In a population-based case-control study that included more than 85,000 participants, researchers found a ninefold increased risk of intracranial tumors among children who received four or more CT scans.
The results “indicate that judicious CT usage and radiation-reducing techniques should be advocated,” Yu-Hsuan Joni Shao, PhD, professor of biomedical informatics at Taipei (Taiwan) Medical University, and colleagues wrote.
The study was published in the Canadian Medical Association Journal.
Dose-response relationship
The investigators used the National Health Insurance Research Database in Taiwan to identify 7,807 patients under age 25 years with intracranial tumors (grades I-IV), leukemia, non-Hodgkin lymphomas, or Hodgkin lymphomas that had been diagnosed in a 14-year span between the years 2000 and 2013. They matched each case with 10 control participants without cancer by sex, date of birth, and date of entry into the cohort.
Radiation exposure was calculated for each patient according to number and type of CT scans received and an estimated organ-specific cumulative dose based on previously published models. The investigators excluded patients from the analysis if they had a diagnosis of any malignant disease before the study period or if they had any cancer-predisposing conditions, such as Down syndrome (which entails an increased risk of leukemia) or immunodeficiency (which may require multiple CT scans).
Compared with no exposure, exposure to a single pediatric CT scan was not associated with increased cancer risk. Exposure to two to three CT scans, however, was associated with an increased risk for intracranial tumour (adjusted odds ratio, 2.36), but not for leukemia, non-Hodgkin lymphoma, or Hodgkin lymphoma. Exposure to four or more CT scans was associated with increased risk for intracranial tumor (aOR, 9.01), leukemia (aOR, 4.80), and non-Hodgkin lymphoma (aOR, 6.76), but not for Hodgkin lymphoma.
The researchers also found a dose-response relationship. Participants in the top quintile of cumulative brain radiation dose had a significantly higher risk for intracranial tumor, compared with nonexposed participants (aOR, 3.61), although this relationship was not seen with the other cancers.
Age at exposure was also a significant factor. Children exposed to four or more CT scans at or before age 6 years had the highest risk for cancer (aOR, 22.95), followed by the same number of scans in those aged 7-12 years (aOR, 5.69) and those aged 13-18 years (aOR, 3.20).
The authors noted that, although these cancers are uncommon in children, “our work reinforces the importance of radiation protection strategies, addressed by the International Atomic Energy Agency. Unnecessary CT scans should be avoided, and special attention should be paid to patients who require repeated CT scans. Parents and pediatric patients should be well informed on risks and benefits before radiological procedures and encouraged to participate in decision-making around imaging.”
True risks underestimated?
Commenting on the findings, Rebecca Smith-Bindman, MD, a radiologist at the University of California, San Francisco, and an expert on the impact of CT scans on patient outcomes, said that she trusts the authors’ overall findings. But “because of the direction of their biases,” the study design “doesn’t let me accept their conclusion that one CT does not elevate the risk.
“It’s an interesting study that found the risk of brain cancer is more than doubled in children who undergo two or more CT scans, but in many ways, their assumptions will underestimate the true risk,” said Dr. Smith-Bindman, who is a professor of epidemiology and biostatistics at UCSF. She said reasons for this include the fact that the investigators used estimated, rather than actual radiation doses; that their estimates “reflect doses far lower than we have found actually occur in clinical practice”; that they do not differentiate between a low-dose or a high-dose CT; and that that they include a long, 3-year lag during which leukemia can develop after a CT scan.
“They did a lot of really well-done adjustments to ensure that they were not overestimating risk,” said Dr. Smith-Bindman. “They made sure to delete children who had cancer susceptibility syndrome, they included a lag of 3 years, assuming that there could be hidden cancers for up to 3 years after the first imaging study when they might have had a preexisting cancer. These are decisions that ensure that any cancer risk they find is real, but it also means that the risks that are estimated are almost certainly an underestimate of the true risks.”
The study was conducted without external funding. The authors declared no relevant financial relationships. Dr. Smith-Bindman is a cofounder of Alara Imaging, a company focused on collecting and reporting radiation dose information associated with CT.
A version of this article first appeared on Medscape.com.