【作者】 李晓裔；

【导师】 周华东；

【作者基本信息】 第三军医大学 ， 神经病学， 2011， 博士

【摘要】 背景与目的:老年期痴呆的原因很多,阿尔茨海默病(Alzheimer’s disease,AD)是最常见的一种[1],通常以记忆障碍开始逐渐发展为全面认知功能衰退,严重影响了老年人的生活质量,给家庭和社会带来很大的负担。轻度认知功能障碍(mild cognitive impairment,MCI)已作为一个独立的疾病受到重视,MCI患者是转化为AD的高危人群[2, 3]。在健康老年人与AD之间有一个较长的发展过程,MCI是介于正常人老年人与AD之间的发展阶段[4]。由于AD的治疗效果较差,MCI是对AD进行早期干预治疗的有利阶段,已成为目前的研究热点。听觉事件相关电位(auditory eventrelated potentials,AERPs)是一种可靠的脑活动对听觉刺激反应的一种无创伤性神经电生理检测,是帮助确认认知障碍和心理状态的评估手段[5]。对反应听觉刺激的oddball任务产生的事件相关电位已被广泛用于定量测量皮层感觉活动和认知功能。近年来很多文献报道了AERPs的不同成分在MCI的研究,主要用来诊断、预测和观察疗效,认为有很高的敏感性和特异性[6-9]。磁共振波谱(magnetic resonance spectroscopy,MRS)是神经影像学中发展迅速的领域,可对神经化学物质及其代谢物进行非侵入性活体分析,在脑结构改变出现之前即可显示出脑代谢变化。在过去的二十年中,已广泛应用于神经及精神疾病,帮助了解这些疾病的病理机制,监测药物治疗或非药物治疗的长期变化,并确定诊断组之间的差异。目前MRS检测已用于认知功能障碍性疾病[10],也用于MCI的诊断、分型和预测转归,肯定了其敏感性[11-14]。本研究通过AERPs的各成分分析,寻找反应MCI早期认知损害的神经电生理指标。通过氢质子磁共振波谱分析(1H-MRS)寻找反应MCI早期认知损害的脑物质代谢指标。了解MCI的脑功能变化和物质代谢变化特征,观察与临床心理学特征之间的联系。探讨AERPs和1H-MRS可否作为MCI诊断的敏感方法。同时观察MCI的脑功能变化和物质代谢变化特征,是否介于健康老人和AD之间;临床心理学评分是否也介于二者之间;脑功能变化和相关脑区物质代谢是否有相关性;脑功能变化和相关脑区物质代谢分别与临床心理学评分是否有相关性。通过分析他们的关系,及早发现MCI大脑电活动的变化及反应其变化的相应脑区物质代谢异常。验证AERPs和1H-MRS异常是否与临床认知测试评分变化一致,以说明AERPs和1H-MRS为早期MCI诊断和鉴别诊断敏感的手段。两者联合应用可说明脑代谢变化是脑功能异常的反应,这些改变特征与临床变化特征一致。联合应用AERPs和1H-MRS可提高MCI诊断的准确性。研究方法1、对符合MCI和AD的患者和年龄匹配的健康老人进行对照研究。MCI组34例,男23例,女11例,年龄62～85岁,平均年龄71.6±5.7岁,教育年限7.8±1.2年。AD组18例,男11例,女7例,年龄60～83岁,平均年龄73.4士5.6岁,教育年限7.1±1.4年。对照组(Normal controls,NC)34例,男21例,女13例,年龄57～80岁,平均年龄71.6±5.7岁,教育年限7.8±1.2年。首先用神经心理学量表对三组受试者进行总体认知功能、分项记忆功能、语言功能、执行功能和视觉空间功能评估。2、对MCI、AD和NC受试者完成心理测试后通过声刺激Oddball模式,即通过听音调序列的靶探测任务诱发出AERPs。在头皮记录刺激所诱发的电位成分,获得P50,N100,P200、N200和P300波。分析三组受试者非靶刺激Cz点记录的P50,N100,P200波潜伏期和波幅的差异性;靶刺激Pz点记录的N200和P300波潜伏期和波幅的差异性。3、对MCI、AD和NC组受试者在AERPs检测后三天内进行1H-MRS检测。感兴趣区选在左额叶、左颞叶和右顶叶皮层。采集三个脑区的脑代谢物N-乙酰天冬氨酸(N-acetyl-aspartate,NAA)、肌酸(Creatine,Cr)和胆碱(Choline ,Cho)数据。分析三组受试NAA、Cr和Cho水平;NAA/Cr和NAA/Cho的差异性。4、分析MCI的AERPs成分P50波幅和P300潜伏期分别与左额叶和左颞叶皮层NAA/Cr的相关性。分析P50波幅、P300潜伏期、左额叶和左颞叶皮层NAA/Cr诊断MCI的准确性。5、分析MCI的AERPs成分P50、N100、P200、N200和P300波的波幅和潜伏期分别与其各项神经心理评分的相关性。6、分析MCI左额叶、左颞叶和右顶叶皮层NAA/Cr分别与其各项神经心理评分的相关性。7、统计学分析:所得数据用均值±标准差( x±s)表示。用SPSS16.0对数据进行处理,多组间进行单因素方差分析,用LSD法进行组间比较;相关性分析用Pearson相关分析;回归分析用线性回归。对单项及联合检测结果作图绘成受试者工作特征曲线,计算曲线下面积和标准误。以P<0.05表示差异具有统计学意义,P<0.01表示统计学差异非常显著。结果1、轻度认知功能障碍神经心理学特征MCI组MMSE、记忆测试得分较NC低,差异非常显著(P< 0.001或P< 0.0001);语言功能的Boston命名得分也降低,差异显著;语言功能的词语流畅试验、执行功能和视觉空间功能得分均无显著差异(均P>0.05)。AD组各项测试得分均较MCI组低,差异显著或非常显著(P< 0.05或P< 0.0001)。2、听觉事件相关电位在轻度认知功能障碍诊断中的作用对非靶刺激的反应:MCI组P50波幅大于NC组,差异非常显著(P< 0.001);P50潜伏期、N100和P200波幅及潜伏期与NC组比较均差异不显著(均P>0.05)。AD组P50波幅较MCI组低,差异显著(P< 0.05);AD组与MCI组比较,P50潜伏期、N100和P200波幅及潜伏期均差异不显著(均P>0.05))。对靶刺激的反应:MCI组P300潜伏期较NC组延长,波幅降低,均差异显著(均P< 0.05);N200波幅和潜伏期与NC组比较均差异不显著(均P>0.05)。AD组P300潜伏期较MCI组延长,差异非常显著,波幅降低,差异显著(均P< 0.01);AD组与MCI组比较N200波幅和潜伏期均差异不显著(均P>0.05)。3、磁共振波谱在轻度认知功能障碍诊断中的作用MCI组与NC组比较,左额叶皮层NAA水平和NAA / Cr均降低,均差异非常显著(均P< 0.01);左颞叶皮层NAA和NAA / Cr降低,均差异显著(均P< 0.05);右顶叶皮层NAA和NAA / Cr均差异不显著(P>0.05);三个脑区的Cr、Cho水平和Cho/Cr均差异不显著(均P>0.05)。AD组与MCI组比较左额叶、左颞叶和右顶叶皮层NAA和NAA / Cr均降低,均差异非常显著(P< 0.0001);三个脑区的Cho水平和Cho/Cr均差异不显著(均P>0.05)。4、MCI听觉事件相关电位和磁共振波谱与认知功能评分的相关性MCI的P50波幅和P300潜伏期与神经心理测试各记忆分项得分均呈负相关关系;MCI左额叶和左颞叶皮层的NAA / Cr与神经心理测试各记忆分项得分均呈正相关关系。P50波幅与左额叶皮层的NAA / Cr呈负相关关系;P300潜伏期与左额叶和左颞叶皮层的NAA / Cr呈负相关关系。在鉴别NC和MCI及MCI和AD时,受试者工作特征曲线下面积显示联合P50波幅、P300潜伏期、左额叶和左颞叶皮层的NAA / Cr四个指标检测时曲线下面积最大。结论1.本组MCI有明显的记忆障碍,部分语言功能损害,执行功能和视觉空间功能无损害,符合MCI的认知特点。2. MCI特点是P50波幅增大,P300潜伏期延长不如AD显著。P50波幅和P300潜伏期可用于MCI的诊断。3. MCI的特点是NAA / Cr降低,但较AD患者NAA / Cr数值高,受影响的脑区少。NAA / Cr可用于MCI的诊断。4. P50波幅、P300潜伏期和NAA / Cr与认知功能中记忆功能相关,P50波幅和P300潜伏期与NAA / Cr相关。联合AERPs和~1H-MRS检测时诊断MCI的准确性明显提高。AERPs和1H-MRS联合应用可作为诊断和跟踪MCI的手段。更多还原

【Abstract】 There are many reasons for elderly dementia, Alzheimer’s disease (AD) is the most common.Widespread cognitive decline appear gradually,seriously affecting the quality of life of older persons.Mild cognitive impairment (MCI) is a disorder in older patients that is initially characterized by cognitive disturbances,such as memory or language function that is a major risk factor for AD.MCI has been paid more attention as an independent disease.MCI patients are at high risk of conversion to AD.It is a long development process between healthy elderly and AD.Because there are no effective treatment on AD, MCI is the intervention stage, has become the focus of research.Auditory event-related potential (AERPs) is a reliable non-invasive detection of nerve electrophysiology to identify cognitive impairment and mental state assessment.Magnetic resonance spectroscopy (MRS) is a non-invasive neural imaging technology to analyze brain metabolites in vivo.This study was aimed to examine the AERPs and 1H-MRS characteristic changes of MCI, to examine the changes in AERPs and their relationship with brain metabolic changes in MCI by analyzing the results of controls,MCI and AD subjects.By examining AERPs and brain biochemical components, We observe the changes of these parameters in MCI subjects in order to find neuroelectrophysiological indexes to reflect cognitive impairment in earlier period of MCI.We analyzed the correlation of AERPs components abnormalities and the biochemical components abnormalities in different brain regions,and the correlation of abnormalities in AERPs and 1H-MRS with the clinical cognitive test scores.We study their relationship in order to find the early metabolic abnormalities of different brain regions that may reflect the changes of brain activities in the MCI.We study the relationship of abnormalities in AERPs and 1H-MRS with the clinical cognitive test scores in order to identify concordance of these changes, identify the role of AERPs and 1H-MRS measurements.They may offer means to diagnosis MCI and track changes of brain activities associated with functional and metabolic changes,and to assess early cognitive impairment in MCI.Combined 1H-MRS and AERPs can improve diagnostic accuracy of MCI.METHODS1.The study involve MCI,mild AD patients and age-matched healthy elderly control subjects.MCI group,34 patients,23 males and 11 females,aged 62 to 85 years,mean age 71.6±5.7 years,years of education 7.8±1.2 years.AD group,18 patients,11 males and 7 females,aged 60 to 83 years,mean 73.4±5.6 years,years of education 7.1±1.4 years.The control group 34 patients,21 males and 13 females,aged 57 to 80 years,mean age 71.6±5.7 years,years of education 7.8±1.2 years.The cognitive functions of the all subjects were assessed by clinical neurological and neuropsychological assessment scales in overall cognitive function,memory function,language function, executive function and visual spatial function.2.After neuropsychological assessments,the event related potentials components P50,N100,P200,N200 and P300 waves of the subjects evoked by a sonic stimulus Oddball mode,a target detection task by listening to a sequence of tones.The stimulus paradigm were recorded in MCI,mild AD and control subjects.We analyzed group differences of P50,N100,P200 latencies and amplitudes recorded at the Cz site;N200 and P300 waves latencies and amplitudes differences recorded at the Pz site.3.Brain metabolic data were collected three days after AERPs were examined in all subjects.The concentrations of brain metabolites and the metabolite ratios were obtained in three different brain areas:left frontal,left temporal and right parietal cortex by magnetic resonance spectroscopy.The brain metabolites were N-acetyl aspartate (NAA),Creatine (Cr) and Choline (Cho).The metabolite ratios were NAA/Cr and NAA/Cho.We analyzed group differences in MCI,mild AD and control subjects.4.We analyzed correlation between AERPs parameters and cerebral NAA/Cr ratios in left frontal and left temporal lobe cortex of MCI.AERPs parameters were P50 amplitude and P300 latency.We analyzed the diagnostic accuracy of P50 amplitude,P300 latency,NAA/Cr of left frontal and left temporal cortex to MCI.5.We correlate P50,N100,P200,N200,P300 latencies and amplitudes with clinical neuropsychological scores of MCI.6.We correlate cerebral NAA/Cr ratios in left frontal,left temporal lobe and right parietal cortex with clinical neuropsychological scores of MCI.7.Statistical analysis:Data are expressed as mean±SD.SPSS16.0 software was used for statistical analysis.One way ANOVA was used with the LSD multiple comparison tests for multiple groups;Pearson correlation was used for correlation analysis;linear regression was used for regression analysis.We drew receiver operating characteristic curve according to the results of individual detection and combined detection,calculated area under the curve and standard error.Statistical significance was accepted at a level of P < 0.05.RESULTS1. Neuropsychological analysis of MCIMCI received significantly lower scores in all memory tests and one of the language test Boston naming test(P< 0.001 or P< 0.0001.MCI and controls showed no significant differences in the results of another language test verbal fluency test,the two visual-spatial function tests and the two executive function tests(all P>0.05.AD patients had significantly lower scores in all tests(P< 0.05 or P< 0.0001).2. Auditory event-related potentials analysis of MCI(1)In response to non-targetsAERPs in response to non-target frequent tones at the Cz site are presented.P50 amplitude in MCI were significantly increased than controls(P< 0.001).But there were no significant group differences in the latency of P50, amplitudes or latencies of N100 and P200(all P>0.05).AD had significant lower amplitude than MCI(P< 0.05).(2)In response to target infrequent tonesMCI showed a significantly prolonged P300 latency and lowered amplitude than controls(all P< 0.05).But there were no significant group differences in the N200 latency and amplitude(all P>0.05.AD had significantly prolonged P300 latency and lowered amplitude than MCI(all P< 0.01).But there were no significant group differences in the N200 latency and amplitude(all P>0.05).3. Metabolite levels and ratios analysis of MCI by 1H-MRSThe levels of metabolites NAA,Cr,Cho and their ratios NAA/Cr, Cho/Cr in left frontal, left temporal and right parietal cortex of controls,MCI and AD patients are shown.Compared with controls, MCI had significant lower NAA and NAA/Cr in the left prefrontal and left temporal cortex(P< 0.05 or P< 0.01), but not in the right parietal cortex.Cr,Cho and Cho/Cr were not significantly different between groups in three brain areas(all P>0.05).AD patients had significantly decreased NAA and NAA/Cr in the left prefrontal,left temporal and right parietal cortex than MCI(all P< 0.0001).But there were no significant group differences of Cr,Cho and Cho/Cr in three brain areas(all P>0.05).4. Relationship between AERPs , 1H-MRS and cognitive tests of MCI P50 amplitude and P300 latency were negatively correlated with all the memory scores of MCI.NAA/Cr in the left prefrontal and left temporal cortex were positively correlated with all the memory scores of MCI.P50 amplitude was negatively correlated with NAA/Cr in the left prefrontal cortex; P300 latency was negatively correlated with NAA/Cr in the left prefrontal and left temporal cortex.The receiver operating characteristic curve showed the largest area under the curve when P50 amplitude, P300 latency, NAA / Cr of left frontal and left temporal cortex were combined in identifying NC and MCI, MCI and AD.CONCLUSIONS1. MCI had obvious disorders of assessment of memory function, had slightly partial damage in assessment of language function.There were no obvious impairment of executive function and visual-spatial function.These results consistent with the cognitive characteristics of MCI.2. MCI had larger P50 amplitude and prolonged P300 latency.AD had more prolonged P300 latency than MCI.P50 amplitude and P300 latency may be used to identify normal, MCI and AD.3. MCI had significantly lower NAA/Cr ratios.AD had lower NAA/Cr ratios and more brain areas affected than MCI.NAA / Cr can be used to identify normal, MCI and AD.4. P50 amplitude ,P300 latency and NAA/Cr were correlated with scores of clinical memory tests of MCI.Combined 1H-MRS and AERPs can improve significantly diagnostic accuracy of MCI.Combined AERPs and 1H-MRS may be used as a means of diagnosis and follow-MCI.P50 amplitude and P300 latency were correlated with NAA/Cr.更多还原