Objective: While Second Generation Antipsychotics (SGAs) have been shown, in some studies, to be helpful for amelioration of impulsiveness, aggressiveness and moodiness of patients with Borderline Personality Disorder (BPD), there are additional studies that believe that the evidence does not currently support effectiveness of SGAs for overall severity of borderline symptoms. In light of these conflicting reports, we conducted a trial to compare the efficacy and safety of olanzapine vs. aripiprazole with regard to management of BPD.
Method: Twenty-four female inpatients, meeting the diagnosis of BPD according to the DSM-IV-TR criteria, were randomly entered into an 8-week parallel group, open-label study for random assignment to olanzapine or aripiprazole (n=12 in each group). The patients were selected from outpatients among clientele of two psychiatric clinics and also inpatients from female wards of Razi psychiatric hospital. Patients were excluded from the trial if any prominent co-morbid mental disorder was present. The primary outcome measure included the Brief Psychiatric Rating Scale (BPRS) for estimation of baseline psychopathology, clinical outcome and treatment response. In addition, the Buss-Durkee Hostility Inventory (BDHI) for the assessment of anger and hostility, and the Clinical Global Impressions-Severity Scale (CGI-S) for assessment of overall illness severity were used as secondary scales. Changes in mean total scores of the BPRS, CGI-S and BDHI were analyzed within each group by means of a paired t test and compared across the two groups by an unpaired t test. Statistical significance was defined as a 2-sided p value < or = to 0.05. In addition, analysis of effect size (ES) by calculation of Cohen’s d was performed and recorded.
Results: All of the patients remained in and completed the study. According to the findings olanzapine and aripiprazole reduced mean total score of the BPRS about 25.7% and 18.8% in their groups, and thus illustrated a significant improvement of primary outcome measure at the end of the trial, in comparison with baseline (p<0.01 and p<0.04, respectively). In addition, the mean total score of the BDHI improved around 15% with olanzapine and 12.2% with aripiprazole, which was again significant as regards the first drug (p<0.04), but not so concerning the second one (p<0.06). Significant changes in mean total scores of the CGI-S as well were observed in the olanzapine group with a reduction around 20.9% (p<0.03). Once more, improvement of the same scale by aripiprazole, even though remarkable (19.3%), was not statistically significant (p<0.07). The analysis of specific BPRS subscales in both groups revealed similar, significantly lower scores in anxiety, tension, depressive mood and hostility. In this regard, olanzapine showed appreciably better results on uncooperativeness and excitement, and aripiprazole showed superior efficacy as regards suspiciousness and unusual thought content. The effect size (ES) analysis for changes on the BPRS indicated a large improvement with both olanzapine and aripiprazole. Post-hoc analysis showed an intermediary power=0.46 forof this trial. The mean modal doses of olanzapine and aripiprazole during the trial were 6.4 mg/day (s.d.=2.7) and 7.0 mg/day (s.d.=2.5), respectively.
Conclusion: In spite of somewhat comparable effects, it seems that olanzapine, in comparison with aripiprazole, is more effective on borderline personality symptoms.