vitamin D supplement: not decrease fractures???
A new article was just published in NEJM finding that vitamin D supplementation did not prevent incident bone fractures (see vit d not dec fx NEJM2022 in dropbox, or DOI: 10.1056/NEJMoa2202106). Though a not-so-great study, the article was notable for its wide popular press, with, for example, the NY Times headlines: Study Finds Another Condition That Vitamin D Pills Do Not Help.
Details:
-- an ancillary study of the benefits of vitamin D and omega-3 supplementation (the VITAL study), this one assessed the role of supplemental vitamin D vs placebo in the risk of fractures
-- for the original study 25,871 participants were randomized to 2000 IU of vitamin D per day, 1 g of n-3 fatty acids per day or both (vs double placebo) to assess primary prevention of cancer or cardiovascular disease in men at least 50 years old and women at least 55 years old
-- participants were not recruited on the basis of vitamin D deficiency, low bone mineral density, or fracture history, but they were excluded if they had a history of cancer, cardiovascular disease, or hypercalcemia
-- 51% women, mean age 67, 20% Black participants/71% non-Hispanic white, BMI 28, current smokers 7%
-- diabetes 14%/parental history of hip fracture 15%/rheumatoid arthritis 4%/history of fragility fractures 10%/unintentional falls in the past year 27%/current use of osteoporosis meds 5%/current use of supplemental vitamin D 43%
-- baseline 25-OH D level 31 ng/mL, baseline calcium level 9 mg/dL
-- for subsequent subgroup analyses, they broke down the overall group into quartiles of 25(OH)D levels:
-- only 2.4% had baseline 25-OH D levels < 12ng/mL
-- Participants submitted annual questionnaires detailing demographic characteristics, medical history, medication use, and supplements (including calcium, vitamin D, and fish oil)
-- primary endpoints: incident total, nonvertebral, and hip fractures
-- secondary endpoints: incident total, not vertebral, and hip fractures but excluding toe, finger, skull, periprosthetic, and pathologic fractures (there were 42 periprosthetic and 29 pathologic fractures, the latter including tumors and Paget’s disease)
-- median follow-up 5.3 years
Results:
-- in a sample of participants on vitamin D 2000IU, their 25(OH)D levels increased from 29 to 41 ng/mL, with minimal decrease in PTH levels
-- 1991 incident fractures in 1551 participants were picked up on patient questionnaires but were confirmed by some combination of radiologic reports, radiologic images, orthopedic notes, operative or procedure reports
--primary endpoints:
-- total fractures: 769 in the vitamin D group and 782 in placebo, HR 0.98 (0.89-1.08), p=0.70, not statistically significant
-- nonvertebral fractures: HR 0.97 (0.87-1.07), p=0.50, not statistically significant
-- hip fractures: HR 1.01 (0.70-1.47), p=0.96, not statistically significant
-- secondary endpoints: essentially the same as primary endpoints, though with about 100 fewer patients in the vitamin D and placebo group for both total and non-vertebral fractures
--subgroup analyses (these were not prespecified): essentially no difference between vitamin D supplementation vs placebo for: age (above 66.7 vs below 66.7), sex, race/ethnicity, BMI, taking osteoporosis meds, history of fragility fractures, baseline use of vitamin D, supplemental calcium, baseline 25(OH)D levels (stratified to median of <31 vs more), or quartile of 25(OH) D levels (divisions as noted above); however some of these subgroups had <100 fractures
Commentary:
-- an estimated 53.6 million Americans have osteoporosis, low bone mass, or both in the US and 2 million osteoporotic fractures annually (this number is likely to increase to 3 million fractures per year by 2040 with the aging population), and related costs of more than $95 billion per year
--vitamin D supports skeletal health and improves bone mineralization by increasing intestinal calcium absorption, reducing secondary hyperparathyroidism, and decreasing bone turnover
-- the Institute of Medicine in 2011 recommended dietary vitamin D of 600-800 IU per day, theoretically leading to a total 25(OH)D level of 20 ng/mL; the 2011 recommendations from the Endocrine Society were for 800-2000 IU/d to achieve a vitamin D level of the least 30 ng/mL(see vit d Endocrine Guidelines 2011 in dropbox, or doi: 10.1210/jc.2011-0385)
-- the conclusion in the above study was that there was no significantly lower risk of fractures by taking vitamin D 2000IU than placebo among patients who were not selected for vitamin D deficiency, low bone mass, or osteoporosis
-- but, unfortunately, many of the studies nowadays have important information buried in the supplementary material, which is much less accessible to people who do not subscribe to the journal or have access to a library having the journal or have the time to scrounge for details
-- for example, they divided their baseline 25(OH)D categories into <20 ng/mL, 20-29 ng/mL, and at least 30 ng/mL. There were equal numbers of fractures with vitamin D supplementation vs placebo in all of those felt to be vitamin D replete by the IOM guidelines of having a 25(OH)D level of at least 20 ng/mL. And the group <20ng/mL was much smaller, with 2161 patients, and they had very small numbers of fractures: 53 nonvertebral fractures in those in the vitamin D group and 49 in placebo; and a total of only 5 hip fractures. These small numbers engendered wide confidence intervals, making statistical analyses less reliable. And, given that the average baseline 25(OH)D level was 31ng/mL, there were likely lots of people in the <20ng/mL group who had 25(OH)D levels quite close to this vitamin D replete category of 20ng/mL
-- and, perhaps most notably, their defined group with the clearest vitamin D deficiency, having 25(OH)D levels <12ng/mL, included only 2.4% of the overall group; in this group there were only 401 patients total having a grand total of 15 nonvertebral fractures and 2 hip fractures. Again, hard to make a strong argument that vitamin D supplementation was “not statistically significant” given that the vast majority of the population assessed had really quite good baseline vitamin D levels for bone health
-- the broad assertion of the NY Times above and in the broader popular press (and likely influencing the perceptions of clinicians as well, especially if they were to read only the abstract in NEJM) is a bit concerning given that vitamin D receptors are on almost all cells in the body, and per the parsimonious view of evolution, are likely there for a reason. and, though this study focused only on bone health, many people (including clinicians) may come away thinking that vitamin D supplements are useless
-- there are several studies suggesting that vitamin D has a role in cancer prevention for example, including one primary substudy of the VITAL study (see http://gmodestmedblogs.blogspot.com/2020/12/vitamin-d-coffee-decreases-advanced.html ). And several studies have also suggested a role for cancer prevention, though these are largely observational studies, and their analysis does suggest there may need to be a higher vitamin D threshold than found in the VITAL study, with levels more in the 50 ng/mL range
-- the real advantage of observational studies is that they can have huge databases of information over many years and typically control for many of the important risk factors, though these studies can only conclude that there are associations and not causality, since there could be important unassessed confounders that led to the result
--their disadvantage is that they are not randomized controlled studies with specific goals and clear inclusion/exclusion criteria that would allow for attributing causality
-- but RCTs are costly to run and very difficult to continue long enough to produce long-term assessments. And they are typically funded by drug companies (now, at least), who have their own agendas (eg designing the study to most likely get the conclusion they want, shorter studies are more likely to produce benefits based on their preliminary findings but decrease the likelihood of finding longer-term adverse effects, and minimizing overall expense)
--one overriding concern about the VITAL study is whether any short-term result for participants on given vitamin D is likely to have important outcomes:
-- in this case of fractures, participants were on average 67yo who were put on vitamin D supplements who overall were very healthy, fit people with high baseline vitamin D levels that increased from 29 to 41 ng/mL but without significant change in PTH levels (the baseline PTH level was 30 ng/mL and in mid-normal range, which typically should be elevated in those with vitamin D deficiency)
-- by age 67, the women were postmenopausal, so had already lost lots of their bone mineralization: the 5-7 year peri- to post-menopausal time is a time when bone loss is accelerated by on the order of 2-3x
-- It is pretty clear in several observational studies that those with higher 25(OH)D levels have less osteoporosis as determined by their bone mineral density, and that supplementation with vitamin D in those deficient leads to lower levels of several bone turnover markers. One other VITAL study did find that vitamin D supplementation increased bone mineral density but only a small amount: see vit d osteoporosis BestPract endo2011 in dropboix or doi:10.1016/j.beem.2011.05.002, and J Bone Miner Res 2020; 35: 883
93.
Limitations:
--patients involved in the study were predominantly non-Hispanic white and Black, limiting generalizability to other groups
-- also, this was a relatively fit group with minimal comorbidities, a BMI of 28, only 7% were smokers, and there was a relatively low osteoporosis risk: 5% on osteoporosis meds, baseline vitamin D and calcium levels were told were normal, 2% were on glucocorticoids, rheumatoid arthritis in 4%, history of fragility fractures in 10%
-- several important factors were not included in the study, including exercise and diet
-- and, per the above, this study seemed to be underpowered to have strong conclusions
-- and, the horse was already out of the barn: these participants had already lost a lot of the bone mineralization
So, given the above-mentioned limitations of this study, I would make the following points:
-- this study does not answer the really basic question: does vitamin D supplementation at an early age in those vitamin D deficient and continued for decades prevent bone demineralization and clinical outcomes of fracture risk? And does it have any utility for preventing other potential vitamin D-deficient concerns (cancer, multiple sclerosis, heart disease, immunologic dysfunction…..)?
-- many of these potential vitamin D-deficient problems begin early in life (eg bone demineralization in women begins in their 30s), and, once established, are insufficiently treated by later interventions. the issue with bones seems to be similar to cardiovascular disease: statins do work for secondary prevention of heart disease, with relative risk reductions for cardiovascular events of about 30%, similar to those found in primary prevention. BUT those in the secondary prevention group who already have heart disease have a much higher absolute risk of further cardiovascular disease outcomes
-- ie, preventing disease, a usual primary care imperative, does make sense
-- vitamin D receptors are throughout the body, on pretty much every cell including the diverse components of the immunologic system
-- vitamin D supplementation is cheap, easy, and basically without adverse effects
-- so a short-term study as above in people 67yo who are healthy, and with the vast majority of patients having really good baseline vitamin D levels, who have likely have already lost a lot of bone mineralization by that age, and assessing outcomes only on bones (in this substudy of VITAL) does not really answer the question…
I think there are a few vitamin D options that still make sense:
-- in areas of high sun exposure, it makes sense to check a vitamin D level, since some people may not get sufficient sun exposure (eg wearing a hat and lots of clothes when they are in the sun: after all, sun is a skin carcinogen) or having genetic problem achieving adequate vitamin D levels
--- in areas with much less sun exposure (eg Boston, where there are only a few midday hours in the summer with adequate UV light), I think there are 2 reasonable options:
-- check vitamin D levels and then treat if needed to a level of at least 30 ng/mL (though immunologic or cancer benefit may require higher levels)
-- or just prescribe vitamin D and at some point make sure the achieved 25(OH)D level is okay
geoff
If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org
to get access to all of the blogs (2 options):
1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order
2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category
3. or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them
or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center. This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.
please feel free to circulate this to others. also, if you send me their emails, i can add them to the list
Comments
Post a Comment
if you would like to receive the near-daily emails regularly, please email me at gmodest@uphams.org