J. Dairy Sci. 2007. 90:2797-2803. doi:10.3168/jds.2006-504
© 2007 American Dairy Science Association ®
Effect of Stillbirths on Dam Survival and Reproduction Performance in Holstein Dairy Cows
R. C. Bicalho*,1,
K. N. Galvão
,
S. H. Cheong,
R. O. Gilbert,
L. D. Warnick* and
C. L. Guard*
* Department of Population Medicine and Diagnostic Sciences, and
Department of Clinical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853
1 Corresponding author: rcb28{at}cornell.edu
 |
ABSTRACT
|
|---|
The objectives of this study were to evaluate the effect of stillbirth on survival and reproductive performance of lactating dairy cows. Data were collected from 2 different regions of the US calving-ease scores (CES) were recorded by farm personnel on a scale of 1 (no problem) to 5 (extreme difficulty). Stillbirths were recorded by farm personnel. The final analysis included 13,608 calvings of which 93.4% were live calves and 6.6% stillbirths. An increasing or decreasing trend in the incidence of stillbirth by parity and by CES was analyzed by Cochran-Armitage trend tests. A significant decreasing trend in the incidence of stillbirth by parity group was detected. The incidence of stillbirth increased as the CES increased. The incidence of stillbirths was 3.6, 11.2, 25.9, and 60.1% for CES score 1, 2, 3, and 4, respectively. Dam survival in the herd and reproductive performance were analyzed by the Cox proportional hazards model. Variables that decreased dam survival time were stillbirths, primiparity, and CES of 3 and 4. The variables that reduced reproductive performance were stillbirths, multiparity, male calves, and CES of 3 and 4. Cows that had stillbirths had significantly increased risk of culling/death throughout the lactation and increased median days open by 88 d compared with cows that had live calves. In conclusion, losses from stillbirths are far greater than just the value of the calf.
Key Words: stillbirth • culling • reproduction • dairy cow
|
INTRODUCTION
|
|---|
Stillbirth is defined as death of a calf that occurs just prior to, during, or within 24 to 48 h of parturition (Philipsson et al., 1979). Economic losses to the dairy industry in the United States just due to the loss of replacement animals were estimated greater than $125 million per year (Meyer et al., 2001). Furthermore, the economic loss due to stillbirths increased by $75.9 million from 1985 to 1996 due to the increase in incidence of stillbirths from 9.5 to 13.2% in primiparous and 5.0 to 6.6% in multiparous cows.
The economic loss is even greater if the detrimental effects of stillbirth on the dam’s lactation performance are considered. Stillbirths were correlated with increased risk of developing metritis and retained placenta (Correa et al., 1993; Emanuelson et al., 1993), increased risk of a primiparous cow being a repeat breeder (not conceiving after second service; Moss et al., 2002), and decreased risk of conception compared with the rest of the population (Maizon et al., 2004). Mangurkar et al. (1984), analyzing a large data set (35,773 records) from the Quebec Dairy Herd Analysis Service, reported that calvings associated with dead calves lead to higher culling rates due to lower milk production and impaired reproductive performance when compared with live calvings.
The objective of the present study was to evaluate the effect of stillbirths on the dam’s subsequent survival and reproductive performance using nonparametric survival analyses.
|
MATERIALS AND METHODS
|
|---|
Farms and Management
Calvings from 7 farms from 2 regions in the United States were used. The farms that participated in the study were enrolled by convenience; databases were donated to the author by farm owner or farm veterinarians. The criteria for selection were 1) use of Dairy Comp 305 (Valley Agricultural Software, Tulare, CA) as the dairy farm record database; 2) the farm had to follow the 5 scale default calving ease scoring (CES) system provided by Dairy Comp 305; and 3) the farm had to have good calf data records, defined as farms with <10% missing calf information (dead or alive, male or female, single or twin calving). Data from 12 dairy farms that originally offered their records were included for the research; however, 5 farms were excluded because of poor records on calving ease. The majority of the farms were from upstate New York, where 6 farms were enrolled, ranging from 350 to 3,000 lactating Holstein cows housed in 6-row free stall barns with mattress stalls. The rolling herd average of the farms was about 12,000 kg of milk. The second region was the Midwest, represented by 1 farm consisting of 4,000 lactating Holstein cows housed in 4-row free stall facility with deep stalls bedded with sand. The rolling herd average was 11,800 kg of milk.
Lactating cow nutrition varied between farms and the 2 regions where the farms were located. All farms fed a TMR that was formulated to meet or exceed the NRC nutrient requirements for lactating Holstein cows weighing 650 kg and producing 45 kg of 3.5% FCM.
Study Design and Data Collection
A retrospective observational study design was used. Data were collected from the dairy records database, Dairy Comp 305, from September 2004 until December 2005. Cows were divided into 2 categories according to the status of the newborn calf: cows that delivered a live calf and cows that delivered a stillborn calf. Stillbirth was defined as death of a calf that occurs just prior to, during, or within 48 h of parturition. To exclude abortions, all cows that had calves born dead before 270 d of gestation were excluded from the data analysis.
Cow survival was measured as days from calving to culling or death. Cows were censored for survival if the cow survived beyond 305 d after calving, were dried off before 305 d after calving, or were still alive at the end of data collection. Reproductive performance was measured as the calving-to-conception interval. Cows were censored for reproductive performance if pregnancy was not diagnosed before culling or death, if not pregnant to an insemination in the first 305 d after calving, or the end of data collection.
Additional information was collected which included parity, season of calving, sex of calf, and CES. A 5-point scale CES system was used by all farms enrolled in the study. A CES of 1 was defined as calvings that occurred easily without assistance. A CES of 2 was defined as unassisted calvings, but with more difficulty for the cow compared with CES 1. A CES of 3 was defined as calvings requiring easy assistance from a person. A CES of 4 was defined as vaginally delivered calvings requiring the calf position to be corrected or hard traction to be applied to deliver the calf. A CES of 5 was defined as calvings requiring fetotomy or caesarean section. There were a total of 9 cows in the 7 farms that had a CES of 5, and these cows were collapsed into the CES 4 group.
Statistical Analyses
An increasing or decreasing trend in the incidence of stillbirth by parity and by CES was analyzed by the Cochran-Armitage trend test using the FREQ procedure (SAS Inst. Inc., Cary, NC).
To facilitate analyses and results interpretation, the variables CES, parity, and season of calving were dichotomized before they were used in all multivariable analyses. A new variable calving ease group (CEG) was created to dichotomize CES to unassisted calvings (CEG = 0), which were CES 1 and 2; and assisted calvings (CEG = 1), which were CES 3 and 4. A parity group (PG) variable was created to dichotomize parity to primiparous (PG = 0) and multiparous (PG = 1) cows. Season of calving (SC) variable was created to dichotomize the season of calving to warm months (SC = 0) between April and September, and cold months (SC = 1) between October and March.
The effect of stillbirth on cow survival was analyzed by the Cox proportional hazards model using the PHREG procedure of SAS. The time variable used in the model was days from calving until culled, death, or censored. A censoring variable was used to identify animals that had been sold or died (censored = 1) from the animals that were alive at the end of data collection (censored = 0). The variables used in this model were: stillbirth, calf sex, parity, CES, and season of calving. The variable of farm was used in the model as a strata variable to adjust for the effect of location. Backward variable elimination (Cantor, 1997) considering main effects and 2-way interactions with offspring sex group and all the other independent variables was performed. Variables were retained in the model when their main effect or their interaction with the variable stillbirth was significant (
0.05).
Effect of stillbirth on days open was analyzed by the Cox proportional hazards model using the PHREG procedure of SAS. The time series variable for this model was the calving-to-conception interval or days from calving to censure. A censoring variable was used to identify animals that had been sold or died (censor = 1) from the animals that were alive at the end of data collection (censor = 0). The variables used in this model were stillbirth, calf sex, parity, CEG, and season of calving. The variable of farm was used in the model as a strata variable to adjust for the effect of location. The variable selection criteria and the interaction tests used for this model were the same as cited above for the dam’s survival model. The outcomes of the Cox proportional hazards model were created for this analysis to identify animals that were confirmed pregnant (censored = 1) from animals that finished the data collection period or end of data collection criteria (censored = 0). Analyses are presented as hazard ratio (Spruance et al., 2004). Survival plots were generated by Kaplan-Meier survival analysis performed using GraphPad Prism version 4.00 for Windows (GraphPad Software, San Diego, CA).
The effect of PG, calf sex, and CEG on the incidence of stillbirth was analyzed by logistic regression using the LOGISTIC procedure of SAS. Backward stepwise variable selection considering main effects and 2-way interactions was used. Independent variables were retained in the model when their main effect or their interaction was significant ( 0.05).
|
RESULTS
|
|---|
Descriptive Statistics
A total of 13,608 calvings was used for the final analyses where 93.4% of the calvings were live calves and 6.6% of the calvings were stillbirths. A decreasing trend (P < 0.001) in the incidence of stillbirth by parity was detected by the Cochran-Armitage trend test (Figure 1
). The incidence of stillbirths for first-parity cows was 10.7%, 3.6% for second-parity cows, 5% for third parity animals, and 4.2% for parity 
4. The incidence of stillbirth by farm ranged from 4.1% (of 3,826) to 10% (of 470).
View larger version (7K):
[in this window]
[in a new window]
|
Figure 1. Stillbirth incidence x parity in Holstein cows. Cows in parity 4 were collapsed into the >3 parity group. A decreasing trend on the incidence of stillbirth was detected as parity increased (P < 0.001). Numbers above the bars represent the total number of animals in that parity grouping.
|
|
An increasing trend (P < 0.001) in the incidence of stillbirth by CES was detected by the Cochran-Armitage trend test. The incidence of stillbirths was 3.6, 12, 25.9, and 60.1% for CES 1, 2, 3, and 4, respectively (Figure 2).
View larger version (7K):
[in this window]
[in a new window]
|
Figure 2. Stillbirth incidence by calving ease score in Holstein cows. The incidence of stillbirths increased as calving ease score increased (P < 0.001). Numbers above the bars represent the total number of animals in that calving ease score.
|
|
Effect of Parity Group, Calf Sex, and Calving Ease Group on the Incidence of Stillbirth
A logistic regression model was used for this analysis. The variables retained in the model by the variable selection method were PG, calf sex, and CEG. The incidence of stillbirth was 10.7 and 4%, respectively, for primiparous and multiparous cows. The adjusted odds ratio (AOR) for stillbirth was 2.56 times higher (P < 0.001) for primiparous cows than multiparous cows (Table 1). The AOR for stillbirth was 0.77 for cows that gave birth to a female calf vs. the cows that had a male calf, which means that cows that had female calves were at a 23% (P < 0.001) decreased odds of having a stillborn calf when compared with the cows that gave birth to male calves. The CEG was highly significant, and the AOR for stillbirth was 0.12; the odds of stillbirths were 88% lower for unassisted calvings (P < 0.001) compared with the assisted calvings (Table 1).
Effect of Stillbirths on Dam’s Survival
A total of 13,608 cows were enrolled for this analysis, of which 2,146 were culled or died, and 11,462 were censored. A complete summary of the data censoring by farm is in Table 2.
View this table:
[in this window]
[in a new window]
|
Table 2. Cox proportional hazards summary of the number of events (culling or death) and number of censored cows by farm for Holstein cows
|
|
The effect of stillbirth on the dam’s survival was analyzed by the Cox proportional hazards model. The variables retained in the model by the backward variable selection method were stillbirth, PG, and CEG. No variable was forced into the statistical model. Of all cows that gave birth to live calves, 15.4% (of 12,712) died or were culled, whereas 20.8% (of 896) of cows that had stillbirths were culled or died. The hazard ratio of death/cull was 1.41 (P < 0.001) for cows that had stillbirths vs. cows that had live calves; therefore the hazard rate of death/cull was 41% higher for cows that had stillbirths. The PG and CEG significantly affected cow survival (Table 3).
A survival plot was generated using Kaplan-Meier survival analysis to better illustrate the difference in cow survival between cows that had stillbirths and cows that had live calves (Figure 3). The mean survival time was 255 and 270 d (P < 0.001) for cows had stillborn or live calves, respectively.
View larger version (9K):
[in this window]
[in a new window]
|
Figure 3. Kaplan-Meier survival analysis plot for Holstein dam survival. Cows that had stillbirths (solid line) had higher culling rate compared with cows that had live calves (dotted line).
|
|
Effects of Stillbirths on Dam’s Reproductive Performance
A total of 13,608 cows were enrolled for this analysis, of which 5,995 cows were diagnosed pregnant, and 7,613 were censored (Table 4). The effect of stillbirth on dam’s reproductive performance was analyzed by the Cox proportional hazards model. The variables retained in the model were stillbirth, calf sex, PG, CEG, and SC. No variables were forced into the statistical model. For the cows that gave birth to live calves, 44.7% (of 12,712) were diagnosed pregnant; and for the cows that had stillborn calves, 35.3% (of 896) were diagnosed pregnant by the end of data collection. The hazard ratio for time to conception was 0.76 (P < 0.001) for cows that had stillbirths vs. cows that had live calves; therefore, the hazard rate of becoming pregnant was 24% lower for the cows that had stillbirth compared with cows that had live calves. The hazard ratio for time to conception was 0.93 (P < 0.001), 0.90 (P < 0.001), 0.84 (P = 0.005), and 0.76 (P < 0.001) for male vs. female calves, multiparous vs. primiparous cows, assisted calving vs. unassisted calving, and cold season vs. warm season, respectively (Table 5).
View this table:
[in this window]
[in a new window]
|
Table 4. Cox proportional hazards summary of the number of events (pregnancies) and number of censored cows by farm
|
|
A Kaplan-Meier survival analysis was performed to illustrate the difference in days open between cows that had a stillbirth and cows that had a live calf. The mean days open was 186 and 212 d (P < 0.001) for cows that had live calves and cows that had stillbirths, respectively. The median calving to conception interval was 170 and 258 d (P < 0.001) for cows with live calves and cows with stillborn calves, respectively (Figure 4).
View larger version (9K):
[in this window]
[in a new window]
|
Figure 4. Kaplan-Meier survival analysis plot for dam’s calving to conception interval. Cows that had live calves (dotted line) had a median calving to conception interval of 170 d compared with 258 d for cows that had stillbirths (solid line).
|
|
|
DISCUSSION
|
|---|
Incidence of stillbirth in the population of Holstein dairy cows is increasing significantly (Hansen et al., 2004). Meyer et al. (2001) reported that the incidence of stillbirths in primiparous cows in 1985 was 9.5% and had increased by 4% in 1996, whereas for multiparous cows the incidence increased over 1%. In our study, stillbirths represented 6.6% of all calvings. Primiparous cows had an incidence of stillbirth of 10.7%, and multiparous cows had 4% incidence of stillbirth. We excluded all twin calvings, including those with 1 or both calves being stillborn. Therefore, the true incidence of stillbirths could be higher if twin calvings were taken into account.
It is likely that many factors play a role in the occurrence of stillbirths. Efforts to identify risk factors have been made over the years. Hansen et al. (2004) reported an unfavorable trend of direct and maternal genetic effects for stillbirths in the Danish Holstein population. In that study, the use of Holstein Friesian sires was the main cause for the increased stillbirth incidence. Our study was restricted to Holstein dairy farms located in the United States. The incidence of stillbirths by farm varied from 4.2 to 10.0%, suggesting that management practice differences among farms may influence stillbirth rates.
First-parity animals were at highest risk of having stillbirths. Hansen et al. (2004) cited that the higher incidence of stillbirth in first-parity cows was due to the disproportion between the size of the calf and the dam’s pelvis, which caused difficult calvings. Meyer et al. (2000) reported that calving difficulty, gestation length of 15 to 12 d below the mean, and male calves were important risk factors for stillbirth.
Farm-recorded CES had a strong association with the proportion of stillbirths within CES groups. The stillbirth rate for CES 1 was 3.6%, whereas the stillbirth rate for CES 4 was 60.1%. Mangurkar et al. (1984) reported a strong association between intensity of calving difficulty and proportion of stillbirth. It is difficult to ascertain whether difficult calvings lead to higher stillbirth rates or whether intrauterine death of calves leads to difficult calvings. Although the proportion of stillbirth undoubtedly increases as CES increases, CES 1 and 2 represented 67.9% of all stillbirth calvings used in the study. Johanson and Berger (2003) conclude that 49% of the perinatal mortality was associated with unassisted births.
In the present study, cows that had female calves were less likely to have stillbirths compared with cows that had male calves. Maltecca et al. (2006) observed that odds of stillbirth were significantly lower for female calves when compared with male calves.
Few studies have attempted to determine the effects of stillbirth calvings on the dam’s subsequent survival and days open. Mangurkar et al. (1984), using a mixed linear model statistical procedure, reported that cows that gave birth to stillborn calves had increased risk of being culled and less chance of getting pregnant. In the present study, stillbirths significantly increased the risk of death/culling throughout the lactation. Cows that gave birth to stillborn calves had a 40.9% higher hazard rate of death/culling compared with cows that gave birth to live calves. It is important to point out that the interaction of CEG and stillbirths on the Cox proportional hazards model for dam survival was not significant, which means that the detrimental effect of stillbirth on survival existed even if the stillbirth calving was unassisted. Stevenson and Call (1988) reported that cows experiencing stillbirths were at increased risk for a number of postpartum disorders such as prolapsed uterus, retained placenta, metritis, and displaced abomasum. Correa et al. (1993) also reported increased odds of developing metritis and retained placenta for cows that had stillbirth. It is possible that the higher incidence of postpartum disorders may decrease the survival of cows that had stillbirths.
To the best of our knowledge, this is the first study that used survival analyses to study the effects of stillbirths on the dam’s reproductive performance. Stillbirth increased median calving-to-conception interval by 88 d. The hazard ratio for being diagnosed pregnant was 24.1% lower for cows that had a stillbirth when compared with cows that had a live calf. The interaction between stillbirth and CEG was not significant, which indicates that the detrimental effect of stillbirth calvings on dam’s reproductive performance is independent of calving difficulty.
Stillbirth may initiate a cascade of effects that will detrimentally affect the cow’s performance, but it is possible that there are common causes to calf mortality and dam’s poor performance (Mangurkar et al., 1984). Retrospective research offers an opportunity to make inferences about a population by analyzing data that were recorded, for different purposes, over a period of time in the past. As a consequence, the data used in retrospective studies may not be as precise as data collected for prospective research. In the present study, data were recorded by farm employees, and the quality of data collected was only evaluated retrospectively. Furthermore, subjective data such as CES can vary from farm to farm and even within farm. Additional prospective studies that include health traits prior to calving and during the postpartum period are necessary to elucidate the causal effects that stillbirth has on periparturient diseases.
|
CONCLUSIONS
|
|---|
Primiparous cows had significantly higher stillbirth rates compared with multiparous cows. Although 67.9% of all stillbirth calvings were from unassisted calvings, the incidence of stillbirth was highly correlated with calving ease score. Cows that had stillbirths had a significantly increased hazard rate of culling/death throughout the lactation, with median days from calving to culling/death of 255 compared with 270 d for cows that had a live calf. Cows that had a stillborn calf had their reproductive performance significantly affected. The hazard rate of being diagnosed pregnant was 24% lower for the cows that had a stillbirth calf when compared with the cows that had a live calf. Kaplan-Meier survival analyses indicated that the median days open was increased by 88 d for cows that had stillbirth calves compared with the cows that had live calves. The losses from stillbirths are far greater than just the value of the stillborn calf.
Received for publication August 2, 2006.
Accepted for publication February 7, 2007.
|
REFERENCES
|
|---|
Cantor, A. B. 1997. Extending SAS® Survival Analysis Techniques for Medical Research, Pages 112–113. SAS Institute Inc., Cary, NC.
Correa, M. T., H. Erb, and J. Scarlett. 1993. Path analysis of seven postpartum disorders of Holstein cows. J. Dairy Sci. 76:1305–1312.[Abstract]
Emanuelson, U., P. A. Oltenacu, and Y. T. Gröhn. 1993. Nonlinear mixed model analyses of five production disorders of dairy cattle. J. Dairy Sci. 76:2765–2772.[Abstract]
Hansen, M., I. Misztal, M. S. Lund, J. Pedersen, and L. G. Christensen. 2004. Undesired phenotypic and genetic trend for stillbirth in Danish Holsteins. J. Dairy Sci. 87:1477–1486.[Abstract/Free Full Text]
Johanson, J. M., and P. J. Berger. 2003. Birth Weight as a predictor of calving ease and perinatal mortality in Holstein cattle. J. Dairy Sci. 86:3745–3755.[Abstract/Free Full Text]
Maizon, D. O., P. A. Oltenacu, Y. T. Gröhn, R. L. Strawderman, and U. Emanuelson. 2004. Effects of diseases on reproductive performance in Swedish Red and White dairy cattle. Prev. Vet. Med. 66:113–126.[CrossRef][Medline]
Maltecca, C., H. Khatib, V. R. Schutzkus, P. C. Hoffman, and K. A. Weigel. 2006. Changes in conception rate, calving performance, and calf health and survival from the use of crossbred Jersey x Holstein sires as mates for Holstein dams. J. Dairy Sci. 89:2747–2754.[Abstract/Free Full Text]
Mangurkar, B. R., J. F. Hayes, and J. E. Moxley. 1984. Effects of calving ease-calf survival on production and reproduction in Holsteins. J. Dairy Sci. 67:1496–1509.[Abstract/Free Full Text]
Meyer, C. L., P. J. Berger, and K. J. Koehler. 2000. Interactions among factors affecting stillbirths in Holstein cattle in the United States. J. Dairy Sci. 83:2657–2663.[Abstract]
Meyer, C. L., P. J. Berger, K. J. Koehler, J. R. Thompson, and C. G. Sattler. 2001. Phenotypic trends in incidence of stillbirth for Holstein in the United States. J. Dairy Sci. 84:515–523.[Abstract]
Moss, N., I. J. Lean, S. W. J. Reid, and D. R. Hodgson. 2002. Risk factors for repeat-breeder syndrome in New South Wales dairy cows. Prev. Vet. Med. 54:91–103.[CrossRef][Medline]
Philipsson, J., J. L. Foulley, J. Lederer, T. Liboriussen, and A. Osinga. 1979. Sire evaluation standards and breeding strategies for limiting dystocia and stillbirth. Report of an E.E.C./E.A.A.P. working group. Livest. Prod. Sci. 6:111–127.[CrossRef]
Spruance, S. L., J. E. Reid, M. Grace, and M. Samore. 2004. Hazard ratio in clinical trials. Antimicrob. Agents Chemother. 48:2787–2792.[Free Full Text]
Stevenson, J. S., and E. P. Call. 1988. Reproductive disorders in the periparturient dairy cow. J. Dairy Sci. 71:2572–2583.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
A. Sewalem, F. Miglior, G. J. Kistemaker, P. Sullivan, and B. J. Van Doormaal
Relationship Between Reproduction Traits and Functional Longevity in Canadian Dairy Cattle
J Dairy Sci,
April 1, 2008;
91(4):
1660 - 1668.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
