Evaluation of an Automated Competitive Protein-Binding Assay for 25-Hydroxyvitamin D

The measurement of 25-hydroxyvitamin D is of increasing importance in the management of patients with mineral disorders. However, there are a great variety of test results for 25-hyroxyvitamin D depending on the method used. In this report a new automated method provided by Roche diagnostics (Elecsys Vitamin D Total assay) and the previously marketed method are compared to a reference method (Immundiagnostik ELISA). Further, we tested the new Roche method for its ability to monitor vitamin D supplementation.

Serum aliquots of 80 consecutive patients were prepared and 25-hydroxyvitamin D was measured by two automated methods provided by Roche diagnostics and by ELISA (Immundiagnostik, Bensheim, Germany). Further, we collected samples from 80 osteoporosis patients on vitamin D supplementation (1000 IU daily) and measured serum 25-hydroxyvitamin D using the Roche Elecsys Vitamin D Total assay.

The new Roche Vitamin D Total assay showed better correlation with the ELISA (r = 0.73) than the old automated method (r = 0.41). The 25-hydroxyvitamin D values obtained with the old automated Roche method were much lower compared to the new method or the ELISA, resulting in overestimation of vitamin D deficiency. In this respect, the new Roche Vitamin D Total assay was in rather good agreement with the ELISA. Moreover, the application of the new Roche Vitamin D Total assay in the monitoring of vitamin D supplementation gave clinically useful results: 90% of the patients receiving 1000 IU of vitamin D3 daily had a 25-hydroxyvitamin D serum concentration of > 50 nmol/L, which is in the expected range. Moreover, the 25-hydroxyvitamin D concentrations were negatively correlated to PTH proving the plausibility of the results.

25-hydroxyvitamin D measurements show a large variability. Results from previous studies obtained with the old Roche automated method should be used with caution. The new automated Roche Vitamin D Total assay exhibits a reasonable concordance with the ELISA and can be used for monitoring patients in clinical practice. However, because of the variability, the results for individual patients are of limited use and general population based screening for vitamin D deficiency cannot be advocated.