October 28, 2009

Antioxidant Vitamins Raise Insulin Resistance and Render Exercise Useless

Nothing has been dearer to the hearts of nutritionists than antioxidants. The concept, like all concepts promoted by nutritionists, is easy to understand. Oxidation is the highfaluting word for what happens when things rust. So taking the antioxidant vitamins, Vitamin C and Vitamin E is supposed to "keep your body from rusting."

Except it doesn't. A slow trickle of bad news has been coming through over the past couple years linking antioxidants with bad outcomes.

The first finding came from a large scale study conducted in England where half of 20,536 people considered high risk for heart disease took vitamin C,E, and beta-carotene supplements and half didn't. It found no difference at all in the rates of heart attack, other signs of cardiovascular disease, cancer or, indeed, hospitalization for any other cause.

Then a February 2007 study found that antioxidant supplements actually seemed to raise the risk of death in those who took them.

Yet another blow was dealt to the idea that antioxidants were helpful by the results of the Physicians Health Study II published in 2008. In this double blind, placebo controlled study of 14,641 male physicians taking Vitamin C or E or a placebo that lasted a decade, the conclusion was, " neither vitamin E nor vitamin C supplementation reduced the risk of major cardiovascular events." Not only that, but "...vitamin E was associated with an increased risk of hemorrhagic stroke."

However, there was some hope that supplementing with these vitamins might be of some use specifically in people with diabetes after studies showed that the beta cell was uniquely vulnerable to oxidative stress because it is poor in the production of antioxidant substances.

A paper published in 2000 that analyzed results of the large scale EPIC-Norfolk study seemed to suggest this was true. It found that the higher the plasma vitamin C level in the 6,458 people they studied, the lower their hba1c seemed to be.

But the question was whether the high level of vitamin C actually caused the lower blood sugar levels, or whether its presence was a marker for something else--for example a diet low in junk food.

A further analysis of EPIC Norfolk data published in 2004--after the early results were in suggesting he ineffectiveness of vitamin supplementation against heart disease, pointed to the latter explanation. The study title says it all: Occupational social class, educational level and area deprivation independently predict plasma ascorbic acid concentration: a cross-sectional population based study in the Norfolk cohort of the European Prospective Investigation into Cancer (EPIC-Norfolk) Shohaimi S, Bingham S, Welch A, et al. Eur J Clin Nutr, Mar 31 2004, e-pub.

Why drag this up again? Because two very intersting studies that came out this past year cast some light on WHY antioxidants may be bad for us.

The first is this study, available in full text:

Antioxidants prevent health-promoting effects of physical exercise in humans Michael Ristow et al. PNAS May 26, 2009 vol. 106 no. 21 8665-8670 doi: 10.1073/pnas.0903485106

The researchers in the PNAS study found that "Exercise increased parameters of insulin sensitivity (GIR and plasma adiponectin) only in the absence of antioxidants in both previously untrained (P < 0.001) and pretrained (P < 0.001) individuals."

Why? A rodent study conducted by Tony Tiganis and published in Cell Metabolism in October of 2009 [full text available online as of Oct 25, 2009] found that high doses of antioxidants may interfere with cellular processes in a way that increases insulin resistance.

Reactive Oxygen Species Enhance Insulin Sensitivity" Kim Loh et al. Cell Metabolism,Volume 10, Issue 4, 260-272, 7 October 2009, doi:10.1016/j.cmet.2009.08.009

To read a description of this study in layman's terms view:
Reuters: Antioxidants may increase diabetes risk

Since it is a mantra repeated throughout Diabetes research that ROS are strongly associated with the damage done by diabetes to our organs, after reading this we find ourselves scratching our heads.

We are told to exercise, but this research suggests exercise is only effective in reducing insulin resistance when antioxidants are not present and ROS are produced.

So what message should we take from this? My guess is that it suggests we should confine our intake of antioxidant vitamins to those that come bound up with foods that have long made up a normal part of the human diet. These foods contain very modest doses of antioxidants.

The studies finding problems with antioxidants all use large doses of supplemental vitamins. It is likely that the amount of Vitamin C in your veggies and the Vitamin E in your sunflower seeds is low enough to avoid overwhelming cellular processes like lab created vitamins do.

Rust is not the only example of oxidation in our environment. There is also flame. And when you are trying to burn nutrients, maybe a bit of oxidation is good for you.



Venkat said...

Thanks for this post Jenny. It is very informative and helpful.



nilaus said...

I think these large scale studies should be swallowed with a huge grain of salt. The role of vitamin C has been the subject of debate for years (with Linus Pauling as a warm proponent for ample ingestion of the simple molecule).

Perhaps the insulin resistance is caused by sugar and vitamin C competing for the same receptors.

I would love to see some more details on the design of the studies/statistics etc.

Also, vitamin C = ascorbic acid both in pills and vegetables.

Jenny said...


Follow the links in the post to the articles. Several are free full text.

Linus Pauling was an expert in a different domain and his knowledge of biology was quite limted. His enthusiasm for Vitamin C always struck me as an old man's peculiarity.

My guess is that "insulin resistance" measured during exercise is really glucose burning and has little to do with insulin receptors. After exercise everyone's "Insulin Resistance" drops back to baseline within a very short period. And it stays pretty much where it always is. The exception is the kind of "Insulin resistance" which is due to very high blood sugar or, conversely, very low sugars which caused a counterregulatory burst.

The whole concept of IR is way overdue for a rethink.

RLL said...

Jenny - while I agree with what you have to say, Linus Pauling was not a distant second in the race with Watson and Crick in the discovery of DNA, and would be seen in their territory as a biologist.

My bigger question regards Ocuvite - born with terrible eyes, glaucoma, mac. degen., retinopathy. I take it because there is some (faint?) evidence that it delays m.d. It is a different mix that what was considered in the studies today. RobLL

nilaus said...

I don't think you are being fair to Pauling's legacy. He was desperately trying to point out huge flaws in published medical papers using the strict scientific discipline that (among other things) brought him his Nobel prize in chemistry.

As usual when presenting more involved scientific arguments to doctors, you are 1) ridiculed or 2) ignored. Pauling experienced this manifold in the later years of his life.

His book "How to live longer and feel better" actually fits well with the wonderful approach of your blog. His approach is very concise and well worth studying.

Unlike the constant stream of published large scale studies he was trying to actually understand what was going on both on a statistical but also on a more fundamental level.

Red Sphynx said...

This doesn't close the book on Vitamin E. Each of these studies used only the α-tocopherol form of vitamin E; there are 8 different forms and two others of them (γ and δ) are abundant in natural foods. Huang and Apell showed that α-only supplements cause the γ and δ blood-levels to drop. The deletrious effects of Vit E observed in these studies could be entirely due to this.

But when I read Huang and Apell, I switched from a daily 400 IU cap of α-tocapherol to one 400 IU cap of mixed tocopherols per week.

More on this: Steve Harris has some good discussion. And useful article on mixed tocopherols in diabetes.

Jim Purdy said...

So vitamin C supplements are bad ... unless they are good?

And Linus Pauling was wrong ... unless he was right?

Jenny, Jenny, why can't you keep things simple for the masses who have been dumbed down my FoxNews?

ItsTheWooo said...

Speaking personally, I take a fair bit of supplements. Myo-inositol has dramatically improved mood and possibly also improved my tendency to PCOS and insulin disorders. I find myself less hungry and hormones are somewhat more regular. Mood is significantly better, and I have observed a complete elimination of a certain type of mood that has plauged me all my life. I have been on it for almost 4 months now so it is not dumb luck. Prior to taking this supplement, I would never go this long without this type of mood. Inositol particularly the myo-inositol form plays an essential role in regulating serotonin receptor function; d-chiro-inositol has roles in the function of insulin receptors and the physical process of follicular rupture during ovulation (it is presently hypothesized that PCOS may represent a neuroendocrine dysregulation originating at impaired inositol metabolism, and this explains the mood, metabolic, and reproductive axes of the disorder).
Myo-inositol is partially converted to d-chiro-inositol.

I have also observed great benefit from GTF chromium. The picolinate form did absolutely nothing, and I took it for months, almost a year. But once I started taking GTF, my appetite dropped a lot and I naturally found I could tolerate more carbohydrate. It isn't a huge difference, I still can't eat normal and feel okay, but it is definitely the difference between feeling like crap after eating raspberries vs feeling normal. I can tolerate a bit more, and that is worth it.

Another supplement that has helped me is vitamin C.

Now, this was purely incidental. I don't even know why I started taking it, I didn't expect anything from it. But I couldn't help observe on the days I took vitamin C, I often had hyper energy. I couldn't figure out what it was, until I realized that the hyper energy days correlated with the vitamin C days.
A quick google shows that vitamin C affects tyrosine and thyroid hormone. Tyrosine is the precursor to "life is wonderful" neurotransmitters (dopamine, norepi) and T3.

For me, C has been a surprisingly effective thing. I usually take popular supplements - d, zinc - and receive no benefit. But lo and behold I take the ubiquitous vitamin C and observe changes so substantial I can't ignore them.

One more thing. Since taking these supplements - inositol, GTF chromium, vitamin C - my skin quality has normalized. Prior to, I was plagued by this odd skin condition where I would often have scaly and flaky skin on my finger tips and face. It would be exacerbated when I ate certain foods, like chicken (weirdly). My mood also was worse when skin was at its most scaly.
Since taking these, my finger tips are supple and soft and I have no scaly appearance to fingers and face.

I have since learned of a condition called inositol deficiency psoriasis. It is mostly relevant to manic depressives on lithium therapy (as part of the antimanic effect of lithium is by blocking inositol metabolism, or so it is hypothesized). However, when these patients are given inositol their lithium psoriasis clears up.

It may very well be that I have some kind of f*ckery with inositol metabolism, by genetics or toxins. That bizarre scaling was so painful and miserable. Fingertips cracking and bleeding in the winter, ugh. Totally gone!
This might also explain my PCOS, mood disorder (depressive) and general carb intolerance and tendency toward obesity.

I always found it significant that my basal state is the mirror image of a manic depressive on lithium or antiepileptic therapy. PCOS, depressive or flat, carb intolerance, weight gain. My health history reads like the side effects of depakote or something, lol. (Depakote also messes with inositol, a lot of the "mood stabilizers" do).

Just food for thought.

I think supplements are very good things, if you know which ones to use. Let your well being be your guide.

dmoak said...

OK. I take a pretty potent multi-vitamin with a lot of antioxidants. I'll leave it off for a while and see what happens. Of course. if insulin sensitivity did improve, I suppose it is possible that my blood sugars would not, but that it would take less insulin to achieve the same results.

trinkwasser said...

I was just reading a load of papers by Ronald Krauss, including this one


and this (not Krauss)


both of which also throw mud on the fire: also another one I can't currently find about oxidation at the level of the mitochondria.

All of which really demonstrates that metabolism is far more complex than the soundbytes you are fed from "official" medical sites, and that attempting to fix one problem may worsen another.

I'm coming to see U curves and J curves everywhere. You need enough oxidation to generate energy but not so much that you generate damage.

dmoak said...

I know this post is a month old, but I did discontinue my supplementation, which included chromium, L-carnitine, and my multi-vitamin (which has a lot of anti-oxidents in it). I continued only the fish oil and vitamin D. My blood sugars have gone down and have stayed there, so it appears to be true for me that there was some interference with blood sugar.

Steampowered god said...

Pauling lived 90+ while the lifespan of males is usually in the 70s(80s if you don't smoke, do little alcohol, exercise, eat your veggies, eat your nuts), both his parents died before the age of 50, iirc(did his parents die of disease, accidents? or was it just natural causes?). He took 3g of vitamin C a day.

Most animals synthesize several grams of C, and of the few that don't many ingest several grams of C from their diet... the C production goes up in stress or disease conditions.(Pauling is said to have raised his C intake on such occasions)

Vitamin E requires complex antioxidant mixture as found in nuts to be effective, alone it's not very good, when present as part of a complex mixture of natural antioxidants and minerals the combo(nuts) appears to EXCEED the lifespan benefits from exercise itself.(seventh day adventist studies)

Jenny said...

Pauling was one man. There is no way of knowing if his supplementation had any impact on his life span. Maybe he would have lived longer without it. My dad who was of the same generation (and didn't supplement) lived to be 100. Meanwhile, the data pointing to the dangers of antioxidants include tens of thousands of people.

The idea that oxidative stress is like rust is one of those simplistic ideas that sounds logical but doesn't hold up to study.

Here's another study that calls into question the whole oxidative stress theory.


Steampowered god said...

"Pauling was one man. There is no way of knowing if his supplementation had any impact on his life span. Maybe he would have lived longer without it. My dad who was of the same generation (and didn't supplement) lived to be 100. Meanwhile, the data pointing to the dangers of antioxidants include tens of thousands of people."

Which is why I brought the issue of his parents death before 50, if this was not disease or accident it would imply the genetics are very very bad, not the same to have good genetics as bad genetics. With good enough genes you can smoke for 95 years and reach 100, bad genes and you're not getting anywhere near that. Though more data is needed, there was also a study of 1000s of men where 800mg C vs the 60mg RDA showed 6 year increased lifespan on men.

Remember not all of C's effect are due to antioxidant capacity, it has other effects on the body, for example by boosting collagen synthesis it would strengthen arteries and stimulate bone formation(collagen synthesis being critical for this).

"The idea that oxidative stress is like rust is one of those simplistic ideas that sounds logical but doesn't hold up to study."

Yes, I agree. Animal studies have suggested that even cutting expression of some endogenous antioxidants by half does not reduce lifespan.

Antioxidants defenses already appear close to optimal.(additional antioxidant capacity seems to merely protect from specific additional stresses or toxin exposures, not from everyday metabolic activities)

The following is an interesting read on a factor that appears to be significant in lifespan variation across organisms(even genetically identical organisms)


According to the study membrane byproducts (toxic reactive aldehydes), can have a half life of minutes( as compared to nanosecond or microsecond half lifes for traditional Reactive oxygen species).

Alteration in membrane composition, not antioxidant defenses appears to be the key to exponential lifespan differences(mammal, birds, naked mole rats, insect queens vs workers, bigger animals, etc)

A nice quote from that article:

"The social insects have been suggested as particularly good model organisms to investigate the mechanisms of aging for two reasons: 1) "queens" can be extraordinarily long living, and 2) there is sometimes tremendous variation in life span between genetically identical "queens" and "workers" (171). In some ant species, queens can live up to 30 yr and frequently live 10 times longer than workers (144). Similarly, queen honeybees are reported to have a maximum longevity an order of magnitude greater than worker (i.e., nonreproductive female) honeybees (373). Enhanced antioxidant defenses in queens are not a likely explanation, as one study has shown queen ants have a reduced expression of Cu/Zn-SOD compared with shorter-living workers (282) while others have shown that queen honeybees generally have the same (or, in some cases, lower) levels of antioxidant defenses than worker honeybees (68, 370). The mass-specific metabolic rates of worker and queen honeybees are essentially the same (96). Recent measurements show that queen and worker honeybees have different membrane fatty acid composition and that the peroxidation index of phospholipids of from queen honeybees is 33% of that of workers (122). If the slope of the relationship between PI and MLSP is mathematically similar in honeybees to that described for mammals and birds (155, see Fig. 7), then this difference is capable of explaining the order of magnitude difference in longevity between queen and worker bees."