Search

The anticoagulant warfarin exemplifies a type of drug that exhibits high affinity to pharmacologic target sites of limited capacity, resulting in unusual concentration-dependent distribution and elimination properties. The time course of warfarin concentrations in the serum, liver, kidneys, muscle, and abdominal fat of male Sprague-Dawley rats was determined by high-performance liquid chromatography after IV injection of a 0.25 or 1.0mg/kg dose. The rats were preclassified on the basis of their serum free fraction of warfarin; animals with free fraction values of approximately 0.004 and 0.01 and corresponding differences in elimination half-life were selected for study, yielding four experimental groups. Several rats of each group were sacrificed periodically over approximately 80–240h for determination of drug concentrations. S-warfarin concentrations in serum declined apparently exponentially over at least one order of magnitude. During this time, concentrations in all other assayed tissues declined much more slowly. In another experiment, S-warfarin concentrations in serum and liver were followed for approximately 50 days after IV injection of a 1mg/kg dose. This revealed a terminal, very slow elimination phase in serum nearly parallel to the decline in liver drug concentrations. Simultaneous physiologic modeling of all data (30 equations) using ADAPT II (Biomedical Simulations Resource, Los Angeles, CA), with intrinsic clearance, the dissociation constant of the warfarin-high affinity binding site complex, and two binding parameters for the (unassayed) remainder tissue compartment as parameters of unknown value, yielded very good fittings and parameter estimates with relatively small standard deviations. The unusual dose-dependent accumulation characteristics of this type of drug during continuous infusion are demonstrated by computer simulation of published results of warfarin infusions in rats. Utilization of a model premised on similar target-mediated drug disposition also allowed characterization of data from the literature for racemic warfarin pharmacokinetics in man.

The anticoagulant warfarin exemplifies a type of drug that exhibits high affinity to pharmacologic target sites of limited capacity, resulting in unusual concentration-dependent distribution and elimination properties. The time course of warfarin concentrations in the serum, liver, kidneys, muscle, and abdominal fat of male Sprague-Dawley rats was determined by high-performance liquid chromatography after IV injection of a 0.25 or 1.0 mg/kg dose. The rats were preclassified on the basis of their serum free fraction of warfarin; animals with free fraction values of approximately 0.004 and 0.01 and corresponding differences in elimination half-life were selected for study, yielding four experimental groups. Several rats of each group were sacrificed periodically over approximately 80–240 h for determination of drug concentrations. S-warfarin concentrations in serum declined apparently exponentially over at least one order of magnitude. During this time, concentrations in all other assayed tissues declined much more slowly. In another experiment, S-warfarin concentrations in serum and liver were followed for approximately 50 days after IV injection of a 1 mg/kg dose. This revealed a terminal, very slow elimination phase in serum nearly parallel to the decline in liver drug concentrations. Simultaneous physiologic modeling of all data (30 equations) using ADAPT II (Biomedical Simulations Resource, Los Angeles, CA), with intrinsic clearance, the dissociation constant of the warfarin-high affinity binding site complex, and two binding parameters for the (unassayed) remainder tissue compartment as parameters of unknown value, yielded very good fittings and parameter estimates with relatively small standard deviations. The unusual dose-dependent accumulation characteristics of this type of drug during continuous infusion are demonstrated by computer simulation of published results of warfarin infusions in rats. Utilization of a model premised on similar target-mediated drug disposition also allowed characterization of data from the literature for racemic warfarin pharmacokinetics in man. © 2003 Wiley-Liss, Inc. and the American Pharmaceutical Association J Pharm Sci 92:985–994, 2003

Drug dosage regimen design is a rational process based on a drug's pharmacokinetics and pharmacodynamics in a specific patient setting and refined by feedback information. This process can be severely affected by medication non-compliance, particularly if the prescriber is unaware of the patient's failure to take medications as directed. Non-compliance takes many forms, including omission or addition of one or more doses, deviation from the prescribed timing of drug administration, and failure to adhere to required dosing conditions with respect to food, water and posture. These problems have many causes, including forgetfulness, lack of motivation, lack of confidence in the medication or prescriber, adverse effects, cost, and dosing complexity. Clinical pharmacology courses should include at least one lecture on medication non-compliance, its patterns, causes, implications and management. From a pharmacokinetic perspective, the impact of missed doses increases with an increase of the medication non-compliance impact factor τ / t 1/2 ( τ is the prescribed dosing interval and t 1/2 is the terminal elimination half-life of the drug). The impact of missed doses is decreased by more frequent dosing despite the fact that compliance decreases with increased dosing frequency. Most drug product package inserts do not address compliance issues. Drug-specific pharmacokinetic/pharmacodynamic simulations can yield valuable information for advising patients what to do if they missed one or more doses of their medication.

OBJECTIVE: To explore and evaluate various strategies for drug concentration- and effect-controlled clinical trials, respectively, in the context of studies of population pharmacodynamics (concentration—effect relationships).METHODS: The relative utility of drug concentration- and pharmacologic effect-controlled, randomized clinical trials with two or three concentration—effect measurements for each subject has been explored by computer simulation. The basis for these simulations was a sigmoid-Emax(maximum effect) pharmacodynamic model with Emax= 100%, EC50(drug concentrations required to produce an effect intensity of 50%) = 10 concentration units, gamma = 2, and no hysteresis. Emaxand gamma were held constant whereas EC50was assumed to be log-normally distributed with a 26% coefficient of variation of the natural log-normalized data. A smaller random variability and variability due to measurement error also were incorporated in the simulations. To explore the implications of variable and unknown and gamma values, the suitability of linear and log—linear interpolation procedures for two-point concentration—effect data in different regions of the sigmoid-Emaxcurve was compared.RESULTS: Pharmacologic effect-controlled clinical trials with 300 hypothetical subjects and targeted effect intensities of 25% and 75% yielded very good estimates of drug concentrations required to produce effect intensities of 35%, 50%, and 65%, whereas concentration-controlled trials yielded much poorer estimates. Moreover, the concentration-controlled trials, despite optimum choice of targeted concentrations, yielded a large number of data points with poor information content (effect intensities of <15% or >85%). Determinations based on targeted effect intensities of 25% and 75% yielded better estimates of individual EC50values than those targeted for 25% and 50% or 50% and 75% effect intensity. Results were not significantly improved by adding a third measurement (targeted to 50% effect) to the 25% and 75% effect design. Estimations of drug concentrations required to produce an effect intensity of 50%, based on log—linear interpolation of exact concentration—effect data at 25% and 75%, yielded exact results independent of gamma value (0.5–8.0) whereas linear interpolation produced large overestimates at gamma = 0.5 or 1.0 but satisfactory estimates at gamma ≤ 2.0. Similar calculations for an effect intensity of 15% based on exact concentration—effect data at 5% and 25% yielded reasonably good estimates by both methods of interpolation over a wide range of gamma values. A review of the clinical literature showed that gamma values are usually 2 or higher.CONCLUSIONS: Population pharmacodynamic studies of reversibly acting drugs without pharmacodynamic hysteresis or time dependency (e.g., tolerance) can be successfully conducted using a pharmacologic effect-controlled randomized clinical trial design with only two properly selected target effect intensities per subject.

Purpose. To explore, by simulation procedures, the feasibility of characterizing, from sparse data, the concentration-effect relationship of drugs with pharmacodynamic hysteresis.

The total clearance of dicumarol was determined in 172 adult male Sprague-Dawley rats. Clearance values ranged from 1.46 to 27.0 ml/hr/kg. Statistical analysis of a histogram of the total clearance values indicated a trimodal distribution, with modes at 6.28, 14.8, and 23.7 ml/hr/kg. The percentage of animals in each of these components was 60.5 33.7, and 5.8. A previous study had shown that the total clearance of dicumarol was proportional to the fraction of nonprotein-bound drug in serum (serum free fraction) and that interindividual differences in total clearance of dicumarol in rats were due almost entirely to corresponding differences in the serum free fraction. Therefore, it is likely that the observed trimodal frequency distribution of total clearance values reflects a similar distribution of serum free fraction values of dicumarol. The frequency distribution curve for dicumarol total clearance is very similar to the trimodal frequency distribution curve for warfarin serum free fraction values in rats. This observation is consistent with the previously demonstrated strong correlation of serum free fraction values of dicumarol and warfarin in individual animals.

The effect of ibuprofen on the pharmacokinetics and anticoagulant action of racemic warfarin was determined in a crossover study on male Sprague-Dawley rats. At average plasma concentrations of 24–83 mg/liter, ibuprofen decreased the biological half-life and increased the total clearance of warfarin. It also increased the anticoagulant effect produced by a given plasma concentration of total (free and protein- bound) warfarin. These effects of ibuprofen appear to be a consequence of its displacing effect on warfarin in plasma.

Inhibition of drug metabolism was studied in adult male Sprague‐Dawley rats. A hydroxylated metabolite of phenylbutazone (oxyphenbutazone) inhibited the elimination of phenytoin, which is metabolized by oxidative pathways. The biotransformation of a relatively polar and only slightly plasma protein‐bound drug, antipyrine, was subject to product inhibition by a hydroxylated metabolite, 4‐hydroxyantipyrine. Neither oxyphenbutazone nor 4‐hydroxyantipyrine measurably affected the elimination kinetics or metabolic fate of a drug (sulfanilamide) that is not metabolized by oxidative pathways.

The distribution, elimination, and anticoagulant effect of dicumarol and warfarin were determined in adult male rats following intravenous injection of single doses of these drugs in crossover experiments. The biological half‐life of dicumarol ranged from 5 to 28 hr; that of warfarin ranged from 9 to 30hr. There was a statistically significant correlation between the following pharmacokinetic characteristics of dicumarol and warfarin in individual animals: biological half‐life, apparent volume of distribution, total plasma clearance, and concentration in plasma eliciting onehalf the maximum anticoagulant effect (effective concentration). The mean ratio of the respective biological half‐lives (warfarin/dicumarol) was 1.42, and that of the apparent volumes of distribution was 1.50. The ratio of the effective plasma concentrations (dicumarol/warfarin) was correlated negatively with the half‐life of dicumarol and positively with the ratio of the half‐life values (warfarin/dicumarol) in individual animals. Additional studies with serum samples from other rats showed pronounced interindividual differences in the serum protein binding of both dicumarol and warfarin and a strong correlation between the protein binding of these two drugs in serum of individual animals. The results of this study, together with the results of previous studies in this series, indicate that serum protein binding is the major determinant of interindividual differences in the pharmacokinetics of dicumarol and warfarin in rats under these experimental conditions.

The relationships between the protein binding, distribution in the body, and kinetics of elimination of warfarin were studied. Individual rats eliminated warfarin by apparent firstorder kinetics, with a biological half-life of 5.9–41hr and a total plasma clearance of 2.4–22ml kg−1hr−1. There is a strong positive correlation between the apparent volume of distribution (Vd) and the elimination rate constant (kel). There was no apparent concentration dependence of warfarin binding to serum proteins over a wide concentration range, but there were pronounced intersubject variations in protein binding, with the free fraction of drug (f) in serum ranging from 0.172 × 10−2to 1.53 × 10−2. There are strong positive correlations between fand kel, fand Vd, and fand the kidney-serum concentration ratio of warfarin. Consistent with theory, there is an excellent positive linear correlation between fand total plasma clearance of the drug. The intersubject variation in fis not related to variations in serum albumin or total protein concentration. There is a strong correlation between values of ffor serum and liver homogenate in individual animals, consistent with the lack of correlation between fin serum and the liver-serum concentration ratio of warfarin. These results show that the pronounced intersubject variation in the elimination of warfarin observed in this investigation was related to interindividual differences in plasma protein binding of the drug. The differences in protein binding cannot be ascribed to differences in plasma protein concentrations and may reflect configurational differences of proteins or the presence of an endogenous displacing agent at different concentrations.

The purpose of this investigation was to determine if the reported prolongation of barbiturate–and ethanol-induced sleeping times by nicotine in rodents are pharmacodynamic or pharmacokinetic interactions. Adult female rats were pretreated with nicotine, either 0.5 or 1.5 mg/kg ip acutely or 0.5 mg/kg ip daily for 6 d, whereas control animals received saline solution. Phenobarbital or ethanol was infused intravenously at a slow rate until the rats lost their righting reflex. Acute pretreatment with nicotine reduced significantly the serum, brain, and cerebrospinal fluid (CSF) concentrations of phenobarbital and ethanol, respectively, at the onset of the loss of the righting reflex. Chronic pretreatment with nicotine had no such potentiating effects, indicative of rapid development of tolerance to nicotine. Neither acute nor chronic pretreatment with nicotine had any apparent effect on the elimination kinetics of phenobarbital or ethanol, on biochemical indices of hepatic integrity and renal function, or on the permeability of the blood–CSF barrier to protein. Nicotine, unlike morphine, did not increase the nociceptive threshold (tail squeeze) of rats under the experimental conditions. It is concluded that acute, but not chronic, pretreatment with nicotine increases the sensitivity of rats to the hypnotic effects of phenobarbital and ethanol, respectively. These interactions are entirely pharmacodynamic and have no apparent pharmacokinetic component.

Theophylline, the widely used antiasthmatic drug, can cause life-threatening, generalized seizures when administered in excessive doses. The plasma concentrations of theophylline associated with these seizures vary widely among patients, thereby complicating efforts to prevent seizures by timely initiation of appropriate treatment. Some investigators suspect that chronic administration increases the neurotoxicity of theophylline but others have suggested the opposite. We have studied this problem in an animal model of theophylline-induced seizures. Osmotic pumps containing theophylline solution or drug-free solvent (for the surgical control group) were implanted in adult female Lewis rats, yielding almost constant serum theophylline concentrations of about 14 mg/liter for 7 days in the treated group. On the seventh day, theophylline was administered by much more rapid iv infusion to the two groups of animals and to one nonimplanted (nonsurgical) control group until onset of maximal seizures. There were no statistically significant differences between the three groups with respect to the concentrations of theophylline in serum, serum water, brain, and cerebrospinal fluid at onset of seizures. The concentrations of theophylline metabolites were either very low or undetectable. Under the experimental conditions, preexposure of rats for 7 days to theophylline in the human therapeutic concentration range had no apparent effect on the acute neurotoxicity of the drug.

Plasma concentrations of propoxyphene and its major metabolite, norpropoxyphene, were determined over at least 12 hr after oral administration of 130 mg dextropropoxyphene hydrochloride to eight men with hepatic cirrhosis, of whom four had a surgically constructed portacaval shunt, and to seven healthy men. Propoxyphene concentrations were appreciably higher and norpropoxyphene concentrations were much lower in the patients than in the normal subjects. The ratio of areas under the plasma concentration-time curve from 0 to 12 hr, norpropoxyphene:propoxyphene, was 0.70 ± 0.46 (\[xmacr] ± SD) in patients and 3.94 ± 0.83 in normal subjects. A similar decrease in this ratio was observed previously in otherwise healthy dogs after surgical construction of a portacaval shunt when propoxyphene was given orally, but not after intravenous injection of the drug. A woman with portacaval shunt and essentially complete renal failure was also studied; she exhibited the highest propoxyphene peak concentration in this investigation and had no detectable norpropoxyphene in plasma. Most of the patients, unlike the normal subjects, experienced considerable sedation after propoxyphene. These results are probably due to increased systemic availability of orally administered propoxyphene in patients with hepatic cirrhosis and possibly to increased receptor response to the drug by these patients. It is concluded that propoxyphene should be administered cautiously and in reduced doses to patients with hepatic dysfunction.Clinical Pharmacology and Therapeutics (1980) 28, 417–424; doi:10.1038/clpt.1980.182

This investigation was designed to determine the effect of experimental hyperthyroidism on the hypnotic activity of a benzodiazepine and on the binding characteristics of the benzodiazepine receptor complex. Rats were made hyperthyroid by subcutaneous implantation of slow release pellets containing L-thyroxine. This treatment produced the characteristic symptoms of hyperthyroidism: increased serum thyroxine concentrations, increased heart weight and body temperature, and decreased serum protein concentrations. The hyperthyroid rats and a parallel group of normal animals (with drug-free pellets implanted subcutaneously) were slowly infused intravenously with oxazepam until they lost their righting reflex. The hyperthyroid rats required a significantly larger dose of the benzodiazepine and their total serum and cerebrospinal fluid concentrations of oxazepam (but not the serum concentration of free drug and the brain concentration) at the onset of loss of the righting reflex were modestly but statistically significantly lower than those of the normal rats. The receptor density and affinity for diazepam in hyperthyroid rats were not significantly different from those of the normal animals. Hyperthyroidism apparently affects the pharmacokinetics of oxazepam but has, at best, only a small effect on the pharmacodynamics (hypnotic activity) of the drug.

This investigation determined the effect of serum protein binding on the kinetics of sulfisoxazole distribution, metabolism, and excretion. Adult rats, whose serum free fraction of sulfisoxazole (at a total concentration of 81 ± 6 μg/ml) was 0.05-0.24, received a rapid intravenous injection of 20 mg/kg. Sulfisoxazole concentrations in plasma declined biexponentially with time. There were pronounced and reproducible interindividual differences in the total, metabolic, and renal sulfisoxazole clearances, each positively correlated with the serum free fraction of sulfisoxazole. The renal sulfisoxazole clearance had a component unaffected by serum protein binding. The apparent central compartment volume increased with an increasing serum free sulfisoxazole fraction, but the latter had no apparent effect on the first exponential term of the biexponential equation describing sulfisoxazole elimination kinetics in rats. Serum protein binding was a major determinant of intersubject differences in sulfisoxazole excretion and biotransformation kinetics.

The effect of salicylic acid injection and infusion on the concentrations of free and total (sum of free and bound) unconjugated bilirubin in plasma was studied in hyperbilirubinémie rats. In vitroaddition of salicylic acid, 5-30 mg/100 ml, to rat plasma containing bilirubin, 12.7 mg/100 ml, caused a pronounced, salicylate concentration-dependent increase in the free bilirubin concentration. Normal rats were made hyperbilirubinémie by continuous intravenous infusion of bilirubin; the concentrations of free and total unconjugated bilirubin in plasma, and of total unconjugated bilirubin in whole blood, were determined as a function of time. Rapid intravenous injection of salicylate caused a rapid and pronounced decrease of total bilirubin concentrations in plasma and whole blood but had no apparent effect on the concentration of free bilirubin in plasma. Similar effects were obtained with slow infusions of salicylate, except that total bilirubin concentrations decreased gradually. These observations are consistent with the theory that the clearance rate of bilirubin by the body is proportional to the concentration of free bilirubin in plasma and that the steady-state concentration of free bilirubin depends only on the formation rate and intrinsic clearance of the pigment and is not affected by displacement of bilirubin from plasma protein binding sites. Occasional cases of acute apparent bilirubin intoxication were associated with unusually elevated concentrations of free bilirubin, reflecting a decreased intrinsic metabolic clearance of the pigment.

The effect of sulfisoxazole on the time course of free (unbound) bilirubin concentrations in plasma was studied. Normal adult rats were made hyperbilirubinémie by continuous intravenous infusion of bilirubin. Sulfisoxazole was administered by either rapid intravenous injection or slow intravenous infusion, and the plasma concentrations of free and total (free plus bound) unconjugated bilirubin were determined as a function of time. Rapid injection of sulfisoxazole caused a rapid and pronounced decrease of total bilirubin concentrations in plasma but had only a transient effect on the concentration of free bilirubin. Slow infusion of sulfisoxazole caused a gradual and eventually pronounced decrease of total bilirubin concentrations in plasma but had no apparent effect on the concentration of free bilirubin at any time. These results are consistent with recently developed pharmacokinetic theory according to which the plasma clearance of total bilirubin should increase upon administration of a displacing agent while the plasma clearance of free bilirubin should remain unchanged. Bilirubin-induced encephalopathy caused by sulfisoxazole or other displacing agents may be due to very transient elevations of free bilirubin concentrations in plasma of infants with elevated plasma concentrations of total bilirubin and the consequent redistribution of the pigment to extravascular sites, including the brain.

This investigation was designed to determine if liver diseases can modify the pharmacodynamics of the central nervous system depressants phenobarbital and ethanol. Two experimental models of liver diseases in rats were used: extrahepatic cholestasis produced by bile duct ligation and hepatic necrosis induced by carbon tetrachloride administration. Phenobarbital (both models) or ethanol (cholestasis model only) was infused slowly intravenously until the rats lost their righting reflex. Drug concentrations in serum, brain, and cerebrospinal fluid at that time were determined in the diseased animals as well as in shamoperated or solvent‐treated controls. Phenobarbital concentrations at the onset of action were not significantly different between controls and either 5‐d or 12‐d cholestatic rats, except for total serum concentrations which were lower in the cholestatic groups due to reduced protein binding. Ethanol concentrations were slightly but statistically significantly lower in 12‐d cholestatic rats as compared with controls. Neither 5‐d nor 12‐d carbon tetrachloride‐induced hepatic dysfunction had any significant effect on phenobarbital concentrations at the onset of loss of righting reflex, except for a marginal decrease in the cerebrospinal fluid concentration of rats that had been treated for 5 d with the hepatotoxin. It was concluded that, under the experimental conditions, the hepatic diseases investigated did not have appreciable effects on the central nervous system response to the hypnotic action of phenobarbital and ethanol.

A comprehensive investigation of the effect of phenylbutazone on warfarin pharmacokinetics and anticoagulant activity was carried out in rats to identify and quantify various aspects of the interaction between these drugs. Adult male rats received intravenous racemic warfarin alone and together with phenylbutazone in a crossover experiment. Prothrombin complex activity and the plasma concentrations of phenylbutazone and of free and total (free plus protein-bound) warfarin were determined repeatedly for up to 60hr. The total plasma clearance, the apparent volume of distribution, and the disposition rate constant (β) of warfarin were significantly increased and the intrinsic plasma warfarin clearance was significantly decreased during phenylbutazone administration. Phenylbutazone decreased the serum protein binding of warfarin both in vitroand in vivo, but the in vivoeffect was much more pronounced, apparently due to the displacing effect of phenylbutazone metabolite(s). Phenylbutazone alone had no apparent effect on prothrombin complex activity in vitrobut caused a modest, yet statistically significant, anticoagulant effect in vivo.The anticoagulant effect-plasma warfarin concentration curves for total and free warfarin were shifted to a considerably lower concentration range during phenylbutazone treatment. Thus, the interaction between phenylbutazone and warfarin involves at least three processes: an inhibition of warfarin biotransformation (decreased intrinsic clearance); displacement of warfarin from plasma protein binding sites (increased free fraction); and apparent potentiation of the anticoagulant action produced by a given plasma warfarin concentration. The latter may have been caused, at least in part, by a direct anticoagulant effect of phenylbutazone and/or its metabolite(s). The net effect of decreased protein binding and decreased intrinsic clearance was an increase in the total plasma warfarin clearance. The results of this investigation demonstrate that drug interactions can be complex and multifactorial.

The purpose of this investigation was to determine the effect of dose on warfarin pharmacokinetics in rats. First, in a crossover experiment, rats received 14C-warfarin, 0.2mg/kg iv, 12hr after an injection of either nonradioactive warfarin (0.5mg/kg) or saline solution. Warfarin concentrations in plasma declined triexponentially as a function of time. Pharmacokinetic analysis revealed that pretreatment with warfarin significantly decreased the apparent volume of distribution, total plasma clearance, and intrinsic plasma clearance of the drug. In the second part of the investigation, rats received single intravenous warfarin injections in the order of 0.1-1.0-0.1 or 1.0-0.1-1.0mg/kg at 2-week intervals. The apparent volume of distribution, total plasma clearance, and intrinsic plasma clearance of the 1.0-mg/kg warfarin dose were appreciably lower than those of the 0.1-mg/kg dose. The decrease in the apparent volume of distribution of warfarin with increasing dose is consistent with the previously observed concentration dependence in hepatic uptake of the drug.

Previous studies in this laboratory demonstrated that the rate of intestinal absorption of theophylline in rats is increased significantly by ethanol in low concentrations and that this absorption enhancing effect is associated with an increased net water flux from the intestine. It is now shown that other alcohols, namely methanol, n‐propanol, n‐butanol, glycerin, propylene glycol, and, to a lesser extent, mannitol and sorbitol, can also increase net water flux from the small intestine of anesthetized rats. Polyethylene glycol 200 and 400 had no such effect, suggesting that these compounds do not penetrate to a site of action that elicits the increased net water flux. At initial concentrations of 0.1 and 1.0 M, glycerin and propylene glycol increase significantly the intestinal absorption rate of theophylline from the small intestine of anesthetized rats. The results show that the theophylline absorption enhancing effect of ethanol is not limited to that particular alcohol.

The purpose of this investigation was to determine the effect of phenobarbital on the systemic availability of orally administered dicumarol in rats. Adult male Sprague‐Dawley rats, matched for dicumarol free fraction in serum, received either phenobarbital sodium, 75 mg/kg, or saline solution, orally or intravenously, daily for 7 days. On Day 6, they also received14C‐dicumarol, 2 mg/kg iv, and unlabeled dicumarol, 50 mg/kg po, in aqueous suspension. Venous blood samples were obtained serially over 32hr through an indwelling cannula. Systemic dicumarol availability was determined from the dose‐normalized ratio of areas under the plasma concentration‐time curves. Phenobarbital treatment almost doubled the total clearance of dicumarol and the intrinsic clearance of free dicumarol, with no significant difference between the inductive effects of oral and intravenous doses of phenobarbital. Systemic dicumarol availability in control rats (mean ±SD) was 84 ± 8% (n= 10) and 84 ± 10% (n= 6) in the oral and intravenous phenobarbital studies, respectively. The systemic dicumarol availability in phenobarbital‐treated rats was appreciably lower: 48 ± 7% (n= 10) and 61 ± 12% (n= 6) for orally and intravenously treated animals, respectively. The effect of oral phenobarbital on systemic dicumarol availability was more pronounced than that of intravenous phenobarbital (p< 0.025). The apparent first‐order absorption rate constants for the fraction of the dose available systemically were similar for control and treated animals. There was a positive correlation between systemic dicumarol availability and total dicumarol clearance in control animals (p< 0.001). Proper matching of control and treated animals is, therefore, important for this type of study. The rat appears to be a good model for investigating the mechanism of the inhibitory effect of phenobarbital on dicumarol absorption observed previously in humans.

Little is known at present about the effects of human exposure to microgravity on the pharmacodynamics of medicinal agents. Considering the known physiologic perturbations associated with spaceflight and the effects of some of these perturbations on the pharmacodynamics of certain medicinals that act on the central nervous system, it is likely that the pharmacodynamics of at least some of these agents will be altered in individuals who are subjected to microgravity and other conditions of spaceflight. It is timely to initiate formal studies, initially in ground‐based animal models and human volunteers during prolonged, head‐down bed rest and eventually in animals and human volunteers exposed to microgravity, of pharmacodynamics and pharmacokinetics during spaceflight.

The purpose of this investigation was to explore theoretically certain strategies for optimizing the therapeutic index of drugs and to assess these strategies experimentally with two diuretics. Diuretic agents allow dosing rate flexibilities because the temporal profile of diuretic action can vary considerably as long as the total diuretic effect per day is the same. They can also be used in combination. Experiments were designed to determine if the therapeutic index of furosemide and hydrochlorothiazide can be optimized by administering one or the other at a certain rate or by administering the two drugs together in a certain ratio and at a certain rate. Male Lewis rats received one or the other drug, or combinations of the two, by i.v. infusion at different rates. Several timed urine collections were made under steady-state conditions, with excreted urine replaced volume for volume by i.v. lactated Ringer's solution. The urine flow rates and the urinary excretion rates of the diuretics and of Na+ and K+ were determined. The relationship between the diuretic effect of either of the two drugs given alone and the respective drug excretion rate could be described by the Hill equation. The ratio of urine flow rate to K+ excretion rate exhibited a marked dependence on hydrochlorothiazide excretion rate (highest ratio at high excretion rates), whereas the K+/Na+ excretion rate ratio was constant over a wide range of hydrochlorothiazide excretion rates. There was no significant change of these ratios with changing excretion rate of furosemide. Infusion of the two diuretics in combination of different ratios and at different combined rates under steady-state conditions revealed proportionality between the urinary excretion rates of Na+ and urine over a wide range and a decreasing K+ /urine excretion rate ratio with increasing urine flow rate. Hence, a favorable increase in the Na+/K+ excretion rate ratio was attained with increasing urine flow rate. These experiments and computer simulations demonstrate in principle the feasibility of optimizing the therapeutic index by appropriate selection of a drug's dosing rate or by dosing a combination of two drugs at an appropriate ratio and rate.

Important concepts of pharmacokinetics, such as the biologic half-life, clearance, and bioavailability of drugs, can be utilized to describe and predict the time course of drug concentrations in plasma as a function of dose and frequency of drug administration. In quantitative terms, they can serve to characterize the individuality of the patient, including the effects of disease, with respect to drug absorption, distribution, excretion, and metabolism.(JAMA 235:1864-1867, 1976)

The pharmacokinetic concepts introduced and defined in a preceding article are now applied to the management of drug therapy for the individual patient. The factors that affect the time course of drug concentrations in plasma produced by repetitive administration include the dosing rate, total clearance, biologic half-life, and systemic availability of the drug. A clinical pharmacokinetics service can monitor drug concentrations in biologic fluids, design individualized drug-dosing regimens, and carry out pharmacokinetic diagnostic work-ups to help determine the reasons for a patient's unusual response to drug therapy.(JAMA 235:1987-1992, 1976)

Salicylate kinetics were determined in 28 subjects 25 to 92 years old who received single, oral doses of sodium salicylate (1 gm/1.73 m2). The serum AUC∞of total salicylate did not correlate with age. There was a weak positive correlation between the AUC∞of free (unbound) drug and age, but there was no apparent difference between the AUC∞values of the 15 women and 13 men. Seven of the 16 subjects > 70 years of age cleared salicylate at about the same rate as the younger subjects. A comparison of these seven subjects with the nine > 70 years old who were slow eliminators of salicylate revealed that the latter group consisted of more bedridden patients and that these patients had somewhat lower serum albumin concentrations, but they did not differ from the more rapid eliminators with respect to serum creatinine or urea nitrogen levels, SGOT, average age, female/male ratio, and average body weight. The serum protein binding of salicylate decreased with increasing age, apparently due mainly to decreasing serum albumin concentrations.Clinical Pharmacology and Therapeutics (1985) 38, 6–11; doi:10.1038/clpt.1985.124

The serum protein binding of three weakly acidic drugs (salicylic acid, sulfisoxazole, and Phenytoin), one weak base (diazepam), and one steroid (dexamethasone) was determined in pregnant women at seven time periods during pregnancy and at two time periods post partum, as well as in a group of nonpregnant women of childbearing age. The serum free fraction values (ratio of concentrations, free to total drug) of all drugs rose during pregnancy, primarily after 15 wk of gestation, and remained elevated for at least 1 to 5 days post partum. Pregnancy had the greatest effect on protein binding of sulfisoxazole, diazepam, and salicylic acid. The magnitude of this effect is such that quantitatively significant changes in the pharmacokinetic and pharmacodynamic characteristics of certain drugs may be expected to occur during pregnancy (in addition to possible changes caused by other pregnancy-related effects such as altered activity of drug-metabolizing enzyme systems). All drugs but dexamethasone exhibited significant negative correlations between free fraction values and serum albumin concentrations during pregnancy. The serum protein binding of salicylic acid, but not the other drugs tested, was more extensive in nonpregnant women who were not taking oral contraceptives than in those who were.Clinical Pharmacology and Therapeutics (1980) 28, 253–261; doi:10.1038/clpt.1980.158

The purposes of this investigation were to determine the magnitude of inter- and intraindividual variations in the relationship between heparin concentration and anticoagulant effect in normal adults, and to determine whether these variations are associated with, and therefore predictable from, certain physiologic characteristics of individual subjects. Citrated plasma was obtained from 12 men and 5 women, 21 to 35 yr old. Heparin was added to the plasma to yield concentrations of 0.05 to 1.0 U/ml and the activated partial thromboplastin time (APTT) was determined. These studies were repeated once or twice over 65 days. Baseline APTT values (i.e., APTT without added heparin) ranged from 25.6 to 36.2 sec and the hematocrit ranged from 39% to 50%. Both measures showed little intrasubject variation on the same day or on different days. There was an excellent linear relationship between in APTT and heparin concentration in the 0.05- to 0.8-U/ml range (r2> 0.987 in all cases). The slope value for this relationship ranged from 1.51 to 3.88 ml/U and these interindividual differences were well reproducible on repeated testing. Women had lower hematocrits (p < 0.05) and higher slope values (p < 0.01) than men. Multiple linear regression analysis revealed a linear relationship between observed slope values and slope values calculated as a function of both hematocrit and baseline APTT. Age, weight, and the concentrations of various plasma proteins did not contribute significantly to the predictability of the slope. A multiple linear regression equation with hematocrit and baseline APTT as independent variables yielded a multiple correlation coefficient of 0.875 (p < 0.01). Thus, it may be possible to predict the APTT value produced by a given concentration of heparin in an individual subject from that subject's baseline APTT and hematocrit.Clinical Pharmacology and Therapeutics (1980) 28, 509–516; doi:10.1038/clpt.1980.195

The serum protein binding of dexamethasone, diazepam, phenytoin, sulfisoxazole, and bilirubin was determined in full-term newborn infants and their mothers immediately after delivery. Diazepam and sulfisoxazole were bound more extensively in infant serum than in maternal serum, dexamethasone was bound more extensively in the maternal serum, and the serum protein binding of phenytoin and bilirubin was much the same in the infants and their mothers. These results are consistent with and explain clinical observations that diazepam concentrations are higher in cord than in maternal serum and that phenytoin levels in newborn and maternal sera are in the same range when the mothers had taken one of these drugs for some time before delivery. There were negative correlations between free fraction of drug and albumin concentration in serum only for phenytoin (infants and mothers), diazepam (infants only), and sulfisoxazole (mothers only). Except for dexamethasone (infants and mothers) and phenytoin (mothers), there were positive correlations between the free fraction values of the various compounds and those of salicylic acid. It was previously determined that salicylic acid is more extensively bound in newborn than in maternal serum. Induction or augmentation of labor with oxytocin was associated with decreased serum protein binding of diazepam and possibly of bilirubin in mothers. There was a significant correlation between serum free fraction values in newborns and their mothers only in the case of phenytoin.Clinical Pharmacology and Therapeutics (1980) 28, 58–63; doi:10.1038/clpt.1980.131

Activated charcoal is an effective inhibitor of acetaminophen absorption while sodium sulfate can prevent the depletion of endogenous inorganic sulfate associated with the formation of acetaminophen sulfate. Administration of activated charcoal plus sodium sulfate soon after acetaminophen overdose may reduce acetaminophen absorption and facilitate the elimination of absorbed acetaminophen by providing sufficient sulfate ion for rapid sulfation of the drug. This investigation was designed to determine if sodium sulfate modifies the inhibitory effect of activated charcoal on acetaminophen absorption or if activated charcoal affects the absorption of sodium sulfate. Eight normal adults received, on separate occasions, 1 g acetaminophen, 1 g acetaminophen and 18 g sodium sulfate (decahydrate), 1 g acetaminophen with 10 g activated charcoal, and 1 g acetaminophen with 10 g activated charcoal and 18 g sodium sulfate, in random order. Urine was collected for 48 hours and assayed for acetaminophen and its major metabolites and for inorganic sulfate. The results confirm that activated charcoal can reduce acetaminophen absorption and show that oral administration of activated charcoal with sodium sulfate does not alter the inhibitory effect of activated charcoal on acetaminophen absorption or the bioavailability of the sulfate. A combination of activated charcoal and sodium sulfate may therefore be useful for the initial management of acetaminophen overdose.

Controlled release pharmaceutical dosage forms may offer one or several advantages over conventional dosage forms of the same drug, including reduced dosing frequency, decreased incidence and/or intensity of adverse effects, greater selectivity of pharmacologic activity and a more constant therapeutic effect. In other instances, controlled release products may have no significant advantages or they may actually be less effective and/or more hazardous than conventional dosage forms of the same drug. In some cases, controlled release products may be therapeutically advantageous primarily for certain subpopulations of patients.

The serum protein binding of phenytoin, salicylic acid, sulfisoxazole, and warfarin was determined in normal human adults, in patients with impaired renal function (kidney donor and recipient), and in adult male Sprague-Dawley rats. The free fraction values for salicylate and sulfisoxazole were significantly correlated in all three groups. The other correlations were statistically significant in only one or two of these groups. There was a statistically significant negative correlation between albumin concentration and the free fraction values of salicylic acid and sulfisoxazole (but not of phenytoin and only under special circumstances with warfarin) in normal human subjects and of phenytoin, salicylic acid, and sulfisoxazole (but not warfarin) in rats. No such correlation was observed for any of the drugs in patients with impaired renal function. These observations show that no single weakly acidic drug can serve as an index for quantitatively determining the effect of disease or species differences on the serum protein binding of other weakly acidic drugs.

□ There are appreciable interindividual variations in rats of baseline activated partial thromboplastin time (APTT) and of the anticoagulant effect of heparin added to plasma (as reflected by the slope of the regression line describing the essentially linear relationship between In APTT and heparin concentration). Determination of baseline APTT and slope value on two occasions, 7 days apart, in the same rats revealed that (unlike in humans) these characteristics were subject also to considerable intraindividual variation. To explore the possible reasons for the observed variability, the effect of citrate concentration (acid citrate solution is used as a blood anticoagulant in the collection of plasma), calcium concentration (in the recalcifying solution used to initiate coagulation), and plasma incubation time (for activating the coagulation system) was determined. All three variables had pronounced effects on the anticoagulant response to heparin. Since rat erythrocytes are almost totally impermeable to citrate, hematocrit is a determinant of plasma citrate concentration when acid citrate solution is added in constant proportion to rat blood. Accordingly, inter- and intraindividual differences in baseline APTT and slope values were measured in another experiment in which the citrate solution to plasma (rather than blood) volume ratio was held constant and blood samples were obtained 30 days apart to permit the return of hematocrit values to normal. Intraindividual variation of the coagulation characteristics was appreciably decreased under these conditions. There are important differences between rats and humans with respect to the effect of citrate concentration and plasma incubation time on baseline APTT and on the anticoagulant action of heparin, as well as with respect to the relationship between these two characteristics.

The essential amino acidl-tryptophan has been widely used as a sleeping aid because it can produce drowsiness and decrease sleep latency. Its concentrations in plasma and brain and its binding to plasma protein are markedly altered in hepatic encephalopathy and renal failure. The purpose of this investigation was to determine ifl-tryptophan can enhance the sensitivity of the central nervous system to the hypnotic actions of a barbiturate and an alcohol. Female rats weighing ∼200 g received an intravenous infusion ofl-tryptophan (0.8 or 0.08 mg/min) for 30 min and then an infusion of phenobarbital (0.824 mg/min) withl-tryptophan (0.8 or 0.08 mg min−1) until the onset of loss of righting reflex (LRR). Control animals received an infusion of saline solution for 30 min and then phenobarbital without the amino acid. Similar experiments were performed with ethanol (16.3 mg/min), with and withoutl-tryptophan (0.8 mg/min).L-Tryptophan infused alone at a rate of 3.8 mg/min for 84 min did not cause LRR. Administration ofl-tryptophan at a rate of 0.8 mg/min with phenobarbital was associated with statistically significant reductions in the total dose and concentrations of phenobarbital in serum, serum water, brain, and cerebrospinal fluid (CSF) at onset of LRR. The 0.08 mg/min infusion ofl-tryptophan had a less pronounced effect, with statistically significant reductions of phenobarbital concentrations at onset of LRR in brain and CSF.L-Tryptophan also significantly reduced the total dose and the concentrations of ethanol in serum, brain, and CSF required to produce LRR. It is concluded thatl-tryptophan can potentiate the central nervous system depressant effects of phenobarbital and ethanol.

Pretreatment with cyclosporine reportedly prolongs the effect of certain general anesthetics in humans and the sleeping time of mice after pentobarbital administration. This investigation was designed to determine the mechanism(s) of the cyclosporine–barbiturate interaction. Adult female Wistar rats received cyclosporine (50 mg/kg im) or saline solution daily for 3 days. On the third day, they were injected with heptabarbital (45 mg/kg iv). Other cyclosporine-treated and control groups were infused with heptabarbital until they lost their righting reflex. Treatment for 3 d with cyclosporine was associated with decreased rectal temperature, decreased magnesium concentrations in serum and CSF, increased serum creatinine and urea nitrogen concentrations, elevated serum aspartate aminotransferase activity and total bilirubin concentration, decreased serum total protein concentration, and increased hematocrit. These physiologic changes are consistent with the clinically observed hypomagnesemia, nephrotoxicity, and hepatotoxicity in patients treated with cyclosporine. Control rats slept for 90 ± 14 min (mean ± SD, n = 9) after heptabarbital injection, whereas cyclosporine-pretreated rats slept for 154 ± 22 min. Compared with controls, cyclosporine-pretreated rats awoke (after heptabarbital injection) and went to sleep (after heptabarbital infusion) with significantly lower barbiturate concentrations in serum and CSF. Pretreatment with a single 60-mg/kg im dose of cyclosporine 2 h before heptabarbital infusion caused no significant biochemical changes ~160 min later, except for elevated serum aspartate aminotransferase (which occurred also after injection of the surfactant-containing vehicle) and serum bilirubin. Again, heptabarbital concentrations at onset of sleep (loss of righting reflex) in serum, brain, and CSF of cyclosporine-treated rats were significantly lower than in saline-treated controls. Two or three days of pretreatment with cyclosporine (50 mg/kg/day, im) had no significant effect on the total clearance of the barbiturate. These results indicate that cyclosporine can prolong barbiturate-induced general anesthesia by a pharmacodynamic mechanism.

To characterize the convulsant interaction between theophylline and caffeine, male Sprague‐Dawley rats received an iv infusion of one of seven different combinations of these drugs and of each drug individually until the onset of maximal seizures (which occurred within 30 to 40min after the start of the infusion). The total infused doses of the two drugs and their respective concentrations in the serum, brain, and cerebrospinal fluid (CSF) were used for isobolographic analysis. The results are consistent with classical dose‐ and concentration‐addition and do not suggest either antagonism or synergism. The potency ratio based on the doses or serum concentrations was appreciably different from that based on brain or CSF concentrations. The brain:serum and CSF:serum concentration ratios of caffeine were appreciably higher than those of theophylline. Similar experiments were performed with seven combinations of theophylline and pentylenetetrazol (PTZ) and with each of these drugs individually. This second set of experiments also yielded essentially linear isobolographs indicative of dose‐ and concentration‐addition. The potency ratio based on CSF concentrations was appreciably different from ratio values based on doses, and from those based on brain or serum concentrations. These results illustrate a useful strategy for the characterization of pharmacodynamic drug interactions and highlight the importance of the choice of sampling site for determinations of the potency of drug.