Treatment of women with leiomyomata with gonadotrophin-releasing hormone agonists (GnRHa) for > 6 months is not recommended because of concerns regarding adverse sequelae of prolonged hypoestrogenism. It has been postulated that addition of low-dose sex steroids to GnRHa treatment, i.e. 'add-back' therapy, may avert some of these adverse effects (accelerated bone resorption, vasomotor flushes) without altering the efficacy of GnRHa therapy. To evaluate the effects of long-term GnRHa therapy on uterine size, bleeding patterns, bone mass and lipids, 51 pre-menopausal women with leiomyomata were treated with the GnRHa leuprolide acetate depot, 3.75 mg every 4 weeks for 2 years. After 3 months of leuprolide therapy, the women were randomized to receive either low-dose continuous oestropipate, 0.75 mg daily, plus cyclic norethindrone, 0.7 mg on days 1-14 each month (the oestrogen-progestin add-back group) or higher-dose norethindrone, 10 mg daily (the progestin add-back group), for the remaining 21 months. Mean uterine volume decreased by 40% in both treatment groups during the first 3 months on leuprolide treatment. There was no significant change in uterine size following oestrogen-progestin add-back. However, mean uterine volume in the progestin add-back group increased to 87% of pre-treatment size by treatment month 12 and 95% of pre-treatment size by treatment month 24. Mean bone density of the lumbar spine as measured by dual X-ray absorptiometry decreased significantly by 2.6% during the first 3 months in all patients, but did not change significantly following steroid add-back in both treatment groups during the final 21 treatment months. There were parallel and significant increases in mean haematocrits (Hct) of 4.8% in the oestrogen - progestin group and 7.8% in the progestin group over the 2-year treatment period. Mean serum high-density lipoprotein (HDL) cholesterol concentration was unchanged in the oestrogen progestin add-back group but decreased by 36% in women receiving progestin add-back. By 6 months after completion of treatment, mean uterine volume, leiomyoma-related symptoms, Hct and bleeding patterns had returned to pre-treatment values. Thus, the oestrogen-progestin add-back regimen was superior or equal to the progestin add-back regimen in all safety and efficacy parameters studied; the latter regimen was associated with regrowth of myomatous uterine volume and with marked depression of cardioprotective HDL cholesterol concentrations. One 50 year old woman in the oestrogen-progestin group developed a leiomyosarcoma which was suspected by sonographic changes in leiomyoma appearance and was thought to be unrelated to treatment. In conclusion, GnRHa plus oestrogen-progestin add-back therapy may provide a long-term (i.e. > 6 month) medical treatment option in women with leiomyomata.