The US is breaking new records in the number of daily deaths from COVID-19. The dizzying speed with which several vaccines have been developed and implemented is nothing short of impressive. However, we still have to face the dire prediction that our national death toll will exceed 500,000 Americans before widespread vaccines can lift us out of this crisis. The response to the pandemic, therefore, must include an effort to aggressively eliminate what is becoming apparent as a risk factor for morbidity and mortality in COVID-19 – vitamin D deficiency.
For any COVID-19 risk factor, such as obesity, hypertension or diabetes, strong correlational data is sufficient to inform clinical care, as in the 1964 Surgeon General Luther Terry’s article Smoking and Health Report. This groundbreaking publication, which saved tens of millions of lung cancer lives, was based on a causality analysis by an advisory committee. The team analyzed the existing data and drew on the work of Sir Austin Bradford Hill and Sir Richard Doll, who examined the rise in lung cancer cases in the UK. Hill later outlined the patterns that resulted from his investigation, now known as Hill’s criteria for causality. He assumed that correlational data can be used to infer causality, satisfying several criteria, such as consistency, specificity, temporality and dose responsiveness. Vitamin D deficiency, associated with deleterious effects on innate and adaptive immunity, has many small but growing data sets that meet all of Hill’s criteria as a risk factor for severe COVID-19. And unlike other risk factors, it can be modified acutely.
Jain and colleagues studied 154 patients who came to a medical center for 6 weeks. When deaths were assessed based on vitamin D deficiency (serum 25-OH-D <20 ng / mL), the lethality rate was 21%, compared with only 3% for those with higher levels. Most striking was that vitamin D deficiency was found in 97% of critically ill patients who required ICU admission, but in only 33% of asymptomatic cases, suggesting that low levels are a necessary component of severe COVID-19. This is one of the numerous studies this year establishing the correlation of low vitamin D levels with an aggravated course of COVID-19, as a meta-analysis by Pereira and colleagues reveals.
Still, to establish Hill’s criteria, some experimental evidence is not only recommended, but necessary, and small randomized trials with aggressive vitamin D replacement have shown positive results. Rastogi and colleagues treated 40 subjects with mild COVID-19 and vitamin D deficiency (25-OH-D <20 ng / mL) with placebo or 420,000 IU cholecalciferol (vitamin D3) in a seven-day fast-acting nanoemulsion. that is, 60,000 IU (1,500 μg) per day. The results showed that supplementation helped to clear the virus more quickly - 63% of treated patients tested negative for SARS-CoV-2 on the 14th day, compared with only 21% of the placebo group. In addition, the treated group showed a decrease in fibrinogen levels, which appears to contribute to the increased risk of thrombotic events in COVID-19.
Castillo’s team in Cordoba, Spain, randomized 76 patients hospitalized with COVID-19 in a 2: 1 ratio to receive open calcifediol or no supplementation, in addition to standard treatment. The intervention group received 1,064 μg of this fast-acting vitamin D analogue in the first week, three times more potent than vitamin D3, followed by 266 μg weekly thereafter. Of the treated patients, only 2% (1 in 50) required admission to the ICU compared to 50% (13 in 26) of the untreated group. In addition, 8% of untreated patients died, compared to none in the intervention group. Although vitamin D deficiency was not identified on admission, the researchers cite a report that the levels of 25-OH-D in Córdoba in winter are deficient, on average 16 ng / mL. The use of a study population that leans towards vitamin D deficiency makes this a good study to examine the benefits of aggressively correcting that deficiency in COVID-19. As far as we know, only one intensive care program in the USA has adopted such an aggressive replacement protocol in the treatment of COVID-19.
There is more evidence pointing in that direction. Using a quasi-experimental approach, Annweiler and colleagues examined frail elderly patients hospitalized for COVID-19 in France. The researchers obtained records for those who regularly received bolus vitamin D3 supplements – 20,000 to 50,000 IU per month, a common practice in French nursing homes – and those who did not. Only 10% of those who received regular supplementation progressed to severe COVID-19 compared to 31% of the non-supplemented group. In addition, mortality rates at 14 days were only 7% in the supplemented group compared to the same 31% in the non-supplemented group. The researchers also identified a third group of patients – those who received a single dose of 80,000 IU of cholecalciferol at the time of the COVID-19 diagnosis. This group did better than the group that received none, but the result did not reach statistical significance, suggesting that the dose may have been too low or arrived too late.
We found a study awaiting peer review that did not show the benefits of treating vitamin D deficiency in COVID-19. The researchers administered a single dose (200,000 IU of vitamin D3) to patients ten days after the first symptoms of COVID-19 appeared. Unlike calcifediol, it can take a week or more for the body to convert vitamin D3 into its active form. In addition, as it is fat-soluble, the organism competes with the adipose tissue to obtain the necessary amount, requiring higher doses in obesity (the average BMI in this study was 31.6). Compare the dose given here with the standard protocol to correct vitamin D deficiency in healthy outpatients, who routinely receive a total of 600,000 IU divided into twelve weeks, at 50,000 IU per week.
Data like this is not new. A 2014 Austrian study of 475 patients showed that supplementation with 540,000 IU of vitamin D3 followed by 90,000 IU per month cut the hospital mortality rate in half for ICU patients with severe vitamin D deficiency (25-OH-D level) <12 ng / ml). Higher-level patients showed no benefits, bringing to light a possible deficiency of many vitamin D tests - should they focus on the results only for those who are deficient?
It is not yet a common practice to check serum 25-OH-D levels in patients hospitalized with COVID-19, although many doctors are prescribing supplementation in typical (and possibly insufficient) doses. A Canadian study of 22,214 supplemented subjects found that 1,000 IU of cholecalciferol daily increased 25-OH-D levels by an average of only 4.8 ng / mL with decreasing returns for each additional increment of 1,000 IU per day. Toxicity was not seen in people who reported taking doses as high as 20,000 IU per day, an amount roughly equivalent to that generated by an afternoon of summer sun on the skin. (Several medical societies claim that doses of only up to 4,000 IU of vitamin D per day are safe without medical supervision and that up to 10,000 IU per day has shown no adverse effects seen.)
It is our responsibility, as doctors, not to wait for perfect evidence when making life and death decisions. Given the safety profile of vitamin D, the prevalence of vitamin D deficiency of 40% in the United States and the fact that this season is likely to be the deadliest phase of the pandemic to date, we need to act now. Identifying and eradicating vitamin D deficiency with early, aggressive supplementation with COVID-19 has the potential to save thousands of lives and should be one of our top public health priorities.
Richard H. Carmona, MD, MPH, was the 17th United States Surgeon General and is now a distinguished professor of public health and incident commander at COVID-19 at the University of Arizona. Vatsal G. Thakkar, MD, is an integrating psychiatrist, founder of Reimbursify, and can be followed at Twitter. John C. Umhau, MD, MPH, is a retired commander of the United States Public Health Service and has published more than forty peer-reviewed research articles.