Vitamin K2 (MK-4 and MK-7): what is proven and who should be careful

Tuesday morning, two weeks ago. A woman of about 60 places a box of vitamin D3 with K2 on the counter, bought on a friend's recommendation, and asks: this K2 they put everywhere lately, does it really do anything for the bones or is it the latest fashion. A precise question. The honest answer is: it does, but less than the advertisements claim; for one specific group of women the data are good, and for people on warfarin this vitamin is a downright no-go zone.

Vitamin K is one of those supplements whose questions at the pharmacy come in waves: first silence for years, then suddenly every second vitamin D product includes it. Here I have gathered what its forms are and how they differ, what is actually proven for bones and blood vessels, what the difference between MK-4 and MK-7 is, and which red line must not be crossed.

The EU-approved health claims for vitamin K are two and you will find them on every label: it contributes to normal blood clotting and to the maintenance of normal bones. Behind the two stands one and the same mechanism. Vitamin K is the helper without which a number of important proteins cannot be activated: the process is called carboxylation and acts like the ignition key for them.

Among these proteins are the clotting factors produced in the liver, osteocalcin in the bones, which builds calcium into the bone tissue, and matrix Gla protein in the vessel wall, which prevents calcium from being deposited where it does not belong. One vitamin, three workplaces: liver, bone and vessel wall. This also explains why the interest in it goes far beyond clotting.

Vitamin K1, phylloquinone, comes mainly from green leafy vegetables: spinach, broccoli, cabbage, lettuce. Vitamin K2 is a whole family of compounds, the menaquinones (MK-n): the number shows how many isoprene units there are in the side chain, and hence the designation from MK-4 to MK-13. They are predominantly of bacterial origin and are found in fermented foods: sauerkraut, aged cheeses, and in smaller amounts in eggs, liver and meat. MK-4 is a special case: the body produces it on its own from K1, without bacterial involvement, while the long-chain forms are also synthesised by the bacteria in our gut.

The popular division of K1 for the blood and K2 for the bones is convenient but inaccurate. Both forms activate the same proteins; the difference is in the logistics. K1 stays mainly in the liver and serves clotting, while the long-chain menaquinones travel longer in the blood, attached to lipoproteins, and reach the bones and the vessel wall. It is telling that EFSA sets an adequate daily intake of 70 mcg for phylloquinone only, because the evidence on menaquinones is insufficient for separate norms, and no upper limit has been established for any form of the vitamin.

The two forms you will meet on labels are MK-4 and MK-7. MK-4 breaks down quickly and stays in the blood for a few hours, which requires taking it several times a day, and at the usual doses of 100-200 mcg it is practically undetectable in serum (Sato, 2012). MK-7 comes from fermentation, its natural sources are natto (fermented soybeans) and sauerkraut, it lasts about three days, and one daily intake is enough. The classic 2007 study by Schurgers compared MK-7 with vitamin K1 in healthy volunteers: MK-7 gives more stable levels, accumulates to values 7-8 times higher than K1 with prolonged intake, and activates osteocalcin more completely.

One detail the MK-4 advertisements skip: the Japanese studies with impressive bone results used 45 mg of MK-4 daily, and that is menatetrenone, a prescription medicine in Japan. The figure is over 200 times higher than the usual 100-200 mcg in supplements, so the results of the medicinal dose cannot be copied onto the capsule of an over-the-counter product at the pharmacy.

The most serious study is the three-year randomised trial by Knapen from 2013: 244 healthy postmenopausal women took 180 mcg of MK-7 daily for three years. The result: slower loss of bone mineral density in the spine and the femoral neck, better bone strength indices, and less loss of vertebral height compared with placebo. It is not common for a supplement to be tested for so long and with measurable results.

The limits must also be stated: the participants were healthy postmenopausal women and the conclusions do not transfer automatically to men and younger people, while the endpoint was bone density, not counted fractures. The EU-approved claim applies to vitamin K as a whole, and the data suggest that the intake makes most sense in postmenopausal women with a low vitamin status.

The interest in its effect on the vessels began with the Rotterdam Study of 2004: among nearly 4,800 people, those with a higher intake of menaquinones from food had less aortic calcification and a lower risk of death from coronary heart disease, while K1 intake showed no such link. It is important to stress: this is an observational study and it shows an association, not cause and effect.

The first interventional study came from the same team in 2015: in the women on MK-7, arterial stiffness decreased and inactive matrix Gla protein fell by 50% compared with placebo, and a one-year study from 2025 confirms that in postmenopausal women the vessels remain more elastic. The picture is promising but not yet final: a K2 supplement replaces neither a cardiologist nor the control of blood pressure and cholesterol.

Here my advice is categorical. Warfarin and similar anticoagulants work precisely by blocking the recycling of vitamin K in the body, and any additional intake of the vitamin works against the medicine. The advice of the British health service is direct: vitamin K supplements are not taken while on warfarin. The Schurgers study shows why: even 50 mcg of MK-7 daily can affect the anticoagulant effect in a clinically relevant way.

For food the rule is different and worth knowing: green leafy vegetables are not removed from the menu but kept at a constant level, because a sharp rise or drop in intake destabilises the INR and the risk of clotting or bleeding. Every change in diet or new supplement for a person on anticoagulants goes through the doctor who monitors their readings. This is not excessive caution but direct safety.

In a healthy person with varied food on the menu, the intake of vitamin K from food usually covers the needs: European estimates for total intake range between 72 and 196 mcg daily in adults. The group for which the supplement has the best evidential footing is postmenopausal women, especially those with a low vitamin status, and the dose in the studies is 180-200 mcg of MK-7 daily.

The D3 with K2 combination is not a mandatory pair, as I explained in detail in the article on vitamin D: D3 does its job on its own, and K2 is a sensible addition when fermented foods are missing from the menu. Sources from food are sauerkraut, aged cheeses, eggs and liver. The absence of an upper limit gives no green light for megadoses: intake stays within the tested doses, and with anticoagulant therapy the doctor decides.

This article was written by Donka Arabadzhova, Master of Pharmacy, and is confirmed by the following sources: Schurgers L.J. et al., Blood, 2007: comparison of synthetic vitamin K1 and natto-derived menaquinone-7; Sato T. et al., Nutrition Journal, 2012: comparison of MK-4 and MK-7 bioavailability; Knapen M.H.J. et al., Osteoporosis International, 2013: three-year MK-7 intake in postmenopausal women; Knapen M.H.J. et al., Thrombosis and Haemostasis, 2015: MK-7 and arterial stiffness; Geleijnse J.M. et al., Journal of Nutrition, 2004: the Rotterdam Study on menaquinones and the heart; one-year randomised study, Nutrients, 2025: MK-7 and vascular stiffness; EFSA, 2017: dietary reference values for vitamin K; NIH Office of Dietary Supplements, Vitamin K Fact Sheet for Health Professionals; NHS: warfarin and vitamin K; Regulation (EU) No 432/2012 (approved health claims for vitamin K).