Periprosthetic bone loss is an important factor that limits implant survival after total hip arthroplasty (THA). In a randomized trial we previously reported that pamidronate therapy prevented periprosthetic bone loss and decreased urinary excretion of N-telopeptide collagen cross-links over the first 6 months after THA, but had no apparent effect on free deoxypyridinoline excretion (J Bone Miner Res 2001; 16:556-564). In this study we investigated this discrepant observation that pamidronate reduced conjugated cross-link excretion but had no effect on free cross-links. Free and total deoxypyridinoline (DPD) were assayed by reverse-phase high-performance liquid chromatography (HPLC) and by immunosorbent assay (ELISA) at preoperative baseline and at week 6 after surgery in 46 subjects who had taken part in the trial. Randomly selected, 22 subjects received a single 90 mg intravenous infusion of pamidronate and 24 received placebo. Acute rises in free and total DPD occurred in both study groups at week 6 (P < 0.05). Total DPD excretion was lower in the pamidronate group than in the placebo group when measured by both HPLC and ELISA (P < 0.05). No difference in free DPD was found between groups. A rise in the ratio of free to total DPD occurred in the pamidronate group at week 6 (P = 0.03), but not in the placebo group. Pamidronate treatment suppresses excretion of total DPD. This is consistent with the effect of pamidronate on other turnover markers and periprosthetic bone loss after THA. Urinary-free DPD is a poor marker of response to treatment as the ratio of free-to-total cross-links is affected by amino-bisphosphonate therapy.