A randomized, double-blind, controlled trial evaluating the effect of intranasal insulin on neurocognitive function in euthymic patients with bipolar disorder

Background: Neurocognitive deficits are prevalent, persistent, and implicated as mediators of functional impairment in adults with bipolar disorder. Notwithstanding progress in the development of pharmacological treatments for various phases of bipolar disorder, no available treatment has been proven to be reliably efficacious in treating neurocognitive deficits. Emerging evidence indicates that insulin dysregulation may be pertinent to neurocognitive function. In keeping with this view, we tested the hypothesis that intranasal insulin administration would improve measures of neurocognitive performance in euthymic adults with bipolar disorder. Methods: Sixty-two adults with bipolar I/II disorder (based on the Mini International Neuropsychiatric Interview 5.0) were randomized to adjunctive intranasal insulin 40 IU q.i.d. (n = 34) or placebo (n = 28) for eight weeks. All subjects were prospectively verified to be euthymic on the basis of a total score of <= 3 on the seven-item Hamilton Depression Rating Scale (HAMD-7) and <= 7 on the 11-item Young Mania Rating Scale (YMRS) for a minimum of 28 consecutive days. Neurocognitive function and outcome was assessed with a neurocognitive battery. Results: There were no significant between-group differences in mean age of the subjects {i.e., mean age 40 [standard deviation (SD) = 10.15] years in the insulin and 39 [SD = 10.41] in the placebo groups, respectively}. In the insulin group, n = 27 (79.4%) had bipolar I disorder, while n = 7 (21.6%) had bipolar II disorder. In the placebo group, n = 25 (89.3%) had bipolar I disorder, while n = 3 (10.7%) had bipolar II disorder. All subjects received concomitant medications; medications remained stable during study enrollment. A significant improvement versus placebo was noted with intranasal insulin therapy on executive function (i.e., Trail Making TestPart B). Time effects were significant for most California Verbal Learning Test indices and the Process Dissociation TaskHabit Estimate, suggesting an improved performance from baseline to endpoint with no between-group differences. Intranasal insulin was well tolerated; no subject exhibited hypoglycemia or other safety concerns. Conclusions: Adjunctive intranasal insulin administration significantly improved a single measure of executive function in bipolar disorder. We were unable to detect between-group differences on other neurocognitive measures, with improvement noted in both groups. Subject phenotyping on the basis of pre-existing neurocognitive deficits and/or genotype [e.g., apolipoprotein E (ApoE)] may possibly identify a more responsive subgroup.