Dezinamide for partial seizures: Results of an n-of-1 design trial

Michael D. Privitera, D. M. Treiman, G. W. Pledger, J. T. Sahlroot, A. Handforth, M. S. Linde, C. R. France, J. J. Cereghino, C. B. McCutchen, J. H. Wood

Research output: Contribution to journalArticlepeer-review

13 Scopus citations


Background. Dezinamide (DZM, ADD 94057) is a potential antiepileptic drug that binds to the voltage-sensitive sodium channel and showed preliminary evidence of efficacy and safety in an open-label study. Methods. Our double-blind, placebo-controlled trial at two sites used an n-of-1 (single-patient) design. All 15 patients had medically intractable partial-onset seizures and were comedicated with phenytoin (PHT) only. Treatment was for six 5-week periods (three active paired with three placebo in random sequence). Assuming nonlinear kinetics, we used an initial pharmacokinetic profile to estimate dosages for reaching target plasma concentrations of DZM. Results. Statistically significant seizure reduction was found by both a randomization test (p = 0.0025) and a signed rank test (p = 0.048). Median seizure frequency decreased 37.9%, and 40% of patients had > 50% seizure reduction, both compared with placebo. Pharmacokinetic predictions were not accurate; mean plasma concentrations fell well below target values. Plasma PHT concentrations increased (mean = 17.1%) during DZM treatment. The most common adverse experiences were fatigue, light-headedness, and abnormal gait; five patients required DZM dosage reductions. Conclusions. DZM showed minimal clinical toxicity and significant efficacy despite lower plasma concentrations than predicted by pharmacokinetics. This trial establishes the suitability of the n-of-1 design to investigational antiepileptic drug trials.

Original languageEnglish (US)
Pages (from-to)1453-1458
Number of pages6
Issue number8
StatePublished - Aug 1994
Externally publishedYes

ASJC Scopus subject areas

  • Clinical Neurology


Dive into the research topics of 'Dezinamide for partial seizures: Results of an n-of-1 design trial'. Together they form a unique fingerprint.

Cite this