Dr. Wilder is Professor Emeritus in the Department of Neurology and Neuroscience at the University of Florida College of Medicine in Gainesville.

Acknowledgments: The author reports no financial, academic, or other support of this work. 



Why and how do antiepileptic drugs (AEDs) work in so many neurological and psychiatric disorders? AEDs are highly effective in various neurological and psychiatric disorders. They are the drugs of choice in several mood disorders (including bipolar disorder), neurological disorders, and neuropathic pain syndromes; they are also choice drugs for the prevention of migraine and cluster headaches and restless legs syndrome. AEDs work by altering neuronal excitability, increasing inhibition, and decreasing excitation. They act on neurotransmitters, primarily those involved in excitation (glutamate) and inhibition (γ-aminobutyric acid) and those that may be associated with mood and pain, such as serotonin, dopamine, and norepinephrine. The AEDs approved for the treatment of epilepsy during and after 1993 have created a profound change in the pharmacotherapy of many neurological and psychiatric syndromes.



Since 1993, eight new antiepileptic drugs (AEDs) have been approved for the treatment of epilepsy. We can now assess the use of these drugs not only in epilepsy, but in many other neurological and psychiatric disorders. AEDs have been used in mood disorders and chronic neuropathic pain syndromes for decades. Valproic acid (VPA) is the drug of choice for the treatment of bipolar disease, and topiramate (TPM) and lamotrigine (LTG) are also effective; LTG is especially efficacious in the depressive side of the disorder without tripping the patient back into mania. Valproate and carbamazepine (CBZ) have been used extensively in the management of aggression and episodic dyscontrol syndrome. CBZ was introduced in the United States for the treatment of trigeminal neuralgia more than 2 decades ago.

The new AEDs (those approved after 1993) have opened a Pandora’s box for the widespread use of certain AED drugs for neurological and psychiatric disorders. At present, 38% of AED prescriptions are written for the management of epilepsies and 62% for nonepilepsy uses. AEDs are the drugs of choice in the treatment of chronic neuropathic pain, trigeminal neuralgia, and prevention of migraine. They are also of value in the management of cluster headaches, familial tremor, restless legs syndrome, anxiety, depression, drug and alcohol withdrawal, posttraumatic mood stabilization, spasticity, tics, dystonia, and social phobias. Gabapentin (GP) is the most commonly prescribed AED, although only 12% of prescriptions are written for the treatment of epilepsy. TPM is gaining acceptance as an effective drug for weight loss, in addition to neurological and psychiatric disorders. Some AEDs are undergoing clinical trials for neuroprotection in acute brain and spinal ischemia and certain progressive neurological diseases. How can we account for the widespread use of these drugs? AEDs modulate brain activity by decreasing excitability, enhancing inhibition, and raising the threshold for neuronal firing. They also alter central nervous system (CNS) neurotransmitters, and two of the drugs are carbonic anhydrase inhibitors. This review includes a discussion of the mechanisms of action of the AEDs and provides general comments on efficacy, adverse effects, drug interactions, and dosing parameters of the newer AEDs.


Antiepileptic Drugs

How They Work: Neuronal Targets and Mechanisms of Action

AEDs decrease excitation and enhance inhibition by various mechanisms. AEDs act on specific neuronal targets (Table 1), and many of them block voltage-sensitive sodium channels. This action decreases the frequency of action potentials, raises the threshold for repetitive action-potential generation, and prevents burst firing of neurons.1 In the treatment of epilepsy, this mechanism is thought to prevent the spread of the electrical discharge from an epileptic seizure focus, thereby preventing the spread and generalization of epileptic discharge into a tonic-clonic or grand mal seizure. In the case of neuropathic pain, prevention of repetitive discharge may prevent the windup phenomenon thought to be characteristic of the development of changes in lamina II of the dorsal horn of the spinal cord, which may be important in the development of neuropathic pain.

Some of the AEDs block different types of voltage-sensitive calcium channels. A partial block of slow-conducting calcium channels (T type) decreases thalamocortical reverberating circuits and thereby controls absence seizures.2 Another target by which AEDs reduce excitation of cortical neurons is by reducing calcium entry by blocking L-type calcium channels.3,4 Excitation can be further reduced by blocking N-type calcium channels, which reduce neurotransmitter release.3

Secondly, AEDs target the γ-aminobutyric acid (GABA) inhibitory neurotransmitter system by augmenting the action of GABA at GABAA receptors. This increases the inward flow of chloride ions (Cl-), resulting in neuronal hyperpolarization.5 Other AEDs increase the synthesis and release of GABA. One of the AEDs specifically blocks the reuptake of GABA,6 while another reverses GABA reuptake when the neuron is depolarized. Two of the AEDs trigger the release of serotonin (5-hydroxytryptamine [5-HT]), which in turn stimulates the release of GABA from neurons.7,8

A third target for decreasing excitation of cortical neurons is inhibitory by blocking excitatory receptors. Several of the AEDs either inhibit N-methyl-D-aspartate (NMDA)-type or non-NMDA-type a-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) or kainic acid receptors.9

A fourth target is a reduction of the concentration of neuronal glutamate. One of the AEDs decreases the amount and release of glutamate, the major excitatory transmitter in the CNS.1,5

TPM and zonisamide are carbonic anhydrase inhibitors. This results in CNS acidosis, with a reduction in excitatory NMDA activity and enhancement of inhibitory GABA activity.

Although AEDs used to be commonly characterized as having one or perhaps several main modes of action,2 it has become increasingly clear that most AEDs have multiple modes of action, many of which may be clinically relevant to various neurological and psychiatric disorders10 (Tables 2 and 3).

It is well known that both barbiturates and benzodiazepines bind to the GABAA receptor at different sites, resulting in the augmentation of inhibitory neurotransmission.1,11 However, barbiturates also block non-NMDA-type glutamate receptors.12 Similarly, phenytoin and CBZ are well known as sodium channel blockers.1 They both have a range of other actions, including L- and N-type calcium-channel blockade, blocking NMDA receptors, and, for CBZ, an indirect GABA-ergic effect via increasing the release of 5-HT.10 Valproate has effects on all of the key antiepileptic mechanisms.

As shown in Table 3, a number of the newer AEDs, namely felbamate, TPM, LTG, and GP, have effects on all of the key antiepileptic mechanisms; the effect of GP in blocking action potentials is not via sodium channel blockade.5 After exposure to GP (intracellular injection or extracellular application), the neuron has a time delay before sodium (Na+) channels are affected and burst firing decreases. This is in contrast to other Na+ channel blockers with which the effect is instantaneous. TPM has a direct effect on Na+ channels and L-type calcium ion (Ca++) channels and also blocks non-NMDA glutamate receptors.13 TPM and GP increase the metabolism of glutamate to GABA by stimulating the enzyme glutamic acid decarboxylase and increasing the intracellular concentration of GABA. TPM is a GABA agonist at the receptor site that is different from the phenobarbital and benzodiazepines sites. Tiagabine (TGB) was designed to inhibit the reuptake of GABA by presynaptic neurons and glial cells, and this is its only known mode of action.14 The mechanism of action of levetiracetam (LVT) is unknown.15 It blocks kindling in the kindling model of epilepsy, although it is not effective in blocking seizures in other models of epilepsy. For example, LVT does not block seizures induced by electrical stimulation or seizures induced by pentylene tetrazole in animal models of epilepsy.

The newer AEDs offer a number of advantages over the older drugs. They are just as efficacious as the older drugs in treating the epilepsies and in the treatment of various other neurological and psychiatric conditions. As a group, they are safer and produce fewer idiosyncratic reactions. GP has been chronically administered to over 5,000,000 patients without a serious life-threatening reaction. TPM, likewise, appears to be very safe, with approximately 1,000,000 chronic patient exposures. Felbamate, the first of the new drugs, is not safe and is treated as a third-line agent for the treatment of epilepsy. The remaining new drugs, with the exception of LTG, appear safe from serious idiosyncratic reactions, although they have not received the same usage as GP and TPM. When initiated at low doses and titrated slowly, LTG rarely produces allergic epidermal reactions. The package insert16 for LTG dosing and titration should be carefully followed. The pharmacokinetic profiles for the newer drugs are superior. Bioavailability is better, protein binding is very low, metabolism is more simple, and many of the new AEDs are predominately excreted by the kidneys. The newer drugs produce far fewer drug interactions than the older drugs. Phenytoin (PHT), CBZ, and phenobarbital (PB) are potent enzyme inducers that interact not only with other AEDs but also with many non-AEDs. VPA is a potent enzyme inhibitor for many AEDs (particularly LTG and CBZ epoxide) and non-AEDs. GP and LVT do not interact with other AEDs or non-AEDs. TPM, LT, zonisamide, oxcarbazepine (OX), and TGB are relatively free from producing drug interactions, although they are acted on by the enzyme-inducing drugs. The adverse-effect profiles of the newer AEDs are generally mild and depend on dosing and titration. The general rule of “start low and go slow” applies to AEDs in general. Doses required to treat many neurological and psychiatric conditions are far less than the doses recommended to treat refractory epilepsy.

Neuropathic Pain

Unlike nociceptive pain, which emanates from the activation of the pain receptors, neuropathic pain is the outcome of injury to the pain-conducting nervous system; traditional therapies provide little relief. AEDs have become the drugs of choice for the management of neuropathic pain syndromes. The older anticonvulsants have been used for the treatment of various neuropathic pain and mood syndromes for 40 years.17-20 When appropriate, CBZ (the drug of choice for lancinating pain associated with trigeminal neuralgia) and occasionally phenytoin have been used, although there are only a few randomized clinical studies to support their use for this purpose.21,22 Clinical efficacy for these agents has also been reported for diabetic neuropathy and poststroke pain. However, the general lack of efficacy, along with significant side-effect profiles, has limited the overall use of these drugs.18 GP and TPX have emerged as effective in many chronic pain syndromes, and OX has shown efficacy in trigeminal neuralgia. Controlled studies for other neuropathic pain syndromes are underway.


Approved in 1993 as add-on therapy for the treatment of generalized and partial seizures for adults, GP was synthesized as a structural GABA analog that would cross the blood-brain barrier. GP blocks the tonic phase of nociception and exerts a powerful inhibitory effect in neuropathic pain models of mechanical and thermal hyperalgesia and allodynia. A number of open-label studies have shown positive results in the effect of GP on various pain conditions, including tonic spasms associated with multiple sclerosis, trigeminal neuralgia, neuropathic cancer pain, reflex sympathetic dystrophy, and human immunodeficiency syndrome-related peripheral neuropathy.23

Two large multicenter, randomized, placebo-controlled trials have demonstrated the efficacy of GP in relieving neuropathic pain and associated symptoms (eg, sleep, mood, and quality of life) in patients with postdiabetic neuropathy (PDN) and postherpetic neuropathy (PHN). Dosages ranged from 900–3,600 mg/day in three divided doses. Pain relief was observed during the second week when the dosage reached 1,800 mg/day. The median effective dosage in the two studies was 900–1,200 mg. The drug was well tolerated with no significant differences in adverse effects between the drug and placebo in patients with PDN, with a slightly higher withdrawal rate (13.3% versus 9.5%) in patients with PHN. The most frequent tolerable adverse events were somnolence and dizziness.24,25

Increasing evidence from experimental and clinical studies has shown that GP is efficacious in the treatment of neuropathic pain, particularly pain due to PDN and PHN. Clinical trials are needed to evaluate the efficacy of the agent in other painful conditions such as facial neuralgias (including trigeminal and glossopharyngeal neuralgias), central pain syndromes secondary to cerebrovascular disease, phantom-limb pain, spinal cord injury, and central poststroke pain syndrome.23,26


TPM, a broad-spectrum anticonvulsant, was approved in 1997 for the adjunctive treatment of partial and secondary generalized seizures in adults. Unlike LTG and GP, TPM was used as an anti-nociceptive drug for humans before systematic research on animal models of pain was undertaken. Thus far, published clinical experience with TPM is limited. One case report describes a 60-year-old male patient with post-thoracotomy pain syndrome that remained unresponsive to a host of other treatments for more than 3 years. Two daily doses of TPM, gradually increased up to 50 mg and 75 mg, provided consistent 80% pain relief around the clock without intolerable side effects for more than 6 months.23,27

A single-center, double-blind, randomized, placebo-controlled, 14-week study designed to evaluate the efficacy and safety of TPM in 27 patients with painful diabetic neuropathy reported a statistically significant reduction in average pain scores as measured on the Short-Form McGill Pain Questionnaire (SFMPQ) and the 100-mm visual analog scale of the SFMPQ. TPM (or placebo) was initiated at 25 mg/day and titrated over a 9-week period to 200 mg BID or a maximum tolerated dose. The most common adverse effects of the drug were asthenia, more than 10% weight loss, and confusion. The results suggest that TPM may represent a new option for the treatment of painful diabetic neuropathy.28

TPM may be a reasonable alternative treatment for patients experiencing various neuropathic pain syndromes, particularly for those who were unresponsive to other neuropathic analgesics (including anticonvulsants). TPM is well tolerated at low dosages (100–200 mg/day) and has few interactions with other drugs. Weight loss has been reported in all controlled clinical trials of TPM.

Migraine Headache Prevention

Migraine headache is a chronic, disabling disorder that affects some 30 million Americans and accounts for 20% of all outpatient visits to neurologists. According to the American Migraine Study29,30 (a 1989–1999 landmark population-based survey), 13% of the United States population (18% of all women and 6% of men) suffer from this disorder. The current concept of migraine pathophysiology holds that the underlying mechanism of migraine (with and without aura) is neurovascular in origin and that the primary dysfunction arises in the CNS. The neurovascular theory emphasizes the essential interaction between nerves and vessels, and integrates the known physiologic and pathophysiologic mechanisms. Migraine responds to various treatments, but tryptans are the drugs of choice for the acute attack. A number of treatments are available for the prevention of migraine, although AEDs are the choice drugs. VPA is approved for migraine prevention, and GP and TPX have had clinical acceptance as excellent preventatives.

Valproic Acid

The efficacy and safety of this established AED as prophylactic monotherapy was evaluated in a multicenter, double-blind, placebo-controlled, parallel-group study of 176 patients. The primary efficacy variable was 4-week migraine attack frequency, and the daily dosages were 500 mg, 1,000 mg, 1,500 mg, or placebo. Mean reductions in migraine attack were 1.7, 2.0, 1.7, and 0.5, respectively (P=.05 versus placebo). The 1,000-mg dosage was superior to the 500- and 1,500-mg dosages. Approximately 45% of the three groups treated with divalproex sodium achieved a >50% reduction in attack frequency versus 21% of the placebo group (P<.05).31

The results of two 12-week, double-blind, placebo-controlled studies and one 36-month open-label extension study (N=248) were as follows: 37% of patients reported nausea, 34% reported tremor, 24% reported weight gain, and 12% reported alopecia. Although nausea and alopecia subsided over time, tremor and weight gain persisted.32


After positive observations in an open-label study, 145 patients were randomized to receive GP in a multicenter, double-blind, placebo-controlled trial. Initial dosages of 300 mg/day were gradually increased to 1,800 or 2,400 mg/day, and patients remained at these fixed dosages for an 8-week stabilization period. Analysis showed that the median headache rate during the last 4 weeks was significantly lower in patients receiving GP versus those receiving placebo (2.7 versus 3.5 attacks, respectively; P=.006). Compared with placebo, GP was also associated with a significantly greater number of patients achieving a >50% reduction in migraine rate (16% versus 46%, respectively; P=.008).33

Somnolence, dizziness, and asthenia were observed in more than 10% of the GP-treated patients in the above study. However, only somnolence occurred significantly more often with the active treatment versus placebo (24.5% versus 8.9%; P=.04).33


Based on positive findings in open-label investigations of patients with refractory migraine or intractable daily headache, two single-center, double-blind, placebo-controlled trials were undertaken.34,35 Although the former evaluated TPM as monotherapy, concomitant prophylaxis medication was permitted in the latter. The target dosage was 100 mg BID. In both studies, a 4-week baseline period was followed by a titration phase of 6–8 weeks and a maintenance phase of 8–12 weeks. The results of the studies showed a mean reduction in migraine frequency per 28 days of 1.2 and 1.8 versus 0.4 and 0.5, respectively, for the placebo groups.

The 50% responder rate was higher in the TPM monotherapy group (46.7% versus 6.7% for placebo) than in the add-on treatment group (26.3% versus 9.5% for placebo). A subsequent analysis of two subsets of patients (with and without prophylactic migraine medications at baseline or during the study) showed no significant difference between treatment groups in patients receiving prophylactic medications. In contrast, the mean reduction (2.1) in 28-day headache frequency for the TPM group was significantly greater than that for the placebo group (0.3, P=.00006) in patients with a washout of baseline prophylactic medications.34 (TPM is currently being evaluated as monotherapy in a large, multicenter, placebo-controlled trial to determine its response.)

Adverse effects were mild to moderate and included paresthesias, drowsiness/ somnolence, decreased appetite/anorexia, diarrhea, weight loss, altered taste, and memory impairment. In contrast to divalproex sodium and GP, weight loss is a characteristic feature; TPM-treated patients in the two studies showed a statistically significant mean weight reduction (compared with placebo) of 4.7% to >10% of their body weight.34

Other Neurological Disorders

The AEDs are effective in various other disorders. GP and TPX given before bedtime in doses of 300–900 mg and 50–100 mg, respectively, are effective in nighttime restless legs syndrome and muscle cramps. Primidone, one of the older drugs, and TPX significantly reduce benign essential tremor.36 Dosages range from 250–500 mg/day for primidone and 200–400 mg/day for TPX. AEDs may be helpful in spasticity. GP has been anecdotally reported to be helpful as adjunctive therapy to baclofen. TG, a GABA reuptake inhibitor, has been very effective both alone and as adjunctive therapy to baclofen in refractory spasticity.

Weight gain is prominent with several of the AEDs, such as VPA, CBX, and GP. VPA has resulted in significant weight gain in many of the neurological and psychiatric disorders for which it is used. TPX results in weight loss when used in these disorders. The drug has also been used primarily for weight loss in dosages of 50–200 mg/day. Zonisamide has also been reported to result in weight loss in controlled clinical trials for epilepsy treatment.


AEDs are effective in a wide variety of neurological and psychiatric disorders. Their mechanisms of action serve to control neuronal excitation. AEDs are generally safe and well tolerated, and when dosed and titrated properly have a low adverse-effect profile.   PP


1.    Macdonald RL. Inhibitory synaptic transmission. In: Engel J, Pedley TA, eds. Epilepsy: A Comprehensive Textbook. Philadelphia, Pa: Lippincott–Raven Publishers; 1997:1383-1391.
2.    Taylor CP. Mechanisms of action of new antiepileptic drugs. In: Chadwick D, ed. New Trends in Epilepsy Management: The Role of Gabapentin. London, England: Royal Society of Medicine Services Limited; 1993:13-40.
3.    Stefani A, Spadoni F, Bernardi G. Voltage-activated calcium channels: targets of antiepileptic drug therapy. Epilepsia. 1997;38:959.
4.    Snutch TP, Reiner PB. Ca2+ channels: diversity of form and function. Curr Opin Neurobiol. 1992;2:247-253.
5.    McLean MJ. Gabapentin in the management of convulsive disorders. Epilepsia. 1999;40(suppl 6):S39-S50.
6.    Taylor CP, Gee NS, Su TZ. A summary of mechanistic hypotheses of gabapentin pharmacology. Epilepsy Res. 1998;29:233-249.
7.    Yan QS, Mishra PK, Burger RL, Bettendorf AF, Jobe PC, Dailey JW. Evidence that carbamazepine and antiepilepsirine may produce a component of their anticonvulsant effects by activating serotonergic neurons in genetically epilepsy-prone rats. J Pharmacol Exp Ther. 1992;261:652-659.
8.    Ropert N, Guy N. Serotonin facilitates GABAergic transmission in the CA1 region of rat hippocampus in vitro. J Physiol. 1991;441:121-136.
9.    Loscher W. Pharmacology of glutamate receptor antagonists in the kindling model of epilepsy. Prog Neurobiol. 1998;54:721-741.
10.     White HS. Mechanisms of action of antiepileptic drugs. Epilepsia. 1999;40(suppl 5):S2-S10.
11.     Olsen RW, DeLorey TM. GABA and glycine. In: Siegel MJ, Agronoff BW, Albers RW, Molinet PB, eds. Basic Neurochemistry: Molecular, Cellular and Medical Aspects. 6th ed. Philadelphia, Pa: Lippincott–Raven Publishers; 1999:335-346.
12.     Marshalec W, Narahashi T. Use-dependent pentobarbital block of kainate and quisqualate currents. Brain Res. 1993;608:7-15.
13.     Kramer LD, Reife RA. Topiramate. In: Engel J, Pedley TA, eds. Epilepsy: A Comprehensive Textbook. Philadelphia, Pa: Lippincott–Raven Publishers; 1997:1593-1598.
14.     Sommerville K. Tiagabine. In: Engle J, Pedley TA, eds. Epilepsy: A Comprehensive Textbook. Philadelphia, Pa: Raven Press; 1997:1645-1653.
15.     Genton P, Van Vleymen B. Piracetam and levetiracetam: close structural similarities but different pharmacological and clinical profiles. Epilepsy Disord. 2000;2:99-105.
16. Lamotrigine [package insert]. Research Triangle Park, NC: Glaxo Wellcome, Inc; 1999.
17.     Calabrese JR, Bowden CL, Sachs GS, Ascher JA, Monaghan E, Rudd GD. A double-blind placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. J Clin Psychiatry. 1999;60:79-88.
18.     Collins SL, Moore RA, McQuay HJ, Wiffen P. Antidepressants and anticonvulsants for diabetic neuropathy and postherpetic neuralgia: a quantitative systematic review. J Pain Symptom Manage. 2000;20:449-458.
19.     Pande AC, Davidson JR, Jefferson JW, et al. Treatment of social phobia with gabapentin: a placebo-controlled study. J Clin Psychopharmacol. 1999;19:341-348.
20.     Sindrup SH, Jensen TS. Efficacy of pharmacological treatments of neuropathic pain: an update and effect related to mechanism of drug action. Pain. 1999;83:389-400.
21.     Bloom S. Trigeminal neuralgia: its treatment with a new anticonvulsant drug. Lancet. 1962;1:839-840.
22.     Bloom S. Tic douloureux treated with new anticonvulsant drug. Lancet. 1963;9:285-290.
23.     Tremont-Lukats IW, Megeff C, Backonja MM. Anticonvulsants for neuropathic pain syndromes. Drugs. 2000;60:1029-1052.
24.     Backonja M, Beydoun A, Edwards KR, Bernstein P, Magnus-Miller L. Gabapentin for the symptomatic treatment of painful neuropathy in patients with diabetes mellitus: a randomized controlled trial. JAMA. 1998;280:1831-1836.
25.     Rowbotham M, Harden N, Stacey B, Bernstein P, Magnus-Miller L. Gabapentin for the symptomatic treatment of postherpetic neuralgia: a randomized controlled trial. JAMA. 1998;280:1837-1842.
26.     Nicholson B. Gabapentin use in neuropathic pain syndromes. Acta Neurol Scand. 2000;101:359-371.
27.     Bajwa ZH, Sami N, Warfield CA, Wootton J. Topiramate relieves refractory intercostal neuralgia. Neurology. 1999;52:1917.
28.     Edwards KR, Glantz MJ, Button J, et al. Efficacy and safety of topiramate in the treatment of painful diabetic neuropathy: a double-blind, placebo-controlled study. Neurology. 2000;54(suppl 3):147.
29.     Lipton RB, Stewart WF. Migraine in the United States: a review of epidemiology and health care use. Neurology. 1993;43(suppl 3):S6-S10.
30.     Lipton RB, Stewart WF, Simon D. Medical consultation for migraine: results from the American Migraine Study. Headache. 1998;38:87-96.
31.     Klapper J. Divalproex sodium in migraine prophylaxis: a dose-controlled study. Cephalalgia. 1997;17:103-108.
32.     Silberstein SD, Collins SD. Safety of divalproex sodium in migraine prophylaxis: an open-label, long-term study. Headache. 1999;39:633-643.
33.     Matthew NT, Rapoport A, Saper J, et al. Efficacy of gabapentin in migraine prophylaxis. Headache. 2001;41:119-128.
34.     Edwards KR. Migraine treatment and prophylaxis. Paper presented at: Headache World 2000; September 2000; London, England.
35.     Potter DL. A double-blind, randomized, placebo-controlled study of topiramate in the prophylactic treatment of migraine with and without aura. Paper presented at: Headache World 2000; September 2000; London, England.
36.     Connor, G. Treatment of benign essential tremor with topiramate. Paper presented at: The 28th Annual Meeting of the Southern Clinical Neurological Society; January 13, 2001; Puntarenas, Costa Rica.