摘要：

目的 系统评价尼莫地平治疗血管性痴呆的有效性和安全性。方法 分别以“nimodipine AND vascular dementia”和“尼莫地平AND 血管性痴呆”为检索式，计算机检索美国国立医学图书馆生物医学信息检索系统、英国Cochrane图书馆、荷兰医学文摘、科学引文索引，以及中国知识基础设施工程、维普中文科技期刊数据库、万方数据库，人工检索部分期刊文献及未发表的研究项目和会议资料，Google Scholar 等搜索引擎检索互联网关于尼莫地平单药治疗血管性痴呆的随机对照临床试验（1995 年1 月-2015 年3 月）。Jadad 量表评价文献质量，RevMan 5.3 统计软件行Meta 分析。结果 最终纳入11 篇文献共10 项临床研究（1333 例患者），尼莫地平与安慰剂随机对照临床试验4 项、尼莫地平与盐酸多奈哌齐随机对照临床试验5 项、尼莫地平与甲磺酸双氢麦角毒碱随机对照临床试验1 项，其中仅2 项具体描述随机方法。Meta 分析显示，（1）尼莫地平对血管性痴呆患者精神智力水平的改善优于安慰剂（3 项研究；MD = 0.270，95%CI：0.070 ~ 0.460，P = 0.007）和空白安慰剂（1 项研究；MD = 2.950，95%CI：1.670 ~ 4.200，P = 0.000），但对日常生活活动能力的改善尚无确切疗效［1 项研究（日常生活活动能力量表评分），MD = 5.800，95%CI：2.480 ~ 9.120，P = 0.001；1 项研究（日常生活活动能力指数），MD = -0.040，95%CI：-0.110 ~ 0.030，P = 0.230；1 项研究（工具性日常生活活动能力量表评分），MD = -0.080，95%CI：-0.110 ~ 0.000，P = 0.060］。（2）尼莫地平和盐酸多奈哌齐均可改善血管性痴呆患者精神智力水平和日常生活活动能力，二者疗效无差异［简易智能状态检查量表评分：4 项研究（观察时间12 周），MD = -4.400，95%CI：-4.870 ~ -3.920，P = 0.000；1 项研究（观察时间24 周），MD = -8.800，95%CI：-8.970 ~ -7.430，P = 0.000；日常生活活动能力量表评分：2 项研究，MD = 1.800，95%CI：1.360 ~ 2.230，P = 0.000］。（3）尼莫地平对血管性痴呆患者精神智力水平（1 项研究；MD = 2.170，95%CI：0.890 ~ 3.450，P = 0.001］和日常生活活动能力（1 项研究；MD = -5.160，95%CI：-7.480 ~ -2.840，P = 0.000）的改善作用均优于甲磺酸双氢麦角毒碱。（4）尼莫地平主要不良反应为心脑血管意外、神经精神症状、胃肠道反应、皮肤异常反应和关节水肿等。结论尼莫地平治疗血管性痴呆的疗效可能优于安慰剂和甲磺酸双氢麦角毒碱，而与盐酸多奈哌齐相当。

关键词: 尼莫地平, 痴呆, 血管性, 循证医学, 药物不良反应（非MeSH 词）

Abstract:

Objective  To systematically evaluate the clinical efficacy and safety of nimodipine in treating vascular dementia (VaD).  Methods  Taking "nimodipine AND vascular dementia" as search terms, retrieve in databases such as PubMed, Cochrane Library, EMBASE/SCOPUS, Science Citation Index (SCI), China National Knowledge Infrastructure (CNKI), VIP and Wanfang Data (January 1995-March 2015). Annual searching was applied to retrieve partial periodical literatures and unpublished studies. Google Scholar was used for randomized controlled trials (RCTs) about nimodipine in treating VaD. Jadad scale was used to evaluate the quality of literature, and Meta-analyses were performed by using RevMan 5.3 software.  Results  Eleven literatures met inclusion criteria, including 10 clinical studies (1333 patients). All 10 studies were RCTs, including 4 nimodipine vs placebo, 5 nimodipine vs donepezil and one nimodipne vs hydergine, but only 2 described randomization methods. The results of Meta-analysis showed: nimodipine had better Mini-Mental State Examination (MMSE) score than before treatment and placebo group (3 studies, MD = 0.270, 95%CI: 0.070—0.460, P = 0.007); one study of blank control, MD = 2.950, 95% CI: 1.670—4.200, P = 0.000). Patients treated with nimodipne had no significantly improved Activities of Daily Living (ADL) score than placebo group [one study of ADL, MD = 5.800, 95%CI: 2.480—9.120, P = 0.000; one study of ADL Index, MD = -0.040, 95%CI: -0.110—0.030, P = 0.230; one study of instrumental ADL (IADL), MD = -0.080, 95%CI: -0.110—0.000, P = 0.060]. Both nimodipine and donepezil can improve MMSE and ADL scores, but the efficacy of nimodipine was not superior to donepezil [4 studies of MMSE (12-week observation), MD = -4.400, 95% CI: -4.870— -3.920, P = 0.000; one study of MMSE (24-week observation), MD = -8.800, 95% CI: -8.970— -7.430, P = 0.000; 2 studies of ADL, MD = 1.800, 95% CI: 1.360—2.230, P = 0.000]. Compared with hydergine, nimodipine had better scores in MMSE (MD = 2.170, 95%CI: 0.890—3.450, P = 0.001) and ADL (MD = -5.160, 95%CI: -7.480— -2.840, P = 0.000) in one study. The main side effects of nimodipine were cardio-cerebrovascular diseases, neuropsychiatric symptoms, abnormal skin reaction, gastrointestinal reactions and joint edema, et al.  Conclusions Current study shows that the effect of nimodipine in the treatment of VaD may be superior to placebo and hydergine, but not better than donepezil. The results of systematic review still need more high-quality, multi-center and large-sample RCTs to further prove.

Key words: