Keto Diet Shown to Help Some Diabetics Stop Meds

Study participants on the high-fat, low-carb diet showed A1C improvement at every medical appointment

With Shabina Roohi Ahmed MD

Avocado sandwich

For most patients with type 2 diabetes, changing their diet to reduce risk of exacerbation of disease and comorbidities is a given. But what kind of diet works best has been the subject of study and debate for years. In particular, the composition of the ideal diet in terms of carbohydrates, fats, and proteins remains up in the air. Indeed, the guidelines of the American Diabetes Association currently state only that meal plans should be nutrient dense with few refined carbohydrates and saturated fats.

There has been a lot of popular interest in high fat, low carbohydrate diets – also known as Keto diets, as they promote ketosis, forcing the body to burn fat rather than carbohydrates for fuel. These regimens have also been of interest to researchers looking at ways to jump-start weight loss, reduce dependence on obesity and diabetes associated medications, and potentially reduce inflammation and even heart disease. However, most of the studies have been the result of academic research, with patients enjoying the benefits of larger organizations and the support systems they offer.

Now, a new study is showing the potential for the diet to assist patients in cooperation with their physicians. The retrospective study included type 2 diabetes patients who adhered to a high fat, low carbohydrate diet for at least 3 months, and the 49 diet subjects were paired with 75 patients who did not use the diet.

Study design

Researchers looked for changes between baseline A1c and that of three follow-up visits at 6-11 weeks, 12-16 weeks, and 17-21 weeks. They also looked for changes in body weight and body mass index, lipid profile, and fasting blood glucose, as well as blood pressure as measured at baseline and at follow-up visits that occurred after 12 weeks.

The diet prescribed required a net carbohydrate intake (total carbohydrates minus fiber) of less than 20 grams, or 5-10% of total calories, whichever was lower. Protein was limited to 20-25% of total calories, and fat to 65-70% of total intake. There was no caloric restriction given to the patients on the diet plan.

The physicians talked to patients about where to find sample menus and recipes, and were told to eat only when hungry. Late night eating was discouraged, while plenty of water was encouraged. Physicians asked patients to keep a food log and to bring it to follow-up visits.

Prior to starting the diet, patients stopped using sulfonylurea medications and reduced insulin doses by up to half to avoid hypoglycemia. Both groups were offered phentermine prescriptions, were told to check glucose at least once a day, ideally when fasting, and to keep a log of their blood sugar readings. Patients taking multiple insulin doses were asked to check glucose before meals and at bedtime, and to come back to the clinic after 2 weeks. Other patients returned after 2 to 4 weeks.

The control group was encouraged to eat high fiber food, low-fat dairy, fish, and foods low in saturated fat. They received standard diabetes self-management counseling.

Lastly, the physicians suggested patients in both groups be active at least 30 minutes a day and get 6-8 hours of sleep per night.


  • The low carb/high fat diet showed A1C improvement at every visit, and by the last visit had a mean A1C of 6.67% compared to 7.8% in controls.
  • The diet group lost more weight – a mean loss of more than 12kg or 11.9% of body weight, compared to a weight gain of .5kg in controls.
  • There was no significant difference in those who used phentermine and those who did not.
  • Fasting blood glucose improved significantly compared to the control group, while lipids and blood pressure remained stable in both groups.
  • The diet group discontinued or reduced glucose lowering medication use more frequently than controls, including a reduction or discontinuation in insulin among all diet patients compared to 23% of controls. More than a third of controls using insulin saw their dosage increase.

Expert discussion

"This is the first time a study on the keto diet with diabetic patients has been done in a real world setting," says lead author Shabina Roohi Ahmed, MD, an assistant professor of medicine at Johns Hopkins University in the division of endocrinology, diabetes and metabolism, and medical director at Johns Hopkins Community Physicians Endocrinology in Bethesda, Maryland.

She has been following patients in her clinic who have used this diet since 2013 or 2014. “We don’t have health coaches or nutritionists or a broad telehealth system,” she says. “This was in a doctor’s office and shows it can be implemented in a real-life community setting.”

It’s not an easy diet to adhere to, says Ahmed. Patients have to be very motivated to want to reduce their use of medications, including insulin. The lack of bread, pasta, and potatoes can be extremely difficult for some. “But we tell patients that they are human, and there will be special occasions. So, you plan for them. If it’s your birthday, eat cake, but be stricter before and after.”

Once patients have lost significant weight, they can often move to a less restrictive 50 grams of net carbohydrates. In addition, she says that she has anecdotal evidence that those who start this diet earlier in their diabetes journey seem to have more leeway to eat more carbohydrates. “They may have more pancreatic cells that are functioning.”

A weekly day of eating anything is probably too often, Ahmed warns. “If it happens that often, you won’t see the full benefit. But once a month? Sure.”

What happens after a patient has recovered some degree of glucose tolerance differs by patient, she says. “I have one patient who says that this diet saved her life, but she will never put anything sweet in her mouth again because if she does, her sugar addiction will start all over again.

Even if they can increase the amount of carbs they eat in time, patients should never return to the typical American diet of 100 grams of carbohydrates or more per day. “That’s what got them into trouble to begin with,” says Ahmed.

Times of stress, like many are experiencing during the current pandemic and accompanying economic downturn, are potential danger points for anyone with a chronic illness to manage, but perhaps even more so for those who are on restrictive diets, Ahmed says. Think of the empty grocery store shelves that accompanied the start of the lockdowns in the spring. For some people, finding appropriate foods may have been difficult.

In general, however, Ahmed says that it is becoming easier and easier to find products created just to fit into a high fat and low carbohydrate lifestyle. They may be more expensive, but they can make adherence easier for some patients.

Ahmed uses and recommends taking advantage of new resources to direct both physicians and patients looking for information on the keto diet, how it works, and how to make it a livable lifestyle choice. Among them are, founded and run by a medical doctor, the books The Art and Science of Low Carbohydrate Performance and The Art and Science of Low Carbohydrate Living by Jeff Volek PhD, RD, a founder of the diabetes company Virta Health, and a TED talk by diabetes researcher and physician Sarah Hallberg, DO.

“As a physician, when you read more about this diet, you nod your head and realize how much sense it makes,” says Ahmed. She is hopeful that future research will be able to pin down whether or not there is a benefit to future cardiac health of patients who choose this diet to control diabetes. Concerns remain about whether saturated fats like that in butter and cheese can cause harm, even though surrogate markers such as lipid profiles and inflammation markers decline among those eating this way. “We still need to see if there is more or less stroke, heart failure, or heart attack,” Ahmed says.

Physicians who want to recommend this diet to patients should be sure to explain that the first two to three weeks can be very difficult. “You are going from using carbs as your main energy source to fat. There’s a thing called the ‘Keto flu’ that people complain of. It’s a feeling of fatigue and of not being able to concentrate well.”

The brain will use glucose until there is none left, Ahmed continues. Only then will it move to using fat. “That can be hard. I tell patients to stay well hydrated and get enough salt, as they lose a lot of water weight initially, and with it electrolytes,” she says. Ahmed also recommends magnesium supplements in the beginning.

She has no worries about patients with liver issues using this diet, and, as this isn’t a particularly high protein diet, it’s not hard on the kidneys. “We saw no issues with creatinine.”

She does recommend it to some patients with type 1 diabetes, but is careful because ketoacidosis is a potential problem. “We monitor insulin levels and follow them very closely,” Ahmed says. The only patients she doesn’t suggest the diet to are those on SGLT2 inhibitors. Those medications suppress glucagon release and may precipitate ketoacidosis.

“I tell patients that for this to work, they have to commit to this,” Ahmed notes. “There are ways to feed a sweet tooth once in a while, like with Keto-friendly cookies or cheesecake. We look for ways to remove barriers to their success, so they don’t become an Achilles heel. But it has to be a very motivated patient to be successful.”

Continue Reading:
Ketogenic Diet: Is It Good for Diabetes and Weight Loss?
close X