Spontaneous recovery or persistence of postpartum endometritis and

risk factors for its persistence in Holstein cows

G. Gautam, T. Nakao *, K. Koike, S.T. Long, M. Yusuf,R.M.S.B.K. Ranasinghe, A. Hayashi

Laboratory of Theriogenology, Department of Veterinary Medicine, Faculty of Agriculture, Yamaguchi University, Yamaguchi, Japan

Received 3 July 2009; received in revised form 8 August 2009; accepted 24 August 2009

Abstract

It has been stated that postpartum endometritis in dairy cows has a tendency to cure without intervention. The objectives of this

field study, therefore, were to determine the proportions of cows with spontaneous clinical recovery or persistence of postpartum

endometritis and to determine some risk factors for its persistency in dairy cows (Bos taurus). Holstein-Friesian cows (n = 441

lactations) from seven dairy herds were examined monthly by vaginoscopy and transrectal palpation. A cow was considered to have

‘‘postpartum endometritis’’ if it had pus in the cervico-vaginal discharge at the first postpartum examination during Days 15 to 60

(Day 0 = day of calving); this was classified as mild, mucopurulent, or purulent endometritis, or endometritis with fluid in uterus.

Furthermore, a cow with evidence of endometritis at least once during Days 61 to 150 was considered to have ‘‘persistence (or

recurrence) of endometritis.’’ A total of 104 (23.6%) lactations had postpartum endometritis, of which 25.3% had persistence or

recurrence of clinical endometritis. Cows with persistence or recurrence of endometritis became pregnant at a slower rate (hazard

ratio [HR] = 0.28; P < 0.001) than those with no endometritis until Day 150. Calving in summer (odds ratio [OR] = 7.00; P = 0.04),

early postpartum complications (OR = 6.58; P = 0.05), moderate (OR = 4.03; P = 0.08) and severe (OR = 30.99; P = 004) degrees

of urovagina, and mucopurulent (OR = 9.54; P = 0.02) and purulent (OR = 5.70; P = 0.04) endometritis were risk factors for the

persistence or recurrence of endometritis. Furthermore, 10.6% of cows that had not shown signs of postpartum endometritis had a

new diagnosis of endometritis during Days 61 to 150. Some risk factors for the new diagnosis of endometritis beyond Day 60 were

early postpartum complications (OR = 2.82; P = 0.03) and moderate (OR = 5.00; P = 0.001) or severe (OR = 12.63; P < 0.001)

degrees of urovagina. In conclusion, approximately one quarter of cows with postpartum endometritis had persistence of

endometritis until or beyond the breeding period. Risk factors for the persistence of clinical endometritis were summer calving,

early postpartum complications, clinically relevant urovagina, and clinically relevant endometritis within 2 mo postpartum.

# 2010 Elsevier Inc. All rights reserved.

Keywords: Endometritis; Holstein cows; Spontaneous recovery; Urovagina; Vaginoscopy

www.theriojournal.com

Available online at www.sciencedirect.com

Theriogenology 73 (2010) 168–179

1. Introduction

Postpartum uterine diseases have been described as a

leading cause of reproductive inefficiency in dairy cattle

* Corresponding author. Tel.: +81 83 933 5935;

fax: +81 83 933 5935.

E-mail address: tnakao@yamaguchi-u.ac.jp (T. Nakao).

0093-691X/$ – see front matter # 2010 Elsevier Inc. All rights reserved.

doi:10.1016/j.theriogenology.2009.08.010

[1–17]. Uterine diseases, postpartum endometritis in

particular, reduce the reproductive efficiency of cows by

extending the calving to conception interval, increasing

the number of inseminations per conception [5,7–

9,15,16], and increasing culling rates [5,7,8]. The

prevalence of endometritis in 43 studies ranged from

2.2% to 37.3%, with a median of 10.1% [18]. In some

herds, 40% of postpartum cows may be diagnosed and

treated for uterine infections [17].

G. Gautam et al. / Theriogenology 73 (2010) 168–179 169

Although the prevalence of postpartum endometritis

and its impact on subsequent fertility has been widely

reported, the follow-up clinical examination regarding

its fate has not been well documented. Many published

reports failed to document an overall significant

improvement in reproductive performance after treat-

ment of endometritis [19–25], suggesting that a high

proportion of endometritis cases have a tendency to

recover spontaneously. Some previous studies sup-

ported the likelihood that a high proportion of cows

have spontaneous resolution of endometritis until at

least 4 wk postpartum [5,19]. However, others demon-

strated that the clinical examination of vaginal contents

even before 3 wk postpartum predicted reduced

reproductive performance in dairy cows [7,13,14,20].

Therefore, it is necessary to determine whether the

postpartum period at a diagnosis of endometritis

influences its recovery or persistence. The central

question is how some postpartum cows can resolve

uterine infections spontaneously without intervention,

whereas others in the same herd, with presumably the

same general management, have persistent infection. It

is also not known what proportion of cows resolve

uterine infection spontaneously and what proportions

have persistent infection. Therefore, the objectives of

this field study were to determine the proportions of

cows with spontaneous clinical recovery or persistence

of postpartum endometritis and to determine some risk

factors for its persistency in Holstein cows in a

subtropical area of Japan.

2. Materials and methods

2.1. Animals and management

This study was conducted in 291 Holstein-Friesian

cows (Bos taurus) and used 441 lactations for cows

calving between October 2005 and December 2008.

These cows were in seven commercial dairy herds in

Yamaguchi Prefecture, southwestern region of Japan,

under subtropical conditions. These herds were visited

monthly for reproductive examination by a team of

three or four persons including faculty members, as well

as graduate and undergraduate students from the

Laboratory of Theriogenology, Yamaguchi University

(headed by the senior author). The study population

consisted of cows used for our previous study [26], as

well as additional cows from the same herds. Cows were

kept in a loose housing system in a small paddock (zero-

grazing system) with wood shavings on the floor

throughout the year (Herds A and B), tie-stall barns with

rubber mattresses on the floor (Herds C, D, and E), or in

tie-stall barns with an open paddock (Herds F and G).

The herd size ranged from 20 to 60 lactating cows. The

cows were nonseasonal, year-round calvers, milked

twice daily with herd average 305-d milk yield between

8700 and 10,200 kg per cow. Parity ranged from one to

seven. Cows were artificially inseminated after detec-

tion of estrus by the herd owners, usually after 35 d

postpartum. In all herds, no estrus/ovulation synchro-

nization programs were used. Cows were observed for

signs of estrus in the morning, at noon, and in the late

afternoon. In Herds A, B, F, and G, standing to be

mounted was interpreted as a sign of estrus, and those

cows were inseminated 8 to14 h later. In Herds C, D,

and E, estrus was detected based on some secondary

estrous signs (swelling, relaxation, and congestion of

vulva, mucus discharge, bellowing, restlessness, or

decreased milk yield), and cows were inseminated 8 to

14 h later. In all herds, cows were inseminated with

frozen-thawed semen of proven sires (three or four sires

in total) from the Livestock Improvement Association

of Japan. In Herds F and G, embryo transfer was also

done in some cows (decided by the herd owners).

Pregnancy was diagnosed by transrectal palpation or

with ultrasonography (Tringa Linear Vet, equipped with

a dual frequency 5/7.5 MHz intrarectal transducer; Pie

Medical Equipment BV, Maastricht, The Netherlands),

usually �30 d after insemination (or embryo transfer).

2.2. Clinical examination

Each herd was visited once a month for regular

reproductive examinations, which included the detec-

tion and treatment of reproductive disorders, pregnancy

diagnosis, and recording of body condition score (BCS;

scale 1 to 5, with 0.25-point increments) [27]. Cows that

were more than 2 wk postpartum were examined by

vaginoscopy and transrectal palpation every month until

confirmation of pregnancy, a decision to cull, or at least

until 5 mo postpartum. Each time, all subjective

assessments were made by the same veterinarian (the

senior author). At each visit, if a cow was within 1 mo

after artificial insemination (AI) and did not return to

estrus, only vaginoscopy was performed. If the cow was

more than 1 mo after AI, pregnancy was diagnosed by

transrectal palpation or ultrasonography. All other cows

were examined vaginoscopically followed by transrec-

tal palpation of the genital tract.

For vaginoscopy, first the cow was restrained in a

feeding stanchion or a halter, the tail was held to one

side, the vulva washed with lukewarm water mixed with

benzylkonium chloride 10% (Osvan; Nihon Seiyaku

Co., Tokyo, Japan) at 1:1000 dilution, wiped with a

G. Gautam et al. / Theriogenology 73 (2010) 168–179170

clean paper towel, sprayed with 2% polyvinyl pyrrol-

iodine, and wiped with cotton soaked in 70% alcohol.

An autoclave-sterilized glass vaginoscope (35 cm long,

2 mm thick, and 4 cm external diameter; L.C.C. Co.

Ltd., Shunan Shi, Yamaguchi, Japan) was inserted into

the vagina up to the level of the external cervical os. A

separate vaginoscope was used for each cow. The

contents of the anterior vagina, if any, were viewed

using a torch light, and a portion was aspirated with a

25-mL or 50-mL syringe fitted to an AI gun sheath and

transferred into a Petri dish. Each time, the nature of the

recovered fluid was scored qualitatively as clear mucus,

predominately clear mucus with some flecks of pus,

mucopurulent (approximately 50% pus and 50%

mucus), or purulent (>50% pus). The recovered fluid

was also checked for its color (yellowish or no color),

consistency (watery or mucoid), smell (uremic or no

smell), and pH using a paper strip (Spezial-Indikator-

papier; Macherey-Nagel GmbH & Co. KG, Duren,

Germany).

After vaginoscopy, transrectal palpation of the

reproductive tract was performed to determine the

diameter of cervix, diameter, location (abdominal or

pelvic), consistency (contraction, elasticity and toni-

city), and symmetry of uterine horns (i.e., differences in

diameter), the presence of any fluid in the uterus (yes or

no), and the presence of any palpable ovarian structures

(follicle, corpus luteum, cystic ovarian follicle, etc.).

2.3. Case definition

Cases having any amount of purulent material in the

cervico-vaginal discharge without any signs of systemic

illness were defined as endometritis and classified into

the following categories, based on vaginoscopic and

transrectal palpation findings:

� Mild endometritis (E1): cows with predominately

clear mucus with some flecks of pus on vaginoscopy.

� Mucopurulent endometritis (E2): cows with muco-

purulent discharge on vaginoscopy.

� Purulent endometritis (E3): cows with purulent

discharge on vaginoscopy.

� Endometritis with fluid in uterus (E4): cows with

purulent or mucopurulent discharge on vaginoscopy

and appreciable amount of fluid in uterus detected by

transrectal palpation.

No treatment was indicated for endometritis within 2

mo postpartum, except for 10 cows, which were

excluded from the analysis. Similarly, after 2 mo

postpartum, no treatment was indicated for endometritis

(except six cows treated for persistent endometritis,

which were excluded from the analyses of reproductive

performance).

Cases having vaginal contents with the presence of

urine or urine mixed mucus (characterized by at least

any two of yellowish in color, watery in consistency,

uremic in smell, and/or pH >7.4) were defined as

urovagina. Cases of urovagina were further classified

into the following different degrees based on the extent

of covering of external cervical os by the urine or urine-

mixed mucus [26]:

� Mild degree of urovagina: small amount of urine or

urine mixed mucus (approximately 10 to 100 mL)

only on the floor of vagina, but not covering the

external cervical os.

� Moderate degree of urovagina: appreciable amount of

urine or urine mixed mucus (approximately 100 to

500 mL) covering almost half or less than half of the

external cervical os (with some risk of urine entering

the uterus).

� Severe degree of urovagina: large amount of urine or

urine mixed mucus (approximately >500 mL) cover-

ing more than half or the whole external cervical os

(high risk of urine entering the uterus).

The vagina of cows with urovagina (except that of

those not returning to estrus after AI) was flushed with 2

L 3% aluminum potassium sulfate (Myouban; Nipro

Farm Co., Osaka, Japan) using an autoclaved glass

vaginoscope and a hose (2 cm in diameter) fitted to a

funnel, and the fluid was removed from the vagina.

2.4. Data collection and management

Data were collected from 441 lactations, which

included herd, parity (1, 2, 3, or �4), season of calving

(autumn: September through November; winter:

December through February; spring: March through

May; summer: June through August), type of calving

(normal, or abnormal [dystocia, abortion or fetal

death]), retention of fetal membranes >24 h (RFM),

early postpartum complications (any one or more of

puerperal metritis, hypocalcemia, abomasal displace-

ment, ketosis, lameness, or clinical mastitis before or at

first postpartum examination), BCS at first postpartum

examination, BCS at first examination beyond Day 60

(Day 0 = day of calving), degree of endometritis, and

degree of urovagina. Cows treated for endometritis

within Day 60 (n = 10) and those culled before having

two examinations beyond Day 60 (culled too early,

n = 9) were excluded from the analysis.

G. Gautam et al. / Theriogenology 73 (2010) 168–179 171

During scheduled monthly herd visits, all postpartum

cows were examined for the first time during Days 15 to

60. Therefore, a cow was considered to have ‘‘post-

partum endometritis’’ if it showed any degree of

endometritis (defined as above) at first postpartum

examination. During Days 61 to 150, a cow with any

amount of pus in the cervico-vaginal discharge at least

once (out of two or more examinations) was considered

to have endometritis (considering that any pus discharge

during the breeding period is abnormal, all degrees of

endometritis beyond Day 60 were simply considered as

‘‘endometritis’’ without any classification).

Based on these criteria, a cow that had postpartum

endometritis but showed no more clinical endometritis

beyond Day 60 (out of two or more examinations) or

conceived within Day 60 was considered to have

‘‘spontaneous clinical recovery of endometritis,’’

whereas a cow that had postpartum endometritis and

was still showing clinical endometritis during Days 61

to 150 was considered to have ‘‘persistence or

recurrence of clinical endometritis.’’ To avoid confu-

sion, hereafter the terms ‘‘spontaneous recovery’’ and

‘‘persistence’’ are used. Likewise, a cow that had no

signs of postpartum endometritis but showed clinical

endometritis at least once during Days 61 to 150 was

considered to have ‘‘new diagnosis of endometritis.’’

2.5. Statistical analysis

The prevalence of different degree of ‘‘postpartum

endometritis’’ was expressed as a percentage of total

examined lactations within Day 60. During Days 15 to

60, the uterine conditions were expected to be dynamic

over the enrollment period. Therefore, for the analysis,

the findings were stratified into three stages, 15 to 20 d

(early), 21 to 28 d (intermediate), and 29 to 60 d (late),

as well as overall. Spontaneous recovery (%) or

persistence (%) of endometritis was expressed as a

proportion of cows with ‘‘postpartum endometritis’’

that had spontaneous clinical recovery or that had

persistence of clinical endometritis during Days 61 to

150, respectively. All analyses were performed with

SPSS software [28], with lactation as the unit of

concern. The reproductive performance of cows

diagnosed with endometritis before 3 wk postpartum

was compared with that of those with no endometritis

during this period by Cox’s proportional hazards

regression models of time to conception and by logistic

regression models of pregnancy proportions by 210 d

postpartum. The variables describing endometritis

during Days 15 to 20 (yes or no), herd, season of

calving, parity group, and BCS (continuous) were

included in a proportional hazards regression model of

time to conception. Models were built by manual

reverse stepwise elimination, using likelihood-ratio test

statistics. At each step, the variable with highest P value

was removed, and the model was run until all the

remaining independent variables and covariates were

significant (P < 0.05). Multiple logistic regression

analysis was used to analyze variables with dichot-

omous outcomes. Risk factors were tested against each

outcome variable in a univariate analysis. All variables

that were associated with an outcome of interest

(P < 0.25) were included in a full model, and a reverse

stepwise procedure using a likelihood-ratio test

statistics was used to arrive at a final model, with a

threshold P value for inclusion being < 0.05 and >0.1

for exclusion. The final models were assessed using

Hosmer-Lemeshow goodness of fit test. For analysis of

pregnancy proportion by 210 d postpartum in cows

examined within 3 wk postpartum, the model included

endometritis status during Days 15 to 20 (yes or no),

herd, season of calving, parity group, and BCS at first

postpartum examination (continuous).

For analysis of risk factors for ‘‘postpartum

endometritis,’’ the model included herd (1 to 7), parity

(1, 2, 3, or �4), season of calving (autumn, winter,

spring, or summer), type of calving (normal or

abnormal), RFM (no or yes), early postpartum

complications (no or yes), BCS at first postpartum

examination (continuous), urovagina within Day 60 (no,

mild, moderate, or severe), and postpartum interval at

examination (15 to 20, 21 to 28, or 29 to 60 d).

For analysis of persistence of endometritis, a model

was constructed for cattle in which the ‘‘postpartum

endometritis’’ (i.e., endometritis within Day 60) was

diagnosed; the model included herd, parity, season of

calving, RFM, early postpartum complications, BCS at

first postpartum examination, BCS at first examination

beyond Day 60, BCS change between first postpartum

examination and first examination beyond Day 60

(increased, unchanged, or decreased), degree of

urovagina until Day 150, degree of postpartum

endometritis, postpartum interval at examination, and

terms of interactions between degree of endometritis

and postpartum interval at examination.

Likewise, a model was constructed for cattle that did

not show the postpartum endometritis to analyze the

risk factors for ‘‘new diagnosis of endometritis’’; the

model included the same explanatory variables

described for the persistence of endometritis except

the degree of postpartum endometritis.

The pregnancy rates among cows with no endome-

tritis until Day 150, those with spontaneous recovery of

G. Gautam et al. / Theriogenology 73 (2010) 168–179172

Table 1

Prevalence of degrees of postpartum endometritis in Holstein cows examined for the first time between Days 15 and 60.

Finding Interval postpartum at examination

15–20 d (Early) 21–28 d (Intermediate) 29–60 d (Late) Overall

Number of cows examined 89 114 238 441

Postpartum endometritis, n (%)* 40 (44.9) 29 (25.4) 35 (14.7) 104 (23.6)

Mild (E1), n (%) 11 (12.4) 10 (8.8) 19 (8.0) 40 (9.1)

Mucopurulent (E2), n (%) 7 (7.7) 7 (6.1) 6 (2.5) 20 (4.5)

Purulent (E3), n (%) 16 (18.0) 8 (7.0) 7 (2.9) 31 (7.0)

Endometritis with fluid in uterus (E4), n (%) 6 (6.7) 4 (3.5) 3 (1.3) 13 (3.0)

E1, almost clear mucus with some flecks of pus; E2, mucopurulent discharge; E3, purulent discharge; and E4, purulent or mucopurulent discharge on

vaginoscopy and fluid in uterus detected by transrectal palpation.* Postpartum endometritis was diagnosed based on vaginoscopic and transrectal palpation findings at first postpartum examination.

endometritis, those with persistence of endometritis,

and those with new diagnosis of endometritis were

compared by Cox’s proportional hazards regression

analysis as above, accounting (if significant) for the

effects of herd, season of calving, parity, and BCS at

first postpartum examination. Likewise, the odds ratios

of pregnancy proportions by 210 d postpartum among

these four categories were compared by multiple

logistic regression model as above, accounting (if

significant) for the effects of herd, season of calving,

parity, and BCS at first postpartum examination.

3. Results

In total, 441 lactations were examined for endome-

tritis, of which 104 (23.6%) had postpartum endome-

tritis (i.e., endometritis within Day 60), with 9.1% mild

(E1), 4.5% mucopurulent (E2), 7.0% purulent (E3)

endometritis, and 3.0% endometritis with fluid in the

uterus (E4). The prevalence of endometritis during the

early (15 to 20 d), intermediate (21 to 28 d), and late (29

Table 2

Odds ratios of the variables included in the final logistic regression model for

cows.

Factor Class Endometriti

n

Herd

Type of calving Normal 82/381

Abnormal 22/60

Retention of fetal membranes No 83/398

Yes 21/43

Early postpartum complications* No 84/387

Yes 20/54

Postpartum interval at examination, d 15 to 20 40/89

21 to 28 29/114

29 to 60 35/238

*Early postpartum complications included any one or more of the followin

lameness, or clinical mastitis before or at the first postpartum examination

to 60 d) postpartum periods was 44.9%, 25.4%, and

14.7%, respectively (Table 1). Cows diagnosed with

clinical endometritis during Days 15 to 20 became

pregnant at a slower rate (HR = 0.60; 95% CI 0.36 to

0.98; P = 0.04) compared with that of those with no

clinical endometritis during this period. When the

endometritis status during Days 15 to 20 was substituted

by the degrees of endometritis in the model, the

respective HRs for pregnancy were 1.33 (95% CI, 0.64

to 2.77; P = 0.44), 0.29 (95% CI, 0.10 to 0.98; P = 0.05),

0.59 (95% CI, 0.31 to 1.12; P = 0.10), and 0.27 (95% CI,

0.10 to 0.91; P = 0.03) for E1, E2, E3, and E4,

respectively. No other factors remained in the final

model. Likewise, cows diagnosed with clinical endo-

metritis during Days 15 to 20 were 0.32 times less likely

to become pregnant by 210 d postpartum (66.7% vs.

86.1%; OR, 0.32; 95% CI, 0.11 to 1.00; P = 0.05)

compared with those with no clinical endometritis

during this period.

Logistic regression analysis for the risk of post-

partum endometritis indicated no significant effects of

the risk of postpartum endometritis (during Days 15 to 60) in Holstein

s (15–60 d) OR 95% CI P value

Percentage, %

0.01

21.5 Reference

36.4 2.20 1.08–4.46 0.03

20.8 Reference

48.8 3.48 1.65–7.37 0.001

21.7 Reference

37.0 2.29 1.08–4.46 0.02

44.9 6.33 3.47–11.57 <0.001

25.4 2.21 1.23–3.98 0.008

14.7 Reference

g: puerperal metritis, hypocalcemia, abomasal displacement, ketosis,

.

G. Gautam et al. / Theriogenology 73 (2010) 168–179 173

Table 3

Frequency distribution for varying degrees of postpartum endometritis diagnosed during various periods in dairy cows and their spontaneous

recovery or persistency during the latter (Days 61 to 150) postpartum period.

Diagnosis at first postpartum examination Endometritis status during

Days 61 to 150, n (%)

Postpartum interval at

first examination (d)

Degree of postpartum

endometritis*

Number of

cowsyNo Yes

15–20 (Early) E1 10 9 1

E2 6 4 2

E3 16 12 4

E4 5 5 0

Total 37 30 (81.0) 7 (19.0)

21–28 (Intermediate) E1 10 8 2

E2 5 3 2

E3 5 4 1

E4 2 1 1

Total 22 16 (72.7) 6 (27.3)

29–60 (Late) E1 19 16 3

E2 5 3 2

E3 6 2 4

E4 2 1 1

Total 32 22 (68.8) 10 (31.2)

Overall 91 68 (74.7) 23 (25.3)

*For definition, see Table 1.yCows treated for endometritis (n = 10) and those culled before having two examinations beyond Day 60 (n = 3) were not included.

parity, season of calving, BCS at first postpartum

examination, and urovagina within Day 60. Variables

that significantly increased the risk of diagnosing

postpartum endometritis were herd, type of calving,

RFM, early postpartum complications, and postpartum

interval at examination (Table 2). There were no

significant interactions (P > 0.1).

Table 4

Odds ratios of the variables included in the final logistic regression model fo

Holstein cows.

Factor Class Persistence o

n*

Season of calving Autumn 3/21

Winter 7/23

Spring 1/18

Summer 12/29

Early postpartum complications No 15/72

Yes 8/19

Urovagina No 8/54

Mild 3/15

Moderate 7/14

Severe 5/8

Endometritis during Days 15 to 60y E1 6/39

E2 6/16

E3 9/27

E4 2/9

*Cows treated for endometritis (n = 10) and those culled before having twoyFor definition, see Table 1.

Of 104 cattle with postpartum endometritis, 10 were

treated for endometritis within Day 60, and three were

culled before having two examinations beyond Day 60.

Thus, 91 cattle remained for the analysis of persistence

or spontaneous recovery of endometritis. Of 91 cows

with postpartum endometritis, 68 (74.7%) recovered

spontaneously, and 23 (25.3%) had persistence of

r the risk of persistence or recurrence of postpartum endometritis in 91

f endometritis OR 95% CI P value

Percentage, %

14.3 Reference

30.4 1.69 0.22–13.03 0.61

5.6 0.08 0.03–1.83 0.11

41.4 7.00 1.05–46.93 0.04

20.8 Reference

42.1 6.58 0.90–48.46 0.05

14.8 Reference

20.0 1.65 0.26–10.30 0.59

50.0 4.03 0.83–19.69 0.08

62.5 30.99 2.91–330.09 0.004

15.4 Reference

37.5 9.54 1.33–67.02 0.02

33.3 5.70 1.11–30.78 0.04

22.2 0.70 0.05–9.66 0.78

examinations beyond Day 60 (n = 3) were not included in the model.

G. Gautam et al. / Theriogenology 73 (2010) 168–179174

Table 5

Odds ratios of the variables included in the final logistic regression model for the risk of new diagnosis of endometritis (during Days 61 to 150)

among cows that had no endometritis within 60 d postpartum.

Factor Class New diagnosis of endometritis

(61–150 d)

OR 95% CI P value

n* Percentage, %

Early postpartum complications No 26/297 8.8 Reference

Yes 9/34 26.5 2.82 1.11–7.18 0.03

Urovagina No 18/263 6.8 Reference

Mild 4/32 12.5 1.84 0.57–5.90 0.30

Moderate 8/26 30.8 5.00 1.85–13.49 0.001

Severe 5/10 50.0 12.63 3.26–48.86 <0.001

*Cows having postpartum endometritis during Days 15 to 60 (n = 104), and those culled before having two examinations beyond Day 60 (n = 6)

were not included in the model.

Fig. 1. Survival curves derived from final Cox’s proportional hazards

regression analysis (adjusted for the effects of herd, season of calving,

and BCS at first postpartum examination) for proportion of nonpreg-

nant dairy cows against days postpartum in cows with no endometritis

until Day 150, cows with spontaneous recovery of endometritis, cows

with persistence or recurrence of endometritis, and cows with a new

diagnosis of endometritis during Days 61 to 150.

endometritis during Days 61 to 150 (Table 3). Among

91 cows with postpartum endometritis, 76 (83.5%) cows

had no abnormal vaginal discharge on the second

postpartum examination. However, 8 of these 76 cows

had an abnormal discharge again on subsequent

examinations.

Logistic regression analysis for the risk of persis-

tence of endometritis during Days 61 to 150 indicated

no significant effects of herd, parity, type of calving,

RFM, BCS at first postpartum examination, BCS at first

examination beyond Day 60, BCS change, postpartum

interval at examination, and the terms of interactions

between degree of endometritis and postpartum interval

at examination. Calving in summer season, early

postpartum complications, moderate and severe degrees

of urovagina, and mucopurulent and purulent endome-

tritis within Day 60 were significant risk factors for the

persistence of endometritis (Table 4). No significant

interactions were found between significant covariates

and degree of endometritis within Day 60 (P> 0.1). The

proportion of persistency was not influenced by the

periods postpartum at diagnosis of endometritis; that is,

whether postpartum endometritis was diagnosed during

Days 15 to 20, during Days 21 to 28, or during Days 29

to 60 (19.0%, 27.3%, and 31.2%, respectively; P = 0.49;

Table 3).

Likewise, among 337 lactations with no signs of

postpartum endometritis, six were culled before having

two examinations beyond Day 60. Of the remaining 331

lactations with no postpartum endometritis, 35 (10.6%)

had a new diagnosis of endometritis during Days 61 to

150. Logistic regression analysis for the risk of new

diagnosis of endometritis beyond Day 60 indicated no

significant effects of herd, season of calving, parity, type

of calving, RFM, BCS at first postpartum examination,

BCS at first examination beyond Day 60, and BCS

change. Early postpartum complications and the

moderate and the severe degrees of urovagina were

the significant risk factors for the new diagnosis of

endometritis during Days 61 to 150 (Table 5). There was

no significant interaction between postpartum compli-

cations and urovagina (P > 0.1).

Accounting for the effects of herd, season of calving,

and BCS at first postpartum examination (parity did not

remain in the final model), cows with the persistence of

endometritis and those with new diagnosis of endome-

tritis but not those recovered spontaneously became

pregnant at a slower rate compared with cows with no

clinical endometritis until Day 150 (Fig. 1 and Table 6).

Likewise, the pregnancy proportions by 210 d post-

partum in cows with persistence of endometritis

(40.0%; OR, 0.10; 95% CI, 0.05 to 0.22; P < 0.001)

G. Gautam et al. / Theriogenology 73 (2010) 168–179 175

Table 6

Final Cox’s proportional hazards regression models of clinical findings that were associated with time to conception (relative pregnancy rate) in

Holstein cows.

Categories n* HR pregnancyy 95% CI P value

No endometritis until Day 150 296 Reference

Spontaneous recovery of endometritis 68 0.89 0.66–1.20 0.43

Persistence or recurrence of endometritis 17 0.28 0.14– 0.53 <0.001

New diagnosis of endometritis (61–150 d) 35 0.44 0.28–0.70 <0.001

*Cows treated for endometritis (n = 16), and those culled before having two examinations beyond Day 60 (n = 9) were excluded from the analysis.yHazard ratio of pregnancy adjusted for the effects of herd, season of calving, and BCS at first postpartum examination.

and in those with a new diagnosis of endometritis during

Days 61 to 150 (53.1%; OR, 0.31; 95% CI, 0.14 to 0.68;

P = 0.004) but not in those that recovered spontaneously

(70.3%; OR, 0.65; 95% CI, 0.34 to 1.24; P = 0.19) were

lower compared with that in cows with no clinical

endometritis until Day 150 (78.3%).

4. Discussion

This study is one of the few reported trials that

conducted follow-up examinations of cows with

postpartum endometritis to determine the subsequent

fate of endometritis until or beyond the breeding period.

Although there are several reports on the prevalence of

postpartum endometritis, its impact on reproductive

performance, and risk factors for postpartum endome-

tritis [5,7–16], to our knowledge, the current study was

the first to investigate risk factors for persistence of

endometritis in Holstein cows. It is difficult to compare

the prevalence of postpartum endometritis with other

reports, due to differences in the methods of diagnosis,

postpartum periods at examination, and lack of

description of diagnostic methods in some studies

[29–31]. When comparing the prevalence rate with

other studies that examined the contents of vagina for

the presence of pus, the prevalence of postpartum

endometritis in this study seemed similar to those

reported by McDougall et al. [13] and Runciman et al.

[14] but slightly lower than some previously reported

rates [5,7,15,32] and higher than that reported by others

[24,33]. There are very few reported trials that

examined the cows prior to 3 wk postpartum using

vaginoscopy, due to concerns regarding diagnosing self-

resolving uterine infection. However, in the current

study, cows diagnosed with endometritis (but not mild

cases) even before 3 wk postpartum had detrimental

effects on future reproductive performance, which was

in agreement with our previous findings [7] and also

with some other reports [13,14,20]. This demonstrated

the usefulness of performing vaginoscopy even prior to

3 wk postpartum. Furthermore, it was also of interest to

determine whether the endometritis diagnosed before 3

wk postpartum had a greater tendency to recover

spontaneously than that diagnosed during later post-

partum periods; however, it was not verified from this

study. It was the degree of endometritis rather than the

postpartum interval at diagnosis of endometritis that

influenced the recovery or persistency of endometritis.

The prevalence of postpartum endometritis in this

study was significantly associated with herd, abnormal

calving, RFM, early postpartum complications, and the

postpartum interval at examination. The increased

incidence of postpartum endometritis with abnormal

calving was consistent with some previous studies

[14,29,30,34,35] but in contrast with others [5,7,16].

Likewise, the increased risk of postpartum endometritis

in cows with RFM was consistent with most previous

studies [5,7,14,16,34]. Abnormal calving and RFM may

serve as a perfect medium for bacterial growth and

thereby increase the susceptibility for uterine diseases.

In the current study, early postpartum complications

increased the risk of postpartum endometritis, which

was consistent with the findings of previous studies that

demonstrated the increased risk of endometritis in cows

with postpartum metabolic disorders [16,29,30,36].

Because of the small number of cows with each

complication, and assuming that each might interfere

with immunity, and therefore ability of the animal to

clear uterine contamination, we combined all these

complications to form ‘‘early postpartum complica-

tions.’’ Similarly, the risk of diagnosing postpartum

endometritis decreased with the time postpartum, which

was in agreement with McDougall et al. [13] and also in

agreement with previous bacteriologic studies demon-

strating that the prevalence of uterine infection declines

with the time postpartum [12,37–40].

Approximately three quarters of the cows with

postpartum endometritis in this study had spontaneous

clinical recovery, whereas the remainder had persis-

tence or recurrence of infection until or beyond the

breeding period. It was not known whether these cases

were the persistence or recurrence of endometritis.

G. Gautam et al. / Theriogenology 73 (2010) 168–179176

Therefore, we simply used the term ‘‘persistence’’ to

include both cases. If clinical resolution had been

considered only on the basis of second examination,

83.5% of cows with postpartum endometritis would

have been classified as spontaneously recovered.

However, 8 (9%) of 91 cows with postpartum

endometritis that had no abnormal discharge on second

examination showed signs of clinical endometritis again

on subsequent examinations. Therefore, we decided to

define spontaneous recovery based on at least two

examinations during the breeding period. The diagnosis

of endometritis in this study was based only on clinical

findings. In the absence of clinical signs of endometritis,

a substantial proportion (up to 50%) of cows had

subclinical endometritis during the postpartum period

that impaired the subsequent reproductive performance

[8–10]. The cytologic criteria for diagnosis of

subclinical endometritis continue to be refined

[10,11,41]. Although the endometrial histology or

cytology for the diagnosis of endometritis was beyond

the scope of this study, we inferred that several

vaginoscopic examinations increased the accuracy of

our criteria of classifying cows into spontaneous

recovery or no endometritis until Day 150. Moreover,

the spontaneous recovery defined here was later

validated by the lack of difference in pregnancy rates

between cows with spontaneous recovery and those

with no endometritis until Day 150. Conversely, cows

with persistence of endometritis had poorer reproduc-

tive performance compared with those with no

endometritis until Day 150. Therefore, we inferred

that cows failing to resolve uterine infection suffer more

detrimental reproductive consequences than those that

resolve uterine infection. It can be assumed that the

persistence of endometritis left alternations in the

uterine lining and decreased the ability of cows to

conceive, and thus reduced reproductive performance.

Cows calving in summer were seven times more likely

to persist with endometritis beyond Day 60 than those

calving in autumn. Roman-Ponce et al. [42] reported that

uterine blood flow was reduced in cows exposed to heat

stress. Because of less blood flow to the uterus, fewer

leukocytes are moved to the endometrium; this may

compromise the clearance of uterine infection, leading to

the persistence of endometritis. Moreover, overall

immune status of cows decreased by the stress of calving

might be further exacerbated during heat stress and could

contribute to the persistence of endometritis. Postpartum

complications continued to be the risk factor for the

persistence of endometritis. Early postpartum complica-

tions might have interfered with immunity and therefore

the ability to clear a uterine infection and thus might have

contributed to its persistence for a prolonged period.

Likewise, cows with moderate and severe degrees of

urovagina were 4 and 31 times more likely, respectively,

to have persistence of endometritis compared with cows

with no urovagina. Cows with signs of postpartum

endometritis (a purulent cervical discharge) usually have

a relaxed cervix, which may allow the entry of

contaminated urine into the uterus in cows with

‘‘moderate’’ and ‘‘severe’’ degrees of urovagina, and

this may lead to the persistence of endometritis.

Similarly, persistence of postpartum endometritis

depended on its severity. Mucopurulent and purulent

endometritis were nine and six times, respectively, more

likely to persist compared with mild endometritis.

Perhaps uterine inflammation in case of mucopurulent

and purulent endometritis may reach a level of severity

where spontaneous recovery of uterine tissue is less

likely. Conversely, most cases of mild endometritis

recovered spontaneously. We concluded that the self-

healing capacity of the uterus in case of mild

endometritis was very high. This might explain previous

reports in which severe degrees of endometritis, but not

mild endometritis, had a clear negative effect on fertility

[5,7,24,36]. However, due to the small number of cows

with endometritis with fluid in the uterus (E4), this

category might not have shown the significant effect on

persistence or recovery of the infection; and this needs

to be confirmed by further study using a large number of

cows. Thus, cows with purulent or mucopurulent

discharge on vaginoscopy during Days 15 to 60 may

be defined as ‘‘clinically relevant endometritis.’’ Based

on the outcome of the current and previous studies, we

concluded that cows with severe endometritis before the

breeding period need to be treated, provided the therapy

is known to improve the fertility of affected cows,

whereas in mild cases, the self-healing potential is very

high, and the affected cows show good reproductive

efficiency without treatment, as long as they do not have

postpartum complications or a moderate or severe

degree of urovagina.

There are very few ‘‘reported’’ trials that examined

cows vaginoscopically during the breeding period

(beyond Day 60). In this study, 10.6% of cows that

did not show the signs of postpartum endometritis had a

new diagnosis of endometritis during the breeding

period. Drillich et al. [43] also found that among the

cows that had no endometritis during 21 to 27 d

postpartum, approximately 5% of cows showed signs of

endometritis during 68 to 74 d postpartum. Cows with a

new diagnosis of endometritis clearly had poorer

reproductive performance compared with cows with

no endometritis until Day 150.

G. Gautam et al. / Theriogenology 73 (2010) 168–179 177

Among the cows without signs of clinical endome-

tritis within Day 60, early postpartum complications

and moderate and severe degrees of urovagina were risk

factors for the new diagnosis of endometritis beyond

Day 60. Perhaps cows with a new diagnosis of

endometritis had subclinical endometritis within Day

60, and because of poor immunity (due to early

postpartum complications and/or other factors) the

infection persisted and was clinically apparent during

the later postpartum period. However, the diagnosis of

subclinical endometritis was beyond the scope of this

study, and thus, these results should not be over-

estimated. Likewise, it was not fully known whether the

new diagnoses of endometritis were really new

infections or were the persistent cases that might not

have shown abnormal discharge within Day 60 because

of luteal phases of estrus cycle in some of the cows

during examination. Although assessment of vaginal

contents for the presence of pus is the most useful and

practical method for the diagnosis of endometritis [3], a

single vaginoscopic examination may miss up to 9% of

cows with uterine discharge [10]. Increased production

of cervical mucus and myometrial contractions during

estrus may increase the probability of observing

discharge externally or in the anterior vagina. Alter-

natively, uterine discharge may not be detected

externally or by vaginoscopy in cows in which the

cervix is closed. Therefore, it is recommended to

perform at least two vaginoscopic examinations at a 2-

to 4-wk interval before the breeding period to detect

cases of clinical endometritis.

Among cows without signs of endometritis within

Day 60, the prevalence of new endometritis (during Days

61 to 150) was significantly higher in cows with moderate

and severe degrees of urovagina but not in cows with mild

urovagina compared with that in cows with no urovagina,

which was similar to the findings of our previous study

[26]. Moderate and severe degrees of urovagina, defined

here, would have a higher risk of urine entering into the

uterus and might have caused chronic endometritis and

infertility [26,44,45]. Thus, similar to our previous study

[26], moderate and severe degrees of urovagina can be

considered as ‘‘clinically relevant urovagina.’’ The

prevalence of urovagina in this study (24%, 11%, 9%,

and 4% for overall, mild, moderate, and severe,

respectively) was consistent with our previous study

[26] that also used cows from the same herds. The reason

for the high prevalence of urovagina in these herds was

not known, although no other published reports were

available regarding the prevalence of urovagina in dairy

cows, and it was difficult to conclude whether this

prevalence rate was high.

The clinically relevant urovagina in this study

(prevalence, 13%) appeared to be one of the main

predisposing factors for the prevalence of endometritis

during the breeding period. Likewise, the competence

of the immune system may be one of the major

differences among cows that have spontaneous recov-

ery, persistence, or a new diagnosis of endometritis, and

postpartum complications might be one of the factors to

compromise the immune function of postpartum cows.

Neutrophil function was impaired in cows that develop

uterine infection [46–49]. However, it is still not known

whether there are any differences in neutrophil

functions between cows that recover uterine infection

spontaneously and those that have persistence of

infection, and therefore further study is recommended.

In conclusion, the prevalence of postpartum endo-

metritis in this study was 23.6%. Approximately three

quarters of cows with postpartum endometritis had

spontaneous clinical resolution, whereas the remainder

had persistence of clinical endometritis until or beyond

the breeding period, leading to poor reproductive

performance. Calving in summer season, early post-

partum complications, ‘‘clinically relevant urovagina,’’

and ‘‘clinically relevant endometritis’’ within 60 d

postpartum were risk factors for persistence of

endometritis. Furthermore, approximately 10% of cows

that did not show the signs of endometritis within 60 d

postpartum had a new diagnosis of endometritis beyond

60 d postpartum; some risk factors for the new diagnosis

of endometritis beyond 60 d postpartum were early

postpartum complications and ‘‘clinically relevant

urovagina.’’

Acknowledgments

The first author was supported by the Monbukaga-

kusho Scholarship of Japan for PhD study, and the

authors are thankful to the Ministry of Education,

Culture, Sports, Science and Technology of Japan for

financial support. The authors thank Mr. Nguyen Cong

Thinh and Ms. Naoko Mitsui for their help during

clinical examination and in collecting data. Our sincere

thanks are due to the herd owners involved in this study

for their cooperation.

References

[1] Sheldon IM, Lewis GS, LeBlanc S, Gilbert RO. Defining post-

partum uterine disease in cattle. Theriogenology 2006;65:

1516–30.

[2] Sheldon IM, Williams EJ, Miller ANA, Nash DM, Herath S.

Uterine diseases in cattle after parturition. Vet J 2008;176:

115–21.

G. Gautam et al. / Theriogenology 73 (2010) 168–179178

[3] Sheldon IM, Dobson H. Postpartum uterine health in cattle.

Anim Reprod Sci 2004;82–83:295–306.

[4] LeBlanc SJ. Postpartum uterine disease and dairy herd repro-

ductive performance: a review. Vet J 2008;176:102–14.

[5] 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;85:2223–36.

[6] Azawi OI. Postpartum uterine infection in cattle. Anim Reprod

Sci 2008;105:187–208.

[7] Gautam G, Nakao T, Yusuf M, Koike K. Prevalence of endome-

tritis during the postpartum period and its impact on subsequent

reproductive performance in two Japanese dairy herds. Anim

Reprod Sci 2009. doi:10.1016/j.anireprosci.2009.02.001.

[8] Gilbert RO, Shin ST, Guard CL, Erb HN, Frajblat M. Prevalence

of endometritis and its effects on reproductive performance of

dairy cows. Theriogenology 2005;64:1879–88.

[9] Gilbert RO, Shin ST, Guard CL, Erb HN. Incidence of endome-

tritis and effects on reproductive performance of dairy cows.

Theriogenology 1998;49:251.

[10] Kasimanickam R, Duffield TF, Foster RA, Gartley CJ, Leslie

KE, Walton JS, et al. Endometrial cytology and ultrasonography

for the detection of subclinical endometritis in postpartum dairy

cows. Theriogenology 2004;62:9–23.

[11] Barlund CS, Carruthers TD, Waldner CL, Palmer CW. A com-

parison of diagnostic techniques for postpartum endometritis in

dairy cattle. Theriogenology 2008;69:714–23.

[12] Williams EJ, Fischer DP, Pfeiffer DU, England GCW, Noakes

DE, Dobson H, et al. Clinical evaluation of postpartum vaginal

mucus reflects bacterial infection and the immune response in

cattle. Theriogenology 2005;63:102–17.

[13] McDougall S, Macaulay R, Compton C. Association between

endometritis diagnosis using a novel intravaginal device and

reproductive performance in dairy cattle. Anim Reprod Sci

2007;99:9–23.

[14] Runciman DJ, Anderson GA, Malmo J, Davis GM. Use of

postpartum vaginoscopic (visual vaginal) examination of dairy

cows for the diagnosis of endometritis and the association of

endometritis with reduced reproductive performance. Aust Vet J

2008;86:205–13.

[15] Nakao T, Moriyoshi M, Kawata K. The effect of ovarian

dysfunction and endometritis on subsequent reproductive per-

formance in high and medium producing dairy cows. Therio-

genology 1992;37:341–9.

[16] Kim IH, Kang HG. Risk factors for postpartum endometritis and

the effect of endometritis on reproductive performance in dairy

cows in Korea. J Reprod Dev 2003;49:485–91.

[17] Lewis GS. Uterine health and disorders. J Dairy Sci

1997;80:984–94.

[18] Kelton DF, Lissemore KD, Martin RE. Recommendations for

recording and calculating the incidence of selected clinical

diseases of dairy cattle. J Dairy Sci 1998;81:2502–9.

[19] 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;85:2237–49.

[20] Thurmond MC, Jameson CM, Picanso JP. Effect of intrauterine

antimicrobial treatment in reducing calving-to-conception inter-

val in cows with endometritis. J Am Vet Med Assoc

1993;121:436–40.

[21] Steffan JMA, Adriamanga S, Thibier M. Treatment of metritis

with antibiotics or prostaglandin F2 alpha and influence

of ovarian cyclicity in dairy cows. Am J Vet Res 1984;45:

1090–4.

[22] Callahan CJ, Horstman LA. Treatment of early postpartum

metritis in a dairy herd: response and subsequent fertility. Bovine

Practitioner 1987;22:124–8.

[23] Mejı́a ME, Lacau-Mengido IM. Endometritis treatment with a

PGF2 alpha analog does not improve reproductive performance

in a large dairy herd in Argentina. Theriogenology 2005;63:

1266–76.

[24] Nakao T, Moriyoshi M, Kawata K. Effect of postpartum intrau-

terine treatment with 2% polyvinylpyrrolidon-iodine solution on

reproductive efficiency in cows. Theriogenology 1988;30:

1033–43.

[25] Runciman DJ, Anderson GA, Malmo J, Davis GM. Effect of

intrauterine treatment with cephapirin on the reproductive per-

formance of seasonally calving dairy cows at risk of endometritis

following periparturient disease. Aust Vet J 2008;86:250–8.

[26] Gautam G, Nakao T. Prevalence of urovagina and its effects on

reproductive performance in Holstein cows. Theriogenology

2009;71:1451–61.

[27] Ferguson JD, Galligan DT, Thomsen N. Principal descriptors of

body condition score in Holstein cows. J Dairy Sci

1994;77:2695–703.

[28] SPSS for Windows, version 11.5. SPSS Inc., Chicago, IL, 2002.

[29] Curtis CR, Erb HN, Sniffen CJ, Smith RD, Kronfeld DS. Path

analysis of dry period nutrition, postpartum metabolic and

reproductive disorders, and mastitis in Holstein cows. J Dairy

Sci 1985;68:2347–60.

[30] Markusfeld O. Periparturient traits in seven high dairy herds.

Incidence rates, association with parity, and interrelationship

among traits. J Dairy Sci 1987;70:158–68.

[31] Bruun J, Ersboll AK, Alban L. Risk factors for metritis in Danish

dairy cows. Prev Vet Med 2002;54:179–90.

[32] Hendricks KEM, Bartolome JA, Melendez P, Risco C, Archbald

LF. Effect of repeated administration of PGF2a in the early

postpartum period on the prevalence of clinical endometritis and

probability of pregnancy at first insemination in lactating dairy

cows. Theriogenology 2006;65:1454–64.

[33] Knutti B, Busato A, Kupfer U. Reproductive efficiency of cows

with endometritis after treatment with intrauterine infusions or

prostaglandin injections or no treament. J Vet Med A

2000;47:609–15.

[34] Correa MT, Erb H, Scarlett J. Path analysis for seven postpartum

disorders of Holstein cows. J Dairy Sci 1993;76:1305–12.

[35] Kaneene JB, Miller R. Risk factors for metritis in Michigan dairy

cattle using herd- and cow-based modeling approaches. Prev Vet

Med 1995;23:183–200.

[36] Murray RD, Allison JD, Gard RP. Bovine endometritis: compara-

tive efficacy of alfaprostol and intrauterine therapies, and other

factors influencing clinical success. Vet Rec 1990;127:86–90.

[37] Elliot K, McMahon KJ, Gier HT, Marion GB. Uterus of the cow

after parturition; bacterial content. Am J Vet Res 1968;29:77–81.

[38] Griffin JFT, Hartigan PJ, Nunn WR. Non-specific uterine infec-

tion and bovine fertility. I. Infection patterns and endometritis

during the first seven weeks post-partum. Theriogenology

1974;1:91–106.

[39] Paisley LG, Mickelsen WD, Anderson PB. Mechanism and

therapy for retained fetal membranes and uterine infections of

cows: a review. Theriogenology 1986;25:353–81.

[40] Sheldon IM, Noakes DE, Rycroft AN, Dobson H. Effect of

postpartum manual examination of the vagina on uterine bacter-

ial contamination in cows. Vet Rec 2002;151:531–4.

G. Gautam et al. / Theriogenology 73 (2010) 168–179 179

[41] Galvao KN, Brittin SB, Frajblat M, Gilbert RO. Defining cutoff

points for subclinical endometritis at different stages of lactation

[abstract]. J Dairy Sci 2007;90(Suppl 1):13.

[42] Roman-Ponce H, Thatcher WW, Caton D, Barron DH, Wilcox

CJ. Thermal stress on uterine blood flow in dairy cows. J Anim

Sci 1978;46:175–80.

[43] Drillich M, Damaris R, Wittke M, Heuwieser W. Treatment

of chronic endometritis in dairy cows with an intrauterine

application of enzymes: a field trial. Theriogenology

2005;63:1811–23.

[44] Gilbert RO, Wilson DG, Levine SA, Bosu WT. Surgical manage-

ment of urovagina and associated infertility in a cow. J Am Vet

Med Assoc 1989;194:931–2.

[45] Easley KJ. Diagnosis and treatment of vesicovaginal reflux in the

mare. Vet Clin North Am Equine Pract 1988;4:407–16.

[46] Cai TQ, Wetson PG, Lund LA, Brodie B, McKenna DJ, Wagner

WC. Association between neutrophil functions and periparturi-

ent disorders in cows. Am J Vet Res 1994;55:934–43.

[47] Hussain AM. Bovine uterine defense mechanisms: a review. J

Vet Med Ser B 1989;36:641–51.

[48] Hussain AM, Daniel RCW. Bovine endometritis: current and

future alternative therapy. J Vet Med A 1991;38:641–51.

[49] Hussain AM, Daniel RCW. Phagocytosis by uterine fluid and

blood neutrophils and hematological changes in postpartum cows

following normal and abnormal parturition. Theriogenology

1992;37:1253–67.