Vitamin K2 and Arterial Calcification: The MGP Mechanism, Rotterdam Cohort, and the JAMA 2026 RCT

Why do arteries calcify?

Calcium in soft tissue is a sign of dysregulated mineral metabolism — not merely "lime deposits." Vascular smooth muscle cells synthesize matrix Gla-protein (MGP) — a potent local inhibitor of mineralization. For MGP to bind calcium ions and hydroxyapatite crystals, five of its glutamic acid residues must be converted to gamma-carboxyglutamic (Gla) acids. This step is catalyzed by gamma-glutamyl carboxylase, which requires vitamin K2 to function.

When K2 is insufficient, MGP remains uncarboxylated (inactive). Its inactive form — dephosphorylated uncarboxylated MGP (dp-ucMGP) — circulates in the blood and is now used as a biomarker of K2 deficiency. The higher the dp-ucMGP, the lower the proportion of active MGP in the arterial wall and the more freely calcium is deposited.

What did the Rotterdam Study show?

Geleijnse and colleagues (J Nutr, 2004) analyzed data from 4807 men and women over 55 from the Rotterdam population cohort with no history of myocardial infarction — followed for 7 to 10 years. Participants were divided into three groups by menaquinone (K2) intake. The highest tertile consumed on average more than 32.7 mcg K2/day versus less than 21.6 mcg in the lowest tertile:

• CHD mortality: RR = 0.43 (95% CI: 0.24–0.77) — a 57% reduction
• All-cause mortality: RR = 0.74 (95% CI: 0.59–0.92) — a 26% reduction
• Severe aortic calcification: OR = 0.48 (95% CI: 0.32–0.71) — a 52% reduction

A key nuance: K1 intake (phylloquinone, from green vegetables) showed no association with any of these outcomes. Only K2 demonstrated an effect. These are observational data: the group with the highest K2 consumed primarily cheeses and other dairy products, so residual confounding cannot be excluded.

Does MK-7 supplementation reduce arterial stiffness?

Knapen and colleagues (Thromb Haemost, 2015) conducted a randomized double-blind placebo-controlled trial with 244 healthy postmenopausal women. Intervention: MK-7 (menaquinone-7) at 180 mcg/day for three years.

Results: carotid-femoral pulse wave velocity (cfPWV) and the Stiffness Index beta were significantly reduced in the MK-7 group compared to placebo. Dp-ucMGP declined by approximately 50% — confirming the biological activity of the supplement. The most pronounced effect was observed in women with a baseline Stiffness Index beta above the median (10.8) — that is, those with initially higher-than-normal arterial stiffness.

What did the VitaK-CAC trial in JAMA 2026 show?

Vossen and colleagues published results in JAMA Cardiology (June 2026) from a randomized double-blind trial in 180 adults with verified coronary artery disease and coronary artery calcification (50–400 Agatston units). Supplement: MK-7 360 mcg/day, duration — 2 years.

Coronary artery calcification progression was significantly slower in the K2 group (p=0.02). Annual calcium accrual declined by approximately 19 Agatston units. Stenosis progression was observed in 33% of K2 group participants versus 41% in the placebo group. Plasma MK-7 levels confirmed compliance. Limitations: small sample size, single country, clinical significance of Agatston index changes remains unclear.

For comparison: the Diederichsen et al. trial (Circulation, 2022) in elderly men with already-established severe aortic valve calcification showed a different result — despite reliable reductions in dp-ucMGP (biomarker effect confirmed), valvular calcification progression was not slowed. This suggests K2 is more effective at early stages rather than capable of reversing an already-established process.

How does K2 differ from K1?

Phylloquinone (K1) has a half-life of approximately 1.5 hours and is predominantly retained by the liver — it serves coagulation factors. Menaquinone-7 (MK-7) remains in the blood for more than 72 hours and redistributes to peripheral tissues: arteries, bones, cartilage — where MGP and osteocalcin are located. It is precisely this tissue bioavailability that allows MK-7 to demonstrate effects in studies that K1 does not produce.

The primary source of MK-7 is natto (fermented soybeans), one of the few foods with a concentration of approximately 1000–1280 mcg per 100 g. Outside Japan, dietary MK-7 intake is extremely low: the "high" tertile in the Rotterdam cohort consumed only about 33 mcg/day. Most Western diets provide considerably less.