Optimal preferred MSG concentration in potatoes, spinach and beef and their effect on intake in institutionalized elderly people

Abstract

Background

Elderly people may benefit from sensory stimulation to increase food intake since anorexia of ageing is prevalent among them. An optimal MSG concentration may increase the palatability of foods but this depends on the food and chemosensory status of the taster. Currently, the results on taste enhancing to increase intake are inconsistent.

Objective

To find an optimal preferred MSG concentration in mashed potatoes, spinach and ground beef and to determine whether this concentration increases consumption of these foods among institutionalized elderly people.

Design

Single blind within subject cross-over study performed at the laboratory and in the residents' own apartments.

Participants

33 elderly and 29 young people in the sensory study and 53 elderly people in the intake study.

Measurements

Pleasantness of the foods was rated of the foods each with 0, 0.5, 0.8, 1.3 and 2.0 g of MSG/100g. Intake was measured by weighing back leftovers of 2 meals with MSG (0.5% in mashed potatoes, 2% in spinach and ground meat) and without MSG.

Results

0.5% MSG (p<0.05) was preferred in mashed potatoes but no optimal preferred concentration was found for spinach and ground beef, possibly because of their complex taste. Intake was not different between the foods with and without MSG or the total meal (all p>0.68).

Conclusion

MSG (0.5% and 2%) does not guarantee a higher intake among elderly. The chemosensory heterogeneity of the elderly population requires more individual flavor enhancement to improve the dietary intake and sensory experience.

Introduction

Gradual loss of smell and taste perception is common among elderly people ( 1., 2., 3.). Overall, the sense of smell is more affected than the sense of taste (4). Since both senses play a major role in the palatability of food (5), the enjoyment and as a consequence the intake may decline because of the changes in chemosensory function (6). This could lead to body weight loss and a poor nutritional status (6).

Previous studies suggest that MSG (mono sodium glutamate) improves the pleasantness and intake of food ( 7., 8., 9., 10., 11., 12.). The optimal concentration of MSG differs for each food product. For example, Bellisle et al. showed that 0.6% MSG increased the intake of mushroom soup and mashed potatoes but not of rice (13). They also demonstrated that 0.6% MSG increased the intake of pasta and semolina but not of vegetables such as green beans, celery and cauliflower (14). Further this group showed that a concentration of 1.2% MSG added to beef jelly and spinach mousse caused an immediate increased consumption but no sustained higher intake (15). Lastly, Yamaguchi and Takahashi (16) found optimum MSG concentrations of 0.3% in clear soup, 0.3% in egg custard and 0.9% in miso soup.

Elderly people prefer higher MSG concentrations in various food items compared to young adults. Murphy (12) found higher preferred MSG levels for older people when 0-0.5% MSG was added to soup. In younger people, Schiffman and Miletic (17) found optimal concentrations of 0.3% MSG in corn, 0.15% in carrots, 0.6% in chicken broth and 0.8% in onion soup while these concentrations were at least 10 times higher in corn and carrots, 2.0% in chicken broth and 1.5% in onion soup for elderly people. However, it is unclear in this last study how these optimal concentrations were determined.

The objective of the two-fold study reported here is to determine whether or not an optimal preferred MSG concentration in several foods increases intake in elderly people. In the sensory study we sought for an optimal preferred concentration of MSG in several food items for young and elderly using MSG concentrations varying from 0.5%-2.0% w/w. In the intake study we examined the effect of the obtained optimal concentrations on intake in elderly participants.

Methods

Participants

The sensory study included elderly people who were residents of nursing home “Dorpsveld” in Rotterdam, the Netherlands. The young adults were recruited by advertisements in Wageningen University buildings. The intake study included only elderly people, recruited from 3 nursing homes in Deventer, Gendringen and Rijssen, the Netherlands.

A general questionnaire was used for screening. The inclusion criteria for the elderly were: ≥ 65 yrs, able to participate in a sensory study, good eyesight, no allergy to MSG or the foods in this study, not on a sodium restricted diet, no disease in terminal phase and no use of antidepressants. The

criteria for the young adults were: < 30 yrs of age and not allergic to MSG or to the foods in this study. After screening, 39 elderly and 29 young people were enrolled in the sensory study and 58 elderly participated in the intake study. Everyone provided their written informed consent. The Medical Ethical Committee of the Division of Human Nutrition of Wageningen University approved the study protocol plus the used materials.

Design

The sensory study started with a 1-day pilot experiment to select three suitable foods and to determine the adequate MSG concentration range to use in the main sensory study. For the latter experiment, the participants were divided over 6 test sessions and rated 15 food samples (3 foods each with 5 MSG concentrations).

The intake study had a within subject cross over design and lasted 4 weeks. Once a week on the same day the elderly people received a hot meal using the 3 foods from the sensory study, either with or without MSG. Everyone received 2 meals with MSG and 2 meals without MSG in random order.

The olfactory performance and the nutritional status (of the elderly) were assessed in both studies. The studies were carried out single blind.

Stimuli

The pilot experiment was conducted with 5 young participants and 5 foods i.e. mashed potatoes (potato powder, Maggi, Nestlé), frozen spinach, minced meat, cream of mushroom soup and bread. All products were tested with 0, 0.5, 0.8, 1.3 and 2.0 g of MSG (Spice Island, Koninklijke Euroma B.V., Wapenveld, the Netherlands)/100 g of product (0.2 log steps), except for bread, which was tested with 0, 0.5, 2.0 and 3.0 g of MSG/100 g of product. Half of the normal salt levels were used in the food products. When adding MSG to a product, less salt could be used without the loss of palatability (18). The reduced added salt levels were 0.2 g salt (NaCl)/100 g in the mashed potatoes (0.07 g of salt in the powder -could not be reduced- and 0.13 added during the preparation), 0.25 g salt/100 g in spinach and 0.37 g of salt in minced meat (19).

Mashed potatoes, minced meat and spinach were found suitable for the sensory and the intake study plus they make up an appropriate hot meal in a Dutch nursing home. The MSG range appeared to be wide enough because the concentrations showed clear effects on the perceived pleasantness ratings (especially in mashed potatoes and minced meat) and the intensity (in all products). Thus, mashed potatoes (0.2 g NaCl/100 g), spinach (0.25 g NaCl/100 g) and minced meat (0.37 g NaCl/100 g) were included, each with 0, 0.5, 0.8, 1.3 and 2.0 g of MSG/100 g. The preffered MSG concentrations for the elderly as determined in the sensory study (0.5% MSG in mashed potatoes (optimal), 2% in minced meat and 2% MSG in spinach) were used in the intake study. For minced meat 2% was chosen because the pleasantness ratings increased upon higher MSG concentrations and was highest at 2%. Spinach, as well as minced meat has a complex taste; therefore a substantial amount was needed to make up a sensory difference between the regular and the MSG enriched spinach.

Procedure

The foods in both studies were prepared according to a standard recipe (19). For the sensory study, the foods were prepared in the nursing home kitchen except for minced meat which was prepared in the university research kitchen 2 days before testing and refrigerated at approx. 5°C. Before each test session, the reheated meat (3 minutes at 900 Watt) and other foods were mixed with an appropriate amount of MSG. All foods were kept warm in aluminium boxes with cardboard lids (labelled with or without MSG in the intake study) au bain-marie (Hufner, Munster, Germany).

For tasting and rating of the food samples, the elderly participants of the sensory study were seated at separate tables in a room (±18m²) in the nursing home. The young people were seated in sensory test cabins in the university laboratory. Everyone was instructed not to talk. The researchers checked this by being present in the room of the nursery home as well as in the laboratory. Approximately 40 g of each sample (65°C) was served in labeled plastic cups with plastic spoons with a 2 min interval for the elderly and 1 minute for the young people. The order of the food products was randomized between the 6 test sessions and the order of the MSG concentrations was randomized within the test sessions.

For the intake study, some elderly were seated in a living room while others stayed in their apartment. These conditions did not change during the study. On a warm plate (0 25 cm), 300 g (Denver Instrument/XP-3000, round off in 0,1 gram) of spinach, 300 g of mashed potatoes and 200 g of minced meat were weighted and put on a tray with a cup of broth (bouillon) and a dessert (yogurt). All the meals, with or without MSG were delivered at the same time every day.

Measurements

Anthropometry

In both studies, the knee-to-floor height (KFH) of the elderly participants was measured twice without shoes using a stadiometer in a sitting position, from the anterior surface of the thigh to the floor with the ankle and the knee each flexed at a 90° angle against the metallic help. Body height was calculated as follows: height (in cm) = 3.16*KFH (cm) (22).

Body weight was measured using a weighing scale (Seca, Hamburg, Germany).

Sensory evaluation of taste intensity and pleasantness (sensory study)

Taste intensity and pleasantness of the 3 foods with varied MSG concentrations were rated on a 10-point scale by the taste and swallow method. Dutch elderly are familiar with this scale because it is used in the Dutch school system. The left anchors (=1) were marked ‘not at all’ and the right anchors (=10) ‘extremely’ for each attribute. After each sample, participants neutralized their taste with tap water.

Dietary intake of the hot meal (intake study)

The portion sizes were twice as large compared to normal sizes so the elderly could eat ad libitum. Intake was measured by weighing the left-overs of each component (minced meat, spinach and mashed potatoes) together and separately. The data were converted into nutrients and energy using the Dutch food composition table (23).

Mini Nutritional Assessment (MNA)

The MNA® is a screening and assessment tool to identify malnutrition in the elderly (24). A calculated score from the screening section indicates possible malnutrition or not. If so, the questions of the assessment section plus two anthropometric measures namely mid arm circumference and calf circumference clarify whether the participant is at risk of malnutrition, or if the participant is malnourished. The MNA was administered to the elderly people.

Olfactory sensitivity

The European test of Olfactory Capabilities (ETOC) (Neurosciences et Systemes Sensoriels, Universite Claude Bernard Lyon 1, France and Healthsense) measured olfactory sensitivity for common odorants. The test contains sixteen blocks of four vials and only one vial (15 ml) contains an odor (25). The odors are: vanilla, cloves, apple, eucalyptus, cinnamon, fuel-oil, pine, garlic, cut grass, anise, orange, fish, rose, thyme, lemon and mint. To evaluate the olfactory performance, a detection task was given followed by an identification task by selecting the right descriptor between the four options. A correct identification answer was only included when the correct accompanying detection answer was given. The ETOC score (maximum score=32) is the sum of the correct detection score and the correct identification score.

Cutoff values resulting from 95% confidence limits for different age groups were established upon 1330 participants with the ETOC (49). The correct answers from our participants were compared to these cutoff values according to age. Normal scores for the young subjects (20-29 y) were 16 for detection and ≥ 14 for identification. Normal scores for 70-79 y: detection score ≥ 14 and identification score ≥ 11. For ≥ 80 y: detection score ≥ 13 and identification score ≥ 9.

Data analysis

For the sensory study, a within-subjects repeated measures design (ANOVA) compared mean attribute ratings measured at 5 different MSG concentration levels. Age served as a between-subject factor and concentration as a within-subject factor. With the cutoff values derived from results of the ETOC, participants within both age groups were divided into a group with low and normal olfactory performance. ANOVA compared pleasantness ratings measured at the 5 MSG levels between the low and normal group.

The Wilcoxon Signed Rank Test compared differences in mean intake between the hot meals with and without MSG. The data of the intake study was not normally distributed and intake was measured within the same subject. A Pearson correlation coefficient determined if the change in energy intake was related to the ETOC score.

All statistical analyses were performed with SPSS for WINDOWS software (version 11.5; SPSS Benelux, Gorichem, The Netherlands). P values < 0.05 were considered significant.

Results

Participants

Thirty-three elderly out of 39 completed the sensory study. Six dropped out, because of health (4) and personal reasons (2). All 29 young participants completed the study.

Dietary intake data of 53 out of 59 elderly were included in the analysis of the intake study. Six people dropped out, 4 did not like the hot meal, 1 person consumed only one meal and the last participant did not save the left-overs. Of the 53 participants, 87% ate all the 4 meals and 13% consumed 2 or 3 meals with at least one hot meal of both conditions.

General characteristics and olfactory performance

The general characteristics of the participants of both studies are given in Table 1. The percentage of elderly wearing dentures was 82 in the sensory study versus 96% in the intake study. The elderly participants had lower ETOC scores with a higher variation (figure 1) compared to the young adults.

Table 1.

Characteristics of the young and the elderly participants of the sensory study and the intake study (mean & SD)

Characteristic	Sensory study		Intake study Elderly (n=53)
	Young (n=29)	Elderly (n=33)
Gender (male/female)	6/23	12/21	13/40
Age (y)	21.3 (2.2)	80.8 (6.2)	85.8 (6.2)
Height (m)	1.74 (0.1)	1.65 (0.1)	1.61 (0.1)
Weight (kg)	64.4 (8.0)	74.6 (14.1)	68.3 (11.0)
BMI (kg/m2)	21.3 (2.0)	27.4 (4.9)	26.5 (4.2)
Dentures (%)a	4	82	96
Smoking (yes/no) (%)	10/90	9/91	2/98
MNA: risk of malnutrition
No risk	-	79	85
Increased risk		21	11
Malnutrition		0	4
ETOC scorec	28.1 (1.9)	20.0 (7.7)	12.0 (7.5)
ETOC score/max score (%)d	88	63	38
Participants with normal ETOC score for age (%)e	21	33	0

^a.Sensory study participants: denture wear not subdivided into complete or partial wear; b. MNA = Mini Nutritional Assessment. Normal score 23.5 (maximum score = 30);

^c.ETOC score= sum of the correct detection score and the correct identification score of the European Test of Olfactory Capabilities. (Range: 0-32);

^d.Mean ETOC score divided by the maximum score (32);

^e.Percentage of young or elderly people with normal ETOC score according to cut off values for age

Figure 1.

Correct detection (♦) and identification (■) responses (both with max score of 16) of the young (n=29) and all the elderly (n=33 in sensory study; n=53 in intake study) in the European Test of Olfactory Capabilities as a function of age

Sensory study

Effect 0-2% MSG on taste intensity

The mean taste intensity scores of all 3 foods increased with higher concentrations of MSG (Figure 2a). The effect of concentration was significant for mashed potatoes and spinach [F(,236)>9.7, p<0.01] and minced meat [F(4,240)=5.5, p<0.01]. The young participants had higher intensity ratings for the five concentrations compared to the elderly. Hence, the age effect for mashed potatoes, spinach [both F(1,59)>12, p<0.01] and for minced meat [F(1,60) =20.3, p<0.01] was significant. For mashed potatoes there was significant age x concentration interaction effect [F(4,236)=2.6, p<0.05], but the slopes of the curves were not different between the young and the elderly for spinach [F(4,236)=1.2, p=0.3] and minced meat [F(4,240)=0.9, p<0.5]

Figure 2a.

Mean intensity ratings of mashed potatoes, spinach and minced meat as a function of MSG concentration in young (– ♦ –, n=29) and elderly (–■–, n=33, and n= 32 in spinach ratings) participants

Effect 0-2% MSG on pleasantness

The effect of concentration was significant for mashed potatoes and minced meat [both F(4,240)>3.1, p<0.05] but not for spinach F(4,236)=0.8, p=0.5]. The pleasantness curve of mashed potatoes for the young participants peaked at 0.5% of MSG but the scores for the elderly people were stable from 0.5% MSG onwards (Figure 2b). The pleasantness of minced meat increased with increasing MSG concentration. There was no clear pleasantness optimum for spinach for both age groups.

Figure 2b.

Mean pleasantness ratings of mashed potatoes, spinach and minced meat as a function of MSG concentration in young (–♦–, n=29) and elderly (–■–, n=33, and n= 32 in spinach ratings) participants

The mean pleasantness responses over the five concentrations for each of the foods were not different between the age groups [mashed potatoes and minced meat F(1,60)>0.03, p>0.6]; spinach F(1,59)=1, p=0.3]. No interaction effect (age x concentration) was found for each of the foods, meaning that the slopes for the two curves were not different [mashed potatoes and minced meat F(4,240)>0.4, p>0.6]; spinach F(4, 236)=0.7, p=0.6].

Effect of olfactory performance on pleasantness ratings

By means of the ETOC cut off values, we found little difference between the low and normal performing group. The averaged pleasantness responses over the concentrations in mashed potatoes were not different between the low and the normal performing group in the young plus elderly group [F(1,60)=0.1, p=0.72] and in the young group [F(1,27)=1.6, p=0.2]. Also, within the elderly group there was no sensory effect [F(1,31)=0.1,p=0.8].

Intake study

There was no difference in intake (kJ) (Table 2) between the minced meat with MSG versus without MSG (p ≥ 0.925), spinach (p ≥ 0.685), mashed potatoes (p > 0.941) and the total meal (p ≥ 0.896).

Table 2.

Mean energy (kJ) (SD) intake of the hot meal components with and without MSG

	Energy (kJ)
	+ MSG	- MSG
Minced meat	1113 (574)	1135(602)
Spinach	83 (44)	85 (46)
Mashed potatoes	559 (315)	564 (322)
Total meal	1756 (881)	1774 (905)

Pearson’s correlation coefficient (r) for the change in energy intake and the ETCO score (correct detection plus identification scores) was -0.24 (p=0.082).

Discussion

This study showed that the consumption of mashed potatoes enhanced with an optimal preferred concentration of 0.5% MSG and spinach and ground beef with 2% MSG did not increase among institutionalized elderly.

Sensory study

Both the young and the elderly people preferred 0.5% MSG in mashed potatoes. For spinach no clear optimum of MSG was found within both age groups but for minced meat the pleasantness ratings given by the elderly people seemed to increase slightly upon higher MSG concentrations. The MSG range up to 2% is chosen partly due to the complex taste of spinach and minced meat. Although no clear optimum was found in these foods, we believe our range was adequate and should not have been higher. Other studies find optimum preferred pleasantness ratings at MSG concentrations of 0.3%-0.9% (28) or 0.6%-1.2% (15) although one study used a range up to 6% or 8% (16).

Our finding is in line with results by Mojet et al. (26) who did not find an optimum for pleasantness among young and elderly people with a range of 0.16% to 1% MSG in broth. Other studies are in contrast with our finding (27, 28, 16). Prescott and Young (11) found that 0.8% MSG added to vegetable soup increased its palatability among young and elderly people. Furthermore, Bellisle et al. (15) found an optimum of 0.6% for beef jelly and spinach mouse when using a range varying from 0-1.2% MSG. The inconsistent results above could be due to a food related issue. As shown by Bellisle et al. (13, 14) optimum concentrations of MSG were more pronounced in starchy dishes than in vegetable dishes. Both spinach and minced meat have a complex taste and are both glutamic acid-rich foods. This may work against the palatability enhancing effect of added MSG.

The effect of the MSG concentration on perceived taste intensity was significant in all 3 foods. This is not entirely in line with results found by Mojet et al. (29). The intensity of the ascending concentrations of MSG dissolved in water was perceived as different but not when MSG was dissolved in a food product (bouillon). The effect of an increasing MSG concentration in a more neutral environment seems more pronounced than when dissolved in a product with a complex taste of itself. This is what our results show in the mashed potatoes (more neutral taste) versus spinach and minced meat.

We found no age related difference for a higher flavor preference while some studies report the opposite (30 - 34,12, 35). The age related difference might depend on the kind of flavor or taste. Perhaps for savory tastes like salt and umami, the difference in optimal preferred concentration between the age groups is less pronounced than for example sweet flavors (31, 26). In line with this, Zallen et al. (36) found no age related difference in optimal preferred salt concentrations and neither did Mojet et al. (26).

In agreement with Koskinen, Kalviainen and Tuorila (37), Koskinen & Tuorila (38), Kremer, Bult, Mojet and Kroeze (39), Forde and Delahunty (40) we found no association between an impaired olfactory sensitivity and higher preferred flavor levels. In line with our results, Mojet et al. (26) did not find an effect of threshold sensitivity and perceived supra threshold intensity on optimal preferred concentrations of several flavors dissolved in water. The variation among elderly in their ability to perceive a flavor may add to the difficulty to relate the impaired sensory sensitivity to an increased liking of flavor enhanced products.

Intake study

This study shows that the energy intake of the total hot meal or any of the separate foods with MSG did not increase compared to the same meal or components without MSG among institutionalized elderly people.

Because of practical reasons it was not possible to perform both the sensory study and the intake study in the same population of elderly people. Although both groups are institutionalized elderly people they are a heterogeneous group and may have a different sensory performance. Therefore, the optimal MSG concentration might have varied between the two groups.

Denture wear among 96% of our elderly population, could have contributed to our finding. Complete and/or partial denture wear has shown to affect food intake in several ways (41, 42, 43). Because of ethical reasons no specific medication use could be asked for. Therefore, we are unaware if certain drugs were taken that affected olfaction and taste which could have interfered with our results. Instead we asked for specific diseases that are related to diminished olfaction, taste and/or intake such as diabetes, cancer and Parkinson disease. Subjects with any of these conditions were not included in this study.

The results of studies on adding extra flavor and/or MSG as a way to increase food intake are inconsistent. While some studies confirm a higher intake with higher sensory stimulation either with MSG and/or flavors (44), others do not (45, 34). Our results are in agreement with a recently performed long term trial (46). In a study where flavor enhancers (primarily odors) were added to the diet of 39 institutionalized elderly during 3 weeks, the intake of only 3 out of 20 enhanced foods significantly increased (8). Although De Jong et al. (33) noted a higher preference for sucrose in breakfast items, no increase was found in the amount eaten because of the higher sweetness. Other studies report the opposite (13, 14). Schiffman (10) showed that an added combination of 0.3% to1.0% MSG and flavor to foods during a week improved energy intake by 10% or more in 40 out of 43 sick elderly. It is unclear which foods she used to enhance. Mathey et al. (44) added a mixture of MSG (0.3%) and flavors to the hot meal of elderly people and found an increased intake of the flavor enhanced foods.

In line with a recently performed study (46) we found that elderly people with a low olfactory performance have an increased intake of food when exposed to a flavor enhancement treatment compared to the elderly with a normal performance. This suggests that for the elderly with low olfactory scores, flavor amplification may hold potential to increase intake (47). However, also in these people olfactory loss does not apply to all odours (48).

In conclusion, we found that an optimal preferred MSG concentration of 0.5% MSG in mashed potatoes and 2% in spinach and minced meat does not necessarily guarantee a higher intake of these foods. Likely, factors such as denture wear, medicine use and the heterogeneity in olfactory functioning contribute as to why flavor enhancement can not be used as a one-size-fits-all approach.

Future research should focus more on the individual older adult and their status of chemosensory functioning to gain insight if and which compensatory strategy could be effectively used.

Financial support

This work was performed with financial support from the European Commission Quality of Life and Management of Living Recourses Fifth Framework Programme. QLK1-CT-1999-0001.

Financial disclosure

None of the authors had any financial interest or support for this paper.