Budoff Matthew J, Gopal Ambarish
Division of Cardiology, Department of Medicine, Los Angeles Biomedical Research Institute at Harbor-UCLA, Torrance, CA 90502, USA.
J Cardiovasc Comput Tomogr. 2007 Oct;1(2):97-105. doi: 10.1016/j.jcct.2007.04.002. Epub 2007 May 18.
Although the intent of cardiac computed tomographic angiography (CTA) is to visualize the coronary, aortic, and cardiac structures, portions of noncardiac structures are visible on the scan. Because cardiac CT scanning is primarily obtained with a small field of view (to maximize coronary visualization with highest spatial resolution), some have argued that the scans should be secondarily reconstructed to further evaluate portions of the lung, breast, and bone. The suggested benefits of a routine radiologist overread of the extracardiac structures for incidental findings have not been scientifically validated and mostly come from anecdotal experiences. The same anecdotal arguments were used to support body scanning; the idea that complete visualization of all structures will lead to earlier cancer detection and therefore better outcomes. Every center that has ever offered body scanning can show a case of early detection of lung cancer, renal cancer, and colon cancer, thus proving their efficacy. However, body scanning has been uniformly discouraged, most strongly by the American College of Radiology and other professional organizations, because of the high number of false-positive findings, low ratio of true positives to false positives, high follow-up costs, and increased anxiety, all without proof of improvement in outcomes. Similar arguments were also made for routine chest x-rays in smokers, until studies showed that earlier detection of lung masses did not lead to improvement in outcomes. Preliminary studies are showing that enlarging the field for CTA scans to look for incidental findings will suffer the same fate as body scanning and chest x-rays, as another form of screening that cannot be medicolegally justified because of severely high false-positive rates and no improvement in outcomes. Until data are available to the opposite, we should use our good judgment and restraint and not perform large-field reconstructions for the explicit purpose of screening.
