By Jay Furst
Computed tomography (CT), which uses specialized X-rays to create images of areas inside the body, has revolutionized medicine in the nearly 50 years since it became commonly available. From kidney stones to cancer, CT has become an indispensable diagnostic tool.
In 1980, an estimated 3 million CT scans were made in the U.S. By 2007, the estimated total was closer to 60 million, and by 2015 the estimate was 80 million. A CT scan combines a series of X-rays from different angles to create cross-sectional images that can be used to discover illness or injury, guide surgery and treatment plans, and monitor outcomes.
Despite its common use, the exposure of patients to ionizing radiation from CT is incompletely understood, says Konrad Stopsack, M.D., formerly a Mayo Clinic internist who now is at Memorial Sloan Kettering Cancer Center in New York City. For that reason, he and James Cerhan, M.D., Ph.D., chair of Mayo Clinic's Department of Health Sciences Research, initiated a retrospective study on the cumulative doses of ionizing radiation from CT that patients in Olmsted County received over a 10-year period.
"We wanted to learn more about exposure to ionizing radiation from CT among adults, as well as which parts of the population were most affected, and for what clinical indications CTs were obtained," says Dr. Stopsack, lead author of the study, which is posted online at the Mayo Clinic Proceedings website and will be published in the journal's October issue. "Previous studies have looked at these questions but either were unable to track people over a long period of time or only looked at those patients who repeatedly came back to the same hospital, so were not necessarily representative of the general population."
The study was made possible by using the resources of the Rochester Epidemiology Project, the vast treasure trove of medical records that have been logged in Rochester and Olmsted County since 1966, and it could not have been done as efficiently elsewhere, says Dr. Cerhan, an epidemiologist and the Ralph S. and Beverley Caulkins Professor of Cancer Research at Mayo Clinic.
"The Rochester Epidemiology Project allows the combination of access to detailed clinical data over a long period of time on a geographically defined population," he says.
Dr. Stopsack, who graduated from Mayo Clinic's Internal Medicine Residency Program in 2017 and conducted the research with Dr. Cerhan while still at Mayo, says the study reviewed all CT examinations performed in Olmsted County from 2004 to 2013. Mayo Clinic, Olmsted Medical Center and affiliated hospitals perform CT scans, and the institutional review boards of the two institutions approved the study.
"The number of people who end up getting a CT scan is pretty impressive," Dr. Stopsack says. "Half the population we sampled received a CT within 10 years. That's quite a lot."
Of the 54,447 adults whose de-identified medical records were examined, almost half — 26,377 — underwent at least one CT. The total number of CTs performed was 107,961 during the 10-year period, with the largest group (44.2 percent) of patients having 2-4 CTs during that period.
The study relied on administrative claims data to determine which CTs people had and then to calculate the radiation dosage the patient received.
"We found quite striking differences in radiation dosage," Dr. Stopsack says. "Patients who were older or had very low or very high body mass index, for example, received higher radiation doses over time." Also factors were education and race; educational levels both lower and higher than four-year college degrees were associated with higher doses, and African-Americans received higher doses than whites.
"The association of age, smoking and perhaps body mass index with CT is not very surprising, even if most of that had not been described previously in research," says Dr. Stopsack. "However, it is quite striking that people with lower educational attainment and African-Americans accumulated higher doses."
Dr. Stopsack believes that this disparity may reflect poorer health on average across these specific groups.
Of 600 CTs among 200 patients who had accumulated high doses over one decade, 70 percent were for restaging or assessing the extent of cancerous tumors and lymphoma, abdominal pain, infection, kidney stones, follow-up of (potentially cancerous) nodules or masses, and to check for pulmonary embolism.
"That's another striking number," Dr. Stopsack says. "Two percent of the population sampled received really high estimates doses over 10 years, in excess of 100 millisieverts (measurement of radiation exposure)."
In terms of potential implications for clinical care, Dr. Stopsack says the high cumulative doses potentially could be reduced if CTs were ordered for a shorter list of indications, or if lower doses were exam were used.
"Replacing CT with ultrasound in patients who may have kidney stones, for example, is one area that has been successfully studied in a randomized trial," he says. "Another is follow-up CTs for lymphoma patients. Our results underline that CTs for these patients are relatively big contributors to cumulative radiation."
Other dose-reduction strategies noted in the study are decreasing doses per exam and more generally, addressing variations in dose between institutions.
Most important, though, is the value of CT to diagnose illness and and treat patients, the researchers say.
"When a CT is clinically indicated, the benefit is almost always greater than any potential harms, including from radiation," Dr. Stopsack says.