INTRODUCTION

Carbohydrates (CHO) are the main nutrients in diets across the world. Their primary function is to act as energy fuel, and according to their structure vary in their palatability, digestion, absorption, release of hormones, and oxidation (¹,²). Glycaemic index (GI) is an indicator of the healthy quality of CHO present in food, and is defined as the increase in the area under the blood glucose curve that is produced by the intake of a fixed amount of available CHO from a given food, usually 50 g, in relation to the same amount of CHO from a standard food (glucose or white bread) (³). Therefore, foods that occupy more than 70 % of the area under the blood glucose curve of the standard food are considered high-GI, and those with a small glycaemic increase, less than 55 %, are referred to as low-GI (⁴).

There are certain factors that affect the GI of food and therefore influence glycaemic response, such as type of sugar ingested, starch nature, cooking method and elaboration of foods that modify gelatinization, gelling, and retrogradation degree of starch (⁵).

Other factors that affect GI are the preliminary culinary techniques or mechanical procedures that increase contact surface, thus increasing the area exposed for enzymatic hydrolysis (^{6 7 8}-⁹). On the other hand, the addition of ingredients with high amounts of dietary fibre, proteins, or fats can decrease the GI of food, reducing its glycaemic response (^{10 11 12}-¹³).

The traditional food prepared in Chile is based on tubers (potatoes) and refined cereals, mainly derived from wheat and rice, which, when applying typical culinary techniques, present high GI values. A local study found that although the food preparations preferred by the Chilean population with type-2 diabetes mellitus (DM2) presented a daily average of CHO within the recommended range (219.8 g/day ± 27.0), average GI was high (74.9% ± 11.3), which in turn showed a strong positive correlation with the metabolic control parameters of DM2 (¹⁴). These preparations are also often consumed by the general population in Chile, which could imply an increased risk of developing DM2 among healthy subjects, and could also induce complications in subjects with DM2.

The purpose of this study was to assess the acceptability of low-GI preparations equivalent to traditional high-GI ones in Chile, elaborated with minimal changes in the ingredients and culinary techniques to significantly diminish GI while maintaining acceptability levels.

MATERIALS AND METHODS

RESULTS

MAIN INGREDIENTS AND THEORETICAL GLYCAEMIC INDEX OF LUNCHES

Twelve high-GI preparations with their respective low-GI counterparts were produced. The contribution of available CHO from the high-GI group was on average 53.4 ± 3.6 g, while in the low-GI group it was 48.8 ± 3.6 g, which showed no statistical significance.

Table I shows the preparations with their glycaemic index values and the complete menus, which were comprised of a salad, a main dish, and a dessert, with detailed modifications made with the aim of obtaining low-GI preparations from the initial high-GI recipes. Desserts had modifications involving the selection of ingredients with a lower GI, quantity of protein incorporated, processing of raw materials, and quantity of fibre contributed. In salads and main dishes refined foods were replaced by wholemeals, tubers by vegetables, cooked vegetables by raw vegetables, peeled vegetables by vegetables with peel, and grated foods by whole foods.

Table I.  Preparations included in the meals and determinant GI main ingredients 

The lunch's name is given according to the main dish. ^(a) “Niño envuelto” refers to a thin slice of meat wrapping vegetables in sticks. CHO: carbohydrate; GI: glycaemic index. * Statistically significant difference between the glycaemic indices of both groups as measured by Student's t-test for independent samples (p < 0.001).

ACCEPTABILITY OF FOOD PREPARATIONS

All the preparations in figure 1 were classified as optimal, exceeding the established cut-off point.

Figure 1.  General acceptability of high and low GI lunches according to a 7-point hedonic scale. Values are expressed as mean ± SD. ■: High glycaemic index; □: Low glycaemic index. ^(a) Lunches are represented with the name of the main dish in each one, and include salad, main dish, and dessert. ^(b) “Niño envuelto” refers to a thin slice of meat wrapping vegetables in sticks. *Statistically significant difference. 

Only two lunches had significant differences in their acceptability levels: the low-GI “legume with CHO” lunch had a higher acceptability against its high-GI version, while the high-GI “chicken with corn” lunch got a higher acceptability value when compared to its low-GI equivalent. The courses that had greater scores included the dessert in the low-GI “legume with CHO” lunch, while in the high-GI “chicken with corn” lunch both the main dish and dessert obtained a better score than their low-GI version (data not shown in the figure).

SENSORY ATTRIBUTES OF FOOD PREPARATIONS

In figure 2 there was a preference for the low-GI salad and dessert dishes, while the main dishes were slightly better praised regarding flavour and texture in the high-GI preparations.

Figure 2.  Acceptability evaluation of sensory parameters on a 9-point scale for the components of the high and low-GI lunches. Values are represented as mean ± SD. ■: High glycaemic index; □: Low glycaemic index. *Statistically significant difference when comparing the attributes of each GI group, as measured by Student's t-test for paired samples. 

ACCEPTABILITY RANKING FOR LOW GLYCAEMIC INDEX LUNCHES

Table II compares the ranking that was compiled based on both score scales. Some preparations were placed in the same position in both rankings, such as “Meat soup” (1st place), while the lowest-ranking positions were held by the same three preparations.

Table II.  Acceptability ranking for low glycaemic index lunches according both scales 

CHO: carbohydrates. ^(a) “Niño envuelto” refers to a thin slice of meat wrapping vegetables in sticks. ^(b) Two or three preparations in the same level represent a tie in that ranking position.

DISCUSSION

It is essential to be able to offer the population healthy food preparations with a low GI that may be widely accepted, so that such options will become the preference over traditional diets, which usually present high GI levels based on their cereals and processed derivatives (¹⁷).

In the present study, 12 preparations that are typically consumed in Chile were used as determined by Pincheira's study (¹⁴) with a high GI, and from them it was possible to obtain 12 low-GI counterpart dishes, with a significant difference when comparing the average GI obtained between both preparations. In order to remove the interference of GL (glycaemic load) from the preparations, and isolate the GI as the main variable, all the menus were prepared with a total of 50 to 60 g of CHO.

In order to modify and convert high-GI lunches into low-GI preparations, the extrinsic and intrinsic factors that modify the speed of digestion for food CHO, and therefore their GI, was used. The mechanical process was considered a variable to modify from the typical preparations habitually used, which by reducing the size of food particles allow a greater surface/volume ratio and thus make them more susceptible to salivary and intestinal amylases, increasing in this way the GI (¹⁸). In this sense, a study showed that a one-inch potato cube increases its GI by 25 % if crushed, depending also on factors such as type of starch, physical form, and gelatinization ability (¹⁹).

For some preparations (legumes with cereals, pastries, mashed potatoes, rice, etc.), starch gelatinization was considered a determinant factor of GI, with granules absorbing water, thus considerably increasing their volume and viscosity. This structural change increases physical accessibility and facilitates the function of intestinal enzymes during the digestive process, thus increasing the GI of food (²⁰). For example, raw potato is less digestible than cooked potato, where the hydrolysis of starch is favoured by the capacity to become gelatinized at high temperatures (55-66 °C) in the presence of water. In this situation, the endogenous amylases are activated during the cooking process, and before being denatured by the effect of temperature they manage to degrade the starch to dextrins (¹⁹).

In the present study, the use of dry heat was encouraged in the preparations of low GI, instead of humid heat (²¹). The insoluble fibre content of each food was also considered, as this could act as a barrier to amylase function, and therefore further reduce glucose absorption, which decreases the GI of the preparations (²²).

On the other hand, the soluble dietary fibre, both at the gastric and intestinal level, as a result of its viscosity, slows gastric emptying and transit time, respectively, in addition to the small intestine, which increases the thickness of the layer of water that should pass the solutes to reach the membrane of the enterocyte, generating a reduction in glucose absorption and thus resulting in a flattened glycaemia curve with an insulin response according to the demand of glycaemia. To achieve the described properties, it was necessary to consider the amount of dietary fibre contained in each food, with or without peel, raw or cooked food, and the consumption of pulp or juice squeezed from fruit (²³,²⁴).

In some of the preparations, the addition of foods rich in proteins and fats played an important role. The proteins to be consumed with CHO generate a delayed increase in insulinemic response, which is on average 3-4 hours postprandial. The phenomenon is especially evident with pasta, where the presence of gluten slowed the action of digestive amylases, which limits the absorption of glucose. In the case of fats, it has been seen that, when consumed in conjunction with CHO, the area under the curve of the glycaemic response diminishes within the first 2-3 hours, which may be due to delay in gastric emptying (¹⁰). It should also be noted that a decrease in GI occurs through the intestinal hormones known as incretins, where the glucagon-like peptide-1 (GLP-1) stimulates the production of insulin, and is secreted in the L-cells of the ileum after ingestion of fats. This peptide promotes a reduction of the stomach's acid secretion and intestinal motility, which decreases the rate of absorption of nutrients, and in turn increases the sensitivity of peripheral tissues to insulin (¹¹,¹²). A clear example can be seen by placing a potato into the oven without fat, which has a higher GI (95 %) than a potato chip to which fat has been added during cooking (75 %) (¹⁹).

Another factor to consider is the production process and manufacturing of the products. Indeed, certain industrial processes lead to an increase in gelatinization, as is the case with the manufacturing of instant mashed potatoes, which increases by about 30 % their GI when compared with natural mashed potatoes (⁴,²⁵). On the other hand, the process of pasta binding decreases its GI due to the extrusion of the dough, which leads to a heating effect that results in the formation of a protective layer that will help slow down starch gelatinization during cooking. This is applicable for long pasta such as spaghetti and noodles; however, research has shown that short pasta made from the same flour may have higher, even 2-fold values due to an increase in contact surface with digestive enzymes and their increased propensity to gelatinize, just as happens with soup noodles (82 %) as compared with spaghetti (40 %) (¹⁶).

The challenge of this study was making culinary changes that would convert a high-GI preparation into a low-GI one, while maintaining a good level of acceptability. Regarding the results of general acceptability for each menu in both groups, a high level of approval was successfully achieved. This result was similar to that shown in a clinical trial in subjects with obesity (²⁶). The averages obtained in most of the high- and low-GI lunches were higher than 6 points (“I moderately like it”); only 3 obtained average scores below 6 points, presenting the most noticeable differences with their counterparts; however they did exceed the established optimal cut-off point. When evaluating the average acceptability of the salads, main dishes, and desserts consumed in the low- and high-GI lunches, it was observed that, while there were no significant differences between both lunches in each group, all of them exceeded 6 points on average, with the main dishes having a higher average than the others.

When comparing the specific acceptability values of the salads, main dishes, and desserts that made up each high- and low-GI lunch, it was observed that the best evaluations applied to main dishes, either high- or low GI. In another hand, the most noticeable differences between high- and low-GI meals were found in desserts, which can be attributed to menu variability, personal preferences in taste, complexity of preparations, and/or the high prevalence of fruits, considering that usual consumption of fruits in Chile has decreased (²⁷).

The “legumes with CHO” menu showed greater acceptability in its low-GI version, possibly due to a significant difference specifically in its dessert (Bavarois), which could probably be attributed to the addition of evaporated milk generating changes in consistency and flavour. On the contrary, the “chicken with corn” high-GI main dish and dessert showed significantly greater acceptability than their low-GI counterparts, possibly because the subjects preferred the preparation containing a whole piece of chicken separated from the stew to small pieces of chicken amongst a mixture of vegetables, which is commonly a preference related to the visual size of the portion of food on each plate (²⁸,²⁹). In relation to the dessert, the high-GI version had a higher score than the low-GI “oats with milk”, most likely because texture and appearance were not well received, this being a dish that is seldom eaten. Evidence showed that this pattern of behaviour repeated itself, with high-GI food obtaining greater acceptability scores than low-GI food when the latter version had ingredients not commonly used on a daily basis (³⁰,³¹).

It should be noted that two methods of sensory evaluation were used, one that assessed general acceptability and one that assessed attributes separately, achieving a consensus in the final ranking of the best and worst menus. These findings confirm the reliability of the results obtained.

CONCLUSION

Given that the population has a strong preference for high-GI food and preparations, the introduction of low-GI diets could be quite challenging, as it would be necessary to implement changes to promote a decrease in GI, something that would be unusual for the population and their eating habits. This study was able to show that it is possible to produce low-GI menus equivalent to traditional ones that are perceived as generally acceptable, and that have some components (salad, main dish or dessert) with high acceptability scores, equal to or even better than typical Chilean culinary preparations, as well as components that could not achieve a greater acceptability.