Hypoglycemia low blood glucose while highly fat adapted with high ketones Research?

(Herb Martin) #1

Do you know of links to any real science or even case studies and reports on people with (technical) hypoglycemia while highly fat adapted with high ketones?

I am talking about people who feel fine, even wonderful, and who may be on long extended fasts, with glucose down below 55-ish (3.0) and ketones up above 5.0-7.0

My guess is this isn’t well studied and few medical people know the actual effects even if it has been researched.

Any links or directions to anything much beyond the standard definitions of hypoglycemia would be appreciated.

(For context: I have been a carnivore for 9 months and fasting 2/3 of the time for the last 2 1/2, commonly running ketones above 6.0 or even 8.1+ at several weeks. So far my glucose hasn’t dropped out of the 50s or isn’t often out of the 60s but I’ve seen reports of others in the 40s and want to do my research ahead of such events.)

What is your lowest fasted blood glucose?
Anyone have a keto BGL ratio for metabolic health?
(Wendy) #2

Hi last May i ended up in the hospital. Heart rhythm out of wack and high blood pressure. Blood sugar was in the 40s. I had No idea the the low blood sugar would cause heart problems. Since i went keto. And to my surprise my blood sugar has leveled out! Heart rhythm and blood pressure great! Have lost 20#! Just got the mojo ketone tester. So far the highest is 2.0
Still feeling great!

(Jane) #3

During my last EF back in January my BG dropped to 52 and my BK was 7.6. I felt great. Later that day my BG went up even though only drinking water so apparently my body decided it needed to make some glucose and it did.

Would be an interesting study, for sure.


I believe @Karim_Wassef has documented many such events in his self-experiment data collecting.

(Karim Wassef) #5

Yes. My 19 day fast and subsequent 5 days a week cyclic fasts showed that this was not unusual for me

Lowest glucose 33, highest ketones 9.3, on different days too.

I used DEXA and RMR to monitor lean mass and metabolic rate decline.

The continuation is in my carnivore reverse fasting thread :

(Herb Martin) #6

Karim’s story is useful since apparently people can and do survive even with blood glucose in the 30s, but I am still seeking actual research in addition.

I haven’t been below 55-ish (as far I remember) but heard of people in the 40s (prior to this thread) and wanted to understand it before it happens.

My ketone meter (Precision Xtra) topped out at 8.0 so once I went above that it just said “HI”.

Friendly but not that useful :slight_smile:


For some of us, glucose may never reach the low levels some have seen for a couple of reasons I can think of off the top of my head. First, different people produce different glucose and ketone combinations and that appears unique to the individual because of hormonal cocktails we can mostly only guess at. Second, the brain does not see the same ketone levels as the plasma due to the blood-brain barrier, and that also varies among people. I’m a believer that brain ketone availability largely determines how much glucose your body decides it needs.

Here is one study of a short fast in humans where such things were measured. https://journals.sagepub.com/doi/full/10.1097/00004647-200010000-00012

And a couple of pictures from the study…


Note: the brain also uses AcAc and lactate during glucose scarcity.


I’ve looked through more studies, including those that administered exogenous ketones to see effect on glucose levels, and they all tend to bottom out at around ~60 mg/dl. I’ll bookmark this in case I run across something.


Dang it, just when I had given up, it occurred to me to look at Angus Barbieri. He got super low.

Blood glucose concentrations around 30 mg/100 ml were recorded consistently during the last 8 months, although the patient was ambulant and attending as an out-patient. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2495396/

(Karim Wassef) #10

what are you looking for?

The reason some people get into the 30s is because we get deeply fat adapted and physiologically glucose intolerant.

Basically, our cells prefer ketones and eschew glucose. The liver reacts by setting the glucose setpoint lower (reducing glucose lower reduces oxidative stress and increases longevity). The only cells that are obligate glucose users are red blood cells, parts of the brain and internal parts of the kidneys. Those don’t have mitochondria so can’t use ketones…

(Bacon by any other name would taste just as great.) #11

Dr. Phinney has a story of some friends of his who used an insulin clamp to drive down the serum glucose of some fat-adapted athletes they were studying. He said that, even though the researchers realized later that they could get into trouble for taking such a risk (and no IRB would ever countenance such shenanigans today), they did publish the results. I’ll try to track down a reference to the paper; I remember coming across it within the last six months.

(Milton Alvis, MD) #12

Personally, I have never seen anyone in ketosis and using a CGM (extremely helpful) with CGM glucose concentrations below about 45 mg/dl.

However, does mean they have any problems or symptoms?
No!. Typically report feeling and thinking better than usual.

All the fear about low glucose concentrations comes from insulin and sulfonylurea medications which can drive glucose low enough in people who are not keto adapted, typically below 40 mg/dL and end up symptomatic. The concerns about low are driven by those using they bad tools and concerns that people go low enough they become confused, comatose and die if their glucose continue to go down
far below 40. The lowest I have ever seen, by finger stick, was 0 mg/dL (accuracy unclear) because was in a VA hospital surgical ward and had been given too much insulin. He was, of course, comatose, but awakened, very confused, after being given D50 IV.

There is Zero evidence that I am aware of the low glucose ever becomes a problem for people adapted to nutritional ketosis.

Be aware that many issues which many physicians believe are not true.

Medicine remains more art than science and overwhelmingly about disease, not health: http://goo.gl/Blh6rW

Milton Alvis

(Bacon by any other name would taste just as great.) #13

Phinney says that when his friends performed that study, they drove the participants’ serum glucose down to levels that normallly cause coma or death, but the guys were fine. I love his line, “But the only people having palpitations were the ones in the white coats.” :grin:

Of course, no IRB in its right mind would authorize such a study today.

I believe this is the writeup of the study I have in mind. If not, it’s a very similar study.
“Resistance to Symptomatic Insulin Reactions after Fasting,” Drenick et al., 1972)

(Bunny) #14

Phinney & Volek: Starvation Ketosis: During total fasting when there is complete absence of any caloric intake for several days, the resulting increase in ketones is referred to as starvation ketosis. The absence of any dietary carbohydrate and protein over a week or more raises ketone concentrations to between 5 and 10 mmol/L, significantly higher than nutritional ketosis, but lower than concentrations in keto-acidosis. Starvation ketosis is an important physiologic process that evolved millions of years ago, enabling humans to survive for prolonged periods on body fat when food was not available.4 Obviously starvation ketosis is not sustainable long term, nor is it advisable to intentionally induce it for shorter periods (i.e., intermittent fasting) because of essential nutrient deprivation, lean tissue loss, and other potentially dangerous side effects.5 …” …More

Megan Ramos: “…But my blood sugar levels are typically between 3 and 3.9 mmol/L, which translates into 50 to …or sorry 54 to 70 mg/dL. So, that 65 to 99 mg/dL is considered to be normal by government standards here in Canada, the 54 to 70 is considered to be common amongst ketogenic population. A lot of my patients too, who follow ketogenic diet, they’re getting sugars between 3 and 3.9 or between 54 and 70.

Sort of branching off the first question about blood glucose levels, what are optimal blood ketone levels? Good question. We don’t really know what any of these ketone levels mean. It’s crazy. I get asked these questions all the time and there’s a occasional post on the keto page or there’s different doctors making different speculations about what are optimal ketone levels. There’s just not a whole lot of data out there really telling us the significance of certain ketone levels. We do have some research and we can see that nutritional ketosis is between 0.5 and 1.5 mmol/L. When your ketones are between 0.5 and 1.5 mmol/L, then you’re in a good state of nutritional ketosis. Nutritional ketosis does not equate to therapeutic ketosis that you would use to treat metabolic syndrome, neurological conditions such as MS or Parkinson’s or Alzheimer’s disease or various cancers. Nutritional ketosis is just for overall health and well-being.

A lot of research shows that fat burning occurs between 1.5 and 3.O mmol/L. There are some research that indicates that once you’re above 3, it doesn’t increase your rate of fat loss. But there’s a difference between 1.5 and 3.O. Yes, 3.O indicates more body fat burning or more fat burning in general than 1.5 mmol/L. The researches done doesn’t really show much more of a significant impact of having ketones of 4, 5 or 6 as compared to 3 for weight loss and fat burning. What should your numbers be? What if your ketones are 6? What if they’re 5? What if they’re 4? Is that a bad thing? We don’t really know. I’ll tell you a little bit about my numbers. My morning ketone numbers are usually between 3 and 6. What gauge is my blood ketone number is in the morning, determines on how late in the evening I eat. I try to eat no later than 7:00 or 7:30 come hell or high water. Most days, my husband and I finish eating at 2:00 PM if we intend to eat at all that day if we’re not fasting. But the weekends, Friday nights, it doesn’t always pan out that way. I really, really try except for the odd time, which ends up being about once or twice month where I end up eating a little bit past 7:30. Rarely, rarely, do I ever eat past 8:00 or 8:30 in the evening. Usually, those are very late nights too where I’m up until 1:00 or 2:00 in the morning at a social event. Those are my morning numbers, between 3 and 6. I feel good and my blood sugar levels are low.

Now, having these high ketone levels can be a problem if the blood sugar levels are going up. I usually tell people that if their ketones hit 4 and their blood sugar levels are trending up and not down, they need to stop and they need to eat. They don’t need to eat carbohydrates but you need to eat a well-balanced ketogenic or low carbohydrate meal. Get in plenty of non-starchy vegetables, get in some good fats, a variety of good fats, eat some good hearty protein and then just hold off on fasting for a little bit. It becomes very dangerous when the blood glucose levels and ketone levels are both going up. Now, my ketone levels can be 6 and my blood glucose levels can be around 3 mmol/L, which is around 65. That’s not a word for concern and I feel fantastic. We are learning more and more what these numbers mean. What is really optimal, we don’t really know. Also, for patients that are fasting, when you’re in a fasting state and you start to get ketones of 0.5 and 1.5, that does actually indicate some body fat burning. If you’re in a fasted state and you see ketones of 1, don’t beat yourself up. You are burning fat. We know that that 1 is not coming from nutritional ketosis because you haven’t eaten. You’re fasting.

​All these numbers are a little funky. We still need to learn a lot. We need to learn what they mean on their own on a high fat diet, on a ketogenic diet. We need to learn what fasting means for these numbers too, a little bit more. If you are fasting, don’t get discouraged when you see ketones over 0.5. That’s usually when I start to be able to correlate patients ketones with weight loss. As soon as my patients start getting ketones of 0.5, I notice weight loss on the scale and body fat loss in terms of taking their measurements. That’s when you’re in a fasted state. That’s really good. Even though that’s considered nutritional ketosis if you’re eating, for weight loss, you’re still losing weight at those numbers if you’re fasting. Don’t go crazy trying to get high numbers. Be very cautious if your ketones go up and your blood sugar levels are also going up. If it looks like that’s a trend, it’s always better to be safe than sorry and stop and have something to eat. There’s never any pain in eating some bacon and eggs and you can always jump back into your fast after that.

​I know many of you are going to now write in asking about diabetic keto acidosis but I’m going to talk about that in another few Q & A’s from now. We’re going to cover a little bit more on the ketogenic diet and what is diabetic keto acidosis. …” - Megan Ramos

(Herb Martin) #15

PaulLOld BaconianAdmin

I need to read and study much of the above but wanted to at least say some kind of thank you now to everyone who is offering help.

This is much more of what I was seeking, and there may even be more in these references and links than I suspect.

I think the main thing we are still missing (it probably doesn’t exist or we’d have found it by now) is:

What is a dangerous (or what’s safe) LOW blood glucose level for a fully fat adapted person who is on a ketogenic diet or fast with ketones at a high level and who is experiencing NO SYMPTOMS?

Apparently Karim (and others) have been into the 30s, I’ve been in the 50’s (which isn’t that low).

I trust Karim’s report but it would be nice to have some non-anecdotal guidelines rather than “below 70 etc is hypoglycemia” and requires possible medical intervention.

MAlvisMilton Alvis, MD
What would you do if someone like Karim walked in with a BG of 33 and felt wonderful while his ketones were 5+?

Thank you Wendy though I am not looking for problems with hypoglycemia unless you were highly ketogenic (>2.0->5.0) AND likely if you had NO SYMPTOMS before it became a serious problem.

What I am trying to discover is SAFE LIMITS for those of us who are monitoring ourselves carefully and what is safe to advise others who are FEELING REALLY GOOD and taking reasonable precautions.

Preferably as citable research papers, but case studies and other reports are also useful both in lieu of peer review hard science and as an adjunct.

Going to re-read and study all of the above but keep it coming please.

(Karim Wassef) #16

glucose alone is not a meaningful threshold. It’s all about energy and there are other key sources of energy: ketones, free fatty acids, and even creatine for sudden bursts of energy.

glucose of 30 with ketones of 0 = comatose
glucose of 30 with ketones of 7 = feeling awesome!

I always measure both because one without the other isn’t so meaningful.

(Herb Martin) #17


glucose alone is not a meaningful threshold. It’s all about energy and there are other key sources of energy: ketones, free fatty acids, and even creatine for sudden bursts of energy.

glucose of 30 with ketones of 0 = comatose
glucose of 30 with ketones of 7 = feeling awesome!

Ok, I get that, but at least a small part of the motivation of this request is to deal with semi-knowledgable (medical) people who freak out at the sight of 30 or even 50 but have no knowledge of nutritional ketosis.

The other half involves ensuring we don’t give poor advice to someone (including especially ourselves) to ignore a real problem.

So while I agree with you I am still looking for [more] citable science.

(Bunny) #18

Megan Ramos: ​Now, the problem with these glucose levels is that they’ve sort of been determined based on people who are following the standard North America diet, which we know is high in carbohydrates, starches or fine carbohydrates. All these really opposite foods of the low carbohydrate, high fat diet that a lot of us are trying to follow or the ketogenic diet. Now, I, myself, am keto adapted and I follow a very strict ketogenic diet 95% of the time. I’m still human, guys, so cut me some slack for the 5% of the time that I’m off my game. ​But my blood sugar levels are typically between 3 and 3.9 mmol/L, which translates into 50 to … or sorry 54 to 70 mg/dL. So, that 65 to 99 mg/dL is considered to be normal by government standards here in Canada, the 54 to 70 is considered to be common amongst ketogenic population. A lot of my patients too, who follow ketogenic diet, they’re getting sugars between 3 and 3.9 or between 54 and 70. What does it mean? What is normal? What is acceptable for someone who’s keto adapted? We’re still learning. I usually gauge how my body’s feeling at a certain blood sugar level. I let that determine whether or not it’s normal for me. Hypoglycemia is not defined as a set of numbers. It’s defined as a set of symptoms, so dizziness, mental fatigue, shaking, palms sweaty, hands. That defines hypoglycemia. It’s a list of symptoms.

​I have so many patients in clinic, who are so used to their blood sugar levels being super high, like somewhere around 360, which is about 20 mmol/L for our Canadian friends, and once they drop down to half of that, 180 or 10, they start to feel hypoglycemic because they’re so used to running higher. Sometimes, it takes a body a little while to adapt. This happened to me when I first went on a keto diet. My blood sugar levels dropped and I didn’t feel so great, so I had to cut back on my fasting a little bit. My body adapted to having these lower blood sugar levels. It started fueling on fat more efficiently at the same time and I started feeling fantastic. Now, I feel really great when my blood sugar level’s 3 or 54, I feel fantastic. I feel on top of the world and my body’s fueling nicely on fat. …More

(Bunny) #19

Megan Ramos was the very first one (a guinea pig as she humorously calls it) in Dr. Jason Fung’s practice to combine therapeutic fasting with the ketogenic diet (to find out its affects on the human body) as she describes it here in this video:

Megan Ramos - 'Practical Fasting: The Use of Therapeutic Fasting in a Clinical Setting’


[1] Therapeutic use of intermittent fasting for people with type 2 diabetes as an alternative to insulin

(Karim Wassef) #20

we are ketone consumers from birth

watch 6:45 minutes in…

glucose is only necessary for a small % of overall cells. The liver regulates down to the minimum needed.