Power Doppler can detect the presence of 7–8 day conceptuses prior to flushing in an equine embryo transfer program

In order to determine whether differences in uterine blood flow between pregnant and non-pregnant mares can be used to predict the presence of the equine embryo prior to flushing in an embryo transfer program, power Doppler ultrasonography was used on a total of 52 mares on days 7 or 8 post-ovulation. Computer analysis of Doppler images was subsequently performed using ImageJ v1.48 software. Vascular perfusion of the endometrium was analyzed using spot meter techniques, measuring mean pixel intensity and area of blood flow. Mares with positive flushings presented a higher uterine blood flow area (one embryo: 54.01 ± 2.27 mm2 or two embryos: 61.01 ± 6.73 mm2) prior to embryo recovery compared to barren mares (21.77 ± 2.22 mm2) (p ≤ 0.05). However, significant differences in vascular perfusion were not detected between single or twin pregnancies. Blood flow area appears to be a good predictor for differentiation between pregnant and non-pregnant mares with an AUC: 0.869; p ≤ 0.001 and an optimal cut-off value of 37.21 mm2. Both the mare's age and day of embryo recovery caused effects on uterine vascular perfusion. According to Youden's J statistics the uterine blood flow area of young pregnant mares was greater than 25.4 mm2 on day 7 (with a sensitivity of 75% and a specificity of 87.5%) and greater than 21.02 mm2 on day 8 post-ovulation (with a sensitivity of 93.8% and a specificity of 100%). The uterine blood flow area in adult pregnant mares was greater than 41.4 mm2 on day 7 (with a sensitivity of 80% and a specificity of 85.5%) and greater than 35.55 mm2 on day 8 after ovulation (with a sensitivity of 97.2% and a specificity of 85.7%). Evaluation on day 8 is therefore considered to be more reliable. Older and middle aged pregnant mares (5–18 years old) had increased uterine vascularization compared to young pregnant mares (2–5 years old) (p ≤ 0.001). Conversely, older barren mares showed higher endometrial vascularity (35.06 ± 2.56 mm2) than young (17.21 ± 1.26 mm2) and middle aged non-pregnant mares (23.84 ± 1.50 mm2) (p ≤ 0.05). We hypothesized that the higher blood flow area seen in older barren mares may be a consequence of a subclinical endometritis due to repeated flushing for embryo recovery. The results of the present study indicate that power Doppler ultrasound combined with computer assisted analysis of images are reliable techniques to detect early pregnancy prior to embryo recovery.