Computer-assisted image analysis of endometrial edema and follicular wall attributes and transrectal follicular tonometry in estrous mare

Abstract

The ability to determine the final stages of follicular maturation and detect impending ovulation in the mare still eludes veterinarians. The objectives were to characterize endometrial edema, and follicular wall attributes of the dominant follicle during the estrous cycle in mares and determine how hormonal manipulation influenced the outcomes of interest. A computer-assisted image analysis system was used to objectively evaluate the echotexture oftransrectal ultrasonographic images of the endometrium in the uterine body, midpoint, and tip of the right and left uterine horns in the mare. Echotexture was expressed as a computer generated a numeric pixel value (NPV) which reflected the brightness of the pixel elements within the measurement spot. A novel transrectal tonometric device was used to measure follicular pressure of the dominant follicle during estrus. Ten light horse mares were examined during 4 successive estrous periods with varying degrees of hormonal manipulation (group 1: natural estrus; group 2: prostaglandininduced estrus; group 3: prostaglandin-induced estrus with hCG administrated when the dominant follicle was 35 mm in diameter or greater; and group 4: prostaglandin-induced estrus with hCG administrated when a subjective grade II endometrial edema pattern was attained). General linear models for repeated measures were used to investigate if there were differences (p < 0.05) in mean values over time from ovulation. Data were retrospectively adjusted by hours from ovulation (ovulation = 0 hours). Post-hoc comparisons between time points were made using the Bonferroni method in SPSS. The following attributes were evaluated: the mean NPV of the endometrium, mean NPV of the follicular wall, follicular wall thickness (mean area under the curve) and follicular tonometry (mean pounds per square inch). Mean endometrial NPV in only the uterine body changed over time from ovulation (p< 0.001) in groups 1, 3, and 4. In all 4 groups, there was an inverse relationship between the changes in subjective endometrial edema and endometrial mean NPV in the uterine body. Mean (± SEM) follicular antral diameters increased until 24 hours before ovulation in groups 1 (p = 0.01), 2 (p = 0.03) and 4 (p = 0.02). Mean antral diameters at time of ovulation for the 4 groups were 45.2 ± 5.5 mm, 46.9 ± 13.5 mm, 39.2 4.1 mm, and 43.0 ± 7.8 mm, respectively. Mean (± SEM) follicular growth rates for the four groups up to 24 hours before ovulation were: 2.5 ± 0.6 mm/day, 1.9 ± 0.5 mm/day, 1. 7 ± 0.1 mm/day, and 4.0 ± 0.1 mm/day, respectively (p < 0.05). Mean (± SEM) NPV of the follicular wall increased over time from ovulation for groups 2 (p =0.05), 3 (p < 0.001), and 4 (p = 0.03). The mean (± SEM) daily changes (24 hrs) was 3.1± 1.9 NPV for group 2, 11.7 ± 5.3 NPV for group 3 and 10.7 ± 4.3 NPV for group 4. There was a progressive increase in the follicular wall thickness in groups 1 (p = 0.005) and 2 (p= 0.01). The mean (± SEM) daily changes (24 hrs) for groups 1 and 2 were 71.3 ± 24.6 pixels and 127.0 ± 45.3 pixels, respectively. Mean (± SEM) intrafollicular pressures decreased 0.5 to 1.0 PSI between 24 and 12 hrs prior to ovulation (p = 0.01, p = 0.01, p < 0.001, p = 0.03 for groups 1, 2, 3, and 4, respectively). Hormonal treatment shortened the duration of estrus and accelerated the changes in the attributes evaluated. In conclusion, computer-assisted image analysis and transrectal tonometry provided a means to objectively evaluate the normal, morphological changes in the endometrium and the preovulatory follicle of the mare during the estrous cycle.