Vitamin K Slows Coronary Calcium: NEW Netherlands Study

Can vitamin K2 slow coronary artery calcium? A new JAMA Cardiology randomized trial found that MK-7 may reduce CAC progression over 2 years — but the real story is more complicated.

TL;DR: In this randomized trial, people with symptomatic coronary artery disease and CAC scores between 50–400 took either MK-7, a form of vitamin K2, at 360 mcg daily or placebo for 2 years. The MK-7 group had slower progression of coronary artery calcium and calcium mass, with no major safety signal. But this study did not prove fewer heart attacks, fewer stents, or lower mortality — so this is interesting, not a miracle. Especially important: do not start vitamin K supplements without medical guidance if you take warfarin/Coumadin.

Article link: https://jamanetwork.com/journals/jamacardiology/fullarticle/2850256

In this video, I break down the VitaK-CAC trial: who was studied, what dose was used, what happened to CAC scores, why plaque calcification is biologically complicated, and what I would — and would not — take away from this as a vascular surgeon.

⏰ Chapters
0:00 - New vitamin K study: why this matters
0:39 - The paper: MK-7 and coronary artery calcium
1:42 - Vitamin K1 vs K2, why MK-7
2:13 - What the VitaK-CAC trial tested
3:08 - Calcified plaque vs total plaque burden
3:28 - Study design: randomized, placebo-controlled, double-blind
4:41 - Who was included and excluded in the trial, plus why it matters
6:32 - What “symptomatic CAD” actually meant here
8:47 - The intervention: 360 mcg MK-7 daily
9:13 - CT scans, CAC scoring, and calcium mass
10:59 - How vitamin K status was measured
12:29 - Primary outcome: CAC progression, not heart attacks
13:03 - Statistical methods and “fast progression”
16:44 - LDL levels, statins, and a possible data discrepancy
18:25 - Baseline characteristics of the study population
22:12 - Adherence, adverse events, and vitamin K blood levels
23:31 - Matrix Gla protein: did MK-7 do what it was supposed to?
24:58 - Primary results: CAC progression slowed with MK-7
26:21 - Why the effect is exciting but modest
29:17 - Fast progressors: no significant difference
30:07 - Calcium mass and intention-to-treat analysis
32:02 - Secondary outcomes: plaque type and stenosis
33:32 - The big question: is less calcification always better?
35:40 - Discussion: promising, but clinical significance unknown
37:13 - Why MK-7 does not “remove” existing calcium
38:23 - Vitamin K biomarkers and possible under-dosing
42:25 - Study limitations
43:41 - Funding, conflicts of interest, and final takeaways
44:48 - My clinical take on vitamin K2/MK-7

This video is for education only and is not personal medical advice. Do not start, stop, or change supplements or medications based on YouTube — especially if you take warfarin or any medication affected by vitamin K. Talk with your own clinician.

High Quality 3rd-party tested supplements at Fullscript (10% discount): https://us.fullscript.com/welcome/ljohnstonmd/store-start

Other tools for optimal health (note these are affiliate links):
Home BP Cuff: https://amzn.to/49Cq7rh
Sonicare Toothbrush: https://amzn.to/3KllfMS
WaterPik: https://amzn.to/4894Xi8
Apple Watch with sleep & HR monitoring: https://amzn.to/3XaUn5d
Oura Ring with sleep, cycle, and recovery monitoring: https://amzn.to/4riyhLS
Intake Breathing Nasal Support: https://amzn.to/48dfQQ1
Personal Blender: https://amzn.to/4pyNGGl
True Nutrition Protein: https://oken.do/q2xzqeqw
Glass Meal Prep Containers: https://amzn.to/4ocEPJ2
Sleep Mask: https://amzn.to/48fSJV5
Air Purifier: https://amzn.to/4puzTQZ

Sign up for more information on my own practice here: https://corsighthealth.com/
___________________________
🧬 About Dr. Lily Johnston
  
Dr. Johnston is a double board-certified vascular and general surgeon in San Diego, specializing in metabolic and cardiovascular prevention. She’s the founder of CorSight Health and a passionate advocate for reimagining how medicine approaches chronic disease.

Transcript

SPEAKER_00 0:00

A brand new study just dropped this week, and it is the first time we have ever seen vitamin K supplementation actually result in a positive outcome for patients. I'm really excited to go through this study with you guys today. And if you're new here, my name is Dr. Lily Johnston. I am a board-certified vascular surgeon, but I specialize in cardiometabolic prevention, so hopefully, you will never need me as a surgeon. We are trying a bit of a new layout today, so let us know what you think. And please bear with me, my internet has been just a little spotty, so the video may lag in some places, but we're gonna do our best. The paper is titled Two Years of Mediquinone 7 Supplementation and Coronary Artery Calcification, a randomized controlled clinical trial. This was published this week online in Jamma Cardiology. It is by Sugars et al. And the abstract talks to us about everything that's in the whole paper. We're gonna skip past that and go straight to the introduction. And the paper starts by reminding us that vitamin K in population studies has been a very powerful predictor of low cardiovascular events, meaning patients who've had higher levels naturally of vitamin K in their bloodstream have lower rates of cardiovascular disease. However, the trials that have been done to date show at best a mixed picture, mostly a negative picture. There have been several randomized controlled clinical trials to date that have not shown any impact of vitamin K on arterial calcification or aortic valve calcification. So they also say here the two major compounds that constitute vitamin K are phyloquinone or vitamin K1 and the menequinones or vitamin K2, which there are several homologues. And although K1 is usually the most common, the menequinone 7 or MK7 has better bioavailability and a longer half-life. And so that has been what's been used in many of the trials and is what they're calling preferable as a supplement to support vitamin K status. So presently it is unclear whether supplementation with MK7 can halt or retard the progression of coronary artery calcification in patients with symptomatic coronary lesions. We're going to talk about what symptomatic means here in a bit. The premise of this whole trial, the Vita K CAC trial, is that they hypothesized in patients with coronary artery disease, supplementation with vitamin K and K7 for a period of two years would attenuate or slow down the progression of coronary artery calcium or in calcium mass compared to a placebo control. They say we did not include measurements of total plaque volume. Changes in plaque volume are less reproducible than those in calcium score or mass, and they are associated with distinct pathophysiological mechanisms that are independent from calcification per se. Let me translate that. They are not looking at total plaque burden, including the soft plaque. This is only about the calcified plaque that shows up in a coronary calcium score. And that becomes important later as we're going to talk about what the implications of this paper are, but let's go into the methods and see what they did. So they say the details of our study have been described before, and the study protocol appears in supplement one. I reviewed supplement one for us, and we can talk about what's in there in a minute. But in brief, this is a double-blind, placebo-controlled, randomized phase three trial. So let's take that piece by piece, double blind. Neither the patients who receive the substance nor the investigators who are giving it know which patient is getting placebo and which one is getting the active agent. So that's the second part, which is placebo-controlled. Everybody's getting a pill of some kind and it's randomized. So patients are randomly assigned to one treatment versus placebo. Phase three trial just means it is already been established as safe, and now we're looking at a more sophisticated endpoint. And the group that they're studying are patients with pre-existent coronary artery calcification. And they recruited from two different hospitals, one university center and one community-dwelling hospital. This is important for generalization of the results. And they evaluated whether over a period of two years, which is a decent length of time, the oral and K7 supplementation would slow the rate of coronary calcium progression compared to placebo. Here are their inclusion criteria. Patients had to be at least 18 years of age, have concerns suggestive of coronary artery disease. Now, this is where they're calling this symptomatic, and I have a little bit of a quibble here because they never say what this is. I assume that this is chest pain, but I even went through the supplement and it doesn't really say what concerns suggestive of coronary artery disease would be. I assume that that's angina or chest pain with activity, but I don't know. And the patients must also have a baseline calcium score between 50 and 400 Agatsten units on a particular kind of CT scanner. The exclusion criteria, who was not allowed to participate? Patients with type 1 diabetes, overt kidney disease, or a filtration rate GFR of less than 60. So this is important because patients with long-standing diabetes and kidney disease accumulate a lot of calcium in the walls of their arteries as part of the pathophysiology of their disease. MK7 or vitamin K has been studied in particularly kidney disease patients in a couple of randomized trials with negative results. And so they wanted to remove that as a confounder in this particular study. Patients with reduced life expectancy, which in the supplement is less than two years, use of vitamin K antagonists, so those are medications that block the action of vitamin K. And the one that's relevant here is warfarin or cumidin, and pregnancy or the wish to become pregnant. They also excluded patients who were scheduled to undergo a revascularization procedure, meaning a coronary bypass operation or a coronary stent, and those who had a myocardial infarction, heart attack, or stroke less than six months before inclusion. Digging a little deeper in the supplement, they also excluded anybody with a history of revascularization, stenting, or coronary artery bypass. So when I think about symptomatic disease, I'm usually thinking of heart attack, stroke, or stenting. That's not what they mean here. What they mean are, I think, chest pain. They really, again, are very vague about what that concern related to coronary disease would be, but I have to assume it's chest pain. So these are not people who have yet had an event. We are still talking about primary prevention here. That means we're trying to avoid a first event like heart attack or stroke. Okay. And then they excluded patients in whom the baseline scan was of insufficient quality because of motion artifact, heart rate, or rhythm disturbances, and interfering problems were excluded as well. I don't know exactly what that means. Here's another important point about this trial. Recruitment started in January 2012, was interrupted a few times for logistical reasons, and the COVID-19 pandemic, close out was in October of 2022. So this spans a decade. And, you know, if you think about what has happened in the last 10 years of your life, like a lot can happen in a decade. And especially as we're thinking about the medical management of patients with coronary disease, I have some concerns that even though they say treatment has been, you know, largely stable, and individually the patients were only followed for two years at a time, a patient who enrolled in 2012 might be a little different than patients who enrolled in 2020. I don't know for sure. They don't really have a concern about that as they present their data, but I just wonder. All right, randomization and masking. Basically, they did a decent job randomizing and making sure that patients were not aware what they were doing and um they stratified for age, sex, body mass index, statin use, and bisphosphonate use. Bisphosphonates are the category of drugs that patients take when they have osteoporosis that are intended to remineralize the bones, the um poorly mineralized bones. It turns out this was not a huge confounder in their study. They only had a couple of patients, I think three or four, who were on bisphosphonates, but uh this was not a huge confounder for them. Okay, the intervention. For a period of two years, patients in the intervention group received a once-daily oral tablet of 360 micrograms of MK7. This is a dose that has been studied in previous trials. And here they're telling you who produced it. While patients in the control group received placebo tablets similar in shape, color, and size. And again, this dose was based on the results of previous studies. Patients came to the hospital every six months. Baseline, 12 and 24 months, they gathered information on tablet intake, adverse effects, and collected blood to assess the vitamin K status. And they also repeated CT scans to quantify the extent of coronary calcification, all using the same equipment. I'm not going to go through the CT scanning protocol in detail. Please know they were also doing CT angiography, so some of the protocol is relevant to that. For example, patients were pre-medicated with a beta blocker to achieve a stable heart rate or nitroglycerin. And first, we determined calcium score on a native scan, contrast material. Here they're talking about how they score these. Calcium scores were determined according to the Agatha method, which considers both the total area and maximal density of calcified areas. We also measured total calcium mass, which is less variable and operator dependent than the volume score. The coronary artery tree was analyzed for the presence and severity of coronary disease. Plaques were defined as visible structures within or adjacent to the coronary artery lumen and quantified by a semi-automated analysis. I'm clear exactly what that means. Dedicated software algorithm, radiologist and cardiologist, both highly experienced in the assessment of coronary angiography, evaluated the scans, decided on the presence and severity of lesions by consensus. So that means two readers were looking at this and they decided together. They were blinded to clinical information and treatment group. All right, and they're using the 2022 CAD RADS system, which is just a standardized way of reporting the outcomes of a CT angio about how severe the disease is. Okay. Assessment of vitamin K status. They did this a couple of ways, one of which is important. They measured plasma levels of vitamin K1 and MK7 to assess the bioavailability of the vitamin. And they also measured plasma levels of dephosphorylated, uncarboxylated matrix GLA protein. Woof, that's a mouthful. Okay. Matrix GLA protein is the key molecule that helps vitamin K do what it's supposed to do, which is take calcium out of the bloodstream and put it where it belongs in the bones and not allow it to get into the walls of the arteries and into the aortic valve. This GLA protein typically circulates in an inactivated form, which is this devosphorylated uncarboxylated form, this uh DPUC, right? So when you take vitamin K, you should decrease the amount of the inactivated form. You don't measure the active form, you just measure a decrease in this inactivated form, and that tells you that the vitamin K you're taking is activating this GLA protein, which is what it's supposed to do, and that the inactive forms of this protein that are circulating, not doing anything, are gonna drop. Okay, so that's how they measure its functional bioactivity. This is some chemistry about how they measured things. And they are giving us some information here about how precise the assays are with the lab standards. That's nice. All right, outcome. What is the trial actually measuring? The primary objective of our trial is to assess whether oral MK7 supplementation would slow CAC progression after 24 months in patients with pre-existing CAC in comparison with placebo. Incidence of new calcifications, as specified in our protocol, was a secondary outcome measure. Notice what is not being measured here. There is no measurement of end events, heart attacks, strokes, revascularization like stenting or coronary bypass, none of that. We are just looking at imaging findings. All right, here's their statistical analysis. I'm gonna skim on by that, uh, except for this one part. I lied. We used several calculations to assess the evolution of the calcium score. These included continuous changes on an absolute scale, which is just what it sounds like, right? Your first score is 100, your next score is 150, a square root scale, a log scale, and a percentage scale. So these are all just different ways of adjusting the numbers to see if there's a way that this becomes significant. This is a little bit helpful because it's possible that you might see the change on a different scale. It's also a little dicey in the sense that you're just doing multiple comparisons to see if something sticks. And you have to be careful with that statistically. You have to adjust for the fact that you're making multiple comparisons on the same data set because the more tests you run just randomly, the more likely one is to be positive, just by chance alone. So if you're gonna do that, you really need to account for that statistically, and they don't say that they do that. Um, and so that's a that's another little quibble that I have with their methods here. They also look at what they're gonna call fast progression. So fast progression was pre-specified as 15% or more annual CAC score progression relative to baseline. So that means going from a score of 100 to 120, right? That's more than a 15% increase. That would be considered a fast progression. You have to keep in mind, though, 15% is about the limit of detection for coronary calcium. So it's maybe that that's really fast. This is, I think, a standardized definition. I don't think they pulled this out of nowhere, but just be aware that it's um some progression at least. I don't know if it's actually fast, but it's definitely some and maybe uh within the realm of what's actually detectable by coronary calcium. Meaning if it's less than 15%, it may or may not represent real change. If you go from 100 to 110, is that real or is that just noise in the system? It's not clear to me. So uh fast progression is greater than 15%. We're gonna come back to that as we go through their results. Okay. In addition, we use the Hawkinson criteria, which is more than 2.5 units change per year on the square root scale to establish fast progression. So a separate definition, again, here of fast progression, and they did a regression analysis to adjust for the variables. All right, results. They screened a thousand and eighteen patients assessed for eligibility. A substantial portion could not be included for a variety of reasons, of which refusal to participate was the most frequent one. They don't really say why people refused, just that they didn't want to. However, the general characteristics of these patients did not differ from those who were included, meaning they don't think that this represents a source of bias. Ultimately, 180 patients were randomized, 90 per group. That's almost a 90% attrition. That's um that's substantial. 13 of them dropped out before any baseline measurements. So 167 patients entered the trial proper. 85 per patients received the MK7, and 82 received the placebo. And their baseline characteristics are summarized in the table. We're gonna go through that in a minute. Um, during treatment, eight patients in placebo group and 10 patients in the MK7 group dropped out, so that's pretty uh even between the two groups. So at the end of the trial, 75 patients from placebo and 75 from NK7 completed the entire follow-up period. Mean low density lipoprotein cholesterol levels in these patients remained stable during the study. And I'm gonna read these numbers out loud because they are important. Placebo group was 93 milligrams per deciliter at baseline versus 97 after two years. The MK7 group was 100 milligrams per deciliter at baseline versus 93 after two years. Statin use also remained unchanged. So if you're looking at this, you're thinking 90 to 100 is probably not optimal for patients with known coronary disease as a level of their LDL cholesterol. When you go look at the baseline characteristics of the intention to treat population, you're gonna see here that LDL cholesterol is 77 milligrams per deciliter in both groups. Well, that's actually much lower than the 90 to 100 we just saw. I am not certain about this discrepancy. It is not clear to me from reading the supplements or uh trying to read in between the lines about which group of patients we're talking about here, whether this is true. So you see that there's the 82 and 85. This was the original group that were assigned. It's possible, I'm gonna go back up here to the results uh section we just looked at, that if we're talking about this 93 to 97, maybe that's just the 75 patients who finished the complete trial. That's a pretty big difference for just a few patients who dropped out of both groups. So I think that this is probably a typographic error or maybe a data reporting error. Um, but it makes me a little suspicious about what else might not be quite correct in the rest of the study, and that is a red flag for me. All right, so this is just the flow diagram for who got enrolled in the study. I'm not gonna belabor that point. It just shows you exactly who was screened, why they were not included, and who was eliminated from the trial. If we're gonna look at who was discontinued or who discontinued treatment, um in the placebo group here on the left, we have uh five who discontinued, two for personal reasons, two for diarrhea, one for a stroke. And in the intervention group, we had two for personal reasons, one for diarrhea, one for atrial fibrillation. What we can glean from that is that there's not a huge side effect profile that would seem to be driving people to stop the intervention. Okay. The reasons for discontinuation are pretty much the same in both groups, and uh there's not a significant signal for any kind of side effect or adverse effect that would be causing patients to drop out of the trial. So let's go back to this baseline characteristics. Uh, age in the trial is about 60 years old. They have about 40% women in the trial, which is good enrollment. The average, or sorry, the median uh BMI is 28 with an intercortile range of 25 to 30, 31, pretty reasonable. Um, systolic blood pressure is in the 130s over 80s, so not perfect control, but probably a reasonable population approximation. A little over half of the patients had a diagnosis of hypertension. Here are the cholesterol numbers that we reviewed. Um, about 45 to 50 percent of the patients had dyslipidemia, and only four to five percent had a diagnosis of diabetes. The average glucose was 97 to 99. So not terrible. Uh creotinine, which is renal kidney function, was pretty normal. They do have, if we're looking here at the smokers line, they have a pretty substantial population of smokers in this study. Uh, 67% in the placebo group and 74% in the intervention group. Um, that's respectable and a little bit higher than I would say is common in a lot of newer studies. And again, maybe this is something that changed a little over the 10-year enrollment period, not sure, but I would say certainly that represents a bit of a higher average than I'm used to seeing in my patient population. Here we're gonna look at the medications. So uh 70, about 80% were on statin therapy. They don't say dose, they don't say whether they're modest, moderate intensity or high intensity, or which drugs that they were using. Um, 40 to 40 to 50% of patients were on ACE ARBs for blood pressure. 30% were on beta blockers, uh, which is again a little higher than I would usually see. Uh, 7 to 20% on calcium channel blockers and 14 to 18% on diuretics. The median calcium score was 135 to 145. So, you know, solidly in the positive but not heart attack equivalent range, right? So over 300 is what I would consider this. You might as well have had a heart attack for the amount of risk it represents, but over 100 is absolutely a solidly increased level of risk for those patients. Calcium mass, again, 25, 26. And then uh based on the CT angiogram, the stenosis severity is two from the CAD RADS category, and segment involved score, the number of arteries that have a narrowing is about three in each of these groups. So overall, pretty well balanced, but there are. Are some aspects of this population that are a little different than what I typically treat? Again, including the number of smokers and slightly lower rates of high blood pressure and elevated cholesterol. All right. Adherence to MK7 treatment was estimated to be about 80%. So that means 80% of the people took it most of the time. It's not entirely clear what they mean by adherence. Does that mean you took it every day? Does that mean you took it at all? They say that they assessed this at their six-month visits, but I'm not entirely sure what they mean by this and why it's estimated rather than something that they're reporting directly from the interviews that they did with their patients. Adverse events were uncommon and not different between the two groups. We already reviewed that as part of the diagram. Okay, so let's talk about plasma levels. Figure two shows the time course, plasma levels of vitamin K1 and MK7. In both groups, the K1 levels fell slightly. That's not here or there because they didn't supplement K1. MK7 levels rose significantly in the active treatment group. Great, that's exactly what we would expect, and not in the placebo group. So that is the middle panel here of figure two. In the blue, you're seeing the intervention group with much higher levels of MK7 in the plasma. The placebo group is in yellow across the bottom, totally flat line, no change, meaning they're not getting vitamin K supplements for somewhere else, right? They followed the instructions and are not getting additional medications or supplements. So that's great. What is a little curious is this next panel. On average, levels of DP, UCMGP, that's the inactivated form of this matrix GLA protein, rose in both groups, although significantly less in the treated group. And remember, this was supposed to go down in the MK7 group. We were supposed to see that their supplementation with vitamin K was going to activate more of this GLA protein, and so there should be less of the inactive pool floating around in the plasma. So the levels are going up in the placebo group, and they're just going up less in the MK7 group. So there is some impact here, but it's not going down. And I would have expected it to go down. Now they do say that the change in this level correlates inversely with those in MK7, but it's not really a profound correlation coefficient. So an R squared of 0.17 means about 17% of that variance is accounted for by this relationship. So that's not really impressive to me. Um, but what they are saying is like if you had a higher level of MK7, you had a lower increase in your uh inactive GLA protein. So this is a little strange. They're gonna come back to this in the discussion. We'll talk about it, um, but it may mean that the dose they chose was not sufficient. Okay, primary outcome. In the placebo group, the absolute unadjusted meaning CAC score increased from 145 at baseline to 173 after one year and to 214 Agathin units after the second year. So overall we went from 145 to 214. In the active treatment group, values were 135 Agathin units, and they ended up at 184 Agaton units after two years, respectively. The beta coefficient uh for difference in difference was 22 Agatsten units, and irrespective of what scale was used, all analyses showed that the evolution of CAC score over time was significantly less in the active treatment group compared with the placebo group. So, first of all, yay, vitamin K does something. Okay, we have been waiting years for some study to show that giving MK7 or vitamin K2 actually does anything for the cardiovascular system. I have never seen it ever demonstrated. So this is the first time we are seeing anything. And on the one hand, I am super excited by this. On the other hand, this is not a huge difference in coronary calcium scores, right? Our end score for the active group is 184 Agastin units versus 214 Agastin units in the placebo group. So is that a clinically meaningful difference? At two years, maybe not. Over many years, perhaps. We can talk about the effect of calcification on total plaque, and we'll get to that in the discussion. But I have to say, I'm a little like, on the one hand, I'm so excited because we finally have some evidence that the vitamin K I've been recommending to my patients for a long time is actually doing something, right? We always knew it should based on this mechanism of action, right? This GLA protein, all of the basic science that we had was supposed to, this was supposed to work. But I would get really annoyed when I would see people online talking about how vitamin K is going to take the calcium out of your arteries and put it back in your bones. It doesn't work like that, to be very clear, but it may slow the deposition of new calcium into the circulation. But we had just never been able to really convincingly show it in a trial, right? All we had was the epidemiology. So this is the first trial that's convincingly showing any impact at all, and that's great. The impact that it's showing, mm, me, me, okay, a little underwhelming at two years. But again, if we took this out to five years, 10 years, maybe we see a bigger impact. Um, is it the same impact we would see if these patients had optimal lipid control or I don't know, quit smoking? Uh, I don't know. That's a good question because truly, for this patient population, there's a lot of intervention that would be possible to help mitigate their risk. But here are overall your uh scores, right? The absolute score and the percent change in baseline. Again, they compared this about six different ways. The root, square root, the log, they did a bunch of these. They did not account for multiple comparisons, but they showed consistently, no matter how they measured it, that there absolutely was an impact on treatment versus placebo. Okay. Now they're gonna do some adjusted analyses, meaning they're going to account for age and smoking and blood pressure and LDL and all of the other things here. In the multivariable model that included all the pre-specified variables, only baseline CAC score and treatment group remained associated with changes in CAC score, meaning patient age, smoking status, none of this impacted the change in coronary calcium score other than their initial score and their treatment group. All right. They say adjusted beta coefficients were 1.1 agatin units for baseline and 19 for treatment group, meaning that yearly increase in CAC was reduced by 19 units during treatment with MK7. Again, 20 agatin units, give or take. Is that clinically meaningful? Nobody knows. All right, including the baseline levels of the inactivated form of this GLA protein did not alter the outcome. All right, now let's talk about the fast progressors, patients who had heightened progression of their coronary calcium score. This did not differ significantly between the groups. 60% in the placebo group versus 56% in the MK7 group after one year, and 65% for those receiving placebo versus 58% for those receiving MK7 after two years. And the same was true when we applied the Hawkinson criteria. So no matter whether you wanted to call it 15% change or using the Hawkinson criteria, however you define a fast progressor, there was no difference in rate of progression or the number of fast progressors, excuse me, between placebo and the treatment group. And that again is a little disappointing. I would have wanted to see that the fast progressors might get more benefit from MK7. Okay. Analyses of the calcium mass score yielded comparable results. I'm gonna kind of skim through that because most people don't actually track calcium mass routinely. Likewise, changes in mass score were significantly smaller than the MK7 group. When we restricted our analyses to only those patients who were deemed to be adherent to treatment or those who did not drop out, which is the usually per protocol or the as-treated groups, we didn't find any different as the intention to treat. So when you're analyzing patients in a randomized clinical trial, the appropriate thing to do is to analyze the intention to treat group, the original 80-something number of patients that were assigned to each group. And even though about, you know, 10 patients dropped out in each group, you actually analyze those patients with it in the intention to treat because that's the least biased sample that you have. That was the randomized group. When people drop out, they drop out for specific reasons, and that introduces bias. So the statistically appropriate thing is to do this intention to treat analysis. But everybody always wants to know well, those patients dropped out. Obviously, they're not contributing to the final data set. What about the patients who actually stuck with the therapy? That is the per protocol group, and you're typically allowed to analyze that as a separate, you know, sub-analysis or secondary analysis later. And it's nice when the results are consistent. Otherwise, there's a question about whether it's the treatment or whether it's bias in the groups that is causing your difference. So they did both, and the results are consistent in both analyses. That is a good sign for the design of the trial and the sense that this is not a biased finding based on who dropped out. All right. These are the same graphs just showing calcium mass instead of calcium score. We're going to skip over that. And then their last piece of results is the secondary outcomes. At baseline, all patients had visible coronary plaques. This is not a surprise, right? They all had a positive calcium score, so they should have visible plaque as well. In most cases, this caused mild stenosis or CAD rads category two and involved three segments or three portions of the coronary artery tree. None of the patients in this trial had high-risk plaque. The evolution over time of stenosis category and segment involvement in the two treatment groups is presented in a supplement. Overall, severity increased among about 41% of patients receiving placebo and 33% of patients receiving MK7, but this difference was not significant. Progression in the number of affected vessels did not differ between the groups either. And again, it's not clear that we would expect to see a big difference between because what we're really looking at is the ability of matrix GLA protein to change where calcium is deposited. But it shouldn't really impact the accumulation of soft plaque in the artery walls. It doesn't really have an impact on that as far as we know. Mostly this is about the calcification. This other table in supplement two provides more detailed information on types of lesions and their location. All three plaque categories, which is non-calcified, partially calcified, and calcified, increased significantly in all segments, but without significant differences between the two groups. Increases in CAC score correlated significantly with the number of non-calcified plaques that became partially calcified during the study, but not with any other types of plaque. Again, this is saying that what they're seeing are the non-calcified plaques becoming partially calcified, and that's what is increasing the coronary calcium score in these patients. Again, that's totally predictable from what we know of the stabilizational plaque, but it does raise the question: is preventing calcium deposition in the artery wall a good thing, right? Yes, coronary calcium score going up as a sign of disease progression is absolutely associated with a change in an increase in cardiac event rates, right? If you're untreated and your calcium score goes from 100 to 200, that's a bad thing and it suggests that your risk for an event is higher. But we know that there are things that could raise your calcium score, like some lipid lowering therapy that is in fact stabilizing your plaque and reducing your chances of having a heart attack, reducing your chances of needing a stent or a bypass. Because we're no longer putting down new plaque, we are just transforming the non-calcified unstable plaque into stabilized plaque. So if we are removing the body's ability to stabilize the plaque by removing its calcium, right? Removing the uh concrete trucks that are coming in and paving over these unstable potholes, is that a good thing? We don't know. And again, this is not an outcomes trial. This is just an imaging trial. So we're seeing a 20-point change in average calcium score between treatment and placebo. And it's unclear whether this is going to represent any reduction in event rates for these patients. Could it even be an unintended harm if they are continuing to deposit plaque that is now not stabilized, not paved over, and they might have more events. We don't know. So this is a very interesting, ongoing question, and the trial here does not provide any evidence one way or the other about that. Let's go through the discussion. The present data support our hypothesis that supplementation with MK7 during a period of two years attenuates or slows down coronary artery calcification in individuals with coronary disease. However, the overall effect is modest, and given that the percentage of fast progressors did not differ between the two treatment groups, the clinical significance of our findings remains to be demonstrated. It is uncertain, for instance, whether the relatively small reduction in calcium score translates okay, we talked about the reduction may not mean it reduces events. Indeed, studies with statins suggest that these agents increase calcification, thus contributing to stabilization of such plaques. Since statins inhibit the invivosynthesis of MK4, which is another isoform of vitamin K2, this calcification-promoting effect may well be opposite to what we observed in the present trial. Nevertheless, a recent systematic review and meta-analysis of studies on the association between statin use and coronary calcification concluded that the evidence is at best controversial. Moreover, CAC remains a powerful predictor of major cardiovascular events. Thus, before dismissing MK7 as a form of treatment with a potentially adverse outcome, we need more information on the effects of this vitamin on hard endpoints. Exactly. Heart attacks, revascularization. Only the results of a well-designed outcome trial can provide an answer to the question as to whether the effect of MK7 is beneficial. Since Agastin's score is determined by the product, progression of CAC may merely reflect increased density and thus more stability of measured plaques. However, we also measured calcium mass, which is a better marker of the calcification process than calcium volume, with lower interscan variability. As both the Agastin and the calcium mass score pointed in the same direction, we are confident that MK7 can slow coronary calcification. We also found that increases in CAC score correlated with the number of non-calcified plaques that became partially calcified. And that suggests that MK7 slows calcification in developing plaques, but does not affect already calcified plaques. This is what I said earlier. It does not change, it won't reduce your calcium score. It's not sucking calcium out of the plaques that already exist, that are already stable and depositing it somewhere else. Okay. Our data further illustrate MK7 had no effect on non-calcified coronary artery plaques, which is arguably the most important thing for events, right? That's what ruptures and causes heart attacks and strokes. But they say this is not surprising. So the development of such plaques is a multifactorial process in which vitamin K only plays one role. All right. They're going to talk here about K1, unlikely this contributed. To assess the functional effect of the administered NK7, we also measure plasma levels of this inactivated matrix GLA protein, given that an increase in this protein is associated with enhanced calcification and is by itself a risk marker of incident cardiovascular complications. The inverse relationship between changes in NK7 and those of DPUC MGP that we observed can be taken as evidence for a positive effect of MK7 supplementation on this biomarker. Nevertheless, the overall increase over time suggests that there was still some form of vitamin K shortage in the vascular wall. This is because the levels kept going up. They went up more slowly, they went up less than in the placebo group. But the fact that they went up, they are saying, the authors, suggests a form of vitamin K shortage in the vascular wall. It is conceivable, they say, that a more pronounced effect might have been observed with an increased dosage or extended treatment duration. Moreover, the efficacy of NK7 supplementation may be contingent on the presence of a genuine vitamin K deficiency, suggesting a differential response based on nutritional status. And finally, they acknowledge that with an 80% adherence and a 17% dropout rate, the true efficacy of NK7 may be underestimated. So overall, what they're saying is we might have seen a bigger effect if we had selected patients for sure who had deficiency, right? Very elevated levels of this GMP protein that was inactivated. We might have seen a bigger impact had we done a longer study and really, you know, repleted those levels, or if we had used a higher dose and really addressed what we're still saying is a deficiency in the wall. Or if more patients had been adherent or had stayed within the study group, right? All of these things might have led to a bigger impact than we actually see. Little is known about the potential effect of NK7 on the progression of coronary calcification. A recent Danish study concluded that in patients with no prior ischemic heart disease, MK supplementation for 24 months had no significant effect on progression in those with a baseline Agathan score less than 400, but that study included only men between 65 and 74 years of age with aortic valve calcification, and that was the primary endpoint. So they're saying our study population is a little bit different. They also mentioned that in the Danish study, patients received vitamin D, and that may have confounded the impact. They go through that the valvular calcification was actually the primary endpoint. And they say here, which is an interesting point, it should be emphasized though, that even in this trial, there was a significant reduction in the progression of CAC score in the treatment group when the square root method was applied. So in a very small subset analysis, right, when they uh do CAC instead of the valve, then they do this square root method, then they do actually see a change in progression. Additionally, CAC score progression was significantly reduced by the vitamin combination in participants using statins. Accordingly, the differences between our trial and the Danish study are less absolute than it may seem at first, which is a nice way to try to reconcile the idea that another study came out with a different answer. And they're like, well, but the answer's not that different, right? Because they actually do see a change in the calcium score when they use the square root method. And in the patients who were on statins, they did see more of an impact. And many of the patients on this troop uh trial were on statins. And so they are saying that even though it looks like these trials came out with discordant results, in fact, they're not as separate as they are. Finally, incidence of adverse events were remarkably low. Although we cannot under exclude underreporting, patients knew they were participating in a vitamin trial, and this may have been associated with a no SIBO effect or a predisposition to perceive side effects. So they mentioned some limitations. Uh, with the current methods, we cannot assess the presence of microcalcifications. It is possible, therefore, that potential effects of MK7 were missed. Furthermore, we don't know to what extent participants modified their dietary intake of foods rich in vitamin K. However, the fact that the vitamin K1 levels did not rise argues against that possibility. So that's a really nice control that they had. Moreover, the changes in plasma NK7 levels are too substantial to be explained by dietary changes. The present study is small in sample size, and the high number of patients who refuse to participate in the trial raises some concern about the generalizability of the results, right? If like 80 to 90% of patients refuse to participate, is that the same number of people who will refuse to take a vitamin K supplement? And although we cannot rule out selection bias, post hoc comparison of the individuals who did and did not participate revealed no differences. They don't mention a 10-year trial period for a pretty small clinical trial. I think that's also a limitation of the study. The fact that it there was confounding by the COVID-19 pandemic in there somewhere also is uh perhaps a bit of a concern. But generally speaking, this is a very profound addition to the literature here. So, in conclusion, again, MK7 has the potential to reduce new coronary artery calcification in people with mild coronary disease. It is inexpensive and safe, can be easily purchased. Whether this translates into clinical benefits yet, we do not know. All right, so the other thing that you're always asking me about in trials is um how is it paid for? This study was supported by the Dutch Hart Foundation from a grant, and uh Natopharma or Gnosis provided free delivery of the MK7 tablets and conflicts of interest. Uh, the lead author reports receiving grants from Gnosis and holding shares in the coagulation profile outside the submitted work. So there is a little bit of a conflict here, right? There is some uh sponsorship of the study in terms of the drugs or the supplements, and there is an author who is affiliated with that group who is in the role of lead author. But neither the Dutch Hard Foundation nor Natopharma had any role in the design and conduct of the study. Um, and this was presented at the European Society of Hypertension. So, what do you guys think about this paper? Let me know down in the comments. I think this is a very exciting new evolution, even though we don't really know what it means clinically. My take-home message for you is I have been recommending vitamin K along with vitamin D for my patients for a number of years because overall I think the risk is low and there potentially is some upside, even though we still don't have any evidence of hard outcomes benefit yet. I don't think that the cost is high and it's potentially helpful. That's my take on it. Let me know what you guys think in the comments below. Until next time, take really good care.