Antiepileptic Drugs, 5th Edition



Clinical Efficacy and USE in Other Neurologic Disorders

Ettore Beghi MD

Chief, Neurophysiology Unit, “San Gerardo” Hospital, Monza, Italy; and Head, Neurological Disorders Laboratory, Institute for Pharmacological Research “Mario Negri,” Milan, Italy

Carbamazepine (CBZ) has been one of the most extensively investigated antiepileptic drugs for the treatment of neurological disorders other than epilepsy. CBZ exerts a use-dependent inhibition of sodium channels and reduces the frequency of sustained repetitive firing of action potentials in neurons. CBZ also may inhibit the release of somatostatin and have some calcium antagonistic effects (1,2). These mechanisms of action may explain the efficacy of CBZ in conditions like neuropathic pain and pain syndromes, hemifacial spasm, restless legs syndrome (RLS), or myotonia. The analgesic effects of CBZ are correlated with the plasma levels of the drug (3). However, most of the evidence on the use of CBZ in these disorders comes from nonrandomized studies or from small randomized trials. This prevents solid conclusions on the clinical efficacy of this drug in many of these conditions.


Neuropathic pain is a spectrum of neuralgic pain syndromes that includes trigeminal neuralgia and neuralgias affecting other cranial or peripheral nerves (glossopharyngeal, superior laryngeal, postherpetic), diabetic neuropathy, thalamic syndrome, phantom limb pain, tabetic pain, cancer pain, and others.

Trigeminal and Other Chronic Neuralgias

CBZ has been extensively investigated for the treatment of trigeminal neuralgia, a paroxysmal form of facial pain commonly affecting the second and third division of the trigeminal nerve. The results of randomized clinical trials on the use of CBZ in patients with trigeminal or other neuralgias are illustrated in Table 24.1 (4, 5, 6, 7, 8, 9, 10, 11). The daily dose ranged from 100 to 2,400 mg. The drug was manifestly superior to placebo and was more effective or better tolerated than other active comparators (tizanidine, tocainide, pimozide). Pain relief was obtained in up to 84% of cases and persisted after prolonged treatment in up to 80% (8,12). In a systematic review of three placebo-controlled studies (13), the odds ratio (OR) for efficacy of CBZ was 4.9 [95% confidence interval (CI), 3.4 to 6.9); however, patients on active treatment had more adverse events (OR, 3.7; 95% CI, 2.2 to 6.2) and were more frequently withdrawn from treatment (OR, 6.2; 95% CI, 1.2 to 31.7). The number-needed-to-treat for effectiveness of CBZ compared with placebo was 2.6 (95% CI, 2.0 to 3.4), and that for adverse events was 3.7 (95% CI, 2.4 to 7.8) (14).

The efficacy of CBZ in neuropathic pain other than trigeminal neuralgia is supported by less consistent findings (Table 24.2). The drug has been compared with placebo, prednisolone, nortriptyline-fluphenazine, or transcutaneous electrical nerve stimulation (TENS) for the management of pain in patients with diabetes (15,16), stroke (17), herpes zoster (18,19), and Guillain-Barré syndrome (20). CBZ 150 to 1,000 mg/day was better than placebo and TENS, similar to nortriptyline-fluphenazine, and less effective than prednisolone and amitriptyline. Nonrandomized studies (mostly case reports and case series) reported CBZ being effective in glossopharyngeal neuralgia, phantom limb pain, multiple sclerosis, thalamic syndrome, and miscellaneous neuralgias (2).

Based on the results of the randomized clinical trials, CBZ can still be considered as a first-line drug for the treatment of trigeminal neuralgia. The starting dose may be 100 to 200 mg. In most cases, the daily dose should be increased gradually until pain relief is achieved, up to 1,000 to 1,200 mg. In this population of predominantly elderly patients, adverse effects may appear if dosing is not very low and gradual . Occasionally, further dose increments may be suggested. The daily dose of 2,400 mg should never be exceeded. Given the spontaneous remission of pain and the toxic effects of long-term treatment with CBZ, treatment withdrawal should be recommended after 2 to 3 months in patients with pain relief. In patients not responding to CBZ, phenytoin, gabapentin, lamotrigine, or topiramate could be attempted as monotherapy or in combination. Patients with intractable pain may require either percutaneous thermocoagulation of the trigeminal ganglion or direct exploration of the trigeminal root.




No. Treated (age, yr) [Disease]

Treatment Duration [Double-Blind Period]

Daily Dose, mg [Comparator]

Significant Results [No. Improved]

Adverse Eventsa [No. Withdrawals]


9 (37-76) [Trigeminal]

3 days

600 [PLC]

CBZ preferred by 8/9; CBZ and PLC equally effective in 1/9

(11 taking CBZ in open trial also assessed) CBZ 14/20; PLC 3/9 [CBZ1/20; PLC 0/9]


77 (20-84) [Trigeminal]

4 wk (two 2-wk periods)

400-800 [PLC]

Reduction of pain severity to 57% with CBZ and 15% with PLC (32% difference)
No. paroxysms reduced by 68% (PLC 26%). Effect of triggers reduced by 68% (PLC 40%) [CBZ 144/268; PLC 35/190]

CBZ 38/77; PLC 20/77 [CBZ 1/77; PLC 0/77]


42 (36-83) [Trigeminal 30; postherpetic 6; tabetic 2; atypical facial pain 4]

10 days

400-1,000 [PLC]

[Mild improvement to complete recovery with CBZ in 100% with trigeminal neuralgia, postherpetic and tabetic pain; PLC response minimal or absent in all patients]

(Only trigeminal neuralgia) CBZ 23/36 [CBZ 3/30]


37 (?) [Trigeminal]

2 wk

100-2,400 [PLC]

[CBZ 15/20; PLC 6/24]

Drowsiness: CBZ 10/37 Ataxia: CBZ 7/37 [CBZ 2/37; PLC 0/24]


71 (30-80+) [Trigeminal typical 40; atypical 16; other facial pain 15]

One or two 5-day periods

400-600 [PLC]

[Typical trigeminal: CBZ 33/40; PLC 8/40; atypical: CBZ 13/15; PLC 1/15; other facial pain: CBZ 8/15; PLC 2/15]

(Including long-term assessment) CBZ 9/71 [CBZ 4/71; PLC 0/71]


12 (47-72) [Trigeminal]

3 wk

900 [TZN 18]

[CBZ 4/6; TZN 1/5]

[TZN 3/6; CBZ 0/6]


12 (41-78) [Trigeminal]

2 wk

Max. tolerated dose [TCN 60/kg]

[12/12 on both drugs]

TCN 3/12; [TCN 1/12; CBZ 0/12]


48 (48-68) [Trigeminal]

8 wk

300-1,200 [PMZ 4-12; PLC]

Arbitrary symptom score reduced by 78% with PMZ and 50% with CBZ [PMZ 18/48; CBZ 27/48]

PMZ 40/48 CBZ 21/48

CBZ, carbamazepine; PLC, placebo; PMZ, pimozide; TCN, tocainide; TZN, tizanidine.

a No. with any event or, if unavailable, with most common events.



No. Treated (age, yr) [Disease]

Treatment Duration [Double-Blind Period]

Daily Dose, mg [Comparator]

Significant Results [No. Improved]

Adverse Eventsa [No. Withdrawals]


30 (21-81) [Diabetic neuropathy]

2 wk

Up to 600 [PLC]

[CBZ 28/30; PLC 19/30]

Somnolence: CBZ 16/30
Dizziness: CBZ 12/30 [CBZ 2/30; PLC 0/30]


16 (?) [Diabetic neuropathy]

4 wk

600 [FPZ/NTP 1.5/30]

Visual analogue scale no difference between treatments

CBZ 3/16; FPZ 8/16 [CBZ 2?]


15 (53-74) [Stroke]

4 wk

Up to 800 [AMT up to 75; PLC]

Mean pain intensity at week 4 (10-step verbal scale): CBZ 4.2; AMT 4.2; PLC 5.3 [CBZ 5/14; AMT 10/15; PLC 1/15]

CBZ 13/14; AMT 14/15; PLC 7/15 [CBZ 1/14; AMT 0/15; PLC 0/15]


29 (NR) [Postherpetic neuralgia]

8 wk

150-1,000+ CLP 10-75 [TNS]

Mean improvement (pain difference) with CBZ+CLP: 43.1 points (TNS 0.2 points) [CBZ+CLP 11/12 TNS 2/4]

[CBZ 3/16; TCN 2/13]


40 (50-86) [Acute herpes zoster]

4 wk

400 [PDN 40 for 10 days, then tapered]

Healing faster with PDN Neuralgia: CBZ 13/20; PDN 3/20]



12 (22-54) [Guillain-Barré syndrome]

7 days

300 [PLC]

Pain score (range 0-5): CBZ 1.7; PLC 3.1
Sedation score (range 0-6): CBZ 2.3; PLC 4.2
Pethidine requirement: CBZ 1.7 mg/kg/day; PLC 3.7 mg/kg/day

None significant

ADL, activities of daily living; AMT, amitriptyline; ASP, aspirin; CBZ, carbamazepine; CLP, clonazepam; FPZ, fluphenazine; NR, not reported; NTP, nortriptyline; PDN, prednisolone; PLC, placebo; TNS, transcutaneous
electrical nerve stimulation.

a No. with any event or, if unavailable, with most common events.



The use of CBZ also can be considered for the treatment of glossopharyngeal neuralgia, based on common pathophysiologic mechanisms. Less consistent findings support the use of the drug in other chronic pain syndromes, for which no definite guidelines can be proposed. Because the risk of adverse treatment events and drug withdrawal is high with CBZ, the drug should be used only with constant reference to its risk-benefit profile.


The frequency of migraine and tension-type headache and their ill-defined pathogenic mechanisms, spectrum of severity, and variable response to treatment prompted the investigation of some anticonvulsant drugs, including valproic acid, clonazepam, lamotrigine, and CBZ.

There is only one double-blind, crossover, placebo-controlled study assessing the efficacy of CBZ in the treatment of migraine (21). In this study, 48 patients aged 14 to 60 years were treated with CBZ 600 mg/day or equivalent placebo for 6 weeks. CBZ was better than placebo in reducing the number of attacks and increasing the number of responders. Adverse events were most common in patients given active treatment (CBZ, 30/45; placebo, 11/48), but only one case on CBZ needed to withdraw from treatment. Despite these encouraging findings, evidence is insufficient to support the use of CBZ in migraine, tension-type, or other chronic headache syndromes.


RLS is a common sensorimotor disorder with an estimated prevalence of 1% to 5%, characterized by unpleasant sensation in the legs occurring predominantly during sleep (22). This syndrome may cluster in families and be secondary to other clinical conditions, like anemia, pregnancy, or peripheral neuropathies. The effects of CBZ in patients with RLS have been tested in two randomized clinical trials. In a small, placebo-controlled, crossover study in six patients, CBZ up to 600 mg/day reduced the number of attacks in three cases (placebo, 0) (23). The apparent efficacy in RLS was confirmed in a larger, double-blind, placebo-controlled study involving 174 patients. In this study, CBZ up to 300 mg/day was significantly more effective than placebo (24). The median daily dose of CBZ was 236 mg and the median drug plasma concentration was 12 µmol/L. Thirty-four patients receiving CBZ experienced unwanted effects (six withdrawals). However, in this trial, even placebo showed a significant therapeutic effect, which indicates the need of placebo-controlled studies to confirm the efficacy of any treatment of this disorder. The treatment of RLS usually is symptomatic and is causal only in the secondary forms. The treatments of choice include dopaminergic drugs and benzodiazepines. CBZ 200 to 300 mg/day can be used as a second-choice drug for idiopathic or cryptogenic RLS.

Nocturnal paroxysmal dystonia and episodic nocturnal wandering are complex motor attacks occurring during sleep and characterized by sudden arousal followed by dystonic posturing and semipurposeful activity (25). The attacks may remit with CBZ, given at very low dosages in open trials, and recur after treatment withdrawal.


Myotonia is a clinical manifestation of different muscle disorders characterized by altered muscle membrane physiology. CBZ (600 and 800 mg/day) has been compared with phenytoin (200 and 300 mg/day) and placebo for the treatment of myotonia in six patients with myotonic dystrophy (26). In this double-blind, crossover study, each treatment was given for 15 days. CBZ was similar to phenytoin (but better than placebo) in decreasing the time of myotonic afterdischarges and improving myotonic symptoms, with no dose differences.

Because there are no standard reference treatments, CBZ might be indicated as a therapeutic option for myotonia. The risk-benefit profile of the drug and the availability of alternative drugs [phenytoin (Chapter 61), diazepam) must be considered when starting treatment of patients with myotonia. Additional randomized trials are needed to assess the efficacy of CBZ and other drugs in the treatment of myotonia.


The use of CBZ has been considered for the symptomatic treatment of agitation and aggressiveness in patients with dementia. Although the rationale for the effects of CBZ in agitation of patients with dementia is unknown, its use has been advocated on the basis of its apparent efficacy on agitation, aggression, irritability, and impulsivity in several clinical disorders, despite the negative results of a small controlled study (27). In a multicenter, double-blind, placebo-controlled, 6-week study of 51 patients with Alzheimer's disease or vascular dementia and agitation living in nursing homes, CBZ was started at 100 mg/day and given at a modal daily dose of 300 mg (mean serum concentration of 5.3 µg/mL) (28). Active treatment was better than placebo in reducing agitation and aggression, as indicated by the significant reduction of the scores of the Brief Psychiatric Rating Scale (CBZ, 7.7 points; placebo, 0.9 points) and by the


improvement of the Clinical Global Impression rating (CBZ, 77%; placebo, 21%). Agitation and hostility were the most affected behavioral abnormalities. Extra time required to attend to behavioral problems was significantly lower with CBZ. However, adverse events (mostly drowsiness, disorientation, and ataxia) were more common with active treatment (59% versus 29%), and dropouts were reported only with CBZ (four cases). Subsequent withdrawal of CBZ showed reversion to baseline state in patients previously showing improved behavior, compared with no change in patients stopping placebo (29). On this basis, chronic CBZ use at low to average daily doses might be considered an alternative option for the treatment of agitation and aggressiveness in patients with dementia.


Attention-deficit/hyperactivity disorder (ADHD) is a rather common clinical condition that is present in approximately 3% of school-age children and adolescents (30). Despite several psychopharmacologic options, there are many nonresponders in whom the persisting behavioral symptoms may affect academic and social functioning. CBZ has been repeatedly assessed as a treatment option in children and adolescents with ADHD, and a meta-analysis has been performed of the international reports concerning the efficacy of the drug in this study population (31). Seven open studies (189 patients; CBZ daily dose, 100 to 800 mg) and three double-blind, placebo-controlled studies (57 patients; CBZ daily dose, 200 to 600 mg) reported CBZ being effective in ADHD, with significant changes from baseline in impulsivity, distractibility, and motor overactivity. The meta-analysis of the three controlled studies showed CBZ being significantly superior (p = .018) to placebo, with 71% of patients experiencing substantial improvement (placebo, 26%). Sedation (7% to 13%), rash (5%), and ataxia (2%) were the most common adverse events, which were reported as mild and transient. Based on this meta-analysis, CBZ 200 to 600 mg/day can be considered safe and effective for children with ADHD.


Hemifacial spasm is a chronic movement disorder of the face characterized by twitching, tonic spasm, and synkinesis of the muscles innervated by the facial nerve. The recommended treatment of hemifacial spasm is the local injection of type A botulinum toxin. Pharmacotherapy of hemifacial spasm includes membrane-stabilizing anticonvulsants like CBZ and phenytoin; however, the efficacy of these drugs is sustained only by small open trials with inadequate follow-up.

CBZ (400 and 600 mg/day) was effective in reducing cerebellar tremor in a small, single-blind, placebo-controlled trial (32). By contrast, CBZ seems ineffective in the treatment of tinnitus (33,34) and stuttering (35).


  1. Waterhouse E, DeLorenzo RJ. Mechanisms of action of antiepileptic drugs: An overview. In: Shorvon S, Dreifuss F, Fish D, et al., eds. The treatment of epilepsy.Oxford: Blackwell Science, 1996:123-137.
  2. Swerdlow M. Anticonvulsant drugs and chronic pain. Clin Neuropharmacol1984;7:51-82.
  3. Tomson T, Ekbom K. Trigeminal neuralgia: time course of pain in relation to carbamazepine dosing. Cephalalgia1981;1:91-97.
  4. Rockliff BW, Davis EH. Controlled sequential trials of carbamazepine in trigeminal neuralgia. Arch Neurol1966;15:129-136.
  5. Campbell FG, Graham JG, Zilkha KJ. Clinical trial of carbamazepine (Tegretol) in trigeminal neuralgia. J Neurol Neurosurg Psychiatry1966;29:265-167.
  6. Killian JM, Fromm GH. Carbamazepine in the treatment of neuralgia. Arch Neurol1968;19:129-136.
  7. Nicol CF. A four year double blind study of Tegretol in facial pain. Headache1969;9:54-57.
  8. Rasmussen P, Riishede J. Facial pain treated with carbamazepine (Tegretol). Acta Neurol Scand1970;46:385-408.
  9. Viming ST, Lyberg T, Lataste X. Tizanidine in the management of trigeminal neuralgia. Cephalalgia1986;6:181-182.
  10. Lindstrom P, Lindblom U. The analgesic effect of tocainide in trigeminal neuralgia. Pain1987;28:45-50.
  11. Lechin F, van der Dijs B, Lechin ME, et al. Pimozide therapy for trigeminal neuralgia. Arch Neurol1989;46:960-963.
  12. Carnaille H, De Coster J, Tyberghein J, et al. Etude statistique de près de 700 cas de facialgies traiteès par le Tégrétol. Acta Neurol Belg1966;66:175-196.
  13. McQuay H, Carroll D, Jadad AR, et al. Anticonvulsant drugs for management of pain: a systematic review. BMJ1995;311: 1047-1052.
  14. Wiffen P, Collins S, McQuay H, et al. Anticonvulsant drugs for acute and chronic pain. In: Cochrane database of systematic reviews.2000:CD00133.
  15. Rull JA, Quibrera R, Gonzales-Millan. Symptomatic treatment of peripheral diabetic neuropathy with carbamazepine (Tegretol): Double blind crossover trial. Diabetologia1969;5:215-218.
  16. Gomez-Perez FJ, Choza R, Rios JM, et al. Nortriptyline-fluphenazine vs carbamazepine in the symptomatic treatment of diabetic neuropathy. Arch Med Res1996;27:525-529.
  17. Leijon G, Boivie J. Central post-stroke pain: A controlled trial of amitriptyline and carbamazepine. Pain1989;36:27-36.
  18. Gerson GR. Studies on the concomitant use of carbamazepine and clomipramine for the relief of post-herpetic neuralgia. Postgrad Med J1977;53[Suppl 4]:104-109.
  19. Keczkes K, Basheer AM. Do corticosteroids prevent post-herpetic neuralgia? Br J Dermatol1980;102:551-555.
  20. Tripathi M, Kaushik S. Carbamazepine for pain management in Guillain-Barre syndrome patients in the intensive care unit. Crit Care Med2000;28:655-658.
  21. Rompel H, Bauermeister PW. Aetiology of migraine and prevention with carbamazepine (Tegretol): Results of a double-blind, cross-over study. S Afr Med J1970;44:75-80.
  22. Wetter TC, Pollmacher T. Restless legs and periodic leg movements in sleep syndromes. J Neurol1997;244[Suppl 1]:S37-S45.
  23. Lundvall O, Abom PE, Holm R. Carbamazepine in restless legs: A controlled pilot study. Eur J Clin Pharmacol1983;25:323-324.



  1. Telstad W, Sorensen O, Larsen S, et al. Treatment of the restless legs syndrome with carbamazepine: A double blind study. BMJ1984;288:444-446.
  2. Montagna P. Nocturnal paroxysmal dystonia and nocturnal wandering. Neurology1992;42[Suppl 6]:61-67.
  3. Sechi GP, Traccis S, Durelli L, et al. Carbamazepine versus diphenylhydantoin in the treatment of myotonia. Eur Neurol1983;22:113-118.
  4. Chambers CA, Bain J, Rosbottom R, et al. Carbamazepine in senile dementia and overactivity: A placebo controlled double blind trial. IRCS Med Sci1982;10:505-506.
  5. Tariot PN, Erb R, Podgorski CA, et al. Efficacy and tolerability of carbamazepine for agitation and aggression in dementia. Am J Psychiatry1998;155:54-61.
  6. Tariot PN, Jakimovich L, Erb R, et al. Withdrawal from controlled carbamazepine therapy followed by further carbamazepine treatment in patients with dementia. J Clin Psychiatry1999;60:684-689.
  7. Safer DJ, Krager JM. A survey of medication treatment for hyperactive/inattentive students. JAMA1988;260:2256-2258.
  8. Silva RR, Munoz DM, Alpert M. Carbamazepine use in children and adolescents with features of attention-deficit hyperactivity disorder: A meta-analysis. J Am Acad Child Adolesc Psychiatry1996;35:352-358.
  9. Sechi GP, Zuddas M, Piredda M, et al. Treatment of cerebellar tremors with carbamazepine: A controlled trial with long-term follow-up. Neurology1989;39:1113-1115.
  10. Hulshof JH, Vermeij P. The value of carbamazepine in the treatment of tinnitus. J Otorhino Laryngol Relat Spec1985;47: 262-266.
  11. Yang DJ. Tinnitus treated with combined traditional Chinese medicine and Western medicine. Chung Hsi I Chieh Ho Tsa Chih [Chin J Mod Dev Trad Med]1989;9:259-260, 270-271.
  12. Harvey JE, Culatta R, Halikas JA, et al. The effects of carbamazepine on stuttering. J Nerv Ment Dis1992; 180:451-457.


Beghi E. The use of anticonvulsants in neurological conditions other than epilepsy: A review of the evidence from randomized controlled trials. CNS Drugs 1999;11:61-82.