Mechanisms involved in treadmill walking improvements in Parkinson's disease.

Patients with Parkinson's disease (PD) improve gait after treadmill training and while they are walking over the treadmill. However, the mechanisms of these improvements have not been addressed. We designed a treadmill simulator without a belt that could move on a walkway in a constant speed, in order to explore the mechanism underlying treadmill walking improvements in PD. All subjects were tested in three different sessions (treadmill, simulator(assisted) and simulator(not assisted)). In each session, subjects first walked overground and then walked using the treadmill or simulator with the hands over the handrails (simulator(assisted)) or with the hands free (simulator(not assisted)). Step length, cadence, double support time, swing time, support time and the coefficient of variation (CV) of step time and double support time were recorded. Over the treadmill PD patients increased their step length and reduced significantly their cadence and CV of double support time in comparison with overground walking. In the simulator(assisted) condition PD patients reduced significantly the CV of double support time in comparison with overground walking. With the simulator(not assisted) both groups decreased their step length and increased their cadence and CV of double support time, compared with walking overground. These findings suggest that the step length improvement observed in PD patients, walking over a treadmill, is due to the proprioceptive information generated by the belt movement, since no improvement was reported when patients using a treadmill simulator.