EFFECTS OF DIAGNOSTIC IRRADIATION

Abstract

The criticisms of Burch (cf. abstract 38810) of the causal effect of diagnostic irradiation and leukemia, previously answered by Wise, are reevaluated. Burch still contends that the data on time intervals, for the spondylitlcs who received radiotherapy and for the Oxford survey population, cannot be reconciled on the hypothesis that the diagnostic x rays to the trunk induced the extra M series (myeloid) leukemias in the Oxford survey. His diagram appears to show that the two distributions of intervals between irradiation and onset are appreciably different. However, his diagram is misleading, and the modes of the two underlying distributions of times to onset following inducing irradiation will probably be found to be nearly the same. The true probability distribution of time intervals is almost certainly much more sharply peaked than is shown in Burch's diagram. He also points out that large doses of irradiation to animals are associated with a relatively short latent period; however, there is no evidence that this holds good in either of the above survey populations. Since it is possible that the two underlying distributions of time intervals are very similar, the radiogenic hypothesis is strengthened even if the intervals for the diagnostic x rays domore » tend to be shorter. It is difficult to explain such a sharp peak at any time interval on the preleukemic fitness hypothesis. On the other hand the data do indicate a real association between M and L (lymphntic) leukemias and respiratory and genitourinary illnesses in the preceding 5 yr, but this association cannot explain the excess exposures in the various subgroups of M leukemias, particularly those for the 3rd and 4th years before the yr of onset. (BBB)« less