Cheating Death
Page 5
Ewy is cantankerous and opinionated, and he’s sure those opinions are right. In other words, he’s like a lot of heavyweights in academic medicine. In professional stature, Ewy is decidedly a heavyweight, even if his office is thousands of miles from any ivy-covered wall. After graduating from the University of Kansas, he completed a medical residency at Georgetown. After finishing his training in Washington, D.C., in 1971, Ewy headed west to help launch the University of Arizona’s new teaching hospital in Tucson. He’s run the cardiology department since 1990, which makes him the longest-serving chief of cardiology in the United States. 5
Ewy first got interested in CPR research as a way to fine-tune guidelines for the public. For years, studies have shown that people are much more willing to do a simplified version of CPR—if you tell them to stick with chest compressions and don’t worry about the mouth-to-mouth part. To Ewy, that itself was more than enough reason to support a change in guidelines. But as he took a closer look at the data, to make sure a modified technique would still be reasonably effective, Ewy started to notice something else, something strange. The survival rates for people getting chest compressions alone weren’t only as good as people getting full AHA-approved CPR, they were better. Almost by accident, the public health campaign had stumbled onto a medical discovery.
To understand Ewy’s theory about CPR, you have to know about the three-phase model of cardiac arrest, developed by our friend Dr. Lance Becker from the University of Pennsylvania’s Center for Resuscitation Science and Myron Weisfeldt of Johns Hopkins University. 6 The three distinct phases are electrical, circulatory, and metabolic. The first lasts approximately four minutes, during which time the heart still pulsates with its own electrical energy, even as it fails to generate a coherent, blood-pumping rhythm. The ensuing circulatory phase lasts from approximately four minutes after cardiac arrest until the ten-minute mark. Whatever oxygen was present in the blood has been consumed, and without oxygen, the heart can no longer generate electrical energy. The absence of oxygen also triggers dangerous chemical reactions throughout the body, as cells turn to sources of stored energy. At a certain point—about ten minutes after cardiac arrest, assuming there is no intervention—the cascade of cell-killing chemical reactions reaches a crescendo. This marks the third step toward death, the metabolic phase. It’s during this time that cell death begins in earnest.
The model helps explain why some interventions do work. During the electrical phase, defibrillation is highly effective; after that, not so much. That’s because defibrillation doesn’t restore electricity to the heart; it just resets the rhythm. For it to work, the heart needs to have enough energy present to resume beating once given the chance, like a car battery with enough juice to still take a jump. No matter how hard you try, you can’t restart a completely dead battery.
When you perform CPR, you are in effect sending oxygenated blood to the heart tissue—sort of like kindling to catch the electrical flame of a defibrillator. In traditional CPR, bystanders and paramedics alike are trained to start by checking to see that the airway is clear and to alternate compressions with rescue breaths—mouth to mouth. Paramedics are trained to insert a breathing tube, as well. The artificial breaths are supposed to add oxygen to the blood, and chest compressions are meant to circulate that oxygen. What Ewy realized is that some of that effort might be wasted. From the three-phase model, we see that when breathing ceases, for several minutes there is still a good amount of oxygen sitting in the bloodstream. The human body stores far more oxygen than we are generally aware of, and that oxygen lingers for some time after we’ve actually stopped breathing. Therein lies an important lesson that turns conventional CPR on its head: maybe, just maybe, those artificial breaths aren’t necessary.
The thing is, in order to help, that oxygen has to circulate in the blood. If your heart has stopped, the oxygen can only circulate if someone pumps the heart artificially, by compressing the chest. When you pause to give an artificial breath, you’re not pressing on the chest. The same goes for inserting a breathing tube, a sometimes awkward process that usually takes anywhere from twenty to thirty seconds. 7 Even after administering a shock, you’re supposed to wait and read the heart rhythm, to see if the shock has worked—and try again, if needed. All these extra steps, says Ewy, waste time. Precious time that has cost too many lives.
You don’t need the three-phase model to understand that time is at a desperate premium after cardiac arrest. Every second is critical. According to the American Heart Association, for every minute that goes by without someone attempting CPR or defibrillation, the odds of survival decrease by 7 to 10 percent. 8 If ten minutes go by, survival is a long shot. Delay means more than just a lower chance of survival. Every moment without oxygen increases the chance of brain damage should the victim survive. The heart itself is also at risk; as cardiologists like to say, “Time is muscle,” heart muscle. Even if a patient survives with brain function nearly intact, extra minutes without oxygen means more dead heart tissue, increasing the severity of cardiovascular disease and the risk of a future heart attack.
In Ewy’s view, getting blood and oxygen moving—-compressing the chest—is virtually the only thing that matters. In theory, all the extra steps and equipment are lifesaving, but Ewy felt they could just as easily be distractions. In fact, when the city of Seattle initially put defibrillators on their ambulances, survival rates for cardiac arrest actually went down. 9 It’s not because defibrillators are a bad idea, just the opposite. But anything that slows down the process of CPR poses a new challenge. “If you stop for anything, it’s a disaster,” Ewy likes to say.
With a mix of admiration and exasperation, Dr. Ben Abella, who works with Becker as research director at the Center for Resuscitation Science, calls Ewy a zealot for chest compressions. It’s an accurate description. The takeaway message from Ewy is if you see someone fall to the ground after a cardiac arrest, just start pushing on the chest as fast as you can. To emphasize the goal of preserving full brain function, he dubbed the method CCR, for cardio-cerebral resuscitation.
In the lab, the Sarver team, led by Ewy and cardiologist Dr. Karl Kern, did controlled experiments with animals. In 1993, they compared research subjects who received only chest compressions during resuscitation from cardiac arrest to subjects who received artificial breaths along with the compressions. The animals who got only compressions didn’t just do as well, they did better. 10 After publishing those startling findings, the team kept plugging along. Between 1993 and 2002, they conducted six more studies with pigs, comparing CCR (chest compressions only) to CPR with artificial breaths. They all found the same thing: the breaths provided no extra benefit. 11
If Ewy’s results had been accepted at that time, I probably would have learned a very different sort of CPR when I was in medical school, but there were admittedly a couple of hurdles still to cross. The first problem was, the research subjects were swine, and no one was paying much attention to doctors trying to revive a bunch of pigs. Another problem was that other researchers tried similar experiments and got different results. Their pigs died. 12 When we asked Ewy about that, he was ready with an answer. He says these other studies failed because researchers paralyzed the pigs’ chest muscles, so they couldn’t gasp in at least a small amount of oxygen during cardiac arrest, as would happen during a “real-world” arrest.
Some doctors still argued that the studies were misleading. One argument that caught my eye was that a pig’s trachea, or windpipe, is shaped differently than a human’s. But Ewy says critics were setting the bar impossibly high. “When you’re talking about cardiac arrest, you just cannot do randomized, controlled trials in people,” he told me. Research on swine, he argues, is a good substitute. “In our animal model, we’ve come up with a lot of different ways of doing CPR that improve survival in man.”
Ewy wanted the American Heart Association to stop telling people to give mouth-to-mouth resuscitation, but for a long time, the AHA didn’t see things his w
ay. When it comes to writing guidelines, the AHA lists six criteria that it will consider: Number one, the gold standard, is a randomized, controlled study involving people. Number six is animal studies. To start changing minds, Ewy and his colleagues needed a real-life experiment. But how? Who would possibly be willing to buck the guidance of the national medical organizations? What they needed was someone outside the establishment, someone willing to take a leap of faith.
And then in November 2002, opportunity knocked, more or less out of nowhere. A week after announcing their desire to break with AHA guidelines, Ewy and Kern were at the American Heart Association meeting in Chicago. As always, the meeting was a big one, a virtual small city at the glass-enclosed McCormick Place convention center. Neither of the Arizona physicians had ever seen the burly, bespectacled doctor who strode up out of the crowd.
He introduced himself as Mike Kellum, an ER doctor from southern Wisconsin, just a two-hour drive from the meeting. For almost a decade, he had been the emergency services director for Mercy Health System, a group of clinics and small hospitals serving Rock and Walworth counties. Kellum didn’t work in academia, but he liked to read medical journals in his spare time. He had the energy of a young man, but he was nearly sixty, and his keen interest in resuscitation research dated back to the 1980s, when he’d read an article about dogs who were successfully revived after flatlining on a heart monitor for several minutes. 13
Kellum was well aware that most times when his paramedics were called to the scene of a cardiac arrest, they couldn’t offer much help. To me, he described a sense of impotence as EMS director, going through case report after case report: “Looking at these cardiac arrests, reviewing these, you’re seeing ‘they’re dead,’ ‘they’re dead,’ one after another. ‘Dead, dead, dead, dead.’ After all this time, why are we spending time trying to bring no one back to life?”
The reports out of Arizona gave Kellum a sense that he might have a way to change the game. In the hallway outside the AHA meeting, he told Ewy that he wanted to see if the new protocol could work among the flat, sprawling dairy farms of southern Wisconsin. Soon after, he flew to Tucson, accompanied by three other EMS directors from Wisconsin. This small group of unknown physicians was launching a major challenge to the field of emergency medicine.
While not a prestigious academic center, Mercy Health System is the dominant medical provider in Rock and Walworth counties, about an hour south of Madison. Its sixty-three facilities handle more than 85,000 patients a year. 14 Kellum’s ambulance squads have a lot of ground to cover. Even with sirens blaring full tilt, it takes an ambulance twenty minutes to get from Mercy’s main hospital in Janesville to the far western edge of Rock County. 15
In the three years prior to Kellum’s experiment, emergency teams had responded to ninety-two cases of witnessed cardiac arrest. Of those patients, only nineteen survived—and only fourteen without serious brain damage. It was, as Kellum’s team wrote in a subsequent paper, an awful record—but no worse, probably better in fact, than the results from the rest of the country. Put bluntly, under the accepted standard of care, the vast majority of patients died. 16
Once home Kellum and the other EMS directors made a radical decision to change the way they responded to cardiac emergencies. They would try the resuscitation method that had formally been tested, at that point, only on pigs. They would teach it to their paramedics, firefighters and police officers—everyone who was part of the counties’ 911 emergency response system. The single focus would be ensuring circulation to the brain. Every effort would focus on chest compressions, and interruptions would be kept to a minimum. When they first came to a patient who had stopped breathing, they would immediately begin by giving not fifteen, not thirty, but two hundred hard and fast compressions to the chest. Emergency responders would follow that with a single shock from a defibrillator rather than the multiple shocks that were considered standard medical procedure.
Defibrillation is a powerful lifesaving tool, but Kellum knew that each shock takes precious time. So after each single shock, emergency responders would give another two hundred chest compressions. Rescue breaths were eliminated entirely. A small device would be inserted into the mouth to pump in additional oxygen, but no breathing tube would be inserted until the patient had a pulse or until he or she had received three rounds of shocks and compressions—six hundred chest compressions in all.
All through 2004, Kellum called down to Tucson with updates. Everything was going great; it was obvious the new technique was working. Paramedics were getting saves they had never gotten before. Ewy was thrilled, but it wasn’t enough. He needed other people to know about the results, so every time he got Kellum on the phone, he would harangue him to submit them to an academic journal. Bouncing in his seat across from me, Ewy reenacted the phone calls. “I’d say, ‘Mike, you gotta get some data!’ After he’d called me up several times, I’d be screaming at him, ‘You gotta get some data!’ He’d say, ‘I’m just an ER doc; I can’t do that.’ But eventually I just wore him down,” said Ewy.
In the fall of 2006, when Kellum finally published his article (Ewy was a coauthor) in the American Journal of Medicine, the results were astounding. In the previous three years, of ninety-two people in Rock and Walworth counties who suffered out-of-hospital cardiac arrest, only 15 percent had survived with intact brain function. After the new protocol was implemented, that rate more than tripled. In thirty-three cases of sudden cardiac arrest, nineteen people survived and sixteen of them—48 percent—walked out of the hospital, more or less as good as new. Ewy recalls, “We had a dickens of a time getting it published. People thought this was just too good to believe.” But a longer follow-up study found nearly identical results, 17 and in the meantime, Ewy had found a bigger stage to test the theory.
He had also found an important ally, Dr. Bentley Bobrow. Bobrow is a serious, small, almost dainty man in his mid-thirties. As director of emergency services for the state of Arizona, he oversees the training of paramedics, and by 2005, he was familiar with Ewy’s research, not to mention the real-life experiments in Tucson and southern Wisconsin. Bobrow decided it was time to try the experiment on a larger scale.
Unlike Gordon Ewy, Bobrow tends to worry out loud—especially about whether a reporter will paint him as a rebel. While Ewy has been described as “a constant thorn in the side of the AHA,” Bobrow likes to emphasize what he shares with mainstream thinking. “I don’t want to tell anyone else what to do,” he says. Still, despite the diplomatic language, the crew cut, the neat white dress shirt and tie, an independent streak sticks out, and not just in medicine. Along with his wife and young son, for example, Bobrow doesn’t own a television set. 18 And he doesn’t mince words.
“Some people felt it was negligent to not follow the existing guidelines, but if with the guidelines 97 percent of everyone died, we felt it was incumbent on us to try something new” is how Bobrow explains the urgency. But “something new” was a hard sell. In most Arizona cities, emergency response is handled by the fire department. As Bobrow drove from firehouse to firehouse, he got an earful from medics and firefighters who had been doing things the same way for more than two decades. There were paramedics on the job who had been part of Glendale’s first-ever CPR class back in the 1970s, when CPR was new and almost miraculous in its promise. The paramedics thought of themselves as medical professionals, and now they weren’t supposed to give breaths? Anyone could do this? 19
Patiently, stop after stop, again and again, Bobrow made the pitch. He kept coming back to the same point: What was there to lose? When 97 percent of the patients died, how could they do worse?
In fact, Bobrow and Ewy were confident they would do much better. They’d both heard too many stories to be convinced otherwise. Ewy’s favorite one dates back to the mid-1990s when an emergency physician from Seattle played a tape recording for him. It was of a phone call from a woman who called 911 after her husband collapsed and stopped breathing. While an ambulance raced to
ward the house, the 911 dispatcher tried to guide the woman through basic CPR. The advice would sound familiar to anyone who has taken a CPR class in the past thirty years: Feel for a pulse. Tilt the head back. Check the airway. Listen for breathing. Pinch the nose. Breath into the mouth twice. Fifteen chest compressions. Repeat.
Frightening as the scene must have been, Ewy can’t stop grinning when he tells the story. The frantic woman would ask how far away the ambulance was, and the dispatcher would send her back to continue CPR. “After a while,” says Ewy, “she came back to the phone and said, ‘Why is it, every time I press on his chest, he opens his eyes, and every time I stop and breathe for him, he goes back to sleep?’ ” He paused and gave a rueful laugh. “This woman in ten minutes learned more about cerebral perfusion [getting blood flow to the brain] than we had in fifteen or twenty years of CPR research.” All that research, Ewy says, points to one thing: “You don’t stop pressing on the chest for anything.”
By March 2007, Bobrow’s firefighters were ready to pull up the curtain on their experiment. The results were better than anyone, except perhaps Ewy, had dreamed. Among all victims of out-of-hospital cardiac arrest, the survival rate more than tripled. Among those whose heart stopped in front of a witness who called 911, it nearly quadrupled. Bobrow set the bar high: he didn’t count people as survivors unless they walked out of the hospital without significant brain damage. And yet, among victims of an illness that typically kills more than 95 percent of those it strikes, one in four was walking around almost as if nothing had happened. 20
As we came to learn, this phenomenon wasn’t isolated to small towns in Wisconsin and Arizona. In 2007, researchers in Japan unveiled the results of a massive study on bystander interventions. They examined cases where a person collapsed of sudden cardiac arrest outside of a hospital but in view of a bystander: of those receiving no help before an ambulance arrived, only 3 percent survived; of those who got traditional mouth-to-mouth resuscitation along with chest compressions, the survival rate jumped to 11 percent; but of those who got chest compressions only, it was even better—19 percent. 21