Two bone alkaline phosphatase assays compared with osteocalcin as a marker of bone formation in healthy elderly women.

Serum bone alkaline phosphatase (ALP; EC 3.1.3.1) was measured with a wheat germ agglutinin (WGA) precipitation assay and with a new IRMA in a group of healthy elderly women. Both assays were correlated with serum total ALP activity and with osteocalcin. The two bone ALP assays have comparable within- and between-run imprecisions (WGA assay within-run CVs 2.6-5.4% and between-run, 4.0-5.1%; IRMA within-run CV 5.0% and between-run, 3.2%). Comparison of the WGA precipitation assay (x) with the IRMA (y) demonstrated a correlation coefficient of 0.87 [Deming regression equation: y = (0.58 +/- 0.02)x - (4.62 +/- 0.45); n = 101; Sy/x = 1.26; P < 0.001). Correlation studies with osteocalcin and total ALP showed correlation coefficients (all P < 0.001) of 0.34 and 0.65, respectively, for the WGA precipitation assay and of 0.36 and 0.68, respectively, for the IRMA. We conclude that the two bone ALP assays have similar imprecision and that neither can be given preference over the other as a marker of bone turnover.