Vitamin D as a prognostic biomarker in COVID-19: single-center study and meta-analyses

Abstract

Vitamin D was investigated as a prognostic biomarker in COVID-19, in relation to both disease susceptibility and outcomes in infected individuals. Patients admitted to the hospital with a confirmed COVID-19 diagnosis were included if they had a vitamin D measurement prior to hospitalization. Using age- and sex-matched controls, vitamin D levels were investigated for an association with COVID-19 related hospitalizations. Further, vitamin D levels were investigated for an association with 30-day mortality in hospitalized COVID-19 patients. Additionally, three meta-analyses were conducted, investigating the association of vitamin D with the following outcomes: Having a positive SARS-CoV-2 test, hospitalization with COVID-19, and mortality in COVID-19 patients. A total of 685 hospitalized COVID-19 patients were included in the single-center study. Compared to controls, they had higher vitamin D levels. Unadjusted analysis of these 685 cases found higher vitamin D levels associated with increased 30-day mortality. This association disappeared after adjusting for age. In the fully adjusted model, no association between vitamin D and 30-day mortality was found. The meta-analyses found significant associations between lower vitamin D and having a positive SARS-CoV-2 test, and mortality among hospital-admitted COVID-19 patients. The relationship between lower vitamin D and COVID-19 related hospital admissions trended towards being positive but was not statistically significant. Many factors seem to influence the associations between vitamin D and COVID-19 related outcomes. Consequently, we do not believe that vitamin D in and of itself is likely to be a clinically useful and widely applicable predictor for the susceptibility and severity of COVID-19 infections.