Increased gastrointestinal absorption of intact antigen with systemic immunization has been considered a major etiologic factor in the development of food sensitivity. We attempted to test this hypothesis in infants with suspected food protein-induced entercolitis by measuring serum ovalbumin (OVA) concentrations after ingestion of egg white (prior to the performance of good challenges to establish this diagnosis). We first noted significant underestimation of serum OVA concentrations in the presence of even low serum anti-OVA antibody concentrations (greater than 1:12). Next, using selected noninhibitory sera, we found that all infants studied absorbed some OVA, there was no correlation between serum OVA levels and age (3-11 months), and there was no significant difference between serum OVA concentrations in infants who subsequently had positive oral food challenge responses (120 +/- 67 ng/ml) and a matched group with negative challenges (102 +/- 80). These data do not support the hypothesis that "intestinal closure" (antigen exclusion) occurs in the neonatal period or the role of antigen absorption as the major etiological factor in the development of food sensitivity. Better methods of quantitating macromolecular absorption must be developed before the role of antigen absorption in food sensitivity can be assessed. Of note, urinary excretion of intact OVA also occurred. This varied greatly from one voiding to the next and continuing for at least 13 hr after ingestion.