Estimated Cancer Risk with Use of CT in Pediatrics

Over the past 20 years, the use of computed tomography (CT) in pediatrics has been increasing. In 2001, 5% to 11% of the 85 million CT scans performed in the United States were on children. The use of CT has improved diagnostic capabilities, but there are risks associated with the procedure. The ionizing radiation doses associated with CT are 100 to 500 times higher than with conventional radiography, ranges that have been linked to an increased risk of cancer.

The increased risks are particularly worrisome for children because compared with adults they are more sensitive to radiation-induced carcinogenesis and have more years of life remaining when cancer could develop. Previous studies have found increased risk of leukemia in children who received an active bone marrow dose from CT of ≥30 mGy, and an increased risk of brain cancer in children who received a brain dose of ≥50 mGy.

There have been recommendations to lower doses in children, but whether those recommendations have been widely implemented is unclear. Researchers recently conducted a study to identify trends in CT imaging among pediatric enrollees of 6 diverse healthcare systems. The retrospective observation study of a random sample of CT scans calculated radiation exposure and lifetime attributable risks of cancer and projected the number of future cancers expected to result from the use of CT in pediatrics. Study results were reported in JAMA Pediatrics [2013;167(8):700-707].

The use of CT was examined for children <15 years of age from 1996 to 2010. During that period, between 152,419 and 371,095 children were included each year, accounting for a total of 4,857,736 child-years of observation. Of the study cohort, half were girls and 29% were <5 years of age.

Between 1996 and 2005, the use of CT in pediatrics increased. CT use remained stable between 2005 and 2007, and then began to decline. For children <5 years of age and those 5 to 14 years of age, the rates between 1993 and 2003 were similar; in 2004, the rates between those age groups diverged, with a greater increase in the number of scans among older children.

Effective doses were highest for abdomen/pelvis scans; mean doses increased from 10.6 mSv among children <5 years of age to 14.8 mSv among children 10 to 14 years of age. Effective doses tended to increase with advancing age for chest and spine scans, but decreased with age for head scans. An effective dose of ≥20 mSv was delivered by 14% to 25% of abdomen/pelvic scans, 3% to 8% of chest scans, and 6% to 14% of spine scans, depending on age.

For head and spine scans, the projected lifetime attributable risk of solid cancer decreased, with a less consistent relationship for abdomen/pelvis and chest scans. Solid cancer risks were higher for girls than for boys and tended to be highest for abdomen/pelvis scans: 25.8 to 33.9 projected cases per 10,000 CT scans for girls versus 13.1 to 14.8 cases per 10,000 CT scans for boys. Solid cancer risk was lowest for head scans for children ≥5 years of age: 1.1 to 2.4 cases per 10,000 CT scans.

Nationally, 4 million pediatric CT scans of the head, abdomen/pelvis, chest, or spine performed each year are projected to cause 4870 future cancers. Reducing the highest 25% of doses to the median might prevent 43% of these cancers, the analysis found.

In summary, the researchers noted, “The increased use of CT in pediatrics, combined with the wide variability in radiation doses, has resulted in many children receiving a high-dose examination. Dose-reduction strategies targeted to the highest quartile of doses could dramatically reduce the number of radiation-induced cancers.”