Print Friendly 

Weight Gain in Bipolar Disorder:
Causes and Treatments

David Printz, MD, Joy Clark, MD, Laurie Stricks, PhD,
and Dolores Malaspina, MD

Primary Psychiatry. 2003;10(11):29-36

Dr. Printz is assistant clinical professor, Dr. Stricks is research scientist, and Dr. Malaspina is professor of psychiatry, all in the Department of Psychiatry at Columbia University in New York City.

Dr. Clark is a psychiatric resident at St. Luke’s/Roosevelt Medical Center in New York City.

Disclosure: Dr. Printz receives research support from GlaxoSmithKline and Pfizer, and is on the speaker’s bureau for AstraZeneca, GlaxoSmithKline, Ortho-McNeil and Pfizer. Dr. Malaspina is a consultant to Bristol-Myers Squibb and Janssen, and is on the speaker’s bureau for Wyeth.

Please direct all correspondence to: David Printz, MD, Bipolar Disorder Research Clinic, Columbia University, 1051 Riverside Drive, Unit 55, New York, NY 10032; Tel: 212-543-5944; Fax: 212-543-6728; E-mail:

Focus Points

Bipolar disorder patients are significantly more likely to be overweight or obese, thus placing them at an increased risk for obesity-related conditions, such as diabetes and hypertension.

Weight gain may be caused by mood stabilizers and/or bipolar disorder-related symptoms, such as increased appetite and decreased activity during depression.

In helping patients evaluate the relative risks and benefits of treatments, clinicians should consider the medical background of the patient, the liability of the medication to promote weight gain, and the ability of the patient to actively participate in pharmacologic and nonpharmacologic weight-reduction strategies.



Bipolar disorder is associated with a high incidence of significant obesity, much of which is likely iatrogenic and caused by mood-stabilizing medications. This obesity produces a significant psychosocial burden, increases the risk for a range of comorbid medical disorders, and leads to higher rates of medication noncompliance and clinical relapse. Fortunately, the range of mood-stabilizing agents is expanding and several of the newer agents have a lower tendency to promote weight gain. Also, a greater focus on medication-associated weight gain has increased the data available to aid in medication selection. This article summarizes epidemiologic data relating obesity to bipolar illness, describe the associations between specific mood stabilizers and weight gain, and provide practical approaches to the management of weight gain in bipolar patients.




Despite increasing awareness of the harmful effects of obesity, rates of obesity1 continue to climb in developed nations.2 The Third National Health and Nutrition Examination Survey found that the rates of being overweight or obese (body mass index [BMI] 25 [Table 1]) were 63% for males and 55% for females. Data derived from the survey3 demonstrated that type 2 diabetes, gallbladder disease, coronary heart disease, hypercholesterolemia, and hypertension all occurred at significantly higher rates in the overweight and obese subgroups of the population. It is important to remember that even the relatively modest excess weight found in the overweight group increased risk for these medical comorbidities. With the exception of hypercholesterolemia, progressively higher rates were seen with increasing obesity class. Indeed, obesity in the United States is credited with approximately 280,000 excess deaths per year.4


It is clearly incumbent upon clinicians to minimize the potential for iatrogenic weight gain induced by medications. This is particularly true for psychiatrists managing chronic psychotic and affective disorders, where symptoms of the illness (altered appetite, poor impulse control, decreased rates of exercise) may conspire with the effects of medications to produce tremendous weight gain in some patients. Since many psychiatric patients are less likely to adhere to regular medical evaluation by an internist, psychiatrists need to be aware of the causes and consequences of weight gain, monitor weight and metabolic parameters, pursue primary prevention of obesity when possible, and be aware of treatment options for obesity.


This review seeks to provide clinically useful information for the practicing psychiatrist by summarizing the literature relating bipolar disorder and obesity, comparing the weight gains associated with different mood stabilizers, and providing options for the treatment of obesity in bipolar patients.


Weight Gain Associated With Bipolar Disorder


Recent studies suggest that obesity occurs at higher rates in patients with bipolar disorder. Elmslie and colleagues5 compared bipolar patients to controls from a population database and found higher rates of being overweight in female patients relative to controls (44% versus 25%) and higher rates of obesity in female (20% versus 13%) and male (19% versus 10%) patients relative to controls. An elevated waist-to-hip ratio (indicative of central obesity and increased cardiovascular risk) was more common in patients relative to controls for both women (59% versus 17%) and men (58% versus 35%). Fagiolini and colleagues6 found that 68% of bipolar subjects were either overweight (36%) or obese (32%) and that BMI correlated with number of previous depressive episodes. McElroy and colleagues7 reviewed the data on 644 patients with bipolar disorder treated in a naturalistic fashion. They found that being overweight, obese, or extremely obese occurred in 58%, 21%, and 5% of the sample, respectively. Comorbid medical conditions that occurred at higher rates in the overweight or obese bipolar patients included hypertension, diabetes, and arthritis. Degree of exposure to psychotropic medications known to cause weight gain correlated positively with current BMI, while frequency of exercise correlated negatively. These data highlight the elevated rates of obesity and associated medical consequences in bipolar patients.


Mood Stabilizers and Weight Gain


Medications which are considered mood stabilizing in nature include lithium, a subset of antiepileptic drugs (AEDs), and the range of atypical antipsychotic medications. Not surprisingly, most of the data relating weight gain to AEDs and antipsychotics have been collected in patients with epilepsy and schizophrenia, respectively.




Lithium use has long been associated with weight gain. In their review of 12 early lithium studies in bipolar patients, Goodwin and Jamison8 reported that weight gain was a subjective complaint of approximately one in five patients. Weight gain was also cited as the second most important cause of medication noncompliance, which can in turn produce relapse.9 Many other studies have reported similar weight gain in bipolar patients treated with lithium.10-16 The reported average increases in body weight range from 4 kg11 to 6.3 kg,13 with the largest increases occurring in the first 2 years of treatment.11 In direct comparisons, average weight gain in patients treated with lithium is similar to those treated with valproate.13,15 In summary, recent data support earlier observations that lithium causes moderate weight gain in a substantial fraction of patients and marked weight gain in a smaller percentage.


Antiepileptic Drugs




Weight gain is a side effect commonly associated with valproate use. Dinesen and colleagues17 observed that 57% of 63 epilepsy patients treated with valproate gained more than 4 kg during treatment. In a study of 100 children with epilepsy treated with valproate,18 44% developed significant weight gain. Corman and colleagues19 retrospectively characterized the weight gain in 70 epilepsy patients treated with valproate for 27 months and found that 71% had gained at least 5% of baseline weight, compared to 43% of patients in a smaller carbamazepine control group. A 4 kg weight gain was seen in 70% of the valproate patients in that study. In a cross-sectional study of women with epilepsy treated with valproate or carbamazepine monotherapy for at least 2 years,20 mean BMI was significantly higher with valproate (24.4) than carbamazepine (22.9). Higher postprandial insulin levels were also observed in the valproate-treated patients, providing evidence of possible insulin resistance.


A similar association was found in studies by Isojarvi and colleagues,21 who observed higher rates of obesity (defined by BMI25) in a group of women with epilepsy naturalistically treated with valproate (59%) relative to those treated with carbamazepine (28%) or to controls (12%). In the subset of valproate patients who gained weight, the average weight gain was 21 kg. The valproate-treated group also had higher rates of insulin resistance, menstrual dysfunction, elevated androgen levels, and polycystic ovaries.22

A subset of these obese valproate-treated women were subsequently switched to lamotrigine and experienced significant weight loss, with resolution of associated endocrine and metabolic abnormalities. In a recent study, Pylvanen and colleagues23 compared valproate-treated epilepsy patients with controls matched by BMI. Obesity was present in 49% of subjects in both groups and leptin values were comparable. However, insulin levels were higher in both normal weight and obese valproate subjects relative to controls. Isojarvi and colleagues21 proposed that valproate-induced weight gain led to insulin resistance and hyperinsulinemia and thereby to increased androgen levels and polycystic ovaries. The finding by Pylvanen and colleagues23 provides further evidence of valproate-associated weight gain and its potential association to other medical problems.


There are relatively little data relating valproate to weight gain in bipolar patients. Valproate has been compared to olanzapine in two multicenter acute mania studies. In a 3-week double-blind, randomized monotherapy study,24 the mean increases in weight in the olanzapine and valproate groups were 2.5 and 0.9 kg, respectively. Weight gain was reported as an adverse event by 12% and 7.9% of subjects in the olanzapine and valproate groups, respectively. In a 12-week acute mania trial, Zajecka and colleagues25 reported significantly greater weight gain in an olanzapine group (4.0 kg) compared to a valproate group (2.5 kg). In that study, weight gain was identified as an adverse event in 25% of olanzapine patients and 10% of valproate patients. Body weight appeared to plateau at approximately 28 days for valproate and 42 days for olanzapine. Taken together, these studies suggest that valproate produces significant weight gain in a minority of bipolar patients, but that it produces less weight gain than olanzapine.



Carbamazepine has demonstrated efficacy in acute mania and maintenance treatment of bipolar disorder. Similar to valproate, carbamazepine is more effective for periods of mood elevation than depression. As noted in a series of studies comparing carbamazepine to lithium16 or valproate,20,21 weight gain is less severe with carbamazepine than with either of lithium or valproate.




Lamotrigine has demonstrated efficacy for bipolar depression and rapid-cycling bipolar disorder. Devinsky and colleagues26 evaluated the weight change in 463 subjects drawn from adjunctive lamotrigine epilepsy trials and found a median weight gain of only 0.1 kg after a mean duration of treatment for 318 days. In a 32-week double-blind, randomized trial of lamotrigine and valproate monotherapy in patients with epilepsy (n=133), Biton and colleagues27 reported that valproate produced a mean weight gain of 12.8 lbs, which had not yet reached a plateau at the end of the study. In contrast, weight gain in the lamotrigine group averaged only 1.3 lbs. In a randomized, double-blind comparison of lamotrigine and valproate for adolescents with epilepsy,28 lamotrigine produced no weight gain while valproate treatment resulted in significant weight gain by week 10, with a further increase over the subsequent 22 weeks.


The bipolar literature supports the minimal weight effect observed in the epilepsy literature. For example, in a 26-week, double-blind monotherapy maintenance study in rapid-cycling patients, Calabrese and colleagues29 observed a 1.1 kg weight gain in lamotrigine patients relative to a loss of 0.3 kg in placebo-treated patients. In contrast, in the lithium versus lamotrigine bipolar maintenance study previously cited,14 11% of lamotrigine-treated subjects gained 7% of body weight over 18 months. The reason for this discrepant finding is unclear. However, the literature suggests that there is minimal weight gain associated with lamotrigine.




Topiramate is a structurally novel AED approved in 1996 for the treatment of epilepsy. Although topiramate failed to separate from placebo in multicenter monotherapy trials for acute mania, several studies suggest antimanic efficacy when the drug is used as an adjunctive treatment.30-32


Appetite suppression and weight loss have been observed in trials using topiramate for both epilepsy and bipolar disorder. A mean loss of 9% of body weight was observed in a naturalistic study of epilepsy patients maintained on topiramate for an average of 2 years.33 In a prospective study, weight loss was evaluated in a sample of epilepsy patients taking adjunctive topiramate for at least 1 year. Significantly more weight was lost in the obese (BMI 30) subset of subjects relative to the entire sample at both 3 months (4.2 versus 3.0 kg) and 12 months (10.9 versus 5.9 kg) of treatment. Weight loss at 3 months correlated with reduced caloric intake, suggesting a direct effect of topiramate on appetite.


In a retrospective review34 of naturalistic treatment of bipolar patients, adjunctive use of topiramate was associated with weight loss in half of the patients. The mean weight change was 14.2 lbs and the mean dose of topiramate was higher in patients who lost weight (138 mg/day) than in those who did not lose weight (70 mg/day). In another adjunctive study in mania, weight loss was approximately 9.4 lbs over 5 weeks.31 In a comparison of topiramate and bupropion augmentation for bipolar depression, weight loss over 8 weeks was 5.8 kg and 1.2 kg, respectively.35 Marcotte36reviewed the data from 298 outpatients with bipolar disorder or cyclothymia treated openly with topiramate for up to 4 years. Overall, the mean weight loss was 12.5 lbs. Greater weight loss was associated with a longer duration of treatment; for patients treated <6 months, 6–12 months, and >12 months, the weight loss was 8.0, 13.8, and 17.3 lbs, respectively. Greater weight loss was also observed with higher doses of topiramate; mean dose was 212 mg/day for patients losing <20 lbs and 282 mg/day for those losing >20 lbs. Lastly, McElroy and colleagues37 conducted a 14-week, double blind, placebo-controlled trial of topiramate (mean dose 212 mg/day) in obese patients with binge-eating disorder. They observed a greater reduction in the topiramate group in binge frequency, binge-eating obsessions, BMI, and body weight. Mean weight loss was almost 6 kg in the topiramate group and 1 kg in the placebo group.


The potential benefits of topiramate for either mood stabilization or weight reduction must be balanced against the potential for troublesome side effects. For topiramate, side effects most often include parasthesias, somnolence, and cognitive difficulties (including memory, concentration, and word-finding difficulties). To some degree, these side effects can be reduced with more gradual titration of the medication.


In summary, weight loss has been associated with topiramate treatment in patients with a variety of disorders when used as either a monotherapy or an adjunctive treatment. The degree of weight loss appears dose dependent, is positively related to baseline weight, and is maintained or improved beyond at least 1 year of treatment.


Antipsychotic Medications


The association between early atypical antipsychotics and weight gain in patients with schizophrenia is well known. This extensive literature has been widely reviewed,38-40 and our discussion will therefore be limited to more recent findings. Comparison of weight gain liabilities across the atypical agents is facilitated by a meta-analysis41 of data from 81 published clinical trials. The predicted mean weight gains at 10 weeks of treatment derived from that analysis are listed in Table 2. Unfortunately, there was insufficient quetiapine data to include that medication in the analysis. Although providing important comparison data to assist in medication selection, this meta-analysis only estimates short-term weight gain.




Multiple studies support the efficacy of olanzapine in the acute and maintenance treatment of bipolar mania. However, olanzapine also produces significant weight gain. In a retrospective chart review42 of 121 schizophrenia patients undergoing naturalistic treatment, olanzapine use was associated with a significantly greater weekly weight increase (0.76 kg/wk) than was seen with typical antipsychotics (0.27 kg/wk), clozapine (0.22 kg/wk), or risperidone (0.15 kg/wk). Other naturalistic data43 from a large sample of schizophrenia patients receiving maintenance antipsychotic monotherapy treatment with haloperidol, olanzapine, or risperidone demonstrated weight gain with each agent of 2.8, 4.6, and 3.1 kg, respectively.


Clinical trials provide more controlled prospective data. In a 3-week, placebo-controlled monotherapy trial for acute mania,44 weight gain was reported in 11.4% of olanzapine patients versus 1.4% of placebo patients. Mean weight change was +1.65 kg for olanzapine and -0.44 kg for placebo. One long-term (mean=6.6 months) prospective open-label extension of an acute mania clinical trial noted a 6.64 kg mean weight gain.45 As mentioned in the above section on valproate, two olanzapine versus valproate acute mania studies also provide evidence for greater weight gain on olanzapine. Of note, the 10-week weight gain with olanzapine (4.0 kg) observed by Zajecka and colleagues25 is nearly identical to that predicted by the meta-analysis of Allison and colleagues.41 This suggests that any effects of diagnosis on weight gain are not as powerful as medication-specific effects. A recent analysis from extended clinical trials46 in schizophrenia suggest that weight gain plateaus by 39 weeks of treatment. With continued treatment of 3 years, mean weight gain with olanzapine was 6.3 kg and with haloperidol, the active comparator, it was 0.7 kg.


Clinical treatment would benefit from the ability to predict which patients would gain significant weight with olanzapine. A retrospective analysis47 of weight gain in 1,189 schizophrenia patients treated with olanzapine (mean=13.2 mg) for 1 year divided patients into rapid and non-rapid weight gain groups based upon whether they gained 7% or more of baseline body weight in the first 6 weeks of treatment. Fifteen percent of the sample fell into the rapid weight gain group, whose members gained 2% or more of body weight in the first 2 weeks (4–7 lbs) and who subsequently reached a plateau of approximately 20 lbs gained by 22 weeks. The non-rapid weight gain group, consisting of the other 85% of subjects, gained an average of 4 lbs over the course of the 52-week study. This study suggests that there may be a subgroup of patients at risk for severe weight gain on olanzapine who can be identified based upon rapid weight gain in the first 2 weeks of treatment. Data from olanzapine clinical trials suggests that weight gain is not dosage related.




Clozapine retains an important role as a treatment for severe, refractory mania or rapid cycling but has been associated with weight gain comparable to, or exceeding,48 that of olanzapine. There is also some indication that weight gain with clozapine may not plateau as early as with other atypicals, but may continue over several years of treatment.49 In a recent retrospective 5-year naturalistic study50 of weight gain and diabetes risk in patients with schizophrenia treated with clozapine, it was determined that weight increased by 1.2 lbs/month. The rate of weight gain decreased after 12 months but weight did not plateau until approximately month 46. No correlation between dose and weight gain was observed. The rate of new onset diabetes in this group was remarkably high (37% over 5 years) but did not correlate with weight gain. This finding, also observed in studies of weight gain and diabetes with other atypicals, suggests that different mechanisms are responsible for these two metabolic adverse events.




As estimated by Allison and colleagues,41 mean 10-week weight gain for risperidone is 2.1 kg. Additional data have supported this estimate as well. Wirshing and colleagues48 retrospectively evaluated weight gain in 92 schizophrenia patients and found that risperidone was associated with less weight gain (4.1 kg) than either clozapine (7.5 kg) or olanzapine (8.0 kg). The mean time to maximum weight gain for these three agents were 15.0, 24.9, and 21.2 weeks, respectively. Recently published data from a 1-year double-blind, placebo-controlled schizophrenia trial comparing haloperidol and risperidone51 found a mean weight change in the two groups of -0.73 kg and +2.3 kg. In patients with bipolar disorder, risperidone appears to produce modest weight gain. In a 3-week placebo-controlled multicenter trial of adjunctive risperidone for acute mania,52 weight gain was 1.7 kg for risperidone and 0.5 kg for placebo.




As recently reviewed by Nasrallah,53 quetiapine is associated with a moderate degree of weight gain (2.08 kg) after 5–6 weeks of treatment, which persists over time (≤12 months) without worsening. This weight change does not appear to be dose related. Based upon this recent data, as well as several studies cited by Allison and colleagues,41 it is reasonable to place quetiapine between risperidone and olanzapine in terms of weight-gain liability.




Overall, ziprasidone has been associated with minimal or no weight gain.41,54 No significant weight gain was observed in a short-term placebo-controlled acute mania study54,55 or in schizophrenia trials. In addition, a mean weight loss of 3.6 kg over 6 months of ziprasidone treatment was observed in 40 patients with mental retardation who had previously developed significant weight gain on other agents.56




A recent meta-analysis of aripiprazole use in patients with schizophrenia57 (n=1648) revealed a 0.7 kg weight gain over 4–6 weeks (versus 0.6 kg and 1.3 kg for haloperidol and risperidone, respectively). In a 26-week comparison of aripiprazole to olanzapine, weight change was -2.2 kg and +4.4, respectively. Based on this early data, aripiprazole appears to possess no weight-gain liability.


Practical Approach to Weight Gain in the Bipolar Patient


As the data illustrates, significant weight gain is associated with many of the medications we employ for the acute and chronic treatment of bipolar disorder. This data should be used to guide medication selection and factored into the risk:benefit analysis undertaken for any proposed treatment. However, there will be times when medications causing weight gain will need to be used due to a poor response to other medications, the need for polypharmacy, or avoidance of other adverse effects. Prior to instituting treatment, it is helpful to clarify the following issues in order to assist in medication selection:

•Patient’s personal and family liability to obesity and to obesity-related disorders (diabetes, heart disease, dyslipidemia)
•Any comorbid eating disorders (especially, binge-eating disorder)

•Baseline weight and BMI

•Baseline vital signs, thyroid function tests, fasting blood sugar, and fasting lipid panel

After the start of treatment, it is critical to monitor weight change closely, particularly for patients who are not as self-observant. Weight monitoring should become part of the regular routine in patients with chronic illnesses, such as bipolar disorder. Weight monitoring should be incorporated into treatment in a fashion analogous to checking blood levels and monitoring liver function tests and other lab values. For drugs with high weight-gain liability, particularly in patients with other risk factors or higher baseline weight, it makes sense to preempt further weight gain with nutritionist consultation, implementation of an appropriate diet, and a regular exercise program. To minimize frustration in both the patient and clinician, the difficulty of implementing such a program (particularly by someone with a chronic psychiatric illness) should be articulated and understood.


If weight gain occurs during the course of treatment, metabolic and endocrine parameters should be reassessed. Some assessments are medication specific (eg, checking for emergent hypothyroidism with lithium) and some are general (eg, monitoring for elevated fasting blood sugar, checking fasting lipids, following blood pressure more closely). In addition to the non-pharmacologic strategies listed above, a variety of medications can be employed to facilitate weight loss.




Topiramate is an AED with substantial data supporting its efficacy for weight loss. Topiramate appears to be a reasonable choice for a weight-reducing medication for bipolar patients due to its potential antimanic efficacy when used as an augmentation agent, the absence of risk for cycle acceleration, and the low potential for the development of tolerance or abuse. Based upon the data cited, the target dose for optimum weight loss is probably 150–250 mg/day. Introduction at a low dose and gradual titration may reduce side effects.




Sibutramine is hypothesized to decrease appetite and increase energy expenditure. A recent 48-week placebo-controlled trial was conducted in 1,102 obese adults. Mean weight loss of 4 kg occurred over a 4-week lead-in period, in line with previous short-term clinical trials. Over a subsequent 44-week maintenance period, additional weight loss with continuous or intermittent sibutramine treatment was 3.8 and 3.3 kg, respectively. Weight increased 0.2 kg in the placebo group. This study reaffirmed the efficacy of sibutramine in acute treatment and suggested that benefits are maintained after discontinuation (ie, there was no marked weight gain during placebo group maintenance) and are extended with continued dosing. Potential side effects include dry mouth, constipation, sweating, and headache. Since sibutramine is a potent serotonin and norepinephrine reuptake inhibitor (similar to venlafaxine) it likely has the potential to induce a switch into mania (which has been reported58) or cause cycle acceleration.




Orlistat is a reversible inhibitor of gastric and pancreatic lipases. It acts by blocking the conversion of triglycerides into free fatty acids and monoglycerides, thereby decreasing the rate of absorption of lipids from the intestine. In an 18-month, placebo-controlled trial59 of overweight patients, orlistat produced greater weight loss than placebo (6.5 versus 3.0% at baseline). Although lacking systemic effects, orlistat is associated with gastrointestinal side effects, must be taken three times a day, and requires compliance with a reduced fat diet.




Phentermine is a sympathomimetic drug, with central nervous system actions similar to dextroamphetamine and other stimulants. It is approved only for the short-term treatment of obesity. When combined with fenfluramine, phentermine was associated with increased rates of valvular heart disease and primary pulmonary hypertension, prompting removal from the market of this combination. The risk of cardiopulmonary disease with phentermine monotherapy cannot be entirely ruled out. Most weight loss occurs in the initial treatment period and there is the potential for tolerance or tachyphylaxis.


Although stimulants are frequently used in the treatment of bipolar disorder (particularly bipolar depression with hypersomnia and anergy), the high rates of comorbid substance abuse with the illness mandate careful evaluation of the potential phentermine patient for stimulant abuse liability.




Weight loss was identified as an adverse event in epilepsy clinical trials of zonisamide, a newer AED. A recent placebo-controlled trial in 60 obese (BMI=36.3) subjects60 documented 5.9 and 0.9 kg weight losses in the zonisamide and placebo groups, respectively. Total weight loss in zonisamide-treated completers was 9.2 kg. This represents promising data, with an effect on weight similar to that of topiramate. However, pending more extensive data, including studies of bipolar patients, it seems prudent to be cautious in its use. Side effects can include sedation, ataxia, dizziness, and difficulty with concentration and memory.




Bipolar disorder is complicated by rates of obesity and related medical conditions which are greater than in the general population. Based upon the studies cited, iatrogenic weight gain due to mood-stabilizing medication is likely responsible for a significant fraction of this added morbidity. There is a pronounced range of weight-gain liability across the spectrum of mood stabilizing agents (Table 3). Lithium, valproate, and olanzapine are associated with the greatest weight gain; carbamazepine, risperidone, and quetiapine with intermediate weight gain; and lamotrigine, ziprasidone, and aripiprazole with no significant weight gain. Topiramate is the only mood stabilizer that produces significant weight loss. It is incumbent upon the clinician to inform and educate the patient about weight gain, monitor weight and related endocrine and metabolic parameters closely, and be aware of pharmacologic and nonpharmacologic strategies to reduce obesity. PP



1. Grundy SM. Obesity, metabolic syndrome, and coronary atherosclerosis. Circulation. 2002;23:2696-2698.


2. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. Jama. 2003;1:76-79.


3. Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. The disease burden associated with overweight and obesity. Jama. 1999;16:1523-1529.


4. Allison DB, Fontaine KR, Manson JE, Stevens J, VanItallie TB. Annual deaths attributable to obesity in the United States. Jama. 1999;16:1530-1538.


5. Elmslie JL, Silverstone JT, Mann JI, Williams SM, Romans SE. Prevalence of overweight and obesity in bipolar patients. J Clin Psychiatry. 2000;3:179-184.


6. Fagiolini A, Frank E, Houck PR, et al. Prevalence of obesity and weight change during treatment in patients with bipolar I disorder. J Clin Psychiatry. 2002;6:528-533.


7. McElroy SL, Frye MA, Suppes T, et al. Correlates of overweight and obesity in 644 patients with bipolar disorder. J Clin Psychiatry. 2002;3:207-213.


8. Goodwin FK, Jamison KR. Manic-Depressive Illness. New York, NY: Oxford University Press; 1990.


9. Johnson RE, McFarland BH. Lithium use and discontinuation in a health maintenance organization. Am J Psychiatry. 1996;8:993-1000.


10. Chen Y, Silverstone T. Lithium and weight gain. Int Clin Psychopharmacol. 1990;3:217-225.


11.Vestergaard P, Poulstrup I, Schou M. Prospective studies on a lithium cohort. 3. Tremor, weight gain, diarrhea, psychological complaints. Acta Psychiatr Scand. 1988;4:434-441.


12. Henry C. Lithium side-effects and predictors of hypothyroidism in patients with bipolar disorder: sex differences. J Psychiatry Neurosci. 2002;2:104-107.


13. Chengappa KN, Chalasani L, Brar JS, Parepally H, Houck P, Levine J. Changes in body weight and body mass index among psychiatric patients receiving lithium, valproate, or topiramate: an open-label, nonrandomized chart review. Clin Ther. 2002;10:1576-1584.


14. Bowden CL, Calabrese JR, Sachs G, et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently manic or hypomanic patients with bipolar I disorder. Arch Gen Psychiatry. 2003;4:392-400.


15. Bowden CL, Calabrese JR, McElroy SL, et al. A randomized, placebo-controlled 12-month trial of divalproex and lithium in treatment of outpatients with bipolar I disorder. Divalproex Maintenance Study Group. Arch Gen Psychiatry. 2000;5:481-489.


16. Coxhead N, Silverstone T, Cookson J. Carbamazepine versus lithium in the prophylaxis of bipolar affective disorder. Acta Psychiatr Scand. 1992;2:114-118.


17. Dinesen H, Gram L, Andersen T, Dam M. Weight gain during treatment with valproate. Acta Neurol Scand. 1984;2:65-69.


18. Egger J, Brett EM. Effects of sodium valproate in 100 children with special reference to weight. Br Med J. 1981;6291:577-581.


19. Corman CL, Leung NM, Guberman AH. Weight gain in epileptic patients during treatment with valproic acid: a retrospective study. Can J Neurol Sci. 1997;3:240-244.


20. Luef G, Abraham I, Haslinger M, et al. Polycystic ovaries, obesity and insulin resistance in women with epilepsy. A comparative study of carbamazepine, and valproic acid in 105 women. J Neurol. 2002;7:835-841.


21. Isojarvi JI, Laatikainen TJ, Knip M, Pakarinen AJ, Juntunen KT, Myllyla VV. Obesity and endocrine disorders in women taking valproate for epilepsy. Ann Neurol. 1996;5:579-584.


22. Isojarvi JI, Laatikainen TJ, Pakarinen AJ, Juntunen KT, Myllyla VV. Polycystic ovaries and hyperandrogenism in women taking valproate for epilepsy. N Engl J Med. 1993;19:1383-1388.


23. Pylvanen V, Knip M, Pakarinen A, Kotila M, Turkka J, Isojarvi JI. Serum insulin and leptin levels in valproate-associated obesity. Epilepsia. 2002;5:514-517.


24.Tohen M, Baker RW, Altshuler LL, et al. Olanzapine versus divalproex in the treatment of acute mania. Am J Psychiatry. 2002;6:1011-1017.


25. Zajecka JM, Weisler R, Sachs G, Swann AC, Wozniak P, Sommerville KW. A comparison of the efficacy, safety, and tolerability of divalproex sodium and olanzapine in the treatment of bipolar disorder. J Clin Psychiatry. 2002;12:1148-1155.


26. Devinsky O, Vuong A, Hammer A, Barrett PS. Stable weight during lamotrigine therapy: a review of 32 studies. Neurology. 2000;4:973-975.


27. Biton V, Mirza W, Montouris G, Vuong A, Hammer AE, Barrett PS. Weight change associated with valproate and lamotrigine monotherapy in patients with epilepsy. Neurology. 2001;2:172-177.


28. Biton V, Levisohn P, Hoyler S, Vuong A, Hammer AE. Lamotrigine versus valproate monotherapy-associated weight change in adolescents with epilepsy: results from a post hoc analysis of a randomized, double-blind clinical trial. J Child Neurol. 2003;2:133-139.


29. Calabrese JR, Suppes T, Bowden CL, et al. A double-blind, placebo-controlled, prophylaxis study of lamotrigine in rapid-cycling bipolar disorder. Lamictal 614 Study Group. J Clin Psychiatry. 2000;11:841-850.


30. McElroy SL, Suppes T, Keck PE, et al. Open-label adjunctive topiramate in the treatment of bipolar disorders. Biol Psychiatry. 2000;12:1025-1033.


31.Chengappa KN, Rathore D, Levine J, et al. Topiramate as add-on treatment for patients with bipolar mania. Bipolar Disord. 1999;1:42-53.


32. Suppes T. Review of the use of topiramate for treatment of bipolar disorders. J Clin Psychopharmacol. 2002;6:599-609.


33. Skavoy T, Brodtkorb E. Weight loss in epileptic patients treated with topiramate. Acta Neurol Scand. 2003;6:428-429.


34. Ghaemi SN, Manwani SG, Katzow JJ, Ko JY, Goodwin FK. Topiramate treatment of bipolar spectrum disorders: a retrospective chart review. Ann Clin Psychiatry. 2001;4:185-189.


35. McIntyre RS, Mancini DA, McCann S, Srinivasan J, Sagman D, Kennedy SH. Topiramate versus bupropion SR when added to mood stabilizer therapy for the depressive phase of bipolar disorder: a preliminary single-blind study. Bipolar Disord. 2002;3:207-213.


36. Marcotte DB. Longer term treatment with topiramate for bipolar disorders. Bipolar Disord. 2001;3(suppl 1):46-47.


37. McElroy SL, Arnold LM, Shapira NA, et al. Topiramate in the treatment of binge eating disorder associated with obesity: a randomized, placebo-controlled trial. Am J Psychiatry. 2003;2:255-261.


38. Allison DB, Casey DE. Antipsychotic-induced weight gain: a review of the literature. J Clin Psychiatry. 2001;62(suppl 7):22-31.


39. Ganguli R. Weight gain associated with antipsychotic drugs. J Clin Psychiatry. 1999;60(suppl 21):20-24.


40. Russell JM, Mackell JA. Bodyweight gain associated with atypical antipsychotics: epidemiology and therapeutic implications. CNS Drugs. 2001;7:537-551.


41. Allison DB, Mentore JL, Heo M, et al. Antipsychotic-induced weight gain: a comprehensive research synthesis. Am J Psychiatry. 1999;11:1686-1696.


42. Simpson MM, Goetz RR, Devlin MJ, Goetz SA, Walsh BT. Weight gain and antipsychotic medication: differences between antipsychotic-free and treatment periods. J Clin Psychiatry. 2001;9:694-700.


43. Bobes J, Rejas J, Garcia-Garcia M, et al. Weight gain in patients with schizophrenia treated with risperidone, olanzapine, quetiapine, or haloperidol: results of the EIRE study. Schizophr Res. 2003;1-2:77-88.


44. Tohen M, Sanger TM, McElroy SL, et al. Olanzapine versus placebo in the treatment of acute mania. Olanzapine HGEH Study Group. Am J Psychiatry. 1999;5:702-709.


45. Sanger TM, Grundy SL, Gibson PJ, Namjoshi MA, Greaney MG, Tohen MF. Long-term olanzapine therapy in the treatment of bipolar I disorder: an open-label continuation phase study. J Clin Psychiatry. 2001;4:273-281.


46. Kinon BJ, Basson BR, Gilmore JA, Tollefson GD. Long-term olanzapine treatment: weight change and weight-related health factors in schizophrenia. J Clin Psychiatry. 2001;2:92-100.


47. Jaton L, Kinon BJ, Rotelli M, Kollack-Walker S, Kaiser C. Differential rate of weight gain present among patients treated with olanzapine. Schizophr Res. 2003;60(suppl):357.


48.Wirshing DA, Wirshing WC, Kysar L, et al. Novel antipsychotics: comparison of weight gain liabilities. J Clin Psychiatry. 1999;6:358-363.


49. Umbricht D, Pollack S, Kane JM. Clozapine and weight gain. J Clin Psychiatry. 1994;9(suppl B):157-160.


50. Henderson DC, Cagliero E, Gray C, et al. Clozapine, diabetes mellitus, weight gain, and lipid abnormalities: A five-year naturalistic study. Am J Psychiatry. 2000;6:975-981.


51. Csernansky JG, Mahmoud R, Brenner R. A comparison of risperidone and haloperidol for the prevention of relapse in patients with schizophrenia. N Engl J Med. 2002;1:16-22.


52. Yatham LN, Grossman F, Augustyns I, Vieta E, Ravindran A. Mood stabilisers plus risperidone or placebo in the treatment of acute mania. International, double-blind, randomised controlled trial. Br J Psychiatry. 2003;182:141-147.


53. Nasrallah H. A review of the effect of atypical antipsychotics on weight. Psychoneuroendocrinology. 2003;28(suppl 1):83-96.


54. Keck P Jr., Buffenstein A, Ferguson J, et al. Ziprasidone 40 and 120 mg/day in the acute exacerbation of schizophrenia and schizoaffective disorder: a 4-week placebo-controlled trial. Psychopharmacology (Berl). 1998;2:173-184.


55. Keck PE Jr., Versiani M, Potkin S, West SA, Giller E, Ice K. Ziprasidone in the treatment of acute bipolar mania: a three-week, placebo-controlled, double-blind, randomized trial. Am J Psychiatry. 2003;4:741-748.


56. Cohen S, Fitzgerald B, Okos A, Khan S, Khan A. Weight, lipids, glucose, and behavioral measures with ziprasidone treatment in a population with mental retardation. J Clin Psychiatry. 2003;1:60-62.


57. Jody D, Carson WH, Iwamoto T, et al. Meta-analysis of weight effects with aripiprazole. Schizophrenia Res. 2003;1:358.


58. Cordeiro Q, Vallada H. Sibutramine-induced mania episode in a bipolar patient. Int J Neuropsychopharmacol. 2002;3:283-284.


59. Krempf M, Louvet JP, Allanic H, Miloradovich T, Joubert JM, Attali JR. Weight reduction and long-term maintenance after 18 months treatment with orlistat for obesity. Int J Obes Relat Metab Disord. 2003;5:591-597.


60. Gadde KM, Franciscy DM, Wagner HR 2nd, Krishnan KR. Zonisamide for weight loss in obese adults: a randomized-controlled trial. JAMA. 2003;14:1820-1825.


Print Friendly 

Diagnosis and Management
of Neuropathic Pain States

Bruce D. Nicholson, MD

Dr. Nicholson is a pain specialist in private practice in Allentown, Pennsylvania.

Disclosure: Dr. Nicholson is a consultant to Endo, Flanding, and Pfizer; is on the speaker’s bureau of Endo, Flanding, Ortho-McNeil, and Pfizer; and receives research grants from Endo, Flanding, and Pfizer.

Please direct all correspondence to: Bruce D. Nicholson, MD, Pain Specialist of Greater Lehigh Valley, 1240 South Cedar Crest Blvd, Suite 307, Allentown, PA 18105; Tel: 610-402-1754; Fax: 610-402-1747; E-mail:

Focus Points

Neuropathic pain is a symptom of peripheral or central neurological pathology.

Neuropathic pain has a prevalence of approximately 1% in the United States and its prevalence increases with age.

Common comorbidities of neuropathic pain include sleep disturbances and depression.

Antiepileptic drugs have shown efficacy in the treatment of neuropathic pain.


Neuropathic pain affects approximately 1% of the population and is most common in those with diabetes. Although knowledge about its pathobiology and molecular biology has increased, the clinical treatment of neuropathic pain remains difficult. The efficacy of opioids in the treatment of neuropathic pain is controversial and side effects limit their use. Other pharmacologic classes that have efficacy in the management of neuropathic pain include tricyclic antidepressants, selective norepinephrine reuptake inhibitors, and antiepileptics. This review discusses the etiology of neuropathic pain and presents evidence for its treatment with drugs from different pharmacologic classes.


Neuropathic pain is the result of neurological dysfunction usually caused by disease, trauma, or injury. Common conditions include diabetic neuropathy, postherpetic neuralgia, trigeminal neuralgia, radiation- and chemotherapy-induced neuropathies, chronic radiculopathies, and idiopathic peripheral neuropathies.1,2 The prevalence of neuropathic pain has been estimated at approximately 1% of the population and its prevalence increases with age.3 Diabetes is the leading cause of neuropathic pain in the United States, where approximately 11 million people have been diagnosed with diabetes mellitus, and and 6 million remain undiagnosed.4 It is estimated that 50% to 80% of diabetics will develop some form of neuropathy, and that 30% to 40% of these patients will suffer painful diabetic neuropathy (PDN). Obesity is a major risk factor for the development of diabetes, and as obesity becomes an increasing problem both in the US and worldwide, the incidence of diabetes and PDN is expected to increase.

Although the knowledge regarding the pathobiology and molecular biology of neuropathic pain has increased, its clinical management has been difficult. Neuropathic pain is usually chronic and responds poorly to standard analgesic pharmacotherapy, such as nonsteroidal antiinflammatory drugs (NSAIDs).5 Opioids have been used with some success6 but side effects have limited their use. Other classes of medications that have been used include tricyclic antidepressants (TCAs), serotonin and norepinephrine reuptake inhibitors (SNRIs), antidepressants, and antiepileptic drugs (AEDs).5 This article discusses the etiology of neuropathic pain and presents evidence supporting the use of various pharmacologic classes in its management.

Pain:Definition and Classification

Pain can be broadly classified into two diagnostic categories: nociceptive and neuropathic.

Nociceptive Pain

Nociceptive pain is most often an appropriate physiologic response that occurs when specific neurons called nociceptors are exposed to a noxious stimulus, such as tissue damage during surgery or trauma. The physiological response to tissue damage is nociceptive pain, which involves activation of the pain-specific A-d and C-fiber neurons, which in turn transmit a stimulus to the spinal cord dorsal horn.5 From the dorsal horn the stimulus undergoes modulation and subsequent transmission to the brain where pain is perceived as a protective mechanism. Nociceptive pain has a positive, protective role in that it signals the potential for tissue damage and alerts the person for the need to prevent further injury.

Neuropathic Pain

Neuropathic pain is initiated or caused by a primary lesion or dysfunction in the nervous system,7 and is an inappropriate pain response. Neuropathic pain may be stimulus-induced, but unlike nociceptive pain, ongoing tissue injury or nociceptive stimulation is not required for its initiation. Whereas nociceptive pain serves a protective function, neuropathic pain serves no useful purpose and is simply a pathologic condition. The precipitating neural injury may arise from peripheral or central nervous system trauma as a result of infection (postherpetic neuralgia, human immunodeficiency virus [HIV]-associated neuralgia), malignancy (neuropathy due to chemotherapy, nerve compression, or invasion), metabolic disturbances (diabetic neuropathy), infarct (stroke), nerve compression or entrapment (spinal stenosis, radiculopathy), or idiopathically.

Although neuropathic pain is a symptom of an underlying pathology, it may also be considered a disease in that it is a manifestation of a pathologic dysfunction within the nervous system. Over the past decade, animal models have expanded our knowledge of both acute and chronic pain states. Several important pathophysiologic mechanisms involved in the initiation and perpetuation of neuropathic pain have been identified, many of which include injury to the peripheral as well as central nervous system. Specific examples of these abnormalities include altered sodium (Na+) channel function at the site of nerve injury (neuroma) and altered Na+ and calcium (Ca++)channel function within the central nervous system.8 Changes in glutaminergic subtypes of α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid and N-methyl-D-aspartate (NMDA) receptors are also known to be involved in the development and maintenance of peripheral and central pain sensitization, which play an important role in the pathology of persistent pain.9 These changes within the nervous system lead to the development of the pathologic conditions referred to as hyperalgesia and allodynia.5

Allodynia is pain due to a stimulus that is not normally considered painful. A patient with allodynia may complain of pain with the lightest touch of clothing moving over the area of involvement. Various anatomical changes due to injury of nociceptive (C-fibers and A-d fibers) as well as non-nociceptive (A-β mechanoreceptors) neurons may lead to the development of allodynia. Hyperalgia describes a heightened response to a painful stimulus and is a consequence of peripheral and central sensitization. Loss of interneurons at the spinal cord dorsal level may lead to a decrease in g-aminobutyric acid (GABA) levels. The injury to GABA reuptake mechanisms can result in the phenomenon of disinhibition manifested by explosive pain.5

Clinically, patients with neuropathic pain may present with diverse symptoms characterized by tingling, numbness, as well as dull or throbbing burning pain, or electrical shock/lancinating pain. The symptoms related to neuropathic pain are typically worse at night (eg, painful diabetic neuropathy) and may significantly affect sleep. At time of examination, the most common signs of neuropathic pain include hyperalgesia and allodynia.

Adding to the complexity of the patient with neuropathic pain is its association with comorbid conditions, such as anxiety, depression, and sleep disruption.10 Disturbed sleep patterns are found in a majority of chronic pain patients and are thought to create a cycle that further exacerbates chronic pain behavior.11 As a consequence, sleep deprivation may become a significant problem, which in turn may contribute to anxiety, stress, and depression.12 Chronic pain (pain that lasts >3 months) may lead to symptoms of anxiety or depression in ≤60% of patients.13

Insight into the mechanisms that sustain the symptoms and signs of neuropathic pain, beginning from the peripheral level leading to the spinal cord and brain, may well begin to improve clinical outcomes. Currently, the goals for the management of neuropathic pain are focused on alleviating the patient’s pain and if possible, preventing further injury to the nervous system.14 Current approaches that not only attenuate the underlying symptoms and signs of a neuropathic pain, but also focus on improving comorbid conditions, such as restoring normal sleep and addressing anxiety and depression, represent desirable therapeutic strategies.

Treatment of Neuropathic Pain

Neuropathic pain can be severe and debilitating and is generally resistant to traditional analgesic pharmacotherapy such as NSAIDs, COX-2 inhibitors, and acetaminophen. Although some patients may respond favorably to opioids,6 concerns regarding side effects, dependence, and the lack of evidence of long-term efficacy have limited their use. Traditional medical models for treatment of neuropathic pain are based on the use of antidepressants and AEDs. TCAs (eg, amitriptyline, nortriptyline, and desipramine) are typically used for burning pain. When appropriate, first-generation AEDs have been utilized with some success, eg, carbamazepine, which previously was considered the drug of choice for lancinating pain. Numerous case reports, retrospective reviews, and randomized-controlled trials have recently focused on the use of the second-generation AEDs in the treatment of neuropathic pain. Newer AEDs may well represent a more effective and better tolerated pharmacologic approach to the treatment of neuropathic pain.15

Tricyclic Antidepressants

TCAs, which include amitriptyline, nortriptyline, and desipramine, have been considered first-line drugs for the treatment of neuropathic pain.16 The mechanism of action of TCAs include increasing postsynaptic concentration of norepinephrine and serotonin (for the tertiary amines) as well as sodium-channel blockade. However, relatively few randomized, placebo-controlled trials have been undertaken to support the use of TCAs for the treatment of neuropathic pain syndromes. In most cases, no parameters other than pain relief, eg, the fraction of patients reporting a 50% improvement in pain response, were measured. Little if any information concerning comorbid conditions and quality of life data in these studies has been obtained or reported.

TCAs appear to be more effective for pain control when there is a balance between noradrenergic and serotonergic effects, as is observed with the tertiary amines, such as amitriptyline, imipramine, and clomipramine.17 However, desipramine and nortriptyline secondary amines have been shown to provide relief from PDN and postherpetic neuropathic (PHN) pain.18 Amitriptyline, the “gold standard” for treating neuropathic pain, has proven efficacy against a number of pain conditions, including PDN and PHN. The dose of amitriptyline, as well as most other TCAs, is normally 10–100 mg/day QHS, with a titration of 10 mg/week. Max and colleagues16 have demonstrated an analgesic effect that is independent of the antidepressive effect. Imipramine, the only other tertiary amine tested in patients with PDN, demonstrated a 50% improvement in pain relief and was more efficacious than amitriptyline.19 Unfortunately, with even the most effective drugs, up to 50% of patients with moderate to severe pain may have only a partial response to initial single drug therapy. With currently available agents for the treatment of neuropathic pain, only 30% of patients will demonstrate a 50% or greater improvement in visual analog scale ratings with single drug therapy.18,19

Clinically, a significant side-effect profile of amitriptyline includes sedation, anticholinergic effects (constipation, pseudodementia), hypotension, cardiac effects, and seizures. This profile may have significant ramifications, particularly when certain types of neuropathies are present. Patients with PHN or PDN are often elderly and may have compromised autonomic nervous systems. Many elderly diabetic patients do not tolerate amitriptyline for long-term use because of constipation and dry mouth, making compliance an important issue. In the elderly population, cardiac effects, particularly dysrhythmias, must be taken into account when using tertiary amines, such as amitriptyline.20

Antiepileptic Drugs

Clinical trials focused on neuropathic pain syndromes have demonstrated the efficacy of several AEDs.21 This pharmacologic group may be divided into first-generation drugs, including carbamazepine, valproic acid/divalproex sodium, phenytoin, and clonazepam, and second generation drugs, including gabapentin, topiramate, lamotrigine, oxcarbazepine, levetiracetam, tiagabine, and zonisamide.21-23


Carbamazepine has been the drug of choice for the treatment of trigeminal neuralgia for >30 years. Several trials have demonstrated improvement in the lancinating component of the neuropathic pain syndrome.24 However, it has limited efficacy in the overall treatment of PDN as well as PHN. Unfortunately, this general lack of efficacy, along with a significant side-effect profile, has limited the overall use of carbamazepine. Its potential side effects include rash, leukopenia, thrombocytopenia, hyponatremia, and hepatic enzyme induction, which may influence the action of other drugs. Recent trials using oxcarbazepine, a second-generation AED, have demonstrated efficacy in the treatment of trigeminal neuralgia.25 A major benefit associated with the use of oxcarbazepine is the lower incidence of side effects when compared to carbamazepine.


Overall, second-generation AEDs have demonstrated improved efficacy with lower side-effect profiles than first-generation AEDs. In placebo-controlled trials, gabapentin has demonstrated efficacy for the treatment of PDN and PHN.26 Gabapentin, which increases whole brain GABA levels and blocks the 2 subunit of voltage-sensitive calcium channels, received Food and Drug Administration approval in 2002 for the treatment of PHN in doses of <1,800 mg/day TID.27 Data analysis has demonstrated positive benefits relating to comorbid conditions, such as sleep and anxiety.28 Common side effects may include somnolence and dizziness upon initiation of therapy.29


The results of a small, double-blind study of 27 patients suggested that topiramate might be effective in the treatment of PDN.30 However, the results of large clinical trials have been inconsistent. In three randomized, placebo-controlled clinical trials involving 1,259 patients, no significant effect of topiramate was observed for the primary measure of pain visual analog (PVA) compared with placebo.31 Vinik and colleagues32 conducted a double-blind, placebo-controlled study of 323 PDN patients. They found that 36% of patients treated with topiramate at doses up to 400 mg/day for 12 weeks had a greater than 50% reduction in the PVA scale score, compared to only 21% in the placebo group (P=.005).

The dosing of topiramate, which is typically given on a BID schedule, is quite variable, ranging from a total daily dose of 50–100 mg for migraine headache therapy to 200–300 for the treatment of PDN.32 Common side effects with titration (25 mg/week) include paresthesias, cognitive slowing, and weight loss.33 A recent study of topiramate for migraine prevention demonstrated weight loss greater than 2 body mass index units over 6–9 months of treatment.34


Lamotrigine therapy for painful HIV neuropathy, PHN, trigeminal neuralgia, and central pain of post-stroke origin has been proven effective in prospective randomized studies.35,36 Early studies were negative at doses of 200 mg/day and lower for PDN, but recent data indicate efficacy at doses of 200–400 mg/day in divided doses. Significant side effects include rash with rapid titration greater than 25 mg/week, and rare occurrences of Stevens-Johnson syndrome.

Other Drug Classes


Selective Serotonin Reuptake Inhibitors

Selective serotonin reuptake inhibitors (SSRIs) are not particularly effective treatments for pain, and only two trials have demonstrated minimal efficacy in painful peripheral neuropathy conditions.19 The lack of efficacy with SSRIs is most likely a result of the predominate effect of this drug class on serotonin.

Serotonin Norepinephrine Reuptake Inhibitors

Early reports with venlafaxine, a combined SNRI, have been encouraging. In an open-label study of 11 patients with PDN, venlafaxine reduced pain by 75% to 100% in all patients.37 Venlafaxine has a better side-effect profile than first-generation TCAs. The importance of this class of antidepressants is yet to be determined but it deserves future consideration as part of combination therapy for neuropathic pain.

Sodium Channel Blockers

Other drugs that have been used to control pain include the sodium channel blockers lidocaine (intravenous or topical) and mexiletine (oral). Intravenous lidocaine was reported to be a useful treatment for central pain syndromes and spinal cord injury pain, but is not a practical therapy in ambulatory patients. Mexiletine demonstrated efficacy against PDN in two randomized trials. Doses as high as 400–675 mg were needed to achieve efficacy, but were associated with a significant adverse-event profile, including dyspepsia.38

Several formulations of topical lidocaine, for instance the eutectic mixture or EMLA (a combination of prilocaine and lidocaine), have been shown to be effective in PHN and PDN. The 5% lidocaine patch was approved by the FDA in 2001 for the treatment of PHN. The ability to block sodium channels allows topical lidocaine to be effective in patients with allodynia secondary to sodium channel disruption and abnormal function in peripheral neurons.39


Capsaicin is an interesting compound that has analgesic activity by virtue of its effect on VNr1 receptors. Variable results were obtained using this agent in PDN clinical trials.40,41 Capsaicin is effective as a treatment for PHN when applied topically four times/day for 4 weeks. This dosing schedule has been shown to alleviate posttraumatic or mastectomy-related pain. However, compliance is a major issue, particularly in patients who have PHN with a burning, allodynic component.42,43


Dextromethorphan, present in some cough syrups, is a weak NMDA antagonist. Good results have been obtained with this drug in trials involving patients with PDN. Unfortunately, dextromethorphan is not available in a formulation appropriate to deliver the 300–600 mg/day required unless capsules are made up by a compounding pharmacist. It produced no benefit over placebo in trials investigating PHN, indicating that different mechanisms are involved in PDN and PHN.44


Opioids are m-receptor agonists and there has been much debate as to their effectiveness for neuropathic pain control. Studies with animal models show that resistance to m-receptor function develops shortly after nerve injury, with downregulation of sensitivity to opioids. Few prospective, placebo-controlled trials have been conducted with these agents in oral formulations. Controlled-release oxycodone was shown to be an effective treatment for PHN in two trials and more recently for PDN.45 Up to 30 mg BID were needed before improvement was seen, which may have contributed to 76% of patients reporting at least one opiate-related adverse event. A recent double-blind, dose-response study of levorphanol in patients whose neuropathic pain was refractory to treatment showed a better analgesic response in the high-dose group compared with the low-dose group.6

Tramadol is a weak m-receptor agonist that also has SNRI properties. In a randomized, double-blind, placebo-controlled trial, tramadol was proven effective when given at 200–400 mg/day for 4 weeks for the treatment of peripheral neuropathy.46 The investigators concluded that both the m-receptor antagonism and the SNRI action of tramadol were required for this effect.


Neuropathic pain can be severe and debilitating, and its clinical management remains difficult. TCAs have been considered the first-line treatment of neuropathic pain, but <50% of patients show a significant improvement and side effects limit their utility in the elderly. Carbamazepine has a long history of use in trigeminal neuralgia but has limited efficacy in diabetic neuropathy and other neuropathic pain conditions. Recent advances in AED pharmacotherapy have yielded significant benefits in the treatment of neuropathic pain conditions. For example, newer generation AEDs, such as gabapentin and topiramate, have demonstrated efficacy in the treatment of PDN or PHN. More effective treatment of neuropathic pain will be possible when knowledge of pathophysiology is combined with new drug therapy. Increased clinician understanding of the individual patient’s symptoms will help to improve treatment as well.47 PP


1. Katz N. Neuropathic pain in cancer and AIDS. Clin J Pain. 2000;16(suppl 2):S41-S48.

2. Leland JY. Chronic pain: primary care treatment of the older patient. Geriatrics. 1999;1:23-37.

3. Davis MP, Srivastava M. Demographics, assessment, and management of pain in the elderly. Drugs Aging. 2003;1:23-57.

4. American Diabetes Association: Diabetes Statistics. Available at: Accessed June 13, 2003.

5. Woolf CJ, Mannion RJ. Neuropathic pain: aetiology, symptoms, mechanisms, and management. Lancet. 1999;9168:1959-1964.

6. Rowbotham MC, Twilling L, Davies PS, Reisner L, Taylor K, Mohr D. Oral opioid therapy for chronic peripheral and central neuropathic pain. N Engl J Med. 2003;13:1223-1232.

7. Merskey H, Bogduk N. Classification of Chronic Pain. Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms. 2nd ed. Seattle, WA: IASP Press; 1994.

8. Chaplan SR. Neuropathic pain: role of voltage-dependent calcium channels. Reg Anesth Pain Med. 2000;3:283-285.

9. Nicholson B. Gabapentin use in neuropathic pain syndromes. Acta Neurol Scand. 2000;6:359-371.

10. Atkinson JH, Ancoli-Israel S, Slater MA, Garfin SR, Gillin JC. Subjective sleep disturbance in chronic back pain. Clin J Pain. 1988;4:225-232.

11. Moldofsky H, Scarisbrick P, England R, Smythe H. Musculosketal symptoms and non-REM sleep disturbance in patients with “fibrositis syndrome” and healthy subjects. Psychosom Med. 1975;4:341-351.

12. Johnson LC. Physiological and psychological changes following total sleep deprivation. In: Kales A, ed. Sleep Physiology and Pathology. Philadelphia, PA: J.B. Lippincott; 1969:206-220.

13. Tunks E. Comorbidity of psychiatric disorder and chronic pain. In: Campbell J, ed. Pain 1996 -An Updated Review. IASP Committee of Refresher Courses. Seattle, WA: IASP Press; 1996:287-296.

14. MacFarlane BV, Wright A, O’Callaghan J, Benson HA. Chronic neuropathic pain and its control by drugs. Pharmacol Ther. 1997;1:1-19.

15. Bennett GJ. Neuropathic pain: new insights, new interventions. Hosp Pract (Off Ed). 1998;10:95-107.

16. Max MB, Lynch SA, Muir J, Shoaf SE, Smoller B, Dubner R. Effects of desipramine, amitriptyline, and fluoxetine on pain in diabetic neuropathy. N Engl J Med. 1992;19:1250-1256.

17. Ross EL. The evolving role of antiepileptic drugs in treating neuropathic pain. Neurology. 2000;5(suppl 1):S41-S46.

18. McQuay HJ, Tramer M, Nye BA, Carroll D, Wiffen PJ, Moore RA. A systematic review of antidepressants in neuropathic pain. Pain. 1996;68(2-3):217-227.

19. Sindrup SH, Jensen TS. Efficacy of pharmacological treatments of neuropathic pain: an update and effect related to mechanism of drug action. Pain. 1999;3:389-400.

20. Beers MH. Explicit criteria for determining potentially inappropriate medication use by the elderly. An update. Arch Intern Med. 1997;14:1531-1536.

21. Backonja MM. Use of anticonvulsants for treatment of neuropathic pain. Neurology. 2002;5(suppl 2):S14-S17.

22. Collins SL, Moore RA, McQuayHJ, Wiffen P. Antidepressants and anticonvulsants for diabetic neuropathy and postherpetic neuralgia: a quantitative systematic review. J Pain Symptom Manage. 2000;6:449-458.

23. Chong MS, Libretto SE. The rationale and use of topiramate for treating neuropathic pain. Clin J Pain. 2003;1:59-68.

24. Campbell FG, Graham JG, Zilkha KJ. Clinical trial of carbazepine (tegretol) in trigeminal neuralgia. J Neurol Neurosurg Psychiatry. 1966;3:265-267.

25. Zakrzewska JM, Patsalos PN. Long-term cohort study comparing medical (oxcarbazepine) and surgical management of intractable trigeminal neuralgia. Pain. 2002;3:259-266.

26. Backonja M, Beydoun A, Edwards KR, et al. Gabapentin for the symptomatic treatment of painful neuropathy in patients with diabetes mellitus: a randomized controlled trial. JAMA. 1998;21:1831-1836.

27. Rice AS, Maton S. Gabapentin in postherpetic neuralgia: a randomised, double-blind, placebo-controlled study. Pain. 2001;2:215-224.

28. Pande AC, Davidson JR, Jefferson JW, et al. Treatment of social phobia with gabapentin: a placebo-controlled study. J Clin Psychopharmacol. 1999;4:341-348.

29. Rowbotham M, Harden N, Stacey B, Bernstein P, Magnus-Miller L. Gabapentin for the treatment of postherpetic neuralgia: a randomized controlled trial. JAMA. 1998;21:1837-1842.

30. Edwards KR, Glantz MJ, Button J. Efficacy and safety of topiramate in the treatment of painful diabetic neuropathy: a double-blind, placebo-controlled study (abstract). Neurology. 2000;7(suppl 3):A81.

31. Data on file. Ortho-McNeil Pharmaceuticals; Raritan, NJ.

32. Vinik A, Hewitt D, Xiang J, Jordan D, Rosenthal N. Topiramate in the treatment of painful diabetic neuropathy: results from a multicenter, randomized, double-blind, placebo-controlled trial (abstract). Neurology. 2003;60(suppl 1):A154-A155.

33. Eads LA, Kramer T. Effects of topiramate on global functioning in treatment-refractory mood disorders (abstract). Int J Neuropsychopharmacol. 2000;3(suppl 1):S340.

34. Rothrock JR, Jacobs D, Neto W, Bhattacharaya S. Safety and tolerability of topiramate in migraine prevention: results from a multicenter, randomized, double-blind, placebo-controlled study (MIGR-002). Abstract presented at: the 55th Annual Meeting of the American Academy of Neurology; April 1-3 2003; Honolulu, Hawaii.

35. Zakrzewska JM, Chaudhry Z, Nurmikko TJ, Patton DW, Mullens EL. Lamotrigine (lamictal) in refractory trigeminal neuralgia: results from a double-blind placebo controlled crossover trial. Pain. 1997;2:223-230.

36. Simpson DM, McArthur JC, Olney R, et al. Lamotrigine for HIV-associated painful sensory neuropathies: A placebo-controlled trial. Neurology. 2003;9:1508-1514.

37. Davis JL, Smith RL. Painful peripheral diabetic neuropathy treated with venlafaxine HCl extended release capsules. Diabetes Care. 1999;11:1909-1910.

38. Jarvis B, Coukell AJ. Mexiletine. A review of its therapeutic use in painful diabetic neuropathy. Drugs. 1998;4:691-707.

39. Rowbotham MC, Davies PS, Verkempinck C, Galer BS. Lidocaine patch: double-blind controlled study of a new treatment method for post-herpetic neuralgia. Pain. 1996;1:39-44.

40. Effect of treatment with capsaicin on daily activities of patients with painful diabetic neuropathy. Capsaicin Study Group. Diabetes Care. 1992;2:159-165.

41. Biesbroeck R, Bril V, Hollander P, et al. A double-blind comparison of topical capsaicin and oral amitriptyline in painful diabetic neuropathy. Adv Ther. 1995;2:111-120.

42. Watson CP, Tyler KL, Bickers DR, Millikan LE, Smith S, Coleman E. A randomized vehicle-controlled trial of topical capsaicin in the treatment of postherpetic neuralgia. Clin Ther. 1993;3:510-526.

43. Treatment of painful diabetic neuropathy with topical capsaicin. A multicenter, double-blind, vehicle-controlled study. The Capsaicin Study Group. Arch Intern Med. 1991;11:2225-2229.

44. Beydoun A. Postherpetic neuralgia: role of gabapentin and other treatment modalities. Epilepsia. 1999;40(suppl 6):S51-S56.

45. Watson CP, Babul N. Efficacy of oxycodone in neuropathic pain: a randomized trial in postherpetic neuralgia. Neurology. 1998;6:1837-1841.

46. Sindrup SH, Andersen G, Madsen C, Smith T, Brosen K, Jensen TS. Tramadol relieves pain and allodynia in polyneuropathy: a randomised, double-blind, controlled trial. Pain. 1999;1:85-90.

47. Richardson JK. The clinical identification of peripheral neuropathy among older persons. Arch Phys Med Rehab. 2002;11:1553-1558.


Print Friendly 

Dylan P. Wint, MD, and Nathan A. Shapira, MD, PhD

Dr. Wint is a staff physician at the Malcolm Randall VA Medical Center, and assistant professor in the Department of Psychiatry at the University of Florida in Gainesville.

Dr. Shapira is assistant professor in the Department of Psychiatry at the University of Florida.

Disclosure: Dr. Shapira is on the speaker’s bureau of Forest and Ortho-McNeil; receives research support from Abbott, Janssen, and Ortho-McNeil; and has received honoraria from Bristol-Myers Squibb, Eli Lilly, Forest, Janssen, and Pfizer.

Acknowledgments: The authors would like to thank Edward Weselcouch, PhD, for his assistance in the preparation of this manuscript.

Please direct all correspondence to: Dylan P. Wint, MD, Department of Psychiatry, PO Box 100256, Gainesville, FL 32610; Tel: 352-392-3681; Fax: 352-379-4170; E-mail:

Focus Points

Impulsivity is found in a broad range of psychiatric syndromes and impulsive behavior can take many forms.

Well-defined biological correlates of impulsive behavior exist.

There are few well-studied treatments for some types of impulsivity and there is much to learn about the causes and treatments of impulsive behaviors overall.


Impulsivity is found across a broad range of psychiatric syndromes. It also encompasses many types of behavior, including aggression, eating, stealing, and gambling. Some types of impulsive behavior are associated with changes in brain activity and neurotransmitter turnover; serotonergic changes are particularly well linked to impulsive aggression. Efforts continue to discover the biological roots of impulsive behavior. Nevertheless, some effective treatments have been developed and antiepileptic drugs in particular have shown promise in ameliorating these disturbing behaviors.


Impaired impulse control, which is common in mental illness and can be life threatening, may be the most important characteristic of psychiatric disorders. Patients who do not possess reliable control of their impulses usually require hospitalization, restraints, forced treatment, or other impositions on their privacy and personal choices. In some psychiatric syndromes impulsivity is rare, while in others it is an essential feature. There are numerous theories about the causes of impulsivity and appropriate treatments for impulse dyscontrol. This review will briefly examine the pathophysiology and treatment of some of the clinically important impulsive behaviors seen in psychiatric illnesses.

Impulsive Anger and Aggression

Anger and aggression are among the most distressing symptoms of psychiatric illness. The lives of patients, families, and whole communities are severely disturbed when violence erupts.1 Unfortunately, anger and aggression are also rather frequent in a subgroup of the psychiatrically ill.2,3 In fact, potential or demonstrated aggressive behavior is one of the most common reasons for involuntary confinement.4 Unlike planned, predatory violence (not addressed in this article), impulsive aggression among psychiatric patients tends to be spontaneous and easily provoked.1 Aggression and violence may be directed at patients, those around them, or both.

Aggression Directed At Others


Outward-directed aggression is probably the best-studied manifestation of impulsivity, because it is relatively easy to observe and quantify. It presents itself in many psychiatric illnesses, with intermittent explosive disorder (IED) being the prototypical form of impulsive aggression. According to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition Text-Revision (DSM-IV-TR),5 IED is characterized by repeated failure to resist aggressive impulses that result in serious assaultive acts or destruction of property. IED?patients act with a degree of aggressiveness that is grossly out of proportion to the aggression-inducing stressor. Less prototypical, but more frequent, is the impulsive aggression seen in dementia, where verbal aggression and physical attack upon caregivers is common.6 Other illnesses with high incidences of outward-directed aggression include bipolar disorder,7 substance-use disorders, personality disorders,8 and psychotic disorders.9

The aggressive actions of psychiatric patients can be as harmless as angry stares and verbal assault, or as destructive as vandalism and physical violence. The type of illness does not necessarily predict the degree or nature of the aggressive acts, but the existing literature, conflicting as it is, suggests that young men with delusions seem to be the most at risk, while patients (both male and female) with depression are at lowest risk.1,10,11 The literature also implies, and personal observation confirms, that psychiatrically disturbed patients often act violently on the basis of a distorted, but genuine, instinct of self-preservation—their perception of threat, and their assessment of the appropriate action, are altered by their illness.


Multiple lines of research indicate that the frontal and temporal lobes play pivotal roles in the mediation of aggressive impulses. The most basic model posits that the orbitofrontal and prefrontal regions of the brain inhibit the aggressive impulses that are generated by the temporal lobe’s limbic structures. Head-injured veterans who have suffered damage to the orbitofrontal regions are more likely to have post injury aggression than those with nonfrontal brain lesions.12 In patients with dementia, frontal hypoperfusion as measured by positron emission tomography (PET) is associated with violent behavior.13 Another PET study showed that personality-disordered patients with decreased orbitofrontal metabolism are, by their own reports, more aggressive.14 In yet another PET study, decreased frontal metabolism distinguished convicted prisoners who committed crimes of passion (one type of impulsive aggression) from those who had planned their violent crimes.15 Magnetic resonance imaging diffusion tensor imaging of schizophrenic subjects suggests that atypicality of frontal white matter is associated with impulsivity and aggression.16

Temporal lobe dysfunction also results in impulsive aggression. Trauma to the temporal lobe has been associated with post injury violence17 and patients with temporal lobe epilepsy may also show aggressive behavior, especially when anterior temporal atrophy is present.18 Thus, impulsive aggression seems to result from an unfavorable balance of inhibitory influences from the prefrontal/ventromedial frontal lobe and pro-aggressive activity from the anterior temporal lobe.19

Knowledge about the relationships between impulsive aggression and neurotransmitters is evolving. The field’s best studied and most robust finding is the negative correlation between impulsive violence and markers of serotonin turnover in the brain. Linnoila and colleagues20 discovered that impulsivity was associated with low cerebrospinal fluid concentrations of 5-hydroxyindoleacetic acid, a serotonin metabolite, in aggressive convicts. This relationship between impulsivity and serotonin has been replicated in other studies of humans21,22 and nonhuman primates.23-25 This is particularly interesting in light of the prefrontal area’s high concentration of serotonin receptors.26 Serotonergic deficiency seems to be a trait phenomenon—consistently present regardless of the individual’s state of mind.27

Other studies have indicated that dopamine turnover in the brain correlates positively with aggression.28 Homozygosity for a low-activity variant of the dopamine-catabolizing enzyme catechol-O-methyltransferase predicts serious aggressive behavior in schizophrenics.29 Some evidence suggests, however, that the increased dopamine metabolism in aggressive subjects merely reflects decreased serotonergic inhibition of dopaminergic tone.30 Because of its association with stress responses, norepinephrine has also been implicated in the pathophysiology of impulsivity although its role is less clear. Alteration of norepinephrine metabolism may be a state phenomenon—occurring around the time of an impulsive action, rather than as a persistent abnormality.31 Acetylcholine may also play a role, as nicotine appears to reduce aggressive behavior.32


Given the multifactorial etiology of impulsive aggression, it is not surprising that the most commonly used treatments are not specific to any particular brain region or neurotransmitter system. The mainstays of pharmacologic treatment for impulsive aggression have been the antiepileptics (AEDs).33 Valproic acid is broadly effective in treating aggression and impulsivity.34 In the treatment of acute mania, valproate may have particular efficacy for impulsivity and irritability.35 In mostly uncontrolled studies, valproate was useful in ameliorating impulsive aggression associated with dementia,36 traumatic brain injury,37 and cluster B personality disorders.38 Carbamazepine may also effectively treat impulsive aggression. Among rehabilitation specialists, it is the preferred drug for patients who have suffered traumatic brain injury39 and it has demonstrated efficacy in aggression associated with dementia40 and epilepsy.41 Gabapentin, one of the “new generation” AEDs, was used successfully in a retrospective study of dementia-related aggression.42 Finally, an open-label reported that topiramate improved behavioral disturbances in patients with Alzheimer’s disease.43 Clinical effectiveness was evident during the first week of therapy and was sustained for at least
4 weeks with doses of 25–100 mg/day.

Although the role of antipsychotics has yet to be fully determined, their influence on dopamine- and serotonin-related neurotransmission has stimulated interest in studying their clinical effects. In a controlled study of 206 subjects with Alzheimer’s dementia, olanzapine 5–10 mg was significantly superior to placebo in controlling psychosis and aggression.44 A similar study with 337 subjects demonstrated the efficacy of risperidone in reducing aggressive behavior.45 Interpreting these results is somewhat difficult, because antipsychotics may improve thought content and processes, rather than impulsive aggression per se.

Other medications that may ameliorate impulsive aggression are lithium,33 propranolol,46 and the antidepressants.37 Lithium has a particularly long history of use for impulsive aggression,47 but recent controlled studies have used only conduct-disordered children as subjects.48-50 Unfortunately, because of the necessity of serum monitoring, multiple drug interactions, and a variety of dangerous side effects, lithium is poorly suited to treating many impulsive patients.51,52 The use of propranolol in patients with aggression due to organic brain disease is supported by a blinded crossover study of 10 otherwise refractory patients. Treatment with propranolol may be limited by bradycardia and hypotension.53 In a well-controlled study of 40 patients with personality disorders, fluoxetine reduced irritability and aggression.54 Although many of the medications used to treat impulsive aggression have sedating qualities, it should be noted that sedation is usually neither necessary nor sufficient to achieve control of aggression.

Environmental, behavioral, and cognitive treatments should also be used in impulsive aggression. Misinterpretation of the intentions of hospital staff can induce violence in psychiatric patients.55,56 Giving patients positive reinforcement of appropriate behavior and social skills training may dramatically reduce verbal and physical aggression.57 Training of those who interact with disturbed patients may also be of benefit.58

Self-Directed Aggression

Suicide attempts and self-mutilation are common precipitants to psychiatric hospitalization and treatment. In the United States, the suicide rate averaged 12.5 per 100,000 people in the 1990s.59 Depression is responsible for 50% of completed suicides, but substance use disorders, mania, personality disorders, and adjustment disorders are also important risk factors.59

For a variety of reasons, psychiatrically disturbed patients cause themselves injury without the intent to die. Although it is casually linked to cluster B personality disorders, self-injurious behavior (SIB) occurs as a result of many psychiatric illnesses, including developmental impairment, substance abuse, mania, and depression.42 Other forms of SIB include skin picking in Prader-Willi syndrome, self-mutilation in Lesch-Nyhan syndrome, and trichotillomania. Patients suffering from psychosis may also seriously injure themselves in response to command hallucinations or delusions.


Even less is known about self-directed aggression than is known about violence directed at others. Suicidality, like outward-directed impulsive aggression, has been repeatedly associated with low serotonin levels in the central nervous system.60 The lethality of suicide attempts is correlated with decreased frontal lobe metabolism and reduced serotonergic reactivity, two frequent findings in major depressive disorder.61,62 Serotonergic tone (probably decreased) also seems to play a role in self-mutilation. In addition, opiate and dopamine activity are thought to be involved in creating a biological reward system that promotes SIB.63


Treatment of suicidality is typically accomplished by treating the underlying disorder (eg, major depressive disorder, alcohol dependence). However, some treatments seem to be of particular benefit when applied to suicidal patients. Chronic lithium therapy seems to have a specific antisuicidal effect.64 Electroconvulsive therapy can treat acute suicidality, but seems to have no effect on long-term suicide risk.65

Clearly, more work must to be done in order to understand the causes and treatments of suicide. British authorities recently banned the prescribing of paroxetine to children because of a paradoxical increase in suicide rates among patients who were treated with the medication.66 The cause of this phenomenon is under investigation but the controversy is somewhat reminiscent of similar concerns about increased violence resulting from fluoxetine treatment; those fears have proven to be unfounded.67

Nonsuicidal SIB lends itself better to long-term psychotherapeutic and pharmacologic treatments. Linehan’s dialectical behavioral therapy is effective in decreasing parasuicidal behavior in women suffering from borderline personality disorder.68 In a study of borderline patients, carbamazepine had a beneficial effect on behavioral dyscontrol; in the same study, alprazolam exacerbated dyscontrol.69 Fluoxetine may also reduce self-injurious behavior in borderline patients.70 In open-label studies, topiramate has shown efficacy in treating the self-injurious behavior of Prader-Willi syndrome,71,72 and carbamazepine has been effective in treating Lesch-Nyhan self-mutilation.73

Impulsive Consumption and Pleasure Seeking

Bulimia Nervosa

Bulimia nervosa is an eating disorder characterized by the DSM-IV-TR5 as binge eating and inappropriate compensatory methods to prevent weight gain occurring 2 times/week and persisting for 3 months.5 In purging-type bulimia, the subject engages in self-induced vomiting or misuse of laxatives, diuretics, or enemas. Patients with nonpurging-type bulimia compensate with other abnormal behaviors, such as fasting or excessive exercise. The prevalence is estimated at 1% to 3% in the general population and 4% to 13% in female college-age students.74 The disorder is more common in women than in men and often causes dental problems due to repeated vomiting, menstrual problems, electrolyte imbalance, and severe mood disturbances. Although not mentioned in DSM-IV-TR, impulsivity and bulimia are clearly associated.75 About 30% of bulimic patients exhibit SIB, compared to 4% to 10% in the general psychiatric population.76 Bulimia’s other common impulsive comorbidities are substance use disorders, kleptomania, and compulsive buying.77-79


Successful treatment for bulimia nervosa should be anchored by cognitive-behavioral therapy (CBT), which is the best-studied remedy for the illness. In a study of 113 women who were treated with CBT, 89% were no longer bulimic and 69% had no eating disorder diagnosis after 3 years of treatment.80 Interpersonal therapy has a similar long-term benefit.81 Many of the studies that have evaluated psychotherapy for bulimia, however, are not blinded.

Until recently, the pharmacologic treatment of bulimia nervosa consisted primarily of antidepressants, especially tricyclic antidepressants (TCAs) and selective serotonin reuptake inhibitors (SSRIs). Several double-blind, placebo-controlled studies have demonstrated the efficacy of TCAs in treating bulimia.82,83 Hughes and colleagues82 reported that desipramine reduced weekly binge frequency by 91% compared to a 19% increase in those treated with placebo (P<.01). Despite the apparent success of TCAs, their use is limited by an array of side effects, including weight gain, drowsiness, and cardiac arrhythmias.84

The SSRI fluoxetine was shown to be effective in a double-blind study of
387 bulimic women.85 In this 8-week trial, fluoxetine 60 mg/day was superior to placebo in reducing the weekly frequency of binge eating and vomiting episodes. Fluoxetine is the only antidepressant approved by the Food and Drug Administration for the treatment of bulimia nervosa. Of interest is the observation that the effectiveness of the antidepressants in the treatment of bulimia appears to be distinct from their effect on depression, ie, their effect is similar in patients regardless of the presence or degree of concomitant depression.82,83

Other pharmacologic classes are being evaluated for use in bulimia nervosa. For example, ondansetron, a 5-HT3 receptor antagonist, resulted in a 50% reduction in binge-purge episodes— an effect attributed to restoring vagal afferent sensitivity and increasing the awareness of satiety.86 AEDs have also been used, although early studies demonstrated minimal benefit.87,88 Recently, topiramate has generated interest based on several case reports and a single clinical study.89,90 In a randomized, double-blind study of 69 patients, topiramate median dose, 100 mg/day reduced the number of mean binge and/or purge days by 45% compared to 11% seen in the placebo group.91 The mechanism of this effect has not been elucidated, but topiramate is an inhibitor of voltage-gated sodium and L-type calcium channels, an enhancer of γ-aminobutyric acid (GABA)A receptor-mediated chloride flux, and an inhibitor of glutamate-mediated excitation of α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)/kainate receptors.92

Binge-Eating Disorder

Binge-eating disorder (BED) is characterized in the DSM-IV5 as recurrent episodes of binge eating in the absence of the inappropriate compensatory behaviors of bulimia nervosa. The prevalence of BED is estimated to be 1.5% to 2% in the general population, but approaches 45% in obese individuals.93 A recent study reported that mutation to the gene coding for the melanocortin 4 receptor is strongly associated with binge eating; all obese subjects with a mutation reported binge eating, compared to only 14% of obese subjects with the normal phenotype.94 In addition to its primary characteristics of bingeing and obesity, BED is strongly associated with mood symptoms, and those who are obese have more comorbid depression and personality disorders.95


Although CBT has been associated with reductions in binge eating, it typically has little effect on weight.93 Pharmacotherapeutic studies have centered on the use of SSRIs and D-fenfluramine. In a 6-week, double-blind study, Arnold and colleagues96 reported that fluoxetine reduced both binge frequency and body weight. In a similar study, D-fenfluramine reduced binge episodes with modest effect on weight loss.97 Other treatments associated with reductions in binge frequency and modest weight loss include fluvoxamine, sertraline, sibutramine, and venlafaxine.98-101

Naltrexone and topiramate have recently been studied for the treatment of BED. Although naltrexone reduced binge frequency in an 8-week, double-blind study, the effect was no different than that seen in the placebo group.102 In an open-label study of women with BED who were treated with topiramate 100–1,400 mg/day, 9 of 13 had at least a 50% decrease in binge-eating episodes.103 The mean body weight of all subjects decreased by an average of 12 kg during the 3–30 month study period.103 In a double-blind study, 61 obese patients were treated with topiramate 25–600 mg/day or placebo for 14 weeks. Those patients in the topiramate group experienced a greater reduction in binge frequency and BMI than placebo-treated patients.104 An open-label extension of this study demonstrated continued efficacy with significant reductions of binge frequency and weight over an additional 42 weeks.105


Kleptomania is defined by the DSM-IV5 as “the recurrent failure to resist impulses to steal items even though the items are not needed for personal use or for their monetary value.” The prevalence of kleptomania in the general population is estimated to be 0.6%, but this may be an underestimate because few people disclose their problem, even to their therapist, and most escape detection.106 Kleptomanic behavior is impulsive and repetitive. It is also associated with anxiety one fights to resist the impulse to steal and suffers guilt and shame over engaging in the act of stealing.107 Women are affected with kleptomania four times as often as men.106


Kleptomania shares comorbidity with mood and anxiety disorders, eating disorders, and substance-abuse disorders.108 It is through these comorbidities that patients with kleptomania are usually referred for treatment.109 The choices of therapy for the treatment of kleptomania have historically been antidepressants, AEDs, lithium, or valproate.110,111 The results of pharmacologic treatment are variable. For example, in a case series of 20 patients, only two cases of remission with fluoxetine were reported whereas seven others were unresponsive.112 The combination of SSRIs and mood stabilizers (lithium with fluoxetine,113 valproate with fluvoxamine114) has had some positive results. Topiramate was reported to reduce both the urge to steal and the pleasure derived from the act in a case series of three patients. The investigator hypothesized that the mechanism might be increased GABA input to the nucleus accumbens.107

Pathological Gambling

Pathological gambling (PG) is characterized in the DSM-IV5 as an impulse-control disorder that is disabling and a public health issue because of its devastating effect on families. Its prevalence in the US has been estimated at between 1% and 3.4% of the adult population, with more men than women affected.115 Alcohol and substance abuse disorders are common in the PG populations, and have been reported in as many as 47% of inpatient pathological gamblers.116 PG is highly comorbid with mood disorders, substance-use disorders, attention-deficit/hyperactivity disorder, and other impulse-control disorders.111,117 Neurobiological studies have suggested the involvement of serotonin, norepinephrine, and dopamine neurotransmitter systems in the etiology of PG.118,119


Treatment of PG has traditionally centered on psychotherapy and self-help groups (eg, Gamblers Anonymous), but the evidence for success is not very convincing. However, the results of recent studies suggest the potential of pharmacotherapy to positively influence PG. Naltrexone has been reported to improve PG symptoms in open-label,120 double-blind, placebo-controlled,121 and retrospective chart review studies.122 The most common adverse effect of naltrexone in these reports was nausea. Naltrexone may, by blocking opioid receptors, blunt behavioral urges or decrease the biological reward associated with pathological gambling.123

At least four recent clinical studies suggest that SSRI therapy may be effective in the treatment of PG. For example, Zimmerman and colleagues124 reported that citalopram, in an open-label study of 15 adult PG subjects, significantly reduced all gambling measures, including number of days gambling, amount of money lost, and number of urges to gamble. The effect was independent of the antidepressant effects of citalopram. Others have reported similar benefits of SSRI treatment,125-127 although one study suggested a benefit only in men and younger patients.125 Nefazodone, a 5-HT2A receptor antagonist, may also be successful in treating PG.128

Lithium and valproate may also be effective in treating patients with PG. In one study, either drug was administered to nonbipolar PG subjects in a single-blind fashion.129 After 14 weeks, significant improvement was reported for both drugs on the Yale-Brown Obsessive Compulsive Scale modified for PG. Fourteen of 23 patients (61%) treated with lithium and 13 of the 19 patients (68%) taking valproate were much improved as assessed by the Clinical Global Impressions scale for Improvement.


Impaired impulse control is a common and prominent element of psychiatric disorders. Knowledge about the neuropathophysiology of impulse-control disorders is slowly emerging, but the complex interaction of the systems involved and the common presence of comorbidities has made the search for appropriate and specific pharmacotherapy difficult. Impulsivity seems to be closely connected to the serotonergic tone, with contributions of dopamine-, opioid-, and noradrenergic-responsive systems. As our ability to analyze the function of the brain improves, we will be able to add our already impressive armamentarium of psychotherapeutic and pharmacologic treatments. In the future, treatments for impulsivity will become more specific and effective, improving the lives of patients and their families. PP


1. Hollander E, Posner N, Cherkasky S. The neuropsychiatry of aggression and impulse control disorders. In: Yudofsky SC, Hales RE, eds. American Psychiatric Press Textbook of Neuropsychiatry. Washington, DC: American Psychiatric Press, Inc; 2002:579-596.

2. Grassi L, Peron L, Marangoni C, et al. Characteristics of violent behaviour in acute psychiatric in-patients: a 5-year Italian study. Acta Psychiatr Scand. 2001;104:273-279.

3. Goldsmith HF, Manderscheid RW, Henderson MJ, et al. Projections of inpatient admissions to specialty mental health organizations: 1990 to 2010. Hosp Comm Psychiatry. 1993;44:478-483.

4. Noble P. Violence in psychiatric in-patients: review and clinical implications. Int Rev Psychiatry. 1997;9:207-216.

5. Diagnostic and Statistical Manual of Mental Disorders. 4th ed, text-rev. Washington, DC: American Psychiatric Association; 2000.

6. Eastley R, Wilcock GK. Prevalence and correlates of aggressive behaviours occurring in patients with Alzheimer’s disease. Int J Geriatr Psychiatry. 1997;4:484-487.

7. Manic-Depressive Illness. New York, NY: Oxford University Press; 1990.

8. Gunderson JG, Phillips KA. Personality Disorders. In: Kaplan HI, Sadock BJ, eds. Comprehensive Textbook of Psychiatry/VI. Baltimore, MD: Williams and Wilkins; 1995:1425-1461.

9. Wallace C, Mullen P, Burgess P, Palmer S, Ruschena D, Browne C. Serious criminal offending and mental disorder. Case linkage study. Br J Psychiatry. 1998;172:477-484.

10. Sheridan M, Henrion RE, Robinson L, et al. Precipitants of violence in a psychiatric inpatient setting. Hosp Comm Psychiatry. 1990;41:776-780.

11. Barlow K, Grenyer B, Ilkiw-Lavalle O. Prevalence and precipitants of aggression in psychiatric inpatient units. Aust N Z J Psychiatry. 2000;34:967-974.

12. Grafman J, Schwab K, Warden D, Pridgen A, Brown HR, Salazar AM. Frontal lobe injuries, violence, and aggression: a report of the Vietnam Head Injury Study. Neurology. 1996;5:1231-1238.

13. Hirono N, Mega MS, Dinov ID, Mishkin F, Cummings JL. Left frontotemporal hypoperfusion is associated with aggression in patients with dementia. Arch Neurol. 2000;6:861-866.

14. Goyer PF, Andreason PJ, Semple WE, et al. Positron-emission tomography and personality disorders. Neuropsychopharmacology. 1994;1:21-28.

15. Raine A, Meloy JR, Bihrle S, Stoddard J, LaCasse L, Buchsbaum MS. Reduced prefrontal and increased subcortical brain functioning assessed using positron emission tomography in predatory and affective murderers. Behav Sci Law. 1998;3:319-332.

16. Hoptman MJ, Volavka J, Johnson G, Weiss E, Bilder RM, Lim KO. Frontal white matter microstructure, aggression, and impulsivity in men with schizophrenia: a preliminary study. Biol Psychiatry. 2002;1:9-14.

17. Tonkonogy JM. Violence and temporal lobe lesion: head CT and MRI data. J Neuropsychiatry Clin Neurosci. 1991;2:189-196.

18. van Elst LT, Woermann FG, Lemieux L, Thompson PJ, Trimble MR. Affective aggression in patients with temporal lobe epilepsy: a quantitative MRI study of the amygdala. Brain. 2000;123(Pt 2):234-243.

19. Amen DG, Stubblefield M, Carmicheal B, Thisted R. Brain SPECT findings and aggressiveness. Ann Clin Psychiatry. 1996;3:129-137.

20. Linnoila M, Virkkunen M, Scheinin M, Nuutila A, Rimon R, Goodwin FK. Low cerebrospinal fluid 5-hydroxyindoleacetic acid concentration differentiates impulsive from nonimpulsive violent behavior. Life Sci. 1983;26:2609-2614.

21. Coccaro EF. Central serotonin and impulsive aggression. Br J Psychiatry. 1989;8:52-62.

22. Brown GL, Linnoila MI. CSF serotonin metabolite (5-HIAA) studies in depression, impulsivity, and violence. J Clin Psychiatry. 1990;51(suppl):31-41.

23. Westergaard GC, Suomi SJ, Chavanne TJ, et al. Physiological correlates of aggression and impulsivity in free ranging female primates. Neuropsychopharmacology. 2003;6:119-148.

24. Higley JD, Mehlman PT, Poland RE, et al. CSF testosterone and 5-HIAA correlate with different types of aggressive behaviors. Biol Psychiatry. 1996;11:1067-82.

25. Fairbanks LA, Melega WP, Jorgensen MJ, et al. Social impulsivity inversely associated with CSF 5-HIAA and fluoxetine exposure in vervet monkeys. Neuropsychopharmacology. 2001;4:370-8.

26. Evans J, Platts H, Lightman S, Nutt D. Impulsiveness and the prolactin response to d-fenfluramine. Psychopharmacology (Berl). 2000;2:147-152.

27. Biver F, Lotstra F, Monclus M, et al. Sex difference in 5-HT2 receptor in the living human brain. Neurosci Lett. 1996;1:25-28.

28. Soderstrom H, Blennow K, Manhem A, Forsman A. CSF studies in violent offenders. I. 5-HIAA as a negative and HVA as a positive predictor of psychopathy. J Neural Transm. 2001;17:869-878.

29. Kotler M, Barak P, Cohen H, et al. Homicidal behavior in schizophrenia associated with a genetic polymorphism determining low catechol-O-methyltransferase (COMT) activity. Am J Med Genet. 1999;6:628-633.

30. Brunner D, Hen R. Insights into the neurobiology of impulsive behavior from serotonin receptor knockout mice. Ann N Y Acad Sci. 1997;836:81-105.

31. Oquendo MA, Mann JJ. The biology of impulsivity and suicidality. Psychiatr Clin North Am. 2000;1:11-25.

32. Cherek DR. Effects of cigarette smoking on human aggressive behavior. Prog Clin Biol Res. 1984;169:333-344.

33. Fava M. Psychopharmacologic treatment of pathologic aggression. Psychiatr Clin North Am. 1997;2:427-451.

34. Lindenmayer JP, Kotsaftis A. Use of sodium valproate in violent and aggressive behaviors: a critical review. J Clin Psychiatry. 2000;2:123-128.

35. Swann AC, Bowden CL, Calabrese JR, Dilsaver SC, Morris DD. Pattern of response to divalproex, lithium, or placebo in four naturalistic subtypes of mania. Neuropsychopharmacology. 2002;4:530-536.

36. Grossman F. A review of anticonvulsants in treating agitated demented elderly patients. Pharmacotherapy. 1998;3:600-606.

37. Wroblewski BA, Joseph AB, Kupfer J, Kalliel K. Effectiveness of valproic acid on destructive and aggressive behaviours in patients with acquired brain injury. Brain Inj. 1997;1:37-47.

38. Hollander E, Tracy KA, Swann AC, et al. Divalproex in the treatment of impulsive aggression: efficacy in Cluster B personality disorders. Neuropsychopharmacology. 2003;6:1186-1197.

39. Fugate LP, Spacek LA, Kresty LA, Levy CE, Johnson JC, Mysiw WJ. Measurement and treatment of agitation following traumatic brain injury: II. A survey of the Brain Injury Special Interest Group of the American Academy of Physical Medicine and Rehabilitation. Arch Phys Med Rehabil. 1997;9:924-928.

40. Tariot PN, Erb R, Podgorski CA, et al. Efficacy and tolerability of carbamazepine for agitation and aggression in dementia. Am J Psychiatry. 1998;1:54-61.

41. Thibaut F, Colonna L. [Carbamazepine and aggressive behavior: a review]. Encephale. 1993;6:651-656.

42. Hawkins JW, Tinklenberg JR, Sheikh JI, Peyser CE, Yesavage JA. A retrospective chart review of gabapentin for the treatment of aggressive and agitated behavior in patients with dementias. Am J Geriatr Psychiatry. 2000;3:221-225.

43. Napolitani BG, Talarico G, Ottaviani D, et al. Topiramate treatment of behavioral disturbances in Alzheimer’s disease. Presented at: the 8th International Converence on Alzheimer’s Disease and Related Disorders; July 17-22, 2002; Stockholm, Sweden.

44. Street JS, Clark WS, Gannon KS, et al. Olanzapine treatment of psychotic and behavioral symptoms in patients with Alzheimer disease in nursing care facilities: a double-blind, randomized, placebo-controlled trial. Arch Gen Psychiatry. 2000;10:968-76.

45. Brodaty H, Ames D, Snowdon J, et al. A randomized placebo-controlled trial of risperidone for the treatment of aggression, agitation, and psychosis of dementia. J Clin Psychiatry. 2003;2:134-43.

46. Silver JM, Yudofsky SC, Slater JA, et al. Propranolol treatment of chronically hospitalized aggressive patients. J Neuropsychiatry Clin Neurosci. 1999;3:328-335.

47. Sheard, MH, Marini JL, Bridges CI, et al. The effect of lithium on impulsive aggressive behavior in man. Am J Psychiatry. 1976;133:1409-1413.

48. Campbell M, Adams PB, Small AM, et al. Lithium in hospitalized aggressive children with conduct disorder: a double-blind and placebo-controlled study. J Am Acad Child Adolesc Psychiatry. 1995;5:445-453.

49. Malone RP, Delaney MA, Luebbert JF, et al. A double-blind placebo-controlled study of lithium in hospitalized aggressive children and adolescents with conduct disorder. Arch Gen Psychiatry. 2000;7:649-654.

50. Rifkin A, Karajgi B, Dicker R, et al. Lithium treatment of conduct disorders in adolescents. Am J Psychiatry. 1997;4:554-555.

51. Tueth MJ, Murphy TK, Evans DL. Special considerations: use of lithium in children, adolescents, and elderly populations. J Clin Psychiatry. 1998;59(suppl 6):66-73.

52. Kores B, Lader MH. Irreversible lithium neurotoxicity: an overview. Clin Neuropharmacol. 1997;4:283-299.

53. Greendyke RM, Kanter DR, Schuster DB, et al. Propranolol treatment of assaultive patients with organic brain disease. A double-blind crossover, placebo-controlled study. J Nerv Ment Dis. 1986;5:290-294.

54. Coccaro EF, Kavoussi RJ. Fluoxetine and impulsive aggressive behavior in personality-disordered subjects. Arch Gen Psychiatry. 1997;12:1081-1088.

55. Davis S. Violence by psychiatric inpatients: a review. Hosp Comm Psychiatry. 1991;6:585-90.

56. Whittington R, Wykes T. An observational study of associations between nurse behaviour and violence in psychiatric hospitals. J Psychiatr Ment Health Nurs. 1994;2:85-92.

57. Vaccaro FJ. Application of operant procedures in a group of institutionalized aggressive geriatric patients. Psychol Aging. 1988;1:22-28.

58. Hoeffer B, Rader J, McKenzie D, Lavelle M, Stewart B. Reducing aggressive behavior during bathing cognitively impaired nursing home residents. J Gerontol Nurs. 1997;5:16-23.

59. Roy A. Psychiatric emergencies. In: Kaplan HI, Sadock BJ, eds. Comprehensive Textbook of Psychiatry/VI. Baltimore, MD: Williams and Wilkins; 1995:1081-1097.

60. Shaw DM, Camps FE, Eccleston EG. 5-Hydroxytryptamine in the hind-brain of depressive suicides. Br J Psychiatry. 1967;505:1407-1411.

61. Oquendo MA, Placidi GP, Malone KM, et al. Positron emission tomography of regional brain metabolic responses to a serotonergic challenge and lethality of suicide attempts in major depression. Arch Gen Psychiatry. 2003;1:14-22.

62. Baxter LR Jr., Schwartz JM, Phelps ME, et al. Reduction of prefrontal cortex glucose metabolism common to three types of depression. Arch Gen Psychiatry. 1989;3:243-250.

63. Winchel RM, Stanley M. Self-injurious behavior: a review of the behavior and biology of self-mutilation. Am J Psychiatry. 1991;3:306-317.

64. Baldessarini RJ, Tondo L, Hennen J. Lithium treatment and suicide risk in major affective disorders: update and new findings. J Clin Psychiatry. 2003;64(suppl 5):44-52.

65. Prudic J, Sackeim HA. Electroconvulsive therapy and suicide risk. J Clin Psychiatry. 1999;60(suppl 2):104-110.

66. Abbott A. British panel bans use of antidepressant to treat children. Nature. 2003;6942:792.

67. Walsh MT, Dinan TG. Selective serotonin reuptake inhibitors and violence: a review of the available evidence. Acta Psychiatr Scand. 2001;2:84-91.

68. Koerner K, Linehan MM. Research on dialectical behavior therapy for patients with borderline personality disorder. Psychiatr Clin North Am. 2000;1:151-67.

69. Cowdry RW, Gardner DL. Pharmacotherapy of borderline personality disorder. Alprazolam, carbamazepine, trifluoperazine, and tranylcypromine. Arch Gen Psychiatry. 1988;2:111-119.

70. Markovitz PJ, Calabrese JR, Schulz SC, Meltzer HY. Fluoxetine in the treatment of borderline and schizotypal personality disorders. Am J Psychiatry. 1991;8:1064-1067.

71. Smathers SA, Wilson JG, Nigro MA. Topiramate effectiveness in Prader-Willi syndrome. Pediatr Neurol. 2003;2:130-133.

72. Shapira NA, Lessig MC, Murphy TK, Driscoll DJ, Goodman WK. Topiramate attenuates self-injurious behaviour in Prader-Willi syndrome. Int J Neuropsychopharmacol. 2002;2:141-145.

73. Roach ES, Delgado M, Anderson L, Iannaccone ST, Burns DK. Carbamazepine trial for Lesch-Nyhan self-mutilation. J Child Neurol. 1996;6:476-478.

74. Hsu LK. Epidemiology of the eating disorders. Psychiatr Clin North Am. 1996;4:681-700.

75. Fahy T, Eisler I. Impulsivity and eating disorders. Br J Psychiatry. 1993;162:193-197.

76. Paul T, Schroeter K, Dahme B, Nutzinger DO. Self-injurious behavior in women with eating disorders. Am J Psychiatry. 2002;3:408-411.

77. Baum A, Goldner EM. The relationship between stealing and eating disorders: a review. Harv Rev Psychiatry. 1995;4:210-221.

78. Wiseman CV, Sunday SR, Halligan P, Korn S, Brown C, Halmi KA. Substance dependence and eating disorders: impact of sequence on comorbidity. Compr Psychiatry. 1999;5:332-336.

79. Lejoyeux M, Tassain V, Solomon J, Ades J. Study of compulsive buying in depressed patients. J Clin Psychiatry. 1997;4:169-173.

80. Carter FA, McIntosh VV, Joyce PR, et al. Role of exposure with response prevention in cognitive-behavioral therapy for bulimia nervosa: three-year follow-up results. Int J Eat Disord. 2003;2:127-135. 

81. Halmi,K. Eating disorders. In: Sadock BJ, Sadock VA, eds. Kaplan & Sadock’s Comprehensive Textbook of Psychiatry Seventh Edition on CD-ROM. New York, NY: Lippincott Williams & Wilkins; 2003:

82. Hughes PL, Wells LA, Cunningham CJ, Ilstrup DM. Treating bulimia with desipramine. A double-blind, placebo-controlled study. Arch Gen Psychiatry. 1986;2:182-186.

83. Agras WS, Dorian B, Kirkley BG, Arnow B, Bachman J. Imipramine in the treatment of bulimia: a double-blind controlled study. Int J Eating Disord. 1987;1:29-38.

84. Mitchell JE, Pyle RL, Eckert ED, Hatsukami D, Pomeroy C, Zimmerman R. A comparison study of antidepressants and structured intensive group psychotherapy in the treatment of bulimia nervosa. Arch Gen Psychiatry. 1990;2:149-157.

85. Fluoxetine Bulimia Nervosa Collaborative Study Group. Fluoxetine in the treatment of bulimia nervosa. A multicenter, placebo-controlled, double-blind trial. Arch Gen Psychiatry. 1992;2:139-147.

86. Faris PL, Kim SW, Meller WH, et al. Effect of decreasing afferent vagal activity with ondansetron on symptoms of bulimia nervosa: a randomised, double-blind trial. Lancet. 2000;9206:792-797.

87. Wermuth BM, Davis KL, Hollister LE, Stunkard AJ. Phenytoin treatment of the binge-eating syndrome. Am J Psychiatry. 1977;11:1249-1253.

88. Kaplan AS, Garfinkel PE, Darby PL, Garner DM. Carbamazepine in the treatment of bulimia. Am J Psychiatry. 1983;9:1225-1226.

89. Knable M. Topiramate for bulimia nervosa in epilepsy. Am J Psychiatry. 2001;2:322-323.

90. Felstrom A, Blackshaw S. Topiramate for bulimia nervosa with bipolar II disorder. Am J Psychiatry. 2002;7:1246-1247.

91. Hoopes SP, Reimherr FW, Kamin M, Karvois D, Rosenthal NE, Karim R. Topiramate treatment of bulimia nervosa. Presented at: the Annual Meeting of the American Psychiatric Society; May 2002; New Orleans, La.

92. Shank RP, Gardocki JF, Streeter AJ, Maryanoff BE. An overview of the preclinical aspects of topiramate: pharmacology, pharmacokinetics, and mechanism of action. Epilepsia. 2000;41(suppl 1):3-9.

93. Agras WS. Treatment of binge-eating disorder. In: Gabbard GO, ed. Treatment of Psychiatric Disorders. 3rd ed. Washington, DC: American Psychiatric Press; 2001:2209-2219.

94. Branson R, Potoczna N, Kral JG, Lentes KU, Hoehe MR, Horber FF. Binge eating as a major phenotype of melanocortin 4 receptor gene mutations. N Engl J Med. 2003;12:1096-1103.

95. Yanovski SZ, Nelson JE, Dubbert BK, Spitzer RL. Association of binge eating disorder and psychiatric comorbidity in obese subjects. Am J Psychiatry. 1993;10:1472-1479.

96. Arnold LM, McElroy SL, Hudson JI, Welge JA, Bennett AJ, Keck PE. A placebo-controlled, randomized trial of fluoxetine in the treatment of binge-eating disorder. J Clin Psychiatry. 2002;11:1028-1033.

97. Stunkard A, Berkowitz R, Tanrikut C, Reiss E, Young L. d-fenfluramine treatment of binge eating disorder. Am J Psychiatry. 1996;11:1455-1459.

98. Hudson JI, McElroy SL, Raymond NC, et al. Fluvoxamine in the treatment of binge-eating disorder: a multicenter placebo-controlled, double-blind trial. Am J Psychiatry. 1998;12:1756-1762.

99. McElroy SL, Casuto LS, Nelson EB, et al. Placebo-controlled trial of sertraline in the treatment of binge eating disorder. Am J Psychiatry. 2000;6:1004-1006.

100. Malhotra S, King KH, Welge JA, Brusman-Lovins L, McElroy SL. Venlafaxine treatment of binge-eating disorder associated with obesity: a series of 35 patients. J Clin Psychiatry. 2002;9:802-806.

101. Appolinario JC, Godoy-Matos A, Fontenelle LF, et al. An open-label trial of sibutramine in obese patients with binge-eating disorder. J Clin Psychiatry. 2002;1:28-30.

102. Alger SA, Schwalberg MD, Bigaouette JM, Michalek AV, Howard LJ. Effect of a tricyclic antidepressant and opiate antagonist on binge-eating behavior in normoweight bulimic and obese, binge-eating subjects. Am J Clin Nutr. 1991;4:865-871.

103. Shapira NA, Goldsmith TD, McElroy SL. Treatment of binge-eating disorder with topiramate: a clinical case series. J Clin Psychiatry. 2000;5:368-372.

104. McElroy SL, Arnold LM, Shapira NA, et al. Topiramate in the treatment of binge eating disorder associated with obesity: a randomized, placebo-controlled trial. Am J Psychiatry. 2003;2:255-261.

105. McElroy SL, Arnold LM, Shapira NA, Keck PE, Wu S-C, Hudson JI, Capece JA, Rosenthal N. Long-term use of topiramate in the treatment of binge eating disorder. Poster presented at: the American Psychiatric Association Annual Meeting; May 19, 2003; San Francisco, CA.

106. Goldman MJ. Kleptomania: making sense of the nonsensical. Am J Psychiatry. 1991;8:986-996.

107. Dannon PN. Topiramate for the treatment of kleptomania: a case series and review of the literature. Clin Neuropharmacol. 2003;1:1-4.

108. McElroy SL, Hudson JI, Pope HG, Keck PE. Kleptomania: clinical characteristics and associated psychopathology. Psychol Med. 1991;1:93-108.

109. Durst R, Katz G, Teitelbaum A, Zislin J, Dannon PN. Kleptomania: diagnosis and treatment options. CNS Drugs. 2001;3:185-195.

110. Chong SA, Low BL. Treatment of kleptomania with fluvoxamine. Acta Psychiatr Scand. 1996;4:314-315.

111. McElroy SL, Hudson JI, Pope H, Jr., Keck PE, Jr., Aizley HG. The DSM-III-R impulse control disorders not elsewhere classified: clinical characteristics and relationship to other psychiatric disorders. Am J Psychiatry. 1992;3:318-327.

112. McElroy SL, Pope HG, Jr., Hudson JI, Keck PE, Jr., White KL. Kleptomania: a report of 20 cases. Am J Psychiatry. 1991;5:652-657.

113. Burstein A. Fluoxetine-lithium treatment for kleptomania. J Clin Psychiatry. 1992;1:28-29.

114. Kmetz GF, McElroy SL, Collins DJ. Response of kleptomania and mixed mania to valproate. Am J Psychiatry. 1997;4:580-581.

115. Volberg RA. The prevalence and demographics of pathological gamblers: implications for public health. Am J Public Health. 1994;2:237-241.

116. McCormick RA, Russo AM, Ramirez LF, Taber JI. Affective disorders among pathological gamblers seeking treatment. Am J Psychiatry. 1984;2:215-218.

117. Specker SM, Carlson GA, Christenson GA, Marcotte M. Impulse control disorders and attention deficit disorder in pathological gamblers. Ann Clin Psychiatry. 1995;4:175-179.

118. Roy A, Adinoff B, Roehrich L, et al. Pathological gambling. A psychobiological study. Arch Gen Psychiatry. 1988;4:369-373.

119. Comings DE, Rosenthal RJ, Lesieur HR, et al. A study of the dopamine D2 receptor gene in pathological gambling. Pharmacogenetics. 1996;3:223-234.

120. Kim SW, Grant JE. An open naltrexone treatment study in pathological gambling disorder. Int Clin Psychopharmacol. 2001;5:285-289.

121. Kim SW, Grant JE, Adson DE, Shin YC. Double-blind naltrexone and placebo comparison study in the treatment of pathological gambling. Biol Psychiatry. 2001;11:914-921.

122. Grant JE, Kim SW. Effectiveness of pharmacotherapy for pathological gambling: a chart review. Ann Clin Psychiatry. 2002;3:155-161.

123. Kim SW. Opioid antagonists in the treatment of impulse-control disorders. J Clin Psychiatry. 1998;4:159-64.

124. Zimmerman M, Breen RB, Posternak MA. An open-label study of citalopram in the treatment of pathological gambling. J Clin Psychiatry. 2002;1:44-48.

125. Blanco C, Petkova E, Ibanez A, Saiz-Ruiz J. A pilot placebo-controlled study of fluvoxamine for pathological gambling. Ann Clin Psychiatry. 2002;1:9-15.

126. Hollander E, DeCaria CM, Mari E, et al. Short-term single-blind fluvoxamine treatment of pathological gambling. Am J Psychiatry. 1998;12:1781-1783.

126. Kim SW, Grant JE, Adson DE, Shin YC, Zaninelli R. A double-blind placebo-controlled study of the efficacy and safety of paroxetine in the treatment of pathological gambling. J Clin Psychiatry. 2002;6:501-507.

127. Pallanti S, Baldini Rossi N, Sood E, Hollander E. Nefazodone treatment of pathological gambling: a prospective open-label controlled trial. J Clin Psychiatry. 2002;11:1034-1039.

128. Pallanti S, Quercioli L, Sood E, Hollander E. Lithium and valproate treatment of pathological gambling: a randomized single-blind study. J Clin Psychiatry. 2002;7:559-564.


Print Friendly 

Stephen D. Silberstein, MD, FACP
Primary Psychiatry. 2003;10(10):64-71

Dr. Silberstein is professor of neurology and director of the Jefferson Headache Center at Thomas Jefferson University Hospital in Philadelphia, Pennsylvania.

Disclosure: Dr. Silberstein is a consultant for Eli Lilly; has received research support and/or unrestricted educational grants from Abbott, Allergan, AstraZeneca, Bristol-Myers Squibb, Eli Lilly, GlaxoSmithKline, Janssen, Johnson & Johnson, Merck, Ortho-McNeil, Parke Davis, Pfizer, Robert Wood Johnson, UCB Pharma, and Vernalis; and is on the advisory panel and/or speaker’s bureau of Abbot, Allergan, AstraZeneca, Elan, GlaxoSmithKline, Johnson & Johnson, and Merck.

Please direct all correspondence to: Stephen D. Silberstein, MD, FACP, Jefferson Headache Center, Thomas Jefferson University Hospital; 111 South 11th St; Gibbon Building, Suite 8130; Philadelphia, PA 19107; Tel: 215-955-2243; Fax: 215-955-1960; E-mail:

Focus Points

Migraine is associated with several neurologic and psychiatric disorders which produce changes in mood, thought, and behavior attributed to poorly-specified alterations of brain function.
A bidirectional causal influence occurs between migraine and depression, suggesting that migraine and psychiatric comorbidities may share overlapping etiologies.
Pharmacologic treatment of migraine may be acute (abortive, symptomatic) and administered 2–3 days/week, or preventive (prophylactic) and given on a daily basis.

Many psychiatric drugs, including antiepileptics and antipsychotics, may be effective in the treatment of migraine headaches, especially in patients with comorbid disorders.


How can the clinician best recognize and treat migraine while taking into account the high prevalence of comorbid illness? A migraine is an episodic headache disorder typically experienced in multiple phases, with warning signs coming hours or even days before the migraine itself. Migraine severity ranges from moderate to marked and the pain is often aggravated by routine physical activity or simple head movement. Migraine is associated with a number of comorbid neurologic and psychiatric disorders, and is linked to traditional psychiatric conditions on many levels. A bidirectional causal influence occurs between migraine and depression. The relationship is due to the presence of the disorder, not attack frequency. Migraines and psychiatric conditions may in fact share underlying neural mechanisms. Pharmacologic treatment of migraine may be acute (abortive, symptomatic) or preventive (prophylactic); patients experiencing frequent severe headaches often require both approaches. Medications used in acute headache treatment include analgesics, antiemetics, anxiolytics, nonsteroidal antiinflammatory drugs, ergots, steroids, major tranquilizers, narcotics, and, more recently, selective serotonin agonists (triptans). Preventive medications with the best documented efficacy are β-blockers, divalproex, topiramate, and amitriptyline. Treatment choice is made based on a drug’s proven efficacy and tolerability, the physician’s knowledge and experience, the patient’s preferences and headache profile, and the presence or absence of coexisting disorders.


Both headache and psychopathology can be divided into primary and secondary disorders. For the primary headaches (migraine, cluster, tension-type headache [TTH]), the headache disorder itself is the problem. These disorders are analogous to the major idiopathic psychiatric disorders.1 In secondary headache the symptoms are due to an underlying condition, such as a metabolic derangement, a brain tumor, a stroke, or other forms of structural brain disease. Secondary headache disorders are therefore analogous to the organic psychiatric syndromes.


Migraine is an episodic headache disorder often accompanied by neurologic, gastrointestinal, autonomic, and psychologic changes. The International Headache Society (IHS)distinguishes seven categories of migraine, the most important of which are migraine without aura, formerly referred to as common migraine (Table 1) and migraine with aura, formerly referred to as classic migraine (Table 2). The aura is the complex of focal neurologic symptoms that initiates or accompanies an attack.2 An individual patient may have headache without aura, headache with aura, and aura without headache. The migraine attack can be divided into four phases: the premonitory phase, the aura, the headache phase, and the postdrome.3


Premonitory Phase

Premonitory (prodromal) phenomena occur in approximately 60% of migraineurs, often hours to days before the onset of headache. These phenomena include psychologic, neurologic, constitutional, and autonomic features.3 Psychologic symptoms include depression, euphoria, irritability, restlessness, mental slowness, hyperactivity, fatigue, and drowsiness. Neurologic phenomena include photophobia, phonophobia, and hyperosmia. Constitutional symptoms include a stiff neck, a cold feeling, sluggishness, increased thirst, increased urination, anorexia, diarrhea, constipation, fluid retention, and food cravings. According to Griffin and colleagues4 some patients report a poorly characterized feeling that a migraine attack is coming. Migraineurs who reported premonitory symptoms were able to accurately predict their full-blown headaches 72% of the time. The most common premonitory symptoms were feeling tired/weary (72%), difficulty concentrating (51%), and stiff neck (50%). Poor functioning commonly predicted headache.


The migraine aura is characterized by focal neurologic symptoms that typically precede but sometimes accompany an attack. Approximately 20% of migraine sufferers experience auras. Most aura symptoms evolve slowly over 5–20 minutes and usually <60 minutes. The aura can be characterized by visual, sensory, or motor phenomena, alone or in combination. Auras may also involve language or brainstem disturbances. Headache usually occurs within 60 minutes of the end of the aura.

The most common migraine aura is visual. The visual aura often has a hemianoptic distribution and includes both positive (scintillations, fortification spectra, photopsia) and negative (scotomata) visual features that often occur together. Elementary visual disturbances can occur; these include colorless scotomata or positive visual phenomena, such as photopsia (the sensation of unformed flashes of light or sparkles before the eyes) or phosphenes (an objective visual sensation that appears with the eyes closed and in the absence of light). Simple flashes, specks, or crude or uniform pseudohallucinations of geometric forms (points, stars, lines, curves, circles, sparks, flashes, or flames) may occur and may be single or number in the hundreds. They may move rapidly across the visual field, sometimes crossing the midline and often preceding a scotoma. More complicated pseudohallucinations include the teichopsia or fortification spectrum, which is the most characteristic visual aura and is almost diagnostic of migraine. An arc of scintillating lights often begins in central vision, sometimes forms a herringbone-like pattern, and may expand to encompass an increasing portion of a visual hemifield. Other complex positive features, such as bright geometric lights, may occur. Objects may occasionally appear to change in size or shape.5,6

Numbness or tingling (paresthesias) over one side of the face and in the ipsilateral hand or arm are the most common of the somatosensory phenomena. Hemiparesis may occur, and, if the dominant hemisphere is involved, dysphasia or aphasia may develop. Olfactory hallucinations have been reported. Odors tend to be unpleasant and can last from 5 minutes to 24 hours. This may be preceded or followed by other aura symptoms. Anxiety, deja vu, and jamais vu have been reported and are presumably of temporal lobe origin.3 Auras may occur repeatedly, even many times an hour, for several months. In cases where persistent auras are present, they are documented and referred to as the “migraine aura status.”6

Headache Phase

The typical migraine headache is unilateral and throbbing. The severity ranges from moderate to marked and the pain is often aggravated by routine physical activity or simple head movement. The pain may be bilateral at its onset (which it is in 40% of patients); it may remain bilateral throughout the attack or begin on one side and become generalized.7 The headache can occur at any time of the day or night, but it occurs most frequently upon arising.7 The onset is usually gradual; the pain peaks and then subsides, lasting usually between 4 and 72 hours in adults and 2 and 48 hours in children.5

The pain of migraine is invariably accompanied by other features. Anorexia is common, although food cravings can occur. Nausea occurs in as many as 90% of migraineurs and vomiting occurs in about one third.8,9 Many patients experience sensory hyperexcitability manifested by photophobia, phonophobia, and osmophobia, and seek a dark, quiet environment.7,10 Other systemic symptoms, including blurry vision, nasal stuffiness, anorexia, hunger, tenesmus, diarrhea, abdominal cramps, polyuria (followed by decreased urinary output after the attack), facial pallor (or, less commonly, redness), sweating, and sensations of heat or cold, may be noted during the headache phase. Impaired concentration is common; memory impairment occurs less frequently. Depression, fatigue, anxiety, nervousness, and irritability are common.


The headache of migraine is often followed by the postdrome. During the postdrome, patients may have many of the symptoms that occurred during the prodrome. The patient may feel tired, washed out, irritable, and listless, and may have impaired concentration, scalp tenderness, or mood changes. Some people feel unusually refreshed or euphoric after an attack, whereas others note depression and malaise. Recurrent psychosis has been reported, with the so-called migraine madness lasting up to 4 weeks and manifested by morbid visual hallucinations and delusions, including the belief that homes or people have been replaced by exact doubles (reduplicative paramnesia or Capgras syndrome).6

Unusual Migraine Auras

The visions of Hildegard of Bingen, an 11th century mystic, have been attributed in part to her migrainous scintillating scotomata. Characteristic of the visions that she and other visionary prophets saw were working, boiling, or fermenting lights. Hildegard indicates that her visions often contained elements of blinding lights in the following manner11:

I saw a great star most splendid and beautiful, and with it an exceeding multitude of falling sparks which with the star followed southward… But sometimes I behold within this light another light which I name “the living light itself

In addition, in the book of the prophet Ezekiel, Ezekiel describes his vision thus12:

I saw a storm wind coming from the north, a vast cloud with flashes of fire and brilliant light about it; and within was a radiance like brass, glowing in the heart of the flames… As I looked at the living creatures, I saw wheels on the ground, one beside each of the four. The wheels sparkled like topaz, and they were all alike: in form and working they were like a wheel inside a wheel, and when they moved in any of the four directions they never swerved in their course.


Migraine prevalence is similar and stable in Western countries and the United States.13 In the US, 17.6% of women and 6% of men had experienced one migraine attack in the previous year.14 Migraine prevalence varies by age, gender, race, and income. Before puberty, migraine prevalence is approximately 4%.15 After puberty, prevalence increases more rapidly in girls than in boys. In the US, migraine prevalence decreases as household income increases.14,16,17

Comorbidity of Migraine

Migraine is associated with a number of neurologic diseases (epilepsy, stroke) and psychiatric disorders (depression, mania, anxiety, and panic). The term comorbidity, coined by Feinstein,18 originally referred to coexistent conditions in clinical trials. For the purpose of this article, comorbidity refers to an association between two disorders that is more than coincidental.19

Migraine comorbidity is important for a number of reasons.19 Co-occurring diseases can complicate diagnosis, influence treatment choices, and give clues about pathophysiology. For example, depression, anxiety disorders, or epilepsy, can all cause headaches, while headaches can change mood and behavior.20,21 However, comorbidity is more than a problem of differential diagnosis; the real challenge for the neuropsychiatrist is recognizing that more than one disease is present.19 Because of the high rate of comorbid disorders in migraine, the principle of diagnostic parsimony (attributing all symptoms to one diagnosis)does not apply. The presence of migraine should increase the suspicion that another disorder may be present. Conversely, patients with depressive disorders often have medically undiagnosed migraine, and there may be a tendency to attribute the migrainous manifestations to the depressive disorder.22

Antidepressants can be used to treat both migraine and depression.23,24 When migraine occurs with manic depressive illness or epilepsy, topiramate or divalproex sodium can be used to treat both conditions.3 However, comorbid illnesses may impose therapeutic limitations. Some treatments may be relatively contraindicated in individuals who have more than one disease. For example, β-blockers may be a less desirable option when treating a migraine patient who also has depression.

Psychopathology and Psychiatric Disorders

A number of studies have examined the relationship between migraine and specific psychiatric disorders.9 Several clinic-based studies have reported an increased prevalence of migraine in patients with major depressive disorder (MDD) and an increased prevalence of MDD in patients with migraine.25-27 Three population-based studies have examined a wide range of psychiatric disorders in addition to MDD.9,24,28-30

Merikangas and colleagues24 reported on the association of migraine with specific psychiatric disorders in a random sample of 457 adults 27–28 years of age in Zurich, Switzerland. Persons with migraine (n=61) were found to have increased 1-year rates of affective and anxiety disorders. Specifically, the odds ratio for MDD (2.2, 95% CI 1.1-4.8), bipolar spectrum disorders (2.9, 95% CI 1.1-8.6), generalized anxiety disorder (2.7, 95% CI 1.5-5.1), panic disorder (3.3, 95% CI 0.8-13.8), simple phobia (2.4, 95% CI 1.1-5.1), and social phobia (3.4, 95% CI 1.1-10.9), were significantly higher in persons with migraine compared to those without migraine.

Migraine with MDD was frequently complicated by an anxiety disorder. Merikangas and colleagues24 suggest that in persons with all three disorders, the onset of anxiety generally precedes the onset of migraine, whereas the onset of MDD most often follows the onset of migraine.

Epidemiologic studies support the association between migraine and MDD previously reported in clinic-based studies. The prospective data indicate that the observed cross-sectional or lifetime association between migraine and MDD could result from a bidirectional influence, from migraine to subsequent onset of MDD and from MDD to first migraine attack (Table 3). Furthermore, these epidemiologic studies indicate that persons with migraine have increased prevalence of bipolar disorder, panic disorder, and one or more anxiety disorders.9,16,23,31

It has been proposed that MDD in patients with migraine might represent a psychologic reaction to repeated, disabling migraine attacks. Migraine generally has an earlier mean age of onset than MDD in the normal population and in patients with comorbid disease. Nonetheless, the bidirectional influence of each condition on the risk for the onset of the other is incompatible with the simple causal model.9 Furthermore, Breslau and Davis31 reported that the increased risk for first episode of MDD (and/or panic disorder) did not vary by the proximity of prior migraine attacks. These findings lessen the plausibility that the migraine-depression association results from the demoralizing experience of recurrent and disabling headaches, suggesting instead that their association might reflect shared etiologies.

Breslau and colleagues32 recently examined migraine and depression comorbidity in a large-scale epidemiologic study, the Detroit Area Study of Headache. The study comprised three groups: persons with migraine (n=536), persons with other severe headaches of comparable pain severity and disability (n=162), and matched controls with no history of severe headache (n=586).

The results suggest the possibility that different causal pathways might account for the comorbidity of MDD in these two headache categories. The results for migraine suggest shared causes, whereas those for other headaches of comparable severity suggest a causal effect of headache on depression.

Another line of evidence comes from a study of a biologic marker of depression in migraineurs. Jarman and colleagues33 administered the tyramine test to 40 migraine patients, 16 of whom had a lifetime history of MDD. Low tyramine conjugation, a trait marker for endogenous depression, was strongly associated with a lifetime history of MDD in subjects with comorbid disease, regardless of their current psychiatric status. The authors argue that the association of the trait marker with MDD in migraineurs rules out the possibility that the depression is a psychologic reaction to migraine attacks.9

Personality Characteristics and Psychopathology

The relationship between migraine and psychopathology has been discussed far more often than it has been systematically studied.9 Over the years, many studies have focused on particular personality traits of migraineurs. The basic assumptions are: (1) migraineurs share common personality traits; (2) these traits are enduring and measurable; and (3) these traits differentiate migraineurs from controls.34

The notion of a migraine personality first grew out of clinical observations of the highly selected patients seen in subspecialty clinics.9 In 1934, Touraine and Draper35 reported that migraineurs were deliberate, hesitant, insecure, detailed, perfectionistic, sensitive to criticism, and deeply frustrated emotionally. They were said to lack warmth and to have difficulty making social contacts. Wolff36 found migraineurs to be rigid, compulsive, perfectionistic, ambitious, competitive, chronically resentful, and unable to delegate responsibility.

Most investigations have used psychometric instruments, such as the Minnesota Multiphasic Personality Inventory (MMPI) or the Eysenck Personality Questionnaire (EPQ). The EPQ is a well-standardized measure which includes four scales: psychoticism, extroversion, neuroticism, and lie.

Brandt and colleagues37 used the Washington County Migraine Prevalence Study to conduct the first population-based case-control study of personality in migraine (Table 4).6,30,37,38 Over 10,000 subjects 12–29 years of age were selected using random-digit dialing and received a diagnostic telephone interview. A sample of subjects who met the criteria for migraine with or without aura [n=162] were compared with subjects without migraine. Each subject received the EPQ, the 28-item version of the General Health Questionnaire, and a question about headache laterality. Migraineurs scored significantly higher than controls on the neuroticism scale of the EPQ scale, indicating that they were more tense, anxious, and depressed than the control group. In addition, women with migraine scored significantly higher than controls on the psychoticism scale of the EPQ, indicating that they were more hostile, less interpersonally sensitive, and out of step with their peers. Overall, studies that used the EPQ or similar personality measures and compared migraineurs to nonmigraine controls have generally reported an association between migraine and neuroticism.9,38-41

Many investigators42-45 have used the MMPI to investigate the migraine personality. These studies have been limited by several factors.1 MMPI studies have usually been clinic-based, limiting their generalizability and creating opportunities for selection bias. Most have not used control groups, relying instead on historical norms. Many have not used explicit diagnostic criteria for migraine. Despite these limitations, most studies show elevation of the neurotic triad, although this is not statistically significant (Table 5).9,42,43,45-50


Studies of migraine and personality have generally not controlled for drug use, headache frequency, and headache-related disability. Furthermore, they have not controlled for major psychiatric disorders (such as MDD or panic disorder), which occur more commonly in migraineurs. The association between major psychiatric disorders and personality disorders may confound the assessment of the relationships between these disorders and migraine. Neuroticism, in particular, is associated with depression and anxiety, which occur with increased prevalence in migraineurs. Differences in neuroticism across studies might reflect variations in the role of comorbid psychiatric disease. The available data suggest that migraineurs may be more neurotic than nonmigraineurs. The stereotypical rigid, obsessional migraine personality might reflect the selection bias of a distinct subtype of migraine that is more likely to be seen in the clinic.

Breslau and Andreski51 examined the association between migraine and personality, taking into account history of co-occurring psychiatric disorders. The results suggest that migraine sufferers might be more vulnerable to psychopathology and poor adjustment to their medical condition. This also suggests that the association between migraine and neuroticism is not attributable to comorbid depression or anxiety disorders.

In a later report, Breslau and colleagues52 presented findings from prospective data on the migraine-neuroticism association from their epidemiologic study of young adults. Controlling for MDD and anxiety disorders at baseline, women scoring in the highest quartile of the neuroticism scale were nearly three times more likely to develop migraine than those scoring in the lowest quartile during a 5-year follow-up. In men, neuroticism did not predict migraine, although the small number of incidence cases in men precluded reliable estimates of the risk for migraine associated with neuroticism.


Pharmacologic treatment of migraine may be acute (abortive, symptomatic) or preventive (prophylactic). Patients experiencing frequent severe headaches often require both approaches. Symptomatic treatment attempts to abort (stop the progression of) or reverse a headache once it has started. Preventive therapy is given on a daily basis, even in the absence of a headache, to reduce the frequency and severity of anticipated attacks. Symptomatic treatment is appropriate for most acute attacks and should be used no more than 2–3 days/week. If attacks occur more frequently, treatment strategies should focus on decreasing the attack frequency.

Medications used in acute headache treatment include analgesics, antiemetics, anxiolytics, NSAIDs, ergots, steroids, major tranquilizers, narcotics, and, more recently, selective serotonin agonists (triptans). One or more of these medications can be used for headaches of differing severities.5 Preventive treatments include a broad range of medications, most notably b-blockers, calcium channel blockers, antidepressants, and antiepileptic drugs (AEDs) (Table 6).

Preventive Treatment

Preventive medications are meant to reduce attack frequency, duration, or severity.5,53 According to the US Headache Consortium Guidelines,54

indications for preventive treatment include:

• Migraine that significantly interferes with the patient’s daily routine despite acute treatment

• Failure of, contraindication to, or troublesome adverse events from acute medications

• Acute medication overuse

• Very frequent headaches (>2/week), indicating risk of medication overuse

• Patient preference

• Special circumstances, such as hemiplegic migraine or attacks with a risk of permanent neurologic injury

Preventive medication groups include β-adrenergic blockers, antidepressants, calcium-channel antagonists, serotonin antagonists, AEDs, and nonsteroidal antiinflammatory drugs. Choice is based on efficacy, adverse events, and coexistent and comorbid conditions (Table 6).55 The medication should be started at a low dose and increased slowly until therapeutic effects develop or the ceiling dose is reached. A full therapeutic trial may take 2–6 months. Acute headache medications should not be overused. If headaches are well controlled, medication can be tapered and discontinued, but dose reduction may provide a better risk:benefit ratio. Women of childbearing potential should be on adequate contraception.

Behavioral and psychological interventions used for prevention include relaxation training, thermal biofeedback combined with relaxation training, electromyography biofeedback, and cognitive-behavioral therapy.56 These measures are effective as monotherapy, but they are more effective when used in conjunction with pharmacologic management.


Propranolol, nadolol, atenolol, metoprolol, and timolol are effective pharmacologic treatments of migraine.57 As their relative efficacy has not been established, choice is based on b-selectivity, convenience, adverse events, and patients’ tolerability.5 b-blockers can produce behavioral adverse events, such as drowsiness, fatigue, lethargy, sleep disorders, nightmares, depression, memory disturbance, and hallucinations, and should be avoided when patients are depressed. Decreased exercise tolerance limits their use by athletes. Less common adverse events include impotence, orthostatic hypotension, and bradycardia. β-blockers are relatively contraindicated for patients with congestive heart failure, asthma, Raynaud’s disease, and insulin-dependent diabetes.



The tricyclic antidepressant amitriptyline is the only antidepressant with fairly consistent support for efficacy in the treatment of migraine,57 although there is one positive trial for fluoxetine.58 Adverse events associated with antidepressants include increased appetite, weight gain, dry mouth, sedation, and sexual dysfunction59; cardiac toxicity and orthostatic hypotension occur occasionally.58 Antidepressants are especially useful for patients with comorbid depression and anxiety disorders.


Divalproex sodium and sodium valproate, in both immediate and extended release formulations, are effective in the treatment of migraine.57 The most frequent adverse events associated with their use are nausea (42%), alopecia (31%), tremor (28%), asthenia (25%), dyspepsia (25%), somnolence (25%), and weight gain (19%).60 Hepatotoxicity and pancreatitis are the most serious adverse events, but irreversible hepatic dysfunction is extremely rare in adults. Baseline liver function studies should be obtained, but follow-up studies of adults on AED monotherapy for headaches are probably not necessary.61 Divalproex carries a high risk of congenital abnormality.

Gabapentin was shown to be effective in a placebo-controlled, double-blind trial.62 Migraine attack frequency was reduced by 50% in about one-third of patients. The most common adverse events were dizziness or giddiness and drowsiness.

Topiramate, a D-fructose derivative, has been associated with weight loss. In two large, double-blind, placebo-controlled, multicenter trials, topiramate, both 100 and 200 mg, was effective in reducing migraine attack frequency by 50% in half of the patients.63,64 Dropouts due to adverse events were common in the topiramate groups, but did not affect statistical significance.

To circumvent the contraindications to β-blockers, divalproex and topiramate may be useful in migraine patients with comorbid epilepsy, anxiety disorder, bipolar disorder, depression, Raynaud’s disease, asthma, and diabetes.

Setting Treatment Priorities

The goals of treatment are to relieve or prevent the pain and associated symptoms of migraine and to optimize the patient’s ability to function normally (Table 6). The preventive medications with the best documented efficacy are β-blockers, amitriptyline (some believe its efficacy is less documented), divalproex, and topiramate. Medication choice is made based on a drug’s proven efficacy, the physician’s informed belief about medications not yet evaluated in controlled trials, the drug’s adverse events, the patient’s preferences and headache profile, and the presence or absence of coexisting disorders.6


The primary headache disorders are linked to traditional psychiatric diseases on many levels. Diagnosis depends upon self-reported symptoms and other clinical features. Both families of disorders produce changes in mood, thought, and behavior attributed to poorly-specified alterations of brain function. Epidemiologic studies show that migraine and mood disorders are powerfully associated. The bidirectionality of risk is incompatible with simple causal models; depression cannot merely be a response to pain, since the onset of depression may precede the onset of migraine. Comorbidity is most likely, at least in part, a consequence of overlapping neural mechanisms. This notion is supported by the many drugs that are effective in the treatment of both headache and psychiatric disease. Antidepressants, AEDs, lithium carbonate, and neuroleptics, are part of the shared therapeutic armamentarium. Many of the drugs developed in psychiatry may find application in the treatment of headache, especially in patients with comorbid disease. PP


1. Silberstein SD, Lipton RB, Breslau N. Neuropsychiatric aspects of primary headache disorders. In: Yudofsky SC, Hales RE, eds. Neuropsychiatry and Clinical Neurosciences. 4th ed. Washington: American Psychiatric Publishing Inc; 2002:451-488.

2. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Headache Classification Committee of the International Headache Society. Cephalalgia. 1988;8(suppl 7):1-96.

3. Silberstein SD, Lipton RB, Goadsby PJ. Migraine: diagnosis and treatment. In: Headache in Clinical Practice. Oxford, UK: Isis Medical Media; 1998:61-90.
4. Giffin NJ, Ruggiero L, Lipton RB, et al. Premonitory symptoms in migraine: an electronic diary study. Neurology. 2003;60:935-940.

5. Silberstein SD, Saper JR, Freitag FG. Migraine: diagnosis and treatment. In: Wolff’s Headache and Other Head Pain. 7th ed. New York: Oxford University Press; 2001:121-237.

6. Silberstein SD, Young WB. Migraine aura and prodrome. Semin Neurol. 1995;15:175-182.

7. Selby G, Lance JW. Observation on 500 cases of migraine and allied vascular headaches. J Neurol Neurosurg Psychiatry. 1960;23:23-32.

8. Lipton RB, Stewart WF, Celentano DD, Reed ML. Undiagnosed migraine headaches: a comparison of symptom-based and reported physician diagnosis. Arch Intern Med. 1992;156:1273-1278.

9. Silberstein SD. Migraine symptoms: results of a survey of self-reported migraineurs. Headache. 1995;35:387-396.

10. Drummond PD. A quantitative assessment of photophobia in migraine and tension headache. Headache. 1986;26:465-469.

11. Singer C. The visions of Hildegarde of Bingen. In: From Magic to Science. New York, NY: Dover; 1958.

12. Ezekiel 1:5-6.

13. Stewart WF, Shechter A, Rasmussen BK. Migraine prevalence. A review of population-based studies. Neurology. 1994;44(suppl):S17-S23.

14. Stewart WF, Lipton RB, Celentano DD, Reed ML. Prevalence of migraine headache in the United States: relation to age, income, race, and other sociodemographic factors. JAMA. 1992;267:64-69.

15. Lipton RB, Hamelsky SW, Stewart WF. Epidemiology and impact of headache. In: Silberstein SD, Lipton RB, Dalessio DJ, eds. Wolff’s Headache and Other Head Pain. 7th ed. New York, NY: Oxford University Press; 2001:85-107.

16. Breslau N, Davis GC, Schultz LR, Peterson EL. Migraine and major depression: a longitudinal study. Headache. 1994;7:387-393.

17. Lipton RB, Stewart WF, Diamond S, Diamond ML, Reed M. Prevalence and burden of migraine in the United States: data from the American Migraine Study II. Headache. 2001;41:646-657.

18. Feinstein AR. The pretherapeutic classification of comorbidity in chronic disease. J Chronic Dis. 1970;23:455-468.

19. Lipton RB, Silberstein SD. Why study the comorbidity of migraine. Neurology. 1994;44:4-5.

20. Welch KM. Relationship of stroke and migraine. Neurology. 1994;44(suppl):S33-S36.

21. Ottman R, Lipton RB. Comorbidity of migraine and epilepsy. Neurology. 1994;44:2105-2110. 

22. Lipton RB, Stewart WF. Migraine in the United States: a review of epidemiology and health care use. Neurology. 1993;43(suppl):S6-S10.

23. Breslau N, Merikangas K, Bowden CL. Comorbidity of migraine and major affective disorders. Neurology. 1994;44(suppl):S17-S22.

24. Merikangas KR, Angst J, Isler H. Migraine and psychopathology. Results of the Zurich cohort study of young adults. Arch Gen Psychiatry. 1990;47:849-853.

25. Marchesi C, De Ferri A, Petrolini N, et al. Prevalence of migraine and muscle tension headache in depressive disorders. J Affect Disord. 1989;16:33-36.

26. Morrison DP, Price WH. The prevalence of psychiatric disorder among female new referrals to a migraine clinic. Psychol Med. 1989;19:919-925.

27. Merikangas KR, Risch NJ, Merikangas JR, Weissman MM, Kidd KK. Migraine and depression: association and familial transmission.

J Psychiatr Res. 1988;22:119-129

28. Stewart WF, Linet MS, Celentano DD. Migraine headaches and panic attacks. Psychosom Med. 1989;51:559-569.

29. Stewart WF, Shechter A, Liberman J. Physician consultation for headache pain and history of panic: results from a population-based study. Am J Med. 1992;92(suppl):35S-40S.

30. Merikangas KR, Merikangas JR, Angst J. Headache syndromes and psychiatric disorders: association and familial transmission.
J Psychiatr Res. 1993;27:197-210.

31. Breslau N, Davis GC. Migraine, major depression and panic disorder: a prospective epidemiologic study of young adults. Cephalalgia. 1992;12:85-90.
32. Breslau N, Schultz LR, Stewart WF, Lipton RB, Lucia VC, Welch KM. Headache and major depression: is the association specific to migraine? Neurology. 2000;54:308-313.

33. Jarman J, Fernandez M, Davies PT, et al. High incidence of endogenous depression in migraine: confirmation by tyramine test.
J Neurol Neurosurg Psychiatry. 1990;53:573-575.

34. Schmidt FN, Carney P, Fitzsimmons G. An empirical assessment of the migraine personality type. J Psychosom Res. 1986;30:189-197.

35. Touraine GA, Draper G. The migrainous patient: a constitutional study. J Nerv Ment Dis. 1934;80:183-204.

36. Wolff HG. Personality features and reactions of subjects with migraine. Arch Neurol Psychiatry. 1937;37:895-921.

37. Brandt J, Celentano D, Stewart W, Linet M, Folstein MF. Personality and emotional disorder in a community sample of migraine headache sufferers. Am J Psychiatry. 1990;147:303-308.

38. Rasmussen BK. Migraine and tension-type headache in a general population: psychosocial factors. Int J Epidemiol. 1992;21:1138-1143.

39. Passchier J, Orlebeke JF. Headaches and stress in schoolchildren: an epidemiological study. Cephalalgia. 1985;5:167-176.

40. Passchier J, Helm-Hylkema H, Orlebeke JF. Personality and headache type: a controlled study. Headache. 1984;24:140-146.
41. Philips C. Headache and personality. J Psychosom Res. 1976;20:535-542.

42. Invernizzi G, Gala C, Buono M, Cittone L, Tavola T, Conte G. Neurotic traits and disease duration in headache patients. Cephalalgia. 1989;9:173-178.

43. Sternbach RA, Dalessio DJ, Kunzel M, Bowman GE. MMPI patterns in common headache disorders. Headache. 1980;20:311-315.

44. Kudrow L, Sutkus BJ. MMPI pattern specificity in primary headache disorders. Headache. 1979;19:18-24.

45. Weeks R, Baskin S, Rapoport A, Sheftell F, Arrowsmith F. A comparison of MMPI personality data and frontalis electromyographic readings in migraine and combination headache patients. Headache. 1983;23:75-82.


Print Friendly 

Primary Psychiatry. 2003;10(10):24-26


FDA Approves Bupropion Extended-Release for Major Depressive Disorder

The United States Food and Drug Administration approved bupropion hydrochloride extended-release (Wellbutrin XL, GlaxoSmithKline) for the treatment of adult major depressive disorder (MDD) in September.

Bupropion XL should be initiated at a dosage of 150 mg/day in the morning, with an increase to 300 mg/day within 4 days. For patients who do not improve at 300 mg/day within 4 weeks, the dosage can be titrated to a maximum of 450 mg/day. Patients must wait 24 hours between each dosage.

Bupropion XL is associated with a dose-related risk of seizures. It should not be taken with any other formulations of bupropion, including sustained-release (SR) or immediate-release.

Bupropion XL has a lower risk of weight gain and sexual side effects than other forms of the drug. The adverse-event profile for bupropion XL is similar to that of bupropion SR and includes headache, dry mouth, nausea, insomnia, and sweating. —CN

Antipsychotic Use Linked to Increased Risk of Diabetes

Diabetes affects approximately 12 million Americans and has been found in two forms; type 1, common in juveniles, and type 2, mostly found in adults. Diabetes has been found to be associated with antipsychotic use in some patients. Case studies have suggested that some atypical antipsychotics may induce type 2 diabetes.

Dr. Francesca Cunningham, of the University of Illinois in Chicago, and colleagues, studied the risk of diabetes in patients taking antipsychotic medications for schizophrenia. They studied 19,878 patients, 5,981 of whom were taking olanzapine, 5,901 of whom were taking risperidone, and 977 of whom were taking quetiapine. Only 110 patients were taking clozapine, thus making it too small of a sample size to conduct a head-to-head comparison of that drug. They analyzed prescription data collected between October 1998 and September 2001. For study inclusion, patients had to be free of antipsychotics for 3 months prior to the study and had to initiate therapy due to the study. Patients with a history of diabetes were excluded from the study.

Cunningham and colleagues found an increased risk for diabetes in patients taking any of the three antipsychotics studied. For example, there was a 27% greater risk of developing diabetes in patients taking olanzapine, a 49% greater chance for diabetes in patients taking risperidone, and a 48% greater chance for diabetes in patients taking quetiapine.

Cunningham and colleagues also determined the hazard ratio by adjusting the patient population based on gender, race, marital status, diabetogenic agents, diabetic screening panels, and age (Table). The adjusted hazard ratio was found to be 1.50 for olanzapine, 1.47 for risperidone, and 1.54 for quetiapine.

A follow-up study is currently comparing the risk of diabetes among patients taking atypical antipsychotics. The study will enroll 3,000 patients from the initial study and conduct another medical chart review. —CN  

(International Society for Pharmaco-epidemiology Meeting; August 2003; Philadelphia, PA)

OCD Symptoms May Be Common in New Parents

Obsessive distressing thoughts are common in both mothers and fathers after the birth of a baby, according to a new study surveying parents of newly born infants. While symptoms of obsessive-compulsive disorder (OCD) have been shown to be higher than expected in postpartum females, the overall prevalence these symptoms postpartum remains unknown. The current study examined the presence and phenomenology of obsessive intrusive thoughts, images, and impulses in a large sample including both male and female parents.

Jonathan S. Abramowitz, MD, and colleagues at the Mayo Clinic Department of Psychiatry and Psychology in Rochester, Minnesota, mailed 6-page surveys to 300 females who had given birth between December 2001 and April 2002. All women had experienced full-term uncomplicated deliveries and had no history of delusional disorders, schizophrenia, or bipolar disorder. An identical survey was mailed to the father of each infant. Participants were asked to provide up to three upsetting intrusive thoughts they had experienced in response to several demographic items. For example, for the item “burping the baby,” one might say “I sometimes think about what would happen if I hit her too hard.”

Severity of intrusive thoughts was assessed with the following four items from the Yale-Brown Obsessive Compulsive Scale: (A) time per day spent on intrusive thoughts (rated as 0=none, 1=≤1 hour, 2=1–3 hours, 3=3–8 hours, or 4=≥8 hours); (B) degree to which intrusive thoughts interfered with functioning socially or at work (rated as 0=none to 4=severe); (C) degree of distress from intrusive thoughts (rated as 0=none to 4=severe); and (D) degree of control over intrusive thoughts (rated as 0=completely to 4=not at all). Participants were also asked about their history of OCD and were assessed for depression using the Center for Epidemiological Studies Depression scale (CES-D).

Approximately 66% of survey completers (20%) reported intrusive unwanted thoughts. There was no significant difference in percentage of males versus females reporting such thoughts.

Seven categories of intrusive thoughts were derived: (1) suffocation or sudden infant death syndrome, eg, maybe the baby rolled over and cannot breathe; (2) accidents, eg, the neighbor’s dog may attack the baby; (3) ideas or urges of intentional harm, eg, will the baby be brain damaged if I throw her out the window?; (4) losing the infant, eg, someone may kidnap the baby at the grocery store; (5) severe illness, eg, the baby may have cerebral palsy; (6) inappropriate sexual thoughts, eg, a thought about the baby’s genitals; (7) contamination, eg, the baby may suffer microbiological contamination from a person or object (Table).

On average, disturbing thoughts lasted 1 hour/day and were mildly distressing, although such thoughts lasted longer and were more distressing for mothers than fathers. Both groups reported that the intrusions caused little functioning interference and that they were able to control the thoughts. CES-D scores were subclinical on average, but higher in women. There was a small to moderate association between severity of intrusive thoughts and depressive symptoms among mothers, but not fathers.

According to Abramowitz and colleagues, these results support the hypothesis that senseless, intrusive, unacceptable thoughts, ideas, urges, and images about infants are common among healthy parents of newborns. In addition, such thoughts are reported with similar prevalence among fathers and mothers. Furthermore, the obsessions resemble those observed in patients with OCD in that the thoughts often center around disastrous consequences and are considered senseless, excessive, and incongruent with the person’s beliefs.

A number of factors may contribute to development of intrusive obsessional thoughts, according to the researchers. While theories have blamed fluctuations in hormone levels during pregnancy and postpartum, the presence of similar obsessions in fathers weakens this theory. Rather, situational factors such as the stress of caring for a newborn may be responsible. In addition, sensitivity to external threat cues tend to play a role in obsessive thought; it is probable that one would be hypersensitive to cues of danger when first faced with the responsibility of caring for a helpless baby. This may also explain the increased frequency of distress reported by mothers, as they also reported spending more time with their infants than fathers. Occurrence of obsessional thoughts may also be more upsetting to mothers due to the pressure of being expected to be happy during the postpartum period.

The authors also point out the importance of knowing the difference between obsessional thoughts, which do not lead to violence, and psychotic thoughts, which may lead to harming a child. People with obsessional thoughts are disgusted and frightened by them, while those with psychotic thoughts view them as rational. For parents who mistakenly interpret their intrusive thoughts as indicators that they are evil people capable of harming their child, these thoughts will elicit further distress.

Abramowitz and colleagues suggest that parents who have difficulty controlling recurring thoughts of harm to their child should seek care from their primary care physician or psychiatrist who may be able to determine whether the thoughts are depressive, obsessional, or psychotic. —DH

(J Clin Psychol Med Settings. 2003; 3:157-164)

Study Finds Common Genetic Roots for Severe Psychoses

A recent study reported what may be the first hard evidence for a similar genetic origin between bipolar disorder and schizophrenia. Sabine Bahn, MD, of the University of Cambridge in the United Kingdom, and a multicentered team of colleagues, found that these major psychotic illnesses both showed dysfunction of key oligodendrocyte and myelination genes, including transcription factors that regulate these genes.

Oligodendrocytes are responsible for myelin development in brain cells; myelin sheaths, composed mostly of fats and proteins, function to insulate the nerve cells enabling them to conduct electric signals between the brain and other parts of the body. The similar expression changes observed in the brains of patients with bipolar disorder and schizophrenia, support the idea that the disorders share a common causative and pathophysiological etiology.

Bahn and colleagues collected prefrontal cortex tissue of 45 brains from the Stanley brain collection (Stanley Medical Research Institute, Bethesda, MD)—15 samples of schizophrenia, 15 of bipolar affective disorder, and 15 controls. Psychiatric diagnoses, using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria, were made independently by two psychiatrists using medical records and telephone interviews with family members of the deceased. Gene-expression profiles of schizophrenia samples were compared with controls using indexing-based differential display polymerase chain reaction (PCR). Results were cross-validated with quantitative PCR, which was also used to determine expression profiles of 16 other oligodendrocyte and myelin genes in schizophrenia and bipolar disorder.

Similar dysfunction of key oligodendrocyte and myelin genes were observed in the brains of patients with schizophrenia and in those with bipolar disorder. The brains also showed a high degree of overlap in expression changes and several transcription factors known to coordinate myelin gene expression showed corresponding alterations. Overall, microarray analysis of the same genes correlated well.

Bahn and colleagues point out that although bipolar disorders and schizophrenia differ in presentation and clinical course, there is some overlap of symptoms and medications used for treatment of the disorders. The current findings suggest there may also be similar mechanisms involved in the disorders and that they may be more closely related than previously thought.

Bahn and colleagues also suggest that these results support those of other studies suggesting abnormalities in expression of lipid- and myelin-related genes in schizophrenia. These results surpass findings of other microarray investigations by showing similar expression changes in bipolar disorder, providing strong evidence that the disorders share a common genetic pathway.

Although the reason for the gene abnormalities remain unknown, the researchers suggest it may be related to infections of the central nervous system or other environmental issues. This is consistent with studies indicating that infections in early infancy may be a risk factor for schizophrenia and bipolar disorder. They posit that the ability to locate abnormalities in myelin may someday help physicians identify children likely to develop these mental disorders and provide treatment before symptoms surface.

Whether a reduction in oligodendrocyte number actually represents cellular dysfunction or whether it is simply a result of death has been debated. The authors note that the selective reduction observed in this study supports the latter view. They suggest that more refined expression studies are needed to confirm that abnormality is a result of dysfunction rather than death and to determine whether the changes are symbolic of a “sick” brain, are disease specific, or are secondary to another disease. —DH

(Lancet. 2003;362:798-805)

Bowel Disease Associated With Emotional Problems In Children

More than one third of children with inflammatory bowel disease (IBD) suffer from psychological problems, including depression and anxiety. IBD is a group of chronic intestinal disorders with symptoms of abdominal pain, rectal bleeding, and/or diarrhea, resulting from inflammation of the bowel. The course of the disorder is often unpredictable, and treatment with medications can cause severe side effects, such as vomiting, rash, hair loss, pancreatitis, bone loss, and low white blood cell count.

Laura Mackner, MD, and colleagues at Ohio State University in Columbus, previously compared emotional health of 41 children who had been diagnosed with IBD at least 1 year prior to the study, with 27 healthy children. The researchers expected the results to indicate that the children had adjusted well to managing all aspects of the disease, but they were dismayed to find that many had psychological issues involving low self-esteem, negative body image, and lower emotional, social, and family functioning.

In an effort to further analyze the behavior, emotional, and social functioning of IBD children, Mackner and colleagues administered questionnaires to 50 children with IBD (ages 11–17) and their parents and teachers. Twenty-seven healthy children and their parents and teachers were used as controls. The researchers were specifically looking for a relationship between the emotional factors and severity of IBD, family functioning, and self-esteem. Children described themselves using a self-esteem and body image checklist; parents rated their child’s behavior socially and emotionally at home; and teachers were asked to describe the child’s behavior at school.

Parents of children with IBD were three times more likely to report that their child had significant and emotional problems compared to parents of healthy children. In addition, teachers of IBD children reported that these children had more attention problems and missed school 1.5 times more often than children without IBD. Social problems, such as being teased and withdrawn were reported by 17% of parents of the IBD group compared to none of those in the control group. Self-esteem and body image were not shown to differ between the groups.

Behavioral and emotional functioning was higher in IBD children who had stronger family relationships and higher self-esteem. However, positive relationships and self-esteem may not protect social functioning, which was shown to be more dependent on whether the IBD was active. Having to take medication during school hours or abide by a special diet differentiates the child from the rest of his or her peers and may cause teasing, which contributes to psychosocial stress.  

Mackner and colleagues suggest that psychosocial interventions, along with medication when necessary, may help improve emotional functioning of children with IBD. They also suggest that children with IBD who show signs of trouble with school and social performance should be evaluated for emotional well-being. —DH

(American Psychological Association Annual Meeting; August 8, 2003; Toronto, Canada)

Stimulants Shown to Be Safe for Children With ADHD and Mania

The presence of mania in children with attention-deficit/hyperactivity disorder (ADHD) has made some physicians skeptical about prescribing stimulants due to the risk of increased side effects. However, research has found that such patients do not suffer increased side effects from methylphenidate.

Cathryn Galanter, MD, of Columbia University in New York City, conducted a placebo-controlled, double-blind study of 270 children 7–10 years of age who had participated in the Multimodal Treatment Study of ADHD. Following a lead-in stage, the children were administered methylphenidate titrated in three dosage levels or placebo for 4 weeks. Side effects in the children were determined by parents and teachers, who rated presence and severity of 10 side effects commonly associated with methylphenidate. The researchers constructed two proxies to distinguish children with ADHD and manic symptoms without bipolar disorder from those with ADHD only. The first proxy was based on the mania subsection of the Diagnostic Interview Schedule for Children (DISC), while the second was based on a response pattern on the Child Behavior Checklist (CBCL).

Manic symptoms were determined in 29 (10%) children according to the DISC proxy and 32 (11%) were diagnosed according to the CBCL. Seven children were shown to have manic symptoms according to both proxies. None of the children discontinued treatment due to adverse side effects during the titration period, although four were removed due to side effects during the lead-in phase. Children with mania and ADHD responded similarly to those without mania. They saw no difference between the groups in terms of irritability or other adverse side effects in response to methylphenidate.

Galanter and colleagues suggest that children with ADHD and manic symptoms may benefit from a carefully monitored methylphenidate trial. They plan to conduct studies investigating long-term response to stimulants in this population. —DH

(J Child Adolesc Psychopharmacol. 2003;2:123-136)

Psychiatric Dispatches is compiled and written by Deborah Hughes and Christopher Naccari.


Print Friendly 

Antiepileptic Treatment of
Posttraumatic Stress Disorder

Jeffrey L. Berlant, MD, PhD
Primary Psychiatry. 2003;10(10):41-49

Dr. Berlant is assistant clinical professor of psychiatry in the Department of Psychiatry at the University of Washington and is in private practice in Boise, Idaho.

Disclosure: Dr. Berlant is a consultant for GlaxoSmithKline and Ortho-McNeil, and is on the speaker’s bureau of Cephalon and GlaxoSmithKline.

Please direct all correspondence to: Jeffrey L. Berlant, MD, PhD, Department of Psychiatry, University of Washington, 4477 Emerals, Suite A300, Boise, ID 83706; Tel: 208-336-9907; Fax: 208-336-1043; E-mail:

Focus Points

Posttraumatic stress disorder (PTSD) is a mental illness that affects some people who have experienced or witnessed a violent or traumatic event.

The pathophysiology of PTSD requires an understanding of the neurobiology of fear and emotions.

γ-aminobutyric acid (GABA) inhibits the fear neurocircuits at many points while glutamate excites them; drugs that enhance GABA activity and/or reduce glutamate activity may be useful in the treatment of PTSD.

Antiepileptic drugs may have utility in the treatment of PTSD.



Do antiepileptic drugs (AEDs) have a role in the treatment of posttraumatic stress disorder (PTSD)? Current treatment of PTSD typically consists of psychotherapy and pharmacotherapy with antidepressants, but better treatments are greatly needed. The neural pathways involved in PTSD are strongly influenced by g-aminobutyric acid and glutamate—neurotransmitters that are implicated in the mechanism of action of many AEDs. This article reviews the brain fear circuitry and neurobiology of PTSD. In addition, the mechanisms of AED action as they relate to potential targets of intervention in treatment of PTSD is discussed. The clinical studies in which AEDs have been evaluated for treatment of PTSD are presented.




Posttraumatic stress disorder (PTSD) is a mental disorder that may develop in response to experiencing or witnessing a variety of physically threatening events, from physical assault, rape, sexual abuse, or motor vehicle accidents to natural disasters, military action, or terrorist attacks. PTSD is defined in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV)1 by three major symptom clusters: criterion B—re-experiencing (nightmares, flashbacks, intrusive recollections), criterion C—avoidance/numbing (avoiding thoughts or reminders, emotional numbing, loss of interest, detachment or estrangement, restricted range of affect), and criterion D—hyperarousal (insomnia, irritability or anger, difficulty concentrating, exaggerated startle reflex, physiologic reactivity, hypervigilance). When PTSD symptoms continue for 3 months, the disorder is considered chronic, and often persists for decades. Although a majority of cases of PTSD resolve on their own, 30% to 50% become chronic, and after 1 year the incidence of spontaneous remission diminishes, particularly for women.2


PTSD is currently treated with a variety of psychotherapeutic and pharmacologic measures. The central focus of psychopharmacologic research in PTSD has been antidepressants, and the only agents approved by the United States Food and Drug Administration for the treatment of PTSD are the selective serotonin reuptake inhibitors (SSRIs) sertraline and paroxetine. Nonetheless, the literature on brain fear circuits and the pathophysiology of PTSD suggests that neuromodulatory amino acids, particularly γ-aminobutyric acid (GABA) and glutamate, may overshadow the role of monoamine synapses in the regulation of fear responses. GABA appears to be a diffusely acting “brake” on fear circuitry activity at many levels, including cortical and limbic levels, if not at midbrain and brainstem levels. Glutamate, through N-methyl-D-aspartate (NMDA) receptors and α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)/kainate receptors, appears to mediate fear responses associated with excessive levels of stress,3 and serves as the “gas pedal” for fear responses. Inappropriate control of GABA and glutamate in various regions of the brain has been implicated in PTSD,4 and the fact that many of the antiepileptic drugs (AEDs) act via these neuromodulatory amino acids offers the intriguing possibility that AED pharmacotherapy may be effective in the treatment of PTSD. Indeed, the results of small preliminary studies with topiramate,5 lamotrigine,6 divalproex,7-9 and other AEDs10-12 suggest that these agents may have a place in the treatment of PTSD. In the remainder of this review, the neuroanatomy and neurophysiology of the parts of the brain involved with fear responses will be discussed, as well as the emerging picture of the pathogenesis of PTSD. How AEDs interact with these neural pathways and potentially influence the signs and symptoms of PTSD will also be explored.


Postulated Fear Circuitry of the Brain


Understanding the pathophysiology of PTSD requires knowing about more than merely the actions of various neurotransmitters on synapses; it requires knowledge about the structures and circuits in the brain that generate emotions and emotional reactions and about the factors that regulate the functioning of those structures and circuits. Vermetten and Bremner13 provided a well-presented review of the circuits and systems in stress, focusing mainly on the circuits involved in PTSD. Their model describes a system that perceives potential threats, cognitively appraises the potential threat, compares the threat with previous memories, selects responses for emotion and action, assigns an emotional valence to the threat, and prepares the organism for response by initiating motor and autonomic responses appropriate to the threat.


Sensory input is gated through the thalamus to the basolateral amygdala (BLA) (olfactory input is directly delivered to the amygdala), with projections to the orbitofrontal cortex, where the cognitive appraisal of the threat, its position and shape, its context, and other attributes are assessed (Figure 1). The amygdala attaches an emotional valence to the threat and triggers outflow tracts that initiate the motor, autonomic, and neurohormonal changes associated with fear response (Figure 2). Outflow tracts from the central nucleus of the amygdala (CeA) mediate these response changes through projections to the locus ceruleus (noradrenergic responses, such as elevated heart rate, blood pressure), the paraventricular nucleus (PVN) of the hypothalamus (activation of hypothalamic-pituitary-adrenal [HPA] dynamics and enhanced adrenocorticotrophin [ACTH] and cortisol release), and the periaqueductal gray (“freeze or flight” response). Internal cortical processing through connections between the amygdala and frontal cortex can influence the excitability of the amygdala and modulate the fear response.

The neurotransmitters glutamate and GABA are involved in controlling fear responses. Local infusions of NMDA or AMPA, GABAA antagonists, and corticotrophin-releasing hormone (CRH) into the BLA increase heart rate and blood pressure. Electrical stimulation of the BLA, combined with adequate norepinephrine and serotonin, elevates plasma cortisol levels. Electrical stimulation of the CeA results in increased attention, freezing behaviors, and brainstem reflexes, such as startle.3

The emerging model is of a system of inputs via a plastic BLA capable of acquired conditioned learning and aversive association, which is subsequently linked to a “hard-wired” output target system, ie, minimally, the locus ceruleus, periaqueductal gray, and PVN of the hypothalamus, that produces the characteristic manifestations of fear. This system of conditioned fear learning appears to be muted by NMDA antagonists at the level of the BLA.


Neurobiology of Posttraumatic Stress Disorder


PTSD is a clinical syndrome associated with diffuse noradrenergic hyperactivity, which is seen physiologically as an increase in heart rate and blood pressure, and in the galvanic skin response as a result of exposure to a traumatic cue. An intense startle response, a large increase in heart rate following exposure to a cue, and an elevated resting heart rate have also been cited as common physiological concomitants of PTSD.14


Abnormalities of the HPA axis may also be involved in PTSD. Although the reported abnormalities vary, there is consistent evidence of hypersuppressibility of plasma cortisol levels following very low dose exposure to dexamethasone. Studies of CRH in cerebral spinal fluid (CSF) suggest that CRH levels are elevated in PTSD, similar to that seen in depression. In contrast to depression, however, release of ACTH following intravenous CRH is augmented rather than blunted.15 This anomaly might be accounted for by the inability of current techniques to identify the proportion of CSF CRH that is extrahypothalamic in PTSD.16 


Neuromodulatory Amino Acids


Many biological factors, including amino acids, neuropeptides, and neurotransmitters, have been implicated in the production of PTSD symptoms, but recent research has increasingly focused on glutamate and GABA. Combined, these neuromodulatory amino acids are the most abundant neurotransmitters found in the brain, with glutamate exhibiting excitatory and GABA inhibitory effects on neuronal activity.


γ-aminobutyric acid


GABA inhibits the fear circuitry at many points, most likely through GABAA receptor activation. Brain structures in which GABA has been shown to exhibit active inhibition include the stria terminalis, preoptic area, hypothalamus, prefrontal cortex, and HPA axis. Serotonin acts to enhance GABA suppression of glutamate excitation in the basolateral amygdala.17 A major role of GABA in PTSD is suggested by the observation that patients with PTSD have reduced plasma levels of GABA compared to healthy controls and trauma controls without PTSD.18 Additional evidence is that PTSD patients are reported to have reduced benzodiazepine-receptor density and/or affinity in the medial prefrontal cortex.19 In summary, these studies raise the hypothesis that, perhaps mediated through plasma GABA levels or through dysfunctional benzodiazepine or GABAA receptors, abnormalities in GABA functioning may be associated with the pathophysiology of PTSD.




Glutamate receptors belong to two families: inotropic and metabotropic (G-protein-coupled) receptors. Three subgroups of ionotropic receptors exist: NMDA, AMPA, and kainate. NMDA and AMPA receptors are found in high density in the cortex, hippocampus, striatum, septum, and amygdala. The three groups of metabotropic receptors (MGlu I, II, and III) collectively affect phosphoinositol cascades and phosphokinase C, modulate adenyl cyclase, and negatively modulate glutamate and GABA. MGlu II/III agonists have anxiolytic, antipsychotic, and neuroprotection properties; both agonists and antagonists modulate serotonin activity.20


Glutamate is actively involved in fear responses. Uncontrollable stress raises glutamate release in the prefrontal cortex and hippocampus.21 Glutamate activates the basolateral amygdala and facilitates activity along pathways to the lateral extended amygdala.22,23 Glutamate also directly stimulates anxiety reactions from the periaqueductal gray of the brainstem and from midbrain nuclei, such as the locus ceruleus, regulating cardiovascular activity and parabrachial nuclei regulating respiratory activity.3 Finally, glutamate stimulates presynaptic inhibition of GABA release via kainate-type glutamate receptors,24-27 potentially crippling the brake of neural activity at least at the level of the hippocampus and probably at the hypothalamus and basolateral amygdala as well.


The NMDA, AMPA, and kainate receptors are involved in fear and anxiety responses. NMDA receptors facilitate activation of fear circuits within the amygdala, and NMDA antagonists inhibit these connections.3 Virtually all antidepressants have been shown to reduce NMDA receptor activity.28 AMPA has been shown to accelerate “kindling” of the amygdala, with progressive sensitization to the activating potential of various stimuli. Experimental AMPA antagonists reduce anxious behavior without evidence of motor slowing suggestive of sedation and block the startle response.29 Kainate receptors mediate different responses depending on location. In animal periaqueductal gray tissues, injection of kainate results in inhibitory avoidance (learned fear), in caudate tissue it reduces benzodiazepine-binding sites (a finding reported in the frontal cortex of humans with PTSD),30 and in the hippocampus it blocks conditional and contextual fear but not unconditioned fear.


Mechanisms of Action of Antiepileptics


A variety of mechanisms have been reported proposing how AEDs may inhibit seizures. Although the exact mechanisms are not fully understood, many of these medications are known to influence brain GABA levels and NMDA receptor activity. Although the principal mechanism of AED action of carbamazepine is usually cited as its sodium-channel gating effects, some studies have found that chronic carbamazepine administration to rats increases GABA concentrations in several brain regions, especially in the limbic areas.31 Valproate is known to elevate brain GABA (inhibitory) activity and may suppress NMDA-type glutamate receptors (excitatory). Felbamate is the first agent known to simultaneously potentiate GABA responsiveness and inhibit NMDA activity in the same pharmacologic dose range. To date, felbamate is the only AED proven active at the NMDA receptor; it may also work through presynaptic inhibition of glutamate release.20 Gabapentin, through unknown mechanisms, increases GABA turnover, at least in some brain regions. Lamotrigine, a known sodium channel inhibitor, may also inhibit release of glutamate. Its broad range of AED activity, however, suggests that there may be other, as yet uncharacterized, mechanisms of action, since other AEDs that also inhibit glutamate release lack lamotrigine’s broad range of seizure control. Topiramate blocks sodium channels, substantially elevates whole-brain GABA, and may work with novel GABAA receptors that do not bind benzodiazepines.32 to enhance GABAergic action. Like felbamate, topiramate inhibits glutamate receptors; in this instance it inhibits AMPA and kainate33 instead of NMDA receptors. Tiagabine works differently, through inhibition of both neuronal and glial GABA reuptake transporter proteins, most specifically on the GAT-1 transporter, one of four GABA transporters. The mechanism of action, therefore, would be assumed to be through both GABAA and GABAB receptors. Zonisamide, in contrast, inhibits sodium channels and blocks T-type calcium channels. It may also inhibit glutamate release from hippocampal tissue, but does not directly inhibit glutamate receptors.34 The above discussion of known mechanisms of action of newer AEDs suggests the effects on both GABA and glutamatergic systems are central to their AED activity.


Examination of the different neurobiological humoral substances suspected to be involved in the pathophysiology of PTSD reveals an extensive list of candidates, including the HPA axis, catecholamines, serotonin, cytokines, neuropeptide Y, cholecystokinin, opiates, somatostatin, GABA, and glutamate.13 Much as the use of antidepressants has probed the utility of monoamine modulators for the treatment of PTSD, AEDs present opportunities for modulating GABA and glutamate as a treatment strategy. To some extent, antidepressants may also work by influencing neuromodulatory amino acid systems as well as through monoamines, insofar as virtually all classes of antidepressants have been found to effectively reduce NMDA receptor activity.


Clinical Work With Antiepilepticsin Posttraumatic Stress Disorder


Despite the theoretical link between modulation of GABA and glutamate and the prominent role these neuromodulatory amino acids play in the limbic system and its larger connections, there are limited reports assessing the clinical value of AEDs in the treatment of PTSD.




Three open-label studies of carbamazepine in PTSD have been reported, including one using quantitative assessment of change in PTSD symptoms. Lipper and colleagues35 reported a 36% reduction in Diagnostic and Statistical Manual of Mental Disorders, Third Edition (DSM-III)36 PTSD symptoms (P<.01) in 10 inpatient subjects, and cited a 60% responder rate at week 5 with carbamazepine monotherapy. These findings are somewhat weakened by the fact that the investigators used an unpublished, unvalidated 48-point PTSD checklist to evaluate the drug effects. Although no subjects were reported to have dropped out during the study, of five subjects located after the end of the study, all had stopped carbamazepine, two because of rash and three in order to resume alcohol use. In a second study of unstated duration with carbamazepine monotherapy, 8 of 10 inpatient Vietnam veterans with DSM-III PTSD completed the study and, according to staff observation and patient self-report, had improved impulse control, anger outbursts, and violent behavior.37 Reasons for not staying in the study and adverse effects were unreported.


A third series, in which 28 adolescents in a state hospital were treated for undefined PTSD, tracked clinical reports of intrusive thoughts, flashbacks, hallucinations, and nightmares. After treatment with carbamazepine (monotherapy in 20) for a mean of 35 days (range 17–92 days), 79% became “asymptomatic” and the remainder “significantly improved.”11 Adverse effects and dropouts were not reported. Concomitant medications included methylphenidate (n=3), sertraline (n=2), clonidine (n=1), fluoxetine (n=1), and imipramine (n=1). In addition to these three open trials, there has been one case report with carbamazepine.12




The results of three open-label studies of divalproex in the treatment of PTSD have been reported. The first was conducted in 16 outpatient Vietnam veterans with Diagnostic and Statistical Manual of Mental Disorders, Third Edition-Revised38 PTSD, using an unpublished, four-point clinician rating scale for clinical outcomes. Three of 14 were treated with monotherapy divalproex. Concomitant medications included perphenazine (n=3), amitriptyline (n=3), desipramine (n=1), trazodone (n=2), clonidine (n=1), thioridazine (n=1), doxepin (n=1), chlorpromazine (n=1), and clonazepam (n=1). Of the 14 evaluated for outcomes, four withdrew, one due to side effects and three without recurrence of PTSD. After a mean treatment duration of 13.6 months (range 2–17 months), 11 (79%) reported significant improvement in hyperarousal, 9 (64%) in avoidance, but none (0%) in re-experiencing symptoms.7


An 8-week trial in 16 male combat veterans as outpatients used the Clinician Administered PTSD Scale (CAPS) to evaluate divalproex and reported a 17% reduction in DSM-IV PTSD symptoms in the 13 completers (P<.01). Five of 16 subjects received monotherapy. Concomitant medications included trazodone (n=5), fluoxetine (n=1), bupropion (n=1), buspirone (n=1), lorazepam (n=1), temazepam (n=1), and nefazodone (n=1). There was a 19% discontinuation rate due to side effects, specifically drowsiness, dizziness, increased irritability, headaches, vomiting, dry mouth, and flushing.8


Another 8-week trial in 30 outpatient combat veterans also used the CAPS and, using completer analysis, found a 24% reduction in DSM-IV PTSD symptoms at week 4 (n=21, P<.0001) and a 29% reduction at week 8 (n=14). The investigators reported a response rate of 43% at week 8. Monotherapy was said to be predominant in this study, notwithstanding the use of benzodiazepines (n=2) and hydroxyzine/diphenhydramine (n=19) for assistance with sleep. Twenty-one completed at least 4 weeks and 14 completed all 8 weeks. The discontinuation rate exceeded 50%, with 6 (20%) dropping out due to rash, diarrhea, or nausea. Reasons for the remaining early terminators were unreported.9 In addition to these open trials, there have been two case reports of divalproex use in PTSD.10,39




Lamotrigine has been tested in one study. In this randomized, double-blind, placebo-controlled trial, 10 DSM-IV PTSD subjects were assigned to lamotrigine monotherapy and five were assigned to placebo. Treatment duration was 10 weeks.6 Adverse effects included rash, sweating, drowsiness, poor concentration, thirst, restlessness, forgetfulness, bad taste in mouth, and erectile difficulty. Two subjects on lamotrigine (20%) and four on placebo (80%) dropped out of the study; all of the lamotrigine dropouts and two of the placebo dropouts occurred due to the onset of rash. The study was presented as positive, with 50% of lamotrigine subjects having a positive outcome using the Duke Global Rating for PTSD Scale compared to 25% on placebo (P<.05). However, closer examination in secondary analysis of the individual cases finds little evidence to support it as such.


On last-observation-carried-forward (LOCF) analysis (n=15), quantitative assessment of improvement in PTSD symptoms using a standardized instrument, the Severity of Illness (SI)-PTSD found a 23% improvement with lamotrigine (n=9) compared to 20% on placebo (n=5), presumably not a significant difference. Of additional concern, on completer analysis (n=9), lamotrigine scored lower than placebo (SI-PTSD improvement was 22% with lamotrigine [n=8] and 33% with placebo [n=1]), although the extremely small sample size precludes any judgment about the statistical significance of this finding. Because the very small sample size in this study makes ruling out a positive effect difficult, a larger study might conceivably find a difference. A larger study may support the finding that four of five responders required lamotrigine doses of ≥500 mg/day. However, the many weeks needed for slow dosage escalation to reduce the risk of triggering rash would be undesirable due to the practical clinical need for more rapidly working pharmacologic agents.




The effect of gabapentin in DSM-IV PTSD was evaluated in a retrospective chart review of 30 outpatient veterans previously treated for a mean duration of 10 months (range 1–36 months).40 A single “uninvolved” evaluator assessed the medical records, using an unnamed, nonstandard 4-point clinical global improvement scale, focusing on nightmares and insomnia as target symptoms, but also citing a variety of other symptoms, including pain, panic attacks, mood swings, and generalized anxiety. Outcome assessment reported a “marked” improvement in 17%, “moderate” in 60%, “mild” in 10%, and “none” in 13%. Half of subjects remained on gabapentin for at least 8 weeks. Gabapentin monotherapy was used in 10% of cases. Concomitant medications included fluoxetine (n=9), risperidone (n=4), buspirone (n=2), doxepin (n=1), diazepam (n=4), paroxetine (n=8), trazodone (n=5), lorazepam (n=3), sertraline (n=3), temazepam (n=2), amitriptyline (n=1), nefazodone (n=2), mirtazapine (n=1), clonazepam (n=4), propranolol (n=1), zolpidem (n=1), bupropion (n=5), imipramine (n=1), carbamazepine (n=2), venlafaxine (n=1), chloral hydrate (n=1), and olanzapine (n=1). Twenty of the 27 patients on comcomitant medications received two or more concomitant agents. Adverse effects included sedation, swelling, and dizziness. The dropout rate was 50% by week 8, with a 7% dropout rate ascribed to adverse effects, specifically excessive daytime sedation and “swelling.” In addition to this open trial of gabapentin, there has been one case report of positive benefit from gabapentin,41 and one of a woman who developed a nontherapeutic recrudescence of PTSD during self-administration of gabapentin for cocaine withdrawal symptoms.42




Topiramate was assessed in an open-label study in 35 outpatients with chronic civilian PTSD who were treated for an average of 33 weeks (range=1–119 weeks). Assessment was done with clinical report of nightmares and intrusive memories and, in 17 subjects, with the Posttraumatic Stress Disorder Checklist—Civilian Version (PCL-C). The PCL-C is reported to correspond to the CAPS with a kappa coefficient of approximately 0.90. Using the LOCF method, there was full improvement in intrusive memories in 54% and of nightmares in 50% of all subjects. Among patients without hallucinations (a globally less ill subset) there was a full response rate of 63% for nightmares and 68% for intrusive memories. For the 17 nonhallucinatory patients completing the PCL-C before and after beginning topiramate, there was a 49% reduction in PTSD symptoms at week 4. Responders, defined as those achieving a 30% or greater reduction in PTSD symptoms on the PCL-C, was 77% at week 4. Topiramate monotherapy was used in 20% (n=7). Concomitant medications, all previously resulting in treatment nonresponse, included benzodiazepines—predominantly zolpidem (n=8), atypical neuroleptics (n=8), SSRIs (n=7), divalproex (n=7), lamotrigine (n=5), stimulants (n=4), venlafaxine (n=3), lithium carbonate (n=2), gabapentin (n=2), verapamil (n=1), mirtazapine (n=1), selegiline (n=2), and nefazodone (n=1). Sixteen of the 28 patients on concomitant medications received two or more additional medications. The dropout rate was 37% overall, and 26% due to adverse effects. Among those who dropped out despite a positive clinical response (n=7), reasons included a variety of concerns, such as cessation of eating, headaches, impaired memory, urticaria (later overcome by switching from topiramate tablets to the sprinkle capsule formulation), and exacerbation of panic attacks. One discontinued topiramate when PTSD symptoms did not recur after his prescription ran out.5


Other Antiepileptics


Tiagabine, a GABA reuptake inhibitor, has been mentioned in two case reports (total N=3) to have at least some benefit for intrusive thoughts and nightmares, hyperarousal, and panic.43,44 Vigabatrin, a GABA transaminase inhibitor not marketed in the United States, has been reported to be helpful for calming, improving sleep, and reducing startle response in a small case series involving five patients.45 A single case report of oxcarbazepine, a GABAergic agent marketed for partial epilepsy, suggested benefit for anxiety in general, although it was unstated whether this was monotherapy and no specific comment on PTSD symptoms or scores was reported.46 No other studies with these drugs have been reported.


Comparative Effectiveness


A summary of literature reports on the efficacy of AEDs in the treatment of PTSD is presented in the Table.5-7,11,35,37,40,45 It appears that PTSD patients may benefit from several of the AEDs, including topiramate, carbamazepine, divalproex, lamotrigine, gabapentin, tiagabine, and vigabatrin, In only one very small study was a treatment comparison made, and that was lamotrigine to placebo6; all other reports were either open-label or case reports. Therefore, controlled clinical research is needed to assess these signals.

It is unclear if any AED is superior to any other AED or to sertraline or paroxetine in the treatment of PTSD. The literature contains little information to provide definitive estimates of relative efficacy of these agents. None of the studies employ head-to-head, properly randomized comparison of responses using adequate doses of different AEDs. However, some of these reports provide stronger signals of efficacy than others, which is an important consideration in selecting agents for future study.

Interestingly, these signals appear different depending on the definition selected for assessing response. Studies using the least discriminating definition of response as an impressionistic judgment of improvement, either as a nominal category or in the form of a global clinical improvement score or other simple ordinal scale, provide quite similar estimates for all AEDs. Carbamazepine has a response rate of 80% in inpatient male veterans and 79% in hospitalized adolescents, divalproex has a response rate of 64% to 79% for avoidance and hyperarousal symptoms and 0% for reexperiencing symptoms in outpatient combat veterans, gabapentin has a moderate to marked response rate of 77% in outpatient veterans, lamotrigine has a 50% response rate compared to a 25% placebo rate in combined civilian and veteran outpatients, and topiramate has a 68% full and 89% full or partial response rate for reexperiencing symptoms in civilians with nonhallucinatory PTSD.


However, some studies use more discriminating definitions of response. A positive response may be defined in terms of a predefined magnitude of reduction in PTSD, as measured by a structured clinical interview or self-report instrument, which attains or exceeds the conventional 30% threshold for clinical response. Studies utilizing this definition use a variety of instruments to measure each of the symptoms listed in the version of the DSM contemporary to the time period when the study was done. Using this definition of response, 60% of carbamazepine patients (based on an unvalidated checklist), 43% of divalproex patients (based on the CAPS), 33% of lamotrigine patients (based on the Severity of Illness-PTSD scale), and 77% of topiramate patients (based on the PTSD?Checklist-Civilian scale) were responders. Viewed by this method, there appears to be a considerable spread in values from 33% to 77%.


Similarly, comparing percentage of reduction of PTSD symptoms as measured by structured clinical interview or self-report instrument suggests there may be differences. Using this method, improvement rates vary from 17% and 29% for divalproex, to 22% for lamotrigine, 36% reduction for carbamazepine, and 50% for topiramate. Viewed this way, there appears to be sufficient spread of values from 17% to 50% to consider the possibility that some agents might be superior to others in terms of efficacy.


Care must also be taken when comparing the results of these studies because of small study samples, different instruments to measure response, different patient populations (eg, military or civilian, adult or adolescent, male or female or both), different definitions of PTSD, and different durations of treatment until assessment. Demonstration of superior efficacy of any agent over another ultimately requires a different methodology—either well-controlled, dose-adequate, head-to-head comparisons, or comparative estimations of effect sizes from future placebo-controlled clinical trials.


If the efficacy of topiramate for treatment of PTSD were to be demonstrated under controlled conditions, it would argue strongly for understanding the critical pieces neuromodulatory amino acids may play in the pathophysiology of PTSD, since topiramate lacks direct NMDA receptor activity.30 The ability to rapidly block PTSD symptoms may be related to its vigorous elevation of brain GABA levels and receptor responsivity, which provides a general quieting of fear responses, and to its AMPA-receptor inhibiting effects,47 which may help normalize neural
plasticity of the basolateral nucleus of the amygdala and quiet periaqueductal gray stimulation by glutamate.48




PTSD is a mental disorder that can affect people who have experienced or witnessed a violent or traumatic event. Treatment of PTSD involves psychotherapy and pharmacotherapy, and although two SSRIs have been approved by the FDA for treatment of PTSD, a consensus on medication choice does not exist, especially in complex situations, such as comorbid bipolar disorder. Although the central focus of research into psychopharmacologic treatment of PTSD has been the use of antidepressants, the literature on brain fear circuits and the pathophysiology of PTSD suggest that neuromodulatory amino acids, particularly GABA and glutamate, may overshadow the role of monoamine synapses in regulation of fear responses. GABA acts to inhibit the brain’s fear circuitry, and because many AEDs act by increasing brain GABA levels, they became targets of interest as potential therapeutic agents in the treatment of PTSD. However, the possible role of AEDs in modulating glutamate activity should not be ignored. Glutamate, acting via NMDA and AMPA/kainate receptors, amplifies fear responses. In addition to inhibiting glutamate release, glutamatergic AEDs may address PTSD symptoms by inhibiting NMDA receptor activity through AMPA-receptor inhibition (because NMDA receptors require AMPA receptor activity to depolarize) and by enhancing GABA release through kainate-receptor inhibition (because kainate receptors can suppress the presynaptic release of GABA). There are likely other mechanisms as well, by which AEDs may modulate activity at different levels of the fear circuit.


The results of several small studies suggest that AEDs may have a role in the treatment of PTSD. Larger, well-controlled studies of AEDs in subjects with PTSD are needed to test the positive signals that these small studies have suggested. If the efficacy of topiramate is confirmed through rigorous, systematic clinical research, it may point toward the critical importance of AMPA/kainate receptor inhibition as a differentiating property for individual AEDs and as a target for new drug development for PTSD. PP




1. Diagnostic and Statistical Manual of Mental Disorders. 4th ed, text-rev. Washington, DC: American Psychiatric Association; 2000.

2. Breslau N. The epidemiology of posttraumatic stress disorder: what is the extent of the problem? J Clin Psychiatry. 2001;62(suppl 17):16-22.
3. Davis M, Whalen PJ. The amygdala: vigilance and emotion. Mol Psychiatry. 2001;1:13-34.

4. Hageman I, Andersen HS, Jorgensen MB. Post-traumatic stress disorder: a review of psychobiology and pharmacotherapy. Acta Psychiatr Scand. 2001;6:411-422.

5. Berlant J, van Kammen DP. Open-label topiramate as primary or adjunctive therapy in chronic civilian posttraumatic stress disorder: a preliminary report. J Clin Psychiatry. 2002;1:15-20.

6. Hertzberg MA, Butterfield MI, Feldman ME, et al. A preliminary study of lamotrigine for the treatment of posttraumatic stress disorder. Biol Psychiatry. 1999;9:1226-1229.

7. Fesler FA. Valproate in combat-related posttraumatic stress disorder. J Clin Psychiatry. 1991;9:361-364.

8. Clark RD, Canive JM, Calais LA, Qualls CR, Tuason VB. Divalproex in posttraumatic stress disorder: an open-label clinical trial. J Trauma Stress. 1999;2:395-401.

9. Petty F, Davis LL, Nugent AL, et al. Valproate therapy for chronic, combat-induced posttraumatic stress disorder. J Clin Psychopharmacol. 2002;1:100-101.

10. Berigan TR, Holzgang A. Valproate as an alternative in post-traumatic stress disorder: a case report. Mil Med. 1995;6:318.

11. Looff D, Grimley P, Kuller F, Martin A, Shonfield L. Carbamazepine for PTSD. J Am Acad Child Adolesc Psychiatry. 1995;6:703-704.

12. Stewart JT, Bartucci RJ. Posttraumatic stress disorder and partial complex seizures. Am J Psychiatry. 1986;1:113-114.

13. Vermetten E, Bremner JD. Circuits and systems in stress. II. Applications to neurobiology and treatment in posttraumatic stress disorder. Depress Anxiety. 2002;1:14-38.

14. Berlant J. New drug development for post-traumatic stress disorder. Curr Opin Investig Drugs. 2003;1:37-41.

15. Yehuda R. Current status of cortisol findings in post-traumatic stress disorder. Psychiatr Clin North Am. 2002;2:341-368.

16. Kasckow JW, Baker D, Geracioti TD Jr. Corticotropin-releasing hormone in depression and post-traumatic stress disorder. Peptides. 2001;5:845-851.

17. Stutzmann GE, LeDoux JE. GABAergic antagonists block the inhibitory effects of serotonin in the lateral amygdala: a mechanism for modulation of sensory inputs related to fear conditioning. J Neurosci. 1999;11:RC8.

18. Vaiva G, Ducrocq F, Cottencin O, Goudemand M, Thomas P. Immediate fright reaction: an essential criterion in the development of posttraumatic stress disorder (PTSD). Can J Psychiatry. 2000;10:939.

19. Bremner JD, Innis RB, Southwick SM, Staib L, Zoghbi S, Charney DS. Decreased benzodiazepine receptor binding in prefrontal cortex in combat-related posttraumatic stress disorder. Am J Psychiatry. 2000;7:1120-1126.

20. Zarate CA, Quiroz J, Payne J, Manji HK. Modulators of the glutamatergic system: Implications for ht development of improved therapeutics in mood disorders. Psychopharmacol Bull. 2002;4:35-83.

21. Pepeu G, Blandina P. The acetylcholine, GABA, glutamate triangle in the rat forebrain. J Physiol Paris. 1998;5-6:351-355.

22. Rodrigues SM, Bauer EP, Farb CR, Schafe GE, LeDoux JE. The group I metabotropic glutamate receptor mGluR5 is required for fear memory formation and long-term potentiation in the lateral amygdala. J Neurosci. 2002;12:5219-5229.

23. Herman JP, Tasker JG, Ziegler DR, Cullinan WE. Local circuit regulation of paraventricular nucleus stress integration: glutamate-GABA connections. Pharmacol Biochem Behav. 2002;3:457-468.

24. Min MY, Melyan Z, Kullmann DM. Synaptically released glutamate reduces gamma-aminobutyric acid (GABA)ergic inhibition in the hippocampus via kainate receptors. Proc Natl Acad Sci U S A. 1999;17:9932-9937.

25. Liu QS, Patrylo PR, Gao XB, van den Pol AN. Kainate acts at presynaptic receptors to increase GABA release from hypothalamic neurons. J Neurophysiol. 1999;2:1059-1062.

26. Rodriguez-Moreno A, Herreras O, Lerma J. Kainate receptors presynaptically downregulate GABAergic inhibition in the rat hippocampus. Neuron. 1997;4:893-901.

27. Cunha RA, Constantino MD, Ribeiro JA. Inhibition of [3H] gamma-aminobutyric acid release by kainate receptor activation in rat hippocampal synaptosomes. Eur J Pharmacol. 1997;2-3:167-172.

28. Paul IA, Nowak G, Layer RT, Popik P, Skolnick P. Adaptation of the N-methyl-D-aspartate receptor complex following chronic antidepressant treatments. J Pharmacol Exp Ther. 1994;1:95-102.

29. Walker DL, Davis M. The role of amygdala glutamate receptors in fear learning, fear-potentiated startle, and extinction. Pharmacol Biochem Behav. 2002;3:379-392.

30. Sperk G, Schlogl E. Reduction of number of benzodiazepine binding sites in the caudate nucleus of the rat after kainic acid injections. Brain Res. 1979;3:563-567.

31. Nagaki S, Kato N, Minatogawa Y, Higuchi T. Effects of anticonvulsants and gamma-aminobutyric acid (GABA)-mimetic drugs on immunoreactive somatostatin and GABA contents in the rat brain. Life Sci. 1990;22:1587-1595.
32. White HS, Skeen GA, Woodhead J, Wolf HH. Topiramate modulates GABA-evoked currents in murine cortical neurons by non-benzodiazepine mechanism. Epilepsia. 2000;41(suppl 1):S17-S20.

33. Shank RP, Gardocki JF, Streeter AJ, Maryanoff BE. An overview of the preclinical aspects of topiramate: pharmacology, pharmacokinetics, and mechanism of action. Epilepsia. 2000;41(suppl 1):S3-S9.

34. Rho JM, Sankar R. The pharmacologic basis of antiepileptic drug action. Epilepsia. 1999;11:1471-1483.

35. Lipper S, Davidson JR, Grady TA, et al. Preliminary study of carbamazepine in post-traumatic stress disorder. Psychosomatics. 1986;12:849-854.

36. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, DC:?American Psychiatric Association; 1980.

37. Wolf ME, Alavi A, Mosnaim AD. Posttraumatic stress disorder in Vietnam veterans clinical and EEG findings; possible therapeutic effects of carbamazepine. Biol Psychiatry. 1988;6:642-644.

38. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed rev. Washington, DC:?American Psychiatric Association; 1987.

39. Szymanski HV, Olympia J. Divalproex in posttraumatic stress disorder. Am J Psychiatry. 1991;8:1086-1087.

40. Hamner MB, Brodrick PS, Labbate LA. Gabapentin in PTSD: a retrospective, clinical series of adjunctive therapy. Ann Clin Psychiatry. 2001;3:141-146.

41. Brannon N, Labbate L, Huber M. Gabapentin treatment for posttraumatic stress disorder. Can J Psychiatry. 2000;1:84.

42. Markowitz JS, Finkenbine R, Myrick H, King L, Carson WH. Gabapentin abuse in a cocaine user: implications for treatment? J Clin Psychopharmacol. 1997;5:423-424.

43. Berigan T. Treatment of posttraumatic stress disorder with tiagabine. Can J Psychiatry. 2002;8:788.

44. Schwartz TL. The use of tiagabine augmentation for treatment-resistant anxiety disorders: a case series. Psychopharmacol Bull. 2002;2:53-57.

45. Macleod AD. Vigabatrin and posttraumatic stress disorder. J Clin Psychopharmacol. 1996;2:190-191.

46. Berigan T. Oxcarbazepine treatment of posttraumatic stress disorder. Can J Psychiatry. 2002;10:973-974.

47. Zullino DF, Krenz S, Besson J. AMPA blockade may be the mechanism underlying the efficacy of topiramate in PTSD. J Clin Psychiatry. 2003;2:219-220.

48. Zullino DF, Cottier AC, Besson J. Topiramate in opiate withdrawal. Prog Neuropsychopharmacol Biol Psychiatry. 2002;6:1221-1223.


Print Friendly 

Pharmacologic Treatment of
Binge-Eating Disorder

William P. Carter, MD, and Lindsay J. Pindyck, BA
 Primary Psychiatry. 2003;10(10):31-36

Dr. Carter is instructor in psychiatry and Ms. Pindyck is a medical student at Harvard Medical School, McLean Hospital, in Belmont, Massachusetts.

Disclosure: Dr. Carter is a consultant for and on the speaker’s bureau of Eli Lilly, Ortho-McNeil, and Wyeth; he receives research support from Eli Lilly and Pfizer.


Please direct all correspondence to: William P. Carter, MD, Department of Psychiatry, Harvard Medical School, McLean Hospital, 115 Mill St, Belmont, MA 02478; Tel: 617-855-2277; Fax: 617-855-3748; E-mail:

Focus Points

Binge-eating disorder (BED) is characterized by recurrent episodes of binge eating with behavioral indicators of loss of control, distress over the binge, and the lack of compensatory mechanisms to prevent weight gain.

BED is highly prevalent (approximately 45%) in the obese population.

Pharmacotherapy with antidepressants and appetite suppressants can reduce binge frequency and body mass index.

Newer-generation antiepileptics, such as topiramate and zonisamide, have been associated with weight loss in clinical studies of epilepsy and obesity and have shown promise in the treatment of BED.



Binge-eating disorder (BED) has a prevalence of 2% to 3% in the general population and is much more common (approximately 45%) in obese individuals. The use of psychotherapy may successfully reduce the incidence of binge eating but this treatment option has not been useful in reducing body weight. The close resemblance of bulimia nervosa and BED suggests a role for the pharmacologic treatment of BED with antidepressants. While treatment recommendations for BED must be considered preliminary pending replication of early findings, empirical data and controlled studies offer support for the use of selective serotonin reuptake inhibitors (SSRIs), centrally acting appetite suppressants, and topiramate. Anecdotal reports also show promise for venlafaxine and zonisamide. The strength of data from controlled clinical trials, tolerability, and ease of administration suggest that SSRIs could be first-line treatments for BED, but that special populations (eg, bipolar patients) may benefit from first-line treatment with topiramate. Combination treatments with antiepileptics and SSRIs may offer further advantages for some patients as well. The optimal combination of pharmacologic treatments and relative roles of pharmacologic and psychotherapy treatments remain to be determined.




The rapidly increasing percentage of Americans who are overweight is generating fervor in the popular press and an equally intense search for solutions in the medical community. A disproportionate number of overweight people have binge-eating disorder (BED)1—the most common eating disorder.2 BED is characterized by recurrent episodes of binge eating with the behavioral indicators of loss of control, distress over the binges, and lack of compensatory behaviors to prevent associated weight gain.3 The number of controlled studies of BED has increased considerably in the last several years, and, more generally, the treatment of overweight conditions continues to be an active area of research. For example, the antiepileptic drugs (AEDs) topiramate and zonisamide have stimulated renewed interest in a possible role for AEDs in the treatment of eating disorders and weight management.


The encouraging results from these studies run counter to the sentiment of many clinicians and patients that there are no sound pharmacologic treatments for overweight condition, with or without BED. A growing body of anecdotal evidence, past controlled trials, and the results from these recent studies suggest that a change in thinking may be in order. These developments have naturally met with some skepticism, given the checkered history of weight-loss treatments, but they have certainly captured the attention of affected patients and the media.4,5


This review addresses aspects of BED and the studies of the pharmacologic treatment of this disorder with three categories of medications: antidepressants, centrally acting appetite suppressants, and AEDs. Based on these findings, preliminary treatment guidelines for the pharmacologic treatment of BED are provided.




Since Stunkard6 first described BED in 1959, researchers have debated the merits of a separate diagnostic entity for BED and related conditions, such as nonpurging bulimia nervosa and obesity with binge eating. In fact, in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV),3 BED still represents a form of eating disorder not otherwise specified, with diagnostic criteria detailed in the appendix as a condition requiring further study (Table 1).3 BED differs mainly from bulimia nervosa by the absence of any compensatory behaviors and by the higher frequency and longer duration of behaviors required for diagnosis (Table 2). These more stringent criteria serve to distinguish BED from mild variants of normal eating. With prevalence rates estimated between 2% and 3%, BED is more common than anorexia nervosa and bulimia nervosa combined.7,2 In the obese population, the prevalence of BED approaches 45%.8 In addition, it affects a broader range of the population in terms of both sex and age distribution. Men represent approximately 33% of patients with BED (though estimates vary), and patients with BED commonly present for treatment in their 40s, though the actual onset of illness may be considerably earlier.9,10 Furthermore, BED also appears to affect similar percentages of African Americans and Caucasians.7,11



It is obvious that the symptoms diagnostic of BED cause patients considerable distress and dysfunction. In addition, the correlation with overweight condition contributes additional health risks, such as hypertension and diabetes.12,13 Therefore, as a common condition associated with considerable morbidity, BED constitutes a significant public health concern.


Pharmacologic Treatment


Until recently, the investigations of possible treatments of BED lagged behind the obvious need. Without established treatments, many patients first turned to a series of calorie-restricted diets; however, many patients with BED cannot comply with these regimens.14 Psychotherapy, specifically cognitive-behavioral therapy (CBT) and interpersonal therapy, has demonstrated some success in addressing binge eating per se, but not in sustained weight reduction.15 Among those seeking the increasingly common surgical intervention of gastric bypass surgery, approximately 50% are believed to have BED.16,17 This overrepresentation of BED patients is not limited to the surgical domain. Estimates for the prevalence of BED in weight loss clinics, for example, range up to 33%,18 and in a setting such as Overeaters Anonymous, up to 70% of participants may have BED.1,7,19 This known concentration of BED patients in various settings could conceivably facilitate screening efforts and subsequent referrals for appropriate treatment. The current pharmacologic options include antidepressants, centrally acting appetite suppressants, and AEDs.




A possible role for antidepressants in the treatment of BED is largely derived from BED’s close resemblance to bulimia nervosa, a disorder for which many controlled trials have demonstrated superiority of antidepressants over placebo. Even medications such as desipramine and phenelzine, which are known to be associated with weight gain in many patients with depression, have demonstrated efficacy in the treatment of bulimia nervosa20; this finding indicates that mere appetite suppression is unlikely to be the mechanism of action at use. Furthermore, while bulimia nervosa also shows considerable comorbidity with major depression, patients with bulimia without major depression also respond to antidepressants,20 countering an explanation of the response as coincidental with the alleviation of a mood disorder.


Several categories of antidepressants, including tricyclic antidepressants (TCAs), monoamine oxidase inhibitors, and selective serotonin reuptake inhibitors (SSRIs), have demonstrated efficacy in the treatment of bulimia nervosa.20 Collectively, this response argues against an idiosyncratic reaction to a medication or a coincidental improvement of a comorbid disorder as a comprehensive explanation of treatment success. Instead, it is more likely that the treatments have a more fundamental mechanism of action in common. In turn, this fundamental mechanism would be more likely to generalize to the treatment of BED, as it is a closely related condition. Subsequent controlled trials in BED are needed to confirm this hypothesis.


Early work on binge eating occurred prior to the DSM-IV4 criteria for BED and, therefore, is found in studies addressing closely related conditions, such as bulimia nervosa, nonpurging type, and obesity in binge eating. As with bulimia nervosa, early trials examined TCAs versus placebo. A high placebo response rate and small sample size limited inferences about imipramine and naltrexone in one negative study.21 However, desipramine was superior to placebo in reducing binge frequency (but not weight) in a study of nonpurging bulimia nervosa.22


A subsequent open trial of fluvoxamine in binge eating without vomiting suggested efficacy for this medication and added evidence that this positive effect on binge frequency was not specific to TCAs.23 The controlled studies of SSRIs that followed added further support. The first such study demonstrated superiority of fluvoxamine over placebo in the primary outcome measure binge frequency, and in secondary outcomes including body mass index (BMI) and score on the Clinical Global Impressions for Improvement (CGI-I) scale.24 McElroy and colleagues25,26 found similar results with both sertraline and citalopram in separate studies. A fourth study of an SSRI failed to show a difference between fluoxetine and placebo on measures of binge frequency or weight,27 but a fifth study demonstrated superiority of fluoxetine on all three measures: binge frequency, BMI, and the CGI-I scale.28 All four positive studies were short-term (6–9 weeks) with relatively small sample sizes (N=38–85). In summary, four of five short-term, placebo-controlled studies of SSRIs demonstrated superiority of medication over placebo on measures of binge frequency, weight, and global improvement (Table 3).21,22,24-31

Centrally Acting Appetite Suppressants


The centrally acting appetite suppressants have properties directly linked to a defining symptom of BED—overeating. Appetite suppressants act at least in part by affecting neurotransmission in appetite and satiety centers in the brain.32 Like some antidepressants, dexfenfluramine inhibits serotonin reuptake, as does sibutramine, which also has noradrenergic activity. Thus, as a dual reuptake inhibitor, sibutramine has properties in common with duloxetine and venlafaxine.33-35 Results from an open study suggest efficacy of venlafaxine in BED as well.36


In an 8-week study, dexfenfluramine was superior to placebo in reducing binge frequency,29 but ironically, this appetite suppressant is the only agent in a positive study of BED by DSM-IV criteria that failed to separate from placebo on a weight measurement. Furthermore, dexfenfluramine was withdrawn from the market shortly after the publication of the study due to its association with heart-valve defects. Sibutramine, on the other hand, in a somewhat longer study (12 weeks) had results superior to placebo on both parameters.30 However, CGI was not assessed in either study. A recent double-blind study of sibutramine has confirmed these observations.37 Side effects of sibutramine present a possible drawback to this line of treatment; common adverse reactions experienced are headache, change in appetite, dry mouth, and insomnia.




In early research on BED, some patients displayed electroencephalographic abnormalities and responded to treatment with the AED phenytoin.38 Subsequently, however, more systematic research with bulimia nervosa patients failed to corroborate the findings that suggested a seizure disorder as the etiology for such dysregulated eating behaviors.39,40 AED use in the treatment of BED then waned until the successful use of these medications in the treatment of mood disorders and impulse-control disorders stimulated renewed interest. In patients with both bipolar disorder and bulimia nervosa, the older AEDs carbamazepine and valproic acid were effective treatments.39 However, as was the case with gabapentin, the possible side effect of increased appetite limited the appeal of these medications to patients.41,42 While lamotrigine is not known to cause increased appetite and has a role in the treatment of bipolar disorder,43 no formal studies exist to guide evaluation of its use in eating disorders.


A newer generation of AEDs, including felbamate, topiramate, and zonisamide, has been associated with significant weight loss in clinical trials for epilepsy.44-46 Felbamate’s association with severe side effects, including aplastic anemia, hepatic failure, and Steven’s Johnson syndrome, has restricted its use to treatment refractory epilepsy patients.44-46 Encouraging findings have emerged from obesity trials, however, where zonisamide and topiramate have demonstrated efficacy in achieving short-term weight loss.47,31 In addition, a placebo-controlled trial showed topiramate to be effective in the treatment of bulimia nervosa.48,49 As with bipolar disorder, where success with certain AEDs did not always generalize to other members of this class of medication, controlled studies are needed to evaluate which particular AEDs might be effective agents in the treatment of BED.


Early evidence of success with phenytoin has not been replicated in subsequent controlled trials of eating disorders. Zonisamide has received favorable reviews in anecdotal reports of BED treatment, but no controlled trials have been published. Topiramate also showed promise in an open trial of 13 BED patients with obesity, during which 9 patients responded on the measures of binge frequency and weight50; these findings were replicated in a second open trial.51 In a placebo-controlled trial, topiramate was superior to placebo on measures of binge frequency, weight, and CGI.47 Topiramate was also superior to placebo in achieving remission of binge eating, with 64% of patients reporting absence of all bingeing versus 30% for placebo patients.




The SSRIs remain the most extensively studied medications in the treatment of BED, with four published placebo-controlled trials favoring medication over placebo on measures of binge frequency, weight, and CGI. (The negative trial has not yet been published). However, the studies were all short term (6–9 weeks) and weight loss was modest (1.7–4.6 kg among completers). Results from the topiramate study compare favorably with the changes with the SSRIs in both the weight loss measure and also remission of binge eating. However, in the absence of a head-to-head study, the validity of such comparisons cannot be directly assessed.


While preliminary, these findings offer a basis for guidelines of pharmacologic treatment of BED (Table 4). First, it should be noted that there are no Food and Drug Administration-approved treatments for BED, and that such guidelines are therefore limited to inferences from the studies reviewed here and from clinical practice. Given the broader range of studies with SSRIs, their favorable side-effect profiles, and the high co-occurrence of BED with major depression, these medications could be considered first-line treatments for BED. Adequate trials will typically require doses and durations at the upper end of the range used to treat depression and bulimia nervosa, perhaps resembling the typical treatments for obsessive-compulsive spectrum disorders rather than uncomplicated unipolar major depression.


Exceptions to the use of SSRIs as first-line agents could apply to patients known to have bipolar disorder, where topiramate might be the preferred agent. Topiramate may offer the advantage of adjunctive mood stabilization, and it certainly would be expected to be associated with less switching into mania or exacerbation of mood cycling.52 Because sibutramine’s mechanism of action resembles that of the antidepressant venlafaxine, topiramate might be preferred over sibutramine in bipolar patients.


Topiramate may offer an alternative or adjunctive treatment for patients who cannot tolerate SSRIs, or who are only partially responsive to these agents. Low starting doses and gradual titration improve tolerability of topiramate. Therefore, doses should start at no more than 25 mg HS, then advance by 25 mg/day weekly until the patient shows full response or reaches a target dose of at least 200 mg HS. Patients treated with topiramate report different subjective states which correlate with improvement in binge eating, whether by decreased appetite or by earlier satiety, for example. Single dosing at bedtime generally minimizes the potential for sedative side effects, but some patients benefit from divided doses to address periods of peak binge urges over the course of the day. In addition to possible sedation, common side effects experienced with topiramate are dizziness, ataxia, paresthesia, cognitive changes, and somnolence.


Third-line treatment options include monotherapy with sibutramine, venlafaxine, and zonisamide, or combinations of AEDs and antidepressants. The combination of pharmacotherapy and psychotherapy has not been adequately addressed, and head-to-head studies of pharmacologic and psychotherapy treatments raise challenging methodologic issues. Still, CBT and interpersonal therapy have demonstrated efficacy in decreasing binges but not weight. Thus, while no definitive conclusions about the relative roles of psychotherapy and pharmacologic treatment can be made at this time, the fact that BED patients without mood disorders respond to medication implies that such treatment should not be reserved only for those who have comorbid mood disorders. Medication treatment should also not be reserved for those who fail to respond to psychotherapy. Intuitively, the combination of psychotherapy and pharmacologic treatment offers the advantage of combining different “mechanisms of action” and the potential to target specific residual symptoms.




The close resemblance of bulimia nervosa and BED suggests a role for the pharmacologic treatment of BED with antidepressants. Empirical data and controlled studies in the treatment of BED offer support for the use of SSRIs, centrally acting appetite suppressants, and the AED topiramate. Anecdotal reports also show promise for venlafaxine and zonisamide. Strength of controlled trial data, tolerability, and ease of administration suggest SSRIs could be first-line treatments for BED, but special populations (eg, bipolar patients) may benefit from first-line treatment with topiramate. Combination treatments with AEDs and SSRIs may offer further advantages for some patients as well. Optimal combination pharmacologic treatments, and relative roles of pharmacologic and psychotherapy treatments, remain to be determined. PP



1. de Zwaan M, Mitchell JE, Raymond NC, et al. Binge eating disorder: clinical features and treatment of a new diagnosis. Harv Rev Psychiatry. 1994;1:310-325.
2. Hsu LKG. Epidemiology of the eating disorders. Psychiatr Clin North Am. 1996;19:681-700.

3. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.

4. Goode E. “To Curb Cravings and Even More.” New York Times.

April 15, 2003:F5.

5. Grady D. “Quest for Weight Loss Drug Takes an Unusual Turn.” New York Times. April 15, 2003:F5.

6. Stunkard AJ. Eating patterns and obesity. Psychiatr Q. 1959;33:284-295.

7. Spitzer RL, Yanovksi S, Wadden T, et al. Binge eating disorder: its further validation in a multisite study. Int J Eat Disord. 1993;13:137-153.

8. Agras WS. Treatment of binge-eating disorder. In: Gabbard GO, ed. Treatments of Psychiatric Disorders. Washington, DC: American Psychiatric Press; 2001: 2209-2219.

9. Bruce B, Agras WS. Binge eating in females: a population-based investigation. Int J Eat Disord. 1992;12:365-373.
10. Yanovksi SZ. Binge eating disorder: current knowledge and future directions. Obes Res. 1993;1:306-324.
11. Pike KM, Dohm FA, Striegel-Moore RH, et al. A comparison of black and white women with binge eating disorder. Am J Psychiatry. 2001;158:1455-1460.

12. Crow S, Kendall D, Praus B, Thuras P. Binge eating and other psychopathology in patients with type II diabetes mellitus. Int J Eat Disord. 2001;30:222-226.

13. Willett WC, Dietz WH, Colditz GA. Guidelines for healthy weight. N Engl J Med. 1999;341:427-434.

14. Ferguson KJ, Spitzer RL. Binge-eating disorder in a community-based sample of successful and unsuccessful dieters. Int J Eat Disord. 1995;18:167-172.

15. Wilfley DE, Agras WS, Telch CF, et al. Group cognitive-behavioral therapy and group interpersonal psychotherapy for the nonpurging bulimic individual: a controlled comparison. J Consult Clin Psychol. 1993;61:296-305.

16. Powers PS, Perez A, Boyd F, Rosemurgy A. Eating pathology before and after bariatric surgery: a prospective study. Int J Eat Disord. 1999;25:293-300.

17. Adami GF, Gandolfo P, Bauer B, Scopinaro N. Binge eating in massively obese patients undergoing bariatric surgery. Int J Eat Disord. 1995;17:45-50.

18. Walsh BT, Devlin MJ. Eating disorders: progress and problems. Science. 1998;280:1387-1390.

19. Spitzer RL, Devlin M, Walsh BT, et al. Binge eating disorder: a multisite field trial of the diagnostic criteria. Int J Eat Disord. 1992;11:191-203.

20. Hudson JI, Pope HG, Jr., Carter WP. Pharmacologic therapy of bulimia nervosa. In: Goldstein D, ed. The Management of Eating Disorders and Obesity. Totowa, NJ: Humana Press; 1999:19-32.

21. Alger SA, Schwalberg MD, Bigaouette JM, et al. Effect of a tricyclic antidepressant and opiate antagonist on binge-eating behavior in normoweight bulimic and obese binge-eating subjects. Am J Clin Nutr. 1991;53:865-871.

22. McCann UD, Agras W. Successful treatment of nonpurging bulimia nervosa with desipramine: A double-blind, placebo-controlled study. Am J Psychiatry. 1990;147:1509-1513.

23. Gardiner HM, Freeman CPL, Jesinger DK, et al. Fluvoxamine: an open pilot study in moderately obese female patients suffering from atypical eating disorders and episodes of bingeing. Int J Obes Relat Metab Disord. 1993;17:301-305.

24. Hudson JI, McElroy SL, Raymond NC, et al. Fluvoxamine in the treatment of binge-eating disorder: A multi-center placebo-controlled, double-blind trial. Am J Psychiatry. 1998;155:1756-1762.

25. McElroy SL, Arnold LM, Shapira NA, et al. Placebo-controlled trial of sertraline in the treatment of binge-eating disorder. Am J Psychiatry. 2000;15:1004-1006.

26. McElroy SL, Hudson JI., Malhotra S, et al. Citalopram in the treatment of binge-eating disorder: A placebo-controlled trial. J Clin Psychiatry. Inpress.

27. Grilo CM. A controlled study of cognitive behavioral therapy and fluoxetine for binge eating disorder. Poster presented at: the Annual Meeting of the Eating Disorders Research Society; November 2002; Charleston, SC.

28. Arnold LM, McElroy SL, Hudson JI, et al. A placebo-controlled randomized trial of fluoxetine in the treatment of binge-eating disorder. J Clin Psychiatry. 2002;63:1028-1033.

29. Stunkard A, Berkowitz R, Tanrikut C, et al. D-fenfluramine treatment of binge eating disorder. Am J Psychiatry. 1996;153:1455-1459.

30. Appolinario JC, Godoy-Matos A, Fontenelle LF, et al. An open-label trial of sibutramine in obese patients with binge-eating disorder. J Clin Psychiatry. 2002;63:28-30.

31. McElroy SL, Arnold LM, Shapira NA, et al. Topiramate in the treatment of binge eating disorder associated with obesity. Am J Psychiatry. 2003;160:255-261.

32. Hansen DL, Toubro S, Stock MJ, et al. The effect of sibutramine on energy expenditure and appetite during chronic treatment without dietary restriction. Int J Obes Relat Metab Disord. 1999;23:1016-1024.

33. Artigas F, Nutt DJ, Shelton R. Mechanism of action of antidepressants. Psychopharmacol Bull. 2002;36(suppl 2):123-132.

34. Bray GA, Blackburn GL, Ferguson JM, et al. Sibutramine produces dose-related weight loss. Obes Res. 1999;7:189-198.

35. Bymaster FP, Dreshfield-Ahmad LJ, Threlkeld PG, et al. Comparative affinity of duloxetine and venlafaxine for serotonin and norepinephrine transporters in vitro and in vivo, human serotonin receptor subtypes, and other neuronal receptors. Neuropsychopharmacology. 2001;25:871-880.

36. Malhotra S, King KH, Welge JA, et al. Venlafaxine treatment of binge-eating disorder associated with obesity: A series of 35 patients. J Clin Psychiatry. 2002;63:802-806.

37. Mitchell JE, Gosnell BA, Roerig JL, et al. Effects of sibutramine on binge eating, hunger, and fullness in a laboratory human feeding paradigm. Obes Res. 2003;11:599-602.

38. Wermuth BM, Davis KL, Hollister LE, Stunkard AJ. Phenytoin treatment of the binge-eating syndrome. Am J Psychiatry. 1977;134:1249-1253.

39. Hudson JI, Pope HG Jr. The role of anticonvulsants in the treatment of bulimia. In: McElroy SL, Pope HG Jr, eds. Use of Anticonvulsants in Psychiatry: Recent Advances. Clifton, NJ: Oxford Health Care; 1988:141-154.

40. Krüger S, Kennedy SH. Psychopharmacotherapy of anorexia nervosa, bulimia nervosa and binge-eating disorder [review]. J Psychiatry Neurosci. 2000;25:497-508.

41. McIntyre RS. Psychotropic drugs and adverse events in the treatment of bipolar disorders revisited. J Clin Psychiatry. 2002;63(suppl 3):15-20.

42. Biton V, Mirza W, Montouris G, et al. Weight change associated with valproate and lamotrigine monotherapy in patients with epilepsy. Neurology. 2001;56:172-177.

43. Calabrese JR, Shelton MD, Rappaport DJ, et al. Bipolar disorders and the effectiveness of novel anticonvulsants. J Clin Psychiatry. 2002;63(suppl 3):5-9.

44. Bergen DC, Ristanovic RK, Waicosky K, et al. Weight loss in patients taking felbamate. Clin Neuropharmacol. 1995;18:23-27.

45. French J, Smith M, Faught E, et al. Practice advisory: The use of felbamate in the treatment of patients with intractable epilepsy. Epilepsia. 1990;40:803-808.

46. Pellock JM, Appleton R. Use of new antiepileptic drugs in the treatment of childhood epilepsy. Epilepsia. 1999;40(suppl 6):29-38.

47. Gadde KM, Francicsy DM, Wagner HR II, et al. Zonisamide for weight loss in obese adults: a randomized controlled trial. JAMA. 2003;289:1820-1825.

48. Hoopes SP, Reimherr FW, Kamin M, et al. Topiramate treatment of bulimia nervosa. Presented at: the Annual Meeting of the American Psychiatric Association; May 2002; Philadelphia, PA.

49. Reimherr FW, Hoopes SP, Karvois D, et al. Topiramate in the treatment of bulimia nervosa: Additional efficacy. Presented at: the Annual Meeting of the American Psychiatric Association; May 2002; Philadelphia, PA.

50. Shapira NA, Goldsmith TD, McElroy SL. Treatment of binge eating disorder with topiramate: A clinical case series. J Clin Psychiatry. 2000;61:368-372.

51. Appolinario JC, Fontenelle LF, Papelbaum M, et al. Topiramate use in obese patients with binge eating disorder: An open study. Can J Psychiatry. 2002;47:271-273.

52. Suppes T. Review of the use of topiramate for treatment of bipolar disorders. J Clin Psychopharmacol. 2002;22:599-609.


Print Friendly 

Erasmo A. Passaro, MD
 Primary Psychiatry. 2003;10(10):72-79

Dr. Passaro is director of the Bayfront Medical Center Comprehensive Epilepsy Program in St. Petersburg, Florida.

Disclosure: Dr. Passaro is a consultant for and on the speaker’s bureau of Ortho-McNeil.

Please direct all correspondence to: Erasmo A. Passaro, MD, Bayfront Medical Center Comprehensive Epilepsy Program, 601 7th St South, St. Petersburg, FL 33701; Tel: 727-824-7149; Fax: 727-824-7133; E-mail:; Web site:

Focus Points

Psychiatric comorbidity in epilepsy is underrecognized and contributes to reduced quality of life in epilepsy patients.

Comorbid psychiatric symptoms most commonly occur in the inter-ictal state of epilepsy, but can also occur in the ictal and post-ictal states.

Inter-ictal depression is 17-fold greater in patients with temporal lobe epilepsy, and the risk for suicide is 5-fold greater than in the general population.

The most common ictal psychiatric symptom is fear and anxiety that typically lasts several seconds and can be accompanied by autonomic symptoms.

The inter-ictal psychosis of epilepsy differs from schizophrenia.

In epilepsy patients, pathology within the hippocampus and the amygdala may increase the risk for psychiatric comorbidity.



A relationship between psychiatric disorders and epilepsy has been recognized for several centuries. This psychiatric comorbidity manifests as psychoses, mood disorders, anxiety disorders, and personality disorders. Psychiatric disorders are classified in three categories with regard to their relationship to seizures: inter-ictal—the state during which the patient is not having seizures; ictal—psychiatric symptoms during the seizure; and3 post-ictal—psychiatric symptoms that are followed by a seizure. The most common inter-ictal psychiatric disorders in epilepsy patients are depression and inter-ictal anxiety. Ictal psychiatric symptoms, such as ictal psychosis and ictal depression, are rare. Ictal fear and anxiety, on the other hand, are common. Post-ictal psychosis has been well described, while post-ictal depression and anxiety have not been well characterized. Although psychiatric disorders in epilepsy have been known for centuries, they are often underrecognized and undertreated. This psychiatric comorbidity contributes to the stigma associated with epilepsy.




The relationship between epilepsy and psychiatric disorders has been recognized for several centuries (Table 1).1-6 Throughout the 20th century, investigators looked for an association between behavioral disturbances, psychiatric syndromes, or personality disorders in patients with temporal lobe epilepsy (TLE).6-8 This relationship continued to be misunderstood; in fact, through part of the 20th century, epilepsy was classified as a major functional psychosis.9

Psychiatric comorbidity manifests as psychoses, mood disorders, anxiety disorders, and personality disorders. These psychiatric disorders are classified with regard to their relationship to the seizures: (1) inter-ictal—psychiatric disorder occurs during a time when the patient is not having seizures; (2) ictal—psychiatric symptoms occur during the seizure; and (3) post-ictal—psychiatric symptoms are followed by a seizure.

Comorbidity in epilepsy is often unrecognized and untreated, contributing to impaired quality of life10 and to the stigma associated with epilepsy.11 In fact, >50% of epilepsy patients report stigma as one of their major concerns.11 The relationship between psychiatric disorders and epilepsy is further illustrated by the psychotropic effect of antiepileptic drugs (AEDs). For example, many AEDs have mood-stabilizing properties. Infrequently, AEDs can produce psychiatric adverse effects, such as psychosis or mood disorders. This article reviews the clinical features of the psychiatric comorbidity of epilepsy.


Psychoses in Epilepsy


The prevalence of psychosis in patients with epilepsy is greater than in the general population.12 Psychosis in epilepsy most often occurs in the inter-ictal or post-ictal state, and rarely in the ictal state. In the 1950s, Bartlett13 reported that 1.1% of a patient group suffered from both epilepsy and psychosis, but others have reported different prevalence findings, such as 6%,14 9.4%,8 and 25%.15 The variability in these early findings can be explained by the heterogeneity of the inclusion criteria for psychosis, and the lack of a refined classification system, such as the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition-Text Revision (DSM-IV-TR).16


Inter-Ictal Psychosis


Inter-ictal psychosis is defined as a chronic psychosis that is not temporally related to the occurrence of seizures or medication side effect. While epilepsy patients can have comorbid schizophrenia, the inter-ictal psychosis of epilepsy can fulfill DSM-IV-TR criteria for schizophrenia or a delusional disorder, the characteristics of which and prognoses can be very different.


Inter-ictal psychosis accounts for 10% to 30% of cases of psychosis in epilepsy.17 In one study,18 the prevalence of schizophrenia, diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders, Third Edition-Revised (DSM-III-R),19 was found to be nine times greater in an epilepsy clinic than in a migraine clinic. Another study found a standardized incidence ratio of 1:48 for all epilepsy patients and 2:35 for TLE patients.20


The inter-ictal psychosis of epilepsy usually presents with paranoid delusions, suspiciousness, and hypervigilance toward the environment; delusions are usually less circumscribed and have a religious or a persecutory content.21 Taylor22 found that mesial temporal sclerosis is an important risk factor for psychosis in epilepsy patients. A more recent study by Kanemoto and colleagues23found similar results.


Ictal Psychosis


Ictal psychosis, defined as psychotic symptoms during a simple partial seizure, is very rare. In one series, it was observed in 4 of 29 cases of simple partial (focal) status epilepticus.24 Sometimes a visual or an auditory aura can be misinterpreted as a psychosis. In one case report, a patient had complex visual hallucinations for 2 weeks associated with an ictal discharge.25 However, unlike patients with psychosis, patients with prolonged visual or auditory simple partial seizures recognize that the phenomena they are experiencing are not real. Some patients in nonconvulsive status epilepticus are mistakenly diagnosed as psychotic because they are intermittently responsive, with inappropriate nonpurposeful behavior and waxy flexibility.26 Although ictal psychosis is rare, it is probably often misdiagnosed since simple partial seizures (focal seizures without loss of awareness) only show an electroencephalogram (EEG) correlate approximately 25% of the time.27 In patients with a history of epilepsy and stereotyped episodes of psychosis, ictal psychosis should be excluded. If the index of suspicion is high, and the EEG shows either no change or a nonepileptiform pattern, ictal subtraction single photon emission computerized tomography should be considered for definitive diagnosis.28


Post-Ictal Psychosis


Post-ictal psychosis occurs following a single seizure or a cluster of seizures, and accounts for approximately 25% of psychosis in epilepsy.29 It may be preceded by a lucid period with psychotic symptoms that follow within 24–72 hours.29 Since there may be a latency of a few days between the seizures and the onset of the psychosis, the relationship between the seizure(s) and the psychosis is often not initially recognized. This is particularly true in patients with nocturnal seizures or those who live alone, cases where seizures are often unrecognized. Post-ictal psychosis is often accompanied by a prominent alteration in mood that usually lasts for an average of 14 days with spontaneous recovery being the usual course.29 Post-ictal psychosis can be disabling since the patient’s relationships at home and at work are often strained by the unpredictable nature of the psychotic episodes. Post-ictal psychosis is most frequently associated with TLE.


Mood Disorders


Inter-ictal depression is the most common mood disorder in epilepsy. Bipolar mood disorder is infrequently observed in epilepsy patients. Ictal mood disorders are rare, and post-ictal mood disorders are uncommon and have been infrequently described.


Inter-Ictal Depression


A relationship between epilepsy and depression was first described in 400 B.C. by Hippocrates, who observed a relatively high frequency of “melancholia” among epilepsy patients.30 Among patients with TLE, a history of depression is 17 times more frequent than controls.31 Patients with epilepsy and depression are at higher risk for psychiatric hospitalization than nonepileptic depressed patients.32 Risk factors for depression include male gender, left-sided lesion, and use of polypharmacy to control the seizures.32 The severity of the depression is not related to the duration of epilepsy, the frequency of seizures, or a positive family history of depression.33 Depression in epilepsy patients is rarely of the bipolar type.


Ictal Depression


An association between ictal depression and the experience of olfactory hallucinations supporting an anteromesial temporal localization has been noted by several authors.34 Gloor and colleagues35 reported on three patients who experienced depression or guilt only when temporolimbic structures were stimulated. While inter-ictal depression and prodromal depressive symptoms are common, ictal depression is rarely observed.36 Rare cases have reported an association between simple partial and absence status epilepticus and prolonged depressive affect.36


Post-Ictal Depression and Mania


One study found that 65% of poorly controlled patients with epilepsy experienced post-ictal depressive symptoms with a mean duration of 37 hours.37 Such symptoms are more common in patients with frontal or temporal lobe seizures without a predominant lateralization.38 Rarely, post-ictal hypomania has been described.39


Inter-Ictal Mania


Although the prevalence of bipolar affective disorder has not been formally assessed, some reports suggest that the prevalence of bipolar affective disorder in patients with epilepsy is between 0.1% and 4.3%.40 The current use of AEDs that are effective in bipolar illness may contribute to the lack of manic episodes in epilepsy patients. However, mania was uncommon even before the use of current AEDs.41 Population-based studies comparing the incidence of bipolar illness in epilepsy patients as compared to the general population have not been done.


Anxiety Disorders


Anxiety is a common comorbid condition in epilepsy patients.14,42,43 Anxiety symptoms are most commonly observed in patients with TLE. It occurs in the inter-ictal, ictal, or post-ictal state. Ictal anxiety and fear represent the most common forms of ictal affect in TLE patients.44,45


Inter-Ictal Anxiety


Anxiety conditions, such as generalized anxiety disorder and panic disorder with or without agoraphobia, are more prevalent in the epilepsy population than in the normal population.46 Currie and colleagues14 found anxiety disorders in 19% of TLE patients. In refractory epilepsy patients considered for epilepsy surgery, anxiety was the most common Axis I DSM-III-R diagnosis present in 10.7% of patients42 and in 32% in another study.43 Seizure severity was found to be a significant predictor of anxiety in 100 patients with medication-resistant epilepsy.47 Anxiety symptoms can develop as anticipation of a seizure sometimes leading to avoidance behaviors.48 AED withdrawal can also precipitate anxiety symptoms.49


Sometimes the autonomic symptoms of simple partial seizures, such as increased heart rate, shortness of breath, and diaphoresis, may simulate panic disorder.50 In the author’s experience, many patients with simple partial seizures with autonomic symptoms are mistakenly diagnosed with anxiety until a secondarily generalized tonic-clonic seizure immediately follows the autonomic symptoms.


Ictal Fear and Anxiety


Anxiety and fear also represent the most common form of ictal affect in patients with epilepsy.45 It can be accompanied by a rising epigastric sensation, palpitations, diaphoresis, mydriasis, and pallor.51 An early study reported ictal fear in up to 35% of patients with temporal lobe epilepsy,45 but a more recent study found an incidence of 10% to 15%.52 Up to 33% of patients with ictal fear have been found to have a comorbid panic disorder.53


The amygdala is important in the regulation of emotion. Stimulation of the amygdala in humans, for example, produces ictal fear.51 Simple partial status has been reported in a patient with depth electrode recording of seizures in the amygdala.54 Cendes and colleagues55 studied 50 patients with TLE and compared volumetric measurements of the amygdala in patients with ictal fear to patients with other types of auras. They found that TLE patients with ictal fear had smaller amygdalar volumes than TLE patients without this aura. The accompanying Figure illustrates the magnetic resonance imaging of a patient with atrophy of the amygdala and ictal fear. It is likely that ictal fear arises from seizures originating or propagating to the amygdala.


Post-ictal Anxiety


Post-ictal anxiety, which lasts for 12–24 hours, has received little attention in the literature.56 In the author’s experience, a prodrome of anxiety that lasts for hours or days and is subsequently relieved after a seizure, is more frequently observed.


Personality Disorders


The existence and specificity of a characteristic behavioral syndrome in TLE patients is controversial. The behavior pattern of many epilepsy patients differs from that of age, sex, and socioeconomic matched control subjects. Kraepelin5 noted “meticulousness and slowing of mental process” in half of his epilepsy patients. Epilepsy patients have been described as circumstantial and tangential in their thinking.57 Gastaut and colleagues58 found that viscosity and irritability was greater in two thirds of 60 outpatients with TLE, and he emphasized that the behavioral and personality changes observed in TLE patients developed at least 2 years after the onset of clinical seizures. He also noted that the majority of his patients were hyposexual. He compared epilepsy patients to Kluver-Bucy syndrome patients who exhibited heightened emotions, viscosity, and hyposexuality. Later, Geschwind59 reported that TLE patients were overly concerned with ethical and religious issues, were unusually serious, lacked a sense of humor, and wrote excessively. Bear and Fedio60 found that patients with right TLE reported more emotional traits and minimized their behavioral changes (ie, polished their image), whereas left TLE patients had more ideational traits and often tarnished their image.


Devinsky and Najar61 recently challenged the characterization of personality changes in epilepsy patients as a disorder, viewing it rather as a change that is not necessarily maladaptive or negative. They hypothesized that the association between hypergraphia, hyposexuality, viscosity, religious concerns, and deepened emotions is unusual and may strongly suggest a limbic seizure origin. They argue that the characterization of TLE patients into narrowly defined personality traits predisposes to further stigmatization.


A study of 52 medically refractory epilepsy patients showed that 21% met the criteria for an Axis II DSM-III-R disorder.62 The presence of auras correlated with the presence of a personality disorder. Dependent and avoidant personality disorders were the most common diagnoses. Further studies are necessary to determine whether these personality changes are due to the chronic disability of epilepsy, treatment with AEDs, cognitive dysfunction, or the seizures themselves.


Antiepileptic-Related Psychiatric Effects


AEDs infrequently cause psychiatric symptoms in epilepsy patients. While these psychiatric adverse effects are sometimes observed in clinical practice, most of the reports are anecdotal. Many of these studies did not assess the confounding effect of baseline psychiatric diagnoses, concomitant medication, and the seizures.


Primidone, tiagabine, topiramate, vigabatrin, and felbamate are other AEDs known to cause depressive symptoms in some patients (Table 2).63-71 There have been no consistent reports of psychiatric adverse effects associated with lamotrigine, oxcarbazepine, gabapentin, valproate, phenytoin, and carbamazepine.64 In fact, one study comparing valproate to lamotrigine in epilepsy patients found that lamotrigine had a greater beneficial effect on health-related quality of life that correlated with measures of mood on several scales.72

A first episode of psychosis may occur in association with changes in AEDs.71 These symptoms can occur with the introduction of AED add-on therapy, abrupt AED discontinuation, and after AED overdose. Some cases of psychosis are related to AED withdrawal.

Barbiturate and benzodiazepine withdrawal may produce anxiety, irritability, psychosis, and delirium.64 There have been anecdotal reports of psychiatric symptoms with the introduction of phenytoin and carbamazepine, but none have been consistent.64


Psychiatric symptoms also occur with AED withdrawal. For example, Ketter and colleagues73 showed that 40% of epilepsy patients withdrawn from phenytoin, carbamazepine, and valproate, developed moderate to severe psychiatric symptoms. Depression and anxiety were most common, and psychosis occurred less frequently. AED withdrawal-induced psychopathology usually occurs in the final week of the AED taper, and tends to resolve within 2 weeks of restarting the original AED.64


When evaluating patients with possible AED-related psychiatric changes, it is important to inquire about a history of similar psychiatric symptoms and their frequency, and to determine if there is a temporal relationship between the initiation or the discontinuation of the AED. In addition, AEDs should be introduced slowly and withdrawn slowly. Both zonisamide and topiramate reduce appetite and cause weight loss, and therefore should be used with caution in patients with eating disorders. Many of the reports of AED-induced psychiatric changes have occurred with AED polytherapy. In this regard, pharmacodynamic interactions between AEDs may be partly responsible for the AED-induced psychiatric changes. Patients who are at greater risk for AED-induced psychiatric changes, such as those with a past psychiatric history, brain injury, or mental retardation, should be closely monitored.


The mechanism for AED-induced psychiatric changes is largely unknown. Some have speculated that AEDs with predominant g-aminobutyric acid (GABA)ergic activity, such as benzodiazepines, barbiturates, tiagabine, and vigabatrin, produce sedative psychiatric adverse effects (ie, depression).64 AEDs that attenuate antiglutamatergic excitatory activity either presynaptically or postsynaptically, such as lamotrigine and felbamate, are likely to produce activating psychiatric effects. Topiramate, which has both GABA-enhancing and glutamate-attenuating properties, can produce a combination of sedating and activating effects.74 However, there is no empirical data to support these mechanistic hypotheses.




The most common psychiatric disorders associated with epilepsy are inter-ictal depression and anxiety. Inter-ictal psychosis is observed with greater frequency in TLE patients than in the general population, while inter-ictal mania is rarely observed. Ictal fear and anxiety, the most common ictal psychiatric symptoms, are usually brief (lasting seconds to minutes) and stereotyped. They can occur in isolation or as part of an amygdala pathology-associated aura prior to a complex partial seizure. Ictal fear/anxiety can be confused with panic disorder since the two conditions share similar features. Ictal psychosis and ictal depression, on the other hand, are rarely observed. Post-ictal psychiatric changes, such as post-ictal psychosis, post-ictal anxiety, and post-ictal depression, occur less frequently and their association with seizures sometimes go unrecognized.


The existence of an inter-ictal personality disorder of TLE is controversial, and such a characterization further contributes to the stigma associated with epilepsy since these traits have a pejorative connotation. However, avoidant and dependent personality disorders are more commonly observed in TLE patients, particularly, those with auras.


AEDs may also produce psychiatric side effects that should always be considered as a cause of psychiatric symptoms. These include depression, irritability, agitation, and psychosis. AED withdrawal can provoke transient psychiatric symptoms, and therefore AEDs should always be slowly tapered when they are discontinued.


Refractory TLE patients, particularly those with hippocampal sclerosis or amygdalar pathology, are at greatest risk for developing psychiatric comorbidity. Early recognition of psychiatric comorbidity and understanding the psychiatric adverse effects of AEDs, will allow for prompt psychiatric referral and treatment to reduce this comorbidity and improve overall quality of life.75 PP



1. Temkin O. The Falling Sickness: A History of Epilepsy From the Greeks to the Beginnings of Modern Neurology. 2nd ed rev. Baltimore, MD: Johns Hopkins University Press; 1971:55.


2. Falret J. De l’Etat mental des l’epileptiques. Arch Gen Med. 1860;16:666-679. French.


3. Jackson JH. West riding lunatic asylum medical reports. In: Selected Writings of John Hughlings Jackson. New York, NY: Basic Books; 1958:119.


4. Clouston TS. Clinical Lectures on Mental Diseases. London, England: J and A Churchill; 1861.


5. Kraepelin E. Psychiatrie. 8th ed. Leipzig, Germany: Barth; 1924.


6. Gibbs FA, Gibbs EL, Lennox WG. Cerebral dysrhythmias of epilepsy: measures for their control. Arch Neurol Psychiatry. 1938:39:298-314.


7. Lennox WG. Phenomena and correlates of the psychomotor triad. Neurology. 1951;1:357-371.


8. Gibbs EL, Gibbs FA, Fuster B. Psychomotor epilepsy. Arch Neurol Psychiatry. 1948;60:331-339.


9. Jaspers K. General Psychopathology. Hoenig J, Hamilton MW, trans. Manchester, England: University Press; 1962:606.


10. Hermann BP, Seidenberg M, Bell B. Psychiatric comorbidity in chronic epilepsy: identification, consequences, and treatment of major depression. Epilepsia. 2000;41:(suppl 2):S31-S41.


11. Baker GA, Brooks J, Buck D, et al. The stigma of epilepsy: a European perspective. Epilepsia. 1999;41:98-104.


12. Stevens JR. Psychosis and the temporal lobe, in neurobehavioral problems in epilepsy. In: Advances in Neurology. Vol 55. Smith D, Treiman D, Trimble M, eds. New York, NY: Raven Press; 1991:79-96.


13. Bartlett JE. Chronic psychosis following epilepsy. Am J Psychiatry. 1957;114:338-343.


14. Currie S, Heathfield KWG, Henson RA, et al. Clinical course and prognosis of temporal lobe epilepsy. Brain. 1971;94:173-190.


15. Hill D, Pond DW, Mitchell W, et al. Personality changes following temporal lobectomy for epilepsy. J Ment Sci. 1957;103:18-27.


16.Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.


17. Dongier S. Statistical study of clinical and electroencephalographic manifestations of 536 psychotic episodes occurring in 516 epileptics between clinical seizures. Epilepsia. 1959;1:117-142.


18. Mendez MF, Grau R, Doss RC, et al. Schizophrenia in epilepsy: seizure and psychosis variables. Neurology. 1993;43:1073-1077.


19.Diagnostic and Statistical Manual of Mental Disorders. 3rd ed rev. Washington, DC:?American Psychiatric Association; 1987.


20. Bredkjaer SR, Mortensen PB, Parnas J. Epilepsy and non-organic affective psychosis. National Epidemiology study. Br J Psychiatry. 1997;170:515-519.


21. Kristensen O, Sindrup HH. Psychomotor epilepsy and psychosis. I. Physical Aspects. Acta Neurol Scand. 1978;57:361-369.


22. Taylor DC. Mental state and temporal lobe epilepsy:?a correlative account of 100 patients treated surgically. Epilepsia. 1972;13:727-765.


23. Kanemoto K, Takeuchi J, Kawasaki J, Kawai I. Characteristics of temporal lobe epilepsy with mesial temporal sclerosis, with special reference to psychotic episodes. Neurology. 1996;47:1199-1203.


24. Wolf P. Status epilepticus and psychosis. In: XIII Epilepsy International Symposium. New York, NY: Raven Press; 1982:211-217.


25. Sowa MV, Pituck S. Prolonged spontaneous complex visual hallucinations and illusions as ictal phenomena. Epilepsia. 1989;30:524-526.


26. Kaplan PW. Nonconvulsive status epilepticus. Semin Neurol. 1996;16:33-40.


27. Devinsky O, Kelley K, Porter RJ, Theodore WH. Clinical and electroencephalographic features of simple partial seizures. Neurology. 1988;9:1347-1352.


28. Thomas P, Zifkin B, Migneco O, Lebrun C, Darcourt J, Andermann F. Nonconvulsive status epilepticus of frontal origin. Neurology. 1999;52:1174-1183.


29. Logsdail S, Toone BK. Post-ictal psychoses: a clinical and phenomenological description. Br J Psychiatry. 1988;152:246-252.


30. Lewis A. Melancholia: a historical review.
J Ment Sci. 1934;80:1-42.


31. Forsgren L, Nystrom L. An incident case referent study of epileptic seizure in adults. Epilepsy Res. 1990;6:66-81.


32. Mendez MF, Cumming J, Benson F. Depression in epilepsy: significance and phenomenology. Arch Neurol. 1986;43:766-770.


33. Indaco A, Carrieri P, Nappi C. Interictal depression in epilepsy. Epilepsy Res. 1992;12:45-50.


34. Weil AA. Depressive reactions associated with temporal lobe-uncinate seizures. J Nerv Ment Dis. 1955;150:173-176.


35. Gloor P, Oliver A, Quesney LF, et al. The role of the limbic system in experiential phenomena of temporal lobe epilepsy. Ann Neurol. 1982;12:129-144.


36. Wells CE. Transient Post-ictal Psychosis. Arch Gen Psychiatry. 1975;32:1201-1203.


37. Soto A, Kanner A, Hershowitz L. Postictal psychiatric symptoms in patients with poorly controlled seizures: a prevalence study. Epilepsia. 1997;38(suppl 8):155.


38. Kanner A, Nieto JC. Depressive disorders in epilepsy. Neurology. 1999;53(suppl 2):S26-S32.


39. Barczak P, Edmunds E, Betts T. Hypomania following complex partial seizures. Br J Psychiatry. 1988;152:137-139.


40. Roberston MM. Affect and mood in epilepsy: an overview with a focus on depression. Acta Neurol Scand Suppl. 1992;140:127-132.


41. Wolf P. Manic episodes in epilepsy. In: Akimoto H, Kazamatsuri H, Seino M, et al, eds. Advances in Epileptology: Thirteenth Epilepsy International Symposium. New York, NY: Raven Press; 1982:237-240.


42. Manchanda R, Schaefer B, Mc Lachlan RS, et al. Psychiatric disorders in candidates for surgery for epilepsy. J Neurol Neurosurg Psychiatry. 1996;61:82-89.


43. Victoroff J. DSM- III-R psychiatric diagnosis in candidates for epilepsy surgery: lifetime prevalence. Neuropsychol Behav Neurol. 1994;7:87-97.


44. Williams D. The structure of emotions reflected in epileptic experiences. Brain. 1956;79:29-67.


45. Bingley T. Mental symptoms in temporal lobe epilepsy and temporal lobe gliomas. Acta Psychiatr Neurol Scand. 1958;33(suppl):1-51.


46. Weilberg JP, Bear DM, Sachs G. Three patients with concomitant panic attacks and seizure disorder: possible clues to the neurology of anxiety. Am J Psychiatry. 1983;144:1053-1056.


47. Smith DF, Baker GA, Dewey M, et al. Seizure frequency, patient perceived severity and the psychosocial consequences of intractable epilepsy. Epilepsy Res. 1991;9:231-241.


48. Hermann BP, Dikmen S, Schwartz MS, et al. Interictal psychopathology in patients with ictal fear: a quantitative investigation. Neurology. 1982;32:7-11.


49. Malow BA, Blaxton TA, Stertz B, et al. Carbamazepine withdrawal: effects of taper rate on seizure frequency. Neurology. 1993;43:2280-2284.


50. Williamson PD, Thadani VM, French JA, et al. Medial temporal lobe epilepsy: videotape analysis of objective clinical seizure characteristics. Epilepsia. 1998;39:1182-1188.


51. Gloor P, Olivier A, Quesney LF, et al. The role of the limbic system in experiential phenomena of temporal lobe epilepsy. Ann Neurol. 1982;12:129-144.


52. Devinsky O, Sato S, Theodore WH, et al. Fear episodes due to limbic seizures with normal ictal scalp EEG: a subdural electrographic study. J Clin Psychiatry. 1989;50:28-30.


53. Mintzer S, Lopez F. Comorbidity of ictal fear and panic disorder. Epilepsy Behav. 2002;3:330-337.


54. Mc Lachlan RS, Blume WT. Isolated fear in complex partial status epilepticus. Ann Neurol. 1980;8:639-641.


55. Cendes F, Andermann F, Gloor P, et al. Relationship between atrophy of the amygdala and ictal fear in temporal lobe epilepsy. Brain. 1994;117:739-746.


56. Blumer D, Altschuler LL. Affective disorders. In: Engel J, Pedley TA, eds. Epilepsy: a Comprehensive Textbook. New York, NY: Lippincott-Raven Press; 1997:2085-2099.


57. Blumer D. Personality disorders in epilepsy. In: Ratey JJ, ed. Neuropsychiatry of Personality Disorders. Boston, Mass: Blackwell Science; 1995:230-263.


58. Gastaut H, Morin G, Lesevre N. Etude du comportement des epileptiques psychomoteurs dans I’intervalle de leurs crises: les troubles de l’activete global et de la sociabilite. Ann Med Psychol (Paris). 1955;113:1-27. French


59. Geschwind N. Behavioural changes in temporal lobe epilepsy. Psychol Med. 1979;9:217-219.


60. Bear DM, Fedio P. Quantitative analysis of inter-ictal behavior in temporal lobe epilepsy. Arch Neurol. 1977;34:454-467.


61. Devinsky O, Najar S. Evidence against the existence of a temporal lobe epilepsy personality syndrome. Neurology. 1999;53(suppl 2):S13-S25.


62. Lopez-Rodriguez F, Altshuler L, Kay J, Delrahim S, Mendez M, Engel J Jr. Personality disorders among medically refractory epileptic patients. J Neuropsychiatry Clin Neurosci. 1999;11:464-469.


63. Beydoun A, Passaro EA. Anticonvulsants in psychiatric disorders. In: Ettinger AB, Kanner AM. Psychiatric Issues in Epilepsy. Philadelphia, Penn: Lippincott Williams and Wilkins; 2001:251-260.


64. Ketter TA, Post RM, Theodore WH. Positive and negative psychiatric effects of anti-epileptic drugs in patients with seizure disorders. Neurology. 1999;53(suppl 2):S53-S67.


65. Wolf SM, Shinnar S, Kang H, Gil KB, Moshe SL. Gabapentin toxicity in children manifesting as behavioral changes. Epilepsia. 1995;12:1203-1205.


66. Shorvon SD, Lowenthal A, Janz D, et al. Multi-center double blind randomized placebo controlled trial of levetiracetam as add-on therapy in patients with refractory partial seizures. Epilepsia. 2000;41:1179-1186.


67. Kossoff EH, Bergey GK, Freeman JM, Vining EP. Levetiracetam psychosis in children with epilepsy. Epilepsia. 2001;42:1611-1613.


68. Dietrich DE, Kropp S, Emrich HM. Oxcarbazepine in affective and schizoaffective disorders. Pharmacopsychiatry. 2001;6:242-250.


69. Dohmeier C, Kay A, Greathouse N. Neuropsychiatric complications of topiramate therapy. Epilepsia. 1998;39(suppl. 6):189.


70. Miyamoto T, Kohsaka, Koyama T. Psychotic episodes during zonisamide treatment. Seizure. 2000;9:65-70.


71. Matsuura M. Epileptic psychosis and anticonvulsant treatment. J Neurol Neurosurg Psychiatry. 1999;76:231-233.


72. Chris Sackellares J, Jacqueline Kwong W, Vuong A, Hammer AE, Barrett PS. Lamotrigine monotherapy improves health-related quality of life in epilepsy: a double-blind comparison with valproate. Epilepsy Behav. 2002;3:376-382.


73. Ketter TA, Malow BA, Flamini R, et al. Anticonvulsant withdrawal-emergent psychopathology. Neurology. 1994;44:55-61.


74. Petroff OA, Hyder F, Mattson RH, Rothman DL. Topiramate increases brain GABA, homocarnosine, and pyrrolidinone in patients with epilepsy. Neurology. 1999;52:473-478.


75. Hermann BP, Seidenberg M, Bell B. Psychiatric co-morbidity in chronic epilepsy: identification, consequences and treatment of major depression. Epilepsia. 2000;41(suppl 2): S31-S41.


Print Friendly 

Hugh Myrick, MD, Robert Malcolm, MD, and Raymond Anton, MD

Dr. Myrick is assistant professor of psychiatry and Drs. Malcolm and Anton are professors of psychiatry in the Department of Psychiatry and Behavioral Sciences at the Medical University of South Carolina in Charleston.


Disclosure: Dr. Myrick is on the speaker’s bureau of Abbott, Forest, and Pfizer and has received research support from Abbott, Pfizer, and UCB Pharma; Dr. Malcolm is a consultant for Bristol-Myers Squibb, Cephalon, and Ortho-McNeil, is on the speaker’s bureau for Cephalon and Eli Lilly, and has received research support from GlaxoSmithKline and Sanofi; and Dr. Anton is a consultant for Ortho-McNeil and Pfizer.

Funding/support: grants P50 AA10761 (HM, RM, RA) and K23 AA00314 (HM) from the National Institute on Alcohol Abuse and Alcoholism; VA Merit Grant from the Medical Research Service, Ralph H. Johnson VAMC (HM)

Please direct all correspondence to: Hugh Myrick, MD, MUSC, Institute of Psychiatry 4N, 67 President St, Charleston, SC 29425; Tel: 843-792-2727; Fax: 843-792-7353; Email:

Focus Points

Benzodiazepines, which enhance the activity of the neurotransmitter
γ-aminobutyric acid (GABA), are the standard treatment for withdrawal from alcohol dependence.

The GABA and glutamate neurotransmitter systems mediate the craving and rewarding effects of alcohol.

Many antiepileptic drugs interact with both the GABA and glutamate neurotransmitter systems and may have utility in treating and preventing substance use disorders.




Interest in the use of antiepileptics (AEDs) to treat addictive disorders has been growing, primarily because of an increased understanding of the neuropathobiology of addiction. The γ-aminobutyric acid (GABA) and glutamate neurotransmitter systems are known to be involved in mediating craving and the rewarding effects of alcohol. Although benzodiazepines are the gold standard for the treatment of alcohol withdrawal, they influence only the GABAergic pathways and also suffer from their own potential for abuse. Many AEDs interact with both the GABA and glutamate neurotransmitter systems, and thus have the potential not only for benefit during withdrawal, but also to prevent relapse to substance use. This review discusses the use of AEDs in alcohol, benzodiazepine, and opiate addiction withdrawal, and in preventing relapse to alcohol or cocaine abuse.




There is an increased understanding surrounding the use of antiepileptics (AEDs) in the treatment of addictive disorders. Preclinical and clinical trials of AEDs have generally focused on the treatment of alcohol withdrawal, alcoholism, and cocaine dependence, although other trials have examined their use in the treatment of benzodiazepine and opiate withdrawal as well.

Routine use of AEDs in the treatment of addictive disorders has been hampered by several factors. First, most of the literature concerning their use in treating addictive disorders has been published in the psychiatric literature rather in than in general medical literature. This is problematic because most of the day-to-day treatment of addictive disorders occurs in the nonpsychiatric setting. Second, with the exception of alcohol withdrawal, few double-blind studies using AEDs in the treatment of addictive disorders have been reported. Lastly, much of the research has been conducted and published in Europe and did not capture the attention of the American treatment community.

This review summarizes the use of AEDs in the treatment of addictive disorders. Specifically, the use of AEDs to treat withdrawal syndromes and to prevent relapse to alcohol and cocaine use is reviewed.


Withdrawal Symptoms



Consensus reports and clinical guidelines report that benzodiazepines are the gold standard in the treatment of alcohol withdrawal.1,2 Benzodiazepines are utilized to decrease alcohol withdrawal symptoms due to their GABAergic activity.3 However, other neurotransmitter systems, such as the excitatory amino acid glutamate, are also implicated in the general central nervous system hyperexcitability of alcohol withdrawal.4,5 Pharmacotherapies that possess novel mechanisms of action on neurotransmitters in addition to g-aminobutyric acid (GABA) may be useful and possibly better suited to the treatment of alcohol withdrawal. There is growing interest in the use of AEDs for this purpose.

There are several potential advantages to using AEDs for the treatment of alcohol withdrawal. First, AEDs do not have the abuse potential of the benzodiazepines.6 This fact is increasingly important as the majority of the treatment of alcohol withdrawal has become an outpatient procedure. Second, AED agents have been found to be useful in the treatment of affective and anxiety disorders. These disorders commonly co-occur with alcohol dependence and many of the core symptoms of affective and anxiety disorders, such as depressed mood, sleep disturbance, irritability, and anxiety, are also commonly seen during alcohol withdrawal. Third, AEDs are less likely than benzodiazepines to have an acute and additive interaction if used concomitantly with alcohol.7 Fourth, AEDs tend to blunt cognition to a greater lesser than benzodiazepines; negative cognitive effects may hamper continued work and family role performance and impede early attempts at rehabilitation counseling. Fifth, there is some indication that benzodiazepine treatment of alcohol withdrawal may lead to rebound withdrawal symptoms, which poses a high risk for relapse.8 Finally, AEDs have been shown to block neuronal sensitization or “kindling” in brain cells. Neuronal sensitization has been shown to occur as a result of multiple alcohol detoxifications. Long-term consequences, such as worsening withdrawal symptoms over repeated episodes of alcohol withdrawal, may be diminished through reduction or inhibition of this sensitization.9


Valproate, valproic acid, and divalproex have a long history of use as therapeutic agents in decreasing the symptoms of alcohol withdrawal. Initial chart reviews reported that valproate treatment decreased seizure rate and more rapidly reduced alcohol withdrawal symptoms.10,11 Reports of open-label studies in the treatment of alcohol withdrawal suggest that valproate shortens the duration and severity of withdrawal symptoms, and reduces the need for use of other medications.12-14

In a double-blind study comparing valproic acid, carbamazepine, and placebo in 138 patients, alcohol withdrawal seizures occurred in only 1 of 46 valproate-treated subjects compared to 2 of 43 carbamazepine-treated and 3 of 49 placebo-treated subjects.15 In another study, Reoux and colleagues16 randomly assigned patients to 7 days of treatment with divalproex or placebo in addition to oxazepam based on the Clinical Institute With-drawal Assessment for Alcohol–revised (CIWAA-r).17 Divalproex-treated patients required significantly less oxazepam and had significantly less increase in the CIWAA-r score, suggesting they had better control of alcohol withdrawal symptoms.


Carbamazepine was one of the first AEDs to be studied for the treatment of alcohol withdrawal syndrome. In a double-blind, placebo-controlled trial, carbamazepine was clinically and statistically superior to placebo in suppressing alcohol withdrawal symptoms, as well as improving depression, fear, and anxiety.18 Other double-blind, placebo-controlled trials have demonstrated that for the treatment of alcohol withdrawal, carbamazepine is more effective than placebo,19

59p60 equal in effectiveness to chlormethiazole (a sedative-hypnotic frequently used in Europe),20 and superior to tiapride (a dopamine antagonist antipsychotic).21 In addition, two trials have demonstrated carbamazepine to be equivalent to oxazepam in decreasing alcohol withdrawal symptoms.22,23

More recently, carbamazepine has been compared to lorazepam for the treatment of alcohol withdrawal in outpatients.8 Both drugs appeared equally effective in acutely reducing the symptoms of alcohol withdrawal. Carbamazepine, however, was superior to lorazepam in significantly reducing the number of drinks per drinking day and increasing the probability of not drinking at all after alcohol withdrawal treatment. These effects were evident in the overall sample but most pronounced in the subgroup of subjects who had undergone two or more prior alcohol detoxifications.


Gabapentin has generated interest as a potential treatment for a variety of psychiatric disorders.24 Preclinical studies have shown that gabapentin may have a role in decreasing both convulsive and anxiety-related aspects of alcohol withdrawal at doses that may cause minimal expected ataxic or sedative effects.25,26 In a small, open-label study, Bonnet and colleagues27 found that the addition of gabapentin to a standard alcohol detoxification regime of clomethiazole reduced the amount of clomethiazole needed. Myrick and colleagues13 treated 11 outpatients with open-label gabapentin. Patient received gabapentin 1,200 mg tapered to 400 mg over the 5 days of the study. Subjects had complete resolution of withdrawal symptoms as measured by the CIWAA-r.


Stuppaeck and colleagues28 conducted a 7-day, open-label trial of vigabatrin (1 g BID for 3 days) in 25 inpatients undergoing alcohol withdrawal. The symptoms of alcohol withdrawal seemed to be suppressed, but the interpretation of results are difficult because of the addition of oxazepam (as allowed by protocol) in 15 subjects. One subject, who had received a total of 250 mg of oxazepam on the first 2 days of the study, suffered a seizure on day 3.


Topiramate may prevent tonic-clonic seizures after cessation of alcohol use.29 In an open-label study, 12 alcoholic subjects experiencing one or two tonic-clonic seizures per year were treated for 30 days with topiramate 100 mg/day. No subject had a tonic-clonic seizure during the study. While the study mentions that subjects were treated with carbamazepine prior to initiation of topiramate, it is unclear if the carbamazepine was discontinued prior to alcohol cessation or was used during the alcohol withdrawal period. No measures of alcohol withdrawal symptoms were reported.




Unlike the treatment of alcohol withdrawal, there is no gold standard medication used in the treatment of benzodiazepine withdrawal. Strategies for benzodiazepine withdrawal include drug tapering and phenobarbital substitution.30 Drug tapering is most commonly used in situations where slow discontinuation of the benzodiazepine can be tolerated and the withdrawal can be managed on an outpatient basis. Phenobarbital substitution is designed to be used under close observation, such as in an inpatient unit. AEDs may provide the ability to more quickly taper the benzodiazepine while providing seizure prophylaxis and anxiolysis. Unfortunately, few studies supporting the use of AEDs in this capacity have been conducted.


Ries and colleagues31 reported on a series of nine individuals in whom the addition or substitution of carbamazepine resulted in a more rapid detoxification and better tolerability of benzodiazepine withdrawal. The only double-blind, placebo-controlled trial found that patients tapered from their benzodiazepine with the addition of carbamazepine were more abstinent at 5 weeks than those treated with placebo.32 Kline and colleagues33 examined the effect of carbamazepine on discontinuation of alprazolam in 36 patients with panic disorder and in 35 patients with generalized anxiety disorder. After the double-blind addition of carbamazepine or placebo for 1 week, alprazolam was tapered by 25% every third day. Results suggested that the panic disorder patients exhibited more withdrawal symptoms than the generalized anxiety patients and seemed to derive greater benefit from carbamazepine.


In a cases series of four patients, Apelt and Emrich34 reported that valproate successfully decreased the symptoms of benzodiazepine withdrawal in patients who took up to 25 mg of lorazepam a day for 1–18 years prior to detoxification. Two additional case reports in patients with comorbid psychiatric illnesses and benzodiazepine dependence provide further evidence of valproate’s potential role in decreasing the symptoms of benzodiazepine withdrawal.35,36 A more recent double-blind trial found that patients treated with valproate or trazodone were significantly more likely to be benzodiazepine-free 5 weeks posttaper than patients treated with placebo.37




In an open-label study, topiramate was compared to clonidine and a combination of carbamazepine/mianserin in the treatment of multidrug withdrawal.38 Fewer doses of antiemetics and sedative/antipsychotic drugs were administered to the topiramate-treated patients (n=10) than to those in the other treatment groups. In addition, Zullino and colleagues39 reported three cases in which topiramate controlled the symptoms of opiate withdrawal. Although speculative, topiramate may accomplish this by inhibiting α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors which would reduce a possible withdrawal-induced hyperglutamatergic activation of the locus coeruleus.


Relapse Prevention


During the weeks or months following cessation, substance-dependent individuals may experience a variety of symptoms, such as mood instability, anxiety, irritability, and sleep disturbance.40 Self-medication with alcohol or drugs to reduce these symptoms during early recovery may lead to relapse. AEDs may be useful in reducing these protracted withdrawal symptoms because of their mood-stabilizing and anxiolytic properties.

In addition, dimensional characteristics, such as impulsivity and irritability, often underlie many psychiatric disorders, including substance use disorders.41 A large literature supports the use of AED agents in treating impulsivity disorders.42,43 Impulsivity may be a risk factor for relapse,44 and treating it may reduce the likelihood of relapse to substance use.

Finally, the search for novel pharmacotherapies for prevention of relapse to alcohol or drug use has emphasized interactions with the dopamine and opiate reward systems, which are modulated by GABA and glutamate. Many AEDs interact with these GABA and glutamate systems.45


Alcohol Dependence


Disulfiram and naltrexone are the only medications approved by the Food and Drug Administration for the treatment of alcohol dependence. However, other pharmacotherapeutic approaches, including AEDs, have been reported to decrease relapse to alcohol use.


In one double-blind study of 29 subjects, those treated with carbamazepine had fewer drinks per drinking day and fewer heavy drinking days at months
2 and 4 of treatment.46


In a recent 12-week, double-blind, placebo-controlled trial, divalproex-treated patients had a smaller incidence of relapse to heavy alcohol use and significant decrease in irritability compared to those treated with placebo.47 In another study,48 patients treated with divalproex for moderate alcohol withdrawal had a more rapid and consistent reduction of symptoms than those treated with a benzodiazepine. In addition, more of those who were maintained on divalproex were completely abstinent after 6 weeks than those who were treated with divalproex or a benzodiazepine during the withdrawal phase only.


Individuals in early abstinence often report sleep disturbances which may lead to relapse as a way of self-medication. Karam-Hage and Brower49 treated 12 alcoholics who were still experiencing insomnia after 4 weeks of abstinence. The addition of gabapentin at night resulted in improvement in a variety of subjective sleep measures. Only two of the subjects relapsed during the 8-week trial.


An open-label study of 24 patients evaluated the effectiveness and tolerability of topiramate in controlling alcohol craving in patients with at least one other psychiatric comorbidity.50 Topiramate was initiated at 50 mg/day and titrated upwards every 3 days to a maximum dose of 400 mg/day (mean final dose=261 mg/day). Over the 10-week treatment period, alcohol consumption decreased from 5.6 drinks/day to 0.7 drinks/day. Komanduri51 reported that two patients with comorbid psychiatric disorders, including bipolar disorder, posttraumatic stress disorder, and alcoholism, became abstinent after the addition of topiramate 300 mg/day to existing medication regimens. The patients had been treated with a variety of AEDs without a reduction in alcohol use. Interestingly, in both cases the patients reported an altered taste of alcohol once topiramate was initiated.

The best evidence for the efficacy of topiramate in the treatment of alcohol dependence comes from the results of a recent double-blind study reported by Johnson and colleagues.52 In this study, 150 alcohol-dependent subjects (based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition53 criteria) were randomly assigned to treatment with placebo or topiramate (target dose=300 mg/day divided into 2 doses) for 12 weeks. Evaluation of drinking behavior occurred on a weekly basis during the study. Subjects treated with topiramate showed improvement in all measures of drinking behavior compared to those treated with placebo. For example, the topiramate group had a larger reduction in drinks per day (-6.24) than the placebo group (-3.36, P=.0006) at study endpoint. The percentage of days abstinent doubled in the topiramate group, from 21.7% at baseline to 44.2% at study end, while it decreased from 26.0% at baseline to 18.0% after 12 weeks in the placebo group (P=.0003). Other measures of improvement in the topiramate group included reduction in the number of drinks per drinking day, reduction in the percentage of heavy drinking days, and improvement in Obsessive-Compulsive Drinking Scale factor scores. In addition, the time to the first day of 14 continuous days of abstinence or nonheavy drinking was significantly shorter in the topiramate group. No serious adverse events were reported during the study. Dizziness, paresthesia, psychomotor slowing, and memory or concentration impairment were reported more frequently in the topiramate group than in the placebo group. The ability of topiramate to diminish alcohol use may be related to the agent’s effect on both the GABA and glutamate neurotransmitter systems54,55—a profile somewhat unique among the AED agents.


Cocaine Dependence


Despite multiple clinical trials, there are no FDA-approved medications for the treatment of cocaine dependence. However, several AED trials have yielded promising results in decreasing cocaine craving and cocaine use.


A recent review of five randomized studies, including 455 subjects in whom carbamazepine was compared to placebo, concluded that results did not support the use of carbamazepine for the treatment of cocaine dependence.56 A recent trial compared carbamazepine to placebo in the treatment of cocaine-dependent subjects with or without affective disorders.57 In subjects with affective illness, carbamazepine was associated with a significantly longer time to first cocaine use and had a trend toward fewer positive cocaine urines (P=.08). Carbamazepine treatment did not have any effect on cocaine use in individuals without affective disorders.


Two open-label studies58,59 suggest that valproate may be effective in controlling cocaine craving. Halikas and colleagues58 found a lower percentage of relapse to cocaine use in 55 subjects treated with divalproex who had valproate serum levels >50 µg/mL compared to subjects with serum levels <50 mg/mL. Higher serum levels were also associated with a decrease in number of days that cocaine was used and with improved levels of subjective functioning. Myrick and colleagues,59 in an 8-week pilot study of divalproex, found a decrease in cocaine craving and cocaine use. Rating of craving and frequency of craving decreased from 69% and 61% at baseline to 14% and 13%, respectively, by week 8 (P=.05). While the percent of positive urine drug screens for cocaine decreased from 64% at baseline to 18% at week 4 and 28% at week 8, this did not reach statistical significance. Retention in the study was 64% at week 4 and 50% at week 8.


Myrick and colleagues60 enrolled 30 subjects in an 8-week, open-label trial of gabapentin (600 mg BID) in cocaine-dependent subjects. Cocaine craving was significantly decreased, and positive urine drug screens for cocaine decreased from 86% at baseline to 29% at week 8 (P=.001). Limitations of the study include a small sample size, high drop-out rate, and open-label design.


Winhusen and Somoza61 compared tiagabine, sertraline, donepezil, and placebo in 67 cocaine-dependent subjects. Tiagabine, a GABA-enhancing agent, was the only compound found to be superior to placebo on measures such as global clinical improvement and subjective measures of cocaine use. In addition, there was a trend toward a reduction in urine drug screens positive for cocaine.


Treatment Considerations


There is generally no difference in the safety of AEDs use in addictive populations versus other patient populations. However, chronic substance use can lead to many health-related problems that may necessitate special laboratory monitoring or the selection of specific AED agents. The most common of these health-related problems is compromised hematopoetic and liver function, although other factors, such as the abuse potential of the prescribed drug, should be recognized.

In addition, chronic alcohol use can often lead to general bone-marrow suppression. Careful monitoring of white blood cell counts should be undertaken if alcohol-dependent patients are prescribed carbamazepine, as the agent is known to cause agranulocytosis. The use of valproate has been associated with a benign decrease in platelet counts and therefore careful monitoring of platelet function should be initiated prior to and during treatment with valproate in alcohol-dependent subjects.

Finally, individuals with addictive disorders may have compromised liver function for a variety of reasons. Liver enzymes should be monitored before initiating treatment and during treatment with AEDs that are known to be hepatotoxic, such as carbamazepine and valproate. In cases of increased liver enzymes, the use of medications with less effect on hepatic metabolism, such as gabapentin, tiagabine, or topiramate, may be warranted.




There has been an increasing amount of attention given to AEDs as potential treatments of addictive disorders. These agents may express their clinical effects by decreasing withdrawal symptoms or influencing reward systems in the brain. Except in the area of alcohol withdrawal, most of the current data had been limited to case reports and small, open-label clinical trials. However, more recent controlled studies in alcohol dependence have provided further evidence supporting the use of the AEDs as relapse-prevention agents. While many of the first-generation AEDs possessed side-effect profiles or toxicities that required laboratory monitoring, many of the newer AEDs lack such adverse effects. It is hoped that the increasing focus on these agents will lead to important advances in identifying new neurochemical targets to decrease withdrawal symptoms and reduce relapse to substance use. PP




1. Anton RF, Becker HC. Pharmacotherapy and pathophysiology of alcohol withdrawal. In:?Kranzler HR, ed. Handbook of Experimental Pharmacology. Berlin, Germany: Springer-Verlag; 1995:114.

2. Mayo-Smith MF. Pharmacological management of alcohol withdrawal. A meta-analysis and evidence-based practice guideline. American Society of Addiction Medicine Working Group on Pharmacological Management of Alcohol Withdrawal. JAMA. 1997;2:144-151.

3. Cooper BR, Viik K, Ferris RM, White HL. Antagonism of the enhanced susceptibility to audiogenic seizures during alcohol withdrawal in the rat by gamma-aminobutyric acid (GABA) and “GABA-mimetic” agents. J Pharmacol Exp Ther. 1979;3:396-403.

4. Hoffman PL, Rabe CS, Grant KA, Valverius P, Hudspith M, Tabakoff B. Ethanol and the NMDA receptor. Alcohol. 1990;3:229-231.

5. Littleton J. Neurochemical mechanisms underlying alcohol withdrawal. Alcohol Health Res World. 1998;1:13-24.

6. Griffiths RR, Sannerud VD. Abuse of a dependence on benzodiazepines and other anxiolytic/sedative drugs. In: Herbert Y, Melzer Y, eds. Psychopharmacology the Third Generation of Progress. New York, NY: Raven Press; 2003.

7. The Pharmacologic Basis of Therapeutics. 7thed. New York, NY: MacMillan Publishing Company; 1985.

8. Malcolm R, Myrick H, Roberts J, Wang W, Anton RF, Ballenger JC. The effects of carbamazepine and lorazepam on single versus multiple previous alcohol withdrawals in an outpatient randomized trial. J Gen Intern Med. 2002;5:349-355.

9. Brown ME, Anton RF, Malcolm R, Ballenger JC. Alcohol detoxification and withdrawal seizures: clinical support for a kindling hypothesis. Biol Psychiatry. 1988;5:507-514.

10. Bastie Y. Suppression by depakene of epileptic crises caused by withdrawal in alcoholic detoxification. Ann Med Psychol (Paris). 1970;128II:200-204.

11. Bonfiglio C, Falli S, Pacini A. The use of sodium dipropylacetate to prevent and treat alcoholic delirium tremens. Minerva Med. 1977;63:4233-4255.

12. Lambie DG, Johnson RH, Vijayasenan ME, Whiteside EA. Sodium valproate in the treatment of the alcohol withdrawal syndrome. Aust N Z J Psychiatry. 1980;3:213-215.

13. Myrick H, Brady KT, Malcolm R. Divalproex in the treatment of alcohol withdrawal. Am J Drug Alcohol Abuse. 2000;1:155-160.

14. Rosenthal RN, Perkel C, Singh P, Anand O, Miner CR. A pilot open randomized trial of valproate and phenobarbital in the treatment of acute alcohol withdrawal. Am J Addict. 1998;3:189-197.

15. Hillbom ME, Hjelm-Jager M. Should alcohol withdrawal seizures be treated with anti-epileptic drugs? Acta Neurol Scand. 1984;1:39-42.

16. Reoux JP, Saxon AJ, Malte CA, Baer JS, Sloan KL. Divalproex sodium in alcohol withdrawal: a randomized double-blind placebo-controlled clinical trial. Alcohol Clin Exp Res. 2001;9:1324-1329.

17. Sullivan JT, Swift RM, Lewis DC. Benzodiazepine requirements during alcohol withdrawal syndrome: clinical implications of using a standardized withdrawal scale. J Clin Psychopharmacol. 1991;5:291-295.

18. Bjorkqvist SE, Isohanni M, Makela R, Malinen L. Ambulant treatment of alcohol withdrawal symptoms with carbamazepine: a formal multicentre double-blind comparison with placebo. Acta Psychiatr Scand. 1976;5:333-342.

19. Björkqvist SE, Isohanni M, Mäkelä R, et al. Ambulant treatment of alcohol withdrawal symptoms with carbamazepine: a formal multicentre, double-blind comparison with placebo. Acta Psychiatrica Scandinavica. 1976;53:333-342.

20. Sonck T, Malinen L, Hanne J. Carbamazepine in the treatment of acute withdrawal syndrome in alcoholics: methodological aspects. In: Rationality of Drug Development. Excerpta Medica International Congress Series No. 383. Amsterdam: Excerpta Medica. 1976:251.

21. Agricola R, Mazzarino M, Urani R. Treatment of acute alcohol withdrawal syndrome with carbamazepine: a double-blind comparison with tiapride. J Int Med Res. 1982;10:160-165.

22. Malcolm R, Ballenger JC, Sturgis ET, et al. Double-blind controlled trial comparing carbamazepine to oxazepam treatment of alcohol withdrawal. Am J Psychiatry. 1989:146;617-621.

23. Stuppaeck CH, Pycha R, Miller C, Whitworth AB, Oberbauer H, Fleischhacker WW. Carbamazepine versus oxazepam in the treatment of alcohol withdrawal: a double blind study. Alcohol and Alcohol. 1991;27:153-158.

24. Letterman L, Markowitz JS. Gabapentin: a review of published experience in the treatment of bipolar disorder and other psychiatric conditions. Pharmacotherapy. 1999;5:565-572.

25. Watson WP, Robinson E, Little HJ. The novel anticonvulsant, gabapentin, protects against both convulsant and anxiogenic aspects of the ethanol withdrawal syndrome. Neuropharmacology. 1997;10:1369-1375.

26. Bailey CP, Molleman A, Little HJ. Comparison of the effects of drugs on hyperexcitability induced in hippocampal slices by withdrawal from chronic ethanol consumption. Br J Pharmacol. 1998;2:215-222.

27. Bonnet U, Banger M, Leweke FM, Maschke M, Kowalski T, Gastpar M. Treatment of alcohol withdrawal syndrome with gabapentin. Pharmacopsychiatry. 1999;3:107-109.

28. Stuppaeck CH, Deisenhammer EA, Kurz M, Whitworth AB, Hinterhuber H. The irreversible gamma-aminobutyrate transaminase inhibitor vigabatrin in the treatment of the alcohol withdrawal syndrome. Alcohol Alcohol. 1996;1:109-111.

29. Rustembegovic A, Sofic E, Kroyer G. A pilot study of Topiramate (Topamax) in the treatment of tonic-clonic seizures of alcohol withdrawal syndromes. Med Arh. 2002;4:211-212.

30. Smith DE, Wesson DR. Benzodiazepines and other sedative-hypnotics. In: Galanter M, Kleber H, eds. American Psychiatric Press Textbook of Substance Abuse Treatment. Washington, DC: American Psychiatric Press; 1995.

31. Ries RK, Roy-Byrne PP, Ward NG, Neppe V, Cullison S. Carbamazepine treatment for benzodiazepine withdrawal. Am J Psychiatry. 1989;4:536-537.

32. Schweizer E, Rickels K, Case WG, Greenblatt DJ. Carbamazepine treatment in patients discontinuing long-term benzodiazepine therapy. Effects on withdrawal severity and outcome. Arch Gen Psychiatry. 1991;5:448-452.

33. Klein E, Colin V, Stolk J, Lenox RH. Alprazolam withdrawal in patients with panic disorder and generalized anxiety disorder: vulnerability and effect of carbamazepine. Am J Psychiatry. 1994;12:1760-1766.

34. Apelt S, Emrich HM. Sodium valproate in benzodiazepine withdrawal. Am J Psychiatry. 1990;7:950-951.

35. Roy-Byrne PP, Lewis N, Villacres E, et al. Preliminary evidence of benzodiazepine subsensitivity in panic disorder. Biol Psychiatry. 1989;7:744-748.

36. McElroy SL, Keck PE, Jr., Lawrence JM. Treatment of panic disorder and benzodiazepine withdrawal with valproate. J Neuropsychiatry Clin Neurosci. 1991;2:232-233.

37. Rickels K, Schweizer E, Garcia EF, Case G, DeMartinis N, Greenblatt D. Trazodone and valproate in patients discontinuing long-term benzodiazepine therapy: effects on withdrawal symptoms and taper outcome. Psychopharmacology (Berl). 1999;1:1-5.

38. Krenz SL, Zimmerman G, Pin M, et al. Topiramate in multidrug withdrawal – comparison with clonidine. Eur Psychiatry. 2002;17(suppl 1):213-214.

39. Zullino DF, Cottier AC, Besson J. Topiramate in opiate withdrawal. Prog Neuropsychopharmacol Biol Psychiatry. 2002;6:1221-1223.

40. Satel SL, Kosten TR, Schuckit MA, Fischman MW. Should protracted withdrawal from drugs be included in DSM-IV? Am J Psychiatry. 1993;5:695-704.

41. Moeller FG, Barratt ES, Dougherty DM, Schmitz JM, Swann AC. Psychiatric aspects of impulsivity. Am J Psychiatry. 2001;11:1783-1793.

42. Keck PE, Jr, McElroy SL, Friedman LM. Valproate and carbamazepine in the treatment of panic and posttraumatic stress disorders, withdrawal states, and behavioral dyscontrol syndromes. J Clin Psychopharmacol. 1992;1:(suppl):36S-41S.

43. Lindenmayer JP, Kotsaftis A. Use of sodium valproate in violent and aggressive behaviors: a critical review. J Clin Psychiatry. 2000;2:123-128.

44. Henderson MJ, Galen LW, DeLuca JW. Temperament style and substance abuse characteristics. Subst Abuse. 1998;2:61-70.

45. Anton RF. Pharmacologic approaches to the management of alcoholism. J Clin Psychiatry. 2001;62(suppl 20):11-17.

46. Mueller TI, Stout RL, Rudden S, et al. A double-blind, placebo-controlled pilot study of carbamazepine for the treatment of alcohol dependence. Alcohol Clin Exp Res. 1997;1:86-92.

47. Brady KT, Myrick H, Henderson S, Coffey SF. The use of divalproex in alcohol relapse prevention: a pilot study. Drug Alcohol Depend. 2002;3:323-330.

48. Longo LP, Campbell T, Hubatch S. Divalproex sodium (Depakote) for alcohol withdrawal and relapse prevention. J Addict Dis. 2002;2:55-64.

49. Karam-Hage M, Brower KJ. Gabapentin treatment for insomnia associated with alcohol dependence. Am J Psychiatry. 2000;1:151.

50. Rubio G, Ponce G, Jimenez-Arriero MA, Palomo T, Ferre F. Effectiveness of topiramate in control of alcohol craving [abstract]. Presented at: the European Congress of Neuropsychopharmacology Regional Meeting; April 19, 2002; Bucharest, Romania.

51. Komanduri R. Two cases of alcohol craving curbed by topiramate. J Clin Psychiatry. 2003;5:612.

52. Johnson BA, Ait-Daoud N, Bowden CL, et al. Oral topiramate for treatment of alcohol dependence: a randomised controlled trial. Lancet. 2003;361:1677-1685.

53. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC:?American Psychiatric Association; 1994.

54. Gibbs JW, III, Sombati S, DeLorenzo RJ, Coulter DA. Cellular actions of topiramate: blockade of kainate-evoked inward currents in cultured hippocampal neurons. Epilepsia. 2000;41(suppl 1):S10-S16.

55. Petroff OA, Hyder F, Rothman DL, Mattson RH. Topiramate rapidly raises brain GABA in epilepsy patients. Epilepsia. 2001;4:543-548.

56. Lima AR, Lima MS, Soares BG, Farrell M. Carbamazepine for cocaine dependence. Cochrane Database Syst Rev. 2001;4:CD002023.

57. Brady KT, Sonne SC, Malcolm RJ, et al. Carbamazepine in the treatment of cocaine dependence: subtyping by affective disorder. Exp Clin Psychopharmacol. 2002;3:276-285.

58. Halikas JA, Center BA, Pearson VL, Carlson GA, Crea F. A pilot, open clinical study of depakote in the treatment of cocaine abuse. Hum Psychopharmacol. 2001;3:257-264.

59. Myrick H, Henderson S, Brady KT, Malcolm R, Measom M. Divalproex loading in the treatment of cocaine dependence. J Psychoactive Drugs. 2001;3:283-287.

60. Myrick H, Henderson S, Brady KT, Malcolm R. Gabapentin in the treatment of cocaine dependence: a case series. J Clin Psychiatry. 2001;1:19-23.

61. Winhusen T, Somoza E. The HPA axis in cocaine use: implications for pharmacotherapy. J Addict Dis. 2001;3:105-119.



Print Friendly 

Jonathan E. Shaywitz MD, and Michael R. Liebowitz, MD
Primary Psychiatry. 2003;10(10):51-56

Dr. Shaywitz is a research fellow in the Anxiety Disorder Clinic at the New York State Psychiatric Institute and instructor in psychiatry at Columbia University, both in New York City.

Dr. Liebowitz is director of the Anxiety Disorder Clinic at the New York State Psychiatric Institute and professor of clinical psychiatry at Columbia University.
Disclosure: Dr. Liebowitz is a consultant for and on the speaker’s bureau of Pfizer, GlaxoSmithKline, and Wyeth; he receives research support and/or honorarium from Pfizer, GlaxoSmithKline, Novartis, Ortho-McNeil, and Wyeth.

Please direct all correspondence to: Michael R. Liebowitz, MD, New York State Psychiatric Institute, Unit 54, 1051 Riverside Dr, New York, NY 10032; Tel: 212-543-5366; Fax: 212-543-6915; E-mail:

Focus Points

Anxiety disorders are common, with an estimated prevalence ranging from 2.5% for obsessive-compulsive disorder to 13% for social anxiety disorder.

Multiple neurotransmitters are involved in anxiety disorders, and the fear neuro-network centered in the amygdala is involved in at least some of the anxiety disorders.
Antiepileptics interact with neurotransmitter systems involved in anxiety disorders and several have demonstrated efficacy in their treatment.


Anxiety disorders are common, with an estimated lifetime prevalence range of 2.5% for obsessive-compulsive disorder to 13% for social anxiety disorders. Since the recognition of these disorders, clinicians have been searching for safe and effective anxiolytic drugs. Benzodiazepines are markedly effective in many patients with generalized anxiety disorder but are associated with withdrawal symptoms upon discontinuation. The selective serotonin reuptake inhibitors have demonstrated efficacy in the treatment of anxiety disorders but only approximately 50% of patients benefit. Many antiepileptics interact with neuropathways thought to be involved with the pathobiology of anxiety disorders, and the results of clinical trials of these agents suggest that they may have utility in the treatment of some anxiety disorders.


Anxiety has been recognized in medicine since the 19th century, but has only recently been officially recognized and classified as a mental disorder.1 Researcher and clinicians have since been searching for safe and effective anxiolytic drugs, beginning with barbiturates in 1903, meprobamate in the 1950s, and benzodiazepines during the later parts of the 20th century.2 The search for effective pharmacotherapy continued with emphasis on finding drugs that were free of addiction or dependence issues. The advent of the use of tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs) was originally thought to address these concerns, but both classes have slow onset of action and dose-limiting toxicity as well as association with discontinuation syndromes.3 Buspirone seemed to solve the dependence and toxicity problems, but has not lived up to its promise in anxiolytic efficacy.4 Other agents used for the pharmacotherapy of anxiety disorders include benzodiazepines, the selective serotonin reuptake inhibitors (SSRIs) and serotonin norepinephrine reuptake inhibitors (SNRIs).5

The understanding of the neurobiology of anxiety has progressed rapidly in recent years from basic knowledge that g-aminobutyric acid (GABA) and serotonin neurotransmitter systems were involved in the expression of anxiety to more fully understanding the complex interaction of multiple neurotransmitters, the fear neuro-network centered in the amygdala, and the genetic predisposition. The amygdala is critical to fear responses and projects to multiple brain systems involved in the physiological and behavioral responses to fear. Its involvement in anxiety is suggested by imaging studies which have shown increased amygdala activation in anxious subjects compared to healthy controls in some anxiety disorders.6

The array of neurotransmitter systems associated with anxiety disorders is illustrated by animal studies, human challenge studies, pharmacologic treatment studies, and imaging investigations.

Despite progress in the understanding of the neurobiology and the pharmacotherapy of anxiety disorders, the response rate to pharmacotherapy remains suboptimal, and the search for novel treatments continues. Antiepileptics (AEDs) are a class of agents with an emerging role in the treatment of anxiety disorders. In addition to their known AED activity, the mechanism of action of several of these drugs suggests that they may be effective in the treatment of anxiety disorders. In the remainder of this review the mechanism of action of the AEDs will be described as will the evidence for their use in the treatment of generalized anxiety disorder (GAD), social anxiety disorder (SAD), and obsessive-compulsive disorder (OCD).

Neurobiology of Anxiety Disorders

The anatomical center for anxiety appears to be the hippocampus and amygdala, which in turn activate the hypothalamic-pituitary-adrenocortical (HPA) axis.7,8 Sensory input is sent through the thalamus to the amygdala, which is known to be critical for the evaluation of stress and fear and involved in some of the anxiety disorders, and to the hippocampus, which is thought to be important in processing emotional memory. When a fearful stimulus is perceived by the amygdala, neural projections to multiple brain systems coordinate the appropriate response, for example, activation of the sympathetic nervous system and the release of stress hormones. Pathological anxiety is different from fear in that the provocative stimulus is not threatening, but the response is just as real as if it were.

The coordination of the neural response to stress or fear depends on carefully orchestrated neurotransmission, and perturbations in any part of the modulation or feedback control can lead to a dysfunctional response. Changes in activity of many neurotransmitters have been reported in anxiety, for example, GABA, glutamine, norepinephrine, serotonin, corticotropin-releasing hormone, and cholecystokinin.7,8 Many of the AED agents are known to affect GABA and glutamate, providing a rationale for investigating their role in the treatment of anxiety.


Research into AED mechanisms has traditionally focused on the ability to inactivate voltage-gated sodium channels, to suppress T-type calcium channels, or to inhibit GABAA receptor-mediated chloride flux. As summarized in Table 1, most AEDs that have been tested in the treatment of anxiety disorders act via one or more of these mechanisms, although how these agents inhibit seizures is not fully understood. For example, the AED mechanism of valproate is unknown, but it enhances brain GABA activity and possibly inhibits glutamate receptors—actions that may contribute to the drug’s efficacy.9 Topiramate enhances neurotransmitter action at GABAA receptors and inhibits glutamate via α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid/kainate receptors. It also inhibits voltage-gated sodium and L-type calcium channels.10 It is the activities that influence brain GABA levels and glutamate activity that may be of interest in the treatment of anxiety disorders. The results of clinical studies in which AEDs have been used to treat anxiety disorders are summarized in Table 211-22 and discussed below.

Generalized Anxiety Disorder

GAD is characterized by a period of excessive anxiety or worry occurring over a period of at least 6 months and is often accompanied by multiple associated symptoms: motor tension, fatigability, poor concentration, autonomic hyperactivity, and hyperarousal.23 In the GAD patient, the degree of anxiety or worry is out of proportion to the likelihood or severity of impact of the feared event and cannot be attributable to any other more focal distress. Thus, GAD is more of an exclusionary disorder defined by what it is not: panic disorder, SAD, OCD, or posttraumatic stress disorder. GAD is common, with an estimated 12-month prevalence of 3.1% and a lifetime prevalence of 5.1%, with women diagnosed 60% more frequently than men.24 GAD is a chronic illness, but, despite its high prevalence, little is understood about the course of the disease. GAD often occurs initially during adolescence and the severity of its symptoms can cycle over the lifetime.

GAD is generally treated with a combination of psychotherapy and pharmacotherapy.25 The medications that are typically used to treat GAD are benzodiazepines, antidepressants, and buspirone. Four benzodiazepine drugs are commonly used to treat GAD: diazepam, lorazepam, clonazepam, and alprazolam.26 They are relatively safe and have fast onset anxiolytic and sedative activity. These compounds exert their effect by enhancing the activity of GABA, resulting in a reduced neuronal firing rate in the locus ceruleus. GABA is the most prevalent and important inhibitory neurotransmitter in the human body. Furthermore, the benzodiazepines, some of the most effective pharmacologic treatments for anxiety disorders, exert their effects via the GABAA receptor.27 Benzodiazepines are reported to be effective in approximately 35% of patients,28,29 and adverse effects associated with their use (sedation, psychomotor impairment) are generally mild. Withdrawal symptoms are sometimes seen upon discontinuation of therapy and are more common with the faster-acting compounds. Because of the potential for dependence, the benzodiazepines are commonly used in combination with antidepressant drugs, and tapered as the effect of the antidepressant medication begins to emerge.

Depression is a common comorbidity of GAD and antidepressants are reported to be effective in its treatment. However, the beneficial effect of the SSRIs and SNRIs in treatment of anxiety occurs independently of any antidepressant effect. The SSRIs, like fluvoxamine and clomipramine, act by inhibiting the neuronal reuptake of serotonin and thus potentiate the activity of serotonergic pathways. Based on their favorable adverse-effect profile, SSRIs are a preferred class of antidepressant drugs used in the treatment of GAD.30

Buspirone was developed as an anxiolytic drug, but unlike the benzodiazepines, has no abuse potential and is not habit forming. Because of its prolonged onset of anxiolytic activity (2–3 weeks), it cannot be used on an as needed basis. The results of recent studies suggest that buspirone is of limited efficacy in the treatment of some anxiety disorders; it is often used in combination with SSRIs.31

Antiepileptic Treatment of Generalized Anxiety Disorder


The results of three studies of the effects of pregabalin in the treatment of GAD have been published. Pregabalin is a lipophilic GABA analogue that was originally developed as an AED.32 It is about 3–10 times more potent than gabapentin in AED activity, and it has recently been shown to be effective against neuropathic pain and GAD. Rickels and colleagues19 studied 455 patients with GAD in a randomized, double-blind trial of three doses of pregabalin compared to alprazolam and placebo. Anxiolytic efficacy was evaluated with the Hamilton Rating Scale for Anxiety (HAM-A). Pregabalin at 300, 450, or 600 mg/day and alprazolam 1.5 mg/day produced significantly greater reductions in HAM-A scores than placebo. The onset of the antianxiety effect of pregabalin was fast, with improvement seen within 1 week of starting treatment. Pregabalin and the SNRI venlafaxine were compared to placebo in a randomized, double-blind study of 426 patients.20 Pregabalin 400 or 600 mg/day or venlafaxine 75 mg/day were significantly superior to placebo in reducing symptoms of GAD as measured by the HAM-A. As was true in the Rickels study, no dose response to pregabalin was seen in the dose range tested.

Pande and colleagues21 have recently reported the results of a comparison of pregabalin 150 or 600 mg/day to lorazepam 6 mg/day or placebo in 276 patients with GAD. Both doses of pregabalin and lorazepam resulted in significantly greater reductions in HAM-A scores compared to placebo over the 4-week study. The most common adverse events for both active treatments were somnolence and dizziness. No withdrawal syndrome was seen when pregabalin was tapered over 1 week at the end of the study. Other adverse effects associated with pregabalin include ataxia and headache, which are mild-to-moderate in severity, and transient.32 The results of these studies suggest that pregabalin is an effective treatment for GAD.


A single case report describes the use of levetiracetam added to a regimen of the SSRI citalopram, in a 42-year-old female suffering from GAD.33 Levetiracetam 250 mg/day reduced anxiety within 4 or 5 days, and the beneficial effect improved and persisted for over 6 months.


Tiagabine, an inhibitor of GABA reuptake, was used to treat 18 patients with GAD in an open-label study (10 mg/day). The patients showed a significant improvement of anxiety symptoms as indicated by decreased HAM-A scores.34 In a similar study, tiagabine was evaluated for use in 10 case patients refractory to conventional antianxiety medications.35 All patients were described as “much” or “very much” improved after
4 weeks of treatment.

Social Anxiety Disorder

SAD, also known as social phobia, is a common anxiety disorder in which subjects are unusually fearful of social interactions and are particularly concerned that their actions might cause embarrassment or humiliation.36 The fear can be overwhelming and interfere with school, work, and other ordinary activities. SAD is a surprisingly common anxiety disorder. Using the Diagnostic and Statistical Manual of Mental Disorders, Third Edition-Revised (DSM-III-R)37 criteria, the 1-year prevalence was estimated to be about 8% and the lifetime prevalence about 13%.24 The onset of SAD is usually in childhood or young adulthood, and it occurs about twice as frequently in women as in men.38

Typical symptoms of SAD include increased sweating, tremor, blushing, dry mouth, hypertension, and tachycardia.36 The causes of SAD are unknown, but there is strong evidence of a biological component, such as heritability, abnormalities in neurotransmitter receptor density, and efficacy of medications used to treat SAD.36 Other anxiety disorders often comorbid with SAD include depression and alcohol abuse.36,39

Treatment of SAD involves both psychotherapy and pharmacotherapy, but the rate of improvement is reported to be slower than with some other psychiatric disorders, such as major depressive disorder (MDD).36 SSRIs have become widely used for SAD,40,41 but only about 50% of SAD patients show improvement with any given SSRI, and high doses are required for extended periods of time.36 Benzodiazepines, MAOIs, SSRIs, and SNRIs have all been reported to be effective in the treatment of SAD.36,39

Antiepileptic Treatment of Social Anxiety Disorder


Topiramate has been tested in the treatment of SAD in a single study. Van Ameringen and colleagues17 evaluated 16 adult outpatients with generalized SAD in an open-label study. The dose of topiramate was titrated from 25 mg/day to 400 mg/day over the first 9 weeks of the 16-week study. Efficacy was measured by changes from baseline in the Liebowitz Social Anxiety Scale (LSAS) score and by the Clinical Global Impression-Improvement Scale (CGI-I). Eleven patients completed the study and seven were considered responders based on CGI-I scores of 1 or 2. In the intent-to-treat analysis, mean CGI-Severity changed from 5.1±0.6 (“markedly ill”) to 3.8±1.4 (“mildly or moderately ill,”
P=.001) after 16 weeks. LSAS score decreased by 29% (P=.013). Four patients withdrew from the study because of adverse events and one withdrew because of lack of efficacy. The most common adverse effects were weight loss, paresthesia, and headache. The authors concluded that results of the study showed the potential for using topiramate as a first-line treatment of SAD.


In a randomized, double-blind study, 135 patients with generalized SAD were randomized to treatment with pregabalin 150 mg/day or 600 mg/day or placebo.22 The primary efficacy measurement was change in LSAS from baseline to the end of study (11 weeks). Patients treated with the high-dose pregabalin had change in LSAS of 28.6±3.2 compared to that seen in the placebo-treated group of 18.4±3.2 (
P<.024). The change in the low-dose pregabalin group was 21.5±3.4 (P>.05 compared to placebo). Based on the CGI-I scale, 43% (20/47) of the high-dose pregabalin group were responders, compared to 22% (10/46) in the placebo group (P=.03, the low-dose pregabalin responder rate was not reported). The magnitude of the response of those treated with high-dose pregabalin was similar to that reported for effective pharmacotherapy in other large clinical studies. A post-hoc analysis indicated that pregabalin was most effective in those patients without comorbid psychiatric disorders. Somnolence was the most common adverse effect and was reported by 43% of the high-dose group, 10% of the low-dose group, and 9% of the placebo group.


Gabapentin has been reported to be effective in the treatment of SAD in a randomized, double-blind study of 69 patients.14 Patients were dosed flexibly with gabapentin (900–3,300 mg/day) or placebo for 14 weeks and were evaluated with both clinician- and patient-rated scales. A significant reduction in the symptoms of SAD was seen in those treated with gabapentin compared to those treated with placebo, with a reduction of total score on the LSAS being -27.3 for the gabapentin-treated group and -11.9 for the placebo group. The low responder rates (gabapentin, 32%; placebo, 19%) compared with those seen in positive studies of other drugs remain unexplained.14 Only dry mouth and dizziness occurred significantly more frequently in the gabapentin group than in the placebo group. The authors concluded that gabapentin has a favorable risk-benefit profile for the treatment of SAD.


Seventeen patients were enrolled in an open-label study of valproate in the treatment of SAD.18 After 12 weeks of treatment at doses between 500 and 2,500 mg/day, both LSAS and CGI-I scores were significantly improved compared to baseline. The response rate was 41% of the intent-to-treat population and 47% of study completers.

Obsessive-Compulsive Disorder

OCD is characterized by unwanted and persistent thoughts or impulses in the mind of the patient (obsessions). The thoughts are unpleasant and cause a high degree of anxiety. The disorder often involves rituals such as repetitive hand washing or checking things (compulsions). Most adults recognize at some point that their obsessions and compulsions are excessive and unreasonable, but are unable to control them. Historically, OCD was thought to be quite rare, but recent studies have estimated the lifetime prevalence to be approximately 2.5%.42 The severity of the disorder may fluctuate, but OCD is a chronic and potentially debilitating illness. Comorbid conditions (eg, depression, other anxiety disorders, eating disorders) are often seen with OCD and can make the diagnosis difficult.

The neurobiology of OCD is complex. The serotonergic neurotransmitter system was first implicated when it was shown that clomipramine, a serotonergic tricyclic compound, reduced the symptoms of OCD.43 Other tricyclic antidepressants with less potent effects on serotonin reuptake, such as amitriptyline, were without effect.43 Support for the role of serotonin in OCD is shown by the fact that the SSRIs all have documented efficacy in the treatment of OCD. However, only 50% to 60% of OCD patients respond to SSRIs, suggesting that other neurotransmitters are involved. Dopamine has been implicated because of the observation that stereotyped movements in animals similar to compulsive behavior in humans are seen after high doses of dopaminergic drugs.44 Similarly, neurological disorders that have a dopamine dysfunction in the basal ganglia, like Tourette’s syndrome, are commonly comorbid with OCD.44 The combined role for dopamine and serotonin in OCD is not surprising based on the extensive interaction of the two neurotransmitters.45 This complexity continues on the neuroanatomical level where the orbitofrontal-limbic-basal ganglia areas are implicated in OCD. PET studies show an increased metabolic rate in the orbitofrontal gyri and caudate nucleus,46 and magnetic resonance imaging studies have shown reduced volumes of orbitofrontal gyri, amygdala, and basal ganglia in adults,47 and an increased thalamic size in children with OCD.47-49 Following successful treatment with SSRIs or behavioral therapy, structural changes in the caudate, orbitofrontal gyri, and cingulate cortex have been described.50-52

Clomipramine, a TCA that is a potent serotonin reuptake inhibitor, is effective in about 40% to 60% of OCD patients.53 Although it appears to be as effective as the SSRIs,54 clomipramine is typically reserved for use in those who fail to respond to SSRIs, because of its side-effect profile. Whereas all SSRIs are demonstrably better than placebo,55 only about 50% of OCD patients respond to them, and adjunct therapy is often added. Commonly added drugs include the atypical antipsychotics or clomipramine.56-58 Despite initial success of pharmacotherapy, there is substantial relapse following drug discontinuation.59,60

Antiepileptic Treatment of Obsessive-Compulsive Disorder

Although OCD and epilepsy have similarities (eg, involuntary forced thinking in epilepsy and obsessional thought in OCD,61 similarities of temporal lobe electroencephalogram [EEG] in epileptic and OCD patients62-64), the early use of AEDs in the treatment of OCD yielded meager results. Pacella and colleagues64 described treating four OCD patients with diphenylhydantoin without response.


Carbamazepine was given to four OCD patients with temporal EEG abnormalities, and improvement was seen in only one.11 In an open-label trial of carbamazepine (400–1,600 mg/day) involving nine OCD patients, only one of the eight patients who completed the 8-week study showed a substantial reduction in obsessions and compulsions.12 No change in mean behavioral or mood scores was seen. Khanna13 treated seven OCD patients with abnormal EEG with carbamazepine (600–1,000 mg/day) and followed them for 12 weeks.2 No changes in mean mood or behavior scores was found, but two patients reported a >50% reduction in their clinical symptoms. Other case reports of the use of carbamazepine have been reported with similar disparate results. The fact that some individuals have responded positively to carbamazepine has led several investigators to suggest that a small subset of patients may benefit from this treatment.12,13


The effects of gabapentin in OCD patients who were partial responders to the SSRI fluoxetine were reported by Corà-Locatelli and colleagues.15 In this 6-week trial, five OCD patients who were currently treated with fluoxetine 30–100 mg/day were given gabapentin, which was started at 900 mg/day (in three doses) and titrated to a maximum of 3,600 mg/day. All patients reported improvement in mood, OCD symptoms, anxiety, and sleep, within 2 weeks of starting treatment. The results of this preliminary study were used as the basis of the design of a double-blind study that is in progress.


Lamotrigine was tested in an open-label study in eight OCD patients who had responded inadequately to SSRI therapy.16 Sertraline (
200 mg/day) and clomipramine (225 mg/day) had been given for at least 14 weeks and were continued during the study. Lamotrigine was added to therapy at a dose up to 100 mg/day. After a mean treatment period of 47 days, only one patient reported improvement, and no significant changes from baseline were found for the mean Yale Brown Obsessive-Compulsive Scale, CGI-S, or CGI-I scores. However, these results should be interpreted with caution as drugs used in OCD tend to be in the higher dose range, and a more typical dose for lamotrigine would be 150–400 mg/day.


The pharmacotherapy of anxiety disorders has centered around benzodiazepines, buspirone, and antidepressants of the SSRI and SNRI classes. These agents are effective for many patients, but dependence issues limit the use of benzodiazepines, the prolonged onset of action of buspirone prevents “as needed” use, and SSRIs and SNRIs are not effective in all patients. Many AEDs interact with neuropathways thought to be involved in the pathobiology of anxiety disorders, and the results of clinical studies suggest that some of these agents may have utility in the treatment of anxiety disorders.

Pregabalin, an AED chemically related to gabapentin, has been tested in several large, randomized, double-blind studies and has shown efficacy in the treatment of GAD and SAD compared to placebo. Topiramate and valproate have been reported to improve symptoms of SAD in open-label studies, and gabapentin was effective in a double-blind study of SAD patients. Carbamazepine, gabapentin, and lamotrigine have been tested in small studies for the treatment of OCD with disparate results. AEDs, with their neurobiological specificity, clinical efficacy, and tolerability, represent an exciting new option in the treatment of anxiety disorders. Additional large, randomized studies will be needed to further explore the scope of their efficacy. PP


1. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, DC: American Psychiatric Association; 1980.

2. Nemeroff CB. Anxiolytics: past, present, and future agents. J Clin Psychiatry. 2003;64:3-6.
3. Haddad PM. Antidepressant discontinuation syndromes. Drug Saf. 2001;24:183-197.

4. Davidson JR, DuPont RL, Hedges D, Haskins JT. Efficacy, safety, and tolerability of venlafaxine extended release and buspirone in outpatients with generalized anxiety disorder. J Clin Psychiatry. 1999;60:528-535.

5. Gorman JM. Treating generalized anxiety disorder. J Clin Psychiatry. 2003;64(suppl 2):24-29.

6. De Bellis MD, Casey BJ, Dahl RE, et al. A pilot study of amygdala volumes in pediatric generalized anxiety disorder. Biol Psychiatry. 2000;48:51-57.

7. Sullivan GM, Coplan JD, Gorman JM. Psychoneuroendocrinology of anxiety disorders. Psychiatr Clin North Am. 1998;21:397-412.

8. Jetty PV, Charney DS, Goddard AW. Neurobiology of generalized anxiety disorder. Psychiatr Clin North Am. 2001;24:75-97.

9. Loscher W. Basic pharmacology of valproate: a review after 35 years of clinical use for the treatment of epilepsy. CNS Drugs. 2002;16:669-694.

10. Shank RP, Gardocki JF, Streeter AJ, Maryanoff BE. An overview of the preclinical aspects of topiramate: pharmacology, pharmacokinetics, and mechanism of action. Epilepsia. 2000;41(suppl):S3-S9.

11. Jenike MA, Brotman AW. The EEG in obsessive-compulsive disorder. J Clin Psychiatry. 1984;45:122-124.

12. Joffe RT, Swinson RP. Carbamazepine in obsessive-compulsive disorder. Biol Psychiatry. 1987;22:1169-1171.

13. Khanna S. Carbamazepine in obsessive-compulsive disorder. Clin Neuropharmacol. 1988;11:478-481.

14. 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.

15. Corà-Locatelli G, Greenberg BD, Martin J, Murphy DL. Gabapentin augmentation for fluoxetine-treated patients with obsessive-compulsive disorder. J Clin Psychiatry. 1998;59:480-481.

16. Kumar TC, Khanna S. Lamotrigine augmentation of serotonin re-uptake inhibitors in obsessive-compulsive disorder. Aust N Z J Psychiatry. 2000;34:527-528.

17. Van Ameringen M, Mancini C, Pipe B. An open trial of topiramate in the treatment of generalized social phobia. Poster presented at: the 52nd Annual Meeting of the Canadian Psychiatric Association; Banff. November 2002; Alberta, Canada.
18. Kinrys G, Pollack MH, Simon NM, Worthington JJ, Nardi AE, Versiani M. Valproic acid for the treatment of social anxiety disorder. Int Clin Psychopharmacol. 2003;18:169-172.

19. Rickels K, Pollack MH, Lydiard RB, et al. Efficacy and safety of pregabalin and alprazolam in generalized anxiety disorder. Presented at: 43rd Annual New Clinical Drug Evaluation Unit Meeting; June2002; Boca Raton, FL.

20. Kasper S, Blagden M, Seghers S, et al. A placebo-controlled study of pregabalin and venlafaxine treatment of GAD. Presented at: the 43rd Annual New Clinical Drug Evaluation Unit Meeting; June 2002; Boca Raton, FL.

21. Pande AC, Crockatt JG, Feltner DE, et al. Pregabalin in generalized anxiety disorder: a placebo-controlled trial. Am J Psychiatry. 2003;160:533-540.

22. Feltner DE, Pollack MH, Davidson JRT, et al. A placebo-controlled, double-blind study of pregabalin treatment of social phobia: outcome and predictors of response. J. Eur. Coll. Neuropsychopharmacol. 2000;10(suppl 3):S345.

23. Diagnostic and Statistical Manual of Mental Disorders. 4th ed text-rev. Washington, DC: American Psychiatric Association; 2000.

24. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry. 1994;51:8-19.

25. Taylor CB. Treatment of anxiety disorders. In: Schatzberg AF, Nemeroff CB, eds. Textbook of Psychopharmacology. 2nd ed. Washington, DC: American Psychiatric Press Inc.; 1998:775-789.

26. Connor KM, Davidson JR. Generalized anxiety disorder: neurobiological and pharmacotherapeutic perspectives. Biol Psychiatry. 1998;44:1286-1294.
27. Rudolph U, Crestani F, Benke D, et al. Benzodiazepine actions mediated by specific gamma-aminobutyric acid(A) receptor subtypes. Nature. 1999;401:796-800.

28. Dubovsky SL. Generalized anxiety disorder: new concepts and psychopharmacologic therapies. J Clin Psychiatry. 1990;51:3-10.

29. Davidson JR. Pharmacotherapy of generalized anxiety disorder. J Clin Psychiatry. 2001;62:46-50. Discussion 51-42.

30. Kent JM, Coplan JD, Gorman JM. Clinical utility of the selective serotonin reuptake inhibitors in the spectrum of anxiety. Biol Psychiatry. 1998;44:812-824.

31. Lydiard RB, Brawman-Mintzer O, Ballenger JC. Recent developments in the psychopharmacology of anxiety disorders. J Consult Clin Psychol. 1996;64:660-668.

32. Lauria-Horner BA, Pohl RB. Pregabalin: a new anxiolytic. Expert Opin Investig Drugs. 2003;12:663-672.

33. Pollack M. Levetiracetam (Keppra) for anxiety. Curbside Consultant. 2002;1:4.

34. Papp L. Tiagabine treatment of GAD. Presented at: the Annual Meeting of the American Psychiatric Association; May 2003; San Francisco, Calif.

35. Crane D. Tiagabine for treatment of anxiety. Presented at: the Annual Meeting of the American Psychiatric Association; May 2003; San Francisco, Calif.

36. Zamorski MA, Ward RK. Social anxiety disorder: common, disabling, and treatable. J Am Board Fam Pract. 2000;13:251-260.

37. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed rev. Washington, DC: American Psychiatric Association; 1987.

38. Faravelli C, Zucchi T, Viviani B, et al. Epidemiology of social phobia: a clinical approach. Eur Psychiatry. 2000;15:17-24.

39. den Boer JA. Social phobia: epidemiology, recognition, and treatment. BMJ. 1997;315:796-800.

40. Van Ameringen MA, Lane RM, Walker JR, et al. Sertraline treatment of generalized social phobia: a 20-week, double-blind, placebo-controlled study. Am J Psychiatry. 2001;158:275-281.

41. Stein MB, Fyer AJ, Davidson JR, Pollack MH, Wiita B. Fluvoxamine treatment of social phobia (social anxiety disorder): a double-blind, placebo-controlled study. Am J Psychiatry. 1999;156:756-760.

42. Regier DA, Boyd JH, Burke JD Jr, et al. One-month prevalence of mental disorders in the United States. Based on five Epidemiologic Catchment Area sites. Arch Gen Psychiatry. 1988;45:977-986.

43. Ananth J, Pecknold JC, van den Steen N, Engelsmann F. Double-blind comparative study of clomipramine and amitriptyline in obsessive neurosis. Prog Neuropsychopharmacol. 1981;5:257-262.

44. Goodman WK, McDougle CJ, Price LH, Riddle MA, Pauls DL, Leckman JF. Beyond the serotonin hypothesis: a role for dopamine in some forms of obsessive compulsive disorder? J Clin Psychiatry. 1990;51:36-43. Discussion 55-38.

45. Kapur S, Remington G. Serotonin-dopamine interaction and its relevance to schizophrenia. Am J Psychiatry. 1996;153:466-476.

46. Baxter LR Jr, Phelps ME, Mazziotta JC, Guze BH, Schwartz JM, Selin CE. Local cerebral glucose metabolic rates in obsessive-compulsive disorder. A comparison with rates in unipolar depression and in normal controls. Arch Gen Psychiatry. 1987;44:211-218.

47. Szeszko PR, Robinson D, Alvir JM, et al. Orbital frontal and amygdala volume reductions in obsessive-compulsive disorder. Arch Gen Psychiatry. 1999;56:913-919.

48. Rosenberg DR, Benazon NR, Gilbert A, Sullivan A, Moore GJ. Thalamic volume in pediatric obsessive-compulsive disorder patients before and after cognitive behavioral therapy. Biol Psychiatry. 2000;48:294-300.

49. Gilbert AR, Moore GJ, Keshavan MS, et al. Decrease in thalamic volumes of pediatric patients with obsessive-compulsive disorder who are taking paroxetine. Arch Gen Psychiatry. 2000;57:449-456.

50. Baxter LR, Jr., Schwartz JM, Bergman KS, et al. Caudate glucose metabolic rate changes with both drug and behavior therapy for obsessive-compulsive disorder. Arch Gen Psychiatry. 1992;49:681-689.

51. Benkelfat C, Nordahl TE, Semple WE, King AC, Murphy DL, Cohen RM. Local cerebral glucose metabolic rates in obsessive-compulsive disorder. Patients treated with clomipramine. Arch Gen Psychiatry. 1990;47:840-848.

52. Perani D, Colombo C, Bressi S, et al. [18F]FDG PET study in obsessive-compulsive disorder.
A clinical/metabolic correlation study after treatment.
Br J Psychiatry. 1995;166:244-250.

53. Clomipramine in the treatment of patients with obsessive-compulsive disorder. The Clomipramine Collaborative Study Group. Arch Gen Psychiatry. 1991;48:730-738.

54. Greist JH, Jefferson JW, Kobak KA, Katzelnick DJ, Serlin RC. Efficacy and tolerability of serotonin transport inhibitors in obsessive-compulsive disorder. A meta-analysis. Arch Gen Psychiatry. 1995;52:53-60.

55. Zohar J, Chopra M, Sasson Y, Amiaz R, Amital D. Obsessive compulsive disorder: serotonin and beyond. World J Biol Psychiatry. 2000;1:92-100.

56. Pfanner C, Marazziti D, Dell’Osso L, et al. Risperidone augmentation in refractory obsessive-compulsive disorder: an open-label study. Int Clin Psychopharmacol. 2000;15:297-301.

57. McDougle CJ, Epperson CN, Pelton GH, Wasylink S, Price LH. A double-blind, placebo-controlled study of risperidone addition in serotonin reuptake inhibitor-refractory obsessive-compulsive disorder. Arch Gen Psychiatry. 2000;57:794-801.

58. Pallanti S, Quercioli L, Paiva RS, Koran LM. Citalopram for treatment-resistant obsessive-compulsive disorder. Eur Psychiatry. 1999;14:101-106.

59. Eisen JL, Goodman WK, Keller MB, et al. Patterns of remission and relapse in obsessive-compulsive disorder: a 2-year prospective study. J Clin Psychiatry. 1999;60:346-351. Quiz 352.

60. Bergeron R, Ravindran AV, Chaput Y, et al. Sertraline and fluoxetine treatment of obsessive-compulsive disorder: results of a double-blind, 6-month treatment study. J Clin Psychopharmacol. 2002;22:148-154.

61. Brickner RM, Rosen AA, Munro R. Physiological aspects of the obsessive state. Psychosom Med. 1940;2:369-383.

62. Epstein AW, Bailine SH. Sleep and dream studies in obsessional neurosis with particular reference to epileptic states. Biol Psychiatry. 1971;3:149-158.

63. Jenike MA. Obsessive-compulsive disorder: a question of a neurologic lesion. Compr Psychiatry. 1984;25:298-304.

64. Pacella BL, Polatin P, Nagler SH. Clinical and EEG studies in obsessive-compulsive states. p521944;100:830-838.