This project test the hypothesis that humoral immunity to rotavirus infection, assessed in serum, is associated with the risk of rotavirus diarrhea in children aged under 2 years who reside in rural Bangladesh. Cases of rotavirus diarrhea were assembled from population-based, comprehensive surveillance of episodes of pediatric diarrhea in Matlab, Bangladesh during 1985-86. All rotavirus diarrhea episodes were serotyped with monoclonal antibodies after cultivating rotavirus isolates. Controls were similarly aged children randomly selected during four surveys of the Matlab community undertaken during 1985-86. Serological correlates of natural immunity to rotavirus diarrhea were assessed by contrasting acute-phase sera collected from the cases and sera from controls. Case-control comparisons revealed that: 1) serum IgG antirotavirus antibodies were correlated inversely with the occurrence of rotavirus diarrhea, suggesting a protective relationship; but 2) protective associations were not serotype-specific. A second goal is to evaluate whether breast feeding is associated with a reduced risk of rotavirus diarrhea in children under the age of 2 years. Using a case-control sampling strategy analogous to that described for the seroepidemiological assessment, comparisons of cases and controls for antecedent histories of breast feeding revealed that: 1) breast feeding was associated overall, with a reduced risk of rotavirus diarrhea, due to the protection conferred by exclusive but not partial breast feeding; but 2) after infancy, the direction of the relationship became reversed, with a higher risk of rotavirus diarrhea among breast-fed than non-breast-fed children. Future analyses will assess whether the protection observed during infancy is correlated with antirotavirus and neutralizing antibodies in breastmilk.
