Neurocognitive markers of psychosis in bipolar disorder: a meta-analytic study

Abstract

Background: It has been suggested that patients with psychotic bipolar disorder (BDP+) might have more severe cognitive deficits than non-psychotic bipolar disorder patients (BDP-). However, only a handful of studies have compared cognition between BDP+ and BDP-. Our aim was to examine cognitive deficits associated with psychosis in BD using meta-analytic methods.

Methods: After a systematic literature review, we conducted a meta-analysis on studies that compared cognition in bipolar disorder (BD) patients with and without a history of psychosis. In addition the effects of clinical and demographic confounder factors were examined with meta-regression analyses.

Results: There were no significant differences for gender and duration of illness between groups. Compared with BDP-, BDP+ patients had more inpatient admissions, a younger onset of illness and used antipsychotics more commonly. BDP+ patients also performed significantly worse in 4 of 6 cognitive domains (planning and reasoning, working memory, verbal memory and processing speed). There were also differences for some individual tasks (List Learning, Delayed Recall, Trail Making B, Wisconsin Card Sorting Test, Digits Backwards, Stroop Interference, Semantic Fluency) with BDP+ patients showing moderately greater impairment on these tasks (d=0.30-0.55).

Conclusions: A history of psychosis is associated with greater severity of cognitive deficits in BD. However, this effect is modest, and these findings do not suggest a complete categorical distinction between BDP+ and BDP-. Psychosis in BD might reflect partly distinct neurobiological processes.