CT Scans: Too Much of a Good Thing?

We still take care of plenty of emergencies in the emergency department (the old "ER" of TV fame): trauma, heart attacks, strokes, seizures. But about 20 years ago, as a growing shortage of primary-care physicians meant long waits to see a doctor even for people with insurance, the emergency department began to take on a new role—a diagnostic center for patients with potentially serious, but not immediately life-threatening, conditions. To their accustomed tools of defibrillator paddles and endotracheal tubes, emergency departments added blood tests and electrocardiograms—and one of the most powerful diagnostic tools of all, the CT scan.

Of 100 million emergency-department visits last year, roughly 30 percent involved a CT scan, raising the question: are some patients getting too much radiation?

There is no question that for speed and diagnostic accuracy, the CT scan is a wonderful tool. Essentially a rotating X-ray machine, a CT scanner uses a powerful computer to fashion detailed cross sections from X-rays passing through the body at many different angles. But the original CT scanners were quite slow: a typical scan took up to 20 minutes. In recent years, however, CT scans have been improved, so the images are sharper and many scans now take just a few seconds.

The clarity and speed of CT scanning has made it a workhorse in emergency departments. "Pan scans" of the head, neck, chest, abdomen and pelvis have largely replaced multiple X-rays for trauma victims. Before CT scans were common, emergency-department doctors diagnosed appendicitis based on symptoms, does-it-hurt-here physical exams and blood tests. Unfortunately, one in five patients in whom appendicitis was diagnosed turned out—after an emergency appendectomy—not to have had appendicitis at all. With CT scanning, that doesn't happen very often.

Chest pain is one of the most common reasons people go to the emergency department. It sometimes can signal a life-threatening condition such as pulmonary emboli (potentially fatal blood clots that lodge in the arteries connecting the heart to the lungs), aortic dissections (tears in the lining of the large artery that comes out of the heart) and, of course, heart attacks. CT scans have greatly improved the detection of pulmonary emboli and aortic dissections. But often, the cause of chest pain isn't anything so urgent; about 80 percent of chest pain patients admitted from the emergency department to the hospital turn out not to have acute heart vessel blockage. Studies led by Dr. Udo Hoffmann at Massachusetts General Hospital have shown that CT scans—along with other tests—can help rule out blocked coronary artery disease as the cause of chest pain and reduce unnecessary hospital admissions.

So the question is not whether CT scans can help diagnose serious conditions efficiently, it's whether benefits of CT scans might outweigh the costs and risks for some patients. There definitely are risks: CT scanning involves exposure to hefty doses of radiation. The radiation from a single chest CT is about 100 times the dose of a standard chest X-ray exam (two views, front and from the side). Columbia University researchers have estimated that in a few decades, up to 2 percent of all cancers in the United States in future years will be caused by the radiation from CT scans being done now.

That calculation is still controversial, and many doctors disagree with it, but most do subscribe to the goal of reducing radiation exposure. CT manufacturers and researchers have developed techniques to reduce radiation exposure per scan. One involves tailoring the X-ray dosage by taking into account a patient's size and thickness in different directions. There's also a trend toward low-radiation scans in some cases—the images may be a little fuzzier but supersharp pictures aren't always necessary to make a diagnosis.

Doctors are also being urged to pay special attention to the groups that are more likely to be harmed by radiation: younger people (whose growing tissues are more vulnerable) and people who get scanned many times. Dr. Aaron Sodickson and collaborators at Harvard-affiliated Brigham and Women's Hospital are developing a system that will keep track of the cumulative radiation exposure from CT scans and other imaging studies to identify patients at higher risk of radiation-induced cancer. The system will alert doctors of the risk when they order a CT scan of one of those patients.

Ultimately, the value of any tool lies in the wisdom of those who wield it. In 2009, researchers will be working hard to identify how best to reduce the radiation exposure from CT scans—and to identify those people who can most benefit from this powerful technology.