Two unusual cases of severe erythroblastosis due to A and B sensitization have been presented. When injections of A and B group substances failed to arrest the disease, exchange transfusions were carried out, using 900 to 1,000 cc. of fresh group O blood. In each case the response was prompt and dramatic, although the convalescence in one was prolonged by an intercurrent diarrhea. Both infants have made complete recoveries and have developed normally both physically and mentally.
Observations have been presented regarding the pathogenesis of erythroblastosis and icterus precox due to A-B sensitization. The following conclusions seem to be warranted on the basis of the evidence presented:
1. The greatest majority of cases of jaundice and anemia of the newborn that cannot be explained on the basis of Rh incompatibility are caused by incompatibility of the major blood groups.
2. High maternal alpha and beta antibody titers per se are not necessarily correlated with disease in the infant.
3. Univalent alpha and beta antibodies present in the maternal serum traverse the placenta and are the cause of the disease in the infant. Bivalent antibodies are held back by the intact placenta and play no or hardly any role in the causation of the disease. Univalent alpha and beta antibodies are demonstrable in the sera of a large proportion of "normal" individuals.
4. A-B sensitization in pregnancy occurs mainly when the infant belongs to the secretor type.
5. A theory is suggested that the quality of the alpha and beta antibodies, namely, whether they are homospecific or heterospecific, may affect the severity of the manifestations in the infant.
Technics of titrating alpha and beta and Rh antibodies are described and discussed. A table has been prepared which converts antibody titers into concentrations of immune globulin in the serum, and demonstrates the impossibility of certain extravagantly high titers claimed in the literature.