Background
Data from case series suggest that clozapine may benefit inpatients with borderline personality disorder (BPD), but randomised trials have not been conducted.
Methods
Multicentre, double-blind, placebo-controlled trial. We aimed to recruit 222 inpatients with severe BPD aged 18 or over, who had failed to respond to other antipsychotic medications. We randomly allocated participants on a 1:1 ratio to receive up to 400mg of clozapine per day or an inert placebo using a remote web-based randomisation service. The primary outcome was total score on the Zanarini Rating scale for Borderline Personality Disorder (ZAN-BPD) at six months. Secondary outcomes included self-harm, aggression, resource use and costs, side effects and adverse events. We used a modified intention to treat analysis (mITT) restricted to those who took one or more dose of trial medication, using a general linear model fitted at six months adjusted for baseline score, allocation group and site.
Results
The study closed early due to poor recruitment and the impact of the COVID-19 pandemic. Of 29 study participants, 24 (83%) were followed up at six months, of whom 21 (72%) were included in the mITT analysis. At six months, 11 (73%) participants assigned to clozapine and 6 (43%) of those assigned to placebo were still taking trial medication. Adjusted difference in mean total ZAN-BPD score at six months was -3.86 (95% Confidence Intervals = -10.04 to 2.32, p=0.22). There were 14 serious adverse events; six in the clozapine arm and eight in the placebo arm of the trial. There was little difference in the cost of care between these groups.
Interpretation
We recruited insufficient participants to test the primary hypothesis. The study findings highlight problems in conducting placebo-controlled trials of clozapine and in using clozapine for people with BPD, outside specialist inpatient mental health units.
ISRCTN18352058. https://doi.org/10.1186/ISRCTN18352058