152: Why Your Ketone Readings Don’t Match

Do you measure your ketones? Here's something a lot of people don't realize. Not all ketone tests are measuring the same thing. Urine strips measure acetoacetate, which can be useful early on but becomes less reliable as your body gets better at using ketones. Breath meters measure acetone, a byproduct of ketone production, but not the main ketone circulating in your blood. Blood meters do measure BHB, the dominant ketone your body actually uses for fuel. And now continuous ketone monitors measure BHB over time, giving you a full curve instead of one isolated reading. That matters because a single number can't show you what happened after a meal or overnight or throughout the day. So the real question isn't just, am I in ketosis? It's, what is my body actually doing over time? This is Lecture 152 of the Metabolic Classroom. Welcome back to the Metabolic Classroom. I'm Ben Bikman, metabolic scientist and professor of cell biology. Today we're going to talk about how to measure ketones. That, of course, is relevant for anyone who is interested in metabolic health and understanding metabolic flexibility and fuel use. Three established consumer methods have been around for years. That's urine strips, breath analyzers, and the finger prick blood meters. But now there is also a fourth, which is the continuous ketone monitor, and that represents a pretty new category and one that I think changes what's possible when it comes to metabolic measurements in ways that the older tools are just limit— too limited to really help with. Collectively though, the four approaches do differ in ways that go beyond convenience or price. They differ in what molecule they're measuring and including when they measure it and how reliably the number on the screen or the number that you're given reflects what's actually happening in your blood or in your circulation. So to get things started, uh, let's discuss how to make sense of some of these methods. And in, in order to do that, we need to just have a brief primer or review of physiology and some biochemistry, which is just to review what the body actually produces when it shifts towards fat-based fuel metabolism, because the four commercially available testing tools don't all measure the same molecule. And that distinction is pretty important, and it's an important one to keep in mind as we work through the strengths and the limitations of each of these approaches. When the liver ramps up ketogenesis in response to low insulin— of course, that could be low-carb diets or fasting— we have the increased delivery of fats from adipose tissue to the liver. So again, that has to happen when insulin is down, and then the liver begins burning all that fat in the midst of the low insulin and turning some of that fat burning into what is commonly called ketone bodies, or we can just call them ketones. Now, only— they're not all true ketones in the strictly chemical sense, but we're just gonna lump it into that. The first of these ketones to appear is acetoacetate, or what we could call the mother of the ketones. That is the direct product of, of that ketogenic pathway, and it's the precursor to the other two ketones. From acetoacetate, we have an enzyme called BHB dehydrogenase, and that generates beta-hydroxybutyrate, which we usually shorten to be just BHB. That becomes the dominant circulating ketone species during sustained ketosis, and it really does Uh, it accounts for about 70% of any energy that the body's getting from ketones. So when you think about the ketones that are, are burning, BHB is the majority of that. But there's a third member of this ketone trio, and that's acetone. It's formed by a spontaneous decomposition or breakdown of acetoacetate. So it's not enzymatic. You do have that one enzyme, the BHB dehydrogenase, producing BHB.. But then the acetoacetate can also just break down spontaneously into acetate. It's very small. It's— or sorry, into acetone. It can break down into acetone. It's small, it's volatile. And because of that unique chemical structure, it can be exhaled through the lungs. Now, here's the important point in the midst of all of this description. Each of these three molecules ends up in a different compartment, if you will, at different proportions. And each of these commercial measurement methods will target a different one with, uh, one redundancy. So the urine strips will measure acetoacetate, the breath analyzers will measure acetone, and then the finger prick blood will measure BHB, and the continuous ketone monitor the newest entry to this field, will also read BHB, but not from the blood like the ketone meter does, but from the interstitial fluid. So that kind of, that watery space that surrounds your cells, but it's outside your blood vessel. That would be the space just under the skin, for example. None of these four measurements is wrong, um, within its own sphere, but they are not interchangeable. Um, they're not measuring the same thing in the same places. And this mismatch between what each one detects and what is actually happening in your blood is the greatest source of confusion when someone tries to interpret their results. As we get into the distinct methods, let's begin with the oldest and least expensive method on the market, which is the urine ketone strip. The chemistry behind the strip has been in clinical use for more than half a century, and it relies on a simple reaction between acetoacetate and and a compound called sodium nitroprusside, which produces a purplish violet color that you can then compare against a reference scale that you're given on the bottle of those urine strips. The intensity of the color is roughly proportional to the concentration of acetoacetate in the urine, although the readout is at best semi-quantitative. You are just left to matching shades of purple by eye, just by what you're seeing, which means that two people, uh, examining the same strip could reasonably arrive at two different conclusions, but it would be roughly in the same range. The advantage is that the strips are very inexpensive. They're well under just a few cents per test. They don't require any equipment beyond the strip itself. They produce a result fairly quickly, within just a few seconds. And in the early days, of a ketogenic state. When, uh, when a person is first adopting, say, a low-carb or ketogenic diet and they're generating a lot of ketones, uh, specifically at a rate that it may exceed what the body is using, so the relative amount of production to use is going to be high, that strip can really give a good, strong, immediate visual confirmation that something's happening, that the liver is producing ketones and that some of them are spilling into the urine. Now, the limitations, of course, are relevant too. The first and most consequential is that the strips read the— not the wrong molecule, but not the main ketone. Acetoacetate is not the dominant circulating ketone. BHB is. Because urine acetoacetate reflects what the kidneys have filtered and excreted, not what is currently fueling your body, the urine readings will always lag behind the actual metabolic state when it comes to ketones, often even by hours. So urine output also is affected by hydration. A well-hydrated person will dilute their ketones in their urine, and so the reading will be a little lighter than you would expect. But a dehydrated person will be concentrating their urine. They'll have more acetoacetate in the urine that they're producing, not because they're producing more acetoacetate, but because they're not making as much urine. So it's more concentrated, and it lead them to a conclusion that their ketones are actually higher than they are. But I mentioned this second limitation, which is this keto adaptation, that when you first start adhering to a ketogenic diet, you're making more ketone than you're using. But over time, the muscles, the brain, other tissues will start using those ketones more. And so the actual amount of acetoacetate that ends up in the urine will drop dramatically, even though your circulating ketones may be just as high as they ever were. In one study of women on a very strict ketogenic diet, these urine sticks failed to detect ketosis in nearly half of the cases where blood BHB was clearly elevated and confirmed through other means. So you have a pretty high false negative rate, and it climbs higher and higher over time. This is the reason that so many people on long-term ketogenic diets become convinced that they've fallen out of ketosis when in fact they've simply just become more efficient at using the ketones. So what to take from this? For someone who just wants to have a yes or no signal in the first few weeks of adopting a ketogenic diet, the urine strips are perfectly fine, if perhaps a little inconvenient. After all, You need to take them into the bathroom with you. The second non-invasive option, and one that has become more accessible in recent years, is the breath ketone analyzer. These devices target acetone, that third member of the ketone trio, which forms by that, that spontaneous breakdown of acetoacetate. Now, because of its unique chemical structure, as I noted, it's just eliminated through the lungs and it is a waste product. The body does not use acetone in any way for energy. The most rigorously studied consumer devices will use a metal oxide little semiconductor, this little sensor that is selective for acetone. So the, the user performs a very controlled exhalation into the device, and that is a point of some problem that you have to be very consistent in how much you're breathing out. And then the meter will report breath acetone in parts per million or in some other measurement. Now, the appeal here is obvious. The device is reusable, so the cost per measurement gets ever closer to zero as you're— as you just continue to use it. It's a— it was a fixed cost, and so every time you're using it, it's a little cheaper, if you will. It is genuinely non-invasive. There's no finger prick, and there's not even a biological collection like it is with a urine sample. And unlike urine, breath acetone reflects systemic ketone state in something close to real time rather than what might have been filtered into the bladder several hours ago. Now, there are very real limitations here. Like urine, breath analyzers are measuring the wrong molecule. It's acetone in this case rather than BHB. So the reading is once again a downstream proxy for the dominant circulating ketone. The relationship between breath acetone and blood BHB is, is moderate at any single point in time. It's not bad, um, although with, with daily exposures, um, as you— if you try to find a direct correlation with BHB, it's, it's not great, although it's kind of in maybe an acceptable range. But what you could say is that as you're using the same device repeatedly and you're getting better, better with your breathing, it's going to give you a consistent directionality, at least in practical terms. The breath measurement gives you a reasonable estimate then of where your ketones sit on average across, you know, a day, for example. But the single reading does not really reliably tell you a precise BHB concentration at that moment. Now, the second issue— I've kind of alluded to this here— is the interference where the breath acetone sensors are vulnerable to other compounds that can be exhaled, like alcohol. If you've been drinking alcohol, that breath ketone analyzer is going to be heavily skewed. Even certain foods— if there's hand sanitizer residue, um, in the air or on your hands as you're holding your device, and other kind of ambient sources of, um, of these organic compounds, that can just lead to some skewing or mistaken measurements. I noted the exhalation technique is important. You have to be consistent in how much— like, how deeply you inhale before you have your exhale, how fast you're exhaling. Anyway, you can start to see how there are a number of variables that can influence the breath acetone or the breath ketone meter in a way that can make it a little fickle. Now, the next step up in both accuracy And Effort is the capillary blood ketone meter. And this just works on essentially the same principle as a glucose meter. You have a small little lancet and it will, um, poke your skin and it will draw a drop of blood from the fingertip. And then that blood is applied to a test strip and that test strip is loaded with an enzyme that reacts with BHB., and then the meter will read the electrochemical signal that comes from that enzyme reaction, and then it will give you a number, and that number will be in millimoles per liter. Because BHB is the species that primarily circulates, that's the main ketone. This is the measurement that most directly answers that, that practical question of just how much ketone is in my blood. Studies that have compared, um, several of these portable meters against laboratory very thorough, rigorous lab techniques do show good agreement across the full physiologically relevant range of BHB. The advantages are obvious with this one. It's a true quantitative reading of the dominant circulating ketone, and you can get it within seconds. For most purposes, verifying nutritional ketosis in, in a person— if you're wanting— wondering whether you're fat adapted or you're tracking the response to even an exogenous ketone supplement, like you, you take some GoBHB and you want to see how well it's worked, that finger prick meter is what, what has been considered a, a practical gold standard outside of the research laboratory. Now, the drawbacks are mostly practical rather than scientific. The strips are pricey, um, and which can limit how often you measure, but the procedure is also uncomfortable. It's invasive.. And if you're trying to measure your ketones multiple times a day, you're going to feel it on your fingertips. As a person who's done that, it can in fact get quite uncomfortable, but the result is also a little episodic, a single number that you get in that moment. Um, it's, it's not giving you any information as to where your ketones were an hour ago or in the minutes between that hour and the moment you're measuring. Um, and so because it also requires a very deliberate effort, like you have to make yourself bleed, uh, most people will end up testing just more when it's convenient rather than it might be, might be, uh, the most informative. And indeed they may end up not measuring as often as they would benefit from. Now that last point brings us to the most recent entry into the field, and that's the continuous ketone monitor. And that's actually the device that I personally have been wearing right now. My sleeves are too long for you to show me, but I have a continuous ketone monitor from SiBio that's made by SiBionics. That's the company. And I'm wearing one now. And to my knowledge, it's the only continuous ketone monitor available to the general consumer, although I know there are other manufacturers working on this as well and that are, you know, at various levels of development. So I'll refer to the Cybiosensor because that's the one I know, and that's probably the one some of you have used, um, previously because it is getting more common. That Cybiosensor works on the same general principle as the continuous glucose monitor. You have a small little filament that's inserted just under the skin, so that's what you kind of punch into your skin with the app— the applicator. It's actually often totally painless, but you put it up against the skin and you push the little button and it just pops it again into your skin. And, and that little sensor that's embedded into your skin just very shallowly, it is in that interstitial fluid space. And then within that sensor you have that enzyme, that same kind of enzyme-based electrochemical reaction that will target BHB. But of course it's getting it from that interstitial fluid, and then you get the electrical signal that is then going to be transmitted via Bluetooth to your phone, to an app on your phone. And then you can get a reading every few minutes. That's the difference. Whether you are deliberately checking or not, the device is sending a reading to your phone. So every time you open up the app, it can— let's say you haven't opened the app up all night and you wake up in the morning and look, it sort of, it will backfill, if you will, and you're able to look back in time to see the measurements all through the evening. Now, when you first put it on, the sensor takes about an hour to warm up., and then it's gonna go continuously for 14 days. Um, and the nice thing about these devices is that they come factory calibrated, so they're designed just to work straight outta the package. You don't need to do any kind of finger prick to calibrate it. Now, there are of course some very real and obvious advantages. The first one is just the sheer amount of data you're getting instead of maybe 2 or maybe 3 discrete numbers per day, like if you were using the blood meter. The continuous ketone monitor will produce near, uh, several hundred of readings every day. You can then see that full trajectory. You can see the pattern of your BHB rather than just any one single isolated point. The, the patterns that you get emerge from continuous data that you simply cannot get from those intermittent spot checks. And once you've seen those patterns, What to me is so valuable is that it becomes much harder, um, to make a mistake. Or rather, I'll say that another way, it becomes better at you being able to see these trends and responses. And the second, the second advantage is just the absence of a finger prick. Once you put the sensor in place, you can forget about it. And indeed, it's easy to forget about it. You don't feel any lingering discomfort or pain. And so that it removes a lot of the friction that limits how often you might test if it is just the blood prick method. And the third is that the device is measuring BHB, that same molecule as the finger prick meters, and it's the main ketone. It's the molecule you want. So the, the, the molecule that is being detected is the right one. The most, um, directly relevant data that we have on this with regards to peer-reviewed papers was a study presented at the American Diabetes Association meetings in which that SiBio, um, ketone monitor was compared against capillary BHB measurements in adults. And in this case, it was in people with diabetes, um, but I'm not giving any kind of medical advice here. And they found that the absolute difference between the continuous sensor and the finger prick reference was within just, you know, about 0.1, 0.2 millimoles., per liter. So pretty darn close, usually within about 20%, um, of a, a relative difference. Now you'd say, well, that's not perfect. Yeah, but the overall trend over time is, I think, what matters most. You having a difference of 0.1 or 0.2 millimolar matters less than you seeing what happened while you slept, or you seeing what happened the hour or two after you ate a specific meal. Now further, with regards to the evidence, there was a small randomized trial that used the SiBio CKM to track the response to an oral ketone supplement, and they found that the sensor captured the expected rise and fall of BHB in real time, but it also reported a gradual day-to-day decline in the area under the curve, um, over that period of time. Which might have been a little bit of drift, or it might have been just a bit of a delayed or a diminished response to the supplement. The device is very good at showing you the— or generating at all just the, the, a curve that you can see the shape of this curve within any given day at any moment. Even if the absolute number is off by 0.1 or so, it, I think the relative trend, that slow trend, that you see, is the most valuable feature. Now, further, I can say from personal experience that wearing the SiBio CKM, I'm able to detect changes in BHB, whether I have taken, a BHB supplement, whether it's the D form of the two enantiomers or even the L form, with the L form potentially being more of the signaling molecule. But I've taken LBHB, and I've detected a big increase in BHB levels. So it, at least in my experience, appears to measure both. Okay, a few additional points. This, um, the SiBio CKM, like any of them, is not currently approved by the, um, US FDA as a medical device. So if you're getting it, it's sold in the US as a wellness or, or a lifestyle product. So it's not meant to be used to diagnose or to treat or monitor any kind of disease state. You're doing it because you're curious. Also, like any interstitial fluid sensor, including a continuous glucose monitor, there is a, a small physiological lag relative to blood on the order of a few minutes. So during periods of rapid change in ketones, the displayed number is going to be a few minutes behind what's truly circulating in your blood. Okay, now what about just kind of more insight into everyday, uh, and I say this as a, as one who's been using these devices for a while, um, beyond the technical specs and the validation, I think that the more interesting question is what wearing the device actually shows you in everyday life, because the advantages of the continuous data become much more concrete once you have watched your own curve unfold over a few days. And there are a handful of patterns that emerge almost immediately for most people. The first is the overnight curve, which is more variable than people often assume. Ketones tend to climb during fasting hours, including sleep, but then in the early morning hours, there also, there is this, is this unexpected, maybe a little bump where you have some, uh, bump, I should say slump. Rather, where in the morning you have the counter-regulatory hormones like cortisol surging. That's the, the dawn phenomenon, and that can blunt, um, not only stop the ketone rise but even bring it down a little bit. The, the single finger prick reading that you take in the morning may, may just— it may lose all of the ketosis or the ketone spike that's been happening overnight. So you may measure it at a peak, or you may measure it at a valley, or anywhere in between. And there's really no reliable way to know without the continuous data. And then the second benefit, as I've alluded to, is the meal response. This is where I think the device delivers the value that no other ketone monitoring device can match. A meal that you may have assumed was clearly within a ketogenic range may produce, may actually suppress ketones over the next 2 to 3 or 4 hours, while foods that you thought might be problematic, may not really register at all. So an example of this is that you may take— I just saw an ad, for example, of keto-friendly donuts. I wouldn't be surprised if you ate a keto-friendly donut and you found your ketones went down. Now, on the other hand, let's say you baked a potato and you refrigerated it. Now you're taking it out again and reheating it. You may find that that does almost nothing to your ketones because there's such a relatively modest glycemic and insulinemic response. And the patterns that, that matter most in all of this in the end is just that these curves are deeply individual. You can have two people on the same diet, they eat the same snack or meal or drink, and they may have different ketone trajectories. They may not look anything alike even though they, they ate the same thing or drank the same thing. And that, I think, is the single best argument for measuring that with, with the CK-M is that it really gives you insight into your own response. Very analogous to partly of why I'm such an advocate of CGMs, that with a continuous glucose monitor, you're just able to get so much personalized information. And then with the CK-M, you're getting the same, albeit in the realm of ketones. So where does this leave a person trying to decide between methods? Yes, urine strips are cheap, um, they're disposable, just sort of binary yes or no signals. Generally, the breath analyzers are, are non-invasive, but of course they're, they're a little fickle. Then the finger prick is a very accurate method, but it is literally painful and it can start to add up. And then you have the continuous ketone monitor, and that is the tool that I think is best. It show— it's the only one that can really give you patterns rather than points., and it lets you just see how your physiology is actually reacting and behaving across very big periods of time. And for that sort of personal experimentation that you want, you just cannot get that when the data points cost a dollar and a finger prick in a fresh lancet every time. That's it, class dismissed. Until next time, more knowledge, Better health.