If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Our objective was to retrospectively compare pregnant versus nonpregnant cattle in terms of vascular and morphometric changes in corpora lutea between d 12 and 20 following timed artificial insemination (TAI). Crossbred (Gir × Holstein) lactating dairy cows (n = 136) and heifers (n = 111) were bred after synchronizing ovulations using an estradiol plus progesterone (P4)-based protocol. Corpus luteum (CL) characteristics (area, echotexture, blood flow) were recorded at 48-h intervals from d 12 to 20 following TAI using an ultrasound equipped with color Doppler. Blood samples were collected to determine CL function (plasma P4). Pregnancy diagnosis was performed at d 30. Quantitative assessment of colored pixels within the CL was performed using ImageJ software (National Institutes of Health, Bethesda, MD) and echotexture was quantified using custom software. Continuous variables such as luteal tissue area (LTA), CL blood flow (CLBF), adjusted CLBF (ratio LTA:CLBF), mean pixel value (MPV), pixel heterogeneity (HETER), and plasma P4 were analyzed retrospectively as repeated measures (d 12 to 20) in pregnant versus nonpregnant females using PROC MIXED (SAS Institute Inc., Cary, NC). Main effects were pregnancy status, day of cycle, and their interaction. Further analyses used only data from d 16, because this was the earliest time point of deviation between CLBF of pregnant and nonpregnant animals. We created quartiles for each variable and calculated the risk of pregnancy within quartile. Differences were determined using the chi-squared test. Plasma P4 was significantly higher in prospective pregnant versus nonpregnant cattle on d 18 and 20, whereas LTA differed only on d 20. On d 16, CLBF and adjusted CLBF diverged between pregnant and nonpregnant, followed by a progressive reduction in the latter until d 20. Mean pixel value was not affected by pregnancy status, but HETER was lower on d 20 in pregnant than in nonpregnant cattle. Likelihood of pregnancy increased from quartile (Q)1 (lowest values) to Q4 (highest) of CLBF (Q4 vs. Q1, odds ratio = 32.8, 95% confidence interval: 9.6 to 112.1) and adjusted CLBF [Q4 vs. Q1, odds ratio = 25.4, 95% confidence interval: 8.1 to 80.4), whereas a lower risk of pregnancy was observed only for animals within Q1 of plasma P4 [Q4 vs. Q1, odds ratio = 3.1, 95% confidence interval: 1.3 to 7.2). Day 16 quartiles of LTA, MPV, and HETER did not affect odds of pregnancy. In conclusion, we identified distinct CLBF patterns as early as 16 d after TAI and confirmed that CL function is lost by a reduction in blood flow, which precedes physical regression.
The ultimate goal in cattle reproductive management is the successful establishment of a pregnancy that is carried to term. One of the primary steps to achieve this goal is driven by the early conceptus, which must signal its presence to the dam in a well-orchestrated biological process referred to as maternal recognition of pregnancy (
Previous studies have used B-mode, grayscale ultrasonography to demonstrate that functional luteolysis (loss of function) precedes structural luteolysis (loss of tissue volume) by at least 24 h after exogenous treatment with PGF2α in cattle (
), a palpable and visually evident CL may still be observed by ultrasound imaging in the proestrus phase and as late as during behavioral estrus. Therefore, to make reliable inferences about loss of luteal function based solely on morphological evaluation, comparisons of data from serial grayscale B-mode exams (24 or 48h intervals) are required.
The advent of color Doppler ultrasound imaging made it possible to examine the intensity of blood flow toward and within the CL by observing the movement of red blood cells within vessels (
). In normal luteal physiology, angiogenesis is positively correlated with systemic progesterone (P4) concentrations during CL development, and apoptosis of endothelial cells is involved in both functional and structural luteolysis (
). The latter approach has the potential to improve cattle reproductive efficiency by allowing earlier resynchronization of open females and, consequently, a reduction in the interval between services.
Importantly, the association between CLBF and morphology (area or diameter) increases the reliability for early prediction of pregnancy success (
). Perhaps the source of inconsistencies among studies lies in the fact that time points for CLBF evaluations varied between different studies, and the criteria adopted to make inferences about CL function by its appearance in color Doppler and B-mode imaging were sometimes divergent among studies. Nevertheless, the association of B-mode and Doppler imaging for assessing the functional status of ovarian structures may provide novel strategies for intensive reproductive management in cattle.
Because color Doppler ultrasonography has been used to identify loss of vascularization within the CL (
), we hypothesize that evaluation of CLBF could be an earlier indicator of physiological luteolysis in cattle compared with evaluation of CL morphology, echotexture, or laboratory assays for P4. The objectives of this study were to (1) evaluate differences in CL features from d 12 to 20 after timed AI (TAI), and (2) determine how early changes in CLBF, morphology (size), echotexture, and function are distinguishable between prospective pregnant and nonpregnant cattle.
MATERIALS AND METHODS
Procedures described in this study were in accordance with the Brazilian Ethics, Bioethics and Animal Care Committee (CEBEA) guidelines and approved by the Embrapa Institutional Ethics in the Use of Animals Committee (CEUA, protocol CEUA-CNPGL 022011).
Location and Experimental Design
This experiment was performed in 2 replicates, within a 72-d period, from July to September (winter in the Southern Hemisphere). Holstein × Gir crossbred lactating dairy cows (n = 136) and heifers (n = 111) with BCS ranging from 2 to 4 (scale 1 to 5;
) were the study subjects. Lactating cows were 45 to 100 d in milk and heifers were cycling, between 18 to 25 mo of age, and with a BW ranging from 330 to 360 kg at the beginning of the experimental period. All females were maintained in outdoor corrals at the Embrapa Research Farm, located in Coronel Pacheco, MG, Brazil (21°25′16″ S and 43°15′56″ W), had access to pasture during the night and were fed corn silage and ration during the day. Water, salt, and minerals were available ad libitum.
Females were synchronized using a standard TAI protocol consisting of 2 mg i.m. of estradiol benzoate (EB; Sincrodiol 1 mg/mL, Ourofino, Ribeirão Preto, SP, Brazil) and insertion of an intravaginal P4 device (1.0 g of P4, Sincrogest, Ourofino) on d −10; 0.526 mg i.m. of sodium cloprostenol (PGF2α analog; Sincrocio 0,25 mg/mL, Ourofino) and P4 device removal on d −2; 1 mg i.m. of EB on d −1 and TAI on d 0, 52 to 54 h after P4 withdrawal. Semen straws were purchased from a commercial company and all TAI were performed by a single technician. Pregnancy diagnosis was performed 30 d after TAI by ultrasonography to visualize the embryonic vesicle and embryo heartbeat. Then, data were retrospectively analyzed based on pregnancy status; that is, pregnant versus nonpregnant females.
Ultrasonography and Color Doppler Flow Imaging
Beginning at d 12 post-TAI, the ovaries of all females were scanned for CL evaluation. Animals that did not present with a CL on d 12 (38 of 247) and 2 cows that had 2 CL were removed from the study. Thus, females with a single CL were examined by ultrasonography at 48-h intervals until d 20 (d 12, 14, 16, 18, and 20). We used a portable ultrasound (MyLab30 Vet Gold; Esaote SpA, Genova, Italy) equipped with a linear-array rectal transducer, capable of creating images in B-mode (grayscale imaging) and in color-flow Doppler mapping mode (blood flow assessment). Imaging settings (gain, depth, frequency, power, focus position, frame rate) were standardized and remained unchanged for the duration of the experiment. In B-mode, transducer frequency was set to 7.5 MHz and total gain 70%, whereas for the color-flow mode, frequency was 6.6 MHz, pulse repetition frequency 1 kHz, frame rate of 23 frames/s, and minimum detectable velocity of 0.06 m/s.
Both ovaries were scanned and the location of the CL was recorded for each animal. Then, B-mode and color-flow mode short video clips (7 s duration) were stored in the internal memory of the ultrasound machine. Later, CL area and area of cavity (if present) were calculated using the internal calipers of the machine. To account for the presence of fluid-filled cavities, luteal tissue area (LTA) was calculated by subtracting the area of cavity from the CL area. As for quantitative assessment of vascularization, an image at the maximum diameter of the CL was analyzed using the ImageJ software (
; National Institutes of Health, Bethesda, MD) to determine the area of colored pixels within the CL, an indirect estimation of CLBF. An adjusted CLBF was calculated dividing CLBF by LTA (CLBF:LTA ratio) to determine the proportion of colored pixels relative to the area of tissue. Last, we performed a quantitative assessment of CL echotexture as described by
. Briefly, a minimum representative area was arbitrarily selected within the luteal tissue and the brightness of each pixel composing that area (5,184 pixels) was quantified by a custom software (Quantporo, Federal University of Vicosa, MG, Brazil) in a scale of 256 shades of gray (0 = black; 255 = white). The mean pixel value (MPV) and pixel heterogeneity (HETER) were quantified for each CL in this experiment.
Blood Samples and Progesterone Analysis
On each day of examination, a blood sample was collected by puncture of the coccygeal vein into 4.5-mL test tubes containing K3 EDTA (Vacutainer Systems, Becton Dickinson, São Paulo, SP, Brazil). Samples were centrifuged at 900 × g for 15 min and the plasma was harvested and stored at −20°C in 1.5-mL Eppendorf tubes. Plasma P4 concentrations were determined using a solid-phase 125I RIA (Coat-a-Count Progesterone, Siemens Medical Solutions Diagnostic, Los Angeles, CA) at the São Paulo State University endocrine laboratory following procedures described elsewhere (
). Sensitivity was 0.02 ng/mL and inter- and intraassay coefficients of variation were 2.7 and 9.7%, respectively.
Data were analyzed retrospectively after pregnancy diagnosis 30 d post-TAI. The binomial variable pregnancy per AI (P/AI) was analyzed using the Fisher exact test and considered all females subjected to the TAI protocol, including those that failed to ovulate and did not have a CL on d 12. Animals were grouped according to pregnancy status (pregnant or nonpregnant) and further analysis consisted of comparisons of all variables between these 2 groups. Continuous outcome variables (LTA, CLBF, adjusted CLBF, MPV, HETER, and plasma P4) were examined for normality and homogeneity of variances using the Shapiro-Wilk and Bartlett tests. No transformation was required for these variables. Statistics was performed using PROC MIXED of SAS (SAS Institute Inc., Cary, NC) with a repeated statement to account for measurements over time (d 12, 14, 16, 18, and 20 post-TAI). The statistical model included animal, pregnancy status, replicate, and day of the cycle as class variables, with animal within pregnancy status and replicates as random variables. All outcome variables were analyzed for the main effects of pregnancy status, day of cycle, and their interaction. Preliminary statistical analysis indicated that animal category (heifer vs. cow) did not affect any variable included in the statistical model; therefore, all data presented here were analyzed using combined data from cows and heifers.
Because partial results demonstrated that the earliest deviation in CL features between pregnant and nonpregnant females occurred on d 16, we further examined data from d 16 separately for all continuous variables and determined cutoff values at the 25th, 50th and 75th percentiles, dividing distinct CL features of each animal into quartiles (Q)1 to 4 (1 = lowest values; 4 = highest values) The risk of pregnancy (P/AI) within each quartile was analyzed using the Chi-squared test to determine differences in P/AI among quartiles. In addition, we analyzed the likelihood of pregnancy within each quartile as follows: Odds of pregnancy = P(1)/P(0), where P(1) is probability of being pregnant (number of pregnant/total number of animals in a given quartile) and P(0) is probability of not being pregnant (number of nonpregnant/total number of animals in a given quartile). If a statistical difference was detected in P/AI between 2 given quartiles, we calculated the odds ratio (OR) by dividing odds of pregnancy in one quartile by odds of pregnancy in the other. Finally, to determine optimal cut-off values for each endpoint at d 16, we performed receiver operating characteristic (ROC) analyses using the GraphPad Prism 8 software (GraphPad Inc., San Diego, CA). A ROC curve was created and the area under the curve (AUC), specificity (Sp), sensitivity (Se), and P-values of the ROC analysis were calculated for each outcome variable. We also determined optimal cutoff values at d 16 that were most predictive of pregnancy status.
Additionally, we determined the occurrence of abnormally short, normal, and long estrous cycles among nonpregnant animals (i.e., those that were bred but did not become pregnant and came back into estrus). For that, we arbitrarily determined that a plasma P4 ≤1.0 ng/mL meant occurrence of functional luteolysis (i.e., loss of function), following the same criterion previously adopted by others (
). Finally, data from animals bred, not pregnant, with luteal phases that ended (P4 ≤1.0 ng/mL) before or on d 20 (short and normal length; n = 46) was analyzed using PROC MIXED of SAS, with a repeated statement to account for the main effect of day of cycle upon each continuous variable described previously. For this analysis, d 0 was considered the day that plasma P4 reached its nadir (≤1.0 ng/mL) and, thus, data from d −2, −4, −6, and −8 relative to P4 nadir were assessed. Comparisons among means from different days relative to luteolysis were performed using the Tukey-Kramer post hoc test. Statistical significance was determined based on a P-value of 0.05, and results are shown as least squares means ± standard errors of the mean, unless otherwise stated.
Of all animals submitted to TAI, 15.4% (38 of 247) did not present with a CL at the ultrasound examination on d 12; that is, ovulation rate in response to TAI protocol was 84.6% (198 of 247). We therefore removed females without a CL on d 12 from further analysis. Additionally, we excluded from data analysis 3 animals that developed cystic ovarian disease during the experiment and 8 animals with plasma P4 <1.0 ng/mL on d 12. Overall P/AI was 42.1% (104 of 247). In the end, 198 females contributed with data on d 12 to 20 (104 pregnant and 94 nonpregnant).
Temporal analysis of CL morphology revealed that LTA remained constant from d 12 to 18, followed by a significant reduction of LTA in nonpregnant animals on d 20 (Figure 1A; P < 0.0001). In addition, we observed an effect of day on plasma P4 concentrations (P < 0.0001) and a significant decrease in nonpregnant animals on d 18 and 20, compared with pregnant females (Figure 1B; P < 0.0001). Quantification of colored pixels as an indicator of vascularization (CLBF) revealed that pregnant and nonpregnant cattle had similar values of CLBF on d 12 and 14, but diverged from d 16 onward; that is, blood flow was increased in pregnant animals simultaneously with a decrease in nonpregnant animals (Figure 1C; P < 0.0001). Similarly, the CLBF:LTA ratio (adjusted CLBF) did not change on d 12 and 14, but the values for pregnant and nonpregnant females differed from d 16 onward (Figure 1D; P < 0.0001). It is noteworthy that the magnitude of differences in CLBF and adjusted CLBF increased progressively over time (d 16, 18, and 20; Figures 1C and 1D). As for CL echotexture, MPV was not affected by pregnancy status (P > 0.10), except on d 18, when pregnant animals had a greater MPV than nonpregnant (Figure 1E; P = 0.0228). Finally, HETER increased from d 12 to 20 (day effect, P < 0.0001) and was affected by pregnancy status (P = 0.0391); that is, it had higher values in pregnant than in nonpregnant females on d 20 (Figure 1F).
We observed that the amount of blood flow to the CL on d 16 post-TAI was an early predictor of loss of CL function because it preceded the divergence in plasma P4. Therefore, we considered d 16 of the cycle a decisive moment for CLBF between animals that eventually became pregnant or not and further divided values of various CL features into quartiles and analyzed risk of pregnancy (P/AI) within quartile on that day. We observed an increase in the risk of pregnancy from Q1 to Q4 for CLBF and adjusted CLBF (Figure 2C and 2D), whereas lower or higher values for LTA did not affect the risk of pregnancy (Figure 2A). Females with the lowest plasma P4 concentrations on d 16 (Q1; below 25th percentile) had a significantly lower risk of pregnancy than females in Q2 and Q4 (Figure 2B; contrast Q1 vs. Q2, P = 0.0381 and Q1 vs. Q4, P = 0.0128). Nonetheless, females in Q1 and Q3 for plasma P4 had similar risk of pregnancy. Risk of pregnancy did not differ among quartiles of echotexture features (MPV and HETER; Figure 2E and 2F).
Considering only features of d-16 CL that affected risk of pregnancy (CLBF, adjusted CLBF, and plasma P4), analyses of OR indicated that the likelihood of pregnancy increases in females with the highest values of CLBF (Q4) compared with Q3 (OR = 6.9, 95% CI: 2.1 to 22.5; P = 0.0013), Q2 (Q4 vs. Q2, OR = 9.2, 95% CI: 2.8 to 30.6; P = 0.0003) and Q1 (Q4 vs. Q1, OR = 32.8, 95% CI: 9.6 to 112.1; P < 0.0001). Similar findings were observed for adjusted CLBF (Q4 vs. Q3, OR = 4.0, 95% CI: 1.3 to 12.2; P = 0.0150; Q4 vs. Q2, OR = 15.0, 95% CI: 4.9 to 45.5; P < 0.0001; and Q4 vs. Q1, OR = 25.4, 95% CI: 8.1 to 80.4; P < 0.0001). Additionally, the likelihood of pregnancy increases in females with intermediate (Q2) and higher plasma P4 (Q4) compared with the lowest quartile (Q2 vs. Q1, OR = 2.6, 95% CI: 1.1 to 5.9; P = 0.0251; and Q4 vs. Q1, OR = 3.1, 95% CI: 1.3 to 7.2; P = 0.0082). Finally, results of ROC analysis (Figure 3) indicated that adjusted CLBF and CLBF were the best possible predictors of pregnancy at d 16 (AUC = 0.81 and 0.78, respectively; P < 0.0001), whereas plasma P4 was a fair predictor (AUC = 0.61; P = 0.0076), and LTA (AUC = 0.52; P = 0.58), MPV (AUC = 0.51; P = 0.87), and HETER (AUC = 0.52; P = 0.59) were poor predictors of pregnancy status if assessed at d 16. Optimal cutoff values were >15.68% of adjusted CLBF (Sp 84% and Se 70%; Figure 3D), >0.55 cm2 of CLBF (Sp 79% and Se 64%; Figure 3C), and >3.45 ng/mL of plasma P4 (Sp 30% and Se 97%; Figure 3B).
Among bred, nonpregnant females, data from a subset of females having cycles of short or normal lengths (n = 46) were used in the analysis of CL features normalized to the time of functional luteolysis (plasma P4 ≤1.0 ng/mL); that is, days relative to luteolysis (d 0, luteolysis; −2, −4, −6, and −8). Within this subset of females, 2 animals had data from d 0 to −4, 18 animals had data from d 0 to −6, and 26 animals had data from all days (d 0 to −8). The remaining bred, nonpregnant females (36 of 94; 38.3%) had a long cycle; that is, plasma P4 >1.1 ng/mL at the end of the study (d 20) and did not enter the analysis with normalized P4. Data of CL features on days relative to luteolysis is shown in Table 1. All outcome variables were affected by day of cycle; that is, as the cycle advanced toward luteolysis, CL features (area, blood flow, echotexture, and function) underwent significant changes.
Table 1Morphology, blood flow, echotexture, and function of bovine corpora lutea (CL) around the time of luteolysis in bred, nonpregnant cattle
All measurements were normalized to d−8, −6, −4, and −2 relative to functional luteolysis (d 0; plasma progesterone ≤1.0 ng/mL). Luteolysis was arbitrarily considered as the day plasma progesterone reached a nadir of ≤1.0 ng/mL. Data are mean ± SEM.
Within a row, means without a common superscript differ (P < 0.05).
a–d Within a row, means without a common superscript differ (P < 0.05).
1 All measurements were normalized to d−8, −6, −4, and −2 relative to functional luteolysis (d 0; plasma progesterone ≤1.0 ng/mL). Luteolysis was arbitrarily considered as the day plasma progesterone reached a nadir of ≤1.0 ng/mL. Data are mean ± SEM.
In this study, we demonstrated that the blood supply of the bovine CL deviates between prospective pregnant versus nonpregnant females as early as at d 16 after TAI, 24 to 48 h before any divergent values could be identified in other CL morphological and functional features. Moreover, we detected a noticeable progressive decrease of CL blood flow from d 16 onward in animals that eventually did not become pregnant, which seems to be a reliable indicator of functional luteolysis. The occurrence of luteolysis is the primary physiological indicator of failure in maternal recognition of pregnancy (
), and Doppler imaging appears to be a suitable tool to study CL regression at early time points, whereas examinations of other CL features do not seem as useful for that purpose.
Our findings corroborate the current knowledge about the physiological mechanisms of luteolysis; that is, lack of vascular perfusion halts P4 secretion and follows degradation and regression of the luteal tissue (
). We also confirmed that the initial signs of pregnancy failure and early events leading to CL lysis may be detected as early as 16 d post-TAI if the right tool is used, an approach that can be of great advantage for aggressive management strategies that use intensive resynchronization schedules to improve reproductive performance (
Reproductive performance of dairy cows resynchronized after pregnancy diagnosis at 31 (±3 days) after artificial insemination (AI) compared with resynchronization at 31 (±3 days) after AI with pregnancy diagnosis at 38 (±3 days) after AI.
Interestingly, we observed a progressive increase in CLBF during the experimental period in pregnant females. This finding is reminiscent of reports indicating a similar increase in uterine blood flow (
) during the first weeks of pregnancy in the cow. Earlier studies indicated that uterine blood flow increased in the gravid uterine horn during early pregnancy, probably in response to the presence of the conceptus (
). Molecular mechanisms involved in the increase in CLBF in pregnant cows remain unclear, but we speculate that they may involve the expression of growth factors (e.g., vascular endothelial growth factor, VEGF) and luteotropic molecules secreted by the conceptus. In humans, rodents, and some primates, angiogenesis increases in early pregnancy in response to prolactin (rodents) or chorionic gonadotropin (primates and humans), but this pattern was not observed in marmosets (reviewed by
). Further studies are required to determine which factors are involved in the increase of CLBF in pregnant cattle.
The physiology of post-PGF2α events has been well described in ruminants. It involves a reduction in angiogenic activity within the CL, a decrease in blood flow to the gland, and, consequently, death of large and small steroidogenic luteal cells, followed by tissue degradation (
). The use of B-mode ultrasonography to monitor CL development and regression is, consequently, an approach of limited usefulness due to its ability to detect CL regression only a couple of days after CL function has already been lost and plasma P4 has decreased to basal levels (
The outcomes of the current study suggest an alternative approach to overcome this bottleneck if we consider that the evaluation of CLBF by Doppler ultrasonography may anticipate, by 24 to 48 h, the imminent decrease in plasma P4 levels in females that will return to estrus. Also, thresholds of CLBF and adjusted CLBF used in this study to sort females into quartiles may be used to determine the likelihood of pregnancy in a given group of animals examined as early as 16 d post-TAI. Moreover, ROC curves were used to better explore the relationship between CL features and pregnancy status at d 16. This analysis indicated optimal cutoff values of CLBF (0.55 cm2) and adjusted CLBF (∼16%) that may be used at d 16 with fairly high sensitivities and specificities to predict pregnancy outcome.
Quantitative assessment of image attributes (e.g., MPV and HETER) were not advantageous for early prediction of luteolysis. Pixel heterogeneity differed between pregnant and nonpregnant females on the last day of examination (d 20), later than other CL attributes. The use of computer-assisted image analysis was proposed as an alternative to optimize interpretation of sonographic images of the CL (
). The lack of a consistent association between the image attribute MPV and loss of CL function was confirmed in the current study by analysis of data from nonpregnant cows, normalized to the day of luteolysis (Table 1). Mean pixel value decreased numerically as the cycle approached functional luteolysis but differences were not statistically significant. Interestingly, HETER increased significantly on the day of luteolysis compared with 8 and 6 d earlier. The increase in HETER at the time of luteolysis is a finding reminiscent of a previous report from our group (
). Other CL morphological and vascular features decreased consistently 24 to 48 h before progesterone concentrations reached values <1.0 ng/mL, in bred, nonpregnant females.
A secondary advantage of ultrasonographic examinations around the time of luteolysis is the early identification of females that either failed to ovulate in response to the TAI protocol (∼15% in this study) or animals that had asynchronous or short estrous cycles. For example, 3.6% (9 of 247) of the females had abnormally short cycles, with plasma P4 below the minimum threshold of 1.0 ng/mL on d 12 post-TAI. For those animals, Doppler imaging was useful to identify failure of timely ovulation; that is, CLBF in most females with short cycles fell into Q1 (7 out of 9) and Q2 (2 out of 9), which indicated that those animals were very unlikely to become pregnant.
An additional finding of this study was the characterization of long cycles (P4 >1.0 ng/mL on d 20) in a considerable proportion of animals (38.3%) that were later diagnosed as nonpregnant at d 30. This observation is particularly important for early diagnosis of nonpregnant animals using CLBF evaluation on d 20, as described previously by our group (
). That previous study demonstrated a considerable incidence of false-positives, which is in accordance with the fair proportion of females with high P4 on d 20 observed in the present study. To overcome this limitation, CLBF evaluations may be postponed, a strategy already adopted by others (
). Nonetheless, this decision of postponing CLBF exams will delay resynchronization of open females. Additionally, this would not overcome the incidence of false-positives associated with animals failing to respond to the TAI protocol; that is, that had had asynchronous ovulations. Finally, it is important to consider that some of the animals with long cycles could have been truly pregnant at the time of CLBF exams but had embryonic loss before pregnancy diagnosis at d 30, a relatively common condition in high-producing dairy herds (
). Our results suggest that Doppler imaging for the assessment of CLBF in the mid-luteal phase is an emerging technological advance in cattle reproductive management and is expected to play an important role to improve the productivity of herds with poor reproductive performance, particularly if early identification of luteolysis is associated with early resynchronization of nonpregnant females.
Evaluation of bred females using Doppler ultrasonography identified divergent patterns of CLBF between pregnant and nonpregnant cattle as early as 16 d post-TAI and anticipated the presumptive diagnosis of functional luteolysis by 24 to 48 h, compared with the gold standard (plasma progesterone). Additionally, evaluation with color Doppler on d 16 was a reliable indicator of the likelihood of pregnancy if females were sorted into quartiles according to their quantitative CLBF and adjusted CLBF; ROC curves indicated that these 2 vascular CL features were the best early predictors of pregnancy status. In contrast, luteal tissue area and echotexture were inconsistent early indicators of luteolysis and may be used for this purpose only close to d 20 post-TAI. We confirmed that luteal function is lost through an evident reduction in blood flow that takes place a couple of days before any detectable changes in CL morphology and echotexture.
The authors thank personnel from Embrapa's experimental farm (Coronel Pacheco, MG, Brazil) for technical assistance and animal handling. This research was supported by grants from FAPEMIG (Minas Gerais, Brazil; project CVZ APQ 02863/09) and Embrapa (Brazil; project 01.13.06.001.03.06).
Vascular control of ovarian function: Ovulation, corpus luteum formation and regression.
Reproductive performance of dairy cows resynchronized after pregnancy diagnosis at 31 (±3 days) after artificial insemination (AI) compared with resynchronization at 31 (±3 days) after AI with pregnancy diagnosis at 38 (±3 days) after AI.