The chest pain and the shortness of breath usually come first. Then a burning, tight feeling right under the breastbone. Sometimes the pain is accompanied by nausea, sometimes just the foreboding sense that something is terribly wrong.
With 5 million people each year rushing into emergency rooms complaining of symptoms such as these, doctors face a difficult task: assessing whether or not it is a heart attack -- the leading cause of death in the United States -- or something much more benign, like indigestion or angina. Often the medical history, physical examination and electrocardiogram do not provide conclusive information, says Dr. Robert Roberts, chief of cardiology at Baylor College of Medicine. This can delay treatment for those who need it most; it can also fill up intensive care units with people who may have just inhaled their spaghetti dinner a little too quickly.
"The nature of chest pain is virtually the same, and both are relieved by nitroglycerin," he explains. In fact, according to Roberts, only 10 percent of all the people who think they're having a heart attack actually are.
"It's not an easy thing to diagnose," concedes Elaine Josephson, a doctor at the ER at St. Luke's Hospital in New York and spokeswoman for the American College of Emergency Physicians. "Sometimes it's clear cut, when you evaluate the patient, look at the EKG, and see, 'Yup, this person is having a heart attack.' Other times you're not sure, the pain comes and goes and you don't have the luxury of an old EKG to compare it to."
For most of Roberts' professional life, he has been trying to develop ways to speed up diagnosis for a disease for which every moment matters. In 1974 he developed an important heart test that many hospitals still currently use. Then, eight years ago, Roberts created a new test -- a blood test that can usually tell within an hour if the person is having a heart attack or not. The problem is that most hospitals are still not using it.
In this week's Circulation, the journal of the American Heart Association, Roberts and his team published the results of the first comprehensive study on this new test. Doctors at the Baylor College of Medicine and the University of Texas Medical School tested patients at four hospitals in Houston and then had the results compiled by an outside source. Neither the lab technicians drawing the blood nor the doctors treating the patients knew about results, treating the patients as they normally would. Roberts estimates that the test, if widely used, could save hospitals billions of dollars, and save lives. Salon Health & Body interviewed Roberts by phone from his office in Houston about why this test has not become a fixture in emergency departments across the country, the dangerous crowding in intensive care units and the psychological effects of thinking you're having a heart attack.
What happens when a person walks into an emergency room and complains about chest pains?
There are several issues that we face. First of all, is this a heart attack that must be taken care of now? Secondly, is this pain of the heart (angina), which is not associated with actual damage to the heart? And thirdly, where do we go from here? We can't obviously admit everybody to the hospital.
It is almost impossible to tell by just looking at their medical history and doing a physical examination. And the electrocardiogram, which is routinely administered, will only identify about 40 percent of all heart attacks.
Why is time so important with heart attacks?
When a clot has formed to block off the blood supply to an artery, that part of the heart looses the blood, or at least most of it. If it loses it for more than 15 minutes, it will die. But if you restore blood flow in less time than that, it will essentially recover. Death marches from the inside of the heart to the outside, and it marches over a period of four to six hours. And that's why if you restore blood flow within six hours -- by using blood clotters or angioplasty -- you still will decrease the amount of damage and the death rate. At one hour, it's a 1 percent death rate, and at six, it's a 10 percent, after that it levels out at about 15 percent.
What was the old test like?
For decades, we used an approach called creatine-kinase MB (CK-MB), which took 10 to 12 hours to find out if the person was having a heart attack. CK-MB is an enzyme that's released from the heart in increased amounts after the beginning of an attack. We measure it. But since it takes so long to get the results, a lot of people get admitted to hospitals -- to wait out that interval -- only to find out nine hours later that's it's negative, and then they are sent home. Our study estimates that about $12 billion a year is spent unnecessarily admitting people who didn't need to be admitted. A lot of that money can be eliminated if you use CK-MB subform [the new test].
So your new test could save money, but can it save lives?
Yes. With CK-MB subform we can characterize more than 80 percent of the heart attacks by just administering one test when they first arrive in the emergency room, and then again an hour later. After six hours of onset, it goes up to 90 percent certainty. Plus, the test only takes about 20 minutes and costs the same as the other tests.
You've been using this at Baylor for a few years now. Tell me about the typical person you've been testing.
Someone comes in with chest pains, they are scared -- they should be, I would be, too, probably even more so. But there are a lot of reasons why you can have pain. You can have angina, or you can have indigestion and you can have all of those things that can give you the same chest pains. With this test, an hour later you can find out that everything is normal and that it's not a heart attack, and that you should electively go sometime this week or next and see a gastroenterologist. That is obviously a tremendous relief to anyone who thought they were having a heart attack. And this is the consistent story of people who have been using this test. Also, you don't want to fill up your intensive care unit with people who end up going home within nine hours, which is what the statistics show.
But isn't it better to take the precaution and admit the patient?
It's obviously better to admit them to the hospital and find out that it's negative and then send them home. The most common thing people sue for is chest pains sent home that are heart attacks, and so we overdo it the other way. All I'm saying is that a lot of this can be avoided by using an early marker, such as the CK-MB subform. It's not acceptable to put someone into a coronary care unit or intensive care if you had some way of knowing that they didn't have to be there. It's not exactly psychologically pleasing to be in a coronary care unit with all those beepers going off, and needles sticking out of you everywhere, and everybody monitoring you thinking you have had [or are having] a heart attack, and find out a few hours later now you didn't have one and it was unrelated.
Why aren't more hospitals using this test?
I don't know the answer to that. I developed this test about eight years ago. Maybe they were waiting for a study like this -- it's the first one to be double-blind [meaning that both the people in the lab administering the test and the doctors don't know] and compared with all the other tests. In medicine, we want this thing called "evidence-based medicine." As you know, there are lots of things in medicine that take five or six years to catch on. There are some other hospitals using it right now, at least 20 to 30 hospitals, and people in other countries, like England, Holland, Germany, Belgium and Japan.
Take me through the basics of your study and describe how you conducted the CK-MB subform study.
Over the course of eight months -- seven days a week and 24 hours a day -- we tested 995 patients with chest pain for CK-MB subform, a molecule the heart releases when it undergoes damage. And if you measure just the total amount of CK-MB in the blood you have to wait about 10 to 12 hours. But if you look at these subforms, you can see how one, of the two, changes and compare it to the other one, which remains unchanged. It's that ratio that we measure to pick up the early diagnosis. And within a couple of hours, the ratio changes. The basis of the new test is to pick it up and quantify that ratio.
How about these other early markers, like CK-MB myoglobin and total CK-MB and troponin T and I? Are they effective tests?
People have the impression that cardiac troponin I and T are early markers. Now the literature doesn't say that, but docs think that. In my study, after six hours, they only show about 50 to 60 percent positivity rate, which means that you have to wait up to 14 hours to get 90 percent sensitivity. Myoglobin, on the other hand, is not a bad test for early diagnosis, it has a 78 percent sensitivity.
You speak a lot about reducing the hospital's costs. Is this what it is about, perhaps at the expense of the patient?
No. No. 1, this is about doing proper treatment. Because if you are having a heart attack, you use a clot buster. But if you aren't having a heart attack and you have unstable angina, and you were given a clot buster, it increases your chance of dying. I don't want to make it look like it's just about saving money.
Most people, outside of the 10 percent having heart attacks and 20 percent with unstable angina, do not need to be admitted to the hospital that night. They can be worked on electively, and it would save billions of dollars a year.
Shares