Albumin-Corrected Phenytoin: the Sheiner-Tozer Equation

What the total level would read if albumin were normal, and why renal failure changes the arithmetic · v1.0

How to use this tool: Phenytoin is roughly 90% protein bound, mostly to albumin. When albumin is low, the total level under-represents how much active, unbound drug is actually present, and a total level that looks low or normal can sit alongside a free level that is already toxic. This tool corrects the measured total level for the albumin actually present, using the Sheiner-Tozer equation.

1. Measured Level and Albumin

Measured total phenytoin (mcg/mL, equal to mg/L)

Serum albumin (g/dL)

2. Renal Function

Severe renal impairment (estimated creatinine clearance below 20 mL/min) or receiving dialysis

Uraemia further reduces phenytoin's binding to albumin, independent of the albumin level itself. Ticking this changes which correction equation is used. See Renal Staging if the creatinine clearance is not already known.

Clinical Disclaimer: This is an estimate, not a measured free level. Where toxicity is genuinely in question and a free phenytoin assay is available, a directly measured free level is more reliable than any corrected estimate.

Clinical Application & Nuances

1. Why This Matters at the Bedside

Phenytoin is about 90% protein bound in a patient with normal albumin. The 10 to 20 mcg/mL total range that most laboratories quote as "therapeutic" assumes that binding. In hypoalbuminaemia, a smaller fraction of the drug is bound, so a total level of 8 mcg/mL in a patient with an albumin of 2.0 g/dL can represent the same free, active drug concentration as a total level of 15 to 16 mcg/mL in a patient with a normal albumin. Reading the total level alone in a hypoalbuminaemic patient risks two opposite errors: dismissing a toxic free level as subtherapeutic, and pushing the dose up in response to a "low" total that is not actually low once corrected.

2. Who This Applies To

Hypoalbuminaemia significant enough to matter here is common in exactly the patients most likely to be on phenytoin for an acute indication: the critically ill, the malnourished, nephrotic syndrome, cirrhosis, and pregnancy. A dose adjusted from an uncorrected total level in any of these patients is being adjusted from the wrong number.

3. Two Equations, Not One

The original Sheiner-Tozer equation assumes normal renal function. Uraemia displaces phenytoin from albumin independently of the albumin concentration itself, which is why end-stage renal disease and dialysis use a separate equation with a smaller coefficient, derived by Liponi, Winter and Tozer in 1984. Using the normal-renal equation in a dialysis patient understates the correction and can read a genuinely toxic free level as safe.

4. Where the Coefficients Come From, and Where They Do Not Agree

This tool uses the coefficients most widely taught and most widely reproduced, 0.2 for normal renal function and 0.1 for renal failure, both against a fixed 0.1 term. A 2016 systematic review found that revised coefficients, derived from larger and more contemporary patient samples, predicted measured free levels more accurately in some populations than the original equation did. There is no single equation shown to be best across every population, including critical illness, and the difference between equations matters most exactly when the correction matters most, at the extremes of albumin. Treat the corrected value here as one estimate among a documented family of them, not as a measured result.

5. A Level Does Not Diagnose Toxicity, and a Level Does Not Rule It Out

Nystagmus, ataxia, dysarthria, lethargy and, at the extreme, coma are the classical progression of phenytoin toxicity, but the level at which each appears varies between patients. A corrected level inside the reference range does not exclude toxicity in a symptomatic patient, and a level above range in an asymptomatic patient is not, on its own, a reason to withhold a needed dose. Correlate with the examination.

Abbreviations: CrCl (Creatinine Clearance) · ESRD (End-Stage Renal Disease)

Algorithm References & Evidence Base

Sheiner LB, Tozer TN. Clinical pharmacokinetics: the use of plasma concentrations of drugs. In: Melmon KL, Morrelli HF, eds. Clinical Pharmacology: Basic Principles in Therapeutics. 2nd ed. New York: Macmillan; 1978:71-72.
Liponi DF, Winter ME, Tozer TN. Renal function and therapeutic concentrations of phenytoin. Neurology. 1984;34(3):395-397.
Winter ME. Basic Clinical Pharmacokinetics. 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2010. Phenytoin chapter.
Kiang TKL, Ensom MHH. A comprehensive review on the predictive performance of the Sheiner-Tozer and derivative equations for the correction of phenytoin concentrations. Ann Pharmacother. 2016;50(4):311-325.
Soriano VV, Tesoro EP, Kane SP. Characterization of free phenytoin concentrations in end-stage renal disease using the Winter-Tozer equation. Ann Pharmacother. 2017;51(8):669-674.
Cheng W, Kiang TKL, Bring P, Ensom MHH. Predictive performance of the Winter-Tozer and derivative equations for estimating free phenytoin concentration. Can J Hosp Pharm. 2016;69(4):271-279.
Sharma PK, Pandey S, Garg RK, Takalkar K. Unexplained encephalopathy with phenytoin toxicity: hyperammonemia, the underlying cause. Neurol India. 2018;66(6):1829.
Winter ME, Tozer TN. Phenytoin. In: Evans WE, Schentag JJ, Jusko WJ, eds. Applied Pharmacokinetics: Principles of Therapeutic Drug Monitoring. 3rd ed. Vancouver, WA: Applied Therapeutics; 1992.

How to Cite This Tool

AMA Style:
Umakanth S. Albumin-Corrected Phenytoin (Sheiner-Tozer Equation). MEDiscuss. Published 2026. Accessed .

Vancouver Style:
Umakanth S. Albumin-Corrected Phenytoin (Sheiner-Tozer Equation) [Internet]. MEDiscuss.org; 2026 [cited ]. Available from:

Category Foundational CalculatorsCalculator

Specialties Internal Medicine, Neurology, Critical Care

Nobody has recorded when this content was last revised, so we cannot tell you how current it is. Check the doses and thresholds against current guidance before you use it.