Research & Education

New study reveals low-carb diet combined with medical visits may improve type 2 diabetes

A study published in Jama Internal Medicine earlier this month adds to the large body of research indicating that low-carb diets are effective for improving glycemic control in people with type 2 diabetes. The study was a randomized clinical trial comparing group medical visits plus intensive weight management compared to group medical visits alone for glycemia in patients with type 2 diabetes. It was a noninferiority trial, meaning that it was intended to determine whether group medical visits combined with intensive weight management (WM/GMV) were noninferior to group medical visits (GMV) alone, and the main outcome was changed in HbA1c at 48 weeks. The study also sought to determine whether WM/GMVs were superior to GMVs alone for hypoglycemic events, diabetes medications, and weight loss.

Study subjects were overweight individuals with type 2 diabetes being treated at two Veteran’s Administration medical centers in North Carolina (n=263; mean age 60.7). At baseline, mean HbA1c was 9.1, mean body weight was 108.5 kg (239 lbs), and mean BMI was 35.3. At baseline, 82% of subjects were being treated with metformin and 45% were treated with sulfonlyureas. Regarding insulin, 38.4% of subjects were not on insulin; 24.7% were taking basal insulin alone and 36.9% were taking mealtime insulin with or without basal insulin. Other medications used at baseline included TZDs, SGLT-2 inhibitors, DPP4 inhibitors, and GLP-1 receptor agonists. 

Regarding the dietary intervention for weight management, a very low-carb diet was the initial focus, with carbohydrate intake limited to 20-30 grams per day, with no calorie restriction. According to the paper, “Participants were taught how to add to daily carbohydrate intake gradually as they approached their weight goal or if adherence was threatened.”

Subjects in the GMV group met every 4 weeks for 16 weeks and after that every 8 weeks for diabetes counseling and medication management. The WM/GMV subjects met every 2 weeks for 16 weeks and every 8 weeks after that for the same purposes as above plus weight management support. Dietary intake was assessed every 16 weeks via 3-day food records and adherence to the intervention was assessed by research personnel using a checklist of key topics covered at each group meeting. With carbohydrate limited to just 20-30 g/day, at least initially, it’s likely subjects who were compliant were in ketosis, but the paper makes no mention of subjects measuring ketone levels nor blood glucose levels, which may have been a more accurate way of assessing adherence.

The results were noteworthy. With regards to HbA1c, WM/GMV was found to be noninferior to GMV alone, but it was not superior. At 16 weeks and 32 weeks, mean HbA1c in the WM/GMV subjects was lower than in the GMV subjects (7.4 compared to 8.1), but this difference nearly evaporated by the end of the 48-week study. Weight loss was substantially greater in the WM/GMV group compared to GMV alone at all assessed time points, with the greatest difference seen at 32 weeks (102.3 kg versus 108.5 kg). By 48 weeks, this difference began to decrease as participants in the WM/GMV arm regained some of the weight they had lost, but the difference was still significant.

Noteworthy as well is that 12 of 109 WM/GMV subjects (11%) could be considered to have put T2D into “remission,” based on achieving HbA1c <6.5% while taking no diabetes medication or solely metformin. No subjects in the GMV group met these criteria. Additionally, subjects in the WM/GMV group experienced an estimated mean of 3.2 hypoglycemic events during the 48-week study, compared to almost double—6.6—in the GMV group. (This equates to one event every 16 weeks compared to one event every 8 weeks, respectively. However, even the higher rate of one event every 8 weeks may be a big improvement compared to an individual self-managing T2D, where the incidence of hypoglycemia is likely much higher.)

Another important result was discontinuation of insulin therapy for about 25% of WM/GMV subjects who were using it at baseline, compared to no insulin discontinuation in the GMV group. In fact, of GMV subjects taking insulin, over 50% had their doses increased. Similar changes were seen in sulfonylureas, with more subjects in the WM/GMV group being able to discontinue these drugs than in the GMV group, and more in the GMV group having their doses increased. 

To summarize, the dietary intervention initially led to better glycemic control (mean HbA1c of 7.4 at 16 weeks compared to 8.1 without the dietary intervention), but this difference evened out by the end of the 48-week study. Even so, the weight management counseling provided other benefits, including greater weight loss, fewer hypoglycemic events, and more reductions or total elimination of various diabetes medications. Achieving similar glycemic outcomes with less medication and fewer hypos means the diabetes is managed more effectively.

Lead study author William Yancy, MD, said, “Intensive weight management using a low-carbohydrate diet can be as effective for glycemic improvement as medication intensification.” This study shows that a low-carb diet can improve glycemic control as well as—if not better than—increasing doses of medication. This is important because most pharmaceutical drugs for T2D come with undesirable side-effects, such as weight gain and hypoglycemia, but also those as severe as euglycemic ketoacidosis requiring hospitalization, or genital gangrene (from SGLT-2 inhibitors).  

Considering subjects were counseled in how to increase their carbohydrate intake to improve general adherence, it’s probable that a substantial number of WM/GMV subjects were not following a truly low-carb diet by the end of the study, which could account for the weight regain and rebound of HbA1c. Difficulty with adherence to dietary changes “in the real world” is increasingly recognized as a problem in nutrition research. Whether any particular dietary protocol—keto, Paleo, vegan, Mediterranean—is effective for a particular goal is irrelevant if no one can actually stick to it outside a hospital metabolic ward. It appears that even meeting as a group every 8 weeks may not be enough to help people stay on track. For those who can adhere for the long-term, though, and who continue to reap the benefits of improved blood glucose control and insulin sensitivity, regarding the power of ketogenic diets to treat T2D, one physician said, “For many patients, it is absolutely life-changing.”