Borderline personality disorder (BPD) has long defined definitive treatment. Such failure is reflected in repeated suicidal crises, often associated with dysphoric symptoms of a chronic fluctuating nature, whose labile intermittent character does suggest a subthreshold bipolar depressive mixed state. For all these reasons, we hypothesized that the anticonvulsant lamotrigine, touted to be a mood stabilizer with antidepressant properties, might be uniquely beneficial for these patients.
From a base rate of about 300 patients in a community mental health center, we identified eight patients meeting seven or more of the DSM-IV criteria for BPD without concurrent major mood disorders. All patients presented with history of severe suicidal behavior, hostile depression and/or labile moods, stimulant and alcohol abuse, as well as multiple unprotected sexual encounters; one patient was actually HIV positive. All had failed previous trials with different antidepressants and mood stabilizers. All current medications were gradually withdrawn--and when necessary--patients kept on a low dose of a conventional neuroleptics for a few weeks, while lamotrigine was being gradually introduced in 25-mg weekly increments until the patient responded (up to 300 mg/day maximum).
Consistent with previous work by us and others, bipolar family history could be documented in three of eight BPD patients, and worsening on antidepressants in four of eight, providing indirect support to our conceptualization of BPD as a bipolar variant. One patient developed a rash on 25 mg and was dropped from the lamotrigine trial, while another patient was noncompliant. Three who failed lamotrigine, subsequently responded, respectively, to sertraline, lithium-thioridazine combination, and valproate. The remaining three patients showed a robust response to lamotrigine, ranging from 75 to 300 mg/day: their functioning jumped from a mean baseline DSM-IV GAF score in the 40's to the 80's during 3-4 months. Among all responders impulsive sexual, drug-taking and suicidal behaviors disappeared and no longer met the criteria for BPD. At an average follow-up of 1 year, they no longer meet criteria for BPD.
Open uncontrolled results on a small number of patients in a tertiary care center may not generalize to BPD patients at large.
Overall, the BPD response to pharmacotherapy in the present case series was 75%. The fact that five of six pharmacotherapy responders required mood stabilizers, argues against the prevalent view that the depressions of borderline patients belong to unipolarity. Of BPD patients who completed the trial, 50% achieved sustained remission from their personality disorder with lamotrigine monotherapy. The dramatic nature of the response in patients refractory to all previous medication trials and maintenance of a robust response over 1 year, argue against a placebo effect. Controlled systematic investigation of lamotrigine in BPD is indicated.