Early life environmental exposures and birth characteristics and childhood lymphoma risk: a systematic review and meta-analysis.

Abstract

Given the limited evidence on causes of childhood lymphoma, this systematic review and meta-analysis synthesized data on suggested factors associated with the risk of Hodgkin's lymphoma (HL) and non-Hodgkin's lymphoma (NHL). PubMed, Web of Science, and Embase were searched for peer-reviewed studies. Eligible articles were appraised using Joanna Briggs Institute tools. Random-effects meta-analysis reported pooled effect sizes (ES) with 95% confidence intervals (CI), combining cohort and case-control studies' results as primary analysis. Sixty-one articles met the inclusion criteria. For HL, factors such as gestational age, gestational size, mode of delivery (caesarean), and birth order were assessed in only a few cohort articles, showing no associations. High birthweight (>4000 g) suggested increased HL risk (ES = 1.20, 1.02-1.41). Younger maternal age (≤25 years) showed no association in cohort studies but evidence of increased risk in case-control studies (ES = 1.81, 1.25-2.63). Maternal smoking (≥10 cig/day) showed no association, and parental occupational and household pesticide exposures were not associated with HL risk. For NHL, no consistent associations with perinatal factors were identified. Maternal smoking (ES ranging from 1.31-1.38) and paternal smoking (1 to 10 cig/day; ES = 1.56, 1.09-2.24) showed some evidence of increased risks. Household paternal pesticide exposure (ES = 1.74, 1.24-2.45) and maternal insecticide exposure (ES = 1.97, 1.42-2.73) were also associated with increased risks. A modest number of eligible articles for most potential risk factors resulted in limited evidence and shows the need for more aetiological studies in international collaborations.