EXCRETION OF URINARY N-TELOPEPTIDES REFLECTS CHANGES IN BONE TURNOVERDURING OVARIAN SUPPRESSION AND INDICATES INDIVIDUALLY VARIABLE ESTRADIOL THRESHOLD FOR BONE LOSS

Objective: To evaluate the effectiveness of N-telopeptides and E(2) in monitoring bone turnover during GnRH agonist- (GnRH-a) or danazol-ind uced hypoestrogenism. Design: Comparative, nonrandomized prospective s tudy. Setting: Institute for the Study and Treatment of Endometriosis Patient(s): Premenopausal women undergoing ovarian suppression with Gn RH-a (n = 16) or danazol (n = 9). Intervention(s): Serum and urine sam ples were collected and bone mineral density was measured before, duri ng, and after treatment Main Outcome Measure(s): N-telopeptide excreti on, serum E(2), and bone mineral density at L1 to L4 and femoral neck. Result(s): During treatment in the GnRH-a group, mean E(2) levels wer e 53% below and N-telopeptides were 38% above the mean baseline. At 1 month post-treatment, L1 to L4 bone mineral density decreased by 3.85% . In the danazol group, E(2), N-telopeptides and L1 to L4 bone mineral density changed nonsignificantly in the opposite direction with the m ean 1.25% increase in L1 to L4 at 1 month post-treatment. In combined groups, L1 to L4 bone mineral density better correlated with other mea sures than femoral neck bone mineral density. N-telopeptide excretion was more predictive of L1 to L4 change, with correlation the highest b etween N-telopeptides at month 4 and bone mineral density at month 1 a fterward, while E(2) appeared more predictive of the less reliable fem oral neck bone mineral density. Individual exceptions to the model of an E(2) threshold for bone loss were observed. Also noted were high co rrelation between on-therapy levels of E(2) and N-telopeptides, as wel l as the presence of a 1-month time lag between E(2) and N-telopeptide changes. Conclusion(s): Bone density decreases during GnRH-a and may slightly increase during danazol treatment. However, E(2) threshold fo r bone loss varies individually. N-telopeptides predict changes in bon e mineral density at L1 to L4 better than E(2).