Register for Updates | Search | Contacts | Site Map | Member Login

ICRP: Free the Annals!

View Comment

Submitted by Nobuhiko Ban, Tokyo Healthcare University
   Commenting as an individual
Document Early and late effects of radiation in normal tissues and organs: threshold doses for tissue reactions and other non-cancer effects of radiation in a radiation protection context

Among epidemiological studies in Table 2.3, none of them provides decisive evidence for the induction of circulatory disease at relatively low doses. For example, the atomic bomb survivor studies by Shimizu et al. did not find significant association with radiation for myocardial infarction. The apparent excess deaths from circulatory disease could be attributed to factors other than radiation, such as misclassification on death certificate and rheumatic heart disease of the infectious nature. Studies of US radiologists and radiologic technologists may have just detected the cohort effect. As to Mayak workers, handling of the internal dose is problematic, and the trend with the external dose alone in ischemic heart disease mortality was not significant.

Without convincing mechanism, it is hard to conclude that circulatory disease can be induced at doses below 12Gy based on these epidemiological data. Our understanding does not much differ from UNSCEAR 2006 in this context. In reality, the report of the 57th session of UNSCEAR to General Assembly says “the associated mechanisms are still unclear and the estimation of risks at low doses remains problematic”.

Extrapolating the risk from the data at high doses is possible, particularly from the observation following radiation therapy, in an analogous fashion to radiation-induced cancer. In the case of cancer, however, there is a rationale for the extrapolation since the disease is thought to be a single cell origin. Currently we do not have a comparable theoretical basis for circulatory disease, and there is no reason to estimate excess relative risk based on the linear dose response. Defining resultant 1% incidence dose as a threshold is even paradoxical.

The estimated risks are quite variable between epidemiological studies, showing confounding factors or other risk factors mainly affect the results. All that’s required now is not to derive “mock threshold” in the name of the precautious principle, but to scrutinize each epidemiological data for elucidation of the causes of the heterogeneity. In the long run, it may be needed to re-examine the basic principle that classifies the radiation effects as stochastic and deterministic.