By Michael H. Crawford, MD, Editor

SYNOPSIS: A randomized, prospective, noninferiority study of the safety of two automated CPR devices (LUCAS and AutoPulse) against the standard manual chest compressions in cardiac arrest victims showed that in cases of severe or life-threatening complications, the LUCAS device was noninferior to standard CPR, but more organ damage with the AutoPulse cannot be excluded.

SOURCES: Koster RW, Beenen LF, van der Boom EB, et al. Safety of mechanical chest compression devices AutoPulse and LUCAS in cardiac arrest: A randomized clinical trial for non-inferiority. Eur Heart J 2017;38:3006-3013.

Lamhaut L, Hutin A. Looking at the force beyond the dark side of mechanical message. Eur Heart J 2017;38:3014-3016.

It is difficult for all members of a healthcare team to perform consistent CPR chest compressions. Automated devices have become more popular and have demonstrated improvements in aortic blood pressure and coronary perfusion compared to manual chest compressions. However, no study has shown increased survival with these devices. Further, some observers have suggested using automated devices causes increased chest wall damage.

Investigators from the Netherlands conducted a randomized, controlled trial of two devices widely available in the United States to test the hypothesis that these devices may cause more chest wall or visceral damage than manual compressions. The study population consisted of in-hospital cardiac arrest patients or those arriving in the ED with the need for ongoing CPR. Patients < 18 years of age with traumatic arrest and those treated with a device by emergency medical personnel in the ambulance were excluded. Patients were randomized 1:1:1 to manual compressions, the LUCAS device, or the AutoPulse device. The depth and rate of manual compressions were calibrated by displacement of a Philips HeartStart MRx defibrillator. All arrests were handled for study purposes by a cardiac arrest team that received appropriate training every six months. The team was not blinded to the device used. Either autopsy or postmortem CT scans were used to identify any injuries in those who didn’t survive. The primary endpoint was damage to visceral organs, large blood vessels, and vertebrae. Secondary endpoints included rib or sternal damage. The study was not powered to analyze survival or compare the two devices. The authors designed this study as a noninferiority investigation.

A total of 1,697 patients were randomized, but after applying the exclusion criteria, AutoPulse was employed in 115, LUCAS in 122, and manual compressions in 137. Serious or life-threatening visceral organ damage was seen in 12% of AutoPulse patients, 7% of LUCAS patients, and 6% of control patients, which was not statistically different (P = 0.75). Investigators observed the secondary outcome in 46% of AutoPulse patients, 40% of LUCAS patients, and 41% of control patients (P = 0.82). The death of three patients was clearly related to CPR. Two patients experienced liver rupture with the LUCAS device and exsanguinated, and one patient on AutoPulse experienced a tension pneumothorax and developed an air embolism in the brain. The authors concluded that the LUCAS device does not cause more serious or life-threatening harm than manual compressions, but more damage cannot be excluded with the AutoPulse device.


The LUCAS device did not exceed the 95% confidence intervals for non-inferiority, but the AutoPulse device did, so the authors could not confirm noninferiority for the AutoPulse device. Interestingly, the first AutoPulse trial ended early because of an increasing mortality trend, and previous trials of the LUCAS device have not shown improved outcomes. However, current guidelines recommend the use of these devices during patient transport and during coronary interventions. Thus, it is pertinent to make sure these devices are not harmful. The LUCAS device did not increase serious organ damage, but did lead to more sternal fractures, which are rarely life-threatening. There were two cases in which the use of the LUCAS device seemed to be lethal. This was due to liver rupture, but surely the device could not have been properly placed to cause this complication. The AutoPulse device raises intrathoracic pressure and in one instance resulted in a tension pneumothorax with lethal air embolism to the brain. This cannot be ascribed to improper device use and may be a small but serious risk of the AutoPulse device.

The main strength of this study is that 90% of the patients underwent autopsy, CT scans, or extensive clinical evaluation of complications. However, there are several weaknesses. The patients were randomized before exclusion criteria were applied, which could have introduced biases. Placement of either device took three to five minutes. During this time, there could have been more return of spontaneous circulation. Everyone was subjected to usual chest compressions before the devices were placed, resulting in a device use time of only about 20 minutes. One would expect more time if the device was used in the catheterization laboratory. Ventilation use specifics were not provided, which could be an issue with the AutoPulse device in particular. Finally, the devices were compared to CPR enhanced by measurement devices, which provided instant feedback of rate and depth. The results compared to regular CPR could have been different.

How does this study inform clinical practice? Clearly, the LUCAS device, when deployed properly, is not harmful when compared to optimized CPR. Therefore, using this device in situations in which CPR is challenging (during coronary interventions) or unsafe for the operators (patient transport), as recommended in the guidelines, is reasonable. Other situations may be appropriate, too. When there is a large size difference between the patient and the operator, device use could be more effective. In this study, even with measured feedback, sternal depression was less than recommended with CPR. One of my colleagues, who is an above-average-sized man, hopes that when he arrests, healthcare providers use the LUCAS device rather than relying on a 90-pound intern to resuscitate him.