Keto enthusiasts -- Are you at greater risk of going into DKA without warning?

Type 1 and LADA
1

So I ask this question for those who are very low carb/keto – how do you make sure that you’re not wandering into DKA territory without realizing it?

I ask because we recently had a scare when our son was in the mountains skiing. We forgot to do the last step on his pump before bed and he was without insulin all night. He’d eaten relatively few carbs at dinner and because he was in the mountains, we think his muscle cells may have been taking in more sugar from the blood than they otherwise would have.

Anyways, long story short, he had a flat, perfect blood sugar all night long but woke up with high ketones. He was not sick – the only indication we had was when we went to look at his pump and realized it was still in the reservoir setup process.

This is the second time he’s tested positive for ketones. The first he also had not had elevated blood sugar in the hours prior to his ketone test – and he also had eaten relatively few carbs.

This made me wonder: the hallmarks for DKA are elevated blood sugar, lack of insulin and high ketones. But what if you can generate DKA without the first? Some organs in the body can be starving for insulin even if blood sugar levels are normal.

For people who are habitually putting themselves into a state of ketosis, how do they ward off the risk of DKA? It just seems like you’re living with less of a safety margin, closer to that tipping point, and my son’s situation shows you can easily develop high levels of ketones even without being sick and with perfectly normal blood sugar levels, so there could be scenarios where you’d start to show symptoms before you realize you are in DKA.

Do you just have a habit of testing your blood ketone levels two or three times a day? Do you check ketones any time you’ve exercised? How do you stay safe?

2 Likes
2

I agree with you about it seeming like there’s a lowered safety margin. I know that for people who are not at risk of severe insulin deficiency (i.e., anyone without full-fledged Type 1) there is no risk of DKA with a low-carb diet because ketone production is normal. But it seems to me that, if you are eating low-carb AND have full-fledged Type 1, especially those on a pump, then you do have to be more careful. If DKA is a ketone level of, say, 5 mmol/L, and you’re already at 2 mmol/L to start with, it seems to me that in the event of a site failure you’d be 3 mmol/L away from DKA rather than 5 mmol/L away from DKA. This just being a hypothetical example, but it’s how my thinking goes. The first time I tried a low-carb diet a few years ago, I had an incident where I had a bad pump site, felt really sick, and the ketone strip turned so dark it looked black. I had never seen it turn that dark before, and it totally freaked me out, so I stopped eating low-carb after that.

I’ve been low-carb to varying degrees for the past six months, much of that time eating less than 50 grams of net carbs per day. My understanding is that when eating low-carb, ketones are fine as long as blood sugars are normal. If blood sugars are high AND ketones are high, then you have to work to get both down. This happened to me when I had a pump site issue while eating low-carb: my blood sugar gradually rose to over 20 mmol/L over about four hours despite corrections, and I finally started feeling a bit sick and checked my blood ketones, which were 2.1 mmol/L. When that happened, I corrected with injections and changed my site and monitored until my ketones were back below 0.5 mmol/L. But at other times I had checked and ketones were, say, 0.8 mmol/L with a perfectly normal blood sugar and feeling fine, and that I didn’t worry about. I do think that with a low-carb diet you would probably always have high blood sugar as a warning. Going into DKA with a normal blood sugar can happen, but I think it’s fairly rare.

I’ve recently started on an SGLT2 inhibitor, which is approved for Type 2 diabetes but, since it doesn’t depend on insulin production, is potentially helpful for Type 1s as well. It works by stopping the kidneys from reabsorbing glucose. One of the side effects for people with Type 1 is going into DKA even with a normal blood sugar, since there is a mechanism other than insulin working to lower blood sugar. While taking this medication, my endocrinologist wants me to check ketones daily or if I’m ever feeling sick, even if my blood sugar is normal. I’ve been checking two to three times a day, and have run into moderate and high ketones a few times, which have gone away with drinking lots of water and taking extra insulin. Due to the fact that I would have no warning signs of DKA if I was on a low-carb diet while taking this medication (I wouldn’t have ketones as a warning since I’d always be producing them, and I wouldn’t have high blood sugar as a warning since it could be totally normal even while in DKA), I’ve decided to raise the level of carbs I’m eating daily so that I’m not producing ketones all the time. I’ve been taking this new medication for a week, and so far it’s been amazing for me. My insulin needs have dropped and my blood sugar is better on this medication than it was while eating low-carb.

Outside of the above scenario, though, I don’t think I would check for ketones daily unless there was some other special circumstance (high blood sugar, pump failure, illness, etc.) or unless your endocrinologist instructed you to do so. If you are concerned I would ask your son’s endocrinologist about it and see what he/she advises.

I’m sure others here know far more about the biology and physiology than I do. I’ll be interested to see what they have to say.

3

It is important to understand that most humans will produce ketones overnight and many will have observable ketones in their blood and even in their urine. Our bodies burn ketones overnight while we fast. And insulin is not needed for ketone uptake.

And it turns out exercise can enable a certain amount of glucose uptake without insulin. Some endurance athletes find they can literally stop basal and bolus insulin during exercise. It is quite possible your son simply had a normal fasted overnight. And there are basically no organs that will “starve” for glucose in the absence of insulin, they will all either burn ketones or use diffusion based mechanisms to take up at least a limited amount of glucose. And low levels of insulin isn’t really a hallmark of DKA, it is a factor that can cause DKA. Insulin downregulates the liver production of glucose and ketones. Generally in DKA insulin deficiency results in high levels of both glucose and ketone production. Without high levels of glucose you basically don’t have DKA.

That being said it is important to understand DKA and HHS. True DKA has a hallmark of elevated blood sugar, highly elevated ketones and dehydration. HHS is characterized by highly elevated blood sugar, elevated ketones and dehydration. The two conditions overlap and are related.

When we hear reports of euglycmic DKA it is quite possible that we are hearing reports of HHS. The SGLT2 drugs can cause dehydration and dehydration can cause insulin resistance which my cause a cascade leading to a condition which is diagnosed as DKA. But in fact it may be the dehydration that causes the problem, not that you stopped taking insulin. And another factor is also that the SGLT2 drugs cause you to excrete glucose which may “mask” a high rate of glucose production.

So in summary, I’ve never seen anyone (competent) suggest that being in ketosis or producing ketones is either unnatural or puts you at risk of either DKA or HHS. If anyone has studies which suggest this I would like to see them. I just have never seen any suggestion that there is a physiological mechanism that ketones can cause DKA. It is a logical mistake that since ketones are “associated” with DKA that people think they cause DKA.

You can talk with your doctor about his but they probably don’t have a strong enough understanding of low carb diets and ketosis. If you want to be safe then you would likely be better served by doing everything you can to make sure you son doesn’t have interruptions in his insulin.

4 Likes
4

Awesome responses! I’m not testing for high ketones on any regular basis but should my BG run over the top of my range despite correction I do take a look at ketone levels. I’ve been LCHF since this past July and have found my fluid requirements, as well as my salt requirements increased. I’ll feel sick to my stomach well before I’m over the top of my range if BG is climbing, which is alleviated quickly by drinking any non caffeinated sugarless fluid. If you are committed to a LCHF program a higher fluid intake will also be necessary to, ah, support normal bowel movement. Sounds like your son’s low carb intake was not part of his normal routine so a few hours of ketone increase without high BG would not indicate the kind of sustained metabolic event that results in DKA.

5

If low insulin levels don’t cause DKA and high ketone levels don’t cause DKA, then what causes it? According to my endocrinologist, it’s caused when insulin levels drop below the baseline that the body needs for basic functions. So insulin levels are “low” compared to what the body needs at that time.

I don’t understand how other conditions like HHS or dehydration could be misdiagnosed as DKA. Isn’t DKA caused by high levels of ketones affecting the acidity of the blood (ketoacidosis) caused by insulin levels that are too low in comparison to what the body needs as a baseline level (diabetic) and treated with insulin and other medications? Wouldn’t dehydration or HHS not have these very high levels of ketones and/or not have the very low levels of insulin and not have the blood acidity affected?

6

@Brian_BSC, the thing is his ketone level was 3.4 mmol/L, which is in the “high range.” While people here are saying that, given his euglycemic state, he was not at risk of “true” DKA, but instead HHS or some equivalent I see that as a kind of semantic splitting hairs. I mean we don’t care whether someone fits the criteria for DKA, we care whether they are at risk of getting very very sick. And I guess no one has given me a credible explanation for why someone who has not gotten sufficient insulin to down-regulate ketone production for hours is not at risk for runaway ketone production and resulting acidosis – even if their blood sugar is normal. Obviously, the advent of normoglycemic “DKA” with SGLT2 inhibitors shows that high ketone production, even with normal blood sugar, can lead to a medical emergency. And indeed, our son was complaining of a tummy ache and feeling sick at 3.4 mmol/L. There is no doubt in my mind that he got to that level of ketones as a result of not having insulin delivered via his pump for 10+ hours, that it had the potential to lead to severe illness, and that this was not a normal “fasted” level of ketone production.
I’d also argue that there are credible mechanisms to explain how this process occurs that do not rely on occult dehydration.

Basically, all I’m suggesting is that people who are on a low-carb diet be alert to the idea that they are closer to DKA and that it could conceivably occur in conditions that are normoglycemic, such as if they’re exercising, at altitude, getting sick, etc. If even people who do not have diabetes can get DKA on a low-carb diet (extremely rarely, but still theoretically possible), then we shouldn’t dismiss it as totally impossible in people with a documented, absolute insulin deficiency. Personally, if I was low-carb I would have some kind of random, spot-check blood ketone testing set up in my protocol, as well as testing daily anytime I’m exercising more than usual, or in circumstances such as being at elevation, coming down with something or feeling an unspecified crumminess. But that’s just me. I guess my tolerance is lower too because my son cannot tell me if he feels “ketone” sick, so there is one less safety mechanism in place.

7

[quote=“Jen, post:5, topic:58326, full:true”]

We think of ketones as “causing” acidosis in DKA but it is more complicated. While ketones are acidic, the body becomes vulnerable to acidosis when there is no buffer. So it is more appropriate to say that in the face of dehydration ketones can cause and acidosis problem. So while lack of insulin can help instigate the onset of DKA, it is just as much the loss of fluids and electrolytes which results in acidosis.

The differential diagnosis between HHS and DKA is that in HHS your blood sugar is higher and your acidosis less. Both involve high blood sugar and acidosis. HHS (and DKA) happens in both T1 and T2. These conditions present as essentially the same, the only difference is the degree of acidosis and blood sugar level.

To diagnose acidosis they test your Anion gap. Acidosis can occur with DKA (HHS) but can also occur as a result of dehydration. This is why I keep suggesting that anyone taking an SGLT2 drug also stay hydrated and make sure that maintain good levels of electrolytes. Dehydration alone can cause acidosis.

1 Like
8

Right. We forget this as diabetics, since we are focused on DKA (for obvious reasons), but in a former life when I was trained as a medical first responder, we were taught that acidosis is the primary danger from dehydration in all people.

Anyhow, I’m firmly in the nutritional ketosis is different from pathological ketosis camp, but I also have significant endogenous insulin production. So I’m pretty certain I don’t have an iron in this fire, at the moment. I have low BG and moderate ketones at any given time (due to a lot of exercise and a low-carb diet). I have been asked by my medical team to carefully monitor myself for symptoms or signs of DKA, but it doesn’t seem to unduly worry anyone other than my wife. I do make sure that I’m well hydrated and full of electrolytes at all times, though :slight_smile:

9

A blood ketone reading of 3 mM is consistent with a low carb diet of 20 g carbs/day (“Art and Science of Low Carb Living”). Blood during starvation go up to 5mM and these levels might be observed during fasts or after exercise. Acidosis levels are 15-25 mM.

While conditions leading to DKA may not depend on dehydration, DKA will present with severe dehydration. And the case you presented, this was a woman that presented with severe dehydration, exactly the situation I argue is of concern. I have not seen any evidence that routine levels of blood ketones up to 3 mM represents any concern and that even levels of 5 mM associated with starvation are an issue. We don’t see people on extended fasts or starving being diagnosed with DKA.

10

Actually, there are cases of people with T1D being diagnosed with DKA while in a starved state.

I guess what you don’t address is that, yes, dehydration is a major player in DKA, but how exactly do people get dehydrated? Sleeping all night without drinking something is certainly a form of dehydration, and I suspect that the ketone production itself is dehydrating.

All I’m saying is that the first domino can be the ketone production, and if you’re running persistently higher ketones due to a LCHF diet, you have tilted that first domino a little, meaning you may need to consider the possibility of DKA in situations that you otherwise wouldn’t.

11

I think you have the wrong end of the stick with dietary ketones and ketoacidosis KA be it from diabetes (called diabetic ketoacidosis DKA) or other causes of KA. Or just acidosis

I can’t add to what has already been said here and to you on the other thread about it.

When the answer is about 4 books worth. I don’t think a forum will answer it to your satisfaction. I would hit google for a few hours.

12

I understand your concern, but I think the example you’ve provided is not really a great example. For one thing, the abstract of the paper mentions that the single case was severely dehydrated, and that was a first treatment in correcting the euglycemic DKA episode. Secondly, there is the following line:

This case involves the complex interplay among type 1 diabetes, depression, ketoacidosis, and starvation physiology resulting in glucose concentrations in keeping with euglycemic diabetic ketoacidosis.

So, while I understand your point, I think, I don’t think it is something most people have to worry about. This PWD was Type 1, depressed, and starving themselves. That’s a fairly unusual set of circumstances, which the paper points out (they call euglycemic diabetic ketoacidosis “relatively uncommon”). So I think the generalized advice you’re trying to make (if you are Type 1 and eating low-carb, be aware of susceptibility to ketoacidosis) would make more sense if you added another qualifier or two. I’d phrase it this way:

If you are Type 1 diabetic, eating a low-carb diet, frequently ill, and prone to dehydration or starvation, be aware that you could be prone to euglycemic DKA.

I think this advice would probably apply most commonly to someone suffering from diabulimia or similar.

13

@David49, I understand that these are just anecdata, and as such, not much to hang your hat on. I’m connecting a lot of docs in a pretty speculative manner. So you can bin this in the category of “medical hypothesis” rather than “well-substantiated, evidence-based assertion.”

Stll, allow me to defend my speculation for a second :slight_smile:
I do think that these intriguing, if isolated case reports, combined with all the data on SGLT-inhibotors, suggest that euglycemic DKA maybe underrreported, as stated in this article

In which case, I think it’s worth it for people with diabetes to consider which circumstances may make that more or less likely.

I’m proposing that euglycemic DKA is probably more likely in people who are on a ketogenic diet compared to those who are not on a ketogenic diet. Also possibly more likely in people who are exercising or have increased oxygen needs, or take very little insulin beyond what is needed to supply their basal needs. That’s not to say the overall rate of DKA will be higher in those on a keto diet; just that keto dieters are likelier to have good blood sugar at the time they’re spiraling into DKA.

Obviously that paper is focused on the SGLT2s, but I thought this section was noteworthy:

There are multiple proposed theories exploring the link between SGLT2 inhibitors and ketoacidosis. One possible mechanism describes a decreased secretion of insulin from pancreatic cells in a response to the lowering of blood glucose via urinary excretion.2 This results in a decrease of circulating insulin and its antilipolytic activity, leading to increased free fatty acid production.2 Other evidence suggests that SGLT2 inhibitors stimulate the secretion of the counterregulatory hormone glucagon, which in turn contributes to the overproduction of ketone bodies.13,14 Another animal study suggests that SGLT2 inhibitors might decrease the renal clearance of ketone bodies.2 The net result is a stimulation of the ketogenesis pathway and an increase in serum ketones, which predispose the body to ketoacidosis. This effect is compounded in the event of physiologic stressors including, for example, starvation or dehydration.6,8,10 Although research on the pathophysiology is currently ongoing and no definitive causation has been established, the association between SGLT2 inhibitors and DKA cannot be ignored.

To me, that paragraph in the study questions the notion DKA is a condition that can easily be seen coming a mile in advance. Only half of these cases of DKA had any known trigger!

It also makes me skeptical of the idea that the ketones are just somehow passive bystanders and that dehydration is the real culprit in DKA. The fact is, no one knows exactly which elements of DKA are most critical to the cascade, but the SGLT2 data suggests ketones are a critical player. Sure dehydration may result and exacerbate the problem, but it’s not the primary cause.

Obviously, people who are not on SGLT-2 inhibitors have different physiological tipping points. But it just seems like all these cases of euglycemic DKA are unmasking some of the necessary ingredients to generating the condition: lack of insulin and ketones, but not necessarily dehydration. And when you’re ketogenic, you’re adding a bit more of one of those ingredients (possibly two) to the pot. Whether you think the risk is worth doing regular blood ketone checks is different question: That’s a risk assessment analysis, based on preliminary or speculative information, and of course will be different for everyone.

But I suspect that in a year or two we’ll see more and more data showing that this is surprisingly common and that ketosis is not merely a “bystander” in DKA as seems to be argued on this forum quite a lot.

This is not to knock a low-carb diet, by the way; every type of intervention is going to have some pros and cons. For LCHF, I’d put: Pros: for many, better blood sugar control, less insulin, reduced weight, etc. Cons: No apple pie! No ice cream! No focaccia bread! And potentially: less warning for DKA, even if the overall risk is small and not elevated.

14

I’m scared to try keto again after my first attempt a few years ago. Short story- hospital with dehydration and mysterious acidosis-like symptoms but without high blood sugar. It was not pretty and I’m not going there again (I was trying it to stave off diabetes and to lose weight).

15

I’m definitely not disagreeing with you on this one! As Dr. Peter Attia says, the difference between DKA and nutritional ketosis is the difference between “a house fire and a fireplace.” But, having even slightly elevated levels of ketones (as I surely do, given the amount I exercise and how few carbs I eat) is necessarily decreasing the distance between “normal” levels of plasma ketones and dangerous levels of plasma ketones.

However, the difference between plasma ketone levels in nutritional ketosis and DKA are rather large. The maximum likely plasma concentration of ketones from nutritional ketosis is about 3.0 mM. The minimum level at which acidosis is possible is about 15.0 mM. That may not sound like a huge difference, but you are at “only” 20% of the “minimum danger level” when fully keto-adapted. Short of starvation, no circulating insulin, or dehydration, it’s unlikely that one could just have an 80% jump in ketone concentration.

I think, really, that every diabetic of any type should be aware of the same things. Not having enough insulin, eating too many carbs, not drinking enough water, and eating too much or not enough will put you at risk of various life-threatening conditions. Diabetic coma and DKA (whether with elevated blood sugar or euglycemic) are killers. So are critical lows. So, I’m aware of them, but I don’t let fear of these conditions affect my day-to-day life. I have made a judgement: eating low-carb has more benefits than the daily risk of going into an unpredictable, euglycemic DKA event.

Now, if I end up as one of the lucky few in a DKA event with a 85 mg/dL blood sugar, I’ll probably come back here and change my tune :slight_smile: Now it’s time for me to hit the gym and raise my ketone levels a bit more haha.

1 Like
16

As far as I can tell from searching, DKA can happen at ketone levels as low as 3 mmol/L or sometimes lower.
https://www.diabetes.org.uk/Documents/About%20Us/What%20we%20say/Management-of-DKA-241013.pdf

17

I’ve always been confused by the ketone levels I see quoted in low-carb groups, too. If ketones aren’t dangerous until they’re 10-25 mmol/L, it makes me wonder why my ketone meter says that >0.5 may indicate a problem, >1.5 requires medical intervention (sick day rules), and >3.0 is grounds to seek emergency medical care because an immediate risk of DKA. These guidelines seem awfully alarmist if you don’t really have to worry until you’re 15 mmol/L. It makes me wonder if there’s a difference in units such as mmol/L and mg/dl.

18

This thread caused me to look at my Abbott Precision Xtra blood ketone tester and it’s package insert. Their warnings all couple a >= 300 mg/dL (16.7 mmol/L) BG with three levels of ketones.

0.0 mmol/L - 0.6 mmol/L – Repeat both ketone and blood tests with new strips – contact health care professional

0.6 mmol/L - 1.5 mmol/L – “may indicate development of a problem that could require medical assistance. Follow your health care professional’s instructions.”

> 1.5 mmol/L – “You may be at risk of developing diabetic ketoacidosis (DKA).”

I’m going to ask my new endo to write a script for 12/90 days.

19

@Terry4, these are even lower than what I was going with by memory. Which makes me wonder if there are different units of measurement. Is mM the same as mmol/L? Becuase 1.5 mmol/L is miles away from what @Brian_BSC quotes of 15-25 mmol/L for DKA. If they are different units, then I think we should be converting to the same units. To my knowledge, all ketone meters available to individuals measure ketones in mmol/L. And if they are the same units, why is one source (low-carb books/forums) saying they are safe while another (endocrinologists and meter manuals) are saying they are potential DKA? Those seem extreme opposites.

My endocrinologist said that, at least in the cases of SGLT2 inhibitor related DKA he has treated in emergency (he splits his time between his practice seeing patients, working at the hospital, and teaching medical school), people have said it seems to come out of nowhere. So, like you, I agree that monitoring for DKA and being aware of it may be a good idea, especially for people who are eating low-carb and have Type 1 (or otherwise have minimal to no insulin production) and are on a pump, or otherwise have multiple risk factors. I think most people who eat low-carb may not have these multiple risk factors, though.

I also do think that, given your son was feeling sick (you said in the original post that he was not sick) caution was warranted. I think when there are multiple factors at play, it can be hard to tell what is causing a high level of ketones, and in those cases (in my opinion) it’s better to be safe than sorry. I outlined in my example of a bad pump site while eating low-carb. Maybe that ketone level of 2.1 mmol/L was perfectly normal given what I was eating, but given I also had a blood sugar of 20 mmol/L and felt a bit sick, I treated it with additional insulin (as per sick day rules).

I know that there are people that never check for ketones because they claim it doesn’t matter, and I don’t agree with that, because if you are following sick day or similar rules, ketones direct one to give additional insulin above and beyond what would normally be given for a high blood sugar with no ketones.

20

Thanks @Jen, yes I agree. In our son’s case, there was just no question – he was getting no insulin for close to 12 hours! That in itself can kick people into DKA, so we never doubted that the ketones were a flashing red warning signal.
It’s definitely scary that SGLT2s can cause DKA without warning.