NEW YORK - CT scans in children can cause small but significant increases in the risk of leukemia and brain cancer, a new study finds.
Researchers say the results do not mean that CT scans should be avoided entirely - they can be vitally important in certain situations, like diagnosing severe head injuries - but that the test should be performed only when necessary, and with the lowest possible dose of radiation.
CT, or computed tomography, scans take X-rays from various angles and combine them to create cross-sectional images, and they involve much more radiation than traditional X-ray techniques. Concern about potential harm from the scans has grown as their use has climbed steeply; at least 4 million children a year receive scans in the United States, and researchers estimate that a third of the scans are unnecessary or could be replaced by safer tests, like ultrasound or magnetic resonance imaging, which do not use radiation.
The new study, published online Wednesday in The Lancet, a British medical journal, is based on the records of nearly 180,000 children who had scans from 1985 to 2002 in Britain. There were 74 cases of leukemia and 135 cases of brain cancer in the group. The authors estimated the radiation doses and found that the more scans the children had and the more radiation they received, the higher their risk.
Children under 15 who had two or three scans of the head had triple the risk of brain cancer compared with the general population, the researchers found, and 5 to 10 scans tripled the risk of leukemia. But the baseline risk is extremely low - 4.5 cases of leukemia per 100,000 people under 20, and 3.5 cases of cancer of the brain or central nervous system - so that even tripled, it remains small.
“The basic message for childhood cancers is that they are rare,’’ said Mark S. Pearce, the first author of the study, from Newcastle University and the Royal Victoria Infirmary.
In response, the American College of Radiology issued a statement urging parents not to refuse needed CT scans, especially for potentially life-threatening conditions like head and spine injuries, pneumonia complications, and chest infections.
The Lancet study has limitations: It is observational, meaning that the researchers just looked at what happened to patients without picking them at random to be treated or not.