein binding

The binding of 2-phenylpropionic acid (2PPA) enantiomers to rabbit albumin has been studied using fatty-acid-free albumin, with and without oleic acid, and using plasma from control rabbits and from rabbits rendered uremic with uranyl nitrate. The models of binding examined included specific binding at one and at two species of binding site, nonspecific binding and with inhibition between enantiomers with competitive or noncompetitive kinetics. Although any one aspect of the data is adequately modeled by nonspecific binding together with a single species of specific site, the simplest physical model consistent with the whole data requires two species of specific site together with nonspecific binding. Oleic acid in vitro, or other modifiers in vivo, inhibit the binding at both specific sites and reduce nonspecific binding. The inhibition at one site is sufficient that the situation in whole plasma simplifies to one site plus nonspecific binding. Competition between enantiomers occurs at this remaining site, at which R-2PPA binds more avidly than S-2PPA. Both specific and non-specific binding are reduced further in uranyl nitrate-induced renal failure. In the light of these findings, we discuss the implications of enantioselective binding and of competition between enantiomers for binding sites on the interpretation of drug disposition studies.     

Enantioselective disposition of lansoprazole in extensive and poor metabolizers of CYP2C19

OBJECTIVE: To evaluate the enantioselective disposition of lansoprazole in relation to the genetic polymorphism of CYP2C19. METHODS: A single oral dose of racemic lansoprazole (30 mg) was administered to 6 extensive metabolizers and 6 poor metabolizers whose genotypes were determined by use of polymerase chain reaction-restriction fragment length polymorphism. The pharmacokinetic parameters were estimated from the plasma concentrations of lansoprazole racemate, its enantiomers, and metabolites, which were measured for 24 hours after drug administration. The unbound fraction of lansoprazole enantiomers was determined by means of ultrafiltration of fresh human serum spiked with racemic lansoprazole. RESULTS: The plasma concentrations of R(+)-lansoprazole were consistently higher than those of the S(-)-enantiomer in both extensive and poor metabolizers of CYP2C19, and the mean area under the plasma concentration-time curve of the (+)- and (-)-enantiomers showed 4.3- and 5.8-fold differences between poor and extensive metabolizers, respectively. The (+)/(-) ratios of lansoprazole clearance were not significantly different between poor and extensive metabolizers (0.19 +/- 0.07 and 0.05 +/- 0.08, respectively). The values for volume of distribution of the (-)-enantiomer were 3- and 10-fold greater, respectively, than those of the (+)-enantiomer in poor and extensive metabolizers, which was related to a 2-fold higher unbound fraction of the (-)-enantiomer. CONCLUSIONS: The effect of CYP2C19 genetic polymorphism on the enantioselective disposition of lansoprazole seems to be less significant than the effect on omeprazole and pantoprazole. The disposition of lansoprazole enantiomers appears to be influenced by enantioselective protein binding and by enantioselective metabolism of lansoprazole.     

Stereospecific analysis and enantiomeric disposition of 3, 4-methylenedioxymethamphetamine (Ecstasy) in humans.

BACKGROUND: Little is known concerning the enantioselective disposition of 3,4-methylenedioxymethamphetamine (MDMA; ecstasy) in humans. In addition, the potential of utilizing the stereochemical composition of an analyte in biological media for forensic purposes requires investigation. METHODS: The enantiomers of MDMA and its demethylated metabolite, 3,4-methylenedioxyamphetamine (MDA), present in plasma and urine extracts were derivatized with (-)-(R)-alpha-methoxy-alpha-trifluoromethylphenylacetyl chloride and analyzed by gas chromatography-mass spectrometry and gas chromatography, respectively. The enantioselective disposition of MDMA and MDA was determined following oral administration of racemic MDMA (40 mg) to eight male volunteers. RESULTS: The plasma concentrations of (R)-MDMA exceeded those of the S-enantiomer [ratio R:S of the area under the curve (AUC), 2.4 +/- 0.3], and the plasma half-life of (R)-MDMA (5.8 +/- 2.2 h) was significantly longer than that of the S-enantiomer (3.6 +/- 0.9 h). The majority of the recovered material in urine was excreted within 24 h after dosing, with the recovery of (R)-MDMA (21.4% +/- 11.6%) being significantly greater than that of (S)-MDMA (9.3% +/- 4.9%), and with (S)- and (R)-MDA accounting for 1.4% +/- 0.5% and 1.0% +/- 0.3% of the dose, respectively. Mathematical modeling of plasma enantiomeric composition vs sampling time demonstrated the applicability of using stereochemical data for the prediction of time elapsed after drug administration. CONCLUSIONS: Analytical methods for determining the enantiomeric composition of MDMA and MDA in plasma and urine were developed. The disposition of MDMA in humans is stereoselective, with the more active S-enantiomer having a reduced AUC and shorter half-life than (R)-MDMA. The determination of stereochemical composition may be applicable for forensic purposes.     

Renal disposition of gentamicin, dibekacin, tobramycin, netilmicin, and amikacin in humans.

The tubular disposition of five aminoglycosides was studied in humans to establish a possible relationship between tubular reabsorption and the nephrotoxicity that has been described in the literature. Thirty-three healthy male volunteers received a continuous intravenous infusion of isotonic saline with inulin, followed 1 h later by inulin plus gentamicin, dibekacin, tobramycin, netilmicin, or amikacin (1 mg/kg per h) or amikacin (4 mg/kg per h) over a period of 2 h. Brain-stem-evoked response audiometry was performed both before and at the end of each infusion. The latency of wave V remained constant whichever antibiotic was considered. The glomerular filtration rate did not vary significantly during the infusion of each drug. The percent fractional excretion was 79 +/- 6, 81 +/- 22, 85 +/- 5, and 99 +/- 9 for gentamicin, dibekacin, tobramycin, and netilmicin, respectively, and 83 +/- 4 and 124 +/- 13 for amikacin at concentrations of 1 and 4 mg/kg per h, respectively. Net balance and renal clearance were similar for the five aminoglycosides when administered at a rate of 1 mg/kg per h. With gentamicin only, fractional excretion was correlated with the urinary flow rate. We can conclude that (i) gentamicin, generally considered the most nephrotoxic agent, had the highest degree of net reabsorption; (ii) netilmicin exhibited a net zero tubular balance; (iii) amikacin had different patterns of tubular disposition according to the dose, i.e., reabsorption at 1 mg/kg per h and secretion at 4 mg/kg per h, raising the hypothesis of a saturable process of reabsorption; and (iv) these differences in tubular reabsorption could account at least in part for the known different nephrotoxic potentials of these five aminoglycosides in humans.     

Kinetics of absorption and elimination of ofloxacin in humans after oral and rectal administrations

Ofloxacin pharmacokinetics have been studied in four healthy subjects after a single oral or rectal dose, each of 200 mg. For the oral dose t _max was about 2 h, C _max 1·96±0·56 μ g/ml and AUC _1–15 15·22  μ g/ml.h. Two‐phase elimination pharmacol kinetics were observed for the oral dose, t_1/2 for the rapid elimination phase was 3·3 h and for the slow phase 10 h. With the rectal dose t _max was 6 h, C _max 0·71±0·44 μ g/ml and AUC _0–15 7·58  μ g/ml.h. The relative rectal bioavailability ( AUC rectal/ AUC oral) was 49·8%.%Elimination rate of the rectal dose was generally slow ( t_1/2=9 h), an observation attributable to the sustained‐release effect of the rectal suppository base, PEG 6000. The indication is that the rectal formulation cannot be substituted totally for the oral without first increasing the rectal dose; the 200 mg suppository can however be employed as a follow‐up therapy to the oral dose in certain situations.     

The effect of sulfinpyrazone on the disposition of pseudoracemic phenprocoumon in humans

The effect of sulfinpyrazone on the pharmacokinetics and disposition of the enantiomers of pseudoracemic phenprocoumon was assessed by analyzing serial plasma, urine, and fecal samples for parent drug and metabolites by GC/MS. Essentially all of the administered dose could be accounted for either as parent drug, known metabolites, or their conjugates. Phenprocoumon and the 7-hydroxymetabolite represented the major materials recovered. All drug-related materials excreted into the urine were extensively conjugated. Sulfinpyrazone treatment did not affect the hypoprothrombinemia produced by phenprocoumon nor did it significantly alter the plasma elimination kinetics of the individual (R)- and (S)-enantiomers. However, an apparent increased free fraction of both enantiomers in plasma and inhibition of 7-hydroxylation of (S)-phenprocoumon were observed in the presence of sulfinpyrazone. The results of this study are contrasted with those of a previous study on the interaction between sulfinpyrazone and the structurally similar coumarin anticoagulant warfarin.     

Bergamottin contribution to the grapefruit juice-felodipine interaction and disposition in humans

OBJECTIVES: Our objectives were to evaluate the contribution of bergamottin to the grapefruit juice-felodipine interaction and to characterize bergamottin disposition. METHODS: In this study 250 mL grapefruit juice; 2-, 6-, or 12-mg capsules of bergamottin plus water; or water was administered with 5 mg extended-release felodipine to 11 volunteers in a partially randomized, 5-way crossover study. Plasma concentrations of felodipine, its primary metabolite (dehydrofelodipine), bergamottin, and 6',7'-dihydroxybergamottin were determined. RESULTS: Grapefruit juice (containing 1.7 mg bergamottin) increased peak plasma concentration (C max ) and area under the plasma concentration-time curve (AUC) of felodipine by 89% (P < .025) and 54% (P < .025), respectively, compared with water. With 2 mg bergamottin, felodipine C max increased by 33% (P < .05). The increase by bergamottin was markedly variable among individuals (range, -33% to 125%). With 6 mg bergamottin, felodipine C max was enhanced by 35% (P < .025), and with 12 mg bergamottin, felodipine C max increased by 40% (P < .05) and AUC increased by 37% (P < .05) compared with water. Bergamottin measured in plasma after administration of 6 and 12 mg produced C max values of 2.1 and 5.9 ng/mL, respectively, and times to reach C max of 0.8 and 1.1 hours, respectively. The bergamottin metabolite 6',7'-dihydroxybergamottin was detected in plasma of some subjects after bergamottin administration. CONCLUSIONS: Bergamottin enhanced the oral bioavailability of felodipine and may cause a clinically relevant drug interaction in susceptible individuals. Grapefruit juice-drug interactions likely also involve other furanocoumarins, possibly acting in combination by additive or synergistic mechanisms. Bergamottin has systemic availability and is metabolized in vivo to 6',7'-dihydroxybergamottin.     

Penetration of ofloxacin into heart valves, myocardium, mediastinal fat, and sternal bone marrow in humans.

Ofloxacin penetration into heart tissue (valve and myocardium), mediastinal fat, and sternal bone marrow was the object of a prospective nonrandomized study. Thirty-six patients undergoing mitral and/or aortic valve replacement were included. Patients were divided into two groups of 18 patients each. Group 1 patients were administered a single 400-mg intravenous dose of ofloxacin over a 30-min period upon anesthesia (n = 6) or at 1 h (n = 6) or 6 h (n = 6) prior to surgery. Group 2 patients received a 200-mg oral dose of ofloxacin every 12 h during the 48 h preceding surgery. In this group, the final dose of ofloxacin was administered 3 h (n = 9) or 8 h (n = 9) before anesthesia. Plasma and tissue ofloxacin concentrations were assayed by high-pressure liquid chromatography. In group 1 patients, the peak level in plasma was 15.9 +/- 2.5 micrograms/ml. Peak ofloxacin levels in tissue were reached by hour 1 and were 8.89 +/- 2.16 micrograms/g in myocardium and 5 +/- 0.75 micrograms/g in heart valves. A significant decrease in ofloxacin levels in heart valve tissue and sternal bone marrow was observed after hour 3. Nevertheless, ofloxacin myocardial, heart valve, and sternal bone marrow levels remained higher than the MICs for the usually susceptible pathogens for at least 3 h. In group 2 patients, myocardial levels were long lasting (6.46 +/- 1.92 micrograms/g [4 to 8 h] and 5.92 +/- 0.95 micrograms/g [8 to 12 h]) and remained higher than those observed in the other tissues over the entire study period. A progressive but insignificant decrease in ofloxacin heart valve levels was observed (from 2.46 +/- 0.40 micrograms/g [4 to 8 h] to 1.57 +/- 0.22 micrograms/g [8 to 12 h]). In both groups, concentration in mediastinal fat were lower and tended to decrease with time. These were 1.83 +/- 0.61 micrograms/g with the first hour and 0.85 +/- 0.43 micrograms/g between hours 8 and 12 in group 1 and 1.74 +/- 0.52 micrograms/g between hours 4 and 8 and 0.67 +/- 0.11 micrograms/g between hours 8 and 12 in group 2. In conclusion, satisfactory diffusion of ofloxacin into heart tissue seems to favor use of the drug in the treatment of bacterial endocarditis due to susceptible pathogens. Furthermore, the progressively decreasing concentrations observed in heart valve and sternal bone marrow and the poor levels achieved in mediastinal fat suggest the need for renewing injection 3 h following initial infusion if the drug is used as an antibiotic prophylactic agent during cardiovascular surgery.