Are ketones dangerous? The answer to that question is it depends. I think it’s important to note that ketones still may pose a risk for those of us on a ketogenic diet, pursuing nutritional ketosis. I have blogged a lot over the years about the benefits of a ketogenic (low carb, high fat) diet for type 1 diabetes. I even wrote an eBook about it. I began my dietary experimentation after 15 years of living with diabetes so my discussion of this topic often assumes that readers are also aware of their own limits and responsibility to establish their limits before inviting complex experimentation. As exciting as the results of a ketogenic diet are, risks must be understood in order to proceed safely.

Ketones can indicate diabetic keto-acidosis or DKA, which is life-threatening insulin deficiency, for someone with type 1 diabetes. Sadly, far too many people with type 1 diabetes die from DKA every year. Although the risk is pretty low on an annual basis, there are large numbers of people who actually die from DKA. So, you can easily see why the medical system believes that ketones (or the condition it is associated with) can absolutely can be dangerous.We all enter into experimentation with ketosis having had the “ketones are dangerous” doctrine drilled into our heads from the time of diagnosis. When you have type 1 diabetes, assuming a standard carbohydrate driven diet, the appearance of ketones in any concentration suggests DKA (diabetic keto acidosis) which is life threatening! DKA occurs as a result of insulin deficiency and correlates with elevated blood glucose. In response, usually over a period of hours, the liver dumps ketones which accumulate in the bloodstream at levels approaching 20 mmol/l . At this level of concentration the blood acidifies and the symptoms of DKA set in. This condition requires insulin to slow the production of ketones and correct the elevated blood glucose.

The problem is that ketones in a standard carbohydrate diet only appear in response to this critical insulin deficiency and so their association is purely negative and tantamount to danger. The idea of pursuing ketone production on a much lower level as a means of health and energy isn’t even in the picture. If I go to the hospital as a Type 1 Diabetic, I have to be very careful to let them know that I am in nutritional ketosis NOT DKA. Often times the mere presence of any ketones is a red flag since it’s the primary marker for DKA. Few nurses or doctors will say, “Yes but how high are his ketones? Only 3 mmol/l? Oh–he might not be in DKA, we should see if he’s in nutritional ketosis!” They are not testing to see how high the ketones are; they are only testing for presence or absence.

To be honest, there are SO FEW people pursuing nutritional ketosis that it is unrealistic to expect that every medical provider will be “hip” to T1D and/or nutritional ketosis. Frankly, even many national experts on T1D believe that nutritional ketones might cause DKA in high enough doses. Whether this is true or not remains to be seen, but you get the idea of the challenges a keto-friendly T1D person faces when intersecting with the carb-burning non-T1D world. Here is an example of the kinds of “cognitive dissonance” you should expect in the mainstream medical world:

I got a concerned email from a person who had been hospitalized with misdiagnosed “Keto Flu” after starting the ketogenic diet. Her blood glucose tested in normal range but feeling the flu-like symptoms that often times coincide with fat adaptation (requiring electrolytes and salt) alarmed her. She decided to play it safe and at the emergency room the staff only focused on her type 1 diabetes and the presence of ketones. They treated for DKA when she was actually not in DKA but had been producing lower levels of ketones through nutritional ketosis.

I experienced these same symptoms when I first started on the ketogenic diet (many people do-it’s called the “Keto Flu”) and I had the same concerns of DKA. I did some research and discovered that this was commonplace and not a danger so long as blood glucose was within acceptable range (which it was). I drank some bullion soup to get some extra salt, went to bed early and the next day the symptoms were gone and I was adapted to burning fat as my main source of energy! Still, it can be a tough call to make given the consequences and prudence is always the better part of valor.

Nutritional ketosis introduces ketones at a low level the body can use for energy and at a concentration that does not acidify the blood. Nutritional ketosis usually refers to ketone levels between 0-3 mmol/l whereas DKA sets in around 20mmol/l. However, without the black and white back drop of “no ketones=good, some ketones=danger” things can get a bit murky. When inviting an increase in blood ketones through nutrition it’s important to note that this CAN contribute (even if only slightly) to slight head start for DKA in an insulin deficient scenario. In the fairly extensive research I have done about the question of “are ketones dangerous” I have not heard of anyone being able to generate anything even close to 20 mmol/l of ketones through nutritional ketosis. Still, that does not mean it’s impossible or that we can afford to be complacent.

Many studies have associated insulin pump use with increased rates of DKA. The convenience of pump use may sadly come at a cost. If you are concerned about DKA and pumps, one possibility is to use long acting and a pump at the same time. This is one of many reasons I choose not to use an insulin pump. Bad sites, occlusions, cooked insulin from hot yoga, kinked tubing, empty insulin cartridges or dead batteries can all put you in a situation where you’re headed for DKA and if you’re already in nutritional ketosis–that very well may serve to grease the rails! If you already have a higher blood ketone level from nutritional ketosis it’s also likely that you can go into DKA at a lower blood glucose level since less extreme highs and less duration of time would seem to be necessary to increase ketones to 20 mmol/l–a point that’s relevant regardless of your chosen method of insulin delivery.

The flip side of this is that vigilance and attention to blood sugar will help interpret the appearance of ketones if you’re prepared for more nuanced information. Nutritional ketosis alone (assuming “normal” titration of insulin and blood glucose levels) will not raise ketones to the point of DKA. Also, when on a low carb diet, blood glucose excursions are much smaller and much less frequent. Assuming the insulin one takes is reaching it’s delivery target, the likelihood of going high enough to enter DKA is very low. Awareness of the change in risk management strategy is key-specifically in the first week or two of adapting to a ketogenic diet.

I’m not writing this to scare anyone or to back-pedal from my usually laudatory discussion of ketones. I’m sipping coffee with butter and beta hydroxybutyrate in it and preparing to go to the gym for a fasted training session later this afternoon. As beneficial as ketones are when understood fully and used properly, they are equally dangerous when approached without proper preparation and caution. I believe that there are not enough disclaimers in the world to replace common sense and individual responsibility. I am frequently torn between a desire to share the nearly magical properties of ketosis and to keep silent about them. Attention to detail is imperative; the ketogenic diet isn’t something you wake up one morning and do haphazardly.

The keto-T1D community is a small slice of a small fraction of the US population. When you enter the carb-burning non-T1D world that most of us live in, be prepared to experience some misunderstanding.

I posted a photo on my facebook page recently discussing some of the benefits of ketosis relating to hypoglycemia. I received a comment complaining that I should offer more disclaimers or else people would take my post as being “prescriptive” leading to dangerous outcomes for the uninitiated. I pointed out that literally the first words of my post read: “First, a couple of disclaimers”

To this the commenter replied, “Oh well it’s on Facebook, so I can’t actually take the time to read anything but the teaser text”

Again, the first words were a disclaimer. Despite all of my best efforts, there will never be enough precautions to replace simply reading and researching. Common sense. Personal responsibility. Still, I wanted to have some information available for those willing to read about the points of caution that I have adhered to in my experiments. I am a big believer in ketogenic eating for type 1 diabetes but that doesn’t mean I throw caution to the wind, much less advocate for others to do so.

  • Go gradually. Test often. Be aware of how your body feels. Insulin puts the brakes on ketone production; if you’re having second thoughts, eat carbs and take insulin.
  • Cut carbs and replace those calories with fat-gradually, over time. Cut carbs by 20% every week or so until you’re fully in the ketogenic zone (less than 50 grams daily). Once you’re comfortable eating around 50 grams of carbs or fewer each day, then experiment with shifting calories from protein to fat while testing for ketones. A slower approach means smaller errors and smaller corrections to the process.
  • Going from high carb to low carb may take a month. Going from low carb to fat adapted (ketogenic) may take another month. Patience and a methodical approach will help it “stick” in the long run.
  • When shifting the majority of calorie intake to fat sources (ketogenic diets are usually a 75/25 ratio of fat to protein) prepare to supplement electrolytes (salt will do in an emergency) with magnesium, potassium and sodium. Failure to do this will lead to the “Keto flu”. I learned this the hard way-stay hydrated and up the electrolytes!
  • Don’t try to workout with any intensity until you’re fully past the point of fat adaptation. It can be hard to resist the feelings of stable blood glucose and energy, but be patient and give it a couple of weeks. If you must, do very limited, low intensity exercise like walking. This helps you deal proactively with changes in insulin sensitivity and allows your body to learn this new “metabolic language” gradually.
  • Exogenous ketones can make a good thing better, but personally I did not try them within my first year on the ketogenic diet. I believe that the benefits are fantastic but the risks are much more significant given my type 1 diabetes. Being totally solid on a ketogenic diet is an important step to really understand how your body responds and if ketosis is something you want to venture deeper into.
  • Never get complacent. It only takes a skipped dose of insulin or a pump site gone bad overnight to put you in range of DKA. This is one reason why a full on keto diet might not be a great idea for teens or those who are more likely to be distracted.

It’s not all bad news though. Plenty of people manage the risks of ketones with no problems. Plenty of people drive their car to work despite the obvious risks that require management. It’s important to point out that a ketogenic diet is a low carb diet but not all low carb diets are ketogenic. Simply eating lower carb without getting into nutritional ketosis can be a much simpler way to gain 70-80% of the benefit of glycemic control with only 20% of the effort–but that’s a topic for another post!

The takeaway from this discussion is that DKA kills many people, year after year. Don’t be complacent. The benefits of nutritional ketosis do not justify a lower level of vigilance against the dangers of diabetic keto-acidosis.