Vitamin K2, D3 and Coronary Calcium: The DANCODE Trial

The DANCODE trial (DANish COronary DEcalcification) was presented at the European Society of Cardiology Congress in August 2026 and published in Circulation. Danish researchers randomised 398 men and women who already had severe coronary artery calcification — a calcium score of at least 400 Agatston units, with a median near 900 — to either daily vitamin K2 (720 micrograms of MK-7) with vitamin D3 (25 micrograms, or 1,000 IU), or a matching placebo. Everyone had a heart CT at the start, at one year and at two years.

After two years, the calcium score had risen by 196 points in the supplement group and 248 points in the placebo group. That is a difference of 52 points, or roughly 21% slower calcium build-up, and it held in both women and men. For a cheap, well-tolerated supplement, that is a genuinely interesting result — but read the next section before you rush to a pharmacy on Orchard Road.

This is the point we make with every patient in Singapore who brings in a supplement bottle: vitamin K2 did not remove calcium, and it did not shrink plaque. The calcium score still went up in the DANCODE trial — just more slowly. Total plaque volume and, crucially, the soft non-calcified plaque that is most closely linked to heart attacks were unchanged.

There is a second gap. DANCODE measured calcium on a CT scan, not heart attacks, strokes or deaths. It was not large enough or long enough to tell us whether 21% slower calcification translates into a single prevented event. Calcified plaque is, if anything, the stable and healed form of plaque, so slowing its accumulation is a surrogate marker — a reasonable one to study, but not proof of benefit. Anyone selling you K2 as heart-attack prevention is ahead of the evidence.

The biology is plausible, which is why the trial was done. Your artery walls make a protein called matrix Gla protein, one of the body's natural brakes on calcification. It only works once vitamin K has activated it — and if you are short of vitamin K, more of it circulates in its inactive form, which several population studies, including the Rotterdam Study, have linked to more coronary artery calcification and more cardiovascular events.

The form used in the trials, MK-7, stays in the blood far longer than the vitamin K1 in leafy greens, with a half-life of about three days, so one daily dose keeps matrix Gla protein topped up. Vitamin D3 was added because it helps regulate where calcium goes in the body. We see a surprising amount of vitamin D deficiency in Singapore despite the sunshine, largely from indoor working hours, so the D3 component is rarely wasted. It is a sensible mechanism — but mechanisms have misled cardiology many times before.

DANCODE studied a specific group: people with a lot of existing coronary calcium (a score of 400 or more) who were, for the most part, already on guideline cholesterol and blood-pressure treatment. It did not test people with a calcium score of zero or a low score, and it was never designed as an alternative to a statin.

So if your recent coronary calcium score in Singapore came back at 15, this trial does not apply to you, and your attention is better spent on LDL cholesterol, blood pressure and not smoking. If it came back at 600, the trial is relevant — but as an add-on to proper medical therapy, not a replacement. We would still want your LDL cholesterol well below 1.8 mmol/L, and often lower.

This matters, so we will be blunt. Vitamin K2 directly opposes warfarin, which works by blocking vitamin K. In a careful dose-response study in healthy volunteers, just 45 micrograms of MK-7 a day lowered the INR by around 40%, and even 10 to 20 micrograms produced a clinically meaningful drop in most people. The DANCODE dose was 720 micrograms — about sixteen times that.

If you are on warfarin, or the related drugs phenprocoumon or acenocoumarol, starting K2 without telling anyone can push your INR down and raise your stroke risk. Do not start it on your own. Speak to whoever runs your anticoagulation monitoring in Singapore first; occasionally a small, fixed dose of vitamin K is used deliberately to stabilise a swinging INR, but that is a supervised decision. If you take a DOAC such as apixaban, rivaroxaban, dabigatran or edoxaban, vitamin K does not interact and this warning does not apply.

For most people the honest answer is that it is optional, and it is not a priority. If your coronary artery calcification score is high, your LDL cholesterol and blood pressure are already at target, and you have realistic expectations, then adding vitamin K2 with D3 is reasonable and low-risk — provided you are not on warfarin. The trials used a combined preparation of 720 micrograms of MK-7 and 25 micrograms of D3; a general multivitamin with a token amount of K2 is not the same thing.

In Singapore, a month of a good-quality MK-7 plus D3 supplement runs about SGD 25 to SGD 60. That is not a large sum, but it is also not doing the heavy lifting. At our clinic at Paragon Medical Centre we are happy to talk it through, usually at the end of a consultation that has already sorted out the things that genuinely change outcomes.

DANCODE is a single trial that moved a marker on a scan rather than a patient outcome. We think it is worth knowing about, and we raise it with the right patients in Singapore, but several questions remain genuinely open: