512 Original Paper Czech J. Anim. Sci., 58, 2013 (11): 512–519 Supported by the Ministry of Education, Youth and Sports of the Czech Republic (S grant). he luteal blood low, area and pixel intensity of corpus luteum, levels of progesterone in pregnant and nonpregnant mares in the period of 16 days after ovulation J. Šichtař 1 , R. Rajmon 1 , K. Hošková 1 , D. Řehák 2 , L. Vostrý 3 , H. Härtlová 1 1 Department of Veterinary Sciences, Faculty of Agrobiology, Food and Natural Resources, University of Life Sciences Prague, Prague, Czech Republic 2 Institute of Animal Science, Prague-Uhříněves, Czech Republic 3 Department of Genetics and Breeding, Faculty of Agrobiology, Food and Natural Resources, University of Life Sciences Prague, Prague, Czech Republic ABSTRACT: he objective of the present study was to examine if luteal blood low (LBF) monitoring could be used as an additional prognostic tool for early pregnancy diagnosis, and we particularly focused on the diferences in LBF between pregnant and nonpregnant mares. Furthermore, other possible developmental dif- ferences of corpus luteum (CL) between pregnant and nonpregnant mares were evaluated. he CL (n = 119) of 27 mares were monitored once daily in B- and Power-Doppler Mode on days 1, 2, 9, 12, and 16 after ovulation (day 0 = ovulation). he data were evaluated using the MIXED Linear Model with repeated measures, and parameters were estimated by the REML method. he course of LBF, area of CL, and pixel intensity difered in nonpregnant mares on a day-to-day basis in contrast to more stable values in pregnant mares. Further, the proiles of the courses were identical until day 9, but since day 12 the diferences between pregnant and non- pregnant mares started to be prominent. he LBF, pixel intensity, and level of progesterone (P4) were similar in all mares until day 16, when smaller LBF, lower pixel intensity, and lower levels of P4 were found in non- pregnant mares (P = 0.04, P = 0.02, P < 0.05, respectively). In pregnant and nonpregnant mares the LBF was weakly (r = 0.29 in both) and pixel intensity strongly (r = 0.48 and 0.59, respectively) correlated to the levels of P4. LBF was strongly correlated to the area of CL in pregnant as well as nonpregnant mares (r = 0.72 and 0.64, respectively). In accordance with the results presented in our study we can state that LBF monitoring is not a suitable tool for early pregnancy diagnosis or prognosis as the diferences between pregnant and nonpregnant mares are notable – similarly to other indicators of CL status – just after the onset of luteolysis (day 16) when embryo itself is detectable. Keywords: horse; pregnancy; ultrasonic imaging; Doppler ultrasonography; image analysis Early pregnancy diagnosis is an important man- agement tool which greatly enhances the efficiency of a breeding program in mares. Breeders want to check their mares for pregnancy as early as possible mostly for economic purposes. There are several more or less reliable methods for early pregnancy diagnosis (reviewed in Ginther, 1992). In the early pregnancy period the CL plays a crucial role (Ginther et al., 1985). Luteal tissue produces progesterone (P4) and generally the plasma values > 10 ng/ml are sufficient for successful embryo development and maintenance of pregnancy in the early period (Squires et al., 1974; Ginther, 1992). Pregnancy at such an early period could therefore