The effect of biphosphonate therapy on bone mineral density (BMD) in patients with primary hyperparathyroidism (PHP) is unknown. Forty postmenopausal women (mean age, 70 yr) with PHP were randomized to receive alendronate 10 mg/d or placebo for 48 wk, followed by treatment withdrawal for 24 wk. The mean (+/-SD) changes in BMD at femoral neck (+4.17 +/- 6.01% vs. -0.25 +/- 3.3%; P = 0.011) and lumbar spine (+3.79 +/- 4.04% vs. 0.19 +/- 2.80%; P = 0.016) were significantly higher with alendronate at 48 wk. Serum calcium was reduced with alendronate but not placebo (-0.09 vs. +0.01 mmol/liter; P = 0.018). Serum bone-specific alkaline phosphatase activity was lower with alendronate from 12 wk onward and increased 24 wk after treatment withdrawal (21.1 +/- 12.8 to 7.3 +/- 4.9 IU/liter at 48 wk, and 15.0 +/- 14.8 IU/liter 24 wk after withdrawal; P = 0.002 for trend). Osteocalcin concentration decreased at 48 wk and increased 24 wk after alendronate withdrawal (P = 0.019 for trend of change over time) but not with placebo. Urinary N-telopeptide/creatinine ratio decreased with alendronate at 48 wk and increased 24 wk after treatment withdrawal (P = 0.008 for trend). N-telopeptide/creatinine ratio did not change with placebo. Alendronate improves BMD and reduces bone turnover markers in postmenopausal women with PHP.