TABLE 1.

Relation between magnesium treatment and vitamin D metabolism, by baseline 25(OH)D concentrations¹

	Group, ng/mL		Difference (treatment − placebo) (95% CI)
Baseline 25(OH)D	Magnesium	Placebo		P
25(OH)D3, ng/mL
20	30.95 ± 2.18	29.37 ± 2.08	1.58 (−2.28, 5.44)	0.42
30	32.20 ± 1.56	29.41 ± 1.61	2.79 (0.25, 5.34)	0.03
40	29.76 ± 1.84	30.31 ± 1.80	−0.54 (−3.26, 2.18)	0.69
50	24.89 ± 2.74	31.76 ± 2.00	−6.87 (−11.30, −2.45)	0.002
24,25(OH)2D3, ng/mL
20	4.35 ± 0.41	3.96 ± 0.39	0.39 (−0.39, 1.17)	0.33
30	4.31 ± 0.32	3.90 ± 0.32	0.41 (−0.10, 0.91)	0.11
40	3.47 ± 0.32	3.91 ± 0.32	−0.44 (−0.99, 0.11)	0.12
50	2.12 ± 0.44	3.97 ± 0.36	−1.85 (−2.70, −0.99)	<0.0001
25(OH)D2, ng/mL
20	−2.22 ± 0.57	−1.65 ± 0.55	0.56 (0.19, 1.69)2	0.30
30	−1.94 ± 0.45	−1.38 ± 0.46	0.57 (0.27, 1.18)2	0.13
40	−1.68 ± 0.44	−2.13 ± 0.45	1.57 (0.72, 3.41)2	0.25
50	−1.43 ± 0.55	−3.56 ± 0.57	8.39 (2.38, 29.63)2	0.001

¹Values are means ± SEMs unless otherwise indicated. Values are based on our multiple linear regression model, adjusting for age = 60 y, sex = male, BMI (kg/m²) = 30, baseline eGFR = 7 · min^–1 · 1.73 m^–2, blood sample collection season = summer, baseline 25(OH)D₃ = 30 ng/mL or 24,25(OH)₂D₃ = 4 ng/mL or 25(OH)D₂ = 0.5 ng/mL among 87 participants randomly assigned to the magnesium treatment arm and 93 participants randomly assigned to the placebo arm. The P values for the interactions between magnesium intake with baseline 25(OH)D in changing 25(OH)D₃, 24,25(OH)₂D₃, and 25(OH)D₂ were 0.001, <0.0001, and 0.009, respectively. Total 25(OH)D, 25(OH)D₃: 1 ng/mL = 2.4959 nmol/L; 25(OH)D₂: 1 ng/mL = 2.4233 nmol/L. 24,25(OH)₂D₃, 24,25-dihydroxyvitamin D₃; 25(OH)D, 25-hydroxyvitamin D; 25(OH)D₂, 25-hydroxyvitamin D₂; 25(OH)D₃, 25-hydroxyvitamin D₃.

²Log-transformation was conducted to appropriately fit the model in the analysis of 25(OH)D₂. The fold-change of 25(OH)D₂ is presented.