Earlier this year, an expanded and updated glycemic index (GI) was published.1 Its tables contain hundreds of entries, including multiple studies on the same foods. It is much more complete than the typical 35 food lists commonly published.
About eight years ago, I was in the health food store next to my office when a local nutritionist (and hairdresser) told the counter girl not to put shredded carrots on his sandwich because they were fattening. Those of you who know me will be stunned that I kept my mouth shut - but my mind sure was racing!
A couple of months prior, the 1995 GI tables had been published. It was very newsworthy to hear that the GIs for foods like rice cakes and carrots were higher than those of white sugar. In the next few years, many popular diet books focused on high-glycemic carbohydrates as the cause of the skyrocketing rates of obesity seen in developed nations.
I do agree that refined carbohydrates are not health foods and should be consumed sparingly. I also agree that the GI of foods is a potentially useful tool. However, it can also be misused. I wanted to tell the guy at the health food store, "Dude, when you roll to the local supermarket and see all that stored energy, too many carrots is not the problem! And the GI of a single food on an empty stomach does not mean the body will respond in kind when it's consumed with a mixed meal at midday."
The GI is the blood glucose response to a given food, and is usually calculated by measuring the blood sugar of test subjects that are fasting, feeding them a measured amount of a test food, and retesting the blood two to three hours later. This number is adjusted to an equal amount in weight of the reference food, either white bread or glucose. The glycemic index is the newest way to classify carbohydrate-rich foods previously categorized as either simple or complex, or as sugars or starches. There has been increased utilization of the glycemic index; however, there are some important limitations. Published glycemic index values can vary to a level that disturbs some researchers and health care professionals. The following is an explanation of why seemingly similar foods can have varied numbers:
Variable Problems and Explanations
Blood Testing Method: In GI studies, blood is extracted from capillaries or veins. Capillaries demonstrate greater changes in postprandial glucose concentrations than do veins.1
Time Period of the Test: Gl testing takes generally two hours; however, tests have ranged from one-and-a-half to three hours. Therefore, if a given laboratory tests a specific food after one-and-a-half hours and another laboratory tests the same food after three hours, there is a good chance that the GI values will differ.
Cooking Time: The degree of cooking affects the degree of starch gelatinization which, in turn, affects the digestibility and, therefore, the rate and level of blood sugar increase.1,14
Differences in Ingredients: If two laboratories test pasta with a red sauce, for example, the presence, absence or amount of oil; the presence or absence of sugars; and the water content of the sauce can vary and affect the GI.1
Botanical Differences: Rice values tend to vary widely, due to the strain of rice and its percentage of the starches amylose and amylopectin. Rice that tends to be higher in amylose will also have a lower GI.14 Again, the content of these starches in the same strains of rice can vary, depending on where they are grown.1,14
Laboratory Methods for Determining Carbohydrate Content of the Test Food: Some laboratories will use published food tables, while others will directly measure the amount of starch and sugar in a given food. This may clearly cause discrepancies due to the individual laboratory methods involved in measuring, as opposed to the methods used in published tables.1,14
Different Portion Sizes: Different portion sizes for fluids and solids, when used in various studies, can affect GI.1
Difficulty Measuring Resistant Starches: The carbohydrate portion of a test food should not include resistant starch; however, it is very difficult for laboratories to calculate the exact amount of resistant starches and may be over- or underemphasized, both of which can result in a different GI.1
Biochemical Individuality: The GIs are an average of the test subjects. For example, the GI of sweet corn2 on the glucose scale was 59, plus or minus 11 for five healthy people. That means the individual GIs varied on an 80-gram portion of sweet corn from 48 to 70.1
Degree of Stored Glycogen: When cyclists were given a high-GI meal, there was a much smaller rise in their blood glucose levels when the test was performed after exhaustive exercise. The researchers felt that due to muscle glycogen depletion, the glycemic response was muted. Therefore, the GI of foods can also depend on the level of stored glycogen present in the body.3
Level of Mastication: If a food is not well masticated, the GI level of that food will be lower.
Even with the above given limitations and variability, research has shown some very interesting trends comparing low-versus-high GI diets:
increased rates of type II diabetes4,5
improvements in glycated hemoglobin concentration in type I diabetics10,11
increased rates of cardiovascular disease6
improved insulin sensitivity in patients at risk for cardiovascular disease10
lower levels of HDL cholesterol in healthy people7
improved blood lipid profiles in patients at risk for cardiovascular disease10
greater oxidative stress9
elevated HDL cholesterol levels in type I diabetic patients12
greater postprandial insulin levels1
decreased rates of breast cancer13
higher levels of triglycerides in healthy people8
decreased rates of colon cancer1
decreased rates of obesity6
About those Carrots
The new GI tables include four studies on carrots.The indices of carrots, on a glucose scale for each study, were 16, 32, 49 and 92. The authors then calculated the mean of the four trials. The GI of carrots is now 47. (I just knew they weren't the problem!)
- Foster-Powell K, Holt S, Brand-Miller J. International table of glycemic index and glycemic load values 2002. American Journal of Clinical Nutrition 2002;76:5-56.
- Jenkins D, Wolever T, Taylor R, et al. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981;34:362-366.
- Burke L, Collier G, Hargraves M. Muscle glycogen storage after prolonged exercise: affect of the glycemic index of carbohydrate feedings. J Applicant Physiol 1993;75:1019-23.
- Salmeron J, Ascherio A, Rimm E, et al. Dietary fiber, glycemic load, and risk of NIDDM in men. Diabetes Care 1997;20:545-50.
- Salmeron J, Manson J, Stampfer M, et al. Dietary fiber, glycemic load, and risk of NIDDM in women. JAMA 1997;277:472-7.
- Ludwig D. Dietary glycemic index and obesity. J Nutr 2000;130:280S-3S.
- Frost G, Leeds A, Dore C, et al. Glycemic index as a determinate of serum HDL cholesterol concentration. Lancet 1999;353:1045-8.
- Gavin J. Pathophysiologic mechanisms of posttranual hyperglycemia. Am J Cardiol 2001;88:4-8.
- Ceriello A, Bortolotti N, Motz E, et al. Meal-induced oxidative stress and low-density lipoprotein oxidation in diabetes: the possible role of hyperglycemia. J Metab 1999;48:1503-8.
- Frost G, Leeds A, Trew G, et al. Insulin sensitivity in women at risk of coronary heart disease and the effect of a low-glycemic diet. J Metab 1998;47: 1245-51.
- Giacco R, Parillo M, Rivellese A, et al. Long-term dietary treatment with increased amount of fiber-rich low-glycemic-index natural foods improves blood glucose control and reduces the number of hypoglycemic events in type I diabetic patients. Diabetes Care 2000;23:1461-6.
- Toeller M, Buyken A, Heitkamp G, et al. Nutrient intakes as predictors of body weight in european people with type I diabetes. Int J Obes Relat Metab Disord 2001;25:1-8.
- Franceschi S, Dal M, Augustin L, et al. Dietary glycemic load and colorectal cancer risk. Ann Oncol 2001;12:173-8.
- Bland J, et al. Clinical Nutrition; A Functional Approach. Institute for Functional Medicine 1999; Gig Harbor, Washington.
Click here for more information about G. Douglas Andersen, DC, DACBSP, CCN.