A Survey of In-Oil Canned Tuna Quality by Sensory Analysis and the Determination of the Oxidative Degradation of the Liquid Medium

Abstract

Sensory analysis and determination of analytical indices to evaluate the oxidative degradation of the liquid medium were used to assess the overall quality of tuna canned in different types of oil. Moreover, the relationships between the quality of the liquid medium and the quality of raw material were studied. The results pointed out that quality of the raw material changed as related to the types of oil used. In particular, the tuna in extra virgin olive oil showed significant higher scores for the descriptors related to the quality of raw materials and lower values for the indices that evinced the degradation of liquid medium.

Keywords:

• Canned tuna
• Sensory evaluation
• Oxidative degradation
• Vegetable oil

INTRODUCTION

For public health reasons, authorities in many countries promote the health benefits of seafood and encourage the consumption of marine products. Such campaigns, together with the recent image problems of the meat and poultry industries in health and animal welfare, have probably contributed to the increasing demand for seafood products.[1] In fact, fish is gaining ground as a protein source in the human diet. It is also noteworthy that it has low saturated fat and contains omega 3 fatty acids, known to support good health.[2] Nevertheless, modern lifestyles and new habits of consumption have had a marked influence on seafood consumption, and tendencies are different with regard to diverse types of seafood. Canned products have risen continuously since 1990. Among the species available in the markets, tuna is the commonest canned seafood product.[1]

Tuna cans may present different liquid media, like brine, oil, vinegar, or tomato.[3] Oil is one of the most used packing medium in canning manufacturing, having a preserving effect and contributing to make the product more palatable. Among the different types of oil, the most commonly used include olive oil (made up of refined olive oil added with virgin oil other than ‘lampante’ oil, in an undefined quantity) and refined seed oils, while extra virgin olive oil is seldom applied.

The international literature paid particular attention to the quality of canned tuna. The most recent papers pertain to the species identification of commercial canned tuna by DNA analysis,[3,4] in order to identify fish species and prevent possible fraudulent and vague labelling of the product, and to the assessment of the heavy metals content of canned tuna fish. The determination of phospholipids was used both for tuna species differentiation[5] and the evaluation of the quality of fish meal.[6] Histamine production by Enterobacteriaceae isolated from tuna was also investigated.[7] Moreover, some papers regard the assessment of the degree of oxidation of tuna lipids, by means of the determination of cholesterol oxidation products,[8] and TBA test and fluorescence measurement of compounds resulting from the interactions of carbonyl and amino groups.[9] Many investigations were carried out with the aim to assess the genuineness and quality of the oil medium used, evincing the occurrence of a diffusion of lipids from the fatty fraction of the fish into the liquid medium and that most oils used include oils resulting from a refining process; moreover, trans isomers were always absent in extra virgin olive oils, whereas they were always present in olive oils and in refined seed oils.[10–14] The antioxidant effectiveness of different packing media on tuna lipids was also verified,[9] and it was seen that the use of extra virgin olive oil may improve canned tuna quality, nutritional and sensorial as well. The aim of this paper was to assess the overall quality of tuna canned in different kind of oils. This was achieved both by sensory analysis and the determination of the analytical indices useful to evaluate the oxidative degradation of the liquid medium. Moreover, the relationships between the quality of the liquid medium and the quality of raw material were considered.

MATERIALS AND METHODS

Samples

Thirty-seven samples of preserved tuna were purchased in different retail stores: eight were canned in extra virgin olive oil, nineteen in olive oil (refined olive oil added with virgin oil other than ‘lampante’ oil) and ten in seed oils. All the samples were purchased about six months after their production.

Sensory Analysis

Quantitative descriptive analysis of the sensory properties was carried out by ten trained panellists, habitual consumers of in-oil canned tuna. The sensory appraisal of tuna canned in different oil types consisted of five phases, described as follows:

1. Identification of the descriptors qualifying the product “in-oil tuna.” This was the result of a survey conducted both among the quality system supervisors of some of the major Italian producers of in-oil tuna and among the usual consumers of this product. The identified descriptors concerned the evaluation of the colour, the flesh firmness, the organoleptic properties, and presence of defects (Table 1). For the descriptors salt taste, sweet taste, bitter taste, and rancid flavour standard solutions of each of them were used as reference; as regards the other descriptors a high quality in-oil tuna was used as reference.

2. Arrangement of a form for sensory analysis. For the sensory appraisal of the identified descriptors, a structureless 10-cm linear scale was applied. The score attributed by panellists to each parameter corresponded to the length (expressed in centimetres) of the line segment comprised between the origin of the scale and the dash indicating the perceived intensity.

3. Standardisation of a method for the preparation and tasting of samples. The samples to be tested for the sensory analysis should be as uniform as possible: the selected cans had the same capacity (80 g). The tasting method was standardised and subdivided into several steps: at the beginning of the panel session, each panellist was asked to open the can and carry on the visual analysis of samples to assess the colour uniformity and the aspect; then he/she was asked to smell the tuna that was still in the can. The panellist put then the sample in a dish and examined it for the other descriptors using a fork to observe internal portions as well. The subsequent step was the actual tasting.

4. Training of panellists. Prior to sensory evaluation, they participated in a cycle of lectures on the sensory attributes of in-oil tuna. During the lectures, references for the selected descriptors (Table 1) were defined and intensity ratings for these references were established by consensus of the panel.

5. Execution procedure of the panel test. The samples were marked with a numerical code of three figures attributed randomly, and the order of their tasting was decided randomly. Each sample was offered at room temperature and anonymously to all panel members. Panellists were asked to judge a sample at a time, and they had access to water and unsalted soda crackers to help cleanse their palates before tasting the subsequent sample.

Table 1 Descriptors identified for the quality assessment of in-oil canned tuna

Descriptors	Definition	Definition/Evaluation procedure
Flesh cohesion	Parameter related to the cohesion of flesh. The lowest score is attributed to very fragmented tuna	Visual assessment
Colour uniformity	Descriptor related to tuna colour homogeneity. The highest score is given to tuna whose colour is highly uniform	Visual assessment
Pink intensity	Parameter referred to the typical pink colour of tuna. The lowest score is attributed to the product that shows a pink shade quite far from the typical one	Visual assessment
Characteristic smell	Sensation concerning the typical odour of fish. The highest score is attributed to the product that has the typical and characteristic smell of tuna	Odour perceived when holding the uncut tuna close to the nose
Global flavour	Sensation relative to the overall organoleptic properties of canned tuna	Assessment relating to olfactory/gustatory perception
Oiliness	Parameter related to an over-soaking of tuna with the liquid medium. Minimum value = lack of defect	Assessment of the quantity of oil soaking the tuna, as perceived during mastication
Hardness	Parameter related to too hard tuna. Minimum value = absence of defect	Assessment of the force required to compress the product
Mealiness	Parameter concerning too friable tuna. Minimum value = absence of defect	Assessment of the force required to reduce a piece of tuna into smaller pieces
Salt taste	The minimum value is given to the product that does not show any defect	The intensity of the taste associated with a NaCl solution
Sweet taste	The minimum value is attributed to the product that does not show any defect	The intensity of the taste associated with sucrose solution
Bitter taste	The minimum value is given to the product that does not show any defect	The intensity of the taste associated with caffeine solution
Rancid flavour	The minimum value is attributed to the product that does not show any defect	Assessment relating to olfactory/gustatory perception linked to lipid oxidation
Damaged fish flavour	Flavour related to the presence of odours and tastes that are not characteristic of fresh fish. Minimum value = lack of defect	Assessment relating to olfactory/gustatory perception linked to off fish
Fish smell	Parameter concerning an excessive taste and smell of fish not necessarily damaged, but so strong as to be unpleasant. Minimum value = lack of defect	Odour perceived when holding the tuna close to the nose

Analytical Determinations

In the laboratory, for each sample the oil/dripped fish (w/w) ratio was determined. The oil recovered from each can, previously filtered on anhydrous sodium sulphate, was used for the determination of the level of trans isomers and of oxidative degradation. The trans isomers were determined by gas chromatographic analysis of methyl esters according to the official method.[15] The chromatographic system was composed of a Fisons HRGC gas chromatograph (Milan, Italy) with a SPTM 2340 fused silica capillary column, film thickness 0.20 μm, 60 m in length × 0.25 mm i.d., from Supelco (Bellefonte, PA, USA). The carrier gas was hydrogen; the temperature of the split injector was 210°C, with a splitting ratio of 1:100; the temperature of flame ionisation detector was 220°C. The oven temperature was programmed from 160 to 200°C with increments of 1.3°C/min and final isothermal of 15 min.

As for the oxidative degradation, the triacylglycerol oligopolymers and the oxidised triacylglycerols were determined by means of high performance size-exclusion chromatography (HPSEC) analysis of polar compounds separated from the oils by silica gel column chromatography according to the AOAC method.[16] CH₂Cl₂ was utilised as eluant at flow rate of 1 ml/min. The HPSEC system consisted of a series 200 pump (Perkin-Elmer, Norwalk, CT, USA) with Rheodyne injector, a 50 μl loop, a PL-gel guard column (Perkin-Elmer, Beaconsfield, UK) of 5 cm in length × 7.5 mm i.d., and a series of 3 PL-gel columns (Perkin-Elmer, Beaconsfield, UK) of 30 cm in length × 7.5 mm i.d. each. The columns were packed with highly cross-linked styrene-divinylbenzene copolymer with particle of 5 μm and a pore diameter of 500, 500 and 100 Å, respectively. The detector was a series 200 refractive index (Perkin-Elmer, Norwalk, CT, USA) connected to an integrator. Peak identification and quantification were carried out as described in previous papers.[14,17]

Statistics

One-way analysis of variance (ANOVA) was used to compare all the experimental data obtained for the canned tuna according to the types of oil used for their preservation. Factorial discriminant analysis was performed using S-Plus 6.1 (Insightful Corp., Seattle, WA, USA).

RESULTS AND DISCUSSION

As shown in Table 1, fourteen descriptors were identified to qualify the product “in-oil tuna.” Some of them concerned the evaluation of colour (colour uniformity, pink intensity); one of them regarded the evaluation of flesh texture (flesh cohesion); some others concerned the aromatic properties of the product (characteristic smell, global flavour); those remaining were necessary to evaluate the presence of defects (oiliness, hardness, mealiness, salt taste, sweet taste, bitter taste, rancid flavour, damaged fish flavour, fish smell). Table 2 reports the mean scores and the standard deviation of all descriptors used for the assessment of the sensory quality of canned tuna, subdivided according to the type of oil used as liquid medium, as well as the results of statistical analysis.

Table 2 Mean scores of the descriptors used for the assessment of the overall quality of canned tuna, subdivided according to the type of oil used as liquid medium and results of statistical analysis

Descriptors	Results of ANOVA	Tuna in extra virgin olive oil (TEVOO)	Tuna in olive oil (TOO)	Tuna in refined seed oil (TSO)
Flesh cohesion	^AB = p < 0.001	6.9 ± 1.3^A,a	6.3 ± 2.1^AB,a	4.7 ± 2.5^B,b
	^ab = p < 0.05
Colour uniformity	^AB = p < 0.001	7.4 ± 1.0^A,a	5.8 ± 2.1^AB,b	5.3 ± 2.1^B,b
	^ab = p < 0.01
Pink intensity	p < 0.05	6.9 ± 1.3^a	5.4 ± 1.2^b	6.0 ± 1.3^b
Characteristic smell	p < 0.05	4.8 ± 1.7^a	6.1 ± 1.5^b	6.3 ± 2.1^b
Global flavour	p < 0.05	6.2 ± 2.0^a	5.8 ± 1.3^a	4.9 ± 1.5^b
Oiliness	NS	1.7 ± 1.8	1.3 ± 1.2	1.6 ± 1.5
Hardness	p < 0.05	0.8 ± 1.0^a	0.6 ± 0.7^a	2.4 ± 3.4^b
Mealiness	NS	1.3 ± 1.5	1.7 ± 1.3	1.4 ± 1.5
Salt taste	NS	0.6 ± 0.7	0.5 ± 0.7	0.5 ± 0.8
Sweet taste	NS	0.4 ± 0.7	0.3 ± 0.4	0.3 ± 0.5
Bitter taste	NS	0.3 ± 0.7	0.2 ± 0.5	0.5 ± 0.6
Rancid flavour	p < 0.05	0.3 ± 0.6^ab	0.2 ± 0.3^a	0.5 ± 0.7^b
Damaged fish flavour	p < 0.05	0.2 ± 0.6^a	0.6 ± 0.8^ab	0.9 ± 1.2^b
Fish smell	p < 0.01	0.4 ± 0.8^a	1.0 ± 1.4^ab	1.6 ± 1.3^b
A-B, a-b: Different superscript letters in row indicate significant differences. NS: not significant (p > 0.05).

As to the descriptors properly correlated to the raw material quality, it results that for flesh cohesion and colour uniformity the mean values of the scores attributed by the panel to the tested samples were higher for the tuna in extra virgin olive oil (TEVOO) as compared to tuna in olive oil (TOO) and, notably, in seed oils (TSO), with significant differences at p < 0.001 only between TEVOO and TSO. However, although at a lower degree of significance, differences were observed between TOO and TSO for the flesh cohesion descriptor (p < 0.05) and between TEVOO and TOO for the colour uniformity descriptor (p < 0.01). These results pointed out that the type of oil used as liquid medium changed as a function of the type and quality of the raw material used for canned tuna production. In particular, with extra virgin olive oil a single piece of tuna with a quite uniform colour seems to be used, whereas in the case of olive oil and seed oil more crumbs and small pieces are used with a non-uniform colour, due maybe to a partial fish bleeding.[18,19] Moreover, also for the pink intensity descriptor the mean scores were significantly higher (p < 0.05) for TEVOO as compared to TOO and TSO. The hardness and mealiness descriptors are also correlated to the quality of the raw material used in the manufacturing of canned tuna; significant differences were found only for the first of the two descriptors. In particular, significantly higher mean scores (p < 0.05) were determined for TSO as compared to TEVOO and TOO. This might be due to the use of fish that had lost too much compositional water during cooking, thus resulting in a product that was chewy, with little flavour and absorbing a limited amount of oil during ripening.[18,19]

As to the characteristic smell descriptor, the mean values attributed by the panel were significantly lower (p < 0.05) for TEVOO than for TOO and TSO. This might be attributed to the extra virgin olive oil used as liquid medium of tuna and to its characteristic organoleptic imprint, which had partially attenuated the typical tuna odour. Nevertheless, this has not influenced the judgement expressed in terms of global flavour descriptor for which the mean scores were higher for TEVOO than for TOO and TSO, with significant differences (p < 0.05) comparing the scores obtained for TSO and those obtained for TOO and TEVOO. These results reflect an enhancement of preserved tuna sensory characteristics when using extra virgin olive oil rather than olive oil or seed oil as liquid medium. The damaged fish flavour and fish smell descriptors showed increasing mean values shifting from TEVOO to TOO and TSO, with differences that became significant when comparing TEVOO and TSO. These results do confirm the lower attention to the raw material quality when you have to produce preserved tuna using oils of lower commercial value and quality. As to the other descriptors considered, no significant differences were observed except for rancid flavour for which significant differences were found (p < 0.05) between TOO and TSO, with higher mean values for the latter.

Figure 1 shows the mean percent values and the results of statistical analysis of the oil/dripped tuna ratio (w/w) according to the different type of oil used. The data showed that the lowest mean value was observed for TEVOO (mean value of 36.6%), followed by TOO (mean value of 38.3%) and lastly by TSO (mean of 46.3%), with significant differences (p < 0.05) only comparing TEVOO and TOO with TSO. This might be attributed both to the greater commercial value of extra virgin olive oil as compared to olive oil (refined olive oil + virgin olive oil) and, especially, to refined seed oils, and to the use of a greater amount of crumbs and tuna small pieces, especially for TSO, that induce a greater presence of free spaces and hence require more oil. This assumption is in agreement, as previously described, with the scores attributed by panellists to the flesh cohesion descriptor. Moreover, in about 80% of TOO and TSO some water was found in the oil medium. This has concerned only about 40% of TEVOO, and might be attributed to an incomplete dripping of tuna before canning.

Figure 1 Mean percent values and results of statistical analysis (p < 0.05) of the oil/dripped tuna ratio (w/w) according to the type of oil used as liquid medium. TEVOO: tuna in extra virgin olive oil; TOO: tuna in olive oil; TSO: tuna in seed oils. A-B: Different superscript letters in row indicate significant differences.

Table 3 reports the mean percent values and the standard deviation of trans oleic isomers (C_18:1), of trans linoleic + trans linolenic (C_18:2+C_18:3), as well as the sum of triacylglycerol oligopolymers and oxidised triacylglycerols, subdivided according to the different types of oil medium of canned tuna. Trans isomers were found in traces in TEVOO, whereas in TOO and TSO the mean percent values of oleic trans were higher and statistically different (p < 0.05), with levels of 0.13% and 0.06%, respectively. In TOO the mean quantity of trans linoleic + trans linolenic isomers was equal to 0.13%, significantly lower (p < 0.001) than in TSO (mean value of 0.77%). The lower content of trans oleic isomers combined with the greater content of trans linoleic + trans linolenic observed in TSO as compared to TOO has to be correlated to the specific different fatty acid composition. Considering the negative implications of trans isomers of unsaturated fatty acids on consumers' health[20,21] their determination is a valuable analytical tool to assess oil quality.

Table 3 Percent mean values and standard deviation of trans isomers and sum of triacylglycerol oligopolymers (TGP) and oxidised triacylglycerols (ox-TG) of different oils used as covering oil in canned tuna

Parameters	Results of ANOVA	Tuna in extra virgin olive oil (TEVOO)	Tuna in olive oil (TOO)	Tuna in refined seed oil (TSO)
C18:1 trans	p < 0.05	tr	0.13 ± 0.06^a	0.06 ± 0.04^b
C18:2 trans + C18:3 trans	p < 0.001	tr	0.13 ± 0.07^a	0.77 ± 0.77^b
TGP + ox-TG	p < 0.001	0.77 ± 0.1^a	1.33 ± 0.31^b	2.22 ± 0.84^c
a-b: Different superscript letters in row indicate significant differences; and tr: traces (not integrated).

As to the level of oxidative degradation of oil medium in the tested canned tuna, assessed through the sum of triacylglycerol oligopolymers and oxidised triacylglycerols, it was significantly higher (p < 0.001) shifting from extra-virgin to olive and seed oils. The mean value was 0.77% for TEVOO, 1.33% for TOO and 2.22% for TSO. The triacylglycerol oligopolymers, stable substances due to the secondary oxidation of oils and fats, and the oxidized triacylglycerols, which presence is related to the products of primary oxidation, represent classes of compounds which levels can be useful indices to measure the real level of oxidative degradation of the edible fats.[14,22] Moreover, these classes of substances of triacylglycerol oxidation, are considered potentially harmful for consumers' health if assumed in large amounts.[23,24]

Figure 2 shows the results of the factorial discriminant analysis of the examined in-oil canned tuna. The analytical parameters utilised enabled to discriminate TEVOO, TOO, and TSO samples. In particular, TSO samples were linked to the descriptors identifying the defects of raw material, and to the sums of C_18:2 trans + C_18:3 trans and of triacylglycerol oligopolymers + oxidised triacylglycerols; on the contrary, TEVOO samples were linked to the descriptors identifying raw material quality (colour uniformity, pink intensity, flesh cohesion) and global flavour; finally, TOO samples showed intermediate characteristics and were linked to C_18:1 trans.

Figure 2 Factorial discriminant analysis of the tested samples. E: tuna in extra virgin olive oil; O: tuna in olive oil; and S: tuna in refined seed oil.

CONCLUSIONS

The results of the analytical determinations and the sensory analysis allow us to remark that the quality of the raw material used for preserved tuna production changed according to the type of oil used as liquid medium; at any rate, the use of extra virgin olive oil enhanced the organoleptic properties of preserved tuna. Among the different typologies of preserved tuna, the descriptors showing significant differences were pink intensity, flesh cohesion, colour uniformity, and hardness, which are properly correlated to the raw material quality; damaged fish flavour, fish smell, and rancid flavour, which are descriptors identifying the defects; finally characteristic smell and global flavour. Tuna preserved in olive oil and refined seed oil showed higher scores for the descriptors concerning the presence of defects, whereas tuna in extra virgin olive oils showed higher scores for the descriptors related to the colour and flesh cohesion. Trans isomers were absent or present in traces in extra virgin olive oil used as medium, while they were constantly present both in olive and refined seed oils; finally, the oxidative state of preserved tuna liquid medium was significantly higher in refined seed oils as compared to olive and extra virgin olive oils.