Abstract
The use of computed tomography (CT) imaging is clearly beneficial for millions of patients. However, the potential adverse health effects, particularly cancer, of ionising radiation exposure from CT early in life are an issue of growing concern in the radiological protection, medical, and public health communities. Although efforts to quantify these effects have been conducted, the precision and accuracy of reported risks needs confirmation. EPI-CT, a European collaborative epidemiological study, was set up to quantify risks from paediatric CT to optimise paediatric diagnostic protocol. The study, coordinated by the International Agency for Research on Cancer, was designed as a multi-national cohort study of children and young adults who underwent CT scanning for long-term follow-up. It combined data from existing and extended cohorts in France, the UK, and Germany, and from new cohorts assembled in Belgium, Denmark, the Netherlands, Norway, Spain, and Sweden using a common protocol. A flexible dose reconstruction approach that can accommodate collection of data from historical sources (prior to 2000) and automatically extract data from the Digital Imaging and Communications in Medicine headers of recorded images available in the Picture Archiving Communication System was developed. Individual organ dose estimates for each child were derived from Monte-Carlo-based radiation transport calculations using hybrid phantoms of different sexes and ages. To account for uncertainties due to missing input data, a simulation method that maintains correlations of doses for persons within subgroups with similar exposure attributes and simulates uncertain dose-model parameter values was used. Simulation studies to evaluate the potential impact of a range of potential confounders (e.g. underlying medical conditions, socio-economic status, missing medical procedures performed outside of participating hospitals) on risk estimates were conducted based on data from some EPI-CT countries and/or reasonable scenarios. In total, 1,170,186 patients (before censorship) were enrolled in the national cohorts. Most patients (75%) had only undergone one CT scan and 29% of all patients were aged <5 years at the time of their first CT examination. The median duration of follow-up was 8 years for the entire cohort, although this varied between countries. Overall, the follow-up accounted for nearly 10 million person-years. This study received partial funding from the European Commission 7th Framework Programme under Grant Agreement No. 269912.