In the present study, one bolus of 100,000 IU vitamin D3 followed by a weekly dose of 10,000 IU vitamin D3 rapidly and significantly increased serum 25(OH)D by an average of 48 nmol/L to an adjusted mean of almost 100 nmol/L, which is, in the range where protective immunomodulatory effects have previously been observed4,5. This intervention permitted 78% of participants in the intervention to achieve vitamin D sufficiency (> 75nmol/L) at endpoint, compared to fewer than 15% of controls, within 27 to 70 days.
Several studies have evaluated the dosage of vitamin D supplements required to safely attain and maintain the optimal serum 25(OD)D level in general population13,16,22. Supplementation of 1000 to 4000 IU/day for 6 to 18 weeks can raise 25(OH)D levels to above 75 nmol/L6,7, depending on baseline levels. Conversely, in a systematic review of oral vitamin D supplementation, a single dose of 100,000 IU induced a rapid peak serum 25(OH)D levels with sustained levels above 75 nmol/L only for 5 to 6 weeks when it was < 50 nmol/L; it only persisted until 3 months if the baseline was > 50 nmol/L8. In two pediatric trials, 100,000 IU bolus combined with daily supplement at age-specific recommended doses of 400 IU achieved a rapid (10 days) and sustained (3 months) serum 25(OH)D above 75nmol/L in all children; but neither boluses alone nor daily supplementation alone were sufficient to lead to a rapid and sustained level about 75 nmol/L13,16. Our findings are in line with the observations and confirm the values of combining a bolus with regular dosing to achieve rapid and sustained vitamin D sufficiency.
Due to the trial's premature termination, we were unable to explore the effects of vitamin D supplementation on COVID-19 infection risk. While numerous studies explored the potential impact of vitamin D in patients hospitalised with COVID-19, studies pointing to the primary prevention remain limited to few observational studies23–25, which provided conflicting results. There are few trials registered at Clinicaltrials.gov addressing the effects of vitamin D supplementation on risk of COVID-19 infection and completion of these ongoing trials is needed to expand our understanding of the effects of vitamin D supplementation on preventing COVID-19.
As one of the potential mechanisms of action of vitamin D in COVID-19 is via its potential anti-inflammatory properties, we explored the impact of supplementation on serum CRP. Indeed, vitamin D supplementation could suppress the nuclear factor kappa B (NF-κB) pathway, which in turns may reduce systemic inflammation and production of CRP26. While our observed group difference of 1.50mg/L in change from baseline in serum CRP did not reach statistically significance, fewer subjects had elevated CRP (> 5mg/L) in the intervention, than in the control, group at endpoint, after adjusting for baseline levels and baseline group imbalances. A previous meta-analysis of 10 trials involving a total of 924 participants indicated that daily vitamin D supplementation (ranging from 400 to 7,143 IU for 8 to 48 weeks) significantly decreased the circulating CRP level by 1.08 mg/L (95% CI, − 2.13, − 0.03), with evidence of significant heterogeneity26. Additional well-designed RCTs are warranted to further investigate the role of vitamin D supplementation in systemic inflammation.
Adherence to intervention was high in both groups probably in part due to weekly rather than daily administration regimen, coupled with electronic reminder, as both techniques have previously been shown to improve adherence27–30. Furthermore, with clear written instruction, step-by-step video guide, as well as virtual supervision during sample collection, our participants demonstrated that they could collect their samples and shipped it with appropriate packaging, with all samples received in acceptable condition. Although slightly more than half of participants completed the first few fortnightly health questionnaires within the requested timeframe, rapid follow-up of delayed completion identified errors that were resolved quickly, which resulted in over 90% timely completion of study questionnaires throughout the trial. Thus, we would strongly recommend close monitoring of protocol adherence to rapidly identify and correct any issues. Moreover, we cannot rule-out the possibility that the adherence could have reduced with a longer study duration. The literature reports mixed results regarding the implementation of virtual trials. Similar to our findings, some studies showed overall success in adherence to intervention and study procedures31–33, while others reported higher dropout rates34,35. Conducting a hybrid trial during COVID-19 pandemic has allowed us to conduct research activities, even during lockdown and curfew periods. With close monitoring and supervision, it could be a promising alternative for in-person visits even after the pandemic.
No adverse health events attributable to vitamin D toxicity were reported throughout the study, which is in line with previous research that reported daily doses of vitamin D up to 10,000 IU were safe in adult participants4,36,37.
The strengths of this trial include the randomised controlled design and high retention rate (97%). We acknowledge several limitations, including the inconsistent study duration among the participants, varying between 27 to 70 days, due to premature study termination. Nonetheless, the duration was adequate to detect important between groups differences in serum 25(OH)D levels. Although data regarding dietary vitamin D intake or sunlight exposure was not collected throughout the study; the serum 25OHD level remained stable in the control study, suggesting no meaningful impact of these factors on study findings.
Our findings can be generalized to both Canadian adults and healthcare workers. Indeed, our observations are consistent with a previous study reporting vitamin D insufficiency in winter and early spring in over 70% of Canadians, with only one third supplementing their diet, usually with < 400IU vitamin D/ day38. Our findings are also concordant with a previous systematic literature review showing that above 92% of nurses and practising physicians had serum vitamin D levels < 75nmol/L39.