Effect of inulin on the textural and sensory properties of mortadella, a Spanish cooked meat product

Abstract

Inulin is a fructooligosaccharide with demonstrable beneficial effects on health. Its effect on the textural and sensory properties of mortadella, a Spanish cooked meat product, was studied. Conventional (23% fat) and reduced-fat sausages (10% fat) were prepared. Fat reduction yields an energy value reduction close to 33%. In both cases, inulin was incorporated, in powder and as gel form, in sufficient amounts to constitute the 2.5%, 5% and 7.5% of the final product. To evaluate the effect of inulin, instrumental measures of texture and sensory analysis (hedonic test) were performed. Textural analysis indicated that powdered inulin increases hardness and this fact was more evident in reduced-fat sausages which showed that this changes even at concentrations of 2.5%. However, when inulin was incorporated as a gel, textural parameter changed only when the high levels were assayed (7.5%) showing softer sausages independent of the fat content. Sensory analysis was highly favourable and the overall acceptability was good in all batches in spite of the changes observed in texture. It can be established that this product can be enriched with inulin to a maximum level of 7.5% and preferably as gel with a good sensory quality.

Introduction

Inulin is the common name for fructooligosaccharide containing a degree of polymerisation of up to sixty monomers of fructose bound by a β-2-1 glycosidic linkage. The major dietary sources of inulin include the Jerusalem artichoke and chicory, which contain sufficient amounts to be used as commercial sources of fructooligosaccharides. Chicory roots contain up to 70% inulin on a dry matter basis (Gibson & Roberfroid, 1995). Its β-configuration makes inulin non-digestible to hydrolysis by human digestive enzymes, even those of the small intestine. Therefore, undigested inulin reaches the large intestine, the most heavily colonised region of the gastrointestinal tract (10¹² bacteria g⁻¹). Bifidobacteria ferment the inulin and produce a wide variety of compounds that affect the intestine and the systemic physiology: short-chain carboxylic acids (acetate, propionate, butyrate) and lactate that play an important role in regulating cellular metabolism as well as cell division and differentiation (Cummings, 1997; Roberfroid & Slavin, 2000). The consequent lowering of pH allows a suitable medium for the development of the bifidogenic flora and limits the growth of those considered pathogenic. As a result, inulin is considered prebiotic (Flamm et al., 2001).

With regard to other physiological functions, inulin reduces the risk of arteriosclerosis; it increases the bio-availability of several minerals such as Ca, thus increasing the mineral density (Frank, 1999; Roberfroid, 1999) of the bone and preventing osteoporosis (Anonymous, 2000; Frank, 2000); it reduces the risk of gastrointestinal diseases Alldrick (2000); it lowers levels of cholesterol and triglycerides in the serum (Carter et al., 1998; Jackson et al., 1999; Desmedt & Jacobs, 2001); it controls the levels of glucose in blood (Carter et al., 1998; Jackson et al., 1999); it reduces the liberation of toxins and carcinogens; and, finally, inulin stimulates the immune system (Desmedt & Jacobs, 2001; Meyer et al., 2001; Shah, 2001).

As inulin is a non-digestible oligosaccharide, it is not absorbed in the small intestine and, therefore, it has no caloric value in the traditional sense. However, because of colonic fermentation, inulin contributes to the energy pool. For nutritional labelling purposes, it has been assigned a caloric value ranging between 1 and 1.5 kcal g⁻¹ (Boeckner et al., 2001; Desmedt & Jacobs, 2001; Flamm et al., 2001).

Currently, inulin is an accepted food ingredient. It has a unique ability to form discrete highly stable particle gels and contribute to the rheological and textural properties of foods. Inulin gel is composed of a three-dimensional network of insoluble submicron crystalline in water. Its characteristics depend on temperature, concentration and the degree of polymerisation but, in general terms, inulin gels result in a white creamy fat-like structure which can easily be incorporated into foods. Thus it has been used to replace fat by up to 100% (Boeckner et al., 2001; Desmedt & Jacobs, 2001).

Inulin is considered to be a functional food ingredient and its industrial use, in human foods, include fat substitution (sauces, milk products like cheese and yoghurt, ice creams, candies, etc.), reduction of caloric value (sugar-free chocolate, meal substitutes), water-holding ability (bakery goods), emulsions (margarine) and, in general, it is used to modify the texture and viscosity of foods (Boeckner et al., 2001; Shah, 2001).

The incorporation of inulin in meat products has not been studied extensively. Some authors have studied inulin as a fat substitute in low-fat dry fermented sausages (Mendoza et al., 2001) and have obtained acceptable products from a sensorial point of view. Only a few research studies have been performed on the effect of inulin in cooked meat products.

On the other hand, fat is one of the main components of food and contribute to their texture and flavour and increases the feeling of satiety during meals. For these reasons, fat reduction implies technological and commercial problems in the manufacture of meat emulsions so as to minimise the sensory and texture modifications occurring as a result of the impact of fat reduction (taste, colour). Inulin should be a good fat replacer for these meat products, and its presence should contribute, as does dietary fibre, to the physiological benefits previously described.

Therefore, the objective of this study was to determine the influence of inulin on the textural and sensory properties of conventional and reduced-fat mortadella, a Spanish bologna-type cooked meat product.

Material and methods

Sausage manufacture

Two experiments were performed in which mortadella was manufactured according to the traditional formula. In experiment 1, the conventional formula was used with the habitual fat level; accordingly, the percentage of ingredients was the following: 46% pork meat, 39% pork fat (jowl), 10% ice water and 5% a commercial mixture of spices and additives (Procavi ‘C50’. Anvisa, Arganda del Rey, Madrid, Spain) specially prepared for this type of meat product. The mixture contained salt, dextrose, soya proteins, spices, glutamic acid, emulgents and gelificants, antioxidants, potassium nitrate, trisodium citrate and carminic acid as colourant. In experiment 2, the jowl percentage was reduced to 15%, representing a fat reduction of close to 44% (Table 1). The percentage of pork meat was increased to 70%; water and spices were added at the concentrations used in experiment 1 (Table 1). The selection of these amounts was based on previous studies in our laboratory (unpublished data) in which no technological problems were observed.

The meat ingredients were obtained from a local abattoir and frozen separately at −18 °C. Before use they were kept at 4 °C for approximately 18 h. The block of meat and the fat were chopped in a high-speed (1500 rpm) bowl (Robot-Coupé Mod 20 v, Jackson, MS, USA) for 1 min under partial vacuum (500–700 mbar). Then, one-half of the ice water and one-half of the spice mixture were added, and the mix was chopped for an additional 2 min, approximately. The remaining ice water and the spices were then added and the batter was mixed until emulsion was complete. Temperature did not exceed 2 °C at any time.

In both experiments, seven batches (2 kg per batch) were prepared: one was manufactured without inulin (control batch) and in the other six batches Raftiline ST© (Orafti Ltd, Oreye, Belgium), with 90–94 g per 100 g of dry matter of inulin, was added. In three of these batches, inulin was incorporated directly as powder mixed with the spices in a quantity sufficient to constitute 2.5%, 5% or 7.5% of the final product. In the other three batches, inulin was incorporated as a gel prepared mixing inulin (55 g) and water (100 mL). Thus, a soft gel was obtained, easy to use at temperatures batters manufacture. Different amounts of this gel were added so as to provide the desired inulin final content: 4.5, 9.1 and 13.6 mL for batches that have 2.5%, 5% or 7.5% of inulin respectively. Gel was incorporated at the same time that of the ice water.

Batters were handstuffed into five pieces of 400 g using 85-mm diameter plastic water impermeable casings (Multibar) (Oskuda GMBH, Georgsmarienhütte, Germany). The mortadella was cooked in a steam oven at a temperature of 78 °C for 35 min, sufficient to reach 72 °C in the interior of the product. The temperature was measured with an Interface EBI-AE 2000 (Ebro Electronic GmbH & Co., D. Ingolstadt, Germany). Finally, sausages were cooled by immersion in a cool water bath and maintained under refrigeration until their analysis.

All batches were manufactured in duplicate.

Chemical analysis

Fat content was determined according to the Soxhlet method (No. 948.22), protein (Kjeldhal nitrogen, No. 991.30), moisture (oven air-drying method No. 925.10) and ash (muffle furnace method No. 920.100) were analysed following AOAC procedures (AOAC, 1997). Carbohydrates were calculated by difference. All determinations were performed in quintuplicate.

Energy values

Total calories (kcal) were calculated for 100 g samples using the Atwater values corresponding to fat (9 kcal g⁻¹), protein (4.02 kcal g⁻¹) and carbohydrates (3.87 kcal g⁻¹).

Textural analysis

The Textural Profile Analysis (TPA) test (Bourne, 1978) was used to evaluate sausages using the texturometer TA.XT 2i/25 (Stable Micro System, London, UK) at room temperature. Central cores of five transversely cut slices of each sausage (1.5 cm high and 2.5 cm diameter) were horizontally placed on the heavy-duty platform HDP/90 of the texturometer. Samples were compressed twice to 50% of their original height. A crosshead of 2 mm s⁻¹ was used. The following parameters were determined: hardness (N cm⁻²), maximum force required to compress the sample (H); springiness (cm), ability of sample to recover its original form after deforming force was removed (S); cohesiveness, extent to which sample could be deformed prior to rupture (A2/A1, A1 being the total energy required for the first compression and A2 the total energy required for the second compression); adhesiveness (N s), work necessary to pull the compressing plunger away from the sample; gumminess (N cm⁻²), force to disintegrate a semisolid meat sample for swallowing (H × Cohesiveness); chewiness (N cm⁻¹), work to masticate the sample for swallowing (S × Gumminess).

Shear analysis was performed with a Warner–Bratzler reversible blade HDP/BS using the same texture analyser. Samples were also five cores (1.5 cm high and 2.5 cm diameter) of each sausage prepared as it has been described for the TPA. The crosshead speed was 2 mm s⁻¹. Maximum force to cut the sample (shear force, N) and the work needed to move the blade through the sample (work of shearing, N s) were recorded.

Sensory analysis

A hedonic test was performed in which a total of sixty untrained panellists evaluated the experimental sausages. Panel members were recruited from faculty, students and staff of our department. All of them were habitual consumers of these types of meat products. Tests were carried out using non-structured 10-cm scales in which the panellists evaluated pleasantness level of different attributes: odour, colour, texture, taste and overall acceptability (0 = very unpleasant and 10 = very pleasant).

Hardness, juiciness and cohesiveness were also evaluated (quantitative descriptive analysis) by twelve trained panellists. They were five male and seven female laboratory co-workers and ranged in ages from 34 to 60 years. Although all of whom had previous sensory panel experience, they received further training prior to the analysis according to the method described by Steenblock et al. (2001). Three open-discussion sessions of 1 h approximately were held so as to familiarise the panellists with the attributes to evaluate. Training was performed using different commercially cooked meat products with several qualities that permit to establish a range of intensity. When it was necessary, experimental sausages were manufactured in which composition was conveniently modified so as to force the sensory characteristics of the product. The scales used for the evaluation of texture parameters were the following: hardness (0 = very soft and 10 = very hard), juiciness (0 = very dry and 10 = very juicy) and cohesiveness (0 = low cohesion and 10 = high cohesion).

Both analyses were performed under white fluorescent lights in individual booths constructed according to the specifications of the International Standards Organisation (ISO) DP 6658 (ISO, 1985). Each panellist received two slices of the samples (85 mm diameter, 2 mm high approximately) which were maintained in hermetic transparent containers and served at room temperature. Unsalted crackers and water at room temperature were also provided to clean the palate between samples.

Two sessions per day were conducted in which panel members were served randomised samples in each test session with a minimum 1-h break between sessions to reduce fatigue.

As the sausages were manufactured in duplicate, a sensorial analysis was performed for each one.

Statistical methods

Experiments 1 and 2 were analysed independently. Textural properties were studied by using a one-way analysis of variance independently for each experiment. The factor represented the fibre percentage with two replicated measurements for each percentages of fibre. In each experiment, the Duncan Test was employed to determine the differences between the mean values of the batches with added fibre and between those batches and their corresponding controls. Significance level was established at P < 0.05.

The effect of the panellists in the descriptive test was removed by rescaling all the scores given by each panellist from 0, which represents the minimum score used by a given panellist, to ten, which is the maximum score used by that panellist. After rescaling, the effect of the panellists in the sensorial responses was analysed by a two-factor analysis of variance according to a complete randomised block experimental design. The factor represented the percentage of fibre, while the block variable had twelve levels representing each of the panellists that collaborated in the trials. The F-test showed that the effect of the panellists was not significant (P > 0.05). As a consequence, the final model considered only one factor (one-way Anova) representing the content in fibre with twenty-four replicates for each percentage. A Duncan test was also employed to determine the differences between the batches with different percentages of fibre.

A principal components analysis (PCA) was performed on the correlation matrix of both the texture parameters and the standardised sensorial responses. This analysis was carried out with the average textural properties (or sensorial responses) for the sausages of both experiments 1 and 2. The results of the PCA were represented as biplots where both the batches (PC-scores) and attributes (PC-loadings) were plotted. The statistical package Statgraphics Plus 5.0 (Statistical Graphics Corporation, Herborn, VI, USA) for Windows was used.

Results and discussion

Sausage composition

Table 2 shows the approximate composition of control and inulin-added batches of both experiments. The differences between the composition of batches belonging to experiments 1 and 2 reflect only the modifications used in the original formula to manufacture reduced-fat sausages, i.e. the increase in water and protein because of the substitution of pork fat by pork meat. These differences also reflect the incorporation of fibre because the percentage of sausage components decreases according to the increase in fibre incorporated.

The average fat content value among all treatments (including the control) was 22.45% in conventional batches (Experiment 1) and 10.10% in reduced-fat ones (Experiment 2). This difference represents the expected fat reduction, an average value close to 44%. These changes in the final composition have been observed by other authors in similarly cooked meat sausages manufactured with cellulose gums (Mittal & Barbut, 1994; Barbut & Mittal, 1996) and in our laboratory when other fibres (fruit fibre and synthetic oligosaccharides) were added (unpublished data).

Caloric value

The energy value shows the expected decrease in relation to the fat reduction levels. In this sense, the caloric value decreased from an average value of 256.7 kcal per 100 g in the conventional batches (experiment 1) to 172.5 kcal per 100 g in reduced-fat products (experiment 2). This difference represents an average energy reduction close to 33% (Table 2). In both the experiments, there was an inverse relationship between the inulin level and the caloric value. Bearing in mind that inulin contributes only with 1.5 kcal g⁻¹, it can be said that its contribution to the final energy value is very low. Similar results has been obtained by other authors using inulin in dry fermented sausages (Mendoza et al., 2001).

Textural analysis

The results of the textural profile analysis in experimental sausages are shown in Table 3. In sausages made with conventional fat content, the incorporation of powdered inulin did not translate into any change in springiness and cohesiveness, which showed similar values to the control (P > 0.05). Nevertheless, a general increase in adhesiveness (P < 0.05) was observed and in hardness, 7.5% powder sausages were significantly harder than the control and the other inulin powder concentrations assayed (P < 0.05). When inulin was added in gel form it did not modify the springiness or the adhesiveness of the control batches; however, cohesiveness and hardness were modified at the highest concentrations: the sausages were significantly softer and had less cohesiveness than the control batch (P < 0.05). Gumminess and chewiness are secondary parameters and showed a behaviour similar to the primary parameters they depend. Therefore, chewiness tends to decrease as the concentration of the inulin in the gel increases, that is to say, the softer the sausages are, the easier it is to chew them.

These data seems to indicate that the 7.5% inulin level is like a inflexion point, because, if it is incorporated as powder, this amount is sufficiently high to make the product harder. However, if it is incorporated as a gel, inulin transfers its creamy characteristic to sausage and makes it softer.

The data were analysed using covariance PCA, comprising a matrix of average scores. Figure 1 is a projection of TPA data of the first two principal components of the PCA developed for the conventional sausages. PC 1 represents 75.5% of variability and PC 2 18.5%. This implies that both PC collect the 94% of the total variability of the data. The remainder 6% is so low that can be discarded and the data showed in two dimensions. Accordingly, to interpret the data, PC 1 is the most important principal component and it includes cohesiveness, hardness and springiness in contrast to adhesiveness. This last attribute shows the largest vector and so it is the most important. PC 2 is mainly correlated with adhesiveness and cohesiveness.

The batches were grouped into two well-differentiated clusters. One of them included batches manufactured with inulin as powder (full symbols) and it is characterised by its cohesiveness, hardness and springiness. The second cluster included those manufactured with inulin as gel (empty symbols), in which adhesiveness is the most important parameter. It can be observed as 7.5% Gel batch is the most distant from the position of the attributes hardness and cohesiveness which indicates that this batch is softer showing at the same time the lower cohesiveness (Table 3).

In experiment 2 (Table 3), when inulin is incorporated in powder form, the sausages were, in all the concentrations assayed, significantly harder than the control batch (P < 0.05). This indicates that powdered inulin tends to harden the product. This effect is more evident in the reduced-fat batches in which the fat level is reduced which indicated that the fat mask the hardening effect of inulin powered and so it can only be observed at the higher concentration (7.5%) in conventional batches. When it is incorporated in gel form, the sausages become softer, especially at 7.5% (P < 0.05). It is noted that in all reduced-fat batches, springiness is lower than in the control (P < 0.05). The adhesiveness was not changed and cohesiveness showed some unimportant changes. Gumminess and chewiness showed a tendency to decrease in the batches made with gel. This is significantly different from the control, especially with respect to chewiness and is due mostly to hardness.

Similar results have been described by several authors in relation to the addition of dietary fibre of different origins (wheat, barley, fruit) in low-fat cooked meat products (Mansour & Khalil, 1997; Griguelmo et al., 1999; Shand, 2000) and inulin in dry fermented sausages (Garcia et al., 2002). In these studies, the authors also describe a decrease in hardness related to the presence of fibre incorporated as aqueous solution.

These data were also analysed using covariance PCA (Fig. 2). Principal component 1 collects 49.3% of the total variability of the data while principal component 2 collects 33.9%. The remainder 16.8% corresponding to the third and fourth principal components were discarded and the data showed in two dimensions. Hardness shows the largest vector and so it is the most important attribute. PC 1 is positively correlated to cohesiveness but not to adhesiveness and hardness. PC 2 is mainly correlated to cohesiveness but not to adhesiveness and hardness. Two clusters can also be observed in which batches are grouped according to the type of inulin used. Batches with powdered inulin (full symbols) are grouped by their hardness and adhesiveness while those with inulin as gel were grouped mainly because of their cohesiveness.

Table 4 shows the results obtained from the shear test. In general terms, the data obtained show that both the shear force and the work of shearing were very similar in all cases. However, it is important to make several qualifications. In the case of the conventional batches, the shear force was similar in the control and in the experimental batches. No significant differences were observed. However, work of shearing increases at the higher concentrations of powdered inulin and decrease as inulin gel levels increase. This behaviour agrees with that observed in the TPA (Table 3). However, in the reduced-fat batches, the situation was different because no significant differences were observed (P > 0.05) either for shear force or for work of shearing. These results are in contrast to those found for similar meat products by other authors who observed how the presence of fat or dietary fruit fibre (Cross et al., 1980; Berry & Leddy, 1984; Griguelmo et al., 1999) caused an increase in the shear force. These differing results could simply be caused by the structure of the gel formed by the fibre used in this study. Fruit fibres have a percentage of insoluble fibre that has a lower capacity to retain water. However, inulin has a high capacity to bond with water and can form gels that are firm, soft and stable all at the same time. This stability makes the gel bond with a greater amount of water thus making it easier to cut the product and reproducing the sensations that fat gives this food.

Sensory analysis

Table 5 shows the effect of inulin in the sensory properties of conventional and reduced-fat sausages. In general terms, a similarity between the panellists’ scores for the different parameters can be observed.

In the conventional batches, no significant differences were found in the majority of the pleasantness attributes evaluated, with panellists noting a very respectable level of overall acceptability for all the samples (P < 0.05).

With respect to hardness, the panellists considered that 7.5% Gel batch was the softer even than the control batch (P < 0.05), as it can be expected by the results obtained in TPA (Table 3). Juiciness was quite similar in all batches. Perhaps it is possible to observe that juiciness decrease when inulin is added as powder and slightly increase when it is added as gel, in both cases at the concentration of 7.5%. No significant differences were observed in relation to the control. It is possible that the gel of inulin influence in the perception of juiciness by the level of water that it physically entraps while this situation does not occur when the inulin is incorporated as powder. When judging cohesiveness, panellists only found significant differences between the 7.5% powder inulin and the control batch. This fact indicates that inulin as powder could modify the cohesiveness when it is added at this concentration. Despite this, the overall acceptability of the products were good.

Likewise, similar results were observed for reduced fat batches. The majority of the pleasantness attributes were evaluated favourably and in a very similar way in all the batches being again 7.5% Gel batch which shows the highest scores in the overall acceptability. In relation to the hardness, batches manufactured with inulin as gel were significantly softer than the control batch which indicate that inulin as gel is able to soften the product achieving similar values to the hardness of the conventional batches. As in experiment 1, juiciness increased with the levels of insulin as gel while cohesiveness showed similar values in all batches (p>0.05).

A broad examination of these data tells us that inulin is well accepted from the sensorial point of view even in amounts until 7.5% of the final product and this acceptability is valid both for manufactured meat products with normal fat levels and those with reduced-fat.

The PCA performed from the sensorial analysis are shown in Figs 3 and 4. In relation to the hedonic test (Fig. 3), PC 1 receives the majority of total variability (51%) while PC 2 receives 27%. The remainder 22% of the variability was discarded and correspond to the third and fourth principal components. First component was mainly represented by odour, texture and taste. These two last sensory attributes, contrary to colour, showed the largest vectors which indicate that they are the most important parameters. Second component was mainly represented by colour. Two clusters were established; conventional batches were grouped by texture and taste while the reduced-fat ones were grouped by colour.

The PCA from the descriptive test (Fig. 4) revealed that the 93% of the variability was accounted for by the two first principal components. PC 1 (65%) is related to juiciness and hardness but these two variables are inversely correlated. PC 2 (28%) is mainly related to the cohesiveness. The remainder 7% of the variability was discarded. In this case, the batches were grouped in relation to the form in which inulin was incorporated. One group collected the batches manufactured with inulin as gel and they were characterised by the juiciness. The other group collected those manufactured with inulin as powder being characterised by their hardness.

These results indicate that inulin could be incorporated into conventional or reduced-fat cooked meat products, as powder or as gel. This last option can be considered as the most adequate because of the juiciness and the softness that is provided to the product.

Research into inulin in reduced-fat dry fermented sausages has produced similar results even to the extent that sausages of up to 11.5% inulin (Mendoza et al., 2001) were sensorially acceptable, because at this concentration the fibre gives the product a softer texture and sensory properties similar to those of conventional fermented sausages.

Other soluble fibres have been assayed by different authors in cooked meat products, such as oat or fruit fibre (Chang & Carpenter, 1997; Griguelmo et al., 1999). The results obtained were different because in both cases a considerable increase in the hardness of the product was described.

Conclusions

It is possible to successfully manufacture cooked meat products like mortadella with inulin preferably as a gel until the constitute is 7.5% of the final product without significant changes in the overall acceptabe pleasantness. Because inulin lends a softer sensation that simulates the fat mouthfeel, it is feasible to incorporate this fibre into hypocaloric meat products. In addition, the presence of inulin enriches these new meat products in a dietary fibre that have a favourable effect on the health.

Acknowledgments

Thanks are expressed to Dr C. Pin for the statistical analysis. This work was supported by the Spanish Ministerio de Educación, Cultura y Deporte (CICYT) Project AGL 2001–1227 and AGL 2004–04154.

References