- Email: [email protected]
Ultrasonic scanning measurements of the longissimus thoracis et lumborum muscle to predict carcass muscle content in sheep
PII:
s0309-1740(97)00074-0
Meat Sciem=e,Vol. 48, No. l/2, 4148, 1998 0 1997 Elsevier Science Ltd All rights reserved. Printed in Great Britain 0309-1740/98 $19.00+0.00
ELSEVIER
Ultrasonic Scanning Measurements of the longissimus thoracis et lumborum Muscle to Predict Carcass Muscle Content in Sheep 0. Mahgoub Department
of Animal and Veterinary Sciences, College of Agriculture, Sultan Qaboos University, PO Box 34, Al-khod 123, Sultanate of Oman
(Received 10 April 1997; revised version received 16 June 1997)
ABSTRACT Nineteen Omani sheep (average 26 kg body weight) were scanned over the 6th rib, 12th rib and the second-last lumbar vertebra, to determine the maximum depth (B) , circumference, area of the cross-section and volume of the longissimus thoracis et lumborum (LD) muscle. The volume of the muscle was estimated by multiplying the average area of its cross-section obtained at the three sites by its length. After slaughter, the LD muscle was dissected out, frozen, and its maximum depth, circumference, length, weight and volume were determined. Correlation and regression analyses were carried out to evaluate relationships between scanning and real measurements and between scanning measurements and carcass total muscle content. Generally there were positive correlations (r =0.43+66) between the LD scanning measurements made at the 12th rib and the corresponding measurements made on the dissected muscle. The volume of the LD muscle estimated from scanning was positively correlated (r = 0.59) with its real volume which was correlated with its weight (r=O.51) and consequently with the total carcass muscle content (r =0.69). Ultrasonic scanning can be used to predict the circumference and volume of the LD muscle. These may be usedfor prediction of total carcass muscle content in sheep in addition to traditional measurements of the muscle (the maximum depth B) and area. 0 1997 Elsevier Science Ltd. All rights reserved
INTRODUCTION Prediction of carcass composition in live animals has an important use in evaluating meat animals. Lean is the most valuable part of the carcass and most sought after by consumers. Therefore, prediction of its content in live animals has significant application to the meat industry. Several techniques have been used to predict carcass composition in live animals. These have included dilution techniques which are based on total body water (Owen, 1976) and ultrasonic scanning. The latter method has been used for measuring 41
42
0. Mahgoub
back-fat thickness in meat animals since the 1950’s (Stouffer and Westerveld, 1976). Introduction of real-time B-mode ultrasonic scanners has allowed measuring the depth and area of muscles. Animals of good conformation or ‘muscularity’ are claimed to produce carcasses with more lean meat and a higher proportion of joints and lean meat in the more expensive cuts (Ward et al., 1992). This trait has traditionally been subjectively assessed in meat animals but recently some workers have proposed objective measurements of muscularity using ultrasonic scanning. An example is using the ratio between muscle depth of a group of muscles surrounding a bone and bone length as a measurement of ‘muscularity’ in sheep (Purchas et al., 1991). Another approach for determining muscularity in live animals could be by estimating the weight of total carcass muscle content using X-ray computer tomography (Young et al., 1996) but it is very expensive. The present study aimed to estimate total carcass muscle content from the weight of the LD muscle by estimating its volume from ultrasonic scans. The muscle lies along the vertebral column and extends from the sacrum to the base of the neck. It has been long used as an indicator for total amount of muscle in the carcass of meat animals (Forrest et al., 1975). Measurements of its maximum width (A), maximum depth (B), area and fat thickness above the muscle (C) have been standardized for this purpose (P&son, 1939). Its structure, anatomical relationships, superficial position and more or less its uniform shape indicated that its volume may be feasibly estimated by ultrasonic scanning.
MATERIALS
AND METHODS
Animals and management Nineteen Omani sheep (6 entire, 9 wethers and 4 ewes) were used. They were fed ad-libitum creep feed (CP 16%) from birth until weaning. From weaning until slaughter, lambs were fed ad-libitum concentrate sheep feed (16% CP) plus chopped Rhodesgrass hay (8% CP). Water and salt blocks were available ad-libitum. Animals were slaughtered when each reached a body weight of 26 kg. Ultrasound measurements A Microimager 2000 (Ausonics) real-time B-mode ultrasound scanner fitted with a 7.5 MHz transducer and connected to a Polaroid printer was used. During scanning the animals were restrained in standing position using a head stall. The scanning sites were clipped and the exact points determined by palpation and marked with a marker pen. Aquasonic 100 (Parker) ultrasonic transmission gel was used as an acoustic couplant. Ultrasonic LD measurements include: maximum width (A), maximum depth (B), circumference and area of the cross sections were taken from frozen images on the left side of the animal (Fig. 1). Measurements A and B were taken by the direct control method, i.e. marking the border points then using the trackball on the control board to measure the distance between them. The circumference and area were measured by the direct control and ellipse best fit methods (Anonymous, 1989). The machine has a facility to warn of abnormally larger fittings. The LD length was estimated by externally measuring the distance between the 6th rib and the ilium tuber coxae. To obtain the average cross sectional area, the LD muscle was scanned at three sites: at the 6th rib, 12th rib and the second-last lumbar vertebrae (located immediate to the ilium tuber coxae). The muscle volume was estimated by multiplying its average cross sectional area by its length.
Ultrasonic measurements in sheep
43
Fig. 1. An image of an ultrasonic scan showing the measurements made on the m. longissimus thoracics et lumborum (LD) including the maximum width (h), maximum depth (X), fat thickness
over LD (+), circumference (C) and area (A) using the direct control and ellipse best-fit methods.
Carcass measurements The left-half carcass was split between the 12th and 13th thoracic vertebrae. The A and B measurements were taken on the muscle surface. The carcass was further broken into seven wholesale cuts according to the methods of Levie (1967). The left-half carcass was dissected for major carcass tissue components. The LD muscle portions on the rack (between 6th and 12th ribs) and the short loin (between the 13th rib and last lumbar vertebrae) were dissected out. They were cleaned from adhering tissue and frozen as close to their natural shape on a flat surface. They were then kept in vacuum-sealed plastic bags and stored at -20°C for further measurements. The LD area at the three sites was determined using a digital-planimeter (Ottplan) to measure the muscle cross section traced on tracing paper. The average of 10 readings was recorded. Length, circumference and weight were taken from both frozen rack and loin LD portions using a vernier calliper, a string and scales, respectively. The muscle volume was determined by water displacement method using a home-made device comprised of a glass beaker fitted with a-U-shape-looped glass &be. Ten readings were taken and the average was recorded. The total volume of the muscle was computed as the total of the rack and loin LD volumes. Biometrical Data were analyzed using correlation and regression options in SAS (1985) personal computers GLM procedure to study relationships between scanning and real measurements with carcass total muscle content. Stepwise regression analysis (SAS, 1985) was used to find the maximum r2 and best parameter estimates for multi-variable models to estimate total carcass muscle from LD scanning measurements (at the 1% significance level).
0. Mahgoub
44
RESULTS
AND DISCUSSION
Omani sheep are a small early maturing breed but they have good potential for meat production under tropical conditions if managed well and slaughtered at light weights (Mahgoub and Lodge, 1994). Twenty-six 26 kg live weight achieved at 4-5 month of age, yielding carcasses of 14.77 kg (a dressing out percentage of about 57%) containing carcass muscle proportion of 54.6% recorded in the current study are similar to reports on the breed in Oman (Mahgoub and Lodge, 1994). The 12th rib LD muscle area recorded in the current study (9.97 cm2 )is smaller than that reported for temperate breeds. This may be attributed to differences in body and carcass weights as well as genotype. For example Ward et al. (1992) reported an LD area of 10.12, 16.26, 11.03, 20.40, 18.67 cm2 in 20 and 29 kg carcass Romney; 14.7 and 32.5 kg
TABLE 1
Means and Standard Deviations (SD) of Liveweights, Carcass Weights, Carcass Muscle Content, Carcass and Ultrasonic
Measurements
of Omani Sheep Slaughtered at an Average 26 kg Body Weight Carcass measurements
Liveweight (kg) Carcass weight (kg) Muscle (% in carcass) LD muscle weight (g) LD measurements at 6th rib A (mm) B (mm) C (mm) Area (cm2) Circumference (mm) LD measurements at 12th rib A (mm) B (mm) C (mm) Area (cm2) Circumference (mm) LD measurements at lumbar A (mm) B (mm) C (mm) Area (cm2) Circumference (mm) Volume (cm3) “Using ellipse best-fit method. bUsing tracing method.
Mean
SD
25.707 14.770 54.60 298.90
1.435 0.832 3 55
37.33 22.06 5.96 108.94
53.58 27.42 6.53 9.97
3 4 0.99 13
5 5 2 1.34
125.24
8
61.68 20.89
8 3
9.70
I.50
141.89
14
242.52
23
Ultrasound measurements Mean
SD
43.11 19.37 6.16 6.26” 5.746 96.28” 105.286
3 2 1.42 1.13 11 12
44.95 19.89 4.95 6.79” 6.59b 101.1 In 11O.83b
3 2 1 0.96 1.16 7 12
43.05 18.74 6.16 6.14” 5.64b 94.00” 102.33b 20264
6 3 2 1.30 1.57 13 13 34
5
45
Ultrasonic measurements in sheep
carcass Texel; and 33.4 kg carcass Oxford sheep, respectively. Koohmaraie et al. (1996) reported an m. longissimus area of 18 cm2 in 25 kg carcass Dorset x Romonov sheep. The dissected muscle in the latter study weighed 745-787 g compared to 300 g in Omani sheep. LD measurements taken at three sites indicated that ultrasonic measurements well matched equivalent carcass measurements (Table 1) which is in line with findings in temperate sheep (Ward et al., 1992; Koohmaraie et al., 1996). That includes the LD area, circumference (determined by tracing and ellipse-fitting methods) and volume proposed in the current study. Table 2 presents the relationships between LD scanning measurements and those made on the dissected carcass. Significant 0, ~0.05) relationships between the LD scanning measurements at the 6th rib and second-last lumbar vertebra with the real measurements of the LD muscle included a positive correlation between the B measurement at lumber region and the muscle weight as well as positive correlations with real B and circumference measurements. There were positive correlations between all the LD scanning measurements at the 12th rib and the weight, B and circumference of the muscle. This indicates that scanning at this site should give satisfactory results for prediction of corresponding carcass measurements and confirms its traditional use by workers in this field (Binnie et al., 1995). Significant correlation between scanning and real measurements at 12th rib rather than at the other two sites may be attributed to the fact that the muscle at this site is scanned at its maximum thickness. This is supported by the greater value of B measurement at 12th rib than at 6th rib or lumbar sites (Table 1). Findings of the current study confirmed reports of positive correlations between scanning LD maximum depth (B) and area with corresponding carcass measurements and their use as indicators for the prediction of carcass muscle content (McEwan et al., 1989; Ward et al., 1992; Young and Deaker, 1994) even without adjustments for body or carcass weights. In the present study, the average area of the three sites had positive correlation with the muscle weight and the scanning volume had a positive correlation with the real TABLE 2
Correlation Coefficients between Real and Scanning LD Measurements as well as their Statistical Significance in Omani Sheep Slaughtered at an Average 26 kg Body Weight LD scanning measurements
LD real measurements Weight
Volume
Length
B
Circumference
LD scan at 6th rib Maximum depth (B Circumference (direct) Area (direct) LD scan at 12th rib Maximum depth (B) Circumference (direct) Area (direct)
0.35 0.29 0.15
0.33 -0.12 0.15
0.13 -0.23 0.22
0.48* 0.60* 0.27
0.43* 0.66** 0.56*
0.03 0.16 -0.04
-0.42 -0.32 -0.23
O&I* 0.66** 0.60*
040 0+56* 0*58*
LD scan at last lumber Maximum depth (B) Circumference (direct) Area (direct)
0.47* 0.43 0.41
0.11 0.34 0.28
-0.17 -0.08 -OX)4
0.27 0.41* 0.24
0.47* 0.49* 0.42
Average area Volume
0.49* 0.51*
0.18 0.59*
-0.03 -o&I
0.46 0.51*
0.42 0.42
*, **, *** Correlation coefficient is significant (p < 0.05; 0.01; 0.001).
-0.17 0.28 -0.07
46
0. Mahgoub
volume, weight and B measurement. This indicates that the estimation of volume and weight of the LD muscle from scanning images is useful and may be confidently used for prediction of total carcass muscle content. The current study proposes two novel scanning measurements; namely the circumference and the volume, as predictors of total carcass muscle content in sheep and possibly in other meat animals in addition to the traditional measurements (maximum depth and area). These two measurements had positive correlations with the corresponding measurement made on the dissected muscle. The circumference at the 12th rib had the highest r value and therefore, was regarded as the best predictor for the muscle weight in this study. The volume also had a significant relationship with the weight of the muscle. The positive correlations between the LD circumference and area measured by tracing (direct method) and those measured by ellipse best fit method indicated that either of the two methods may be confidently adopted. Table 3 shows regression equations and correlation coefficients between scanning measurements and total carcass muscle content. The LD scanning measurements made at the 12th rib also had the highest positive correlations with total carcass muscle content compared to measurements taken at the other two sites. This confirms that scanning at this TABLE 3 Regression Equations Describing the Relationships between Scanning and Real LD Measurements with Total Carcass Muscle Content of Omani Sheep Slaughtered at an Average 26 kg Body Weight Independent variable
Constanl (a)
Regression coefficient (b * sb)
Dependent variable
r
Scanning measurements At 6th rib Maximum depth (B)
3062.44
16.13 (30.03)
Circumference (D) Area (D) Circumference (F) Area (F)
2037.58 2922.26 2426.66 2920.90
12.35* (4.85) 0.73 (0.58) 10.00 (6.67) 0.73 (0.51)
0.56* 0.31 0.36 0.35
At 12th rib Maximum depth (B) Circumference (F) Area (F) Circumference (D) Area (D)
2017.49 1670.17 2094.25 1284.25 2370.36
68.02 (46.41) 1511** (3.95) 1.89*** (0.34) 20.73* (9.72) 1.49 (0.73)
0.34 0.48* 0.47* 0.70** 0.82***
At last lumber Maximum depth (B) Circumference (D) Area (D) Circumference (F) Area (F)
2521.47 2738.34 3081.13 250 1.62 2941.21
4588 5.90 0.46 9.23 0.72
0.40 0.29 0.27 0.32 0.30
Average area Volume External length
2421.18 2421.87 1198.32
Total carcass muscle
(26.66) (5.02) (0.42) (7.14) (0.59)
1.54* (0.57) 0.003***(0.0008) 3.70* (1.51)
*, **, *** Regression is significant (p < 0.05; 0.01; 0.001). D: circumference and cross-sectional area were measured by direct method (tracing). F: circumference and cross-sectional area were measured by ellipse best-fit method.
0.13
0.57* 0.69** 0.47*
41
Ultrasonic measurements in sheep
TABLE 4 Stepwise Regression Analysis (maximum r;? option) for the Best 4-variable Model of Scanning LD Measurements to Predict Total Carcass Muscle Content of Omani Sheep Slaughtered at an Average
26 kg Body Weight Variable
Intercept Maximum Depth (B) at 12th rib Circumference (by tracing) at 12th rib Average cross-sectional area (3 sites) Scanning volume
Parameter estimate
Standard error
Probability of F
R2
-231.000 56.915
506.283 29.031
0.659 0.081
0.904
28.219
4745
0.0002
-8.869
1.922
0.0013
0.024
0.005
0.0021
site should provide satisfactory results for prediction of carcass muscle content. At this site, the cross-sectional area and circumference measured by direct method had the highest r values (0.82 and 0.70, respectively) and therefore, were regarded as the best predictors of carcass muscle content. The volume, length, and average area of the cross-section, obtained by scanning were positively correlated with total carcass muscle content. The volume had a high r value (0.69) which indicates it as a good predictor of carcass muscle content. The current study indicates that these three measurements (circumference, cross sectional area at 12th rib and volume) could have a significant application in the meat industry for prediction of carcass muscle content in live animals, especially breeding stock. The method used in this study for estimation of muscle weight from its volume by ultrasonic scanning may be utilized for other individual muscles or muscle groups to determine muscularity and carcass muscle content. Stepwise regression analysis, using the maximum r* option, of LD scanning measurements (at the 0.1 significance level) is presented in Table 4. The analysis confirmed findings of correlation and regression analyses for best predictors of carcass muscle content. Maximum depth (B), circumference at 12th rib (by tracing method), average crosssectional area and volume accounted for 90% of the variation in the total carcass muscle content. The four measurements represented the best parameter estimates in a 4-variable model for prediction of total carcass muscle content. They may be adopted in a commercial model for evaluation of live sheep.
CONCLUSIONS This study showed significant relationships between longissimus muscle maximum depth and area measurements obtained from ultrasound scanning with corresponding carcass measurements. Two other scanning measurements, circumference and volume, also had significant relationships with corresponding carcass measurements. All these significant relationships indicate that they may be used for prediction of carcass muscle content in live sheep. ACKNOWLEDGEMENTS The author wishes to thank Dr Q. M. Ali of the College of Medicine, Sultan Qaboos University for valuable discussion at preliminary stages of study. MS A. L. Olvey,
48
0. Mahgoub
Mr N. M. Al-Saqry and Mr R. M. Al-Busaidi of the Deptartment. of Animal and Veterinary Sciences provided technical assistance. Published with the approval of the College of Agriculture, SOU as paper number 120497.
REFERENCES Anonymous (1989) Microimager 2000 Operation Manual. Ausonics Pty Ltd, Australia. Binnie, D. B., Farmer, R. J. and Clarke, J. N. (1995) Proceedings of the New Zealand Society of Animal Production 55, 111. Forrest, J. C., Aberle, E. D., Hedrick, H. B., Judge, M. D. and Merkel, R. A. (1975) Principles of Meat Science. Freeman & Co., New York, USA. Koohmaraie, M., Shackelford, S. D. and Wheeler, T. L. (1996) Journal of Animal Science 74, 70. Levie, (1967) The Meat Handbook. Avia Publishing Co., Westport, CT, USA. McEwan, J. C., Clarke, J. N., Knowler, M. A. and Wheeler, M. (1989) Proceedings of the New Zealand Society of Animal Production 43, 113. Mahgoub, 0. and Lodge, G. A. (1994) Animal Production 58,365. Owen, J. B. (1976) Sheep Production. Bailiere Tindall, London, UK. Palsson, H. (1939) Journal of Agricultural Sciences (Cambridge), 29, 544. Purchas, R. W., Davies, A. S. and Abdullah, A. Y. (1991) Meat Science 30, 8 1. SAS Institute Inc., (1985) SAS User’s Guide: Statistics, Version 5, ed. SAS Institute Inc., pp. l-956. NC, Cary. Stouffer, J. R. and Westerveld, R. G. (1976) Journal of Clinical Ultrasound 5, 124. Ward, B. G., Purchas, R. W. and Abdullah, A. Y. (1992) Proceedeings of the New Zealand Society of Animal Production. 52, 33.
Young, M. J. and Deaker, J. M. (1994) Proceedings of the New Zealand Society of Animal Production 54, 215.
Young, M. J., Nsoso, S. J., Logan, C. M. and Beatson, P. R. (1996) Proceedings of the New Zealand Society of Animal Production 56, 205.
s0309-1740(97)00074-0
Meat Sciem=e,Vol. 48, No. l/2, 4148, 1998 0 1997 Elsevier Science Ltd All rights reserved. Printed in Great Britain 0309-1740/98 $19.00+0.00
ELSEVIER
Ultrasonic Scanning Measurements of the longissimus thoracis et lumborum Muscle to Predict Carcass Muscle Content in Sheep 0. Mahgoub Department
of Animal and Veterinary Sciences, College of Agriculture, Sultan Qaboos University, PO Box 34, Al-khod 123, Sultanate of Oman
(Received 10 April 1997; revised version received 16 June 1997)
ABSTRACT Nineteen Omani sheep (average 26 kg body weight) were scanned over the 6th rib, 12th rib and the second-last lumbar vertebra, to determine the maximum depth (B) , circumference, area of the cross-section and volume of the longissimus thoracis et lumborum (LD) muscle. The volume of the muscle was estimated by multiplying the average area of its cross-section obtained at the three sites by its length. After slaughter, the LD muscle was dissected out, frozen, and its maximum depth, circumference, length, weight and volume were determined. Correlation and regression analyses were carried out to evaluate relationships between scanning and real measurements and between scanning measurements and carcass total muscle content. Generally there were positive correlations (r =0.43+66) between the LD scanning measurements made at the 12th rib and the corresponding measurements made on the dissected muscle. The volume of the LD muscle estimated from scanning was positively correlated (r = 0.59) with its real volume which was correlated with its weight (r=O.51) and consequently with the total carcass muscle content (r =0.69). Ultrasonic scanning can be used to predict the circumference and volume of the LD muscle. These may be usedfor prediction of total carcass muscle content in sheep in addition to traditional measurements of the muscle (the maximum depth B) and area. 0 1997 Elsevier Science Ltd. All rights reserved
INTRODUCTION Prediction of carcass composition in live animals has an important use in evaluating meat animals. Lean is the most valuable part of the carcass and most sought after by consumers. Therefore, prediction of its content in live animals has significant application to the meat industry. Several techniques have been used to predict carcass composition in live animals. These have included dilution techniques which are based on total body water (Owen, 1976) and ultrasonic scanning. The latter method has been used for measuring 41
42
0. Mahgoub
back-fat thickness in meat animals since the 1950’s (Stouffer and Westerveld, 1976). Introduction of real-time B-mode ultrasonic scanners has allowed measuring the depth and area of muscles. Animals of good conformation or ‘muscularity’ are claimed to produce carcasses with more lean meat and a higher proportion of joints and lean meat in the more expensive cuts (Ward et al., 1992). This trait has traditionally been subjectively assessed in meat animals but recently some workers have proposed objective measurements of muscularity using ultrasonic scanning. An example is using the ratio between muscle depth of a group of muscles surrounding a bone and bone length as a measurement of ‘muscularity’ in sheep (Purchas et al., 1991). Another approach for determining muscularity in live animals could be by estimating the weight of total carcass muscle content using X-ray computer tomography (Young et al., 1996) but it is very expensive. The present study aimed to estimate total carcass muscle content from the weight of the LD muscle by estimating its volume from ultrasonic scans. The muscle lies along the vertebral column and extends from the sacrum to the base of the neck. It has been long used as an indicator for total amount of muscle in the carcass of meat animals (Forrest et al., 1975). Measurements of its maximum width (A), maximum depth (B), area and fat thickness above the muscle (C) have been standardized for this purpose (P&son, 1939). Its structure, anatomical relationships, superficial position and more or less its uniform shape indicated that its volume may be feasibly estimated by ultrasonic scanning.
MATERIALS
AND METHODS
Animals and management Nineteen Omani sheep (6 entire, 9 wethers and 4 ewes) were used. They were fed ad-libitum creep feed (CP 16%) from birth until weaning. From weaning until slaughter, lambs were fed ad-libitum concentrate sheep feed (16% CP) plus chopped Rhodesgrass hay (8% CP). Water and salt blocks were available ad-libitum. Animals were slaughtered when each reached a body weight of 26 kg. Ultrasound measurements A Microimager 2000 (Ausonics) real-time B-mode ultrasound scanner fitted with a 7.5 MHz transducer and connected to a Polaroid printer was used. During scanning the animals were restrained in standing position using a head stall. The scanning sites were clipped and the exact points determined by palpation and marked with a marker pen. Aquasonic 100 (Parker) ultrasonic transmission gel was used as an acoustic couplant. Ultrasonic LD measurements include: maximum width (A), maximum depth (B), circumference and area of the cross sections were taken from frozen images on the left side of the animal (Fig. 1). Measurements A and B were taken by the direct control method, i.e. marking the border points then using the trackball on the control board to measure the distance between them. The circumference and area were measured by the direct control and ellipse best fit methods (Anonymous, 1989). The machine has a facility to warn of abnormally larger fittings. The LD length was estimated by externally measuring the distance between the 6th rib and the ilium tuber coxae. To obtain the average cross sectional area, the LD muscle was scanned at three sites: at the 6th rib, 12th rib and the second-last lumbar vertebrae (located immediate to the ilium tuber coxae). The muscle volume was estimated by multiplying its average cross sectional area by its length.
Ultrasonic measurements in sheep
43
Fig. 1. An image of an ultrasonic scan showing the measurements made on the m. longissimus thoracics et lumborum (LD) including the maximum width (h), maximum depth (X), fat thickness
over LD (+), circumference (C) and area (A) using the direct control and ellipse best-fit methods.
Carcass measurements The left-half carcass was split between the 12th and 13th thoracic vertebrae. The A and B measurements were taken on the muscle surface. The carcass was further broken into seven wholesale cuts according to the methods of Levie (1967). The left-half carcass was dissected for major carcass tissue components. The LD muscle portions on the rack (between 6th and 12th ribs) and the short loin (between the 13th rib and last lumbar vertebrae) were dissected out. They were cleaned from adhering tissue and frozen as close to their natural shape on a flat surface. They were then kept in vacuum-sealed plastic bags and stored at -20°C for further measurements. The LD area at the three sites was determined using a digital-planimeter (Ottplan) to measure the muscle cross section traced on tracing paper. The average of 10 readings was recorded. Length, circumference and weight were taken from both frozen rack and loin LD portions using a vernier calliper, a string and scales, respectively. The muscle volume was determined by water displacement method using a home-made device comprised of a glass beaker fitted with a-U-shape-looped glass &be. Ten readings were taken and the average was recorded. The total volume of the muscle was computed as the total of the rack and loin LD volumes. Biometrical Data were analyzed using correlation and regression options in SAS (1985) personal computers GLM procedure to study relationships between scanning and real measurements with carcass total muscle content. Stepwise regression analysis (SAS, 1985) was used to find the maximum r2 and best parameter estimates for multi-variable models to estimate total carcass muscle from LD scanning measurements (at the 1% significance level).
0. Mahgoub
44
RESULTS
AND DISCUSSION
Omani sheep are a small early maturing breed but they have good potential for meat production under tropical conditions if managed well and slaughtered at light weights (Mahgoub and Lodge, 1994). Twenty-six 26 kg live weight achieved at 4-5 month of age, yielding carcasses of 14.77 kg (a dressing out percentage of about 57%) containing carcass muscle proportion of 54.6% recorded in the current study are similar to reports on the breed in Oman (Mahgoub and Lodge, 1994). The 12th rib LD muscle area recorded in the current study (9.97 cm2 )is smaller than that reported for temperate breeds. This may be attributed to differences in body and carcass weights as well as genotype. For example Ward et al. (1992) reported an LD area of 10.12, 16.26, 11.03, 20.40, 18.67 cm2 in 20 and 29 kg carcass Romney; 14.7 and 32.5 kg
TABLE 1
Means and Standard Deviations (SD) of Liveweights, Carcass Weights, Carcass Muscle Content, Carcass and Ultrasonic
Measurements
of Omani Sheep Slaughtered at an Average 26 kg Body Weight Carcass measurements
Liveweight (kg) Carcass weight (kg) Muscle (% in carcass) LD muscle weight (g) LD measurements at 6th rib A (mm) B (mm) C (mm) Area (cm2) Circumference (mm) LD measurements at 12th rib A (mm) B (mm) C (mm) Area (cm2) Circumference (mm) LD measurements at lumbar A (mm) B (mm) C (mm) Area (cm2) Circumference (mm) Volume (cm3) “Using ellipse best-fit method. bUsing tracing method.
Mean
SD
25.707 14.770 54.60 298.90
1.435 0.832 3 55
37.33 22.06 5.96 108.94
53.58 27.42 6.53 9.97
3 4 0.99 13
5 5 2 1.34
125.24
8
61.68 20.89
8 3
9.70
I.50
141.89
14
242.52
23
Ultrasound measurements Mean
SD
43.11 19.37 6.16 6.26” 5.746 96.28” 105.286
3 2 1.42 1.13 11 12
44.95 19.89 4.95 6.79” 6.59b 101.1 In 11O.83b
3 2 1 0.96 1.16 7 12
43.05 18.74 6.16 6.14” 5.64b 94.00” 102.33b 20264
6 3 2 1.30 1.57 13 13 34
5
45
Ultrasonic measurements in sheep
carcass Texel; and 33.4 kg carcass Oxford sheep, respectively. Koohmaraie et al. (1996) reported an m. longissimus area of 18 cm2 in 25 kg carcass Dorset x Romonov sheep. The dissected muscle in the latter study weighed 745-787 g compared to 300 g in Omani sheep. LD measurements taken at three sites indicated that ultrasonic measurements well matched equivalent carcass measurements (Table 1) which is in line with findings in temperate sheep (Ward et al., 1992; Koohmaraie et al., 1996). That includes the LD area, circumference (determined by tracing and ellipse-fitting methods) and volume proposed in the current study. Table 2 presents the relationships between LD scanning measurements and those made on the dissected carcass. Significant 0, ~0.05) relationships between the LD scanning measurements at the 6th rib and second-last lumbar vertebra with the real measurements of the LD muscle included a positive correlation between the B measurement at lumber region and the muscle weight as well as positive correlations with real B and circumference measurements. There were positive correlations between all the LD scanning measurements at the 12th rib and the weight, B and circumference of the muscle. This indicates that scanning at this site should give satisfactory results for prediction of corresponding carcass measurements and confirms its traditional use by workers in this field (Binnie et al., 1995). Significant correlation between scanning and real measurements at 12th rib rather than at the other two sites may be attributed to the fact that the muscle at this site is scanned at its maximum thickness. This is supported by the greater value of B measurement at 12th rib than at 6th rib or lumbar sites (Table 1). Findings of the current study confirmed reports of positive correlations between scanning LD maximum depth (B) and area with corresponding carcass measurements and their use as indicators for the prediction of carcass muscle content (McEwan et al., 1989; Ward et al., 1992; Young and Deaker, 1994) even without adjustments for body or carcass weights. In the present study, the average area of the three sites had positive correlation with the muscle weight and the scanning volume had a positive correlation with the real TABLE 2
Correlation Coefficients between Real and Scanning LD Measurements as well as their Statistical Significance in Omani Sheep Slaughtered at an Average 26 kg Body Weight LD scanning measurements
LD real measurements Weight
Volume
Length
B
Circumference
LD scan at 6th rib Maximum depth (B Circumference (direct) Area (direct) LD scan at 12th rib Maximum depth (B) Circumference (direct) Area (direct)
0.35 0.29 0.15
0.33 -0.12 0.15
0.13 -0.23 0.22
0.48* 0.60* 0.27
0.43* 0.66** 0.56*
0.03 0.16 -0.04
-0.42 -0.32 -0.23
O&I* 0.66** 0.60*
040 0+56* 0*58*
LD scan at last lumber Maximum depth (B) Circumference (direct) Area (direct)
0.47* 0.43 0.41
0.11 0.34 0.28
-0.17 -0.08 -OX)4
0.27 0.41* 0.24
0.47* 0.49* 0.42
Average area Volume
0.49* 0.51*
0.18 0.59*
-0.03 -o&I
0.46 0.51*
0.42 0.42
*, **, *** Correlation coefficient is significant (p < 0.05; 0.01; 0.001).
-0.17 0.28 -0.07
46
0. Mahgoub
volume, weight and B measurement. This indicates that the estimation of volume and weight of the LD muscle from scanning images is useful and may be confidently used for prediction of total carcass muscle content. The current study proposes two novel scanning measurements; namely the circumference and the volume, as predictors of total carcass muscle content in sheep and possibly in other meat animals in addition to the traditional measurements (maximum depth and area). These two measurements had positive correlations with the corresponding measurement made on the dissected muscle. The circumference at the 12th rib had the highest r value and therefore, was regarded as the best predictor for the muscle weight in this study. The volume also had a significant relationship with the weight of the muscle. The positive correlations between the LD circumference and area measured by tracing (direct method) and those measured by ellipse best fit method indicated that either of the two methods may be confidently adopted. Table 3 shows regression equations and correlation coefficients between scanning measurements and total carcass muscle content. The LD scanning measurements made at the 12th rib also had the highest positive correlations with total carcass muscle content compared to measurements taken at the other two sites. This confirms that scanning at this TABLE 3 Regression Equations Describing the Relationships between Scanning and Real LD Measurements with Total Carcass Muscle Content of Omani Sheep Slaughtered at an Average 26 kg Body Weight Independent variable
Constanl (a)
Regression coefficient (b * sb)
Dependent variable
r
Scanning measurements At 6th rib Maximum depth (B)
3062.44
16.13 (30.03)
Circumference (D) Area (D) Circumference (F) Area (F)
2037.58 2922.26 2426.66 2920.90
12.35* (4.85) 0.73 (0.58) 10.00 (6.67) 0.73 (0.51)
0.56* 0.31 0.36 0.35
At 12th rib Maximum depth (B) Circumference (F) Area (F) Circumference (D) Area (D)
2017.49 1670.17 2094.25 1284.25 2370.36
68.02 (46.41) 1511** (3.95) 1.89*** (0.34) 20.73* (9.72) 1.49 (0.73)
0.34 0.48* 0.47* 0.70** 0.82***
At last lumber Maximum depth (B) Circumference (D) Area (D) Circumference (F) Area (F)
2521.47 2738.34 3081.13 250 1.62 2941.21
4588 5.90 0.46 9.23 0.72
0.40 0.29 0.27 0.32 0.30
Average area Volume External length
2421.18 2421.87 1198.32
Total carcass muscle
(26.66) (5.02) (0.42) (7.14) (0.59)
1.54* (0.57) 0.003***(0.0008) 3.70* (1.51)
*, **, *** Regression is significant (p < 0.05; 0.01; 0.001). D: circumference and cross-sectional area were measured by direct method (tracing). F: circumference and cross-sectional area were measured by ellipse best-fit method.
0.13
0.57* 0.69** 0.47*
41
Ultrasonic measurements in sheep
TABLE 4 Stepwise Regression Analysis (maximum r;? option) for the Best 4-variable Model of Scanning LD Measurements to Predict Total Carcass Muscle Content of Omani Sheep Slaughtered at an Average
26 kg Body Weight Variable
Intercept Maximum Depth (B) at 12th rib Circumference (by tracing) at 12th rib Average cross-sectional area (3 sites) Scanning volume
Parameter estimate
Standard error
Probability of F
R2
-231.000 56.915
506.283 29.031
0.659 0.081
0.904
28.219
4745
0.0002
-8.869
1.922
0.0013
0.024
0.005
0.0021
site should provide satisfactory results for prediction of carcass muscle content. At this site, the cross-sectional area and circumference measured by direct method had the highest r values (0.82 and 0.70, respectively) and therefore, were regarded as the best predictors of carcass muscle content. The volume, length, and average area of the cross-section, obtained by scanning were positively correlated with total carcass muscle content. The volume had a high r value (0.69) which indicates it as a good predictor of carcass muscle content. The current study indicates that these three measurements (circumference, cross sectional area at 12th rib and volume) could have a significant application in the meat industry for prediction of carcass muscle content in live animals, especially breeding stock. The method used in this study for estimation of muscle weight from its volume by ultrasonic scanning may be utilized for other individual muscles or muscle groups to determine muscularity and carcass muscle content. Stepwise regression analysis, using the maximum r* option, of LD scanning measurements (at the 0.1 significance level) is presented in Table 4. The analysis confirmed findings of correlation and regression analyses for best predictors of carcass muscle content. Maximum depth (B), circumference at 12th rib (by tracing method), average crosssectional area and volume accounted for 90% of the variation in the total carcass muscle content. The four measurements represented the best parameter estimates in a 4-variable model for prediction of total carcass muscle content. They may be adopted in a commercial model for evaluation of live sheep.
CONCLUSIONS This study showed significant relationships between longissimus muscle maximum depth and area measurements obtained from ultrasound scanning with corresponding carcass measurements. Two other scanning measurements, circumference and volume, also had significant relationships with corresponding carcass measurements. All these significant relationships indicate that they may be used for prediction of carcass muscle content in live sheep. ACKNOWLEDGEMENTS The author wishes to thank Dr Q. M. Ali of the College of Medicine, Sultan Qaboos University for valuable discussion at preliminary stages of study. MS A. L. Olvey,
48
0. Mahgoub
Mr N. M. Al-Saqry and Mr R. M. Al-Busaidi of the Deptartment. of Animal and Veterinary Sciences provided technical assistance. Published with the approval of the College of Agriculture, SOU as paper number 120497.
REFERENCES Anonymous (1989) Microimager 2000 Operation Manual. Ausonics Pty Ltd, Australia. Binnie, D. B., Farmer, R. J. and Clarke, J. N. (1995) Proceedings of the New Zealand Society of Animal Production 55, 111. Forrest, J. C., Aberle, E. D., Hedrick, H. B., Judge, M. D. and Merkel, R. A. (1975) Principles of Meat Science. Freeman & Co., New York, USA. Koohmaraie, M., Shackelford, S. D. and Wheeler, T. L. (1996) Journal of Animal Science 74, 70. Levie, (1967) The Meat Handbook. Avia Publishing Co., Westport, CT, USA. McEwan, J. C., Clarke, J. N., Knowler, M. A. and Wheeler, M. (1989) Proceedings of the New Zealand Society of Animal Production 43, 113. Mahgoub, 0. and Lodge, G. A. (1994) Animal Production 58,365. Owen, J. B. (1976) Sheep Production. Bailiere Tindall, London, UK. Palsson, H. (1939) Journal of Agricultural Sciences (Cambridge), 29, 544. Purchas, R. W., Davies, A. S. and Abdullah, A. Y. (1991) Meat Science 30, 8 1. SAS Institute Inc., (1985) SAS User’s Guide: Statistics, Version 5, ed. SAS Institute Inc., pp. l-956. NC, Cary. Stouffer, J. R. and Westerveld, R. G. (1976) Journal of Clinical Ultrasound 5, 124. Ward, B. G., Purchas, R. W. and Abdullah, A. Y. (1992) Proceedeings of the New Zealand Society of Animal Production. 52, 33.
Young, M. J. and Deaker, J. M. (1994) Proceedings of the New Zealand Society of Animal Production 54, 215.
Young, M. J., Nsoso, S. J., Logan, C. M. and Beatson, P. R. (1996) Proceedings of the New Zealand Society of Animal Production 56, 205.