Lack of pharmacokinetic interaction between ropinirole and theophylline in patients with Parkinson's disease

Objective: Ropinirole and theophylline have the potential to interact, because they use the same hepatic cytochrome P450 (CYP1A2) as their major metabolic pathway. The present study investigated the effect of steady-state oral theophylline on the pharmacokinetics of ropinirole at steady state and the effect of steady-state ropinirole on the pharmacokinetics of a single intravenous (i.v.) dose of theophylline, both in patients with idiopathic Parkinson's disease (PD). Methods: Pharmacokinetic parameters (AUC and C_max) for i.v. theophylline were compared before and after a 4-week period of oral treatment with ropinirole (2 mg t.i.d.) in 12 patients with PD. Patients were then maintained at this dose of ropinirole, and oral theophylline was co-administered at doses of up to 300 mg b.i.d. The parameters AUC, C_max and t_max for ropinirole were compared before, during and after oral theophylline co-treatment. Results: Co-administration of ropinirole did not significantly change the pharmacokinetics of i.v. theophylline (mean AUC with and without ropinirole: 68.6 μg · h⁻¹ · ml⁻¹ and 70.0 μ· h⁻¹ · ml⁻¹, respectively; mean C_max with and without ropinirole: 11.07 μ g · ml⁻¹ and 11.83 μg · ml⁻¹, respectively). Similarly, there were no significant changes in ropinirole pharmacokinetics when the drug was co-administered with oral theophylline (mean AUC for ropinirole with and without theophylline: 21.91 ng · h⁻¹ · ml⁻¹ and 22.09 ng · h⁻¹ · ml⁻¹, respectively; mean C_max for ropinirole with and without theophylline: 5.65 ng · ml⁻¹ and 5.54 ng · ml⁻¹, respectively; median t_max for ropinirole with and without theophylline: 2.0 h and 1.5 h, respectively). Conclusion: These results suggest a lack of significant pharmacokinetic interaction between the two drugs at current therapeutic doses. Received: 10 August 1998 / Accepted in revised form: 27 January 1999     

Lack of a pharmacokinetic interaction between atovaquone and proguanil

Objective: To assess the magnitude of the putative effect of atovaquone on the pharmacokinetics of proguanil and to determine whether the pharmacokinetics of atovaquone are affected by concomitant administration of proguanil, with both drugs administered for 3 days to healthy adult volunteers. Methods: This was an open-label, randomized, three-way cross-over study, in which 18 healthy volunteers received 400 mg proguanil, 1000 mg atovaquone and 1000 mg atovaquone + 400 mg proguanil. Each treatment was given once daily for 3 days with a 3-week wash-out period between each occasion. For the assay of proguanil, cycloguanil and atovaquone, blood was sampled before dosing and at regular intervals over 8 days when proguanil was given, and over 17 days when atovaquone was given. Results: The geometric mean of the area under the atovaquone plasma concentration-time curve calculated from 0 to 24 h after the last dose (AUC_0→24h) was 180 μg · ml⁻¹ · h following administration of atovaquone alone and 193 μg · ml⁻¹ · h following atovaquone with proguanil. The geometric mean AUC_0→24h for proguanil was 6296 ng · ml⁻¹ · h after proguanil alone and 5819 ng · ml⁻¹ · h following proguanil with atovaquone. The corresponding values for the metabolite cycloguanil were 1297 ng · ml⁻¹ · h and 1187 ng · ml⁻¹ · h, respectively. The geometric mean elimination half-life (t_1/2) of atovaquone was 57.1 h when given alone and 59.0 h when administered together with proguanil. The corresponding geometric mean values of t_1/2 for proguanil were 13.7 h and 14.5 h. Exploratory statistical analysis showed no important gender effects on the pharmacokinetics of atovaquone, proguanil, or cycloguanil. Conclusion: The pharmacokinetics of atovaquone and proguanil and its metabolite, cycloguanil, were not different when atovaquone and proguanil were given alone or in combination. Received: 14 October 1998 / Accepted in revised form: 8 February 1999     

Pharmacokinetic and pharmacodynamic evaluation of fluticasone propionate after inhaled administration

Objective: To evaluate the pharmacokinetic and systemic pharmacodynamic properties of inhaled fluticasone propionate (FP). Methods: Single doses of 0.25, 0.5, 1.0 and 3.0 mg FP were administered to groups of six healthy subjects. Serum concentration profiles of FP were monitored over 24 h by means of high-performance liquid chromatography/mass spectrometry (HPLC/MS–MS). Systemic pharmacodynamic effects were evaluated by measuring endogenous serum cortisol and circulating white blood cells, and analyzed with previously developed integrated pharmacokinetic/pharmacodynamic (PK/PD) models. Results: FP showed a dose-independent terminal half-life with a mean (SD) of 6.0 (0.7) h. Maximum serum concentrations occurred 1.0 (0.5) h after administration, ranging from 90 pg · ml⁻¹ for the 0.25 mg dose to 400 pg · ml⁻¹ for the 3.0 mg dose. This, together with an estimated mean absorption time of nearly 5 h and a known oral bioavailability of less than 1%, indicates prolonged residence at and slow absorption from the lungs. In the investigated dose range, the cumulative systemic effect was dose-dependent for both markers of pharmacodynamic activity. For doses of 0.25, 0.50, 1.0 and 3.0 mg FP, the PK/PD-based cumulative systemic-effect parameters were 159, 186, 257 and 372% · h for lymphocyte suppression, 107, 186, 202 and 348% · h for granulocyte induction and 23.6%, 33.8%, 51.0% and 73.6% for cortisol reduction, respectively. The time courses of lymphocytes, granulocytes and endogenous cortisol could be sufficiently characterized with the applied PK/PD models. The measured in vivo EC₅₀ values, 30 pg · ml⁻¹ and 7.3 pg · ml⁻¹ for white blood cells and cortisol, respectively, were in good agreement with predictions based on the in vitro relative receptor affinity of FP. Conclusion: After inhalation, FP follows linear pharmacokinetics and exhibits dose-dependent systemic pharmacodynamic effects that can be described by PK/PD modeling. Received: 27 January 1997 / Accepted in revised form: 5 August 1997     

Pharmacokinetics of oxypolygelatine in healthy volunteers

Objective/methods: The pharmacokinetics of the plasma substitute oxypolygelatine (OPG) were studied in 12 healthy volunteers after single-dose administration of 27 ml · kg⁻¹ body weight, with a maximum of 2000 ml. OPG was determined in plasma and urine over 48 h after the infusion. Peak plasma OPG concentrations at the end of the infusion were determined to 4.600 (623) μg · ml⁻¹, the area under the plasma concentration/time curve (AUC_0∞) was calculated to 70.135 (15.861) μg · h · ml⁻¹. Results: The model-independently calculated volume of distribution came to 23.1 (4.8) l with a clearance total is (Cl_tot) of 24.6 (6.8) ml · min⁻¹. The initial half-life according to a three-compartment model came to 0.3 (0.2) h, followed by a distribution half-life of 3.1 (2.6) h and a terminal elimination half-life of 13.4 (2.2) h. Cumulative urinary excretion of OPG was 64% after 48 h. Conclusion: This low recovery rate may be explained by the distribution of OPG into the extravascular space and subsequent degradation in tissue sites. Received: 9 June 1998 / Accepted in revised form: 23 November 1998     

Penetration of ciprofloxacin, norfloxacin and ofloxacin into the aqueous humour of patients by different topical application modes

Objectives: A prospective study was undertaken to determine the transcorneal penetration of three topically applied fluoroquinolones into aqueous humour. Methods: Two hundred and twenty-four patients undergoing cataract extraction received 0.3% ciprofloxacin, norfloxacin or ofloxacin eye drops by two different administration modes with different frequencies and intervals of application. At the beginning of cataract extraction (0.5–3 h after the last drop), 50–100 μl aqueous fluid was aspirated from the anterior chamber and immediately stored at −80 °C. Antibiotic concentrations were measured using high-performance liquid chromatography. Results: Generally, topical ofloxacin and ciprofloxacin yielded aqueous humour levels higher than topical norfloxacin. The highest concentrations of all tested fluoroquinolones were measured after using an application mode, in which one drop was given every 15 min between 0600 hours and 0800 hours, prior to operation. When applied by this mode, ciprofloxacin achieved a mean aqueous level of 0.380 (±0.328) μg · ml⁻¹ (range 0.033–1.388 μg · ml⁻¹), norfloxacin 0.182 (0.118) μg · ml⁻¹ (range 0.038–0.480 μg · ml⁻¹) and ofloxacin 0.564 (0.372) μg · ml⁻¹ (range 0.064–1.455 μg · ml⁻¹). These mean concentrations were above the minimum inhibitory concentration (MIC₉₀), concentrations required for inhibition of 90% of pathogen strains in vitro of gram-negative bacteria, such as Proteus mirabilis and Escherichia coli. Therapeutic values above the MIC₉₀ of Staphylococcus epidermidis, the pathogen causing eye infections most frequently, were reached by 67.5% of patients after ofloxacin and by 41% after ciprofloxacin, but never after norfloxacin treatment. Conclusion: Of the currently available topical fluoroquinolones, ofloxacin achieved the highest aqueous humour concentration. This fluoroquinolone may be an useful ophthalmic agent for topical antibacterial management, but it does not seem to be prophylactically effective against Streptococcus pneumoniae or Pseudomonas aeruginosa. Received: 22 April 1997 / Accepted: 8 June 1997     

Improved penetration of aminoglycosides and fluoroquinolones into the aqueous humour of patients by means of Acuvue contact lenses

Objectives: In order to improve the penetration of topically applied drugs in ophthalmology, the suitability of hydrophilic contact lenses (Acuvue, Vistacon, power −1.0 D) as a drug delivery system for antibiotics was tested. A prospective study was undertaken to determine the transcorneal penetration of five topically applied aminoglycosides and fluoroquinolones into the aqueous humour of patients. Methods: Two hundred and sixty-five patients undergoing cataract extraction received 0.3% gentamicin, kanamycin, tobramycin, ciprofloxacin or ofloxacin solution by two different modes of administration: either as eye drops (nine drops every 15 min, starting 2 h prior to surgery) or by means of a drug delivery system (Acuvue contact lenses soaked for 1 h in eye drop solution without preservatives, 1–5 h prior to surgery). At the beginning of cataract extraction, 50–100 μl aqueous fluid was aspirated from the anterior chamber and immediately stored at −80 °C. Antibiotic concentrations were measured using fluorescence polarisation immuno-assays (aminoglycosides) or high-performance liquid chromatography (fluoroquinolones). Results: After soaking for 1 h in 0.3% eye drop solutions, Acuvue contact lenses released about 190–250 μg aminoglycoside and ofloxacin and 1000 μg ciprofloxacin. These amounts are considerably lower or in the same order of magnitude than obtained with application of eye drops (1350 μg). From the aminoglycosides tested, only gentamicin and tobramycin, but not kanamycin, were able to penetrate into the aqueous humour of patients. After the wearing of antibiotic-soaked lenses, mean aqueous humour concentrations were higher than after the use of eye drops. This difference reached significance in tobramycin (1.09 (1.30) μg · ml⁻¹ vs 0.49 (0.79) μg · ml⁻¹), ciprofloxacin (1.23 (0.60) μg · ml⁻¹ vs 0.38 (0.33) μg · ml⁻¹) and ofloxacin (5.55 (2.53) μg · ml⁻¹ vs 0.56 (0.37) μg · ml⁻¹). The percentage of patients with aqueous humour concentration above the MIC₉₀ of Staphylococcus epidermidis, the most common cause of postoperative endophthalmitis, was 92% and 100% after wearing ciprofloxacin- or ofloxacin-soaked lenses, respectively. Conclusion: Gentamicin and tobramycin penetrated into the aqueous humour of patients, whereas kanamycin was not able to overcome the corneal barrier. Acuvue contact lenses soaked in 0.3% eye drop solutions can release sufficient amounts of gentamicin, ciprofloxacin and ofloxacin to produce bacteriostatic concentrations in the humor aquosus. Acuvue contact lenses can be recommended as a drug delivery system for fluoroquinolones. Received: 15 October 1998 / Accepted in revised form: 16 December 1998     

The effect of the apolipoprotein E phenotype on cholesteryl ester transfer protein activity, plasma lipids and apolipoprotein A I levels in hypercholesterolaemic patients on colestipol and lovastatin treatment

Background: Apolipoprotein E (apo E) allele E 4 is associated with high atherogenic lipid levels and coronary heart disease. Cholesteryl ester transfer protein (CETP) transfers cholesteryl esters from (high density lipoprotein) HDL to other lipoproteins. CETP gene expression is enhanced in hypercholesterolaemia and correlates with plasma apo E concentration. Objective: The effect of the apo E phenotype on plasma CETP activity and the hypolipidaemic efficacy of colestipol and lovastatin was studied in patients with type II a or II b hypercholesterolaemia. Results: The baseline mean plasma total, low density lipoprotein (LDL) and HDL cholesterol, triglyceride, apolipoprotein A I (apo A I) concentrations and CETP activity were 8.89 mmol · l⁻¹, 6.78 mmol · l⁻¹, 1.39 mmol · l⁻¹, 1.59 mmol · l⁻¹, 1.49 g · l⁻¹ and 114 nmol · h⁻¹ · ml⁻¹, respectively. The colestipol-induced changes were −26%, −36%, +5%, +12%, −1% and −17%, and the lovastatin-induced changes −34%, −44%, +6%, −18%, +1% and −19%. The lipid and apo A I concentrations or the CETP activity did not differ statistically significantly according to the apo E phenotype, although the HDL cholesterol and apo A I levels were lowered in patients with apo E 4/4 but elevated in patients with the other phenotypes. The CETP activity correlated with the LDL cholesterol concentration (r = 0.52, P = 0.01) and the change in the LDL cholesterol during colestipol (r = 0.51, P = 0.02) and lovastatin (r = 0.65, P = 0.001) treatment, but only in patients without the apo E 4 allele. Conclusion: Colestipol and lovastatin reduced CETP activity to the same amount, regardless of the apo E phenotype. The apo E phenotype seems to modify the interaction between CETP activity and LDL cholesterol in hypercholesterolaemia and during pharmacological lowering of cholesterol. Received: 13 May 1998 / Accepted in revised form: 3 October 1998     

Pharmacokinetics of glibenclamide and its metabolites in diabetic patients with impaired renal function

Background: Glibenclamide (Gb) may provoke long-lasting hypoglycaemic reactions, and one of the known risk factors is impaired renal function. We have demonstrated Gb to have a terminal elimination half-life of 15 h, and the main metabolites have a hypoglycaemic effect. With few exceptions, detailed studies on second generation sulphonylureas in diabetics with impaired renal function are lacking. Therefore, we analysed the pharmacokinetics of Gb and its active metabolites, 4-trans-hydroxyglibenclamide (M1) and 3-cis-hydroxyglibenclamide (M2) in this patient group. Methods: Two groups of 11 diabetic patients with impaired renal function (IRF, iohexol clearance range 7–42 ml · min⁻¹ · 1.73 m⁻²) or normal renal function (NRF, iohexol clearance range 75–140 ml · min⁻¹ · 1.73 m⁻²) were compared. A single oral 7-mg dose of Gb was administered after overnight fasting. Serum samples and urine collections were obtained over 48 h and 24 h, respectively. Concentrations of Gb, M1 and M2 were determined by a sensitive and selective high-performance liquid chromatography assay. Results: Peak serum values of M1 (24–85 ng · ml⁻¹ vs 16–57 ng · ml⁻¹), M2 (7–22 ng · ml⁻¹ vs <5–18 ng · ml⁻¹) and M1 + M2 (32–100 ng · ml⁻¹ vs 23–76 ng · ml⁻¹) were higher in the IRF group. AUC and C_max of Gb were lower and the clearance to bioavailability ratio (CL/f) was higher in the IRF group. AUC and C_max of M1 were higher and CL/f lower in the IRF group. Much lower amounts of M1 and M2 were excreted in the urine in the IRF group (7.2% vs 26.4% in 24 h). The fraction of the Gb dose excreted as metabolites (fe(met) 0–24 h), ranged between 0.005 and 0.36 and correlated significantly with renal function measured by iohexol clearance. No other pharmacokinetic differences were found. Conclusion: The differences in AUC, C_max and CL/f of Gb may be explained by a higher free fraction in the IRF group which would increase Gb metabolic clearance. The inverse findings regarding M1 may be explained by the fact that the metabolites are primarily eliminated by the kidneys. After a single dose of Gb, neither Gb, M1 nor M2 seemed to accumulate in diabetic subjects with IRF. As only small amounts of M1 and M2 were excreted in the urine, this indicates one or several complementary non-renal elimination routes, e.g. shunting of metabolised Gb to the biliary excretion route and/or enterohepatic recycling of both metabolites and unmetabolised Gb. Received: 21 April 1997 / Accepted in revised form: 14 October 1997     

Plasma esterases in cystic fibrosis: the impact of a respiratory exacerbation and its treatment

Objectives: To determine the effect of an exacerbation of respiratory symptoms in cystic fibrosis (CF) on the activities of plasma benzoylcholinesterase and butyrylcholinesterase. Methods: Twenty-nine patients with CF in a respiratory exacerbation and 27 healthy volunteers matched for age and sex were recruited. Blood was obtained from the patients when commencing antibiotic treatment and 14 days later on completion of treatment. One blood sample was taken from the healthy volunteers. The activities of benzoylcholinesterase and butyrylcholinesterase were determined by spectrophotometric assay. The circulating inflammatory markers, C-reactive protein and neutrophil elastase-α₁antiproteinase complex were also measured. Results: Benzoylcholinesterase activity was significantly (P = 0.001) lower in patients at the start of a respiratory exacerbation, compared with healthy controls [mean (SD): 917 (274) versus 1191(298) nmol · ml⁻¹ · min⁻¹]. Benzoylcholinesterase activity increased significantly in patients to 1013 (237) nmol · ml⁻¹ · min⁻¹, following a course of antibiotic treatment (P = 0.006). Butyrylcholinesterase activity was also lower (P = 0.001) in patients at the start of a respiratory exacerbation, compared with healthy controls [5.54 (1.64) versus 7.01 (1.79) μmol · ml⁻¹ · min⁻¹], and increased significantly in the patients to 6.31 (1.58) μmol · ml⁻¹ · min⁻¹ following treatment (P = 0.006). Conclusion: We demonstrated significant suppression of plasma esterase activities during an exacerbation of respiratory symptoms in CF, which was only partially reversed after antibiotic treatment. Further studies are needed to examine other pathways of drug metabolism in this group of chronically infected patients. Received: 8 June 1998 / Accepted in revised form: 18 September 1998     

Helicobacter pylori clearance and serum gastrin and pepsinogen I concentrations in omeprazole treatment of duodenal ulcer patients

Objective: To determine which demographic factors may influence serum gastrin and pepsinogen I (PGI) levels in duodenal ulcer patients undergoing omeprazole treatment. Methods: We conducted an outpatient-based prospective study in the Veterans General Hospital, Taipei, to investigate the pharmacological effects on patients with duodenal ulcers receiving omeprazole treatment for 4 weeks. Sixty-eight patients (61 males/7 females, aged 25–73 years) with endoscopically confirmed duodenal ulcer were included. Gastrin and pepsinogen I levels were measured before and after treatment. Demographic factors including age, sex, smoking, ulcer healing and antral Helicobacter pylori colonization/clearance were analyzed, in order to measure their probable influences on serum gastrin and pepsinogen I levels. Results: Ulcer healing was seen in 92.6% of patients while 48 (70.6%) antral clearances were seen in 66 H. pylori colonized patients at the end of trial. Omeprazole monotherapy led to a marked elevation of serum gastrin (85.8 pg · ml⁻¹, SD 32.0 pg · ml⁻¹ vs 133.9 pg · ml⁻¹, SD 71.6 pg · ml⁻¹, P < 0.01), and pepsinogen I (111.0 ng · ml⁻¹, SD 36.7 ng · ml⁻¹ vs 253.6 ng · ml⁻¹, SD 64.8 ng · ml⁻¹, P < 0.01) levels when measured on day 29. Only patients showing antral H. pylori clearance exhibited an influence on the magnitude of pepsinogen I elevation following omeprazole monotherapy (143.9%, SD 67.3% vs 78.6%, SD 51.2%, P < 0.01). Moreover, the sensitivity and specificity of serum pepsinogen I variations were plotted on a receiving operating characteristic (ROC) curve. The 140% increased pepsinogen I level yielded a maximum accuracy of 80% specificity or 50% sensitivity to predict antral H. pylori clearance. Conclusion: Antral H. pylori clearance is at least partially responsible for the omeprzaole-induced hyperpepsinogenemia I. The magnitude of hyperpepsinogenemia I probably provides a non-invasive alternative for predicting H. pylori clearance. Received: 22 August 1996 / Accepted in revised form: 1 October 1998     

Population analysis of the non-linear red blood cell partitioning of draflazine following various infusion durations

Objective: The pharmacokinetics and non-linear red blood cell partitioning of the nucleoside transport inhibitor draflazine were investigated in 19 healthy male and female subjects (age range 22–55 years) after a 15-min i.v. infusion of 1 mg, immediately followed by infusions of variable rates (0.25, 0.5 and 1 mg · h⁻¹) and variable duration (2–24 h). Methods: The parameters describing the capacity-limited specific binding of draflazine to the nucleoside transporters located on erythrocytes were determined by NONMEM analysis. The red blood cell nucleoside transporter occupancy of draflazine (RBC occupancy) was evaluated as a pharmacodynamic endpoint. Results: The population typical value for the dissociation constant K _d (%CV) was 0.648 (12) ng · ml⁻¹ plasma, expressing the very high affinity of draflazine for the erythrocytes. The typical value of the specific maximal binding capacity B_max (%CV) was 155 (2) ng · ml⁻¹ RBC. The interindividual variability (%CV) was moderate for K _d (38.9%) and low for B_max (7.8%). As a consequence, the variability in RBC occupancy of draflazine was relatively low, allowing the justification of only one infusion scheme for all subjects. The specific binding of draflazine to the red blood cells was a source of non-linearity in draflazine pharmacokinetics. Steady-state plasma concentrations of draflazine virtually increased dose-proportionally and steady state was reached at about 18 h after the start of the continuous infusion. The t_1/2βaveraged 11.0–30.5 h and the mean CL from the plasma was 327 to 465 ml · min⁻¹. The disposition of draflazine in whole blood was different from that in plasma. The mean t_1/2β was 30.2 to 42.2 h and the blood CL averaged 17.4–35.6 ml · min⁻¹. Conclusion: Although the pharmacokinetics of draflazine were non-linear, the data of the present study demonstrate that draflazine might be administered as a continuous infusion over a longer time period (e.g., 24 h). During a 15-min i.v. infusion of 1 mg, followed by an infusion of 1 mg · h⁻¹, the RBC occupancy of draflazine was 96% or more. As the favored RBC occupancy should be almost complete, this dose regimen could be justified in patients. Received: 6 February 1997 / Accepted in revised form: 12 May 1997     

Variability of morphine disposition during long-term subcutaneous infusion in terminally ill cancer patients

Objective: To study the plasma concentrations of morphine and its glucuronides to assess the intra- and interindividual variability of the disposition of morphine administered by subcutaneous infusion in cancer patients. Methods: Blood samples were taken repeatedly in eight patients with severe cancer pain who were being treated with morphine (60–3000 mg per day) via chronic (8–160 days) subcutaneous infusion. Venous blood samples were collected at least weekly and, when possible, on 3 consecutive days after dose adaptation or any other major change in the patients' treatment. Concentrations of morphine and its glucuronides in plasma were measured after solid-phase extraction using a validated high-performance liquid chromatography assay. The stability of the morphine solutions was determined by repeated measurement of the concentrations of morphine and its degradation products in the solutions. Results: The morphine concentration in the infusion solutions remained unchanged during storage and infusion. The plasma concentrations of morphine and its glucuronides were within the ranges reported in the literature. There was, as expected, a large interindividual variability: from patient to patient, the mean of the normalised plasma concentrations ranged from 0.3 ng · ml⁻¹ · mg⁻¹ to 0.8 ng · ml⁻¹ · mg⁻¹ for morphine, from 1.0 ng · ml⁻¹ · mg⁻¹ to 3.1 ng · ml⁻¹ · mg⁻¹ for morphine-6-glucuronide and from 6.8 ng · ml⁻¹ · mg⁻¹ to 24.3 ng · ml⁻¹ · mg⁻¹ for morphine-3-glucuronide. Intraindividual variability was also important. The residual standard deviation of the mean normalised plasma concentrations calculated for each patient ranged from 26% to 56% for morphine, from 20% to 51% for morphine-6-glucuronide and from 20% to 49% for morphine-3-glucuronide. The normalised plasma concentrations of morphine and its glucuronides did not increase with dose or time, and no explanation for the pronounced pharmacokinetic intraindividual variability was found. Conclusion: During subcutaneous infusion of morphine, there is a large intra- and interindividual variability of the morphine disposition which could be of clinical relevance. Received: 5 August 1997 / Accepted in revised form: 8 October 1997     

Pharmacokinetics and pharmacodynamics of acetazolamide in patients with transient intraocular pressure elevation

Objective: To characterize the pharmacokinetics and pharmacodynamics of acetazolamide in patients with transient intraocular pressure (IOP) elevation and to provide individual patients with the optimal dosage regimen for this drug. Methods: We studied 17 patients with transient IOP elevation, who were given 62.5–500 mg acetazolamide orally as single or repetitive doses. Plasma acetazolamide concentration and IOP were measured at approximately 1, 3, 5, and 9 h after the last acetazolamide administration. Pharmacokinetics and pharmacodynamics were analyzed by nonlinear mixed-effect modeling using the program NONMEM. Results: The plasma concentration profile of acetazolamide was characterized by a one-compartment model with first-order absorption. The apparent oral clearance was related to the creatine clearance (CCR) which was estimated by the Cockcroft and Gault equation, as follows: 0.0468 · CCR l · h⁻¹. The estimated apparent oral volume of distribution, first-order absorption rate constant, and absorption lag time were 0.231 l · kg⁻¹, 0.821 · h⁻¹, and 0.497 h, respectively. IOP after oral acetazolamide administration was characterized by an E_max model. The maximal effect in lowering the IOP (E_max) was 7.2 mmHg, and the concentration corresponding to 50% of the maximal effect (EC₅₀) was 1.64 μg · ml⁻¹. As 70% of E_max was achieved at a plasma concentration of 4 μg · ml⁻¹, this concentration was considered satisfactory for lowering IOP. The recommended dosage was calculated so that the minimum plasma concentration at steady state exceeded this target concentration; 250 mg t.i.d., 125 mg t.i.d., 125 mg b.i.d., and 125 mg once daily for the patients with CCR values of 70, 50, 30, and 10 ml · min⁻¹, respectively. Conclusion: Measuring plasma concentrations of acetazolamide and subsequent pharmacokinetic and pharmacodynamic analyses are useful for estimating its concentration-dependent effectiveness in lowering the IOP in individual patients. The dosage regimen presented in this study is expected to improve the benefits of acetazolamide pharmacotherapy in most elderly patients with transient rises in IOP following intraocular surgery. Received: 10 April 1997 / Accepted in revised form: 21 October 1997     

Distribution and metabolism of N G-nitro-l-arginine methyl ester in patients with septic shock

Objective: The pharmacokinetics of N ^G-nitro-l-arginine methyl ester (l-NAME), an inhibitor of nitric oxide (NO) synthesis, was investigated in patients with septic shock. Methods: Blood was sampled at intervals before, during and after 12-h infusion of l-NAME 1 mg · kg⁻¹ · h⁻¹ in nine septic shock patients for determination of plasma concentrations by high-performance liquid chromatography (HPLC). In three patients the renal clearance of the drug was determined. Results: Incubation of l-NAME with plasma and blood in vitro revealed hydrolysis to N ^G-nitro-l-arginine (l-NOARG), the active inhibitor of NO synthesis. l-NOARG did not undergo further degradation. Continuous intravenous infusion of 1 mg · kg⁻¹ · h⁻¹ of l-NAME for 12 h in patients with septic shock increased blood pressure and resulted in increasing plasma concentrations of l-NOARG (C_max 6.2 μg · ml⁻¹ at 12 h) whereas l-NAME concentrations reached a plateau within 1.5 h (C_max 1.0 μg · ml⁻¹). After the infusion was stopped l-NAME disappeared from the plasma rapidly (half-life 19.2 min) whereas l-NOARG concentration declined slowly (half-life 22.9 h). The calculated volume of distribution for l-NAME was 0.45 l · kg⁻¹ body weight and 1.96 l · kg⁻¹ for l-NOARG. The renal clearance for l-NOARG was 3.5% of total body clearance for l-NOARG, whereas l-NAME could not be detected in urine. Conclusion: We conclude that vasoconstriction with l-NAME in septic patients may result from hydrolysis to l-NOARG, the active inhibitor of NO synthesis. The long plasma half-life and large volume of distribution for l-NOARG suggests extensive distribution to extravascular tissues. Since renal excretion is minimal, elimination of the metabolite l-NOARG follows other pathways. Received: 13 March 1998 / Accepted in revised form: 30 June 1998     

Temperature dependency of the release and bioavailability of nicotine from a nicotine vapour inhaler; in vitro/ in vivo correlation

Objective: To investigate the temperature dependency of the dose released and the plasma levels of nicotine from a vapour inhaler. Methods: In an open, randomised, three-way cross-over pharmacokinetic study 18 healthy subjects inhaled nicotine for 20 min (80 inhalations) every hour for 10 h (11 administrations) at three different environmental temperatures: 20°, 30° and 40 °C. In the in vitroexperiment, 5, 10, 15 and 20 l air were forced through the inhaler. With a 15 l air volume, the average amount of nicotine released was 1.44, 3.49, 4.80 and 6.99 mg at 10 °C, 22 °C, 29 °C and 40 °C, respectively. The maximum dose released at the highest temperature (40 °C) and the largest air volume investigated (20 l) was approximately 7.5 mg. Results: In vivo peak plasma levels obtained at 30° and 40 °C were 29.7 and 34.0 ng · ml⁻¹, compared with 22.5 ng · ml⁻¹ at ambient room temperature (20 °C). At 20 °C, the area under the plasma concentration–time curve (AUC) of the last dosing interval was 20.5 ng · ml⁻¹ · h. At 30 °C and 40 °C, the AUCs were 26.5 and 30.3 ng · ml⁻¹ · h, respectively. The results thus showed a mean increase of the in vivo AUC by 29% at 30 °C and by 48% at 40 °C compared with the AUC at 20 °C. These increases should be compared to the in vitro results, showing a mean increase of 59% and 122%, respectively, at 30° and 40 °C. The in vitro results also showed that a relatively larger fraction of the dose was released into the first 5 l of air at the higher temperatures, at 40 °C, about 50% of the total amount released into 20 l. Conclusion: It was concluded that the in vitro/in vivo discrepancy was most probably due to increased aversive effects at elevated temperatures, causing the subjects to inhale smaller puff volumes. Further, the inhaler would not produce nicotine plasma levels exceeding those achieved following cigarette smoking, even in a hot climate. Received: 20 September 1996 / Accepted in revised form: 5 March 1997     

Pharmacokinetics and safety of candesartan cilexetil in subjects with normal and impaired liver function

Objective: The influence of liver disease on the pharmacokinetics of candesartan, a long-acting selective AT₁ subtype angiotensin II receptor antagonist was studied. Methods: Twelve healthy subjects and 12 patients with mild to moderate liver impairment received a single oral dose of 12 mg of candesartan cilexetil on day 1 and once-daily doses of 12 mg on days 3–7. The drug was taken before breakfast. Serial blood samples were collected for 48 h after the first and last administration on days 1 and 7. Serum was analyzed for unchanged candesartan by HPLC with UV detection. Results: The pharmacokinetic parameters on days 1 and 7 revealed no statistically significant influence of liver impairment on the pharmacokinetics of candesartan. Following single dose administration on day 1, the␣mean␣C_max was 95.2 ng · ml⁻¹ in healthy subjects and 109 ng · ml⁻¹ in the patients. The AUC_0−∞ was␣909 ng.h · ml⁻¹ in healthy volunteers and 1107 ng.h · ml⁻¹ in patients and the elimination half-life was 9.3 h in healthy volunteers and 12 h in the patients. At steady state on day 7, mean C_max values were similar in both groups (112 vs 116 ng · ml⁻¹); the AUC_τ was 880 ng.h · ml⁻¹ in healthy subjects and 1080 ng.h · ml⁻¹ in patients while the elimination half-life was 10 h in healthy subjects and 12 h in the patients with liver impairment. The AUC_0−∞ on day 1 was almost identical to the AUC_τ on day 7. A moderate drug accumulation of 20%, which does not require a dose adjustment, was observed following once-daily dosing in both groups. No serious or severe adverse events were reported. Conclusion: Mild to moderate liver impairment has no clinically relevant effect on candesartan pharmacokinetics, and no dose adjustment is required for such patients. Received: 24 November 1997 / Accepted in revised form: 18 February 1998     

Comparative pharmacokinetics of dirithromycin and erythromycin in normal volunteers with special regard to accumulation in polymorphonuclear leukocytes and in saliva

Objective: In a randomized cross-over study, we assessed pharmacokinetics and intracellular concentrations in polymorphonuclear leukocytes (PMN) and saliva of erythromycin and erythromycylamine, the active metabolite of dirithromycin. Methods: Ten healthy volunteers received 1 g erythromycin b.i.d. or 500 mg dirithromycin qd for 5 days (wash out period, 35 days). Concentrations of erythromycin and erythromycylamine were measured in serum, urine, saliva, and granulocytes by bioassay and high-performance liquid chromatography (HPLC) on days 1, 3, and 5 of each study period, respectively. Results: While maximal serum concentrations (C_max) and the area under the data (AUD_tot) of erythromycin were significantly higher (C_max 1.44 mg · l⁻¹, AUD_tot 5.66 mg · h · l⁻¹) than those of erythromycylamine (C_max 0.29 mg · l⁻¹, AUD_tot 1.96 mg · h · l⁻¹), erythromycylamine had a significantly higher mean residence time (21 h) than erythromycin (5.5 h). Erythromycylamine accumulated significantly more in PMN than erythromycin;␣the accumulation factor of erythromycylamine was 100 with a maximal intracellular concentration of 13.4 mg · l⁻¹, whereas the maximal accumulation factor of erythromycin was 4 with a maximal intracellular concentration of 6.1 mg · l⁻¹. There were no significant differences in maximal saliva concentrations (erythromycin 0.35 mg · l⁻¹, erythromycylamine 0.31 mg · l⁻¹). Received: 16 September 1996 / Accepted in revised form: 12 February 1997     

Renal handling of drugs in the healthy elderly Creatinine clearance underestimates renal function and pharmacokinetics remain virtually unchanged

Objective: It is commonly assumed that renal function, and in parallel the excretion of drugs, is considerably reduced in the elderly. Endogenous creatinine clearance or indirect estimates of this parameter are generally recommended for adapting drug dosage. The present study evaluates the validity of both assumptions. Methods: We compared pharmacokinetics (and pharmacodynamics) of 50 mg atenolol, 800 mg piracetam and 25 mg hydrochlorothiazide plus 50 mg triamterene in ten healthy young [25 (2) years] and 11 healthy elderly subjects [68 (5) years]. Inulin (C_in) and para-aminohippurate [PAH (C_PAH)] clearance (infusion clearance technique), endogenous (C_Cr) and calculated (Cockroft-Gault) creatinine clearance, analysis of drugs and their metabolites (HPLC), were performed. Renal haemodynamics and the pharmacokinetics of β-adrenergic blocking agent, diuretics and the nootropic agent piracetam, respectively, were measured on separate days. Results: C_in was significantly (P < 0.01) lower in the healthy elderly subjects [104 (12) vs 120 (14) ml · min⁻² · 1.73 m⁻² in the young], but remained within the normal range (>90 ml · min⁻² · 1.73 m⁻²). In contrast, C_Cr was even lower in healthy elderly subjects [95 (24) vs 121 (20) ml · min⁻¹ in the young], and the Cockroft-Gault clearance underestimated true glomerular filtration rate (GFR) even more seriously [74 (17) vs 122 (16) ml · min⁻¹]. For atenolol the mean area under the curve (AUC) was similar in both groups [3.16 (0.48) μg · h⁻¹ · ml⁻¹ in the elderly vs 3.01 (0.30) in the young], as was the mean maximal plasma concentration [0.42 (0.07) vs 0.44 (0.06) μg · ml⁻¹], but the proportion of the drug excreted in urine was marginally (P < 0.025) lower in the elderly. Similar results were obtained for hydrochlorothiazide, whereas no marked differences between the groups were found for triamterene and its metabolite. Furthermore, the pharmacodynamic action of diuretics was not significantly altered in the elderly. Conclusions: The true GFR of the healthy elderly remains within the normal range and is underestimated by creatinine clearance and more so by its surrogate (Cockroft-Gault clearance). In parallel, pharmacokinetics of renally excreted drugs are not affected in the healthy elderly to a clinically significant extent. For drugs with a narrow therapeutic window, indirect estimates of GFR appear to be an unreliable means for calculating correct dosage in the elderly. Received: 2 April 1998 / Accepted in revised form: 19 October 1998     

The cardiac effects of terfenadine after inhibition of its metabolism by grapefruit juice

Objective: To determine whether the pharmacokinetics and electrocardiographic pharmacodynamics of terfenadine are affected by the concomitant administration of grapefruit juice. Methods: Six healthy volunteers were recruited for a balanced cross-over study. Each volunteer received 120 mg terfenadine 30 min after drinking 300 ml of either water or freshly squeezed grapefruit juice. The alternative treatment was administered on the second study day 2 weeks later. Measurements of the area under the terfenadine plasma concentration-time curve (AUC), maximum terfenadine concentration (C_max) and the time to maximum concentration (t_max) were made, and the corrected QT (QTc) interval was measured from the surface electrocardiogram. Results: Terfenadine was quantifiable in plasma in all 6 subjects on both study days for up to 24 h post-dosing. The AUC of terfenadine was significantly increased by concomitant grapefruit administration (median values 40.6 vs 16.3 ng · ml⁻¹ · h), as was the C_max (median values 7.2 vs 2.1 ng · ml⁻¹). The t_max was not significantly increased and there was no significant change in the median QTc interval despite the increased terfenadine levels. The 95% confidence interval for the difference in the change in QTc interval at C_max was −13 to +38 ms. Conclusion: Administration of grapefruit juice concomitantly with terfenadine may lead to an increase in terfenadine bioavailability, but the increase observed in this study did not lead to significant cardiotoxicity in normal subjects. However, this does not exclude the risk of cardiotoxicity in high-risk subjects given greater doses of grapefruit juice over longer periods of time. Received: 14 October 1996 / Accepted in revised form: 10 December 1996     

Age-dependent differences in the anticoagulant effect of phenprocoumon in patients after heart valve surgery

Objective: An enhanced response to warfarin and an increased risk of major bleeding has been observed in older patients. The reason for this increase in sensitivity remains unknown. It could be due to pharmacodynamic reasons, pharmacokinetic reasons, or both. Methods: We therefore followed an anticoagulant regimen with phenprocoumon in 19 older (76 years) and 19 younger patients (50 years) following heart valve replacement. INR values were determined frequently. At the 4th and around the 24th day after starting treatment with phenprocoumon, we also measured the total and unbound plasma concentration of phenprocoumon. Results: The dose requirement to obtain the desired anticoagulant effect was significantly lower in the older patients than in the younger patients (26.3 vs. 37.3 μg · kg⁻¹ · day⁻¹). The total plasma concentration (2.19 vs. 2.43 μg · ml⁻¹), the percentage unbound drug in the plasma (0.61 vs. 0.64%) and the unbound plasma concentration (13.8 vs. 15.1 ng · ml⁻¹) did not differ significantly between older and younger patients. The dose-adjusted INR (INR/dose) was higher in the older patients (110 vs. 67) but the INR adjusted for the unbound plasma concentration (INR/C_uss) which reflects the intrinsic sensitivity to the drug, was not significantly different (192 vs. 173). However, the older patients had an about 30% significantly lower metabolic clearance based on unbound drug (84 vs. 115 ml · kg⁻¹ · h⁻¹). Conclusions: Older patients (> 70 years) require a dose approximately 30% lower than younger patients (< 160␣years). Pharmacokinetic reasons (reduced metabolic clearance) are mainly responsible for the lower dose requirement of the older patients after heart valve surgery. Received: 23 August 1996 / Accepted November 11 1996