Premium
Excessive Variability in a Popular Direct Immunoassay for Total 25 Hydroxy Vitamin D (t25OHD)
Author(s) -
Holmes Earle W,
Garbincius Jean,
McKenna Kathleen M.
Publication year - 2012
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.26.1_supplement.643.6
Subject(s) - immunoassay , linear regression , cluster (spacecraft) , chromatography , regression analysis , chemistry , zoology , mathematics , analytical chemistry (journal) , statistics , medicine , biology , immunology , antibody , computer science , programming language
This study was undertaken to compare t25OHD results determined by the DiaSorin Liaison direct immunoassay(IA) (DiaSorin Inc., Stillwater, MN 55082) with results that were measured by LC – MS (Quest Diagnostics, Wood Dale, IL 60191).166 consecutive clinical specimens were assayed for t25OHD by both IA and LC‐MS. The paired results were analyzed by linear regression analysis. Cluster analysis was used to classify the specimens into 5 groups based on the magnitudes of the biases in the t25OHD results determined by the two methods. The mean bias in t25OHD for each group was compared to each group's mean concentrations of 25 OHD2 and 25 OHDD3. Regression parameters for the paired results measured by IA(y) and LC‐MS(x) were: y=0.85x+5.9; r 2 =0.64; SE = 9.1 ng/mL. The inter‐method biases between the t25OHD results by IA and LC‐MS ranged from +30 ng/mL to −30 ng/mL Evaluation of the 5 groups derived from cluster analysis showed that the two groups with the highest positive mean inter‐method biases had significantly increased 25OHD2 concentrations and significantly increased 25OHD2/25OHD3 ratios compared to the other 3 groups. t25OHD was under‐recovered by the IA in about half of the clinical specimens with 25OHD2 concentrations that were > 20 ng/mL. The inter‐method bias between IA and LC‐MS results exceeded the total allowable error specification for 25OHD measurements in 28 % of the clinical specimens that were tested. The results showed that t25OHD concentrations determined by the Liaison IA can be falsely elevated or falsely decreased relative to LC‐MS determined values by heterophile antibody interference and under‐recovery of 25OHD2, respectively. These analytical limitations emphasize the need to carefully validate the accuracy of direct immunoassays for total 25OHD against a definitive reference method before they are approved for routine clinical use.