Death by glucose – understanding (and reversing) type 2 diabetes

Regular readers will know that glucose from starchy carbohydrates or sugars can be dangerous and damaging for us. Glucose does this damage by a combination of inflammation, glycation, fat accumulation and insulin resistance. This can either result in, or contribute to, non-communicable diseases that are rife in western societies and that are increasing in prevalence across the world (in parallel with our increasing carbohydrate consumption). Examples: diabetes (type 2), obesity, heart disease, arterial disease, non-alcoholic fatty liver disease and cancer. As well, inflammation and insulin resistance damage the brain and are linked to depression (including severe depression), anxiety, ADHD and Alzheimer’s disease (as a few examples). To illustrate how strong this connection can be, it has been proposed by some scientists that Alzheimer’s disease be renamed Type 3 Diabetes.

It may seem crazy that one dietary nutrient can cause such widespread damage. However, glucose has a pivotal role in the metabolism of every cell in every organ of our bodies. Get this wrong, and there are going to be consequences. Dire ones, as it turns out.

The evolution of glucose metabolism

Why do we even metabolise glucose if it is so harmful? Evolution doesn’t find the best solution to a problem, it uses trial-and-error to incrementally improve the current position. I’m not going into details here, but essentially the first way cells started making energy was by fermenting glucose without a need for oxygen. Later, cells evolved mechanisms for burning other fuels (fatty acids, amino acids, ketones) with oxygen to generate much larger amounts of energy. They also evolved mechanisms for burning the byproduct of glucose fermentation. Because evolution elaborates on previous mechanisms, we retain glucose as a fuel. There is more on this evolutionary development in my ‘fermenting vegetables‘ post.

The goldilocks level of blood glucose

It is widely thought that the brain can only burn glucose, but this is wrong. It has a significant alternative fuel – ketones, derived from fat. However, there is a class of cells that are indeed dependent on glucose for their energy – our red blood cells. This is because they have jettisoned their internal machinery for burning fuels with oxygen (and even their DNA) in order to make more room for the haemoglobin that they transport around the body. Therefore, they can only use the glucose-fermentation step for energy and, consequently, we need a supply of glucose in our blood. But, we need a goldilocks amount – not too much and not too little – just right.

The just right amount is about a teaspoon of glucose distributed across our entire blood circulation. So, we don’t need much. Exceeding this amount chronically can lead to the non-communicable disorders mentioned. The remainder of this post is directed at glucose and diabetes, particularly type 2, and how a better understanding of glucose metabolism gives us a simple, practical and inexpensive way to prevent, ameliorate, or even reverse, type 2 diabetes.

Measuring blood glucose

There are three main glucose measures: 1. Fasting plasma glucose (FPG); 2. The oral glucose tolerance test (OGTT) and; 3. The percent of haemoglobin that is glycated (abbreviated A1c). Glycated means that a molecule of glucose has randomly stuck itself onto a molecule of haemoglobin and damaged it.

These measure different aspects of glycemic control. The FPG indicates the level that glucose has fallen to after a fast of at least 8 hours (or overnight). It represents a lower limit to blood glucose for an individual. The OGTT measures plasma glucose 2 hours after ingesting 75g of glucose. It is a measure of the impact of glucose and how effectively it can be controlled (although this test is unphysiological – it would be a challenge to consume that much glucose at once and without other macronutrients in daily life). A1c is a measure of the damage that glucose has done to the haemoglobin in red blood cells over their lifetime (of about 3 months). It indicates the average blood glucose level day and night over this period. Taking into account a host of other factors (e.g. age, sex, weight, family history, ethnicity, gestational diabetes, metabolic syndrome, blood pressure, blood lipid profile), a diagnosis of diabetes may be made if FPG is greater than 7.0 mmol/L, OGTT is greater than 11 mmol/L and/or A1c is greater than 6.4%.

The new concept of pre-diabetes

Pre-diabetes occupies a grey area between normal and diabetic. Pre-diabetes is a recent concept, and not without controversy. There are two main viewpoints – that of the World Health Organisation (WHO) and that of the influential American Diabetes Association (ADA). According to WHO, a FPG of 6.1–6.9 mmol/L indicates a condition called impaired fasting glucose (IFG). Likewise, an OGTT of 7.8–10.9 mmol/L indicates impaired glucose tolerance (IGT). One or both of these may lead to a diagnosis of pre-diabetes. The ADA sets a lower limit for IFG (5.6 mmol/L) and includes A1c levels of 5.7–6.4%.

What this means is that many more individuals (millions more) will be diagnosed pre-diabetic under the ADA vs. WHO criteria. When the ADA lowered its limit (compared to WHO) for defining pre-diabetes in 2003, it was estimated that about 25 million more Americans were newly labelled pre-diabetic. The prevalence of IFG in the US increased from 8% to 35% with the 2003 criterion. Over half the US population is now classified either pre-diabetic or diabetic by the ADA rules. This is good news for drug companies manufacturing diabetes medicines.

As it happens, the ADA receives significant funding from drug companies manufacturing diabetes medicines. They list 10 ‘Elite’ sponsors, nine of which are drug companies (the tenth is a medical technology company making diabetes devices). To be ‘Elite’, these companies must donate at least 1 million dollars (US) annually to the ADA, giving the ADA at least a 10-million dollar pool that depends on the profitability and co-operation of their donors. There are other levels of ‘sponsorship’. These conflicts of interest are not confined to the ADA. Every diabetic association I have looked at is compromised in the same way. Something to ponder – is diabetes a disease or an industry?

Further, the term pre-diabetes is misleading. Many people labelled pre-diabetic do not progress to diabetes, some will continue to live with their slightly elevated glucose levels, and some may even go on to re-normalise their blood glucose. As well, diabetes risk does not differ substantially between those labelled pre-diabetic and those without that label but with other risk factors. The WHO preferred the term ‘Intermediate Hyperglycaemia’ over ‘pre-diabetes’, as this simply describes the situation without implying anything causal or inevitable, however the use of pre-diabetes is now the norm.

Finally, there is only poor to fair reproducibility for pre-diabetic IFG and IGT measures. Reproducibility is greater for people with diabetes and for people with normal blood glucose, indicating less stability in the twilight zone between diabetic and normal. Caution should be exercised when interpreting a single test result, or even multiple results if they are inconsistent.

Diagnosis vs. Treatment

The diagnostic categories used for diabetes have not translated into meaningful treatment. We know this is so, because the prevalence of diabetes continues to rise, and diabetes management continues to stress health services everywhere.

The reason is clear – the treatment (to medically reduce blood glucose) does not address the cause (increasing insulin resistance and diminished pancreatic function). High blood glucose is a symptom, not a cause, of diabetes. As a result, the condition continues to progress and deteriorate over time. This is the fundamental flaw in current thinking – using the marker for diagnosis as the target for treatment:

Blood glucose is a convenient marker for detecting insulin resistance, but it is an inappropriate target for treating insulin resistance.

This is not to say that diabetic drugs have no role in the management of diabetes. For example, metformin can be thought of as an analogue of dietary glucose restriction as it prevents the liver releasing glucose into the circulation. Non-diabetic drugs will be needed to manage complications arising. And, in general, getting glucose out of the blood is a good thing too. It is just that these strategies don’t treat the cause.

Additionally, there’s a hidden catch if diabetic drugs (such as insulin or sulfonylureas) are used to lower blood glucose, because the question that is rarely asked is this:

If medications are getting glucose out of the blood, then where is it going?

Medications don’t get glucose out of the body, just out of the blood.

There are really only 2 main places to put it – into cells throughout the body, or stored as fat. Cells have a limited need for glucose, and excessive glucose causes as much damage to cells as it does in the blood. Indeed, this may be the basis for insulin resistance itself – cells opposing insulin’s efforts to stuff them full of glucose they don’t want. If instead, it goes into fat stores, then the eventual obesity does harm of its own (mostly inflammatory) and can further contribute to insulin resistance.

This mechanism (cellular insulin resistance and fat storage) makes sense if it only occurs intermittently throughout life – there will be times when glucose is superfluous to need, and insulin thinks ‘well, cells are resisting glucose now but they may need it later, so I’ll store it as fat until then’. The problem arises when this mechanism becomes active chronically.

In any case, good blood glucose control might be achieved medically, but systemic harm is being done silently everywhere else (diabetes is often referred to as the ‘silent disease’). This explains why, even with good glycemic control, people with type 2 diabetes eventually suffer damage to eyes, kidneys, nerves, blood vessels and the heart over the longer term.

Pre-diabetes and early prevention

What is the purpose of labelling otherwise-healthy asymptomatic people pre-diabetic? The standard response is – to identify people thought to be at risk of diabetes and to instigate early preventative measures. However, what constitutes early preventative measures?

There is broad consensus across authorities that this amounts to losing weight by calorie restricted diets and exercise. However, as I explained in my previous post, this is a poor approach to weight loss and stands an excellent chance of failing in the longer term. It is a particular challenge for overweight diabetics. Further, it gets the cause wrong again. Like high blood glucose, obesity is a symptom, not a cause, of diabetes. It results from insulin’s efforts to get glucose out of the blood by storing it as fat.

When lifestyle changes fail, the profession can legitimately turn to drugs to normalise blood glucose.

The unfolding story of insulin resistance and glycemic control

Assuming that a person is not overloading their body with dietary sources for glucose, and that their cells are sensitive to insulin, everything works nicely – insulin enables glucose to enter cells to be used as fuel or, in the case of muscle (or liver), to be stored as glycogen for later needs.

However, under constant dietary glucose overload, and augmented by a variety of risk factors, cells start to become resistant to insulin’s action. The pancreas responds by releasing more insulin – it tries harder.

This is the early silent phase – blood glucose will still be in the normal range because insulin resistance is low and because the pancreas is functional enough to compensate by increasing insulin secretion. This stage can be detected if insulin resistance (instead of glucose levels) is measured or calculated, which it rarely is.

With continued pressure to dispose of dietary glucose, and with unnaturally-high levels of insulin, insulin resistance increases. The pancreas will keep up-regulating insulin secretion while it can, but insulin may resort to its second strategy now – getting glucose out of the blood by converting it (in the liver) to fat and securely storing in the liver, or later in our adipose fat tissue. This is the beginning of non-alcoholic fatty liver disease and abdominal obesity.

The carbohydrate-eater is now gaining weight, however blood glucose could still be normal because insulin has been able get rid of the glucose into cells or fat. Blood glucose may be at the higher-end of normal though. A diagnosis of pre-diabetes is not made. This could be going on for years or decades.

Finally, the system can no longer cope, the pancreas starts to become dysfunctional, there is not enough insulin secreted to store glucose in cells or fat, and blood sugar measures now rise. This is the point at which a diagnosis of pre-diabetes might be made. However, we are now well into the pathology of insulin resistance.

With high insulin resistance and a functionally-limited pancreas, medicine turns to drugs to lower blood glucose. This does not improve insulin sensitivity, and the progression of diabetes continues until medical complications arise.

This is not to say that everyone’s story will (or needs to) unfold in this way, but rather it is to offer a perspective of the underlying processes. I hope this knowledge is empowering.

Is there something better to do?

Yes. If we’re having trouble getting glucose out of our bodies, then don’t put it in.

The solution is dietary, not pharmacological. Minimise dietary sources for glucose, and make up for calorie-deficits with fats. While insulin resistance is the term used to describe what is going on, what cells are actually resisting is glucose. So, take it out of the diet. With circulating glucose chronically low, insulin sensitivity may be improved or restored.

Our high-carbohydrate low-fat diets are a recent fad, and run contrary to our evolutionary biology. If we reverted to our heritage eating pattern (low-carbohydrate, high-fat) then there is every chance that our biology would breathe a sigh and return to its normal regulation of blood glucose, just as it has done successfully for the last 2-3 million years (excluding the last 100).

The degree of carbohydrate restriction will depend on the individual and their level of insulin resistance, however, even people with normal blood glucose (especially high-normal), or those gaining weight inexplicably, may be insulin resistant and could benefit from a lower carbohydrate diet.

The strategy is expected to be successful for people in the pre-diabetic range, but even manifest diabetes can be reversed with sufficiently-aggressive carbohydrate restriction. I usually refer readers to for further information, motivation, recipes and meal plans.

While Diabetes Associations contribute significantly to the problem of diabetes, one notable exception is This is a community of people living with diabetes that have set up a forum for the exchange of experiences and information outside the tightly-managed control of industry-funded organisations such as the American Diabetes Association or its Australian, Canadian, UK and International equivalents. The number 1 forum on this site is the ‘Low Carb Program’ for type 2 diabetes, which currently has 192,000 people participating. This is the future – people empowering themselves and rejecting the Standard Approach to Diabetes (SAD) because they know it doesn’t work.

There is no need for anxiety or alarm with any of this. Just make a start with a degree of application that is probably proportional to your FPG. Monitor as you go. The expectation is for a sustainably good outcome.

Surely we need carbohydrates?

We metabolise glucose, however, it is not an essential dietary macronutrient. The liver can make the glucose needed for daily living – we don’t need much of it (remember the teaspoon?). There is no such thing as ‘carbohydrate-deficiency syndrome’. The brain does not depend on glucose. There are no essential carbohydrates, whereas there are essential fatty acids (like omega-3) and there are essential amino acids (from proteins). It is usually recommended that diabetics limit carbohydrates to about 20-25g per day. This is mostly about the co-nutrients (such as magnesium) and fibre. Replacing carbohydrates with fats will not increase the risk of heart disease.

Current diabetes dietary guidelines

Where does this leave the current dietary guidelines promoted by diabetes associations worldwide? The ones that recommend eating carbohydrates and even condone sweets? I consider them to be a crime against humanity. Considering the prevalence of type 2 diabetes, and the dreadful progression of systemic symptoms leading to a premature (and unnecessary) death, I think that the definition of genocide should be extended to include the consequences of this officially-endorsed dietary program. The WHO estimated that 1.5 million people worldwide died (directly) from diabetes in 2012. That is the most recent number I can find (and the most recent number referred to by WHO in 2016). It can only have been increasing. It is happening annually. The organisations (and people) responsible for this need to be held accountable. And, in the meantime, ignored.

Back to the future

There is nothing new about treating diabetes with a low-carbohydrate diet. With type 2 diabetes beginning to emerge at the beginning of the 20th century, and because there were no drugs at the time, diet became a natural approach (just as it was for epilepsy).

It was well recognised that dietary carbohydrates increased blood glucose. The problem was – what to replace the carbohydrates with. There was a fear of replacing them with fats because some diabetics (it turns out these were type 1 diabetics, although that classification wasn’t known at the time) can have dangerously-elevated levels of ketones in their urine, and ketones can only come from fat. If ketones get too high it can lead to ketoacidosis (ketones are acids), coma and death. So, carbohydrates were replaced with protein, which had a deleterious effect because the liver converts excess protein to glucose (the body has no way to store unneeded protein, so it converts it into something else it can use).

The first account I have come across that directly studied the risk and benefits of dietary fat was published in two papers (in the Archives of Internal Medicine) in 1920 and 1921 by Newburgh and Marsh. While they do not report on the type of diabetes, it seems from their clinical descriptions that they were studying what we now know as type 2. The 1920 paper describes their daring objective:

… we have dared to ignore the belief concerning the danger of fat in the diet of diabetics, and have investigated in the clinic the effect of a diet whose energy comes largely from fat, to which is added sufficient protein to maintain nitrogen equilibrium and the minimal carbohydrate necessitated in making up a diet that a human being can eat over a long period of time.

This is an elegant description of the modern ketogenic diet. Note that their diet was designed to be sustainable over a long period of time.

They report success, stating: “(1) that glycosuria [high urine glucose] is avoided in severe diabetics; (2), that this diet does not precipitate [keto]acidosis; (3), that nitrogen equilibrium is maintained, and (4) that the patients are able to lead at least a moderately active, comfortable life.”

The 1921 paper reports the beneficial effects of their diet on blood (rather than urine) glucose.

Altogether they report on 73 cases of diabetes admitted to their clinic at the University of Michigan. The cases were all severe and advanced, because physicians only referred patients to them when all else had failed. Except for 3 patients who died within a day of admission and one who “went into coma after eating a bag of oranges brought by a relative,” glucose was normalised in everyone on their diet, none developed ketoacidosis and the patients maintained “excellent condition months after leaving the clinic.

Why didn’t this sensible dietary approach persist? Well, 1921 was also the year that insulin was discovered. In 1922, the first (type 1) diabetic child was successfully treated. As soon thereafter as 1923, the Nobel Prize for Medicine was awarded for the discovery of insulin.

The rest is history.

Print This Post Print This Post

Note: The content of this post is intended to be educational and informative. It should not be taken as medical advice.

Further resources:

My latest post on obesity is relevant companion reading.

Two older posts on diabetes: Diabetes and diet, Diabetes – update

The ketogenic diet page:

Appallingly, the two pioneering low-carb high-fat dietary studies from 1920/21 are still behind a pay-wall, despite being ~100 years old. Here are the references anyway:

Newburgh, LH and Marsh, PL (1920) “The use of a high fat diet for the treatment of diabetes mellitus, first paper” Archives of Internal Medicine 26(6): 647

Newburgh, LH and Marsh, PL (1921) “The use of a high fat diet for the treatment of diabetes mellitus, second paper, blood glucose” Archives of Internal Medicine 27(6): 699

Tabak, AG et al.(2012) “Prediabetes: A high-risk state for developing diabetes”. The Lancet 379(9833): 2279–2290. An interesting title because they do not argue that strongly for it to be a high-risk (they seem to be addressing it more as a conjecture).

Kim, SH et al. (2006) “Comparison of the 1997 and 2003 American Diabetes Association Classification of Impaired Fasting Glucose”. Journal of the American College of Cardiology 48(2): 293

Jason Fung, Intensive Dietary Management, for the insight that diabetes medicines are just hiding glucose by taking it out of the blood and putting it somewhere else in the body.

I didn’t discuss early starvation diets for diabetes because they were later discredited. However, if interested, see: Mazur, A (2011) Why were “starvation diets” promoted for diabetes in the pre-insulin period? Nutrition Journal 10:23

Abbreviation List:
A1c: Glycated haemoglobin
ADA: American Diabetes Association
FPG: Fasting Plasma Glucose
IFG: Impaired Fasting Glucose
IGT: Impaired Glucose Tolerance
OGTT: Oral Glucose Tolerance Test
WHO: World Health Organisation
Go Back