The objective of this study was to further evaluate the relative safety of extracorporeal photopheresis in the treatment of patients with AIDS-related complex. Twenty patients with AIDS-related complex, three from the initial report and 17 additional patients, were enrolled. The patient population had various risk factors. There were nine homosexuals, five heterosexual consorts of human immunodeficiency virus (HIV)-positive patients, four reformed i.v. drug abusers, and two hemophilia patients. The patients received monthly treatment with extracorporeal photopheresis. In 16 of the 19 patients, this study provides evidence of clinical stability over a longer period. The relative stability of beta 2-microglobulin and neopterin levels demonstrates that photopheresis therapy does not have an untoward effect on the degree of activation of the immune system with respect to induction of HIV replication. Antibody titers to the major viral antigens, envelope glycoproteins (gp) 120 and 41, reverse transcriptase enzyme gp 66/31 and 55, and the core protein p24 remained stable throughout the course of therapy. A subjective improvement was noted in the majority of patients. The evaluation of T-cell subsets revealed that the photopheresis treatment did not have a detrimental effect on CD3 and CD8 cells. Some decreases were noted in the CD4 cell counts but the decline may be less than is normally seen at corresponding stages of HIV infection. Skin test responsivity improved in 11 patients, remained unchanged in seven patients, and declined in two. The preliminary results suggest that in HIV disease, extracorporeal photopheresis is safe and warrants a prospective controlled trial.