The aim of the study was to compare alendronate, hormone replacement therapy (HRT), and their combination in treatment of osteoporosis in elderly postmenopausal women. Ninety patients, aged 65-80 yr (mean 71), with a T-score of bone mineral density (BMD) of 2.5 or less at either the lumbar spine or the femoral neck were randomized to receive daily 10 mg alendronate (n = 30), 2 mg estradiol plus 1 mg norethisterone acetate (n = 30) (HRT), or their combination (n = 30) for 2 yr. BMD of the lumbar spine and the upper femur was measured at baseline and after 1 and 2 yr of treatment. Urinary excretion of type I collagen aminoterminal telopeptide as related to creatinine and serum type I procollagen aminoterminal propeptide was assayed at baseline and at 6-month intervals thereafter. Increases of 9.1-11.2% in lumbar spine BMD at 2 yr were similar in the study groups. Only HRT increased femoral neck BMD statistically significantly (P < 0.0001 for a change from baseline) at both 1 (+4.9%; P =NS vs. the other groups) and 2 yr (+5.8%; P < 0.05 vs. the other groups). Total hip BMD increased similarly in all study groups. Percentage reductions in urinary type I collagen aminoterminal telopeptide in the HRT group (60.2-62.7%) were significantly smaller than those in the combination group (78.1-80.4%) (P < 0.0001-0.0069) and the alendronate-only group (72.4-76.1%) (P = 0.047 at 24 months). Serum type I procollagen aminoterminal propeptide decreased less in the HRT group (53.6-59.8%) than in the other groups [73.0-75.0% in the alendronate group (P < 0.001 at 12 months); 67.0-71.5% in the combination group (P < 0.0001 at 12 months, P = 0.013 at 24 months)]. We conclude that in elderly postmenopausal women with osteoporosis, the combination of HRT and alendronate did not offer an extra gain of bone mass over either treatment alone. In terms of BMD changes, the single treatments were equally effective, but the reductions in bone markers were less with HRT than with alendronate.