Effects of Late Gestation Heat Stress on Postpartum Milk Production and Reproduction in Dairy Cattle R. B. MOORE,l J. W. FUQUAY,l and W. J. DRAPALA* Mississippi State University
Mississippi State 39762 ABSTRACT
INTRODUCTION
Carry-over effects of late gestation heat stress on postpartum productive and reproductive traits were estimated from DHI records using 341 lactations from six sites in Mississippi. Climatological data were gathered from records of weather stations near the sites. Using multiple linear regression analyses, predictor variables for lactations were age at calving, lactation number, maximum degreedays (above 32.2'C) during the periods 30 and 60 d prepartum, and precipitation 30 and 60 d prepartum. Months and sites were indicator variables. Dependent variables included milk and fat production during early, mid, and late lactation; days to peak lactation; days open; services per conception; and body weight. Age at calving affected milk and fat production in mid and late lactation and services per conception. Degreedays for 60 d prepartum had the greatest negative influence on production variables; its statistical significance was shown in predictions of milk and fat production in early and midlactation. Days open were higher for July than for cows calving in August or September. Sites had effects on many milk and fat measurements and some reproductive traits. These results indicate that heat stress in the last 60 d of gestation has negative effects on some production variables. (Key words: late gestation, heat stress, milk production, reproduction)
The detrimental effects of high ambient temperature on the physiological functions of the cow have been the subject of much research. Reviews on hyperthermal stress have shown both acute and prolonged negative effects on lactation in dairy cattle with numerous workers citing a depression in milk production when average daily temperatures exceeded 21'C (1, 4, 15, 16). Adverse responses to summer temperatures appear to be more severe when the minimum daily temperature fails to drop into the thermoneutral zone (11). This is supported by research indicating a positive response to night cooling (4, 5, 18). Temperatures above the thermoneutral zone of cattle adversely affect duration and intensity of estrus (4, 9), conception rate (2, 7, 10, 13), embryo survival (7), and days open (2). More estrous cycles of irregular length occur in summer (6). The focus of most of the research has been on failure of fertilization and early embryonic mortality (7, 8, 10). Fans to provide night cooling improved estrous response rates in lactating Holsteins during summer (18). Research on the effects of heat stress during late pregnancy on postpartum reproductive changes and milk production has been limited. The effects of shade and other types of cooling on heat stress relief during the last trimester of pregnancy, as measured by postpartum response of Holstein cows, were examined (3, 11, 17). The relief from heat stress during this period increased calf birth weight and postpartum milk production. However, days to first service, days open, or services per conception were not altered (3, 12). In those earlier studies, the degree of exposure to high ambient temperature was not quantified. The objective of this study was to examine the carry-over effects of late gestation stress on postpartum milk and fat production and on certain reproductive traits and, in so doing, to estimate the level of stress for a specified prepartum period.
Received December 23, 1991. Accepted March 9,1992. bcpartment of ~nimaland airy Sciences. 2Lkpamnent of F i x ~ t a statistics. l
1992 J Dairy Sci 751877-1882
1877
1878
MOORE ET AL. MATERIALS AND METHODS
Data
Source
Prepartum climatological data and postpar-
aun cow performance data were collected from 7 cow-years in six Mississippi dairy herds. The herds were the Mississippi State University herd located at Starkville (site l), at Coastal Plains Branch Experiment Station at Newton (site 2), and at commercial dairies at Newton (site 3), Pontotoc (site 4). Jackson (site 5), and Poplarville (site 6). AU herds were located near official weather stations. Only Holsteins were used in this research. The range in rolling herd average within the six sites was 6361 kg of milk at site 6 to 8442 kg of milk at site 1. The range in milk fat rolling herd average was 210 kg at site 6 to 291 at site 2. These were above average herds for this region; the state average of cows on DHI was 5952 kg of milk and 215 kg of fat. Management in these herds also was considered to be above average for this region. Cow data were obtained from DHI Form 200 and DHI Form 203 (North Carolina State University Dairy Records Processing Center) on 341 lactations. There were 132 lactations from site 1, 59 from site 2, 28 from site 3, 51 from site 4, 27 from site 5 , and 44 from site 6. Only complete 305d lactations were used. B e cause the study dealt with only late gestation stress, cows that calved from July through September were included for 1976 through 1982. Climatological data taken at weather stations near each site were obtained from publication number ISSN 0145-0425, published monthly by the National Oceanographic and Atmospheric Administration.
Minimum degreedays was the cumulative number of d e p s that the minimum temperature was below 21.1'C for the last 30 or 60 d of pregnancy. Maximum degreedays was the cumulative number of degrees that the maximum temperature was above 32.2"C for the specified period. Precipitation was the centimeters of total rainfall during the specified period. Milk and fat production was calculated from monthly test day data. The values for milk in early, mid, and late lactation were calculated by adding the test day milk weights in the specified period (i.e., 102 d), dividing by the number of tests in that period, and then multiplying by the length of the period. The fat values were calculated in the same manner except that kilograms of fat produced during each period were used. The remaining variables listed were obtained from DHI data as they appeared on the forms. Data were recorded with calving month and site indicated. Statistical Analysis
Because the weather data were in a format that was difficult to extrapolate, the following statistical procedure was employed to identify variables that contributed significantly to a regression analysis. The Linear Least Squares Select Program (Mississippi State University Statistics Library, Mississippi State) was used for detecting the best subset of k predictors on 83 of the 341 records. Using this procedure, the coefficient of multiple determination was not significantly lowered after minimum degreedays was deleted from the analysis ("able 1); therefore, minimum degreedays was deleted from later analyses. The statistical procedure used was a multiple linear regression analysis, which pennitted the estimation of combined and separate Data Collection effects of the predictor variables on pertinent The following information was obtained or dependent variables. calculated for each record: age at calving, lacThe general equation for the regression tation number, minimum degreedays for 30 model was and 60 d prepartum, maximum degreedays for 30 and 60 d prepartum, precipitation for 30 6 and 60 d prepartum, early (first 102 d of biXq + m% + sitel + ej, Yj = bo + lactation following stress), mid (second 102 d :=1 of lactation), and late (last 101 d of lactation) lactation milk and fat production, number of where Y variables were early, mid, and late days until cow peaks in production, number of lactation milk production; early, mid, and late days open, number of services per conception, lactation fat production; days to peak production; days open; services per conception; body and body weight at calving. Journal of Dairy Scimce VoL 75, No. 7. 1992
LATE GESTATIOPT HEAT STRESS
TABLE 1.The effect of deleting minimum degreedays on the multiple cornlation coefficient.
After deleting
AI variables
minimum degreedays
~~
Milk production Early lactation Midlactation Late lactation Fat Early lactation Midlactation Late lactation Services to conception Days open Days to peak BW
6682 .6129 S232
.6590
.3733 .2750 .a78 .2813 2.559 25% .7699
.3O58 ,2713 .0618 .2268 .2093 .1730 .7581
.5945
SO56
weight; bo is the Y intercept; are the partial regression coefficients; site is the effect of site 1 (1 = 1.5); m% is effect of calving month (k = July, August, and September), and e, is the random error associated with the observation j (i = 1, 2 . . . n , where n is the number of observations). In all analyses, July was used as the monthly standard, and site 6 was used as the site standard. These standards were chosen arbitrarily. In a trial run,September was used as the monthly standard. Results of this analysis showed no change in significance. The X variables were X 1 = age at calving, X 2 = lactation number, X 3 = maximum degree-days (30 d prepartum), X , = maximum degreedays (60 d prepartum), X 5 = precipitation (30 d prepartum), and X, = precipitation (60 d prepartum). RESULTS AND DISCUSSION
Means, standard deviations, and standard errors of the 341 lactations are shown in Table 2. Age at Calving
The mean age at calving for all cows in the analysis was 52.2 mo. Age at calving affected mid and late lactation milk and fat production and services per conception (Table 3). This may be because cows produce more milk in succeeding lactations with lower producing
1879
cows being culled earlier (14). The significance in services per conception illustrates that older cows require more services per conception (14). Age at calving tended to have an effect on days to peak lactation (P < .lo) (Table 3). Lactation Number
Because lactation number and age at calving were highly correlated (r = .973), they affected each model similarly except for midlactation milk and fat production and services per conception (Table 3). Lactation number approached significance (P < .lo) in the latter three models. Factors such as reproductive problems and management control of calving interval likely created some differences between age at calving and lactation number. Both variables made significant contributions to late lactation miLk and fat production (Table 3). DegreeDays
This variable was used to estimate the level of heat stress on a cow during a particular period. Results of the regression analysis indicated that these values had a significant impact on several milk and fat production variables. Because degree-days had not been used by other physiologists, it was not possible to compare data. Researchers in Florida used temperature readings in shade and in no shade to evaluate the effects of prepartum heat stress on postpartum milk production and reproductive changes (3, 12). However, they did not report estimates of the level of stress as in this research. Maximum degree-days at 30 d prepartum was not a predictor in any model. The length of stress was probably too short to affect postpartum variables that were studied. Maximum degree-days 60 d prepartum had an impact on early (P < .01)and midlactation milk production (P .05) and on early (P < .05)and midlactation fat production (P < .05) flable 3). We conclude that, because it was not significant in predicting late lactation milk and fat production, late gestation heat stress does not affect late lactation. Although maximum degreedays 60 d prepartum affected most mi& and fat production variables, it did Journal of Dairy Science Vol. 75, No. 7. 1992
1880
MOORE ET AL.
TABLE 2. Means, staodard deviations, and stsndard errors for each variable in the model (all sites analysis).
Variable Independent variables Age at calving, mo Lactation number Maximum degreedays' 30 d Prepartnm 60 d prepartum Recipitation 30 d ~ ~ a r t u m ~m, 60dprepartum,cm Dependent vpriabks
x
SD
SE
52.25 2.75
23.15 1.65
1.25
85.09 16257
7358 135.77
3.98 7.35
4.09 8.01
2.% 427
.16 23
Early laclation'
287759
Midlactation Late lactation Fat production, kg Early lactation Midlactation Late lactation
2522.80
1937.60
607.01 604.71 523.25
32.87 32.75 28.34
91.84 87.40 70.09 58.61 124.89 1.99 562.32
24.37 19.80 20.43 30.66 66.36 1.38 85.94
1.32 1.07 1.11 1.66 3.59 .07 4.65
.09
Milk,@
Days to peak Days open Services per conception BW, 4 3
~~
~~~
'Maximum degreedays = The cumulative number of degrees that the maximum temperature was above 322'C for the specified period. 'Early lactation = Milk or fat in the 6rst 102 d of lactation following stress, midlactation = milk or fat in the second 102 d of lactation, and late lactation = milk or fat in the last 101 d of lactation.
not affect reproductive measurements. Similarly, Lewis et al. (12) found that, although heat stress prepartum affected some sensitive measures of postpartum reproductive function, it did not alter days to first estrus, days open, or services per conception.
Site Effects
In all analyses, site 6 was used as the benchmark The regression coefficients in Table 3 indicate that sites l and 2 had the highest milk and fat production as indicated by positive coefficients, and sites 5 and 6 had the lowest. Sites affected almost all milk and fat Precipitation measurements and some reproductive traits Precipitation in any of the specified periods (Table 3). The differences among sites within had little effect on the regression analysis for each model could be attributed to the differany of the models. It was anticipated that ences in management of the herds, such as precipitation could affect some physiological quality of nutrition and housing. measures by increasing the humidity. However, the increased humidity may have been modulated by reduced solar radiation because CONCLUSIONS of cloud cover. Maximum degreedays at 60 d prepartum Month Effects had the greatest negative influence on variables Month of calving (July, August, or Septem- affecting production of milk and milk fat durber) did not affect any of the milk and fat ing early and midlactation. It was not signifitraits, but it did affect days open (P c .05). The cant in predicting late lactation milk and fat negative regression coefficients in Table 3 in- production; therefore, late gestation heat stress dicate that days open were higher for cows does not affect late lactation. Cows calving in calving in July than in August or September. July had more days open than those calving in Journal of Dairy Science Vol. 75. No. 7, 1992
1881
LATE GESTATION HEAT STRESS
* +
B II
Journal of Dahy Science Vol. 75, No. 7, 1992
1882
m R F 4 ET AL.
August and September. Heat stress in the last 60 d of gestation does negatively affect some production variables. REFERENCES 1 Bond, T. E. 1969. Livestock environment research review. Proc. Annu. Mtg. Am. Soc. Agric. Eng.,
Pacific Coast Regioq Phoenix# Az. ZCavestany, D., A. B. El-Wishy, and R. H. Foole. 1985. Effect of season and high enviromnental temperahue on fertility of Holstein cattle. I. Dairy Sci. 68: 1471. 3Collier, R. J., S. G. Doelger, H. H. Head, W. W. Thatcher, and C. J. Wilcox. 1982. Effects of heat swess during pregnancy on maternal hormone concentrations,calf birthweight and postpartum milk yield of Holstein cows. J. Anim. Sci. 54309. 4FUqnay. J. W. 1981. Heat stress as it affects animal production. J. Anim. Sci. 52164. SFuquay, J. W., W. H. McGee, and F. M. Hutson. 1970. Seasonal effect on reproductive performawe in dairy cattle. Mississippi Agric. For. Exp. Stn. Info. Sheet 1114, Mississippi State. 6Fuquay, J. W., A. B. zook, J. W. Daniel, W. H. Brown, and W. E. Poe. 1979. Modifications in freestall housing for dairy cows daring this summer. J. Dairy Sci. 62577. 7 Gwazdaush, F. C. 1985. Effects of climate on reproduction in cattle. J. Dairy Sci. 68:1568. 8 G w a z d a ~ ~F. h ,C., C.J. W i l ~ xand , W. W. 'Ihatcher. 1975. Environmental and management factors affecting conception rate in a subtropical climate. J. Dairy Sci. 58:88. 9HaU, J. G., C. Branton, and E. J. Stone. 1959. Estrus, estrous cycles, ovulation time, time of service and
Journal of Dairy Science Vol. 75, No. 7, 1992
fertility of dairy cattle in Louisiana J. Dairy Sci. 42: 1086. lOIOgraham, R H., D. D. Tdette, and W. D. Wagner. 1974. Relationship of temperature and humidity to conception rate of Holstein cows in subtropical climate. J. Dairy Sci. 57476. 11Johnson, H. D., H. H. Kebler, A. C. Ragsdale, and M. D. Shanllin.1960. Effects of various combinations of temperatare and humidities on milk production. J. Dairy Sci. 43:871. 12lewis, G. S., W. W. Thatcher, E. L.Bliss, M Drost, and R. J. Collier. 1984. Effects of heat stress during pregnancy on postpaturn reproductive changes in Holstein cows. J. Anim. Sci. 58174. 13Monty. D. E., and L. K. Wolff. 1974. Summer heat stress and reduced fertility in Holstein-Friesian cows in Arizona. Am. J. Vet. Res. 35:1496. 14 Salisbury, G.W., and N. L. V a n D m k . 1%1. Physiology of Reproduction and Artificial Insemination of Cattle. W. H. Freeman, San Francisco, CA. 15Skwart, R E. 1973. Environment research in cattle since 1968. Resented 1973 Annu. Mtg. Am. SOC. Agric. Eng. Univ. Kentucky, Lexington. Tech. Paper 73-415. Am. Soc. Agric. Eng., St. Joseph, MI. 16Thatcher, W. W., F. C. Gwazdauskas, C.J. Wilcox, J. Toms, H. H. Head, D. E. Buffington, and W. B. Fredrickwn. 1974. Milking performance and reproductive efficiency of dairy cows in an environmentally controlled structure. J. Dairy Sci. 57304. 17 Wolfemon, D., I. Flamenbaum, and A. Berman. 1988. Dry period heat stress relief effects on prepartum progesterone, calf birth weight and milk production. I. Dairy Sci. 71:809. lSYounas, M., J. W. Fuquay, and A. B. Moore. 1991. Estrus and endocrine response as modified by forced ventilation in lactating Holsteins during summer. J. Dairy Sci. 7qSuppl. 1):194.(Abstr.)