2023 年 29 巻 3 号 p. 247-256
Espuma was examined for its suitability in meals for older adults with masticatory disturbance or dysphagia. Espuma is foam-style food prepared from pureed ingredients mixed with emulsifying agents and combined with nitrous oxide gas. Three whipping creams were used to prepare espuma: milk fat, vegetable fat, and soy milk cream. Boiled salmon was mixed with each cream to create the espuma foods and the resultant mixtures were examined. The soy milk cream espuma sample did not show syneresis, indicating that it was extremely suitable as a coagulating agent for espuma food. The texture of the boiled salmon in soy milk cream satisfied the Consumer Affairs Agency standard for people with dysphagia. Endoscopic examination of swallowing revealed little to no laryngeal residue. The results showed that boiled salmon in soy milk cream prepared as espuma food is suitable for use in meals for older adults with masticatory disturbance or dysphagia.
Hyper-aging societies are becoming increasingly common in developed nations (Kudo et al., 2015). In older adults, continuing to eat delicious meals safely and with peace of mind helps to maintain, or even improve, quality of life and leads to a longer healthy life expectancy. Extending the healthy life expectancy of older adults is becoming increasingly important to enable them to live independently for as long as possible i) (Tominaga, 2016). Many countries are experiencing demographic shifts toward aging populations, which highlights the importance of understanding the relationship between diet and quality of life in older adults (Govindaraju et al., 2018). The primary cause of protein-energy malnutrition that is often seen in older adults in nursing homes is due mostly to degradation or dysfunction of their ability to chew or swallow. Dysphagia is common among nursing home residents and is associated with increased mortality (Peladic et al., 2019). Thus, the relationship between dysphagia and malnutrition is strong (Brody, 1999). It is therefore necessary to prepare meals that correspond to the physical status of older adults with masticatory disturbance or dysphagia so that they can continue oral intake of food. It is desirable to develop foods, cooking methods, and meal preparation methods that facilitate easy chewing and swallowing in older people with masticatory disturbance (Egawa et al., 2007; Tominaga, 2016).
The suitability of espuma food in meals for older adults with masticatory disturbance or dysphagia has been examined. Espuma is a foam-style food prepared from pureed ingredients mixed with foaming or stabilizing ingredients (e.g., cream, gelatin, thickening agents, and albumen) and combined with nitrous oxide gas (Komagome et al., 2019). Usually, milk fat cream is used as the coagulating agent (Stankov et al., 2020). Espuma food created using this technique does not require chewing and quickly melts in the mouth. In addition, the flavor of espuma food is well preserved (Angeloni et al., 2020). For these reasons, espuma food is considered suitable for use in meals for individuals who have difficulty swallowing, including older adults and infants or children with disabilities.
Mousses and soft foods are widely used in meals for older adults, but espuma food is not. Some studies have reported on the preparation of natto (fermented soybeans) and bread pap (porridge) as espuma food (Touma and Yamamura, 2012; Komagome et al., 2019); however, studies on preparation methods or use of other food materials are scarce (Sato et al., 2022). The Spanish chef Ferran Adrià who invented espuma—also known as “culinary foam”—provides espuma food recipes freely to the public (Adrià, 2004). However, these recipes are for fruits, vegetables, and beverages; none are for fish or meat dishes. New espuma food recipes that suit the taste and food culture of Japanese people need to be formulated so that this food preparation technique can be used to make meals for Japanese people with swallowing difficulties. Therefore, this study aimed to develop an espuma dish made with fish.
Three types of whipping cream were tested at two temperature conditions to select the cream to be used in preparing boiled salmon in cream as espuma. Next, the food was evaluated in terms of density, texture, and amount of syneresis. Sensory evaluations were also conducted and the suitability of the sample in meals for older adults with masticatory disturbance or dysphagia was assessed.
The aim of this study was to contribute to improving the quality of life of older people who have difficulties swallowing food by providing them with espuma preparations of Japanese food.
Materials Three whipping creams were used to prepare espuma: milk fat cream (Special Selection Hokkaido Fresh Cream 35, Takanashi Milk Products Co., Ltd.; 1 432 kJ per 100 g, 2.1 g protein, 35.6 g fat, 3.2 g carbohydrate, and 35 % milk fat), vegetable fat cream (Whip Vegetable Fat Cream, Megmilk Snow Brand Co., Ltd.; 1 633 kJ per 100 g, 1.5 g protein, 41.6 g fat, and 2.4 g carbohydrate), and soy milk cream (Ko-cream Whip, Fuji Oil Co., Ltd.; 1 516 kJ per 100 g, 1.8 g protein, 37.1 g fat, and 4.8 g carbohydrate). Rice miso (Honba Sendai Miso, Sendai Miso Shoyu Co., Ltd.; 737 kJ per 100 g, 10.0 g protein, 6.2 g fat, 20.8 g carbohydrate, and 5038 mg sodium) was used as seasoning.
Boiled salmon in cream was prepared as espuma, using fresh salmon from Hokkaido and the abovementioned creams as coagulating agents. Hidaka kombu (seaweed) (Fujicco Co., Ltd.) was used to prepare kombu broth, which counteracts fishy odors (Shirota and Mineki, 2017). The preparation of kombu broth is described below.
Preparation of cream to be used for espuma Each whipping cream (100 g each, at 5 °C or 25 °C) was placed in an espuma dispenser (Espuma Advance TAH-CG, Toho Acetylene Co., Ltd.) with a 390-mL container. The dispenser was filled with nitrous oxide gas at constant pressure using the included gas charger (Randhawa and Bodenham, 2016). The dispenser was shaken manually 20 times and the contents were dispensed as espuma. The milk fat cream was labeled “A,” the vegetable fat cream “P,” and the soy milk cream “S”.
For seasoning, 3.1 g of miso (0.4 % of total salinity for each 100 g sample) was added to the espuma food samples. These samples were labeled “AM,” “PM,” and “SM,” respectively. The samples were also labeled to indicate temperature, with “C” indicating 5 °C and “R” indicating 25 °C.
Preparation of boiled salmon in cream as espuma The amounts of the prepared samples were set according to the results of preliminary experiments: 50 g of salmon, 90 g of coagulating agent, and 3.2 g of miso (0.3 % of total salinity per serving). In the preliminary experiments, 40, 50, and 60 g of salmon were used; at 40 g, the salmon flavor was considered weak, while at 60 g, the salmon fish meat did not adequately form a foam. Thus, 50 g was used.
Kombu broth was prepared by immersing Hidaka kombu in distilled water for 30 min at 25 °C and then boiling the water until it reached 90 °C. After removing the kombu, a freshly skinned salmon was boiled for 5 min in the kombu broth. Next, the boiled salmon was mixed with cream and miso for 1 min in a blender (TM8200, Tescom Co., Ltd.) set at 12 000 rpm. The mixture was filtered using a strainer and 100-g filtered samples were placed into an espuma dispenser. The dispenser was charged with nitrous oxide gas shaken manually 20 times, and the contents were dispensed as espuma.
Samples prepared with milk fat cream were excluded from the study because the foam did not hold its shape. Samples prepared using vegetable fat cream were labeled “X” and those with soy milk cream were labeled “Y.”
Appearance and nutrient contents of prepared samples Using a digital camera, photographs were taken of samples X and Y to document their appearance. The 15 g samples were placed in a glass container with a diameter of 6 cm. The amounts of energy, protein, fat, and carbohydrate in each prepared sample were calculated according to the tables of nutritional components and food compositions to calculate the weights per 418 kJ.
Density calculations The density [kg/m3] of each sample was calculated. A stainless steel bowl with a diameter of 40 mm and a depth of 15 mm was filled with each sample. The surface was leveled using a spatula. The sample was weighed and the density calculated using the following formula: density = sample weight / water weight (Kodama et al., 2016). Assuming an eating time of 30 min, the espuma samples were left to rest at room temperature (25 °C) and their densities were measured at 0, 10, 20, and 30 min after preparation.
Amount of syneresis The amount of syneresis [mL] of each foam sample was measured by pouring 30 g of the sample through a funnel with a diameter of 90 mm after the abovementioned resting times.
Texture of prepared samples Immediately after espuma preparation, each sample was placed into a stainless steel bowl with a diameter of 40 mm and a depth of 15 mm to determine whether it satisfied the texture standard for foods for people with dysphagia established by the Consumer Affairs Agency (CAA)ii) (Table 1). The top was leveled using a spatula and the texture was measured using a creepmeter (RE2-3305B-1, Yamaden Co., Ltd.) under measurement conditions of a 20 N load cell and a plunger cylinder with a diameter of 20 mm and a height of 8 mm. A pressure speed of 10 mm/sec was applied at two times the pressure load at 20 ± 2 °C. This method follows the CAA test procedure for evaluating food for people with masticatory disturbance or dysphagia (CAA, 2020).
| Standard | Level I | Level II | Level III |
|---|---|---|---|
| Hardness [N/m2] | 2.5×103–1×104 | 1×103–1.5×104 | 3×102–2×104 |
| Adhesiveness [J/m3] | Less than 4×102 | Less than 1×103 | Less than 1.5×103 |
| Cohesiveness | 0.2–0.6 | 0.2–0.9 | – |
Microstructure of prepared samples To examine the microstructure of each espuma sample, the samples were frozen using liquid nitrogen. They were immediately cut and placed on a sample table. A tabletop electron microscope (TM-3030Plus, Hitachi High-Tech Corporation) with the accelerating voltage set at 10 kV was used to observe the samples.
Sensory evaluation of prepared samples A series of analytic and palatability sensory evaluations of the prepared samples was conducted using the scoring method. Tests were conducted twice by 20 fourth-year students in the Department of Food and Nutrition at the authors' university as well as a panel of 10 older adults (1 man and 9 women; age 79.6 ± 4.9 years). The older adults did not reside in nursing homes and had no issues with swallowing and mastication functions when consuming daily meals. The sensory evaluation test was conducted under the approval of the Ethics Committee of Tokyo Kasei University (Approval number: H30-24). Each panel member was briefed on the purpose of the study, and written informed consent for their participation in the study was obtained.
Samples X and Y (20 g each) in the sensory evaluation conducted by the students were evaluated using a 5-point scale, where 1 = Weak, 2 = Slightly weak, 3 = Neither weak nor strong, 4 = Slightly strong, and 5 = Strong. Ten items were set for the evaluation targets: levels of color depth, salmon aroma, softness, smoothness of melting in the mouth, stickiness, smoothness in swallowing, rough sensation on the tongue, salmon flavor, sweetness, and saltiness. Different samples were used for the palatability sensory evaluation tests, for which both appearance and overall evaluation (deliciousness) were evaluated as 1: Unfavorable, 2: Slightly unfavorable, 3: Neither unfavorable nor favorable, 4: Slightly favorable, or 5: Favorable.
Sample Y (20 g) was served for the sensory evaluation tests conducted by the panel of older adults. The evaluation targets were reduced and grades at three levels were set because the gustatory sensitivity of older adults differs from that of younger adults. Also, the older adults who had been recruited to the panel were not familiar with these sensory evaluation tests (Kobayashi, 2015). The panel evaluated the sample using a 3-point scoring method to evaluate aroma and flavor of salmon, level of smoothness in swallowing, and smoothness in consuming. The grades were set as 1 = Weak, 2 = Neither weak nor strong, and 3 = Strong in the analytic sensory evaluation, and 1 = Unfavorable, 2 = Neither unfavorable nor favorable, and 3 = Favorable in the palatability sensory evaluation for evaluating deliciousness of the sample. In both tests, panelists were allowed to give open-ended responses.
Endoscopic examination of swallowing prepared samples Endoscopic examination was performed using a video nasopharyngo-laryngoscope (VNL8-J10, PENTAX Medical) while two older women (subject W [age 68 years] and subject Z [age 65 years]) swallowed sample Y from a spoon two times. For each test, 1.5 g of the sample was used, and the presence or absence of laryngeal residue after swallowing was observed.
Statistical analysis One-way analysis of variance was performed using SPSS ver. 24 (IBM Corp.) to analyze the measurement data. Tukey's test was used to compare differences among the three creams. A t-test was performed using Excel 2016 (Microsoft Corp.) to compare differences between samples X and Y. Tests were performed using 5 % and 1 % levels of significance.
Density of cream samples Fig. 1 shows the chronological changes in the foam density of the samples. Sample AR began to show a higher density than that of samples PR and SR from 10 min after preparation (p < 0.05). Sample AC also showed a higher density than that of sample SC (p < 0.05), whereas the density of sample PC was not significantly different from that of sample AC.
Density of samples A, P, and S using creams stored at 5 °C and 25 °C.
n = 9, a–d: six samples (AC, AR, PC, PR, SC, SR) showed significant differences for different letters at the same elapsed time (p < 0.05)
A comparison of foam density changes according to storage temperature for all samples revealed that samples AC and SC were less dense compared with samples AR and SR (p < 0.05). Sample AC showed a lower density compared with sample AR immediately after espuma preparation in group A (p < 0.05). There was no significant difference in the foam density of samples PC and PR in group P from 10 min after preparation (p < 0.05). As for foam density in group S, sample SC began to show a lower density compared with sample SR from 10 min after preparation (p < 0.05).
The density [× 103 kg/m3] of the miso-containing cream samples immediately after espuma preparation was 0.30 ± 0.01 in sample AMR, 0.28 ± 0.01 in sample PMR, and 0.30 ± 0.01 in sample SMR, compared with 0.27 ± 0.01 in sample AMC, 0.28 ± 0.01 in sample PMC, and 0.30 ± 0.01 in sample SMC. Thus, the density of the miso-containing cream was higher than any sample without miso (p < 0.05).
As for the influence of cream storage temperature, the density of samples AMC, PMC, and SMC was lower than that of samples AMR, PMR, and SMR immediately after espuma preparation (p < 0.05).
Amount of syneresis in cream samples Fig. 2 shows the chronological changes in the syneresis amount of the foam. Sample AR showed a larger amount of syneresis compared with samples PR and SR at the same elapsed time (p < 0.05). Syneresis was not observed in sample SR after 30 min of resting time.
Amounts of syneresis for samples A, P, and S using creams stored at 5 °C and 25 °C.
n = 9, a,b: six samples (AC, AR, PC, PR, SC, SR) showed significant differences for different letters at the same elapsed time (p < 0.05), measured with samples of 30 g each placed on the funnel.
For the influence of sample storage temperature on amount of syneresis, the results showed smaller amounts in samples AC and PC compared with samples AR and PR, respectively, at the same elapsed time (p < 0.05). Neither sample SR nor sample SC showed syneresis.
The amounts of syneresis in samples AMR and AMC were respectively 2.0 ± 0.2 and 1.5 ± 0.6 at 10 min after preparation, 5.0 ± 0.2 and 4.5 ± 0.4 at 20 min, and 5.4 ± 0.4 and 5.4 ± 0.6 at 30 min. At 20 min after preparation, smaller amounts of syneresis were observed in samples AMR and AMC compared with samples AR and AC, respectively (p < 0.05). Samples PMR, PMC, SMR, and SMC did not show syneresis after 30 min of resting time.
Texture of cream samples Table 2 shows the measurement results of the foam texture. The hardness of sample AR (0.94 [× 103 N/m3]) was higher than that of samples PR and SR. Cohesiveness was not observed to be significantly different among the cream samples. The adhesiveness of sample AR (4.51 [× 102 J/m3]) was higher than that of sample PR (p < 0.05), but was not significantly different from that of sample SR. The hardness of sample PC (0.51 [× 103 N/m3]) was lower than that of samples AC and SC (p < 0.05). In this temperature category, cohesiveness was not significantly different among the cream samples. Adhesiveness was in the order of SC > AC > PC.
| Sample | Hardness [× 103 N/m2] | Cohesiveness | Adhesiveness [× 102 J/m3] |
|---|---|---|---|
| AC | 1.06 ± 0.29c | 0.75 ± 0.04 | 4.07 ± 1.26d |
| AR | 0.94 ± 0.16c | 0.75 ± 0.01 | 4.51 ± 1.77bcd |
| AMC | 0.77 ± 0.15d | 0.68 ± 0.05 | 4.00 ± 1.21d |
| AMR | 1.44 ± 0.37a | 0.68 ± 0.03 | 5.67 ± 0.95bc |
| PC | 0.51 ± 0.06e | 0.79 ± 0.02 | 2.51 ± 0.41e |
| PR | 0.47 ± 0.06e | 0.79 ± 0.00 | 2.54 ± 0.37e |
| PMC | 1.70 ± 0.29a | 0.75 ± 0.03 | 7.49 ± 1.90a |
| PMR | 1.11 ± 0.18bc | 0.73 ± 0.01 | 4.72 ± 1.01bcd |
| SC | 1.18 ± 0.13b | 0.76 ± 0.07 | 4.27 ± 1.56cd |
| SR | 0.63 ± 0.24d | 0.82 ± 0.06 | 4.60 ± 0.06bcd |
| SMC | 1.50 ± 0.22a | 0.75 ± 0.04 | 6.09 ± 2.05ab |
| SMR | 1.43 ± 0.33ab | 0.74 ± 0.04 | 5.95 ± 1.89ab |
n = 9, Values are means ± SD,
Hardness and adhesiveness: significant differences between labels (p < 0.05).
Cohesiveness: no evident difference between labels.
a–e: Differences were significant (p < 0.05).
A: milk fat cream; P: vegetable fat cream; S: soy milk cream;
C: samples stored at 5 °C; R: samples stored at 25 °C;
M: samples added with miso
The hardness of miso-containing sample AMR (1.44 [× 103 N/m3]) was higher than that of sample AR (p < 0.05), while the hardness of sample AMC (0.77 [× 103 N/m3]) was lower than that of sample AC (p < 0.05). Neither cohesiveness nor adhesiveness was significantly different between the cream A samples, with or without miso.
Among the cream P samples, the hardness of sample PMR (1.11 [× 103 N/m3]) was higher than that of sample PR (p < 0.05), and the hardness of sample PMC (1.70 [× 103 N/m3]) was higher than that of sample PC (p < 0.05). Cohesiveness was not significantly different between the cream P samples with or without miso. The adhesiveness of samples PMC (7.49 [× 102 J/m3]) and PMR (4.72 [× 102 J/m3]) was higher than that of samples PC and PR (p < 0.05).
The hardness of samples SMR (1.43 [× 103 N/m3]) and SMC (1.50 [× 103 N/m3]) was higher than that of samples without miso, (p < 0.05), regardless of storage temperature. Cohesiveness was not significantly different among the cream S samples with or without miso. The adhesiveness of sample SMR (5.95 [× 102 J/m3]) showed almost no difference from that of sample SR, whereas the adhesiveness of sample SMC (6.09 [× 102 J/m3]) was higher than that of sample SC (p < 0.05).
Microstructure of cream samples The microstructure of the cream samples was examined using a tabletop electron microscope (Fig. 3). Samples AC and AR consisted of small-sized bubbles, whereas samples PC and PR consisted of spherical bubbles of relatively large and consistent size. Sample SC consisted mainly of fine bubbles with occasional large bubbles, whereas in sample SR, a substantial fraction of small bubbles was observed along with large bubbles that were distributed relatively evenly.
SEM images of samples A, P, and S
Espuma samples were frozen with liquid nitrogen; the frozen samples were sliced and placed on the sample table to examine the torn surface.
A: milk fat cream; P: vegetable fat cream; S: soy milk cream;
C: samples stored at 5 °C; R: samples stored at 25 °C
Table 3 shows the mean long diameters of 30 bubbles observed under a scanning electron microscope. The bubbles in the AC, PC, and SC cream samples were larger than those in samples AR, PR, and SR (p < 0.05). The bubbles in the S cream samples were smaller than those in the other samples, which possibly contributed to foam stability.
| Sample | A | P | S |
|---|---|---|---|
| C | 120 ± 30b | 160 ± 40a | 80 ± 30cd |
| R | 80 ± 40c | 110± 40b | 50 ± 40cd |
| MC | 70 ± 30c | 80 ± 30cd | 50 ± 10de |
| MR | 70 ± 30c | 40 ± 20de | 30 ± 10ef |
Values are means ± SD, n = 30,
a–e: Differences were significant (p < 0.05).
A: milk fat cream; P: vegetable fat cream; S: soy milk cream;
C: samples stored at 5 °C; R: Samples stored at 25 °C;
M: samples added with miso
The mean long diameter of the bubbles was 70 µm in samples AMC and AMR. The bubbles in sample AMC were smaller than those in sample AC (p < 0.05), but the bubbles in samples AMR and AR were not significantly different in size. Among the P cream samples, the mean long diameter of the bubbles was 80 µm in sample PMC and 40 µm in sample PMR, and both were smaller than the bubbles in samples PC and PR (p < 0.05). Likewise, among the S cream samples, the mean long diameter of the bubbles was 50 µm in sample SMC and 30 µm in sample SMR, and both were smaller than those in samples SC and SR (p < 0.05). The addition of miso showed the smallest standard deviation of bubble sizes in each sample, indicating the formation of uniform bubble sizes.
The long-to-short ratio of cream samples was investigated in relation to the amount of syneresis. The results showed that the closer the long-to-short ratio was to 1, the lower the amount of syneresis and the higher stability of the sample. The correlation coefficient (r) between the long-to-short ratio and the amount of syneresis was −0.94877. This was measured by using an electron microscope to determine the ratio of the area of the bubbles and the successive layers surrounding these bubbles, after which the rate of bubble inclusion was calculated. The density of the cream sample decreased as the rate of bubble inclusion increased. The correlation coefficient (r) between the rate of bubble inclusion and the density was ∓0.94573.
Appearance and nutrient calculations The appearances of samples X and Y are shown in Table 3. Visual observation of the surface of sample Y revealed it to be smoother than that of sample X.
The weight per 100 kcal of samples X and Y was respectively 33.0 g and 35.0 g, the amount of protein was 2.8 g and 3.2 g, the amount of fat was 9.2 g and 8.7 g, and the amount of carbohydrate was 0.7 g and 1.2 g. The amounts of protein and carbohydrate in sample Y were relatively higher than those in sample X.
Density of prepared samples Table 3 shows the density measurement results of samples X and Y. The density of sample X (0.46 [× 103 kg/m3]) was higher than that of sample Y (0.42 [× 103 kg/m3]; p < 0.05).
Amount of syneresis in prepared samples At 30 min after preparation, the amount of syneresis of sample X was 0.8 ± 0.4 mL, whereas that of sample Y was 0.0 ± 0.1 mL. Both samples showed minimal syneresis, which maintained the stability of the foam.
Texture of prepared samples Table 4 shows the results of texture measurements of samples X and Y. The hardness and adhesiveness of sample Y were both lower than those of sample X (p < 0.05). However, cohesiveness was not significantly different between the two samples.
| Sample | X | Y | |
|---|---|---|---|
| Appearance |  |
 |
|
| Density [× 103 kg/m3] | 0.46 ± 0.03a | 0.42 ± 0.01b | |
| Hardness [×103 N/m2] |
2.26±0.13a | 0.81 ±0.03b | |
| Texture | Cohesiveness | 0.73 ±0.05 | 0.78 ±0.05 |
| Adhesiveness [×102 J/m3] |
6.41 ± 1.77a | 1.93±0.03b | |
n = 9, Values are means ± SD,
a,b: Differences between the examined samples were significant (p < 0.05).
Notably, the texture measurement results of sample X satisfied the Level II standard of CAA-approved food for people with dysphagia, and those of sample Y satisfied the Level III standard (Table 1). The three levels of CAA-approved food for people with dysphagia are as follows: Level I is jelly-like food, Level II is mousse-like food, and Level III is porridge and paste-like food. Accordingly, Level I foods are suitable for people with the most severe dysphagia.
Microstructure of prepared samples Fig. 4 shows the microstructures of samples X and Y, as observed with a scanning electron microscope. The bubbles in sample X were small, whereas those in sample Y were large. The successive layers surrounding these bubbles were thin in sample X and thick in sample Y, suggesting that the foam was more stable in sample Y than in sample X. Tissue flakes that were likely salmon fish meat were observed in the successive layers of sample Y. The mean long diameter of bubbles in sample X (n = 30; 90 ± 20 µm) was longer than that in sample Y (150 ± 30 µm; p < 0.05).
Microstructure of samples X and Y. Both samples were frozen with liquid nitrogen; frozen samples were sliced and placed on the sample table to examine the torn surface. X, sample X; Y, sample Y, ↑: indicates likely fragments of salmon meat.
Sensory evaluation of prepared samples Fig. 5 shows the results of the analytic sensory evaluation conducted by the students. Sample X was evaluated to be weaker than sample Y in terms of “Softness” and “Smoothness in melting in the mouth” and stronger in terms of “Stickiness” and “Rough sensation on the tongue” (p < 0.05). In the sensory evaluation of palatability, the appearance of sample Y (3.2 ± 1.0) was preferable to that of sample X (2.8 ± 0.9; p < 0.1). The results for the overall evaluation (deliciousness) showed that sample Y (3.8 ± 0.8) was preferable to sample X (3.3 ± 1.0; p < 0.05).
Analytic sensory evaluation of samples X and Y by university students using five-grade scoring Analysis evaluation 1: Weak 3: Neither 5: Strong n = 20, **: p < 0.01, *: p < 0.05
In the free-response column, several negative opinions regarding sample X were noted, such as sample X was greasy without smoothness, and some panelists mentioned that they felt a certain degree of roughness. In contrast, positive remarks were obtained for sample Y, such as that the texture was smooth and pleasant to taste, and that it retained the flavor of Japanese cooking or the deep flavor of miso.
Fig. 6 shows the results of the analytic and palatability sensory evaluations by the older adults. In the analytic sensory evaluation, sample Y was evaluated to be strong in terms of salmon aroma and flavor, as well as ease in swallowing and eating. The palatability sensory evaluation revealed that sample Y was favorable overall, with a score of 2.40 ± 0.07. Half of the panelists evaluated sample Y as delicious. Some mentioned in the free-response column that they wished the food was in the shape of fish fillets.
Sensory evaluation of sample Y by older people using three-grade scoring.
Analysis evaluation 1: Weak 2: Neither 3: Strong Palatability evaluation 1: Unfavorable 2: Neither 3: Favorable
n = 10
Endoscopic examination of swallowing prepared samples Swallowing was evaluated using sample Y. Table 5 shows the presence and/or absence of laryngeal residue after the first and second swallows of 2 older adults (W and Z). In both adults, almost no laryngeal residue remained after the second swallow.
| Subject | First swallowing | Second swallowing |
|---|---|---|
| W |  |
 |
| Z |  |
 |
◯: laryngeal residue
Properties of espuma prepared using three types of cream and selecting the cream for preparing boiled salmon in cream The density of espuma samples P and S was low, with minimal syneresis. Samples stored at 5 °C showed better foam stability than those stored at 25 °C. Samples SC and SR exhibited no syneresis at all, indicating that soy milk cream is extremely suitable as a coagulating agent for preparing espuma. The emulsifiers contained in creams S and P were found to have a stabilizing effect. However, the emulsion can be destroyed by stirring and/or aeration, which induces the stability of air bubbles (Noda, 1993). These effects were thought to have been responsible for the light, low-density air bubbles in samples S and P. Furthermore, sample S was thought to show the least syneresis because the structure for cream S consists of spherical proteins such as soybean protein, in addition to its stable molecular film (Noda, 1988; Tsuge and Ogoshi, 2011). Cream P also contained soy protein. Typical culinary foams, such as meringue, soufflé, and cake, have a stable structure due to the presence of adsorbed proteins at the fluid interface between the gas and water phases (Dickinson, 1989). Also, proteins are either hydrophilic or hydrophobic, and function as surfactants that maintain the stability of bubbles by lowering the surface tension of water (Hatae, 2017). These protein characteristics are thought to have brought about the bubble stability effect in cream S. Cream A is conventionally used as the coagulating agent for producing espuma. However, sample A was inferior in quality because it had a higher bubble density and a higher amount of syneresis compared with samples P and S. On the other hand, sample A stored at 5 °C showed similar values to samples P and S. Milk fat has a lower solid fat index compared with other types of fat and easily melts and undergoes syneresis as the temperature rises (Noda, 1993). Also, fat coagulation is a primary factor in inducing the foaming of cream, which occurs at lower temperatures and enhances formability. Accordingly, the density of the cream samples stored at 5 °C was lower than at 25 °C.
Hence, espuma sample A should be stored at 5 °C and served as soon as possible. Sample P also showed better bubble density and amount of syneresis at this low temperature, indicating that it should also be stored at 5 °C. Sample S was not affected by temperature and did not show any sign of syneresis, and thus proved to be the superior coagulating agent for creating the espuma form.
There are no recent publications regarding cream foam density. Therefore, in this study, cream foam density was determined by stirring creams using an electric hand mixer for 5 min. The density of cream foam A was 0.48 [× 103 kg/m3] at 5 °C, whereas that of creams foam P and S was 0.38 [×103 kg/m3] and 0.47 [× 103 kg/m3], respectively. The results indicate that cream foam produced by the espuma method includes a larger amount of air than normal whipped cream, thereby producing a lighter foam.
The addition of miso reduced the amount of syneresis in the espuma samples and improved their stability. The surfactant effect of the proteins in the miso improved the foam stability of the espuma.
We compared the texture of cream foams made using an electric hand mixer with the espuma samples. The hardness of the A, P, and S cream foams was 3.13 [× 103 N/m3], 1.47 [× 103 N/m3], and 1.15 [× 103 N/m3], respectively. These results were higher than those for the espuma samples. Based on these results, it was considered that cream foams made using the espuma method have a light and soft texture.
Although sample A by itself did not satisfy the Level II standard of CAA-approved food for people with dysphagia, it did with the addition of miso. Foam stability has been found to improve when the size of the foam bubbles becomes uniform (Tsuge and Ogoshi, 2011). Therefore, in the present study, the addition of miso reduced the amount of syneresis, the size of the foam bubbles became uniform, and the stability of the foam improved. Using miso as a condiment should be considered effective in the preparation of espuma foods.
Properties of prepared samples A relatively small amount of boiled salmon in cream espuma food (31.0–33.0 g) contains 100 kcal, according to the calculation results for the amounts of protein, fat, and carbohydrates. Accordingly, the espuma method is considered to be an ideal cooking method, and the food prepared using this technique is appropriate for people who are undernourished or who can ingest only a small amount of food.
The espuma food samples X and Y showed little to no syneresis at 30 min after preparation. Therefore, both samples are considered appropriate for intake by people with impaired chewing and/or swallowing abilities within 30 min of preparation. However, espuma food dries easily on surfaces, so it should be consumed as soon as possible.
The texture of both samples X and Y satisfy the CAA-approved standard of food for people with dysphagia, which indicates that their form is well suited for older adults who have difficulty in chewing and/or swallowing. Those who require Level I standard food should be given slightly more solid foods than those prepared in this study.
Regarding the microstructure of the prepared samples of espuma, the mean long-diameter of the bubbles in the foam of sample Y was significantly longer than that of sample X. This difference in diameter is likely due to larger amounts of air in sample Y than in sample X, which resulted in lighter, lower-density bubbles. The mean long-diameter of the bubbles for only cream sample P was longer than that for sample S; however, the mean long diameter for sample X prepared as boiled salmon in cream was shorter than that for sample Y. Thus, the addition of boiled salmon in cream affected the mean long diameter of the bubbles.
Sample Y was evaluated more favorably than sample X in the sensory evaluation by the university students. Sample Y was assessed as being softer, easily melted in the mouth, and had a less rough sensation on the tongue compared with sample X. Thus, sample Y is considered an easy-to-swallow espuma food that can be used in meals for older people. Sample Y was also considered favorably overall in the palatability sensory evaluation and was preferred over sample X.
Sample Y was also evaluated favorably in the sensory evaluation by the panel of older adults. The ease of swallowing sample Y enables a stress-free diet, which is another result that supports its suitability for use in meals for older adults. In the evaluation of food taste, only 1 of 10 participants responded that sample Y was “Unfavorable,” half of the remaining participants answered “Favorable,” and the rest answered “Neither.” Thus, there was no evident issue with palatability.
The results of the endoscopic examination of swallowing using sample Y showed little to no laryngeal residue after the second swallow, indicating the low possibility of dysphagia. Endoscopic examination of swallowing was conducted with two healthy older women without incident, suggesting that sample Y can used in meals for older adults with swallowing dysfunction. Further sensory and endoscopic examinations of swallowing in older adults with swallowing dysfunction are necessary to support the future widespread use of espuma food in this population.
The results of this study suggest that boiled salmon in cream prepared as espuma food is appropriate for use in meals for older adults with masticatory disturbance or dysphagia.
This study demonstrated the suitability of making espuma foods from vegetable fat cream and soy milk cream because they have low density and their foam shows minimal syneresis. Boiled salmon in cream prepared as espuma food satisfied the CAA-approved food-texture standard for people with dysphagia. The findings of this study suggest that boiled salmon in cream prepared as espuma has the potential to contribute to improving quality of life and preventing malnutrition in older adults with masticatory disturbance or dysphagia by supplying safe and delicious meals.
Acknowledgements The authors would like to thank the students and older adults who patiently cooperated with the sensory evaluation tests. The authors also thank Fuji Oil Co., Ltd. for providing the soy milk cream for espuma food preparation.
Conflict of interest There are no conflicts of interest to declare.
