It has been more than a quarter of a century since the first strains of human cytomegalovirus (CMV) were isolated.1-3 Although its pathogenic potential was recognized immediately,4,5 the prevalence of congenital infection and the considerable morbidity associated with "silent" fetal infections were not appreciated until some years later.6,7 This morbidity takes the form of diminished intellectual potential and deafness. We also have learned that, unlike rubella and toxoplasmosis, systemic infection can be transmitted to a fetus by an immune mother.8 This point is critical in considering the problem of CMV vaccine development. If we vaccinate a woman of childbearing age, will it induce immunity and protect her unborn children from disease? We do not know. Cytomegalovirus infection, characteristically nonconforming, may not necessarily follow the common sequence of antigen processing, antibody formation, and clinical immunity.
In 1974, Elek and Stern inoculated medical students with a virus lysate