The use of automatic chest compressions for cardiac arrest began 15 years ago. Since it is available in emergency medicine, it has raised a fascinating debate. This fascination comes from the first animal study in which the use of this device showed a great haemodynamic effect. It was rapidly approved worldwide by the Emergency Medical Systems (EMS) for the quality of the cardiopulmonary resuscitation (CPR) with a reduced EMS team. For the first time, patients could be transported to the hospital while undergoing good quality CPR in safe conditions for the EMS team. For this reason, automated CPR with mechanical chest compressions was generalized in most of the developed EMS. However, in 2014 the negative results of two large randomized studies with the two different major devices available at that time were published: the Autopulse VC (CIRC trial) and the LUCAS VC (LINC trial). These two studies compared survival after CPR using either automatic chest compressions or optimized manual chest compressions. After these results, the value of automatic chest compressions was debated: should we use the devices if there is no increase in survival? Furthermore, these devices are expensive, and can be harmful, as some data suggest that the use of these devices can lead to internal damage. More recently, the last cardiac arrest international guidelines outlined that extracorporeal cardiopulmonary resuscitation (ECPR) or rapid transfer to angiography for CPR should be considered for refractory cardiac arrest. The transfer of a patient undergoing high quality CPR cannot be considered in safe conditions without mechanical devices. Therefore, the safety aspect of the use of mechanical chest compression is a critical issue, for the patient and the team in charge of the patient. With this in mind, in this issue of the journal, Koster et al. tried to answer this question. In a monocentric randomized study they compared the use of the LUCAS and the Autopulse to optimize manual CPR. The main endpoint was ‘serious or lfe-threatening resuscitation-related damage to visceral organs’. The second endpoint was the damage to the chest/ribs. This study can be considered as a safety study. The authors used a smart methodology to evaluate the damage by a combination of computed tomography (CT) autopsy, classical autopsy, and clinical course. In this study, neither of the two devices produced a significant increase of serious or life-threatening resuscitation damage: the main endpoint occurred in 6.4% of patients of the control group, 11.6% for the Autopulse P = 0.15 (compared with the control group), and 7.4% P = 0.75 for the LUCAS group (compared with the control group). For the second endpoint, the results were also negative (P = 0.51 and P = 0.82 compared with the control group for Autopulse and LUCAS, respectively). However, for the authors, ‘To accept the non-inferiority hypothesis, the upper bound of the 95% confidence interval of an observed rate difference of damage between mechanical or manual chest compressiontreated patients should not exceed þ10%’. They used this endpoint to power this study. For this reason, they conclude that there is a difference in the results between the two devices used. The LUCAS is non-inferior in terms of safety compared with manual CPR with this criterion, whereas the Autopulse cannot reach the objective of non-inferiority for this safety issue. This difference can easily be understood by the technique used by these two devices. On the one hand, for the LUCAS device, the objective is to concentrate pressure on the sternum. Surprisingly, with this device, two patients died from liver rupture. The mechanism of this lesion can be attributed to the high pressure transferred to the diaphragm and more probably to mispositioning of the LUCAS device. This calls for better training of the user. In this study, the rescuers were trained every 6 months. On the other hand, the Autopulse increases the pressure in the entire chest, with potentially

中文翻译：

---

超越机械按摩黑暗面的力量

15 年前开始使用自动胸外按压进行心脏骤停。由于它可用于急诊医学，因此引发了一场引人入胜的辩论。这种魅力来自第一个动物研究，其中使用该装置显示出巨大的血液动力学效应。它在全球范围内迅速获得了紧急医疗系统 (EMS) 的心肺复苏 (CPR) 质量认证，并缩减了 EMS 团队。第一次，病人可以被运送到医院，同时为 EMS 团队在安全的条件下接受高质量的心肺复苏术。出于这个原因，在大多数已开发的 EMS 中普遍采用了带有机械胸外按压的自动 CPR。然而，在 2014 年，使用当时可用的两种不同主要设备的两项大型随机研究的负面结果发表了：Autopulse VC（CIRC 试验）和 LUCAS VC（LINC 试验）。这两项研究比较了使用自动胸外按压或优化的手动胸外按压进行 CPR 后的存活率。在这些结果之后，人们对自动胸外按压的价值进行了辩论：如果生存率没有增加，我们应该使用这些设备吗？此外，这些设备价格昂贵，而且可能有害，因为一些数据表明使用这些设备可能会导致内部损坏。最近，最新的心脏骤停国际指南概述了难治性心脏骤停应考虑体外心肺复苏 (ECPR) 或快速转移到血管造影进行 CPR。在没有机械设备的情况下，不能考虑在安全条件下转移接受高质量 CPR 的患者。所以，对于患者和负责患者的团队而言，使用机械胸外按压的安全性是一个关键问题。考虑到这一点，在本期杂志中，Koster 等人。试图回答这个问题。在一项单中心随机研究中，他们比较了使用 LUCAS 和 Autopulse 来优化手动 CPR。主要终点是“内脏器官严重或危及生命的复苏相关损伤”。第二个终点是胸部/肋骨的损伤。本研究可视为安全性研究。作者使用一种智能方法通过计算机断层扫描 (CT) 尸检、经典尸检和临床过程相结合来评估损伤。在这项研究中，两种设备均未显着增加严重或危及生命的复苏损伤：主要终点发生在 6. 对照组 4% 的患者，Autopulse P = 0.15（与对照组相比）为 11.6%，LUCAS 组为 7.4% P = 0.75（与对照组相比）。对于第二个终点，结果也是阴性的（与 Autopulse 和 LUCAS 的对照组相比，P = 0.51 和 P = 0.82，分别）。然而，对于作者来说，“为了接受非劣效性假设，机械或手动胸外按压治疗患者之间观察到的损伤率差异的 95% 置信区间的上限不应超过 +10%”。他们使用这个端点来支持这项研究。出于这个原因，他们得出结论，使用的两种设备之间的结果存在差异。与具有此标准的手动 CPR 相比，LUCAS 在安全性方面不逊色，而 Autopulse 在这个安全问题上无法达到非劣效性的目标。通过这两种设备使用的技术可以很容易地理解这种差异。一方面，对于 LUCAS 装置，目标是将压力集中在胸骨上。令人惊讶的是，使用这种装置，两名患者死于肝破裂。这种损伤的机制可归因于传递到隔膜的高压，更可能是由于 LUCAS 装置的错误定位。这需要对用户进行更好的培训。在这项研究中，救援人员每 6 个月接受一次培训。另一方面，Autopulse 会增加整个胸部的压力，可能 目的是将压力集中在胸骨上。令人惊讶的是，使用这种装置，两名患者死于肝破裂。这种损伤的机制可归因于传递到隔膜的高压，更可能是由于 LUCAS 装置的错误定位。这需要对用户进行更好的培训。在这项研究中，救援人员每 6 个月接受一次培训。另一方面，Autopulse 会增加整个胸部的压力，可能 目的是将压力集中在胸骨上。令人惊讶的是，使用这种装置，两名患者死于肝破裂。这种损伤的机制可归因于传递到隔膜的高压，更可能是由于 LUCAS 装置的错误定位。这需要对用户进行更好的培训。在这项研究中，救援人员每 6 个月接受一次培训。另一方面，Autopulse 会增加整个胸部的压力，可能