Comparison of total and bone-specific alkaline phosphatase in patients with nonskeletal disorder or metabolic bone diseases.

To evaluate the diagnostic validity of new assays for bone-specific alkaline phosphatase (BAP), we compared measurements of total alkaline phosphatase (TAP) in serum with results for three different assays of serum BAP in healthy adults (n = 119), patients with chronic nonskeletal disorders (n = 123), and patients with metabolic bone diseases (n = 113). Serum TAP was determined by a standard colorimetric assay, BAP by the methods of lectin precipitation (L-BAP), enzyme immunoassay (E-BAP), and immunoradiometric assay (I-BAP). Impairment of liver function resulted in significant increases of all alkaline phosphatase (AP) measurements, with the smallest changes being exhibited by E-BAP. Compared with the results by TAP, diagnostic sensitivity (i.e., of values exceeding the reference interval) was not improved by BAP, but receiver-operating characteristic (ROC) curve analyses revealed improved discrimination for primary hyperparathyroidism by E-BAP. These results indicate that, in the presence of liver disease, the specificity of AP measurements is improved by measuring BAP. In most other clinical situations, serum TAP appears to provide sufficient clinical information; however, the cross-sectional study design used here allows no statement about the usefulness of BAP in serial measurements.