Pre-morbid height and weight as risk factors for development of central nervous system neoplasms.

Information on pre-morbid height and weight from a national screening of tuberculosis between 1963 and 1975 was linked with the registrations in the population-based Norwegian Cancer Registry. For each case with a primary central nervous system (CNS) neoplasm, 10 matched controls were taken from the non-cases. Analyses were done by a Cox regression model for the total group of CNS neoplasms and the various histological groups. In the case of the total group, height emerged as a significant risk factor for both sexes. Within each sex, a similar trend was found for each histological group although statistical significance was retained only for glioblastoma among males and for other types of glioma (astrocytoma, oligodendroglioma, mixed glioma and ependymoma) for females. A significant negative association was revealed between Quetelet's index (weight/height squared) and 'other gliomas' in females. No association with body mass, as expressed by Quetelet's index, was found for patients with meningioma.