OBJECTIVE: To measure "on" freezing during unassisted walking (UW) and test
if two devices, a modified inverted stick (MIS) and a visual laser beam st
ick (LBS) improved walking speed and number of "on" freezing episodes in pa
tients with Parkinson's disease (PD).
BACKGROUND: Multiple visual cues can overcome "off" freezing episodes and c
an be useful in improving gait function in parkinsonian patients. These dev
ices have not been specifically tested in "on" freezing, which is unrespons
ive to pharmacologic manipulations.
METHODS: Patients with PD, motor fluctuations and freezing while "on," atte
mpted walking on a 60-ft track with each of three walking conditions in a r
andomized order: UW, MIS, and LBS. Total time to complete a trial, number o
f freezes, and the ratio of walking time to the number of freezes were comp
ared using Friedman's test.
RESULTS: Twenty-eight patients with PD, mean age 67.81 years (standard devi
ation [SD] 7.54), mean disease duration 13.04 years (SD 7.49), and mean mot
or Unified Parkinson's Disease:Rating Scale score "on" 32.59 (SD 10.93), pa
rticipated in the study. There was a statistically significant correlation
of time needed to complete a trial and number of freezes for all three cond
itions (Spearman correlations: UW 0.973, LBS 0.0.930, and MIS 0.842). The m
edian number of freezes, median time to walk in each condition, and median
walking time per freeze were not significantly different in pairwise compar
isons of the three conditions (Friedman's test). Of the 28 subjects, six sh
owed improvement with the MIS and six with the LBS in at least one outcome
measure.
CONCLUSION: Assisting devices, specifically based on visual cues, are not c
onsistently beneficial in overcoming "on" freezing in most patients with PD
. Because this is an otherwise untreatable clinical problem and because occ
asional subjects do respond, cautious trials of such devices under the supe
rvision of a health professional should be conducted to identify those pati
ents who might benefit from their long-term use.