Symposium on infectious complications of neoplastic disease (Part II). Protected environments are discomforting and expensive and do not offer meaningful protection.

An environment that is germ free and a patient who is germ free have been goals in managing immunocompromised patients, especially those who are neutropenic. Such a germ-free environment, inside and out, can be achieved in laboratory animals after considerable manipulation, including delivery by cesarean section. This is rarely indicated or achieved in humans, but investigators have tried to decrease the numbers of organisms on the outside of patients (both environmental and on the skin) and to do the same with the inside--or at least to selectively decrease the organisms most likely to invade from mucous membranes and the gastrointestinal tract. Most studies of protected environments have included patients who receive prophylactic oral nonabsorbable antibiotics, prophylactic systemic antibiotics, or oral absorbable antibiotics (selective decontamination). Another variable that needs to be considered is the use of sterile food--included in some studies, not used in others, and not mentioned in still others. No studies have used the only appropriate control, which is the same laminar airflow room, but with the airflow turned off and without gowns, masks, or gloves, only assiduous hand-washing. The same staff should care for the control patients as well as those in protected environments. In addition, control patients in some studies have been clearly different in the severity of the underlying disease from patients placed in protected environments. Even without these proper controls, differences in infection rates have varied considerably among the studies both inside and outside the protected environment and efficacy has also varied considerably; however, the one constant in almost every controlled study is that life has not been prolonged, remission induction increased, nor remission duration prolonged.(ABSTRACT TRUNCATED AT 250 WORDS)