Previous page 119

page_200

next page >

Table 12.2 The composition, normal serving size and weight of the 50 g carbohydrate portion of the test foods based on manufacturers' data.

Food

White bread (control)

Candy-coated chocolate peanut pieces

Chocolate-covered caramel and nougat bar

Chocolate-covered caramel and nougat bar1 ('lite' formulation)

Chocolate-covered caramel cookie bar

Chocolate-covered peanut and caramel bar

Chocolate granola bar

Energy bar 1 (cocoa-based)

Energy bar 2 (cocoa-based)

Energy drink2 (chocolate milk)

Milk chocolate bar

1 Includes polydextrose.

Energy Protein Fat Carbohydrate Normal

  • serving size (g)
  • kcal) (%) (%) 2.4

244 507 460 382 477 477 432 354 353 88 542

43.8

60.9

70.5 76.41

59.6

58.7

100 ml

36.9

Weight of 50 g carbohydrate portion (g)

114.2

82.1

71.0

81.6

73.2

72.3

69.4 385 ml

83.9

low glucose responses, the average GI being 49 and average II being 59. The candy-coated chocolate peanut pieces gave the lowest GI (GI = 33) while the chocolate granola bar and the chocolate-covered caramel and nougat bar had the highest (GI = 62). The latter value was comparable to that obtained by Jenkins et al. in 1981 for a similar product (Mars Bar) sold in Canada (4).

One reason for the relatively low GI values is the high fat content of the products. Fat in food is well known to slow gastric emptying, thereby producing lower glucose responses, but importantly, insulin responses are not always correspondingly low (25). Hence, high-fat foods such as chocolate may appear to be very desirable if GI were the only criterion for food selection (it shouldn't be).

One would normally expect the GI and II to be roughly similar since the blood glucose rise is the major stimulus for insulin secretion. Plasma glucose and insulin responses have correlated well in studies of sugary foods (5, 10) and starchy foods (26). Surprisingly, among this group of foods there was no positive correlation between glucose and insulin (r = 0.3, NS, Fig. 12.2). In fact, in three cases the II

< previous page page_201

next page >

Table 12.3 The glycemic and insulin indices of chocolate products. Chocolate product

Candy-coated chocolate peanut pieces

Chocolate-covered caramel and nougat bar

Chocolate-covered caramel and nougat bar ('lite' formulation)

Chocolate-covered caramel cookie bar

Chocolate-covered peanut and caramel bar

Chocolate granola bar

Energy bar 1 (cocoa-based)

Energy bar 2 (cocoa-based)

Energy drink (chocolate milk)

Milk chocolate bar

Plasma glucose response to milk chocolate in ten healthy subjects.

Plasma glucose response to milk chocolate in ten healthy subjects.

was double that predicted (chocolate bar, candy-coated chocolate peanut pieces, chocolate-covered peanut and caramel bar). These unexpected findings therefore corroborate other reports that chocolate products sometimes produce disproportionately high insulin responses (10).

While the fat content might seem the obvious reason for the disparity between glucose and insulin, it is not the entire explanation. Foods with a similarly high fat content, such as potato chips and peanut butter, do not produce as much insulinemia as certain chocolate products (9). Furthermore, Brand Miller et al. (5) found that some low-fat chocolate products (<2% fat w/w) displayed the same disparity between glucose and insulin. For example, Kellogg's Coco-pops (a

< previous page page_202

next page >

etin

Plasma insulin response to milk chocolate in ten healthy subjects.

H read

ChycjliJ[e har

Time (min)

Plasma insulin response to milk chocolate in ten healthy subjects.

chocolate-coated puffed rice breakfast cereal) gave a GI of 77 but an II almost twice as high as expected (II = 124), the highest value seen in that study of 39 common foods). Similarly, the GI of a low-fat chocolate milk was 34 while the II was 69. In these instances, the high insulin response cannot be explained by large amounts of fat.

The quantity and quality of the protein may be another explanation for the high insulin response to some chocolate products. Like carbohydrate, protein also elicits an insulin response (27) and cocoa powder contains significant amounts of protein (18.5 g/100 g) (20). In healthy subjects, the degree of insulinemia (AUC) after consuming a 'pure' protein load is about 30% of that seen with the same amount of glucose (27). However, the amount of protein per 50 g carbohydrate portion was similar in most of the products that we tested. It is possible that cocoa is a source of non-nutritive substances that stimulate insulin secretion, but given the variation in insulin responses among similar chocolate products, other food ingredients may be responsible for the effect (28). Further research in this area is needed.

While a small number of chocolate-containing products appear to raise insulin levels higher than anticipated, their small contribution of calories to the diet suggests that they are not a significant health threat.

Significance of High Insulin Responses

Individual variation in insulin responses to foods is determined to a large degree by the level of insulin resistance in the individual (29). Insulin resistance is

Delicious Diabetic Recipes

Delicious Diabetic Recipes

This brilliant guide will teach you how to cook all those delicious recipes for people who have diabetes.

Get My Free Ebook


Post a comment