Abstract: Objective To investigate the effect of cognitive-motor dual-task on gait automaticity in patients with Parkinson's disease (PD).Methods A total of 45 patients with primary PD in Tianjin Huanhu Hospital were selected from June 2020 to December 2021. Cognitive single task, motor single task and cognitive-motor dual-task in the "off" and "on" states of dopaminergic medication were performed. The correct number of subtractions was recorded to measure cognitive performance. A three-dimensional gait analyzer was used to extract the parameters of stride length and gait velocity, and the dual-task cost (DTC) was calculated to measure gait automaticity.Results Compared with single task, patients with PD showed slower gait velocity (t = 11.909, P = 0.000; t = 11.621, P = 0.000), shortened stride length (t = 9.027, P = 0.000; t = 9.991, P = 0.000), while there wasnosignificant difference in thecorrect number of subtraction(t = 0.651, P = 0.519; t = 1.205, P = 0.234) during both "off" and "on" states.In patients with PD, the DTCof gait velocity (Z = - 4.148, P = 0.000) and stride length (Z = - 4.114, P = 0.000) in the "off" state was higherthan that in the "on" state, while the DTC forsubtraction hadno significant difference (Z = - 1.005, P =0.315). The DTC of gait velocity in the "off" and "on" states of patients with PD was negatively correlatedwith the Mini-Mental State Examination (MMSE) score (r_s = - 0.474, P = 0.000; r_s = - 0.461, P = 0.000), butpositively correlated with Unified Parkinson's Disease RatingScale Ⅲ (UPDRS Ⅲ) score (r_s = 0.446, P =0.002; r_s= 0.428, P = 0.003).Conclusions Patients with PDhave a significant decrease in stride lengthand gaitvelocity, and higherrisk of fallingwhen performingdual-task.Dopaminergic medication can improve the gait automaticity of patients, and cognitive decline may lead to gait automaticity disorder in patients with PD.

Key words: Parkinson disease, Gait analysis, Cognitive reserve

摘要： 目的 探讨认知-运动双任务对帕金森病患者步态自动性的影响。方法 共纳入2020年6月至2021年12月在天津市环湖医院诊断与治疗的45例原发性帕金森病患者，分别在拟多巴胺药“关”期和“开”期执行认知单任务、运动单任务和认知-运动双任务，记录减数正确个数以及采用三维步态分析仪提取步速和步幅参数，并计算双任务成本。结果 与单任务相比，帕金森病患者“关”期和“开”期执行双任务时均步速减慢（t = 11.909，P = 0.000；t = 11.621，P = 0.000）、步幅缩短（t = 9.027，P = 0.000；t = 9.991，P = 0.000），而减数正确个数差异均无统计学意义（t = 0.651，P = 0.519；t= 1.205，P = 0.234）。帕金森病患者“关”期步速双任务成本（Z = - 4.148，P = 0.000）和步幅双任务成本（Z = - 4.114，P = 0.000）均高于“开”期，而减数双任务成本差异无统计学意义（Z = - 1.005，P = 0.315）。帕金森病患者“关”期和“开”期步速双任务成本均与简易智能状态检查量表评分呈负相关（r_s = - 0.474，P = 0.000；r_s =- 0.461，P = 0.000），而与统一帕金森病评价量表第三部分评分呈正相关（r_s = 0.446，P = 0.002；r_s = 0.428，P = 0.003）。结论 帕金森病患者在执行双任务时步速减慢、步幅缩短，存在较高的跌倒风险，拟多巴胺药可有效改善步态自动性，认知功能下降可能导致帕金森病患者步态自动性下降。

关键词: 帕金森病, 步态分析, 认知储备