Objectives: To evaluate the efficacy and safety of very low dose danazol in improving pelvic pain in women with endometriosis, the benefit of preceding the treatment by a short course of a GnRH agonist, symptoms recurrence after drug withdrawal, and variations in menstrual pattern. Design: Open-label, randomized study. Setting: University hospital endometriosis center. Patients: Forty-two women with moderate or severe pelvic pain and laparoscopically diagnosed endometriosis. Interventions: Treatment with oral danazol, 50 mg/d, for 9 months (group I, n = 21) or leuprolide depot for 3 months followed by oral danazol, 50 mg/d, for 6 months (group II, n = 21), and a 6-month follow-up. Main Outcome Measures: Variations in severity of symptoms during treatment and at the end of follow-up as shown by a linear analog scale and a verbal rating scale; menstrual blood loss as assessed by a pictorial chart. Results: Four patients withdrew from the study, one in each group at the fifth month of treatment (for persistent pain) and one in each group during follow-up (they requested additional therapy); one woman in group I was lost to follow-up. Significant improvements were obtained in dysmenorrhea, deep dyspareunia, and nonmenstrual pain in both treatment schedules without differences between the groups. Also menstrual blood loss was significantly reduced in both groups. A temporary fall in high and rise in low density lipoprotein cholesterol was observed in the study population. At the end of follow-up symptoms recurred without significant differences in median pain scores with respect to baseline. Conclusion: Very low dose danazol may be an alternative for temporary relief of endometriosis-associated pain. Ovulation is not always inhibited and barrier contraception is needed. Side effects occur but are rarely severe. Further data are required to evaluate the influence of long-term administration on the lipid profile.