扩散张量成像预测轻度认知损害向阿尔茨海默病痴呆转化的系统评价和Meta分析

Meta-Analysis of the Predictive Value of Diffusion Tensor Imaging in the Conversion of Patients with Mild Cognitive Impairment to Alzheimer’s Disease

目的系统评价扩散张量成像(DTI)对轻度认知损害(MCI)患者转化为阿尔茨海默病(AD)的预测价值。方法PubMed、Cochrane Library、Embase、Web of Science、Medline、Ovid Journal、中国知网、万方、维普、中国生物医学文献数据库,检索时限为2010年1月1日至2020年8月30日,查找关于DTI对MCI转归预测价值的队列研究及巢式病例对照研究,根据纳入和排除标准筛选符合标准的文献,对其进行数据提取、质量评价和统计分析。用纽卡斯尔-渥太华量表(NOS)进行质量评价,使用Stata 15.0进行数据合并。结果共纳入5篇文献,共包括MCI患者266例,其中,最终转变成AD患者的有106例,仍稳定在MCI阶段的有160例。Meta结果分析显示,DTI对MCI患者向AD患者转化具有中等的敏感度为82.23%(95%CI 73.38%~88.54%,P=0.000),遗漏率仅为17.77%;预测特异度为60.72%(95%CI 40.35%~70.79%,P=0.447),显著性水平不具有统计学意义。其中亚组分析所得结果中,其中左侧钩状束各向异性分数(FA)值的预测敏感度最高,为86.72%(95%CI 74.74%~93.51%,P=0.000),遗漏率仅为13.28%,其预测特异度为26.99%(95%CI 19.19%~36.52%,P=0.000),误判断率高达73.01%。结论基线DTI参数的改变对于MCI患者1~4年内转归的预测敏感度较好,可用于指导临床早期干预,但因其特异度较低,因而需要更多较好的预测指标以降低误判率。

Objective To systematically evaluate the predictive value of diffusion tensor imaging(DTI)in the conversion of patients with mild cognitive impairment(MCI)to Alzheimer’s disease(AD).Methods PubMed,the Cochrane Library,Embase,Web of science,Medline,Ovid Journal,CNKI,Wanfang data,VIP and CBM were used,Cohort studies and nested case control studies from January 1,2010 to August 30,2020 about the predictive value of DTI to the outcome of MCI were searched.The literatures that met the criteria were selected according to the inclusion criteria,and the data was extracted,quality evaluation and statistical analysis were performed.The quality was evaluated with the Newcastle-Ottawa scale(NOS)and the data was merged with Stata 15.0.Results A total of 5 articles were included,including 266 MCI patients,106 of them eventually became AD patients and 160 remained stable in the stage of MCI.The results of Meta analysis showed that DTI had a moderate sensitivity(82.23%,95%CI 0.7338-0.8854,P=0.000)for the conversionfrom MCI to AD patients,the omission rate is 17.77%,the prediction specificity was 60.72%(95%CI 0.4035-0.7079 P=0.447),and there was nostatistically significance in the significance level.Among the results of subgroup analysis,the prediction sensitivity of the FA value of left uncinate fascicle was highest,which was 86.72%(95%CI 0.7474-0.9351,P=0.000),the omission rate wasonly 13.28%,and its prediction specificity was 26.99%(95%CI 0.1919-0.3652 P=0.000),the misjudgment rate was 73.01%.Conclusion The change of baseline DTI parameters was sensitive to the prognosis of MCI patients within 1 to 4 years,and can be used to guide early clinical intervention.More predictors are needed to reduce the misjudgment rate because of DTI’s low specificity.