Evaluation of sensory and biochemical changes in freshwater catfish stored under vacuum and different modified atmospheres

Abstract

The present study was carried out to compare the influence of six different packaging atmospheres (air, vacuum and MAPs including 5% O₂ + 40% CO₂ + 55% N₂, 5% O₂ + 60% CO₂ + 35% N₂, 5% O₂ + 80% CO₂ + 15% N₂ and 100% CO₂) on the biochemical and sensory attributes of freshwater catfish fillets stored at 4 °C. Fillets were monitored for biochemical parameters (pH, total volatile bases nitrogen (TVBN), lipid oxidation) and sensory attributes for 21 days. Proximate and fatty acid composition were also determined in fresh fillets. The sensory quality of all fillets was acceptable during the first 13 ± 1 days of storage in air, 16 ± 1 days of storage in vacuum and MAP1, 18 ± 1 days of storage in MAP2 and 20 ± 1 days of storage in MAP3. The overall sensory scores for fillets which were packed under 100% CO₂ were higher than the acceptable limit at the end of storage. It was found that fillets consisted of 5.71 g lipid per 100 g which is susceptible to oxidation due to the high amount of unsaturated fatty acids (63.86%) versus saturated fatty acids (31.14%). Vacuum packed and 100% CO₂ fillets showed the lowest TBARS values while air-stored samples showed the highest TBA values. TVBN increased negligibly during storage in all treatments and never exceeded the acceptability limit (35 mg N per 100 g). It can be concluded that 100% CO₂ was the best evaluated atmosphere for storage of catfish fillets at 4 °C with superior biochemical and sensory attributes.

---

1 Introduction

Fresh fish has a limited shelf life when compared to other fresh food products. This is because of some intrinsic factors such as the presence of a large amount of water in fish tissues, which leads to a high a_w, a low content of carbohydrates resulting in a neutral pH (>6.0) in post-mortem fish and a high content of non-protein nitrogen. Deterioration of fish muscle is mainly related to changes brought about by both chemical and microbiological reactions, such as oxidation of lipids, activities of the fish's own enzymes and microorganism proliferation.¹ The consumer tendency towards utilisation of high quality fresh products, which are minimally processed and without chemical preservatives, has intensified a particular technique known as modified atmosphere packaging (MAP). MAP is performed by replacing the normal atmospheric air surrounding the pack of food with a different gas mixture and is able to extend the shelf life of fresh meat, poultry and fish products.² It has been shown by several investigators that the spoilage of fish can be delayed using MAP.^3,4 The gas mixture in MAP for fish products is highly specific for each fish species. Although gas compositions of 40% CO₂ + 30% N₂ + 30% O₂ and 60% CO₂ + 40% N₂ have been recommended for lean and fatty fish, respectively,^5,6 Arashisar et al.⁷ reported that using 40% CO₂ + 30% N₂ + 30% O₂ was the worst choice for freshwater rainbow trout, which is considered to be a low fat fish.

Pangasius pangasius, which is known as patin in Malaysia, is a freshwater catfish and belongs to the semi-fatty fish category. It is popular for its juicy taste and is regarded as an economically important species in Malaysia. Presently, there are no data available in the literature regarding the effects of different MAPs and vacuum packaging on the shelf life of the freshwater patin fish. Thus, this study was undertaken to determine the influence of vacuum and MAP with different gas mixtures on the biochemical and sensory attributes of patin fillets stored at 4 °C.

2 Material and methods

2.1 Preparation of fish samples and packaging procedures

Two hundred and seventy live freshwater catfish (Pangasius pangasius), each weighing approximately 1 kg and 30 cm in length, were purchased from Taman Bukit Serdang fish ponds. The fish, which were still alive, were directly placed in polystyrene boxes in layers interleaved on ice and then transferred to the Food Safety and Quality Laboratory of the Faculty of Food Science and Technology within 20 min. The fish died during transportation, since they suffered from thermal shock and were maintained in the ice condition until they were taken out for filleting. To ensure uniformity of fish characteristics, they were bought from the same farm and were reared under similar conditions in aquaculture farms located in Kampar, Perak. On arrival to the laboratory, the fish were washed under running tap water and immediately filleted manually under strict aseptic conditions. Two fillets were obtained from each fish which totalled to 540 fillets. The fillets were then divided into 6 equal groups. In the first group, fillets were placed in the packaging film and stored aerobically without sealing for use as control; the remaining 5 groups of fish were packaged under five different atmospheres using an automatic vacuum packaging machine (DZQ400H, China). Twenty eight packs of fish were used for each group and 3 fillets were placed in each packaging film and packed. The packaging material was a three-layer nylon barrier film/bag, nylon 15/polyethylene 15/linear low density polyethylene 60 (nylon 15/pe15/lldpe60), 25 cm × 31 cm, 0.09 mm in thickness, with an oxygen transmission rate of 1.55 cm³ m⁻² atm⁻¹ per 24 h, a carbon dioxide transmission rate of 6.15 cm³ m⁻² atm⁻¹ per 24 h at 25 °C and a water vapour transmission rate of 15 g m⁻² atm⁻¹ per 24 h at 38 °C. The ratio between the volume of the gas and product (G/P ratio) was 2 : 1 (v/w) for MAP conditions. The gas mixture compositions used in this study were defined as MAP1 (5% O₂ + 40% CO₂ + 55% N₂), MAP2 (5% O₂ + 60% CO₂ + 35% N₂), MAP3 (5% O₂ + 80% CO₂ + 15% N₂) and MAP4 (100% CO₂). Fillets packed under vacuum conditions (VP) were inserted in a film and sealed after the removal of air. All of the treatments and control were stored in a cooled room at a constant temperature (4 ± 0.5 °C) for a period of 21 days.

2.2 Sampling

All of the fillets were subjected to chemical and sensory analyses at time points of 0, 3, 6, 9, 12, 15, 18 and 21 days. Biochemical analyses were performed in triplicate. At each sampling time, 4 packs from each treatment were taken randomly and analysed. Two packs were used for chemical analyses and the other two for sensory evaluations. Six fresh fillets (unpacked) were analyzed immediately for chemical and sensory parameters and the results obtained were considered as the data for day 0. Experiments were repeated twice at different times within a period of 2 months.

2.3 Chemical analyses

2.3.1 Proximate analyses. Before packaging the fillets in different gas mixtures and vacuum, the compositions of the constituents in patin fillets used in this study were determined. Fillets from three fish were chopped into small portions and completely mixed. Samples of minced fish were taken randomly and the proximate compositions were determined in triplicate. The moisture content was determined using the AOAC⁸ standard method. Accurately weighed samples of fish muscle (4–6 g) were dried in an oven at 105 °C overnight till a constant weight was obtained. The ash content was measured by burning the residue from the moisture determination in a furnace at 550 °C for 24 h.⁸ The protein content of fish samples was determined by the micro-Kjeldahl method.⁹ Total lipid was obtained from 50 g of the minced fish flesh using the chloroform–methanol extraction method based on a procedure described by Kinsella et al.¹⁰ with minor modifications by Kim.¹¹

2.3.2 Determination of fatty acid composition. The fatty acid methyl esters (FAMEs) of lipid samples were extracted according to the method described by Christie.¹² The fatty acid profile was analysed using gas chromatography (GC) according to the procedure proposed by Ariffin et al.¹³ A gas chromatograph (Agilent 6890) equipped with a flame ionisation detector (FID) was used at a split ratio of 1 : 40. A fused silica capillary BPX-70 column (60 ml × 0.32 mm ID, 0.25 μm film thickness; SGE, International Pty. Ltd, Victoria, Australia) was used. The temperature program was set from 115 to 180 °C at a rate of 8 °C min⁻¹ and maintained for 10 min and again raised to 240 °C at a rate of 10 °C min⁻¹ with a final holding time of 30 min. Next, 1 μl of FAME samples were injected and flashed through with helium as the carrier gas at a rate of 0.4 ml min⁻¹. The temperatures of the injection part and the detector were set at 250 °C and 280 °C, respectively. Fatty acids were identified and quantified by comparing their retention times with those of pure standards purchased from Supelco (Supelco™ 37 component FAME mix) and analysed under the same conditions.

2.3.3 pH. The pH of the fish flesh was measured after preparation of the fish homogenate using 10 g muscle and 100 ml of distilled water. A Mettler Toledo (Switzerland) pH meter was used in this study.

2.3.4 Total volatile bases nitrogen. Total volatile bases nitrogen (TVBN) is one of the most commonly used parameters in assessing seafood quality. It represents the measurement of volatile amines including trimethylamine (TMA), dimethylamine, ammonia and other volatile basic nitrogenous compounds associated with seafood spoilage. Analysis was carried out according to the steam distillation procedure described by Antonacopoulos and Vyncke.¹⁴ Distillation was carried out after the addition of MgO to minced fish flesh (10 g). The distillate was collected in a flask containing 25 ml of 3% aqueous solution of boric acid and a few drops of the mixed indicator (methyl red and methylene blue). After distillation, the solution in the receiver flask was titrated using an aqueous solution of 0.1 N hydrochloric acid. The amounts of TVBN were expressed as mg nitrogen per 100 g sample.

2.3.5 Lipid oxidation. Lipid oxidation was determined based on the concentration of thiobarbituric acid reactive substances (TBARS) in fish fillets according to the method proposed by Goulas and Kontominas.¹⁵ The results were expressed as mg malondialdehyde (MDA) per kg of fish flesh. In this assay, one molecule of MDA, the product of lipid oxidation, reacted with two molecules of 2-thiobarbituric acid (TBA), which were added to the sample, and formed a pink complex that could be detected using a spectrophotometer at 532 nm as an index of lipid oxidation.

2.4 Sensory evaluation

Sensory evaluation was performed based on the method described by Maqsood and Benjakul¹⁶ and Soccol and Oetterer⁶ with slight modifications. For example, prior to evaluation, fillets were removed from the cooled room and maintained at room temperature (25–28 °C) for 30 minutes. Fish fillets were presented individually to each panellist in plastic trays and were randomly coded with three digit numbers unrelated to the storage conditions. A patin fillet which had been frozen at 30 °C and thawed at room temperature prior to sensory analyses was also presented to the panellists as the control sample. Duplicate samples of patin fish fillets were evaluated by 30 untrained panellists. At each time point, two packs of fillets from each treatment were used for sensory analyses. One fillet from each pack was removed and presented to the panellists (n = 2). Panel members, aged between 24 and 31 years, were postgraduate students of the Universiti Putra Malaysia in the Faculty of Food Science and Technology who studied different fish species in their projects and were familiar with fish consumption and spoilage. Fish samples were analysed for colour, texture, odour and overall acceptability, using a nine-point hedonic scale, where 1 represents dislike extremely, 2: dislike very much, 3: dislike moderately, 4: dislike slightly, 5: neither like nor dislike, 6: like slightly, 7: like moderately, 8: like very much, and 9: like extremely. A score of 5 and lower was considered as the limit of acceptability for each attribute.

2.5 Statistical analysis

All data were subjected to one-way analysis of variance (ANOVA with Duncan's test) to evaluate the differences between mean values. SPSS software (SPSS 15.0 for windows, SPSS Inc., Chicago, IL, USA) was used as the statistical analysis system and the significance was defined at p < 0.05.

3 Results and discussion

3.1 Chemical changes

3.1.1 Proximate composition. The moisture, protein, lipid and ash contents of Pangasius pangasius are shown in Table 1. A few studies have been carried out on the chemical composition of Malaysian patin fish. These include a study by Tee et al.,¹⁷ who carried out the proximate analysis of this fish and reported a fat content of 5.5%. In another study which was carried out by Suriah et al.,¹⁸ 5.68 g per 100 g lipid was reported in this species. Thus, the fat percentage (5.7%) of the fish obtained in the present work is in agreement with the above-mentioned studies. The moisture, protein and ash contents of patin fish in this study were in line with the findings of the last research by Tee et al.¹⁷ They reported that the chemical composition of patin was 76.2% moisture, 16.6% protein and 1.1% ash. It is well known that the chemical composition of fish muscle is influenced by many factors such as age, sex, geographical regions, fish size, catching season, diet and other environmental conditions.¹⁹

Table 1 Chemical composition of patin fillets^a

Proximate composition	% mean value	SD
a Data are expressed as the mean value of six samples.
Protein content	16.71	0.17
Moisture content	76.11	0.09
Lipid content	5.7	0.13
Ash content	1.33	0.06

---

3.1.2 Fatty acid profile of patin fillets. The fatty acid composition of the lipid extracted from patin (Pangasius pangasius) fillets is presented in Table 2. The gas chromatogram of the fatty acid profile of patin fish is shown in Fig. 1. Lipid of patin fillets contained 42.87% monounsaturated fatty acids (MUFAs), 31.14% saturated fatty acids (SFAs) and 21.08% polyunsaturated fatty acids (PUFAs). The principal fatty acid of patin oil was oleic acid (38.89%), followed by palmitic acid (22.71%) and linoleic acid (15.30%). This is similar to the results of a study carried out by Maqsood and Benjakul.¹⁶ Based on their results, the fatty acid composition of patin resembled that of the striped catfish (Pangasius hypophthalmus), which is a closely related family of patin. The percentage of PUFAs in freshwater patin fish was lower (21.08%) compared to the other fatty acid groups (SFAs and MUFAs), and it is not surprising that low levels of omega-3 fatty acids (1.22%) were observed in the freshwater patin fish in the present study. It has been shown by several investigators that freshwater fish contain lower values of omega-3 polyunsaturated fatty acids compared to marine fish (seawater). This is because marine fish obtain these omega-3 fatty acids from zooplankton, whereas freshwater fish feed mainly on vegetation and plant materials.^18,20,21 Furthermore, fish require polyunsaturated fatty acids to tolerate low water temperatures; therefore, it could be expected that in the warm waters of some countries located near the equator, such as Malaysia, PUFA concentrations in lipids become lower than in cold water fish.²¹ The total content of unsaturated fatty acids in the lipid of freshwater patin was approximately two-fold higher than that of saturated fatty acids, thus indicating that the lipids of patin fish were susceptible to oxidation.

Table 2 Fatty acid profile of lipids extracted from patin fillets^a

Fatty acids	Formula	Fatty acid content (g per 100 g lipid)
a Data obtained from duplicate samples; values are means ± standard deviation (n = 2).
Myristic acid	C14:0	0.84 ± 0.02
Palmitic acid	C16:0	22.71 ± 0.21
Stearic acid	C18:0	6.32 ± 0.01
Tricosylic acid (tricosanoic acid)	C23:0	0.53 ± 0.01
Lignoceric acid (tetracosanoic acid)	C24:0	0.72 ± 0.01
Σ Saturated fatty acids (SFAs)		31.14
Palmitoleic acid	C16:1	3.35 ± 0.03
Oleic acid	C18:1	38.89 ± 0.07
Cis-11-eicosenoic acid	C20:1 n-9	0.62 ± 0.005
Σ Monounsaturated fatty acids (MUFAs)		42.87
Linoleic acid(LA)	C18:2 n-6	15.3 ± 0.003
γ-Linolenic acid (GLA)	C18:3 n-6	0.50 ± 0.01
α-Linolenic acid	C18:3 n-3	0.58 ± 0.01
Cis-11,14-eicosadienoic acid	C20:2 n-6	0.73 ± 0.01
Cis-8,11,14-eicosatrienoic acid (hGLA)	C20:3 n-6	1.09 ± 0.01
Arachidonic acid	C20:4 n-6	2.21 ± 0.02
Docosahexaenoic acid (DHA)	C22:6	0.64 ± 0.02
Σ Polyunsaturated fatty acids (PUFAs)		21.08

---

Fig. 1 Chromatogram of the fatty acid composition of patin (Pangasius pangasius) fish oil.

3.1.3 Changes in thiobarbituric acid reactive substances (TBARS). Changes in the TBARS values of patin fillets during storage under different MAP conditions, vacuum and air at 4 °C are shown in Fig. 2(a). The initial TBA value of patin fillets at day 0 was very low (0.35 mg malondialdehyde (MAD) per kg). As can be seen in Fig. 2, the TBA values of all treatments increased steadily throughout the storage period up to day 18. Afterward, TBA values were found to decrease as a consequence of the interaction between malonaldehyde (MDA) and protein degradation products.²² Similarly, Goulas and Kontominas,¹⁵ who investigated TBA values of MAP and air-stored sea bream fillets at 4 °C, reported that there was a trend towards an increase in TBA values up to a certain point during storage, followed by either a decrease in these values or a reduced rate of increase. In the current study, it was found that vacuum and different MAP conditions were very effective in retarding the lipid oxidation of patin fillets during refrigerated storage. Fillets which were packed under vacuum displayed lower (but not significant, p > 0.05) TBARS values compared to those packed with different gas mixtures. The absence of oxygen in the vacuum packed samples could be an explanation for the lowest TBARS values in these fillets in the present research. This result is in agreement with a report by Gimenez et al.²³ The results of this study indicated that TBA values in air-stored fillets were significantly higher (p < 0.05) than those of other treatments. This finding is indicative of the fact that the presence of high levels of oxygen (∼21%) in an air atmosphere has a clear influence on the oxidation level of air-stored fillets. According to Connell,²⁴ TBA values of 1–2 mg MDA per kg of fish flesh are usually regarded as the limit beyond which the fish will normally develop an objectionable odour/taste. In the present research, when TBA values of air-stored fillets reached a value of 2 mg MDA per kg (∼13 days of storage), rancidity was evident in the patin fillets. The current results show that oxidative rancidity in fillets which were stored under different MAP and vacuum conditions remained relatively low during the storage period at 4 °C, and their level never reached a value of 2 mg MDA per kg. Several investigators have reported that oxidative rancidity in some fish may become a problem if higher than normal levels of oxygen are used in the modified atmosphere.^25,26 Adjusting low level of oxygen in MAP-stored patin fillets (5%) could be a possible reason for the small amounts of TBA in these treatments.

Fig. 2 Changes in TBARS levels (a), TVBN levels (b), and pH (c) of patin fillets stored under different conditions at 4 °C.

3.1.4 Changes in pH. Changes in the pH of patin fillets during storage under different MAP conditions, vacuum and air at 4 °C are shown in Fig. 2(c). In the present study, the initial pH of patin fillets (at day 0) was 6.52, which increased gradually in air-stored samples at 4 °C and reached 6.97 on the final sampling day (21 day). In general, the highest pH values were obtained in the control, and significant differences (p < 0.05) were observed in the pH between air-stored samples and other treatments from the 12^th day of storage onward. The higher pH observed in air-stored patin fillets can be partly attributed to the accumulation of basic compounds such as ammonia and other alkaline substances as a consequence of bacterial growth. Goulas and Kontominas,²⁷ Torrieri et al.²⁸ and Siah and Ariff²⁹ have also reported higher pH values in aerobically stored samples compared to MAP-stored samples. In contrast, lower pH levels were found in MAP samples, and the rate of increase in pH in MAP treatments was lower compared to the control. This low pH in MAP-stored fillets was consistent with the low bacterial counts which were observed, particularly in the 100% CO₂ treatment. This may be due to the bacteriostatic effect of carbon dioxide, resulting in the prevention of protein breakdown and therefore amine formation. Similar findings were also observed by Chen and Xiong,³⁰ Hovda et al.³¹ and Gimenez et al.²³ It is noteworthy that a slight drop in pH took place in those packaging atmospheres which were enriched in carbon dioxide at the beginning of the storage period. This reduction in pH can be ascribed to the dissolution of CO₂ in fish muscle, where it is converted to carbonic acid and results in a drop in pH. Several studies have recorded a fall in the pH of MAP samples, including those by Siah and Ariff²⁹ in barramundi fillets, Kostaki et al.⁴ in sea bass and Gimenez et al.²³ in rainbow trout.

3.1.5 Changes in total volatile bases nitrogen (TVBN). Changes in the TVBN content of patin fillets during storage under different MAP conditions, vacuum and air at 4 °C are shown in Fig. 2(b). The initial TVBN content of patin fillets on day 0 was 18 mg per 100 g. According to Huss³² and Connell,²⁴ the level of TVBN in freshly caught fish is generally between 5 and 20 mg N per 100 g, while concentrations of 30–35 mg N per 100 g are considered to be the acceptable limit above which fishery products are regarded as unfit for human consumption.³³

The TVBN content of patin fillets ranged from 18 to 33.5 mg N per 100 g in air-stored samples throughout the 21-day storage period. As shown in Fig. 2(b), the TVBN values of the control sample did not increase greatly during the first 12 days of storage (∼4 mg N per 100 g), whereas it rose more rapidly from 12^th to the 21^st day (∼11 mg N per kg). Similarly a minor increase in TVBN levels occurred in other treatments (MAP1, MAP2, MAP3, MAP4 and VP) in the present study. Similarly, Castro et al.³⁴ found low concentrations of TVBN during the edible storage life of European sea bass (21 days); they discovered an increase in TVBN after 20–22 days of storage (beyond their shelf lifetime). It is believed that a slow increase in TVBN occurs in some fish with little or no trimethylamine oxide (TMAO) or in those where spoilage is due to a non-TMAO reducing flora. Thus, low levels of TVBN were liberated during storage, probably resulting from the deamination of amino acids.³⁵ According to Gram and Huss,³⁶ TMAO is present in all marine fish, but only some freshwater fish possess this compound. Trimethylamine (TMA), which constitutes a major part of the TVBN content, is produced by some spoilage bacteria such as Shewanella putrefaciens, Photobacterium phosphoreum and Vibrionaceae under anaerobic conditions. According to Kyrana et al.³⁷ low levels of TMAO in cultured fish are probably due to the composition of the feed.

It could be speculated that since the Pangasius pangasius used in the present study is a cultured freshwater fish, the level of TMAO present in the fish flesh is not considerable, therefore little TMA and as a consequence negligible TVBN is produced during storage. The results of the present study indicate that TVBN values in all treatments did not exceed the acceptability limit at the time of spoilage, therefore it can be concluded that TVBN is not a useful indicator for quality assessment of patin (Pangasius pangasius) fish. Similar observation was also achieved by Lalitha et al.³⁸ in Pearl spot during storage under different MAP conditions and air at 0 °C. They attributed low TVBN content to extremely small quantities of TMA and H₂S producing a bacterial count of <10⁸ CFU per g. Other investigators like Kyrana and Lougovois,³⁹ Papadopoulos et al.⁴⁰ and Castro et al.³⁴ have reported that TVBN is not a suitable indicator for establishing the quality of sea bass (Dicentrarchus labrax) during storage at 0 °C.

3.2 Sensory evaluation

The sensory evaluation of patin fillets during storage under different MAP conditions, vacuum and air at 4 °C is shown in Table 3. Patin fillets on day 0 of storage had a firm texture, a pink yellowish colour and possessed no off odour as described by panellists. Thus, the day 0 samples received the highest overall sensory scores (8.64). All of the sensory attributes for filleted patin samples stored under air, vacuum and different MAP conditions indicated a similar pattern of decreasing acceptability. In the present study, the sensory quality of fillets stored under aerobic conditions and vacuum packaging deteriorated faster than the MAP samples. Air-stored samples exhibited lower sensory scores for all attributes and they got spoiled after 13 days of storage. Rejecting the air-stored patin fillets by panellists on day 13 of storage was in accordance with the development of rancidity which was observed with high TBARS values on the same day of the storage period. Regarding the vacuum-packaged patin fillets, however, the levels of TBARS in these samples were comparable to MAP samples, but they garnered lower sensory scores and deteriorated faster. A similar finding was also obtained in vacuum-packed rainbow trout fillets by Gimenez et al.²³ Considering a score of 5 as the acceptable limit, the shelf life of patin fillets based on the overall sensory scores was the longest for MAP4 samples (more than 21 days) followed by MAP3 (between 18 and 21 days), MAP2 (18 days), MAP1, vacuum (between 15 and 18 days) and control samples (between 12 and 15 days). In general, the present sensory results were in good agreement with studies which have reported a longer shelf life for MAP-stored fish compared to fish under aerobic conditions and vacuum packaging, such as those of Gimenez et al.²³ in rainbow trout and Ozogul et al.³ in sardine.

Table 3 Sensory evaluation of patin fillets stored at 4 °C^a

Sensory parameter	Packaging conditions	Storage time (days)
		0	3	6	9	12	15	18	21
a Values are mean ± SD of scores of 30 panelists. Values within a column with different superscript letters (^abc) are significantly different (p < 0.05).
Odour	Air	8.55 ± 0.1^a	7.97 ± 0.4^a	6.75 ± 0.3^a	6.10 ± 0.4^a	5.50 ± 0.2^a	4.60 ± 0.5^a	4.12 ± 0.4^a	3.40 ± 0.7^a
	VP	8.55 ± 0.1^a	8.00 ± 0.2^a	7.44 ± 0.6^ab	6.90 ± 0.8^ab	6.09 ± 0.2^ab	5.26 ± 0.9^ab	4.81 ± 0.2^ab	4.07 ± 0.4^ab
	MAP1	8.55 ± 0.1^a	8.11 ± 0.1^a	7.48 ± 0.2^ab	6.94 ± 0.3^ab	6.21 ± 0.6^ab	5.47 ± 0.6^ab	4.90 ± 0.4^ab	4.33 ± 0.6^abc
	MAP2	8.55 ± 0.1^a	8.20 ± 0.5^a	7.61 ± 0.2^ab	7.02 ± 0.2^ab	6.25 ± 0.2^ab	5.47 ± 0.3^ab	5.07 ± 0.7^ab	4.30 ± 0.4^abc
	MAP3	8.55 ± 0.1^a	8.34 ± 0.3^a	7.77 ± 0.4^ab	7.16 ± 0.3^b	6.45 ± 0.4^ab	5.71 ± 0.5^b	5.46 ± 0.3^b	4.55 ± 0.8^bc
	MAP4	8.55 ± 0.1^a	8.42 ± 0.1^a	8.00 ± 0.3^b	7.52 ± 0.4^b	7.01 ± 0.5^b	6.30 ± 0.2^b	5.90 ± 0.8^b	5.25 ± 0.1^c
Texture	Air	8.73 ± 0.4^a	8.17 ± 0.4^a	6.72 ± 0.3^a	5.80 ± 0.4^a	5.08 ± 0.1^a	4.31 ± 0.5^a	3.81 ± 0.8^a	2.90 ± 0.1^a
	VP	8.73 ± 0.4^a	8.29 ± 0.3^a	7.34 ± 0.9^ab	6.64 ± 0.4^ab	5.85 ± 0.4^ab	5.10 ± 0.6^ab	4.72 ± 0.3^ab	4.05 ± 0.2^b
	MAP1	8.73 ± 0.4^a	8.33 ± 0.2^a	7.43 ± 0.6^ab	6.71 ± 0.4^ab	6.00 ± 0.8^abc	5.18 ± 0.9^ab	4.70 ± 0.5^ab	3.96 ± 0.5^b
	MAP2	8.73 ± 0.4^a	8.40 ± 0.1^a	7.60 ± 0.3^b	6.90 ± 0.4^b	6.17 ± 0.7^bc	5.42 ± 0.2^bc	4.97 ± 0.4^b	4.19 ± 0.3^b
	MAP3	8.73 ± 0.4^a	8.52 ± 0.4^a	7.77 ± 0.5^b	7.12 ± 0.4^b	6.43 ± 0.3^bc	5.65 ± 0.2^bc	5.19 ± 0.5^b	4.46 ± 0.7^b
	MAP4	8.73 ± 0.4^a	8.63 ± 0.2^a	7.86 ± 0.4^b	7.37 ± 0.4^b	6.96 ± 0.2^c	6.20 ± 0.1^c	5.75 ± 0.1^b	5.47 ± 0.4^c
Color	Air	8.62 ± 0.5^a	8.05 ± 0.6^a	6.76 ± 0.1^a	5.95 ± 0.4^a	5.55 ± 0.6^a	4.82 ± 0.2^a	4.10 ± 0.6^a	3.55 ± 0.5^a
	VP	8.62 ± 0.5^a	8.12 ± 0.5^a	7.52 ± 0.1^ab	7.19 ± 0.2^b	6.39 ± 0.9^ab	5.75 ± 0.1^ab	5.07 ± 0.4^ab	4.72 ± 0.8^b
	MAP1	8.62 ± 0.5^a	8.26 ± 0.5^a	7.73 ± 0.2^ab	7.14 ± 0.1^b	6.53 ± 0.8^ab	5.82 ± 0.3^ab	5.13 ± 0.5^ab	4.81 ± 0.3^b
	MAP2	8.62 ± 0.5^a	8.39 ± 0.4^a	7.87 ± 0.8^ab	7.36 ± 0.3^b	6.71 ± 0.3^ab	6.06 ± 0.2^b	5.32 ± 0.6^b	4.95 ± 0.5^b
	MAP3	8.62 ± 0.5^a	8.46 ± 0.3^a	7.99 ± 0.3^b	7.48 ± 0.3^b	6.85 ± 0.5^b	6.31 ± 0.6^b	5.54 ± 0.7^b	4.97 ± 0.3^b
	MAP4	8.62 ± 0.5^a	8.53 ± 0.3^a	8.18 ± 0.2^b	7.68 ± 0.7^b	7.12 ± 0.1^b	6.51 ± 0.1^b	5.79 ± 0.5^b	5.33 ± 0.5^b
Overall	Air	8.64 ± 0.2^a	8.02 ± 0.5^a	6.71 ± 0.2^a	5.93 ± 0.4^a	5.25 ± 0.1^a	4.45 ± 0.4^a	4.00 ± 0.2^a	3.33 ± 0.7^a
	VP	8.64 ± 0.2^a	8.10 ± 0.3^a	7.38 ± 0.8^ab	6.84 ± 0.1^ab	6.00 ± 0.5^ab	5.31 ± 0.3^ab	4.75 ± 0.1^ab	4.18 ± 0.2^ab
	MAP1	8.64 ± 0.2^a	8.18 ± 0.6^a	7.50 ± 0.4^ab	6.91 ± 0.4^ab	6.27 ± 0.3^abc	5.54 ± 0.9^ab	4.89 ± 0.5^ab	4.29 ± 0.1^abc
	MAP2	8.64 ± 0.2^a	8.25 ± 0.1^a	7.73 ± 0.1^b	7.02 ± 0.6^b	6.44 ± 0.3^bc	5.61 ± 0.2^b	5.19 ± 0.1^b	4.39 ± 0.3^bc
	MAP3	8.64 ± 0.2^a	8.33 ± 0.4^a	7.82 ± 0.7^b	7.18 ± 0.1^b	6.61 ± 0.2^bc	5.93 ± 0.3^b	5.42 ± 0.4^b	4.53 ± 0.5^bc
	MAP4	8.64 ± 0.2^a	8.55 ± 0.1^a	8.00 ± 0.3^b	7.46 ± 0.5^b	7.10 ± 0.2^c	6.37 ± 0.1^b	5.76 ± 0.6^b	5.31 ± 0.4^c

---

4 Conclusion

From the results of this study, it is clear that the shelf life of patin fillets could be extended significantly by using a gas mixture rich in carbon dioxide compared to vacuum packaging and air-storage. Sensory and biochemical characteristics of patin fillets improved when they were stored under higher CO₂ concentration. 100% CO₂ was found to be the best condition for the preservation of fillets. A shelf life of more than 21 days was achieved for patin fillets stored under 100% CO₂ compared to 13 days for air stored patin based on sensory data. This shelf life extension (8 days) could provide the consumers with an acceptable product, while conferring economic advantages during distribution and retailing at the same time.

Acknowledgements

We gratefully acknowledge the laboratory assistances of Faculty of Food Science and Technology, Universti Putra Malaysia (UPM), for the technical support.

References

1. M. Sivertsvik, W. K. Jeksrud and M. J. T. Rosnes, A review of modified atmosphere packaging of fish and fishery products significance of microbial growth, activities and safety, Int. J. Food Sci. Technol., 2002, 37, 107–127.
2. K. Fernández, E. Aspe and M. Roeckel, Shelf-life extension on fillets of Atlantic Salmon (Salmo salar) using natural additives, superchilling and modified atmosphere packaging, Food Control, 2009, 20, 1036–1042.
3. F. Ozogul, A. Polat and Y. Ozogul, The effect of modified atmosphere packaging and vacuum packaging on chemical, sensory and microbiological changes of sardines (Sardina pilchardus), Food Chem., 2004, 85, 49–57.
4. M. Kostaki, V. Giatrakou, N. L. Savvaidis and M. G. Kontominas, Combined effect of MAP and thyme essential oil on the microbiological, chemical and sensory attributes of organically aquacultured sea bass (Dicentrarchus labrax) fillets, Food Microbiol., 2009, 29, 475–482.
5. D. C. Cann, N. C. Houston, L. Y. Taylor, G. L. Smith, A. B. Thomson and A. Craig, Studies of Salmonids Stored Under a Modified Atmosphere, Ministry of Agriculture, Fisheries and Food, Torry Research Station, Aberdeen, U.K., 1984.
6. M. C. H. Soccol and M. Oetterer, Use of modified atmosphere in seafood preservation, Braz. Arch. Biol. Technol., 2003, 46, 569–580.
7. S. Arashisar, O. Hisar, M. Kaya and T. Yanik, Effect of modified atmosphere and vacuum packaging on microbiological and chemical properties of rainbow trout (Oncorynchus mykiss) fillets, Int. J. Food Microbiol., 2004, 97, 209–214.
8. AOAC, Official Methods of Analysis of the Association of Official Analytical Chemists, Washington, DC, 15th edn, 1990.
9. Y. Pomeranz and C. Meloan, Food Analysis: Theory and Practice, New York, Chapman & Hall, 1994.
10. J. E. Kinsella, J. L. Shimp, J. Mai and J. Weihrauch, Fatty Acid Content and Composition of Fresh Water Finfish, American Oil Chemists' Society, 1977, vol. 54, pp. 424–429.
11. L. L. Kim, Extraction of Lipids. In Laboratory Manual of Analytical Methods and Procedures for Fish and Fish Products, ed. Katsutoshi Miwa and Low Su Ji, Marine Fisheries Research Department, Southeast Asian Fisheries Development Center, C-2. Singapore, 1992.
12. W. W. Christie, Preparation of Ester Derivatives of Fatty Acids for Chromatographic Analysis, in Advances in Lipid Methodology – Two, ed. W. W. Christie, Oily Press, Dundee, 1993, pp. 69 – 111.
13. A. A. Arrifin, J. Bakar, C. P. Tan and R. A. Rahman, Essential fatty acids of pitaya (dragon fruit) seed oil, Food Chem., 2009, 114, 561–564.
14. N. Antonacopoulos and W. Vyncke, Determination of volatile basic nitrogen in fish: a third collaborative study by the West European Fish Technologists’ Association (WEFTA), Z. Lebensm.-Unters. Forsch., 1989, 189, 309–316.
15. A. E. Goulas and M. G. Kontominas, Effect of salting and smoking-method on keeping quality of chub mackerel (Scomber japonicas): biochemical and sensory attributes, Food Chem., 2005, 93, 511–520.
16. S. Maqsood and S. Benjakul, Synergistic effect of tannic acid and modified atmospheric packaging on the prevention of lipid oxidation and quality losses of refrigerated striped catfish slices, Food Chem., 2010, 121, 29–38.
17. E. S. Tee, S. Siti Mizura, R. Kuladevan, S. I. Young, S. C. Khor and S. K. Chin, Nutrient composition of Malaysian freshwater fishes, Proc. Nutr. Soc. Mal., 1989, 4, 63–73.
18. A. Suriah Rahman, T. S. Huah, O. Hassan and N. M. Daud, Fatty acid composition of some Malaysian freshwater fish, Food Chem., 1995, 54, 45–49.
19. K. H. I. Sallam, A. M. Ahmed, M. M. Elgazzar and E. A. Eldaly, Chemical quality and sensory attributes of marinated Pacific saury (Coloabis saira) during vacuum-packaged storage at 4 °C, Food Chem., 2007, 102, 1061–1070.
20. P. Vlieg and D. B. Brody, Lipid contents and fatty acid composition of some New Zealand freshwater finfish and marine finfish, shellfish and roes, N. Z. J. Mar. Freshwater Res., 1988, 22, 151–162.
21. F. Jabeen and A. S. Chaudhry, Chemical composition and fatty acid profiles of three freshwater fish species, Food Chem., 2011, 125, 991–996.
22. J. Fernandez, J. A. Perez-Alvarez and J. A. Fernandez-Lopez, Thiobarbituric acid test for monitoring lipid oxidation in meat, Food Chem., 1997, 59, 345–353.
23. B. Gimenez, P. Roncales and J. A. Beltran, Modified atmosphere packaging of filleted rainbow trout, J. Sci. Food Agric., 2002, 82, 1154–1159.
24. J. J. Connell, Control of Fish Quality, Fishing News Books Limited, London, 4th edn, 1995.
25. G. Finne, Modified- and controlled-atmosphere storage of muscle foods, Food Tech., 1982, 36, 128–133.
26. K. Stammen, D. Gerdes and F. Caporaso, Modified atmosphere packaging of seafood, Int. J. Food Sci. Nutr., 1990, 29, 301–331.
27. A. E. Goulas and M. G. Kontominas, 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., 2007, 100, 287–296.
28. E. Torrieri, S. Cavella, F. Villani and P. Masi, Influence of modified atmosphere packaging on the chilled shelf life of gutted farmed bass (Dicentrarchus labrax), J. Food Eng., 2006, 77, 1078–1086.
29. W. M. Siah and W. M. Ariff, Modified atmosphere storage of barramundi (Lates calcarifer) fillets, J. Trop. Agr. Food Sci., 2002, 30, 201–207.
30. G. Chen and Y. L. Xiong, Shelf- stability enhancement of precooked red claw crayfish (Cherax quadricarinatus) tails by modified CO₂/O₂/N₂ gas packing, LWT–Food Sci. Technol., 2008, 41, 1431–1436.
31. M. B. Hovda, B. T. Lunestad, M. Sivertsvik and J. T. Rosnes, Characterisation of the bacterial flora of modified atmosphere packed farmed Atlantic cod (Gadus morhua) by PCR-DGGE of conversed 16S rRNA gene regions, Int. J. Food Microbiol., 2007, 117, 68–75.
32. H. H. Huss, Fresh Fish Quality and Quality Changes, FAO, FAO Fisheries Series, No. 29, Rome, 1998.
33. J. J. Connell and J. M. Shewan, Sensory and Non-Sensory Assessment of Fish, in Advances in Fish Science and Technology, ed J. Connell, Fishing News Books, Farnham, 1980, pp. 55–56.
34. P. Castro, J. C. P. Padron, M. J. C. Cansino, E. S. Velasquez and R. M. De Larriva, Total volatile base nitrogen and its use to assess freshness in European sea bass stored in ice, Food Control, 2006, 17, 245.
35. H. H. Huss, Quality and Quality Changes in Fresh Fish. FAO Fisheries Technical Paper 348, Rome, 1995, pp. 195.
36. L. Gram and H. H. Huss, Microbiological spoilage of fish and fish products, Int. J. Food Microbiol., 1996, 33, 121–137.
37. V. R. Kyrana, V. P. Lougovois and D. S. Valsamis, Assessment of shelf life of maricultured gilthead sea bream (Sparus aurata) stored in ice, Int. J. Food Sci. Technol., 1997, 32, 339–347.
38. K. V. Lalitha, E. R. Sonaji, S. Manju, L. Jose, T. K. S. Gopal and C. N. Ravisankar, Microbiological and biochemical changes in pearlspot (Etroplus suratensis Bloch) stored under modified atmosphere, J. Appl. Microbiol., 2005, 99, 1222–1228.
39. V. R. Kyrana and V. P. Lougovois, Sensory, chemical, microbiological assessment of farm-raised European sea bass (Dicentrarchus labrax) stored in melting ice, Int. J. Food Sci. Technol., 2002, 37, 319–328.
40. V. Papadopoulos, I. Chouliara, A. Badeka, I. N. Savvaidis and M. G. Kontominas, Effect of gutting on microbiological, chemical, and sensory properties of aquacultured sea bass (Dicentrarchus labrax) stored in ice, Food Microbiol., 2003, 20, 411–420.

---