Skip to content

Mechanical CPR and the LINC trial



ExpertPeerReviewStamp2x200The first time I saw the Thumper performing CPR on a patient I thought “well, that makes sense.” Since then we have seen other devices, most notably the Zoll AutoPulse and the Physio-Control LUCAS. It was disappointing to many in 2005 when the AutoPulse trial was halted early due to harm. 1 Although four-hour survival was similar between groups, the hospital discharge survival rate in the manual CPR group was 9.9% compared to 5.8% in the mechanical CPR group. Many hypotheses were proposed to explain the results, which included Hawthorne effect, prolonged device deployment time, and enrollment bias. Last month, the results of the LUCAS in Cardiac Arrest (LINC) trial were published in JAMA, breathing new life into the mechanical vs manual CPR debate. 2

Study Methodology and Results

6 month survival was 8.5% and 8.1% for mechanical vs manual CPR respectively

The LINC trial was a multicenter, randomized trial, which enrolled 2,589 cardiac arrest patients in Europe into either a mechanical CPR or manual CPR arm.

  • The primary outcome measure was survival after four hours.
  • Secondary outcomes included neurologic function at ICU discharge, hospital discharge, 1 month, and 6 months.
  • There was no significant difference in any of the outcome measures between the two groups.
    • 4-hour survival rate: Mechanical compression group (23.6%) versus the conventional CPR group (23.7%)
    • 6-month survival rate: Mechanical compression group (8.5%) versus conventional CPR group (8.1%)
    • 6-month good neurologic status (based on Cerebral Performance category of 1 or 2): Mechanical compression group (99%) versus conventional group (94%)
  • There are some discrepant details:
    • The mechanical CPR group received 3 minute cycles of compressions as opposed to 2 minutes in the manual group (so rescuer fatigue would not be an issue)
    • In the mechanical CPR group, defibrillations were given to all patients after 90 seconds of compression without rhythm check to reduce interruptions in compressions (as an unnecessary shock was thought to have less harm than interruptions).

Prehospital CPR is a unique entity

Manual CPR performance in the field creates unique challenges for EMTs and paramedics. In the hospital, the patient generally remains on a gurney with adequate staff to rotate every two minutes per the AHA standards. In a six-story walkup apartment, the situation is different: the initial team consists of two people, one of which needs to gather information, ventilate, obtain access, and give medications while the other does compressions. This is after they just carried their equipment up six flights of stairs. Maybe they have upset or angry bystanders to control as well. Eventually backup arrives, which may only be another pair of providers.

Many of these resuscitations are pronounced in the field, but not all. On scene, a mechanical CPR device can free up a provider to perform other tasks. During transport, however, is where the benefit of mechanical CPR truly emerges. Imagine performing compression in a stairwell, stopping on each landing to “catch-up” compressions since you can’t carry the person and do compressions simultaneously. Imagine performing compressions in an elevator on an upright patient because there is not enough room to lay the patient flat. Imagine standing up unrestrained in a moving ambulance driving lights and sirens through traffic to perform compressions on a patient.

None of these situations lend towards effective compressions, and some are even dangerous to providers. This is the reality of prehospital care that cannot be nicely quantified, and this is why proof of equivalency in neurologic outcomes between manual and mechanical CPR is so important.

Bottom Line

The LUCAS device can provide effective and uninterrupted compressions during the entire EMS scene and transport experience. Although the LINC trial demonstrated that mechanical CPR is equivalent to manual CPR, mechanical compressions may reduce the risk of injury to providers without compromising patient outcome. From the perspective of the EMS arena, mechanical CPR may be more than just equivalent.

Hallstrom A, Rea T, Sayre M, et al. Manual chest compression vs use of an automated chest compression device during resuscitation following out-of-hospital cardiac arrest: a randomized trial. JAMA. 2006;295(22):2620-2628. [PubMed]
Rubertsson S, Lindgren E, Smekal D, et al. Mechanical chest compressions and simultaneous defibrillation vs conventional cardiopulmonary resuscitation in out-of-hospital cardiac arrest: the LINC randomized trial. JAMA. 2014;311(1):53-61. [PubMed]

Expert Peer Review

Mechanical CPR is a concept that simply makes sense. Why not use a machine that creates the simple up and down motion of traditional CPR to free up providers to do other interventions? For years we’ve worked to maximize the quality of compressions. Rate is important so use a metronome or keep a song in your head that mimics the rate (the most frequently referred to is “Stayin’ Alive” by the Beegees but others have been cited). Humans get tired so let’s switch providers frequently. Pauses in CPR to check for pulse, cardiovert etc are bad. Unfortunately, based on Salim Rezaie’s ALiEM post and Scott Weingart’s peer review, it doesn’t look like hands-on defibrillation is ready for prime time. The device, however, allows for shocks to be delivered while compressions are administered. Solving these human factors with a simple automated device makes sense but does it work?

The two studies cited here by Dr. Kivlehan suggest these devices don’t help. The first study from 2005 demonstrated harm with an increased mortality rate in the group receiving treatment from the mechanical device.[1] The more recent study (LINC trial) demonstrated equivalence of the LUCAS device with traditional CPR.[2] Additionally, some advantages were given to the device: no blinding (clearly can’t be done), longer cycles of compressions (debatable advantage – increases pauses in traditional group), shock given without pulse check (decreases pauses in compressions). Bottom-line, though, is that there was no benefit in the primary outcome of survival at four hours.

Dr. Kivlehan, and many others, argues that use of this device is safer in the prehospital setting because compressions can continue while the patient is being carried, loaded or driven to the ED without interruptions and without harm to providers. I’m not a prehospital provider but it seems to me that patients without ROSC in the field shouldn’t be transported. This is supported by a recent study in Critical Care showing that only 0.49% of patients without ROSC in the field had a good neurologic outcome.[3] A longer commentary on this paper can be found on Ryan Radecki’s EM Lit of Note site.

So what’s the real bottom line on mechanical devices? At this point in time, there’s no data that supports their use over traditional compressions. That being said, I guarantee there will be more studies looking for benefit. Why? The LUCAS device costs roughly $15,000. That’s a big price tag. Proving a benefit will lead to huge device sales. One in every ambulance in NYC (about 300 on the street at any one time) would carry a price tag of $4.5 million. How many should each hospital have? There’s a lot of money around this device but no numbers to support it’s widespread adoption.

For more information on the LUCAS trial, please see Rory Spiegel’s post on EM Lit of Note.


  1. Hallstrom A, Rea TD, Sayre MR, et al. Manual Chest Compression vs Use of an Automated Chest Compression Device During Resuscitation Following Out-of-Hospital Cardiac Arrest: A Randomized Trial. JAMA 2006 Jun 14 (295)22:2620-2628. Pubmed
  2. Rubertsson S, Lindgren E, Smekal D, et al. Mechanical Chest Compressions and Simultaneous Defibrillation vs Conventional Cardiopulmonary Resuscitation in Out-of-Hospital Cardiac Arrest: The LINC Randomized Trial. JAMA 2013 Nov 17 [Epub ahead of print]. Pubmed
  3. Goto Y, Maeda T, Nakatsu-Goto, Y. Neurological outcomes in patients transported to hospital without prehospital return of spontaneous circulation after cardiac arrest. Critical Care 2013; 17:R274 doi: 10.1186/cc13121 [Open Access]
Anand Swaminathan, MD MPH
Assistant Residency Director and Assistant Professor of Emergency Medicine, Bellevue/NYU ; Faculty Editor of EM Lyceum
Sean Kivlehan, MD MPH

Sean Kivlehan, MD MPH

Associate Director
International Emergency Medicine Fellowship
Department of Emergency Medicine
Brigham and Women's Hospital
Harvard Medical School
Sean Kivlehan, MD MPH

Latest posts by Sean Kivlehan, MD MPH (see all)

  • Christopher

    I think looking at this another way, if your providers pay attention to their chest compression quality they end up doing a good job. Perhaps we can even say mechanical CPR is as good as a trained provider.

    As for the last assertion, I’ve got some problems with it. It is hard to fathom what risk to injury I have while providing manual CPR, unless you’re referring to carpal tunnel syndrome or a repetitive stress injury. If you mean risk due to transport…that’s odd, because nobody should be transporting an active cardiac arrest.

    Transporting cardiac arrest patients is practically negligent given the wealth of evidence of harm caused by inadequate compressions during movement. The hallmark of a low performing system is a “load and go” strategy for cardiac arrest, and shows the pre-hospital providers have not stepped up to own resuscitation from OOHCA. I’m not aware of any system which transports active cardiac arrests and has good survival to discharge numbers.

    I don’t think you can say mechanical CPR allows for transport either; there have been no studies which show transporting patients with ongoing mechanical CPR is equivalent to the high survival to discharge rates in systems which “stay and play”. Small mannequin studies have shown M-CPR provides consistent chest compressions in vehicles, but that is it. I’m not really sure who M-CPR frees up either, because the only two pre-ROSC interventions which are known to improve survival to discharge are compressions and defibrillation. Everything else can wait.

    Regardless, the reality is only a very few patients will benefit from additional in-hospital resuscitation and most hospitals are not prepared to provide the truly “additional” therapies with on-going CPR (ECMO, PCI, PE lysis). Some areas do, and I think inclusion criteria for transport should be placed into their general arrest protocols but only when you’ve applied maximal on-scene therapy (25? 30 minutes? Probably more if you’re good.). So until M-CPR can beat my amazing EMT’s and firefighters at a price which fits my budget, I have no choice but to go with the human component!

    I’ll end with saying I’m not a total negative nancy about M-CPR, and some outstanding physicians and paramedics I respect are proponents (simply their love of the device demands I pay attention to it). Obviously there should be little practical difference between well executed manual and mechanical CPR; in fact logic points to M-CPR’s superiority in consistency of delivery and absence of fatigue.

    Our gains in survival to discharge have been thru a return to the basics and a focus on bringing the resuscitation to the patient. If M-CPR is used to continue the broken load-and-go cardiac arrest care, we’re very likely to lose any gains we’ve made.

    • Matt Anderson

      I think your argument is accurate for PEA/Asystole as well as unwitnessed arrest. However, transporting actively arresting patients ie VF/VT and witnessed arrest to a higher level of care and where resources are robust makes more sense then continuing OOCHA until ROSC is obtained vs withdrawal of care. Just my opinion – I have seen transports occur with the LUCAS and there’s no difficulty in accomplishing this task. Moreover, the device allows less staff in the ambulance given no need for manual CPR.

      I have seen the LUCAS work and I’m a believer. Manual CPR is resources heavy with multiple ED staff having to take part in manual CPR. This device frees up those staff to either help with other logistical arrest activities or to continue to help support the flow in the emergency department.

      • Christopher

        The problem is the ED is not a higher level of care for witnessed or primary VF/VT arrest, or at least there is no compelling reason to believe that an ED is a better place to resuscitate an OOHCA patient than where you found them (nothing is done differently in the ED for non-refractory patients.)

        But to your point, the logistics of OOHCA are certainly simpler if I don’t have to allocate humans to CPR. However, making it easier to move the patient isn’t a concern even remotely on my radar during an arrest. Honestly, if we’re looking to improve ED flow we shouldn’t bring them a patient in active arrest anyways 🙂

        I think there is utility today in its application for refractory arrest, CPR+PCI, helo/fixed wing CPR, and resource constrained environments. Otherwise you’re asking services to spend $15-25k on a device which does not modify outcomes.

        • Matt Anderson

          Nothing from the ACLS standpoint may not be being done differently. But, there are a number of things that happen in the ED that do not regularly happen in the OOCHA setting – cardiac ultrasound, E-FAS, complete documentation with times and drugs given with associated rhythm, definitive airway when needed ie we use a CCR protocol. IMHO getting a patient to where robust resources are available simply makes since to me. With that said, there should be OOCHA algorithms used by EMS systems as this has shown to prevent unnecessary transfers. To transport vs not transport truly depends on the patient and the comfort of the OOCHA provider to give max aggressive care.

          If a system cannot afford the device then you have to use what you have and that’s manual CPR but if the system can afford it then it is worth looking at as all manual CPR is not created equally.

          • Christopher

            US/eFAST/labs is about the only thing our system does not provide in the field which conceptually makes a difference in survival (albeit in your undifferentiated PEA/Asystole crowd). I’d personally rather have POC labs or US than M-CPR if they were equivalently priced.

            When a simple move to field resus and focused chest compressions tripled our survival to discharge, we must take a skeptical view of a return to the old ways.

          • Sean Kivlehan

            Another great point that I hope to address in a future post. Availability of POC testing in the field is on the horizon, and some systems are already using POC lactates, troponins, and potassium levels in the field to help with decision making. As technology becomes more portable and affordable, we will likely see more of this. I look forward to seeing data on whether these interventions improve outcomes. Ultrasound is another modality that may have a future in EMS, although it will require training as it is operator dependent. Evaluating for pneumothorax, effusion, hemoperitoneum, and assistance in vascular access are uses that come to mind. Some systems are already experimenting with this and I am curious to hear their opinions.

          • The inclination to transport OOHCA is a strong one for reasons that have never been clear to me. Recent article about mechanical CPR vs manual CPR in moving ambulance. I am surprised this study received IRB approval and the willing consent of subjects to accept considerable risk to participate.

    • Sean Kivlehan

      Thanks so much for your comments Christopher – this is exactly the type of discussion I was hoping we would create. The risk of injury I mention is two fold – back injuries sustained by providers performing compressions in awkward positions (such as a narrow hallway etc) and doing compressions unrestrained in a moving ambulance. Although carpal tunnel syndrome wasn’t what I was picturing, I’m sure a provider with the condition would welcome the LUCAS!

      All kidding aside, you bring up a really good point with the transport of cardiac arrests. We have become more aggressive with field pronouncement over recent year, which I think is a good thing and the research supports. There will still be times when arrest patients do need to be transported though, in particular patients felt to be viable, for example PEA with a proximate cause such as toxins or STEMI or hypothermia. Also, there is the issue of in-transport arrest of the unstable peri-arrest patient as well as a re-arrest of a ROSC patient once transport has been initiated.

      I. like you, am a strong proponent of aggressive EMS management of cardiac arrest in the field and feel that paramedics are extremely well trained in ACLS. For patients with a poor prognosis, transport is not going to be an issue. But even on scene, if planning on terminating, it is likely that the team has performed a few round of standard ACLS care on the patient which would include more than just compressions and defibrillation (although you are correct that those are the only two interventions clearly shown to improve outcomes. Therefore, the device would free up a provider to assist with these task as well as speaking with family.

      I don’t expect you or anybody to choose to use the LUCAS based on my opinion. In fact only a few months ago I gave a lecture in which I recommended against mechanical CPR, which was based on the evidence at the time. However, new and more complete data is now here and providers need to assess it to form their own opinions.

      • If OOHCA is a rare event and PEA with a proximate cause such as toxins, STEMI, or hypothermia is a rare subset of the OOHCA does having a mechanical compression device encourage expanding the inclusion criteria for patients to transport? Also there is the added burden and potential transport of getting the patient to a facility capable of and prepared to treat the underlying cause.

  • Teri

    Sean: Great article, We have been using LUCAS as the device of choice for 2 years now. We have had some positive outcomes with its use. Additionally during transport (YES WE TRANSPORT cardiac arrests-for many reasons) the EMT/Medic is not standing doing compressions, the driver is now concentrating just on the road and not worrying about throwing people around in the back, and the consistency of the depth/rate is accurate. GREAT DEVICE!!! Now I can say that it’s DR. SEAN APPROVED!!! 🙂

    • What are some of the reasons your organization transports patients with no pulse?

  • Christian

    We used the LUCAS a bit in the ICU on my last rotation. The anecdotal evidence from the awesome Drs. there was pretty profound. Additionally, the anesthesia attendings were pretty into it and it totally made sense in difficult positioning areas, like the ICU. It actually made sense in a lot of scenarios, and I have to admit, I’m a fan. They also commented on its more ideal positioning while attempting to intubate patients as the backboard prevents bouncing, and you are welcome to shock while the LUCAS is running as well (minimizing delays even more).

    That being said, I entered the a different ICU for a code the other day and walked right up to the bed, moved someone out of he way who was doing the old “1-hand chest compressions”, and gave solid compressions for 1 cycle and we had ROSC. One might argue that quickly approaching the patient and giving quality chest compressions without delay would be better than going to grab the LUCAS, although, I guess, if the person in my scenario had just gone to get the LUCAS and put it on instead of giving poor compressions, it would have been time better spent.

    Anyway, thanks for the great review!

  • Seth Trueger

    Interesting post & comments. I had a nice discussion about MCPR with Howie Mell at ACEP, and our consensus was this: the big question is whether the benefits of MCPR (no fatigue, good compressions, provider safety, etc) are worth the delays to starting compression in getting the device set up (and the monetary costs). This study suggests that on balance, it’s not worth it.

    • Dr. Howie Mell

      Thanks Seth. I stand by what we came up with. Even slight delays in the initiation of CPR are shown to produce worse outcomes. Push hard, push fast. Using a pit crew approach and compression only CPR, I have large fire departments with >20% survival (defined as discharged sufficiently intact as to not be on hospice nor require 24 hour nursing care) for all comers in OOHCA. If we use Ulstein criteria, closer to 70%. And I completely disagree with Matt (above). There is no reason to transport OOHCA. Stay where you are and solve the problem!

      • Sean Kivlehan

        Thank you everyone for your thoughtful comments and Dr Swaminathan for the review. I agree that this is not evidence that mechanical CPR is better than manual CPR, but to say that there is never an indication to transport OOHCA is untrue. There will be re-arrests of ROSC patients and circumstances (as I mentioned above) in which transport is indicated. Yes, these are infrequent, and the majority should not be transported.

        Whether or not the cost of the device is worth it is another question, one which I cannot answer. Even in the context of non-ROSC arrests that are not transported, CPR will still be done on scene for a period of time: Is the device really more expensive than sending a fully staffed fire truck to a medical call just so there are more hands to rotate through compressions? Is that truly the most effective use of resources? From that perspective it may be useful and cost effective for arrests on scene.

        Perhaps the best thing that this discussion shows is that EMS is continuing to mature, and that there are smart people using evidence to intelligently debate what is best for our patients. The future of EMS is bright.

        • Seth Trueger

          I will add the defensive caveat that I only brought up financial cost as a parenthetical- if we can save 1 patient then $15k is probably worth it. The big question is whether the non-financial cost of getting the device set up / not compressing immediately is worth the advantages down the road. My guess (which this study supports) is no.

  • Cliff Reid

    Hi Dr Kivlehan
    Thanks for the post. You state in your bottom line: “…the LINC trial demonstrated that mechanical CPR is equivalent to manual CPR”. As this wasn’t an equivalence study, isn’t it the case that all we can say is that the study did not show a difference between mechanical and manual CPR? This difference in interpretation of the results is subtle but arguably important.


  • Mike G

    If a patient arrests enroute to the hospital after ROSC, it sure is a LOT safer to have a LUCAS in place than to suddenly need to place providers in harms way to resume managing a cardiac arrest.