Insulin load index / most ketogenic foods

[tim2000s]

The danger of the low carb / high carb debate is that the high carb proponents will run a straw man argument and say that the body needs insulin (which is correct) and it's nuts to try to eliminate it. I just think you need to reduce it so you can achieve optimal blood sugars and minimise the risks of insulin toxicity.

Surely the counter argument to that is pretty clear though.

The body needs insulin. Yes, it does.
It's nuts to try and eliminate it. That's correct, but how is fat deposited? Insulin causes fat cells to convert excess carbohydrate to fat and store it while stopping the body from using it as energy.
What is logically the best way to reduce obesity? Reduce insulin or carbs.
Well if you reduce carbs, then you reduce insulin. That way you reduce fat conversion and therefore obesity.

Ignoring insulin toxicity, I'd posit that this is simply a side effect of there being too much insulin in the system, rather than anything else, and in a normal, functioning adult, the only real way to reduce insulin is to reduce carbs (and other glucogenic foods, but predominantly carbs).

I think that much of the issue relating to insulin and the high carb diet is that the majority of HCPs have little idea of how important Insulin is in the lipid metabolism. I encourage all T1 and T2 diabetics to try and have a discussion with DSNs, GPs and consultants where they bring this into the conversation. I think people will be surprised at the results.

I don't think I'd go near the insulin toxicity discussion as the immediate impact of LC is that you reduce obesity (which gets more focus than even diabetes). As a side effect, based on the Southport GP and Newcastle Diet experiences, you can reverse diabetes. If LC is therefore put forward as the correct way to eat, then you wouldn't see anywhere near as much obesity or insulin resistance, and therefore T2 wouldn't be developed to anything like the extent that we have been seeing.

 

[zand]

@Spiker Thanks for starting this thread. Some of it is a bit above my head, but I'm enjoying the challenge nevertheless.

 

[LucySW]

Yes, it's incredibly interesting. I'll have to come back to this all at a later stage. These issues of carb v insulin in management are so important.

 

[Spiker]

@Spiker Thanks for starting this thread. Some of it is a bit above my head, but I'm enjoying the challenge nevertheless.

Same here on both counts!

 

[Spiker]

Yes. Particularly given the recent realisation regarding the risks of insulin toxicity I would think for type 2s insulin should be seen as a short term crutch until they can modify diet and lifestyle to eliminate the need for it and their own pancreas an supply the requirements. I wonder if there's any research on the relative risk of high insulin versus high blood sugars?

Well, as @phoenix showed, above, there is a disagreement about the evidence for high insulin. So I think it is premature to be calling for everyone to run from the right hand side of the boat to the left hand side of the boat. Again, I believe the high insulin theory. I believe it because I read Taubes, and he puts up a lot of evidence. But, @phoenix and others have pointed out that people like Jenny Ruhl have criticised Taubes. And I rate Jenny Ruhl very highly. So Marty I would suggest you should familiarise yourself with the arguments that Taubes references and the criticisms that Jenny Ruhl levels against Taubes, then you will be in a much stronger position to argue this from evidence - what evidence exists, at least.

 

[Spiker]

Spiker said:
Do you have any links to evidence for this other than anecdotal? I ask not because I don't believe you, but because this question is often asked and there is not much solid evidence on the subject one way or the other.

Spiker, only anecdotal. But it makes sense to me to not burn out your remaining beta cells.

Here are some 'anecdotal' videos from people that I have the privilege of calling friends from a seminar in Brisbane.

I asked Troy and he said that he thought he would still be in the honeymoon phase of his diabetes an that the dietary approach that he's taking would be extending that. I think Lisa's son Daniel would be in the same boat.

Ok so that's two people, in fact it's one person's hypothetical speculation and another person talking about a third person. So that is definitely firmly in the world of anecdotal, at best.

The fact that different countries have different approaches might make it possible for someone to design a study that compares the different approaches. The difficulties would be first, that only two options would be looked at, early insulin or late insulin, and low carb would not be an option (since no country except Sweden is close to advising it as a treatment); second, that the honeymoon phase is not itself something that public health workers measure, and doesn't have a clear definition or methodology for measuring it. You would have to infer it by looking at some kind of transition from lower levels to higher levels of insulin, or something.

 

[Spiker]

It makes sense to me that Bernstein would advise to eat protein rather than carbs for T1s as you can get glucose from protein via GNG at a slower rate.

I think you may be missing the point about Bernstein. Bernstein's pioneering method would today be called LCHP, Low Carb High Protein, and would be differentiated from LCHF, Low Carb High Fat. What many of us have found is that LCHP is ineffective precisely because the protein still metabolises to glucose. Protein is not so much an alternative to carbs, as "Carbs Lite". Many of us have found that LCHF is more effective, as it reduces the total glucogenic intake (carbs and protein) and so, through Bernstein's very own "law of small numbers", provides us with better blood sugar control. And, equally important, escaping from carb dependency, including GNG carbs, makes it psychologically and physiologically easier to maintain a low carb diet.

For example, for me, the Bernstein diet is a permanent battle with hunger. It's no different than being on a calorie restricted high carb diet. I get hunger cravings all the time and overeat protein to feed my carb addiction via GNG. Now I realise the straight Bernstein diet has and does work for a lot of people. I am not dissing it. But for me LCHF is more effective and more sustainable.

That's the reason that Bernstein is starting to become somewhat isolated within the low carb diabetic community. Which is a shame because he is practically our messiah and in many cases literally is our saviour.

 

[LucySW]

I feel I owe Bernstein my health and sanity in the appalling mess I found myself in.

And I also couldn't manage the straight Bernsteinian diet - hungry all the time. I just switched over to HF: 30g carb/5%, 15% protein, 80% fat. Tho the high fats, esp sat fats, do make me more insulin resistant. So I'm feeling my way - not wanting that, but not wanting any more weight loss.

 

[Spiker]

The Origin trial is also interesting and doesn't suggest toxicity. This trial, had a very different population to Accord .It was longer, Subjects were people with pre diabetes or early T2 though they did select a large proportion of participants with a high CVD risk. The starting HbA1cs weremuch lower than in Accord and the use of insulin was definitely supplementary .
They were hoping to find that Glargine had a protective effect against CVD. It didn't but it was neutral.
http://professional.diabetes.org/CongressReport_Display.aspx?CID=91304

I don't see that ORIGIN result as any kind of disproof of insulin toxicity. They were taking a bunch of newly diagnosed diabetics and nondiabetics (with IGT or pre D or CV issues) and giving them 1-2 units of Lantus, twice a week. I think it's very clear that when Fung talks about insulin toxicity he is talking about the common situation of advanced T2D and taking a hundred or more units a day. This ORIGIN study really has no bearing on that scenario at all.

As far as I can see the ORIGIN study proves that a tiny amount of insulin has no discernable effect. I think the money for the study (presumably put up by Lantus manufacturers?) could have have been better spent elsewhere. :-/

 

[phoenix]

@Spiker
HIgh fat one side,high carbs the other.
I'd rather sit in the middle of the boat, maybe not so rocky !
(Actually it annoys me that there is this two camp mentality. If you don't eat very high fat then you must eat very low fat. I assure you that the French cheese I eat is normally full fat; but I also eat French portion sizes)
Off to eat cold chicken, very large salad, with vinaigrette , new potatoes followed by strawberries.
Be interesting to compare with yesterday where I was very erratic because of unpredictable events.. No lunch, grabbed a raspberry tart at 4pm , ate dinner with a private party in a restaurant from 7pm . I had small amounts of bread, pasta, beef casserole, some cheese, some icecream ; no veg except in the casserole. Small portions but high percentage of carbs. Ended up taking normal amount of insulin fora day with no exercise; 26U but went to bed on 80mg/dl and woke at 77mg/dl. which is a improvement on recently.

 

[Spiker]

@Spiker
HIgh fat one side,high carbs the other.
I'd rather sit in the middle of the boat, maybe not so rocky !

Same here!

 

[Spiker]

The danger of the low carb / high carb debate is that the high carb proponents will run a straw man argument and say that the body needs insulin (which is correct) and it's nuts to try to eliminate it.

I don't think so. I don't think I've ever heard a high carb advocate (or opponent of low carb) say the body needs insulin. It would be a straw man argument, you're right, since no one is arguing to the contrary (no one is arguing "the body doesn't need insulin"). But I don't think anyone is deploying that argument.

Much more commonly, people (often they are HCPs) will claim "the body needs carbs" or specifically "the brain needs carbs" and they will say the body needs (RDA) carbs per day or the brain needs (eg) 100g of carbs a day. This is all spurious. Humans need zero g of ingested carbs to survive indefinitely. The glucose needs of the brain, nervous system and a few other specialised cells are modulated down during carb scarcity, supplanted by ketones and other products. The brain's residual need for glucose is easily met by GNG.

So yes I have often heard "the body/brain needs carbs" argument but never heard "the body needs insulin" as an argument. The immediate response from a low carb proponent would be "yeah, so?", so it really wouldn't have any mileage even as a straw man argument.

 

[Spiker]

Spiker, great link and website. I spent quite a bit of time on it this morning.

This article was really interesting too, "Diet Wars...Which One Is Optimal": https://optimisingnutrition.wordpress.com/2015/03/22/best_diet/ It provides a comparison between the different types of diets. Thanks!

@martykendall those graphs are superb and your number crunching is incredible.
I notice you don't have the Bernstein diabetic diet in there (unless it's called something else?), it would be interesting to see where the good doctor B fits in to the picture.

 

[Spiker]

The first published research using the FII to predict insulin response (normal subjects) is here: http://ajcn.nutrition.org/content/90/4/986.full#T1
The next one used carb counting/dosing for a high carb meal using this as a control , then compared carb counting and FI dosing for a test meal with a similar food insulin index to the control meal but far fewer carbs.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3177729/?report=reader

These results are very impressive in a number of ways. First they show that FII, and also fat content, allow bolus dosing to be much more predictable. Secondly the first article tells us that carb counting, protein counting, just don't work effectively to predict insulin demand. :-(
So maybe it's not our fault that we struggle to maintain good Hba1C even when conscientiously carb counting.

 

[martykendall]

Same here on both counts!

Me too.

I'm definitely enjoying the interaction with you guys.

 

[Spiker]

My review of the data from the University of Sydney didn't show that fat has an impact on insulin

The findings from the ACJN articles quoted above by @phoenix show that fat has a significant negative correlation with insulin demand and a significant positive correlation with predictability of dosing.

[Edit: the ACJN articles are U of Sydney's own analysis of its FII data set]

Though probably a lot of us knew both of those things already, they are not far off just being common sense.

 

[Spiker]

I think this is Bernstein's latest on dosing for protein - . If you asked me to write down what this means next time you go to calculate a protein dose I'm not sure I could. What I get from this is that it's something and not nothing.

Yeah. Sigh...

Here he says 1 oz of protein food typically needs 0.5u of Regular (non analogue) insulin vs 8g of (net) carbs needs 1.0u of Regular

Given that protein food is 20 to 25% protein by weight he is saying in effect:

6g - 7g of protein = 0.5u R; 8g carb = 1.0u R

So dividing through, he is stating what I would call a "protein ratio" in a range (~57% - ~66%) that is consistent with the figure he gives elsewhere in his book of 58%

He does also elsewhere in his book (slightly contradicting this video) say you have to subtract off the daily structural [protein!] requirement, and he gives a formula to guesstimate that daily structural need.

 

[martykendall]

I think you may be missing the point about Bernstein. Bernstein's pioneering method would today be called LCHP, Low Carb High Protein, and would be differentiated from LCHF, Low Carb High Fat. What many of us have found is that LCHP is ineffective precisely because the protein still metabolises to glucose. Protein is not so much an alternative to carbs, as "Carbs Lite". Many of us have found that LCHF is more effective, as it reduces the total glucogenic intake (carbs and protein) and so, through Bernstein's very own "law of small numbers", provides us with better blood sugar control. And, equally important, escaping from carb dependency, including GNG carbs, makes it psychologically and physiologically easier to maintain a low carb diet.

For example, for me, the Bernstein diet is a permanent battle with hunger. It's no different than being on a calorie restricted high carb diet. I get hunger cravings all the time and overeat protein to feed my carb addiction via GNG. Now I realise the straight Bernstein diet has and does work for a lot of people. I am not dissing it. But for me LCHF is more effective and more sustainable.

That's the reason that Bernstein is starting to become somewhat isolated within the low carb diabetic community. Which is a shame because he is practically our messiah and in many cases literally is our saviour.

I like the way you think Spiker!!!

There has been a discussion about LCHF vs LCHP on the TYPEONEGRIT Facebook page over the last couple of days after Bernstein said that you shouldn't limit protein for growing T1D kids, which largely makes sense. It's akin to saying that you shouldn't limit carbs for normal growing kids, because they need the energy. Bernstein does advocate winding back the protein if you want to lose weight, which makes sense as it manipulates the insulin load if you're already low carb.

So one thing I've been thinking about is... what is the optimal insulin load / total glucose? There's got to be a "sweet" spot.

Taubes focuses on carbs as the insulin culprit but now people have dropped their carbs and upped their protein to compensate they still may have an unacceptably high amount of glucose from GNG which leaves them with high blood sugars and high insulin and obesity.

I've been mulling around some thoughts for a new blog post titled "why we get fat and what to do about it" that brings protein into the picture as well as carbs.

I really like Paul Jaminet. He often talks about an optimal dose curve. Your blood glucose can be too high or too low. Your insulin can be too high or too low.

Similarly I think there must be a sweet spot for insulin load / total glucose (what is the right term?). In his Perfect Health Diet Jaminet does a really nice job of defining optimal ranges for carbs and protein but then doesn't bring the two together into a total glucose load. For example if you're on the upper end of the range for carbs and protein then I think you'll end up with a glucose load that's too high.

The way I see it total glucose load = [carbs - indigestible fibre] + [glucogenic amino acids from protein - amino acids used for repair of the body].

How do you determine the "goldilocks zone" for total glucose?

I think you have too much when your HbA1c is greater than 5.4mmol/L or your average blood sugar is greater than 5.4mmol/L and you are never showing ketones greater than 0.4mmol/L. See https://optimisingnutrition.wordpress.com/2015/03/22/diabetes-102/

On the other end of the scale you would have too little total glucose if you were under these blood sugar targets AND you had low energy, low mucus, low thyroid or any of the other (anectodal?) criticisms of ultra low carb.

The total glucose load would obviously vary based on your activity level and you would have to dial it in based on your own n=1 testing.

I'm thinking of running some numbers on a range of dietary approaches to see what a typical total glucose load would be for a range of dietary appoaches.

Thoughts?

 

[tim2000s]

I like the way you think Spiker!!!

There has been a discussion about LCHF vs LCHP on the TYPEONEGRIT Facebook page over the last couple of days after Bernstein said that you shouldn't limit protein for growing T1D kids, which largely makes sense. It's akin to saying that you shouldn't limit carbs for normal growing kids, because they need the energy. Bernstein does advocate winding back the protein if you want to lose weight, which makes sense as it manipulates the insulin load if you're already low carb.

So one thing I've been thinking about is... what is the optimal insulin load / total glucose? There's got to be a "sweet" spot.

Taubes focuses on carbs as the insulin culprit but now people have dropped their carbs and upped their protein to compensate they still may have an unacceptably high amount of glucose from GNG which leaves them with high blood sugars and high insulin and obesity.

I've been mulling around some thoughts for a new blog post titled "why we get fat and what to do about it" that brings protein into the picture as well as carbs.

I really like Paul Jaminet. He often talks about an optimal dose curve. Your blood glucose can be too high or too low. Your insulin can be too high or too low.

Similarly I think there must be a sweet spot for insulin load / total glucose (what is the right term?). In his Perfect Health Diet Jaminet does a really nice job of defining optimal ranges for carbs and protein but then doesn't bring the two together into a total glucose load. For example if you're on the upper end of the range for carbs and protein then I think you'll end up with a glucose load that's too high.

The way I see it total glucose load = [carbs - indigestible fibre] + [glucogenic amino acids from protein - amino acids used for repair of the body].

How do you determine the "goldilocks zone" for total glucose?

I think you have too much when your HbA1c is greater than 5.4mmol/L or your average blood sugar is greater than 5.4mmol/L and you are never showing ketones greater than 0.4mmol/L. See https://optimisingnutrition.wordpress.com/2015/03/22/diabetes-102/

On the other end of the scale you would have too little total glucose if you were under these blood sugar targets AND you had low energy, low mucus, low thyroid or any of the other (anectodal?) criticisms of ultra low carb.

The total glucose load would obviously vary based on your activity level and you would have to dial it in based on your own n=1 testing.

I'm thinking of running some numbers on a range of dietary approaches to see what a typical total glucose load would be for a range of dietary appoaches.

Thoughts?

I believe that your statement regarding too much protein causing significant GNG only holds true in the diabetic population. Due to the use of exogenous insulin, we are not as efficient at regulating GNG as non-diabetics.

But to focus on gng and insulin alone is a little like trying to live off water alone. Amylin plays an important part in the glucose metabolism and the rate of gut emptying and there are additional hormones that need to be considered. Amylin is particularly pertinent to the diabetic population as it is produced in beta cells so any damage there reduces amylin production and could be one of the causes of increased hunger, amongst other things.

There is an interesting article here on the hormones affecting the glucose metabolism and it is, as ever, more complex than we like to think. http://m.spectrum.diabetesjournals.org/content/17/3/183.full

 

[martykendall]

That's fine @martykendall . I was going to say that I've not tried it with a 100% fat trial, and then I remembered that most mornings I make a bulletproof coffee with Butter and Coconut oil

Unfortunately I can't show you the traces for those, but it's an easy enough experiment to redo. Just as a from memory exercise, I don't recall seeing any real impact on BG levels with that. Certainly not anything that made me stop, take a photo and post it on here. It usually contains 25g of butter and 25g of Coconut oil....

I just thought it would be interesting to compare protein versus BPC or something high fat given that the TAG approach says that you should allow 10% of fat as carbs. From what I understand you can't convert fat to glucose in the blood that would require insulin?!?!? Type 1s with CMGs are the perfect metabolic experiment (read: lab rat, guinea pig, canary in the coal mine) to demonstrate the action of insulin.