Project topics and materials
Get Complete Project Material File(s) Now! »
CHAPTER THREEnFACTORS AFFECTING THE USEFULNESS OF REPRODUCTIVE TRACT SCORING AS A CULLING TOOL ANALYSED BY LONG TERM REPRODUCTIVE PERFORMANCE IN BEEF HEIFERS
Abstract
In a 7-year longitudinal study 292 beef cows in a restricted breeding system were observed from 1-2 d before their first breeding season, when reproductive tract scoring (RTS) was performed, until weaning their 5th calves. The objectives were to determine whether pre-breeding RTS in heifers is a valid tool to predict long-term reproductive performance, and secondly to investigate factors that may influence its predictive ability. Outcomes measured were failure to show oestrus during the first 24 d of the first 50-day AI season, failure to become pregnant during each yearly AI season (reproductive failure), days to calving from the start of each calving season, and years to reproductive failure. The effect of RTS on each outcome was adjusted for year of birth, pre-breeding age, BW and BCS, and for 24-day anoestrus, bull, gestation length, previous days to calving and previous cow efficiency index where applicable. During their first breeding season, heifers with RTS 1 were more likely to be in anoestrus for the first 24 d (OR 6.1, 95% CI 2.2, 16.7), and were also more likely to fail to become pregnant even after adjusting for 24-day anoestrus (OR 3.2, 95% CI 1.2, 8.6), compared to those with RTS 4 or 5. Animals with RTS 1 or 2 were at increased risk of early reproductive failure compared to those with RTS 4 or 5 (HR = 1.4, 95% CI 1.0, 1.9) despite the fact that RTS was not associated with calving rate or days to calving after the second calving season. Although RTS at a threshold of 1 had consistent specificity of 94-95% for both 24-day anoestrus and pregnancy failure, its predictive value was lower in the age cohort with a higher prevalence of anoestrus. Most animals with RTS 1 or 2 that were subsequently detected in oestrus were in early to mid di-oestrus at the time of scoring; repeating RTS on low scoring animals after 7 d may therefore improve specificity. We conclude that RTS is a valid culling tool to improve long-term reproductive success in a seasonal breeding system, by excluding heifers that are likely to fail to become pregnant or likely to calve late in their first calving season. We further conclude that the predictive value of RTS decreases with increasing prevalence of anoestrus and at certain stages of the oestrous cycle, and that RTS may predict pregnancy failure due to causes other than anoestrus.
Keywords: beef cattle; culling; fertility; heifer selection; predictive ability; reproductive tract score
Introduction
Reproductive traits are 10 times more economically important than production traits in beef cows (Wiltbank, 1994). Restricted breeding and calving during the optimal season are key principles in good cow-calf management (Denham et al., 1991, Engelken et al., 1991). Proper management and selection of heifers (using BW, conformation, EBV, reproductive tract score and pelvimetry) before breeding are essential to the success of such systems (Grass et al., 1982, Larson, 2005).
The onset of puberty in heifers is initiated by a decrease in oestradiol receptors in the hypothalamus and pituitary, ending the prepubertal negative feed-back and resulting in the first LH surge and ovulation (Day et al., 1984, Day et al., 1987). This shift occurs at a specific critical BW (as a proportion of adult BW) and critical age which varies amongst animals (Pence et al., 2007). Various factors affect the age at puberty in individuals, and reproductive tract scoring (RTS) provides an indirect measure of pubertal development (Andersen et al., 1991, Pence and BreDahl, 1998, Holm et al., 2009). Weaknesses of RTS include imperfect repeatability, subjectiveness and inconsistent associations with reproductive outcome (Rosenkrans and Hardin, 2003, Holm et al., 2009).
Short term reproductive performance may be predicted by RTS (Andersen et al., 1991, Pence et al., 2007, Holm et al., 2009) and we hypothesised that RTS may predict long-term survival in restricted bred heifers due to its association with pregnancy outcome and days to calving after first breeding, combined with reports that heifers calving early tend to calve early in subsequent seasons and have increased lifetime production (Lesmeister et al., 1973, Pence et al., 2007, Stevenson et al., 2008, Cushman et al., 2013). To the knowledge of the authors, a long-term study of the performance of heifers by RTS category has not been reported.
The objectives of this study were to determine the usefulness of RTS as predictor of long-term reproductive performance, and to investigate factors that may influence its predictive value.
Materials and methods
This was an observational study of 292 uniquely identified Bovelder beef cows born in either 2002 or 2003 (2002 and 2003 cohorts) that were followed from just prior to their first breeding season until they had weaned up to five calves. The farming system and breed type have been described previously (Paterson et al., 1980, Schoeman and Jordaan, 1998, Holm et al., 2008, 2009).
Reproductive tract scoring by transrectal palpation using a 5-point scale was performed on all heifers either 1 or 2 d before the onset of their first breeding season (Andersen et al., 1991). Scores 4 and 5 were combined in the analyses, after assuming that both categories were pubertal at the time of scoring (Stevenson et al., 2008), and were used as the reference category in Cox proportional hazards and logistic regression models. It was further assumed that heifers with RTS 1 or 2 were prepubertal, whereas those with RTS 3 were peripubertal (Stevenson et al., 2008). Body condition score (BCS) was determined at the same time using a 9-point scale (Marston, 2005). For the purpose of regression models and survival analysis, BCS was categorised into 2 categories: BCS ≤ 6 and BCS ≥ 7. Farm management and staff were blinded to RTS and BCS data throughout the study.
Animals with parity 0, 1, 2, and ≥ 3 were managed in separate groups, and a single inseminator was assigned to each group. The breeding season for heifers started on 15 October every year and consisted of 50 d of continuous visual oestrus observation, with once daily AI at 09h00. The breeding season for cows started on 1 November and consisted of 60 d of oestrus observation and AI in a similar way. Inter-oestrus periods of nulliparous heifers ranged from 16 to 24 d (mean 20 d). Days to first oestrus was defined as either the days to first insemination if it resulted in a pregnancy, or the days to the first insemination that was followed by a normal (16 to 24 d) inter-oestrus interval, or if neither of the above occurred it was the days to the last insemination. An animal was defined in oestrus or met-oestrus at the time of RTS if days from RTS to first oestrus ranged from 18 to 24 d. Similarly she was defined in early di-oestrus on the day of RTS if days to first oestrus ranged from 14 to 17 d, mid-cycle for days to oestrus from 9 to 13 d, late di-oestrus for days to oestrus from 5 to 8 d and pro-oestrus if days to first oestrus ranged from 1 to 4 d. If a heifer’s days to first oestrus was more than 24 d, it was assumed that she was not yet cycling on the day of scoring, and this was defined as 24-day anoestrus.
Bulls were placed with cows for a period of 42 d in a multisire system at a maximum ratio of 1:40 cows, starting 5 to 10 d after the end of the AI period. All the bulls used for natural breeding and AI originated from the same herd and were allocated to cows based on growth performance and conformation, while controlling for inbreeding. Semen for AI was collected, processed and stored in a purpose-built facility on the farm. Seventeen AI bulls were allocated to 10 to 30 heifers each, and the ratio decreased to 1 to 10 cows per AI bull by the fifth parity.
Pregnancy diagnoses (PD) were performed by transrectal palpation (Sheldon and Noakes, 2002) between 23 March and 26 April of every year. Artificial insemination records of cows were available to the veterinarian during pregnancy diagnosis to assist in the differentiation between AI and clean-up bull pregnancies. Animals that were not pregnant to the AI season, as well as those that aborted, or that were confirmed pregnant to AI but failed to calve during the calving season, were sold as soon as their status was known.
Data collected during every AI and calving season included the following: bull allocated, first to fourth AI day (numbered from the first day of the AI season), pregnancy diagnosis, abortion and culling dates, calving date, dystocia score, twinning data, calf gender, calf BW at birth and BW of the cow and calf at weaning. Cow efficiency index (CEI) determined at each weaning event was defined as the weaning weight of the calf corrected to an age of 205 d divided by the metabolic weight of the cow at weaning (BW0.75) (Kleiber, 1947).
Days to pregnancy was defined as the days from the start of the AI season to the last insemination for animals that were confirmed pregnant after the end of the breeding season. Gestation length (GL) was defined as the number of days from the last recorded AI until calving. Animals with GL < 266 d were either changed from “calved” to “aborted” if the birth weight of the calf was below 25kg and the calf did not survive, otherwise an earlier conception date was assigned if this was available from the AI records, or else the GL data was removed if neither was possible. For animals with GL > 299 d the pregnancy diagnosis data was changed from pregnant to not pregnant, as we assumed that the cow did not conceive during the AI season, or if the only recorded AI was early in the season and the calving date was too early for a bull pregnancy, the GL data was removed, in which case it was assumed that additional AI’s performed were not recorded.
Days to calving was defined as the number of days from the start of the calving season until each cow calved, and the first day of the calving season was defined as the day on which the first cow calved within each age cohort.
The study was terminated after the fifth intercalving interval had occurred for all remaining cows, which occurred in April 2009 and April 2010 for the 2002 and 2003 cohorts, respectively.
Analytical procedures
Data were analysed using NCSS 2007 (NCSS, Kaysville, UT, USA) and STATA 11.1 (StataCorp, Texas, USA). Independent proportions, means and medians were compared using the Fisher exact test, ANOVA and Kruskal-Wallis ANOVA, respectively. Standard deviations were provided with the means. Pregnancy proportions, RTS, BW and age of heifers were compared between different AI bulls used during the first breeding season.
Reproductive tract score, being the variable of interest, was initially used in univariable models of days to calving and pregnancy outcome for the first to the fifth breeding season, for each heifer cohort as well as for the combined data. The individual effects of other possible covariates were also estimated (pre-breeding age, BW, BCS and GL, and also the preceding season’s days to calving and CEI in the case of the second to fifth calving seasons), whereafter the effect of RTS on the outcome was adjusted for covariates that were significant (P < 0.05) predictors on their own, using multivariable models. Artificial insemination bull was added as a random effect to the logistic regression models of pregnancy failure during the first AI season. The fit of the logistic regression models of pregnancy failure during the first AI season was evaluated using the Hosmer-Lemeshow goodness-of-fit test.
For the Cox regression model of years to reproductive failure, reproductive failure (the event of interest) was defined as a negative pregnancy diagnosis after the limited AI breeding season. Observations for reproductive failure were done once every year on the day of pregnancy diagnosis, and all data from animals that had left the herd since the previous observation (or the start of the study in the case of the first pregnancy diagnosis) were interval censored to the following day of pregnancy diagnosis. Censored data included those from cows that had aborted their previous pregnancy, cows that died or cows that were culled for any other reason. Artificial insemination bull used during the first AI season was added as a shared frailty to the Cox regression model of years to reproductive failure, whereas the proportional hazards assumption of the model was evaluated using Schoenfeld residuals, and by evaluating the log cumulative hazards plot of the curves of the RTS categories. Data from cows that were still in the herd (and confirmed pregnant) at the study termination were right censored to the last observation (Dohoo et al, 2003c). Sensitivity (Se), specificity (Sp) and positive predictive value (PV+) were calculated for the ability of RTS 1, or 1 and 2 combined, to predict either anoestrus or pregnancy failure.
Results
Heifers with low RTS had higher rates of 24-day anoestrus and pregnancy failure when compared to those with higher RTS (Table 3.1). Lower RTS was independently associated with an increased odds of 24-day anoestrus in both years (Table 3.2). None of the other variables was independently associated with 24-day anoestrus even when RTS was removed from the models (P > 0.05), whereas most of the odds ratios presented in Table 3.2 changed by more than 30% when RTS was removed from the models.
Declaration
Dedication
Acknowledgements
SUMMARY
List of figures
List of tables
List of abbreviations
CHAPTER ONE GENERAL INTRODUCTION
1.1. Introduction
1.2. Fertility in female cattle
1.2.1. Antral follicle count
1.2.2. Genotype x environment interaction
1.2.3. The onset of puberty in cattle
1.3. Restricted breeding seasons in beef cattle
1.4. Reproductive tract scoring
1.4.1. Accuracy and precision
1.5. Dystocia in cattle
1.5.1. Factors affecting dystocia in cattle
1.5.2. Factors affecting calf birth weight
1.5.3. Pelvimetry
1.5.4. Relative pelvis size
1.6. Hypothesis and research questions
1.7. Description of the study herd
1.8. Study objectives and outline of the thesis
CHAPTER TWO THE VALUE OF REPRODUCTIVE TRACT SCORING AS A PREDICTOR OF FERTILITY AND PRODUCTION OUTCOMES IN BEEF HEIFERS
2.1. Abstract
2.2. Introduction
2.3. Materials and methods
2.4. Results
2.5. Discussion
2.6. Conclusion
CHAPTER THREE FACTORS AFFECTING THE USEFULNESS OF REPRODUCTIVE TRACT SCORING AS A CULLING TOOL ANALYSED BY LONG TERM REPRODUCTIVE PERFORMANCE IN BEEF HEIFERS
3.1. Abstract
3.2. Introduction
3.3. Materials and methods
3.4. Results
3.5. Discussion
3.6. Conclusions
CHAPTER FOUR A NEW APPLICATION OF PELVIS AREA DATA AS CULLING TOOL TO AID IN THE MANAGEMENT OF DYSTOCIA IN HEIFERS
4.1. Abstract
4.2. Introduction
4.3. Materials and methods
4.4. Results
4.5. Discussion
4.6. Conclusions
CHAPTER FIVE ULTRASONOGRAPHIC REPRODUCTIVE TRACT MEASURES AND PELVIS MEASURES AS PREDICTORS OF PREGNANCY FAILURE AND ANESTRUS IN RESTRICTED BRED BEEF HEIFERS
5.1. Abstract
5.2. Introduction
5.3. Materials and methods
5.4. Results
5.5. Discussion
5.6. Conclusions
CHAPTER SIX GENERAL DISCUSSION
6.1. Introduction
6.2. The value of RTS as pre-breeding management tool
6.3. Accuracy of RTS
6.4. Repeatability of RTS
6.5. The use of relative pelvis size to predict calving ease
6.6. Adding ultrasonography to pre-breeding examination of beef heifers
6.7. Cost effectiveness of pre-breeding examination of beef heifers
6.8. Limitations of the study
6.9. Practical applications of pre-breeding examination of beef heifers
CHAPTER SEVEN CONCLUSIONS, RECOMMENDATIONS AND FURTHER RESEARCH QUESTIONS
7.1. Conclusions
7.2. Recommendations
7.3. Further research questions resulting from this work
LITERATURE CITED
GET THE COMPLETE PROJECT
