The Characterization and Incidence of Pale, Soft, Exudative Turkey Meat in a Commercial Plant 1

The Characterization and Incidence of Pale, Soft, Exudative Turkey Meat in a Commercial Plant 1 C. M. Owens,* E. M. Hirschler,* S. R. McKee,* R. Martinez-Dawson, and A. R. Sams*,2 *Department of Poultry Science, Texas A&M University, College Station, Texas 77843-2472; and Department of Experimental Statistics, Clemson University, Clemson, South Carolina 29634-0367 ABSTRACT Pale, soft, exudative (PSE) turkey meat is processing line to determine the commercial incidence of a growing problem for industry and has been associated PSE meat based on color. The pale fillets had significantly with rapid postmortem ph decline and loss of protein lower ph, greater L* value, and less WHC but equal functionality, similar to PSE pork. This study was designed to estimate the incidence of PSE meat in a commer- color. The L* value and ph were correlated with WHC temperature when compared with the fillets with normal cial plant and use response surface methodology to characterize the relationship between ph, lightness (L* value), as measured by expressible moisture, but L* value seems to have more predictive value. By using the L* value and water-holding capacity (WHC). One hundred thirtyfour turkey breast fillets were selected from the pro- range (>53) from the pale-selected fillets as an indication cessing line so that 67 had normal color (lightness), and of paleness, approximately 40% of the 2,995 fillets would the other 67 were more pale than normal. Fillets were exhibit poor WHC. These results suggest that PSE meat analyzed at time of deboning (1.5 h postmortem) and can represent a significant portion of commercially processed turkey breast meat and that the L* value measure- at 24 h postmortem for color (L* value), ph, drip loss, expressible moisture, and temperature. Additionally, L* values were measured on 2,995 turkey breasts from the ment could be used to sort turkey meat so that PSE type meat could be used in specialized formulations. (Key words: pale, soft, exudative; turkey; meat; postmortem ph; color) 2000 Poultry Science 79:553 558 INTRODUCTION Recently, the turkey industry has been faced with the increasing problem of pale, soft, and exudative meat (PSE) similar to the PSE condition observed in pork (Penny, 1969; Cassens et al., 1975; Mitchell and Heffron, 1982; Warriss and Brown, 1987). This PSE turkey meat is pale in color, has lower water-holding capacity (WHC), and forms softer gels (Sosnicki and Wilson, 1991; Barbut, 1993; Ferket and Foegeding, 1994; Barbut, 1997a; McKee and Sams, 1997, 1998). The increasing demands of turkey products by consumers have shifted turkey sales from whole birds to further processed products that now compose a large proportion of the turkey market. As the turkey industry increases the production of further processed cooked products, such as formed breast loaves and rolls, the PSE problem becomes more apparent. The PSE meat can cause problems during cooking by increasing purge in the cook-in bags, which Received for publication May 28, 1999. Accepted for publication December 8, 1999. 1 This research was supported by a grant from the US Poultry and Egg Association. 2 To whom correspondence should be addressed: asams@poultry. tamu.edu. results in a product with increased yield losses during cooking, poor meat binding, and dry, soft texture. These defective products must be repackaged, which increases costs because of packaging material and labor. The PSE in swine has been associated with antemortem and postmortem stressors including environmental temperatures (Cassens et al., 1975), preslaughter handling practices (Backstrom and Kauffman, 1995; D Souza et al., 1998), stunning methods (Backstrom and Kauffman, 1995), and chilling regimes (Honikel, 1987; Offer, 1991). The PSE is the result of accelerated postmortem glycolysis, which results in a rapid postmortem ph decline while carcass temperatures are still high. This ph and temperature combination results in protein denaturation in the muscle that leads to paler meat color, decreased WHC, and poorer texture (Warriss and Brown, 1987; Santos et al., 1994). Turkeys may be susceptible to the same types of antemortem and postmortem stressors that can cause accelerated rigor development. McKee and Sams (1997) observed that preslaughter heat stress accelerated postmortem ph decline and yielded paler meat with an elevated cook loss. McKee and Sams (1998) reported that Abbreviation Key: EM = expressible moisture; PSE = pale soft exudative; WHC = water-holding capacity. 553

554 OWENS ET AL. inadequate chilling temperatures (>20 C) resulted in lighter meat with higher drip loss and cook loss. Many turkey processing plants are now hot-boning meat to send the meat directly to further processing. Meat can typically be greater than 20 C at the time of deboning (approximately 1.5 to 2 h postmortem). A rapid decline in ph combined with this higher temperature for a long period of time could result in PSE meat. If PSE meat could be sorted before further processing, the PSE meat could be used with special formulations that contain ingredients or conditions to restore meat quality and protein functionality to reduce yield losses and improve texture. However, in developing strategies to use this defective meat, it will be important to understand the relationship between meat quality characteristics including cook loss, drip loss, color, and expressible moisture and biochemical characteristics of the meat including ph and temperature. Therefore, the objective of this study was to use response surface methodology to describe the relationships between these parameters. In addition, color was used to estimate the incidence of PSE meat in a commercial turkey processing plant. Experiment 1 MATERIALS AND METHODS In two trials, 134 turkey breast fillets were collected from the deboning line of a commercial turkey processing plant at 1.5 h postmortem. These fillets were placed into one of two subjective color categories: paler than normal color (n = 67) or normal color (n = 67). The L* values (objective lightness of color) of all fillets in both color groups were measured at 1.5 and 24 h postmortem using a Minolta colorimeter. 3 For each fillet, the colorimeter was programmed to collect an average of three separate color readings that were scattered over the surface of the fillet. Muscle ph was directly measured at 1.5 h postmortem using a Metoxy ph spear probe 4 and meter. 5 Fillets were weighed at collection, placed in individually numbered and perforated (to allow for drip loss) plastic bags, and kept overnight in a cooler at 2 C. The fillets were reweighed at 24 h postmortem for drip loss determination, and expressible moisture was determined using the filter paper press method described by Urbin et al. (1962). Experiment 2 The L* value was measured on a total of 2,995 fillets on the deboning line (1.5 h postmortem) over 3 sampling d to determine the incidence of pale fillets in this same commercial plant. The percentage of pale fillets was determined based on L* value range for the pale fillets collected in the first part of the experiment. The minimum L* value 3 Model CR-200, Minolta Corp., Ramsey, NJ 07446. 4 Model #MXT-6700F, Cypress Systems, Lawrence, KS 66047. 5 Model #HM-17MX, Cypress Systems, Lawrence, KS 66047. TABLE 1. Means of meat quality attributes from normal and pale turkey breast fillets 1 Measurement Normal 2 Pale 2 SEM L* value (1.5 h) 47.31 b 56.85 a 0.46 L* value (24 h) 48.99 b 54.72 a 0.37 ph (1.5 h) 6.09 a 5.72 b 0.01 Expressible moisture (%) 23.41 b 32.31 a 0.70 Drip loss (%) 0.72 b 2.52 a 0.18 Postchill muscle temperature (C) 28.6 a 28.2 a 0.23 a,b Means with no common superscript differ significantly (P < 0.05). 1 Groups selected based on subjective color evaluation. 2 n = 67 per mean, excluding cook loss. for the pale group was 53. This value was used as the threshold between pale and normal meat. Data were subjected to regression and correlation analyses with procedures of SAS (1985). First- and secondorder effects were tested, and response surface models were selected based on Mallow s Cp statistic (Martinez et al., 1995; Dawson and Martinez-Dawson, 1998). The residual mean square error was used as the error term, and trials were pooled because no significant interaction was observed. Means of the color groups were separated with the analysis of variance F-test (SAS, 1985). Significance was determined using P < 0.05. Response surface graphs were generated to graphically describe the relationship between multiple variables. RESULTS AND DISCUSSION Characterization Experiment 1 Means of meat quality attributes from normal and pale turkey breast fillets are in Table 1. In all parameters, except temperature and cook loss, the two groups, pale and normal, were significantly different. As expected, the pale turkey fillets had higher L* values, lower muscle ph, and higher expressible moisture and drip loss compared with the normal colored fillets. All carcasses were chilled similarly; therefore, a lack of temperature difference could be expected. Pale turkey meat has been previously associated with lower muscle ph and lower WHC (Barbut, 1993, 1996; McCurdy et al., 1996; McKee and Sams, 1997, 1998; Rathgeber et al., 1999). Color is an important meat quality attribute, and it is often used as an indicator of PSE meat (Barbut, 1993; Chizzolini et al., 1993; Kauffman et al., 1993). L* values measure lightness and range from 0 (black) to 100 (white). Pale meat is the result of denaturation of sarcoplasmic proteins, which leads to greater scattering of light (Bendall, 1973; Swatland, 1993). Protein denaturation can occur as a result of a rapid ph decline while carcass temperatures are still high (Bendall and Wismer-Pedersen, 1962; Penny 1969; Mitchell and Heffron, 1982; Offer, 1991). The L* value was not affected by muscle temperatures in the observed range (22 to 36 C). However, L* value was significantly affected by muscle ph (Figure 1). The L* value increased as muscle ph decreased. The relationship between L* value and muscle ph is curvilinear with a

CHARACTERIZATION OF PALE SOFT EXUDATIVE TURKEY MEAT IN COMMERCIAL PLANTS 555 TABLE 2. Correlation coefficients (P values) between expressible moisture, drip loss, ph, and L* value (1.5 and 24 h) of turkey breast fillets Parameter ph L* value (1.5 h) L* value (24 h) Expressible moisture 0.52 (0.0001) 0.57 (0.0001) 0.60 (0.0001) Drip loss 0.38 (0.0001) 0.50 (0.0001) 0.54 (0.0001) ph... 0.70 (0.0001) 0.64 (0.0001) FIGURE 1. The relationship between ph and L* value at 1.5 h postmortem of turkey breast fillets (L* value = 659.1679 187.1478 * ph + 14.2517 * ph * ph; R 2 = 0.5642; P = 0.0001). Each point represents average L* value for every 0.05 increment of ph. The regression line represents predicted values based on raw data. steeper slope at lower muscle ph. This form of relationship suggests that lightening of the meat is only related to abnormally low ph values at 1.5 h. If rigor development is accelerated, resulting in lower muscle ph, then it is likely that sarcoplasmic and myofibrillar proteins will begin to denature, resulting in pale meat. Muscle color continues to change over time, whereas rigor develops in muscle; therefore, 24-h L* value was related to 1.5-h muscle ph (Figure 2). L* value (24 h) increased as muscle ph decreased. This relationship between 24-h L* value and 1.5-h ph was similar to the relationship between 1.5-h L* value and 1.5-h ph; however, the 24-h L* value relationship is more linear, rather FIGURE 2. The relationship between ph and L* value of turkey breast fillets at 24 h postmortem (L* value = 109.7166 9.7957 * ph; R 2 = 0.4124; P = 0.0001). Each point represents average L* value for every 0.05 increment of ph. The regression line represents predicted values based on raw data. than the more curvilinear 1.5-h L* value relationship. This linear relationship is probably due to relating 24-h L* value to 1.5-h muscle ph. Temperature at this range (22 to 36 C) had no effect on 24-h color. The L* values were significantly correlated with ph at 1.5 and 24 h postmortem ( 0.70 and 0.64, respectively) (Table 2). Similar relationships have been previously reported in poultry and swine (Warriss and Brown, 1987; Barbut, 1993, 1996, 1997b; McCurdy et al., 1996). The L* value was significantly but less correlated with postchill temperature and only at 1.5 h postmortem (data not shown). The lower correlation between L* value and temperature may have resulted from the fact that all fillets were observed to have a warmer temperature range (22 to 36 C) than would normally be expected in a rapidly and thoroughly chilled turkey breast fillet. The plant at which this study was conducted did not rapidly chill their carcasses before deboning. This procedure resulted in warm, boneless fillets that were chilled off the carcasses. Expressible moisture (EM) is a useful measurement of WHC that involves using force to express water from the meat. Water that is bound by proteins is not expressed from the meat. Expressible moisture is the percentage of total water in the meat that can be expressed by applied force. Generally, an increase in EM indicates a greater proportion of the water that is held more loosely and, thus, indicates a lower WHC (Honikel and Hamm, 1994). The relationship between L* value at 1.5 h, ph, and EM (Figure 3) indicates that EM increases with increasing L* values. However, the EM increase is more linear at higher ph values but reaches a plateau with lower ph. By plotting the EM of each incremental increase in L* value (Figure 4), we determined that EM increased with L* value and plateaued at about L* = 54. This plateau may suggest that at an L* value 54, there is minimal protein functionality or that the influence of some other factor begins to dominate EM. For example, at high L* values, a portion of expressible water may have already been lost as drip prior to EM analysis at 24 h postmortem. The relationship between L* value at 24 h, ph, and EM (Figure 5) is similar to that shown in Figure 3 except that the effect of L* value is more linear at all ph levels. The difference in the relationships (between Figures 3 and 5) may result from the dynamic nature of the muscle metabolism at 1.5 h postmortem. Rigor development is still ongoing at this time and will not be complete until 4 to 8 h postmortem (Stewart et al., 1984; Owens and Sams, 1997). Regardless of these subtle differences in the shapes of these graphs, changes in L* value clearly coin-

556 OWENS ET AL. FIGURE 3. The relationship between expressible moisture (EM), ph, and L* value (1.5 h) of turkey breast fillets (EM = 14.8935 0.0236 * color * color 2.5438 * ph * ph + 0.5435 * color * ph; R 2 = 0.3714; P = 0.0001). Each point represents average ph and EM for every one increment of L* value. The surface plot represents predicted values based on raw data. cide more with changes in EM than do changes in ph for turkey breast meat. Drip loss is another indicator of WHC. Drip loss was significantly affected by L* value at both 1.5 and 24 h postmortem (Figures 6 and 7); however, ph had no effect. As L* value (1.5 and 24 h) increased, drip loss increased. The lack of an effect of ph on drip loss was unexpected. However, ph decline in poultry is generally more rapid than in pork (Addis, 1986) and, therefore, the range of FIGURE 5. The relationship between expressible moisture (EM), ph and L* value (24 h) of turkey breast fillets (EM = 24.7291 + 0.0083 * color * color 0.5549 * ph * ph; R 2 = 0.3861; P = 0.0001). Each point represents average ph and EM for every one increment of L* value. The surface plot represents predicted values based on raw data. ph values at an early postmortem time (1.5 h) could be smaller in poultry than in pork. This rapid decline in ph may diminish the correlation between drip loss and ph measured at 1.5 h postmortem. The correlations among EM, drip loss, and ph and L* value (1.5 and 24 h) are shown in Table 2. Expressible moisture was significantly correlated with ph and with L* value at 1.5 and 24 h, which is to be expected. A rapid decline in ph early postmortem, while carcass temperatures are still high, can result in protein denaturation in FIGURE 4. The relationship between expressible moisture (EM) and L* value (1.5 h) of turkey breast fillets (EM = 163.2122 + 6.5531 * color 0.0544 * color * color; R 2 = 0.3512; P = 0.0001). Each point represents average EM for every one increment of L* value. The regression line represents predicted values based on raw data. FIGURE 6. The relationship between drip loss and L* value (1.5 h) of turkey breast fillets (drip = 8.3729 + 0.1920 * color; R 2 = 0.2488; P = 0.0001). Each point represents average drip loss for every one increment of L* value. The regression line represents predicted values based on raw data.

CHARACTERIZATION OF PALE SOFT EXUDATIVE TURKEY MEAT IN COMMERCIAL PLANTS 557 FIGURE 7. The relationship between drip loss and L* value (24 h) of turkey breast fillets (drip = 11.8026 + 0.2589 * color; R 2 = 0.2881; P = 0.0001). Each point represents average drip loss for every one increment of L* value. The regression line represents predicted values based on raw data. the muscle. This denaturation can affect color as well as WHC (Bendall and Wismer-Pedersen, 1962; Penny, 1969; Mitchell and Heffron, 1982; Offer, 1991; Warriss and Brown, 1987; Santos et al., 1994). Incidence Experiment 2 In addition to evaluating the relationship between various meat quality attributes, the commercial incidence of pale meat was assessed. Because L* value was significantly correlated to EM and drip loss in Experiment 1 and has been previously used to identify meat with PSE characteristics, L* value was measured on 2,995 fillets on 3 sampling d over a 3-mo period to estimate the incidence. The percentage of fillets with 1.5 h L* value >53 was calculated to determine incidence. This threshold was chosen based on an L* value of 53. This L* value was on the low end of the range for the pale group of fillets selected in the first part of this study. The frequency of L* values is shown in Figure 8. Forty-one percent of the turkey breast fillets surveyed had L* values >53, suggesting that a substantial percentage of the turkey meat in this plant could potentially have lower WHC. However, more study is needed to determine the L* value threshold above which protein damage and associated loss in WHC become commercially unacceptable. This determination will allow a better assessment of which of these 41% of fillets are truly PSE. In conclusion, these results suggest that a substantial portion of commercially produced turkey breast meat may have compromised WHC. Although L* value and ph may be correlated with WHC, as measured by expressible moisture, L* value seems to have more predictive value. This measurement could be used to sort turkey meat so that PSE-type meat could be used in specialized formulations. FIGURE 8. Frequency of L* values (1.5 h) of turkey breast fillets in a commercial processing plant. The dashed line represents the threshold between pale and normal meat (L* value = 53). ACKNOWLEDGMENTS The authors are grateful for the support of Plantation Foods and US Poultry and Egg Association. REFERENCES Addis, P. B., 1986. Poultry muscle as a food. Pages 371 404 in: Muscle as Food. P. J. Bechtel, ed. Academic Press, New York, NY. Backstrom, L., and R. Kauffman, 1995. The porcine stress syndrome: A review of genetics, environmental factors, and animal well-being implications. Agri-Practice 16(8):24 30. Barbut, S., 1993. Colour measurements for evaluating the pale soft exudative (PSE) occurrence in turkey meat. Food Res. Int. 26:39 43. Barbut, S., 1996. Estimates and detection of the PSE problem in young turkey breast meat. Can. J. Anim. Sci. 76:455 457. Barbut, S., 1997a. Occurrence of pale soft exudative meat in mature turkey hens. Br. Poult. Sci. 38:74 77. Barbut, S., 1997b. Problem of pale soft exudative meat in broiler chickens. Br. Poult. Sci. 38:355 358. Bendall, J. R., 1973. Postmortem changes in muscle. Pages 244 309 in: The Structure and Function of Muscle. Vol. 2, part 2. 2nd ed. G. H. Bourne, ed. Academic Press, New York, NY. Bendall, J. R., and J. Wismer-Pedersen, 1962. Some properties of the fibrillar proteins of normal and watery pork muscle. J. Food Sci. 27:144 159. Cassens, R. G., D. N. Marple, and G. Eikelenboom, 1975. Animal physiology and meat quality. Adv. Food Res. 21:71 155. Chizzolini, R., E. Novelli, A. Badiani, P. Rosa, and G. Delbano, 1993. Objective measurements of pork quality: Evaluation of various techniques. Meat Sci. 34:49 77. Dawson, P. L., and R. Martinez-Dawson, 1998. Using response surface analysis to optimize the quality of ultrapasteurized liquid whole egg. Poultry Sci. 77:468 474. D Souza, D. N., F. R. Dunshea, R. D. Warner and B. J. Leury, 1998. The effect of pre-slaughter handling and carcass processing rate post-slaughter on pork quality. Meat Sci. 50:429 437. Ferket, P. R., and E. A. Foegeding, 1994. How nutrition and management influence PSE in poultry meat. Pages 64 78 in: Proceedings from BASF Technical Symposium, Multi-State Poultry Feeding and Nutrition Conference. Indianapolis, IN.

558 OWENS ET AL. Honikel, K. O., 1987. Influence of chilling on meat quality attributes of fast glycolysing pork muscles. Pages 273 283 in: Evaluation and Control of Meat Quality in Pigs. P. V. Tarrant, G. Eikelenboom, and G. Monin, ed. Martinus Nijhoff, Dordrecht, The Netherlands. Honikel, K. O., and R. Hamm, 1994. Measurement of waterholding capacity and juiciness. Pages 125 161 in: Quality Attributes and Their Measurement in Meat, Poultry and Fish Products. A. M. Pearson and T. R. Dutson, ed. Blackie Academic & Professional, Bishopbriggs, Glasgow, UK. Kauffman, R. G., W. Sybesma, F.J.M. Smulders, G. Eikelenboom, B. Engel, R.L.J.M. van Laack, A. H. Hoving-Bolink, P. Sterrenburg, E. V. Nordheim, P. Walstra, and P. G. van der Wal, 1993. The effectiveness of examining early post-mortem musculature to predict ultimate pork quality. Meat Sci. 34:283 300. Martinez, R. M., P. L. Dawson, H. R. Ball, Jr., K. R. Swartzel, S. E. Winn, and F. G. Giesbrecht, 1995. The effects of ultrapasteurization with and without homogenization on the chemical, physical, and functional properties of aseptically packaged liquid whole egg. Poultry Sci. 74:742 752. McCurdy, R. D., S. Barbut, and M. Quinton, 1996. Seasonal effect on pale soft exudative (PSE) occurrence in young turkey breast meat. Food Res. Int. 29:363 366. McKee, S. R., and A. R. Sams, 1997. The effect of seasonal heat stress on rigor development and the incidence of pale, exudative turkey meat. Poultry Sci. 76:1616 1620. McKee, S. R., and A. R. Sams, 1998. Rigor mortis development at elevated temperatures induces pale exudative turkey meat characteristics. Poultry Sci. 77:169 174. Mitchell, G., and J.J.A. Heffron, 1982. Porcine stress syndrome. Adv. Food Res. 28:167 230. Offer, G., 1991. Modelling of the formation of pale, soft, and exudative meat: Effects of chilling regime and rate and extent of glycolysis. Meat Sci. 30:157 184. Owens, C. M., and A. R. Sams, 1997. Muscle metabolism and meat quality of pectoralis form turkeys treated with postmortem electrical stimulation. Poultry Sci. 76:1047 1051. Penny, I. F., 1969. Protein denaturation and water-holding capacity in pork muscle. J. Food Technol. 4:269 273. Rathgeber, B. M., J. A. Boles, and P. J. Shand, 1999. Rapid postmortem ph decline and delayed chilling reduce quality of turkey breast meat. Poultry Sci. 78:477 484. Santos C., L. C. Roserio, H. Goncalves, and R. S. Melo, 1994. Incidence of different pork quality categories in a Portuguese slaughterhouse: A survey. Meat Sci. 38:279 287. SAS Institute, 1985. SAS/STAT Guide for Personal Computers. 6th ed. SAS Institute Inc., Cary, NC. Sosnicki, A. A., and B. W. Wilson, 1991. Pathology of turkey skeletal muscle: Implications for the poultry industry. Food Struct. 10:317 326. Stewart, M. K., D. L. Fletcher, D. Hamm, and J. E. Thomson, 1984. The influence of hot boning broiler breast meat muscle on ph decline and toughening. Poultry Sci. 63:1935 1939. Swatland, H. J., 1993. Paleness, softness, and exudation in pork Review. Pages 273 286 in: Pork Quality: Genetic and Metabolic Factors. E. Poulanne and D. I. Demeyer, ed., C.A.B. International, Wallingford, UK. Urbin, M. C., D. A. Zessin, and G. D. Wilson, 1962. Observations on a method of determining the water-binding properties of meat. J. Anim Sci. 21:9. Warriss, P. D., and S. N. Brown, 1987. The relationships between initial ph, reflectance and exudation on pig muscle. Meat Sci. 20:65 74.