“Push Hard, Push Fast?”

“Push Hard, Push Fast”

We all know the mantra. It’s catchy.

Compressions not deep enough? Push hard.


Not fast enough? Push fast.

Great. Except… One thing many of us have learned as professional rescuers is that the bigger issue is not pushing too slow, but pushing too fast.

“But…push FAST!” I mean, faster has to be better right?

How many times have we witnessed CPR administered in a way that seems like a race to set the world record for fastest compression rate? How many times have we seen someone compress at an appropriate rate only to be chided to go faster?

While we know that compressing at a rate <100/min is too slow to provide effective CPR, how fast should we actually compress?

At some point, can fast be too fast?

I did not find a lot of definitive research on this. What I did find indicates that the optimal rate of compressions are between 100-120 per minute. If the rate rises above 125, ROSC declines sharply (1). The classic law of diminishing returns kicks in, and bad things start to happen to decrease the chances of a successful outcome:

  • Increased rescuer fatigue
  • Compressions become more shallow and less effective
  • Full recoil of the chest is diminished

Clearly, these negative factors work against effective resuscitation and best practices. The only problem is that compression rates often reach the 140-180 range during resuscitative efforts.

I conducted an experiment a couple of years back utilizing an instrumented manikin. I’ll admit that the sample size was small, but I believe the results to be representative of the average rescuer. Compressors were told to provide guideline CPR, and results were recorded. No metronome was utilized. The average rate delivered over the first two minutes was 140. As a BCLS/ACLS instructor, I have witnessed “too fast” compressions often, and I have heard of similar experiences elsewhere. Why does this happen? A few reasons are apparent to me:

  • Adrenaline surge
  • Poor temporal awareness
  • Gap in knowledge

Adrenaline Surge

The adrenaline surge is easy to understand. It makes us want to go fast, and leap tall buildings in a single bound. Some of the fastest compression rates will be at the beginning of the code, just when we need effective compressions most. While it happens to all of us, not everyone has the same ability to control it, and prevent it from hindering our performance.

Poor temporal awareness

Related to the adrenaline surge, when under stress our ability to perceive time is distorted. Against our best efforts, time seems to slow down or speed up to the point where we can not accurately sense time lapses. It becomes difficult to effectively sense the difference between a compression rate of 110 and 140. This requires training, and even better the use of a metronome. This is akin to the problem of inadvertently over-ventilating patients at a too-fast rate.

Think about it: At a compression rate of 120/minute, at the edge of the effective zone, we are providing 2 compressions per second. At a rate of 150 compressions/minute, outside of the effective zone, we are giving 2.5 compressions per second. I don’t think it’s easy to perceive the difference between 2 and 2.5 compressions per second under non-stressful conditions, do you? How do you think we do under stress? Without an effective mechanism to control this, we can easily slip outside the effective range.

Knowledge Gap

Many providers still do not realize that “push hard, push fast” has an upper limit, and that giving compressions at a rate > 125 can be detrimental to outcomes. “The faster the better” approach is still pervasive out in the field. More of an educational emphasis needs to be placed on the idea that faster isn’t always better.

Compression rates matter. While I would like to see more research in this area, for what we know now, we need to have a mechanism in place to prevent “over-compressing”. A metronome or similar device would seem to fit the bill. Simple, affordable, but still very under-utilized.

(1) Study determines optimal chest compression rate (Idris, M.D.)
Image credit: foxnews.com


  • I’m right there with you. I stress this in all my classes. “Don’t go too slow or too fast.” Too slow does not create the necessary perfusion pressures, and too fast does not allow the heart enough time to refill with blood. They have to find the sweet spot…..100/min.

  • Jake says:

    I would add that we are probably now at the point where CPR feedback tools (like Zoll’s “Real CPR Help” available with their monitors, or their new standalone “PocketCPR” feedback device*) have been available long enough that they should be considered standard-of-care for EMS agencies and hospitals. Having used Zoll’s version on several codes, I can say for certain that they make an immediately visible difference in CPR quality.

    * I picked Zoll as an example because that’s what my agency has used for decades and it’s what I’m familiar with. I assume other monitor/defibrillator manufacturers have their own versions.)

  • David Baumrind says:

    Yes, CPR feedback tools are definitely helpful. However, from what I’ve seen, they do not do a great job of preventing someone from compressing too fast.

  • Jake says:

    The ones I have seen at least have a metronome function.

  • Austintr says:

    Our service is in the process of implementing “high performance CPR/pit crew”, part of which entails performing compressions at a rate of 110/min due to a well-observed inconsistency of 10% in rate due to the “human factor”. This allows us to be steadily at a rate between 100-120. So far with post-code review analysis, we are achieving 90% compression fractions, so it will hopefully produce good results.

  • James Emerson says:

    As far as feedback monitor defibrillators are concerned, why are we making adjustments on people when it comes to compressions. Perfect practice on Feedback TRAINING Devices should be accomplished prior to playing “in the game”

  • Mark says:

    David, what rate would you set your metronome to? I would argue that the “sweet spot” is 120, not 100. This is consistent with the findings of Idris, A, et al. “The Relationship Between Chest Compression Rates and Outcomes from Cardiac Arrest” Circulation. Jun 19, 2012; 125(24): 30043012.


  • David Baumrind says:

    @Austin: Glad to hear the changes you are making. Keep us informed of the progress!

    @James: While we “play like we practice”, it can be difficult some of the challenges that occur, especially among those departments that don’t run a lot of codes. The feedback device helps keep us in the “zone”.
    It is utilized by some of the highest performing systems out there.

  • David Baumrind says:

    I would agree that a rate of 100 is on the low end of the range. I have to point out though, that the guidelines say to compress at a rate of “at least 100″, that does not mean push at “100″. That is something that we have created.
    Having said that, there is not much out there regarding the exact “optimal” rate, I wish there was. While the research you cite mentions 120, that puts you very close to 125 which is the edge of the cliff. I have seen some systems have success with 110. I am not sure how much it matters between 110 and 120 (not studied), but we do know it is important to stay <125.

  • Peter Hammarlund says:

    I would add that, theoretically, since the coronary vessels are perfused during diastole (or in between the compressions in the setting of CPR) too fast CPR would propabably diminish the chance for the heart to recover.

  • Tor P says:

    There are numerous metronome apps available for smartphones, which work really well (I use it for all my in-hospital codes). The one I have (Mobile Metronome, for Android) even lets me set a different tone for every 9th beat, which gives an optimal rate for ventilations once the patient is intubated…

  • Mike Copper says:

    I agree with @Andrew Randazzo that Too slow does not create the necessary perfusion pressures, and too fast does not allow the heart enough time to refill with blood. They have to find the sweet spot..100/min.

  • Jonathan Jung says:

    You are right on the money…. we don’t have a to slow problem, we have a to fast problem. I set up the Resuscitation System and teams at our hospital. Part of my job is recording the metrics so I can benchmark our performance both at mock codes and live codes. Our compression rates were averaging over 140 cpm. I have recorded compression rates as fast as 218 in our cathlab. I just presented this to the hospital and everyone seemed quite stunned. Part of the problem is what AHA teaches… >= 100. the other part of the problem is that our team leaders are not intervening and slowing people down. ERC teaches 100-120 cpm, so why not AHA? We are starting to use metronome apps at 110 cpm which seems to be helping. Compression depth has similar problems when people are compressing to deep. Unnecessary broken ribs lead to slower chest recoil not to mention other hazards. I am still waiting for my Physio-Control “true cpr” units which have been back ordered since Feb 2014. They will give you true compression depth using radio freq vs the not so accurate accelerometers (especially on hospital beds). Again part of the problem is AHA teaching >= 5 cm …. ERC teaches 5-6 cm, why not AHA?…. Maybe AHA will address this in 2015? I hope so.

Leave a Reply

Your email address will not be published. Required fields are marked *

EMS 12-Lead

Cardiac Rhythm Analysis, 12-Lead ECG Interpretation, Resuscitation

JEMS Talk: Google Hangout

know it all parapup
83 Year Old Male: Shortness of Breath
@ Kyle I would question your authority to call out people for not having a license or being a know it all parapup when your tx basically entails "call medical control." I think we can both agree that his cardiac output is not great at all. I assume your reluctance to give him any other…
2014-10-30 20:26:11
83 Year Old Male: Shortness of Breath
Well st elevation in avr and v1 associated with anterior and lateral depression would call for possible posterior wall MI. 15 lead would be in order. Also check all the leads for appropriate placing. If v7, v8, and v9 show the elevation i would treat as a STEMI per my protocol. Asprin only until medical…
2014-10-30 18:14:05
The most awesome STEMI test on the internet!
Thanks for the app. It made me think about all that one may see in the field. The only problem was I never got a score or saw the results of how I did other than saying I had completed the test. Anyway a great way to get the old brain working.
2014-10-30 13:14:27
83 Year Old Male: Shortness of Breath
I mostly agree with dustin. I believe this is may be an isolated posterior MI. The R wave in V2 points to it being a posterior MI. otherwise it is a 1st degree av block with a LAHB. I am somewhat concerned with the concordant t segment depression noted and in fact if you were…
2014-10-30 04:22:44
Karl Brennan
Understanding Amiodarone
Great article , however in VF caused by hyperkalemia it should be avoided along with lidocaine , Since it shuts down the K channels, the eiteiology of the arrest hyper K, K channels are needed to exchange K in the cell. Calcium , Bicarbonate, dextrose and insulin should be used to decrease K levels along…
2014-10-30 03:04:45

STEMI Expert?

  • Click here to find out!
  • 12-Lead ECG Challenge Smartphone App


    12-Lead ECG Challenge Smartphone App - $5.99

  • Apple iOS
  • Android
  • Amazon
  • Web Based

  • FRN-TV video review
  • iMedicalApps.com review
  • Interested in resuscitation?

    FireEMS Blogs eNewsletter

    Sign-up to receive our free monthly eNewsletter

    Visitor Map / Stats

    Locations of visitors to this page