Radiation safety and cancer risks in pediatric computed tomography: a systematic review of large-scale cohort studies

Background The increasing use of computed tomography (CT) in pediatric populations has raised concerns regarding radiation-induced malignancies due to higher radiosensitivity and longer post-exposure lifespan in children. Objective This study aimed to synthesize evidence from large-scale cohort studies on the association between pediatric CT exposure and subsequent cancer risk, with a focus on leukemia and brain tumors. Methods A systematic review was conducted following PRISMA 2020 guidelines. Literature searches were performed in PubMed and the Cochrane Library from database inception to 30 June 2025. Seven large cohort studies, including more than 11 million individuals from Europe, Asia, and Australia, were included. Data on radiation exposure, cancer outcomes, follow-up, effect measures, and methodological quality were extracted independently by two reviewers. Results A consistent dose-response relationship was observed across studies. Excess relative risk (ERR) per mGy ranged from 0.033 to 0.045 for leukemia and from 0.006 to 0.023 for brain tumors in studies reporting organ-dose estimates. The risk was higher in children exposed at younger ages, particularly under 5 years, and increased with cumulative exposure. Scan-count analyses also suggested higher incidence with repeated CT, although scan count is a proxy for dose and should not be interpreted as equivalent to organ-dose ERR. Although indication bias was identified, adjusted and lagged analyses generally supported a persistent association. Conclusion Pediatric CT exposure is associated with a small but measurable increase in long-term cancer risk. These findings should be interpreted alongside the diagnostic benefit of CT in trauma, oncology, and life-threatening illness. Clinically justified CT remains essential; the results support strict justification, dose optimization, and the use of non-ionizing alternatives when they can answer the clinical question.