Abstract
Purulent vaginal discharge (PVD) is a common uterine disease in dairy cattle that has negative effects on reproductive performance. Reproductive management programs that synchronize ovulation use gonadotropin-releasing hormone (GnRH) to induce ovulation and prostaglandin F2α (PGF2α) to induce luteolysis. The objectives of this study were to evaluate ovarian response to treatment with GnRH and the odds of bearing a corpus luteum or being inseminated in dairy cows with or without PVD. Another objective was to determine the hazard of insemination after administration of PGF2α in dairy cows with or without PVD. Primiparous (n = 291) and multiparous (n = 402) cows were evaluated for PVD using a Metricheck device at 46 ± 3 and 35 ± 3 days in milk (DIM) (study day 0), respectively. On study day 14, primiparous (n = 107) and multiparous (n = 197) cows were treated with GnRH and subsequent ovulation was recorded. Primiparous (n = 178) and multiparous (n = 368) cows not inseminated by study day 21 were administered PGF2α and response to PGF2α treatment was determined by detection of estrus. Furthermore, cows were categorized by the presence of a CL or being inseminated by study days 14, 21, and 35. Overall prevalence of PVD was 28.5% and 13.4% for primiparous and multiparous cows, respectively. Projected 305-d milk yield was less (P < 0.01) in PVD+ multiparous cows compared with PVD– multiparous cows, however, no (P = 0.26) difference was detected between primiparous PVD+ and PVD– cows. Ovulatory response to GnRH treatment was 51.8% and 47.8% for primiparous and multiparous cows, respectively. Primiparous PVD– cows tended (P = 0.06) to be less likely to ovulate to GnRH than primiparous PVD+ cows, whereas multiparous PVD+ cows were less (P = 0.04) likely to ovulate to GnRH than PVD– multiparous cows. The odds of bearing a corpus luteum or being inseminated by study days 14, 21, or 35 was not associated with PVD in primiparous cows. In contrast, the odds of bearing a corpus luteum or being inseminated by study days 14 and 21 was (P ≤ 0.03) associated with PVD in multiparous cows, but not (P = 0.11) on study day 35. Hazard of insemination after PGF2α was not (P ≥ 0.38) associated with PVD in primiparous or multiparous cows. Purulent vaginal discharge is associated with response to treatment with GnRH in dairy cattle. Purulent vaginal discharge might negatively affect reproductive management programs that use GnRH to induce ovulation.
Keywords: dairy cow, endometritis, estrus detection, gonadotropin-releasing hormone, ovulation, prostaglandin F2α
INTRODUCTION
Clinical endometritis (CE) is a common uterine disorder in dairy cows, which has been defined as the presence of purulent vaginal discharge (PVD) 26 d postpartum or cervical diameter ≥ 7.5 cm determined by transrectal palpation 20 d postpartum (LeBlanc et al., 2002). Prevalence of CE across herds ranges from 5% to 26% (LeBlanc et al., 2002). Dubuc et al. (2010) suggested that using the term CE is not accurate in cases that uterine health is only determined by presence of PVD. Previous studies showed that PVD impacts reproductive efficiency of dairy cows by decreasing pregnancy per AI (McDougall et al., 2007) and increasing days to first AI (LeBlanc et al., 2002) and pregnancy (Dubuc et al., 2010).
Bacterial infection of the uterus is associated with impaired ovarian function including prolonged anestrous, decreased follicle growth, reduced plasma concentration of estradiol, and increased risk of cystic ovarian follicles (Mateus et al., 2002; Sheldon et al., 2002). Furthermore, function and life span of the corpus luteum (CL) can be influenced by bacterial infection of the uterus (Lüttgenau et al., 2016). Reproductive management programs commonly used in dairies depend on inducing ovulation with gonadotropin-releasing hormone (GnRH) and luteolysis with prostaglandin F2α (PGF2α) to synchronize ovulation. The influence of PVD on the success of reproductive management strategies is not completely understood, except that fertility is decreased in dairy cows with PVD. Ovarian responses to treatment with GnRH and PGF2α of dairy cows with PVD have not been fully investigated, therefore, we hypothesized that dairy cows with PVD would be less likely to ovulate in response to treatment with GnRH and less likely to be detected in estrus after treatment with PGF2α. The objectives of the current study were to evaluate GnRH-induced ovulation risk and odds of bearing a CL or being inseminated in cows with or without PVD. A third objective was to determine hazard of insemination after administration of PGF2α in cows with or without PVD.
MATERIALS AND METHODS
All animal care and management procedures were approved by the Kansas State University Institutional Animal Care and Use Committee (Manhattan, KS).
Animals
Cows utilized in this retrospective observational study were a subset of cows from a previous study (Voelz et al., 2016). The current study was conducted at two commercial dairy farms located in southwest Kansas from June through August 2014. Cows from dairy A were housed in dry-lot corrals with shade and milked twice daily. Cows from dairy B were housed in a free-stall barn with fans and sprinklers. Furthermore, cows from dairy B had access to a dirt exercise lot and were milked thrice daily. Both dairies fed a TMR once daily with ad libitum access to feed and water. Complete reproductive management procedures for primiparous and multiparous cows have previously been described (Voelz et al., 2016). Briefly, in the previous study (Voelz et al., 2016), cows were randomly assigned to one of two treatments: control and Gpresynch. Control primiparous and multiparous cows were synchronized with two injections of PGF2α given 14 d apart starting at 67 ± 3 and 56 ± 3 days in milk (DIM), respectively. Cows not detected in estrus 14 d after the last injection of PGF2α were submitted to the Cosynch-72 timed AI protocol, which consisted of starting the protocol with a GnRH injection, PGF2α 7 d later, and GnRH and timed AI 10 d after the initiation of the protocol. Cows submitted to the Gpresynch protocol were treated with GnRH 1 wk before the first injection of PGF2α, then followed the same protocol as control cows. Therefore, Gpresynch primiparous and multiparous cows were treated with GnRH at 60 ± 3 and 49 ± 3 DIM, respectively. Estrus detection was conducted once daily based on tail paint removal. Cows in both treatments (control and Gpresynch) were deemed eligible to be inseminated at 53 DIM. Thus, cows detected in estrus after 53 DIM were inseminated and received no further treatment.
Experimental Procedures
Experimental procedures are illustrated in Fig. 1. Weekly cohorts of primiparous (n = 291) and multiparous (n = 402) cows were evaluated for PVD 21 d before first injection of PGF2α. Primiparous and multiparous cows were examined at 46 ± 3 and 35 ± 3 DIM (study day 0), respectively. Purulent vaginal discharge was determined using the Metricheck device (Simcro, Hamilton, New Zealand) as previously described (McDougall et al., 2007). Vaginal discharge was scored by a single technician on a four-point scale (0 = clear mucus; 1 = clear mucus with flecks of white pus; 2 = ≤ 50% white or off-white exudate; 3 = > 50% purulent white, yellow; or 4 = blood-colored exudate; Williams et al., 2005). Cows with vaginal discharge score ≥ 2 were categorized as being purulent vaginal discharge positive (PVD+) and cows with a vaginal discharge score < 2 were classified as being purulent vaginal discharge negative (PVD−). Furthermore, body condition score (BCS) (0.25-point increments; 1 = thin and 5 = fat) was assessed at enrollment (study day 0; Ferguson et al., 1994). BCS was categorized as high (≥3.00) or low (<3.00). For primiparous cows, ultrasound examinations of ovaries were performed at 46 ± 3 (study day 0), 60 ± 3 (study day 14), and 67 ± 3 (study day 21) DIM. Ovaries of multiparous cows were examined at 35 ± 3 (study day 0), 49 ± 3 (study day 14), and 56 ± 3 (study day 21) DIM. Number of CL and follicles ≥ 10 mm in diameter were recorded for each ovary. Cyclic status (yes vs. no) was determined by the presence of an ultrasonically detected CL on either study day 0 or 14. Cows without a CL on study day 0 and study day 14 were considered to be noncyclic at study day 14. Furthermore, projected 305-d milk yield (P305M) was extracted from the on-farm record keeping software (DairyComp 305, Valley Agriculture Software, Tulare, CA) at enrollment (study day 0).
Figure 1.
Schematic of the experimental design. Cows were eligible to be inseminated after 53 days in milk.
In a subgroup of primiparous (n = 107) and multiparous (n = 197) cows, 100 µg i.m. GnRH (2 mL Factrel, Zoetis Inc., Florham Park, NJ) was administered at study day 14. Ovulation to GnRH (yes vs. no) was determined on study day 21 by the presence of a new CL in the same ovary where a previous follicle ≥ 10 mm was identified on study day 14. This subgroup consisted of cows randomly assigned to be submitted to the Gpresynch treatment in the previous study (Voelz et al., 2016).
Estrus detection was performed once daily and estrus was determined by removal of tail paint. All cows were eligible to be inseminated after 53 DIM. Primiparous (n = 178) and multiparous (n = 368) cows not inseminated by study day 21 (67 ± 3 or 56 ± 3 DIM, respectively), were administered 25 mg PGF2α i.m. (dinoprost tromethamine, 5 mL Lutalyse, Zoetis Inc.). Response to PGF2α (yes vs. no) was determined by the detection of estrus and insemination of cows within 14 d after receiving PGF2α.
Cows were further categorized by the presence of a CL or being inseminated by study days 14 (CL + AI14), 21 (CL + AI21), and 35 (CL + AI35). At each time point (study days 14, 21, and 35), cows bearing a CL or receiving AI were classified as a success (yes or no).
Statistical Analyses
Data for primiparous and multiparous cows were analyzed in separate models because of differences in DIM at enrollment. Descriptive statistics were performed using the FREQ procedure of SAS version 9.3 (SAS Institute Inc., Cary, NC) with Chi-square analysis. Continuous variables were analyzed using the GLM procedure of SAS. Dichotomous data were analyzed using the LOGISTIC procedure of SAS. For analysis of the ovulatory response to GnRH treatment, models included PVD (positive vs. negative), cyclic status (cyclic vs. noncyclic), BCS category (high vs. low), and the two-way interactions between PVD, cyclic status and BCS category. Models for evaluating odds of success at study day 14 (CL + AI14) included PVD, BCS category, and the interaction between PVD and BCS category. For analysis of being a success at study day 21 (CL + AI21) and day 35 (CL + AI35), models included PVD, GnRH treatment at study day 14 (yes vs. no), BCS category, and the two-way interactions between PVD, GnRH treatment at study day 14, and BCS category. Dairy was included as a fixed effect in all models. Variables were removed from the model when P > 0.10 using a backward elimination method. Dairy was forced to remain in the final model. Analysis of the Cox proportional hazard of insemination after PGF2α was performed using the PHREG procedure of SAS. Models included PVD, GnRH treatment at study day 14, cyclic status, BCS category, and dairy. Models for the hazard of insemination also included interactions between PVD and GnRH treatment at study day 14, PVD and BCS category, and PVD and cyclic status. Variables were removed using backward elimination based on the Wald criterion when P > 0.10 with dairy being forced to stay in the final model. Survival analysis for time to insemination after PGF2α was performed using the LIFETEST procedure of SAS. Statistical significance was determined as P ≤ 0.05 and statistical tendencies as 0.05 < P ≤ 0.10.
RESULTS
Three primiparous cows were sold between study day 21 to 35 and were removed from the CL + AI35 analysis. In addition, 24 primiparous and 19 multiparous cows were excluded from the analysis of ovarian response to GnRH treatment because ultrasonography was not performed on study day 21. These cows were cyclic according to the ultrasound examinations before study day 21; thus, they were included in the analysis for percentage of cows bearing a CL or inseminated by study days 14, 21, or 35.
Descriptive Results
Descriptive results are presented in Table 1. Overall prevalence of PVD was 28.5% and 13.4% for primiparous and multiparous cows, respectively. Prevalence of PVD in dairy A was 26.2% and 9.9%, whereas prevalence of PVD in dairy B was 30.3% and 18.2% for primiparous and multiparous cows, respectively. Proportion of multiparous PVD– cows cyclic at study day 14 was greater (P = 0.01) than multiparous PVD+ cows (Table 1). No difference (P = 0.24) was detected in proportion of cows cyclic at study day 14 between primiparous PVD+ and PVD– cows. Projected 305-d milk yield was less (P < 0.01) in PVD+ multiparous cows compared with PVD– multiparous cows (Table 1). In contrast, P305M did not (P = 0.26) differ between primiparous PVD+ and PVD– cows.
Table 1.
Projected 305-d milk yield (P305M), BCS and proportion of cows with BCS ≥ 3 at enrollment, and proportion of cows cyclic at study day 14 based on parity for cows with or without PVD
| Item | PVD | P-value | |
|---|---|---|---|
| Positive | Negative | ||
| Primiparous | n = 83 | n = 208 | |
| P305M, kg (± SE) | 7,808 ± 131 | 7,979 ± 80 | 0.26 |
| BCS (± SE) | 3.06 ± 0.04 | 3.06 ± 0.02 | 0.85 |
| BCS ≥ 3, % | 75.9 | 77.4 | 0.78 |
| Cyclic at study day 14, % | 75.3 | 81.5 | 0.24 |
| Multiparous | n = 54 | n = 348 | |
|---|---|---|---|
| P305M, kg (± SE) | 9,455 ± 200 | 9,994 ± 68 | <0.01 |
| BCS (± SE) | 2.87 ± 0.06 | 2.95 ± 0.02 | 0.16 |
| BCS ≥ 3, % | 57.4 | 58.3 | 0.90 |
| Cyclic at study day 14, % | 73.6 | 87.1 | 0.01 |
Ovarian Responses to GnRH Treatment
Ovulatory response to GnRH treatment was 51.8% and 47.8% for primiparous and multiparous cows, respectively. Primiparous PVD– cows tended (P = 0.06) to be less likely to ovulate after GnRH treatment than primiparous PVD+ cows (Table 2). In contrast, multiparous PVD+ cows had reduced (P = 0.04) ovulatory response to GnRH treatment than multiparous PVD– cows (Table 2). Both primiparous and multiparous noncyclic cows were more likely (P < 0.01) to ovulate to GnRH treatment than cyclic cows (Table 2). Primiparous cows with low BCS had decreased (P = 0.02) odds of ovulating after GnRH treatment compared with high BCS, whereas in multiparous cows, BCS was not associated (P = 0.96) with ovulation after GnRH (Table 2). Ovulatory response to GnRH did not differ (P > 0.64) between dairies. The two-way interactions between PVD, cyclic status, and BCS category were not significant (P > 0.10) for ovulatory response to GnRH treatment.
Table 2.
Adjusted odds of ovulation from study days 14 to 21 after treatment with GnRH for primiparous and multiparous cows positive or negative for PVD
| Item | Level | Ovulation, % (no./no.) | AOR3 (95% CI) | P-value |
|---|---|---|---|---|
| Primiparous | ||||
| PVD | Negative | 40.8 (20/49) | Referent | 0.06 |
| Positive | 67.6 (23/34) | 2.75 (0.95–7.93) | ||
| Cyclic Status1 | Cyclic | 37.5 (21/56) | Referent | <0.01 |
| Noncyclic | 81.5 (22/27) | 18.82 (3.76–94.13) | ||
| BCS2 | High | 57.6 (34/59) | Referent | 0.02 |
| Low | 37.5 (9/24) | 0.15 (0.03–0.74) | ||
| Dairy | A | 51.6 (16/31) | Referent | 0.95 |
| B | 51.9 (27/52) | 1.04 (0.35–3.06) | ||
| Multiparous | ||||
| PVD | negative | 49.7 (75/151) | Referent | 0.04 |
| positive | 37.0 (10/27) | 0.35 (0.13–0.95) | ||
| Cyclic Status1 | cyclic | 39.6 (57/144) | Referent | <0.01 |
| noncyclic | 82.4 (28/34) | 8.75 (3.17–24.15) | ||
| BCS2 | high | 45.9 (45/98) | Referent | 0.96 |
| low | 50.0 (40/80) | 1.02 (0.53–1.96) | ||
| Dairy | A | 44.1 (41/93) | Referent | 0.64 |
| B | 51.8 (44/85) | 1.17 (0.61–2.21) | ||
1Cows with a corpus luteum present on study day 0 or 14 were classified as cyclic.
2Body condition score category: high (≥3.00) or low (<3.00).
3Adjusted odds ratio.
Percentage of Cows Bearing a CL or Inseminated by Study days 14 (CL + AI14), 21 (CL + AI21), or 35 (CL + AI35)
For primiparous cows, CL + AI14 was not (P ≥ 0.28) associated with PVD or dairy, however, was associated (P < 0.01) with BCS (Table 3). Primiparous cows with high BCS were (P < 0.01) more likely to be classified as a success on study day 14 than low BCS cows (85.3% vs. 62.7%; Table 3). For multiparous cows, CL + AI14 was associated (P ≤ 0.03) with PVD and BCS. Multiparous PVD+ cows were (P = 0.03) less likely to be classified as a success on study day 14 than PVD– cows (Table 3). Furthermore, multiparous cows with low BCS had decreased (P < 0.01) odds of being classified a success on study day 14 than high BCS cows (Table 3). Dairy was associated (P < 0.01) with CL + AI14 as multiparous cows from dairy A were (P < 0.01) more likely to be classified a success than multiparous cows from dairy B (90.1% vs. 79.4%; Table 3). The interaction between PVD and BCS category was not significant (P > 0.10) for CL + AI14.
Table 3.
Adjusted odds of bearing a corpus luteum or being inseminated by study day 14 for primiparous and multiparous cows positive or negative for PVD
| Item | Level | Success, % (no./no.) | AOR2 (95% CI) | P-value |
|---|---|---|---|---|
| Primiparous | ||||
| PVD | Negative | 81.7 (170/208) | Referent | 0.28 |
| Positive | 75.9 (63/83) | 0.70 (0.37–1.32) | ||
| BCS1 | High | 85.3 (191/224) | Referent | <0.01 |
| Low | 62.7 (42/67) | 0.29 (0.16–0.55) | ||
| Dairy | A | 78.6 (99/126) | Referent | 0.87 |
| B | 81.2 (134/165) | 1.05 (0.57–1.91) | ||
| Multiparous | ||||
| PVD | Negative | 87.4 (304/348) | Referent | 0.03 |
| Positive | 74.1 (40/54) | 0.46 (0.22–0.94) | ||
| BCS1 | High | 90.6 (212/234) | Referent | <0.01 |
| Low | 78.6 (132/168) | 0.35 (0.19–0.62) | ||
| Dairy | A | 90.1 (209/232) | Referent | <0.01 |
| B | 79.4 (135/170) | 0.41 (0.23–0.74) | ||
1Body condition score category: high (≥3.00) or low (<3.00).
2Adjusted odds ratio.
Among primiparous cows, CL + AI21 was not (P ≥ 0.14) associated with PVD, treatment with GnRH at study day 14, or dairy (Table 4). BCS was associated (P < 0.01) with CL + AI21. Similar to study day 14, cows with high BCS were more likely to be classified a success on study day 21 compared with low BCS cows (94.6% vs. 79.1%; Table 4). For multiparous cows, CL + AI21 was associated (P ≤ 0.01) with PVD, treatment with GnRH at study day 14, and BCS. Low BCS and PVD+ cows were (P ≤ 0.01) less likely to be classified as a success on study day 21 compared with high BCS and PVD– cows, respectively (Table 4). Multiparous cows treated with GnRH on study day 14 had increased (P < 0.01) odds of being classified as a success on study day 21 than multiparous cows not treated with GnRH (Table 4). Dairy tended (P = 0.10) to be associated with CL + AI21 for multiparous cows (Table 4). Two-way interactions between PVD, GnRH treatment at study day 14, and BCS category were not significant (P > 0.10) for CL + AI21 for primiparous and multiparous cows.
Table 4.
Adjusted odds of bearing a corpus luteum or being inseminated by study day 21 for primiparous and multiparous cows with or without PVD
| Item | Level | Success, % (no./no.) | AOR2 (95% CI) | P-value |
|---|---|---|---|---|
| Primiparous | ||||
| PVD | Negative | 91.8 (191/208) | Referent | 0.40 |
| Positive | 89.2 (74/83) | 0.68 (0.28–1.67) | ||
| GnRH | No | 89.7 (165/184) | Referent | 0.14 |
| Yes | 93.5 (100/107) | 2.00 (0.79–5.10) | ||
| BCS1 | High | 94.6 (212/224) | Referent | <0.01 |
| Low | 79.1 (53/67) | 0.21 (0.09–0.48) | ||
| Dairy | A | 91.3 (115/126) | Referent | 0.61 |
| B | 90.9 (150/165) | 0.80 (0.35–1.87) | ||
| Multiparous | ||||
| PVD | Negative | 97.4 (339/348) | Referent | <0.01 |
| Positive | 87.0 (47/54) | 0.17 (0.06–0.51) | ||
| GnRH | No | 93.7 (192/205) | Referent | <0.01 |
| Yes | 98.5 (194/197) | 5.81 (1.54–22.22) | ||
| BCS1 | High | 97.9 (229/234) | Referent | 0.01 |
| Low | 93.5 (157/168) | 0.23 (0.07–0.73) | ||
| Dairy | A | 97.4 (226/232) | Referent | 0.10 |
| B | 94.1 (160/170) | 0.40 (0.13–1.20) | ||
1Body condition score category: high (≥3.00) or low (<3.00).
2Adjusted odds ratio.
For primiparous cows, CL + AI35 was not associated (P = 0.29) with PVD (Table 5). In contrast, GnRH and BCS were associated (P ≤ 0.03) with CL + AI35 for primiparous cows (Table 5). Primiparous cows treated with GnRH or with high BCS were (P < 0.04) more likely to be a success on study day 35 than primiparous cows not treated with GnRH or with low BCS. Dairy tended to be associated with CL + AI35 for primiparous cows (Table 5). For multiparous cows, CL + AI35 was not associated (P ≥ 0.11) with PVD or dairy (Table 5). Treatment with GnRH at study day 14 and BCS were associated (P ≤ 0.03) with CL + AI35 for multiparous cows (Table 5). Two-way interactions between PVD, GnRH treatment at study day 14, and BCS category were not significant (P > 0.10) for CL + AI35 for primiparous and multiparous cows.
Table 5.
Adjusted odds of bearing a corpus luteum or being inseminated by study day 35 for primiparous and multiparous cows with or without PVD
| Item | Level | Success, % (no./no.) | AOR2 (95% CI) | P-value |
|---|---|---|---|---|
| Primiparous | ||||
| PVD | Negative | 95.2 (197/207) | Referent | 0.29 |
| Positive | 92.6 (75/81) | 0.55 (0.18–1.66) | ||
| GnRH | No | 92.4 (170/184) | Referent | 0.03 |
| Yes | 98.1 (102/104) | 5.43 (1.18–24.39) | ||
| BCS1 | High | 96.4 (215/223) | Referent | <0.01 |
| Low | 87.7 (57/65) | 0.18 (0.08–0.66) | ||
| Dairy | A | 96.8 (119/123) | Referent | 0.08 |
| B | 92.7 (153/165) | 0.35 (0.10–1.14) | ||
| Multiparous | ||||
| PVD | Negative | 98.3 (342/348) | Referent | 0.11 |
| Positive | 94.4 (51/54) | 0.29 (0.07–1.32) | ||
| GnRH | No | 96.1 (197/205) | Referent | 0.03 |
| Yes | 99.5 (196/197) | 10.42 (1.25–90.91) | ||
| BCS1 | High | 99.2 (232/234) | Referent | 0.02 |
| Low | 95.8 (161/168) | 0.15 (0.03–0.78) | ||
| Dairy | A | 98.7 (229/232) | Referent | 0.12 |
| B | 96.5 (164/170) | 0.31 (0.07–1.33) | ||
1Body condition score category: high (≥3.00) or low (<3.00).
2Adjusted odds ratio.
Hazard of Insemination After PGF2α
For primiparous cows, hazard of insemination after PGF2α was not (P = 0.38) associated with PVD. Hazard of insemination after PGF2α tended (P = 0.06) to be associated with cyclic status because noncyclic cows tended to be inseminated at a slower rate (adjusted hazard ratio [AHR] = 0.65, 95% CI = 0.41 to 1.03) than cyclic cows. Furthermore, BCS, treatment with GnRH, dairy, and all interactions were not (P > 0.38) associated with hazard of insemination after PGF2α for primiparous cows. The survival analyses for primiparous cows revealed that PVD was not (P = 0.40) associated with the interval from PGF2α to insemination (Fig. 2A). Median and mean (± SE) days to insemination after PGF2α were 4 and 7.9 ± 0.5 d for PVD– primiparous cows, and 11 and 8.6 ± 0.7 d for PVD+ primiparous cows. Interactions between PVD and GnRH treatment at study day 14, PVD and BCS category, and PVD and cyclic status were not significant (P > 0.10) for hazard of insemination of primiparous cows.
Figure 2.
Survival analysis for days to insemination for (A) primiparous and (B) multiparous cows positive (PVD+) or negative (PVD–) for purulent vaginal discharge. Median days to insemination of primiparous cows were 11 and 4 d for PVD+ or PVD– cows, respectively (P = 0.40). Median days to insemination of multiparous cows were 10 and 6 d for PVD+ or PVD– cows, respectively (P = 0.37).
Hazard of insemination after PGF2α was not (P = 0.76) associated with PVD for multiparous cows. In addition, treatment with GnRH, BCS, and all interactions were not (P > 0.15) associated with hazard of insemination. Hazard of insemination after PGF2α for multiparous cows was associated (P = 0.02) with cyclic status. Noncyclic multiparous cows were inseminated at a slower rate (AHR = 0.58, CI = 0.37 to 0.90) than cyclic cows. The survival analyses for multiparous cows revealed that PVD was not (P = 0.37) associated with the interval from PGF2α to insemination (Fig. 2B). Median and mean (± SE) days to insemination after PGF2α were 6 and 8.2 ± 0.3 d for PVD– multiparous cows, and 10 and 6.7 ± 0.5 d for PVD+ multiparous cows. Interactions between PVD and GnRH treatment at study day 14, PVD and BCS category, and PVD and cyclic status were not significant (P > 0.10) for hazard of insemination of multiparous cows.
DISCUSSION
Previous studies demonstrated a negative association between PVD and reproductive efficiency in dairy cattle (LeBlanc et al., 2002; McDougall et al., 2007; Dubuc et al., 2010). Decreased reproductive efficiency of cows with PVD can be a result of several factors, including delayed resumption of cyclicity (Galvão et al., 2010), decreased pregnancy per AI (LeBlanc et al., 2002; McDougall et al., 2007), and altered uterine microbiota (Bicalho et al., 2017). Although several reports demonstrated a negative effect of PVD on fertility of dairy cows, there is a paucity of evidence of PVD influencing ovarian responses to GnRH or PGF2α treatments. The current study provides evidence that PVD in multiparous cows is negatively associated with ovarian responses to hormonal treatments, which may potentially influence success of reproductive management programs. In the United States, ovulation synchronization programs are dependent on exogenous GnRH and PGF2α to successfully synchronize ovulation for timed AI. Ovulation response to the initial GnRH of the Ovsynch protocol is important for synchronizing ovulation at the time of AI and improving fertility (Bello et al., 2006). The current study raises concern that PVD might decrease synchronization of multiparous cows submitted to reproductive programs by decreasing the ovulatory response to GnRH in multiparous cows, but not in primiparous cows.
In dairy cattle, PVD is associated with bacterial infection of the uterus, with the most common pathogens being Escherichia coli and Arcanobacterium pyogenes (Williams et al., 2005). Increased bacterial load of the uterus can result in the presence of lipopolysaccharide (LPS) in follicular fluid (Herath et al., 2007). In an in vitro study, Herath et al. (2007) demonstrated that exposure of granulosa cells to LPS caused a decrease in mRNA expression of aromatase and decreased production of estradiol from granulosa cells isolated from medium (4 to 8 mm) and larger follicles (>8 mm). Decreased concentrations of estradiol because of LPS exposure could attenuate the release or block the LH surge; therefore, preventing ovulation. Infusion of LPS into the uterus has been shown to suppress the preovulatory surge of LH in estrus-synchronized Holstein heifers (Peter et al., 1989). Peter et al. (1989) suggested that an increased concentration of cortisol in LPS-infused heifers resulted in suppression of LH secretion, however, a direct effect on the ovary and follicle cannot be ignored (Herath et al., 2007). Administration of GnRH during ovulation- synchronization programs causes release of LH in all cows (Stevenson and Pulley, 2016); however, magnitude of LH release is dependent on the concentration of progesterone (Giordano et al., 2012; Pulley et al., 2015; Stevenson and Pulley, 2016). Of particular interest, ewes infused with LPS had greater concentration of progesterone when compared with noninfused controls (Battaglia et al., 1997, 1999, 2000). The observed LPS-induced increase in concentration of progesterone was attributed to adrenal secretion (Battaglia et al., 1997). Therefore, the association between GnRH-induced ovulatory responses and PVD could be related to direct effects of PVD on the follicle, or differences in concentrations of progesterone and cortisol. The finding that primiparous PVD+ cows were more likely to ovulate to GnRH than primiparous PVD– cows is bewildering, especially because the interaction between PVD and cyclic status was not significant. Distinct hormonal profiles of primiparous and multiparous cows may be the main reason for differences in GnRH-induced ovulatory responses of PVD+ and PVD– cows. Indeed, in the previous study, concentration of progesterone was not similar between primiparous and multiparous cows (Voelz et al., 2016). Nevertheless, it is unknown whether differences exist in concentrations of progesterone between primiparous and multiparous PVD+ cows.
In addition to demonstrating that multiparous PVD+ cows have decreased GnRH-induced ovulatory responses, the current study presents evidence that multiparous PVD+ cows have decreased overall ovulatory response, regardless of treatment with GnRH. Multiparous PVD+ cows had lesser percentage of cows bearing a CL or inseminated by study days 14, 21, and 35, when controlling for BCS. It is important to note that these time points (study days 14, 21, and 35) may coincide with the end of the voluntary period in some dairy herds. Decreased ovulatory responses may influence DIM at first AI and pregnancy outcomes, thus impacting reproductive efficiency of dairy herds. These findings indicate that PVD in multiparous cows may have carryover effects on ovulatory responses. Ribeiro et al. (2016) demonstrated evidence that uterine disease has carryover effects on fertility by altering uterine environment. Indeed, the negative impact of uterine disease on fertility is independent of anovular condition or low BCS after calving (Ribeiro et al., 2016). Collectively, results from the present study and those reported by Ribeiro et al. (2016) demonstrate that multiparous cows diagnosed with uterine disease have decreased ovulatory responses and fertility, respectively, regardless of cyclic status or BCS. Furthermore, these data demonstrate uterine disease has a prolonged impact in the reproductive tract because of the observed negative effects in the ovary and uterus several days after diagnosis of uterine disease. Nevertheless, it is unclear why ovarian responses differed in PVD+ and PVD– cows between parities. Similar rates of success were recorded in primiparous PVD+ cows, based on presence of a CL or insemination, compared with PVD– cows at study days 14, 21, and 35.
Although PVD was determined and categorized at greater DIM for primiparous than multiparous cows, prevalence of PVD was greater in primiparous than multiparous cows. Multiparous cows are more likely to have metabolic diseases, such as hyperketonemia and hypocalcemia, compared with primiparous cows (Markusfeld, 1987). Despite PVD being less prevalent in multiparous cows, uterine disease and other disorders might have an additive effect in altering ovarian responses. In fact, Ribeiro et al. (2016) demonstrated an additive negative effect of uterine and nonuterine disease on pregnancy per AI.
In the present study, PVD was associated with decreased P305M in multiparous cows. In contrast, no differences were detected in P305M between PVD+ and PVD– primiparous cows. Other studies have not observed an effect of PVD on milk production (Fourichon et al., 1999; Dubuc et al., 2011). It is possible that multiparous PVD+ cows from the present study experienced other disorders besides uterine disease. It is important to consider that other disorders may have confounded the association between PVD and outcomes evaluated in this experiment. Unfortunately, the authors did not evaluate cows for immediate postpartum disorders in this study, which limits the possibility of exploring potential confounders. Nonetheless, it is still unknown whether early postpartum diseases are associated with ovarian responses to treatment with GnRH and PGF2α. The hypothesis that PVD affects multiparous cows more than primiparous cows is interesting, however, one should be cautious in drawing any inferences from the present study because of differences in DIM at PVD diagnosis. Cows used in this study were part of a previous experiment (Voelz et al., 2016), which DIM at initiation of the reproductive programs differed between parities. Because the authors of the current study intended to maintain a similar interval between evaluation of PVD and the first injection of the reproductive protocol, DIM at PVD diagnosis differed between parities. Therefore, separate statistical analyses by parity were required to avoid the confounding of differences in DIM at diagnosis of PVD. Further research is needed to evaluate the effects of PVD on ovarian responses of primiparous and multiparous diagnosed with PVD at similar DIM.
Peripheral concentration of prostaglandin E2 (PGE) has been shown to be increased in cows with bacterial infection of the uterus during the first 2 wk postpartum (Herath et al., 2009). Using an in vitro model, Herath et al. (2009) also demonstrated a shift from production of PGF2α to PGE by epithelial cells of the bovine endometrium treated with LPS. In contrast, repeated intrauterine infusion of LPS in heifers did not modulate mRNA expression of PGE synthase in the endometrium of the uterus on day 6 of the estrous cycle, however, mRNA expression of PGE synthase was increased in day 6 CL compared with controls (Lüttgenau et al., 2016). Although these studies evaluated the effects of LPS on prostaglandin secretion and mRNA modulation of the endometrium, limited experiments exist that have evaluated responses to exogenous PGF2α in PVD+ cows (Kaufmann et al., 2010; McDougall et al., 2013). Even though the current study did not detect differences in hazard of insemination between cows with or without PVD in response to PGF2α, caution should be considered when interpreting these results. One limitation of the current study was the extended time (21 d) between classification of PVD and administration of PGF2α. Endometrial inflammation or decreased bacterial load of the uterus in PVD+ cows may have partly resolved before PGF2α administration. Furthermore, the small sample size of cows with PVD receiving PGF2α also should be considered. A large proportion of cows were inseminated before PGF2α treatment, resulting in a small number of cows treated with PGF2α. Experiments with a shorter interval between PVD diagnosis and PGF2α treatment with an adequate sample size are required to fully elucidate the role of PVD on ovarian responses to PGF2α.
In conclusion, purulent vaginal discharge is associated with GnRH-induced ovulatory response and spontaneous ovulation in dairy cattle. Purulent vaginal discharge in multiparous cows had a more pronounced detrimental effect on ovarian responses than in primiparous cows. Dairy herds with a high prevalence of PVD, especially in multiparous cows, might observe decreased reproductive efficiency and responses to reproductive management programs. Response to PGF2α was not associated with PVD in primiparous or multiparous cows, however, more research is required to replicate and verify these findiwngs.
Conflict of interest statement: None declared.
Footnotes
This research was partially funded by Zoetis Animal Health (Florham Park, NJ) and the Kansas Dairy Commission (Hays, KS).
The authors thank the owners and managers of the collaborating dairies, Gary Neubauer of Zoetis Animal Health, and Robert Briggeman of ABS Global (DeForest, WI).
LITERATURE CITED
- Battaglia D. F., Bowen J. M., Krasa H. B., Thrun L. A., Viguié C., and Karsch F. J.. 1997. Endotoxin inhibits the reproductive neuroendocrine axis while stimulating adrenal steroids: a simultaneous view from the hypophyseal portal and peripheral blood. Endocrinol. 138:4273–4281. [DOI] [PubMed] [Google Scholar]
- Battaglia D. F., Beaver A. B., Harris T. G., Tanhehco E., Viguié C., and Karsch F. J.. 1999. Endotoxin disrupts the estradiol-induced luteinizing hormone surge: interference with estradiol signal reading, not surge release. Endocrinol. 140:2471–2479. [DOI] [PubMed] [Google Scholar]
- Battaglia D. F., Krasa H. B., Padmanabhan V., Viguié C., and Karsch F. J.. 2000. Endocrine alterations that underlie endotoxin-induced disruption of the follicular phase in ewes. Biol. Reprod. 62:45–53. doi:10.1095/biolreprod62.1.45 [DOI] [PubMed] [Google Scholar]
- Bello N. M., Steibel J. P., and Pursley J. R.. 2006. Optimizing ovulation to first GnRH improved outcomes to each hormonal injection of Ovsynch in lactating dairy cows. J. Dairy Sci. 89:3413–3424. doi:10.3168/jds.S0022-0302(06)72378-5 [DOI] [PubMed] [Google Scholar]
- Bicalho M. L. S., Lima S., Higgins C. H., Machado V. S., Lima F. S., and Bicalho R. C.. 2017. Genetic and functional analysis of the bovine uterine microbiota. Part II: purulent vaginal discharge versus healthy cows. J. Dairy Sci. 100:3863–3874. doi:10.3168/jds.2016–12061 [DOI] [PubMed] [Google Scholar]
- Dubuc J., Dufffield T. F., Leslie K. E., Walton J. S., and LeBlanc S. J.. 2010. Definition and diagnosis of postpartum endometritis in dairy cows. J. Dairy Sci. 93:5225–5233. doi:10.3168/jds.2010–3428 [DOI] [PubMed] [Google Scholar]
- Dubuc J., Duffield T. F., Leslie K. E., Walton J. S., and LeBlanc S. J.. 2011. Effects of uterine diseases on milk production and culling in dairy cows. J. Dairy Sci. 94:1339–1346. doi:10.3168/jds.2010–3758 [DOI] [PubMed] [Google Scholar]
- Ferguson J. D., Galligan D. T., and Thomsen N.. 1994. Principal descriptors of body condition score in Holstein cows. J. Dairy Sci. 77:2695–2703. [DOI] [PubMed] [Google Scholar]
- Fourichon C., Seegers H., Bareille N., and Beaudeau F.. 1999. Effects of disease on milk production in the dairy cow: a review. Prev. Vet. Med. 41:1–35. [DOI] [PubMed] [Google Scholar]
- Galvão K. N., Frajblat M., Butler W. R., Brittin S. B., Guard C. L., and Gilbert R. O.. 2010. Effect of early postpartum ovulation on fertility in dairy cows. Reprod. Dom. Anim. 45:e207–e211. doi:10.1111/j.1439-0531.2009.01517.x [DOI] [PubMed] [Google Scholar]
- Giordano J. O., Fricke P. M., Guenther J. N., Lopes G. Jr., Herlihy M. M., Nascimento A. B., and Wiltbank M. C.. 2012. Effect of progesterone on magnitude of the luteinizing hormone surge induced by 2 different doses of gonadotropin-releasing hormone in lactating dairy cows. J. Dairy Sci. 95:3781–3793. doi:10.3168/jds.2011–5155 [DOI] [PubMed] [Google Scholar]
- Herath S., Williams E. J., Lilly S. T., Gilbert R. O., Dobson H., Bryant C. E., and Sheldon I. M.. 2007. Ovarian follicular cells have innate immune capabilities that modulate their endocrine function. Reprod. 134:683–693. doi:10.1530/REP-07-0229 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Herath S., Lily S. T., Fischer D. P., Williams E. J., Dobson H., Bryant C. E., and Sheldon I. M.. 2009. Bacterial lipopolysaccharide induces an endocrine switch from prostaglandin F2α to prostaglandin E2 in bovine endometrium. Endocrinol. 150:1912–1920. doi:10.1210/en.2008-1379 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kaufmann T. B., Westermann S., Drillich M., Plöntzke J., and Heuwieser W.. 2010. Systemic antibiotic treatment of clinical endometritis in dairy cows with ceftiofur or two doses of cloprostenol in a 14-d interval. Anim. Reprod. Sci. 121:55–62. doi:10.1016/ j.anireprosci.2010.04.190 [DOI] [PubMed] [Google Scholar]
- LeBlanc S. J., Duffield T. F., Leslie K. E., Bateman K. G., Keefe G. P., Walton J. S., and Johnson H.. 2002. Defining and diagnosing postpartum clinical endometritis and its impact on reproductive performance in dairy cows. J. Dairy Sci. 85:2223–2236. [DOI] [PubMed] [Google Scholar]
- Lüttgenau J., Lingemann B., Wellnitz B., Hankele A. K., Schmicke M., Ulbrich S. E., Bruckmaier R. M., and Bollwein H.. 2016. Repeated intrauterine infusion of lipopolysaccharide alter gene expression and lifespan of the bovine corpus luteum. J. Dairy Sci. 99:6639–6653. doi:10.3168/jds.2015–10806 [DOI] [PubMed] [Google Scholar]
- Markusfeld O. 1987. Periparturient traits in seven high dairy herds. incidence rates, association with parity, and interrelationships among traits. J. Dairy Sci. 70:158–166. [DOI] [PubMed] [Google Scholar]
- Mateus L., Lopes da Costa L., Bernardo F., and Silva J. R.. 2002. Influence of puerperal uterine infection on uterine involution and postpartum ovarian activity in dairy cows. Reprod. Dom. Anim. 37:31–35. [DOI] [PubMed] [Google Scholar]
- McDougall S., Macaulay R., and Compton C.. 2007. Association between endometritis diagnosis using a novel intravaginal device and reproductive performance in dairy cattle. Anim. Reprod. Sci. 99:9–23. doi:10.1016/j.anireprosci.2006.03.017 [DOI] [PubMed] [Google Scholar]
- McDougall S., de Boer M., Compton C., and LeBlanc S. J.. 2013. Clinical trial of treatment programs for purulent vaginal discharge in lactating dairy cattle in New Zealand. Theriogenology 79:1139–1145. doi:10.1016/j.theriogenology.2013.02.002 [DOI] [PubMed] [Google Scholar]
- Peter A. T., Bosu W. T. K., and DeDecker R. J.. 1989. Suppression of preovulatory luteinizing hormone surges in heifers after intrauterine infusions of Escherichia coli endotoxin. Am. J. Vet. Res. 50:368–373. [PubMed] [Google Scholar]
- Pulley S. L., Keisler D. H., and Stevenson J. S.. 2015. Concentrations of luteinizing hormone and ovulatory responses in dairy cows before timed artificial insemination. J. Dairy Sci. 98:6188–6201. doi:10.3168/jds.2015–9473 [DOI] [PubMed] [Google Scholar]
- Ribeiro E. S., Gomes G., Greco L. F., Cerri R. L. A., Vieira-Neto A., Monteiro P. L. J. Jr., Lima F. S., Bisinotto R. S., Thatcher W. W., and Santos J. E. P.. 2016. Carryover effects of postpartum inflammatory diseases on developmental biology and fertility in lactating dairy cows. J. Dairy Sci. 99:2201–2220. doi:10.3168/jds.2015–10337 [DOI] [PubMed] [Google Scholar]
- Sheldon I. M., Noakes D. E., Rycroft A. N., Pfeiffer D. U., and Dobson H.. 2002. Influence of uterine bacterial contamination after parturition on ovarian dominant follicle selection and follicle growth and function in cattle. Reprod. 123:837–845. [PubMed] [Google Scholar]
- Stevenson J. S., and Pulley S. L.. 2016. Feedback effects of estradiol and progesterone on ovulation and fertility of dairy cows after gonadotropin-releasing hormone- induced release of luteinizing hormone. J. Dairy Sci. 99:3003–3015. doi:10.3168/jds.2015–10091 [DOI] [PubMed] [Google Scholar]
- Voelz B. E., Rocha L., Scortegagna F., Stevenson J. S., and Mendonça L. G. D.. 2016. Treatment of lactating dairy cows with gonadotropin-releasing hormone before first insemination during summer heat stress. J. Dairy Sci. 99:7612–7623. doi:10.3168/jds.2016–10970 [DOI] [PubMed] [Google Scholar]
- Williams E. J., Fischer D. P., Pfeiffer D. U., England G. C. W., Noakes D. E., Dobson H., and Sheldon I. M.. 2005. Clinical evaluation of postpartum vaginal mucus reflects uterine bacterial infection and the immune response in cattle. Theriogenology 63:102–117. doi:10.1016/j.theriogenology.2004.03.017. [DOI] [PubMed] [Google Scholar]