The steps for CPR are well known to caregivers and many bystanders. Some are so familiar with the orchestrated movements, they can perform them spontaneously when needed. But are those efforts effective?
The introduction of equipment designed to measure the effectiveness of chest compressions and ventilations has shined a light on the quality of the most elemental steps in performing effective CPR. In the harsh glare, many professionals are learning that although they can execute all of the steps in a CPR call while managing a cardiac arrest, improvements in the quality of the essential CPR components they deliver—compressions and ventilations—can significantly improve return of spontaneous circulation (ROSC).
Therefore, many agencies are seeking ways to improve the quality of their CPR. Following are case studies of some interesting techniques and guidelines from throughout the world.
Kansas City (Mo.) EMS
In 2006, Kansas City (Mo.) EMS (KCMO) made significant changes to its adult cardiac arrest resuscitation protocol—minimizing interruptions to compressions, increasing the ratio of compressions to ventilation, and de-emphasizing or even delaying intubation. But changing the ratio of compressions to ventilations from 30:2 to 50:2 hasn’t been easy, says Joseph Salomone, MD, the KCMO EMS medical director. The transition required a significant amount of training and research review.
The EMS crews needed convincing, and Salomone also got push-back from his fellow physicians. After four or five years of re-education and demonstrated evidence of survival rates, attitudes are finally changing. “It has taken a bit of time,” he admits. “Missouri is the ‘Show Me’ state.”
These days, Salomone has data on his side. A paper he published in 2009 with Alex G. Garza, MD, et al in Circulation states that survival rates jumped from an average of 22% in the three years prior to implementing adult CPR protocol changes to 44% after the changes.(1) In the first year of the new protocols, Kansas City EMS reported that 38% of cardiac arrest patients survived with favorable neurological outcomes.
Salomone puts strong emphasis on consistent, focused training. “You have to … develop an entire training and verification process.”
It’s also important that the prehospital providers understand what happens in a cardiac arrest; the dangers of over-ventilation; and why timing, rate and depth are critical. “Once the heart stops, circulation is king. Everything else is secondary,” Salomone says.
Citizen CPR begun in the first few minutes of cardiac arrest is another component vital to successful outcomes. Kansas City EMS hopes that starting a hands-only CPR education program in high schools will improve acceptance.
Salomone remains optimistic. “It’s an evolutionary process,” he says. “We have to be willing to adapt to the changes in science and … make improvements. Our success with the initial change allowed us to move to straight continuous chest compression cardiocerebral resuscitation (CCC-CCR) using passive oxygenation in 2008. We continue to demonstrate improved outcomes over our baseline AHA data.”
UC San Diego Medical Center
Dan Davis, MD, has been involved in research and training in both the hospital and prehospital settings. He calls the current training model for advanced cardiac life support (ACLS) fairly rigid, saying there isn’t enough room for customization. “Are the AHA course offerings supposed to be the sole training model? Or can we use the recommended treatment algorithms as a basis from which we can build ‘customized’ training and practice recommendations?” Davis asks. “Flexibility isn’t one of the strengths of the current model.”
The problem, according to Davis, is that the current model doesn’t allow the agency to integrate its own data, tracking or best practices.
Davis developed a role-specific resuscitation training program, one adapted from a version-unique, institution-specific algorithm that incorporates performance improvement data and clinical outcomes. As a result, he says survival rates have tripled at the medical center. What’s more, the program helps identify at-risk patients, which has cut the number of arrests in half.
“It’s not our intent to go against the AHA,” Davis says. He points out that intuitively it makes sense to incorporate your own data into training. However, he admits that the added openness and flexibility carry an inherent potential for abuse.
To avoid this, Davis has worked to develop guidelines for the Advanced Resuscitation Training (ART) course. Sub-courses are modified to accommodate the unique roles and clinical knowledge of specific providers. Examples: Basic Advanced Resuscitation Training (BART), Respiratory Advanced Resuscitation Training (RART) and Pharmacy Resuscitation Training (PHART).
Davis has found that the most successful way to effect change in people’s performance—and subsequent clinical outcomes—is to improve their understanding of the need for the change. “You have to tell people why you want them to do something different,” he says.
To illustrate this point, Davis refers to a small experiment he conducted with an agency in San Diego County. It had recorded the fastest compression times in all of the ROC-PRIMED (Resuscitation Outcomes Consortium Prehospital Resuscitation) study. The responders had heard the message loud and clear about pushing fast and deep, but somehow had failed to incorporate the third component, recoil.
To address the problem, Davis contacted their EMS liaison, who sent memos to all agency personnel, resulting in no change in performance. Next, a trainer with a metronome and resuscitation manikin was sent to provide additional training. Still, there was no change in compression rates. Finally, Davis gave a lecture to describe the science behind the concept. That worked. “It was their understanding that had to change,” Davis says.
In 2009, Davis presented his findings at the AHA Patient Science Symposium and received the award for best presentation. “I think [the award] reflects the desire by the upper echelon of [the] AHA to find an alternative to the current model,” he says.
As this monograph goes to press, Davis is involved in a collaborative effort with the AHA to see how his training program can be incorporated into the current AHA model.
MedStar, Fort Worth, Texas
In Fort Worth, Texas, MedStar first began qualitatively assessing resuscitation provided by its crews when it added a CPR measurement and feedback tool that’s used in conjunction with their monitor/defibrillators. As the assistant to the medical director, Daniel Ebbett, EMT-P, was tasked with incorporating the device into a focused training program designed to stress the basics of CPR. By using it as a tracking tool, Ebbett was able to provide crews with real-time feedback on the quality of CPR provided during cardiac arrest calls and make changes that would benefit the patient.
Ebbett says that through its participation in the CARES registry, MedStar could track 100% of CPR calls from beginning to end. The feedback loop included everyone from first responders to the medical director.
By reviewing a visual display that indicates pauses to specific interventions, crews could identify areas that delayed compressions and determine the best ways to streamline care. Example: To avoid lengthy pauses in chest compressions while ventilating a patient, MedStar has replaced ET tubes with King airways—and now allows the airways to be placed by both ALS and BLS personnel.
Other changes in CPR policy include treating a cardiac arrest on scene rather than transporting the patient to the hospital. (Ebbett says that CPR can be managed fairly effectively standing still, but not in a moving vehicle.) If after 30 minutes on scene the patient remains in cardiac arrest, the medics are cleared to terminate treatment. If the medics get a pulse back, the patient is transported to the appropriate facility.
Ebbett says that all pediatric patients are transported immediately and that pulseless patients under some other conditions are as well, including patients in public settings or patients whose families are uncomfortable that their loved one is not being transported.
This summer, MedStar began a re-education program for the entire system and a reorganization of the way CPR calls are run. They now use a “pit crew” method, assigning roles and switching the person providing the compressions more often. “The focus is on what works,” Ebbett says. The policy includes mandatory use of the CPR feedback device and lots of hands-on practice.
Colorado Springs (Colo.) Fire Department
Battalion Chief Rick Martinez, Colorado Springs (Colo.) Fire Department (CSFD), says his department evaluates and tracks every CPR call and ST segment elevation myocardial infarction (STEMI) call. He knows how long it takes to put on the leads, if compressions are performed correctly and when information is transmitted to the appropriate hospital.
By pushing to become more data-driven, Martinez says he can quickly address areas that need improvement and identify where things are going right. “I really needed to be able to tell the community what their return on investment is,” he says.
Tracking helps improve the fire department’s response to cardiac arrest calls, but CSFD is also making a concerted effort to educate the public in CPR. The department’s goal is to train at least 10% of the 400,000 residents of Colorado Springs in the first year of the program. Within the next two or three years, CSFD officials plan to train between 25 and 30% of their population in CPR.
Tampere University Hospital & University of Tampere Medical School Teiskontie, Finland
When Sanna Hoppu, MD, PhD, began studying the quality of CPR at Tampere University Hospital, she found that there was room for improvement. “We all needed some practice. The number of resuscitation attempts was so low, it was always like [the first time],” she says.
Her solution was to develop a small, well-trained team that could respond to cardiac arrests anywhere in the hospital. Starting with approximately 30 ICU nurses, Hoppu trained with her group once a month for two years. “Soon, this group started to do everything better,” Hoppu says. “Better training, treatment and compressions all help.”
She has since expanded the program to the rest of the hospital.
Hoppu requires the nurses to use a CPR feedback and measurement device with audio voice prompts.
Anesthesiologists are also part of the resuscitation group, intubating the cardiac arrest patients. “As part of advanced resuscitation, we always intubate the patient,” she says. However, chest compressions now continue during the intubation attempt. Hoppu says other devices are being explored.
Along with improvements in the quality of CPR have come discussions regarding do not resuscitate (DNR) orders. According to Hoppu, CPR is not performed on certain critical patients. “It’s not ethical to start a resuscitation attempt if it goes no further,” she says. “However, if you start CPR, you must do it very well. It’s not ethical to work as a nurse or doctor if you don’t know how to do good CPR. That’s my opinion.”
Fire Department, City of Münster, Germany
The City of Münster Fire Department has found that a combination of feedback and training is effective in improving survival rates of cardiac arrest patients. According to Andreas Bohn, MD, emergency medical director for the Münster Fire Department and anesthesiologist and professor in the Department of Anesthesiology and Intensive Care at Münster University Hospital, a back-to-basics approach, augmented by a feedback system, helps the firefighter-EMTs and emergency physicians who respond to emergency medical calls, monitor—and improve—the quality of CPR they provide.
“I am convinced that we can give the patient the best chance to survive by doing the best CPR possible,” says Bohn.
When Bohn took over the office of emergency medical director in 2005, he noticed that the responders were more intent on starting an IV or intubating a cardiac arrest patient, and less intent on chest compressions. In fact, the least experienced responder was often given the task of doing chest compressions during a call. That was the first thing Bohn changed.
At the same time, Bohn began a randomized trial with two feedback systems to compare the quality of CPR with extended feedback, limited feedback and no feedback. The purpose was to identify the optimal amount of feedback needed for prehospital providers to improve the quality of CPR.
He found that even limited acoustic and visual feedback leads to improved compression depth compared with no feedback. Further, chest compression depth of less than 40 mm resulted in decreased ROSC rates.
“We were only able to show that improved chest compressions improve ROSC, but that happened with and without feedback,” Bohn says. “Feedback can only help [responders] realize that compressions [are] poor. The rescuer needs to respond to that.”(2)
Bohn says that after a CPR call, there is a tendency to think that it went well, sometimes in spite of the facts. He uses debriefing sessions with the crews and real-time data gathered during the call to help crews see where they can improve. “These are very well-trained people,” he says. “It’s hard for them to believe me when I say they are too tired to continue chest compressions. But if a machine tells them, there’s no discussion. It’s a fact,” he says.
The results of the department’s efforts are impressive. According to the National CPR Registry of Germany, the Münster Fire Department crews and the physicians who provide ALS care in the field are posting the best survival and ROSC rates in the country.
Chesterfield County (Va.) Fire & EMS
Chesterfield County Fire and EMS has made a concerted effort to improve the chances of survival for cardiac arrest patients, including hiring Allen Yee, FAAEM, as a full-time operational medical director. “For a moderately sized fire department, that’s quite an investment in EMS,” Yee says.
To get a better understanding of the department’s resuscitation efforts, Yee and his coworkers examined the system from beginning to end. They found that training was disjointed and not as intense as traditional trauma instruction. All personnel had gone through the standard AHA courses, but had not come away with an understanding of the physiology of cardiac arrest and the definitive care of the patient.
First, Yee introduced a wholesale change in philosophy, requiring all providers to work as a team. Other changes included updating dispatch protocols; employing new technology, such as a mechanical CPR assist device and intraosseus device; and introducing hypothermia for cardiac arrest patients. “We will continue to look at technology,” Yee says.
All cardiac arrest survivors are transported to specialty hospitals that will best continue post-resuscitative care.
Responders are taught that the first few minutes make an important difference. They are required to use a bag-valve mask for the first 10 minutes of the call and focus on chest compressions. If the airway is still compromised after 10 minutes, they may use an airway device. According to Yee, the department has moved away from intubating cardiac arrest patients, in favor of the King airway or other supraglottis airway devices that can be placed by either ALS or BLS providers.
“Maintaining skills is extremely important,” Yee says. Responders are required to spend a year training on high-fidelity simulators. Mandatory BLS training is supplied by ALS providers during in-station drills to keep skills sharp.
To ease training and avoid errors, all equipment has been standardized across the department. “Whether you are a career or volunteer [member], on the job for one day or [you’re] a 20-year veteran, we do things the same way,” Yee says, “at least, that’s the expectation.”
Yee says that initially, responders viewed CPR as a futile effort because most of their patients died. And they were right. When Yee began his work at Chesterfield, only 7–12% of cardiac arrest patients survived to hospital discharge. One year later, discharge rates had surged to 40% and have remained relatively consistent in the high 20s to lower 30s, he says.
University of Pennsylvania Health System, Philadelphia
In 2006, the University of Pennsylvania Health System began studying the concept of CPR debriefing intervention for hospital providers directly involved in resuscitation events. It was a collaborative, non-confrontational method of reviewing data provided by recording and feedback devices, and looking for ways to improve the quality of CPR provided. According to Benjamin S. Abella, MD, MPhil, clinical research director, Center for Resuscitation Science and Department of Emergency Medicine, the process gave people a chance to find out how well they really did. “Most of us never get a chance to look in the mirror,” he says. The initial program was so successful that the hospital began debriefing in earnest in 2007.
The program is named RAPID, for “resuscitation with actual performance integrated debriefing.” Using a process similar to the military and aviation industries, participants evaluate their performance during stressful and relatively infrequent cardiac events in order to affect future performance. Using technology that collects detailed CPR quality and performance data, responders participate in an integrated debriefing following every resuscitation attempt.
Ideally, the team involved in a cardiac arrest response would meet immediately after or within 24 hours following the event. “For practical reasons, we’ve gone with a once-a-month debriefing,” Abella says.
Abella works with in-hospital providers, primarily ICU and ED personnel, who respond to cardiac arrests at the facility. An important aspect, he says, is to debrief the entire team, including nurses, doctors and respiratory therapists. “People respond well when they receive specific feedback about their specific performance,” Abella says. “It could be a promising opportunity for an EMS system to do as well.”
In 2008, Abella, et al published the results of a study on improving
in-hospital cardiac arrest process and outcomes with performance debriefing. They found that ventilation rates decreased and compression depth increased among the RAPID group, with no significant change in survival to discharge among the baseline cohort.(3)
Following the launch of the program, the hospital has seen an uptick in performance at cardiac arrests. But Abella sees something almost as important. “Although it’s difficult to measure, we feel we’ve changed the culture,” he says.
Sunnybrook-Osler Centre for Prehospital Care, Brampton, Ontario, Canada
Sheldon Cheskes, MD, CCFP (EM), FCFP, is medical director for Peel and Halton Region paramedics at Sunnybrook-Osler Centre for Prehospital Care in Ontario, Canada, and is active in CPR research. “Once the CPR process and quality of CPR provided could be measured through newer technology CPR pads, I became more interested,” he says. It isn’t a surprise then that the region became an early user of CPR data to study cardiac arrest.
Previously, Cheskes says, the only objective measure was obtained from a training manikin. New feedback tools allow providers to see what happens in real-life, chaotic situations. “I felt intuitively that providing feedback to the paramedics would provide a positive effect on survivability,” he says.
However, Cheskes agrees with others that reviewing data is only part of the quality improvement equation. “Whatever device you use isn’t as important as providing feedback and studying the quality of CPR,” he says. Every time the EMS providers at Sunnybrook-Osler participate in a resuscitation, they receive a report card on their performance. “We find that this type of positive feedback is of greater benefit to EMS providers. The paramedics want to do better,” Cheskes says. He cautions that positive feedback doesn’t mean overlooking issues in care that are also identified in the CPR process files.
According to Cheskes, data and feedback work. Since 2006, survival rates have shown improvement. Currently, survival rates are 6.9% for all cardiac arrest patients and 22.6% for arrests with shockable rhythms. In 2006, 5.0% of all arrests survived, and 17.2% of those with a shockable rhythm survived.
In 2008, Cheskes and others published an abstract on the role of chest compression depth during out-of-hospital CPR. The study evaluated 615 cardiac arrest cases in six U.S. and Canadian cities. In more than half of the cases, recorded compression depths were less than 38 mm. Further, one-quarter of the responders who were providing shallow compressions were found to have a compression rate of greater than 120 per minute. According to the abstract, the researchers found an inverse association between compression depth and rate.(4) The more shallow the compressions, the faster the rate.
The researchers concluded that further studies are required to determine the optimum chest compression depth and the relationship between compression rate and chest compression fraction.
South Jordan (Utah) Fire Department
In July 2007, the South Jordan Fire Department launched an ambitious and unique program called, “Project Cardiac React.” The goal of the community initiative is to improve patient outcomes in cardiovascular emergencies through a comprehensive system of care that involves the entire community.
According to South Jordan Fire Chief Chris Evans, Americans have invested billions of dollars over the past few decades in fire prevention, and, as a result, the death toll from fire in the U.S. has significantly decreased. Today, cardiovascular disease claims more lives in a few days than fires do in an entire year.(5) Evans set about to do something about that.
He says the inspiration for the project came from a cardiac arrest call he ran in December 2005. A woman called 9-1-1 to say that her husband had collapsed. While she was on the line with the dispatcher, the man went into cardiac arrest. His wife performed CPR until the crews arrived. The department was field testing a mechanical CPR assist device at the time, so Evans tagged along on the call. It took seven shocks, but the patient’s heart eventually restarted. Five months later, the man walked out of the hospital.
“When that incident occurred, I had been on the fire service for 22 years. I’d never seen a cardiac arrest patient leave the hospital,” he says. Until then, he had accepted as fact that cardiac arrest was always fatal.
Evans began looking for a program to implement, but found that most focused on cardiac arrest or STEMI, but not both. So he built a program from the ground up.
He made sure the fire department was ready by upgrading the EMT-B personnel to EMT-Intermediate and trained all personnel how to acquire, interpret and transmit a 12-lead ECG. “I wanted to ensure consistent care regardless of which apparatus arrived first,” he said.
Next, he had to train the citizens. The department taught CPR classes, but the turnout was pitiful. So he dropped the fee, used on-duty crews to teach the class and used a different program, decreasing class time from four hours to two.
Class participation went from 29 citizens in 2006 to 753 by 2008. Evans credits the increase to targeted advertising in the newspaper. He had preprinted PostIt notes placed on the front page of the local newspaper to point to an insert describing the program. The insert was also included in a packet of coupons delivered to non-subscribers.
“We saw another 300% increase in participation compared with the same time period in the previous year,” Evans says. “It was the most effective marketing tool we have used to date.”
Evans needed funding to launch the next step—placing hundreds of public-access AEDs. This time, he went to area business leaders for support. During one presentation, a local CEO was so impressed with the program that his company donated $150,000. “It really gave us the shot in the arm we needed to get it off the ground,” Evans says.
Evans convinced a local hospital to provide free medical oversight and partnered with a vendor to purchase the AEDs and other equipment.
Eight members of the department were trained as public defibrillator specialists, to provide site surveys, write guidelines and assist businesses in implementing an AED maintenance program.
Concerns about liability led Evans to work with local legislators to update a Utah state law to protect anyone who uses an AED. A local ordinance was enacted to mandate that certain occupancies have an AED available, and established penalties for tampering with an AED. Compliance checks have been added to the fire inspection program.
Evans says that in the past 18 months, 190 AEDs have been deployed. “That’s growing every day,” he says.
In November, Evans began yet another step in the program, adding therapeutic hypothermia for cardiac arrest patients treated by South Jordan paramedics.
Recently, a South Jordan city council member commented during a council meeting that he hopes nobody has a heart attack, “but if they do, South Jordan is the best place to have it.”
Evans has prepared a CD with everything a department needs to start their own program. If interested, contact him at [email protected].
1. Garza AG, Gratton MC, Salomone JA, et al: “Improved patient survival using a modified resuscitation protocol for out-of-hospital cardiac arrest.” Circulation. 2009;119:2597–2605.
2. Lukas RP, Sengelhoff C, Döpker S, et al: “[Chest compression quality: Can feedback technology help?]” [Article in German.] Anaesthesist. 2010;59(2):
3. Edelson DP, Litzinger B, Arora V, et al: “Improving in-hospital cardiac arrest process and outcomes with performance debriefing.” Archives of Internal Medicine. 2008;168(10):1063–1069.
4. Stiell IG, Everson-Stewart S, Christenson J, et al: Abstract P192: “What is the role of chest compression depth during out-of-hospital CPR?” Circulation. 2008;118:S_1487.
5. Lazar R: Legislative Strategies for Modernizing U.S. AED Laws. December 2006. Self-published. www.tvfr.com/safetytips/docs/Lazar-Strategies_for_Modernizing_US_AED_Laws_1206_upd.pdf.
This article originally appeared in an editorial supplement to December 2010 JEMS, FireRescue, Journal of Emergency Nursing and ACEP News as Retraining Pays Off: Systems improve their resuscitation performance.
The steps for CPR are well known to caregivers and many bystanders. Some are so familiar with the orchestrated movements, they can perform them spontaneously when needed. But are those efforts effective?