47 replies to this topic

[stephen_b]

Intuition lends itself to the belief that too much or too little of any vitamin has potential to be harmful. I remember reading excessive folate could increase the risk of some cancers with regard to the methyl cycle, I think they were the fast growing types, like intestinal, oral, prostate etc. I'd have to look for the study.

Agreed that there is an inverted J-shaped response (for everything that actually does you good at least; I doubt that cyanide has one, but who knows). For the folate studies, be aware that the version often used is synthetic folic acid, which 60% of the population can't fully metabolize.

[chroncile]

Okay so what's the conclusion here? I take 5000 IU Vitamin D3 and 1000 IU Vitamin D2 (from LEF Multivitamin 2-per-day; I only take one) everyday. Am I taking too much?

Also, I started taking 5000 IU Vitamin D3 a couple of months ago and really, I have not noticed any effects from taking it. Should I just drop it all together or only take one per week?

[GK77]

^^TBD
However, some would say it's too much.
I would say you don't really know until you assess your risk for specific cancers and other diseases that higher levels of Vit D have an inverse relationship with and get labs. Then you can set an appropriate personalized goal for your serum Vit D and shoot for it.

Edited by GK77, 13 June 2012 - 03:33 AM.

[Michael]

The Vitamin D council's objections about the lack of adjustment for vitamin A in a Nordic population (cod liver oil) initially sound reasonable, but it's very unlikely that anyone got the very high doses that would be required to reach such high levels in Denmark from cod liver oil recommended . Their suggestion about the possible role of reverse causation sound reasonable, but the distribution of serum D by age group doesn't seem to support it. IAC, however, these are not really reasonable objections in most of the studies included in the recent meta-analysis (PMID: 22170374) showing optimal serum vitamin D levels are around 31 ng/mL. Nor is osteoporosis, in particular, terribly cogent as a driver of such an association, granted the studies showing higher risk of prostate, pancreatic, and some other cancers at high serum levels, and or the association between very high vitamin D and atrial fibrillation.

Okay so what's the conclusion here? I take 5000 IU Vitamin D3 and 1000 IU Vitamin D2 (from LEF Multivitamin 2-per-day; I only take one) everyday. Am I taking too much?

Almost certainly, unless you're a shut-in with a weak 25-(OH)D3 1-alpha-hydroxylase variant. 1000 IU is for most reasonably young white people plenty for chronic use. But as was said already to others: there's no sense declaring a given dose too much or too little for any one person: get yourself tested and see.

Edited by Michael, 16 June 2012 - 01:58 PM.

[kismet]

By the way, there are two recent meta-analyses. Zitterman 2012 (referenced by Michael) and Schöttker 2012 (below). The latter is more consistent with a slightly higher range of benefit, e.g. 30-40ng instead of ~30ng/ml, but I haven't looked into any details.


“There appeared to be a roughly inverse linear relationship of 25(OH)D levels and mortality in all studies except for the studies of Michaelsson et al., 2010, that observed a u-shaped association, and Cawthon et al., 2010, that found no association between 25(OH)D levels and all-cause mortality. For all other studies, the group with lowest 25(OH)D levels had the highest mortality risk. [but there seems to be no formal analysis of this?!]”

all-cause mortality reduced with a pooled
HR of 0.92 (95% confidence interval: 0.89-0.95) for a 20nmol/l increase in 25(OH)D levels.

“we only included population-based cohort studies addressing the outcome all-cause mortality, whereas the meta-analysis of Zittermann et al., 2012 included also studies with community-dwelling adults that do not reflect random population samples (over-sampling of subjects with ≥2 domains in disability in Semba et al., 2009, recruitment of ambulatory patients in Kuroda et al., 2009 and inclusion of high-risk patients for vitamin D deficiency because the vitamin D status was determined at the provider's discretion for clinical indications, like, e.g. osteoporosis, in Anderson et al., 2010), other study types (Bolland et al., 2010 reported on a trial of calcium supplementation) and another outcome (cardiovascular mortality in Kilkkinen et al., 2009).”

Ageing Res Rev. 2012 Feb 17. [Epub ahead of print]
Serum 25-hydroxyvitamin D levels and overall mortality. A systematic review and meta-analysis of prospective cohort studies.
Schöttker B, Ball D, Gellert C, Brenner H.

Edited by kismet, 02 July 2012 - 11:56 PM.

[platypus]

I just got my reading, 125 nmol/l. I haven't taken much vit-D supplements since last winter but sunbathing at noon seems to have done the trick. I'd like to do some more sunbathing, do you think that would be detrimental for my health (from a vit-D perspective)?

Edited by platypus, 27 September 2012 - 08:01 PM.

[niner]

I just got my reading, 125 nmol/l. I haven't taken much vit-D supplements since last winter but sunbathing at noon seems to have done the trick. I'd like to do some more sunbathing, do you think that would be detrimental for my health (from a vit-D perspective)?

To convert from the SI unit (nM) to the older style (ng/ml), divide by 2.5.

125 nM /2.5 = 50 ng/ml.

At one point in time, I would have said this was a fine 25-OH-D level, and it still might be for some people. Personally, I'm targeting 30 ng/ml now, but that's because of a strong family history of prostate cancer. Consider protecting your face and hands if you sunbathe, since they usually accrue the most lifetime damage.

[JLL]

Well this is interesting, haven't been keeping up with the vitamin D news lately... I also started taking 5,000 IU back in the day when it was the norm on this forum, and got my levels up to 45 ng/ml. Then again, I haven't been taking vitamin D lately so it's probably gone down from those levels. I have friends and family who are still taking the 5,000 IU daily without having blood tests done... guess I should warn them about these studies.

[Guest]

So is it at all clear what drives the excess mortality we see above 31 ng/ml? It sounds as if its an effect by which too much vitamin D causes more CVD and a range of cancers above that range (I'm being intentionally mean here; I know that its not causation in itself).

Can we say which effects dominate the curve? Is the higher cancer mortality at the high range responsible for the higher all cause mortality or can one describe the rising CVD mortality as driving factor for all cause mortality?

Personally I suppose that it's easier for me to modify a CVD risk by life style measures than a cancer risk. Also research targetting CVD apears to be much more promising than those targetting cancer. So if CVD mortality would be the dominating factor I'd be much less concerned about overdosing (say, getting 45 ng instead of 31 ng).

[Guest]

Just to bump the last question I posted. Going over the studies on total mortality and known cancer-related effect again I suspect that yes, indeed, the excess mortality can be attributed largely to an increase in CVD. I see it this way:

The vast majority of studies on Vitamin D and cancer (though there is only one RCT in humans) point to a much higher level than 31 ng for optimal cancer prevention, especially concerning colon and breast cancer, but also other kinds of tumors. Conversely the level deemed to be optimal for CVD prevention always almost is set below the level optimal for various cancers. Its just a hunch though, but considering that levels as high as 60 ng likely will result in elevated calcium this could facilitate CVD compared to its CVD-optimal level (but nothing that could be treated with some Vitamin K2).

I'd appreciate any opinions on this one. In the long run cancer seems to be a much harder problem to takle than CVD (even within in the SENS approach). Considering that the 31 ng/ml stem from a general population sample - which clearly are not popping K2 or generally optimizing their CVD risk lifestyle wise - it might seem to be a wise choice to up your D-intake to reach levels above 50 ng/ml. Unless of course you have a particular (familiy) history of hearth disease.


Thoughts?

[goobicii]

I am 80 kg low BF,if I drop 5000iu everyday will it be bad? many days I get nearly none sun other times I get couple hours

[david ellis]

Just to bump the last question I posted. Going over the studies on total mortality and known cancer-related effect again I suspect that yes, indeed, the excess mortality can be attributed largely to an increase in CVD. I see it this way:

The vast majority of studies on Vitamin D and cancer (though there is only one RCT in humans) point to a much higher level than 31 ng for optimal cancer prevention, especially concerning colon and breast cancer, but also other kinds of tumors. Conversely the level deemed to be optimal for CVD prevention always almost is set below the level optimal for various cancers. Its just a hunch though, but considering that levels as high as 60 ng likely will result in elevated calcium this could facilitate CVD compared to its CVD-optimal level (but nothing that could be treated with some Vitamin K2).

I'd appreciate any opinions on this one. In the long run cancer seems to be a much harder problem to takle than CVD (even within in the SENS approach). Considering that the 31 ng/ml stem from a general population sample - which clearly are not popping K2 or generally optimizing their CVD risk lifestyle wise - it might seem to be a wise choice to up your D-intake to reach levels above 50 ng/ml. Unless of course you have a particular (familiy) history of hearth disease.


Thoughts?

It's not a research study, but Track Your Plaque folks, think that higher Vitamin D levels will reduce plaque. They are verifying their protocols with calcium scoring CT scans.

Edited by david ellis, 05 June 2013 - 05:52 PM.

[niner]

I am 80 kg low BF,if I drop 5000iu everyday will it be bad? many days I get nearly none sun other times I get couple hours

If you are sometimes getting a couple hours of sun, you might not need any supplemental D. The only way to really know would be to get your 25-OH-D3 level checked. If you don't get checked, I wouldn't go over 1000 IU/day. Skip it entirely if you have a family history of prostate cancer.

It's not a research study, but Track Your Plaque folks, think that higher Vitamin D levels will reduce plaque. They are verifying their protocols with calcium scoring CT scans.

That's right- Dr. Davis is a proponent of vitamin D. I'll bet they have the appropriate amount of vitamin K in the mix, though.

[nameless]

It's not a research study, but Track Your Plaque folks, think that higher Vitamin D levels will reduce plaque. They are verifying their protocols with calcium scoring CT scans.

That's right- Dr. Davis is a proponent of vitamin D. I'll bet they have the appropriate amount of vitamin K in the mix, though.

The thing about the Track your Plaque folks is they are doing a ton more things than simply increasing Vit D. Major diet changes, exercise, fish oil, etc. ... have they done enough scans (or any scans) using the same exact protocol, multiple times, but changing just Vit D serum levels, to see if the Vit D is having an effect?

[timar]

Please remember that correlation is not causation! I think there are far more evident explanations for the U-shaped mortality curve than hypothetical interactions between vitamin D and vitamin A or K or other ill effects of vitamin D itself.

First, interventional studies with vitamin D generally had favorable outcomes with regard to all-cause mortality as well as specific endpoints, many of them targeting much higher blood levels than the 80 nmol/l associated with lowest all-cause mortality in the Danish cohort. Primates, hunter-gatherer populations as well as people occupationally exposed to sunshine (i.e. life savers) naturally obtain blood levels between 150 and 200 nmol/l from sun exposure (and no, dear Weston A. Price fellows, primates usually don't eat organ meat and many hunter-gatherers discard it too).

Second, we know that certain gene variants have an marked impact on cholesterol homeostasis. Vitamin D is a cholesterol metabolite and thus any modulation to cholesterol homeostasis at large is expected to have an impact on vitamin D metabolism. We know of at least two gene variants (there are certainly more) which have an either a positive or negative impact on cholesterol metabolism and longevity, and the exact opposite effect on vitamin D levels (ApoE4[1] and a variant of CYP2R1[2]), showing a plausible metabolic correlation between a favorable blood lipid profile and lower blood levels of calcidiol (either because of reduced cholesterol synthesis or because of enhanced clearance through the cytochrome pathway).

Thus, the nonlinear increase in all-cause mortality risk positively associated with vitamin D levels is most probably the effect of underlying genetic variations regarding cholesterol homeostasis, such as CYP2R1 and ApoE.

[1] http://inflammation-...min-d-und-apoe4
[2] http://www.ncbi.nlm....les/PMC3519162/

Edited by timar, 08 June 2013 - 08:12 PM.

[timar]

What I would really like to see is an epidemiological study where mortality risk associatiated with vitamin D (calcidiol) levels is corrected for cholesterol (LDL/HDL ratio, or even better, ApoB/ApoA ratio). My prediction is that the U-shaped curve would give way to an inverse J-shaped curve, or a least a U-shaped curve with its minimum at a much higher vitamin D level and less adjacent increase in mortality risk at higher levels. This would have to exclude statin users though - which is almost impossible nowadays.

Edited by timar, 11 June 2013 - 08:06 AM.

[Guest]

Second, we know that certain gene variants have an marked impact on cholesterol homeostasis. Vitamin D is a cholesterol metabolite and thus any modulation to cholesterol homeostasis at large is expected to have an impact on vitamin D metabolism. We know of at least two gene variants (there are certainly more) which have an either a positive or negative impact on cholesterol metabolism and longevity, and the exact opposite effect on vitamin D levels (ApoE4[1] and a variant of CYP2R1[2]), showing a plausible metabolic correlation between a favorable blood lipid profile and lower blood levels of calcidiol (either because of reduced cholesterol synthesis or because of enhanced clearance through the cytochrome pathway).

Thanks for your contribution - I did not had those interactions on the radar. I knew that especially in northern populations we see a relatively high percentage of people which are Apoe4 homo-/heterocygote (somthing in the 20%-25% if I remember correctly), but somehow missed the connection to elevated Vitamin D status in those groups. It would have been valuable, too, to have a detailed analysis of causes of death or general health status in the epidemological studies - though "Alzheimers" might not be registered seperately. If those 20%-25% would all be in the high-Vitamin-D percentiles this would certainly limit the conclusions you can draw from those kind of analysis - hell, Apoe4-homocygotes would have had the highest "natural" Vitamin-D levels and would be among the lowest in life expectancy.

I'd really like to contact the authors of the danish study to see if they inofficially have more information at hand.

[NocicepticBoss]

I have a paper from Canada here (2011):

Abstract:

The Institutes of Medicine (IOM) recently revised the recommended
dietary allowances (RDA) for vitamin D, to maintain
serum 25-hydroxyvitamin D (25(OH)D) at or above 50 nmol/L, to
sustain bone density, calcium absorption, and to minimize risk of
osteomalacia and rickets. However there are compelling reasons
why 25(OH)D should preferably exceed 75 nmol/L: (A) Scrutiny of
actual data specified by the IOM relating 25(OH)D to bone density
and osteomalacia shows the desirable minimum 25(OH)D to be
75 nmol/L (30 ng/mL). (B) Humans are primates, optimized
through evolution to inhabit tropical latitudes, with serum 25(OH)
D over 100 nmol/L. © Epidemiologic relationships show health
benefits if 25(OH)D levels exceed 70 nmol/L; these include fewer
falls, better tooth attachment, less colorectal cancer, improved
depression and wellbeing. Some studies of populations at highlatitude
relate higher 25(OH)D to risk of prostate cancer, pancreatic
cancer or mortality. Those relationships are attributable to the
dynamic fluctuations in 25(OH)D specific to high latitudes, and
which can be corrected by maintaining 25(OH)D at steady, high
levels throughout the year, the way they are in the tropics. (D)
There are now many clinical trials that show benefits and/or no
adversity with doses of vitamin D that raise serum 25(OH)D to
levels beyond 75 nmol/L. Together, the evidence makes it very
unlikely that further research will change the conclusion that risk
of disease with serum 25(OH)D higher than 75 nmol/L is lower
than the risk of disease if the serum 25(OH)D is approximately
53 nmol/L.

Conclusion:

Conclusion
Dr. Bouillon argues elsewhere in this journal for the conservative perspective, that a lower desirable
level – 50 nmol/L (20 ng/mL) for 25(OH)D – is justified unless and until much stronger RCT evidence
can support the efficacy and the safety of widespread use of vitamin D intakes beyond 800 IU/day (see
Why modest but widespread improvement of the vitamin D status is the best strategy in this issue). Since
most modern populations exhibit 25(OH)D levels that are on average, higher than 50 nmol/L,46,47 the
conservative perspective adheres to the notion that endemic levels of 25(OH)D are inherently the
healthiest ones. The conservative approach to vitamin D recommendations does not take into account
all of the evidence available, because it downplays the usefulness of existing knowledge of human
biology, epidemiology, and even the existing evidence from clinical trials. In at least 7 clinical trials the
25(OH)D level in the control group were at or above the IOM value of 50 nmol/L, yet additional vitamin
D produced significant health effects.32–37,43 To paraphrase the highest level criterion for the GRADE
evidence based approach,10 “Is further research likely to alter the conclusion that risk of disease is
lower if serum 25(OH)D exceeds 75 nmol/L than if the level is at 51 nmol/L?” A desirable minimum
25(OH)D level of 75 nmol/L is readily justifiable now, based on the very criteria used by the IOM, as well
as by the basic biology of humans, by epidemiologic relationships and by existing evidence from
randomized clinical trials. There will always be further ongoing studies to await, but the conservative
perspective begs the question, “Exactly how much evidence is needed before the IOM and the IARC
accept that a minimally desirable serum 25(OH)D of 75 nmol/L is preferable to 50 nmol/L?” The
conservative view that 50 nmol/L is appropriate is not scientifically defensible. Modest scrutiny of the
key criteria that the IOM committee specified to justify its conclusion shows that healthy bones require
a 25(OH)D level substantially higher than what they claim.

Vieth, R. Why the minimum desirable serum 25-hydroxyvitamin D level should be 75 nmol/L (30 ng/ml). Best Practice & Research Clinical Endocrinology & Metabolism 25 (2011) 681–691

Edited by NocicepticBoss, 11 June 2013 - 01:55 PM.