A randomised crossover trial: Exploring the dose-response effect of carbohydrate restriction on glycaemia in people with well-controlled type 2 diabetes

J Hum Nutr Diet. 2023 Feb;36(1):51-61. doi: 10.1111/jhn.13030. Epub 2022 May 30.


Background: Trials investigating the role of carbohydrate restriction in the management of glycaemia in type 2 diabetes (T2D) have been confounded by multiple factors, including degree of calorie restriction and dietary protein content, as well as by no clear definition of a low-carbohydrate diet. The present study aimed to provide insight into the relationship between carbohydrate restriction and glycaemia by testing the effect of varying doses of carbohydrate on continuous glucose concentrations within a range of intakes defined as low-carbohydrate at the same time as controlling for confounding factors.

Methods: This was a randomised crossover trial in participants with T2D (HbA1c: 6.6 ± 0.6%, 49 ± 0.9 mmol mol-1 ) testing five different 6-day eucaloric dietary treatments with varying carbohydrate content (10%, 15%, 20%, 25%, and 30% kcal). Diets exchanged %kcal from carbohydrate with fat, keeping protein constant at 15% kcal. Daily self-weighing was employed to ensure weight stability throughout each treatment arm. Between dietary treatments, participants underwent a washout period of at least 7 days and were advised to maintain their habitual diet. Glycaemic control was assessed using a continuous glucose monitoring device.

Results: Twelve participants completed the study. There were no differences in 24-h and post-prandial sensor glucose concentrations between the 30 and 10%kcal doses (7.4 ± 1.1 mmol L-1 vs. 7.6 ± 1.3 mmol L-1 [p = 0.28] and 8.1 ± 1.5 mmol L-1 vs. 8.5 ± 1.4 mmol L-1 [p = 0.28], respectively). In our exploratory analyses, we did not find any dose-response relationship between carbohydrate intake and glycaemia. A small amount of weight loss occurred in each treatment arm (range: 0.4-1.1 kg over the 6 days) but adjusting for these differences did not influence the primary or secondary outcomes.

Conclusions: Modest changes in dietary carbohydrate content in the absence of weight loss at the same time as keeping dietary protein intake constant do not appear to influence glucose concentrations in people with well-controlled T2D.

Summary: This study randomised people with T2D to receive five different doses of carbohydrate from 10% to 30% of calories in random order to see what effect it had on their blood glucose.

Keywords: blood glucose; diabetes mellitus; diet; dietary carbohydrates; dietary proteins; post-prandial period.

Publication types

  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Blood Glucose / metabolism
  • Blood Glucose Self-Monitoring
  • Cross-Over Studies
  • Diabetes Mellitus, Type 2*
  • Diet, Carbohydrate-Restricted
  • Dietary Carbohydrates
  • Dietary Proteins
  • Humans
  • Weight Loss / physiology


  • Blood Glucose
  • Dietary Proteins
  • Dietary Carbohydrates

Associated data

  • ISRCTN/ISRCTN11067343