Applied Radiology

Dr. White is the Director of Thoracic Imaging, Diagnostic Imaging Department, University of Maryland Medical Center, Baltimore, MD.He is also a member of the editorial board of this journal.

For those of us who have practiced radiology for more than 10 years, the rapid increase in the volume of our chest computed tomography (CT) business has been truly extraordinary. When I arrived at the University of Maryland as a neophyte attending thoracic radiologist in 1991, we would routinely pass our days contemplating the diagnoses of perhaps 2 or 3 chest CT scans. Fast-forwarding to 2006, we typically interpret 40 or more CT studies at a much less leisurely pace.

This remarkable increase in chest CT volume has been facilitated by rapidly advancing technological capabilities, such as multidetector CT, and the attendant expansion of indications for pulmonary CT. For example, high-resolution CT was in its infancy in 1991; now, it is considered a standard part of the CT armamentarium. Perhaps the largest contributor to the increased volume of studies has been the use of chest CT to evaluate patients with suspected pulmonary embolism. This indication did not exist in 1991, but it now accounts for approximately 20% of our total CT volume.

This flood of CT studies naturally raises the question of whether clinicians have come to rely too heavily on the "truth" provided by CT images. Those who decry CT overutilization suggest that the art of history taking and physical diagnosis is being cast aside in favor of objective data whose value may be overrated. Even "simpler" imaging technology is often abandoned. Today, many patients with suspected pulmonary embolism proceed directly to CT without even undergoing a chest radiograph.

In actuality, both defenders and critics of the current level of CT utilization are probably oversimplifying a highly nuanced set of circumstances. Critics correctly point to the arbitrary nature of many of the clinical requests for chest CT, some of which are often motivated by medicolegal considerations. Defenders recognize that greater experience can lead to improved diagnostic acumen. The reality is, however, that radiologists are hesitant to serve as utilization gatekeepers for fear of alienating referring clinicians.

Ultimately, we must rely on rigorous scientific inquiry to determine the proper use of pulmonary CT scanning. Too often, imaging decisions are made on an empiric basis. The current debate on overutilization of CT in general, and chest CT in particular, speaks strongly to the need for well-designed outcomes research to define the appropriate uses for our increasingly powerful technologies. Prospective studies in large patient populations are needed to determine appropriate imaging strategies-considering risk factors, laboratory studies, and clinical signs and symptoms-that can guide rational use of this technology to optimize accuracy of diagnosis, limit costs and radiation exposure, and minimize the potential for medicolegal actions.