【文章】疑似心肌梗死患者0/1小时与0/3小时路径的性能比较：随机对照试验的系统综述与个体参与者数据荟萃分析
完整摘要 · 中文译文
背景：对于疑似心肌梗死患者，当前指南推荐采用0/1小时路径（于0小时和1小时采血）进行早期风险分层。然而，各单项试验采用了不同的安全性结局，疗效结局尚不确定或未予评估。我们对所有随机对照试验进行了系统评价和个体参与者数据荟萃分析，以综合全部证据、评估试验间的异质性，并采用跨试验统一的定义来评价临床相关结局。 方法：在本系统评价和个体参与者数据荟萃分析中，我们检索了PubMed、Cochrane对照试验中心注册库、Scopus和国际临床试验注册平台，检索时间从建库至2026年7月6日，无语言限制，纳入比较0/1小时与0/3小时路径（0小时和3小时采血，即标准诊疗）安全性和有效性的随机试验。获取个体参与者数据并统一结局定义。主要分析采用两阶段随机效应荟萃分析。在收到个体参与者数据后、但在数据统一或分析之前，主要安全性结局由最初规定的全因死亡或任何新发心肌梗死复合终点，细化为30天内全因死亡或新发1型心肌梗死的复合终点。主要有效性结局为急诊科停留时间和从急诊科直接出院。本荟萃分析已在PROSPERO注册（CRD420261383452）。 研究结果：我们的系统综述识别出五项随机试验，所有这些试验均提供了个体受试者数据。其中，主要安全性结局的定义各不相同，且四项试验未报告整个分析人群的急诊科停留时间。我们纳入了110 933例患者就诊（55 489例采用0/1小时路径管理，55 444例采用0/3小时路径管理）。30天内，0/1小时路径组有446例患者就诊（0.8%）发生全因死亡或新发1型心肌梗死，而0/3小时路径组为562例患者就诊（1.0%）（随机效应合并比值比[OR] 0.83 [95% CI 0.45-1.53]；I2=62.9%；τ2=0.143）。0/1小时路径组的急诊科停留时间为314分钟（IQR 211-481），而0/3小时路径组为308分钟（209-453）（随机效应合并几何均值比1.03 [95% CI 0.85-1.25]；I2=99.2%；τ2=0.024）。0/1小时路径组有37 190例患者就诊（67.0%）从急诊科直接离院，而0/3小时路径组为36 485例患者就诊（65.8%）（随机效应合并OR 1.31 [95% CI 0.79-2.16]；I2=97.8%；τ2=0.156）。 解释：在现有的随机试验中，两种路径的短期不良事件发生率均较低，但汇总的安全性估计仍不精确。平均而言，更早复查心肌肌钙蛋白并未缩短急诊科停留时间，也未明显提高直接出院率，且研究间存在显著异质性，提示治疗效果可能不会在不同医疗系统中一致转化。对于考虑从0/3小时路径转为0/1小时路径的医院，实施决策应综合考虑更广泛的诊断和组织管理因素。 资助：瑞士国家科学基金会、瑞士心脏基金会、巴塞尔大学医院、巴塞尔大学和巴塞尔心血管研究基金会。
完整摘要 · 英文原文
BACKGROUND: In patients with suspected myocardial infarction, the 0/1 h pathway (blood draws at 0 h and 1 h) is recommended in current guidelines for early risk stratification. However, individual trials used different safety outcomes, and efficacy outcomes were inconclusive or not assessed. We conducted a systematic review and individual participant data meta-analysis of all randomised controlled trials to synthesise the totality of the evidence, assess heterogeneity across trials, and evaluate clinically relevant outcomes using harmonised definitions across trials. METHODS: In this systematic review and individual participant data meta-analysis, we searched PubMed, Cochrane Central Register of Controlled Trials, Scopus, and the International Clinical Trials Registry Platform from inception to July 6, 2026, without language restrictions, for randomised trials comparing the safety and efficacy of 0/1 h versus 0/3 h pathways (blood draws at 0 h and 3 h, ie, standard care). Individual participant data were obtained and outcome definitions were harmonised. The primary analysis used a two-stage random-effects meta-analysis. After receipt of the individual participant data but before data harmonisation or analysis, the primary safety outcome was refined to a composite of all-cause death or new type 1 myocardial infarction within 30 days, from the originally specified composite of all-cause death or any new myocardial infarction. The primary efficacy outcomes were emergency department length of stay and direct discharge from the emergency department. This meta-analysis is registered with PROSPERO (CRD420261383452). FINDINGS: Our systematic review identified five randomised trials, all of which provided individual participant data. Among these, the definition of the primary safety outcome varied, and in four trials emergency department length of stay was not reported for the whole analysis population. We included 110 933 patient presentations (55 489 managed with the 0/1 h pathway and 55 444 managed with the 0/3 h pathway). Within 30 days, all-cause death or new type 1 myocardial infarction occurred in 446 patient presentations (0·8%) with the 0/1 h pathway versus 562 patient presentations (1·0%) with the 0/3 h pathway (random-effects pooled odds ratio [OR] 0·83 [95% CI 0·45-1·53]; I2=62·9%; τ2=0·143). Emergency department length of stay was 314 min (IQR 211-481) in the 0/1 h pathway group versus 308 min (209-453) in the 0/3 h pathway group (random-effects pooled geometric mean ratio 1·03 [95% CI 0·85-1·25]; I2=99·2%; τ2=0·024). Direct discharge from the emergency department occurred in 37 190 patient presentations (67·0%) with the 0/1 h pathway versus 36 485 patient presentations (65·8%) with the 0/3 h pathway (random-effects pooled OR 1·31 [95% CI 0·79-2·16]; I2=97·8%; τ2=0·156). INTERPRETATION: Across the available randomised trials, short-term adverse event rates were low with both pathways, but the pooled safety estimate remained imprecise. On average, earlier cardiac troponin retesting did not reduce emergency department length of stay or clearly increase direct discharge, and substantial between-study heterogeneity indicates that treatment effects might not translate uniformly across health-care systems. For hospitals considering transitioning from a 0/3 h to a 0/1 h pathway, implementation decisions should take broader diagnostic and organisational considerations into account. FUNDING: Swiss National Science Foundation, Swiss Heart Foundation, University Hospital Basel, University of Basel, and Foundation for Cardiovascular Research Basel.