See

discussions, stats, and author profiles for this publication at:

https://www.researchgate.net/publication/41808042

Risk factors for clinical endometritis in

postpartum dairy cattle

Article in Theriogenology March 2010

DOI: 10.1016/j.theriogenology.2010.01.023 Source: PubMed

CITATIONS READS

58 379

5 authors, including:

Timothy J Potter Javier Guitian

University of Surrey Royal Veterinary College
30 PUBLICATIONS 150 CITATIONS 122 PUBLICATIONS 1,720 CITATIONS

SEE PROFILE SEE PROFILE

John Fishwick Iain Martin Sheldon

Royal Veterinary College Swansea University
18 PUBLICATIONS 216 CITATIONS 135 PUBLICATIONS 4,712 CITATIONS

SEE PROFILE SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Javier Guitian
letting you access and read them immediately. Retrieved on: 21 November 2016
 Available online at www.sciencedirect.com

Theriogenology 74 (2010) 127134

www.theriojournal.com

Risk factors for clinical endometritis in postpartum dairy cattle

Timothy J. Potter a,*, Javier Guitian a, John Fishwick a,
Patrick J. Gordon b, I. Martin Sheldon a,c
a
Department of Veterinary Clinical Sciences, Royal Veterinary College, Hawkshead Lane, North Mymms, Hatfield AL9 7TA, UK
b
Shepton Veterinary Group, Shepton Mallet, Somerset BA4 4EZ, UK
c
Present address: School of Medicine, Swansea University, Swansea, SA2 8PP, UK
Received 23 September 2009; received in revised form 15 January 2010; accepted 31 January 2010

Abstract
Bacterial contamination of the uterine lumen after parturition occurs in most dairy cattle. The presence of clinical endometritis
beyond three weeks post partum depends on the balance between microbes, host immunity, and other environmental or animal
factors. The present study tested the hypothesis that clinical endometritis is associated with animal factors, such as retained fetal
membranes, assisted calving and twins, as well as fecal contamination of the environment. The association between selected risk
factors and the lactational incidence risk of clinical endometritis was examined in 293 animals from four dairy herds. Multivariate
analysis was used to identify risk factors and quantify their relative risk (RR) and population attributable fraction (PAF) based on the
proportion of cows exposed to each factor. The lactational incidence of clinical endometritis was 27% and significant risk factors for
clinical endometritis were retained fetal membranes (RR = 3.6), assisted calving (RR = 1.7), stillbirth (RR = 3.1), vulval angle
(RR = 1.3), primparity (RR = 1.8), and male offspring (RR = 1.5) but not the cleanliness of the environment or the animal. The
highest PAF was associated with male offspring (0.6) so the use of sexed semen has the greatest potential to reduce the incidence of
clinical endometritis. The dominant association between retained fetal membranes and clinical endometritis was supported by an
expert panel of clinicians. The risk factors for clinical endometritis appear to be associated with trauma of the female genital tract
and disruption of the physical barriers to infection rather than fecal contamination.
# 2010 Elsevier Inc. All rights reserved.

Keywords: Bovine; Clinical endometritis; Postpartum; Risk factors

1. Introduction passage of lochia out of the uterus, and through the

Bacterial contamination of the uterine lumen in the
first two weeks after parturition occurs in 8090% of dairy
cattle [1]. For several weeks after parturition there is a
cycle of bacterial contamination, clearance, and re-
contamination. In many animals this bacterial contam-
ination is gradually resolved by uterine involution,
* Corresponding author. Tel.: +441707666523;
fax: +441707666285.
E-mail address: tpotter@rvc.ac.uk (T.J. Potter).
mobilization of immune defenses. However, failure to
resolve the contamination can compromise uterine
function, and the persistence of pathogenic bacteria for
at least 3 weeks postpartum causes clinical endometritis
in 1020% of postpartum dairy cattle [14]. Clinical
endometritis is associated with tissue damage, delayed
uterine involution, disruption of endometrial function,
and perturbation of ovarian cycles [58]. Clinical
endometritis is characterized by a purulent discharge
from the uterus, which is often detected in the vagina of
affected animals [1,3,4]. Clinical endometritis causes
infertility at the time the disease is present, and

0093-691X/$ see front matter # 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.theriogenology.2010.01.023
128 T.J. Potter et al. / Theriogenology 74 (2010) 127134
subfertility even after successful resolution of the disease.
In several studies, the impact of clinical endometritis on
dairy herds was about 20% lower conception rate, 30 d
longer median calving to conception interval, and 3%
more animals culled for failure to conceive [1,3,4].
The bacteria associated with clinical endometritis
are Escherichia coli, Arcanobacterium pyogenes,
Fusobacterium necrophorum, and Prevotella species
[7,9,10]. It is assumed that these organisms are acquired
from the feces and fecal contamination of the animal
coat, bedding, and the environment. However, the role
of the animals environment and, in particular, whether
fecal contamination is a risk for clinical endometritis,
has often been neglected. This is unfortunate because a
key question for designing programmes to prevent
disease is whether animal or environmental factors are
most important. The risk factors that have been
previously established for uterine disease include
abnormal length of gestation, stillbirth, twins, assisted
parturition, retained fetal membranes, or a caesarean
section operation [1116]. However, in many studies,
the disease outcome that is examined is metritis during
the first 3 weeks postpartum, rather than clinical
endometritis, which is limited to disease after 3 weeks
postpartum [17]. In some studies the case definition of
the uterine disease encompasses metritis and clinical
endometritis. In others, risk factors for uterine disease
have only been identified individually by unadjusted or
univariate analysis of their association with the disease
The present study tested the hypothesis that clinical
endometritis is associated with animal factors, such as
retained fetal membranes, assisted calving, and twins,
as well as fecal contamination of the environment. We
determined the association between selected risk factors
and the lactational incidence risk of clinical endome-
tritis as defined by the presence of >50% purulent
material in the uterine discharge detectable in the vagina
21 d or more after parturition, or mucopurulent
(approximately 50% pus, 50% mucus) discharge
detectable in the vagina after 26 d postpartum [17].
In addition, expert veterinarian opinions of the risk
factors considered important for clinical endometritis
were examined. The purpose of the work is to generate
knowledge that will inform the design of control
programmes to prevent clinical endometritis.
2. Materials and methods
2.1. Animals
A prospective study enrolled all cows that calved
during 1 yr from 4 Holstein-Friesian dairy herds. The
convenience sample of herds in Somerset and
Hertfordshire was selected based upon 2 main
criteria. First, the farm had a good recording system
with veterinary involvement in management of data
using herd health computer software (Interherd;
NMR, Chippenham, UK). Second, the farm had an
established arrangement of weekly visits to the farm
by a veterinarian for fertility management. The herds
had all-year-round calving patterns with average
calving to first insemination intervals ranging from
7688 days, and calving to conception intervals of
92140 days in the year preceding the present study.
Cows received maize and grass silage while housed
during winter, and grazed grass in the summer.
Animals were enrolled at the weekly farm visits when
animals 2128 d postpartum were presented for
veterinary examination.
2.2. Clinical examination
Animals were examined by a veterinarian (TP) and
all procedures conformed to EC Directive 86/609/
EEC. The cleanliness of the hind quarters was
assessed and assigned a cow cleanliness score based
upon previously published criteria: (1) completely
free of dirt or has very little dirt; (2) slightly dirty; (3)
mostly covered in dirt; or (4) completely covered,
caked in dirt [18]. Fecal consistency for each animal
was assessed and assigned a numerical fecal
consistency score based upon the following published
criteria: (1) very dry, lumpy; (2) dry, stiff, semi-
formed pats; (3) circular, moist raised pat with petal
like symmetrical rings surrounding a dipped centre;
(4) flat, loose, thinly spread; or (5) liquid pools of
feces [19]. Body condition was assessed by examina-
tion of eight areas of the cows body, using
established criteria [20]. Briefly, animals were
assigned a body condition score along a 1 to 5 scale
with 0.25 unit increments; score 1 indicated
emaciated condition and score 5 an obese condition.
The angle of the vulva was recorded as abnormal if
less than 708 to the horizontal axis judged using a
protractor.
The presence of clinical endometritis is usually
determined by identifying pus in the lumen of the
vagina with a speculum or by withdrawing the contents
of the vagina by hand or with a Metricheck device
[7,2123]. There is a high level of agreement for
diagnosis of clinical endometritis between the speculum
method and manual examination of the contents of the
vagina [23]. The presence of vaginal pus also correlates
with the presence of polymorphonuclear cells detected
 T.J. Potter et al. / Theriogenology 74 (2010) 127134
by endometrial cytobrush [24], but the latter method
also detects subclinical as well as clinical endometritis.
The manual method to examine the contents of the
vagina does not cause iatrogenic bacterial contamina-
tion [25]. Furthermore, the vaginal mucus examined by
this method correlates with the presence and number of
pathogenic bacteria in the uterine lumen [26]. Briefly,
the vulva was cleaned thoroughly with a dry paper towel
before insertion of a clean, lubricated, gloved hand. The
vagina was assessed for any signs of damage or injury
by palpation of the lateral, dorsal, and ventral walls.
Animals with palpable vaginal injury (n = 6) were
excluded from the final data set due to the risk of
misidentifying vaginitis as endometritis. The mucus
content of the vagina was withdrawn manually for
examination. The vaginal mucus was characterized
using an endometritis scoring system [26]. Character
score was assigned as follows: unaffected animals: clear
or translucent mucus; score 1: mucus containing flecks
of white or off-white pus; 2: mucopurulent exudate
containing  50% white or off-white mucopurulent
material; and 3: exudate containing >50% purulent
material, usually white or yellow, but occasionally
sanguinous. Clinical endometritis was defined by the
presence of >50% purulent material (score 3) in the
uterine discharge detectable in the vagina 21 d or more
after parturition, or mucopurulent discharge (score 2 or
3) detectable in the vagina after 26 d postpartum [17].
For ethical and animal welfare reasons, animals were
treated intramuscularly with cloprostenol (Estrumate;
Schering Plough Animal Health, Welwyn Garden City,
UK.) or infused intrauterine with Cephapirin (Metri-
cure; Intervet UK Ltd., Walton, UK.) until normal
vaginal mucus was detected in the vagina.
2.3. Farm, clinical, and fertility data
Farm and fertility data were recorded by farm staff
and entered into a herd health monitoring computer
programme (Interherd; NMR, Chippenham, UK).
Information on the animals included parity, milk yield,
somatic cell count, and month of calving. Information
on the periparturient period was recorded, including
cases of hypocalcaemia, retained fetal membranes,
calving assistance, and other disease conditions:
lameness, mastitis, and abdominal disorders. Records
also included information on the calf sire (beef or
dairy), status at delivery (alive or stillborn), calf sex, and
the presence of a twin. Records were also kept of the
wearing of arm-length gloves when performing vaginal
examinations during parturition and the disinfection of
calving equipment.
129
2.4. Expert opinion
A survey was prepared to assess current expert
opinion on the relevance of the selected risk factors for
endometritis. Experts were veterinarians selected based
upon achievement of a United Kingdom postgraduate
qualification in bovine reproduction or cattle health and
production. Surveys were distributed by post to all
eligible veterinarians. Participants were asked to place
in rank order what they considered to be the most
important of 21 risk factors for clinical endometritis
(see Results section for list of factors).
2.5. Statistical analysis
The lactational incidence risk of clinical endome-
tritis was calculated for each farm as the proportion of
cows recruited in the farm that developed clinical
endometritis. The distribution of clinical endometritis
scores was compared between farms by means of the x-
squared test of association. Descriptive statistics for
qualitative and quantitative variables under study were
computed, before odds ratios (OR) were estimated
using logistic regression. The risk factors tested were:
farm identity (14), parity, month of calving, calf sire
(beef or dairy), status at delivery (alive or stillborn),
calving assistance, calf sex, the presence of a twin,
wearing of arm-length gloves when performing vaginal
examinations during parturition, disinfection of calving
equipment, lactation milk yield, somatic cell count, cow
cleanliness score, fecal consistency score, body condi-
tion score, angle of the vulva, and the occurrence of
hypocalcaemia, retained fetal membranes, lameness,
mastitis or abdominal disorders.
The model-building strategy involved two steps [27].
In the first step, all variables were screened using
univariable conditional logistic regression with the data
grouped by herd. Only variables significant at P < 0.35
by Walds test (two-tailed) were included in the second
step of the analysis. Before continuing with the second
step of the analysis, associations between exposure were
examined, and where a pair of variables was
significantly associated (P < 0.05 by x2-test, two-
tailed) and both with P < 0.25, the variable judged as
most biologically plausible was used as a candidate in
the multivariable analysis. The subsequent selection of
the most parsimonious multivariable model was
achieved by backward stepwise elimination of non-
significant (P > 0.1, two-tailed tests) factors. The
analyses were carried out using version 9.0 of the
statistical package Stata (Stata Corporation, College
Station, TX). The logistic regression models produced
130 T.J. Potter et al. / Theriogenology 74 (2010) 127134

odds ratios as estimates of the strength of association Table 1

between the potential risk factors and endometritis. Univariate conditional logistic regression Odds Ratios (OR) and
statistical probability (P) values for risk factors associated with
However, because endometritis was common in the clinical endometritis that were eligible for subsequent multivariate
study population (80 of the 293 cows studied), the analysis.
adjusted OR are not a good approximation of the risk
Variable OR P
ratio and so an established correction [28] was used to
obtain the risk ratio, where P0 is the lactational Retained fetal membranes 34.29 < 0.001
Stillbirth 7.94 < 0.001
incidence of endometritis:
Twins 4.98 0.003
Assisted calving 2.83 0.002
RR OR=1  P0 P0  OR Primiparity 2.03 0.021
Mastitis 1.77 0.225
From the risk ratio and the proportion of cows exposed Hyopcalcaemia 1.56 0.349
to the different risk factors (Pe) the Population Attrib- Male calf 1.40 0.215
utable Fraction [28] was calculated as an estimate of the Fecal consistency score a 1.27 0.349
potential impact of single risk factors on endometritis Vulval angle 0.96 0.085
Cow cleanliness score a 0.83 0.294
occurrence in the herd, assuming that their association
Dairy vs beef cross calf 0.70 0.254
with endometritis is causal:
a
Included as linear effect since no departure from linearity was
PAF RR  1=Pe RR  1 1 detected by LR test.

3.2. Analysis of risk factors for endometritis

3. Results
3.1. Incidence and impact of endometritis
A total of 293 animals were recruited in the study and
the average lactational incidence of endometritis was
27%. The farm incidence of endometritis was: farm 1,
24/75 (32%); farm 2, 18/65 (28%); farm 3, 18/77 (23%);
and farm 4, 20/76 (26%). The distribution of clinical
endometritis scores did not differ significantly between
the four farms (Fig. 1; P > 0.05).
Fig. 1. Spine plot to indicate the distribution of clinical endometritis
scores between animals for each of the four study farms. The clinical
endometritis score was based on examination of the vaginal mucus
character as previously described [26]. Briefly, unaffected animals
(&) had clear or translucent mucus; score 1 ( ): mucus containing
flecks of white or off-white pus; 2 ( ): mucopurulent exudate contain-
ing  50% white or off-white mucopurulent material; and 3 (&):
exudate containing >50% purulent material, usually white or yellow.
The number of animals on each farm is indicated (N).
The odds ratios for the association between single
potential risk factors and clinical endometritis, derived
from univariate conditional logistic regression models
and eligible for multivariate analysis are presented in
Table 1. The following remaining factors were not
significant (P > 0.35) and had odds ratios close to 1:
body condition score, month of calving, lameness,
abdominal disease, disinfection of calving equipment,
gloves worn during partition examination, milk yield or
somatic cell count.
For the multivariate analysis, several significant
correlations were evident among the risk factors for
endometritis identified during univariate analysis
(Table 2). Taking these correlations and biological
plausibility into consideration, two sets of non-collinear
but significant variables were identified and a model
was fit for each of them. The first model included:
vulval angle, calving assistance, a live calf, calf sex,
retained foetal membranes, and mastitis. The second
model included the variables; vulval angle, parity, calf
sex, retained foetal membranes, and mastitis. The two
final models are presented in Table 3. Markers of
hygiene including the fecal consistency score, cow
cleanliness score, disinfection of calving equipment or
the wearing of gloves when assisting parturition, were
not significant in the univariate or multivariate models.
3.3. Survey results
A total of 75 surveys were sent out to veterinarians
with postgraduate qualification in bovine health and
 T.J. Potter et al. / Theriogenology 74 (2010) 127134 131

Table 2
Correlation matrix showing correlation coefficients for pairs of variables that were significantly associated (NS, not significant, P > 0.05).
Vulval Parity Calving Multiple Stillbirth Male Retained fetal Mastitis
angle assistance birth sex membranes
Vulval angle X NS NS NS NS NS NS NS
Parity X - 0.15 0.10 0.08 NS NS NS
Calving assistance X 0.12 0.20 NS 0.1 NS
Multiple birth X NS NS 0.24 NS
Stillbirth X 0.12 NS NS
Male sex X NS NS
Retained fetal membranes X NS
Mastitis X

Table 3
Results of two conditional logistic regression models for the association between selected potential risk factors and endometritis. The odds ratios
(OR) and the derived risk ratios (RR) with 95% confidence intervals (CI) are presented, as well as the proportion of cows exposed to each risk factor
(Pe) used to derive the population attributable fractions (PAF) as described (24).
Model Variable OR SE P-value 95%CI RR 95%CI Pe PAF
(OR) (RR)
1 Retained fetal membranes 40.3 42.1 0.001 5.0-326.4 3.96 2.58-4.24 0.042 0.148
Stillbirth 7.5 3.5 0.001 3.0-18.5 3.09 2.09-3.83 0.091 0.236
Assisted calving 2.1 0.8 0.064 1.0-4.4 1.67 0.97-2.51 0.206 0.302
2 Retained fetal membranes 43.5 46.6 0.001 5.3-355.8 3.97 2.65-4.23 0.042 0.148
Primiparity 2.3 0.8 0.009 1.2-4.4 1.78 1.17-2.46 0.259 0.384
Male calf 1.8 0.6 0.044 1.0-3.3 1.54 1.01-2.15 0.528 0.632
Vulval angle 1.5 0.4 0.086 0.9-2.4 1.31 0.96-1.71 0.084 0.107

Table 4 reproduction. A total of 58 responses were received; a

Ranking of 21 perceived risk factors for endometritis (most important response rate of 77.3%. Table 4 displays the rank order
first) based on postal survey of current experts (n = 57).
of risk factors considered by the experts to be important
Retained fetal membranes for endometritis. Three of the top four risk factors
Dystocia
considered most important by the experts matched those
Twins
Stillbirth from the multivariate model.
Hypocalcaemia
Induced Calving 4. Discussion
Vaginal trauma at the time of calving
Low Body Condition Score
Cattle are particularly susceptible to contamination
High Body Condition Score
Dirty Calving Equipment of the uterine lumen with bacteria after parturition, with
Vitamin E/Selenium Defiiciency 8090% of dairy cattle affected [7,9]. More importantly,
Vulval Conformation 1020% of animals have clinical endometritis, which
Calving Indoors compromises animal welfare, causes infertility, and
Insufficient Structural Fibre in Diet
incurs economic costs for the dairy industry [1,3]. The
Parity
Fecal consistency score development of clinical disease depends on the balance
Breed of sire between the pathogenic organisms that contaminate the
Staff not wearing gloves when examining/assisting calving animals uterine lumen, the host immune response, and the
Movement between management groups at the time of calving environment. The identity of the bacteria that con-
Breed
taminate the uterine lumen, their pathogenicity and host
Calving Outside
immune responses have been characterized and appear
132 T.J. Potter et al. / Theriogenology 74 (2010) 127134
to be consistent across the world [7,9,10,29,30]. It is
assumed that these bacteria come from the skin, feces,
and fecal contamination of the environment. The
present study tested the hypothesis that clinical
endometritis is associated with animal factors, such
as retained fetal membranes, assisted calving, and
twins, as well as fecal contamination of the environ-
ment. The most significant risk factors included retained
fetal membranes, assisted calving, stillbirth, vulval
angle, primiparity, and male offspring. Factors asso-
ciated with hygiene or fecal contamination were not
significantly associated with clinical endometritis.
These observations lead us to suggest that the design
of control programmes for uterine disease should focus
on sire selection, sexed semen, and the management of
cattle around the time of parturition rather than their
hygiene.
Retained fetal membranes were the most important
risk factor for clinical endometritis in the two multi-
variate models derived in the present study. The risk of
retained fetal membranes for uterine disease has been
established previously [14,16]; and was also recognized
as the most important factor by the expert panel. The
presence of necrotic material in the uterine lumen
associated with retained fetal membranes likely
provides a favourable environment for bacterial growth.
Furthermore, retained fetal membranes maintain the
patency of the cervix, which is a physical barrier to
infection in unaffected animals. Retained fetal mem-
branes also delay uterine involution, expulsion of lochia
and the regeneration of the endometrium [31].
Unfortunately, the causes of retained fetal membranes
are multifactorial, mostly unquantified, and unlikely to
respond to simple control measures [15,32]. Indeed,
there may be a genetic component to retained fetal
membranes that would require considerable effort to
address [33]. Another possibility is a deficit in immune
function because cows with retained fetal membranes
have impaired neutrophil function, which may com-
promise their ability to reject the placenta after
parturition [34].
Factors other than retained fetal membranes that
were identified as significant risk factors for clinical
endometritis from the multivariate analysis were
calving assistance, stillbirth, a male calf, a vulval
angle < 708 to the horizontal, and a primiparous dam.
These observations also agree with previous reports
[13,14,35]. Many of the significant factors identified in
the present analysis, including calving assistance,
stillbirth, a male calf, and a primiparous dam are likely
to be associated with some degree of trauma to the
female genital tract around the time of parturition.
Indeed the size of the calf implied by a male calf as a
risk for disease in Model 2 may be mediated by calving
assistance risk factor in Model 1. The correlation matrix
for the multivariate modelling indicated associations
between primiparity and calving assistance or stillbirth,
probably related to the smaller pelvis of such animals
than more mature cows. Calving assistance also tended
to be associated with twins and retained fetal
membranes, and as expected the birth of twins was
highly correlated with retained fetal membranes.
Although the odds ratios are indicated in the
multivariate models, the relative risk and population
attributable fraction were also calculated because
clinical endometritis was common in the study
population [28]. The population attributable fraction
takes into account the prevalence of the particular risk
factors along with the relative risk associated with that
risk factor, this allows for the investigation of the
potential role of a commonly occurring risk factor
which has only a small impact on the incidence of
disease compared to an infrequently occurring risk
factor that has a large impact on disease incidence. The
population attributable fraction represents the propor-
tion of cases of clinical endometritis that could be
prevented by removing the risk factor, assuming
causality. Therefore, considering model 2, avoiding
the birth of male calves, perhaps by using sexed semen,
could have a substantial impact on clinical endometritis,
dependent on the degree of causality. The idealized
interpretation of a population attributable fraction of 0.6
for male calves is that 60% of cases in the study
population would be prevented if all calves were female
and if the distribution of other risk factors would not
change as a result of the elimination of this factor. On
the other hand, the remaining factors in model 2
retained fetal membranes, primiparity, and vulval
angleare not amenable to control by the farmer or
veterinarian. In model 1, assisted parturition and
stillbirth had the greatest population attributable
fraction. While these factors may seem difficult to
address for the farm management, the widespread use of
artificial insemination in dairy herds and the selection of
easy-calving sires could rapidly reduce the risk of
clinical endometritis in dairy herds.
The bacteria that cause uterine disease such as E. coli
and A. pyogenes are likely derived from the environ-
ment that cattle are kept in or from the animals skin [1].
However, a key set of observations, even from the
univariate analysis was that clinical endometritis was
not significantly associated with the cleanliness of the
animals or fecal consistency. These observations are
surprising at an intuitive level but support a previous
 T.J. Potter et al. / Theriogenology 74 (2010) 127134
study, which concluded that the cleanliness of the farm
did not influence the incidence of clinical endometritis
[36]. As the cleanliness of the animals was not a
significant risk factor within each farm, presumably
sufficient bacteria are available to contaminate the
uterus and establish disease on most farms. The next
step to consider in the pathogenesis of uterine disease is to
evaluate how these bacteria reach the uterine lumen, so it
is not surprising that human assistance with parturition is
a risk factor for uterine disease. However, whether the
farmer wore protective gloves for assisting the animal
was not significant. This observation leads us to speculate
that the farmers skin is not a source of bacteria for the
cow, but that the bacteria come directly from the
environment or cows skin or feces. The other factor
identified in the multivariate analysis that could directly
relate to fecal contamination was the angle of the vulva.
Perhaps the greater fecal contamination of the less
vertical vulva leads to feces entering the vagina and so
ascending bacterial infection of the female genital tract.
The expert panel also ranked retained fetal mem-
branes, assisted calving, twins, and stillbirth as
important for clinical endometritis. However, the
experts also considered hypocalcemia and body
condition score relatively important for clinical
endometritis, but this did not match the findings in
the present study. However, other studies have reported
that metabolic factors may have a role in uterine disease
[14,37].
The 27% lactational incidence of clinical endome-
tritis was greater for each farm than the the usually
quoted figure of 1020% [1,22]. However, a high
incidence of clinical endometritis is not infrequently
reported when animals are examined carefully [14,38].
Furthermore, there appears to be an increasing number
of dairy cattle treated for clinical endometritis [39]. The
mucus character scores showed a similar pattern across
the farms with score 3 most common, agreeing with
previous observations on 19 farms [40].
In conclusion, the present study sought to determine
which risk factors are important for clinical endome-
tritis in dairy cattle. Retained fetal membrane was the
most important factor and its dominant association with
uterine disease is consistently supported by the
literature and our clinical expert panel. Other risk
factors for clinical endometritis included calving
assistance, stillbirth, primiparity, male offsping, and
vulval angle. All the significant risk factors appear to be
associated with trauma of the female genital tract and/or
disruption of the physical barriers to infection, rather
than the cleanliness of the animal. Control programmes
should focus on the selection of appropriate sires and
133
management of animals around the time of parturition
to limit damage to the female genital tract.
Acknowledgments
This study was supported by the Royal College of
Veterinary Surgeons Trust. Martin Sheldon is a BBSRC
Research Development Fellow. The authors thank Clive
Keevil, Steve Craddock, Graham Bicknell, Charlie
Verity, and Frances Stephenson for technical assistance.
TJP, PJG, JF, and IMS designed the research; TJP and
PJG performed the study; TJP and JG analysed the data;
TJP, JG, IMS, and JF wrote the paper.
References
[1] Sheldon IM, Cronin J, Goetze L, Donofrio G, Schuberth HJ.
Defining Postpartum Uterine Disease and the Mechanisms of
Infection and Immunity in the Female Reproductive Tract in
Cattle. Biol Reprod 2009;81:102532.
[2] Gilbert RO, Shin ST, Guard CL, Erb HN. Incidence of endome-
tritis and effects on reproductive performance of dairy cows.
Theriogenology 1998;49:251.
[3] Borsberry S, Dobson H. Periparturient diseases and their effect
on reproductive performance in five dairy herds. Vet Rec
1989;124:2179.
[4] LeBlanc SJ, Duffield TF, Leslie KE, Bateman KG, Keefe GP,
Walton JS, Johnson WH. The effect of treatment of clinical
endometritis on reproductive performance in dairy cows. J Dairy
Sci 2002;85:223749.
[5] Herath S, Lilly ST, Fischer DP, Williams EJ, Dobson H, Bryant
CE, Sheldon IM. Bacterial lipopolysaccharide induces an endo-
crine switch from prostaglandin F2a to prostaglandin E2 in
bovine endometrium. Endocrinology 2009;150:191220.
[6] Bonnett BN, Martin SW, Meek AH. Associations of clinical
findings, bacteriological and histological results of endometrial
biopsy with reproductive performance of postpartum dairy cows.
Prev Vet Med 1993;15:20520.
[7] Sheldon IM, Noakes DE, Rycroft AN, Pfeiffer DU, Dobson H.
Influence of uterine bacterial contamination after parturition on
ovarian dominant follicle selection and follicle growth and
function in cattle. Reproduction 2002;123:83745.
[8] Opsomer G, Grohn YT, Hertl J, Coryn M, Deluyker H, de Kruif
A. Risk factors for post partum ovarian dysfunction in high
producing dairy cows in Belgium: a field study. Theriogenology
2000;53:84157.
[9] Griffin JFT, Hartigan PJ, Nunn WR. Non-specific uterine infection
and bovine fertility. I. Infection patterns and endometritis during
the first seven weeks post-partum. Theriogenology 1974;1:91106.
[10] Huszenicza G, Fodor M, Gacs M, Kulcsar M, Dohmen MJW,
Vamos M, Porkolab L, Legl T, Bartyik J, Lohuis JACM, Janosi S,
Szita G. Uterine bacteriology, resumption of cyclic ovarian
activity, and fertility in postpartum cows kept in large-scale
dairy herds. Reprod Dom Anim 1991;34:23745.
[11] Hussain AM, Daniel RCW, OBoyle D. Postpartum uterine flora
following normal and abnormal puerperium in cows. Therio-
genology 1990;34:291302.
[12] Peeler EJ, Otte MJ, Esslemont RJ. Inter-relationships of peri-
parturient diseases in dairy cows. Vet Rec 1994;134:12932.
134 T.J. Potter et al. / Theriogenology 74 (2010) 127134
[13] Markusfeld O. Periparturient traits in seven high dairy herds.
Incidence rates, association with parity, and interrelationships
among traits. J Dairy Sci 1987;70:15866.
[14] Kim IH, Kang HG. Risk factors for postpartum endometritis and
the effect of endometritis on reproductive performance in dairy
cows in Korea. J Reprod Devel 2003;49:48591.
[15] Han IK, Kim IH. Risk factors for retained placenta and the effect
of retained placenta on the occurrence of postpartum diseases
and subsequent reproductive performance in dairy cows. J Vet
Sci 2005;6:539.
[16] Paisley LG, Mickelsen WD, Anderson PB. Mechanisms and
therapy for retained fetal membranes and uterine infections of
cows: A review. Theriogenology 1986;25:35381.
[17] Sheldon IM, Lewis GS, LeBlanc S, Gilbert RO. Defining
postpartum uterine disease in cattle. Theriogenology 2006;65:
151630.
[18] Schreiner DA, Ruegg PL. Effects of tail docking on milk quality
and cow cleanliness. J Dairy Sci 2002;85:250311.
[19] Hughes J. A system for assessing cow cleanliness. In Pract
2001;23:51724.
[20] Edmondson AJ, Lean IJ, Weaver ID, Farver T, Webster G. Body
condition scoring chart for Holstein dairy cows. J Dairy Sci
1989;72:6878.
[21] Runciman DJ, Anderson GA, Malmo J. Comparison of two
methods of detecting purulent vaginal discharge in postpartum
dairy cows and effect of intrauterine cephapirin on reproductive
performance. Aust Vet J 2009;87:36978.
[22] LeBlanc SJ, Duffield TF, Leslie KE, Bateman KG, Keefe GP,
Walton JS, Johnson WH. Defining and diagnosing postpartum
clinical endometritis and its impact on reproductive performance
in dairy cows. J Dairy Sci 2002;85:222336.
[23] Pleticha S, Drillich M, Heuwieser W. Evaluation of the Metri-
check device and the gloved hand for the diagnosis of clinical
endometritis in dairy cows. J Dairy Sci 2009;92:542935.
[24] Barlund CS, Carruthers TD, Waldner CL, Palmer CW. A com-
parison of diagnostic techniques for postpartum endometritis in
dairy cattle. Theriogenology 2008;69:71423.
[25] Sheldon IM, Noakes DE, Rycroft AN, Dobson H. Effect of
postpartum manual examination of the vagina on uterine bacte-
rial contamination in cows. Vet Rec 2002;151:5314.
[26] Williams EJ, Fischer DP, England GCW, Dobson H, Pfeiffer DU,
Sheldon IM. Clinical evaluation of postpartum vaginal mucus
reflects uterine bacterial infection and the inflammatory response
to endometritis in cattle. Theriogenology 2005;63:10217.
[27] Hosmer DW, Lemeshow S. Applied Logistic Regression. New
York: John Wiley & Sons, Inc; 1989.
[28] Zhang J, Yu KF. Whats the relative risk? A method of correcting
the odds ratio in cohort studies of common outcomes. JAMA
1998;280:16901.
[29] Bonnett BN, Martin SW, Gannon VP, Miller RB, Etherington
WG. Endometrial biopsy in Holstein-Friesian dairy cows. III.
Bacteriological analysis and correlations with histological find-
ings. Can J Vet Res 1991;55:16873.
[30] Herath S, Fischer DP, Werling D, Williams EJ, Lilly ST, Dobson
H, Bryant CE, Sheldon IM. Expression and function of Toll-like
receptor 4 in the endometrial cells of the uterus. Endocrinology
2006;147:56270.
[31] Sheldon IM, Williams EJ, Miller ANA, Nash DM, Herath S.
Uterine diseases in cattle after parturition. Vet J 2008;176:115
21.
[32] Laven RA, Peters AR. Bovine retained placenta: aetiology,
pathogenesis and economic loss. Vet Rec 1996;139:46571.
[33] Joosten I, Sanders MF, Hensen EJ. Involvement of major histo-
compatibility complex class I compatibility between dam and
calf in the aetiology of bovine retained placenta. Anim Genet
1991;22:45563.
[34] Kimura K, Goff JP, Kehril Jr ME, Reinhardt TA. Decreased
neutrophil function as a cause of retained placenta in dairy cattle.
J Dairy Sci 2002;83:53350.
[35] Correa MT, Erb H, Scarlett J. Path analysis for seven postpartum
disorders of Holstein cows. J Dairy Sci 1993;76:130512.
[36] Noakes DE, Wallace L, Smith GR. Bacterial flora of the uterus of
cows after calving on two hygienically contrasting farms. Vet
Rec 1991;128:4402.
[37] Markusfeld O. Factors responsible for post parturient metritis in
dairy cattle. Vet Rec 1984;114:53942.
[38] Gilbert RO, Shin ST, Guard CL, Erb HN, Frajblat M. Prevalence
of endometritis and its effects on reproductive performance of
dairy cows. Theriogenology 2005;64:187988.
[39] Williams EJ, Herath S, England GC, Dobson H, Bryant CE,
Sheldon IM. Effect of Escherichia coli infection of the bovine
uterus from the whole animal to the cell. Animal 2008;2:11537.
[40] Sheldon IM, Noakes DE. Comparison of three treatments for
bovine endometritis. Vet Rec 1998;142:5759.