AIMS: To compare the effects of a low carbohydrate diet (LCD < 100 g carbohydrate/d) and a high carbohydrate diet (HCD > 250 g carbohydrate/d) on glycaemic control and cardiovascular risk factors in adults with type 1 diabetes.

MATERIALS AND METHODS: In a randomized crossover study with two 12-week intervention arms separated by a 12-week washout, 14 participants using sensor-augmented insulin pumps were included. Individual meal plans meeting the carbohydrate criteria were made for each study participant. Actual carbohydrate intake was entered into the insulin pumps throughout the study.

RESULTS: Ten participants completed the study. Daily carbohydrate intake during the two intervention periods was (mean ± standard deviation) 98 ± 11 g and 246 ± 34 g, respectively. Time spent in the range 3.9-10.0 mmol/L (primary outcome) did not differ between groups (LCD 68.6 ± 8.9% vs. HCD 65.3 ± 6.5%, P = 0.316). However, time spent <3.9 mmol/L was less (1.9 vs. 3.6%, P < 0.001) and glycaemic variability (assessed by coefficient of variation) was lower (32.7 vs. 37.5%, P = 0.013) during LCD. No events of severe hypoglycaemia were reported. Participants lost 2.0 ± 2.1 kg during LCD and gained 2.6 ± 1.8 kg during HCD (P = 0.001). No other cardiovascular risk factors, including fasting levels of lipids and inflammatory markers, were significantly affected.

CONCLUSIONS: Compared with an intake of 250 g of carbohydrate per day, restriction of carbohydrate intake to 100 g per day in adults with type 1 diabetes reduced time spent in hypoglycaemia, glycaemic variability and weight with no effect on cardiovascular risk factors.

Original languageEnglish
JournalDiabetes, Obesity and Metabolism
Issue number7
Pages (from-to)1680-1688
Number of pages9
Publication statusPublished - Jul 2019


  • clinical trial
  • randomized trial
  • type 1 diabetes


Dive into the research topics of 'Low versus High Carbohydrate Diet in Type 1 Diabetes: A 12-week randomized open-label crossover study'. Together they form a unique fingerprint.

Cite this