| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
PHARMACOKINETICS AND PHARMACODYNAMICS |
From GlaxoSmithKline Pharmaceuticals, Clinical Pharmacokinetics/Modeling and Simulation (Dr. Cox, Dr. Boyle, Dr. Fossler, Mr. Aluri), Clinical Pharmacology and Discovery Medicine, King of Prussia, Pennsylvania; Clinical Development and Medical Affairs (Dr. Parchman), Cardiovascular/Genito-urinary Therapeutic Area, King of Prussia, Pennsylvania; Biomedical Data Sciences (Dr. Holdbrook), Biostatistics and Programming, King of Prussia, Pennsylvania; and Baylor College of Medicine, Houston, Texas (Dr. Kleiman). GlaxoSmithKline Pharmaceuticals (Philadelphia, PA) and Encysive Pharmaceuticals (Houston, TX) provided study support.
Address for reprints: Donna S. Cox, PhD, GlaxoSmithKline, Clinical Pharmacokinetics Modeling and Simulation, Clinical Pharmacology and Discovery Medicine, 709 Swedeland Road, UW 27-1013, King of Prussia, PA 19406.
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
10%). By covariate analysis, clearance increased linearly with body weight. Plasma argatroban and ACT effect were well described using a sigmoidal Emax model. For argatroban in combination with platelet glycoprotein IIb/IIIa receptor blockade in patients undergoing PCI, population PK parameters are consistent with values reported for argatroban in healthy subjects. A predictable relationship exists between argatroban concentration and effect in this setting.
Key Words: Argatroban • percutaneous coronary intervention • platelet GPIIb/IIIa receptor antagonists • activated clotting time • pharmacokinetics • pharmacodynamics
The most common anticoagulant used during PCI is unfractionated heparin.1,2,4 However, heparin has several significant limitations that make the use of other anticoagulants more attractive, including the inability to bind and inhibit fibrin-bound thrombin or platelet-bound factor Xa, nonspecific binding to plasma proteins, the stimulation of platelet activation and aggregation, dependence on adequate antithrombin levels, interaction with platelet factor 4 (PF4), and the risk of heparin-induced thrombocytopenia (HIT).2,16,17 DTIs such as argatroban are a class of anticoagulants that offer several advantages when compared with heparin. Argatroban is a synthetic DTI derived from L-arginine that reversibly binds the active site of thrombin.18 Argatroban prevents fibrin formation and platelet aggregation, has predictable pharmacodynamic responses as measured by either the activated partial thromboplastin time (aPTT) or activated clotting time (ACT), and has no inhibitory effect on other serine proteases at therapeutic concentrations.18,19 Unlike the DTIs hirudin and bivalirudin, which are renally cleared, argatroban is predominantly hepatically metabolized.20 Last, argatroban is not inhibited by PF4 and does not potentiate HIT.21-23 Argatroban is approved in the United States as an anticoagulant for the prophylaxis or treatment of thrombosis in patients with HIT and in patients with or at risk for HIT undergoing PCI.
Several studies have assessed the effect of argatroban anticoagulation in patients undergoing PCI. In a pooled analysis of 3 prospective studies,21 91 patients who had or who were at risk for HIT underwent 112 separate PCIs using argatroban anticoagulation (350 µg/kg initial bolus followed by 25 µg/kg/min infusion). Outcomes compared favorably with those historically reported for heparin anticoagulation during PCI. More recently, a pilot study of argatroban in combination with platelet GPIIb/IIIa receptor antagonists in patients undergoing PCI has suggested that a lower dose of argatroban (250-300 µg/kg bolus followed by a 15-µg/kg/min infusion) provides adequate anticoagulation and is well tolerated.24 Because platelet GPIIb/IIIa receptor antagonists are now routinely used for anticoagulation during PCI, particularly in patients with unstable angina or with other high-risk clinical characteristics,25,26 additional evaluation of argatroban in combination with platelet GPIIb/IIIa receptor antagonists is warranted.
The purpose of this pharmacokinetic-pharmacodynamic (PK-PD) analysis was to characterize the relationship between argatroban plasma concentrations and pharmacodynamic effect, as assessed by ACTs in patients undergoing PCI using argatroban in combination with a platelet GPIIb/IIIa receptor antagonist. In addition, we developed a population PK-PD model to help guide dose selection in future clinical trials of argatroban in patients undergoing PCI.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Patients 
18 years of age with documented coronary artery disease who were scheduled for PCI were eligible. Exclusion criteria included thrombolytic therapy within 24 hours, current heparin use if the aPTT was 35 seconds or ACT was > 160 seconds, planned use of any other anticoagulant during the study period, international normalized ratio > 1.2, use of any GPIIb/IIIa antagonist in the previous 3 weeks, use of an investigational drug within 30 days or 5 half-lives, laboratory evidence of clinically relevant hepatic dysfunction, Q-wave myocardial infarction with cardiogenic shock, use of interventional devices other than a balloon or stent, > 50% stenosis of an unprotected left main coronary artery, planned staged interventional procedures, and a condition associated with increased risk for bleeding.
The study had an adaptive design with the option of changing the dose of argatroban if target ACTs (275-325 seconds in at least 80% of patients) were not met after administration of the initial bolus dose. Following provision of written informed consent, patients received 160 to 325 mg aspirin and either 300 mg clopidogrel or 150 mg ticlopidine. Part 1 of the study consisted of 101 patients receiving an initial 250-µg/kg bolus followed by a 15-µg/kg/min infusion of argatroban during the procedure, in combination with abciximab (bolus dose of 0.25 mg/kg followed by a 0.125-µg/kg/min infusion for 12 hours; n = 99) or eptifibatide (double bolus of 180 µg/kg followed by a 1.5-µg/kg/min infusion for 12 hours; n = 2). In part 2, an additional 51 received an initial bolus dose of 300 µg/kg followed by a 15-µg/kg/min infusion of argatroban during the procedure, in combination with abciximab. In each part, patients may have received up to 3 bolus doses of 150 µg/kg to achieve or maintain target ACTs.
Simultaneous pharmacokinetic and pharmacodynamic (ACT) sample collection occurred within 5 to 15 minutes following the start of argatroban therapy and approximately every 30 minutes thereafter. Samples were also collected every 30 minutes following the end of the infusion until the sheath was removed, which typically occurred within 3 hours of discontinuing argatroban. This sparse sampling scheme ranged from 3 to 6 samples per patient. Blood specimens for PK analysis were collected into citrate-containing tubes and centrifuged within 2 hours to separate plasma. Plasma was frozen at -70°C, shipped frozen to GlaxoSmithKline (King of Prussia, Pa), and stored at -70°C until assayed for argatroban.
The total data set used in the population PK-PD model assessment consisted of 152 patients (106 males, 46 females) with a mean (standard deviation [SD]) age of 63 (11) years and weight of 89 (18) kg. Other baseline characteristics and clinical outcomes of the study population are reported elsewhere.24
Analytical Methods
Argatroban was isolated from human plasma by protein precipitation and was quantified using turboionspray interface liquid chromatography/tandem mass spectrometry (LC/MS/MS). Positive ion multiple-reaction monitoring was employed from the MS/MS detection of argatroban and the internal standard, nitrazepam. Based on 50 µL of plasma aliquot, the lower limit of quantification for argatroban was 5.00 ng/mL. Linear responses in analyte/internal standard peak area ratios were observed for analyte concentrations ranging from 5.00 ng/mL to 1000 ng/mL. Intra-assay precision ranged between 5.84% and 7.96%. Interassay accuracy and precision ranged between 1.43% and 3.26% and between 3.97% and 8.72%, respectively.
The investigator at each individual center measured whole-blood ACTs using a calibrated Hemochron (tube system; International Technidyne Corp, Edison, NJ) or HemoTec (cartridge system; Medtronic, Englewood, Colo) device according to the manufacturer's instructions.
Pharmacostatistical Analysis
Software. Preliminary PK analysis was done using WinNonlin (Pharsight Corporation, Mountain View, Calif). NONMEM (version 5.0, level 1.0; GloboMax LLC, Hanover, Md) was used to conduct all population PK and PD analysis. The statistical assumptions for the models are valid, irrespective of the drug being dosed to a target effect. Actual, rather than protocol-specified, sampling times were used in all analyses.
Pharmacokinetic model development. Initial pharmacokinetic model development was performed using graphical analysis. Plots of argatroban plasma concentration versus time were constructed for each individual and examined to determine the appropriate descriptive model. Patients who received additional bolus doses were included in the analysis. All PK data were fit to a 2-compartment pharmacokinetic model using nonlinear mixed-effects modeling; model fitting employed ADVAN3 and TRANS4 subroutines within NONMEM. The model was parameterized in terms of clearance (CL), volume of central compartment (V1), volume of peripheral compartment (V2), and inter-compartmental clearance (Q). Initial estimates of CL and Vss (V1 + V2) were obtained from previous data in healthy volunteers.20,27,28 Interindividual variability (IIV) in PK parameters was evaluated using several models, but it was best described using an exponential model as depicted below:
 | (1) |
where Pi is the parameter estimate of the ith individual, P is the typical (mean) value for the population, and 
i is the proportional difference between Pi and P and has a mean value of 0 and a variance = 
2. Random residual variability, which includes model misspecification, errors in sampling times, and within-subject variability, was also evaluated using several error structures. A combined additive and proportional (CCV) model was found to best describe the data as depicted below:
 | (2) |
where Cji and C0ji are the ith measured and model-predicted argatroban plasma concentrations for the jth patient, respectively. The parameters 
1ji and 2ji denote the random residual error for the constant CV portion of the error model and the additive portion with respective variances of 
and 
and a mean of zero. Comparisons between 3 different estimation methods (ie, first-order [FO], first-order conditional estimation [FOCE], and first-order conditional estimation method with interaction [FOCE INT]) were evaluated. The influence of covariates such as body weight, age, gender, and race were also evaluated on argatroban PK. All of these fixed effects were added individually to the model centered at their median value.
Pharmacokinetic model development was carried out in a stepwise fashion. An initial (baseline) model was run, then parameters were added one at a time and the model was refit. Criteria for selecting the final model included improvement in the weighted residual plots, increased precision in parameter estimates (with smaller standard error estimates), reduced interresidual and/or random residual variances for the parameter estimates, and a reduced objective function (decrease of at least 7 points was considered significant, P < .01). Goodness of fit and model validity were assessed by graphical analysis of predicted versus observed data plots and assessments in the weighted residual and residual plots. All of these criteria were taken into account when evaluating alternate models. Once a satisfactory fit was obtained for the plasma concentration data, the individual predictions were used as input for the pharmacodynamic model.
Pharmacokinetic-pharmacodynamic model. Initial pharmacodynamic model development was carried out using graphical analysis. Plots of argatroban plasma concentration and pharmacodynamic effect (ACT) were constructed for each individual and examined to determine the appropriate descriptive model. Based on this graphical analysis, the baseline sigmoid Emax model was chosen and is described as follows:
 | (3) |
where E(t) is the ACT at time t, E0 is the baseline effect (ACT) measured in the absence of drug, Emax is the maximal increase in ACT response from baseline, EC50 is the concentration of argatroban causing a 50% increase in the ACT, and 
is a shape factor characterizing the slope of the response.
Individual predicted argatroban concentrations from the final pharmacokinetic model were used as input into the PK-PD model, and pharmacodynamic parameters and their associated variability were estimated. This sequential design (ie, did not include Bayesian estimates) was appropriate for the sparse pharmacokinetic and pharmacodynamic sampling scheme used in this study. Among the common models examined, the CCV model was determined to provide the best description of the residual variability. Comparisons between 3 different estimation methods (ie, FO, FOCE, and FOCE INT) were evaluated at the end of the model development process. Patient demographic characteristics such as body weight, age, gender, and race were examined to describe their influence on argatroban PD. Model development was carried out in a stepwise fashion, and the goodness of fit was assessed in a similar manner as the PK portion of the modeling using standard plots.
Model validation. The available data (both PK and PK-PD) were divided into 2 groups for the analysis. The index data set (70% of the total data) was used for model development. The validation data set (30% of the total data) was set aside and not used for model building. Once a satisfactory fit was obtained using the index data, the model was used to predict the validation data set. The 2 data sets were then combined (n = 152, 100% of the total data) and the final model refit to refine the final population pharmacokinetic parameter estimates. The model validation procedure was done for both the PK and PK-PD data sets. All results described herein used the combined data set (n = 152 patients) unless otherwise stated.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
The partial residuals of each PK parameter were plotted against patient age, weight, gender, and race to determine if a relationship between these fixed effects and any of the PK parameters could be discerned. Only weight and age appeared to affect CL and V1, respectively. Model 4 incorporated the addition of weight to clearance in a linear fashion. This resulted in a 76-point drop in the objective function and an increase in the precision of the pharmacokinetic parameters. As depicted in model 5, the covariance step was aborted when the addition of other covariates (age, race, and gender) was assessed. Hence, model 4 was chosen for subsequent analysis. Last, additional estimation methods were evaluated, with minimization terminated in both models 6 (FOCE) and 7 (FOCE with interaction). Therefore, model 4 was selected as the final PK model. Although there was a target ACT range in the study, this did not cloud the modeling assumptions. Such modeling is considered valid irrespective of there being a target ACT range.
Table II summarizes the argatroban population PK parameters for patients undergoing PCI, using the fitting of model 4. The PK parameters and interindividual variability were well defined, with %CVs of 10% and 11% to 34%, respectively. The linear effect of weight on clearance explained additional interindividual variability, with the CV decreasing from 41% to 38%. Residual error, reflecting intrasubject variability, measurement error, and model misspecification, was moderate at 21%, with a small additive component (14 µg/L).
|
Population Pharmacokinetics/Pharmacodynamics
Table III describes the major steps of the model development process for the combined PK-PD fit. Model 8, the initial baseline model, incorporating a baseline sigmoid Emax model, included estimates of all pharmacokinetic parameters and their intersubject variabilities (IIV). The precision of estimating the IIV on both Emax and EC50 was poor (%CV > 1000). In addition, a matrix plot of all of the IIVs plotted against one another revealed a correlation between the interindividual variability of Emax and EC50. Several approaches were attempted to account for this correlation. A covariance (block) was tested in model 9. This model resulted in a 25-point increase in the objective function and a decrease in the precision of the parameter estimates. Model 10 was considered by removing IIV on E0. This approach resulted in a 37-point decrease in the objective function and an increase in the precision on the pharmacodynamic parameters and their IIVs. Both a hybrid method (model 11) and a method incorporating a scaling factor (model 12) were also assessed to try and characterize the correlation between the IIV between Emax and EC50. Both models 11 and 12 resulted in a significant drop in the objective function (33 and 130 points, respectively) but unreasonable pharmacodynamic parameter estimates and low precision on IIV. Therefore, model 10 was chosen for subsequent analysis.
|
The partial residuals of each PD parameter were plotted against patient age, weight, gender, and race to determine if a relationship between these fixed effects and any of the PD parameters could be discerned. There was no apparent relationship between these covariates and PD parameters, and no further covariate analysis was assessed. Last, additional estimation methods were evaluated, with minimization terminated in models 13 (FOCE) and 14 (FOCE with interaction). Therefore, model 10 was selected as the final PKPD model.
Table IV summarizes the argatroban population PKPD parameters for patients undergoing PCI, from the fitting of model 10. The PD parameters were well defined, with %CV estimates < 15% for each parameter. Random residual variability was 12%. Interindividual variability for Emax and was 36% and 23%, respectively, with moderate precision (18% and 30%, respectively).
|
Figure 1 shows population-predicted and individual-predicted concentrations of argatroban plotted against observed effect (ACT) obtained from model 10. The fit was judged to be very good. In addition, a featureless plot of the weighted residuals versus predicted effect indicated that both the structural and statistical models were satisfactory.
|
Model Validation
Model validation was performed on both the PK and PK-PD data sets. Predictions of the validation data were plotted together with the observed validation data. The prediction of the validation data set (30% of total data) by the index data set (70% of total data) was reasonably good by visual inspection. In addition, 1000 simulations were performed to calculate the 95% prediction intervals of the observed data. Figure 2 depicts the individual-observed and model-predicted effect versus argatroban concentration profile for all patients, together with the 97.5 and 2.5 percentiles calculated from the simulations. Approximately 95% of the observed data appear to fall within this interval, suggesting that the final model accurately describes the observed data.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Using a sequential population approach, the pharmacokinetics and pharmacodynamics of argatroban administered in combination with a platelet GPIIb/IIIa receptor antagonist were well characterized. First, a 2-compartment PK model consisting of error models to characterize both residual and intersubject variability was developed. Population PK parameters were well characterized with standard errors (%CV) being 10%, and interindividual variability was also reasonably defined by the model with %CVs < 45%. Argatroban mean population PK estimates for CL (22.0 L/h), V1 (11.0 L), and V2 (13.0 L) were reasonable and consistent with values previously obtained in healthy volunteers (CL, 19.3-24.4 L/h; Vss, or V1 + V2, 18.6-25.0 L) and also suggests that there was no effect of the GPIIb/IIIa receptor antagonist on the PK of argatroban.20,27,28 Previous data suggest that there is little difference between ACT values obtained from either the Hemochron or HemoTec device in patients undergoing PCI using argatroban anticoagulation.29 Therefore, the type of device used to measure ACT was not assessed as a covariate during model development. However, covariate analysis suggested that the linear effect of body weight on clearance explained additional interindividual variability. In a previous study in healthy subjects, clearance was approximately 20% lower in elderly men than elderly women, most likely due to differences in body weight.20 It is noted that the recommended dosing of argatroban for its FDA-approved indications is based on body weight, and guidance is available by 10-kg increments of body weight for standard infusion rates of argatroban 1 mg/mL, final concentration.30 The current study supports the conclusion that there is also a linear effect of weight on clearance in patients undergoing PCI. However, only 3% of the residual variability in the PK-PD model was accounted for by body weight. Therefore, the impact of body weight in this patient population is not significant enough to warrant modification of current dosing guidance.
The target ACT range of 275 to 325 seconds was met in a majority of patients administered a single 250-µg/kg bolus of argatroban followed by a 15-µg/kg/min infusion; however, 22% of the patients received additional 150-µg/kg boluses to reach target ACTs.24 Among the patients who received an initial 300-µg/kg bolus dose, 27% required additional boluses to reach the target ACTs. Patients exhibiting low ACTs (in either bolus dose group) who required additional boluses did not have argatroban concentrations that were significantly different (eg, lower) from those who only received 1 bolus. These differences in the response or effect of argatroban from patient to patient may be explained by the 
26% variability captured in the population PK-PD model. Irregardless of the level of ACT achieved, all patients in the study who completed PCI had angio-graphic success and were considered to have adequate anticoagulation by the site investigators.
Clinical studies in healthy volunteers have demonstrated that argatroban, with or without a bolus dose, predictably increases coagulation parameters (ACT and aPTT) relative to baseline in a similar, dose- and concentration-dependent manner.20,31 These studies suggest that the relationship between argatroban concentration and effect is linear. However, results from the present study strongly suggest that the relationship between argatroban concentration and effect (ACT) for patients undergoing PCI with platelet glycoprotein IIb/IIIa blockade is best described by a baseline sigmoid Emax model. Further studies assessing the relationship between concentration and effect at higher doses of argatroban may be necessary to complement this study. Nonetheless, argatroban population PD parameters were well defined by this model, with %CVs being less than 15% for each parameter. Intersubject variability was moderate and no greater than 26% where estimated. With a sigmoid Emax model, there is an increase in response with drug concentration at lower drug concentrations, and a plateau effect level is approached at higher drug concentrations. The final PK-PD model suggests that in this study population, a mean plasma argatroban concentration of 1920 ng/mL, which corresponds to a typical steady-state level in healthy subjects receiving 10 µg/kg/min,31 is expected to increase the mean ACT by 50%. In addition, an effect plateau is expected to occur at a mean ACT of approximately 464 seconds, reflecting the sum of the Emax and E0 estimates.
All pharmacological responses have a maximum effect in which no further increment in response is achieved despite increased drug dose or concentration. In general, recognition of this maximal effect is helpful in avoiding ineffectual increases of dose that may be associated with toxicity or unwarranted side effects. The mean maximal effect (ACT) in patients undergoing PCI with argatroban in combination with GPIIb/IIIa inhibition was 288 and 297 seconds for the 250-µg/kg and 300-µg/kg bolus groups, respectively. The mean maximal effect predicted by the population PK-PD model (464 seconds) suggests that a higher bolus and/or a higher bolus followed by a higher infusion of argatroban may produce a greater effect in this patient population.
In summary, argatroban administered in combination with a platelet GPIIb/IIIa receptor antagonist in patients undergoing PCI was safe and effective. Population PK parameters are comparable with values reported for argatroban in healthy subjects. By covariate analysis, patient age, gender, and race did not affect clearance, which increased linearly with body weight. Given the small amount of variability accounted for by body weight in the model, no additional adjustment beyond the currently available weight-adjusted dosing guidance30 is required. Furthermore, a predictable relationship exists between mean plasma argatroban concentration and mean effect (ACT) in this setting. The population PK-PD model suggests that an increase in dose may produce higher ACT levels. Hence, the population PK-PD model may be a useful tool in guiding dose selection of argatroban for future studies in patients undergoing PCI.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Submitted for publication October 29, 2003; Revised version accepted May 23, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
1. McKeage K, Plosker G: Argatroban. Drugs 2001;61: 515-522.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
2. Wittkowsky AK: The role of thrombin inhibition during percutaneous coronary intervention. Pharmacotherapy 2002;22(6, Pt. 2): 97S-104S.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
3. Ruef J, Katus HA: New antithrombotic drugs on the horizon. Expert Opin Pharmacother 2003;12: 781-797.
4. Choo JK, Yong JJ, Kereiakes DJ: A guide to drug use during percutaneous coronary intervention. Drugs 2002;62: 2589-2601.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
5. Kaplan K: Direct thrombin inhibitors. Expert Opin Pharmacother 2003;4: 653-666.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
6. Popma JJ, Ohman M, Weitz J, Lincoff M, Harrington, RA, Berger P: Antithrombotic therapy in patients undergoing percutaneous coronary intervention. Chest 2001;119: 321S-336S.
7. Popma JJ, Weitz J, Bittl JA, Ohman M, Kuntz RE, Lansky AJ, et al: Antithrombotic therapy in patients undergoing coronary angioplasty. Chest 1998;114: 728S-741S.
8. Eikelboom J, French J: Management of patients with acute coronary syndromes: what is the clinical role of direct thrombin inhibitors? Drugs 2002;62: 1839-1852.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
9. Mehta WR, Yusuf S: Short- and long-term oral antiplatelet therapy in acute coronary syndromes and percutaneous coronary intervention. J Am Coll Cardiol 2003;41: 79S-88S.
10. Cohen M: The role of low-molecular-weight heparin in the management of acute coronary syndromes. J Am Coll Cardiol 2003;41: 55S-61S.
11. Nappi J: Heparin, bivalirudin, and reduction of complications during percutaneous coronary intervention. Pharmacotherapy 2002;22(6, Pt. 2): 89S.[CrossRef][Web of Science]
12. Deutsch E: The emerging role of low-molecular-weight heparin and antiplatelet therapies in the cardiac catheterization laboratory. Am Heart J 1999;138: S577-S585.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
13. Tcheng JE, Talley D, OShea J, Gilchrist I, Kleiman N, Grines C, et al: Clinical pharmacology of higher dose eptifibatide in percutaneous coronary intervention (The PRIDE study). Am J Cardiol 2001;88: 1097-1102.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
14. Kleiman N, Llem J, Fernandes L, Rubin H, Challa S, Solomon S, et al: Pharmacodynamic profile of the direct thrombin antagonist bivalirudin given in combination with the glycoprotein IIb/IIIa antagonist eptifibatide. Am Heart J 2002;143: 585-593.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
15. Gilchrist I, Oshea C, Kosoglou T, Jennings L, Lorenz T, Kitt M, et al: Pharmacodynamics and pharmacokinetics of higher-dose, double-bolus eptifibatide in percutaneous coronary intervention. Circulation 2001;104: 406-411.
16. Bauer KA: Selective inhibition of coagulation factors: advances in antithrombotic therapy. Semin Thromb Hemost 2002;28(Suppl. 2): 15-24.
17. Hirsh J: New anticoagulants. Am Heart J 2001;2(Supp.): S3-S8.
18. Hursting MJ, Alford K, Becker JC, Brooks RL, Joffrion JL, Knappenberger GD, et al: Novastan (brand of argatroban): a small-molecule, direct thrombin inhibitor. Sem Thromb Hemost 1997;23: 503-516.[Medline] [Order article via Infotrieve]
19. Moledina M, Chakir M, Gandhi PJ: A synopsis of the clinical uses of argatroban. J Thromb Thrombolysis 2001;12: 141-149.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
20. Swan SK, Hursting MJ: The pharmacokinetics and pharmacodynamics of argatroban: effects of age, gender, and hepatic or renal dysfunction. Pharmacotherapy 2000;20: 318-329.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
21. Lewis BE, Matthai WH Jr, Cohen M, Moses J, Hursting M, Leya, F: Argatroban anticoagulation during percutaneous coronary intervention in patients with heparin-induced thrombocytopenia. Catheter Cardiovasc Interv 2002;57: 177-184.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
22. Lewis BE, Wallis DE, Berkowitz SD, Matthai WH, Fareed J, Walengo JM, et al: Argatroban anticoagulant therapy in patients with heparin-induced thrombocytopenia. Circulation 2001;103: 1838-1843.
23. Lewis BE, Wallis DE, Leya F, Hursting MJ, Kelton JG: Argatroban anticoagulation in patients with heparin-induced thrombocytopenia. Arch Int Med 2003;163: 1849-1856.
24. Jang IK, Lewis BE, Matthai WH, Kleiman NS: Argatroban and glycoprotein IIb/IIIa inhibitor are effective and safe in patients undergoing percutaneous coronary intervention: a prospective, open-label, dose-response study [abstract]. Am J Cardiol 2003;92(Suppl.): 128L.[CrossRef]
25. The EPILOG Investigators: Platelet glycoprotein IIb/IIIa receptor blockade and low-dose heparin during percutaneous coronary revascularization. N Engl J Med 1996;336: 1689-1696.[Web of Science]
26. Reginelli JP, Chew D: Direct thrombin inhibitors for percutaneous coronary intervention in the current era of platelet glycoprotein IIb/IIIa inhibition. J Interv Cardiol 2002;15: 141-146.[Medline] [Order article via Infotrieve]
27. Tran JQ, DiCicco RA, Sheth SB, Tucci M, Peng L, Jorkasky DL, et al: Assessment of the potential pharmacokinetic and pharmacodynamic interactions between erythromycin and argatroban. J Clin Pharmacol 1999;39: 513-519.[Abstract]
28. Inglis AM, Sheth SB, Hursting MJ, Tenero DM, Graham AM, DiCicco RA: Investigation of the interaction between argatroban and acetaminophen, lidocaine, or digoxin. Am J Health Syst Pharm 2002;59: 1258-1266.
29. Iqbal O, Ahmad S, Lewis BE, Walenga JM, Rangel Y, Fareed J: Monitoring of argatroban in ARG310 study: potential recommendations for its use in interventional cardiology. Clin Appl Thromb Hemost 2002;8: 217-224.
30. Prescribing information for argatroban. Accessed May 19, 2004, at www.argatroban.com
31. Swan SK, St. Peter JV, Lambrecht LJ, Hursting MJ: Comparison of anticoagulant effects and safety of argatroban and heparin in healthy subjects. Pharmacotherapy 2000;20: 756-770.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Rice, M. J. Hursting, G. M. Baillie, and D. A. McCollum Argatroban Anticoagulation in Obese Versus Nonobese Patients: Implications for Treating Heparin-induced Thrombocytopenia J. Clin. Pharmacol., August 1, 2007; 47(8): 1028 - 1034. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
