The proportion of carboxylated to total or intact osteocalcin in serum discriminates warfarin-treated patients from control subjects.

We assessed the serum concentration of gamma-carboxylated osteocalcin (OC), total OC, and full-length OC in a clinical setting of 37 patients on continuous warfarin treatment (international normalized ratio 2.0-3.8). A comparison was done with the results from 30 untreated age-matched controls. Four monoclonal antibodies, previously generated and characterized as to their ability to recognize different human OC forms and fragments, were used in three two-site immunofluorometric assays. The warfarin-treated patients had significantly lower levels of carboxylated OC 4.9 +/- 3.8 (+/- 1 SD) ng/ml compared with the controls 13.1 +/- 9.7 (p < 0.0001). There was no difference in the levels of total OC or full-length OC between the two groups of patients. A strong correlation was found between the serum concentration of carboxylated OC and total OC, both for the warfarin-treated patients (r = 0.98) and for the controls (r = 0.99). There was a distinct cut-off level at 0.80, in the quotient carboxylated OC/total OC, at which all warfarin-treated patients fell below and all controls above this level. Hence, the concentration or ratio of serum gamma-carboxylated OC in clinical settings such as warfarin-treated patients could be measured using two-site immunoassays.