Do Keto Diets Work For Diabetes?

Featured Video Play Icon

The reason this topic deserves a blog all in itself is because with diabetes the stakes are arguably higher. Hypergylcaemia is a major risk factor for many health problems. This is true regardless of the type of diabetes you have. We also know in type 2 diabetes 90% of people suffering with the disease do so because they are overweight and/or inactive. This leads to an accumulation of fat around the pancreas, liver and cells of the body creating an insulin resistance. As a result, the body is unable to control blood glucose levels adequately resulting in hyperglycaemia. 

Obesity is a major risk factor for multiple diseases and health conditions. Throw diabetes and hyperglycaemia into the equation and suddenly that initial risk is multiplied greatly. Therefore, it is really important to help people with diabetes control their blood glucose levels (both type 1 and type 2). 

What is a Keto Diet?

The lines a murky when it comes to defining low carbohydrate diets. Usually a normal diet will contain somewhere between 120-280g per day. A Keto diet is defined as very low carbohydrate diets, usually below 50g per day. So these diets are very low carbohydrate. 

If you are unsure what carbohydrates are, I have blogged about this previously here. Carbohydrates are one of the main energy providing nutrients alongside fat. By minimising the carbohydrates in the diet, the body must turn to fat as it’s main energy source. 

Usually fat metabolism follows a particular series of events which ultimately results in energy production. This process is disrupted during a low carbohydrate diet. In order to compensate for the low carbohydrate intake some of the fat is made into glucose by the liver. This means the liver needs to find an alternative way of meeting it’s usual fat/energy quota. It does this by producing ketone body’s. When your body is producing ketone body’s you are said to be in ‘ketosis’. 

It is this process the keto diet manipulates. Ketosis occurs naturally usually seen during periods of starvation, low carbohydrate diets, in type 1 diabetes when insufficient insulin is present and in some cases during excessive alcohol intake or exercise.

Why should a Keto Diet work in diabetes? 

With type 1 or type 2 diabetes, it is the carbohydrates in your diet that cause glucose levels to increase.

Fat and protein only have a minimal effect on blood glucose levels and thus do not cause large spikes. 

Ketone body’s replace glucose as an energy source to help fuel the central nervous system and take up the slack in times of low glucose. These also do not increase glucose levels. 

Therefore, by removing the carbohydrates in your diet you minimise the potential for glucose rises.

There is some evidence to suggest keto diets help people lose weight because the body relies upon fat as it’s primary energy source (see my previous blog on this here). This may be relevant in type 2 diabetes where weight loss is one of the cornerstones of  treatment. However, this evidence isn’t without its flaws.

Some research I looked at reported improved HbA1c’s in patients with diabetes following a keto diet even without weight loss. Another key benefit was people following a keto diet were able to reduce the number of medications they were taking for diabetes. This could be a useful strategy for patients with multiple daily therapies assuming they are able to stick to the diet. 

However, such articles were very short term and did not have many participants included in them. Therefore, the validity is questionable and shouldn’t be taken as gospel. 

So why isn’t this first line treatment in diabetes? 

Pros of a keto diet in diabetes

With every diet there are always positives and negatives and the keto diet is no exception.

Looking specifically at diabetes, most of the benefits come from the medicine burden associated with diabetes and ill health.

Type 1 diabetes

People with type 1 diabetes will need to take insulin regardless but the amount they need to take may be much less. This is particularly true of their rapid acting insulin. In fact, some patients following a keto diet with type 1 diabetes may find they barely need to take rapid insulin for food alone. However, the management of rapid insulin is complex and it may still be required for the carbohydrate converted from the higher levels of protein and fat consumed on the keto diet.

They may also still need to take rapid insulin to correct high levels associated with stress, illness or any other cause of hyperglycaemia.

Type 2 diabetes

In type 2 diabetes, the benefits are a bit more pronounced for some medications. Some patients with type 2 diabetes take medications specifically designed to prevent glucose spikes after eating. Namely sulphonylureas (such as Gliclazide – or any diabetes medication ending in ‘ide’ or ‘mide’) or insulins with a rapid acting element to them (mixed insulins and rapid insulin themselves). For a look at what these may be we have a medication section on this site which you can find here.

By eating less foods that cause definitive elevations in their glucose levels you may find you can reduce your dose or even stop taking such medications. If you’re taking insulin, it might be that you can be converted to a different insulin not concerned with dealing with the glucose spikes concerned with food.

Overall, keto diets have been associated with lower HbA1c results. This means average glucose control improves for patients who follow these diets for the short term (1-12months). This is certainly promising and in my opinion such data shouldn’t be ignored.

Problems with keto diets

Although Keto diets reduce the scope for blood glucose level increases there is a down side.

Long term sustainability 

The first concerns are about long term sustainability. When we look at chronic diseases like diabetes we are analysing the long term effects. Controlling blood glucose levels for only short periods are unlikely to yield the benefits of long term control. This is why we encourage a more sustainable approach to dieting that can be maintained long term.

Most research shows low carbohydrate diets are initially better at reducing body weight. However, there is no benefit compared to other diets over 1 year because people consistently cannot sustain such diets. 

Study Design

Another major problem is most of the articles reporting benefits are very short term and study very few people. When manipulating diet in research studies it is very difficult to use a study design which analysis cause and effect long term. This is because you can’t get people to adhere to such diets for too long. Therefore, it’s tricky to isolate one dietary factor independent of other lifestyle factors. Thus, the real world application is also questionable. 

This is why most studies looking at diet are observational studies where we look at the dietary traits of different populations and form conclusions that way. A good example of observational studies is when looking at smoking. If you studied someone for 5 years from starting smoking to quitting 5 years later, you are unlikely to find any significant lung disease. However, if you look at a population of smokers over the long term you start to notice the significantly increased level of death from all causes. 

A recent major observational review study raised concerns about low carbohydrate diets. In this review, people who replaced carbohydrates with animal fat and protein were much more likely to die or have a cardiovascular event. This risk was removed if replacing animal fat and protein with plant based sources such as beans. pulses, lentils, oils etc. However, in doing so, they would have increased their carbohydrate intake and thus be following a comparatively higher carbohydrate diet. In fact, this article showed the lowest risk of death or cardiovascular problems was when people consumed around 50% of their total energy intake from carbohydrates. 

This article was important because it superseded previous articles and accounted for previous study design flaws, had 15,000 people and was relatively longer term. 

Remission trials

The DIRECT study was published in late 2017 and showed people with newly diagnosed type 2 diabetes can push their diabetes into remission. 

The study prescribed an 800kcal per day liquid diet for participants over 3-5 months. They found patients who lost a significant amount of weight were able to significantly reduce their medications and even push their type 2 diabetes into remission. This was followed by a maintenance phase much resembling current UK dietary guidelines. After 1 year participants were able to maintain such loses.

Returning to the keto diet topic, the reason this is relevant is because this diet consisted of 59% carbohydrates in the trial phase and 50% carbohydrate in the maintenance phase. Therefore, the diet was not ketotic. In fact, it was far from it. So how did these patients put their diabetes into remission despite eating and drinking carbohydrates?  

In 90% of type 2 diabetes it is fat accumulation causing the problem. Therefore, you can remove most of carbohydrate from the diet (like in a keto diet) and demonstrate good glucose levels but it hasn’t really addressed the underlying issue. Without significant weight loss in most people, a keto diet merely masks the problem. Therefore, the DIRECT study addressed the underlying cause of the disease. This is why despite both the diet and maintenance phase containing 50% plus carbohydrates, the participants were able to sustain normal glucose levels. 

However, the DIRECT study was not perfect. It only studied just over 300 participants and like other such research, is reasonably short term. 800kcal per day may also be considered an extreme measure and no more sustainable than a keto diet. 

I think why this regime has made such waves in the diabetes community is because it supports what we have known for some time in type 2 diabetes. Weight loss is the key for most people with the disease. 

We have a separate blog on the additional downsides of a keto diet specially looking at long term health that probably shouldn’t be ignored. You can find the long version here or the condensed version here.

SGLT-2 inhibitors

This is an important topic to cover and if you take any medication ending in ‘gliflozin’ keep reading. SGLT-2 inhibitors lower glucose levels by increasing the amount of glucose urinated out. This class has many benefits and if you’re interested you can learn more about them here.

However, in the context of a keto diet, they can be dangerous. Due to their mechanism of making you urinate more, you can become dehydrated. Ketones plus dehydration rarely go together very well and there has been a number of cases of diabetic keto acidosis (DKA) reported in patients taking this medication.

DKA if left untreated, will be fatal within 24-48 hours. Luckily, fatalities are very rare these days as medical teams are good at spotting it early. However, the experience is not a good one and you will be very acutely unwell. Furthermore, an episode of DKA can cause longer term damage to your body and therefore is best avoided.

Usually DKA is only discussed in type 1 diabetes but we have seen a number of patients with type 2 diabetes going into DKA when following keto diets and taking an SGLT-2 inhibitor.

Can you stop the SGLT-2 inhibitor?

Ultimately, this will be your decision. I’d encourage you to follow the link above looking at the benefits of this class of medication. Rarely are SGLT-2 inhibitors started only for glucose lowering reasons. They are also proven to significantly improve cardiovascular and renal health. Therefore, stopping this medication for dietary purposes will lose these benefits and can be dangerous to your longer term health.

Speak with your medical team if you have further questions.

Type 1 diabetes

There is very few articles published looking at this topic. The 2 articles I did find were not overly convincing but did support what I expected to find. Low carbohydrate diets led to improved HbA1c’s in the short term but did increase episodes of hypoglycaemia and increased circulating blood lipids. Therefore, it seemed the benefits were outweighed by the cons.

Hypoglycaemia is sometimes not taken as seriously by many patients as high glucose levels. However, hypos can be just as or even more dangerous than high glucose levels. Low glucose levels can be fatal or significantly impact your body and cognitive functioning both short and longer term. Therefore, increased rates of hypos are not to be taken lightly, even if your glucose levels do not run as high as before by commencing a keto diet.

There are also concerns regarding people with type 1 diabetes sailing too close to the wind if they are naturally ketotic. The margin of error before tipping into a diabetic ketoaoidodsis state may be smaller if ketones are already present in the system. 


So what does this all mean?

From my research it seems there are many ways to approach improving your diabetes control. What appears most important is making sure you are able to stick with the changes you make long term. 

Try not to get locked into one dimensional thinking where glucose levels is all you focus on. Yes, these are important but if by improving your glucose levels you increase your risk of other complications you haven’t made huge progress. 

Naturally there will always be those individuals who thrive on such diets such a keto diets. These are usually the folks I meet in my daily practice who argue we should be rolling this out nationwide. 

However, looking at the general research, I’d argue there’s no need for keto diets or even low calorie diets and in real life they aren’t overly sustainable long term. They can be effective in the short term and possibly even longer term for some people. However, the most important thing for most people with type 2 diabetes is weight loss and embracing a new way of life.

Those with type 1 diabetes the situation is slightly more complex. Their condition is not driven by weight gain but rather an insufficient insulin production. There is yet any evidence to suggest low carbohydrate diets improve type 1 outcomes. However, like our type 2 counterparts, there are likely some folk who will do well on such a diet. However, please remember the risk factors I outlined above. 

Keto diets and low calorie diets are quite extreme measures. That said, with 10% of the total NHS budget spent on diabetes, perhaps extreme measures is exactly what we need. Particularly in the short term to help kick start change.

If you decide to embark on a keto diet I would strongly advise making sure you have a long term maintenance plan in mind. One useful strategy which appears consistently in my research is using short term cycles of such diets followed by longer term maintenance phases. This can then be repeated. Remember, if you are changing your lifestyle, please speak with your diabetes team so they can recommend adjustments to your medications. 

Need more help

If you need more support we offer a range of 1:1 consultancy services and online programmes focused on helping you improve your glucose control and health for the long term. These include our type 2 diabetes recovery programme, type 1 glucose stability programme and winning weight loss programme. Go to the pages using the links above or in the headers or get in touch if you have questions.


A review of the carbohydrate–insulin model of obesity

Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis

Dietary carbohydrate restriction as the first approach in diabetes

management: Critical review and evidence base

Effect of low-calorie versus low-carbohydrate ketogenic diet in type 2 diabetes.

Hyperketonemia and ketosis increase the risk of complications in type 1 diabetes

Primary care-led weight management for remission of type 2

diabetes (DiRECT): an open-label, cluster-randomised trial

Overweight and diabetes prevention: is a low-carbohydrate–high-fat

diet recommendable?–high-fat+diet+recommendable%3F

Revealing the molecular relationship between type 2 diabetes and the metabolic changes induced by a very-low-carbohydrate low-fat ketogenic diet

Short-term safety, tolerability and efficacy of a very low-calorie-ketogenic diet interventional

weight loss program versus hypocaloric diet in patients with type 2 diabetes mellitus

Short-term safety, tolerability and efficacy of a very low-calorie-ketogenic diet interventional weight loss program versus hypocaloric diet in patients with type 2 diabetes mellitus.

The effect of a low-carbohydrate, ketogenic diet versus a low-glycemic index diet on glycemic control in type 2 diabetes mellitus

The glycaemic benefits of a very?low?carbohydrate ketogenic diet in adults with Type 1 diabetes mellitus may be opposed by increased hypoglycaemia risk and dyslipidaemia

5 thoughts on “Do Keto Diets Work For Diabetes?”

  1. A Keto diet absolutely does help people with diabetes because I have type 1 diabetes for 25 years now. I started my keto diet about 1 year ago and feel great and it’s boosted my energy tremendously. Oh, and lowered my A1C to 6.4% from 7%.

  2. Nevermind the Inuit, the Masai, who ate this way for centuries before the white came long with his flour and sugar. Also ignore the Arctic explorers Schwatka and Steffansen who wrote about their experience and the results of Steffansen’s proving the efficacy of this diet in New York after his return. Also, what was the standard of care for diabetics prior to the manufacturing of insulin – ketogenic diets – diabetes used ot be called carbohydrate intolerance. I suggest doing further research.

    • diabetesdietguy July 19, 2019 at 2:47 pm

      Hey Kay

      Thanks for comment. Review here says it lowers glucose and can help reduce medications. So I’m not saying there aren’t pros. Hence the pro section!

      However there are also cons to the diet. No diet is complete so to not discuss cons implies a bias.

      You can choose any select group to prove your own point. Hence why I look at the reviews with 100,000s people to help inform our discussions.

      I’m not putting across my own opinion but what has been researched and studied. I also have the benefit of seeing ALOT of diabetes patients in clinic following every sort of diet. So again to ignore both sides of the coin is just bad science.

      Thanks for reading.


  3. Hiya! I know this is kinda off topic nevertheless I’d figured
    I’d ask. Would you be interested in exchanging links or maybe guest writing a blog post or
    vice-versa? My blog covers a lot of the same topics as yours
    and I feel we could greatly benefit from each other. If you happen to be interested feel
    free to send me an email. I look forward to hearing from you!

    Excellent blog by the way!

Comments are closed.

%d bloggers like this: