Intravenous topiramate: Safety and pharmacokinetics following a single dose in patients with epilepsy or migraines taking oral topiramate


  • Anne M. Clark,

    1. Upsher-Smith Laboratories, Minneapolis, Minnesota, U.S.A
    Search for more papers by this author
  • Robert L. Kriel,

    1. Center for Orphan Drug Research, University of Minnesota, Minneapolis, Minnesota, U.S.A
    2. Department of Experimental and Clinical Pharmacology, College of Pharmacy, University of Minnesota, Minneapolis, Minnesota, U.S.A
    Search for more papers by this author
  • Ilo E. Leppik,

    1. Department of Experimental and Clinical Pharmacology, College of Pharmacy, University of Minnesota, Minneapolis, Minnesota, U.S.A
    2. MINCEP Epilepsy Care, Minneapolis, Minnesota, U.S.A
    3. Department of Neurology, University of Minnesota, Minneapolis, Minnesota, U.S.A
    Search for more papers by this author
  • James R. White,

    1. Minnesota Epilepsy Group, St. Paul, Minnesota, U.S.A
    Search for more papers by this author
  • Thomas R. Henry,

    1. Department of Neurology, University of Minnesota, Minneapolis, Minnesota, U.S.A
    Search for more papers by this author
  • Richard C. Brundage,

    1. Department of Experimental and Clinical Pharmacology, College of Pharmacy, University of Minnesota, Minneapolis, Minnesota, U.S.A
    Search for more papers by this author
  • James C. Cloyd

    Corresponding author
    1. Center for Orphan Drug Research, University of Minnesota, Minneapolis, Minnesota, U.S.A
    2. Department of Experimental and Clinical Pharmacology, College of Pharmacy, University of Minnesota, Minneapolis, Minnesota, U.S.A
    3. Department of Neurology, University of Minnesota, Minneapolis, Minnesota, U.S.A
    • Address correspondence to James Cloyd, Center for Orphan Drug Research, University of Minnesota, 2001 6th St SE, Minneapolis, MN 55455, U.S.A. E-mail:

    Search for more papers by this author



Although topiramate is widely prescribed for epilepsy and migraine, there is no intravenous product. We have developed an injectable topiramate formulation in which the drug is solubilized in a cyclodextrin matrix, Captisol® (Ligand Pharmaceuticals, Inc., La Jolla, CA). Our long-term goal is to evaluate intravenous topiramate for the treatment of neonatal seizures. Prior to studies in newborns, we carried out an investigation of injectable topiramate's safety and pharmacokinetics in adult patients.


Twenty adult volunteers with epilepsy or migraine on stable, on maintenance topiramate therapy were given 25 mg of a stable-labeled intravenous topiramate over 10 min, followed by their usual oral doses. Vital signs were taken, electrocardiography studies (ECGs) were recorded, and the infusion sites were periodically examined prior to and up to 24 h after dosing. Blood samples were collected prior to administration and serially for 96 h thereafter. Plasma concentrations of both stable-labeled and regular topiramate were measured using liquid chromatography-mass spectrometry (LC-MS). Concentration-time data were analyzed using a noncompartmental approach with WinNonlin 5.2 (Pharsight Corporation, Mountain View, CA, U.S.A.).

Key Findings

Seven patients experienced one or more of the following minor adverse events including nausea and vomiting (1), tingling around the lips (1), paresthesia in the arms and legs (1), and a mild vasovagal response with intravenous catheter placement (1). Included in the adverse events were four patients with epilepsy who had seizures consistent with their histories. There were no changes in heart rate, blood pressure, or ECG results, and there were no infusion site reactions. Pharmacokinetic parameters (mean ± standard deviation [SD]) determined following the intravenous dose included absolute bioavailability: 110 ± 16%, distribution volume: 0.79 ± 0.22 L/kg, clearance: 2.03 ± 1.07 L/h, and elimination half-life: 27.6 ± 9.7 h. Distribution volume, half-life, and clearance were significantly altered by enzyme-inducing drugs.


A single 25-mg dose of intravenous topiramate caused minimal infusion site or systemic adverse effects in patients taking oral topiramate. Pharmacokinetic results show that oral topiramate is completely absorbed and that its steady-state elimination half-life is longer than previously assumed, which permits once or twice daily dosing even in the presence of enzyme-inducing drugs. The information from this study can inform the design of subsequent studies in adults, older children, and newborns, including controlled clinical trials intended to determine the efficacy and safety of intravenous topiramate for neonatal seizures.

Topiramate (TPM) is used in adults and children, age 2 and older, for the treatment of epilepsy and, in adults, for the prevention of migraine headaches. It possesses a broad range of pharmacologic actions: blockade of voltage-gated sodium channels, augmentation of certain subtypes of γ-aminobutyric acid (GABA)A receptors, inhibition of α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) and kainate glutamate ionotropic receptors, and weak inhibition of some isoenzymes of carbonic anhydrase (CA-II and CA-IV). There is a growing body of evidence that TPM is effective in controlling seizures and is neuroprotective in newborn laboratory animal models of status epilepticus and cerebral ischemia (Yang et al., 1998; Niebauer & Gruenthal, 1999; Lee et al., 2000; Cha et al., 2002; Follett et al., 2004; Koh et al., 2004).

Neonatal seizures are a relatively rare, but serious medical problem. The most common cause is hypoxic-ischemic brain injury, which has a high mortality rate and grave neurologic sequelae (Ronen et al., 1999). The occurrence of seizures in conjunction with hypoxic-ischemic injury is thought to exacerbate subsequent neurologic problems (Holden et al., 1982; Sheth et al., 1999; Tekgul et al., 2006). In an open label trial, the current drugs of choice, phenobarbital and phenytoin, for treating neonatal seizures were effective in less than 50% of newborns (Painter et al., 1999). In addition to their uncertain efficacy, both drugs are associated with serious adverse effects (Painter et al., 1999; Bittigau et al., 2003; Glier et al., 2004).

Based on promising results from laboratory studies, orally administered TPM has been used to treat neonatal seizure when first-line therapy fails (Silverstein & Ferriero, 2008; Glass et al., 2011). Silverstein & Ferriero (2008) reported the results of a survey of 55 child neurologists regarding their choice of a second-line, add-on antiepileptic drug in the treatment of neonatal seizures. The leading add-on candidates were topiramate (recommended by 55%) and levetiracetam (recommended by 47%). Among those recommending topiramate, 70% perceived treatment to be beneficial and 63% perceived no side effects. In addition to use as an antiepileptic drug, oral topiramate has recently been studied for use in hypothermic newborns with hypoxic-ischemic encephalopathy, but thus far studies have focused on safety and pharmacokinetics (Filippi et al., 2009, 2010). When TPM is used to treat neonatal seizures an extemporaneously compounded oral solution of unknown stability is typically administered through a gastrointestinal tube.

An injectable TPM formulation could improve the treatment of neonatal seizures by ensuring more precise, reliable dosing (Glass et al., 2011). Furthermore, injectable TPM would be useful as bridge therapy in older children and adults on oral topiramate when they are unable to take medications by mouth. Although TPM solubility in water (maximum = 9.8 mg/ml) would permit manufacturing of an intravenous solution suitable for adults and older children, a more concentrated formulation is desirable for newborns so as to limit the fluid volume administered with a dose (Janssen Pharmaceuticals, Inc, 2012).

New drugs and formulations typically must first be evaluated in adults and older children before beginning studies in neonates, Therefore, this report presents the results of a first-in-human study of intravenous topiramate (IV TPM). The aims were to determine the safety and pharmacokinetics of IV TPM in adult patients on maintenance TPM therapy. To accomplish the latter aim, we used a stable-labeled TPM formulation, which permitted rigorous characterization of oral and IV TPM pharmacokinetics, including absolute bioavailability, distribution volume, clearance, and elimination half-life under steady-state conditions without interrupting maintenance therapy (Baillie, 1981).



Twenty patients 18 years of age and older were enrolled. They were taking oral TPM for treatment of migraines or epilepsy. Patients were excluded if they were pregnant or breastfeeding. They could be on monotherapy or taking other medications along with TPM, but had to be on a stable regimen of all medications for at least 2 weeks prior to the first day of the study.

Patients were recruited from clinics at Fairview University Medical Center, MINCEP Epilepsy Care, and other neurology clinics in the Minneapolis-St. Paul metropolitan area.

Study drug

Stable-labeled TPM was formulated as a 10 mg/ml solution dissolved in a 10% sulfobutyl cyclodextrin (Captisol, Ligand, La Jolla, CA, U.S.A.). This formulation is sufficiently concentrated to limit the injection volume but does not require amounts of Captisol® larger than those in currently approved products. The formulation was manufactured at the University of Iowa, Division of Pharmaceutical Sciences under good manufacturing practices.

Study design

This protocol was approved by the University of Minnesota Institutional Review Board. It was conducted under IND #78993. All subjects provided written informed consent. They were admitted to the University of Minnesota General Clinical Research Center on the morning of the study. A brief physical and neurologic examination including electrocardiography (ECG) was performed, and catheters were inserted in each arm; one was used for blood sampling and the other for delivery of the stable-labeled TPM. Routine blood tests included measurements of sodium, potassium, glucose, serum creatinine, chloride, urea nitrogen, hemoglobin, albumin, carbon dioxide, alanine aminotransferase (ALT), aspartate transaminase (AST), alkaline phosphatase, bilirubin, and total plasma protein.

A 25 mg (2.5 ml) intravenous dose was selected so as to limit the increase in TPM exposure, since it was added to patients' oral TPM regimens. Drug was administered intravenously over 10 min with a syringe pump. The volume of intravenous injection was measured and the syringe was weighed before and after administration as verification of the dose administered. One hour after the end of the infusion, each patient took his/her usual morning TPM dose.

Subjects remained at the clinical research unit for 12 h following dosing. They returned 24, 48, 72, and 96 h after dosing for additional blood draws and safety assessments including blood pressure, pulse, and self-reported adverse events. Central nervous system toxicity and ECG were assessed by a neurologist prior to, during, and 15 min after the infusion. Blood pressure and pulse were monitored every 2 min during infusion, every 15 min for 1 h after infusion, and then every 8 h until discharge. A research nurse examined the infusion site for irritation and extravasation and asked the patient about infusion site discomfort.

Blood sampling

Ten-milliliter blood samples were collected for determination of plasma TPM concentrations at predose, 5, 15, and 30 min, and 1, 2, 4, 6, 10, 12, 24, 48, 72, and 96 h after administration of the intravenous dose. Samples were immediately centrifuged and plasma was frozen until analysis.

Analytic methods

The analytical method was designed to measure TPM and stable-labeled TPM in plasma using Hewlett-Packard 1100 series (Agilent Technologies, Santa Clara, CA, U.S.A.) with a degasser (Model G1322A), quaternary pump (Model G1311A), autosampler (Model G1313A), and a quadrupole mass spectrometer (model G1946A). CHEMSTATION Software (Agilent Technologies) was used to operate the modules and for data analysis. The analytes were separated using a Zorbax Eclipse XDB C18 column (150 × 3.0 mm, 3.5 μm particle size) (Agilent Technologies) and eluted isocratically with a mobile phase consisting of 20 mm Ammonium Acetate (Fisher Scientific, Pittsburgh, PA, U.S.A.) buffer and methanol (Fisher) (50:50, v/v). The flow rate was 0.4 ml/min, and the run time was 10 min. The analytes were subsequently detected in the electrospray ionization-mass spectrometry system; the mass spectrometer was run in the positive ion mode. Nitrogen was used as the drying gas and was supplied at a flow of 12 L/min at a temperature of 300°C. The capillary voltage was set at 4,000 V. The high-performance liquid chromatography (HPLC) eluent was introduced in the source via the electrospray interface, generating the negatively charged molecular ion [M-H]−. The samples were analyzed using Selected Ion Monitoring mode (SIM) and quantitation ions were m/z 338, m/z 344, and m/z 350 for TPM, TPMSI, and TPMd12 internal standard, respectively. The samples were run with a seven-concentration standard curve (run in triplicate) and nine quality control samples (low, med, and high run in triplicates). The calibration curve was linear in the concentration range of 0.1–20 μg/ml. The limit of detection was 0.5 ng/ml; the limit of quantification was 0.04 μg/ml. Assay precision for TPM and stable-labeled TPM ranged from 2–5% and 3–5%, respectively, with accuracy values of TPM recovery for TPM in spike plasma ranged from 94–105% and 90–106% for stable-labeled TPM.

Pharmacokinetic analysis

Topiramate concentration-time data were analyzed using a noncompartmental pharmacokinetic approach with WinNonlin software (version 5.2; Pharsight Corporation, Mountain View, CA, U.S.A.). The elimination rate constant (λz) was determined from the slope of the terminal log-linear portion of the plasma concentration–time curve, and the terminal half-life (t1/2) was calculated as ln 2/(λz). Maximum plasma concentrations (Cmax) and time to maximum concentration (tmax) were obtained by direct observation of the data. The area under the concentration–time curve (AUC) to the last nonzero plasma concentration (Clast) above the lower limit of quantification was calculated as AUClast. The area under the concentration–time curve extrapolated to infinity (AUC0–∞) was calculated as AUClast + (Clast/λz). Clearance was obtained by dose divided by AUC0–α. Volume of distribution was calculated by dose divided by k*AUC0–α. Absolute bioavailability (F) was determined by calculating the ratio of the dose normalized oral steady-state oral area under the concentration time curves for the dosing interval (AUCss0–tau) to the dose normalized intravenous area under the concentration time curves (AUC0–α).

display math

Bioavailability was computed only for patients taking TPM once daily or if they were on equal oral TPM doses taken either every 8 or 12 h. Five patients were excluded from bioavailability analysis because the failed to meet one of these criteria.

Statistical analysis

Means and standard deviations for the parameters were calculated obtained using the descriptive statistics tool in WinNonlin version 5.2

The effects of sex, age, comedications, weight, creatinine clearance, and indication (epilepsy or migraines) on elimination half-life, clearance, volume of distribution, and bioavailability was examined using a univariate analysis of variance. A p < 0.05 was considered significant.


All 20 subjects (13 female, 7 male, all Caucasian) enrolled completed the study. There mean age and weight were: 39.9 years (range 26–74 years), and 92.3 kg (range 54.5–150.3 kg). Twelve patients were being treated for epilepsy among whom eight were taking an enzyme-inducing antiepileptic drug: 3 on phenytoin, 4 on carbamazepine, and one on oxcarbazepine. Eight were taking TPM for migraine headaches. The mean TPM dose was 178 mg/day (range 50–600). The mean creatinine clearance was 107.2 ml/min (range 66–158).

Safety results

No serious adverse events occurred. There were no changes in heart rate, blood pressure, ECG, or infusion site reactions. Seven patients (35%) experienced one or more adverse event during the study: one each had nausea and vomiting, tingling around the lips, paresthesia in the arms and legs, and mild vasovagal response with intravenous catheter placement. Four patients with intractable epilepsy experienced a typical seizure during the study and the site recorded these as adverse events. All adverse events fully resolved quickly.

Pharmacokinetics results

The concentration-time profile (0–96 h) following the intravenous dose is shown in Fig. 1. Table 1 presents the pharmacokinetic parameters calculated from the intravenous and oral TPM concentration-time data. The mean (±standard deviation) absolute bioavailability for orally administered TPM was 110 ± 16%, with a range of 77–143% (Fig. 2). The mean clearance was 2.03 ± 1.07 L/h, but ranged fivefold from 0.84 to 4.23 L/h. Subjects with the greatest clearances were on at least one enzyme-inducing antiepileptic drug (carbamazepine, oxcarbazepine, or phenytoin). The half-life ranged from 15.5 to 53.5 h with a mean half-life of 27.6 ± 9.7 h (mean ± standard deviation). Half-life was also influenced by inducing comedications.

Table 1. Mean topiramate pharmacokinetic parameters
ParameterAll patientsInducersNoninducersp-valuea
  1. CL, clearance; Vd, volume of distribution; T1/2, half-life; Cmax IV, maximum plasma concentration following the IV dose; AUC0–α IV, area under the plasma concentration time curve from zero to infinity.

  2. a

    Comparisons between inducers and noninducers.

Cl (L/h)2.03 ± 1.073.31 ± 0.641.35 ± 0.37<0.001
Vd (L/kg)0.79 ± 0.221.01 ± 0.200.67 ± 0.12<0.001
T1/2 (h)27.6 ± 9.721.2 ± 7.131.1 ± 9.20.023
Cmax IV (μg/ml)0.56 ± 0.150.47 ± 0.080.61 ± 0.160.052
AUC0–∞ IV (h*μg/ml)16.0 ± 7.07.8 ± 1.619.8 ± 5.3<0.001
Figure 1.

Mean topiramate concentration-time profile after 25 mg intravenous. Error bars signify one standard deviation of the mean.

Figure 2.

Absolute topiramate bioavailability by patient.

Once enzyme-inducing comedication was incorporated into the model, no other factor, that is, age, creatinine clearance, weight, sex, or indication had an effect on TPM pharmacokinetics.


An intravenous TPM formulation would be useful in the treatment of neonatal seizures as well as in providing bridge therapy for older children and adults in whom oral TPM is interrupted. Before studies in neonates can be done, the pharmacokinetics and safety of TPM typically must be evaluated in adults. This pilot phase I study is a first-in-human investigation of intravenous TPM safety and pharmacokinetics.

Topiramate therapy is associated with significant verbal function, memory, and attention impairment in healthy volunteers and patients with epilepsy (Meador et al., 2003; Blum et al., 2006). We choose to first study intravenous TPM in patients on long-term oral TPM therapy in order to minimize the potential for adverse effects. Furthermore, we used a small dose so as to limit the effects of increased plasma TPM concentrations. Although we observed only minor adverse events; a larger dose or administration to subjects naive to TPM would more likely result in clinically important adverse events.

The use of an injectable, stable-label formulation given simultaneously to patients on maintenance TPM therapy along with a liquid chromatography-mass spectroscopy assay allowed us to measure both labeled (IV) and unlabeled (oral) TPM and, thus characterize aspects of TPM pharmacokinetics not previously known.

The use of an IV formulation along with the oral product, permitted the first ever determination of TPM absolute bioavailability, which is approximately 100% with modest interpatient variability. This suggests that clinicians can be reasonably confident that poor or wide variability in absorption is not a likely cause of poor response. Knowing that TPM is completely absorbed also permits use of the same dose intravenously when oral therapy is interrupted.

For the same reasons that apply to absolute bioavailability, this is the first study reporting the absolute volume of distribution. The distribution volume, 0.8 L/kg, we determined agrees with previous reports, of 0.6–0.8 L/kg. (Garnett, 2000) and exhibits relatively low variability. Information about distribution volume permits calculation of intravenous loading doses to quickly and accurately attain targeted TPM concentrations. However, studies are needed to investigate the safety of using higher loading doses in patients.

Our study is also the first to characterize TPM absolute clearance and elimination half-life under steady-state conditions. TPM elimination involves both metabolism and excretion of the parent drug in the urine. Approximately 20% of TPM is metabolized when administered in the absence of enzyme inducers (Wu et al., 1994). When TPM is administered with an enzyme inducer, the apparent metabolic clearance doubles (Gisclon et al., 1994; Sachdeo et al., 1996; Garnett, 2000). Our study, using an intravenous formulation which avoids intestinal metabolism, confirms that the effect of enzyme inducers is primarily in the liver and not in intestinal tissue. We found a fivefold range in clearances, from 0.84 to 4.23 L/h, which likely contributes to the substantial variability in dosing requirements.

The use of a stable-labeled intravenous formulation also allowed us to characterize TPM elimination half-life under steady-state conditions. Previous studies in either healthy volunteers or patients given a single or multiple TPM doses reported an average half-life of 21 h (Easterling et al., 1988; Doose et al., 1996; Johannessen, 1997). Information about the steady-state half-life of TPM in the presence of enzyme-inducing drugs had not been reported. We found half-life in induced patients was similar to that in earlier reports, but TPM half-life in uninduced patients was substantially longer, 31 h, than values obtained following single doses. Our results indicate, that regardless of concomitant therapy many patients may be able to take immediate release oral TPM on a once daily basis.

Impaired renal function and advanced age decrease TPM clearance (Doose & Streeter, 2002). However, we found no effect of age or renal function on TPM pharmacokinetics, which was expected given that our patients had were relatively young and had creatinine clearances in the normal range.

We used a small intravenous dose for this first-in-human study to obtain preliminary safety and pharmacokinetic data prior to initiating the next phase of the project, a safety and bioequivalence study comparing a clinically relevant intravenous TPM dose with the commercially available oral tablet, which has now been completed (Clark et al., 2013). Results from these pilot phase I studies in adults support further evaluation of intravenous TPM in adults, children, and newborns.


The assistance of Usha Mishra, MS in performing the analysis of TPM is gratefully acknowledged. This study was supported by an FDA Office of Orphan Products Development Research Grant (R01 FD003540-01).


R. C. Brundage owns stock in Glaxo Smith Kline. Dr. Cloyd is entitled to royalties under a licensing agreement pertaining to injectable topiramate between Ligand Pharmaceuticals, Inc, and the University of Minnesota. This relationship has been reviewed and managed by the University of Minnesota in accordance with its conflict of interest policies. Dr. Cloyd is also a consultant to Upsher Smith Laboratories regarding the development of an extended release topiramate product. The remaining authors have no conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.