The Safety Profile of Vitamin D Supplements Using Real-World Data from 445,493 Participants of the UK Biobank: Slightly Higher Hypercalcemia Prevalence but Neither Increased Risks of Kidney Stones nor Atherosclerosis

Abstract

Background: Potential calcium-related adverse events of vitamin D supplement use have not been addressed in large-scale, real-world data so far. Methods: Leveraging data from the UK Biobank, encompassing 445,493 individuals aged 40-69, we examined associations of high 25-hydroxyvitamin (25(OH)D) levels ≥ 100 nmol/L and vitamin D supplementation with hypercalcemia (serum calcium > 2.6 mmol/L), kidney stones, and atherosclerosis assessments (pulse wave arterial stiffness index and carotid intima-medial thickness). Regression models were comprehensively adjusted for 49 covariates. Results: Approximately 1.5% of the participants had high 25(OH)D levels, 4.3% regularly used vitamin D supplements, and 20.4% reported regular multivitamin use. At baseline, the hypercalcemia prevalence was 1.6%, and 1.1% was diagnosed with kidney stones during follow-up. High 25(OH)D levels were neither associated with calcium-related adverse events nor atherosclerosis assessments. Vitamin D and multivitamin supplementation were associated with an increased prevalence of hypercalcemia (odds ratios and 95% confidence intervals: 1.46 [1.32-1.62] and 1.11 [1.04-1.18], respectively) but were neither associated with atherosclerosis nor future kidney stones. Conclusions: High 25(OH)D levels observable in routine care were not associated with any adverse outcome. Vitamin D users have a slightly higher prevalence of hypercalcemia, possibly due to co-supplementation with calcium, but without a higher atherosclerosis prevalence or risk of kidney stones.

Keywords: adverse events; atherosclerosis; hypercalcemia; kidney stone risk; real-world evidence; serum 25-hydroxyvitamin D levels; vitamin D supplementation.

Figure 1

A flow chart of the study population inclusion and exclusion. Abbreviation: 25(OH)D: 25-hydroxyvitamin D, CIMT: carotid intima medial thickness, ICD-10: the 10th revision of the international classification of disease, PASI: pulse wave arterial stiffness index. ^a Sourced from self-reported, primary care, hospital admission, and death register data.

Figure 2

The dose–response relationship of serum 25-hydroxyvitamin D concentration with (a) hypercalcemia (cross-sectional analysis) and (b) kidney stones (longitudinal analysis). Notes: 3 knots were used and located at the 25th, 50th, and 75th serum 25-hydroxyvitamin D percentile, and the 75 nmol/L was used as the reference. Horizontal lines represent the odds ratio or hazard ratio of 1. Solid lines are estimates of the odds ratio or hazard ratios, and dashed lines are their 95% confidence intervals. Knots are represented by dots. The models are adjusted for all covariates included (n = 49, see legend of Table 2).

Figure 3

The dose–response relationship of serum 25-hydroxyvitamin D concentration with (a) pulse wave arterial stiffness index and (b) carotid intima medial thickness. Notes: 3 knots were used and located at the 25th, 50th, and 75th serum 25-hydroxyvitamin D percentile, and the 75 nmol/L was used as the reference. Horizontal lines represent the odds ratio of 1. Solid lines are estimates of odds ratios, and dashed lines are their 95% confidence intervals. Knots are represented by dots. The models are adjusted for all covariates included (n = 49, see legend of Table 2).