The use of ozone therapy is reported to be effective in a variety of viral illnesses, including HIV disease. We performed a phase I study of ozone blood treatments in 10 patients in whom no significant toxicity was observed. Three patients with moderate immunodeficiency showed improvement in surrogate markers of HIV-associated immune disease. A phase II controlled and randomized double-blinded study was initiated comparing reinjection of ozone-treated blood, and reinjection of unprocessed blood for 8 weeks, followed by a 4-week observation period. Ozone had no significant effect on hematologic, biochemical or clinical toxicity when compared with placebo. CD4 cell count, interleukin-2, γ-inteferon, β2-microglobulin, neopterin and p24 antigen were also unaffected by both treatment arms. In conclusion, ozone therapy does not enhance parameters of immune activation nor does it diminish measureable p24 antigen in HIV-infected individuals.