Antiepileptic Drugs, 5th Edition



Efficacy and Use in Psychiatric Disorders

Melvin D. Shelton MD, PhD*

Joseph R. Calabrese MD**

* Clinical Trials Section, Mood Disorders Program; and Assistant Professor of Psychiatry, Case Western Reserve University School of Medicine, University Hospitals of Cleveland, Cleveland, Ohio

** Professor of Psychiatry, Department of Psychiatry, Case Western Reserve University School of Medicine; and Director, Mood Disorders Program, Department of Psychiatry, University Hospitals of Cleveland, Cleveland, Ohio

During clinical trials in the development of lamotrigine as an antiepileptic drug, it was observed that patients demonstrated improvements in mood and sense of well-being. There are theoretical reasons to suppose that lamotrigine might possess mood-stabilizing properties; mood swings may be associated with the kindling process, as are seizures. Bipolar disorders are defined as cyclic affective patterns comprising different combinations of mania, hypomania, depression, mixed states (coincident mania and depression), and rapid cycling (four or more cycles per year). These different mood states are not equally amenable to treatment; bipolar depression and rapid-cycling bipolar disorder are particularly treatment refractory. The first prospective, randomized, placebo-controlled study of lamotrigine monotherapy in bipolar-I depressed patients has shown that patients receiving 200 mg and 50 mg daily showed significant improvement compared with placebo on several measures of depression. In the first double-blind maintenance study of lamotrigine and the first controlled study of any medication in a cohort of patients with rapid-cycling bipolar disorder, results indicated lamotrigine was a useful treatment for those patients with bipolar II disorder. Lamotrigine has not been shown to have clear efficacy in the treatment of mania or unipolar depression. A series of controlled studies investigating the use of lamotrigine in bipolar disorder is ongoing.

There are theoretical reasons to postulate that antiepileptic drugs (AEDs) might be effective in the treatment of mood disorders. Post and colleagues have suggested that mood instability leading to bipolar affective disorder (BAD) may be causally related, as are seizures, to progressive neural kindling (1). The Diagnostic and Statistical Manual for Mental Disorders, 4th edition, Text Revision (DSM-IV-TR) (2) describes bipolar disorder as a recurrent condition with a lifetime prevalence of approximately 0.4% to 1.6% in community samples. Patients with the disorder experience marked mood fluctuations, which have been described as type I (BP-I, showing depressions, possibly euthymias and manias), type II (BP-II, with depressions, possibly euthymias, and hypomanias, but not manias), and mixed states (simultaneously meeting criteria for both depression and mania). In men, the first episode is likely to be manic, and the proportion of manias to depressions is 1:1 or higher. In women, the index episode is likely to be depressive, and depressive episodes predominate during the course of the disorder (2). The sex ratio in patients with BP-I is equal; female patients with BP-II are more common (3). Any of these three variants may show a rapid cycling course (RC), defined as four or more mood states (i.e., at least two full bipolar cycles) per year (2). Patients with RC, comprising 13% to 20% of patients with bipolar disorder (4), are more likely to be female (5). There is evidence of ultrarapid cycling, with cycle periods as short as several days, and ultradian cycling, with periods of less than 24 hours (6). RC patients were initially identified by Dunner and Fieve as having poor prognoses, and poor responses to treatment with lithium (7), the earliest pharmacologic gold standard for treatment of bipolar disorder. It appears that only approximately 50% of patients with bipolar spectrum disorder, 30% to 40% of patients with mixed-state disorder, and 20% to 30% of patients with RC bipolar disorder respond to lithium treatment. Lithium as a first-line mood stabilizer has been joined by carbamazepine and divalproex. Like lithium, they appear to be less effective in the depressed phase of the illness, particularly in patients with RC bipolar disorder (8). Antidepressant medications are useful, but run the risk of inducing hypomania, mania, or increased cycle frequency (9). Because there are no published reports of any single mood stabilizer equally efficacious


in all phases of bipolar illness, polypharmacotherapy has been increasingly used (10,11). Clearly, there is a need for such a mood stabilizer.


Initial findings (12) of the efficacy of lamotrigine in the treatment of intractable seizures gave anecdotal evidence of mood improvement in treated patients. These findings were extended (13) in a study of 81 lamotrigine-treated patients with mixed epilepsy. The primary efficacy variable was seizure frequency; a secondary measure was quality of life. Although the quality-of-life examination was methodologically limited, there was evidence that patients thought of themselves as both happier and more in control of their situation, and that they experienced improvements in selfesteem. These reports led to closer scrutiny of the drug's effects in patients with mood disorders, especially bipolar affective disorder (13, 14, 15, 16, 17, 18).

There have been approximately 16 open studies of lamotrigine's efficacy as a mood stabilizer. The largest of these (18) was published in 1999. A 48-week prospective, open-label trial was conducted using lamotrigine as monotherapy (n = 15) or as an adjuvant (n = 60) in patients with BP-I and BP-II. Depressed patients exhibited a 42% improvement in scores on the Hamilton Depression Rating Scale (HAMD); patients with mania, hypomania, or mixed states showed a 74% improvement in scores on the Mania Rating Scale from the Schedule for Affective Disorders, Change Version (MRS). Eighty percent of nondepressed patients showed a marked response to treatment; only 60% of depressed patients showed similar improvement.


The first randomized, placebo-controlled, parallel-group trial evaluating the efficacy of lamotrigine monotherapy in the treatment of bipolar disorder has been completed (19). One hundred ninety-five adult outpatients diagnosed with BP-I depression and moderately to severely depressed at study entry were randomized to either 50 mg (LTG50, 63 patients) or 200 mg (LTG200, 66 patients) daily of lamotrigine, or to placebo (66 patients) at 15 U.S. centers and 6 centers in Australia, the United Kingdom, and France. Treatment groups were stratified to detect any lithium effect on subsequent lamotrigine treatment. Lamotrigine dosing was titrated according to the protocol depicted in Figure 38.1. The test instruments were the HAMD, MRS, Clinical Global Impression—Severity (CGI-S), and the Montgomery-Asberg Rating Scale for Depression (MADRS).

Lamotrigine was significantly more effective than placebo on most outcome measures. Patients on LTG200 showed significant improvement by all measures except for the 31-item HAMD [observed and last observation carried forward (LOCF) analysis] and the LOCF analysis of the 17-item HAMD. Over 50% of the LTG200 patients met response criteria for the HAMD-17, Clinical Global Impression—Intensity (CGI-I), and MADRS. On the latter two instruments, the degree of improvement was nearly twice that shown by patients in the placebo arm, and the difference reached statistical significance (p < .05). Patients in the LTG50 group showed less robust improvement than the LTG200 group (Figure 38.2). By design, both groups received the same lamotrigine doses for the first 3 weeks of treatment; the pooled patients on lamotrigine showed significant improvement over placebo by the end of week 3.


FIGURE 38.1. Lamotrigine at 50 and 200 mg/day versus placebo. (From Calabrese J, Bowden C, Sachs G, et al. A double-blind, placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. J Clin Psychiatry 2000;20:607-614, with permission.)

Two frequent impediments to successful treatment of patients with mood stabilizers are (a) idiopathic switches to mania, hypomania, or mixed states; and (b) poor medication


compliance secondary to intolerable side effects. In this study, approximately 5% of patients cycled into a manic, hypomanic, or mixed state. In contrast, switch rates as high as 25% for monoamine oxidase inhibitors and 21% for tricyclic antidepressants have been reported. (17). In each group, over 91% of patients were over 70% medication compliant. There also was no intergroup difference in the number of study withdrawals due to adverse effects.


FIGURE 38.2. Percentage of patients on placebo, lamotrigine 50 mg/day (LTG50), and lamotrigine 200 mg/day (LTG200) meeting end point response criterion (see text). (From Calabrese J, Bowden C, Sachs G, et al. A double-blind, placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. J Clin Psychiatry2000;20:607-614, with permission.)


As noted previously, the RC variant of bipolar disorder was initially identified as a characteristic of patients resistant to lithium treatment (7), so the development of a more effective mood stabilizer for use in this population is desirable. The first controlled, prospective study of adult RC patients has been published; it also is the first double-blind maintenance study of lamotrigine (20). Three hundred twenty-four patients entered treatment at 24 U.S. (n = 292) and three Canadian (n = 32) sites. The initial phase of the study was an open-label stabilization phase; the second phase was double-blind, placebo-controlled, and randomized (Figure 38.3). After screening, patients entered a 6-week titration to a target dosage of 100 to 300 mg daily. Flexible dosing was permitted to allow for differences in tolerability. During the first 4 weeks of treatment, patients were permitted additional psychotropic medications for acute episodes; over 4 to 8 weeks, all other psychotropics were tapered. To meet randomization criteria, subjects scored 14 or less on the HAMD, 12 or less on the MRS, and remained compliant on lamotrigine 100 mg/day, all over a 2-week period.


FIGURE 38.3. Schema for lamotrigine versus placebo comparison study. (From Calabrese J, Suppes T, Bowden C, et al. A double-blind, placebo-controlled prophylaxis study of lamotrigine in rapid-cycling bipolar disorder. J Clin Psychiatry 2000;61:841-850, with permission.)

Patients successfully negotiating the taper of psychotropics while retaining affective stability were offered advancement to the 26-week randomized phase, in which they underwent discontinuation of open-label lamotrigine and double-blind 1:1 randomization to lamotrigine or placebo treatment. In contrast to the monotherapy study, this study was open to patients with either BP-I or BP-II, and groups were appropriately stratified. As in the preliminary phase, lamotrigine dosing was flexible, and ranged from 100 to 500 mg/day. Use of up to 2 mg/day of lorazepam was permitted for agitation, hostility, and insomnia. The primary outcome variable was time to pharmacologic intervention to prevent a mood episode. A secondary outcome variable was survival time in study, that is, time to premature discontinuation for any reason. Secondary test instruments were the HAMD, MRS, CGI-S, and Global Assessment Scale.

The lamotrigine and placebo groups did not differ significantly in time to intervention; however, mean survival times were 18 weeks for lamotrigine and 12 weeks for placebo. The mean group survival times were 14 weeks for lamotrigine and 8 weeks for placebo; the difference was significant (p < .05). Figure 38.4 summarizes the differences in


response between patients with BP-I and BP-II disorder. There was no significant difference in lamotrigine versus placebo group survival times for patients with BP-I. For patients with BP-II, there was a trend toward significance in the time to intervention analysis, and a significant difference (p < .015) in the median survival in study times (lamotrigine 17 weeks, placebo 7 weeks).


FIGURE 38.4. Kaplan-Meier curves for survival time in study for bipolar disorder type I (BP-I), bipolar disorder type II (BP-II), and BP-I = BP-II groups. (From Calabrese J, Suppes T, Bowden C, et al. A double-blind, placebo-controlled prophylaxis study of lamotrigine in rapid-cycling bipolar disorder. J Clin Psychiatry 2000;61:841-850, with permission.)

Figure 38.4 summarizes findings from the 6-month stability data. The percentage of patients with BP-I stable for 6 months was not significantly different between lamotrigine and placebo groups (39% and 31%, respectively), but was significantly different (46% and 18%, respectively; p < .04) for patients with BP-II. Median survival time in study for patients with BP-I was not significantly different for those in lamotrigine and placebo groups; the mean for patients with BP-II was 17 weeks for lamotrigine and 7 weeks for placebo. However, these differing survival times may have been due to intergroup differences in placebo response rates. Thirty-seven of 60 patients stable for 6 months without relapse were in the lamotrigine group; this constituted 41% of the lamotrigine group. Twenty-three of 60 stable patients were in the placebo group; this was 26% of the placebo sample. The difference was significant (p < .03).

This study design had limitations in addition to the elevated BP-I placebo response rate. Because of the lack of previous lamotrigine monotherapy data in this population, the primary outcome measure study was discovered retrospectively to have been underpowered to detect the observed differences, although secondary measures were not. The abrupt discontinuation of lamotrigine at randomization may have produced artifactual affective changes, and the immediate intervention to prevent relapse prevented measurement of actual time to full relapse.

Comparisons between lamotrigine's efficacy in unipolar and in bipolar depression are of interest. There is one case report (21) of efficacious lamotrigine monotherapy in unipolar depression, and a report of two patients receiving effective lamotrigine adjuvant therapy (22). During development (Glaxo-Smith Kline, Research Triangle Park, North Carolina, data on file), evidence supported antidepressant activity in unipolar disorder. However, this conclusion was weakened by differences in the LOCF and observed analyses, and by a high placebo response rate. In a randomized,


double-blind, crossover trial, Frye and colleagues (23) studied lamotrigine and gabapentin monotherapy of patients with BP-I (n = 13), BP-II (n = 15), RC (n = 21), and unipolar depression (n = 10). Forty-five percent of lamotrigine-treated patients were assessed as improved or very much improved on the CGI. Response for patients with unipolar disorder (29%) was less than that for patients with bipolar disorder. In this study, more patients with BP-I (73%) improved than those with BP-II (47%). This result must be interpreted carefully, however, because the measure of improvement was of overall clinical impression of illness rather than a straightforward measurement of depression. Lamotrigine also has been evaluated in the acute management of mania (18). In a placebo-controlled designs using lamotrigine treatment as monotherapy and as add-on therapy, minimal efficacy was observed; lamotrigine was well tolerated in this study and did not worsen the symptoms of mania.

Additional evidence supporting the premise that lamotrigine might be expected to be more effective in patients with BP-II than in those with BP-I has been offered in several studies. The most recent examination of the question is the multisite study referred to previously (21): 57% of subjects were depressed at study entry, compared with 21% manic/hypomanic, 18% euthymic, and 5% mixed. Also in that study, a high percentage of both the lamotrigine and control groups (37% and 45%, respectively) reached study end point because of depression. Because depression appears to be a hallmark of BP-II disorder, it is reasonable to postulate that patients with BP-II disorder may respond to lamotrigine treatment. The issue of whether lamotrigine is more efficacious in BP-I or BP-II disorder awaits results of further controlled studies. However, the cumulative results of the studies done so far provide evidence that lamotrigine is effective in the acute management of the depressed phase of BP-I depression, and effective in the long-term stabilization of mood in patients with RC bipolar disorder.


  1. Post R, Weiss S, Pert A. Differential effects of carbamazepine and lithium on sensitization and kindling. Prog Neuropsychopharmacol Biol Psychiatry1984;8:425-434.
  2. American Psychiatric Association. Diagnostic and statistical manual of mental disorders,4th ed, text revision. Washington, DC: American Psychiatric Association, 2000.
  3. Goodwin F, Jamison K. Manic-depressive illness.New York: Oxford University Press, 1990.
  4. Bauer M, Calabrese J, Dunner D, et al. Multisite data reanalysis of the validity of rapid cycling as a course modifier for bipolar disorder in DSM-IV. Am J Psychiatry1994;151:506-515.
  5. Coryell W, Endicott J, Keller M. Rapidly cycling affective disorder: demographics, diagnosis, family history, and course. Arch Gen Psychiatry1992;49:126-131.
  6. Kramlinger K, Post R. Ultra-rapid and ultradian cycling in bipolar affective illness. Br J Psychiatry1996;168:214-323.
  7. Dunner D, Fieve R. Clinical factors in lithium prophylaxis failure. Arch Gen Psychiatry1974;30:229-233.
  8. Calabrese J, Bowden C, Woyshville M. Lithium and anticonvulsants in bipolar disorder. In: Bloom F, Kupfer D, eds. Psychopharmacology: the fourth generation of progress.CD-ROM Version 3,, 2000.
  9. Post R, Denicoff K, Leverich G, et al. Drug-induced switching in bipolar disorder: epidemiology and therapeutic implications. CNS Drugs1997;8:352-365.
  10. Frye M, Ketter K, Leverich G, et al. The increasing use of polypharmacotherapy for refractory mood disorders: 22 years of study. J Clin Psychiatry2000;61:9-15.
  11. Shelton M, Calabrese J. Current concepts in rapid cycling bipolar disorder. Curr Psychiatry Rep2000;2:310-315.
  12. Jawad S, Richens A, Goodwin G, et al. Controlled trial of lamotrigine (Lamictal) for refractory partial seizures. Epilepsia1989; 30:656-653.
  13. Smith D, Chadwick D, Baker G, et al. Seizure severity and the quality of life. Epilepsia1993;34[Suppl 5]:S31-S35.
  14. Calabrese J, Fatemi S, Woyshville M. Antidepressant effects of lamotrigine in rapid cycling bipolar disorder. Am J Psychiatry1996;153:1236.
  15. Fogelson D, Sternback H. Lamotrigine treatment of refractory bipolar disorder. J Clin Psychiatry1997;58:271-273.
  16. Kusumakar V, Yatham L. An open study of lamotrigine in refractory bipolar depression. Psychiatry Res1997;72:145-148.
  17. Sporn J, Sachs G. The anticonvulsant lamotrigine in treatment-resistant manic-depressive illness. J Clin Psychopharmacol1997; 17:185-189.
  18. Calabrese, J, Bowden C, McElroy S, et al. Spectrum of activity of lamotrigine in treatment refractory bipolar disorder. Am J Psychiatry1999;156:1019-1023.
  19. Calabrese, J, Bowden C, Sachs G, et al. A double-blind placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. J Clin Psychiatry1999;60:2:79-88.
  20. Calabrese J, Suppes T, Bowden C, et al. A double-blind, placebo-controlled, prophylaxis study of lamotrigine in rapid cycling bipolar disorder. J Clin Psychiatry2000;61:841-850.
  21. Rapport D, Calabrese J, Clegg K, et al. Lamotrigine in unipolar major depression. Prim Psychiatry1999;6(4):41-42.
  22. Maltese, T. Adjunctive lamotrigine treatment for major depression. Am J Psychiatry1999;156:1833.
  23. Frye M, Ketter T, Kimbrell, et al. A placebo-controlled evaluation of lamotrigine and gabapentin monotherapy in refractory mood disorders. J Clin Psychopharmacol2000;20:607-614.