MK-4 vs MK-7: I Dug Through the Research to Settle the Vitamin K2 Debate

MK-4 vs MK-7: The Short Answer (For People Who Scroll)

I’ve been asked about mk-4 vs mk-7 more than almost any other supplement comparison, and I get why. Both are vitamin K2. Both show up on labels making bone and heart claims. Both cost real money. And most articles comparing them are either copy-paste jobs or thinly disguised ads for whichever form the site happens to sell.

So I went through the actual trials. Here’s the short version for people who scroll.

My position, stated up front: MK-7 wins on practicality for most people. It stays in your blood for days, works at microgram doses, and once-daily dosing actually does something. But here’s what surprised me when I first dug into this years ago. MK-4 has the largest bone fracture trials ever run on any form of K2. Not MK-7. MK-4. That complicates the neat little story most supplement brands want to tell you.

The catch (and there’s always a catch) is dose. The MK-4 fracture data comes from 45 mg per day, which is about 500 times what you’ll find in a typical K2 capsule. That gap matters more than almost anything else in this debate.

In this article I’ll walk through the absorption science, the actual human trials on bone and heart health, dosing, safety, and who should pick which form. No hedging, no both-sides-are-great cop-out.

MK-4 vs MK-7: The Short Answer (For People Who Scroll)

What Is Vitamin K2, and Why Are There Two Versions Fighting for Your Money?

The menaquinone family explained without the chemistry lecture

Vitamin K2 isn’t one molecule. It’s a family called menaquinones, running from MK-4 all the way up to MK-13. The number tells you the length of the molecule’s side chain, and that side chain length changes almost everything about how your body handles it: how it’s absorbed, how long it circulates, and which tissues it reaches.

MK-4 has a short side chain. MK-7 has a longer one. That single structural difference drives most of the practical differences you’ll read about below.

So what does K2 actually do? Two jobs, mainly. It activates osteocalcin, a protein that pulls calcium into your bones where you want it. And it activates matrix Gla protein (MGP), which keeps calcium out of your arteries where you very much don’t want it. Same vitamin, two locations, one theme: putting calcium in the right place. That’s the whole pitch, and honestly, it’s a good one.

Without enough K2, these proteins sit around in their inactive, uncarboxylated forms. You can actually measure the inactive versions in blood, which gives researchers a handy marker of K2 status (more on that when we get to the heart data).

Where MK-4 and MK-7 come from

MK-4 is the animal form. You’ll find it in egg yolks, butter, liver, and meat, especially from pastured animals. Your body also converts some vitamin K1 (the leafy green form) into MK-4 on its own, which is a detail I think gets underappreciated. Supplemental MK-4, though, is synthetic. Nothing wrong with that, but it’s worth knowing.

MK-7 is the fermentation form. Its most famous source is natto, the sticky, pungent fermented soybean dish that gives Japanese natto eaters some of the highest K2 intakes on Earth. Supplemental MK-7 is produced by bacterial fermentation, usually with Bacillus subtilis, the same organism that makes natto what it is.

Here’s the analogy I keep coming back to. Think of MK-4 as a sprinter: fast in, fast out, gone within hours. MK-7 is a marathon runner: slower to peak, but still going three days later. Neither is “better” in the abstract. It depends on what race you’re running.

And that brings us to the pharmacokinetics, which is where this comparison stops being theoretical.

What Is Vitamin K2, and Why Are There Two Versions Fighting for Your Money?

The Absorption and Half-Life Battle: This Is Where MK-7 Pulls Ahead

What the pharmacokinetic data shows

The key head-to-head comparison came from Schurgers et al. (2007), published in Blood. They gave healthy volunteers equal doses of MK-4 and MK-7 and tracked serum levels. The results weren’t close. MK-7 produced roughly 6-fold higher serum levels over time and stayed detectable for days. MK-4, at nutritional doses, was barely detectable in serum at all.

Barely detectable. That stopped me the first time I read it.

The half-life numbers explain why. MK-7 hangs around for about 68-72 hours. MK-4 clears in roughly 1-2 hours. Take an MK-7 capsule every morning and your blood levels build to a stable plateau within a couple of weeks, so every tissue gets steady exposure around the clock. Take a single MK-4 capsule and it’s mostly gone before lunch.

Why half-life matters for real people taking real pills

This is the part I want you to actually internalize. Once-daily dosing only works if the compound sticks around. MK-7’s long half-life means the pill you take with breakfast is still working at 3 a.m. MK-4’s short half-life means single daily doses at typical supplement strengths probably leave you uncovered for 20+ hours a day. That’s why the serious MK-4 trials dosed it three times daily at massive amounts.

Now, a counterpoint I don’t see enough articles make. Serum levels aren’t the whole story. Sato’s research on MK-4 tissue distribution suggests MK-4 concentrates in specific tissues, including the brain and glandular tissue, faster than it clears from blood. Low serum doesn’t automatically mean no activity. Your bloodstream is the highway, not the destination, and MK-4 may simply exit the highway quickly. Animal work backs this up: MK-4 shows distinct tissue accumulation patterns that MK-7 doesn’t fully replicate.

Does that rescue MK-4 at supplement doses? I’m skeptical, but it keeps me from dismissing it entirely.

One practical point before we move on, and this applies to both forms. K2 is fat-soluble. Uenishi and colleagues, along with the broader absorption literature, have shown that vitamin K absorbs dramatically better with dietary fat. Take your K2 with a fat-free breakfast and you’re flushing money down the toilet. Pair it with eggs, yogurt, avocado, anything with actual fat in it. This one habit probably matters more than which brand you buy.

So MK-7 dominates the blood-level battle. But blood levels are a surrogate. What happens to actual bones?

The Absorption and Half-Life Battle: This Is Where MK-7 Pulls Ahead

Bone Health: MK-4 Has the Bigger Trials, MK-7 Has the Better Doses

The Japanese MK-4 fracture data

Here’s the fact that surprises almost everyone: MK-4 at 45 mg/day is an approved osteoporosis treatment in Japan. Has been for decades. Not a supplement, an actual prescribed therapy, given as 15 mg three times daily.

The evidence behind it looked strong for a while. A 2006 meta-analysis by Cockayne et al. in Archives of Internal Medicine pooled the Japanese trials and found significant reductions in vertebral fractures (about 60%), hip fractures (77%), and all nonvertebral fractures (81%). Those numbers are enormous. Suspiciously enormous, frankly, and the authors themselves flagged concerns about trial quality.

Then reality intervened. The 2009 Inoue study, a much larger Japanese trial with over 4,000 women, showed considerably weaker results. No significant fracture reduction overall, though a subgroup with existing vertebral fractures did seem to benefit. That’s a familiar pattern in nutrition research: small early trials dazzle, big later trials deflate.

And I keep coming back to the dose. 45 mg is a pharmacological dose, roughly 500 times higher than the 90-100 mcg of MK-4 you’ll find in most supplement bottles. There is essentially no fracture evidence for MK-4 at the doses people actually buy. None. If a brand cites the Japanese fracture trials to sell you a 100 mcg MK-4 capsule, they’re citing evidence for a different intervention entirely.

The MK-7 bone density trials

The landmark MK-7 trial came from Knapen et al. (2013) in Osteoporosis International. 244 healthy postmenopausal women took 180 mcg of MK-7 daily for 3 years. The MK-7 group showed significantly less decline in bone mineral density at the lumbar spine and femoral neck, plus improved bone strength indices. Vertebral height loss slowed too.

Notice the dose: 180 micrograms. Not milligrams. MK-7 produced measurable bone effects at a dose you can actually buy in a normal capsule, taken once a day, because that 72-hour half-life keeps levels steady with minimal effort.

That said, let me temper expectations. Knapen measured bone density and strength indices, not fractures. Fracture prevention is the outcome that actually matters, and no MK-7 trial has been large or long enough to prove it yet. A three-year BMD trial is genuinely good evidence. It’s not fracture proof.

My honest read on the bone evidence

Look, if you judge purely by trial size and hard outcomes, MK-4 wins on paper. Fracture data beats density data.

But that’s the wrong comparison for a supplement buyer. The MK-4 fracture evidence exists only at 45 mg/day, a dose that requires prescription-strength products and three-times-daily dosing, and even that evidence weakened as trials got bigger. The MK-7 evidence, while limited to bone density endpoints, was generated at exactly the dose sitting in supplement bottles right now.

If you’re comparing what’s actually in the capsules you can buy, at the doses printed on the label, the MK-7 evidence is simply more relevant to what you’ll swallow. That’s my read, and I’ll stand by it.

Bone is only half the K2 story, though. The heart data is where things get genuinely interesting.

Bone Health: MK-4 Has the Bigger Trials, MK-7 Has the Better Doses

Heart and Artery Health: MK-7 Is Doing Most of the Heavy Lifting Here

The Rotterdam Study and what it started

Every conversation about K2 and heart health traces back to one paper. Geleijnse et al. (2004) followed 4,807 Dutch adults in the Rotterdam Study and found that people with the highest menaquinone intake had a 57% lower risk of dying from coronary heart disease compared to the lowest intake group. They also had less severe aortic calcification. Vitamin K1 intake, interestingly, showed no such association. Only the menaquinones.

Observational data, yes. It can’t prove causation, and I won’t pretend it does. But a 57% difference in coronary death got researchers’ attention, and it kicked off two decades of K2-heart research that continues today.

MK-7 and arterial stiffness trials

The best interventional evidence so far is Knapen’s 2015 trial in Thrombosis and Haemostasis (same research group as the bone trial, different endpoint). 244 postmenopausal women took 180 mcg of MK-7 daily for 3 years. Arterial stiffness, measured by pulse wave velocity and other markers, improved in the MK-7 group. The effect was strongest in women who started with stiffer arteries, which makes intuitive sense. You can’t improve what isn’t impaired.

The mechanism is refreshingly clean for a nutrition story. Matrix Gla protein is your arteries’ built-in calcification inhibitor, but it only works when K2 activates it. Inactive MGP, measured as dp-ucMGP in blood, is a direct readout of arterial K2 status. Trial after trial shows that MK-7 lowers dp-ucMGP efficiently at practical doses of 90-360 mcg. Imagine MGP as a security guard posted in your artery walls; K2 is what wakes him up, and dp-ucMGP tells you how many guards are asleep on the job.

Now for the honest part. I’ll be straight about the gap: nobody has run a long-term trial proving that either form of K2 prevents heart attacks or cardiovascular death. We have observational associations, a mechanism, a validated biomarker, and improved arterial stiffness. That’s a genuinely promising chain of evidence. It is not a finished one.

And MK-4? The cardiovascular data at supplement doses is thin to the point of nearly not existing. Almost every cardiovascular intervention trial, from the stiffness studies to the ongoing calcification research, uses MK-7. If your primary reason for taking K2 is arterial health, this category isn’t a close contest.

Which raises the obvious next questions. How much should you actually take, is any of this safe long-term, and what happens if you’re on blood thinners? That’s where we’re headed next.

Beyond Bones and Arteries: Where Each Form Might Have Unique Turf

This is the part of the K2 conversation where I have to slow down and separate what’s known from what’s marketed. Because both camps, the MK-4 loyalists and the MK-7 crowd, have claims here that outrun the receipts.

MK-4’s niche: hormones, brain, and cell studies

Here’s a fact that made me sit up: MK-4 is the form your body actually manufactures. Your tissues convert other vitamin K forms into MK-4 and then concentrate it in the brain, pancreas, salivary glands, and reproductive organs. That’s not random. Biology doesn’t stockpile a molecule in specific tissues for no reason, and this pattern fuels the argument that MK-4 has jobs beyond activating osteocalcin and MGP.

The testosterone angle is the one that gets shared endlessly online. Ito et al. reported in 2011 that MK-4 (but not vitamin K1) stimulated testosterone production in rats and in testicular cell cultures. Interesting? Yes. Human proof? No. That was rodent and cell work, and more than a decade later we still don’t have a solid human trial showing MK-4 raises testosterone in men. If you see a supplement marketed on that claim, you’re paying for an extrapolation.

The brain story is similar. MK-4 is involved in sphingolipid synthesis, and sphingolipids matter for neurons. There’s animal data linking vitamin K status to cognition. There is not, as of now, a human trial showing MK-4 supplements improve memory or protect against dementia. I’d call this a hypothesis worth watching, nothing more.

MK-7 and inflammation, insulin sensitivity

MK-7 has its own frontier claims. A small trial published in Nutrition Journal gave healthy adults MK-7 and found reduced inflammatory markers in circulating monocytes. Separate pilot work has poked at insulin sensitivity, since osteocalcin (the protein K2 activates in bone) appears to talk to the pancreas. Suggestive stuff. Also small, short, and unreplicated.

Look, I’ll say it plainly: this whole section is where marketing outruns evidence on both sides. If someone sells you MK-4 for testosterone or MK-7 for blood sugar, ignore them. Buy K2 for bones and arteries, where the trials actually exist, and treat everything else as a possible bonus.

One pairing that does make mechanistic sense: vitamin D3. Most quality K2 supplements include it, and for good reason. D3 increases your body’s production of osteocalcin and MGP, the very proteins K2 activates. D3 builds the workforce; K2 puts them to work. Taking one without the other leaves half the system idle.

Dosage: How Much MK-4 or MK-7 Do You Actually Need?

Practical stuff. This is where the two forms stop being a scientific debate and become a lifestyle question.

Evidence-based MK-7 dosing

The sweet spot for MK-7 is 90 to 180 mcg daily. That’s not a guess; it’s the range used in the 3-year Knapen trials on bone density and arterial stiffness. Theuwissen’s dose-finding work showed that doses as low as 90 mcg meaningfully improved carboxylation of MGP and osteocalcin over 12 weeks, while 180 mcg did it faster and more completely. Some trials went to 360 mcg with no safety issues, but I haven’t seen evidence you need that much unless a blood test says your dp-ucMGP is stubbornly high.

One capsule, once a day. That’s the whole protocol.

The MK-4 dosing problem

Now here’s what bugs me about the MK-4 market. The dose with actual fracture outcome data behind it, the dose used across those Japanese osteoporosis trials, is 45 mg per day, split into three 15 mg doses. That’s milligrams. Forty-five thousand micrograms, or roughly 250 to 500 times what’s in a typical MK-7 capsule.

Yet walk through any supplement aisle and you’ll find MK-4 products dosed at 1 to 5 mg. That dose was never tested for bone outcomes. It’s too low to replicate the Japanese protocol and, given MK-4’s absorption problems, probably too low to do much of anything measurable. These products exist in an evidence vacuum, priced as if they carry the Japanese trial data on their backs. They don’t.

Timing and how to take it

Both forms are fat-soluble, so take them with food, ideally your largest or fattiest meal. Dinner is a fine default. For MK-7, once daily is enough because of the 3-day half-life. For MK-4, the roughly 1 to 2 hour half-life means the studied protocol requires dosing three times a day, every day. Miss a dose and your blood levels are back near zero within hours.

Cost matters too. A quality MK-7 supplement at 90-180 mcg runs about $10-25 per month. Replicating the 45 mg/day MK-4 protocol typically costs more, often $30-50 monthly, and demands three-times-daily compliance that most people (myself included) won’t sustain past week two.

Safety: Is Either Form Risky?

Short answer: for most people, no. But there’s one group where this stops being a casual question.

What the toxicity data shows

Vitamin K2 has no established tolerable upper intake limit, which is unusual for a fat-soluble vitamin and tells you something about its track record. EFSA reviewed MK-7 safety and found no concerns at studied doses. Human trials up to 360 mcg of MK-7 daily ran clean. And the Japanese have prescribed 45 mg of MK-4 daily, a dose hundreds of times higher, to osteoporosis patients for years without a toxicity signal. That’s a wide margin.

Let me also kill a persistent myth: K2 does not cause dangerous blood clots in healthy people. The clotting system saturates at low vitamin K intakes. Once your clotting factors are fully activated, extra K2 doesn’t push clotting further; the surplus goes to osteocalcin and MGP. Your blood doesn’t get “thicker” because you took a K2 capsule.

Side effects reported in trials are minimal. Occasional mild GI upset, that’s about it. I’ll also mention something readers ask about: scattered anecdotal reports of sleep disturbance or heart palpitations with MK-7. I’ve looked, and there’s no trial-level evidence behind these. If it happens to you, drop the dose or switch forms, but the data says this is rare and unconfirmed.

The one group that needs to be careful

Warfarin users. This is non-negotiable. Warfarin works by blocking vitamin K recycling, and K2 directly counteracts it. Theuwissen’s group showed that MK-7 doses as small as 10 to 45 mcg, less than what’s in many multivitamins, measurably affected INR in people on vitamin K antagonists. MK-7’s long half-life makes it especially potent here. If you take warfarin, K2 supplementation is a conversation with your prescriber, full stop.

The newer anticoagulants (DOACs like apixaban, rivaroxaban, dabigatran) don’t work through vitamin K at all, so this interaction doesn’t apply to them. The warning is specific to warfarin-type drugs.

How Long Until You Notice Anything?

I need to set expectations here, because K2 is not a supplement you feel. No energy boost, no mood lift, nothing subjective. The benefits show up in blood markers first and hard outcomes years later.

Blood markers vs real outcomes

The markers move fast. Published dose-finding work by Theuwissen and colleagues showed dp-ucMGP and undercarboxylated osteocalcin improving within weeks of starting MK-7, with clear dose-dependent changes by 12 weeks. Your arteries’ security guards wake up quickly once the K2 shows up.

The outcomes take patience. In the Knapen trials, bone density and arterial stiffness differences emerged over 2 to 3 years of daily supplementation. Not 30 days. Not 90. Years. Anyone promising visible results from K2 in a month is selling something, and it isn’t science.

MK-4 follows the same logic with an extra catch: because levels crash within hours of each dose, marker improvements depend entirely on hitting all three daily doses consistently. Skip your midday capsule regularly and you’re undercutting the whole protocol.

Think of K2 like compound interest. Boring in any given month, meaningful over a decade.

My Verdict: Which One Should You Take?

I opened this article as a genuine agnostic on the mk-4 vs mk-7 question. After going through the trials, I’m not agnostic anymore.

Choose MK-7 if…

You’re most people. MK-7 wins on once-daily dosing, stable blood levels across the full 24 hours, and, this is the decisive part, bone and artery trials conducted at doses that actually match what’s in the bottle you’d buy. 180 mcg was studied for 3 years; 180 mcg is what quality supplements contain. That alignment between evidence and product is rarer in the supplement industry than it should be.

Choose MK-4 if…

You’re deliberately following the Japanese high-dose bone protocol, 45 mg daily under medical supervision, or you specifically want the exact form used in that fracture research. That’s a legitimate choice with real outcome data behind it. It’s also expensive, demands three daily doses, and in most countries requires sourcing pharmaceutical-grade material.

What I wouldn’t do is buy the low-dose MK-4 products at 1 to 5 mg. They match neither the Japanese protocol nor the MK-7 evidence. They sit in a gap, and I’d skip them entirely.

What I’d do

I’d take 90 to 180 mcg of all-trans MK-7 with dinner, paired with vitamin D3, and I’d plan on years, not weeks. Combination MK-4/MK-7 products are fine if you like them, but understand that the MK-7 content is doing the evidenced work at those doses.

Label checklist: look for “all-trans” MK-7 specifically, since cheap synthesis can produce cis isomers that are biologically inactive (some tested products have contained mostly useless cis forms). Third-party testing matters for the same reason. And take it with fat, always.

Remember the analogy I started with: MK-4 is the sprinter, explosive but gone from your bloodstream in hours; MK-7 is the marathon runner, unglamorous but still working three days later. For a protein-activation job that needs coverage every hour of every day for years, I’m betting on the marathon runner. The data agrees with me.

Frequently Asked Questions

What’s the main difference between MK-4 and MK-7?

MK-4 has a short side chain and a half-life of about 1 to 2 hours, requiring high doses (45 mg) taken three times daily in studies. MK-7 has a longer side chain, a half-life of roughly 3 days, and works at small once-daily doses of 90-180 mcg. MK-7 also produces far more stable blood levels.

How do MK-4 and MK-7 work in the body?

Both activate vitamin K-dependent proteins through carboxylation. The two most important are osteocalcin, which binds calcium into bone, and matrix Gla protein (MGP), which blocks calcium from depositing in artery walls. Same mechanism, different pharmacokinetics.

Are MK-4 and MK-7 safe to take together?

Yes. There’s no known interaction between the two forms, and combination products are common. No upper intake limit has been established for vitamin K2. The one exception is anyone on warfarin, where any K2 form interferes with the drug.

What is the best dosage for MK-4 vs MK-7?

For MK-7, 90-180 mcg once daily, the range validated in 3-year clinical trials. For MK-4, the only dose with fracture outcome data is 45 mg per day split into three doses. Low-dose MK-4 products (1-5 mg) have no outcome trials behind them.

How long does MK-7 take to work for bone and heart health?

Blood markers like dp-ucMGP and undercarboxylated osteocalcin improve within weeks, with clear changes by 12 weeks. Measurable improvements in bone density and arterial stiffness took 2 to 3 years of daily use in the Knapen trials.

Can I take MK-4 or MK-7 if I’m on blood thinners?

Not with warfarin unless your prescriber approves and monitors your INR. Even 10-45 mcg of MK-7 measurably affected INR in trials. DOACs like apixaban and rivaroxaban don’t work through vitamin K, so K2 doesn’t interfere with them.

Is MK-7 better absorbed than MK-4?

Yes. At typical supplement doses, MK-7 reliably raises serum vitamin K2 levels while equivalent microgram doses of MK-4 often produce no detectable rise at all. MK-4 only shows meaningful blood levels at milligram-range doses.

Do I need to take vitamin K2 with vitamin D3?

You don’t need to, but the pairing makes sense. Vitamin D3 increases production of osteocalcin and MGP, the proteins K2 activates. D3 creates the workforce, K2 activates it. Both should be taken with a meal containing fat.

The Bottom Line

I went into this comparison expecting a murkier answer. The supplement industry loves a false equivalence, and “MK-4 vs MK-7” gets framed as a matter of taste. It isn’t. The pharmacokinetics, the dosing practicality, and the alignment between trial doses and real products all point the same direction for the average person: all-trans MK-7, 90-180 mcg, with a fatty meal, paired with D3, for years.

MK-4 earned its place in history with the Japanese fracture trials, and at 45 mg daily it remains a legitimate clinical tool. But the 1-5 mg capsules sold on the strength of that research are borrowing credibility they haven’t earned.

Pick the marathon runner. Give it time. Check your dp-ucMGP if you want proof it’s working. That’s the whole playbook.

Frequently Asked Questions

MD, PhD

Dr. Dimitar Marinov

Medical Reviewer • Chief Assistant Professor, Medical University of Varna

Dr. Marinov is a licensed physician and scientist specializing in nutrition and dietetics with years of experience in clinical and preventive medicine. He references every statement with high-quality research.

LinkedIn Instagram Google Scholar ResearchGate Website