Deep Brain Stimulation of the Subthalamic Nucleus Modulates Reward-Related Behavior: A Systematic Review

Abstract

Deep brain stimulation of the subthalamic nucleus (STN-DBS) is an effective treatment for the motor symptoms of movement disorders including Parkinson's Disease (PD). Despite its therapeutic benefits, STN-DBS has been associated with adverse effects on mood and cognition. Specifically, apathy, which is defined as a loss of motivation, has been reported to emerge or to worsen following STN-DBS. However, it is often challenging to disentangle the effects of STN-DBS per se from concurrent reduction of dopamine replacement therapy, from underlying PD pathology or from disease progression. To this end, pre-clinical models allow for the dissociation of each of these factors, and to establish neural substrates underlying the emergence of motivational symptoms following STN-DBS. Here, we performed a systematic analysis of rodent studies assessing the effects of STN-DBS on reward seeking, reward motivation and reward consumption across a variety of behavioral paradigms. We find that STN-DBS decreases reward seeking in the majority of experiments, and we outline how design of the behavioral task and DBS parameters can influence experimental outcomes. While an early hypothesis posited that DBS acts as a "functional lesion," an analysis of lesions and inhibition of the STN revealed no consistent pattern on reward-related behavior. Thus, we discuss alternative mechanisms that could contribute to the amotivational effects of STN-DBS. We also argue that optogenetic-assisted circuit dissection could yield important insight into the effects of the STN on motivated behavior in health and disease. Understanding the mechanisms underlying the effects of STN-DBS on motivated behavior-will be critical for optimizing the clinical application of STN-DBS.

Keywords: deep brain stimulation (DBS); motivation; operant; reward; rodent; subthalamic nucleus (STN).

Figure 1

PRISMA flow diagram of the inclusion criteria of studies eligible for systematic review.

Figure 2

Experiments investigating STN-DBS effects on reward-related behavior. Graphical representation of experiments assessing the effects of STN-DBS on reward-related behavior. Experiments are split according the task and the reward provided, and whether the main effect was an increase (green frames), decrease (red frames) or no change (black frames) in motivational state induced by STN-DBS. FR, Fixed ratio; PR, progressive ratio.

Figure 3

Experiments investigating STN lesion effects on reward related behavior. Graphical representation of experiments assessing the effects of STN lesion on reward-related behavior. Experiments are split according the task, the reward provided and whether the main effect was an increase (green frames), decrease (red frames) or no change (black frames) in motivational state following lesion of the STN. FR, Fixed ratio; PR, progressive ratio.

Figure 4

Experiments investigating STN pharmacological inhibition effects on reward related behavior. Graphical representation of experiments assessing the effects of pharmacological inactivation of the STN on reward-related behavior. Experiments are split according the task, the reward provided and whether the main effect was an increase (green frames), decrease (red frames) or no change (black frames) in motivational state following pharmacological inactivation of the STN. FR, Fixed ratio; PR, progressive ratio.

Figure 5

STN-DBS decreases reward related behavior. Forest plot of the Cohen's d standardized mean difference for reward seeking effect of STN-DBS. Mean effect size is depicted by the dashed red line. Key details of the experimental design, and DBS stimulation parameters are summarized in the associated table. All the studies provided water ad libitum. 5-CSRTT, Five choice serial reaction time; CRTT, choice reaction time task; PD, Parkinson's disease; FR, Fixed ratio; PR, progressive ratio; rIGT, rat Iowa Gambling Task.

Figure 6

STN lesion does not consistently affect reward related behavior. Forest plot of the Cohen's d standardized mean difference for reward seeking effect of STN lesion, ranked in order of positive to negative effect. Mean effect size is depicted by the dashed red line. 5-CSRTT, five choice serial reaction time; FR, Fixed ratio; PR, progressive ratio.

Figure 7

Pharmacological inactivation of the STN does not consistently affect reward related behavior. Forest plot of the Cohen's d standardized mean difference for reward seeking effect of STN pharmacological inactivation, ranked in order of positive to negative effect. Mean effect size is depicted by the dashed red line. 5-CSRTT, five choice serial reaction time test; FR, fixed ratio; PR, progressive ratio.

Figure 8

Afferent and efferent connections of STN functional subdivisions. The subthalamic nucleus (STN) is subdivided into a dorsolateral motor territory, a ventromedial associative territory, and a medial limbic territory. Each functional territory receives input from different cortical regions or the external segment of the globus pallidus (GPe), and in turn projects to different downstream structures, including the internal segment of the globus pallidus (GPi), substantia nigra pars reticulata (SNr), nucleus accumbens (NAc) and ventral pallidum (VP). These input-output interactions provide for parallel control of motor, cognitive, and emotional functions. The STN is composed of interneurons and glutamatergic projection neurons whose dendrites may arborize over a distance of up to 500 μm. This is important, because individual STN neurons may physically span into adjacent territories and be effected by DBS applied to these adjacent subdivisions.