• Resuscitation
• Accident & Emergency
• Clinical Procedures
• Evidence Based Medicine

The report by Martin et al1 describing the accuracy of resuscitation interventions by Advanced Paediatric Life Support instructors raises some interesting questions, not least for the writers of resuscitation guidelines.

The most obvious question is: Does it really matter that external chest compressions (ECCs) during cardiopulmonary resuscitation (CPR) are performed slightly too fast, not deep enough or in the wrong ratio? From the perspective of the guideline writer, the answer is complex. For example, there is very good evidence that CPR performed by bystanders improves survival in infants and children, but there is no direct evidence in humans that strict compliance with the international recommendations on ECC rate, depth, duty cycle (the ratio between compression and relaxation) and relaxation force will achieve improved patient outcomes during resuscitation.2 Indeed, it would be impossible to design an ethical study to prove this. There is, however, evidence from laboratory animal studies, and some human observations, that inform best methods of optimising pressure and flow during ECC and it is a reasonable assumption that better perfusion should create the potential for improved outcomes.3 Hence, the guideline writer has the difficult task of translating such research evidence into a number or, as in this case,1 a small range of numbers for the interventions during ECC. If a resuscitation provider performs ECC outside these ranges, there will be no sudden failure of blood flow—it is likely the ratio of blood flow to, say, ECC rate is approximately normally distributed and small deviations from the optimum will not have catastrophic consequences. Nevertheless, if the evidence suggests better performance is found with certain values it would seem reasonable for us to strive to achieve those values. But, of course, it is not as simple as that. One mathematical model found that the formula for optimum blood flow during chest compression in children could be achieved with a compression-to-ventilation ratio calculated by the formula of 1.6× square root of body weight in kilograms.4 Clearly this approach is not suitable as a guideline, which should be simple to understand, easy to teach and uncomplicated to apply. The International Liaison Committee on Resuscitation (ILCOR) fully understands this objective and incorporates a working party on implementation, education and training in its deliberations in order to provide a consensus on science and treatment recommendations, which takes into account not just the evidence but also the ease of teaching and practicality. It could still be argued that the current guideline for ECC in infants is too complex however; there are, after all, two different methods of compression using either two fingers or two thumbs, depending on the number of rescuers.5 Nevertheless, the instructions on location, rate, compression depth, duty cycle and relaxation are the same for both. For all paediatric resuscitations, ILCOR made a concerted attempt to distil all the information into a simple sentence when it coined the phrase ‘Push hard, push fast, minimise interruptions of chest compression; allow full chest recoil, and don't provide excessive ventilation’.6 It would seem, from the report by Martin et al1 that the advice on rate has been adopted enthusiastically but the advice of depth is still not heeded.

Another important question is: What should happen when new evidence on resuscitation is presented? There are three possibilities: Ignore it; Change the guidelines immediately; or, Review it, along with any other evidence in a planned review cycle. Clearly, the third option is the only practical solution when one considers the scale of widespread clinical implementation and is the approach adopted by ILCOR. It is, however possible that a piece of evidence emerges that is so compelling that it cannot be ignored. In this case, an interim statement could be issued. An example of this approach was the change in recommendation about the use of automated external defibrillators in children less than 8 years of age.7 In this instance, it was discovered that children were being denied the use of automated external defibrillators unnecessarily and a statement was issued between the 2000 and the 2005 guideline reviews.

Next we must ask: How can the poor outcomes described by Martin et al1 be improved? We are strong advocates for the idea that CPR training should be part of ‘citizenship’ education. Mortality decreases if the general public (ie, ‘lay people’) can do CPR, as seen in the increased survival from out-of-hospital cardiac arrests in Seattle, where bystander rates in performing and delivering CPR is one of the highest in the world. Promoting the benefits of ‘lay CPR’ would likely reduce morbidity as well. There are several programmes directed towards teaching lay people: for adults, there are Red Cross and St John's ambulance courses; for children, there are a variety of courses, some of which incorporate school activities. An example is the Injury Minimisation Program that teaches children how to do CPR. The Injury Minimisation Program delivers public health messages in the context of the school curriculum to year-6 children and, in association with a visit to the local hospital, teaches this receptive age group to deliver and to instruct older people in undertaking CPR. The public health message must be that lay CPR is an integral part of the initial link in the chain-of-care, and that doing something is better than doing nothing.

Ultimately, beyond the public health perspective, saving a life and performing CPR is down to the provider and the report by Martin et al1 highlights the very important fact that we need to know how well we are doing, as we are doing it. This is where we need help from biomedical engineers to develop the technology that gives us real-time feedback about the effectiveness of ECC. Unfortunately, it is not a simple problem to solve. The challenges faced are producing accurate instrumentation that can quickly ascertain the important parameters of CPR to ensure that effective ECC is delivered along with a rapid feedback system that will influence the delivery of the next compression. The problems are compounded by variation in body size and physiology seen within the age ranges of children and adults. In the healthcare environment, we have some alternatives, if a tracheal tube is in place we can incorporate end-tidal carbon dioxide monitoring during CPR to follow the adequacy of ventilation and pulmonary blood flow. We can also combine end-tidal carbon dioxide monitoring with ultrasound measurement of cardiac output (or at least femoral artery volume) and thus follow pulmonary and systemic circulations to assist in determining the effectiveness of CPR. But these advanced technologies are not available for lay CPR. Therefore, we consider that biofeedback is part of the future for delivering resuscitation. This problem is an opportunity for technology developers in biomedical and engineering science; that is, to allow anyone doing CPR to know that they are doing their best in a stressful situation. We now know that if ECC is performed effectively, good neurological outcome is possible, even after prolonged CPR.8 The ultimate hope is that delivering CPR in conjunction with biofeedback mechanisms will further favourably alter the odds of return of spontaneous circulation, survival and improved neurological outcome.