Introduction: Increased mortality in patients with low serum concentrations of S-25(OH)D has been described, though no causal relationship has been shown. The aim of this cohort study was to investigate the possible association between S-25(OH)D status and all-cause mortality in 5,147 patients attending a university hospital in Denmark from 2003-2008.
Methods: Serum was analyzed for ionized calcium (S-Ca(2+)), parathyroid hormone (S-PTH) and S-25(OH)D. Kaplan-Meier curves were used to analyze the association between S-25(OH)D quartiles, S-PTH-quartiles and all-cause mortality. Cox regression analyzed associations of age, gender, concentrations of ionized calcium, S-PTH and S-25(OH)D with all-cause mortality.
Results: Log rank tests for both S-25(OH)D and S-PTH quartiles showed a significant difference across the quartiles. The Cox regression analysis showed a hazard ratio (HR) for death of 1.02 (95% CI: 1.01; 1.03) for S-PTH per pmol/L increase, 1.07 (95% CI: 1.05; 1.10) for S-25(OH)D per 10 nmol/L decrease and 0.54 (95% CI: 0.24; 1.20) for S-Ca(2+) per mmol/L increase.
Discussion: This study supports growing evidence, that low serum concentrations of 25(OH)D are associated with an increased all-cause mortality; however the nature of the study prevents the finding of a causal relationship. In the Cox regression analysis S-25(OH)D concentrations in the blood were inversely associated with all-cause mortality. Prospective controlled studies are needed to confirm a causal relationship between low S-25(OH)D concentrations and mortality.