Mechanical CPR Devices - What is the scientific evidence so far with regards to its effectiveness in real life clinical settings?

There are two types of CPR devices:

1. Load-distributing band CPR devices (LDB)
Provide circumferential thoracic compression

2. Piston-driven CPR device (PD)
Provide sternal compressions

In preclinical settings, CPR devices improve coronary perfusion, cardiac output, ROSC.

But how effective are these in real-life clinical settings? What’s the evidence so far?

Let's look at three latest scientific papers:

1. A meta-analysis by Westfall et al in Crit Care Med. 2013;41(7) 

P = OHCA victims (N = 6,538)
I =  CPR devices (both LDB & PD) CPR
C = Manual CPR
O = ROSC (defined as palpable pulse with measurable BP for at least 1 min)

Results:
12 papers analyzed (caveat: only English language papers were selected)
A total of 6,538 subjects with 1,824 ROSC events

Combined analysis of both types of CPR devices:
Treatment effect in favor of higher odds of ROSC with mechanical CPR devices (odds ratio 1.53 [95% CI, 1.32, 1.78]; p < 0.001)

But when analyzed separately, only the LDB-CPR device was found to be superior to manual chest compressions with odds of achieving ROSC being 1.6 times greater

Ref:
Westfall M, Krantz S, Mullin C, Kaufman C. Mechanical versus manual chest compressions in out-of-hospital cardiac arrest: a meta-analysis. Crit Care Med. 2013;41(7):1782-9.

2. The CIRC Trial (2014)
Randomized, unblinded, controlled group sequential trial involving 5 centers: 3 US sites, 2 European sites, N = 4231 cases; industry-funded
P = OHCA victims
I = IA*-LDB device (n = 2099)
C = Manual CPR (n = 2132)
O = survival to hospital discharge (primary)
Caveat: excluded cases where the body sizes were too big for the device
*integrated = manual CPR was provided while device was applied

Results:
The survival to hospital discharge in the Manual CPR group (n = 2132) is 233 (11.0%) compared to LDB CPR (n = 2099) where the result is 196 (9.4%). The adjusted OR is 1.06 (CI 0.83 - 1.10).

Conclusion: LDB device is as good as manual compression

Ref:
Wik L, Olsen JA, Persse D, Sterz F, Lozano M, Jr., Brouwer MA, et al. Manual vs. integrated automatic load-distributing band CPR with equal survival after out of hospital cardiac arrest. The randomized CIRC trial. Resuscitation. 2014;85(6):741-8.

3. The PARAMEDIC Study (2015)
Randomized, pragmatic design to study the clinical outcomes of a piston-driven CPR device (LUCAS-2) vs manual CPR
Intention-to-treat analysis
P = OHCA victims
I = Piston-driven CPR device, Lucas-2 (n = 1652)
C = Manual CPR (n = 2819)
O = survival at 30 days

Results:
In this intention-to-treat pragmatic analysis, 30 day survival was similar in the LUCAS-2 group (104 [6%] of 1652 patients) and in the manual CPR group (193 [7%] of 2819 patients; adjusted odds ratio [OR] 0·86, 95% CI 0·64-1·15).

Conclusion:
No evidence of improvement in 30 day survival with LUCAS-2 compared with manual compressions. On the basis of ours and other recent randomised trials, widespread adoption of mechanical CPR devices for routine use does not improve survival.

Ref:
Perkins GD, Lall R, Quinn T, Deakin CD, Cooke MW, Horton J, et al. Mechanical versus manual chest compression for out-of-hospital cardiac arrest (PARAMEDIC): a pragmatic, cluster randomised controlled trial. Lancet. 2015;385(9972):947-55.

The PARAMEDIC study was conducted as a pragmatic study design. What is a pragmatic study design? Pragmatic study design is getting more popular nowadays due to its real-life generalizability. 

Compared to the more familiar explanatory study design,

1. Pragmatic study is aimed to to evaluate the effectiveness of the intervention in a broad and diverse routine clinical practice whereas the explanatory study design is aimed to evaluate the efficacy of an intervention in a well-defined and controlled setting (ideal/optimal environment)
2. Because pragmatic study is done in diverse clinical settings, it often requires a larger sample size as compared to the explanatory study design that often require a smaller sample size
3. Because pragmatic study design is done in real life settings, the confounders are often poorly controlled whereas in an explanatory study design, the confounders are controlled as much as possible
4. As such, because of its real life settings, pragmatic study is probably often more generalizable as compared to the explanatory study design where it is less generalizable.
5. The bottomline is: pragmatic study design answers the questions of whether an intervention actually works in real life versus the explanatory study design that seeks to answer if and how an intervention works

Summary of evidences so far:

1. Meta-analysis by Westfall et al (2013): LDB device seems better than manual compression in achieving ROSC. Piston-driven device is no better.
2. CIRC trial (2014): IA-LDB is no better than manual compression in survival to hospital discharge
3. PARAMEDIC study (2015): Piston-driven device is no better than manual compression in survival to 30 days

What do these evidences mean to us?
CPR devices in general (not just the piston-driven types like LUCAS, LUCAS-2, but the load-distributing band type like Autopulse) is as good as, but no better, than good quality manual CPR in survival to hospital discharge.   The caveat here is: "good quality". As long as you are able provide good quality CPR, it doesn't really matter whether it is your good old pair of hands or the CPR devices, the outcome is the same. But if you are short of staffs to rotate around for chest compression or you anticipate the prehospital journey is rather long and not conducive for manual CPR, CPR device is a good alternative.

In other words, CPR devices don't work magic. Probably the only good reason to buy one is that it frees the hands of the rescuers for other procedures such as defibrillation with on-going CPR or IV access.

Practically, in my setting, I have experienced with both auto-pulse and LUCAS-2 and both may have technical difficulties like the device cannot be fitted properly especially if the patient is rather big size, and thus wasting valuable time when manual CPR could be provided. In other words, there's a learning curve to overcome.  I have also seen cases where the paramedics applied the device, but unfortunately, half-way through the resuscitation process, the device stopped working, or the paramedics realized the device was not positioned centrally, and so, he removed it, reapplied it and restarted it. Technical faults. Wasting time.

Anyone with anecdotal experiences that CPR devices achieve more ROSC than manual CPR should really ask if this is in fact, more of a reflection of the fact that we are not providing good quality manual CPR than anything else. Unfortunately, not many hospitals in Malaysia are using feedback devices yet to evaluate the equality of our manual CPR.

The bottomline is, I have no conflict of interest with or against any of the companies of these devices. I think these are all great devices to have but I just think that in our quest to look into this Man vs Machine issue, we should be more balanced in our perspective, not to be carried away and lose sight of that it is really the quality of the CPR that makes the difference.