The association of plasma glucose, BHBA, and NEFA with postpartum uterine diseases, fertility, and milk production of Holstein dairy cows

Accepted Manuscript The association of plasma glucose, BHBA, and NEFA with postpartum uterine diseases, fertility, and milk production of Holstein dairy cows M.L.S. Bicalho, E.C. Marques, R.O. Gilbert, R.C. Bicalho PII:

S0093-691X(16)30455-1

DOI:

10.1016/j.theriogenology.2016.09.036

Reference:

THE 13835

To appear in:

Theriogenology

Received Date: 10 June 2016 Revised Date:

7 September 2016

Accepted Date: 21 September 2016

Please cite this article as: Bicalho MLS, Marques EC, Gilbert RO, Bicalho RC, The association of plasma glucose, BHBA, and NEFA with postpartum uterine diseases, fertility, and milk production of Holstein dairy cows, Theriogenology (2016), doi: 10.1016/j.theriogenology.2016.09.036. 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

revised 1

The association of plasma glucose, BHBA, and NEFA with postpartum uterine diseases,

2

fertility, and milk production of Holstein dairy cows

4

RI PT

3

Bicalhoa M. L. S., E. C. Marquesb, R.O. Gilberta and R.C. Bicalhob,1

5 6

a

7

NY 14853.

9

SC

b

Department of Population Medicine and Diagnostic Sciences, Cornell University, Ithaca, NY

M AN U

8

Department of Clinical Sciences, College of Veterinary Medicine, Cornell University, Ithaca,

14853-6401.

10 11 12

TE D

13 14 15

17 18 19 20 21

1

AC C

EP

16

Corresponding author: Rodrigo Carvalho Bicalho, Department of Population Medicine and Diagnostic Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 148536401, Phone: 607 253-3140, Fax: 607 253-3982, e-mail: [email protected]

1

ACCEPTED MANUSCRIPT

revised 22

ABSTRACT The objective of this study was to investigate the association between the metabolic

24

indicators non-esterified fatty acids (NEFA), β-hydroxybutyrate (BHBA), and glucose during

25

the transition period and the development of uterine diseases. In total, 181 Holstein dairy cows

26

were enrolled in the study. Plasma glucose, NEFA, and BHBA concentrations were measured at

27

-50, -6, 3, 7 and 14 d relative to parturition. All cows enrolled in the study were evaluated for

28

retained placenta (RP), metritis, and endometritis. RP and metritis were diagnosed and treated by

29

trained farm personnel. Clinical endometritis was evaluated by a veterinarian at 35 DIM using a

30

Metricheck device. We found plasma glucose concentration to be associated with the occurrence

31

of metritis and clinical endometritis. Moreover, cows with an increased calving to conception

32

interval (>150 d) presented higher plasma glucose concentrations than cows that became

33

pregnant within the first 150 d. BHBA and NEFA were not associated with the occurrence of any

34

uterine disorder. Receiver operating characteristic (ROC) curves were used in an attempt to

35

determine the cow-level critical thresholds for the occurrence of metritis, and endometritis.

36

Additionally, pairwise comparisons of area under the curve (AUC) of ROC curves for the critical

37

thresholds for glucose, BHBA, and NEFA predicting the same uterine disease were performed.

38

Glucose at 3 DIM was the best predictor for metritis and endometritis diagnosis, with AUC

39

values of 0.66 and 0.67, respectively. Multivariable logistic regressions were performed and

40

showed that cows with higher levels of glucose at day 3 were at 6.6 times higher odds of being

41

diagnosed with metritis, and 3.5 times higher odds of developing clinical endometritis, compared

42

to cows with lower glucose levels. Finally, a simple linear regression analysis demonstrated a

43

negative correlation between daily milk yield in the first and second weeks of lactation and

AC C

EP

TE D

M AN U

SC

RI PT

23

2

ACCEPTED MANUSCRIPT

revised 44

plasma glucose concentrations measured at days 7 and 14, respectively. NEFA and BHBA

45

concentrations were not found to be associated with milk production.

47

EP

TE D

M AN U

SC

Key words: Glucose, NEFA, BHBA, metritis

AC C

48

RI PT

46

3

ACCEPTED MANUSCRIPT

revised 49

1. INTRODUCTION Postpartum uterine diseases such as metritis, endometritis, purulent vaginal discharge and

51

retained placenta (RP) are associated with substantially infertility, reduced milk yield, and

52

increased culling rates [1]. These diseases have complex multifactorial causes which include

53

exposure to bacterial pathogens [2], mineral and vitamin deficiencies[3], negative energy balance

54

[4], and immunosuppression[5-7]. Previous studies have extensively demonstrated associations

55

between negative energy balance markers, particularly non-esterified fatty acids (NEFA) and β-

56

hydroxybutyric acid (BHBA) in plasma, and the incidence of postpartum diseases such as

57

clinical ketosis and displaced abomasum[8, 9] . Excessive fat mobilization in dairy cows has also

58

been associated with clinical endometritis and linked to immunosuppression [10-12] . The

59

inflammation of the endometrium has been showed to have a detrimental effect on reproductive

60

performance, reducing both the first service conception rate and the overall pregnancy risk

61

mainly due to dysregulated ovarian function and oocyte development [13-16].

TE D

M AN U

SC

RI PT

50

The liver is responsible for metabolizing circulating NEFA, which can be completely

63

oxidized for ATP production, exported from the liver as lipoproteins, or partially oxidized into

64

BHBA and other ketone bodies [17, 18]. When the liver is overload with NEFA, hepatocytes

65

increase the level of partial oxidation of NEFA, which leads to the accumulation of BHBA and

66

other ketone bodies, eventually causing sub-clinical and clinical ketosis [19]. Therefore,

67

hyperketonemia is a marker of liver health and has been associated with infectious diseases (e.g.

68

mastitis, metritis) [3, 20].

AC C

EP

62

69

In addition to fat mobilization in early post-partum cows, liver gluconeogenesis increases

70

to provide glucose for synthesis of milk lactose [21]. The large demand for glucose may lower

71

the amount of glucose available to other tissues in the body, including those that are involved in

4

ACCEPTED MANUSCRIPT

revised postpartum immune defense.

During the period of negative energy balance, dairy cows

73

experience a reduction in blood glucose levels and neutrophil function. Low glucose levels

74

observed during the transition to lactation may be associated with immunosuppression.

75

Granulocytes depend on uptake of exogenous glucose and intracellular glycogen stores for the

76

energy required for chemotaxis, phagocytosis, and microbial killing [22, 23]. Conversely, many

77

studies have demonstrated impaired adherence and neutrophil dysfunction associated with high

78

glucose levels [24-26].

SC

RI PT

72

A strong association between decreased immune response and a greater degree of

80

negative energy has been reported in cows that developed uterine disease compared with healthy

81

cows [10, 12, 27, 28]. For instance, cows that developed uterine disease experienced a greater

82

degree of negative energy balance, increased serum levels of inflammatory markers, greater

83

blood glucose concentration at calving and had lower intracellular neutrophil glycogen levels

84

[20, 28]. Moreover, cows with metritis and cows with cystic ovaries had increased levels of

85

ketone bodies than unaffected cows[29].

TE D

M AN U

79

Besides, relationships between blood metabolites and the reproductive performance of

87

dairy cows have been reported earlier [30, 31]. For instance, in a multivariate description of

88

factors that influence fertility in dairy cows, Westwood et al. (2002) showed that increased

89

concentrations of plasma glucose was associated with greater probability of estrous expression at

90

first ovulation, whereas higher serum concentrations of NEFA lowered the probability of

91

conception by d 150 [31], supporting previous observations where more mobilization of body

92

tissue delayed resumption of ovarian activity [32], However, in another study days-to-conception

93

was not associated with glucose levels but was inversely related to milk production[33] .

AC C

EP

86

5

ACCEPTED MANUSCRIPT

revised Thus far, most current research has focused on the evaluation of the markers of negative

95

energy balance, NEFA and BHBA as predictors of inflammation and postpartum diseases [8, 34,

96

35]; while few studies have evaluated the importance of glucose as a potentially significant risk

97

factor for the development of uterine diseases. Therefore, the objective of this study was to

98

investigate associations between the metabolic indicators NEFA, BHBA, and glucose during the

99

transition period and the occurrence of uterine diseases and the subsequent effect on fertility.

100

Because uterine diseases have negative effect on milk production, and milk production is

101

accompanied by changes in glucose and energy metabolism, milk yield was also evaluated.

M AN U

SC

RI PT

94

102

AC C

EP

TE D

103

6

ACCEPTED MANUSCRIPT

revised 104 105

2. MATERIALS AND METHODS 2.1 Farm, management, and sample collection This study was conducted from October 2012 until January 2013 on a dairy farm located

107

near Ithaca, New York. In total, 181 Holstein dairy cows (108 dry cows and 73 pregnant heifers)

108

were enrolled in the study. The farm milked 3,300 Holstein cows 3 times daily in a double 52-

109

stall parallel milking parlor. The cows were housed in freestall barns, with concrete stalls

110

covered with mattresses and bedded with composted manure solids. All cows were offered a

111

TMR consisting of approximately 55% forage (corn silage, haylage, and wheat straw) and 45%

112

concentrate (corn meal, soybean meal, canola, cottonseed, and citrus pulp) on a dry matter basis

113

of the diet. The diet was formulated to meet or exceed the NRC nutrient requirements for

114

lactating Holstein cows weighing 650 kg and producing 45 kg of 3.5% fat corrected milk (NRC

115

2001). The farm reproductive management used a combination of Presynch, Resynch, and

116

detection of estrus, with 25% to 30% of cows bred via timed AI and the remainder bred after

117

detection of estrus solely by activity monitors (Alpro; DeLaval, Kansas City, MO). Pregnancy

118

was diagnosed by rectal palpation at 39 ± 3d since the last insemination. All study cows were

119

followed until 300 days postpartum or the date of culling (if <300 days) from the herd. Data

120

regarding health traits, reproductive performance (the cow being diagnosed pregnant within 150

121

d postpartum) and milk yield during the subsequent lactation were obtained from DairyComp

122

(DairyComp 305 Tulare, CA) records for the herd, and descriptive statistics were calculated

123

(version 9.3, SAS Institute Inc., Cary, NC).

AC C

EP

TE D

M AN U

SC

RI PT

106

124

Blood was sampled from all study subjects at -50 (±3), -6 (±3), 3 (±3), 7 (±3), and 14

125

(±3) d relative to parturition. Blood collection was performed via the coccygeal vein/artery using

126

a 10-ml vacuum tube with lithium heparin, and a 20-gauge × 2.54 cm needle (Becton, Dickinson

7

ACCEPTED MANUSCRIPT

revised and Company, Franklin Lakes, NJ). After collection, all blood samples were transported to the

128

laboratory on ice, and plasma was harvested after centrifugation at 2,000 × g for 15 min at 4oC.

129

Plasma was frozen at -80oC. Body condition scores (BCS) were determined for all study cows at

130

blood collection time by a single investigator using a five-point scale with a quarter-point system

131

as previously described[36].

132

2.2 Case definitions

RI PT

127

RP and metritis were diagnosed and treated by trained farm personnel according to

134

specific protocols designed by the Ambulatory and Production Medicine Clinic at Cornell

135

University. RP was defined as cows that failed to release their fetal membranes within 24 h of

136

calving. Metritis was defined as the presence of fetid, watery, red-brown uterine discharge and a

137

rectal temperature greater than 39.5oC. Clinical endometritis was defined by the presence of

138

purulent or worse vaginal discharge, after retrieving vaginal mucus using the Metricheck device

139

(Metricheck, SimcroTech, Hamilton, New Zealand) as described in a previous study [37].

140

Diagnosis of clinical ketosis was attributed to cow that was off feed, had sudden weight loss and

141

decreased milk production, but had no other detectable signs of disease and it was tested positive

142

for ketones bodies in the urine using a reagent strip for urinalysis (Ketostix,Bayer, Pittsburgh,

143

PA). Ketotic cows were treated orally with 300ml of propylene glycol per day for 3-4 days.

144

2.3 Plasma glucose, BHBA, and NEFA

M AN U

TE D

EP

AC C

145

SC

133

Plasma glucose (PGO enzyme preparation, Sigma Aldrich, St. Louis, MO) and NEFA

146

(NEFA-C® kit; Wako Pure Chemical Industries, Osaka, Japan) concentrations were measured by

147

enzymatic colorimetric assays. Intra and interassay coefficients of variation were 2.0% and 2.9%

148

and 5.2% and 8.3% for NEFA and glucose respectively. Plasma samples were tested for BHBA

8

ACCEPTED MANUSCRIPT

revised concentrations using an electronic BHBA measuring system (Precision Xtra, Abbott, Abingdon,

150

UK) previously validated for animal use [38].

151

2.4 Statistical analysis

152 153

RI PT

149

Descriptive statistical analysis was completed in SAS using the FREQ and UNIVARIATE procedures (SAS Institute INC., Cary, NC).

Three groups of mixed general linear models were fitted to the data using the MIXED

155

procedure of SAS (SAS Inst., Inc., Cary, NC) to evaluate the effects of the independent

156

variables, RP, metritis and clinical endometritis, on the plasma concentrations of glucose,

157

BHBA, and NEFA. Each outcome was analyzed in separate models over the whole experiment

158

period. Additionally, a fourth mixed general linear model was fitted using the MIXED procedure

159

of SAS to evaluate the effect on pregnancy within 150 d on glucose plasma concentrations. To

160

facilitate data analysis and interpretation, a specific variable was created to indicate pregnancy

161

status within 150 d after parturition (Preg ≤150d); also, cows were categorized according to their

162

BCS at enrollment as nBCS1 (BCS ≤ 2.75), nBCS2 (3.00 ≤ BCS ≤ 3.25), and nBCS3 (BCS ≥

163

3.50). Potential confounders evaluated were, nBCS and parity (1, ≥2). The data were

164

longitudinally collected and comprised a series of repeated measures of the dependent variable

165

throughout the time points of each blood parameter measured. The within-cow correlation was

166

accounted for by imposing a first-order autoregressive covariance structure (which assumed that

167

the within-cow correlation of the repeated measures of the dependent variables decreased as the

168

time between the test dates increased) to the error term. Residuals were tested for normality by

169

visual evaluation of the distribution plot, and homogeneity of variances was determined by

170

comparing graphs (Levene’s test).

AC C

EP

TE D

M AN U

SC

154

9

ACCEPTED MANUSCRIPT

revised Results are presented as least squares means followed by the respective standard error of

172

the mean. Effects were considered significant at P ≤ 0.05, and a tendency was considered when

173

0.05 < P ≤ 0.10. Interactions were kept in the model when P ≤ 0.15. Graphs were built with

174

GraphPad Prism 5 software (GraphPad Software Inc., La Jolla, CA).

RI PT

171

Receiver Operator Characteristic (ROC) curves were obtained using MedCalc version

176

9.5.2.0 (Schoonjans, Mariakerke, Belgium). The continuous variables glucose, NEFA, and

177

BHBA plasma concentration were evaluated with ROC analysis to determine a critical threshold

178

for predicting disease at each study time point.

M AN U

SC

175

The higher area under the curve (AUC) values associated with each metabolite and

180

disease were used to determine the most predictive critical threshold for disease identification.

181

Sensitivity was defined as the proportion of animals diagnosed with disease that were above a

182

given metabolite threshold, and specificity was the proportion of animals that did not have

183

disease below a given threshold[39]. The point on the ROC curve that had the highest combined

184

sensitivity and specificity was considered the critical threshold. Interpretation of this critical

185

threshold was based on the AUC, as previously described [40]. Likelihood ratios (LR) were also

186

calculated.

EP

TE D

179

Glucose concentration at Day 3 was found to be the most predictive threshold based on

188

the diagnostic likelihood ratios and areas under the curves of the ROC analysis of all curves.

189

Therefore, to evaluate the effect of glucose levels on the probability of metritis and clinical

190

endometritis, logistic regressions were fitted to the data using JMP®PRO 10. To control for

191

repeated measures, the animal identification number was included in the model as a random

192

effect; and the models included the fixed effects of parity (1, ≥2) and glucose concentration at

193

Day 3 relative to parturition.

AC C

187

10

ACCEPTED MANUSCRIPT

revised Finally, simple linear regression analyses were used to identify the correlation between

195

plasma glucose concentrations at day 7 and day 14 relative to parturition and daily milk yield.

196

Scatter diagrams were prepared using MedCalc version 9.5.2.0 (Schoonjans, Mariakerke,

197

Belgium).

198 199

RI PT

194

3. RESULTS 3.1 Descriptive analysis

In total, 181 cows were studied, of which 73 were primiparous (40.3%) and 108 (59.6%)

201

were of second parity and greater (multiparous). The overall incidences of clinical ketosis, RP,

202

metritis, and clinical endometritis in the study population were 7.1%, 9.4%, 15%, and 22.3%,

203

respectively. The percentage of pregnant cows within 150 d was 51.1% (n=92).

M AN U

SC

200

At enrollment the average days of gestation was 229.7 d (SE = ± 2.19). The median BCS

205

at enrollment for both primiparous and multiparous cows was 3.125 (range 2.25 to 4). The

206

postpartum median BCS for primiparous cows was 3.25, 3.25, and 3.125 at day 3, 7, and 14

207

relative to parturition, respectively, and for multiparous cows the BCS was 3.25, 3.25, and 3.0

208

for the same post-parturition days as above.

209

3.2 Association between plasma glucose and uterine diseases

EP

TE D

204

Cows diagnosed with RP had greater (P ≤ 0.01) concentrations of glucose in plasma

211

compared with healthy herdmates (Table 1). Primiparous cows that presented RP had higher

212

glucose concentrations at days -6, 3, 7 and 14 relative to parturition (P ≤ 0.05) when compared to

213

primiparous cows that did not present RP(Figure 1). Multiparous cows that presented RP did

214

significantly differ in glucose concentration (P ≤ 0.05) on days -50 and 3 relative to parturition

215

from multiparous cows that did not present RP; and tended to have higher glucose concentrations

216

at day 7 and 14 (P = 0.06 and P = 0.08, respectively; Figure 1).

AC C

210

11

ACCEPTED MANUSCRIPT

revised Overall glucose concentrations for cows diagnosed with metritis were significantly higher

218

than for cows without metritis (P ≤ 0.01, Table 1). Primiparous cows affected with metritis

219

exhibited higher plasma glucose concentrations on day -6 and day 7 (P ≤ 0.05; Figure 2) than

220

their herdmates not diagnosed with metritis; and had a tendency to exhibit higher glucose

221

concentrations at day 14 (P = 0.06). Whereas, multiparous metritic cows had significantly higher

222

glucose concentrations on days 3 and 14(P ≤ 0.05), and had a tendency to exhibit higher glucose

223

concentrations at day 14 in comparison to multiparous cows without metritis (P = 0.06; Figure

224

2).

M AN U

SC

RI PT

217

Plasma glucose was higher for cows with clinical endometritis when compared to healthy

226

cows (P ≤ 0.01, Table 1). Although glucose concentration was not affected by parity (P = 0.40),

227

an interaction between parity, disease and time on glucose concentration was evident (P = 0.02).

228

Multiparous cows that did develop clinical endometritis had significantly higher glucose

229

concentrations on day 3, 7, and 14 when compared to healthy multiparous cows (P ≤ 0.01;

230

Figure 3).

231

3.3 Association between plasma BHBA and uterine diseases

TE D

225

Plasma concentration of BHBA did not differ between cows with RP compared to cows

233

without RP (P = 0.48, Table 2). BHBA concentration, however, tended to be lower for

234

primiparous cows than for multiparous cows (P = 0.09). The increase in BHBA concentration at

235

day 3 relative to parturition for primiparous cows with RP in comparison to healthy primiparous

236

cows was not significant (P = 0.24; Figure 1). Additionally, multiparous cows with RP tended to

237

exhibit a lower BHBA concentration at day 14 relative to parturition (P ≤ 0.10, Figure 1).

238

Overall plasma BHBA was not associated with metritis (P = 0.36) or clinical endometritis (P =

239

0.80, Table 2). BHBA concentrations did not significantly differ between primiparous and

AC C

EP

232

12

ACCEPTED MANUSCRIPT

revised multiparous cows with metritis and healthy cows (Figure 2). Neither variation was seen on

241

serum concentrations of BHBA for primiparous and multiparous cows with or without

242

endometritis.

243

3.4 Association between plasma NEFA and uterine diseases

RI PT

240

Plasma NEFA concentration did not differ between cows with RP compared to cows

245

without RP (P = 0.48, Table 3). Primiparous cows that developed metritis had a tendency to have

246

a higher NEFA concentration at day 7 relative to parturition in comparison to healthy cows (P =

247

0.08, Figure 2). Healthy multiparous cows presented higher NEFA concentrations at days 3 and

248

7 relative to parturition (P ≤ 0.05) and tended to maintain a higher NEFA concentration at day 14

249

relative to parturition (P ≤ 0.10; Figure 2).

250

3.5 Association between plasma glucose and reproduction

M AN U

SC

244

Cows with an increased calving to conception interval (not pregnant within 150 d)

252

presented a higher plasma glucose concentration than cows that became pregnant within 150 d (P

253

≤ 0.05, Table 4). Glucose concentrations at day 3 and day 7 were significantly higher for

254

primiparous cows that were not diagnosed pregnant within 150 d (P ≤ 0.05; Figure 4).

255

3.6 Clinical thresholds

EP

TE D

251

ROC analysis was used to determine cow-level critical thresholds for glucose, BHBA,

257

and NEFA (combined highest sensitivity and specificity) that would better predict RP, metritis

258

and clinical endometritis presence. The results of ROC curve determination of critical thresholds,

259

sensitivity, specificity, AUC, and LR are presented in Table 5.

AC C

256

260

Likelihood ratios were calculated based on critical thresholds determined by univariable

261

ROC analysis. For instance, LR positive (LR+) indicates the likelihood that a positive test result

262

(value at or above the threshold) would come from an animal diagnosed with disease in

13

ACCEPTED MANUSCRIPT

revised comparison to a healthy animal. Additionally, pairwise comparisons of ROC curves for the

264

critical threshold for glucose, BHBA, and NEFA predicting the same uterine disease were

265

performed (Figures 5 and 6). For predicting metritis, the ROC curve for glucose also presented a

266

higher AUC (0.66) compared to those of BHBA (AUC = 0.58) and NEFA (AUC = 0.60).

267

Similarly, for predicting clinical endometritis, the ROC curve for glucose had a higher AUC

268

(0.64) compared to those of BHBA (AUC = 0.52) and NEFA (AUC = 0.57).

RI PT

263

Glucose at day 3 relative to parturition was the best predictor for metritis and clinical

270

endometritis diagnosis, with an AUC of 0.66 and 0.67, respectively, when compared to the other

271

two metabolites evaluated in this study (Table 5). Cows with glucose ≥56 mg/dL at day 3

272

relative to parturition were 6.6 times more likely to have metritis compared to cows with glucose

273

< 56 mg/dL. Moreover, cows at day 3 with glucose levels ≥ 45 mg/dL were at a 3.5 higher odds

274

of developing clinical endometritis compared to cows with glucose levels <45 mg/dL.

275

3.7 Association between plasma glucose and early lactation milk production

TE D

M AN U

SC

269

Simple linear regression showed that plasma glucose was negatively associated with milk

277

yield. A negative correlation between the weekly average for daily milk yield for all cows during

278

the first week of lactation and plasma glucose concentration measured at day 7 relative to

279

parturition is depicted in Figure 7A (R2 = 0.31, P ≤ 0.001). Figure 7B displays a negative

280

correlation between the weekly average for daily milk yield for all study cows during the second

281

week of lactation and plasma glucose concentration at day 14 relative to parturition (R2 = 0.27, P

282

<0.001). However, plasma glucose concentration is associated with only 30% of the variation in

283

milk production.

284

AC C

EP

276

4. DISCUSSION

14

ACCEPTED MANUSCRIPT

revised The results of our study demonstrate that plasma glucose concentration is a significant

286

risk factor for metritis, and clinical endometritis. Cows that were diagnosed with metritis and

287

subsequently with clinical endometritis had higher plasma glucose levels after parturition

288

compared to healthy counterparts. Glucose plasma concentrations after parturition were also

289

found to be higher in cows that had RP than in cows without RP. For predicting RP, the ROC

290

curve with a higher AUC and likelihood ratios was for glucose concentrations at day 3.

291

However, it is unclear how changes in blood glucose at day 3 after parturition can affect RP on

292

the day of calving. Retained placenta leads to inflammation and bacterial contamination in the

293

early postpartum period that contribute to greater uterine inflammation; therefore, high glucose

294

concentrations in cows with retained placenta might be a consequence of the greater odds of

295

subsequent endometrial, cervical, or vaginal inflammation. Moreover, cows with an increased

296

calving to conception interval (not pregnant within 150 d) presented higher plasma glucose

297

concentrations than cows that became pregnant within the first 150 d. BHBA and NEFA plasma

298

concentrations did not differ between metritic or endometritic cows and their healthy

299

counterparts. Our data further established that cows with higher levels of glucose at day 3 after

300

parturition were 6.6 times more likely to be diagnosed with metritis, and 3.5 times more likely to

301

be further diagnosed with clinical endometritis compared to cows with lower glucose levels.

SC

M AN U

TE D

EP

AC C

302

RI PT

285

During the first week of lactation the overall plasma glucose concentration decreased by

303

18% and then kept decreasing until the end of the study period (day 14) to about 23%, in

304

comparison to the first measurement at day -50. This reduction in overall plasma glucose

305

concentration after parturition that we observed has also been reported by others [28, 41, 42].

306

Galvao et al. (2010) showed that cows that developed metritis had a lower neutrophil glycogen

307

concentration at calving when compared to healthy cows, and this was assumed to be reflective

15

ACCEPTED MANUSCRIPT

revised of low pre-partum blood glucose concentration, since in the same work, blood glucose was

309

shown to be positively associated with neutrophil glycogen store[28]. However, in our study

310

diseased cows (cows affected by at least one of the uterine diseases) exhibited higher glucose

311

concentration at day -50 and -6 relative to parturition than the cows that remained healthy.

RI PT

308

It is unknown whether the post-calving high glucose levels are directly responsible for

312

the observed increased risk of uterine diseases, but this is a possibility.

For instance,

314

hyperglycemia may be mediated through increased cortisol levels around calving concurrent to

315

increased insulin resistance after parturition and this is reported to be associated with

316

immunosuppression [28, 43, 44]. Recent research has shown evidence that cytokines, elicited by

317

the presence of endotoxins, directly stimulate lipolysis and the breakdown of liver glycogen [45,

318

46]. Metritis is an acute process caused by bacterial contamination around the time of calving

319

[47], when potential exposure to pathogens is most likely. For example, the intra-uterine

320

presence of E. coli at 1 to 3 days after parturition was found to be strongly associated with the

321

incidence of metritis [2]. Additionally, cows with retained placenta have higher levels of LPS

322

and E.coli in the uterine lochia than healthy cows[48]; hence, pathogenic bacteria can evade the

323

cow’s immune system by overriding neutrophil function. Additionally, microvascular damage

324

resulting from impaired endothelial function has been associated with increased glucose

325

concentrations in humans [49]. The negative effects of high glucose levels found in our study

326

may be associated with postpartum health through other mechanisms not assessed in the present

327

study.

AC C

EP

TE D

M AN U

SC

313

328

Although in the present study we found high glucose levels to be associated with clinical

329

endometritis, Senosy et al. (2012) and Galvao et al (2010) reported low blood glucose to be a

330

risk factor for cows with endometritial inflammation at week 4 and 7 postpartum[28, 50],

16

ACCEPTED MANUSCRIPT

revised respectively, whereas Burke et al. (2010) and Akbar et al. (2014) did not observe differences in

332

glucose concentrations between healthy cows and cows with subclinical endometritis[51, 52].

333

Those authors did not include cows with clinical signs of disease; therefore, the lack of

334

agreement may be associated with disease severity and the degree of the immune response to

335

infection. Studies of glucose metabolism indicate that, in the early phase of exposure to

336

endotoxin, there is increased hepatic glucose output as a consequence of increased hepatic

337

glycogenolysis

338

inflammatory events are accompanied by dramatic changes of hepatic protein synthesis.

339

Cytokines are responsible for this enhanced gluconeogenesis concomitant to changes in hepatic

340

protein metabolism, characterized by heightened acute-phase protein synthesis [54].

RI PT

331

resulting

in

hyperglycemia

SC

gluconeogenesis,

[53].

Additionally,

M AN U

and

Engagement of microbial molecules by Toll-like receptors of phagocytes induces the

342

production of pro-inflammatory cytokines, interleukin-1 (IL-l), tumor necrosis factor-alfa

343

(TNFα), and IL-6, which in turn trigger a systemic inflammatory response characterized by a

344

cascade of immune and metabolic changes to fuel immune-cell activation and elicit effective

345

antioxidant responses [55]. IL-1, IL-6, and TNFα can alter glucose homeostasis, both indirectly

346

by stimulating counter-regulatory hormone secretion and by direct action themselves [56].

347

Furthermore, IL-1 was also found to be released by endometrial cells in response to combined

348

host cell damage and bacterial infection [57]. TNFα has been shown to induce the production of

349

IL-1 in various cells [58]. Moreover, injections of recombinant bovine TNF have been shown to

350

induce extensive changes in the levels of plasma glucose, triglycerides, and NEFA, and also in

351

insulin and growth hormone concentrations in Holstein calves [59].

AC C

EP

TE D

341

352

The systemic inflammatory response leads to protein catabolism (weight loss, muscle

353

wasting) and appetite suppression [60]. It would be reasonable to expect that the inflammatory

17

ACCEPTED MANUSCRIPT

revised process happening in our study cows would lead to lower DMI, increased fat mobilization and

355

consequently increased plasma concentrations of NEFA and BHBA. We found no significant

356

link between plasma BHBA or NEFA and the occurrence of any uterine disease in the present

357

study, consistent with results of previous studies [52, 61]. Burke et al. (2010) and Dubuc et al.

358

(2010) suggested that energy status per se is not a risk factor for clinical endometritis; however,

359

both groups reported increased prepartum NEFA levels as a risk factor for metritis [52, 62].

360

Ospina et al. (2010) found that cows with NEFA concentration ≥0.6 mEq/L within 14 d before

361

calving had a greater risk of RP, metritis, or both [8]. On the other hand, Chapinal et al. (2011)

362

did not find a herd-level threshold for cows in the high NEFA group for the prediction of metritis

363

and RP, whereas pre-calving high NEFA was a predictor of both diseases at the cow level [35].

M AN U

SC

RI PT

354

Milk yield was negatively associated with blood glucose concentration in our study. A

365

similar finding was reported by Galvao et al. (2010). The impact of metabolic inflammation on

366

production has been shown recently, of which cows with higher levels of inflammatory markers

367

had reduced milk production in the first month, compared to cows presenting lower levels of

368

inflammation [63]. Moreover, Farney et al. (2013) showed that glucose levels were significantly

369

reduced by anti-inflammatory treatment; and anti-inflammatory treatment increased milk yield

370

by about 7% [64, 65]. Additionally, the presence of uterine disease has been shown to be

371

negatively associated with milk production [28, 52, 66]. In the present study, healthy cows

372

presented overall lower plasma glucose levels during the entire study period in comparison to

373

diseased cows, in agreement with previous reports[67, 68]. Because the mammary gland of

374

dairy cows is responsible for 50% to 85% of whole-body glucose consumption [68] it is

375

conceivable that healthier cows that produce more milk will present lower plasma glucose

376

concentrations. Furthermore, in the present study, plasma NEFA and BHBA concentrations were

AC C

EP

TE D

364

18

ACCEPTED MANUSCRIPT

revised 377

not found to be associated with milk loss, which is at odds with the findings of previous reports

378

[8, 69]. Finally, we reported herein that cows with increased calving to conception interval (days

380

at conception >150) had higher plasma glucose levels when compared to cows that got pregnant

381

within 150 days. Our results are in line with those of Garverick et al. (2013), who showed higher

382

early postpartum glucose concentrations in cows with reduced fertility[70]. Slower glucose

383

clearance and increased insulin resistance have been shown to be associated with the

384

development of ovarian cysts in postpartum cows [71]. Moreover, increased glucose

385

concentrations have been linked to early embryonic death accompanied by increased expression

386

of TNFα contributing to activation of apoptotic signaling in mice [72].

M AN U

SC

RI PT

379

Taken together, these results may indicate that elevated plasma glucose concentration is a

388

consequence of altered inflammatory responses leading to impaired health and reproductive

389

performance in postpartum dairy cows. We postulated that those negative effects are likely a

390

complex effect of systemic inflammatory condition, since all cows around parturition develops

391

NEB and over half of the cows become impervious to uterine infection. We acknowledged that

392

one limitation of the current study is the fact that all study cows were allocated in a single farm

393

and the metabolic markers evaluated here are highly dependent on animal’s energy status and

394

therefore, are variable with herd level of production or breed. The absence of association

395

between NEFA and prediction of postpartum uterine health reported in this study may be a

396

reflect of other mechanisms and parameters that were not evaluated in the present study; for

397

example, NEFA concentrations has been shown to be dependent on the degree of illness

398

experienced by cows after calving; and the onset of ovarian activity has been significantly

399

associated with prolonged exposure to NEB, rather than magnitude of NEB[73, 74]. Last, further

AC C

EP

TE D

387

19

ACCEPTED MANUSCRIPT

revised 400

molecular/clinical studies may help explain how glucose levels, specifically pathophysiological

401

glucose concentrations, affect metabolic dynamics and cellular functions in dairy cows.

403

RI PT

402

CONCLUSION

The present study found that metabolic glucose concentration was the only risk factor for

405

metritis, and clinical endometritis versus two other measured metabolic indicators of negative

406

energy balance, NEFA and BHBA. We found plasma glucose concentration to be associated

407

with the occurrence of metritis and clinical endometritis. Moreover, cows with an increased

408

calving to conception interval (>150 d) presented higher plasma glucose concentrations than

409

cows that became pregnant within the first 150 d. BHBA and NEFA did not differ between

410

metritic or endometritic cows compared to their healthy counterparts. Finally, milk yield was

411

negatively associated with blood glucose concentration. Neither NEFA nor BHBA concentration

412

was found to be associated with milk loss. Further molecular/clinical studies may help explain

413

how glucose levels, specifically pathophysiological glucose concentrations, affect metabolic

414

dynamics and cellular functions.

417 418 419

M AN U

TE D

EP

416

AC C

415

SC

404

420 421 422

20

ACCEPTED MANUSCRIPT

revised 423 424

Table1. Least squares means for glucose for cows that developed uterine diseases or remained

426

healthy at days relative to parturition

RP Healthy Primiparous Multiparous

Parity Sample collection day (SCD)1 Parity*RP Parity*RP*SCD Model 2

Metritis Healthy Primiparous Multiparous

Metritis Parity

Clinical Endometritis(CE)

432

43.43-50.17 42.10-44.67 41.95-47.69 43.14-47.56

CE Healthy Primiparous Multiparous

46.92 42.72 44.17 45.47

44.18-49.66 41.39-44.05 41.76-46.59 43.61-47.32

P-value

≤ 0.01 0.91

≤ 0.01 0.68 0.17 0.02 0.77 ≤ 0.01 0.21 0.15 ≤ 0.01 0.40

EP

Sample collection day (SCD)1 ≤ 0.01 Parity* CE Figure 3 0.32 Parity* CE *SCD 0.02 1 Days relative to parturition at sample collection. Blood samples were collected at 20 and 6 days prior to estimated parturition day and at 3, 7, and 14 days postpartum. 2 LSM = least squares mean 3 95% C.I. = 95% Confidence interval of the mean

AC C

427 428 429 430 431

46.79 43.39 44.82 45.35

Figure 2

TE D

Sample collection day (SCD)1 Parity*metritis Parity*metritis*SCD Model 3

Parity

Figure 1

95% C.I.3 45.54-55.66 42.05-44.51 42.42-51.73 44.48-49.13

M AN U

Retained placenta

LSM2 51.60 43.28 47.07 46.80

SC

Model 1

RI PT

425

21

ACCEPTED MANUSCRIPT

revised 433

Table2. Least squares means for BHBA for cows that developed uterine diseases or remained

434

healthy at days relative to parturition

Parity Sample collection day (SCD)1 Parity*RP Parity*RP*SCD Model 2

Parity Sample collection day (SCD)1 Parity*metritis Parity*metritis*SCD Model 3

440

CE Healthy Primiparous Multiparous

0.7115 0.7198 0.7333 0.6981

P-value 0.48 0.09 ≤ 0.01 0.17 0.82 0.36 0.33 ≤ 0.01 0.60 0.76

0.6518-0.7713 0.688-0.7517 0.6826-0.7839 0.6515-0.7447

0.80 0.30

EP

Sample collection day (SCD)1 ≤ 0.01 Parity* CE Figure 3 0.52 Parity* CE*SCD 0.90 1 Days relative to parturition at sample collection. Blood samples were collected at 20 and 6 days prior to estimated parturition day and at 3, 7, and 14 days postpartum. 2 LSM = least squares mean 3 95% C.I. = 95% Confidence interval of the mean

AC C

435 436 437 438 439

0.6051-0.7524 0.6864-0.7443 0.6533-0.7794 0.6278-0.7276

Figure 2

TE D

Clinical Endometritis(CE)

0.6787 0.7154 0.7164 0.6777

M AN U

Metritis Healthy Primiparous Multiparous

Metritis

Parity

Figure 1

95% C.I.3 0.6469-0.8354 0.6786-0.7347 0.6869-0.8462 0.6198-0.7428

RI PT

RP Healthy Primiparous Multiparous

Retained placenta

LSM2 0.7412 0.7066 0.7665 0.6813

SC

Model 1

22

ACCEPTED MANUSCRIPT

revised Table3. Least squares means for NEFA for cows that developed uterine diseases or remained healthy

Parity Sample collection day (SCD)1 Parity*RP Parity*RP*SCD Model 2

Figure 1

Parity Sample collection day (SCD)1 Parity*metritis Parity*metritis*SCD Model 3 Clinical Endometritis(CE) Parity

0.6787 0.7154 0.7164 0.6777

M AN U

Metritis Healthy Primiparous Multiparous

Metritis

95% C.I.3 0.6469-0.8354 0.6786-0.7347 0.6869-0.8462 0.6198-0.7428

RI PT

RP Healthy Primiparous Multiparous

Retained placenta

LSM2 0.7412 0.7066 0.7665 0.6813

0.6051-0.7524 0.6864-0.7443 0.6533-0.7794 0.6278-0.7276

SC

Model 1

Figure 2

CE Healthy Primiparous Multiparous

0.7115 0.7198 0.7333 0.6981

0.6518-0.7713 0.688-0.7517 0.6826-0.7839 0.6515-0.7447

P-value 0.48 0.09 ≤ 0.01 0.17 0.82 0.36 0.33 ≤ 0.01 0.60 0.76 0.80 0.30

AC C

441

EP

TE D

Sample collection day (SCD)1 ≤ 0.01 Parity* CE Figure 3 0.52 Parity* CE*SCD 0.90 1 Days relative to parturition at sample collection. Blood samples were collected at 20 and 6 days prior to estimated parturition day and at 3, 7, and 14 days postpartum. 2 LSM = least squares mean 3 95% C.I. = 95% Confidence interval of the mean

23

ACCEPTED MANUSCRIPT

revised 442

Table4. Least square means for plasma glucose concentrations for cows that were diagnosed

443

pregnant or not pregnant within 150 DIM.

Yes No Primiparous Multiparous

Pregnancy <150 d Parity

95% C.I.3 44.10-46.59 41.01-43.39 43.09-45.27 42.02-44.66

P-value <0.01 0.33

SC

Sample collection day (SCD)1 <0.01 Parity*Preg<150 Figure 4 0.43 Parity*Preg<150*SCD 0.78 1 Days relative to parturition at sample collection. Blood samples were collected at 20 and 6 days prior to estimated parturition day and at 3, 7, and 14 days postpartum. 2 LSM = least squares mean 3 95% C.I. = 95% Confidence interval of the mean

M AN U

444 445 446 447 448 449

LSM2 45.33 42.20 44.18 43.34

RI PT

Model 1

AC C

EP

TE D

450

24

ACCEPTED MANUSCRIPT

revised Table5. Receiver operator characteristic (ROC) curve analysis for the determination of critical

452

thresholds for glucose, NEFA, and BHBA; as predictors of uterine diseases in dairy cows. The

453

results of ROC curve determination of critical glucose, NEFA, and BHBA thresholds for the

454

prediction of an individual disease at each time point were tabulated and ranked by their

455

respective AUC. The higher AUC values referred to each metabolite and disease are listed

456

below. Additional information on levels of glucose, NEFA, and BHBA showing sensitivity,

457

specificity, and LR+ is provided. Disease Days1

Threshold2

467 468 469 470 471

21.8-66.0 43.9 - 80.1 32.5-70.6 29.5-63.1 45.4 - 80.8 22.1 - 53.1

84.5-94.7 59.7 - 76.5 55.8-70.6 60.7-77.5 46.3 - 61.5 63.2 - 78.3

90.5 68.5 63.4 69.6 54 71.7

0.66 0.67 0.54 0.57 0.58 0.52

0.018 0.005 0.41 0.18 0.11 0.7

TE D

4.5 2.01 1.41 1.51 1.4 1.27

EP

466

95%CI for Sp LR5₊ AUC6 Pvalue

AC C

458 459 460 461 462 463 464 465

MET 3 56 42.9 CE 3 45 63.3 NEFA MET 3 0.40 51.7 (mEq/L) CE -6 0.23 45.9 BHBA MET 14 0.6 64.5 (mmol/L) CE 3 0.9 36.6 Disease: MET: Metritis, CE: Clinical Endometritis 1 Days relative to parturition 2 Highest combined specificity and sensitivity 3 Se: epidemiologic sensitivity 4 Sp: epidemiologic specificity 5 Likelihood ratio positive 6 AUC: area under the curve

95%CI for Se Sp4

M AN U

Glucose (mg/dL)

Se3

SC

Metabolite

RI PT

451

472 473 474 475

25

ACCEPTED MANUSCRIPT

revised Figure1. Least square means ± SEM for plasma glucose, BHBA, and NEFA concentrations for

477

A) primiparous and B) multiparous cows that presented Retained Placenta or remained healthy

478

at days relative to parturition. The gray line represents the cows that presented RP and the black

479

line represents the cows that remained healthy.

M AN U

SC

RI PT

476

-6

3

7

14

AC C

EP

TE D

0 -5

480 481

26

ACCEPTED MANUSCRIPT

revised

483

A) primiparous and B) multiparous cows that developed metritis or remained healthy at days

484

relative to parturition. The gray line represents cows that developed metritis and black line

485

represents cows that remained healthy.

14

3

7

SC -6

14

7

3

-6

BHBA (mmol/L)

-5

0

M AN U

14 14

7

EP 3

14

7

3

-6

-5 0

14

7

3

-6

-5

0

AC C

-6

TE D

7

3

-6

0 -5

BHBA (mmol/L)

RI PT

Figure2. Least square means ± SEM for plasma glucose, BHBA, and NEFA concentrations for

Glucose, mg/dl

482

486 27

ACCEPTED MANUSCRIPT

revised Figure3. Least square means ± SEM for plasma glucose, BHBA, and NEFA concentrations for

488

A) primiparous and B) multiparous cows that showed clinical endometritis or remained healthy

489

at days relative to parturition. The gray line represents cows that developed endometritis and

490

black line represents cows that remained healthy.

RI PT

487

491 70

A

B

40 35 30

50 40

14

1.0

B

0.4 0.2

AC C

Days relative to parturition

14

7

0.0

14

-6

3

0.0

7

EP

0.5

0.6

3

1.0

0.8

-6

TE D

A

Days relative to parturition 0.8

A

0.4

7

Days relative to parturition

1.5

0.6

3

-5 0

14

7

3

-6

-5 0

30

Days relative to parturition

B 0.6 0.4

0.2 0.2

Days relative to parturition

14

7

3

-6

-5 0

14

7

3

0.0 -6

0.0 -5 0

BHBA (mmol/L)

SC

Glucose, mg/dl

45

60

M AN U

50

BHBA (mmol/L)

Glucose, mg/dl

55

-6

60

Days relative to parturition

28

ACCEPTED MANUSCRIPT

revised Figure4. Least square means ± SEM for plasma glucose concentrations for A) primiparous and

493

B) multiparous cows that were pregnant <150 d. The gray line represents cows that did not

494

become pregnant and the black line represents cows that became pregnant

RI PT

492

AC C

EP

TE D

M AN U

SC

495

29

ACCEPTED MANUSCRIPT

revised Figure5. Display of receiver operator characteristic (ROC) curves that determined the critical

497

threshold for concentrations of glucose at day 3 (≥56 mg/dL; gray line), NEFA at day 3

498

(≥0.40mEq/L; dotted line), and BHBA at day 14 (≥0.6 mmol/L; black solid line) predicting

499

metritis in dairy cows. Pairwise comparison of AUC values of ROC curves determine the

500

difference between the AUC for glucose and NEFA (0.165; SE: 0.0951; 95%CI:-0.0211 to

501

0.352, P = 0.0821); the difference between AUC for glucose and BHBA (0.0961; SE: 0.0954;

502

95%CI:-0.0909 to 0.283, P = 0.3138); the difference between AUC for NEFA and BHBA

503

(0.0693; SE: 0.103; 95%CI:-0.133 to 0.272, P = 0.5022). The diagonal line represents the

504

sensitivity and specificity level at which the test in non-informative.

AC C

EP

TE D

M AN U

SC

RI PT

496

505 506 507 508 509 510 511 512 513 514

30

ACCEPTED MANUSCRIPT

revised Figure6. Display of receiver operator characteristic (ROC) curves that determined the critical

516

threshold for concentrations of glucose at day 3 (≥45 mg/dL; gray line), NEFA at day -6

517

(≥0.23mEq/L; dotted line), and BHBA at day 3 (≥0.9 mmol/L; black solid line) predicting

518

clinical endometritis in dairy cows. Pairwise comparison of AUC values of ROC curves

519

determine the difference between the AUC for glucose and NEFA (0. 0950; SE: 0.102; 95%CI:-

520

0.106 to 0.295, P = 0.3532); the difference between AUC for glucose and BHBA (0. 0370; SE:

521

0.0947; 95%CI:-0.149 to 0.223, P = 0.6960); the difference between AUC for NEFA and BHBA

522

(0.0579; SE: 0.0961; 95%CI:-0.130 to 0.246, P = 0.5464). The diagonal line represents the

523

sensitivity and specificity level at which the test in non-informative.

M AN U

SC

RI PT

515

AC C

EP

TE D

524

31

ACCEPTED MANUSCRIPT

revised Figure7. Correlation between average daily milk production for the first and second week of

526

lactation and plasma glucose at 7 and 14 d relative to parturition, respectively. Bivariate

527

correlation was estimated by the REML method. A) Milk Wk1= 47.657 - 0.169 x Glucose day7;

528

N = 159; R = 0.31; P <0.001. B) Milk Wk2= 50.903 - 0.174 x Glucose day14; N = 159; R =

529

0.27; P = 0.001.

RI PT

525

530

A) Glucose at day 7

SC

531

M AN U

532 533 534 535

539

B) Glucose at day 14

EP

538

AC C

537

TE D

536

32

ACCEPTED MANUSCRIPT

revised 540

6. 7.

8.

9. 10. 11.

12. 13. 14. 15.

16. 17.

18. 19. 20.

RI PT

5.

SC

4.

M AN U

3.

TE D

2.

Bicalho, R.C., et al., Molecular and epidemiological characterization of bovine intrauterine Escherichia coli. J Dairy Sci, 2010. 93(12): p. 5818-30. Bicalho, M.L., et al., Association between virulence factors of Escherichia coli, Fusobacterium necrophorum, and Arcanobacterium pyogenes and uterine diseases of dairy cows. Vet Microbiol, 2012. 157(1-2): p. 125-31. Sordillo, L.M., G.A. Contreras, and S.L. Aitken, Metabolic factors affecting the inflammatory response of periparturient dairy cows. Anim Health Res Rev, 2009. 10(1): p. 53-63. Butler, W.R., R.W. Everett, and C.E. Coppock, The relationships between energy balance, milk production and ovulation in postpartum Holstein cows. J Anim Sci, 1981. 53(3): p. 742-8. Cheong, S.H., et al., Cow-level and herd-level risk factors for subclinical endometritis in lactating Holstein cows. J Dairy Sci, 2011. 94(2): p. 762-70. Kimura, K., et al., Decreased neutrophil function as a cause of retained placenta in dairy cattle. J Dairy Sci, 2002. 85(3): p. 544-50. Martinez, N., et al., Evaluation of peripartal calcium status, energetic profile, and neutrophil function in dairy cows at low or high risk of developing uterine disease. J Dairy Sci, 2012. 95(12): p. 7158-72. Ospina, P.A., et al., Evaluation of nonesterified fatty acids and beta-hydroxybutyrate in transition dairy cattle in the northeastern United States: Critical thresholds for prediction of clinical diseases. J Dairy Sci, 2010. 93(2): p. 546-54. LeBlanc, S.J., K.E. Leslie, and T.F. Duffield, Metabolic predictors of displaced abomasum in dairy cattle. J Dairy Sci, 2005. 88(1): p. 159-70. Cai, T.Q., et al., Association between neutrophil functions and periparturient disorders in cows. Am J Vet Res, 1994. 55(7): p. 934-43. Kimura, K., J.P. Goff, and M.E. Kehrli, Jr., Effects of the presence of the mammary gland on expression of neutrophil adhesion molecules and myeloperoxidase activity in periparturient dairy cows. J Dairy Sci, 1999. 82(11): p. 2385-92. Hammon, D.S., et al., Neutrophil function and energy status in Holstein cows with uterine health disorders. Vet Immunol Immunopathol, 2006. 113(1-2): p. 21-9. Kasimanickam, R., et al., Endometrial cytology and ultrasonography for the detection of subclinical endometritis in postpartum dairy cows. Theriogenology, 2004. 62(1-2): p. 9-23. Gilbert, R.O., et al., Prevalence of endometritis and its effects on reproductive performance of dairy cows. Theriogenology, 2005. 64(9): p. 1879-88. Bromfield, J.J., et al., PHYSIOLOGY AND ENDOCRINOLOGY SYMPOSIUM: Uterine infection: linking infection and innate immunity with infertility in the high-producing dairy cow. J Anim Sci, 2015. 93(5): p. 2021-33. Ribeiro, E.S., et al., Carryover effect of postpartum inflammatory diseases on developmental biology and fertility in lactating dairy cows. J Dairy Sci, 2016. 99(3): p. 2201-20. von Soosten, D., et al., Effect of trans-10, cis-12 conjugated linoleic acid on performance, adipose depot weights, and liver weight in early-lactation dairy cows. J Dairy Sci, 2011. 94(6): p. 2859-70. Reynolds, C.K., et al., Splanchnic metabolism of dairy cows during the transition from late gestation through early lactation. J Dairy Sci, 2003. 86(4): p. 1201-17. Andersen, B., G.H. Goldsmith, and P.J. Spagnuolo, Neutrophil adhesive dysfunction in diabetes mellitus; the role of cellular and plasma factors. J Lab Clin Med, 1988. 111(3): p. 275-85. Contreras, G.A., et al., Nonesterified fatty acids modify inflammatory response and eicosanoid biosynthesis in bovine endothelial cells. J Dairy Sci, 2012. 95(9): p. 5011-23.

EP

1.

AC C

541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586

33

ACCEPTED MANUSCRIPT

revised

26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.

38. 39. 40. 41. 42. 43. 44.

RI PT

25.

SC

24.

M AN U

23.

TE D

22.

Bauman, D.E. and W.B. Currie, Partitioning of nutrients during pregnancy and lactation: a review of mechanisms involving homeostasis and homeorhesis. J Dairy Sci, 1980. 63(9): p. 151429. Borregaard, N. and T. Herlin, Energy metabolism of human neutrophils during phagocytosis. J Clin Invest, 1982. 70(3): p. 550-7. Weisdorf, D.J., P.R. Craddock, and H.S. Jacob, Granulocytes utilize different energy sources for movement and phagocytosis. Inflammation, 1982. 6(3): p. 245-56. Andersson, L., Subclinical ketosis in dairy cows. Vet Clin North Am Food Anim Pract, 1988. 4(2): p. 233-51. Lin, X., J.K. Candlish, and A.C. Thai, Superoxide production by neutrophils from diabetics and normal subjects in response to glucose and galactose. Exp Mol Pathol, 1993. 58(3): p. 229-36. Ortmeyer, J. and V. Mohsenin, Glucose suppresses superoxide generation in normal neutrophils: interference in phospholipase D activation. Am J Physiol, 1993. 264(2 Pt 1): p. C402-10. Kimura, K., T.A. Reinhardt, and J.P. Goff, Parturition and hypocalcemia blunts calcium signals in immune cells of dairy cattle. J Dairy Sci, 2006. 89(7): p. 2588-95. Galvao, K.N., et al., Association between uterine disease and indicators of neutrophil and systemic energy status in lactating Holstein cows. J Dairy Sci, 2010. 93(7): p. 2926-37. Dohoo, I.R. and S.W. Martin, Subclinical ketosis: prevalence and associations with production and disease. Can J Comp Med, 1984. 48(1): p. 1-5. Rabiee, A.R. and I.J. Lean, Uptake of glucose and cholesterol by the ovary of sheep and cattle and the influence of arterial LH concentrations. Anim Reprod Sci, 2000. 64(3-4): p. 199-209. Westwood, C.T., I.J. Lean, and J.K. Garvin, Factors influencing fertility of Holstein dairy cows: a multivariate description. J Dairy Sci, 2002. 85(12): p. 3225-37. Butler, W.R. and R.D. Smith, Interrelationships between energy balance and postpartum reproductive function in dairy cattle. J Dairy Sci, 1989. 72(3): p. 767-83. Kappel, L.C., et al., Relationship between fertility and blood glucose and cholesterol concentrations in Holstein cows. Am J Vet Res, 1984. 45(12): p. 2607-12. Duffield, T.F., et al., Impact of hyperketonemia in early lactation dairy cows on health and production. J Dairy Sci, 2009. 92(2): p. 571-80. Chapinal, N., et al., The association of serum metabolites with clinical disease during the transition period. J Dairy Sci, 2011. 94(10): p. 4897-903. Edmonson, A.J., I. J. Lean, L. D. Weaver, T. Farver, and G. Webster, A body condition scoring chart of Holstein dairy cows. . Anim. Physiol. Anim. Nutr. (Berl), 1989. 90: p. 459-466. McDougall, S., R. Macaulay, and C. Compton, Association between endometritis diagnosis using a novel intravaginal device and reproductive performance in dairy cattle. Anim Reprod Sci, 2007. 99(1-2): p. 9-23. Gordon, J.L., S.J. Leblanc, and T.F. Duffield, Ketosis treatment in lactating dairy cattle. Vet Clin North Am Food Anim Pract, 2013. 29(2): p. 433-45. Greiner, M., D. Pfeiffer, and R.D. Smith, Principles and practical application of the receiveroperating characteristic analysis for diagnostic tests. Prev Vet Med, 2000. 45(1-2): p. 23-41. Swets, J.A., Measuring the accuracy of diagnostic systems. Science, 1988. 240(4857): p. 128593. Skaar, T.C., et al., Seasonal effects of prepartum and postpartum fat and niacin feeding on lactation performance and lipid metabolism. J Dairy Sci, 1989. 72(8): p. 2028-38. Vazquez-Anon, M., et al., Peripartum liver triglyceride and plasma metabolites in dairy cows. J Dairy Sci, 1994. 77(6): p. 1521-8. Roth, J.A. and M.L. Kaeberle, Effect of glucocorticoids on the bovine immune system. J Am Vet Med Assoc, 1982. 180(8): p. 894-901. Galvao, K.N., et al., Association between endometritis and endometrial cytokine expression in postpartum Holstein cows. Theriogenology, 2011. 76(2): p. 290-9.

EP

21.

AC C

587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636

34

ACCEPTED MANUSCRIPT

revised

50.

51. 52. 53. 54. 55. 56. 57. 58. 59.

60. 61.

62. 63. 64. 65.

66.

RI PT

49.

SC

48.

M AN U

47.

TE D

46.

Bradford, B.J., et al., Daily injection of tumor necrosis factor-{alpha} increases hepatic triglycerides and alters transcript abundance of metabolic genes in lactating dairy cattle. J Nutr, 2009. 139(8): p. 1451-6. Eckel, E.F. and B.N. Ametaj, Invited review: Role of bacterial endotoxins in the etiopathogenesis of periparturient diseases of transition dairy cows. J Dairy Sci, 2016. 99(8): p. 5967-90. Sheldon, I.M., A.N. Rycroft, and C. Zhou, Association between postpartum pyrexia and uterine bacterial infection in dairy cattle. Vet Rec, 2004. 154(10): p. 289-93. Dohmen, M.J., et al., Relationship between intra-uterine bacterial contamination, endotoxin levels and the development of endometritis in postpartum cows with dystocia or retained placenta. Theriogenology, 2000. 54(7): p. 1019-32. van den Oever, I.A., et al., Endothelial dysfunction, inflammation, and apoptosis in diabetes mellitus. Mediators Inflamm, 2010. 2010: p. 792393. Senosy, W.S., Y. Izaike, and T. Osawa, Influences of metabolic traits on subclinical endometritis at different intervals postpartum in high milking cows. Reprod Domest Anim, 2012. 47(4): p. 666-74. Akbar, H., et al., Postpartal subclinical endometritis alters transcriptome profiles in liver and adipose tissue of dairy cows. Bioinform Biol Insights, 2014. 8: p. 45-63. Burke, C.R., et al., Relationships between endometritis and metabolic state during the transition period in pasture-grazed dairy cows. J Dairy Sci, 2010. 93(11): p. 5363-73. Hijmans, B.S., et al., Zonation of glucose and fatty acid metabolism in the liver: mechanism and metabolic consequences. Biochimie, 2014. 96: p. 121-9. Andus, T., J. Bauer, and W. Gerok, Effects of cytokines on the liver. Hepatology, 1991. 13(2): p. 364-75. Johnson, R.W., The concept of sickness behavior: a brief chronological account of four key discoveries. Vet Immunol Immunopathol, 2002. 87(3-4): p. 443-50. Webber, J., Abnormalities in glucose metabolism and their relevance to nutrition support in the critically ill. Curr Opin Clin Nutr Metab Care, 1998. 1(2): p. 191-4. Healy, L.L., J.G. Cronin, and I.M. Sheldon, Endometrial cells sense and react to tissue damage during infection of the bovine endometrium via interleukin 1. Sci Rep, 2014. 4: p. 7060. Le, J. and J. Vilcek, Tumor necrosis factor and interleukin 1: cytokines with multiple overlapping biological activities. Lab Invest, 1987. 56(3): p. 234-48. Kushibiki, S., et al., Administration of recombinant bovine tumor necrosis factor-alpha affects intermediary metabolism and insulin and growth hormone secretion in dairy heifers. J Anim Sci, 2000. 78(8): p. 2164-71. Horn, J., et al., Influence of insulin and substrate concentration on protein synthetic rate in fetal tissues. Res Vet Sci, 1983. 35(1): p. 35-41. Bicalho, M.L., et al., Effect of trace mineral supplementation on selected minerals, energy metabolites, oxidative stress, and immune parameters and its association with uterine diseases in dairy cattle. J Dairy Sci, 2014. 97(7): p. 4281-95. Dubuc, J., et al., Risk factors for postpartum uterine diseases in dairy cows. J Dairy Sci, 2010. 93(12): p. 5764-71. Bertoni, G., et al., Effects of inflammatory conditions on liver activity in puerperium period and consequences for performance in dairy cows. J Dairy Sci, 2008. 91(9): p. 3300-10. Farney, J.K., et al., Anti-inflammatory salicylate treatment alters the metabolic adaptations to lactation in dairy cattle. Am J Physiol Regul Integr Comp Physiol, 2013. 305(2): p. R110-7. Carpenter, A.J., et al., Hot topic: Early postpartum treatment of commercial dairy cows with nonsteroidal antiinflammatory drugs increases whole-lactation milk yield. J Dairy Sci, 2016. 99(1): p. 672-9. Dubuc, J., et al., Effects of postpartum uterine diseases on milk production and culling in dairy cows. J Dairy Sci, 2011. 94(3): p. 1339-46.

EP

45.

AC C

637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686

35

ACCEPTED MANUSCRIPT

revised

69. 70.

71.

72. 73. 74.

RI PT

68.

Drackley, J.K., ADSA Foundation Scholar Award. Biology of dairy cows during the transition period: the final frontier? J Dairy Sci, 1999. 82(11): p. 2259-73. Zhao, F.Q. and A.F. Keating, Expression and regulation of glucose transporters in the bovine mammary gland. J Dairy Sci, 2007. 90 Suppl 1: p. E76-86. McArt, J.A., et al., Elevated non-esterified fatty acids and beta-hydroxybutyrate and their association with transition dairy cow performance. Vet J, 2013. 198(3): p. 560-70. Garverick, H.A., et al., Concentrations of nonesterified fatty acids and glucose in blood of periparturient dairy cows are indicative of pregnancy success at first insemination. J Dairy Sci, 2013. 96(1): p. 181-8. Vanholder, T., et al., Hormonal and metabolic profiles of high-yielding dairy cows prior to ovarian cyst formation or first ovulation post partum. Reprod Domest Anim, 2005. 40(5): p. 4607. Torchinsky, A., et al., TNF-alpha acts to prevent occurrence of malformed fetuses in diabetic mice. Diabetologia, 2004. 47(1): p. 132-9. Canfield, R.W. and W.R. Butler, Energy balance and pulsatile LH secretion in early postpartum dairy cattle. Domest Anim Endocrinol, 1990. 7(3): p. 323-30. Huzzey, J.M., et al., Associations of prepartum plasma cortisol, haptoglobin, fecal cortisol metabolites, and nonesterified fatty acids with postpartum health status in Holstein dairy cows. J Dairy Sci, 2011. 94(12): p. 5878-89.

SC

67.

M AN U

687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706

AC C

EP

TE D

707

36

ACCEPTED MANUSCRIPT Most current research has focused on the evaluation of NEFA and BHBA as markers of negative energy balance and predictors of postpartum diseases, while few studies have evaluated the importance of glucose as a potentially significant risk factor for the development of uterine diseases. Therefore, the objective of this study was to investigate

RI PT

associations between the metabolic indicators NEFA, BHBA, and glucose during the transition period and the occurrence of uterine diseases and the subsequent effect on fertility. Because uterine diseases have negative effect on milk production, and milk production is

SC

accompanied by changes in glucose and energy metabolism, milk yield was also evaluated. Plasma glucose, NEFA, and BHBA concentrations were measured at -50, -6, 3, 7 and

M AN U

14 d relative to parturition. We found plasma glucose concentration to be associated with the occurrence of metritis and clinical endometritis; and with an increased calving to conception interval (>150 d). BHBA and NEFA were not associated with the occurrence of any uterine disorder. Additionally, pairwise comparisons of area under the curve (AUC) of ROC curves

TE D

for the critical thresholds for glucose, BHBA, and NEFA predicting the same uterine disease were performed. Glucose at 3 DIM was the best predictor for metritis and endometritis diagnosis, with AUC values of 0.66 and 0.67, respectively.

EP

Multivariable logistic regressions showed that cows with higher levels of glucose at

AC C

day 3 were at 6.6 times higher odds of being diagnosed with metritis, and 3.5 times higher odds of developing clinical endometritis, compared to cows with lower glucose levels. Finally, a simple linear regression analysis demonstrated a negative correlation between daily milk yield in the first and second weeks of lactation and plasma glucose concentrations measured at days 7 and 14, respectively. NEFA and BHBA concentrations were not found to be associated with milk production.