S: Sensory & Food Quality

Transcription

1 Effect of Modified Atmosphere Packaging on Index Method (QIM) Scores of Farmed Gilthead Seabream (Sparus aurata L.) atlow and Abused Temperatures Marco Campus, Elia Bonaglini, Roberto Cappuccinelli, Maria Cristina Porcu, Roberto Tonelli, and Tonina Roggio Abstract: A Index Method (QIM) scheme was developed for modified atmosphere packaging (MAP) packed gilthead seabream, and the effect of MAP gas mixtures (60% CO 2 and 40% N 2 ; 60% CO 2, 30% O 2, and 10% N 2 ), temperature (2, 4, and 8 C), and time of storage on QI scores was assessed. QI scores were crossed with sensory evaluation of cooked fish according to a modified Torry scheme to establish the rejection point. In order to reduce redundant parameters, a principal component analysis was applied on preliminary QIM parameters scores coming from the best performing MAP among those tested. The final QIM scheme consists of 13 parameters and a maximum demerit score of 25. The maximum storage time was found to be 13 d at 4 C for MAP 60% CO 2 and 40% N 2.Storageat 2 C do not substantially improved sensory parameters scores, while storage under temperature abuse (8 C) accelerated drastically the rate of increase of QI scores and reduced the maximum storage time to 6 d. Keywords: farmed gilthead seabream, modified atmosphere packaging, quality index method, sensory analysis Introduction The farming of gilthead seabream (Sparus aurata L.) in the European Union account to 67% of total fish production (Ernst and Young, AND-I, Eurofish, Indemar 2008) with the demand increasing every year. In parallel, many new farms are entering in what is becoming a very competitive market, making the evaluation of farmed fish quality very important. Consumer satisfaction is usually based on overall quality of fish as assessed directly by consumers through appearance, texture, and taste (Reineccius 1990). Up to recent years, the most widely used scheme to evaluate fish freshness in the European Union was the quality grading scheme (Anonymous 1996), based on a rating system that expresses a comprehensive level of product compared to a standard. There are 3 assessment levels: E, A, and B, with E indicating the highest quality, while fish rated below B is considered unfit for human consumption. The major limitation of this method is its lack of specie specificity, while it is well known that sensory characteristics may vary tremendously among various species, and the modifications during storage as well (Hyldig and others 2007). Index Method (QIM) is a structured scale based upon a scheme originally developed by Tasmanian Food Research Unit (Bremner 1985), and used to assess the freshness of fish utilizing significant sensory parameters and a score system of demerit points. The scores for all the characteristics are summed-up to give an overall sensory score, the so-called QI (Botta 1995). A QI of 0 indicates a very fresh fish and score increases as the freshness s characteristics lapses. The choice of parameters is determined by the combination of best MS Submitted 7/14/2010, Accepted 12/13/2010. Authors are with Porto Conte Ricerche Srl, S.P. 55 Porto Conte/Capo Caccia, Km , Loc. Tramariglio, Alghero (SS) 07041, Italy. Direct inquiries to author Campus ( campus@portocontericerche.it). descriptors related to deterioration of fish, relatively to its species. The purpose of the QIM is to establish a more accurate scheme for production management and fish inspection (Hyldig 1998). Also, the increasing tendency to the e-commerce of fish calls for an harmonization of accurate sorting and grading systems between the markets within the European Union and interconnected auctioning instruments (Vader and others 2003). QIM for different species has been developed, including cod (Gadus morhua), frozen and filletted (Warm and others 1998), mediterranean hake (Merluccius merluccius) (Baixas-Nogueras and others 2003), frozen hake (Merluccius capensis and Merluccius paradoxus) (Herrero and others 2003), farmed atlantic salmon (Salmo salar) (Sveinsdóttir and others 2002), tub gurnard (Chelidonichthys lucernus) (Bekaert 2006), raw and whole herring (Clupea harengus) (Martinsdóttir and others 2001), whole scad (Trachurus trachurus)(inácio and others 2003), processed herring (Clupea harengus), stored both in air and under modified atmosphere packaging (MAP) (Lyhs and Schelvis-Smit 2005), and raw gilthead seabream (Huidobro and others 2000, 2001; Alasalvar and others 2000). MAP, either alone or in combination with other hurdle technologies (Drosinos and others, 1997; Özogul and others 2004; Goulas and Kontominas 2007; Tsironi and Taoukis 2010), has become an increasingly popular preservation technique, bringing major changes in storage, distribution, and marketing of raw products to meet consumers demand, showing positive effects on fish shelf life (Sivertsvik and others 2002). Carbon dioxide is the most important gas used in MAP due to the bacteriostatic and fungistatic effect. Carbon dioxide is highly soluble in water and fat and solubility is higher at lower temperature. Moreover, microorganism growth inhibition in MAP is dependent upon CO 2 concentration in the product (Devlieghere and others 1998a, 1998b). The effect of CO 2 inhibition on bacterial growth is complex and includes alteration of intracellular ph, changes in physicochemical properties of proteins, alteration of cell membrane, and direct inhibition of enzyme activity (Parkin C 2011 Institute of Food Technologists R doi: /j x Further reproduction without permission is prohibited Vol. 76, Nr. 3, 2011 Journal of Food Science S185

2 and Brown 1982; Daniels and others 1985; Dixon and Kell 1989; Farber 1991). It has been demonstrated that packaging in CO 2 leads to the selection of CO 2 tolerant anaerobes (Sivertsvik and others 2004; Pournis and others Different gas mixtures have been proposed for MAP of stored fish, with CO 2 concentrations ranging from 11.5% (Nelson and Tretsven 1977), 20% (Brown and others 1980), 25% (Villemure and others 1986), and up to 100% CO 2 content. However, in most cases, CO 2 levels higher than 25% have been found to be necessary. Gas compositions in the headspace of 30% O 2, 40% CO 2, and 30% N 2 or 40% to 60% CO 2 and 60% to 40% N 2 have been proposed for low fat fish and for fatty fish, respectively (Sacks and Gore 1987; Robertson 1993; Davis 1998; Pastoriza and others 1998). Debevere and Boskou (1996) identified a mixture of 40% O 2 and 60% CO 2 as the most effective for the inhibition of trimethylamine (TMA) production in cod fillets. On the other hand, Torrieri and others (2006) showed that a mixture of 30% O 2 and 50% CO 2 results in the best sensory scores of gutted farmed bass packed under MAP. Recommendations in the use of O 2 in MAP are due to the beneficial influence in spoilage microflora, as assessed in white fish (Tiffney and Mills 1982; Sacks and Gore 1987). Boskou and Debevere (1997) suggested a combination of 60% to 70% CO 2 and 30% to 40% O 2 to prevent TMA production by Shewanella spp., basing on test carried out in vitro. The usefulness of QIM is to determine how different ways of handling, processing, and storage conditions affect the fish shelf life such as under MAP and temperature abuse during storage (Martinsdóttir and others 2003). The QI is not only species related, but also related to the processing technology adopted, so different schemes should take into account products specific differences (Lyhs and Schelvis-Smit 2005). Although a QIM scheme has been proposed for raw gilthead seabream (Huidobro and others 2000), it remains to be fully defined a QIM scheme specifically designed for gilthead sea bream packed under MAP. The objective of the present work was to develop a QIM scheme specific for gilthead seabream packed in MAP and to investigate how the packaging technique affects the QIM scores during chilled storage and under temperature abuse. The scheme should allow to predict the residual shelf life using simple sensory evaluations. Tests performed were aimed to develop a preliminary QIM scheme relying on several sensory parameters. A principal component analysis (PCA) was performed to reduce redundancies, and the scores coming from the best performing MAP were used to elaborate the final QIM scheme. The scheme was then used to assess the quality of MAP packed gilthead seabream at different storage temperatures, 2 C, and under temperature abuse (8 C). Materials and Methods Samples Farmed gilthead seabream were obtained from a fish farm located in northwestern Sardinia (Alghero, Italy). A total of 72 fish (18 mo of age) were used in 2 independent trials performed across November 2009 and February Average weight and standard length were g and cm in the 1st trial and g and 21.9 cm in the 2nd trial, respectively. Fish were killed by dipping in ice-cold water (hypothermia). After death, the fish were transported to the laboratory in melting ice within 2 to 3 h from harvesting. Modified atmosphere packaging and storage conditions Samples were packed individually in PP trays using a Retray 250 (Reepack, Packaging Machinery Supplies, Seriate, Italy) and sealed with a plastic film with gas barrier and antifog properties (multilaminated PET/PE/EVOH/PE Cxtop64, Coopbox, Italy; O 2 transmission rate: 1.5 cm 3 /m 2 to 24 h/atm; CO 2 transmission rate 5 cm 3 /m 2 to 24 h/atm; N 2 transmission rate <1 cm 3 /m 2 to 24 h/atm; water vapor transmission rate: 5.5 g/m 2 atm, at 23 C and 0% relative humidity). Gas to fish volume ratio was 2:1. Two independent storage experiments were performed. In the 1st, 2 gas concentrations were used to package the samples: 18 samples were packaged in 60% CO 2 and 40% N 2 (MAP1), while 18 samples were packaged in 60% CO 2, 30% O 2, and 10% N 2 (MAP2). Immediately after packaging, gas concentrations were measured with a Combicheck gas analyzer (PBI Dansensor, Denmark), and samples were stored at 4 Cinthedark. The 1st experiment was aimed to pick the best performing MAP among the 2 selected (MAP1 and MAP2), reduce the number of redundant parameters and develop the final QIM scheme. A 2nd experiment was aimed to assess the effect of storage temperature over QI scores in samples packed under the best performing MAP, according to the results of the 1st trial: 36 samples were gas packed and divided in 2 groups, each of 18 samples. The 2 groups were stored, respectively, at 2 and 8 C, in the dark. Development of QIM scheme for MAP packed gilthead seabream Sensory assessment was performed on days 0, 1, 3, 7, 10, and 13. A completely randomized sampling scheme was adopted coding samples using random number tables. Samples were evaluated under natural day light conditions in a climatized room at 22 C. At each sampling time, 3 fish for each condition tested were randomly chosen to evaluate their sensorial attributes, both raw and cooked. Water has been served between tests to clean palates. A panel made up of 4 panelists, selected according to international standards (ISO 8586 part ), trained in fish quality assessment, and familiar with Quantitative Descriptive Analysis methods, developed the QIM scheme. During the 1st trial, a continuous training was carried out to acquaint the panelists with the sensory parameters, with the aim of developing a preliminary QIM scheme for gilthead seabream under MAP. The set of preliminary assessed parameters is reported in Table 2. Furthermore, high definition photos were taken in each sensory evaluation session as references for visual sensory attributes changes during storage. The preliminary scheme has been modified to a final scheme, according to the results of a PCA, in order to eliminate redundant parameters. In the 2nd trial, the final QIM scheme (calibration curve) was used to evaluate fish stored at different temperatures (2 and 8 C), packed under the best performing MAP from the 1st trial, selected basing on QI scores and sensory scores of cooked fish, assessed using a modified Torry scheme, according to Alasalvar and others (2000), in order to establish the rejection point. Briefly, a hedonic scale from 10 to 3 was used, 10 stating absolutely fresh and 3 completely putrid or spoiled, assessing odor, flavor, and texture scores. Fish fillets were cooked in a microwave oven for 3 min then served to the panelists after 5 min. The fish were assessed for odor, flavor, and texture. The rejection threshold was set <6. Statistical analyses In order to asses differences among fish and/or panelists in the evaluation of the QIM data, the QI scores of each evaluation session (QI score in each day) and for the 2 modified atmospheres were tested with 2 factor analysis of variance (ANOVA 2-factors), with no replications. The QI best fit regression line and the correlation coefficient were computed and the predictability of QI was S186 Journal of Food Science Vol. 76, Nr. 3, 2011

3 calculated with a 95% level of confidence, which corresponds to about 2 d precision. In order to minimize the number of parameters, PCA was performed together with cross-correlation analysis, based on the 5 d measurements. All the statistical analysis has been performed using the MATLAB statistical tools (Mathworks Inc., Natick, Mass., U.S.A.). Results and Discussion Development of QIM scheme for MAP packed gilthead seabream Design of the preliminary QIM scheme. The preliminary QIM scheme for packed gilthead seabream at 4 C, stored under 2 different atmospheres, MAP1 and MAP2, was developed using the scores from the 1st storage experiment. Eventual biases related to fish or to panelist variability were taken into account performing an ANOVA 2 factors test, where the factors were the panelist and the fish sample, in each session. The analysis showed little differences for one panelist, in 1 d of testing alone, at a significance level of 99%, in both MAP1 and MAP2 samples. Data in Table 1 are referred to samples packed in MAP1 samples. In the preliminary scheme, 17 parameters were assessed as reported in Table 2. Demerit points were assigned according to the expected rate freshness loss and to the relative importance of each parameter in stating the overall fish quality. QI sensory assessment of MAP packed gilthead seabream showed rapid early changes for some parameters, both in MAP1 and MAP2, namely pupil and muscle color, which reached the maximum score at day 3. Scores for odor, eye shape, eye clarity, skin appearance, mucus on skin, flesh elasticity, color, odor, and mucus on gills increased continuously during storage, although at different rates. Parameters, such as gaping and rancidity, did not change at all, while blood on skin followed an unpredictable trend during storage. Both QI and sensory scores indicate that storage under MAP1 provides a longer shelf life. In fact, the rejection threshold for cooked fish was 10 d for MAP2 samples and 13 d for MAP1 samples (Figure 1). In the MAP1 group, the rejection threshold corresponded to the attainment of the maximum demerit points for some sensory parameters (odor, eyes clarity, appearance of the skin, mucus on skin, elasticity, color of gills, odor of gills, and mucus on gills), and the reaching of a QI score of 20. Notably, MAP2 samples reached the rejection threshold at a QI score of Design of the final QIM scheme Since the sensory parameters for MAP1 and MAP2 groups were similar, in order to reduce the number of redundant parameters, we focused our analysis on the MAP1 data alone. We performed a PCA analysis on the 17 parameters of MAP1 and examined their cross-correlations, averaging the score of each indicator in each session using the evaluations of each panelist. The PCA revealed 4 principal components accounting for more than 99% of Table 1 P-values for ANOVA two factor analysis on fish samples and panelists. Days of storage Fish samples Panelists Table 2 Preliminary QIM scheme developed for MAP packed gilthead seabream. Demerit Parameter Characteristics points Odor Odor Fresh 0 Spoiled 3 Rancidity Not rancid 0 A little rancid 1 Rancid 2 Very rancid 3 Eyes Clarity Clear-translucent 0 Slightly opaque 1 Opaque 2 Form Convex 0 Flat 1 Concave 2 Pupil Black, bright, shiny 0 Slightly grayish 1 Gray-white 2 Blood Absent 0 Slightly bloody 1 Bloody 2 Very bloody 3 Appearance Skin Very bright 0 Bright 1 Slightly dull 2 Dull 3 Blood Absent 0 Present 1 Mucus No mucus 0 Not clotted, clear-transparent 1 Slightly clotted, yellowish 2 Clotted, brown 3 Flesh Rigor Rigor 0 Postrigor 1 Soft 2 Elasticity Not marked by pressure 0 Slightly marked by pressure 1 Clearly marked by pressure 2 Color Fresh, translucent 0 Waxy, milky 1 Dull, slightly dicolored, yellowish 2 Opaque, discolored, yellow, brown 3 Belly Firmness Firm 0 Soft 1 Sunken 2 Burst 3 Gills Color Dark red 0 Brownish red 1 Discolored 2 Odor Fresh 0 Spoiled 3 Mucus Absent 0 Slight 1 Moderate 2 Excessive or sticky 3 Gaping Gaping No gaping 0 Slight gaping 1 Gaping noticeable 2 Gaping pronounced, disrupted 3 Total 0 to 43 Table 3 Results of the factor analysis (PCA) on the QI parameters assessed. Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Eigenvalue Proportion Cumulative Vol. 76, Nr. 3, 2011 Journal of Food Science S187

4 variance, meaning that the number of parameters can be reduced (Table 3). In fact, 2 parameters (gaping and eyes-blood) showed a value of zero that did not change during the evaluation sessions and therefore did not contribute to the QIM score, and for these reasons were discharged. The cross-correlation data revealed that very few indicators were independent, while many were strongly correlated. A clustering analysis performed using the PCA results (Figure 2) showed that 2 more parameters, appearance-blood and Figure 1 Preliminary Index for chilled gilthead seabream under MAP. = dashed line. Torry scores for MAP2 packed samples. = dashed line. QI scores for MAP2 packed samples. = continuous line. Torry scores for MAP1 packed samples. = continuous line. QI scores for MAP1 packed samples. Horizontal lines = rejection threshold for cooked samples; Vertical lines = maximum storage days at 4 C; Red lines = linear regression of QI scores. Figure 2 Loadings of preliminary QIM scheme sensory parameters (MAP1 packed samples stored at 4 C). S188 Journal of Food Science Vol. 76, Nr. 3, 2011

5 odor-rancidity, presented a different behavior with respect to the others. Their loading relative to the 1st principal component is strongly negative; thus, they appear well separated into a different cluster (red spots). Since the 1st principal component accounts for the most part of the variance of the system (Table 3), the 2 parameters do not contribute to the system variability with time and are therefore useless or negligible for contributing to the QIM score. Consequently, we eliminated the 4 parameters discussed above and decided to hold 13 of the original 17. The final QIM scheme for MAP1 (60% CO 2, 40% N 2 ) packed gilthead seabream consists of 13 descriptors (Table 4). QI score evolved linearly during storage and has a good fit with the theoretical QI (y = ; R 2 : 0.985) (Figure 3). Compared to the QIM scheme proposed by Huidobro and others (2000) for raw gilthead seabream stored in ice, the QIM scheme developed has a higher number of parameters (13 instead of 8) and higher total demerit points (25 instead of 15). We have included in the scheme some parameters that are useful to discriminate early fish freshness modifications, namely rigor state, pupil, and muscle color. Moreover, belly firmness, which was found always to sunken in the latter stages of storage in MAP, and the gills mucus, which is related to microbial proliferation and Table 4 Final QIM scheme developed for MAP packed gilthead seabream. Demerit Parameter Characteristics points Odor Odor Fresh 0 Eyes Clarity Clear-translucent 0 Slightly opaque 1 Opaque 2 Form Convex 0 Flat 1 Concave 2 Pupil Black, bright, shiny 0 Slightly grayish 1 Gray-white 2 Appearance Skin Very bright 0 Bright 1 Slightly dull 2 Mucus No mucus 0 Not clotted, clear-transparent 1 Clotted, brown 2 Flesh Rigor Rigor 0 Postrigor 1 Elasticity Not marked by pressure 0 Slightly marked by pressure 1 Clearly marked by pressure 2 Color Fresh, translucent 0 Waxy, milky 1 Dull, slightly dicolored, yellowish 2 Belly Firmness Firm 0 Soft 1 Sunken 2 Gills Color Dark red 0 Brownish red 1 Discolored 2 Odor Fresh 0 Mucus Absent 0 Slight 1 Copious 2 Total 0 to 25 degradation of gills tissues, were also included in the parameter list. Although a shelf life of 13 d was determined for MAP packed in 60% CO 2 and 40% N 2 basing on QI scores, the QI evaluation of shelf life should match microbiological standards in respect to product safety, since pathogens outgrowth may occur while the product is still acceptable under QI scores. Indeed, use of MAP with improper cold chains may introduce an increased risk of botulism. The Advisory Committee on the Microbiological Safety of Food recommended in 1994 that, when chilled storage conditions is the only control factor for C. botulinum growth and toxin production, storage at temperatures between 5 and 10 C should be limited to 5 d, and shelf life up to 10 d could be assigned only when storage temperatures are maintained below 5 C(Davis 1998). Effect of storage temperature on QI scores The shelf life limiting critical point, from production to consume, is fish exposure in bench and home storage, in which temperatures well above the 4 C set temperature are likely to be found (Riva and Limbo 2007). Higher temperatures lead to a less dissolved CO 2 in the product and in accelerated microbial and enzymatic activity (Sivertsvik and others 2002). Hence, we decided to store MAP1 packed samples at very low (2 C) and under abuse conditions (8 C) temperatures, to test the effect of storage temperature on QI scores, using the final scheme for MAP1. QI scores followed a linear increase that fit the model (R 2 :0.95 and 0.99, at 2 and 8 C, respectively). Fish stored at 8 C suffered a very fast loss of freshness, resulting in a rapid increase of QI score. In fact, on day 7, the QI score was 21.5 and the Torry score had passed the rejection threshold (Figure 4). Therefore, QIM scores higher than those predicted by the scheme developed at 4 C, which is the correct refrigeration temperature, should aware that a temperature abuse has occurred. Conversely, fish stored at 2 C showed on day 7 a QI score of 15, a Torry score of 6.83 and the rejection threshold (<6) were reached after 13 d (data not shown). Therefore, while thermal abuse reduced drastically the shelf life, due to decreased solubility of CO 2 in the product and acceleration of microbial metabolism and endogenous enzymes activities, storage at 2 C was found not to improve sensorial parameters compared to storage at 4 C. Figure 3 Final Index for chilled gilthead seabream under MAP (60% CO2, 40% N2). = Torry scores; = QI scores; Horizontal line = rejection threshold for cooked samples; Vertical line = maximum storage days at 4 C; Red line = linear regression of QI scores. Vol. 76, Nr. 3, 2011 Journal of Food Science S189

6 Figure 4 Index for gilthead seabream under MAP (60% CO2, 40% N2) stored at 8 C. = Torry scores; = QI scores; Horizontal line = rejection threshold for cooked samples; Vertical line = maximum storage days at 8 C; Red line = linear regression of QI scores. Conclusions A QIM scheme was developed for gilthead seabream packed in MAP (60% CO 2 to 40% N 2 )storedat4 C, with 2 d precision. MAP1 (60% CO 2 to 40% N 2 ) was found to be better than O 2 - containing MAP2, resulting in a longer shelf life. The scheme developed consists of 13 parameters, which gave a total demerit point of 25. The scheme allows to predict the residual shelf life of MAP packed gilthead seabream stored at 4 C, using the regression equation obtained (y = ), with a 2 d precision. The limit of acceptability was found to be 13 d at 4 C and storage at 2 C was found not advantageous in respect to 4 C. Under temperature abuse storage, the shelf life was limited to 6 d. The knowledge of shelf life limiting factors of different fish species from different aquatic environments and under different packaging conditions is still limited, so studies are needed to optimize the packaging technique and to assess the quality of fish during storage. Specific QIM schemes for variable combinations of fish speciesaquatic environment (related to specific microbial contamination) packaging technology need to be developed. Acknowledgments We wish to thank Dr. Luca Pretti for technical assistance. This work was supported by Sardegna Ricerche. References Alasalvar C, Taylor KDA, Öksüz A, Garthwaite T, Alexis MN, Grigorakis K Freshness assessment of cultured sea bream (Sparus aurata) by chemical, physical and sensory methods. Food Chem 72: Anonymous Council regulation (EC) No 2406/96 of November 26, 1996, laying down common marketing standards for certain fisheries products. Off J Eur Comm L334:1 14. Baixas-Nogueras S, Bover-Cid S, Veciana-Nogues T, Nunes ML, Vidal-Carou MC Development of quality index method to evaluate freshness in Mediterranean Hake (Merluccius merluccius). J Food Sci 68: Bekaert K Development of quality index method scheme to evaluate freshness of tub gurnard (Chelidonichthys lucernus). In: Luten JB, Jacobsen C, Bekaert K, Sæbø A, Oehlenschläger J, editors. Seafood research from fish to dish. HA Wageningen: Wageningen Academic Publishers. p Boskou G, Debevere J Reduction of trimethylamine oxide by Shewanella spp. under modified atmospheres in vitro. Food Microbiol 14: Botta JR Evaluation of seafood freshness quality. New York: VCH Publishers Inc. 180 p. Bremner HA A convenient, easy to use system for estimating the quality of chilled seafood. Fish Proc Bull 7: Brown WD, Albright M, Watts DA, Heyer B, Spruce B, Price RJ Modified atmosphere storage of rockfish (Sebastes miniatus) and silver salmon (Oncory-hynchus kisutch). J Food Sci 45:93 6. Daniels JA, Krishnamurthi R, Rizvi SSH A review of effects of carbon dioxide on microbial growth and food quality. J Food Protect 48: Davis HK Fish and shellfish. In: Parry RT, editor. Principles and applications of modified atmosphere packaging of foods. Glasgow: Blakie Academic & Professionals. p Debevere J, Boskou G Effect of modified atmosphere packaging on the TVB/TMAproducing microflora of cod fillets. Intern J Food Micro 31: Devlieghere F, Debevere J, Van Impe J. 1998a. Effect of dissolved carbon dioxide and temperature on the growth of Lactobacillus sake in modified atmospheres. Int J Food Microbiol 41: Devlieghere F, Debevere J, Van Impe J. 1998b. Concentration of carbon dioxide in the waterphase as a parameter to model the effect of a modified atmosphere on micro-organisms. Int J Food Microbiol 43: Dixon NM, Kell DB The inhibition of CO2 of the growth and metabolism of microorganisms. J Appl Bacteriol 67: Drosinos EH, Lambropoulou K, Mitre E, Nychas GJE Attributes of fresh gilt-head seabream (Sparus aurata) fillets treated with potassium sorbate, sodium gluconate and stored under modified atmosphere at 0 ± 1 C. J Appl Microbiol 83: Ernst & Young, AND-I, Eurofish, Indemar Etude des performances économiques et de la compétitivité de l aquaculture de l Union Européenne. Publisher: European Commission DG MARE, Available from: mances-conomiques-et-de-la-comptitivit-de-laquaculture-de-lunion-europenne/. Accessed Feb 23, Farber JM Microbiological aspects of modified atmosphere packaging technology a review. J Food Protect 54: Goulas AE, Kontominas MG Combined effect of light salting, modified atmosphere packaging and oregano essential oil on the shelf-life of sea bream (Sparus aurata): biochemical and sensory attributes. Food Chem 100: Herrero AM, Huidobro A, Careche M Development of a quality index method for frozen hake (M. capensis and M. paradoxus). J Food Sci 68: Huidobro A, Pastor A, Tejada M index method developed for raw gilthead seabream (Sparus aurata). J Food Sci 65: Huidobro, A, Pastor A, López-Caballero ME, Tejada M Washing effect on the quality index method (QIM) developed for raw gilthead seabream (Saprus aurata). Eur Food Res Technol 212: Hyldig G index method a rapid sensory method for quality management. Sensory and instrumentally measured quality of farmed fish. NJF Seminar Nr 285. Available from: Accessed Feb 23, Hyldig G, Bremner A, Martinsdóttir E, Schelvis R index methods. In: Nollet LML, editor. Handbook of meat, poultry and seafood quality. Ames: Blackwell Publishing. p Lyhs U, Schelvis-Smit R Development of a quality index method (QIM) for maatjes herring stored in air and under modified atmosphere. J Aquat Food Prod Technol 14: Inácio P, Bernardo F, Vaz-Pires P Effect of washing with tap and treated seawater on the quality of whole scad (Trachurus trachurus). Eur Food Res Technol 217: ISO Sensory analysis-general guidance for the selection, training and monitoring of assessors. Part 1: selected assessors. First edition. Geneva, Switzerland: Intl. Organization for Standardization. Martinsdóttir E, Sveinsdóttir K, Luten J, Schelvis-Smit R, Hyldig G Sensory evaluation of fish freshness. Reference manual for the fish sector. Ijmuiden: QIM-Eurofish. p 49. Martinsdóttir E, Luten JB, Schelvis-Smit AAM, Hyldig G Developments of QIM past and future. In: Luten JB, Oehlenschlägern J, Ólafsdóttir G, editors. of fish from catch to consumer labelling, monitoring and traceability. Enfield: Enfield Pub & Dist Co Inc. p Nelson RW, Tretsven WI Storage of pacific salmon in controlled atmosphere. In: Progress in refrigeration Science and Technology, Proceedings of the XIV international Congress of refrigeration. Moscov: International Institute of Refrigeration Aberdeen. Özogul F, Polat A, Özogul Y The effect of modified atmosphere packaging and vacuum packaging on chemical, sensory and microbiological changes of sardines (Sardina pilchardus). Food Chem 85: S190 Journal of Food Science Vol. 76, Nr. 3, 2011

7 Parkin KL, Brown WD Preservation of seafood with modified atmospheres. In: Martin RE, Flick GJ, Hebard CE, Ward DR, editors. Chemistry and biochemistry of marine food products. Westport, Conn.: AVI Publishing Cp. p Pastoriza L, Sampedro G, Herrera JJ, Cabo ML Influence of sodium chloride and modified atmosphere packaging on microbiological, chemical and sensorial properties of aquacultured sea bass (Dicentrarchus labrax) stored in ice. Food Microbiol 20: Pournis N, Papavergou A; Badeka A; Kontominas MG; Savvaidis IN Shelf-life extension of refrigerated mediterranean mullet (Mullus surmuletus) using modified atmosphere packaging. J Food Protect 68: Reineccius G Off-flavors in food. Crit Rev Food Sci 29: Riva M, Limbo S Shelf life study of fresh European sea bass (Dicentrarchus labrax) and its modelling. Acts of: 3CIGR Section VI Intl. Symposium on Food and ASSgricultural Products: Processing and Iinnovations; 2007 September 24 26; Naples, Milan: AIDIC. Robertson GL Deteriorative reactions in foods. In: Robertson GL, editor. Food packaging. Principles and practice. New York: Marcel Dekker Inc. p Sacks B, Gore A Gas packaging techniques and trends. Food Rev 14: Sivertsvik M, Jeksrud WK, Rosnes JT A review of modified atmosphere packaging of fish and fishery products-significance of microbial growth, activities and safety. Int J Food Sci Tech 37: Sivertsvik M, Rosnes JT, Jeksrud WK, Solubility and absorption rate of carbon dioxide into non-respiring foods. Part 2: raw fish fillets. J Food Eng 63: Sveinsdóttir K, Martinsdóttir E, Hyldig G, Jørgensen B, Kristbergsson K Application of Index Method (QIM) scheme in shelf-life study of farmed atlantic salmon (Salmo salar). J Food Sci 67: Tiffney P, Mills A Storage trials of controlled atmosphere packaged fish products. Technical report nr 191. Edinburg: Sea Fish Industry Authority. Available from: Accessed Feb 23, Torrieri E, Cavella S, Villani F, Masi P Influence of modified atmosphere packaging on the chilled shelf-life of gutted farmed bass (Dicentrarchus labrax). J Food Eng 77: Tsironi T, Taoukis PS Modeling microbial spoilage and quality of gilthead seabream fillets: combined effect of osmotic pre-treatment, modified atmosphere packaging and nisin on shelf life. J Food Sci 75: Vader J, Carrapat H, Maraschio O, Thygesen G, Guirrec Y, Mìguez Lopez E, Bergmendsson KD grading and e.commerce in European fish auctions. In: Luten JB, Oehlenschläger J, Ólafsdóttir G, editors. of fish from catch to consumer labelling, monitoring and traceability. Enfield: Enfield Pub & Dist Co Inc. p Villemure G, Simrad RE, Picard G Bulk storage of cod fillets and gutted cod (Gadus morhua) under carbon dioxide atmosphere. J Food Sci 51: Warm K, Bøknæs N, Nielsen J Development of quality index methods for evaluation of frozen Cod (Gadus morhua) and Cod Fillets. J Aqua Food Prod Technol 7: Vol. 76, Nr. 3, 2011 Journal of Food Science S191