OBJECTIVE Our purpose was to compare the effects of leuprolide acetate in patients with symptomatic uterine leiomyoma before hysterectomy. METHODS Group I (n = 90) included patients with a pretreatment uterine size of 14 to 18 gestational weeks and group II (n = 60) included patients with uteri > 18 weeks' gestational size. Patients in both groups were randomized to either immediate hysterectomy or 2 months of preoperative gonadotropin-releasing hormone agonist. RESULTS All patients in the two groups with a pretreatment hemoglobin < 11.0 gm/dl randomized to agonist had a significant (p < 0.05) increase (> or = 1.5 gm/dl) in hemoglobin level. Patients in group I who received preoperative agonist were more likely to undergo vaginal hysterectomy (80% vs 13%, p < 0.05) than were patients who did not receive preoperative agonist. Patients undergoing vaginal hysterectomy had a shorter hospital stay, decreased operative blood loss, and a shorter convalescence period than did those undergoing abdominal hysterectomy. In group II, in spite of a mean uterine volume reduction of 51.3%, intraoperative morbidity, operative blood loss, hospital stay, and postoperative convalescence period did not differ between treatment arms. CONCLUSIONS The preoperative administration of gonadotropin-releasing hormone agonist in patients with a uterus of 14 to 18 weeks' size increases the use of vaginal hysterectomy, decreases intraoperative blood loss, and shortens hospital stay and convalescence. Preoperative gonadotropin-releasing hormone agonist for patients with a preoperative hemoglobin < 11.0 gm/dl reduces the risk of preoperative transfusion. Preoperative gonadotropin-releasing hormone use in the nonanemic patient with a uterine size > or = 18 weeks' gestational size doses not appear to lower operative morbidity.