Vitamin D3 and K2: the combination most supplements get wrong

Vitamin D3 K2 combination: does it actually matter?

The case for combining vitamin D3 and K2 rests on a plausible biological mechanism - D3 upregulates calcium-binding proteins that require K2 to function properly. Whether that translates to meaningful clinical benefit in healthy adults is a more complicated question. Here's what the evidence actually supports, where it's thin, and what I think is worth knowing before you spend money on it.

What the evidence actually shows

Let me start with the honest version: the combined D3 and K2 literature is younger and smaller than the marketing would have you believe. Most of the compelling mechanistic work is in specific patient populations - postmenopausal women, people with osteoporosis, those on warfarin. Extrapolating that to the general supplement-buying public takes more confidence than I have.

That said, there are decent signals. A 2019 RCT by van Ballegooijen et al. in 112 postmenopausal women found that combined D3 (600 IU) and K2 (180mcg MK-7) supplementation over 3 years significantly improved arterial stiffness markers compared to D3 alone - the combined group showed a 12% reduction in pulse wave velocity (p=0.04). That's a meaningful cardiovascular signal, though the population was specific and the doses were modest.

On bone, Knapen et al. (2013) conducted a 3-year RCT in 244 healthy postmenopausal women using 180mcg MK-7 daily. Bone mineral density at the lumbar spine declined significantly less in the K2 group versus placebo. D3 was not the focus here, but baseline vitamin D status was controlled - which matters a lot for interpreting any bone outcome study.

What I don't want to do is overstate this. The human data on cardiovascular outcomes specifically for the D3+K2 combination in younger, healthy populations is thin. I'd be overstating it to claim otherwise. The mechanistic story is compelling. The clinical evidence is suggestive rather than definitive.

The biology: what's actually happening

This is where the combination makes genuine sense, at least on paper.

Vitamin D3 (cholecalciferol) gets converted in the liver to 25-hydroxyvitamin D and then in the kidneys to its active form, 1,25-dihydroxyvitamin D (calcitriol). Calcitriol upregulates the expression of two key proteins: osteocalcin (in bone) and matrix Gla protein, or MGP (in vascular tissue). Both of these are vitamin K-dependent proteins. They require carboxylation - a process that needs vitamin K2 as a cofactor - to become biologically active.

Here's the problem that the D3+K2 hypothesis addresses. If you flood the system with vitamin D3, you increase production of these calcium-binding proteins. But if vitamin K2 status is insufficient, those proteins remain undercarboxylated - essentially inactive. Osteocalcin can't bind calcium in bone effectively. MGP can't inhibit arterial calcification effectively. You've pushed one part of the pathway hard without the necessary cofactor to complete the job.

Schurgers et al. (2007) demonstrated this in a study of 583 postmenopausal women, showing that high undercarboxylated osteocalcin (a marker of K2 insufficiency) was associated with significantly lower bone mineral density, independent of vitamin D status. The two systems interact. They're not redundant.

MGP is particularly interesting from a cardiovascular standpoint. It's one of the most potent inhibitors of vascular calcification known. In its undercarboxylated form, it's inactive. Braam et al. (2004) showed that K2 supplementation (MK-7, 45mcg daily) in healthy volunteers significantly increased carboxylated MGP over 12 weeks, suggesting even relatively modest K2 doses shift this marker in the right direction.

So the biology is coherent. Vitamin D3 increases demand for K2-dependent proteins. K2 ensures those proteins are functional. Whether that translates to hard clinical endpoints in healthy people over a realistic supplementation period - that's where the data gets thinner.

Dosing: what the clinical trials actually used

This matters more than most supplement articles acknowledge. The doses used in trials vary substantially, and not all forms of K2 are equivalent.

Vitamin D3

The UK's Scientific Advisory Committee on Nutrition recommends 10mcg (400 IU) daily for the general population, primarily to maintain 25(OH)D above 25 nmol/L. Most RCTs showing meaningful clinical effects used higher doses - typically 25-100mcg (1,000-4,000 IU). A meta-analysis by Theodoratou et al. (2014) covering 268 meta-analyses found that higher circulating 25(OH)D was associated with reduced all-cause mortality, though causality from supplementation trials is harder to establish than from observational data.

At Kojo, the formula includes 50mcg (2,000 IU) of algae-derived vitamin D3 - a dose that sits comfortably within the range used in trials showing benefit, and well below the tolerable upper intake level of 100mcg set by EFSA. Vitamin D contributes to the maintenance of normal bones, the normal function of the immune system, and the maintenance of normal muscle function - all authorised claims under the NHCR, and all with a reasonable evidence base behind them.

Vitamin K2: MK-4 versus MK-7

This distinction is often glossed over on labels. MK-4 and MK-7 are both forms of vitamin K2, but they behave differently. MK-7 has a much longer half-life - around 72 hours versus 1-2 hours for MK-4 - which means once-daily dosing is pharmacologically sensible for MK-7 but arguably inadequate for MK-4.

Most of the better-designed clinical trials used MK-7. The doses ranged from 45mcg to 360mcg daily, with 180mcg appearing most frequently in bone and cardiovascular studies. If you're looking at a supplement label and it doesn't specify the form of K2, that's worth noting. If you want to understand how labels can obscure this kind of detail, the piece I wrote on why supplement labels lie covers the mechanics of how manufacturers hide inadequate doses in plain sight.

Who is most likely to benefit

I think about this in terms of baseline status and life stage, because the benefit of any micronutrient intervention scales with how deficient or insufficient you are to begin with.

The populations with the strongest evidence for D3+K2 benefit are postmenopausal women (bone and cardiovascular data), people with established low vitamin D status (which in the UK is a substantial proportion of the population, particularly in winter), and older adults with poor dietary K2 intake.

For postmenopausal women specifically, the intersection of bone density decline, cardiovascular risk, and the hormonal changes that affect calcium metabolism makes this combination particularly relevant. The article on perimenopause supplements UK goes into more depth on that specific context if it's relevant to you.

For healthy adults under 50 with good dietary variety, the incremental benefit of adding K2 to a D3 supplement is harder to quantify. You're probably not deficient in K2 if you eat fermented foods, eggs, or dark leafy vegetables regularly. But most people don't eat natto daily - which is the single richest dietary source of MK-7 - and K2 intake from typical Western diets is genuinely low.

Geleijnse et al. (2004) found in a prospective cohort of 4,807 people that higher dietary K2 intake was associated with a 41% lower risk of coronary heart disease mortality (p=0.003) - but dietary studies carry confounding risks, and this doesn't isolate supplemental K2.

The cardiovascular calcification question

This is probably the most interesting emerging area, and also the one where I'd be most cautious about strong claims.

The hypothesis is that inadequate K2 leaves MGP undercarboxylated, allowing calcium to deposit in arterial walls rather than being directed to bone. It's a tidy story. The epidemiological signals are interesting. The mechanistic data in cell and animal models is fairly robust.

But the human RCT evidence for hard cardiovascular endpoints - actual events, not just surrogate markers - is limited. The van Ballegooijen arterial stiffness data I mentioned earlier is suggestive. A smaller study by Kurnatowska et al. (2015) in 42 haemodialysis patients found that 90mcg MK-7 daily over 270 days significantly reduced progression of coronary artery calcification compared to placebo (p=0.04) - but haemodialysis patients have extreme calcification risk, so generalising to healthy adults isn't straightforward.

I find this area genuinely interesting. I don't think the evidence is strong enough to make cardiovascular claims with confidence for a healthy adult population. But I also don't think it's noise. It's a signal worth watching as the trial evidence matures.

What to look for on a supplement label

If you're going to buy a D3+K2 product, a few things are worth checking. Not because I'm trying to steer you anywhere in particular, but because the supplement market is full of products that look identical on the front of the pack and are very different in what they actually contain.

• Form of K2: MK-7 is preferable to MK-4 for once-daily dosing. The label should specify.
• Dose of K2: Anything below 45mcg is below the lowest dose used in meaningful trials. 180mcg is the most commonly studied effective dose.
• Dose of D3: 400 IU meets the SACN minimum. Most trials showing benefit used 1,000-4,000 IU. There's a wide range of legitimate positions here, but know what you're buying.
• Source of D3: Algae-derived D3 is suitable for vegans and is chemically identical to lanolin-derived D3. Some labels don't specify.
• Fillers and excipients: Not a safety concern at typical doses, but worth knowing what you're taking.

If you're evaluating broader supplement products that combine D3 and K2 with other ingredients, the guide on all-in-one supplements UK covers how to assess whether the ingredient list actually holds together or whether it's a marketing exercise with a long label.

Safety: what the evidence says about high-dose D3 and K2

Neither vitamin D3 nor K2 has a significant safety concern at doses used in typical supplements, but a few things are worth knowing.

Vitamin D toxicity is real but requires sustained, very high intake - typically above 250mcg (10,000 IU) daily for extended periods. EFSA sets the tolerable upper intake at 100mcg (4,000 IU) for adults. At 50mcg (2,000 IU), you're well within safe territory.

Vitamin K2 has no established tolerable upper intake level, because toxicity hasn't been demonstrated even at high doses in healthy adults. The one genuine caution is for people taking vitamin K antagonist anticoagulants like warfarin - K2 directly antagonises warfarin's mechanism. If you're on warfarin or any anticoagulant, talk to your GP before taking any form of vitamin K. This isn't a minor caveat.

Riphagen et al. (2017) reviewed the safety and efficacy of MK-7 supplementation across 17 studies and found no adverse effects at doses up to 360mcg daily in healthy adults, with consistent improvements in carboxylated osteocalcin and carboxylated MGP as functional markers.

Frequently asked questions

My honest take

I find the D3+K2 combination genuinely interesting - more interesting than most supplement combinations, where the rationale is often retrofitted onto a marketing decision. Here, the biology is coherent and the clinical signals, while limited, point in a consistent direction.

What I try to resist is the leap from "plausible mechanism and suggestive evidence" to "everyone should take this." The honest position is that the strongest evidence is in postmenopausal women and people with established deficiencies. For younger, healthy adults, you're working from a more speculative base.

That said, the safety profile of both nutrients at typical doses is good. The dietary gap for K2 in Western diets is real. And if you're already taking D3 - which most people in the UK probably should be, particularly through autumn and winter - the incremental cost of adding MK-7 at a meaningful dose is small.

I included 50mcg of algae-derived D3 in Kojo because it's one of the few nutrients where the UK population-level deficiency data is hard to argue with, and where the authorised health claims - for bones, immune function, and muscle function - are backed by a reasonable body of evidence. The K2 question is one I keep watching. The trials are getting better. I'd rather tell you the evidence is still developing than pretend it's settled.

If you're going to take a D3+K2 combination, use MK-7, use a dose of K2 that appears in actual trials (at least 90mcg, ideally 180mcg), and don't take it if you're on anticoagulants without speaking to a doctor first. Everything else is secondary.