Objectives: The aim of this study is to compare the various methods that nay be used for determination of the contents of glycemic carbohydrates (GC) in food products. The glycemic carbohydrates are those carbohydrates which after ingestion are broken down to glucose.
Methods: The contents of glycemic carbohydrates in food products may be calculated by the indirect method as the difference between total content of carbohydrates and the content of non-glycemic carbohydrates (fiber). In order to calculate the glycemic load the glycemic index and the total content of carbohydrates in food product have to be determined (10). In the direct determination of the content of glycemic carbohydrates (11 – 15) the masses of tested food and of the reference glucose are to known and only the areas of the respective glucose response function should be measured.
Results: The indirect metod of the determination of contents of glycemic carbohydrates (1-3) cumulates errors in determination of the nutritive and non-nutritive components of food product (2). Glycemic load is equal to the content of glycemic carbohydrates, however it inherits the errors in determinations of glycemic index and the total content of carbohydrates in food product (10) which is calculated by (1). The direct method of determination of the content of glycemic carbohydrates in food products does not have the drawbacks of the previous methods.
Conclusions: Glycemic indexes are ambiguous and misleading indicators of glycemic properties of food products. The direct method of determination of glycemic carbohydrates is simple, it has no drawbacks of glycemic indexes and glycemic loads and it may be carried out even in any kitchen.
Key words: glycemic index, glycemic load, glycemic corbohydrates, blood glucose, food products, diabetics
A lot of work has been done in order to determine the glycemic indexes of all kinds of foods [1-3]. From the obtained data the International tables of glycemic indexes and glycemic loads have been compiled . In this paper, we intend to answer a few questions that are important for diabetics:
- a) Which of the methods for determination of the glycemic carbohydrates (GC), glycemic indexes and glycemic loads of food products may provide diabetics with the most reliable data useful in the planning of their everyday diabetic diets. The glycemic (or diabetic) carbohydrates (DC) are those carbohydrates which after ingestion are broken down to glucose.
- b) What are in fact the glycemic indexes of food products and why they do not meet expectations attach to them.
- c) To what extent glycemic loads of food products may replace the glycemic indexes which have turned out to be misleading and useless?
We shall also discuss the features of a new direct method for the determination of glycemic carbohydrates. That method can be used not only by healthy volunteers but also by diabetics of type 2 to determine the glycemic carbohydrates in food products . The contents of glycemic carbohydrates in food products are important for diabetics of type 2 and type1 and all persons interested in planing the proper diets and feeding .
The plan of this paper is as follows. In the following two sections we will review the commonly used methods for the determination of the total content of carbohydrates and the glycemic carbohydrates through the glycemic index and glycemic load. We will also highlight some inherent problems in these methods. Then we will proceed to propose a new method which aims to overcome these problems. We close the paper with a discussion and conclusions.
Determination of the total contents of carbohydrates in food products
The total carbohydrate contents in food products are not determined with the independent method. They are calculated as the difference between the mass of the sample of food product and the masses of the nutrients (fats and proteins) and other ingredients (water and ash) contained in it. Carbohydrates, proteins, and fats, as well as non-nutritive components water and ash are usually expressed per 100 g of food product. The total content of carbohydrates in 100 g of food product is calculated from the following relationship :
mTC g = 100 g – ash g – water g – proteins g – fats g (1)
mTC = %TC total content of carbohydrates in 100 g of food product.
Carbohydrate structures are very complex. Oligosacharides (carbohydrates of low molecular weight with short chains) are easily broken down to glucose and other simple sugars. Polysacharides (carbohydrates, with very long and branched chains) are only partially digested and decomposed into glucose by the digestive juices. For diabetics it is important to know the contents of glycemic carbohydrates in food products from which they are used to prepare their every day meals. The total content of all carbohydrates in food products is not of great importance.
Molecules of glucose that are produced from carbohydrates in the stomach are transfered into the blood, and next into the cells of the body. Those processes are controlled by the pancreas and insulin. Only t (GC), which are broken down to glucose, cause the increase of the concentration of the blood glucose. The non-glycemic carbohydrates (NGC) that are resistant to gastric juices – although they perform various useful functions – do not participate in the energy balance of the body and are excreted from it. The food fiber is the most known component of the non-glycemic carbohydrates. There are also those carbohydrate structures which are only partially decomposed to glucose.
The content of the glycemic carbohydrates (DC) in food product can be calculated from the simple relation:
mGC g = 100 g – ash g – water g – proteins g – fats g – fiber g (2)
mGC g = mTC g – mNGC g (3)
mGC – the mass of the glycemic carbohydrates in a food product
mNGC – the mass of fiber and of carbohydrate structures that are resistant to the gastric juices
Only food fiber can be determined quantitatively. It is therefore assumed, that the content of non-glycemic carbohydrates is equal to the content of food fiber in the food product. It is also worth noting that the values on the left side of equations (1) – (3) cumulate in themselves all errors in the the determination of the nutrient and non-nutrient components of a food product.
Glycemic Indexes (GI) and glycemic loads (GL)
Glycemic indexes of food products have been introduced in the eighties to help diabetics to control glucose blood levels. The glycemic index of a food product is the percentage ratio of the area between the glucose response function (GRF) and its baseline after ingestion by a healthy volunteer the portion of the food product which contains 50 g of carbohydrates to the area between the glucose response function and its baseline after ingestion by the same person of 50 g of reference glucose:
GI = 100*ApG* / AG* (4)
GI – the glycemic index
ApG* – the area between the glucose response function and its baseline after ingestion by
a healthy volunteer the portion of the food product which contains 50 g of carbohydrates
in [min * mmol / l] (1 mmol / l = 18 mg / dl)
AG* – the area between the glucose response function and its baseline after ingestion by
the same person 50 g of reference glucose [min * mmol / l]
The stars * in the subscripts of the respective variables mean that they cannot take any values, but they have to incorporate the restrictions imposed on the respective values by the mass of a food product which should contain 50 g of carbohydrates.
The definition of the glycemic index (4) explains how it should be determined and calculated, but it does not define the glycemic index neither in physical, nor in chemical terms.
The data contained in the definition of the glycemic index have to obey the following relations:
mp* : mpC* = 100 : %TC (5)
ApG* : AG* = mpG* : mG* (6)
mp* – the mass of the portion of a food product, that contains 50 g of carbohydrates;
mpG* the mass of glucose that is obtained after injestion of the tested portion of a food product;
mG* – the mass of 50 g of pure glucose which volunteers consume as the reference substance;
%TC – the percent of total carbohydrates in a food product.
mpC* = mG* – the mass of 50 g of carbohydrates contained in the portion of a food product mp* .
The masses of portions of a food product mp* that are consumed by volunteers are calculated from proportion (5) as follows:
mp* = (100*mpC* )/%TC (7)
and the glycemic index is calculated according to the relation (4).
Multiplying the relation (6) by 100 we get:
100 (ApG* : AG*) =100 (mpG* : mG*) (8)
GI = 100 mpG* / mG* (9)
In the physico-chemical terms the glycemic index of a food product is equal to the percentage of glycemic carbohydrates contained in that portion of a product being calculated in respect to the 50 g of the reference glucose – but not in respect to the mass of the portion of a food product mp*. Equation (9) defines the glycemic index in the physico-chemical terms.
Substituting into equation (9) the mass of reference glucose mG* with the value from relation (5) we obtain:
PpG* – 100* mppG*/mp* = GI*%TC/100 = GL (10)
PpG* – the percentage of glycemic carbohydrates contained in the food product;
GL – the glycemic load.
The relation (10) shows that the glycemic load is equal to the percentage of glycemic carbohydrates in a food product calculated by using the glycemic index (GI) and the content of total carbohydrates %TC (1).
The determination of glycemic carbohydrates in food products
The pure glucose that is used as the reference substance, and the glucose produced by the stomach from glycemic carbohydrates undergo the same regulating mechanism of the pancreas and insulin. In other words, the glucose molecules are indistinguishable in the blood despite their origin.
According to that observation we may write the following relationships[5 – 8];
AG = kG*mG (11)
AG – the area of the glucose response function after consumption of the reference
mG – the mass of the consumed glucose [g];
kG – the proportionality factor of the dimension of [min*mmol/(g*l)]
A similar relationship may be written for the area of the glucose response function after consumption of a food product by the same person:
ApG = kG*mpG (12)
ApG – the area of the glucose response function [min*mmol/(g*l)] after consumption by the
same person of the portion of a food product;
mpG – the mass of glycemic carbohydrates contained in the portion of a ested food [g].
The dimension of the proportionality factor kG depends on units, in which the areas of the glucose response functions are expressed. There are no restrictions imposed on the mass of reference glucose mG (mG* = 50 g) and on the mass of portion of a tested food mp (7).
The mass of glycemic carbohydrates contained in the portion of a tested food can be obtained dividing the relation (12) by the relation (11):
mpG = (mG/AG)* ApG (13)
and hence the percentage of glycemic carbohydrates PpG contained in a food product is given by the following formula
PpG = (mpG*100)/mp = 100*(mG/AG)*ApG /mp (14)
That expression may be written in a bit more practical form:
PpG = (mpG*100)/mp = 100*(mG/AG)/ (mp/ApG) (15)
since only the ratios of the mass to the area of the glucose response function are needed for the calculation.
In relations (11) and (12) we assume that the glucose response function is a linear function of blood glucose. This is justified for healthy persons. The consumption of 50 g of glucose by a healthy person produces the glucose response function, whose the maximum height is of the order of about 9.4 mmol/l (170 mg/ dl)  and therefore it does not exceed the kidney threshold (KT) at about 10 mmmol/l (180 mg/dl). The glucose response function is approximately 120 minutes wide at the level of the baseline.
The same sample of glucose consumed by a diabetic of type 2 produces a much larger glucose response function with the maximum height of about 13.33 mmol/l (240 mg/dl) and approximately is 180 minutes wide at the baseline . The maximum height of such glucose response function exceeds the kidney threshold by about 3.33 mmol/l (60 mg/dl).
As soon as the concentration of blood glucose exceeds the kidney threshold the kidneys begin to remove excessive glucose from blood. Glucose appears not only in the body cells – where it is desirable – but also in the urine – where it is not desired. The excessive concentrations of blood glucose after some time may initiate various health complications and diseases associated with diabetes.
When food products or glucose are consumed in such quantities so that the generated glucose response functions are much bigger than kidney threshold, then in equations (11) and (12) apart fof the linear factor that corresponds to the insulin regulating mechanism ki , an additional factor has to be included that takes into account the removal of glucose by the kidneys kk. That last factor may depend on the concentration of glucose in the blood.
To avoid those complications, diabetics ought to eat only such quantities of pure glucose and food products which generate the glucose response functions that are smaller than the kidney threshold, since then:
kG = ki and kk = 0 (16)
To determine the area of a glucose response function, the concentrations of blood glucose may be measured with any glucometer at least every half hour until it returns to the baseline. The time is measured starting from the first bite of food or a first drink of the glucose solution. The concentration of glucose that was measured after fasting is accepted as the concentration of blood glucose at the zero time. The areas of the glucose response functions of a tested food and that of reference glucose ought to be measured in the same way during two days in any order.
From the values presented in Tables 1 and 2 it is evident that the glycemic indexes do not properly and reliably characterize glycemic properties of carbohydrates contained in food products.
Table 1. The contents of glycemic carbohydrates (columns 4 -8) in food products (A – F) at the respective glycemic indexes (5 – 90) depending on the mass of food products that ought to be consumed (column 3) for their determination .
The letters A to F in Table 1 stand for food products, which differ in the contents of carbohydrates (column 2), and the masses of the portions of food products. It is true that products A – F of the same glycemic index of 25 contain the same amount of glucose of 12.5 g is transfered to blood. But that amount of glucose is contained in very different portions beginning from 1000 g and ending with 56 g (column 3). Normally, a person does not eat such large portions of food as 1000 or 500 grams. However, the portions of food products consumed to determine glycemic indexes are not reported in various tables with GI.
Table 2. The glycemic loads (or the percentages of glycemic carbohydrates contained in 100 g of food products A – F) in respect to the total contents of carbohydrates in food products (column2) and the glycemic indexes (5 – 90, columns 3 – 7).
The 100 g portion of product F of glycemic index 25 supplies the blood with 22.5 g of glucose (column 4), that is much more than the same portions of product A, B and C with glycemic indexes as large as 50, 75 and even 90 (columns 5,6 and 7). Therefore, any referring to the values of glycemic indexes of food products compiled in numerous tables without taking into account the sizes of consumed portions (or the total contents of carbohydrates) seems to be very confusing, if of any value at all.
The values that result from determination of the glycemic loads of food products inherit the errors of the determination of glycemic indexes (4) as well as the errors of the determination of total contents of carbohydrates in those products (1). The errors in the determination of glycemic indexes are rather large since the areas of glucose response functions for the reference glucose and tested foods are very different. They are very large for pure glucose but they are 10 and even 20 times smaller for food products of glucemic indexes of 10 and 5, respectively. This is one of reasons why the glycemic indexes, and in consequence the glycemic loads are loaded with very large errors as well. For GI the following values were obtained: 101.0 ± 30; 74 37.98 ± 70.3 (SD) , and 54.0 ± 6.1 (SEM) .
In the direct method of determination of glycemic carbohydrates in food products (11 – 15), proposed in the current paper, there is no reference to the total content of carbohydrates in the tested foods, which is in marked contrast to the case of both the glycemic indexes and the glycemic loads, where those values are required for their determination.
Moreover, the masses of tested food and the reference glucose are not fixed by any relationships such as (5) – (7). They may be of any size, provided that the glucose response functions do not exceed the kidney threshold. Therefore, the contents of glycemic carbohydrates in food products may be determined not only by healthy persons but also by diabetics of type 2 provided that their glucose response functions are below the kidney threshold.
The accuracy of the determination of the glycemic carbohydrates according to equation (11) – (15) depends mainly on the accuracy in the determination of the areas of glucose response functions. The largest errors in the determination of the areas of glucose response functions are caused by difficulties in the cutting off their baselines . Those errors can be minimized when the areas of glucose response functions both for the food product and for the reference glucose are of the same or similar sizes. That important requirement is practically never fulfilled in the determinations of the glycenic indexes and the glycemic loads, as well.
The same amount of glucose consumed by a healthy person and a diabetic of type 2 usually generates glucose response functions of very different areas and heights. Those ones of healthy persons are much smaller than those of the diabetics type 2. Therefore, the values of the ratio(mG/AG) in equation (14) and (15) characterizes the effectiveness with which the glucose from blood is transferred to the body cells. Therefore, the values of that ratio may be used for diagnostic purposes for diabetics type 2. The smaller is the value of that ratio, the larger is the area of glucose response function and that means that the insulin regulatory system is less efficient.
Conclusions and outlook
The percentages of glycemic carbohydrates may be determined:
- by means of relations (2) and (3) as the difference between total contents of carbohydrates (1) and the content of non-glycemic carbohydrates,
- by determination of glycemic loads (10),
- and by the direct method of determination of glycemic carbohydrates (11) – (15) proposed in the present paper
Glycemic indexes and glycemic loads have genetically built-in defects (compare relations (4) – (10) with the data presented in Tables 1 – 2). Glycemic loads (GL) inherit all the errors associated with the determination of glycemic indexees (4) and the total contents of carbohydrates in tested foods (1).
The direct method of determination of glycemic carbohydrates in food products, or in other words of the carbohydrates which are decomposed to glucose in the stomach, seems to be the most simple and the most accurate of all discussed methods. Moreover, it may be used not only by healthy volunteers but also by diabetics of type 2 in contrast to the determinations of glycemic indexes. The direct method is very simple and accurate and what more it can be carried out in a any household equipped with an electronic balance weighing with accuracy of at least ±1 g.
Diabetics need to know which of the methods, that may be used to characterize glycemic properties of food products are the most reliable in planing their diabetic diets. The accuracy of those methods need to be, therefore, checked and compared at least for such common foods as potatoes, rice, bread and other cereals.
Each food product may be prepared for consumption in various ways [7,8,9]. The way of preparation of some foods may seriously effect the border that separates the glycemic and non-glycemic carbohydrates since it is not fixed and stiff. To some extent it depends on the way the meal is prepared and especially on the temperature and the time of cooking.
It seems to be useful to introduce some necessary standardization in preparation of some foods (at least the temperature and the time of cooking). The time of consumption may be also an important factor. Such commonly used descriptions as raw or cooked carrot are rather incomparable .
We hope that the presented new direct method of determining glycemic carbohydrates will turn out to be useful both for diabetic patients and for medical practitioners.
FS Atkinson, K Foster-Powell, JC Brand-Miller – Diabetes care, 2008, International Tables of Glycemic Index and Glycemic Load Values: 2008, http://care.diabetesjournals.org/content/31/12/2281.short
 R. Chlup, J. Bartek, M. Reznícková, J. Zapletalová, B. Doubravová, L. Chlupováet al, Determination of the glycaemic index of selected foods (white bread and cereal bars) in healthy persons. Biomed Papers 148(1) 17 – 25 (2004)
 Alkaabi et al. Glycemic indices of five varieties of dates in healthy and diabetic subjects, Nutrition Journal 2011 10 – 59, http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3112406/
 H. Kunachowicz, I. Nadolna, K. Iwanow, B. Przygoda, Wartość odżywcza wybranych produktów spożywczych i typowych potraw PZWL. Wydanie IV Warszawa 2005: p. 11
 A. Janik, Blog: Trzeba wiedzieć, jak skutecznie kontrolować swoją cukrzycę, https://cukrzycadieta.wordpress.com,
 ibid., Nieoczekiwane i różniące się między sobą impulsy glukozowe zwyczajnej marchwi
Correspondence with Regional Editor
Dear Dr. Janik,
Thank you for your e-mail submitting your paper, Glycemic indexes and glycemic loads in comparison with the direct determination of glycemic carbohydrates in food products, for consideration by Nutrition.
Upon conducting an internal review with Nutrition’s Regional Editors, your paper unfortunately received a relatively low priority score. The demands on Journal space are great and we attempt to keep our material balanced among various disciplines in nutrition. To attain these goals, sometimes we have to refuse publication of material that is more appropriate for another journal and its readers. We are thus returning it to you without delay so you may send it to another journal. We wish you all the best in your future endeavors.
Lubos Sobotka, M.D., Ph.D.
Dear Dr. Sobotka,
Glycemic indexes and their practical usability for diabetics
Thank you for your last e-mail. I do not intend to ask you to change your final decision or to reconsider the scores that you and other Edidors have attached to my paper.
However, I would like to ask you and the other Nutrition’s Regional Editors to read thoroughly my study once again and in the case of any doubt please contact the experts of FAO and WHO that created the International Tables of Glycemic Indexes to ask them for any explanation and advice that you may need.
Aleksander Janik PhD.
Dear dr Janik,
Your article was not accepted, because formula for glycemic index calculation is outside the scope of major group of our readers.
The main interest of our journal is to publish the articles devoted to clinical nutrition and metabolism.
Our suggestion is that glycemic index (which importance for clinical practice in diabetology it not still properly defined) is special
field of experts and physicians working in diabetes – which is not primary group of our readers.
Dear Dr, Sobotka,
Glycemic indexes and glycemic loads in comparison with the direct determination of glycemic carbohydrates in food products (Ref. No.: NUT-D-15-00272)
I am sorry but I can not accept your explanations (see also ref. [2. 3]). You should realize that your final decision is directed against the nutritional needs of millions of diabetics in the world.
From my part, I consider that matter as closed, but it is still open to Nutrition’s Editors.
Aleksander Janik PhD.
Dear Dr Janik,
I agree with you. However, if the article is important for millions of diabetics, the publication should be submitted to the journal which is devoted to this disease.
The publication in our journal would have only small impact to specialists in diabetes and therefore millions of diabetics will not be informed.
Moreover, we can accept less than 25% of all manuscript submitted and oriented mainly in clinical and general nutrition.
My decision is final, however I wish you successes in more appropriate journal.