Consequences of two or four months of finishing feeding of culled dry dairy cows on carcass characteristics and technological and sensory meat quality

Meat Science 76 (2007) 635–643 www.elsevier.com/locate/meatsci

Consequences of two or four months of Wnishing feeding of culled dry dairy cows on carcass characteristics and technological and sensory meat quality M. Vestergaard a

a,¤

, N.T. Madsen b, H.B. Bligaard b, L. Bredahl c, P.T. Rasmussen d, H.R. Andersen a

Department of Animal Health, Wealth and Nutrition, Danish Institute of Agricultural Sciences, Research Centre Foulum, PO Box 50, Foulum, DK-8830 Tjele, Denmark b Danish Meat Research Institute, DK-4000 Roskilde, Denmark c The Aarhus School of Business, The MAPP Centre, DK-8000 Aarhus, Denmark d Danish Cattle Research Centre, Foulum, DK-8830 Tjele, Denmark Received 17 July 2006; received in revised form 29 December 2006; accepted 2 February 2007

Abstract Finishing feeding was evaluated as a way to improve carcass-, meat- and eating quality of culled dairy cows. In total, 125 Danish Friesian cows were purchased from commercial dairy herds. Cows were culled for various typical reasons at diVerent stages of lactation, were non-pregnant and had milk yield at culling ranging from 1 to 25 kg/d and had LW varying from 330 to 778 kg. Cows were housed in tiestalls and had free access to barley straw and water during a 7-d drying-oV period. Cows were allocated to three equal treatment groups based on parity, LW, BCS, and culling reason. A control group (C) was slaughtered immediately after drying-oV (n D 43), a group (F2) was Wnishing-fed for 63 days (n D 41), and a group (F4) was Wnishing-fed for 126 days (n D 41). In the Wnishing period, cows had free access to a TMR (10.6 MJ ME and 130 g CP per kg of DM). Cows on treatment, F2 and F4 gained 1.16 § 0.05 kg/d in the Wnishing period. Compared with C-cows, F2- and F4-cows had 56 and 97 kg higher carcass weight, 10% and 21% larger Longissimus muscle area, and 14 and 70% more backfat, respectively, at time of slaughter. EUROP conformation scores were 2.2 (C), 3.4 (F2) and 4.4 (F4) and EUROP fat scores were 1.9, 3.0 and 3.7. Finishing feeding increased IMF, improved meat Xavour and colour, and tended to reduce shear force value and improve tenderness and juiciness. The F4 cows also had higher fat trim than C- and F2-cows. Cows were divided into two parity groups (1st parity and older cows). Compared with 1st parity cows, older cows ate 12% more feed, had similar daily gain, were heavier, and had higher BCS and fatness including IMF. The results show that it is possible to dry-oV and Wnish-feed culled dairy cows resulting in larger muscles, increased fatness, improved overall carcass quality and better technological as well as sensory quality characteristics. © 2007 Elsevier Ltd. All rights reserved. Keywords: Cull dairy cows; Finishing feeding; Carcass quality; Eating quality

1. Introduction Most of the beef meat sold in Denmark originates from dairy cows. These cows are typically culled by the dairy farmer and slaughtered without any Wnishing feeding. Thus, *

Corresponding author. Tel.: +45 89 99 15 07; fax: +45 89 99 11 66. E-mail address: [email protected] (M. Vestergaard).

0309-1740/$ - see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.meatsci.2007.02.001

cows representing diVerent breeds, ages, stages of lactation, body condition and fattening levels are sent for slaughter and give rise to large variation in carcass and technological quality characteristics and most likely also in eating quality. Finishing feeding can be practised during lactation and will improve carcass quality of cows and the marginal feed conversion ratio is high due to the concomitant milk production (Liboriussen & Klastrup, 1988). However, the

636

M. Vestergaard et al. / Meat Science 76 (2007) 635–643

present production system at the dairy farm as well as decisions made by the dairy farmer often excludes this as a real possibility. One possible way of improving carcass quality of these cows could be to dry-oV the cows and Wnish-feed them in a specialised beef unit (e.g., multi-site production). The production performance and carcass quality eVects have been studied primarily in beef breeds (Cranwell, Unruh, Brethour, & Simms, 1996; Graham & Price, 1982; Habermann, Luger, Frickh, Zollitsch, & Lettner, 2000; Price & Berg, 1981; Sawyer, Mathis, & Davis, 2004), and the consequences for technological and eating quality characteristics is less well studied, especially in dairy breeds. Some experiments in dry beef cows (Bowling, Smith, Carpenter, Dutson, & Oliver, 1977; Cranwell et al., 1996; Habermann et al., 2000; Matulis, McKeith, Faulkner, Berger, & George, 1987; Miller, Cross, Crouse, & Jenkins, 1987) have found that high-energy Wnishing feeding improved colour characteristics of the meat, increased marbling and IMF, reduced shear force value, improved collagen solubility, and/or sensory-panel tenderness compared with maintenance-fed cows. Another but more practical problem is to identify cows that have a potential for Wnishing feeding and to identify those that have to be slaughtered immediately. These aspects have been dealt with in relation to animal age (Graham & Price, 1982; Malterre, Robelin, Agabriel, & Bordes, 1989; Price & Berg, 1981; Sawyer et al., 2004), body condition and fatness (Gleeson & McCarthy, 1979; Graham & Price, 1982) or various cow characteristics (Van Arendonk, Stokvisch, Korver, & Oldenbroek, 1984). The problem of Wnding good cow characteristics as predictors of Wnishing performance are described elsewhere for the cows used in the present study (Vestergaard, Andersen, Rasmussen, Baymler, & Madsen, 2005), and will not be covered herein. Some preliminary data from this study were presented earlier (Bang, Madsen, Oksama, & Vestergaard, 2000). With the overall purpose of Wnding ways to improve meat and eating quality of beef meat for the domestic market, it was the objective of the present study to investigate the eVects of 2 or 4 months of Wnishing feeding in dry Danish Friesian cull cows on performance, carcass, and meat and eating quality characteristics. 2. Materials and methods The experiment complied with the Danish Ministry of Justice Law no. 382 (June 10, 1987), Act no. 726 (September 9, 1993) concerning experiments with animals and care of experimental animals.

condition score (BCS), herd and culling reason. There was no sorting among the cows purchased, so all cows initiated the experiment. At start of the experiment, cows weighed from 330 to 770 kg and BCS ranged from 1.0 to 4.5 (1 D thin, 2, 3, 4 and 5 D fat). Cows from 1st parity (n D 60) and 2nd to 8th parity (mean D 3.2; n D 65) were equally represented on the three treatments. As planned, there was a large variation in stage of lactation (viz. from 22 to 395 days in milk (DIM)). One group (n D 43) which represented control (C) cows was slaughtered after 7 days of drying oV; a second group (n D 41) was dried oV and Wnishing fed for 2 months (F2); and a third group (n D 41) was dried oV and Wnishing fed for 4 months (F4). It was planned to have a diVerent mean stage of lactation (DIM) in the three treatment groups (i.e., 226 (C), 213 (F2) and 173 (F4) DIM) in order not to get too large diVerence in DIM at time of slaughter among treatment groups. All 129 cows completed the experiment, but four cows did not fulWl the requirements for being included in the data analyses. The exclusion was unrelated to the treatments: pregnancy (F4), calving date could not be identiWed (F2), total discarding at slaughter due to sarcosporids in the muscles (F2), and severe leg disorder at start and during Wnishing with partial condemnation of one rear leg at slaughter (F4). Thus, the Wnal dataset comprised 125 cows. As the complete dry-oV period exceeded 7 days for some cows, the actual realized dry-oV and Wnishing periods were: 11.4 and 57.6 days for F2, and 9.8 and 121.8 days for F4, respectively. Thus the mean drying-oV period for Wnishingfed cows was 10.6 days, with a variation from 3 to 30 days. 2.2. Housing, feeding and live animal recordings Cows were housed in tie-stalls equipped with sawdustbedded rubber mattresses. During dry-oV, cows had free access to barley straw and water only. In the Wnishing period, cows had free access to a total mixed ration (TMR) consisting of soybean meal (11.0%), sugar beet pulp (20.0%), rolled barley (23.5%), sugar beet molasses (22.0%), chopped barley straw (20.0%), and vitamin and mineral mixtures (3.5%). The TMR contained 10.6 MJ ME/kg DM, 0.83 Scandinavian Feed Units (SFU)/kg DM (a measure of NE) and 130 g crude protein/kg DM. These mean Wgures are based on weekly samples frozen and later pooled for the Wnal feed stuV analyses. Live weight, BCS and backfat area (BFA) and Longissimus muscle area (LMA) measured by ultrasound (Aloka) were recorded at the beginning of the experiment (all cows) and again immediately before slaughter (F2 and F4 cows).

2.1. Animals and experimental design In total 129 Danish Friesian cows purchased over two years (65 in the Wrst and 64 in the second year) and coming from nine commercial herds each year were transported to the experimental station and divided into three similar groups of cows based on parity, live weight (LW), body

2.3. Slaughter procedures, slaughter remarks, recordings and carcass quality Cows were loaded and transported tied in the morning on the day of slaughter. The Wrst year cows were slaughtered 35 km and the second year 200 km away from the

M. Vestergaard et al. / Meat Science 76 (2007) 635–643

experimental farm. The Wrst year eight batches of 9, 11, 6, 15, 12, 2, 2, and 6 cows were slaughtered on diVerent dates from May to November and the second year Wve batches of 7, 14, 17, 17 and 7 cows were slaughtered from January to August. Although two diVerent abattoirs were used, the slaughter procedures were standardized and similar both years. The same personnel from the Danish Meat Research Institute inspected, watched, supervised the slaughtering and performed the commercial cutting and rib dissection each year. Five cows received remarks from the veterinary inspection but were otherwise normal; two cows had an abscess in the intestine (F4), one had an abscess in a kidney (F4), one had muscle injury (F2), and one had an abscess in the liver and a ‘foreign body’ (C). Cows were slaughtered within 30 min after arrival to the abattoir. Low voltage electrical stimulation was applied to the carcass within 10 min after stunning. Weight of whole carcass and of caul fat and kidney fat was recorded. The carcass was chilled at an air temperature of 10 °C so that all parts of the carcass were kept above this temperature for the Wrst 12 h. Chilled carcasses were then kept at 3–4 °C. Carcasses were classiWed according to the EUROP system for conformation and fatness and for lean/fat colour by a Danish 5-point scale (1 D light, 2, 3 D normal, 4, 5 D dark). Daily carcass gain was calculated for F2- and F4-cows utilizing F0-cows as an initial slaughter group. The ratio of carcass gain to total daily gain was also calculated to get an estimate of the relative deposition in the carcass vs. noncarcass parts of the body. Two days after slaughter, pH was measured in the Longissimus muscle (LM). Overall pH was 5.75 (SE D 0.008). Half carcasses were cut, and a photograph was taken between the 1st and 2nd lumbar vertebrae to record the cross-sectional area of Longissimus muscle (rib eye area, REA) and the subcutaneous backfat thickness (BT). A semi-commercial boning of the right side carcasses was performed and weights of 14 cuts, fat trim, and bone + tendons were recorded and the proportion of saleable meat estimated (Therkildsen, Vestergaard, Jensen, Andersen, & Sejrsen, 1998) The prime rib section (5th/6th thoracic to 1st/2nd lumbar vertebrae) was separated into muscle (sirloin + residue muscle), fat, and bone + tendons. Samples from the LM taken at the 12th and 13th thoracic vertebrae and at the 1st lumbar vertebrae were selected for the technological analyses. Another sample of LM (2nd to 5th lumbar vertebrae) was used for sensory analyses (see below). All samples were vacuum-packed.

637

2000 after the muscle had been allowed to bloom at 4 °C for 80 min. The caudal portion of LM was vacuum-packed and then frozen at ¡20 °C for later measurement of texture. Shear force values of cooked meat was obtained from a Volodkevich shear attachment on a Karl Frank 81559 apparatus as described earlier (Vestergaard, Sejrsen, & Klastrup, 1994). The mean shear force value of six bites was used for each steak. The remainder of the LM sample was minced and used for the determination of intramuscular fat (IMF) by using the Soxtec HT-H+ method and total pigmentation (mg/g) using the Hornsey method. Samples for sensory analyses were frozen before evaluation. Then samples were thawed at 4 °C for 20–21 h, cut into 23 mm steaks and trimmed of subcutaneous fat (so all three treatment groups looked the same) to reach a Wnal temperature of 10–15 °C. The trimming loss was recorded in the 1st year only. Steaks were roasted on a 160 °C open pan, turned every 2 min, and cooked to an internal temperature of 62– 65 °C (8–9 min total). A trained 8-member panel judged the meat on an unstructured line-scale (0–15), where a higher value equals a higher intensity of the trait (Table 4). 2.5. Statistical analysis Data were analysed as a 2 £ 2 factorial design by Proc GLM/MIXED (SAS Institute Inc., Cary NC) using a model which included the systematic eVects of treatment group (C, F2 and F4), parity (1st and later) and their interaction as well as block (1st and 2nd year of experiment). The initial LW of the cow within parity group was included as a covariant. Within Wrst parity cows the mean LW at day 0 was 509.4 kg and within later parity cows the mean LW at day 0 was 595.4 kg. The covariate adjustment was made because there were small (<10 kg) unplanned diVerences in the mean LW at day 0 between the three treatment groups (Table 1). However, the adjustment made by the covariate was rather small, and for most traits non-signiWcant. There were large diVerences between parity groups, due to diVerences in size of cows, which will only be dealt with brieXy. Only a few parity times treatment interactions were detected, and these were related to some fat measures. Residual correlations (Proc MANOVA of SAS) between various traits were made within the group of control cows or within the group of Wnishing-fed cows after adjusting for parity and block eVects. 3. Results

2.4. Technological and sensory analyses

3.1. Growth, muscle and fat development during Wnishing feeding

Longissimus muscle samples used for sensory and technological analyses were stored at 4 °C until 16 days after slaughter. Then Wnal pH was measured and samples were distributed as follows: a 2 cm mid-section steak of LM was cut for colour measurements. The Hunterlab-colour (L, a¤, and b¤-values) was measured on a Datacolour DataXash

Cows on diVerent treatments were not diVerent in age (P > 0.29), parity (P > 0.10), live weight (P > 0.24), BCS (P > 0.18), LMA (P > 0.52), BFA (P > 0.43), and milk yield (P > 0.55) at start of the experiment. As planned, DIM was shorter (P < 0.05) for Wnishing-fed cows compared with control cows (196 vs. 222 d). However, an earlier analysis

638

M. Vestergaard et al. / Meat Science 76 (2007) 635–643

Table 1 Live weight, daily gain, body condition score (BCS), carcass weight, dressing percentage, carcass gain, EUROP conformation and fatness scores, lean/fat colour and caul- and kidney fat in culled dry dairy cows without (C) or with a 2 (F2) or 4 (F4) month Wnishing perioda Number of cows LW at d 0 (dry), kgb LW at slaughter, kg ADG Wnishing period, kg/d ADG total period, kg/dc BCS (1–5) Carcass weight, kg Dressing, % Net carcass gain, g/dd EUROP conformation EUROP fatness Lean/fat colour Caul fat, kge Kidney–heart–pelvic fat, kgf

C

F2

F4

SEM

P-Value

43 537 552x,y,zg 0g ¡1.82g 2.7g 248g 45.1g – 2.28g 1.99g 3.49g 5.1g 5.3g

41 558 624h 1.16h 0.94h 3.6h 295h 47.3h 754 3.40h 3.02h 3.13h 7.8h 8.3h

41 560 696i 1.16h 1.10h 4.2i 337i 48.5i 713 4.43i 3.69i 3.01h 12.2i 14.5i

10.7 4.2 0.05 0.25 0.07 2.7 0.34 30 0.13 0.10 0.059 0.72 0.49

0.24 0.001 0.001 0.001 0.001 0.001 0.001 0.43 0.001 0.001 0.001 0.001 0.001

a The analyses included 125 observations unless otherwise speciWed in a footnote. Values presented are least squares means with SEM and P-values for the main eVect of treatment from type III F-tests. b The actual mean LW at start, which is included as a covariate in all other analyses as LW at start adjusted within parity group. c Only 123 observations available. d Only data from the 82 Wnishing-fed cows available. e Only 94 observations available. f Only 117 observations available. g–i Means with a diVerent superscript within a row are diVerent (P < 0.05).

revealed that these diVerences in DIM did not aVect the production performance (Vestergaard et al., 2005). In the 7-day drying-oV period, C-cows lost on average 1.8 kg/d (Table 1). The F2- and F4-cows started Wnishing feeding, when they were declared completely dry. For F2and F4-cows, the mean drying-oV period was 10.6 days. However, some Wnishing-fed cows needed 4 days of dryingoV, whereas two cows needed up to 30 days. During Wnishing, the growth performance (ADG) for F2- and F4-cows was on average high (1.16 kg/d in both groups), but there was a large variation among cows. Thus 6 cows gained above 1.8 kg/d and 7 cows gained below 0.6 kg/d. In fact two cows in the F2-group, hardly gained during the Wnishing period due to the long drying-oV period with weight loss. This aVected the feed eYciency of these two cows substantially. The gain to feed ratio (kg ADG:kg DMI) was 0.082 and 0.078 for F2- and F4-Wnishing-fed cows (P > 0.05), respectively. The weight loss during drying-oV aVected the overall daily gain from start to slaughter more for F2- than for F4cows due to a proportionally longer drying-oV period, so the overall ADG was 0.94 kg/d for F2- and 1.10 kg/d for F4cows (Table 1). 3.2. Carcass quality characteristics All carcass quality characteristics were improved by the Wnishing feeding (Table 2). The higher weight of the carcass and higher dressing percentage show that especially the net carcass gain was markedly improved during Wnishing-feeding. Thus, although daily carcass gain was not diVerent between F2- and F4-cows, the carcass gain to total ADG

accounted for almost 80% in F2-cows and only 65% in F4cows. This suggests excessive fat deposition in the non-carcass depots as evidenced by a 139% to 174% higher cauland kidney fat deposition after the long Wnishing period of F4-cows compared with a modest 53–57% increase in F2cows (Table 1). Also carcass fat increased, so the proportion of carcasses classiWed in EUROP fat class 3 or class 4 was only 25% in C-cows and 83% and 93% in F2- and F4cows, respectively. However, also 7% of F4-cows became too fat leading to excessive carcass fat trim at slaughter. The proportional highest carcass gain was evidenced in the forepart (wing) as the hindquarter (pistol) percentage was reduced from 47.7 in C-cows to 46.1 and 45.2 in F2and F4-cows (Table 1). The total carcass composition showed improved carcass yield, markedly reduced bone and increased fat trim proportion (Table 2). The dissection of the prime rib showed a doubling of the proportion of dissectible fat from 12.1% in C-cows and up to 25.7% in F4cows. With Wnishing feeding, the LMA and BFA increased by 11–23% and 14–70%, respectively. 3.3. Technological and sensory quality characteristics The changes in ultimate pH and pigment content could probably not explain the slight improvement in meat colour characteristics of meat from Wnishing-fed cows (Table 3). Both the visual marbling and the analysed fat content (IMF) showed a 42% and 112% higher IMF in F2- and F4cows, respectively. The shear force value tended (P < 0.06) to be reduced but only by 4–13% with Wnishing feeding. However, the shear force value of all treatment groups was low and indicated an overall good tenderness in these cows.

M. Vestergaard et al. / Meat Science 76 (2007) 635–643

639

Table 2 Carcass quality characteristics and in vivo measurements of Longissimus muscle and backfat area in control (C) and Wnishing-fed (F2 and F4) culled dry dairy cowsa C

F2 d

Hindquarter (pistol), kg Forequarter (wing), kgb Pistol, % Saleable meat, %c Carcass fat trim, %c Prime rib section, kg Meat, % Fat, % Bone, % LMA, cm2 (ultrasound) REA, cm2 (at slaughter) BFA, cm2 (ultrasound) BT, mm (at slaughter)

F4 e

f

67.8 80.7e 46.1e 74.6e 5.7e 7.92e 59.9e 18.1e 22.0e 54.2e 55.0e 8.0e 5.4e

59.0 66.4d 47.7d 73.3d 4.2d 6.18d 62.1d 12.1d 25.8d 49.5d 49.5d 7.0d 3.7d

76.0 93.6f 45.2f 75.4f 6.9f 9.79f 55.4f 25.6f 18.9f 59.2f 61.2f 11.9e 9.8e

SEM

P-Value

0.59 0.89 0.18 0.21 0.21 0.14 0.55 0.70 0.40 0.97 1.02 0.44 0.62

0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001

a

The analyses include 125 observations unless otherwise speciWed in a footnote. Values presented are least squares means with SEM and P-values for the main eVect of treatment from type III F-tests. b Only 124 observations available. c Only 123 observations available. d–f Means with a diVerent superscript within a row are diVerent (P < 0.05).

Table 3 Meat quality characteristics of Longissimus muscle from control (C) and Wnishing-fed (F2 and F4) culled dry dairy cowsa C

F2 5.55b

pH ultimate Meat colour Lightness (L) Redness (a¤) Yellowness (b¤) Pigmentation, ppm Visual marbling (1–5) Intramuscular fat, % Shear force value, kg

32.6b 20.6b 9.5b 217 2.3b 2.6b 5.93

F4

5.50c 33.9c 21.3c 10.1c 204 2.9c 3.7c 5.68

5.51c 34.3c 21.9c 10.4c 204 3.7d 5.5d 5.13

SEM

P-Value

0.007

0.001

0.30 0.21 0.11 4.6 0.12 0.25 0.23

0.001 0.001 0.001 0.07 0.001 0.001 0.06

a The analyses include 125 observations. Values presented are least squares means with SEM and P-values for the main eVect of treatment from type III F-tests. b–d Means with a diVerent superscript within a row are diVerent (P < 0.05).

Table 4 Sensory quality of steaks from Longissimus muscle from control (C) and Wnishing-fed (F2 and F4) culled dry dairy cowsa Tenderness Bite resistance Chewing time Juiciness Colour (roasted) Meat Xavourb Smell Cooking loss, %b

C

F2

8.5 6.1 7.9 10.2 5.4c 8.5c 8.9 16.0c

9.0 5.8 7.4 10.4 6.0d 9.0d 9.1 17.4d

F4 9.2 5.8 7.3 9.9 6.6e 9.4e 9.2 17.5e

SEM

P-Value

0.27 0.19 0.23 0.15 0.20 0.11 0.11 0.30

0.16 0.41 0.17 0.12 0.001 0.001 0.24 0.001

a The analyses include 124 observations unless otherwise stated in a footnote. Values presented are least squares means with SEM and P-values for the main eVect of treatment from type III F-tests. b Only 123 observations available. c–d Means with a diVerent superscript within a row are diVerent (P < 0.05).

The overall eating quality was at a high level in all three treatment groups. However, the sensory panel detected improved colour and meat Xavour and tendencies for better juiciness, tenderness and reduced chewing time with Wnishing feeding (Table 4).

3.4. EVect of parity The experimental design and the pooling of all older parity cows does not allow for a real parity comparison. Comparing 1st parity cows with older cows, that were at the

640

M. Vestergaard et al. / Meat Science 76 (2007) 635–643

same stage of lactation at start, showed the older cows ate 12% more feed (P < 0.001), but ADG was similar, and net carcass gain tended (P < 0.13) to be lower for older cows. At slaughter, older cows were heavier (665 vs. 583 kg; P < 0.001), had higher BCS (3.6 vs. 3.4; P < 0.04), higher EUROP conformation score (3.5 vs. 3.2; P < 0.02), and higher EUROP fatness score (3.1 vs. 2.7; P < 0.001). Although, LMA was 8% larger and BT was 28% thicker (both P < 0.001) for older cows at slaughter, the estimated changes in LMA and BFA during Wnishing (data not shown) were not aVected by parity. The fat percentage of prime rib (20.3 vs. 16.9%, P < 0.001) as well as IMF in Longissimus muscle (4.5 vs. 3.4%) was higher in older cows, but other meat- and eating quality characteristics were not aVected by parity group.

4. Discussion The diVerence between overall ADG and ADG in the Wnishing period is caused by the inevitable weight loss associated with drying-oV the cows and highlights the importance of minimizing the drying-oV period before Wnishing feeding (Habermann et al., 2000; Sawyer et al., 2004). In the case of a long drying-oV period, a somewhat increased Wnishing period is needed to counterbalance this by utilizing catch-up growth. Furthermore, it is important that all cows respond to the Wnishing ration in terms of a high ADG in order to get a good overall performance. In the present experiment, two F2-cows had a very low growth response (i.e., ADG: 180 g/d) and should have been slaughtered initially in order to improve overall performance. It is, however, diYcult to judge if a cull cow is worth a Wnishing feeding or not, as we could not detect these non-qualiWed cows based on a priori knowledge about LW, parity, stage of lactation, BCS, and milk yield (Vestergaard et al., 2005). There was, however, a tendency for cows being culled for reasons related to udder health (e.g., mastitis) to perform less well than cows culled for other reasons. In agreement with our results, others found that initial fatness did not aVect Wnishing performance (Graham & Price, 1982) or that thin cows performed slightly better (Gleeson & McCarthy, 1979). Performance of these dairy cows (i.e., ADG, DMI and G:F) was similar to that of other dairy cows fed similar rations for similar periods of time (Garnsworthy, Cole, Grantley-Smith, Jones, & Peters, 1986), higher than for dairy cows fed high-silage rations (Drennan, Nicoll, & CaVrey, 1983; Robelin, Agabriel, Malterre, & Bonnemaire, 1990), but generally poorer than observed for Wnishing-fed dry beef cows (Faulkner, McKeith, Berger, Kesler, & Parrett, 1989; Habermann et al., 2000; Matulis et al., 1987; Sawyer et al., 2004). ADG was also higher in lean Holstein cows Wnishing-fed a more energy-dense 70% high moisture corn-based rations for 68 d that gained 1.43 kg/d after being dried-oV for 10 days (Jones & Macleod, 1981). A compensatory growth response after a period of low-energy feeding

could also be responsible for some of the diVerences in level of ADG between diVerent studies. The length of the Wnishing period did not aVect the ADG in the present and another dairy cow study (Drennan et al., 1983), but others have seen decreasing ADG after longer fattening periods, especially with high concentrate feeding (Faulkner et al., 1989; Graham & Price, 1982; Malterre et al., 1989). The feeding regime was chosen not to be aggressive for the rumen function, in order to avoid bloat, rumen acidosis, and other digestive malfunctions and to reduce the risk of liver abscesses. In accordance, no cows suVered from digestive problems and only one of 82 Wnishing-fed cows (1.2%) had a liver abscess compared with 26% in Wnishing-fed and 10% in control cows (Graham & Price, 1982), 50% in fed and 29% in non-fed cows (Price & Berg, 1981), and more than 30% with Wnishing feeding (Sawyer et al., 2004) in other experiments. Other results (Drennan et al., 1983; Habermann et al., 2000) suggest that a less intensive Wnishing feeding compared with more intensive feeding results in similar improvements in performance and carcass characteristics. We saw no reduction in ADG or carcass accretion rates (data not shown) in older compared with younger cows. A sub-division of the group of older cows into 2nd, 3rd, and 4th + older parity cows (Vestergaard et al., 2005) did not show any signiWcant diVerences in ADG or in changes in LMA and BFA among these three parity sub-groups. Furthermore, for the 8 cows of parity 5 or higher in the present study, ADG in the treatment period was 1.15 kg/d which is very similar to the overall ADG of all Wnishing-fed cows. This is in accordance with Wndings in other dairy (Holstein) cows (Jones, 1983; Jones & Macleod, 1981) and beef cows studies (Graham & Price, 1982), but in contrast to Wndings in other beef cow studies (Malterre et al., 1989; Sawyer et al., 2004), where younger cows responded better and more eYciently than older cows. However, these cull beef cows were generally much older than the typical cull dairy cows. LMA and BFA measured by ultrasound were positively correlated before Wnishing feeding (r D 0.36, P < 0.001). Although both LMA and BFA was positively correlated to LW at start (r D 0.42–0.45, P < 0.001) and at slaughter (both r D 0.49, P < 0.001), LMA and BFA were not correlated at time of slaughter. This suggests that muscle development could occur without necessarily stimulating fat deposition, as seen in the Wrst parity cows. A study with beef cows found similar REA but lower BT in older compared with younger cows (Sawyer et al., 2004). Generally this and other experiments support that duration of the Wnishing period as well as age of cows are important for the relative eVect on muscle vs. fat deposition (Garnsworthy et al., 1986; Jones & Macleod, 1981; Malterre et al., 1989; Robelin et al., 1990; Sawyer et al., 2004). In the present study, we saw an interaction eVect on some carcass fat measures, which increased more with Wnishing-feeding in older compared with 1st parity cows. Taken together these Wndings suggest that a some-

M. Vestergaard et al. / Meat Science 76 (2007) 635–643

what shorter Wnishing period is probably needed for older compared with 1st parity dairy cows to avoid excessive fat trim at slaughter. The eVects seen on carcass weight, lean, bone and fatness as well as composition of prime rib are similar to Wndings seen in previous studies (Boleman, Miller, Buyck, Cross, & Savell, 1996; Cranwell et al., 1996; Drennan et al., 1983; Garnsworthy et al., 1986; Graham & Price, 1982; Habermann et al., 2000; Jones & Macleod, 1981; Miller et al., 1987). Furthermore, the increased proportion of forequarter after Wnishing feeding was also seen earlier (Jones, 1983). The lighter, brighter and redder colour of meat from Wnishing-fed cows has also been detected by others (Cranwell et al., 1996; Habermann et al., 2000) as has the general increase in IMF of Longissimus muscle (Matulis et al., 1987). The higher juiciness and higher Xavour intensity, detected in the present study, were also seen in other studies together with a better overall desirability (Boleman et al., 1996; Faulkner et al., 1989; Habermann et al., 2000). Shear force values were slightly improved and eVects on trained-panel tenderness were also positive but non-signiWcant with Wnishing feeding. Others also saw small or nonsigniWcant eVects of Wnishing on tenderness (Boleman et al., 1996; Dryden et al., 1979; Faulkner et al., 1989; Schnell, Belk, Tatum, Miller, & Smith, 1997) although some positive eVects were seen after 28 and 56 d of Wnishing in another study (Cranwell et al., 1996). Generally Wnishing feeding improves shear force values of Longissimus muscle (Boleman et al., 1996; Cranwell et al., 1996; Dryden et al., 1979; Faulkner et al., 1989; Habermann et al., 2000; Matulis et al., 1987) with few exceptions (Malterre et al., 1989). It could be considered surprising that diVerences in such important meat quality characteristics like tenderness were so small between control and Wnishing-fed cows in the present study. At least a dilution eVect of the collagen content due to muscle Wbre hypertrophy was expected to be positive as well as increases in the proportion of soluble collagen (Cranwell et al., 1996; Malterre et al., 1989; Miller et al., 1987; Schnell et al., 1997). However, the low eVect of Wnishing feeding on meat shear force values and tenderness can probably be partly explained by the optimized slaughtering, cooling and ageing procedures applied to all treatment groups in the present study. These procedures ensured the tenderization process related to the contribution from the muscle-Wbres similarly in all treatment groups and led to a general good overall eating quality. It has been reported that diVerences between Wnishing-fed and non-fed cows in shear force values is minimized if electrical stimulation is applied (Boleman et al., 1996) as was also seen in the present study. Besides this fact, other explanations could be that increased protein breakdown, increased calpain- or reduced calpastatin activities likely have contributed to the low shear force values and high tenderness in all treatment groups. Thus, there is evidence that a short fasting period, like the C-cows were exposed to during drying-oV, will increase muscle protein breakdown as seen in fasted lambs (Oddy, Lindsay, Barker, & Northrop, 1987) leading to either increased cal-

641

pain levels and/or reduced calpastatin activity at time of slaughter which in turn leads to higher myoWbril fragmentation index (MFI) and lower shear force values. This was seen in lambs fed at 60 compared with 200% of maintenance level one week prior to slaughter (McDonagh, Fernandez, & Oddy, 1999). Also results from cull beef cows indicated that Wnishing feeding for 28–56 days did not improve tenderness and shear force values compared with cows slaughtered immediately after purchase (Schnell et al., 1997). Feeding at or near maintenance level as seen in the 14 days on feed group by Schnell et al. (1997) or in cows gaining 0.1 kg/d (Miller et al., 1987) seem to be negative for meat tenderness. In the present study, we saw that a high LW loss during the 7-d drying-oV period was correlated to higher tenderness (r D 0.47, P < 0.003) and lower shear force values (r D ¡0.28, P < 0.10) in the group of C-cows (n D 43). To further study if the C-cows actually have better tenderness than non-dry cows, it requires another experiment where some cows slaughtered directly from lactation are compared with fasted cows (drying-oV). It is also known that high compared with moderate feeding level in young growing animals (Therkildsen, Larsen, & Vestergaard, 2002) and high vs. maintenance level in Wnishing-fed cull beef cows (Miller et al., 1987) will reduce shear force values and improve meat tenderness. However, there was no signiWcant correlation between actual ADG during Wnishing-feeding and Wnal tenderness or shear force for the 82 Wnishing-fed cows in the present study (data not shown). So both the short fast in control cows and the increased ADG in the Wnishing-fed cows will probably improve the in vivo muscle protein degradation compared with that of maintenance-fed cows and will in turn favour post mortem protein fragmentation and the tenderization process. Besides diVerences in protein turnover between treatment groups, the increased IMF may be responsible for the slightly improved tenderness with Wnishing feeding as the correlation was positive between IMF and tenderness (r D 0.36, P < 0.001) and negative between IMF and shear force value (r D ¡0.37, P < 0.001). These somewhat weak correlations are similar to results obtained in other studies (Jeremiah, 1996; Wheeler, CundiV, & Koch, 1994). In support of our Wndings with the trained panel, an inhouse test panel of 298 consumers also experienced the meat from the Wnishing-fed cows to be of higher eating quality than that from non-fed cows (Grunert et al., 2002) when evaluating the meat on a 7-point scale for each of eight quality dimensions. However, the diVerences were bigger and thus the consumer-ratings for ‘meat tenderness’, ‘juiciness’, ‘good taste’ and ‘global experience’ signiWcantly and linearly increased with time on Wnishing. The reason for a more convincing positive eVect of Wnishing feeding in the in-house testing is unknown, but we speculate that this could be due to a higher robustness of the high IMF content steaks being able to better tolerate the less well-controlled private cooking conditions. However, it is interesting that the consumers least appreciated steaks from Wnishing-fed cows when looking at pic-

642

M. Vestergaard et al. / Meat Science 76 (2007) 635–643

tures of the raw steaks probably due to the high backfat thickness and IMF content, and they also rated the cooked steaks signiWcantly less lean and less healthy (Grunert et al., 2002). This shows that when buying meat and when eating meat, the consumers could very well rank meat with diVerent fat content diVerently. In the end, the ambivalent eVects of fat on consumers’ quality perception introduce a risk of unfulWlled quality expectations at the moment of consumption, and, hence, prevent consumer demand. This fact together with Wndings that relationships between marbling or fat thickness and palatability traits are generally low (Jeremiah, 1996) makes it diYcult to Wnd the optimum level of fatness. However, with beef carcasses it was also shown that with 7.6 to 10.2 mm fat thickness at 12th to 13th rib more than 90% of the steaks would be acceptable in palatability traits (Jeremiah, 1996; Tatum, Smith, & Carpenter, 1982). For the present experiment, this means that some of the F2-cows should have been fed for another month to reach this level of backfat. 5. Conclusion The results show that it is possible to dry-oV and Wnishfeed cull dairy cows with a total mixed ration resulting in an overall daily gain of 1 kg/day, larger muscles, increased fatness and marbling, improved overall carcass quality, and better technological as well as sensory quality characteristics. It is, however, necessary to estimate the optimum Wnishing period for the individual cows in order not to end up with excessive fat trimming of carcass and steaks. The rather small diVerence, especially in tenderness, is related to the fact that both a short fast during the drying-oV period and a high daily gain most likely increase muscle protein degradation and thereby also improve meat tenderness. Acknowledgements We acknowledge the Wnancial support from the Danish Cattle Federation. The practical skills of H. Baymler and co-workers in managing the experiment and the recordings on the animals at Ammitsbøl Skovgaard experimental station are acknowledged. Danish Crown is acknowledged for the good collaboration with registrations and sampling at the slaughter plants. B. Christensen, M. Baltzer and C. Bejerholm are acknowledged for leading dissection and performing sensory panel analyses. For data-management and data-analysis, the authors wish to thank M. Oksama, C. Middelhede and L.B. Gildbjerg. References Bang, H. G., Madsen, N. T., Oksama, M., & Vestergaard, M. (2000). EVect of feeding period on carcass and meat quality of dry cull cows. In Proceedings of the 46th international congress of meat science and technology (pp. 158–159). Argentina: Buenos Aires. Boleman, S. J., Miller, R. K., Buyck, M. J., Cross, H. R., & Savell, J. W. (1996). InXuence of realimentation of mature cows on maturity, color,

collagen solubility, and sensory characteristics. Journal of Animal Science, 74, 2187–2194. Bowling, R. A., Smith, G. C., Carpenter, Z. L., Dutson, T. R., & Oliver, W. M. (1977). Comparison of forage-Wnished and grain-Wnished beef carcasses. Journal of Animal Science, 45, 209–215. Cranwell, C. D., Unruh, J. A., Brethour, J. R., & Simms, D. D. (1996). InXuence of steroid implants and concentrate feeding on carcass and longissimus muscle sensory and collagen characteristics of cull beef cows. Journal of Animal Science, 74, 1777–1783. Drennan, M. J., Nicoll, G. B., & CaVrey, P. J. (1983). EVects of level of barley, trenbolone acetate and duration of feeding on beef production from cull cows fed silage. Irish Journal of Agricultural Research, 22, 79– 94. Dryden, F. D., Marchello, J. A., Tinsley, A., Martins, C. B., Wooten, R. A., Roubicek, C. B., et al. (1979). Acceptability of selected muscles from poor condition and realimented cull range cows. Journal of Food Science, 44, 1058–1062. Faulkner, D. B., McKeith, F. K., Berger, L. L., Kesler, D. J., & Parrett, D. F. (1989). EVects of testosterone propionate on performance and carcass characteristics of heifers and cows. Journal of Animal Science, 67, 1907–1915. Garnsworthy, P. C., Cole, D. J. A., Grantley-Smith, M., Jones, D. W., & Peters, A. R. (1986). The eVect of feeding period and trenbolone acetate on the potential of culled dairy cows for beef production. Animal Production, 43, 385–390. Gleeson, P. A., & McCarthy, D. (1979). Does it pay to fatten cull cows? Farm and Food Research, 10, 173–175. Graham, W. C., & Price, M. A. (1982). Feedlot performance and carcass composition of cull cows of diVerent ages. Canadian Journal of Animal Science, 62, 845–854. Grunert, K.G., Bredahl., L., & Scholderer, J. (2002). ’Færdigfedning af slagtekøer – slutrapport fra forbrugerundersøgelsen’ (in Danish). Project Paper no. 3-2002. ISSN 0907 2101. Centre for research on customer relations in the food sector. The Aarhus School of Business, pp. 40. Habermann, W., Luger, K., Frickh, J., Zollitsch, W., & Lettner, F. (2000). ’Lohnt sich ausmast von altkühen? Untersuchungen zur fütterungsintensität, XeischbeschaVenheit und wirtschaftlichkeit’ (in German). Die Bodenkultur, 51, 59–69. Jeremiah, L. E. (1996). The inXuence of subcutaneous fat thickness and marbling of beef. Palatability and consumer acceptability. Food Research International, 29, 513–520. Jones, S. D. M. (1983). Tissue growth in young and mature cull Holstein cows fed a high energy diet. Journal of Animal Science, 56, 64–70. Jones, S. D. M., & Macleod, G. K. (1981). The feedlot performance and carcass composition of young and mature cull Holstein cows. Canadian Journal of Animal Science, 61, 593–599. Liboriussen, T., & Klastrup, S. (1988). ‘Fedning af slagtekøer’ (in Danish). Medd. Nr. 715, Statens Husdyrbrugsforsøg, Copenghagen, p. 4. Malterre, C., Robelin, J., Agabriel, J., & Bordes, P. (1989). ‘Engraissement des vaches de réforme de race Limousine’ (in French). INRA Productions Animales, 2, 325–334. Matulis, R. J., McKeith, F. K., Faulkner, D. B., Berger, L. L., & George, P. (1987). Growth and carcass characteristics of cull cows after diVerent times-on-feed. Journal of Animal Science, 65, 669–674. McDonagh, M. B., Fernandez, C., & Oddy, V. H. (1999). Hind-limb metabolism and calpain system activity inXuence post-mortem change in meat quality in lamb. Meat Science, 52, 9–18. Miller, M. F., Cross, H. R., Crouse, J. D., & Jenkins, T. G. (1987). EVect of feed energy intake on collagen characteristics and muscle quality of mature cows. Meat Science, 21, 287–294. Oddy, V. H., Lindsay, D. B., Barker, P. J., & Northrop, A. J. (1987). EVect of insulin on hind-limb and whole-body leucine and protein metabolism in fed and fasted lambs. British Journal of Nutrition, 58, 437–452. Price, M. A., & Berg, R. T. (1981). On the consequences and economics of feeding grain ad libitum to culled beef cows. Canadian Journal of Animal Science, 61, 105–111.

M. Vestergaard et al. / Meat Science 76 (2007) 635–643 Robelin, J., Agabriel, J., Malterre, C., & Bonnemaire, J. (1990). Changes in body composition of mature dry cows of Holstein, Limousin and Charolais breeds during fattening. I. Skeleton, muscle, fatty tissues and oVal. Livestock Production Science, 25, 199–215. Sawyer, J. E., Mathis, C. P., & Davis, B. (2004). EVects of feeding strategy and age on live animal performance, carcass characteristics, and economics of short-term feeding programs for culled beef cows. Journal of Animal Science, 82, 3646–3653. Schnell, T. D., Belk, K. E., Tatum, J. D., Miller, R. K., & Smith, G. C. (1997). Performance, carcass, and palatability traits for cull cows fed high-energy concentrate diets for 0, 14, 28, 42, or 56 days. Journal of Animal Science, 75, 1195–1202. Tatum, J. D., Smith, G. C., & Carpenter, Z. L. (1982). Interrelationships between marbling, subcutaneous fat thickness, and cooked beef palatability. Journal of Animal Science, 54, 777–784. Therkildsen, M., Larsen, L. M., & Vestergaard, M. (2002). InXuence of growth rate and muscle type on muscle Wbre type characteristics, protein synthesis capacity and activity of the calpain system in Friesian calves. Animal Science, 74, 243–251.

643

Therkildsen, M., Vestergaard, M., Jensen, L. R., Andersen, H. R., & Sejrsen, K. (1998). EVect of feeding level, grazing and Wnishing on growth and carcass quality of young Friesian bulls. Acta Agriculturae Scandinavica, Section A, Animal Science, 48, 193–201. Van Arendonk, J. A. M., Stokvisch, P. E., Korver, S., & Oldenbroek, J. K. (1984). Factors determining the carcass value of culled dairy cows. Livestock Production Science, 11, 391–400. Vestergaard, M., Andersen, H. R, Rasmussen, P.T., Baymler, H., & Madsen, N. T. (2005). ‘Kan man se på en malkeko om den er egnet til færdigfedning’ (in Danish), KvægInfo nr. 1526, . Vestergaard, M., Sejrsen, K., & Klastrup, S. (1994). Growth, composition and eating quality of Longissimus dorsi from young bulls fed the adrenergic agonist cimaterol at consecutive developmental stages. Meat Science, 38, 55–66. Wheeler, T. L., CundiV, L. V., & Koch, R. M. (1994). EVect of marbling degree on beef palatability in Bos taurus and Bos indicus cattle. Journal of Animal Science, 72, 3145–3151.