Skip to content

Mechanical CPR in children


Sometimes a single tweet can stir up something deep inside me that I want to know the answer to.  This time it was Jim DuCanto, airway guru.

Since this is a blog about paediatrics I won’t go into my opinions about mechanical CPR (mCPR) in grown ups. I’ll leave it to Robbie Simpson to rant for me. I have very limited experience in its use which really leads to opinion based rather than evidence based medicine.

My limited experience of mechanical CPR

I have been at the airway end when it was used as a bridge to ECMO and the cath lab in a case of VF storm in a man younger than me. He walked out of the hospital neurologically intact (N=1). I have also seen many, many dead – if not for the rhythmic mechanical thumping on the chest – that I have had to pronounce when they get to hospital instead of them being left in their own home/residential facility.

Cardiorespiratory arrests in kids are rare with an incidence of around 1-20 per 100,000 person years. The majority are respiratory or due to progressive circulatory shock and occur when physiological compensation can no longer occur. Other than a few rare instances cardiopulmonary resuscitation is liable to be futile. One large study of paediatric arrests found that of those children transported to hospital by EMS survival to discharge was 7.8% (3.5% of infants, 10.4% of children and 12.6% of adolescents). According to the ROC Epistry – Cardiac Arrest the incidence of a shockable rhythm was 4-5% in infants and 15% in adolescents.

Jim posited that mechanical CPR might buy staff breathing room to plan interventions, but the most likely effective intervention is airway control and ventilation.

Both devices in current use – the Autopulse and LUCAS-2 – specifically mention paediatric arrests as a contraindication to use. And it is easy to see why. With such variability in sizes it would be near impossible to build a machine that could work on every size of child. So is there an alternative?

Perhaps, rather than relying on technology to do our job better we should focus on incremental gains – doing the basics well.


Even in a large Melbourne quaternary paediatric centre endotracheal intubation is a rare event with only 71 reported events over a one year time period. The majority of these were due to trauma or status epilepticus rather than cardiac arrest. Even then, the first pass success rate was 78%. Other studies have shown an even lower incidence of first pass success when video review was used. They also are associated with prolonged pauses in CPR. You might think that using video laryngoscopy might improve things but that doesn’t seem to be the case in simulation based studies.


Once the tube has gone through the cords we often breath a sigh of relief. If you are anything like me you may have been holding your breath for the attempt and there is a tendency to hyperventilate the patient. A review of simulated paediatric codes found that every single one of them ended up with the patient being bagged around a rate of 40 breaths per minutes rather than the recommended 8-20 breaths. If we want to up our game perhaps we could consider adding an impedance threshold device. These valves attach to the endotracheal tube and limit air entering the lung during the passive expansion phase. This creates a reduction in negative intrathoracic pressure thus improving venous return.


There still seems to be a fear, amongst some healthcare providers, of inserting an intraosseous needle.  It can be confronting, crunching through the outer cortex of bone, but it can (and should) be completed in seconds.

High quality chest compressions

Most studies of paediatric CPR involve small numbers of patients and the heterogenous nature of the circumstances surrounding the arrests make it difficult to combine the data in a meta-analysis. The higher quality studies rely upon video review of the events rather than bystander or scribe feedback. A number of common themes emerge. Chest compressions are performed more slowly than recommended around 10% of the time and too fast around 44% of the time. Once simple low-tech way of improving this is the use of a metronome (or a metronome app if you are so inclined).

We are used to practising CPR on adult mannequins and know how far to compress the chest. This is less obvious in children, and we tend to lean on the chest during the decompression phase. Force transducer/accelerometer technology can provide real-time feedback

It’s hard to get real-time feedback regarding the quality of compressions. So practitioners could take a leaf from the world of adult resuscitation and use end-tidal CO2 as a surrogate marker for perfusion.

Low-dose, high-frequency training can help staff retain their skills. Rather than mandated tri-yearly refresher courses, brief booster training is equally as effective in improving skill retention.

One other effect of technology has not been considered. It is that it removes us from our patients. It distances us at a time when our empathy needs to be at its greatest. When CPR fails (and it will), how will we feel if we have done everything, using a machine to pump the chest? Would we feel better or worse than if we had laid on hands? Would it make our young patient more of a person, of a life lived, or less, in our eyes?

So how can we create a calmer resuscitation? Come to DFTB17 and listen to Tim Horeczko to find out.


Atkins DL, Everson-Stewart S, Sears GK, Daya M, Osmond MH, Warden CR, Berg RA, Resuscitation Outcomes Consortium Investigators. Epidemiology and outcomes from out-of-hospital cardiac arrest in children. Circulation. 2009 Mar 24;119(11):1484-9

Leman P, Morley P. Review article: Updated resuscitation guidelines for 2016: A summary of the Australian and New Zealand Committee on Resuscitation recommendations. Emergency Medicine Australasia. 2016 Aug 1;28(4):379-82.

Long E, Sabato S, Babl FE. Endotracheal intubation in the pediatric emergency department. Pediatric Anesthesia. 2014 Dec 1;24(12):1204-11.

Kerrey BT, Rinderknecht AS, Geis GL, Nigrovic LE, Mittiga MR. Rapid sequence intubation for pediatric emergency patients: higher frequency of failed attempts and adverse effects found by video review. Annals of emergency medicine. 2012 Sep 30;60(3):251-9.

Niebauer JM, White ML, Zinkan JL, Youngblood AQ, Tofil NM. Hyperventilation in pediatric resuscitation: performance in simulated pediatric medical emergencies. Pediatrics. 2011 Nov 1;128(5):e1195-200

Schuerner P, Grande B, Piegeler T, Schlaepfer M, Saager L, Hutcherson MT, Spahn DR, Ruetzler K. Hands-off time for endotracheal intubation during CPR is not altered by the use of the C-MAC video-laryngoscope compared to conventional direct laryngoscopy. A randomized crossover manikin study. PloS one. 2016 May 19;11(5):e0155997.

Milander MM, Hiscok PS, Sanders AB, Kern KB, Berg RA, Ewy GA. Chest compression and ventilation rates during cardiopulmonary resuscitation: the effects of audible tone guidance. Academic Emergency Medicine. 1995 Aug 1;2(8):708-13.

Sutton RM, Niles D, Meaney PA, Aplenc R, French B, Abella BS, Lengetti EL, Berg RA, Helfaer MA, Nadkarni V. Low-dose, high-frequency CPR training improves skill retention of in-hospital pediatric providers. Pediatrics. 2011 Jul 1;128(1):e145-51.

Bhende MS, Thompson AE. Evaluation of an end-tidal CO2 detector during pediatric cardiopulmonary resuscitation. Pediatrics. 1995 Mar 1;95(3):395-9.[



No data was found

Leave a Reply

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

2 thoughts on “Mechanical CPR in children”

  1. Thanks, Jim, for your expert input.

    I think medical ego plays a part in practitioners wanting to intubate when a SGA/SAD would be just as effective. Teaching/empowering all levels of practitioner from EMS to nurses and doctors to use them in order to minimize hands off time is a constant challenge.

    I’ll definitely look out for a Paeds SALAD in Berlin. In my experience it’s usually pool/seawater that is the problem rather than vomit. I’ll be trying out the approach at the Alfred CIA course in a couple of weeks.

  2. Great post Andrew! I’m thankful to have stimulated your thought processes here.

    I’m totally your student on this one when it comes to Peds life support. I do have an interesting option that has shown similar performance in elevating the coronary petfusion pressures during CPR as the impedance threshold device (pig model, (Paul Dorian, Toronto and

    I utilize the Oxylator in routine and emergency cases quite a bit, and its mechanism actually will allow face mask ventilation with continuous chest compressions (thereby being the only resuscitator capable of being used in such a manner), and it functions well with SGA/SAD and tracheal tube. It is not commercially available in Australia yet, but has CE and FDA certifications. Mihn LeCong has one of you’d like to borrow it (I’m sure he’d share it).

    SGA/SAD design continue to advance and amaze in utility and ability. My home-grown tests with various types in a TruCorp SALAD model suggest that the LMA style SGA/SAD’s reduce aspiration on large vomit boluses (150 ml over 2 seconds–unpublished data) by allowing the simulated airway contaminant (SAC) to flow under and around the mask and out of the mouth near the corners of the lips. The double balloon type (Laryngeal Tube in this case) HELD the SAC in the hypopharynx, causing a 50 percent fraction (or more) to enter the collection containers on the lungs. We need a cadaver model of this to really validate this simulation.

    We may have a Peds SALAD system at SMACC this summer if you are interested?