Effect of polydextrose and stevia on quality characteristics of low-calorie biscuits

Rana, M.S., Das, P.C., Yeasmin, F. and Islam, M.N. Department of Food Technology and Rural Industries, Bangladesh Agricultural University, Mymensingh 2202, Bangladesh Department of Agro Product Processing Technology, Jashore University of Science and Technology, Jashore 7408, Bangladesh Department of Chemical and Food Process Engineering, Rajshahi University of Engineering and Technology, Rajshahi 6204, Bangladesh


Introduction
Nowadays, food industries are focusing on consumer health issues. Besides, people are conscious about the foods which have higher fat, cholesterol and sugar because of the possibility of being affected by different diseases including coronary heart disease, fatness, hypertension, different cancers, gall bladder diseases etc. (Ding and Malik, 2008). Excess fat and sugar consumptions have adverse effects on human health and there is a clear link between excessive dietary fat consumption and risk of cardiovascular disease (Viana et al., 2003). Researches also illustrate linkages between highly glycemic diets and various types of cancer (Eyre et al., 2004;Bonovas et al., 2004;Ding and Malik, 2008). By concerning these alarming conditions, instead of eating high fat and sugar, people are so much interested in low-calorie foods (Akoh, 1995).
Biscuits are small cakes, prepared by wheat flour, fat, sugar and other ingredients by mixing, conditioning and passing between rollers to make a sheet before being baked . Consumer demand is changing for the quality of biscuits with taste, safety, convenience and nutrition (Masoodi and Bashir, 2012). Fats and sugar impart many functional, nutritional and sensory properties of bakery foods. Fats improve texture, mouthfeel, structure and lubricity in foods (Giese, 1996), whereas sugar provides volume, texture, color and sweetness (Zoulias et al., 2000).
Foods with low fat and sugar are often lack of palatability. Several natural and artificial components are generally used as fat replacers to substitute the amount of fat in food items (Aggarwal et al., 2016). Fat substitutes are food ingredients that can replace fat totally or partially with similar organoleptic properties in foods (Lim et al., 2010). Several carbohydrates, protein, or lipid -based fat replacers are being used individually or in combination (Mitchell, 1996;Akoh, 1998). Polydextrose, processed sugars, derivatives of starch, cellulose, hemicellulose, gums, etc. are generally used as carbohydrate -based fat replacer (Nonaka, 1997) helps in lowering calorie for similar weight swap 4 kcal/g, instead of 9 kcal/g (Lindsay, 1996).
Stevia, leaves of Stevia rebaudiana, has been getting scientific interest to replace sugar (Michaud et al., 2002;Bell and Sears, 2003). Steviol glycosides are the active compounds of stevia and are around 150 -200 times sweeter than table sugar (Romieu et al., 2004). It is heat, pH stable and non-fermentable (Franceschi et al., 2001). As the body is unable to metabolize the glycosides of stevia, it provides no calories at all. Utilization of zerocalorie sweeteners in bakery goods may have considerable effects on tenderness, color and flavor of the final products (Mariotti and Alamprese, 2012). Artificial sweetener does not take part in Maillard reaction and caramelization resulting in lighter color in food products after cooking (Gallagher et al., 2003;Lin et al., 2010).
Few researchers tried to develop biscuits by replacing fat with different fat replacers such as maltodextrin, dairytrim, pectin etc. (Zoulias et al., 2000), and also by their combination with or without other components like corn fiber, lupine extract, guar gum etc. (Forker et al., 2012). Application of polydextrose and stevia as a source of fat replacement in biscuits is yet to be studied. However, using of polydextrose and stevia to replace fat and sugar in biscuits will be an important area to be investigated. Considering above standpoints, this research was conducted under the following objectives: (a) to prepared low-calorie biscuits by using of polydextrose and stevia as fat and sugar replacers and (b) to evaluate the physicochemical and sensory features of the developed biscuits.

Formulation and preparation of biscuits
The biscuits were made using the process by Sarker et al. (2013) and Das et al. (2018) with slight modifications. The formulation was done by the following experimental design as given in Table 1. Firstly, the daldah (fat) was finely mashed. Blended sugar was then added to it. Thereafter, other ingredients including egg, salt, polydextrose, stevia solution (2.5% solution), milk powder and vanilla essence were incorporated and well blended. The flour and baking powder were then added and combined well into a smooth dough. After that, the dough was shaped into thin, 3 mm thick consistent layer and cut out using a 3 cm diameter circular biscuit cutter. The biscuits were then baked in a baking oven (Sharp Electric Oven EO -257 CT-BK) at 180°C for 15 mins. Finally, the produced biscuits were cooled at ambient temperature and enclosed in airtight packets.

Nutritional analysis of developed biscuits
Proximate nutritional parameters such as moisture, ash, protein and fat content of the developed biscuits were analyzed on the basis of the method as narrated by AOAC (2012). Carbohydrate content was calculated by allowing the subtraction techniques mentioned by Pearson (1976) as carbohydrate = 100 -(protein + fat + ash + moisture content). Amount of energy for the consumption of 100 g biscuits was calculated by considering the total energy given by per g protein, fat and carbohydrate (Okoye, 1992). Each analysis was performed in triplicate.

Physical analysis
Physical parameters of biscuits were analyzed by AACC (2000) methods. Spread -ratio was determined by dividing the average diameter value by the average thickness value. Digital weighing balance was used to measure the weight (g). Thickness (cm) and diameter (cm) were measured by slide caliper. Volume and density were calculated using thickness, diameter and mass of the biscuits. A Stable Micro Systems Texture Analyser TAXT2i (Texture Technologies Corp, England) was used to analyze biscuit texture by following Oliveira (2013).

Sensory analysis
Sensory parameters like color, taste, texture and overall acceptability were evaluated by following the FULL PAPER method described by Ranganna (2005). A semi -trained jury of 15 panelists assessed sixteen biscuit samples containing various proportions of ingredients for their sensory attributes. The 9-point hedonic rating test (Amerine et al., 1965;Sarker et al., 2013;Begum et al., 2018;Marufa et al., 2019) was used to assess the extent of acceptance at a significance level of 5% (p≤0.05).

Statistical analysis
Single factor Analysis of Variance (ANOVA) and Fisher's Least Significant Difference (LSD) Multiple Comparison Test procedures of the Method of Statistical (MSTAT) system was conducted to find the significant variation among biscuit samples for different quality parameters by following Gomez and Gomez (1984) and Das et al. (2019).

Moisture
The moisture content (Table 2) of the developed biscuit samples ranged as 4.36 -4.99% in wb and 4.56 -5.25% in db. The moisture content of control biscuit (P) was 4.69% (wb) and 4.92% (db). Moisture content was decreased slightly with the fat and sugar reduction, while the addition of polydextrose results increasing of moisture. This is happened due to higher water holding capacity of polydextrose because polymers such as CMC, xanthan gum, maltodextrin, polydextrose etc. have high water empathy and can maintain foods' moisture (Nonaka, 1997). The moisture content of sample C was the highest (4.99% in wb and 5.25% in db) than other samples whilst the lowest moisture content was observed for sample M (4.36% in wb and 4.56% in db). The moisture content of the developed biscuits was closely related to that reported by Grah et al. (2014) as 5.13 -7.17% (db), Hussein et al. (2011) as 3.80 -4.62% (db), and Das et al. (2018) as 4.35 -4.91% (wb) and 4.55 -5.16% (db). Hussain and Kaul (2018) reported comparatively lower moisture content in biscuits as 2.56 -3.42% (wb), whereas Radhika et al. (2019) found even higher moisture content in the range of 9.30 -11.60% (wb). Aggarwal et al. (2016) studied the production of low-calorie biscuits using artificial sweeteners and milkmultigrain fat substitutes and found 4.30% (wb) moisture for control biscuit and 4.63% (wb) for optimized lowcalorie biscuits. The variation in the moisture content of this study with that of other authors might be due to change of recipe, baking time and temperature etc.

Protein
Different biscuits samples were analyzed for protein content and values were found in the range of 7.81 -8.91%. Protein content was increased significantly (p≤0.05) with the reduction of fat and sugar, while decreased significantly (p≤0.05) with the addition of polydextrose and stevia since wheat flour contains higher protein content. Control biscuit has the lowest (7.81%) protein, while sample M had the highest protein (8.91%). Grah et al. (2014) recorded a protein content of 6.88 -11.45% in biscuits, whereas Kabirullah et al. (1995) reported 6.88 -11.78%. The lower protein content of 6.60% in biscuit was reported by Seevaratnam et al. (2012). Hence, the protein contents in the prepared FULL PAPER biscuits are in agreement with the protein content of the biscuits reported by the other workers.

Fat
The fat content was in the range of 11.37 -17.88% (Table 2) in different biscuit samples. The analysis showed that fat content was the highest (17.88%) in sample P (control) and was lowest in the sample O (11.37%). Significant differences were noticed for the fat content of the biscuit samples at 5% level of significance. Using of polydextrose to reduce the amount of fat in recipe resulted significantly (p≤0.05) on the final fat content of the biscuits and attributed in lowering of maximum 36.41% fat than control. The differences in fat content among different samples was due to the difference in fat content among the sample's formulation. Lourencetti et al. (2013) found the fat content of inulin fortified reduced-fat biscuits in the range of 4.0 -14.2%. Kabirullah et al. (1995), reported that the fat content of biscuits ranges between 5.66 -26.67%, while Seevaratnam et al. (2012) measured 25.8%. Aggarwal et al. (2016) recorded 21.5% fat in control biscuit and 14.1% low-calorie, low -fat biscuits.

Ash
The biscuits samples showed the ash content with a range of 1.09 -1.24%. It was observed that the ash content of sample M was the highest (1.24%) and the sample P (control) gave the lowest ash content (1.09%) and differed significantly (p≤0.05) from all other samples. Ash content of the biscuits increased slightly with the reduction of fat and sugar but decreased with the increasing of the polydextrose and stevia level in the formulation. Kabirullah et al. (1996), analyzed biscuits and found ash content in the range of 1.09 to 2.78%. Lourencetti et al. (2013) found ash content of inulin fortified reduced -fat biscuits in the range 1.54 -4.18%. Accordingly, it is concluded that the ash content of developed biscuits was closer to the ranges as mentioned above.

Total carbohydrate
The amount of carbohydrate in the developed biscuits was ranged as 68.65 -73.72% (Table 2). Kabirullah et al. (1995) reported that the total carbohydrate content of biscuits in the range of 64.31 -83.45%. Total carbohydrate content was the highest (73.72%) in the case of sample B, while sample P (control) gave the lowest carbohydrate content (68.65%). The difference in the total carbohydrate content of the sample of biscuits can result from the different protein, fat, ash and moisture content levels.  lowering of the total calorie content. Control sample P had the highest amount of energy as 479.16 cal/100 g biscuit, while the sample M attributed with the minimum energy content of 418.84 cal/100 g. That is, using 3 g polydextrose in combination with 5 mL 2.5% stevia solution indicated the best formulation as low-calorie biscuits, and resulted in lowering around 12.59% calorie value. Energy value of the processed low-calorie biscuit with the usage of sugar substitutes and milk -multigrain fat replacer was recorded by Aggarwal et al. (2016) as 485.6 cal/100 g biscuit, which was around 7.59% lower than the energy value of their processed control biscuits. However, the total energy content of this study was in conformity with the energy values reported by Das et al. (2018) as 465.28 -466.82 cal/100 g.

Physical properties of the biscuits
The results of physical properties analysis are presented in Table 3. The diameter and thickness of the biscuits decreased significantly (p≤0.05) with the decreasing of fat and sugar content, but the addition of the polydextrose and stevia resulted in a slight increase of diameter and thickness. The highest diameter was found in case of sample P (3.9 cm) followed by sample C (3.85 cm) and sample F (3.83 cm). The lowest diameter was represented by 50% fat and sugar reduction sample, M (3.57 cm). Thickness was highest in case of sample P (0.7 cm) followed by sample C (0.67 cm) and sample B (0.66 cm). Sample M showed the lowest thickness value (0.5 cm). A similar type of result was found by Lourencetti et al. (2013) in the formulations of cookies with 25%, 50% and 75% reduced -fat by inulin, whereas Zoulias et al. (2000) reported relatively smaller diameter and thickness in the biscuits with sugar and fat replaced by inulin and maltodextrin than the standard formulation. Results of this study for diameter and thickness were more or less similar to that reported by Hussain and Kaul (2018) and almost complied with Das et al. (2018) as the results were in the range of the values reported by these authors. This reduction in dimension can be correlated with the elastic recoil capability provided by carbohydrate -based fat replacer. Spread ratio is considered as one of the most important parameters of biscuit consistency, significantly (p≤0.05) influenced by the reduction of fat and sugar. Spread ratio was increased with the reduction of fat and sugar, while in addition of polydextrose and stevia, there was a slight decrease in spread ratio. Spread ratio was the highest (7.1) in case of sample M followed by the sample N (7.03) and sample O (7.0), while the lowest value (5.58) was given by sample P (control). Spread ratio recorded by Das et al. (2018) as 4.66 -7.24 for composite biscuits complied with the values of this study, whilst Radhika et al. (2019) reported even higher spread ratio of 8.84 -10.22. Probably polydextrose and stevia effects resulted in increasing of spread ratio. However, the difference in the size of mold and baking conditions may be responsible for the difference of spread ratio, thickness and diameter values with that reported by other authors.
Mass of the biscuit samples was decreased with the reduction of fat and sugar, while further increased with the increasing of the polydextrose level since fat, sugar and polydextrose has higher density compared to wheat flour. Mass was highest in case of the sample C (4.2 g), FULL PAPER Sample Diameter (cm) Thickness (cm) Spread ratio Mass (g) Volume (cc) Density (g/cc) Compression force (  followed by sample B and F (4.12 g). The lowest value was found in the case of sample M (3.63 g). The highest volume was in the case of sample P (8.34 cc) followed by sample C (7.98 cc) and sample B (7.51 cc). The lowest volume was represented by the sample M (5.03 cc). The volume of the biscuits decreased with the decreasing of fat and sugar, but the addition of the polydextrose resulted in a further slight increase in volume.
Density is the function of mass and volume and was increased with the reduction of fat and sugar, while in addition of polydextrose density decreased slightly. The highest value was found in case of sample O (0.74 g/cc) followed by sample M and N (0.72 g/cc). The lowest value was shown by sample P (0.49 g/cc). The values of density were in conformity with the reported values of Das et al. (2018).
The highest compression value is associated with a higher percentage of fat and sugar replacement since fat shorts the gluten activity. The compressive value was found in the range 9.38 -16.44 kgf where sample M and sample P shows the highest (16.44 kgf) and the lowest (9.38 kgf) value respectively. Lourencetti et al. (2013) found the compressive force of the inulin fortified reduced -fat biscuits in the range 12. 58 -18.55 (kgf). The values of compressive strength also complied with Forker et al. (2012). Significant (p≤0.05) increasing of compressive values represent the effectiveness of polydextrose and stevia to increase the toughness of bakery products.

Sensory evaluation
The scores obtained from sensory analysis for control, and polydextrose and stevia supplemented lowcalorie biscuits are presented in Table 4. It is seen that there were significant differences in color, flavor, texture and overall acceptability preference among the samples at 5% level of significance. Concerning color preference, the score of biscuits was decreasing with the reduction of fat and sugar, while in addition of polydextrose and stevia resulted further enhance of color scores. The highest color score was obtained by the sample P (8.20) and the lowest score was found for the sample M (5.20). Among the biscuits with low-calorie formulation, sample C which was formulated by reducing 10% fat and sugar, and by using 9 g polydextrose and 1 mL 2.5% stevia solution secured the highest score of 8.07, also it was equally acceptable to control sample P. In case of flavor, the control biscuit sample P showed the highest score (8.47) while sample M showed the lowest score (5.47). The flavor of the biscuit was decreased with the decrease in fat and sugar. At the same time, in addition to polydextrose and stevia further improves the flavor of the biscuits. Texture, an important quality feature of biscuit which directly affects the acceptance and sales. The composition and texture of the biscuits are substantially directly related. The texture of the prepared biscuits was degrading due to the reduction of the fat and sugar quantity in the formulation, which subsequently resulted in decreasing the scores provided by the taste panelist. But, with the addition of polydextrose and  stevia level resulted in improvement of the biscuit texture as found as for color and flavor also. The highest and the lowest scores were secured by the control sample P (8.07) and the sample M (5.33), respectively. The overall acceptability of the biscuits was justified on the basis of the scores of the color, flavor and texture where the control sample P and the sample M showed the highest (8.13) and the lowest (5.4) scores respectively. Sample C having 9 g polydextrose and 1 mL 2.5% stevia solution as formulated for 10% fat and sugar reduction was the best among the developed biscuits rather than control. However, all of the biscuits samples got a score higher than 5 (neither like nor dislike). A more or less similar type of result was also noted by Lourencetti et al. (2013), where biscuits with 50% reduced -fat supplemented with inulin obtained highest preference to the panelist among low-calorie biscuits. Aggarwal et al. (2016) also noticed that the sensory values decrease in the low-calorie biscuits. Based on sensory properties, it can be said that 10% fat and sugar reduction from the recipe of the biscuits would be equally acceptable to the panelists, though up to 50% is acceptable as sensory scores were above the preferable value.

Conclusion
From the study, it is apparent that polydextrose and stevia can be considered as potential and effective replacer of fat and sugar respectively in biscuit preparation. Significant changes in physical and chemical properties were noticed due to the reduction of fat and sugar amount in the initial formulation. In addition, polydextrose containing biscuits were more preferable to the taste panelists, also showed better physical and chemical characteristics. However, further research is recommended in this area, specifically on the development and optimization of low fat and low sugar processed food with more combination of polydextrose and stevia.