Nosocomial infections in the immunocompromised adult.

It is universally accepted that the immunocompromised host is more susceptible to nosocomial infection than the normal host. Descriptive clinical or epidemiologic studies that utilize the "organism oriented" approach are now abundant. Generalizations about risk factors are difficult to make, however, because of the wide range of host defects that are usually compounded by therapeutic intervention. Comparative data are lacking on whether certain procedures or treatments are associated with a higher incidence of infection in immunocompromised hosts versus normal hosts. Aggressive diagnostic and therapeutic approaches to nosocomial infection in immunocompromised patients are usually advocated, but there have been few prospective studies of their effectiveness and the possible increased hazard of nosocomial infection. The paramount questions that must be addressed are, "What infections are preventable and how?" Two derivative issues relate to (1) the cost to benefit aspects of preventive measures, and (2) the impact of such measures on survival from underlying disease. Rigorously controlled studies of specific preventative measures are clearly required because the appearance of improvement may be related to concurrent improvements in the treatment of underlying disease or its complications. The ultimate value of infection control programs in immunocompromised patients with depend on the demonstration that intervention not only reduces infection rates but also has a favorable impact on the treatment of the underlying disease.