Age-dependent discordance in chemiluminescent immunoassay-electrochemiluminescence immunoassay correlation for 25-hydroxyvitamin D measurement reveals methodological considerations for assessment in early infancy.

Vitamin D (VD) status is assessed by measuring serum 25-hydroxyvitamin D (25(OH)D) levels using immunoassays. In December 2024, a Japanese diagnostic laboratory transitioned from chemiluminescent immunoassay (CLIA: LIAISON® 25 OH Vitamin D Total Assay) to an electrochemiluminescent immunoassay (ECLIA: Elecsys® Vitamin D Total III assay). Although these methods yield comparable results in adults, it remains unclear whether this applies to children, particularly newborns and infants. This study aims to clarify the impact of transition from CLIA to ECLIA on 25(OH)D levels during early infancy. Serum 25(OH)D levels were measured in 84 infants under three months old (67 under four days old) using both CLIA and ECLIA. Correlation and agreement were assessed using Passing-Bablok regression and Bland-Altman analyses, respectively. The obtained regression equation was applied to 252 age- and sex-matched cases from 2,253 historical controls in the registry to evaluate changes in VD classification. The regression equation was YECLIA = 0.846XCLIA - 2.281 (τ = 0.53, p < 0.001). Bland-Altman analysis revealed a constant negative bias of -4.89 ± 3.53 ng/mL [-5.66, -4.12] (mean ± standard deviation [95% confidence interval]) for ECLIA versus CLIA. Reclassification revealed an increase in VD deficiency (25(OH)D < 12 ng/mL) from 27.8% to 61.9%, and VD insufficiency (25(OH)D < 20 ng/mL) from 75.4% to 91.7%. The transition from CLIA to ECLIA led to lower measured 25(OH)D levels in newborns and infants and increased VD deficiency and insufficiency diagnoses. Method-specific differences should be considered when evaluating VD status during early infancy, particularly in infants under four days.