Vitamin K2 with D3: Do You Need to Take Them Together, and Does It Make Sense?

Vitamin D3 increases the absorption of calcium from the gut and prompts cells to produce certain calcium-binding proteins, including osteocalcin in bone and the protein MGP (matrix Gla protein) in vessel walls. These proteins themselves, however, do not work: to be able to bind calcium they must be 'activated' by carboxylation, and the enzyme that does this requires vitamin K as a cofactor. Hence the idea: if D3 raises the amount of proteins and K2 activates them, together they should more effectively direct calcium to bone and prevent its deposition in arteries.

That is a good description of what we know: the mechanism is plausible and well described in laboratory studies and in measurements of markers (e.g. the proportion of undercarboxylated osteocalcin), but the review itself is narrative, and the authors admit that synergy requires further research. Biological plausibility is not yet proof of clinical benefit. Many ideas that make sense at the level of proteins do not translate into fewer fractures or heart attacks, which is why randomized trials with hard or at least surrogate endpoints are so important.

This is so far one of the strongest pieces of evidence in favor of combining K2 with D3, but it must be read with two caveats. First, the intervention group received both vitamins at once and the control group placebo, so the trial neither separates the effect of K2 from that of D3 nor shows that the combination is better than K2 alone. Second, the result concerns a surrogate endpoint (the rise in a calcium score on CT), not the number of heart attacks or deaths, and concerns people with severe, advanced calcification, aged 71 on average. A difference of 52 Agatston units is about one fifth of the increase in the placebo group (from 248 AU), and the clinical significance of such a difference has not yet been confirmed in trials with hard endpoints. The dose of K2 (720 µg) was also many times higher than in typical supplements (usually 100-200 µg MK-7).

The 2023 meta-analysis concerned different forms and doses of vitamin K and different populations, so its result does not translate directly to a healthy person taking 100 µg of K2 prophylactically. It does, however, point in a direction consistent with DANCODE: a benefit in slowing coronary calcification, but modest and subject to uncertainty.

For bone the picture is less favorable than for calcification and, importantly, not uniform. Two well-known three-year randomized trials with MK-7 gave conflicting results. In the older Dutch study by Knapen and colleagues, 244 healthy postmenopausal women took 180 µg of MK-7 a day or placebo for 3 years; MK-7 improved vitamin K status and reduced the decline in bone mineral density in the lumbar spine and femoral neck (but not in the total hip) and improved bone strength indices.

The Rønn study is particularly important for our question, because all participants received D3, and what was tested was precisely the addition of K2 to D3 and calcium. The result is that K2 really changed the biochemistry (less undercarboxylated osteocalcin), but this did not translate into better bone density over 3 years. Together with Knapen's positive result this gives an ambiguous picture: there is no convincing evidence that adding K2 to D3 reduces bone loss, still less the number of fractures. The small size of the studies and differences in dose and population (healthy women versus women with osteopenia) make it hard to reconcile the results.

Doses in the studies varied and were often much higher than those found in the most popular preparations (usually 45-200 µg of K2 per capsule), so the results cannot be directly translated to a standard supplement. We found no study that directly compared a 'K2 with D3 together' schedule with a 'K2 and D3 at different times of day' schedule. In practice, then, convenience decides the combination: one capsule with both vitamins means less to remember, and that increases regularity. If, however, you need a larger dose of D3 than of K2 in a ready-made capsule (or the reverse), nothing prevents taking them separately.

Both vitamins are fat-soluble, and their absorption in the gut depends on the presence of fat in the meal and on bile salts. The standard advice is therefore to take K2 and D3 with a meal containing fat (eggs, fish, olive oil, nuts, full-fat dairy), not on an empty stomach and not with a fat-free meal. For D3 we have a separate text on time of day and fat; for K2 we found no study comparing times of day, so the recommendation rests on the general principle of absorption of fat-soluble vitamins, not on direct evidence for a K2 supplement.

From a practical point of view this means that the best time is the one at which you eat the fattiest meal of the day and remember your supplements — often breakfast or lunch. Time of day has no known biological significance beyond this rule. If you have disorders of fat absorption (pancreatic, liver or biliary disease, after bariatric surgery), the effectiveness of oral fat-soluble vitamins may be weaker and it is worth discussing monitoring with a doctor.

This is the most important warning in the whole article. Vitamin K antagonists (warfarin, acenocoumarol) work by blocking the vitamin K cycle, so K2 from a supplement acts in exactly the opposite way and can weaken the drug's effect, lowering INR and increasing the risk of clots. The study concerned healthy volunteers on acenocoumarol, but the mechanism is the same for warfarin. So do not start K2, nor 'with D3 in one preparation', if you take such a drug, without the consent of your treating doctor, who can plan INR checks. The new oral anticoagulants (apixaban, rivaroxaban, dabigatran) do not act through the vitamin K cycle, but in their case too it is worth telling your doctor about every supplement.

Combining K2 with D3 is an example of an idea that makes biological sense and has obtained the first promising data, but has still not passed the test of hard endpoints. The large, well-designed DANCODE trial nicely shows how carefully such a result has to be read: the effect is real at the level of imaging, but it concerns people with advanced disease, a high dose of K2 and a combination, not a single vitamin. For a healthy person with a normal D3 level, a sensible strategy is not 'adding K2 just in case' but a conscious decision after talking to a doctor, especially when there are cardiovascular or osteoporosis risk factors.