Capani F, Casalini G, Consoli A, D'Emilio A, La Nave G, Loragno M, Vitacolonna E, Zappone G
Istituto di Medicina Interna, Università degli Studi di Chieti, Italy.
Acta Diabetol Lat. 1991 Jan-Mar;28(1):47-53. doi: 10.1007/BF02732113.
The aim of this work was to observe whether different types of carbohydrates might require different insulin doses. Five type 1 CSII-treated diabetic subjects (age 39 +/- 4 years), C-peptide negative and in optimal metabolic control (HbA1c 7.5 +/- 0.2) were selected for the study. They were connected to the Biostator 6 times with an interval of 4-10 days between sessions and fed a meal containing 75 g of carbohydrates of different types: bread, pasta, potatoes, apples, oranges and sucrose. The following net (above basal) insulin requirement for the 30 meals were found (IU - mean + SD): bread 9.15 +/- 1.97; pasta 6.00 +/- 1.37; potatoes 7.05 +/- 2.58; apples 4.54 +/- 1.42; oranges 6.21 +/- 2.62; sucrose 7.83 +/- 2.33. A statistically significant difference was found by ANOVA among insulin requirements for foods (p less than 0.05). Single comparisons between bread and the other foods showed a statistically significant difference only between bread and apple (p less than 0.05). Mean coefficient of variation was 33.9% for the subjects and 30.7% for the meals. A significant correlation was found between Jenkins' glycaemic index and insulin requirement (r = 0.897; p less than 0.001). In conclusion, the high intraindividual variability of insulin requirement does not advice the use of the glycaemic index during optimized insulin therapy.