Table 1.
Selected studies evaluating the role of vitamin D status or vitamin D supplementation in patients with COVID-19
| Author | Study region | Study design | Study population | Sample size (women/men) | Age in years* | Vitamin D status | Results (RR, OR, or HR; 95% CI) |
|---|---|---|---|---|---|---|---|
| Observational studies | |||||||
| Ilie et al. [127] | 20 European countries | Ecological study | Populations with data on mean 25(OH)D concentrations and COVID-19 | NA | NA | 25(OH)D: 22.7 ± 4.2 ng/mL |
Negative correlation between 25(OH)D concentrations and • COVID-19 cases: r = –0.44 (P = 0.05) • COVID-19 mortality: r = –0.44 (P = 0.05) |
| Meltzer et al. [7] | USA | Retrospective cohort study | Individuals with 25(OH)D or 1,25(OH)2D concentrations | 489 (366/123) | 49.2 ± 18.4 | Vitamin D deficiency (< 20 ng/mL): 35% | Increased risk of test ( +) for COVID-19 when vitamin D likely deficient vs likely sufficient (RR, 1.77; 1.12–2.81) |
| Kaufman et al. [6] | USA | Retrospective cohort study | Individuals tested for COVID-19 with matching 25(OH)D results from the preceding 12 months | 191,779 (130,473/61,306) | 54.0 (40.4–64.7) | Mean seasonally adjusted 25(OH)D: 31.7 ± 11.7 ng/mL |
Association of vitamin D concentrations with SARS-CoV-2 positivity rates • 25(OH)D < 20 ng/mL (39,190 patients): 12.5%; 12.2%–12.8% • 25(OH)D 30–34 ng/mL: 8.1%; 7.8%–8.4% • 25(OH)D ≥ 55 ng/mL: 5.9%; 5.5%–6.4% |
| Merzon et al. [12] | Israel | Population-based retrospective study | Individuals tested for COVID-19 with plasma 25(OH)D concentrations | 7,807 (4,573/3,234) | COVID-19 test ( +): 35.6 (34.5–36.7); (–): 47.4 (46.9–47.9) | 25(OH)D < 20 ng/mL: 13%; 25(OH)D 20–29 ng/mL: 72% |
In patients with vitamin D < 30 ng/mL, • Likelihood of COVID-19: aOR, 1.50; 1.13–1.98 • Likelihood of hospitalization for COVID-19: aOR, 1.95; 0.99–4.78 |
| Radujkovic et al. [13] | Germany | Consecutive case series with prospectively collected data | Hospitalized patients with symptomatic COVID-19 | 185 (90/95) | 60 (49–70) | 25(OH)D: 16.6 (12.4–22.5) ng/mL | Association of low vitamin D (< 12 ng/mL) with IMV and/or death (HR, 6.12; 2.79–13.42) and death (HR, 14.73; 4.16–52.19) |
| Jain et al. [56] | India | Prospective observational study | (A) Asymptomatic patients with COVID-19 or (B) COVID-19 patients requiring ICU admission | 154 (69/95) | (A) 42.3 ± 6.4; (B) 51.4 ± 9.1 | (A) 25(OH)D: 27.9 ± 6.2 ng/mL; (B) 25(OH)D: 14.4 ± 5.8 ng/mL |
Markedly low vitamin D concentrations in patients with severe COVID-19 In patients with vitamin D deficiency • Higher levels of IL-6, ferritin, and TNF-α • Higher fatality rate (21% vs 3%) |
| Hastie et al. [129] | UK | Retrospective study | UK Biobank participants | NA | NA | NA | No association with 25(OH)D concentrations with severe COVID-19 or mortality |
| Hernandez et al. [151] | Spain | Retrospective case–control study | Patients hospitalized for COVID-19 | 216 (86/130); 19 were on vitamin D supplementation | Vitamin D supplementation ( +): 61.0 (47.5–70.0); (–): 60.0 (59.0–75.0) | Vitamin D < 20 ng/mL: 82% |
Higher prevalence of vitamin D < 20 ng/mL than population-based controls (82.2% vs 47.2%, P < 0.01) 25(OH)D concentrations • Inverse correlation with ferritin (P = 0.01) and D-dimer levels (P = 0.03) • No relationship with COVID-19 severity |
| Angelidi et al. [152] | USA | Retrospective cohort study | Patients hospitalized for COVID-19 | 144 (80/64) | 66 (55–74) | 25(OH)D: 30.4 ± 17.0 ng/mL |
Association with mortality • 25(OH)D < 30 ng/mL vs ≥ 30 ng/mL: 9.2% vs 25.3%, P = 0.02 • Association of increased vitamin D concentrations with in-hospital mortality (OR, 0.94; 0.90–0.98) and IMV (OR, 0.96; 0.93–0.99) |
| Abdollahi et al. [153] | Iran | Prospective case–control study | Hospitalized patients tested (A) positive or (B) negative for COVID-19 | 402 (132/270) |
(A) 48.0 ± 17.0; (B) 46.3 ± 13.5 |
(A) 25(OH)D: 24 (19–29) ng/mL; (B) 25(OH)D: 26 (21–35) ng/mL |
Association of low vitamin D concentrations with COVID-19 infection (P = 0.02) |
| Reis et al. [131] | Brazil | Prospective cohort study | Patients hospitalized for moderate-to-severe COVID-19 | 220 (103/117) | 55.1 ± 14.6 | 25(OH)D < 10 ng/mL: 16 (7%); > 10 ng/mL: 204 (93%) |
Hospital length of stay • 25(OH)D < 10 ng/mL vs ≥ 10 ng/mL: 9.0 days vs 7.0 days, P = 0.057 • No association with IMV and mortality |
| Experimental studies | |||||||
| Castillo et al. [16] | Spain | Pilot RCT (intervention: high-dose oral calcifediol) | Patients hospitalized for COVID-19 | 76 (31/45) | 53 ± 10 | NA |
Intervention vs control • Reduced requirements for ICU admission (P < 0.001) |
| Rastogi et al. [17] | India | RCT (intervention: 60,000 IU/day with therapeutic target of 25(OH)D > 50 ng/mL) | Asymptomatic or mild COVID-19 patients with 25(OH)D < 20 ng/mL | 40 (20/20) | Intervention group: 50.0 (36.0–51.0); Control group: 47.5 (39.3–49.2) | 25(OH)D: Intervention group 8.6 ng/mL; Control group 9.5 ng/mL* |
Intervention vs control • Higher negative conversion of SARS-CoV-2 RNA (62.5% vs 20.8%; P < 0.02) • A significant decrease in fibrinogen levels (difference: 0.70 ng/mL, P = 0.007) |
| Annweiler et al. [154] | France | Quasi-experimental study (intervention: bolus vitamin D administration) | Frail elderly nursing-home residents with COVID-19 | 66 (15/51) | 87.7 ± 9.0 | NA |
Intervention vs control • Survival rate: 82.5% vs 44.4%, P = 0.023 • Mortality: aHR, 0.11; 0.03–0.48, P = 0.002 |
| Annweiler et al. [121] | France | Quasi-experimental study: vitamin D supplementation (A) over the preceding year or (B) after COVID-19 diagnosis | Patients hospitalized for COVID-19 in a geriatric unit | 77 (38/39) | 88 (85–92) | NA |
Survival at day 14 • (A) vs (B): 93.1% vs 81.2%, P = 0.33 • (A) vs control: 93.1% vs 68.7%, P = 0.02 Mortality for 14 days • (A) vs control: aHR, 0.07; 0.01–0.61 • (B) vs control: aHR, 0.37; 0.06–2.21 |
| Murai et al. [125] | Brazil | RCT (intervention: a single oral dose of 200,000 IU of vitamin D3) | Patients hospitalized for COVID-19 who were moderately to severely ill | 237 (104/133) | 56.2 ± 14.4 | 25(OH)D: 20.9 ± 9.2 ng/mL |
Vitamin D3 vs placebo • Length of hospital stay: 7.0 days vs 7.0 days • In-hospital mortality: 7.6% vs 5.1%, P = 0.43 • ICU admission: 16.0% vs 21.2%, P = 0.30 • IMV: 7.6% vs 14.4%, P = 0.09 |
| Lakkireddy et al. [155] | India | RCT (intervention: 60,000 IU/day of vitamin D) | Patients hospitalized for COVID-19 and vitamin D < 30 ng/mL | 87 (22/65) | 45 ± 13 | Intervention group 16 ± 6 ng/mL; Control group: 17 ± 6 ng/mL |
Inflammatory markers (CRP, LDH, IL-6, ferritin, N/L ratio) • Significant reduction in the intervention group (P < 0.01) but not in the control group (P > 0.05) except CRP |
| Sánchez-Zuno et al. [122] | Mexico | RCT (intervention: 10,000 IU/day of vitamin D3) | Asymptomatic or mildly symptomatic patients with COVID-19 | 42 (22/20) | 43 (20–74) | Vitamin D: 22.4 (12.1–45.9) ng/mL |
• > 3 symptoms of COVID-19 vs control: 0% vs 4%, P = 0.04 • SARS-CoV-2 RNA positivity vs control: 0% vs 5%, P = 0.47 • SARS-CoV-2 seropositivity vs. control: 72.7% vs 75.0%, P > 0.05 |
| Mendelian randomization study | |||||||
| Butler-Laporte et al. [132] | Two sample Mendelian randomization study | Individuals of European ancestry | GWAS of genetic variants associated with vitamin D concentrations: 443,734 (including 401,460 from the UK Biobank); GWAS of COVID-19 susceptibility, hospitalization, and severe diseases: 1,299,010 (from the COVID-19 Host Genetic Initiative) |
Genetically increased 25(OH)D concentrations by one SD (logarithmic scale) • No association with COVID-19 susceptibility: OR, 0.95; 0.84–1.08 • No association with hospitalization for COVID-19: OR, 1.09; 0.89–1.33 • No association with severe COVID-19: OR, 0.97; 0.77–1.22 |
|||
*(mean ± SD or overall range), 1,25(OH)2D 1,25-hydroxyvitamin D, 25(OH)D 25-hydroxyvitamin D, aHR adjusted hazard ratio, aOR adjusted odds ratio, CI confidence interval, COVID-19 coronavirus disease 2019, CRP C-reactive protein, GWAS genome-wide association study, HR hazard ratio, ICU intensive care unit, IMV invasive mechanical ventilation, IL-6 interleukin-6, LDH lactate dehydrogenase, NA not applicable, N/L ratio neutrophil/lymphocyte ratio, OR odds ratio, RCT randomized controlled trial, RR relative risk, SARS-CoV-2 severe acute respiratory syndrome coronavirus 2, SD standard deviation, TNF-α tumor necrosis factor-α