Centralised bulk pre-packaging of fresh pork retail cuts in various gas atmospheres

Meat Science 36 (1994) 293-308

Centralised Bulk Pre-Packaging of Fresh Pork Retail Cuts in Various Gas Atmospheres E, M. Buys, J. Kriiger & G. L. Nortj6 Meat ScienceCentre, Animal and Dairy ScienceResearch Institute. Private Bag X2, Irene, 1675, Republic of South Africa (Received 4 March 1992; revisedversion received 5 November 1992; accepted 8 November 1992)

ABSTRACT A centralised bulk pre-packaging technique (laboratory method), utilising various gas mixtures (c. 100% C02; c. 75% C02 : 25% N2; c. 80% 02 : 20% C02 and c. 25% C02:50% N2 : 25% 02), was evaluated in terms of quality attributes such as microbiology, colour, odour and consumer acceptability. According to the bacterial counts recorded, all four packaging treatments were successful in prolonging the storage life (21 days c. O°C) of centralised bulk pre-packaged pork retail cuts, while still ensuring a subsequent shelf life of at least 3 days (c. O°C). The gas mixture comprising c, 25% C02 : 50% N2 : 25% 02 was the most successful treatment in terms of acceptability and colour scores.

INTRODUCTION Fresh meat packaging is developing rapidly, both for large scale distribution and for retailing purposes. The growing dominance of supermarkets and self-service retailing in Europe has encouraged the centralised preparation and packaging of meat with techniques being developed to lengthen the shelf life accordingly (Taylor, 1990). In South Africa a similar trend is being observed with the introduction of a bulk prepackaging technique to the South African meat industry. During a previous trial (Scholtz et aL, 1992) it was concluded that pork retail cuts could be bulk stored in a 100% CO2 at 0°C for 21 days, 293 Meat Science 0309-1740/94/$07.00 © 1994 Elsevier ScienceLtd, England. Printed in Great Britain

294

E. M. Buys, J. Kriiger, G. L. Nortjd

with a subsequent retail shelf life of 4 days (0°C). The commercial storage life of the 100% CO2 bulk stored pork chops was, however, limited owing to a loss in the colour of the pork chops. According to Taylor (1990) a system which successfully combines bulk pre-packaging with the high O2:CO2 of modified atmosphere packaging (MAP) and low meat temperature conditions could provide colour stability of bulk prepackaged samples. The present study was designed to evaluate a laboratory bulk pre-packaging technique (utilising various approximate gas mixtures, i.e. 100% CO2; 75% CO2:25% N2; 80% O2:20% C02 and 25% C02:50% N2:25% 02) in terms of quality attributes such as microbiology, colour, odour and consumer acceptability.

MATERIALS AND METHODS

Meat Twelve pig carcasses were selected according to a c. 30 min post-mortem pH (>6, in the M. longissimus thoracis, in the area of the last three ribs) and a carcass mass of c. 65 kg. The 12 carcasses represented three repetitions of the experimental design. Only the loin cuts were used.

Packaging Treatments Both loins of each carcass were randomly allocated to a specific bulk pack treatment (c. 100% CO2; c. 75% CO2: 25% N2; c. 80% 02:20% CO 2 and c. 25% CO2:50% N2:25% 02) and each loin cut into 18 chops. Each chop was placed in a shallow styrofoam tray and overwrapped with PVC (OTR--c. 5000 ml/m2/24 h/1 atm at 22°C 75% RH).

Bulk Packaging (mother bag) Of the PVC-overwrapped chops from each loin 12 were bulk packed with a laboratory method, six per bulk packed (BB4L Cryovac barrier bag, OTR--39 ml/m2/24 h/atm at 23°C 75% RH). Twelve bulk packs, representing three repetitions, were vacuum packed, heat sealed (R6chermatic vacuum machine) and subsequently filled with c. 100% CO2 by insertion of a needle (Scholtz et al., 1992). The ballooned bulk pack was again heat sealed to close off the puncture. The volume ratio of gas headspace to meat was approximately 3:1. The other bulk packs were also filled with gas using the same method, i.e. 12 bulk packs with c. 75% CO2: 25% N2; c. 12 with 80% 02: 20% CO 2 and c. 12 with 25% C02:50°//0 N2:25% 02.

Pre-packaging of fresh pork retail cuts

295

Foamtray/P VC overwrap controls In order to evaluate the validity of the comparison of the different bulk packaging treatments (c. 100% CO2 versus c. 75% CO2:25% N2 versus c. 80% 02:20% C O 2 versus c. 25% CO2:50% N2:25% 02) o v e r the different pork carcasses, the other six PVC-overwrapped chops from each loin served as PVC-control samples.

Storage and shelf life study Bulk Packaging (mother bag) Three replicate bulk packs from each treatment (c. 100% CO2; c. 75% CO2 : 25% N2; c. 80% 0 2 : 20% C O 2 and c. 25% CO2 : 50% N 2 : 25% 02) were opened after c. 1 h of saturation with the specific gas mixture. One set (2) of PVC-overwrapped samples from each bulk pack was assessed 30 min after opening the bulk pack. The remaining four samples were displayed in an open deck retail display cabinet (c. 0°C) for 2 or 4 days. The remainder of the bulk packs (36) were stored at 0°C for either 7, 14 or 21 days. After each relevant storage period, three replicate bulk packs from each treatment were opened and assessed as noted above. Foamtray/P VC overwrap controls The Day 0 PVC-overwrapped control samples were assessed without a display period, while the rest of the samples were displayed in an open deck retail display cabinet (c. 0°C) for 2 or 4 days.

Quality Attributes Following the 0, 7, 14 and 21 days bulk pack storage periods, the PVCoverwrapped samples were displayed for each specified period, withdrawn and assessed according to microbiological, colour, odour and acceptability parameters.

Microbiological analysis A measured area of c. 12 cm 2 was removed aseptically to a depth of c. 5 m m from the upper surface of the sample (Nortj6 et al., 1982). This was homogenised in a Stomacher 400 ( D H K (Pty) Ltd) with a 100 ml of quarter strength Ringer's (Merck) diluent. Counts were obtained as follows: Total aerobic counts on Standard 1 nutrient agar (Std 1; Merck), incubated for 3 days at 25°C; total anaerobic counts on Std 1 agar, incubated for 5 days at 25°C, in an anaerobic jar using BBL GasPack plus gas generator envelopes (Bactlab Systems (Pty) Ltd) as H 2 and CO2 generators, while M R S agar (de Man et al., 1960) was used for determination of lactic acid bacteria (5 days at 30°C). Pseudomonas spp. were monitored on Kielwein

296

E. M. Buys, J. Krfiger, G. L. NortjO

agar (Kielwein, 1971) incubated for 3 days at 25°C and D H L agar (Sakazaki et al., 1960)was used to determine Enterobacteriaceae (2 days, 37°C). The Brochothrix thermosphacta count was only determined on the bulk packed samples after each specified storage period, the counts being done on streptomycin-thallous acetate agar (STAA, Gardner, 1966) following incubation 25°C for 5 days. Colour The colour of each unopened sample was assessed by a trained panel consisting of 10 people 30 min after each pack was opened. A colour chart was provided and the colour scored on a 5 point scale ranging from 1 = 'extremely pale'; 2 = 'pale'; 3 = ' n o r m a l ' ; 4 = 'dark' to 5 = 'extremely dark' (Anon., 1981). Odour Samples were assessed by a trained panel of 10 people 30 min after each pack was opened. Odour was scored on a 6 point scale which ranged from 1 = 'no odour'; 2 = 'fresh meat'; 3 = 'slightly off'; 4 = 'moderately off'; 5 = 'strongly off' to 6= 'completely off'. Acceptability The acceptability of each unopened sample was assessed by an untrained panel of eight people to give an indication of the consumer acceptability of the PVC-overwrapped sample. This was done 30 min after each pack was opened, according to an 8 point scale ranging from 1 = 'extremely unacceptable'; 2 = 'unacceptable'; 3 = 'moderately unacceptable'; 4 = 'slightly unacceptable'; 5--'slightly acceptable'; 6 = 'moderately acceptable'; 7 = 'acceptable' to 8 = 'extremely acceptable'.

Statistical analysis The data were analysed by analysis of variance to determine which factors (bulk packaging, storage period, display period) and interaction between factors contributed significantly to the different parameters determined. Levels of P _< 0.05 were taken to be significant.

RESULTS

PVC-Overwrapped Control Samples The mean total counts of the four control groups did not differ significantly over the whole retail shelf life period (Days 0, 2 and 4) (P = 0.6867), although they did on the individual sampling days (0 versus 2

Pre-packaging of fresh pork retail cuts

297

versus 4) (P = 0.0001). The results obtained for the mean p s e u d o m o n a d (P--0-8788) and mean lactic acid (P=0-1412) counts were similar. These results indicated that no significant differences were encountered regarding the mean total count of the different pig carcasses used, although the p o p u l a t i o n structure differed significantly regarding the mean anaerobe (P = 0-0060) and Enterobacteriaceae (P=0.0024) counts. The initial total count (24 h post mortem, Day 0) for the first control group (PVC-control 1) was log 2.57 cm -2, which steadily increased throughout the display to reach log 6-90 cm -2 on Day 5 (P=0.0001) (Table 1). The total counts of the second and fourth control groups, as well as the lactic acid bacteria, pseudomonad, anaerobe and Enterobacteriaceae counts, followed a similar trend i.e. steadily increasing from Day 0 onwards (Table 1). The third control g r o u p had a higher initial total count (log 4.71 cm -2) which did not increase over a subsequent 2 days display but reached higher levels on Day 4 of the display. Similar trends were also observed for all the other counts recorded for the third control group (Table 1). Therefore PVC-overwrapped control samples could serve as a general control to compare the results of the four packaging treatments that were assessed.

Bulk Packaging Microbiological assessment Mean total count. Although the packaging treatments did not influence the mean total counts significantly, significant differences were found regarding the different bulk storage days (P =- 0-0134) and retail shelf life days (P -- 0.0001)). The total counts obtained from the retail cuts from all the applications (c. 100% CO2; c. 75% C 0 2 : 2 5 % N2; c. 80O//o02:20% CO2 a n d c. 25% CO2 : 50% N2 : 25% 02) after 0 days bulk storage were initially (Day 0) at low levels, namely < log 3 cm 2 (Tables 2a and 2b). These counts increased to reach significantly higher levels after a subsequent 4 days retail display. The only exception was the retail cuts from the 80% 0 2 : 2 0 % CO2 treatment which reached an unacceptable high total count on Day 4 of the retail shelf life study (log 7-24 cm-2; P = 0.0001). The mean total counts recorded for the retail samples after 7 days bulk storage followed a similar trend to the counts recorded initially (0 days bulk storage) for all four bulk packaging treatments. The initial total counts (0 days display) after 14 and 21 days bulk storage were still at acceptable levels for all four bulk packaging treatments. These counts increased over a subsequent 2 days of retail display and then remained stable for the following 2 days of retail display (Tables 2a and 2b).

298

E. M. Buys, J. Kriiger, G. L. Nortj~

0

~

d d d

d

d d

d d d

d o o

0

~

~

0

N

.

~

.

d d 6

.

.

d

6 6

.

.

.

o o 6

.

d 6 d

Pre-packaging of fresh pork retail cuts

299

Mean anaerobe and lactic acid bacteria counts. The statistical analyses for the main effects of the anaerobe and lactic acid bacteria counts were similar to that noted for the total counts. The packaging treatments × storage days (P < 0-05) and storage days × display period (P < 0.05) interactions of these counts were however also significant. According to the packaging treatments × storage days interaction ( P = 0-0140) of the anaerobic counts, the samples stored in e. 25% CO2:50% N2:25% 02 for 0 days had lower mean anaerobe counts (log 2.07 ca'n-z) after display (Days 0, 2 and 4) than the samples from all the other treatments (log 3-00 cm-2). After 14 and 21 days storage and subsequent display the latter, however, had lower mean counts (log 4.00 cm -2) than those stored in c. 25% CO2: 50% N2:25% O2 (log 5-60 cm 2). A similar observation was made for the packaging treatments × storage days (P=0.0140) interaction of the lactic acid bacteria counts. The mean anaerobe and lactic acid bacteria counts (for all the packaging treatments) followed the same increasing trend as the mean total counts after 0 days bulk storage (Tables 2a and 2b). Mean pseudononad count. Only the display period (P=0-0001) main effect had a significant influence on the mean pseudomonad counts. The mean pseudomonad counts followed the same increasing trend as the mean total counts after 0 days bulk storage for all the packaging treatments (Tables 2a and 2b). Mean Enterobacteriaceae count. The statistical analysis of the mean Enterobacteriaceae counts was similar to that recorded for the mean total counts. The mean Enterobacteriaceae counts of the samples from all the packaging treatments (c. 100% CO2; 75% CO2:25% N2; c. 80% O2 20% CO2 and c. 25% CO2:50% N2:25% O2) following 0 days bulk storage were initially (Day 0) low (
E. M. Buys, J. Kriiger, G. L. NortjO

300

=~.

~

~~

" ~

~

~

&

o

~

,

~

~

.

.

o

.

.

~

.

.

.

~

.

.

~

.

.

~

.

.

o

.

.

.

~

.

.

~

.

.

Pre-packaging of fresh pork retail cuts

301

6 2

~

~

°

~

o

~

~

m

~

°

°

°

m

~

~

.~

. . . . . . . . . .

.~~

o

~

~

~

°

o

~

~

~

~**>**~***~

~

E. M. Buys, J. Kr~iger, G. L. NortjO

302

After 14 days storage the count recorded for the samples bulk packed in c. 75% CO2:25% N2; c. 80% 02:20% CO2 was higher (log 1.34 cm 2) and differed from the count recorded for all the other treatments (
1 = 'extremely

pale',

5 •

'extremely dark'

:1 1

WIDO

I

I

I

r

I

I

I

I

I

F

2

4

W2D0

2

4

W3D0

2

4

W4DO

2

4

DISPLAY PERIOD (0"C)/ DAYS Fig. 1. Colour, odour and acceptability assessment of centralised bulk pre-packaged stored (0?C) and subsequent retail shelf life study (0°C) of pork loin cuts. ¢ , 100% CO2, [ , 2 5 % CO 2 75% N2; - - . - - , 25% CO 2 25% 02 50% N2; [] , 20% CO 2 80% 02; W l , day 0 storage, W2, day 7 storage, W3, day 14 storage, W4, day 21 storage.

Pre-packaging of fresh pork retail cuts

303

ODOUR ASSESSMENT ODOUR SCORE

t - 'no odour', 6 - 'completely o f f '

6

/

,

WID0

2

4

W2D0

2

4

W3D0

2

4

W4D0

2

4

DISPLAY PERIOD (0 °C)/ DAYS ACCEPTABILITY ASSESSMENT ACCEPTABILITY SCORE

t • 'totally unacceptable', 8 • 'very acceptable'

8

7

6

5

4

WID0

--

2

4

W2DO

2

4

W3D0

2

4

W4DO

DISPLAY PERIOD (0 °C)/ DAYS Fig. l.

contd.

2

4

304

E. M. Buys, J. Kriiger, G. L. Nortj~

samples stored in c. 80% 02:20% CO2 for 0 days was also found to be normal initially (Day 0), but as display progressed the colour paled. This trend seemed to persist throughout the rest of the storage period (Days 7, 14 or 21) and subsequent display periods. Odour assessment

The statistical analysis of the odour scores indicated that the packaging treatments (P--0.0004), storage period (P=0-0001) and display period (P --0.0001) effected the odour of the samples significantly. This was also true for the packaging treatment × storage period (P = 0-0001) and storage period × display period (P-0-0024) interactions. A fresh meat odour of the samples stored in 100% C 0 2 and in c. 75% CO2:25% N2 persisted through the whole extended bulk storage period (Days 0, 7, 14 or 21) and subsequent display periods, except on Day 4 of display following 21 days storage when the odour was judged as slightly off (Fig. 1). The samples from the c. 80% 02:20% CO2 and c. 25% CO2:50% N2:25% O2 bulk packaged treatments had a fresh meat odour until after 14 days bulk storage. Thereafter the odour scores increased as time progressed, i.e. after 14 days storage from fresh meat odour (Day 0) to slightly off (Day 4) and after 21 days storage from slightly off (Day 0) to moderately off (Day 4; c. 25% CO2: 50% N 2 : 25% 02) and slightly off (Day 4; c. 80% 02 : 20% C02). Acceptability assessment

According to the statistical analysis of the acceptability scores the packaging treatments (P -- 0-0001), storage period ( P = 0.0001), display period (P= 0-0001), packaging treatment × storage period (P--0-0001), packaging treatment × display period (P= 0.0001)and storage period x display period (P--0-0001) interactions all influenced acceptability significantly. All the samples were initially acceptable after 0 days storage (Fig. 1). The samples from the 100% CO 2 and 25% CO2:50% N2:25% O2 packaging treatments were judged moderately acceptable to slightly acceptable as the storage and subsequent display periods progressed. Only after 21 days storage and 2 days subsequent display were these samples moderately unacceptable to the consumer panel. The samples from the c. 75% CO2:25% N2 bulk packaging treatment also became unacceptable as time progressed, but were already slightly unacceptable after 7 days storage and moderately unacceptable after 14 days storage. The acceptability score for the samples from the c. 80% 0 2 : 2 0 % C O 2 followed the same trend as observed for the colour assessment. After 0 days storage the samples were acceptable but became slightly acceptable

Pre-paekaging of fresh pork retail cuts

305

as display progressed (Day 4). Similarly, after 7 days storage, the samples were found to be moderately acceptable to the panel, but after 4 days of display they judged the samples to be slightly unacceptable. This trend persisted throughout the remaining storage (Day 14 or 21) and subsequent display periods.

DISCUSSION The statistical analysis of the total bacterial counts indicated that the different bulk packaging treatments (c. 100% CO2 versus c. 75% CO2:25% N 2 versus c. 80% 02:20% CO2 versus c. 25% CO2 50% N2:25% 02) did not differ significantly over the whole bulk storage period. According to the mean total counts the 100% CO2 and c. 75% CO2:25% N 2 stored bulk packed samples could be stored for 21 days (c. 0°C) with a 4 day shelf life (c. 0°C) and the c. 25% CO2:50% N2:25% 02 and 80% 02:20% CO2 bulk packed samples could b e stored for 21 days with a 2 day shelf life. The mean total count was used as an indication of the end of retail shelf life, because according to Dainty et al. (1983), lean meat off-odours become evident during aerobic storage of lean meat when microbial numbers reach c. log 7 cm -2. It is thus assumed that the end of shelf life is reached when the retail cut reaches a psychrotrophic count of log 5 cm -2. This criterion is based on the fact that after being sold, the retail cut might still have a 1-2 day shelf life in a commercial refrigerator. The lactic acid bacteria counts did not differ significantly between bulk packaging treatments. This might indicate that the inclusion of 20-25% CO2 in a gas mixture is sufficient to effectively inhibit the aerobic spoilage organisms and thus to prolong the storage and shelf life of fresh pork retail cuts. Seideman & Durland (1984) reported that a CO2 concentration of 20% or higher will inhibit the growth of putrefactive bacteria (associated with meat spoilage) provided the meat is stored in the chilled condition. According to Kropf (1980) CO2 in increasing concentrations is more effective in microbial control. Brody (1989) reported that studies of respiration rates of a number of different organisms (such as Pseudomonas, Acinetobacter, Alteromonas, Yersinia and Enterobacter) as a function of CO2 in the atmosphere indicate that the level of CO2 that gave maximum inhibition for the common spoilage organisms was approximately 26% CO2, which gives the same inhibition as higher levels of CO2. The pseudomonad counts did not differ significantly between the bulk packaging treatments over the whole extended storage period. This might indicate that the laboratory gas replacement procedure was inefficient

306

E. M. Buys, J. Krfiger, G. L. NortjO

and thus sufficient residual 0 2 w a s present in the c. 100% CO 2 bulk packs, which promoted proliferation of the pseudomonads. Higher mean p s e u d o m o n a d counts were recorded for the c. 25% CO2:50% N2:25% 02 and c. 80% O2:20% CO2 bulk packed samples during display following 14 (
Pre-packaging of fresh pork retail cuts

307

acceptable 1 h after each bulk storage period than on the subsequent display days. This trend was also observed for the colour scores. As this trend was not observed for the other bulk packaging treatment which had 02 in the gas mixture (c. 25% CO2:50% N2:25% 02) it might be that storage in high percentages of 02 and subsequent display in air could adversely affect the colour of retail PVC-overwrapped pork chops. To conclude, it seems that according to the bacterial counts recorded, all four packaging treatments were successful in prolonging the storage life (21 days, c. 0°C) of centralised bulk pre-packaged pork retail cuts, while still ensuring a subsequent shelf life of at least 3 days (c. 0°C). Silliker et aL (1977) previously reported that the exposure of pork to CO2-enriched atmospheres not only resulted in anti-microbial effects but a residual effect as evidenced by continued inhibition of microbial growth after such meat is placed in an air atmosphere. The c. 100% CO2 bulk packaging treatment was the most successful, regarding storage and subsequent shelf life extension of fresh pork at c. 0°C, but as previously reported the colour-life (9 days) of these pork samples was shorter than the microbial shelf life (21 days) (Silliker et al., 1977; Gill & Harrison, 1989; Seideman & Durland, 1984; Scholtz et al., 1992). It thus seems that the c. 25% CO2:50% N2:25% O2 gas mixture was the most successful bulk packaging treatment regarding acceptability and colour scores, although a storage period of 14 days (c. 0°C) and subsequent shelf life of 2 days (c. 0°C) is recommended regarding the odour scores. According to Taylor (1990) bulk pre-packaging with high O2/CO2 gas mixtures, which is less expensive than individually packed MA trays, could be equally effective in prolonging the red meat colour and could provide a serious challenge to some of the longer established methods used in the centralised packing and distribution of fresh meat.

REFERENCES Anon. (1981). Agriculture Canada, Pub. 5/80/B. Blickstad, E. & Molin, G., (1983). J. Food Prot., 46, 756 Brody, A. L. (Ed.) (1989). In Controlled~Modified Atmosphere~Vacuum Packaging of Foods. Food and Nutrition Press, Trumbull, CT. USA. Dainty, R. H., Shaw, B. G. & Roberts, T. A. (1983). In Food Microbiology: Advances and Prospects, ed. T. A. Roberts, & Skinner, F. A. Academic Press, London. De Man, J. C., Rogosa, M. & Sharpe, M. E. (1960). J. Appl. Bacteriol., 30, 500. Gardner, G. A. (1966). J. Appl. Bacteriol., 29, 455. Gill, C. O. & Harrison, J. C. L. (1989). Meat Sci., 26, 313. Grau, H. F. (1979). Appl. Environ. Microbiol., 38, 818. Kielwein, G. (1971). Arch. Lebensmittelhyg., 22, 29.

308

E. M. Buys, J. Krgiger, G. L. Nortjd

Kropf, D.H. (1980). Proc. Rec. Meat Conf., 33, 15. Newton, K. G. & Gill, C. O. (1978). J. Appl. Bacteriol., 44, 91: Nortj6, G. L., Swanepoel, E., Naud6, R. T., Holzapfel, W. H. & Steyn, P. L. (1982). J. Food Prot., 45, 1016. Sakazaki, R., Nanioka, S., Osada, A. & Yamada, C. A., (1960). J. Exp. Med., 30, 13. Scholtz~ E. M., Jordaan, E., KrUger, J., Nortj6, G. L. & Naud6, R. T. (1992). Meat Sci., 32, 11. Seideman, S. C & Durland, P. R. (1984). J. Food Qual., 6, 239. Silliker, J. H., Woodruff, R. E., Lugg, J. R., Wolfe, S. K. & Brown, D. (1977). Meat Sci., 1, 195. Taylor, A. A. (1990). 36th Int. Congress Meat Sci. Technol. Havana, Cuba, p. 346.