Maternal and childhood consumption of coffee, tea and cola beverages in association with childhood leukemia: a meta-analysis.

Objective: To systematically review studies and meta-analyze the literature on the association of maternal and/or index child's coffee, tea, and cola consumption with subsequent development of childhood leukemia and its major subtypes.

Methods: Eligible studies were identified through a detailed algorithm and hand-search of eligible articles' references; thereafter, summary-effect estimates were calculated by leukemia subtype and dose-response meta-analyses were performed.

Results: Twelve case-control studies, comprising a total of 3649 cases and 5705 controls, were included. High maternal coffee consumption was positively associated with acute lymphoblastic leukemia (ALL; OR: 1.43, 95%CI: 1.22-1.68) and acute myeloid leukemia (AML; OR: 2.52, 95%CI: 1.59-3.57). Any or low to moderate maternal cola consumption was also positively associated with overall leukemia (AL) and ALL, A linear trend between coffee and cola consumption and childhood leukemia was observed in the dose-response analyses. On the contrary, low to moderate tea consumption was inversely associated with AL (OR: 0.85, 95%CI: 0.75-0.97), although the trend was non-significant. A null association between offspring's cola consumption and leukemia was noted.

Conclusions: Our findings confirm the detrimental association between maternal coffee consumption and childhood leukemia risk and provide indications for a similar role of maternal cola intake. In contrast, an inverse association with tea was found, implying that other micronutrients contained in this beverage could potentially counterbalance the deleterious effects of caffeine. Further research should focus on the intake of specific micronutrients, different types of coffee and tea, specific immunophenotypes of the disease, and the modifying effect of genetic polymorphisms.