Glycemic response of volunteers to the consumption of supplements and food formulas for oral and/or enteral nutrition

The objective of this study was to determine the glycemic index (GI) and glycemic load (GL) of three products from the Brazilian market used as a supplement and food formula for oral and/or enteral nutrition. The volunteers (n = 16) attended Food Research Center weekly for six weeks after a 10–12-h overnight fasting. Blood was sampled in the fasting state (t = 0) and at 15 min, 30 min, 45 min, 60 min, 90 min, and 120 min after starting to eat each evaluated meal: glucose solution (reference food, three times) and three products: Cubitan® vanilla (specific for wounds healing), Diasip® chocolate, and Diasip® vanilla (diabetic supplements). GI was determined by calculating the area under the glycemic response curve using the trapezoidal rule and ignoring the areas below the fasting line and considering the GI of glucose to be 100. To determine GL, it was considered the amount of carbohydrates available in a standard serving of the product and GI. The three products studied showed low GI and low GL (Cubitan® GI = 35, GL = 6; Diasip® chocolate GI = 49, GL = 7; Diasip® vanilla GI = 47, GL = 7), with significant differences from those and the reference food, but no significant difference between them. Similar results were also observed for the blood glucose peak, which occurred 30 min after the consumption of all products. GI and GL of the products were considerably lower than those of the reference food. The products evaluated presented a low glycemic response, shown by a glycemic response curve with a slightly accentuated shape and no high peaks.


Introduction
The glycemic index (GI) is an indicator of the relative impact of carbohydrates present in foods on plasma glucose concentration. It allows the quality of carbohydrates to be assessed since equal amounts of available carbohydrates are compared between a test food and a reference food [1]. The amount and the quality of these carbohydrates is responsible for up to 90% of the postprandial insulin response of the meal [2]. The glycemic load (GL), in turn, also considers the amount of carbohydrates available in a standard serving of food, and not only the quality of the carbohydrates [3,4].
A low GI and/or low GL diet has beneficial effects on several metabolic and physiological aspects associated with chronic non-communicable diseases [5][6][7][8][9]. These benefits include reduced insulin demand, improved plasma glucose control, and reduced serum levels of lipids [10]. These aspects are closely related to diabetes mellitus control, a disease that is considered an epidemic by the World Health Organization [11].
Diabetes management involves dietary and lifestyle modification, which is sometimes combined with pharmacological intervention. The use of food supplements by people with diabetes has become a common practice for controlling the disease by helping to control blood glucose or balancing nutritional status. These products are also used by hospitalized patients with hyperglycemia.
Chronic hyperglycemia interferes with the homeostasis of glucose metabolism in endothelial cells, nerve cells, and peripheral blood mononuclear cells, as it allows excessive intracellular glucose entry and reduces the process of glycolysis; these factors end up promoting the production of reactive oxygen species and other toxic compounds. These changes lead to damage to various tissues, including the skin, which may present a persistent inflammatory state, compromising wound healing [12].
The main guidelines on the care of diabetic patients suggest the use of a specialized oral supplement to improve glycemic control [13][14][15][16] which is the proposal of Diasip® and that is related to the composition of the formula, because it contains lower carbohydrate content, in addition to low GI carbohydrates and high fiber content, which improve postprandial blood glucose levels. Patients with both pressure injuries and other wounds, common in diabetics patients, require a careful diet that must be rich in energy, protein, arginine, zinc, and vitamins A, C, and E [15][16][17][18], which can be supplied with the use of specific food supplements, and it was this goal that Cubitan® was developed. Therefore, it is essential that diabetic patients have options for an oral nutritional supplement that assists in wound healing and contains low GI and GL. In addition, many critically ill patients who have healing problems should also have extreme care with hyperglycemia so that this product can meet both objectives.
Diabetes-associated morbidity and mortality include serious injury complications, financially impacting patients and their families and healthcare systems worldwide [19]. Cereda et al. [20,21] have demonstrated that an adequate supply of arginine and micronutrients is crucial in promoting wound healing. In addition, they stated that nutritional intervention should be considered an integral part of wound care since its cost-effectiveness makes it advantageous for both patients and the healthcare system. Other experts have also found that specialized nutritional therapy with the addition of arginine, zinc, and antioxidant vitamins can promote the wound healing of patients [16,22].
There are few data published about the GI and GL of supplements and food formulas for oral and/or enteral nutrition, especially in Brazil. Knowing these parameters is important for healthcare professionals establishing a nutritional strategy to manage a clinical condition such as diabetes mellitus. In addition, populations in different regions of the world may have different characteristics and lifestyles, which makes it important to evaluate the glycemic response of foods in their region of consumption and the different presentations available in each market.
The aim of this study was to determine GI and GL produced by the consumption of a specific oral supplement for glycemic control (Diasip®) and a formula for oral and/ or enteral nutrition specific for wound healing (Cubitan®) available on the Brazilian market.

Materials and methods
Three products were studied: Cubitan® (vanilla flavor, liquid presentation) and Diasip® (chocolate and vanilla flavors, powder presentation), produced by Danone Nutricia Ltda. The glucose solution was used as the reference food.
The products are registered in the Brazilian Health Regulatory Agency of the Ministry of Health (ANVISA), are available on the Brazilian market, and meet the microbiological standards established by RDC 12/01 of ANVISA for category 26 (foods for specific population groups), which includes the determination of Salmonella, Staphylococcus coagulase-positive, coliforms at 35 °C, and aerobic mesophilic bacteria.
The Laboratory of Food Chemistry and Biochemistry of the Food and Experimental Nutrition Department at FCF/ USP performed specific chemical analyses to determine the available carbohydrates of each product. The amount of product to be consumed by participants in our study was calculated from the content of available carbohydrates as the sum of the soluble sugars and available oligosaccharides present in the product.

Determination of soluble sugars
Soluble sugars were quantified by high-performance liquid chromatography. The determination was carried out in triplicate, and the results were expressed in grams (g) per 100 g of sample on a whole basis.

Determination of resistant starch (RS) and available oligosaccharides
The analysis of RS and available oligosaccharides was performed using the AOAC 2002.02 method [23] after washing each product with ethanol and performing an enzymatic process for 16 h. The free glucose content was quantified by an enzymatic method (glucose oxidase/peroxidase/ABTS) [24].

Determination of blood glucose
Capillary blood was obtained by finger capillary puncture to determine blood glucose using a Roche® Accu-Chek glucometer according to the recommended methodology [4,25] after a 10-to 12-h overnight fasting (T = 0) and 15 min, 30 min, 45 min, 60 min, 90 min, and 120 min after starting to eat the food reference or test foods.

In vivo study
For obtaining data for the project "Glycemic response of volunteers to the consumption of supplements and food formulas for oral and/or enteral nutrition," trials with volunteers were carried out weekly at the Food Research Center (FoRC)/Faculty of Pharmaceutical Sciences of the University of São Paulo for 6 weeks, with weekly attendance and a stay of 2 to 3 h. In the three initial weeks of screening, doses of glucose solution were consumed (280 mL of water with 25 g of glucose), and in the subsequent weeks, the supplements, objects of this study, were consumed. A similar volume of all products was consumed (~ 300 mL) in a manner ensuring that one serving provided 25 g of available carbohydrates according to the results of chemical analyses.

Basic criteria for selection
Volunteers were aged between 18 and 45 years old of both genders and in good health condition according to a report indicating the absence of diabetes, kidney and gastrointestinal diseases, hyperthyroidism, abnormally low or high weight, and family history of diabetes. Female volunteers were not pregnant, breastfeeding, menopausal, or receiving hormone therapy. All volunteers also confirmed that they were not taking any medication that could affect the digestion and absorption of food (antibiotics or medication for diarrhea and constipation) during the study period.
After screening with the consumption of the reference food (glucose), it was confirmed that all volunteers had a normal blood glucose concentration, 70-99 mg/100 mL in the morning (after fasting for 10 h), a postprandial state with a maximum blood glucose of 140 mg/100 mL, and 2 h after the consumption the glycemia had value similar to fasting.
This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of the Faculty of Pharmaceutical Sciences (Approval CEP/FCF 2.814.784). On the first day of the clinical trial, all eighteen volunteers signed a free and informed consent. It was used data from 16 participants (10 women and 6 men) with a mean age of 33 ± 6 years, mean weight of 69.4 ± 17.1 kg, and mean height of 1.69 ± 0.1 m. Data from two volunteers were not included, as their glycemic responses were extremely inconsistent when they consumed two and three test food, which did not occur during screening.

Calculation of glycemic response, glycemic index, and glycemic load
The glycemic curve was plotted using blood glucose data obtained at times 0 (fasting) min, 15 min, 30 min, 45 min, 60 min, 90 min, and 120 min after the ingestion of food. The incremental area under the curve (AUC) was calculated geometrically using the trapezoidal rule and ignoring the areas below the fasting line [4,26].
To determine GI, the volunteers consumed similar portions of all products: 297 mL of Cubitan® (liquid product, consumed in full), 97 g of Diasip® vanilla, and 93 g of Diasip® chocolate. The Diasip® products were diluted in 300 mL of water and contained 25 g of available carbohydrates, similar to the reference food (glucose).
GI of each product was calculated using the equation: GI = (glycemic AUC of test food/glycemic AUC of glucose) × 100 [4].
The mean GI of each food was calculated from the AUC values of each individual, taking the area for glucose as a reference (GI = 100).
To determine the GL of each food, we used the equation GL = (GI of food × grams of available carbohydrate per serving)/100 [27].

Statistical analysis
Assumptions of normality and homogeneity were investigated by the Shapiro-Wilk test (n = 16) and Levene test for each product's outcome. Only the AUC data showed no normal distribution and were evaluated by the Friedman Test for paired non-parametric measures. The other quantitative data are presented as mean and standard error, as they present a profile of normality. StatSoft's Statistica 12 software was used to perform an analysis of variance (ANOVA for repeated measures) for comparisons between three groups with the Bonferroni post hoc test used to determine significant differences between the groups. Values of p < 0.05 were considered significant.

Results
The available carbohydrates of the three products studied are shown in Table 1. No resistant starch or maltodextrin was found.
The clinical trial showed that the samples studied had a low GI (Table 2) with a significant difference from the reference food, but no significant difference was observed between the products. Considering the recommended serving portion in each product label information, the three products also had a low GL, with values between 6 and 7 ( Table 2).
In Fig. 1, it can be seen that the AUC of the products is less than half of that for the reference food. Figure 2 shows the average glycemic response of the three products, where a blood glucose peak can be observed 30 min after the end of intake, with a statistically significant difference in relation to the reference food.

Discussion
The three products evaluated in the present study had a low content of available carbohydrates (from 14.9 to 15.8 g in a 200-mL portion) and a low GI, GC, and glycemic response. Thus, they may be indicated for inpatients or outpatients to promote nutritional balance and glycemic control. The value of GI = 35 for Cubitan® is the same as that in a European study (London, 2006) conducted previously by the  [30]. There is a difference between the macronutrient compositions of the liquid and powder versions of Diasip®, such as a lower lipid and higher protein and fiber contents in the powder presentation may justify the differences observed in the GI. The sources and proportions of carbohydrates also differ, comprising maltodextrin (75%), isomaltulose (21%), and lactose (4%) in the powder presentation and isomaltulose (38%), lactose (31%), and tapioca starch (27%) in the liquid presentation. In this study, Cubitan® had the lowest impact on the glycemic response, with a very flat curve (Fig. 2) and GI = 35, despite containing 9.9 g of sucrose per serving. These findings demonstrate that the consumption of this product should not impact blood glucose or cause episodes of hyperglycemia.
Fourteen articles were included in a systematic review that aimed to evaluate the effectiveness of diabetes-specific formulas compared to standard formulas. The results suggested that diabetes formulas are effective in controlling fasting blood glucose, glycated hemoglobin, and the increase in HDL cholesterol; some studies also revealed positive outcomes for postprandial blood glucose values. The authors considered that the presence of low GI carbohydrates, relatively low amounts of carbohydrates, relatively high amounts of protein, the presence of monounsaturated fatty acids, and different amounts and types of fiber in the diabetes-specific formulas may be responsible for the positive changes in cardiometabolic parameters observed compared to the standard formulas [31].
A study evaluated the impact of oral dietary supplements with a low proportion of available carbohydrates, isomaltulose, and fortified with soluble dietary fiber on blood glucose levels after ingestion in 20 individuals with prediabetes. It was observed a significant reduction in AUC compared with control with a correspondent reduction of insulin response. The oral supplement tested showed significantly lower blood glucose levels than the standard oral dietary supplement at all postprandial time points, as it was observed in the present study [32].
A review article involving patients with T2DM included 10 randomized controlled trials, according to the inclusion criteria, and demonstrated that most of these clinical trials resulted in significant improvement in postprandial hyperglycemia when low GI meals are consumed. Low GI meals were those that were low in carbohydrates or high in soluble fiber. Both provided favorable postprandial glucose excursions. The study's main recommendation for clinical practice is "Reduce the glycemic index (GI) of each meal by replacing high GI foods with low GI foods" [33]. In this regard, low GI oral supplements offer a very safe option.
Diasip® oral and/or enteral feeding formula was evaluated by Lansink et al. (2011) in a randomized, doubleblind, and controlled trial with a duration of four weeks that involved 44 type 2 diabetes patients who used oral hypoglycemic agents. The case group received the liquid version of Diasip® (2 units/day), and the control group received an equal volume of isocaloric formula with dietary fiber. Plasma glucose levels were assessed after formula consumption throughout the follow-up period. The researchers found that plasma glucose levels were significantly lower in the group receiving Diasip® than in the control group [34]. A double-blind, randomized, and controlled trial was Fig. 2 Mean glycemic response of volunteers (n = 16) to the intake of 25 g of available carbohydrates for 120 min after the intake of reference food and products. *Significant difference for all products from that of glucose (p < 0.05); # significant difference for all products from that of glucose, except for Diasip® chocolate; ## significant difference between Cubitan and glucose (ANOVA with Bonferroni post hoc, p < 0.05) also performed in Brazil with 40 patients, in which the case group received two units a day of liquid Diasip®, complementing their usual diet, and the control group received an isocaloric and fiber-free product. The glycemic response and secondary parameters were measured at T = 0 and after 6 weeks and 12 weeks. Patients who received Diasip® had lower plasma glucose levels in all the evaluations performed [35].
Persistent glycemic imbalance leads to oxidative stress and endothelial dysfunction, which are precursor events of tissue damage and determinants for the development of micro-and macrovascular complications. Diabetes alters the structure of the cell membrane, modifying the inflammatory response and inhibiting the vascularization process, which favors the development of wounds, especially lower limb injuries [36]. Furthermore, because of the decrease in metabolism and nutrient delivery, as well as inadequate concentrations of growth factors and cellular abnormalities, diabetes causes continuous trauma to the injured area [37].
In addition to diabetic patients, patients with malnutrition, infection, elderly, poor mobility, and other conditions have wound healing problems, making them susceptible to pressure injuries. Patients with both pressure injuries and other wounds require a careful diet that must be rich in energy, protein, arginine, zinc, and vitamins A, C, and E, which can be supplied with the use of specific food supplements [16,22].
In the multicenter Oligo Element Sore Trial carried out with 138 malnourished patients with a mean age of 81 years, 17 of them had diabetes; it was observed that the use of a specific formula developed for wound healing that was enriched with arginine, zinc, and antioxidants (Cubitan®) resulted in improved wound healing and lower non-nutritional care costs, including antibiotic and professional care costs [38]. In another study, elderly people living in long-term-care residences in Europe (n = 245, mean age of 82 years, 40 of them with diabetes mellitus) with grade II-IV pressure injuries were followed for 9 weeks of treatment, during which they received three daily units of Cubi-tan® (250 kcal, 20 g protein, 3 g arginine, 250 mg vitamin C, 38 mg vitamin E, and 9 mg zinc). After the intervention with the supplement, the lesions showed a 53% reduction in the affected area and a significant reduction in exudation [39].
In addition to the consumption of a balanced diet that meets nutritional needs, hyperglycemia must be avoided in patients diagnosed with diabetes and those without diabetes. Therefore, the offer of nutritional supplements that support glycemic control or wound healing without the risk of promoting a high glycemic response is desirable.
It is needed more studies involving metabolic responses from supplements and food formulas since several products are currently being commercialized.

Conclusion
The oral supplement Diasip®, vanilla and chocolate flavors, and the formula Cubitan® for oral and/or enteral feeding induce a reduced glycemic response, with flattened blood glucose peaks, a low GI (47, 49, and 35, respectively), and low GL (between 6 and 7). These greatly low values indicate that these products are suitable for avoiding glycemic peaks. Data availability Datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request and with authorization from the sponsor company.

Declarations
Ethics approval and consent to participate Approval CEP/FCF 2.814.784 was granted by the Ethics Committee of the Faculty of Pharmaceutical Sciences.