Mrs. Smith (not her real name) fidgeted in her chair in my examination room as I scanned the radiology report she had given me. She had visited the emergency room the previous evening with severe abdominal pain that had eventually been diagnosed as gastritis, or swelling of the stomach lining due to a virus. During her evaluation, the ER physician had ordered a CT scan of her abdomen and pelvis. Although Mrs. Smith's liver and intestines appeared normal, the radiologist had noted a tiny mass on one of her kidneys.
The report stated that the mass was consistent with a harmless cyst, but concluded with a statement that was all too familiar to me: "Cannot rule out malignancy. Clinical correlation required." Translation: it was almost certainly nothing serious, but there was a very small chance that it might be cancer, and now it was my job to make sure it wasn't. But further investigation of this incidental finding, which had no relationship to Mrs. Smith's original symptoms, would involve a painful biopsy, and if the biopsy was inconclusive, surgery to remove her kidney. In similar situations with other patients, I had suggested the alternative of regular monitoring with additional scans to make sure that the mass wasn't growing; however, this option requires that a patient live each day with the anxiety of not knowing if she has cancer.
That episode happened almost a decade ago. Yet the dilemma that my patient faced is, if anything, much more common today. A study published recently in the journal Radiology found that children visiting U.S. emergency rooms had five times as many CT scans in 2008 as in 1995. By 2008, 6 percent of pediatric ER visits involved a CT scan. The same research group, led by Dr. David Larson at Cincinnati Children's Hospital Medical Center, previously found an even greater rise in scanning during adult ER visits, with 25 percent of patients age 65 and older, and 12 to 16 percent of younger adults, getting a CT scan in 2007.
In addition to increasing risks associated with radiation exposure, all of those CT scans turn up an awful lot of "incidentalomas," the term that doctors use for incidental findings that could be (but probably aren't) cancer. A study published last year in the journal Archives of Internal Medicine found that nearly 40 percent of CT and MRI scans performed for research purposes at the Mayo Clinic from January through March 2004 turned up at least 1 incidental finding. In the 35 patients in whom doctors took further action (additional testing, specialist consultation, or surgery), only 6 were judged by researchers to have clearly benefited from an investigation, while in the rest there was no clear benefit or clear harm, such as complications from surgery for a benign tumor. Of all types of scans, CT of the abdomen and pelvis - the very same scan that my patient got - was the most likely to turn up an incidental finding.
In fact, the American College of Radiology has become so concerned about the problem of incidentalomas on CT scans of the abdomen and pelvis that they recently published detailed guidance for clinicians about how to approach such findings. "Subjecting a patient with an incidentaloma to unnecessary testing and treatment can result in a potentially injurious and expensive cascade of tests and procedures," the radiology group warns, advising that doctors carefully consider an individual patient's risk for cancer in deciding whether or not to recommend further evaluation.
So what can you do to reduce the chance you will be harmed by an incidentaloma? Three experts in diagnostic medicine at the the Dartmouth Institute for Health Policy and Clinical Practice recently recommended that patients who are told about an incidental finding always seek a second opinion to verify that the radiologist's interpretation of their scan is correct, and understand that clinical observation of an incidentaloma is often a safer option than more testing or surgery. Also, they advise that patients adopt a "healthy skepticism" about testing and only consent to scans that are absolutely necessary to establish a diagnosis or plan of action, rather than ordered “just to be sure.”
To be honest, I don’t remember what Mrs. Smith chose to do about her incidentaloma. If I saw her as a patient today, I would definitely consult a second radiologist to be sure that the kidney mass was actually there. If it was, I would probably recommend a wait-and-see approach, given that additional testing could create more risk than reward.
And if I had the power to turn back to clock and warn my patient before she arrived in the ER, I’d advise her to ask the physician there if the CT scan was really needed at all.