Thursday, May 7, 2009

Dihydro-Vitamin K1

Step right up ladies and gents; I have a new miracle vitamin for you. Totally unknown to our ignorant pre-industrial ancestors, it's called dihydro-vitamin K1. It's formed during the oil hydrogenation process, so the richest sources are hydrogenated fats like margarine, shortening and commercial deep fry oil. Some of its benefits may include:
Dihydro-vitamin K1 accounts for roughly 30% of the vitamin K intake of American children, and a substantial portion of adult intake as well. Over 99 percent of Americans have it in their diet. Research on dihydro-vitamin K1 is in its infancy at this point, so no one has a very solid idea of its effects on the body beyond some preliminary and disturbing suggestions from animal experiments and brief human trials.

This could be another mechanism by which industrially processed vegetable oils degrade health. It's also another example of why it's not a good idea to chemically alter food. We don't understand food, or our bodies, well enough to know the long-term consequences of foods that have been recently introduced to the human diet. I believe these foods should be avoided on principle.


Aaron Blaisdell said...
This comment has been removed by the author.
Aaron Blaisdell said...

I guess there's a good reason it's called shortening--its effect on lifespan. I'm glad I've dramatically reduced its presence in my and my family's diet.

Tom Moertel said...

Regarding, "It's also another example of why it's not a good idea to chemically alter food," does homogenization of milk qualify for something that ought to be avoided?

The research I've seen is inconclusive. In a recent (2007) survey article, for example, the last line of the abstract reads: "The impact of homogenization, as well as heating and other treatments such as cheesemaking processes, on the health properties of milk and dairy products remains to be fully elucidated." [1] To me, that reads a lot like, "We don't know whether modern dairy treatments have harmful effects."

I wish I knew the truth, because one of the most convenient ways to get the right vitamin K is by drinking whole milk from pastured cows. Unfortunately, most of what's available is homogenized and pasteurized, thanks to our dairy laws.

[1] On the supposed influence of milk homogenization on the risk of CVD, diabetes and allergy, 2007, Br. J. of Nutrition

Stan Bleszynski said...

Quote from the paper:

"Another dietary form of vitamin K, 2',3'-dihydrophylloquinone, is a product of commercial hydrogenation of phylloquinone-rich oils and does not occur in nature (9).Interestingly, they did not put this statement in the abstract!

Perhaps it is a purely accidental omission or sloppiness, but one does not get any ideas from just their abstract, of the importance of this issue!

If the authors wished to obfuscate or bury their work, they have chosen a very good strategy. I am sure they will be "rewarded" with more grants. 8-:)

Last but not least the effect on vitamin K2 production status in serum (in rats) was not just significant but HUGE! Almost 1:3!

It would be very very interesting to analyze K2 status versus hydrogenated oils in humans. Since humans do not convert K1 --> K2MK4, my guess is that the result will probably be even worse!

Thanks for digging it out Stephan!

Robert McLeod said...


Very interesting, I will have to dig through those links again later.

This does bring up my previous question as to whether or not linoliec acid itself is harmful or rather our modern processing of it, the refining, deoderization, etc. process. Then we take that unstable polyunsaturated fat and store it in air, exposed to light, for a long time and expect it to not be rancid?

Yet it's possible to treat and store vegetable oils properly (chilled in a dark bottle), and many people in the paleo community are enjoying nut and flax oils without signs of troubles from it. I still would agree that polyunsaturated fats shouldn't be cooked with.

After all we know saturated fats are ok when unprocessed, but the transaturated versions have similar problems. However, early studies that ignored the difference gave saturated fats a bad name.

From this point of view, food sterilization through irradiation also looks like a bad idea. At minimum it will likely break micronutrients, with the potential danger that the fragments are physiologically harmful.


I don't think pasteurization can cause chemical changes, it just kills bacteria.

Hydrogenation involves: much higher temperatures, the presence of a catalyst, hydrogen gas (a very strong reducing agent). All three of these aspects combine to make chemical reactions possible that otherwise wouldn't occur in nature.

Monica said...

"I don't think pasteurization can cause chemical changes, it just kills bacteria."

Not an issue I've researched intensively, but I think it probably can cause chemical changes. Pasteurization, in the sense of heating milk to 70 C or so, is probably not that dangerous, I would agree. After all, most of us have cooked milk in the past. (I think even heating will still reduce vitamin content, though.) But ultrapasteurization is a different animal, I think. The milk is swiftly taken to 140 degrees C... over the course of a couple of seconds. I can imagine that would have quite an effect on proteins in terms of denaturation or even fracturing.

I did look up some studies awhile back on the effect of pasteurized and ultrapasteurized milk on calves. My recollection is that the calves did much more poorly on the ultrapasteurized milk in comparison to pasteurized.

TedHutchinson said...

In case others have not found it yet Dihydrophylloquinone in foods leads to an online chapter of a book on Vitamin K
The leading expert in Vitamin K appears to be Booth SL whose work is funded by US Dept of Agriculture services. A pubmed search for
Booth sl dihydrophylloquinone
produces several interesting online papers

TedHutchinson said...

In trying to research this issue a little more I have found this
USDA National Nutrient Database for Standard Reference, Release 18
Content of Selected Foods per Common Measure, sorted alphabetically Vitamin K (phylloquinone)(µg)

And those who life within travelling distance of Tufts Boston MA (If it had been Boston UK I'd be there) may care to participate in this research
2445: Vitamin K Absorption
Interesting to see the use of statins excludes you from participating in this study.

TedHutchinson said...

Sorry Wrong link this is the Vitamin K trial at Tuft's

Stephan Guyenet said...


I'm a little bit suspicious of homogenization on principle, but as far as tampering with food goes, it's pretty mild. Physically altering a food is not as bad as chemically altering it in my book.


Yes, it's possible that the effect could be magnified in humans. No one really knows at this point. It may not actually actively interfere with K2 status, but based on the rodent work, I feel it's a possibility.


I'm open to the idea that components of industrial vegetable oils other than linoleic acid are major contributors to their health effects. I still think excess linoleic acid is very likely to be a problem in and of itself. But other things like dihydro-K1 could very well play a role.


Another thing pasteurization does is destroy the milk's enzyme content, including lactase. I know several lactose-intolerant people who can drink raw milk without any trouble.

Guy Adamson said...

Sounds like a great dietary supplement to send to your worst enemy! Or be promoted by government nutrition standards...oh wait...

There's definitely something to a more "gentle" form of processed milk. I was lactose intolerant but can drink nonhomogonized, low-temp, vat pasteurized milk from grass-fed cows with no problems. In fact, it appears the enzymes in the regular milk I do drink helps me process the small amount of processed dairy I also consume.

Thanks for your posts Stephan!

Monica said...

"Another thing pasteurization does is destroy the milk's enzyme content, including lactase. I know several lactose-intolerant people who can drink raw milk without any trouble."

Definitely. My fiance is very lactose intolerant and doesn't have much of a problem with raw milk, which we get on a weekly basis. He still uses a couple of lactase pills with it, which I'm not sure he absolutely needs, but he can't even drink 1/4 cup pasteurized milk even with the pills.

gallier2 said...


any ideas on micro-filtered milk? I don't know if that kind of milk is available where the readers here live, but in France (and newly also in Luxemburg) it is available in several supermarkets. AFAIK the milk is not heated but is filtered through microscopic tube that stop potential pathogenes but let everything through.
I found some links on the process, but they look to come from the providers of this technology, so I would like to know if what problems could be hidden.

From a taste point of view, I have to tell it tastes quite like real raw milk (or so I think, because the last time I tasted raw milk was some 20 years ago), it hasn't the typical after taste of cooked milk (ultra-pasteurized milk doesn't taste good at all, it has foul after taste, that I noticed since I use high quality fresh milk).

Unknown said...



It will work but since the fat micelles is about the same diameter as the pore diameter of the filter membrane I guess that the milk will be homogenized to some extent (given that sterile filtration require a pore diameter below 0.22 micron).

I don't think it will taste the same since the size distribution of the fat micelles is changed. You can feel taste differences between pasteurized milk and non-homogenized, pasteurized milk.

I also suspect that pasteurized milk is to some little extent being homogenized during the pasteurization process since you are allowing the milk to flow under higher pressure through small pipes.

Unknown said...


Will the filtered milk form cream if you allow it to stand?

What is the price difference between filtered milk and pasteurized milk?

gallier2 said...


In the description of the process, it is explicitly stated that the milk is skimmed before filtering, exactly because the fat particles are too big. The cream is pasteurized separatly and added back later. As the taste alteration of milk comes mainly from the modifications of the proteins not of the fat, the taste remains that of raw milk. The texture might be quite different, due to the homogenization I suppose.

From a price view, the milk cost is in the same range as the other brands of fresh milk (about 1,20/l) . In Luxemburg I now found a brand that is even organic, it's a bit more expensive though (€1,39/l).

Jenny Light said...

Please check out this list of snack and fast foods containing dihydrophylloquinone (dK), giving total content in each product here:

Just further reason the stay away from industrial foods!

As always, thanks Stephan for the heads up!

Anna said...

I've not seen this filtered milk, but wow, it sure sounds highly processed. I'll stick to my simple raw milk.

Monica said...

Hey Gallier2 --

My guess is that for lactose intolerant people, this might be a less ideal option. That's because I believe some of the lactase activity comes from "good" lactobacilli. My guess is that you would almost completely eliminate any probiotic benefit by filtering.

Robert Andrew Brown said...

Thanks Stephan more fascinating material.

Thanks Stan some really useful links.

Laura in Arizona said...

Thank you so much for this blog and the time and effort you put into it. I have learned so much. Your words and wisdom are changing peoples lives.

I am sorry if this is not the best place but I have a question about anti-nutrients in raw, leafy greens. I consume about 3-4 cups in a breakfast shake along with an egg, cod liver oil, coconut milk and berries. I wouldn't be so concerned but I have gotten in the habit of having this almost every day.

I know you research a lot of things and are interested in things like anti-nutrients and healthy eating. You have a wonderful way of distilling the basic truths out and presenting them in a thoughtful way.

Is this too much? should I cook them first? or should I not worry about it.

Thanks so much, Laura.

Stephan Guyenet said...


Thanks for the encouragement. My opinion is that greens that have been bred to be eaten raw can be eaten raw, and greens that have been bred for cooking should be cooked. Lettuce, baby spinach, mache etc. are fine raw, although spinach does contain oxalic acid which can interfere with mineral absorption. Probably not a problem for the average person, but large quantities every day might be overdoing it. Kale, chard and other greens of that nature are probably best eaten cooked in my opinion.

I'm in favor of eating greens and I personally love romaine or butter lettuce with a fresh vinaigrette!

gallier2 said...


the separation of the cream and the milk is nothing specific to the micro-filtered milk. Standard milk process does also handle both separately, the cream is added back when botteling to give the different classes, skimmed, semi-skimmed and full. You will notice that what is sold as full milk has 3.5% fat when cows milk is about 4%.

As for real raw milk, it's not possible for everyone to get some, the regulation are quite severe and unless one knows directly a farmer, there is no possibility to get some.

Anna said...

I am fortunate to live in CA, where currently few raw milk can be sold in stores (though there are constant legal and regulatory challenges to raw dairy lately). There are two farm/dairies that serve many stores throughout the state, though only one of them delivers as far south as my location (otherwise one has to find someone with a cowshare operation). I have visited/toured the farm/dairy that our milk comes from (it's hard to get information about the milk production from most dairies, let alone tour the place and see the operation first hand). The cows are milked in the field in a custom-made mobile milking parlor (pulled to the cows by a tractor), the milk is immediately chilled as well as filtered once.

In the store, next to the glass bottles of the only other non-homogenized (but pasteurized) milk available locally, I can see the raw milk has far more cream content, so it has not been standardized to only 3.5% milkfat. The dairy owner told me it is as high as 8% sometimes and I can believe it based on the cream layer variations throughout the year.

It's incredibly delicious! I do buy some pasteurized dairy products (not UHT though) if they don't have added nonfat milk solids, gums, or preservatives (harder and harder to find), but for milk, it's strictly raw for us.

Laura in Arizona said...

Thanks so much for your input Stephan. That makes a lot of sense. I think I will start cooking the kale/collards and eating as a side dish and cut the spinach down. This all started as a way to increase the the dark greens in my diet. At first I was happy but since then I have been concerned that it is too much too often. I love baby salad greens with a vinaigrette too!

Anonymous said...

Regarding the formation of Didydra-vitamin K1 during the hydrogenation process:

The bleaching, refining, and deodorization process that all processed vegetable oils undergo creates a significant amount of trans-fats. Does this process also result in the formation of dihydra-vitamin K?

Deodorization is not a chemical process, but it involves high pressure and very high tempurature for a significant period of time, and results in dramatic changes in the composition of the oil.

This study seems to have some detail on this question, but I don't really want to pay the $30! Does anyone have access?

Conversion of Vitamin K1 to 2‘,3‘-Dihydrovitamin K1 during the Hydrogenation of Vegetable Oils

Stephan Guyenet said...

U.S. Food Trends,

I took a look at the paper. There is no dihydro-K1 in unhydrogenated soybean oil. As soon as it's lightly hydrogenated, it's 50-50 K1/dihydro-K1. I believe most commercial canola oil is lightly hydrogenated, which could potentially make it a source of dihydro-K1. Soybean oil may commonly be lightly hydrogenated as well, I'm not totally sure. They tend to hydrogenate the oils that contain omega-3 like canola and soybean, because it stabilizes them.

stevea said...

Stephan, I applaud your blog, as one of the very few taking a rational approach and pointing so serious references, not the pseudo-science and faux-science websites.

I do take exception on one matter. You entitle the link to Booth, Peterson et al as "Inhibiting vitamin K2 metabolism", but the paper does not test for or show inhibition. Briefly, rats fed dihydro-K1(dK1) produce less MK-4 end product. Very roughly 20-70% as much MK-4 appeared in various tissues of the dK1 fed rats. They suggest higher urine clearance rate for dK1 is the cause. dK1 might be colloquially said to be "less nutritious", but not inhibitory.

To test for inhibition one would need to study the relative utilization of the two forms. Most likely K1 is preferentially used if urine clearance times for dK1 is the issue, but still it's possible that dK1 is preferentially used in MK-4 metabolism and inhibits K1 use; this paper provides no evidence on the matter.

Unrelated to the above, this is a nice reference, indiicating in references that all dK1 is from hydrogenation of K1 in oils.

Sadly the common dietary hydrogenated oils sources (soy, canola and cottonseed) are all rich in K1. The paper has a very limited list of prepared foods analyzed for K1 and dK1 the surprising fact is that significant dK1 appears in foods like frozen pizza, dried onion soup and canned spaghetti (margarine and chicken nuggets seems obvious sources). It seems it will require a carefully reading of ingredient lists to avoid hydrogenated oil products containing dK1.

Keep up the great work !

Stephan Guyenet said...


Yes, I agree that the paper does not demonstrate directly that dihydro-K1 inhibits K2 metabolism. What it does is suggest the possibility that it could inhibit K2 production. This needs to be tested directly, by feeding rodents K1 and dihydro-K1 together. But it's not hard to imagine that a form of K1 that can't be converted to K2 might compete with K1 for absorption through the digestive tract, into tissues, or for enzymes, thereby reducing overall K2 production.

I didn't mean to imply that the study had proven that dihydro-K1 inhibits K2 metabolism, simply that the possibility is there and it should be a concern.

Unknown said...

"I wish I knew the truth, because one of the most convenient ways to get the right vitamin K is by drinking whole milk from pastured cows. Unfortunately, most of what's available is homogenized and pasteurized, thanks to our dairy laws."

Actually, I believe the best source of K is butter from raw cream. Vitamin K is mostly in the cream and therefore highly concentrated in butter.
There are some people who believe that the homogenization of milk is worse than the pasteurizing, due to the process of forcing the fat at high speeds through fine mesh screens in order to keep it from ever separating. I suggest reading "The Untold Story of Milk" if you want to know more about how milk is processed.
Personally, I think pasteurization is worse. The proteins in milk are pretty delicate, which is one of the reasons pasteurization was pushed- you can transport cooked milk farther than raw milk- and the intense heating process, especially nowadays, can denature the proteins and make them basically unrecognizable to our bodies. Enter milk allergies.
All I can say is, I'm really glad to have access to a good raw dairy!

Unknown said...

Very interesting post Stephan, I learn a lot from your blog.

I'd like to add a Swedish perspective on canola oil or rapeseed oil. On the bottle we have at home (I didn't buy it) it says:

"Zeta rapeseed oil is refined, which gives a number advantages. The oil tolerates heat better and stays fresh for a longer time. [...] The cleaning process does not alter the composition of fatty acids in the oil. Therefore, this oil is just as healthy as an unrefined (cold pressed) oil. [...]Zeta rapeseed oil is healthy beause of its high content of monounsaturated fat and omega-3 fatty acids."

My comments:

1. Are they lying?

2. Should I worry about the fact that they use the word "refined" in one place and "cleaning process" in another place?

I see no reason to use rapeseed oil (since there is butter, olive oil and coconut oil) for cooking at home, but rapeseed oil (unhydrogenated, but still) is all over the place in the processed foods I still do allow, like fish fingers and mayo, the kind of food that my children love to eat.

Anna said...

Anna Delin,

Have you considered making fish fingers and mayo at home?

You could make a big batch of fish fingers (search for homemade fish sticks and you'll find lots of recipe ideas online). Freeze them in a single layer on a flat pan, then store in a zip bag in the freezer for convenience. I do this with meatballs and I've done it with chicken "nuggets" & fiskeboller in the past.

For what it's worth, when I started making big reductions in the amount of industrially processed food for my family, my son was 5 yo and not very interested in making the change with us. ;-)

I'm not proud of this, but like most "industrial" kids these days, my son was becoming too accustomed to food that was easy to open, heat up, and serve. I rationalized "fast food at home" in various ways, but I wanted to change that. He's now 11 yo and his interest in eating a wider variety of "adult" foods has gradually increased, but it didn't happen until I found ways to reduce the amount of packaged "juvenile" food that was available to him throughout the day (including the awful heat & serve school lunches). At times it was not easy to get him to try the "real foods" (and I'm not one to push detested food on an unwilling kid - at the same time, I don't want to be a short-order cook). One strategy was to periodically "run out" of some juvenile favorite (he was more resistant to what I preferred to serve him if he knew his favorite was in the house). Another was to create homemade versions of "juvenile" or highly processed favorites (homemade catsup, homemade tartar sauce, homemade ice cream, meatballs, baked custard, home-flavored yogurt, etc.).

Eventually he grew to understood more about what I was trying to keep out of our family's diet, and he's been more cooperative and less resistant (mostly). Now my major challenge is interesting him in helping me in the kitchen so he learns what goes into food prep.

The transition does require planning ahead, though, and creating strategies for those times when it's tempting to fall back on the packaged and convenience items. Mayo is a quick and easy one, though - I make one or two small batches of mayonnaise weekly and store them in an jar in the fridge - I can now do it in my sleep - it only takes about 3 minutes or less with a handheld "stick" blender to make a cup of really good mayo, which I use as a base for a wide variety of salad dressings and sauces.

I also found I saved money that I could use for buying higher quality foods, though it didn't always seem like it at first. But buying a good lunch box system (laptop lunch boxes) and a good quality yogurt by the quart and flavoring it at home with seasonal fruit, a teaspoon of maple syrup and some vanilla and serving only the amount a child will eat instead of popping the foil lid off an individual pre-sweetened container that might only be half-eaten, - can be quite cost-effective in the long run (just one example).