The Professional Animal Scientist 34:156–166 https://doi.org/10.15232/pas.2017-01635 ©2018 American Registry of Professional Animal Scientists. All rights reserved.
Use of residual feed intake as a selection criterion on the performance and relative development costs of replacement beef heifers D. Damiran,*† G. B. Penner,† K. Larson,* and H. A. (Bart) Lardner*†1 *Western Beef Development Centre, Humboldt, SK, Canada, S0K 2A0; and †Department of Animal and Poultry Science, University of Saskatchewan, Saskatoon, SK, Canada, S7N 5A8
ABSTRACT Two heifers groups differing in residual feed intake (RFI) were compared with a third control (CON; n = 20) group of randomly selected heifers for performance, reproductive efficiency, and system economics to first calving and repeatability of RFI ranking, with all 3 groups selected from the same cohort. Following weaning, 70 Angus heifers (initial BW = 260 ± 3 kg; 6 mo of age) from a single cohort were fed a forage-based diet (10.0% CP; 65.2% TDN) for 93 d (period 1) where BW, DMI, ADG, G:F, and RFI were evaluated. After period 1 RFI testing, 40 heifers were classified into 2 groups [20 efficient heifers (low RFI; RFI = −1.01 ± 0.10 kg/d) and 20 inefficient heifers (high RFI; RFI = 0.77 ± 0.08 kg/d)] and then selected for a second feeding trial (period 2) and compared with the 20 CON heifers. All 60 heifers in period 2 (BW = 322 ± 2.9 kg; 10 mo of age) were fed for 93 d on a similar forage-based diet (11.0% CP; 66.5% TDN). Low-RFI heifers had the lowest (P = 0.01) RFI value of −0.33 kg/d, followed by CON and high-RFI heifers, −0.09 and 0.42 kg/d, respectively. Control heifers tended (P = 0.08) to have lower ADG (0.83 kg/d) compared with low-RFI (0.92 kg/d) or highRFI heifers (0.91 kg/d), and low-RFI heifers tended (P = 0.08) to have greater G:F (0.10 ± 0.003) than either CON (0.9 ± 0.003) or high-RFI heifers (0.09 ± 0.003). Spearman rank correlation for RFI between period 1 and 2 was 0.58 (P < 0.01); however, 51% of heifers had a different RFI value in period 2 compared with period 1. First-calf pregnancy rates were 80% for low RFI, 93% for CON, and 100% for high RFI (χ2; P = 0.09). Winter feed costs were ~Can$25 per heifer lower for low-RFI heifers compared with high-RFI animals. Heifers with increased feed efficiency may exhibit reduced reproductive performance, suggesting further research is needed.
The authors declare no conflict of interest. Corresponding author:
[email protected]
1
Key words: feed efficiency, heifer, reproductive efficiency, residual feed intake
INTRODUCTION Efficient use of feed and reduced feed costs could improve the economic sustainability of the beef cattle industry. Residual (or net) feed intake (RFI) is a measure of feed efficiency and is defined as the difference between actual and expected feed intake to support maintenance and ADG for a group of cattle (Archer et al., 1999). Measurement and prediction of RFI has gained popularity as a selection tool to improve feed efficiency in beef cattle (Blair et al., 2013). It has been reported that cattle with low RFI had similar rates of BW gain to those with high RFI, even though feed intake was lower for the low-RFI cattle (Kelly et al., 2010; Durunna et al., 2012). Thus, selection for RFI may present an opportunity to reduce feed costs along the entire beef supply chain, including the cow-calf sector. Despite potential to improve feed efficiency, adoption of RFI in the cow-calf sector has been constrained, because measuring RFI is technically challenging and costly. Although RFI is a moderately heritable trait (Archer et al., 2002; Blair et al., 2013), substantial reranking of RFI status has been reported for cattle fed the same diet over 2 consecutive periods (Durunna et al., 2012). Moreover, previous research has been conducted with backgrounding or finishing diets that are not representative of the high-forage diets used for developing replacement heifers. In fact, there have been few published studies evaluating RFI with high-forage diets (Kelly et al., 2010; Manafiazar et al., 2015) and with respect to the potential effect on replacement heifer reproductive performance (Kelly et al., 2010; Basarab et al., 2011; Loyd et al., 2011; Black et al., 2013; Hafla et al., 2013; Randel and Welsh, 2013). The objectives of this study were to evaluate performance, reproductive efficiency, and system economics for heifers classified as having either high or low feed efficiency based on RFI values in comparison with heifers randomly selected as replacements.
157
Feed efficiency of beef heifers
MATERIALS AND METHODS Study Site and Management All experimental procedures were approved by University of Saskatchewan Animal Research Ethics Board (Protocol No. 20090107), and heifers were cared for according to the guidelines of the Canadian Council on Animal Care (2009). The study was conducted at the Western Beef Development Centre’s Termuende Research Ranch located near Lanigan (lat. 51°51′N, long. 105°02′W), Saskatchewan, Canada. Daily average temperatures were obtained from Environment Canada (www.climate.weatheroffice.ec .gc.ca) for Watrous, Saskatchewan, approximately 50 km southeast of the study site (51°48′N, 104°51′W).
Animals and Management—Period 1 Spring-born (April to late May) Angus heifers (n = 90), suitable for herd replacements, were sourced from the main Western Beef Development Centre herd, weaned in early October, and allocated to the study. There were 2 consecutive feeding periods during the study, with data collected from November 21, 2012, to May 25, 2013. At weaning, each heifer was identified with a half-duplex radio frequency transponder button (Allflex USA Inc., Dallas/Ft. Worth Airport, TX) in the right ear. All heifers were considered as cohorts for each of the RFI calculations in the 2 periods. For the study, 3 drylot pens were used, each pen (50 × 120 m) was surrounded by wood slatted fences with 20% porosity and contained an open-faced shed in one end, and water was supplied to each pen in a heated water bowl. In 2 of the pens, feed intake was measured, with 8 GrowSafe Intake (GrowSafe Systems Ltd., Airdrie, Alberta, Canada) feed bunks per pen. The remaining pen had a fence-line bunk (0.5 m of bunk space per animal). Wood chips were used as bedding during inclement weather conditions. Out of the original cohort of 90 heifers, 20 (control; CON) were randomly selected before the start of period 1. The
CON heifers were group fed; therefore, measurement of individual DMI during period 1 was not possible. However, the amount of feed provided to the CON group, which was fed twice daily ad libitum, was recorded daily to estimate average DMI. The remaining 70 heifers were then randomly allocated to the 2 drylot pens fitted with GrowSafe feed bunks. During period 1 (November 21, 2012, to February 22, 2013; postweaning period), a 21-d adaptation period was followed by a 72-d feeding period, where daily DMI and cumulative BW gain were measured (Archer et al., 1999).
Animals and Management—Period 2 After completing period 1 only 40 heifers [20 most efficient heifers (low RFI) and 20 least efficient heifers (high RFI)] of the 70 animals continued to be evaluated in a second feeding trial (period 2; prebreeding period). The 20 CON heifers were also included in the period-2 trial. In period 2, the low-RFI, high-RFI, and CON heifers were equally divided into 1 of 2 pens with GrowSafe bunks as described previously. All heifers were provided a 21-d adaptation period to ensure all heifers consumed feed from the GrowSafe bunks. Subsequently, a 72-d feeding period was conducted, just before the start of the breeding season. In both period 1 and period 2, heifers were fed a similar diet formulated to support growth rates of 0.8 kg/d (NRC, 2000) and reach a prebreeding target BW of 62% (~395 kg) of mature BW (~637 kg). The forage-based diet contained processed bromegrass–alfalfa hay (10.1% CP; 55.0% TDN) and rolled barley (11.0% CP; 75.0% TDN) (DM basis) and was fed ad libitum twice daily (0800 and 1500 h) for the next 185 d (period 1 and period 2; Table 1). However, daily temperatures during the study ranged from −10.0 to −21.3°C (average −15.6°C) and from 5.6 to −6.7°C (average −0.7°C) for period 1 and period 2, respectively. Therefore the hay:barley ratio in diet was adjusted as predicted by CowBytes Beef Ration Balancing Program Version 5.3.1 (AAFRD, 2011; Table 1).
Table 1. Composition of diet fed to heifers during study (DM basis) Item Ingredient composition, %, as-fed basis Mixed hay Barley grain Nutrient composition, mean ± SD CP, % TDN,1 % NEm, Mcal/kg NEg, Mcal/kg 1
Period 1
72.0 28.0 10.0 ± 0.3 65.2 ± 0.4 1.48 ± 0.01 0.89 ± 0.01
Calculated using the Weiss equation (Weiss et al., 1992).
Period 2
70.0 30.0 11.0 ± 0.3 66.5 ± 0.6 1.52 ± 0.02 0.93 ± 0.02
158
Damiran et al.
In addition, heifers had ad libitum access to a commercial 2:1 mineral supplement [15.5% Ca, 7.0% P, 30 mg/ kg Se, 20 mg/kg Co, 200 mg/kg I, 1,500 mg/kg Cu, 5,000 mg/kg Mn, 5,000 mg/kg Zn, 1,000 mg/kg Fe, 1 mg/kg F, 500,000 IU/kg vitamin A (minimum), 50,000 IU/kg vitamin D3 (minimum), 2,500 IU/kg vitamin E (minimum); Cargill Animal Nutrition, Winnipeg, Manitoba, Canada] and cobalt-iodized salt [99.0% NaCl (minimum), 39.0% Na, 180 mg/kg I, 120 mg/kg Co; The Canadian Salt Company Ltd., Pointe-Claire, Quebec, Canada] over the duration of the study. Overall, for the entire period heifers were fed a diet formulated to support growth to reach a prebreeding target BW of 62% (~395 kg) of mature BW (~637 kg; Lardner et al., 2014). Diet ingredients were sampled every 21 d, placed in paper bags, dried in a forced-air oven at 55°C for 72 h for DM, and then ground to pass through a 1-mm screen using a Wiley mill (Model 4, Thomas Scientific, Swedesboro, NJ) for chemical analysis. Dry matter (method 930.15), ash (method 942.05), ether extract (method 920.02), CP (method 984.13), and ADF (method ID 954.01) were analyzed according to AOAC (1990). Neutral detergent fiber was determined according to Van Soest et al. (1991) with sodium sulfite and heat stable α amylase and expressed including residual ash. Total digestible nutrients and DE were determined according to Weiss et al. (1992), and NEm and NEg were estimated using the NRC beef model (NRC, 2000). Chemical composition of all feedstuffs used in the study is presented in Table 1.
BW and BCS Measurement Measures of BW were taken over 2 consecutive days at the beginning and end of each feeding period (period 1 and period 2), before morning feeding, and every 14 d during each period. Average daily gain was calculated by period for each heifer by subtracting the initial BW from the end BW and dividing by the number of days in each period. Body condition score was determined by a trained technician at the start and end of the trial on a scale of 1 to 5 (1 = emaciated to 5 = grossly fat; Lowman et al., 1976; Marx, 2004).
Heifer Management from Breeding to Calving Following period 2 (during the breeding season and until pregnancy diagnosis), heifers were managed as a single group on mixed crested wheatgrass [Agropyron cristatum (L.) Gaertn.] and smooth bromegrass (Bromus inermis Leyss.) pasture. From pregnancy determination to calving, pregnant heifers grazed in field paddocks on swathed barley (10.8% CP, 69.3% TDN, DM basis) from early November 2013 to mid-February 2014, followed by being housed in a drylot and fed free-choice grass–legume hay (86.6% DM, 9.7% CP, 58.5% TDN, DM basis) with a daily provision of a range pellet (2.7 kg/d; 13.6% CP, 79.5% TDN, DM basis) from mid-February 2014 to end of calving (early June 2014). The winter and calving diets were
designed to meet NRC (2000) recommended protein and energy requirements for pregnant beef heifers similar to the heifers used in the current study. Before breeding in June, all heifers (n = 60) received a vaccine against bovine respiratory syncytial virus, infectious bovine rhinotracheitis, bovine viral diarrhea, and parainfluenza 3 (Express 5; Boehringer Ingelheim Vetmedica Inc., St. Joseph, MO); a Clostridium 8-way modified live vaccine (Covexin 8; Schering-Plough Animal Health, Guelph, Ontario, Canada); and an anthrax spore vaccine (Colorado Serum Company, Denver, CO). Heifers were managed as a single group and were exposed to 3 bulls that passed a breeding soundness evaluation for a 63-d breeding season. Estrus was synchronized with a single 2-mL injection of cloprostenol sodium, an analog of prostaglandin F2α (Estroplan, Parnell Technologies Pty. Ltd., Alexandria, NSW, Australia) administered 5 d after bulls were placed with heifers. Pregnancy rates were determined by rectal palpation at approximately 50 d after bulls were removed. Body weight and BCS were also measured at this time. All BW data at pregnancy diagnosis was adjusted for conceptus weight using the following equation from the NRC (2000):
conceptus weight (kg) = (actual calf birth weight × 0.01828) × e[(0.02×t)− (0.0000143×t×t)],
where t = day of pregnancy. Date of conception was determined by subtracting 282 d from the subsequent calving date (DeRouen et al., 1994). Calving began in March 2014, and calving difficulty was recorded. Calving difficulty was evaluated on a 1 to 5 scale, where 1 = no assistance, 2 = easy pull, 3 = mechanical pull, 4 = hard mechanical pull, and 5 = caesarean section. All calves were weighed within 24 h of birth and received an i.m. injection of vitamin A, D (Vitamin AD3 Forte, Alfasan, Woerden, Holland), and E (Dystosel, Zoetis, Canada Inc. Kirkland, Quebec, Canada) equating to 250,000, 38,000, and 68 IU/calf, respectively. All male calves were castrated using rubber elastrator rings, and all calves were individually identified with a visual plastic management ear tag. Calving span and calving distribution were calculated. Calf birth BW was adjusted for sex according to the Beef Improvement Federation (1990) guidelines. All calf weaning BW are reported as 205-d adjusted weaning BW.
Feed Cost Analysis of Heifer Groups Winter development costs were calculated using a similar procedure as described by Lardner et al. (2014) and Larson (2014). All dollar values expressed are in Canadian dollars. Calculated feed costs for heifer groups were based off feeding records for the entire 185-d replacement heifer development period (period 1 + period 2). Using individual feed intake data and feed ingredient costs, daily feed costs were calculated for each heifer, followed by feed
Feed efficiency of beef heifers
cost for the entire heifer development period. The price of hay and rolled barley grain was Can$94/t and Can$259/t, respectively. In this study, the feed was the only cost that differed between the groups. Costs that did not vary between groups (i.e., costs associated with bedding, yardage, labor, equipment use, infrastructure, and manure removal) were adapted from previous studies (Lardner et al., 2014) conducted at the Western Beef Development Centre.
Calculations, Reranking, and Statistical Analysis Residual feed intake was calculated as the difference between actual feed intake of each individual and the expected feed intake of the group, as described by Arthur et al. (2001), where actual DMI was regressed on mid-test metabolic BW (BW0.75) and ADG to calculate an expected DMI for each heifer using the PROC REG procedure (SAS Institute Inc., Cary, NC). The model for expected feed intake was
yi = b0 + b1ADGi + b2MWTi + ei,
where b0 = the regression intercept, b1 = the partial regression coefficient of feed intake on ADG, b2 = the partial regression coefficient of feed intake on mid-test BW0.75, ADGi = the ADG of animal i, MWTi = the mid-test metabolic (BW0.75) BW of animal i, and ei = the uncontrolled error of the ith animal. Based on the performance during period 1 for the 40 selected heifers, individual feed intake and efficiency was reclassified in period 2 to determine whether period influenced the phenotypic classification of animals. Heifers identified as efficient (low RFI) and inefficient (high RFI) based on phenotypic classification from period 1 were used in the period-2 feeding trial to evaluate the repeatability of RFI ranking. Standard deviations above and below the mean RFI were used to group the selected animals into low-RFI (RFI 0.5 SD above the mean) groups by several authors (Nkrumah et al., 2006; Lancaster et al., 2009; Kelly et al., 2010; Durunna et al., 2012). In the current study, differences in heifer RFI between period 1 and period 2 were classified into (1) changed by >−0.5 SD (assuming shifted to low-RFI group), (2) changed by +0.5 SD (shifted to high-RFI group) to investigate the characteristics of heifers with substantial changes in RFI value. It is important to note that only the 40 heifers included in both periods were used in this assessment (data from all other heifers were removed before calculating the RFI rank in periods 1 and 2). Data were analyzed using the MIXED procedure of SAS 9.2. The model used for the analysis was Yij = μ + Ti + eij, where Yij was an observation of the dependent variable ij; μ was the population mean for the variable; Ti was
159
the fixed effect of the contemporary heifer group (CON, low-RFI, and high-RFI group); and eij was the random error associated with the observation ij. Heifer was considered an experimental unit. Percentage data variables, pregnancy rate, and calving distribution within cycle were analyzed by Chi-squared generated by ordinal logistic fit with JMP software (SAS Institute Inc.), with differences declared at P < 0.05. For all data, when a significant difference was detected (P < 0.05), means were separated using the Tukey-Kramer post hoc test. The correlations between feed efficiency traits of heifer groups were calculated using the CORR procedure of SAS, and correlation coefficients were classified as strong (r >0.6), moderate (0.6 > r > 0.4), or weak (r −0.5 SD (shifting to low-RFI group), ±0.5 SD (maintaining in the same RFI group), and >0.5 SD (shifting to high-RFI group) or switched RFI classes were calculated.
RESULTS AND DISCUSSION Selection of Low- vs. High-RFI Heifers The objectives of the current study were to compare the performance and economics of replacement heifer selection systems that were sourced from conventionally selected heifers with a system that included selection of heifers from the same cohort, based on RFI status. During the initial RFI evaluation of heifers in period 1, there was substantial variation in RFI status, ranging between −2.2 and 1.5 kg/d (Table 2). This range is similar to that reported previously for cattle fed backgrounding diets (Durunna et al., 2011, 2012). Overall, heifers had a mean initial BW of 260 kg (SD = 21.6), DMI of 7.7 kg/d (SD = 1.1), and G:F of 0.08 kg of BW gain/kg of DMI (SD = 0.01) (Table 2). Average heifer ADG in period 1 was 0.64 kg (Table 2) and did not reach the targeted level of 0.80 kg/d, suggesting the colder temperatures experienced in period 1 might have affected heifer performance. Heifers had ad libitum access to feed delivered twice daily; therefore, heifer DMI was adequate (2.7% of BW, data not shown) in period 1. Lardner et al. (2014) suggested that to achieve targeted BW gains, heifers need a diet with higher nutrient density (i.e., increased energy) than that of the current study. In agreement with heifer growth in the current study, Block et al. (2010) found a lack of accuracy and precision in the
160
Damiran et al.
NRC (2000) predictions of energy requirements and DMI of wintering beef cattle in western Canada. From the calculated results in period 1, the 20 most efficient heifers (low RFI; n = 20; RFI = −1.01 ± 0.10 kg/d) and 20 least efficient heifers (high RFI; n = 20; RFI = 0.77 ± 0.08 kg/d) were selected for the second feeding period (period 2).
Heifer Performance, Feed Intake, and Feed Efficiency Average performance between heifer groups in period 2 is shown in Table 3. The RFI values measured in period 2 were greatest for (P < 0.001) high RFI (0.42 ± 0.13 kg/d) but did not differ (P > 0.05) between CON and low RFI (−0.09 ± 0.15 and −0.33 ± 0.12, respectively). There were no statistical differences (P > 0.05) among the 3 heifer groups for initial BW, DMI, and BCS. In addition, final BW did not differ (P = 0.34) among heifer groups, averaging 391 kg (SE ±3.5 kg) across all groups, equating to 61.4% of mature BW (targeted BW) before breeding. The low-RFI and high-RFI heifers tended (P = 0.08) to have greater ADG compared with CON heifers, and G:F
tended (P = 0.08) to be lower for CON and high RFI than low RFI. As described previously, CON heifers were group fed twice daily ad libitum (full access to feed bunk) during period 1, whereas those heifers in low- and high-RFI treatment groups were fed in a GrowSafe system (less than full access to feed bunk). Thus, at the start of period 2, CON heifers would have less experience than low- and high-RFI heifers (21 vs. 131 d) in accessing feed in the GrowSafe system. Even though there was a 21-d adaptation period in period 2, before intake data collection, experience in management systems could affect ADG of heifer groups during period 2. For all heifer groups, a strong positive relationship existed between RFI and DMI (r >0.66; P < 0.01). As can be expected, RFI was very weakly or not correlated (r 0.05) with ADG (Table 4), and a moderate and negative relationship (r = 0.4 to 0.6; P < 0.05) existed between RFI and G:F for all groups of heifers (Table 4). The results of the current study agree with other results (Kelly et al., 2010; Durunna et al., 2012) that suggested that cattle with low RFI had similar rates of gain to cattle with high-RFI status. Thus, the ranges in intake, perfor-
Table 2. Phenotypic summary statistics of beef heifer performance (±SD) of heifers fed a forage-based diet during period 1 RFI quintile1 Item n Initial BW, kg Mean Minimum Maximum Final BW, kg Mean Minimum Maximum ADG, kg/d Mean Minimum Maximum DMI, kg/d Mean Minimum Maximum G:F, kg Mean Minimum Maximum RFI, kg/d Mean Minimum Maximum
1 14 261 ± 23.9 215 303 305 ± 32.0 246 366 0.62 ± 0.16 0.38 0.93 6.5 ± 1.07 4.07 8.44 0.096 ± 0.004 0.070 0.110 −1.14 ± 0.42 −2.18 −0.57
2 14
260 ± 20.2 220 287 305 ± 25.8 265 340 0.64 ± 0.12 0.49 0.92 7.5 ± 0.75 6.54 9.12 0.084 ± 0.002 0.070 0.100 −0.28 ± 0.15 −0.51 −0.08
3 14
265 ± 27.7 229 316 311 ± 32.6 264 369 0.65 ± 0.13 0.43 0.88 8.0 ± 0.86 6.38 9.11 0.082 ± 0.003 0.070 0.100 0.11 ± 0.09 −0.02 0.25
4 14
255 ± 19.0 229 289 299 ± 20.4 268 330 0.62 ± 0.10 0.46 0.86 8.0 ± 0.55 7.14 9.2 0.076 ± 0.002 0.060 0.090 0.41 ± 0.14 0.25 0.68
5 14
258 ± 17.4 232 288 304 ± 23.1 275 350 0.64 ± 0.20 0.24 1.09 8.7 ± 1.07 6.88 10.91 0.072 ± 0.005 0.030 0.100 0.98 ± 0.26 0.70 1.52
Overall 70
260 ± 21.6 215 316 305 ± 26.7 246 369 0.64 ± 0.14 0.24 1.09 7.7 ± 1.12 4.07 10.9 0.082 ± 0.014 0.030 0.110 0.00 ± 0.75 −2.18 1.52
Data are divided into 5 groups as categorized by residual feed intake (RFI) ranking (n = 70): quintile 1 = lowest 20% of RFI values; quintile 2 = next 21 to 40% of RFI values; quintile 3 = next 41 to 60% of RFI values; quintile 4 = next 61 to 80% of RFI values; quintile 5 = highest 81 to 100% of RFI values.
1
161
Feed efficiency of beef heifers
Table 3. Performance of a randomly selected population of heifers (control; CON) and heifers selected for low or high residual feed intake (RFI) fed a forage-based diet during period 2 Item n Initial BW, kg Final BW, kg DMI, kg/d ADG, kg/d RFI,2 kg/d Final BCS3 G:F, kg of BW gain/kg of DM
CON
Low RFI1
High RFI1
SEM
P-value
20 325.2 389.9 9.5 0.83 −0.09b 2.65 0.09
20 326.2 398.1 9.5 0.92 −0.33b 2.55 0.10
20 314.6 385.5 10.0 0.91 0.42a 2.58 0.09
5.07 6.09 0.19 0.030 0.134 0.044 0.002
0.21 0.34 0.13 0.08 0.05) and ADG (r = 0.33; P < 0.05) than for RFI, which was similar to the results of Kelly et al. (2010) and Durunna et al. (2011, 2012). This suggests that greater precision in selection for improved feed efficiency may be achieved with using RFI as a criterion than with using G:F and ADG. It is also important to evaluate the reranking of RFI status for heifers from period 1 to period 2, and to understand the characteristics of heifers that had slight (changed by −0.5 SD or changed by >0.5 SD) changes (Table 6). Based on a change of 0.40 kg of DM/d (changed by >|0.5| SD; Table 6), 28% (n = 11) of the heifers changed from high- to low-RFI rankings (by −0.5 SD) and 23% (n = 9) of the heifers changed from a low- to high-RFI rank (by >0.5 SD). Approximately
Table 5. Spearman correlations among residual feed intake (RFI), G:F, and ADG of heifers between 2 feeding periods (period 1, after weaning; period 2, before breeding) Item1 P1-RFI P2-RFI P1-G:F P2-G:F P1-ADG
P2-RFI P1-G:F P2-G:F P1-ADG P2-ADG 0.58*
−0.61* −0.24 −0.20 −0.45† 0.10
−0.13 0.26 0.75* −0.07
0.06 0.03 0.21 0.74* 0.33‡
P1 = period 1; P2 = period 2. *P < 0.001, †P < 0.01, ‡P < 0.05: correlations significantly different from zero.
1
162
Damiran et al.
Table 6. Least squares means for heifers that changed their residual feed intake (RFI) ranking from period 1 to period 2 by >−0.5 SD, +0.5 SD (SD = 0.83 kg of DM/d) Item n Initial BW, kg Final BW, kg ADG, kg/d DMI, kg/d G:F RFI, kg/d a,b
Changed by >−0.5 SD
Changed by +0.5 SD
11 320.6 392.2 0.92 9.7ab 0.09 −0.25b
19 319.9 390.6 0.90 9.5b 0.10 −0.07b
9 319.4 391.8 0.93 10.4a 0.09 0.60a
SEM
7.10 8.43 0.043 0.26 0.004 0.172
P-value
0.10 0.99 0.92 0.06 0.57 0.01
Means within a row with different superscripts differ (P < 0.05).
49% (n = 19) of the heifers maintained their RFI class (by ±0.5 SD). Likewise, heifers that changed by >+0.5 SD had a significantly greater (P < 0.05) change in RFI than heifers that changed by >−0.5 SD or heifers that changed by ±0.5 SD. These data indicate that although there is a moderate correlation, almost 50% of the time, heifers will change RFI rank among consecutive measurement periods. Results of the current study are in agreement with those reported by Durunna et al. (2012), who indicated that reranking exists in heifers despite receiving the same type of diet in 2 different feeding periods and that the reranking may be exaggerated in heifers with an extreme RFI status in each period. Previous research (Durunna et al., 2012) also reported that about 49% of heifers maintained their RFI class, whereas 51% of the heifers had a different RFI class in period 2, implying that diet and feeding period can affect the RFI status of heifers. Similarly, according to Durunna et al. (2011), when steers had their diet switched from a grower to a finisher diet, the Spearman rank correlation was 0.33 between 2 feeding periods. The switch from one RFI classification to another may be attributed to some factors that may limit any ability of the heifer to adjust to a different maturity stage or environment. Guan et al. (2008) reported that the ability of cattle to use feed is associated with the population of rumen microbes. The remixing imposed in period 2 (removal of 30 heifers and introduction of 20 CON heifers) might have changed the social structure within each pen and induced reranking of RFI status due to feeding behavior (Hegarty, 2004). For these reasons, different cattle may perform differently in different periods, thereby affecting the RFI status of an individual. In the current study, low-RFI heifers had 19% lower DMI (6.87 vs. 8.51 kg/d; P < 0.01; data not shown) compared with the high-RFI heifer group in period 1, which was in agreement with Black et al. (2013) and Hafla et al. (2013), who found similar results for growing beef heifers. Even though heifer RFI reranking was affected, the numerical
difference in DMI between low- and high-RFI groups still existed (5%; 9.50 vs. 10.0 kg/d; P > 0.05) during period 2. However, the DMI for low RFI was not different (P > 0.05) from CON, further suggesting that although selection for low RFI may reduce feed intake relative to high RFI, differences between randomly sampled populations are unlikely to occur. An explanation for the lower difference for DMI between the low- and high-RFI treatments was the greater BW gain observed in period 2 than period 1. Others (Hughes and Pitchford, 2004) found that when cattle were fed at maintenance requirements alone, the low RFI line had the advantage in efficiency, but when requirements increased due to BW gain and milk production during gestation and lactation, they became less efficient. In agreement with the current study, in another trial in which RFI was determined on nonlactating, open cows after weaning consuming a pelleted hay–wheat ration, the authors found that low RFI–line cows had only a 4.5% lower DMI compared with their high RFI–line counterparts (Arthur et al., 1999). Those heifers that maintained their RFI classes in both periods, whether efficient or not, could offer a platform for understanding the genetic mechanisms surrounding using RFI as a selection tool. Finally, the reranking reported for RFI in this study may call into question the appropriate time to measure the trait, especially for individuals intended to be used as replacements. Durunna et al. (2011) suggested that feed efficiency evaluations may be more appropriate at an older age when the animals are close to their mature BW; therefore, in the current study, heifer feed efficiency data obtained during period 2 may be more representative than data obtained during period 1, when heifers were younger in age. However, when data were pooled for the entire heifer development period (185 d), the DMI of low-RFI heifers was ~9% less (8.4 vs. 9.28 kg/d; P < 0.01; data not shown) than that of high-RFI heifers. Whereas, the DMI of the CON group (8.7 kg/d; data not shown) was similar (P > 0.05) to the DMI of
163
Feed efficiency of beef heifers
Table 7. Growth and reproductive performance of heifer groups Heifer group1 Item
CON
Low RFI
High RFI
SEM
P-value
Age at breeding, mo Pregnancy rate, % Pregnancy-diagnosis BW,2 kg Pregnancy-diagnosis BCS Calf birth date, Julian date Calf birth BW,3 kg Calving-ease score4 Calving distribution, % of total 1 to 21 d 22 to 42 d 43 to 63 d Calf weaning BW5
14.1 92.9 474 2.8 97 34 1.0 90 10 — 256
14.5 79.5 475 2.8 106 32 1.0 49 44 7 241
14.3 100.0 460 2.7 98 35 1.4 72 28 — 257
0.15 4.7 7.8 0.07 6.24 2.25 0.23 8.8 8.4 2.4 7.6
0.21 0.10 0.33 0.85 0.18 0.58 0.47 0.02 0.16 0.26 0.27
CON = control heifer group; low RFI = low residual feed intake (RFI) group; high RFI = high RFI group. 2 Heifer BW adjusted for conceptus according to NASEM (2016). 3 Calf birth BW adjusted for sex following Beef Improvement Federation (1990) guidelines. 4 Scoring system with a scale of 1 to 5: 1 = no assistance; 2 = easy pull; 3 = mechanical pull; 4 = hard mechanical pull; and 5 = caesarean section. 5 The 205-d adjusted weaning weight. 1
the low- and high-RFI group. This observation suggests that the former explanation must be examined in further research studies.
Heifer Reproductive Performance Reproductive performance of heifers is a critical variable for the profitability of cow-calf producers (Patterson et al., 1992). Although RFI might present an opportunity to reduce feed costs, there have been mixed results regarding the effect of RFI status on reproductive performance (Arthur et al., 2005; Basarab et al., 2011; Loyd et al., 2011; Blair et al., 2013). In the present study, the 3 heifer groups were similar (P > 0.05) for age at breeding, averaging 14.3 mo (SE = 0.1). However, pregnancy rates for CON, low-RFI, and high-RFI heifers were 93, 80, and 100%, respectively (Table 7). There was a tendency (P = 0.10) for the low-RFI heifers to exhibit lower pregnancy rates than CON or high-RFI heifers. There were also fewer (P = 0.02) low-RFI heifers calving in the first cycle compared with CON or high-RFI heifers (Table 7), yet there was no difference (P = 0.26) among heifer groups in the proportion calving in the second and third cycles. It appears that the reduced proportion of low-RFI heifers calving in the first cycle resulted in a numerically greater proportion of heifers calving in the second and third cycles. However, the pregnancy rates and calving distribution rates in the current study should be interpreted with caution given the relatively small number of heifers (n = 20) in each group. Arthur et al. (2005) and Blair et al. (2013) reported no
differences between high- and low-RFI lines for pregnancy rate. In contrast, Randel and Welsh (2013), in a review, stated selection for low RFI results in selection of leaner heifers that reach puberty later and concluded that selection for low RFI may impair reproductive efficiency. This appears to be supported by Arthur et al. (2005), who observed that low-RFI cows calved 8 d later (P = 0.18) than high-RFI cows and the progeny from low-RFI cows calved 5 to 6 d later in the calving season, an effect that was attributed to a delay in first estrus (Loyd et al., 2011). The current study data do not support observed differences in calving date (P = 0.18) based on selection for RFI relative to the unselected population. Moreover, BW and BCS were similar among groups when measured at pregnancy diagnosis (Table 7). Finally, the 205-d adjusted weaning weights (251.2 ± 4.4 kg) of calves born to heifers from the 3 groups did not differ (P = 0.27). In the current study RFI was calculated using models, which included ADG and mid-metabolic BW. Other studies (Basarab et al., 2011) have examined the effects of RFI, when RFI was adjusted for off-test backfat thickness and feeding event frequency, on heifer fertility and productivity. The authors reported (Basarab et al., 2011) that when RFI was adjusted for backfat thickness and feeding behavior, no differences (P > 0.05) were observed in pregnancy rate, calving pattern, and productivity in beef heifers. However, if RFI is not adjusted for body fatness and feeding behavior, then selection for efficient heifers may contribute to reduced pregnancy rates. Therefore, future studies with larger sample sets of animals (should also
164
Damiran et al.
Table 8. Economic analysis of heifer development from weaning to breeding (Can$/heifer per day) Heifer group1 Item Total feed cost2 Labor3 Other3,4 Manure cleaning3 Total cost Total development costs, 185 d
CON
Low RFI
High RFI
SEM
P-value
1.43ab 0.07 0.16 0.03 1.69 312.5ab
1.38b 0.07 0.16 0.03 1.64 304.2b
1.52a 0.07 0.16 0.03 1.78 328.6a
0.027 — — — 0.027 4.904
