Children undergoing computed tomography (CT) scans with cumulative radiation doses of about 50 mGy had about triple the risk for leukemia, and those who received doses of about 60 mGy had nearly triple the risk for brain cancer, according to the results of a retrospective cohort study published online June 7 in the Lancet. Radiation exposure from CT scans in childhood could triple the risk of leukemia and brain cancer.
The study authors note that CT scans are very useful diagnostically, but that children are more radiosensitive than adults and may therefore have additional potential risks for cancer from ionizing radiation. The study goal was to determine the excess risk for leukemia and brain tumors after CT scanning in a cohort of children and young adults.
Youth younger than 22 years and without previous diagnoses of cancer who first underwent CT scanning in National Health Service (NHS) centers in England, Wales, or Scotland between 1985 and 2002 were included in the analysis. The investigators estimated absorbed brain and red bone marrow radiation doses per CT scan in megagrays.
The NHS Central Registry provided data for cancer incidence, mortality, and loss to follow-up from January 1, 1985, to December 31, 2008. Use of Poisson relative risk models allowed assessment of excess incidence of leukemia and brain cancer. To exclude CT scans associated with cancer diagnosis, follow-up for leukemia started 2 years after the first CT, and for brain cancer 5 years after the first CT.
Diagnosis of leukemia during follow-up occurred in 74 of 178,604 patients, and diagnosis of brain tumor in 135 of 176,587 patients. There was a positive association between radiation dose from CT scans and leukemia (excess relative risk [ERR] per mGy, 0.036; 95% confidence interval [CI], 0.005 - 0.120; P = .0097) and brain tumors (ERR, 0.023; 95% CI, 0.010 - 0.049; P < .0001).
Compared with patients who received a radiation dose of less than 5 mGy, those who received a cumulative dose of at least 30 mGy (mean dose, 51.13 mGy) had a relative risk of leukemia of 3.18 (95% CI 1.46 - 6.94). For patients who received a cumulative dose of 50 to 74 mGy (mean dose, 60.42 mGy), the relative risk of brain cancer was 2.82 (95% CI, 1.33 - 6.03).
"Our main findings were confined to children under the age of 15 years, and showed that the risk of brain cancer is tripled with 2 or 3 CT scans, and the risk of leukemia is tripled with 5 to 10 CTs," Dr. Pearce said. He added that the risks vary with age and with radiation exposure of a particular type of CT scan to a given target organ.
Limitations and Implications
The investigators suggest that applying the dose estimates for 1 head CT scan before the age of 10 years would translate into about 1 excess case of leukemia and 1 excess brain tumor per 10,000 patients.
Limitations of this study include the lack of data about the reasons for CTs, exposure to radiography, and other clinical variables.
This study should reduce the debates about whether risks from CT are real. Pearce and colleagues confirm that CT scans almost certainly produce a small cancer risk.
The US National Cancer Institute and UK Department of Health funded this study. The study authors have disclosed no relevant financial relationships. Dr. Einstein is supported by the US National Heart, Lung, and Blood Institute; a Victoria and Esther Aboodi Assistant Professorship; and the Louis V. Gerstner Jr Scholars Program. He has also received research support from GE Healthcare and Spectrum Dynamics and has been a consultant for Kyowa Hakko Kirin Pharma and the International Atomic Energy Agency.
Lancet. Published online June 7, 2012.