胆碱酯酶抑制剂改善轻中度阿尔茨海默病患者认知疗效的Meta分析

Effect of Cholinesterase Inhibitorson on Cognition of Patients with Mild to Moderate Alzheimer's Disease:a Meta-analysis

目的探讨4种胆碱酯酶抑制剂(多奈哌齐、加兰他敏、石杉碱甲、卡巴拉汀)治疗轻中度阿尔茨海默病(AD)患者的疗效。方法计算机检索Pub Med、EMBase、Cochrane Library、Web of Science、中国知网、万方数据知识服务平台、维普网、中国生物医学网,检索时限为1994年1月—2016年3月,纳入关于胆碱酯酶抑制剂治疗轻中度AD患者的随机双盲安慰剂对照试验。对纳入文献进行数据提取和质量评价,提取包括第一作者、发表时间、药物、日给药剂量、性别、年龄、观察时间、基线简易精神状态评价量表(MMSE)评分、结局指标等资料。采用RevMan 5.3软件分析多奈哌齐、加兰他敏、石杉碱甲、卡巴拉汀改善轻中度AD患者阿尔茨海默病评定量表-认知项目(ADAS-cog)和MMSE评分的效果。结果共纳入30篇文献,其中多奈哌齐15篇,加兰他敏7篇,石杉碱甲4篇,卡巴拉汀4篇。Meta分析结果显示,多奈哌齐5 mg组、多奈哌齐10 mg组、加兰他敏24 mg组、加兰他敏32 mg组、卡巴拉汀12 mg组ADAS-cog评分均低于安慰剂组[加权均数差(WMD)=-2.13,95%CI(-2.69,-1.57),P<0.001;WMD=-2.03,95%CI(-2.80,-1.25),P<0.001;WMD=-3.05,95%CI(-3.54,-2.55),P<0.001;WMD=-2.64,95%CI(-3.55,-1.73),P<0.001;WMD=-2.17,95%CI(-3.30,-1.04),P<0.001]。石杉碱甲400μg组和安慰剂组ADAS-cog评分比较,差异无统计学意义[WMD=-2.95,95%CI(-8.19,2.29),P=0.270]。多奈哌齐5 mg组、多奈哌齐10 mg组、石杉碱甲400μg组、卡巴他汀12 mg组MMSE评分均高于安慰剂组[WMD=1.07,95%CI(0.55,1.59),P<0.001;WMD=1.20,95%CI(0.91,1.50),P<0.001;WMD=2.47,95%CI(1.15,3.78),P<0.001;WMD=1.23,95%CI(0.35,2.11),P=0.006]。Begg秩相关检验、Egger线性回归分析,ADAS-cog纳入文献P值分别为0.511、0.755;MMSE纳入文献P值分别为0.441、0.212,未观察到发表偏倚。结论除石杉碱甲外,其他胆碱酯酶抑制剂治疗轻中度AD患者认知功能减退效果显著。石杉碱甲治疗效果仍有待纳入高质量文献进一步论证。

Objective To discuss the effect of cholinesterase inhibitors （ donepezil , galanthamine , huperzine A , rivastigmine） on the treatment of patients with mild to moderate Alzheimer＇s Disease （ AD） . Methods From January 1994 to March 2016 , PubMed , EMBase , Cochrane Library , Web of Science , CNKI , Wanfang Data Knowledge Srvice Platform , VIP Network , and China Biomedical Network was searched by computer retrieval to enroll randomized , double - blind , placebo - controlled trials of cholinesterase inhibitors treating patients with mild to moderate AD. Data extraction and quality evaluation wereconducted to the included literatures. The data including the first author, the time of publication, medicine, one - day dose , gender , age , the time of observation , the baseline MMSEand the outcome indicators were extracted. RevMan 5. 3 software was used to analyze the effect of donepezil , galanthamine , huperzine A , and rivvstigmine on improving the scores of Alzheimer＇sDisease Assessment Scale - cognitive Subscale （ADAS-cog） and mini - mental state examination （ MMSE） among patients with AD. Results We enrolled 30 literatures totally , of which 15 on donepezil , 7 on galanthamine , 4 on huperzine A and 4 on rivastigmine. Meta - analysis showed that the ADAS-cog scores of the group of 5 mg donepezil , group of 10 mg donepezil , group of 24 mg galanthamine , group of 32 mg galantamine and group of 12 mg rivastigmine were lower than those of the placebo group [WMD= -2. 1 3 , 95% Cl （ - 2. 69 , - 1. 5 7 ） , P 〈0. 0 01; WMD = - 2. 03 , 95% Cl （ - 2 . 80 , - 1 . 2 5 ） , P 〈0 . 0 01;WMD 3 . 05 , 95% Cl （ - 3 . 54 , - 2. 55 ） , P 〈 0 . 001 ; WMD = - 2. 64 , 95% Cl （ - 3 . 55 , - 1. 73 ） , P 〈 0 . 001 ;WMD = -2 . 1 7 , 95% C/ （ -3. 30 , - 1. 04） , P 〈0 . 0 0 1 〕. There was no significant difference in the ADAS-cog score betweengroup of 400 ^g huperzine A and the placebo group [ WMD = - 2. 95 , 95% Cl （ - 8. 19 , 2. 29） , P =0. 27 0 ]. The MMSEscores of the group of 5 mg donepezil , group of 10 mg donepezil , group of 400 jxg huperzine A and group of 12 mg rivastigminewere higher than those of the placebo group [ WMD = 1. 07 , 95% Cl （0. 55 , 1. 59 ） , P 〈 0 . 0 1 ; WMD = 1. 20 , 95% Cl （0.91,1.50）,P〈0. 0 01; WMD=2.47, 95%C/ （ 1 .1 5,3 .7 8 ）,P 〈 0 .0 0 1; WMD = 1 .2 3 , 95%C/ （ 0 .3 5,2 .1 1 ） , P = 0.006]. Begg rank correlation test and Egger linear regression test showed that P value of the literatures included in scale was 0.511 and 0. 755 respectively; P value of the literatures included in MMSE scale was 0. 441 and 0. 212 respectively. No publication bias was observed. Conclusion Except for huperzine A, other cholinesterase inhibitors have significant effect on the treatment of cognitive decline among patients with mild to moderate AD. The treatment effect of huperzidemonstrated by including high quality literatures.