Do CT scans increase childhood cancer risk? A UF researcher has the answer

A recent study links exposure to radiation from medical imaging to a small-but-significant risk of blood cancers among children and adolescents.

But do not panic. The study concludes the benefits of medical imaging outweigh the minimal risks.

Funded by the National Cancer Institute, the study will help medical personnel make informed decisions about using imaging on children. The study concluded that while ionizing radiation is a carcinogen, the benefit-to-risk ratio favors CT imaging of children when imaging is justified and the technique minimizes adverse effects.

The paper, “Medical Imaging and Pediatric and Adolescent Hematologic Cancer Risk,” was published last week in the New England Journal of Medicine. For his part of the study, Bolch used virtual patient anatomic models to reconstruct bone marrow doses in more than 3.7 million children who underwent CT imaging between 1996 to 2016.

“We used a library of 3D anatomic whole-body computerized patient models developed in the early 2010s under a contract with the National Cancer Institute,” said Wesley Bolch, Ph.D., a distinguished professor in biomedical and radiological engineering with the University of Florida's J. Crayton Pruitt Family Department of Biomedical Engineering.

In Bolch’s Advanced Laboratory for Radiation Dosimetry Studies, researchers developed hundreds of models representative of U.S. adults and children across all combinations of ages, heights and weights.

“We call this an organ dose reconstruction,” said Bolch, who is also a member of the UF Health Cancer Center.

The study is unique because it directly measures leukemia risks in these pediatric patients. Previously, researchers used established cancer-risk models, many of which were based on data from the atomic bomb survivors in 1945 Japan. Needless to say, Bolch noted, diagnostic X-ray exposures are vastly different from atomic weapon radiation fields.

“This is the very first study of its kind in the U.S. and Canada, and the very first study of cancer risks in children undergoing medical imaging where each patient was considered in a unique manner regarding their sex, body size and medical imaging exposure technique factors,” he said.

While CT imaging contributes a greater fraction of total radiation, doses from nuclear medicine, radiography and fluoroscopy were also included for bone marrow-dose calculations.

“Dr. Bolch’s role in this landmark study highlights UF BME’s leadership in making medical imaging safer for children. It demonstrates UF’s dedication to research that safeguards patients and informs global healthcare practices,” said BME chair Cherie Stabler, Ph.D.

In the study, the highest CT doses to bone marrow were seen in head-and-neck imaging, where the average dose was 30.8 milligray (a unit measuring ionizing radiation deposited in tissue). CT imaging of the head – one of the more common pediatric CT scans – showed an average dose of 13.7 milligray.

The incidence of hematologic cancers by age 21 years was 0.3% among those children exposed to bone marrow doses more than 30 milligray. However, fewer than 1% of the 3.7 million children in this study had cumulative doses exceeding 30 milligray.

Plus, CT imaging doses today are much lower, and imaging systems are much faster than in the late-1990s and early-2000s.

This research comes 25 years after a Columbia University research paper made the link between leukemia and some radiology scans, thus scaring “every mother in this country,” Bolch said.

The key issue in that study showed imaging technologists and radiologists were not making adjustments to X-ray techniques that explicitly considered the size of the patient.

“Consider that you've just imaged an obese a male with a high-intensity and high energy X-ray beam, and now a petite 7-year-old girl becomes the next patient to be imaged. In the late 1990s, very few clinics would adjust the X-ray energies and intensities from the previous adult patient. In this case, the girl received a much larger amount of imaging than was really needed to form a diagnostic-quality image,” Bolch said. “This paper scared a lot of people, but it really was a great service because it says, ‘Oh, we're doing this wrong.’”

Starting in the early 2000s, physicians adjusted the energy and intensities of the X-ray beam based on the size of the patient. CT system manufacturers, Bolch said, also made technological improvements to lower patient CT doses.

The paper’s lead authors are radiologist and epidemiologist Rebecca Smith-Bindman, M.D., from the University of California, San Francisco, and biostatistician Diana Miglioretti, Ph.D., from UC, Davis.

“Their work involved collecting and organizing the medical records that showed what imaging exams were performed on the children, when they occurred, how they were acquired and what modality was used – CT, radiography, nuclear medicine or fluoroscopy,” Bolch said.

They also led the team that linked these patients to cancer registries that indicated which patients later developed bone marrow cancers.

“This is where my laboratory came in," Bolch said. “We ran computer simulations of these imaging procedures to provide estimates of bone marrow radiation doses for each child, for each form of medical imaging and for each imaging examination.

“Everybody went into action to figure out what is the appropriate lowest dose of radiation that would give us a good image. These risks are low, and when justified by the imaging physician, patient benefits, such as disease detection, will greatly outweigh these very small risks.”

Reference:

Rebecca Smith-Bindman, Medical Imaging and Pediatric and Adolescent Hematologic Cancer Risk, New England Journal of Medicine, DOI: 10.1056/NEJMoa2502098.