脑卒中患者认知障碍的评估及危险因素研究

发布时间：2018-04-24 08:06

本文选题：脑卒中 + 血管性认知障碍　；参考：《南京大学》2016年博士论文

【摘要】：背景与目的：脑卒中患者约有三分之二发生血管性认知障碍(vascular cognitive impairment, VCI),但认知评估缺乏统一标准,因此,美国国立神经疾病与卒中研究院-加拿大卒中网络(the National Institute of Neurological Disorders and Stroke and the Canadian Stroke Network, NINDS-CSN)推荐了规范化的认知心理评估方案(包括60 min,30 min和5 min方案),并倡议在不同语言环境中推广。目前,NINDS-CSN认知心理学量表已在韩国、香港、新加坡和法国等人群得以普及,但仍缺乏普通话版量表及人群验证。本研究旨在编译、修订并验证NINDS-CSN神经心理学量表,并在脑卒中患者中,分析不同神经心理学定义对认知障碍发生率的影响,以及验证认知筛查量表(面对面评估和电话评估)的信效度,而后进一步分析脑卒中后发生VCI的危险因素。方法：首先进行NINDS-CSN普通话版量表的编译和修订,并检测其信效度。外在效度定义为各方案总分对脑卒中患者和健康对照的区分能力,体现为受试者工作特征曲线分析的曲线下面积(area under the curve, AUC)。信度分析包括内部一致性信度和重测信度,分别由统计量克朗巴赫α系数和组内相关系数(intraclass correlation coefficients, ICC)进行评估。其次,基于验证的普通话版量表,分析不同神经心理学定义,即低于对照人群均值的1个标准差(standard deviation,SD)、1.5 SD或2 SD以及认知领域的受损由单个或多个测验定义,对认知障碍发生率的影响；同时,使用简易精神状态检查(mini-mental state examination, MMSE)和蒙特利尔认知评估量表(Montreal cognitive assessment test, MoCA)进行面对面评估、5 min方案和认知六项筛查(six item screener, SIS)进行电话评估,验证在脑卒中人群中作为认知筛查工具的有效性,并根据最高约登指数确定最佳阈值。最后,在脑卒中患者中,按是否存在VCI分为两组,比较其人口学、临床和影像等资料(急慢性脑梗塞和腔梗的数目、大小和部位,脑白质高信号和脑萎缩的严重程度),使用单因素分析和多因素logistic回归分析确定脑卒中后发生VCI的危险因素。结果：普通话版量表的验证基于50例轻度脑卒中患者和50例健康对照,外在效度用AUC表示,60 min方案为0.88(95%可信区间[confidence interval,CI],0.82-0.95),30 min方案为0.88(95%CI,0.81-0.95),而5 min方案为0.86(95%CI,0.79-0.93)。各个测验之间的克朗巴赫a系数为0.87:60 min,30 min和5 min方案的重测信度ICC分别为0.90,0.83和0.75。认知筛查量表的验证共纳入89例脑卒中患者(年龄,62.9±8.6岁；男性,65.2%)进行面对面评估,其中80例完成间隔一个月的电话评估。不同神经心理学定义下,认知障碍的发生率从46.3%-76.3%不等,其中遗忘型单领域的认知障碍均较少见；定义越严格,认知障碍发生率越低,并以单领域认知受损的患者为主。不同定义下,面对面量表和电话量表均提示较好的外在效度(AUC0.7)。作为面对面评估,MoCA比MMSE与综合认知评估的一致性更好,认知障碍的阈值以MMSE27分,MoCA≤19分为最佳；电话评估中,5 min量表的评估时间为4.3±1.0分钟,SIS需时57.3±17.7秒钟,但5 min方案比SIS与综合认知评估的一致性更好,最佳阈值为5 min方案≤23分以及SIS≤4分。为了探讨脑卒中后发生VCI的危险因素,选取具备影像资料的68例患者(年龄,62.7±8.8岁；女性,36.8%),距脑卒中事件中位时间7个月后进行NINDS-CSN认知心理评估,其中42例(61.8%)存在VCI,以女性居多(P=0.001),教育程度更低(P0.001),有较多淡漠症状(P=0.008),陈旧性大梗塞灶的数目也更多(P=0.046)。多因素回归分析发现,教育程度(比值比[odds ratio,OR],0.728；95% CI,0.575-0.922;P=0.008)、女性(OR,6.477；95% CI,1.275-32.902;P=0.024)、淡漠评估分数(OR,0.905;95% CI,0.823-0.995;P=0.039)和皮层萎缩(OR,6.131;95% CI,1.351-27.828;P=0.019)是脑卒中后发生VCI的独立影响因素。结论：本研究证实,普通话版NINDS-CSN认知心理学量表(包括60 min,30 min和5 min方案)适用于轻度脑卒中患者的认知评估,不同神经心理学定义下,认知障碍的发生率不同,可相差1.6倍。MMSE、MoCA作为面对面评估,5 min方案和SIS作为电话评估,都是简便有效的认知筛查工具,相应的认知障碍阈值分别为27分,19分,23分和4分。较低的教育程度、女性、淡漠症状和皮层萎缩是脑卒中后发生认知障碍的独立危险因素。
[Abstract]:Background and purpose: about 2/3 of cerebral apoplexy patients have vascular cognitive impairment (VCI), but there is a lack of unified standard for cognitive assessment. Therefore, the National Institute of neurodisease and apoplexy of the United States, Canada Stroke Network (the National Institute of Neurological Disorders and Stroke) Troke Network, NINDS-CSN) recommends a standardized cognitive psychological assessment scheme (including 60 min, 30 min and 5 min), and advocates promoting in different language environments. Currently, the NINDS-CSN cognitive psychology scale has been popularized in Korea, Hongkong, Singapore and France, but still lacks the Putonghua scale and population verification. To compile, modify and verify the NINDS-CSN neuropsychological scale, and to analyze the effects of different neuropsychological definitions on the incidence of cognitive impairment in stroke patients, and to verify the reliability and validity of the cognitive screening scale (face-to-face assessment and telephone evaluation), and then further analyze the risk factors for the occurrence of VCI after stroke. Carry out the compilation and revision of the NINDS-CSN Putonghua scale and examine its reliability and validity. The external validity is defined as the ability to distinguish between stroke patients and health controls by the total score of each scheme, which is the area under the curve of area under the curve (AUC). Reliability analysis includes internal consistency reliability and retest letter. Degrees were evaluated by the statistics Krone Bach alpha coefficient and intraclass correlation coefficients (ICC). Secondly, based on a verifying Putonghua scale, the definition of different neuropsychology was analyzed, that is, 1 standard deviations (standard deviation, SD), 1.5 SD or 2 SD, and cognitive domain. The effect on the incidence of cognitive impairment was impaired by a single or multiple test definition; meanwhile, the face-to-face assessment was performed using the simple mental state examination (Mini-Mental State Examination, MMSE) and the Montreal cognitive assessment scale (Montreal cognitive assessment test, MoCA), and the 5 min scheme and cognitive six screening (six item) A telephone evaluation was conducted to verify the effectiveness of the cognitive screening tool in the stroke population and to determine the best threshold according to the highest index. Finally, in the stroke patients, the VCI was divided into two groups, and the number, size and location of the acute cerebral infarction and the infarct and the infarct, and the high white matter were compared. Signal and brain atrophy severity), using single factor analysis and multiple factor Logistic regression analysis to determine the risk factors for VCI after stroke. Results: the verifying of the Putonghua scale was based on 50 cases of mild stroke patients and 50 healthy controls, the external validity was expressed with AUC, and the 60 min scheme was 0.88 (95% confidence interval [confidence InterVA). L, CI], 0.82-0.95), the 30 min scheme was 0.88 (95%CI, 0.81-0.95), and the 5 min scheme was 0.86 (95%CI, 0.79-0.93). The Krone Bach a coefficient between each test was 0.87:60 min. The 30 and 5 scheme's retest reliability included 89 cases of stroke patients (age, 62.9 + 8.6 years old; male, 6). 5.2%) face to face assessment, of which 80 patients completed a one month telephone assessment. Under different neuropsychological definitions, the incidence of cognitive impairment ranged from 46.3%-76.3%, of which all cognitive disorders in the amnestic single domain were less common; the more strict the definition, the lower the incidence of cognitive impairment, and the difference in the patients with single domain cognitive impairment. Under the definition, both the face scale and the telephone scale showed good external validity (AUC0.7). As a face to face assessment, the consistency of MoCA was better than that of the comprehensive cognitive assessment. The threshold of cognitive impairment was MMSE27 and MoCA < 19 as the best. The evaluation time of the 5 min scale was 4.3 + 1 minutes in the telephone evaluation, and 57.3 + 17.7 seconds when SIS was required, but it was 57.3 + 17.7 seconds for SIS. The 5 min scheme was better than the SIS and the comprehensive cognitive assessment, the best threshold was 5 min schemes less than 23 and SIS < 4. To explore the risk factors of VCI after stroke, 68 patients with imaging data (age, 62.7 + 8.8 years, women, 36.8%) were selected and the cognitive psychology of the event was 7 months after the event in the stroke. Among them, 42 (61.8%) had VCI, with P=0.001, P0.001, P=0.008, and older large infarcts (P=0.046). Multiple regression analysis found that education (ratio ratio [odds ratio, OR], 0.728; 95% CI, 0.575-0.922; P=0.008), and women (OR, 6.477; 95%, 1.275) -32.902; P=0.024), the indifference assessment score (OR, 0.905; 95% CI, 0.823-0.995; P=0.039) and cortical atrophy (OR, 6.131; 95% CI, 1.351-27.828; P=0.019) are independent factors of VCI after stroke. Conclusion: This study confirms that the Putonghua NINDS-CSN cognitive psychology scale (including 60, 30 and 5) is suitable for mild stroke. Patients' cognitive assessment, different neuropsychological definitions, the incidence of cognitive impairment is different, can differ 1.6 times.MMSE, MoCA as a face-to-face assessment, 5 min scheme and SIS as a telephone evaluation, is a simple and effective cognitive screening tool, the corresponding cognitive impairment threshold of 27, 19, 23 and 4 points. Lower education, women, Apathy symptoms and cortical atrophy are independent risk factors for cognitive impairment after stroke.

【学位授予单位】：南京大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：R743.3

【相似文献】