The influence of levodopa-induced dyskinesias on manual tracking in patients with Parkinson's disease.

The influence of peak-dose, levodopa-induced dyskinesias (LID) on manual tracking (MT) was examined in 10 dyskinetic patients with Parkinson's disease (DPD), and compared to 10 age/gender-matched non-dyskinetic patients with Parkinson's disease (NDPD) and 10 healthy controls. Whole-body movement (WBM) and MT performance were recorded simultaneously with a 6-degrees-of-freedom magnetic motion tracker and forearm rotation sensors, respectively. Subjects were asked to match the length of a computer-generated line with a line they controlled via wrist rotation. Results show that DPD patients had greater WBM magnitude at rest and during the motor task, both in displacement and in velocity. All groups displayed some increase in WBM displacement from rest to MT, but only the DPD group had a significant increase in WBM velocity during movement. As for MT performance (determined by assessing the positional mismatch between subjects' and target lines), ERROR in displacement was statistically similar between groups. There was no correlation between ERROR and the magnitude of WBM within the DPD group. The DPD group showed significant increased ERROR when the velocity of the subject's line was compared with that of the velocity of the target line. When two distinct target pace segments were examined (FAST/SLOW), no significant differences were found in ERROR for displacement for either group, but both the NDPD and DPD group showed increased ERROR from SLOW to FAST for velocity. This was accompanied with an increase in WBM velocity only in the DPD group. The lack of increased ERROR during the SLOW tracking portion in the DPD group supports the notion that the dyskinesias themselves were not primarily responsible for the ERROR seen in the patients. When examining the positive or negative values of ERROR (i.e., faster or slower than the target), we found that the increased ERROR in velocity observed in the DPD group was the result of excess velocity rather than bradykinesia, manifested as isolated deviations from the target trace in the DPD group that were coherent in time with increased ERROR in velocity. In conclusion, evidence presented in the present study shows that the LID was not the primary cause of the ERROR seen in the DPD group. Accordingly, we propose that the increased ERROR seen in the DPD group resulted from a mechanism distinct from the one generating LID.