An Open-Label, Randomized, Active Controlled Trial Evaluating Sublingual Dexmedetomidine for Moderate to Severe Agitation in Inpatients With Schizophrenia or Bipolar Disorder.

PURPOSE: Agitation can occur in patients with schizophrenia, schizoaffective disorder, and bipolar disorder. This study evaluated the efficacy of sublingual dexmedetomidine, a selective alpha-2 adrenergic receptor agonist, for the treatment of acute agitation in "real-world" psychiatric inpatients. METHODS: This was a pragmatic, randomized, open-label, active-controlled study, conducted at Temple University Episcopal Hospital in Philadelphia, PA. Participants were inpatients with a diagnosis of schizophrenia, schizoaffective disorder, or bipolar disorder, and were randomized in a 1:1 manner to receive sublingual dexmedetomidine or oral lorazepam. Participants received sublingual dexmedetomidine 180 mcg (n = 7), sublingual dexmedetomidine 120 mcg (n = 5), or oral lorazepam 2 mg (n = 12) during an episode of agitation. The primary efficacy end point was absolute change from baseline in Positive and Negative Syndrome Scale-Excited Component (PANSS-EC) total score at 2 hours after medication administration. There were no restrictions on concurrent substance use disorders, suicidal ideations, or psychiatric comorbidities. RESULTS: Twenty-four participants self-administered study medication. Diagnoses included schizophrenia 8/24 (33%), schizoaffective disorder 5/24 (21%), and bipolar disorder 11/24 (46%). At 2 hours post-dose, the PANSS-EC total change from baseline was [median (IQR)] -13.0 (-17.0, -10.0) for sublingual dexmedetomidine and -14.0 (-19.0, -12.0) for oral lorazepam. There were no spontaneously reported adverse effects in either group. CONCLUSIONS: At 2 hours post-dose, sublingual dexmedetomidine was effective at reducing acute agitation in "real-world" psychiatric inpatients with schizophrenia, schizoaffective disorder, and bipolar disorder, who may have had psychiatric comorbidities. There were no spontaneously reported adverse effects. The study was not powered to detect differences between treatment groups.