Do-It-Yourself Diagnosis
A CT scan can send a powerful message. So what's wrong with this picture?
Feeling OK but maybe sweating the effects of a lifetime of cheeseburgers on your coronary arteries? You smoked when you were young and dumb, and now you wake up nights wondering what you might have done to your lungs. Modern technology has the answer. Step right in, ma'am. Try one of our special fast-CT scans. Find little problems before they become big ones; it could save your life. And we take credit cards.
Walk-in CT scanning centers have taken root in cities and tony suburbs where the worried well are willing to pay from several hundred to more than a thousand dollars out of pocket. (Neither private health insurance nor Medicare will cover the cost.) In states where a doctor's referral is required, a physician at the center will be delighted to write one. Many centers offer a smorgasbord of tests, typically a heart scan, lung scan, whole-body scan, and colon scan, or virtual colonoscopy. Their ads stress convenience, speed, and comfort. After the scan, a physician consults with you, raising points to discuss with your doctor. And if you need to alter your unhealthful ways, the scan will help nudge you to do that, too. Right?
Some people see the light. But in last week's Journal of the American Medical Association, the first large study to investigate scanning's potential to change behavior suggested that the majority don't. The discouraging conclusion: The initial alarm at an ominous scan typically fizzles into apathy. For many clinical researchers, it is one more reason, along with marketing techniques they consider dubious, to slam the scans.
The study investigators examined 450 U.S. Army regulars stationed near Washington, D.C., who were due for their annual physical. Besides the usual exam and counseling, half received an electron-beam computed tomography (EBCT) heart scan. After a year, those in the EBCT group at elevated risk for a heart attack, indicated by calcified deposits in the coronary arteries that glowed in white on the images, were no better than high-risk members of the unscanned group at shedding their bad behaviors.
It was a blow to the "teachable moment" that physicians yearn for, when a patient is uniquely receptive to a doctor's entreaties to do the right thing. Allen Taylor, a study coauthor and director of cardiovascular research at Walter Reed Army Medical Center, believes the results can be applied to the general population. "There is this impression of military people as being in better shape than most of us, that they watch their health, that they take advice seriously," he says. "Well, this was a population that had a number of real risk factors. The military guy down the block could be anybody."
Teachable moments do exist, especially if someone is scared to death. "A good example is the patient who has had a heart attack," says Philip Greenland, chairman of the department of preventive medicine at Northwestern Memorial Hospital in Chicago, who wrote an editorial praising the study. If Marian Hayden doesn't exactly fit the category of someone shocked into action, her children do. Starting in high school, the 72-year-old Chicago resident smoked for more than 40 years before she quit. In late 1999, her smokers' cough prompted her children to push her to have a $325 walk-in lung scan they had seen advertised by Rush North Shore Medical Center in nearby Skokie, Ill. A hospital radiologist found a suspicious half-inch nodule in her right lung, and in January Hayden had a lobectomy to remove the bottom third of the lung. Twice-yearly CT scans since then have been clear. The scan very possibly saved her life.
Scan-happy. Rush North Shore still offers lung scans, but radiology chairman Leonard Berlin no longer promotes them. He has come to feel increasingly squeamish about scanning symptomless individuals. "Everybody has lung scars, and it's very difficult to differentiate them from little tiny nodules," he says.
Lung scans, it seems, are overly sensitive; they catch almost everything, no matter how small. And they aren't very specific. This combination produces a very high rate of "false positives"--findings that look like something but aren't. In the March issue of Radiology, an ongoing Mayo Clinic study reported that after two years, CT lung scans had identified suspect nodules in 1,049 out of 1,520 current or recent smokers. There were 2,832 nodules in all--nearly three per person.
How many mattered? The scans turned up 36 cancerous nodules. The other 2,796 were determined to be benign, a 98.7 percent rate of false positives. Eight individuals had to have surgery to remove nodules that then proved benign. "My profession is chest radiology, so I see all these people dying of lung cancer," says Stephen Swensen, the lead study author and radiology chair at the Mayo Clinic. "It's like a 747 full of people diving into a mountain every day. It drives me crazy. That's why I started looking into the possibility that scanning might save lives."
So far, Swensen believes, "there's no conclusive evidence that screening for lung cancer saves lives, and it could be doing more harm than good" when complications and deaths from unnecessary procedures are factored in. Leonard Berlin calls it the "cascade of testing"--one scan leading to another, followed by increasingly invasive procedures.
William Casarella, radiology chair of Emory University Hospital in Atlanta, is familiar with the phenomenon. Casarella, 65, is faithful about having a colonoscopy because of a family history of colon cancer. For his exam in November 2001, he opted for a virtual colonoscopy instead of the usual kind, in which a long tube, or endoscope, is snaked through the colon to examine the inner surface and snip suspicious growths for biopsy. He finds that it hurts--but knows that's because he declines sedation so he can get back to work quickly. "That's not very smart, I guess," he admits.
Mixed results. The colon scan was negative. But it also displayed a growth on Casarella's kidney, another on his liver, and several in his lungs. With dye pumped through his system for better contrast, a second scan showed that the one on the kidney was a harmless cyst but the liver growth wasn't. Tissue drawn from the liver didn't settle the issue. Neither did a PET (positron emission tomography) scan. A third CT scan, of the lungs, verified numerous nodules. Because the mass on the liver might indicate cancer that had spread to the lungs, three chunks of tissue were removed from the right lung.
"It was far more painful than I thought it would be," says Casarella. "It was good to go through it. Physicians talk about lung biopsies in a casual way because we don't have a good idea how painful and debilitating it is." Casarella was hospitalized for four days and needed narcotics for pain relief for two more weeks. It was five weeks before he felt normal. The final bill, completely paid by insurance, was $46,000. All the findings were negative.
The irony of virtual colonoscopy is that while scanning centers promote it as easy and comfortable, bowel cleansing is still a hated part of the routine, as with a regular colonoscopy. The day before, patients must swallow pills or several quarts of liquid that tastes like seawater, followed by frequent opportunities for bathroom reading. When 696 patients who had a virtual and a regular colonoscopy on the same day were asked whether they would repeat the exam annually, according to a survey in the latest Radiology, the percentage who said they would more than doubled if offered the possibility of no bowel prep. The patients also said ordinary colonoscopy wasn't so bad, and no more uncomfortable than virtual colonoscopy.
Peter Cotton, a gastroenterologist and medical director of the digestive disease center at the Medical University of South Carolina in Charleston, is no fan of virtual colonoscopies. With about a 30 percent chance that something will be found, he says, a biopsy must be done, so a follow-up conventional colonoscopy will be necessary anyway. "When they discover that," he says, "a lot of people figure, what the hell, I'll just go ahead." The only arguments favoring virtual colonoscopy, say experts, are the tiny possibility--1 in 750 or smaller--of perforation in the regular procedure, which can have fatal consequences, and that occasionally the cecum, a pouch at the beginning of the colon, can't be inspected.
Negative publicity for scanning centers has led some states, such as Pennsylvania and Texas, to claim they prohibit self-referral centers. State press releases even boast that they have closed offending centers. Complying with state law, however, usually requires nothing more than a referral slip that can be written by any physician--even if employed by a center. Short of a bad economy or a skeptical public, scanning centers seem poised to continue to reap profits from the worried well.
This story appears in the May 19, 2003 print edition of U.S. News & World Report.