Twenty-eight patients were included in the study. These subjects had received fluoxetine for ≥12 weeks, reaching 80 mg/day if tolerated (mean=62.5 mg/day, SD=20.1). Nineteen of these patients received fluoxetine in a separate placebo-controlled study (4
); 16 did not respond to fluoxetine, and the three responders still had severe enough body dysmorphic disorder to participate in this pimozide study. Nine of the patients received the fluoxetine in my practice. An additional 16 patients in the separate placebo-controlled fluoxetine study (13 nonresponders and three responders), plus two nonresponders from my clinical practice, did not enter the pimozide study for different reasons (e.g., lack of interest).
The 28 patients were randomly assigned to 8 weeks of double-blind pimozide (N=11) or placebo (N=17) augmentation while remaining on a fixed fluoxetine dose. Pimozide and placebo were furnished in identical-appearing tablets (2 mg for pimozide). Subjects were started on 1 mg/day, with an attempt made to raise the dose to 2 mg/day after 1 week and then by 2 mg per week to a maximum of 10 mg/day if tolerated. After a complete description of the study, written informed consent was obtained from all participants.
Inclusion/exclusion criteria were standard for efficacy studies (e.g., reference 4
). Following fluoxetine treatment, subjects had a body dysmorphic disorder score of ≥20 on the Yale-Brown Obsessive Compulsive Scale Modified for Body Dysmorphic Disorder (8
) with fair, poor, or absent insight and were at least moderately ill according to the Clinical Global Impression (CGI) scale. Subjects took no other psychotropics. They could not begin psychotherapy during the study or have begun it within the past 4 months. The Yale-Brown Obsessive Compulsive Scale Modified for Body Dysmorphic Disorder score was the primary outcome measure; a decrease in score of ≥30% determined treatment response (8
). The Brown Assessment of Beliefs Scale (9
) assessed delusionality of appearance beliefs and categorized subjects at baseline as delusional (N=12) or nondelusional (N=16) according to an empirically derived cutoff point. Other measures were the CGI, Hamilton Rating Scale for Depression, Brief Psychiatric Rating Scale, and Structured Clinical Interview for DSM-III-R.
Excluding the augmented fluoxetine trial, 18 (64.3%) of the 28 subjects had previously received a total of 58 psychotropic medications. Fifteen subjects received a total of 26 SRIs, and three received a neuroleptic (one trial each). Only two SRI trials improved body dysmorphic disorder, but only five trials were considered minimally adequate for body dysmorphic disorder (2
), one of which led to improvement. Three non-SRI medications (a non-SRI tricyclic, lithium, and a neuroleptic) improved body dysmorphic disorder.
Analyses were based on the intent-to-treat study group and used analysis of covariance (ANCOVA) with baseline measures as the covariate. The effect size (d) was based on ANCOVA. Continuous variables were analyzed with independent-sample t tests, and dichotomous variables were analyzed with chi-square test and Fisher’s exact test. All tests were two-tailed; the alpha level was 0.05.