Management of maternal-offspring behavior to improve lamb survival ...

5 downloads 66 Views 292KB Size Report
Oct 26, 2007 - 3Corresponding author: Julie[email protected]. Received ...... Dwyer, C. M., A. B. Lawrence, S. C. Bishop, and M. Lewis. 2003.
Management of maternal-offspring behavior to improve lamb survival in easy care sheep systems J. M. Everett-Hincks and K. G. Dodds J Anim Sci 2008.86:E259-E270. doi: 10.2527/jas.2007-0503 originally published online Oct 26, 2007;

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://jas.fass.org/cgi/content/full/86/14_suppl/E259

www.asas.org

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

Management of maternal-offspring behavior to improve lamb survival in easy care sheep systems1,2 J. M. Everett-Hincks3 and K. G. Dodds AgResearch Limited, Invermay Agricultural Centre, AgResearch, Puddle Alley, Private Bag 50034, Mosgiel, New Zealand

ABSTRACT: This paper examines the environmental and management factors affecting lamb survival on high-performing sheep farms in New Zealand. Improved lambing percentage is the biggest contributor to higher profits on New Zealand sheep farms. Many sheep breeders have selected and bred ewes for increased fecundity over the last 4 decades. The increased proportion of ewes having triplets is of concern to farmers and to industry because neonatal lamb mortality is highest in triplets. The majority of lamb deaths occur in the first 3 d after birth and range from 5 to 30% for individual sheep flocks. The ability of a lamb to survive to weaning is determined by genetics, behavior, physiology, and the environment, including on-farm management practices. We investigated the effects of dam body condition in pregnancy, weather during lambing, lamb

birth weight, and maternal behavior on single, twin, and triplet lamb viability at birth, lamb death risks from dystocia, and starvation exposure and survival through to weaning for 20 industry flocks from 2003 to 2004 (15,821 lambs). Ewes with higher body condition scores in mid pregnancy had heavier lambs at birth (P < 0.01). Lambs weighing 5.5 to 6 kg at birth (P < 0.01) were more likely to be viable at birth and survive to weaning than heavier or lighter lambs. Weather conditions during late pregnancy (P < 0.05) proved more important than conditions during lambing (P < 0.05) in determining lamb viability and survival through to weaning. Older ewes and ewes with triplets require considerably more attention for farmers to realize their production potential. This information can help formulate appropriate management programs to improve lamb survival rates under easy care farming systems.

Key words: birth weight, body condition score, dystocia, heat loss, lamb survival, starvation exposure ©2008 American Society of Animal Science. All rights reserved.

INTRODUCTION Improved lambing percentage is the biggest contributor to higher profits on New Zealand sheep farms. Many sheep breeders have selected and bred ewes for increased fecundity over the last 4 decades. Lamb survival is an important issue in highly fecund sheep flocks. The national mean lambing percentage from 2004 to 2006 was 125%, compared with 100% from 15

1

Financial assistance and support was provided by Ministry of Agriculture and Fisheries Sustainable Farming Fund in 2005 and 2006 and Ovita Ltd. in 2003 and 2004. A very special thank you to all of the farmers that contributed flock records to this study. Many very special thanks to the lamb survival team, in particular S. Duncan, A. Anderson, and J. Kerslake. 2 Presented at the Sheep symposium at the annual meeting of the American Society of Animal Science, San Antonio, TX, July 8 to 12, 2007. 3 Corresponding author: [email protected] Received August 6, 2007. Accepted October 14, 2007.

J. Anim. Sci. 2008. 86(E. Suppl.):E259–E270 doi:10.2527/jas.2007-0503

yr earlier (1990 to 1993; Anonymous, 2006). Davis et al. (1983) reported that as mean litter size increases above 1.7, the decline in single-bearing ewes is offset by an increase in triplet-bearing ewes. The increased proportion of ewes having triplets is of concern to farmers and to industry because lamb mortality is greatest in triplets (Everett-Hincks et al., 2005a,b; Kerslake et al., 2005) and twin- and triplet-born lambs have greater mortality rates than singles (Johnson et al., 1982; Hinch et al., 1983; Scales et al., 1986; Hall et al., 1988). Many studies report lower survival to weaning in lambs weighing less than 3 kg at birth (Hight and Jury, 1970; Dalton et al., 1980; Johnson et al., 1982; Nowak and Lindsay, 1992), but overall, the relationship between lambing rate and lamb survival in highly fecund ewes is poorly understood. This paper provides an investigation into the environmental and management effects on lamb survival and mortality on high performing sheep farms in New Zealand. With this information appropriate animal management programs can be formulated to reduce lamb mortality rates.

E259

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

E260

Everett-Hincks and Dodds

Table 1. Lamb traits recorded between birth and 3 d of age Lamb trait

Abbreviation

Description

Lamb birth weight Lamb viability at birth

LBW LVB

Dystocia death risk factor

LDD

Starvation and exposure death risk factor

LDSE

Lamb survival to 3 d post birth

LS3day

Lamb survival to weaning

LSWWT

Recorded after birth and measured to the nearest 0.1 kg. Lambs with no lung aeration at postmortem were given a score of 0, whereas lambs with partial or full lung aeration were given a score of 1. All lambs alive at birth were also given a score of 1. All surviving lambs to 3 d of age were given a score of 0, whereas those that died between birth and 3 d of age from difficult births (using evidence of moderate to severe localized subcutaneous edema, organ ruptures, or hemorrhages) were given a score of 1. Lambs dying from risk factors other than a difficult birth were also given a score of 0, except that lambs with a death risk factor of starvation exposure were assigned as missing. All surviving lambs to 3 d of age were given a score of 0, whereas those that died between birth and 3 d of age from the starvation exposure complex (diagnosed based on the degree of brown adipose tissue depletion and lack of food in the gastrointestinal tract) were given a score of 1. Lambs dying from risk factors other than starvation and exposure complex were also given a score of 0. Lamb survival to 3 d post birth based on days survived between date of birth and date of death. Lambs that survived to 3 d of age were given a score of 1, whereas those that did not were given a score of 0. Lamb survival to weaning (approximately 100 d of age) based on a record for weaning weight for the lamb; then the lamb was assumed to be alive and given a score of 1. If there was not a weaning weight recorded, the lamb was assumed to be dead and given a score of 0.

MATERIALS AND METHODS Animal Care and Use Committee approval was not obtained for this study because the data were obtained from an existing database. Animal performance records were obtained from the AgResearch Lamb Survival Database and from the national sheep recording database, Sheep Improvement Limited, for 20 flocks (15,821 lambs), from 2 yr of lambing data (2003 to 2004) and including many breeds and breed compositions (predominantly Romney, Coopworth, and Texel) to investigate management and environmental effects for the lamb traits in Table 1. The flocks are in the Otago Southland regions of New Zealand and were performance recorded. All lambs that died between birth and 3 d of age were collected, tagged, weighed, and autopsied to determine the cause of death. The autopsy procedure was modified from that described by McFarlane (1965), in consultation with veterinary practitioners. The autopsy procedure first diagnosed lamb viability at birth based on the presence of lung aeration and then assigned a primary cause of death, which included dystocia (organ rupture, hemorrhage, moderate to severe localized subcutaneous edema on head, neck, brisket, or rib cage), starvation exposure (no brown adipose tissue on heart and kidneys and no food in the gastrointestinal tract), and other causes (infection, congenital abnormality, other). The study by Kerslake et al. (2005) characterized primary dystocia as localized moderate to severe (>3-mm thickness) subcutaneous edema on the lamb’s body at postmortem. Data were edited to remove missing records and low subclass numbers. Ewes 2 to 6 yr of age were included,

and ewes older than 6 yr were grouped into age group 6. Lambs from litters greater than 3 were removed from the data set. Lambs were also removed from the data set if they were fostered, hand-reared, aborted, resulted from embryo transfer, or where their dam was assisted by the shepherd with lambing. Weather conditions were tested in the survival analyses in the form of a sheep heat loss (HL) calculation described by Coronato (1999), which incorporates temperature, wind speed, and precipitation data collected from the nearest weather station. In this study, sheep HL values ranged from a minimum of 10 Wⴢm−2 to a maximum of 108 Wⴢm−2, with a mean of 45 Wⴢm−2. Heat loss variables, calculated for each flock according to date of birth, were average HL 2 wk before birth (HL2wb), average HL 1 wk before birth (HL1wb), average HL on the day of birth (HLdob), and average HL over the 3 d after birth (HL3day). All ewes were scored for body condition (1 = emaciated to 5 = grossly fat, at 0.5 intervals) at mid pregnancy and again 2 wk before lambing (Jeffries, 1961). Ewe body condition score in mid pregnancy (BCSm), and 2 wk before lambing, and the change in body condition from mid to late pregnancy were tested as covariates and fixed effects in the models. A maternal behavior score (MBS), similar to that described by O’Connor et al. (1985), was recorded on dams from 5,691 lambs and 8 flocks and fitted as a class effect for lamb death risk to dystocia (LDD), lamb death risk to starvation exposure (LDSE), lamb survival to 3 d of age (LS3day), and lamb survival to weaning (LSWWT). The MBS was scored on a 5-point scale based on the distance that a ewe retreats from her lambs when the shepherd is tagging them (Table 2).

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

E261

Management to improve lamb survival

Table 2. The maternal behavior scores (MBS) recorded for dams at lamb tagging Description of MBS1 Ewe Ewe Ewe Ewe Ewe

MBS1

flees at the approach of the shepherd, shows no interest in the lambs, and does not return retreats further than 10 m but comes back to her lambs as the shepherd leaves them retreats to such a distance that tag identification is difficult (5 to 10 m) retreats but stays within 5 m stays close to the shepherd during handling of her lambs

1 2 3 4 5

1

O’Connor et al. (1985).

A pedigree file containing the parents and grandparents of each lamb was used to form a relationship matrix. Random effects included a direct additive genetic effect and a maternal genetic effect. Analyses were performed using ASREML (Gilmour, 2006). Fixed effects and covariates were tested and retained in the model if statistically significant at P < 0.05, with the objective of accounting for most of the variation in the trait. Date of birth, lamb sex, and age of dam effects, as well as year of birth by litter size and by flock interactions and birth weight within litter size and dam age interactions, were fitted regardless of significance. The final model for lamb birth weight (n = 15,821) included the fixed effects of flock (P < 0.01), year of birth (P < 0.01), litter size at birth (P < 0.01), flock × year of birth (P < 0.01), flock × litter size (P < 0.01), year of birth × litter size (P < 0.05), flock × year of birth × litter size (P < 0.01), age of dam (P < 0.01), and lamb sex (P < 0.01). Date of birth (P < 0.01) and date of birth within flock × year of birth (P < 0.01), and BCSm (P < 0.01) were fitted as covariates. Statistical analyses showed the effect of lamb birth weight fitted as a quadratic effect for the traits below to be more important and to explain more of the variation than birth weight fitted as a linear effect. The final model for lamb viability at birth (LVB; n = 12,159) included the fixed effects of flock (P < 0.01), year of birth (P < 0.01), litter size at birth (P < 0.01), flock × year of birth (P < 0.01), flock × litter size (P > 0.05), year of birth × litter size (P > 0.05), flock × year of birth × litter size (P < 0.01), age of dam (P > 0.05), and lamb sex (P > 0.05). Date of birth date (P > 0.05) and date of birth within flock × year of birth (P < 0.01), BCSm (P < 0.05), birth weight (P < 0.01), and HL2wb (P < 0.05) were fitted as covariates. The final model for LDD (n = 14,987) included the fixed effects of flock (P < 0.01), year of birth (P < 0.01), litter size at birth (P < 0.01), flock × year of birth (P < 0.01), flock × litter size (P < 0.01), year of birth × litter size (P > 0.05), flock × year of birth × litter size (P < 0.01), age of dam (P < 0.01), and lamb sex (P < 0.01). Date of birth (P > 0.05) and date of birth within flock × year of birth (P < 0.01), birth weight within litter size, age of dam (both P < 0.01), and average HL (HL2wb; P < 0.01 and HL3day; P < 0.01) were fitted as covariates. The final model for lamb death risk to starvation exposure (LDSE; n = 10,511) included the fixed effects of flock (P < 0.01), year of birth (P < 0.01), litter size at

birth (P < 0.01), flock × year of birth (P < 0.01), flock × litter size (P < 0.01), year of birth × litter size (P > 0.05), flock × year of birth × litter size (P < 0.01), age of dam (P < 0.01), and lamb sex (P > 0.05). Date of birth (P > 0.05) and date of birth within flock × year of birth (P < 0.01), birth weight (P < 0.01), birth weight within dam age (P < 0.05), and HL3day (P < 0.05) were fitted as covariates. The final model for LS3day (n = 15,821) included the fixed effects of flock (P < 0.01), year of birth (P < 0.01), litter size at birth (P < 0.01), flock × year of birth (P < 0.01), flock × litter size (P < 0.01), year of birth × litter size (P > 0.05), flock × year of birth × litter size (P < 0.01), age of dam (P < 0.01), and lamb sex (P > 0.05). Date of birth (P > 0.05) and date of birth within flock × year of birth (P < 0.01), birth weight deviation within litter size (P < 0.01), birth weight within age of dam (P < 0.05), and average HL (HL1wb; P < 0.01 and HL3day; P < 0.01) were fitted as covariates. The final model for LSWWT (n = 15,821) included the fixed effects of flock (P < 0.01), year of birth (P < 0.01), litter size at birth (P < 0.01), flock × year of birth (P < 0.01), flock × litter size (P < 0.01), year of birth × litter size (P < 0.05), flock × year of birth × litter size (P < 0.01), age of dam (P < 0.01), and lamb sex (P < 0.05). Date of birth (P < 0.01), date of birth within flock × year of birth (P < 0.01), birth weight within litter size (P < 0.01), birth weight within age of dam (P < 0.05), dam BCSm (P < 0.05), and average HL (HL1wb; P < 0.01 and HL3day; P < 0.05) were fitted as covariates.

RESULTS Table 3 provides a summary of the effects of litter size at birth, lamb sex, and dam age on the lamb traits recorded between birth and 3 d of age and for lamb survival to weaning age.

Lamb Birth Weight Lamb birth weight was lowest for triplet-born lambs, ewe lambs, and lambs born to 2-yr-old dams (Table 3). Twin-born lambs were on average 0.70 kg heavier than triplet-born lambs, and single-born lambs were 1.69 kg heavier than triplet-born lambs. Ram lambs were 0.37 kg heavier than ewe lambs (Table 3). Dam BCSm had a significant effect on lamb birth weight where greater body condition score ewes had heavier lambs at birth (β = 0.11 ± 0.021 kg of birth weight/BCSm; P < 0.01).

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

E262

Everett-Hincks and Dodds

Table 3. Effect of litter size at birth, lamb sex, and dam age relative to the lowest value within each class on lamb traits recorded between birth and 3 d of age and survival to weaning age Trait1 Item Trait mean Litter size at birth Single Twin Triplet P-value Lamb sex Ram Ewe P-value Dam age, yr 2 3 4 5 6 P-value

LBW

LVB

LDD

LDSE

LS3day

LSWWT

4.8

0.98

0.04

0.01

0.91

0.84

1.69 ± 0.048 0.70 ± 0.039 0.00 ± 0.044 **

0.05 ± 0.013 0.07 ± 0.009 0.00 ± 0.011 **

0.03 ± 0.014 0.00 ± 0.011 0.09 ± 0.013 **

0.00 ± 0.012 0.00 ± 0.01 0.01 ± 0.011 **

0.07 ± 0.017 0.08 ± 0.013 0.00 ± 0.016 **

0.13 ± 0.026 0.11 ± 0.018 0.00 ± 0.021 **

0.37 ± 0.037 0.00 ± 0.037 **

0.00 ± 0.008 0.01 ± 0.009 ns

0.02 ± 0.01 0.00 ± 0.01 *

0.00 ± 0.009 0.00 ± 0.009 ns

0.00 ± 0.012 0.02 ± 0.012 ns

0.00 ± 0.017 0.06 ± 0.017 *

± 0.014 ± 0.014 ± 0.014 ± 0.015 ± 0.016 **

0.00 ± 0.023 ne2 0.04 ± 0.019 0.02 ± 0.020 0.02 ± 0.022 **

0.00 0.43 0.51 0.52 0.58

± 0.043 ± 0.041 ± 0.041 ± 0.042 ± 0.046 **

0.01 0.02 0.02 0.00 0.03

± 0.01 ± 0.01 ± 0.01 ± 0.011 ± 0.012 ns

0.01 0.00 0.02 0.01 0.01

± 0.016 ± 0.011 ± 0.017 ± 0.012 ± 0.013 **

0.00 0.02 0.00 0.03 0.01

± 0.01 ± 0.01 ± 0.009 ± 0.011 ± 0.011 **

0.03 0.03 0.04 0.00 0.01

1 LBW = lamb birth weight (kg), LVB = lamb viability at birth (proportion), LDD = lamb death risk from dystocia (proportion), LDSE = lamb death risk from starvation exposure complex (proportion), LS3day = Lamb survival from birth to 3 d of age (proportion), and LSWWT = lamb survival to weaning age (proportion). 2 ne = nonestimable. *P < 0.05, **P < 0.01, ns = not significant at P < 0.05.

Lamb Viability at Birth Lamb viability was lowest for triplets (Table 3). Twin LVB was 7% greater than for triplets and single LVB was 5% greater than for triplets (Table 3). Dam BCSm had a significant effect on LVB where ewes of greater BCSm had less viable lambs than ewes with lower BSCm (β = −0.02 ± 0.005 units of LVB/BCSm; P < 0.05). Lamb birth weight had a significant effect on LVB, where heavier and lighter lambs were less viable at birth than the optimum birth weight of 0.5 kg above the mean (P < 0.01; Figure 1). The HL2wb significantly affected LVB where greater heat loss over this period led to lower lamb viability (β = −0.001 ± 0.0007 units of LVB/Wⴢm−2; P < 0.05).

Lamb Death Risk to Dystocia Dystocia rates were greatest in triplets, ram lambs, and lambs born to 3-yr-old dams (Table 3). Triplet LDD was 9% greater than for twins (Table 3). Lambs of optimum birth weight, that is 0.5 to 1 kg above the overall mean (regardless of litter size), had less death risk to dystocia. Death risk to dystocia was greatest in triplet lambs 2 kg lighter than the mean (Figure 2; P < 0.01). Mean HL during the 2 wk before birth (HL2wb, β = 0.005 ± 0.0008 units of LDD/Wⴢm−2; P < 0.01) and 3 d from birth (HL3day; β = 0.001 ± 0.0003 units of LDD/Wⴢm−2; P < 0.01) increased lamb death risk to dystocia, with HL leading up to birth having the greater impact of the two. Unfavorable dam MBS increased lamb death risk to dystocia (P < 0.01) with greater values for triplet lambs with dams having a MBS lower than 3. Single

born lambs were not affected by the MBS of their dam (Figure 3).

Lamb Death Risk to Starvation Exposure Starvation exposure rates were lowest in single lambs and lambs born to 2-yr-old dams (Table 3). Triplet LDSE was 1% greater than single and twin lambs (Table 3). Lamb death risk to starvation exposure was lowest for lambs born of optimum birth weight (1 kg heavier than the mean) and greatest for lambs born 2 kg lighter than the mean (Figure 4; P < 0.01). Mean HL 3 d from birth (HL3day) significantly affected lamb death risk to starvation exposure for single (β = −0.001 ± 0.0005 units of LDSE/Wⴢm−2), twin (β = 0.0003 ± 0.0002 units of LDSE/Wⴢm−2), and triplet lambs (β = 0.00003 ± 0.0004 units of LDSE/Wⴢm−2) to differing degrees (P < 0.05). Lamb death risk to starvation exposure increased for twin lambs as HL increased 3 d from birth, whereas the reverse relationship was true for single lambs; that is, as HL increased, LDSE decreased for single-born lambs. Dam MBS had a highly significant effect on lamb death risk to starvation exposure (P < 0.01) and was greatest for triplet lambs if their dams had a MBS less than 3 (Figure 5).

Lamb Survival to 3 Days of Age Lamb survival from birth to 3 d of age was lowest for triplet lambs and lambs born to 5 yr old dams (Table 3). Twin LS3day was 8% greater than for triplets (Table 3). The optimum birth weight for survival to 3 d of age

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

Management to improve lamb survival

E263

Figure 1. Quadratic regression of lamb viability at birth (LVB) on lamb birth weight deviation from the mean (LBWdev; ± SE): LVB = −0.009LBWdev2 + 0.0072LBWdev + 0.0503. Mean lamb birth weight = 4.8 kg.

Figure 2. Quadratic regression of lamb death risk to dystocia (LDD) on lamb birth weight deviation from the mean (LBWdev) for single, twin, and triplet lambs, respectively (± SE): LDD = 0.0122LBWdev2 − 0.0288LBWdev + 0.0394; LDD = 0.0243LBWdev2 − 0.0376LBWdev + 0.0135; and LDD = 0.0202LBWdev2 − 0.0435LBWdev + 0.099. Mean lamb birth weight = 4.8 kg. Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

E264

Everett-Hincks and Dodds

Figure 3. Least squares means (± SE) of lamb death risk to dystocia (LDD) by dam maternal behavior score (MBS) for single, twin, and triplet lambs.

Figure 4. Quadratic regression of lamb death risk to starvation exposure (LDSE) on lamb birth weight deviation from the mean (LBWdev; ± SE): LDSE = 0.0141LBWdev2 − 0.0308LBWdev + 0.0168). Mean lamb birth weight = 4.8 kg. Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

E265

Management to improve lamb survival

Figure 5. Least squares means (± SE) of lamb death risk to starvation exposure × dam maternal behavior score for single, twin, and triplet lambs.

was 1 kg above the mean (Figure 6; P < 0.01). Lamb survival to 3 d of age was lowest for triplet lambs born 2 kg lighter than the mean. Mean HL 1 wk before birth (HL1wb, β = −0.003 ± 0.0005 units of LD3day/Wⴢm−2; P < 0.01) had a highly significant effect on LS3day, where greater HL led to lower survival rates and HL1wb had a greater impact on survival than HL3day (β = −0.002 ± 0.0005 units of LS3day/Wⴢm−2; P < 0.01). Dam MBS had a significant effect on lamb LS3day (P < 0.05; Figure 7). Triplet lambs born to dams with MBS lower than 3 had the lowest survival rates to 3 d of age.

Lamb Survival to Weaning Age Lamb survival from birth to weaning age was lowest for triplets, ram lambs, and lambs born to 2-yr-old dams (Table 3). Twin LSWWT was 11% greater than for triplets (Table 3). Lamb survival to weaning age was greatest for lambs born to dams of greater BCSm (β = 0.001 ± 0.009 units of LSWWT/BCSm; P < 0.05). Lambs born of optimum birth weight (1 kg above the mean) had the greatest survival rates to weaning (Figure 8; P < 0.01). Lamb survival to weaning was lowest for triplet lambs born 2 kg lighter than the mean. Mean HL 1 wk before birth (HL1wb) had a highly significant effect on lamb survival to weaning (β = −0.002 ± 0.0007 units of LSWWT/Wⴢm−2; P < 0.01), where greater HL led to lower survival rates and HL1wb had a greater impact on survival than HL3day (β = −0.001

± 0.0007 units of LSWWT/Wⴢm−2; P < 0.05). Dam MBS had a significant effect on lamb survival to weaning for all lambs (P < 0.05; Figure 9).

DISCUSSION The majority of lamb deaths from birth to weaning occur in the first 3 d after birth and range from 5 to 30% for individual sheep flocks (Kerslake et al., 2005). Previous research has shown that under New Zealand conditions starvation/exposure accounts for approximately 30% of newborn lamb losses (McCutcheon et al., 1981). Dalton et al. (1980) reported dystocia rates of 27% in dead single lambs and 17% in dead multiple lambs. A recent study by Kerslake et al. (2005) showed that the predominant cause of death from birth to 3 d of age was dystocia, accounting for 57% of single and 47% of multiple lamb deaths. Haughey (1983) suggested that between 20 to 60% of neonatal lamb deaths pathologically categorized as starvation or exposure are actually consequences of birth stress. Differing results between studies may be due to differences in clinicopathological diagnoses. It has long been recognized by New Zealand sheep farmers who farm outdoors all year round that weather conditions have the largest influence on lamb survival. Alexander (1984) has also reported the importance of weather on newborn lamb survival. Maternal care can minimize the impact of detrimental environmental factors on lamb losses (Poindron et

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

E266

Everett-Hincks and Dodds

Figure 6. Quadratic regression of lamb survival to 3 d of age (LS3day) on lamb birth weight deviation from mean (LBWdev) for single, twin, and triplet lambs, respectively (± SE): LS3day = −0.016LBWdev2 + 0.0598LBWdev + 0.369; LS3day = −0.0386LBWdev2 + 0.077LBWdev + 0.3792; and LS3day = −0.0376LBWdev2 + 0.0763LBWdev + 0.303. Mean lamb birth weight = 4.8 kg.

Figure 7. Least squares means (± SE) of lamb survival to 3 d of age × dam maternal behavior score for single, twin, and triplet lambs. Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

Management to improve lamb survival

E267

Figure 8. Quadratic regression of lamb survival to weaning age (LSWWT) on lamb birth weight deviation from the mean (LBWdev) for single, twin, and triplet lambs, respectively (± SE): LSWWT = −0.0238LBWdev2 + 0.0881LBWdev + 0.4936; LSWWT = −0.047LBWdev2 + 0.0893LBWdev + 0.4743; and LSWWT = −0.0397LBWdev2 + 0.1011LBWdev + 0.3609. Mean lamb birth weight = 4.8 kg.

Figure 9. Least squares means (± SE) of lamb survival to weaning age × dam maternal behavior.

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

E268

Everett-Hincks and Dodds

al., 1984). However, the proper expression of adequate maternal behavior is made more difficult by modern livestock systems, which increase fecundity and thus increase demands upon the mothering ability of ewes (Chenoweth and Landaeta-Hernandez, 1998). Therefore, a better understanding of environmental and management conditions throughout pregnancy and at lambing may allow a reduction in lamb losses seen in highly fecund ewes. Lamb survival is predominantly controlled by the environment (Lopez-Villalobos and Garrick, 1999; Morris et al., 2000; Everett-Hincks et al., 2002, 2005b). This study has shown that weather conditions, measured in the form of sheep HL (Coronato, 1999), have a highly significant effect on newborn lamb survival. Heat loss leading up to lambing had a greater impact on lamb survival than HL recorded on the day of birth or after birth. This finding is in agreement with Coronato (1999) who also suggested that bioclimatic conditions during late pregnancy are at least as important as conditions during lambing in determining the survival of lambs. It is likely that weather conditions in late pregnancy are influencing the energy balance of the ewe when her nutritional requirements are greatest, particularly for ewes with twins (Holmes, 1975) and triplets (EverettHincks et al., 2005a). The energy cost during pregnancy in the ewe is largely met by increased feed intake, except in very late pregnancy when intake may decline. It is likely that ewes with multiples will mobilize maternal tissues to support their energy demands and those of their growing fetuses (Jelbart and Dawe, 1984). However, under severe HL the energy cost is likely not to be met for ewes with larger litters even if there is sufficient feed. Therefore, it is likely that an energy imbalance at this time, exacerbated by environmental HL, is significantly affecting LVB and subsequent lamb survival. The last 4 to 6 wk before lambing are critical, and Scales et al. (1986) showed that when multiple-bearing ewes were offered additional feed in late pregnancy, lamb mortality was reduced, indicating merit in improved prelamb feeding for ewes carrying more than 1 lamb. Scales and coworkers reported that a 10-kg increase in ewe liveweight during the last 6 wk of pregnancy resulted in an increased birth weight of 0.46 kg for singles and 0.52 kg for twins. In a 2002 study with Romney ewes conducted by Everett-Hincks et al. (2005a), twin litter weight at birth was similar for ewes grazing 2-, 4-, and 6-cm sward heights; however, increasing pasture allowance from just 2 to 4 cm for ewes with triplet litters increased litter weight at birth by 2 kg and improved litter survival to tagging by 4%. Birth weight, fitted as a quadratic effect in this study, emphasizes the importance of an optimum birth weight for survival traits where lesser and greater birth weights are not favorable for these traits. The optimum birth weight where lamb death risk to starvation exposure and dystocia were lowest and lamb viability and survival were highest was 0.5 to 1 kg above the mean.

An optimum birth weight of around 5.5 kg for all lamb traits in this study is significantly heavier than the optimum birth weight of 3 kg reported nearly 3 decades earlier by Dalton et al. (1980). Increased birth weight is an advantage to the survival of twin and triplet lambs, indicating that selection for multiple births should be accompanied by selection for increased birth weight. Hight and Jury (1970) reported that lambs of heavier birth weights are better equipped to survive conditions predisposing them to exposure and starvation because they have more energy stored as brown fat reserves and maintain their suckling drive for a greater duration than lighter lambs. Lamb birth weight has been reported as the predominant factor leading to dystocia in single lambs (Fogarty, 1992). However, birth weight is unlikely to be the only reason for high dystocia rates in larger litters. This study showed that lighter lambs had higher dystocia rates. Dystocia can be a consequence of lamb birth weight, sire breed, dam pelvic conformation (Fogarty and Thompson, 1974), malpresentation, maternal overfeeding, or prolonged parturition (Sargison, 1997; Everett-Hincks et al., 2007). In addition, lambs that endure difficult births have trouble maintaining body temperature and have inhibited behaviors in teat searching and suckling (Eales et al., 1982). This can increase the chances of death when subjected to cold stress or undernutrition. The high rate of deaths from dystocia in triplet lambs was unexpected. It may indicate that dystocia as a primary cause of lamb death is underestimated within highly fecund sheep flocks and is particularly prevalent in lighter lambs possibly resulting from prolonged parturition (Everett-Hincks et al., 2007). Inadequate feeding of ewes can result in poor maternal behavior, and poor ewe behavior is an important cause of lamb death (Nowak, 1996; Everett-Hincks et al., 2005a). Dwyer et al. (2003) demonstrated that a moderate reduction in maternal nutrition during late pregnancy resulted in a measurable reduction in the expression of maternal behaviors, in particular MBS at parturition, under intensive conditions indoors. The same authors did not observe a change in neonatal lamb behavior from the effect of maternal undernutrition and concluded that neonatal progress was affected by birth weight. Dam maternal behavior score had a significant effect on lamb death risk to starvation exposure and dystocia where ewes with lower scores had higher lamb mortality rates. Maternal behavior score gives an indication of the strength of ewe-lamb attachment when separated (Everett-Hincks et al., 2005b). Nowak (1996) observed a considerable improvement in twin lamb bonding with their dam and twin lamb survival when the mother remained on the birth site for a minimum of 6 h. If paddock pasture quality and feeding levels are good and the paddock has adequate shelter, then the ewe will stay at the birth site with her lambs for a greater period (Putu et al., 1988; Pollard and Littlejohn, 1999).

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

Management to improve lamb survival

E269

Coronato, F. 1999. Environmental impacts on offspring survival during the lambing period in central Patagonia. Int. J. Biometeorol. 43:113–118. Dalton, D. C., T. W. Knight, and D. L. Johnson. 1980. Lamb survival of sheep breeds on New Zealand hill country. N. Z. J. Agric. Res. 23:167–173. Davis, G. H., R. W. Kelly, J. P. Hanrahan, and R. M. Rohloff. 1983. Distribution of litter size within flocks at different levels of fecundity. Proc. N. Z. Soc. Anim. Prod. 43:25–28. Dwyer, C. M., A. B. Lawrence, S. C. Bishop, and M. Lewis. 2003. Ewe-lamb bonding behaviors at birth are affected by maternal undernutrition in pregnancy. Br. J. Nutr. 89:123–136. Eales, F. A., J. S. Gilmour, R. M. Barlow, and J. Small. 1982. Causes of hypothermia in 89 lambs. Vet. Rec. 110:118–120. Everett-Hincks, J. M., H. T. Blair, K. J. Stafford, N. Lopez-Villalobos, P. R. Kenyon, and S. T. Morris. 2005a.The effect of pasture allowance during pregnancy on maternal behavior and lamb rearing performance in highly fecund ewes. Livest. Prod. Sci. 97:253–266. Everett-Hincks, J. M., K. G. Dodds, and J. I. Kerslake. 2007. Parturition duration and birthing difficulty in twin and triplet lambs. Proc. N. Z. Soc. Anim. Prod. 67:55–60. Everett-Hincks, J. M., N. Lopez-Villalobos, H. T. Blair, and K. J. Stafford. 2002. Genetic variation in maternal behavior score and lamb survival. 7th World Congress on Genetics Applied to Livestock Production, August 19–23, Montpellier, France. Session 14. Behavior and Welfare. Communication No. 14-14. Everett-Hincks, J. M., N. Lopez-Villalobos, H. T. Blair, and K. J. Stafford. 2005b. The effect of maternal behavior score on lamb and litter survival. Livest. Prod. Sci. 93:51–61. Fogarty, N. M. 1992. Breeding for reproductive performance. Pages 226–233 in Reproduction in sheep. D. R. Lindsay and D. T. Pearce, ed. Canberra Australian Academy of Science in conjunction with the Australian Wool Corporation. Fogarty, N. M., and J. M. Thompson. 1974. Relationship between pelvic dimensions, other body measurements and dystocia in Dorset Horn ewes. Aust. Vet. J. 50:502–506. Geenty, K. G. 1997. 200 by 2000. A Guide to Improved Lambing Percentage. Wools of New Zealand and the New Zealand Meat Producers Board., Wellington, New Zealand. A. R. Gilmour, ed. 2006. ASREML. VSN, Australia. Hall, D. G., A. R. Egan, B. M. Bindon, L. R. Piper, R. Nethery, M. A. Hillard, and G. C. Uphill. 1988. Immunoglobin levels as an index of survival of multiple born lambs. Proc. Aust. Soc. Anim. Prod. 17:413. Div. Anim. Prod., CSIRO, Armidale, Australia. Haughey, K. J. 1983. Selective breeding for rearing ability as an aid to improving lamb survival. Aust. Vet. J. 60:361–363. Hight, G. K., and K. E. Jury. 1970. Hill country sheep production: II. Lamb mortality and birthweights in Romney and Border Leicester × Romney flocks. N. Z. J. Agric. Res. 13:735–752. Hinch, G. N., R. W. Kelly, J. L. Owens, and S. F. Crosbie. 1983. Patterns of lamb survival in high fecundity Booroola flocks. Proc. LITERATURE CITED N. Z. Soc. Anim. Prod. 43:29–32. Holmes, R. J. 1975. Relationship of maternal behavior to neonatal Alexander, G. 1984. Constraints to lamb survival. Pages 199–209 in loss in prolific sheep. Proc. N. Z. Vet. Assoc. Sheep Soc. 5:67–75. Reproduction in Sheep. D. R. Lindsay and D. T. Pearce, ed. Aust. Jeffries, B. C. 1961. Body condition scoring and its use in manageAcad. Sci., Canberra, Australia. ment. Tasmanian J. Agric. 32:19. Anonymous. 2006. Annual review of the New Zealand sheep and beef Jelbart, R. A., and S. T. Dawe. 1984. Management strategies for high industry 2005–2006. Meat and Wool Economic Service of New lambing rates. Proc. Aust. Soc. Anim. Prod. 15:75–79. Zealand, Wellington, New Zealand. Johnson, D. L., J. N. Clarke, K. S. Maclean, E. H. Cox, N. C. Amyes, Aspin, M. 1997. Best practice lambing survey and scientific review. and P. V. Rattray. 1982. Level of nutrition of the ewe and lamb New Zealand Meat Res. Dev. Counc., Wellington, New Zealand. survival. Proc. N. Z. Soc. Anim. Prod. 42:13–14. Beilharz, R. G., and B. G. Luxford. 1984. The inheritance of reproducKerslake, J. I., J. M. Everett-Hincks, and A. W. Campbell. 2005. tive performance. Pages 240–242 in Reproduction in Sheep. D. Lamb survival: A new examination of an old problem. Proc. N. R. Lindsay and D. T. Pearce, ed. Aust. Acad. Sci., Canberra, Z. Soc. Anim. Prod. 65:13–18. Australia. Canberra Australian Academy of Science in conjuncLopez-Villalobos, N., and D. J. Garrick. 1999. Genetic parameter tion with the Australian Wool Corporation. estimates for lamb survival in Romney sheep. Proc. N. Z. Soc. Chenoweth, P. J., and J. Landaeta-Hernandez. 1998. Maternal and Anim. Prod. 59:121–124. reproductive behavior of livestock. Pages 145–165 in Genetics McCutcheon, S. N., C. W. Holmes, and M. F. McDonald. 1981. The and Behavior of Domestic Animals. T. Grandin, ed. Academic starvation-exposure syndrome and neonatal lamb mortality. A review. Proc. N. Z. Soc. Anim. Prod. 41:209–217. Press. Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

Studies have attempted to define suitable husbandry practices for larger litter sizes and to encourage ewelamb attachment and subsequently survival immediately after birth. In a New Zealand survey of farmer practices at lambing, Aspin (1997) identified that when selecting paddocks for lambing, priority considerations were pasture cover, shelter, stocking rate, and topography; however, these factors were not defined. Pollard and Littlejohn (1999) investigated sheltering behavior of lambing ewes under extensive conditions and found that types of shelter did not affect lamb productivity in the 1 yr of study, where weather was moderate. They did, however, observe that the least used artificial shelters were near roadways and human activity, reinforcing the ewe’s need to seek isolation. However, for New Zealand conditions the effect of weather on lamb survival is an important source of variability, which could be alleviated by carefully selecting more sheltered north-facing paddocks for ewes in the late stages of pregnancy, not just at lambing and to reconsider prelamb shearing. Ideally, ewes with triplets should be identified as early as possible, separated, and preferentially fed (Geenty, 1997). Ewes with higher BCS at pregnancy scanning had heavier lambs at birth; therefore, it is important to manage ewes to ensure they are a BCS 3 at mid pregnancy and body condition is maintained throughout the remainder of pregnancy. Ewe BCS will likely act as an energy buffer in times of environmental and nutritional stress. Beilharz and Luxford (1984) discussed the limiting effect of the environment on reproductive progress. They proposed that larger litter sizes could only be utilized successfully if the environment was improved. They maintained that even relatively small increases in litter size would need improvement in the environment if deleterious side effects were to be avoided. Our study has confirmed the importance of the maternal environment for the growth and survival of larger litters and that further research is needed to better understand the biochemical mechanisms in the latter stages of gestation.

E270

Everett-Hincks and Dodds

McFarlane, D. 1965. Perinatal lamb losses. 1. An autopsy method for the investigation of perinatal losses. N. Z. Vet. J. 13:116–135. Morris, C. A., S. M. Hickey, and J. N. Clarke. 2000. Genetic and environmental factors affecting lamb survival at birth and through to weaning. N. Z. J. Agric. Res. 43:515–524. Nowak, R. 1996. Neonatal survival: Contributions from behavioural studies in sheep. Appl. Anim. Behav. Sci. 49:61–72. Nowak, R. F., and D. R. Lindsay. 1992. Discrimination of Merino ewes by their newborn lambs: Important for survival? Appl. Anim. Behav. Sci. 34:61–74. O’Connor, C. E., N. P. Jay, A. M. Nicol, and P. R. Beatson. 1985. Ewe maternal behavior score and lamb survival. Proc. N. Z. Soc. Anim. Prod. 45:159–162. Poindron, P., P. Le Neindre, and F. Levy. 1984. Maternal behavior in sheep and its physiological control. Pages 191–199 in Reproduction in Sheep. D. R. Lindsay and D. T. Pearce, ed. Australian

Academy of Science in conjunction with the Australian Wool Corporation, Canberra, Australia. Pollard, J. C., and R. P. Littlejohn. 1999. Shelter for lambing in southern New Zealand. II. Sheltering behavior and effects of productivity. N. Z. J. Agric. Res. 42:171–177. Putu, I. G., P. Poindron, and D. R. Lindsay. 1988. High level of nutrition during late pregnancy improves subsequent maternal behavior of Merino ewes. Proc. Aust. Soc. Anim. Prod. 17:294–297. Sargison, N. D. 1997: Lamb Mortality – Conception to Weaning. Pages 77–89 in Proc. 27th Annu. Seminar. Society of Sheep and Beef Cattle Veterinarians, New Zealand Veterinary Association (NZVA), Wellington, New Zealand. Scales, G. H., R. N. Burton, and R. A. Moss. 1986. Lamb mortality, birth weight and nutrition in late pregnancy. N. Z. J. Agric. Res. 29:75–82.

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.

References

This article cites 27 articles, 1 of which you can access for free at: http://jas.fass.org/cgi/content/full/86/14_suppl/E259#BIBL

Downloaded from jas.fass.org at AgResearch Library (Ruakura) on April 7, 2008. Copyright © 2008 American Society of Animal Science. All rights reserved. For personal use only. No other uses without permission.