- Email: [email protected]
The colour of the adductor muscle as a predictor of pork quality in the loin
MEAT SCIENCE Meat Science 73 (2006) 565–569 www.elsevier.com/locate/meatsci
The colour of the adductor muscle as a predictor of pork quality in the loin P.D. Warriss a
a,*
, S.N. Brown a, P. Pas´ciak
b
School of Veterinary Science, University of Bristol, Langford, Bristol BS40 5DU, UK b Ecopig, 42-510 Wojkowice Kos´cielne 28, Poland Received 8 August 2005; accepted 6 February 2006
Abstract The relation between measurements of colour made in the m. adductor (AD) at 45 min or 20 h post mortem and the quality, assessed subjectively in terms of colour and waterholding capacity, of the m. longissimus (LD) in the loin was examined. The study used data from 100 pig carcasses exhibiting a wide range of meat quality from extreme PSE (pale, soft and exudative) to extreme DFD (dark, firm and dry). The subjective assessments were confirmed by objective measures of paleness (reflectance) and waterholding capacity (drip loss in storage) in the LD. Lightness (L*) measured at 20 h post mortem in the AD was the best potential predictor of loin muscle quality, explaining 59% of the variation in subjective and objective quality measures. Comparable measurements at 45 min post mortem explained between 21% and 44% of the variation. The equation that described the relation between AD Lightness (L*) and subjectively assessed LD quality was derived. This could be used to transpose the AD L* values from a population of slaughtered pigs into nominal subjective scores for the LD, allowing the frequency of the five subjective quality groups (extremely DFD, slightly DFD, normal, slightly PSE, and extremely PSE) in the population of carcasses to be defined. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Pork; Colour; Quality; Prediction
1. Introduction Whether pork has normal characteristics or is PSE (pale, soft and exudative) is one of the most important determinants of its eating quality. It also has implications for consumer acceptability of the raw meat (Topel et al., 1976; Wachholz, Kauffman, Henderson, & Lockner, 1978) and its technological and processing properties (Kauffman, Wachholz, Henderson, & Lochner, 1978). The Pork Chain Quality Audit, carried out in the USA and reported by Cannon et al. (1996), identified inadequate muscle colour and water holding capacity as ranking second, after excessive fat, as the primary Industry concern about pork quality. Because of this there is considerable concern over the *
Corresponding author. Tel.: +44 0117 928 9261; fax: +44 0117 928 9582. E-mail address: [email protected] (P.D. Warriss). 0309-1740/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.meatsci.2006.02.009
current apparent high prevalence of the problem (Warriss, 2000, chap. 7), although precise and reliable estimates of this for most countries are difficult to find. One reason for this is that the term PSE is a subjective description rather than an objective definition. Different individuals’ perceptions of what is PSE therefore to some degree vary. This has led to proxy measurements, such as muscle pH or paleness (Kauffman et al., 1993), often being used as indicators of meat with potential PSE characteristics, especially when there is a need to sample large numbers of carcasses. However, the relationships between pH, and colour and water holding capacity (WHC), are complex and non-linear, probably because the effects on colour and WHC relate to denaturation of different protein components of the muscle (Murray, 1995, chap. 4). The pH value measured soon after the death of the animal (645 min) is therefore not a completely reliable indicator of meat quality and measurements of pH tend to give
566
P.D. Warriss et al. / Meat Science 73 (2006) 565–569
higher estimates of the prevalence of PSE carcasses than subjectively assessed colour and WHC. Recent studies have therefore concentrated on looking for techniques that predict WHC specifically, rather than the occurrence of PSE, using techniques such as near infrared spectroscopy (Forrest et al., 2000), electrical conductivity (Lee et al., 2000) or Fourier transform infrared spectroscopy (Pedersen & Engelsen, 2001). Another problem is that it is difficult to examine the appearance of the commercially important muscles in the intact, freshly dressed pork carcass. In particular, the m. longissimus dorsi (LD), which is one of the commercially most valuable muscles, and one that is especially prone to exhibiting PSE characteristics, is not visible. Various electronic probes have therefore been developed, based on differences in the optical or electrical properties of PSE meat (see, Swatland, 1995). Examples of their commercial use have been given by Oliver, Gispert, Tibau, and Diestre (1991), Chizzolini, Novelli, Badiani, Rosa, and Delbono (1993) Whitman, Forrest, Morgan, and Okos (1996). However, it has been found (Chizzolini et al., 1993; Swatland, 1985), that both pH electrodes and other electronic probes suffer from disadvantages when they are used routinely in the relatively harsh environments and conditions found in slaughter and processing plants. By contrast with most large muscles, the surface of the m. adductor (AD) in the pelvic region is inevitably exposed in at least one side when the carcass is split by cutting down through the backbone. We have therefore examined the potential usefulness of objective measures of colour made on the cut surface of the AD to predict the subjectively assessed quality of the LD in intact pig carcasses. Colour can be objectively and reliably measured using portable tristimulus colour analysers (Warriss, 1995). As well as the value of these measures to predict PSE carcasses, for completeness we also looked at their value to identify carcasses in which the LD showed DFD (dark, firm and dry) characteristics. The use of measurements of electrical capacitance, resistivity and paleness (CIE percent Y) in the AD to predict LD quality, have been described previously by Swatland (1982), although based on a relatively small sample of eight pigs. 2. Materials and methods Measurements were made in 100 carcasses (mean weight 69.9 ± 4.6 (sd) kg) from female pigs of predominantly white breeding killed using normal commercial practices in the University abattoir at Langford. Measurements made on these carcasses have been reported previously (Brown, 1992; Lopez-Bote, Warriss, & Brown, 1989; Warriss, Brown, Lopez-Bote, Bevis, & Adams, 1989). At 45 min post mortem a sample of LD was removed and frozen in liquid nitrogen pending later measurement of the pH (pH45) using a glass electrode after homogenisation in 5 mM sodium iodoacetate, 150 mM KCl, pH 7.0. The colour of the freshly exposed surface of the AD was deter-
mined by making triplicate readings using the CIELAB L*a*b* colour space (Warriss, 1995) with a Chromameter Reflectance II instrument (Minolta (UK) Limited, Milton Keynes) after being allowed to ‘‘bloom’’ for about 15 min. Blooming allows the reaction of reduced purple myoglobin and haemoglobin at the muscle surface with oxygen to convert them into the bright red oxygenated forms. Brewer, Zhu, Bidner, Meisinger, and McKeith (2001) showed that the rate of blooming was constant over a wide range of pig muscles and moreover did not affect L* value. Small changes in a* and b* chromaticity coordinates were complete within 10 min after exposure of the fresh muscle surface to the air. At about 20 h post mortem a thin slice of the surface of the same AD was removed to expose a fresh cut surface and, after allowing the meat to ‘‘bloom’’ for 15 min, a further three colour readings were made on the fresh meat surface. A slice of the LD about 2.5 cm thick, just posterior to the head of the last rib, was removed and used for subjective assessment of colour-structure. This was done by a panel of seven judges, experienced in the assessment of meat quality, working independently and using a five-point scale (1 = extremely DFD, 2 = slightly DFD, 3 = normal, 4 = slightly PSE, 5 = extremely PSE). This scale has been characterised and validated by relating it to objective quality measures (Warriss & Brown, 1993). The subjective assessment score for each sample was the mean of the values given by the seven judges. A 10 cm length of LD posterior to the subjectively assessed slice was also removed and used for determination of ultimate pH (pHu), drip loss (Warriss, 1982), reflectance using an EEL Reflectometer (MacDougall, Cuthbertson, & Smith, 1969) and CIELAB colour, as described above for the AD. We measured reflectance using the EEL Reflectometer because values above 50 have been used to define PSE meat (MacDougall et al., 1969). Moreover, the UK Meat and Livestock Commission (MLC) blueprint standards for high quality pork specify that the colour of the LD should be consistent with EEL values between 30 and 55. For analysis the carcasses were divided into three groups (PSE, normal and DFD) based on the quality characteristics of the LD. For the purpose of this work, meat was classified as PSE if six out of the seven judges assessed it as slightly or extremely PSE (scores of 4 or 5). Similarly it was classed as DFD if it was assessed as slightly or extremely DFD (scores of 1 or 2) by six out of the seven judges. Linear correlation coefficients were calculated between selected measurements made in the LD and those made in the AD. 3. Results Sixteen carcasses were classed as having PSE loins, 61 normal loins and 23 DFD loins. The characteristics of the loins in the three classes have been reported previously (Lopez-Bote et al., 1989) but, for clarity, the most important are shown in Table 1. These confirmed the subjective
P.D. Warriss et al. / Meat Science 73 (2006) 565–569
567
Table 1 Characteristics of the LD muscles of the carcasses classed as PSE, normal or DFD (means ± SD) PSE (n = 16)
Normal (n = 61)
a
5.60 ± 0.236 5.37a ± 0.080 59.5a ± 7.79 61.0a ± 3.35 13.0a ± 1.76
pH45 pHu Reflectance (EEL value) L* measured at 20 h pm % drip loss in storage
DFD (n = 23)
b
6.39c ± 0.169 6.33b ± 0.292 27.2c ± 3.56 45.4c ± 2.38 2.27c ± 1.42
6.05 ± 0.256 5.45a ± 0.159 44.2b ± 6.22 54.3b ± 3.12 10.3b ± 2.88
Means with different superscripts are significantly different (P < 0.001).
assessments and are consistent with the values and criteria for PSE and DFD meat given by Warner, Kauffman, and Russell (1993). Compared with normal loins, PSE loins had low pH45 values, high reflectance (EEL values) and lost large amounts of drip (exudate) during storage. DFD loins had a high pHu, low reflectance and very low loss of drip during storage. The mean values for the colour measurements made in the AD muscle at 45 min and 20 h post mortem are given in Table 2. At 45 min post mortem there were significant differences between PSE and normal groups of carcasses for all colour measurements made in the AD (L*, a*, b*, hue and saturation). However, differences between normal and DFD groups of carcasses were significant for only L*, b* and hue. L* and a* values measured at 20 h post mortem were significantly different between all three carcass groups. The b* value, hue or saturation could not differentiate between PSE and normal groups, although carcasses in the DFD group had significantly lower values. Correlation coefficients were calculated between the colour measurements made at both 45 min and 20 h post mortem in the AD and the subjective assessment score, reflectance (EEL value) and drip loss in the LD (Table 3). Correlations were higher at 20 h than at 45 min post mortem between all measurements. At both times the L* values of the AD were the best potential predictors of
Table 2 Colour measurements in the AD muscles of the carcasses with loins classed as PSE, normal or DFD (means together with the overall residual mean square (rms) and overall significance of differences between means) PSE (n = 16)
Normal (n = 61)
DFD (n = 23)
rms
Table 3 Correlation coefficients between measurements made in the AD at two times post mortem and the characteristics of the LD from the same carcass Reflectance (EEL value)
Drip loss
Subjective assessment score
Measured at 45 min pm L* 0.66 a* 0.29 b* 0.56 Hue 0.57 Saturation 0.35
0.46 0.20 0.38 0.38 0.24
0.56 0.26 0.48 0.47 0.31
Measured at 20 h pm 0.77 L* a* 0.48 b* 0.59 Hue 0.55 Saturation 0.54
0.78 0.54 0.64 0.60 0.60
0.78 0.53 0.67 0.63 0.60
For P = 0.05, r = 0.195; for P = 0.01, r = 0.254; and for P = 0.001, r = 0.321.
LD reflectance, drip or subjectively assessed quality. At 20 h post mortem, the L* value explained 56% of the variation in each of these variables. The next best potential predictor was the b* chromaticity coordinate. However, this explained only between 35% and 50% of the variation in LD quality. The a* chromaticity coordinate was less valuable and there was no improvement in the potential predictive value if the information from a* and b* values was combined to calculate hue or saturation. Measurement of L* value at 45 min post mortem explained only 16–38% of variation in LD quality. The relation between the L* value measured in the AD at 20 h and the subjective score based on the LD (Fig. 1) was described by the regression equation: Subjective score of LD
Measured at 45 min pm 42.6a L* a* 11.3a * b 4.6a Hue 21.5a Saturation 12.2a
38.5 10.2b 3.2b 17.2b 10.7b
37.3 9.8b 2.7c 15.3c 10.2b
2.97* 1.09*** 12.61*** 3.61**
Measured at 20 h pm L* 46.6a * a 13.1a b* 6.5a Hue 26.0a Saturation 14.6a
44.6b 12.3b 5.9a 25.4a 13.6a
38.3c 10.5c 3.3b 17.3b 11.0b
7.82*** 2.07*** 1.82*** 19.9*** 2.89***
b
c
5.86***
*P < 0.05, **P < 0.01, ***P < 0.001; means with different superscripts are significantly different (P < 0.05).
¼ 0:1747 L value of AD 4:6571 ðR2 ¼ 0:60Þ From this it can be calculated that the L* values of the AD corresponding to the five subjective scores for the LD are: score 1 (extremely DFD) L* = 32.4, score 2 (slightly DFD) L* = 38.1, score 3 (normal) L* = 43.8, score 4 (slightly PSE) L* = 49.6 and score 5 (extremely PSE) L* = 55.3. Using this equation the AD L* values from a population of slaughtered pigs could similarly be transposed into nominal subjective scores for the LD, allowing the frequency of the five quality groups in the population of carcasses to be defined.
568
P.D. Warriss et al. / Meat Science 73 (2006) 565–569
4.0
Table 4 The range of L* values measured in the AD at two times post mortem in carcasses classed as PSE, normal or DFD based on the characteristics of the LD
3.0
L*
Subjective score in loin
5.0
45 min L* 20 h
2.0
1.0
0.0 30.0
40.0
50.0
60.0
AD lightness (L*) Fig. 1. The relationship between Lightness (L*) measured in the AD at 20 h post mortem and the mean subjective assessment score made in the LD of the loin (1 = extremely DFD, 2 = slightly DFD, 3 = normal, 4 = slightly PSE, 5 = extremely PSE).
4. Discussion The average values for the colour measurements made in the AD at both 45 min and 20 h post mortem varied progressively in the meat quality groups (PSE, normal and DFD) as defined by subjective assessment of the LD, and confirmed by objective measures of pH45, reflectance (EEL value) and waterholding (drip loss). This implied that the quality of the AD reflected that of the LD. This was confirmed by the significant correlations found between subjective and objective measures of LD quality and colour measurements made in the AD. In all cases these were higher at 20 h than at 45 min post mortem when the whole population of 100 carcasses was considered. This is in accord with previous findings (see, Kauffman et al., 1993) and probably for reasons discussed by Swatland (1997). The highest correlations were with CIELAB L* values, indicating that, overall, it is the lightness component of colour of the AD, rather than hue or saturation, that is potentially the best indicator of poor quality in the loin. Brewer et al. (2001) also found that the L* value in all the muscles they studied (not unfortunately including the AD) was the instrumental value most highly correlated with subjectively assessed colour when this was based on the Japanese Colour Standards. Swatland (1982) likewise found that paleness in the AD measured 7 days post mortem explained 56% of the paleness in the LD. L* value has the added advantage that it is not apparently affected by whether the muscle surface has had a chance to bloom (Brewer et al., 2001), so no delay between cutting a fresh surface and measurement is necessary. Unfortunately, despite the correlations found in the present study between L* value and loin quality, because of the overlap of the distribution of values between the quality groups, it was not possible to use L* values made at either time post mortem to identify categorically individual carcasses as PSE, normal or
PSE (n = 16)
Normal (n = 61)
DFD (n = 23)
37–48 42–51
34–45 38–50
35–41 35–45
DFD (Table 4). However, because the LD and AD have different fibre compositions (Ruusunen & Puolanne, 1988), the AD being a redder muscle with a more oxidative metabolism, the AD would not be expected to be a perfect predictor of LD quality. In conclusion, the use of colour measurements, particularly L* values, made on the cut surface of the AD exposed on splitting the pig carcass appears to have potential for predicting the meat quality in the more valuable, but less accessible loin. Better prediction would generally result from measurements made at 20 h post mortem after chilling, but carcasses showing extreme PSE would probably be identifiable using measurements made at 45 min post mortem, just after carcasses had entered the chill room. Measurements of L* values on the cut surface of the AD could be used to predict the relative prevalence of carcasses showing PSE and DFD in populations of slaughtered pigs. The method would however probably not be sufficiently precise for selection of individual carcasses on the slaughter line so that those that showing poor meat quality could be processed differently. Portable tristimulus colour meters are relatively widely available, and the CIELAB system for measuring colour is well established and it is easy to standardise measurements reliably using standard white and coloured tiles. The method could therefore be useful for measuring the quality of populations of pork carcasses under commercial conditions, for example to monitor dayto-day variation, or variation between plants, or potential changes in quality caused by changes in production procedures. It might have advantages over current techniques using pH electrodes or electronic quality probes such as carcass grading probes and the Fibre Optic Probe (Fortin & Raymond, 1988; Warriss et al., 1989) or the Pork Quality Meter (Warriss, Brown, & Adams, 1991). References Brewer, M. S., Zhu, L. G., Bidner, B., Meisinger, D. J., & McKeith, F. K. (2001). Measuring pork colour: effects of bloom time, muscle, pH and relationship to instrumental parameters. Meat Science, 57, 169–176. Brown, S. N. (1992). A note on the use of subjective methods for assessing pig meat quality on the slaughterline. Meat Science, 32, 195–202. Cannon, J. E., Morgan, J. B., McKeith, F. K., Smith, G. C., Sonka, S., Heavner, J., et al. (1996). Pork Chain Quality Audit: quantification of pork quality characteristics. Journal of Muscle Foods, 7, 29–44. Chizzolini, R., Novelli, E., Badiani, A., Rosa, P., & Delbono, G. (1993). Objective measures of pork quality: evaluation of various techniques. Meat Science, 34, 49–77. Forrest, J. C., Morgan, M. T., Borgaaerd, C., Rasmussen, A. J., Jespersen, B. L., & Andersen, J. R. (2000). Development of technology for the early postmortem prediction of waterholding capacity and drip loss in fresh pork. Meat Science, 55, 115–122.
P.D. Warriss et al. / Meat Science 73 (2006) 565–569 Fortin, A., & Raymond, D. P. (1988). The use of the electrical characteristics of muscle for the objective determination of PSE and DFD in pork carcasses under commercial conditions. Canadian Institute of Food Science and Technology Journal, 21, 260–265. Kauffman, R. G., Sybesma, W., Smulders, F. J. M., Eikelenboom, E., Engel, B., van Laack, R. J. L. M., et al. (1993). The effectiveness of examining early post mortem musculature to predict ultimate pork quality. Meat Science, 34, 283–300. Kauffman, R. G., Wachholz, D., Henderson, D., & Lochner, J. V. (1978). Shrinkage of PSE, normal and DFD hams during transit and processing. Journal of Animal Science, 46, 1236–1240. Lee, S., Norman, J. M., Gunasekaran, S., van Laack, R. L. J. M, Kim, B. C., & Kauffman, R. G. (2000). Use of electrical conductivity to predict water-holding capacity in post-rigor pork. Meat Science, 55, 385–389. Lopez-Bote, C., Warriss, P. D., & Brown, S. N. (1989). The use of muscle protein solubility measurements to assess pig lean meat quality. Meat Science, 26, 167–175. MacDougall, D. B., Cuthbertson, A., & Smith, R. J. (1969). The assessment of pig meat paleness by reflectance photometry. Animal Production, 11, 243–246. Murray, A. C. (1995). The evaluation of meat quality. In S. D. Morgan Jones (Ed.), Quality and grading of carcasses of meat animals (pp. 83–107). Boca Raton, USA: CRC Press. Oliver, M. A., Gispert, M., Tibau, J., & Diestre, A. (1991). The measurement of light scattering and electrical conductivity for the prediction of PSE pig meat at various times postmortem. Meat Science, 29, 141–151. Pedersen, D. K., & Engelsen, S. B. (2001). Early prediction of the quality of porcine meat by FT-IR spectroscopy. In Proceedings of the 47th international congress of meat science and technology (pp. 204–205). Krakow, Poland, Paper 3-15.. Ruusunen, M., & Puolanne, E. (1988). Fiber-type distribution in porcine muscles. In Proceedings of the 34th international congress of meat science and technology (pp. 52–54). Qld, Australia.. Swatland, H. J. (1982). Intermuscular variation in physical properties of pork. Canadian Institute of Food Science and Technology Journal, 15, 92–95. Swatland, H. J. (1985). Optical and electronic methods of measuring pH and other predictors of meat quality in pork carcasses. Journal of Animal Science, 61, 887–891.
569
Swatland, H. J. (1995). On-line evaluation of meat. Lancaster and Basel: Technomic Publishing Company, Inc., 347 pp. Swatland, H. J. (1997). Observations on rheological, electrical, and optical changes during rigor development in pork and beef. Journal of Animal Science, 75, 975–985. Topel, D. G., Miller, J. A., Berger, P. J., Rust, R. E., Parrish, F. C., & Ono, K. (1976). Palatability and visual appearance of dark, normal and pale colored porcine m. longissimus. Journal of Food Science, 41, 628–630. Wachholz, D., Kauffman, R. G., Henderson, D., & Lockner, J. V. (1978). Consumer discrimination of pork color at the market place. Journal of Food Science, 43, 1150–1152. Warner, R. D., Kauffman, R. G., & Russell, R. L. (1993). Quality attributes of major porcine muscles: a comparison with the Longissimus lumborum. Meat Science, 33, 359–393. Warriss, P. D. (1982). The relationship between pH45 and drip in pig muscle. Journal of Food Technology, 17, 573–578. Warriss, P. D. (1995). Instrumental measurement of colour. In A. A. Taylor, A. Raimundo, M. Severini, & F. J. H. Smulders (Eds.), Meat quality and meat packaging (pp. 221–232). Nijmegen, Utrecht, The Netherlands: European Consortium for Continuing Education in Advanced Meat Science and Technology. Audet Tijdschriften b.v.. Warriss, P. D. (2000). Meat science: An introductory text. Wallingford, Oxon, OX10 8DE: CAB International, 312pp. Warriss, P. D., & Brown, S. N. (1993). Relationships between the subjective assessment of pork quality and objective measures of colour. In J. D. Wood & T. L. J. Lawrence (Eds.), Safety and quality of food from animals (pp. 98–101). Edinburgh: Occasional Publication of the British Society of Animal Production No. 17, British Society of Animal Production. Warriss, P. D., Brown, S. N., & Adams, S. J. M. (1991). Use of the Tecpro pork quality meter for assessing meat quality on the slaughterline. Meat Science, 30, 147–156. Warriss, P. D., Brown, S. N., Lopez-Bote, C., Bevis, E. A., & Adams, S. J. M. (1989). Evaluation of lean meat quality in pigs using two electronic probes. Meat Science, 25, 281–291. Whitman, T. A., Forrest, J. C., Morgan, M. T., & Okos, M. R. (1996). Electrical measurements for detecting early postmortem changes in porcine muscle. Journal of Animal Science, 74, 80–90.
The colour of the adductor muscle as a predictor of pork quality in the loin P.D. Warriss a
a,*
, S.N. Brown a, P. Pas´ciak
b
School of Veterinary Science, University of Bristol, Langford, Bristol BS40 5DU, UK b Ecopig, 42-510 Wojkowice Kos´cielne 28, Poland Received 8 August 2005; accepted 6 February 2006
Abstract The relation between measurements of colour made in the m. adductor (AD) at 45 min or 20 h post mortem and the quality, assessed subjectively in terms of colour and waterholding capacity, of the m. longissimus (LD) in the loin was examined. The study used data from 100 pig carcasses exhibiting a wide range of meat quality from extreme PSE (pale, soft and exudative) to extreme DFD (dark, firm and dry). The subjective assessments were confirmed by objective measures of paleness (reflectance) and waterholding capacity (drip loss in storage) in the LD. Lightness (L*) measured at 20 h post mortem in the AD was the best potential predictor of loin muscle quality, explaining 59% of the variation in subjective and objective quality measures. Comparable measurements at 45 min post mortem explained between 21% and 44% of the variation. The equation that described the relation between AD Lightness (L*) and subjectively assessed LD quality was derived. This could be used to transpose the AD L* values from a population of slaughtered pigs into nominal subjective scores for the LD, allowing the frequency of the five subjective quality groups (extremely DFD, slightly DFD, normal, slightly PSE, and extremely PSE) in the population of carcasses to be defined. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Pork; Colour; Quality; Prediction
1. Introduction Whether pork has normal characteristics or is PSE (pale, soft and exudative) is one of the most important determinants of its eating quality. It also has implications for consumer acceptability of the raw meat (Topel et al., 1976; Wachholz, Kauffman, Henderson, & Lockner, 1978) and its technological and processing properties (Kauffman, Wachholz, Henderson, & Lochner, 1978). The Pork Chain Quality Audit, carried out in the USA and reported by Cannon et al. (1996), identified inadequate muscle colour and water holding capacity as ranking second, after excessive fat, as the primary Industry concern about pork quality. Because of this there is considerable concern over the *
Corresponding author. Tel.: +44 0117 928 9261; fax: +44 0117 928 9582. E-mail address: [email protected] (P.D. Warriss). 0309-1740/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.meatsci.2006.02.009
current apparent high prevalence of the problem (Warriss, 2000, chap. 7), although precise and reliable estimates of this for most countries are difficult to find. One reason for this is that the term PSE is a subjective description rather than an objective definition. Different individuals’ perceptions of what is PSE therefore to some degree vary. This has led to proxy measurements, such as muscle pH or paleness (Kauffman et al., 1993), often being used as indicators of meat with potential PSE characteristics, especially when there is a need to sample large numbers of carcasses. However, the relationships between pH, and colour and water holding capacity (WHC), are complex and non-linear, probably because the effects on colour and WHC relate to denaturation of different protein components of the muscle (Murray, 1995, chap. 4). The pH value measured soon after the death of the animal (645 min) is therefore not a completely reliable indicator of meat quality and measurements of pH tend to give
566
P.D. Warriss et al. / Meat Science 73 (2006) 565–569
higher estimates of the prevalence of PSE carcasses than subjectively assessed colour and WHC. Recent studies have therefore concentrated on looking for techniques that predict WHC specifically, rather than the occurrence of PSE, using techniques such as near infrared spectroscopy (Forrest et al., 2000), electrical conductivity (Lee et al., 2000) or Fourier transform infrared spectroscopy (Pedersen & Engelsen, 2001). Another problem is that it is difficult to examine the appearance of the commercially important muscles in the intact, freshly dressed pork carcass. In particular, the m. longissimus dorsi (LD), which is one of the commercially most valuable muscles, and one that is especially prone to exhibiting PSE characteristics, is not visible. Various electronic probes have therefore been developed, based on differences in the optical or electrical properties of PSE meat (see, Swatland, 1995). Examples of their commercial use have been given by Oliver, Gispert, Tibau, and Diestre (1991), Chizzolini, Novelli, Badiani, Rosa, and Delbono (1993) Whitman, Forrest, Morgan, and Okos (1996). However, it has been found (Chizzolini et al., 1993; Swatland, 1985), that both pH electrodes and other electronic probes suffer from disadvantages when they are used routinely in the relatively harsh environments and conditions found in slaughter and processing plants. By contrast with most large muscles, the surface of the m. adductor (AD) in the pelvic region is inevitably exposed in at least one side when the carcass is split by cutting down through the backbone. We have therefore examined the potential usefulness of objective measures of colour made on the cut surface of the AD to predict the subjectively assessed quality of the LD in intact pig carcasses. Colour can be objectively and reliably measured using portable tristimulus colour analysers (Warriss, 1995). As well as the value of these measures to predict PSE carcasses, for completeness we also looked at their value to identify carcasses in which the LD showed DFD (dark, firm and dry) characteristics. The use of measurements of electrical capacitance, resistivity and paleness (CIE percent Y) in the AD to predict LD quality, have been described previously by Swatland (1982), although based on a relatively small sample of eight pigs. 2. Materials and methods Measurements were made in 100 carcasses (mean weight 69.9 ± 4.6 (sd) kg) from female pigs of predominantly white breeding killed using normal commercial practices in the University abattoir at Langford. Measurements made on these carcasses have been reported previously (Brown, 1992; Lopez-Bote, Warriss, & Brown, 1989; Warriss, Brown, Lopez-Bote, Bevis, & Adams, 1989). At 45 min post mortem a sample of LD was removed and frozen in liquid nitrogen pending later measurement of the pH (pH45) using a glass electrode after homogenisation in 5 mM sodium iodoacetate, 150 mM KCl, pH 7.0. The colour of the freshly exposed surface of the AD was deter-
mined by making triplicate readings using the CIELAB L*a*b* colour space (Warriss, 1995) with a Chromameter Reflectance II instrument (Minolta (UK) Limited, Milton Keynes) after being allowed to ‘‘bloom’’ for about 15 min. Blooming allows the reaction of reduced purple myoglobin and haemoglobin at the muscle surface with oxygen to convert them into the bright red oxygenated forms. Brewer, Zhu, Bidner, Meisinger, and McKeith (2001) showed that the rate of blooming was constant over a wide range of pig muscles and moreover did not affect L* value. Small changes in a* and b* chromaticity coordinates were complete within 10 min after exposure of the fresh muscle surface to the air. At about 20 h post mortem a thin slice of the surface of the same AD was removed to expose a fresh cut surface and, after allowing the meat to ‘‘bloom’’ for 15 min, a further three colour readings were made on the fresh meat surface. A slice of the LD about 2.5 cm thick, just posterior to the head of the last rib, was removed and used for subjective assessment of colour-structure. This was done by a panel of seven judges, experienced in the assessment of meat quality, working independently and using a five-point scale (1 = extremely DFD, 2 = slightly DFD, 3 = normal, 4 = slightly PSE, 5 = extremely PSE). This scale has been characterised and validated by relating it to objective quality measures (Warriss & Brown, 1993). The subjective assessment score for each sample was the mean of the values given by the seven judges. A 10 cm length of LD posterior to the subjectively assessed slice was also removed and used for determination of ultimate pH (pHu), drip loss (Warriss, 1982), reflectance using an EEL Reflectometer (MacDougall, Cuthbertson, & Smith, 1969) and CIELAB colour, as described above for the AD. We measured reflectance using the EEL Reflectometer because values above 50 have been used to define PSE meat (MacDougall et al., 1969). Moreover, the UK Meat and Livestock Commission (MLC) blueprint standards for high quality pork specify that the colour of the LD should be consistent with EEL values between 30 and 55. For analysis the carcasses were divided into three groups (PSE, normal and DFD) based on the quality characteristics of the LD. For the purpose of this work, meat was classified as PSE if six out of the seven judges assessed it as slightly or extremely PSE (scores of 4 or 5). Similarly it was classed as DFD if it was assessed as slightly or extremely DFD (scores of 1 or 2) by six out of the seven judges. Linear correlation coefficients were calculated between selected measurements made in the LD and those made in the AD. 3. Results Sixteen carcasses were classed as having PSE loins, 61 normal loins and 23 DFD loins. The characteristics of the loins in the three classes have been reported previously (Lopez-Bote et al., 1989) but, for clarity, the most important are shown in Table 1. These confirmed the subjective
P.D. Warriss et al. / Meat Science 73 (2006) 565–569
567
Table 1 Characteristics of the LD muscles of the carcasses classed as PSE, normal or DFD (means ± SD) PSE (n = 16)
Normal (n = 61)
a
5.60 ± 0.236 5.37a ± 0.080 59.5a ± 7.79 61.0a ± 3.35 13.0a ± 1.76
pH45 pHu Reflectance (EEL value) L* measured at 20 h pm % drip loss in storage
DFD (n = 23)
b
6.39c ± 0.169 6.33b ± 0.292 27.2c ± 3.56 45.4c ± 2.38 2.27c ± 1.42
6.05 ± 0.256 5.45a ± 0.159 44.2b ± 6.22 54.3b ± 3.12 10.3b ± 2.88
Means with different superscripts are significantly different (P < 0.001).
assessments and are consistent with the values and criteria for PSE and DFD meat given by Warner, Kauffman, and Russell (1993). Compared with normal loins, PSE loins had low pH45 values, high reflectance (EEL values) and lost large amounts of drip (exudate) during storage. DFD loins had a high pHu, low reflectance and very low loss of drip during storage. The mean values for the colour measurements made in the AD muscle at 45 min and 20 h post mortem are given in Table 2. At 45 min post mortem there were significant differences between PSE and normal groups of carcasses for all colour measurements made in the AD (L*, a*, b*, hue and saturation). However, differences between normal and DFD groups of carcasses were significant for only L*, b* and hue. L* and a* values measured at 20 h post mortem were significantly different between all three carcass groups. The b* value, hue or saturation could not differentiate between PSE and normal groups, although carcasses in the DFD group had significantly lower values. Correlation coefficients were calculated between the colour measurements made at both 45 min and 20 h post mortem in the AD and the subjective assessment score, reflectance (EEL value) and drip loss in the LD (Table 3). Correlations were higher at 20 h than at 45 min post mortem between all measurements. At both times the L* values of the AD were the best potential predictors of
Table 2 Colour measurements in the AD muscles of the carcasses with loins classed as PSE, normal or DFD (means together with the overall residual mean square (rms) and overall significance of differences between means) PSE (n = 16)
Normal (n = 61)
DFD (n = 23)
rms
Table 3 Correlation coefficients between measurements made in the AD at two times post mortem and the characteristics of the LD from the same carcass Reflectance (EEL value)
Drip loss
Subjective assessment score
Measured at 45 min pm L* 0.66 a* 0.29 b* 0.56 Hue 0.57 Saturation 0.35
0.46 0.20 0.38 0.38 0.24
0.56 0.26 0.48 0.47 0.31
Measured at 20 h pm 0.77 L* a* 0.48 b* 0.59 Hue 0.55 Saturation 0.54
0.78 0.54 0.64 0.60 0.60
0.78 0.53 0.67 0.63 0.60
For P = 0.05, r = 0.195; for P = 0.01, r = 0.254; and for P = 0.001, r = 0.321.
LD reflectance, drip or subjectively assessed quality. At 20 h post mortem, the L* value explained 56% of the variation in each of these variables. The next best potential predictor was the b* chromaticity coordinate. However, this explained only between 35% and 50% of the variation in LD quality. The a* chromaticity coordinate was less valuable and there was no improvement in the potential predictive value if the information from a* and b* values was combined to calculate hue or saturation. Measurement of L* value at 45 min post mortem explained only 16–38% of variation in LD quality. The relation between the L* value measured in the AD at 20 h and the subjective score based on the LD (Fig. 1) was described by the regression equation: Subjective score of LD
Measured at 45 min pm 42.6a L* a* 11.3a * b 4.6a Hue 21.5a Saturation 12.2a
38.5 10.2b 3.2b 17.2b 10.7b
37.3 9.8b 2.7c 15.3c 10.2b
2.97* 1.09*** 12.61*** 3.61**
Measured at 20 h pm L* 46.6a * a 13.1a b* 6.5a Hue 26.0a Saturation 14.6a
44.6b 12.3b 5.9a 25.4a 13.6a
38.3c 10.5c 3.3b 17.3b 11.0b
7.82*** 2.07*** 1.82*** 19.9*** 2.89***
b
c
5.86***
*P < 0.05, **P < 0.01, ***P < 0.001; means with different superscripts are significantly different (P < 0.05).
¼ 0:1747 L value of AD 4:6571 ðR2 ¼ 0:60Þ From this it can be calculated that the L* values of the AD corresponding to the five subjective scores for the LD are: score 1 (extremely DFD) L* = 32.4, score 2 (slightly DFD) L* = 38.1, score 3 (normal) L* = 43.8, score 4 (slightly PSE) L* = 49.6 and score 5 (extremely PSE) L* = 55.3. Using this equation the AD L* values from a population of slaughtered pigs could similarly be transposed into nominal subjective scores for the LD, allowing the frequency of the five quality groups in the population of carcasses to be defined.
568
P.D. Warriss et al. / Meat Science 73 (2006) 565–569
4.0
Table 4 The range of L* values measured in the AD at two times post mortem in carcasses classed as PSE, normal or DFD based on the characteristics of the LD
3.0
L*
Subjective score in loin
5.0
45 min L* 20 h
2.0
1.0
0.0 30.0
40.0
50.0
60.0
AD lightness (L*) Fig. 1. The relationship between Lightness (L*) measured in the AD at 20 h post mortem and the mean subjective assessment score made in the LD of the loin (1 = extremely DFD, 2 = slightly DFD, 3 = normal, 4 = slightly PSE, 5 = extremely PSE).
4. Discussion The average values for the colour measurements made in the AD at both 45 min and 20 h post mortem varied progressively in the meat quality groups (PSE, normal and DFD) as defined by subjective assessment of the LD, and confirmed by objective measures of pH45, reflectance (EEL value) and waterholding (drip loss). This implied that the quality of the AD reflected that of the LD. This was confirmed by the significant correlations found between subjective and objective measures of LD quality and colour measurements made in the AD. In all cases these were higher at 20 h than at 45 min post mortem when the whole population of 100 carcasses was considered. This is in accord with previous findings (see, Kauffman et al., 1993) and probably for reasons discussed by Swatland (1997). The highest correlations were with CIELAB L* values, indicating that, overall, it is the lightness component of colour of the AD, rather than hue or saturation, that is potentially the best indicator of poor quality in the loin. Brewer et al. (2001) also found that the L* value in all the muscles they studied (not unfortunately including the AD) was the instrumental value most highly correlated with subjectively assessed colour when this was based on the Japanese Colour Standards. Swatland (1982) likewise found that paleness in the AD measured 7 days post mortem explained 56% of the paleness in the LD. L* value has the added advantage that it is not apparently affected by whether the muscle surface has had a chance to bloom (Brewer et al., 2001), so no delay between cutting a fresh surface and measurement is necessary. Unfortunately, despite the correlations found in the present study between L* value and loin quality, because of the overlap of the distribution of values between the quality groups, it was not possible to use L* values made at either time post mortem to identify categorically individual carcasses as PSE, normal or
PSE (n = 16)
Normal (n = 61)
DFD (n = 23)
37–48 42–51
34–45 38–50
35–41 35–45
DFD (Table 4). However, because the LD and AD have different fibre compositions (Ruusunen & Puolanne, 1988), the AD being a redder muscle with a more oxidative metabolism, the AD would not be expected to be a perfect predictor of LD quality. In conclusion, the use of colour measurements, particularly L* values, made on the cut surface of the AD exposed on splitting the pig carcass appears to have potential for predicting the meat quality in the more valuable, but less accessible loin. Better prediction would generally result from measurements made at 20 h post mortem after chilling, but carcasses showing extreme PSE would probably be identifiable using measurements made at 45 min post mortem, just after carcasses had entered the chill room. Measurements of L* values on the cut surface of the AD could be used to predict the relative prevalence of carcasses showing PSE and DFD in populations of slaughtered pigs. The method would however probably not be sufficiently precise for selection of individual carcasses on the slaughter line so that those that showing poor meat quality could be processed differently. Portable tristimulus colour meters are relatively widely available, and the CIELAB system for measuring colour is well established and it is easy to standardise measurements reliably using standard white and coloured tiles. The method could therefore be useful for measuring the quality of populations of pork carcasses under commercial conditions, for example to monitor dayto-day variation, or variation between plants, or potential changes in quality caused by changes in production procedures. It might have advantages over current techniques using pH electrodes or electronic quality probes such as carcass grading probes and the Fibre Optic Probe (Fortin & Raymond, 1988; Warriss et al., 1989) or the Pork Quality Meter (Warriss, Brown, & Adams, 1991). References Brewer, M. S., Zhu, L. G., Bidner, B., Meisinger, D. J., & McKeith, F. K. (2001). Measuring pork colour: effects of bloom time, muscle, pH and relationship to instrumental parameters. Meat Science, 57, 169–176. Brown, S. N. (1992). A note on the use of subjective methods for assessing pig meat quality on the slaughterline. Meat Science, 32, 195–202. Cannon, J. E., Morgan, J. B., McKeith, F. K., Smith, G. C., Sonka, S., Heavner, J., et al. (1996). Pork Chain Quality Audit: quantification of pork quality characteristics. Journal of Muscle Foods, 7, 29–44. Chizzolini, R., Novelli, E., Badiani, A., Rosa, P., & Delbono, G. (1993). Objective measures of pork quality: evaluation of various techniques. Meat Science, 34, 49–77. Forrest, J. C., Morgan, M. T., Borgaaerd, C., Rasmussen, A. J., Jespersen, B. L., & Andersen, J. R. (2000). Development of technology for the early postmortem prediction of waterholding capacity and drip loss in fresh pork. Meat Science, 55, 115–122.
P.D. Warriss et al. / Meat Science 73 (2006) 565–569 Fortin, A., & Raymond, D. P. (1988). The use of the electrical characteristics of muscle for the objective determination of PSE and DFD in pork carcasses under commercial conditions. Canadian Institute of Food Science and Technology Journal, 21, 260–265. Kauffman, R. G., Sybesma, W., Smulders, F. J. M., Eikelenboom, E., Engel, B., van Laack, R. J. L. M., et al. (1993). The effectiveness of examining early post mortem musculature to predict ultimate pork quality. Meat Science, 34, 283–300. Kauffman, R. G., Wachholz, D., Henderson, D., & Lochner, J. V. (1978). Shrinkage of PSE, normal and DFD hams during transit and processing. Journal of Animal Science, 46, 1236–1240. Lee, S., Norman, J. M., Gunasekaran, S., van Laack, R. L. J. M, Kim, B. C., & Kauffman, R. G. (2000). Use of electrical conductivity to predict water-holding capacity in post-rigor pork. Meat Science, 55, 385–389. Lopez-Bote, C., Warriss, P. D., & Brown, S. N. (1989). The use of muscle protein solubility measurements to assess pig lean meat quality. Meat Science, 26, 167–175. MacDougall, D. B., Cuthbertson, A., & Smith, R. J. (1969). The assessment of pig meat paleness by reflectance photometry. Animal Production, 11, 243–246. Murray, A. C. (1995). The evaluation of meat quality. In S. D. Morgan Jones (Ed.), Quality and grading of carcasses of meat animals (pp. 83–107). Boca Raton, USA: CRC Press. Oliver, M. A., Gispert, M., Tibau, J., & Diestre, A. (1991). The measurement of light scattering and electrical conductivity for the prediction of PSE pig meat at various times postmortem. Meat Science, 29, 141–151. Pedersen, D. K., & Engelsen, S. B. (2001). Early prediction of the quality of porcine meat by FT-IR spectroscopy. In Proceedings of the 47th international congress of meat science and technology (pp. 204–205). Krakow, Poland, Paper 3-15.. Ruusunen, M., & Puolanne, E. (1988). Fiber-type distribution in porcine muscles. In Proceedings of the 34th international congress of meat science and technology (pp. 52–54). Qld, Australia.. Swatland, H. J. (1982). Intermuscular variation in physical properties of pork. Canadian Institute of Food Science and Technology Journal, 15, 92–95. Swatland, H. J. (1985). Optical and electronic methods of measuring pH and other predictors of meat quality in pork carcasses. Journal of Animal Science, 61, 887–891.
569
Swatland, H. J. (1995). On-line evaluation of meat. Lancaster and Basel: Technomic Publishing Company, Inc., 347 pp. Swatland, H. J. (1997). Observations on rheological, electrical, and optical changes during rigor development in pork and beef. Journal of Animal Science, 75, 975–985. Topel, D. G., Miller, J. A., Berger, P. J., Rust, R. E., Parrish, F. C., & Ono, K. (1976). Palatability and visual appearance of dark, normal and pale colored porcine m. longissimus. Journal of Food Science, 41, 628–630. Wachholz, D., Kauffman, R. G., Henderson, D., & Lockner, J. V. (1978). Consumer discrimination of pork color at the market place. Journal of Food Science, 43, 1150–1152. Warner, R. D., Kauffman, R. G., & Russell, R. L. (1993). Quality attributes of major porcine muscles: a comparison with the Longissimus lumborum. Meat Science, 33, 359–393. Warriss, P. D. (1982). The relationship between pH45 and drip in pig muscle. Journal of Food Technology, 17, 573–578. Warriss, P. D. (1995). Instrumental measurement of colour. In A. A. Taylor, A. Raimundo, M. Severini, & F. J. H. Smulders (Eds.), Meat quality and meat packaging (pp. 221–232). Nijmegen, Utrecht, The Netherlands: European Consortium for Continuing Education in Advanced Meat Science and Technology. Audet Tijdschriften b.v.. Warriss, P. D. (2000). Meat science: An introductory text. Wallingford, Oxon, OX10 8DE: CAB International, 312pp. Warriss, P. D., & Brown, S. N. (1993). Relationships between the subjective assessment of pork quality and objective measures of colour. In J. D. Wood & T. L. J. Lawrence (Eds.), Safety and quality of food from animals (pp. 98–101). Edinburgh: Occasional Publication of the British Society of Animal Production No. 17, British Society of Animal Production. Warriss, P. D., Brown, S. N., & Adams, S. J. M. (1991). Use of the Tecpro pork quality meter for assessing meat quality on the slaughterline. Meat Science, 30, 147–156. Warriss, P. D., Brown, S. N., Lopez-Bote, C., Bevis, E. A., & Adams, S. J. M. (1989). Evaluation of lean meat quality in pigs using two electronic probes. Meat Science, 25, 281–291. Whitman, T. A., Forrest, J. C., Morgan, M. T., & Okos, M. R. (1996). Electrical measurements for detecting early postmortem changes in porcine muscle. Journal of Animal Science, 74, 80–90.