The present study was conducted to test the hypothesis that exposure t
o influenza in pregnancy increases the risk of tumour of certain type
in childhood. Children ages 17 years or less diagnosed in Greece with
brain tumours or neuroblastomas from 1982 to 1993 (n = 94) were contra
sted to 210 controls selected from the same hospitals. Mothers of thes
e children were interviewed about a variety of possible etiologic fact
ors. The prevalence of influenza in Greece for each year during the pe
riod 1984-1992 was also compared with the number of children born duri
ng the same year who subsequently developed brain tumour or neuroblast
oma. The results indicate a significant association between influenza
in pregnant women and occurrence of tumour in index child (OR: 3.15, 9
5% CI: 1.13-8.77). These results persisted when adjustment for potenti
al confounding factors was made. The findings should be interpreted ca
utiously because of lack of serologic documentation of information abo
ut infection obtained in interviews. A positive correlation (r = 0.74)
of the number of tumour births by year of birth with the prevalence o
f influenza during the same year was also noted. This exploratory stud
y is one of the few case-control studies of the epidemiology of childh
ood tumours in children, and the results suggest directions for future
epidemiologic studies in this relatively uncharted field.