Vitamin K2 · Menaquinone
Evidence Fact Sheet
MK-4 · MK-7
Vitamin K2 (menaquinone; MK-7, MK-4) is a fat-soluble vitamin that acts as a cofactor for γ-glutamyl carboxylase, activating Gla-proteins such as osteocalcin and matrix Gla protein. Human evidence spans bone-density meta-analyses, arterial-stiffness and calcification RCTs, and honest null findings. Typical research dose 90-200 μg/day MK-7. EFSA/ANVISA authorized bone and coagulation claims.
Also known as: Menaquinone-7 · MK-7 · MK-4 · Menatetrenone
Overview
Vitamin K2 (menaquinone) denotes a family of fat-soluble vitamers, principally MK-7 and MK-4 (menatetrenone). Its mechanism is to serve as a cofactor for γ-glutamyl carboxylase, the enzyme that γ-carboxylates vitamin K-dependent Gla-proteins — most notably osteocalcin (which incorporates calcium into bone matrix) and matrix Gla protein (which inhibits vascular calcification) — and it also supports normal blood coagulation when vitamin K1 intake is low. It is studied across musculoskeletal and cardiovascular systems. Typical supplemental research doses are 90-200 μg/day of MK-7, with some bone trials using higher menaquinone doses. Regulatory status: FDA GRAS / DSHEA dietary supplement with a cleared NDI for MK-7; EFSA authorized claims under Reg 432/2012 for maintenance of normal bone and normal blood clotting (subject to the "source of vitamin K" threshold), NRV 75 μg; ANVISA functional coagulation claim under IN 28/2018; and in China permitted as a food fortifier and health-food raw material. It is contraindicated with vitamin K antagonist (warfarin) therapy.
Mechanism of Action
Cofactor for γ-glutamyl carboxylase activating Gla-proteins · Carboxylates osteocalcin (bone matrix incorporation of calcium) · Carboxylates matrix Gla protein (inhibits vascular calcification)
Body systems: MUSCULOSKELETAL · Cardiovascular
Evidence-Based Benefits
Each benefit below is anchored to a specific PubMed-indexed study. Effect sizes, sample sizes, and p-values are reported as published; no values are inferred. Honest negatives and null results are kept alongside the positive findings, and disease-research populations are described as such — Vitamin K2 · Menaquinone is not characterized as a treatment for any disease.
Bone Mineral Density (Postmenopausal Osteoporosis)
Meta-analysis supported- 16 RCTs6,425 subjects pooled
- P = 0.006lumbar spine BMD · 10 studies
- RR = 0.43, P = 0.01fracture (after excluding 1 study)
A meta-analysis of randomized trials in postmenopausal osteoporosis found vitamin K2 significantly improved lumbar spine bone mineral density and reduced undercarboxylated osteocalcin. Across six fracture trials there was no overall difference, but excluding one heterogeneous study yielded a significant fracture reduction — illustrating how fragile the fracture signal is. Adverse reactions did not differ from control.
Reported effect: 16 RCTs / 6,425 subjects; lumbar spine BMD improved (P=0.006, 10 studies); fracture RR=0.96 (P=0.65) overall, RR=0.43 (P=0.01) after excluding one heterogeneous study
“Sixteen RCTs with a total of 6,425 subjects were included in this meta-analysis. The overall effect test of 10 studies showed a significant improvement in lumbar spine BMD (BMD LS) (P = 0.006) with VK2. The overall effect test of the six RCTs showed no significant difference in fracture incidence between the two groups (RR=0.96, P=0.65). However, after excluding one heterogeneous study, the overall effect test showed a significant reduction in fracture incidence with VK2 (RR = 0.43, P = 0.01).”
Source: PMID 36033779 · Ma 2022 · Front Public Health
Bone Mineral Density (Long-Term Meta-Analysis)
Meta-analysis supported- 9 RCTs6,853 participants
- WMD 2.17lumbar BMD %change · 95% CI 1.59-2.76
A separate long-term meta-analysis in postmenopausal women with osteoporosis associated vitamin K2 with a significantly greater percentage change in lumbar bone mineral density versus control. This corroborates the lumbar-spine BMD direction seen in the larger meta-analysis, anchored to a pooled weighted mean difference.
Reported effect: 9 RCTs / 6,853 participants; lumbar BMD %change WMD 2.17 (95% CI 1.59-2.76)
“Nine RCTs with 6853 participants met the inclusion criteria. Vitamin K2 was associated with a significantly increased percentage change of lumbar BMD (WMD 2.17, 95% CI [1.59-2.76])”
Source: PMID 35711002 · Zhou 2022 · J Bone Miner Metab
Bone Loss in Healthy Postmenopausal Women (RCT)
RCT supported- n = 2443-year double-blind
- 180 μg/dayMK-7 dose
In a 3-year double-blind RCT, low-dose MK-7 improved vitamin K status and slowed the age-related decline in bone mineral content and density at the lumbar spine and femoral neck, but not at the total hip. The abstract reports the direction of effect without a single headline effect-size number for BMD.
Effect size: this study reports the direction of the finding but does not state a specific numeric effect size, so none is given here rather than estimated.
“Healthy postmenopausal women (n = 244) received for 3 years placebo or MK-7 (180 μg MK-7/day) ... MK-7 intake significantly improved vitamin K status and decreased the age-related decline in BMC and BMD at the lumbar spine and femoral neck, but not at the total hip.”
Source: PMID 23525894 · Knapen 2013 · Osteoporos Int
Bone Density & Microarchitecture in Osteopenia — Honest Negative (RCT)
Null / no benefit RCT supported- p > 0.09aBMD vs placebo · 3 yr
- −65.2 ± 23.5%ucOC fell (p<0.01) yet BMD null
A 3-year RCT in 142 postmenopausal women with osteopenia is the key honest negative for bone: high-dose MK-7 (375 μg/day) dramatically lowered undercarboxylated osteocalcin yet produced no difference in bone mineral density or microarchitecture versus placebo. Improving the biomarker did not translate into a structural bone benefit in this population.
Reported effect: n=142; ucOC −65.2 ± 23.5% (MK-7) vs −0.03 ± 38.5% (placebo), p<0.01; after 3 years aBMD decreased at all sites without differences between groups (p>0.09); microstructure changes similar
“Undercarboxylated osteocalcin decreased in the MK-7-group (- 65.2 ± 23.5%) compared with the placebo group (- 0.03 ± 38.5%), p < 0.01 after 1 year. After 3 years, aBMD decreased at all sites without differences between the MK-7 and placebo-treated women (p > 0.09). Changes in microstructure over 3 years were similar between the two groups, as assessed by both HRpQCT and DXA trabecular bone score.”
Source: PMID 33030563 · Rønn 2021 · Osteoporos Int
Arterial Stiffness & Vascular Gla-Protein (RCT)
RCT supported- n = 2443-year RCT · 180 μg/day
- 50 %dp-ucMGP fall vs placebo
In a 3-year double-blind RCT of healthy postmenopausal women, MK-7 significantly decreased carotid-femoral pulse wave velocity and the Stiffness Index β in the total group, with the largest improvements in women whose baseline stiffness was above the median. MK-7 also cut dp-ucMGP — the inactive form of vascular matrix Gla protein — by 50% versus placebo.
Reported effect: n=244 (180 µg MK-7/day, 3 years); cfPWV and Stiffness Index β significantly decreased; MK-7 decreased dp-ucMGP by 50% vs placebo
“After three year MK-7 supplementation cfPWV and the Stiffness Index β significantly decreased in the total group ... MK-7 decreased dp-ucMGP by 50 % compared to placebo”
Source: PMID 25694037 · Knapen 2015 · Thromb Haemost
Vascular Stiffness & Blood Pressure (Recent RCT)
RCT supported- p = 0.035stiffness: +9.4% MK-7 vs +49.1% placebo
- −3.0% ± 9.0 (p=0.007)brachial BP · high-stiffness subgroup
A 2025 one-year RCT in 165 post-menopausal women with low vitamin K status found MK-7 significantly attenuated the rise in vascular stiffness (β-stiffness index) versus placebo, and in the high-stiffness subgroup lowered brachial blood pressure and increased arterial distensibility. This replicates the arterial-stiffness direction of the earlier Knapen trial.
Reported effect: n=165 (180 µg MK-7/day, 1 year); stiffness placebo +49.1% ± 77.4 vs MK-7 +9.4% ± 67.1 (p=0.035); brachial BP −3.0% ± 9.0 (p=0.007) and distensibility +13.3% ± 32.3 (p=0.040) in high-stiffness subgroup
“MK-7 treatment significantly attenuated vascular stiffness in post-menopausal women (placebo +49.1% ± 77.4; MK-7 +9.4% ± 67.1; p = 0.035) ... decreased blood pressure at brachialis (-3.0% ± 9.0; p = 0.007) ... increased distensibility coefficient (+13.3% ± 32.3; p = 0.040)”
Source: PMID 40077685 · de Vries 2025 · Nutrients
Cardiovascular Events & Coronary Heart Disease (Meta-Analysis)
Meta-analysis supported- 21 articles222,592 participants
- HR 0.70menaquinone & CHD · 95% CI 0.53-0.93
A meta-analysis of observational cohorts found higher dietary menaquinone (vitamin K2) intake associated with lower total coronary heart disease, while phylloquinone (K1) showed only a marginal association and neither was significantly tied to all-cause mortality. The K2-CHD estimate rests on only two studies, so this is an association from epidemiology, not proof of a supplement effect.
Reported effect: 21 articles / 222,592 participants; menaquinone & total CHD pooled HR 0.70 (95% CI 0.53, 0.93; I²=32.1%, two studies); phylloquinone & CHD HR 0.92 (95% CI 0.84, 0.99); no significant association with all-cause mortality
“Twenty-one articles were included with 222,592 participants ... menaquinone and total CHD (0.70; 95% CI 0.53, 0.93; I² = 32.1%; two studies) ... No significant association was observed between dietary vitamin K and all-cause mortality”
Source: PMID 31119401 · Chen 2019 · Eur J Nutr
Aortic Valve Calcification — Honest Negative (RCT)
Null / no benefit RCT supported- 365 men24-month RCT · 720 μg MK-7 + D
- P = 0.64AVC progression difference: 17 AU
A Circulation RCT randomized 365 elderly men with an aortic valve calcification score above 300 AU to 720 μg MK-7 plus vitamin D or placebo for 24 months. The progression of aortic valve calcification did not differ between groups — a clear null for the primary calcification endpoint despite high-dose K2.
Reported effect: 365 men, 720 µg MK-7 + 25 µg vitamin D vs placebo, 24 months; AVC increase 275 AU (95% CI 225-326) vs 292 AU (95% CI 246-338), difference 17 AU (95% CI −86 to 53; P=0.64)
“In elderly men with an AVC score >300 AU, 2 years MK-7 plus vitamin D supplementation did not influence AVC progression.”
Source: PMID 35465686 · Diederichsen 2022 · Circulation
Arterial Stiffness in Chronic Hemodialysis — Honest Negative (RCT)
Null / no benefit RCT supported- n = 96375 mcg MK-7 · 24 weeks
- p = 0.24cPWV change vs standard care (null)
A multicenter RCT of MK-7 (375 mcg/day) in 96 hemodialysis patients with arterial stiffness found no significant difference in carotid-femoral pulse wave velocity versus standard care over 24 weeks. A pre-specified diabetic subgroup did show a significant stiffness reduction, but the overall primary outcome was null — consistent with other CKD vitamin K trials.
Reported effect: n=96 (MK-7 375 mcg/day, 24 weeks); cPWV change −6.0% (−20.2, 2.3) MK-7 vs −6.8% (−19.0, 7.3) control, p=0.24 (overall null); diabetes subgroup −10.0% (−15.9, −0.8) vs 3.8% (−5.8, 11.6), p=0.008
“no significant difference in the change in cPWV at 24 weeks between the MK-7 group and standard care [-6.0% (-20.2, 2.3) vs. -6.8% (-19.0, 7.3), p = 0.24] ... significantly decreased cPWV in patients with diabetes [-10.0% (-15.9, -0.8) vs. 3.8% (-5.8, 11.6), p = 0.008]”
Source: PMID 37299386 · Naiyarakseree 2023 · Nutrients
Insulin Sensitivity & Adiponectin — Honest Negative (RCT)
Null / no benefit RCT supported- n = 148375 µg MK-7 · 12 months
- p = 0.03adiponectin +6.1% vs −0.7% (HOMA-IR null)
In a 12-month RCT, MK-7 (375 µg/day) raised plasma adiponectin versus placebo but did not change HOMA-IR or leptin, indicating no effect on insulin sensitivity in healthy postmenopausal women. The authors concluded that lowering undercarboxylated osteocalcin does not improve insulin sensitivity in this population.
Reported effect: n=148 (MK-7 375 µg/day, 12 months); adiponectin +6.1% ± 20.1% (MK-7) vs −0.7% ± 15.5% (placebo), p=0.03; HOMA-IR and leptin did not change
“P-adiponectin increased in the MK-7 group (6.1 ± 20.1%) (mean ± SD) compared to the placebo group (-0.7 ± 15.5%) after 12 months (p = 0.03). ... HOMA-IR and p-leptin did not change in the two groups. ... did not change insulin sensitivity”
Source: PMID 33664429 · Rønn 2021 · Eur J Clin Nutr
Systemic Inflammation in Type 2 Diabetes — Honest Negative (RCT)
Null / no benefit RCT supported- n = 45 completers200 mcg/day · 12 weeks
- PIVKA-II p<0.05inflammation between-group: NS
A 12-week RCT of MK-7 (200 mcg/day) in type 2 diabetes patients found within-group falls in hsCRP, IL-6 and TNF-α, but the between-group differences in inflammatory markers, dp-ucMGP and body composition were not statistically significant — only PIVKA-II differed significantly between groups. This is an honest null for the anti-inflammatory claim.
Reported effect: 60 randomized / 45 completers (MK-7 200 mcg/day, 12 weeks); only PIVKA-II significant between groups (p<0.05); between-group differences in hsCRP, IL-6, TNF-α and body composition not statistically significant
“serum levels of the inflammatory markers (hsCRP, IL-6, and TNF-α) were significantly lower in the MK-7 group (p < 0.05) ... between group differences in the inflammatory markers were not statistically significant ... between group differences were only significant for PIVKAII (p < 0.05)”
Source: PMID 35365594 · Karamzad 2022 · Nutr Diabetes
Dosage (research context · not a recommendation)
90 μg/day NRV (men) · 90 μg/day (women); typical MK-7 supplement dose 90-200 μg/day. Geleijnse 2004 PMID 15514282 J Nutr Rotterdam Study cohort (n = 4,807) reported the highest tertile of dietary menaquinone intake associated with lower CHD mortality (RR 0.43, 95% CI 0.24-0.77) and severe aortic calcification (OR 0.48). Rønn 2021 PMID 33030563 Osteoporos Int 3-year RCT (n = 142 postmenopausal women with osteopenia, 375 μg MK-7/day + D3 + Ca) found osteocalcin carboxylation improved but bone mineral density and microarchitecture did not differ vs. placebo — honest mixed signal
Regulatory Status · 4 Markets
- US · FDA
- GRAS · DSHEA dietary supplement · NDI for MK-7 cleared
- EU · EFSA
- Authorized claims Reg 432/2012 bone maintenance + normal coagulation (vitamin K source threshold required); NRV 75 μg
- CN · China
- Permitted as a food nutrition fortifier under GB 14880-2012 and as a health-food raw material (vitamin K is in the Health-Food Raw-Material Catalogue). Health-food approvals exist mainly as combinations (calcium + D3 + K2); standalone K2 registrations are limited. SAMR recognized.
- BR · ANVISA
- RDC 243/2018 dietary supplement · IN 28/2018 Anexo V alegação funcional verbatim: "A vitamina K auxilia na coagulação do sangue." Bone claim not separately authorized — vitamina K within bone-health stack must rely on Anexo V coagulação claim only
Authorized Claims
EFSA — “Vitamin K contributes to the maintenance of normal bones.” (Reg 432/2012)
EFSA — “Vitamin K contributes to normal blood clotting.” (Reg 432/2012)
ANVISA — “A vitamina K auxilia na coagulação do sangue.” (IN 28/2018 Anexo V alegação funcional)
Safety
Contraindicated in patients on vitamin K antagonist therapy (warfarin); inform clinician before starting MK-7 if on anticoagulants; otherwise excellent safety profile within typical supplement doses
Related
Goals: joint-bone · heart-health · menopause-support · longevity-stack
Lifestyles: senior-60-plus · menopause
References
PubMed-indexed citations anchoring the benefit findings above. Effect sizes are reported as published.
- PMID 36033779 · Ma 2022 · Front Public Health — Bone Mineral Density (Postmenopausal Osteoporosis)
- PMID 35711002 · Zhou 2022 · J Bone Miner Metab — Bone Mineral Density (Long-Term Meta-Analysis)
- PMID 23525894 · Knapen 2013 · Osteoporos Int — Bone Loss in Healthy Postmenopausal Women (RCT)
- PMID 33030563 · Rønn 2021 · Osteoporos Int — Bone Density & Microarchitecture in Osteopenia — Honest Negative (RCT)
- PMID 25694037 · Knapen 2015 · Thromb Haemost — Arterial Stiffness & Vascular Gla-Protein (RCT)
- PMID 40077685 · de Vries 2025 · Nutrients — Vascular Stiffness & Blood Pressure (Recent RCT)
- PMID 31119401 · Chen 2019 · Eur J Nutr — Cardiovascular Events & Coronary Heart Disease (Meta-Analysis)
- PMID 35465686 · Diederichsen 2022 · Circulation — Aortic Valve Calcification — Honest Negative (RCT)
- PMID 37299386 · Naiyarakseree 2023 · Nutrients — Arterial Stiffness in Chronic Hemodialysis — Honest Negative (RCT)
- PMID 33664429 · Rønn 2021 · Eur J Clin Nutr — Insulin Sensitivity & Adiponectin — Honest Negative (RCT)
- PMID 35365594 · Karamzad 2022 · Nutr Diabetes — Systemic Inflammation in Type 2 Diabetes — Honest Negative (RCT)
Frequently Asked Questions
1. What is vitamin K2 and how does it differ from vitamin K1?
Vitamin K2 is the menaquinone family of fat-soluble vitamins, principally MK-7 and MK-4 (menatetrenone), whereas vitamin K1 is phylloquinone. Both act as cofactors for γ-glutamyl carboxylase, the enzyme that activates Gla-proteins such as osteocalcin (bone) and matrix Gla protein (blood vessels). K2 forms like MK-7 have longer circulation than K1, and in the Chen 2019 meta-analysis dietary menaquinone — not phylloquinone — carried the stronger association with lower coronary heart disease.
2. What dose of vitamin K2 is used in research?
Supplement studies commonly use 90-200 μg/day of MK-7; the EFSA NRV is 75 μg. Some trials went higher — for example the postmenopausal arterial-stiffness and bone-loss RCTs used 180 μg/day MK-7, while several osteopenia and diabetes trials used 375 μg/day or 200 mcg/day. This is an evidence-reporting page, not dosing guidance — the doses listed are simply the amounts the cited studies tested.
3. Can vitamin K2 reverse arterial or aortic valve calcification?
Here the honest negatives matter most. The Diederichsen 2022 Circulation RCT in 365 men found 720 μg MK-7 plus vitamin D did not slow aortic valve calcification progression (difference 17 AU, P=0.64), and MK-7 did not improve arterial stiffness overall in hemodialysis patients (Naiyarakseree 2023, p=0.24). What K2 does reliably do is reduce dp-ucMGP — the inactive vascular form of matrix Gla protein — by 50% versus placebo and improve arterial stiffness in healthy postmenopausal women (Knapen 2015; de Vries 2025, p=0.035). Biomarker and stiffness effects exist; reversing established calcification has not been shown.
4. Does vitamin K2 help blood sugar or inflammation?
The controlled evidence is largely null. A 12-month RCT (Rønn 2021) raised plasma adiponectin (+6.1% vs −0.7%, p=0.03) but did not change HOMA-IR or leptin — no improvement in insulin sensitivity. A 12-week trial in type 2 diabetes (Karamzad 2022) saw within-group falls in hsCRP, IL-6 and TNF-α, but the between-group differences in inflammatory markers and body composition were not statistically significant. So claims of metabolic or anti-inflammatory benefit are not supported by between-group results.
Last evidence review: 2026-06-04