Conference Coverage

Carbohydrate restriction a viable choice for reversal of type 2 diabetes, expert says



Carbohydrate restriction is a viable patient choice for type 2 diabetes reversal, according to Sarah Hallberg, DO.

Medical director, supervised weight loss program, Indiana University Health Arnett Doug Brunk/MDedge Medical News

Dr. Sarah Hallberg

“Nutritional ketosis supports diabetes reversal by reducing insulin resistance while providing an alternative fuel to glucose with favorable signaling properties,” she said at the World Congress on Insulin Resistance, Diabetes, and Cardiovascular Disease.

Low-carbohydrate nutritional patterns including ketosis have extensive clinical trial evidence for improvement of type 2 diabetes, including preliminary results from a 5-year study of 465 patients enrolled in the Indiana Type 2 Diabetes Reversal Trial that Dr. Hallberg is overseeing in her role as medical director and founder of the medically supervised weight-loss program at Indiana University Health Arnett, Lafayette.

“The ketogenic diet is not a fad diet, it’s what we used to treat people with before the advent of insulin,” said Dr. Hallberg, who has been recommending and counseling patients with type 2 diabetes to follow a ketogenic diet for nearly 10 years. “Of course, insulin has been wonderful. It’s saved so many people with type 1 diabetes. But we also misused it in type 2 diabetes. Instead of counseling people the way we used to about the food that they’re taking in to control their blood sugar, we’ve just been putting [them] on medication, including insulin.”

The American Diabetes Association and other organizations have updated their guidelines to include low-carbohydrate eating patterns for type 2 diabetes treatment, she continued. Veterans Affairs/Department of Defense recommend carbohydrate levels as low as 14%.

Dr. Hallberg, who is also medical director for Virta Health, defined a very-low-carbohydrate or ketogenic diet as less than 50 g of carbohydrates per day, or fewer than 10% of calories consumed. A low-carbohydrate diet is 51-130 g of carbohydrates per day, or 25% or fewer calories consumed, whereas anything above 25% calories consumed is a not a low-carbohydrate diet. A well-formulated ketogenic diet, she continued, consists of 5%-10% carbohydrates (or less than 50 g), 15%-20% protein, and 70%-80% fat. The carbohydrates include 5-10 g per day of protein-based food, 10-15 g of vegetables, 5-10 g of nuts/seeds, 5-10 g of fruits, and 5-10 g of miscellaneous nutrients. “When we’re talking about a total carbohydrate intake per day of under 50 g, you can get a lot of vegetables and nuts in,” she said. “I like to tell my patients they’re not eating GPS: no grains, no potatoes, and no sugar.”

Recently, Dr. Hallberg and colleagues published a review in which they sought to evaluate the appropriateness of sources cited in the ADA’s guidelines on eating patterns for the management of type 2 diabetes, identify additional relevant sources, and evaluate the evidence (Diabetes Obes Metab. 2019;21[8]:1769-79). “We looked at how much evidence there is for the low-carb diet, the Mediterranean diet, the DASH [Dietary Approaches to Stop Hypertension] diet, and a plant-based diet,” she said. “We found a wide variation in the evidence for each eating pattern, but the low-carb eating pattern for diabetes has so much more evidence than any of the other eating patterns.”

In an earlier study, researchers followed 10 inpatients with diabetes in a metabolic ward for 3 weeks. Their mean age was 51 years, and their mean body mass index was 40.3 kg/m2. The patients were fed a standard diet for 7 days, then a low-carbohydrate diet (21 g per day) for 14 days (Ann Intern Med 2005; 142[6]:403-11). After 2 weeks of the low-carbohydrate diet, their mean fasting blood glucose dropped from 7.5 to 6.3 mmol/L, and their mean hemoglobin A1c (HbA1c) fell from 7.3% to 6.8%. “The levels came down very fast,” said Dr. Hallberg, who was not involved with the study. “This is an important part of the intervention, because when you get a patient who’s tried everything, who’s injecting hundreds of units of insulin every day, you can make a huge difference in the first couple of weeks. It is not unusual for us to pull patients off of 200-plus units of insulin. This is as motivating as all get out. It also affects their pocketbook right away. This is one of the reasons our patients are able to sustain a ketogenic diet along with support: early motivation and satisfaction.”

In a longer-term trial, researchers evaluated the impact of a ketogenic diet in 64 obese patients with diabetes over the course of 56 weeks (Moll Cell Biochem. 2007;302[1-2]:249-56). The body weight, body mass index, and levels of blood glucose, total cholesterol, LDL cholesterol, triglycerides, and urea showed a significant decrease from week 1 to week 56 (P less than .0001), while the level of HDL cholesterol increased significantly (P less than .0001).

A separate trial conducted in Israel evaluated the effects of a low-carbohydrate diet, compared with a Mediterranean or low-fat diet in 322 moderately obese patients over the course of 2 years (N Engl J Med. 2008;359:229-41). The rate of adherence to a study diet was 85% at 2 years. The mean weight change was greatest for those on the low-carbohydrate diet, followed by the Mediterranean and low-fat diets. Fasting glucose was best for those on the Mediterranean diet at the end of 2 years, whereas change in HbA1c was best among those on the low-carbohydrate diet.

Another study randomized patients to a low-carbohydrate ketogenic diet (less than 20 g per day with no calorie restriction) or to a low–glycemic index diet (55% carbohydrate restriction of 500 kcal from baseline) over the course of 24 weeks (Nutr Metab [Lond]. 2008 Dec 19. doi:10.1186/1743-7075-5-36). Between baseline and week 24, the mean HbA1c fell from 8.8% to 7.3% in the very-low-carbohydrate diet group, and from 8.3% to 7.8% in the low–glycemic diet group, for a between-group comparison P value of .03. In addition, 95% of patients in the low-carbohydrate diet group were able to reduce or eliminate the number of medications they were taking, compared with 62% of patients in the low–glycemic diet group (P less than .01).

Dr. Hallberg and colleagues are currently in year 4 of the 5-year Indiana Type 2 Diabetes Reversal Study, a prospective, nonrandomized, controlled trial of carbohydrate restriction in 465 patients, making it the largest and longest study of its kind. Of the 465 patients, 387 are in the continuous-care arm, which consists of a diet from Virta Health based on principles of nutritional ketosis, and 87 patients in a usual care arm who are followed for 2 years. The trial includes patients who have been prescribed insulin and who have been diagnosed with diabetes for an average of 8 years.

At the meeting, Dr. Hallberg presented preliminary results based on 2 years of data collection. The retention rate was 83% at 1 year and 74% at 2 years. In the treatment arm, the researchers observed that the level of beta hydroxybutyrate, or evidence of ketogenesis, was the same at 2 years as it had been at 1 year. “So, people were still following the diet, as well as being engaged,” she said.

At the end of 2 years, the mean HbA1c reduction was 0.9, the mean reduction for the Homeostatic Model Assessment of Insulin Resistance was 32%, and 55% of completers experienced reversal of their diabetes. Overall, 91% of insulin users reduced or eliminated their use of insulin, and the average weight loss was 10% of baseline weight. “Medication reduction was across the board,” she added. “This is huge from a cost-savings and a patient-satisfaction standpoint. We were improving A1c levels in patients who have had diabetes for an average of over 8 years while we were getting [them] off medication, including insulin. Low carb is now the standard of care.”

Even patients who did not experience a reversal of their diabetes were conferred a benefit. They had an average reduction of 1.2 in HbA1c level, to 7%; their average weight loss was 9.8%; 45% of patients eliminated their diabetes prescriptions; 81% reduced or eliminated their use of insulin; there was an average reduction of 27% in triglyceride levels; and they had a 17% reduction in their 10-year risk score for atherosclerotic cardiovascular disease.

In the overall cohort, the 10-year Atherosclerotic Cardiovascular Disease risk score improved by 12%; almost all markers for cardiovascular disease improved at 1 year. “We were giving these patients appropriate support, which I think is key,” Dr. Hallberg said. “No matter what you do, you have to have a high-touch intervention, and supply that through technology. We do better than medication adherence. Putting patients on a carbohydrate-restricted diet with the appropriate support works for sustainability.”

Dr. Hallberg disclosed that she is an employee of Virta Health and that she is an adviser for Simply Good Foods.

Recommended Reading

Top picks for online diabetes information for doctors and patients
MDedge Endocrinology
SGLT2 inhibitors for diabetes safe, effective in older adults
MDedge Endocrinology
Farxiga granted Priority Review for treatment of adults with HFrEF
MDedge Endocrinology
Fast-acting, mealtime insulin aspart is approved for kids
MDedge Endocrinology
Score predicts bariatric surgery’s benefits for obesity, type 2 diabetes
MDedge Endocrinology