Vitamin K spent most of its history known for exactly one job: helping blood clot. The K is even from Koagulation, the German spelling used by the researcher who first described it. That reputation is accurate but incomplete, and the incomplete part is where most of the current supplement interest lives.
Here’s what vitamin K does, the real difference between K1 and K2, how much you need, and the one interaction that genuinely matters.
What vitamin K actually does
Vitamin K is a cofactor for a single chemical reaction — carboxylation — performed on a specific family of proteins. Without vitamin K, these proteins get built but never activated. They’re like tools that come out of the box without a handle.
Three groups of these proteins matter:
Clotting factors. Several of the proteins in the coagulation cascade require vitamin K activation. This is the classic role, it happens in the liver, and it’s the reason newborns receive a vitamin K shot — babies are born with very low stores and can’t yet make their own via gut bacteria.
Osteocalcin. A protein made by bone-building cells that, once activated by vitamin K, binds calcium into the bone matrix. Low vitamin K status shows up as more undercarboxylated (inactive) osteocalcin circulating.
Matrix Gla protein. Made in blood vessel walls and cartilage, where its activated form appears to inhibit calcium from depositing in soft tissue.
That third one is the interesting part. Vitamin K doesn’t add calcium to your diet or pull it from anywhere — it’s involved in the machinery that determines where calcium ends up. The shorthand you’ll see everywhere is that vitamin K “directs calcium to bone, not arteries.” That’s a reasonable summary of the mechanism. It is not a demonstrated clinical outcome from taking a supplement, and the two get conflated constantly.
K1 vs K2: a real distinction, oversold
Vitamin K1 (phylloquinone) is the plant form. It’s what’s in leafy greens, and it makes up roughly 75-90% of vitamin K in a typical diet. K1 is absorbed and taken up heavily by the liver, where it services clotting factors. Comparatively little reaches bone or arterial tissue.
Vitamin K2 (menaquinones) comes from fermented foods, some animal products, and bacterial synthesis. The forms are labeled MK-4 through MK-13 by side-chain length, and two matter commercially:
- MK-4 is found in animal foods (liver, egg yolk, butter) and can also be converted from K1 in some tissues. It has a short half-life — a few hours — so blood levels spike and fall. Research using MK-4 has often used pharmacological doses far above what supplements provide.
- MK-7 comes from bacterial fermentation; natto, the Japanese fermented soybean dish, is by far the richest food source. Its half-life is measured in days rather than hours, so a modest daily dose builds and holds steady blood levels. That’s why nearly every K2 supplement on the shelf is MK-7.
The honest state of the evidence: K2’s better distribution to bone and vascular tissue is well demonstrated, and supplementation reliably improves markers of vitamin K status. Whether that translates to fewer fractures or less arterial calcification in ordinary healthy adults is still an open question. Some trials in postmenopausal women have shown effects on bone markers with less consistent effects on actual fracture rates. Treat K2 as biologically sensible and clinically unproven — not as an established preventive.
How much you need
There’s no RDA for vitamin K, because there wasn’t enough data to set one. Instead there’s an adequate intake:
- Men 19+: 120 mcg/day
- Women 19+: 90 mcg/day
Two things to understand about that number. First, it was set based on what’s needed for normal clotting — the liver function. It was not set based on what fully activates osteocalcin or matrix Gla protein, and some researchers argue those tissues need more. Second, it’s easy to hit. Half a cup of cooked kale or spinach can deliver several hundred micrograms.
Frank vitamin K deficiency is rare in healthy adults. It shows up mainly in people with significant fat malabsorption (celiac disease, cystic fibrosis, short bowel, cholestatic liver disease), in people on long courses of broad-spectrum antibiotics, and in newborns.
There is no tolerable upper limit for vitamin K, because no toxicity has been established from food or supplemental K1 and K2. That’s genuinely unusual for a fat-soluble vitamin — and it should not be read as “unlimited is fine,” only that a threshold hasn’t been identified. (Synthetic K3, menadione, is a different molecule with real toxicity concerns and isn’t sold as a human supplement.)
Food sources
K1 (per typical serving):
- Cooked kale, collards, spinach, turnip greens — several hundred to over 1,000 mcg per cooked cup
- Broccoli, Brussels sprouts — roughly 100-200 mcg per cooked cup
- Raw leafy salads — 50-100 mcg per couple of cups
- Vegetable oils, especially soybean and canola — meaningful amounts in a tablespoon
K2:
- Natto — far and away the richest source, several hundred to over 1,000 mcg of MK-7 per serving
- Hard and soft aged cheeses — tens of micrograms
- Egg yolks, liver, butter from pasture-raised animals — modest MK-4
- Fermented foods generally, in variable amounts
Because vitamin K is fat-soluble, absorption improves substantially when it’s eaten with fat. Dressing on the salad isn’t optional if the salad is your K source. This is the same principle covered in our guide to fat-soluble versus water-soluble vitamins.
Supplements: forms and sensible doses
If you supplement, the practical version:
- K2 as MK-7: 90-180 mcg/day is the range most products use and most trials have studied. Higher isn’t clearly better.
- K1 supplements exist but rarely make sense for someone eating vegetables.
- Combination products typically pair K2 with vitamin D3, often 1,000-5,000 IU D3 with 100-180 mcg MK-7. The rationale — vitamin D increases calcium absorption, vitamin K influences where it goes — is coherent, though whether the pairing produces better outcomes than D alone is still being worked out. We look at that combination specifically in vitamin D with K2.
- Take with a meal containing fat. Any meal with real fat in it works. Timing within the day doesn’t matter much for MK-7 given its long half-life.
Who might reasonably consider it: people who genuinely eat very few green vegetables and no fermented foods, people taking high-dose vitamin D and calcium, and people with malabsorption conditions — that last group under medical guidance, not on their own. Everyone else is likely already covered by food, and food remains the better vehicle.
The warfarin interaction — and the thing people get backwards
This is the most important safety point in this guide.
Warfarin works by blocking vitamin K recycling. Its dose is calibrated to your habitual vitamin K intake and monitored with INR blood tests. So the danger isn’t vitamin K itself — it’s change.
Someone on warfarin who suddenly starts a K2 supplement, or suddenly starts eating a large kale salad daily, can push their INR down and raise clot risk. Someone who abruptly stops a habitual source can push their INR up and raise bleeding risk. The clinical guidance is to keep vitamin K intake steady day to day, not to eliminate it. Cutting greens entirely is not the goal and can make dosing harder to stabilize.
If you’re on warfarin: don’t start or stop a vitamin K supplement without telling your prescriber, and expect more frequent INR monitoring if anything changes. This does not apply the same way to newer direct oral anticoagulants (apixaban, rivaroxaban, dabigatran), which don’t work through vitamin K — but they still warrant a conversation before adding anything. Our supplement-drug interactions guide covers the broader picture.
A few other notes: high-dose vitamin E supplementation may interfere with vitamin K function at very high intakes; bile acid sequestrants and orlistat can reduce absorption of fat-soluble vitamins including K; and people with kidney disease taking calcium or vitamin D should coordinate any K supplement with their care team.
Bottom line
Vitamin K’s clotting role is settled and easily covered by a diet with regular green vegetables — 90-120 mcg/day is the target, and a single serving of cooked greens often exceeds it. The K2 story about bone and arterial calcium is mechanistically credible and clinically unproven, which makes MK-7 at 90-180 mcg/day with a fatty meal a reasonable low-risk experiment for people with thin dietary intake or those already taking vitamin D and calcium — but not something to expect measurable results from. If you take warfarin, keep your intake consistent and change nothing without your prescriber.
This guide is educational and is not medical advice. Talk with a healthcare provider before starting any supplement — especially if you are pregnant, nursing, taking medication (particularly anticoagulants), or managing a health condition.