Publikationsdatenbank

Optimization of reproductive management of lactating dairy cows on farms using automated activity monitoring systems for estrus detection or protocols for timed artificial insemination (2021)

Art

Hochschulschrift

Autor

Tippenhauer, Christie Marie (WE 19)

Quelle

Berlin, 2021 — 128 Seiten

Sprache

Englisch

Verweise

URL (Volltext): https://refubium.fu-berlin.de/handle/fub188/31421

Kontakt

Tierklinik für Fortpflanzung

Königsweg 65
Haus 27
14163 Berlin
+49 30 838 62618
fortpflanzungsklinik@vetmed.fu-berlin.de

Abstract / Zusammenfassung

The overall objectives of this thesis were 1) to determine the association between the timing of artificial insemination (AI) and different characteristics of an estrus event (i.e., onset, peak, and end) using an automated activity monitoring (AAM) system on pregnancy per AI (P/AI) in lactating Holstein cows inseminated with either fresh or frozen semen, 2) to identify plausible factors associated with estrous expression and subsequent P/AI in lactating Holstein cows using an AAM system, and 3) to evaluate two modifications (i.e., double dose, second administration 24 h apart) of the prostaglandin F2α (PGF) treatment using cloprostenol in a 7-d Ovsynch protocol compared with a standard single PGF dose on progesterone (P4) concentration at the time of the second GnRH treatment (G2) and P/AI in lactating Holstein cows. To determine the association between the timing of AI with either fresh or frozen semen and different characteristics of an estrus event (i.e., onset, peak, and end) using an AAM system on P/AI, an observational study was conducted on 4 commercial dairy farms in Germany. The evaluation included a total of 3,607 AI services based on the alert of a neck-mounted AAM system (Heatime; SCR Engineers Ltd., Netanya, Israel). Pregnancy diagnosis was performed following farm individual strategies either by transrectal palpation 38 ± 3 d after AI or by transrectal ultrasonography 30 ± 3 d after AI. To evaluate whether cows were exposed to heat stress, loggers were installed within the barns that continuously recorded ambient temperature and relative humidity, thereby calculating the temperature-humidity-index (THI). The software of the AAM system converted individual cow’s raw activity data into an activity change index value using a proprietary algorithm. These index values ranged from 0 (lowest) to 100 (highest). Onset of estrus (OE) was defined as a cow exceeding an activity change index value of 35. End of estrus (EE) was defined by the first instance at which the index value fell below 35 again. The intensity of an estrus event was represented by the peak of the activity change index value (PAE) during an estrus event. Furthermore, PAE was classified into low intensity (35 to 89 index value) and high intensity (90 to 100 index value). Duration of an estrus event (DE) was defined as the interval from OE to EE. The mean (± standard deviation) DE was 14.3 ± 4.6 h. The mean interval from OE to AI was 16.8 ± 8.0 h, from PAE to AI was 11.9 ± 8.1 h, and from EE to AI was 2.5 ± 8.7 h. Primiparous cows (31.4%) had greater P/AI than multiparous cows (27.3%) and first postpartum AI (31.5%) yielded greater P/AI than subsequent AI services (27.2%). Heat stress decreased P/AI, such that exposure to a low THI of < 60 (36.1%) resulted in greater P/AI than exposure to a medium THI of 60 to 68 (28.6%) or a high THI of > 68 (23.9%) 1 wk before AI. For each farm, cumulative milk yield within the first 100 days in milk (DIM) was classified into quartiles. Cows with either low (29.8%) or high (31.6%) 100-d milk yield had greater P/AI than cows with intermediate (25.7%) 100-d milk yield. Estrous intensity was associated with P/AI. Cows with low PAE (26.0%) were less fertile compared with cows showing high PAE (32.8%). Type of semen was not associated with P/AI, as cows inseminated with frozen semen (28.8%) achieved similar P/AI compared with cows inseminated with fresh semen (29.8%). There was a quadratic effect of the interval from OE to AI on P/AI. Pregnancy per AI was greatest for cows inseminated from 7 to 24 h after OE. We did not observe an interaction between the interval from OE to AI and type of semen. Furthermore, greatest P/AI was achieved for cows inseminated within 18 h after PAE or from 5 h before EE to 12 h after EE. There tended to be an interaction between type of semen and the intervals from PAE to AI and from EE to AI. Cows inseminated with fresh semen within 5 h before EE had greater P/AI than cows inseminated with frozen semen. In addition, cows inseminated with frozen semen from 13 to 18 h after EE had greater P/AI than cows inseminated with fresh semen. Results from this study allow one to draw the conclusion that inseminating cows from 7 to 24 h after OE or within 18 h after PAE yields greatest P/AI irrespective of type of semen. In addition, high PAE was favourable for P/AI. As only two-thirds of cows expressed high PAE, further research is warranted to assess factors associated with suboptimum estrous intensity and the underlying physiology. The second study aimed to identify factors associated with estrous expression (i.e., DE and PAE) and subsequent fertility in lactating Holstein cows using the AAM system Heatime. Therefore, a total of 5,933 estrus events representing 3,132 lactating Holstein cows located on 8 commercial dairy farms in Germany were evaluated. All farms participated in monthly dairy herd improvement association’s (DHIA) testing. To calculate the THI, climate loggers installed within the barns recorded ambient temperature and relative humidity hourly. Depending on individual farm management, pregnancy diagnosis was performed either by transrectal palpation ranging from 35 to 46 d after AI or transrectal ultrasonography at 30 ± 3 d after AI. In agreement with the first study, the same categories were used to categorize PAE (i.e., low = index values from 35 to 89; high = index values from 90 to 100). Overall, the mean (± standard error of the mean) DE was 14.94 ± 0.06 h. More than two-thirds of estrus events (73.5%) were of high PAE. Because we found a strong correlation (r = 0.67) between DE and PAE, only PAE was used in the final statistical model as a fixed effect. Cows with high PAE (28.5%) had 1.35 greater odds of pregnancy compared with cows with low PAE (22.8%). Heat stress indicated by an increased THI 1 wk before AI was associated with shorter DE, lower PAE and subsequently decreased P/AI. The change of weighted rumination measured by the AAM system was associated with estrous expression, as a lower nadir was associated with a greater risk for high PAE and long DE. There was no association, however, between the nadir of change of weighted rumination and P/AI. The AAM system detected a small percentage of cows (4.7%) with short inter-estrus intervals (i.e., more than one activity peak within 7 d close to the event of estrus). Cows with a short inter-estrus interval had reduced DE and PAE and showed decreased P/AI (19.5%) compared with cows with no short inter-estrus interval (32.8%). Furthermore, primiparous cows (29.4%) had greater P/AI than multiparous cows (22.1%). Interestingly, PAE did not differ among parities, but multiparous cows expressed estrus with longer DE (13.15 ± 0.31 h) than primiparous cows (12.52 ± 0.32 h). An estrus event with long DE or high PAE was more likely later in lactation. Dairy herd improvement association’s test data from the date closest to AI was associated with estrous expression or P/AI in different ways. We did not find an association between milk fat, milk urea nitrogen, or lactose with estrous expression or P/AI. There was an association, however, between milk protein, milk yield, and somatic cell count from DHIA test date before AI and estrous expression, but not P/AI. Increasing milk protein increased DE and the likelihood for an estrus event with high PAE, whereas an increase in milk yield and somatic cell count (especially > 1,000 x 103 cells/mL) was negatively associated with estrous expression. In summary, DE and PAE were highly correlated, and particularly cows with high PAE were associated with greater odds for pregnancy. Whereas risk factors, such as DIM at AI, THI 1 wk before AI, and short inter-estrus intervals were associated with both estrous expression and P/AI, change of rumination, udder health, and milk yield were only associated with estrous expression but not with subsequent P/AI. The objective of the third study was to evaluate two modifications (i.e., double dose, second administration 24 h apart) of the PGF treatment using cloprostenol in a 7-d Ovsynch protocol compared with a standard single PGF dose on P4 concentration at the time of G2 and P/AI in lactating Holstein cows. This study was subdivided into 2 experiments with a total of 8 participating commercial dairy farms in Germany. Cows from both experiments were assigned on a weekly basis in a consecutive manner to receive 1 of the 3 treatments: 1) a traditional Ovsynch including a standard single PGF dose of 500 µg cloprostenol (control: GnRH; 7 d, PGF; 9 d, GnRH), 2) an Ovsynch with a double PGF dose of 1,000 µg cloprostenol (GDPG: GnRH; 7 d, 2xPGF; 9 d, GnRH), or 3) an Ovsynch with a second PGF treatment of 500 µg cloprostenol 24 h later (GPPG: GnRH; 7 d, PGF; 8 d, PGF; 32 h, GnRH). All cows received timed AI (TAI) approximately 16 h after G2. Pregnancy diagnosis was performed either by transrectal palpation at 38 ± 3 d after TAI (experiment 1) or transrectal ultrasonography at 35 ± 7 d after TAI (experiment 2). To facilitate first postpartum TAI, farms from experiment 1 used a Presynch-Ovsynch protocol (PGF, 14 d later PGF, 12 d later GnRH, 7 d later PGF, 2 d later GnRH, and 16 to 18 h later TAI), whereas no presynchronization protocol was used on farms from experiment 2. A total of 1,581 lactating dairy cows (60 experimental units) from 2 dairy farms were enrolled in experiment 1. Blood samples were collected from a subsample of these cows (n = 491; 16 experimental units) at the time of G2 to determine P4 concentration. In experiment 2, 1,979 lactating dairy cows (252 experimental units) from 6 dairy farms were enrolled. Transrectal ultrasonography was performed in all of the latter cows to determine the presence or absence of a corpus luteum (CL) at the time of the first GnRH treatment (G1). In experiment 1, treatment affected P4 concentration at G2. The percentage of cows with very low P4 concentration (0.00 to 0.09 ng/mL) at G2 was increased by both treatments (n (GDPG) = 44/179, n (GPPG) = 45/154) compared with the control group (n = 23/158). Only the GPPG treatment, however, decreased the percentage of cows with high P4 concentration (0.6 ng/mL) at G2 compared with the control group (n (control) = 70/158, n (GDPG) = 66/179, n (GPPG) = 41/154). In addition, P/AI was greater for GDPG (38.2%) and GPPG (38.9%) than for control cows (29.8%). In experiment 2, GPPG cows (37.4%) had greater P/AI than control cows (31.0%) and tended to have greater P/AI than GDPG cows (31.8%). Pregnancy per AI did not differ for GDPG and control cows. Interestingly, cows with a CL at G1 (n = 989/1,662) had similar P/AI (34.1%) to cows without a CL at G1 (32.6%) and we found no interaction between treatment and presence of a CL at G1 on P/AI. Because both experiments were conducted using a similar experimental design, results were combined to achieve more statistical power to test the effect of treatment on P/AI. To avoid a possible effect of presynchronization on treatments, cows receiving first postpartum TAI from experiment 1 (n = 572) were excluded from the combined analysis (n = 2,573). Combined results indicated greater P/AI for GPPG (40.3%) than for control (31.8%) and GDPG cows (33.4%). There was no difference for P/AI between GDPG and control cows. Results from this study demonstrate that compared with the traditional 7-d Ovsynch including a single standard PGF dose on d 7 adding a second PGF treatment on d 8 during a 7-d Ovsynch protocol decreased the proportion of cows with high P4 concentration at G2 and increased P/AI by 8.5 percentage units. Adding of a second PGF treatment on d 8 also seems to increase P/AI compared with a double PGF dose (+ 6.9 percentage units). Doubling the PGF dose on d 7 in a 7-d Ovsynch protocol only increased the proportion of cows with very low P4 concentration at G2, but did not affect P/AI compared with a single PGF dose (+ 1.6 percentage units). After all, it cannot be ruled out that the use of a presynchronization protocol influences the effect of a PGF dose frequency modification in an Ovsynch protocol. This was indicated by presynchronized cows receiving first postpartum TAI from experiment 1 having similarly increased P/AI treated with a double PGF dose compared with a second PGF dose. To confirm a possible interaction between the treatment effect and presynchronization, future research is warranted. Overall, this thesis shows that 1) inseminating cows from 7 to 24 h after OE or within 18 h after PAE yields greatest P/AI irrespective of type of semen using a neck-mounted AAM system, 2) estrous expression is highly associated with greater odds for pregnancy, but the proportion of cows showing high estrous expression is limited as influenced by many risk factors, and 3) modification of the 7-d Ovsynch protocol using a second PGF treatment on d 8 is beneficial for P/AI compared with a single or a double PGF dose on d 7. Taken together, these results have the potential to improve reproductive performance on dairy farms using AAM systems or TAI protocols or a combination of both.