Type 1 Keto

Healthcare Professionals’ Manual


Therapeutic Nutrition in Type 1 Diabetes; A Guide for Health Care Professionals.

PLEASE NOTE. The content of the Healthcare Professional Pack provides information that might be of use to someone with Type 1 Diabetes who is looking to transition to a therapeutic nutrition approach to their condition. It is intended to facilitate discussion between the patient and clinician and not to either provide advice or get in the way of the clinician to patient relationship.

“People with Type 1 Diabetes should have the right to normal blood glucose” – Dr. Richard Bernstein, pioneer of a lower-carb approach in Type 1 Diabetes.


This Type 1 Keto HCP pack has information to provide healthcare professionals with an overview of a low carbohydrate approach to Type 1 Diabetes. It details the benefits, principles, and practice of a therapeutic nutritional approach to the condition.

For many people, reducing carbohydrates in the diet is the beginning of a journey towards improved glycaemic control, increased day to day safety, and improved quality of life.

The pack provides the rationale, evidence, and practical information needed to work with people with Type 1 diabetes to help them achieve their goals. There is a wealth of dietary information on the types of food that would be ideal for someone with Type1 looking to restrict the carbohydrate in their diet. There are no recipe calendars or fixed regimens, but there is an abundance of recipes that set out to provide examples of recipes and food choices that someone with Type 1 will find helpful when shopping and in the preparation of food. The emphasis is on a practical approach to food preparation and selection that will suit those with time pressure and financial pressure. There are plenty of hints and tips on being opportunistic to get the best value out of a shopping trip.

This course has been produced around both personal and professional observation of the needs of people with Type 1 who are interested in a low carb lifestyle. Data collected elsewhere has shown that the concerns of people with Type 1 diabetes are not necessarily aligned with those of the clinician. Type 1’s were predominantly concerned with mental health and lifestyle issues compared with biological markers that clinicians usually are more used to.

Not all people are at the same stage in their journey to achieve this aim straight away, and therefore the setting of goals that are in alignment with the individual’s hopes is essential, as with any change process. A simple way to do this is to use the GRIN model: Find out what Goals the patient has concerning their diabetes management at present; what Resources can be utilised to help them achieve that goal; a practical and realistic way of making changes in small Increments if needed; and finally, making arrangements for following up, emphasising the positives, reassessing progress and reviewing the goal (Noticing ).


Nutritional ketosis is a healthy state and happens irrespective of diabetes status. It is the production of ketone bodies as the end result of the regulated metabolism of fat in the absence of carbohydrates. Ketone bodies (acetone, acetoacetate, and beta-hydroxybutyrate) are water-soluble and energy-dense. Some ketone bodies are a valuable source of energy in themselves. For example, Beta-hydroxybutyrate (BHB) is easily transported to the brain, which can use this molecule for its energy needs.

Ketone levels for people in normoglycemia will vary from 0.3 to 3 mmol/l in the blood of someone in nutritional ketosis. This is safe. It is not uncommon to find up to 3+ of ketones in the urine of patients who are using a very low carbohydrate diet, but the glucose levels will be normal. If ketones are present in excess, they are simply converted to acetoacetate, which is readily excreted in the urine. Ketones can also be converted to acetone, which is excreted through breathing.

Nutritional ketosis is not a state of starvation but represents healthy metabolism. It is the result of choosing to take energy in the form of fats instead of carbohydrates.

In diabetic ketoacidosis (DKA), however, there is a state of insufficient insulin which can happen in Type 1 diabetes due to overconsumption of carbohydrates, infection, or underdosing with insulin, uncontrolled fat-burning ensues. Uncontrolled fat metabolism will lead to increased ketone bodies. These excess ketone bodies will cause acidosis. Insufficient insulin for the needs of the body at that time leads to increased glucose partly generated by the removal of the insulin ‘brake’ on glucagon, partly by dietary carbohydrates. Polyuria, dehydration, and electrolyte disturbances follow. This is a medical emergency. Contrast this with the controlled and intended natural production of ketones in nutritional ketosis when sufficient insulin is present.

It is imperative, however, for patients to understand that they need to be vigilant when in nutritional ketosis. Glucose levels rising might be an indication of insufficient insulin. DKA can also, by itself, cause temporary insulin resistance. It is also essential to make the patient aware of the possibility of attempting to overcorrect a raised blood glucose in this situation. Recovery from DKA can be sudden and might lead to a hypoglycaemic state if treatment is too aggressive. As always, close supervision is recommended.

All people with Type 1 Diabetes will already have been taught how to calculate a corrective dose of insulin for raised glucose levels as part of their routine care, but there is no harm done by enforcing this point.

Sleep, stress, and physical activity play a significant role in the day to day diabetes management, and there are ideas on the varied ways to manage this.



We do know that the EPIC Norfolk paper concluded that ‘Glycated haemoglobin concentration seems to explain most of the excess mortality risk of diabetes in men and to be a continuous risk factor through the whole population distribution.’ It is therefore desirable to try to achieve near normoglycaemia in Type 1 Diabetes if at all possible. http://www.bmj.com/content/322/7277/15?ijkey=34136a7ae31af9ef58502ca8a32324070317d353&keytype2=tf_ipsecsha

Unfortunately, this ambitious target received a setback after reporting of the Accord Trial, which attempted to treat participants with type 2 diabetes to normal levels but did not demonstrate any improvement of cardiovascular outcomes and worryingly an increase in mortality. https://www.nhlbi.nih.gov/news/press-releases/2008/accord-clinical-trial-publishes-results

One reason for this could have been the carbohydrate content of the diet (55%), which would have increased the plasma insulin levels and hence systemic inflammation.

An important trial was published in 2012, looking at the potential to lower blood glucose safely to improve glycaemic control in Type 1 Diabetes. It concluded that:

.An educational program involving a low-carbohydrate diet and correspondingly reduced insulin doses for informed individuals with Type 1 Diabetes gives acceptable adherence after four years. One in two people attending the education achieves a long-term significant HbA1c reduction.

This was a trial in Sweden, which reported a significant reduction of 1.3% (7.7-6.4%)(61-46mmol/mol) in HbA1c with adherent persons in the experiment (48%). The normalisation of blood glucose was immediate on day 1 of the trial. https://dmsjournal.biomedcentral.com/articles/10.1186/1758-5996-4-23

In another trial by the same author reporting of severe hypoglycaemic episodes reduced from 2.9 per week to 0.5 per week after one year in participants on a relatively modest carbohydrate intake of 70-90g per day (current guideline 250- 300g/day). Daily bolus insulin doses reduced from 21.1 units per day to 12.4 units per day at 12 months. https://www.ncbi.nlm.nih.gov/pubmed/16454166

A short study by Krebs reported similar results on HbA1c reduction for low carbohydrate diets in Type 1 Diabetes. They also reported that there was no detrimental effect on lipid profiles. The study was very short, just three weeks, but anecdotally people with T1 using a low carbohydrate diet report either no change or favourable change in lipid profiles. Data obtained from this Type 1 Program will shed more light on this area. https://www.ncbi.nlm.nih.gov/pubmed/26965765

Carbohydrate restriction diets compare well with the DAFNE trial. DAFNE teaches insulin management around the currently recommended conventional food. In the DAFNE trial, there was a reduction of HbA1c of only 0.36% (4 mmol/ mol) at four years.

The REPOSE Trial combining DAFNE with intensive insulin management using a conventional diet reported an HbA1c reduction of 0.85%-0.42% at two years (multiple injections v’s pumps in those with T1 with a similar starting HbA1c to the Swedish trials mentioned earlier). Analysis of the data from the REPOSE trial shows that a tiny 3% of those participants achieved the NICE guideline target of 48mmol/mol (6.5%). http://www.bmj.com/content/356/bmj.j1285

Compare this with data from the Facebook community, Type1grit where almost the exact opposite percentage achieve or commonly exceed the

NICE guideline target. Type 1Grit responders using a diet averaging 35g of carbohydrate had a mean daily carbohydrate intake of 36 g. The reported mean HbA1c was 5.67% ± 0.66%. http://www.thejournalofdiabetesnursing.co.uk/media/content/_master/3071/files/pdf/jdn16-9-364-9.pdf

Improving glycaemic control reduces not only physical complications but also mental health measures, which are equally, if not more important, to the individual concerned. Depression is more common in diabetes but improved glycaemic control, and indeed, a sense of being in control makes a significant difference in Type 1 Diabetes (Bernstein University Module 42-YouTube).

To back this up, a study from the University of Illinois on young adults found that as fear of hypoglycaemia and fear of complications increase, quality of life decreases, and as self-efficacy, or the person’s belief in themselves to achieve goals and positive outcomes increases, so does the quality of life. https://journals.sagepub.com/doi/full/10.1177/0145721718808733

Quality of life was not associated with either diabetes self-management behaviours or diabetes knowledge. Thus, fear of hypoglycaemia, fear of complications, and self-efficacy may be appropriate primary outcomes to use in interventions designed to improve the quality of life in young adults. https://www.ncbi.nlm.nih.gov/pubmed/26628250

One person with Type 1 put it like this; ‘a sense of personal pride comes about when a patient takes control of their own health, they then will be more likely to seek out other positive health behaviours, and to pass on the experience to others as their confidence improves. Personally, I believe part of the problem in healthcare surrounds the deferral of responsibility to the HCP. I look back on my own experience of this in T1D management as having been nothing short of disastrous’.

Health Care Professionals will have the opportunity through collaboration to guide their Type 1 patients towards achieving their own goals but also towards fulfilling the recognised biological measurements that are known to reduce complications. This is important because it has been shown that a reduction in HbA1c and a positive change in hope is associated with an increased frequency of monitoring in young people.

The latest NICE guideline for adults with Type 1 Diabetes NG17 recommends a target HbA1c of 48mmol/mol or lower. This is pragmatic figure worked around a compromise between the frequency of complications and hypos. This guideline also expressly excluded diets involving carbohydrate restriction with no explanation for why. This can be found in Appendix C section 1.2.2 p46. https://www.nice.org.uk/guidance/ng17/evidence/appendices-af-pdf-435400238

However, there is good evidence, provided above, that carbohydrate-restricted diet is a safe and effective strategy in Type 1 Diabetes, and there is also a wealth of anecdotal evidence involving thousands of individuals with Type 1 Diabetes using low carbohydrate diets. It is becoming increasingly recognised that daily glycaemic variability might be worse than even constant high glucose. Ketogenic diets smooth out peaks and troughs of blood glucose and also can maintain near normoglycemic levels.



Type 1 Diabetes is, in essence, a hormone deficiency condition. Unfortunately, that hormone, insulin, acts by reducing blood glucose rises in response to food, amongst other things. And the food makes it difficult to manage Type 1 in practice. It is challenging to estimate carbohydrates by estimating the amount in a meal, and it is just as tricky to time insulin injections and doses to achieve smooth control of blood glucose.

Although many endogenous metabolic pathways raise blood glucose, dietary carbohydrates are still the primary source of glucose found in the body. Whereas internal metabolism can produce up to 180g of glucose per day if required, the typical diet contains around 300g of carbohydrate a day. Most of that will be absorbed as glucose, even though it may be eaten as starchy food such as grain and potato. Starchy carbs are ultimately absorbed into the bloodstream as glucose.

In the management of Type 1 Diabetes, insulin is required not only to control glucose produced endogenously (typically basal insulin), but it is also needed to manage glucose ingested as carbohydrates (usually bolus insulin). Lowered carbohydrate means lower doses of bolus insulin. The less insulin, the less visceral fat, which is not only associated with increased insulin resistance but also significantly increased cardiovascular risk.

SATIETY. Lowered carbohydrate means increased fat. People drop their craving for carbohydrate food, although managing what is essentially carbohydrate addiction needs explanation and patient support. This makes managing the eating, carb counting, bolus injecting cycle so much easier to cope with. People will be able to drop to 2 meals a day instead of maybe eating six times a day, including snacks.


So, it makes sense to reduce the carbohydrate content in the meal to increase the chance of getting the numbers right. Smaller doses of insulin lead to smaller swings of blood glucose, which lead to a reduction in too-high and too-low blood glucose levels, which leads to better and ultimately safer control.

It is fortunate that of all macronutrients, it is carbohydrate that is the only non-essential macro-nutrient. It is also fortunate that this non-essential macro-nutrient is by far the most likely to raise blood glucose. So, removing most carbohydrates from the diet can make a large difference to glycaemic control without leaving the patient nutrient deficient. 10.1136/openhrt-2015-000273 https://openheart.bmj.com/content/2/1/e000273

Protein also raises blood glucose but, to a lesser extent and takes longer to achieve that effect. On a higher carbohydrate diet, its effect is masked by carbs but will become apparent with the move to lower carbohydrate.

Fat has a minimal effect on blood glucose levels, so it is easy to replace carbohydrates with fat.

Low carbohydrate diets based around real, are nutrient-dense. Fat and protein contain more than enough nutrients to replace those lost in the carbohydrates. But it must be emphasised that low carbohydrate diets are not NO carb.

Leafy, fibrous foods typically have some carbohydrates, but the quantity can be managed well with lowered doses of insulin.



The beauty of the carbohydrate management approach is in its simplicity. The method is familiar to both clinicians and patients and has the usual three aspects:

1) Basal insulin

2) Carbohydrate counting

3) Bolus insulin


The basal dose that the patient is already taking is probably accurate but can be refined by periods of fasting. Most people nowadays would recommend twice-daily basal insulin split 50:50, even for the latest ultra-long-acting insulins. This method of fine-tuning basal dosing is explained in the core module.


The bolus dosing principle is nearly the same as the person is already doing: count the carbs and inject insulin in the ratio that they have previously been taught in the clinic. On a very low carbohydrate diet, the blood glucose will start to normalise from day one, and because of the familiarity of the method, people with Type 1 Diabetes should take to it very quickly. The exchange between patient and clinician can be adapted to the individual’s needs and frequency of contact tailored to suit.


Protein ingestion should be considered at an early point. As the participant gets better at managing carbohydrates to a lower level, they will start to notice the effect of the protein, a slow rise in blood glucose at 2-3 hours due to glucose production, which is a metabolic consequence of dietary protein.

The protein plateau would have always been there but masked by the more substantial glucose-raising effect caused by the carbohydrate.

Modern rapid-acting insulins are designed for high carbohydrate diets and have a necessary rapid time of onset. When carbs are removed from the menu, it might be required to alter the timing of injection. Ultra-rapid insulin might necessitate doses during or even after a meal. Consider less aggressive insulin such as Actrapid or Humulin S. Some patients prefer this as it enables them to control their protein intake more effectively.

A rough guide is to count the grams of protein and inject about half of the usual ratio that has been worked out for carbohydrates, e.g., if the ratio of insulin to carbohydrate is 1 unit to 10g carbs, then it is 1 unit of insulin to 20g protein.

This is a good starting point and can be refined through the timing of the dose and amount as required for each individual.


Typically, a patient will find that they are injecting less insulin than before. If they have access to a continuous glucose meter, the learning may be faster.

We should be aiming for a situation where every T1 should at least be eligible for a CGM.

Patients are recommended to allow 5 hours between bolus insulin injections to allow the previous dose to be degraded. If not, it could be that the hypo risk increases due to insulin stacking.

It is inevitable that the daily doses of bolus insulin will drop. The basal insulin dose is likely to stay the same. There are reports online of people coming off insulin altogether. These are mostly people with type 2 misdiagnosed as insulin deficient. It is vital to reinforce to the patient that the object of this programme is solely to normalise glucose levels. The patient will almost certainly reduce their insulin doses but should not be given the impression that this diet is a cure.


Most people who have had diabetes for over a decade will have some insulin resistance as a result of the chronic over-injection of insulin required to manage the higher than needed amounts of carbohydrate in their diet. As the HbA1c normalises, some people find that they get more sensitive to their insulin and need to reduce the basal dose, this happens over months if at all. It is something to bear in mind. They also need to be aware that insulin sensitivity to bolus doses improves and they might find that the time to act of their basal insulin decreases

Once a person has learned to perfect their dose of basal insulin, some opt to skip meals, especially at times of day when diurnal insulin resistance is highest (most often breakfast when required bolus doses are usually higher most people). This can be used in some cases to improve glycaemic control, and this is a frequent topic on some low carb forums.


Reported beneficial effects of lowered carbohydrate in Type 1 are increased clarity of thought, improved visual acuity, a sense of energy, and a reduction in aches and pains. The latter is likely due to a decrease in the glycation of tissue. Age-Related Glycation End-products (AGEs) formed as a result of chronic hyperglycaemia can damage tissue function and promote inflammation. Glycation of tissues is not unique to erythrocytes (which are easy to access to measure HbA1c). But arranging an HbA1c blood test every three months can be very motivating

There can be an improvement in mood swings and also depression, especially in people who previously found control difficult with higher carbohydrate diets. It might be useful to measure wellbeing through a suitable scoring system such as Warwick Edinburgh or similar. Measurement of this at intervals throughout the year can also be insightful.


Type 1 Diabetes follows the same principles as all low carbohydrate diets: real food, healthy fats, reduced carbohydrate.

There is no science behind three meals a day. This has evolved around social habits and working patterns. Two meals a fast of 16-18 hours is often practiced by those eating low carb. It is perfectly safe in type 1 on a very low carb diet. Three meals a day plus snacks is even worse and is nothing but extreme marketing. Avoid snacks if you can. You might not be hungry at all, likely bored, sugar-addicted, stressed, or dehydrated. But it all depends on what suits you as an individual.

The most crucial factor in managing the diet in a keto lifestyle is to cut the carbs. This is the most effective strategy for achieving satisfactory blood glucose control. Very low carb or ketogenic diets are the most effective way to do this.

There is no single dietary strategy that can be recommended for all. The body is a unique organism and might have different nutritional needs at various stages of life. We, therefore, have individual nutrient requirements. Some do well with meat, some dairy, some plants. Within each food group, there will be certain types of food that are unsuitable. An obvious example would be a vegetarian with coeliac disease. They can eat most plant-based foods except wheat. But wheat might have been perfectly okay at an earlier phase of their life. This principle operates more subtlety for a whole range of foods. Individuals will react differently to the same food ingredient. We have to factor in the metabolic changes of ageing and the effects of hormonal profiles as we age.

The majority of the population have a mixed diet, and this is the subject of this article. The principle approach to a ketogenic diet is to work with your personal food preferences and reduce the carbs around those preferences. It might be the case that you will need to experiment around the different nutrients to get the formula right. It takes time and some degree of experimentation.

Some people will be u

altogether and still manage ketosis with nutritional completeness.

The guidance below gives information on the foods that are suitable for a ketogenic diet.


All green vegetables and salads are great. Eat as much as you want. Try substituting mash, pasta, or rice for vegetables such as broccoli, courgettes, cauliflower, or green beans. Cauliflower can be made into a rice substitute. Pizza can be kept in the diet by substituting the carb-dense base for a cauliflower version. Courgettes can be a substitute for spaghetti if they are spiralised.

Root vegetables, in general, contain significant amounts of carbohydrate, especially potato and parsnip. Swede and carrot are less carby, but you do need to count the carbs. Celeriac is low in carb. Again, count the carbs.


Meat, preferably the fatty cheaper cuts, is excellent food on a very low carb diet. Offal such as liver, kidneys, and heart, is often overlooked but is full of nutrition and extremely cheap. Eggs (three eggs a day is not too much), and oily fish such as salmon, mackerel or tuna, are all great sources of protein and can be eaten freely. Most protein building block molecules called amino acids are not essential. These can be made from other nutrients. But there are essential amino acids that need to form part of the diet. There are proteins in plants, including green vegetables. Other sources of protein from plants are soy, natto( fermented soy), black soybeans, lupini beans, pea protein, hemp, and chia seeds—also nuts and nut butters. Take care to note the carbs in some nut kinds of nut butter.

The absolute minimum requirement for protein is 0.61g/kg of ideal body weight. Various organisations have added a ‘safety buffer,’ and often 0.75-1g/kg is often quoted. This is the recommended amount to consume to maintain weight and preserve muscle mass. Experts in nutritional ketosis have recommended slightly higher amounts of protein, 1-2g/kg of ideal body weight. Some people might have other reasons to eat more than this. Children and growing adults, pregnant or breastfeeding women, and those looking to gain muscle mass for whatever reason should consume more protein. Adults will need to adjust their protein intake to suit their own needs.


Healthy, naturally occurring, and unprocessed fat-containing foods are okay. Fats usually come as part of food such as cheese, nuts, and oily fish. It is often said that the fat comes along with the protein. So if you choose a protein, there will mostly be fat with it. Fat is a source of the fat-soluble vitamins, A, D, E and K

Olive oil is especially useful. Butter is a better choice than margarine. Avoid margarine as it is a processed food with additives. Coconut oil is excellent for frying, as is olive oil. Olives eaten whole contain 15% fat ( Olive oil). Avocado oil, linseed, and some nut butter, especially almond butter and coconut butter, are also good sources of fat.

Processed ‘vegetable oils’ are not the healthy option that was once believed.


Some tropical fruits like bananas, oranges, grapes, mangoes, or pineapples are very carby and should be avoided. Berries are better and can be eaten in small amounts. A lightly cupped handful of around ten blueberries or 2-3 strawberries is acceptable. But count the carbs. These things can soon add up over a day. There are many other berry fruits. Apples and pears must be carb counted as some varieties are overly sweet. Fruit juices are to be avoided.


The occasional treat of dark chocolate (85% cocoa or more) in small quantity is fine to have if the carbs are allowed for. When sugar intake is reduced or stopped, many people find their tastes change, and higher cocoa chocolate becomes the most enjoyable. Nuts such as almonds, macadamia, walnuts, pecans, hazelnuts, and brazil nuts are the less carby options. Cashews and peanuts can be higher in carbs. You will need to count the carbs as they are easy to overeat. Be mindful that snacking is causing continual stimulation of the digestive system. This might lead to glucose swings and possibly insulin resistance. Best to stick to mealtimes, if possible.


Processed meats such as bacon, ham, sausages, or salami are okay. They are not as healthy as unprocessed real meats because of the additives used in the processing. The scientific consensus on their acceptability changes almost daily, so enjoy in moderation.


Cheese is an excellent source of healthy fat. It can make a tasty snack. Note that it is easy to consume a lot of calories when eating cheese inadvertently.



Sugar needs to be cut out completely, cakes, biscuits, cereals, and snack bars. Yes, even those ‘healthy’ ones (marketing again).

Table sugar contains 50% fructose. Fructose does not stimulate insulin secretion, but the high quantities of fructose being consumed nowadays are leading to insulin resistance and non-alcoholic fatty liver disease (NAFLD). While this is mostly associated with type 2 diabetes, it also complicates the management of Type 1 diabetes. If you have had Type 1 over a decade, then you are likely to have some insulin resistance because of the higher quantities of insulin injected to cover unnecessary carbs. If you have gained too much weight, then you are likely to be insulin resistant.

BEIGE CARBS. It should be your goal as someone with Type 1 diabetes to aim to cut out carby foods. There are unavoidable carbs in so-called low carb foods, and adding carbs will make it challenging to reach keto levels of carbs every day. Carby foods include bread, pasta, and rice, including whole-grain versions of these foods, as they will still increase blood sugar levels to the same degree, albeit over a slightly more extended period. There are substitutes. Rice can be made from cauliflower. Linseed can be used in a bread mix. See the recipe pages.


Sweeteners have been shown to increase hunger and may make weight loss more difficult. ‘Natural’ sweeteners such as Stevia, erythritol, and xylitol will not raise blood glucose in most cases. Artificial sweeteners such as Saccharin, Aspartame Acesulfame, and Neotame have zero calorific value but can cause insulin resistance. They are many times sweeter than table sugar and can fool the brain into preparing the bowel to digest carbohydrates even though there are none in the meal.


Many alcoholic drinks contain significant amounts of carbohydrates. Beer contains the same amount of glucose as a slice of bread. But low carb beers are becoming available. Another choice would be the occasional spirit. Take care with mixers, which often contain a lot of sugar. Use low-calorie mixers if needed. Go for extra dry Prosecco, which is the lowest carb of any wines at 1%. A full-bodied wine could contain as much as 5% carbohydrate. The best red choice might be a Pinot Noir. The best white wine choice

would be a dry Sauvignon Blanc.

Be aware that alcohol itself blocks glycogenolysis and gluconeogenesis. This means that alcohol blocks the body’s ability to raise blood glucose in a hypoglycaemic attack. Grain and grape drinks that are fermented, such as beer and sweet wines can contain significant amounts of sugar. The combination of a glucose-raising substance such as the grain in beer, and the alcohol, can make diabetes management quite trick.


Cholesterol is not water-soluble, so it must be transported in the blood with lipoproteins. These include low-density lipoproteins (LDL) and high-density lipoproteins (HDL).

LDL-cholesterol transports cholesterol from the liver out to the cells of the body, while HDL-cholesterol recycles/returns cholesterol to the liver. Traditionally LDL-cholesterol has been known as ‘bad cholesterol’ and HDL-cholesterol as ‘good cholesterol.’ (although this is by and large a crude oversimplification).

LDL-cholesterol might no longer be as important as other lipid markers such as HDL and Triglyceride.

Cholesterol is commonly misunderstood. Although it becomes problematic when coupled with inflammation, it is required for many bodily functions

The vast majority of patients who use a ketogenic diet experience an improvement of critical lipid parameters or remain stable ( HDL/TG ratio increases). 10% demonstrate an increase in total cholesterol and LDL and are regarded as hyper responders. Whether this phenotype is indeed associated with an increased risk of cardiovascular disease is not known. But LDL particle size regularly demonstrates a beneficial pattern.


LDL particles vary in size, from ‘small-dense’ through to ‘large-fluffy.’

The number of small-dense particles appears to correlate with cardiovascular risk directly, and this may be a better measure than total LDL-cholesterol level.

Typically, we do not measure LDL particle number or size. However, small-dense LDL particle number seems to directly correlate with fasting blood triglyceride level, which could, therefore, be used as a proxy measure.

A diet high in sugar and refined carbohydrate is thought to be a driving factor for raised fasting triglyceride levels and increased small-dense LDL particle number. Dietary saturated fat has no role in the causation of heart disease and might be protective against stroke. Journal of the American College of Cardiology 17 June 2020 online review. https://www.onlinejacc.org/content/early/2020/06/16/j.jacc.2020.05.077

Considering the vital role cholesterol plays, it is important we take an educated approach to cholesterol. It has a crucial role in cell membranes, including the brain and peripheral nerve myelin sheaths. Cholesterol is also a precursor for bile salts and steroid hormones; progestogens, glucocorticoids, mineralocorticoids, androgens, and oestrogens,

Total cholesterol is a poor marker for cardiovascular risk overall. There is no such thing as ‘serum cholesterol’ as it is not transported as a free form but as a passenger in lipoproteins or chylomicrons.

The gut absorption of fat into the lymphatic system and then the thoracic duct means that it enters the arterial circulation first and is absorbed by tissues before modification in the liver. Quality dietary fats are, therefore, essential in this respect. Cholesterol carried in LDL lipoproteins with apoB100 expressed on the surface is linked to heart disease. This is a stronger link than the association with unfractionated LDL.

The sole origin of apoB100 is the liver. A significant amount of liver fat is derived from the conversion of excess carbohydrates to fat. This will be packaged as apoB100 for dispersal to tissue receptors along with hepatic derived cholesterol. Increased cholesterol is produced through the genetic up-regulation of HMGCoA reductase (the same enzyme blocked by statins) under the influence of raised insulin levels stimulated by carbohydrates. So, it can be argued that carbohydrate intake is driving cardiovascular disease much more strongly than cholesterol, which is ‘downstream’ of the causative metabolic upset. There is an emerging consensus that carbohydrate intake in excess is leading to insulin resistance, which is the prime driver of cardiovascular disease. Highly processed foods amplify the effects of insulin resistance and cause oxidation of the LDL particles, shifting their particle size distribution and causing them not to be cleared in the usual way. Low carbohydrate healthy fat, real food might go a long way to correcting this imbalance.


It depends entirely upon the confidence of the patient and what the patient wants to do. There are some people with Type 1 diabetes who seem to be able to achieve control and eat a low-fat, high carbohydrate diet and do not seem to get complications with slightly higher HbA1c.

These are probably happy the way they are and will not be looking to change things.

Then others are keen to start as soon as possible because they feel that their control is poor. It is up to the partnership between the clinician and patient

to work out how quickly the transition to a lowered carbohydrate or ketogenic diet should happen.

If patients manage their basal insulin levels to the optimum and are confident

in carbohydrate counting, they can achieve ketosis within 2-3 weeks.

Improved metabolic control will happen from day one if the patient decides on a more radical approach.

Others might want to go more slowly and reduce the carbohydrate in their meal in a stepwise fashion over several weeks. Some might prefer to reduce carbohydrates in their meals just one meal at a time. It is an individual choice. Many people might not want to go to very low carb diets in the 30g a day range, but prefer to use more carbohydrates.

The aim is to achieve near normoglycemia 24 hours a day, and it is not unrealistic to expect this with a ketogenic (30g carbohydrate per day) diet. Do not encourage the patient to become obsessed with ketosis.

Inevitably by reducing dietary carbohydrate and burning fat, there will be ketogenesis, but ketosis is not the sole aim of the very low carb diet. The aim is near-normal or normoglycemia (with the least amount of insulin possible). Having to balance insulin with carbohydrates means that blood glucose levels will sometimes be suboptimal, and the patient will be briefly non-ketotic, this is unimportant as long as glycaemic control is improved.


People with Type 1 Diabetes in the early years might be more sensitive to insulin. Still, as time progresses, (over roughly ten years), some degree of insulin resistance might accompany the condition. (there are useful publications on that) It might be the case that people with Type 1 Diabetes and insulin resistance (sometimes called double diabetes), will notice that their insulin takes longer to act after injection because of insulin resistance. On a lower carbohydrate diet, the metabolic changes caused by the

reduced insulin requirements will tend to restore insulin sensitivity. The patient will begin to notice that their insulin appears more ‘potent’ in that it will act more quickly and also that they will need reduced insulin doses over and above what would be required by the dietary carbohydrate reduction. Over time they might need to reduce their basal insulin. It can be a good idea to revise basal dosing every few months if you suspect insulin resistance.

The patient will be eating a higher proportion of their diet as fat, and this, too, can cause temporary insulin resistance until it is fully metabolised. Until the patient becomes adept at managing lower carbohydrates, it can help if fat is split relatively evenly between meals.


Cutting most carbohydrates from the diet, particularly refined sugars, can be a significant change for both the brain and body. Sugar is thought to affect the brain reward system. Some people find that giving up sugar can cause withdrawal symptoms just like those of any other substance that acts via the brain reward system. Fortunately, in the majority of people, these cravings disappear in the first few weeks; in others, it can take months or longer.

Possible solutions:

Helping the patient to understand why dietary change is necessary can help to combat cravings. Many people find that removing all sugar from the diet is their best option. There are addiction questionnaires that would be suitable for exploring a person’s susceptibility to sugar addiction. One example is the Yale Food Addiction Scale.


The adjustment phase to lowered carbohydrates can lead to a relative lack of sodium. This is because insulin acts to retain sodium (and with it, fluid). In most people with Type 1 Diabetes who reduce their carbohydrates, their insulin total daily will decrease, and there will be a diuresis phase as the sodium is excreted through the renal tract. Fluid replacement with water is adequate. Adding extra salt in the first two weeks might also be useful. Relatively high insulin levels might be a driver for hypertension, and reducing insulin might lower blood pressure. It is good practice to monitor blood pressure in this phase if the patient feels particularly lightheaded and consider whether to review anti-hypertensive medication, especially if the patient is taking those anti-hypertensives, which cause salt loss. And also, if they are on diuretics, especially thiazides.


Any significant change in diet will cause a temporary change in bowel habits. Fluid replacement with water is adequate. Laxatives might be needed (which one ?). But this phase does pass. At 3-4 weeks, the patient often feels much more vital than previously.


The clinician’s assessment of the patient will include having knowledge of the blood tests and status of the liver, gall bladder, pancreas, kidneys, and gut. Active disease in any digestive organ might be approached with a more cautious transition if planning very low carbohydrates. However, many of these conditions will respond well. For example, non-alcoholic fatty liver can be reversed relatively quickly on a ketogenic diet. Individual assessment is key to safe management.

Caution is also recommended in those with ‘brittle’ diabetes and hypoglycaemic unawareness. Until you can be sure that the patient has an adequate glucagon response to low blood glucose, it is not recommended to proceed with reducing carbs to a level that might pose a risk. Some people are highlighting the possibility of glucagon resistance, especially in steatohepatitis, which again should be borne in mind in Type 1.

Gall bladder disease and gallstones can sometimes worsen if the transition to very low carbohydrate is too swift. It might be prudent to plan several months of gradual weaning down to lowered carbohydrate to avoid precipitating complications.

There have been rare reports of euglycemic ketoacidosis in people with Type 1 Diabetes. This is a diagnosis of exclusion with metabolic acidosis and is difficult to diagnose. It should be on the list of differential diagnoses of a Type 1 who has symptoms of DKA with normal glucose. It has been described in infection, insulin pump failure, and also when insulin is combined with an SGLT2 inhibitor. None of these patients were on a low carb diet, but there is the possibility that this could happen on all types of diet. Ketone testing should be encouraged for all people with Type 1 Diabetes when they are unwell.

People who use very low carbohydrate diets can experience asymptomatic hypoglycaemia. While this is not specific to Type 1 Diabetes, exogenous insulin on a very low carbohydrate diet might add a further complicating dimension to this phenomenon. If someone with Type 1 can access a CGM, it will make not only this but also all other aspects of management much more relaxed.


There are medical conditions that do not suit a low carbohydrate diet down to ketogenic levels. These are mainly inherited disorders of lipid metabolism, pyruvate carboxylase deficiency, plus porphyria.

For people with advanced diabetes complications, it is crucial to manage the transition with great care. In some situations, as we know, improvement of diabetes control can temporarily worsen complications. The body has taken years to get to the state of advanced diabetes and will have tried to adapt to high blood glucose levels by storing excess glucose as fat or urinating out glucose. Not to mention the insulin resistance brought about by chronic insulin overdosing, and vagal neuropathy. A rapid transition to ketone metabolism could possibly upset that balance. So, the risk: benefit has to be balanced carefully. Very small steps might work, or it may be that the risk is too high, and it might be best to revert to the patient’s usual regimen if it seems safer to do so.


Type 1’s who are struggling with their control often ‘get it’ very quickly. After all, they have to live with their condition 24 hours a day and often have an awareness of what advice works by their observations. Once they are introduced to lowered carb as an option, it can be a light bulb moment. The NICE guideline NG17 emphasises patient preference, and there is strong evidence that balancing carbohydrate intake with insulin injections is an effective management practice in Type 1 Diabetes.

Consider providing glucose test strips on-demand to enable patients to test frequently. It helps them gain confidence and is very motivating for both clinicians and patients. Ideally, become an advocate for your patient and make a

case for continuous glucose monitoring. It makes learning rapid and improves safety.

Provide blood ketone testing strips. All people with Type 1 Diabetes should have these to be able to manage potential DKA during those times of vulnerability (infection, etc.). Patients will feel more confident when they are aware of their blood ketone levels when on a low carbohydrate diet. Indeed, also on a conventional higher carb diet. As the body adapts to ketogenic metabolism, the blood ketone levels tend to reduce. There is no need to test ketones regularly as the strips are expensive, but they are a useful monitoring tool at certain times.

Arrange HbA1c testing three monthly. It is helpful as it is motivating; patients often look forward to HbA1c testing when following a lower carbohydrate lifestyle. For patients with CGM devices, the Coefficient of Variability and Time in Range are beginning to emerge as better markers of control.

If you know of other people with Type 1 Diabetes in your practice who are interested in lower carbohydrate management, it is a good idea to put them in touch with each other locally if they so choose.

©Ian Lake 29/10/2018

Leave a Reply

Your email address will not be published. Required fields are marked *