Theriogenology 81 (2014) 407–418

Contents lists available at ScienceDirect

Theriogenology journal homepage: www.theriojournal.com

Characterization of follicle and CL development in beef heifers using high resolution three-dimensional ultrasonography Stephanie Scully a, Alex C.O. Evans b, Patrick Duffy a, Mark A. Crowe a, * a b

School of Veterinary Medicine, University College Dublin, Belfield, Dublin, Ireland School of Agriculture and Food science, University College Dublin, Belfield, Dublin, Ireland

a r t i c l e i n f o

a b s t r a c t

Article history: Received 2 April 2013 Received in revised form 1 October 2013 Accepted 11 October 2013

The aim was to characterize dominant follicle (DF) and CL development through the estrous cycle of cattle using three-dimensional (3D) ultrasonography while making a comparison with conventional two-dimensional (2D) B-mode ultrasound (US) and to relate the measures taken to systemic concentrations of steroid hormones and gonadotropins. After synchronization of estrus, the ovaries of crossbred beef heifers (N ¼ 5) were assessed using daily US with a GE Voluson i US scanner until the end of the first follicle wave, then every other day until emergence of the final (ovulatory) wave, when daily US resumed until ovulation. Follicle and CL growth were recorded and mapped. Measures of diameter (2D) and volume (3D) of the DF from the first and ovulatory waves of the cycles; and CL development were captured and stored for further analysis. Blood flow to the DF and CL were assessed using 3D power Doppler US measuring vascularization index (VI; %), vascularization flow index (0/100) and flow index (0/100). Jugular blood samples were collected every 24 hours for progesterone from the first estrus until the second ovulation. Concentrations of estradiol (E2) and follicle stimulating hormone (FSH) were measured every 8 hours from estrus to second follicle wave emergence; then, E2 only was measured from final follicle wave emergence until ovulation. Data were analyzed using PROC MIXED and PROC REG in SAS. Dominant follicle blood flow tended to decrease during follicle wave emergence and DF VI increased (P < 0.05) 24 hours before ovulation after peak E2. Measures of the DF and CL volume (3D) were highly predictive of 2D diameter measures throughout the cycle (P < 0.0001). Predictive values (r2) for day of wave emergence and day from ovulation were similar for 2D and 3D measures; however, 2D measures had higher repeatability when compared with 3D measures. There was no relationship between CL VI and progesterone early in the cycle (r2 ¼ 0.12; P ¼ 0.1); however, there was a strong positive relationship approaching ovulation (r2 ¼ 0.77; P < 0.0001). In conclusion, 3D power Doppler measures of blood flow appears to be representative of vascular changes in the DF and CL throughout the estrous cycle. However, the extra time required to acquire and analyze a 3D image and the relatively little additional information obtained over that achievable with 2D imaging in terms of follicle and CL development might preclude its widespread use other than for detailed research purposes. Ó 2014 Elsevier Inc. All rights reserved.

Keywords: Follicle CL 3-D power doppler ultrasound Blood flow Cattle Estrous cycle

1. Introduction * Corresponding author. Tel.: þ353 1 7166255; fax: þ353 1 7166253. E-mail address: [email protected] (M.A. Crowe). 0093-691X/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.theriogenology.2013.10.012

The use of ultrasound imaging for the noninvasive examination of reproductive organs in cattle has contributed to

408

S. Scully et al. / Theriogenology 81 (2014) 407–418

substantial advances in biology over the past two decades [1–3]. Conventional B-mode imaging was first used in cattle in the 1980s [4]. It has since played an integral role in reproductive management of cattle. The characterization of bovine follicular waves and pregnancy diagnosis using ultrasound has been particularly important [5,6], giving rise to the development of improved synchronization protocols and a new understanding of follicular dynamics during the estrous cycle [7]. Before the availability of ultrasound examination it was only possible to observe the ovaries after surgery or slaughter and so the proposed existence of waves of follicular growth could not be verified [8]. It is now well known that in cattle, two or three follicle waves occur during each estrous cycle, identifiable by the emergence of a cohort of follicles (>5 mm), that is stimulated by a transient increase in follicle stimulating hormone (FSH) concentrations [9,10]. The growing cohort secrete inhibins and estradiol (E2) that decrease FSH concentrations to basal levels and during this period, selection of a dominant follicle (DF) occurs [10]. The follicle undergoing selection preferentially develops LH receptors in its granulosa cell layer and benefits from enhanced insulin-like growth factor-I bioavailability [11]. Ovulation of the DF is dependent on LH pulse frequency and amplitude which is typically not sufficient for ovulation early in the cycle during the luteal phase [12]. However, during the follicular phase, LH pulse frequency increases, allowing for final maturation and ovulation of the DF [10,13]. An adequate supply of blood to the DF is vital for its survival and development and an increase in blood flow to the follicle is known to be associated with ovulation [14]. Survival of the CL is dependent on successful angiogenesis after ovulation and then maintenance of the vascular network to allow for sufficient progesterone (P4) secretion [15], therefore blood flow should be identifiable in corpora lutea during the luteal phase. Two-dimensional (2D) B-mode ultrasound is only capable of allowing visualization of external features of an organ whereas, three-dimensional (3D)/four-dimensional (4D) imaging can reveal internal features, and possibly provide new information about the organ of interest [16]. Threedimensional imaging was first developed in the 1990s and is now a standard tool in obstetrics and gynecology used routinely for the diagnosis of various clinical problems such as fetal abnormalities [17] and tumor growth [18] in humans. A 3D ultrasound examination produces a series of still volumes that can be displayed in any plane after the examination. Real-time 3D imaging, also referred to as 4D imaging, incorporates the fourth dimension of time meaning that the volume data set of a region of interest can be stored and later visualized as one single dynamic image [16]. Recent animal studies have adopted the use of 3D/4D imaging in their research, successfully characterizing events such as equine and elephant fetal development [19,20] and pregnancy diagnosis in cats and dogs [16]. Image acquisition using 3D/ 4D technology has improved with the introduction of high frequency ultrasound transducers, allowing volume data sets to be produced much faster which might help to overcome commonly reported problems of breathing artifacts and animal movement. The aim of this study was to characterize DF and CL development in beef heifers using 3D/4D imaging

technology and to establish the relationship between 3D and 2D measurements to assess the potential advantages of 3D imaging compared with the standard B-mode techniques in bovine research. 2. Materials and methods 2.1. Animals and treatments All experimental procedures involving animals were licensed by the Department of Health and Children, Ireland. Protocols were in accord with the Cruelty to Animals Act (Ireland 1876) and the European Community Directive 86/609/EC and were sanctioned by the Institutional Animal Research Ethics Committee in University College Dublin. All animals were housed indoors on straw bedding for the duration of the experiment and were offered ad libitum access to a diet consisting of 50:50 maize silage (Dry matter (DM); 344 g/kg, Crude Protein (CP); 76 g/kg, and Metabolisable Energy (ME); 11.8 MJ/kg DM) and grass forage (DM 239 g/kg, CP 101 g/kg, and ME 11.1 MJ/kg DM) in addition to 8-kg concentrates (DM 883 g/kg, CP 281 g/kg, and ME 12.9 MJ/kg DM) per day. The estrous cycles of crossbred beef heifers (N ¼ 8), predominantly Charolais and Limousin breeds aged between 18 and 24 months, were synchronized using an 8-day Controlled Internal Drug release (CIDR 1.9 g; Pfizer) intravaginal device with administration of a PGF2a analogue (2 mL; Estrumate, Schering Plough Animal Health, Hertfordshire, UK) equivalent to 0.5 mg cloprostenol 1 day before CIDR removal. Heifers were observed for signs of estrus four times per day commencing 30 hours after CIDR removal and only those recorded in standing estrus within the next 24 hours (Day ¼ 0; N ¼ 5) were used. 2.2. Ultrasound examinations Transrectal ultrasound scanning began on Day 0 (i.e., day of estrus) and was carried out daily throughout the first follicular wave then every other day until emergence of the ovulatory wave when daily scanning was resumed until ovulation. Follicle wave emergence was determined as the day when a cohort of between five and 10 follicles larger than 5 mm emerged on the ovaries. Three-dimensional power Doppler and 2D images of the DFs and CL during the first follicular wave and approaching ovulation were captured and stored for further analysis. Examinations were carried out using a Voluson i portable ultrasound system (GE Healthcare, Vienna, Austria) equipped with a convex RNA 5-9 probe which has 144  90 field of view. Identical preinstalled instrument settings (color gain, pulse repetition frequency, color-power, wall motion filter, frame rate) were used during all examinations. Examinations were carried out by the same operator at the same time (between 9 AM and 11 AM) each day, taking on average 30 minutes per animal. Heifers were sedated with a low dose of xylazine hydrochloride (0.2 mL per animal; Chanelle Pharmaceuticals Manufacturing Ltd., Galway, Ireland) which was diluted in 5 mL of Norocaine (Norbrook Laboratories Ltd., Newry, Northern Ireland) and administered as

S. Scully et al. / Theriogenology 81 (2014) 407–418

an epidural treatment at each ultrasound examination to minimize heifer movement while acquiring a 3D image. A map of the follicles present on each ovary was recorded at each ultrasound examination to help in identifying the beginning and end of each follicular wave and to help in retrospectively identifying the DF of each follicle wave. Follicle diameter was measured using integrated, electronic calipers on the ultrasound machine. Three still images of the DF at its maximum diameter were measured. Three diameter measurements were calculated for each follicle and a mean calculated. An average of the mean of three follicle diameter measures was calculated to determine follicle diameter at each time point. The mean of these three measures of follicle diameter and CL diameter were used for statistical analysis and for the production of all figures and tables. 2.3. Three-dimensional and power Doppler imaging During the ultrasound examination, the ovaries were first visualized in 2D B-mode. The “angio-mode” was switched on; the mobile sector appeared and was set up to cover only the region of interest. The 3D facility was engaged by switching to “volume mode”. The volume sector angle was preset to 90 and the fast volume acquisition (low resolution) setting was selected to avoid artifacts. The duration of the volume acquisition was between 4 and 6 seconds. At this stage, it was important that the animal remained as still as possible while the 3D images of the ovary were acquired. The resultant multiplanar display was examined to ensure that the whole ovary had been captured in the volume acquisition. The operator attempted to acquire three of these volume image sets for each animal on each scanning day. As the CL and DF of interest grew, separate volume measurements were taken of each structure to ensure a full volume set was acquired. The 3D volumes were stored on the hard disk of the ultrasound machine and later transferred to a personal computer for analysis. Each of the 3D images captured for the DF and CL were measured at each time point in individual animals. The mean of three measures of the DF and CL were used for statistical analysis and production of all relevant tables and figures. 2.4. Three-dimensional image analyses All analyses of stored ultrasound volumes were carried out off-line by the same operator using a personal computer. The virtual organ computer-aided analysis (VOCAL Version 7.0; GE Healthcare) imaging program was used to calculate the volume and vascularity of the DF and CL. The acquired volumes yielded multiplanar views of the ovaries (Fig. 1A and B). All calculations were done using these multi-planar images. The longitudinal view was used as the reference image. The rotation step was selected at 15 , resulting in the definition of 24 contours of the DF and CL. Contours of the DF and CL were manually drawn using the “manual mode” (allowing for any irregularities to be outlined) in all 24 sections using the computer mouse. When all contours were drawn, the volumes of the follicle (Fig. 1B) and CL (Fig. 1A) were calculated automatically.

409

The stored ultrasound volume obtained using 3D power Doppler sonography was defined by voxels (smallest unit of volume). Gray-scale voxels are 3D gray-scale information grades from black to white, the lowest value (intensity) being 0 and the highest 100 (g0–g100). A similar scale was used for color values (c0–c100). Using these values, the histogram facility of VOCAL software generated the vascularization index (VI), vascularization flow index (VFI), and the flow index (FI). Vascularization index was calculated as color values (color voxels þ gray values), VFI was calculated as weighted color values (color values þ gray scale values), and FI was calculated as weighted color values (color values). The following are the descriptions of the equations used for calculation of blood flow indices. The VI is a measure of the percentage of the structure from which a 3D power Doppler image has been captured that is vascularized relative to the overall volume. The FI is the amount of blood flow through the vessels of interest. The VFI is an intermediate measure incorporating the VI and FI in one single parameter:

P100 VI ¼ P100

c¼1 hcðcÞ P þ 100 c¼100 hcðcÞ

g¼1 hgðgÞ

P100

VFI ¼ P100

c¼1 c$hcðcÞ P100 c¼0 hcðcÞ

g¼1 hgðgÞ þ

P100 c$hcðcÞ FI ¼ Pc¼1 100 c¼1 hcðcÞ where c indicates color value (0–100), g grey scale value (0–100), hc the frequency of a color value, and hg frequency of a gray scale value. 2.5. Blood samples and analyses Blood samples were collected by jugular venipuncture into vacutainers (Becton Dickinson, Plymouth, UK). Concentrations of P4 were measured in serum samples taken from Day 0 to ovulation (at 24-hour intervals). Circulating E2 and FSH concentrations were measured in samples collected from Day 0 until the second follicle wave emergence (at 8-hour intervals) and from final wave emergence until ovulation (at 12-hour intervals). After collection, blood samples were refrigerated (4  C) for 12 to 24 hours before being centrifuged at 1500  g for 20 minutes at 4  C. Concentrations of P4 were measured using a commercially available solid-phase RIA (Coat-ACount Progesterone, Diagnostic Products Corp., Los Angeles, CA, USA) as previously described [21]. Intra-assay coefficients of variation (CV) were 11.4%, 4.6%, and 6.4% for low, medium, and high P4 concentration serum pools, respectively. The inter-assay CV for the same quality control serum pools were 16.9%, 9.8%, and 6.5%, respectively. The sensitivity of the P4 assay was 0.03 ng P4/mL serum. Concentrations of FSH were quantified using a previously validated RIA [22]; the inter-assay CV for low, medium, and high FSH concentration serum pools were 6.3%, 5.7%, and 7.6%, respectively, and intra-assay CV for the same quality

410

S. Scully et al. / Theriogenology 81 (2014) 407–418

Fig. 1. Multiplanar display of (A) CL obtained using three dimensional ultrasound on Day 4 after ovulation and (B) the dominant follicle of the ovulatory wave 1 day before ovulation. Longitudinal section is shown in the upper left quadrant, transverse section is shown in the upper right quadrant, and the coronal section is the lower left quadrant. The resultant vascular area of the CL and dominant follicle is shown in the lower right quadrant.

control pools were 10.8%, 9.1%, and 16.7%, respectively. The sensitivity of the FSH assay was 0.07 ng FSH/mL serum. Circulating concentrations of E2 were determined using RIA after the extraction method [23] using the Adaltis MAIA E2 kit (Biostat, Stockport, UK). The interassay CV for low and high E2 concentration serum pools were 18.5% and 19.1%, respectively. The intra-assay CV for the same quality control pools were 10.8% and 16.8%, respectively. The sensitivity of the E2 assay was 0.2 pg E2/mL serum. Peak circulating P4 concentrations were calculated as the mean of the two greatest circulating P4 concentrations. 2.6. Statistical analyses All statistical analyses were carried out using SAS (SAS Institute Inc., Cary, NC, USA). Data were checked for

normality. Data for days from follicle wave emergence and days from ovulation were analyzed separately. During the period from first follicle wave emergence, a Box-Cox transformation [24] was used to normalize the distribution of P4, E2, FSH, DF diameter, DF volume, DF VFI, and DF FI. During the period approaching ovulation the same transformation was used to normalize P4, E2, DF diameter, DF volume, DF VI, DF FI, and CL VFI. The effect of day of wave emergence or days from ovulation for all variables such as DF and CL measurements and blood hormone concentrations were determined using mixed models with animal as a random effect. A compound symmetry covariance structure for the data based on Akaikes’ information criterion values was used. The transformed data were used to calculate P values. The effect of day and animal was included in the final model where

S. Scully et al. / Theriogenology 81 (2014) 407–418

411

Fig. 2. Characterization of (A) dominant follicle diameter (2D), (B) dominant follicle volume (3D), (C) E2 and FSH (diamonds indicate E2 and squares indicate FSH) during initial wave emergence and approaching ovulation in beef heifers. Follicle diameter (2D) and volume (3D) increased during the first and ovulatory wave (P ¼ 0.0001) and approaching ovulation (P < 0.05). Concentrations of FSH increased (P ¼ 0.0001) before emergence of the first follicle wave. Concentrations of E2 increased (P ¼ 0.0001) between Days 2 and 3 of the first wave. Concentrations of E2 increased acutely (P ¼ 0.0001) during the ovulatory wave approximately 48 hours before ovulation of the ovulatory follicle. 2D, two-dimensional; 3D, three-dimensional; E2, estradiol.

significant (P < 0.1). The relationship between 2D and 3D measures of the DF and CL was determined using linear regression. For presentation of these data the transformed data were used to calculate r2 values. The nontransformed values were used for illustrative purposes. The reliability (the ability of a test to give the same result for a single observer) of the measures for volume, diameter, and vascularity were assessed using interclass correlation coefficients (ICCs). 3. Results All but one heifer had three follicular waves; data for this animal were included in all analyses with the exception of follicle data during first follicle wave emergence.

first wave of follicle growth or approaching ovulation are illustrated in Figure 2A–C. During follicle wave emergence, diameter of the future DF (Fig. 2A) increased (P ¼ 0.001) as did its volume (Fig. 2B; P ¼ 0.0001). Diameter of the DF (Fig. 2A) increased (P ¼ 0.0001) and volume (Fig. 2B) increased (P ¼ 0.0001) from 5 days before ovulation. During follicle wave emergence circulating concentrations of E2 (Fig. 2C) increased between Days 2 and 3 (P ¼ 0.001) and between Days 7 and 8 (P ¼ 0.0001). There was a significant rise in E2 concentrations (Fig. 2C) 2 days before ovulation (P ¼ 0.001). Concentrations of FSH (Fig. 2C) were high before wave emergence then decreased (P ¼ 0.0001) as the DF increased in size. Concentrations of FSH began to increase from approximately Day 3 of the first follicle wave.

3.2. Follicle blood flow measures 3.1. Follicle measures, E2, and FSH concentrations Temporal changes in 2D and 3D DF measurements and circulating concentrations of FSH and E2 during either the

The changes associated with day of first follicle wave emergence and days from ovulation are illustrated in Figure 3A–C. Dominant follicle VI (Fig. 3A) decreased

412

S. Scully et al. / Theriogenology 81 (2014) 407–418

Fig. 3. Characterization of (A) dominant follicle VI (%), (B) dominant follicle FI (0–100) and (C) dominant follicle VFI (0–100) during initial wave emergence and approaching ovulation in beef heifers. Vascularization index refers to the percentage of vessels relative to follicle volume that are present, VFI is a measure of the amount of vessels while taking flow into account, and FI is an indication of blood flow intensity. Follicle VI decreased during wave emergence (P ¼ 0.05) then increased before ovulation (P ¼ 0.14), DF FI did not appear to change during wave emergence (P ¼ 0.6) or approaching ovulation (P ¼ 0.56). Dominant follicle VFI decreased during wave emergence (P ¼ 0.2) but began to increase approaching ovulation (P ¼ 0.02). DF, dominant follicle; FI, flow index; VFI, vascularization flow index; VI, vascularization index.

(P ¼ 0.05) during the first follicle wave. The ovulatory DF VI did not appear to change approaching ovulation (P ¼ 0.14). Measures for DF FI (Fig. 3B) did not change significantly during follicle wave emergence (P ¼ 0.6) or approaching ovulation (P ¼ 0.56). Although DF VFI (Fig. 3C) did not change during follicle wave emergence (P ¼ 0.2). The ovulatory DF VFI initially decreased and then began to increase as ovulation approached (P ¼ 0.02). 3.3. CL measures and P4 concentrations Temporal changes in CL development measured using 2D and 3D methods and serum concentrations of P4 during the early luteal phase and later luteal phase relative to follicle growth are illustrated in Figure 4A–C. During the first follicular wave CL tissue area (2D; Fig. 4A) and volume (3D; Fig. 4B) increased (P ¼ 0.0001 and P ¼ 0.0001, respectively). CL tissue area decreased (P ¼ 0.0001) as did

volume (P ¼ 0.01) during the 5 days before ovulation that were measured. Concentrations of P4 (Fig. 4A) increased during the period of first wave emergence (P ¼ 0.0001) then decreased from 5 days before ovulation (P ¼ 0.0001). 3.4. CL blood flow measures Changes associated with first follicle wave emergence and days from ovulation are illustrated in Figure 5A–C. CL VI (Fig. 5A) was highest on Day 0 of wave emergence (i.e., Day 2.4  0.2 mean  SEM of the estrous cycle) after which time it began to decrease (P ¼ 0.0004). During the last 5 days approaching ovulation there was a further decrease (P ¼ 0.0001). CL FI (Fig. 5B) did not change during the first follicular wave (P ¼ 0.45) or as ovulation approached (P ¼ 0.09); CL VFI (Fig. 5C) decreased (P ¼ 0.02) during wave emergence, then continued to decrease from 4 days before ovulation (P ¼ 0.02).

S. Scully et al. / Theriogenology 81 (2014) 407–418

413

Fig. 4. Characterization of (A) CL tissue area (2D), (B) CL volume (3D), and (C) progesterone during initial wave emergence and approaching ovulation. CL tissue area was calculated from measurements of CL tissue diameter and excluded any fluid-filled lumen if present. CL volume was calculated from the 3D volume set captured using the ultrasound probe and excluded any lumen volume if present. CL tissue area and volume increased (P < 0.05) during the first follicle wave and decreased during the ovulatory wave approaching ovulation (P < 0.05). Progesterone concentrations increased (P ¼ 0.0001) during the first follicle wave and decreased during the ovulatory wave approaching ovulation (P < 0.0001). 2D, two-dimensional; 3D, three dimensional.

During the first follicle wave, circulating concentrations of P4 increased (P ¼ 0.0001). Serum concentrations of P4 then decreased to basal levels after luteolysis, from 5 days before ovulation (P ¼ 0.0001). 3.5. Comparison of 3D and 2D measures The relationship between volume and diameter for DF and CL measures is illustrated in Figure 6A and B. Follicle volume had a high predictive ability for follicle diameter throughout the cycle (r2 ¼ 0.84; P ¼ 0.0001; Fig. 6A). CL volume was also highly predictive of CL diameter (r2 ¼ 0.64; P ¼ 0.0001; Fig. 6B). The ICCs for 2D and 3D measures of the DF and CL during both periods of measurement are summarized in Table 1. Measures taken in 2D for CL tissue area had a higher correlation coefficient than 3D measures and 2D measures of the DF diameter had a lower ICC value than 3D follicle volume during the period of first wave emergence. However, DF diameter was highly repeatable compared with

DF volume during the period approaching ovulation. Blood flow indices for the CL had higher correlation coefficients during the early luteal phase relative to first follicle wave emergence than later in the luteal phase, approaching ovulation. Blood flow indices for the DF had poor correlation coefficients during both periods of measurements. Values never increased above 0.42 during first follicle wave emergence and never more than 0.1 approaching ovulation. 3.6. CL blood flow and P4 The relationship between CL VI and P4 concentrations is illustrated in Figure 7A and B. There was a poor relationship (r2 ¼ 0.12; P ¼ 0.1) between CL blood flow VI and P4 early in the cycle during formation of the corpus luteum (Fig. 7A). There was also a poor correlation between CL FI and VFI indices and P4 during the period early in the cycle, CL FI and VFI had r2 values of 0.01; P ¼ 0.0001 and 0.04; P ¼ 0.0001, respectively.

414

S. Scully et al. / Theriogenology 81 (2014) 407–418

Fig. 5. Characterization of (A) CL VI, (B) CL FI, and (C) CL VFI during initial wave emergence and approaching ovulation in beef heifers. Vascularization index refers to the percentage of vessels relative to follicle volume that are present, VFI is a measure of the amount of vessels while taking flow into account, and FI is an indication of blood flow intensity. CL VI decreased (P ¼ 0.0004) during the first wave and continued to decrease from Day 5 before ovulation (P < 0.0001). CL FI remained static during the first wave (P ¼ 0.45) then decreased from Day 5 before ovulation (P ¼ 0.09). CL VFI decreased during the first wave (P ¼ 0.02) and continued to decrease to basal levels from Day 4 before ovulation (P ¼ 0.02). FI, flow index; VFI, vascularization flow index; VI, vascularization index.

During the period approaching ovulation (Fig. 7B), CL VI became highly predictive of P4 concentrations (r2 ¼ 0.77; P ¼ 0.0001). CL VFI and FI were moderately predictive of P4 with values of r ¼ 0.45; P ¼ 0.0001 and r ¼ 0.45; P ¼ 0.0001, respectively. Table 1 summarizes the ICCs for measures taken using 3D and 2D imaging during the period of first follicle wave emergence and approaching ovulation.

dynamics; (2) 3D measures for DF blood flow might be more accurate than visual scoring for detectable blood flow which might be indicative of health and ovulatory status; and (3) CL blood flow did not show a strong positive relationship with P4 initially in the luteal phase but there was a strong positive relationship between these measures in the late luteal phase, as P4 concentrations declined.

4. Discussion

4.1. Relationship between 3D and 2D measures

This is the first study to characterize follicle and CL development during the estrous cycle using 3D ultrasound technology while also making a comparison with 2D ultrasound for the measurement of DF and CL size. The main findings from this study were that: (1) 3D imaging does not appear to offer any advantage over 2D measures for characterization of changes in DF or CL

The 3D and 2D imaging showed similar temporal changes for DF and CL growth throughout the estrous cycle [5,25]. Further investigation of interobserver correlation coefficients revealed a poor level of repeatability within the repeated measures for 3D measures compared with repeated 2D measures. This is despite the fact that for the 3D measurement technique each examination consisted of

S. Scully et al. / Theriogenology 81 (2014) 407–418

Fig. 6. Relationship between 3D (DF and CL volume) and 2D (DF and CL diameter) imaging processes for (A) dominant follicle size measures and (B) CL size measures throughout the cycle (includes data from the first and preovulatory follicle waves). There was a strong positive relationship (r2 ¼ 0.84; P < 0.0001) between 3D and 2D measures of DF size. A positive relationship was found (r2 ¼ 0.64; P < 0.0001) between 3D and 2D measures of CL size. 2D, two-dimensional; 3D, three-dimensional; DF, dominant follicle.

three serial volume data sets which were each calculated from 12 measurement planes. The use of 2D imaging for the measurement of ovarian structures is accepted as a viable method of measurement [10]. Results from the present study indicate that this is still the case and that 3D imaging is not an adequate alternative to 2D analysis of DF and CL size measurements in cattle. Previous work carried out a similar comparison in women and found a strong agreement between 2D and 3D measures of the ovaries [26]. Measures taken in 3D had a higher repeatability with correlation coefficients never falling below 0.9 [26–29]. Furthermore, studies have shown that volume measured using 3D imaging is more accurate than crude calculations of volume from 2D diameter measures [30]. In these studies, the women remained perfectly still allowing the best possible image to be captured. However, in the current study, even with the use of a low dose sedative, artifacts associated with breathing and animal movements were still present, and that might contribute to the poor repeatability measures. 4.2. Blood flow to the DF During the early to midluteal phase, the increasing concentrations of P4 ensures a low LH pulse frequency environment [13]. This means that the DF of the first follicular wave will undergo atresia. Initiation of DF atresia

415

begins before any detectable change in follicle size [31]. Previous visual scoring of the DF in the first wave of lactating dairy cows describes an increase in detectable blood flow signals to the DF after deviation [32]. In the present study during the first follicular wave, DF VI decreased significantly during emergence of the first DF and DF FI and VFI remained unchanged. The difference here could be attributed to the nature of the 3D measurements which take into account the size of the follicle relative to blood vessels while also measuring rate of flow. The decreased blood flow is likely due to the absence of increasing concentrations of E2 along with the ovulatory surge of LH to help stimulate a further increase in blood flow [33,34]. Previous reports have demonstrated the expression of angiogenic peptides either in response to or in association with the LH surge in cattle [35]. Measurements of the largest subordinate follicle were not taken in this study but it is possible that follicle blood flow might reflect the fate of the first wave DF before onset of diameter deviation [36]. In the period approaching ovulation, blood flow to the DF appeared to undergo a similar decrease in blood flow as had been observed in the first follicular wave. However, after the preovulatory rise in E2 blood flow, indices tended to increase until ovulation. The increase in blood flow to the DF has previously been shown to be regulated by LH and E2 [37]. Doppler ultrasound studies of the DF before ovulation would suggest that the functional changes that take place in the DF before ovulation are closely related to a local increase in blood flow within the follicle wall [14]. Histological studies have shown the DF to have more vessels on the theca layer approaching ovulation compared with subordinate follicles [38]. Previous studies have shown that assessment of size of the largest follicle is not always indicative of its estrogen-P4 ratio and relative dominance [10]. Increased vascularity and vascular permeability are likely to be part of the process which gives the DF its developmental edge, allowing for more nutrients, gonadotropins, and growth factors to permeate into the theca and granulosa cells of the follicle [39,40]. In the current study, the DF of the first follicle wave did not experience any apparent increase in blood flow. In comparison, the DF that emerged during the period approaching ovulation did appear to experience an increase in the percentage of blood vessels. Therefore blood flow might be a viable method by which to assess the health or ovulatory capacity of a preovulatory follicle. 4.3. Blood flow to the CL Blood flow indices for the CL were highest during the early luteal phase while the first DF wave was emerging. This peak was then followed by a slight decline. After the initial decline, CL blood flow indices remained steady as the CL approached the static phase of growth. During this static phase of growth, CL diameter and volume, and concentrations of P4 increased. The poor relationship between CL blood flow and P4 during the static phase was likely due to this inverse relationship. Previous investigations of CL blood flow have found it to be poorly related to P4 concentrations and CL size during the early and midluteal

416

S. Scully et al. / Theriogenology 81 (2014) 407–418

Fig. 7. Relationship between CL blood flow and progesterone (A) during initial wave emergence and (B) from Day 5 before ovulation. No relationship was found between CL VI and progesterone during the first follicular wave (r2 ¼ 0.12; P ¼0.1) while a strong positive relationship was found from Day 5 before ovulation (r2 ¼ 0.77; P < 0.0001). VI, vascularization index.

phase [41,42]. For the present study, total CL blood flow was calculated relative to CL size. Therefore during initial growth the proportion of CL area with associated blood flow relative to its size was quite high. As the CL increased in diameter and volume, the percentage of blood vessels relative to size decreased. During the period approaching ovulation, CL VI and VFI declined rapidly in association with decreasing P4 and CL size. Previous studies measuring the CL using 2D have reported similar profiles as the CL undergoes luteolysis [42].

Table 1 The ICCs for repeated measures of volume and VI for the follicle and CL during the days from wave emergence and during the days approaching ovulation. Variable

Day from wave emergence ICC

Day from ovulation ICC

CL volume CL diameter CL VI CL FI CL VFI Follicle volume Follicle diameter Follicle VI Follicle FI Follicle VFI

0.49 0.76 0.58 0.58 0.54 0.66 0.06 0.4 0.09 0.42

0.87 0.93 0.6 0.27 0.4 0.66 0.96 0.1 0.05 0.06

Abbreviations: FI, flow index; ICC, interclass correlation coefficient; VFI, vascularization flow index; VI, vascularization index.

There was a strong positive relationship between CL blood flow indices and serum concentrations of P4. A similar relationship has been proposed with regard to CL blood flow area suggesting that it would be an appropriate indicator of P4 concentrations particularly during luteal regression [42]. When measuring DF and CL blood flow using 2D power Doppler, vascularity was measured on one plane that was deemed representative of the vascularity of the entire structure [32,41,43]. This technique, although it provides useful information, relies heavily on the subjective impression of the examiner. It should be acknowledged that positive strides in the measurement of ovarian blood flow of domestic animals have been made over the past 10 years. Previous studies have attempted to quantify the vascularity to the developing follicle in cattle and horses [14,44,45]. However, measurements of DF and CL blood flow in 3D offers the potential to measure the vascularity of a structure in a more objective manner. This is especially the case for angiogenesis during growth of the bovine follicle that is quite unevenly distributed on different regions of the theca layer and so might be difficult to quantify on a 2D plane [38]. The assessment of VI was of particular interest because validation studies have proposed that it might be related to the number of vessels present [46]. In this study, CL blood flow volume indices had reasonable ICCs particularly during the early luteal phase then decreasing slightly during the period approaching ovulation. 4.4. Using the 3D probe The convex 3D abdominal probe used in the experiment was ideal for transrectal use in cattle and was used without any perceived discomfort to the heifers. The acquisition of 3D images was time-consuming, however, with most examinations lasting between 20 and 30 minutes even with the use of a sedative. The use of sedation during the study was not expected to have had any effect on measurements of DF or CL blood flow data [47,48]. 4.5. Conclusions In summary, the results from this study give the first characterization of DF and CL development using 3D ultrasound techniques during follicle wave emergence and the early luteal phase and approaching ovulation during the late luteal phase in heifers. The main findings suggest that 3D imaging is not a sufficient replacement for 2D imaging to characterize changes in DF and CL dynamics throughout the estrous cycle considering the extra labor involved in the capture of the 3D images. However, results are largely in agreement with previous research regarding the temporal changes in follicle and CL blood flow during the period of follicle wave emergence and during the period approaching ovulation. CL measures were slightly different than those in previous studies likely because of the measurement of CL blood flow volume as opposed to measurement of blood flow from a 2D color image. The role of increased blood flow to the DF might be an important regulatory factor after selection and during differentiation and continued growth

S. Scully et al. / Theriogenology 81 (2014) 407–418

toward becoming a preovulatory follicle [32] and might be a method to assess its health status.

[22]

Acknowledgments [23]

Funded by Science Foundation Ireland (07/SRC/B1156). The authors thank the staff at UCD Lyons Research farm for care and maintenance of the experimental animals. They also thank Ms. Penny Furney for assay of FSH, estradiol, and progesterone concentrations in serum samples. References [1] Ginther OJ, Knopf L, Kastelic JP. Temporal associations among ovarian events in cattle during oestrous cycles with two and three follicular waves. J Reprod Fertil 1989;87:223–30. [2] Ginther OJ, Utt MD. Doppler ultrasound in equine reproduction: principles, techniques, and potential. J Equine Vet Sci 2004;24: 516–26. [3] Ginther OJ, Silva LA, Araujo RR, Beg MA. Temporal associations among pulses of 13,14-dihydro-15-keto-PGF2alpha, luteal blood flow, and luteolysis in cattle. Biol Reprod 2007;76:506–13. [4] Pierson RA, Ginther OJ. Ultrasonography of the bovine ovary. Theriogenology 1984;21:495–504. [5] Savio JD, Keenan L, Boland MP, Roche JF. Pattern of growth of dominant follicles during the oestrous cycle of heifers. J Reprod Fertil 1988;83:663–71. [6] Pieterse MC, Szenci O, Willemse AH, Bajcsy CS, Dieleman SJ, Taverne MA. Early pregnancy diagnosis in cattle by means of lineararray real-time ultrasound scanning of the uterus and a qualitative and quantitative milk progesterone test. Theriogenology 1990;33: 697–707. [7] Roche JF, Mihm M, Diskin M, Ireland JJ. A review of regulation of follicle growth in gattle. J Anim Sc 1998;76:16–29. [8] Ireland JJ, Roche JF. Development of nonovulatory antral follicles in heifers: changes in steroids in follicular fluid and receptors for gonadotropins. Endocrinology 1983;112:150–6. [9] Adams GP, Matteri RL, Kastelic JP, Ko JC, Ginther OJ. Association between surges of follicle-stimulating hormone and the emergence of follicular waves in heifers. J Reprod Fertil 1992;94:177–88. [10] Sunderland SJ, Crowe MA, Boland MP, Roche JF, Ireland JJ. Selection, dominance and atresia of follicles during the oestrous cycle of heifers. J Reprod Fertil 1994;101:547–55. [11] Lucy MC. The bovine dominant ovarian follicle. J Anim Sci 2007;85: E89–99. [12] Rahe CH, Owens RE, Fleeger JL, Newton HJ, Harms PG. Pattern of plasma luteinizing hormone in the cyclic cow: dependence upon the period of the cycle. Endocrinology 1980;107:498–503. [13] Forde N, Beltman ME, Lonergan P, Diskin M, Roche JF, Crowe MA. Oestrous cycles in Bos taurus cattle. Anim Reprod Sci 2011;124: 163–9. [14] Acosta T, Hayashi K, Ohtani M, Miyamoto A. Local changes in blood flow within the preovulatory follicle wall and early corpus luteum in cows. Reproduction 2003;125:759–67. [15] Schams D, Berisha B. Angiogenic factors (VEGF, FGF and IGF) in the bovine corpus luteum. J Reprod Dev 2002;48:233–42. [16] Hildebrandt TB, Drews B, Kurz J, Hermes R, Yang S, Göritz F. Pregnancy monitoring in dogs and cats using 3D and 4D ultrasonography. Reprod Domest Anim 2009;44:125–8. [17] Merz E, Weber G, Bahlmann F, Miric-Tesanic D. Application of transvaginal and abdominal three-dimensional ultrasound for the detection or exclusion of malformations of the fetal face. Ultrasound Obstet Gynecol 1997;9:237–43. [18] Hata T, Hata K, Noguchi J, Kanenishi K, Shiota A. Ultrasound for evaluation of adnexal malignancy: from 2D to 3D ultrasound. J Obstet Gynaecol Res 2011;37:1255–68. [19] Kotoyori Y, Yokoo N, Ito K, Murase H, Sato F, Korosue K, et al. Threedimensional ultrasound imaging of the equine fetus. Theriogenology 2012;77:1480–6. [20] Drews B, Hermes R, Göritz F, Gray C, Kurz J, Lueders I, et al. Early embryo development in the elephant assessed by serial ultrasound examinations. Theriogenology 2008;69:1120–8. [21] Forde N, Beltman ME, Duffy GB, Duffy P, Mehta JP, Ó’Gaora P, et al. Changes in the endometrial transcriptome during the bovine

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

417

estrous cycle: effect of low circulating progesterone and consequences for conceptus elongation. Biol Reprod 2011;84:266–78. Crowe MA, Padmanabhan V, Hynes N, Sunderland SJ, Enright WJ, Beitins IZ, et al. Validation of a sensitive radioimmunoassay to measure serum follicle-stimulating hormone in cattle: correlation with biological activity. Anim Reprod Sci 1997;48:123–36. Prendiville DJ, Enright WJ, Crowe MA, Vaughan L, Roche JF. Immunization of prepubertal beef heifers against gonadotropin-releasing hormone: immune, estrus, ovarian, and growth responses. J Anim Sci 1995;73:3030–7. Box GE, Cox DR. An analysis of transformations revisited, rebutted. J Am Stat Assoc 1982;77:209–10. Sirois J, Fortune JE. Ovarian follicular dynamics during the estrous cycle in heifers monitored by real-time ultrasonography. Biol Reprod 1988;39:308–17. Raine-Fenning NJ, Campbell BK, Clewes JS, Kendall NR, Johnson IR. The interobserver reliability of three-dimensional power Doppler data acquisition within the female pelvis. Ultrasound Obstet Gynecol 2004;23:501–8. Jokubkiene L, Sladkevicius P, Rovas L, Valentin L. Assessment of changes in volume and vascularity of the ovaries during the normal menstrual cycle using three-dimensional power Doppler ultrasound. Hum Reprod 2006;21:2661–8. Raine-Fenning NJ, Campbell BK, Clewes JS, Kendall NR, Johnson IR. The reliability of virtual organ computer-aided analysis (VOCAL) for the semiquantification of ovarian, endometrial and subendometrial perfusion. Ultrasound Obstet Gynecol 2003;22:633–9. Raine-Fenning NJ, Campbell BK, Clewes JS, Johnson IR. The interobserver reliability of ovarian volume measurement is improved with three-dimensional ultrasound, but dependent upon technique. Ultrasound Med Biol 2003;29:1685–90. Amer A, Hammadeh M, Kolkailah M, Ghandour AA. Three-dimensional versus two-dimensional ultrasound measurement of follicular volume: are they comparable? Arch Gynecol Obstet 2003;268: 155–7. Sartori R, Fricke PM, Ferreira JC, Ginther OJ, Wiltbank MC. Follicular deviation and acquisition of ovulatory capacity in bovine follicles. Biol Reprod 2001;65:1403–9. Acosta TJ, Hayashi K, Matsui M, Miyamoto A. Changes in follicular vascularity during the first follicular wave in lactating dairy cows. J Reprod Dev 2005;51:274–80. Acosta TJ. Studies of follicular vascularity associated with follicle selection and ovulation in cattle. J Reprod Dev 2007;53: 39–44. Mattioli M, Barboni B, Turriani M, Galeati G, Zannoni A, Castellani G, et al. Follicle activation involves vascular endothelial growth factor production and increased blood vessel extension. Biol Reprod 2001; 65:1014–9. Siqueira LC, dos Santos JT, Ferreira R, Souza dos Santos R, dos Reis AM, Oliveira JF, et al. Preovulatory changes in the angiotensin II system in bovine follicles. Reprod Fertil Dev 2013;25:539–46. Mihm M, Evans AC. Mechanisms for dominant follicle selection in monovulatory species: a comparison of morphological, endocrine and intraovarian events in cows, mares and women. Reprod Domest Anim 2008;43:48–56. Siddiqui MA, Ferreira JC, Gastal EL, Beg MA, Cooper DA, Ginther OJ. Temporal relationships of the LH surge and ovulation to echotexture and power Doppler signals of blood flow in the wall of the preovulatory follicle in heifers. Reprod Fertil Dev 2010;22:1110–7. Jiang J, Macchiarelli G, Tsang B, Sato E. Capillary angiogenesis and degeneration in bovine ovarian antral follicles. Reproduction 2003; 125:211–23. Berisha B, Schams D, Kosmann M, Amselgruber W, Einspanier R. Expression and localisation of vascular endothelial growth factor and basic fibroblast growth factor during the final growth of bovine ovarian follicles. J Endocrinol 2000;167:371–82. Zimmermann RC, Xiao E, Bohlen P, Ferin M. Administration of antivascular endothelial growth factor receptor 2 antibody in the early follicular phase delays follicular selection and development in the Rhesus monkey. Endocrinology 2002;143:2496–502. Luttgenau J, Ulbrich SE, Beindorff N, Honnens A, Herzog K, Bollwein H. Plasma progesterone concentrations in the mid-luteal phase are dependent on luteal size, but independent of luteal blood flow and gene expression in lactating dairy cows. Anim Reprod Sci 2011;125:20–9. Herzog K, Brockhan-Lüdemann M, Kaske M, Beindorff N, Paul V, Niemann H, et al. Luteal blood flow is a more appropriate indicator for luteal function during the bovine estrous cycle than luteal size. Theriogenology 2010;73:691–7.

418

S. Scully et al. / Theriogenology 81 (2014) 407–418

[43] Bollwein H, Lüttgenau J, Herzog K. Bovine luteal blood flow: basic mechanism and clinical relevance. Reprod Fertil Dev 2012;25:71–9. [44] Acosta TJ, Gastal EL, Gastal MO, Beg MA, Ginther OJ. Differential blood flow changes between the future dominant and subordinate follicles precede diameter changes during follicle selection in mares. Biol Reprod 2004;71:502–7. [45] Gastal EL, Gastal MO, Ginther OJ. Relationships of changes in Bmode echotexture and colour-Doppler signals in the wall of the preovulatory follicle to changes in systemic oestradiol concentrations and the effects of human chorionic gonadotrophin in mares. Reproduction 2006;131:699–709.

[46] Raine-Fenning NJ, Nordin NM, Ramnarine KV, Campbell BK, Clewes JS, Perkins A, et al. Determining the relationship between three-dimensional power Doppler data and true blood flow characteristics: an in-vitro flow phantom experiment. Ultrasound Obstet Gynecol 2008;32:540–50. [47] Reno R, Ginther OJ. Vascular perfusion of reproductive organs in pony mares and heifers during sedation with detomidine or xylazine. Am J Vet Res 2009;70:141–7. [48] Braun U, Fohn J. Duplex ultrasonography of the common carotid artery and external jugular vein of cows. Am J Vet Res 2005; 66:962–5.