Texture Analysis of Food Samples Used for the Evaluation of Masticatory Function

Introduction Food questionnaire surveys are often used to evaluate masticatory function. In daily clinical practice in Japan, a survey is performed using a list of food groups suitable for the Japanese diet. The foods on the list were categorized into five food groups based on their mastication index. The patient's masticatory function is determined by the food groups that can be eaten. The masticatory index, which indicates chewability, was defined based on the percentage of 110 denture wearers who responded that they could eat food normally. A survey with this list is useful because of its simplicity; however, there is a lack of objective data on the physical properties of food samples. Consequently, to make the results of the food questionnaire survey more objective indicators, we performed a texture analysis of the food samples on the list. Methods We performed a texture analysis of 93 samples from 77 food items on the list. Compression tests were performed using a texture analyzer, and hardness, cohesiveness, adhesiveness, viscosity, and gumminess were calculated by a texture profile analysis. Results Even with the same ingredients, the results differed depending on the presence or absence of food skin, the direction of pressing (vertical or horizontal), cooking methods, and temperature differences. However, the masticatory index was negatively correlated with hardness (-0.4157, p<0.001) and gumminess which is determined as the product of hardness×cohesiveness (-0.4980, p<0.001). Conclusion This study suggests that the masticatory index indicating chewability may be related to the hardness and cohesiveness of food samples. Even for foods with the same hardness, the degree of difficulty in forming a food mass is expected to vary depending on differences in cohesiveness. Moreover, the presence or absence of food skin, the direction of food fibers, cooking methods, and temperature differences change the physical properties of the food. Therefore, the composition and structure of the foods or eating habits of patients should be taken into consideration when conducting a food questionnaire survey.


Introduction
Mastication, to crush and mix food fragments, requires an appropriate occlusal relationship between the upper and lower jaws, a fitted tooth crown shape, and an accurate movement of the tongue with the buccal mucosa and the lips, which cooperates with jaw movements.In addition, saliva also plays an important role in bolus formation.These oral functions are greatly impaired by the loss of teeth, jawbone, and oral soft tissues, and most patients with oral tumors have severe trouble with mastication and bolus formation.In particular, segmental mandibulectomy causes anatomical distortion resulting in a functional deficit [1].Radiotherapy to the oral cavity also reduces the secretory capacity of the salivary glands.These changes in oral function must be appropriately evaluated in clinical practice.Direct and indirect methods are used to evaluate masticatory function [2].Indirect examination methods evaluate mastication by judging jaw movement, electromyography, the occlusal contact state, and occlusal force.In contrast, direct examination involves the subject actually chewing a masticatory sample (e.g., gum or gummy jelly) [3][4][5] and methods based on a food questionnaire survey.
Gum and gummy jelly are very useful because they can be objectively evaluated, but they are too hard to evaluate in patients with significant deterioration in the oral function.In fact, when the masticatory function after mandibular resection was measured with gummy jelly, most patients scored ≤2 out of a maximum score of 9 [4].This was because the gummy jelly was so hard that the patient could not bite it off in multiple small pieces.Therefore, the test results may have underestimated the actual masticatory function of patients with oral dysfunction.On the contrary, even in the same patients, different results were obtained from a food questionnaire survey using the masticatory function evaluation table reported by Sato et al. which is often used in daily clinical practice in Japan [3,4,[6][7][8].The questionnaire with this table made it possible to evaluate the chewability of foods at levels 2, 3, and 4 of the five categories using the masticatory index described below.At levels 2, 3, and 4, there were foods that were neither too soft nor too hard, even in patients with oral dysfunction [8].Ingredients that were equivalent to level 2 are, for example, sausages and hamburgers, etc., level 3 are bacon and bamboo shoots, etc., and level 4 are pickled radish and fresh abalone, etc.Therefore, the table by Sato et al. may be useful in evaluating the function of chewing such moderately hard foods by patients with oral dysfunction.On the other hand, there is a lack of objective data on the physical properties of the food samples in the table.In this table, the masticatory index is used as a measure of masticatory function; however, it was determined based on the percentage of 110 denture wearers who responded that they could eat the food normally in the study of Sato et al.The masticatory index is useful because of its simplicity, but it is not supported by objective data on the physical properties of foods.Consequently, to make the results of the food questionnaire survey more objective indicators, we performed a texture analysis of the food samples listed on the questionnaires to examine their physical properties.

Samples
Of the 100 items listed in the masticatory function evaluation table for complete dentures [6], we first excluded non-food items (i.e., cotton thread, mandarin orange bags, etc.).Moreover, we deleted food items that are not compatible with modern Japanese eating styles, such as okoshi (a type of Japanese cracker) or amanatto (a type of Japanese sweet), while we added new food items that suit our modern diet (e.g., avocado [9], croissant [10], etc.).Food items that were difficult to mold into the form for the measurement described below (e.g., peanuts, potato chips, lettuce, etc.) were also excluded.On the other hand, we used multiple types within the same category in some food items.For example, four types of gummy jelly and two types of jelly from different manufacturers were used for the measurement.It should be noted that Engelead jelly, which is a pharmaceutical, as defined by the Japan Consumer Affairs Agency, for patients with difficulty swallowing [11], was also employed.Additionally, two types of cheese (cream cheese and processed cheese) and two types of tofu (firm tofu and silken tofu) were included.In the same manner, baumkuchen and financier were included in the cakes.Finally, 77 food items that could be commercially available were used as samples (Table 1).Sample pieces were processed into cubes of 1.5×1.5×1.5 cm as a rule.If the thickness of the food is less than 1.5 cm, only the length and width were molded to 1.5×1.5 cm.Products that are difficult to mold, such as pudding, were measured as they were.Some food items were measured both raw and cooked.Some food items were measured both skinned and unskinned, such as grapes and pumpkins.Bread both with and without the crust was used.For some ingredients for which the lengths in the vertical and horizontal directions differed, measurements were taken both longitudinally and transversely.Finally, this study was conducted using 93 samples of 77 food items.

Cooking methods
Samples such as frozen foods were heated using a water oven, AX-GA1-W (Sharp Corporation, Japan).The water oven was set to the microwave function (power setting: 600 W).During heating, food samples were placed on a dessert plate (diameter: 19.5 cm) and lightly covered with a wrap film for food packaging, Dia Wrap Eco Pita! (Mitsubishi Plastics Co., Ltd., Japan).After heating, a stick thermometer, digital thermometer TT-533 (Tanita Co., Ltd., Japan), was used to insert the temperature part (tip 20 mm) into the sample, and the temperature was measured when the displayed temperature stabilized.For samples normally eaten at room temperature, the temperature was finally adjusted to 23±1°C.On the other hand, the actual temperature was considered when eating.For example, the temperature of dumplings and Hamburg steaks was set at 65°C, and the temperature of takoyaki was set at 80°C.
In the cases of grilling and boiling, we used an IH cooker, KZ-PH30P (Panasonic, Japan), with a medium heat setting of 700 W. A frying pan (diameter: 26 cm, depth: 4 cm) was used for grilling.Taking into account the thermal conductivity of the frying pan, the sample was added 30 seconds after the ignition of the IH cooker.The food samples were cooked according to general cooking methods.For example, bacon was heated over medium heat for 30 seconds on each side.In the case of boiling, 1 liter of distilled water was placed in a pot (diameter: 16 cm, depth: 7 cm), and the sample was placed in the pot after boiling.However, if there was a description of the cooking method on the package, it was followed.For example, tsukune was boiled in a pot containing 400 ml of water for 10 minutes.
Foods that were likely to lose their shape after cooking were molded into cubes of 1.5 cm after cooking, but the other samples were molded before cooking.For example, silken tofu, firm tofu, hanpen, and konnyaku were first cut into large pieces, then boiled, and finally shaped into 1.5 cm 2 cubes.The sizes of these samples before boiling were as follows: silken tofu (length 7.5 cm, width 7.5 cm, thickness 2.5 cm); firm tofu (length 8.5 cm, width 5.5 cm, thickness 3.2 cm); hanpen (length 11.0 cm, width 11.0 cm, thickness 1.5 cm); and konnyaku (length 13.5 cm, width 7.0 cm, thickness 2.0 cm).The following samples were molded before cooking, but the boiling time was varied according to the ease of boiling: eggplant (one minute); bacon (five minutes), carrot (30 minutes), radish (20 minutes), and potato (10 minutes).

Measurement of food texture
A compression test was performed using a texture analyzer, TA.XTplusC (Eiko Seiki Co., Ltd., Japan), and the following items (items 1-5) were calculated according to the texture profile analysis method [12]: (1) hardness (defined as the force necessary to achieve a given deformation), ( 2) cohesiveness (defined as the strength of the internal bonds that make up the body of the product), ( 3) adhesiveness (defined as the work necessary to overcome the attractive forces between the surface of the food and the surface of other materials with which the food comes in contact), ( 4) viscosity (defined as the flow rate per unit force), and ( 5) elasticity (defined as the rate at which a deformed material returns to its undeformed condition after the deforming force is removed).Additionally, gumminess (the energy required to disintegrate a semi-solid food product to a state ready for swallowing) was determined as the product of hardness×cohesiveness, and chewiness (the energy required to masticate a solid food product to a state ready for swallowing) was determined as the product of hardness×cohesiveness×elasticity.As shown in Figure 1, hardness is the height of the first peak, cohesiveness is the area of the second positive peak divided by the area of the first positive peak (A2/A1), adhesiveness is the area of the negative peak immediately after the first positive peak (A3), viscosity is the height of the negative peak, and elasticity is the time from start to peak of the second positive peak divided by the time from start to peak of the first positive peak (T2/T1).

FIGURE 1: Illustration of the texture profile analysis method.
Hardness is the height of the first peak.Cohesiveness is calculated by A2/A1.Adhesiveness is the area of A3.
Viscosity is the height of the negative peak.Elasticity is calculated by T2/T1.S: second; A1: the area of the first positive peak; A2: the area of the second positive peak; A3: the area of the negative peak immediately after the first positive peak; T1: the time from start to peak of the first positive peak; T2: the time from start to peak of the second positive peak Measurement conditions such as the plunger size and compression speed were specified in detail.A cylindrical probe (diameter: 20 mm) was used, assuming clenching of the molars.Compression was performed twice, and the compression speed was set to 2.00 mm/s.The distance the probe moved after contacting the sample was set to 5 mm.The measurement was repeated eight times per sample type, and the maximum and minimum data were deleted.The remaining six measurements were used to calculate the mean values and standard deviations for hardness, cohesiveness, adhesiveness, and elasticity.

Statistical analysis
All statistical analyses were performed using Excel Statistics, Version 3.2 (Social Information Service Co., Ltd., Tokyo, Japan), and the Spearman correlation coefficient (r) was calculated.P-values of <0.05 were considered to indicate a statistically significant difference.

Results
The hardness, cohesiveness, adhesiveness, and viscosity measurements are shown in Table 1.The maximum hardness was 4639.171g (sweet chestnut).This was followed by pickled radish, bamboo shoots, pear, green soybeans (frozen food), and Allium chinense (Chinese onion).In contrast, the minimum hardness was 24.621 g (the central part of untoasted sliced bread), and croissant and sliced bread with crust (untoasted) also had low hardness values.The average hardness value was 918.926 g.Regarding cohesiveness, the maximum value was 2.368 (skinned Kyoho grape), and skinned Pione grapes and daifuku (Japanese sweet) also had high values, while the minimum value was 0.349 (monaka (Japanese sweet)), and boiled May queen potato and boiled taro also had also low values.The average value of cohesiveness was 1.057.The maximum adhesiveness value was 205.569 (boiled radish) and the minimum was 0.004 (marshmallow, TOUGH® gummy, croissant, and untoasted sliced bread crusts).The average adhesiveness value was 15.644.The maximum viscosity was 175.243 (bananas) and the minimum was 0.158 (untoasted sliced bread crusts).The average viscosity value was 15.572.Elasticity values were approximately 1.0 for almost all food items, and there was no significant difference; therefore, the gumminess values (hardness×cohesiveness) and chewiness values (hardness×cohesiveness×elasticity) were almost the same.The maximum gumminess was 4798.488(boiled bamboo shoots), followed by pear and Allium chinense (Chinese onion).The minimum gumminess was 27.068 (central part of untoasted sliced bread), followed by pudding crust, croissant, and untoasted sliced bread.The average gumminess value was 931.156.
In this research, we occasionally found that even with the same ingredients, the results differed depending on the presence or absence of the skin, the direction of pressing (vertical or horizontal), the cooking methods, and the difference in temperature.For example, in the case of Kyoho grapes, differences were observed in all measured values, depending on the presence or absence of the skin.Skinless Kyoho grapes (with seeds) had the following values: hardness, 443.781; cohesiveness, 2.368; adhesiveness, 3.387; and viscosity, 4.371.In contrast, Kyoho grapes with skin (with seeds) had the following values: hardness, 761.227; cohesiveness, 1.029; adhesiveness, 0.057; and viscosity, 0.695.In the case of white bread, the crust remaining slice was harder, but there were no significant differences in other indicators.However, differences were only found in the hardness between the vertical and horizontal directions in which the samples had different internal fiber orientations.A typical example was pickled turnip.A vertical push produced the following values: hardness, 2178.565;cohesiveness, 1.429; adhesiveness, 4.148; and viscosity, 10.632.In contrast, a horizontal push produced the following values: hardness, 590.052; cohesiveness, 1.420; adhesiveness, 4.278; and viscosity, 9.632.Regarding cooking methods, many ingredients were found to be softer when boiled than when raw, but both silken and firm tofu became harder when boiled.In comparison to baking and boiling, baked bacon was harder than boiled bacon.In terms of the food temperature at the time of measurement, the adhesiveness and viscosity of the Hamburg steak were extremely high at higher temperatures, but the hardness remained almost the same.
In the table for evaluating the masticatory function for complete dentures [6], the masticatory index of the food was determined based on the percentage of 110 denture wearers who responded that they could eat the food normally.We investigated a Spearman correlation coefficient between this quantitative index and the data of this study.The correlation coefficients with the masticatory index were as follows: hardness, -0.4157 (p<0.001);cohesiveness, -0.2799; adhesiveness, 0.1572; and viscosity, -0.0082.The masticatory index showed a negative significant correlation with hardness (-0.4157, p<0.001) and gumminess which is determined as the product of hardness×cohesiveness (-0.4980, p<0.001) (Figure 2).

Discussion
From ingestion to swallowing, the processes of biting, crushing, mixing, and bolus formation of foods are necessary.However, these oral functions are greatly reduced in patients with oral cancer.One of the reasons is that surgical resection is the standard treatment for oral cancer [13].If the jawbone is surgically resected, hard reconstruction should be performed using a fibular flap or reconstruction plate; furthermore, a jaw prosthesis using dental implants may be needed.However, it has been shown that the improvement in masticatory function is limited in such case [3,4,8,14].Additionally, radiotherapy causes a dry mouth, which is one of the main oral function problems, since sufficient saliva is required for bolus formation.Consequently, it is necessary to accurately evaluate the oral functions, including the masticatory function, after treatment for oral cancer.Thus far, we have evaluated the oral function using gum, gummy jelly, and food questionnaires.Gum and gummy jelly are very reliable because they can be objectively evaluated, but they are too hard to evaluate in patients with significant deterioration in the oral function.In other words, the food questionnaire survey, whose items include foods with a variety of characteristics, including hardness, could allow a more realistic and detailed assessment of the masticatory function in patients with oral dysfunction.However, there is the possibility that the results of the food questionnaire survey may change depending on the subjectivity of the patient.In addition, objective data on the physical properties of food samples described in the questionnaires are lacking.Therefore, in this study, to perform a more objective evaluation, we analyzed the texture of the food samples of the questionnaire and examined their physical properties.
We found that the physical properties of individual foods differed depending on the presence or absence of food skin, the direction of food fibers, the cooking methods, and the temperature.For example, all the values of all measured items differed depending on the presence or absence of grape skin.The skin protects the fragile and soft pulp, and it is natural that the presence of skin would increase the hardness.In contrast, the skin itself reduced the values of cohesiveness, adhesiveness, and viscosity.It can be considered that the grape skin prevented the pieces of pulp from mixing and adhering to each other.Accordingly, the grape skin should be a suppressor of bolus formation.Regarding bread, each numerical value of texture measurement was different depending on the structure and material of the breads [15].In this study, when comparing breads with and without crust, the hardness of the bread with crust was more than twice as high, but the cohesiveness, adhesiveness, and viscosity were almost the same.Unlike grape skins, bread crusts probably do not have the effect of preventing food particles from mixing or sticking to each other.Differences in hardness were also observed according to the orientation of the internal fiber whether vertical or horizontal, even for the same food material.Foods with fibers, such as vegetable stems, may become harder if the chewing direction matches the direction of the fibers.A typical example of such cases in this study was pickled turnip; however, bolus formation after fine chewing is thought to be independent of the direction of the fiber, because there were no differences in cohesiveness, adhesiveness, and viscosity.Considering the cooking method, when comparing the methods of baking and boiling bacon, the former was found to be harder.This is thought to be due to a change in the amount of water contained in the ingredient.It was considered that baking rather than boiling tended to decrease the amount of moisture contained in the ingredients.In the case of tofu, both the silken and the firm tofu became slightly firmer after boiling.Tofu is made by hardening soy milk with calcium salt.It is known that heating promotes further bonding between soy protein and calcium ions, and as a result, the hardness should increase even after the protein has returned to room temperature.However, even when Hamburg steak was measured at a high temperature (65℃), the hardness did not change much in comparison to room temperature.In contrast, the adhesiveness and the viscosity of the Hamburg steak increased significantly at a high temperature.It is thought to be due to the oil becoming lubricated by heating and the crumbly texture of the ingredients being reduced.On the basis of these results, it is considered necessary to confirm the composition and structure of the foods or eating habits when conducting a questionnaire survey.
In this study, the masticatory index described in the complete denture masticatory function evaluation table by Sato et al. showed a significantly negative correlation with both hardness and gumminess, and gumminess resulted in a higher correlation coefficient value.Gumminess is defined as the energy required to break down a semi-solid food into fragments until it is ready to swallow [12], and its value is calculated by the formula of hardness×cohesiveness [16].Since the masticatory index by Sato et al. has been useful for evaluating masticatory function, cohesiveness is considered to be as important as hardness when evaluating masticatory function.The cohesiveness was shown to indicate the strength of internal bonds that make up the body of food and the degree to which a food can be deformed before it ruptures (breaks) [16].Yoshimine et al. demonstrated that cohesiveness should be evaluated as the index of the ability to dilute a bolus with saliva and also stated that multiple parameters should be measured for a comprehensive assessment of mastication [17].According to the results of this study, even with foods of almost the same hardness, for example, boiled eggplant (the masticatory index is 90 and hardness was 535.941 g) had high cohesiveness (1.938) and was likely to form a bolus, while fish sausage (the masticatory index is 67 and hardness was 440.972) had low cohesiveness (0.984) and would be difficult to become a bolus.It may be difficult to form a bolus from foods with low cohesiveness, especially for people with low saliva production.Engelead is a jelly product approved as a food for special dietary use, "Food for persons who have difficulty swallowing: approval standard I," as stipulated by the Consumer Affairs Agency [11].Among the food samples in this study, this jelly was the sixth softest (78.662 g) and had the 23rd highest cohesiveness (1.146).Therefore, from these results, it can be reasonably said that this jelly is suitable for people who have difficulty swallowing.To evaluate the actual mastication in detail, it is important to also investigate the cohesiveness of food samples in addition to their hardness.However, only 93 food samples were examined in this study, which is not a sufficient number.Further research is needed in the future to increase the number of subjects.

Conclusions
This study suggests that chewability may be related to the hardness and cohesiveness of food samples.Even for foods with the same hardness, the degree of difficulty in forming a food mass is expected to vary depending on differences in cohesiveness.Moreover, the presence or absence of food skin, the direction of food fibers, cooking methods, and temperature differences change the physical properties of the food.Therefore, the composition and structure of the foods or eating habits of patients should be taken into consideration when conducting a food questionnaire survey.

Additional Information
Author Contributions

FIGURE 2 :
FIGURE 2: Correlation between the masticatory index and each texture analysis result.