脑栓塞保护装置在TAVI中减少致残性卒中的贝叶斯荟萃分析
该系统综述与贝叶斯荟萃分析纳入7项随机试验(n=3996),评估经导管主动脉瓣植入术(TAVI)中脑栓塞保护(CEP)对致残性卒中的影响。主要终点致残性卒中的相对风险(RR)为0.56(95%CrI 0.28-1.19),绝对风险差(ARD)0.56%,需治疗人数(NNT)179;任何获益的后验概率为94.4%,但临床相关效应的概率仅为0-0.1%。结论认为CEP获益可能性高但临床相关性不大。
要点
- CEP降低致残性卒中的RR为0.56,但95%可信区间跨1。
- 绝对风险差0.56%,NNT 179,未达预设的临床重要差异阈值(ARD 1.1%,NNT 91)。
- 临床相关效应的后验概率极低(0-0.1%),结果在敏感性分析中稳健。
核心结论
- CEP治疗效应有益的概率高,但不太可能具有临床相关性。
研究设计与样本
- 系统检索三个电子数据库,纳入将TAVI患者随机分配至CEP组或标准治疗组的随机试验。
- 共纳入7项随机研究进行荟萃分析,总样本量n=3996(CEP组n=2126,对照组n=1870)。
- 主要分析采用 vague prior 的贝叶斯荟萃分析,次要分析采用两个基于文献的 informed priors。
主要发现
- 在 vague prior 下,CEP用于致残性卒中的估计中位RR为0.56(95%CrI 0.28至1.19),推导ARD为0.56%,NNT为179,I²=0%。
- 任何获益的后验概率为94.4%,但临床相关效应的概率在 vague 和 informed priors 下均为0-0.1%。
- 结果在多项敏感性分析中稳健。
关键数值与推荐
- 最小临床重要差异设定为ARD 1.1%(NNT 91),基于专家共识。
临床怎么用
- CEP在TAVI中虽可能降低致残性卒中风险,但绝对获益幅度小,临床相关性有限。
- 未来试验应聚焦于识别基线卒中风险增高的TAVI患者以及开发新一代装置。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVES: The use of cerebral embolic protection (CEP) during transcatheter aortic valve implantation (TAVI) has been studied in several randomised trials. We aimed to perform a systematic review and Bayesian meta-analysis of randomised CEP trials, focusing on a clinically relevant reduction in disabling stroke. METHODS: A systematic search was applied to three electronic databases, including trials that randomised TAVI patients to CEP versus standard treatment. The primary outcome was the risk of disabling stroke. Outcomes were presented as relative risk (RR), absolute risk differences (ARDs), numbers needed to treat (NNTs) and the 95% credible intervals (CrIs). The minimal clinically important difference was determined at 1.1% ARD, per expert consensus (NNT 91). The principal Bayesian meta-analysis was performed under a vague prior, and secondary analyses were performed under two informed literature-based priors. RESULTS: Seven randomised studies were included for meta-analysis (n=3996: CEP n=2126, control n=1870). Under a vague prior, the estimated median RR of CEP use for disabling stroke was 0.56 (95% CrI 0.28 to 1.19, derived ARD 0.56% and NNT 179, I2=0%). Although the estimated posterior probability of any benefit was 94.4%, the probability of a clinically relevant effect was 0-0.1% under the vague and informed literature-based priors. Results were robust across multiple sensitivity analyses. CONCLUSION: There is a high probability of a beneficial CEP treatment effect, but this is unlikely to be clinically relevant. These findings suggest that future trials should focus on identifying TAVI patients with an increased baseline risk of stroke, and on the development of new generation devices. PROSPERO REGISTRATION NUMBER: CRD42023407006.
引用
- Heuts S, Gabrio A, Veenstra L, et al. Stroke reduction by cerebral embolic protection devices in transcatheter aortic valve implantation: a systematic review and Bayesian meta-analysis. Heart. 2024. PMID: 37996242. doi:10.1136/heartjnl-2023-323359
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。