The purpose of this study was to estimate clinical validity of a new available immunoradiometric assay for circulating intact human BGP (N-tact Osteo SP) by measuring this protein in a large number of normal subjects and patients with the most common metabolic bone diseases. One hundred normal subjects were studied in order to obtain our normal ranges (4.9 +/- 1.7 ng/ml). The mean values found in 28 patients with primary hyperparathyroidism (17.5 +/- 22.8 ng/ml, P < 0.001), 15 glucocorticoid-treated patients (1.9 +/- 1.5, P < 0. 001), 10 patients with hypoparathyroidism (1.5 +/- 0.7, P < 0.001), 9 with hyperthyroidism (8.3 +/- 3.8, P < 0.001), 8 with skeletal metastases (7.2 +/- 2.3, P < 0.001), and 4 with humoral hypercalcemia of malignancy (2.42 +/- 1.91, P < 0.005) were significantly different from mean values found in normal subjects. Mean decrease of serum osteocalcin T-score values was significantly greater when evaluated by N-tact Osteo SP assay in 15 steroid-treated patients (-1.4 +/- 1.0) and 19 primary hyperparathyroid (PHPT) patients (3.6 +/- 1.9), compared with the mean values obtained with the Elsa-Osteo assay (-0.67 +/- 1.2, P < 0. 002 and 4.3 +/- 2.8, P < 0.04, respectively). We found significant correlations between the global skeletal uptake of 99mTc-methylendiphosphonate and serum BGP levels assayed by both N-tact Osteo SP (P < 0.01) and Elsa-Ost-Nat assay (P < 0.05). Our results indicate that this new immunoradiometric assay for the intact human osteocalcin has the potential for good discrimination between normal subjects and patients with both low and high bone turnover. Furthermore, our findings emphasize the fact that, in the absence of available standardized commercial assays, one should rely on only one assay because different results are obtained by different assays under different clinical conditions.