Antiepileptic Drugs, 5th Edition



Clinical Efficacy and Use in Nonepileptic Disorders

Kekneth W. Sommerville MD

Medical Director, Neuroscience, Department of Marketed Products, Clinical Research, Abbott Laborarories, Abbott Park, Illinois

Tiagabine, a derivative of nipecotic acid, is an inhibitor of γ-aminobutyric acid (GABA) uptake that was developed specifically for use as an antiepileptic drug (AED). The drug discovery program for tiagabine was established to identify inhibitors of GABA uptake that were sufficiently lipophilic to cross the blood-brain barrier and to produce anticonvulsive effects in the central nervous system (CNS).


Because GABA is the major inhibitory neurotransmitter in the CNS, GABAergic mechanisms have been extensively investigated in research on seizure suppression and epilepsy therapy (1,2). A brief description of GABA-related pharmacology can provide a basis for understanding the role of GABAergic drugs in the treatment of epilepsy and other neuropsychological disorders.

The two principal postsynaptic GABA receptors are the GABAA and GABAB receptor complexes. The GABAB receptor is G-protein coupled; its activation by GABA causes hyperpolarization and resultant inhibition of neurotransmitter release. The GABAA receptor complex is a pentameric heterooligomer that exists in multiple subtypes in vivo. In addition to binding sites for GABA, the GABAA receptor has binding sites for benzodiazepines, barbiturates, and neurosteroids, and it is coupled to a chloride ion channel. Activation of the GABAA receptor induces increased inward chloride ion flux and results in membrane hyperpolarization and neuronal inhibition. Biochemical and clinical evidence indicates that the initiation and spread of seizure activity may involve alterations in the activity of the GABAA receptor (3). Compounds that directly stimulate the GABAA receptor complex tend to have anticonvulsant activity, whereas GABAA receptor antagonists (including bicuculline and pentylenetetrazol) produce convulsions in animals (4).

Once released into the synapse, free GABA that does not bind rapidly to the GABAA or GABAB receptor complexes may diffuse from the synapse or may be actively taken up by neurons and glial cells. Four different membrane transporter proteins that mediate the uptake of synaptic GABA into neurons and glial cells have been identified, differing in their CNS distribution and localization; they are known as GAT-1, GAT-2, GAT-3, and BGT-1 (5). Tiagabine interacts primarily with the GAT-1 transporter.


Because a reduction in GABAergic neuronal activity has been implicated in various neuropsychological disorders, including epilepsy, anxiety, and pain, the enhancement of such activity has been the focus of numerous pharmacologic research initiatives (1,6,7). These have included investigations into the direct stimulation of the GABAA receptor, the modulation of GABAA receptor activity (by compounds such as the benzodiazepines), and the inhibition of GABA metabolism (by compounds such as vigabatrin). For the most part, these areas have yielded compounds of only limited utility in the treatment of epilepsy and other disorders, because of side effects caused by generalized GABA receptor stimulation, rapid development of anticonvulsant tolerance (in the case of the benzodiazepines), or other unacceptable effects.

The enhancement of GABAA receptor activity through the inhibition of GABA uptake represents a promising alternative approach for two principal reasons. First, the inhibition of GABA uptake preferentially accentuates the patterns of endogenous GABA release in the brain, instead of inducing generalized, possibly nonphysiologic, receptor stimulation. Second, the extent to which GABA uptake inhibitors stimulate GABAA receptors is ultimately limited by the amount of synaptic GABA available, which is presumably under tight physiologic control (8).

Tiagabine was synthesized after the identification of several cyclic amino acids (including nipecotic acid), that


inhibit GABA uptake in certain tissues in vitro. Tiagabine is a derivative of nipecotic acid with a lipophilic “anchor” covalently attached to the amino nitrogen (Figure 75.1). Tiagabine strongly inhibits GABA uptake in synaptosomal preparations from rat brain and in cultured neurons and glial cells in vitro (9). It also generates a dose-dependent increase in extracellular GABA levels in rat brain in vivo, as well as inhibiting some chemically induced seizures (10). Tiagabine appears to interact preferentially with the GAT-1 GABA transporter, and this may limit its activity to regions of the CNS in which GAT-1 plays a significant role (the cortex, cerebellum, and hippocampus) (11,12). In clinical trials, tiagabine has been shown to be effective in the treatment of a range of partial seizure types, both as adjunctive therapy (13) and as monotherapy (14, 15, 16). A schematic diagram of the synaptic action of tiagabine is shown in Figure 75.2.


FIGURE 75.1. Chemical structure of tiagabine.


FIGURE 75.2. Schematic diagram of the action of tiagabine at the synapse. GAD, glutamic acid decarboxylase; SSA, succinic semialdehyde. (From Suzdak PD, Jansen JA. A review of the preclinical pharmacology of tiagabine: a potent and selective anticonvulsant GABA uptake inhibitor. Epilepsia 1995;36:612-626, with permission.)


The possibility that modulation of GABAergic responses may be useful in therapeutic approaches to conditions other than epilepsy led to the suggestion that tiagabine be considered for clinical evaluation of efficacy in these conditions. Meldrum and Chapman summarized the case for exploring alternate uses for tiagabine (12). The specificity of


tiagabine's antiepileptic activity seems to be related to the importance of GAT-1 as its principal GABA transporter. Tiagabine specifically suppresses various chemically induced and kindled seizures in rats, it is also effective against certain genetic models of reflex epilepsy in other species, and it has some activity against maximal electroshock seizures. Other GABA uptake inhibitors working through transporters different from GAT-1 show a different spectrum of anticonvulsant activity. This specificity may determine how useful tiagabine may be in other conditions possibly responsive to GABAergic treatment (12).

Therapeutic areas for continuing tiagabine research include the following:

  • Expansion of epilepsy-related indications (e.g., infantile spasms) (12).
  • Suppression of status epilepticus and epileptogenesis (12).
  • Exploration of disorders known or suspected to be related to GABAergic mechanisms: sleep disorders (12), pain (including postherpetic neuralgia and diabetic neuropathy) (17, 18,19), movement disorders (particularly those related to basal ganglia disease, e.g., tardive dyskinesia) (20,21), spasticity (22), bipolar disorder (23), and anxiety (24).
  • Neuroprotection against ischemia-induced cell loss (25).



The most extensive area of research with tiagabine outside of epilepsy is neuropathic pain. Tiagabine was evaluated in several standard models of pain in rats and mice (17). Tiagabine demonstrated antinociceptive effects in moderate doses (7.2 and 24.3 µmol/kg) against thermal pain in a mouse hot-plate test, but it was ineffective in rat tail-flick tests at high doses (72.8 µmol/kg). Because the hot-plate test is thought to involve supraspinal processing and the tail-flick test is considered a spinal reflex, these findings suggest that tiagabine selectively modulates supraspinal GABAergic responses. Tiagabine was also effective at moderate doses in reducing the acetic acid-induced abdominal stretching response in mice and the first phase of formalininduced paw flinches in rats. In the second phase of the formalin paw-flinch test, which involves an active inflammatory component, high doses of tiagabine were required to produce a response, a finding suggesting that tiagabine does not possess significant antiinflammatory activity. Tiagabine was also moderately effective (compared with morphine) in reducing the allodynic response to von Frey filaments in a spinal nerve ligation model of rat neuropathic pain. The specificity of the effects of tiagabine in moderating pain perception suggest that only certain pain pathways may be strongly dependent on GAT-1-mediated GABAergic mechanisms (17).

The effectiveness of tiagabine in the management of diabetic neuropathy was evaluated in a combination open-label and double-blind crossover trial using a once-daily, extended-release formulation. Thirty-five adult patients (≥18 years of age) with painful diabetic polyneuropathy were titrated with tiagabine in the initial open-label phase to determine their maximum tolerated dose and initial pain relief response. The rate of discontinuation for adverse events was higher than expected from earlier evaluations of tiagabine (primarily CNS events thought to be related to too-rapid titration), so only 17 patients completed the open-label phase, and 11 were randomized to the doubleblind portion of the study. The high rate of drug discontinuation may also have been related to the lack of extended-release dosage forms other than the 12-mg strength. All adverse events were transient and resolved with tiagabine discontinuation or dose reduction. Pain assessments were based on the Brief Pain Inventory (BPI) and McGill Pain Questionnaire (MPQ). Patients who tolerated a tiagabine dose of at least 12 mg/day, and who demonstrated pain relief of at least three units on the worst/average pain items of the BPI (Figure 75.3), were randomized, underwent drug washout, and were retitrated to their previous dose in a two-period, double-blind placebo-controlled crossover study. Significant differences (p < .05) were observed in the treatment group with respect to percentage of pain relief in the last 24 hours and the degree to which pain had interfered with sleep in the last 24 hours. Improvements, although not statistically significant, were noted in other preplanned BP-I and MPQ variables, most notably those dealing with the extent to which pain had interfered with social relationships and enjoyment of life on the BPI (Figure 75.3) and the affective subscale of the MPQ. The researchers concluded that the generally favorable results of tiagabine treatment warrant further evaluation in larger-scale studies with greater dosing flexibility (19).

In another small study of neuropathic pain, 17 patients with painful sensory neuropathy received an initial dose of 4 mg tiagabine after a 1-week washout of all previous pain medications. Patients were subsequently titrated to 16 mg/day by increments of 4 mg/wk. Nine patients completed the titration, and eight discontinued for adverse events, mostly nausea, dizziness, or disorientation. In patients who completed the titration, tiagabine reduced symptoms of surface pain (34.2%), skin sensitivity (28%), burning (36%), cold (35.5%), pain sharpness (29%), and discomfort (18.4%). There was no effect on dull or deep pain, or on autonomic tests, although two patients reported less perspiring and four noted improved sleep patterns and skin temperature. The best results were found at doses of 8 to 12 mg/day. The investigators suggested that tiagabine may improve small-fiber neuropathy and recommended further study (18).




FIGURE 75.3. Changes in brief pain inventory, mean group scores, double-blind period. [From Kirby LC, Collins SD, Deaton RL, et al. Tiagabine (Gabitril) in the management of painful diabetic polyneuropathy. Presented at the poster session of the American Pain Society meeting, Fort Lauderdale, FL, 1999.]


The use of tiagabine in spasticity was evaluated in a pilot study of 14 children <12 years of age who suffered from both static spastic quadriplegia and epilepsy intractable to treatment by other AEDs. The number of AEDs previously used unsuccessfully for seizures ranged from one to eight, with a mean of 4.5. The children had various seizure types, and all had two or more different types of seizures. Tiagabine was titrated from a starting dose of 0.1 to 0.2 mg/kg/day to seizure abatement, onset of adverse events, or the maximum dose of 1.1 mg/kg/day. Spasticity was assessed by motor function improvements (incorporating gross, fine, and oral motor control) scored by a participating pediatric neurologist on a scale of 0 to 4 (0, no improvement; +1, 25% improvement; +2, 50% improvement; +3, 75% improvement; and +4, 100% improvement). Seizure control was also scored on a scale of 0 to 4 incorporating ranges (0, 0% to 24% reduction; +1, 25% to 49% reduction, etc.). All children demonstrated at least some improvement in motor function (range, +1 to +4), with a mean improvement of +2.3 (~50%). Seizure control improvement scores ranged from 0 to +4, with a mean improvement of +2.4 (reduction of 50% to 74%). The authors noted that the encouraging results in both motor control and seizure control for spasticity and epilepsy of varying origin suggest that longerterm, randomized trials of tiagabine efficacy in spasticity should be undertaken (22).


The demonstrated efficacy of divalproex sodium in migraine prophylaxis has led to interest in assessing the ability of other AEDs, including tiagabine, to reduce migraine frequency and severity. In an open-label trial of tiagabine use in 41 patients who had experienced adverse effects (n = 22) or incomplete relief from migraine (n = 14) using divalproex sodium, tiagabine was used at moderate doses (8 to 16 mg/day) over a minimum 3-month period. Because no lead-in baseline phase was monitored, no estimate of reduction in migraine frequency could be reported. However, improvement (measured using global evaluations by both patient and physician) was noted in 28 of 41 patients. They reported improvements of >50%. Twelve patients experienced 14 adverse events, resulting in discontinuation of tiagabine therapy in nine patients. Adverse events included fatigue (n = 9), weight gain (n = 2), confusion (n = 2), and poor memory (n = 1) (26).

Another open-label trial evaluated tiagabine use in 49 patients who were unresponsive to current migraine prophylactic and abortive therapy. Tiagabine dosage was titrated to 16 mg/day. This study allowed the continued use of comedications and abortive treatment. After 6 weeks of treatment, 23% of the patients were headache free; after 12 weeks, 32% were headache free; an additional 34% reported improvement. Five patients discontinued tiagabine because of side effects (sedation, dysphoria, agitation, or tremor). Investigators observed that patients who


responded to tiagabine therapy were significantly more likely to have comorbid epilepsy, affective disorder, posttraumatic stress disorder, or fibromyalgia than nonresponders (27).

A case-study report found that tiagabine was effective in four patients with migraine and other types of headache who were refractory to multiple medications, including other AEDs. In two patients with refractory cervicogenic headache, daily headache frequency declined 80% to 90% with relatively high (36 and 80 mg/day) tiagabine doses (28). It was not clear from the report whether these patients were taking concomitant enzyme-inducing drugs, which could account for the high doses.

The apparent efficacy of tiagabine in migraine prophylaxis in small uncontrolled trials led all researchers involved to suggest that larger-scale, well-controlled trials are warranted. The GABAergic mechanism of tiagabine and the frequent comorbidity of migraine and epilepsy suggest that tiagabine has potential in this therapeutic area.


Tiagabine has also been evaluated in psychiatric disorders, to date, exclusively in a small-scale trial and case studies. A case-study evaluation of three patients reported benefit from tiagabine low-dose (8 mg/day) add-on therapy in all cases. The patient with the most severe case, diagnosed as having schizophrenic disorder (bipolar type), had an extensive history of psychotic episodes and multiple courses of medication, often with significant side effects. Tiagabine was used to replace lamotrigine as an add-on medication to paroxetine and olanzapine and successfully controlled paranoid features that appeared in the absence of lamotrigine. In the other two cases, complete remission of the symptoms of bipolar disorder (severe mania in the first case and a mixed bipolar state with psychotic depressive features in the second) was reported with tiagabine adjunctive to a regimen of valproate, bupropion, and methylphenidate (the first case), and carbamazepine and paroxetine (the second case) (23).

In two patients with refractory bipolar disorder, low-dose tiagabine was evaluated as adjunctive treatment to regimens of lamotrigine and alprazolam (in one patient) and venlafaxine, lithium, and flurazepam (in the other). Both patients showed remission of bipolar symptoms, with improvement in mood. One patient began to experience a manic episode while taking 3 mg/day of tiagabine; he condition apparently was restabilized by an increase to 4 mg/day (29).

A small-scale trial in Europe evaluated the use of tiagabine in bipolar disorder, with relatively rapid titration in eight acutely manic patients. Tiagabine was started at 20 mg/day (described as a loading dose) on day 1 and was further increased in steps of 5 mg/day when tolerated by the patient, up to a maximum of 40 mg/day in two patients. Tiagabine therapy was adjunctive in all but one patient. There was a slight improvement in three patients, all with moderate manic syndrome, but none of the severely manic patients showed a clear benefit. Side effects were significant in several patients, including a generalized tonic-clonic seizure in one patient that may have been attributable to tiagabine administration. The authors stressed the small scale of the study and compared their results with those reported in the three-patient report cited earlier. The high initial dose and rapid titration may have been a factor in the results and could have contributed to the seizure and other adverse events. These investigators suggest that tiagabine may not be useful as an acute therapy for manic episodes, but it deserves further study in general psychiatric practice (30). The high doses and rapid titration in this study contrast with other, more favorable reports in which drug administration was more cautious. If tiagabine is useful in patients with mania, it would need to be tested at lower doses than in this report and with slower titration.

Tardive dyskinesia is a hyperkinetic motor disorder associated primarily with long-term administration of neuroleptic medications. A double-blind, placebo-controlled trial of γ-vinyl γ-aminobutyric acid (vigabatrin), another GABAergic drug, was carried out in seven schizophrenic patients with tardive dyskinesia. A significant decrease in dyskinetic symptoms (involuntary movements), as measured by Smith Extrapyramidal Scale ratings, occurred with the administration of vigabatrin. This was associated with a twofold increase in cerebrospinal fluid levels of GABA. A significant reduction in cerebrospinal fluid levels of GABA was also observed in the dyskinetic patients with schizophrenia compared with nondyskinetic controls. These results support the view that a GABA deficit plays an important role in the pathophysiology of tardive dyskinesia, and GABA agonists and GABA-receptor potentiators are potentially effective in treating this disorder (21). Although vigabatrin has GABAergic action through a different mechanism than that of tiagabine, these promising data led to preclinical work on tardive dyskinesia with tiagabine.

Haloperidol-induced oral dyskinesias (vacuous chewing movements) in Sprague-Dawley rats have been used in several laboratories as an animal model for tardive dyskinesia. In a study of 48 rats, tiagabine, at a dosage of 75/mg/kg/day, significantly inhibited the onset of vacuous chewing movements and decreased the average severity from 11.2±2.0 vacuous chewing movements per 5 minutes in rats receiving haloperidol alone to 4.4±1.4 in rats receiving haloperidol and tiagabine. By comparison, the placebo rate was 1.3±0.5. The investigators concluded that these data suggest the potential usefulness of tiagabine as a therapeutic agent in the treatment or prophylaxis of tardive dyskinesia (20). Clinical trials of tiagabine in tardive dyskinesia have been initiated, but, at the time of publication, results have not been reported.




The relatively specific range of activity of tiagabine in the laboratory and clinic and its ability to be well tolerated at efficacious dose ranges are presumably related to the targeted effect on the uptake of GABA. The important role played by GABAergic mechanisms, in normal and abnormal CNS function, suggests that tiagabine may be of clinical benefit to many patients who are refractory to current therapies for many different disorders.

Both the development of tiagabine and the regulatory approval for tiagabine use in treating epilepsy occurred only recently; therefore, reports on its efficacy in the treatment of other conditions are preliminary. However, the encouraging early results observed in the treatment of pain, spasticity, and migraine and other headaches clearly indicate the need for well-controlled clinical trials. The possibility that tiagabine may be efficacious in the treatment of psychiatric and movement disorders also deserves further study, especially in the promising areas of bipolar disorder and tardive dyskinesia. Further understanding of the pharmacology and specificity of tiagabine, particularly against the background of steadily increasing knowledge about the multiple roles of GABA, should help to guide research into new clinical areas.


  1. Meldrum B. Pharmacology of GABA. Clin Neuropharmacol1982;5:293-316.
  2. Krogsgaard-Larsen P, Falch E, Larsson OM, et al. GABA uptake inhibitors: relevance to antiepileptic drug research. Epilepsy Res1987;1:77-93.
  3. Tunnicliff G, Raess BU. GABA neurotransmitter activity in human epileptogenic brain. In: Tunnicliff G, Raess BU, eds. GABA mechanisms in epilepsy.New York: Wiley-Liss, 1991: 105-120.
  4. Schwartz RD. The GABA-A receptor-gated ion channel: biochemical and pharmacological studies of structure and function. Biochem Pharmacol1988;27:3369-3378.
  5. Borden LA. GABA transporter heterogeneity: pharmacology and cellular localization. Neurochem Int1996;29:335-356.
  6. Krogsgaard-Larsen P. GABA synaptic mechanisms: stereochemical and conformational requirements. Med Res Rev1988;8: 27-56.
  7. Krogsgaard-Larsen P, Scheel-Krüger J, Kofoed H. GABA-neuro-transmitters: pharmacochemical, biochemical and pharmacological aspects.Copenhagen: Munksgaard, 1979.
  8. Suzdak PD, Jansen JA. A review of the preclinical pharmacology of tiagabine: a potent and selective anticonvulsant GABA uptake inhibitor. Epilepsia1995;36:612-626.
  9. Braestrup C, Nielsen EB, Sonnewald U, et al. (R)-N-[4,4-bis(3-methyl-2-thienyl)but-3-en-1-yl]Nipecotic acid binds with high affinity to the brain γ-aminobutyric acid uptake carrier. J Neurochem1990;54:639-647.
  10. Fink-Jensen A, Suzdak PD, Swedberg MDB, et al. The γ-aminobutyric acid (GABA) uptake inhibitor, tiagabine, increases extracellular brain levels of GABA in awake rats. Eur J Pharmacol1992;220:197-201.
  11. Borden LA, Dhar TGM, Smith KE, et al. Tiagabine, SK&F 89976-A, CI-966, and NNC-711 are selective for the cloned GAB transporter GAT-1. Eur J Pharmacol Mol Pharmacol1994; 269:219-224.
  12. Meldrum BS, Chapman AG. Basic mechanisms of gabitril (tiagabine) and future potential developments. Epilepsia1999;40 [Suppl 9]:S52-S56.
  13. Ben-Menachem E. International experience with tiagabine add-on therapy. Epilepsia1995;36[Suppl 6]:S14-S21.
  14. Alarcon G, Binnie CD, Elwes RDC, et al. Monotherapy antiepileptic drug trials in patients undergoing presurgical assessment: methodological problems and possibilities.Seizure1995;4: 293-301.
  15. Schachter SC. Tiagabine monotherapy in the treatment of partial epilepsy. Epilepsia1995;32[Suppl 6]:S2-S6.
  16. Brodie MJ, Bomhof MAM, Kalviainen R, et al. Double-blind comparison of tiagabine and carbamazepine monotherapy in newly diagnosed epilepsy: preliminary results. Paper presented at the 22ndInternational Epilepsy Congress, Dublin, 1997.
  17. Giardina WJ, Decker MW, Porsolt RD, et al. An evaluation of the GABA uptake blocker tiagabine in animal models of neuropathic and nociceptive pain. Drug Dev Res1998;44:106-113.
  18. Kanard R, Mendell J, Kissel J, et al. Treatment of painful sensory neuropathy with tiagabine hydrochloride. Paper presented at the International Symposium on the Autonomic Nervous System, Rio Grande, Puerto Rico, 2000.
  19. Kirby LC, Collins SD, Deaton RL, et al. Tiagabine (Gabitril) in the management of painful diabetic polyneuropathy. Presented at the poster session of the American Pain Society meeting, Fort Lauderdale, FL, 1999.
  20. Gao XM, Kakigi T, Friedman MB, et al. Tiagabine inhibits halieridol-induced oral dyskinesias in rats. J Neurol Transm1994; 95:63-69.
  21. Thaker GK, Tamminga CA, Alphs LD, et al. Brain γ-aminobutyric acid abnormality in tardive dyskinesia. Arch Gen Psychiatry1987;44:522-529.
  22. Holden KR, Titus MO. The effect of tiagabine on spasticity in children with intractable epilepsy: a pilot study. Pediatr Neurol1999;21:728-730(abst).
  23. Kaufman KR. Adjunctive tiagabine treatment of psychiatric disorders: three cases. Ann Clin Psychiatry1998;10:181-184.
  24. Neilson EB. Anxiolytic effect of NO-328, a GABA-uptake inhibitor. Psychopharmacology (Berl)1988;96:S42(abst).
  25. Johansen FF, Diemer NH. Enhancement of GABA neurotransmission after cerebral ischemia in the rat reduces loss of hippocampal CA1 pyramidal cells. Acta Neurol Scand1991;84:1-5.
  26. Freitag FG, Diamond S, Diamond ML, et al. An open use trial of tiagabine in migraine. Headache Q2000;11:133-134.
  27. Drake ME Jr, Kay AM, Knapp MS, et al. An open-label trial of tiagabine for migraine prophylaxis. Headache1999;39:352(abst).
  28. Krusz JC, Belanger J, Scott VB. Tiagabine in the treatment of very refractory migraines and other headaches. Headache1999; 39:363(abst).
  29. Schaffer LC, Schaffer CB. Tiagabine and the treatment of refractory bipolar disorder [Letter]. Am J Psychiatry1999;156: 2014-2015.
  30. Grunze H, Erfurth A, Marcuse A, et al. Tiagabine appears not to be efficacious in the treatment of acute mania. J Clin Psychiatry1999;60:759-762.