PURPOSE: To assess the influence of different total-body irradiation (TBI) regimens on interstitial pneumonitis (IP), we retrospectively analyzed our clinical data concerning an homogeneous group of patients conditioned with cyclophosphamide (CY) alone and single-dose or fractionated TBI before autologous bone-marrow transplantation (ABMT). METHODS AND MATERIALS: One hundred eighty-six patients with acute nonlymphoblastic leukemia (n = 101), acute lymphoblastic leukemia (n = 62), chronic myeloid leukemia (n = 11), non-Hodgkin's lymphoma (n = 10), and multiple myeloma (n = 2) referred to our department between May 13, 1981 and September 16, 1992, underwent TBI before ABMT. The male-to-female ratio was 123:63 (1.95), and mean and median age was 33 +/- 12 (6-63 years) and 35 years, respectively. Cyclophosphamide alone (60 mg/kg/day on each of 2 successive days) was used as conditioning chemotherapy in all patients. Patients were irradiated according to two techniques: either with single-dose (STBI) (n = 124; 10 Gy administered to the midplane at the level of L4, and 8 Gy to the lungs) or with fractionated (FTBI) (n = 62; 12 Gy in 6 fractions over 3 consecutive days to the midplane at the level of L4, and 9 Gy to the lungs) TBI. The mean instantaneous dose rate was 0.057 +/- 0.0246 Gy/min (0.0264-0.1692 Gy/min). It was < or = 0.048 Gy/min in 48 patients (LOW group), > 0.048 and < or = 0.09 Gy/min in 129 patients (MEDIUM group), and > 0.09 Gy/min in 9 patients (HIGH group). The median follow-up period was 5 years (24-120 months). RESULTS: In January 1994, the 5-year overall (including all causes of death) and disease-free survival (DFS) rates were 50 and 48%, respectively. The 5-year DFS was 47.9% in the STBI group, and 47.8% in the FTBI group (p = 0.77). It was 44% in the HIGH group, 53% in the MEDIUM group, and 34% in the LOW group (LOW vs. MEDIUM, p = 0.009). The 5-year IP incidence was 17% in all patients, 16% in the STBI group and 18% in the FTBI group (p = 0.37), but it was significantly higher in patients receiving high instantaneous dose rate TBI (56% in the HIGH, 13% in the MEDIUM, 20% in the LOW groups; HIGH vs. MEDIUM, p = 0.002). However, sex (p = 0.37), age (18% for > 20 vs. 10% for < or = 20 years, p = 0.37), and body weight (> 60 kg vs. < or = 60 kg, p = 0.09) did not influence the IP incidence in univariate analyses. Multivariate analysis (Cox model) revealed that the instantaneous dose rate (p = 0.05), and the age (p = 0.04) were the two independent factors influencing the incidence of IP. CONCLUSION: This retrospective study including only the patients transplanted with ABMT conditioned with CY alone and STBI or FTBI concluded that instantaneous dose rate and age significantly influenced the incidence of IP, whereas sex, body weight, and fractionation did not.