Vitamin D Flops in Moderate-Severe COVID


A close up of a blister pack of vitamin D capsules

Among patients with moderate to severe COVID-19, there was no difference in length of hospital stay in those given a single 200,000-IU dose of vitamin D3 versus those given placebo in a randomized trial in Brazil.

Median length of hospital stay was about 7 days for both groups, with no significant differences in secondary outcomes including in-hospital mortality, admission to ICU, or need for mechanical ventilation, reported Rosa Pereira, MD, PhD, of Universidade de Sao Paulo, and colleagues in an online edition of JAMA.

An accompanying editorial by David Leaf, MD, of Brigham and Women’s Hospital in Boston, and Adit Ginde, MD, MPH, of University of Colorado School of Medicine in Aurora, noted that prior research showed that lower levels of vitamin D were independently associated with worse outcomes in patients with acute illness, but also that vitamin D supplementation as a treatment had no impact on clinical outcomes.

Leaf and Ginde said that COVID-19 has generated renewed interest in vitamin D, adding that this study was “the largest published randomized, double-blind, placebo-controlled trial of vitamin D3 administration among hospitalized patients with COVID-19 to date.”

While the findings taken in isolation “may appear ambiguous” and show no evidence of benefit for moderately to severely ill patients, they said, perhaps the door should not be closed altogether.

“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),” they wrote, adding that the study did not address outpatients with mild COVID-19 or the use of vitamin D as prophylaxis against the disease.

Pereira and colleagues examined data from hospitalized patients with moderate to severe COVID-19 at two sites in Sao Paulo from June to August 2020, with the final follow-up in October. Participants were eligible if they were adults who tested positive for SARS-CoV-2 via PCR or had a CT scan finding “compatible with the disease,” as well as “diagnosis of flu syndrome with criteria for hospitalization.” Primary outcome was length of hospital stay, defined as date of randomization to hospital discharge.

Overall, 240 patients were randomized: 120 to a single oral dose of 200,000 IU of vitamin D and 120 to placebo. Of these, 237 were included in the primary analysis. Participants’ mean age was 56, about 44% were women, and mean baseline 25-hydroxyvitamin D level was 20.9. They were given the intervention after a mean of 10.3 days from symptom onset.

The adjusted hazard ratio for hospital discharge was 0.99 (95% CI 0.71-1.37, P=0.94). There were also non-significant differences between the intervention and placebo groups in secondary outcomes, including in-hospital mortality (7.6% vs 5.1%, respectively, P=0.43), admission to the ICU (16.0% vs 21.2%, P=0.30), and need for mechanical ventilation (7.6% vs 14.4%, P=0.09).

In the intervention group, 86.7% achieved 25-hydroxyvitamin D sufficiency versus 10.9% of the placebo group, meaning “the present null findings cannot be attributed to the failure of increasing serum 25-hydroxyvitamin D levels.”

Researchers noted no adverse events, except a single episode of vomiting linked to the intervention. They also noted no significant differences between groups in any health-related laboratory markers following the intervention.

Limitations to the data included that a minimal clinically important difference in hospital length of stay among COVID-19 patients has yet to be determined, as well as the small and heterogeneous sample size, with different patients with co-existing diseases on different medication regimens.

Leaf and Ginde noted that 30 studies on vitamin D interventions in COVID-19 are currently listed on, globally and across the disease spectrum. However, they suggested that many of these studies may be “underpowered or will not achieve target enrollment,” based on experience during the pandemic.

Leave a Reply

Your email address will not be published. Required fields are marked *