A major difficulty with the detection of circulating galactomannan, a cell-wall polysaccharide released by Aspergillus sp during growth, in the serodiagnosis of invasive aspergillosis is the occurrence of false-positive ELISA results, especially in neonates and infants. On the basis of molecule similarity, we postulate that a lipoteichoic acid of Bifidobacterium sp can act as epitope for the monoclonal antibody used in the ELISA. The neonatal gut is heavily colonised with Bifidobacterium sp and these bacteria or their lipoteichoic acid might cause ELISA reactivity with serum after translocation because of immaturity of the intestinal mucosa. If our hypothesis is correct, we might find a method to discriminate between false-positive and true-positive ELISA results and thereby prevent unnecessary pre-emptive treatment of patients.