This article is more than 5 years old. Information may no longer be current.
Low-carb diets show promise in improving glycemic measures
Carbohydrate restriction does not affect HbA1c for adults with type 2 diabetes, at least in the long term, but low-carbohydrate diets exhibit the potential to produce improvements, according to findings published in Diabetic Medicine.
“There is significant current interest in the role of dietary carbohydrates for weight control and in the context of type 2 diabetes for the control of glycemia; however, the ideal amount of dietary carbohydrate remains unclear,” Paul D. McArdle, BSc, MA, RD, MBDA, the lead clinical dietitian and deputy head of nutrition for the Birmingham Community Healthcare NHS Foundation Trust in the U.K., and colleagues wrote. “Current U.S. and European diabetes organizations do not make strong recommendations about the quantity of carbohydrate, but rather state that monitoring of total carbohydrate is a key strategy in achieving glycemic control, with the focus on dietary changes instead targeted at weight loss in those who are overweight.”
McArdle and colleagues conducted a systemic review and meta-analysis of 25 randomized controlled trials of carbohydrate restriction in adults with diagnosed type 2 diabetes (n = 2,132 combined participants) published between 1976 and 2018. All trials had an intervention duration of at least 8 weeks with outcomes reported at 12 weeks or more and included HbA1c as an outcome.
Restricting carbohydrates did not influence HbA1c in the meta-analysis (weighted mean difference of –0.09%; 95% CI –0.27 to 0.08). However, in a subgroup of five trials that utilized a “low-carbohydrate” diet (50-130 g per day), the researchers noted a clinically significant result (weighted mean difference of –0.49%; 95% CI, –0.75 to –0.23). Each of the subgroup trials was conducted over 6 months or less and outcomes were measured at 6 months. The subgroup analysis also showed that low-carbohydrate eating plans had a significant effect on body weight (weighted mean difference of –0.43 kg; 95% CI, –0.74 to 0.12), whereas no overall pooled effect was observed for all 25 trials (weighted mean difference of –0.13 kg; 95% CI, –0.33 to 0.08).
“Controversy in the area of dietary carbohydrate is likely to persist,” the researchers wrote. “Data from studies of carbohydrate-restricted diets raise important questions about the long-term sustainability of such diets. ... In order to add value, any future trials should be long term (>12 months in duration), should adopt the prevailing definitions of low carbohydrate and should intend to keep both the caloric content of the diets in study arms and any changes in body weight equal.” – by Phil Neuffer
Disclosure: McArdle reports he has received honoraria from Healthspan, Eli Lilly and Novo Nordisk.
Perspective
Back to TopIn my most recent position as clinical services manager for a titration service, operating under a study protocol, I worked exclusively with patients with type 2 diabetes who needed their insulin titrated to achieve glycemic control. The patients were referred to the service because they were not in glycemic control (HbA1c > 7.5%). Most patients were under-dosed. All patients required increased insulin doses to achieve glycemic control. Patients were not prescribed any dietary recommendations whatsoever. Whatever they were doing food-wise was not addressed. The insulin was titrated according to their normal eating habits.
Basically, what I would say is that long-term studies are lacking and should be done. However, from a registered dietitian’s perspective, my experience has shown me that when patients attempt a low carbohydrate diet — and the definition of “low carbohydrate” differs from patient to patient — they come to the realization that they are eating way more carbohydrate than they thought and, of course, more calories. If you reduce carbohydrates, you reduce calories, so weight loss is no surprise.
Personally, I did not see a huge difference in HbA1c results when patients reduced carbohydrates. From my perspective, depending on how under-dosed the patient’s insulin dose was, the short-term studies may not cover enough time to make a difference in HbA1c result. It took most of my patients more than 3 months to get the insulin titrated to a level of acceptable HbA1c results (< 7.5%). If there is not enough insulin or medication on board, the amount of carbohydrates may not be a factor.