Wednesday, August 11, 2010

Children and adolescents on intensive insulin therapy maintain postprandial glycaemic control without precise carbohydrate counting

The study has demonstrated that, in free-living children who are adjusting insulin dose for carbohydrate intake, an individually calculated insulin dose covers 10-g variations in carbohydrate amount in a meal, with no differences in area under the postprandial glucose curves and in postprandial BGLs for 2.5 h. Even at 2.5–3.0 h, although there was a small difference in BGLs for the 70-g meal, the postprandial BGLs remained well within current internationally defined targets for optimal glycaemic control of 5–10 mmol/l [11]. Tight glycaemic control was achieved for all test meals (Fig. 2), with no statistical difference in AUC for any meal.
This study has shown for the first time that a mealtime insulin dose can cover a range of carbohydrate quantities in children and adolescents on intensive insulin therapy. It has been suggested that precise carbohydrate counting in grams is preferable to estimations of 10-g CHO portions or 15-g CHO exchanges to achieve optimal postprandial glycaemic control [1,10]. However, there are no studies to support this and the findings of the current study indicate that this degree of accuracy in carbohydrate quantification may be unnecessary to maintain glycaemic control in daily life. Studies are required to examine the impact of more widely variable carbohydrate amounts, such as 15-g CHO exchanges, on postprandial control.
The current study found no difference between postprandial glucose levels for children on CSII vs. MDI when insulin dose was matched to carbohydrate amount. These findings are supported by another study that compared daily glycaemic patterns in children on CSII with matched subjects on MDI [14]. However, in other studies, CSII has been reported to be associated with small improvements in postprandial control [15,16]. Larger studies are needed to determine if there is an association between insulin treatment modality and postprandial glycaemic control.
Several studies in adults have demonstrated that the pre-meal insulin requirement is proportional to the carbohydrate content of the meal [17–19]. Our findings may also be applicable to adults on intensive insulin therapy and do not conflict with these studies, but suggest a particular amount of insulin will cover a range in carbohydrate quantity. In addition, although the glucose response curve may have differed for a meal of a different glycemic index, our findings should be applicable to meals of varying carbohydrate types.
A potential limitation of our study may have been sample size. However, we had sufficient power to detect a difference in glucose excursions of 2.2 mmol/l at 2 h between carbohydrate loads. Education programmes, such as DAFNE [3], cite a rise in BGLs of 2–3 mmol/l as the potential impact of one 10-g CHO portion, although there have been no studies to demonstrate this. A further limitation of our study is that only the effect of a 10-g variation in carbohydrate amount for a 3-h postprandial period was examined. Further studies are needed into the frequency and amount by which children under- or overestimate carbohydrate quantity in meals in their daily lives and the impact this has on glycaemic control.
In clinical practice, precise carbohydrate counting often involves the inclusion of foods that were not counted in the portion or exchange systems [20]. Acquiring such a detailed knowledge of the carbohydrate content of all foods on a meal-to-meal basis increases the burden of management already placed on a child with diabetes and this additional complexity may affect adherence [8]. Furthermore, it may increase fat consumption from packaged foods that have the precise carbohydrate amount specified on the label. This is a concern as children with diabetes higher than recommended fat intakes in a number of studies [21,22].
Education programmes for children and adults that teach adjustments in insulin dose for carbohydrate intake without using precise grams, have demonstrated improved quality of life [23] and glycaemic outcomes [3,24,25]. There is limited evidence regarding the ability of adults and children to count carbohydrate, although some research suggests estimations in grams are often inaccurate [26] and teaching in precise grams fails to improve accuracy compared with portion or exchange estimations [27].
This study demonstrates that an individually calculated insulin dose for meals with ±10-g variations in carbohydrate amount results in maintenance of postprandial BGLs. We conclude that small errors in carbohydrate quantification of less than 10 g are unlikely to make a significant difference to postprandial glycaemic control for meals of approximately 60 g of carbohydrate. Precise carbohydrate counting in gram increments appears unnecessary to maintain postprandial blood glucose control in children and adolescents using intensive insulin therapy.