RELATIONSHIPS BETWEEN LEAN GROWTH
AND EFFICIENCY OF PORK PRODUCTION


Sam Buttram, Ph.D.
DEKALB Swine Breeders, Inc.

INTRODUCTION

The goal of the swine industry must be the economical production of high quality, lean pork. The traits that are most economically important to the commercial swine industry are sow productivity, growth rate, feed conversion, lean percentage, and meat quality. Other traits such as lean growth and lean tissue feed conversion (LTFC) that have been suggested as selection objectives (Fowler et al., 1976) are essentially combinations of the traits listed above. Nonetheless, both lean growth and LTFC serve as important tools for characterizing pig genotypes. Lean growth in particular is an essential ingredient in developing swine growth models designed to optimize production systems and minimize costs (Schinckel and de Lange, 1996). The purposes of this presentation are to review and discuss: 1) how to measure lean growth; 2) relationships between lean growth and other traits of economic importance; and 3) selection for lean growth.

MEASURING LEAN GROWTH

Lean growth rate is generally defined as the increase in weight of lean per unit of time. Average daily lean gain (ADLG) is calculated much the same as average daily live weight gain (ADG) except that weight of lean is more difficult to measure than live weight. The following formula is used to compute ADLG:

Both initial and final lean weights are determined most accurately by carcass dissection and chemical analysis. Chemical analysis is necessary to standardize the fat content of the lean. Lean is usually standardized to 0% (fat-free), 5% or 10% fat. Carcass dissection is extremely expensive and thus limited in its use. A less costly approach to estimate ADLG is to predict initial and final weights of lean using traits that can be more easily measured on the live animal or carcass.

Initial weight of lean (containing 5% fat) can be predicted for 40-60 lb. feeder pigs from a prediction equation (NPPC, 1991) that uses only live weight as a dependent variable. This equation assumes that there is little variation in lean percentage of 40-60 lb. feeder pigs. Unpublished data from Schinckel and de Lange (1996) showed that the standard deviation of the 14 genotype x sex means for weight of fat-free lean in 60 lb. pigs representing seven genotypes and two sexes was about 1.34 lb., with a range of 16.1 to 22.0 lb. Therefore, there is indeed some variation in lean percentage of 40-60 lb. feeder pigs, but any error associated with using the above equation for predicting initial weight of lean is not likely to affect ADLG dramatically.

Final weight of lean can be predicted using either live animal or carcass traits. Equations are available for predicting the weight of lean (containing 5% fat) using realtime ultrasonic measurements of backfat thickness and loin muscle area on the live hog or using carcass measurements of backfat thickness and loin muscle area on ribbed carcasses (NPPC, 1991). In addition, most pork slaughter plants measure either backfat thickness and loin muscle depth with an optical probe, or midline backfat thickness, along with warm carcass weight, which can be used to predict the percentage of fat-free lean in the carcass (NPPC, 1994). Attention must be paid to the percentage of fat to which lean is standardized when using these equations. While these equations may be useful for characterizing pigs for lean growth, they may also carry some bias when used across genotypes and sexes (Wagner et al., 1993). For most accurate predictions, it is necessary to develop specific equations for each genotype and sex.

DEKALB Swine Breeders has developed separate equations to predict lean content of its different genetic lines and sexes using weight and realtime ultrasonic measurements of backfat thickness, loin muscle area and loin muscle depth and weight of the live animal. In addition, research at DEKALB has shown that loin muscle area is superior to loin muscle depth for predicting lean when combined with live weight and backfat thickness. When loin muscle depth is included in the prediction equations in addition to weight, backfat thickness, and loin muscle area, the R-square is not significantly improved.

Swine growth models require estimates of daily whole-body protein accretion, or protein accretion curves. Lean growth curves are very similar to protein accretion curves and can be used to construct protein accretion curves. Schinckel and de Lange (1996) described two economical methods for estimating protein accretion curves. One method is to obtain serial live weights and realtime ultrasonic measurements of backfat thickness and loin muscle area on pigs at three-week intervals during the growing-finishing period. These measurements are used for predicting fat-free lean growth curves and protein accretion rates. The second method is to determine mean fat-free ADLG during the growing-finishing period using procedures like the ones described above. Mean ADLG can then be used to estimate the daily protein accretion curve (Schinckel et al., 1996).

RELATIONSHIPS WITH OTHER TRAITS

Lean growth is essentially a function of growth rate and lean percentage. Pigs that grow fast and stay lean have high lean growth rates. Table 1. illustrates how both growth rate (ADG) and carcass lean percentage influence ADLG. It can be seen that increasing either ADG or carcass lean percentage results in an increase in ADLG and that when both are increased, their effects are additive. In addition, pigs with identical lean growth rates may have very different ADG's and/or carcass lean percentages.

Schinckel (1991) sampled 236 market hogs representing a wide range of genetic sources and breed crosses to study the variation in ADLG and LTFC that existed in the industry. Table 2. shows the relationships he found between ADLG, LTFC and other production traits. Lean growth rate and LTFC were highly correlated (r=-.91) with each other, but had different relationships with the other traits measured. Lean growth rate was moderately associated (r=.37) with ADG, but was not associated (r=-.04) with average daily feed intake. This suggests that pigs which ate more feed did not necessarily have the potential to deposit more lean. Conversely, LTFC was associated (r=.42) with average daily feed intake, but showed no association (r=-.09) with ADG. This indicates that pigs which ate more feed were less efficient than pigs which ate less feed, presumably because the excess feed was going to produce fat which requires more energy than lean deposition. Both ADLG and LTFC were highly correlated (r=.78 and r=-.88, respectively) with carcass lean percentage and moderately correlated (r=-.56 and r=.55, respectively) with feed/gain.

One might predict from these relationships (Table 2.) that increasing lean growth should result in improvements in ADG, carcass lean percentage, and feed/gain, with no little or effect on feed intake. Improvement in LTFC should result in increased carcass lean percentage and improved feed/gain, but would result in decreased feed intake and no improvement in ADG. In the long run, further decreases in LTFC might even cause a reduction in ADG as a result of the decreased feed intake.

It is important to note that these pigs were tested in an environment that allowed the pigs to maximize feed intake and to express their genetic potential for ADLG. Under sub-optimal conditions, the relationships between feed intake and ADLG or LTFC will change. For instance, if feed intake is limited by the environment to a level that does not allow maximum lean growth, one might expect that pigs which eat more feed would also have higher ADLG, resulting in a positive correlation between feed intake and ADLG. Likewise, those same pigs would not be expected to have such poor LTFC because some of the additional feed is now being used to deposit lean tissue instead of fat.

Clutter (1994) reviewed estimates of genetic correlations between lean growth and sow productivity traits. Although he concluded that these estimates were based on small populations and lacked precision, there was little evidence that the genetic relationships between lean growth and sow productivity traits were other than zero.

The relationships between lean growth and meat quality traits are not well defined, but some estimates of genetic correlations between ADLG and several meat quality traits were provided by the NPPC's recent terminal sire evaluation program (NPPC, 1995). These estimates (Table 3.) indicate that ADLG was not closely related to any of the meat quality traits measured. The closest relationships were between ADLG and marbling score (r=-.20) or lipid content of loin muscle (r=-.20), both of which are measures of intramuscular fat.

SELECTION FOR LEAN GROWTH

Several selection studies have demonstrated that lean growth can be improved through selection. Cleveland et al. (1982) selected on an index of ADG and backfat thickness for five generations. ADG increased by .03 lb. per generation and backfat thickness decreased by .018 in. per generation. After five generations of selection, barrows from the select and control lines were evaluated (Table 4.). Barrows selected for increased growth rate and decreased backfat thickness had 19.8% higher ADLG and 18% better LTFC than nonselected barrows, but there was little difference in feed intake.

Cameron and Curran (1994) selected Large White and British Landrace pigs for four generations of divergence selection for ADLG, LTFC and average daily feed intake (DFI). Selection for increased ADLG increased ADG and decreased backfat thickness and feed/gain (Table 5.), but had little or no effect on DFI. Selection for improved LTFC had no effect on ADG, but backfat thickness, feed/gain and DFI were reduced. Selection for increased DFI increased ADG and feed/gain in both populations, and increased backfat thickness in Large White pigs, but not in Landrace pigs. Overall conclusions were that efficiency of lean growth was improved by selection for either ADLG or LTFC. This improvement was due to increasing growth rate with little change in feed intake from selection for ADLG, but was due primarily to reduced feed intake with selection on LTFC.

These two experiments illustrate three themes that are found throughout the literature with regard to selection for lean growth rate and LTFC. First, selection for increased lean growth will result in improvements in ADG, lean percentage, feed/gain and LTFC, with little or no effect on feed intake. Second, selection for improved LTFC will result in improvements in lean percentage and feed/gain, but will cause no improvement in ADG and will decrease feed intake. Third, selection for increased growth rate and increased lean percentage (decreased backfat thickness) will result in simultaneous improvements in both traits and in improved ADLG.

Clutter (1994) reviewed the effects of selection for lean growth on reproductive performance in a number of selection experiments. In this extensive review, he found no conclusive evidence that selection for lean growth will have a detrimental effect on sow reproduction. He did note, however, a general lack of adequate precision regarding many of the studies. This is in agreement with his previously mentioned findings regarding genetic correlations.

Meat quality has only recently become a major issue for the swine industry. As a result, no long term selection experiments have measured meat quality traits in order to observe correlated responses that may take place when selecting for lean growth. Based on the correlations in Table 3., selection for lean growth may reduce intramuscular fat content, but would not be expected to affect other measures of meat quality.

Based on the information presented here, lean growth would seem to be a good selection objective. In terms of the economically important traits listed at the beginning of this paper, selection for lean growth should result in improvements in growth rate, feed conversion, and lean percentage, with little or no effect on sow productivity. Its effect on meat quality remains a question. However, when it comes to implementing a selection program, how should lean growth be used? Should ADLG be measured on each pig and used as a composite trait for BLUP analysis? Should ADLG be given an economic value and used in an index with sow productivity or meat quality traits? Probably not.

First, selection for ADLG as a composite trait does not allow for appropriate economic weightings to be given to the individual component traits. For example, as shown in Table 1., two pigs may have the same ADLG, but one pig may have higher growth rate and the other, higher lean percentage. Unless the two traits have the same economic value, the two pigs should not be considered of equal value. Second, separation of ADLG into its component traits allows one to account for different phenotypic and genetic correlations and different heritabilities of the component traits. Furthermore, since ADLG cannot be measured directly, but must be predicted from indirect measurements, responses in important correlated traits may depend upon the choice of indirect measurements (Bennett, 1992). This problem can be somewhat overcome using a statistical method described by Bennett (1992) for restricting the regression of correlated traits on predicted ADLG to be equal to the regression of the correlated traits on ADLG measured directly.

Perhaps a more simple, yet practical and effective, approach is to go back to the basics of index selection and genetic improvement. First, accurately measure the traits that are economically important to the swine industry today and that are anticipated to be important in the future. Next, estimate genetic parameters for the traits or use parameter estimates that are available (NSIF, 1987). Then, compute best linear unbiased estimates of breeding values for each of the traits using complete pedigrees. Finally, establish relative economic values or use recommended values (NSIF, 1987) to weight the breeding values of each trait in order to construct an index that will optimize genetic progress toward the selection objective. This process will be effective whether the selection objective is increased lean growth or LTFC of a sire line, increased litter size of a dam line, or improved meat quality. It will also reduce the problems of uncertainty associated with selection on a composite trait as described above.

SUMMARY

Lean growth is related to the efficient production of pork in at least two ways. First, lean growth curves are essential components of swine growth models. Characterizing genotypes according to their lean growth potential, i.e. their lean growth curves will greatly enhance the effectiveness of these models to optimize production systems and minimize costs. Second, improvement of lean growth continues to be an important selection objective for the swine industry. However, the most effective strategy will likely be to select for lean growth in combination with sow productivity and meat quality traits using an index of breeding values for traits that are appropriately weighted according to their economic values.

REFERENCES

Bennett, G. L. 1992. A model for selecting for improved lean gain. Proc. National Swine Improv. Fed. p. 31.

Cameron, N. D., and M. K. Curran. 1994. Selection for components of efficient lean growth in pigs 4. Genetic and phenotypic parameter estimates and correlated responses in performance traits with ad-libitum feeding. Anim. Prod. 59:281.

Cleveland, E. R., P. J. Cunningham, and E. R. Peo, Jr. 1982. Selection for lean growth in swine. J. Anim. Sci. 54:719.

Clutter, A. C. 1994. The effect of lean growth selection on reproductive performance. Proc. National Swine Improv. Fed. P. 45.

Fowler, V. R., M. Bichard and A. Pease. 1976. Objectives in pig breeding. Anim. Prod. 23:365.

NPPC. 1991. Procedures to evaluate market hogs. National Pork Producers Council, Des Moines, IA.

NPPC. 1994. Fat-free lean index users guide. National Pork Producers Council, Des Moines, IA.

NPPC. 1995. Genetic evaluation terminal line program results. National Pork Producers Council, Des Moines, IA.

NSIF. 1987. Guidelines for uniform swine improvement programs. National Swine Improv. Fed.

Schinckel, A. P. 1991. Optimizing lean growth in swine - a genetic perspective. Proc. Can. Annu. Conf. for Feed Manufacturers - Guelph, Ontario.

Schinckel, A. P., P. V. Preckel and M. E. Einstein. 1996. Prediction of daily protein accretion rates of pigs from estimates of fat-free lean gain between 20 and 120 kilograms live weight. J. Anim. Sci. 74:498.

Schinckel, A. P. and C. F. M. de Lange. 1996. Characterization of growth parameters needed as inputs for pig growth models. J. Anim. Sci. 74:2021.

Wagner, J. R., A. P. Schinckel, and J. C. Forrest. 1993. Genotype and sex biases in the estimation of pork carcass composition. Proc. National Swine Improv. Fed. p. 47.

TABLE 1. Average Daily Lean Gain of Pigs with Different Growth Rates (ADG) and Carcass Lean Percentages.
ADG.lb
Lean Percentage, %
48
50
52
54
56
58
60
1.60
.58
.61
.64
.67
.70
.73
.76
1.70
.61
.65
.68
.71
.75
.78
.81
1.80
.65
.68
.72
.75
.79
.83
.86
1.90
.69
.72
.76
.80
.83
.87
.91
2.00
.72
.76
.80
.84
.88
.92
.96
2.10
.76
.80
.84
.88
.92
.96
1.00
2.20
.79
.84
.88
.92
.97
1.01
1.05

TABLE 2. Phenotypic Correlations Between Performance Traits and Lean Growth (ADLG) and Lean Tissue Feed Conversion. (LTFC).a
Trait
ADLG
LTFC
Average daily gain
.37
-.09
Backfat thickness
-.61
.82
Loin muscle area
.73
-.69
Lean percentage
.78
-.88
Feed/gain
-.56
.55
Daily feed intake
-.04
.42
LTFC
-.91
a Adapted from Schinkel, 1991.

TABLE 3. Genetic Correlations Between Performance Traits and Lean Growth (ADLG).a
Trait
ADLG
Average daily gain
.61
Backfat thickness
-.61
Loin muscle area
.62
Color
-.07
Marbling
-.20
Firmness
-.10
Ultimate pH
-.13
Water holding capacity
.12
Drip loss
.12
Tenderness
-.09
Juiciness
.04
Flavor
.08
Intramuscular fat
-.20
a Adapted from NPPC, 1995.

TABLE 4. Performance of Select and Control Line Barrows.a
Trait
Select line
Control line
ADG, lb.
1.70
1.55
Backfat, in.
.98
1.18
Feed/gain
3.18
3.44
DFI, lb.
5.41
5.33
ADLG, lb.
.61
.51
LTFC
8.83
10.77
a Adapted from Cleveland et al., 1982.

TABLE 5. Responses to Selection for Lean Growth Rate, Lean Feed Conversion or Daily Feed Intake in Large White (LW) and British Landrace (LR) Lines.a
Line Selection Group
ADG,lb.
DFI,lb.
FCR
BF,in.
LW Lean growth rate
.12
-.04
-.26
-.15
Lean feed conversion
.00
-.20
-.11
-.15
Daily feed intake
.21
.70
.13
.15
LR Lean growth rate
.22
.15
-.25
-.08
Lean feed conversion
.02
-.38
-.22
-.08
Daily feed intake
.11
.51
.12
-.01
a Adapted from Cameron and Curran, 1994.


Return to NSIF HOME PAGE
Last modified July 1997.
Ping Liang, Department of Animal Science.