Two hormonal protocols for timed artificial insemination in mares under different residence times of the progesterone intravaginal device

Abstract

The study aimed to evaluate two hormonal protocols for oestrus and ovulation synchronisation for timed artificial insemination (TAI) and their effect on the embryo recovery rate (ERR) in horses. Eleven crossbreed mares were allocated in three groups, taking 3 different oestrus cycles per mare, during two breeding seasons, totalling 17 cycles followed per group, with 11 cycles in the first season and 6 cycles in the second season. Conventional follicular follow-up in the control group (CG; n = 17) was performed by transrectal palpation and ultrasound (US) examination; the ovulation was induced with 500 mu g histrelin (intramuscularly: IM) when the diameter of the follicles was >= 35 mm, the uterine oedema was grade 3 and cervical opening had begun. Artificial insemination with fresh semen (250 x 10(6) sperm cells) was performed 24 h after ovulation inductor administration, and embryos were collected eight days after ovulation (D8). The experimental groups had a progesterone-releasing intravaginal device (PRID) inserted for 9 (PRID9) or 11 days (PRID11). Day zero (D0) was defined as the day of TAI for both groups. The PRID was inserted at the beginning of the protocol (D-14 for PRID9 or D-16 for PRID11) plus US examination of the ovaries; the PRID was removed on D-5, 75 mu g prostaglandin were administered (PGF2 alpha; IM) and US was performed; 500 mu g (IM) of histrelin (ovulation inductor) was administered on D-1; after 24 h (D0), TAI was performed with fresh semen (250 x 10(6) sperm cells) plus US; nine days after TAI (D9), transcervical embryo collection was performed. After the embryo collection, a dose of prostaglandin F2 alpha was administered, with a delay of seven days before the start for the next group in order to desynchronise the mare's oestrus cycles between the protocols, so as not to interfere in the synchronisation results from one group to another. The order of submission of mares to protocols was randomly defined by sortition in each group. The groups were worked simultaneously: PRID11, PRID9 and CG in the first season and PRID11, CG and PRID9 in the second, during three consecutive months within the reproductive season. Treatment in both experimental groups (PRID9 and PRID11) was performed at any stage of the oestrous cycle. The administration of the protocols was applied to cyclical mares regardless of the US findings at the beginning of the study, and ovulation induction was performed regardless of the size of the follicle or uterine oedema. The US examinations of ovulations were only performed to verify whether they were early or late in relation to the protocol. Responsiveness (presence of preovulatory follicle diameter >= 30 mm on the day of ovulation induction D-1) for CG, PRID9 and PRID11 was 100 %, 58.8 % and 41.1 %, respectively (P > 0.05). The ERRs were 52.94 %, 29.41 % and 17.64 % for the CG, PRID9 and PRID11, respectively. There was a statistical difference only between the CG and PRID11 (P < 0.05). It was concluded that the hormonal protocol using PRID for nine days combined with histrelin administration four days after the PRID's removal showed promising results related to the ERR in mares, while reducing conventional follicular monitoring, the number of gynaecological examinations, animal management and visits to the properties. Because no statistical difference was detected between the PRID9 protocol and the CG, it could be used in large equine herds, reducing animal management and the number of visits by professionals. Effects such as early ovulations, the small size of preovulatory follicles at TAI, the interval between P4 removal and ovulation induction, and low ERR in treatment groups might be improved in future studies, aiming at the possibility using the method for commercial purposes under the acceptance of TAI protocols by veterinarians and breeders. Further studies about TAI in horses are recommended.