In Practice

The patient isn't the only one with a history

By Dr. Suzanne Koven
April 25, 2011

E-mail this article

Invalid E-mail address
Invalid E-mail address

Sending your article

Your article has been sent.

Text size +

The man, a retiree in his 70s, sat shirtless on an exam table as his wife hovered nearby . He didn’t look sick , but his story was worrisome. He’d been having chest pain on and off for several days. What’s more, he had virtually every possible risk factor for coronary artery disease: diabetes, high blood pressure, high cholesterol, a long history of smoking, and several relatives who’d had heart attacks.

The resident who saw him in the clinic offered to wheel him to the emergency room. But the man didn’t want to wait around in the noisy ER and he really didn’t want to be admitted overnight to the hospital. He felt fine, he told the resident. He’d go home, he said, and come back if the pain returned. That’s when the resident asked for my assistance.

Chest pain is one of the most common complaints that bring people into doctors’ offices and emergency rooms. Most of the time the cause of the pain isn’t dangerous. Two of the frequent causes, gastroesophageal reflux (heartburn) and costochondritis (inflammation of the joints between the ribs and breastbone), are uncomfortable but not very serious.

But three not infrequent causes of chest pain — myocardial infarction (heart attack), aortic aneurysm, and pulmonary embolism (a blot clot in the lungs) — can be fatal. One study estimated that up to a quarter of patients who are sent home with a missed diagnosis of acute myocardial infarction will die soon afterward.

So evaluating chest pain is no simple matter. A doctor has to navigate between overreacting (hospitalizing someone with heartburn, for example) and under-reacting (sending someone home who’s having a heart attack). The challenge is to estimate the likelihood that something serious is going on and then, if there is any uncertainty, to err on the side of keeping the patient in the hospital overnight. The diagnosis “R.O.M.I’’ (“rule out myocardial infarction’’) is one of the most frequent reasons for hospitalization.

A cardiologist I once worked with quipped: “Evaluating chest pain is like making love — everyone does it a little differently, but everyone is trying to reach the same conclusion.’’ In the case of chest pain, the “conclusion’’ is determining if the patient is having or at risk for having a heart attack, the most common of the serious conditions causing chest pain.

The three most useful tools we use to figure this out are the electrocardiogram (EKG), the blood levels of two cardiac muscle proteins, CPK and troponin, and the patient’s history. An abnormal EKG or blood tests can clearly indicate damage to the heart. The patient’s history is harder to interpret. Classically, someone having a heart attack has pressure beneath his breast bone, often radiating to the back, jaw, or left arm, and accompanied by sweating and shortness of breath.

But many people, especially women, don’t have textbook symptoms. Up to half of women having heart attacks have no chest pain at all, which accounts for why they are so often misdiagnosed and contributes to the fact that rates of death from heart disease in women haven’t fallen in the past 30 years, as they have for men.

The man in the clinic had a normal EKG, his blood tests hadn’t been run yet, and his chest pain was not typical for heart disease. It lasted for only a minute or two at a time — shorter than usual for cardiac pain — and was closer to his left armpit than the center of his chest. Still, I felt he should stay in the hospital.

Sometimes, paradoxically, acknowledging the limits of what doctors know and can do is the best way to win a patient’s confidence. “Look,’’ I told the man, “There are a few possible scenarios here, and I can’t tell you for sure which one it will be.’’ His choices:

1) You go to the emergency room, get inconvenienced, and we find you haven’t had a heart attack in which case you’re home tomorrow night having a beer and laughing at us.

2) You come into the hospital and you are having a heart attack — which we will likely have caught early, before your heart muscle has been damaged.

3) You go home and you have a heart attack that damages your heart muscle or kills you. The patient, and particularly his wife, found this argument compelling and elected to go to the emergency room.

What I didn’t tell the patient was that, in addition to EKG, the blood tests, and his history, there was another factor I took into consideration in making my recommendation: my own experience. Just as generals are said to always be “fighting the last war,’’ doctors are always correcting their past mistakes.

When it comes to diagnosing acute myocardial infarction, I’ve made two mistakes of which I’m aware, and both have made me more cautious about evaluating patients with chest pain. The first occurred many years ago when I was a junior resident, on duty overnight in the coronary care unit (CCU). I’d been up all night and at 5 in the morning I was called to see a woman in the emergency room who the ER staff thought should go to the CCU.

She had chest pain and, they thought, an abnormal EKG. I disagreed with their interpretation and told them to send the woman to a regular hospital floor. An hour later she was up in the CCU with a more clearly abnormal EKG, which the intern on the floor had copied and posted in the residents’ office with my name on it — a public drubbing I’ve never forgotten.

The second case of coronary disease I misdiagnosed was my own mother’s. Granted, I wasn’t my mom’s doctor, but when she gave up tennis at 75 (too tired to run for the ball), started seeing a physical therapist for left shoulder pain a year later (“rotator cuff’’), and began taking tranquilizers for shortness of breath caused by “anxiety’’ not long after that, I should have connected the dots that led to her heart attack. She turned out to be one of those many women with atypical presentations of heart disease — as was correctly diagnosed when I (finally) took her to the emergency room of my own hospital.

The 70-something retiree with the worried wife was admitted overnight and had normal blood tests and EKGs. He had a stress test, too, and it indicated that he did not have coronary artery disease. But the stress test did show an abnormal heartbeat, which can be treated in a variety of ways including with medication and with an electrical shock called cardioversion.

The decision about which to choose was made according to the available scientific evidence, the patient’s preference, and — I’m certain — the cardiologist’s own professional and perhaps even personal experiences. But that’s another story.

Dr. Suzanne Koven is a primary care internist at Massachusetts General Hospital. She writes a monthly column about the uncertainties, dilemmas, and stories that patients and doctors share in practice. She can be reached at

Health & Wellness video

loading video... (please wait a moment)

Weekly video series | Be Well, Boston
Tips for safe running

Tips for safe running

How to run safely -- whether you're a pro or beginner.

Be Well Boston on Twitter

    waiting for twitterWaiting for to feed in the latest...