Dopamine-agonists and impulsivity in Parkinson's disease: impulsive choices vs. impulsive actions

Abstract

The control of impulse behavior is a multidimensional concept subdivided into separate subcomponents, which are thought to represent different underlying mechanisms due to either disinhibitory processes or poor decision-making. In patients with Parkinson's disease (PD), dopamine-agonist (DA) therapy has been associated with increased impulsive behavior. However, the relationship among these different components in the disease and the role of DA is not well understood. In this imaging study, we investigated in PD patients the effects of DA medication on patterns of brain activation during tasks testing impulsive choices and actions. Following overnight withdrawal of antiparkinsonian medication, PD patients were studied with a H2 ((15)) O PET before and after administration of DA (1 mg of pramipexole), while they were performing the delay discounting task (DDT) and the GoNoGo Task (GNG). We observed that pramipexole augmented impulsivity during DDT, depending on reward magnitude and activated the medial prefrontal cortex and posterior cingulate cortex and deactivated ventral striatum. In contrast, the effect of pramipexole during the GNG task was not significant on behavioral performance and involved different areas (i.e., lateral prefrontal cortex). A voxel-based correlation analysis revealed a significant negative correlation between the discounting value (k) and the activation of medial prefrontal cortex and posterior cingulate suggesting that more impulsive patients had less activation in those cortical areas. Here we report how these different subcomponents of inhibition/impulsivity are differentially sensitive to DA treatment with pramipexole influencing mainly the neural network underlying impulsive choices but not impulsive action.

Keywords: Parkinson's disease; dopamine agonists; impulsivity.

Figure 1

rCBF activation: Pramipexole activated two different networks depending on the task at hand: during DDT, DA strongly increased rCBF in right medial prefrontal cortex (mPFC) and left posterior cingulate (PCC); during GNG, DA had a minor role by activating the left lateral prefrontal cortex (lPFC). The image shows areas activated in the contrast DDT ON‐DDT OFF and GNG ON‐GNG OFF and individual DA‐rCBF changes extracted by 10 mm sphere from the statistical peak.

Figure 2

rCBF deactivation: Pramipexole during DDT de‐activated the left ventral striatum. No changes were observed during the GNG. The image shows areas de‐activated in the contrast DDT ON‐DDT OFF and GNG ON‐GNG OFF.

Figure 3

Behavioral results: (a) Pramipexole increased impulsivity during DDT depending on reward's amplitude (F _1,5 = 5.65; P=0.006): in large rewards choices, patients ON medications made more impulsive choices; (b) Pramipexole did not affect the performances to GNG, in terms of number of commission errors (CE) and anticipation errors (AE).