Are ketones insulinogenic (and does it matter)?
There has been a lot of hype around the interwebs about exogenous ketones and whether they are health promoting, particularly for folks with conditions that relate to excess insulin such as diabetes, traumatic brain injury, Alzheimer’s, cancer, epilepsy, obesity etc.
A couple of people asked me whether I thought exogenous ketones are insulinogenic.
Roger Unger’s 1964 paper the Hypoglycemic Action of Ketones. Evidence for a Stimulatory Feedback of Ketones on the Pancreatic Beta Cells indicates that ketone levels are controlled by insulin and that ketones suppress lipolysis:
Ketone bodies have effects on insulin and glucagon secretions that potentially contribute to the control of the rate of their own formation because of antilipolytic and lipolytic hormones, respectively. Ketones also have a direct inhibitory effect on lipolysis in adipose tissue.
It seems that exogenous ketones are indeed insulinogenic to some degree.
But how do we test this hypothesis to find out whether they are just slightly insulinogenic like fats or more insulinogenic like carbohydrates?
How to test the insulin response to exogenous ketones in someone with Type 1 Diabetes
If someone with Type 1 Diabetes stops taking their insulin both their blood glucose levels and blood ketones spiral out of control as they slip into ketoacidosis which can be dangerous and fatal before very long without exogenous insulin injection. In metabolically healthy people, high levels of ketones suppress mobilisation of body fat (lipolysis).
In someone with Type 1 Diabetes, taking exogenous insulin brings both ketones and blood glucose under control. So, based on what we see in people with Type 1 Diabetes, it seems logical that exogenous ketones would provoke an insulin response to keep ketones and glucose under control.
One way to test whether exogenous ketones are insulinogenic would be to have a Type 1 Diabetic to take a significant amount of exogenous ketones and monitor how much additional insulin they need to keep the continuous blood glucose monitor stable with the same amount of calories in glucose.
As an aside, I initially became interested in exogenous ketones after hearing a number of podcasts with Patrick Arnold and Dominic D’Agostino thinking that it may be a useful alternative source of energy that does not rely on insulin for my wife, who has Type 1 Diabetes. However, the one time she tried it resulted in such bad gut distress she never touched it again. So scratch that n = 1.
Food insulin index testing with exogenous ketones
Another way to test whether exogenous ketones are insulinogenic would be to run a food insulin index test  using ketones rather than food. This would involve giving 1000kJ of exogenous BHB (e.g. 48g of KetoCaNa) and measuring the insulin response over two or three hours.
The chart below compares the results of previous food insulin index tests undertaken for different foods. Comparing the area under the curve insulin response for the exogenous ketones to the insulin area under the curve for glucose would give you the insulin index for exogenous beta hydroxybutyrate.
I am surprised that the companies marketing exogenous ketones to people with metabolic issues like diabetes and epilepsy, as part of their due diligence, haven’t already done this testing to understand to what degree exogenous ketones are insulinogenic.
But wait, the food insulin index testing with exogenous ketones has already been done!
Then I came across this figure in a paper, Nutritional Ketosis Alters Fuel Preference and Thereby Endurance Performance in Athletes (Cox et al, 2016), where they have effectively done the food insulin index testing with exogenous ketones.
Thirty-nine athletes took an isocaloric dose of ketone esters, carbs and fat in three different sessions. In the chart on the right we can see that the ketones provoked about half the insulin response compared to the carbohydrate drink and a little more than the fat drinks. This test is different to normal food insulin index testing in that the participants started to exercise ten minutes after taking the drinks (i.e. at T = 0) at which point insulin and glucose start to decline.
Updated insulin load formula, including exogenous ketones
The chart below shows the relationship between the food we eat and our insulin response based on the previous food insulin index testing.
insulin load = carbohydrate – fibre + 0.56 * protein – 0.725 * fructose + 0.5 * exogenous ketones
It appears that exogenous ketones provide about half the insulinogenic impact of carbohydrates (i.e. about the same as protein).
Exogenous ketones stimulate insulin, but BHB also inhibits lipolysis directly via the nicotinic acid receptor PUMA-G in adipose.
While exogenous ketones may be equally as insulinogenic as protein, they’ll also be a counterproductive use of insulin.
Whereas the insulin response to protein is a positive use of insulin to build and repair muscle, with exogenous ketones, insulin simply reduces oxidation of other fuels to allow ketones to be burned.
Exogenous ketones displace the burning of other substrates. You know what else displaces the burning of other substrates? Glucose. Carbs reduce the amount of fat you burn. Similarly, exogenous ketones displace both fat and carbs/glucose.
That’s a double whammy in the wrong direction! Substrate competition is key.
Total energy = glucose + ketones
In a healthy metabolism, endogenous ketones are generated as fat stores are mobilised to compensate for a decreased energy availability from glucose. When glucose is not available, ketones come to the rescue to ensure survival.
If you’re insulin resistant, you might have trouble releasing free fatty acids due to the high levels of insulin circulating in your bloodstream. This inability to access your own fat stores will reduce your ability to create ketones and likely lead you to be more hungry and eat more than you otherwise would if you were insulin sensitive. If you are insulin sensitive you can more easily access your own body fat stores.
This chart demonstrates the concept of total energy (i.e. glucose + ketones) using more than a three thousand combined ketone and glucose readings from people following a low carb/keto lifestyle. Other than in the extremes of extended fasts or major feasting, the body seems to use insulin to maintain a homoeostasis of around 5 to 6 mmol/L of total energy in the blood.
On the left-hand side of the chart, when our blood glucose levels drop, we get a rise in ketones, but an increase in autophagy and all the good stuff that comes with fasting and ketosis.
On the right-hand side of the chart, when we drive our total energy high with excess energy (be it from processed carbs, Bulletproof Coffee, or exogenous ketones) the body releases insulin to stop stored body fat and glucose being released into our bloodstream.
People with the highest levels of metabolic health tend to walk around with a lower total energy in their bloodstream. It seems you don’t need to buffer lots of energy in the blood if you can easily mobilise body fat and glycogen stores quickly when required.
Having high levels of energy sitting around in the blood stream is far from ideal and leads to glycation in the case of high blood glucose levels and oxidation in the case of free fatty acids.
The total energy concept also seems to hold up with laboratory testing in rat pancreas islet cells, where exogenous ketone bodies promoted insulin secretion when there was greater than 5.0 mmol/L of glucose.
It appears that if your blood glucose levels are greater than 5.0 mmol/L (or 90 mg/dL), then exogenous ketones will be insulinogenic (at least if you’re a rat, but we have no reason to believe this wouldn’t occur in humans as well).
So if your blood glucose levels are greater than 5.0mmol/L (or 90 mg/dL), then those expensive exogenous ketones will be working just like a quick burning insulinogenic fuel, just like a dose of carbs.
Do exogenous ketones “help” with fasting?
If exogenous ketones raise insulin and reduce blood glucose, then where does the glucose go? It gets stuffed back into the liver.
Think about all of these people who fast with the intent of depleting liver glycogen but drinking Keto/OS. They’re literally preserving glycogen stores! No wonder we were seeing whacky glucose and ketone response to fasting with exogenous ketones.
Instead of the normal trajectory of a fast that would result in depleted liver glycogen we see exogenous ketones keeps this from happening, so you would get purges of glucose out of the liver throughout the fast when people were fasting using exogenous ketones.”
Let’s take a quick look at what Mike means by “the normal trajectory of glucose”. In the chart below, we can see how blood glucose levels drop and ketone increase in four people.
Where things get interesting is when you step look at the longer-term glucose and ketone trajectory of the fourth person who was taking exogenous ketones during the fast.
What’s causing this anomaly in glucose and ketone response? Is it a unique level of insulin resistance, or could this simply be explained by the use of exogenous ketones which are down regulating release of free fatty acids and endogenous ketone production?
One theory is that exogenous ketones are switching off lipolysis, which drives the liver to release more glucose and ramp up gluconeogenesis to fuel the system during fasting?
The addition of exogenous ketones seem to actually be suppressing the endogenous ketones production?!?!
The glucose : ketone index is the measure that Dr Thomas Seyfried encouragescancer patients to use during a fast to measure its therapeutic effect. The lower the better. For most people the GKI continues to drop during extended fasting, but in this case the GKI dropped and then starts to rise over time when taking exogenous ketones.
I would like to see some more thorough studies to understand if this is typical in people taking exogenous ketones when they “fast”. It’s not conclusive, but n = 1s are useful to build a hypothesis that can be tested in a more controlled environment.
Ray Cronise and David Sinclair recently published an intriguing article, Oxidative Priority, Meal Frequency, and the Energy Economy of Food and Activity: Implications for Longevity, Obesity, and Cardiometabolic Disease (2016) where they detailed the basis for the oxidative priority of different fuel substrates.
In terms of oxidative priority, it seems that:
- Alcohol will be burned first because the body has limited storage capacity for it. It sees it as a toxin that needs to be cleared.
- Protein will be burned second because you can only store a few hundred calories worth of amino acids in the bloodstream (though I think most people struggle to overeat protein when from whole food sources).
- Carbohydrate will then be burned before we can access our virtually unlimited stores of body fat.
So, I think it would be logical that exogenous ketones would be first in line (before or just after alcohol) to be burned off because the body has no way of storing the exogenous ketones other than circulating in the blood stream. Based on this logic, Craig Emerich has updated the Cronise chart to include ketones as the second fuel to burn off after alcohol.
The bottom line of all this is that exogenous ketones neither lower insulin nor promote fat burning. They’re just another fuel that will be burned before the fat on your bum and your belly! It seems that exogenous ketones are simply a fuel source, not a weight loss product as some people had hoped.
Do exogenous ketones boost exercise performance?
Exogenous ketones in sports performance is an interesting area of research. Rumour has it the Tour de France cyclists and British Olympic rowers are using ketone ester drinks (though it’s worth noting that the people spreading these rumours are selling the ketone esters).
Some people use exogenous ketones as a preworkout, like caffeine, to give them a cognitive boost. Research by Richard Veech and Kieren Clarke suggests that there may be a small athletic boost if you provide both exogenous ketones and exogenous glucose at the same time to provide a “dual fuel”.  This situation provides a fuel oversupply that would force the body to burn off the excess fuel quickly.
Dr Mike T Nelson suggest that driving a chronic energy surplus from high ketones and high glucose might be problematic in the long term as there is no precedent in nature for this condition.
I have dabbled with exogenous ketones (i.e. KetoCaNa, Pruvit and the Ketone Aid ketone ester). The chart below shows how my blood ketones rose to 3.5mmol/L and then back down to normal levels after about 3 hours. Note how my body tries to remove the excess energy from the blood stream and bring the total energy back down to around 5.0mmol/L.
I didn’t find a massive boost in performance in my workouts with any of the ketone products. My best performance is when I was fasted without supplementation and it seems I could easily access my fat stores and breathe more effortlessly.
I’m far from a high level athlete, but when I perform at my best cycling or in my kettlebell workouts, my breathing seems effortless and my time to exhaustion increases. When we are insulin sensitive and / or don’t have excess glucose in our system and burn more fat for fuel we use less oxygen than when we burn glucose for energy. This reduction in oxygen usage is critical to make sure you don’t get out of breath and fatigue. It seems that too much exogenous ketones or glucose in the system will mean that we’re less reliant on burning our own body fat.
I think the future of exogenous ketones in athletic performance will revolve around finding the right dose to boost ketones enough to get a performance benefit, without switching off lipolysis, which is where the real performance powerhouse lies. If you put in so much fuel in line in front of your virtually unlimited fat stores, then you may risk gassing out because you can’t access your fat stores as easily.
Perhaps someone who is a normal carb burner might benefit from having ketones raised to the 3 or 4 mmol/L range, while someone who is more fat adapted might benefit more with ketones in the 1 to 2 mmol/L range so as to get a dual fuel boost without switching off fat burning?
It’s still early days. Time and more experimentation will tell.
Edit: I understand from a friend who was involved as a test subject that there was a trial underway to test the effect of high dose ketone esters over a month. While they initially experienced a massive boost in energy and performance when they used the exogenous ketones subsequent doses morning and night day after day no longer yielded the same result. It seems your body becomes adapted and learns to clear the high level of energy quickly with no performance benefit. I have yet to see this research published though.
Does it matter?
If you’re metabolically healthy and you enjoy the brain buzz of exogenous ketones more than alcohol or caffeine and want to use exogenous ketones as a pre-workout, then I say go your hardest if you can afford it.
However, if you are looking for improvements in your metabolic health or magical weight, I think you should be cautious.
Companies like Ketopia are marketing exogenous ketones as a “bridge” through the keto flu, which I think is a more ethical approach (although many people say you can eliminate the ‘keto flu’ with good mineral supplementation).
But you probably haven’t heard of Ketopia, because selling seven days’ worth of exogenous ketones isn’t a great business model in comparison to getting people to sign up for an ongoing subscription as a distributor buying thousands of dollars of ‘inventory’ up front so they can take it… Every. Single. Day.
Perhaps when exogenous ketones are no longer the realm of over-hyped multi level marketing campaigns people will be able to experiment with exogenous ketones and find whether they live up to any of the claims and are indeed worth the money? This is starting to happen now with Julian Baker’s InstaKetones coming in at about one sixth of the price of Pruvit’s products.
Should you take exogenous ketones?
If you’re using exogenous ketones with the hope of reducing insulin levels or reversing metabolic disease (e.g. Type 2, cancer, Alzheimer’s, obesity), then maybe think again. Exogenous ketones may alleviate your symptoms while they’re in your system (about 2 hours), but I fear they might worsen the conditions that people are using them for in the hope of improved metabolic health.
Mike Sheridan’s article in T-Nation makes a number of excellent points:
Ketones may be depressing dieters’ hunger and giving them a hit of energy and cognitive enhancement, but it’s INHIBITING their ability to burn fat, providing zero nourishment, and doing nothing for their metabolic health. There’s an assortment of evidence suggesting that it’s probably making things worse.
Think of exogenous ketones kind of like alcohol. When they’re consumed, everything is stored and nothing else is burned. So any lipolysis (fat burning) that would be taking place is halted; any glucose and fatty acids in your blood that were circulating are stored; and the ingested ketones are burned until there aren’t any left.
But suggesting individuals already fasting, restricting calories, or cutting carbs will get anything other than a brain buzz is misleading. And to serve up exogenous ketones to an obese, insulin-resistant general population with promises of fat-burning and disease prevention is potentially damaging.
Sure, it might suppress hunger and give a damaged brain a useable fuel source, but what happens when pre-diabetic Pete starts adding ketones to his glucose-rich blood? Or anaerobic Andy continues reloading with the same amount of carbs post-workout even though the liver glycogen he normally burns during his sessions is now suppressed?
Exogenous ketones might provide energy to the muscles or brain cells of someone with Type 2 or Alzheimer’s who can’t use glucose well because of decades of hyperinsulinemia. But, if someone already has super high glucose and insulin levels, will they worsen the condition by chasing high ketone levels with large doses of insulinogenic exogenous ketones?
If someone is trying to shrink their brain tumour by reducing growth stimulating insulin, will ingesting large amounts of exogenous insulinogenic ketones have the side effect of accelerate growth in the brain tumours? It’s a scary thought, but recent studies suggest that this may in fact be the case.
At the current rate, it looks like we will be able to confirm the long term effects of exogenous ketones sooner rather than later.
But by then, the people running the MLMs will have driven off into the sunset and be on to another scheme.
Good news… endogenous ketones for free!
The good news is that all the benefits of endogenous ketosis is freely available risk free! It’s not easy, but you can get the benefits of ketosis (e.g. autophagy, apoptosis, increased NAD+, mitogenesis etc) by keeping the insulin load of your normal diet down to the point that you can maintain normal blood glucose and insulin levels, then occasionally you can push the time between meals than usual in order to derive some extra benefits (i.e. intermittent fasting).
Postscript… What to do if you really want to drop your insulin levels!
A couple of years on, and it seems that the hype around exogenous ketones is dying down a little.
Virta’s long term trial of their ketogenic diet approach has confirmed that people on a ketogenic diet tend to adapt to using fat for fuel and see their blood ketone levels drop back down to near baseline levels well below what most would consider optimal ketosis or even mild nutritional ketosis (though I have yet to see this fact discussed by Virta!).
We recently ran a weight loss challenge with the Nutrient Optimiser where some people tracked ketones on a high satiety diet. As shown in the chart below, the daily average blood ketone levels increased to around 0.8 mmol/L over the first three weeks of the challenge but then dropped to around 0.4 mmol/L as participants continued to lose fat.
Blood ketones took a little longer to rise for people who identified as insulin resistant and stayed higher for longer compared to people who identified as insulin sensitive.
However, as shown in the chart below, weight loss continued throughout the challenge, regardless of ketone levels and regardless of whether users identified as insulin resistant or insulin sensitive.
Fat loss does not seem to be correlated with:
- blood ketones levels,
- insulin sensitivity, or
- perceived or actual insulin resistance status.
What was most interesting was that we were able to measure insulin status in my wife Monica (who has Type 1 diabetes) during the challenge. The chart below shows Monica’s daily insulin dose during the challenge with a 14 day moving average (red line).
Her insulin levels requirements dropped significantly when she changed to a nutrient dense high satiety diet with adequate protein to promote satiety. Towards the right-hand side of the chart, we see that her insulin demand began to drift back up as her energy intake stabilised at a lower level, even though she was continuing to lose weight.
The chart below shows Monica’s weight (blue line) and daily insulin requirement (orange line) during the challenge plotted together. Weight loss tends to occur after a reduced daily insulin demand. Lower insulin levels allow stored body fat to flow into circulation to meet the energy needs.
It seems that insulin decreases in response to an abrupt change in energy intake, but may return to baseline levels once energy intake stabilises.
To provide some context, the chart below shows Monica’s daily insulin dose over the last ten years. Switching from a standard western diet to a low carbohydrate diet around eight years ago enabled her to decrease her daily insulin from 50 to 30 units per day. However, during the challenge, her insulin requirements decreased to the lowest levels ever!!!
How did she do it? Monica’s primary focus during the challenge was on prioritising nutrient dense, high satiety foods. The key difference from her previous approach during the challenge was a significant reduction in her intake of nuts, cheese and cream.
While high-fat foods cause a slower insulin demand, they still affect insulin requirements. Low carb foods like nuts, cheese and cream can be a good way for people managing diabetes to obtain adequate energy, they do seem to trigger a long term insulin response and focusing on higher satiety foods tends to reduce overall demand.
Overall, Monica was able to lose 7.5 kg or 10.7% of her body weight during the Nutrient Optimiser challenge (isn’t she cute! I’m a lucky man)!!
post updated July 2019