When phenytoin and valproic acid are used concurrently, how should total and free phenytoin levels be calculated and monitored?

Medical Advisory BoardAll articles are reviewed for accuracy by our Medical Advisory Board
Educational purpose only • Exercise caution as content is pending human review
Article Review Status
Submitted
Under Review
Approved

Last updated: October 7, 2026 • View editorial policy

Have a follow-up question?

Our Medical A.I. is used by practicing medical doctors at top research institutions around the world. Ask any follow up question and get world-class guideline-backed answers instantly.

Monitoring Phenytoin Levels with Concurrent Valproic Acid Use

When phenytoin and valproic acid are used together, you must measure or calculate free (unbound) phenytoin levels rather than relying on total phenytoin levels, as valproic acid significantly increases the free fraction of phenytoin through protein displacement and metabolic inhibition.

The Critical Drug Interaction

Valproic acid displaces phenytoin from plasma albumin binding sites and inhibits its hepatic metabolism. Coadministration increases the free fraction of phenytoin by approximately 60%, while total plasma clearance and apparent volume of distribution of phenytoin increase by 30% 1. This means total phenytoin levels become unreliable indicators of therapeutic effect or toxicity.

Monitoring Strategy

Preferred Approach: Direct Measurement

  • Measure free (unbound) phenytoin levels directly whenever this assay is available
  • Target therapeutic range for free phenytoin: 1-2 mcg/mL (compared to 10-20 mcg/mL for total phenytoin in patients not on valproic acid)
  • Free phenytoin measurement is the gold standard and avoids estimation errors 2, 3

Alternative: Calculation When Direct Measurement Unavailable

If free phenytoin assays are not readily available, use the Haidukewych equation for estimation:

Free phenytoin (mcg/mL) = 0.095 × Total phenytoin + 0.001 × Valproic acid × Total phenytoin

(All concentrations in mcg/mL)

This equation demonstrates the least bias and most precision compared to alternatives 2. The May equation consistently underpredicts free phenytoin concentrations and should be avoided.

Important Caveats About Calculations

  • Prediction equations have significant limitations, particularly in critically ill patients where mean differences between actual and estimated free phenytoin can exceed 0.8 mcg/mL 3
  • ICU patients show greater discrepancies than non-ICU patients
  • Hypoalbuminemia (common in critically ill patients) further increases free drug concentrations 4
  • Even with the best equation, the free phenytoin fraction is not constant and usually exceeds the expected 10% in patients comedicated with valproic acid 2

Clinical Monitoring Algorithm

  1. Obtain baseline measurements: Total phenytoin, valproic acid, and albumin levels
  2. If free phenytoin assay available: Measure directly and target 1-2 mcg/mL
  3. If free phenytoin assay unavailable:
    • Use Haidukewych equation for estimation
    • Consider albumin levels in interpretation (hypoalbuminemia increases free fraction)
    • Monitor closely for clinical signs of toxicity regardless of calculated values
  4. Adjust phenytoin dosage based on clinical response and free (not total) phenytoin levels 1
  5. Monitor frequently after any dose changes or changes in valproic acid dosing

Critical Warning Signs

Watch for breakthrough seizures despite "therapeutic" total phenytoin levels—this may indicate the combination is ineffective 1. Conversely, 57% of patients may have supratherapeutic free phenytoin levels even when total levels appear within normal range 4.

Signs of phenytoin toxicity include ataxia, nystagmus, diplopia, and altered mental status 5. These can occur even with "normal" total phenytoin levels when valproic acid is present.

Method-Specific Considerations

The accuracy of predictive equations depends on the analytical method used. HPLC measurements provide more accurate predictions than standard TDx assays, which tend to yield higher total and free phenytoin concentrations 6. Ensure consistency in laboratory methodology when monitoring over time.

Related Questions

For a patient with seizure disorder on divalproex 500 mg three times daily and phenytoin (Dilantin) 300 mg at bedtime, with serum albumin 32 g/L, total valproic acid concentration 256 µmol/L and adjusted phenytoin concentration 50 µmol/L, should the medication doses be changed?
What is the interaction between valproic acid (VPA) and phenytoin?
What effect does valproic acid have on phenytoin serum concentrations?
What labs should be monitored for patients on phenytoin (Dilantin)?
How should phenytoin (Dilantin) serum levels be drawn for therapeutic drug monitoring?
What is the recommended management for a patient with diverticular disease presenting as a severe complicated acute recurrence (SCAR)?
Is there a difference between ferrous sulfate 65 mg delayed‑release (enteric‑coated) and regular (immediate‑release) ferrous sulfate?
Given a normal sinus CT, an endoscopy showing gastro‑esophageal reflux disease (GERD) with inflamed adenoids and tonsils, and antibiotics that reduced adenoid swelling but left the tonsils enlarged, what is the likely cause of my persistent tonsillar swelling?
Can a diabetic patient who cannot afford Jardiance (empagliflozin) be managed with glyburide 5 mg and metformin 500 mg twice daily to achieve an HbA1c of 7.5%?
In an otherwise healthy patient who has gained weight and has no alarm features, does this argue against gastric malignancy?
What is the recommended dosing, route, and duration of diazepam for acute musculoskeletal spasm in adults?

Professional Medical Disclaimer

This information is intended for healthcare professionals. Any medical decision-making should rely on clinical judgment and independently verified information. The content provided herein does not replace professional discretion and should be considered supplementary to established clinical guidelines. Healthcare providers should verify all information against primary literature and current practice standards before application in patient care. Dr.Oracle assumes no liability for clinical decisions based on this content.