Low-Carb Diet Promising for Lowering Blood Sugar in Prediabetes
In a randomized trial, declines in HbA1c over 6 months were “modest” but warrant further study and longer follow-up.
“There's good evidence that in people with type 2 diabetes a low-carb diet can reduce blood sugar levels. But there's much less known and understood about people in that [lower] range and how low carb could affect their blood sugar,” said Kirsten S. Dorans, ScD (Tulane University School of Public Health and Tropical Medicine, New Orleans, LA), lead author of the study published this week in JAMA Network Open.
Among the 150 participants randomized to fewer than 60 g of carbs per day, or their usual diet, a greater reduction from baseline HbA1c levels was achieved in the low-carb group at 6 months. Dorans and colleagues say the mean decrease of 0.23%, which was significant (P < 0.001), was also “modest” and accompanied by a lower calorie intake and greater weight loss.
“The key message to take away here is that a low-carb diet, if it's maintained, may be a useful approach to preventing and treating type 2 diabetes, but more research is needed,” she added.
Interest in low-carb diets for weight loss and cardiometabolic diseases date back decades, but trials thus far have produced mixed results. Recent diabetes screening recommendations from the US Preventive Services Task Force advise clinicians to refer patients with prediabetes to preventive interventions, including lifestyle changes, diet and exercise leading to weight loss, noting that about a third of those who are prediabetic aren’t even aware of it.
Differences Beyond Just HbA1c
Patients randomized into the study (mean age 59 years; 72% female; 59% Black) had untreated HbA1c between 6.0% and 6.9% (42 to 52 mmol/mol). The low-carb group (n = 75) received behavioral counseling and key supplemental food, including nuts and olive oil, to improve odds of compliance. For the first 3 months, the low-carb group was asked to limit daily carbohydrate intake to < 40 g/day, and then from months 4 through 6, their daily intake goal was < 60 g. For the first 4 weeks, participants received weekly individual behavioral counseling sessions, which transitioned to small group sessions with other participants every other week for the next 4 weeks, and then to primarily telephone follow-up for the final 4 weeks. The usual-diet group (n = 75) received standardized written dietary advice and were offered monthly information sessions unrelated to weight loss, but no other dietary support.
At baseline, 3 months, and 6 months, all participants were assessed by 24-hour dietary recall questionnaire, had height, weight, and waist circumferences measured, and underwent three blood pressure measurements. HbA1c also was measured at this time, as were other biomarkers. Urine testing was done to measure ketones, and in a subgroup of 59 participants, 24-hour glucose levels were measured by a continuous glucose monitor over 14 days.
Total calorie intake decreased in the low-carb group by a mean of 389 calories per day compared with the usual-diet group at 3 months, and was a mean of 456 calories per day lower by 6 months (P < 0.001 for both). Participants in the low-carb group ate fewer total and net carbs, added sugars, and sugar-sweetened beverages. Conversely, the low-carb group had higher percentages of calories from protein and monounsaturated and polyunsaturated fats.
In addition to the decrease in HbA1c, the low-carb group had a greater decrease in fasting plasma glucose at 6 months (P < 0.001). While the usual-diet group only lost a mean of 0.5 kg by 6 months, the low-carb group averaged a loss of 6.4 kg (P < 0.001). Other differences favoring the low-carb group included decreases in fasting insulin, homeostasis model assessment of insulin resistance (HOMA-IR), and waist circumference at 6 months. The low-carb group also had lower mean 24-hour glucose and nighttime glucose, and spent 9.8% more time in the glycemic range of 70 mg/dL to 120 mg/dL compared with the usual-diet group. By 6 months, 53% of those in the low-carb group had an had HbA1c < 6.0% compared with only 32% of those in the usual-diet group (P = 0.01).
In subgroup analyses, the change in HbA1c at 6 months was greater in Black than in white participants and in men compared with women. However, given the small numbers, Dorans said those findings, while hypothesis-generating, are difficult to interpret in the context of the trial.
Which Diet Is Really Best?
To TCTMD, Dorans said longer follow-up is needed both to clarify the impact of a low-carb diet on HbA1c and to see if calories and weight begin to creep back up after 6 months, something she said is frequently seen in weight-loss trials.
“While it is true that carb reduction can control blood glucose in diabetics, it's not the whole story,” said Andrew Freeman, MD (National Jewish Health, Denver, CO), who commented on the findings for TCTMD. “It's a subset of the bigger truth, which is why do people who eat carbs have higher sugars when they have diabetes? It's typically because of fatty liver and fatty muscle infiltrate.”
Freeman said studies going back to the 1940s have shown that diet can improve diabetes and that it doesn’t necessarily have to include drastic reduction of carbohydrates. One reason many people may lose a lot of weight on a low-carb diet is from cutting back on what he calls “carbage,” or highly processed carb-rich foods known to spike blood glucose.
“That’s why it really is difficult in this study and in others to tease out if the low-carb diet is the reason for the improvement in diabetes or if it just helps people to lose weight and cut out the bad stuff,” he added.
Freeman said while low-carb diets may be an adjunct for some people trying to control their blood glucose, what people substitute for the carbs is important, since very-low-carb keto-type diets high in meat, cheese, and processed food can increase cholesterol and cancer risk, and are environmentally irresponsible in the long term.
“I think there's a lot of potential in this space to try to really figure out what the best diet is to help people fight their diabetes,” he added. “But, the effect seen in this study was relatively small. What would have happened with a more plant-based, no processed-white carb type of diet? I suspect the results would have been bigger.”
Dorans KS, Bazzano LA, Qi L, et al. Effects of a low-carbohydrate dietary intervention on hemoglobin A1c: a randomized clinical trial. JAMA Network Open. 2022;5(10):e2238645.
- Dorans reports receiving grants from the National Heart, Lung, and Blood Institute outside the submitted work.
- Freeman reports no relevant conflicts of interest.