If you are on a ketogenic or very low carb (VLC) diet (e.g. with 50-100gr carb/day and/or eating ketone producing MCT oils such as coconut oil), you may have a dilemma of having high Blood Glucose (BG) despite eating LC: If you are keto adapted, that is, your body is using ketones and even though you have sufficient insulin (say >5 microU/ml) your body tries to keep your BG higher than necessary, e.g. above 100-110 mg/dl. That is your BG set-point is always high. If you try to lower the set-point to say 80s, by water Intermittent Fasting (IF), then your body starts to convert your muscles into glucose to keep its high BG set-point. So, you may have a slightly lower BG, but you lose some muscle mass. Having a high set-point has many other problems, e.g. if you eat something with a little bit more carb, say a small fruit, your BG shoots up to 130s and stays there for hours.
This may be due to something called "Physiological Insulin Resistance (PhIR) by Petro Dobromylskyj. He wrote many good articles about it -–thanks Petro--in his blog Hyperlipid. Apparently, PhIR is a normal reaction of the body and quite different from Pathological Insulin Resistance (PaIR). It seems that the main difference between PhIR and PaIR is that insulin is at a normal level in the former and abnormally high in the latter. (PaIR is obviously type2 diabetes.) If I understand correctly, PhIR is kind of IR only in the muscle tissue, that is only the muscles do not react to insulin and NOT use glucose even though it is available. However, if you are eating too much protein, the liver may also be considered IR, because it tries to keep the BG high by converting proteins to glucose, even though BG is already too high, that is, it also may not be responding to insulin. (I think working muscles can use glucose with or without insulin. Would that mean that PhIR does not affect working muscles? If not, then one can reduce his high BG [due to PhIR] by exercising.)
The purpose of this Paleohack question is to understand how this happens to whom and how it can be avoided?
Several things seem to cause PhIR:
“… LC eating rapidly induces [Physiological] insulin resistance. This is a completely and utterly normal physiological response to carbohydrate restriction. Carbohydrate restriction drops insulin levels. Low insulin levels activate hormone sensitive lipase. Fatty tissue breaks down and releases non esterified fatty acids (NEFA). These are mostly taken up by muscle cells as fuel and automatically induce insulin resistance in those muscles. “…palmitic acid is the primary NEFA released from human adipose tissue during fasting. Think of palmitic as a signal molecule to tell the muscles that inhibition of glucose uptake is needed and to tell the liver that increased gluconeogenesis is required because there is no food coming in. Chris Kresser also supports this view.
Ketone bodies e.g. “Beta hydroxy butyrate. … ketone bodies, one of the hall marks of carbohydrate or total calorie restriction, channel glucose away from muscles, toward brain and add a modest supplementary energy supply to brain tissue too. It's exactly what you expect on an adaptive basis, exactly the same function as palmitic acid performs and clearly the two metabolic pathways are closely linked, though ketones seem to work downstream of the action of palmitic acid.”
Methylglyoxal: “Methylglyoxal is elevated in ketosis, but the bulk is produced by glycolysis… Methylglyoxal is an inhibitor of glycolysis.”
According to Paul Jaminet, the high BG problem is caused by too low carbs in VLC diets and can be avoided by eating some more “safe starches”. According to Dr Ron Rosedale, it is caused by too much protein in the diet. This discussion was taken up also by one Paleohack question, which lists many good links about it. But, I think the best and the most up-to-date link on this discussion is this one, where Dr Rosedale uses color coding to distinguish who said what. It is a good scientific discussion and definitely worth reading to understand many background issues. The amount of protein Dr Rosedale recommends is:
Dr. Rosedale defines his protein recommendation as: “So what’s high [protein]. Certainly above 1 gram [of protein] per kilogram of lean mass is probably high. Most people, I’ll put on 0.7 or 0.75 grams per kilogram of lean body mass. But if I’ve got a diabetic, and I really want to reverse their aging, which means reverse their diabetes, because diabetics is a model of aging, I’ll put them down to 0.5 or 0.6 grams per kilogram of lean body mass per day.” For downloading the full PDF version of this article see: "Protein: The Good, The Bad and The Ugly" http://drrosedale.com/Protein_The_Good_The_Bad_and_The_Ugly.htm
A few other points need to be noted in the above Jaminet-Rosedale (JR) debate in view of several excellent articles recently written by Chris Masterjohn on glycation, glucose synthesis and oxaloacetate :
They generally agree that glycation is important, but ignore methylgloxal, which is higher in VLC and ketogenic diets and 20000 times more glycating than glucose (existence of methylgloxal supports Jaminet’s position) Methylglyoxal is a reactive aldehyde that is very toxic to cells, it can inhibit growth in E. coli at milimolar concentrations. The excessive intake of glucose by a cell is the most important process for the activation of the methylglyoxal pathway. (It is strange that “methylglyoxal is produced by glycolysis… and yet also an inhibitor of glycolysis.” What does it mean?) But, there is also some evidence that methylglyoxal kills some types of cancer cells.
They talk about post prandial blood glucose (PPBG), but ignore PhIR effect on PPBG, which often causes longer PPBG spikes than high glycemic foods (existence of PhIR supports Jaminet’s position).
They do not discuss oxaloacetate (supports Jaminet’s position)
They do not discuss TCA cycle (supports Rosedale’s position)
Although not directly related to the JR debate, apparently, conversion of glucose to fat is not significant in humans.
It should be noted that Dr Rosedale always emphasizes deleterious effects of hormonal changes e.g. of insulin, leptin caused by PPBG spikes, (rather than glycation caused by glucose,) such as mTOR, autophagy etc. He notes that: “However a main point of my argument is that BG levels are only a small part of the story; What higher carb intake does to insulin and leptin is even more important; it raises them promoting insulin and leptin [pathologic] resistance.” " …only measuring blood glucose without knowing what insulin and leptin are doing gives very incomplete and often misleading information when it comes to effects of any intervention such as diet.”
So my question is how can we avoid the high BG caused by PhIR, do we have to keep exercising to reduce BG? Or is high protein the real issue, if yes, then what exactly should be the protein amount. Or should we try to avoid ketosis, but, if you eat MCT oils (e.g. coconut oil) the body produces ketones even at higher carb levels. Also does this happen in other LC diets, such as Atkins and the Optimal Diet of Jan Kwasniewski.
Other possibilities are that you've developed IR. You may have been over the hump before you started low-carbing. Or it may have developed even while low-carbing. These are all plausible but it's all guesswork. Get tested first.
The recent breakthrough I'm referring to is the finding that elevated BCAAs are associated with IR. Even years before diabetes diagnosis. This would be like pre-pre-diabetes.
There are animal models which implicate BCAA supplementing with IR. Just a few days ago, the NE J of M published findings regarding bariatric surgery patients that reversed their pre-existing diabetes (many of them were insulin-dependent T2 diabetics), whose beta cells were largely thought to be non-functional. After the surgery, these patients showed significantly decreased BCAA levels, got their BG control back, and got off of insulin. Most were "cured" of their diabetes in the sense that they attained normal BG even before losing any weight from the surgery.
Implication: increased BCAAs could promote IR, and a gut restructuring could immediately lower BCAAs and restore normal BG control and reverse IR. We don't know how this works but all we know is that there is association between high serum BCAAs and IR. Paleos might pooh pooh this finding. In fact, there were studies indicating exactly the opposite, how leucine benefits IR, until a few years ago. Howver, there seems to be something else going on. And no one knows for sure.
It is possible that IR can develop and even accelerate while doing low-carb, high-fat, and high-protein diets. One thing I wouldn't do is supplement with BCAAs while low-carbing, as weightlifters do.
The overall issue is insulin sensitivity and the effect is somewhat exaggerated. The prior studies cited by Jaminet and others had rather large A1c ranges: 0.5 intervals. I doubt a VLC diet will push your FBG too much above 110. If you do, you may have failing BG control. You're less sensitive to insulin, for sure. And it is true that on a high carb (i.e., safe starch) diet, your FBG could be 70-80, as your body is just primed for you to ingest carbs. It's like always having your gun loaded, lest you ingest a large glucose heaping.
But I fail to understand the underlying issue. VLCing should not deteriorate your BG control. In other words, VLCing will flatten out your BG volatility, while a high-carb / safe starch diet will have much higher peaks (esp. at 1h pp) but lower overall FBG. And when measuring your A1c, the postprandial peaks seem to be averaged out and you end up with lower overall A1c when doing safe-starch / high carb diets, especially when combined with HIT. But you WILL have peaks above 140 at 1h pp even with normal BG control. Whether that's harmful is another matter, as some premise their entire VLC effort on "preserving" remaining pancreatic beta cells; supposedly, even breaches above 110 will render some of your beta cells non-functional. But I think the recent breakthrough in diabetic reserach has shown that it's the incretins, BCAAs and gut hormones which really control insulin secretion, not the beta cells, whose functionality could be restored.
If you see your FBG go up above 110 consistently and closer to 150, then you may be losing BG control for reasons other than the lack of insulin sensitivity caused by the VLC diet. If you're more closer to 110-120, I would think it is the VLC diet. People on VLC diets still progress on their paths to diabetes. Many T2 diabetics become insulin dependent or "double diabetics." Many LADAs or MODYs not previously diagnosed have to be on insulin and will graudually or abruptly lose BG control. But for the healthy, the difference in FBG should be within the normal ranges mentioned above, not at the diabetic range (129+).
And as mentioned above, the lower A1c with higher carb diets are achieved via higher volatility but the regress toward the mean phenomemon. It's not clear which one is superior: the lower volatility but higher A1c and higher FBG vs. the higher volatility but lower A1c and lower FBG.
VLC eating raised my fasting BG from 95 to 115 - just adding my data to the discussion.
Personally, after spending a lot of time reading about this issue, I have decided to side with Dr. Jaminet, and I will be adding safe starches soon.
This issue is kind of scary. This post pushed me to order a glucose monitor from Amazon to make sure I'm still in normal range - it's kind of scary not knowing what this way of eating is doing. Yeah, I've lost a ton of weight and I've got a 6 pack starting to show, but what long term negative health effects are there with high blood glucose?
I don't have time to read the whole thing so I'm sorry if I missed something (I'm heading to work) but one thing caught my eye:
"According to Dr Ron Rosedale, it is caused by too much protein in the diet."
Well, then don't eat a cow for dinner, right? Being Paleo isn't about eating massive quantities of protein or eating it to an excess. I think many people here don't eat a cow for dinner. They just eat LC and replace some of the sugar/starches with fats.
Not enough glucokinase? It would seem the liver would want to store the glycogen rather than leave it in the blood stream. I think paleo amounts of fructose are necessary for the release of glucokinase.
Namby Pamby, thank you for your note. Please see my replies to the points you raise after the >> signs below.
I doubt a VLC diet will push your FBG too much above 110: >>It does not, in fact it is almost always between 100-110.
But I fail to understand the underlying issue. VLCing should not deteriorate your BG control:>>Apparently, muscles do not use glucose, but liver continues to make glucose. Please see the Hyperlipid blog links I gave.
But you WILL have peaks above 140 at 1h pp even with normal BG control.: >>Yes I have, if I eat more than 25g of carbs in one meal. The problem is that it lingers between 130-140 for 3-4 hours, unless I do some exercise. Maybe with little carbs in the meal, protein is converted to glucose as it is digested and its digestion takes 3-4 hours. My A1C is about 5.6.
But I think the recent breakthrough in diabetic research has shown that it's the incretins, BCAAs and gut hormones which really control insulin secretion, not the beta cells, whose functionality could be restored: >>That’s interesting. Could you please give some links/references about this recent research.
Actually I agree with Rosedale's observation regarding protein's effect on BG: longer PP BG rises. But unless you're eating 200g+ protein, the effect should be small. >> I eat 80-95g/d of protein on average, but maybe my calorie intake is too low, about 1500 cal/d, so %protein cal maybe a bit high and as I noted above, it may be converted to glucose to compensate for lack of carbs, although I doubt it as I eat 30-40g/d coconut oil and about 60-100g carbs/d. This is exactly the point I do not understand and that is why I wrote this paleohack question.
So, yes, with proper protein restriction, your BG should improve while VLCing, than under VLC diets with no protein upper limit. >> I agree. Apparently Rosedale has reduced his protein recommendation. So, probably I need to eat about 65-70g/d protein (and more fat to compensate for its calories).
Thanks again for your response..
I'm posting this here because although its an old post it's the first one that comes up when you search for high glucose on a low-carb diet. I have been struggling with the same thing. I read and re-read this post over and over trying to make sense of what was going on. How could I go from regularly getting readings in the mid 70's in the late afternoon to never having a reading below 85, period.
Why were my fasting glucose numbers so high?
I found this information (IFG means impaired fasting glucose):
"As mentioned above, HOMA-IR largely reflects resistance of glucose production by liver and kidney to suppression by insulin, whereas hyperglycemic clamp-determined insulin sensitivity largely reflects the sensitivity of muscle glucose uptake to stimulation by insulin (22). Our finding that in IFG, HOMA-IR was increased but hyperglycemic clamp-determined insulin sensitivity was not decreased lead us therefore to postulate that in IFG individuals, there may be selective or preferential hepatorenal insulin resistance."
Since my fasting readings are in the pre-diabetes stage I researched that and determined that I would be considered to have Impaired Fasting Glucose but normal glucose tolerance (since my post prandials are fine). Apparently they are very different. IFG results when you have a reduced basal insulin secretion and impaired glucose-stimulated first phase insulin secretion. And most importantly:
People with impaired fasting glucose have insulin resistance to glucose that their liver makes.
So if that is part of my problem it would explain it quite well since that is just about the only glucose I'm getting at 5am....the glucose that my liver makes.
Here is the citation:
Different mechanisms for impaired fasting glucose and impaired postprandial glucose tolerance in humans.
John Gerich, Christian Meyer, Asimina Mitrakou, Walkyria Pimenta, Ervin Szoke, Timon Van Haeften,, et al. Diabetes Care. 29.8 (Aug. 2006) p1909. Word Count: 3730.
Hope this helps somebody out there.
Morning Blood Sugar 6 Answers