Accuracy of two equations in determining normalized phenytoin concentrations

Characterization of Free Phenytoin Concentrations in End-Stage Renal Disease Using the Winter-Tozer Equation

BACKGROUND

The Winter-Tozer (WT) equation has been shown to reliably predict free phenytoin levels in healthy patients. In patients with end-stage renal disease (ESRD), phenytoin-albumin binding is altered and, thus, affects interpretation of total serum levels. Although an ESRD WT equation was historically proposed for this population, there is a lack of data evaluating its accuracy.

OBJECTIVE

The objective of this study was to determine the accuracy of the ESRD WT equation in predicting free serum phenytoin concentration in patients with ESRD on hemodialysis (HD).

METHODS

A retrospective analysis of adult patients with ESRD on HD and concurrent free and total phenytoin concentrations was conducted. Each patient's true free phenytoin concentration was compared with a calculated value using the ESRD WT equation and a revised version of the ESRD WT equation.

RESULTS

A total of 21 patients were included for analysis. The ESRD WT equation produced a percentage error of 75% and a root mean square error of 1.76 µg/mL. Additionally, 67% of the samples had an error >50% when using the ESRD WT equation. A revised equation was found to have high predictive accuracy, with only 5% of the samples demonstrating >50% error.

CONCLUSION

The ESRD WT equation was not accurate in predicting free phenytoin concentration in patients with ESRD on HD. A revised ESRD WT equation was found to be significantly more accurate. Given the small study sample, further studies are required to fully evaluate the clinical utility of the revised ESRD WT equation.

Predictive Performance of the Winter-Tozer and Derivative Equations for Estimating Free Phenytoin Concentration

BACKGROUND

The Winter-Tozer equation for estimating free phenytoin concentration is biased and imprecise. Alternative predictive equations are available, but most remain unvalidated.

OBJECTIVES

To assess the bias and precision of the Winter-Tozer equation and selected derivative equations in predicting free phenytoin concentration and to derive new equations with better predictive performance.

METHODS

A retrospective chart review (for patients with samples drawn for free phenytoin concentration between September 2008 and September 2013) was conducted for 3 subpopulations (critical care, general medicine, neurology) in one hospital. Patients were included if older than 18 years with values for free phenytoin concentration available and were excluded if phenytoin was not at steady state or if they were undergoing hemodialysis or receiving enzyme inhibitors or inducers that would affect phenytoin clearance. The predictive performance measures used were mean prediction error (MPE), root mean square error, and Bland-Altman plots. Spearman rank correlation and multiple linear regression were performed with log-transformed data.

RESULTS

In total, 133 patients were included (70 men [53%]; mean age ± standard deviation 64 ± 19 years; serum creatinine 90.4 ± 64.0 µmol/L; albumin 26.4 ± 7.0 g/L). In the combined population, the Winter-Tozer equation (MPE 1.7 µmol/L, 95% confidence interval [CI] 1.5 to 1.9) and the Anderson equation (MPE 0.5 µmol/L, 95% CI 0.3 to 0.7) over-predicted free phenytoin concentration, whereas the first Kane equation tended to underpredict free phenytoin (MPE -0.2 µmol/L, 95% CI -0.4 to 0.0), and the second Kane equation significantly underpredicted free phenytoin (MPE -0.3 µmol/L, 95% CI -0.5 to -0.1). In each subpopulation, the Winter-Tozer equation overpredicted true concentration with greater bias and imprecision. All equations performed poorly in the critical care subpopulation. Only albumin (R (2) = 0.09) and total phenytoin concentration (R (2) = 0.53) were correlated with free phenytoin concentration. The equation derived by multiple linear regression exhibited significantly less bias and imprecision than the Winter-Tozer equation in the validation set (p < 0.05). A new, user-friendly equation, specific to the authors' patient population, was derived, which had an albumin coefficient of 0.275.

CONCLUSIONS

Relatively poor predictive performance of the Winter-Tozer and derivative equations calls for more precise and less biased equations. The novel equations presented here, which had better predictive performance for free phenytoin concentration and were based on a large sample of adult patients, should be further validated in other institutions.

A Comprehensive Review on the Predictive Performance of the Sheiner-Tozer and Derivative Equations for the Correction of Phenytoin Concentrations

OBJECTIVE

In settings where free phenytoin concentrations are not available, the Sheiner-Tozer equation-Corrected total phenytoin concentration = Observed total phenytoin concentration/[(0.2 × Albumin) + 0.1]; phenytoin in µg/mL, albumin in g/dL-and its derivative equations are commonly used to correct for altered phenytoin binding to albumin. The objective of this article was to provide a comprehensive and updated review on the predictive performance of these equations in various patient populations.

DATA SOURCES

A literature search of PubMed, EMBASE, and Google Scholar was conducted using combinations of the following terms: Sheiner-Tozer, Winter-Tozer, phenytoin, predictive equation, precision, bias, free fraction.

STUDY SELECTION AND DATA EXTRACTION

All English-language articles up to November 2015 (excluding abstracts) were evaluated.

DATA SYNTHESIS

This review shows the Sheiner-Tozer equation to be biased and imprecise in various critical care, head trauma, and general neurology patient populations. Factors contributing to bias and imprecision include the following: albumin concentration, free phenytoin assay temperature, experimental conditions (eg, timing of concentration sampling, steady-state dosing conditions), renal function, age, concomitant medications, and patient type. Although derivative equations using varying albumin coefficients have improved accuracy (without much improvement in precision) in intensive care and elderly patients, these equations still require further validation.

CONCLUSIONS

Further experiments are also needed to yield derivative equations with good predictive performance in all populations as well as to validate the equations' impact on actual patient efficacy and toxicity outcomes. More complex, multivariate predictive equations may be required to capture all variables that can potentially affect phenytoin pharmacokinetics and clinical therapeutic outcomes.

Pharmacokinetic of antiepileptic drugs in patients with hepatic or renal impairment

Many factors influence choice of antiepileptic drugs (AEDs), including efficacy of the drug for the indication (epilepsy, neuropathic pain, affective disorder, migraine), tolerability, and toxicity. The first-generation AEDs and some newer AEDs are predominately eliminated by hepatic metabolism. Other recent AEDs are eliminated by renal excretion of unchanged drug or a combination of hepatic metabolism and renal excretion. The effect of renal and hepatic disease on the dosing will depend on the fraction of the AED eliminated by hepatic and/or renal excretion, the metabolic isozymes involved, as well as the extent of protein binding, if therapeutic drug monitoring is used. For drugs that are eliminated by renal excretion, methods of estimating creatinine clearance can be used to determine dose adjustments. For drugs eliminated by hepatic metabolism, there are no specific markers of liver function that can be used to provide guidance in dosage adjustments. Based on studies with probe drugs, the hepatic metabolic enzymes are differentially affected depending on the cause and severity of hepatic disease, which can aid in predicting dose adjustment when clinical data are not available. Several AEDs are also associated with laboratory markers of mild hepatic dysfunction and, rarely, more severe hepatic injury. In contrast, the risk of renal injury from AEDs is generally low. In general, co-morbid hepatic or renal diseases influence the decision for the selection of an AED. For some patients dosing changes to their existing AEDs may be appropriate. For others, a change to another AED may be a better option.

Use of antiepileptic drugs in hepatic and renal disease

The use of antiepileptic drugs in patients with renal or hepatic disease is common in clinical practice. Since the liver and kidney are the main organs involved in the elimination of most drugs, their dysfunction can have important effects on the disposition of antiepileptic drugs. Renal or hepatic disease can prolong the elimination of the parent drug or an active metabolite leading to accumulation and clinical toxicity. It can also affect the protein binding, distribution, and metabolism of a drug. The protein binding of anionic acidic drugs, such as phenytoin and valproate, can be reduced significantly by renal failure, causing difficulties in the interpretation of total serum concentrations commonly used in clinical practice. Dialysis can further modify the pharmacokinetic parameters or result in significant removal of the antiepileptic drugs. Antiepileptic drugs that are eliminated unchanged by the kidneys or undergo minimal metabolism include gabapentin, pregabalin, vigabatrin, and topiramate when used as monotherapy. Drugs eliminated predominantly by biotransformation include phenytoin, valproate, carbamazepine, tiagabine, and rufinamide. Drugs eliminated by a combination of renal excretion and biotransformation include levetiracetam, lacosamide, zonisamide, primidone, phenobarbital, ezogabine/retigabine, oxcarbazepine, eslicarbazepine, ethosuximide, and felbamate. Drugs in the latter group can be used cautiously in patients with either renal or liver failure. Antiepileptic drugs that are at high risk of being extracted by hemodialysis include ethosuximide, gabapentin, lacosamide, levetiracetam, pregabalin and topiramate. The use of antiepileptic drugs in the presence of hepatic or renal disease is complex and requires great familiarity with the pharmacokinetics of these agents. Closer follow-up of the patients and more frequent monitoring of serum concentrations are required to optimize clinical outcomes.

Phenytoin loading doses in adult critical care patients: does current practice achieve adequate drug levels?

Phenytoin is regularly employed in the critically ill for prophylaxis against or treatment of seizure disorders. No prior studies have examined current dosing practices in an Australasian intensive care unit (ICU) setting. The aims of this study were to: a) describe the adequacy of contemporary dosing in respect to free and total serum phenytoin concentrations; b) identify factors associated with therapeutic drug concentrations; and c) examine the accuracy of predictive equations that estimate free concentrations in this setting. All patients receiving a loading dose of phenytoin in a tertiary-level ICU were eligible for enrolment; 53 patients were enrolled in the study. Serum samples to determine free and total phenytoin concentrations (measured by high performance liquid chromatography) were then drawn prior to the following dose. Free concentrations below the recommended target (<1 mg/l) were considered as suboptimal. The most common indication for phenytoin loading was traumatic brain injury (49%) and the mean administered dose was 14.5 (3.66) mg/kg. Twenty-six patients (49%) had suboptimal trough free concentrations, although this subgroup was significantly heavier and therefore received a lower per kilogram dose (12.8 [3.1] vs 16.3 [3.4] mg/kg, P=0.001). In multivariate analysis, larger weight adjusted doses (P=0.018), higher albumin concentration (P=0.034) and receiving phenytoin for an indication other than seizure (P=0.035), were associated with a greater likelihood of adequate concentrations. In conclusion, phenytoin dosing remains complex in critically ill patients, although lower per kilogram loading doses are strongly associated with free concentrations below the desired target.

Characterization of unbound phenytoin concentrations in neurointensive care unit patients using a revised Winter-Tozer equation

BACKGROUND

Prior studies examining the accuracy of the Winter-Tozer (WT) equation for correcting total phenytoin concentrations in critically ill patients have yielded conflicting results and are limited by small sample sizes and stringent exclusion criteria, which lessen external validity.

OBJECTIVE

To determine whether the traditional WT equation is appropriate in correcting total phenytoin concentrations in a large sample of patients in a neurointensive care unit (NICU) and whether a new equation may be more predictive.

METHODS

In a retrospective study, NICU patients with reports of a concurrent total and unbound phenytoin concentration and albumin level were analyzed. Two new predictive equations were generated using a revised WT equation and regression model of baseline and laboratory characteristics. Prediction error analysis using a 20% validation cohort was conducted on all 3 equations for comparison.

RESULTS

A total of 140 adults were included for data analysis, with data on 80% used for derivation and 20% as validation of all equations. The mean unbound phenytoin concentration was 1.4 μg/mL, which represented a free fraction of 10%. Most samples were collected within 24 hours of NICU admission. Multivariate regression analysis demonstrated that albumin, total phenytoin concentration, sex, and creatinine clearance were predictive of measured unbound phenytoin concentrations. The traditional WT equation significantly underpredicted true unbound phenytoin concentrations, with 32.1% of patients having a prediction error of more than 50% in the validation cohort.

CONCLUSIONS

The traditional WT equation was significantly biased in underpredicting true unbound phenytoin concentrations in neurointensive care unit patients and should not be used in this setting. Two modified equations were more accurate and precise and should be considered for use when unbound phenytoin concentrations are not readily available in an NICU population.

Phenytoin-rifampin drug interaction in a hypoalbuminemic, renal failure patient: a complex clinical case

Phenytoin, a commonly used antiepileptic, is difficult to dose optimally due to its narrow therapeutic window, nonlinear pharmacokinetics, extensive protein binding, and participation in clinically significant drug interactions. Although clinicians are aware of the interaction with two widely used antituberculosis agents, rifampin and isoniazid, few reports have described the implications for managing phenytoin dosing in this situation. To our knowledge, only two reports of the clinical experience with this interaction have been published, and only one of these reports involved the addition of isoniazid. We present a case of a 60-year-old man treated with triple antiepileptic therapy, including phenytoin, who experienced seizures shortly after hospital admission. Dosing of phenytoin proved difficult in this patient due to an acute kidney injury and severe hypoalbuminemia requiring hemodialysis. A further complexity was the addition of antituberculosis therapy (rifampin, isoniazid, pyrazinamide, and ethambutol [RIPE]) for suspected tuberculosis meningitis after the patient experienced persistent encephalopathy. Phenytoin concentrations decreased steadily after rifampin and isoniazid initiation despite dose increases, and the free concentration of phenytoin reached a low of less than 0.5 µg/ml on day 8 of RIPE therapy. The patient continued on a stable dose of phenytoin and RIPE therapy for unconfirmed tuberculosis meningitis until discharge. This report is the first description of this drug interaction in 20 years and highlights the need for appropriate management of phenytoin in a patient with complicated needs for pharmacotherapy.

Monitoring free drug concentrations: challenges

Measurement of drug concentrations in biological samples is of utmost importance in many research areas. The information about the amount of drug in a biological sample can be given as either total concentration, which ignores the interaction of the drug with the sample matrix, or as free concentration, which shows the portion of molecules able to diffuse through membranes and exert biological activity. Although the historical trend has been towards determining total concentrations, measurement of free concentrations is becoming more important since it correlates better with pharmacological and toxicological effects. This review will discuss the most popular experimental approaches for monitoring free drug concentrations, based on the type of sample to be investigated and the kind of information to be collected. It is shown that the current challenges in measuring free concentrations are: convenience, accuracy, precision, wide applicability, availability of accurate and precise reference methods, ruggedness, and standardized sample conditions.

Therapeutic drug monitoring of phenytoin in critically ill patients

Therapeutic drug monitoring of phenytoin is necessary to ensure therapeutic and nontoxic levels. Hypoalbuminemia, renal failure, and interactions with other highly protein-bound drugs (e.g., valproic acid) alter protein binding of phenytoin. When these conditions are present, free serum concentrations, which represent the pharmacologically active entity, cannot be predicted from total serum concentrations. Besides general alterations in drug distribution and elimination, protein binding is often altered in critically ill patients. Case reports describe phenytoin toxicity secondary to inappropriate dosage adjustments based solely on total serum concentrations in patients with hypoalbuminemia. Free drug measurements and theoretical equations to facilitate the interpretation of total phenytoin serum levels have been introduced. However, they are not widely implemented in clinical practice because evidence of improvements in patient outcomes is limited. Knowledge of the pharmacokinetic properties of phenytoin is indispensable for correct interpretation of total serum concentrations when protein binding is altered. Free serum concentrations should be measured, or theoretically calculated if measurements are unavailable, to avoid misinterpretation of total serum levels and consequent inappropriate adjustments in the dosage of phenytoin in critically ill patients.

Total phenytoin concentrations do not accurately predict free phenytoin concentrations in critically ill children

OBJECTIVE

To determine the relationship between estimated free, measured free, and measured total phenytoin levels in critically ill pediatric patients, assess the utility of the Sheiner-Tozer equation in predicting free phenytoin levels, and identify comedications that may influence phenytoin binding or confound attempts to maintain therapeutic concentrations.

DESIGN

Retrospective chart review.

SETTING

Twenty-four-bed medical-surgical pediatric intensive care unit.

PATIENTS

Sixty critically ill pediatric patients receiving phenytoin for treatment of seizures in a large multidisciplinary intensive care unit.

INTERVENTIONS AND MAIN RESULTS

The linear correlation between free and total phenytoin concentrations was moderate (r = .795), but the mean difference between actual free concentrations and those estimated from total concentrations using the Sheiner-Tozer equation was -0.31 +/- 0.5 microg/mL (95% confidence interval, -1.3 to 0.7). This difference was of concern, as 10% of patients had toxic free levels (>2 microg/mL) when simultaneously measured total levels were therapeutic (<20 microg/mL). The mean free/total phenytoin ratio was 0.13 +/- 0.07 (range, 0.06-0.42) and varied considerably among patients. Free fractions were particularly elevated in children whose serum albumin concentrations were <2.5 g/dL (0.22, p < .001). However, the relationship between free phenytoin and serum albumin concentration appeared to be nonlinear. Coadministration of valproic acid and cefazolin also increased free fraction (p < .001).

CONCLUSIONS

Measured total phenytoin concentrations are unreliable for directing therapy in critically ill children. In part, this is because phenytoin binding shows greater variability in this population than has been reported in adults. This phenomenon is exacerbated by coadministration of other highly protein-bound drugs. Instead, free phenytoin concentrations should be routinely measured in critically ill children to prevent possible intoxications and ensure therapeutic dosing. Corrections using the Sheiner-Tozer equation were unreliable.