Laboratory considerations in the assessment of 25-hydroxyvitamin D in pregnant women by automated immunoassays

Abstract

Background: Because of the pathophysiological role of vitamin D in health, there is an increased interest to check the clinical status of this vitamin. Immunochemical assays are commonly employed to determine 25-hydroxyvitamin D (25 (OH) D) in clinical laboratories and its testing could be influenced by pre-analytic and analytic issues. The aim of this study was to compare the 25(OH)D results obtained from three commonly used immunoassays in pregnant women to check a possible discrepancy between tests.

Material and methods: A group of 50 pregnant women who were in their third trimester were included in this study. The quantification of serum vitamin D was performed utilizing three immunochemistry-based assays including Elecsys, VIDAS and Alegria. We also involved 21 non-pregnant volunteers to clinically assess the vitamin D status in this group of people.

Results: Our findings revealed a significant inconsistency between the obtained results from three assays for serum 25(OH)D. The 25(OH)D showed higher values when measured by the Elecsys assay while the VIDAS assay had lower values compared to the other immunoassays. More notably, the 25(OH)D testing in non-pregnant subjects showed consistent results in all three immunoassays.

Conclusions: The results of the 25(OH)D measurements in pregnant women should be interpreted carefully due to a great inaccuracy in immunoassay testing. There is no such disagreement in non-pregnant people. Standardization of vitamin D testing in various settings is a crucial matter for clinical laboratories.

Keywords: 25-Hydroxyvitamin D; Analytical method; Immunoassay; Inconsistency; Pregnant women.

Graphical abstract

Fig. 1

The concentration of 25(OH)D determined using three different methods. Serum 25(OH)D level (in ng/mL unit) had the highest values (26.8 ± 11.49 ng/mL) when measured by the Elecsys and decreasing concentration to 21.13 ± 9.59 and 17.13 ± 7.75 ng/mL for the Alegria and the VIDAS assays with a significant p value lower than 0.0001 for their difference.

Fig. 2

The correlation of the 25(OH)D results obtained from various assays: a) There was a significant correlation between the results obtained from the Elecsys and the Alegria assays, b) the VIDAS results had a direct statistically significant correlation with the Alegria results in the studied subject, c) the Elecsys results had also a statistically significant direct correlation with the results obtained from the VIDAS assay, and d) between all assays.

Fig. 3

Comparison of 25(OH)D levels (ng/mL) results obtained from different immunoassays using the Bland-Altman difference plot: A) 25(OH)D concentrations obtained by Elecsys were generally higher (mean difference: +9.646 ng/mL) than those of VIDAS. B) Lower level of bias was also demonstrated for Elecsys compared to the Alegria results (mean bias: +5.671 ng/mL). C) On the contrary, the results obtained by VIDAS, were on average lower (mean bias: −3.975 ng/mL) than the Alegria.

Fig. 4

Serum 25(OH)D concentration in different assays: the 25(OH)D level (in ng/mL unit) starting with the Elecsys assay (22.14 ± 6.53 ng/mL) and decreasing down 18.38 ± 7.69 ng/mL for the Alegria followed by similar results obtained by the VIDAS assay (17.73 ± 8.35 ng/mL) with non-acceptable significance (p value = 0.065). According to the error plot, concentration of 25(OH)D in non-pregnant volunteers showed an agreement between results obtained from the three immunoassay tests.