ORIGINAL RESEARCH published: 24 September 2021 doi: 10.3389/fpubh.2021.758347 Previous Vitamin D Supplementation and Morbidity and Mortality Outcomes in People Hospitalised for COVID19: A Cross-Sectional Study Juan Antonio Arroyo-Díaz 1 , Josep Julve 2 , Bogdan Vlacho 3 , Rosa Corcoy 4,5,6 , Paola Ponte 7 , Eva Román 7 , Elena Navas-Méndez 8 , Gemma Llauradó 9,10 , Edited by: Josep Franch-Nadal 3,10,11 , Pere Domingo 1* and Didac Mauricio 3,5,10,12* Zisis Kozlakidis, International Agency for Research on 1 Infectious Diseases, Department of Medicine, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 2 Institut de Recerca Cancer (IARC), France de l’Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 3 Diabetis des de l’Atenció Primària (DAP)-Cat Group, Unitat de Reviewed by: Suport a la Recerca Barcelona, University Institute for Primary Care Research (Institut Universitari per a la recerca a l’Atenció Ronan Lordan, Primària (IDIAP) Jordi Gol), Barcelona, Spain, 4 Center for Biomedical Research in the Network in Bioengineering, University of Pennsylvania, Biomaterials, and Nanomedicine, Madrid, Spain, 5 Department of Endocrinology and Nutrition, Hospital de la Santa Creu i United States Sant Pau, Barcelona, Spain, 6 Department of Medicine, Faculty of Medicine, Autonomous University of Barcelona, Barcelona, William B. Grant, Spain, 7 Department of Medicine, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 8 Unitat de Suport a la Recerca Sunlight Nutrition and Health Barcelona, University Institute for Primary Care Research (IDIAP Jordi Gol), Barcelona, Spain, 9 Department of Endocrinology Research Center, United States and Nutrition, Hospital del Mar, Parc de Salut Mar, Barcelona, Spain, 10 Centro de Investigación Biomédica en Red (CIBER) of Diabetes and Associated Metabolic Diseases, Instituto de Salud Carlos III, Madrid, Spain, 11 Primary Health Care Center *Correspondence: Raval Sud, Gerència d’Atenció Primaria, Institut Català de la Salut, Barcelona, Spain, 12 Departament of Medicine, Universitat Pere Domingo de Vic - Universitat Central de Catalunya, Barcelona, Spain pdomingo@santpau.cat Didac Mauricio didacmauricio@gmail.com Aim: The study aim was to assess the association of vitamin D supplementation before hospital admission and severe outcomes in subjects admitted for COVID-19. Specialty section: This article was submitted to Methods: We performed a cross-sectional analysis of pseudonymised medical record Infectious Diseases - Surveillance, data from subjects admitted to the Hospital de la Santa Creu i Sant Pau (Barcelona, Prevention and Treatment, a section of the journal Spain) for COVID-19 during March and April 2020. The composite primary study outcome Frontiers in Public Health was defined as death and/or invasive mechanical ventilation (IMV). Association between Received: 13 August 2021 risk factors and study outcomes was evaluated by bivariate analysis, followed by logistic Accepted: 30 August 2021 Published: 24 September 2021 regression analysis. Citation: Results: In total, 1,267 persons were hospitalised during the observation period. Arroyo-Díaz JA, Julve J, Vlacho B, Overall, 14.9% of the subjects were on active vitamin D supplementation treatment Corcoy R, Ponte P, Román E, Navas-Méndez E, Llauradó G, before admission. The subjects in the vitamin D group were significantly older Franch-Nadal J, Domingo P and than subjects without vitamin D supplementation. We observed higher rates of Mauricio D (2021) Previous Vitamin D Supplementation and Morbidity and the primary outcome (death and/or IMV) among the persons with previous use Mortality Outcomes in People of vitamin D (30.1 vs. 22.9% in those not receiving treatment). In the bivariate Hospitalised for COVID19: A analysis, previous use of vitamin D was positively associated with death and/or Cross-Sectional Study. Front. Public Health 9:758347. IMV [odds ratio (OR): 1.45 95% CI: 1.03; 2.04]; however, after adjustment for doi: 10.3389/fpubh.2021.758347 other risk factors this association disappeared (OR: 1.09 95%CI: 0.65; 1.81). Frontiers in Public Health | www.frontiersin.org 1 September 2021 | Volume 9 | Article 758347 Arroyo-Díaz et al. Vitamin D Supplementation and Severity of COVID19 Conclusion: We did not find an association between vitamin D supplementation before hospital admission and death and/or IMV in subjects admitted for COVID-19. The age and the burden of age-associated comorbidities were independently associated with the in-hospital events. Keywords: COVID-19, hospitalisation, mortality, invasive mechanic ventilation, vitamin D INTRODUCTION of acquiring COVID-19 infection (10, 11). Similar findings were observed in two other observational studies from the US. In From the start of the COVID-19 pandemic to July 2021, more one of them, the probability of testing positive for COVID-19 than 191,158,708 new cases and ∼4.2 million of deaths have increased with decreasing levels of vitamin D among black been reported worldwide (1). Although several different vaccines women (12), while in the other among black individuals, lower are in use, the number of COVID-19 cases globally remains levels of vitamin D were associated with increased risk of high, and deaths continue to increase (2). The United States COVID-19 (13). Recently, the results of two meta-analyses have and the European Union drug regulatory agencies have already been reported. One of them included a total of 29 observational authorised new indications for treatments such as remdesivir, studies, with exclusion of those without recent measurement lopinavir/ritonavir and interferon for subjects hospitalised with of vitamin D levels; this meta-analysis found lower levels of COVID-19. However, little or no effect on overall mortality, vitamin D among subjects with an active COVID-19 infection initiation of ventilation and duration of hospital stay was and also among those with severe COVID-19 (14). However, in observed in the corresponding clinical trials (3). In the the other meta-analysis that included 31 observational studies, RECOVERY clinical trial, dexamethasone reduced the mortality the authors did not find a significant association between vitamin among subjects hospitalised for COVID-19 receiving either D deficiency and COVID-19 health outcomes (15). It should invasive mechanical ventilation (IMV) or oxygen alone (4). be underlined that the design of these observational studies During the pandemic, different therapies have been explored does not allow a cause-effect relationship between vitamin D to prevent or treat the disease, including the use of vitamin status and the risk of COVID-19 or the outcomes of the D supplementation. disease. Regarding the relationship between vitamin D status and Vitamin D is a fat-soluble vitamin that is either formed in the COVID-19 outcomes, several observational studies found that skin or ingested in the diet; this vitamin enters the bloodstream patients with COVID-19 requiring hospitalisation or those with and travels to the liver and kidney where it is hydroxylated on severe disease had lower vitamin D concentration or vitamin D carbons 25 and 1 to form 25-hydroxyvitamin D (25(OH)D) and deficiency (16–19). In this respect, a systematic review and meta- 1,25-dihydroxy vitamin D (1,25(OH)D), respectively (5). The analysis of 39 studies concluded that despite high heterogeneity latter is the active hormone with a wide range of effects. Both and methodological differences, there was a relationship between children and adults are at risk of developing vitamin D deficiency, low 25 (OH) D concentrations and SARS-CoV-2 infection, which is very common globally. For instance, in Spain 33.9% of severity of the disease, and mortality (20). However, the authors the population is at risk for vitamin D deficiency (6). concluded that further studies should investigate the association In addition to its actions on calcium absorption and bone between vitamin D and COVID-19, especially in subgroups mineralization, this vitamin has pleiotropic actions, including of age and sex (20). Moreover, two quasi-experimental studies the regulation of immune responses and inflammation. It has and a pilot randomised clinical trial reported that vitamin D3 been hypothesised that its deficiency can diminish immune supplementation was associated with less severe COVID-19 (7, responses to respiratory viruses (7). A favourable effect of 8, 21). vitamin D supplementation on the Acute Respiratory Distress As we are in need of further evidence on the association of Syndrome (ARDS) has been reported, which may be elicited vitamin D and COVID-19 outcomes, we undertook the current through activation of the vitamin D receptor (VDR) signalling study with the aim of assessing the association of vitamin pathway, and a consequential decrease of cytokine/chemokine D supplementation prior to hospital admission and adverse hypersecretion, modulation of the activity of neutrophils and outcomes in subjects admitted for COVID-19. preservation of the integrity of the pulmonary epithelial barrier (8). In addition, vitamin D has been implicated in preventing or improving adverse outcomes of COVID-19 by activating or METHODS repressing genes via the VDR that are involved in regulating the Study Design and Settings renin-angiotensin system (RAS) or innate and adaptive cellular We analysed cross-sectional data from hospitalised individuals immunity (9). infected with SARS-CoV-2, stratified by previous vitamin D The effects of vitamin D in preventing COVID-19 infection supplementation before hospital admission. Data were obtained have been addressed in a few observational studies with from an anonymised electronic health records database from the discordant observations. An initial observational study reported Hospital de la Santa Creu i Sant Pau (Barcelona, Spain). The that vitamin D deficiency was associated with an increased risk database included retrospective clinical information, information Frontiers in Public Health | www.frontiersin.org 2 September 2021 | Volume 9 | Article 758347 Arroyo-Díaz et al. Vitamin D Supplementation and Severity of COVID19 TABLE 1 | Baseline characteristics of all subjects and according to vitamin D supplements. Characteristic All subjects N = 1,267 Group not receiving vitamin D supplements Group receiving vitamin D supplements p-value N = 1,078 N = 189 Age, mean (SD), years 64.7 (16.3) 63.2 (16.3) 73.3 (13.7) <0.001 Sex (male), n (%) 696 (54.9) 634 (58.8) 62 (32.8) <0.001 Smoking, n (%) 129 (10.2) 111 (10.3) 18 (9.52) 0.76 Comorbidities, n (%) Hypertension 602 (47.5) 484 (44.9) 118 (62.4) <0.001 Hyperlipidaemia 509 (40.2) 404 (37.5) 105 (55.6) <0.001 Obesity 567 (44.8) 488 (45.3) 79 (41.8) 0.38 Diabetes 252 (19.9) 205 (19.0) 47 (24.9) 0.07 Cardiovascular diseases 152 (12.0) 118 (10.9) 34 (18.0) 0.01 Heart failure 107 (8.45) 80 (7.42) 27 (14.3) 0.003 Atrial fibrillation 126 (9.94) 100 (9.28) 26 (13.8) 0.07 Chronic kidney disease 155 (12.2) 105 (9.74) 50 (26.5) <0.001 COPD 219 (17.3) 179 (16.6) 40 (21.2) 0.13 Charlson comorbidity Index (>2) 687 (54.2) 525 (48.7) 162 (85.7) <0.001 Laboratory parameters, mean (SD) Lymphocytes (µL) 1,000 (2,000) 970 (790) 1,200 (4,900) 0.28 C-reactive protein (mg/L) 145 (110) 145 (111) 148 (109) 0.72 Interleukin 6 (pg/mL) 293 (752) 292 (775) 300 (412) 0.97 Ferritin (µg/L) 1,832 (6,115) 1,840 (6,473) 1,789 (3,440) 0.93 D-dimer (ng/mL) 6,200 (1,7900) 5,900 (18,000) 7,400 (17,000) 0.34 Blood lactate (mmol/L) 1.11 (1.23) 1.09 (1.09) 1.22 (1.87) 0.24 PaO2 (mmHg) 77.0 (38.2) 76.0 (34.7) 82.9 (54.7) 0.046 PaO2 /FiO2 (mmHg) 283 (120) 284 (121) 280 (110) 0.77 COPD, chronic obstructive pulmonary disease; SD, standard deviation. on admission, diagnostic and procedure codes, prescribed Outcome events during the hospital stay were predefined medications, and laboratory parameters from 1,267 subjects as: death, admission to the intensive care unit (ICU), days admitted between March and April 2020. Outcome events were of ICU stay, hospitalisation duration, and invasive mechanical documented from hospital admission until discharge or death. ventilation (IMV). The primary study outcome was predefined The study was approved by the Ethics Committee of the as death and/or IMV. Hospital de la Santa Creu i Sant Pau (Re. Nr. HSCSP-20/117). Statistical Methods Inclusion and Exclusion Criteria We analysed demographic and clinical characteristics according The study enrolled subjects between 19 and 101 years with to vitamin D supplementation prior to admission. Quantitative a confirmed PCR SARS-CoV-2 diagnostic test [asymmetric variables were summarised according to their distribution reverse-transcription polymerase chain reaction (RT-PCR) of [median, first and third quartile or mean and standard deviation ART] as documented in the medical record. (±SD)] and or categorical variables as frequency, number and percent (%). The association between events (mechanical Study Variables ventilation, mortality, and mortality or mechanical ventilation) On admission, the following baseline variables were collected: and previous use of vitamin D was evaluated by bivariate age, sex, smoking status, information on comorbidities recorded analysis, followed by logistic regression analysis adjusted for in the electronic records [hypertension, hyperlipidaemia, obesity, sex and age and associated risk factors (age, sex, obesity, diabetes, cardiovascular disease (CVD), heart failure, chronic hypertension, diabetes, hyperlipidaemia, CVD, CKD, COPD, kidney disease (CKD), chronic obstructive pulmonary disease cancer, Charlson comorbidity index and PaO2 /FiO2 index). (COPD), and the Charlson comorbidity index], and blood Several models of interest were tested, with the sequential laboratory parameters (i.e., lymphocytes, C-reactive protein, inclusion of different covariates and the estimated differences interleukin-6, ferritin, D-dimer, lactate dehydrogenase, PaO2 , expressed as adjusted odds ratio (AOR) and their respective 95% and the PaO2 /FiO2 index). Vitamin D exposure was defined as a confidence intervals (CI). Additionally, we performed the same current prescription of any form of active vitamin D supplement bivariate analysis followed by logistic regression analysis (using at the time of admission to the hospital. the same previously mentioned risk factors) for a subgroup of Frontiers in Public Health | www.frontiersin.org 3 September 2021 | Volume 9 | Article 758347 Arroyo-Díaz et al. Vitamin D Supplementation and Severity of COVID19 TABLE 2 | Intra hospital events overall and according to Vitamin D supplements in the whole study population and in the subgroup of 60 years or more. All ≥60 years All subjects Group not receiving Group receiving All subjects Group not receiving Group receiving Vitamin D vitamin vitamin vitamin D supplements D supplements D supplements supplements N = 1,267 N = 1,078 N = 189 N = 797 N = 638 N = 159 Death, n (%) 217 (17.1) 167 (15.5)* 50 (26.5)* 201 (25.2) 154 (24.1) 47 (29.6) IMV, n (%) 124 (9.8) 113 (10.5) 11 (5.8) 93 (11.7) 85 (13.3) 8 (5.0) Death and/or IMV, 304 (23.9) 247 (22.9)* 57 (30.1)* 262 (32.8) 211 (33.0) 51 (32.0) n (%) ICU, n (%) 146 (11.5) 133 (12.3) 13 (6.9) 104 (13.0) 95 (14.9) 9 (5.7) Days in ICU, mean 1.87 (7.1) 2.00 (7.4) 1.08 (5.2) 2.32 (7.9) 2.65 (8.4) 1.01 (5.3) (SD) Hospitalisation 8.75 (10.4) 8.91 (10.7) 7.86 (8.5) 9.17 (11.5) 9.56 (12.2) 7.64 (8.1) duration (days), mean (SD) ICU, intensive care unit; IMV, invasive mechanical ventilation. *p-value < 0.05. subjects who were 60 years of age or older. Data management and and/or IMV) and for death: both events were more frequent statistical analyses were performed using the R statistical software among subjects with previous use of vitamin D (22.9 vs. 30.1% version 3.6.1 (https://www.r-project.org/). and 15.5 vs. 26.5%, respectively). However, the percentage of subjects admitted to the intensive care unit (ICU), days in RESULTS ICU, as well as duration of hospitalisation was lower in the vitamin D group. During March–April 2020 (corresponding to the first COVID The sub-analysis of subjects over 60 years showed a wave), 1,267 persons were hospitalised due to COVID- slightly higher percentage for the death and/or IMV in the 19 at the Hospital de la Santa Creu i Sant Pau. One group receiving vitamin D. However, no statistically significant hundred and eighty-nine (14.9%) subjects were receiving differences for the primary outcome were observed between active vitamin D supplementation on admission (Vitamin groups in this subgroup. D group). Supplementary Figure 1 presents the flowchart of the study. Factors Associated With Events by Previous Use of Vitamin D Baseline Characteristics The bivariate and logistic regression analysis results for main The baseline clinical characteristics of the study subjects are composite outcome are presented in Supplementary Table 1 and shown in Table 1. The mean age of subjects was 64.7 years, Figure 1A. In the bivariate analysis, we observed that vitamin and the majority were male (54.9%). However, we observed D and most of the risk factors and co-morbidities (older age, differences between the groups for age and sex, whereby subjects hypertension, diabetes, hyperlipidaemia, CVD, CKD, COPD, in the vitamin D group were older and were more frequently cancer, higher Charlson comorbidity index score) were positively female. Subjects in the vitamin D group had a higher rate of associated with the composite primary outcome (death and/or most concomitant diseases of interest, except for obesity, which IMV). We observed negative associations only for the PaO2 /FiO2 was more frequent in the group without vitamin D. Regarding Index and female sex. the laboratory parameters, the majority of the parameters In the logistic regression analyses, after adjustment for all (lymphocytes, C-reactive protein, interleukin 6, D-dimer, Blood associated risk factors, only age, diabetes, CVD, and a higher lactate, PaO2 , and PaO2 /FiO2 Index) were higher in the vitamin Charlson comorbidity index score were positively associated D group, however, no statistically significant differences were with the primary outcome. We did not observe a significant observed between the groups, except for PaO2 where borderline association between previous use of vitamin D and death and/or differences were observed. IMV (AOR: 1.09, 95% CI: 0.65; 1.81). Obesity and PaO2 /FiO2 index were negatively associated with the main composite Events During in-hospital Stay outcome in this analysis. The events and complications during the hospital stay are The results for the separate events (death and IMV) presented in Table 2. A total of 217 (17.1%) subjects died in all subjects are presented in Supplementary Table 1 and during in-hospital stay. We observed statistically significant Figures 1C,D. Similar associations were observed in the differences between groups in the primary study outcome (death bivariate analysis for mortality, except for obesity, where a Frontiers in Public Health | www.frontiersin.org 4 September 2021 | Volume 9 | Article 758347 Arroyo-Díaz et al. Vitamin D Supplementation and Severity of COVID19 FIGURE 1 | Risk factors for severity outcomes. Frontiers in Public Health | www.frontiersin.org 5 September 2021 | Volume 9 | Article 758347 Arroyo-Díaz et al. Vitamin D Supplementation and Severity of COVID19 negative association with death was observed. For the IMV Moreover, in the logistic regression models, PaO2 /FIO2 ratio outcome, obesity was positively associated, while vitamin D and obesity were associated with the study outcomes. These supplementation was negatively associated. No associations were findings are in line with other previously published reports. The observed between death and/or IMV and vitamin D use when the PaO2 /FIO2 ratio has been reported as an independent factor model was adjusted for all factors. related to death in COVID-19 subjects receiving intensive care (22). In addition, obesity has been associated with substantially Factors Associated With Events by increased risks of severe outcomes in patients with COVID- Previous Use of Vitamin D Among Persons 19 (23). Therefore, it is perhaps surprising that in the logistic regression models, we found that the obesity was negatively at Least 60 Years or Older associated with the study outcomes. This could be due to the Supplementary Table 2 and Figure 1B show the bivariate and J-shaped association between BMI and the mortality curve. logistic regression analyses for persons at least 60 years or It was already reported that survival rates of subjects with older. In subjects aged 60 years or older, the same factors (age, moderate obesity were higher than for those with normal body hypertension, diabetes, CVD, CKD, cancer, a higher Charlson mass index (BMI), overweight or severe obesity subjects (23). comorbidity index score) were positively associated with the However, a clear limitation of our study is that data on BMI primary outcome (death and/or IMV), although with a lower were not available and the analyses were only based on a clinical unadjusted OR. In the adjusted logistic regression analyses, recorded diagnosis of obesity in the clinical records, which is only age and diabetes were positively associated with the not a solid variable to analyse the contribution of adiposity on main outcome. the outcomes. So far, regarding studies that tested the use of vitamin D DISCUSSION as a preventive therapy for COVD-19, one pilot randomised clinical trial in Spain with 76 participants reported that The results of the present cross-sectional study of 1,267 admission to the ICU was reduced in the group receiving a persons infected with COVID-19 during March and April high dose of vitamin D (7). Two additional quasi-experimental 2020 showed higher in-hospital mortality and/or IMV among studies from France observed that only regular use of vitamin subjects treated with vitamin D prior to hospital admission in D was associated with less severe COVID-19 and a lower the crude analysis. However, no significant association between mortality rate (9, 21), while the administration of vitamin D use of vitamin D supplementation prior to hospitalisation after COVID-19 diagnosis did not affect severity outcomes and risk of death and/or IMV was observed in the fully (9). It is important to mention that the authors in the adjusted model. Instead, age and the burden of age-associated French study had full information on supplemental vitamin D comorbidities were independently associated with the in- exposure (dose and duration of supplementation) (9), which hospital events. was not the case of our study, resulting again in another The importance of vitamin D on the clinical severity of important limitation. COVID-19 remains controversial. So far, the studies published Several systematic reviews and meta-analyses have been on this topic are heterogeneous in their methodology and the published on this topic as well. In a systematic review including findings obtained. Some studies support the association between 4 observational studies to assess the association between vitamin D and favourable disease outcomes, whereas others vitamin D supplementation or level with COVID-19 infection did not observe significant associations. For example, in an susceptibility and outcomes (clinical course, morbidity and observational study from Spain with 216 hospitalised subjects mortality outcomes), no robust association was found between with COVID-19, the researchers did not observe an association vitamin D and COVID-19 severity of symptoms or mortality between vitamin D status (circulating levels of 25OHD) and (24). Another systematic review with 11 studies about vitamin the severity of COVID-19 infection (ICU admission, the need D levels/supplementation and risk of COVID-19 infection, for mechanical ventilation, or mortality); these results are in adverse outcomes and possible benefits among subjects aged line with the results observed in our study where we did 60 years or over concluded that supplementation with vitamin not find any signal for an association (17). Despite the small D during COVID-19 reduced the risks for mortality, high number of subjects (only 19) identified as receiving vitamin D flow oxygen therapy needs, or ICU support (25). Nevertheless, supplementation in that study, similarities could be observed the authors pointed out the importance of the characteristics in terms of demographic and clinical characteristics with of the supplementation regimen, with aspects that remain subjects included in current study, such as higher percentage unresolved such as the dose, frequency of administration, and of women and more severe comorbidity profile (hypertension, duration (25). In our sub-analysis of subjects aged 60 years cardiovascular diseases) compared with COVID-19 subjects not or over, the composite outcome of death and/or IMV was receiving vitamin D supplementation. Moreover, the authors more frequent for both groups (on vitamin D and without) reported lower percent of persons with PaO2 /FIO2 ratio <300, than in the total study group; however, no differences were lower median serum ferritin and less need for tocilizumab observed regarding the use of previous vitamin D. Independently therapy among the persons with oral vitamin D supplements of previous vitamin D intake, evidence suggests that subjects (17). In contrast, we did not observe any differences in the over 60 years are more prone to severe complications and PaO2 /FIO2 ratio and ferritin levels between our study groups. longer duration of COVID-19 compared with younger persons Frontiers in Public Health | www.frontiersin.org 6 September 2021 | Volume 9 | Article 758347 Arroyo-Díaz et al. Vitamin D Supplementation and Severity of COVID19 (age 60 or younger) (26–28); this was also observed in our the recommendation on the dose and duration of the study. Subjects on vitamin D in the current study were supplementation with vitamin D should be based on the levels 10 years older compared with subjects without vitamin D of vitamin D in the given subject. In subjects from the general supplementation. A recent meta-analysis of three randomised population with severe deficiency (<10 ng/mL) and a target controlled trials found no effect of vitamin D supplementation level of vitamin D >25 ng/mL, the recommended dose is either in participants with COVID-19 for all-cause mortality (an 16.000 IU/weekly during 5 weeks for calcifediol or 50.000 outcome in two of the studies), and the need for invasive IU/weekly during 4–6 weeks for cholecalciferol. In those with mechanical ventilation and adverse events (one of the studies) vitamin D insufficiency (10–25 ng/mL), the recommended dose (29). Moreover, the authors reported high inconsistency in for cholecalciferol is 25.000 IU/daily and 16.000 IU/monthly for reporting of adverse reactions or safety of vitamin D and a calcifediol (37). Since the group receiving vitamin D supplements lack of evidence that this supplementation prevented death from in our study consisted mainly of elder subjects (mean age 73.3 COVID-19 (29). years) and predominantly of female sex (67.2%), we assume Based on the currently published observational studies, that the dose of vitamin D prescribed by the treating physician vitamin D deficiency seems to be associated with an increased was probably in the range recommended by guidelines. Since risk of severe disease and mortality from COVID-19 (16–20, this is a cross-sectional study, the logistic regression models 22, 23, 25, 30). In a randomised clinical trial from Brazil, should be interpreted with caution; accordingly they only that included patients already hospitalised with moderate to describe the possible association between the risk factors and the severe COVID-19 in whom a single high dose of vitamin D studied outcomes. was administrated, the authors reported that it was too early to conclude whether vitamin D supplementation may improve the disease prognosis (31). Further, treatment with a high dose CONCLUSIONS of vitamin D in subjects who have already presented with complications of COVID-19 may not be effective since the organ No association between vitamin D supplementation prior to damage by virus had already initiated. The inverse relationship hospital admission and important outcomes, i.e., death and/or between the risk of acute respiratory infections and vitamin IMV, or secondary in-hospital events was observed in this cross- D blood levels is well-reported (7, 10, 32, 33). The possible sectional study of inpatients with COVID-19. Well-designed biological explanation for this effect could be the influence randomised clinical trials are needed to confirm the potential role of genes regulating the immune system and the inflammatory of vitamin D on outcomes of COVID-19. response, which might modify the susceptibility to and severity of infections (34). Vitamin D deficiency is highly prevalent, especially in older adults, mainly due to the decreased synthesis DATA AVAILABILITY STATEMENT in the skin or deficient dietary intake (35), and it could be easily The data analysed in this study is subject to the following prevented with regular supplementations (25). In our study, licences/restrictions: The data controller for Hospital de la Santa another important limitation is the lack of information on the Creu i Sant Pau does not allow the sharing of raw data. levels of vitamin D among the study subjects. Our data suggest Requests to access these datasets should be directed to Pere that the main drivers of severe outcomes in COVID-19 are Domingo, pdomingo@santpau.cat. age-associated comorbidities. Since most of our subjects taking vitamin D were older adults with a high burden of comorbidities, the potential benefits of vitamin D supplementation may be ETHICS STATEMENT very difficult to demonstrate. Additionally, it is possible that in many of these subjects the reason for prescription of vitamin D The studies involving human participants were reviewed and supplementation was a status of vitamin D deficiency. approved by the Ethics Committee of the Hospital de la Apart from those already stated, this study has additional Santa Creu i Sant Pau (Re. Nr. HSCSP-20/117). Written limitations. The database was created from anonymised, informed consent for participation was not required for this routinely collected medical data of hospitalised subjects during study in accordance with the national legislation and the the first wave of the pandemic; therefore, the amount of institutional requirements. incomplete data was high. Consequently, some important clinical variables were not considered in the analysis, such as BMI, dose and time of exposure related to pharmacological AUTHOR CONTRIBUTIONS treatment, completeness of information on medical conditions or procedures prior to admission. Moreover, we did not have JA-D, JJ, JF-N, PD, and DM: conceptualisation. EN-M: formal variables related to vitamin D such as duration of vitamin analysis. JA-D, DM, PD, ER, and PP: resources and data D supplementation, doses, adherence to supplementation or curation. BV: writing—original draught preparation. BV, RC, blood levels of vitamin D. It should be pointed out that the JA-D, GL, JF-N, PD, and DM: writing—review and editing. vitamin D intake in Spain is among the lowest compared with DM and JF-N: supervision. JA-D: project administration. other European countries (36). According to the SEIOMM All authors contributed to the article and approved the (Spanish Society for Research on Bone and Mineral Metabolism), submitted version. Frontiers in Public Health | www.frontiersin.org 7 September 2021 | Volume 9 | Article 758347 Arroyo-Díaz et al. Vitamin D Supplementation and Severity of COVID19 ACKNOWLEDGMENTS SUPPLEMENTARY MATERIAL The authors acknowledge Hospital de la Santa Creu The Supplementary Material for this article can be found i Sant Pau, Barcelona (Spain) for providing the online at: https://www.frontiersin.org/articles/10.3389/fpubh. study data. 2021.758347/full#supplementary-material REFERENCES 17. Hernández JL, Nan D, Fernandez-Ayala M, García-Unzueta M, Hernández- Hernández MA, López-Hoyos M, et al. Vitamin D status in hospitalized 1. COVID-19 Situation Update Worldwide, as of week 28, European Centre subjects with SARS-CoV-2 infection. J Clin Endocrinol Metab. (2021) for Disease Prevention and Control website. (2021). Available online at: 106:e1343–53. doi: 10.1210/clinem/dgaa733 https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-cases 18. Radujkovic A, Hippchen T, Tiwari-Heckler S, Dreher S, Boxberger M, Merle (accessed July 25, 2021). U. Vitamin D deficiency and outcome of COVID-19 subjects. Nutrients. 2. World Health Organization. COVID-19 Weekly Epidemiological Update 22. (2020) 12:2757. doi: 10.3390/nu12092757 World Heal. Organ (2021). Available online at: https://www.who.int/docs/ 19. Macaya F, Espejo Paeres C, Valls A, Fernández-Ortiz A, González del default-source/coronaviruse/situation-reports/weekly_epidemiological_ Castillo J, Martín-Sánchez J, et al. Interaction between age and vitamin D update_22.pdf (accessed July 25, 2021). deficiency in severe COVID-19 infection. Nutr Hosp. (2020) 37:1039–42. 3. WHO Solidarity Trial Consortium. Repurposed antiviral drugs for covid-19— doi: 10.20960/nh.03193 interim WHO solidarity trial results. N Engl J Med. (2021) 384:497–511. 20. Kazemi A, Mohammadi V, Aghababaee SK, Golzarand M, Clark CCT, 4. RECOVERY Collaborative Group, Horby P, Lim WS, Emberson JR, Mafham Babajafari S. Association of vitamin D status with SARS-CoV-2 infection or M, Bell JL, et al. Dexamethasone in Hospitalized Patients with Covid-19. N COVID-19 severity: a systematic review and meta-analysis. Adv Nutr. (2021). Engl J Med. (2021) 384:693–704. doi: 10.1056/NEJMoa2021436 doi: 10.1093/advances/nmab012. [Epub ahead of print]. 5. Caballero B. Guide to Nutritional Supplements. Cambridge, MA: Academic 21. Annweiler C, Hanotte B, Grandin de l’Eprevier C, Sabatier JM, Lafaie Press (2009). L, Célarier T. Vitamin D and survival in COVID-19 subjects: a quasi- 6. González-Molero I, Morcillo S, Valdés S, Pérez-Valero V, Botas P, Delgado E, experimental study. J Steroid Biochem Mol Biol. (2020) 204:105771. et al. Vitamin D deficiency in Spain: a population-based cohort study. Eur J doi: 10.1016/j.jsbmb.2020.105771 Clin Nutr. (2011) 65:321–8. doi: 10.1038/ejcn.2010.265 22. Gu Y, Wang D, Chen C, Lu W, Liu H, Lv T, et al. PaO2 /FiO2 and IL-6 are 7. Gunville C, Mourani P, Ginde A. The role of vitamin D in prevention risk factors of mortality for intensive care COVID-19 subjects. Sci Rep. (2021) and treatment of infection. Inflamm Allergy-Drug Targets. (2013) 12:239–45. 11:1–8. doi: 10.1038/s41598-021-86676-3 doi: 10.2174/18715281113129990046 23. Dana R, Bannay A, Bourst P, Ziegler C, Losser M-R, Gibot S, et al. Obesity 8. Entrenas Castillo M, Entrenas Costa LM, Vaquero Barrios JM, et al. Effect of and mortality in critically ill COVID-19 patients with respiratory failure. Int J calcifediol treatment and best available therapy versus best available therapy Obes. (2021) 45:2028–37. doi: 10.1038/s41366-021-00872-9 on intensive care unit admission and mortality among subjects hospitalised 24. Grove A, Osokogu O, Al-Khudairy L, Mehrabian A, Zanganeh M, Brown for COVID-19: a pilot randomised clinical study. J Steroid Biochem Mol Biol. A, et al. Association between vitamin D supplementation or serum vitamin (2020) 203:105751. doi: 10.1016/j.jsbmb.2020.105751 D level and susceptibility to SARS-CoV-2 infection or COVID-19 including 9. Annweiler G, Corvaisier M, Gautier J, Dubée V, Legrand E, Sacco G, et al. clinical course, morbidity and mortality outcomes? A systematic review. BMJ Vitamin d supplementation associated to better survival in hospitalised frail Open. (2021) 11:e043737. doi: 10.1136/bmjopen-2020-043737 elderly covid-19 subjects: the geria-covid quasi-experimental study. Nutrients. 25. Dramé M, Cofais C, Hentzien M, Proye E, Coulibaly PS, Demoustier-Tampère (2020) 12:3377. doi: 10.3390/nu12113377 D, et al. Relation between vitamin D and covid-19 in aged people: a systematic 10. Meltzer DO, Best TJ, Zhang H, Vokes T, Arora V, Solway J. review. Nutrients. (2021) 13:1339. doi: 10.3390/nu13041339 Association of vitamin D status and other clinical characteristics 26. Mahase E. Covid-19: death rate is 0.66% and increases with age, study with COVID-19 test results. JAMA Netw Open. (2020) 3:e2019722. estimates. BMJ. (2020) 369:m1327. doi: 10.1136/bmj.m1327 doi: 10.1001/jamanetworkopen.2020.19722 27. Verity R, Okell LC, Dorigatti I, Winskill P, Whittaker C, Imai N, et al. 11. Kaufman HW, Niles JK, Kroll MH, Bi C, Holick MF. SARS-CoV-2 positivity Estimates of the severity of coronavirus disease 2019: a model-based analysis. rates associated with circulating 25-hydroxyvitamin D levels. PLoS ONE. Lancet Infect Dis. (2020) 20:669–77. doi: 10.1016/S1473-3099(20)30243-7 (2020) 15:e0239252. doi: 10.1371/journal.pone.0239252 28. Liu Y, Mao B, Liang S, Yang JW, Lu HW, Chai YH, et al. Association between 12. Cozier YC, Castro-Webb N, Hochberg NS, Rosenberg L, Albert MA. age and clinical characteristics and outcomes of COVID-19. Eur Respir J. Palmer JR. Lower serum 25(OH)D levels associated with higher risk of (2020) 55:2001112. doi: 10.1183/13993003.01112-2020 COVID-19 infection in US Black women. PLoS ONE. (2021) 16:e0255132. 29. Stroehlein JK, Wallqvist J, Iannizzi C, Mikolajewska A, Metzendorf MI, doi: 10.1371/journal.pone.0255132 Benstoem C, et al. Vitamin D supplementation for the treatment of 13. Meltzer DO, Best TJ, Zhang H, Vokes T, Arora VM, Solway J. COVID-19: a living systematic review. Cochrane Database. Syst Rev. (2021) Association of vitamin D levels, race/ethnicity, and clinical characteristics 24:CD015043. doi: 10.1002/14651858.CD015043 with COVID-19 test results. JAMA Netw Open. (2021) 4:1–18. 30. Oscanoa TJ, Amado J, Vidal X, Laird E, Ghashut RA, Romero-Ortuno R, et al. doi: 10.1001/jamanetworkopen.2021.4117 The relationship between the severity and mortality of SARS-COV-2 infection 14. Crafa A, Cannarella R, Condorelli RA, Mongioì LM, Barbagallo F, Aversa and 25-hydroxy vitamin D concentration - a meta-analysis. Adv Respir Med. A, et al. Influence of 25-hydroxy-cholecalciferol levels on SARS-CoV- (2021) 89:175–87. doi: 10.5603/ARM.a2021.0037 2 infection and COVID-19 severity: a systematic review and meta- 31. Murai IH, Fernandes AL, Sales LP, Pinto AJ, Goessler KF, Duran CSC, et al. analysis. EClinicalMedicine. (2021) 37:100967. doi: 10.1016/j.eclinm.2021. Effect of vitamin D3 supplementation vs placebo on hospital length of stay 100967 in patients with severe COVID-19: a multicenter, double-blind, randomized 15. Bassatne A, Basbous M, Chakhtoura M, El Zein O, Rahme M, El- controlled trial. medRxiv [preprint]. (2020). doi: 10.1101/2020.11.16.20232397 Hajj Fuleihan G. The link between COVID-19 and VItamin D (VIVID): 32. Ingham TR, Jones B, Camargo CA, Kirman J, Dowell AC, Crane J, et al. a systematic review and meta-analysis. Metabolism. (2021) 119:154753. Association of vitamin D deficiency with severity of acute respiratory doi: 10.1016/j.metabol.2021.154753 infection: A CASE-control study in New Zealand children. Eur Respir J. (2014) 16. Karahan S, Katkat F. Impact of serum 25(OH) vitamin D level on mortality 44. in patients with COVID-19 in Turkey. J Nutr Health Aging. (2021) 25:189–96. 33. Sabetta JR, DePetrillo P, Cipriani RJ, Smardin J, Burns LA, Landry doi: 10.1007/s12603-020-1479-0 ML. Serum 25-hydroxyvitamin d and the incidence of acute viral Frontiers in Public Health | www.frontiersin.org 8 September 2021 | Volume 9 | Article 758347 Arroyo-Díaz et al. Vitamin D Supplementation and Severity of COVID19 respiratory tract infections in healthy adults. PLoS ONE. (2010) 5:e11088. Cilag, and Boehringer Ingelheim. DM has received advisory and/or speaking fees doi: 10.1371/journal.pone.0011088 from Astra-Zeneca, Ascensia, Boehringer Ingelheim, GSK, Lilly, MSD, Novartis, 34. Petrelli F, Luciani A, Perego G, Dognini G, Colombelli PL, Ghidini A, et al. Novo Nordisk, and Sanofi; they received research grants to the institution from Therapeutic and prognostic role of vitamin D for COVID-19 infection: a Astra-Zeneca, GSK, Lilly, MSD, Novartis, Novo Nordisk, Sanofi, and Boehringer. systematic review and meta-analysis of 43 observational studies. J Steroid Biochem Mol Biol. (2021) 211:105883. doi: 10.1016/j.jsbmb.2021.105883 The remaining authors declare that the research was conducted in the absence of 35. Meehan M, Penckofer S. The role of vitamin D in the aging adult. J Aging any commercial or financial relationships that could be construed as a potential Gerontol. (2014) 2:60–71. doi: 10.12974/2309-6128.2014.02.02.1 conflict of interest. 36. Spiro A, Buttriss JL. Vitamin D: an overview of vitamin D status and intake in Europe. Nutr Bull. (2014) 39:322–50. doi: 10.1111/nbu.12108 Publisher’s Note: All claims expressed in this article are solely those of the authors 37. Casado E, Quesada JM, Naves M, Peris P, Jódar E, Giner M, et al. and do not necessarily represent those of their affiliated organizations, or those of Recomendaciones de la SEIOMM en la prevención y tratamiento del the publisher, the editors and the reviewers. Any product that may be evaluated in déficit de vitamina D. Rev Osteoporos Metab Min. (2021) 13:84–97. this article, or claim that may be made by its manufacturer, is not guaranteed or doi: 10.4321/S1889-836X2021000200007 endorsed by the publisher. Conflict of Interest: RC has received advisory and/or speaking fees from Abbott, Ascensia, Lilly, MSD, Novo Nordisk and Sanofi. JF-N has received advisory Copyright © 2021 Arroyo-Díaz, Julve, Vlacho, Corcoy, Ponte, Román, Navas- and or speaking fees from Astra-Zeneca, Ascensia, Boehringer Ingelheim, GSK, Méndez, Llauradó, Franch-Nadal, Domingo and Mauricio. This is an open-access Lilly, MSD, Novartis, Novo Nordisk, and Sanofi; they received research grants article distributed under the terms of the Creative Commons Attribution License (CC to the institution from Astra-Zeneca, GSK, Lilly, MSD, Novartis, Novo Nordisk, BY). The use, distribution or reproduction in other forums is permitted, provided Sanofi, and Boehringer. PD has received lecture and Advisory Board fees from the original author(s) and the copyright owner(s) are credited and that the original Gilead Sciences, Roche, Merck, Sharp and Dohme, ViiV Healthcare, Janssen and publication in this journal is cited, in accordance with accepted academic practice. Cilag, Theratechnologies, Boehringer Ingelheim, and Ferrer International. PD has No use, distribution or reproduction is permitted which does not comply with these received research grants from Gilead Sciences, ViiV Healthcare, GSK, Janssen and terms. Frontiers in Public Health | www.frontiersin.org 9 September 2021 | Volume 9 | Article 758347