Vitamin K2, ucMGP, and Arterial Calcification Risk

Vitamin K status may matter to vascular health long before a scan shows calcium in an artery. In community-based adults, higher uncarboxylated Matrix Gla Protein (ucMGP)—a marker of functionally inactive MGP—has tracked with arterial stiffness and later heart failure with preserved ejection fraction (HFpEF). But the most demanding test of the idea, giving vitamin K2 to people on dialysis with established calcification, has repeatedly produced a different result: the biomarker improves, while arteries do not.

That split defines the honest case for vitamin K2. The biology and observational data justify interest. Supplementation has not yet proved that it prevents heart attacks, strokes, HFpEF, or cardiovascular death in otherwise healthy adults.

MGP is a calcification brake, not a drain cleaner

Matrix Gla Protein helps inhibit calcium crystal formation in vascular tissue. It needs vitamin K-dependent carboxylation to do that job. When MGP remains uncarboxylated, it is less able to bind calcium and restrain mineral deposition.

This mechanism fits experimental work, but it does not establish what a supplement can accomplish in humans. In a partial MGP-loss mouse model, reduced MGP activity increased aortic pulse-wave velocity and altered arterial structure, including elastin fragmentation and collagen-matrix disruption, before obvious mineralization. The implication is plausible: restoring MGP activity may prevent new damage more readily than it repairs an artery already remodeled and calcified. It remains an interpretation, not a prevention trial result.

Community cohorts found a signal before kidney failure

The strongest reason to take the biomarker seriously comes from the 2022 Framingham Heart Study analysis. Among 7,066 participants in the Offspring and Third Generation cohorts, plasma ucMGP was independently associated with carotid-femoral pulse-wave velocity, a direct measure of arterial stiffness, and with incident HFpEF during follow-up. The analyses adjusted for age, sex, systolic blood pressure, antihypertensive treatment, diabetes, body-mass index, smoking, and lipid measures; associations across the prespecified cardiovascular traits met Bonferroni-corrected significance criteria.

Researchers excluded warfarin users and people with estimated glomerular filtration rates below 30 mL/min/1.73 m². That makes the finding more relevant to adults outside end-stage kidney disease, but it does not make it causal. ucMGP could partly reflect dietary patterns, metabolic health, kidney function within the non-severe range, or another process that also raises cardiovascular risk.

A 2024 MESA cohort study extended the association prospectively in a diverse community sample: higher dp-ucMGP predicted both incident and progressive coronary-artery and aortic calcification over long-term follow-up. Prospective data reduce one concern—that existing calcification itself drove the biomarker level—but confounding still remains. Cohorts can identify a useful warning light; they cannot prove that changing the warning light changes the outcome.

The biomarker also appears in sicker populations. A 2026 heart-failure cohort associated higher dp-ucMGP with worse outcomes, while a study in patients with hypertension linked higher dp-ucMGP with lower axial skeletal muscle mass. These findings broaden the clinical context, but they do not establish that vitamin K deficiency caused either outcome.

Dialysis trials lowered dp-ucMGP without improving vascular endpoints

Randomized trials in hemodialysis patients offer the clearest corrective to supplement marketing. Across four trials, MK-7 lowered dp-ucMGP, showing that participants absorbed the intervention and that it affected the intended biomarker. Yet the trials did not find significant improvement in arterial stiffness or calcification progression.

In the double-blind, placebo-controlled RenaKvit trial, 48 dialysis patients received 360 micrograms of MK-7 daily or placebo for two years. MK-7 significantly reduced dp-ucMGP but did not significantly improve carotid-femoral pulse-wave velocity, coronary artery calcium, or abdominal aortic calcification. The trial was small, so its null vascular findings were imprecise; they cannot settle the question alone.

The larger open-label Trevasc-HDK trial randomized 178 hemodialysis patients with baseline coronary artery calcium scores of at least 30 to MK-7 at 360 micrograms three times weekly or usual care for 18 months. MK-7 reduced dp-ucMGP, but investigators found no significant benefit for the primary coronary-calcification endpoint or secondary pulse-wave velocity and augmentation-index outcomes. Blinded radiologic assessment helps, though the lack of participant blinding can still affect adherence and care behavior.

Two more trials point in the same direction. The multicenter placebo-controlled pilot iPACK-HD trial found no demonstrated benefit on coronary-calcification progression, and a multicenter MK-7 trial in chronic hemodialysis found no significant improvement in arterial stiffness. A pilot cannot exclude a modest effect, but four trials spanning calcification and stiffness endpoints make reversal of advanced dialysis-related vascular disease an unsupported claim.

Why might the trials and cohorts disagree?

Timing is a credible explanation, although it has not been proved. Dialysis patients often have severe, longstanding vitamin K insufficiency, established calcification, and structural damage to the arterial matrix. An intervention that helps prevent new mineral deposition may be unable to dissolve existing calcium or rebuild disrupted elastin. Trevasc-HDK deliberately enrolled people who already had measurable coronary calcification, which makes it a poor test of primary prevention.

That explanation should not become an escape hatch for every null trial. The dialysis evidence shows that a large drop in dp-ucMGP does not guarantee a vascular benefit over 18 months to two years in this population. It does not show whether earlier supplementation works, because an adequately powered trial with clinical cardiovascular endpoints in healthy or early-risk adults has not been completed.

Earlier-stage trials are interesting, not decisive

A double-blind, placebo-controlled trial in vitamin K-deficient kidney-transplant recipients reported improvement in serum calcification propensity and arterial-stiffness markers with vitamin K supplementation. These are surrogate outcomes, not fewer cardiovascular events, and transplant recipients do not stand in for the general public.

The double-blind AVADEC trial reported attenuated aortic-valve calcification progression with MK-7 plus vitamin D3. Because both supplements were given together, the study cannot isolate an MK-7 effect. Its later substudy found no significant improvement in epicardial adipose tissue volume or systemic inflammatory markers, which argues against selling the combination as a broad cardiovascular anti-inflammatory treatment.

A one-year MK-7 trial in postmenopausal women examined vascular stiffness and blood pressure, a more prevention-relevant setting than dialysis. Its results do not yet supply a hard-endpoint answer for the wider population.

The biomarker itself has an unresolved problem

dp-ucMGP is not the only candidate marker of vitamin K status. In a cross-sectional CKD study, Nyvad and colleagues found that PIVKA-II, which reflects hepatic vitamin K status, was associated with aortic calcification, whereas dp-ucMGP was not. Different vitamin K pools may matter in different disease settings. At minimum, this result weakens the assumption that lowering dp-ucMGP alone proves that the calcification-relevant pathway has been corrected.

What to do with this evidence

MK-7 has a biologically plausible role in MGP carboxylation, and 360 micrograms per day produced substantial dp-ucMGP suppression in dialysis research. It has not reversed established vascular calcification in dialysis trials. For healthy adults, the rationale is prevention based on biomarker associations and mechanism—not demonstrated reductions in cardiovascular events.

Anyone taking warfarin or another vitamin K antagonist should not start vitamin K supplements without the clinician managing anticoagulation. For everyone else, the evidence supports a measured conclusion: vitamin K2 may be worth studying as an early preventive intervention, but it has not earned a claim that it protects healthy arteries or reverses calcium already there.