- Email: [email protected]
Associations among Trueperella pyogenes, endometritis diagnosis, and pregnancy outcomes in dairy cows
Accepted Manuscript Associations among Trueperella pyogenes, endometritis diagnosis and pregnancy outcomes in dairy cows M.L.S. Bicalho, F.S. Lima, V.S. Machado, E.B. Meira, Jr., E.K. Ganda, C. Foditsch, R.C. Bicalho, R.O. Gilbert PII:
S0093-691X(15)00516-6
DOI:
10.1016/j.theriogenology.2015.09.043
Reference:
THE 13348
To appear in:
Theriogenology
Received Date: 23 February 2015 Revised Date:
17 September 2015
Accepted Date: 18 September 2015
Please cite this article as: Bicalho MLS, Lima FS, Machado VS, Meira Jr EB, Ganda EK, Foditsch C, Bicalho RC, Gilbert RO, Associations among Trueperella pyogenes, endometritis diagnosis and pregnancy outcomes in dairy cows, Theriogenology (2015), doi: 10.1016/j.theriogenology.2015.09.043. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 1
Revised
2
Associations among Trueperella pyogenes, endometritis diagnosis and pregnancy outcomes
3
in dairy cows
Foditschb, R.C. Bicalhob and R.O. Gilberta
5 6
a
Department of Clinical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853, USA.
SC
7 8
ᵇ Department of Population Medicine and Diagnostic Sciences, Cornell University,
11
Ithaca, NY 14853-6401, USA. c
M AN U
9 10
RI PT
M.L.S. Bicalhoa, F.S. Limac, V.S. Machadob, E.B. Meira Jrb, E.K. Gandab, C.
4
Department of Veterinary Clinical Medicine, College of Veterinary Medicine, University of Illinois, Urbana, IL 61802, USA.
12
TE D
13
Corresponding author: Robert O. Gilbert, Department of Clinical Sciences, College of Veterinary
15
Medicine, Cornell University, Ithaca, NY 14853-6401, Phone: 607 253-3435, Fax: 607 253-
16
3531, e-mail: [email protected]
18 19 20
AC C
17
EP
14
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 21
Abstract The objective of this study was to evaluate three commonly used methods for
23
endometritis diagnosis by comparing each one’s association with the presence of intrauterine
24
Trueperella pyogenes (TP) and reproductive performance. Lactating Holstein cows (n=452)
25
were evaluated a single time at 35 ± 3 days postpartum to diagnose endometritis based on three
26
criteria: presence of purulent vaginal discharge (PVD) detected by a Metricheck® device,
27
presence of purulent uterine lavage fluid (PUL), presence of cytological endometritis (CE) based
28
on relative abundance of polymorphonuclear leukocytes (PMNL) in uterine lavage fluid. A
29
threshold of PMNL ≥ 5% was used to diagnose the occurrence of CE. Also, a swab of the uterine
30
lavage was cultured to evaluate the presence of TP and determine its association with
31
endometritis diagnosis criteria and pregnancy outcomes. The results showed that cows positive
32
for TP had increased prevalence of PVD and PUL and tended to have greater prevalence of CE.
33
Median time to pregnancy was 56 days longer and hazard of pregnancy was 34% lower for TP-
34
positive cows than for TP-negative cows. Presence of PUL led to a 35% lower hazard of
35
pregnancy and 34 days longer median time to pregnancy than cows without PUL. Likewise,
36
cows diagnosed with PVD had a 47% lower hazard of pregnancy and 57 days longer median
37
time to pregnancy than cows without PVD. Cows diagnosed with CE had a 27% lower hazard of
38
pregnancy and 42 days longer median time to pregnancy than cows without CE. When the three
39
different diagnostic methods were used as independent variables in a Cox’s proportional hazard
40
model that evaluated hazard of pregnancy, the variable PVD was the only statistically significant
41
variable. Combined PUL and CE or combined PUL and PVD had no additional effects on hazard
42
of pregnancy when compared with only PUL, only PVD, or only CE as the criterion to determine
43
endometritis. However, combined PVD and CE had an additive, detrimental effect on
AC C
EP
TE D
M AN U
SC
RI PT
22
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis reproductive performance. The sensitivity and specificity of each of TP, PUL, PVD and CE for
45
predicting non-pregnancy at 300 days postpartum were all similar. In conclusion, TP-positive
46
cows had impaired reproductive performance and increased likelihood of PVD and PUL.
47
Additionally, TP-positive cows tended to have an increased prevalence of CE. Cows with PVD
48
had the lowest reproductive performance when compared to cows with PUL or CE, suggesting
49
that PVD at 35 days postpartum is a better criterion for the diagnosis of endometritis and
50
reproductive failure.
SC
RI PT
44
51
Keywords: Trueperella pyogenes; purulent uterine lavage; purulent vaginal discharge;
53
cytological endometritis; pregnancy
54 55
59
EP
58
AC C
57
TE D
56
M AN U
52
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 60
1. Introduction Postpartum uterine diseases such as metritis and endometritis are highly prevalent in
62
high-producing dairy cows and negatively affect reproductive performance and milk yield [1].
63
Bacteria commonly contaminate the uterine lumen after parturition, and Escherichia coli,
64
Trueperella pyogenes and Fusobacterium necrophorum are considered the most relevant species
65
involved in the pathogenesis of uterine diseases. It seems that T. pyogenes acts synergistically
66
with F. necrophorum, Bacteriodes spp. and Prevotella spp. to enhance the severity of uterine
67
disease [2-4]. Endometritis affects 15% to 42% of dairy cows, and T. pyogenes is recognized as
68
an important pathogen associated with this disease [5,6] because of its persistence in the
69
contaminated uterus [7,8].
M AN U
SC
RI PT
61
The presence of bacteria in the uterus causes inflammation, histological lesions of the
71
endometrium, and delays in uterine involution [4,9,10]. It has been shown that inflammation of
72
the endometrium has a detrimental effect on reproductive performance, reducing both the first
73
service conception rate and the overall pregnancy risk [11,12].
TE D
70
A recent study by de Boer et al. found that cows with any bacterial growth at 21 days
75
after parturition, irrespective of bacterial species, were less likely to conceive [13]. However,
76
several earlier studies have repeatedly described a specific association between the presence of T.
77
pyogenes and occurrence of purulent vaginal mucous exudate [2,3,14,15]. Moreover, cows
78
infected with T. pyogenes at 34-36 days postpartum were at greater odds of developing clinical
79
endometritis [6]. Machado et al., using metagenomics analysis, reported that the relative
80
abundance of Trueperella spp. was increased in endometritic cows [16]. It seems that the
81
presence of purulent contents in the uterine lumen and vagina of dairy cows is intertwined with
82
the presence of T. pyogenes. Therefore, a diagnostic tool to accurately identify the presence of
AC C
EP
74
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 83
purulent contents in the uterus and vagina would potentially be a good predictor of the presence
84
of this major pathogen and the detrimental impact it appears to have on the fertility of dairy
85
cows. Recently, various diagnostic tools such as uterine lavage, ultrasonographic examination,
87
bacterial culture, uterine cytology, and the Metricheck® device (Metricheck, SimcroTech,
88
Hamilton, New Zealand) have been used to identify the occurrence of endometritis in dairy cows
89
[7,16-19]. While some diagnostic tools may offer advantages (e.g. Metricheck is inexpensive,
90
non-invasive, and easy to perform), no consensus regarding the best predictor of negative impact
91
on reproductive outcomes has emerged. Dubuc et al. reported that cows having purulent vaginal
92
discharge, based on a Metricheck evaluation, did not necessarily a have higher percentage of
93
polymorphonuclear leukocytes (PMNL) in the uterine lumen when a cytological evaluation was
94
also performed [18]. Those authors also suggested that purulent vaginal discharge and
95
cytological endometritis constitute different risk factors, and that both conditions affect
96
subsequent reproductive performance [18].
TE D
M AN U
SC
RI PT
86
Another study used uterine bacteriology and cytology results to investigate false-positive
98
findings of clinical endometritis based on vaginoscopy [7]. Additionally, ultrasonographic
99
assessment of uterine fluid volume and endometrial thickness combined was not a good predictor
100
of reproductive performance [17]. Endometrial biopsy is not practical in the field, and neither
101
microbiological examination of uterine swabs nor the cytobrush technique yields a quick
102
diagnosis under field conditions, nor are the results they provide an accurate representation of the
103
uterine environment (e.g. sampling at different uterine locations could give different results). In
104
previous studies, our group performed uterine lavage in an attempt to obtain a more
AC C
EP
97
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 105
representative sample of luminal contents compared to samples obtained from uterine swabs and
106
biopsies [6,16,20]. Machado et al. demonstrated that uterine lavage sample optical density (ULSOD) is a
108
useful tool for assessment of uterine health in cows with greater than 18% neutrophils in uterine
109
lumen samples [16]. Cheong et al. investigated the use of a reagent strip test on uterine lavage
110
samples for the diagnosis of cytological endometritis, and reported strong associations of the test
111
results with cytological endometritis and reproductive impairment; however, in comparison with
112
conventional cytology, the performance of the reagent strip was relatively poor [21].
SC
RI PT
107
Availability of a simple, reliable tool to predict negative impacts of endometritis could
114
offer bovine veterinarians a practical aid to assess herd uterine health and estimate reproductive
115
outcomes. We identified three approaches as the most promising diagnostic tools for
116
endometritis/purulent vaginal discharge complex. These are uterine lavage (a more accurate
117
means of uterine sample collection), uterine cytology (which enables evaluation of uterine
118
inflammation not evident to the naked eye) and Metricheck (easiest to perform).
TE D
M AN U
113
Therefore, the objective of this study was to evaluate the above three commonly used
120
methods for endometritis diagnosis by comparing the strength of their associations with presence
121
of intrauterine T. pyogenes and with time to pregnancy in dairy cows.
AC C
122
EP
119
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 123 124
2. Materials and methods
125
2.1. Farm, management, and body condition (BCS) measurement In total, 452 Holstein lactating dairy cows from a single dairy farm located near Ithaca,
127
New York, were enrolled in the study at 35 ± 3 d postpartum. Cows were enrolled weekly from
128
October 2012 to January 2013. The farm was selected because of its longstanding working
129
relationship with the Ambulatory and Production Medicine Clinic at Cornell University. Cows
130
were housed in freestall barns with concrete stalls covered with mattresses and bedded with
131
waste dry manure solids. Cows were milked three times daily in a double 52-stall parallel
132
milking parlor. All cows were offered free-choice access to a TMR consisting of approximately
133
55% forage (corn silage, haylage, and wheat straw) and 45% concentrate (corn meal, soybean
134
meal, canola, cottonseed, and citrus pulp) on a dry matter basis of the diet. The diet was
135
formulated to meet or exceed the National Research Council nutrient requirements (NRC, 2001)
136
for lactating Holstein cows weighing 650 kg and producing 45 kg of 3.5% fat-corrected milk
137
[22].
TE D
M AN U
SC
RI PT
126
Reproductive management used a combination of Presynch [23], Ovsynch [24], Resynch
139
[25], and detection of estrus, with 25% to 30% of cows bred by timed artificial insemination
140
(TAI) and the remainder bred after detection of estrus solely by using activity monitors (Alpro,
141
DeLaval, Kansas City, MO). Estrus detection was based on electronic activity sensors worn
142
around the neck; no cows were artificially inseminated based on visual detection of estrus.
143
Activity sensors determined the level of activity in a cow, and the computer compared it to her
144
rolling activity over the previous week. Cows were then classified as having an activity of 0, 1,
145
2, or 3 and artificially inseminated when they attained an activity level of 3. Inseminations were
AC C
EP
138
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis conducted once daily. Cows bred by TAI were separated by automatic sort gates during the first
147
milking of the day for breeding, and cows that triggered the estrus activity monitoring system
148
were artificially inseminated after the milking following detection of estrus. Survival analysis
149
techniques were used to analyze the relationship between different diagnostic methods for
150
endometritis and reproductive performance. The reproductive outcome of interest was days from
151
calving to conception. Conception day was defined as the insemination day that resulted in
152
pregnancy. Pregnancy was diagnosed by rectal palpation at 39 ± 3-d since the last insemination.
153
All study cows were followed until 300 days postpartum or the date of culling (if ≤300 days)
154
from the herd. Data regarding health traits and appropriate measures of reproductive
155
performance (days from calving to conception) were obtained from DairyComp (DairyComp 305
156
Tulare, CA) records for the herd, and descriptive statistics were calculated (version 9.3, SAS
157
Institute Inc., Cary, NC).
SC
M AN U
Body condition scores (BCS) were recorded at time of enrollment using a five-point scale
160
TE D
158
RI PT
146
161
2.2. Criteria for endometritis diagnosis, Trueperella pyogenes identification
with a quarter-point system as described by Edmonson et al. (1989)[26] .
EP
159
Cows were enrolled on a weekly basis. All cows at 35 ± 3 d postpartum during the
163
enrollment period were included in the study. BCS was recorded and uterine samples were
164
collected at a single time for each cow under investigation and processed independently, as
165
described in the next sections. Study cows were identified and restrained in a head lock
166
stanchion, the perineum and vulva were cleansed with a paper towel and subjected to sample
167
collection according to the following procedures in succession:
AC C
162
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis First, vaginal discharge was retrieved using the Metricheck® device to determine the
169
presence of purulent vaginal discharge (PVD) as described by Williams et al. [3]. Briefly, a
170
scoring system based on a 0 to 3 scale was used (0 = no discharge and clear mucus, 1 = flecks of
171
purulent material within otherwise clear mucus, 2 = mucopurulent but <50% purulent material, 3
172
= mucopurulent with >50% purulent material and/or fetid discharge). Cows with a score of 3
173
were classified as PVD positive.
RI PT
168
Next, uterine lavage samples were collected to evaluate the presence of purulent content
175
in the uterine lavage and the presence of polymorphonuclear leukocytes (PMNL) and uterine
176
epithelial cells. Briefly, restrained cows had their perineum area re-cleansed and disinfected with
177
70% ethanol, and a plastic infusion pipette was introduced into the cranial vagina and
178
manipulated through the cervix into the uterus. A total of 20 mL of sterile saline solution was
179
infused into the uterus and agitated gently, and a sample of the fluid was aspirated. The volume
180
of recovered fluid ranged from 5 to 15 mL. All samples were visually scored by one investigator,
181
who assessed the presence of a purulent or mucopurulent secretion in the uterine lavage sample.
182
The score ranged from 0 to 2, with 0 indicating absence of a purulent or mucopurulent secretion
183
in the lavage sample, 1 indicating a bloody but non purulent sample, and 2 the presence of pus in
184
the lavage sample. Cows with a score of 2 were classified as purulent uterine lavage (PUL)
185
positive.
M AN U
TE D
EP
AC C
186
SC
174
Aliquots of the uterine lavage samples were kept on ice until they were processed in the
187
laboratory. Samples were cultured on Mueller-Hinton agar plates supplemented with 5%
188
defibrinated sheep blood selective for T. pyogenes (BBL TM 7, Becton, Dickinson and Co.,
189
Sparks, MD) and incubated aerobically in 7% CO2 at 38°C for 48 h. Colonies of T.
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 190
pyogenes were distinguished based on colony morphology, hemolysis, and Gram stain. Cows
191
infected with intrauterine Truperella pyogenes (TP) were classified as TP positive. Additionally, uterine lavage samples were examined to determine the presence of
193
cytological endometritis. Briefly, once in the laboratory, the uterine lavage samples were
194
vortexed to mix and resuspend cellular material, and an aliquot of 100 µl was transferred to a
195
cytocentrifuge chamber (Cyto-Tek®cytocentrifuge, Electron Microscopy Sciences, Hatfield, PA).
196
Samples were centrifuged onto glass slides at 1000 rpm for 3 min. The slides were then air-dried
197
and stained with modified Wright-Giemsa stain (Diff-Quik 1, Dade Diagnostics, West Monroe,
198
LA). All slides were examined by the same examiner using bright-field microscopy at 400×
199
magnification. The examiner counted 200 cells from each slide and the percentage of PMNL
200
among all other cells was calculated. Cows with PMNL ≥ 5% were classified as cytological
201
endometritis (CE) positive [12].
203
2.3. Statistical analyses
TE D
202
M AN U
SC
RI PT
192
Binary data such as prevalence of PUL, PVD and CE according to TP presence were
205
analyzed by logistic regression using the GLIMMIX procedure of SAS (version 9.3, SAS
206
Institute Inc., Cary, NC) fitting a binary distribution. The reproductive responses of interest were
207
time to pregnancy and hazard of pregnancy according to TP, PUL, PVD and CE. Time to
208
pregnancy was analyzed by survival analysis with the Cox proportional hazard model using the
209
PHREG procedure of SAS. Cows that left the study (due to death or sale of the cow) before 300
210
days postpartum were censored. The hazard ratio and respective 95% CI were calculated for
211
time-dependent categorical data. The model assumption of proportionality of hazards was
212
assessed by visual inspection of the Kaplan-Meier survival curves, which were generated using
AC C
EP
204
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis the LIFETEST procedure of SAS. Values of median days to pregnancy were also generated by
214
the Kaplan-Meier method using the LIFETEST procedure of SAS. A different Cox proportional
215
hazard model was built for TP presence and each of the endometritis diagnosis criteria. Three
216
additional models were built to evaluate potential additive effects of new variables that
217
combined: 1) PUL and CE, 2) PUL and PVD, and 3) PVD and CE.
RI PT
213
Kaplan-Meier survival graphs were generated using MedCalc version 10.4.0.0
219
(Mariakerke, Belgium). The time series variable for this model was the calving-to-conception
220
interval or days from calving until the end of the follow-up period; the minimum follow-up
221
period was 150 days and the maximum was 250.
M AN U
SC
218
Analyses in 2×2 tables for each of the endometritis diagnosis criteria were performed
223
using the Pearson chi-squared test (FREQ procedure of SAS). Sensitivity, specificity, negative
224
and positive likelihood ratio, negative and positive predictive value of TP presence and each
225
endometritis diagnosis criterion for predicting non-pregnancy were calculated.
227
Statistical significance was declared when the P-value was ≤ 0.05, whereas 0.05 < Pvalue ≤ 0.10 was regarded as a tendency.
228
EP
226
TE D
222
3. Results
230
3.1. Associations between endometritis diagnosis criteria and Trueperella pyogenes presence
231
AC C
229
As expected, the presence of TP was associated with increased risk of endometritis for all
232
criteria used for its diagnosis. Cows that were TP-positive had an increased prevalence of PUL
233
and PVD (P < 0.001) when compared with their counterparts that were TP-negative (Table 1).
234
Moreover, cows that were TP-positive tended to have a greater prevalence of CE than did TP-
235
negative cows (P= 0.09) (Table 1).
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 236 237
3.2. Reproductive performance according to endometritis diagnosis criteria and Trueperella
238
pyogenes presence The occurrence of endometritis defined by PUL, PVD and CE was associated with lower
240
reproductive performance. Cows with PUL had a 35% lower hazard of pregnancy (P < 0.01)
241
than cows without PUL (Fig. 1). Moreover, median time to pregnancy in cows with PUL was
242
increased (P < 0.02) by 44 d (168 vs. 124) when compared with cows without PUL (Fig. 1).
243
Likewise, cows diagnosed with PVD had a 47% lower hazard of pregnancy (P < 0.001) than
244
counterparts without PVD, and the median time to pregnancy was 57 d longer (175 vs. 118) for
245
cows with PVD (Fig.2). Cows diagnosed with CE had a 27% lower hazard of pregnancy (P =
246
0.05) than cows without CE, and the median time to pregnancy was 42 d longer (168 vs. 126) for
247
cows with CE (Fig. 3). When PUL, PVD and CE were used as covariates in the same Cox’s
248
proportional hazard model, only PVD was statistically different (P < 0.01) in the final model
249
(Table 2). In that model, PVD significantly reduced the hazard of pregnancy (hazard ratio = 0.59;
250
95% CI = 0.40 to 0.86), as shown in Table 2.
TE D
M AN U
SC
RI PT
239
Combined PUL and PVD or combined PUL and CE had no additive effects on hazard of
252
pregnancy when compared with using only PUL, only PVD, or only CE to determine
253
endometritis (Fig. 4 and 5). However, combined PVD and CE had a significant additive effect;
254
cows that were positive for both PVD and CE had the longest interval from calving to conception
255
when compared to non-affected cows or cows affected only with CE or PVD (Fig. 6).
257
AC C
256
EP
251
Reproductive performance of TP-positive cows was dramatically reduced compared to
TP-negative cows. The median time to pregnancy in TP-positive cows was increased (P < 0.02)
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 258
by 56 d (175 vs. 119) when compared with TP-negative cows (Fig. 7). Additionally, the hazard
259
of pregnancy was 34% lower for TP-positive cows than for TP-negative cows (Fig. 7).
260 3.3. Sensitivity, specificity, positive and negative likelihood ratios, positive and negative
262
predictive value for predicting pregnancy at 300 days postpartum
RI PT
261
The presence of TP and the endometritis diagnosis criteria PUL, PVD and CE had very
264
similar values for sensitivity, specificity, positive likelihood ratio, negative likelihood ratio,
265
positive predictive value, and negative predictive value for predicting non-pregnancy at 300 days
266
postpartum (Table 3). Overall, TP, PUL, PVD and CE had relatively high sensitivity (88.2% to
267
90.7%) and positive predictive value (69.6% to 79.2%), but relatively low specificity (13.6% to
268
22.1%) and negative predictive value (22.1% to 37.1%) for predicting non-pregnancy by 300
269
days postpartum (Table 3).
M AN U
SC
263
AC C
EP
TE D
270
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 271 272 273
4. Discussion Three commonly used criteria for diagnosis of endometritis (PVD, PUL and CE) were evaluated in this study by comparing their association with the presence of intrauterine T.
275
pyogenes (TP) and their deleterious effects on reproductive performance.
RI PT
274
T. pyogenes was associated with an increased likelihood of clinical endometritis (PUL
277
and PVD) and reproductive failure. Additionally, clinical endometritis (PUL and PVD) was
278
significantly associated with lower reproductive performance, and CE had a strong tendency to
279
be associated with decreased reproductive performance. However, when PUL, PVD and CE
280
were used as covariates in the same Cox’s proportional hazard model, only PVD was retained in
281
the model (P-value <0.05) and significantly associated with reduced hazard of pregnancy.
282
Moreover, an additive effect of PVD and CE was identified: cows affected with both PVD and
283
CE had a significantly longer interval from calving to conception when compared to healthy
284
cows and cows affected with only PVD or CE.
M AN U
TE D
286
Our results support an etiological role for T. pyogenes in the pathogenesis of endometritis/purulent vaginal discharge complex and subsequent reproductive failure.
EP
285
SC
276
These results are in general agreement with the findings of others, who showed that cows
288
that were culture positive for T. pyogenes were at a higher risk of having clinical endometritis as
289
diagnosed by vaginoscopy or ultrasound [27,28]. Olson and Williams [2,3] reported that cows
290
that were heavily infected (high colony-forming unit counts) with T. pyogenes and other
291
anaerobic bacteria were at much higher risk of having mucopurulent or purulent vaginal
292
discharge. Similarly, Mateus et al observed that cows with mild or severe endometritis had a
AC C
287
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 293
higher prevalence and persistence of T. pyogenes [27]. Another study identified T. pyogenes as
294
the most common bacterial species isolated from uterine biopsies [9]. In a previous study, we demonstrated that most of the bovine intrauterine T. pyogenes
296
isolates we detected carried the gene FimA, which encodes a fimbrial subunit that mediates
297
bacterial adhesion to endothelial cells [29]. FimA is an important virulence factor associated with
298
increased risk of both metritis and endometritis [6,29]. Moreover, in a subsequent study, we
299
found that cows at 34–36 days postpartum that were infected with intrauterine T. pyogenes had
300
higher odds of developing clinical endometritis compared to TP-negative cows [6].
SC
RI PT
295
Despite the strong association of T. pyogenes and endometritis, a study by Galvao et al.
302
reported that only 41% of cows with PVD were T. pyogenes culture positive at the time of PVD
303
diagnosis [30]. Moreover, Sens and Heuwieser found alpha-hemolytic Streptococcus to be
304
strongly associated with reduced reproductive performance compared to T. pyogenes [31].
305
Bacterial pathogens and inflammatory mediators affect reproductive performance apparently
306
independently and additively and may interrupt an established pregnancy [32]. It is known that
307
endometrial cells are capable of directly recognizing PAMPs (pathogen-associated molecular
308
patterns) via Toll-like receptors [33] and secreting inflammatory cytokines to initiate an
309
inflammatory response [34]. T. pyogenes was found to initiate a cellular inflammatory response
310
in contrast to the exotoxin pyolisin [35].
TE D
EP
AC C
311
M AN U
301
The impact of endometritis on reproductive performance is well described in the
312
literature. Purulent vaginal discharge detected around 21 days post-parturition has been linked to
313
a lower pregnancy rate and a longer time to pregnancy [36,37], and cows with PVD required
314
more inseminations per pregnancy [38]. McDougall and coworkers reported that conception to
315
first-service is negatively associated with increasing endometritis score detected by vaginal
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis discharge [39]. Moreover, Studer and Morrow found a positive association between days open
317
and conception rate and increased purulent vaginal discharge score in cows after parturition [14].
318
In the present study, combined PVD and CE had an additional effect of further reducing
319
the probability of pregnancy. This is in an agreement with the results of other studies that found
320
that cows with both purulent vaginal discharge and subclinical endometritis had the lowest
321
pregnancy per artificial insemination, highest pregnancy loss, and delayed time to pregnancy
322
compared with cows diagnosed with only one of the two diseases, or compared with cows having
323
no diagnosis of uterine disease [40,41]. Only one study did not find any effect of increased
324
endometrial PMNL infiltration on reproductive performance [42]. Elevated PMNL count at
325
cytological examination was associated with an increased time to pregnancy [43], and subclinical
326
endometritis was directly associated with poor body condition and had a negative effect on
327
reproductive traits of dairy cows [44].
M AN U
SC
RI PT
316
Likewise, cows with cytological endometritis diagnosed by the uterine lavage technique
329
were less likely to become pregnant before 120 days postpartum than non-endometritic cows
330
[12], and they took longer to become pregnant [12,43,45]. Also, the proportion of cows that
331
required more than three services was higher in cows that had uterine inflammation [44]. In
332
addition, Machado et al., who used a visual scoring system to evaluate uterine lavage samples,
333
demonstrated that cows with purulent uterine lavage were less likely to conceive and had a
334
greater median calving-to-conception interval [16]. The low-volume uterine lavage technique
335
used in that study and in the present study assesses the presence of a mucopurulent secretion only
336
inside the uterus, thereby helping to avoid false-positive diagnoses that might be more common
337
with other methods of diagnosis such as rectal palpation and visual inspection of vaginal
338
discharge.
AC C
EP
TE D
328
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis In a recent comprehensive review of the diagnostic criteria and methods for diagnosing
340
endometritis [46], the authors stated that more reports on reproductive outcomes are necessary to
341
improve the comparative data among diagnostic criteria. In the present study, overall, TP, PUL,
342
PVD and CE had relatively high sensitivities (88.2% to 90.7%) but relatively low specificities
343
(13.6% to 22.1%) for predicting non-pregnancy at 300 days postpartum. Only a few papers have
344
evaluated the impaired reproductive outcomes and PVD diagnosis by Metricheck [37,39,40].
RI PT
339
The validity of PVD evaluation for diagnosis of endometritis is confirmed by the
346
substantial impact of the condition on reproductive parameters found in our study and by very
347
similar values for sensitivity, specificity, positive likelihood ratio, negative likelihood ratio,
348
positive predictive value, and negative predictive value for predicting non-pregnancy in
349
comparison to PUL and CE. Moreover, when the three evaluated methods were included in the
350
same statistics model, only PVD was significantly associated with reduced hazard of pregnancy.
M AN U
SC
345
The use of Metricheck to evaluate PVD is a less invasive, and the characteristic color
352
and odor of vaginal discharge as well as endometritis score are highly correlated with T.
353
pyogenes in the uterus [1,36,47,48].
TE D
351
Finally, the strong association of purulent content with reduced fertility described in the
355
present study is likely the combined detrimental effect of all conditions that lead to purulent
356
uterine/vaginal discharge (e.g. vaginitis, cervicitis, endometritis, and inadequate uterine
357
involution). Although factors that cause lesions in the genital tract (uterus, cervix , and vagina)
358
such as dystocia, twin parturition, and retained placenta were not evaluated in this study,
359
previous studies have reported those factors to be highly associated with PVD [6,8,49].
AC C
EP
354
360
In conclusion, TP-positive cows had impaired reproductive performance and increased
361
likelihood of PVD and PUL. Additionally, TP-positive cows tended to have increased prevalence
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 362
of CE. Cows affected with PVD had the lowest reproductive performance when compared to
363
PUL and CE, suggesting that PVD at 35 d postpartum is a better criterion for the diagnosis of
364
endometritis and reproductive failure.
RI PT
365 ACKNOWLEDGMENTS
367
This research was supported in part by the Cornell University Agricultural
368
Experiment Station federal formula funds, Project No NYC-480450 (to ROG) received
369
from Cooperative State Research, Education and Extension Service, U.S. Department
370
of Agriculture. Any opinions, findings, conclusions, or recommendations expressed in
371
this publication are those of the authors and do not necessarily reflect the view of the
372
U.S. Department of Agriculture.
AC C
EP
TE D
M AN U
SC
366
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 373 References
375 376 377
[1] LeBlanc SJ, Duffield TF, Leslie KE, Bateman KG, Keefe GP, Walton JS, et al. Defining and diagnosing postpartum clinical endometritis and its impact on reproductive performance in dairy cows. J. Dairy Sci. 2002;85(9):2223-36.
378 379 380
[2] Olson JD, Ball L, Mortimer RG, Farin PW, Adney WS, Huffman EM. Aspects of bacteriology and endocrinology of cows with pyometra and retained fetal membranes. Am. J. Vet. Res. 1984;45(11):2251-5.
381 382 383
[3] Williams EJ, Fischer DP, Pfeiffer DU, England GC, Noakes DE, Dobson H, et al. Clinical evaluation of postpartum vaginal mucus reflects uterine bacterial infection and the immune response in cattle. Theriogenology 2005;63(1):102-17.
384 385
[4] Sheldon IM, Lewis GS, LeBlanc S, Gilbert RO. Defining postpartum uterine disease in cattle. Theriogenology 2006;65(8):1516-30.
386 387 388
[5] Bicalho RC, Machado VS, Bicalho ML, Gilbert RO, Teixeira AG, Caixeta LS, et al. Molecular and epidemiological characterization of bovine intrauterine escherichia coli. J. Dairy Sci. 2010;93(12):5818-30.
389 390 391
[6] Bicalho ML, Machado VS, Oikonomou G, Gilbert RO, Bicalho RC. Association between virulence factors of escherichia coli, fusobacterium necrophorum, and arcanobacterium pyogenes and uterine diseases of dairy cows. Vet. Microbiol. 2012;157(1-2):125-31.
392 393 394
[7] Westermann S, Drillich M, Kaufmann TB, Madoz LV, Heuwieser W. A clinical approach to determine false positive findings of clinical endometritis by vaginoscopy by the use of uterine bacteriology and cytology in dairy cows. Theriogenology 2010;74(7):1248-55.
395 396
[8] Potter TJ, Guitian J, Fishwick J, Gordon PJ, Sheldon IM. Risk factors for clinical endometritis in postpartum dairy cattle. Theriogenology 2010;74(1):127-34.
397 398 399
[9] Bonnett BN, Martin SW, Gannon VP, Miller RB, Etherington WG. Endometrial biopsy in holstein-friesian dairy cows. III. bacteriological analysis and correlations with histological findings. Can. J. Vet. Res. 1991;55(2):168-73.
400 401
[10] Sheldon IM, Noakes DE, Rycroft AN, Dobson H. The effect of intrauterine administration of estradiol on postpartum uterine involution in cattle. Theriogenology 2003;59(5-6):1357-71.
402 403 404
[11] 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(1-2):9-23.
AC C
EP
TE D
M AN U
SC
RI PT
374
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis [12] 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(9):1879-88.
407 408 409
[13] de Boer M, Buddle BM, Heuer C, Hussein H, Zheng T, LeBlanc SJ, et al. Associations between intrauterine bacterial infection, reproductive tract inflammation, and reproductive performance in pasture-based dairy cows. Theriogenology 2015;83(9):1514-24.
410 411 412
[14] Studer E, Morrow DA. Postpartum evaluation of bovine reproductive potential: Comparison of findings from genital tract examination per rectum, uterine culture, and endometrial biopsy. J. Am. Vet. Med. Assoc. 1978;172(4):489-94.
413 414 415
[15] Cordero OX, Wildschutte H, Kirkup B, Proehl S, Ngo L, Hussain F, et al. Ecological populations of bacteria act as socially cohesive units of antibiotic production and resistance. Science 2012;337(6099):1228-31.
416 417 418
[16] Machado VS, Knauer WA, Bicalho ML, Oikonomou G, Gilbert RO, Bicalho RC. A novel diagnostic technique to determine uterine health of holstein cows at 35 days postpartum. J. Dairy Sci. 2012;95(3):1349-57.
419 420
[17] Barlund CS, Carruthers TD, Waldner CL, Palmer CW. A comparison of diagnostic techniques for postpartum endometritis in dairy cattle. Theriogenology 2008;69(6):714-23.
421 422
[18] Dubuc J, Duffield TF, Leslie KE, Walton JS, LeBlanc SJ. Definitions and diagnosis of postpartum endometritis in dairy cows. J. Dairy Sci. 2010;93(11):5225-33.
423 424 425
[19] Machado VS, Oikonomou G, Bicalho ML, Knauer WA, Gilbert R, Bicalho RC. Investigation of postpartum dairy cows' uterine microbial diversity using metagenomic pyrosequencing of the 16S rRNA gene. Vet. Microbiol. 2012;159(3-4):460-9.
426 427 428 429
[20] Machado VS, Bicalho ML, Meira Junior EB, Rossi R, Ribeiro BL, Lima S, et al. Subcutaneous immunization with inactivated bacterial components and purified protein of escherichia coli, fusobacterium necrophorum and trueperella pyogenes prevents puerperal metritis in holstein dairy cows. PLoS One 2014;9(3):e91734.
430 431 432
[21] Cheong SH, Nydam DV, Galvao KN, Crosier BM, Ricci A, Caixeta LS, et al. Use of reagent test strips for diagnosis of endometritis in dairy cows. Theriogenology 2012;77(5):85864.
433 434
[22] Institute of Medicine (US) Committee on Examination of the Evolving Science for Dietary Supplements. 2002.
435 436 437
[23] Moreira F, Orlandi C, Risco CA, Mattos R, Lopes F, Thatcher WW. Effects of presynchronization and bovine somatotropin on pregnancy rates to a timed artificial insemination protocol in lactating dairy cows. J. Dairy Sci. 2001;84(7):1646-59.
AC C
EP
TE D
M AN U
SC
RI PT
405 406
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis [24] Pursley JR, Mee MO, Wiltbank MC. Synchronization of ovulation in dairy cows using PGF2alpha and GnRH. Theriogenology 1995;44(7):915-23.
440 441 442
[25] Fricke PM, Caraviello DZ, Weigel KA, Welle ML. Fertility of dairy cows after resynchronization of ovulation at three intervals following first timed insemination. J. Dairy Sci. 2003;86(12):3941-50.
443 444
[26] Edmonson A.J., Lean I.J., Weaver L.D., Farver T., Webster G.1989. A body condition scoring chart for Holstein dairy cows. J. Dairy Sci., 72 (1989), pp. 68–78
445 446 447
[27] Mateus L, da Costa LL, Bernardo F, Silva JR. Influence of puerperal uterine infection on uterine involution and postpartum ovarian activity in dairy cows. Reprod. Domest. Anim. 2002; 37(1):31-5.
448 449 450
[28] Brick TA, Schuenemann GM, Bas S, Daniels JB, Pinto CR, Rings DM, et al. Effect of intrauterine dextrose or antibiotic therapy on reproductive performance of lactating dairy cows diagnosed with clinical endometritis. J. Dairy Sci. 2012;95(4):1894-905.
451 452 453
[29] Santos TM, Caixeta LS, Machado VS, Rauf AK, Gilbert RO, Bicalho RC. Antimicrobial resistance and presence of virulence factor genes in arcanobacterium pyogenes isolated from the uterus of postpartum dairy cows. Vet. Microbiol. 2010;145(1-2):84-9.
454 455
[30] Galvao KN, Greco LF, Vilela JM, Sa Filho MF, Santos JE. Effect of intrauterine infusion of ceftiofur on uterine health and fertility in dairy cows. J. Dairy Sci. 2009;92(4):1532-42.
456 457 458
[31] Sens A, Heuwieser W. Presence of escherichia coli, trueperella pyogenes, alpha-hemolytic streptococci, and coagulase-negative staphylococci and prevalence of subclinical endometritis. J. Dairy Sci. 2013;96(10):6347-54.
459 460
[32] Gilbert RO. The effects of endometritis on the establishment of pregnancy in cattle. Reprod. Fertil. Dev. 2011;24(1):252-7.
461 462 463
[33] Davies D, Meade KG, Herath S, Eckersall PD, Gonzalez D, White JO, et al. Toll-like receptor and antimicrobial peptide expression in the bovine endometrium. Reprod. Biol. Endocrinol. 2008;6:53,7827-6-53.
464 465 466
[34] Wira CR, Grant-Tschudy KS, Crane-Godreau MA. Epithelial cells in the female reproductive tract: A central role as sentinels of immune protection. Am. J. Reprod. Immunol. 2005;53(2):65-76.
467 468 469
[35] Amos MR, Healey GD, Goldstone RJ, Mahan SM, Duvel A, Schuberth HJ, et al. Differential endometrial cell sensitivity to a cholesterol-dependent cytolysin links trueperella pyogenes to uterine disease in cattle. Biol. Reprod. 2014;90(3):54.
AC C
EP
TE D
M AN U
SC
RI PT
438 439
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis [36] LeBlanc SJ, Duffield TF, Leslie KE, Bateman KG, Keefe GP, Walton JS, et al. The effect of treatment of clinical endometritis on reproductive performance in dairy cows. J. Dairy Sci. 2002;85(9):2237-49.
473 474 475
[37] 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(9):369-78.
476 477 478
[38] Gautam G, Nakao T, Yusuf M, Koike K. Prevalence of endometritis during the postpartum period and its impact on subsequent reproductive performance in two japanese dairy herds. Anim. Reprod. Sci. 2009;116(3-4):175-87.
479 480 481
[39] 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(1-2):9-23.
482 483 484
[40] Lima FS, Bisinotto RS, Ribeiro ES, Greco LF, Ayres H, Favoreto MG, et al. Effects of 1 or 2 treatments with prostaglandin F(2)alpha on subclinical endometritis and fertility in lactating dairy cows inseminated by timed artificial insemination. J. Dairy Sci. 2013;96(10):6480-8.
485 486 487
[41] Dubuc J, Duffield TF, Leslie KE, Walton JS, Leblanc SJ. Randomized clinical trial of antibiotic and prostaglandin treatments for uterine health and reproductive performance in dairy cows. J. Dairy Sci. 2011;94(3):1325-38.
488 489 490
[42] Plontzke J, Madoz LV, De la Sota RL, Drillich M, Heuwieser W. Subclinical endometritis and its impact on reproductive performance in grazing dairy cattle in argentina. Anim. Reprod. Sci. 2010;122(1-2):52-7.
491 492 493
[43] Galvao KN, Frajblat M, Brittin SB, Butler WR, Guard CL, Gilbert RO. Effect of prostaglandin F2alpha on subclinical endometritis and fertility in dairy cows. J. Dairy Sci. 2009;92(10):4906-13.
494 495 496
[44] Bacha B, Regassa FG. Subclinical endometritis in zebu x friesian crossbred dairy cows: Its risk factors, association with subclinical mastitis and effect on reproductive performance. Trop. Anim. Health Prod. 2010;42(3):397-403.
497 498 499
[45] Cheong SH, Nydam DV, Galvao KN, Crosier BM, Gilbert RO. Cow-level and herd-level risk factors for subclinical endometritis in lactating holstein cows. J. Dairy Sci. 2011;94(2):76270.
500 501 502
[46] de Boer MW, LeBlanc SJ, Dubuc J, Meier S, Heuwieser W, Arlt S, et al. Invited review: Systematic review of diagnostic tests for reproductive-tract infection and inflammation in dairy cows. J. Dairy Sci. 2014;97(7):3983-99.
503 504
[47] Sheldon IM, Dobson H. Postpartum uterine health in cattle. Anim. Reprod. Sci. 2004;8283:295-306.
AC C
EP
TE D
M AN U
SC
RI PT
470 471 472
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 505 506
[48] Djonov V, Andres AC, Ziemiecki A. Vascular remodelling during the normal and malignant life cycle of the mammary gland. Microsc. Res. Tech. 2001;52(2):182-9.
507 508
[49] Dubuc J, Duffield TF, Leslie KE, Walton JS, LeBlanc SJ. Risk factors and effects of postpartum anovulation in dairy cows. J. Dairy Sci. 2012;95(4):1845-54.
RI PT
509
AC C
EP
TE D
M AN U
SC
510
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis Table 1: Prevalence of endometritis diagnosed by purulent uterine lavage (score>2), purulent
512
vaginal discharge (score>3), and cytology (PMNL ≥ 5%) at 35 ± 3 d postpartum in cows that
513
were negative or positive for the presence of Trueperella pyogenes.
514 Trueperella pyogenes Endometritis prevalence
Total n=452
Negative n=403
Positive n= 49
Purulent uterine lavage (PUL)
13.8%
8.9%
55.1%
Purulent vaginal discharge (PVD) 14.0%
8.4%
61.2%
< 0.001
Cytological endometritis (CE)
12.7%
11.9%
20.4%
0.09
PUL&PVD
8.1%
3.5%
47%
< 0.001
PUL&CE
3.5%
2.0%
16.3%
< 0.001
PVD&CE
4.4%
2.7%
18.4%
< 0.001
PUL&PVD&CE
2.9%
1.2%
16.3 %
< 0.001
P-value < 0.001
SC
M AN U
AC C
EP
TE D
515 516
RI PT
511
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis Table 2: Final Cox proportional hazard model of the association of purulent uterine lavage,
518
purulent vaginal discharge and cytological endometritis (considered together) with the hazard
519
ratio (HR) of pregnancy by 300 d after parturition in Holstein cows examined at 35 ± 3 d (n =
520
452) after calving. Coefficient
SE
HR
Purulent uterine lavage
-0.15
0.19
0.86
Purulent vaginal discharge
-0.53
0.20
0.59
Cytological endometritis
-0.15
0.18
0.86
521
AC C
EP
TE D
M AN U
522
95% CI
P-value
0.60-1.25
0.43
0.40-0.86
<0.01
0.61-1.22
0.41
SC
Variable
RI PT
517
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis Table 3: Sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR),
524
positive predictive value (PPV), negative predictive value (NPV) for predicting non-pregnancy
525
status 300 d after parturition for presence of Trueperella pyogenes (TP), purulent uterine lavage
526
(PUL), purulent vaginal discharge (PVD), and cytological endometritis (CE) at a single
527
examination performed on Holstein cows 35 ± 3 d after parturition.
Sensitivity, %
90.7 (87.7-93.8)
88.2(84.4-91.4)
Specificity, %
13.6 (6.9-20.2)
19.2 (12.2-28.1)
PLR
1.1 (0.9-1.4)
NLR
0.7 (0.4-1.2)
532 533
22.1 (14.6-31.3)
17.3 (10.6-26.0)
1.1 (1.0-1.2)
1.1 (1.0-1.3)
1.1 (1.0-1.2)
0.6 (0.4-1.0)
0.5 (0.3-0.8)
0.7 (0.4-1.1)
79.2 (74.9-83.1)
78.2 (73.8-82.2)
NPV, %
22.1 (18.0-26.5)
32.8 (21.2-46.1)
37.1 (25.1-50.4)
31.6 (19.8-45.4)
TE D
77.0%
EP
531
88.8 (85.0-91.9)
78.5 (74.1-82.5)
AC C
530
88.8 (85.0-91.9)
69.6 (54.2-82.5)
Number of cows
529
CE
PPV, %
Percent pregnant 528
PVD
SC
PUL
M AN U
TP
RI PT
523
452
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis
TE D
EP
545
Fig.1. Kaplan-Meier survival curves for proportion of non-pregnant cows at 300 days postpartum according to the presence of purulent uterine lavage (PUL) at 35 ± 3 d postpartum. Median time to pregnancy for cows with PUL and without PUL was 168 d (95% confidence interval [CI] = 134 to 177) and 124 d (95% CI = 114 to 134), respectively. The hazard of pregnancy at 300 d postpartum was lower (P < 0.01) for cows with PUL than for cows without PUL (hazard ratio = 0.65; 95% CI = 0.49 to 0.86).
AC C
535 536 537 538 539 540 541 542 543 544
M AN U
SC
RI PT
534
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis
EP
TE D
Fig.2. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the presence of purulent vaginal discharge (PVD) at 35 ± 3 d postpartum. Median time to pregnancy for cows with PVD and without PVD was 175 d (95% confidence interval [CI] = 154 to 220) and 118 d (95% CI = 112 to 131), respectively. The hazard of pregnancy at 300 d postpartum was lower (P < 0.001) for cows with PVD than for cows without PVD (hazard ratio = 0.53; 95% CI = 0.41 to 0.70).
AC C
547 548 549 550 551 552 553 554 555
M AN U
SC
RI PT
546
ACCEPTED MANUSCRIPT
EP
TE D
Fig.3. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the presence of cytological endometritis (CE) at 35 ± 3 d postpartum. Median time to pregnancy for cows with CE and without CE was 168 d (95% confidence interval [CI] = 114 to 209) and 126 d (95% CI = 115 to 135), respectively. The hazard of pregnancy at 300 d postpartum was lower (P = 0.05) for cows with CE than for cows without CE (hazard ratio = 0.73; 95% CI = 0.54 to 0.98).
AC C
556 557 558 559 560 561 562 563 564 565
M AN U
SC
RI PT
Therio-14-Bicalho-Endometritis Diagnosis
ACCEPTED MANUSCRIPT
EP
TE D
Fig.4. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to combined presence of purulent vaginal discharge (PVD) and purulent uterine lavage (PUL) at 35 ± 3 d postpartum. Median time to pregnancy for cows classified as healthy, only PVD, only PUL, and combined PVD and PUL were: 118 d (95% confidence interval [CI] = 111 to 131), 211 d (95% CI = 134 to 223), 121 d (95% CI = 95 to 218), 174 d (95% CI = 154 to 217), respectively. The hazard of pregnancy at 300 d postpartum did not differ between combined PVD + PUL and only PVD (hazard ratio = 1.19; 95% CI = 0.73 to 1.93) or only PUL (hazard ratio = 0.78; 95% CI = 0.45 to 1.33).
AC C
566 567 568 569 570 571 572 573 574 575 576
M AN U
SC
RI PT
Therio-14-Bicalho-Endometritis Diagnosis
ACCEPTED MANUSCRIPT
TE D
EP
588
Fig.5. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the combined presence of purulent uterine lavage (PUL) and cytological endometritis (CE) at 35 ± 3 d postpartum. Median time to pregnancy for cows classified as healthy, only PUL, only CE, and combined PUL and CE were: 119 d (95% confidence interval [CI] = 113 to 134), 139 d (95% CI = 132 to 197), 126d (95% CI = 95 to 209), 177 d (95% CI = 135 to 177), respectively. The hazard of pregnancy at 300 d postpartum did not differ between combined PUL + CE and only PUL (hazard ratio = 0.67; 95% CI = 0.38 to 1.17) or only CE (hazard ratio = 0.58; 95% CI = 0.32 to 1.03).
AC C
577 578 579 580 581 582 583 584 585 586 587
M AN U
SC
RI PT
Therio-14-Bicalho-Endometritis Diagnosis
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis
603 604 605
EP
TE D
Fig.6. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the combined presence of purulent vaginal discharge (PVD) and cytological endometritis (CE) at 35 ± 3 d postpartum. Median time to pregnancy for cows classified as healthy, only PVD, only CE, and combined PVD and CE were: 118 d (95% confidence interval [CI] = 112 to 131), 172 d (95% CI = 136 to 220), 118 d (95% CI = 93 to 190), 207 d (95% CI = 168 to 223), respectively. The hazard of pregnancy at 300 d postpartum did not differ between combined PVD + CE and only PVD (hazard ratio = 0.67; 95% CI = 0.38 to 1.17), but combined PVD and CE had a lower hazard of pregnancy than only CE (hazard ratio = 0.52; 95% CI = 0.30 to 0.90).
AC C
591 592 593 594 595 596 597 598 599 600 601 602
M AN U
SC
RI PT
589 590
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Therio-14-Bicalho-Endometritis Diagnosis
606
TE D
EP
614
Fig.7. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the presence of Trueperella pyogenes (TP) at 35 ± 3 d postpartum. Median time to pregnancy for TP positive and TP negative was 175 d (95% confidence interval [CI] = 136 to 211) and 119 d (95% CI = 114 to 134), respectively. The hazard of pregnancy at 300 d postpartum was lower (P = 0.02) for TP-positive than TP-negative cows (hazard ratio = 0.66; 95% CI = 0.48 to 0.90).
AC C
607 608 609 610 611 612 613
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis
AC C
EP
TE D
M AN U
SC
RI PT
615
S0093-691X(15)00516-6
DOI:
10.1016/j.theriogenology.2015.09.043
Reference:
THE 13348
To appear in:
Theriogenology
Received Date: 23 February 2015 Revised Date:
17 September 2015
Accepted Date: 18 September 2015
Please cite this article as: Bicalho MLS, Lima FS, Machado VS, Meira Jr EB, Ganda EK, Foditsch C, Bicalho RC, Gilbert RO, Associations among Trueperella pyogenes, endometritis diagnosis and pregnancy outcomes in dairy cows, Theriogenology (2015), doi: 10.1016/j.theriogenology.2015.09.043. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 1
Revised
2
Associations among Trueperella pyogenes, endometritis diagnosis and pregnancy outcomes
3
in dairy cows
Foditschb, R.C. Bicalhob and R.O. Gilberta
5 6
a
Department of Clinical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853, USA.
SC
7 8
ᵇ Department of Population Medicine and Diagnostic Sciences, Cornell University,
11
Ithaca, NY 14853-6401, USA. c
M AN U
9 10
RI PT
M.L.S. Bicalhoa, F.S. Limac, V.S. Machadob, E.B. Meira Jrb, E.K. Gandab, C.
4
Department of Veterinary Clinical Medicine, College of Veterinary Medicine, University of Illinois, Urbana, IL 61802, USA.
12
TE D
13
Corresponding author: Robert O. Gilbert, Department of Clinical Sciences, College of Veterinary
15
Medicine, Cornell University, Ithaca, NY 14853-6401, Phone: 607 253-3435, Fax: 607 253-
16
3531, e-mail: [email protected]
18 19 20
AC C
17
EP
14
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 21
Abstract The objective of this study was to evaluate three commonly used methods for
23
endometritis diagnosis by comparing each one’s association with the presence of intrauterine
24
Trueperella pyogenes (TP) and reproductive performance. Lactating Holstein cows (n=452)
25
were evaluated a single time at 35 ± 3 days postpartum to diagnose endometritis based on three
26
criteria: presence of purulent vaginal discharge (PVD) detected by a Metricheck® device,
27
presence of purulent uterine lavage fluid (PUL), presence of cytological endometritis (CE) based
28
on relative abundance of polymorphonuclear leukocytes (PMNL) in uterine lavage fluid. A
29
threshold of PMNL ≥ 5% was used to diagnose the occurrence of CE. Also, a swab of the uterine
30
lavage was cultured to evaluate the presence of TP and determine its association with
31
endometritis diagnosis criteria and pregnancy outcomes. The results showed that cows positive
32
for TP had increased prevalence of PVD and PUL and tended to have greater prevalence of CE.
33
Median time to pregnancy was 56 days longer and hazard of pregnancy was 34% lower for TP-
34
positive cows than for TP-negative cows. Presence of PUL led to a 35% lower hazard of
35
pregnancy and 34 days longer median time to pregnancy than cows without PUL. Likewise,
36
cows diagnosed with PVD had a 47% lower hazard of pregnancy and 57 days longer median
37
time to pregnancy than cows without PVD. Cows diagnosed with CE had a 27% lower hazard of
38
pregnancy and 42 days longer median time to pregnancy than cows without CE. When the three
39
different diagnostic methods were used as independent variables in a Cox’s proportional hazard
40
model that evaluated hazard of pregnancy, the variable PVD was the only statistically significant
41
variable. Combined PUL and CE or combined PUL and PVD had no additional effects on hazard
42
of pregnancy when compared with only PUL, only PVD, or only CE as the criterion to determine
43
endometritis. However, combined PVD and CE had an additive, detrimental effect on
AC C
EP
TE D
M AN U
SC
RI PT
22
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis reproductive performance. The sensitivity and specificity of each of TP, PUL, PVD and CE for
45
predicting non-pregnancy at 300 days postpartum were all similar. In conclusion, TP-positive
46
cows had impaired reproductive performance and increased likelihood of PVD and PUL.
47
Additionally, TP-positive cows tended to have an increased prevalence of CE. Cows with PVD
48
had the lowest reproductive performance when compared to cows with PUL or CE, suggesting
49
that PVD at 35 days postpartum is a better criterion for the diagnosis of endometritis and
50
reproductive failure.
SC
RI PT
44
51
Keywords: Trueperella pyogenes; purulent uterine lavage; purulent vaginal discharge;
53
cytological endometritis; pregnancy
54 55
59
EP
58
AC C
57
TE D
56
M AN U
52
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 60
1. Introduction Postpartum uterine diseases such as metritis and endometritis are highly prevalent in
62
high-producing dairy cows and negatively affect reproductive performance and milk yield [1].
63
Bacteria commonly contaminate the uterine lumen after parturition, and Escherichia coli,
64
Trueperella pyogenes and Fusobacterium necrophorum are considered the most relevant species
65
involved in the pathogenesis of uterine diseases. It seems that T. pyogenes acts synergistically
66
with F. necrophorum, Bacteriodes spp. and Prevotella spp. to enhance the severity of uterine
67
disease [2-4]. Endometritis affects 15% to 42% of dairy cows, and T. pyogenes is recognized as
68
an important pathogen associated with this disease [5,6] because of its persistence in the
69
contaminated uterus [7,8].
M AN U
SC
RI PT
61
The presence of bacteria in the uterus causes inflammation, histological lesions of the
71
endometrium, and delays in uterine involution [4,9,10]. It has been shown that inflammation of
72
the endometrium has a detrimental effect on reproductive performance, reducing both the first
73
service conception rate and the overall pregnancy risk [11,12].
TE D
70
A recent study by de Boer et al. found that cows with any bacterial growth at 21 days
75
after parturition, irrespective of bacterial species, were less likely to conceive [13]. However,
76
several earlier studies have repeatedly described a specific association between the presence of T.
77
pyogenes and occurrence of purulent vaginal mucous exudate [2,3,14,15]. Moreover, cows
78
infected with T. pyogenes at 34-36 days postpartum were at greater odds of developing clinical
79
endometritis [6]. Machado et al., using metagenomics analysis, reported that the relative
80
abundance of Trueperella spp. was increased in endometritic cows [16]. It seems that the
81
presence of purulent contents in the uterine lumen and vagina of dairy cows is intertwined with
82
the presence of T. pyogenes. Therefore, a diagnostic tool to accurately identify the presence of
AC C
EP
74
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 83
purulent contents in the uterus and vagina would potentially be a good predictor of the presence
84
of this major pathogen and the detrimental impact it appears to have on the fertility of dairy
85
cows. Recently, various diagnostic tools such as uterine lavage, ultrasonographic examination,
87
bacterial culture, uterine cytology, and the Metricheck® device (Metricheck, SimcroTech,
88
Hamilton, New Zealand) have been used to identify the occurrence of endometritis in dairy cows
89
[7,16-19]. While some diagnostic tools may offer advantages (e.g. Metricheck is inexpensive,
90
non-invasive, and easy to perform), no consensus regarding the best predictor of negative impact
91
on reproductive outcomes has emerged. Dubuc et al. reported that cows having purulent vaginal
92
discharge, based on a Metricheck evaluation, did not necessarily a have higher percentage of
93
polymorphonuclear leukocytes (PMNL) in the uterine lumen when a cytological evaluation was
94
also performed [18]. Those authors also suggested that purulent vaginal discharge and
95
cytological endometritis constitute different risk factors, and that both conditions affect
96
subsequent reproductive performance [18].
TE D
M AN U
SC
RI PT
86
Another study used uterine bacteriology and cytology results to investigate false-positive
98
findings of clinical endometritis based on vaginoscopy [7]. Additionally, ultrasonographic
99
assessment of uterine fluid volume and endometrial thickness combined was not a good predictor
100
of reproductive performance [17]. Endometrial biopsy is not practical in the field, and neither
101
microbiological examination of uterine swabs nor the cytobrush technique yields a quick
102
diagnosis under field conditions, nor are the results they provide an accurate representation of the
103
uterine environment (e.g. sampling at different uterine locations could give different results). In
104
previous studies, our group performed uterine lavage in an attempt to obtain a more
AC C
EP
97
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 105
representative sample of luminal contents compared to samples obtained from uterine swabs and
106
biopsies [6,16,20]. Machado et al. demonstrated that uterine lavage sample optical density (ULSOD) is a
108
useful tool for assessment of uterine health in cows with greater than 18% neutrophils in uterine
109
lumen samples [16]. Cheong et al. investigated the use of a reagent strip test on uterine lavage
110
samples for the diagnosis of cytological endometritis, and reported strong associations of the test
111
results with cytological endometritis and reproductive impairment; however, in comparison with
112
conventional cytology, the performance of the reagent strip was relatively poor [21].
SC
RI PT
107
Availability of a simple, reliable tool to predict negative impacts of endometritis could
114
offer bovine veterinarians a practical aid to assess herd uterine health and estimate reproductive
115
outcomes. We identified three approaches as the most promising diagnostic tools for
116
endometritis/purulent vaginal discharge complex. These are uterine lavage (a more accurate
117
means of uterine sample collection), uterine cytology (which enables evaluation of uterine
118
inflammation not evident to the naked eye) and Metricheck (easiest to perform).
TE D
M AN U
113
Therefore, the objective of this study was to evaluate the above three commonly used
120
methods for endometritis diagnosis by comparing the strength of their associations with presence
121
of intrauterine T. pyogenes and with time to pregnancy in dairy cows.
AC C
122
EP
119
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 123 124
2. Materials and methods
125
2.1. Farm, management, and body condition (BCS) measurement In total, 452 Holstein lactating dairy cows from a single dairy farm located near Ithaca,
127
New York, were enrolled in the study at 35 ± 3 d postpartum. Cows were enrolled weekly from
128
October 2012 to January 2013. The farm was selected because of its longstanding working
129
relationship with the Ambulatory and Production Medicine Clinic at Cornell University. Cows
130
were housed in freestall barns with concrete stalls covered with mattresses and bedded with
131
waste dry manure solids. Cows were milked three times daily in a double 52-stall parallel
132
milking parlor. All cows were offered free-choice access to a TMR consisting of approximately
133
55% forage (corn silage, haylage, and wheat straw) and 45% concentrate (corn meal, soybean
134
meal, canola, cottonseed, and citrus pulp) on a dry matter basis of the diet. The diet was
135
formulated to meet or exceed the National Research Council nutrient requirements (NRC, 2001)
136
for lactating Holstein cows weighing 650 kg and producing 45 kg of 3.5% fat-corrected milk
137
[22].
TE D
M AN U
SC
RI PT
126
Reproductive management used a combination of Presynch [23], Ovsynch [24], Resynch
139
[25], and detection of estrus, with 25% to 30% of cows bred by timed artificial insemination
140
(TAI) and the remainder bred after detection of estrus solely by using activity monitors (Alpro,
141
DeLaval, Kansas City, MO). Estrus detection was based on electronic activity sensors worn
142
around the neck; no cows were artificially inseminated based on visual detection of estrus.
143
Activity sensors determined the level of activity in a cow, and the computer compared it to her
144
rolling activity over the previous week. Cows were then classified as having an activity of 0, 1,
145
2, or 3 and artificially inseminated when they attained an activity level of 3. Inseminations were
AC C
EP
138
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis conducted once daily. Cows bred by TAI were separated by automatic sort gates during the first
147
milking of the day for breeding, and cows that triggered the estrus activity monitoring system
148
were artificially inseminated after the milking following detection of estrus. Survival analysis
149
techniques were used to analyze the relationship between different diagnostic methods for
150
endometritis and reproductive performance. The reproductive outcome of interest was days from
151
calving to conception. Conception day was defined as the insemination day that resulted in
152
pregnancy. Pregnancy was diagnosed by rectal palpation at 39 ± 3-d since the last insemination.
153
All study cows were followed until 300 days postpartum or the date of culling (if ≤300 days)
154
from the herd. Data regarding health traits and appropriate measures of reproductive
155
performance (days from calving to conception) were obtained from DairyComp (DairyComp 305
156
Tulare, CA) records for the herd, and descriptive statistics were calculated (version 9.3, SAS
157
Institute Inc., Cary, NC).
SC
M AN U
Body condition scores (BCS) were recorded at time of enrollment using a five-point scale
160
TE D
158
RI PT
146
161
2.2. Criteria for endometritis diagnosis, Trueperella pyogenes identification
with a quarter-point system as described by Edmonson et al. (1989)[26] .
EP
159
Cows were enrolled on a weekly basis. All cows at 35 ± 3 d postpartum during the
163
enrollment period were included in the study. BCS was recorded and uterine samples were
164
collected at a single time for each cow under investigation and processed independently, as
165
described in the next sections. Study cows were identified and restrained in a head lock
166
stanchion, the perineum and vulva were cleansed with a paper towel and subjected to sample
167
collection according to the following procedures in succession:
AC C
162
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis First, vaginal discharge was retrieved using the Metricheck® device to determine the
169
presence of purulent vaginal discharge (PVD) as described by Williams et al. [3]. Briefly, a
170
scoring system based on a 0 to 3 scale was used (0 = no discharge and clear mucus, 1 = flecks of
171
purulent material within otherwise clear mucus, 2 = mucopurulent but <50% purulent material, 3
172
= mucopurulent with >50% purulent material and/or fetid discharge). Cows with a score of 3
173
were classified as PVD positive.
RI PT
168
Next, uterine lavage samples were collected to evaluate the presence of purulent content
175
in the uterine lavage and the presence of polymorphonuclear leukocytes (PMNL) and uterine
176
epithelial cells. Briefly, restrained cows had their perineum area re-cleansed and disinfected with
177
70% ethanol, and a plastic infusion pipette was introduced into the cranial vagina and
178
manipulated through the cervix into the uterus. A total of 20 mL of sterile saline solution was
179
infused into the uterus and agitated gently, and a sample of the fluid was aspirated. The volume
180
of recovered fluid ranged from 5 to 15 mL. All samples were visually scored by one investigator,
181
who assessed the presence of a purulent or mucopurulent secretion in the uterine lavage sample.
182
The score ranged from 0 to 2, with 0 indicating absence of a purulent or mucopurulent secretion
183
in the lavage sample, 1 indicating a bloody but non purulent sample, and 2 the presence of pus in
184
the lavage sample. Cows with a score of 2 were classified as purulent uterine lavage (PUL)
185
positive.
M AN U
TE D
EP
AC C
186
SC
174
Aliquots of the uterine lavage samples were kept on ice until they were processed in the
187
laboratory. Samples were cultured on Mueller-Hinton agar plates supplemented with 5%
188
defibrinated sheep blood selective for T. pyogenes (BBL TM 7, Becton, Dickinson and Co.,
189
Sparks, MD) and incubated aerobically in 7% CO2 at 38°C for 48 h. Colonies of T.
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 190
pyogenes were distinguished based on colony morphology, hemolysis, and Gram stain. Cows
191
infected with intrauterine Truperella pyogenes (TP) were classified as TP positive. Additionally, uterine lavage samples were examined to determine the presence of
193
cytological endometritis. Briefly, once in the laboratory, the uterine lavage samples were
194
vortexed to mix and resuspend cellular material, and an aliquot of 100 µl was transferred to a
195
cytocentrifuge chamber (Cyto-Tek®cytocentrifuge, Electron Microscopy Sciences, Hatfield, PA).
196
Samples were centrifuged onto glass slides at 1000 rpm for 3 min. The slides were then air-dried
197
and stained with modified Wright-Giemsa stain (Diff-Quik 1, Dade Diagnostics, West Monroe,
198
LA). All slides were examined by the same examiner using bright-field microscopy at 400×
199
magnification. The examiner counted 200 cells from each slide and the percentage of PMNL
200
among all other cells was calculated. Cows with PMNL ≥ 5% were classified as cytological
201
endometritis (CE) positive [12].
203
2.3. Statistical analyses
TE D
202
M AN U
SC
RI PT
192
Binary data such as prevalence of PUL, PVD and CE according to TP presence were
205
analyzed by logistic regression using the GLIMMIX procedure of SAS (version 9.3, SAS
206
Institute Inc., Cary, NC) fitting a binary distribution. The reproductive responses of interest were
207
time to pregnancy and hazard of pregnancy according to TP, PUL, PVD and CE. Time to
208
pregnancy was analyzed by survival analysis with the Cox proportional hazard model using the
209
PHREG procedure of SAS. Cows that left the study (due to death or sale of the cow) before 300
210
days postpartum were censored. The hazard ratio and respective 95% CI were calculated for
211
time-dependent categorical data. The model assumption of proportionality of hazards was
212
assessed by visual inspection of the Kaplan-Meier survival curves, which were generated using
AC C
EP
204
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis the LIFETEST procedure of SAS. Values of median days to pregnancy were also generated by
214
the Kaplan-Meier method using the LIFETEST procedure of SAS. A different Cox proportional
215
hazard model was built for TP presence and each of the endometritis diagnosis criteria. Three
216
additional models were built to evaluate potential additive effects of new variables that
217
combined: 1) PUL and CE, 2) PUL and PVD, and 3) PVD and CE.
RI PT
213
Kaplan-Meier survival graphs were generated using MedCalc version 10.4.0.0
219
(Mariakerke, Belgium). The time series variable for this model was the calving-to-conception
220
interval or days from calving until the end of the follow-up period; the minimum follow-up
221
period was 150 days and the maximum was 250.
M AN U
SC
218
Analyses in 2×2 tables for each of the endometritis diagnosis criteria were performed
223
using the Pearson chi-squared test (FREQ procedure of SAS). Sensitivity, specificity, negative
224
and positive likelihood ratio, negative and positive predictive value of TP presence and each
225
endometritis diagnosis criterion for predicting non-pregnancy were calculated.
227
Statistical significance was declared when the P-value was ≤ 0.05, whereas 0.05 < Pvalue ≤ 0.10 was regarded as a tendency.
228
EP
226
TE D
222
3. Results
230
3.1. Associations between endometritis diagnosis criteria and Trueperella pyogenes presence
231
AC C
229
As expected, the presence of TP was associated with increased risk of endometritis for all
232
criteria used for its diagnosis. Cows that were TP-positive had an increased prevalence of PUL
233
and PVD (P < 0.001) when compared with their counterparts that were TP-negative (Table 1).
234
Moreover, cows that were TP-positive tended to have a greater prevalence of CE than did TP-
235
negative cows (P= 0.09) (Table 1).
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 236 237
3.2. Reproductive performance according to endometritis diagnosis criteria and Trueperella
238
pyogenes presence The occurrence of endometritis defined by PUL, PVD and CE was associated with lower
240
reproductive performance. Cows with PUL had a 35% lower hazard of pregnancy (P < 0.01)
241
than cows without PUL (Fig. 1). Moreover, median time to pregnancy in cows with PUL was
242
increased (P < 0.02) by 44 d (168 vs. 124) when compared with cows without PUL (Fig. 1).
243
Likewise, cows diagnosed with PVD had a 47% lower hazard of pregnancy (P < 0.001) than
244
counterparts without PVD, and the median time to pregnancy was 57 d longer (175 vs. 118) for
245
cows with PVD (Fig.2). Cows diagnosed with CE had a 27% lower hazard of pregnancy (P =
246
0.05) than cows without CE, and the median time to pregnancy was 42 d longer (168 vs. 126) for
247
cows with CE (Fig. 3). When PUL, PVD and CE were used as covariates in the same Cox’s
248
proportional hazard model, only PVD was statistically different (P < 0.01) in the final model
249
(Table 2). In that model, PVD significantly reduced the hazard of pregnancy (hazard ratio = 0.59;
250
95% CI = 0.40 to 0.86), as shown in Table 2.
TE D
M AN U
SC
RI PT
239
Combined PUL and PVD or combined PUL and CE had no additive effects on hazard of
252
pregnancy when compared with using only PUL, only PVD, or only CE to determine
253
endometritis (Fig. 4 and 5). However, combined PVD and CE had a significant additive effect;
254
cows that were positive for both PVD and CE had the longest interval from calving to conception
255
when compared to non-affected cows or cows affected only with CE or PVD (Fig. 6).
257
AC C
256
EP
251
Reproductive performance of TP-positive cows was dramatically reduced compared to
TP-negative cows. The median time to pregnancy in TP-positive cows was increased (P < 0.02)
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 258
by 56 d (175 vs. 119) when compared with TP-negative cows (Fig. 7). Additionally, the hazard
259
of pregnancy was 34% lower for TP-positive cows than for TP-negative cows (Fig. 7).
260 3.3. Sensitivity, specificity, positive and negative likelihood ratios, positive and negative
262
predictive value for predicting pregnancy at 300 days postpartum
RI PT
261
The presence of TP and the endometritis diagnosis criteria PUL, PVD and CE had very
264
similar values for sensitivity, specificity, positive likelihood ratio, negative likelihood ratio,
265
positive predictive value, and negative predictive value for predicting non-pregnancy at 300 days
266
postpartum (Table 3). Overall, TP, PUL, PVD and CE had relatively high sensitivity (88.2% to
267
90.7%) and positive predictive value (69.6% to 79.2%), but relatively low specificity (13.6% to
268
22.1%) and negative predictive value (22.1% to 37.1%) for predicting non-pregnancy by 300
269
days postpartum (Table 3).
M AN U
SC
263
AC C
EP
TE D
270
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 271 272 273
4. Discussion Three commonly used criteria for diagnosis of endometritis (PVD, PUL and CE) were evaluated in this study by comparing their association with the presence of intrauterine T.
275
pyogenes (TP) and their deleterious effects on reproductive performance.
RI PT
274
T. pyogenes was associated with an increased likelihood of clinical endometritis (PUL
277
and PVD) and reproductive failure. Additionally, clinical endometritis (PUL and PVD) was
278
significantly associated with lower reproductive performance, and CE had a strong tendency to
279
be associated with decreased reproductive performance. However, when PUL, PVD and CE
280
were used as covariates in the same Cox’s proportional hazard model, only PVD was retained in
281
the model (P-value <0.05) and significantly associated with reduced hazard of pregnancy.
282
Moreover, an additive effect of PVD and CE was identified: cows affected with both PVD and
283
CE had a significantly longer interval from calving to conception when compared to healthy
284
cows and cows affected with only PVD or CE.
M AN U
TE D
286
Our results support an etiological role for T. pyogenes in the pathogenesis of endometritis/purulent vaginal discharge complex and subsequent reproductive failure.
EP
285
SC
276
These results are in general agreement with the findings of others, who showed that cows
288
that were culture positive for T. pyogenes were at a higher risk of having clinical endometritis as
289
diagnosed by vaginoscopy or ultrasound [27,28]. Olson and Williams [2,3] reported that cows
290
that were heavily infected (high colony-forming unit counts) with T. pyogenes and other
291
anaerobic bacteria were at much higher risk of having mucopurulent or purulent vaginal
292
discharge. Similarly, Mateus et al observed that cows with mild or severe endometritis had a
AC C
287
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 293
higher prevalence and persistence of T. pyogenes [27]. Another study identified T. pyogenes as
294
the most common bacterial species isolated from uterine biopsies [9]. In a previous study, we demonstrated that most of the bovine intrauterine T. pyogenes
296
isolates we detected carried the gene FimA, which encodes a fimbrial subunit that mediates
297
bacterial adhesion to endothelial cells [29]. FimA is an important virulence factor associated with
298
increased risk of both metritis and endometritis [6,29]. Moreover, in a subsequent study, we
299
found that cows at 34–36 days postpartum that were infected with intrauterine T. pyogenes had
300
higher odds of developing clinical endometritis compared to TP-negative cows [6].
SC
RI PT
295
Despite the strong association of T. pyogenes and endometritis, a study by Galvao et al.
302
reported that only 41% of cows with PVD were T. pyogenes culture positive at the time of PVD
303
diagnosis [30]. Moreover, Sens and Heuwieser found alpha-hemolytic Streptococcus to be
304
strongly associated with reduced reproductive performance compared to T. pyogenes [31].
305
Bacterial pathogens and inflammatory mediators affect reproductive performance apparently
306
independently and additively and may interrupt an established pregnancy [32]. It is known that
307
endometrial cells are capable of directly recognizing PAMPs (pathogen-associated molecular
308
patterns) via Toll-like receptors [33] and secreting inflammatory cytokines to initiate an
309
inflammatory response [34]. T. pyogenes was found to initiate a cellular inflammatory response
310
in contrast to the exotoxin pyolisin [35].
TE D
EP
AC C
311
M AN U
301
The impact of endometritis on reproductive performance is well described in the
312
literature. Purulent vaginal discharge detected around 21 days post-parturition has been linked to
313
a lower pregnancy rate and a longer time to pregnancy [36,37], and cows with PVD required
314
more inseminations per pregnancy [38]. McDougall and coworkers reported that conception to
315
first-service is negatively associated with increasing endometritis score detected by vaginal
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis discharge [39]. Moreover, Studer and Morrow found a positive association between days open
317
and conception rate and increased purulent vaginal discharge score in cows after parturition [14].
318
In the present study, combined PVD and CE had an additional effect of further reducing
319
the probability of pregnancy. This is in an agreement with the results of other studies that found
320
that cows with both purulent vaginal discharge and subclinical endometritis had the lowest
321
pregnancy per artificial insemination, highest pregnancy loss, and delayed time to pregnancy
322
compared with cows diagnosed with only one of the two diseases, or compared with cows having
323
no diagnosis of uterine disease [40,41]. Only one study did not find any effect of increased
324
endometrial PMNL infiltration on reproductive performance [42]. Elevated PMNL count at
325
cytological examination was associated with an increased time to pregnancy [43], and subclinical
326
endometritis was directly associated with poor body condition and had a negative effect on
327
reproductive traits of dairy cows [44].
M AN U
SC
RI PT
316
Likewise, cows with cytological endometritis diagnosed by the uterine lavage technique
329
were less likely to become pregnant before 120 days postpartum than non-endometritic cows
330
[12], and they took longer to become pregnant [12,43,45]. Also, the proportion of cows that
331
required more than three services was higher in cows that had uterine inflammation [44]. In
332
addition, Machado et al., who used a visual scoring system to evaluate uterine lavage samples,
333
demonstrated that cows with purulent uterine lavage were less likely to conceive and had a
334
greater median calving-to-conception interval [16]. The low-volume uterine lavage technique
335
used in that study and in the present study assesses the presence of a mucopurulent secretion only
336
inside the uterus, thereby helping to avoid false-positive diagnoses that might be more common
337
with other methods of diagnosis such as rectal palpation and visual inspection of vaginal
338
discharge.
AC C
EP
TE D
328
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis In a recent comprehensive review of the diagnostic criteria and methods for diagnosing
340
endometritis [46], the authors stated that more reports on reproductive outcomes are necessary to
341
improve the comparative data among diagnostic criteria. In the present study, overall, TP, PUL,
342
PVD and CE had relatively high sensitivities (88.2% to 90.7%) but relatively low specificities
343
(13.6% to 22.1%) for predicting non-pregnancy at 300 days postpartum. Only a few papers have
344
evaluated the impaired reproductive outcomes and PVD diagnosis by Metricheck [37,39,40].
RI PT
339
The validity of PVD evaluation for diagnosis of endometritis is confirmed by the
346
substantial impact of the condition on reproductive parameters found in our study and by very
347
similar values for sensitivity, specificity, positive likelihood ratio, negative likelihood ratio,
348
positive predictive value, and negative predictive value for predicting non-pregnancy in
349
comparison to PUL and CE. Moreover, when the three evaluated methods were included in the
350
same statistics model, only PVD was significantly associated with reduced hazard of pregnancy.
M AN U
SC
345
The use of Metricheck to evaluate PVD is a less invasive, and the characteristic color
352
and odor of vaginal discharge as well as endometritis score are highly correlated with T.
353
pyogenes in the uterus [1,36,47,48].
TE D
351
Finally, the strong association of purulent content with reduced fertility described in the
355
present study is likely the combined detrimental effect of all conditions that lead to purulent
356
uterine/vaginal discharge (e.g. vaginitis, cervicitis, endometritis, and inadequate uterine
357
involution). Although factors that cause lesions in the genital tract (uterus, cervix , and vagina)
358
such as dystocia, twin parturition, and retained placenta were not evaluated in this study,
359
previous studies have reported those factors to be highly associated with PVD [6,8,49].
AC C
EP
354
360
In conclusion, TP-positive cows had impaired reproductive performance and increased
361
likelihood of PVD and PUL. Additionally, TP-positive cows tended to have increased prevalence
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 362
of CE. Cows affected with PVD had the lowest reproductive performance when compared to
363
PUL and CE, suggesting that PVD at 35 d postpartum is a better criterion for the diagnosis of
364
endometritis and reproductive failure.
RI PT
365 ACKNOWLEDGMENTS
367
This research was supported in part by the Cornell University Agricultural
368
Experiment Station federal formula funds, Project No NYC-480450 (to ROG) received
369
from Cooperative State Research, Education and Extension Service, U.S. Department
370
of Agriculture. Any opinions, findings, conclusions, or recommendations expressed in
371
this publication are those of the authors and do not necessarily reflect the view of the
372
U.S. Department of Agriculture.
AC C
EP
TE D
M AN U
SC
366
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 373 References
375 376 377
[1] LeBlanc SJ, Duffield TF, Leslie KE, Bateman KG, Keefe GP, Walton JS, et al. Defining and diagnosing postpartum clinical endometritis and its impact on reproductive performance in dairy cows. J. Dairy Sci. 2002;85(9):2223-36.
378 379 380
[2] Olson JD, Ball L, Mortimer RG, Farin PW, Adney WS, Huffman EM. Aspects of bacteriology and endocrinology of cows with pyometra and retained fetal membranes. Am. J. Vet. Res. 1984;45(11):2251-5.
381 382 383
[3] Williams EJ, Fischer DP, Pfeiffer DU, England GC, Noakes DE, Dobson H, et al. Clinical evaluation of postpartum vaginal mucus reflects uterine bacterial infection and the immune response in cattle. Theriogenology 2005;63(1):102-17.
384 385
[4] Sheldon IM, Lewis GS, LeBlanc S, Gilbert RO. Defining postpartum uterine disease in cattle. Theriogenology 2006;65(8):1516-30.
386 387 388
[5] Bicalho RC, Machado VS, Bicalho ML, Gilbert RO, Teixeira AG, Caixeta LS, et al. Molecular and epidemiological characterization of bovine intrauterine escherichia coli. J. Dairy Sci. 2010;93(12):5818-30.
389 390 391
[6] Bicalho ML, Machado VS, Oikonomou G, Gilbert RO, Bicalho RC. Association between virulence factors of escherichia coli, fusobacterium necrophorum, and arcanobacterium pyogenes and uterine diseases of dairy cows. Vet. Microbiol. 2012;157(1-2):125-31.
392 393 394
[7] Westermann S, Drillich M, Kaufmann TB, Madoz LV, Heuwieser W. A clinical approach to determine false positive findings of clinical endometritis by vaginoscopy by the use of uterine bacteriology and cytology in dairy cows. Theriogenology 2010;74(7):1248-55.
395 396
[8] Potter TJ, Guitian J, Fishwick J, Gordon PJ, Sheldon IM. Risk factors for clinical endometritis in postpartum dairy cattle. Theriogenology 2010;74(1):127-34.
397 398 399
[9] Bonnett BN, Martin SW, Gannon VP, Miller RB, Etherington WG. Endometrial biopsy in holstein-friesian dairy cows. III. bacteriological analysis and correlations with histological findings. Can. J. Vet. Res. 1991;55(2):168-73.
400 401
[10] Sheldon IM, Noakes DE, Rycroft AN, Dobson H. The effect of intrauterine administration of estradiol on postpartum uterine involution in cattle. Theriogenology 2003;59(5-6):1357-71.
402 403 404
[11] 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(1-2):9-23.
AC C
EP
TE D
M AN U
SC
RI PT
374
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis [12] 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(9):1879-88.
407 408 409
[13] de Boer M, Buddle BM, Heuer C, Hussein H, Zheng T, LeBlanc SJ, et al. Associations between intrauterine bacterial infection, reproductive tract inflammation, and reproductive performance in pasture-based dairy cows. Theriogenology 2015;83(9):1514-24.
410 411 412
[14] Studer E, Morrow DA. Postpartum evaluation of bovine reproductive potential: Comparison of findings from genital tract examination per rectum, uterine culture, and endometrial biopsy. J. Am. Vet. Med. Assoc. 1978;172(4):489-94.
413 414 415
[15] Cordero OX, Wildschutte H, Kirkup B, Proehl S, Ngo L, Hussain F, et al. Ecological populations of bacteria act as socially cohesive units of antibiotic production and resistance. Science 2012;337(6099):1228-31.
416 417 418
[16] Machado VS, Knauer WA, Bicalho ML, Oikonomou G, Gilbert RO, Bicalho RC. A novel diagnostic technique to determine uterine health of holstein cows at 35 days postpartum. J. Dairy Sci. 2012;95(3):1349-57.
419 420
[17] Barlund CS, Carruthers TD, Waldner CL, Palmer CW. A comparison of diagnostic techniques for postpartum endometritis in dairy cattle. Theriogenology 2008;69(6):714-23.
421 422
[18] Dubuc J, Duffield TF, Leslie KE, Walton JS, LeBlanc SJ. Definitions and diagnosis of postpartum endometritis in dairy cows. J. Dairy Sci. 2010;93(11):5225-33.
423 424 425
[19] Machado VS, Oikonomou G, Bicalho ML, Knauer WA, Gilbert R, Bicalho RC. Investigation of postpartum dairy cows' uterine microbial diversity using metagenomic pyrosequencing of the 16S rRNA gene. Vet. Microbiol. 2012;159(3-4):460-9.
426 427 428 429
[20] Machado VS, Bicalho ML, Meira Junior EB, Rossi R, Ribeiro BL, Lima S, et al. Subcutaneous immunization with inactivated bacterial components and purified protein of escherichia coli, fusobacterium necrophorum and trueperella pyogenes prevents puerperal metritis in holstein dairy cows. PLoS One 2014;9(3):e91734.
430 431 432
[21] Cheong SH, Nydam DV, Galvao KN, Crosier BM, Ricci A, Caixeta LS, et al. Use of reagent test strips for diagnosis of endometritis in dairy cows. Theriogenology 2012;77(5):85864.
433 434
[22] Institute of Medicine (US) Committee on Examination of the Evolving Science for Dietary Supplements. 2002.
435 436 437
[23] Moreira F, Orlandi C, Risco CA, Mattos R, Lopes F, Thatcher WW. Effects of presynchronization and bovine somatotropin on pregnancy rates to a timed artificial insemination protocol in lactating dairy cows. J. Dairy Sci. 2001;84(7):1646-59.
AC C
EP
TE D
M AN U
SC
RI PT
405 406
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis [24] Pursley JR, Mee MO, Wiltbank MC. Synchronization of ovulation in dairy cows using PGF2alpha and GnRH. Theriogenology 1995;44(7):915-23.
440 441 442
[25] Fricke PM, Caraviello DZ, Weigel KA, Welle ML. Fertility of dairy cows after resynchronization of ovulation at three intervals following first timed insemination. J. Dairy Sci. 2003;86(12):3941-50.
443 444
[26] Edmonson A.J., Lean I.J., Weaver L.D., Farver T., Webster G.1989. A body condition scoring chart for Holstein dairy cows. J. Dairy Sci., 72 (1989), pp. 68–78
445 446 447
[27] Mateus L, da Costa LL, Bernardo F, Silva JR. Influence of puerperal uterine infection on uterine involution and postpartum ovarian activity in dairy cows. Reprod. Domest. Anim. 2002; 37(1):31-5.
448 449 450
[28] Brick TA, Schuenemann GM, Bas S, Daniels JB, Pinto CR, Rings DM, et al. Effect of intrauterine dextrose or antibiotic therapy on reproductive performance of lactating dairy cows diagnosed with clinical endometritis. J. Dairy Sci. 2012;95(4):1894-905.
451 452 453
[29] Santos TM, Caixeta LS, Machado VS, Rauf AK, Gilbert RO, Bicalho RC. Antimicrobial resistance and presence of virulence factor genes in arcanobacterium pyogenes isolated from the uterus of postpartum dairy cows. Vet. Microbiol. 2010;145(1-2):84-9.
454 455
[30] Galvao KN, Greco LF, Vilela JM, Sa Filho MF, Santos JE. Effect of intrauterine infusion of ceftiofur on uterine health and fertility in dairy cows. J. Dairy Sci. 2009;92(4):1532-42.
456 457 458
[31] Sens A, Heuwieser W. Presence of escherichia coli, trueperella pyogenes, alpha-hemolytic streptococci, and coagulase-negative staphylococci and prevalence of subclinical endometritis. J. Dairy Sci. 2013;96(10):6347-54.
459 460
[32] Gilbert RO. The effects of endometritis on the establishment of pregnancy in cattle. Reprod. Fertil. Dev. 2011;24(1):252-7.
461 462 463
[33] Davies D, Meade KG, Herath S, Eckersall PD, Gonzalez D, White JO, et al. Toll-like receptor and antimicrobial peptide expression in the bovine endometrium. Reprod. Biol. Endocrinol. 2008;6:53,7827-6-53.
464 465 466
[34] Wira CR, Grant-Tschudy KS, Crane-Godreau MA. Epithelial cells in the female reproductive tract: A central role as sentinels of immune protection. Am. J. Reprod. Immunol. 2005;53(2):65-76.
467 468 469
[35] Amos MR, Healey GD, Goldstone RJ, Mahan SM, Duvel A, Schuberth HJ, et al. Differential endometrial cell sensitivity to a cholesterol-dependent cytolysin links trueperella pyogenes to uterine disease in cattle. Biol. Reprod. 2014;90(3):54.
AC C
EP
TE D
M AN U
SC
RI PT
438 439
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis [36] LeBlanc SJ, Duffield TF, Leslie KE, Bateman KG, Keefe GP, Walton JS, et al. The effect of treatment of clinical endometritis on reproductive performance in dairy cows. J. Dairy Sci. 2002;85(9):2237-49.
473 474 475
[37] 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(9):369-78.
476 477 478
[38] Gautam G, Nakao T, Yusuf M, Koike K. Prevalence of endometritis during the postpartum period and its impact on subsequent reproductive performance in two japanese dairy herds. Anim. Reprod. Sci. 2009;116(3-4):175-87.
479 480 481
[39] 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(1-2):9-23.
482 483 484
[40] Lima FS, Bisinotto RS, Ribeiro ES, Greco LF, Ayres H, Favoreto MG, et al. Effects of 1 or 2 treatments with prostaglandin F(2)alpha on subclinical endometritis and fertility in lactating dairy cows inseminated by timed artificial insemination. J. Dairy Sci. 2013;96(10):6480-8.
485 486 487
[41] Dubuc J, Duffield TF, Leslie KE, Walton JS, Leblanc SJ. Randomized clinical trial of antibiotic and prostaglandin treatments for uterine health and reproductive performance in dairy cows. J. Dairy Sci. 2011;94(3):1325-38.
488 489 490
[42] Plontzke J, Madoz LV, De la Sota RL, Drillich M, Heuwieser W. Subclinical endometritis and its impact on reproductive performance in grazing dairy cattle in argentina. Anim. Reprod. Sci. 2010;122(1-2):52-7.
491 492 493
[43] Galvao KN, Frajblat M, Brittin SB, Butler WR, Guard CL, Gilbert RO. Effect of prostaglandin F2alpha on subclinical endometritis and fertility in dairy cows. J. Dairy Sci. 2009;92(10):4906-13.
494 495 496
[44] Bacha B, Regassa FG. Subclinical endometritis in zebu x friesian crossbred dairy cows: Its risk factors, association with subclinical mastitis and effect on reproductive performance. Trop. Anim. Health Prod. 2010;42(3):397-403.
497 498 499
[45] Cheong SH, Nydam DV, Galvao KN, Crosier BM, Gilbert RO. Cow-level and herd-level risk factors for subclinical endometritis in lactating holstein cows. J. Dairy Sci. 2011;94(2):76270.
500 501 502
[46] de Boer MW, LeBlanc SJ, Dubuc J, Meier S, Heuwieser W, Arlt S, et al. Invited review: Systematic review of diagnostic tests for reproductive-tract infection and inflammation in dairy cows. J. Dairy Sci. 2014;97(7):3983-99.
503 504
[47] Sheldon IM, Dobson H. Postpartum uterine health in cattle. Anim. Reprod. Sci. 2004;8283:295-306.
AC C
EP
TE D
M AN U
SC
RI PT
470 471 472
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis 505 506
[48] Djonov V, Andres AC, Ziemiecki A. Vascular remodelling during the normal and malignant life cycle of the mammary gland. Microsc. Res. Tech. 2001;52(2):182-9.
507 508
[49] Dubuc J, Duffield TF, Leslie KE, Walton JS, LeBlanc SJ. Risk factors and effects of postpartum anovulation in dairy cows. J. Dairy Sci. 2012;95(4):1845-54.
RI PT
509
AC C
EP
TE D
M AN U
SC
510
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis Table 1: Prevalence of endometritis diagnosed by purulent uterine lavage (score>2), purulent
512
vaginal discharge (score>3), and cytology (PMNL ≥ 5%) at 35 ± 3 d postpartum in cows that
513
were negative or positive for the presence of Trueperella pyogenes.
514 Trueperella pyogenes Endometritis prevalence
Total n=452
Negative n=403
Positive n= 49
Purulent uterine lavage (PUL)
13.8%
8.9%
55.1%
Purulent vaginal discharge (PVD) 14.0%
8.4%
61.2%
< 0.001
Cytological endometritis (CE)
12.7%
11.9%
20.4%
0.09
PUL&PVD
8.1%
3.5%
47%
< 0.001
PUL&CE
3.5%
2.0%
16.3%
< 0.001
PVD&CE
4.4%
2.7%
18.4%
< 0.001
PUL&PVD&CE
2.9%
1.2%
16.3 %
< 0.001
P-value < 0.001
SC
M AN U
AC C
EP
TE D
515 516
RI PT
511
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis Table 2: Final Cox proportional hazard model of the association of purulent uterine lavage,
518
purulent vaginal discharge and cytological endometritis (considered together) with the hazard
519
ratio (HR) of pregnancy by 300 d after parturition in Holstein cows examined at 35 ± 3 d (n =
520
452) after calving. Coefficient
SE
HR
Purulent uterine lavage
-0.15
0.19
0.86
Purulent vaginal discharge
-0.53
0.20
0.59
Cytological endometritis
-0.15
0.18
0.86
521
AC C
EP
TE D
M AN U
522
95% CI
P-value
0.60-1.25
0.43
0.40-0.86
<0.01
0.61-1.22
0.41
SC
Variable
RI PT
517
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis Table 3: Sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR),
524
positive predictive value (PPV), negative predictive value (NPV) for predicting non-pregnancy
525
status 300 d after parturition for presence of Trueperella pyogenes (TP), purulent uterine lavage
526
(PUL), purulent vaginal discharge (PVD), and cytological endometritis (CE) at a single
527
examination performed on Holstein cows 35 ± 3 d after parturition.
Sensitivity, %
90.7 (87.7-93.8)
88.2(84.4-91.4)
Specificity, %
13.6 (6.9-20.2)
19.2 (12.2-28.1)
PLR
1.1 (0.9-1.4)
NLR
0.7 (0.4-1.2)
532 533
22.1 (14.6-31.3)
17.3 (10.6-26.0)
1.1 (1.0-1.2)
1.1 (1.0-1.3)
1.1 (1.0-1.2)
0.6 (0.4-1.0)
0.5 (0.3-0.8)
0.7 (0.4-1.1)
79.2 (74.9-83.1)
78.2 (73.8-82.2)
NPV, %
22.1 (18.0-26.5)
32.8 (21.2-46.1)
37.1 (25.1-50.4)
31.6 (19.8-45.4)
TE D
77.0%
EP
531
88.8 (85.0-91.9)
78.5 (74.1-82.5)
AC C
530
88.8 (85.0-91.9)
69.6 (54.2-82.5)
Number of cows
529
CE
PPV, %
Percent pregnant 528
PVD
SC
PUL
M AN U
TP
RI PT
523
452
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis
TE D
EP
545
Fig.1. Kaplan-Meier survival curves for proportion of non-pregnant cows at 300 days postpartum according to the presence of purulent uterine lavage (PUL) at 35 ± 3 d postpartum. Median time to pregnancy for cows with PUL and without PUL was 168 d (95% confidence interval [CI] = 134 to 177) and 124 d (95% CI = 114 to 134), respectively. The hazard of pregnancy at 300 d postpartum was lower (P < 0.01) for cows with PUL than for cows without PUL (hazard ratio = 0.65; 95% CI = 0.49 to 0.86).
AC C
535 536 537 538 539 540 541 542 543 544
M AN U
SC
RI PT
534
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis
EP
TE D
Fig.2. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the presence of purulent vaginal discharge (PVD) at 35 ± 3 d postpartum. Median time to pregnancy for cows with PVD and without PVD was 175 d (95% confidence interval [CI] = 154 to 220) and 118 d (95% CI = 112 to 131), respectively. The hazard of pregnancy at 300 d postpartum was lower (P < 0.001) for cows with PVD than for cows without PVD (hazard ratio = 0.53; 95% CI = 0.41 to 0.70).
AC C
547 548 549 550 551 552 553 554 555
M AN U
SC
RI PT
546
ACCEPTED MANUSCRIPT
EP
TE D
Fig.3. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the presence of cytological endometritis (CE) at 35 ± 3 d postpartum. Median time to pregnancy for cows with CE and without CE was 168 d (95% confidence interval [CI] = 114 to 209) and 126 d (95% CI = 115 to 135), respectively. The hazard of pregnancy at 300 d postpartum was lower (P = 0.05) for cows with CE than for cows without CE (hazard ratio = 0.73; 95% CI = 0.54 to 0.98).
AC C
556 557 558 559 560 561 562 563 564 565
M AN U
SC
RI PT
Therio-14-Bicalho-Endometritis Diagnosis
ACCEPTED MANUSCRIPT
EP
TE D
Fig.4. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to combined presence of purulent vaginal discharge (PVD) and purulent uterine lavage (PUL) at 35 ± 3 d postpartum. Median time to pregnancy for cows classified as healthy, only PVD, only PUL, and combined PVD and PUL were: 118 d (95% confidence interval [CI] = 111 to 131), 211 d (95% CI = 134 to 223), 121 d (95% CI = 95 to 218), 174 d (95% CI = 154 to 217), respectively. The hazard of pregnancy at 300 d postpartum did not differ between combined PVD + PUL and only PVD (hazard ratio = 1.19; 95% CI = 0.73 to 1.93) or only PUL (hazard ratio = 0.78; 95% CI = 0.45 to 1.33).
AC C
566 567 568 569 570 571 572 573 574 575 576
M AN U
SC
RI PT
Therio-14-Bicalho-Endometritis Diagnosis
ACCEPTED MANUSCRIPT
TE D
EP
588
Fig.5. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the combined presence of purulent uterine lavage (PUL) and cytological endometritis (CE) at 35 ± 3 d postpartum. Median time to pregnancy for cows classified as healthy, only PUL, only CE, and combined PUL and CE were: 119 d (95% confidence interval [CI] = 113 to 134), 139 d (95% CI = 132 to 197), 126d (95% CI = 95 to 209), 177 d (95% CI = 135 to 177), respectively. The hazard of pregnancy at 300 d postpartum did not differ between combined PUL + CE and only PUL (hazard ratio = 0.67; 95% CI = 0.38 to 1.17) or only CE (hazard ratio = 0.58; 95% CI = 0.32 to 1.03).
AC C
577 578 579 580 581 582 583 584 585 586 587
M AN U
SC
RI PT
Therio-14-Bicalho-Endometritis Diagnosis
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis
603 604 605
EP
TE D
Fig.6. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the combined presence of purulent vaginal discharge (PVD) and cytological endometritis (CE) at 35 ± 3 d postpartum. Median time to pregnancy for cows classified as healthy, only PVD, only CE, and combined PVD and CE were: 118 d (95% confidence interval [CI] = 112 to 131), 172 d (95% CI = 136 to 220), 118 d (95% CI = 93 to 190), 207 d (95% CI = 168 to 223), respectively. The hazard of pregnancy at 300 d postpartum did not differ between combined PVD + CE and only PVD (hazard ratio = 0.67; 95% CI = 0.38 to 1.17), but combined PVD and CE had a lower hazard of pregnancy than only CE (hazard ratio = 0.52; 95% CI = 0.30 to 0.90).
AC C
591 592 593 594 595 596 597 598 599 600 601 602
M AN U
SC
RI PT
589 590
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Therio-14-Bicalho-Endometritis Diagnosis
606
TE D
EP
614
Fig.7. Kaplan-Meier survival curves for the proportion of non-pregnant cows at 300 days postpartum according to the presence of Trueperella pyogenes (TP) at 35 ± 3 d postpartum. Median time to pregnancy for TP positive and TP negative was 175 d (95% confidence interval [CI] = 136 to 211) and 119 d (95% CI = 114 to 134), respectively. The hazard of pregnancy at 300 d postpartum was lower (P = 0.02) for TP-positive than TP-negative cows (hazard ratio = 0.66; 95% CI = 0.48 to 0.90).
AC C
607 608 609 610 611 612 613
ACCEPTED MANUSCRIPT Therio-14-Bicalho-Endometritis Diagnosis
AC C
EP
TE D
M AN U
SC
RI PT
615