• Guest - w'd love to know what you think about the forum! Take the 2025 Survey »

Concern Over Euglycemic DKA & Low-Carb Diet With Type 1 Diabetes

T

T1DConquerer

Newbie
Messages
3
Type of diabetes
Type 1
Treatment type
Insulin
I'm a recently diagnosed T1D (been a year now since I was diagnosed, 24M), and while my last A1c was pretty good (5.4), lately the sugar has been often quite high no matter what I do. I'm tired of the roller coaster and want greater stability with my condition so I'm looking to go low-carb. I've picked up Dr. Bernstein's book and have been reading it, but I'm concerned about the possible increased risk of euDKA from a low-carb diet, especially the kind of carb-restriction Dr. Bernstein recommends.

From the studies I have read, euDKA is usually brought on by prolonged fasting (much reduced caloric intake) and often the fasting occurs as result of an infection or illness where symptoms of nausea, vomiting, fatigue, abdominal pain are present. And despite nominally taking sufficient insulin dosage (for example only long-acting or long-acting AND rapid-acting only from increased glucose due to illness), this amount of insulin is not enough to stop ketone formations which causes metabolic acidosis even at normal sugar levels (<200 mg/dL for diabetics). If I have any of this wrong or am leaving some key information out please correct me as I am not an endocrinologist or medical professional.

But could euDKA still occur in the absence of such restricted calorie-intake and simply just eating a normal calorie diet as it pertains to the individual but with low-carbohydrate intake (say <50g carbs a day for example?). Does a low-carb diet put you at increased risk for euDKA? Each individual is different so wondering if maybe for example, <30g carbs a day for some people might not be enough carbs to prevent dangerously high ketones from ketogenesis even despite covering meals with necessary insulin.

Would be awesome for low-carb type 1 practitioners to respond from their experience and knowledge, but all responses are appreciated!
 
Last edited:
Hi @T1DConquerer and welcome to the forums.

I've got to say that I thought euDKA was mainly caused by SGLT2 inhibitors (eg flozins) and hadn't heard of it being caused by keto on its own.

This is the article that comes up from google (which I'm guessing you've read)

So it looks like pregnancy (I guess not an issue for you ) and fasting are possible triggers?

I'm going to tag @Oldvatr who is quite knowledgeable about euDKA .

Have you tried joining the Typeonegrit facebook group (committed T1 ketoers) to see what they say about the issue?

I go low carb but not keto and increase carbs if ill to ensure that I'm not at risk of conventional DKA, so my experience probably isn't helpful.

But there are a few keto T1s here so hopefully they will post soon.

Once more welcome.
 
Hi @T1DConquerer

Researchers from my Diabetes Clinic recently published this short study on different diets for Type 1s and the effect on ketone production. This is a preliminary study, but the authors conclude that the levels of ketones in the low carb groups stayed lower than they had anticipated and below the clinical threshold for DKA.

I have been given a ketone meter to track my ketones and was recommended to test in the morning (after overnight fast). My levels do rise as I decrease carbs, but I've noticed that being dehydrated can also cause ketones to rise.

 
I am not aware of eDKA being associated with keto diets. The range of blood ketones due to ketosis rarely exceed 4 mmol/l (72 mg/dl) and even intense sport activity will not raise it much higher. eDKA is usually levels of 10 mmol/l and above. There is no real evidence of DKA of any flavour occurring with blood sugar levels lower than 10 mmol.l (180 mg/dl) which is not as high as normal DKA caused by insulin deficiency.

, So it is usual for T1D and sometimes T2D to experience DKA when the pancreas output is too low, but is nearly always caused by very high bllood sugar levels of 20 mmol/l or higher, combined with insufficient insulin. eDKA is different as it occurs with an apparently working pancreas that has been interferred with such as when taking certain medications. The Gliflozins are one med that can give this, but there seems to be some evidence that the new meds that force GLP-1 to override the normal beta cell operation may overdrive the beta cells leading to a loss of insulin output temporarily. So maybe if you are in the honeymoon period, then I suppose it might affect you like that, but unless the GLP-1 is being forced by some other mechnism, you would probably experience high blood sugars at the same tinme. so not euglycemic.

The use of keto diet by someone using insulin is not generally linked to eDKA, but will increase ketone levels naturally and this is a known and expected event. Obviously, there is risk of hypo and matching insulin dose to diet can be tricky, especially when carb counting, and I believe that keto insulin users find they may have to add in a bit extra to cover protein intake as well (I have seen 50% meantioned as a rough guide, but I am no expert in this)

Now, weight loss injections like Ozempic and wegovy are contraindicated for T1D, probably because of this increased risk of eDKA. But there are T1D using these meds, and so far any incidence seems to be rare. But apparently possible, as described in the StatPearls document linked by @EllieM previous.

To quote from it
"Therefore, conditions like anorexia, gastroparesis, fasting, use of a ketogenic diet, and alcohol use disorder can lead to states of carbohydrate starvation" So the operative word here is starvation. a proper keto diet will trigger fat burning mode, which causes gluconeogenesis to create glucose for the brain and vital functions, I suppose there is a risk that there may be other conditions that interfere with the glucose generation operations, but then glucose would be below the 10 mmol/l threshold for eDKA. And this is not being reported.

I did find this case study report for DKA in a Type 2 patient

Ketogenic Diet as a Trigger for Diabetic Ketoacidosis in a Misdiagnosis of Diabetes: A Case Report

www.ncbi.nlm.nih.gov

But he had bgl levels of 424 mg/dl on admission, so hardly eDKA. He was also re=classified as T1D on discharge-
 
Thanks for citing this great study
 

Appreciate your insight!

Just as a disclaimer, I don't claim to know all about physiology and human physiological processes, in fact, I'm still a newb when it comes to these things. But I've learned quite a bit in a short-time and want to be sure of all of the factors involved before I transition to a low-carb diet. Especially euDKA, how it can happen, and how it actually happens physiologically, whether or not the risks of euDKA are higher on a low-carb diet (which is probably true), etc.

I understand that during nutritional ketosis ketone levels of 0.5 to 3 mmol/L occur and this isn't really cause for concern in a non-diabetic person. If a type 1 diabetic for example was to go low-carb, he/she would also go into ketosis and have blood ketone levels within this same range (provided the carbohydrate is restricted enough to even go into ketosis). So long as this type 1 diabetic takes the insulin to cover the carb-intake, plus basal requirements, there shouldn't really be cause for concern, but it is more complicated than this:

I cannot directly post a link to the study below as I don't have enough posts to do so just yet, but traditional DKA (the hyperglycemic variety), although often associated with blood ketone levels 3 mmol/L and higher, can also occur with blood ketone levels as low as 1 mmol/L. This is why various leading diabetic institutions and societies put blood ketone requirements for a DKA diagnosis at >1.0 mmol/L for the sake of conservatism in diagnosis.

Per study:

"Our results showed that diabetic ketosis patients have blood ketone levels of 1.05–5.13 mmol/L, with 21.7% of ketotic patients having a blood ketone level of 3 mmol/L or more. If all the DKA patients (typical, atypical, and lactic acidosis-combined, n=255) were analyzed together, blood ketone levels ranged from 1.02–15.9 mmol/L, with approximately 30% of patients having less than 3.0 mmol/L."

So some of the patients were actually in diabetic ketosis rather than DKA and their ketone levels were anywhere from 1-5 mmol/L of which 22% were above 3 mmol/L and not in any DKA. This might be encouraging for soon to be/existing low-carbers. However, the fact that 30% of patients with actual DKA had blood ketone levels below 3.0 mmol/L is quite alarming.

Since DKA can also occur with blood ketone levels below 3.0 mmol/L, as low as 1 mmol/L even (well within nutritional ketosis), it might be safe to assume the same is true for euDKA.

And this goes to my main concern with euDKA is that it seems the risk of it occurring rises with a low-carb diet due to individual sensitivity and variability.

Say for example a Type 1 diabetic is eating a low-carb diet of a maximum of 50g carbs a day (I get that traditionally this is known as keto but there seems to be some marginal interchangeability, suffice to say, keto is certainly low-carb, but very-low-carb ) This individual is taking sufficient bolus insulin doses for every meal, along with basal insulin requirements. As this person is in nutritional ketosis, the metabolic processes of lipolysis, beta-oxidation, and ketogenesis are going to be more prevalent, but sufficient insulin dosage should be enough to prevent excessive states of the processes. Here's the kicker: What if it doesn't? What if for whatever reason this individual, despite nominally taking sufficient basal and bolus insulin cannot prevent excessive states of lipolysis, beta-oxidation, and ketogenesis in his/her body because personally for him/her the carbohydrate-intake is just too low, and so since these fat-burning metabolic processes continue to occur excessively, a dangerously high ketone concentration is formed in the blood negatively altering the acid-base balance of the blood to metabolic acidosis. What if even 70g carbs a day is too low? It's so individual specific.

Also while the above mentions insulin insufficiency (which it's important to note, was not readily known that it was insufficient likely due to the presence of normal glucose levels), as the primary reason for euDKA, does insulin insufficiency even have to occur for euDKA to happen as opposed to hyperglycemic DKA? That is for some people, do the blood ketone levels associated with nutritional ketosis alone result in negative changes to the person's acid-base balance, and therefore metabolic acidosis? What if some people for whatever reason have a reduced ability to clear ketones efficiently, leading to their toxic accumulation? I suppose the entire scenario described in this paragraph may or may not be labeled euDKA, but likely another form of acidosis, but labels aside, it's still a risk-factor with a low-carb diet, no?

There are a lot of factors here at interplay and it can be hard to discern the main culprit/s. I understand that euDKA is rare even on a low-carb diet and that it's primarily caused by prolonged fasting (whether deliberate or due to illness), among other reasons, but it's important to dissect everything that's connected, so not only me but others can understand what they are getting into with a low-carb diet as a diabetic.

What do you make of all of this?
 
Hi @T1DConquerer

I will be very interested to read the paper you mention above. Please post a link when you are able to.

What makes DKA and euDKA so dangerous is the resulting acidosis, not blood ketones per se. There seems to be numerous potential causes that can result in acidosis, elevated ketones is just one of them and as you suggest, in this study even if all the patients suffered from diabetes, ketones may not have been the source or sole source of their metabolic acidosis. Acidosis is dangerous for anyone, not just diabetics. What symptoms led to their hospital admissions? How were they treated?

My ketone levels typically run 0.5 mmol/L to very occasionally 5 - 4.0 mmol/L. Average 1.5 mmol/L. Mostly linked to my carb intake, but I will go higher after an especially high protein meal and/or if I'm dehydrated. Extra hydration is always my first action if I feel I'm developing DKA symptoms, especially vomiting, regardless of ketone levels.

Low carb way of eating (WOE) is recognised as a recommended therapy in Type 2 diabetes (but not in conjunction with SGLT2 inhibitors) and for certain forms of epilepsy. As you know it is widely practised as is intermittent fasting in the general population.

A point to note that ketone levels can be highest when first starting a low carb WOE. Many people find that once their liver is adapted they can find a balance between gluconeogenisis from protein/fat intake and their energy expenditure so that the ketones they produce are metabolised for fuel.

Are you still interested in trying low carb? Aside from the very rare risk of euDKA, being aware of adjusting your insulin doses appropriately will be important to avoid the increased risk of hypos.

Edit to add link.

Acidosis: MedlinePlus Medical Encyclopedia

Acidosis is a condition in which there is too much acid in the body fluids. It is the opposite of alkalosis (a condition in which there is too much base in the body fluids).
medlineplus.gov
 
Personally, I cannot get my head around why a T1D would want to use a keto diet anyway. i can see that carb reduction could help lower the swings and roundabouts and perhaps enhance control, but keto seems to be pushing it too far especially as you say, insulin users on a variable treatment regime adjust basal and bolus to suit, But fixed dose users should not attempt to go keto or ultra low carb IMHO.

i note that the study you quoted talks about lacto acidosis which is a different matter to dietary ketosis. Many T1D are placed on Metformin, which is known to increase lactic acid levels in the blood, so this could be a factor. intense exertion or exercise is another possible factor.

Also the study is talking low levels of ketones, but does not give any indication of what the glucose levels or insulin levels are at that time. and are these readings taken from admission states, or later. i suspect that maybe it is a problem with gluconeogenesis causing low glucose levels, and this may be related to other influences i.e. other medications or medical conditions.

one final point, and that is that starvation ends in eDKA, so if the body is not able to scavenge the building blocks it needs from protein or fat, then that is true starvation.
 
Oldvatr said:
Personally, I cannot get my head around why a T1D would want to use a keto diet anyway.
Click to expand...
Lots of non diabetics want to do a keto diet, so why not T1's? We're pretty much like normal people.

I think I would be very close to keto levels on most days if it weren't for the two high carb Belgian beers I have almost every day.
I have no idea if I ever have been in ketosis, and it's not something I care much about either way, but I do know that more than about 7 grams of carbs for my first meal will spike me, no matter how I dose. So I have a very good reason to keep my breakfast very low carb.
Other carbs I eat come mainly from lower carb veggies and condiments, and hummus for my midnight snack.
 
According to NHS advice, ketone levels of 3 mmol/l or higher requires an urgent request for a blue light taxi to A&E. Also a reading of 2+ or more on weestix is a similar emergency.

I am using SGLT2 med, and my pee regularly registers 8 on the stix, but I am not in DKA. it is also a regular feature of ketosis that readings up to 4 are normal, and can go higher through exercise. So is that why the NHS states that keto diets are dangerous?
 
T1D are normally at higher risk of DKA than normal people. The possibility of missing a dose can trigger it as can fasting. It is not so simple for a T1D to keep in ketosis anyway and adjusting doses can be tricky at low carb levels due to fat and protein interference, Keto is not just about carbs. T1D have higher glycogenesis due to Glucagon anyway.
 

Edit to add: One common trigger for DKA is urinary tract infection, which can cause the kidneys to stop removing / excreting ketones. Some meds like DPP-4 and SGLT2 can also interfere with kidney excretion which is why they are associated with euDKA.
 
Hi Oldvatr

Sadly I have been blue lighted into A&E with DKA, but ketone levels by then were well over 13 mmol/L! This was absolutely not due to a keto diet - back from SE Asia on a short trip, bit too much partying with close friends - way too much wine, vodka and pizza - by the time I figured the vomiting and breathing difficulties were more likely DKA than just overindulgence I luckily was able to have a close friend call 999. Relatively quickly fixed in resus, discharged to regular ward - next morning felt well enough to argue about discharge conditions - hospital wanted my ketones under 1.5 even though all I had eaten in the previous 36 hours was scrambled eggs which made that challenging.

I have had an interesting discussion with my endo - one of the authors of the paper I posted earlier - he is concerned about euDKA in Type 1s so asked what I thought about this - my answer was DKA symptoms are first priority - even if ketones aren't above 3-5 and BG levels look normal. This has served me well over the intervening years since the 'bluelight' episode as I believe I have been able to nip a few potential DKA incidents in the bud, primarily through aggressive hydration.
 
Oldvatr said:
T1D are normally at higher risk of DKA than normal people.
Click to expand...
As far as I know, this is only true when either running high or being ill.
Oldvatr said:
The possibility of missing a dose can trigger it as can fasting.
Click to expand...
Missing needed insulin doses will make you rise, which is a risk for DKA. Why would fasting trigger DKA in T1's? (Provided we keep our BG at normal levels (so still inject for the glucose we produce ourselves.)
I don't need to eat just because I'm T1, unless I'm hypo.
Oldvatr said:
adjusting doses can be tricky at low carb levels due to fat and protein interference
Click to expand...
I find dosing for protein and fats much, much easier than dosing for carbs. Keeps me level well below the prediabetic range.
Some others find higher carb lower fat more predictable to dose for.
There are T1's who are very good at dosing and timing for carbs, but most of the ones staying at non diabetic levels do so with the help of reduced carbs.
 
@T1DConquerer - Have you heard of Dr Ian Lake?

If not, Ian Lake is an experienced GP and long-term Type 1 who lives a keto lifestyle. He is active and has done a couple of extreme duration challenges - all on keto, or less. One he was fasted throughout.

I have had the pleasure to meet him on a few ocassions and he's a lovely knowledgeable guy.

His website is here: https://type1keto.com/ He's on social media, including X/Twitter, call it whatever it is called today. I wonder what his views are on euDKA. He'll definitely have a view on it.

Good luck with it, whatever you decide to do.
 
T1d have a higher risk of DKA due to having higher levels of glucagon. Normal people, and also T2D, when they eat a meal, will see insulin levels rise, and glucagon levels decrease or remain static. But paradoxically, T1D can suffer a marked increase in glucagon following a meal, and this triggers gluconeogenesis to generate extra than necessary glucose from fatty acids in the liver, and this process also creates ketones as a byproduct.

Ketone bodies - Wikipedia

en.wikipedia.org

This effect may be exacerbated by a ketogenic diet, and the kidneys may become swamped and unable to excrete the excess ketones.
 
Oldvatr said:
T1D can suffer a marked increase in glucagon following a meal, and this triggers gluconeogenesis to generate extra than necessary glucose from fatty acids in the liver, and this process also creates ketones as a byproduct.
Click to expand...
This implies that the extra glucagon increases BG, so it's still about high BG and DKA, not eDKA.
If BG rises above normal from either food or gluconeogenesis it means the T1 isn't adequately injecting to keep their BG in the normal range.

There are many T1s doing very well on keto type diets with hba1c's in the thirties whithout ever going into DKA.
 
I am aware that it is perfectly possible for T1D to follow keto. I am talking Risk, not certainty. But should all T1D be encouraged to go for keto?

The glucagon issue is the same issue that SGLT2 meds face, and they definitely give rise to euDKA in about 30% of DKA cases associated with that med. I too would expect the extra glugagon sourced glucose to be additive thus giving higher bgl. But if your keto diet is starting with a baseline of 5 or 6 mmol/l then adding to that may still be significant and yet be below the 20+ that normal DKA presents as. Remember that the glucagon runs while metabolic glucose sleeps, i,e, it is happening during fasting periods. It is the extra ketones that drop out of gluconeogenesis ;process that are the trouble, not the glucose.

We are all generating low level ketones due to gluconeogenesis all the time, and normally our kidneys can filter them out, For normal and T2D we actually resynthesise the ketones into COa and ATP for use in the mitochondria but that needs insulin to open the door for it to happen. This too is a risk factor at play for T1D and other insulin users.

So should we be saying to all T1D reading this forum that keto for T1D is hunky ~ dory or should we at least put up some caveats and warnings? at the end of the day, it is a nmatter of individual choice (unless you are institutionalised
 
But we T1's don't only inject for food, we inject based on BG and predicted BG as well. So there is no significant rise.
This thread is about eDKA, not about DKA with high numbers.
Also, most T1's are warned to be wary of DKA with numbers starting between 13 and 15, not only above 20.
Oldvatr said:
For normal and T2D we actually resynthesise the ketones into COa and ATP for use in the mitochondria but that needs insulin to open the door for it to happen. This too is a risk factor at play for T1D and other insulin users.
Click to expand...
Insulin users inject the insulin needed for this, it really isn't much different than endogenous insulin, if not exactly the same.
Oldvatr said:
But should all T1D be encouraged to go for keto?
Click to expand...
No, why?
We're all individuals, with our own preferred approaches.

But so far, nothing in what you have shared has shown that eating a keto type of diet, provided BG is kept at normal levels, increase the risk of eDKA.
The one risk I see is that when you're in ketosis you lose the easiest way to check for possible risk of DKA, namely testing for ketones.
 
As a T1 I've ;always been told that DKA was caused by insufficient insulin and that high bgs and ketones were a marker. (Admittedly, my consultant was horrified when I told her a few years ago that I'd never really tested for ketones, and promptly gave me a glucometer that could test for them. (And I haven't had a DKA in 53 years of T1, don't know if that is luck or not).

So, assuming no other meds than insulin, is a T1 actually more likely to go into euDKA than a non diabetic? (Assuming that a T1 is injecting insulin to maintain their bgs and therefore presumably not low on the stuff)


Non-medication causes of euDKA appear to be starvation, pregnancy or a weird metabolosm.. I'm not sure whether a T1 is more likely to have any of those than a non diabetic?
 
You must log in or register to reply here.
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn More.…