Vitamin K2: MK-4 vs MK-7 for Bone and Heart Health

The Vitamin K Family

Vitamin K exists as K1 (phylloquinone), abundant in leafy greens and used mainly by the liver for coagulation factors, and K2 (menaquinones), a family distinguished by side chain length and produced by bacteria and found in fermented foods and animal products.

The important K2 forms are MK-4, which mammals can synthesise from K1 in some tissues and which is present in animal foods, and MK-7, produced by bacterial fermentation and abundant in natto. Longer menaquinones such as MK-8 and MK-9 appear in fermented dairy.

The functional distinction that matters is tissue distribution. K1 is cleared rapidly by the liver, so relatively little reaches bone or arterial tissue. MK-7, with its long side chain, binds lipoproteins in circulation, giving it a long half-life and meaningful delivery to tissues outside the liver. This is the mechanistic reason that K2 is discussed for bone and vascular outcomes while K1 is discussed for coagulation.

What K2 Does: Two Proteins, Two Destinations

Vitamin K is the cofactor for gamma-glutamyl carboxylase, the enzyme that converts glutamate residues into gamma-carboxyglutamate, which is what allows a protein to bind calcium. Without adequate vitamin K, these proteins are made but remain uncarboxylated and non-functional.

Osteocalcin is produced by osteoblasts. Carboxylated osteocalcin binds calcium and hydroxyapatite and participates in bone matrix mineralisation. Uncarboxylated osteocalcin circulates and is used as a marker of inadequate vitamin K status.

Matrix Gla protein is produced in vascular smooth muscle and is one of the strongest known inhibitors of vascular calcification. In its carboxylated form it binds calcium and prevents crystal formation in the arterial wall. Circulating dephosphorylated-uncarboxylated matrix Gla protein is a validated marker of vitamin K status in vascular tissue, and elevated concentrations track with arterial calcification measures and with cardiovascular outcomes in observational cohorts.

This is the "calcium traffic" mechanism in one sentence: vitamin K2 does not add calcium or remove it, it determines whether the calcium already present is directed into bone matrix or left available to deposit in vessel walls.

MK-4 Versus MK-7

Note the thousandfold difference in dose. MK-4 evidence comes from doses of 45 milligrams per day, while MK-7 works at 90 to 360 micrograms. Products offering "100 mcg MK-4" are delivering a dose far below anything tested for bone outcomes, with a compound cleared within hours. If the intent is MK-4 at trial doses, that means 15 mg three times a day, which is a different product category entirely.

The Evidence, by Outcome

Arterial elasticity and calcification

A three-year randomised trial in postmenopausal women using 180 mcg per day of MK-7 reported improved arterial stiffness measures and reduced carotid artery elasticity decline relative to placebo. Trials in populations with high calcification burden, including dialysis and aortic valve cohorts, have produced more mixed results, suggesting established calcification is harder to influence than its progression.

Bone density and fracture

The MK-4 evidence at 45 mg per day includes Japanese trials reporting reduced fracture incidence and maintained bone mineral density, which is why it has clinical use in that country. Trials with MK-7 at microgram doses consistently improve biochemical markers, including reducing uncarboxylated osteocalcin, with more modest and less consistent effects on bone mineral density itself. Meta-analyses of vitamin K for fracture prevention are heterogeneous and depend heavily on which trials are included.

Insulin sensitivity and metabolic markers

Osteocalcin has hormonal activity affecting insulin sensitivity, and some trials report modest metabolic improvements with K2. This is an interesting secondary line of evidence rather than an established use.

Status markers

The most reliable finding across the literature is that supplemental K2 reduces circulating uncarboxylated osteocalcin and dephosphorylated-uncarboxylated matrix Gla protein in a dose-dependent way, meaning it demonstrably improves vitamin K functional status. Whether that translates into hard outcomes is where the evidence thins, which is the familiar shape of a biomarker-versus-outcome gap.

Who Is Most Likely to Benefit

Vitamin K1 adequacy for coagulation is achieved by almost anyone eating vegetables. Functional K2 adequacy in bone and vascular tissue is a different and more demanding standard, and several groups are plausibly short of it.

• People supplementing vitamin D and calcium, since increased calcium absorption raises the importance of directing it correctly.
• People with low intake of fermented foods and organ meats, which is most Western diets. Natto is by far the richest MK-7 source and is not widely eaten outside Japan.
• Postmenopausal women, given accelerated bone remodelling.
• People with reduced kidney function, in whom matrix Gla protein status is often poor, though this group requires clinical supervision rather than self-management.
• People with existing arterial calcification on imaging, where the objective is slowing progression.

Safety and the Warfarin Absolute

Vitamin K2 has no established toxicity at supplemental doses. Even the 45 mg per day MK-4 regimen has decades of clinical use without a characteristic adverse profile, and no tolerable upper intake level has been set because no toxicity threshold has been identified.

There is one absolute contraindication and it matters. Warfarin and other vitamin K antagonists work by blocking vitamin K recycling. Supplemental vitamin K directly opposes their mechanism and can make anticoagulation unstable and unsafe. Anyone taking warfarin should not add vitamin K2 without their prescriber managing it, and consistency of vitamin K intake is the relevant principle for that population. Direct oral anticoagulants such as apixaban and rivaroxaban do not act on vitamin K and are not affected in the same way, but the medication list should still be checked rather than assumed.

Practical Protocol and Review Points

The defensible default for most people is MK-7 at 90 to 200 mcg per day, taken with a fat-containing meal since it is fat soluble, once daily given the long half-life. Where the objective is specifically bone density and the person wants the regimen with fracture data behind it, that means MK-4 at 45 mg per day in three divided doses, which is a considerably bigger commitment.

Combination products pairing D3 with K2 are reasonable and convenient, provided the K2 dose is meaningful rather than a token amount added for the label. Twenty micrograms of MK-7 in a vitamin D capsule is marketing rather than dosing.

Review triggers are mostly contextual rather than symptomatic, since vitamin K status is not perceptible. Starting an anticoagulant is a hard stop. Starting or substantially increasing vitamin D and calcium raises the case for K2. Imaging that shows arterial calcification changes the objective from prevention to slowing progression. And a change in diet toward regular natto or fermented dairy reduces the supplemental requirement.