Delay implant strategy in calf-fed Holstein steers: growth performance and carcass characteristics

ABSTRACT The influence of live weight (LW) at first implanting on growth performance and carcass characteristics was evaluated in calf-fed Holstein fed a steam-flaked corn-based diet. Treatments were: (1) control (not implanted); (2) first implanted at 267 kg LW; (3) first implanted at 291 kg LW, and (4) first implanted at 321 kg LW. All implanted cattle were re-implanted on d-112 of trial. Both the initial and final implants contained 120 mg of trenbolone acetate and 24 mg of oestradiol. Compared with non-implanted controls, implanting increased (P < .01) overall dry matter intake (DMI, 5.9%), average daily gain (ADG,16.7%), gain efficiency (ADG:DMI; 9.4%), and estimated dietary net energy for maintenance (NEm, 8.6%) and gain (NEg, 9.7%). Increasing LW at first implanting decreased overall DMI (linear effect; P = .01), but did not affect overall ADG (P = .17) or gain efficiency (P = .32). Within the range of 267–321 kg, weight at first implanting did not affect growth-performance or carcass characteristics. Interval growth performance of implanted and non-implanted calves was a predictable function of LW.


Introduction
Calf-fed Holstein steers enter the feedlot at characteristically light weights (115-180 kg), where they are fed for periods typically in excess of 300 days (Zinn et al. 2005;Duff & McMurphy 2007). As with beef breeds, application of growth implants is among the most important management tools for enhancement of ADG, gain efficiency, carcass weights, and longissiumus muscle area (LMA) of Holstein steers (Chester-Jones et al. 1990;Perry et al. 1991;Zinn et al. 1999). However, the optimal weight at first implant application has received limited research attention. Beckett and Algeo (2002) observed that growth performance of calf-fed Holstein steers (initial weight 156 kg) was not negatively impacted by delaying first implant application until calves had been on feed 120 d. The objective of the present study was to evaluate the weight of calves at first implant application on growth-performance and carcass characteristics of calf-fed Holstein steers.

Animals processing, housing, and feeding
Ninety-six calf-fed Holstein steers (264 ± 3 kg) were used in a 224-d feeding trial, to evaluate the effects of weight at first implanting on growth performance, and carcass traits of calffed Holstein steers. Average daily minimum and maximum air temperature and relative humidity during the trial were was 9.5 and 27.6°C, and 42%, respectively. Upon arrival, steers were vaccinated for IBR-PI 3 (TSV-2, Zoetis, Florham Park, NJ), clostridials-haemophilus (Ultrabac 7, Zoetis, Florham Park, NJ), pasteurella hemolytic (One Shot, Zoetis, Florham Park, NJ), treated for internal and external parasites (Dectomax, Zoetis, Florham Park, NJ), injected with 500,000 IU of vitamin A (Vita-Jec A&D 500, RXV Products, Porterville, CA), branded and eartagged. Calves were grouped by weight into four weight blocks of four pens each (six calves per pen). Pens were 50 m 2 with 26.7 m 2 overhead shade, equipped with automatic drinkers, and 4.3 m fence-line feed bunks. Treatments were: (1) control (not implanted); (2) first implanted at 267 kg live weight (LW) (I-267); (3) first implanted at 291 kg LW (I-291); and (4) first implanted at 321 kg LW (I-321). All implanted cattle were re-implanted on d-112 of trial. Both the initial and final implant was Revalor-S (containing 120 mg of trenbolone acetate and 24 mg of oestradiol; Merck & Co. Inc., Millsboro, DE). Composition of the growing-finishing diet is shown in Table 1. Calves were provided ad libitum access to feed and water. Fresh feed was added to the feed bunk twice daily. Measures of LW and hip height (HH, cm) were obtained at 28d intervals. HH was measured from the ileum tuber coxae to the floor of the chute, using a metal ruler with a mobile crossbar. For calculation of growth performance, initial LW is the off-truck arrival weight. Interim and final LW were reduced 4% to account for digestive tract fill. Final weights were adjusted to a constant dressing percentage (final LW = carcass weight/ 0.623). ADG 1.097 (NRC 1984). Net energy content of the diet for maintenance and gain were calculated by assuming a constant maintenance energy (EM, Mcal/d) cost of 0.084W 0.75 (NRC 2000). The NE values of the diets for maintenance and gain were obtained by means of the quadratic formula: , where a = -0.877DMI, b = 0.877EM + 0.41DMI + EG, c = -0.41EM, and NE g = 0.877NE m -0.41.

Carcass data
Hot carcass weights were obtained at time of slaughter. After carcasses chilled for 48 h, the following measurements were obtained: (1) LM area, by direct grid reading of the muscle at the 12th rib; (2) subcutaneous fat over the eye muscle at the 12th rib taken at a location 3/4 the lateral length from the chine bone end (adjusted by eye for unusual fat distribution); (3) kidney, pelvic and heart fat (KPH) as a percentage of HCW; and (4) marbling score (USDA 1965; using 3.0 as minimum slight, 4.0 as minimum small, etc.).

Statistical design and analysis
The trial was analysed as a randomized complete block design, using pen as the experimental unit. Treatments effects were tested using the following orthogonal contrasts: (1) Control (non-implanted) vs. I-267, I-291, and I-321; (2) Linear effect of LW at first implant and (3) Quadratic effect of LW at first implant. Data for WT, and HH traits were analysed using the MIXED procedure of SAS System (SAS Inst. Inc., Cary, NC) as a repeated measures analysis, to allow for heterogeneous variances and correlations among different time intervals on test (Littell et al. 1998). The linear mixed model used for this analysis includes the overall mean, treatment, day, and treatment with day interaction as fixed effects and pen, pen by treatment, and steers within pen by treatment as random components. Estimation was carried out using the method of REML, assuming a variance-covariance structure and correlations over time on the test. The variance-covariance components were subjected to a test of hypothesis using the option COVTEST. The resulting covariance between random parameters was different from zero (P < .01). Therefore, different variance-covariance structures were used in TYPE option of REPEATED statement. Final structure was selected based on values closest to zero for the Akaike's and Schwarz's Bayesian information criteria. Least square means for treatment, days and interaction effects were used in multiple comparison and significance was declared at P < .05, unless otherwise. To test hypothesis about parallel trends over time for the treatments, day effect as linear and quadratic regression was introduced into linear mixed model, looking for statistical evidence of treatment by day interaction. Full models containing quadratic regression were reduced by removing factors that did not contribute significantly to the model. Hypothesis for equal regressions was realized. Also, one prediction equation for WT trait was obtained by inclusion of HH traits plus interactions and determining best fit applying coefficient of determination (R 2 ), MSE and Mallow's coefficient [C(p)] criterions via the STEPWISE option of REG procedure.
Procedures for animal care and management were conducted under protocols approved by the University of California, Animal Use and Care Advisory Committee.

Results and discussion
Treatment effects on growth performance of calf-fed Holstein steers are presented in Table 2. Compared with non-implanted controls, implanting increased (P < .01) overall 224-d DMI (5.9%), ADG (16.7%), gain efficiency (ADG:DMI; 9.4%), and estimated dietary NE m (8.6%) and NE g (9.7%). These results are consistent with previous studies involving calf-fed Holstein steers, wherein implanting improved ADG by 12-18%, and gain efficiency by 7-12% (Chester-Jones et al. 1990;Perry et al. 1991;Zinn et al. 1999). Likewise, Ainslie et al. (1992) observed that compared with non-implanted steers, implanting Holstein steers enhanced estimated NE m and NE g (8.7 and 11.1%, respectively). They imputed this improvement to a 22% reduction in energy requirements for maintenance. Alternatively, the improved apparent dietary NE for implanted steers may be a reflection of the non-nutritional action of implants on composition of gain, enhancing net protein retention, and hence, leaner-than-expected tissue growth for the specified LW and ADG (Reinhardt 2007).
Increasing LW at first implanting decreased overall DMI (linear effect; P = .01), but did not affect overall ADG (P = .17) or gain efficiency (P = .32). Beckett and Algeo (2002) likewise observed that delaying implanting during the initial and growing phase did not influence overall growth-performance of calf-fed Holstein steers. However, very little work has been reported evaluating the limiting initial weight or weight range when the initial implant might be applied before overall growth-performance is impacted. In the present study, growth-performance response was numerically optimal in steers receiving their first implant at an average weight of 291 kg.
The relationship between AW and DMI as affected by dietary treatments is shown in Figure 2. Daily DMI of non-implanted steers was maximal as steers approached an AW of approximately 470 kg. In the case of implanted steers, daily DMI was maximal as steers approached an AW of approximately 548 kg. Thereafter, daily DMI began a decline. As a function of AW, incremental DMI, kg/d can be explained by the equations: Implanted steers DMI, kg/d = 4.026 -0.012223AW +0.0000967AW 2 -0.000000103AW 3 (P , .01, r 2 = .99), Non -implanted steers DMI, kg/d = 19.93 -0.13977AW +0.000437895AW 2 -0.000000408AW 3 (P , .01, r 2 = .98).
As can be seen from the above two equations, average shrunk weight, alone, does a very good job of explaining DMI. However, the practicality of their use is limited as they do not incorporate differences in dietary NE and ADG.
For implanted and non-implanted steers, the ratio of predicted:observed DMI averaged 1.06 ± 0.08 and 0.98 ± 0.06, respectively. With implanted steers, the conventional equation consistently overestimated DMI across the range of LW, whereas with non-implanted steers, DMI was largely underestimated. Although application of this conventional approach has yielded consistent results for calf-fed Holstein steers with respect to overall average observed vs. expected dietary NE and DMI (Ramirez et al. 1998;Zinn et al. 2000;Carrasco et al. 2013), it may be less reliable in estimation of DMI at specific points across the practical LW range (Carrasco et al. 2013), or in comparisons where treatments are expected to affect composition of gain (as is the case with growth implants).
For implanted and non-implanted steers, the ratio of predicted:observed DMI averaged 1.00 ± 0.03 and 1.00 ± 0.03, respectively can be used to estimate all coefficients and exponents in the application of steer calf equation for calf-fed Holsteins.
The relationship between average AW and gain efficiency is shown in Figure 3. As expected, gain efficiency of both nonimplanted and implanted steers decreased with increasing AW. For non-implanted steers, gain efficiency declined linearly (gain feed = 0.335-0.000405 AW, P < .01, r 2 = .89) until steers achieved an average LW of 510 kg. Subsequently, gain efficiency declined more precipitously (0.00228 units/kg increase in LW above 510 kg). In the case of implanted steers, gain efficiency declined linearly across the weight range of the study (gain efficiency = 0.3478-0.000368AW, P < .01, r 2 = .83).  As mentioned previously, overall gain efficiency was not different among implanted groups. This effect was due to a compensating increase in gain efficiency, particularly noticeable during the final 28 days on feed (linear effect, P < .01). It is expected that relative performance response to growth implants declines with time post-implantation (Montgomery et al. 2001). Reinhardt (2007) observed that re-implanting cattle around d 70 during these declines boosted performance response to a level equal to or greater than that of the previous implant. In the present study, treatments I-291 and I-321 were re-implanted at 84 and 56 d following the initial implant.
Consistent with Schlegel et al. (2006), HH was a good predictor of LW in Holstein calves. For non-implanted calves, HH explained 77% of the variation in LW. For calves first implanted at an average weight of 267, 291, and 321 kg, HH explained 82%, 83%, and 84%, respectively, of the variation in LW. The close relationship between LW and LW:HH ratio of nonimplanted and implanted steers is shown in Figure 4. As LW increases, LW:HH ratio linearly increases (LW:HH, kg/cm = 0.405 + 0.00607LW, r 2 = .997). Consistent with Solis et al. (1989), implanting enhances muscularity largely to the extent that final LW of implanted steers is greater than that of nonimplanted steers, allowing for greater LW:HH ratio.
Treatment effects on carcass characteristics are shown in Table 3. LW at first implanting did not affect (P > .10) carcass characteristics of implanted steers. Implanting increased (8.8%, P < .01) carcass weight and LMA (9.2%, P < .01), and tended to decrease KPH fat (14%, P = .08) compared with non-implanted steers. LW at first implanting did not affect (P > .10) dressing percentage, fat thickness, yield grade and quality grade. Increased LMA has been a consistent response in comparisons of implanted vs. non-implanted cattle (Bartle et al. 1992;Samber et al. 1996;Foutz et al. 1997;Hermesmeyer et al. 2000;Roeber et al. 2000). This response is largely attributable to increased carcass weight (Campbell 2005). Consistent with the present study, growth implants have shown little or no effect on fat thickness (Eversole et al. 1989;Wilson et al. 1999), KPH percentage (Bartle et al. 1992) and dressing percentage (Samber et al. 1996).
Reduced marbling scores, resulting in lower quality grades, is a concern when combination implants (trenbolone acetateestrogen implants) are applied (Apple et al. 1991;Bartle et al. 1992), and is exacerbated by re-implant frequency (Zinn et al. 1999). In the present study, marbling score was numerically lower for implanted steers (5.5 vs. 5.0 for non-implanted vs. implanted steers). However, this difference was not detected as statistically significant (P = .17). This effect may be attributable to differences in terminal LW between non-implanted and implanted steer treatments. Perry et al. (1991) observed that implanted Holstein steers required an additional 25-45 kg LW to achieve choice grade compared to non-implanted steers. In the present study the average terminal weight of implanted steers was 49 kg greater than that of non-implanted steers. Consistent with Nour et al. (1983) and Thooney (1987), implanted calf-fed Holstein steers deposited adequate marbling to meet industry standards.

Implications
Application of growth implants in calf-fed Holstein steers has marked positive effects on daily weight gain, gain efficiency, carcass weight and LMA. Within the range of 267-321 kg, weight at first implanting did not affect growth-performance or carcass characteristics. Interval growth performance of implanted and non-implanted calves is a predictable function of LW.

Disclosure statement
No potential conflict of interest was reported by the authors.