Pediatric patients who had CT scans had a higher-than-expected risk of developing developing cancer later in life, results of a large retrospective Dutch study suggested.
The risk for any type of cancer was 47% higher and ranged from 11% to more than three times higher for specific types of cancer. The calculations did not achieve statistical significance for all of cancers analyzed but did yield trends toward increased risk in all cases, reported Michael Hauptmann, MD, of the Netherlands Cancer Institute in Amsterdam, and colleagues in the Journal of the National Cancer Institute.
The authors offered a cautious interpretation of the findings, noting that “CT scans for children represent a potentially life-saving and quality of life-improving technique for many patients. In addition, the tumors evaluated here are associated with small absolute excess risks. Nonetheless, careful justification of pediatric CT scans and dose optimization, as are customary in many hospitals, are essential to minimize risks.”
The introduction of CT technology to medicine greatly enhanced diagnostic capabilities, but CT imaging exposes patients to higher levels of radiation than any other type of diagnostic radiology procedure. Children are particularly susceptible to the potential adverse effects of radiation, including radiation-induced malignancies, particularly leukemia and brain tumors, the authors noted.
Several epidemiologic studies showed an increased cancer risk in pediatric patients who have CT imaging. The results have been somewhat mixed, as some data showed increases in specific types of cancer whereas other studies demonstrated an increased risk across multiple types of cancer. However, studies consistently showed an excess cancer risk in association with CT imaging in childhood, the authors noted.
Continuing the investigation of the issue, Hauptmann’s group analyzed data from the Dutch Pediatric CT Study, which focuses on brain cancer and leukemia but also tracks other cancers. The prospective cohort analysis comprised 168,394 patients who were <18 years when they had CT scans (unrelated to cancer) during 1979 to 2012. The authors assimilated data on cancer incidence, vital statistics, and confounding factors by means of record linkage with external national registries.
For various types of cancer, investigators calculated excess relative risk (ERR) per 100 mGy organ dose of radiation and standardized incidence ratios (SIRs) derived from cancer incidence in the general Dutch population. The data comprised 262,227 CT scans performed at 42 Dutch hospitals. After excluding patients followed for less than 2 years, data analysis included 140,612 patients for leukemia analyses and 106,530 patients for analyses of solid tumors.
Investigators identified a total of 454 cancers. Considering all tumor sites, CT imaging was associated with a SIR of 1.47 (95% CI 1.34 to 1.61). Statistically significant increases were observed for cancers of the colon, bone, large bowel, soft tissues, thyroid, and central nervous system (CNS, including malignant and benign brain tumors), as well as hematolymphoproliferative cancers and nonmelanoma skin cancers.
Comparisons of expected versus observed cancer incidence showed that rates were at least twice as high as expected for cancers of the large bowel (SIR 2.05), connective tissues (SIR 2.03), and CNS (SIR 2.05), as well as nonmelanoma skin cancer (SIR 3.36). Additionally, the SIR for colon cancer was 1.95, which also achieved statistical significance. When leukemia was separated from myelodysplastic syndromes, CT imaging did not have a significant impact on incidence.
The study was supported by the European Community Seventh Framework Program and by WorldwideCancer Research.