A Preliminary Double-Blind, Placebo-Controlled Trial of Divalproex Sodium in Borderline Personality Disorder

Background: Borderline personality disorder ischaracterized by affective instability, impulsivity, andaggression and is associated with considerable morbidity andmortality. Since anticonvulsant agents may be helpful in suchsymptomatology, we compared divalproex sodium with placebo inpatients with borderline personality disorder.

Method: A 10-week, parallel, double-blinddesign was conducted. Sixteen outpatients meeting StructuredClinical Interview for DSM-IV Axis II Personality Disorderscriteria for borderline personality disorder were randomlyassigned to receive placebo (N = 4) or divalproex sodium (N =12). Change was assessed in global symptom severity (ClinicalGlobal Impressions-Improvement Scale [CGI-I]) and functioning(Global Assessment Scale [GAS]) as well as in specific coresymptoms (depression, aggression, irritability, and suicidality).

Results: There was significant improvement frombaseline in both global measures (CGI-I and GAS) followingdivalproex sodium treatment. A high dropout rate precludedfinding significant differences between the treatment groups inthe intent-to-treat analyses, although all results were in thepredicted direction.

Conclusion: Treatment with divalproexsodium may be more effective than placebo for globalsymptomatology, level of functioning, aggression, and depression.Controlled trials with larger sample sizes are warranted toconfirm these preliminary results.