高流量鼻导管与无创通气在急性心力衰竭相关呼吸衰竭中的比较：一项系统综述和荟萃分析。
完整摘要 · 中文译文
背景：无创通气（NIV）被推荐为急性心力衰竭（AHF）相关呼吸衰竭的一线呼吸支持。然而，其耐受性常常受限。经鼻高流量湿化氧疗（HFNC）已成为一种潜在的替代方案，但比较性证据仍不一致。本研究旨在评估HFNC与NIV在AHF相关呼吸衰竭患者中的有效性和安全性。 方法：我们按照PRISMA指南进行了一项系统评价和荟萃分析。我们检索了四个主要电子数据库，检索时间从建库至2026年4月。我们纳入了随机对照试验（RCT）和观察性研究，这些研究纳入的成年患者患有继发于AHF或急性心源性肺水肿的呼吸衰竭，并比较了HFNC与NIV。主要结局为复合治疗失败。次要结局包括气管插管、短期死亡率、动脉血气（ABG）参数、呼吸频率、呼吸困难评分和住院时间。使用随机效应模型。采用推荐分级评估、制定与评价（GRADE）方法评估证据确定性。 结果：共纳入十项研究（n = 1,282例患者），包括五项随机对照试验和五项观察性研究。在复合治疗失败结局（RR = 1.22，95% CI：0.81-1.84）、气管插管（RR = 0.86，95% CI：0.58-1.28）或短期死亡率（RR = 0.99，95% CI：0.69-1.42）方面，HFNC与NIV相比差异无统计学显著性。针对治疗失败按设计分层的估计值... 失败在两项随机对照试验（RR = 1.00, 95% CI: 0.61-1.63）和观察性研究（RR = 1.44, 95% CI: 0.77-2.69）中也均未跨越无效应线。在动脉血气参数、呼吸频率、呼吸困难评分或住院时长方面未观察到统计学显著差异。GRADE评估显示，随机对照试验比较主要为中等确定性证据，而PaCO2和pH值变化为高确定性证据。 结论：现有证据未发现高流量鼻导管（HFNC）与无创通气（NIV）在急性心力衰竭相关呼吸衰竭患者临床疗效方面存在统计学显著差异。对于不耐受NIV的特定患者，可在密切监测并能及时升级至NIV的前提下考虑使用HFNC。尽管随机对照试验比较的证据确定性总体为中等，但仍需开展更多效能充分的多中心随机对照试验，以更好地明确患者选择标准及具有临床意义的差异。 系统评价注册：www.crd.york.ac.uk/PROSPERO/view/CRD420261330512，注册号CRD420261330512。
完整摘要 · 英文原文
BACKGROUND: Noninvasive ventilation (NIV) is the recommended first-line respiratory support for acute heart failure (AHF)-related respiratory failure. However, its tolerance is often limited. High-flow nasal cannula (HFNC) has emerged as a potential alternative, but comparative evidence remains inconsistent. This study aimed to evaluate the efficacy and safety of HFNC versus NIV in patients with AHF-related respiratory failure. METHODS: We conducted a systematic review and meta-analysis following PRISMA guidelines. We searched four major electronic databases from inception to April 2026. We included randomized controlled trials (RCTs) and observational studies enrolling adult patients with respiratory failure secondary to AHF or acute cardiogenic pulmonary edema, comparing HFNC and NIV. The primary outcome was composite treatment failure. Secondary outcomes included endotracheal intubation, short-term mortality, arterial blood gas (ABG) parameters, respiratory rate, dyspnea scores, and hospital length of stay. A random-effects model was used. The certainty of evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. RESULTS: Ten studies (n = 1,282 patients) were included, comprising five RCTs and five observational studies. HFNC showed no statistically significant difference compared with NIV in the composite outcome of treatment failure (RR = 1.22, 95% CI: 0.81-1.84), endotracheal intubation (RR = 0.86, 95% CI: 0.58-1.28), or short-term mortality (RR = 0.99, 95% CI: 0.69-1.42). Design-stratified estimates for treatment failure also crossed the line of no effect in both RCTs (RR = 1.00, 95% CI: 0.61-1.63) and observational studies (RR = 1.44, 95% CI: 0.77-2.69). No statistically significant differences were observed in ABG parameters, respiratory rate, dyspnea scores, or hospital length of stay. The GRADE assessment showed predominantly moderate certainty evidence for RCT comparisons, with high-certainty evidence for changes in PaCO2 and pH. CONCLUSION: The available evidence did not find a statistically significant difference between HFNC and NIV in terms of clinical efficacy in patients with AHF-related respiratory failure. HFNC may be considered in selected patients who do not tolerate NIV, provided that close monitoring and prompt access to NIV escalation are available. Although the certainty of evidence was generally moderate for RCTs comparisons, further adequately powered multicenter RCTs are needed to better define patient selection and clinically important differences. SYSTEMATIC REVIEW REGISTRATION: www.crd.york.ac.uk/PROSPERO/view/CRD420261330512, identifier CRD420261330512.