T Nation

Long-Term Low Carb Diet and Glucose Intolerance/Insulin Resistance

Hey guys, just wondering if anyone has experience with being on a low carb diet for a while then switching back to carbs. I have been eating fairly low carb for the better part of two years now, the other day I thought I would try a carb refeed day and I basically ate nothing but starchy carbs all day. I was very bloated for the next two or three days. Just wondering if anyone has any experience with this type of thing. I’ve been doing some research and some people seem to say Long term keto can cause insulin resistance and others say keto with improve insulin sensitivity. Any advice is much appreciated. Thx everyone

Making the switch back must be done slowly over 2-3 weeks or your body will store tons of fat. I switched from keto to full on carbs. Went from 12% to 18% bf way too fast. Since then I learned that you need to slowly add carbs back over a period of 2-3 weeks

This was probably secondary to your gut flora responding to the sudden increase in FODMAPs.

Hard to imagine how a low carb or keto diet could lead to insulin resistance. Is there anything in PubMed to this effect?

After a re-feed, isn’t everyone bloated for two or three days? LOL I mean really… I know I am.

Low carb, high fat dieting tends to produce a state called “physiological insulin resistance.” Muscles load up with fatty acids for energy and enzymes that burn glucose are produced much less, while fat burning enzymes increase in muscles. (Insulin itself has a function of up-regulating the synthesis of glucose burning enzymes). Note that fatty acids and glucose both end up providing raw materials for the Kreb’s cycle of energy production.

Since the muscles are full of fatty acids and are not burning glucose very fast, if you eat carbs in this state, their won’t be room to put them in muscles and there won’t be a mechanism to burn them off so extra glucose sits in the blood and in the extra-cellular fluid of the body. The glucose also pulls more fluid into the extra-cellular spaces.

Medical practice has recently changed with regard to glucose tolerance tests such that a glucose tolerance test taken while on a low carb (maintenance calorie) diet is invalid. Fasting blood sugar can be elevated to 90-125, and individuals will fail a glucose tolerance test often with blood sugar well over 140 at the 30 minute and at the 90-120 minute range. Some long term low carb high fat dieters have had diabetic blood sugars (200-265) on a glucose tolerance test. This condition largely goes away with a few days of higher carb intake. Medical practice now requires at least 150 grams of carbs a day for 2 days prior to a glucose tolerance test to get a valid result. There is some suggestion that carbs should be even higher in the days prior to a glucose tolerance test with failed tests occurring in people chronically consuming around 30% carbs.

Individuals who eat fewer carbs and more fat on a maintenance diet will secrete more insulin per gram of carbs ingested, often 200%-300% as much for a given amount of carbohydrate. This means that insulin exposure does not drop much on a low carb high fat diet if calories remain at maintenance level. In fact, insulin dependent diabetics who drop from a 50% carb diet to a 20% carb diet tend to use the same amount of insulin within 2-3 days because the body switches over to burning and storing fat in muscles. If they go back to a 50% carb diet, they tend to need much more insulin for 2-3 days, but then revert to their old level. Insulin needs do tend to drop off more dramatically at under about 20% carbs on a maintenance diet.

Physiological insulin resistance therefore is temporary, and a high fat, low carb diet may reduce insulin resistance over a long period of time because it reduces blood sugar spikes. Pathologically insulin resistance individuals often over-secrete insulin in response to blood sugar spikes and then experience late post-prandial hypoglycemia. So controlling blood sugar spikes may improve long term insulin resistance by preventing oversecretion.

High fat, low insulin diets can also cause de novo lipogenesis in the liver which CAN lead to non alcoholic fatty liver disease which is a chronic state of insuln 'resistance.

Now, keep in mind that eating in a calorie deficit will reduce insulin exposure and improve insulin resistance, but a low carb, maintenance calorie diet does not have much effect on insulin exposure because of physiological insulin resistance.

Low carb (maintenance calorie) diets do not improve insulin sensitivity on their own, but they can work 1) by helping to control calorie intake since they will reduce blood sugar swings and help reduce hunger, 2) by reducing peak blood sugar levels after meals which cause most of the damage associated with high blood sugar (blood sugar over 140 has been shown to destroy pancreatic cells) 3) by reducing pathological over-secretion of insulin due to higher blood sugar spike, and also stress hormone release in response to rebound hypoglycemia, 4) by reducing inflammation that triggers cortisol release by reducing intake of inflammatory/allergenic carb sources, like wheat and 5) by reducing net fructose intake mostly from sucrose as fructose tends to cause fatty liver disease over the years because fructose gets turned into fatty acids in and by the liver.

There are 4 ways that somebody can become pathologically insulin resistance and it is important to identify which one is at play with an individual. The most common way is hormonal-stess and poor sleep raise cortisol and adrenaline levels that block the effects of insulin. The second is fatty buildup in the liver which is usually due to high fructose diets and alcohol, but can be caused by high fat diets if you overeat. The third is high insulin exposure which is usually the result of simple overeating over a period of years. This causes desensitization of insulin receptors on the cell. The fourth is genetically un-sensitive cell receptors.

The first 2 are intrinsically linked to food choices which is why “calories in and out” is only relatively true. Eliminating fructose and alcohol and foods that stimulate cortisol can solve insulin resistance issues in most people. They can also help prevent overeating (mechanism 3).


Thanks for your response

Great post @mertdawg
Also would like to point out that constantly keeping insulin low will tend to lead to higher blood sugar, higher Cortisol, higher SHBG and TBG.

Anyways, no population ever has been on keto long term. Agreed that it could be a temporary useful tool to improve some cases of insulin resistance.

1 Like

Assuming you’re not referring exclusively to Type I diabetics, I think you’re confusing cause and effect here.

I was interested in that comment as well. I am looking into it but it does appear that low insulin levels raise SHBG https://www.ncbi.nlm.nih.gov/pubmed/7962291

Low insulin levels tend to slightly raise fasting blood sugar by turning up gluconeogenesis which gets its supply of amino acids from cortisol, http://high-fat-nutrition.blogspot.com/2007/10/physiological-insulin-resistance.html.

And there appears to be a complicated relationship between cortisol and insulin levels. Low insulin may raise cortisol because it is a sign to the body of low glucose intake and so cortisol may be released to turn up gluconeogenesis to provide glucose for the brain and working tissues. There have been some conjectures that low insulin levels leading to keto-adaptation eventually turn down gluconeogenesis since ketones take over for amino acid derived glucose, but I have not seen any proof of this. On the other hand, high cortisol can reduce insulin secretion as a counter regulatory mechanism (lowering the insulin makes it easier to turn on gluconeogenesis without wasting as much cortisol (raw materials) In this case, basically high cortisol raises insulin levels to maintain euglycemia, but eventually the body “gives in” and reduces insulin so that it doesn’t have to push as much cortisol. As you can see, there is some conflicting data on this which may depend on the period of time of higher cortisol levels.

Thirdly, high insulin can RAISE cortisol if it results in rebound hypoglycemia.

I am pretty sure that low insulin also does raise TBG. What are your thoughts?

Now low insulin levels due to a hypocaloric intake are likely to correspond to lower fasting blood sugar because the hypocaloric state clears out room in the muscles and liver to dispose of blood glucose even if gluconeogenesis and ketogenesis turn up.

I’m not confusing cause and effect as I believe that insulin manipulation (through food) is causative to a lot of adaptations from the body. I’m mostly basing myself on my knowledge (from pharma school) on what is insulin, what are insulin’s functions (well known or not). Note that higher SHBG or TBG aren’t necessarily a bad thing - you don’t want low SHBG nor low TBG which are obviously associated with diabetes.

I’d say I’m only stating what’s happening when people spike insulin too little; none of the above is desirable long term but could be interesting short term in the current insulin resistance epidemics. Caloric restriction also works very well indeed. I believe that’s tied to the point #3 that was made above - don’t overeat.

@mertdawg I appreciate the papers you’re sharing. I’ll need to get off the phone and bring in some homework myself when I have more time :slight_smile:

My issue concerned the suggestion that we (humans) can exert direct control over insulin levels. Absent exogenous use in a T1DM situation, we can’t. Thus, the assertion that “constantly keeping insulin low will tend to lead to higher blood sugar” is puzzling to me, as the only way to ‘keep insulin low’ is to reduce blood sugar levels (ie, the “hypocaloric intake” to which you referred).

As for the other signaling/carrier entities, I believe all are elevated by stress. Of course, a chronic hypocaloric state is a potent stressor.

Surely you can see (from above) why it seemed to me that you had.

Does spiking insulin prior to stress reduce cortisol levels? I really have never delved into these questions before but I do know that when type I diabetics reduce carb intake progressively, there is evidence that they start to experience hormonal blood sugar spikes that do not correspond to food intake. From what I read from the Mayo Clinic, it is believed that cortisol starts to cause non-mealtime blood sugar spikes when glucose intake drops to somewhere around 20-30% of daily caloric expenditure. It seems to abate when ketogenesis turns up though.

I don’t know. I would guess that it doesn’t, but it would be just that–a guess.

In the situation you describe, what are the T1DMs doing vis a vis exogenous insulin use?

So regarding T1Ds, first off, they will take mealtime insulin and basal insulin. Basal insulin will either be given as a single (or sometimes 2) long acting injection that provide a low amount of basal insulin 24 hours a day, or they will have a pump that provides a controlled amount delivered every 5 minutes or so to give an hourly rate.

Let’s just say that that hourly rate is .5 units per hour during the day and 1.0 units per hour when they are sleeping. The basal needs are typically fairly predictable and stable over the course of the day and from one day to the next, but when carbs cross that turbulent zone of around 100 grams a day or somewhere in the mid to low 20% range, there can be sudden blood sugar spikes at unpredictable times. One day it might be at 1:00 am for example. This makes programming basal insulin harder because the hormonal spikes do not occur at the same time of day and the basal needs become erratic throughout the day. If a T1D is monitoring blood sugar with a monitor they will get an alarm for this spike wherever they set their monitor alarm. Perhaps it is set at 130, so they may wake up to an alarm at 1:00 am because their blood sugar hits 130 and deliver a correction that should push it back down.

A problem though is that hormonal spikes tend to deplete counter-regulatory hormones over the next several hours. It also takes only a tiny deficit of insulin to bump up blood sugar. Typically a type 1 diabetic using a pump will have their blood sugar rise 25-40 points an hour if they turn off their insulin pump for an hour (occuring within about 2-3 hours of turning off insulin), so a deficit in basal insulin of just 4% over the course of the day will result in a gradual rise of 25-40 points.

So the situation I am describing is one where lower carbs may stimulate gluconeogenesis at unpredictable times creating a net deficit in basal insulin which they would manually add additional insulin to compensate for, but would also have to be alert to running lower over the next 4-6 hours due to acute depletion of cortisol, glucagon or adrenaline, or an acute desensitization to them.

1 Like

This is a pretty good summary of why low FRUCTOSE consumption is critical to long term improvements in insulin sensitivity.

1 Like

I’d say that because of the inverse relationship between insulin and cortisol, it would work this way. It works the other way around for sure, cortisol is indeed positively correlated with glucose

I’m very confused. So you’re saying we can’t lower fasting insulin levels, thus preventing/reversing insulin resistance, without drugs? Because that’s a quite mainstream way to approach IR.

This type of publications seem to show a significant Cortisol suppression after a standard glucose tolerance test, -43% in the control group


Whether stress significantly reverses that I can’t tell, it would depend on the nature of the stress I guess. Adrenaline is a powerful player too.

As I was at pain to emphasize above, I’m saying we can’t control insulin directly. We can indirectly influence insulin levels (to some degree) via manipulation of diet. So to claim (as you did above) that “constantly keeping insulin low will tend to lead to higher blood sugar” is tantamount to saying ‘constantly keeping blood sugar low will tend to lead to higher blood sugar’–a statement I’m sure you will agree is puzzling.

I think it comes down to chronic versus acute. Low blood sugar means low insulin levels on a relative basis, but chronically low insulin levels downregulate glucose disposal and so there is a self regulating drift back up somewhat. Ketogenic diets can raise insulin sensitivity in the short run by exposing insulin receptors to less insulin which makes them more sensitive (probably low insulin levels increase the number of insulin receptors on cells), however long term low insulin levels dramatically reduce the levels of enzymes for glycolysis and so fasting blood sugar (as well as blood sugar after a given amount of carbs such as a glucose challenge) will be much higher.

I interpret the cortisol info the same way too. Cortisol acutely raises insulin to manage blood sugar levels, but if cortisol is chronically high there is likely to be (and the study I posted supports) a decrease in insulin levels since this can accomplish the same thing as cortisol, ie elevating blood sugar levels. It would be efficient for the body that is chronically exposed to cortisol in response to lowish blood sugar levels to SOMEWHAT reduce insulin levels so as to maintain blood sugar for the brain. This allows the cortisol to also dial back.

But yea, its kind of like saying that high insulin levels cause low insulin levels because when you raise insulin you don’t need as much any more.

Oh I see where the confusion comes from with my wording.

Indeed, I had in mind a longer term period as per the title of this topic, but I assumed we were discussing ketogenic diets whereas low carb diets should spike insulin well enough via protein intake, and that shall do. Plus, proteins trigger glucagon too. So my bad, wrong context ha.