摘要目的:比较脑小血管病（CSVD）患者认知以及运动双重任务行走（DTW）的步态特征差异，并判断何种步态参数能够更好地诊断CSVD及判断疾病的严重程度。方法:纳入2020年9月1日至2021年7月1日在解放军总医院第七医学中心神经内科就诊的经磁共振成像证实有CSVD的患者106例以及同期的体检人员21名，根据Fazekas评分，将研究对象分为轻度组（34例，1分）、中度组（34例，2分）、重度组（38例，3分）以及健康对照组（21名）。在单任务行走（STW）和DTW条件下记录参与者参数，并通过单任务和双任务之间的差异计算双重任务效应（DTC）。采用广义估算方程评估疾病严重程度以及行走条件对基本步态参数及DTC的影响，事后分析通过Bonferroni方法校正。通过Spearman相关分析确定疾病严重程度与步态参数及DTC间的相关性。通过Logistic回归建立联合预测因子，应用联合诊断试验计算步态参数及其DTC预测CSVD的敏感度、特异度及受试者工作特征（ROC）曲线下面积。结果:试验组中STW的基础步态参数显著优于认知或运动DTW（均 P<0.05），而对照组中不同任务下的基础步态参数差异无统计学意义（均 P>0.05）。在认知DTW中，时间步态参数（步频和跨步时间）仅在中重度组出现显著恶化［步频：中度组（100.220±1.795）步/min，重度组（94.525±2.139）步/min；跨步时间：中度组（1.227±0.024）s，重度组（1.299±0.031）s］，但空间参数（跨步长及步速）在各组（除对照组和轻度组外）间［跨步长：对照组（1.050±0.021）m，轻度组（0.974±0.022）m，中度组（0.903±0.025）m，重度组（0.793±0.026）m；步速：对照组（0.944±0.028）m/s，轻度组（0.866±0.030）m/s，中度组（0.751±0.027）m/s，重度组（0.606±0.022）m/s］差异均有统计学意义（均 P<0.05）。同样无论疾病严重程度，认知DTW条件下所有步态参数DTC均高于运动DTW（均 P<0.05），而对照组中不同任务下的步态参数DTC差异无统计学意义（均 P>0.05）。其中认知DTW的时间参数DTC仅在中重度组出现异常，而空间参数DTC在各组间差异均有统计学意义（包括对照组和轻度组；均 P<0.05）。认知DTW条件下，空间参数及DTC与疾病严重程度的相关性显著高于时间参数（0.50< r<0.64比0.15< r<0.39）。应用空间参数对CSVD的判别进行联合诊断试验发现，联合诊断因子ROC曲线下面积显著大于单一指标。 结论:认知DTW能够更好地反映CSVD患者的步态异常。认知DTW的空间参数及其DTC能更有效地诊断CSVD并区分其严重程度，DTC可能是更好的指标。对于诊断CSVD而言，空间参数及其DTC之间差异并无统计学意义，但联合因子可以显著提高敏感度，降低假阴性率。

abstractsObjective:To compare the gait characteristics of cognitive and motor dual task walking (DTW) in patients with cerebral small vessel disease (CSVD), and determine the best gait parameters to diagnose CSVD and judge the severity of the disease.Methods:A total of 106 patients with CSVD and 21 healthy individuals were included from September 1, 2020 to July 1, 2021 in the Seventh Medical Center of Chinese People′s Liberation Army General Hospital. According to the Fazekas scores, the subjects were divided into mild ( n=34, 1 point), moderate ( n=34, 2 points), severe ( n=38,3 points) groups and control group ( n=21). Participants were recorded parameters under single task walking (STW) and DTW conditions, and calculated dual task effect (DTC) through the difference between single task and dual task. The differences in gait variances and their DTC were shown by generalized estimation equations when performed in STW and DTW and 4 groups of the severity of disease. Post-hoc comparisons were corrected using Bonferroni′s method. Spearman analyses were applied to explore the correlations between gait parameters and their DTC during STW or DTW and severity of disease. Based on the Logistic model, combining predictors or probabilities were gained and applied to establish receiver operating characteristic curve in order to calculate sensitivity, specificity, and the area under the curve. Results:In the control group, there was no statistically significant difference in gait parameters between STW and DTW. In the CSVD group, the gait parameters of STW were significantly better than cognitive or motor DTW (all P<0.05). In the control group, there was no statistically significant difference in basic gait parameters under different tasks (all P>0.05). In cognitive DTW, temporal gait parameters (stride frequency and stride time) deteriorated significantly only in moderate and severe groups [stride frequency:moderate group 100.220±1.795/min,severe group 94.525±2.139/min;stride time:moderate group (1.227±0.024) s, severe group (1.299±0.031) s], but spatial parameters [stride length: control group (1.050±0.021) m, mild group (0.974±0.022) m, moderate group (0.903±0.025) m, severe group (0.793±0.026) m; stride speed: control group (0.944±0.028) m/s, mild group (0.866±0.030) m/s, moderate group (0.751±0.027) m/s, severe group (0.606±0.022) m/s] were significantly different among all groups (except the control group and mild group;all P<0.05). The DTC of all gait parameters during cognitive DTW was higher than that during motor DTW (all P<0.05) for CSVD patients. While no any difference was found between cognitive DTW and motor DTW in the control group (all P>0.05). Similarly, the temporal parameters′ DTC of cognitive DTW was abnormal only in the late stage of disease, while the spatial parameters′ DTC showed statistically significant difference among all the groups (including the control group and the mild group;all P<0.05). Correlation coefficients of the spatial parameters and their DTC in condition of cognitive DTW were significantly higher than temporal parameters and their DTC (0.50< r<0.64 vs 0.15< r<0.39). The area under curve of the combined predictor was significantly higher than that of any single index. Conclusions:Cognitive DTW can better reflect the abnormal gait of CSVD patients. The spatial parameters and DTC of cognitive DTW could effectively diagnose CSVD and distinguish the disease of severity. And DTC might be better indicators. For diagnosis of CSVD, there was no significant discrepancy between the spatial parameters and DTC, but the combined predictor could significantly improve the sensitivity and reduce the false negative rate.