Less than a week after JAMA Network Open published a small study showing zinc and vitamin C were not associated with improved mild COVID-19 infections, a 240-person JAMA study also found that a single dose of vitamin D did not have any significant effect on moderate to severe COVID-19 infections.
The study, published yesterday by Igor Murai, PhD, a Sao Paul rheumatologist, and colleagues, reported that hospital stay was a median 7.0 days for both those in the intervention and the placebo group, and while there were differences up to 8.4 percentage points across in-hospital mortality, intensive care unit (ICU) admission, and mechanical ventilation needs, they were all statistically not significant.
"When this clinical trial is taken [in] isolation, the findings may appear ambiguous; that is, the findings do not exclude clinically important benefit (or harm) from high-dose vitamin D3 administration in hospitalized patients with moderate to severe COVID-19," US clinicians David Leaf, MD, MMSc, and Adit Ginde, MD, MPH, write in a related JAMA commentary.
"However, taken together with existing randomized clinical trials of vitamin D administration in hospitalized patients with respiratory infection and critical illness, the results reported by Murai et al do not support routine administration of vitamin D in hospitalized patients with moderate to severe COVID-19."
Supplement safe but ineffectual
The researchers recruited adult patients in Sao Paulo who were hospitalized with COVID-19 but did not require mechanical ventilation or ICU admittance at the time of enrollment. From Jun 2 to Aug 27, 2020, half of the 240 patients were randomly assigned a single dose of 200,000 international units (IU) of 25-hydroxyvitamin D (25[OH]D) in addition to the standard of care, and the other half were assigned a placebo. Three people were excluded from the final results because they withdrew consent.
Patients were an average of 56.2 years old, and they were randomized a median of 10.3 days after symptom onset and a median of 1.4 days after hospitalization. The most common comorbidities were obesity (57.8% of the intervention group vs 53.7% of the placebo group), high blood pressure (56.3% vs 49.2%), and diabetes (41.2% vs 29.7%).
None of the primary or secondary outcomes achieved statistical significance. In-hospital mortality was 7.6% in the treatment group and 5.1% in the placebo group (95% confidence interval [CI], -4.1% to 9.2%; P = 0.43), ICU admission after enrollment was 16.0% vs 21.2% (95% CI, -15.1% to 4.7%; P = 0.30), and 7.6% vs 14.4% needed mechanical ventilation (95% CI, -15.1% to 1.2%; P = 0.09).
For patients requiring mechanical ventilation, there was no significant difference between duration (15.0 average days vs 12.8; 95% CI, -8.4 to 12.8; P = 0.69).
Before vitamin D doses were administered, patients' average vitamin D level was 21.2 nanograms per milliliter (ng/mL) in the intervention group and 20.6 ng/mL in the placebo group—20 ng/mL is considered healthy. And only about a quarter of all participants had a severe deficiency (less than 12 ng/mL). While 25(OH)D levels significantly increased to an average of 44.4 ng/mL in the vitamin D group, Leaf and Ginde point out that this does not measure how much the body was actually able to convert.
The only adverse effect reported was one instance of vomiting.
Largest RCT to date, but still underpowered
In the JAMA commentary, Leaf and Ginde say that, although the results seem to show vitamin D's insignificance for COVID-19 treatment and is the largest randomized, double-blind, placebo-controlled trial assessing the vitamin published so far, the study had noteworthy design limits, including being underpowered.
"With 208 participants they would have 80% power to detect a 50% difference in hospital length of stay, which is a highly improbable result," they write.
Additionally, they write, not only does the study's focus on moderate to severe patients limit its generalization, but other COVID anti-inflammatory studies have had results highly dependent on the severity of the illness. The lack of vitamin D deficiency could also have had an impact on the results.
While this does not detract from the researchers' final conclusions, they write, "Given the lack of highly effective therapies against COVID-19, except perhaps for corticosteroids, it is important to remain open-minded to emerging results from rigorously conducted studies of vitamin D (despite smaller sample sizes and important limitations of some studies)."