Cytomegalovirus (CMV) in the compromised host(s).

Cytomegalovirus infection in compromised hosts occurs most frequently as asymptomatic shedding of a reactivated virus. Reactivation occurs when cellular immunity is compromised owing to either immune-deficiency disease or iatrogenic intervention. When intervention is modest, most antibody-positive patients do not shed virus (19); when the group is treated with a higher mean dose of cytotoxic agents, almost all antibody-positive patients shed virus (6). In the instance where the chemotherapeutic program includes massive doses of these agents, as in heart or marrow transplantation, or when dysfunction of all cellular activity is extreme, as in the case of severe combined immune deficiency and far-advanced neoplasia, dissemination of infection with multiple-organ involvement ensues (17, 20-22). Primary infection may occur in some patients with neoplasia or in marrow-transplant recipients from either blood transfusions or from marrow, although data are insufficient to be certain (22). It has been conclusively shown, however, that seronegative renal-allograft recipients usually develop primary infection when they acquire a kidney from a seropositive donor. Clinical illness related to virus infection in the first few months after transplantation occurs in almost all patients with primary infection and seldom in those with reactivation infection (6, 25, 26). Three very intriguing observations have been made that should suggest avenues for future research: (a) Since illness occurs more closely related to development of antibody than it does to onset of virus shedding, this suggests that host response plays an important role in the disease that is obviously caused by a virus; a definition of the mechanism of this interaction could lead to an understanding of host-induced disease processes in man; (b) Graft-versus-host disease in marrow transplantation; and (c) allograft rejection in renal-graft recipients correlate with development of CMV infection (16, 22, 24, 26). Taken together, these observations are consistent with the hypothesis that virus antigens interrelate with most antigens in such a way as to lead to immune response to both antigens. Exploration of this clinical observation could lead to a greater understanding of the function of the HL-A system in man.