A Matter of Time

When a heart attack strikes, cardiologist William O’Neill wants to see the patient quickly. But it doesn’t always work out that way. Once, for instance, a man whom O’Neill had previously treated was transported to a hospital in a different suburb of Detroit, even though the patient’s wife pleaded with the ambulance company to take him straight to Beaumont Hospital in Royal Oak, Mich., where O’Neill works. At the first facility, doctors evaluated the man and decided that he needed more care than they could provide. So, they sent him back on the road. All told, it took 3 hours for the patient to reach O’Neill’s cardiac catheterization unit. Yet it was just 30 minutes from the man’s home.

Once the man reached Beaumont, the cardiac team used a catheter to insert a balloonlike device into his blocked heart artery and inflate it. That procedure, known as angioplasty, prized open the vessel and restored blood flow to the suffocating heart muscle downstream of the congestion. Although the man survived the ordeal, his heart suffered.

“There was lots of damage to the heart muscle that could have been avoided if he had gotten to our place sooner,” O’Neill says.

Emergency angioplasty is more effective in treating the most serious kind of heart attack than is its main alternative, which relies on intravenous drugs called thrombolytics to break up arterial clots. Angioplasty is particularly effective when it’s followed by the insertion of a mesh support tube, or stent, into the artery. “This is the preferred therapy,” says Alice K. Jacobs, president of the American Heart Association.

Most studies haven’t found a benefit to using a combination of clot-busting drugs and angioplasty.

Perhaps a fifth of U.S. hospitals offer angioplasty on an emergency basis. Most hospitals either don’t have the facilities or staff to perform the procedure or find it impractical to have them available 24 hours a day, says cardiologist Ellen C. Keeley of the University of Texas Southwestern Medical Center in Dallas.

As a result, in the United States, the majority of severe heart attacks—those known as ST-segment-elevated myocardial infarctions—get treated only with clot-busting agents.

For patients found to need angioplasty, a few U.S. cities have local agreements among hospitals and ambulance companies that facilitate immediate transfer from a hospital not offering it to one that does. Furthermore, in Boston, medics riding in ambulances have the tools to identify patients with severe heart attacks and take those patients only to hospitals that perform angioplasty.

But shuttling a patient from place to place in the midst of a heart attack is risky. “Heart muscle is dying as the clock is ticking,” says cardiologist Elliott M. Antman of Brigham and Women’s Hospital in Boston.

Over the past few years, data accumulated in several carefully orchestrated clinical trials have revealed that systematically redirecting patients to angioplasty-ready hospitals can improve health outcomes—at least when the associated delays are brief. Evidence gathered outside of such trials, however, suggests that transfer-related delays are significant in much of the United States.

Dane to consider

Cardiologist Henning Rud Andersen lives and works in Aarhus, a Danish city with about 400,000 residents and two hospitals. A decade ago, his center, Skejby Hospital, was the only one in Denmark to offer emergency angioplasty. The hospital across town, like others in the nation, used thrombolytic drugs instead. Which hospital the ambulance went to determined which treatment a patient got for a heart attack. That generated a discussion among local cardiologists, Andersen says.

“It was an unethical strategy within one city to have two different treatments,” he says. “So, we agreed within this city that we would have only one strategy. If a patient came into the other hospital in our city, we transferred that patient.”

But that resolution only confronted the physicians with a broader geographic dilemma, Andersen recalls. The next question to arise regarded a hospital 30 kilometers to the north of Aarhus. Should that facility administer drugs or transfer patients to Skejby Hospital? “The argument was, ‘It’s unethical to have two different treatments within the same county,'” says Andersen. In the end, he says, “we had to decide for the whole country of Denmark what is the best strategy.”

The result of that debate was a clinical trial, dubbed DANAMI-2, that included Skejby Hospital and 28 other Danish medical centers that collectively served more than 60 percent of the nation’s population. To join the study, teams at four hospitals developed the capability to perform emergency angioplasty and demonstrated their proficiency to international experts.

The non-angioplasty hospitals gave thrombolytic drugs to half the patients, selected randomly, that they diagnosed with ST-segment-elevated attacks and sent the rest to the closest of the five angioplasty centers.

By the end of the 4-year study, the hospitals offering only drug treatment had sent away 567 of their patients. Within a month of the initial crisis, 1 in 12 of the transferred, angioplasty-treated patients had a second heart attack, suffered a disabling stroke, or died. Among the patients who stayed put and got thrombolytic drugs, 1 in 7—nearly twice the proportion—met such a fate within the month.

In 2003, Andersen and his colleagues concluded that transferring patients for angioplasty had been more effective than treating them with the drugs at hand.

DANAMI-2 was the largest clinical trial to test whether transfer for angioplasty is superior to on-site drug treatment. Smaller studies, including one conducted at Beaumont Hospital, had previously supported transfer.

In the Danish study, researchers also recorded the time from the onset of symptoms to the moment that physicians administered drugs or inflated an angioplasty balloon. Time to treatment was typically 55 minutes longer for a patient transferred for angioplasty than for a patient assigned to get thrombolytic drugs. About 32 minutes of the transfer-associated delay was spent in transit between hospitals.

Since the study concluded, Andersen says, Danish ambulances have been outfitted with equipment that enables onboard diagnosis of ST-segment-elevated myocardial infarctions. Now, he says, “there’s no need for these patient to go to their local hospital. They go directly from the field into the [regional] cath lab. It actually speeds up the whole process by approximately 1 hour.”

Today, Andersen says, emergency angioplasty has replaced thrombolysis throughout mainland Denmark and most of its islands. Similar shifts have occurred in France, the Netherlands, and several other European countries.

But that revolution hasn’t reached most of North America.

Burst balloon

“The data from Europe show that if you truly can implement [transfer] quickly, patients do well,” comments Antman. “It remains to be seen whether we can replicate those short transport times” in the United States.

“I don’t think that DANAMI-2 gives us the definitive answer,” adds cardiologist James Brophy of McGill University in Montreal. “There’s definitely a role for transferring patients. I’m just not convinced that … we can zip patients around the countryside.”

Harlan M. Krumholz of Yale University points out that recent U.S. heart attack–treatment guidelines recommend transfer only when angioplasty can be expected within 90 minutes of arrival at the original hospital. Both Krumholz and Antman contributed to those guidelines, which the American College of Cardiology and the American Heart Association jointly published in 2004.

In part to see how closely U.S. practice matches guidelines, Krumholz and his colleagues recently gathered data on treatment delays associated with transfers for angioplasty. Using a national database, they focused on 419 centers that received 4,278 transferred patients. In each case, staff at the initial hospital had decided to transfer the patient instead of administering clot-busting drugs.

The median time between presentation at one hospital and treatment at another was 3 hours, Krumholz and his colleagues reported in the Feb. 15 Circulation. In DANAMI-2, it was just shy of 2 hours, including the 55-minute delay associated with transfer.

“Only 4 percent [of U.S. patients] are treated within 90 minutes … raising the question whether the other 96 percent of the people who were transferred should have gotten the drug instead,” Krumholz says. “Either we need to get faster, or we need to rethink the wisdom of transferring patients,” Krumholz says.

Quicker transfer of heart attack patients is the solution, according to Keeley. But numerous obstacles stand in the way. For one thing, emergency medical services (EMS) can be agonizingly slow to provide ambulance transfers.

“Transferring a patient from one hospital to another is not an EMS priority,” Keeley says. In the Beaumont trial, “waiting for EMS took up most of the transfer time,” she notes.

Other delays occur when patients wait before seeking help, when ER staffs are slow to diagnose patients, and when people drive themselves to a hospital rather than calling 911. Patients arriving by ambulance get first priority in emergency departments.

Furthermore, catheterization teams at the second hospital aren’t always notified promptly that a patient is on the way, Krumholz says. In their study, he and his colleagues noted unnecessary delays between patients’ arrival at the second hospital and their appearance in the cath lab. Better communication between the hospitals could shave valuable minutes from the process, he says.

Ideally, Keeley says, ambulances should handle heart attacks much as they handle trauma, where certain hospitals are automatically bypassed so that patients can be treated in designated regional centers. That would require full-scale, onboard, diagnostic electrocardiograph equipment, which only a small fraction of ambulances currently have.

Boston’s EMS system—which covers up to 1.5 million people occupying 45 square miles—has equipped its vehicles and trained its paramedics to identify ST-segment-elevated heart attacks and redirect those patients to any of Boston’s seven medical centers that provide emergency angioplasty.

For more than 2 years, researchers led by Peter Moyer, who runs Boston’s EMS operations, have been tracking how much time elapses between local 911 calls and resulting angioplasties. They use the 90-minute guideline as a benchmark for their performance and usually “beat it by a good margin,” Moyer says.

Transferring patients expediently should be feasible in any major city, he says, but “we have a long ways to go as a country.”

At the frontier

One day last year, O’Neill fielded an urgent phone call from Windsor Regional Hospital in Ontario. A 44-year-old man with a heart attack, he learned, was headed his way.

The patient had arrived at Windsor with chest pain, and his wife—a nurse who happened to be on duty in the intensive care unit (ICU)—determined its cause. Even after the man received thrombolysis, the only therapy available there, his condition deteriorated. “This ICU nurse was watching her husband dying right in front of her eyes,” O’Neill says.

Windsor and Beaumont hospitals have an arrangement that speeds the transfer of heart attack patients. In addition to contacting O’Neill, Windsor staff arranged for a Canadian ambulance and notified officials at the U.S.-Canadian border.

Minutes later, the patient was rushed through a tunnel that separates Windsor from Detroit. O’Neill and his angioplasty team were waiting in the Beaumont cath lab when the patient arrived half an hour after the call.

“There’s no question that if he had stayed at Windsor he would have died,” O’Neill says. Although the patient’s initial recovery was slow, he’s now back at work. “That’s the kind of case that’s extremely gratifying,” the cardiologist says.

It’s also the kind that makes him think that rapid, routine transfer for heart attack care is achievable. Says O’Neill, “If you can do it across an international border, you should be able to do it across a city limit.”