Vitamin D and COVID-19: evidence and recommendations for supplementation

Abstract

Vitamin D is a hormone that acts on many genes expressed by immune cells. Evidence linking vitamin D deficiency with COVID-19 severity is circumstantial but considerable-links with ethnicity, obesity, institutionalization; latitude and ultraviolet exposure; increased lung damage in experimental models; associations with COVID-19 severity in hospitalized patients. Vitamin D deficiency is common but readily preventable by supplementation that is very safe and cheap. A target blood level of at least 50 nmol l^-1, as indicated by the US National Academy of Medicine and by the European Food Safety Authority, is supported by evidence. This would require supplementation with 800 IU/day (not 400 IU/day as currently recommended in UK) to bring most people up to target. Randomized placebo-controlled trials of vitamin D in the community are unlikely to complete until spring 2021-although we note the positive results from Spain of a randomized trial of 25-hydroxyvitamin D3 (25(OH)D3 or calcifediol) in hospitalized patients. We urge UK and other governments to recommend vitamin D supplementation at 800-1000 IU/day for all, making it clear that this is to help optimize immune health and not solely for bone and muscle health. This should be mandated for prescription in care homes, prisons and other institutions where people are likely to have been indoors for much of the summer. Adults likely to be deficient should consider taking a higher dose, e.g. 4000 IU/day for the first four weeks before reducing to 800 IU-1000 IU/day. People admitted to the hospital with COVID-19 should have their vitamin D status checked and/or supplemented and consideration should be given to testing high-dose calcifediol in the RECOVERY trial. We feel this should be pursued with great urgency. Vitamin D levels in the UK will be falling from October onwards as we head into winter. There seems nothing to lose and potentially much to gain.

Keywords: COVID-19; immunology; vitamin D.

Figure 1.

Relationship between body weight and serum vitamin D. Estimated mean serum 25-hydroxyvitamin D (25(OH)D) level, with 95% confidence interval, of greater than or equal to 1-year-old US residents (n = 31 934) from the National Health and Nutrition Examination Survey, 2003–2010. The lines at 20 ng ml⁻¹ (50 nmol l⁻¹) and 30 ng ml⁻¹ (75 nmol l⁻¹) represent the sufficiency thresholds recommended by the US National Academy of Medicine and the Endocrine Society, respectively. From [16] with permission.

Figure 2.

Estimated mean serum 25-hydroxyvitamin D (25(OH)D) level, with 95% confidence interval, of greater than or equal to 1-year-old US residents (n = 31 934), by skin colour, over the normal range of body weights from the National Health and Nutrition Examination Survey, 2003–2010. From [16] with permission.

Figure 3.

Impact of various factors, including supplementation, on blood vitamin D levels in 5382 community-dwelling Irish adults greater than or equal to 50 years. From [19] with permission.

Figure 4.

Arterial oxygen saturation in wild-type (WT) and vitamin D-deficient mice (VDD) given intratracheal lipopolysaccharide followed by 1500 IU of intra-peritoneal liquid cholecalciferol (Vigantol, VIG) rescue therapy 6 h post-injury. From [44] with permission.

Figure 5.

(a) Blockade of angiotensin II signalling ameliorates lung injury in vitamin D receptor-null mice. Wild-type (WT) and VDR knock-out (KO) mice were pretreated with saline or L1–10 (angiotensin II antagonist) for two weeks followed by LPS challenge. Interleukin-6 (IL-6) levels in the broncho-alveolar lavage (BAL) fluid. ***, p < 0.001 versus LPS-treated WT; ###, p < 0.001 versus LPS-treated KO; n = 5–6 in each genotype. Ctrl, control. IL-6 levels were approximately fivefold increased in vitamin D receptor knock-out mice. From [42] with permission. (b) VitD receptor (VDR) inactivation leads to severe acute lung injury after lipopolysaccharide (LPS) challenge. Wild-type (WT) and VDR knock-out (KO) mice were treated with saline or LPS (20 mg kg⁻¹, intra-peritoneal injection). The lungs in animals lacking vitamin D receptor (KO) show a marked increase in lung congestion with inflammatory cells following lipopolysaccharide (LPS) challenge. From [42] with permission.

Figure 6.

Seasonality of respiratory virus infection in temperate regions. Respiratory viruses are classified in three groups according to their seasonal epidemics. Influenza virus, human coronavirus (HCoV) (such as strains OC43, HKU1, 229E and NL63), and human respiratory syncytial virus (RSV) show peaks in winter (winter viruses). Adenovirus, human bocavirus (HBoV), parainfluenza virus (PIV), human metapneumovirus (hMPV) and rhinovirus can be detected throughout the year (all-year viruses). From [51] with permission.

Figure 7.

Historic prevalence of vitamin D deficiency in European countries compared with mortality per million from COVID-19 by April 2020. From [54] with permission.

Figure 8.

Vitamin D status and risk of all-cause mortality in a cohort of 365 530 individuals with median follow-up 8.9 years from the UK Biobank. From [82] with permission.