The effect of standard-dose wintertime vitamin D supplementation on influenza infection in immunized nursing home elderly residents

Aim To investigate whether three-month oral vitamin D supplementation (800 IU in drops) reduces the risk of influenza infection in elderly nursing home residents vaccinated against influenza. Methods This cross-sectional observational study enrolled 97 participants (73.2% women) who received one dose of seasonal trivalent 2016-2017 influenza vaccine. The patients were randomized into an experimental group, which received vitamin D supplementation for three months starting on the day of vaccination, and a control group, which did not receive vitamin D supplementation. The primary outcome was the number of influenza infections laboratory-confirmed using a rapid point-of-care test based on nasal swabs collected during vitamin D supplementation. The secondary outcome was serum 25-hydroxyvitamin D level at the end of the study. Results The mean age ±standard deviation was 78.5± 8.8 years. All participants had vitamin D deficiency at baseline. Twenty-three participants who developed signs of respiratory infections during the study were tested for influenza virus. Although the number of influenza-positive participants was lower in the group receiving vitamin D supplementation as compared with the control group (5 vs 12), this difference was not significant. Vitamin D supplementation failed to increase 25(OH)D levels after three months of supplementation. Conclusion Elderly nursing home residents in Zagreb County have a significant vitamin D deficiency. The recommended national supplementation of 800 IU daily failed to lead to vitamin D sufficiency and did not reduce the risk of influenza infection among the vaccinated elderly.

Vitamin D is a fat-soluble vitamin with hormonal qualities, playing a well-documented role in bone health and homeostasis of calcium and phosphate. In addition, vitamin D deficiency is associated with various chronic agingrelated noncommunicable diseases in the elderly, such as dementia, cardiovascular morbidity, diabetes, cancer, autoimmune diseases, and increased mortality (1,2). Recently, vitamin D deficiency has been linked to the susceptibility to and course of coronavirus disease 2019 (3).
Humans derive the most of vitamin D from endogenous synthesis in the skin under the influence of ultraviolet radiation, whereas less than 10% is derived from food. Regardless of its source, vitamin D is biologically active only after hydroxylation in the liver and kidney. Total vitamin D supply is assessed by the serum concentration of 25-hydroxyvitamin D (25(OH)D), with a half-life in plasma of about two weeks (4). A concentration below 75 nmol/L is considered vitamin D deficiency (5). The risk factors for vitamin D deficiency are older age, male sex, living at a high altitude, darker skin, lack of sun exposure, diet, and lack of regular vitamin D supplementation (6). Since 25(OH)D concentration depends primarily on the synthesis in the skin, in countries such as Croatia, this concentration is expected to be lowest during winter, which was confirmed in postmenopausal women (7).
In developed countries, the elderly are defined as people aged ≥65 years (8). This population is at risk of vitamin D deficiency due to a reduced capacity of vitamin D formation in the skin after sun exposure, reduced 25(OH)D hydroxylation in the kidney, and a decreased number of vitamin D receptors (9). Nursing home residents and frail persons are at a particularly high risk of vitamin D3 deficiency (10,11). Many countries, including Croatia, have recognized this problem and most of them recommend taking 700-800 IU of vitamin D per day in order to improve bone health (12)(13)(14).
People of all ages are susceptible to influenza infection, but complications such as secondary infections, exacerbations of chronic comorbid conditions, and death are more prevalent in the frail elderly, particularly nursing home residents (15). The 2010 Cochrane review on influenza vaccination in the elderly by Jefferson et al confirmed vaccine safety, but failed to demonstrate its effectiveness and challenged the ongoing efforts to vaccinate the elderly (16). However, a more recent meta-analysis of the same clinical studies showed that influenza vaccine reduced the risk of infection, as well as of influenza-related dis-ease and mortality in the elderly (17). Although the impact of influenza vaccine was evaluated with different outcome measures, certain outcomes, such as laboratory-confirmed influenza, are considered more specific in comparison with seroconversion to the circulating virus strain (18).
The immunogenicity of influenza vaccine in the elderly is considerably lower (30-40%) than in their younger counterparts (70-90%) (19). Nursing home residents are especially vulnerable given their immunosenescence and multimorbidity (20). A chronic antigenic stimulus important for immunosenescence is cytomegalovirus (CMV) infection, which usually leads to asymptomatic infection in healthy people. Namely, the majority of the elderly have antibodies against this virus, reflecting latent infection. Latent CMV infection is proposed to further enhance aging-induced changes, primarily of T-cell immunity, which increases the risk of respiratory viral infections and decreases the responsiveness to influenza vaccination in the elderly (21).
Based on in vitro data on modulatory actions of 25(OH)D on immune cells and several epidemiologic studies, an adequate 25(OH)D level could protect against influenza infections (22). Other micronutrients, such as zinc, selenium, and vitamin C have also recently been shown to protect against acute respiratory infections (23). However, a recent systematic review showed no significant impact of vitamin D deficiency on immunogenic response to influenza vaccination, although possible strain-specific differences were identified (24). Data on the effect of vitamin D supplementation on the incidence of influenza infection in the influenza-vaccinated elderly are lacking. Therefore, the aim of this study was to investigate whether three-month oral vitamin D supplementation (800 IU in drops) reduces the risk of influenza infection in elderly nursing home influenzavaccinated residents. org/). The experimental group started vitamin D supplementation (800 IU in drops) in the morning on the day of immunization under supervision of a nurse and continued it for three months. The control group did not receive vitamin D supplementation. Vitamin D supplementation was administered according to the respective national recommendations (14).

PAtients And MetHods
Throughout the three winter months, the participants were monitored by nursing home employees (certified nurses and once a day by a physician) for the signs of febrile acute respiratory viral illness. In participants presenting with febrile respiratory illness, samples of the posterior nasopharynx were obtained via nylon flocked swab in addition to blood samples for whole blood analysis, C-reactive protein, serum immunoglobulin (IgM, IgG, and IgA) levels, CMV-specific IgM and IgG levels (ETI-Cytok M/ETI-Cytok G; DiaSorin, Saluggia, Italy), and biochemical urine analysis. Based on these findings, along with clinical examination, breathing frequency, and oxygen saturation (determined by percutaneous transcutaneous oximetry) findings, a diagnosis of upper or lower respiratory infection was established by an attending physician. Nasopharyngeal swabs were obtained in symptomatic participants within the first five days of acute respiratory illness, placed in transport media, and stored in a refrigerator until transfer to the microbiology laboratory within 24 hours.

ResuLts
Out of 97 participants enrolled, 91 completed the study. Four individuals did not undergo the second blood sampling since they died from their chronic illness or influenza infection complications. Two participants left the nursing home to live with their families (Table 1).
At study entry, all participants had lower 25(OH)D levels than those currently recommended for optimal health. In addition, the prevalence of very severe and severe vitamin D deficiency was very high (94%) (Figure 1).
Following randomization, the experimental group (n=41) received daily vitamin D supplementation (800 IU in drops per os) in the morning for three months. The control group (n=57) received no vitamin D supplementation. Vitamin D supplementation increased the 25(OH)D levels in 26 (63.41%) experimental group participants (Figure 2). Vitamin D concentration significantly decreased in the group without vitamin D supplementation (from 11.    Twenty-three participants who developed signs of respiratory infections during the study were tested for influenza virus using a point-of-care test for respiratory viral infections. Seventeen participants tested positive for influenza A virus. Participants with influenza infection had a higher incidence of death (p<0.001) and heart failure (p=0.014). Although the number of influenza-positive participants was lower in the group receiving vitamin D supplementation than in the control group (5 vs 12), the difference was not significant (Table 3).
A total of 79 (81.4%) participants tested positive for IgG against CMV at baseline, while none had CMV-specific IgM antibodies. Furthermore, none had a decrease in major serum immunoglobulins (IgM, IgG, and IgA) at the time of randomization. Sociodemographic variables, changes in vitamin D concentrations, and clinical evidence for respiratory infection did not significantly correlate with the degree of immunosenescence as assessed by serum IgG against CMV (results not shown).

disCussion
Our study showed that residents of two nursing homes had a very high prevalence (94%) of very severe and severe forms of vitamin D deficiency irrespective of age, sex, number of medications, and Clinical Frailty Scale score. This prevalence rate significantly exceeded the prevalence of vitamin D deficiency (<30 nmol/L) observed in postmen- opausal women (14.2%) reported in the only previously published study on vitamin D status in our country (7). This discrepancy was expected as our cohort included older participants characterized by significant frailty, reduced mobility, and consequential reduction of sun exposure, as a well-documented risk for vitamin D deficiency. However, the rate of severe deficiency in our study was similar to those reported in some other European countries (26,27).
Although we found an increased vitamin D concentration among the participants having received the recommended vitamin D supplementation, these values were critically below the recommended values in our country, thus necessitating a different supplementation approach in this vulnerable population. More recent Croatian recommendations, published after this study had been approved by the Ethics Committee, recommend higher doses of vitamin D (1500-2000 IU/day) for patients at risk of vitamin D deficiency and therefore supporting the Endocrine Practice Guidelines Committee guidelines instead of those issued by the American Institute of Medicine (800 IU/day) (28). Indeed, 2000 IU/daily of vitamin D achieved 25(OH)D3 values of >80 nmol/L in most nursing home residents (29). However, these authors did not determine the baseline level of 25(OH)D, so it is not clear if this approach could be used in all nursing home residents regardless of the baseline 25(OH)D levels.
An alternative approach using the loading dose of 200,000 IU and maintenance dose (100,000 IU) administered every 13 weeks also failed to reach and maintain the vitamin D concentration of 75-220 nmol/L in nursing home residents (30). Another study suggested that an individualized loading dose of vitamin D be given to those with severe 25(OH) D deficiency (31). At the time of the study, only one vitamin D preparation exclusively in the form of drops available for oral treatment was covered in total by the national health insurance, so we were not able to prescribe alternative forms of vitamin D. It should also be stressed that no consensus has been reached on the optimal 25(OH)D level and vitamin D deficiency. Whatever the threshold selected, there is a consensus on vitamin D deficiency being associated with poor health in the elderly, particularly in nursing home residents (32).
A recent systematic review of 26 studies showed a highly variable incidence (from 1.21% to 85.2%) of acute respiratory infections among older adults in nursing homes. Infections caused by influenza and respiratory syncytial viruses are most common, but other viruses (eg, parainfluenza, human metapneumovirus, and rhinoviruses) are also associated with infection-dependent morbidity and mortality (20). Many vulnerable people with underlying chronic diseases in nursing homes share common spaces for daily activities, and guidelines for prevention of respiratory infections are difficult to follow, which may lead to spread of infections. Furthermore, daily visitors and the nursing home staff can introduce viral infections from the community (33). Given the limited treatment options and consequences of the most common viral infections in the elderly, early detection of infections and vaccination before the influenza season, which in Croatia is free for the elderly, are key measures to prevent morbidity and mortality, reduce the use of unnecessary antibiotics, and rationally implement isolation measures (34,35). The number of influenza-positive participants was lower in the group receiving vitamin D supplementation than in the control group; however, this difference was not significant. Furthermore, two participants who had influenza and died due to complications had received vitamin D supplementation. These results indicate that in our study vitamin D supplementation demonstrated no anti-infective or anti-inflammatory effect, as recently discussed (37). More recent national recommendations for vitamin D supplementation in the groups at risk, which include the frail elderly nursing home residents (28), as well as the findings of a recent meta-analysis, indicate that individuals with profound vitamin D deficiency and those not receiving bolus doses of vitamin D benefit most from vitamin D supplementation (38). Therefore, it is tempting to speculate that this approach would be more effective in preventing influenza infection in vaccinated elderly nursing home residents.
The physiological processes of immunosenescence, as well as comorbidities, contribute to the fragility of the elderly and to the increased risk and adverse outcome of viral res-piratory infections (20). Humoral immunity, which plays an important role in preventing influenza virus transmission and infection, declines with age (39). In our study sample, none of the participants had decreased levels of serum immunoglobulins at randomization, so we excluded the possible contribution of an age-associated deficit of humoral immunity as a risk factor for respiratory viral infection (data not shown). However, a recent study demonstrated that vitamin D supplementation induced a higher plasma level of transforming growth factor beta as an indicator of tolerogenic immune response following influenza vaccination without improving antibody production and thereby influenced predominantly cell-mediated immunity (40).
Cytomegalovirus is considered as an important contributor to chronic antigenic stimulation important for immunosenescence (41). Influenza vaccine inefficiency also has been linked to the reactivation of CMV infection in the elderly (42). We observed no link between CMV seropositivity and the lack of influenza-vaccine protection either in vitamin D supplemented participants or in the control group.
We showed for the first time that the great majority of elderly nursing home residents in Zagreb County had a severe vitamin D deficiency and that a three-month supplementation with 800 IU of cholecalciferol daily according to the current national guidelines failed to increase 25(OH)D levels to the recommended range. Our study demonstrated that 800 IU/day of oral vitamin D supplementation during winter did not prevent influenza infections in the vaccinated elderly. declaration of authorship AGH, MM, and AG conceived and designed the study; AGH acquired the data; ALG, JV, and MM analyzed and interpreted the data; AGH drafted the manuscript; ALG, JV, MM, and AG critically revised the manuscript for important intellectual content; all authors gave approval of the version to be submitted; all authors agree to be accountable for all aspects of the work.

Competing interests All authors have completed the Unified Competing
Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years; no other relationships or activities that could appear to have influenced the submitted work.