Mechanisms behind intrauterine device-induced luteal persistence in mares

Intrauterine glass balls are used to prevent oestrous signs in sports mares, but the mechanism of action is unknown. It has been suggested that the glass ball can mimic an embryo or act via an induced chronic uterine inflammation and absent or continuous low-grade prostaglandin (PG) release. The purpose of this study was to induce prolonged luteal function in mares using a small intrauterine device (IUD) and to study the mechanisms behind prolonged IUD-induced luteal function. A uterine swab and a biopsy specimen were obtained in early oestrus. A water-filled plastic ball, diameter 20 mm and weight 3.6 g, was inserted into the uterus 2–4 days after ovulation; the control mares underwent similar cervical manipulation without ball insertion. The mares were examined three times per week until day 23 and twice weekly thereafter until they returned to oestrus (transrectal palpation, ultrasonography and progesterone determination). The location of the IUD was recorded and ultrasound scans were video-recorded to assess the frequency of uterine contractions. When the mare returned to oestrus, a uterine swab and biopsy specimen were obtained and the bacteriological, cytological and histological (inflammation and glandular dilation) results compared with the samples obtained before the IUD insertion. The PG F2α metabolite levels were measured in the plasma of four control mares and eight IUD mares on days 11–16.The IUD induced a prolonged luteal phase in 75% of the mares (9/12; IUD-P); the mean dioestrous length was 57.0 days. The three mares that did not respond to the IUD (IUD-N) showed a mean dioestrous length of 15.7 days and the 12 control mares 16.1 days. The inflammation and glandular dilation scores were not significantly different in pre- and post-manipulation biopsy specimens. Although locational changes of the IUD were observed, they occurred over very small distances and were mostly limited within the body-bifurcation area. The IUD-N and control mares showed increased uterine contractility 11–16 days post-ovulation, whereas the IUD-P mares did not. The control mares (n = 4) and IUD-N mares (n = 2) showed increased PG levels from day 14 post-ovulation, while the IUD-P mares (n = 6) showed basal levels only. We concluded that the IUD did not cause continuous PG release and suggest that close contact of the IUD with the endometrium may prevent the endometrial cells from releasing PGF2α.