WEDNESDAY, May 31, 2006 (HealthDay News) -- The latest version of computerized X-ray scanning does a fine job of detecting potentially fatal blood clots in the lungs, but the technology still leaves ample room for a doctor's clinical judgment about individual patients, a new study suggests.
Multi-detector computed tomography gives multiple slices, or images, of the lungs to help doctors diagnose pulmonary embolism, and it does so better than older, single-slice machines. However, the report in the June 1 issue of the New England Journal of Medicine indicated that the new technique may be too good at detecting possible problems, which is where the doctor's judgment comes in.
In the study, 824 people suspected of having blood clots in the lung underwent the procedure. In terms of determining who did not have such a clot, the method was 96 percent accurate. But in 17 percent of cases, the test identified a clot in someone who was later found not to have one.
"Exactly why this happens is not known," said Dr. Charles A. Hales, chief of pulmonary and critical care at Massachusetts General Hospital, who took part in the study. "The scan may be finding clots that are too small to matter. Or it may be showing a defect in a blood vessel, something that can be difficult to interpret in a small vessel."
When a scan identifies a possible lung clot, standard procedure is to prescribe an anti-clotting drug such as heparin. But the high percentage of clots that weren't clots in the multi-detector computed tomography study means a physician should think twice about anti-clotting therapy in some cases, Hales said.
When the scan reads positive but the doctor thinks the probability of a clot in a certain patient is low, "you should do further tests rather than put someone on anticoagulants," he said. For example, there might be call for a scan of blood vessels in the legs, where most lung clots originate.
That strategy reflects a basic change in thinking about pulmonary embolisms, which occur in at least 300,000 Americans each year, said Dr. H. Dirk Sostman, a professor of radiology at Weill-Cornell Medical College in New York City, and another member of the research team.
"Back 10 or 15 years ago, it was widely believed that physician judgment was inadequate to deal with this problem," Sostman said. "It was regarded as entirely an imaging question."
Sostman had high praise for the new scanning technology, which he said "has revolutionized a lot of clinical areas," such as diagnosis of brain circulation problems. But, he added, "research has shown that you can get a huge amount out of an informed clinician's judgment." In recent years, that judgment has been codified in programs "that are teachable, not just one doctor's 20 years of information," Sostman said.
The study results show that "the interaction of clinical judgment and the imaging test is the way to go," he said.
Dr. John G. Weg is professor of pulmonary and critical care medicine at the University of Michigan, and another study participant. He said it's not that a doctor should ignore a scan result that conflicts with his assessment of a patient. "But if the clinical impression of the doctor disagrees with the test finding, the doctor should go back over the information because there is a good likelihood that the test might be wrong," Weg said.
"This is very reassuring to the doctor," Sostman added. "The facts of life are that no technology is so good that a physician's judgment of an individual patient doesn't improve things."
Or, as Weg put it, "It's good that we can still think."
You can learn about pulmonary embolism from the National Library of Medicine.