Observed and expected combined effects of clinical mastitis and low body condition on pregnancy loss in dairy cows

m

a

Available online at www.sciencedirect.com

Theriogenology 77 (2012) 115–121

0d

Observed and expected combined effects of clinical mastitis andlow body condition on pregnancy loss in dairy cows

Jorge A. Hernandeza, Carlos A. Riscoa,*, Fabio S. Limab, and Jose E.P. Santosb

a Department of Large Animal Clinical Sciencesb Department of Animal Sciences, University of Florida, Gainesville, Florida, USA

Received 26 January 2011; received in revised form 10 June 2011; accepted 10 July 2011

Abstract

The objective was to compare the observed and expected combined effects of clinical mastitis before timed artificialinsemination (TAI) and low body condition at 70 d postpartum (dpp) on pregnancy loss in dairy cows. Cows were examined forpregnancy by ultrasonographic examination 28–32 d after TAI; the presence of an embryo with a heartbeat was the criterion usedto determine pregnancy. Cows diagnosed pregnant were re-examined by transrectal palpation of the uterus and its contents 28 dlater to confirm pregnancy status and to identify pregnancy loss. Eighty-eight (17%) of 512 cows were diagnosed with pregnancyloss. Cows affected with clinical mastitis before insemination and a body condition score (BCS) � 2.75 at 70 dpp were 2.03 times

ore likely to experience pregnancy loss, compared to cows without clinical mastitis and with a BCS � 2.75 (RR � 2.03; 95%CI � 1.15, 3.60; P � 0.01). This observed combined effect for pregnancy loss (RR � 2.03) was higher than the expectedcombined effect based on adding (RR � 1.39) or multiplying (RR � 1.42) absolute independent excesses due to clinical mastitisor low body condition.© 2012 Elsevier Inc. All rights reserved.

Keywords: Body condition score; Clinical mastitis; Pregnancy loss; Embryonic loss; Dairy cows

www.theriojournal.com

oac

llclmaigdvi

1. Introduction

Pregnancy loss can cause economic burden to thedairy farmer due to repeat breeding, a prolonged calv-ing-to-conception interval, or culling of dairy cows. Instudies conducted in dairy herds in the USA, the prev-alence of pregnancy loss in dairy cows varied from 6 to39% [1–3]. Known and suspected causes of pregnancyloss in dairy cows include genetic factors (deficiency ofuridine monophosphate synthase), infections causedby Leptospira spp, Campylobacter fetus venerealis,Trichomonas fetus, Ureaplasma spp, Mycoplasma spp,nd Hemophilus spp, environmental factors (heat stress

* Corresponding author. Tel.: 352-294-4387; fax: 352-392-7551.

093-691X/$ – see front matter © 2012 Elsevier Inc. All rights reserved.oi:10.1016/j.theriogenology.2011.07.023

r transrectal manipulation of the reproductive tract),nd toxic factors (mycotoxins, high blood urea nitrogenoncentration, teratogens, and nitrates) [3].

Several studies have identified mastitis [1–5] andow body condition [6,8] as risk factors for pregnancyoss in dairy cows. Based on in vitro studies, mastitisan cause embryonic mortality through the release ofipopolysaccharide (LPS), proteoglycans and otherolecules of bacterial origin that activate inflammatory

nd immune responses [9,10]. As a result, there is anncrease in cytokine synthesis from the mammaryland, and, perhaps, at other sites, including the repro-uctive tract [10]. Cytokines can affect pregnancy sur-ival by disrupting function of the hypothalamus, pitu-tary, ovary and uterus [10]. Reduced LH secretion may

116 J.A. Hernandez et al. / Theriogenology 77 (2012) 115–121

and thereby alter uterine function [10] resulting in em-bryonic mortality. The association between low bodycondition and pregnancy loss is less understood thantime to conception after calving. It is known that timeto conception is related to postpartum energy balancestatus. In a previous study [11], a short interval fromcalving to conception in dairy cows was associated witha small degree of energy balance nadir and a fastrecovery rate of energy balance from nadir to equilib-rium; therefore, it was inferred that the degree andtiming of negative energy balance determine the extentto which this condition limits luteal function in dairycattle.

In commercial dairy farms, cows can be affected byboth clinical mastitis and low body condition before orafter insemination. Although several studies have in-vestigated the independent effect of these two factorson pregnancy loss in dairy cows [1,3–8,12] their com-bined effect on pregnancy loss has not been examined.This knowledge is of importance, as it could improvereproductive management of dairy herds. The objectiveof the present study was to compare the observed andexpected combined effects of clinical mastitis beforeTAI and low body condition at 70 dpp on pregnancyloss in dairy cows.

2. Materials and methods

2.1. Study herds

One dairy herd from Florida (herd No. 1) and twodairy herds from California (herds No. 2 and No. 3) thatused the same TAI reproductive management protocolwere used for this study. Briefly, cows were presyn-chronized with two injections of PGF2�, given 14 dapart. Fourteen days after the last PGF2� injection,cows were enrolled in an Ovsynch program (Day 0GnRH; 7 d later, PGF2�; 56 h after PGF2� injection, 2nd

dose of GnRH; and 16 h after second GnRH cows wereTAI. Cows in the Florida herd received a controlledinternal progesterone-releasing insert (CIDR® EAZI-BREED™; Pfizer Animal Health, New York, NY,USA) followed by insert removal and GnRH adminis-tration 7 d later, 25 d after TAI. Cows in the three studyherds were examined by ultrasonography on Day 32after TAI; non-pregnant cows received PGF2� andGnRH 56 h later followed by TAI 16 h after the GnRHinjection.

The study in Florida was conducted between No-vember 2006 and March 2008 in a commercial dairy of2,200 Holstein cows. Cows were housed in free-stall

cooling during the warm season. Cows (n � 1055) thatcalved during November 2006–March 2008 were ini-tially considered for inclusion in the study. Lactatingcows bred by natural service were excluded (n � 512).Lactating cows bred by using a TAI protocol [13] wereincluded (n � 543); cows bred by TAI that were clas-sified as non-pregnant 32 d after TAI were excluded(n � 340). The final study enrollment in herd No. 1included 203 cows classified as pregnant.

The studies in California were conducted from No-vember 2005 to July 2006 (herd No. 2) and fromJanuary to October of 2005 (herd No. 3) [8]. In bothstudy herds, cows were housed in free-stall barnsequipped with fans and sprinklers for forced evapora-tive cooling during the warm season. A total of 147cows in herd No. 2 and 162 cows in herd No. 3 clas-sified as pregnant were enrolled in the study. Overall,the final enrollment for all three study herds was 512cows classified as pregnant following the first postpar-tum insemination.

In all study herds, diets were based on corn silage,alfalfa hay, alfalfa silage, steam-rolled corn, citrus pulp,corn distiller’s grains, whole cottonseed, calcium saltsof palm oil, and a mineral-vitamin-protein mixture.Lactating cow diets were formulated using the CPM-Dairy cattle ration analyzer (Cornell-Pen-Miner Ver.3.0.8, Cornell, University, Ithaca, NY, USA) to meet orexceed the nutrient requirements established by NRC(2001) [14] for lactating Holstein cows weighing 650kg, consuming 24 kg of DM/d, and producing 45 kg/dof milk containing 3.5% fat and 3.1% true proteinduring the first 80 d of lactation.

2.2. Study design

This investigation was designed as an observationalcross-sectional study. To accomplish the study objec-tive, the frequency of clinical mastitis, low body con-dition and other investigated exposure factors werecompared between 88 cows diagnosed with pregnancyloss 60 d after TAI and 424 cows that were not. In theanalysis, the observed combined effect of clinical mas-titis and low body condition on pregnancy loss wasadjusted for herd and breeding season effects.

2.3. Diagnosis of pregnancy and pregnancy loss

In herds No. 1, No. 2 and No. 3, transrectal ultra-sonographic pregnancy examinations were done at 32,30, and 28 d after TAI, respectively. The observation ofan embryonic vesicle with a viable embryo based onpresence of heartbeat was used to classify a cow as

aampc

2

f((nmbawmTd

2

wcamr

bwpeppaHwF

117J.A. Hernandez et al. / Theriogenology 77 (2012) 115–121

tive management protocols, cows classified as pregnantwere re-examined to confirm pregnancy by transrectalpalpation 28 d later [15]; failure to detect an embryonicvesicle on palpation was used as the criterion to classifya cow as having a pregnancy loss [15].

2.4. Diagnosis of clinical mastitis

In all three herds, cows were examined daily forclinical mastitis by farm personnel during each milkingduring the first 70 dpp. A case of mastitis was charac-terized by one or both the presence of abnormal milk orsigns of inflammation in one or more quarters. Cowswith systemic signs of illness, a hot swollen quarterwith a watery secretion received fluid therapy, anti-inflammatory agents, oxytocin for complete milk out,and antibiotics parentally. Cows with abnormal milk(flakes, pus) without sings of systemic illness receivedintra mammary antibiotic treatment. All treatments fol-lowed farm operational procedures established by theherd veterinarian.

2.5. Body condition score

In all three herds, cows underwent a BCS evaluationat 70 � 3 dpp by the same veterinarian (FSL) by using

scale of 1 to 5 [16]. In the analysis, cows weressigned into one of two groups (thin, BCS � 2.75;oderate or overconditioned, BCS � 3.00) based on a

revious study on embryonic survival in lactating dairyows [17].

.6. Data collection

Farm records were used to collect the following dataor each study cow: herd and cow identification, parityprimiparous, multiparous), retained fetal placentayes, no), metritis (yes, no), ketosis (yes, no), lame-ess (yes, no), respiratory disease (yes, no), clinicalastitis (yes, no), BCS at 70 � 3 dpp (1 to 5), and

reeding season (cool months, warm months). Theverage daily temperature– humidity index (THI)as calculated as previously described [8,13]. Theaximum daily THI was categorized as cool whenHI � 72, or warm when THI � 72. The THI on theay of TAI was used in the statistical analysis.

.7. Statistical analysis

The null hypothesis that number of cows affectedith pregnancy loss is the same in cows affected with

linical mastitis before insemination and a BCS � 2.75t 70 dpp, compared to cows non-affected with clinicalastitis and a BCS � 2.75 was tested with logistic

egression. Initial screening of clinical mastitis and low

ody condition and other investigated exposure factorsere performed using Poisson regression. Initially, ex-lanatory variables with a value of P � 0.20 with werentered into the model, and a forward stepwise ap-roach was used to identify variables associated withregnancy loss by use of two-sided P-values-to-enternd P-values to-remove of 0.05 and 0.10, respectively.erd, parity, breeding season, clinical mastitis and BCSere included as required variables in the final model.or all analyses, P-values � 0.05 were considered sig-

nificant.In this study, the risk ratio (RR) was used an epide-

miologic measure of association between an explana-tory variable (i.e., clinical mastitis) and the outcome ofinterest (pregnancy loss). In each variable, the referencecategory had an RR � 1. The RR was used instead ofthe odds ratio (OR) because the OR can over-estimatethe true effect when the outcome of interest (i.e., preg-nancy loss) is common.

The expected combined effect of clinical mastitisand low body condition on pregnancy loss in all studycows was calculated as follows: observed RR for clin-ical mastitis � RR for low BCS - 1 (additive model) orobserved RR for clinical mastitis � RR for low BCS(multiplicative model) [18]. In order to confirm if thecombined effect of clinical mastitis and low body con-dition on pregnancy loss was present in each studyherd, three models were examined (one for each studyherd).

Finally, the number of cows that experienced preg-nancy loss, that were inseminated in warm months, thatwere diagnosed with clinical mastitis, and that had aBCS � 2.75 was compared between herds by using achi-square test. Median number of days postpartumwhen cows were first diagnosed with clinical mastitiswas compared between herds by using a non-paramet-ric Kruskal-Wallis test. For all analyses, P � 0.05 wasconsidered significant.

3. Results

A total of 512 cows were enrolled in this study (203cows in herd No.1, 147 cows in herd No. 2. and 162cows in herd No. 3).

3.1. Pregnancy loss

Overall, 88 of 512 cows (17%; 95% CI � 14, 20)were diagnosed with pregnancy loss between 56 and60 d after TAI. The incidence of pregnancy loss as

cNi2(

3cp

pB0taenwmc(T(

e(c1(i

mechTmsaN

4

om7hobTh

neawpbTrossppiotswc

f

118 J.A. Hernandez et al. / Theriogenology 77 (2012) 115–121

significantly different between herd No. 1 (14%), herdNo. 2 (18%) and herd No. 3 (21%; P � 0.09).

3.2. Breeding season

Sixty-eight cows (13%; 95% CI � 10, 16) wereinseminated in warm months. All cows in herd No. 2were inseminated during cool months. The frequencyof cows inseminated in warm months differed (P �0.05) between herd No. 1 (59 cows or 78%) and herdNo. 3 (9 cows or 6%). Overall, 2 cows only (in herd No.1) which were inseminated in warm months were alsoaffected with clinical mastitis before insemination andhad a BCS � 2.75 at 70 dpp.

3.3. Clinical mastitis

Eighty cows (16%; 95% CI � 12, 19) were classi-fied as affected with clinical mastitis. The prevalence ofclinical mastitis differed (P � 0.05) among herds: 11cows (5%) in herd No. 1, 25 cows (17%) in herd No. 2,and 44 cows (37%) in herd No. 3. Median number ofdays postpartum when cows were first diagnosed withclinical mastitis was 11 d in herd No. 1, 25 d in herdNo. 2, and 44 d in herd No. 3, but these differenceswere not significant (P � 0.32).

3.4. Body condition

Two hundred and twenty-seven cows (44%; 95% CI �40, 49) had a BCS � 2.75 at 70 dpp. The prevalence ofows with BCS � 2.75 was 49% (99 cows) in herdo.1, 47% (69 cows) in herd No. 2, and 36% (59 cows)

n herd No. 3. The prevalence of cows with a BCS �.75 was different (P � 0.05) between herd No. 149%) and herd No. 3 (36%).

.5. Observed and expected combined effects oflinical mastitis and low body condition onregnancy loss

In the univariable analysis, the variables for herd,arity, lameness, respiratory disease, clinical mastitis,CS at 70 dpp, and breeding season had a P value �.20 (Table 1) and were further examined in the mul-ivariable analysis of pregnancy loss. In the multivari-ble analysis, we examined the observed combinedffect of clinical mastitis and BCS at 70 dpp on preg-ancy loss, and the analysis revealed that cows affectedith clinical mastitis and a BCS � 2.75 were 2.03 timesore likely to experience pregnancy loss, compared to

ows without clinical mastitis and with a BCS � 2.75RR � 2.03; 95% CI � 1.15, 3.60; P � 0.01) (Table 2).his observed combined effect for pregnancy loss

RR � 2.03) was higher than the expected combined d

ffect based on adding (RR � 1.39) or multiplyingOR � 1.42) absolute independent excesses due tolinical mastitis or low BCS (i.e., 1.24 � 1.15 � 1 �.39 or 1.24 � 1.15 � 1.42). As the expected joint RR1.42) is fairly close to the observed (2.03), interaction,f present, is weak on the multiplicative scale.

Finally, we examined the combined effect of clinicalastitis and low body condition for pregnancy loss in

ach study herd, and the analysis revealed that theombined effect for pregnancy loss did apply to studyerd No. 1, but it did not to herd No. 2 and herd No. 3.he observed and expected combined effects of clinicalastitis and low body condition were RR � 6.20 (ob-

erved) and 5.95 (expected) in herd No. 1; RR � 2.26nd 0.32 in herd No. 2; and RR � 1.35 and 1.69 in herdo. 3.

. Discussion

The objective of this study was to compare thebserved and expected combined effects of clinicalastitis before insemination and low body condition at

0 dpp on pregnancy loss in dairy cows on three dairyerds. Overall, cows were at high risk of pregnancy lossnly when they were affected with clinical mastitisefore insemination and low body condition at 70 dpp.his study finding, however, was observed in studyerd No. 1, but not in herd No. 2 and herd No. 3.

Eighty-eight of 512 cows (17%) classified as preg-ant between 28 and 32 d after AI by ultrasonographyxamination were diagnosed with a pregnancy loss 28 dfter. In this study, cows classified as non-pregnantere excluded; perhaps some cows in this group wereregnant after insemination, but they lost their embryoefore they were examined 32 d after insemination.herefore, perhaps the prevalence of pregnancy loss

eported in this study was underestimated. Comparisonf the prevalence of pregnancy loss observed in thistudy and that in previous studies must take into con-ideration differences in research methods. For exam-le, in two studies in California [2,3] the prevalence ofregnancy loss in dairy cows was 13% (45 d afternsemination in cows that were classified as pregnantn Day 30) and 18% (28 d after insemination in cowshat were classified as pregnant on Day 27). In onetudy in Ireland [6], the prevalence of pregnancy lossas 7% (28 to 84 d after insemination in cows that were

lassified as pregnant on Day 28).Eighty of 512 cows (16%) were classified as af-

ected with clinical mastitis before insemination. It is

6ipfikdpg

Bpieptdpn

fasa

H

P

R

M

K

L

R

C

B

B

R

AI) to

119J.A. Hernandez et al. / Theriogenology 77 (2012) 115–121

ical mastitis in this study and that in previous studies,which have examined the relationship between clinicalmastitis and pregnancy loss, because the time of expo-sure to clinical mastitis during lactation was different.For example, in a previous study in Florida [1] theprevalence of clinical mastitis was 60/2027 or 3% incows during first 45 d after insemination. In a study inCalifornia [3] the prevalence of clinical mastitis was/103 or 6% in cows during the first 35 d after insem-nation. In another study conducted in California [2] therevalence of clinical mastitis in 1465 cows during therst 45 d after insemination was not reported. To ournowledge, the effects of clinical mastitis occurring atifferent time intervals before or after insemination onregnancy loss in dairy cows have not been investi-

Table 1Univariable analysis of investigated exposure factors and risk of pre

Variable Pregnancy loss

Yesn � 88 (100%) n �

erd1 28 (32) 12 26 (30) 13 34 (38) 1

arityPrimiparous 22 (25) 1Multiparous 66 (75) 2

etained placentaNo 82 (93) 3Yes 6 (7)etritisNo 76 (86) 3Yes 12 (14)

etosisNo 81 (92) 3Yes 7 (8)

amenessNo 82 (93) 4Yes 6 (7)

espiratory diseaseNo 85 (97) 4Yes 3 (3)

linical mastitisNo 67 (76) 3Yes 21 (24)

CS at 70 dpp�2.75 43 (49) 2�2.75 44 (51) 1

reeding seasonCool 72 (82) 3Warm 16 (18)

R � risk ratio; 95% CI � 95% confidence interval; NA � not appa Cows diagnosed pregnant by ultrasound at d 32 (herd 1), 30 (

uterus and its contents 28 d later (ie, 56, 58, and 60 d after T

ated and reported. fl

Two hundred and twenty-seven cows (44%) had aCS � 2.75 at 70 dpp. It is difficult to compare therevalence of cows with a low BCS at 70 dpp observedn this study and that in previous studies that havexamined the relationship between body condition andregnancy loss in dairy cows. The time (days postpar-um) when the BCS evaluation was performed wasifferent between this and other studies [3,6,7] and therevalence of cows with a low BCS in other studies wasot reported [3,4,8].

In contrast with other studies [1–3], after controllingor parity and breeding season, mastitis alone was notssociated with pregnancy loss in study cows. In othertudies [1–3], clinical mastitis was diagnosed 35 to 45 dfter insemination. It is known that an increase in in-

lossa in dairy cows.

RR 95% CI P

0%)

) 1.00 Reference NA) 1.28 0.78, 2.09 0.31) 1.52 0.96, 2.39 0.07

) 1.00 Reference NA) 1.53 0.97, 2.39 0.06

) 1.00 Reference NA0.99 0.46, 2.12 0.99

) 1.00 Reference NA) 1.16 0.67, 2.01 0.57

) 1.00 Reference NA) 0.78 0.38, 1.59 0.49

) 1.00 Reference NA2.26 1.16, 4.40 0.01

) 1.00 Reference NA2.97 1.30, 6.76 � 0.01

) 1.00 Reference NA) 1.69 1.10, 2.59 0.01

1.00 Reference NA1.28 0.87, 1.87 0.20

1.00 Reference NA1.45 0.89, 2.33 0.12

; ND � not determined.

and 28 (herd 3) were reexamined by transrectal palpation of theconfirm pregnancy status and to identify pregnancy loss.

gnancy

No424 (10

75 (4421 (2628 (30

51 (3673 (64

95 (9329 (7)

75 (8849 (12

80 (9044 (10

14 (9810 (2)

20 (993 (1)

65 (8659 (14

41 (57)83 (43)

72 (88)52 (12)

licable

herd 2)

ammatory cytokines from the mammary gland can

S

M

nfirm p

120 J.A. Hernandez et al. / Theriogenology 77 (2012) 115–121

affect pregnancy survival by disrupting function of thehypothalamus, pituitary, ovary, and uterus [10]. How-ever, current knowledge in the scientific literature onthe dynamics of inflammatory cytokines in response tomastitis before or after insemination in dairy cows isvery limited. In one study, 10 cows were experimen-tally infected with Streptococcus uberis and inflamma-tory markers such as tumor necrosis factor; alpha(TNF-alpha) and interleukin 8 (IL-8) concentrationswere followed-up for 6 d after experimental challenge[19]. Peak concentrations of TNF-alpha and IL-8 oc-curred 5 d after challenge and coincided with onset ofclinical signs of mastitis. Furhtermore, TNF-alpha con-centrations dropped significantly 6 d after challengeand IL-8 concentrations remained high 6 d after chal-lenge. In this study, clinical cases of mastitis occurredbetween 11 and 44 dpp in study cows. Duration ofmastitis in study cows was not determined, which wasa limitation in this and other clinical studies [1–5] thatexamined the relationship between mastitis and preg-nancy loss. In this study, however, it seems unlikelythat most cows affected with mastitis remained sickuntil the time of insemination. Thus, although it ispossible for clinical mastitis alone to be a risk factor forpregnancy loss, we inferred that the event of mastitishas to occur at the time of insemination or thereafter.

In the current study, mastitis was only associatedwith increased risk of pregnancy loss when cows alsohad a low BCS at 70 dpp. To our knowledge, this was

Table 2Final logistic regression model for risk of pregnancy lossa in dairy c

Variable Pregn

Yesn � 88 (100%)

Herd1 28 (32)2 26 (30)3 34 (38)

ParityPrimiparous 22 (25)Multiparous 66 (75)

easonCool 72 (82)Warm 16 (18)astitis BCS at 70 dppNo �2.75 34 (38)No �2.75 32 (36)Yes �2.75 9 (12)Yes �2.75 12 (14)

a Cows diagnosed pregnant by ultrasound at d 32 (herd 1), 30 (herand its contents 28 d later (ie, 56, 58, and 60 d after TAI) to co

the first study that provided evidence that a combined

effect of clinical mastitis and low body condition (andnot one exposure alone) increased the risk of pregnancyloss in dairy cows. Based on previous studies in vitro[9,10], it was expected that mastitis can cause embry-onic mortality through the release of lipopolysaccharide(LPS), proteoglycans and other molecules of bacterialorigin that activate inflammatory and immune re-sponses [9,10]. These observations in vitro, however,did not apply to events of clinical mastitis that occurduring early lactation. Based on study results reportedhere, we offer one explanation by which an event ofclinical mastitis during early lactation might mediate areduction in maintenance of pregnancy in dairy cows:events of clinical mastitis during the first 10–50 dppcan induce a prolonged negative energy balance incows leading to losses of body weight and body con-dition days before or after insemination. An associationof mastitis events with severe drops in body weight andmilk yield in dairy cows during early lactation has beenreported [20]. It is known that cows with increasedlosses of body condition or cows with a low bodycondition at AI are more likely to experience embry-onic mortality [6,15,21,22].

Finally, in this study, the observed combined effectbetween clinical mastitis and low body condition onpregnancy loss was present in herd No. 1, but it wasabsent in herd No. 2 and herd No. 3. It is difficult toexplain the reasons for this observed difference be-tween herds. For example, the number of cows affected

ss RR 95% CI P

No� 424 (100%)

175 (41) 1.00 Reference NA121 (29) 1.80 1.00, 3.23 0.04128 (30) 1.88 1.08, 3.25 0.02

151 (36) 1.00 Reference NA273 (64) 1.54 0.97, 2.44 0.06

372 (88) 1.00 Reference NA52 (12) 2.24 1.25, 4.00 0.01

204 (48) 1.00 Reference NA161 (38) 1.15 0.73, 1.80 0.5337 (9) 1.24 0.64, 2.40 0.5122 (5) 2.03 1.15, 3.60 0.01

d 28 (herd 3) were reexamined by transrectal palpation of the uterusregnancy status and to identify pregnancy loss.

ows.

ancy lo

n

d 2) an

with clinical mastitis was significantly higher in herd

[

[

[

[

[

[

[

[

[

[

[

[

[

121J.A. Hernandez et al. / Theriogenology 77 (2012) 115–121

No. 3, compared to herd No. 1 and herd No. 2. How-ever, the number of cows affected with low body con-dition at 70 dpp was significantly lower in herd No. 3,compared to herd No. 1. Overall, the study reportedhere highlighted the need to better characterize theeffects of mastitis and low body condition, before andafter insemination, and the subsequent risk of preg-nancy loss in dairy cows. This information is importantto enhance our current understanding of the mecha-nisms that can cause pregnancy loss in dairy cows andfor a better reproductive management of dairy herds.

References

[1] Risco CA, Donovan GA, Hernandez JA. Clinical mastitis asso-ciated with abortion in dairy cows. J Dairy Sci 1999;82:1684–9.

[2] Chebel RC, Santos JEP, Reynolds JP, Cerri RLA, Juchem SO,Overton M. Factors affecting conception rate after artificialinsemination and pregnancy loss in lactating dairy cows. AnimReprod Sci 2004;84:239–55.

[3] Moore DA, Overton MW, Chebel RC, Truscott ML, BonDurantRH. Evaluation of factors that affect pregnancy loss in dairycattle. J Am Vet Med Assoc 2005;226;1112–8.

[4] Santos JEP, Cerri RLA, Ballou MA, Higginbotham GE, KirkJH. Effect of timing of first clinical mastitis occurrence onlactational and reproductive performance of Holstein dairycows. Anim Repro Sci 2004;80:31–45.

[5] Pinedo PJ, Melendez P, Villagomez-Cortes JA, Risco CA. Ef-fect of high somatic cell counts on reproductive performance ofChilean dairy cattle. J Dairy Sci 2009;92:1575–80.

[6] Silke V, Diskin MG, Kenny DA, Boland MP, Dillon P, KF Mee,Sreenan JM. Extent, pattern and factors associated with preg-nancy loss in dairy cows. Anim Repro Sci 2002;71:1–12.

[7] Lopez-Gatius F, Santolaria P, Yániz J, Rutllant J, López-BejarM. Factors affecting pregnancy loss from gestation day 38 to 90in lactating dairy cows from a single herd. Theriogenology2002;57:1251–61.

[8] Rutigliano HM, Lima FS, Cerri LA, Greco LF, Vilela JM,Magalhaes V, Silvestre FT, Thatcher WW, Santos JEP. Effectsof method of presynchronization and source of selenium onuterine health and reproduction in dairy cows. J Dairy Sci2008;91:3323–36.

[9] Soto P, Natzke RP, Hansen PJ. Identification of possible medi-ators of pregnancy mortality caused by mastitis: actions of

erator, sodium nitroprusside dehydrate, on oocyte maturationand pregnancy development in cattle. Am J Repro Immunol2003;50:263–72.

10] Hansen PJ, Soto P, Natzke RP. Mastitis and fertility in cattle –possible involvement of inflammation or immune activation inpregnancy mortality. Am J Repro Immunol 2004;51:294–301.

11] Reist M, Erdin DK, von Euw D, Tschumperlin KM, Leuen-berger H, Hammon HM, Morel C, Philipona C, Zbinden Y,Künzi N, Blum Jürg. Postpartum reproductive function: associ-ation with energy, metabolic and endocrine status in high yield-ing dairy cows. Theriogenology 2003;59:1707–23.

12] Barker AR, Schink FN, Lewis MJ, Dowlen HH, Oliver SP.Influence of clinical mastitis during early lactation on reproduc-tive performance of Jersey cows. J Dairy Sci 1998;81:1285–90.

13] Lima FS, Risco CA, Thatcher MJ, Benzquin ME, Archbald LF,Santos JEP, Thatcher WW. Comparison of reproductive perfor-mance in lactating dairy cows bred by natural service or timedartificial insemination. J Dairy Sci 2009;93:5456–66.

14] NRC. Nutrient Requirements of Dairy Cattle. 7th Rev, Ed., Natl.Acad. Sci. Washington, DC, 2001.

15] Zemjanis R. 1970. Diagnostic and therapeutic techniques inanimal reproduction. Williams and Wilkins Company, Balti-more, MD., p. 42–4.

16] Ferguson JD, Galligan DT, Thomsen N. Principal descriptors ofbody condition score in Holstein dairy cattle. J Dairy Sci 1994;77:2695–703.

17] Santos JEP, Rutigliano HM, Sá Filho MF. Risk factors forresumption of postpartum cyclicity and embryonic survival inlactating dairy cows. Anim Reprod Sci 2009;110:207–21.

18] Szklo M, Nieto FJ. Defining and assessing heterogeneity ofeffects: interaction. In: Epidemiology: beyond basics. Gaithers-burg, Maryland: Aspen Publishers, Inc; 2000, p. 211–51.

19] Rambeaud M, Almeida RA, Pighetti GM, Oliver SP. Dynamicsof leukocytes and cytokines during experimentally inducedStreptococcus uberis mastitis. Vet Immunolology Immunopa-thology 2003;96:193–205.

20] Maltz E, Devir S, Metz JHM, Hogeveen H. The body weight ofthe dairy cow: I. Introductory study into body weight changes indairy cows as a management aid. Livestock Prod Science 1997;48:175–86.

21] Lopez-Gatius F, Santolaria P, Yaniz J Lopez-Bejar M. Factorsaffecting pregnancy loss from gestation Day 38 to 90 in lactat-ing dairy cows from a single herd. Theriogenology 2002;57:1251–61.

22] Santos JEP, Thatcher WW, Chebel RC, Cerri RLA, Galvão KN.The effect of embryonic death rates in cattle on the efficacy ofestrous synchronization programs. Animal Reprod Science