29 Mar, 17 | by EBM
By Dr. Geoffrey Modest
A recent meta-analysis found that vitamin D supplementation, especially in those who were quite deficient, led to lower risk of acute respiratory infections (see doi.org/10.1136/bmj.i6583).
- 25 eligible RCTs were found, with a total of 10,933 patients, aged 0-95. The researchers were able to access individual participant data from the studies. Outcome data was obtained for those participants experiencing at least one acute respiratory tract infection. The trials were from 14 countries on 4 continents. All studies supplied oral vitamin D3 to those in the intervention arm. This was given as bolus doses every month to every 3 months in 7 studies, weekly doses in 3 studies, and daily doses in 12 studies. Study duration range from 7 weeks to 1.5 years.
- Overall demographics: 50-50 male/female, 50% < 1yo/15% 1-16yo/28% 16-65yo/10% >65yo, 5% baseline 25(OH) level of <25 nmol/L, 33% >25 nmol/L, 62% not recorded
- There was a 12% decreased risk of acute respiratory tract infections in those receiving vitamin D supplementation, adjusted OR 0.88 (0.81-0.96), p<0.001
- The protective effect was seen in those receiving either daily or weekly vitamin D, without additional bolus doses (bolus was defined as at least 30,000 IUs), adjusted OR 0.81 (0.72-0.91), with number needed to treat =20, though there was no benefit in those receiving 1 or more bolus doses
- In those on either daily or weekly vitamin D, there was differential benefit depending on the baseline 25-OH vitamin D level, with p=0.006 for the interaction:
- If <25 nmol/L (equivalent to 10 ng/ml), 70% decrease, adjusted OR 0.30 (0.17-0.53)
- If >25 nmol/L, 25% decrease, adjusted OR 0.75 (0.60-0.95)
- There was no significant difference between the groups by daily dose of vitamin D (<20mg, equivalent to <800 IU; 20-50 mg, equivalent to 800-2000IU; >50 mg, equivalent to > 2000 IU), age, BMI, the presence of asthma or COPD, or having received influenza vaccination.
- No serious adverse events in the vitamin D group, with 0.5% having hypercalcemia and 0.2% renal stones, though both events were evenly represented in the intervention and control arms.
- The body of evidence contributing to these results was considered to be of high quality. And a review of the individual participant data from their included studies showed that the preponderance of them showed benefit from vitamin D therapy (i.e., it was not just a couple of large studies dominating the meta-analysis)
- In differentiating acute respiratory illnesses, the significant benefit was in 11 studies for upper respiratory infections, with a 12% benefit limited to those on daily or weekly dosing [OR 0.88 (0.78-1.00), p=0.05], not in any of the other secondary outcomes they assessed (lower respiratory tract infections, use of antimicrobials, absence from school/work, serious adverse events)
- This meta-analysis basically found that in those patients who were quite vitamin D deficient, there was a pretty remarkable benefit to either daily or weekly vitamin D therapy in preventing acute respiratory infections (70% decrease), and specifically upper respiratory infections. There was also a 25% statistically significant decrease in acute respiratory infections in those with >25 nmol/L, though this was significantly less so than in the more deficient group. Although overall there was no benefit reported in those with 25(OH) vitamin D levels in the 25-50 nmol/L, 50-75 nmol/L or >75 nmol/L groups, they did not break this down by whether they were in the daily/weekly or bolus therapy groups (i.e., it is likely that as with the above analysis, there was benefit if these subgroups were limited to those on daily/weekly therapy. It would have been useful to see if there were a cutpoint of vitamin D level below which there was benefit from daily/weekly supplementation). And it does not appear that participants were randomized by their entry 25(OH) vitamin D levels in the individual studies in the meta-analysis.
- This very low level of vitamin D deficiency (<25 nmol/L) is quite common in my practice (though I do work in Boston, a pretty high latitude, and with many patients with darker skin pigmentation and who are older, all 3 of which predispose to low vitamin D levels), and estimates are that around 6% (20 million) US adults are markedly vitamin D deficient (i.e. <10 ng/ml)
- These conclusions are somewhat limited because of the small numbers of many racial or ethnic subgroups in the individual studies included in the meta-analysis
- One pretty striking conclusion (and very unexpected to me) was that there would be such a difference between those given bolus supplements vs daily/weekly vitamin D. Other studies had suggested that giving the equivalent of about 800 IU/day, whether daily, monthly, or every 6 months, didn’t matter much, at least in terms of bone health. This data suggests that there may be a difference, at least for other outcomes.
- Though I have not sent out any blogs on vitamin D for quite some time (see here for a slew of those blogs), there are some suggestive studies showing there are vitamin D receptors are all over the body (including the immune system, and also being associated with the induction of antimicrobial peptides and other antimicrobial chemicals such as reactive oxygen intermediates), observational studies of mortality benefit as well as benefit for specific autoimmune and other diseases, and even a study finding that vitamin D supplementation improves the response of patients with TB to medications (see doi:10.1073/pnas.1200072109/-/DCSupplemental). Several newer studies have had some mixed results: some showing higher vitamin D levels are associated with decreased multiple sclerosis, better breast cancer survival, fewer relapses from ulcerative colitis, etc.; but some showing no benefit for pain in those with knee osteoarthritis, walking distance in those with heart failure; and several with contradictory results, e.g. for falls in the elderly, asthma in kids. A recent JAMA study (see Lappe J. JAMA. 2017; 317(12):1234) of 2303 healthy postmenopausal women >55yo in rural Nebraska found that vitamin D supplementation did not lead to decreased cancer (there are several other studies finding decreased breast, colon, and prostate cancer in those with higher vitamin D levels). But, despite the headline in Physician’s First Watch on 3/29/17 that “vitamin D supplements fail to prevent cancer in 4-year trial”, this trial looked at women who had a baseline 25-OH vitamin D level of 32.8 ng/ml (much higher than what is considered an adequate vitamin D level), were given 2000 IU of vitamin D and 1500 mg of calcium daily, and achieved a quite high vitamin D level of 43.9 ng/ml. So not so surprising that there was no benefit, since they were not even close to being vitamin D deficient. Of note the 2011 Endocrine Society suggested that the cutpoint for vitamin D deficiency should be 20 ng/ml, though there are some suggestive data that it should be higher, at 30 ng/ml, though they felt that the data were not strong enough to push for this higher cutpoint (see doi: 10.1210/jc.2011-038). And even the 30 ng/dl cutpoint is lower than the mean in the Nebraska study!!
So, my general sense is that we probably should replete vitamin D levels (either more sun, if in the right latitudes and people are able, or by vitamin supplementation) in those who are deficient, since:
- We do have vitamin D receptors all over our bodies for a likely evolutionary reason (and humans started off in climates where there was lots of sun)
- There are lots of observational studies showing benefit for adequate vitamin D levels (though observational studies do not confer clear conclusions of benefit)
- We should be careful about being too reductionist (i.e., there is not one magic bullet for health, but a constellation of healthful activities in concert, and this likely includes adequate vitamin D levels, either by adequate sun exposure or by taking supplements to increase vitamin D levels when the sun exposure is inadequate),
- There seems to be very few adverse effects of vitamin D supplementation, including in the above Nebraska study where they achieved quite high vitamin D levels
- And, unlike almost every other medical intervention, there is minimal cost, with potentially large benefit.