Canadian study of cancer following multiple fluoroscopies

Records of patients treated in Canadian Sanatoria during the period 1930-1952 have been linked with the National Death Index maintained by Statistics Canada to provide fact and cause of death information for the years 1950-1980. Of 31,710 women known to be under observation on January 1, 1950, 13,795 were exposed to fluoroscopy for control of collapse therapy, while the remaining 17,915 were unexposed. The unexposed had the similar mortality from breast cancer to that expected from general population rates. Those exposed to fluoroscopy had increasing mortality with increasing radiation dose to the breast, the best fit to the dose-response curve being a quadratic function. Estimates of risk at doses above 300 rads were largely derived from patients treated in Nova Scotia, where fluoroscopy was administered antero-posterior, as distinct from the more usual postero-antero practiced elsewhere. There is evidence of age-related susceptibility to radiation-induced breast cancer. The risk was maximal for those who first received fluoroscopy in their teens or twenties, but it was similar to expectation for those first exposed at age 30 or more. The latent period from onset of exposure to first increase in the death rate from breast cancer was 15 years for those first exposed at ages 10-24 and 10 years for those first exposed at ages 25 or more. However, these periods coincide with years when mortality from breast cancer normally rises and may therefore not be a true latent period effect. Estimates of predicted excess deaths from breast cancer per million women first exposed at ages 10-29 vary depending on the model used to represent the effect and whether or not data from the Nova Scotia Series are included in the computations