Here's a followup to this post: To Reiterate, Just In Case You Missed It: No Elevated Ketone Levels in the Inuit. In retrospect, I wonder what the reaction would have been had I titled the post different, like: "Inuit have normal glucose tolerance on their natural diet, but it goes to shit when in objective ketosis."
In that post, I showed that at no time have Inuit been measured to be in ketosis. It's merely assumed. On the other hand, there's no doubt that after an 80 hr. fast that they were in ketosis.
For me, the big message was not the state of ketosis on their natural diet, but what happened to their glucose tolerance when there could be no dispute about their deeply ketogenic state. Here's the plots, again. The first plot is their tolerance on their normal diet. The second plot is the same people after an 80-hr fast.
Normal glucose tolerance with adults receiving about 120 grams 12 hrs after last meal
Yet, when I put that post up, everyone was all on about the suggestion that the Inuit are not actually in a state of ketosis on their normal diet, while completely ignoring the glucose tolerance. All sorts of stuff was offered up, from them being "keto adapted," to Vilhjalmur Stefansson's obituary in the NYT.
Well, thanks to Duck, again, we actually get to see what happened to Stefansson and Anderson themselves in similar circumstances, only reversed.
At the conclusion of the "Bellevue Experiment," another lesser known study was published by Edward Tolstoi, who got to test Stefansson and Anderson with a glucose tolerance test once the year of meat was completed. Anderson's glycosuria is discussed heavily in the paper. Tolstoi also discusses the Heinbecker study that was referenced by Richard in the article.
The Effect Of An Exclusive Meat Diet Lasting One Year On The Carbohydrate Tolerance Of Two Normal Men, by Edward Tolstoi (J. Biol. Chem. 1929, 83:747-752)
Heinbecker studied the tolerance of Eskimos to carbohydrate. His subjects, by necessity, lived on a practically exclusive meat diet for years, before their carbohydrate tolerance tests were made. In spite of the fact that their diets were low in carbohydrate, the results of the tests indicated that they assimilated carbohydrate well. The blood sugar curves were within the normal range and the urine remained free of sugar. Is it possible that Heinbecker’s subjects derived sufficient carbohydrate-forming substance from the protein in their diet to keep the insulin producing mechanism sufficiently stimulated to handle large quantities of carbohydrate? His Eskimos consumed about 280 gm. of protein, 135 gm. of fat, and 54 gm. of carbohydrate of which more than half is obtained from the glycogen of the meat. This seems a likely explanation.
Well, let's see what happened to Stefansson and Anderson, in reverse.
The results obtained by these procedures are presented in Table I and Charts 1 and 2. Both show that a marked rise in the blood sugar occurred after the test meal following the previous high fat, low carbohydrate diet. The curve of K. A. not only increased in height, but the hyperglycemia was prolonged also. He presented a glycosuria. V. S. had only an increase in the height of the curve. His urine was sugar-free. After a mixed diet for the periods mentioned above, both men reacted normally to the glucose tolerance tests, the height and the duration of the blood sugar curve being well within normal limits, and the urine sugar-free.
And just as I've been saying over and over, based on simple intuition, the paper affirms.
The above results are not new. They were emphasized by Odin (3), Malmros (4), Stenstrom (5), Staub (6), Kageura (7), Greenwald, Gross, and Samet (8) and others. It was believed that during a period of a low carbohydrate, high fat diet, the need for insulin was diminished, with a resulting decrease in its production. Then, upon administration of a large quantity of carbohydrate to a subject subsisting on such a diet, the carbohydrate mechanism is heavily taxed. The production of insulin cannot keep up with the demand, the result being a hyperglycemia and often a glycosuria as well. The extensive and excellent work of Malmros (4) also supports the above view. He worked with normal human beings who had glucose tolerance tests after general diets as well as after high fat, low protein, and low carbohydrate mixtures. The duration of such diets varied from 1 to 23 days.
In every case a lowering of the tolerance to carbohydrate was noted irrespective of the duration of the preceding diet. Greenwald, Gross, and Samet (8) were of the same opinion. They believed that a diet consisting chiefly of protein does not lower the tolerance for glucose as much as a diet consisting principally of fat. This inference might be used to explain the difference between the results of Heinbecker (9) and the ones presented in this paper. [...]
If then a low carbohydrate, low protein, and high fat diet lowers the tolerance to carbohydrate in normal human beings, is it reasonable to assume that a high carbohydrate diet will raise the tolerance?
This too has been demonstrated. John (11) performed glucose tolerance tests on two normal subjects to whom he gave 100 gm. of glucose on 5 successive days. And although the peak of the blood sugar curve was 260 mg. on the 1st day, the maximum was only 90 mg. on the 5th day. Traugott’s (12) experiments concur with this work. He gave an initial dose of 20 gm. of glucose, obtaining an increase in the height of the blood sugar curve. When, an hour later, after the first 20 gm. additional amounts of glucose were given in various quantities, from 20 to 100 gm., no hyperglycemia was noted. Apparently the initial stimulus of 20 gm. evoked sufficient amounts of sugar-metabolizing hormone, to take care of the additional quantities of glucose given later.
Such experiments and the ones presented above lead to the belief that in normal human beings, the quantity of insulin produced is dependent upon the amount of carbohydrate ingested irrespective of whether this foodstuff is administered as such or derived from the protein fraction of the diet.
So there you have it, again and again. We're dealing with a convenient and clever, self-serving confirmation bias on a MASSIVE scale in the VLC communities. Everyone has convinced themselves that they're pre-diabetic—or even diabetic—because after a while on LCHF and they have some birthday cake, feel awful and measure, they have skyrocketing BG and conclude that they've gone and Caught the Debeetus; when most likely, they are suffering a reversible condition—just as it was with me and now, many others. Yep, I fixed it right up with a regular intake of beans, potatoes and rice. Still pretty moderate carbohydrate, but at least enough that my glucose tolerance now performs as a normal human should.
I'll finish with a comment I received on Facebook about these study results, because I like it and it introduces something else we've learned recently too, that gastric bypass cures T2D.
What that diet wrought is probably delayed first insulin response, which is what you see when these guys go on VLC diets. They're still clinging to the progressive beta cell die off theory when gastric bypasses have made it clear that glucose intolerance is multifactorial. This is why these low-carb ignorami still believe such procedures work only by limiting carb intake by reducing stomach capacity. Tell them to explain how an insulin-dependent T2 diabetic now gushes out insulin. One foundation that used to provide rationale for low-carbing gives way and you see how desperately these people are in denial.
Help victims of chronic VLC and LCHF dieting and spread the word.