Joseph Tennyson, MD
Mechanical cardiopulmonary resuscitation (CPR) devices have seen increased use in recent years. With the emphasis placed on the performance of uninterrupted high-quality CPR, it naturally follows that a mechanical device which is not subject to variation and fatigue from which human rescuers suffer would improve on manual CPR and increase survival. With this reasoning, their use has blossomed in the EMS Community.
As with any innovation, we must consider the potential negative effects of its implementation. The most obvious amongst these potential negatives is the cost of the devices. The two most common mechanical CPR devices in use are the piston-driven Lucas 2 device manufactured by Physio-Control and the load distributing band (LDB) Autopulse device manufactured by the Zoll Corporation. The Physio-Control product sells for approximately $15,000 per unit.1 The Zoll product can sell for as much as $20,000 per unit.2 While these numbers may seem approachable when taken by themselves, they do not include the costly replacement supplies required for each of the products. Additionally, when you consider the size of some EMS agencies, equipping each ambulance in a larger ambulance service becomes exceedingly costly. When considering the devices for in-hospital use, the question becomes whether they should be equipped on each code cart within the hospital. Again, depending on the size of the facility, this cost can become astronomical.
Another concern is the mechanical force applied to the patient’s body. These mechanical CPR devices have been associated with a significant rate of injuries.3-7 Several postmortem studies have revealed increases in specific injury patterns. One postmortem study reviewing patients who had been treated with a load distributing band (LDB) style mechanical device evaluated patients using postmortem computed tomography (CT). This study found that when compared to manual CPR the odds ratio (OR) for rib fracture of 30:1 for patients who received mechanical CPR as compared to manual CPR.4 The increased risk is not limited to the LDB type of device.
Two other postmortem studies found an increase in rib fractures in patients having been treated with the piston driven mechanical CPR device.4,6 These results are not fully conclusive, however. Other studies have shown the absence of this increase in rib fractures. One study by Baumeister of postmortem patients who had received piston driven mechanical CPR showed no difference in rib fractures. This study show a significant increase in substernal hematomas noted on postmortem computed tomography.3 Other injury types seen with increased frequency in these postmortem studies include thoracic and abdominal aortic rupture, pericardial bleeding,6 liver parenchymal injuries, and vertebral body fractures.6,7 Additionally several case reports have been published which demonstrate other injury patterns found after mechanical CPR. These include reports of fatal internal hemorrhage secondary to liver parenchymal injury,8 hollow organ rupture,9 and rupture of both liver and spleen in a single patient.10
An increase in injury patterns may be acceptable if the result is an increase in survival. Our ultimate goal is to benefit the patient and if the cost of a life saved is a few extra rib fractures, many would believe that a reasonable trade-off. Unfortunately, these devices have yet to provide the clear benefit we seek. Single center observational study published by Zeiner in 2015 found mechanical CPR to be associated with an increase in mortality, a decrease in survival to discharge, and a decrease in surviving patients achieving favorable cerebral performance categories (CPC) score of one or two.11 Randomized controlled trials have also failed to demonstrate a benefit. Both the PARAMEDIC trial and the CIRC trial demonstrated statistical equivalence between mechanical and manual CPR.12,13 Two separate reviews of statewide EMS databases have demonstrated a lack of benefit as well. Buckler, with data from Pennsylvania which was published in abstract form, demonstrated an OR for survival of 0.75% (95% CI 0.59-0.95) and for good neurologic recovery also of 0.75 (95% CI 0.57-0.98).14 Youngquist et al demonstrated an even more dramatic decrease in neurologically intact survival in their data from Utah with a propensity score for neurologically intact survival of 0.41 (95% CI 0.24-0.70).15 Finally a recently published Meta-analysis by Bonnes reviewed the above randomized trials as well as some others. When evaluating only the high-quality evidence, this analysis found a lack of benefit for return of spontaneous circulation, survival to admission, survival to discharge and favorable neurological outcome.16
Finally, a review of the published Emergency Cardiac Care Guidelines reveals that the lack of evidence for the benefit of these devices has been recognized by the guideline authors. The 2015 American Heart Association Emergency Cardiovascular Care Guidelines state that the evidence “does not demonstrate a benefit” for these devices and that manual compressions “remain the standard of care.”17 The same recommendation was published in the Australian and New Zealand Committee on Resuscitation.18 In evaluating any therapy, one must weigh the benefits and risks. These devices demonstrate an increase in the risk of harm and have failed to show the anticipated benefits.