Department of Cardiovascular Disease, University of Rome "Sapienza", Rome, Italy.
Eur Rev Med Pharmacol Sci. 2018 Apr;22(8):2405-2414. doi: 10.26355/eurrev_201804_14833.
Intra-aortic balloon pump (IABP) is the device most commonly investigated in patients with cardiogenic shock (CS) complicating acute myocardial infarction (AMI). Recently meta-analyses on this topic showed opposite results: some complied with the actual guideline recommendations, while others did not, due to the presence of bias. We investigated the reasons for the discrepancy among meta-analyses and strategies employed to avoid the potential source of bias.
Scientific databases were searched for meta-analyses of IABP support in AMI complicated by CS. The presence of clinical diversity, methodological diversity and statistical heterogeneity were analyzed. When we found clinical or methodological diversity, we reanalyzed the data by comparing the patients selected for homogeneous groups. When the fixed effect model was employed despite the presence of statistical heterogeneity, the meta-analysis was repeated adopting the random effect model, with the same estimator used in the original meta-analysis.
Twelve meta-analysis were selected. Six meta-analyses of randomized controlled trials (RCTs) were inconclusive because underpowered to detect the IABP effect. Five included RCTs and observational studies (Obs) and one only Obs. Some meta-analyses on RCTs and Obs had biased results due to presence of clinical and/or methodological diversity. The reanalysis of data reallocated for homogeneous groups was no more in contrast with guidelines recommendations.
Meta-analyses performed without controlling for clinical and/or methodological diversity, represent a confounding message against a good clinical practice. The reanalysis of data demonstrates the validity of the current guidelines recommendations in addressing clinical decision making in providing IABP support in AMI complicated by CS.
主动脉内球囊反搏（IABP）是治疗急性心肌梗死（AMI）合并心源性休克（CS）患者最常用的设备。最近关于这个主题的荟萃分析结果相反：一些符合实际指南建议，而另一些则由于存在偏倚而不符合。我们调查了荟萃分析之间差异的原因，并研究了避免潜在偏倚来源的策略。
在 AMI 合并 CS 的 IABP 支持的荟萃分析中搜索科学数据库。分析了临床多样性、方法学多样性和统计学异质性的存在。当我们发现临床或方法学多样性时，我们通过比较选择同质组的患者来重新分析数据。当尽管存在统计学异质性但仍采用固定效应模型时，我们采用随机效应模型重复元分析，使用原始元分析中相同的估计器。
选择了 12 项荟萃分析。6 项随机对照试验（RCT）的荟萃分析因缺乏检测 IABP 效果的能力而没有定论。其中 5 项纳入 RCT 和观察性研究（Obs），1 项仅纳入 Obs。一些关于 RCT 和 Obs 的荟萃分析由于存在临床和/或方法学多样性而产生有偏倚的结果。重新分析为同质组分配的数据与指南建议不再矛盾。
未控制临床和/或方法学多样性的荟萃分析，代表了与良好临床实践相悖的混淆信息。对数据的重新分析证明了当前指南建议在处理临床决策方面的有效性，即在 AMI 合并 CS 中提供 IABP 支持。
