摘要： 目的 总结帕金森病伴头晕患者视眼动功能障碍特点，并分析帕金森病患者发生头晕的危险因素。方法 纳入2019年1月至2022年10月北京大学第一医院收治的80例帕金森病患者，根据入组前1个月是否存在头晕主诉分为头晕组（35例）和非头晕组（45例）。对头晕患者采用视觉模拟评分和头晕残障量表（DHI）进行主观头晕程度评估；对所有患者进行视频眼震电图检查，包括自发试验、凝视试验、反射性扫视试验、记忆扫视试验、平稳跟踪试验、视动试验。采用单因素和多因素线性回归分析筛查头晕患者DHI评分的影响因素，单因素和多因素Logistic回归分析筛查帕金森病患者发生头晕的影响因素，根据头晕患者是否伴神经源性直立性低血压对临床特点及视眼动参数进行亚组分析。结果 多因素线性回归分析显示，神经源性直立性低血压（标准化偏回归系数=0.385，P=0.031）和平均扫视潜伏期（标准化偏回归系数=0.400，P=0.028）对头晕患者DHI评分产生正向影响。多因素Logistic回归分析显示，神经源性直立性低血压（OR=4.701，95% CI：1.398~15.806；P=0.012）和平均扫视潜伏期较长（OR=1.013，95% CI：1.002~1.025；P=0.022）是帕金森病患者发生头晕的危险因素。根据是否存在神经源性直立性低血压将头晕患者分为无直立性低血压亚组（20例）和直立性低血压亚组（15例），与无直立性低血压亚组相比，直立性低血压亚组阻塞性睡眠呼吸暂停低通气综合征比例较高（Fisher确切概率法：P=0.032），平均扫视潜伏期10°（t=2.147，P=0.039）和平均扫视潜伏期25°（t=2.698，P=0.011）较短。结论 神经源性直立性低血压和平均扫视潜伏期延长是帕金森病患者发生头晕的危险因素，并可增加头晕严重程度；帕金森病伴头晕患者易合并快速眼动睡眠期行为障碍和阻塞性睡眠呼吸暂停低通气综合征等睡眠障碍。

关键词: 帕金森病, 头晕, 眼球运动检测, 眼扫视, 低血压，直立性, 睡眠障碍, 危险因素, Logistic模型, 回归分析

Abstract: Objective To summarize the visuo-oculomotor dysfunction in Parkinson's disease (PD) patients with dizziness, and to analyze the risk factors for dizziness. Methods A total of 80 patients with PD visiting Peking University First Hospital between January 2019 and October 2022 were enrolled and divided into dizziness group (n=35) and non-dizziness group (n=45) based on the presence or absence of dizziness complaints within one month prior to enrollment. Subjective dizziness severity was assessed using the Visual Analog Scales (VAS) and the Dizziness Handicap Inventory (DHI) for patients with dizziness. All patients underwent video-nystagmography (VNG) examinations, including without-fixation test, fixation test, reflexive saccade test, memory-guided saccade test, smooth pursuit test and optokinetic test, to obtain visuo-oculomotor function related parameters. Univariate and multivariate linear regression analyses were used to analyze influencing factors for DHI score. Univariate and multivariate Logistic regression analyses were used to screen influencing factors for dizziness in PD patients. Subgroups analysis of clinical characteristics and visuo-oculomotor function related parameters was performed based on the presence or absence of neurogenic orthostatic hypotension (nOH) in dizziness group. Results Multivariate linear regression analysis showed that the presence of nOH (standardized partial regression coefficient=0.385, P=0.031) and the average saccade latency (standardized partial regression coefficient=0.400, P=0.028) had positive influences on the DHI score. The presence of nOH (OR=4.701, 95%CI: 1.398-15.806; P=0.012) and a prolonged average saccade latency (OR=1.013, 95%CI: 1.002-1.025; P=0.022) were risk factors for dizziness in PD patients. Patients in the dizziness group were further divided into 2 subgroups based on the presence of nOH: subgroup without nOH (n=20) and subgroup with nOH (n=15). Compared to the subgroup without nOH, the subgroup with nOH had a higher proportion of obstructive sleep apnea-hypopnea syndrome (OSAHS, Fisher's exact probability: P=0.032), and shorter average saccade latencies at 10° (t=2.147, P=0.039) and 25° (t=2.698, P=0.011) targets. Conclusions The complication of nOH and prolonged average saccade latency are risk factors for dizziness in PD patients and increase the severity of dizziness. PD patients with dizziness are prone to manifest sleep disorders such as rapid eye movement sleep behavior disorder (RBD) and OSAHS.

Key words: Parkinson disease, Dizziness, Eye movement measurements, Saccades, Hypotension, orthostatic, Sleep disorders, Risk factors, Logistic models, Regression analysis