A prospective randomized trial comparing dephereline and busereline for ovulation induction in heat-stressed lactating dairy cows

Abstract

Climate change is causing heat stress (HS) in dairy cattle. This study aimed to compare the clinical efficacy of two GnRH synthetic analogs, dephereline and busereline, as ovulation inducers under HS conditions. The study population comprised 1,000 lactating dairy cows showing signs of spontaneous estrus which were assigned to the groups: DEPH (489 cows receiving 100 µg of dephereline) and BUS (511 cows receiving 10 µg of busereline) at the time of insemination. Cows were included only once in the study. Treatment with busereline increased the risk of multiple ovulations and twin pregnancies, with an odds ratio (OR) of 1.6, and twin pregnancies, with an OR of 2.8, when compared with dephereline. The likelihood of pregnancy in multiple-ovulating cows was significantly higher in the DEPH group than the BUS group. Collectively, our results comparing two ovulation inducers showed that dephereline treatment may improve the fertility of lactating dairy cows under HS conditions.

Hyperthermia, defined as in increase in body temperature above the homeothermic point (heat stress [HS]), is a major factor that impairs reproduction in animals and humans [1, 2], particularly in dairy cattle [3, 4]. Climate change and global warming have increased the intensity, frequency, and duration of HS [2]. In this scenario, research has focused on improving reproductive efficiency to meet dairy herd economic requirements, while making more efficient use of land resources. Effective reproductive management reduces greenhouse gas emissions [5]. Cooling strategies, such as short-term spraying of water followed by its evaporation from the skin using air from fans, can improve the fertility of lactating cows under HS conditions [3]. However, not all producers can incorporate cooling systems into their farms.

The failure of cows in estrus to ovulate is a major consequence of HS [4]. Gonadotropin-releasing hormone (GnRH) and its analogs are often used as ovulation inducers in veterinary reproductive medicine [6]. Two synthetic GnRH analogs, dephereline and busereline, are more active than GnRH, favoring fertility and embryo survival during summer [7]. This study aimed to compare the clinical efficacy of dephereline and busereline for ovulation induction at the time of artificial insemination (AI) in heat-stressed lactating dairy cows. The experiment was performed in a herd without a cooling system in northeastern Spain.

The study population involved 1,000 lactating dairy cows showing spontaneous estrus which were alternately assigned on a weekly rotational basis to the groups: DEPH (489 cows receiving 100 µg of dephereline) and BUS (511 cows receiving 10 µg of busereline). Seven hundred and eighty-seven cows (78.7%) were inseminated under HS conditions (maximum temperature-humidity index [THI] > 72 units [8]). During the study period (June 27 to November 13, 2023), the maximum THI values ranged from 72.2 to 85.9 units from June 27 to October 18 (114 days), whereas they were less than 72 units from October 19 to November 13 (26 days). Mean milk production, days in milk (DIM), maximum THI values at AI, the number of AI events and the number of lactations were 35.9 ± 7.2 (20–62) kg, 170.0 ± 90.6 (49–442) days, 74.6 ± 6.9 (58–86) units, 4.3 ± 2.9 (1–17), and 2.0 ± 1.3 (1–8), respectively. No significant differences in these values between the DEPH and BUS groups were detected using the Student’s t-test.

A total of 29 cows (2.9%) experienced ovulation failure. No pregnancy was observed in any of the cows. No young corpus luteum (CL) (delayed ovulation) was detected in the ovulating cows. Of the 971 ovulating cows at 7–13 days post-AI, 679 (69.9%) had one CL and 292 (30.1%) had two or more CL (279 [95.5%] with two CLs and 13 [4.5%] with three CLs). Pregnancy was recorded at 28–34 days post-AI in 189 (19.5%) ovulating cows, of which 171 (90.5%) carried singletons and 18 (9.5%) had twins. All the embryos were alive at this time. Pregnancy loss was recorded at 49–55 days post-AI in 15 pregnant cows (7.9%). Of the 811 non-pregnant cows, 439 (54.1%) returned to estrus. Table 1 shows the associations between multiple ovulations, pregnancy after AI of ovulating cows, the number of non-pregnant cows returning to estrus, the number of pregnant cows carrying twins, and independent variables included in the final model using binary logistic regression procedures. No factors were associated with ovulation or pregnancy loss. Parity, HS, repeat breeding, sires, and technicians had no impact on any dependent variable. No interactions were observed.

Table 1. Association between multiple ovulations (n = 292) ^a), pregnancy (n = 189) ^b), non-pregnant cows returning to estrus (n = 439) ^c), or pregnant cows carrying twins (n = 18) ^d) and independent variables included in the final logistic regression model for inseminated cows (n = 1,000).

Factor		Class	n	%	Odds ratio	95% Confidence interval	P
Multiple ovulations a)
	DIM at AI e)	< 90	33/229	14.4	Reference
		≥ 90	259/771	33.6	3.8	2.3–5.6	< 0.0001
	Treatment f)	DEPH	117/489	23.9	Reference
		BUS	175/511	34.2	1.6	1.2–2.7	< 0.0001
Pregnancy b)
	DEPH	1 CL	65/360	18.1	Reference
	DEPH	≥ 2 CL	36/117	30.8	2.5	1.8–4.9	0.01
	BUS	1 CL	58/319	18.2	1	0.9–1.1	0.18
	BUS	≥ 2 CL	30/175	17.1	0.9	0.8–1.2	0.15
Return to estrus c)
	Milk at AI	< 34 kg	148/371	39.9	Reference
		≥ 34 kg	291/440	66.1	2.8	1.9–4.8	< 0.0001
	BUS	≥ 2 CL	54/145	37.2	Reference
	BUS	1 CL	151/278	54.3	1.8	1.4–2.7	0.003
	DEPH	≥ 2 CL	51/81	63	2.2	1.2–4.4	0.001
	DEPH	1 CL	183/308	59.4	2.2	1.6–2.9	< 0.0001
Twins d)
	Treatment f)	DEPH	5/101	5	Reference
		BUS	13/88	14.8	2.8	1.9–4.3	0.015

^a) Hosmer–Lemeshow goodness of fit = 27.2; 3 degrees of freedom (df); P = 0.94. Nagelkerke R² = 0.17. ^b) The interaction between treatment and multiple ovulation rate (≥ 2 CL) was significant (P < 0.01); Hosmer–Lemeshow goodness-of-fit = 22.6; 2 df; P = 0.95. Nagelkerke R² = 0.14. ^c) The interaction between treatment and multiple ovulation rate (≥ 2 CL) was significant (P = 0.01); Hosmer–Lemeshow goodness-of-fit = 26.7; 2 df; P = 0.93. Nagelkerke R² = 0.12. ^d) Hosmer–Lemeshow goodness-of-fit = 24.6; df = 2; P = 0.90. Nagelkerke R² = 0.09. ^b) Ovulating cows (n = 971). ^d) Pregnant cow. ^e) DIM: days in milk. ^f) Treatment: DEPH, cows receiving 100 µg of dephereline; BUS, cows receiving 10 µg of busereline. AI, artificial insemination.

Using cows in the DEPH group as a reference, the odds ratio (OR) for multiple ovulations (two or more CLs) of cows in the BUS group was 1.6 (P < 0.0001), whereas DIM ≥ 90 days increased the incidence of multiple ovulations, with an OR of 3.8 (P < 0.0001). A significant interaction (P < 0.01) was observed between the treatment (dephereline vs. busereline) and multiple ovulations or a single ovulation for the likelihood of pregnancy. Thus, using single-ovulating cows in the DEPH group as a reference, the OR for pregnancy in multiple-ovulating cows in the DEPH group was 2.5 (P = 0.01). The pregnancy rates of single- and multiple-ovulating BUS cows were similar to those of single-ovulating DEPH cows. A significant interaction (P = 0.01) was observed between treatment and multiple ovulations for the likelihood of returning to estrus. The OR for return to estrus in multiple-ovulating cows in the DEPH group was 2.2 when compared with multiple-ovulating cows in the BUS group, whereas the OR for the return to estrus of high milk producers (≥ 34 kg) was 2.8 (P < 0.0001), compared to the remaining low-producer cows (< 34 kg). Using cows in the DEPH group as a reference, the OR for twin pregnancy in cows in the BUS group was 2.8 (P = 0.015).

Three CLs were registered in three (2.6%) of the 117 multiple-ovulating cows in the DEPH group and in 10 (5.7%) of the 175 multiple-ovulating cows in the BUS group. No significant differences in these values were detected between the DEPH and BUS groups using the chi-square test.

To obtain a baseline of how HS affects the cows in the herd, data were obtained from 124 untreated cows inseminated just prior to the start of the study during the first 20 days of June. Mean milk production, DIM, and maximum THI values at AI, and the number of AI events and lactations in untreated cows were 41.1 ± 7.8 (23–60) kg, 136.0 ± 65.3 (60–333) days, 74.6 ± 2.4 (70–81) units, 3.7 ± 2.9 (1–13), and 2.3 ± 1.3 (1–7), respectively. Twenty-five cows experienced ovulation failure (20.2%), and delayed ovulation was recorded in 24 cows (19.4%). No pregnancy was observed in any of the cows. In ovulating cows (n = 75), 71 (94.7%) had one CL and four (5.3%) had two CLs. Pregnancy was recorded in 18 (24%) cows, of which one (5.6%) carried twins and five (27.8%) experienced pregnancy loss. Of the 106 nonpregnant cows, 11 (10.4%) returned to estrus. No factors were found to affect ovulation failure, return to estrus, pregnancy, or pregnancy loss rates. The final model included the effect of the maximum THI value at AI as the only factor influencing the delayed ovulation rate with an OR of 1.1 (P < 0.01) for each unit of maximum THI value increase (95% confidence interval: 1.01–1.15; Nagelkerke R² = 0.12).

In the present study, 2.9% (29/1,000) of the cows experienced ovulation failure, which was much lower than the 12.4% (82/663) described during the warm period in cows showing spontaneous estrus with no treatment at AI in the same geographical area [9] and 20.2% of untreated cows experienced ovulation failure. No significant differences were observed between the treatment groups. This suggests that both dephereline and busereline have strong positive effects as ovulation inducers under HS conditions, eliminating the high risk of delayed ovulation in this herd. In fact, multiple ovulations were registered in 29.2% of the cows, a much higher figure than the 10.9% registered in the reference study [9] and the 5.3% observed for untreated cows. However, a significantly higher proportion of multiple ovulations, and consequently, twin pregnancies, were observed in cows in the BUS group than those in the DEPH group. Regardless of the treatment, as expected, cows with DIM > 90 showed a greatly increased rate of multiple ovulations, reinforcing previous results [9, 10]. A higher pregnancy rate was also expected in cows with multiple ovulations in the DEPH group. Double ovulation has been linked to increased pregnancy rates per AI [11,12,13]. Therefore, cows in the DEPH group experiencing multiple ovulations showed normal behavior, both in terms of fertility and the rate of return to estrus in non-pregnant cows. The return to estrus was similar for monovular cows in the DEPH and BUS groups and for multiple-ovulating cows in the DEPH group. However, it is more difficult to determine the reason for the higher incidence of multiple ovulations in cows in the BUS group. Although it resulted in an increased twin pregnancy rate, it was not associated with a further increase in the pregnancy rate. One possible explanation for this is the impairment in the treatment of normal folliculogenesis in superovulating cows in the BUS group. The increased multiple ovulation rate in cows in the BUS group suggests a tendency of the treatment to induce superovulation. This can compromise follicular maturation and subsequent luteal function, both of which are strongly altered when superovulation is induced [14]. The fact that non-pregnant multiple-ovulating cows in the BUS group had less chances to show estrus, with an OR of 0.5 (1/2.2) when compared with their counterparts in the DEPH group, reinforces this idea. In addition, although not significant, the proportion of multiple-ovulating cows with three CLs was higher for the BUS (5.7%) group than the DEPH group (2.6%). If our interpretation is correct, the mechanism by which busereline treatment increases multiple ovulations needs to be established.

Milk production was positively correlated with the rate of return to estrus in non-pregnant cows. Although HS greatly reduces the ability of cows to show estrus [3, 4], cows that are better adapted to HS can also show estrus while simultaneously being high producers. Unexpectedly, HS reduction at the end of the study (26 days) had no positive effect on any of the parameters studied. This was likely due to the long-lasting effect of HS after the end of the HS conditions [15].

Collectively, our data revealed differences between the two ovulation inducers. Both dephereline and busereline showed strong positive effects as ovulation inducers, whereas busereline treatment increased the risk of multiple ovulation and twin pregnancies when compared with dephereline. In addition, dephereline treatment improved the fertility of lactating cows under HS conditions. However, the clinical perspective of this comparative study should be further confirmed during winter with a positive photoperiod.

Methods

Experimental animals

This study was performed in a commercial dairy herd of lactating Holstein dairy cows reared in north-eastern Spain (41.39° latitude, 0.52° longitude). During the study period, the mean number of lactating cows in the herd was 1,850, and the mean annual milk production was 10,150 kg per cow. The mean annual culture rate was 34%. The cows were grouped according to age (primiparous plus secundiparous versus multiparous), milked three times daily, and fed a complete diet. Cows were included if they were healthy with a body condition score of 2.0–3.5 on a scale of 1 to 5 [16], produced more than 20 litres of milk per day, and were free of detectable reproductive disorders and clinical diseases during the study period (days –5 to +50 from insemination). The exclusion criteria were mastitis, lameness, digestive disorders, and pathological abnormalities of the reproductive tract that were detectable on ultrasonography. Only cows that were inseminated during spontaneous estrus were included in the study. The cows did not receive any hormone treatment for at least 21 days before AI. Estrus was detected using accelerometer data from cow collars (DelPro^TM, DeLaval, Tumba, Sweden), and confirmed using rectal palpation of the uterine tone (highly turgid and contractile to touch) and obtaining mucus from the vagina using transrectal uterine massage. The transparency and viscosity of the mucus were examined. If these tests indicated that the cow was not in estrus and ready for service, the cow was rejected for AI. All cows were inseminated by three technicians using freeze-thawed semen from two bulls.

Experimental design

A clear negative effect of heat stress on the reproductive performance of lactating dairy cows has been described from May to September in our geographical area [10, 17], and ovulation failure of preovulatory follicles has been found to increase dramatically during this period [9, 18]. In addition, a negative photoperiod (decreased day length from June 22 to December 21) has been linked to increased ovulation [10]. Thus, this study was performed over a negative photoperiod (June 27 to November 13) involving most AI events during the warm period to increase the chances of both ovulation failure and multiple ovulations. All cows received either a dose of dephereline (gonadorelin acetate [6-D-Phe]; Gonavet Veyx; Ecuphar, Barcelona, Spain) or busereline (busereline acetate; Receptal; Intervet, Salamanca, Spain) at the time of AI to promote ovulation.

Ovulation, defined based on the presence of at least one mature CL > 14 mm in diameter and the number of CLs was assessed using ultrasound 7–13 days post-AI using a portable B-mode ultrasound scanner equipped with a 5–10 MHz transducer (E.I. Medical IBEX LITE; E.I. Medical Imaging, Loveland, CO, USA). The dimensions of the CL were recorded as the mean of two measurements, approximating the maximum length and width. As the presence of a central cavity is not functionally important [19, 20], CLs with a cavity were recorded in a similar manner to that of a solid CL. Delayed ovulation was defined as the presence of one young CL 7–13 days post-AI. The lack of association with a young CL [21, 22] was used as a reference to confirm the state of CL maturity. Pregnancy was diagnosed using transrectal ultrasound 28 days post-estrus and revised 21 days later. Pregnancy loss was recorded when the second examination result was negative. The pregnancy rate was defined as the percentage of cows that became pregnant for the total number of ovulating cows in the corresponding group. If a cow returned to estrus, its status was confirmed by rectal examination, and the animal was recorded as non-pregnant. Because twin births are not desirable in dairy herds and twin pregnancies have been linked to hormonal treatments [23], the number of embryos was recorded at the time of pregnancy diagnosis in pregnant cows. All gynecological examinations and pregnancy diagnoses were performed by the same surgeon. The viability of the embryo/fetus was confirmed by observing the heartbeat in all examinations.

The working conditions for the group of untreated cows were the same as those for the experimental cows, except that AI was performed during the positive photoperiod (increasing day length).

Data collection and statistical analyses

The following data were recorded for each animal: parturition and AI dates; parity (primiparous versus pluriparous); treatment (dephereline versus busereline); maximum THI at AI (≤ 72 units vs. > 72 units); milk production (low producers < 34 kg versus high producers ≥ 34 kg), DIM (≤ 90 days vs. > 90 days); repeat breeder at AI (≤ 3 AI vs. > 3 AI); sire; AI technician; ovulation failure; delayed ovulation; multiple ovulations (two or more CLs); return to estrus; pregnancy; presence of twins in pregnant cows; and pregnancy loss. The threshold of 34 kg for milk production was the median for primiparous cows, whereas the threshold value of 90 DIM was based on previous studies of possible factors influencing double ovulation [9, 10]. Possible inter-group differences in terms of milk production, DIM, maximum THI values at the time of AI, and number of AI events and lactations were analyzed using the Student’s t-test. Temperature and relative humidity data were obtained from a meteorological station located less than 800 m away from the herd.

The effects of treatment on the rates of ovulation failure, multiple ovulations, return to estrus, pregnancy, twin pregnancy, and pregnancy loss (dependent variables) were analyzed using binary logistic regression (PASW Statistics for Windows Version 18.0; SPSS Inc., Chicago, IL, USA) according to the method described by Hosmer and Lemeshow [24]. The factors entered into the model as independent dichotomous variables (where 1 denotes presence and 0 denotes absence) were parity (pluriparity), HS (THI > 72 units), DIM (> 90 days), repeat breeder (> 3 AI events), and milk production (≥ 34 kg) at AI. Treatment, AI technician, and sire (class variables) were considered factors in the analyses. Possible interactions between treatment and the dichotomous variables parity, HS, DIM, repeat breeding, and milk production were also examined. Only non-pregnant cows were included in the analysis of the dependent variable that returned to estrus. Only ovulating cows were included as dependent variables in the analysis. Only pregnant cows were included in the analyses for the dependent variables of twin pregnancy and pregnancy loss.

As the incidence of multiple ovulations and twins was very low in untreated cows, the effects of HS were analyzed using binary logistic regression, as in the experimental cows, only on the rates of ovulation failure, delayed ovulation, return to estrus, pregnancy, and pregnancy loss.

Conflicts of interests

The authors declare no conflicts of interest.