- Email: [email protected]
Survey of Intramammary Infections in Dairy Heifers at Breeding Age and First Parturition1
Survey of lntramammary Infections in Dairy Heifers at Breeding Age and First Parturition' L. K. FOX Washington State University Pullman 991644610
S. T. CHESTER and J. W. HALLBERG The Upjohn Co. Kalamazoo. MI 49001 S. C. NICKERSON Louisiana State University Homer 71040 J. W. PANKEY University of Vermont Burlington O5405-0148 L. D. WEAVER University of California Visalia 93291 ABSTRACT
The prevalence of IMI was greatest during the last trimester of pregnancy, ranging from 49.2%in the winter to 36.8% in the summer. The significant effects of herd location and season suggest that management variables influence prevalence of heifer IMI. Because prevalence of IMI was greatest during the last trimester of pregnancy compared with prevalence during earlier stages of pregnancy, the heifer may be most susceptible to this disease during this period of first gestation. (Key words: mastitis, freshening, pathogens)
A survey was conducted to determine and contrast prevalence of IMI in nulligravid and primigravid dairy heifers pre- and postpartum. Contrasts were made to evaluate the risk factors of location of dairy, trimester of gestation, and season of sampling on IMI. Twentyeight dairies in California, Louisiana, Vermont, and Washington were studied. Lacteal secretions were collected aseptically from heifers at breeding age (8 to 19 mo) from one side of the gland and again at 4 d postpartum from all quarters. Of the quarters sampled, 65.6% prepartum and 64.0% postpartum were free of IMI. The percentages of quarters with IMI from coagulase-negative staphylococci or Staphylococcus aureus IMI were 27.1 and 9% prepartum and 2 1.8 and 2.9% postpartum. Staphylococcus aureus IMI were most prevalent in Louisiana during the months other than summer. Location, herd, and season significantly influenced prevalence of IMI.
Abbreviation key: CNS = coagulase-negative staphylococci. INTRODUCTION
Received July 13, 1994. Accepted November 18, 1994. ]This research was supported in part by the Upjohn Co., Kalamazoo, MI 49001. 1995 J Dairy Sci 78:1619-1628
Heifer mastitis IMI can be detected either pre- or postpartum (27) and is most often subclinical (2, 13). Although heifer mastitis has been recognized as a problem for >SO yr, dairy managers traditionally have viewed heifers at parturition as being free of IMI (24). Perhaps with the rise in expectations for improved control of mastitis, as evidenced by a decrease in the legal limit for bulk tank milk SCC under the Pasteurized Milk Ordinance, dairy managers and researchers have become more
1619
1620
FOX ET AL.
concerned with heifer mastitis. Nickerson and Boddie (15) have stated that a reduction in mastitis of heifers is critical because mammary growth and development is greatest during first pregnancy. Recently, a number of studies have estimated the prevalence of heifer IMI at first parturition (Table l), and these estimates varied widely. A large proportion of mammary quarters of heifers were reported to be free of IMI at calving in one study (20), but >75% were found to have IMI in another study (27). Comparison of results from these and other studies (Table 1) is confounded by problems associated with comparisons for prevalence of IMI from several herds at different locations, during different seasons, and in different years. Moreover, the results of these studies do not explain the risk factors associated with IMI at first parturition. One salient observation of Table 1 is that IMI by coagulase-negative staphylococci (CNS)is the most prevalent pathogen type associated with mastitis in heifers. These pathogens readily colonize multiple body sites of cattle (24). Roberson et al. (22) found an association between mammary colonization at breeding age (12 to 18 mo) and Staphylococcus aureus IMI at parturition. Thus, the breeding age period may be critical with respect to risk of IMI. The purpose of this study was to determine the prevalence of IMI at breeding age and at parturition of primiparous heifers. Effects of season of sampling, age, and stage of pregnancy at first sampling; herd and location were all risk factors considered to explain the variation in prevalence of IMI.
MATERIALS AND METHODS Herd Selection
Twenty-eight herds enrolled in DHI participated in the study: 7 from California, 5 from Louisiana, 7 from Vermont, and 9 from Washington. Herds were selected by investigators, and selection criteria included 1) a history of collaborative efforts, 2) availability of animal restraint facilities, 3) distance from the investigator's laboratory, and 4) herd size and annual calving pattern. The goal was to sample between 20 to 40 heifer calves of breeding age per herd per year, distributed evenly over four seasonal periods. Sample Collectlon
Heifers between the ages of 8 to 38 mo (mean = 19 mo; median = 19.4 mo) were randomly selected at the time of herd visits. Herd visits were made once per season @ecember through February was winter, March through May was spring, June through August was summer, and September through November was autumn). Heifers may have been pregnant at their first sampling; the second sample was conducted at parturition. Lacteal secretions were aseptically collected from one-half (right side or left side) of the udder at first sampling. Teat ends were immersed in a 1% iodophor solution and then scrubbed clean with a cotton swab soaked in 70% isopropyl alcohol. Secretion samples were collected in sterile test tubes. Samples were
TABLE 1. Prevalence of MI of primiparous heifers at fvst parturition. ~~
Reference
Nickerson et al. (15) Cook et al. (1) Oliver et al. (IT+ Pankey et al. (20) Roberson et al. (22)
Heifers sampled
Milk sample'
(no.) 600
Q
525
41 382 828
Prevalence of IMI by pathogen typeZ None
CNS
sa
58.4 43.0 55.4 81.7 54.0
27.9 43.0 39.0 11.4 39.0
8.0
Env
Other
(8) C
Q
Q C
4.2
6.0
.6 .7 8.0
4.9 4.8 13.0
1.4 8.0 1.7
IQ = Quarter; C = composite. V N S = Coagulase-negative staphylococci, Sa = Staphylococcus aureus, Env = environmental pathogens (coliforms and streptococci other than agalactiae). and Other = all other pathogens. 3IMI in heifers not receiving antimicrobial therapy prior to parturition to cure infections were only included. Journal of Dauy Science Vol. 78, No. 7, 1995
SYMPOSIUM: MASTITIS IN DAIRY HEIFERS
kept on ice at 4°C and then frozen or processed within 24 h. Following sample collection, teats were immersed in an acrylic latex solution containing 1% lauricidin (Teat Shield Germicide? 3M Co., St. Paul, h4N). Milk samples were aseptically collected from all functional mammary quarters using standard aseptic collection techniques (9) within 4 d of parturition by the dairy manager or designated employee. Signs of clinical mastitis were recorded at the time of postpartum sample collection.
1621
Statlstlcal Analysis
Not all prevalence data from all mammary quarters were included in the statistical analysis. Reasons for excluding data were contaminated samples, missing samples, samples collected >4 d postpartum, and samples with no data on age of the heifer. To analyze prevalence, isolated pathogens were grouped into four categories, and the following IMI dichotomies were created: infected compared with uninfected, infected with S. aureus compared with uninfected, infected Sample Analysis with a noncontagious pathogen compared with A .Ol-ml portion of all lacteal secretions uninfected, and infected with CNS compared was plated on blood agar plates and incubated with uninfected. These analyses were conat 37°C for 48 h. Isolates were presumptively ducted to identify the effects of location, seaidentified as staphylococci, streptococci, coli- son, gestation trimester, calving, and age at forms, or other genera based on colony charac- calving, as well as their interactions on the teristics (9). Mastitis pathogens were identified proportion of quarters infected in those according to standard protocols of the National dichotomies. The Pearson chi-square statistic Mastitis (9) and then enumerated. Pathogens was used to test the equality of infection rates were classified as contagious (S.aureus, and among herds within the same season and locaCorynebacterium bovis), environmental (coli- tion. Because these tests were statistically sigforms, streptococcal species, and enterococci), nificant, it was not possible to pool quarters CNS, or other. The significance of isolation of across herds and to perform traditional contina mastitis pathogen as the agent causing IMI gency table analyses. Because herds are a ranand classification of contaminants were deter- dom effect, an ANOVA was conducted using mined by the procedures outlined (14). Briefly, variation among herds in IMI by quarter as the the contagious pathogen S. aureus was considerror term. For each of the dichotomies, the ered to cause IMI if one colony (100 cfu/ml) proportion of IMI for each herd was computed was isolated. Isolation of >200 cfu of environand transformed via the Freeman-Tukey procemental mastitis pathogendm1 of milk or isolation of C. bovis in pure culture were dure (23). A weighted ANOVA was computed considered significant, and the isolated agent for which weights equaled n + 1/2, where n is considered to have caused the IMI. Isolation of the number of quarters associated with herd >loo0 cfu/ml of C. bovis was considered sig- proportion, and the form is as outlined in nificant if other bacteria also were present. Table 4 (23). For each step, a sequential analyMoreover, isolation of environmental patho- sis was conducted to determine the model. gens was considered significant when isolated Specifically, an initial model was fit including at >loo0 cfdml and when another bacterial all main effects and interactions, using the type was also isolated. Staphylococci other pooled herd variance as error. If the three-way than S.aureus were only considered significant interaction was not significant (P > .05), then if isolated in pure culture and >loo0 cfu/ml. A this interaction term was pooled with the error, priori, the investigators thought that it was unlikely that more than two types of pathogens and only two-way interactions and main efwould cause IMI. Thus, to differentiate be- fects were included in the model. If at least tween contaminated samples and quarters with one two-way interaction was not significant (P possible multiple infections, a mixed infection > .05), then a third analysis was performed, could have only two types of pathogens. Infec- pooling those nonsignificant two-way interactions were designated based on those deemed tions with the error. This process continued as significant isolations of mastitis pathogens until either the remaining interactions were according to the aforementioned criteria. statistically significant or only the main effects Journal of Daily Science Vol. 78, No. 7, 1995
1622
FOX ET AL.
were in the model. The Pearson chi-square statistic was used to test differences in IMI prevalences between herds within the same location, season, and time category. In a separate analysis, the hypothesis was tested that the effect of sampling at breeding age would predispose the mammary quarter to IMI. Within those cows with four functional quarters, the proportion of sampled quarters at breeding age that subsequently had IMI at parturition was contrasted with the proportion of unsampled breeding age mammary quarters having IMI at parturition. Additionally, the equivalence interval between IMI prevalences in the sampled and unsampled quarters was computed. RESULTS
Two thousand four hundred thirty-five quarter samples from 1583 breeding age heifers were used in the analysis. Based on the quar-
ters sampled, the majority of all heifers at breeding age appeared to be free of IMI (Table 2). Most of those quarters with IMI were caused by CNS (Table 2). Less than 3% of the quarters from heifers of breeding age had IMI by the contagious pathogen S. aureus; Sfreptococcus agalacriae were not found (Table 2). Usable milk samples from the heifers at first parturition numbered 4950. The majority of quarters were also free from IMI at parturition (Table 3). Similar prevalences were seen for IMI by S. aureus and CNS at breeding and at parturition (Tables 2 and 3). The prevalence of IMI in mammary quarters at parturition from those quarters sampled at first breeding age was 37.3%, which was significantly more (P = .07)than the 34.4% prevalence at parturition in mammary quarters that were not sampled at breeding age. However, the difference in IMI at parturition between sampled and unsampled udder halves at breeding age was within 5% of equivalence (P = .024),indicating that, al-
TABLE 2. Prevalence of intramammary infections (IMI)by season and location in bovine nulligravida and primigravid calves at breeding age.' Pathogen type3 Season Winter
Spring
Summer
Fall
Total
LocationZ CA LA VT WA CA LA VT WA CA LA VT WA CA LA VT WA
Uninfected
(96) (no.) 67.3 70 48.0 85 37 61.7 63.0 51 102 68.5 60.3 1430 33 67.3 78.8 82 205 68.1 73.4 278 74.4 64 66.8 133 75 60.0 55.5 106 70.7 65 68.3 69 65.6 1598
Staphylococcus aureus (no.) 0 13 0 1 .7 1 10.1 24 0 0 0 0 0 0 1.3 5 0 0 1.0 2 5 4.0 7.9 15 1.1 I 3.0 3 2.9 70
(5%)
0 7.3 0 1.2
Environmental (%)
0 3.4 5.0 1.2 .7 3.0 2.0 0 .3 1.3 1.2 2.0 0 3.1 1.1 0 1.5
(no.) 0 6 3 1 1 7 1 0 1 5 1 4 0 6 1 0 37
CNS3
(no.) 26 25.0 63 35.6 17 28.3 28.4 23 40 26.8 60 25.3 12 24.5 19.2 20 81 26.9 23.0 87 24.4 21 56 28.8 40 32.0 62 32.5 22 23.9 29 28.7 27.1 659
Other (%)
7.7 5.6 5.0 6.2 3.4 1.3 6.1 1.9 4.7 1.1 0 2.0 4.0 1.0
3.3 0 2.9
(no.) 8 10 3 5
5 3 3 2 14 4 0 4 5 2 3 0 71
'Lacteal secretions were collected from one side of the udder (right or left) from heifers at 8 to 38 mo of age. *Heifers from 7 dairies in California, 5 dairies in Louisiana, 7 dairies in Vermont, and 9 dairies in Washington were part of the study. Revalence is expressed as a percentage of mammary quarters sampled, number of positive samples is also listed. 3Environmental= Coliforms and environmental streptococci, CNS = coagulase-negative stmhvlococci. and other =
1623
SYMPOSIUM: MASTITIS IN DAIRY HEIFERS
-
-
70-
i'
I
E m
1
n
dl ~,I;,I ,
.
I
r/r
LA
WA
LOCATION
_1 UNINFECTED
I-1
~~
CNS
S AUREUS OMER
E UNINFECTED
EWRONMENTAL
ENVIRONMENTAL^
I S AUREUS
0OTHER
CNS
Figure 1. Prevalence of IMI in heifers of breeding age by location. S. aureus = Sraphylococcus aureus, environmental = coliforms and environmental streptococci, and other = all other types of pathogens.
Figure 3. Prevalence of IMI by season of year in breeding age heifers. S. aureus = Staphylococcus aureus, environmental = coliforms and environmental streptococci, and other = all other types of pathogens.
though IMI prevalences in sampled and unsampled mammary quarters were significantly different, the difference was small, 4%. Location of the heifers had a significant effect on prevalence of IMI both before and after parturition (Tables 2, 3, and 4). The prevalence of IMI in heifers in Louisiana was greatest prior to parturition, a result of a greater prevalence of IMI by S. aureus, environmental mastitis pathogens, and CNS (Figure 1). Following parturition, the prevalence of IMI in heifers was again greatest in Louisiana, which had the greatest percentage
of heifers with environmental and S. aureus IMI and the second greatest prevalence of CNS IMI (Figure 2). Season significantly influenced the prevalence of IMI following parturition, but not for heifers of breeding age (Table 4, Figures 3 and 4). The IMI were more prevalent in summer when IMI from S. aureus or environmental pathogens were most prevalent (Figure 4). At breeding age, the stage of pregnancy significantly affected prevalence of IMI and was highest for heifers in their last trimester of pregnancy (Figure 5). Specifically, prevalence
r
,
1
,
I '
L
li:
CA
0UNINFECTED
Winter
WA
S AUREUS
ENVlRONMENTAL
1
I \
Figure 2. Prevalence of IMI in heifers at first parturition by location S. aureus = Srclphylococcus aureus, environmental = coliforms and environmental streptococci, and other = all other types of pathogens.
i
= 0
Summer
SEASON
___________ UNINFECTED
IO CNS
111 Fall
Spring
S AUREUS
~
_
_
_
_
ENVIRONMENTAL
OTHER
Figure 4. Prevalence of IMI in heifers at first parturition by location. S. aureus = Staphylococcus aureus, environmental = coliforms and environmental streptococci, and other = all other types of pathogens. Journal of bairy Science Vol. 78, No. 7, 1995
~
1624
FOX ET AL.
TABLE 3. Prevalence of intramammary infections (JMI) by season and location in bovine primiparae immediately following parturition. Pathogen type2 Season Winter
SPW3
Summer
Fall
Total
Location' CA LA VT WA CA LA VT WA CA LA VT WA CA LA VT WA
Uninfected (96) 70.8 54.3 53.6 80.4 73.1 51.9 64.0 71.7 69.1 42.0 61.3 63.1 79.7 65.8 61.8 73.5 64.0
(no.) 213 350 67 160 335 209 16 185 325 198 49 207 310 239 105 200 3168
Smphylococcus aureus
(96) 1.3 1.6 .8 1.0 1.1 8.2 0 1.6 1.3 6.6 5.0 1.2 1.0 6.6 1.2 2.3 2.8
(no.) 4 10 I 2 5
33 0 4 6 31 4 4 4 24 2 6 140
Environmental (%)
6.0 17.8 5.6 2.5 3.5 9.2 0 0 2.3 21.0 7.5 4.3 2.8 6.9 8.2 2.6 7.7
(no.) 18 115 7 5 16 37 0 0 11 99 6 14 11 25 17 9 385
CNS (%)
19.3 20.8 33.6 10.6 20.3 27.8 36.0 25.2 24.5 24.6 22.2 26.5 14.9 19.0 27.1 15.1
21.8
Other (no.) 48 134 42 21 93 112 9 65 115 116 18 87 48 69 46 40 1083
(%)
2.7 5.6 6.4 5.5 2.0 3.0 0 1.6 2.8 5.7 3.8 4.9
1.5 1.7 1.8 4.5 3.5
(no.) 8 36 8 11 9 12 0 4 13 27 3 16 6 6 3 12 174
'Heifers from 7 dairies in California, 5 dairies in Louisiana. 7 dairies in Vermont, and 9 dairies in Washington were part of the study. Prevalence expressed as a percentage of mammary quarters sampled, and number of positive samples is also listed. 2Environmental = Coliforms and environmental streptococci, CNS = coagulase-negative staphylococci, and other = all other types of mastitis pathogens.
of CNS M I was significantly affected by trimester (Table 4) and, with the exception of fall, was substantially greater during the last trimester of pregnancy (Table 5). Season by trimester of pregnancy significantly influenced prevalence of IMI (Table 4). Prevalence of IMI precalving was lowest for second trimester heifers during the fall and next lowest for third trimester calves during the winter, spring, and fall months (Table 5). Differences were found in prevalence of IMI among herds. The test statistic pooled across all groups (location, season, and time category) was statistically significant (P c .OOol).
tors for this disease. Our study was conducted to determine how the effects of location, season, stage of pregnancy, and herd might affect IMI in heifers. Location significantly affected the prevalence of IMI in heifers. Location was an expected effect on prevalence of IMI in heifers, given the wide range in dairy management styles in the four states participating in this study. Moreover, within location, effect of herd on heifer IMI was significant. Management is the major factor controlling the incidence of IMI in lactating cows (21). More generally, the genetic effect on prevalence of mastitis has been estimated to be small relative to the environmental effect (25). Thus, the estimates for the lactating cow appear to hold DISCUSSION true for heifers as well. In the past, S. aureus Knowledge of how, why, and when an ani- IMI in heifers tended to be higher in Louisiana mal is predisposed to a disease is a necessary (Table 1). In this study, S. aureus prevalence prerequisite for establishing control strategies was severalfold greater in Louisiana than in and disease prevention. To date, a number of other locations (Figures 1 and 2). Previously, a studies have been made to estimate the preva- report from Washington (22) indicated a relalence of M I in heifers (Table I), but no studies tively high prevalence of S. aureus in heifers at have been conducted to identify the risk fac- calving of 8% (Table l), although, in our
Journal of Dairy Science Vol. 78, No. 7, 1995
1625
SYMPOSIUM: MASTITIS IN DAIRY HEIFERS
study, the mean prevalence in Washington herds was <3% (Table 2). Streptococcus agalacriae was not isolated from any heifer samples, which would seem unusual, given that Strep. agalactiae has been found in herds in each region (6, 7, 12, 19). This result indicates that the vectors of transmission of S. aureus in heifers, although similar to those vectors of transmission of other contagious mastitis pathogens in lactating cows, differ from those vectors of transmission of Strep. agalactiae among heifers. In addition to differences in management systems between locations, differences in climatic conditions might influence the incidence of IMI. For example, Smith et al. (26) reported an increase in IMI by environmental pathogens during the warm rainy months of summer in Ohio. Several reports (3, 10, 18) have established a problem with summer mastitis in Europe. Data from Fox and Hancock (4) indicated an increased prevalence of S. aureus IMI during acute cold weather, indicating that season influenced the prevalence of IMI. Indeed, season of sampling did significantly influence the
prevalence of IMI at parturition. At parturition, the number of IMI was greatest during the summer, primarily because of a relative increase in IMI by the staphylococci, S. aureus, and CNS (Figure 4). Presumably, season and climate did not have a direct effect on the mammary gland. Rather. conditions associated with climate exposed the gland to an increased bacterial load that management practices failed to control (21). Some of the conditions that were associated with increased prevalence among heifers at calving did not influence the prevalence of IMI among breeding age heifers, because prevalence of IMI within the latter group was not influenced by season. Nickerson et al. (16) suggested that flies may be responsible for the transmission of S.aureus to breeding age heifers. In that study, the prevalence of S. aureus IMI in breeding age heifers was lowest in the summer, presumably a period when a fly problem would be at the apex. Thus, either flies were not a major vector of S. aureus in Louisiana in this study, or the time necessary from transmission of S. aureus to establishment of IMI was of sufficient duration
TABLE 4. Tests for significance of factors affecting prevalence of IMI. Precalvingl df
Infected3
3 3 3 9 7 9 17 147 198
<.001 .317 <.001 ,407 ,111 ,018 ,804 1.9516
Postcalving2
CNS4
df
Infected
-P
-P Location Q Season ( S ) Times (T) L X S L x T S x T L x S x T Error Total
~~
CNS
.039 ,999 <.001 .587 .024
3
3
.045
,250 1.4236
2 9 6 6 18 182 229
C.001 <.@I ,732 .320 366 .471 ,367 2.8146
,002 ,001
.635 ,366 ,945 ,890 .559 2.4336
~~
'Lacteal secretions were collected from one side of the udder (right or left) from heifers at 8 to 19 rno of age *Heifers from 7 dairies in California, 5 dairies in Louisiana, 7 dairies in Vermont. and 9 dairies in Washington were part of the study. 3Mammary quarters were considered as being either infected or free of infection. 4Mammary quarters were considered as having a coagulase-negative staphylococcal (CNS) IMI, as having another infection, or as being uninfected. sPrecalving time is trimester of pregnancy: interval 1 = 1 to 95 d (n = 483). interval 2 = 96 to 190 d (n = 769), interval 3 = 191 to 285 d (n = 482); interval 4 (n = 515) includes all heifers with prolonged gestation periods, heifers not bred at the time of sampling, or heifers that had multiple conceptions between time of sampling and first calving. Postcalving time 1 = parturition at <24 mo (n = 1803); 2 = parturition at 24.5 to 26 mo (n = 1485); and 3 = parturition at >26 mo (n = I63 1). 6Mean square error. Journal of Dairy Science Vol. 78, No. 7, 1995
1626
FOX ET AL.
pregnant heifers were more prone to IMI than were nonpregnant heifers. However, the mean prevalence in this study for group 4 heifers was similar to those of groups 1 and 2 (Figure 5). Rather, heifers in their third trimester of pregnancy had the highest prevalence of M I . Pregnancy does affect mammary gland development, and thus a mammary gland of a gravid heifer would be physiologically much different from that of a nongravid heifer. Differences in mammary physiology have been associated with differences in susceptibility to IMI in lactating cows and could explain why differences might have existed in prevalence of IMI between gravid and nongravid heifers. Alternatively, the rate of mammary gland development varies with different stages of pregnancy; the greatest rate of growth occurs at the end of gestation (28). The more rapid development of the mammary gland in heifers in the third trimester of pregnancy appears to be as-
that new IMI would not be detected until the next sampling period. Indeed, the prevalence of S. aweus IMI in breeding age heifers was much greater in fall than in summer in Louisiana (Table 2). Trimester of pregnancy did have a significant effect on prevalence of IMI in heifers (Table 4). Most of the heifers in gestation category 4 of Table 5 were assumed to be not pregnant because their sampling to calving interval was >285 d. However, a few might have had a prolonged gestation period. Moreover, some heifers may have been in their first few days of pregnancy at the time of sampling or, alternatively, may have conceived earlier but, after loss of conceptus, were rebred. Accurate breeding records for heifers were not always available from the dairies studied. Therefore, all cows with extended intervals from sampling to calving in this fourth category were included. Evidence (3, 10, 18) suggested that
TABLE 5. Prevalence of intramammary infections by season and stage of pregnancy at first sampling.’ ~~
Pathogen type2 Season Fall
Spring
Summer
Winter
Stage
Uninfected
(dY
(96) 75.8 50.3 52.5 69.1 71.7 70.0 51.95 69.9 72.3 78.8 63.2 61.3 64.1 70.7 50.8 61.4
0-95 96190 191-285 Other 0-95 96-190 191-285 Other 0-95 96-190 191-285 Other 0-95 96-190 191-285 Other
(no.) 25 85 61 47 129 84 55 65 138 249 84 155 50 116 65 62
Staphylococcus aureus
(96) 0
6.5 2.6 0 5.6 5.0 8.5 0 0 1.3 0 1.2 2.6 4.9 6.3 5.0
(no.) 0
11 3 0 10 6 9 0 2 4 0
3 2 8 8 5
Environmental
CNS
6) (no.)
(%I
6.1 2.4 2.6 0 1.7 4.2 0 1.1 1.0 1.6 1.5 .8 3.8 1.8 .8
2 4 3 0 3 5 0
0
0
15.2 34.9 36.2 25.0 20.6 16.7 37.7 26.9 25.7 17.4 32.3 32.8 26.9 21.3 39.1 31.7
1
2 5 2 2 3 3 1
Other (no.) 5 59 42 17 37 20 40 25 49 55 43 83 21 35 50 32
@) 3.0 5.9 6.0 5.9 .6 4.2 1.9 2.2 1.0 .9 3.0 4.0 2.6 1.2 3.1 2.0
(no.) 1
10
7 4 1 5 2 2 2 3 4 10
2 2 4 2
‘Lacteal secretions were collected from one side of the udder (right or left) from heifers at 8 to 38 mo of age. Heifers from 7 dairies in California, 5 dairies in Louisiana, 7 dairies in Vermont, and 9 dairies in Washington were part of the study. 2Environmentalpathogens = Coliforms and environmental streptococci, CNS = all coagulase-negative staphylococci, and other = other mastitis pathogens. 3F’recalving time is trimester of pregnancy, 1 to 95 d = first trimester (n = 483), 96 to 190 d = second trimester (n = 769), and 191 to 285 d = third trimester (n = 482); other (n = 515) includes all heifers with prolonged gestation periods, heifers not bred at the time of sampling, or heifers that had multiple conceptions between time of sampling and first calving. Journal of Dairy Science Vol. 78, No. 7, 1995
SYMPOSIUM: MASTITIS IN DAIRY HEIFERS
1627
CONCLUSIONS
n
2 3 GESTATION PERIOD
~
~
__-_ _
0UNINFECTED 0CNS
S
-_ AUREUS
ENVIRONMENTAL '
0OTHER
Figure 5 . Prevalence of IMI (S. aureus = Sraphylococcus aureus, environmental = coliforms and environmental streptococci. and other = all other types of pathogens) by trimester of gestation [l = 1 to 95 d, 2 = 96 to 190 d, 3 = 191 to 285 d; 4 = heifers with prolonged gestation periods (>285 d), heifers not bred at the time of sampling, or heifers that had multiple conceptions between time of sampling and first calving].
sociated with increased prevalence of MI, as seen in the current study. Most of the increase in prevalence in IMI during the last trimester was due to CNS. Hogan et al. (11) indicated that these pathogens are opportunists and that heifers were more likely to have IMI by CNS than were multiparous cows. Hogan et al. (11) reported that, for all lactating cows, the prevalence of CNS IMI decreased as days of lactation increased and decreased during the dry period as the heifers began their second lactation. Presumably, dry cow therapy eliminated the remaining CNS IMI, as this group of pathogens is susceptible to dry cow therapy (8). Although there was a significant effect of sampling at breeding age on IMI at calving, the effect was small, and prevalence of IMI at breeding age and at parturition was similar. The lack of difference in prevalence would suggest few changes in IMI status from breeding age to parturition. The apparent lack of dynamics in IMI could be explained by the results of a study by Pankey et al. (20). Pankey reported that the number of new IMI are equivalent to the number of cures, thus giving the impression of IMI stasis between breeding age and parturition. Similar results were reported by Weaver et al. (29).
The hypothesized risk factors, season, location, trimester of pregnancy, and herd, all significantly affected the prevalence of IMI in dairy heifers. Management strategies associated with rearing of heifers appear to have major impact on control of this disease as evidenced by significant location and herd effects. Thus, as with mastitis in lactating cows, heifer IMI are controlled through management of heifer environment. Although results of this study did not indicate what management variables most influence mastitis IMI in heifers, the finding that the last trimester of pregnancy of heifers was more predisposed to IMI is significant. Clearly, future studies should concentrate on this critical period to determine what control practices could reduce the prevalence of IMI in heifers. ACKNOWLEDGMENTS
Appreciation is extended to J. R. Roberson for his help in sample collection, to Kenneth J. Dame for data organization, to personnel at the Washington State University dairies, and to all other cooperating dairies. REFERENCES 1 Cook, W. F., E. A. Fiez, and L. K. Fox. 1992. Mastitis in fmt lactation southwest Idaho dairy cows. J. Dairy Sci. 75(Suppl. 1):158.(Abstr.) 2 Daniel, R.L.W., D. A. Bamum, and K. E. Leslie. 1986. Observations on intramammary infections in first calf heifers in early lactation. Can. Vet. J. 27:112. 3Egan, J. 1990. Prevalence of summer mastitis in selected herds in the Province of Leinster. Ir. Vet. J. 43:74. FOX, L. K., and D. D. Hancock. 1989. Effect of segregation on prevention of intramammary infections by Staphylococcus aureus. J. Dairy Sci. 72540. 5 Fox, L. K., and J. R. Roberson. 1993. Heifer mastitis: is it a problem? Page 187 in Proc. Natl. Mastitis Counc., 32nd Annu. Mtg.. Kansas City, MO. Natl. Mastitis Counc., Inc., Arlington, VA. 6Goldberg. J. J., J. W. Pankey, P. A. Drechslesr, P. A. Murdough, and D. B. Howard. 1991. An update survey of bulk tank milk quality in Vermont. J. Food Prot. 54549. 7Gonzalez. R. N., D. E. Jasper, R. B. Bushnell, and T. B. Farber. 1986. Relationship between mastitis pathogen numbers and bulk tank milk in bovine udder infections in California dairy herds. J. Am. Vet. Med. Assoc. 189:442. 8 Harmon, R. J., W. L. Christ, R. W. Hemken, and D. E. Langlois. 1986. Prevalence of minor udder pathogens after intramammary dry treatment. J. Dairy Sci. 69: 843. Journal of Dairy Science Vol. 78, No. 7, 1995
1628
FOX ET AL
9Harmon. R. J., R. J. Eberhart. D. E. Jaspesr, B. E. Langlois, and R. A. Wilson. 1990. Microbiologcal procedures for diagnosis of bovine udder infection. Natl. Mastitis Counc., Inc., Arlington, VA. 10Hillerton. J. E., J.G.H. West, and M.F.H. Sheam. 1992. The cost of summer mastitis. Vet. Rec. 131:315. 11 Hogan, J. S., J. W. Pankey, and K. L. Smith. 1987. Stuphylococcus species, other than Staphylococcus uureus. Page 21 in Proc. Natl Mastitis Counc. 26th Annu. Mtg., Orlando, FL. Natl. Mastitis Counc., Inc., Arlington, VA. 12Hutton. C. T., L. K. Fox, and D. D. Hancock. 1990. Mastitis control practices: differences between herds with high and low milk somatic cell counts. J. Dairy Sci. 73:1135. 13 Munch-Petersen, E. 1970. Mastitis in bovine primiparae. Vet. Rec. 87:568. 14 National Mastitis Council. 1987. Laboratory and Field Handbook on Bovine Mastitis. W. D. Hoard and Sons, Fort Atkinson, WI. 15 Nickerson, S. C., and R. L. Boddie. 1992. Prevalence of heifer mastitis in Northwest Louisiana. Louisiana Cattlemen (Apr.):2. 16 Nickerson, S. C., W. E. Owens, and R. L. Boddie. 1995. Mastitis in dairy heifers: initial studies on prevalence and control J. Dauy Sci. 78:1599. 17 Oliver, S. P., M. J. Lewis, B. E. Gillespie, and H. H. Dowlen. 1992. Influence of prepartum antibiotic therapy on intramammary infections in primigravid heifers during early lactation. J. Dairy Sci. 75406. ISO’Rourke, D. J., R. J. Chamings, and J. M. Booth. 1984. Summer mastitis surveys in England and Wales: 1978-1983. Vet. Rec. 115:62. 19 Owens. W. E., and S. C. Nickerson. 1994. Use of bulk tank cultures for mastitis management. Dairy Res. Rep.. Hill Farm Res. Stn.,Homer, LA.
Journal of Dairy Science Vol. 78, No. 7, 1995
20 Pankey, J. W., P. A. Drechler, and E. Wildman. 1991. Mastitis prevalence in primigravid heifers at parturition. J. Dairy Sci. 74:1550. 21 Reneau, J. K. 1986. Effective use of dairy herd improvement somatic cell counts in mastitis control. J. Dairy Sci. 69:6. 22 Roberson, J. R., L. K.Fox, D. D. Hancock, C. C. Gay, and T. E. Besser. 1994. Coagulase-positive staphylococcus intramammary infections in primiparous dairy cows. J. Dairy Sci. 77:958. 23 SAS” User’s Guide: Statistics, Version 5 Edition. 1985. SAS Inst., Inc., Cary, NC. 24Shearer. J. K.,and R. J. Harmon. 1993. Mastitis in heifers. Vet. Clin. North Am. Food Anim. Pract. 9: 583. 25 Shook, G. E. 1993. Genetic improvement of mastitis through selection on somatic cell count. Clin. North Am. Food Anim. Pract. 9:563. 26Smith, K. L., D. A. Todhunter, and P. S . Schoenberger. 1985. Environmental mastitis: cause, prevalence, and prevention. J. Dairy Sci. 68:1531. 27 Trinidad, P., S . C. Nickerson, T. K.Alley, and R. W. Adkinson. 1989. Efficacy of intramammary treatment in unbred and primigravid dairy heifers. J. Am. Vet. Med. Assoc. 197:465. 28Tucker, H. A. 1981. Physiological control of mammary growth, lactogenesis, and lactation. J. Dairy Sci. 64:1403. 29 Weaver, L. D., C. A. Holmberg, J. S . Cullor. L. K. Fox, S . C. Nickerson, J. W. Pankey, J. Hallberg, and S. T. Chester. 1994. Effect of intramammary infection in dairy heifers, location, and season of calving on intramammary infection at calving. J. Dairy Sci. 77(Suppl. 1):197.(Abstr.)
S. T. CHESTER and J. W. HALLBERG The Upjohn Co. Kalamazoo. MI 49001 S. C. NICKERSON Louisiana State University Homer 71040 J. W. PANKEY University of Vermont Burlington O5405-0148 L. D. WEAVER University of California Visalia 93291 ABSTRACT
The prevalence of IMI was greatest during the last trimester of pregnancy, ranging from 49.2%in the winter to 36.8% in the summer. The significant effects of herd location and season suggest that management variables influence prevalence of heifer IMI. Because prevalence of IMI was greatest during the last trimester of pregnancy compared with prevalence during earlier stages of pregnancy, the heifer may be most susceptible to this disease during this period of first gestation. (Key words: mastitis, freshening, pathogens)
A survey was conducted to determine and contrast prevalence of IMI in nulligravid and primigravid dairy heifers pre- and postpartum. Contrasts were made to evaluate the risk factors of location of dairy, trimester of gestation, and season of sampling on IMI. Twentyeight dairies in California, Louisiana, Vermont, and Washington were studied. Lacteal secretions were collected aseptically from heifers at breeding age (8 to 19 mo) from one side of the gland and again at 4 d postpartum from all quarters. Of the quarters sampled, 65.6% prepartum and 64.0% postpartum were free of IMI. The percentages of quarters with IMI from coagulase-negative staphylococci or Staphylococcus aureus IMI were 27.1 and 9% prepartum and 2 1.8 and 2.9% postpartum. Staphylococcus aureus IMI were most prevalent in Louisiana during the months other than summer. Location, herd, and season significantly influenced prevalence of IMI.
Abbreviation key: CNS = coagulase-negative staphylococci. INTRODUCTION
Received July 13, 1994. Accepted November 18, 1994. ]This research was supported in part by the Upjohn Co., Kalamazoo, MI 49001. 1995 J Dairy Sci 78:1619-1628
Heifer mastitis IMI can be detected either pre- or postpartum (27) and is most often subclinical (2, 13). Although heifer mastitis has been recognized as a problem for >SO yr, dairy managers traditionally have viewed heifers at parturition as being free of IMI (24). Perhaps with the rise in expectations for improved control of mastitis, as evidenced by a decrease in the legal limit for bulk tank milk SCC under the Pasteurized Milk Ordinance, dairy managers and researchers have become more
1619
1620
FOX ET AL.
concerned with heifer mastitis. Nickerson and Boddie (15) have stated that a reduction in mastitis of heifers is critical because mammary growth and development is greatest during first pregnancy. Recently, a number of studies have estimated the prevalence of heifer IMI at first parturition (Table l), and these estimates varied widely. A large proportion of mammary quarters of heifers were reported to be free of IMI at calving in one study (20), but >75% were found to have IMI in another study (27). Comparison of results from these and other studies (Table 1) is confounded by problems associated with comparisons for prevalence of IMI from several herds at different locations, during different seasons, and in different years. Moreover, the results of these studies do not explain the risk factors associated with IMI at first parturition. One salient observation of Table 1 is that IMI by coagulase-negative staphylococci (CNS)is the most prevalent pathogen type associated with mastitis in heifers. These pathogens readily colonize multiple body sites of cattle (24). Roberson et al. (22) found an association between mammary colonization at breeding age (12 to 18 mo) and Staphylococcus aureus IMI at parturition. Thus, the breeding age period may be critical with respect to risk of IMI. The purpose of this study was to determine the prevalence of IMI at breeding age and at parturition of primiparous heifers. Effects of season of sampling, age, and stage of pregnancy at first sampling; herd and location were all risk factors considered to explain the variation in prevalence of IMI.
MATERIALS AND METHODS Herd Selection
Twenty-eight herds enrolled in DHI participated in the study: 7 from California, 5 from Louisiana, 7 from Vermont, and 9 from Washington. Herds were selected by investigators, and selection criteria included 1) a history of collaborative efforts, 2) availability of animal restraint facilities, 3) distance from the investigator's laboratory, and 4) herd size and annual calving pattern. The goal was to sample between 20 to 40 heifer calves of breeding age per herd per year, distributed evenly over four seasonal periods. Sample Collectlon
Heifers between the ages of 8 to 38 mo (mean = 19 mo; median = 19.4 mo) were randomly selected at the time of herd visits. Herd visits were made once per season @ecember through February was winter, March through May was spring, June through August was summer, and September through November was autumn). Heifers may have been pregnant at their first sampling; the second sample was conducted at parturition. Lacteal secretions were aseptically collected from one-half (right side or left side) of the udder at first sampling. Teat ends were immersed in a 1% iodophor solution and then scrubbed clean with a cotton swab soaked in 70% isopropyl alcohol. Secretion samples were collected in sterile test tubes. Samples were
TABLE 1. Prevalence of MI of primiparous heifers at fvst parturition. ~~
Reference
Nickerson et al. (15) Cook et al. (1) Oliver et al. (IT+ Pankey et al. (20) Roberson et al. (22)
Heifers sampled
Milk sample'
(no.) 600
Q
525
41 382 828
Prevalence of IMI by pathogen typeZ None
CNS
sa
58.4 43.0 55.4 81.7 54.0
27.9 43.0 39.0 11.4 39.0
8.0
Env
Other
(8) C
Q
Q C
4.2
6.0
.6 .7 8.0
4.9 4.8 13.0
1.4 8.0 1.7
IQ = Quarter; C = composite. V N S = Coagulase-negative staphylococci, Sa = Staphylococcus aureus, Env = environmental pathogens (coliforms and streptococci other than agalactiae). and Other = all other pathogens. 3IMI in heifers not receiving antimicrobial therapy prior to parturition to cure infections were only included. Journal of Dauy Science Vol. 78, No. 7, 1995
SYMPOSIUM: MASTITIS IN DAIRY HEIFERS
kept on ice at 4°C and then frozen or processed within 24 h. Following sample collection, teats were immersed in an acrylic latex solution containing 1% lauricidin (Teat Shield Germicide? 3M Co., St. Paul, h4N). Milk samples were aseptically collected from all functional mammary quarters using standard aseptic collection techniques (9) within 4 d of parturition by the dairy manager or designated employee. Signs of clinical mastitis were recorded at the time of postpartum sample collection.
1621
Statlstlcal Analysis
Not all prevalence data from all mammary quarters were included in the statistical analysis. Reasons for excluding data were contaminated samples, missing samples, samples collected >4 d postpartum, and samples with no data on age of the heifer. To analyze prevalence, isolated pathogens were grouped into four categories, and the following IMI dichotomies were created: infected compared with uninfected, infected with S. aureus compared with uninfected, infected Sample Analysis with a noncontagious pathogen compared with A .Ol-ml portion of all lacteal secretions uninfected, and infected with CNS compared was plated on blood agar plates and incubated with uninfected. These analyses were conat 37°C for 48 h. Isolates were presumptively ducted to identify the effects of location, seaidentified as staphylococci, streptococci, coli- son, gestation trimester, calving, and age at forms, or other genera based on colony charac- calving, as well as their interactions on the teristics (9). Mastitis pathogens were identified proportion of quarters infected in those according to standard protocols of the National dichotomies. The Pearson chi-square statistic Mastitis (9) and then enumerated. Pathogens was used to test the equality of infection rates were classified as contagious (S.aureus, and among herds within the same season and locaCorynebacterium bovis), environmental (coli- tion. Because these tests were statistically sigforms, streptococcal species, and enterococci), nificant, it was not possible to pool quarters CNS, or other. The significance of isolation of across herds and to perform traditional contina mastitis pathogen as the agent causing IMI gency table analyses. Because herds are a ranand classification of contaminants were deter- dom effect, an ANOVA was conducted using mined by the procedures outlined (14). Briefly, variation among herds in IMI by quarter as the the contagious pathogen S. aureus was considerror term. For each of the dichotomies, the ered to cause IMI if one colony (100 cfu/ml) proportion of IMI for each herd was computed was isolated. Isolation of >200 cfu of environand transformed via the Freeman-Tukey procemental mastitis pathogendm1 of milk or isolation of C. bovis in pure culture were dure (23). A weighted ANOVA was computed considered significant, and the isolated agent for which weights equaled n + 1/2, where n is considered to have caused the IMI. Isolation of the number of quarters associated with herd >loo0 cfu/ml of C. bovis was considered sig- proportion, and the form is as outlined in nificant if other bacteria also were present. Table 4 (23). For each step, a sequential analyMoreover, isolation of environmental patho- sis was conducted to determine the model. gens was considered significant when isolated Specifically, an initial model was fit including at >loo0 cfdml and when another bacterial all main effects and interactions, using the type was also isolated. Staphylococci other pooled herd variance as error. If the three-way than S.aureus were only considered significant interaction was not significant (P > .05), then if isolated in pure culture and >loo0 cfu/ml. A this interaction term was pooled with the error, priori, the investigators thought that it was unlikely that more than two types of pathogens and only two-way interactions and main efwould cause IMI. Thus, to differentiate be- fects were included in the model. If at least tween contaminated samples and quarters with one two-way interaction was not significant (P possible multiple infections, a mixed infection > .05), then a third analysis was performed, could have only two types of pathogens. Infec- pooling those nonsignificant two-way interactions were designated based on those deemed tions with the error. This process continued as significant isolations of mastitis pathogens until either the remaining interactions were according to the aforementioned criteria. statistically significant or only the main effects Journal of Daily Science Vol. 78, No. 7, 1995
1622
FOX ET AL.
were in the model. The Pearson chi-square statistic was used to test differences in IMI prevalences between herds within the same location, season, and time category. In a separate analysis, the hypothesis was tested that the effect of sampling at breeding age would predispose the mammary quarter to IMI. Within those cows with four functional quarters, the proportion of sampled quarters at breeding age that subsequently had IMI at parturition was contrasted with the proportion of unsampled breeding age mammary quarters having IMI at parturition. Additionally, the equivalence interval between IMI prevalences in the sampled and unsampled quarters was computed. RESULTS
Two thousand four hundred thirty-five quarter samples from 1583 breeding age heifers were used in the analysis. Based on the quar-
ters sampled, the majority of all heifers at breeding age appeared to be free of IMI (Table 2). Most of those quarters with IMI were caused by CNS (Table 2). Less than 3% of the quarters from heifers of breeding age had IMI by the contagious pathogen S. aureus; Sfreptococcus agalacriae were not found (Table 2). Usable milk samples from the heifers at first parturition numbered 4950. The majority of quarters were also free from IMI at parturition (Table 3). Similar prevalences were seen for IMI by S. aureus and CNS at breeding and at parturition (Tables 2 and 3). The prevalence of IMI in mammary quarters at parturition from those quarters sampled at first breeding age was 37.3%, which was significantly more (P = .07)than the 34.4% prevalence at parturition in mammary quarters that were not sampled at breeding age. However, the difference in IMI at parturition between sampled and unsampled udder halves at breeding age was within 5% of equivalence (P = .024),indicating that, al-
TABLE 2. Prevalence of intramammary infections (IMI)by season and location in bovine nulligravida and primigravid calves at breeding age.' Pathogen type3 Season Winter
Spring
Summer
Fall
Total
LocationZ CA LA VT WA CA LA VT WA CA LA VT WA CA LA VT WA
Uninfected
(96) (no.) 67.3 70 48.0 85 37 61.7 63.0 51 102 68.5 60.3 1430 33 67.3 78.8 82 205 68.1 73.4 278 74.4 64 66.8 133 75 60.0 55.5 106 70.7 65 68.3 69 65.6 1598
Staphylococcus aureus (no.) 0 13 0 1 .7 1 10.1 24 0 0 0 0 0 0 1.3 5 0 0 1.0 2 5 4.0 7.9 15 1.1 I 3.0 3 2.9 70
(5%)
0 7.3 0 1.2
Environmental (%)
0 3.4 5.0 1.2 .7 3.0 2.0 0 .3 1.3 1.2 2.0 0 3.1 1.1 0 1.5
(no.) 0 6 3 1 1 7 1 0 1 5 1 4 0 6 1 0 37
CNS3
(no.) 26 25.0 63 35.6 17 28.3 28.4 23 40 26.8 60 25.3 12 24.5 19.2 20 81 26.9 23.0 87 24.4 21 56 28.8 40 32.0 62 32.5 22 23.9 29 28.7 27.1 659
Other (%)
7.7 5.6 5.0 6.2 3.4 1.3 6.1 1.9 4.7 1.1 0 2.0 4.0 1.0
3.3 0 2.9
(no.) 8 10 3 5
5 3 3 2 14 4 0 4 5 2 3 0 71
'Lacteal secretions were collected from one side of the udder (right or left) from heifers at 8 to 38 mo of age. *Heifers from 7 dairies in California, 5 dairies in Louisiana, 7 dairies in Vermont, and 9 dairies in Washington were part of the study. Revalence is expressed as a percentage of mammary quarters sampled, number of positive samples is also listed. 3Environmental= Coliforms and environmental streptococci, CNS = coagulase-negative stmhvlococci. and other =
1623
SYMPOSIUM: MASTITIS IN DAIRY HEIFERS
-
-
70-
i'
I
E m
1
n
dl ~,I;,I ,
.
I
r/r
LA
WA
LOCATION
_1 UNINFECTED
I-1
~~
CNS
S AUREUS OMER
E UNINFECTED
EWRONMENTAL
ENVIRONMENTAL^
I S AUREUS
0OTHER
CNS
Figure 1. Prevalence of IMI in heifers of breeding age by location. S. aureus = Sraphylococcus aureus, environmental = coliforms and environmental streptococci, and other = all other types of pathogens.
Figure 3. Prevalence of IMI by season of year in breeding age heifers. S. aureus = Staphylococcus aureus, environmental = coliforms and environmental streptococci, and other = all other types of pathogens.
though IMI prevalences in sampled and unsampled mammary quarters were significantly different, the difference was small, 4%. Location of the heifers had a significant effect on prevalence of IMI both before and after parturition (Tables 2, 3, and 4). The prevalence of IMI in heifers in Louisiana was greatest prior to parturition, a result of a greater prevalence of IMI by S. aureus, environmental mastitis pathogens, and CNS (Figure 1). Following parturition, the prevalence of IMI in heifers was again greatest in Louisiana, which had the greatest percentage
of heifers with environmental and S. aureus IMI and the second greatest prevalence of CNS IMI (Figure 2). Season significantly influenced the prevalence of IMI following parturition, but not for heifers of breeding age (Table 4, Figures 3 and 4). The IMI were more prevalent in summer when IMI from S. aureus or environmental pathogens were most prevalent (Figure 4). At breeding age, the stage of pregnancy significantly affected prevalence of IMI and was highest for heifers in their last trimester of pregnancy (Figure 5). Specifically, prevalence
r
,
1
,
I '
L
li:
CA
0UNINFECTED
Winter
WA
S AUREUS
ENVlRONMENTAL
1
I \
Figure 2. Prevalence of IMI in heifers at first parturition by location S. aureus = Srclphylococcus aureus, environmental = coliforms and environmental streptococci, and other = all other types of pathogens.
i
= 0
Summer
SEASON
___________ UNINFECTED
IO CNS
111 Fall
Spring
S AUREUS
~
_
_
_
_
ENVIRONMENTAL
OTHER
Figure 4. Prevalence of IMI in heifers at first parturition by location. S. aureus = Staphylococcus aureus, environmental = coliforms and environmental streptococci, and other = all other types of pathogens. Journal of bairy Science Vol. 78, No. 7, 1995
~
1624
FOX ET AL.
TABLE 3. Prevalence of intramammary infections (JMI) by season and location in bovine primiparae immediately following parturition. Pathogen type2 Season Winter
SPW3
Summer
Fall
Total
Location' CA LA VT WA CA LA VT WA CA LA VT WA CA LA VT WA
Uninfected (96) 70.8 54.3 53.6 80.4 73.1 51.9 64.0 71.7 69.1 42.0 61.3 63.1 79.7 65.8 61.8 73.5 64.0
(no.) 213 350 67 160 335 209 16 185 325 198 49 207 310 239 105 200 3168
Smphylococcus aureus
(96) 1.3 1.6 .8 1.0 1.1 8.2 0 1.6 1.3 6.6 5.0 1.2 1.0 6.6 1.2 2.3 2.8
(no.) 4 10 I 2 5
33 0 4 6 31 4 4 4 24 2 6 140
Environmental (%)
6.0 17.8 5.6 2.5 3.5 9.2 0 0 2.3 21.0 7.5 4.3 2.8 6.9 8.2 2.6 7.7
(no.) 18 115 7 5 16 37 0 0 11 99 6 14 11 25 17 9 385
CNS (%)
19.3 20.8 33.6 10.6 20.3 27.8 36.0 25.2 24.5 24.6 22.2 26.5 14.9 19.0 27.1 15.1
21.8
Other (no.) 48 134 42 21 93 112 9 65 115 116 18 87 48 69 46 40 1083
(%)
2.7 5.6 6.4 5.5 2.0 3.0 0 1.6 2.8 5.7 3.8 4.9
1.5 1.7 1.8 4.5 3.5
(no.) 8 36 8 11 9 12 0 4 13 27 3 16 6 6 3 12 174
'Heifers from 7 dairies in California, 5 dairies in Louisiana. 7 dairies in Vermont, and 9 dairies in Washington were part of the study. Prevalence expressed as a percentage of mammary quarters sampled, and number of positive samples is also listed. 2Environmental = Coliforms and environmental streptococci, CNS = coagulase-negative staphylococci, and other = all other types of mastitis pathogens.
of CNS M I was significantly affected by trimester (Table 4) and, with the exception of fall, was substantially greater during the last trimester of pregnancy (Table 5). Season by trimester of pregnancy significantly influenced prevalence of IMI (Table 4). Prevalence of IMI precalving was lowest for second trimester heifers during the fall and next lowest for third trimester calves during the winter, spring, and fall months (Table 5). Differences were found in prevalence of IMI among herds. The test statistic pooled across all groups (location, season, and time category) was statistically significant (P c .OOol).
tors for this disease. Our study was conducted to determine how the effects of location, season, stage of pregnancy, and herd might affect IMI in heifers. Location significantly affected the prevalence of IMI in heifers. Location was an expected effect on prevalence of IMI in heifers, given the wide range in dairy management styles in the four states participating in this study. Moreover, within location, effect of herd on heifer IMI was significant. Management is the major factor controlling the incidence of IMI in lactating cows (21). More generally, the genetic effect on prevalence of mastitis has been estimated to be small relative to the environmental effect (25). Thus, the estimates for the lactating cow appear to hold DISCUSSION true for heifers as well. In the past, S. aureus Knowledge of how, why, and when an ani- IMI in heifers tended to be higher in Louisiana mal is predisposed to a disease is a necessary (Table 1). In this study, S. aureus prevalence prerequisite for establishing control strategies was severalfold greater in Louisiana than in and disease prevention. To date, a number of other locations (Figures 1 and 2). Previously, a studies have been made to estimate the preva- report from Washington (22) indicated a relalence of M I in heifers (Table I), but no studies tively high prevalence of S. aureus in heifers at have been conducted to identify the risk fac- calving of 8% (Table l), although, in our
Journal of Dairy Science Vol. 78, No. 7, 1995
1625
SYMPOSIUM: MASTITIS IN DAIRY HEIFERS
study, the mean prevalence in Washington herds was <3% (Table 2). Streptococcus agalacriae was not isolated from any heifer samples, which would seem unusual, given that Strep. agalactiae has been found in herds in each region (6, 7, 12, 19). This result indicates that the vectors of transmission of S. aureus in heifers, although similar to those vectors of transmission of other contagious mastitis pathogens in lactating cows, differ from those vectors of transmission of Strep. agalactiae among heifers. In addition to differences in management systems between locations, differences in climatic conditions might influence the incidence of IMI. For example, Smith et al. (26) reported an increase in IMI by environmental pathogens during the warm rainy months of summer in Ohio. Several reports (3, 10, 18) have established a problem with summer mastitis in Europe. Data from Fox and Hancock (4) indicated an increased prevalence of S. aureus IMI during acute cold weather, indicating that season influenced the prevalence of IMI. Indeed, season of sampling did significantly influence the
prevalence of IMI at parturition. At parturition, the number of IMI was greatest during the summer, primarily because of a relative increase in IMI by the staphylococci, S. aureus, and CNS (Figure 4). Presumably, season and climate did not have a direct effect on the mammary gland. Rather. conditions associated with climate exposed the gland to an increased bacterial load that management practices failed to control (21). Some of the conditions that were associated with increased prevalence among heifers at calving did not influence the prevalence of IMI among breeding age heifers, because prevalence of IMI within the latter group was not influenced by season. Nickerson et al. (16) suggested that flies may be responsible for the transmission of S.aureus to breeding age heifers. In that study, the prevalence of S. aureus IMI in breeding age heifers was lowest in the summer, presumably a period when a fly problem would be at the apex. Thus, either flies were not a major vector of S. aureus in Louisiana in this study, or the time necessary from transmission of S. aureus to establishment of IMI was of sufficient duration
TABLE 4. Tests for significance of factors affecting prevalence of IMI. Precalvingl df
Infected3
3 3 3 9 7 9 17 147 198
<.001 .317 <.001 ,407 ,111 ,018 ,804 1.9516
Postcalving2
CNS4
df
Infected
-P
-P Location Q Season ( S ) Times (T) L X S L x T S x T L x S x T Error Total
~~
CNS
.039 ,999 <.001 .587 .024
3
3
.045
,250 1.4236
2 9 6 6 18 182 229
C.001 <.@I ,732 .320 366 .471 ,367 2.8146
,002 ,001
.635 ,366 ,945 ,890 .559 2.4336
~~
'Lacteal secretions were collected from one side of the udder (right or left) from heifers at 8 to 19 rno of age *Heifers from 7 dairies in California, 5 dairies in Louisiana, 7 dairies in Vermont. and 9 dairies in Washington were part of the study. 3Mammary quarters were considered as being either infected or free of infection. 4Mammary quarters were considered as having a coagulase-negative staphylococcal (CNS) IMI, as having another infection, or as being uninfected. sPrecalving time is trimester of pregnancy: interval 1 = 1 to 95 d (n = 483). interval 2 = 96 to 190 d (n = 769), interval 3 = 191 to 285 d (n = 482); interval 4 (n = 515) includes all heifers with prolonged gestation periods, heifers not bred at the time of sampling, or heifers that had multiple conceptions between time of sampling and first calving. Postcalving time 1 = parturition at <24 mo (n = 1803); 2 = parturition at 24.5 to 26 mo (n = 1485); and 3 = parturition at >26 mo (n = I63 1). 6Mean square error. Journal of Dairy Science Vol. 78, No. 7, 1995
1626
FOX ET AL.
pregnant heifers were more prone to IMI than were nonpregnant heifers. However, the mean prevalence in this study for group 4 heifers was similar to those of groups 1 and 2 (Figure 5). Rather, heifers in their third trimester of pregnancy had the highest prevalence of M I . Pregnancy does affect mammary gland development, and thus a mammary gland of a gravid heifer would be physiologically much different from that of a nongravid heifer. Differences in mammary physiology have been associated with differences in susceptibility to IMI in lactating cows and could explain why differences might have existed in prevalence of IMI between gravid and nongravid heifers. Alternatively, the rate of mammary gland development varies with different stages of pregnancy; the greatest rate of growth occurs at the end of gestation (28). The more rapid development of the mammary gland in heifers in the third trimester of pregnancy appears to be as-
that new IMI would not be detected until the next sampling period. Indeed, the prevalence of S. aweus IMI in breeding age heifers was much greater in fall than in summer in Louisiana (Table 2). Trimester of pregnancy did have a significant effect on prevalence of IMI in heifers (Table 4). Most of the heifers in gestation category 4 of Table 5 were assumed to be not pregnant because their sampling to calving interval was >285 d. However, a few might have had a prolonged gestation period. Moreover, some heifers may have been in their first few days of pregnancy at the time of sampling or, alternatively, may have conceived earlier but, after loss of conceptus, were rebred. Accurate breeding records for heifers were not always available from the dairies studied. Therefore, all cows with extended intervals from sampling to calving in this fourth category were included. Evidence (3, 10, 18) suggested that
TABLE 5. Prevalence of intramammary infections by season and stage of pregnancy at first sampling.’ ~~
Pathogen type2 Season Fall
Spring
Summer
Winter
Stage
Uninfected
(dY
(96) 75.8 50.3 52.5 69.1 71.7 70.0 51.95 69.9 72.3 78.8 63.2 61.3 64.1 70.7 50.8 61.4
0-95 96190 191-285 Other 0-95 96-190 191-285 Other 0-95 96-190 191-285 Other 0-95 96-190 191-285 Other
(no.) 25 85 61 47 129 84 55 65 138 249 84 155 50 116 65 62
Staphylococcus aureus
(96) 0
6.5 2.6 0 5.6 5.0 8.5 0 0 1.3 0 1.2 2.6 4.9 6.3 5.0
(no.) 0
11 3 0 10 6 9 0 2 4 0
3 2 8 8 5
Environmental
CNS
6) (no.)
(%I
6.1 2.4 2.6 0 1.7 4.2 0 1.1 1.0 1.6 1.5 .8 3.8 1.8 .8
2 4 3 0 3 5 0
0
0
15.2 34.9 36.2 25.0 20.6 16.7 37.7 26.9 25.7 17.4 32.3 32.8 26.9 21.3 39.1 31.7
1
2 5 2 2 3 3 1
Other (no.) 5 59 42 17 37 20 40 25 49 55 43 83 21 35 50 32
@) 3.0 5.9 6.0 5.9 .6 4.2 1.9 2.2 1.0 .9 3.0 4.0 2.6 1.2 3.1 2.0
(no.) 1
10
7 4 1 5 2 2 2 3 4 10
2 2 4 2
‘Lacteal secretions were collected from one side of the udder (right or left) from heifers at 8 to 38 mo of age. Heifers from 7 dairies in California, 5 dairies in Louisiana, 7 dairies in Vermont, and 9 dairies in Washington were part of the study. 2Environmentalpathogens = Coliforms and environmental streptococci, CNS = all coagulase-negative staphylococci, and other = other mastitis pathogens. 3F’recalving time is trimester of pregnancy, 1 to 95 d = first trimester (n = 483), 96 to 190 d = second trimester (n = 769), and 191 to 285 d = third trimester (n = 482); other (n = 515) includes all heifers with prolonged gestation periods, heifers not bred at the time of sampling, or heifers that had multiple conceptions between time of sampling and first calving. Journal of Dairy Science Vol. 78, No. 7, 1995
SYMPOSIUM: MASTITIS IN DAIRY HEIFERS
1627
CONCLUSIONS
n
2 3 GESTATION PERIOD
~
~
__-_ _
0UNINFECTED 0CNS
S
-_ AUREUS
ENVIRONMENTAL '
0OTHER
Figure 5 . Prevalence of IMI (S. aureus = Sraphylococcus aureus, environmental = coliforms and environmental streptococci. and other = all other types of pathogens) by trimester of gestation [l = 1 to 95 d, 2 = 96 to 190 d, 3 = 191 to 285 d; 4 = heifers with prolonged gestation periods (>285 d), heifers not bred at the time of sampling, or heifers that had multiple conceptions between time of sampling and first calving].
sociated with increased prevalence of MI, as seen in the current study. Most of the increase in prevalence in IMI during the last trimester was due to CNS. Hogan et al. (11) indicated that these pathogens are opportunists and that heifers were more likely to have IMI by CNS than were multiparous cows. Hogan et al. (11) reported that, for all lactating cows, the prevalence of CNS IMI decreased as days of lactation increased and decreased during the dry period as the heifers began their second lactation. Presumably, dry cow therapy eliminated the remaining CNS IMI, as this group of pathogens is susceptible to dry cow therapy (8). Although there was a significant effect of sampling at breeding age on IMI at calving, the effect was small, and prevalence of IMI at breeding age and at parturition was similar. The lack of difference in prevalence would suggest few changes in IMI status from breeding age to parturition. The apparent lack of dynamics in IMI could be explained by the results of a study by Pankey et al. (20). Pankey reported that the number of new IMI are equivalent to the number of cures, thus giving the impression of IMI stasis between breeding age and parturition. Similar results were reported by Weaver et al. (29).
The hypothesized risk factors, season, location, trimester of pregnancy, and herd, all significantly affected the prevalence of IMI in dairy heifers. Management strategies associated with rearing of heifers appear to have major impact on control of this disease as evidenced by significant location and herd effects. Thus, as with mastitis in lactating cows, heifer IMI are controlled through management of heifer environment. Although results of this study did not indicate what management variables most influence mastitis IMI in heifers, the finding that the last trimester of pregnancy of heifers was more predisposed to IMI is significant. Clearly, future studies should concentrate on this critical period to determine what control practices could reduce the prevalence of IMI in heifers. ACKNOWLEDGMENTS
Appreciation is extended to J. R. Roberson for his help in sample collection, to Kenneth J. Dame for data organization, to personnel at the Washington State University dairies, and to all other cooperating dairies. REFERENCES 1 Cook, W. F., E. A. Fiez, and L. K. Fox. 1992. Mastitis in fmt lactation southwest Idaho dairy cows. J. Dairy Sci. 75(Suppl. 1):158.(Abstr.) 2 Daniel, R.L.W., D. A. Bamum, and K. E. Leslie. 1986. Observations on intramammary infections in first calf heifers in early lactation. Can. Vet. J. 27:112. 3Egan, J. 1990. Prevalence of summer mastitis in selected herds in the Province of Leinster. Ir. Vet. J. 43:74. FOX, L. K., and D. D. Hancock. 1989. Effect of segregation on prevention of intramammary infections by Staphylococcus aureus. J. Dairy Sci. 72540. 5 Fox, L. K., and J. R. Roberson. 1993. Heifer mastitis: is it a problem? Page 187 in Proc. Natl. Mastitis Counc., 32nd Annu. Mtg.. Kansas City, MO. Natl. Mastitis Counc., Inc., Arlington, VA. 6Goldberg. J. J., J. W. Pankey, P. A. Drechslesr, P. A. Murdough, and D. B. Howard. 1991. An update survey of bulk tank milk quality in Vermont. J. Food Prot. 54549. 7Gonzalez. R. N., D. E. Jasper, R. B. Bushnell, and T. B. Farber. 1986. Relationship between mastitis pathogen numbers and bulk tank milk in bovine udder infections in California dairy herds. J. Am. Vet. Med. Assoc. 189:442. 8 Harmon, R. J., W. L. Christ, R. W. Hemken, and D. E. Langlois. 1986. Prevalence of minor udder pathogens after intramammary dry treatment. J. Dairy Sci. 69: 843. Journal of Dairy Science Vol. 78, No. 7, 1995
1628
FOX ET AL
9Harmon. R. J., R. J. Eberhart. D. E. Jaspesr, B. E. Langlois, and R. A. Wilson. 1990. Microbiologcal procedures for diagnosis of bovine udder infection. Natl. Mastitis Counc., Inc., Arlington, VA. 10Hillerton. J. E., J.G.H. West, and M.F.H. Sheam. 1992. The cost of summer mastitis. Vet. Rec. 131:315. 11 Hogan, J. S., J. W. Pankey, and K. L. Smith. 1987. Stuphylococcus species, other than Staphylococcus uureus. Page 21 in Proc. Natl Mastitis Counc. 26th Annu. Mtg., Orlando, FL. Natl. Mastitis Counc., Inc., Arlington, VA. 12Hutton. C. T., L. K. Fox, and D. D. Hancock. 1990. Mastitis control practices: differences between herds with high and low milk somatic cell counts. J. Dairy Sci. 73:1135. 13 Munch-Petersen, E. 1970. Mastitis in bovine primiparae. Vet. Rec. 87:568. 14 National Mastitis Council. 1987. Laboratory and Field Handbook on Bovine Mastitis. W. D. Hoard and Sons, Fort Atkinson, WI. 15 Nickerson, S. C., and R. L. Boddie. 1992. Prevalence of heifer mastitis in Northwest Louisiana. Louisiana Cattlemen (Apr.):2. 16 Nickerson, S. C., W. E. Owens, and R. L. Boddie. 1995. Mastitis in dairy heifers: initial studies on prevalence and control J. Dauy Sci. 78:1599. 17 Oliver, S. P., M. J. Lewis, B. E. Gillespie, and H. H. Dowlen. 1992. Influence of prepartum antibiotic therapy on intramammary infections in primigravid heifers during early lactation. J. Dairy Sci. 75406. ISO’Rourke, D. J., R. J. Chamings, and J. M. Booth. 1984. Summer mastitis surveys in England and Wales: 1978-1983. Vet. Rec. 115:62. 19 Owens. W. E., and S. C. Nickerson. 1994. Use of bulk tank cultures for mastitis management. Dairy Res. Rep.. Hill Farm Res. Stn.,Homer, LA.
Journal of Dairy Science Vol. 78, No. 7, 1995
20 Pankey, J. W., P. A. Drechler, and E. Wildman. 1991. Mastitis prevalence in primigravid heifers at parturition. J. Dairy Sci. 74:1550. 21 Reneau, J. K. 1986. Effective use of dairy herd improvement somatic cell counts in mastitis control. J. Dairy Sci. 69:6. 22 Roberson, J. R., L. K.Fox, D. D. Hancock, C. C. Gay, and T. E. Besser. 1994. Coagulase-positive staphylococcus intramammary infections in primiparous dairy cows. J. Dairy Sci. 77:958. 23 SAS” User’s Guide: Statistics, Version 5 Edition. 1985. SAS Inst., Inc., Cary, NC. 24Shearer. J. K.,and R. J. Harmon. 1993. Mastitis in heifers. Vet. Clin. North Am. Food Anim. Pract. 9: 583. 25 Shook, G. E. 1993. Genetic improvement of mastitis through selection on somatic cell count. Clin. North Am. Food Anim. Pract. 9:563. 26Smith, K. L., D. A. Todhunter, and P. S . Schoenberger. 1985. Environmental mastitis: cause, prevalence, and prevention. J. Dairy Sci. 68:1531. 27 Trinidad, P., S . C. Nickerson, T. K.Alley, and R. W. Adkinson. 1989. Efficacy of intramammary treatment in unbred and primigravid dairy heifers. J. Am. Vet. Med. Assoc. 197:465. 28Tucker, H. A. 1981. Physiological control of mammary growth, lactogenesis, and lactation. J. Dairy Sci. 64:1403. 29 Weaver, L. D., C. A. Holmberg, J. S . Cullor. L. K. Fox, S . C. Nickerson, J. W. Pankey, J. Hallberg, and S. T. Chester. 1994. Effect of intramammary infection in dairy heifers, location, and season of calving on intramammary infection at calving. J. Dairy Sci. 77(Suppl. 1):197.(Abstr.)