Prenatal factors have been implicated in childhood eczema, but the relationship between maternal cytokine production during pregnancy and infant eczema is unknown. Non-selected women in their third trimester were enrolled in the Tucson Infant Immune Study. Data from three sources were used to define MD-eczema: parent-completed illness questionnaires at age 2, 3, 4, 6 and 9 months regarding physician-seen eczema, parent-completed questionnaires at 12 months regarding physician-diagnosed eczema, and medical record reviews. Blood samples were taken from mothers during their third trimester and from the umbilical cord at birth. Maternal peripheral blood mononuclear cells and cord blood mononuclear cells were stimulated with ConA/PMA, and supernatants were assayed for IFN-gamma and IL-4, -5, -10, and -13. Of 364 children, 28% were seen by a physician for eczema by 1 yr of age. After adjustment for potential confounders using logistic regression, the odds for development of eczema in infancy were significantly higher when mothers had active eczema in pregnancy (OR, 2.46, CI 1.0-5.8, p <0.042) and when mothers were in the highest tertile of serum IgE production (OR 2.28, CI 1.2-4.4, p <0.013). Colds in the third trimester were associated with lower odds of eczema (OR 0.32, CI 0.16-0.63, p <0.001). Our findings from this cohort study suggest that in utero factors, including maternal IgE, colds, and eczema, may influence the risk of infant eczema.