Intravenously administered digoxin in patients with acute atrial fibrillation: a population pharmacokinetic/pharmacodynamic analysis based on the Digitalis in Acute Atrial Fibrillation trial

Population Pharmacokinetics of Digoxin in Nonagenarian Patients: Optimization of the Dosing Regimen

OBJECTIVE

The aim of this study was to develop a population pharmacokinetic model of digoxin in patients over 90 years old and to propose an equation for adjusting digoxin dose in this population.

METHODS

We included 326 nonagenarian patients admitted to Severo Ochoa University Hospital (Spain) who received digoxin and were under therapeutic drug monitoring. All data were retrospectively collected, and population modeling was performed with non-linear mixed-effect modeling software (NONMEM®). One- and two-compartment models were tested to calculate digoxin clearance (Cl), volume of distribution (Vd), absorption rate constant (Ka), and bioavailability (bioavailable fraction, F). The covariates were evaluated by stepwise covariate model building, and the final model was internally validated by bootstrap analysis with 1000 resamples. External validation was performed with another population of 95 patients with the same characteristics as the modeling group.

RESULTS

The population was 26% males, with a mean age of 93.2 years (90-103 years), mean creatinine 1.11 mg/dL (0.42-3.81 mg/dL), and mean total body weight 61.2 kg (40-100 kg). The pharmacokinetics of digoxin were best described by a one-compartment model (ADVAN2 TRANS2), with first-order conditional estimation with interaction. The covariates with influence on our model were creatinine clearance based on the Cockcroft-Gault equation (CG), serum potassium (K), co-administration of loop diuretics, and sex: Cl/F = 4.55 · (CG/36.4)0.468 · 0.83LD · 1.21SEX; Vd/F = 355 · (K/4.3)-0.849; Ka = 1.22 h-1 [where LD indicates loop diuretics (1 for administered, 0 for otherwise) and SEX indicates patient sex (1 for male, 0 for female)]. Based on our results, we proposed an equation to adjust the digoxin dosing regimen in nonagenarian patients: dose (mg) = 0.144 · (CG/36.4)0.468 · 0.83LD · 1.21SEX.

CONCLUSIONS

The greatest influence on digoxin clearance came from renal function calculated by the Cockcroft-Gault equation. Vd was decreased by K. The model developed showed a precise predictive performance to be applied for therapeutic drug monitoring.

Pathophysiology and treatment of adults with arrhythmias in the emergency department, part 1: Atrial arrhythmias

PURPOSE

This article, the first in a 2-part review, aims to reinforce current literature on the pathophysiology of cardiac arrhythmias and various evidence-based treatment approaches and clinical considerations in the acute care setting. Part 1 of this series focuses on atrial arrhythmias.

SUMMARY

Arrhythmias are prevalent throughout the world and a common presenting condition in the emergency department (ED) setting. Atrial fibrillation (AF) is the most common arrhythmia worldwide and expected to increase in prevalence. Treatment approaches have evolved over time with advances in catheter-directed ablation. Based on historic trials, heart rate control has been the long-standing accepted outpatient treatment modality for AF, but the use of antiarrhythmics is often still indicated for AF in the acute setting, and ED pharmacists should be prepared and poised to help in AF management. Other atrial arrhythmias include atrial flutter (AFL), atrioventricular nodal reentry tachycardia (AVNRT), and atrioventricular reentrant tachycardia (AVRT), which warrant distinction due to their unique pathophysiology and because each requires a different approach to utilization of antiarrhythmics. Atrial arrhythmias are typically associated with greater hemodynamic stability than ventricular arrhythmias but still require nuanced management according to patient subset and risk factors. Since antiarrhythmics can also be proarrhythmic, they may destabilize the patient due to adverse effects, many of which are the focus of black-box label warnings that can be overreaching and limit treatment options. Electrical cardioversion for atrial arrhythmias is generally successful and, depending on the setting and/or hemodynamics, often indicated.

CONCLUSION

Atrial arrhythmias arise from a variety of mechanisms, and appropriate treatment depends on various factors. A firm understanding of physiological and pharmacological concepts serves as a foundation for exploring evidence supporting agents, indications, and adverse effects in order to provide appropriate care for patients.

Dosage Optimization of Digoxin in Older Patients with Heart Failure and Chronic Kidney Disease: A Population Pharmacokinetic Analysis

BACKGROUND

Renal function is an important index for digoxin dose adjustment, especially in patients with chronic kidney disease (CKD). Decreased glomerular filtration rate is common in older patients with cardiovascular disease.

OBJECTIVE

The aim of this study was to establish a digoxin population pharmacokinetic model in older patients with heart failure and CKD and to optimize the digoxin dose strategy.

METHODS

Older patients with heart failure and CKD aged > 60 years from January 2020 to January 2021 and who had an estimated glomerular filtration rate (eGFR) < 90 mL/min/1.73 m² or urine protein production were enrolled in this retrospective study. Population pharmacokinetic analysis and Monte Carlo simulations (n = 1000) were performed using NONMEN software. The precision and stability of the final model were analyzed by graphical and statistical methods.

RESULTS

Overall, 269 older patients with heart failure were enrolled. A total of 306 digoxin concentrations were collected, with a median value of 0.98 ng/mL (interquartile range [IQR] 0.62-1.61, range 0.04-4.24). The median age was 68 years (IQR 64-71, range 60-94) and eGFR was 53.6 mL/min/1.73 m² (IQR 38.1-65.2, range 11.4-89.8). A one-compartment model with first-order elimination was developed to describe the digoxin pharmacokinetics. Typical values for clearance and volume of distribution were 2.67 L/h and 36.9 L, respectively. Dosage simulations were stratified by eGFR and metoprolol. Doses of 62.5 and 125 μg were recommended for older patients with eGFR < 60 mL/min/1.73 m².

CONCLUSIONS

A population pharmacokinetic model of digoxin in older patients with heart failure and CKD was established in this study. A novel digoxin dosage strategy was recommended in this vulnerable population.

Rapid Atrial Fibrillation in the Emergency Department

Atrial fibrillation (AF) is the most common rhythm disorder seen in doctors' offices and emergency departments (EDs). In both settings, an AF holistic pathway including anticoagulation or stroke avoidance, better symptom management, and cardiovascular and comorbidity optimization should be followed. However, other considerations need to be assessed in the ED, such as haemodynamic instability, the onset of AF, the presence of acute heart failure and pre-excitation. Although the Advanced Cardiovascular Life Support guidelines (European Society of Cardiology guidelines, Acute Cardiac Care Association/European Heart Rhythm Association position statements) and several recent AF publications have greatly assisted physicians in treating AF with rapid ventricular response in the ED, further practical clinical guidance is required to improve physicians' skill and knowledge in providing the best treatment for patients. Herein, we combine multiple strategies with supporting evidence-based treatment and experiences encountered in clinical practice into practical stepwise approaches. We hope that the stepwise algorithm may assist residents and physicians in managing AF in the ED.

Clinical experience with titrating doses of digoxin antibodies in acute digoxin poisoning. (ATOM-6)

INTRODUCTION

For acute digoxin poisoning, it has been recommended to give bolus doses of 10-20 vials or potentially larger than needed doses calculated from dose ingested or the measured concentration. However, a recent revision of internal Poisons Information Centre guidelines prompted a change of our recommendations, specifically instead of large boluses, to use titrating repeated low doses of digoxin antibodies(Digoxin-Fab) based on bedside assessment of cardiac toxicity.

METHODS

This is a prospective observational study of patients with acute digoxin poisoning identified through two Poisons Information Centres and three toxicology units. Patient demographics, signs and symptoms of digoxin toxicity, doses and response to Digoxin-Fab, free and bound serum digoxin concentrations. Outcomes were recorded and analysed.

RESULTS

From September 2013 to September 2020, 23 patients with 25 presentations (median age 56 years, females 56%) were recruited. Median dose ingested was 13 mg(IQR: 9.5-25). Median heart rate (HR) was 41 beats/min before treatment. Initial median digoxin and potassium concentrations were 14.5 nmol/L (IQR: 10.9-20) [11.2 µg/L(IQR: 8.4-15.4)] and 5 mmol/L (IQR: 4.5-5.4 mmol/L), respectively. Gastrointestinal symptoms and acute kidney injury were present in 22 patients (88%) and 5 patients (20%), respectively. Four patients received an initial bolus dose of Digoxin-Fab of 5-20 vials. Twenty-one patients received repeated titrated doses (1-2 vials) of Digoxin-Fab and the median total dose was 4 vials (IQR: 2-7.5). Median maximal change in HR post-Digoxin-Fab was 19 beats/min. The median potassium concentration decrease post-Digoxin-Fab was 0.3 mmol/L. Total dose used in the titration group was 25% and 35% of the predicted doses based on the amount of digoxin ingested or measured serum concentration, respectively. Twelve had free digoxin concentrations measured. Free digoxin concentrations dropped to almost zero after any dose of Digoxin-Fab. Ten patients had a rebound of digoxin >2.6 nmol/L (2 µg/L). There were no deaths from acute digoxin toxicity.

CONCLUSIONS

The new practice of using small, titrated doses of Digoxin-Fab led to a considerable reduction in total usage and major savings. The clinical response to titrated doses was safe and acceptable in acute digoxin poisoning.

Population Pharmacokinetic Studies of Digoxin in Adult Patients: A Systematic Review

BACKGROUND

Digoxin is a cardiac glycoside that was introduced to cardiovascular medicine more than 200 years ago. Its use is associated with large variability, which complicates achieving the desired therapeutic outcomes.

OBJECTIVES

To present a synthesis of the available literature on the population pharmacokinetics of digoxin in adults and to identify the sources of variability in its pharmacokinetics.

METHODS

This is a PROSPERO registered systematic review (CRD42018105300). A literature search was conducted using the ISI Web of Science, Science Direct, PubMed, and SCOPUS databases to identify digoxin population pharmacokinetic studies of adults that utilized the nonlinear mixed-effect modeling approach.

RESULTS

Sixteen articles were included in the present analysis. Only two studies were conducted in elderly subjects as a separate population. Both the pharmacokinetics and pharmacodynamics of digoxin were investigated in one study. Furthermore, the reviewed studies were mostly conducted in East Asian populations (68.8%). Digoxin's pharmacokinetics were usually described by a one-compartment model because of the nature of the collected data. Weight, age, kidney function, presence of heart failure, and co-administered medications such as calcium channel blockers were the most commonly identified predictors of digoxin clearance. The value of apparent clearance in a typical study individual ranged from 0.005 to 0.2 l/h/kg, while the value of the apparent volume of distribution ranged from 3.14 to 15.2 l/kg. The quality of model evaluation was deemed excellent only in 31.3% of the studies.

CONCLUSION

This review provides information about variables that need to be considered when prescribing digoxin. The results highlight the need for prospective studies that allow two-compartment pharmacokinetic/pharmacodynamic models to be established, with a special focus on the elderly subpopulation.

Variability Attribution for Automated Model Building

We investigated the possible advantages of using linearization to evaluate models of residual unexplained variability (RUV) for automated model building in a similar fashion to the recently developed method “residual modeling.” Residual modeling, although fast and easy to automate, cannot identify the impact of implementing the needed RUV model on the imprecision of the rest of model parameters. We used six RUV models to be tested with 12 real data examples. Each example was first linearized; then, we assessed the agreement in improvement of fit between the base model and its extended models for linearization and conventional analysis, in comparison to residual modeling performance. Afterward, we compared the estimates of parameters’ variabilities and their uncertainties obtained by linearization to conventional analysis. Linearization accurately identified and quantified the nature and magnitude of RUV model misspecification similar to residual modeling. In addition, linearization identified the direction of change and quantified the magnitude of this change in variability parameters and their uncertainties. This method is implemented in the software package PsN for automated model building/evaluation with continuous data.

Model-Based Residual Post-Processing for Residual Model Identification

The purpose of this study was to investigate if model-based post-processing of common diagnostics can be used as a diagnostic tool to quantitatively identify model misspecifications and rectifying actions. The main investigated diagnostic is conditional weighted residuals (CWRES). We have selected to showcase this principle with residual unexplained variability (RUV) models, where the new diagnostic tool is used to scan extended RUV models and assess in a fast and robust way whether, and what, extensions are expected to provide a superior description of data. The extended RUV models evaluated were autocorrelated errors, dynamic transform both sides, inter-individual variability on RUV, power error model, t-distributed errors, and time-varying error magnitude. The agreement in improvement in goodness-of-fit between implementing these extended RUV models on the original model and implementing these extended RUV models on CWRES was evaluated in real and simulated data examples. Real data exercise was applied to three other diagnostics: conditional weighted residuals with interaction (CWRESI), individual weighted residuals (IWRES), and normalized prediction distribution errors (NPDE). CWRES modeling typically predicted (i) the nature of model misspecifications, (ii) the magnitude of the expected improvement in fit in terms of difference in objective function value (ΔOFV), and (iii) the parameter estimates associated with the model extension. Alternative metrics (CWRESI, IWRES, and NPDE) also provided valuable information, but with a lower predictive performance of ΔOFV compared to CWRES. This method is a fast and easily automated diagnostic tool for RUV model development/evaluation process; it is already implemented in the software package PsN.

An automated sampling importance resampling procedure for estimating parameter uncertainty

Quantifying the uncertainty around endpoints used for decision-making in drug development is essential. In nonlinear mixed-effects models (NLMEM) analysis, this uncertainty is derived from the uncertainty around model parameters. Different methods to assess parameter uncertainty exist, but scrutiny towards their adequacy is low. In a previous publication, sampling importance resampling (SIR) was proposed as a fast and assumption-light method for the estimation of parameter uncertainty. A non-iterative implementation of SIR proved adequate for a set of simple NLMEM, but the choice of SIR settings remained an issue. This issue was alleviated in the present work through the development of an automated, iterative SIR procedure. The new procedure was tested on 25 real data examples covering a wide range of pharmacokinetic and pharmacodynamic NLMEM featuring continuous and categorical endpoints, with up to 39 estimated parameters and varying data richness. SIR led to appropriate results after 3 iterations on average. SIR was also compared with the covariance matrix, bootstrap and stochastic simulations and estimations (SSE). SIR was about 10 times faster than the bootstrap. SIR led to relative standard errors similar to the covariance matrix and SSE. SIR parameter 95% confidence intervals also displayed similar asymmetry to SSE. In conclusion, the automated SIR procedure was successfully applied over a large variety of cases, and its user-friendly implementation in the PsN program enables an efficient estimation of parameter uncertainty in NLMEM.

The role of digitalis pharmacokinetics in converting atrial fibrillation and flutter to regular sinus rhythm

This report examined the role of digitalis pharmacokinetics in helping to guide therapy with digitalis glycosides with regard to converting atrial fibrillation (AF) or flutter to regular sinus rhythm (RSR). Pharmacokinetic models of digitoxin and digoxin, containing a peripheral non-serum effect compartment, were used to analyze outcomes in a non-systematic literature review of five clinical studies, using the computed concentrations of digitoxin and digoxin in the effect compartment of these models in an analysis of their outcomes. Four cases treated by the author were similarly examined. Three literature studies showed results no different from placebo. Dosage regimens achieved ≤11 ng/g in the model's peripheral compartment. However, two other studies achieved significant conversion to RSR. Their peripheral concentrations were 9-14 ng/g. In the four patients treated by the author, three converted using classical clinical titration with incremental doses, plus therapeutic drug monitoring and pharmacokinetic guidance from the models for maintenance dosage. They converted at peripheral concentrations of 9-18 ng/g, similar to the two studies above. No toxicity was seen. Successful maintenance was achieved, using the models and their pharmacokinetic guidance, by giving somewhat larger than average recommended dosage regimens in order to maintain peripheral concentrations present at conversion. The fourth patient did not convert, but only reached peripheral concentrations of 6-7 ng/g, similar to the studies in which conversion was no better than placebo. Pharmacokinetic analysis and guidance play a highly significant role in converting AF to RSR. To the author's knowledge, this has not been specifically described before. In my experience, conversion of AF or flutter to RSR does not occur until peripheral concentrations of 9-18 ng/g are reached. Results in the four cases correlated well with the literature findings. More work is needed to further evaluate these provocative findings.

A cohort study of digoxin exposure and mortality in men with prostate cancer

OBJECTIVE

To examine the association between digoxin exposure and mortality in men with prostate cancer using linked Irish National Cancer Registry and pharmacy claims data.

PATIENTS AND METHODS

Prostate cancer cases were identified from the database and digoxin exposure at prostate cancer diagnosis was identified from prescription claims. Digoxin users were matched to non-users using a propensity score to identify men with similar cardiovascular comorbidity. Adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated for the association between digoxin exposure and all-cause and prostate cancer-specific mortality (PCSM). Analyses were repeated in the propensity score-matched cohort. Effect modification of treatment with radiation or androgen-deprivation therapy by digoxin exposure was also assessed.

RESULTS

In all, 5732 men with a prostate cancer diagnosis (2001-2006) were identified (digoxin exposed, 391). The median follow-up was 4.3 years. Digoxin exposure was associated with a small non-significant increase in PCSM in the full cohort (HR 1.13, 95% CI 0.91, 1.42) and the propensity. score-matched cohort (HR 1.17, 95% CI 0.88, 1.57). Adjusted HRs for all-cause mortality were increased for digoxin exposed men (HR 1.24, 95% CI 1.07, 1.43). Interactions with treatments received were not significant.

CONCLUSIONS

These results suggest digoxin exposure is not associated with reduced PCSM. Further investigation of other cardiac glycosides that have shown anti-cancer potential may be warranted.

Some comments and suggestions concerning population pharmacokinetic modeling, especially of digoxin, and its relation to clinical therapy

Population pharmacokinetic and dynamic modeling is often employed to analyze data of steady-state trough serum digoxin concentrations in the course of what is frequently regarded as routine therapeutic drug monitoring (TDM). Such a monitoring protocol is extremely uninformative. It permits only the estimation of a single parameter of a 1-compartment model, such as clearance. The use of D-optimal design strategies permits much more information to be obtained, employing models having a really meaningful structure. Strategies and protocols for routine TDM policies greatly need to be improved, incorporating these principles of optimal design. Software for population pharmacokinetic modeling has been dominated by NONMEM. However, because NONMEM is a parametric method, it must assume a shape for the model parameter distributions. If the assumption is not correct, the model will be in error, and the most likely results given the raw data will not be obtained. In addition, the likelihood as computed by NONMEM is only approximate, not exact. This impairs statistical consistency and reduces statistical efficiency and the resulting precision of model parameter estimates. Other parametric methods are superior, as they provide exact likelihoods. However, they still suffer from the constraints of assuming the shape of the model parameter distributions. Nonparametric methods are more flexible. One need not make any assumptions about the shape of the parameter distributions. Nonparametric methods also provide exact likelihoods and are statistically consistent, efficient, and precise. They also permit maximally precise dosage regimens to be developed for patients using multiple model dosage design, something parametric modeling methods cannot do. Laboratory assay errors are better described by the reciprocal of the assay variance of each measurement rather than by coefficient of variation. This is easy to do and permits more precise models to be made. This also permits estimation of assay error separately from the other sources of uncertainty in the clinical environment. This is most useful scientifically. Digoxin has at least 2-compartment behavior. Its pharmacologic and clinical effects correlate not with serum digoxin concentrations but with those in the peripheral nonserum compartment. Some illustrative clinical examples are discussed. It seems that digitalis therapy, guided by TDM and our 2 compartment models based on that of Reuning et al, can convert at least some patients with atrial fibrillation and flutter to regular sinus rhythm. Investigators have often used steady-state trough concentrations only to make a 1-compartment model and have sought only to predict future steady-state trough concentrations. Much more than this can be done, and clinical care can be much improved. Further work along these lines is greatly to be desired.

Rapid sample size calculations for a defined likelihood ratio test-based power in mixed-effects models

Efficient power calculation methods have previously been suggested for Wald test-based inference in mixed-effects models but the only available alternative for Likelihood ratio test-based hypothesis testing has been to perform computer-intensive multiple simulations and re-estimations. The proposed Monte Carlo Mapped Power (MCMP) method is based on the use of the difference in individual objective function values (ΔiOFV) derived from a large dataset simulated from a full model and subsequently re-estimated with the full and reduced models. The ΔiOFV is sampled and summed (∑ΔiOFVs) for each study at each sample size of interest to study, and the percentage of ∑ΔiOFVs greater than the significance criterion is taken as the power. The power versus sample size relationship established via the MCMP method was compared to traditional assessment of model-based power for six different pharmacokinetic and pharmacodynamic models and designs. In each case, 1,000 simulated datasets were analysed with the full and reduced models. There was concordance in power between the traditional and MCMP methods such that for 90% power, the difference in required sample size was in most investigated cases less than 10%. The MCMP method was able to provide relevant power information for a representative pharmacometric model at less than 1% of the run-time of an SSE. The suggested MCMP method provides a fast and accurate prediction of the power and sample size relationship.

Understanding the time course of pharmacological effect: a PKPD approach

The key concepts that underpin the choice of drug and dosing regimen are an understanding of the drugs' effectiveness, the potential for adverse effects, and the expected time course over which both desired and adverse effects are likely to occur. Research in clinical pharmacology should therefore address three fundamental questions: (1) What is the magnitude of drug effects (beneficial or adverse) from a given dose? (2) How quickly will any given effects occur? (3) How long will these effects last? Under steady-state conditions, only the magnitude of drug effects can be examined. This requires researchers to consider non-steady-state conditions, which require more complex models and an understanding of the mechanisms that drive the time course of drug effect. The aim of this review is to provide a conceptual framework for understanding the time course of drug effects using pharmacokinetic-pharmacodynamic models. Key examples will illustrate how this can inform the optimal use of drugs in the clinic.

Clinical manifestations of elderly patients with digitalis intoxication in the emergency department

This study aimed to determine the clinical characteristics of elderly patients diagnosed with digitalis intoxication, on the Emergency Department, University Hospital Complex, A Coruña, Spain. During the study period (January-September 2008) cases were included in which digitalis intoxication was confirmed by plasma digoxin levels. We collected data on age, gender, base-line diseases, therapeutic indications for digoxin, functional classification, ejection fraction, plasma digoxin levels, creatinine clearance, ions, gasometry, electrocardiogram, concomitant medication, symptomatology and treatment. The results were: mean age 82.0 ± 6.6 years, predominantly female subjects (83.7%). The most prevalent pathologies were cardiac valvulopathy (81.0%), hypertension (68.3%) and ischemic cardiopathy (46.3%), 95.1% had a background of cardiac insufficiency, and 52.6% were in functional grade III. The mean digoxin level was 2.7 ± 0.69 ng/ml, 23.1% of the patients had a creatinine clearance of less than 60 ml/min/1.73 m(2) and 2.6% had a severely reduced glomerular filtration rate (GFR) (clearance<30 ml/min/1.73 m(2)). A negative correlation was found between digoxin levels and clearance (r = -0.22; p = 0.18) and between the levels and cardiac frequency (r = -0.35; p = 0.026). Of the patients, 47.5% presented bradycardia and 87.8% arrhythmias, most frequently auricular fibrillation. The most frequent symptoms were nausea (54.8%), fatigue (42.9%), vomiting (33.3%) and anorexia (28.6%). We conclude that clinical digestive symptoms in elderly women who are taking digitalis, with bradycardia and impaired renal functioning, should lead us to suspect digitalis intoxication.

Population pharmacokinetic model of digoxin in older Chinese patients and its application in clinical practice

AIM

To establish a population pharmacokinetic (PPK) model of digoxin in older Chinese patients to provide a reference for individual medication in clinical practice.

METHODS

Serum concentrations of digoxin and clinically related data including gender, age, weight (WT), serum creatinine (Cr), alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), albumin (ALB), and co-administration were retrospectively collected from 119 older patients taking digoxin orally for more than 7 d. NONMEM software was used to get PPK parameter values, to set up a final model, and to assess the models in clinical practice.

RESULTS

Spironolactone (SPI), WT, and Cr markedly affected the clearance rate of digoxin. The final model formula is Cl/F=5.9x[1-0.412 x SPI] x [1-0.0101x(WT-62.9)] x [1-0.0012x(Cr-126.8)] (L/h); Ka=1.63 (h(-1)); V(d)/F=550 (L). The population estimates for Cl/F and V(d)/F were 5.9 L/h and 550 L, respectively. The interindividual variabilities (CV) were 49.0% for Cl/F and 94.3% for V(d)/F. The residual variability (SD) between observed and predicted concentrations was 0.365 microg/L. The difference between the objective function value and the primitive function value was less than 3.84 (P>0.05) by intra-validation. Clinical applications indicated that the percent of difference between the predicted concentrations estimated by the PPK final model and the observed concentrations were -4.3%-+25%. Correlation analysis displayed that there was a linear correlation between observed and predicted values (y=1.35x+0.39, r=0.9639, P<0.0001).

CONCLUSION

The PPK final model of digoxin in older Chinese patients can be established using the NONMEM software, which can be applied in clinical practice.

Importance of shrinkage in empirical bayes estimates for diagnostics: problems and solutions

Empirical Bayes ("post hoc") estimates (EBEs) of etas provide modelers with diagnostics: the EBEs themselves, individual prediction (IPRED), and residual errors (individual weighted residual (IWRES)). When data are uninformative at the individual level, the EBE distribution will shrink towards zero (eta-shrinkage, quantified as 1-SD(eta (EBE))/omega), IPREDs towards the corresponding observations, and IWRES towards zero (epsilon-shrinkage, quantified as 1-SD(IWRES)). These diagnostics are widely used in pharmacokinetic (PK) pharmacodynamic (PD) modeling; we investigate here their usefulness in the presence of shrinkage. Datasets were simulated from a range of PK PD models, EBEs estimated in non-linear mixed effects modeling based on the true or a misspecified model, and desired diagnostics evaluated both qualitatively and quantitatively. Identified consequences of eta-shrinkage on EBE-based model diagnostics include non-normal and/or asymmetric distribution of EBEs with their mean values ("ETABAR") significantly different from zero, even for a correctly specified model; EBE-EBE correlations and covariate relationships may be masked, falsely induced, or the shape of the true relationship distorted. Consequences of epsilon-shrinkage included low power of IPRED and IWRES to diagnose structural and residual error model misspecification, respectively. EBE-based diagnostics should be interpreted with caution whenever substantial eta- or epsilon-shrinkage exists (usually greater than 20% to 30%). Reporting the magnitude of eta- and epsilon-shrinkage will facilitate the informed use and interpretation of EBE-based diagnostics.

Evaluation of the nonparametric estimation method in NONMEM VI: application to real data

The aim of the study was to evaluate the nonparametric estimation methods available in NONMEM VI in comparison with the parametric first-order method (FO) and the first-order conditional estimation method (FOCE) when applied to real datasets. Four methods for estimating model parameters and parameter distributions (FO, FOCE, nonparametric preceded by FO (FO-NONP) and nonparametric preceded by FOCE (FOCE-NONP)) were compared for 25 models previously developed using real data and a parametric method. Numerical predictive checks were used to test the appropriateness of each model. Up to 1000 new datasets were simulated from each model and with each method to construct 90% and 50% prediction intervals. The mean absolute error and the mean error of the different outcomes investigated were computed as indicators of imprecision and bias respectively and formal statistical tests were performed. Overall, less imprecision and less bias were observed with nonparametric methods than with parametric methods. Across the 25 models, t-tests revealed that imprecision and bias were significantly lower (P < 0.05) with FOCE-NONP than with FOCE for half of the NPC outcomes investigated. Improvements were even more pronounced with FO-NONP in comparison with FO. In conclusion, when applied to real datasets and evaluated by numerical predictive checks, the nonparametric estimation methods in NONMEM VI performed better than the corresponding parametric methods (FO or FOCE).

Modelling the influence of MDR1 polymorphism on digoxin pharmacokinetic parameters

OBJECTIVES

Digoxin is a well-known probe for the activity of P-glycoprotein. The objective of this work was to apply different methods for covariate selection in non-linear mixed-effect models to study the relationship between the pharmacokinetic parameters of digoxin and the genotype for two major exons located on the multi-drug-resistance 1 (MDR1) gene coding for P-glycoprotein.

METHODS

Thirty-two healthy volunteers were recruited in three pharmacokinetic drug interaction studies. The data after a single oral administration of digoxin alone were pooled. All subjects were genotyped for the MDR1 C3435T and G2677T/A genotypes. The concentration-time profile of digoxin was established using 12-16 blood samples taken between 15 min and 72 h after administration. We modelled the pharmacokinetics of digoxin using non-linear mixed-effect models. Parameter estimation was performed using the stochastic approximation EM method (SAEM). We used three methods to select the covariate model: selection from a full model using Wald tests, forward inclusion using the log-likelihood ratio test and model selection using the Bayesian Information Criterion.

RESULTS

The three covariate inclusion methods led to the same final model. Carriers of two T alleles for the C3435T polymorphism in exon 26 of MDR1 had a lower apparent volume of distribution than carriers of a C allele. The only other covariate effect was a shorter absorption time-lag in women.

CONCLUSION

The apparent volume of distribution of digoxin is lower in TT subjects, probably reflecting differences in bioavailability. Non-linear mixed-effect models can be useful for detecting the influence of covariates on pharmacokinetic parameters.

The safety of digoxin as a pharmacological treatment of atrial fibrillation

Digoxin has traditionally been the drug of choice for ventricular rate control in patients with chronic atrial fibrillation (AF), with or without heart failure (HF) with systolic dysfunction. In patients with permanent AF, digoxin monotherapy is ineffective to control ventricular rate during exercise, but the combination of digoxin with a beta-blocker or a non-dihydropyridine calcium channel antagonist can control heart rate both at rest and during exercise. Only a few randomised, controlled studies have evaluated the adverse effects of digoxin in patients with AF in a systematic way and side effects requiring drug withdrawal have rarely been reported. When reported, the most frequent adverse effects were cardiac arrhythmias (ventricular arrhythmias, AV block of varying degrees and sinus pauses). This evidence suggested that, in contrast to other antiarrhythmic drugs, digoxin is a safe drug in patients with AF. However, this safety profile can be erroneous due to the short follow-up of the studies and patient selection. Because patients with HF have been excluded in most studies, the safety profile of digoxin in this population has not been directly addressed. Early recognition that an arrhythmia is related to digoxin intoxication as well as recognition of concomitant medications or medical conditions that may directly alter the pharmacokinetic profile of digoxin, or indirectly alter its cardiac effects by pharmacodynamic interactions remain essential for safe and effective use of digoxin in patients with AF.

Pharmacological rate control of atrial fibrillation

To control ventricular rate in patients with AF, physicians should seek to control heart rate at rest and with exertion. The goal has to be achieved while minimizing costs and adverse effects. For emergency use, i.v. diltiazem or esmolol are drugs useful because of their rapid onset of action. They have to be used with caution in patients with concomitant left ventricular failure symptoms, however. For most patients with AF, chronic control of the ventricular rate can be achieved with one drug. For the chronic control of ventricular rate in patients with AF and normal ventricular function, diltiazem, atenolol, are metoprolol are probably the drugs of choice. For patients with AF and structurally abnormal hearts, atenolol, metoprolol, or carvedilol are appropriate choices. Adequate ventricular rate control by pharmacological agents should be evaluated by either 24-hour Holter monitoring or a submaximal stress test to determine the resting and exercise ventricular rate. If the mean ventricular rate is not close to 80 beats per minute, or the heart rate on moderate exertion is not between 90 to 115 beats per minute, a second agent to control the rate should be added. Excessive reductions in ventricular rates that could limit exercise tolerance should be avoided.