Hypoactivation of the ventral and dorsal striatum during reward and loss anticipation in antipsychotic and mood stabilizer-naive bipolar disorder

Abstract

Increased activity within known reward-processing neurocircuitry (eg, ventral striatum, VS) has been reported among medicated individuals with bipolar disorder (BD) I and II. However, such findings are confounded by the potential ameliorative effects of mood-stabilizing and antipsychotic medications on neural activations. This study tests the hypothesis that a pathophysiological locus of alterations in reward processing is present within the striatum in antipsychotic and lithium-naive individuals with BD. Twenty antipsychotic and lithium-naive individuals with BD II or BD not-otherwise specified (NOS) and 20 matched healthy comparison individuals participated in functional magnetic resonance imaging during the performance of a monetary incentive delay task. Between-group comparisons were conducted using small-volume correction focusing on orthogonal a priori regions of interest centered in the VS and dorsal striatum (DS), respectively. During reward anticipation, unmedicated individuals with BD II/NOS had decreased activity within the DS (but not VS). During loss anticipation, on the other hand, decreased activation within both the VS and DS was observed. Across all participants, DS activity (during reward anticipation) was positively associated with putamen volume. This is the first report of decreased dorsal and ventral striatal activity among unmedicated individuals with BD II/NOS. These data contradict a simple 'reward hypersensitivity' model of BD, and add to a growing body of literature suggesting that blunted reward processing may be a vulnerability factor for both mood- and addiction-related disorders.

Figures

Figure 1

Reduced dorsal and ventral striatal activations during gain and loss anticipation (A2 phase of the MID task). Whole-brain, task-related activations (k>50; p<0.001, uncorrected) across all participants are shown in coronal slices. Dorsal and ventral ROIs are shown in red (dorsal) and blue (ventral), and average peak activations are plotted for HC and BD II/NOS groups. Group-level random-effects models were conducted using the small-volume correction (SVC) tool in SPM8, and p-values were thresholded using family-wise error correction (pFWE<0.05). These analyses indicated significantly decreased activity among BD II/NOS participants (vs controls) in the dorsal striatum during win anticipation, and within both the dorsal and ventral striatum during loss anticipation. BD, bipolar II or not-otherwise-specified disorder; HC, healthy control.

Figure 2

Correlation between putamen volumes and BOLD response (shown in u=arbitrary units) within the right dorsal striatum (9 mm ROI) during the anticipation of winning outcomes. BD, bipolar II or not-otherwise-specified disorder; HC, healthy control.