Despite concerns about the risk of radiation-induced cancer among frequently scanned patients, most CT-induced cancers occur among rarely—rather than frequently—scanned young adults, according to a study published in the July issue of Journal of American College of Radiology. The findings led the authors to call for enhanced education focused on appropriate utilization of CT scanning.

Radiation biologists have surmised that 1.5 percent to 2 percent of cancers in the U.S. may be attributable to CT studies, with children and adults younger than age 35 incurring several-fold higher risks.

Given the clinical significance of the situation, Robert L. Zondervan, BA, of the department of radiology at Massachusetts General Hospital in Boston, and colleagues designed a retrospective study to determine the magnitude of frequent body CT scanning of young adults and identify patient diagnoses, exam indications, short-term outcomes and estimated radiation-induced cancer risk.

The study cohort included 25,104 patients age 18 to 35 years who underwent chest or abdominopelvic CT at three academic medical centers between Jan. 1, 2003 and Dec. 31, 2007. The researchers categorized patients as rarely (< 5 exams of a given body part), moderately (> 5 and <15 exams of a given body part) or frequently (>15 exam of a given body part) scanned.

Zondervan and colleagues assigned effective dose estimates of 7 mSv and 10 mSv to chest and abdominopelvic CT studies, respectively, and used the Biological Effects of Ionizing Radiation method to calculate expected cancer incidence and death in each patient category.

Among the study cohort, 18,592 chest and 27,040 abdominopelvic CT scans were identified. The researchers found that 95 percent of patients overall were rarely scanned while 0.3 percent of all patients were frequently scanned.

The most common diagnoses for frequent chest CT patient scanning were cancer, lung transplantation for cystic fibrosis and leukemia or lymphoma with bone marrow transplantation, reported Zondervan et al. Among abdominopelvic patients, the most common diagnoses were cancer, inflammatory bowel disease, pancreatitis and organ transplantation.

The most common indications for frequent chest CT exams were cancer follow-up, signs of infection and lung transplant evaluation. The most common indications for frequent abdominopelvic CT exams were cancer follow-up, postoperative pain or fever and abdominal pain.

The researchers projected that 36 cancers and 19 deaths from additional cancers would occur among the 23,851 rarely scanned patients, accounting for 76 percent and 73 percent of cancer cases and deaths, respectively. They predicted two cases and one death among the 70 frequently scanned patients, accounting for 3 percent each of cancer cases and deaths.

“The vast majority of young adults who undergo frequent body CT scanning (average >3 examinations/year) carry a diagnosis of locally advanced or metastatic cancer, have undergone transplantation, or both,” wrote Zondervan et al, who noted that this cohort faces a significant lifetime cancer case and death risk of 2.2 percent and 1.3 percent, respectively.

The authors stated that physicians should consider the context of the patients’ conditions, adding that defining indications to minimize frequent scanning could have the most impact in this group. Current guidelines, they explained, fail to specify frequency, duration or clinical indication for CT follow-up to monitor and assess treatment response in cancer patients. “More important, they rarely define scenarios in which imaging is unlikely to guide management.”

However, the majority of radiation-induced cancers are projected to occur among rarely scanned patients, which is the result of the very large number of patients in this category. The goal, emphasized Zondervan et al, of reducing cancer risk must be tied to reducing the number of patients scanned or ensuring that only medically necessary CT scans are performed. “Much evidence exists to suggest that up to 20 percent to 50 percent of advanced imaging procedures fail to provide information that improves patient welfare.”

Zondervan and colleagues called for enhanced education such as emphasizing appropriate utilization during residency training and continuing medical education. They also suggested radiologists take a more active role in “establishing procedures to review imaging requests and consulting with referring physicians as to the appropriateness of the indication [and developing] opportunities to educate themselves and the referring service on the clinical scenarios in which CT may or may not be indicated.”

The researchers acknowledged several limitations to the study: The academic medical centers in the study disproportionately house advanced modalities while serving patients with complex conditions, likely resulting in an overestimation of repeat scanning. In addition, dose estimates are variable and inexact and the widely used linear-no-threshold radiation-induced cancer risk model has not been proven true.