It’s been more than 50 years since the first report of successful resuscitation using “closed chest massage.” In describing the early success of this thing called cardiopulmonary resuscitation (CPR), the authors noted that “anyone anywhere can now initiate cardiac resuscitative procedures. All that is needed are two hands.”1
Little has changed with regard to CPR technique in the past half century, but our understanding of the importance of quality CPR has taken center stage in recent years. Beginning with the “push hard, push fast” approach in the 2005 American Heart Association (AHA) Guidelines for CPR and continuing with such new areas of emphasis as the compressions, airway, breathing (CAB) approach to initial resuscitation described in the 2010 Guidelines, it seems we’re beginning to appreciate the link between quality CPR and survival. Now the challenge becomes finding ways to ensure that such quality is provided during every resuscitation.
Scope of the Problem
Only a few months before the release of the 2005 Guidelines, a landmark study published in JAMA sought to describe the performance of CPR in the out-of-hospital setting in three major European cities.2 They did so after providing updated ACLS training for all the providers and informing them that their CPR performance was the subject of the study. This made its results all the more remarkable.
Very few patients received CPR that met the standard of care at the time. The 2000 Guidelines called for a specific target rate of 100 compressions per minute. The compression rate during CPR typically exceeded 120 compressions per minute, and no compressions were being performed nearly half of the time. These factors combined to yield a net compression rate of just 64 compressions per minute.
Even the compressions that were delivered were of questionable quality. More than 60% of the compressions provided were too shallow. And when you combine adequate depth with complete chest-wall recoil, barely one in four compressions met this definition.
Merging all this information together, despite the fact that providers had just been retrained in standard resuscitation practices and knew that their CPR was being measured, the net result was that patients received adequate chest compressions at a rate of not even 18 per minute. You can imagine what their CPR performance might have been when more removed from their training and without the knowledge that their CPR was being monitored. Both of these were likely the case for most of us in the field “¦ until now.
Defining Quality CPR
“All that is needed are two hands.” Seemingly an oversimplification, this remains true to a great degree today. Mechanical CPR devices currently available in the U.S. haven’t been proven to work better than manual CPR. If performing CPR were simple, however, its delivery would be much better than what we’ve seen in studies, such as the one we just described–or what many of us have witnessed in real life. Put simply, if it’s to make a difference, quality CPR requires attention to the details.
Rendering effective chest compressions involves the optimal performance of five key aspects: ensuring the correct compression rate, allowing for complete chest-wall recoil, pressing to the correct depth, minimizing interruptions and maintaining an appropriate duty cycle. Let’s take a moment to look at each of these facets of quality compressions.
Not too fast, not too slow, but just right: Just right seems easy enough to define. We simply need to find a rate that achieves the best possible forward flow of blood while allowing the heart to fill between compressions. Too fast and the heart won’t fill sufficiently, and there will be no blood to move forward. Too slow and the heart will fill but won’t move that blood sufficiently to maintain effective circulation. So we need a rate that’s just right.
Over the past two decades, numerous studies have sought to define the correct rate for chest compressions during CPR. The worksheets that review these studies and form the basis of the 2010 AHA Guidelines support the recommendation that compressions be delivered at a rate of “at least 100/minute.”3 But they also support the concept that compressions can be too fast and probably shouldn’t exceed 120 per minute. And so our “just right” rate falls within that range of 100—120 compressions per minute.
Don’t lean on the chest, just press on it: Remember that generating blood flow during CPR requires both compression and relaxation of the chest wall. It’s during this latter phase, as a result of the slightly negative resting intrathoracic pressure, that blood is pulled back into the chest, filling the heart prior to the next compression.
It turns out that even the slightest bit of leaning on the chest between compressions can result in positive pressures that eliminate this natural “pull.” To put it in perspective, keep in mind that the negative intrathoracic pressure at rest is only ~4 cm H2O. And although this may not mean much, consider that each of us usually generates nearly 20 times that amount of pressure in our abdomen when we urinate. So it is, in fact, a miniscule amount of pressure–any leaning on the chest–that will prevent chest-wall recoil, eliminate that negative resting pressure and prevent blood return to the chest. We have to allow the chest wall to fully recoil after each compression.
Just deep enough: Ensuring the correct compression depth is difficult, as the JAMA study mentioned above highlights with more than half the compressions failing to do so. And similar to compression rate, it’s important to deliver compressions that are neither too deep nor too shallow. Compressions that are too deep increase the risk of injury (e.g., rib fractures, pneuomothorax, liver or splenic lacerations), and compressions that are too shallow will result in inadequate blood flow.
The 2010 AHA Guidelines recommend that compressions be delivered at a depth of “at least 2 inches.” And reviewing the aforementioned worksheets, it’s clear that the depth should probably not exceed 2.5 inches.
Don’t stop: Only half of the resuscitation time in the JAMA study was spent actually performing compressions, and the reasons for this are all too common. As was likely the cause in that study, compressions are frequently “held” for airway management, pulse checks, rhythm interpretation and patient movement, as well as to change provider roles, charge the defibrillator and defibrillate. But we know that interruption of chest compressions reduces perfusion and survival.4,5 This means we must limit the interruption of chest compressions, ideally to no more than 10 seconds.
One and two and “¦ : The use of a cadence like this isn’t coincidental. It accomplishes an important final goal of delivering chest compressions with the appropriate rhythm, maintaining the appropriate duty cycle, which is the percentage of time spent pressing downward during each compression. Said another way, it’s the percentage of time you spend applying pressure to the chest in order to deliver the compressions. Ideally that percentage will fall between 40—50% of the compression time, and the use of a cadence (out loud or in your head) will help achieve that percentage. However, many providers might not need the cadence because delivering 100—120 compressions per minute actually produces a natural duty cycle of ~50%. So attention to one aspect of compressions (rate) may actually help to define quality in this area as well.
Bringing It All Together
It turns out that you do need just “two hands” to deliver effective compressions and to meet the recommendations set forth in the 2010 AHA Guidelines. These two hands must provide compressions at an appropriate rate, allow for complete chest-wall recoil, deliver compressions of sufficient depth, not stop for more than 10 seconds at a time and maintain a rhythm that ensures the ideal duty cycle. And knowing how hard it can be to get all of those things right, we need to look for a way to measure our performance and improve on it.
Reality (Quality) Check
Chest compressions may actually be the most difficult part of resuscitation. They require attention to all the details described above, and they’re essential for maintaining circulation. Without them, the rest of the resuscitation becomes pointless. But most providers have difficulty achieving those goals, as described in the JAMA article.
Here’s where two other important recommendations from the 2010 AHA Guidelines come into play–quality improvement and the potential benefits of real-time CPR prompting and feedback. In fact, it was recognition of these important items that contributed to the decision by the Fire Department of New York (FDNY) to adopt a new ALS monitor–the Philips HeartStart MRx and its Q-CPR technology.
Inherent to efforts to improve on the quality of the CPR being delivered within a system is the ability to accurately measure and review the various aspects of the chest compressions described above. Via the HeartStart MRx, its Q-CPR technology and the Event Review Pro software, the FDNY is now able to review all these compression parameters for every resuscitation in which our paramedics are involved.
Looking at the resuscitation data after the event is important, but perhaps an even more valuable tool is the ability to measure the CPR performance in real time, to provide feedback to the providers and to guide their resuscitation efforts. And although the AHA Guidelines appropriately note that “there are no studies to date that demonstrate a significant improvement in patient survival related to the use of CPR feedback devices during actual cardiac arrest events “¦ real-time CPR feedback technology, such as visual and auditory prompting devices, can improve the quality of CPR.”
2010 Guidelines & the FDNY
During the past several years, the New York City 9-1-1 system has implemented a number of initiatives to improve cardiac arrest survival throughout the five boroughs. The result of these changes and initiatives has been a significant increase in the number of patients who have survived following out-of-hospital cardiac arrest.
The most recent of these initiatives was the decision to introduce the Philips HeartStart MRx with the Q-CPR feature. We believed this technology would provide us with the data necessary to appropriately oversee the resuscitations by our providers, deliver real-time feedback to those providers to optimize each resuscitation effort and collectively develop a data set that would allow us to define CPR benchmarks for a “quality” resuscitation effort.
Quality Improvement Efforts
In my opinion, one of the most impressive parts of the 2010 AHA Guidelines was the specific mention of the need for quality improvement measures. “This process of quality improvement consists of “¦ (1) systematic evaluation of resuscitation care and outcome, (2) benchmarking with stakeholder feedback, and (3) strategic efforts to address identified deficiencies.”
As a result of the introduction of the Philips MRx within the FDNY, combined with our interest in improving cardiac arrest outcomes, these principles are becoming part of our ongoing quality assurance (QA) and quality improvement (QI) efforts.
In New York City, cardiac arrest patients are transported only to cardiac arrest centers, which are hospitals that have partnered with the FDNY to provide therapeutic hypothermia and are required to provide outcomes and other data points for all cardiac arrest patients. This data is added to prehospital data and can now be combined with the CPR performance data derived from the Q-CPR feature. This master data set will allow us to analyze the various aspects of CPR performance to establish benchmarks that reflect “quality CPR” and that are defined by their likelihood to improve cardiac arrest outcomes. We will then be able to measure performance during each resuscitation against these benchmarks.
This has been done by others in other systems. In 2010, Ahamed H. Idris, MD, and his Resuscitation Outcome Consortium (ROC) colleagues reported that a 60% flow time should be considered a minimum standard for CPR performance to improve survival.3 We look forward to validating such statements and hope to address other aspects of CPR performance, such as compression depth, compression rate, duty cycle, incomplete chest-wall recoil, and CPR pauses before and after defibrillation. All of these are measured by the MRx and reported for each resuscitation in the Q-CPR report. The Q-CPR data also allows for close attention to the details of the resuscitation by providing a large data set for every 30-second interval. Interruptions that result from airway management, patient movement and rhythm check can be identified and used to discuss the resuscitation with the providers involved as part of a thorough QA review.
Finally, we plan to address the issue of deficiencies and successes by providing these reports to the field providers, including the EMTs, paramedics, firefighters and EMS officers. Because this data is aggregated from the entire resuscitation, it won’t identify specific individuals or times in which less-effective CPR was delivered. But we believe that it will help reinforce the concept of the resuscitation team by making all the providers on the scene responsible for the overall quality of the CPR being provided during the resuscitation.
Central to any effective resuscitation is the delivery of quality CPR, including adequate rate, depth, chest-wall recoil, duty cycle and limited interruptions. Technologies, such as the Philips MRx with Q-CPR, allow for both real-time feedback to the providers and post-hoc QA and QI efforts, which are central to overseeing a resuscitation system. This dual approach to resuscitation oversight was one of the reasons that the FDNY chose to implement this device this past year. And these types of technologies mean any EMS system can implement similar oversight mechanisms to help ensure the best possible care for their patients. After all, that report from more than 50 years ago appears to have been correct when it stated, “All that is needed is two hands.”2 Or perhaps two hands and a way to make sure those hands do what they’re supposed to do.
Disclosure: The author has reported receiving past honoraria and/or research support, either directly or indirectly, from the sponsor of this supplement. FDNY has received prior grant funding from Philips Healthcare for research for which Dr. Freese was the principal investigator.
1. Kouwenhoven WB, Jude JR, Knickerbocker G. Closed-chest cardiac massage. JAMA, 1960;173(10):1064—1067.
2. Wik L, Kramer-Johansen J, Myklebust H, et al. Quality of cardiopulmonary resuscitation during out-of-hospital cardiac arrest. JAMA. 2005;293(3):299—304.
3. American Heart Association. (n.d.) Advanced Life Support Worksheets. In American Heart Association. www.americanheart.org/.presenter.jhtml?identifier=3060062#CPR
4. Berg RA, Sanders AB, Kern KB, et al. Adverse hemodynamic effects of interrupting chest compressions for rescue breathing during cardiopulmonary resuscitation for ventricular fibrillation cardiac arrest. Circulation. 2001;104(20):2465—2470.
5. Christenson J, Andrusiek D, Everson-Stewart S, et al. Chest compression fraction determines survival in patients with out-of-hospital ventricular fibrillation. Circulation. 2009;120:1241—1247.
This article originally appeared in an editorial supplement to the September 2011 JEMS as “Resuscitation in the City: How technology helps maintain quality CPR in New York.”