The Effect of Levodopa on Improvements in Protective Stepping in People With Parkinson's Disease.

Only 22 participants were tested, and all could stand without aid for more than 1 hour, so it may not apply to people with more severe balance impairment. Training was only one day, and retention was measured only at 24 hours, so longer-term learning and real-world carryover were not tested. The study measured laboratory stepping responses, not falls, so it does not show fall prevention. Some improvement may have carried over between visits: center of mass displacement at follow-up improved nonsignificantly (p=0.10), and margin of stability improved significantly (p=0.006), although testing order was randomized. Early changes could reflect habituation to perturbations rather than learning, although participants had more than 10 familiarization perturbations before training. Fatigue was measured in only 16 of 22 participants, and overall fatigue was low, but higher OFF levodopa. Washout periods varied, with mean 9.47±3.79 weeks, which may add variability. Generalization to sideways tasks may be partly explained by worse baseline center of mass displacement ON levodopa. Step latency was missing for 2 participants because of force-plate malfunction.

Dr. Horak and OHSU have an equity or interest in APDM, a company that may have a commercial interest in the results. This potential conflict was reviewed and managed by the Research and Development Committee at the Portland VA Medical Center and OHSU. No other authors declared a conflict of interest.

Do not conclude that levodopa before balance training will prevent falls. The study measured laboratory stepping responses after one day, not real-world falls, and generalization may reflect worse baseline performance ON levodopa.