CT scans for lung cancer produce many false positives and much anxiety, researchers say
Smokers who have a single CT scan to look for lung cancer have a one in five chance of having a false positive diagnosis, while those who have two such scans have a one in three chance of a false positive, federal researchers reported Monday. The false positives, which indicate that a tumor is present when there actually is none, often lead to unnecessary invasive medical procedures and produce high levels of anxiety among patients who fear they may have a tumor, the researchers reported Monday in the Annals of Internal Medicine. The false positives are twice as likely with a CT scan as they are with a conventional chest X-ray, the team said.
Early diagnosis of lung cancer--as is the case with virtually all types of tumors--leads to a much higher cure rate. In the past, smokers and workers with high exposure to carcinogenic chemicals have used chest X-rays to find out whether they have a small tumor. CT scans, however, can detect much smaller tumors,so there has been a growing trend toward their use in such patients--even though no national body recommends they be used and most insurance companies refuse to pay for the expensive procedure because it has not been demonstrated to be effective. Several studies have produced conflicting results about whether such scans actually save lives, and many researchers are concerned because the high doses of X-rays used in the CT scans can themselves cause cancer.
The National Cancer Institute is conducting a major clinical trial in nearly 60,000 patients to determine whether CT scans do save lives. Results from the study should be available in another year or two. But in preparation for that study, Dr. Jennifer M. Croswell of the National Institutes of Health's Office of Medical Applications of Research and her colleagues performed a pilot study on 3,190 patients who were randomly assigned to receive either a CT scan or a chest X-ray. The patients, all smokers or ex-smokers between age 55 and 74, received one test at the beginning of the study and a second one a year later. They were then followed for another year. Preliminary results were reported last May at an Orlando meeting of the American Assn. for Cancer Research.
Among those who received CT scans, 21% had a false positive after one scan and 33% after two scans. For those who received chest X-rays, 9% had a false positive after one and 15% had a false positive after two. More than half of those with a positive finding had a follow-up scan or chest X-ray. About 7% in the CT scan group and 4% in the chest X-ray group had an unnecessary invasive medical procedure, typically bronchoscopy.
The team could not determine whether the scans actually saved lives. That will come in the larger study.
But because an estimated 159,000 Americans die of lung cancer each year, it is crucial to find a way to detect them earlier. Researchers are looking at a variety of other techniques, including sniffing for certain chemicals in a patient's breath and searching for unique chemical markers in their blood. But until more accurate tests are developed, physicians will have to rely on these current flawed tests.
-- Thomas H. Maugh II