Assessing new treatments for central sleep Apnoea in heart failure: sample size considerations for reliable detection of safety signals

评估治疗心力衰竭合并中枢性睡眠呼吸暂停的新疗法:可靠检测安全性信号的样本量考量

阅读:2

Abstract

PURPOSE: Central sleep apnoea (CSA) and Cheyne-Stokes respiration (CSR) are prevalent in patients with heart failure with reduced ejection fraction (HFrEF) and are associated with increased mortality. Previous trials evaluating adaptive servo-ventilation (ASV) for CSA/CSR in HFrEF have yielded conflicting safety outcomes. This analysis aims to determine the sample size and number of events required to reliably detect safety signals, specifically all-cause mortality, in future trials assessing new treatments for CSA/CSR in HFrEF patients. METHODS: Utilizing data from the SERVE-HF trial, we conducted event-driven, non-inferiority sample size calculations focusing on all-cause mortality as the primary endpoint. The analysis employed a one-sided alpha of 0.025 and 80% power, considering hazard ratios (HRs) ranging from 1.15 to 1.35, in line with FDA guidelines for cardiovascular safety assessments. Calculations assumed a control group event rate of 0.093 events/year, a 5-year accrual period, and a minimum follow-up of 3 years. RESULTS: To detect a non-inferiority margin with an HR of 1.35, a trial would require at least 851 patients and 349 events. More stringent margins necessitate larger sample sizes; for an HR of 1.15, 3,924 patients and 1,607 events are needed. These findings underscore the importance of adequately powered studies to detect potential safety concerns in this patient population. CONCLUSION: Future clinical trials investigating treatments for CSA/CSR in HFrEF patients must incorporate sufficient sample sizes and event counts to ensure reliable detection of safety signals, particularly concerning all-cause mortality. This approach is critical for accurately assessing the risk-benefit profile of new therapeutic interventions.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。