One-point feedback control method for phenytoin dosage adjustment.

Routine clinical pharmacokinetic data collected from patients receiving phenytoin have been analysed to propose a new and simple equation to aid the dosage adjustment of this drug. The data were analysed using NONMEM, a computer program designed for population pharmacokinetic analysis that allows pooling of data. The rate equation for the elimination of phenytoin can be written as Do = kCssn, which fits the steady-state serum concentration (Css) and daily dose data (Do). The parameter n is the kinetic order and the parameter k is an arbitrary rate constant. From the above equation, D2 = D1C1 -nC2n can be derived, which forms the basis of predicting the dosage, D2, to obtain a desired Css, C2, using one initial Css, C1, obtained with an initial dose, D1, and using a population value of n. The value of n for phenytoin was estimated to be 0.312 in this study. The predictive performance of this equation was compared with the Richens and Dunlop nomogram and Bayesian feedback method using two or more steady-state concentration/dose pairs from each of 78 outpatients. This equation allowed the prediction of a dose needed to produce a desired steady-state concentration with errors comparable with the Bayesian feedback method for therapeutic drug monitoring.     

Phenytoin dosage predictions in paediatric patients

Phenytoin dosing in paediatric patients is complicated both by alterations in patient requirements due to growth and maturation changes and by the capacity-limited characteristics of phenytoin metabolism. This study examines 2 pharmacokinetic methods to adjust phenytoin dosage based on a single dosing-rate/steady-state serum phenytoin concentration pair. A Bayesian forecaster and a fixed parameter [rate of metabolism (Vmax)] method were examined with previously published sets of a priori parameter estimates. The fixed Vmax method was utilised with the parameter derived from native Japanese (method 1), US Caucasian (method 2) and European (method 3) patients. The Bayesian forecaster used a priori parameter estimates obtained from native Japanese (method 4) and European (method 5) patients. Each method was examined retrospectively in 34 paediatric patients with a total of 48 predictions possible. Measures of absolute predictability, bias (mean error, % dose) and precision (root mean squared error, % dose), were -3.58/12.2, -1.51/12.2, 4.06/9.96, -4.38/13.2, and -3.10/11.5, for methods 1, 2, 3, 4 and 5, respectively. There was no significant difference among the 5 methods. However, the Bayesian algorithm tended to be more robust over a broad range of situations, providing predictions in all cases. The fixed Vmax methods could not provide predictions in every case. Finally, all methods had a significant number of overpredictions of dosage. Poorer results were observed when prediction of steady-state serum concentrations were performed, partly due to the retrospective nature of the study. We conclude that close monitoring of patients, regardless of the method chosen to adjust dosage, is recommended.     

Evaluation of a proposed method for phenytoin maintenance dose prediction following an intravenous loading dose

Summary A large clinical study, designed to investigate the induction of theophylline metabolism by phenytoin, provided the opportunity to test a previously proposed method for estimating dose requirements of phenytoin. This method involves prediction of the oral maintenance dosage from data obtained following the administration of an intravenous loading dose.In 30 subjects, trough plasma concentration at steady-state were 12.0±4.9 µg·ml^–1 (mean ±SD) and differed by –2.7±39.3% from a mean target plasma concentration of 12.5±1.5µg·ml^–1.A Bayesian regression programme was used to forecast an estimate of each subject's individual pharmacokinetics. These were then used to predict the steady-state plasma concentrations which would be expected from a standard dosing regimen (4 mg per kg per day). When compared to the results expected from the use of this standard dosage, the proposed method gave acceptable steady-state plasma phenytoin concentrations with significant reductions in deviations from target concentrations.This method for the rapid individualization of phenytoin dosage requirements provides an improvement over more traditional methods of choosing an arbitrary dose adjusted for body weight followed by dosage adjustments based on achieved plasma concentration.     

Comparison of four single-point phenytoin dosage prediction techniques using computer-simulated pharmacokinetic values

The predictive abilities of the following four single-point phenytoin dosage adjustment methods were compared using computer-simulated data: the Bayesian feedback method of Vozeh et al. (B), a linearized version of the Bayesian method (LB), the population-clearance method of Graves et al. (G), and the Rambeck nomogram (R). A series of 512 "subjects" with normally distributed values for volume of distribution, weight, and the Michaelis-Menten variables Vmax and Km were simulated. The steady-state serum concentration (SSSC) resulting from the administration of a standard dose of phenytoin sodium (5 mg/kg/day) was calculated, and "subjects" with SSSCs less than or equal to 12 mg/L or greater than or equal to 17 mg/L were entered in the study. If the concentration was greater than 50 mg/L or the standard dosage exceeded Vmax, the dosage was reduced empirically by 25%. Normally distributed random errors were introduced into the SSSC values to simulate actual patient data. The pharmacokinetic values, dosages, and SSSCs were used for predicting the dosage required to attain an SSSC of 14.9 mg/L. In the unstratified population, the mean error and mean-squared error were lowest for methods G and B, followed by methods LB and R. Methods B and LB gave the highest percentages of satisfactory dosage predictions based on the resultant SSSC value. The performance of all methods was superior at initial SSSCs greater than 8 mg/L.(ABSTRACT TRUNCATED AT 250 WORDS)     

A comparison of the accuracy of a least squares regression, a Bayesian, Chiou's and the steady-state clearance method of individualising theophylline dosage

We compared a least squares regression method, used prospectively to individualise the intravenous aminophylline and oral theophylline dosage of 48 patients, with 3 other pharmacokinetic methods - Chiou's, the steady-state clearance and the Bayesian - used retrospectively to analyse the same patient data. Methods were compared on the basis of the similarity of their parameter estimates and the accuracy with which serum concentrations during subsequent intravenous and oral therapy could be forecast, assuming each method's parameter estimates. The least squares and Bayesian programs were able to fit data from all but 4 and 2 patients, respectively. Mean absolute prediction errors were of the order of 20% for serum concentrations during intravenous therapy, and of the order of 40% for serum concentrations during oral therapy. The accuracy of the least squares, Bayesian and steady-state clearance methods were similar, but the accuracy of Chiou's method was comparable only when the 2 serum concentrations needed for the method were measured between 11 and 17 hours apart; an interval which corresponds to the 1.0 to 1.5 half-lives previously suggested as desirable for implementation of the Chiou method.     

Individualization of theophylline dosage: a Bayesian method evaluated in patients with once-daily dosing

A Bayesian drug dosing program was prospectively evaluated in 16 hospitalized patients with obstructive respiratory disease treated with a novel sustained-release preparation Euphylong. Theophylline was given as a single evening dose. By means of the Bayesian method, the early morning peak was predicted from a steady-state theophylline serum level determined at 8:00 a.m. The prediction errors were in all cases within a clinically acceptable range (mean prediction error +/- standard deviation: -0.2 +/- 1.3 mg/L). The results were compared with predictions based on steady state theophylline serum levels at 8:00 p.m. The prediction errors observed were on average greater than the prediction errors based on the serum levels at 8:00 a.m. (-0.6 +/- 2.5 mg/L). These findings suggest that the Bayesian method is useful for dosage predictions in patients treated once a day with Euphylong.     

Steady-state serum levels of anticonvulsant drugs in Chinese epileptic patients living in Taiwan

Steady-state serum concentrations of phenytoin and phenobarbital were obtained in 70 epileptic patients in Taiwan and evaluated with respect to age, sex, and body weight. As demonstrated, the steady-state serum level of phenytoin is not affected by the age and sex of the patients or by phenobarbital used as a comedication. Thus, the main contributing factor influencing the steady-state serum concentration of phenytoin is the drug dosage. Among Chinese epileptic patients, the saturation of phenytoin metabolism occurs at the dosage of 4 mg/kg/day.     

A general method for optimal drug dose computation

A general method is presented for computing drug regimens which are optimal in the sense of minimizing the sum of the squared deviations of the predicted drug concentration in a compartment from a desired concentration in that compartment. Constraints on the predicted drug concentration in other compartments can be specified. The method is designed to allow bolus doses (parenteral or oral) to be given or intravenous infusion rates to be changed only at prespecified times. This feature permits use of the proposed method to develop dosage regimens which may be useful clinically as compared with other proposed optimal dose computation methods which yield very complex regimens.This work was supported in part by Grant GM-16496 and Training Grants GM-00001 and GM-01791.     

基于群体药动学模型的万古霉素个体化给药软件研制及应用

目的:研发基于建立的成人和老年群体药动学(population pharmacokinetics,PPK)模型的万古霉素(vancomycin,VCM)个体化给药软件。方法:根据已建立的成人和老年VCM的PPK模型信息,运用MyEclipse、SQL Server、JRE等工具软件研发VCM给药软件。软件开发方案包括需求分析,概要设计,详细设计,软件编码,软件测试以及软件维护和二次开发。结果:研制的VCM给药软件可实现感染患者信息输入和管理,软件通过接口调用非线性混合效应模型(NONMEM)软件,不仅能预测多种具体VCM给药方案下的血药浓度,供临床医师制定初始用药方案参考,而且能结合已有的血药浓度监测信息和贝叶斯反馈法更精准地预测血药浓度,辅助临床医师进一步优化给药方案。软件应用于VCM血药浓度解读,药师向临床做出剂量调整建议。采纳建议组患者复查的血药浓度均达到目标血药浓度范围。结论:本研究基于VCM的PPK模型研制的给药软件能快速方便地辅助成人和老年感染患者VCM的个体化给药。     

Predicting phenytoin dosages using Bayesian feedback: a comparison with other methods

A Bayesian feedback technique for predicting phenytoin dosage was compared to other dosing methods. Sixty-nine cases were selected on the basis of apparent reliability from 103 medical charts of epileptic patients with multiple phenytoin levels on different dosage regimens. Two published nomograms and a graphical, or computational, technique were compared to the Bayesian technique. Each method was assessed for absolute predictability using measures of bias and precision, i.e., mean percent error and root mean squared percent error, respectively. For a single previous data pair, the Bayesian method was similar to a published nomogram with regard to bias and precision. For multiple data pairs, the graphical or simultaneous equation technique tended to be less biased, but the Bayesian method had better precision. However, none of these differences was statistically significant (p greater than 0.05). The Bayesian method yielded the lowest percentage of predicted doses that exceeded 110% of the actual dose. The Bayesian method conveniently provides a single method applicable to the use of either single or multiple concentration-dosage data pairs and results in fewer extreme dosing errors.     

癫痫患者苯妥英的治疗监测

特殊癫痫患者ＤＰＨ的给药方案主要依据临床效应，无须强调靶范围。群体药动学参数是优化个体给药的前提，稳态Ｂａｙｅｓｉａｎ法能较准确的预测给药速率，非稳态Ｂａｙｅｓｉａｎ法能估算Ｖｍａｘ、Ｋｍ、Ｖｄ、Ｆ，并预测中毒患者停药后何时恢复用药及用药剂量。游离浓度较总浓度更好地指导临床治疗，超滤法结合ＦＰＩＡ适用于常规监测ＤＰＨ游离浓度。唾液可取代血液样品用于游离药物监测。     

An updated comparison of drug dosing methods. Part I: Phenytoin

The relationship between a dose of phenytoin and the resultant serum concentration is difficult to predict, and numerous dosing methods have been developed to quantify the dose required to achieve a specific concentration. This review brings up to date the earlier article in the Journal regarding predictive algorithms, various pharmacokinetics-based dosing techniques and Bayesian feedback methods for phenytoin dosing. The latest data support the original conclusions that dosing methods for phenytoin which incorporate an individualised approach or Bayesian principles tend to offer results superior to those from predictive algorithms. Bayesian methods have the additional advantage of using only 1 serum concentration, obtained under either steady-state or non-steady-state conditions. There is still a need for future investigations that include prospective evaluations of predictive performance and cost-effectiveness data.     

Maximum a posteriori Bayesian estimation of mycophenolic Acid area under the concentration-time curve: is this clinically useful for dosage prediction yet?

This review seeks to summarize the available data about Bayesian estimation of area under the plasma concentration-time curve (AUC) and dosage prediction for mycophenolic acid (MPA) and evaluate whether sufficient evidence is available for routine use of Bayesian dosage prediction in clinical practice. A literature search identified 14 studies that assessed the predictive performance of maximum a posteriori Bayesian estimation of MPA AUC and one report that retrospectively evaluated how closely dosage recommendations based on Bayesian forecasting achieved targeted MPA exposure. Studies to date have mostly been undertaken in renal transplant recipients, with limited investigation in patients treated with MPA for autoimmune disease or haematopoietic stem cell transplantation. All of these studies have involved use of the mycophenolate mofetil (MMF) formulation of MPA, rather than the enteric-coated mycophenolate sodium (EC-MPS) formulation. Bias associated with estimation of MPA AUC using Bayesian forecasting was generally less than 10%. However some difficulties with imprecision was evident, with values ranging from 4% to 34% (based on estimation involving two or more concentration measurements). Evaluation of whether MPA dosing decisions based on Bayesian forecasting (by the free website service https://pharmaco.chu-limoges.fr) achieved target drug exposure has only been undertaken once. When MMF dosage recommendations were applied by clinicians, a higher proportion (72-80%) of subsequent estimated MPA AUC values were within the 30-60?mg?·?h/L target range, compared with when dosage recommendations were not followed (only 39-57% within target range). Such findings provide evidence that Bayesian dosage prediction is clinically useful for achieving target MPA AUC. This study, however, was retrospective and focussed only on adult renal transplant recipients. Furthermore, in this study, Bayesian-generated AUC estimations and dosage predictions were not compared with a later full measured AUC but rather with a further AUC estimate based on a second Bayesian analysis. This study also provided some evidence that a useful monitoring schedule for MPA AUC following adult renal transplant would be every 2 weeks during the first month post-transplant, every 1-3 months between months 1 and 12, and each year thereafter. It will be interesting to see further validations in different patient groups using the free website service. In summary, the predictive performance of Bayesian estimation of MPA, comparing estimated with measured AUC values, has been reported in several studies. However, the next step of predicting dosages based on these Bayesian-estimated AUCs, and prospectively determining how closely these predicted dosages give drug exposure matching targeted AUCs, remains largely unaddressed. Further prospective studies are required, particularly in non-renal transplant patients and with the EC-MPS formulation. Other important questions remain to be answered, such as: do Bayesian forecasting methods devised to date use the best population pharmacokinetic models or most accurate algorithms; are the methods simple to use for routine clinical practice; do the algorithms actually improve dosage estimations beyond empirical recommendations in all groups that receive MPA therapy; and, importantly, do the dosage predictions, when followed, improve patient health outcomes?     

Integrated equation to evaluate accumulation profiles of drugs eliminated by Michaelis-Menten kinetics

Using an equation for the calculation of plasma profiles of drug based on the zero-order input and Michaelis-Menten kinetic output in a one-compartment open model system, the times required to reach various degrees of several steady-state plasma concentrations of phenytoin are calculated. The effect of the apparent volume of distribution (due to intersubject variation or change in protein and/or tissue binding, etc.) on the time required to reach various fractions of steady-state plasma concentrations is discussed.     

Computer-Aided Dosage Form Design. III. Feasibility Assessment for an Oral Prolonged-Release Phenytoin Product

Previous publications described computer-aided methodology for assessing the feasibility of designing prolonged release oral dosage forms containing linear-disposition drugs. Those methods determined all useful release rates and examined those rates to decide whether product development was warranted. The present study developed software to obtain similar information for phenytoin, which exhibits Michaelis–Menten disposition. The values for V _max, K _m, and V _d in 27 patients were employed to assess the ability of prolonged absorption to maintain steady-state plasma concentrations between 10 and 20 mg/liter following oral administration at 8-, 12-, and 24-hr intervals. Phenytoin steady-state plasma concentrations in this range were controlled by elimination and were not extended by prolonged absorption. Furthermore, single i.v. bolus doses resulting in an initial plasma level of 20 mg/liter provided concentrations above 10 mg/liter for ~1 to 3 days. When an oral multiple-dose regimen was found to maintain steady-state concentrations between 10 and 20 mg/liter, that dose and interval produced concentrations within that range regardless of the absorption rate. While absorption rate was not important, each patients dose ranges were extremely narrow, emphasizing that dose size was the dominant factor in the control of phenytoin levels.     

Failure of traditional trough levels to predict tacrolimus concentrations

The objective of this study was to estimate tacrolimus population parameter values and to evaluate the ability of the maximum a posteriori probability (MAP) Bayesian fitting procedure to predict tacrolimus blood levels, using the traditional strategy of monitoring only trough levels, for dosage individualization in liver transplant patients. Forty patients treated with tacrolimus after liver transplantation were studied during the early posttransplant phase (first 2 weeks). This phase was divided into four time periods (1-4 days, 5-7 days, 8-11 days, 12-14 days). Tacrolimus was administered twice daily. Approximately one determination of a tacrolimus trough level on whole blood was performed each day. The NPEM2 program was used to obtain population pharmacokinetic parameter values. With each individual pharmacokinetic parameter estimated by the MAP Bayesian method for a given period, the authors evaluated the prediction of future levels of tacrolimus for that patient for the next period. This evaluation of Bayesian fitting predictive performance was performed using the USC*PACK clinical software. Mean pharmacokinetic parameter values were in the same general range as previously published values obtained with richer data sets. However, during each period, the percentage of blood levels predicted within 20% did not exceed 40%. The traditional strategy of obtaining only trough whole blood levels does not provide enough dynamic information for the MAP Bayesian fitting procedure (the best method currently available) to be used for adaptive control of drug dosage regimens for oral tacrolimus. The authors suggest modifying the blood concentration monitoring scheme to add at least one other concentration measured during the absorptive or distributive phase to obtain more information about the behavior of the drug. D-Optimal design and similar strategies should be considered.     

The accuracy and stability of Bayesian theophylline predictions

Pharmacokinetic parameters for theophylline were determined in 33 patients (3 women), mean age 61.2 years and weight 74.6 kg using the following three methods: (a) standard one-compartmental model calculations, assuming 100% bioavailability, after a single dose of theophylline syrup (mean dose 413 mg); (b) drug nomogram; and (c) Bayesian analysis. Patients entered a randomised study of three two-monthly dosage regimens using low, medium, and high theophylline twice daily doses. These doses produced mean (+/- SE) steady-state serum theophylline concentrations of 6.3 (+/- 0.4), 12.1 (+/- 0.3) and 18.3 (+/- 0.5) mg/L, respectively. A fourth period of placebo (2-month duration) was also included. At the end of each treatment period the measured serum theophylline concentration of each patient was compared with those predicted by each of the above three methods. The revised estimates derived from Bayesian analysis produced the least biased [mean prediction error (ME)] and most precise (mean squared prediction error) predictions for all three dosage periods. Statistical analysis of relative performance demonstrated that the difference in precision between the revised estimates and those of the other two methods was significant (p less than 0.05) with the magnitude of the difference increasing with dose. The revised estimates were also found to be less biased (p less than 0.05) than those of the nomogram. The ME (+/- SE) of the revised estimates for the low, medium, and high dosage periods was 0.34 (+/- 0.30), -0.02 (+/- 0.22) and -0.48 (+/- 0.31) mg/L, respectively.     

Novel method for rapid reversal of drug toxicity: A case report

Drug toxicity is traditionally treated by reducing the amount of the drug absorbed, enhancing elimination, and providing supportive care. Once the drug has been absorbed, there are few methods that help decrease morbidity and mortality caused by a toxic drug level. Albumin infusion is a new approach that changes that, as it can rapidly reverse a toxic drug level back to a therapeutic level. It is believed with most drugs that the toxic effects are related to the total amount of the free drug. In this method, albumin binds to the free drug and acts as a reservoir or depot from which the drug is slowly released to the free form, thereby limiting the effects of drug toxicity. In this case report, an elderly female patient who experienced phenytoin toxicity was treated with albumin infusion, after which her phenytoin level returned to a therapeutic level with corresponding improvements in her symptoms. Based on our calculations, it was predicted that a small amount of albumin would reverse the patient's toxic symptoms. With this approach, the patient's toxic symptoms improved when free phenytoin levels dropped from 15 to 8 μmol/L. Albumin infusion is a promising new therapy that can rapidly reverse a toxic drug level back to a therapeutic level by binding the free drug to albumin.     

Phenytoin clearances in a compliant population: description and application

The saturable elimination kinetics of phenytoin (PHT) makes accurate dosing difficult. A simple equation has been derived to predict dosing requirements if one dose--steady-state serum concentration pair is known [new dose = first dose X Cpss (desired)0.2 X Cpss (achieved)-0.2]. This equation is based on an exponential relation between 177 pairs of PHT clearance and steady-state serum concentrations from 59 compliant patients. To evaluate this population clearance method (PCM), patients were drawn from two independent populations (Minnesota and Iowa, U.S.A.). Predicted doses obtained from PCM were compared with predictions from Bayesian, average Vmax, and average Km methods. The Bayesian method was the most precise and least biased of all methods, under- and overpredicting doses in equal frequency. PCM was biased to produce underpredictions. However, clinically achievable doses can be obtained by consistently rounding the calculated dose upward. The mean underprediction for the Bayesian method was 6 mg, for PCM 46 mg, for average Vmax 21 mg, and for average Km 11 mg. PCM is relatively precise, somewhat biased, and very easy to use.