Skip secondary menu

The Effect of Vertical and Horizontal Lines in Virtual Reality on Freezing of Gait for Parkinson’s Disease Patients


Abigail H. Fulp1,2,3, Kathrin Czarnecki, MD3, Christopher J. Stanley MS2, Lindsey A. Curatalo MS2, Beverly Elroy, CNRN2, Hyung S. Park, PhD2, Diane L. Damiano, PhD2, Mark Hallett, MD3

1North Carolina State University, College of Engineering, Biomedical Engineering Department, 2Rehabilitation Medicine Department, CC, NIH, 3National Institute of Neurological Disorders and Stroke, NIH

Objective: To study freezing of gait (FOG) in Parkinson's disease (PD) under virtual reality (VR) conditions with horizontal and vertical lines, using a patient speed-controlled treadmill system.

Background: Freezing of gait is a symptom experienced by some Parkinson’s Disease patients that can be disabling and greatly affect simple daily activities.  It occurs when patients shorten their step and sometimes come to a stop while walking, approaching a destination, turning, entering a doorway, or in a stressful situation.  It is believed that visual input is the biggest factor causing these freezes and that vertical lines often cause freezes while horizontal lines can potentially help prevent freezes. 

Methods: 3 PD patients with FOG (FOG+), and 3 PD patients without FOG (FOG-) walked on a self-selected speed treadmill using 3D goggles to see the virtual reality environments projected in front of them.  They walked through a baseline hallway and conditions with transverse, perpendicular, horizontal and vertical lines.  Using the marker coordinates and the force plate data, the toe offs and heel strikes for each step were marked in Visual 3D.  Step length, step time, velocity, and cadence were then calculated in Matlab. FOG episodes (defined as a drop in step length below 10% of their average step length for that trial) were counted by review of gait analysis and videotapes.

Results: No authentic freezing episodes were observed, however other gait parameters were analyzed.  Both FOG+ and FOG- patients walked with a faster velocity, longer step length, longer step time, and lower cadence overground than they did on the treadmill.  FOG- patients walked with a faster velocity, longer step length, longer step time, and lower cadence in all conditions compared to FOG+.  The trends between horizontal, vertical, transverse, and perpendicular environments were not notable.  Asymmetry of step length between left and right steps was also examined and found that FOG+ patients had 0.0472 m of asymmetry and FOG- patients had 0.0446 m of asymmetry. The study findings are preliminary and need to be confirmed with a greater sample size.

Conclusion: Although freezing was unable to be observed in this study, FOG- patients walked with more ease and more similar to their overground gait while on the treadmill.   Both patient groups were walking slower on the treadmill compared to overground walking which is consistent with previous studies. The insignificant difference of asymmetry between the two groups shows that asymmetry is a symptom observed in all PD patients and may not predict freezing episodes. Future goals are to incorporate patient self-reporting for freezing and further improve the treadmill controllers and visual environments to provide more realistic walking conditions.

Last updated December 14, 2012