Today I was made aware of a preprint* that was most recently revised yesterday with specific data on vitamin D status. This is the first observational study that has, to my knowledge, addressed how vitamin D levels correlate with COVID-19 severity.
A researcher from the Philippines used the data from three South Asian hospitals to examine the association between 25(OH)D, the principal marker of vitamin D status, and severity of COVID-19 among 212 confirmed cases.
The categories of COVID-19 severity were defined as follows:
- Mild: no pneumonia
- Ordinary: pneumonia, fever, and respiratory symptoms
- Severe: hypoxia and respiratory distress
- Critical: respiratory failure requiring intensive care
In ng/mL (multiply by 2.5 to get nmol/L), the average 25(OH)D among the cases was 23.8.
Among mild cases, it was, on average, 31.2; among ordinary cases, it was 27.4; among severe cases, it was 21.2; among critical cases, it was 17.1.
Vitamin D status was also categorized as follows:
- Normal: >30
- Insufficient: 21-29
- Deficient: <20
Only 26% of the cases had 25(OH)D in the normal range, and virtually all of them (86%) had a mild outcome. A small proportion (7.3%) had an ordinary outcome, and smaller proportions had severe (3.6%) or critical (3.6%) outcomes.
Just over a third (38%) had 25(OH)D in the insufficient range, and only one of them (1.3%) had a mild outcome. Almost half of them (44%) had an ordinary outcome. Just over a quarter (29%) had a severe outcome, and another quarter (26%) had a critical outcome.
Just over a third (36%) had 25(OH)D in the deficient range, and their distribution was pushed further toward severe and critical outcomes. Only one patient (1.4%) had a mild outcome, while 26% had an ordinary outcome, 40% had a severe outcome, and 33% had a critical outcome.
What Is the Optimal Level of 25(OH)D?
This study is small, cross-sectional rather than prospective, and non-randomized, and therefore should not be used to infer cause and effect relationships.
However, if we take the numbers at face value, what 25(OH)D is associated with the best outcome?
The mild cases had a mean 25(OH)D of 31.2, with a standard deviation of 1.08. Assuming a nicely shaped bell curve for the distribution, this suggests that 95% of all mild cases fell between the mean plus or minus two standard deviations, which is between 29 and 33.4.
2.5% of people would fall outside the distribution on the high end, and another 2.5% would fall outside it on the low end. So, two things are true:
- 97.5% of mild cases had 25(OH)D that was 29 ng/mL or higher.
- 97.5% of mild cases had 25(OH)D that was 33.4 ng/mL or lower.
This study provides no information on how vitamin D levels above 34 ng/mL associate with COVID-19 severity. We cannot rule out a U-shaped curve, where higher levels associate with greater severity, at this time.
Does This Represent Cause and Effect?
The relationship between vitamin D and inflammation is a two-way street. The activation of immune cells causes them to use up vitamin D by converting it to its active hormonal form, calcitriol. Inflammation thus decreases 25(OH)D. Nevertheless, the fact that immune cells activate vitamin D for their own purposes means that vitamin D is essential to immune function.
Whenever we see an association between low vitamin D and infectious disease or immune dysfunction, it is possible that the disease itself is depleting the vitamin D. It is also possible that low vitamin D as a result of low dietary intake or low sunshine exposure led to the disease.
If the disease caused the low vitamin D in the first place, this doesn't rule out the possibility that vitamin D supplementation would improve the course of the disease. It may in some cases, and may not in others.
Let's consider some cause-and-effect hypotheses.
What Might Explain These Results?
As I noted previously, the best predictors of severity are low lymphocytes and high interleukin-6 (IL-6). As I noted in the elderberry post, one study identified the strongest predictor as the ratio of neutrophils to CD4 and CD8 T cells, which are subsets of lymphocytes. We don't know for sure that altering these markers will alter the course of the disease, but it's a very reasonable hypothesis that it would.
What do we know about how vitamin D impacts these markers in other contexts?
Vitamin D and Lymphocytes
In the context of HIV, vitamin D deficiency is associated with lower CD4 T cell counts in some but not all studies. However, vitamin D supplementation does not raise the CD4 T cell counts (here, here, or here).
Similarly, in postmenopausal women, vitamin D does not alter lymphocyte counts.
That at least one study in HIV found that low vitamin D levels correlate with low CD4T cells, yet at least three studies found that supplementation had no effect, shows that correlations between vitamin D and immune parameters do not always represent cause-and-effect relationships, and could represent the immune dysfunction itself lowering both vitamin D status and CD4 T cells. This warrants caution for assuming that vitamin D supplementation could help with COVID-19.
On the other hand, the fact that vitamin D supplementation does increase lymphocytes and does decrease the neutrophil-to-lymphocyte ratio in some contexts makes it plausible that vitamin D might do those same things in the context of COVID-19, which might reduce the severity of the disease.
Vitamin D and IL-6
A meta-analysis of four trials in middle-age and older-adults found no effect of vitamin D on IL-6. Another meta-analysis reported four studies that all found no effect of vitamin D on IL-6 in hemodialysis patients. Another pooled the results of eight studies and reported no effect in obese and overweight subjects. In diabetes, three out of five studies found that IL-6 was lower in vitamin D supplementation groups than in controls, but statistical significance was only achieved in one study, and when the results of the five studies were pooled together, they were not statistically significant.
Intramuscular injection of 300,000 IU vitamin D reduced IL-6 in patients with ventilator-associated pneumonia. Since COVID-19 causes pneumonia, this is arguably the context that is most relevant to COVID-19.
The Most Likely Explanation
It seems as though vitamin D has very little impact on IL-6 in studies of chronic illness where the inflammation is typically moderate, while it sometimes has a dramatic effect on IL-6 in severe or acute inflammatory conditions.
Since IL-6 can rise 10- or 20-fold higher than normal levels in COVID-19, it seems quite likely that vitamin D could help keep it closer to normal levels. Since IL-6 could be playing a direct role in hypoxemia and respiratory failure, that makes a strong case for testing whether vitamin D could prevent the disease from progressing in severity.
This seems like a stronger case than vitamin D altering lymphocyte counts, because the effect of vitamin D on IL-6 appears to depend on the severity of the inflammatory disease. By contrast, there a number of negative studies for vitamin D and lymphocyte counts in HIV, where lymphocyte counts really matter, and the contexts where it helps (tuberculosis, adolescent girls but not postmenopausal women) do not seem to have a consistent pattern.
However, these hypotheses are not mutually exclusive and there is no reason to assume vitamin D is limited to one important role.
Should Vitamin D Be Taken Preventatively, or in Response to COVID-19?
In principle, I find it most likely that vitamin D exerts most of its protection after someone becomes ill with COVID-19. In the IL-6 study I reviewed, IL-6 appeared to spike very early in the disease far higher than anyone would have normally, but it was the peak IL-6 reached over the course of the disease that determined whether someone wound up on a ventilator. Since vitamin D seems to reduce IL-6 in acute and severe cases, but not in chronic and moderate cases, I believe maintaining vitamin D levels during the active infection is what is likely to restrain IL-6 and possibly prevent the progression to a severe outcome.
We still do not have any data on how vitamin D impacts the likelihood someone will become infected or symptomatic.
It is entirely possible that a risk factor can protect against infection but promote severe progression, or increase the likelihood of infection while preventing severe progression.
For example, in COVID-19 and the Smoking Paradox, I hypothesized that smoking may decrease infection risk because of the toxic effects that free radicals in the smoke could have on the virus, while promoting disease severity by increasing ACE2 and thus increasing viral load.
Conversely, vitamin D may increase infection risk because it increases ACE2 but does not, unlike smoke, have directly toxic effects on the virus. It may, further, increase viral load, since the ACE2 would be expected to seed exponential growth of the virus. However, if it also restrains IL-6 from reaching extremely high levels, it may ultimately prevent the worst manifestations of the disease.
Nevertheless, given that the 25(OH)D levels associated with a mild outcome (29-34 ng/mL) are consistent with those needed to maximize the absorption of calcium from food, protect bone health, and minimize all-cause mortality (as I pointed out in my response to Rhonda Patrick), it seems wisest and simplest to me to keep 25(OH)D in this range prophylactically.
With that said, I strongly recommend that physicians involved in COVID-19 treatment test and publish the effect of correcting low 25(OH)D on COVID-19 outcomes.
How I am Changing My Position
My position that vitamin D may raise ACE2, which may increase the risk of getting COVID-19, remains the same. I remain concerned that high vitamin D levels may, through this mechanism, make COVID-19 outcomes more severe.
However, I believe this new data justifies putting a tentative floor of 29 ng/mL on the acceptable 25(OH)D level during the COVID-19 crisis.
Rather than using food and sunshine as a general default, I think the best default is to shoot for a 25(OH)D of 30 ng/mL. The average person needs 900 IU/d to achieve this, and 97.5% of people need no more than 1700 IU to achieve this. You can test your level with a home kit, such as those offered by Grassroots Health.
I do not recommend shooting for higher levels and I remain concerned that higher levels could be associated with higher risk through a U-shaped curve.
I plan to update The Food and Supplement Guide for the Coronavirus tomorrow to reflect this. If you purchase it now, you will get the current version now and will then get the free update immediately upon release.
The Bottom Line
Vitamin D levels below 29-34 ng/mL are associated with more severe COVID-19 outcomes. This is based on a small, cross-sectional, observational study, and could easily be revised by data that comes in later.
We do not know if higher levels are also associated with more severe disease through a U-shaped curve, and we currently have no data on how vitamin D impacts the risk of getting COVID-19 in the first place.
Although we cannot be sure that vitamin D is protective, the most likely explanation for this is that maintaining healthy vitamin D levels restrains the massive increase in IL-6 that may occur in severe cases of COVID-19, and thereby stops IL-6 from contributing to the increased blood clotting and low oxygen levels that can render the disease course critical or deadly.
My position is that we should maintain 25(OH)D levels close to 30 ng/mL.
Please Support This Service
Support the service by purchasing a copy of The Food and Supplement Guide for the Coronavirus.
Get the guide for free when pre-ordering my Vitamins and Minerals 101 book.
Or, you can offer support by buying Testing Nutritional Status: The Ultimate Cheat Sheet
Get the guide for free, get 30-50% off the pre-orders of my book, and get the Cheat Sheet for 50% off when you sign up for the CMJ Masterpass. Masterpass membership also includes a storefront where 305 brands of supplements are all 35% off, permanently. These include over 500 different vitamin D supplements. Use the code COVID19 to save 10% off the membership price.
Here's a list of other ways to support my work: How You Can Support My Work
I am not a medical doctor and this is not medical advice. I have a PhD in Nutritional Sciences and my expertise is in conducting and interpreting research related to my field. Please consult your physician before doing anything for prevention or treatment of COVID-19, and please seek the help of a physician immediately if you believe you may have COVID-19.
If you aren't subscribed to the research updates, you can sign up here.
You can access an archive of these updates here.
Comments and Questions
To leave a comment or question, please use the Facebook post for this newsletter.
* The term “preprint” is often used in these updates. Preprints are studies destined for peer-reviewed journals that have yet to be peer-reviewed. Because COVID-19 is such a rapidly evolving disease and peer-review takes so long, most of the information circulating about the disease comes from preprints.