Glycemic Index and AUC Glucose/Blood Curve: Why curve differences between foods?

[atod 0]

I understand the AUC Glucose/Blood Level curve Integral (area under curve) is used to calculate the Glycemic Index of certain foods. I have a question on why two sample foods with the same mass of carbohydrates can have different AUC Integrals or Glycemic Indices. I have some possible theories and would appreciate guidance.

To be clear on definitions, the Glycemic Index is calculated using the relative rise of a blood glucose level two hours after consuming a food and it is the integral of (area under) a 2 hours glucose/blood response curve, following a 12 hour fast and ingestion of a sample food with 50 g of carbohydrates. The Index is calculated specifically using the ratio of a blood glucose AUC integral, ignoring the area below the fasting baseline (start of curve rise), for some carbohydrate source and a reference source. The reference source is usually either glucose or white bread. The trapezoidal rule is typically used to calculate the integral, although any method can be used. The Glycemic scale places pure Glucose at 100. The index is calculated using a ratio of two integrals using this formula: (SAMPLE_FOOD mmol/L * time)/(REFERENCE_FOOD mmol/L * time) * 100.

Where I'm confused is that two sample foods with 50 g of carbohydrates should have the same energy in calories, typically 200 kcal (4 kcal/g carb). I see the integral of the AUC curves for different foods can be different, indicating different glucose/blood level changes for each food over the two hour window. This makes me think, the foods have the same energy content, but are changing the body's blood glucose levels differently, resulting in different curves and Glycemic Indices. I first questioned whether the reason for this was that part of a food's curve lied outside the 2-hour window, raising the Glycemic Index slower than another food. I also read that the glucose levels will change differently depending on the type of carbohydrate, and some form of carbohydrates are even impossible for the body to digest and are either digested by bacteria in the gut or simply excreted in the feces, which would resulting in a lower glucose/blood level rise.

Therefore I have open questions on the AUC curve and Glycemic Index: are these numbers different between the same sample size of 50 g of carbohydrates due to differing blood glucose levels, yet the same energy delivered to the body (perhaps in another form other then rising glucose)? Are the levels different due to differences in the body's digestion of the food? Or, is part of the glucose/blood integral of some lower Glycemic Index foods passing beyond the 2-hour window of measurement, which would cause a smaller integral and lower Glycemic Index?

[RickM 1,747]

I think that it mostly gets down to the relative amounts of simple vs. complex carbohydrates in the food - a big difference between 50g in soda pop vs. a plate of broccoli or raspberries where the sugars are mostly bound up with the Fiber, which takes longer to break down and absorb. The free sugars in the soda starts being absorbed in the mouth and all the way down, so that starts tickling your insulin responses immediately while the more complex fruits and veg need to be chewed and digested before much of that happens.

[atod 0]

Interesting. If the digestion is slower for some foods, does that cause the AUC glucose/blood curve to slide right beyond the 2-hour window, which would result in a smaller integral (less area under curve) and a lower glycemic index?

[RickM 1,747]

I believe that is the idea - I've never looked at it from the mathematical perspective (though I tend to be a numbers guy,) but that makes sense. Lower GI foods have a lower impact on your insulin responses than high GI foods. Then there is the related idea of Glycemic Load, where the important factor is the combined GI of an entire meal, so that high GI foods are averaged with the low GI foods to estimate the response to more real world eating. That way, pineapple on your ham has a much less impact on your system than it does by itself.

There is a related concept that I have heard offered more recently that the beyond the basic food statistics, that processing of the foods, including cooking, impacts its absorbability and can/should be considered. This makes some sense in the apparent problems with packaged or processed foods, beyond just the added sugars that are often included. The problem with this seems to be objectively measuring these differences on a wide enough scale to be useful, beyond the basic "avoid packaged/processed foods" advice, or raw is better than cooked as the body needs to do more to break it down.

This whole topic does illustrate the futility of simply counting carbs as a guide to health or weight loss.