The main objective of this study was to evaluate the safety and efficacy of a perfluorochemical emulsion, Fluosol, with short-term high inspired oxygen tension as an adjuvant to radiation therapy in the treatment of high-grade tumors of the brain. Radiation was delivered to the whole brain at 1.8 Gy per daily treatment for 5 weeks to a total dose of 45 Gy. The radiation portals were then reduced in size to encompass the known volume of tumor, as determined by the presurgical contrast-enhancing ring on computed tomography (CT), plus a 3-cm margin. An additional 10 treatments of 2 Gy each were given to the smaller volume, to bring the total tumor dose to 65 Gy in 7 weeks. This report describes the experience of the first 18 patients treated at the University of Kansas Medical Center on this study, whose median follow-up time from the date of surgery is 77 weeks (62-115 w). Immediately following Fluosol administration on a Monday, patients breathed 100% oxygen for at least 45 minutes prior to and throughout their radiation treatment. On each subsequent day of the weeks in which they received Fluosol, patients breathed 100% oxygen. Hematology and blood chemistries were also drawn prior to Fluosol treatment each Friday during treatment and at the 2-week, 3-month, and 6-month follow-up visits. The median age of the patients was 45 years (16-72); 13 patients were male and 15 carried the diagnosis of glioblastoma multiforme (3 had anaplastic astrocytoma). Two thirds of the patients had an initial allergic reaction to the Fluosol consisting of back pain, shortness of breath, and flushing, but all responded to 50-100 mg of Benadryl. During radiation therapy, all patients developed scalp erythema and complete alopecia by the end of 3 weeks, but no patient required a treatment rest. The serum levels of SGOT, SGPT, and alkaline phosphatase were examined before and throughout the Fluosol treatment and, by week 5, 11/18 of the patients had increased values of all three enzymes above the upper range of normal. These increases persisted through the end of treatment, but most values returned to essentially normal by the 3-month follow-up visit. We conclude that Fluosol, given in the manner described above, appears to be associated with minimal significant side effects and no changes could be detected in the white matter of any of the patients at the time of their magnetic resonance imaging study at 6 months follow-up.(ABSTRACT TRUNCATED AT 400 WORDS)