As everyone involved in diabetes education knows, diet is a member of the trifecta (diet, exercise, medication) in the treatment of all types of diabetes. The treatment of type 1 diabetes before the discovery of insulin by Banting, Best, and McCleod was via the "starvation" diet, where very small amounts of carbohydrate was doled out to avoid hyperglycemia and prevent Diabetic Ketoacidosis and death. Over the years, the "ADA" diet was developed along with other dietary variations including "carbohydrate counting." Although, carbohydrate counting was developed in the 1960s, carbohydrate counting is particularly in vogue due to the emergence of basal bolus therapy: multiple insulin injections and insulin pump therapy. To take advantage of insulin to carbohydrate ratios in association with insulin sensitivity factors, one must learn to carbohydrate effectively. Practice is important and many books, web links, and applications have been published to help patients (see "Meal Snap", "Carb Counting with Lenny" etc.). It is always reassuring to find evidence-based literature to validate theory into practice. There have been few (if any) controlled studies that evaluated the efficacy of carbohydrate counting in type 1 diabetes.
"Effects of Carbohydrate Counting on Glucose Control and Quality of Life Over 24 weeks in Adult Patients with Type 1 Diabetes on Continuous Subcutaneous Infusion" (a randomized, prospective clinical trial-GIOCAR) was published by Dr. Laurenzi et. al in Diabetes Care 34, April 2011, 823-827). According to the authors, this is the first study to validate the efficacy of carbohydrate counting in adult patients using insulin pump therapy. The goal of the study was to "test the effect of carbohydrate counting on glycemic control and (most importantly) the quality of life in these adults on continuous subcutaneous infusion (CSII)." 61 adult patients with T1DM were randomly assigned to 2 groups:
1. Intervention group: learn carbohydrate counting
2. Control group: estimation of pre-meal insulin dose empirically using a 1:1 ratio
At baseline, 12 weeks, and 24 weeks, the authors measured hb A1c, fasting blood sugar, BMI, waist circumference, recorded total daily does of insulin, capillary glucose data, and administered the DSQOLS questionnaire (Diabetes-Specific-Quality-of-Life-Scale- a validated tool to measure quality of life).
In this Italian study, patients were given the Complete Guide to Carb Counting (2nd edition) as a reference for carbohydrate counting. The dieticians and diabetologists on the team worked with the intervention group to instruct how to count carbs and use the insulin/carb ratios to determine pre-meal insulin doses in association with an insulin sensitivity factor to correct blood sugars. Thus, combination of pre-prandial insulin based on the insulin/carbohydrate ratio plus correction insulin based on insulin sensitivity factor determined the dose to give pre-meal. Blood glucose was measured 6 times/day. Before randomization into groups, all 61 patients attended a group lesson with dietician to learn about the recommended diet for patients with diabetes.
What were the results? Intention-to-treat analysis showed:
1. Improvement of the DSQOLS score related to diet restriction. At week 24: baseline difference, p=0.008 in the intervention group as compared with control subjects.
2. Reduction of BMI (p=0.003) and waist circumference (p=0.002) in the intervention group as compared with control subjects.
3. No changes in hb A1c, fasting blood glucose, daily insulin dose, and hypoglycemic episodes were observed.
4. Per-protocol analysis, including only patients who continuously used carbohydrate counting and insulin pump therapy during the study, confirmed improvement of the DSQOLS score and reduction of BMI, waist circumference, and demonstrated a significant reduction of HbA1c (-0.35% versus control subjects, p=0.05)
The authors have concluded that in this cohort of adult patients with type 1 diabetes, treated with insulin pump therapy, "carbohydrate counting is safe, improves quality of life, reduces BMI and waist circumference, and in per-protocol analysis, reduces HbA1c. (It was hypothesized that there was not an observed improvement of hb A1c in the authors Intention to Treat Analysis due to the relatively short six-month study or not all participants trained in carbohydrate counting used it.)
As with all studies, there are limitations:
1. Patients in the intervention group had more interactions with the diabetes team during carbohydrate counting instruction. Can the improvement in metabolic control and weight loss be secondary to extra attention from the team versus carbohydrate counting?
2. Short duration of study, which leads to not evaluating long term results of carb counting.
3. Physical activity in relation to food intake was not measured.
Note, however, that this is the first study to report a randomized clinical trial testing the effects of carbohydrate counting in adults with type 1 diabetes treated with insulin pump, which does indeed demonstrate "that a reduction in Hb A1c, not accompanied by an increase in hypoglycemic events, may be expected when patients continuously use carbohydrate counting in daily management of their diabetes." It is my belief that we can extrapolate these results to my pediatric population as well. It is always reassuring to be able to demonstrate efficacy of a certain world wide practice assumed to be effective with evidence-based medicine.
Published On: June 21, 2011