儿童抽动障碍安慰剂效应的系统评价和Meta分析

Systematic review and Meta-analysis of placebo effect in children with tic disorders

目的 系统评价儿童抽动障碍随机对照试验的安慰剂效应,并探索其影响因素。方法 检索中国学术期刊全文数据库(CNKI)、维普中文期刊全文数据库(VIP)、万方数据库、中国生物医学文献数据库(CBM)、PubMed、Cochrane图书馆、Embase中关于儿童抽动障碍随机对照试验(RCT)中英文文献,检索时间均从各数据库创建至2021年11月。使用R软件对耶鲁综合抽动严重程度量表抽动总分(YGTSS-TTS)、耶鲁综合抽动严重程度量表(YGTSS)总分及基于其定义的有效率进行定量资料或定性资料的Meta分析,探索影响安慰剂效应的因素,并对影响因素进行亚组分析。结果 共纳入11项RCTs,安慰剂组总样本量606例。对于YGTSS-TTS变化均值及其有效率,Meta分析结果分别为7.2[95%CI(6.23,8.18)]和36%[95%CI(28%,45%)];根据有效率的不同定义阈值(减分率≥25%、≥30%和≥50%)做亚组分析,其合并效应值分别为48%[95%CI(43%,53%)]、42%[95%CI(35%,49%)]和20%[95%CI(12%,29%)]。对于YGTSS总分及其有效率,Meta分析结果分别为13.89[95%CI(12.4,15.38)]和32%[95%CI(24%,41%)];根据有效率的不同定义阈值(减分率≥30%和≥50%)做亚组分析,其合并效应值分别为40%[95%CI(34%,47%)]和22%[95%CI(14%,31%)]。对各结局指标分别进行了Meta回归分析,结果提示疾病严重程度和疗程,与安慰剂效应呈正相关。结论 儿童抽动障碍具有一定的安慰剂治疗效应,其安慰剂效应可能与入组严重程度、疗程等因素有关,其效应值随着定义阈值的升高而降低。这些结果,可以为以安慰剂为对照的临床试验设计和样本量估算,提供借鉴与参考。

Objective To systematically evaluate the placebo effect of randomized controlled trials of tic disorders in children and explore its influencing factors. Methods The Chinese and English literatures on children’s tic disorder randomized controlled trial(RCT) in CNKI, VIP, Wanfang data, CBM, PubMed, Cochrane Library and Embase were searched from the establishment of each database to November 2021. Using R software, the total tic scores of Yale global tic severity scale(YGTSS-TTS), the total scores of Yale global tic severity scale(YGTSS) and the effective rate based on their definitions were analyzed by Meta-analysis of quantitative or qualitative data, to explore the factors affecting the placebo effect, and the influencing factors were analyzed by subgroup. Result A total of 11 RCTs were included, with a total sample size of 606 cases in the placebo group. For the mean change and effective rate of YGTSS-TTS, the results of Meta-analysis were 7.2 [95%CI(6.23, 8.18)] and 36% [95%CI(28%, 45%)]respectively;subgroup analysis was performed according to different defined thresholds of efficiency(score reduction rate ≥ 25%, ≥30% and ≥ 50%),the combined effect values were 48% [95%CI(43%, 53%)], 42% [95%CI(35%, 49%)] and 20% [95%CI(12%,29%)] respectively. For the mean change and effective rate of total YGTSS score, the results of Meta-analysis were 13.89 [95%CI(12.4, 15.38)] and 32%[95%CI(24%, 41%)] respectively;subgroup analysis was performed according to different defined thresholds of efficiency(score reduction rate ≥ 30% and ≥ 50%), the combined effect values were 40% [95%CI(34%, 47%)] and 22% [95%CI(14%, 31%)]. Meta regression analysis was carried out for each outcome index and the results showed that the disease severity and course of treatment were positively correlated with the placebo effect. Conclusion Children with ic disorders have a certain placebo effect. The placebo effect may be related to factors such as enrollment severity and course of treatment. The effect size decreases with the increase of the defined threshold. These results can provide reference for placebo-controlled clinical trial design and sample size estimation.