OBJECTIVE:To investigate whether regular endurance-type exercise can benefit rats submitted to a model of ovariectomy (OVX)-induced obesity with or without estrogen replacement. SUBJECTS: OVX Sprague-Dawley rats were compared to an ovariectomized-estradiol-treated group (OVXE2) and a Shamoperated (Sham) group. Each of these groups were subdivided into a sedentary and a treadmill-trained (8 wk) group. DESIGN AND MEASUREMENTS: An experimental study in which various parameters, including fat depots, blood lipids and several organ weights were measured. RESULTS: Plasma levels of 17b-estradiol and uterus weights were significantly (P < 0.05) lower in OVX compared to Sham and significantly (P < 0.01) higher in OVXE2 (hyperestrogenic) compared to Sham rats. Body weights were significantly (P < 0.01) different among groups, in the following decreasing order: OVX, Sham and OVXE2. The average daily food intake and food efficiency were significantly (P < 0.01) increased in OVX compared to Sham, whereas estradiol treatment diminished this effect (P < 0.01). Exercise training did not alter any of the above-mentioned variables in any of the three estrogen groups. Mesenteric and subcutaneous fat weights were significantly (P < 0.01) increased by OVX. This increase was abolished by estrogen replacement or by exercise training. Exercise training also decreased fat weights in OVXE2 and Sham rats. OVX resulted in a decrease in the weights of several other tissues (femur, heart, lungs, liver and adrenal glands) while hyperestrogenic replacement resulted in an increase in weight of all measured tissues. Aside from fat depots, exercise training did not affect any of the tissue weights with the exception for an increase in the weight of the plantaris muscle and adrenal glands and a decrease in lung weight in all three estrogen groups. CONCLUSION: In OVX animals, exercise training may bring about positive changes in body composition (ie reduction in fat weights) despite an ovariectomy-induced increase in body weight.