a aE pe eT Tee eee CELLAR Lae LY, TET toh Lo, 1 ‘ ¢° kbd lad, tO) or lar a ine dd bh Su et ld a a eat ab a SL Duishsdet debedebubeded dedeh delle dodduat Ledskak lel Leia ert eye Le! rer eepeewe Pyoeeee enero Rerepere? } : wr . aie pe : , is pep A e3 GEDA eee os ee ot eee ae TPE ery PAAR S. Pe PTS CY - , > ° : A lh Meg r sre , +e 5 i yt Le Le ae ty dh . * P , Peet Tre y P , Lae et y a 3 ae cae * iia eee z ce rh : Fe r y : , 5 LEY 1 $ 4 iF i mF Bete sinha eae RADE per eete tee Pear bret Shes bebe eas Et RTI LE EIT STV aLe HL TUL Cae OES TELER EST TY LEER OC EERE TLL GERTLER OTA LLP Eber LIA . napa be pe me mph ere Pe ela : eaten weg ; ; y ; $ p ; é Bch. 3 Saat BS 0. , mua ne 5 ae nee SEES wv rotor Le seGabaVeaeeeee : , ; F EPC PT Peer es ba tae Sl Hee aL SULER EL ee Do: Cr he. Be Ste cae Ee ee Ce Teo. ota SISPAVAPR SMD SSP ASEM HII FIIae PIES PEPe PCr Pal rahul te run atau rae wre rr ees asbedibnbeh oping gare ae ian GA ODSRRD HONE IN OS SHAT TED se saz 7 Tee t DLS COTS DE Taha ty eee Ths hana oe , ETE SL bb te eed bl ddd Pee et a oa Pe ee ee a Le ae ps UNIVERSITY ey PENNSYLVANIA LIBRARIES ; ; ' ULTRASONIC IMAGING and REPRODUCTIVE EVENTS IN THE MARE by O. J. GINTHER, V.M.D., Ph.D. Professor of Veterinary Science Department of Veterinary Science University of Wisconsin - Madison Madison, Wisconsin 53706 Jean Austin duPont Veterinary Medicine Library Me viet eal Rod es Senet Square, PA 19348-1 691 Copyright © 1986 by Equiservices All rights reserved 3rd printing Published and distributed by Equiservices 4343 Garfoot Road Cross Plains, WI 53528 USA Telephone: 608-798-4026 Library of Congress Catalog Card No. 85-91070 Printed in the United States of America m ‘ 2 _ : € * - a j o « ; ’ ‘ every equine veterinarian who ever said, "I wish I could see what's going on in there!" nag Preface No one was prepared --- nor could they have been. The proliferation of ultrasound scanners in equine theriogenology in the early 1980s was unexpected. The availability of a real-time noninvasive technology for visualizing internal structure and content of the reproductive organs generated enthusiasm and high expectations. Purchase costs exceeded $20,000. Today costs exceed $10,000, which is still a considerable sum. Unless a durable, high quality scanner is purchased and reasonable use is made of its capabilities, a fair return on investment may not be attained. This book should help practitioners and researchers make an appropriate decision regarding purchase. Those who already have a scanner will be able to expand and sharpen their techniques. Even if one does not have access to a scanner, much information can be gained from this text; ultrasonic morphology is thoroughly integrated with gross anatomy, histology, physiology, endocrinology, and pathology. Moreover, recent research findings obtained with the help of ultrasonic imaging are reviewed. These findings add the dimension of dynamics to our understanding and appreciation of the beauty of equine reproductive function. It will be seen that dynamics --- active motions and forces --- are essential in the functioning of the reproductive tract. The book is intended for research scientists and graduate students as well as clinicians and veterinary students. Except for introductions, the book progresses from one expanded legend to another without the use of intervening text. It is, in effect, an elaborate slide show. This format eliminates the nuisance and discontinuity associated with searching for tables and figures. Acknowledgments Our laboratory functions without the help of a full-time technician. Procedures and chores --- sophisticated to menial --- are done by graduate students with the help of undergraduate student helpers. Appreciation is expressed to the many student helpers who assisted us with the experiments that used the ultrasound technology. Some of the experiments reviewed in this text were done during thesis research. The following graduate students made extensive use of ultrasound scanners: Greg Adams, Don Bergfelt, Chris Bessent, David Cross, Sandy Curran, Karen Hayes, John Kastelic, Gayle Leith, Roger Pierson, Stan Scraba, and David Townson. Thanks are also extended to Lisa Buckley and Joyce Jackson for illustrative work, Roger Pierson and Mark Sumner for preparation of photographs, Ellen Porath for editorial help, Adrianna Rademakers for histology preparations, and Willow Press, Middleton, Wisconsin, for careful and expeditious printing. The research summarized in this book would not have been possible without the loan of ultrasound scanners from several companies. Fred Chavez, formerly with Fischer Imaging, and now with Equisonics, Inc., got us started with the loan of a Fischer Vetscan (Equisonics 210) and customized photographic equipment. Owen Parker of Equisonics has provided continuing consultation and service. David Tharpe of Pitman-Moore, Inc. and Don Zahorik of Bion Corporation also provided scanners for the research effort. Roger Pierson deserves special thanks for providing hard-copy skills. He also prepared Chapter 7, which presents needed information on preparing ultrasound photographs and videotapes. One of the ultrasound scanners and the animals and facilities for the majority of the research were provided by EquiCulture, Inc. CONTENTS Part One PRINCIPLES me APTER 1 --- INTRODUCTION... 4222.0 = see ee 1.1 An Overview .........:252. >. senate 9 1.2 US€S... cc ce ep eee onan coos) eee 1.3 A Revolutionary Research Tool... >. . 3. sage 1.4 History ......2.-2- 409s apa tees SHAPTER 2 --- WAVES AND ECHOES ...............--<-.-- 2.1 Origin of Sound .....s36°2 99922. ,..5... 2.2 Propagation and Anatomy of Sound Waves................. 2.5 Receptiomof Echoes .~”.i34 2.9.5 2 oe ee 24 Attenuation. ..2..-42.4:-3424454 555700 2.» Production of Pulses t29i¢23555.-..3.4,,......2 eee mEAPTER 3 --- THE TRANSDUCER .......-. o42.5. 3547) 3.1 Types of Transducers . oo... 2204 aeanGiuet 5 3 3.2 Sequential Firing of Linear-amay Transducers - 5-3 ee 4 3.3 Beamsand Real-time ....,......:.4. + s55,09Ge 3.4 Beam Resolution: piesb4c55-35 8 ee 3.5. Near Field.and Par Field... .. - s |. ea tg 3.6° Focusing: sss3ee5eeoumeee 55 75s aes 1 See 3.7 Dynamic Focusite@sc 02.4 aaa oa ee ee ee 3.8 Production of Echo Sienals , 225544... 3 50 ee —) Gy Ua be 13 14 15 18 19 21 23 24 20 28 30 31 31 32 Be x Contents CHAPTER 4 --- SIGNAL PROCESSING ............ccccce0. Be APMIS OL A SCAMMET ons. eee ec ccscccceeccsvsevecs ee eS ARAE Paes our WA Ag teete oS o y os bes es cee cos Be ABIG Mics eoGall COOMVEUIET oc cass ecesscnescurssvevecnsvuess oe ET Ly od oc on bene denc ensues s cabeecnsc, oie MOE 5 roy os ces ecbeveh ch sevsusbcssvecteuess 4.6 Scanning Lines 4.7 The Net Result ee eRe eTee eee Caeser ec eee ee seecaer tates ve ee Par esi S —— INTERPRETATION ....55600ccce cc eceaes en eee RE, MSEC es a's veg wr cara ob bso bes Ke ccc cece, 5.2 Nonspecular Reflections Pe MMe PSUR sw oc Fis a reek ovo hk csc eek, 5.4 Enhancement Artifacts 5.5. Reverberation Artifacts 5.6 Beam-Width Artifacts 5.7. Other Artifacts =-@ PSP eCeCeeCePeCT eee eC ae eee aes et i ae tS OP PePeSPT etree ewe ecececeae Castres a ae £4.45 8. 6S OA SO SMe’ SHSESEC HE CAEP AR SARS SCARE BK OEE D6 OE OES =e eee eRe ST See Oe CSSA ARE THRACE HACE S e Part Two INSTRUMENTS AND TECHNIQUES CHAPTER 6 --- INSTRUMENTATION ..............2.20.¢ eee Meee MU UNNOETS 4 3 Pees ens es oes ne G2 - Geiection of Transducer Frequency . 2025 80.8 2002. A Sree See I ees Ve ee ee ed es geal eee ee at oe oe 6.3B Contrast, brightness, calipers, and annotations ......... oe sire ON ¢ 6 2 oi tai re ee ee ete Mee 6 5 5s 5 5 kk hk BO LeE LY. Spee eer ON 5 5955555 2k aS SEES. . d: GA Selectine a Make and Model... ..- 0050202. SF od. =e bis casa PAMAAY Contents xi SHAPTER 7 --- HARD COPY .20..05. ois 0554 ee 87 7.1. Polaroid Photography ... 2.20.45 ie 88 7.2 Negativé-Based Photography 2.uu ass ge 91 7.3 Projection Slides . . 15.40.1550 ko 5 ale 93 14. Videotapés”. &/2vi1.5 CRI O SS). . elas oe 94 75 Other Hard Copy Methods... .. .. 20s <35 54 9a oF BUAPTER 8: =---TECHNIQUES: 9.54 od a5 ee 99 8.1 Orientation of Reproductive Tract .........,1..950a0e ee 100 8.2 The Coupling: Meditint,... on550 14,5) oo 103 8.3. Centralized Examining Areas .._....5.. 70.90 ee 104 8.4 Examining Technique .....,......4..05.9. ee 106 8.5 Water Bath, Biopsy, and Transabdominal Techaiques...i7.5.43 107 8.6 Transrectal Approach in Other Species .................... 109 Page bie OVARIES AND UTERUS SeXAPTER 9 --- OVARIES (22 Woo he... ee 115 9.1 Gross Anatomy (7y.2. 2722995... 116 9.1A Attachment and orientation ..__....-..... ee 116 9.1B Relationship of cortical and medullary areas........... 117 9.1C_ Follicles and luteal:siands. 2-55. 3 54, se 118 9:2. Fommand Functions. 9). 2 ae «ig Seat eee 120 9.2A Estrouscyele! iy. 2... «. + - 20.3 fee 120 9.2B Pregnancy. <...caatyinsl% gical ee 122 9.2C . The uterine luteolytic mechanism 5.55 58 124 9.2D:; Anovulatory season. .....+..4-...+, 4 eee 126 9.3. . Ovarian Scanning Techniques. ....15.. 55.9 128 9.4 Ultrasonic Detection of Abnormalities .. 25 45.9.55.050 130 xii Contents (ia Ae eR Irae BASIC LIS wee ec a cc ee eee se ecco eees 13: Ae Winsome nnmomy OF POLICIES... 6.00 e ee i eect 13 10.2 Reliability of Ultrasound Follicular Measurements ........... 135 IGS Uiirasound Studies of Folliculogenesis .. 2.0.00. 0000.0. wae 136 EES SN See K ED EET EVEL TUES Se Owe eee ea eel 136 iints tise GyCie VETSUS PICANANCY Bs. ee ee es nee 139 ies toca adet OVINANON OL (he yea v2 se cwecescccecees 14) ee fecansune pending (yalabon 4. Fe ee es dae 142 OVO Rei iod i Cee Cle Tae seth ee cod sat tea ws oe os 148 Was seecuns aie Opiingl Breeding Time... ee ee does 150 BAIRE “OBeAAS VINO Vek 66a SUNT E ea EES EN eee ee oo ok 152 en Ae 2 ae AL GUANDS 80. ee es PEN as 155 Lage Mec ceri AS Re ee eae ina 1S¢ 11.2 Ultrasonic Anatomy .........22. 2... cece e cece eee e eee 158 11.3. A Study of Two Types of Ultrasonic Morphologies........... 16] ee RO BANC OT EVCONANICY oa n'e'e-o 0's oe ns aie pw ala ee wag een owe 166 11.5 Applications of Luteal Detection and Evaluation............. 167 11.5A Determining if mare has entered the ovulatory season... 167 Ba okaey A Pt tar ed CISCO 2 os os xc as os no es be eos 168 11.5C Estimating the day of the estrous cycle ............... 169 ne ee a a ee ie eek oe 2 er 173 Rae re Ne 5 5 3 4b 4 Ns ee Ree ee er ene 174 Se eC ie aCe ECE Sek chee ee eae eter a eee 176 igo -ilivsome musiomy OF ie Carvin oo ee 177 74 Uiitasomic Atiaiomy of the Ulerds Fis ee ee 178 Lee Ree ae CCIE ECHOES Oy 0 LP PT PI a Pi 182 Rees | San ee TI 55a, hc hy aes hee EEE OES Soe es we 183 12:7 « Siadies of Cyclic Ultrasonic Changes... 000000. avec eae 183 Pe er OEY FI PO, PO SF ee v8 os tes 188 Rs eS PI rg ee EEE Le OE TEN: CRI ce oe 188 12.8B Intraluminal collections of fluid.................... 191 ae Weta rs Pod yy AIR ek BEA TS A Ps 192 Contents xiii Part Four SINGLETONS CHAPTER 13 --- THE SINGLE EMBRYO ................. 13,1 First Detectiowe:.. ..a29y 5 ee 13.2 Location soos. ak A ea 13.3. Growth Profile im@ross: Section 23559215901 a. 13.4. Shape of the: Vesicles: sasuselegia7. sca 13.5 Growth of Embrya 02.0.2. 295.. 94 205 2 13.6 Days.9 fo16.~ >... 22205 Tees oe a 13.7. Days 171019: . 222) ee ee 13.8. Days:20.to 223! seg 103 ae ee ee 13.9. Days: 24 and 25:2. sn oe ee ee 13.10 Days 28:10:33" ccs bso ieee oy 0G 13.11 Day 360s ne eee. i eh ee 13.12 Days 40 f050) seek. ose oe aces 13.13 Review of Ultrasonic Anatomy over Days 16 to 50 ........... 13.14 Predicting Day of Presnancy, 922-2... ..-. 13.15 After Day 50 nam... 2 ona s sae set CHAPTER 14 --- EMBRYO-UTERINE INTERACTIONS ...... 14.1 Characteristics of Embryo Mobility ....9( 3373. ee 14.1A Effect. of day of presnancy 2 253... es. 14.1B Examples of mobility: 2355 522).22 9-3 14.1C Progressive nature of mobility 43" .44934). sae 14.1D: Expansion and contraction of vesicle 43555 A 14.2 Control of Vesicle Mobility .......4) j@seeeee 14.3 Role. of Embryo Mobility .._.. 35226 sy:g3 ee 14.4 Fixation 3i5.8 os woasonges bog ere 14.4A Day of occurence saciahisias eee ee 14.4B: Site of fixation , ,:, 2200355003 ssgneutee oe 14.5 UterinePones3: 2,35. 3 a ee ee ee 14.5A Descriptionivcs 2yaeScink so to. ee Gee 145B: Relationship.of toneto fixations. ane, Se. 14.5C Control of utetine fone... ...43.......,.3 > ee 14.6 Orientation -.....-....3..:.+55........-.. xiv Contents oop amine Oe Om TV NIC LASS cc tt ce tee ase ceenes 25 IS RG Siw dees sdensssaersrenreserernsés 254 Se Fe Oe el rc tls i eae rere a ara re 25 15.3 Factors Associated with Early Pregnancy Loss .............. 256 ieee REMAND 0a ew los EDC Sa be Feo w le teh’ -: 15.5 Pseudopregnancy following Embryonic Loss ............... 258 19,0 Huibtyonic Loss dimns Days 11 10.15... 2 ees SVU oe 260 15.6A Size and location of the embryonic vesicle ............ 260 15.6B Intraluminal fluid collections versus embryonic loss .... —2¢ a4 Linpiyour Loss during Days 151020. ier. ie 266 3.0 limon Loss during Days 25 tod0...... 60. aoe oo ei ia. A.anuay OF induced Emorvonic Loss... 6. bee AEE ot 27: 15.9A Induction of embryonic loss on Day 12 .............. 273 15.9B Induction of embryonic loss on Day 21............... Bi 15.9C Induction of embryonic loss on Day 30............... 27 Part Five TWINS CHAPTER 16 --- TWINS: ORIGIN AND DEVELOPMENT. .... 287 id pce Donbic Ovulations G00. VG Ai..... 288 16.2 Factors Affecting Incidence of Double Ovulations ........... 290 Se ent A se PEE URIS Fv vel ae v's 290 Ae iis oan dG Ys Pees cove ed brl soak 291 OES et ee OITING HIATUS , 5 0X0 CU Te ewe Oe i are 292 16.3 Double Ovulations and Pregnancy Rate.................05. 293 0 eae aney calculations 200270500. Wa edn en 293 eo vncironous ovulations .. .. 225602019 Ba. PA Ed 5k os 294 ies Asveckronous ovolations: 2..4..... 0806.8 Ged. ni. 295 16.3D Number of ovulations versus embryos ............... 295 16.3E Patterns of ovulations versus number of embryos ...... 296 Contents xv 16.4 Interplay between the Uterus and Twin EMmDLyOs >, sss 55.22 298 164A Mobility ........5t:.10usees 298 16.46 Fixation... 0. c1.2s0nsae stone 302 16.5 Embryo Reduction ........1..55.,.. 303 165A Time of occurretice .....-....5505 303 16.5B Pre-mobility embryo reduction ..:................. 304 16.5C Embryo reduction during Days 11to 40 ............. 305 16.5D Nature of unilateral embryo reduction............... 306 16.5E. Fetalteduction ....2.2.......1.1,.. SEZ CHAPTER 17 --- TWINS: MANAGEMENT AND CORRECTION ae 17.1 Management of Double Follicles and Ovulations I'7.1A Past practices ..2-...3.5.4),.... a 17.1B A suggested ultrasound approach 17.2 Detection of Twin Conceptises ........95).25500 ee 17.2A During mobility phase and on day of fixation.......... 17.2B After fixation .............,..3...., 17.3 Artificial Correction of Twin Embryos 17.3A During the mobility phase 17.3B After fixation OCHS CTH ECGCECBESCAH COS Part Six BIOEFFECTS AND SUMMARY See xPTER 18 --- BIOEFFECTS .....>5...........2 0 18.1 The Cell as a Potential Vareet 52) ee So saseee oe ee 18.2 Risks versus Benefits ......,2.........,. = 18.3. Minimum Intensities for Biological Effects................. 18.4 Recent Reports of Possible Side Effects.................... 18.5 .BioeffectsimMares ...:.... 25435 xvi Contents CHAPTER 19 -- SUMMARY 6. ic eee Cee eee cece cee es 343 Uo MON oie emacs es eneseesessssseesueees a re 343 19D Waves ad ECHOES .. 0... ccc cw cece ees sect eweseneeses 344 $55 Tie WAMSOUCET oe ee ere See e dese recess 346 19.4 Signal Processing .......... cece eee cece recente e cece: 347 19.5 Interpretation .......... cee eee e cece cree een e erences 348 19.6 tees. nw Pe ee re teens 350 19,7 Paar Cay oa ais 2 FI 0d a os ete e ong 351 PTS, PR on sae os in a d's TEU 8 eee ows 352 er ge ee A ee ee 352 SR) WSS Fi Fs hE DIN ea ee SN TGS 353 163 ft fated Glands: is 2 ss 80 Ee oo 355 AE tee Se as was ea VOUS. FEES CA 9 os ev ew ae ss 357 19.13 The Single Embryo. ............ ccc cece cece eect cence: 359 19.14 Embryo-Uterine Interactions ........--e eee eee eee eee 361 19,15 Embryonic Loss. 2.00. 055 sec eee ee ce ee eee e en enes 363 19.16 Twins: Origin and Development........-.-.0-e +e eeeeeees 366 19.17 Twins: Management and Correction ........--++ee+eeeee 368 1 Te ORION nk con co oo ORS AD A Ses Fs was 369 ResCT INDEX ... cowboy. eG yt oe eS wis 7% 37 Part One PRINCIPLES Peer Chapter 1 INTRODUCTION Gray-scale diagnostic ultrasonography is the most profound technological advance in the field of large-animal research and clinical reproduction since the introduction of transrectal palpation and radioimmunoassay of circulating hormones (author's opinion). We now have a method for non-invasive visualization of the internal anatomy of the reproductive organs and their payload --- the conceptus. Dramatic events can be visualized while they are occurring and apparently without interference. A high point in the author's career occurred when a Day-14 equine embryo approached a cyst in the middle of the uterine body, encroached upon it with appropriate distortion of the conceptus, squeezed past the cyst, continued moving in the same direction until it reached the cervix, and then reversed its direction. This event happened during a six-minute span and was observed and photographed as it occurred. It is not surprising to hear veterinarians say that ultrasound scanners are the spice in their brood-mare practice. They're fun. But ultrasonography is a complex and expensive technology in which operator and scanner must interact to produce a useful and pleasing image. The first major thrust of this report is to provide a thorough working knowledge of the instruments so that the operator-scanner interaction is optimal, and the financial investment receives maximal return. Once suitable images of an area of interest are obtained, the next step is interpretation --- relating the ultrasonic information to the tissues and thereby reaching a conclusion. The second major thrust therefore involves the principles of image interpretation, including differentiating between artifactual and representative echoes. Third, the integration of the technology into clinical and research programs is given considerable attention. The possibility of biological side effects of diagnostic ultrasound must not be ignored, and a chapter is devoted to this topic. An underlying knowledge of the physics and terminology of ultrasound is needed not only in the attainment and interpretation of high-quality images, but also in the evaluation of reports and discussions on side effects. — 2 Chapter 1 1.1 An Overview Depiction of the manner in which the ultrasonic beams of a hand-held intrarectal transducer sample a cross-sectional "slice" of uterine horn. Transrectal, diagnostic ultrasound uses high-frequency sound waves to produce images of soft tissues and internal organs. Because of the large size of horses, the transducer can be held in the rectum directly over the organs of interest, as shown. Electric current is applied to crystals in the transducer, producing vibrations characteristic of the crystals and resulting in sound waves. The operator directs the sound waves through the tissues by moving or varying the angle of the transducer as desired. The short distance from rectal wall to viewing area allows the use of high- Introduction 3 frequency scanners that produce images with much detail. The sound beams that pass through the tissues are quite thin (e.g., 2 mm), and a thin "slice" of tissue is sampled, as shown. The two-dimensional image seen on the screen is analogous to a histological section. Tissues have different abilities to either propagate or reflect sound waves. The proportion of the sound wave that is reflected is received by the transducer, converted to electric impulses, and displayed as an echo on the ultrasound screen. The characteristics of various tissue interfaces determine what proportion of the sound wave will be reflected. The reflected portion is represented on the ultrasound image by shades of gray, extending from black to white. Liquids (follicular fluid, yolk-sac fluid) do not reflect sound waves and are said to be nonechogenic or anechoic; therefore, the image of a liquid-containing structure appears black on the screen. At the other extreme, dense tissues (cervix, fetal bone) reflect much of the sound beam and appear white on the screen. Such tissues are said to be echogenic. Other tissues are seen in various shades of gray depending upon their echogenicity or ability to reflect sound waves. Certain tissue formations may cause the sound waves to bend or bounce back and forth or to become weakened or entirely blocked. Therefore, artifacts appear on the screen and will challenge the interpreting abilities of the ultrasonographer. Modern ultrasound instruments for examining the equine reproductive tract are B-mode, real-time scanners. B-mode refers to brightness modality, in which the ultrasonic imaging is a two-dimensional display of dots. The brightness of the dots is proportional to the amplitude of the returning echoes. Real-time imaging refers to the "live" or moving display in which the echoes are recorded continuously, and events such as fetal leg movements and heartbeat can be observed as they occur. Some scanners have videotaping capabilities so that the moving images can be preserved. They may be played back through the ultrasound scanner or a television set. The moving image also can be "frozen" to facilitate measurements or photographic reproduction. The ability to study "living," detailed, sequential, sectional views of the organs demands a working knowledge of anatomy much more extensive than that required for rectal palpation. Veterinarians and animal scientists specializing in large animal reproduction are entering an era in which they must be knowledgeable not only in gross and histological anatomy and pathology, but also in ultrasonic anatomy and pathology. They must be able to relate the image on the viewing screen to normal and abnormal form and function. 4 Chapter 1 ndietiebiacinnniadidieeeen ne Leet Or | keke Examples of ultrasound images from the reproductive tract of mares . The images are of an ovary (A), estrous uterus (B), and Day-45 conceptus (C). The images were taken with a 5 MHz linear-array transducer and are intended as 21 introduction to gray-scale imaging. Note the various shades of gray, extending from black to white. The shade of gray is dependent on the proportion of the sound wav that is reflected back to the transducer --- the more reflected, the lighter the shade o* gray. The reflecting surfaces are called tissue interfaces. An interface resul's wherever tissues of different density are in contact. Fluid-filled structures (follicle allantochorion) do not reflect sound waves and are said to be non-echogenic. Fluid ‘s therefore dark or black on the images. The various tissues are white or shades of gray depending on their echogenicity (ability to reflect sound waves). The interfaces of denser structures are more reflective and therefore lighter on the images. The ultrasonic slice of the ovary is delineated by arrows and has a corpus luteum (cl) and two 8 to 10 mm follicles (f). The corpus luteum is beginning to regress and is very dense and therefore highly echogenic. The image of the cross-section of a uterine horn (delineated by arrows) was taken during estrus. The edematous endometrial folds are prominent. The center of each fold is more reflective and therefore relatively echogenic (light gray or white), whereas the outer portions are relatively nonechogenic (dark gray or black). The allantochorionic fluid of the Day-45 conceptus is black (nonechogenic), whereas the various tissues are shades of gray. f = fetus. uc = umbilical cord. Introduction 5 1.2 Uses When first introduced into theriogenology, ultrasound scanners were used primarily for early pregnancy diagnosis, detection of twins, and photographic documentation of pregnancy in mares. As a result of recent equipment modifications and evaluations, it is becoming clear that ultrasound scanners are destined for a broader, more fundamental role in equine clinical and research programs in reproduction. Detailed information of the many uses of ultrasound in evaluating and monitoring reproductive structures and events in mares is given in later chapters. The following is a summary of some specific uses: 1. Determining whether a mare has entered the ovulatory season. 2. Determining whether a filly has reached puberty. 3. Detecting and differentiating single and double preovulatory follicles and ovulations. 4. Diagnosing failure of ovulation. 5. Monitoring development, maintenance, and regression of the corpus luteum. 6. Differentiating persistence of the corpus luteum from anovulatory or anestrous conditions. 7. Estimating the stage of the estrous cycle by consideration of follicles, corpus luteum, and endometrium. 8. Estimating the extent of estrogen exposure at the endometrium. 9. Evaluating the time and suitability for breeding. 10. Detecting semen in the uterus. 11. Evaluating postpartum uterine involution. 6 Chapter 1 12. Evaluating the suitability of a mare to serve as an embryo-transfer recipient. 13. Differentiating pseudopregnancy from pregnancy. 14. Early detection of the embryo (Day-11). 15. Detecting twins and manually eliminating one. 16. Diagnosing embryonic death at the time of occurrence (absence of heartbeat). 17. Diagnosing ovarian pathological conditions such as peri-ovarian cysts, ovarian tumors, and anovulatory hemorrhagic follicles. 18. Diagnosing pathology of tubular organs such as hydrosalpinx, pyometra, uterine cysts, small collections of intraluminal uterine fluid, fetal debris. 1.3 A Revolutionary Research Tool Animal and veterinary scientists are only beginning to realize the tremendous potential of ultrasonography as a research tool. Totally unsuspected discoveri:s already have been made, and the technology is being used for conventional scientific testing of hypotheses. Many research areas now being investigated by the use of ultrasonography were once largely inaccessible. Following are some recent research discoveries in equine reproduction that were made possible by ultrasonography: 1983 Dynamic interactions between embryo and uterus (embryo mobility, fixation, and orientation) and the underlying physical controlling mechanisms (1,2,3,4). 1984 Dynamic interactions between twin embryos (5). Natural embryo reduction of twin embryos during the mobility phase and after fixation (6). Introduction 7 Sequential changes in follicular populations (follicles >2 or 3 mm) (7). 1985 Changing morphology of the preovulatory equine follicle (8). Incidence, changing morphology, and role of the corpus hemorrhagicum (9). Use of ultrasonic morphology of endometrial folds as an instant biological indicator of estrogen levels (10). Nature of early embryonic loss, including documentation of expulsion through cervix (11). 1.4 History A brief outline of the early history of the development of diagnostic ultrasound (12,13) is given below: 1880 Discovery of the piezoelectric effect. 1940s Refinement of SONAR; SOund NAvigation and Ranging. Ultrasonic detection of flaws in metals. SONAR equipment used to demonstrate echoes deep within body tissues. 1950s Echo-encephalography and echocardiography. Water-path scanner. 1960s Contact scanner. Two-level (black and white) images. 8 Chapter 1 Early 1970s Electronic scan converter and gray-scale imaging. Late 1970s _ Real-time ultrasound. Without the discovery of the piezoelectric properties of certain crystals, diagnostic ultrasonography would not have evolved. It is this property that permits the conversion of electric current to ultrasound waves and the subsequent conversion of the mechanical energy of echoes into electric current. The research activities necessitated by World War II led to the refinement of SONAR, in which ultrasound waves were used to detect submarines. This principle was then slowly adapted to the detection of tissue reflectors. The earliest systems for abdominal and pelvic imaging used a water-path scanner, a medical extension of the SONAR technique in which ‘he patient was seated in a tank of water. The submerged transducer moved in a circle around the patient. The contact scanner represented an important advance, because the transducer could be applied directly to the subject without passage of the sound waves and echoes through a water bath. A layer of gel between the skin surface and the transducer served to eliminate air, which would have blocked the passage of tlie ultrasound waves. Older scanners in the 1960s used a system in which the wide range of echo amplitudes was compressed into two levels, so that various parts of the ima ze were either black or white. This approach was limited because it defined only major differences in tissue densities. The development of gray-scale imaging in the early 1970s was a major advance. Many amplitudes of echoes were represented by levels of the gray scale. The signals were stored in a scan converter and then displayed on a television monitor. In the past few years, the original analog converters were largely replaced by digital scan converters, which store the echo information (location, amplitude) in a memory similar to that used in personal computers. Tue digital scan converters were more stable and more resistant to electronic noise. Another major breakthrough was real-time or dynamic imaging, which became available in the late 1970s. This approach allowed the operator to observe movements as they occurred (e.g., heartbeat, moving fetal limbs). Among other things, the viability of an embryo could be established immediately. Introduction 9 Transrectal, diagnostic, ultrasonic imaging in horses has a very short history, as shown in the following list: 1980 First report of intrarectal imaging of reproductive tract of mares (14). Early Rapid increase in the use of ultrasound by equine 1980s_theriogenologists (15,16). 1983 Publication of detailed descriptions of ultrasonic anatomy of the equine embryo (1). Use of ultrasound for research hypothesis testing (1,2). Short-course on use of ultrasound to evaluate the equine reproductive tract. Introduction of 5 MHz intrarectal linear-array transducer for large animals (17). 1984 Publication of detailed descriptions of ultrasonic anatomy of equine ovaries (7). Publication of ultrasonic anatomy and pathology of the equine uterus (18). 1985 Publication of ultrasonic morphology of embryonic death in horses (19). 1986 Textbook on ultrasonic imaging of reproductive tract in mares. In 1980, one hundred years after the discovery of the piezoelectric effect, Palmer and Driancourt described the use of a hand-held intrarectal transducer and gray- Scale scanner for monitoring reproductive events in mares (14). Their pioneering work opened the way for the increasing use of real-time ultrasonography on brood- 10 Chapter 1 mare farms. Several reports given at the 1982 International Symposium on Equine Reproduction (15,16) subsequently generated excitement over the potential of the technique. In 1983, research and clinical reports began to appear that describec in detail the ultrasonic anatomy and pathology of the reproductive tracts of mare (1,7,18). The interest in the technology soon led to the development of a short course in equine reproductive ultrasonography at Colorado State University. Sufficient interest and information is now available (1986) to justify a text and reference book on the narrow subject of transrectal diagnostic ultrasonography of the reproductive tract of mares. Although the diagnostic ultrasonography described in this text uses the B-mode for two-dimensional imaging, two other modes are available for study of soft tissue (20). The A-mode (amplitude mode) produces a one-dimensional display of echo amplitudes for various depths and is in widespread use for evaluating the fat and lean portions of meat animals. The A-mode also has been used for pregnancy diagnosis in ewes. In the late 1970s, A-mode machines were overpromoted for diagnosis of pregnancy in mares. The unsuitability of these machines for this purpose resulted in some resistance to the real-time, B-mode scanners when introduced to the veterinary market in the early 1980s; prospective buyers tended to recall their unfavorable experiences with the A-mode machines. The M-mode (motion mode) form of imaging is an adaptation of the B-mode and is used for evaluating moving structu‘es of the heart. Doppler ultrasound systems are used to monitor fetal heartbeat and blood flow in large vessels; this system has been used for pregnancy diagnosis in large animals in the late fetal stage by detecting the enlarged uterine vessels. _ pte — © REFERENCES 1. Ginther, O. J. 1983. Fixation and orientation of the early equine concepius. Theriogenology 19:613-623. 2. Ginther, O. J. 1983. Mobility of the early equine conceptus. Theriogenology 19:603-611. 3. Ginther, O. J. 1984. Intrauterine movement of the early conceptus in barren and postpartum mares. Theriogenology 21:633-643. 10. el . a2. 3. 14, i. 16. Introduction 11 . Leith, G. S. and O. J. Ginther. 1984. Characterization of intrauterine mobility of the early conceptus. Theriogenology 22:401-408. . Ginther, O. J. 1984. Mobility of twin embryonic vesicles in mares. Theriogenology 22:83-95. . Ginther, O. J. 1984. Postfixation embryo reduction in unilateral and bilateral twins in mares. Theriogenology 22:213-223. . Ginther, O. J. and R. A. Pierson. 1984. Ultrasonic anatomy of equine ovaries. Theriogenology 21:471-483. . Pierson, R. A. and O. J. Ginther. 1985. Ultrasonic evaluation of the preovulatory follicle in the mare. Theriogenology 24:259-268. . Pierson, R. A. and O. J. Ginther. 1985. Ultrasonic evaluation of the corpus luteum of the mare. Theriogenology 23:795-806. Hayes, K. E. N., R. A. Pierson, S. T. Scraba, and O. J. Ginther. 1985. Effects of estrous cycle and season on ultrasonic uterine anatomy in mares. Theriogenology 24:465-477. Ginther, O. J. 1985. Embryonic loss in mares: Incidence, time of occurrence, and hormonal involvement. Theriogenology 23:77-89. Winsberg, F. and P. L. Cooperberg. 1982. Real-Time Ultrasonography. Churchill Livingstone, New York City, NY. Athey, P. A. and L. McClendon. 1983. Diagnostic Ultrasound for Radiographers. Multi-Media Publishing, Inc., Denver, CO. Palmer, E. and M. S. Driancourt. 1980. Use of ultrasound echography in equine gynecology. Theriogenology 13:203-216. Simpson, D. J., R. E. S. Greenwood, S. W. Ricketts, P. D. Rossdale, M. Sanderson, and W. R. Allen. 1982. Use of ultrasound echography for early diagnosis of single and twin pregnancy in the mare. J. Reprod. Fertil. Suppl. 32:431-439, Chevalier, F. and E. Palmer. 1982. Ultrasonic echography in the mare. J. Reprod. Fertil. Suppl. 32:423-430. 12 Chapter 1 17. Ginther, O. J. and R. A. Pierson. 1983. Ultrasonic evaluation of the reproductive tract of the mare: Principles, equipment and techniques. J. Equine Vet. Sci. 3:195-201. 18. Ginther, O. J. and R. A. Pierson. 1984. Ultrasonic anatomy and pathology of the equine uterus. Theriogenology 21:505-515. 19. Ginther, O. J. 1985. Embryonic loss in mares: Incidence and ultrasonic morphology. Theriogenology 24:73-86. 20. Rantanen, N. W. and R. L. Ewing. 1981. Principles of ultrasound application i1 animals. Vet. Radiol. 22:196-203. Chapter 2 WAVES AND ECHOES The principles of ultrasonography are discussed in Chapters 2,3, and 4. The concepts developed in these chapters are based on reviews and books that were intended for ultrasonographers in human medicine. Additional resources included ultrasound physicists and specialists at the University of Wisconsin Medical School and electrical engineers at veterinary ultrasound companies. The concepts were modified where indicated, and the diagrams were designed to relate directly to ultrasound scanners being used to examine the reproductive tracts of horses. In this chapter, background information is developed on the nature of ultrasound waves and echoes. This is done by comparing the origin, traveling, and echoing of audible sound in air with the origin, traveling, and echoing of ultrasound in tissues. The origin of audible sound from a drumhead, traveling of the resulting waves through air, echoing of the waves from a mountain side, reception of the echoes by the ear drum, and processing of the ear-drum vibrations by the internal auditory system are compared with the origin of ultrasound from transducer crystals, traveling of the resulting waves through tissue, echoing of the waves from tissue reflectors, reception of the echoes by the transducer crystals, and processing of the crystal vibrations by the ultrasound scanner. The published sources for the concepts developed in this chapter were references 1 through 6. 14 Chapter 2 2.1 Origin of Sound AUDIBLE ULTRASOUND SOUND Drum head Crystal RESTING i Tred Ai Ti rrp eaet STATE = omlatules totes ile’ ae Striking AD ACTIVATION ey Alternating electric pulses Compression DISTURBANCE OF MOLECULES Rarefaction (Decompression) A comparison between the origin of audible sound waves and tie origin of ultrasound waves. In this example, audible sound is produced by a diaphragm stretched over a hoop, as in a drumhead. During the resting state or equilibrium, the drumhead is still. The adjacent air molecules are evenly dispersed and maintain an equilibrium due to mutually attracting and repelling forces. However, when the drumhead is activated by striking, it vibrates and disturbs the adjacent air molecules. As the drum head moves in one direction, it causes compression of the molecules. When the drum head moves in the opposite direction, it creates a void. The adjacent compressed molecules rush into the void, resulting in Waves and Echoes 15 an area of decompression or rarefaction. In comparison, diagnostic ultrasound originates from crystals that have piezoelectric properties. Piezoelectric means literally "squeeze-electric" and is pronounced pi é’z0. The crystals expand and contract when subjected to an electric current and, conversely, produce an electric current when compressed by returning echoes. A crystal is activated by an electric charge, which causes expansion or contraction of the crystal according to the alternating polarity of the electric signal. Crystal expansion causes compression of neighboring tissue molecules, and contraction causes rarefaction similar to the response of air molecules near the drumhead. 2.2 Propagation and Anatomy of Sound Waves EXPANDING AND VIBRATING CONTRACTING DEPICTION DRUM HEAD CRYSTAL OF WAVES e®. »£ 2) Sen Secoes Sy oper ee) oe ee eee eee —_o2 oo. ee Se as? = ° o A ~ — Oo oD pao Oe @ | Oo. 05 — ; = @ o fo es > | o © Oo | oc. QO > Hm) 0 0 £.- -Y¥--—----——> Amplitude Longitudinal vibration of molecules Comparative propagation or traveling of audible sound waves and ultrasound waves. The disturbance of air molecules caused by the vibrating drumhead, in turn causes a disturbance of neighboring molecules. In this way, the 16 Chapter 2 disturbance, consisting of alternating compression and rarefaction, passes through the air away from the drumhead. It is the disturbance of the air molecules that is propagated -- not the molecules themselves. The air molecules vibrate back ar forth; these oscillations are longitudinal in the direction of the sound wave. Similarly, ultrasound waves are propagated through tissue by the alternatin: expansion and contraction of the piezoelectric crystal in response to the alternatir polarity of the electric signal. The association between the propagated waves o compression and decompression and the conventional method of diagrammatical! depicting waves is shown. Amplitude refers to the strength or power of the sound waves and is similar volume or "loudness" in audible sound systems. Decibel is a logarithmic unit fo measuring amplitude. The amplitude of echoes from soft-tissue reflectors varic from zero to approximately 60 decibels. Intensity refers to the rate of flow energy through a unit area in association with propagation of the sound wave ai involves the rate of molecular vibration. Intensity is measured in watts per squa centimeter. Intensity measurements are often used in discussions of possib biological effects of ultrasound. Amplitude and intensity are directly related. A wavelength is the distance encompassed by an area of compression and the accompanying area of rarefaction and is depicted as a sine wave. Frequency refers to the number of vibrations or oscillations of the sound source (drumhead or ultrasound crystal) per second. It is therefore identical to the number of wavelengt!is or cycles that pass a given point in the medium per second and to the number of vibrations made by the molecules in the medium per second. Frequency is measured in hertz (Hz) units. One hertz is one cycle per second and a megahertz (MHz) is one million cycles per second. Ultrasound is defined as any sound with a frequency of more than 20,000 Hz. Diagnostic ultrasound uses frequencies of | to 10 MHz, and recently designed equipment for detailed study in ophthalmology uses frequencies of 10 to 25 MHz. Speed or velocity is the time required for a wavelength to pass a given point. Speed is determined by the characteristics of the medium (elasticity and density). In soft biological tissue, speed averages approximately 1540 meters/second, excluding bone (4080 m/sec) and lung tissue (600 m/sec because of air). That is, ultrasound waves travel through tissue at approximately 1.5 millimeters in one millionth of a second. Although these complexities and characteristics should be kept in mind, sound waves will be depicted as simple arrows in some of the diagrams which follow. — ~ mh US Ud -_—— ww ae — hy w Ke Waves and Echoes 17 Type of sound Medium Wavelength Frequency Speed Audible Air 2 to 2000 cm 20 to 20,000 Hz 330 m/sec Diagnostic ultrasound Tissue Less than 1 mm 1 to 10 MHz 1540 m/sec Differences in the characteristics of audible sound and diagnostic ultrasound. The sound waves resulting from a drumhead (audible sound) and those resulting from an ultrasound crystal were depicted in the above illustration on the Same scale for didactic purposes. However, the characteristics of the two types of sound differ profoundly, as shown in the table. In addition to the listed differences, it is noteworthy that high-frequency ultrasound waves, unlike audible sound waves, tend to behave like a directed beam rather than spreading out during propagation. THis important characteristic allows the ultrasonographer to precisely locate small teflectors deep within the tissue; two small reflectors can be differentiated only when the space between them is greater than the width of the beam (Section 3.4). The heterogeneous nature of the medium for ultrasound (tissue), in contrast with the homogeneous nature of the medium for audible sound (air), causes extensive interaction with the passage of ultrasound waves. As described in subsequent sections and chapters, it is these interactions that are fundamental to diagnostic ultrasonography. 18 Chapter 2 2.3 Reception of Echoes AUDITORY SYSTEM ULTRASOUND SYSTEM Sending Receiving Sending Receiving Ear Drum head drum ae Crystal Crystal ee —— == == = —— we «ew «ew —~jd Lens electric crystal Viewed from end Viewed from side of transducer of transducer Methods of focusing. Focusing narrows a portion of the beam profile and thereby increases the amplitude of the echoes from reflectors at a certain depth. Beams may be focused in the thickness plane to give better lateral resolution at a given depth. This is done by using curved crystals (internal focusing) or by placing an acoustic lens beneath the crystals (external focusing), as shown. By modifying the element-firing sequence, beams can also be focused in the width plane (electric focusing). The example shown uses the firing sequence of 1-7, 2-6, 3-5, and 4. 32 Chapter 3 Thus, some scanners provide lateral focusing in two dimensions -- the width of the beam is focused by the firing sequence and the thickness by lens or curved crystals. It is important to select a transducer that is focused closest to the depth of interest. With higher frequency transducers, the focal region is closer to the transducer; however, there is some latitude to this rule and a 7.5 MHz transducer, for example, can be designed to have a focal region comparable to that of a5 MHz transducer. For example, the focal distance of transducers for one veterinary scanner used for intrarectal evaluation of the reproductive tract are as follows: 3.5 MHz, 70 mm; 5 MHz, 35 mm; 7.5 MHz, 20 mm. As an approximation, the effective depth for high-quality imaging is equal to two times the focal distance. A 3 MHz transducer is more suited for studying the large postpartum uterus or a large fetus by either external or intrarectal placement of the transducer. A 5 MHz transducer is more suited for detailed transrectal study of the reproductive tract or early conceptus. 3.7 Dynamic Focusing DYNAMIC FOCUS Short Medium Long focus focus focus Equivalency of a single dynamic focus with three fixed focuses of different depths. Dynamic focusing is a form of electric focusing that results in varying focal distances rather than the fixed distance described in Section 3.6. This is accomplished by systematically varying the delay in the firing sequence, which in turn progressively alters the curvature of the beam front and therefore the total focal distance. Such focusing may be switch-selectable, emphasizing detail at a selected depth of interest only. However, in some scanners, dynamic focusing is designed to occur automatically, according to the distance traversed by the returning echoes. The Transducer 33 3.8 Production of Echo Signals Time: 0 ———— 20 ps —> 40 us ——> 80 Urs First signal Second signal Fire elements 1to 7 = beam #1 —_—--—-— —_ 4 Rectal wall | | / / \ / Relationships among locations of reflectors, amplitudes of echo signals, and time interval between signals. In this simplified example, a 31 mm follicle is located 31 mm from the transducer. A pulse is produced by firing the first seven elements. As the pulse travels through the tissues, it conforms to the confines of the imaginary beam. For simplicity, it is assumed that there are only two reflectors in the field and that the speed of sound is a uniform 1.54 mmj/microsecond. The pulse reaches the follicle in approximately 20 microseconds (31 mm x 1 us/1.54 mm). When the pulse strikes the first reflector (upper surface of the follicle), part of the pulse is reflected back as the first echo, and the remainder continues on as a transmitted pulse. The first echo reaches the transducer at approximately the same time (40 us) that the transmitted pulse reaches the far wall of the follicle, because the two distances traversed are equal (62 mm). When the echo strikes the quiescent crystals of the transducer, an electric signal is produced. The transmitted pulse is weaker than the original because of attenuation at the first reflector. Therefore, the resulting second echo and the corresponding echo signal are weaker than the first. As shown later for this example (Section 4.2), the disparity in amplitude between the two echo signals will be corrected by manual adjustments of the receiver. The associated time intervals serve to assign the proper spatial relationships for the two reflectors on the ultrasound screen (Section 4.6). Note the consistency in time intervals according to the distances traveled, whether as a pulse or the resulting echo. 34 Chapter 3 REFERENCES 1. Zagzebski, J. A. 1983. Properties of ultrasound transducers. In Textbook of Diagnostic Ultrasound. Ed. S. Hagen-Ansert. C. V. Mosby, St. Louis, MO. 2. Zagzebski, J. A. 1983. Pulse-echo ultrasound instrumentation. In Textbook of Diagnostic Ultrasound. Ed. S. Hagen-Ansert. C. V. Mosby, St. Louis, MO. 3. Bartrum, R. J. and H. C. Crow. 1983. Real-time Ultrasound. 2nd Ed., W. B. Saunders Co., Philadelphia, PA. Chapter 4 SIGNAL PROCESSING The intrarectal transducer functions as a remote unit connected by a coaxial cable to the main body or console of the ultrasound scanner. The console contains the components needed for both activating the transducer crystals at a predetermined rate and receiving and processing the returning echo signals. The console contains the controls used by the operator and the main center of attention --- the viewing screen. This chapter discusses the functions of the components of the console. Consideration is given to signal amplification and the adjustment of the gain controls, the memory storage system in analog scan converters and in the more modern digital scan converters, the nature of the B-mode echo-display system, and the transfer of information in the scan converter to the echo display screen including the relationships between scanning lines and raster lines. As in earlier chapters, the presentation is a development of concepts based on published reports for medical ultrasonographers (1 through 5) with modifications appropriate to the veterinary ultrasound scanners that are being used for transrectal imaging of the reproductive tract of mares. 36 Chapter 4 4.1 Components of a Scanner TRANSDUCER Long cord (e.g., 3 meters) Receiver L_-—.|- Amplifier ~ Myf sseitae Digital scan converter Near field Components of a linear-array ultrasound scanner. The pulser or transmitter provides electric signals at a predetermined rate for driving the piezoelectric crystals of the transducer. The pulsating rate is such that the display of echo signals on the screen appears to be continuous. Echoes from tissue reflectors are received by the piezoelectric crystals of the transducer. The echoes are initially processed by the receiver, where the echo signal is amplified and compensation is made for loss of intensity due to attenuation. This initial handling of the signals Signal Processing 37 occurs before storage and is called preprocessing. The degree of amplification is called gain and is comparable to volume in the control of audible sound. The concept of gain must be thoroughly understood because proper adjustment of the gain knobs is one of the most important variables under the continuous control of the veterinary ultrasonographer. The scan converter stores the amplified Signals and shows the resulting information on an echo display screen. The scan converter functions like the memory of a computer by recording the strength of each echo signal at an address that corresponds to the location of the echo source. The television display screen uses a similar address system, so that the addresses and the accompanying code for level of gray are read from the converter to the corresponding address on the television screen. The digital scan converter is also the major synchronizer in the system; the timing of events within the converter involves regulating the transducer Scanning rate on one side and the television format on the other. Storage of signals provides the opportunity not only to see the completed image but also to manipulate the image into a preferred form. A manipulation done after the signals are stored is called postprocessing. Postprocessing includes automatic electronic smoothing of the image and operator-controlled positioning of calipers on the image to measure distances. Recently manufactured veterinary ultrasound scanners have digital scan converters, which utilize the digital system that is used in personal computers. However, some of the veterinary ultrasound scanners that were marketed before 1983 used an analog scan converter in which an electron beam is deflected by signals onto a plate that contains addresses corresponding to an X (horizontal) and Y (vertical) axis. The development of analog scan converters led to sophisticated gray- scale processing in the late 1970s. This was accomplished by measuring the amplitude of the electric charge at a given XY location; this charge then was assigned a particular level of gray on the display screen. The analog converter is less stable and less resistant to electronic noise, because the quantitative representations of signal amplitude and time involve a continuum. In comparison, digital scan converters utilize discrete increments and stable electronic circuits. A potentially important disadvantage of the analog System is that the resulting images are not directly adaptable to videotaping. However, excellent still photographs can be prepared from images produced by the analog system. Most of the ultrasonograms (images from an ultrasound scanner) in this text were taken with an analog system. 38 Chapter 4 4.2 Signal Amplification Second First signal signal | | | | € 40 => ey a us | Near gain Far gain Adjustment of gain controls to equalize signal amplitude. The depicted signals are 40 microseconds apart, corresponding to reflectors 31 mm and 62 mm from the transducer, as shown in Section 3.8. Although the echogenicity of the two reflectors was similar, the amplitude of the second signal is weaker because of attenuation resulting from the interaction between the pulse and the first reflector. The amplifier greatly increases the strength of both signals in preparation for further processing. Because the two signals represent similar reflectors, the gain controls are manually adjusted, as shown, so that the amplified signals are of similar strength. This adjustment is made while viewing the resulting echoes on the image display. As noted in Section 4.1, proper adjustment of the gain control knobs is crucial in building a balanced and pleasing image. On some scanners, a control is provided for time gain compensation (TGC). This control allows the operator to compensate for attenuation by equalizing the echoes coming from different tissue depths; echo signals from distant reflectors are amplified more by the TGC control system than are those from close reflectors. On most scanners, near-gain, far-gain, and overall gain controls are provided for the near field, far field, and the overall image, respectively. Proper balancing of the controls is needed to provide maximum clarity at various depths and to minimize artifactual responses. The near-gain setting affects primarily the amplification in the first few centimeters and is needed especially to reduce the large-amplitude echoes near the transducer. The far-gain control is less important when viewing tissues in the reproductive tract that are close to the intrarectal transducer. Far gain is more important when visualizing large or distant structures (e.g., fetus, postpartum uterus). Signal Processing 39 4.3 Digital Scan Converter XY address corresponding to location of reflector Multi-bit word (Z axis) defining ere Tete amplitude of echo signal Memory storage system in digital scan converters. The scan converter memory stores the appropriate information for location of the reflector and amplitude of the echo. The digital system uses discrete circuits known as bits. "Bit" is an abbreviation for binary digit; each bit represents one of two levels of a quantity. A string of bits forms a multibit word. The number of discrete values represented by various numbers of bits in a word is as follows: 2 bits, 4 levels; 4 bits, 16 levels; 5 bits, 32 levels; 6 bits, 64 levels. Therefore, a six-bit word can represent up to 64 shades of gray according to the amplitude of the incoming echo signal. The capacity of the memory storage system of scanners can be judged on the basis of the number of bits used. The data are stored in computer chips mounted on circuit boards. Mathematically, but not physically, the organization of the memory for entry and subsequent reading of echo data can be illustrated, as shown. The example relates to the previously described tissue reflectors located 31 and 62 mm from the transducer (Section 3.8). Addresses are represented by the XY axes and echo amplitude by the Z axis (multi-bit word). Each multi-bit word is shown consisting of six bits, which describe, according to echo intensity, the shade of gray for each memory location (reflector site). Most current instruments use words of four bits (16 echo levels) to six bits (64 echo levels) to represent the gray-scale value at each address. In this example, the amplitude of the two processed signals was similar, as a result of 40 Chapter 4 manual adjustments of the gain control knobs (Section 4.2). Veterinary scan converters may use, for example, a memory size that is 114 units wide and 450 units high (51,300 addresses). Spatial resolution, which is ultimately transferred to the image screen, is limited not only by the transducer (Section 3.4), but also by the quality of the addressing system of the scan converter. 4.4 B-Mode Display IMAGE DISPLAY TRANSDUCER SCREEN ATTA Fy - aC thd ob 1 Echo issue reflector Density of tissue reflector: Brightness of pixel: gg Very dense —— ——_—_—__—_> White intermediate ——————————- Gre y B-mode echo-display system. The ultrasound scanners described in this text use sequential linear-array transducers or sector transducers (Section 3.1), real-time imaging (Section 3.3), and B-mode echo display. The term B-mode refers to brightness modulation of the dots or pixels on the echo display screen. Each pixel or picture element corresponds to a location in the scan converter memory, which in turn corresponds to the location of a tissue reflector. Echo signals are presented as brightened pixels with the distance from the top of the image to the pixel representing the distance from transducer to tissue reflector, as shown. The brightness of a pixel corresponds to the amplitude of that individual echo signal. Brightness is represented by shades of gray extending from white (very bright or Signal Processing 41 highly echogenic) to black (no discernible echo; nonechogenic or anechoic). For clarity, the size of the pixels is greatly exaggerated in the illustration. Tissue reflectors that are very dense are shown reflecting all of the sound pulse, resulting in very bright (white) pixels. Reflectors of intermediate density are shown returning only a portion of the pulse, resulting in gray pixels. The terms hypoechogenic, hyperechogenic, and isoechogenic are sometimes used to describe low, high, and uniform intensities of echogenicity, respectively. Images from a sector scanner showing two formats. The specimen is an ultrasonic phantom containing fluid-filled objects (largest = 20 mm) at a depth of 80 mm. The image on the left is in the white-on-black format used on current linear- array and sector scanners for examination of the reproductive tract of horses. However, some ultrasonographers in human medicine prefer the reverse black-on- white format shown on the right; the background is white and the brightest pixels are black. Some scanners are equipped with a video inverse switch so either format can be chosen. An ultrasonographer who has worked with only one format may have difficulty when first presented with the opposite format. Apparently, white-on- black provides better boundary information and black-on-white provides better textural information. In the above images, the outlines of the non-echogenic, fluid- filled spheres seem more distinct on the white-on-black image (left), whereas the ultrasonic texture of the remaining simulated tissue seems more pronounced on the black-on-white image (right). Sometimes the availability of both formats aids in the interpretation of echoes from a suspicious area. It is easier for a beginner to use the white-on-black format because of familiarity with light rays. Thus, an intense reflection of either light or an ultrasound echo (as seen on the image) is a bright spot, and a shadow is a black area behind a light barrier or beneath a dense ultrasound reflector (e.g., bone). 42 Chapter 4 4.5 Display Screen Electron gun Electron beam Raster lines Deflection plates Transfer of the information in the scan converter to the echo-display screen. Electrons ("cathode rays") are produced by heating a filament in the electron gun of the oscilloscope. The cathode ray tube provides a vacuum for free passage of the electrons toward a positively charged screen. The electrons are focused into a well-defined electron beam. The beam strikes the phosphor in the screen, causing it to emit light. The beam is directed across the screen by electric signals applied to the deflection plates. This system is the same as that used in television sets. The pattern of movement of the electron beam across the viewing screen is called a raster scan. The scan pattern begins in the upper left corner of the screen, moves steadily across, snaps back, and then begins another line, as shown. The result is a pattern of beam movement over the screen that forms a sweep or set of horizontal raster lines. In contrast to a simple oscilloscope display, the electron-beam scanning arrangement of most ultrasound viewing screens involves two sets of raster lines (sweeps) across the screen to produce an image frame. In a 450-line screen, each sweep results in the filling of 225 lines -- every other one. The raster scan then _jumps to the top of the screen and starts another sweep, which fills in the remaining lines. Each sweep takes 1/60th of a second and both sweeps require 1/30th of a second. Therefore, 30 complete frames are produced in one second, each frame being a composite of the two sweeps. This arrangement reduces flicker on the screen. As noted in Section 3.3, the rapid frame rate is needed for the real-time aspect of modern diagnostic ultrasonography. Signal Processing 43 4.6 Scanning Lines Scanning lines,each corresponding (eee i elt to the center of a beam Cm scale fe ee eee) Echo representing the location and density of a tissue reflector Scanning lines and the placement of echo data on the viewing screen. This example of the placement of the echo information onto the viewing screen corresponds with the tissue reflectors described earlier. For an overview of the simplified example extending from generation of an ultrasound pulse to the appearance of an echo on the screen, review the figures in the following sections in sequence: Section 3.8, events at the transducer and tissue; Section 4.2, echo signal amplification; Section 4.3, storage of location and intensity data in the scan converter; this Section, formation of the image. Each vertical scanning line on the TV image records the information gathered by the corresponding transducer beam. Note that the echo display for each tissue reflector is placed in the appropriate location (first scanning line, 31 and 62 mm from rectal wall). As the raster lines move across the screen, the electron beam turns on and off and thereby produces the vertical scanning lines. Therefore, each point at which a raster line crosses a scanning line represents one pixel. The intensity of the electric signals originating from the memory of the scan converter (Section 4.2) provides the appropriate information for the intensity (gray-scale level) of the oscilloscope's electron beam for each pixel. Thus, the information stored at each address of the converter is transferred to each pixel of the viewing screen as the sweeping beam of electrons passes over the appropriate pixel. Digital systems are very fast, so that recording into the memory, modifying data, and transferring data to the screen can all be done without one process interfering with the other. 44 Chapter 4 € 28942 1 626 0944 $42 i chk ee i biel agi! Hit 7 Ti, Hele TEE Lee a MN ‘ Signal Processing 45 Example of scanning lines. The image of an equine corpus luteum is delineated by arrows in the smaller ultrasonogram. The scale marks are in centimeters. This luteal image, which has the appearance of a ghostly face, was seen by R. A. Pierson during thesis research on the ultrasonic anatomy of the luteal glands. Note that the scanning lines are much more obvious in the enlarged -ultrasonogram. There are 114 scanning lines over the width of the image. The width of the field of view encompassing the 114 scanning lines is 56 mm. This Scanner utilized 64 elements. Firing consisted of the even-odd format, and therefore the scanning lines are separated by half-element spacings (Section 3.2). The number is less than 128 (half-element Spacing for 64 elements) because there is a mathematical loss of some elements on each end of the transducer; each scanning line conforms to the center of each fired cluster of elements (Section 3.3). A comparison between the small and large images demonstrates that scanning lines become more obvious when a sonogram is enlarged, if the distance from the sonogram to eyes is held constant. Note, however, that the scanning lines in the enlargement begin to disappear as the distance from sonogram to eyes is increased. For this reason, if the scanning lines on a given scanner are bold and considered unpleasant, increasing the distance from screen to eyes may be more acceptable. This same principle applies to scanners with zoom or magnification controls. The scanning lines may be more obvious upon magnification, unless the distance from observer to screen is increased. At the optimal distance, the eye tends to provide a smoother picture. The sonograms from some scanners using digital scan converters do not show distinct scanning lines. Such scanners may use an electronic system to average the intensities between adjacent lines, thereby tending to fill in the spaces between lines. The image, therefore, has a smoother appearance than the one shown in this figure. The quality of engineering in the display system is a source of variation in image quality. Part of the smoothness of the digital television image comes from the use of very small pixels. Some scanners in the veterinary market are undesirable because large pixels give the image a checkered appearance. In some veterinary scanners, the conventional.format of vertical scanning lines and horizontal raster lines is rotated 90°, so that the scanning lines are horizontal and the raster lines are vertical. When videotapes prepared from such scanners are played back on the scanner monitor, they are properly oriented. However, when shown on a television viewing screen with the conventional format, the image is oriented on its side. The image can be made upright for viewing purposes, only by setting the television monitor on the appropriate side. 46 Chapter 4 4.7 The Net Result Cbs ot Ki at OS ** Pre i rad Lad * Pes tAae. Neabh ah Moving image Beam (path followed by pulse) | | I Direction of sequential production , of beams \ | een | | | ! \ Relationship between placement of the transducer and the development of an image of a 45-day equine fetus. A 64-element linear- array transducer is held in the rectum over a uterine horn. An ultrasound pulse is generated by firing a cluster of elements. The resulting pulse follows the confines of the imaginary focused beam, and echo signals are returned to the same elements. After completion of echo-gathering data, the cluster moves down the array and a second pulse is fired, producing a second beam. This sequential, segmental firing of clusters of elements moves along the array, completing 30 passes per second. The resulting echo signals are processed, resulting in a displayed image. Each picture element (pixel) corresponds to a location of a tissue reflector, and the brightness of each pixel corresponds to the density (echo-producing ability) of the tissue reflector. The information generated by each pulse or beam is displayed by a corresponding Signal Processing 47 scanning line on the screen. The image shown represents a thin, two-dimensional slice through the tissue. Slowly moving the transducer produces sequential slices, which are displayed as images corresponding to the orientation of the transducer. Thus, a mental image of the three-dimensional aspects of the fetus is obtained. The movements of the fetus (heart, legs) are observed in real-time -- that is, as they occur. REFERENCES 1. McDicken, W. N. 1981. Diagnostic Ultrasonics: Principles and Use of Instruments. John Wiley and Sons, New York, NY. 2. Sarti, D. A. and W. F. Sample. 1980. Diagnostic Ultrasound: Text and Cases. G. K. Hall Co., Boston, MA. 3. Zagzebski, J. A. 1983. Pulse-echo ultrasound instrumentation. In Textbook of Diagnostic Ultrasound. Ed. S. Hagen-Ansert. C. V. Mosby, St. Louis, MO. 4. Bartrum, R. J. and H. C. Crow. 1983. Real-time Ultrasound. W. B. Saunders Co., Philadelphia, PA. 5. Powis, R. L. and W. J. Powis. 1984. A Thinker's Guide to Ultrasonic Imaging. Urban and Schwarzenberg, Baltimore, MD. Chapter 5 INTERPRETATION Proper interpretation of the echoes on an ultrasound screen is crucial. Interpretation requires knowledge of the relationships between tissues and echoes and the ability to differentiate between true and artifactual responses. Interpretation of ultrasound images is one of those disciplines for which learning is never complete. Even after much experience, an ultrasonographer may see a structure or image formation that he had not noticed before --- only to find that thereafter the formation is disconcertingly common and obvious. This chapter will describe the two types of reflections (specular and nonspecular) which are presented as echoes on the screen and the associations between them and tissue structure. Details of the ultrasonic anatomy of specific structures are given in later chapters. The principles of ultrasonography center on the ability of sound waves to be either reflected from or propagated through various tissue interfaces. However, certain tissue formations also cause waves to bend (refract), bounce back and forth or re-echo (reverberate), become weakened (attenuated) or entirely blocked. As a result, distortions appear on the ultrasound image which can be mistaken for normal or pathological structures or changes. These artifactual echoes complicate the interpretation process --- so much so, that individuals who are in the forefront of evaluating and utilizing ultrasound scanners for research and clinical purposes must accept the risk that they may describe an interesting biological structure which others may later show was an artifact. In this regard, lay publications appeared in the early 1980s in which the specular echo on an image of an embryonic vesicle was erroneously labeled the embryonic disc. Such misinterpretations are easily made. However, knowledge of the nature and origin of artifacts, as well as true echoes, should minimize the occurrence of such misinterpretations. Artifacts are especially common during imaging of the reproductive tract because of the many pockets of bowel gas, fluid-filled structures, and the pelvic bone. This chapter will describe the nature and origin of the artifacts common in this area, including shadowing, enhancement, reverberation, and beam-width artifacts. The principle sources for developing the various concepts were references 1, 2, and 3. 50 Chapter 5 5.1 Specular Reflections , Ultrasound beam Angle of impact ; 4 1 Angle 4 of reflection 3 Reflection Transmitted fire : bean uid-filled structure Relationships between angle of impact of a sound beam and specular reflections. A ular reflection results when a pulse strikes an interface that is smooth, wider than the pulse, and parallel to the transducer. Usually, only a small portion of the pulse that strikes such an interface is reflected. The major portion of the pulse continues past the interface as a transmitted pulse or beam, as shown. If the wall of the opposite side of an encapsulated, fluid-filled structure is smooth, the opposite wall also will act as a specular reflector. Only a pulse that strikes a specular reflector at a right angle or very close to it will be recorded as an echo on the image display. Therefore, the intensity of the echo is determined not only by the difference in acoustic impedance between the two tissues making up the interface, but also by the angle of incidence (angle of impact). Pulses that strike the interface at other than a right angle are reflected at an equal angle into the tissue; the smooth surface in these areas will not be detected unless the orientation of the transducer is changed. Specular reflections are very common in images of the female reproductive tract. The smooth surface of the uterus, embryonic vesicles, and uterine cysts are all examples of specular reflectors. Interpretation 51 Day 10 vesicle wy cu ; as . ~ . RR oa nr a Examples of specular echoes from the reproductive tracts of mares. Note that the echoes (arrows) are bright (highly echogenic) and parallel to the surface of the transducer (upper edge of image). The echoes can be seen both on the upper and lower surfaces of the Day-10 and Day-13 embryonic vesicles (A, B). A specular echo is not discernible on the bottom of the Day-16 embryonic vesicle (C), probably because the surface is not smooth or because the echo is obscured by the high degree of echogenicity in the soft tissue beneath the vesicle. This echogenicity is due to enhancement and is discussed below. Specular echoes are also seen on the surface of the images of ovarian follicles (D). Sometimes the transducer must be rotated slightly to demonstrate a specular reflection from a follicle; the orientation of the transducer and surface of the follicle must be compatible, as shown in the diagram. The specular echo on the large compartment of the uterine cystic complex (E) is attributable to the smooth surface of a dome-like fluid-filled projection into the uterine lumen. Frequently, the uterine lumen is seen as a bright echogenic line when the uterus is viewed longitudinally (F). Presumably, the bright line results from specular reflections from the free surface of a uterine fold. 52 Chapter 5 5.2 Nonspecular Reflections SPEC ULAR REFLECTOR NONSPECULAR REFLECTORS sy . Oo °O oO ° z O Smooth interface Rough interface Small interfac es Comparison of the origins of specular and nonspecular echoes. Nonspecular or diffuse reflections originate when a pulse strikes a rough interface or one that is narrower than the pulse. In contrast to specular reflectors, echo amplitude is independent of beam angle. Ultrasound pulses in the focal zone have dimensions of 2 or 3 mm. Interfaces smaller than these approximate figures therefore give rise to nonspecular reflections. Examples are the small interfaces between parenchymal cells (luteal, endometrial) and the surrounding small vessels. When the ultrasound pulse enters a heterogeneous medium or strikes a rough surface, the effective interfaces are narrower than the beam, and scatter of echoes occurs, as shown. Scatter is defined as the redirection of sound in many directions. A very small portion of the scattered echoes are directed back toward the sound source and these are referred to as backscatter. The amplitude of the echoes reaching the transducer from a scatterer is very low (e.g., 1/100 of the amplitude of a specular echo). Since the pulse encounters many scatterers simultaneously, many echoes are generated at once. Some of these may arrive at the transducer at one time and may either reinforce or interfere with one another. The net result is a displayed pattern, texture, or speckling that may help to identify a given tissue or to distinguish one tissue from another (e.g., corpus luteum versus surrounding stroma). This delineation may be possible even in the absence of an intervening capsule acting as a specular reflector. Gray-scale imaging fully utilizes the scatter phenomenon of nonspecular or diffuse reflections. Because the echo amplitude is independent of Interpretation 53 beam angle (unlike specular reflectors), the shade of gray (echo amplitude) for a given cellular tissue is more nearly constant regardless of transducer orientation. Backscatter causes the vast majority of diagnostic echoes in cellular organs. However, the internal tissues of organs can cause some specular as well as nonspecular reflections. Uterine body ai Uterus\ Cervix D pope 7 res Examples of nonspecular echoes from mares (A, B, D, E, F) anda cow (C). CL =corpus luteum. Compare the ultrasonic textures of the indicated tissues. Note the marked differences in echo texture between the uterine body and the denser cervix (D). The urinary bladder (F) of horses normally contains particles large enough to act as diffuse reflectors. The bladder was subjected to ballottement by the transducer just before the sonogram was taken. The particles therefore formed a swirled pattern. The cross section of a uterine horn (B) contains a Day-11 embryonic vesicle with a specular echo on the upper and lower surfaces. 54 Chapter 5 5.3 Shadow Artifacts DENSE OBJECT REFLECTION REFRACTION Ultrasound beam Fluid | Bone an I i Origin of shadow artifacts. A shadow is caused by a noticeable decrease or absence of ultrasonic waves due to blockage or deviation of the sound beams. An ultrasonic shadow is comparable to a shadow that occurs behind a light barrier. The area of a sonogram resulting from blocked sound waves appears dark, similar to a shadow resulting from blocked light waves. The acoustic shadow shown on the left is caused by reflection of most of the sound beam back to the transducer, combined with some absorption of the sound by the reflecting structure. Such massive reflections require a marked mismatch in acoustic impedance (density) at the interface between two tissues or entities (e.g., gas and soft tissue; soft tissue and a very dense material, such as bone or a foreign body). The shadow shown in the middle is caused by reflection of the beam from the side of a smooth curved structure (Section 5.1). Reflection may be the entire or major reason for beam deviation when — the speed of sound is similar in the tissues on both sides of the reflecting surface (e.g., curved side of an ovary or cross-section of a uterine horn). However, if there is a mismatch in the speed of sound on each side of the reflecting surface, the portion of the sound pulse that is transmitted through the surface may also undergo refraction (Section 2.4), as shown on the right. This commonly occurs in association with fluid-filled structures (e.g., equine yolk sacs, uterine cysts, ovarian follicles). Shadowing, as well as other artifacts, may be more pronounced when the offending object is located in the focal zone. Because pronounced shadow depends on complete blockage of a beam, a small object (e.g., 4 mm diameter) may be adequate to block the beam in the narrow focal zone, but not in areas proximal or distal to the focal zone. Interpretation 55 a e sa Pe se eee aa ie 5.4 Enhancement Artifacts Fluid Soft tissue Fe aD Attenuation of beams and the origin of enhancement artifacts. Enhancement or through-transmission artifacts are common in sonograms of the reproductive tract because of the presence of fluid-filled structures (Ovarian follicles, cysts, embryonic vesicles). These artifacts result when the ultrasound beams pass through a reflector-free structure (i.e., fluid-filled), as shown. The beam is not depleted (attenuated ) by echo production while passing through the fluid; that is, the fluid is nonechogenic. Therefore, when the beam emerges from the far side, the amplitude of the pulse is greater than in the tissues on each side. The relatively greater amplitude or strength of the sound beams distal to the fluid-filled structure results in a column of relatively brighter echoes beneath the structure. Enhancement artifacts are often useful because they provide a diagnostic criterion of fluid proximal to the column of enhancement. The sonographer must be careful, however, not to mistake the area of increased echogenicity for a structure. Enhancement artifacts tend to obscure the wall of the fluid-filled structure at the point of emergence of the beam. This difficulty is compounded when the wall also produces a specular echo. Such saturation with echo signals in an area of interest can be diminished by adjusting the gain controls to obtain a more reasonable compromise between the saturated area and other areas of interest. The degree of enhancement, as well as shadowing, can also be diminished by using a lower frequency transducer, because ultrasonic attenuation is directly proportional to frequency. In addition, these artifacts will be more pronounced and sharper in the focal zone of the beams. 56 Chapter 5 Usp itis; Wy HE ive | ‘ Ua! i H'), sont i aang, hati nt tv nr i ‘| Hl se naa b i a pe 4 Hi a 5 iy , H ; " A ty Leer) pees eres , ; ‘ FEHR ORR a minut ey ee arc uatiriid cr AWS 4 Le f ! silly, ‘iil Hi : ef ee , 4h a Walia wei i Ee Hp Hiciy:! iy yl a ; i nei i ‘ 4 Vesicle Follicles DE Follicles Interpretation 57 Examples of shadowing (A-F) and enhancement (D-F) in sonograms from the mare reproductive tract. In Sonograms A, B, and C, the shadowing is caused by complete blockage of the ultrasound beams by very dense or reflective objects. Specifically, the shadowing is due to a fetal bone remnant in the uterus from an abortion (A, arrow), fetal ribs in a normal 120-day conceptus (B), and reflection of the beams from the highly reflective soft tissue and gas interface of two loops of bowel (C, arrows). In D, E, and F, the shadowing is due to reflection or refraction from the sides of follicles and a Day-14 embryonic vesicle. Note the distinct columns of echogenicity (enhancement artifacts) beneath the fluid-filled structures. The enhancement artifacts are further accentuated by the columns of shadowing on each side of the column of enhancement, especially in D and E. Note that enhancement results in greater tissue penetration, as shown by the length and the ultrasonic detail of the echogenic columns. For this reason the phenomenon of enhancement has been used to project ultrasound beams deeper into a tissue mass. Transducers have been designed that are encased in a water bath. The water bath between the transducer face and the skin surface causes greater ultrasound penetration into the tissues. Shadows and enhancement are major artifacts in images of the female reproductive tract and surrounding area. Shadowing is a problem because the tissue in the involved area is not accessible to imaging unless the sound beam can be adequately redirected. Enhancement is usually not a problem, unless the area of _ increased echogenicity is mistaken for a structure. If enhancement obscures the distal wall of a fluid-filled structure, adjustment of the gain controls will minimize the problem. Overall, shadowing and enhancement are probably more of a help than a hindrance. Shadows provide information on the density of an offending object and may delineate the lateral boundaries of a round, solid structure. Either artifact may serve to draw the operator's attention to a fluid-filled structure. 58 Chapter 5 5.5 Reverberation Artifacts INTRARECTAL REVERBERATIONS DISPLAY TRANSDUCER SCREEN Upper interface = transducer and rectal wall Soft -—-> tissue R_——" Lower interface = soft tissue and gas Acoustic shadow Reverberation Origin of reverberation artifacts. Reverberation is a process wherein an echo bounces between two strong interfaces until the ultrasound pulses are exhausted by attenuation. The illustration shows the origin of reverberation artifacts on the display screen as a result of the sound waves bouncing between a soft tissue-gas interface and the interface consisting of the transducer and rectal wall. The pulse is shown making three round trips between interfaces. Round trip number | results in a legitimate echo on the display screen (echo 1). The second and third trips, however, result in false or reverberation marks on the screen. At the completion of the first round trip, the echo strikes the transducer crystals, produces a small voltage, which in turn results in an echo at the appropriate pixel on the screen. However, much of the sound energy is reflected back into the tissue, because the transducer-rectal wall interface represents a profound acoustic impedance mismatch and therefore a very effective ultrasonic reflector. On the second return trip, the waves again strike the crystals, resulting in another echo on the screen (echo 2). This time, however, the scan converter assigns an address that is twice as far from the top of the screen as for the legitimate echo; the second trip resulted in a doubling of the interval from the time the pulse originally left the transducer (trip 1) until the echo struck the transducer after trip 2. In addition, the amplitude of the second echo is weaker because of additional attenuation. Therefore the resulting echo on the screen is not as Interpretation 59 distinct as the first. If there is sufficient amplitude remaining after the second trip, this process may continue producing additional echo signals that are equidistant from one another but progressively weaker. The following three distinguishing features help identify reverberation artifacts: 1) they are equidistant, 2) they gradually diminish in intensity, and 3) they are oriented parallel to the reflective interface. If a very reflective interface is involved (soft tissue-gas), none of the pulse is transmitted through the interface. Therefore, an acoustic shadow results and the reverberation echoes on the screen are placed in the shadow. If part of the pulse is transmitted, as in a soft tissue-liquid interface, the resulting reverberations are placed in the nonechogenic image of the fluid. The more reflective the interface, the greater the likelihood of reverberations. In addition, when the reflective interface is close to the transducer, attenuation is minimal, and the number and intensity of reverberations are increased. Reverberation artifacts are more likely to be seen when they are contrasted with an acoustic shadow or nonechogenic fluid. However, they may be present also in the more echogenic portions of the image and may not be noticed because they are masked. Reverberation artifacts are very common in the pelvic area because of pockets of bowel gas. In addition, the many fluid-filled structures in the female reproductive tract increase the potential for legible reverberations. Reverberation echoes within fluid-filled structures would be troublesome because they could be mistaken for echoes representing internal structure. Fortunately, the distance from transducer to the surface of a fluid-filled structure is usually too great and the reflective properties of the structure are usually not adequate to produce visible reverberation echoes. Apparently, reverberation artifacts can appear on an image even though the source of the reverberations (bowel gas, pelvic bone) is beneath the displayed depth of the image (personal communication, O. Parker, Equisonics, Inc.). The echoes from such structures may reach the transducer after another pulse has been emitted. In this event, the reverberation marks would be superimposed on the image resulting from the subsequent beams, and the origin of the reverberations would not be apparent. Perhaps many of the unexplained, bright echoes on images of this area of the body are from this source. Although less important, internal reverberations also can occur due to rebounding between two interfaces within the animal's tissue. The resulting artifacts are difficult to distinguish from real echoes, but their number is small. Since continuing reverberation depends on adequate signal amplitude, the number and brightness of reverberation artifacts increase as the gain is increased. 60 Chapter 5 Appropriate adjustment of the gain controls may help diminish the reverberation problem. In addition to proper gain control adjustments, reverberations sometimes can be minimized by manual reorientation of the organs of interest (ovaries, uterus) in relation to the offending loops of bowel. Distinct, multiple reverberations often are caused by gas or fecal material between the transducer and the rectal wall. This problem usually can be corrected by reorientation of the transducer or by running a finger across the face of the transducer. Offending material on the face of the transducer sometimes can be removed with a finger without withdrawing the transducer from the rectum. Incident Reflected beam ———— beam j Reflected Tissue scatterer wave from = scatterer Echo 1 Reflective eae 2 ' surface of bowel TISSUE IMAGE Origin of artifactual scattered echoes in a reverberation path. Scattered echo signals often appear beyond highly reflective surfaces between a tissue reflector and the first reverberation echo or between successive reverberation echoes. The origin of the artifactual patterns (2) is depicted. When the incident beam traverses the soft tissue, scatters are encountered which produce nonspecular echoes as described in Section 5.2. When the incident beam strikes the highly reflective surface, a large fraction of its energy is reflected. The reflected beam also encounters the scatterers when returning toward the transducer. Some of the resulting waves from the scatterers also reflect off the highly reflective surface, as shown. These latter echoes (echo 2) trail the main echo (echo 1) back to the transducer. The scanner, therefore, places the origin of the diffuse echoes beyond the reflecting surface. Interpretation 61 etic tet a Reverberation . Reflector Reverberation Ay B Examples of reverberation echoes. In sonograms A and B, the reverberation echo originates from a reflection off the surface of a gas-filled loop of bowel. Note for both A and B that the distance from transducer face (top of image) to the reflector is equal to the distance from reflector to the reverberation echo and that the reverberations fall within the shadow cast by the reflecting surface. In C, the reverberations resulted from a pocket of air or fecal material between the transducer face and the rectal wall. Note that the reverberation echoes are equidistant and progressively weaker. The reverberation echoes in A and B are curved because the highly reflective surface of the air-filled bowel is curved. In contrast, the reverberation echoes in C are straight, because the contact area between the transducer face and the offending material (air, feces) is straight. Artifactual echoes due to scatters, as described above, are prominent beneath the reflector (A, B). \e 62 Chapter 5 5.6 Beam-Width Artifacts Area where beam intersects with both fluid and tissue oe Fluid-filled | \ structure Origin of beam-width artifacts. The periphery of large fluid-filled structures or the entire fluid volume of a small structure often casts a foggy appearance due to partial fill-in of the nonechogenic fluid with echogenic artifactual spots. As a result, the expected sharp, distinct outline is not obtained. The lack of detail is especially noticeable on the lateral walls (with respect to the image) of the fluid-filled structures. As shown above, this is a problem of lateral resolution resulting from portions of the beam sampling both wall and fluid at a given depth (2, 3). When two echoes reach the transducer at the same time, they are treated as one echo and result in one signal. Because the beam fans out beyond the focal zone, beam- width artifacts sometimes appear at the bottom of large structures (e.g., preovulatory equine follicle). Such artifacts may assume a meniscus-like shape, resulting from a change in the solid:fluid ratio as the beams move along the linear array of elements (3). Beam-width artifacts can be generated not only in the plane of the scan or image (width of field of view), but also in the opposite plane corresponding to the thickness of the ultrasonic slice of tissue (2). Beam-width artifacts can mimic solid projections (4) or disorganization of the wall. If beam- width artifacts are misinterpreted as an indication of partial collapse or disorganization of the wall, an erroneous conclusion of impending ovulation, follicular atresia, or embryonic death may be made. The origin of echogenic spots within a fluid-filled structure sometimes can be determined by ballottement of the UETEEEEEEer een sees re Interpretation 63 structure with the transducer; true reflectors may respond by floating. Since the artifact is a function of beam width, the artifactual fill-in can be reduced by using a transducer that has a narrower beam at the depth of greatest interest. As noted above, beam width also contributes to shadowing (Section 5.3) and enhancement artifacts (Section 5.4). In the focal zone, the beam is the narrowest and the intensity of the sound pulses is the greatest. Solid structures (bone) and fluid- filled structures that interact with the beam in the focal zone produce more intense artifacts. In addition, the small offending structures are more likely to involve the entire width of the beam in the focal zone. Occasionally, a band-like artifactual area of increased echogenicity is observed in the focal zone when scanning large solid structures. et HON waitiatlli Examples of apparent beam-width artifacts. Note the echoes scattered near the periphery of the fluid-filled equine follicles (left) and near the bottom of an embryonic vesicle (right). Perhaps the echogenic spots on the bottom of the embryonic vesicle are due to a curvature of the vesicle in the plane associated with the thickness of the ultrasonic slice; the curvature in the given plane does not seem to adequately account for a beam-width artifact. any ee 64 Chapter 5 5.7 Other Artifacts Examples of artifacts due to faulty equipment or outside interference. Ultrasound scanners are highly complex, and the images are subject to aberrations due to engineering flaws or shortcuts, malfunctioning, and outside electrical interference. In addition, a certain amount of noise seems to be a natural byproduct of the reception, processing, and display of ultrasound echoes. The diagonal bright column (arrow) in sonogram A was due to an engineering problem in a transducer. The dark vertical line in sonogram B was due to improper contact between the connection of the transducer's coaxial cable and the receptacle in the console. It was corrected by proper seating. A similar black line can be caused by malfunctioning elements at the origin of the line. The noise in sonogram C (diagonal, bright lines) was due to electrical interference from attempting to run another scanner on the same electric circuit. Refrigerators are a common source of such interference. Sonogram C was made by imaging an ultrasonic tissue phantom. The bright horizontal lines near the bottom of the image represents the phantom design and are not artifacts. Large shadows or fluid-filled structures (cysts, embryonic vesicles) serve as good test areas for the presence of extraneous signals. If the shadow or fluid contains echogenic areas that do not appear to be caused by Interpretation 65 reverberations or beam-width artifacts, electric noise or engineering artifacts may be superimposed on the picture. Such noise may be present, but not noticeable, in the more echogenic portions of the image. REFERENCES 1. Bartrum, R. J. and H. C. Crow. 1983. Real-time Ultrasound. W. B. Saunders Co., Philadelphia, PA. 2. Zagzebski, J. 1983. Images and artifacts. In Textbook of Diagnostic Ultrasound. Ed. S. Hagen-Ansert. C. V. Mosby Co., St. Louis, MO. 3. Wicks, J. D. and K. S. Howe. 1983. Fundamentals of Ultrasonic Technique. Year Book Medical Publishers, Inc., Chicago, IL. 4. Sarti, D. A. and W. F. Sample. 1980. Diagnostic Ultrasound; Text and Cases. G. K. Hall and Co., Boston, MA. Part Two _ INSTRUMENTS and TECHNIQUES Chapter 6 INSTRUMENTATION The encouragement in Chapter 5 to develop and hone image-interpreting abilities assumes the availability of images worth trying to interpret. This chapter provides information on the selection, operation, and care of scanners with a view toward attaining high-quality images. Examples are shown of linear-array transducers designed for intrarectal use in horses, and attention is given to transducer care and maintenance. Transducer frequency is discussed with the aid of sonograms from scans of ultrasonic phantoms. The console and its manual adjustments are considered with emphasis on the contrast, brightness, and gain controls. Some of the veterinary ultrasound scanners now being sold are high in price, but too low in quality. A potential buyer is faced with selecting a make and model that is suitable to specific needs. This chapter includes a section on considerations for instrument selection. Photographic examples of commercial transducers and scanners are included. It is emphasized that use of images prepared from a given make and model or photographs or other references to specific models should not be taken as a recommendation for purchase. Our laboratory has used instruments that were loaned to us or provided at a reduced cost for research. With one cited exception, the photographs of instruments in this chapter were prepared by us, using scanners available when the photographs were taken. Discussion of techniques for producing photographs and videotapes of images is deferred to Chapter 7. 68 Chapter 6 6.1 The Transducer * gies PGT ENS Gene Er CaO Side (left) and face (right) views of linear-array transducers for intrarectal use in horses. Top to bottom: 3.5 MHz, 5.0 MHz, and 7.5 MHz transducers used with the Equisonics 300; 5.0 MHz transducer used with the Technicare 210DX. The linear-array of piezoelectric crystals is located beneath the white (upper three transducers) and black (lower transducer) rectangular area on the face views. The scale is in inches. Transducers for intrarectal use must be designed for ease of insertion and manipulation within the rectum and for minimization of trauma. However, the instrument should have adequate bulk and grip for easy manipulation and to prevent hand cramping during prolonged use. One should be able to readily identify the active surface by feel alone. Transducers should be waterproof and resistant to corrosion. Transducers that require insertion into a plastic cover before intrarectal use are undesirable. Waterproofing and electric insulation are important to prevent electric shocks to the animal or operator and damage to the delicate interior of the transducer. In this regard, users should consider the chance of damage to the scanner, operator, and horse during lightning storms. Although transducers are complex and refined instruments, they also must be durable for veterinary use. The coaxial cables, shown on the right of the transducers, connect the transducer to the console. The cable contains many fine electric wires because of the necessity for independent connections to each crystal. The junction of the cable and transducer must be well-insulated and sealed. The cable should be flexible for intrarectal use and must not be bulky; a thick cable will Instrumentation 69 interfere with intrarectal manipulations and will cause discomfort to the forearm due to the action of the anal sphincter. On some scanners, the cables are too long and drag on the floor when the operator works with the horse close to the scanner (Section 8.3). The transducer should be withdrawn from the rectum by the hand that is gripping the transducer and not by pulling on the cable. This will minimize the likelihood of loosening the attachment of the cable to the transducer. Insertion of a transducer into a protective sleeve. Transducers are precision instruments and expensive (e.g., $3,000) and must be handled with care. Dropping the instrument or marring the covering over the elements can cause a malfunction. A cracked transducer could cause an electric shock (1). The ceramic crystals are brittle, and there are many connecting points of fine wire within the transducer. A defective element or connector may result in a vertical, spurious line through the image corresponding to the defective area (Section 5 .7). R. A. Pierson, in our laboratory, has devised a simple and effective method to protect the transducer during transport. A length of the foam cylinders that are used to insulate water pipes is obtained from a hardware or plumber's store. Such cylinders come in various diameters. If the optimum diameter is used, the transducer is easily inserted, yet the cover will stay firmly in place and provide a thick, protective cushion. The transducer is washed and dried before insertion into the transport cylinder. Occasionally, one may wish to examine a portion of the reproductive tract during surgery. For such use, the transducer should be cold-sterilized (e.g., with ethylene oxide). Autoclaving or heating to temperatures of 100°C may not affect the properties of the crystals, but could have a drastic effect on the bonding cements (2). 70 Chapter 6 6.2 Selection of Transducer Frequency Comparisons of resolution and penetration of 3.5, 5.0, and 75 MHz transducers. The medium is an ultrasonic phantom, and the nonechogenic objects are cross- sections of fluid-filled cylinders with the indicated diameters. The scale is in centimeters. The centers of the cross-sectional views are at 4 cm and 8 cm. The 3.5 MHz transducer barely defines the 5 mm object and does not detect the 2 mm object. The greater resolution of the 5.0 MHz transducer is demonstrated by the clearly defined 5 mm object and the barely defined 2 mm object. The 7.5 MHz transducer provides additional resolution, as indicated by the improved definition of the 2 mm object. However, the penetration of the 3.5 MHz transducer is much greater than that of the other transducers. The 20 mm object at 8 cm was clearly visible with the 3.5 MHz transducer but barely visible with the others (not shown). The 10 mm object is more distinct at 8 cm than at 4 cm with the 3.5 MHz transducer. The hazy appearance at 4 cm is attributable to beam-width artifacts (Section 5.6) because the focal zone is beyond the object. These comparisons demonstrate that the focal depths of the 5.0 MHz and 3.5 MHz transducers were close to 4 cm and 8 cm, respectively. Note that engineering modifications have been made (Section 3.4) so that the field of view of the 7.5 MHz transducer is comparable to that of the 5.0 MHz transducer, despite the increased frequency and apparently improved resolving power. However, at this writing, our experience with this 7.5 MHz transducer is limited and a critical evaluation is not available. Instrumentation 71 The difference in resolving power between a 3.5 MHz and 5.0 MHz transducer can be appreciated on a practical basis in the detectability of structures in the reproductive tract, as shown in the following table: Item 3.5 MHz transducer 5.0 MHz transducer Minimum diameter of detectable follicles 6-8 mm 2-3 mm Detectability of corpus luteum For 5-6 days Day 0 to regression Earliest detection of conceptus Day 11 (6-7 mm) | Day 9 or 10 (3-4 mm) ad , | re Ultrasonograms of an ovary taken with a 35 and 75 MHz transducer (Equisonics 300). The resolving capabilities of the 7.5 MHz transducer were much greater than those of the 3.5 MHz transducer. The focal point of the 3.5 MHz transducer greatly exceeds the 2 cm from transducer to center of ovary. This comparison further illustrates the advantage of the higher frequency transducers, noted earlier in this section. The distance from the transducer face during transrectal imaging to the center of the ovary or the lumen of a nongravid uterus is only a few centimeters. Therefore, a higher frequency transducer (e.g., 5.0 MHz) with a focal point of 3 or 4 cm is well suited for examination of the reproductive tract in nonpregnant or early pregnant mares. The lower frequency transducers (e.g., 3.5 MHz) are more suited for examining the uterus during late pregnancy or soon after parturition. 72 Chapter 6 6.3 The Console 6.3A Overview Instrumentation 73 Examples of linear-array ultrasound scanners. A) Equisonics 210, B) Equisonics 300, C) Technicare 210DX, D) Equiscan 9100. (The photograph of the- Equiscan 9100 was provided courtesy of the Bion Corporation.) The console contains the components for coordinating pulse emission from the transducer, processing the signals from the transducer, and displaying the resulting image on a viewing screen (Section 4.1). Diagnostic ultrasound scanners therefore contain complex electronic circuitry. For most veterinarians, portability is important because the scanner must be moved from farm to farm and sometimes from mare to mare on a given farm. Modern scanners therefore are equipped with a handle and other provisions for transport. Examples of weights of current models are 8 kg (18 Ibs, Technicare 210DX) and 18 kg (40 lbs, Wesmed Medical Systems, WIC50). Despite the complexity of the console, it should be durable and require minimal servicing. Because the scanners are used in barns, they should be designed to exclude as much dust and dirt as possible and should be readily cleaned. ease 74 Chapter 6 6.3B Contrast, brightness, calipers, and annotations FAR(aBcm) Lis FREEZE Po ee Instrumentation 75 EQUISCAN MODEL 9100 Front views of scanners showing examples of operator controls. A) Equisonics 210, B) Technicare 210DX, C) Equiscan 9100. All three models have near, far, and overall gain controls; a magnification or zoom feature; a freeze button; brightness and contrast controls; and electronic Calipers. The gain controls for the 9100 are behind the lower face panel (shown in Section 6.3A). The brightness and contrast controls on the 210DX and 9100 are on the side or rear of the console and can be adjusted with a screwdriver or special tool. The integral electronic calipers for measuring linear distances are controlled by omnidirectional press switches on the 210DX and 9100 and by toggle switches on the Equisonics 210. Toggle controls are faster and more maneuverable; unfortunately, all recent models that we have seen use press controls. The keyboard is for making annotations on the image to permanently identify videotapes and photographs (e.g., animal and farm identity, day of pregnancy, date, operator). Various annotation schemes are provided with different makes of scanners. A simplified and less expensive system is probably adequate for most purposes. Research scientists, for example, may desire the more elaborate systems. Newer Systems are becoming available that allow the operator to make free-hand marks on the screen with a pencil-like instrument. 76 Chapter 6 OVERLAY Viewing screen showing the shade bar, annotations, and centimeter scale for a 35 MHz and a 5.0 MHz trans- ducer (Equisonics 300). The transducer is not attached and the screen is devoid of echoes. The brightness control affects the amount of light associated with echoes on the screen. This control may be increased until light just begins to appear on the screen. Contrast is adjusted until all segments of the gray-scale shade bar are clear or until the darkest bar just begins to come into view. Proper adjustment of the contrast assures that the operator is using the maximum range of the gray i scale. As noted above, on some scanners contrast and brightness are preset or adjusted with a tool. The 3.5 MHz screen shows the labeling categories and the automatic recording of date and time. Examples of annotations made by keyboard entries are shown on the 5.0 MHz screen. The electronic caliper markers were placed on the screen by the operator and the distance between them (23 mm) was recorded automatically. The accuracy of electronic calipers involves a fraction of a millimeter (2), but in practice considerable error may be Instrumentation 77 introduced by uncertainty on proper placement. Cross-sectional areas can be measured with an error of approximately 5% (2), but error is increased by difficulties in precise steering of the caliper. Volume may be estimated by measuring a number of cross-sectional areas and applying an appropriate mathematical formula. An imaginary structure was outlined in the above illustration and the area is given. An example of shade bars and annotations is shown also on the above front view of the Equiscan 9100 (page 75). 6.3C Image inversion An original and an inverted image. Some scanners are equipped with an inversion control, so that the images can be oriented with the cranial aspect of the tissues either to the left or right of the viewing screen. This capability allows the ultrasonographer to more readily associate the image with the tissues being viewed. It is convenient to have the caudal aspect of the tissues appear on the right of the screen when the scanner is placed obliquely on the right of the operator and to the left of the screen when the scanner is on the left. The inversion setting can easily be tested by sliding a finger over the face of the transducer. 78 Chapter 6 6.3D Gain controls Instrumentation 79 Examples of various gain settings. Proper adjustment of the gain controls is crucial to building a balanced and pleasing image. The gain adjustments equalize the signal amplitude at various depths (Section 4.2). When the amplifier gain is too high, the echoes are too strong and the display is overloaded; conversely, if the gain is too low, the echoes are inadequate. The sonograms were taken with a 5.0 MHz (A,B,C) or 3.5 MHz (D,E,F) transducer and a scanner that had controls for overall, near, and far gain. [As noted in Section 4.2, some scanners have a total gain control (TGC) that uses a single knob to equalize the echoes at various depths.] A,D) Overall gain adjusted to an optimum level, but near gain is too low; objects in the first 2 cm (A) or 4 cm (D) from the transducer would not be detected or would just begin to come into view. B,E) Same as A and D, but near gain now is adjusted. C) Overall gain too high; note that the cross-sectional views of the cylinders have lost their normal nonechogenic (black) appearance. This same problem would occur if the brightness control were set too high. F) Far gain has been increased so that the 20 mm object at 12 cm is just beginning to come into view (arrow). Usually, the gain controls can be adjusted at the beginning of the examination of a number of mares with no adjustment thereafter except for occasional fine-tuning. All controls (brightness, contrast, gains) must be considered together. The adjustment of the brightness control, for example, directly influences the optimal setting for the gain controls. A systematic adjustment procedure can be used, followed by fine-tuning during examinations. One systematic approach is as follows: 1. Turn all controls down. 2. Adjust the brightness control until a light background just begins to appear on the screen. 3. Adjust the contrast according to the gray-scale bar. 4. Adjust the overall gain until the major portion of the screen is optimally saturated (sonogram A). | 5. Adjust the near gain so that the top area of the screen matches the area beneath it (sonogram B). 6. Adjust the far gain until echoes begin to appear on the bottom of the screen (sonogram F). 80 Chapter 6 6.3E Freeze control Side-by-side freezing of two images on an Equisonics 300. A Day-29 corpus luteum (left) and embryonic vesicle (right) were photographed simultaneously. Modern scanners have a provision for freezing an image, allowing study or photography. The control button may be located on the console or in a remote box. We prefer to have the control button on the console, because we normally have the console close to the operator (Section 8.3). Freezing is done through an image-storage function known as freeze frame memory. In some scanners, the freeze frame memory may involve only one sweep of beams across the transducer. Therefore, only every other scan line is seen and the resulting image is of low quality. This problem may also occur when attempting to freeze a videotaped image for detailed study or for still photography. In some scanners, several images may be frozen simultaneously and recalled as needed. A side-by-side provision on some scanners, shown above, allows the operator to freeze an image and then to continue scanning by means of an adjacent image. This provision is especially useful for comparing areas or to select the most desirable image for photography. In addition, a structure too wide for one screen can be photographed on two contiguous screens. The images from digital scan converters will remain in storage until released from the memory or until the scanner is turned off. However, images from analog scan converters begin to deteriorate, slightly, after approximately 10 minutes. Instrumentation 6.4 Selecting a Make and Model Points to ponder in selecting a scanner. 1. General SS Ronn so & & . Linear-array versus sector transducers (Sections 3.1 and 6.1) . Transducer and cable design (Section 6.1) . Transducer frequency (Section 6.2) . Portability (Section 6.3A) Annotation and labeling provisions (Section 6.3B) . Measuring provisions (Section 6.3B) . Freeze frame memory provisions and quality (Section 6.3E) . Zoom and magnification provisions and quality (Section 4.5) Photography and videotaping provisions (Chapter 7) Battery power-pack option 2. Quality of images a Specifications 1) Number of elements (Section 3.1) 2) Frame rate (Sections 3.3 and 4.5) 3) Focusing methods (Sections 3.6 and 3.7) 4) Number of shades of gray (Section 4.3) . Results of personal trial 1) Images smooth, pleasant, and free from checkered appearance 2) Able to meet specific needs. Examples: a) Detect mature corpus luteum b) Detect small follicles (e.g., 3 mm) c) Detect small embryonic vesicles - 3. Durability and reliability of scanner 4. Durability and service record of the company 5. Suitability for other uses 81 82 Chapter 6 The purchase of an ultrasound scanner is a major investment (e.g., $16,000), so potential buyers are justified in thoroughly researching the purchase. Veterinary ultrasonography is a new and volatile marketing area. There have been several turnovers in companies or marketing lines, and technological changes are frequent. This chapter, as well as most chapters, contains information that should be useful in deciding whether or not to purchase a scanner and in selecting the most appropriate make and model. References to appropriate sections are noted in the above list. Because of the magnitude of the investment, prospective buyers should expect to have scanners demonstrated to them, preferably on site. This approach combined with consultations with experienced colleagues and study of published information should enable a potential buyer to make reasonable decisions. Makes and models of scanners vary widely in capabilities and incidental provisions, as listed above under Topic 1. All individuals will desire high quality and good service, for example, but may differ in their general needs. This is especially true for clinical versus research requirements. Researchers may want elaborate provisions for annotations, measurement (e.g., area), photography and videotaping, and multiple freeze frame memories. Clinicians may prefer less elaborate schemes, especially if it means a lower price and fewer breakdowns. Provisions should be consistent with the needs --- neither inadequate nor excessive. Some scanners have zoom controls to enlarge a selected area, or magnification controls to enlarge the entire image. However, the resulting images may be poor, especially if the pixels (Section 4.4) become so large that they are visible. Postprocessing magnification of a portion (zoom control) or all (magnification control) of an image in such scanners may provide no real improvement in resolution --- just an increase in size. Other scanners may use all or most of the pixels on the screen when providing an enlarged area, and therefore there is a true increase in resolution. In regard to linear-array versus sector systems, it should be noted that some scanners can be used with either type of transducer. We prefer the linear-array system for imaging the reproductive tract of mares because the resulting cross-sectional images of the uterine horns serve as guides to assure that the entire length of each uterine horn has been searched (Sections 8.4 and 12.1). Our experience with the sector system, however, is limited. For comparisons of linear- array and sector transducers refer to Section 3.1. The expected image quality (Topic 2 in the above list) should take into account the scanners’ specifications, especially in regard to meeting certain minimums. A scanner should not, however, be selected on the basis of its specifications, alone. For example, a scanner with many elements will provide poor quality if other aspects “a ATTN ay a Instrumentation 83 of the engineering (e.g., damping, electrical insulation) are poor. Good resolving power and penetrating capability are not assured on the basis of transducer frequency. Many other factors are involved. It is emphasized that the most reliable information on quality is obtained from a personal trial under the potential buyer's conditions. Ideally, a potential buyer should learn first what a good image looks like by observing images on scanners owned by experienced colleagues or scientists. Ultrasonographers in human medicine also may be conveniently available for this purpose. A buyer should require that the images are smooth, pleasant, and free from the obnoxious checkered appearance of large pixels (Section 4.4). Buyers can prepare a list of needs and then determine by trial whether the scanner performs as desired. An excellent criterion for evaluating the quality of a scanner centers around its ability to consistently image a developing or mature corpus luteum. It is suggested that several horses be selected with known date and side of Ovulation. The -mares should not exceed Day 10, because the corpus luteum may become undetectable in an occasional mare after Day 10. The buyer then can test the ability of the scanner to consistently present a clear and sharp image of the corpus luteum. Poor scanners, even those with a 5.0 MHz transducer, will fail the test. Topics 3 and 4 in the above list concern the durability of the scanner and the reliability and speed of the repair service. Ultrasound scanners are highly specialized instruments and local repair service usually will not be available. Consultation with owners and users is probably the best way to obtain information on the relative durability of the model and the service record of the company. The selection of a scanner for transrectal use also may include consideration of other possible uses (Topic 5). Scanners are being adapted for use in many other species and body systems. Other clinical applications include diagnosing and monitoring: 1) urinary bladder calculi and other diseases, 2) aortic aneurysms, 3) heart diseases, 4) abdominal problems such as biliary and renal calculi, ascites, and neoplasia, 5) joint and tendon problems, and 6) eye or orbital diseases or injuries. The suitability of ultrasonography for evaluating the reproductive tract of stallions needs to be studied. If a scanner is to be used heavily for purposes other than imaging the reproductive tract, sector transducers or Systems that are compatible with both sector and linear-array transducers may have advantages. 84 Chapter 6 i Ci. mo SP Gas = icy roe i mo inti a Sa nt cae yoy £ a Images of an ultrasound phantom taken with an Acuson 128. This scanner is used in human medicine and is not equipped with a transducer foi transrectal use in horses. A 5.0 MHz, phased-array, sector transducer was used to produce the images. The phantom is the same one used with a veterinary scanner for producing the images shown in Sections 6.2 and 6.3C. The fluid-filled objects are 2, 5, 10, and 20 mm in diameter. The Acuson 128 has a control for varying the focal depth. Settings of 4 cm and 7 cm were used for the images on the top and bottom, respectively, as indicated by the arrowheads on the centimeter scale. The values for several variables are automatically displayed (upper right). The curved vertical line _on the right. represents the TGC (total gain control) settings, and the diagonal line represents the power slope. Note the sharp outlines of the fluid-filled objects and the near absence of artifactual echoes within the fluid. The resolution and flexibility of this scanner far exceed those of currently available instruments on the veterinary Instrumentation 85 arket --- but so does the cost (e.g., 10 times more expensive). Perhaps in the future ch scanners will be technically and financially adapted for practical use in large- imal theriogenology. FERENCES Zagzebski, J. A. 1983. Properties of ultrasound transducers. In Textbook of Diagnostic Ultrasound. Ed. Hagen-Ansert. C. V. Mosby, St. Louis, MO. VMcDicken, W. N. 1981. Diagnostic Ultrasound: Principles and Use of nstruments. 2nd Ed., John Wiley & Sons, New York, NY. Chapter 7 HARD COPY R. A. Pierson ‘ard copy is defined as the readable printed copy from a machine, such as a iputer. In ultrasonography, the term is commonly used for photographs or sotapes of images. Communication and documentation of results are important ects of ultrasonography. Specialized instruments and knowledge are required to luce high-quality hard copy, whether a Polaroid image for a mare owner's crapbook or a set of images for a professional or scientific publication. The aration of projection slides and videotapes is very useful, if not a requirement, sducational purposes both for the clinician and researcher. Scientific or lay ‘cles on research or clinical reports involving ultrasound are usually very ndent on the photographic skills of the ultrasonographer and the publisher. It is sssing to produce high-quality images only to have much of the quality and ition lost in the production of photographs or videotapes. It is even more essing to succeed in producing a high-quality photograph only to have much of » Quality lost during duplication in a publication. his chapter describes the instruments and techniques commonly used in ‘‘erinary ultrasonography for recording and storing images. Attention is given to > following: 1) Polaroid photography, because this rapid, but expensive, proach is widely used by veterinarians; 2) negative-based photography, including nization of files and record keeping; 3) preparation of slides for projection; 4) production of videotapes and preparation of photographs from videotapes. © alternate forms of hard copy which are not widely used in the veterinary field are noted. These include thermal printers and multi-format cameras. In the uction of photographs, a freeze-frame memory on the scanner is used. The ‘ity of the photographs therefore depends upon the quality of the real-time zes and the quality of the freeze-frame provision (Section 6.3E), as well as the ity of photographic instrumentation and technique. 88 Chapter 7 7.1 Polaroid Photography Examples of Polaroid cameras used with veterinary ultrasou instruments. These cameras are used with the Technicare 210DX, Equisonics < and Equisonics 210 (left to right). In some systems, the camera shroud is hand-! over the viewing screen; in others, the shroud sets on a hinge system which prov: a stable support for the camera. Polaroid film is much more expensive t negative-based film. In this regard, at least one scanner used for transrectal imas of the mare reproductive tract has a left-to-right switching mechanism so that images can be recorded on one film. This provision reduces the photographic cx Polaroid photography provides almost immediate viewing of the images. Th advantageous in practice situations and in some research applications. Pola images are also an effective way of backing up other photographic systems, when essential that a record of the ultrasound scan not be lost. Unexposed Polaroid films have a shelf-life of approximately nine months at r temperature. The processing chemistry integral to instant film is the limiting fa in storing the film. Freezing Polaroid films will not appreciably extend the lif the film for this reason. Archival qualities of developed Polaroid film are depenc upon storage conditions. If images are stored in acid-free paper under contro temperature, light, and humidity, they should last indefinitely. However, if store traditional paper files, the life of the print may not exceed five years. Prints exposed to light and fluctuations in temperature and humidity will fade appreciably within the Hard Copy 89 irst year, although some Polaroid films may be coated to increase the archival ualities of the print. Film types 611 and 667 were designed for photographing gray-scale images. ype 611 provides extended gray-scale information, but is more difficult to use ecause it has no effective ISO (film-speed) rating. Shutter-speed and aperture justments are made from basic recommendations for exposure, included with the m. Settings depend on the type of viewing screen and brightness and contrast ‘tings. The camera controls are adjusted from the initial recommendation cording to the ultrasonographer's interpretation of the gray scale. Film type 667 is cher in contrast and is rated at 3000 ISO. The short exposure times with this film - advantageous in systems in which the camera shroud is hand held against the wing screen or when there is a likelihood of vibrations in the camera-scanner nplex. Technical questions about specific applications and characteristics of Polaroid as may be addressed to Polaroid Technical Hotline at (800) 225-1618. xamples of sequential Polaroid photographs made by using a ‘ching mechanism for photographing two images on one film. The of sonograms on the left are of Day-48 unilaterally fixed twins. The same tographs taken with a negative-based film are shown and described in Section 5D. The pair of sonograms on the right are of Day-14 and Day-15 twin ryonic vesicles. 90 Chapter 7 me + BS eB. CRIES el A series of cropped Polaroid photographs, showing the sequential he early conceptus (1). The number in changes in ultrasonic anatomy of t the lower-right corner is the number of days from ovulation. This series represents the original detailed study of the changing ultrasonic anatomy of the early conceptus. Film type 667 was used. Hard Copy 91 1.2 Negative-Based Photography Examples of customized 35 mm camera systems. In one system, the ‘Mera is fitted to a shroud that attaches to the scanner as in some Polaroid systems. ne shroud is closed and an exposure made. The camera and shroud may be either moved or swung away from the screen on its mounting hinges. The second system an enclosed auxiliary arrangement, consisting of a high-resolution six-inch 92 Chapter 7 television monitor, close-focusing lens, and camera body. This system is attached to the ultrasound scanner through video connections and acts as a slave monitor. In both systems, the desired image must be "frozen" on the ultrasound viewing screen before the exposures are made. In scanners with multiple freeze-frame memories. side-by-side images can be taken with one exposure. Both systems provide high quality hard copy. The auxiliary system is more suited to research needs 01 centralized examining areas. In our laboratory, the record-keeping system involves an entry onto ; photographic record sheet as each exposure is made. There is a separate line for eac! exposure and a separate page for each roll of film. Thus, each exposure is identifie: with animal number, project, date, and general description. Depending on the dat: display or annotation capabilities of the ultrasound scanner, each negative contain specific information that uniquely identifies it and may be used to cross reference th: description in the log book. After film processing, the negatives are placed i transparent polyethylene sheets that hold an entire roll of 36 exposures (Print Fil: Archival Negative Preservers, Photo Plastic Products, Orlando, FL). Contact sheets which display all of the images on the roll of film on one 8 x 10-inch sheet o photographic paper, may be made without removing the negatives from the holder The log sheets, negatives, and contact sheets are stored in three-ring binders for eas} reference. When desired, selected images may be photographically enlarged fo further study or publication. Individual ultrasound images may be viewed a contact-size or enlarged prints. For display purposes we have enlarged some image: to 16 x 20 inches, with excellent results. Many emulsions currently available in 35 mm film can record the range of gray. scale shades used in veterinary ultrasonography. More than a cursory description oi 35 mm films is beyond the scope of this text. Detailed descriptions of films and thei: individual characteristics are available in the photographic literature. An excellent starting point is the Photographic Lab Handbook, Sth edition by J. S. Carro! (American Photographic Book Publishing Company, Inc., New York, NY). It is our current practice to photographically record ultrasound images on Plus-X Panchromatic film (Eastman Kodak, Rochester, NY). This film appears to give accurate rendition of the gray shades encountered in our use of ultrasound. It is rated at ISO 125 which, with our systems, allows us to use a medium aperture with a reasonably fast shutter speed (for example, £5.6 at 1/2 second). This is important because the central portion of most lenses is sharper than the edge and because longer exposures may result in blurring, especially if the stand for the camera-ultrasound complex is attached to the restraining chute. Faster (e.g., Kodak Tri-X, ISO 400) Hard Copy 93 ad slower (e.g., Kodak Pan-X, ISO 32) films are also available. The increased -ain in Tri-X negatives makes high-quality enlargements unlikely, and the lengthy not available in 35 mm format. Although larger-format camera systems are aptable to the systems used for recording ultrasound images, they are more -xpensive and generally more difficult to use. Technical questions about Kodak films may be answered by contacting Kodak echnical Service at (800) 242-2424. | 3 Projection Slides There are five methods of preparing projection slides of ultrasound images. 1) Ektachrome ER or EPR slide film (Eastman Kodak, Rochester, NY) may be used directly in the camera system. 2) Photographic prints prepared from negatives may be rephotographed using a copy stand and tungsten-balanced Ektachrome EPY or EPT slide film (Eastman Kodak). 3) Pan-X film may be used in the camera system and developed with a reverse processing kit available through Kodak product distributors. 4) Photographic negatives may be temporarily mounted in slide holders and rephotographed using a slide duplicator and negative film. The result is a positive image suitable for projection. 5) A 35 mm instant-slide system that was recently released by Polaroid Corporation may be used. The entire roll of slide film is developed --- in only a few minutes --- with a separate processing kit. We have used all five methods, and each produced acceptable results. We prefer to use the second method because of the reduction in labor involved in preparing the same images for projection slides and publication prints. If images are recorded on X-ray or diagnostic imaging film using a multiformat camera, slides can be prepared by rephotographing the transilluminated images. 94 Chapter 7 7.4 Videotapes An example of a three-quarter-inch videotape recorder and high resolution 19-inch monochrome video monitor. The videocassette record is a Sony U-matic, VO-5600, and the monitor is a Panasonic Video Monitor W‘ 5490. Three-quarter-inch videotape recorders have resolution of 330 lines | monochrome and cost $1,800 to $2,500. The quality of ultrasound images recorde on three-quarter-inch videotape approaches that of the image on the ultrasour viewing screen. The three-quarter-inch system, therefore, is well-suited for most research situations. Videotape recorders are available in two one-half-inch formats --- Beta and VHS. The one-half inch formats are less expensive and are more commonly available. Both tape formats have resolution of approximately 260 lines in monochrome and may be purchased for $400 to $2,000 depending on the desired options. They are adequate for many purposes not requiring great detail. Video recorders specially designed for medical applications (e.g., Panasonic NB 9240 XD) are capable of resolving 500 lines, although the cost of the machine is higher (approximately $4,500). Some recorders have a shuttle knob for low- or high-speed search scan. The ability to do frame-by-frame analysis with these recorders is often desirable in research. Broadcast quality video systems use one- and two-inch videotapes but are prohibitively expensive (e.g., $50,000 and up) for routine taping of ultrasound images. Hard Copy 95 Videotape recording is an excellent means of storing real-time ultrasound scans. ideo recorders may easily be connected to ultrasound instruments that have digital an converters and video monitors. However, this System is not adaptable to older anners that use analog scan converters (Section 4.1). Also, as noted in Section 4.6, some scanners the video format is upright when played back through the scanner, it is rotated 90° when played back through a conventional System. Many ultrasound examinations can be Stored on a sin : five-minute scans may be made on a one-hour three- ch videotapes in VHS or Beta format can store two, four, or six hours of aminations, depending upon the tape speed. However, there is a significant loss of ‘age quality when extended-play recording options are used. Taped ultrasound scans can be reviewed on the video monitor of the scanner or on “eparate monitor. Household television sets are adequate for some purposes. However, if more detail is desired, high-resolution monitors are used. High- resolution monitors are loosely defined as having more than 500 lines. Most commercial high-resolution monitors have 650 to 850 lines of resolution and are commonly available with seven- to 19-inch screens. High have underscan Capability, gle videotape. For example, quarter-inch tape. One-half- “quality monitors generally which allows all the information recorded on the viceotape to be displayed on the monitor screen. When Standard-scan monitors, such as a conventional television set, are used, there is a loss of 10 to 15 percent of the ‘.ormation at the periphery. In lower-quality monitors, imprecise registration of projection system causes "blooming," or distortion of the image. Exceedingly ‘1 contrast settings also may result in blooming. 96 Chapter 7 Sequential photographs from a videotape with time coding. The tin code is placed on each frame, as shown. The numbers indicate (left to right) the taj number, minutes, seconds, and frame number (30 frames/second). Note that the tv images that were selected for photography were taken at an interval of 38 seconc The time recordings show the overall movement of twin embryonic vesicles relati to one another during this time. We routinely install time code on the video track of a duplicate videotape and the audio track of the original tape. Thus, the original tape does not conta distracting numbers, but precise frames can be identified by reference to the cod copy. Time code is an eight-bit audio signal that can be converted to a video sign: It is placed on the tape by a time-code generator and represents an industrial standa (SMPTE) for elapsed time. The code is generated at 2,400 bits per second; 80 bi per frame at 30 frames per second (video play speed). There are even and ocd frames for each field. Time code is a highly accurate 24-hour clock that establishe precisely the time for each frame. An individual frame or series of frames can be located easily in the visual field on duplicate tapes with this system. Computerized editing facilities also can be used. The time-code system is very useful for editing or for analyzing motion, as in the assessment of uterine contractions or conceptus mobility, as shown. Assistance in developing our system was provided by J. Schultz in the video production department of the Instructional Media Center at the University of Wisconsin. Other universities likely provide similar assistance. Hard Copy 97 7.5 Other Hard Copy Methods Multiformat cameras are capable of recording very high-quality images, but they aust be viewed with transmitted light as are X-ray films. This system is commonly ised in hospitals which are equipped with X-ray viewing screens and by operators vho are accustomed to viewing X-ray films. Some examples of multiformat -ameras are those made by Dunn, Schiff, and Matrix instrument companies. Prices ary from $7,000 to $15,000, depending on individual specifications, such as single ormat (fixed-lens system) or multiformat (lens system changes to fit several ormats). Images are sent to the camera either through video connections (digital can converters) or XYZ connections (analog scan converters) and are recorded on 8 10-inch X-ray film. Commonly, four, six, nine, or 25 images are placed on one ieet. The size of an image is approximately four by five inches for the four-on-one ts ormat and approximately the size of a 35 mm slide for the 25-on-one format. : Thermal printers (e.g., Mitsubishi P 50 U) transfer images from the ultrasound instrument to a printer by a screen dump of one video frame (two video fields) using video connections. The image is printed onto thermally sensitized paper using a 280 x 234 dot matrix. The unit is capable of printing 16 shades of gray. Archival qualities of the paper are not precisely known, and the image will fade rapidly when subjected to elevated temperature. This type of copy provides a very rapid set of images at an extremely low cost. The system may be useful in research applications of motion analysis, area calculation, and fetal movement. Floppy disks also are used occasionally. They require a specialized recording ‘strument and must be played back through the scanner. Multiformat cameras, thermal printers, and floppy disk storage are not ommonly used in veterinary applications, although at least one veterinary ‘ltrasound company offers a thermal-printer option. X shape of the excised tract lying on a table, as shown. As the transducer is moved ‘ward over the surface of the organs, the cervix and uterine body are seen in igitudinal view. Then, when the transducer is swept to the one side, a horn is seen ially in cross-section. Usually the ovaries are viewed from the medial surface. parently the transducer is oriented over the dorsal, dorso-medial, or medial ‘ace of a uterine horn, and therefore any of these aspects of the horn can be ard the top of the image. However, because the horns may lie on the irregular ‘ace of viscera, a section of horn can be twisted, causing a change in the expected ntation. More details on these relationships are given in the chapters on the ries, uterus, and the single embryo. ess L438 G3) UZ The Coupling Medium Dipping the transducer and hand into a lubricant and coupling lium. The reflection coefficient of air approaches 100%. Therefore, air between the transducer face and the skin or rectal wall blocks the entry of the u‘rasonic pulses into the tissues. When a transducer is to be applied to skin, the hair is clipped and a liberal application of a coupling medium is applied to prevent air interferences. Because the rectal wall usually is moist, using a coupling medium is not as important for transrectal examinations. However, better contact sometimes is ot tained if a coupling medium is used. In addition, the coupling medium also can 104 Chapter 8 serve as a lubricant, which is needed for intrarectal examinations. Either an oil o water-soluble gel may be used for both purposes --- lubricating and coupling. Th anal area is lubricated with the gel before inserting the hand, and the transducer : dipped into the gel. In our laboratory, a gel is prepared from carboxy methylcellulose. The powder (carboxy-methylcellulose 7H3-SF) for preparing ‘h gel may be obtained from Hercules, Inc., Wilmington, DE (phone, 800-441-7600) The powder is combined with water using a mixer (e.g., paint mixer). One pound o powder will yield approximately five gallons of gel. Fecal material and excess x¢ should be removed from the transducer before they dry. Inserting the lubricated hand and transducer into the rectum requires the sam precautions that are used during rectal palpation to prevent rectal tears and rupture: Presumably, ultrasound examination has an advantage over rectal palpation in ‘ regard, because folds of the rectal wall are not grasped as in palpating. Howe\ special precautions should be taken to ensure that the transducer is advanced wit! the lumen of the rectum and not into a blind pocket. This can be done by plac fingers beyond the tip of the transducer during major forward movements. 8.3 Centralized Examining Areas Walk through 1’ FG=Front gate, 6’ high fe KG=Kick gate, 32” high Scanner: height of operator’s annus \ aiid Stand | Scanner. Techniques 105 Example of a centralized work area with placement of scanner next to 1 restraining chute. Although modern veterinary ultrasound scanners are nortable, they provide much incentive for the development of a centralized ‘xamining area on breeding farms. In this example, the mares are corralled as a ‘Toup into an area which leads into a long lane. The corralling area is wide enough 9 that the operator, without entering the corral, can prod the mares into the lane. he lane is filled with mares, and then a mare is boxed into a restraining chute. The ine can be modified to any length; it can contain any number of restraining chutes ositioned end-to-end, as shown in the example for the first two positions. A onvenient chute is selected for placement of the scanner. If preferred, an individual are can be led into an isolated chute. A platform or cabinet can be built immediately next to the rear of the Chute, as own. The height of the platform should position the scanner so that the screen and trols are approximately at eye level. The scanner should be close to the chute so ‘iat the controls can be adjusted and the details on the screen Closely scrutinized by © operator while the transducer is being held in position. The platform should be ‘idependent of the chute, so that it will not vibrate when the animal moves. Selection o! the method of restraint and the design of an examining area must give consideration to the protection of the instrument as well as those who will use it. ibdued light is another requirement, because visible image details are diminished if © screen is directly exposed to light. Light from doors and windows should not hit ‘ne screen directly and should not hit the operator's eyes when the screen is being ewed. Examination areas that were designed primarily for rectal palpation may erefore require some modifications. Preparing a mare for transrectal ultrasound examination (restraint, evacuation of -tum) is similar to preparing for transrectal palpation. Fecal material can cause stortions on the ultrasound image and must be removed. A pronounced shadow tending from the upper edge of the image may be due to intervening fecal matter ection 5.3). Often the debris may be removed by running a finger over the face of & transducer without removing the transducer from the rectum. The use of a -parate electric circuit may help reduce interference from simultaneous use of ‘ectric appliances (e.g., refrigerators). Some makes of scanners are available with attery power packs for use in areas without electric lines. 23 Sy ei Y Oita iS fey 106 Chapter 8 8.4 Examining Technique Medial surface Dorsal surface Right ovary Right uterine horn A systematic transrectal examining technique. The broken lines represent the orientation of the face of a linear-array transducer over the surface of the organs. The arrows show direction of movement of the transducer over the reproductive tract. This approach allows examining each area twice --- while advancing the transducer and while returning it. As the transducer is advanced over the surface of the uterine body it is moved from side to side, as shown, to ensure viewing the entire width of the body. When the corpus-cornual junction is reached, the transducer is moved laterally along the length of a uterine horn. Sometimes the ovary is reached immediately upon leaving the tip of the horn, but usually it is necessary to move the transducer forward after leaving the horn. Each operator should develop a systematic procedure, depending on whether selected areas or the entire tract is of interest at any given examination. An alternate approach, for example, is to examine the entire uterus and then the ovaries. This modification allows the operator to concentrate on locating specific structures or changes in a given organ. Details for examining certain organs or structures are given in subsequent chapters. During the examination, the viewing screen is the center of visual attention. At the same time, the location and orientation of the transducer Techniques 107 and the resulting field of view are considered. This sense of hand-eye coordination, which relates the image to the tissues, becomes well- -developed with experience. .S Water Bath, Biopsy, and Transabdominal Techniques Images of excised ovaries taken in a water bath. A) Mature, solid corpus ‘cum (arrows); B) Corpus hemorrhagicum filled with a blood clot and fibrinous stwork; C) Two solid corpora lutea (arrows). The images were taken with the ansducer applied to excised ovaries in a water bath. The ovaries were manipulated / grasping the mesovarium so that the fingers did not obscure the areas of interest. ‘tifacts are more of a problem in water baths because of reflective surfaces, pecially when the imaging field encounters the sides of the container or air bubbles ‘om recent agitation. The water-bath technique is especially useful for research ad education. The water serves as a coupling medium, excluding air between the ‘‘ansducer face and the surface of the ovaries. Because water is an excellent conductor of electricity, manufacturers should be consulted regarding the insulating properties of their transducers. Cracked transducers or those with faulty insulation 108 Chapter 8 between the transducer and connecting cable should not be used. For educationa purposes or to study the echo texture of certain structures, the tissue can be sliced in plane approximating that of an image. Research uses include counting an classifying ovarian follicles without destroying the ovaries, and locating certai structures within organs as an aid in sampling for assay (e.g., follicular fluid) or histological purposes. Tissues that have been fixed for histology can be examined b moving the tissue to a container of physiological saline. We are currently using this technique to locate embryos within a fixed uterus as an aid to histological sectioning Ultrasound has been used to guide needles or instruments in human medicine since 1977 (2). Transducers with channels for insertion of a needle are used. The tip of the needle is visualized as it passes through the tissue to the target. Perhaps such a technique could be used in mares for sampling ovarian, uterine, or placenta! structures by introducing an instrument through a nonrectal route (flank, cervi vaginal fornix). Plastic and metal needles and instruments are hyperechogenic. V have used a scanner to visualize the positioning of a biopsy instrument in the equi uterus. Simulated structures also can be monitored by ultrasound. Simulate embryonic vesicles are being used to study the mechanisms involved in intrauterir mobility of the embryo (3). The vesicles are made by injecting water into sma rubber balloons made from the finger tips of surgical gloves. Contrast media ai used with X-ray technology but not with ultrasound. Small bubbles can be detecte when a fluid is injected quickly through a catheter (4). We have visualized th deposition of semen through an insemination pipette into the uterus of heifers. Although this text is devoted to transrectal examination of the reproductive trac a transabdominal approach can be used to visualize the equine fetus after 100 da; (5,6). In the most extensive study (6), excessive hair was clipped and a coupling ge: was used. The transducers (2.5 to 3.0 MHz) were placed in the inguinal area. Feta! parts, amniotic fluid, placental membranes, and motion patterns were visualized. Fetal heart rates decreased from 180 beats/minute at 100 days to 60 to 80 beats/minute two weeks before parturition. It was concluded that this approach was practical in later pregnancy for detecting fetal orientation and viability and the presence of twins. We were unable to visualize the fetus with this approach in fat ponies using a 3.5 MHz transducer. Perhaps the known deleterious effects of fat on ultrasonic beams interferes with the practicality of this approach in fat mares. eae Techniques 109 The transabdominal approach can be used for animals or Species that are too small intrarectal insertion and manipulation of the transducer. This approach is just inning to be investigated for such species as dogs, cats, sheep, and swine. In ne, for example, the gestational sac was distinguished as early as Day 19 after us (7). The technique, using a 3.5 MHz transducer, approached 100% accuracy liagnosing pregnancy after 22 days (8). In dogs, the embryos were observed at 10 (9). The ultrasound approach would be an excellent research tool for study varian and tubal development and transport of the egg in birds because of the : size of the egg and small size of the bird. i Transrectal Approach in Other Species ages of heifer ovaries. A,B) Follicles of various sizes. In image A, the ovary is delineated by arrows; the largest follicle is 12 mm and is surrounded by 12 follicles in the 3 to 4 mm range. C,D) Corpora lutea (arrows). In image D, the Corpus luteum is apparently regressing; as indicated by the increased echogenicity (compare with C). E,F) Stimulated ovaries in heifers undergoing a superovulation teemen. The ultrasound instrument (5.0 MHz) was judged effective for monitoring 110 Chapter 8 and evaluating ovarian follicles and corpora lutea in normal and superovulat heifers. The corpus luteum became visible approximately three days after ovulati and was identifiable throughout the rest of the interovulatory interval. In two of the five heifers, the corresponding corpus albicans was identified for three days after | second ovulation. Irregular, nonechogenic areas were seen on the images of uter: horns during the periovulatory period. These nonechogenic areas were presumat due to intraluminal fluids because they coincided with the discharge of clear, ViSCC mucus preceding ovulation and blood-tinged mucus after ovulation. Normal an: abnormal aspects of folliculogenesis, the ovulatory process, luteinization, ; uterine changes during various reproductive states are now being studied extensiv by ultrasound in cattle. sities nae KS ee) pe Images of bovine embryos (11). A 5.0 MHz transducer was used. The number in the lower right corner of each image is the day of pregnancy (number of Techniques 111 1ys after ovulation). The scale on the left of each image is in 10-mm increments. ay 17. The elongated, nonechogenic embryonic vesicle (arrow) is visualized ‘hin an approximately cross-sectional view of a uterine horn (22 mm, round, gray ttled area). Day 23. The embryonic vesicle has increased in size. Convolutions the uterine horn prevented visualization of the entire vesicle in one view. Day 30. e embryo proper (arrow) within the embryonic vesicle (black). Day 35. The oryo and umbilical cord (arrow) imaged in an approximately frontal plane. Day A sagittal view, showing embryo, amnion (upper arrow), and chorioallantoic mbrane (lower arrow). The chorionic vesicle extended into the cranial and dal portions of the curved uterine horn. Day 48. Sagittal view of the fetus iowing head, front limbs (arrow), rear limbs and umbilical cord, and tail. discrete, nonechogenic areas were first visible within the uterus between Days 12 ind 14, when they were approximately 2 mm in diameter. These discrete structures vere identified as the embryonic vesicle, because they were observed only in heifers ater confirmed pregnant and were always in the uterine horn ipsilateral to the ‘orpus luteum. The presence of an embryo within the embryonic vesicle was ciected as a hyperechogenic area with rhythmic pulsations (heartbeat). The moryonic vesicle gradually increased in length from the day of first observation ntil Day 26 when it extended past the curvature of the horn and began to encroach ito the contralateral horn. By Day 32, the vesicle extended to the tip of the ontralateral horn in all heifers. The embryo proper was first visible between Days > and 29 when the mean length was 10 mm. The embryo increased in length an age of 1.1 mm per day. A heartbeat was detectable in the embryo on the first observed. In one superovulated heifer, five vesicles were visible in the uterine is by Day 14, and by Day 33 seven embryos were observed; two of the seven TyOs were lost by Day 43. ‘Tansrectal scanning can be done in any species or individual large enough to mmodate rectal palpation. We have imaged the conceptus and ovaries in large S with a handheld intrarectal transducer. In small sows and gilts, we have inserted the transducer into the rectum by stiffening the coaxial cable with a heavy pce Of hose and manipulating the transducer, externally. Presumably, such an é) proach could also be used in sheep and goats. The value of such techniques and the possibility of damage to the rectal tissue have not been explored. Recently, however, the use of an intrarectal transducer (5.0 MHz) with a stiffened coaxial cable for diagnosing pregnancy in sows was reported in an abstract (12); 10 of 10 sows were correctly diagnosed pregnant at 12 to 20 days (mean day of first detection, 15.4 +0.7 iS Livery 112 Chapter 8 days). The same workers obtained a mean day of detection of 29.6 +7.1, using the transabdominal approach (3.5 MHz transducer). Exciting research findings anc clinical applications for ultrasonic imaging of the reproductive tract in the non equine species can be expected in the future --- experimentation has just begun. REFERENCES 1. Ginther, O. J. 1979. Reproductive Biology of the Mare: Basic and Applied Aspects. Equiservices, Garfoot Rd., Cross Plains, WI. 2. Holm, H. H. 1982. Real-time biopsy techniques. In Real-Time Ultrasonography, Ed. F. Winsberg and P. L. Cooperberg. Church! Livingston, New York, NY. 3. Ginther, O. J. 1985. Dynamic physical interactions between the equine embry 0 and uterus. Equine Vet. J. Suppl. 3:41-47. 4. McDicken, W. N. 1981. Diagnostic Ultrasound: Principles and Us yf Instruments. 2nd Ed. John Wiley & Sons, New York, NY. 5. O'Grady, J. P., C. H. Yeager, L. Findleton, J. Brown, and G. Esra. 1981. In utero visualization of the fetal horse by ultrasonic scanning. Equine Prcct. 3:45-49. Cu 6. Pipers, F. S. and C. S. Adams-Brendemuehl. 1984. Techniques an applications of transabdominal ultrasonography in the pregnant mare. J. Am. Vet. Med. Assoc. 185:766-771. 7. Irie, M., K. Ohmoto, and S. Kumagaya. 1984. Diagnosis of pregnancy in pigs by real-time ultrasonic B-mode scan. Jap. J. Zootech. Sci. 55:381-388. 8. Inaba, T., Y. Nakazima, N. Matsui, and T. Imori. 1983. Early pregnancy diagnosis in sows by ultrasonic linear electronic scanning. Theriogenology 20:97-101. 9. Cartee, R. E. and T. Rowles. 1984. Preliminary study of ultrasonic diagnosis of pregnancy and fetal development in the dog. Am. J. Vet. Res. 45: 1259-1265. Techniques 113 Q. Pierson, R. A. and O. J. Ginther. 1984. Ultrasonography of the bovine Ovary. Theriogenology 21:497-504. . Pierson, R. A. and O. J. Ginther. 1984. Ultrasonography for detection of pregnancy and study of embryonic development in heifers. Theriogenology 22:225-233. . Thayer, K. M., D. D. Zalesky, D. A. Knabe, and D. W. Forrest. 1985. Early pregnancy examination in sows by intrarectal and abdominal ultrasonic evaluation. J. Ultrasound in Med. (Suppl.) 4:186. Proceedings AIUM/SDMS Ann. Conv., Dallas, TX, October 8-1 1, 1985. wit Sn Case aul 35 Part Three " - ” , - o = a ree . er er a . Chapter 9 OVARIES Although the embryo has been the focus of attention for ultrasonography in es, some of the most profound clinical and research applications involve the ies. Follicles as small as 2-3 mm can be seen, and the corpus luteum can usually jentified throughout its functional life. It will become clear in the study of -hapter 10 on follicles and Chapter 11 on the corpus luteum that ultrasonic aluation of the ovaries is superior to digital evaluation by transrectal palpation. sh ct — UQ my ed rsa — tha inf Scar disc folli he ovaries are the master organs of the mare's reproductive tract. They duce the ova that justify the existence of the remainder of the tract. The care, ‘ization, and subsequent development of the ovum is assigned to the tubular ans, but the ovaries through their hormonal role integrate and control the ctions of the tubular genitalia. The complexity of form and function of the ovary eightened by its dual role. Its functions are gametogenic (development of metes or ova) and endocrine (production of hormones). Of the two principal arian components, the follicles play a dual role (production of ova and estrogens), creas the role of the corpus luteum is endocrine only (production of progestins). )vary's role as the master organ of the reproductive tract necessitates a ‘phology more dynamic than for any other organ in the body. A very large ure, 50 mm in diameter (mature follicle), can be here today and gone TOW (ovulation). Soon the follicle is replaced by a temporary luteal gland. - cyclic changes are reflected in the ultrasonic anatomy and are a basis for onically evaluating the mare's reproductive status or estimating the stage of an is cycle. iis chapter reviews those general aspects of the role and anatomy of the ovaries hould be known for effective scanning and for interpreting and utilizing the nation derived by ultrasonography. Emphasis is given to overall ovarian ling techniques. Use of ultrasound to detect ovarian abnormalities also is issed. Specific detailed information on the ultrasonic aspects of the ovarian les and ovulation is given in Chapter 10 and on the luteal glands in Chapter 11. w— ineasdaal 116 Chapter 9 9.1 Gross Anatomy 9.1A Attachment and orientation Drawing of lateral view of ovary and associated structures (1). amp = ampulla inf = infundibulum ist = isthmus luh = left uterine horn mo = mesovarium ms = mesosalpinx rl = round ligament tm = tubal membrane tuj = tubouterine junction The relationships among ovary, oviduct, ovarian bursa, and tip of the uterine horn are shown in normal position (left) and after exposure of the ovarian bursa and proper ligament of ovary (medial wall of bursa). The ovary is kidney-bean-shaped with the prominent ovulation fossa on the free or ventral border. Note that the broad attachment of the mesovarium is to the dorsal, greater curvature. The surfaces of the ovary between the free and attached borders are called the medial and lateral surfaces. The rectum is located on the midline between the left and right broad ligaments (Section 8.1). Therefore, an intrarectal transducer usually is applied to the medial or dorsal surface of an ovary (Section 9.3). The ovaries in the nonpregnant peo eae Ovaries B Relationship of cortical and medullary areas Dorsal curvature Mesovarium with Mesothelium vessels Medulla Cortex QOS NW Ovulation fossa with germinal epithelium 117 early pregnant mare may ride on the intestines, and the mesovarium may be loose. e orientation of the ovaries is therefore quite variable. Because of the extreme arian mobility, it can be difficult to firmly identify poles and surfaces by rectal pation --- and especially by ultrasound. However, the border areas are readily ntified by palpation and occasionally by ultrasound because of their characteristic ms (concave ventrally and convex dorsally). Because the ovaries may be lifted by intestines, their actual location, as well as orientation, is quite changeable. The ‘ance from ovary to uterine horn may vary from 0 to 5 cm. am jneasdaul iagram of a midsagittal view of ovary. The relationship between cortical noncortical areas of the ovary is unusual in the mare. The medullary or vascular 1s superficial, and the cortical zone, which contains the follicles (parenchyma), the interior of the gland, as shown. The parenchyma reaches the surface only at ovulation fossa on the ventral border. This is the only area from which ovulation rs and the germinal epithelium is confined to the fossa, as shown. The vessels nerves reach the ovary through the broad ligament and so enter the ovary at the sal border and spread out over the lateral and medial surfaces. The convex ace, therefore, is the hilus of the ovary. As shown in Section 9.1A, the chment of the mesovarium is very broad and extends for a considerable distance t the medial and lateral surfaces. The visceral layer of the peritoneum of the ry is loosely attached over more than half the surface of the ovary, and there is considerable fat and loose connective tissue between the visceral layer and the ovary. This area of loose attachment also contains a complex of arteries and veins. 118 Chapter 9 9.1C Follicles and luteal glands Intact trimmed ovary (A) and midsagittal sections (B-I). In A-H, the ovaries and sections are shown with the dorsal greater curvature (hilus) at the top. The attachment of the mesovarium to the greater curvature has been removed. The ventral border has the pronounced depression or ovulation fossa. Although the overall shape resembles that of a kidney bean, the shape and size change dramatically with the development and ovulation of large follicles. The occurrence of ovulation from the ovulation fossa affects the ultrasonic anatomy of the follicles and corpora lutea, as well as the overall shape of the ovary. Note that the large follicles (B,C) and a Ovaries 119 uteal glands (D-H) impinge on the fossa. The preovulatory follicle may reach the ossa by virtue of its large size (C) or perhaps special mechanisms may cause the ollicle to expand along a line of least resistance, so that a portion of the follicle saches the fossa. Bands of connective tissue radiate toward the fossa and may rovide boundaries for directing the follicle's growth. Following ovulation, the yriform follicular cavity may fill with blood (corpus hemorrhagicum; D). As the avity luteinizes, a distinct neck-like process may extend from the fossa to the main ody of the gland, resulting in a gourd- or mushroom-shaped corpus luteum (F,H). ne luteinized process varies considerably in length and prominence. The luteal gland of mares often is called the corpus hemorrhagicum, corpus ‘eum, and corpus albicans, as it moves through successive changes of development, aintenance, and regression. Recent ultrasound studies have shown, however, that any mares do not form a discernible corpus hemorrhagicum (Section 11.3). The 2 early corpus hemorrhagicum has the appearance of a blood clot (D). Luteal tissue : begins to form at the periphery of the cavity over the next few days and has a ‘rabeculated or folded appearance (E,F). The folded appearance probably results ‘rom the collapse of the large ovulatory follicle, so that the wall is thrown into folds which project toward the central cavity. In some mares, luteinization involves the entire structure, and the mature corpus luteum has a solid, although usually folded, appearance (G,H). Other mares develop a corpus hemorrhagicum, and luteinization accompanied by organization and eventual shrinkage of the central blood clot »—E,F). Organization of the clot involves the development of fibrinous networks ), which are often a distinctive feature of the ultrasonic anatomy. When luteal gression occurs, the gland takes on a lighter gross appearance because of creasing vascularization and increasing connective tissue. These changes result in ‘reased tissue density which also alters the ultrasonic anatomy. By the time estrus curs, the luteal structure is straw-colored. Regression of the corpus albicans ‘itinues during the subsequent diestrus. It eventually becomes a highly pigmented ‘eak with the long axis oriented toward the ovulation fossa. During pregnancy, various numbers of secondary corpora lutea may form, oeginning on approximately Day 40. The secondary glands may form following ‘her ovulation or luteinization of unovulated follicles. As a result, glands may be solid or contain a central cavity with a blood clot or nonhemorrhagic fluid (I). The secondary corpora lutea regress after Day 160, along with the primary corpus luteum. WitNwic Gn iusasdaal 120 Chapter 9 9.2 Form and Function 9.2A Estrous cycle | < Gonadotropins ‘ SO —LH 7 \ SoStFSH quccccccccccccccccccnccccscccuscscscssssseessseeanng,, °* ¢, ine “aN |A. x : Vous Progesterone curve i ola é iL; ‘ \ and Ei 7 ¢ ' ‘ : / ‘ \ Luteal dynamics; Fs peeks \ & / e ee Progesterone e J ht \ \ Estrogen curve : and Follicular dynamics ° | ° oe o ile | Growth of large fols_20 mm) follicles increase kedly between Days 10 to 60 of pregnancy and then decrease to low values by /S 180 to 200. The decrease in numbers of large follicles can be attributed to the nation of secondary corpora lutea beginning at approximately Day 40. Some of secondary corpora lutea (perhaps 30%) result from ovulations (representative in, 2.4 +0.5 ovulations/mare) and the remainder from luteinization of vulatory follicles. Ovulation usually is confined to Days 40 to 70. The mean ber of secondary corpora lutea in one Study was 2.8 at 70 days and 10.2 at 140 ; . Numbers increase linearly until both the secondary and primary corpora lutea | ress at approximately 180 days. he first portion of the luteal progesterone profile reflects the production of the mary Corpus luteum and the second portion reflects the formation and output of secondary corpora lutea plus the continued production of the primary corpus luteum. The regulation of the life span of the primary corpus luteum involves several sequential mechanisms. Presumably the factors regulating the corpus luteum ‘the first 12 or 14 days of pregnancy are similar to those operating during the ‘responding days of the estrous cycle. The divergence in progesterone curves ‘ween nonpregnant and pregnant mares at Days 12 to 14 involves the activation pregnant mare) or blockage (pregnant mare) of the uterine luteolytic ‘anism. The embryo retards the production or release of the uterine luteolysin, 2a, into the uterine veins. The importance to the veterinary ultrasonographer of ctivation or blockage of the uterine luteolytic mechanism includes the following iderations: 1) ultrasonic detection of a persisting corpus luteum (Section 5), 2) occurrence of both pseudopregnancy (prolonged interovulatory ‘vals) and shortened interovulatory intervals in association with ultrasonically detected embryonic loss before Day 15 (Chapter 15), 3) early luteolysis in association with ultrasonically detected, small, intraluminal uterine fluid collections (Chapter 15), and 4) role of the ultrasonically detected mobility of the embryonic vesicle. Because of its importance, the uterine luteolytic mechanism is discussed separately in the next section. — pate 124 Chapter 9 9.2C The uterine luteolytic mechanism CORPUS _— (F) LUTEUM EMBRYO ae a UTERINE LUTEOLYTIC MECHANISM The luteo-embryo-uterine triad. The corpus luteum produces progesterone, which is essential to the survival of the embryo (Chapter 15). In ‘he absence of an embryo, the uterus produces a luteolysin (prostaglandin Fra) tat induces luteal regression. The mare then returns to estrus and is afforded anotier opportunity to become pregnant. The embryo must block the uterine luteoly ic mechanism, so that the corpus luteum is maintained along with its production of ‘e vital hormone --- progesterone. Since progesterone begins to decline by approximately Day 12 in nonpregnant mares (Section 9.1C), the embryo must been blocking the uterine luteolytic mechanism by at least Day 11. Removal of ‘ie embryo on Day 15 or 16 by flushing results in luteal maintenance (2), suggesting ‘at blockage of luteolysis by the embryo is completed by Day 15 (1). More recent! it was found that uterine flushing caused the release of PGF 2a; however, the presence of an embryo inhibited such release when flushing was done on Day 11 or 12, but not when done earlier (3). The critical period during which the embryonic vesicle or its products must be present to effectively block luteolysis, therefore, apparently extends from approximately Day 11 to Day 14, inclusive (1). The luteolytic mechanism can be activated precociously by intrauterine foreign matter or pathological processes which irritate the endometrium, including infusion of fluids, plastic intrauterine devices, intrauterine manipulation of probes (biopsy,fiberoptic), and inflammation. This effect is due to early release of the luteolysin, PGF2a. For this reason, short estrous cycles may occur in response to uterine flushing procedures, digital dilation of the cervix, or endometritis. However, the corpus luteum is resistant to early induced regression during the first few days of its Seca cnneN enna uN SKN uenn inno Ovaries 125 velopment. Therefore, exogenous PGF2a does not cause luteolysis during the velopmental stage of the corpus luteum. SHEEP HORSE wes ivr Susncdans yp Ovarian artery Ovarian artery Ovarian vein ie Ovarian vein Comparison of arteries (clear) and veins (cross-bars) of a uterine n and adjacent ovary in a sheep and a horse (1). In sheep (and cattle), re is a unilateral pathway between a horn and the adjacent ovary for uterine- iced luteolysis. In these species, the ovarian artery is tortuous and closely applied ‘the vein that drains the uterine horn. Mares do not have a unilateral luteolytic away and use instead a whole-body pathway between uterus and ovaries. In res, the ovarian artery is relatively straight and caudal to the ovarian vein. These \parative differences in the anatomy of the utero-ovarian vascular pedicle vided some of the initial rationale for the hypothesis that the local pathway from a utcrine horn to the adjacent ovary was veno-arterial, involving the veins that drain uterus and the arteries that supply the ovary (4). That the luteolytic pathway is systemic in mares was suggested by the failures to demonstrate a unilateral pathway ‘h the techniques that successfully demonstrated a unilateral pathway in cattle and cep (4). These experimental approaches included unilateral hysterectomy, nilateral stimulation of the endometrium by foreign material, and administration of | GF2a by intrauterine versus systemic routes. (eal eI oo 126 Chapter 9 9.2D Anovulatory season Se No. of follicles < 20 mm . s a —=-NO. Of FOTLiCIES > 20 MM \ _Basccrereererereen, 6 —— se8 *-., o Oe ee a, wt Oo" @ 8 a: Ss re oO s ob = * S os = nt ay 2 4 ae s a “a, a a fie yee ° s pt 7 ° » ° ° oe : 2 °f ed i ee / 2s Zo 4 4 oO Moco Ped > beer see 9° Os os Be > eer Diameter of largest follicle baited = = —Mean diameter of all follicles a eo 20 ae o E © Oo 10 Jan 1 Feb 1 Mar 1 Apr 1 May Follicular changes before the first ovulation occurring in a group of 14 mares (1). During deep anestrus (e.g., January) the ovaries may be small and devoid of follicles >5 mm. Such ovaries are sometimes difficult to detect by ultrasound; rectal palpation may be needed. Termination of the anovulatory season involved a gradual increase in number of small follicles (<20 mm) during January, February, and March. During the last half of March, the number of small follicles receded and a rapid increase occurred in number of large follicles and diameter of largest follicle. In mares in which the first ovulation occurred in early spring with an associated prolonged estrus, there was an extended period of considerable follicular activity. Several days before ovulation the number of large follicles decreased, similar to what occurs during the ovulatory season, except that the period of decline in numbers of large follicles was more prolonged. The protracted period characterized by large follicles during the transition between anovulatory and ovulatory seasons is a serious obstacle to estimating when ovulation will occur. As 4 result, mares are often bred repeatedly during this time. SEIS EYEE Sr TT Ovaries 127 12 i ~~ al | 6 i} | pit Isd’s | _ 10 a Fi 1 oy = bot Ai 4 E ey i/ \ 42 - ps. Tee x — ] — 1 3 i i i j 4 2 2 1 at2iS Liasicai » Diameter of largest follicle (mm) ~~ No. of follicles 15-25 mm 60 52 44 36 28 20 12 41 Days prior to ovulation Mean concentrations of FSH and LH and follicular changes in 10 pony res during the transition between anovulatory and ovulatory seasons The decline in FSH preceded the decline in number of large follicles. Mean LH /es remained consistently low prior to Day -8 and then increased to high values by y -1 with an associated increase in diameter of the ovulatory follicle. The changes ceding the first ovulation seem to differ from those preceding subsequent lations in the following ways: 1) large follicles are present for a much longer period of time, 2) estrogen production occurs over a longer period of time, 3) the iiterval from FSH decline to ovulation is more prolonged, 4) the decrease in umber of large follicles occurs sooner, and 5) the magnitude of the preovulatory | curve is less for the first ovulation. 128 Chapter 9 9.3. Ovarian Scanning Techniques | Image of Position of ultrasonic transducer tissue slice F Medial (ca? . © Of © a Tb Ue San S oO»: © =>: Oo Caudal e Dorsal @a: © aw~s T = Os Sd = O>5 © O35: Oo a Caudal Relationships between orientation of transducer and the ultrasonic image. When the transducer is applied to the medial surface of the ovary, tic medial aspect of the image appears toward the top of the screen. The caudal pole of the ovary can appear on either the right or left of the screen, depending on whether the scanner is located on the operator's right or left and on the position of the inversion button (Section 6.3C). Similarly, when the transducer is on the dorsal surface, the corresponding surface on the image is toward the top of the screen. The ovary is quite mobile in its suspension from the mesovarium and often the orientation is not clear. The position of the ovary may change even while an examination is in progress. It is difficult to deliberately place the transducer in a given orientation on the ovary because of the extreme mobility. Fortunately, for most purposes, it is not important to view the ovary in a known plane. .pparently segments of intestine or fat may obscure the ova EES Ovaries 129 It is occasionally (less than 1% in one trial) necess ary to reposition the ovary by gital manipulation per rectum in order to locate it or to obtain a clear image. ry, requiring positioning. Occasionally an ovary that is not found with the transducer in one nd is readily found when the other hand is used. Generally, either Ovary can be ‘sily scanned with the transducer in either hand. In this regard, scanning the two aries with one hand is not as awkward as for palpating the two ovaries. Sometimes 7 in one trial), the orientation of the Ovary relative to the transducer may Cause an portant structure (large follicle, corpus luteum) to be missed. The ultrasound am samples sequential "slices" of the Ovary while the transducer is being rotated; re must be taken to ensure that the entire Ovary is exposed. The time required for scanning the ovaries is Similar to requirements for rectal ‘pation. For example, in one trial, the mean time from introduction of the nsducer into the rectum to location of an Ovary averaged four seconds (range, 3 to n=24). The mean time required to move from one Ovary to the other was also ir seconds. As with any type of examination, if the operator feels compelled to ry, mistakes will be made and structures will be missed. Images of ovaries. A, B) Two ovaries showing a longitudinal mid-sagittal age. The transducer was placed on the dorsal curvature extending from pole to ‘©. Note the kidney-bean shape and the ventral ovulation fossa. C) Image of an ary when the tranducer was placed on the medial surface extending from pole to ‘©. The medial surface is at the top of the image. 130 Chapter 9 9.4 Ultrasonic Detection of Abnormalities Ultrasonography is useful, not only for monitoring normal seasonal and cyclic events, but also for diagnosing ovarian irregularities and pathologicat changes (5). These include: 1) double ovulation, 2) ovulation failure, 3) quiet ovulation, 4) hemorrhagic follicles, 5) prolonged maintenance of the corpus luteum or pseudopregnancy, 6) ovarian tumors, and 7) cystic peri-ovarian structure Double ovulation is discussed in Chapter 16 and pseudopregnancy in Sections 11.48 and 15.4. Unfortunately, we have not had the opportunity to ultrasonically examin< mares with ovarian tumors, and published information is not available. Presumably. however, ultrasonic imaging could provide much information on the intern structure of ovarian tumors. Hemorrhagic anovulatory follicles from two mares. The echogenic line: are from fibrinous material that has compartmentalized the blood clots. Note that the large size of the structures exceeds the width of the ultrasound screen (55 mm). This is a form of apparent ovulatory failure, wherein the preovulatory follicle grows to an unusually large size (e.g., 70-100 mm), fails to ovulate but fills with blood, and gradually recedes (1). The resulting hematoma occasionally becomes extremely large. We have diagnosed this condition by ultrasonography in 12 mares. The blood became increasingly organized, and fibrinous echogenic bands formed in the clot. The unruptured hemorrhagic follicle may develop a complete or partial thin wall of luteal tissue or may remain devoid of ultrasonically or grossly (at necropsy) visible W Sir in Str fo: larg Stru and If the cyst is in the peri-ovarian tissue, however, it may readily be confused with an Ovaries 131 ‘eal tissue. The structures gradually regress and the mare may subsequently return estrus. This may be the same phenomenon that has been called autumn follicles. stic ovaries, comparable to what occurs in cattle, have not been tity in mares (1). Failure of ovulation or anovulatory estrus during the Ovulatory season occurs ‘asionally (incidence, 1-3%) (1). The ability to detect the absence of a Corpus cum by ultrasound makes the condition more subject to diagnosis. Fortunately, mare is a dependable ovulator once the transition between anovulatory and latory seasons is complete. Following the last ovulation of the year, mares may clop a large follicle at the expected time, but the follicle does not ovulate and the © enters the anovulatory season. documented as an TY bath. The orientation of the specimen and the orientation of the image are lar. The large cystic structure on the left is the dilated portion of the idibulum, and the structure on the right is the ovary. The Ovary contained a 25 follicle as shown on the ultrasound image. The cleavage line between the two tures is a line of adhesion of the infundibulum to the area around the ovulation 4. Based on.in vivo ultrasound examination alone, the cyst was mistaken for a © follicle. Rectal palpation, however, indicated that the largest fluid-filled -ture was probably not an integral part of the ovary. Since peri-ovarian cysts hydrosalpinx involve fluid-filled structures, they are detectable by ultrasound. Overian follicle, as in this case. Here, rectal palpation has an advantage, because the disital Ot enatm A m e ee LCM ‘ydrosalpinx in an excised tract and an ultrasound image taken in a Ahr See ke ge 132 Chapter 9 ovary. This example illustrates that when ultrasound is used as the primary technique for routine examination of the ovaries, rectal palpation skills should not be abandoned. Detection of normal oviducts does not seem feasible with available technology, but abnormal, fluid-filled oviducts may be detectable. REFERENCES 1. Ginther, O. J. 1979. Reproductive Biology of the Mare: Basic and Applied Aspects. Equiservices, Garfoot Rd., Cross Plains, WI. 2. Hershman, L. and R. H. Douglas. 1979. The critical period for the maternal recognition of pregnancy in pony mares. J. Reprod. Fert. Suppl. 27:395-40 3. Betteridge, K. J.. A. Renard, and A. K. Goff. 1985. Uterine prostaglancin release relative to embryo collection, transfer procedures and maintenance o/ ‘he corpus luteum. Equine Vet. J. Suppl. 3:25-33. 4. Ginther, O. J. 1981. Local versus systemic uteroovarian relationships in farm animals. Acta Vet. Suppl. 77:227-228. Int on ete Chapter 10 FOLLICLES he ovarian follicles of mares are excellent subjects for ultrasonic imaging use they are large, filled with fluid, and readily accessible by the transrectal . Ultrasound provides a rapid, noninvasive, and reliable method of measuring -ounting follicles for both clinical and research purposes, but some errors are ‘able and occasionally even a large follicle can be missed . Some of the potential nical applications of ultrasonic examination of the follicles include: 1) helping to ermine whether a mare has entered the ovulatory season, 2) aiding in estimating ‘age of the estrous cycle, 3) predicting the imminence of Ovulation, 4) ting double preovulatory-sized follicles that are in Close apposition and difficult scern by palpation, 5) detecting failure of ovulation or anovulatory estrus, 6) snitoring small follicles as an aid in judging whether ovarian Sterility or iescence has occurred, 7) evaluating whether a mare is in a condition to respond tcatments for follicular stimulation, and 8) monitoring the results of stimulatory ‘-atnents. Research applications center around the ability to sequentially monitor cular populations, including follicles as small as 2 mm, and the changes in ‘phology of individual follicles (e.g., shape, thickness of follicular wall). The arch potential is especially exciting because small follicles, including those y embedded within the ovary, can be morphologically monitored without \vading the ovarian tissue. is chapter considers the ultrasonic anatomy of follicles, the prospects for cting impending ovulation by ultrasonography, and the ultrasonically visible 2&8 associated with the ovulatory process. Recent research which utilized ‘Sonography to study the dynamics of follicular populations is reviewed. mation concerning multiple ovulations is given in Chapter 16 and information ctecting and evaluating the corpus luteum is given in Chapter 11. 134 Chapter 10 10.1 Ultrasonic Anatomy of Follicles CEG. RRRERAERRHRRRRRET TURRET eTPeTeT TTP SEE EENRRNRRNRETT ETS C POETS SG RAAT ORERERRT ATONE IRIE MN age atl MRI (ne Hill SH cma =) Ultrasound images of ovaries showing follicles of various shapes and diameters. The follicles are comparable to what may be seen during early diestrus (A), late diestrus (B), early estrus (C,D), and just before ovulation (E,F). A) Four follicles, 6-10 mm; two follicles (arrows), 3 and 4 mm. B) Four follicles 10-15 mm, two follicles (arrows), 5 mm. C) Three follicles, 15 and 20 mm; two follicles, 4 and 6mm. Several other small (2-3 mm) follicles are also visible. D) One follicle, 30 mm; three follicles 15-18 mm. E) Single preovulatory follicle. F) Double preovulatory follicles. Follicles, like other fluid-filled structures, appear on the ultrasound images as black (nonechogenic) or dark (hypoechogenic) areas. The follicles are roughly circumscribed. Irregular shapes are attributable to compression between adjacent follicles or between a follicle and the luteal structure or stroma. Some of the apposed walls of similar-sized adjacent follicles are straight. Follicles 135 ‘he apposing walls sometimes are not visible, Causing irregular forms consisting of wo or more follicles. Many of the missing walls were visible on the ultrasound nage during real-time scanning but were lost during freezin »production. Determining the diameter of follicles is Subject to variation because of the regular shapes. The diameter of follicles that are not spherical can be estimated by entally adjusting the irregular shape to an approximately equivalent circular form. ure is required to minimize confusing follicles with other nonechogenic areas, such the nonluteinized central portion of a corpus luteum. Follicles usually can be stinguished from other nonechogenic areas by a defined, relatively smooth outline. adow artifacts, caused by echoing of the sound waves from the side of a curved ‘face, may obscure the follicular outline parallel to the direction of beams (Section 5). Apparent beam-width artifacts (Section 5.6) also may obscure the follicular ‘lines. Follicles as small as 2-3 mm were seen with a 5 MHz transducer, whereas with the same ultrasound instrument equipped with a 3.5 MHz transducer, the minimum observable diameter was 6 mm (1). In an earlier report, the resolution of the instrument prevented detection of follicles that were less than 8 mm (2). 10.2 Reliability of Ultrasound Follicular Measurements LL Method of Mean diameter counting Mean number follicles for the following categories: of largest follicles 2-Smm 6-10mm 11-15mm 16-20mm 320mm follicle (mm) ee eee Necropsy 55 3.8 iS 0.3 0.3 17.0 Ultrasound ff 2.8 0.3 0.3 0.2 14.7 a Se ee ee Comparison of follicular diameters determined by slicing six ovaries necropsy versus in situ ultrasonography (1). The correlations between cropsy and ultrasound values were high for all end points (greater than 0.9). Ina evious study, there was good agreement between rectal palpation and ultrasound ‘stimations of diameters of large follicles (2). A slightly smaller estimate of ameter can be expected by ultrasonography, because only the follicular antrum is ‘easured, whereas other techniques include the walls. g and photographic mo meee seit 136 Chapter 10 10.3 Ultrasound Studies of Folliculogenesis 10.3A Estrous cycle NUMBER OF FOLLICLES 21 18 15 12 “3 OW, 24 oO oS 10.12 14.16 168.20. OF 3 DAYS Profiles of numbers of follicles of various sizes during the interovulatory interval (3). Daily examinations were done with a 5.0 MHz transducer in 40 single-ovulating mares. Data were normalized to the mean length of Follicles 137 ie interovulatory interval (22 days). There were significant differences among days ‘or number of follicles 2 to 5 mm, 16 to 20 mm, and >20 mm, but not for the other wo Categories (6 to 10 mm and 11 to 15 mm). The number of 2 to 5 mm follicles egan to increase just before ovulation. This increase corresponds temporally with 1¢ reported time of increase in FSH (Section 9.2A). The number of large follicles .6 to 20, and >20 mm) began to increase midway between ovulations. At the same me the number of 2 to 5 mm follicles decreased, indicating that small follicles were ‘owing into large follicles. Note that the large follicles decreased in number before vulation occurred. A decrease in number of large follicles through atresia would ontribute to the number of small follicles. “ae 2-10mm Ht, AY + NUMBER OF FOLLICLES rs er “eds DAYS Composite data showing the reciprocal relationship between numbers ‘ follicles 2 to 10 mm and >10 mm (3). Note the reciprocal relationship ‘between the two size classifications caused by cyclic patterns of growth and atresia. 138 Chapter 10 50 45 40 35 30 25 Largest follicle DIAMETER(mm) gt tttag ¥ pat di t follicle x ; nd largest fo 4. rs Phe gate? i 20 15 10 -3 Ov 2 4 6 8 10 12 14 16 18 20 OV 3 DAYS Mean diameters of largest and second largest follicles (3). Follicles both classifications increased in diameter at mid-cycle, corresponding to the time increase in number of large follicles. Note the divergence in the growth proii'c between the largest and second largest follicles beginning after Day 16 (six days before ovulation). This divergence temporally corresponds to the decrease 1 number of large follicles in the previous graph. Day -6, therefore, was the mean ~ 24 te E co . = dix es oh “> w1s = ES Q 12 6 11 w» 10 Lu 2 Oxo 3 9 a ae onl 7 i u 6 | ‘ O 8 o #5 ao 4 LJ : 220mm : speneDagr Paper OE GeO On 1 O os. 6 8 12°15. 18.2124 27 3032 30 DAYS Follicular profiles in nonpregnant and pregnant mares (unpublished). Daily examinations were made with a 3.5 MHz transducer. Data were discontinued in all nonpregnant mares on the day corresponding to the shortest interovulatory interval in the group, and in all pregnant mares on the day corresponding to the shortest interval from the primary ovulation to the first secondary ovulation. This event occurred on Day 17 in the nonpregnant group and on Day 36 in the pregnant Th firs trai pro’ 2 35 ~ 30 —~ 25 EIS!" i nce Follicles 141 up. Note the significant difference between groups in diameter of largest follicle Days 15, 16, and 17. Nonpregnant and pregnant mares did not differ in follicular files until the preovulatory growth spurt in the nonpregnant mares. Pregnant es apparently differ from nonpregnant mares in the lack of a selection ‘hanism for designating an ovulatory follicle and causing regression of other e follicles. In pregnant mares, the largest and second largest follicle and the iber of large follicles (>10 mm) continued to grow, reaching a plateau at oximately Day 20. Thereafter, there were no detectable waves in the mean les --- only a prolonged plateau. Ovulation occurred from this plateau, rather from a growth spurt by a selected follicle. 3C Before first ovulation of the year e@-e-e ist ovulation of year (n=8) ae b---B--f" : O-OO 2nd ovulation of year (n=8) _ 20 “1O.. “Oy eBxce?s cpg Sei 4 DAYS ameter of largest follicle before the first and second ovulations. liameter of the largest follicle was significantly greater for all days before the ovulation of the year, except on the last two days before ovulation. During the ‘tional period before the first ovulation, large follicles may be present for a acted duration, as shown, greatly hampering attempts to determine when to 142 Chapter 10 begin the breeding program. In some mares, especially when the first ovulatio occurs early in the year, large follicles may be present for a month or more before ovulation occurs (5). Apparently a large follicle that undergoes atresia during ‘he transitional period can be confused occasionally with an ovulation when dig 't: examination is used. The number of such errors can be minimized by us diagnostic ultrasound. — 10.4 Predicting Impending Ovulation Predicting the day of ovulation would have considerable use in coordinating the time of breeding with the expected time of ovulation. Palpation is an aid in ‘his regard. In one study (4,5), the preovulatory follicle became soft 12 hours betore ovulation in 40% of the mares. However, it was concluded that this approach was not reliable in individual mares. The availability of transrectal, diagno ultrasonography has raised the question whether this technology can be use predict impending ovulation. Hopes were raised by a report that there wa: increase in the echogenic patterns of the follicular fluid near the periphery o! follicle one or two days before ovulation (6). In women, ultrasonography has | used for this purpose (7,8,9); the preovulatory follicle was described as havi! flattened appearance a few days before ovulation. Because of the interest in ultrasonography as an aid in predicting impenc:ng ovulation, an experiment was designed to determine whether ultrasonica ly detectable changes occurred consistently in the preovulatory follicle (10). Forty riding-type horse mares were used, yielding a total of 79 preovulatory periods v ith one ovulation per period. The preovulatory follicle was defined as the follicle ‘hat became and remained the largest follicle by at least 5 mm in the ovary from which ovulation later occurred. The gain, brightness, and contrast controls were set {0 4 predetermined standard. Shape was described as being more nearly spherical or nonspherical. The mean diameter of two dimensions was used to estimate diameter of nonspherical follicles. Thickness of the follicular wall was scored from 0 to 6 (O= wall not discernible; 6 = thickness equivalent to approximately 3 mm). Gray-scale values for the wall and fluid were made by reference to the standardized gray-scale bar (Section 6.3B). Assessments of gray-scale values for the wall were made at the Jateral borders of the follicles because of enhancement artifacts beneath the follicles (Section 5.4). The following discussion was developed primarily from the results of this experiment. on OS Oaae Follicles 143 Sonograms of preovulatory follicles. A,B,C) Follicles which maintained a /nerical shape prior to ovulation. A) 27 mm follicle. The thickness score of the icular wall is 2. B) 35 mm follicle. Note the increased thickness of the follicular I. C) 42 mm follicle. The thickness score is 6 and the shape is approximately erical. D,E,F) Follicles which changed shape from spherical to nonspherical or to ovulation. D) 38 mm follicle. The thickness score is 4. The nonspherical pe may be partially due to the presence of small follicles visible at 7 and 8 o'clock. 36 mm follicle. The follicular wall is pronounced. Note the nipple-like ‘ojection of the follicle at approximately 2 o'clock. F) The approximate spherical meter of the follicle was calculated to be 37 mm. The thickness score is 6. Note ‘ pronounced neck-like process at 10 o'clock. 144 Chapter 10 15% GROWTH RATE: 3mm/ DAY A et ) “ew 85% DIAMETER (mm) 27 3a 39 45 =a -5 -3 -1 DAY Diagrammatic presentation of the growth rate and changes in shape of the preovulatory follicle. Day -1 is the day before ovulation. Note that th average growth rate was 3 mm/day for the seven days preceding ovulation. Th preovulatory follicle became the largest follicle six or more days prior to ovulatio in 82% of the preovulatory periods (mean, Day -7). This is consistent with previou findings utilizing ovarian palpation and ultrasonography (Section 6.4). The mea: diameter of the preovulatory follicle was 29.4 mm on Day -7 and 45.2 mm or Day -1. The follicle increased in diameter linearly over the seven days at an averag¢ rate of 2.7 mm per day. Ninety-six percent of the follicles were 36 to 50 mm in diameter on Day -1. No preovulatory follicles were smaller than 35 mm or larger than 58 mm on Day -1. The growth rate and mean diameter of the preovulatory follicle on Day -1 was consistent with reported growth rates and Day -1 diameters determined by ovarian palpation and ultrasonography (1,2,5). However, the reported decrease in diameter between Days -2 and -1 (1) was not found in the present study. These divergent results may be due to the use of a two-way measuring technique for nonspherical follicles in the present study. Follicles 145 Eighty-five percent of the preovulatory follicles exhibited a pronounced change n shape from approximately spherical to nonspherical (pear-shaped or conical) at ome time during the preovulatory period. This is consistent with the observation nat preovulatory follicles in women become flattened prior to ovulation (7). There vaS a progressive increase in the number of preovulatory follicles exhibiting a nange in shape as the interval from Day -7 to ovulation decreased. Six of the reovulatory follicles exhibited a pronounced neck-like process, ostensibly preovulatory period. /alues representing the width of the follicular wall increased from Day -6 to Day _ The mean scores did not increase from Day -2 to Day -1, although the mean oophorus was reported for preovulatory follicles in women (8,9), it was not detected in this study in mares. 148 Chapter 10 In summary, the hypothesis that the preovulatory follicle undergoes a change in shape prior to ovulation was supported, although changes occurred at various times over the entire preovulatory period. The results also supported the hypothesis that images corresponding to the follicular wall increase in thickness as the follicle gro and the interval to ovulation decreases. Within the constraints imposed b; instruments, the results did not support the hypotheses that changes in gray-sc value of the follicular wall or the echogenicity of the follicular fluid were predict of impending ovulation. The combination of diameter, shape changes, and thickn of the wall appeared to be valuable for assessing the status of the preovulat follicle, although no ultrasonically visible reliable predictor of impending ovu! was found. In retrospect, the diameter of the follicle appeared to be as useful fo predicting impending ovulation as any of the other criteria. Set A OO ODO OO DH = 10.5 Ovulation Images of ovulation sites. Images were taken on the day the large follicle was missing (Day 0), using daily examinations. A) Collapsed wall of ovulatory follicle, probably beginning to luteinize. Note the intense echogenicity and the apparent area of contact between walls (arrows). B) Ovulation site containing fluid. Note the hyperechogenicity (arrows) of the wall. C) Ovulation site delineated by arrows. The occurrence of ovulation is readily detected by ultrasound by the disappearance of a large follicle that had been present at a previous examination. In addition, the newly forming corpus luteum was visible on Day 0 for all ovulations in one study (17); however, the operator was sometimes aware that a large follicle was present Follicles 149 previous day. Intense echogenicity of the ovulation site was seen on Day 0 in > of 32 ovulations (17). Identification of an ovulation site in the blind has not been attempted except for a limited study (17) in which the newly forming corpus um was identified in three of three mares on Day 0. In other experiments, ovulation occurred in three mares while the ovaries were being examined. In all », the ovulation site became hyperechogenic within five minutes after ovulation, lar to the images shown above. One mare was observed continuously during ovulation. A pear-shaped, 37 mm follicle collapsed over a period of a few seconds, ing a small, irregular, nonechogenic central area (10 mm) surrounded by a ierately echogenic area. Within four minutes the central portion was reduced to 2 om and the remaining area of the site became hyperechogenic (bright white). “he dramatic ultrasonic changes following rupture and collapse of the follicle can be used to detect or confirm ovulation. For example, the prolonged period of development of large follicles that precedes the first ovulation of the year is a particularly challenging theriogenology problem (Section 10.3C). During this time, a clinician who is limited to the tactile sense may have doubts as to whether an ovulation has occurred, whereas visual inspection by ultrasonography may provide definitive clarification. Although ovulation detection by either the tactile approach (rectal palpation) or the visual approach (ultrasonography) has a good degree of reliability, there will always be a number of mares that will challenge the decision- making process. Sometimes the ovulation site may begin filling with blood on Day 0 (Section 11.3), and there may be difficulty in distinguishing by palpation the new corpus hemorrhagicum from a preovulatory follicle. Ultrasound examination, however, will readily detect the echogenic fibrinous bands in the nonechogenic blood clot (Section 11.3). In those instances where the follicle has collapsed but the wall has not become echogenic, the ultrasonographer may sometimes have doubts whether there has been a very recent ovulation. Such cases are readily confirmed by diz'tal palpation of a large, soft area. Some veterinarians may elect to use both dicenostic methods routinely, whereas others may use one as the sole or primary approach and the other as a secondary or supplementary method. It seems reasonable to urge, however, that regardless which technique is selected as the primary approach, an effort should be made to develop or retain skills in the other. 150 Chapter 10 10.6 Selecting the Optimal Breeding Time 35 Exper. 1 Exper. 2 eee oe No.: 103 78 30 Mean: 43mm 45mm SD: t6mm +5mm 25 35-70mm 35-56mn 20 15 10 Number or ovulatory periods 35-39 45-49 55-59 65-69 >7 Diameter (mm) of follicle on day before ovulation Diameter of follicle on the day before ovulation. Follicles were measured by ultrasound daily in horses (primarily Quarter Horses) in two separate experiments within the ovulatory season. Only ovulatory periods with a single ovulation were included in the data. For nonspherical follicles, the average o! ‘wo dimensions was used. Over the two experiments, the range of diameters on Day -1 was 35 to 70 mm. It is especially noteworthy that in a total of 181 single ovulations none of the follicles ovulated before reaching 35 mm. In this regard, however, the preovulatory follicle is smaller on Day -1 in double ovulators. In the series described in Section 16.1, the mean diameter of double follicles on Day -1 was 39.7 mm; 11 of 24 ovulated before reaching 35 mm and nine of the 11 were 25 to 30 mm on Day -1. Follicles 151 ee EB Sa gg eee aati Breeding No. days from No. mares missed when the number of days initiated indicated size ——_——between ovarian examinations was; vhen largest to ovulation One Two Three Four \licle reaches: (mean tSEM) day days days days ee LL ge RR ee ere eee 225 mm hdd ae Atel 0 0 0 1(1%) >30 mm 38 £02 0 0 0 5(6%) >35 mm 4.1+0.2 0 2(3%) 16(21%) 28(35%) >40 mm 22 £02 8(10%) 19(24%) 36(46%) 49(63%) Follicular data relevant to the development of a program for uitiation of breeding. Data are from the same mares as for Experiment 2 in the vove illustration. The first ovulation of the year is not considered. The table shows ne number of days to ovulation and the number of mares that would have been nissed (mare ovulates before being bred), if the breeding program had been initiated t various times based on diameter of largest follicle. For example, if the follicle was neasured every day and breeding had been delayed until the largest follicle reached ‘S mm, none of the mares would have been missed and the mean interval to ovulation vould have been minimized to four days. If the measurements were made every ree days, breeding would have had to begin when the largest follicle reached 30 mim in order to avoid missing any mares. In this way, the table can be used as a suideline to develop a program that takes into account the preferred interval between examinations, the preferred interval between breedings, and the number of mares ‘ie operator is prepared to miss. It is emphasized that these data were obtained after ‘he first ovulation of the year. During the transitory period before the first lation of the year, large follicles can be present for a greatly prolonged time --- sometimes for as long as a month. In the data shown in Section 10.3C, the largest ollicle reached 35 mm nine days (average) before ovulation. eg Factors determining the optimal time to breed are the life span of the sperm and yum and the requirements for sperm capacitation (5). Breeding after ovulation ecreases the possibility of a successful pregnancy. Some field trials indicate that perm may remain viable in the mare for several days --- perhaps five or six days or ven longer. However, the life span may be much shorter for certain stallions. It is. enerally agreed that it is far better to breed before ovulation than after ovulation, \' the entire question of the time and frequency of breeding during estrus requires ao Gg oO (a eA 152 Chapter 10 further study. Apparently the goal on most breeding farms is to breed the mar before ovulation occurs, yet within two or three days before ovulation. Thus, it not uncommon to breed some mares every two or three days. Currently, there ar no reliable ovarian criteria for estimating when ovulation will occur that are m« accurate than judgments based on size, alone (Section 10.4). However, size can measured more accurately by ultrasound than by palpation, thereby minimizing © of the most important variables influencing a decision. 10.7 Superovulation Ovaries from a mare that was stimulated to superovulate. The mare was treated with a pituitary extract (18). A) Stimulated ovary with two preovulatory follicles in this view. B,C) Two blood-filled luteal glands (corpora hemorrhagica) in one ovary (B) and two nearly solid luteal glands in the opposite ovary (C) on Day 2. The peripheries of the individual corpora lutea in image C are not discernible on the photograph, but were during real-time imaging. Ultrasonography is useful for monitoring the results of stimulatory treatments (18), especially when more than one large follicle is present in an ovary. Follicles 153 (EFERENCES . Ginther, O. J. and R. A. Pierson. 1984. Ultrasonic anatomy of equine ovaries. Theriogenology 21:471-483. . Palmer, E. and M. A. Driancourt. 1980. Use of ultrasound echography in equine gynecology. Theriogenology 13:203-216. 3. Pierson, R. A. and O. J. Ginther. 1986. Folliculogenesis during the estrous cycle in the mare. Theriogenology (submitted). '. Parker, W.G. 1971. Sequential changes of the ovulating follicle in the estrous mare as determined by rectal palpation. Proc. Ann. Conf. Vet., College Vet. Med. and Bio-Med. Sci., Colorado State Univ., Fort Collins, CO, cited in (5): 149-150. . Ginther, O. J. 1979. Reproductive Biology of the Mare: Basic and Applied Aspects. Garfoot Rd., Equiservices, Cross Plains, WI. . Squires, E. L., J. L. Voss, M. D. Villahoz, and R. K. Shideler. 1983. Use of ultrasound in broodmare reproduction. Proc. Am. Assoc. Eq. Pract., Las Vegas, NE: 27-43. . Crespigny, L. C., C. O. O'Herlihy, and H. P. Robinson. 1981. Ultrasonic observation of the mechanism of human ovulation. Am. J. Obstet. Gynecol. 139:636-639. ». Hill, L. M., R. Breckle, and C. B. Coulam. 1982. Assessment of human follicular development by ultrasound. Mayo Clin. Proc. 57:176-180. ’, Hackeloer, B. J., R. Fleming, H. P. Robinson, A. H. Adam, and J. R. T. Coutts. 1979. Correlation of ultrasonic and endocrinologic assessment of human follicular development. Am. J. Obstet. Gynecol. 135:122-128. 10. Pierson, R. A. and O. J. Ginther. 1985. Ultrasonic evaluation of the preovulatory follicle in the mare. Theriogenology 24:259-268. 154 Chapter 10 ee le 13. 14. 15, 16. 17. 18. Prickett, M. E. 1966. Pathology of the Equine ovary. Proc. Am. Assoc. Fq Pract., Los Angeles, CA: 145-153. Ginther, O. J. and R. A. Pierson. 1983. Ultrasonic evaluation reproductive tract of the mare: principles, equipment and techniques. J Equine Vet. Sci. 3:195-201. Zweibel, W. J. (Ed.). 1983. Introduction to Vascular Ultrasound. Grune ajc Stratton, Orlando, FL. Baum, G. 1975. Fundamentals of Medical Ultrasonography. G. P. Putman anc Sons, New York, NY. Goss, S. A., R. L. Johnston, and F. Dunn. 1978. Comprehensive compilation © empirical ultrasonic properties of mammalian tissues. J. Acoustic Soc. A 64:423. Pierson, R. A. and O. J. Ginther. 1985. Ultrasonic evaluation of the cor; luteum of the mare. Theriogenology 23:795-806. Ginther, O. J. and Pierson, R. A. Ultrasonic evaluation of the reproductive tract of the mare: ovaries. 1984. J. Equine Vet. Sci. 4:1 1-16. Woods, G. L., and O. J. Ginther. 1983. Induction of multiple ovulations dur the ovulatory season in mares. Theriogenology 20:347-355. y > ult not fol! esti Chapter 11 LUTEAL GLANDS 1e corpus luteum produces progesterone which terminates estrous behavior and ains the mare in a nonreceptive state. Progesterone also prepares the tubular alia for the potential reception of an embryo. If the mare does not become gnant, the corpus luteum regresses. The mare then returns to estrus and is orded another opportunity to become pregnant. If the mare becomes pregnant, corpus luteum is maintained and continues to produce progesterone, which is ential to the continuation of pregnancy. Because of the pivotal role of the corpus ‘um, its detection provides much information to the diagnostician. The presence stage of the luteal gland cannot be readily ascertained by rectal palpation, except ing the first few days after ovulation. Progesterone assays provide the most — ul information for assessing the corpus luteum, but are not convenient for nediate evaluation. One of the major uses of ultrasonography, therefore, involves immediate detection and evaluation of the luteal gland. ‘his chapter is devoted to this vital gland with discussion of its detectability and asonic anatomy. The results of two studies involving the detectability of the al gland are reviewed. One study was done blind and the other without reference previous day's records. The ultrasonic anatomy of the gland is discussed in with reference to studies done in situ and in a water bath using excised ovaries. types of luteal ultrasonic morphologies are described. In one type, a central y of the gland fills with blood which becomes organized, but remains onically detectable throughout luteal life. In the other, a central blood clot is etectable at any time. Finally, some applications are discussed, including the ving: 1) determining whether a mare has entered the ovulatory season, 2) ating the day of the estrous cycle, and 3) detecting luteal persistence. ——————E<——— i#-- | 156 Chapter 11 11.1 Detectability a ee Number of mares in which there was: Uncertainty or Reproductive status Agreement not identified Disagreement i Diestrus or pregnancy Day 0 3 0 0 Days 1-6 iz 1 0 Days 7-10 12 3 0 Days 11-14 8 nit Kae a. Total 35 (88%) 5 (12%) 0 (0%) Returned to estrus (preovulatory) 0 12 (100%) 0 Pregnant Days 18 - 27 9 2 1 Days 28 - 44 9 ey ge Total 18 (64%) 8 (29%) 2 (7%) ee Results of a blind study on the detectability of the corpus luteum The location of the corpus luteum (left or right ovary) was established b: operator by daily palpation per rectum to determine the side of ovulation. ultrasound examinations were done by another operator on one day using a5.0 } transducer. The ultrasonographer recorded the location of the corpus lute! indicated that one was not found or that there was uncertainty about the identific and therefore location. The extent of agreement on location of the corpus luteun evaluated by comparing the palpation records and the ultrasound records. The e of agreement was classified as: 1) agreement, 2) uncertainty or not identified 3) disagreement. The location of the corpus luteum as determinec K me he Hz or ion was tent and by ultrasonography agreed with the previous determination of side of ovulation by palpation in 88% of 40 mares on Days 0 to 14. In the remaining 12%, the ultrasonographer recorded the location as uncertain. There were no disagreements. In all of 12 mares that were in estrus (failed to conceive), the location of the corpus luteum was recorded as uncertain. ~] Ce vai dai y obser’ ult desc1 Th fo inte was was disas ove corr scan and trans days; 100% o . £ 75% 7 = a © teed 2 al S 50% | ° O o 2 8 25% = ® ; 2 Luteal Glands 157 0% OV 2 4 ©. & 0 2 14 16 te OV = 2 Days ‘centage of mares with an ultrasonically visible luteal gland for is days of the interovulatory interval (2). Forty mares were examined vith a 5.0 MHz transducer for a total of 55 interovulatory intervals. Each day's rations were made without knowledge of the previous day's results. The asonographer was more experienced than the operator of the blind study, bed above. The luteal gland was identified for a mean of 17 days (SEM 0.6). mean interovulatory interval was 21.7 days (40.5). The luteal gland was visible | mares from the day of ovulation until at least halfway through the vulatory interval. At the time of the subsequent ovulation, the luteal structure jentified in five of the 55 intervals, and in three of the five the luteal structure ‘dentifiable until one day after the second ovulation. There were no eements between days on the side of location of the luteal gland in any mare he first 10 days, postovulation. These studies demonstrate that a functional ; luteum is usually detectable with a 5.0 MHz transducer and high-quality er. Itis emphasized that identification of the corpus luteum throughout diestrus arly pregnancy requires at least a 5.0 MHz transducer. With a 3.5 MHz ducer, the corpus luteum was identified for a mean of only 5.7 days (range, 4-8 n=19), 158 Chapter 11 11.2 Ultrasonic Luteal Glands 159 Gross appearance and ultrasonograms of excised ovaries (1). Images ere recorded in a water bath (upper case letter) and the ovaries were then sectioned )wer case letter) in a plane approximating that of the image. aA) Arrows mark the ter limits of the image of the corpus luteum. The corpus luteum apparently is ery young (perhaps Day 0 or 1) and is seen as a bright white (hyperechogenic) area the image. bB) Arrows mark the outer limits of a corpus hemorrhagicum srhaps Day 1 or 2). The center of the corpus hemorrhagicum is filled with blood ich is nonechogenic (black) except for scattered echogenic (white) spots. cC) A pus hemorrhagicum that is further developed (perhaps Day 2 or 3), as indicated the thicker outer wall of luteal tissue. The arrow points to a wall of fibrin-like iterial separating the blood clot into a dark, nonechogenic area (above division), vhich probably contains a high proportion of red blood cells, and a lighter area onsisting apparently of plasma or entrapped follicular fluid with a fibrin-like ietwork. The fibrin-like network in the lighter area is characterized by a white, yperechogenic network on the ultrasound image. The luteinized wall (lw) is cb schogenic. A fibrin-filled corpus albicans (ca) is discernible as a small white ircumscribed area. f = follicle. dD) Corpus luteum sectioned in a plane that avolves the neck-like process leading to the ovulation fossa. f = follicle. eE) ature corpus luteum delineated by arrows on the ultrasound image. There is no ntral blood-filled cavity. Arrow on both e and E points to apposed walls of two cjacent follicles. fF) Mature corpus luteum, possibly beginning to regress. Arrow n f and F points to fibrinous center which is highly echogenic. f = apparent ‘covulatory, vascularized follicle. The ultrasonic anatomy of the corpus luteum affected by the presence and nature of a central cavity. Blood was nonechogenic, sulting in the black image in the central area of the corpus hemorrhagicum. vrin-like material in the blood clot was echogenic so that the images of some of the ora lutea had a central white network. The ultrasonic properties of mature »ora lutea were similar to those of the stroma. However, even in the absence of a central cavity, the corpus luteum could be distinguished from stroma by its defined _ der. CM bel Examples of images of corpora lutea taken during transrec examination (3). A) A corpus hemorrhagicum at Day 1, consisting approximately 55% luteal tissue (white) and 45% blood (black). B) Same cor hemorrhagicum at Day 4, containing a nonechogenic, organized (white sp: central area. C) A mature corpus luteum that does not contain a central cavity. structure is ultrasonically homogeneous. Note the neck-like process of the co! luteum extending from the luteal body at 8 o'clock. The process is surroundec several small follicles. In a few instances (incidence unrecorded) distinct small mm or less) circumscribed bodies have been seen. In the study of excised ova! similar structures were determined to be remnants of corpora albicantia w! imparted a hyperechogenic image. This is consistent with the greater reflectioi ultrasound waves by tissue of greater density. It was not possible to follow ti highly echogenic areas on a daily basis. In many cases, it appeared as though areas were identifiable because of the location of contiguous small (2-5 mm) follic It seems feasible that improvements in imaging technology will enable an operato follow the process of luteal regression further into a subsequent interovulaic interval. ty * oe a = 2 | 4 s { Luteal Glands 161 1.3. A Study of Two Types of Ultrasonic Morphologies soy [- 7 OO UNIFORM LUTEAL TISSUE CENTRAL BLOOD CLOT -1 0 2 4 8 12 16 DAY ‘ntroductory diagrammatic summary of the development of two types luteal gland morphologies(2). The luteal structure was evaluated daily ring 55 estrous cycles. Approximately 50% of the glands developed a central \d clot and 50% did not. The clots remained throughout the apparent functional 'e of the gland, but became increasingly organized and reduced in volume. It opears that the corpus hemorrhagicum is not functionally important, because it >veloped in only one-half of the luteal glands. In addition, the development of a corpus hemorrhagicum did not alter the length of time the luteal structure was ultrasonically visible, the length of the interovulatory interval, or the volume of \uteinized tissue. Furthermore, in approximately one-half of the mares with sequential ovulations, a corpus hemorrhagicum formed after one ovulation but not cr the other ovulation. Similarly, in approximately one-half of the mares that uble ovulated, one ovulation was followed by the formation of a corpus norrhagicum and the other was not. These observations indicate that the ‘ation of a corpus hemorrhagicum occurs by chance and is not peculiar to certain mares Or to certain ovulatory periods. Perhaps the extent of rupture of vascular components of the follicular wall during ovulation determines whether or not a corpus hemorrhagicum will form. Our interpretation of the results, as described in the following pages, is that corpus hemorrhagicum formation is an incidental phenomenon in luteogenesis. “sy ,_ — = » oO = Fa) 162 Chapter 11 i CM ssc a GT ie eo Ultrasound images showing various stages of two types of luteal morphologies (2). Luteal glands that formed after 48.5% of 95 ovulations were uniformly echogenic over 90 to 100 percent of the gland's image throughout the period of detectability (G,H,I). The remaining luteal structures (51.5%) exhibited a centrally located nonechogenic area (D,E,F), which was attributable to a blood clot H Luteal Glands 163 EN oO. enmrmwosvwm=deweryHypewmtyewrF=pwrewew Bev YP. sorpus hemorrhagicum). The borders of luteal structures are indicated by arrows. \ site of ovulation (Day 0). The image is highly echogenic. Note the neck-like ess extending from the structure at approximately 2 o'clock. B) Luteal glands , 3) in the ovary of a mare which synchronously double ovulated. The luteal | on the top left of the image has a large (approximately 80% of the structure) ally nonechogenic area. The luteal gland on the bottom right is ultrasonically iogeneous. C) A discrete, hyperechogenic area, identified as a corpus albicans. the contiguous small (2-5 mm) follicles. D-F) Centrally nonechogenic luteal ids on various days postovulation. D) Day 3. The echogenic portion comprises ‘oximately 20% of the gland. E) Day 8. The echogenic area represents proximately 40% of the gland and has a gray-scale value of zone 3. F) Day 13. “he echogenic portion of the gland represents approximately 70% of the area. The y-scale value is zone 4. G-I) Uniformly echogenic luteal glands on various days stovulation. G) Day 3. The echogenic area comprises the entire luteal structure. te the neck-like process extending from the body at approximately 10 o'clock. H) 5: y 7. Mature luteal gland with a homogeneous appearance. The gray-scale value of | a ne echogenic area is zone 3. I) Day 15. A regressing luteal structure of uniformly chogenicity and zone 5 gray-scale value. Approximately 50% of the luteal glands were uniformly echogenic throughout the period during which they were identifiable. The morphology of the remaining luteal glands involved a prominent central nonechogenic area which was attributed to blood on the basis of a previous study (1). Fibrin-like material within the blood clot was echogenic, so images of corpora lutea with this morphology often had an echogenic network interspersed within the nonechogenic area. The echogenic portion of the luteal gland in both luteal morphologies was differentiated from ovarian stroma by assessing the ultrasonic properties of the tissues. Luteal tissue was distinguished from stroma by a distinct border formed in part by the difference in acoustic impedance of the tissues. The ultrasonic texture of the luteal gland was characterized by an echo pattern indicative of loosely organized, well vascularized tissue. Ovarian stroma yielded generally brighter echoes in a pattern representative of denser tissue. There is agreement with observations from this study and the previous experiments (1) and with the known properties of ultrasound in tissue. Glands of both types often showed a distinct mushroom or gourd shape. ogni = Swee Ste st © = <= St S= t= Bue WEtusis bt EWSisE Tas 9 ee3 8 164 Chapter 11 e-==Luteal glands with uniform echogenicity 100% : peobemtnnn taebe, | ote, | potemdendet! “abandons 2 90% 3 | ae ” a 80% /\ 5 ie isd ‘S 70% o Pre a o 60% /| ‘ a 3 50% /\ | ° ~ = I Lu 40% e= Luteal glands with central nonechogenic area a OV 2 4 6 8 10 12 14 16 Days Percentage of tissue that was echogenic (luteinized) for the two types of luteal glands (2). Luteal glands classified as uniformly echogenic (n = 26) were echogenic over 95 to 100 percent of their area throughout the time thai the gland was ultrasonically visible. In glands classified as centrally nonechogenic (1 = 29), the nonechogenic area was first visible on Day 0 (28%), Day 1 (62%), Day 2 (6%), or Day 3 (4%). On the day of ovulation, centrally nonechogenic glands were echogenic over 75 to 100 percent of the image of the gland. Thus, in most mares, before the central cavity appeared, the glands showed an initially high proportion of echogenic tissue which apparently represented the collapsed, luteinizing walls of the recently ovulated follicle. The nonechogenic area increased over Days 1 to 3, due to increasing enlargement of the central area with the nonechogenic blood ciot. Thus, the proportion of luteinized tissue was lowest on Day 3, when the luteal glands were echogenic over an average of 45 percent of their area. The proportion of echogenic area (luteinized tissue) increased during the remaining period of ultrasonic visibility; that is, the central blood clot decreased, proportionally. The glands appeared to retain some of the clot throughout the period of detectability. The proportion of echogenic area reached a mean value of 95% before becoming unidentifiable. Oo &a &t 3 gla Luteal Glands 165 5.0 | . e==: Luteal glands with uniform echogenicity sd N, “ x | |» \S s 4.0 | / ty ot / at 3.5 fs | i ted ? rs A Ee i a J. he te | oat 3.0 t e— Luteal glands with central nonechogenic area OV... 2 a 6 S& 10... 12 {4 16 Days Mean values for gray-scale score for the luteinized portion of the two es of gland. Settings for gain, contrast, and brightness were standardized each Both gland types (excluding blood clots) were highly echogenic on the day of lation. The echogenicity decreased over the first six days, remained at the mum level for several days, and then increased over Days 12 to 16. The gray- value appeared to give an indication of luteal hemodynamics. The very bright, serechogenic echoes on Day 0 may have resulted from apposition of the collapsed : with relatively low vascular perfusion. This result is consistent with the report ‘very bright" and "bright" luteal structures one to two days and three to five days *r ovulation, respectively (4), and the detectability of the luteal gland for a mean ly 5.7 days when a 3.5 MHz transducer was used (2). The image brightness creased to a mean value just beyond zone 3, which was retained ostensibly for the od of highest vascular perfusion (the period of maximum progesterone uction). There was a tendency for the image of the luteal gland to become ighter again, as regression progressed. This was indicative of decreased blood )w, increased tissue density, and fibrin infiltration. From Day 12 until ultrasonic appearance, the gray-scale values reached zone 6 or 7 (very bright) in 33% of 55 nds, zone 5 in 15%, zone 4 in 16%, and no increase in brightness in 36%. 166 Chapter 11 11.4 Corpora Lutea of Pregnancy Images of ovaries from pregnant mares. A) The primary corpus luteum at Day 40. B) Numerous large follicles at Day 60. C) An ovary from a mare ai Day 170, showing several corpora lutea (each arrow points to a corpus luteum). Large follicles and corpora hemorrhagica are common on Days 40 to 60. The combination of corpora hemorrhagica, corpora lutea, and large follicles complicates attemp‘s to discern and count the various structures. The primary and secondary corpora ‘utea cannot be differentiated on gross inspection of a sectioned ovary (5), and it is not expected that they can be differentiated ultrasonically. Secondary corpora lutea vary widely in morphology because they may form from an ovulation site or by luteinization of an intact follicle (Section 9.1C). As a result, they may be solid or contain a central cavity with a blood clot or a clear fluid. The mean number of secondary corpora lutea in one study (5) was 2.8 at Day 70 and 10.2 at Day 140. Because of the massive luteinization, the number of corpora lutea sometimes cannot be determined by ultrasound. However, when needed, as for research purposes, the degree of luteinization can be scored. Luteal Glands 167 |.5 Applications of Luteal Detection and Evaluation SA Determining if mare has entered the ovulatory season Diameter (mm) of largest follicle on day the corpus luteum disappeared End point $20 21-24 =25-29 30-34 35-39 40-44 45-49 >50 No. mares 0 7 8 18 10 9 6 1 Percent per group 0% 12% 14% 30% 17% 15% 10% 2% Percent accumulative 0% 12% 26% 56% 73% 80% 98% 100% No. days to ovulation Mean -- 8 7 5 3 a 3 Zz Range -- 7-10 6-10 1-9 1-7 1-7 1-5 -- Size of largest follicle on the day of ultrasonic disappearance of the uteal structure during the estrous cycle. These data may be used to help stinguish a cycling mare with a regressed undetectable corpus luteum from an 1ovulatory-season mare. If the largest follicle is 20 mm or less and a corpus luteum ‘s not found, the mare can be considered anovulatory; in all of 59 cycling mares, the \argest follicle was greater than 20 mm on the day the luteal structure disappeared. With increasing size of the largest follicle, this approach becomes increasingly less able for distinguishing between the two reproductive states. As expected, the ber of days from disappearance of the luteal structure to ovulation decreased as > diameter of the largest follicle increased. If ovulation does not occur within the umber of days shown in the table, the mare is likely in an anovulatory state. nother approach, is to re-examine the mare in 10 days. If the mare is cycling, it is ions and increases in size at an expected rate. Recording the location and size of each time the mare is examined before pregnancy detection is scheduled will ase the probability of later mistaking a cyst for an embryonic vesicle. ¢ Ultrasonic Anatomy of the Cervix ‘trasonograms of the cervix. A) Day 12 of pregnancy. The cervix is the rechogenic area on the right (arrows), and the uterine body is to the left. B) \4 of pregnancy. The embryonic vesicle is in the uterine body lying against the <. The embryonic vesicle is mobile at this stage and it is not unusual to find it in area. C) Estrus. The cervix (arrows) is indistinct and difficult to differentiate. hypoechogenic area beneath the cervix is the bladder. The echogenicity of the x is similar during diestrus and pregnancy and is attributable to tissue density. cen the cervix becomes flaccid during estrus, it loses its hyperechogenicity and is fficult to differentiate from surrounding tissues. The changing echo texture of the cervix reflects the prevailing levels of estrogens and progestins. 178 Chapter 12 12.4 Ultrasonic Anatomy of the Uterus Oe Bee: Se a PREGNANCY: ey eT) Ultrasonograms of the uterus during the estrous cycle and pregnancy. The images are of the uterine body (A,B,D,E,G) and uterine horns (CBE). Fos orientation, the cranial aspect of the tissues is to the left of the images. Each row of images is from the given reproductive state. Estrus. The echo texture is characterized by alternating and intertwining areas of hyperechogenicity and Ny Posed yey The hyperechogenic areas are attributable to tissue-dense central portions of the endometrial folds. The hypoechogenic areas are attributable to edematous outer portions of the folds. Image B represents an advanced stage of Uterus 179 ‘rual edema, and was taken one day later than image A. Occasionally small ections of free fluid are seen in the lumen; these probably result from pockets of us fluid, but also can represent collections of an exudate (Section 12.8B). The isonic anatomy indicates that there is considerable edematous development of the ymetrial folds during estrus even though the uterine walls, as a whole, are usually more flaccid during estrus than during diestrus (Section 14.5A). The uterine folds alpable when the flaccid walls are compressed between the thumb and finger, ne hand is moved along the horns. There is a positive relationship between the ‘pated size of the folds and their prominence on the image. Long echogenic lines resenting the lumen, as described below for diestrus, usually are not discernible ing estrus, although occasionally a short line will be seen. Diestrus. Individual endometrial folds are less distinct or not discernible, due to the loss of the folding effect characteristic of estrus. The echo texture is therefore homogeneous. In image D, the upper and lower limits of the uterine body are indicated by arrows. The uterine lumen often is identifiable by a hyperechogenic or ite line when the uterus is viewed longitudinally, as shown in image D. Because of the orientation of the uterus, the white line is most prominent in the uterine body. ng segments of the white line sometimes can be seen (D), but usually the line is broken at irregular intervals. Only occasionally is the reflection seen in the horns as a short white line, because the horns are routinely examined in transverse planes. The ultrasonic origin of the echogenic line representing the lumen probably involves specular echoes and is discussed in Section 12.5. Image E shows the echo texture of the caudal portion of the uterine body in contrast to the hyperechogenic cervix to the right of the uterus. Note the short length (5 mm) of specular echoes representing the uterine lumen (arrow) at the junction with the cervix. “regnancy. The ultrasonic characteristics of the uterus during early pregnancy indistinguishable from those of the corresponding days of diestrus. Note the interrupted white line between the two embryonic vesicles (Days 12 and 13) due to the intraluminal location of the vesicles. After the embryonic vesicle becomes fixed (Day 15 or 16), the echo texture of the endometrium begins to change, so that by Days 18 to 20 the endometrial folds become prominent. Compare image H (Day 14) with image I (Day 18). Note that the uterine wall in image H is similar to a diestrous uterine wall (F), but in image I the endometrial folds are more distinct (arrows). The folds are not as prominent as during estrus, however. Perhaps the prominence of the folds at this time is due to the estrogen exposure either from the conceptus or Ovaries. The development of the folds occurs throughout the uterus. This phenomenon is under study to document its occurrence and its characteristics. oe ep earen 9 pm ree >: = rm ar @ CD 180 Chapter 12 ANESTUS: Ultrasonograms of uterus during anestrus. During deep anesir (anovulatory season) the endometrial folds are not discernible and the echo text similar to that of the uterus during diestrus or early pregnancy. However, the c sectional views of the horns may be irregular or flat. In comparison, Cross sec of the horns during the other reproductive states are round. The flat or irre images are caused by the relatively flaccid walls which conform to the irregula of the adjacent viscera. In image A, the horn is lying on the pole of an ovar) image B, the horn is distorted by an adjacent segment of intestine. The periphe the uterine horn is indicated by arrows. x The ultrasonic anatomy of the uterus is profoundly influenced b} reproductive state (estrus, diestrus, pregnancy, anestrus). These changes c attributed to the prevailing levels of ovarian steroids and possibly may be influe also by the products of the conceptus. The echo texture of the uterus and the she the cross-sectional views can be used to help determine reproductive status. Uterus 181 Layers of uterine wall. Images A (same as image F in previous section) and B e taken during diestrus, and image C during estrus. The periphery of each cross A pposition of layers is indicated by arrows pointing outward. The layering effect be due to differences in echo texture between endometrium and myometrium or e presence of a vascular area between the circular and longitudinal muscles of nyometrium. Critical work on the echo texture of various layers of the wall has een done. Often such layers are indistinct or not discernible. ages of uterus on Day 3 (A) and Day 7 (B) after parturition (3). mages have a black, mottled appearance resulting from the collection of artum fluids. The mare ovulated and conceived five days after view B was Evaluation of postpartum involution is a promising potential use of ound that has not yet been investigated. ta ¥ Ser ezt “Es tos 3 ESS eee 182 Chapter 12 12.5 Origin of Specular Echoes Ultrasound image Uterus Specular echoes Position of transducer Specula reflectio: Endomet ma ~wunwownw ow ~~ Myometri Surface view Side view (sides reflected) Drawings of various views of uterine folds. Note the longiti arrangement of the uterine folds in the view of the internal surface. The fol shown photographically in Sections 13.6 and 13.7. There are seven major ! During the estrous cycle and early pregnancy, the folds are relatively prom whereas during the anovulatory season the folds are not. Specular echoes are c: by the reflection of ultrasound pulses from a smooth surface that is wider tha pulse and is parallel to the face of the transducer (Section 5.1). Because conditions can be met by a portion of the surfaces of endometrial folds, the bi ial are ds. nt, sed the 1eSe ight white lines seen in images of longitudinal views of the uterus are attributable to specular echoes. As shown in the diagram, specular echoes appear on the image for those portions of the folds that are parallel to the transducer face. Sometimes long continuous lines may be seen; at other times there may be short, interrupted segments or no lines. During deep anestrus, the folds may not be prominent enough to act as specular reflectors. The reason for the decrease in specular echoes during estrus is not known. Perhaps the engorged folds are less likely to present a long, smooth, uninterrupted surface parallel to the transducer. Uterus 183 2.6 Ejaculate in Uterus ch ta QO fi u es PS Ultrasonograms of semen in the uterus. Images are of the uterine body 3) and horn (C) of a mare in estrus immediately before (A) and after breeding C). That the ejaculate can be visualized within the uterine lumen suggests another ential research and clinical application for the ultrasound technology. Although position of semen directly into the uterus during mating in mares has long been med (1), the ultrasound technology has provided the first direct evidence of itrauterine deposition. Studies of Cyclic Ultrasonic Changes wo experiments were done to characterize the temporal associations among ‘ation, estrous behavior, and the ultrasonically visible changes of the uterus (2). aracterization was made of the morphological changes in the ultrasound images of uterus during periovulatory and interovulatory periods at various times of the A 5.0 MHz transducer was used. At each sequential examination, the uterus given a score of 1, 2, or 3. A score of 1 indicated a homogeneous image, devoid vious endometrial hypertrophy and folding (diestrous-type uterus). A score of dicated that the ultrasound image of the endometrium was hypertrophied and led (estrous-type uterus). An intermediate score of 2 indicated that the \‘rasound image was heterogeneous, but without distinct folding. A score of | is represented by D and F and a score of 3 by A,B, and C in the images shown in Section 12.4. The uterine scoring was done independently by an operator who was not involved in determining estrous behavior or the day of ovulation. 184 Chapter 12 rire 1 St OV aa-2nd OV NN N ULTRASOUND ENDOMETRIAL SCORES eb: ek ed ee O > AGO O26) Nos Oo) DAYS Ultrasound uterine scores associated with the first and second ovulations of the season. A lower-case letter indicates a difference between groups for that day (a = P<0.1; b = P<0.05; c = P<0.01; n= 23). The ultrasonic profile for the second ovulation of the year was similar to that reported in the initial study (3). The scores were low between Days -15 and -9, increased progressively over Days -8 to -3, rapidly decreased between Day -1 and the day of ovulation, continued to decline after Day 0, and reached the low values characteristic of diesirus on Day +2. The maximum score of 3 was reached by 61% of the mares during the second period compared to only 22% during the first period. A decline in scores (e.g., 3 to 2) occurred before ovulation in 65% of the mares. Note that scores associated with the first ovulation of the year increased sooner but did not rise as high as for the second ovulation. The early rise is consistent with the well-known phenomenon of prolonged estrus before the first ovulation of the year (Sections 9,2D and 10.3C). The lower uterine scores on Days -6 to 0 preceding the first ovulation of the year provide rationale for the hypothesis that circulating estrogens are on the average lower before the first ovulation. Published data do not adequately consider this possibility (1,4). In this regard, circulating LH concentrations are lower in beh ney para Uterus © w zt = | Oo ” c¢ ° z kes Ww a 2.0 WW ce O° 24.75 a < fe WW 3 1.5 li 31.25 3 - Sad -15 -12 -9 -6 -3 OV 3 DAYS ssociations between uterine scores and behavioral scores. 185 ciation with the first ovulation than with the second (5). Exogenous estradiol sases the concentrations of LH (6), providing additional support for the ibility that the magnitude of the estrogen influence is lower prior to the first ation. An alternate, unexplored possibility is that the lower uterine scores in ciation with the first ovulation resulted from the absence of adequate ssterone priming. Sexual vior in response to teasing by a stallion was scored from -7 for an aggressively ative response to +6 for maximal sexual receptivity. The uterine scores closely ‘leled the behavioral scores. The data were combined for the first and second ovulations for this illustration, but the parallelism also occurred within each ovulatory period. Thus, the prolonged period of high uterine scores prior to the first ovulation was accompanied by a prolonged period of high behavioral scores. i Li) et 186 Chapter 12 Estrous behavior is common at irregular intervals in mares throughout the anovulatory season even when the ovaries contain only small follicles (<10 mm) ( A similar phenomenon occurs in ovariectomized mares (7) and fillies (8). In| experiment, such paradoxical estrous behavior was noted during the winter, m: weeks before the occurrence of the first ovulation. When estrous behavior occurr earlier than Day -15 before the first ovulation, the uterine scores were consistently low. A uterine score of 2 or more was not obtained for any mare before Day even though the maximum estrous behavioral score of +6 was obtained on 257 186 examinations between Days -48 and -18. Ultrasonic uterine scoring there! may have some value for distinguishing paradoxical estrus from true estrus. lack of agreement in the occurrences of behavioral estrus and an ultrasonical! detected estrous uterus during the winter contradicts the above conclusion th behavioral estrus and uterine estrus are attributable to a common cause --- increas concentrations of circulating estrogens. Perhaps the estrogen threshold or i} 118 ny — pee CD Co Ore duration of exposure may be less for paradoxical estrus than for the uterine effect. The absence of ovarian progesterone priming also may be involved in the failur uterine development, as noted above. An alternate possibility is that paradox estrus is not due to fluctuations in circulating estrogens. In this regard, paradox estrus occurs also in ovariectomized mares and may be due to steroids of adr: origin (7,9). ULTRASOUND ENDOMETRIAL SCORES cSeacOY 4.3 6 onthe oD) AS OV.23 Uterus 187 Jitrasound uterine scores during interovulatory periods in the summer and fall. Data are for 20 horse mares. The lower case letters denote s.\istical differences between groups. There was a small but significant increase in scores in the May-June group on Days 4 and 5. The reason for the small diestrus surge in May-June is not known. Perhaps it is related to the surge of FSH which has been reported to occur in early diestrus during May-July, but not during August- September (10). The two seasons also differed in the ultrasonic uterine profiles curing the periovulatory period; the September-October profile was broader due to ooth an earlier increase and a later decrease in scores. This result indicates a need to iy differences between seasons in ovarian hormone levels. rH These experiments indicate that the morphological changes in the uterus, as determined by ultrasound, are attributable to changes in the exposure of the uterus to estrogens, as indicated by the following: 1) parallelism between behavioral scores aid uterine scores and 2) similarities between the ultrasonic uterine profiles and the reported profiles of circulating estrogen concentrations. The similarity includes a crop in the profiles beginning one or two days before ovulation (Section 9.2A). Although further study is needed, these results indicate that the ultrasonic aracteristics of the uterus may be utilized as a biological indicator of estrogen posure. Perhaps a decline in uterine scores can help predict imminent ovulation in scheduling of breeding. As noted above, a decline in scores occurred before ation in 65% of the mares. Thus, if a uterine score decreases, ovulation may be imminent. The role of progesterone in the cyclic profile of uterine scores is not known. In this regard, preliminary observations did not disclose a difference in ulirasonic uterine scores between the anovulatory season and mid-diestrus. This suggests that low levels of estrogen may be more important than high levels of progesterone in the development of an endometrium that is ultrasonically characteristic of diestrus. 188 Chapter 12 12.8 Uterine Pathology 12.8A Cysts Examples of uterine cysts. A) Two cysts (arrows) in an excised tract viewed through a dorsal longitudinal incision. The cyst on the left is in the ventral wall of the uterus. It forms a circumscribed bulge covered with normal-appearing epithelium. The cyst on the right is in the dorsal wall. It is pedunculated and partially translucent. B,C,D) A cystic complex in another mare seen after opening the uterus (B), on side view as a large dimple on the ventral aspect of the caudal portion of a horn (C), and as the ultrasound image taken in a water bath (D). The orientation in C and D is comparable. Note that the complex in B,C,D encompasses the entire uterine wall, so that a bulge is present on both the external (C) and internal (B) surfaces. hh prop: ages of cystic uterine structures in six mares. A) Cystic complex in ventral uterine body. B) Cystic complex in the caudal portion of the right > horn. Portions of the complex on the left (arrow) appear to involve the thickness of the uterine wall. The large compartment on the right appears to ithin the uterine lumen. C) Small cyst projecting into the uterine lumen. This ould be confused with a Day-11 embryonic vesicle. A follow-up examination ' clarify its identity. An embryonic vesicle would have increased in size and oe in a different location. Identifying cysts before the time of pregnancy sis would alert the operator to their presence and minimize the chances of en identity. D) Cyst in a horn that could be confused with a Day-21 vesicle. | study would disclose that the echogenic nodule on the ventral aspect does not (no heartbeat) and the surrounding uterine wall is not disproportionately ned (Section 13.8). E) Two compartments of a cyst (upper left) within a > horn. The largest black area is a 20-day embryonic vesicle that has become in the uterine horn in the same area, so that the cystic structures are aching upon the vesicle. F) Cystic complex (upper left) indented into a Day-30 ryonic vesicle. The arrow is in the allantoic sac and is pointing to the embryo ‘‘. Note that in addition to being compartmentalized or multiple in form, a cyst / cast a bright specular echo (white line on upper surface, seen especially in A,B). 190 Chapter 12 Focal, fluid-filled cysts of the uterus of mares have been studied at necropsy (11), by uterine palpation through the rectum or through the cervix during estrus (11), and by biopsy (11) and fiber-optic techniques (12). Cysts are well-suited to study by ultrasound because they are fluid-filled and nonechogenic. The ultrasonic pathology of the cystic structures is characterized by compartmentalized or multilobulated, nonechogenic images. The cysts are located most frequently on the ventral aspect of the uterus. Cysts in this location frequently bulge into the lumen, as shown in the first illustration. Cysts that are dorsally located usually are pedunculated, as shown, probably from the gravitational effects associated with the weight of the collecting fluid. Sometimes such cysts move to-and-fro in response to uterine contractions. The images of cysts differed from those of free intraluminal collections of fluid by | the irregular compartments, the frequent involvement of the uterine wall, anc ‘he presence of specular echoes. A few cystic structures were found which were solitary and appeared to be confined to the luminal area. These were easily confused with embryonic vesicles. Such structures can be differentiated from a conceptus by ‘he knowledge that they were present at previous examinations, by failure o! ‘he : structure to develop and grow, and by lack of the mobility characteristic of an | |- to 15-day vesicle (Section 14.1). In addition, compartmentalized cysts also ca mistaken for an embryonic vesicle because a wall between compartments may ©\ve the appearance of the division between yolk sac and allantois. Cystic complexes or lymphatic lacunae frequently are seen by ultrasonography on the exterior aspec's of the uterus. These can be very extensive. Uterine cysts are receiving considerably more attention now than they did a few years ago. This renewed interest in cysts is due to their frequent observance during ultrasound examinations. Isolated cysts and the smaller complexes do not interiere with the establishment of pregnancy, according to limited study in our laboratory. However, much more study on the pathogenesis and importance of cysts is needed. The ultrasound technology likely will play an important role in such future studies. 191 8B Intraluminal collections of fluid iges of free intrauterine fluid collections. A,B,C) Small collections of lid in the uterine lumen. D,E,F) Large fluid collections of purulent exudate tra); note the scattered echogenic material within the nonechogenic areas, ing the presence of debris. The small collections likely would not be detected ’ by any approach other than ultrasound. They often appeared during late s and tended to recur repeatedly within individuals. They were usually - within the lumen, moving occasionally between horns and body. Often the ool or group of pools is elongated and irregular along a portion of the lumen iay be barely discernible (e.g., 3 mm high and 20 mm long). These elongated usually are detected in the uterine body. The free collections of fluid may be guished from cysts by their mobility, irregular shape, lack of ipartmentalization and involvement of the uterine wall, and lack of 492 Chapter 12 r echoes). Mares with small fluid collections gnosed pregnant and frequently had short interovulatory intervals. In three of three mares in which an embryonic vesicle was found floating in a small pool of free fluid on Day 11, the vesicle disappeared by Day was taken on Day 15 from one of the three mares (Section 15.6B) It is all pockets of fluid represented an exudate of an inflammatory minal fluid collections are discussed in Section 15.6B because n mares with such collections and mares that lost the a surrounding membrane (specula during the luteal phase were seldom dia 15; image B likely that the sm process. Small intralu of the similarities betwee embryo during Days 11 to 15. The large collections of fluid are attributable to pyometra or perhaps occasionally mucometra. Pyometra in the mare has been studied extensively (13). A resurgence of research activity should be stimulated by the availability of ultrasound scanners. The technique would allow undisturbed visualization and estimation of the amount of purulent material in the uterus. In several mares, it was noted that the amount of material in the uterus changed considerably between examinations. For example, in one mare an 80-mm diameter distention of the uterus at one examination was ‘reduced to a5 x 30-mm dilation one week later. Such rapid fluctuations are attributable to periodic discharge of much of the exudate through an open cervix. Some of the small fluid collections therefore may have been residual p\ lent exudates. 12.8C Fetal debris ute inc RE Uterus 193 etal bones in the uterus following death of the fetus. Note the shadow /act (arrow) beneath the bone in the sonogram on the left. The shadow indicates | the responsible object is an excellent reflector, which is characteristic of cified bone. The sonogram on the right shows an image of a fetal mandible in the nen of an estrous uterus. The bone was degenerative and porous and therefore did | cast a pronounced shadow. The mandible was removed and the uterus treated antibiotics. The mare conceived at the next estrus. Fetal bones can remain in the is for an indefinite time --- more than a year in one mare --- and render the mare vtile until removed. The bones are difficult to remove by flushing. Apparently »y become trapped or embedded in the endometrial folds and sometimes can be oved only by grasping with the fingers or an instrument. On one occasion, a ‘al bone was removed with a forceps while guiding the forceps onto the bone by ansrectal imaging. The incidence of retained fetal bones is not known, but it is pected that the condition will be diagnosed more frequently because of the ‘easing use of ultrasound scanners. “ERENCES . Ginther, O. J. 1979. Reproductive Biology of the Mare: Basic and Applied Aspects. Equiservices, Garfoot Rd., Cross Plains, WI. jayes, K. E. N., R. A. Pierson, S. T. Scraba, and O. J. Ginther. 1985. Effects of Strous cycle and season on ultrasonic uterine anatomy in mares. -heriogenology 24:465-477. . Cinther, O. J. and R. A. Pierson. 1984. Ultrasonic anatomy and pathology of ‘he equine uterus. Theriogenology 21:505-515. . Oxender, W. D., P. A. Noden, and H. D. Hafs. 1977. Estrus, ovulation and scrum progesterone, estradiol, and LH concentrations in mares after an increased photoperiod in winter. Am. J. Vet. Res. 38:203-207. - Freedman, L. J., M. C. Garcia, and O. J. Ginther. 1979. Gonadotropin levels in response to season and photoperiod in ovariectomized mares. Biol. Reprod. 20:567-574. 194 Chapter 12 10. ils AD. 13: _ Garcia, M. C. and O. J. Ginther. 1978. Regulation of plasma LH by estradiol and progesterone in ovariectomized mares. Biol. Reprod. 19:447-453. _ Asa, C. S., D. A. Goldfoot, M. C. Garcia, and O. J. Ginther. 1980. Sexual behavior in ovariectomized and seasonally anovulatory mares. Horm. Behav. 14:46-54. Wesson, J. and O. J. Ginther. 1981. Puberty in the female pony: reproductive behavior, ovulation, and plasma gonadotropin concentrations. Biol. Reprod. 24:977-986. mAs C07. A. Goldfoot, M. C. Garcia, and O. J. Ginther. 1980. Dexamethasone suppression of sexual behavior in the ovariectomized mare. Horm. Behav. 14:55-64. Turner, D. D., M. C. Garcia, and O. J. Ginther. 1979. Follicula ind gonadotropic changes throughout the year in pony mares. Am. J. Vet. Res. 40:1694-1700. | Kenney, R. M. and V. K. Ganjam. 1975. Selected pathological change: the mare uterus and ovary. J. Reprod. Fert., Suppl. 23:335-339. Mather, E ©. Kok Refsal, B. K. Gustafsson, B. E. Sequin, and i. Ly Whitmore. 1979. The use of fibre-optic techniques in clinical diagnosis and visual assessment of experimental intrauterine therapy in mares. J. | orod. Fert., Suppl. 27:293-297. Hughes, J. P., G. H. Stabenfeldt, H. Kindahl, P. C. Kennedy, L. E. Edq ist, D. P. Neely, and O. L. W. Schalm. 1979. Pyometra in the mare. J. Reprod. Fertil., Suppl. 27:321-329. | i at \ ’ ‘i 3 i ; 5 4 a of 5 ¢ 5 5 ; ) d 5 F i a 4 i ’ f i : ! i , f ‘ b 4 : ‘ ' WA oO Chapter 13 THE SINGLE EMBRYO ‘ly pregnancy diagnosis, monitoring the progress of the conceptus, detection as, and photographic documentation of pregnancy in mares were the primary uses of real-time, B-mode ultrasound scanners in large-animal iogenology. For several years, the equine embryonic vesicle received almost co = ive attention as the target of ultrasonic imaging of the mare's reproductive t. The early fascination with the equine embryo is attributable to several factors: ne large, spherical, fluid-filled form and therefore ease of detectability of the ‘conceptus, 2) the intriguing aspects of viewing embryos (heartbeat, fetal ments), as opposed to viewing other organs, 3) the conformation of the equine and the convenient access to it by the transrectal route, 4) the intensive rinary involvement in establishment of pregnancy in this species, and 5) the omics of the horse industry. ause of the early and continued attention, much is known about ultrasonic ing of the equine embryo. Five chapters are devoted to the embryo with n to the ultrasonic anatomy of the single embryo (this chapter), the physical ics of embryo-uterine interactions (Chapter 14), early embryonic death er 15), and twinning (Chapters 16 and 17). This chapter discusses the cts for early detection, the peculiar growth profile, the changes in the early s shape, and the changes in the embryo's position within the embryonic . The changing ultrasonic anatomy of the embryonic vesicle is described by ‘isons with gross and histological changes. Extensive use is made of diagrams otographs. Some of the diagrams and photographs of specimens were taken ie book "Reproductive Biology of the Mare" (1). However, the diagrams edrawn for compatibility with the companion ultrasonic images for each 196 Chapter 13 13.1 First Detection Se ee 2) eee Item 9 10 11 LZ 13 Ee eee First day of detection No. ponies 6(10%) 34(60%) 16(28%) 0(0%) 1(2% cumulative % 10% 10% 98% 98% 100% No. horses 1(5%) 12(63%) 5(27%) 1(5%) 0(0% cumulative % 5% 68% 95% 100% -- No. total 71(9%) 46(61%) 21(28%) 1(1%) 1(1%) cumulative % 9% 10% 98% 99% 100% Mean (+ SEM) diameter (mm) Ponies 3.3:20;5 3.9+0.2 4.6+0.3 --- 5.0+( Horses 3 3.9+0.3 4.8+0.4 6.00 --- ee ei ee Day and diameter at first detection (unpublished). A 5.0 MHz transcucer was used in 57 ponies and 19 horses. The embryonic vesicle was first detecied on Days 9 to 11, when it reached 3 to 5 mm in diameter (overall mean, 4 mm). Sixteen mares, later shown to be pregnant, also were examined on Days 7 and 8, Dut no vesicles were found. The vesicle was first detected by Day 11 in 98% of the mares. The few vesicles that were not detected until after Day 11 were smaller than the average for that day. There was no difference between ponies and horses in the day of first detection or in the diameter on the day of detection. It is emphasized that detecting the embryonic vesicle this early requires at least a 5.0 MHz transducer and 4 high-quality scanner. Perhaps the vesicle could be detected a day earlier with a 7.5 MHz transducer, but this possibility has not yet been investigated. Detection on Day 8 with improved technology would have considerable use in the embryo transplant industry. In this regard, in a series of 12 sonograms the mean distance from the transducer face to the uterine lumen was 2 cm (range 1.5 to 3 cm), which is compatible with the focal distance of some 7.5 MHz transducers. The Single Embryo 197 2 Location uterine horn NO. OF TIMES EMBRYO WAS IN UTERINE BODY OR HORN 9... .10.-:-P1. Aes Be 14 oe Oks DAY OF PREGNANCY ‘ean number of times that the embryonic vesicle was found in the uterine body or a uterine horn for various days (3). Twenty-five location ceterminations were made (one every five minutes for two hours) for each of five to 99 oD pe See S ven pony mares for each day. The equine embryonic vesicle is highly mobile n the uterine lumen from the day of first detection (Days 9 to 11) through Day onies) or 15 (horses) (Section 14.1). Thereafter, the vesicle is fixed in the il portion of one of the horns. Because of the mobility, the vesicle may be found here within the uterine lumen from the tips of the horns to the cranial aspect of stvix. However, the younger, smaller vesicles (Days 9 to 11; 3 to 6 mm) were frequently (60% of the time) found in the uterine body, as shown. Thereafter, were found in the body with decreasing frequency. The transducer should be 2d slowly when searching for early embryonic vesicles. The ultrasound beam at ocal point is thin (e.g., 2 mm) and passes quickly over a small structure. The esicle therefore may present only a "flashing" image which may be missed. A siematic scanning procedure should be developed, as previously described (Section ), to ensure that the entire uterine lumen is searched. 198 Chapter 13 13.3 Growth Profile in Cross Section MEAN OF HEIGHT AND WIDTH (mm) OF ULTRASOUND IMAGE 62 60 58 56 54 52 50 48 46 44 42 40 38 36 34 32 30 28 26 24 22 20 18 16 14 12 0 It Regression lines Horses —— Ponies —-—-—— Actual means Horses @ Ponies a Days 11-16 Days 16-28 Days 28-45 ae ee ee ae ae ee ee Oe ee er er ee 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 DAY OF PREGNANCY The Single Embryo 199 Growth profile of embryonic vesicle ( unpublished). Ultrasound images e taken from a cross section of a uterine horn or a longitudinal section of the u crine body. The vesicles were measured at the maximum height parallel to the side o ‘he ultrasound screen and at the maximum width parallel to the top of the screen. mean of height and width for each observation was used. The number of o servations for ponies and riding-type horses, respectively, was 9 and 9 on Day 9, id 19 on Day 10, 80 and 20 on Days 11 to 15, and 30 and 11 on Days 16 to 45. > were no significant differences between ponies and horses for any of the days. nathematical regression lines that best characterized the data for each mare type or each of the three indicated portions of the curve are shown. The regression were identical for ponies and horses on Days 11 to 16, and only one line is ‘bie. The rate of expansion appeared to increase gradually over Days 9 to 11, and ‘en expansion increased linearly over Days 11 to 16 (3.4 mm/day). However, the regression lines for both ponies and horses between Days 16 and 28 were S-shaped, with a distinct plateau during approximately Days 18 to 26. After Day 28, the regression lines were linear with a mean growth rate of 1.8 mm/day. The distinct p.ateau in the cross-sectional growth or expansion profile between Days 18 to 26 is peculiar, During this time, the image of the in situ vesicle did not change significantly in size in several experiments, whether measured by vertical (height) or horizontal (width) linear dimensions or by area (4,5). Size of vesicle is not useful for estimating age of the vesicle during Days 18 to 26, and morphological indicators must be used (Section 13.14). It is emphasized that the expansion failure is seen when the vesicle is viewed in a cross-sectional plane of the uterine horn. This phenomenon is readily observed, therefore, when a linear-array transducer is used. Lack of expansion is attributable to the increasingly tense uterine horn and resulting Tesis'ance to vesicle expansion in the cross-sectional plane (4). Actually, the vesicle does grow and expand during this time as indicated by measurements taken of excised vesicles (1). Apparently the turgidity of the vesicle decreases after Day 16. Therefore, the growing membranes are able to begin to conform to the irregularities of ‘he uterine lumen due to resistance of the uterine wall to cross-sectional expansion (5). The phenomenon of the cross-sectional growth plateau is directly related to changes in cross-sectional shape of the conceptus and is discussed further in Section 13.4. _ nul! 200 Chapter 13 Longitudinal and cross sections of two fixed Day-18 vesicles. vesicles are wider in the longitudinal views. Exaggerated expansion along the leng! of the uterine lumen may account for some of the failure of cross-sectional expan during Days 18 to 26. This hypothesis of disproportional longitudinal expansion offered by D. R. Pascoe during a discussion at the Equine Embryo Trar Symposium at Cornell University, 1984. 13.4 Shape of the Vesicle 100 stig 90 a ete 80 3 ” : 70 t the operator's attention. Without the specular echoes, the small yolk sacs ably would be missed. Specular echoes also may be present on the ventral aspect ‘he vesicle, but may be obscured by the echogenicity of an enhancement artifact section 5.4). In the past, specular echoes have been mistaken for the embryonic ; however, the embryonic disc is not ultrasonically visible. 208 Chapter 13 13.7 Days 17 to 19 Bilaminar omphalopleure Sinus terminalis Trilaminar omphalopleure Endoderm Splanchnic Trophoblast mesoderm Somatic mesoderm Chorion Exocoelom Chorioamnion Primitive gut Amniotic sac Embryo Mesoderm Ectoderm aoa Vascular DOaeODd- Nonvascular ——— Endoderm eeccccece Yolk sac Diagram of a Day-19 embryonic vesicle. The derivation of the somatopleure, splanchnopleure, and amniotic cavity are depicted. The somatopleure (literally, "body wall") consists of ectoderm (trophoblast) and mesoderm. The splanchnopleure (literally, "visceral wall") consists of endoderm and mesoderm. The terms bilaminar and trilaminar omphalopleure are used for the two- and three- walled portions of the yolk sac membrane (Section 13.6). The sinus terminalis is a prominent vein that encircles the yolk sac at the leading edge of the mesoderm. The yolk sac is three-layered and vascularized proximal to the sinus terminalis and two- layered without vascularization distally. At this stage, the chorion and chorioamnion are two-layered, nonvascularized membranes consisting of trophoblast and mesoderm. Later these membranes will fuse with a membrane that has not yet emerged (allantois) and then will become vascularized. Note that the amniotic cavity results from a union of the two folds of chorion which pass over the embryo. The Single Embryo 209 Histological sections from a Day-18 conceptus. A) Cross section through the embryo and adjacent membranes, showing the continuity of the mesoderm of the somites (s) with the mesoderm that surrounds the exocoelom (ex). The somatopleure is ventral to the exocoelom and the splanchnopleure of the yolk sac is dorsal. B,C) The three layers of the trilaminar omphalopleure are shown from bottom to top: trophoblast, mesoderm, and endoderm. Note the islets of developing blood cells in the mesoderm (B) and the early blood cells in the lumen of vessels (C). The cells of the internal or endodermal lining of the yolk sac are cuboidal, whereas the cells of the trophoblast are columnar with the characteristics of absorptive cells. The increasing absorptive function of the trilaminar omphalopleure is suggested by the changes in the trophoblastic cells, which are in contact with uterine milk, and by dev and conn pment of the vitelline (yolk sac) vascular system. The blood islands enlarge coalesce to form a continuous network in the yolk sac wall. This network ‘cts with similar channels in the embryo. Thus a vitelline-embryo circulatory system does nx efficie is established, complete with an increasingly powerful heart. The yolk sac ot contain stored food as in bird eggs, but with vascularization it becomes an ont organ for purveying nutritive material to the rapidly developing embryo. 210 Chapter 13 Exposed Day-18 conceptus. The arrows indicate the periphery of the exposed vesicle in the caudal portion of the right uterine horn. Noie that the free-standing conceptus does not sand under its own weight as it did at Day 14. Instead, it flattens and tends to spread, indicating a relative decrease in the volume of yolk sac ‘luid and loss of turgidity of the vesicle wall. This characteristic, combined with disproport l hypertrophy of the dorsal uterine wall (shown in the images below), may play a role in vesicle orientation, as described in Section 14.6. - ~ Cross-sectional ultrasonic views at Day 18. The cross-sectional shape of the conceptus at this stage is irregular and inconsistent. On average, however, the vesicles tend toward a triangular or guitar-pick shape. The apex is oriented dorsally and the smooth, rounded base is oriented ventrally. This is seen especially well in the sonogram in the center; the periphery of the uterine horn is delineated by arrows. The shape change is attributable to hypertrophy of the dorsal uterine wall, especially on either side of the mesometrial attachment (5). Other images of Day-18 vesicles are shown on page 200. The Single Embryo 211 .8 Days 20 to 22 Developing allantois Somatopleure Hindgut of amnion Ectoderm ET e = embryo st = sinus terminalis Endoderm ecccccce va = vitelline artery Mesoderm : ; : Vascular Q.Or vv = Vitelline vein Nonvascular ---—- Yolk sac Allantoic sac Diagram and photograph of a Day-21 conceptus. The removed vesicle was submerged and tilted slightly to expose the embryonal pole. The labels designate the embryo (e), yolk sac vasculature (vv, vitelline vein; va, vitelline artery) and sinus terminalis (st), the prominent collecting vein that encircles the vesicle. The sinus term nalis of the opposite wall can be seen through the translucent membranes and yolk sac fluid (arrow). The vesicle was spherical and 2.6 cm in diameter when submerged. It flattened to a diameter of 3.5 cm upon removal of the fluid in which it was submerged. The diagram shows the completion of amniotic cavity and the emergence of the allantois from the hind gut. The allantois grows into the exocoelom between somatopleure and splanchnopleure. The membrane distal to the sinus terminalis consists of two cell layers (ectoderm and endoderm) and is avascular. 212 Chapter 13 Thick dorsa! uterine wal! Yolk sac Embryo Emerging allantoic sac Thin ventra! uterine wail Cross-sectional ultrasonic view of a Day-22 conceptus. Note the thick, encroaching dorsal endometrial folds and the resulting guitar-pick shape. ‘he structure is primarily yolk sac, but the allantoic sac is emerging and is beginning to lift the embryo from the floor of the vesicle. Allantoic membrane is visible as a |ight (echogenic) line beneath the embryo. The embryo proper, which first becomes ultrasonically visible on approximately Day 20 (Section 13.5), is almost always in the ventral hemisphere of the vesicle when first detected. 13.9 Days 24 and 25 “Developing chorionic Allantochorion girdle Ectoderm cme | YOIK sac allantoic sac pM) ul embryo Endoderm. eeecccee | Allantoic sac © Wl Mesoderm Vascular O7-C* Nonvascular --—- The Single Embryo 213 iagram of the conceptus at Day 25 and specimen from Day 24. The a\\antois has become prominent and forms a cup under the amnion and embryo, | ing them from the floor of the embryonic vesicle. The union of the allantois and chorion (somatopleure) results in an allantochorionic placenta. At this stage the onic girdle, an important band of cells approximately 1mm wide, forms around the conceptus (6). The cells of the girdle will later invade the Ee deaettiue and give ( 1 to the endometrial cups. The cups produce eCG (equine chorionic ¢ otropin or PMSG). The specimen is viewed from the embryonic pole, ing embryo (e) and allantois (a). At this stage the allantoic sac is already larized and quite large compared with the embryo. Crown-rump length of the embryo was approximately 6 mm, which agrees with the ultrasonic measurements »icted in Section 13.5. The pulsating embryonic heart and the passage of blood in the circulatory system were readily observed in the newly obtained specimens. A color close-up of this embryo, showing its vascularity, appears in Section 13.5. Yolk sac Embryo Allantoic sac ‘ross-sectional ultrasound image of a Day-24 embryonic vesicle. The ‘illed allantoic sac is usually well-defined on the ultrasound image as a nonechogenic area beneath the embryo. The yolk sac and allantoic sac are separated by an echogenic line that represents the apposed walls of the two placental sacs. The embryo is in the form of an echogenic nodule on the separating line. The heartbeats (e.g., 150 beats/minute) can be detected with a 5.0 MHz transducer and high-quality scanner. flu 214 Chapter 13 13.10 Days 28 to 33 A specimen d a diagram of an embryonic vesicle at Day 30. Note the increased size of the allantoic sac anc the continued movement of the embryo away from the floor of the vesicle. The isolated specimen shows that the sinus terminalis (st) is still prominent, but the nonvascular bilam nar omphalopleure 0-0—0-0-0—> ee” 2 @ sscccccccccevecs [prover és (RM) Mare A 5.10 ( | | | | | | | I | | | | | | | | 15 <—_—_—$ rE eoccsese [frre . j ! ™3 20 ( | | 0 10 20 30 40 50 60 70 80. 90 100 11 mm. from cervix eai ‘yp I l % << UTERINE BODY —___ > i> Diagrammatic profiles showing changes in location of the embryonic ri vesicle in the uterine body in two mares (5). The first location determination cs was made at time 0, and sequential determinations were made every minute (Mare 3) | or every five minutes (Mare A). The cranial end of the cervix is depicted on the left and was used as the point of reference. RM = right middle (vesicle in m ile segment of right horn); LP = left posterior. At the beginning of the trial, the vesicle was at LP in mare A and 7 mm cranial to the cervix in mare B. Note the varia) ity in rate of movement. In an extensive series of trials, the mean rate of movement of the embr) nic vesicle in the uterine body was 3.4 mm/min, using the cervix or uterine cysts as fixed points of reference (5). Rapid point-to-point movement was seen occasionally. For example, in one mare the vesicle moved from a point next to the cervix to the posterior segment of a horn in approximately two minutes. In another, the time spent in moving from the tip of a horn to the posterior segment was only a few seconds. Embryo-Uterine Interactions 233 in yt yt to ement of a Day-14 vesicle (EV) over a cyst (5). The number in the loy ght corner of each image is the number of minutes from the beginning of the SE e. At the initial determination (min: 0), the vesicle was in the caudal portion of rn. The cyst is shown (upper central image) on the ventral aspect of the ute body 50 mm from the corpus-cornual junction. The relationship of the vesic.c to the cyst at each sequential determination was as follows: 5 minutes, approaching cyst; 6 minutes, beginning to encroach; 7 minutes, beginning to move Over ‘he top; 8 minutes, on top; 9 minutes, moving down the other side; 10 minutes, moviig away; 11 minutes, continuing to move away. This sequence demonstrated that sometimes considerable propulsive force is involved in embryo mobility --- a force ‘apable of distorting the vesicle when an impediment is reached. * z Be gue me ‘ a ———“( it” 234 Chapter 14 14.1C Progressive Nature of Mobility Percentage of observations in which the location change of an embryonic vesicle was in the same or opposite direction as tie previous location change (6). Data for the left and right horns were combined, and left and right sides in the diagram should not be taken literally. Movement was studied by first establishing an ini‘ial direction of movement; the next location change was recorded as being in the same or opposite direction. The presence of a star at the largest percentage of a pair indicates that the ‘wo percentages differ from equality. For example, when the initial direction of movement was from the cranial third of a horn to the middl« urd (movement in caudal direction), 83% of the ves cles continued to move progressively in the ‘ idal direction and 17% moved in the cranial dir ion (significantly different from equality). The bold arrows symbolically denote that the vesicle tended to continue moving progre: vely ‘nthe same direction when it was moving caudally in a horn and when it was moving cranially in the body. The reason for this dichotomy in progressive movement depending on location (horn versus body) is not known. Perhaps there is less intraluminal resistance to movement when the vesicle is moving caudally in the horns and cranially in the uterine body; alternatively, there may be an inequality in the magnitude of uterine contractions in one direction versus the other. When a vesicle enters a uterine horn from the body, the horn is apparently selected by chance. No significant preference was detected in statistical comparisons of right horn versus left horn, horn on the side of corpus luteum versus the opposite horn, horn which the vesicle had entered previously versus the opposite horn, and horn of eventual site of fixation versus the opposite horn. Embryo-Uterine Interactions 235 iD Expansion and contraction of vesicle Day-14 embryonic vesicle during expansion (left) and contraction. ‘pansion and contraction of the larger vesicles (Days 13 and 14) during the ity phase were observed in the uterine body and in the horns when viewed \gitudinally (3). Such contractions were uncommon with smaller vesicles (Days 9 all Qu @ pho The contractions occurred in cycles, lasting 5 to 14 seconds. Although iction of the vesicles also occurred under pressure when the transducer was d down against the uterus, the periodic contractions seemed to be due to local, e forces rather than to external pressures (e.g., pressure from transducer, nent of intestines). The contraction and expansion phenomenon was associated ultrasonically observable uterine contractions and to-and-fro movements of the e. The to-and-fro movements encompassed several millimeters and were mposed on the major location changes described above. The uterine ctions and resulting expansion and contraction and to-and-fro movements of onceptus were accompanied by a flowing or streaming appearance of the ietrium. The minor to-and-fro movements as well as the major, rapid point- it movements gave the vesicle the appearance of a flowing sphere, as though it in a slow-moving stream of water. This is illusionary, however, because onic observations indicate that normally the uterine lumen is obliterated and not contain measurable free fluid (7). Continuous ultrasonic observation of the in longitudinal sections indicated that the uterine contractions are present at ages of the estrous cycle and during early pregnancy (3), but it has not been termined whether the magnitude or nature of this apparent intrinsic activity is ected by stage of the cycle or pregnancy. These observations and the sequential ‘ographs of a conceptus being forced past a uterine cyst (Section 14.1B) indicate that the vesicle is subjected to periodic and profound uterine contractions. 236 Chapter 14 14.2 Control of Vesicle Mobility ee Ee Number of location chan mean tSEM Experiments | No. From horn Between body Among nin: and groups trials to horn and a horn segments ee Experiment I: Day 13 qd 0.6 £0.2 3.4 £0.4 9.9 +0.2* Day 10 Z. Oars DAD dalek 52S Experiment 2: Controis (Days 12-14) 10 O70 53205" 10.9 £1.1 Clenbuterol 10 0.0 +0.0 0.9 0.3 6021.1 Experiment 3: Controls (Days 12-14) 10 0.7:20;3* 2.8 £0.4* 8.0 £0.9 Simulated vesicle 12 OAgDd 1.4 £0.4 6.7 £1.0 ik Seeea os 2k ee Results of three experiments on the mechanisms involved in mob Approaches involved comparisons of the mobility of Day-10 and Day-13 ve (Experiment 1) (6), administration of clenbuterol (a suppressor of ut contractions; Experiment 2) (6), and insertion of an inert simulated vé (Experiment 3) (3). Location determinations were made every five minutes fc hours. The clenbuterol was given in a single intravenous injection on Day 12 and the mobility of the vesicle was compared immediately after treatment to controls. The simulated vesicles were prepared from the finger tips of su: gloves and were filled with water. The simulated vesicles were inserted into 1e on the day of insertion and on the following day. An asterisk indicates a significant uterus through the cervix on Days 12 or 13 of diestrus. Mobility trials were do! r 13; at of sical the difference between the group with the asterisk and the group below it. The greater mobility of normal vesicles on Day 13 than on Day 10 was confirmed. However, the vesicles were quite mobile within the body and tended to move from one segment to 4 more caudal segment more often than for the Day-13 vesicles (48% versus 10%). Clenbuterol decreased the number of location changes, indicating that uterine contractions are the propulsive force for vesicle mobility. act len ves inc! Sphe Embryo-Uterine Interactions 237 -13 embryonic vesicle (left) and simulated embryonic vesicle in a » horn (middle) and 20 minutes later in the uterine body (right). ulated vesicles were mobile, but the rate of movement was significantly less, min the table. Although other interpretations can be offered, the results of hree experiments are consistent with the hypothesis that there are intrinsic contractions capable of moving even inert objects. However, since the rate of ent was greater for an embryonic vesicle than for a simulated vesicle, it may the embryonic vesicle provides a stimulus that increases uterine contractility. imulus apparently is not due to simple distention, because the real and ed vesicles were of comparable diameters. The hypothetical stimulus d by the embryonic vesicle apparently is greater during the period of 1m mobility, whereas at Days 9 and 10, the vesicle may be primarily under the -e of the intrinsic uterine contractions --- the same contractions that moved iulated vesicle. Note that the mobility of the Day-10 vesicles seemed able to that of simulated Day-13 vesicles. Furthermore, Day-10 vesicles and ed Day-13 vesicles spent most of their time in the uterine body. Both were / mobile within the uterine body, however, and frequently traversed the full of the body. The hypothetical stimulating properties of the larger embryonic 's (Days 11 to 14), therefore, may not only increase mobility, but may also se the entries into the uterine horns. tors that may favor mobility of the early embryonic vesicle include the cal shape of the vesicle during the mobility phase, the encapsulation of the vesicle (7), which may give it more rigidity, and the longitudinal arrangement of the uterine folds (Section 12.5). 238 Chapter 14 14.3 Role of Embryo Mobility MARE COW Diagrammatic species comparison between mares and cows, showing the postulated manner in which systemic uterine-induced luteolys's is blocked by a mobile embryonic vesicle in mares and unilateral uterine- induced luteolysis is blocked by an expanding vesicle in cows (3). inthe absence of an embryo, a uterine luteolytic mechanism induces regression of the corpus luteum by release of a luteolysin that is believed to be prostaglandin [20. In the presence of an embryo, the luteolytic mechanism is blocked, maintaining the corpus luteum and its production of the vital hormone, progesterone. In mares, the luteolysin reaches the ovaries through a systemic or conventional whole-body pathway, whereas in cattle and sheep a direct unilateral utero-ovarian pathway is used (8). The uterine luteolytic mechanism is discussed in Section 9.2C. Embryo-Uterine Interactions 239 is postulated, therefore, that the mobility of the equine embryonic vesicle ts the small vesicle (3 to 12 mm) to contact all parts of the endometrium to at uterine-induced luteolysis, and that contacting the entire length of the netrium is important because of the systemic delivery of the luteolysin to the luteum. In contrast, in cattle, the conceptus needs to contact the endometrium a the ipsilateral side, because of the unilateral pathway. This is accomplished ly fixation of the embryonic vesicle (apparently by Day 12) followed by ion of the placental membranes toward each end of the horn (9). Temporal iships support this comparative postulate for the two species. In pony mares, iod of maximum mobility (Days 11 to 14) corresponds to the time of blockage e uterine luteolytic mechanism by the embryo (10; Section 9.2C). In cattle, the al membranes apparently cover the ipsilateral lumen by Day 18 after estrus. onceptus increases rapidly in length over Day 13 (not elongated) to Day 18 2), and some begin to extend into the opposite horn by Day 18 (11). In this the uterine luteolytic mechanism is blocked by the conceptus between Days 117 (12). results of a study by Florida workers (13) are compatible with the hypothesis »vement of the embryonic vesicle to various parts of the uterus is part of the nanism for blocking uterine-induced luteolysis in mares. The vesicle was d to certain areas of the uterus by ligation of the uterine horns. Six of nine egnant mares returned to estrus when the vesicle was restricted to the ipsilateral (orn on side of ovulation) or to the ipsilateral horn and the uterine body. In ‘’ experiment, the vesicle was restricted to the ipsilateral horn and some mares ».ven an exogenous progestin. Three of three pregnant mares that were not subst tone (fixation) after the blockage of luteolysis is complete (Section. 14.5C).. The movement of the embryo to all parts of the uterus may also aid metabolic exchange between the endometrium and yolk sac. ed with exogenous progestin lost the embryo compared to only one of five e given the progestin. \ddition to blocking luteolysis, the traveling vesicle may also distribute ces, throughout the uterus, that are involved in the gradual increase in uterine The increased tone in turn eventually results in the cessation of mobility 240 Chapter 14 14.4 Fixation 14.4A Day of occurrence Item Ponies Horses Total No. mares 47 14 Day of fixation (No. mares) Day 13 5 0 Day 14 13 2 Day 15 Zi, 1 Day 16 7 q Day 17 2 4 Day 18 0 0 Mean day (tSEM) 14.8 40.1 -#- 15.9 40.3 Diameter (mm) of vesicle on day of fixation 19.54+0.7 -*- 23.3 +1.3 Day of fixation in ponies and horses (unpublished). _ Fixation is cc‘ined as the cessation of mobility. In this study, the day of fixation was defined as the first day that the embryonic vesicle was in the same uterine segment during ever; daily examination. The difference between ponies and horses was significant for mean day of fixation and for diameter of embryonic vesicle on the day of fixation. The mean day of fixation was approximately one day later in horses (Day 16) than in ponies (Day 15), and the vesicle of horses was equivalent to one day larger on the day of fixation. In a previous study using a more critical definition of fixation (no location changes during a two-hour mobility trial), fixation occurred on Day 15 in five of seven ponies and on Day 16 in two of seven ponies. Day 17 was the latest day of fixation detected for single embryos in several experiments. Movement of the embryonic vesicle from one horn to another was not detected in any mare after the day of fixation in several experiments involving daily examinations until Day 40 in more than 100 mares. However, a high incidence of transuterine movement of the Embryo-Uterine Interactions 241 e conceptus was reported to occur after Day 20 (14), and some veterinary tioners have stated that such movement is common. We have been unable, ver, to detect location changes after Day 20. 3 Site of fixation ‘osiulated manner in which fixation occurs in the caudal portion of one of the horns. Fixation occurs almost always in the caudal portion of one of the uterine horns. The caudal portion of a horn contains a flexure or curvature, as shown in Section 8.1. It is postulated that fixation occurs at this site because it represents the greatest intraluminal impediment to continued mobility of the emoryonic vesicle; uterine contractions force the vesicle against the ventral wall of the site, as shown in the diagrams, thereby trapping the conceptus. It is not likely that fixation results from a cessation of uterine contractions; ultrasonic observations indicated that uterine contractions are present for many days after fixation occurs. However, no critical studies on the patterns of uterine contractions have been done in the earl /-pregnant mare. It is not known, for example, whether contractions tend to move ‘oward the fixation site from both directions. 242 Chapter 14 Side of Side of Reproductive No. __ ovulation _ te Tieaoneeas status mares eit Right Left Right a Lactating 421 47% 53% 60% —*— 40% Barren 268 45% 55% 41% —+— 59% Maiden 104 62% —-+—38% 33% —+—6/% All mares 793 50% 50% 50% 50% Se Effect of reproductive status on the side (left or right horn) of i fixation (1). Percentages that differ significantly are indicated by an asterisk. i Fixation occurred with greater frequency in the right horn in barren and maiden + mares: the difference was most exaggerated in the maidens. Perhaps the | intraluminal resistance of the flexure to mobility is greater for the right horn in maiden and barren mares; the difference in resistance between left and right horns may be less exaggerated in barren mares than in maidens because of the distending effects of previous pregnancies. In postpartum mares, the embryo became ‘ixed f more often in the left horn. This is attributable to the finding (14,15) that embryo a attachment is more frequent in the formerly nongravid horn. Perhaps the formerly ‘i nongravid horn is smaller, and its flexure is a greater impediment to vesicle m« oility than that of the formerly gravid horn. No difference was found in the time the vesicle spent in the gravid versus nongravid horns during the mobility hase, indicating that selection of the horn in which fixation occurred was not a function of the amount of time the vesicle spent in a horn before Day 15. The extensive embryo mobility phase prior to fixation likely accounts for the perplexing effects of reproductive status on the side of embryo fixation, as well as the consistent fixation of the embryo in the caudal portion of one of the horns. FACING PAGE: Ventral views of uterus on Days 3, 16, and 30 after ovulation. At Day 3 the uterus is relatively flaccid, whereas at Day 16 of pregnancy the uterus is more turgid. Note the extreme turgidity of the nongravid portions of the uterus at Day 30. The equine uterus is flaccid during estrus, increases in tone to an intermediate level by mid-diestrus and, if the mare is pregnant, further increases in tone over Days 12 to 25 (10). Embryo-Uterine Interactions 14.5A Description GO 243 244 Chapter 14 14.5B Relationship of tone to fixation OW ol Uterine tone a Sa : 30 ; j—t——4 ‘ Isd : 25 ot et eee -— mbryonic vesicle I I i Isd | I I I P Mean day 10 > ° . ae of ace Sy’ 20 15 UTERINE TONE (CODED) DIAM. EMBRYONIC VESICLE (MM) = ‘yn? gn? eh vt My 4a 10 13.) 145 4 7 io S20 —O 271 DAY OF PREGNANCY 4 Temporal relationships among vesicle diameter, uterine tone, an: the 5 day of fixation (3). Ten pony mares were examined daily on Days 11 DN. in Uterine tone and vesicle expansion increased over Days 11 to 16, and fixation ; occurred on the average on Day 15.8. These data demonstrate the temporal associations among vesicle expansion, the development of tone, and the occurrence of fixation. The results are consistent with the postulate (16) that fixation is a function of increasing expansion of the embryonic vesicle combined with incr¢ asing intraluminal resistance to mobility due to the development of uterine tone. Perhaps fixation occurs a day later in horses than in ponies (Section 14.4A) because the horse has a larger uterus, and the size of the vesicle is not different between the two types of mares (Section 13.3). In mares with twins, the role of vesicle size in determining when fixation occurred was indicated by the observation that the diameter of the largest vesicle on Day 15 was greater for mares in which movement was not detected after Day 15 (23.7 40.7 mm) than for mares in which movement was detected after Day 15 (19.4 +0.7 mm) (17). In one mare, the larger vesicle (25 mm) was fixed in the caudal right horn and the smaller vesicle (13 mm) was mobile within the same horn cranial to the site of fixation of the larger vesicle. These case histories, as well as the observance of continuing uterine contractions by ultrasound after the day of fixation, indicate that fixation is not due to a cessation of uterine contractions. Embryo-Uterine Interactions 245 i. 5C Control of uterine tone | 4.0 | ; eee | P, = PROGESTERONE (100 mg) is ‘< PREGNANT 3.5 1E, = 1mgESTRADIOL = eoeeece 5E, = 5mgESTRADIOL : | C = NONPREGNANT CONTROL . ; P, +1E 4 2 | 3.0 | | | | | | | | | ; | P | = | c 1 in | YAR | : - 2 r 5 -. | LN *, | uty | * | 1.5 } a 7 \ * | *C | ; l I 1:0an - = treatment period QO 4 8. 12 46 20 24) 25) ee DAYS inges in uterine tone in mares treated on Days 10 through 29 post- ovuation with combinations of progesterone or estradiol (18). There were ‘ive mares per group. Tone was scored from | (flaccid, as during estrus) to 4 (maximum tone, as in early pregnancy). The tone of early pregnancy was mimicked mos’ closely by administration of progesterone plus the low dose (1 mg) of estradiol. 246 Chapter 14 x a . é P4 = PROGESTERONE (100 ma) 2 ' wn E> = eEsTRADIOL (1 mg) e é ae 2a C = contro ¢ A eT ee ee a . j i" Pa a. wR, J it ie , ve a, ! WW z a \ za = s | frre cc Isd e Lu z : ae “ uw 1.54 } = = | <4 4a oe oon “A o° * saaneeertE no Sg. Seett +@ 1.0 | 1 é. Ziv Onl Oa Opa 220. ea I DAYS Changes in uterine tone in seasonally anovulatory mares treated with progesterone or estradiol (18). Maximum response was obtained wiih a regimen of progesterone followed by progesterone plus estradiol. It was concluded from these studies that the development of uterine tone curing early pregnancy is attributable to a priming effect of progesterone followed by exposure to a small amount of estrogen plus continued progesterone. The corpus luteum is the likely source of progesterone during this time. Perhaps the conceptus is the source of estradiol --- the conceptus does produce estradiol during this time (19). Our current working hypothesis is that estrogens produced by the conceptus begin to alter uterine contractions at Day 11. As a result, the rate of embryo mobility and the number of entries of the conceptus into the uterine horns increase (Section 14.2). The mobile embryonic vesicle blocks luteolysis in all parts of the uterus and at the same time distributes estrogens or some other substance that gradually increases uterine tone. By the time the blockage of luteolysis is complete (Day 15), the vesicle has expanded and the uterine tone has increased to the point that fixation occul’. Fixation is prerequisite to the vesicle orientation that begins shortly thereafter. Orientation is discussed in the next section. Embryo-Uterine Interactions 247 14.6 Orientation Mesometrium Uterine wall —— 3 Layers _ yok sac wall and embryo Day: 14 i 20 -ostulated mechanism for orientation of the equine embryonic vesicle (3). Orientation is defined as rotation of the embryonic vesicle so that the embryo proper is on the ventral aspect of the yolk sac. On Day 14, the vesicle is highly mobile and probably is not oriented. Shortly after the end of the mobility phase, the dorsal uterine wall begins to enlarge and encroaches upon the yolk sac. The encroachment is enhanced by the increasing uterine tone. The disproportional encroachment of the wall plus the massaging action of uterine contractions cause the yolk sac to rotate so that the thickest portion of the yolk sac wall (embryonic pole) assumes a ventral position. This concept is supported by the observations given in the remainder of this section. In one study, thickness of the uterine wall in cross section was measured at eight points relative to the face of a clock around the vesicle on Days 15 to 21, using nine or 10 ponies for each day. There were no significant differences in thickness of the wall among the eight locations on Days 15 and 16 (range of means, 7to° mm). However, significant distortion of wall thickness was detected on each of Days |7 through 21. The wall was an average of 3 mm thinner at 12:00, 4:30, 6:00, 7:30, and 9:00 than at 1:30, 3:00, and 10:30. This study confirmed that the uterine wall became thicker in the dorsal aspect of the cross-sectional view of the uterus beginning on Day 17 or 18. The hypertrophy of the wall was especially prominent On each side of the dorsal midline, accounting for the midline location of the apex of the triangular-shaped vesicle and the thinness of the uterine wall at 12 o'clock. 248 Chapter 14 Cross-sectional views showing the disproportional hypertrophy o, uterine wall between Day 17 (left) and Day 22 resulting in a chan shape of the conceptus. The outer limits of the uterine wall are delinea arrows. Note that the guitar-pick shape of the conceptus at Day 22 is du thickening of the dorsal portions of the uterine wall. The shape of the conce; cross-sectional views, as seen on the ultrasound screen, is not static. The dé ni shape of the wall is continuously altered, even while the vesicle is being observ Fi These alterations were superimposed on the generalized forms described in S« = 13.4 and 13.7 and shown in the image for Day 22. The continuous changes 11 : can be illusory due to slight movements of the transducer, which alter the v: area. In addition, external pressures on the uterine wall, such as those resultin: movement of the abdominal viscera, temporarily alter the shape of the vesicle. observation, however, indicates that shape changes are also due to contractions uterus. The uterine wall appeared to exert a kneading or massaging-like action © fixed vesicle. Contractions of the uterine wall also were present at sites far frox vesicle. These contractions seemed similar to those that occurred during the mo phase. Embryo-Uterine Interactions 249 iere sae: cement of the embryo toward the dorsal pole of the vesicle in a m ith normal orientation of the vesicle (top row) and movement to) the ventral pole in a mare with disorientation of the vesicle (3). Th yS Of pregnancy are 29, 32, and 35, left to right, for each mare. The dis ation was associated with natural embryo reduction on Day 24 in a mare that hac aterally fixed twins. The embryo proper on Day 29 in the mare with the inv , disoriented vesicle is indicated by an arrow; the yolk sac is beneath the eml Disorientation of the surviving conceptus was noted in association with unil lly fixed twins in five mares (20; Chapter 16). Apparently, sometimes there is an ‘nterference with the orientation process when two vesicles are present during the time that orientation is postulated to occur. 250 Chapter 14 HEY Disorientation of the embryonic vesicle at Day 33 in a mare that did not develop the uterine tone characteristic of pregnancy (3). In addition to being disoriented or upside down (left), the allantoic sac extended partially into the uterine horn (middle and right). This observation supports the concept that uterine tone and the resulting disproportional encroachment upon the conceptus tend to s/iape the vesicle and are necessary for the orientation to occur. REFERENCES 1. Ginther, O. J. 1983. Effect of reproductive status on twinning and on sice of ovulation and embryo attachment in mares. Theriogenology 20:383-395. 2. Ginther, O.J. 1983. Mobility of the equine conceptus. Theriogenology 19:603- 611. 3. Ginther, O. J. 1985. Dynamic physical interactions between the equine embryo and uterus. Equine Vet. J., Suppl. 3:41-47. 4. Leith, G. S. and O. J. Ginther. 1984. Characterization of intrauterine mobility of the early equine conceptus. Theriogenology 22:401-408. 5. Ginther, O.J. 1984. Intrauterine movement of the early conceptus in barren and postpartum mares. Theriogenology 21:633-644. ¢ 14. Feo, J. C. S. A. 1980. Contralateral implantation in mares mated during 15. Embryo-Uterine Interactions 251 eith, G. S. and O. J. Ginther. 1985. Mobility of the conceptus and uterine ontractions in the mare. Theriogenology (In press). rinther, O. J. and Pierson, R. A. 1984. Ultrasonic anatomy of the equine uterus. neriogenology 21:505-515. inther, O. J. 1981. Local versus systemic uteroovarian relationships in farm iimals. Acta Vet. Scand., Suppl. 77:103-115. ierson, R. A. and O. J. Ginther. 1984. Ultrasonography for detection of regnancy and study of embryonic development in heifers. Theriogenology 22:225- 255. cinther, O. J. 1979. Reproductive Biology of the Mare: Basic and Applied Aspects. Equiservices, Garfoot Rd., Cross Plains, WI. nw ty oO pm P = aon os oche, J. F., M. P. Boland, and T. A. McCready. 1981. Reproductive wastage \lowing artificial insemination of heifers. Vet. Rec. 109:401-404. ). Northey, D. L. and L. R. French. 1980. Effect of embryo removal and trauterine infusion of embryonic homogenates on the lifespan of the bovine corpus luteum. J. Anim. Sci. 50:298-302. 3. McDowell, K. J., D. C. Sharp, L. S. Peck, and L. L. Cheves. 1985. Effect of restricted conceptus mobility on maternal recognition of pregnancy in mares. quine Vet. J., Suppl. 3:23-24. ostpartum estrus. Vet. Rec. 106:368. en, W. E. and J. R. Newcombe. 1981. Relationship between early pregnancy sife in consecutive gestations in mares. Equine Vet. J. 13:51-52. ather, O. J. 1983. The twinning problem: From breeding to Day 16. Proc. ->‘h Ann. Conf. Am. Assoc. Equine Pract., Las Vegas, NV. .CGinther, O. J. 1984. Mobility of twin embryonic vesicles in mares. Theriogenology 22:83-95. 18. Hayes, K. E. N. and O. J. Ginther. 1985. Role of progesterone and estrogen in development of uterine tone in mares. Theriogenology (In press). 252 Chapter 14 19. Sharp, D. C. 1980. Factors associated with the maternal recognition of pregnancy in mares. Vet. Clinics of No. Am.: Large Anim. Pract.. 2:277-290. Chapter 15 EMBRYONIC LOSS “mbryonic loss in mares has been a neglected research area. This is regrettable, especially when considering the economic setbacks and anguish associated with loss of a pregnancy. Progress in this area has been slow because of the need for nonterminal methods of monitoring the progress of the embryo and the near absence of ‘inancial support for research. In the other farm species, progress has been made by assaying hormones and recovering the conceptuses at known stages. The slaughter-and-recovery approach apparently has not been financially feasible in mares, but embryo recovery by uterine flushing is beginning to be used. Rectal palpation has permitted diagnosis of pregnancy and embryonic loss after Day 20. Losses before Day 20, however, could not be differentiated from fertilization failure, except by inference when the mare became pseudopregnant. Ultrasonography has cut in half the formidable interval from Day 0 to Day 20; the emoryo now can be detected and monitored beginning on Day 9 or 10. As a result, several experiments on losses before Day 20 were reported in 1985 and more progress can be expected in the near future. The ability to detect and monitor the embryo by ultrasonic imaging beginning on Day 10 is fortuitous because Day 10 precedes the time that the embryo must block the uterine luteolytic mechanism. ‘he current status of the embryonic-loss problem is reviewed in this chapter. The discussion is limited to singletons, during the embryonal and very early fetal Stages. Loss of one embryo or both embryos in mares carrying twins is deferred to Chapters 16 and 17. Incidence, associated factors, time of occurrence, pseucopregnancy, and role of hormonal mechanisms are considered. The principle emphasis is on the ultrasonic pathology and the sequence of events from death to loss of the debris. Recent studies on the ultrasonic changes following induction of embryonic death are described. Discussions are available elsewhere on the roles of specific pathogens and toxins (1,2) and the associations between histopathology of the endometrium and pregnancy loss (3, 4). 254 Chapter 15 15.1 Incidence i Tim n Pregnancy Day of first Day of loss during pregnancy determination the given Ref. Breed diagnosis of loss time span 5 Standardbreds and Thoroughbreds Days 20 to 24 Days 40 to 50 5.5% 6 Trotters, draft, ponies Mean, Day 23 Mean, Day 43 5.3% a Thoroughbreds Approx. Day 20 Approx. Day 45 5.0% 8 Mixed (horses) Day 15 Day 50 14.4% ‘ 9 Quarter Horses Days 18 to 20 Days 34 to 36 8.6% ‘ 10 Ponies, horses Day 15 Day 50 10.4% 11 Thoroughbreds Days 14 to 18 Days 44 to 48 10.5% Standardbreds Days 14 to 18 Days 44 to 48 10.4% Results of recent studies of early pregnancy loss. Except for references 5 and 9, the data were obtained by ultrasound. The loss rates for the surveys that began after Day 20 and extended to approximately Days 40 to 50 (references 5,6,7) were 5.0 to 5.5%. Incontrast, the loss rates for the surveys that began well before Day 20 (references 8,10,11) were 10.4 to 14.4%. This comparison suggests that there may be a high rate of loss before Day 20. From examination of the table, it is suggested that one should expect an embryo-loss rate of 5% to 10% during the last half of the embryonal stage (Days 20 to 40). The data in this table, however, must be interpreted with caution because of divergent examination programs and reference points among studies and inconsistencies within studies. The loss rate for embryo recipients that were diagnosed pregnant by ultrasound was 13.3% between Days 15 and 50 (8). This loss rate was close to that of bred mares (14.4%) at the same research facility. 13.2 ultr Tese was bet rate in diff high (11) to 2% work pregr 20. prov ultras. Time of Occurrence ‘Days of egnancy Ponies | to 15 28/154 (18.2%) ) to 20 4121 (3.3%) 0 to 25 3/113 (2.6%) to 30 0/108 (0%) 30 to 35 1/94 (1.1%) 35 to 40 1/92 (1.1%) 40 to 45 0/54 (0%) 45 to 50 0/54 (0%) Horses 0/27 (0%) 1/27 (3.7%) 1/26 (3.8%) 1/25 (4.0%) 1/24 (4.2%) 1/23 (4.3%) 0/8 (0%) 0/5 (0%) regnancy loss during various days (10). sound from a herd of 154 research ponies and a contemporaneous herd of 27 rch horses. Pregnancy loss refers to the number of mares in which the embryo oresent on the first given day and dead or absent on the next. The difference sen the pony herd and horse herd on Days 11 to 15 was significant; the high loss Embryonic Loss 255 Total 5/148 4/140 1/134 2/119 2/116 0/62 0/59 Data were (3.4%) (2.9%) (0.7%) (1.7%) (1.7%) (0%) (0%) obtained by the pony herd is discussed in Section 15.6. There were no significant sices among day-groups over Days 15 to 50, although losses tended to be on Days 15 to 25. The results are consistent with another ultrasound study ‘a higher loss rate in Standardbred and Thoroughbred mares between Days 14 ‘3/42 losses) than between Days 28 to 42 (9/42) or 42 to 56 (10/42). Other rs did not find indications of susceptible periods (6,12); however, the ncies were diagnosed by rectal palpation and were initially detected after Day ne two ultrasound studies (10,11) and the inference noted in Section 15.1 \de rationale for the hypothesis that the incidence of embryonic loss is greater before Days 20 or 25 than after. This consideration emphasizes the usefulness of nography for early detection of embryonic loss, both for clinical and research purposes. Another ultrasound study (8) found a high rate of loss for Days 15 to 20 (26% of 61 losses over Days 15 to 50); a similar high rate was found for Days 30 to 35, but this has not been confirmed as a susceptible period in other studies. It is established that the embryo-loss rate exceeds the fetal-loss rate in mares with singletons. In an extensive transrectal palpation study in Australia, 75% of the detected losses occurred by Day 49. 256 Chapter 15 15.3. Factors Associated with Early Pregnancy Loss ————E—E—E—E—E——————————————————————————————— Maintained Lost Factors pregnancy pregnancy mR Number of pregnancies 359 45 Mean (+ SE) age (years) 8.4+40.2 —+— 10.1 +0.7 Mean (+ SE) day of conception (Jan. 1 = Day 1) 116.3 £4.1 121.7 +5.4 Reproductive status Maiden 52 (14%) 3 Foaling 229 (64%) 30 (67%) Barren 77 (22%) 12. (2%) Mares with Caslick's operation 246 (72%) 27 (66%) AOS RE ae a ee eee Effect of various factors on loss of conceptus up to Day 150. data are from an ultrasound study conducted at Cornell University in collabo with veterinarians (11). There was no significant difference in loss rates be Standardbreds and Thoroughbreds or among the groups of mares in the veterinary practices. Of the factors listed, only age was statistically associate pregnancy loss. In an earlier report (12), the percentage of pregnancy los significantly less for mares three to six years (15.1%) than for mares seven {to lese ition een hree with was nine years (19.3%), 10 to 12 years (21.5%), or older (22.2%). Failure to find an effect of reproductive status on pregnancy loss agrees with some previous reports and disagrees with others (14). Another consideration associated with embryonic loss is a history of loss. In an ultrasound study (13) in a herd with a high rate of loss during Days 11 to 15, the loss rate was significantly higher for re-established pregnancies in mares with a history of loss of an embryo (9/18) than for all pregnancies (38/154). Similarly, in another study (11), the loss rate for mares with re-established pregnancies was 40% C2) compared to 11% for all pregnancies. These findings indicate the presence of a chronic problem in at least some of the mares in these groups. Embryonic Loss 257 1.4 Losses before Day 10 ye incidence of embryonic loss before Day 10 is unknown. Collection of er ryos from subfertile donors before Day 10 resulted in a low embryo-recovery ra the donors and a high pregnancy-loss rate in the recipients (18). In a study by Woods and associates (11), subfertile mares were compared to maiden mares. Subfertility was defined as two consecutive years of reproductive failure; several mares also had histories of repeated pregnancy loss. Mares were evaluated by uterine culture and biopsy before breeding. Embryo recovery attempts were made on Days 7 and 8 by uterine flushing. Abnormality of embryos was based on diffuse necrosis, presence of organisms within the embryos, and absence of normal structure. - Item Maiden mares Subfertile mares Number mares: Positive culture 0/14 4/14 Endometritis 0/14 aa 6/14 Endometrial fibrosis 0/14 —*— 8/14 Diameter of embryos (mm) Day 7 0.6 £0.1 0.3 +0.1 2 et Day 8 1.2 +0.2 0.5 +0.1 Number of uterine flushes with: One or more embryos 29 (69%) —-*— 17 (40%) One or more normal embryos 28 (67%) —+— 8 (19%) Only abnormal embryos 1 (2%) —*— 9 (21%) Culture and biopsy results and condition of embryos in maiden and subjertile mares. An asterisk indicates a significant difference. Only the subferiile mares had indications of endometrial pathology, and more of them had embryos that were classified as abnormal. The subfertility seemed attributable, at least in part, to uterine pathology; however, biopsy samples from fertile, older mares were not available for comparisons. Embryos from subfertile mares were smaller than those from maiden mares. The abnormal embryos seemed irreversibly damaged. However, it is not known whether some of them might have continued to grow so that they could have been detected by ultrasound or palpation before being lost. Some of the subfertile mares had histories of pregnancy failure, suggesting that Some of the abnormal embryos may have been capable of some additional growth. 258 Chapter 15 15.5 Pseudopregnancy following Embryonic Loss 25 20 15 10 PROGESTERONE ng/ml 7 11 15 20 25 30 35 40 45 50 DAYS Progesterone levels and ultrasonic morphology. The large dots are the mean progesterone concentrations in 13 control, pregnant mares, and the vertical bars represent two standard deviations above and below the mean. The lines represent individual progesterone concentrations for mares that became pseudopregnant after embryonic loss on Days 12 to 25. In one mare (dotted line), the progesterone levels dropped between Days 15 and 20, but adequate progesterone was maintained to keep the mare in a pseudopregnant state. The ultrasonograms are for this mare. The vesicle was first detected on Day 13 and was undersized (3 mm). The thickness of the uterine wall surrounding the vesicle on Day 20 is uniform (arrows), unlike Day 20 in normal pregnancies. Size decreased over Days 20 to 27 and an embryo proper was not detected. On Day 28, the vesicle was 12 x 12 mm and appeared to be collapsing or fragmenting; the vesicle disappeared on the same day. Apparently the undersized vesicle was only partly successful in preventing luteolysis. Embryonic Loss 259 Period of Number Number loss (Days) lost pseudopregnant 11 to 15 38 11 (26%) 15 to 20 3 i (33%) 20 to 25 4 4 (100%) 25 to 30 0 a 30 to 35 p 2 (100%) 35 to 40 2 2 (100%) ncidence of pseudopregnancy following embryonic loss (10). In this dif! the phe on embr cons embry the prc pseudc facing ‘ey, pseudopregnancy was defined as maintenance of uterine tone and the corpus (14) and was diagnosed as described (Section 11.5B). The incidence of dopregnancy was greater for losses during the later periods (after Day 20). sumably pseudopregnancy would not be a consequence of loss before faintenance of the corpus luteum during pregnancy and presumably \dopregnancy is dependent upon blockage of uterine-induced luteolysis by the (Section 9.2C). ‘The occurrence of pseudopregnancy can be attributed to essful blockage of luteolysis by the embryo or its remnants and the absence of factors to actively initiate luteolysis (e.g., endometritis; Section 15.6B). © appears to be initiated by Day 11 and is completed by Day 15. For mares in he day of loss was known during Days 11 to 15, there was no significant ice between mares that became pseudopregnant (12.2 days, n=5) and mares not (12.7 days, n=16). Perhaps in some mares the products or fragments of ryonic vesicle lingered long enough to adequately block luteolysis. A similar ienon occurs in ewes (15). In mares, manual rupture of the embryonic vesicle ys 12 to 14 resulted in pseudopregnancy (16). Surgical removal of the ynic vesicle on Day 24 also was followed by pseudopregnancy (17), which is ‘ent with the occurrence of pseudopregnancy in all eight mares following onic loss after Day 20. In some mares, blockage of luteolysis is complete, and ogesterone profiles are not distinguishable from those of pregnant mares until pregnancy ends in approximately two months. As shown in the figure on the page, however, sometimes blockage of luteolysis is incomplete. 260 Chapter 15 15.6 Embryonic Loss during Days 11 to 15 15.6A Size and location of the embryonic vesicle The data in this section and in the rest of this chapter are from recent ultrasound studies of natural (10,13) and induced (19) embryonic loss. The losses described in this section are from the group of pony mares with an 18% loss rate during Days (1 to 15 (Section 15.2). The mares that became pseudopregnant following natural embryonic loss are excluded. The pseudopregnant mares were discussed in Section 1320; Be Ee Diameter of embryonic vesicle (mm) on: z Item Day 11 Day 12 Day 13 Dayi4 n ee SE eee 3 Controls 6.2 40.2 (90) —-9.3 40.3 (61) =—:12.5404 (1) 16.4 40.5 (41 : Day of loss: : od 12 5.1 40.5 (11) oe ss — c 13 Bo f04e dg) = sit) a —_ 14 s0t10. 3)... 80106 @) 107218, 3B) ats 15 48305 (4) 7.0409 (5) 10.2406 (5) 134107 (6 Total 5.2403 (25) 7.640.6 (15) 10.8407 (8) 134207 (5 ey" Mall 5) My! sd Diameter of embryonic vesicles before loss on Days 12 to 15. The number in parentheses is number of embryos. The average diameter of embryonic vesicles that were lost during Days 11 to 15 was reduced significantly during the examinations preceding loss. Although the average diameter was reduced, many of the vesicles were well within the range of sizes found in the controls. Most of the vesicles (84%), whether undersized or not, appeared to grow at a normal daily rate until the day of loss. Loss was diagnosed on the basis of failure to find a vesicle ina mare in which a vesicle was previously detected. There were no instances in which a vesicle was found on one day, not found on another day, but was subsequently found again. A reduction in size, collapse, or fragmentation of a vesicle prior to loss were not detected. du fre be in sn nur Wi 15 of (S ectic oe ‘Dp S Embryonic Loss 261 Controls Embryonic loss Day (% in uterine body) (% in uterine body) Day 11 39/81 (48%) —-*— 20/27 (74%) Days 12 to 14 31/108 (29%) 10/23 (43%) Total 70/189 (37%) —-*— 30/50 (60%) mber of mares in which embryo was located in the uterine body daily examinations. Embryos that were later lost were detected more ‘ently in the uterine body than embryos that were maintained. This may have » oo < pen =~ 15.6) dur col! breeding slated to the smaller mean diameter; normal vesicles spend more time (60%) uterine body on Days 9 and 10, and this phenomenon is believed related to size (Section 13.2). No differences were found between the two groups in the er of day-to-day location changes. However, further study using mobility trials e needed to critically evaluate whether some embryos that are lost before Day ve reduced mobility. This question is important because of the postulated role bryo mobility in the crucial blocking of the uterine luteolytic mechanism s 9.2C and 14.3). Intraluminal fluid collections versus embryonic loss. © presence of small (e.g., 5 to 20 mm) collections of fluid in the uterine lumen diestrus in some mares was described in Section 12.8B. Similar small, free mares ons of fluid in the uterine lumen were detected at least once during the 2 season in 40 mares during pregnancy examinations (Days 9 to 11). The vere from the group with the high incidence (18%) of embryonic loss on Days to 15 (Section 15.2). It will be shown in this section that there were similarities betwe embry a the mares with intraluminal collections of fluid and mares that lost the »on Days 11 to 15. 262 Chapter 15 Embryonic loss in association with small collections of fluid in ‘he uterine lumen. At Day 11, the fluid collection in the cranial uterine 5 dy contained a free-floating 5-mm embryonic vesicle. The vesicle was detectable within the fluid because of the two prominent specular echoes (arrows). One hour later, the fluid and vesicle were in the caudal right horn, indicating that both the fluid and its ! contained vesicle were mobile. The vesicle grew at an apparently normal rate over 2 Days 11 to 14. The vesicle was not detected on Day 15, but the fluid collection was present in the left horn (arrow). All of three mares which had both an intraluminal fluid collection and an embryonic vesicle lost the vesicle by Day 15. The other mares } with histories of small collections of fluid had several characteristics similar to ‘10s¢ a of mares that lost the embryo during Days 11 to 15, as shown in the following ‘hree tables: er Oe Pregnant Nonpregnant mares mares _ Pregnancy Group (number) (number) rate a" Controls 100 a 56% Mares with history of embryonic loss 10 4S 30% Mares with history of fluid collections Zz a 6% Fe Pregnancy rates in mares with a history, at some time during the breeding season, of embryonic loss or a small collection of intrauterine fluid. Pregnancy rate was significantly reduced in both groups and was extremely low in the mares with a history of collections of luminal fluid or exudate. Embryonic Loss 263 Number of Mean length ___interovulatory intervals _ (days) of 14 to 16 17 to 19 interovulatory Group Total days days interval (+SEM) ntrols 71 4 (6%) 6 (8%) 23.5 0.6 ervals with >mbryonic loss aS 6 (26%) 6 (26%) 19.6 +0.7 er intervals in mares vith a history of smbryonic loss a2 3 (14%) 5 (23%) 21.0 +1.0 tervals in mares with , history of intrauterine \uid collections 29 7 (24%) 7 (24%) 20.4 £0.9 ‘ ae igth of interovulatory intervals. The mean length of the intervals was -antly reduced for mares with embryonic loss, both for the intervals associated with loss and for intervals in which an embryonic vesicle was not detected. Similarly, the intervals were reduced in mares with a history of intrauterine fluid collections. Short interovulatory intervals tended to occur repeatedly in individuals with fluid collections. One mare had successive intervals of 18, 15, 14, 15, and 20 days and another had intervals of 23, 16, 17, 23, 15, and 16 days. Both mares were bred during each preovulatory period, but no embryos were found. The probability of the occurrence of an interovulatory interval of 14 to 16 days in nonpregnant mares without detected intrauterine fluid collections was only 5.6% (4/71). bake ona! sign he peed Cireulat eaiascciel Group Day 7 Day 11 Pregnant 17.24 40.9 (20) 17.42 40.9 (24) Nonpregnant; no history of embryonic loss or fluid collections 17.52 40.1 (13) 13.0 +1.4 (13) Embryonic loss 12.19 £1.1419) 10.2> +1.2 (25) intrauterine fluid collections 8.8°+1.8 (9) 2.3¢ 40.8 (9) Circulating concentrations of progesterone. Within each column, any two means with a different superscript letter are significantly different. 264 Chapter 15 Mares with embryonic loss during Days 11 to 15 were similar to mares with intrauterine fluid collections in the following ways: 1) pregnancy rate was reduced for the ovulatory periods not associated with loss, 2) mean progesterone concentration on Days 7 and 11 was reduced for the periods associated with loss, 3) mean length of the interovulatory interval was reduced both for periods associated with loss and periods in which an embryo was not detected, and 4) the condition was repeatable within individuals. These similarities suggest that the factors responsible for embryonic loss on Days 11 to 15 in this herd were the same as those responsible for intraluminal collections of fluid. MN CORPUS (*) LUTEUM EMBRYO (=) =) UTERINE LUTEOLYTIC MECHANISM The luteo-embryo-uterine triad. The physiology associated with this triad was discussed in Section 9.2C. The triad is shown again, here, because of its kely involvement in embryonic loss. The high rate of embryonic loss in this herd occurred during the time that the embryo must block uterine-induced luteolysis (Days 11 to 15). It is difficult, however, to determine the primary lesion in the complex interdependent events involving the corpus luteum, embryo, and the uterine luteolytic mechanism. The pathogenesis of embryonic loss on Days 11 to 15 could have involved any of the following: 1) Failure of the embr lock uterine- induced luteolysis. Such failure could be due to underdevelopment, degeneration, ot poor intrauterine mobility of the embryo. If this was the pathogenesis without other complications, the mares would be expected to return to estrus at the normal time. The embryonic vesicle was undersized in some of the mares with embryonic loss. The embryos that were lost were more frequently located in the uterine body. Perhaps the small size and more frequent body location led to a failure to block luteolysis in some mares. 2) Primary luteal inadequacy. This possibility cannot be assessed despite the low progesterone levels in some mares at Day 7, as well as ric reocysy oS YU inc ma dete fou 11 t Embryonic Loss 265 11. The low levels could have been secondary to early induced regression of an rwise adequate corpus luteum. The uterine luteolytic mechanism can be iciously activated by an early release of PGF2« in association with the presence uterine irritant (e.g., endometritis). Low progesterone values at Days 7 and 11 ‘t distinguish between a primary luteal inadequacy and early activation of lysis. Primary luteal inadequacy is further discussed in Section 15.9A. 3) 1e pathology. Our preferred hypothesis is that most of the embryonic loss in mares that did not become pseudopregnant was caused by uterine inflammation. similarities between mares with embryonic loss and mares with small auterine collections of fluid provide the rationale for this hypothesis. Uterine < e m4 2 \\ammation could account for the reduced mean length of the interovulatory rvals in mares with a history of either embryonic loss or intrauterine fluid ‘tions. The short intervals are attributable to early luteolysis induced by ation of the uterine luteolytic mechanism by the inflammatory process. A onic uterine pathological process could acccount for the tendency toward repeated onic loss and repeated shortening of the estrous cycles. Three of the yryonic vesicles that were subsequently lost by Day 15 were within a collection of atly free fluid in the uterine lumen. There were no instances in which the cle survived in mares with similar fluid collections. These observations provide ‘ional rationale for the hypothesis that the presence of small, ultrasonically cted fluid collections in the uterine lumen during diestrus or early pregnancy are ative of a pathological process. Uterine inflammation could also account for the osregnancy rates in mares with a history of such fluid collections; inflammation direc of tl ve interfered with sperm viability or caused death of the embryo before it was able by ultrasound. In addition, indications of uterine inflammation were on Day 16 in uterine biopsies of five of five mares that lost the embryo on Days ‘5. Subsequent studies of early embryonic loss, therefore, should include ind critical evaluation of the inflammatory condition of the uterus, regardless » primary focus of the studies. The work described in Section 15.4 also indica ©s that uterine inflammation may cause very early embryonic loss. 266 Chapter 15 15.7 Embryonic Loss during Days 15 to 25 concern ees? EMBRYONIC VESICLE HEIGHT (mm) ie oe ered ia NTS 0 DAYS POSTOVULATION Individual growth profiles for embryonic vesicles that were ost during Days 15 to 25. The black dot and associated heavy vertical lines for cach day represent the mean and two standard deviations above and below the mean for 35 mares that did not abort. Individual vesicles were defined as undersized when the diameter for a given day was two or more standard deviations below the mean for mares that maintained the embryo. The solid and broken lines represent the growth profiles for individuals with eventual embryonic loss and involve two seemingly different populations (normal-sized, solid lines; undersized, broken lines). An asterisk denotes the last day on which the embryonic vesicle was believed to be viable. Five of the nine vesicles that were lost between Days 15 and 25 were undersized during some or all of the preceding examinations, the remaining four embryonic vesicles seemed normal in size prior to loss. The vesicles that were undersized increased in diameter at an apparently normal rate. The size discrepancy between undersized vesicles and normal-sized vesicles was equivalent 0 approximately three days’ growth. The number of mares with embryonic loss after Day 15 was limited, and further studies involving vesicle measurements before loss v 4 Embryonic Loss 267 ceded; perhaps the loss of normal-sized versus undersized embryonic vesicles is vutable to different causative factors. There were no previously undersized les for mares that lost the embryo after Day 25; that is, all of the losses siated with undersize occurred before Day 25. a Number of Vesicle height (mm) undersized vesicles No. Mean Embryo Embryo ay mares Mean SD¢ -2 SD maintained lost 2 6.3 LS Jud 6/90 4/32 34 oo 2.0 aS 4/62 4/22 a 12.6 3.4 5.8 0/51 3/16 a5, 16.1 3.4 9.3 1/41 3/31 35 [7.9 3.6 1 1/84 5/14 2 35 24.8 oa 17.8 1/87 ae Total 13/415 (3%) 21/106 (20%) — - a EEE eee Relationship between undersized embryonic vesicles and subsequent oryonic loss. Undersized embryonic vesicles were more likely to be lost than niained. The difference in proportional number of undersized vesicles between mates that later lost versus maintained the embryo was significant for Days 13, 15, anc fot €MoDry the det detect occas embryonic loss and the incidence was not greater than what would be expected -) and tended to be significant (P<0.1) for Day 14. Undersized vesicles were “ in 3% of the examinations during Days 11 to 20 in mares that maintained the o and in 20% of the examinations in mares that lost the embryo. In retrospect, ‘obability that undersize indicated eventual loss, therefore, was 62% (21/34). ugh undersize was not a reliable indicator of impending loss of the embryo, the ‘.on of an undersized vesicle warrants frequent subsequent examinations if early ‘10n Of possible embryonic loss is desired. Oversized embryos (greater than the plus two standard deviations) were found in 13 mares and during only one and ionally two examinations per mare. Oversize was not associated with mathematically. A single case of an oversized embryonic vesicle during the yolk sac Stage has been reported (31). A similar vesicle was not found in the present series. Nee EE EEE 268 Chapter 15 Degenerating embryonic vesicle on Days 20, 22, and 23. The apparent degenerating embryonic vesicle is indicated by an arrow. The apparent vesicle had the appearance of a 10-mm echogenic ring (Day 20) or mass (Day 22) floating in an approximately 12 x 25-mm collection of fluid in the uterine body. By Day 23, the size of the fluid collection and the hyperechogenic mass was reduced. The mass and fluid collection were mobile, having moved from the right horn (lower left image) to the caudal uterine body (lower right image). There was no trace of a degenerating mass or uterine fluid on Day 24. Another mare with a collection of fluid and 4 degenerating vesicle at Day 20 had returned to estrus on Day 16 and reovulated on Day 25. The vesicle seemed normal in size and appearance during estrus on Days 16 to 19, except that it was within a collection of fluid, and both the vesicle and fluid changed locations between daily examinations. The presence of an apparent exudate in these two mares suggests that inadequate progesterone was due to uterine pathology leading to stimulation of the luteolytic mechanism. Embryonic Loss 269 — MOBILITY ——»: Ly POOR UTERINE TONE 10 1\ FIXATION FAILURE i EXPULSION? 1 ' OVULATION 5 ' ! 1 i : 3 1 5 i i ! be = 1 ; 1 10 15 20 DAYS Case history and ultrasonogram at Day 20. Note that the progesterone profile is similar to what would be expected in a nonpregnant mare. The embryonic vesicle was undersized when first detected (Day 12, 3 mm) but increased in diameter at 2 normal rate. The uterine tone characteristic of early pregnancy did not develop, probably because of regression of the corpus luteum and a resulting decrease in progesterone. Fixation did not occur at the expected time and the vesicle remained mobile, probably because of the poor uterine tone. On Day 20, the vesicle rhythmically expanded and contracted approximately every four seconds (height at expansion, 30 mm; at contraction, 18 mm). The vesicle is shown in the uterine body in contracted form. On Day 20 the mare was in estrus, the uterus was flaccid, and the uterine folds were ultrasonically edematous as for a normal estrus. The vesicle was still present on Day 21, but was gone on Day 22 --- the day of ovulation. Perhaps the vesicle was expelled through the open estrous cervix. It is noteworthy that the vesicle continued to grow at an apparently normal rate during estrus and until it was lost. The continued mobility and expansion and contraction in this case of natural loss were similar to what occurred following an injection of PGF2a at Day 12 (Section 15.9A). 270 Chapter 15 NORMAL PREGNANCY EMBRYONIC LOSS OD G2 = a Pc) = > Junction of : wns Boe Vesicle mobility Vesicle mobility Inadequate 6D Cee Inadequate uterine tone DAY 15 Vesicle fixation Fixation failure ao OD © f= > ing of < pueeen walls a Vesicle orientation Continued expansion and mobility AD Cy Estrus = Flaccid 2 uterus Expulsion < through a cervix ©) Orientation complete Embryonic loss associated with inadequate progesterone before the expected time of fixation. Compare the events of a normal pregnancy with the postulated sequence of events associated with the embryonic loss. The depicted events associated with this form of embryonic loss are consistent with the case history on page 269 and the results of experimental induction of embryonic loss (pages 276 and 277). 1 em See Day men kno volu estru Embryonic Loss 271 Embryonic Loss during Days 25 to 40 . ll eneration of an embryonic vesicle during Days 42 to 52. The proper is indicated by an arrow on Days 42, 43, 44, dnd 46. The embryo normal on Day 35 and had a heartbeat. The heartbeat was not detected on , and echogenic membranes were prominent above the embryo. These anes were discernible on Days 40 to 44. Their anatomical origin is not but seems to be related to a breakdown of the umbilical cord. The fluid 2 of the vesicle decreased over Days 40 to 52. On Day 52, the mare was in and remnants of the vesicle were not found; however, an intraluminal collection of fluid was present. 272 Chapter 15 Decrease in allantoic fluid between Days 40 and 45. The embry dead on Day 40 (no heartbeat). Note the echogenic membranes above th embryo at Day 40 and compare with the series on the previous page. Although of the fluid was not detectable at Day 45, there was still much solid debris (: Natural embryonic death occurred on Day 38, as indicated by cessation o! beat, in a pony that was being examined daily but was not part of the present The volume of placental fluid decreased approximately 75% by Day 46. Th: fetus and the associated small amount of placental fluid were present until Day occasionally changed locations (e.g., from caudal right horn to uterine bod Day 75 the mare began to show signs of estrus and only a small collection of fl detectable. The mare ovulated on Day 82. In another mare, the fetus seemed on Day 55, but on Day 56 a heartbeat was not detected and the allantoic flu reduced approximately 50%. Over the following 50 days, the dead fetus and vc of fluid remained constant and echogenic spots appeared in the fluid (pr fragments of placental membranes or fetus). The fetal mass and associated flui debris were mobile, sometimes involving the horns and other times the uterine The fetal debris was retained until Day 117, two months after death of the en The mare was beginning to show estrous signs on the day the debris disappeared. was ead uch )W). vart- ries, jead and On was rmal was ‘ume ably is and body. abryo. The mare ovulated on Day 124. These observations raise questions regarding the assumption that dead embryos and early fetuses are absorbed. Apparently, the solids and at least some of the fluids were retained sometimes for weeks or months until the debris was expelled through an open cervix associated with a return to estrus. In summary, the characteristics of loss during the late embryonal and early fetal stages included the following: 1) cessation of heartbeat, 2) dislodgment followed by mobility, 3) gradual decrease in placental fluids, 4) disorganization of placental membranes, and 5) hyperechogenic areas in embryo and membranes. Embryonic Loss 273 15.9 A Study of Induced Embryonic Loss heoretically, embryonic loss could involve any of the following: generation and absorption of the debris and placental fluids, 2) expulsion through the cervix of an entire, ruptured, or fragmented embryonic vesicle while in a viaple or degenerative state, or 3) a combination of absorption and expulsion. A fixed vesicle could degenerate in place, or it could be dislodged while viable or after death has occurred. To study the sequence of events associated with death of an embryo, 22 pregnant pony mares were ovariectomized or given an injection of PGi20 to induce embryonic loss. The sequence of events was monitored by daily ultrasound examination (19). Collapse or separation of the yolk sac membrane from the uterine wall before fixation (cessation of mobility), dislodgment of a fixed vesicle, and cessation of heartbeat were used as indicators of embryonic death or impending death. The day of complete embryonic loss was defined as the day when the vesicle or its apparent tissue remnants and fluids were no longer ultrasonically tected. On the day of complete loss, the cervix was examined digitally, and a cervix that readily accommodated one or more fingers was defined as patent. — ee 15.9A Induction of embryonic loss at Day 12 Number of mares Number of Embryonic Cervix days to Group Total loss open complete loss Control 3 0 iy a Ovariectomy a 3 3 3.0 40.6 Ovariectomy plus progesterone 3 0 --- --- PGF2e. 4 4 4 6.8 £0.2 Effect of ovariectomy or a Day-12 injection of PGF2« on embryonic loss. For unknown reasons, the embryo was lost sooner after ovariectomy than after PGF. treatment. The cervix was patent on the day of complete loss of the vesicle in 274 Chapter 15 all mares. Expulsion of a viable or dead vesicle through the cervix, therefore, could have occurred in at least some of the mares. Expulsion, while it was occurring, was not seen in this series, but the mares were examined only once per day. Expulsion of an intact Day-14 vesicle through the cervix and into the vagina was monitored by ultrasound in a mare that was not part of this study, demonstrating that expulsion can occur. Administration of progesterone (100 mg/day on Days 12 to 40) prevented the embryonic loss associated with ovariectomy. There were no significant differences between the ovarian-intact, control group and the ovariectomized, progesterone- treated group in diameter of the embryonic vesicle over Days 12 to 40, development of uterine tone, or day of fixation, and there were no apparent differences in ultrasonic morphology. These limited data indicate that progesterone may be the only ovarian substance essential for the development and survival of the embryo after Day 12. In another recent preliminary experiment, embryos developed in ovariectomized recipients treated with progesterone (20). The necessity of ovarian progesterone to survival of the embryo has been demonstrated also during i‘s later stages (21,22). Ovariectomized mares that were treated with progesterone lost the embryo more frequently when circulating progesterone concentrations wer: less than 2 ng/ml than when concentrations were higher (22). Concentrations ©: more than 4 ng/ml were associated consistently with survival of the embryo. Primary luteal progesterone deficiency has been reported in cattle (23,24) and sheep (25) and has been implicated in embryonic loss. Luteal insufficiency leading to infertility has been described in women but apparently is not fully accepted (26,27). Altnough progestin regimens are sometimes used in abortion-prone mares (28), neither a cause-and-effect association between diminished progesterone concentrations and natural embryonic loss nor a beneficial effect of progestin treatment during the embryo stage has been adequately documented. In an initial study in mares during the fetal stage (29), plasma progestin concentrations prior to abortion were lowet than in control mares. Other studies by the same workers during the fetal stage have raised the question whether pregnancy loss could result from progestin depression in association with stressful conditions (pain, disease, weaning, 30; withdrawal of the concentrate portion of a ration, 31). Although these trials were done during the fetal stage, they provide rationale for testing the hypothesis that stressful conditions (€.8. shipping, severe diet changes) could depress the progesterone output of the primary corpus luteum during the embryonic stage. This question must be resolved because of the obvious implications to breeding farms. Embryonic Loss 275 bryonic loss after ovariectomy on Day 12. The only indication of en nic death in all mares ovariectomized on Day 12 was the disappearance of the ve in a mean of three days without a subsequent ultrasonically detectable trace, su a uterine dilation. There were no significant differences between the control or nd the ovariectomized group in vesicle mobility or growth rate. Natural los smbryos during Days 11 to 15 also frequently involves disappearance without atrace after apparently normal growth and mobility (Section 15.6A). sansion (left) and contraction (middle and right) of an embryonic vesicle in the uterine body on Day 18 in a mare that was treated with PGF 20 on Day 12. The vesicle remained mobile and did not become fixed at the expecied time (Day 15). Note the separation of the yolk sac membrane from the uterine wall (middle image, at bottom of vesicle). The vesicle was gone on Day 19. 276 Chapter 15 Embryonic Loss 277 Ultrasonic appearance of the embryonic vesicle in another mare that as treated with PGF2« on Day 12. Note the apparently normal rate of growth. ne vesicle is shown while contracted and expanded for each of Days 15, 16, and 18. \ese expansions and contractions occurred in cycles lasting approximately 10 conds. Note the separation of the yolk sac membrane from the uterine wall at the ‘tom portion of the vesicle at Day 18. The vesicle was gone on Day 19. In all four mares treated with PGF2a.0n Day 12, the vesicles remained mobile il the day of loss on mean Day 19 (6.8 days after treatment). In contrast, all of the itrol vesicles became fixed by Day 15, which is the expected time in ponies. The ‘ended period of mobility in the group treated with PGF2. was characterized by »ythmic contraction and expansion of the embryonic vesicle every four to fifteen econds. The mobile vesicles were found frequently in the uterine body and ometimes next to the cervix. Similar mobility of the vesicle beyond the expected ime occurred in association with natural embryonic loss (Section 15.7). Expansion d contraction and intrauterine mobility occur also in normal embryonic vesicles before the day of fixation and probably are caused by uterine contractions. Uterine > increased between Days 12 and 17 in the control mares, similar to what has been ported. However, tone did not appear to increase beyond the Day-12 level in the roup treated with PGF20, probably because of PGF2a-induced luteolysis and loss of varian progesterone. The failure of fixation is attributable to the lack of evelopment of uterine tone, since fixation is believed to be caused by increasing ‘erine tone together with increasing expansion of the embryonic vesicle (Section (5B). Thus, the pathogenesis of embryonic loss following the removal of ovarian »gesterone before fixation was similar to that postulated for natural embryonic : (see diagram on page 270). Separation of a portion of the yolk sac membrane m the uterine wall was noted on the day before loss in two of the four mares ated with PGF2a on Day 12. In retrospect, this sign was probably indicative of vending loss, because it was not seen in this or previous studies in mares that maintained the embryo. The origin of the fluid (nonechogenic area) beneath the yolk sac membrane is not known. Perhaps it was yolk sac fluid that escaped through a cegenerating yolk sac membrane and gravitated to the lowest area. ht > > Y 278 Chapter 15 15.9B Induction of embryonic loss on Day 21 22 Embryonic vesicle in two mares treated with PGF2a. The vesicles appeared to develop normally until they disappeared on Day 23. On the day of loss, the endometrial folds were prominent and apparently edematous, as indicated by ultrasound, and the uterus had a doughy consistency on palpation. In one of the four mares in this group, there was a small pocket of fluid in the uterine lumen (mare 3). Loss of the embryonic vesicle in the four mares occurred on Days 22, 23, 23, and 24, respectively. There was no prior indication of impending loss except in one mare in which the vesicle apparently collapsed or fragmented on the day before complete loss. Heartbeat was detected each day prior to the day of loss or collapse. The cervix was patent on the day of loss except for the mare with the collapsed vesicle. Although loss by expulsion could have occurred in at least three of the mares, actual expulsion was not seen. Embryonic Loss 279 5.9C Induction of embryonic loss on Day 30 Dislodged and mobile vesicle at Day 32. The mare was given an injection ’‘GF20. on Day 30. The vesicle became dislodged on Day 32. The heartbeat of the isiodged embryo (arrow) seemed normal on Day 32, but was not detected the next ay. The series from left to right encompasses approximately five minutes and shows radual movement from the uterine body to the caudal right horn. At the beginning ‘the sequence (left), the vesicle is in the uterine body. In the middle image, the area 1e body is decreasing while the area in the horn is increasing. The arrow indicates arrow channel connecting the two areas. In the right image, the vesicle has completed moving into the horn. n one of the four mares treated with PGF2a on Day 30, embryonic loss occurred iree days, similar to the response in the Day-21 group. In the remaining three es treated with PGF2a and in another mare that was ovariectomized, the intact ryonic vesicle was dislodged on Day 31 (two mares) or 32 (two mares). odgment presumably resulted from a decrease in progesterone and uterine tone, er than from the direct stimulation of uterine contractions by the PGF20; the odgment occurred in the ovariectomized mare as well as in the PGF2a-treated es. The intact dislodged vesicle was mobile within the uterine lumen and was ad frequently in the uterine body. The dislodged and mobile vesicle became igated, sometimes encompassing much of the length of the uterine body and part 1 horn. The heart was beating in two of the four embryos on the day of odgment and continued to beat for one and two days, respectively, after odgment. The uterine folds were prominent by Day 32 and remained enlarged ’ three or four days. This apparent edema of the endometrium cannot be attributed -strus, because estrus was not detected during this time. The placental membranes ‘Zan to separate from the uterine wall on Day 32 or 33. LGM Embryonic Loss 281 Sequential changes in the embryonic vesicle in another mare treated 1) PGF2a on Day 30. The vesicle was apparently developing normally in the c»dal portion of a uterine horn on Day 31. On Day 32 the heart was beating, but the v-oicle was dislodged and mobile. The vesicle is shown extending the length of the uerine body (arrow indicates embryo proper). The two views at Day 33 show the vesicle upside down (left; embryo proper indicated by arrow) and a network of p.cental membranes (right). At Day 35 and thereafter, the embryo proper and p\.cental membranes were not visible. The resulting apparent fluid dilation decreased in size over Days 35 to 41. The small fluid collection at Day 39 is indicated by an arrow. There was no trace of the vesicle or uterine dilations on Day 42. In ths group, the fluid volume of the vesicles gradually decreased until the day of complete loss (Days 38, 38, 41, and 42, respectively, including the ovariectomized mare). The cervix was closed on the day of complete loss indicating that resorption may have played a role in eliminating the embryonic fluids. However, it was not mined whether the embryonic tissues and membranes were eliminated by a ir process, because in the absence of fluids the tissue remnants may not have on discernible ultrasonically. ae a Seek ee ep booed iere were no differences among the three PGF20-treated groups (Day 12, 21, or 30) in the interval from treatment to ovulation. The prolonged presence of embryonic debris in the uterus of the mares treated at Day 30 did not delay ovulation. If embryonic remnants that were not ultrasonically visible were present during the periovulatory period, they were likely eliminated at that time through the cervix. Additional study is required to clarify the relative roles of absorption and ulsion in embryonic loss. Expulsion may have occurred following treatment on Ww wer Days 12 and 21 because the cervix was patent on the day of complete loss, but actual expulsion was not observed. Absorption may have occurred following treatment on Day 30 because placental fluids disappeared while the cervix was closed; however, em ryonic tissues may have been ultrasonically invisible after the absorption of fluid an ay have been retained until cervical patency occurred during the periovulatory period. In this regard, natural late embryonic or early fetal death was followed by retcn'ion of the dead fetus, solid debris, and at least some of the fluids for weeks or mons until the mare returned to estrus (Section 15.8). The assumption, therefore, that cead embryos and early fetuses are absorbed may not be correct. ssscssasteda aosascteatanacseseine 282 Chapter 15 - GC Pymaatititon —_—_—_————— Normal Loss of uterine tone OD OD > — Mobility Gradual loss Estrus of placental fluids EP a of debris t Degenerative changes rough cervix Postulated sequence of events following loss of progesterone ‘tel fixation of the embryonic vesicle. In the normal pregnancy, the vesicle is “ixe in place by the turgid nongravid portions of the uterus. With loss of progeste:on the tone decreases and as a result the vesicle becomes dislodged. For the first cay so, the embryo of the dislodged vesicle sometimes remains viable, as indicated | heartbeat. The dislodged vesicle moves about inside the uterine lumen, althou gh probably is not as mobile as a normal vesicle during the mobility phase (Days 11 14). The dislodged vesicle spends considerable time in the uterine body, entering horn occasionally. The placental fluids gradually decrease in volume over ma days. The cervix remains closed until the mare returns to estrus sometimes weeks months (especially if eCG is present) after cessation of heartbeat. Although some the placental fluids are absorbed, the solid debris is retained until the cervix Ops with the return to estrus. The debris is expelled through the cervix. y ie Embryonic Loss 283 FERENCES Rossdale, P. D., and S. W. Ricketts. 1980. Equine Stud Farm Medicine. 2nd ed., Lea and Febiger, Philadelphia, PA. Roberts, S. J. 1971. Veterinary Obstetrics and Genital Diseases. 2nd ed., Edwards Brothers, Inc., Ann Arbor, MI. Shideler, R. K., A. C. McChesney, J. L. Voss, and E. L. Squires. 1982. Relationship of endometrial biopsy and other management factors to fertility of broodmares. J. Equine Vet. Sci. 2:5. . Kenney, R.M. 1978. Cyclic and pathologic changes of the mare endometrium as detected by biopsy, with a note on early embryonic death. J. Am. Vet. Med. Assoc. 172:241-262. _ Irwin, C. F. P. 1975. Early pregnancy testing and its relationship to abortion. J. Xeprod. Fert., Suppl. 23:485-488. “hevalier, F. and E. Palmer. 1982. Ultrasonic echography in the mare. J. teprod. Fert., Suppl. 32:423-430. . Simpson, D. J., R. E. S. Greenwood, S. W. Ricketts, P. D. Rossdale, M. ‘anderson, and W. R. Allen. 1982. Use of ultrasound echography for early iagnosis of single and twin pregnancy in the mare. J. Reprod. Fert., Suppl. 32:431-439. Villahoz, M. D., E. L. Squires, J. L. Voss, and R. K. Shideler. 1985. Some ybservations on early embryonic death in mares. J. Equine Vet. Sci. (In press). jinther, O. J. 1985. Embryonic loss in mares: Incidence, time of occurrence, ind hormonal involvement. Theriogenology 23:77-89. Sinther, O. J. 1985. Embryonic loss in mares: Incidence and ultrasonic norphology. Theriogenology 24:73-86. Woods, G. L., C. B. Baker, R. B. Hillman, and D. H. Schlafer. 1985. Recent studies relating to embryonic death in the mare. Equine Vet. J., Suppl. 3:104-107. 284 Chapter 15 Id, 133 14. lS: 16. 7. Ss oe 20. 21. oF 23: Bain, A. M. 1969. Foetal losses during pregnancy in the Thoroughbred mare: A record of 2,562 pregnancies. N. Z. Vet. J. 17:155-158. Ginther, O. J. 1985. Embryonic loss in mares: Pregnancy rate, length of interovulatory interval, and progesterone concentrations associated with (oss during Days 11 to 15. Theriogenology 24:409-417. Ginther, O. J. 1979. Reproductive Biology of the Mare: Basic and Applied Aspects. Equiservices, Garfoot Rd., Cross Plains, WI. Edey, T. N. 1967. Early embryonic death and subsequent cycle length in the ewe. J. Reprod. Fert. 13:437-443. Ginther, O. J. 1983. The twinning problem: From breeding to Day 16. Proc. 29th Ann. Conf. Am. Assoc. Equine Prac., Las Vegas, NV. Kooistra, L. and O. J. Ginther. 1976. Termination of pseudopregnanc administration of prostaglandin F2« and termination of early pregnancy by administration of prostaglandin F2a or colchicine or by removal of the embryo in mares. Am. J. Vet. Res. 37:35-39. S —— Squires, E. L., K. J. Imel, M. F. Tuliano, and R. K. Shideler. 1982. Factors affecting reproductive efficiency in an equine embryo transfer programme. J. Reprod. Fert., Suppl. 32:409-414. Ginther, O. J. 1985. Embryonic loss in mares: Nature of loss after experimental induction by ovariectomy or prostaglandin F2a. Theriogenology 24:87-9° minnicks. Ko, PIL, Sertick, MR. Cummings, and R. M. Kenney. 1985. Pregnancy in ovariectomized mares achieved by embryo transfer: A preliminary study. Equine Vet. J., Suppl. 3:74-75. Holtan, D. W., E. L. Squires, D. R. Lapin, and O. J. Ginther. 1979. Effect of ovariectomy on pregnancy in mares. J. Reprod. Fert., Supl. 27:457-463. Shideler, R. K., E. L. Squires, J. L. Voss, D. J. Eikenberry, and B. W. Pickett. 1982. Progestagen therapy of ovariectomized pregnant mares. J. Reprod. Fert., Suppl. 32:459-464. Sreenan, J. M. and M. G. Diskin. 1983. Early embryonic mortality in the cow: Its relationship with progesterone concentration. Vet. Rec. 112:517-521. Embryonic Loss 285 +, Maurer, R. R. and S. E. Echternkamp. 1984. Factors Causing repeat-breeder females in beef cattle. 10th Int. Cong. Anim. Reprod. AI. III:461. . Ashworth, C. J., D. I. Sales, and I. Wilmut. 1984. Embryo survival is influenced by the progesterone profile in ewes. Soc. Study Fert., Winter Meeting, IL. 2:74. . Gautray, J. P., J. DeBrux, G. Tajchner, P. Robel, and M. Mouren. 1981. Clinical investigation of the menstrual cycle. III. Clinical, endometrial, and endocrine aspects of luteal defect. Fertil. Steril. 35:296-303. Jones, G. S. 1975. Luteal phase defects. In: Behrman, S. J. and R. W. Kistner, (Eds.) Progress in Infertility. Little, Brown and Company, Boston, MA. Shideler, R. K., E. L. Squires, J. L. Voss, and D. J. Eikenberry. 1981. Exogenous progestin therapy for maintenance of pregnancy in ovariectomized mares. Proc. 27th Ann. Conv. Am. Assoc. Equine Pract., New Orleans, LA. Morgenthal, J. C. and C. H. Van Niekerk. 1984. Twinning, infectious and habitual abortions as related to total plasma progestagens in the Thoroughbred mare. 10th Int. Cong. Anim. Reprod. AI II:92. Van Niekerk, C. H. and J. C. Morgenthal. 1982. Fetal loss and the effect of stress on plasma progestagen levels in pregnant Thoroughbred mares. J. Reprod. Fert., Suppl. 32:453-457. Van Niekerk, C. H., J. C. Morgenthal, and C. J. Starke. 1983. The effect of nutritional stress on the plasma progestagen levels and embryonic mortality in twin pregnancies in mares. J. S. Afr. Vet. Assn. 54:65-66. Part Five TWINS : ; ¢ Chapter 16 TWINS: ORIGIN AND DEVELOPMENT She has twins." These three feared little words are being heard on breeding 1s with increasing frequency in association with pregnancy diagnosis --- not use of an increasing incidence of twinning, but because of the increasing ilarity of ultrasound scanners. In the past, many sets of twin embryos were detected and eventually were reduced to a singleton by natural processes. \trasonography, however, is displaying twin sets at much earlier stages and more reliably than by palpation. Because we are seeing more sets of twin embryos than before, our anxiety over twins may be greater. Fortunately, ultrasonography has allowed us to learn much about the natural outcome of twin embryos and has greatly improved our capabilities for handling diagnosed twin sets. In addition, double ulations, especially those occurring from a single ovary, are more readily jiagnosed by ultrasonography than by rectal palpation. Ultrasound scanners are a ecessity for effective twin-prevention programs. Their role involves detection of double ovulations and twin embryos, evaluating and monitoring the status of twin embryos, and providing for early manual elimination of one member of a twin set. is chapter and the next chapter will discuss recent findings on the twinning problem for the period extending from ovulation through the end of the embryo stage (Day 40). This chapter will consider the patterns of double ovulations and the factors affecting incidence, including breed, repeatability, and reproductive status. fi 0 bet oo “ae he relationships of synchronous and asynchronous ovulations to the establishment ' (win embryos will be discussed. The dynamic interplay between uterus and twin embryos (mobility, fixation) will be described. Much attention will be given to the incidence and processes of natural embryo reduction (elimination of one member of a twin set) in preparation for the discussion in the next chapter on methods of intervention for correction of twins. 288 Chapter 16 16.1 Patterns of Double Ovulations SPATIAL TEMPORAL STAGE OF CYCLE Same ovary Same day Both during estrus (unilateral) (synchronous) (double primary ovulations) Opposite ovaries Different days One during estrus (bilateral) (asynchronous ) and one during diestrus Methods of classifying the relationships of one ovulation to the other. True double ovulations are primary ovulations. That is, both occur in association with what is considered to be a single estrous period. Double primary ovulations may be defined as synchronous when they occur on the same day and asynchronous when they occur on different days. However, in some studies mares were examined every other day (1,2); therefore, synchronous ovulations were defined as ovulations that were zero or one day apart, since the two intervals could not be differentiated. Double ovulations one to 10 days apart with continuous intervening estrus or with estrus in association with each ovulation are classifed as asynchronous, double primary ovulations. Circulating progesterone concentrations do not increase adequately until after the second ovulation (3). The mare therefore usually rema ns in estrus, the two ovulations being associated with what is considered to be a sinzle estrous period. Distinct from double, asynchronous primary ovulations is the occurrence ©: an ovulation during diestrus, subsequent to the occurrence of the primary ovulation. Diestrous ovulations were first reported in 1972 (3) and their occurrence has been confirmed (4). Because diestrous ovulations occur during the progestational phase of the cycle, the mares do not show estrus and the cervix remains pale, dry, and sticky. Length of diestrus was not altered (3), indicating that the corpus luteum ‘hat results from the diestrous ovulation regressed at the same time as the primary corpus luteum. In the original study (3), a diestrous ovulation occurred in 24% of the diestrous periods (n = 261). However, in an ultrasound study of folliculogenesis in Quarter Horses and Appaloosas, apparent diestrous ovulations were detected in only 5/80 cycles (6%) (8). The incidence of diestrous ovulations, as well as double primary ovulations, may be markedly influenced by certain intrinsic or extrinsic factors. Diestrous ovulations could be responsible for some cases of twin embryos. J Twins: Origin and Development 289 Jitrasonograms of double preovulatory follicles. The two follicles lated on the same day. These double follicles and resulting double ovulations id have been more difficult to detect by rectal palpation than by ultrasound, use the two structures were in close apposition. As discussed in Section 17.1B, ction of double ovulations and consideration of the number of days between ations can be important components of a twin-prevention program. ___._ avey LO ee Single Double Single Double Item ovulations ovulations ovulations ovulations umber of ovulations 103 12 78 12 ameter (mm) of ‘ollicle on Day -1 Mean 42.8 —+— 35.5 453 —«— 36.0 SEM +0.6 as +0.6 £17 Range 35 to 70 25 to 50 35 to 56 29 to 42 No.<35 mm 0 5 0 6 diameter of ovulatory follicle on the day before ovulation. Two eys were done by ultrasound in riding-type horses (primarily Quarter Horses). diameter of the preovulatory follicle on Day -1 was compared between single ators and double ovulators. In both surveys, the preovulatory follicles in double ‘ators ovulated when at a smaller mean diameter. The reason for ovulating at a aller size is not known, and the number of observations was not adequate to permit ner study. This information is clinically important in determining when to iate breeding. If two large follicles are present, breeding should be initiated arlier if the intent is to breed the mare before an ovulation occurs. 290 Chapter 16 16.2 Factors Affecting Incidence of Double Ovulations 16.2A Breed Farm Type Number of Incidence of or or ovulatory multiple Ref. Technique group breed periods ovulations eee a” = 1 Palpation A Thoroughbred 153 22% 7 Palpation B Thoroughbred DiS 25% C Standardbred 237 13% D Standardbred 147 15% 1 Palpation E Quarter Horse 1872 9% F Quarter Horse 622 10% Appaloosa — 103 8% Thoroughbred 150 15% Ultrasound G Quarter Horse 85 8% Slaughterhouse H Pony S24 10% Slaughterhouse i Pony 127 11% Draft a 23% ieee Se Oe eee Results of recent studies on the effect of breed on rate of multiple ovulations. The great majority of the multiple ovulations were double (e.g., 94% in a slaughterhouse study)(5). Thoroughbreds and draft mares had the highest incidence (15% to 25%), Quarter Horses, Appaloosas, and ponies had the lowest incidence (8% to 11%), and Standardbreds were intermediate (13% and 15%). ‘The propensity for double ovulation in Thoroughbreds was demonstrated by 4 statistically greater incidence in Thoroughbreds on the same farm as other breeds (farm F) and on different farms in which examinations were done by the same veterinarian in the same season (farms B and C). Although supporting data were not obtained, two independent veterinarians and operators of Arabian farms concluded that double ovulations were not common in Arabians (10). In a recent report (11), the percentage of twins births recorded in stud books in Poland was 3% fot Thoroughbreds and 0.8% for Arabians. Twins: Origin and Development 291 (6.2B Repeatability =) bo ih es ne wi I id f Itip] Iti Mares with multiple Farm All mares ovulations during previous estrus A 92/875 (11%) . 10/41 (24%) B 158/1703 (9%) 11/102 (11%) C 44/196 (22%) 11/29 (38%) Totals 294/2774 (11%) 32/172 (19%) Frequency in which a set of multiple ovulations was followed by iother set during the next ovulatory period (1). Summed over the three arms, the probability of the occurrence of multiple ovulations for a given cycle was most doubled (P<0.05) when the preceding cycle also had multiple ovulations. nere was a tendency toward repeatability of multiple ovulations within individuals. epeatability of double ovulations has been reported for certain family lines (1) as ell as for individuals (1,5). For example, a mare with a history of two sets of twin ‘etuses in four years had five ovulatory periods during one season and all invoived iouble follicles or ovulations. Among the mares on another farm with records for everal (three to five) ovulatory cycles, two mares had an unusual predisposition for 1 Ovulations and pregnancies. These two mares were the only mares on the farm vat had three or more cycles with double ovulations and were the only mares in f } uch twin pregnancies were established twice during one breeding season. The two ‘es were dam and daughter. Similar case histories were reported recently (22). oarently the occurrence of double ovulations and the establishment of twin enancies are heritable characteristics which could be selected against. Selection inst twinning, however, is not likely to occur because of circumstances peculiar to © equine industry. The tendency toward repeatability is an important consideration ‘win-prevention programs --- problem mares can be given more critical attention. 292 Chapter 16 16.2C Reproductive status i Reproductive Incidence of multiple ovulations on: status Farm A Farm B Farm C Totals ee cap annnnnnEIT ean ae Foaling mares 26/358 (7%) 15/95 (16%) 55/1039 (5%) 96/1492 (6% Barren mares 53/370 (14%) 17/57 (30%) }-- 103/664 (16%) 196/1244 (16% Maiden mares 11/109 (10%) 12/44 (27%) See ee Effect of reproductive status on the incidence of multiple ovulation: on three farms (1). The double ovulation rate was much reduced in foaling mar« (6% versus 16%). This reduction occurred during the first 80 days postpartum I was not detectable thereafter. Similarly the rate of twin conceptuses is lower mares bred during the postpartum period (10, 22). The reduced incidence of dou ovulations in foaling mares may be due to a suppressive action of nursing reproductive function. No effect of age on double-ovulation rate was found in | study. However, in a more recent slaughterhouse study (5), double ovulation was more common in mares aged 6 to 10 years (18%) than in mares aged 2 to 5 ye (14%). Also, the rate of twins births was recently reported to be higher in olce mares (11). —_ wm ln ee 2 In summary, breed, individual repeatability, heritability, reproductive status, and age are factors which have been demonstrated to affect the double-ovulation rate. At the present time, convincing demonstrations of the involvement of other factors, such as level of nutrition, are not available. Some reports have concluded that multiple-ovulation rate or rate of twin births is affected by month or season (3, 12), whereas other studies failed to find a seasonal effect (1, 5, 9, 11). However, in a study of postpartum mares (12), the incidence of multiple ovulations for the first postpartum ovulatory period was lower for January to March (7%) than for April to May (22%). In addition to the known influences, some of the disparity in the wide range of reported frequencies of double ovulations can be attributed to differences in the data-gathering process. Some studies, for example, do not differentiate between double ovulations and a single primary plus a diestrous ovulation. Twins: Origin and Development 293 16.3 Double Ovulations and Pregnancy Rate No. ovulations/mare (based on 16% double- ovulation rate) .6.3A Expectancy calculations No. bred mares No. embryos/mare None One (based on 50% conception rate/ovum) No. or % mares with 46 50 4 given no. of embryos Pregnancy rate: All mares= 54%(54/100) Double ovulators= 75% (12/16) Twin-embryo rate: All bred mares= 4% (4/100) All pregnant mares:7% (4/54) Bred double ovulators= 25% (4/16) Pregnant double ovulators= 33% (4/12) Percentage of mares expected to have a given number of embryos. 2se expectancy calculations assume a 16% double ovulation rate and a 50% conception rate per ovum. The 16% and 50% figures are convenient, representative estimates taken from a review of the literature (6); 50% is the pregnancy rate for an estrous period with a single ovulation. The calculations also assume that in double ovulators the presence of one ovum or resulting embryo does not affect the other. 294 Chapter 16 16.3B Synchronous ovulations N lations / ovul eri Double Predominant Single synchronous Ref Farm breed (Pregnancy rate) (Pregnancy rate) 1 A Quarter Horse 851/1516 (56%) 12/14 (86%) 13 A Quarter Horse 967/1776 (54%) 19/23 (83%) 1 B Thoroughbred 85/127 (67%) 3/3, (100%) 1 C Thoroughbred 143/242 (59%) 21/27 (78%) 1 D Thoroughbred 241/451 (45%) 3/3 (100%) 7 E Thoroughbred 128/235 (55%) 29/39 (74%) a Pr Standardbred 94/124 (74%) 16/21 (76%) d G Standardbred 124/206 (60%) i2ZA7 Ti Totals 2631/4677 (56%) 115/147 (78%) Pregnancy rates in mares with single and synchronous doubl: ovulations based on breeding-farm records. Synchronous ovulations wei defined as those occurring on the same or consecutive days, because most mares wei examined at two-day intervals. Pregnancy rate was defined as percentage of mares diagnosed pregnant (usually by Day 22) regardless of the number of embryos. The pregnancy rate was significantly higher for synchronous double ovulations (787%) than for single ovulations (56%). This phenomenon was seen in postpartum, barren, and maiden mares (13), and in mares in which multiple ovulations were induced by injections of a pituitary extract (2,10,14). The proportion of pregnant, double- ovulating mares was not significantly different from what would have been expected if each ovum of the double ovulators had the same chance for development as the ovum in single ovulators (10). Also, the observed pregnancy rate in the double ovulators (78%) is close to that derived from the expectancy calculations in Section 16.3A (75%). These results indicate that each ovum resulting from synchronous double ovulations has the same potential for becoming fertilized and yielding a viable embryo as the ovum from a single ovulation. The presence of two large preovulatory follicles can be considered desirable because the probability of establishing pregnancy is much improved. Twins: Origin and Development 295 3C Asynchronous ovulations when number of ween 1 lations was: 0 2 4 6 8 12/14 (86%) 13/13(100%) 11/13 (85%) 13/17 (76%) 7TN5 (47%) 19/23 (83%) 13/15 (87%) 12/18 (67%) = 12/22 (55%) 11/29 (38%) 31/37 (84%) 26/28 (93%) 23/31 (74%) 25/39 (64%) 18/44 (41%) ffect of number of days between ovulations on the pregnancy rate. ‘Tne farm is the same as Farm A in Section 16.5. The double ovulations were apparently primary ovulations. The mares were bred before each ovulation when the ovulations were four or more days apart. Note the high pregnancy rates when the terval between ovulations was four to six days or less. These data indicate that each ovum from asynchronous as well as each ovum from synchronous double ovulations viable, except when the interval between ovulations exceeds four to six days. 10.3D Number of ovulations versus embryos Incidence (pregnant mares only) Double Twin con ingl 1 lation Breed ovulations Internally detected | Externally observed Quarter Horse 91/947 (10%) --- 2/947 (0.2%) Standardbred 20/144 (14%) 3/144 (2%) --- 18/110 (16%) 0/10 (0%) --- ublished data showing the discrepancy between the high incidence of double ovulations and low incidence of detected twin conceptuses. Data are from farms in which all mares were bred regardless of the number of preovulatory-sized follicles. Note that the proportion of mares with twins falls 296 Chapter 16 far short of the expected number (twins expected in 4% of all bred mares and 7% all pregnant mares; Section 16.3A). Both ova in double ovulators are viable and fertilizable, as indicated by the high pregnancy rates (Sections 16.3B and 16.3C), Therefore, the low incidence of externally observed twins, when compared to the incidence of double ovulations, can be attributed to a natural biological process ‘or eliminating an excess embryo. This process has been termed embryo reduction (3). 16.3E Patterns of ovulations versus number of embryos Note: This section has been updated in this (third) printing to include new information from the following reference: Ginther, O. J. 1987. Relationships among number of days between multiple ovulations, number of embryos, and type of embryo fixation in mares. Journal Equine Veterinary Science, 7:82-88. a Number of days Number of between pregnant - Number of embryonic ovulations mares vesicles per mare 0 a] 2 et 1 ZZ 1.6 40.1 Z Ly oe 3 2 LD a) Effect of number of days between ovulations on the maximum number of embryonic vesicles per mare in mares that became pregnant. Some of the mares were stimulated with a pituitary extract, but this did not appear to affect ‘he results. Number of ovulations was based on daily ultrasonic scanning and the numer of vesicles was determined by daily scanning on Days 11 to 15 (mobility phase). ‘The number of days between ovulations did not alter the number of embryonic vesicles, whether data were categorized by individual days or by groups of days. Disconcertingly, two previous studies (2,13) of breeding-farm records in our laboratory indicated that twin embryos are far more likely to occur after asynchronous double ovulations than after synchronous ovulations; the records were obtained by transrectal palpation of the ovaries at two-day intervals for detection of double ovulations and palpation of the uterus beginning on approximately Day 20 for detection of twin embryos. Summed over the two previous studies, twin embryonic icles were diagnosed in 67/2360 (3%) pregnant mares with one detected c Twins: Origin and Development 297 ulation, 18/98 (18%) with asynchronous ovulations, and 0/58 (0%) with chronous ovulations. In addition, in retrospective examination of ovulatory erns for mares with observed twin abortions or foals, one ovulation was recorded 31/36 (86%) mares, two asynchronous ovulations in 5/36 (14%), and two chronous ovulations in 0/36 (0%). The reasons for the profound discrepancy veen the ultrasound results (shown in the table) and those (2,13) which led to the ichronous/synchronous hypothesis do not seem resolvable. The ultrasound study a planned test of the asynchronous/synchronous hypothesis utilizing the most cal technology available for the in vivo detection of ovulations and early ryonic vesicles. For these reasons, the results of the ultrasound study seem more ible. The conflicting results would be resolvable if asynchronous twin vesicles were more likely than synchronous vesicles to become bilaterally fixed and therefore mo: feta was diar that > likely to be detected by transrectal palpation and more likely to survive to the stage (18). However, a preference for bilateral fixation of asynchronous twins not found in the ultrasound study -- to the contrary, twins with a difference in eter of 23mm underwent unilateral fixation more frequently than did vesicles liffered by <3mm. The possibility that the development of twins following one detected ovulation is due are |] follc T! dt SI O' Th $0) Ov tw CO! Occ SOr 9 the splitting of a single fertilized ovum seems unlikely because equine twins ‘ported to be almost always dizygotic (15). In our laboratory, there has been no insti dete ace where twin embryos were found in a mare with only one ultrasonically ‘ed ovulation or corpus luteum. Most likely, the reported cases of twins owing one detected ovulation (2,13) resulted from a second undetected ovulation. econd ovulation could have occurred on the same or a later day during estrus or g diestrus (Section 16.1). Sperm apparently have a long survival time in mares, it is not uncommon for pregnancy to develop in mares bred six days before tion (6). Furthermore, the ova from diestrous ovulations are fertilizable (16). fore, it is likely that sperm deposited before an estrous ovulation may times survive long enough to fertilize an ovum from a subsequent estrous ‘tion or diestrous ovulation. In this regard, some veterinarians believe that are more likely to occur when highly fertile stallions are used (1). These derations encourage critical testing of the hypothesis that a second ovulation ‘ting many days after breeding, whether during estrus or diestrus, can ‘times result in twin embryos. 298 Chapter 16 16.4 Interplay between the Uterus and Twin Embryos 16.4A Mobility Ultrasound images of multiple embryonic vesicles. Images A, B, C were taken with a 3.5 MHz transducer and D, E, F with a 5.0 MHz transducer; note the centimeter scale to the left of images A and D. Images A, B, C show the location changes in a set of Day-13 twins. One individual is in the uterine body (A), and one is in the middle of the right horn (B); five minutes later both are in the uterine body (C). Sonograms D, E, F show multiple vesicles in the uterine body at Days 11 and 13 (D), Days 12 and 13 (E), and Days 14 to 16 (F). The quadruplets (F) were from a mare that was superovulated with a pituitary extract. Twins: Origin and Development 299 Example of the sequential locations of each embryonic vesicle in a twin set during the mobility phase (17). Location determinations were made every five minutes for two hours. The solid arrows are for one vesicle, and the open ows are for the other vesicle. The numbers at each position are the number of nutes a vesicle spent in each segment. The series starts at the white star. Initially, » two vesicles moved together, spending five minutes in the left horn/middle segment, five minutes in the anterior body, and 20 minutes in the right 1orn/posterior segment. One vesicle then returned to the anterior body and the other sicle remained in the posterior right horn for 25 additional minutes. Note that one bryonic vesicle was traversing the left horn at the time the other was traversing e right horn. This example illustrates that sometimes members of a twin set will ove independently. ba) je | “+ = 300 Chapter 16 Embryo 2 ee ee LP ea 15 20,25 30 °——_ —_—__ > e=//¢ RP | | | | | | | | | | | | | | Embryo 1 | | | | | | | | | | | | | | 20,25 15 0 e€: nana! mee: Tr a mec esses ec ee ne cmc [[=:=:@ So, LP "mm, 30 “=: Dde 35, 40 oo ae oe 55 60.65, 70 ms Visa Smee eee Orr rim eye a 0 10 20>. 30. 240 50 60 70 80 90 100 110 mm. from cervix <—$$$___—_—__ UTERINE BODY <> Diagrammatic profiles showing an example of the movement of twi embryos in the uterine body (17). The cranial end of the cervix was used as reference point. The first location determination was made at time 0 and sequenti determinations were made every five minutes thereafter. At time 0, both embry: were together at LP (left horn/posterior segment); at five minutes, embryo 2 w still at LP, and embryo | was in the uterine body 90 mm from the cervix, and so on. The mobility phenomenon must be well understood by the ultrasonographer in t! search for twin embryos and in the manual elimination of one vesicle during the mobility phase. The mobility patterns and time of occurrence of mobility of eac member of a twin set are similar to those of singletons. That is, the vesicles are mobile from the day of first detection by ultrasound (Day 9 or 10), but the extent of mobility is greatest over Days 11 to 14 or 15 (Section 14.1). Mobility ceases on approximately Day 15 (ponies) or 16 (horses). During the extensive mobility phase, the vesicle travels throughout the uterine lumen. In one study (17), twin vesicles moved from one horn to the other an average of 0.9 times per two-hour trial (equivalent to 11 times per day). The smaller vesicles (3 to 10 mm) did not move as much as the larger (>10 mm) and spent more time in the uterine body. This difference in rate of mobility is probably related to the more sluggish mobility of Day 9 or 10 singletons. There is one known exception to the similarities of mobility patterns of twins versus singletons: The number of times both members of a twin set Twins: Origin and Development 301 are located in the same uterine segment is greater than what one would expect to occur by chance. In one study (17), the frequency of appearance of both embryos of 1 twin set in the same segment was significantly greater than expected (28% versus 8%). This result indicates that sometimes (36% of the time) twin vesicles move ogether rather than independently. a icf NARA Ni Smaller Larger Item embryo embryo Singletons Eman oe ee Number of times vesicle was in uterine body/2-hour trial 11.2442.3 3.45 +0.8 5.45+1.0 Number of location changes/2 hours 5.44 +1.0 6.52b 40.9 8.65 +0.6 a a Differences in mobility patterns among singletons and the smaller and arger member of a twin set (n=13) (17). Data are from two-hour mobility rials (one location determination every five minutes) on Days 11 to 14. The uterus vas divided into seven segments (three for each horn and one for the body). The maller vesicle spent more time in the uterine body than did the larger vesicle or single vesicles. The smaller vesicle of a twin set made fewer location changes than d singletons. The preference for the body by the smaller member of twin sets scemed due more to vesicle diameter than to day. These data further indicate that cach vesicle plays an independent role in its mobility or responds differently to u‘erine contractions. 302 Chapter 16 16.4B Fixation Pee ge et es Successive Number of embryos Number of embryos in examinations on days: in same location different locations ee 11 and 12 10 (56%) 8 (44%) 12 and 13 12 (43%) 16 (57%) 13 and 14 10 (34%) 19 (66%) 14 and 15 12 (41%) 17. (99%) 15 and 16 22 (76%) 7 (24%) 16 and 17 28 (97%) I, (5%) 17 and 18 28 (97%) i tom 18 and 19 29 (100%) 0 (0%) eee ee Change in location of individual embryos between successive dai examinations in mares with multiple embryos. Fixation occurred by Day in 97% of the embryos, based on failure to detect a location change after Day ! Fixation occurred for all embryos by Day 18. The day of fixation is therefc similar for singletons and twins. Fixation pattern Number of mares Both vesicles in one horn (unilateral) 23 (70%) One vesicle in each horn (bilateral) 10 (30%) pee Proportion of mares in which twin embryos became fixed unilateral! y versus bilaterally. Significantly more vesicles became fixed unilaterally (70% than bilaterally (30%). The conclusion that unilateral fixation of twins is more common than bilateral fixation is apparently contrary to the experience of practitioners. However, many twins that become fixed in one horn are reduced to a singleton by natural embryo reduction before Day 20 and would not have been detected if the first examination was done after Day 20 (Section 16.5C) (18). In addition, if the first examination is done on Days 17 to 19, unilaterally fixed twins are easily overlooked (Section lB): Twins: Origin and Development 303 Ultrasonogram of twin Day-14 and Day- 16 vesicles. The larger vesicle (Day 16, 25 mm) was fixed in the caudal right horn and the smaller vesicle (13 mm) was mobile within the same horn cranial to the site of fixation of the larger vesicle. The smaller vesicle did not move past the larger fixed vesicle and both vesicles became fixed unilaterally. The preference for unilateral fixation of twins may be due to the first vesicle becoming fixed and acting as an impediment to movement of the other. 16.5 Embryo Reduction \6.5A Time of occurrence Days Phase Terminology sefore Day 11 Before extensive mobility phase Pre-mobility embryo reduction ysllto15 During mobility phase (No embryo reduction detected in 33 mares) ys 16to40 = After fixation and before end Postfixation embryo reduction of embryo stage ‘er Day 40 During fetal stage Fetal reduction Relationships between age of conceptus and type of reduction. The given terminology will be used in this text to describe the processes for biological €| mination of one member of a twin set according to time of occurrence. 304 Chapter 16 16.5B Pre-mobility embryo reduction Day Item q 11 No. mares with: >1 normal-sized embryo 10/14 1/14 Some undersized embryos 1/14 5/14 No. embryos/mare Normal-sized embryos 2.9 £0.41 0.7 40.2 Undersized embryos 0.1 204 dV.3 Size and number of embryos in multiple-ovulating mares on Days and 11 and an example of one normal-sized and three undersiz embryos at Day 11 from induced multiple ovulations (19). Synchronc multiple ovulations were induced with a pituitary extract and hCG. The resv supported the existence of an embryo-reduction mechanism in induced multi ovulating mares, occurring between Days 7 and 11. This was demonstrated by decrease in the number of normal-sized and the increase in the number of undersi’ vesicles between Days 7 and 11. Although the embryos likely were in the uterus Day 7, the possibility that the reduction mechanism was initiated at a pre-uterine cannot be excluded. In this regard, the mechanism appeared to operate in b bilateral and unilateral ovulators, suggesting the mechanism exerted its effect « the blastocysts entered the uterus. Further work is required to more critically this hypothesis. In other studies, multiple embryos collected from induced multip ovulators at Day 7 had a lower success rate per embryo when transferrec ARES i i ll Ne Le to recipients than for single embryos (20). However, naturally occurring mult! ple embryos had a high success rate when transferred at Day 7 (21). The postulate of a naturally occurring mechanism for intrauterine embryo reduction before Day 11 requires further study. Indirect rationale for the existence of a form of reduction before the mobility phase is the failure to find reduction during the mobility phase (Section 16.5C), a 64% incidence of postfixation embryo reduction (Section 16.59), and a 6% incidence of fetal reduction (Section 16.5E). Thus, the combined incidence of postfixation and fetal reduction does not seem to adequately account for the wide discrepancy between the double ovulation rate and twinning rate. Twins: Origin and Development 305 9.9C Embryo reduction during Days 11 to 40 a Item Singletons Twins Abortion rate (loss of all embryos) Days 11 to 16 0/36 1/31 Days 17 to 40 4/36 2/30 Total 4/36 (11%) 3/31 (10%) Embryo reduction rate Days 11 to 16 --- 0/28 Days 17 to 40 --- ISLA Total --- 18/28 (64%) oo _. LD Abortion (loss of all embryos) and reduction rates (18). Abortion rate was not different between mares with Singletons and mares with twins. Embryo reduction did not occur during the mobility phase (Days 11 to 15) or on the day of i xation (Day 16) in any of 28 mares. However, reduction occurred in 64% of the mares after fixation and before the end of the embryo stage (Days 17 to 40). ll Oe Time of Number of m with: occurrence of Unilateral Bilateral embryo reduction fixation fixation Total Days 11 to 16 0 (0%) 0 (0%) 0 (0%) Days 17 to 19 9 (47%) 0 (0%) 9 (32%) Days 20 to 29 5 (26%) 0 (0%) 5 (18%) Days 30 to 39 3 (16%) _t__f43%) 4 (4%) Total 17 (89%) 1 (11%) 18 (64%) No reduction 2 (11%) 8 (89%) 10 (36%) See ‘neidence of embryo reduction (18). The incidence of reduction was much greater (P<0.01) for unilateral fixation. Apparently, close apposition of the two vesicles favors reduction. This assumption is also consistent with the absence of tecuction during the mobility phase. 306 Chapter 16 16.5D Nature of unilateral embryo reduction Postfixation embryo reduction (18). Days are given in the lower-rig corners. Days 11-13: Two vesicles in uterine body. Days 17-19: Vesicles are fix in the caudal portion of one horn. The membrane representing the apposed walls the two vesicles is visible only in part (arrows). Days 20-22: The apposed walls 2 more distinct (arrows). Days 26-28, left: Well-defined wall between vesicles (arrow). Embryo in the ventral portion of one vesicle (arrow). Days 26-28, middle: Another view showing extensive membranes which are indicative of twins. The extensive membranes are attributable to the wall between individual vesicles plus a wall within each vesicle separating the yolk and allantoic sacs. Days 26-28, right: Another view showing the second embryo (arrow). Days 29-31: A single view showing a well-developed embryo and a smaller embryo (arrows). Days 34-36: oD Twins: Origin and Development 307 imbryo-reduction has occurred; only one embryo was found (arrow). a = allantoic ac, y = yolk sac. Days 38-40: Normal-appearing remaining embryo suspended rom the dorsal wall by the umbilical cord. y = probably regressing yolk sac. ostfixation embryo reduction (18). Days 1: 13: Two vesicles. Days 14- ind 17-18: Two vesicles. The apposing walls are not visible. Days 19-20 and 31- | The membranes resulting from two adjacent vesicles. Days 33-34: Two cmoryos (arrows). Days 35-36: Only one embryo was found. It is lying on the bottom of the vesicle with the yolk sac beneath it and the allantoic sac above. Days 40: Attachment of umbilical cord (arrow) on ventral aspect of vesicle. The sicle was ballotted so that the embryo would float upwards and expose the ichment of the umbilical cord. Attachment to the ventral hemisphere of the vesicle y indicate disorientation due to the earlier presence of two vesicles in one location. ioe) gS fe =< 308 Chapter 16 . om 5 oe 6 ee ee ee ee ee ee i ; i A 4 TET 7 Ie on 6 oe 8 oe oe ee ee ee | of ¢ eee Ci | Twins: Origin and Development 309 Outcome of multiple embryos (18). The series of images for a given day is isolated by a broken line. RP = right horn, posterior segment; RM = right horn, middle segment; LM = left horn, middle segment; B = body of uterus; LP = left norn, posterior segment. Days 10-14: Five embryonic vesicles in various segments of the uterus. Days 13-17: Three vesicles fixed in the left horn and two fixed in the right horn. Days 16-20: The three vesicles in the left horn have been reduced to two. lhe vesicles which were in the right horn are disappearing so that only a remnant is visible (black arrow). Days 29-33: Two embryos (arrows) in left horn. The smaller vesicle is indented into the larger vesicle and apparently is regressing. Days 31-35: nly some apparent debris of the regressing vesicle is visible to the right of the emaining embryo. Days 34-38: Normal-appearing remaining embryo (arrow). \ttachment is at approximately 3 o'clock and may represent disorientation of the irvivor. The above three photographic sequences of unilateral embryo reduction illustrate ‘wo characteristics of reduction. First, the survivor is sometimes spatially cisoriented in that its umbilical cord may attach toward the ventral aspect of the uterine wall. As described in Section 13.12, the umbilical cord normally attaches corsally. In 4/18 fetuses that remained after embryo reduction of a unilateral twin set, the umbilical cord was attached in the ventral hemisphere of the vesicle, as shown this example. In comparison, the umbilical cord was attached in the dorsal nemisphere in all of 16 mares with singletons and all of 16 fetuses in mares with bilateral twins. In addition, in 8/18 fetuses that originated from unilateral reduction, ‘ie attachment of the umbilical cord was more ventrally located than in any of the 16 vilateral fetuses. Second, the survivor seems unaffected. This aspect is further ‘uStrated in the following table: No. Height of Length of Item vesicles vesicle (mm) embryo (mm) Singletons 32 S521 9 22311 Bilateral twins 16 54 42.2 20 +1.6 Surviving conceptus after embryo reduction 18 54 42.4 20 +0.8 Effect of unilateral embryo reduction on the size of the surviving vesicle and embryo at Day 40 (18). There were no significant differences 310 Chapter 16 among groups. The surviving conceptus was not affected by the reduction process based on its size and appearance. In addition, the incidence of abortion (loss of bot! embryos) was not greater than for singletons (Section 16.5C). If simple two-wa competition or overcrowding was involved in reduction, one would expect a higher incidence of loss of both embryos or a detectable effect on the survivor. Yolk sac wall 2 layers 3 layers and embryo A hypothesis for the mechanism of embryo reduction in unilateral. fixed twins. On the day of fixation (Day 16; illustrated on the left), the two vesicles are in contact, but reduction has not begun. Over the next few days, a singlet would become oriented due to disproportionate thickening of the dorsal uterine w and the massaging action of uterine contractions, as described in Section 14.6. When twin vesicles are present, the same factors (contractions and disproportiona‘e thickening) cause one or both of the vesicles to rotate until one vesicle is located e e —_ adjacent to the thin two-walled portion of the other. Due to the pressure from (1 tense uterine wall, the vesicle that is destined to be eliminated is compressed into f! thin-walled portion of the vesicle that is destined to survive. Death occurs as a result of the reduced surface area between the vesicle wall and the endometrium. The survivor is unaffected because the two-walled portion of the yolk sac is not crucial to the exchange of nutrients, and this area is only temporally blocked while the dying vesicle is becoming invaginated into the survivor. The survivor may be disoriented because the presence of two vesicles during this time interferes with the orientation process. The two-walled portion of the vesicle and eventually the umbilical cord of the survivor, therefore, may be located in either the dorsal portion of the vesicle, as illustrated, or the ventral portion. If this hypothesis is correct, it is expected that the smaller vesicle (twins are usually asynchronous and disparate in size) more likely Twins: Origin and Development 311 would be compressed into the larger. This hypothesis is currently untested. Rationale include the following: 1) In all of three available photographed sequences ‘shown in Section 16.5D), the embryo that was eliminated was located in the area of he eventual attachment of the umbilical cord of the survivor. The umbilical cord of | fetus reaches the allantoic sac in the area that was the two-walled portion of the yolk ac (Section 13.12). The umbilical attachment thus serves as a marker that indicates 1¢ orientation of the embryonic vesicle during the yolk-sac Stage. 2) There is a high icidence of embryo reduction on Days 17 to 20. During that time, there is onsiderable disproportionate hypertrophy of the uterine wall with a pressing gether of the two vesicles , much massaging action of the uterus on the vesicles, and relatively large proportion of the yolk sac wall consisting of only two cell layers. Disorientation of the survivor is a common occurrence with unilateral embryo duction. Unilateral twin fetuses at Day 48 (no embryo reduction; 18). The three ‘ges are three views of the same area taken by slight rotation of the transducer. {: One fetus (arrow) lying on bottom of allantoic sac. Middle: Membrane Ow) representing the apposed walls of the two vesicles. Right: Second fetus ow). Note the similarity between the two conceptuses in size and appearance. ‘Te is no indication that embryo reduction is underway. 312 Chapter 16 16.5E Fetal reduction Outcome Incidence a et Fetal abortion (loss of both fetuses) 10/16 (63%) Birth of two foals 5/16 (31%) Fetal reduction 1/16 (6%) Outcome of the presence of twin fetuses on Days 40 to 42 (22). The twin sets were diagnosed by rectal palpation. Note the relatively high rate o: abortion compared to fetal reduction. As noted in Section 16.5C, embryo reductio is common during the embryo stage, but loss of both embryos is not. It is conclude that during the embryo stage twins are corrected by pre-mobility or post-fixatio: embryo reduction and not by abortion of both embryos; however, twins that are 1c corrected by embryo reduction and enter the fetal stage (>Day 40) intact are m« likely to undergo abortion than fetal reduction or birth of twins. REFERENCES 1. Ginther, O. J., R. H. Douglas, and J. R. Lawrence. 1982. Twinning in mares: A survey of veterinarians and analyses of theriogenology recorcs. Theriogenology 18:333-347. 2. Ginther, O. J., R. H. Douglas, and G. L. Woods. 1982. A biological mechanism for the elimination of excess embryos in mares. Theriogenology 18:475-485 3. Hughes, J. P., G. H. Stabenfeldt, and J. W. Evans. 1972. Clinical and endocrine aspects of the estrous cycle of the mare. Proc. Ann. Conv. Am. Assoc. Equine Pract.: 119-150. 4. Vandeplassche, M., M. Henry, and M. Coryn. 1979. The mature mid-cycle follicle in the mare. J. Reprod. Fert., Suppl. 27:157-162. 5. Henry, M., M. Coryn, and M. Vandeplassche. 1982. Multiple ovulation in the mare. Zbl. Vet. Med. 29:170-184. Twins: Origin and Development 313 6. Ginther, O. J. 1979. Reproductive Biology of the Mare: Basic and Applied Aspects. Equiservices, Garfoot Rd., Cross Plains, WI. 7. Woods, G. L., T. A. Sprinkle, and O. J. Ginther. 1983. Prevention of twin pregnancy in the mare. Proc. Ann. Conf. Am. Therio. Soc., Nashville, TN. 8. Pierson, R. A. and O. J. Ginther. 1985. Ultrasonic evaluation of the preovulatory follicle in the mare. Theriogenology 24:259-268. 9. Wesson, J. A. and O. J. Ginther. 1981. Influence of season and age on reproductive activity in pony mares on the basis of a slaughterhouse survey. J. Anim. Sci. 52:110-129. ). Ginther, O. J. 1982. Twinning in mares: A review of recent studies. J. Equine Ver sch, 2:17 3-145. . Deskur, S. 1985. Twinning in Thoroughbred mares in Poland. Theriogenology 23:711-718. ‘2. Loy, R. G. 1980. Characteristics of postpartum reproduction in mares. Vet. Clin. of No. Amer.: Large Anim. Pract. 2:345-359. Ginther, O. J. 1983. Effect of reproductive status on twinning and on side of ovulation and embryo attachment in mares. Theriogenology 20:383-395. Woods, G. L. and O. J. Ginther. 1982. Ovarian response, pregnancy rate, and incidence of multiple fetuses in mares treated with an equine pituitary extract. J. Reprod. Fert., Suppl. 32:415-421. Jeffcott, L. B. and K. E. Whitwell. 1973. Twinning as a cause of fetal and neonatal loss in the Thoroughbred. J. Comp. Path. 83:91-105. Hughes, J. P. and G. H. Stabenfeldt. 1977. Conception in a mare with an active corpus luteum. J. Amer. Vet. Med. Assoc. 170:733-734. Ginther, O. J. 1984. Mobility of twin embryonic vesicles in mares. Theriogenology 22:83-95. Ginther, O. J. 1984. Postfixation embryo reduction in unilateral and bilateral twins in mares. Theriogenology 22:213-223. 314 Chapter 16 1D; 20. 2ie 22; Woods, G. L. and O. J. Ginther. 1983. Intrauterine embryo reduction in the mare. Theriogenology 20:699-705. Woods, G. L. and O. J. Ginther. 1984. Collection and transfer of multiple embryos in the mare. Theriogenology 21:461-469. Squires, E. L., R. H. Garcia, and O. J. Ginther. 1985. Factors affecting succes: of equine embryo transfer. Equine Vet. J., Suppl. 3:92-95. Ginther, O. J. and R. H. Douglas. 1982. The outcome of twin pregnancies in mares. Theriogenology 18:237-244. Chapter 17 TWINS: MANAGEMENT AND CORRECTION Many clinical uses of diagnostic ultrasonography were described in previous napters. Perhaps no use carries a more satisfying sense of accomplishment than the lanagement and correction of twin embryos. This generalization is especially true 1 veterinarians who work with high-risk breeds, family lines, and individuals. By ‘ie same token, the use of ultrasonography for optimal management of the twinning , yom ct < oblem demands high-quality instrumentation and well-honed interpretive abilities. le operator must be knowledgeable in the fundamentals of ultrasonography and in ‘he ultrasonic anatomy and pathology of the ovaries, uterus, and embryonic vesicle. ore specifically, training or experience must be gained in the recognition of arian follicles and corpora lutea and in the early detection of embryonic vesicles. ie ability to differentiate singletons from twins and to differentiate embryonic sicles from other structures, especially uterine cysts, is essential. In addition, the operator must be proficient in transrectal palpation. Twin detection and correction ‘s an excellent example of the combined use of the two skills --- transrectal imaging nd transrectal palpation. This chapter discusses the management of double follicles and ovulations, ‘luding a review of recent past practices and a suggestion for an ultrasound proach. Information is given that will aid in detecting twin conceptuses during the obility phase and after fixation and in recording the appropriate data on ‘rlogenology records. A manual technique is described for elimination of one -mber of a twin set during the mobility phase. Finally, an ultrasound approach for 6 Management and correction of twins is outlined. 316 Chapter 17 17.1 Management of Double Follicles and Ovulations 17.1A Past practices 1982 survey on twin-prevention practices associated with breeding 1. Incidence: 21/22 (95%) veterinarians utilized a program 2. Methods: Withholding breeding. Recycling. 3. Cost: High (on one farm 17% of estrous periods were lost) 4. Effectiveness: Not known Results of a 1982 survey in the USA on programs to manage doubl« follicles and ovulations (1). Almost all of the breeding farm veterinarians wh responded to the survey modified the breeding program in an effort to prevent th development of twins when double ovulations seemed likely. One approach was t withhold breeding during estrous periods judged to have a high probability of doub! ovulations and then to wait until the next estrus or to shortcycle with an injection of prostaglandin during diestrus. Another approach was to postpone breeding unt after one of the follicles ovulated. These approaches involved consideration of th size of the second-largest follicle and an attempt to judge whether the second follic! would ovulate and whether the two ovulations would be synchronous. As modification, some preferred to continue palpating after an ovulation and the treating with a prostaglandin if a second estrous ovulation occurred, but none of the veterinarians routinely checked for diestrous ovulations. The management restrictions usually were eased as the end of the operational breeding season approached. The economic aspects of a restricted breeding program need to be considered. Such programs could result in failure to establish pregnancy during a breeding season, especially for habitual double ovulators. The tendency for certain mares to double ovulate repeatedly is an exasperating aspect of such programs. In addition, the cost of establishing pregnancy would be increased in proportion to the number of cycles lost. One farm did not breed, because of double follicles, for 24/145 (17%) of the ovulatory estrous periods. Twins: Management and Correction 317 Apparently, the practice of withholding mares from breeding when double ovulations were anticipated seemed such a reasonable approach to the prevention of twins that no test of its usefulness was undertaken. It is understandable, therefore, that the practice became widespread and well engrained, and that information on the outcome of breeding to double follicles lay dormant and unscrutinized in farm records. In conjunction with this survey, two farms which practiced a restricted breeding program were compared with another which bred mares regardless of the number of follicles. The three farms had a similar incidence of twins. ‘7.1B A suggested ultrasound approach Management program for double follicles 1. Breed the mare without regard to the number of anticipated ovulations 2. Administer hCG to favor synchronous double ovulations 3. Determine the length of the interval between ovulations ne Suggested approach for handling mares in which two large follicles are present and double ovulation seems likely. The recent studies reviewed in Section 16.3 do not support the practice of withholding breeding when it is thought ‘iat synchronous double ovulations may occur. To the contrary, the results dicated that synchronous double ovulations provide an improved opportunity to »stablish pregnancy because mares with synchronous double ovulations had a 22% gher pregnancy rate. Apparently when twin embryos developed, often one of them ‘s eliminated before Day 11 by a natural embryo-reduction process (pre-mobility ibryo reduction, Section 16.5). Higher pregnancy rates also occurred in mares with asynchronous ovulations even 1en the ovulations were several days apart. However, the incidence of twin abryos was much greater in association with asynchronous ovulations ection 16.3E). These findings indicate that double synchronous ovulations and esynchronous ovulations that are no more than six days apart are desirable because of ine greatly improved probability of establishing pregnancy. However, asynchronous ovulations are less desirable because of the increased likelihood of developing twin embryos. As described in Section 17.3A, however, ultrasonography has provided an e‘ective method of manually eliminating one member of a twin set. 318 Chapter 17 End point hCG No hCG Number of mares with multiple ovulations 10/13 La Number of multiple ovulators with synchronous ovulations 10/10 —#—~ 3/12 Number of days between first and last ovulations 0.0 £0.0 —-*- 1.6 +0.4 Effect of hCG on synchronization of multiple ovulations in mares superovulated with a pituitary extract. Multiple follicles were induced witl pituitary extract on Days 15 to 19 and hCG was given on Day 20. Ovulations wer defined as synchronous when the first and last ovulations occurred on the same day The proportion of mares with synchronous ovulations was considerably increase and the interval between ovulations was reduced by treatment with hCG Unfortunately, a similar experiment has not been done with naturally occurrin double ovulations. Nevertheless, practitioners may want to consider administratio: of hCG when two large follicles are present. Induction of synchronous doub! ovulations may increase the likelihood that the resulting twin sets will be corrected b natural pre-mobility embryo reduction. In addition, twins originating fro: Synchronous or near synchronous ovulations are more readily manage: (Section 17.2A). Regardless of whether hCG is given, determining the interval between doub! ovulations will be an aid in the later detection of twin embryos. In the studies cited i: Section 16.4, the disparity in embryo diameter was almost always accounted for b the length of the interval between ovulations. For example, if the interval betwee ovulations was three days, the difference in diameter of the two vesicles was usual! equivalent to approximately three days' growth. Knowledge of the interval between ovulations enables the ultrasonographer to select the optimal day for searching for the potential twin embryos while they are still in the mobility phase. As discussed in Section 17.3A, twin embryos are readily detected and can be manually corrected during the mobility phase. Twins: Management and Correction 319 17.2 Detection of Twin Conceptuses 17.2A During mobility phase and on day of fixation S Ovulations Twin search v Vv oT Te TT Teer iar. Te — eee. 0 1 2 3 = 5 6 7 8 S WwW th Bre eG D. 6 Ovulations Twin search v 4 5 6 + & 9 0 i 2 ff eo DAYS Examples of selection of day to begin the search for twins. It is suggested that the search be done when the younger vesicle is 11 or 12 days old. At that time, the younger vesicle usually will be detectable. The older vesicle will not \ J q SI t yet be fixed, unless the interval between ovulations was more than four days. -arching early in the mobility phase allows subsequent searching on a later day, if ‘ne operator is in doubt. In addition, searching early in the mobility phase maximizes ‘1¢ number of opportunities to manually eliminate one of the vesicles. Furthermore, ‘wins are easiest to detect during the mobility phase. Ina study of 20 sets of twins (3), vesicles were identified when they were 11 days old, except for four (10%) which ‘re not found until they were 12 days old. All twins that were detected on Day 11 ‘so were found on Days 12 through 16. There was no instance in which a vesicle was ind on one day, missed the next day, and found again the third day. 320 Chapter 17 Diameter of Vesicle (mm) a Located in Uterine Body (%) Days Mean diameter of individual embryonic vesicles in mares wit multiple vesicles, and percentage that were located in the uterine boc on Days 11 to 16 (3). Mares were primarily Quarter Horses. Number embryonic vesicles is indicated at the top of the figure. Diameters and locations were not significantly different from those of singletons. Note that the vesicles were initially detected primarily in the uterine body on Days 11 and 12 and in a horn on Days 13 to 16. The preference for the uterine body by individual embryonic vesici°s in mares with multiple vesicles was characteristic of vesicles that were 3 to 9 mm ‘1 diameter. Thereafter, the preference for the body declined. The proportion of 3- ‘o 9-mm vesicles that were located in the uterine body was not significantly different among Days 11 to 14 (Day 11, 34/49; Day 12° 31/54; “Day fae ey 32; Day 14, 5/9). Preferential location in the body, therefore, seemed primarily dependent on the size of the vesicle. These data emphasize the importance of searching the full length of the uterine body as well as the horns, especially for the smaller vesicles. Twins: Management and Correction 321 Recording LAI2 BP6 BP Example of recording the location and diameter of twin vesicles. The ‘erus has been divided into nine segments and systematically scanned as described in Section 12.2. It is important to remember that the scanner samples a "slice" of the uterus much like a gross histological specimen (Section 1.1). The thickness of the ultrasound beam and the resulting slice is quite narrow in the focal zone (e.g., 2 mm). /hen twin vesicles are close together, a double "flashing" effect may be noted as the transducer is moved over the vesicles. This effect may not be noted and the twins aissed if the transducer is moved rapidly. A systematic search is important to ensure at every part of the uterine lumen has been examined. A technique for such a arch is described in Section 12.2. A coding system for recording the location of nbryonic vesicles and other structures (i.e., cysts) is an important component of the anual, prefixation, embryo-reduction technique described in Section 17.3A. A ethod of dividing the uterus into segments is depicted above and is described in ore detail in Section 12.2. The diameter of each vesicle is recorded in millimeters, shown. Differentiating between embryonic vesicles and uterine cysts section 12.8A) is especially important in the diagnosis and manual correction f twins. 322 Chapter 17 i semanas seer oivannsoH Lasgeay Min : eye in é as il glia Twin embryonic vesicles during the mobility phase and on the day « fixation. A) Embryonic vesicle (4 mm) in the tip of a uterine horn. This vesic would have been missed if the entire uterus, including the tips of the horns, had r been searched. The bright white spots on the upper and lower surfaces of the ima; of small vesicles (specular echoes) help in the detection and identification proce (Section 5.1). B,C) Twin vesicles on Days 14 and 15 in the middle of the right horn (B) and the caudal portion of the left horn (C). When one vesicle is found, the searci must continue for the second potential vesicle. D,E,F) Twin vesicles in close apposition on Days 14 and 15 (D,E) and Days 16 and 18 (F). When the vesicles are in close apposition, whether mobile or fixed, the individuals are readily identified by the double lobulated appearance. Sometimes, however, various orientations of the transducer must be tried in order to clarify the presence of twins. Note that at this stage the walls in the area of apposition of two vesicles are not usually detectable. This lack of visibility is attributable to the thinness of the walls of the early yolk sac. The wall consists of two layers of cells (ectoderm and endoderm) throughout most of its circumference with no intervening connective tissue (mesoderm; Section 13.6). Twins: Management and Correction 323 17.2B After fixation 324 Chapter 17 Unilateral twin embryonic vesicles after fixation. Also see the sequential! photographs in Section 16.5D. In contrast to the ease of detection betwee: Days 11 and 16, older twins were often difficult to distinguish when fixatio: occurred unilaterally; sometimes they were not differentiated from a singleton. Th walls of apposition between the two vesicles tended to be faint or not discernible o Day 17 and sometimes on Day 18 (sonogram A, Days 17 and 19). An apparently oversized vesicle often provided the only indication of twins at Day 17. On Days 1° to 20, the walls of apposition became much more defined, probably because of the invasion and development of mesoderm in the yolk sac wall (B, Days 22 and 23). The walls of apposition tended to be vertically oriented (A,B), but not always (C, Days 2 and 23). At this stage, the walls of apposition of twins could be confused with the division into yolk and allantoic sacs in a singleton. However, knowledge of the number of days since ovulation and failure to find an embryo proper on the echogenic dividing line are diagnostically important. Sonogram C, for example, h: the appearance of a Day-30 singleton; however, the ovulations occurred 21 and “ days before the examination. The embryo proper of twins sometimes did not become distinguishable un Day 25 or 26. The embryos of singletons usually were identifiable on Day - (Section 13.5), apparently because the images of the vesicles of singletons were le complex. The embryo proper of one member of a twin set is shown for Days 22 a’ 23 (D; arrow); the other embryo is not in this plane of view. On Days 24 to 3\, excess membranes (white lines running across the black vesicular image) were sometimes the only indication that twins were present. An example is shown ior Days 30 and 32 (E). In singletons, the only membrane within the vesicular image «‘ this time results from the division between yolk sac and allantoic sac. The appearance of excess membranes with unilateral twins is attributable to the presence of a wail between each individual vesicle plus a wall within each vesicle separating it into two placental sacs (yolk sac and allantoic sac). Consequently, the ultrasonic anatomy of twin vesicles at this stage is often confusing. The embryo proper of each member of a twin set can be seen in a single plane of a sonogram only occasionally (F; arrows). Usually it is necessary to search much of the vesicular bulge to locate both embryos. Regression of one of the unilateral twin vesicles usually occurred over five to seven days and was characterized by gradual reduction in size. Occasionally the smaller or regressing vesicle was partially surrounded by the larger vesicle. An embryo undergoing reduction is indicated by an arrow in sonograms G (Days 28 and 31), H (Days 32 and 34), and I (Days 32 and 34). In two mares in which the unilateral twin sets were well developed after 40 days (no embryo reduction), Twins: Management and Correction 325 each fetus and vesicle was equivalent in size and appearance. In one of the mares, each fetus was seen only after the transcucer was rotated from side to side; the membrane (apposed walls) separating the two vesicles was distinct. Images of this twin set are shown in Section 16.5D. 17.3. Artificial Correction of Twin Embryos \7.3A During the mobility phase Technique for correcting twins by prefixation manual embryo eduction (5). The two vesicles are located by ultrasonography on Days 11 to 14. f at least one of the two vesicles is in a horn and the vesicles are judged to be ufficiently separated, one of them is ruptured immediately. Rupture of a yolk sac in ne uterine body is difficult because of the larger uterine mass and problems issociated with grasping the body. The rupturing procedure begins by grasping the indicated uterine horn distal (caudal) to the predetermined location of the selected vesicle and proximal to the vesicle that is to be retained. It is most convenient to use the right hand for the left horn and the left hand for the right horn. The horn is 326 Chapter 17 moderately compressed between the thumb and index finger, and the hand is then moved toward the tip of the horn. Movement toward the tip of the horn is important to avoid pushing the vesicle into the inaccessible uterine body. Sometimes (e.g., 6/14 mares) the vesicle. ruptures in place, and sometimes (e.g., 8/14 mares) it moves toward the tip of the horn as the finger and thumb are advanced. This is to be expected because the vesicle is in the mobility phase. Only moderate compression is required to move the vesicle (approximately equivalent to what would be used t palpate endometrial folds during estrus). Sometimes the intact embryonic vesicle cai be felt as the finger and thumb move over it, especially when it is in the middle o cranial portion of ahorn. Vesicles as small as 12 mm (Day 12 or 13) were Teles abine uterine wall is still relatively thin and flaccid at this time, unlike after fixation, so th« mobile structure sometimes can be discerned. With the application of mor compression than what is required to move or feel the vesicle, the vesicle ruptures i) place or after reaching an impediment as it is pushed up the horn. Often (e.g., 8/1. mares) a distinct popping sensation is felt upon rupture. In the ideal situation, th selected vesicle is in the cranial portion of a horn, but even when it is in the caudz portion reduction is relatively easy if one is skilled in rectal palpation techniques The smaller vesicles (9 to 11 mm) are more difficult to rupture and requir considerably more compression than the larger ones (12 to 17 mm). Two 9-mr vesicles required eight and 12 passes over the area before rupture occurred, where the larger structures ruptured immediately or after only a few passes. If a sma vesicle does not rupture after a few attempts, the operator can wait until the next day Usually there were no visible remnants of the ruptured vesicle upon re-examinatio with ultrasound, but occasionally a small, transient dilation was seen. In approximately 50% of the examinations of twin sets, the two vesicles we! well-separated (examples: in opposite horns, anterior and posterior portions of the same horn, a uterine horn and posterior or middle uterine body, anterior body anc middle or anterior portion of a horn). If both vesicles are in the body or if they are judged to be too close for safe, selective rupture, the mare is re-examined later the same day (e.g., 30-60 minutes) or the next day. Ina series of 13 mares, the vesicles were sufficiently separated in six mares at the first examination. In the remaining seven mares, the vesicles separated in an average of 48 minutes in six mares, and in the other mare the twins did not separate adequately during two hours. ae ee Eo —< Twins: Management and Correction 327 —— en sacncliaeae Number Number of embryos per mare of mares Originally Eliminated § Remaining Results eee ii 1 1 0 7/7 became pseudopregnant 5 Zi 1 1 5/5 successes 4 3 1 Z 3/4 successes 1 4 Zs A 1/1 successes a rm Results of manual embryo reduction during the mobility phase (5). sduction was successful in all of four mares which originally had twins, as indicated apparently normal development of the remaining embryo until the last day of xamination (Day 34 to 40). The procedure also was used as part of an ongoing perovulation experiment that required the manual elimination of all but two mbryos in mares that originally had more than two. In four of five such mares, ostfixation development of the remaining two embryos to Day 40 was similar to the evelopment of embryos in contemporary mares with twins. In the one classified as nsuccessful, the remaining embryo developed normally and was not lost until two veeks after the manual correction. Its loss, therefore, may not have been related to ic procedure. In conclusion, rupturing embryos on Days 12 to 14 apparently did ot alter the development of the remaining embryos in five mares which originally ad twins, and in four and perhaps five of five mares which originally had three or yur embryos. Manual embryo reduction during the mobility phase has several advantages: 1) - technique is rapid and minimal compression is required because of the turgidity ' Spherical shape of the vesicle, the thinness of the yolk sac wall, and the relatively ., Compressible uterine walls. The procedure should therefore be atraumatic to uterus. 2) The procedure is done very early in pregnancy. Therefore, if the hod fails, a postfixation procedure can be used or the mare can be recycled with imal loss of time. 3) Other treatments, such as administration of an “prostaglandin are not necessary. 4) Although field trials with large numbers of “es are required, the success rate of the experimental trial was encouraging. 5) chaps most important, the unwanted embryo is eliminated before the twins can come fixed in one horn, making manual reduction more difficult. 328 Chapter 17 17.3B After fixation Day of manual Ref reduction 15 to 18 23 to 31 before 38 32 to 45 30 to 49 Ney oS) ee) SS) Published results on the effectiveness of manual elimination of o1 embryonic vesicle by rupture. The twins were located bilaterally. embryonal bulge was ruptured per rectum by compression between thumb and fin; or between the hand and the mare's pelvis. In some of the studies, a prostaglanc inhibitor, tranquilizer, or an anticholinergic agent was used with the rupturi procedure. Note that the procedure was most effective when done early (by Day 3 An advantage of postfixation reduction is that it can be done with bilateral tw a as UC CONS Tt al Day determined Result 30-34 4/5 = (80%) Term 33/47 (70%) Term 12/40 (30%) Term 31/134 (23%) Term WG. uw (17) without using an ultrasound scanner. However, ultrasound capability allows not on’y manual correction during the mobility phase, but also earlier use of postfixati embryo reduction. In addition, scanners permit diagnosis of unilaterally fixed twins. Available information indicates that manual elimination of an embryonic vesi: should be done as early in gestation as possible. Other methods for managing diagnosed twin embryos include aborting both conceptuses, draining one vesicle with a needle through the vaginal fornix, injecting a destructive agent, and surgically removing one conceptus. These methods are reviewed elsewhere (10,11) and seem less desirable than early manual rupture. Restricting the diet also has been reported as a method of converting a twin pregnancy to a singleton (12), but further study of this promising approach is needed. Regardless of the method, it is important to remember that if intervention results in loss of both conceptuses after the formation of the endometrial cups (approximately Day 36), the resulting production of chorionic gonadotropin (eCG) may interfere with a rapid return to the cyclic condition. Twins: Management and Correction 329 17.4 Summary of Suggested Ultrasound Approach - Breed mares without regard to the number of follicles. With double ovulations, the chances of establishing pregnancy are much improved (Section 16.3). Furthermore, many of the resulting twin sets are naturally reduced to a singleton before Day 11, especially if the ovulations are synchronous (Section 16.5B). . After the first ovulation, continue to examine periodically for a second ovulation. Determining the time of the second ovulation will be a great aid in the later timely detection of potential twin embryos (Section 17.2A). Begin the twin search early in the mobility phase (e.g., Day 11 or 12 after the first ovulation; Section 17.2A). The twin search should be done even if the ovulations were synchronous. 4. If twins are detected, manually eliminate one during the mobility phase (Section 17.3A). ». If Step 4 was not attempted or fixation occurs before successful completion of Step 4, determine whether fixation is unilateral or bilateral. A. If fixation is bilateral, rupture one vesicle immediately (Days 16 to 18). The earlier the vesicle is ruptured, the greater the chance for success (Section 16.5C). The odds on the occurrence of natural reduction in mares with bilateral fixation are low (Section 16.5C), and the odds on successful manual reduction, if done early, are high (Section 17.3A). B. If fixation is unilateral, consider the probabilities that the problem will be corrected by natural embryo reduction before taking other action. As a guide, many (e.g., 50%) can be expected to be corrected naturally between Days 17 to 20 and most (e.g., 90%) by Day 40 (Section 16.5C). These probabilities are in reference to twins present during the mobility phase. If a decision is made to intervene, an attempt can be made to rupture or to separate and rupture one of the vesicles. The chances for success are not 330 Chapter 17 known, but this approach can be attempted before recycling with 2 prostaglandin. The mental stress associated with waiting for natura! unilateral reduction and the possibility of failure of natural reduction can be avoided by adherence to Steps 2-4. This approach requires a high-quality ultrasound scanner equipped with a hig - resolution, high-quality transducer. The operator must be knowledgeable in the fundamentals of ultrasonography and ultrasonic anatomy and pathology of the ovaries, uterus, and embryonic vesicle. More specifically, training or experience must be gained in the recognition of ovarian follicles and corpora lutea and in the early detection of embryonic vesicles. The ability to differentiate singletons from twins and to differentiate embryonic vesicles from other structures, especia'|) uterine cysts, is essential. The extent of the effort made to prevent twins will be determined by the extent the concern for the problem on a given farm or for an individual mare. 7 program can be used when a maximum effort is indicated and can be simplified modified according to need and economic considerations. For example, if twinni is only a minor problem, Steps 2 to 4 can be omitted. This suggested program has not been subjected to extensive testing. Howev based on the trials described in Section 17.3A, judicious use of this ultrasoui oriented approach --- and especially adherence to Step 4 --- can potentially remove the ominous implications associated with the declaration, "She has twins." REFERENCES 1. Ginther, O. J., R. H. Douglas, and J. R. Lawrence. 1982. Twinning in mares: A survey of veterinarians and analyses of theriogenology records. Theriogenology 18:333-347. 2. Woods, G. L. and O. J. Ginther. 1983. Induction of multiple ovulations during the ovulatory season in mares. Theriogenology 20:347-355. 3. Ginther, O. J. 1984. Mobility of twin embryonic vesicles in mares. Theriogenology 22:83-95. 4. Ginther, O. J. 1984. Postfixation embryo reduction in unilateral and bilateral twins in mares. Theriogenology 22:213-223. 10. Ei; i Twins: Management and Correction 331 . Ginther, O. J. 1983. The twinning problem: From breeding to Day 16. Proc. 29th Ann. Conv. Am. Assoc. Equine Pract. Las Vegas, NV. . Woods, G. L., T. A. Sprinkle, and O. J. Ginther. 1983. Prevention of twin pregnancy in the mare. Proc. Ann. Conf. Am. Therio. Soc., Nashville, TN. . Roberts, C. J. 1982. Termination of twin gestation by blastocyst crush in the brood mare. J. Reprod. Fert., Suppl. 32:447-449. . Pascoe, R. R. 1983. Methods for the treatment of twin pregnancy in the mare. Equine Vet. J. 15:40-42. . Ginther, O. J. and R. H. Douglas. 1982. The outcome of twin pregnancies in mares. Theriogenology 18:237-244. Ginther, O. J. 1982. Twinning in mares: A review of recent studies. J. Equine Vet. Sci. 2:127-135. Zent, W. W. 1983. Use of ultrasound in broodmare practice. Proc. 29th Ann. Conv. Am. Assoc. of Equine Pract., Las Vegas, NV. Merkt, H., S. Jungnickel, and E. Klug. 1982. Reduction of early twin pregnancy to single pregnancy in the mare by dietetic means. J. Reprod. Fert., Suppl. 32:451-452. : | | lamb BIOEFFECTS and SUMMARY Chapter 18 BIOEFFECTS In human medicine, diagnostic ultrasound has been popular for more than 15 years. It is estimated that over 50% of pregnant women in the United States have at least one ultrasound examination during pregnancy (1). Because of the vast exposure in human medicine and the rapidly increasing use in veterinary medicine, the possibilities of adverse effects must be diligently and continuously evaluated. The piological effects of ultrasound were recently (1985) reviewed in a book containing contributions from leading authorities (2). The book is recommended for in-depth -onsideration of this important topic. Evaluations of research reports are published innually by the Bioeffects Committee of the American Institute of Ultrasound in Mledicine (AIUM). These literature reviews are available from the AIUM and are so published in the Journal of Ultrasound in Medicine. The possibility of risk in applying the technique of diagnostic ultrasonography is inder continual scrutiny. Recently, the National Institutes of Health solicited grant sroposals specifically for objective evaluation of the possible harmful effects of the echnique. It is expected, therefore, that there will be an increased output of the esults of investigations on the bioeffects of ultrasound. In evaluating such results, it s important to determine whether the experimental protocol approximates the onditions used in evaluating the mare's reproductive tract. Exposure time and ntensity are important considerations. In the past, many experiments on bioeffects tilized continuous waves rather than the short pulses of a few microseconds that are sed in diagnostic instruments (3). This chapter will consider the philosophy of benefits and risks in arriving at a ecision regarding the use of a product. The cell as a potential target and the iteraction of ultrasound with cells or tissues will be discussed, including the ohenomena of thermal and cavitation mechanisms. Current knowledge on the ninimum ultrasound intensities that produce biological effects will be reviewed. ecent (1985) reports of a possible effect of ultrasound on pregnancy rates in women vill be summarized and observational experiences with mares will be noted. 334 Chapter 18 18.1 The Cell as a Potential Target 012345pm aS Lysosome Nucleus Nucleolus Mitochondrion Centrioles Vacuole Space > — Le betweencells oo" / oe Golgi Endoplasmic Plasma apparatus reticulum membrane Basement membrane Internal structure of a cell. The fine internal workings of cells involve structures and functions that could be sensitive to ultrasound. Theoretical examples include the microfibril cytoskeleton, delicate membranes, electron transport chains in the energy-converting processes of the mitochondria, mechanical and chemica! processes in the formation of amino acids, and small tubules of contractile proteins (centrioles) that pull DNA molecules to opposite poles during cell division (4). The propagation of ultrasound waves in tissue involve alternating areas of compression and decompression of tissue molecules and vibrations of the molecules longitudinally in synchrony with wave frequency. Also, the energy generated by sound absorption #s converted to heat in the tissue. Concerns about the technology's side effects are heightened upon consideration of the delicate internal details of cells together with the responses of tissue to ultrasound waves (vibration of tissue molecules, absorption of heat; 4). Such concerns seem especially warranted when considering rapidly growing, dividing, or highly functional or active cells, such as those in the ovaries (oocytes, follicular cells, luteal cells) and, most notably, in an embryo or fetus. Bioeffects 335 The interaction of ultrasound with cells or tissues includes thermal and cavitation mechanisms (1). However, thermal effects with diagnostic ultrasound are apparently inconsequential; it is believed that temperature elevations would not reach 1° Celsius (1). In this regard diagnostic ultrasound should not be confused with therapeutic ultrasound, which involves selective deep heating resulting from much higher intensities. Cavitation is a non-thermal mechanism involving an interaction with tiny gas bubbles that is capable of disrupting cells and tissues. Cavitation has been demonstrated in in vitro ultrasound experiments, but there have been no reports of cavitation in connection with in vivo diagnostic ultrasound examination (1). Apparently it has not been demonstrated that micron-sized gas bodies, which could serve as cavitation nuclei, exist in the tissues of interest to theriogenologists. There are also concerns that the genetic apparatus of cells could be altered, resulting in changes that would be inherited. Such damage has been reported in experiments using intensities well in excess of those used for diagnostic applications (1). 18.2 Risks versus Benefits Benefits Research results Clinical experience Positive Negative decision decision Consideration of benefits and risks. Professionals use research and clinical findings to weigh potential or known benefits against potential or known risks. In this way, they arrive at a positive or negative decision on using a technology or product. On the benefit side, there is considerable weight favoring a positive decision to use diagnostic ultrasonography as an aid in evaluating the equine female reproductive tract. These benefits were detailed in previous chapters. However, on the risk side, there are unknowns. We cannot prove a negative --- that is, that something undesirable will not happen. The decision is ultimately based on informed judgment. Informed judgment implies an adequate background knowledge plus consideration of new findings on both sides of the scale. 336 Chapter 18 18.3. Minimum Intensities for Biological Effects e 7» 10;000 Biological > E effect me 1000 7 Soe Ww © t 3 100 = E = Oe SS ee 7 ultrasound 1 10 1 1 min min hr day EXPOSURE TIME Minimum ultrasound intensity levels reported to produce measurable biological effects (adapted from 5). Exposure to ultrasound is measured ir units of intensity and duration. Intensity defines the acoustic power passing through a unit area of tissue (5). It is commonly expressed in milliwatts per square centimeter. The graph indicates that the minimum intensity required to produce : measurable effect on tissues is 100 milliwatts/cm?, and that the tissues must be exposed for hours. Profound harmful effects have been reported in laboratory) animals and in in vitro studies at these intensities and durations (5,6). However diagnostic ultrasound scanners emit only 1 to 10 milliwatts/cem?; apparently nc observable effects had been confirmed at these intensities even when tissue was exposed much longer than for diagnostic examinations (6). The Biological Effects Committee of the American Institute of Ultrasound in Medicine stated the following: "In the low megahertz frequency range [0.5-10 MHz], there have been no independently confirmed significant biological effects in mammalian tissues exposed to intensities below 100 milliwatts/cm2” (6). This statement was reaffirmed in October 1982. However, as more sensitive biological end points are used, it is reasonable to expect some reduction of this level. Bioeffects 337 18.4 Recent Reports of Possible Side Effects Cumulative pregnancy rates (%) 2 4 6 8 10 12 14.16 18 20 22 242625 Number of cycles Cumulative pregnancy rates in women exposed to ultrasound for monitoring follicular growth (7). The results of this study apparently were reported subsequent to the 1985 publication of the book, "Biological Effects of Ultrasound," cited in the introduction to this chapter. The study involved retrospective examination of the cumulative pregnancy rates per cycle in women inseminated with donor semen. A comparison was made between a group that was monitored by ultrasound and a group that was not. The duration of the ovarian ultrasound examinations was approximately seven minutes. A 3.5 MHz transducer was used. Both groups were monitored by cervical scores and basal body ‘temperature to estimate imminent ovulation. The number of ovulatory cycles required to establish pregnancy was higher and the pregnancy rate per cycle was 338 Chapter 18 lower in the group examined by ultrasound. Although the study was weakened by its retrospective and non-random nature, the results emphasize the need for further, more critical study. In another project, rats were exposed to ultrasound similar in intensity to that used in the human beings (7). Exposure time was five minutes/day. The number of fetuses per rat was significantly lower on Day 14 in the rats exposed to ultrasound (11 + 0.7, n=22) than in controls (14 + 0.3, n=25). In a review of these and other recent experiments (8), a report was cited in which mouse oocytes were exposed to ultrasound in vitro. The ova were fertilized, cultured to the blastocyst stage, and transplanted. Development to the blastocyst stage was not altered, but the frequency of embryonic loss after transplantation was significantly greater in the ultrasound- treated group. The reviewer noted that critical study of the influence of ultrasound on the meiotic process is needed. In mammals, including horses (9), meiotic division is arrested in the fetal ovary and is reinitiated in the preovulatory follicle. Perhaps one of the steps of meiosis in the oocyte just before ovulation is particularly sensitive to ultrasonic waves. The reviewer (8) did not believe that the recent finding: justified the conclusion that ultrasound is innocuous to oocytes during the periovulatory period. Clearly, however, further study is needed. These recen reports emphasize the need for veterinary ultrasonographers to be continually aware of and receptive to new findings. In another retrospective study (10), premature ovulation in women was attribute: to ultrasound examinations. In patients that were not exposed to ultrasound withi: three days of the ovulatory stimulus, ovulation was not observed in any patien between 30 and 37 hours after the ovulatory stimulus (n=23). In a group that wa: examined by ultrasound between the ovulatory stimulus and ovulation, ovulatior occurred in 25 to 36 hours after the ovulatory stimulus in 42% of the patient (n=19). This report was published in 1982 and apparently has not been confirmed Sham controls will be needed in future studies to partition direct ultrasound effect from other variables associated with the examinations and to ensure that patients are randomly assigned to groups. 18.5 Bioeffects in Mares After decades of use, there has been no known instance of injury in humans due to ultrasound, with the possible exception of the recent work involving an effect of Bioeffects 339 ultrasound examination of the preovulatory follicle on fertility in women (Section 18.4). However, the routine transabdominal examination of the reproductive tract in humans differs profoundly from that described herein for transrectal evaluation of the reproductive tract in horses. First, in horses higher frequency transducers and fewer intervening tissues are involved because of the intrarectal placement of the transducer. Second, embryos in horses are detected and monitored at much earlier stages than in humans. Indeed, for these reasons equine embryos would probably serve well as experimental models for in vivo evaluation of the bioeffects of diagnostic ultrasound. Apparently, only one study has examined the safety of using ultrasound to diagnose pregnancy in mares. Colorado workers (11) compared the effects of transrectal palpation to the effects of transrectal ultrasound scanning using a 3.0 MHz transducer on days 15 and 20. There were no significant differences, as shown in the following table: Technique used Palpation plus Palpation probing with an Palpation and Item alone inactive transducer ultrasound No. of mares pregnant 17/30 16/30 20/30 Pregnancy rate 56% 53% 67% The pregnant mares were examined also on days 25, 30, 35, 40, and 50. The embryonic loss rate seemed similar to what has been reported by others who did not use ultrasound. Several research laboratories and many practitioners have examined thousands of mares by ultrasonography to detect and evaluate embryos; no adverse consequences have been reported. In our laboratory, the embryos of several hundred mares have been examined daily from Day 11 to Day 40 with no apparent effect. Since the ultrasonic morphology of embryos was being studied and photographed in detail, the exposure time per examination was much longer than for simple pregnancy diagnosis. The dynamics of embryo-uterine interactions have been studied in more than 100 mares. These studies involved locating the embryo every five minutes for two hours each day on Days 11 to 16. In some mares, the embryo was monitored almost continuously for 15 minutes to one hour. None of these experimental trials produced 340 Chapter 18 known detrimental effects on the embryo in terms of disrupted viability and growth. The ovaries of more than 100 horses have been examined daily for two or three estrous cycles or through Day 40 or 50 of pregnancy. The ovaries were under lengthy scrutiny because all visible follicles were being counted and classified, and the ultrasonic anatomy of corpora lutea was being studied. The estrous cycle lengths, ovulation incidence, time of luteal regression, and pregnancy rate did not seem to be different from data obtained by us and others using other techniques. 118 emphasized that these were secondary observations; bioeffects were not under specific critical study. These experiences are reassuring, but do not negate the need for critical study of the bioeffects of diagnostic ultrasonography when used to transrectally evaluate the reproductive tract in horses. Studies involving prolonged exposure and potentially sensitive end points are needed (e.g., circulating level of ovarian hormones, viability of oocytes, detailed assessments of embryo morphology). Perhaps there is an exposure time at which sensitive changes will be detected. As noted above, however, available clinical and secondary information indicates that such a time will far exceed the duration of exposure involved in clinical evaluation. REFERENCES 1. Ziskin, M. C. 1985. Ultrasonic bioeffects and their clinical relevance: an overview. In Biological Effects of Ultrasound. Eds. W. L. Nyborg and M. C. Ziskin, Churchill and Livingston, New York, NY. 2. Nyborg, W. L. and M. C. Ziskin, Eds. 1985. Biological Effects of Ultrasound. Churchill and Livingston, New York, NY. 3. McDicken, W. N. 1981. Diagnostic Ultrasonics; Principles and Use of Instruments. 2nd Ed. John Wiley and Sons, New York, NY. 4. Powis, R. L. and W. J. Powis. 1984. A Thinker's Guide to Ultrasonic Imaging. Urban and Schwarzenberg, Baltimore, MD. 5. Athey, P.A. and L. McClendon. 1983. Diagnostic Ultrasound for Radio- graphers. Multi-Media Publishing, Inc., Denver, CO. 6. Kremkau, F. W. 1984. Diagnostic Ultrasound. Grune and Stratton, Inc., New York City, NY. 10. Lis Bioeffects 341 . Demoulin, A., R. Bologne, J. Hustin, and R. Lambotte. 1°85. Is ultrasound monitoring of follicular growth harmless? In In Vitro Fertilization and Embryo Transfer. Eds. Seppala and Edwards. Annals NY Acad. Sci., pp. 146-152. . Demoulin, A. 1985. Is diagnostic ultrasound safe during the periovulatory period? Res. Reprod. 17:(No. 2, Apr.)1-2. . Ginther, O. J. 1979. Reproductive Biology of the Mare: Basic and Applied Aspects. Equiservices, Garfoot Rd., Cross Plains, WI. Testart, T., A. Thebault, and R. Frydman. 1982. Premature ovulation after ovarian ultrasonography. Br. J. Obstet. Gynaecol. 89:694. Cited in review by Bioeffects Committee of AIUM, J. Ultrasound in Med., May, 1985. Squires, E. L., J. L. Voss, M. D. Villahoz, and R. K. Shideler. 1983. Use of ultrasound in broodmare reproduction. Proc. 29th Ann. Conf. Am. Assoc. Equine Pract., Las Vegas, NV. Chapter 19 SUMMARY NOTE: Numbers in brackets refer to pages with detailed information 19.1 Introduction Gray-scale diagnostic ultrasonography is a method for non-invasive visualization of the internal anatomy of the reproductive organs and their payload --- the conceptus [1]. Transrectal, diagnostic ultrasound uses high-frequency sound waves to produce images of soft tissues of internal organs [2]. Because of the large size of horses, the transducer can be held in the rectum directly over the organs of interest. Electric current is applied to crystals in the transducer, producing vibrations characteristic of the crystals and resulting in sound waves. The operator directs the sound waves through the tissues by moving or varying the angle of the transducer as desired. The short distance from rectal wall to viewing area allows the use of high- frequency scanners that produce images with much detail. The sound beams that pass through the tissues are quite thin (e.g., 2 mm), and a thin "slice" of tissue is sampled. The two-dimensional image seen on the screen is analogous to a histological section [3]. Tissues have different abilities to either propagate or reflect sound waves. The proportion of the sound waves that is reflected is received by the transducer, converted to electric impulses, and displayed as an echo on the ultrasound screen. The characteristics of various tissue interfaces determine what proportion of the sound waves will be reflected. The reflected portion is represented on the ultrasound image by shades of gray, extending from black to white. Liquids (follicular fluid, yolk sac fluid) do not reflect sound waves and are said to be nonechogenic; therefore, the image of a liquid-containing structure appears black on the screen. At the other extreme, dense tissues (cervix, fetal bone) reflect more of the sound waves and appear white on the screen. Such tissues are said to be echogenic. Other tissues are seen in various shades of gray depending upon their echogenicity or ability to 344 Chapter 19 reflect sound waves. Modern ultrasound instruments for examining the equine reproductive tract are B-mode, real-time scanners. B-mode refers to brightness modality, in which the ultrasonic imaging is a two-dimensional display of dots. The brightness of the dots is proportional to the amplitude of the returning echoes. Real-time imaging refers to the "live" or moving display in which the echoes are recorded continuously, and events such as fetal leg movements and heartbeat can be observed as they occur. Some scanners have videotaping capabilities so that the moving images can be preserved. They may be played back through the ultrasound scanner or a television set. The moving image also can be "frozen" to facilitate measurements or photographic reproduction. 19.2 Waves and Echoes The origin of audible sound from a drumhead, traveling of the resulting waves through air, echoing of the waves from a mountain side, reception of the echoes by ear drums, and processing of the ear-drum vibrations by the internal auditory system are comparable to the origin of ultrasound from transducer crystals, traveling of the resulting waves through tissue, echoing of the waves from tissue reflectors, reception of the echoes by the transducer crystals, and processing of the crystal vibrations by the ultrasound scanner [13]. Diagnostic ultrasound originates from crystals that expand and contract when subjected to an electric current and, conversely, produce an electric current when compressed by returning echoes. Crystal expansion causes compression of neighboring tissue molecules, and contraction causes rarefaction similar to the response of air molecules near a drumhead. In this way, the ultrasound waves travel through tissue. The delay between origin of the waves and reception of the echoes is used to determine the distance from the crystals to the reflector. A wavelength is the distance encompassed by an area of compression and the accompanying area of rarefaction and is depicted as a sine wave [16]. Frequency refers to the number of vibrations or oscillations of the sound source (ultrasound crystals) per second. Frequency is measured in hertz (Hz) units. One hertz is one cycle per second and a megahertz (MHz) is one million cycles per second. Ultrasound is defined as any sound with a frequency of more than 20,000 Hz. Frequencies used for transrectal examination of the mare's reproductive tract are 3.0 to 7.5 MHz. Speed or velocity is the time required for a wavelength to pass a given Summary 345 point. In soft biological tissue, speed averages approximately 1540 meters/second; that is, ultrasound waves travel through tissue at approximately 1.5 millimeters in one millionth of a second. Attenuation is defined as progressive weakening of the ultrasound waves as they travel through tissue, thereby limiting the depth of penetration [19]. Body tissue is a complex medium, and ultrasonic waves in tissue undergo modifications. The heterogeneous nature of tissues results in tissue interfaces wherever tissues of different density are in contact. When sound waves cross an interface, a portion is returned to the transducer in the form of a reflection or echo. The magnitude of the difference in acoustic impedance between the tissues on each side of the interface determines how much of the waves will be reflected. Usually only a small amount is reflected, and the remainder is available to interact with other interfaces deeper in the tissue mass. The difference in impedance is sometimes so great that most of the wave is reflected. For this reason, one cannot "see" through a soft tissue and air interface. The tissue-gas interface may result in reflection of more than 99% of the wave. This occurs commonly in ultrasonic examination of the reproductive tract due to air pockets in the intestines [20]. Air is also a troublesome barrier when trapped between the crystals and the tissue being examined. For this reason, a coupler or gel may be applied between the crystals and the skin in preparation for a transabdominal examination. Refraction refers to the bending of the sound waves as they cross tissues or fluids with different acoustic velocities. This is comparable to the bending of a light beam as it passes from air into water. Refraction is a common cause of an artifact in which a distinct shadow appears below the lateral edges of fluid-filled structures. Scatter occurs when a sound wave encounters an interface that is irregular or smaller than the wavelength. Scatter is very important in imparting ultrasonic textures that are characteristic for a given tissue. The crystals are subjected to a short series of electric excitations, resulting in a short series of vibrations. As a result, a well-defined burst or pulse of waves is produced, which travels through the gel coupler and into the medium. The short, pulsed nature of the generated waves allows an interval of quiescence to permit reception of potential echoes by the same crystals. For example, the pulse of waves may be 2 to 3 mm in length and contain four cycles. The frequency of ultrasound crystals is so great that 1,000 pulses, each consisting of three or four cycles, may be emitted per second despite the requirements that there must be a pause between pulses for echo detection [22]. Higher-frequency crystals (e.g., 5.0 MHz) have the same number of waves per pulse, but the pulse is shorter than that of lower- 346 Chapter 19 frequency crystals (e.g., 3.0 MHz). These relationships are the basis for one of the tenets of ultrasound --- higher frequency results in better resolution. 19.3. The Transducer Most of the ultrasound scanners used for transrectal examination of the reproductive tract in horses use linear-array systems [25]. Linear array refers to the side-by-side arrangement of the rectangular piezoelectric crystals along the length of the transducer. The examining field and the two-dimensional image are in the shape of a rectangle. At least one company in the United States markets a sector transducer for intrarectal use in large animals. Sector refers to the pie-shaped examining field and image. Intrarectal transducers are oriented primarily lengthwise with respect to the animal. The image obtained from a sector transducer represents a cross-sectional sample of tissue with respect to the body, and the image from a linear-array transducer represents a longitudinal sample. Linear-array transducers consist of 60 to 120 or more elements (crystals). Two of the current models of veterinary scanners utilize 64 elements [26]. Elements are fired in small clusters beginning at one end, and then the cluster format is moved in one-element increments along the length of the linear array. For example, a cluster of elements 1, 2, 3, 4, 5, 6, and 7 may be fired. After the cluster is activated, the same cluster is used to detect the echoes by allowing suitable time for echo reception. Then the second cluster (elements 2 to 8) is fired by moving one step down the array [27]. The sound field resulting from firing a cluster of elements is termed a beam [29]. Beams are imaginary; they merely outline the three-dimensional path followed by each pulse. A complete sweep of beams across all the elements in the array produces one image or frame. The use of frames permits real-time images; that is, images that move as the structures move. Detection of rapid movements requires rapid frame rates. Typical frame rates are 20 to 30 frames per second. Resolution refers to the ability of an ultrasound pulse to distinguish between two closely spaced reflectors [30]. High-frequency transducers give better resolution, due to the relatively shorter pulses. Shorter pulses are less likely to overlap two neighboring reflecting surfaces. Focusing narrows a portion of the beam profile and thereby increases the amplitude of the echoes from reflectors at a certain depth [31]. Beams may be focused in the thickness plane to give better lateral resolution at a given depth. This is done by using curved crystals (internal focusing) or by placing an acoustic lens beneath the crystals (external focusing). By modifying the element- Summary 347 firing sequence, beams also can be focused in the width plane (electric focusing). With higher frequency transducers, the focal region is closer to the transducer [32]. A 3.0 or 3.5 MHz transducer is more suited to studying the large postpartum uterus or a large fetus by either external or intrarectal placement of the transducer. A 5.0 MHz transducer is more suited to detailed transrectal study of the reproductive tract or early conceptus. 19.4 Signal Processing Echoes from tissue reflectors are received by the piezoelectric crystals of the transducer. The echoes are initially processed by the receiver, where the echo signal is amplified and compensation is made for loss of intensity due to attenuation [36]. The degree of amplification is called gain and is comparable to volume in the control of audible sound. The concept of gain must be thoroughly understood because proper adjustment of the gain controls is one of the most important variables under the continuous control of the ultrasonographer. On most scanners, near-gain, far- gain, and overall gain controls are provided for the near field, far field, and the overall image, respectively. Proper balancing of the controls is needed to provide maximum clarity at various depths and to minimize artifactual responses. The scan converter stores the amplified signals and shows the resulting information on an echo display screen [37]. Recently manufactured veterinary scanners have digital scan converters that utilize the digital system that is used in personal computers. The term B-mode refers to brightness modulation of the dots or pixels on the echo display screen. Each pixel or picture element corresponds to a location in the scan converter memory, which in turn corresponds to the location of a tissue reflector [40]. Echo signals are presented as brightened pixels with the distance from the top of the image to the pixel representing the distance from transducer to tissue reflector. The brightness of a pixel corresponds to the amplitude of that individual echo signal. Tissue reflectors that are very dense reflect all of the sound pulse, resulting in very bright (white) pixels. Reflectors of intermediate density return only a portion of the pulse, resulting in gray pixels [41]. The terms hypoechogenic and hyperechogenic are sometimes used to describe low and high intensities of echogenicity, respectively. Brightness is represented by shades of gray extending from white (very bright or highly echogenic) to black (no discernible echo; nonechogenic). Information in the scan converter is transferred to the echo-display screen [42]. Electrons ("cathode rays") are produced by heating a filament in the electron gun of the oscilloscope and 348 Chapter 19 are focused into a well-defined electron beam. The beam strikes the phosphor in the screen, causing it to emit light. The beam is directed across the screen by electric signals applied to the deflection plates. The pattern of movement of the electron beam across the viewing screen is called a raster scan. The scan pattern begins in the upper left corner of the screen, moves steadily across, snaps back, and then begins another line. The result is a pattern of movement over the screen that forms a sweep or set of horizontal raster lines. As the raster lines move across the screen, the electron beam turns on and off, thereby producing the vertical scanning lines [43]. Therefore, each point at which a raster line crosses a scanning line represents one pixel. The intensity of the electric signals originating from the memory of the scan converter provides the appropriate information for the intensity (gray-scale level) of the oscilloscope's electron beam for each pixel. Thus, the information stored at each address of the converter is transferred to each pixel of the viewing screen as the sweeping beam of electrons passes over the appropriate pixel. Digital systems are very fast, so that recording into the memory, modifying data, and transferring data to the screen can all be done without one process interfering with the other. The quality of engineering in the display system is a source of variation in image quality [45]. Part of the smoothness of the digital television image comes from the use of very small pixels. Some scanners in the veterinary market are undesirable because large pixels give the image a checkered appearance. 19.5 Interpretation Proper interpretation of the echoes on an ultrasound screen is crucial. Interpretation requires knowledge of the relationships between tissues and echoes and the ability to differentiate between true and artifactual responses [49]. There are two types of reflections (specular and nonspecular) which are presented as echoes on the screen and represent tissue structure. Certain tissue formations also cause waves to bend (refract), bounce back and forth or re-echo (reverberate), become weakened (attenuated) or entirely blocked. As a result, distortions appear on the ultrasound image which can be mistaken for normal or pathological structures or changes. These artifactual echoes complicate the interpretation process. Artifacts are especially common during imaging of the reproductive tract because of the many pockets of bowel gas, fluid-filled structures, and the pelvic bone. A specular reflection results when a pulse strikes an interface that is smooth, wider than the pulse, and parallel to the transducer [50]. Usually, only a small Summary 349 portion of the pulse that strikes such an interface is reflected. The major portion of the pulse continues past the interface as a transmitted pulse or beam. If the wall of the opposite side of an encapsulated, fluid-filled structure is smooth, the opposite wall also will act as a specular reflector. The specular reflection appears as a hyperechogenic echo on the viewing screen. Specular echoes are commonly seen on the upper and sometimes lower surfaces of round (Days 10 to 16) embryonic vesicles. Nonspecular or diffuse reflections originate when a pulse strikes a rough interface or one that is narrower than the pulse [52]. When the ultrasound pulse strikes such a surface, the effective interfaces are narrower than the beam, and scatter of echoes occurs. The net result is a displayed pattern, texture, or speckling that may help to identify a given tissue. Gray-scale imaging fully utilizes the scatter phenomenon of nonspecular or diffuse reflections. A shadow artifact is caused by a noticeable decrease or absence of ultrasonic waves due to blockage or deviation of the sound beams [54]. Shadow artifacts appear as a black column beneath a very dense structure (e.g., bone) or beneath the lateral walls of fluid-filled structures. Enhancement or through-transmission artifacts are common in sonograms of the reproductive tract because of the presence of fluid-filled structures (ovarian follicles, cysts, embryonic vesicles). These artifacts result when the ultrasound beams pass through a reflector-free structure (.e., fluid-filled). The beam is not depleted (attenuated ) by echo production while passing through the fluid. Therefore, when the beam emerges from the far side, the amplitude of the pulse is greater than in the tissues on each side. The relatively greater amplitude or strength of the sound beams distal to the fluid-filled structure results in a column of relatively brighter echoes beneath the structure. Reverberation is a process wherein an echo bounces between two strong interfaces until the ultrasound pulses are exhausted by attenuation [58]. The following three distinguishing features help identify reverberation artifacts: 1) they are equidistant, 2) they gradually diminish in intensity, and 3) they are oriented parallel to the reflective interface. If a very reflective interface is involved (soft tissue-gas), none of the pulse is transmitted through the interface [59]. Therefore, an acoustic shadow results and the reverberation echoes on the screen are placed in the shadow. Reverberation artifacts are very common in the pelvic area because of pockets of bowel gas. 350 Chapter 19 19.6 Instrumentation Transducers for intrarectal use must be designed for ease of insertion and manipulation within the rectum and for minimization of trauma [68]. The difference in resolving power between a 3.5 MHz and 5.0 MHz transducer can be appreciated on a practical basis in the detectability of structures in the reproductive tract [71]. Capabilities of 3.5 and 5.0 MHz transducers, respectively are as follows: 1) minimum diameter of detectable follicles, 6 to 8 mm and 2 to 3 mm; 2) detectability of corpus luteum, from Day 0 to Day 5 and from Day 0 to regression; and 3) earliest detection of conceptus, Day 11 (6 to 7 mm) and Day 9 or 10 (3 to4 mm). The distance from the transducer face during transrectal imaging to the center of the ovary or the lumen of a nongravid uterus is only a few centimeters. Therefore, a higher frequency transducer (e.g., 5.0 MHz) with a focal point of 3 or 4 cm is well suited for examination of the reproductive tract in nonpregnant or early pregnant mares. The lower frequency transducers (e.g., 3.5 MHz) are more suited for examining the uterus during late pregnancy or soon after parturition. The console contains the components for coordinating pulse emission from the transducer, processing the signals from the transducer, and displaying the resulting image on a viewing screen [73]. Diagnostic ultrasound scanners therefore contain complex electronic circuitry. Despite the complexity of the console, it should be durable and require minimal servicing. Because the scanners are used in barns, they should be designed to exclude as much dust and dirt as possible and should be readily cleaned. Most veterinary models have near, far, and overall gain controls; a magnification or zoom feature; a freeze button; brightness and contrast controls; electronic calipers; and a keyboard for making annotations on the image to permanently identify videotapes and photographs. Proper adjustments of the gain controls are crucial to building a balanced and pleasing image. The gain adjustments equalize the signal amplitude at various depths [79]. When the amplifier gain is too high, the echoes are too strong and the display is overloaded; conversely, if the gain is too low, the echoes are inadequate. Usually, the gain controls can be adjusted at the beginning of the examination of a number of mares with no adjustment thereafter except for occasional fine-tuning. All controls (brightness, contrast, gains) must be considered together. The adjustment of the brightness control, for example, directly influences the optimal setting for the gain controls. The purchase of an ultrasound scanner is a major investment, so potential buyers are justified in thoroughly researching the purchase [82]. Makes and models of scanners vary widely in capabilities and incidental provisions. All individuals will Summary 351 desire high quality and good service, for example, but may differ in their general needs. This is especially true for clinical versus research requirements. Researchers may want elaborate provisions for annotations, measurements (e.g., area), photography and videotaping, and multiple freeze frame memories. Clinicians may prefer less elaborate schemes, especially if it means a lower price and fewer breakdowns. A scanner should not, however, be selected on the basis of its specifications, alone. For example, a scanner with many elements will provide poor quality if other aspects of the engineering (e.g., damping, electrical insulation) are poor. Good resolving power and penetrating capability are not assured on the basis of transducer frequency. Many other factors are involved. It is emphasized that the most reliable information on quality is obtained from a personal trial under the potential buyer's conditions. An excellent criterion for evaluating the quality of a scanner centers on its ability to consistently image a developing or mature corpus luteum [83]. 19.7 Hard Copy Hard copy is defined as the readable printed copy from a machine, such as a computer. In ultrasonography, the term is commonly used for photographs or videotapes of images [87]. Communication and documentation of results are important aspects of ultrasonography. Specialized instruments and knowledge are required to produce high-quality hard copy, whether a Polaroid image for a mare owner's scrapbook or a set of images for a professional or scientific publication. The preparation of projection slides and videotapes is very useful, if not a requirement, for educational purposes both for the clinician and researcher. Consideration should be given to the following: 1) Polaroid photography. This rapid, but expensive, approach is widely used by veterinarians; 2) negative-based photography, including organization of files and record keeping; 3) preparation of slides for projection; and 4) production of videotapes and preparation of photographs from videotapes. In the production of photographs, a freeze-frame memory on the scanner is used. The quality of the photographs depends upon the quality of the real-time images and the quality of the freeze-frame provision, as well as the quality of photographic instrumentation and technique. 352 Chapter 19 19.8 Techniques Attainment of the maximum capabilities of the instruments requires not only proper adjustments of.the controls, but also good scanning techniques [99]. The operator must develop a thorough and realistic mental impression of the anatomy and orientation of the organs. Orientation is especially important in transrectal imaging of the uterus and ovaries because of their mobility. The scanning techniques described in this text are intended for linear-array transducers. Although modern veterinary ultrasound scanners are portable, they provide much incentive for the development of a centralized examining area on breeding farms [105]. A platform or cabinet can be built immediately next to the rear of the restraining chute. The height of the platform should position the scanner so that the screen and controls are approximately at eye level. The scanner should be close to the chute so that the controls can be adjusted and the details on the screen closely scrutinized by the operator while the transducer is being held in position. Preparing a mare for transrectal ultrasound examination (restraint, evacuation of rectum) is similar to preparing for transrectal palpation. Fecal material can cause distortions on the ultrasound image and must be removed. A pronounced shadow extending from the upper edge of the image may be due to intervening fecal matter. Each operator should develop a systematic procedure, depending on whether selected areas or the entire tract is of interest at any given examination [106]. During the examination, the viewing screen is the center of visual attention. At the same time, the location and orientation of the transducer and the resulting field of view are considered. 19.9 Ovaries Although the embryo has been the focus of attention for ultrasonography in mares, some of the most profound clinical and research applications involve the ovaries [115]. Follicles as small as 2 to 3 mm can be seen, and the corpus luteum can usually be identified throughout its functional life. The rectum is located on the midline between the left and right broad ligaments of the abdominal reproductive organs. Therefore, an intrarectal transducer usually is applied to the medial or dorsal surface of an ovary [116]. The ovaries in the nonpregnant or early pregnant mare may ride on the intestines, and the orientation of the ovaries is quite variable. Because of the extreme ovarian mobility, it is difficult to identify poles and surfaces. For this reason, a large structure (preovulatory follicle, corpus luteum) may be Summary 353 missed due to failure to examine the entire ovary. Ultrasonography is useful, not only for monitoring normal seasonal and cyclic events, but also for diagnosing ovarian irregularities and pathological changes. These include: 1) double ovulation, 2) ovulation failure, 3) quiet ovulation, 4) hemorrhagic follicles, 5) prolonged maintenance of the corpus luteum or pseudopregnancy, 6) ovarian tumors, and 7) cystic peri-ovarian structures [130]. A hemorrhagic anovulatory follicle is a form of apparent ovulatory failure, wherein the preovulatory follicle grows to an unusually large size (e.g., 70-100 mm), fails to ovulate but fills with blood, and gradually recedes. The resulting hematoma occasionally becomes extremely large. The blood becomes increasingly organized, and fibrinous echogenic bands form in the clot. The structures gradually regress and the mare may subsequently return to estrus. Failure of ovulation or anovulatory estrus during the ovulatory season occurs occasionally (incidence, 1-3%). The ability to detect the absence of a corpus luteum by ultrasound makes the condition more subject to diagnosis [131]. 19.10 Follicles The ovarian follicles of mares are excellent subjects for ultrasonic imaging because they are large, filled with fluid, and readily accessible by the transrectal route [133]. Ultrasound provides a rapid, noninvasive, and reliable method of measuring and counting follicles for both clinical and research purposes. Some of the potential clinical applications of ultrasonic examination of the follicles include: 1) helping to determine whether a mare has entered the ovulatory season, 2) aiding in estimating the stage of the estrous cycle, 3) predicting the imminence of Ovulation, 4) detecting double preovulatory-sized follicles that are in close apposition and difficult to discern by palpation, 5) detecting failure of ovulation or anovulatory estrus, 6) monitoring small follicles as an aid in judging whether ovarian sterility or senescence has occurred, 7) evaluating whether a mare is in a condition to respond to treatments for follicular stimulation, and 8) monitoring the results of stimulatory treatments. Research applications center around the ability to sequentially monitor follicular populations, including follicles as small as 2 mm, and the changes in morphology of individual follicles (e.g., shape, thickness of follicular wall). The research potential is especially exciting because small follicles, including those deeply embedded within the ovary, can be morphologically monitored without invading the ovarian tissue. 354 Chapter 19 In recent ultrasound studies, there were significant differences among days for number of follicles 2 to 5 mm, 16 to 20 mm, and >20 mm, but not for the other two categories (6 to 10 mm and 11 to 15 mm) [137]. The number of 2 to 5 mm follicles began to increase just before ovulation. This increase corresponds temporally with the reported time of increase in FSH. The number of large follicles (16 to 20, and >20 mm) began to increase midway between ovulations. At the same time the number of 2 to 5 mm follicles decreased, indicating that small follicles were growing into large follicles. The large follicles decreased in number before ovulation occurred. A divergence in the growth profile between the largest and second largest follicles began after Day 16 (six days before ovulation). This divergence temporally corresponded to the decrease in number of large follicles [138]. Day -6, therefore, was the mean day of selection of the ovulatory follicle to the detriment of all other large follicles. This phenomenon (preovulatory decrease in number of large follicles) can be used to help estimate imminent ovulation and the optimal time to breed. Nonpregnant and pregnant mares did not differ in follicular profiles until the preovulatory growth spurt in the nonpregnant mares [141]. Pregnant mares apparently differ from nonpregnant mares in the lack of a selection mechanism for designating an ovulatory follicle and causing regression of other large follicles. The growth rate of the preovulatory follicle was 3 mm/day for the seven days preceding ovulation. In a recent study, no preovulatory follicles were smaller than 35 mm or larger than 58 mm on the day before ovulation. Eighty-five percent of the preovulatory follicles exhibited a pronounced change in shape from approximately spherical to nonspherical (pear-shaped or conical) at some time during the preovulatory period [145]. However, the change in shape prior to ovulation occurred at various times over the entire preovulatory period [148]. The follicular wall increased in thickness as the follicle grew and the interval to ovulation decreased. Within the constraints imposed by the instruments, the results did not support the hypotheses that changes in gray-scale value of the follicular wall or the echogenicity of the follicular fluid were predictive of impending ovulation. The combination of diameter, shape changes, and thickness of the wall appeared to be valuable for assessing the status of the preovulatory follicle, although no ultrasonically visible reliable predictor of impending ovulation was found. In retrospect, the diameter of the follicle appeared to be as useful for predicting impending ovulation as any of the other criteria. However, size can be measured Summary 355 more accurately by ultrasound than by palpation, thereby minimizing one of the most important variables influencing a decision [152]. The occurrence of ovulation is readily detected by ultrasound by the disappearance of a large follicle that had been present at a recent examination. In addition, the newly forming corpus luteum was visible on Day 0 for all ovulaltions in one study; however, the operator was sometimes aware that a large follicle was present the previous day. Intense echogenicity of the ovulation site was seen on Day 0 in 88% of 55 ovulations. The dramatic ultrasonic changes following rupture and collapse of the follicle can be used to detect or confirm ovulation. For example, the prolonged period of development of large follicles that precedes the first ovulation of the year is a particularly challenging theriogenology problem. During this time, a clinician who is limited to the tactile sense may have doubts as to whether an ovulation has occurred, whereas visual inspection by ultrasonography may provide definitive clarification. 19.11 Luteal Glands Because of the pivotal role of the corpus luteum, its detection provides much information to the diagnostician [155]. The presence and stage of the luteal gland cannot be readily ascertained by rectal palpation, except during the first few days after ovulation. Progesterone assays provide the most useful information for assessing the corpus luteum, but are not convenient for immediate evaluation. One of the major uses of ultrasonography, therefore, involves the immediate detection and evaluation of the luteal gland. In a recent study, 40 mares were examined daily with a 5.0 MHz transducer for a total of 55 interovulatory intervals [157]. Each day's observations were made without knowledge of the previous day's results. The luteal gland was identified for a mean of 17.0 days. The mean interovulatory interval was 21.7 days. The luteal gland was visible in all mares from the day of ovulation until at least halfway through the interovulatory interval. Identification of the corpus luteum throughout diestrus and early pregnancy requires at least a 5.0 MHz transducer. With a 3.5 MHz transducer, the corpus luteum was identified for a mean of only 5.7 days (range, 4-8 days; n=19). Approximately 50% of 55 luteal glands developed a central blood clot and 50% did not [161]. The clots remained throughout the apparent functional life of the gland, but became increasingly organized and reduced in volume. It appears that the corpus hemorrhagicum is not functionally important, because it developed in cece NT 356 #Chapter 19 only one-half of the luteal glands. In addition, the development of a corpus hemorrhagicum did not alter the length of time the luteal structure was ultrasonically visible, the length of the interovulatory interval, or the apparent volume of luteinized tissue. Furthermore, in approximately one-half of the mares with sequential ovulations, a corpus hemorrhagicum formed after one ovulation but not after the other ovulation. Similarly, in approximately one-half of the mares that double ovulated, one ovulation was followed by the formation of a corpus hemorrhagicum and the other was not. These ultrasound observations indicate that the formation of a corpus hemorrhagicum occurs by chance and is not peculiar to certain mares or to certain ovulatory periods. Ultrasound data may be used to help distinguish a cycling mare with a regressed undetectable corpus luteum from an anovulatory-season mare [167]. If the largest follicle is 20 mm or less and a corpus luteum is not found, the mare can be considered anovulatory; in all of 59 cycling mares, the largest follicle was greater than 20 mm on the day the luteal structure disappeared. Pseudopregnancy can be readily diagnosed by ultrasonic demonstration of a mature and persisting corpus luteum and absence of an embryonic vesicle combined in most cases with digital detection of a turgid uterus [168]. Ultrasonography is especially valuable for determining that luteal regression can be safely induced with prostaglandin without fearing an inadvertent induction of abortion. Echogenicity of the luteal tissue can be used to help estimate the age of either a solid or a centrally hemorrhagic corpus luteum during diestrus. The echogenicity is high (hyperechogenic) during luteal development (first few days after ovulation) and again during regression (last few days). The first few days usually can be distinguished from the last few days on the basis of gland size; the echogenicity on the last few days is associated with increased density of the gland due to luteal regression. The echogenicity criterion does not always apply, however. In a series of 55 luteal glands, 12% were not hyperechogenic during development and 36% were not hyperechogenic during regression. The ratio of luteal tissue to blood clot and the degree of organization of the clot can be used to help estimate the age of glands that have central blood clots [170]. The blood clots develop during the first few days and then progressively become more organized and proportionally smaller. The fibrinization or organization of the blood clot is based on the development of a network of fibrinous highly echogenic bands. Summary 357 19.12 Uterus A thorough understanding of the ultrasonic anatomy and dynamic changes in the uterus is essential in the ultrasonic evaluation of the mare's reproductive tract [173]. The dynamic changes visualized with ultrasonography mirror the prevailing ovarian steroid milieu, thereby aiding in estimating reproductive status. The function of the uterus as the site of embryonal and fetal development further emphasizes the importance of the uterus to the ultrasonographer. The uterus also serves as a guide to the location and orientation of the transducer during ultrasound examinations. Only through a working knowledge of uterine anatomy and orientation can the ultrasonographer be assured that the entire uterus has been searched for embryonic vesicles or pathological processes. The orientation of the uterine body is reasonably stable, and the top of the image may be taken as representing the dorsal aspect of the uterus [175]. The uterine horns, however, may float on the viscera and as a result the top of the image may represent the dorso-medial or medial aspect. When the uterus is not floating on viscera, but instead is in a suspended position, the flexure of the caudal uterine horn is oriented down. If the transducer is oriented longitudinally over the caudal aspect of a uterine horn and the flexure is suspended, the flexure may exceed the focal depth and may cause the operator to miss an the embryonic vesicle or a pathological process. For clinical purposes, the operator must be able to estimate the location of the transducer over the surface of the uterus, but the orientation (dorsal, medial, etc.) may be of little practical importance. The uterine body (and cervix) is examined in longitudinal planes as the transducer is being inserted into the rectum. The transducer is moved slightly from side to side so that the entire width of the body is examined. When the corpus-cornual junction is reached, the transducer is moved laterally along one uterine horn, so that the horn is examined in cross-sectional planes. The horn is then re-examined as the transducer is returned toward the midline. The opposite horn is similarly examined. The uterine body is scanned again as the transducer is being removed. The transducer should be moved slowly, especially when searching for small structures. The location of a structure or area of interest (embryonic vesicle, cyst, intraluminal fluid collection) is assigned to one of nine segments [176]. The segment is estimated by moving the transducer from the structure to each end of a horn or the body. A coding system is used to record the type of structure and its location and size. For example, EV12BA could be used to indicate that an embryonic vesicle 12 mm in diameter is located at BA (uterine body, anterior third). Acc nceennen TN 358 Chapter 19 The echogenicity of the cervix is similar during diestrus and pregnancy and is attributable to tissue density [177]. When the cervix becomes flaccid during estrus, it loses its hyperechogenicity and is difficult to differentiate from surrounding tissues. The changing echo texture of the cervix reflects the prevailing levels of estrogens and progestins. During estrus, the echo texture of the uterus is characterized by alternating and intertwining areas of hyperechogenicity and hypoechogenicity [178]. The hyperechogenic areas are attributable to tissue-dense central portions of the endometrial folds. The hypoechogenic areas are attributable to edematous outer portions of the folds. During diestrus, individual endometrial folds are less distinct or not discernible, apparently due to the loss of the layering effect characteristic of estrus [179]. The echo texture is therefore homogeneous. The uterine lumen often is identifiable by a hyperechogenic or white line when the uterine body or a horn is viewed longitudinally. The ultrasonic characteristics of the uterus during early pregnancy are indistinguishable from those of the corresponding days of diestrus. After the embryonic vesicle becomes fixed (Day 15 or 16) the echo texture of the endometrium begins to change, so that by Days 18 to 20 the endometrial folds become prominent. During deep anestrus (anovulatory season) the endometrial folds are not discernible and the echo texture is similar to that of the uterus during diestrus or early pregnancy [180]. However, the cross-sectional views of the horns may be irregular or flat. In comparison, cross sections of the horns during the other reproductive states are round. The ultrasonic anatomy of the uterus is profoundly influenced by the reproductive state (estrus, diestrus, pregnancy, anestrus). These changes can be attributed to the prevailing levels of ovarian steroids and possibly may be influenced also by the products of the conceptus. The echo texture of the uterus and the shape of the cross-sectional views can be used to help determine reproductive status. In a recent study, the ultrasonic profile of the uterus was characterized by scoring (1 = diestrous type, 2 = intermediate, 3 = estrous type) [184]. The scores were low between Days -15 and -9, increased progressively over Days -8 to -3, rapidly decreased between Day -1 and the day of ovulation, continued to decline after Day 0, and reached the low values characteristic of diestrus on Day +2. A decline in scores (ee, O10 2) occurred before ovulation in 65% of the mares. The uterine scores closely paralleled the behavioral scores [185]. When estrous behavior occurred earlier than 15 days before the first ovulation of the year, the uterine scores were consistently low [186]. A uterine score of 2 or more was not obtained for any mare earlier than 15 days before the first ovulation; however, the maximum estrous Summary 359 behavioral score of +6 was obtained on 25% of 186 examinations between Days -48 and -18. Ultrasonic uterine scoring therefore may have some value for distinguishing paradoxical estrus of the anovulatory season from true estrus. Cysts are well-suited to study by ultrasound because they are fluid-filled and nonechogenic [190]. The ultrasonic pathology of the cystic structures is characterized by compartmentalized or multilobulated, nonechogenic images. The cysts are located most frequently on the ventral aspect of the uterus. Cysts in this location frequently bulge into the lumen. Small collections of free uterine fluid likely would not be detected directly by any approach other than ultrasound [191]. They often appeared during late diestrus and tended to recur repeatedly within individuals. They were usually mobile within the lumen, moving occasionally between horns and body. Often the fluid pool or group of pools is elongated and irregular along a portion of the lumen and may be barely discernible (e.g., 3 mm high and 20 mm long). These elongated forms usually are detected in the uterine body. The free collections of fluid may be distinguished from cysts by their mobility, irregular shape, lack of compartmentalization and involvement of the uterine wall, and lack of a surrounding membrane (specular echoes). Large collections of fluid are attributable to pyometra or perhaps occasionally mucometra [192]. Fetal bones in the uterus following death of the fetus may be detected by ultrasound [193]. Shadow artifact may be seen beneath the bone in the sonogram. The shadow indicates that the responsible object is an excellent reflector, which is characteristic of calcified bone. Fetal bones can remain in the uterus for an indefinite time --- more than a year in one mare --- and render the mare infertile until removed. 19.13. The Single Embryo Early pregnancy diagnosis, monitoring the progress of the conceptus, detection of twins, and photographic documentation of pregnancy in mares were the primary initial uses of real-time, B-mode ultrasound scanners in large-animal theriogenology [195]. In a recent study with a 5.0 MHz transducer, the embryonic vesicle was first detected on Days 9 to 11 when it reached 3 to 5 mm in diameter (overall mean, 4 mm) [196]. Sixteen mares, later shown to be pregnant, also were examined on Days 7 and 8, but no vesicles were found. The vesicle was first detected by Day 11 in 98% of the mares. There was no difference between ponies and horses in the day of first detection or in the diameter on the day of detection. Detecting the 360 Chapter 19 embryonic vesicle this early requires at least a 5.0 MHz transducer and a high-quality scanner. The growth profile of the embryonic vesicle was not different between ponies and horses [199]. Expansion increased linearly over Days 11 to 16 (3.4 mm/day). However, the regression lines between Days 16 and 28 were S-shaped, with a distinct plateau during approximately Days 18 to 26. After Day 28, the regression lines were linear with a mean growth rate of 1.8 mm/day. Size of vesicle is not useful for estimating age of the vesicle during Days 18 to 26, and morphological indicators must be used. It is emphasized that the expansion failure is seen when the vesicle is viewed in a cross-sectional plane of the uterine horn. The vesicle begins to conform to the irregularities of the uterine lumen beginning at Day 17, resulting in triangular and irregular shapes [201]. The tenseness of the yolk sac wall or the relative fluid volume of the yolk sac reduces after Day 16, allowing the membranes to conform to the irregular uterine wall, as viewed in cross section, and to elongate slightly within the lumen, as viewed in longitudinal section [202]. The characteristics of both a growth plateau and changes in shape after Day 16 are probably caused by the same factors: 1) increased turgidity of the uterus, 2) disproportional thickening of the dorsal uterine wall, and 3) decreased turgidity of the vesicle. The embryo proper is first detectable with a 5.0 MHz transducer on Days 19 to 24 and only rarely on Day 18 [202]. When first detected, the embryo almost always is in the ventral hemisphere of the embryonic vesicle [202]. The ultrasonic anatomy of the vesicle must be thoroughly understood to attain the full value of the ultrasound technology. On Days 10 to 15, the vesicles are spherical and produce a black circumscribed image resulting from the enclosed collection of yolk sac fluid [207]. A bright white (echogenic) spot often is present on the dorsal surface of the yolk sac image. This spot is due to specular reflections and is an aid in locating the early yolk sac. At Day 18, the cross-sectional shape of the conceptus is irregular and inconsistent [210]. On average, however, the vesicles tend toward a triangular or guitar-pick shape. The apex is oriented dorsally and the smooth, rounded base is oriented ventrally. Thick, encroaching dorsal endometrial folds are responsible for the guitar-pick shape. At Day 22, the structure is primarily yolk sac, but the allantoic sac is emerging and begins to lift the embryo from the floor of the vesicle. On Day 24 the fluid-filled allantoic sac is usually well-defined on the ultrasound image as a nonechogenic area beneath the embryo [213]. The yolk sac and allantoic sac are separated by an echogenic line that represents the apposed walls of the two placental sacs. The embryo is in the form of an echogenic nodule on the Summary 361 separating line. The heartbeats (e.g., 150 beats/minute) can be detected with a 5.0 MHz transducer and high-quality scanner. The allantoic sac enlarges and the yolk sac recedes over Days 24 to 33 so that the embryo and the echogenic line separating the two sacs move toward the dorsal aspect of the embryonic vesicle [215]. The separating line tends toward a horizontal orientation. This is noteworthy because the apposed walls of twin embryonic vesicles tend toward a vertical orientation, thus serving as an aid in differentiating a singleton from unilaterally fixed twins. The Day-40 embryo proper is close to the dorsal pole of the vesicle, and the umbilical cord is beginning to form [219]. A gradual descent of the fetus and associated lengthening of the umbilical cord then occur. The umbilical cord remains attached to the dorsal aspect of the vesicle. Due to lengthening of the umbilical cord, the fetus comes to rest on the bottom of the vesicle. During ballottement of the uterus with the intrarectal transducer, the fetus floats in the allantoic fluid. Age can be estimated (95% accuracy) within +1.5 days for 6 to 23 mm vesicles and +4 days for 27 to 56 mm vesicles [223]. During the S-shaped portion of the curve (Days 16 to 28), the cross-sectional height did not change enough to be an adequate indicator of age. However, profound morphological changes occurred during this time. Useful morphological indicators of day of pregnancy are first detection of the embryo proper at the ventral pole, the proportion of the vesicle that consists of yolk sac versus allantoic sac, and the descent of the fetus as indicated by the length of the umbilical cord. 19.14 Embryo-Uterine Interactions Research utilizing ultrasonic imaging has demonstrated that the equine embryo constantly interacts with the uterus in a dynamic, physical fashion [229]. These interactions are seen in the phenomena of embryo mobility (movement of the embryonic vesicle throughout the length of the uterine lumen many times per day), fixation (cessation of mobility), and orientation (rotation of the embryonic vesicle so that the embryonic pole is on the ventral aspect of the vesicle). In an initial study, mobility already was occurring when the embryonic vesicle was first detected by ultrasonography on Days 9 or 10 [230]. Mobility at this time was limited, but increased between Days 9 and 11 and reached a plateau of maximum 362 Chapter 19 mobility on Days 11 through 14. On Days 9 and 10, the embryo spent 60% of its time in the uterine body. During the maximum mobility phase (Days 11 to 14), the vesicle moved throughout the length of the uterine lumen many times per day. Expansion and contraction of the larger vesicles (Days 13 and 14) during the mobility phase were observed in the uterine body and in the horns when viewed longitudinally [235]. Such contractions were uncommon with smaller vesicles (Days Otol rie contractions occurred in cycles, lasting 5 to 14 seconds. The contraction and expansion phenomenon was associated with ultrasonically observable uterine contractions and to-and-fro movements of the vesicle [235]. The uterine contractions and resulting expansion and contraction and to-and-fro movements of the conceptus were accompanied by a flowing or streaming appearance of the endometrium. The minor to-and-fro movements as well as the major, rapid point-to- point movements gave the vesicle the appearance of a flowing sphere, as though it were in a slow-moving stream of water. This is illusionary, however, because ultrasonic observations indicate that normally the uterine lumen is obliterated and does not contain measurable free fluid. It is postulated that the mobility of the embryonic vesicle permits the small vesicle (3 to 12 mm) to contact all parts of the endometrium to prevent uterine-induced luteolysis. In addition to blocking luteolysis, the traveling vesicle also may distribute substances that are involved in the gradual increase in uterine tone. The increased tone in turn eventually results in the cessation of mobility (fixation) after the blockage of luteolysis is complete. The movement of the embryo to all parts of the uterus may also aid metabolic exchange between the endometrium and yolk sac. Fixation is defined as the cessation of mobility [240]. The mean day of fixation was one day later in horses (Day 16) than in ponies (Day 15), and the vesicle of horses was equivalent to one day larger on the day of fixation. Movement of the embryonic vesicle from one horn to another was not detected in any mare after the day of fixation in several experiments involving daily examinations until Day 40 in more than 100 mares. Fixation occurs almost always in the caudal portion of one of the uterine horns [241]. The caudal portion of a horn contains a flexure or curvature. It is postulated that fixation occurs at this site because it represents the greatest intraluminal impediment to continued mobility of the embryonic vesicle. It is not likely that fixation results from a cessation of uterine contractions; ultrasonic observations indicated that uterine contractions are present for many days after fixation occurs. Fixation is apparently a function of increasing expansion of the embryonic vesicle combined with increasing intraluminal resistance to mobility due to the development of uterine tone [244]. Summary 363 Orientation is defined as rotation of the embryonic vesicle so that the embryo proper is on the ventral aspect of the yolk sac [247]. On Day 14, the vesicle is highly mobile and probably is not oriented. Shortly after the end of the mobility phase, the dorsal uterine wall begins to enlarge and encroaches upon the yolk sac. The encroachment is enhanced by the increasing uterine tone. It is postulated that the disproportional encroachment of the uterine wall plus the massaging action of uterine contractions cause the yolk sac to rotate so that the thickest portion of the yolk sac wall (embryonic pole) assumes a ventral position. 19.15 Embryonic Loss Ultrasonography is a powerful tool for Studying embryonic loss because the embryo can be detected and monitored beginning on Day 9 or 10 [253]. This is fortuitous because Day 10 precedes the time that the embryo must block the uterine luteolytic mechanism. In a recent survey, the incidence of pseudopregnancy was greater for losses during the later periods (after Day 20; 100%) than for losses during Days 11 to 15 (26%) or 15 to 20 (33%) [259]. Maintenance of the corpus luteum during pregnancy and presumably pseudopregnancy is dependent upon blockage of uterine-induced luteolysis by the embryo. The occurrence of pseudopregnancy can be attributed to successful blockage of luteolysis by the embryo or its remnants and the absence of other factors to actively initiate luteolysis (e.g., endometritis). The nature of early embryonic loss was studied in a herd with a high rate (18%) of loss during Days 11 to 15. The average diameter of embryonic vesicles was smaller in mares that lost the embryo than in mares that maintained the embryo [260]. However, most of the vesicles (84%), whether undersized or not, appeared to grow at a normal daily rate until the day of loss. Embryos that eventually were lost were detected more frequently in the uterine body than embryos that were maintained [261]. Small, free collections of fluid in the uterine lumen were detected in this herd at least once during the breeding season in 40 mares during pregnancy examinations (Days 9 to 11) [264]. Mares with embryonic loss during Days 11 to 15 were similar to mares with intrauterine fluid collections in the following ways: 1) pregnancy rate was reduced for the ovulatory periods not associated with loss, 2) mean progesterone concentration on Days 7 and 11 was reduced for the periods associated with loss, 3) mean length of the interovulatory interval was reduced both for periods associated with loss and periods in which an embryo was not detected, and 364 Chapter 19 4) the condition was repeatable within individuals. These similarities suggest that the factors responsible for embryonic loss on Days 11 to 15 in this herd were the same as those responsible for intraluminal collections of fluid. The high rate of embryonic loss occurred during the.time that the embryo must block uterine-induced luteolysis (Days 11 to 15). It is difficult, however, to determine the primary lesion in the complex interdependent events involving the corpus luteum, embryo, and the uterine luteolytic mechanism. The pathogenesis of embryonic loss on Days 11 to 15 could have involved any of the following: 1) failure of the embryo to block uterine- induced luteolysis, 2) primary luteal inadequacy, and 3) uterine pathology. Our preferred hypothesis is that most of the embryonic loss in the mares that did not become pseudopregnant was caused by uterine inflammation. The similarities between mares with embryonic loss and mares with small intrauterine collections of fluid provide the rationale for this hypothesis [265]. Uterine inflammation could account for the reduced mean length of the interovulatory intervals in mares with a history of either embryonic loss or intrauterine fluid collections. The short intervals are attributable to early luteolysis induced by activation of the uterine luteolytic mechanism by an inflammatory process. A chronic uterine pathological process could acccount for the tendency toward repeated embryonic loss and repeated shortening of the estrous cycles. Five of nine vesicles that were lost between Days 15 and 25 in this same herd were undersized during some or all of the preceding examinations [266]. The vesicles that were undersized increased in diameter at an apparently normal rate. The size discrepancy between undersized vesicles and normal-sized vesicles was equivalent to approximately three days' growth. Undersized vesicles were found in 3% of the examinations during Days 11 to 20 in mares that maintained the embryo and in 20% of the examinations in mares that lost the embryo. In retrospect, the probability that undersize indicated eventual loss was 62% [267]. In some of the mares in this group, fixation of the embryonic vesicle did not occur at the expected time. This was attributable to decreased progesterone (perhaps secondary to uterine inflammation) and, as a result, failure of the development of uterine tone. In these mares the vesicle remained mobile until the. mares returned to estrus. The vesicles were lost during estrus. Embryonic death was diagnosed in a mare on Day 38 on the basis of cessation of heartbeat [272]. The volume of placental fluid decreased approximately 75% by Day 46. The dead fetus and the associated small amount of placental fluid were present until Day 74 and occasionally changed locations (e.g., from caudal right horn Summary 365 to uterine body). On Day 75 the mare began to show signs of estrus and only a small collection of fluid was detectable. The mare ovulated on Day 82. The characteristics of loss during the late embryonal and early fetal stages included the following: 1) cessation of heartbeat, 2) dislodgment followed by mobility, 3) gradual decrease in placental fluids, 4) disorganization of placental membranes, and 5) hyperechogenic areas in embryo and membranes [272]. To study the sequence of events associated with death of an embryo, pony mares were ovariectomized or given an injection of PGF2« to induce embryonic loss [273]. The cervix was patent on the day of complete loss of the vesicle in all mares ovariectomized or treated with PGF2a on Day 12 [274]. Expulsion of a viable or dead vesicle through the cervix, therefore, could have occurred in at least some of the mares. The only indication of embryonic death in all mares ovariectomized on Day 12 was the disappearance of the vesicle in a mean of three days without a prior or a subsequent ultrasonically detectable indication. There were no significant differences between the control group and the ovariectomized group in vesicle mobility or growth rate. Natural loss of embryos during Days 11 to 15 also frequently involved disappearance without a trace after apparently normal growth and mobility. In all of four mares treated with PGF2a on Day 12, the vesicles remained mobile until the day of loss on mean Day 19 [277]. In contrast, all of the control vesicles became fixed by Day 15, which is the expected time in ponies. The extended period of mobility in the group treated with PGF2a was characterized by thythmic contraction and expansion of the embryonic vesicle every 4 to 15 seconds. The mobile vesicles were found frequently in the uterine body and sometimes next to the cervix. Similar mobility of the vesicle beyond the expected time occurred in association with natural embryonic loss. Expansion and contraction and intrauterine mobility occur also in normal embryonic vesicles before the day of fixation and probably are caused by uterine contractions. Uterine tone increased between Days 12 and 17 in the control mares, similar to what has been reported. However, tone did not appear to increase beyond the Day-12 level in the group treated with PGF2a, probably because of PGF2a-induced luteolysis and loss of ovarian progesterone. The failure of fixation is attributable to the lack of development of uterine tone; fixation is believed to be caused by increasing uterine tone together with increasing expansion of the embryonic vesicle. In mares treated with PGF2a on Day 21, the vesicles appeared to develop normally until they disappeared on Day 23 [278]. The cervix was patent on the day of loss. In three mares treated with PGF2a on Day 30 and in another mare that was ovariectomized on Day 30, the intact embryonic vesicle was dislodged on Day 31 366 Chapter 19 (two mares) or 32 (two mares) [279]. Dislodgment presumably resulted from a decrease in progesterone and uterine tone. The intact dislodged vesicle was mobile within the uterine lumen and was found frequently in the uterine body. The dislodged and mobile vesicle became elongated, sometimes encompassing much of the length of the uterine body and part of a horn. The heart was beating in two of the four embryos on the day of dislodgment and continued to beat for one and two days, respectively, after dislodgment. A postulated sequence of events following loss of progesterone after fixation of the embryonic vesicle is as follows: The uterine tone decreases due to inadequate progesterone [282]. As a result, the vesicle becomes dislodged. For the first day or so, the embryo of the dislodged vesicle sometimes remains viable, as indicated by heartbeat. The dislodged vesicle moves about inside the uterine lumen, although it probably is not as mobile as a normal vesicle during the mobility phase (Days 11 to 14). The dislodged vesicle spends considerable time in the uterine body, entering a horn occasionally. The placental fluids gradually decrease in volume over many days. The cervix remains closed until the mare returns to estrus sometimes weeks or months (especially if eCG is present) after cessation of heartbeat. Although some of the placental fluids are absorbed, the solid debris is retained until the cervix opens with the return to estrus. Apparently the debris is expelled through the cervix. 19.16 Twins: Origin and Development Ultrasonography is displaying twin sets at much earlier stages and more reliably than by palpation [287]. Because we are seeing more sets of twin embryos than before, our anxiety over twins may be greater. Fortunately, ultrasonography has allowed us to learn much about the natural outcome of twin embryos and has greatly improved our capabilities for handling diagnosed twin sets. In addition, double ovulations, especially those occurring from a single ovary, are more readily diagnosed by ultrasonography than by rectal palpation. Ultrasound scanners are a necessity for effective twin-prevention programs. Their role involves detection of double ovulations and twin embryos, evaluating and monitoring the status of twin embryos, and providing for early manual elimination of one member of a twin set. Breed, individual repeatability, heritability, reproductive status, and age are factors that have been demonstrated to affect the double-ovulation rate [292]. In recent studies, the pregnancy rate was significantly higher for synchronous double ovulations (78%) than for single ovulations (56%) [294]. A similar result was Summary 367 obtained for asynchronous ovulations, except when the interval between ovulatons exceeded four to six days. The proportion of pregnant, double-ovulating mares was not significantly different from what would have been expected if each ovum of the double ovulators had the same chance for development as the ovum in single ovulators. The presence of two large preovulatory follicles can be considered desirable because the probability of establishing pregnancy is much improved. The proportion of mares with twin embryos falls far short of the expected number [295]. The low incidence of externally observed twins, when compared to the incidence of double ovulations, can be attributed to a natural biological process for eliminating an excess embryo. This process has been termed embryo reduction. In studies of breeding-farm records, there was a 0% incidence of detected twins in mares with synchronous double ovulations contrasted with a 16% incidence in mares with asynchronous ovulations [296]. It has therefore been postulated that embryo reduction before pregnancy diagnosis is far more likely to occur with synchronous than with asynchronous double ovulations. The mobility phenomenon must be well understood by the ultrasonographer in the search for twin embryos and in the manual elimination of one vesicle during the mobility phase. The mobility patterns and time of occurrence of mobility of each member of a twin set are similar to those of singletons [300]. That is, the vesicles are mobile from the day of first detection by ultrasound (Day 9 or 10), but the extent of mobility is greatest over Days 11 to 14 or 15. Inone Study, twin vesicles moved from one horn to the other an average of 0.9 times per two-hour trial (equivalent to 11 times per day). The smaller vesicles (3 to 10 mm) did not move as much as the larger (>10 mm) and spent more time in the uterine body. Fixation occurred by Day 16 in 97% of the embryos in one study, based on failure to detect a location change after Day 16 [302]. Fixation occurred for all embryos by Day 18. The day of fixation is therefore similar for singletons and twins. Abortion rate (loss of all embryos) was not different between mares with singletons and mares with twins [305]. Embryo reduction did not occur during the mobility phase (Days 11 to 15) or on the day of fixation (Day 16) in any of 28 mares. However, reduction occurred in 64% of the mares after fixation and before the end of the embryo stage (Days 17 to 40). The incidence of reduction was much greater for unilateral fixation than for bilateral (89% versus 11%) [309]. It was concluded that during the embryo stage, twins are corrected by pre-mobility or post-fixation embryo reduction and not by abortion of both embryos; however, twins that are not corrected by embryo reduction and enter the fetal Stage (>Day 40) intact are more likely to undergo abortion than fetal reduction or birth of twins [312]. 368 Chapter 19 19.17 Twins: Management and Correction Perhaps no use of diagnostic ultrasound carries a more satisfying sense of accomplishment than the management and correction of twin embryos [315]. This generalization is especially true for veterinarians who work with high-risk breeds, family lines, and individuals. By the same token, the use of ultrasonography for optimal management of the twinning problem demands high-quality instrumentation and well-honed interpretive abilities. The operator must be knowledgeable in the fundamentals of ultrasonography and ultrasonic anatomy and pathology of the ovaries, uterus, and embryonic vesicle. More specifically, training or experience must be gained in the recognition of ovarian follicles and corpora lutea and in the early detection of embryonic vesicles. The ability to differentiate singletons from twins and to differentiate embryonic vesicles from other structures, especially uterine cysts, is essential. A suggested ultrasound approach for optimal management of the twinning problem is as follows: 1. Breed mares without regard to the number of follicles [317]. With double ovulations, the chances of establishing pregnancy are much improved. Furthermore, many of the resulting twin sets are naturally reduced to a singleton, especially if the ovulations are synchronous. 2. After the first ovulation, continue to examine periodically for a second ovulation. Determining the time of the second ovulation will be a great aid in the later timely detection of potential twin embryos. 3. Begin the twin search early in the mobility phase (e.g., Day 11 or 12 after the first ovulation) [319]. The twin search should be done even if the ovulations were synchronous. 4. If twins are detected, manually eliminate one during the mobility phase [325]. 5. If Step 4 was not attempted or fixation occurs before successful completion of Step 4, determine whether fixation is unilateral or bilateral. Summary 369 A. If fixation is bilateral, rupture one vesicle immediately (Days 16 to 18). The earlier the vesicle is ruptured, the greater the chance for success [328]. The odds on the occurrence of natural reduction in mares with bilateral fixation are low, and the odds on successful manual reduction, if done early, are high. B. If fixation is unilateral, consider the probabilities that the problem will be corrected by natural embryo reduction before taking other action. As a guide, many (e.g., 50%) can be expected to be corrected naturally between Days 17 to 20 and most (e.g., 90%) by Day 40. These probabilities are in reference to twins present during the mobility phase. If a decision is made to intervene, an attempt can be made to rupture or to separate and rupture one of the vesicles. The chances for success are not known, but this approach can be attempted before recycling with a prostaglandin. The mental stress associated with waiting for natural unilateral reduction and the possibility of failure of natural reduction can be avoided by adherence to Steps 2-4. The extent of the effort made to prevent twins will be determined by the extent of the concern for the problem on a given farm or for an individual mare. The full program can be used when a maximum effort is indicated and can be simplified or modified according to need and economic considerations. For example, if twinning is only a minor problem, Steps 2 to 4 can be omitted. 19.18 Bioeffects Because of the vast exposure in human medicine and the rapidly increasing use in veterinary medicine, the possibilities of adverse effects of ultrasound must be diligently and continuously evaluated [333]. Concerns about the technology's side effects are heightened upon consideration of the delicate internal details of cells together with the responses of tissue to ultrasound waves (vibration of tissue molecules, absorption of heat) [334]. Such concerns seem especially warranted when considering rapidly growing, dividing, or highly functional or active cells, such as those in the ovaries (oocytes, follicular cells, luteal cells) and, most notably, in an embryo or fetus. The interaction of ultrasound with cells or tissues includes thermal and cavitation mechanisms [335]. However, thermal effects with diagnostic ultrasound are apparently inconsequential; it is believed that temperature elevations would not reach 1° Celsius. Cavitation is a non-thermal mechanism 370 Chapter 19 involving an interaction with tiny gas bubbles that is capable of disrupting cells and tissues. Cavitation has been demonstrated in in vitro ultrasound experiments, but there have been no reports of cavitation in connection with in vivo diagnostic ultrasound examinations. Exposure to ultrasound is measured in units of intensity and duration [336]. Intensity defines the acoustic power passing through a unit area of tissue. It is commonly expressed in milliwatts per square centimeter. The minimum intensity required to produce a measurable effect on tissues is 100 milliwatts/cm?, and the tissues must be exposed for hours. Profound harmful effects have been reported in laboratory animals and in in vitro studies at these intensities and durations. However, diagnostic ultrasound scanners emit only 1 to 10 milliwatts/cm2; apparently no observable effects were confirmed at these intensities even when tissue was exposed much longer than for diagnostic examinations. Retrospective results of the cumulative pregnancy rates per cycle in women inseminated with donor semen were compared between a group that was monitored by ultrasound and a group that was not [337]. The number of ovulatory cycles required to establish pregnancy was higher and the pregnancy rate per cycle was lower in the group examined by ultrasound. Although the study was weakened by its retrospective and non-random nature, the results emphasize the need for further, more critical study. Apparently, only one study has examined the safety of using ultrasound to diagnose pregnancy in mares [339]. The effects of transrectal palpation were compared to the effects of transrectal ultrasound scanning using a 3.0 MHz transducer on Days 15 and 20. There were no significant differences. Several research laboratories and many practitioners have examined thousands of mares by ultrasonography to detect and evaluate embryos; no adverse consequences have been reported. In our laboratory, the embryos of several hundred mares have been examined daily from Day 11 to Day 40 with no apparent effect. The ovaries of more than 100 horses have been examined daily for two or three estrous cycles or through Day 40 or 50 of pregnancy [340]. The estrous cycle lengths, ovulation incidence, time of luteal regression, and pregnancy rate did not seem to be different from data obtained by us and others using other techniques. It is emphasized that these were secondary observations; bioeffects were not under specific critical study. These experiences are reassuring, but do not negate the need for critical study of the bioeffects of diagnostic ultrasonography when used to transrectally evaluate the reproductive tract in horses. SUBJECT INDEX Air, as a reflector, 20 Allantoic sac, 208-221 Amniotic sac, 208, 213, 226 A-mode, 10 Amplification, 38, see also gain Amplitude, 16 Analog scan converter, 37 Anechoic, 41 Anestrus, see anovulatory season Angle of impact, 50 Angle of incidence, 50 Annotations, 75, 76 Anovulatory season and folliculogenesis, 126 and hormones, 126 diameter preovulatory follicle, 141 distinguishing from estrous cycle, 167 transitional period, 142 Anovulatory follicles, 130, 131 Artifacts and faulty equipment, 64 and focal zone, 55 beam width, 62 enhancement, 55, 57 general, 54-65 interference, 64 reverberation, 58 shadow, 54, 57 Attenuation, 38 and reverberation artifacts, 58 definition of, 19 Audible sound compared to ultrasound, 17 origin of, 14 propagation of, 15 B-mode, 40 Backscatter, 52, 53 Beam-width artifacts, 62 Beams and real time, 28, 29 and scanning lines, 44 definition of, 28, 29 dimensions of, 29 resolution of, 30 versus frames, 29 versus pulses, 29 Bilaminar omphalopleure, 214, 216 Bioeffects, 333-340 and cells, 334 in mares, 338-340 minimum intensity for, 336 recent reports of, 337, 338 risks versus benefits, 335 Biopsy, 108 Birds, 109 Bits, 39 Black-on-white format, 41 Bladder, urinary, 53, 101 Blastocoele, 205 Blastocyst, 205 Breeding, initiation of, 151 Brightness, 40 adjustment of, 76 examples of controls for, 74 Broad ligaments, 101, 102 Calipers, 76 372 Subject Index examples of controls, 74 Cathode rays, 42 Cats, 109 Cattle, 109, 111 (etvix, 177 Chorionic girdle, 212, 213, 216 Coaxial cable, 68 Console examples of, 73 portability of, 73 Contrast adjustment, 76 examples of controls of, 79 Contrast media, 108 Corpus albicans, 119 Corpus-cornual junction, 106 Corpus hemorrhagicum gross morphology of, 118, 119 incidence of, 161 ultrasonic anatomy of, 159, 160, 163 Corpus luteum, see luteal glands Coupling gel, 20, 103 Crystals, see piezoelecric crystals Damping, 21 Decibel, 16 Diathermy, 20 Diffraction, 27 Diffuse reflection, see nonspecular reflection Digital scan converter, 36, 37, 39 Dislodgment of embryonic vesicle, 272, 281, 282 Disorientation, 249, 250 Dogs, 109 Doppler ultrasound, 10 Double ovulations diameter on day before, 289 effect of age on, 292 effect of hCG on, 318 effect of reproductive status on, 292 effect of season on, 292 factors affecting rate of, 290 past management of, 316 patterns of, 288 repeatability of, 291 suggestion for management of, 317 ultrasonograms of, 289 versus number of embryos, 295 Double ovulations and pregnancy rate, 293-297 with asynchronous ovulatons, 295 with synchronous ovulations, 294 Echo texture, 52 Echoes and piezoelectric crystals, 18 of audible sound, 18 origin of, 18 production of, 33 Echogenic, 41 Ectoderm, 205, 211 Elements, 25, see also piezoelectric crystals Embryo proper first detection of, 202 growth profile of, 202 orientation of, see orientation position changes in vesicle, 212, 214, PMS 20 221 Embryo reduction, 303-311 and type of fixation, 305 effect on survivor, 309 hypothesis for 310 incidence of, 307 nature of, 306-311 postfixation, 305-311 premobility, 304 ultrasonograms of, 306-308 versus abortion, 305 Embryo, see embryonic vesicle and embryo proper Embryonic disc, 205 Embryonic loss, 253-282 and failure of uterine tone development, 269, 270 and fixation failure, 269, 270 and intraluminal fluid collections, 261-265, 268 and location of vesicle, 261 and luteal inadequacy, 264, 265, 274 and mobility of vesicle, 261 and pseudopregnancy, 258, 259 and undersize of visicle, 257, 260, 266, 267 and uterine luteolytic mechanism, 259, 264, 265 before Day 10, 257 dislodgment during, 279, 282 during days 11 to 15, 260-265 during days 15 to 25, 266-270 during days 25 to 40, 271, 272 factors affecting, 256 incidence of, 254 loss of heartbeat during, 272, 279 repeatability of, 256 susceptible periods for, 254, 255 through the cervix, 273, 277, 278, 281 ultrasonic morphology of, 268, 276, 262; 2] 1272.75, Sis 288 Embryonic loss, induction of, 273-282 on Day 12, 273-276 on Day 21,278 on Day 30, 279-282 Embryonic vesicle - capsule of, 206 first detection of, 196 fixation of, 240-246 growth profile of, 198-199 Subject Index 373 interactions with uterus, 229-250 location of, at first detection, 197 longitudinal versus cross sections of, 200 loss of, 253-282 mobility of, 230-239 orientation of, 229, 247 shape of, 200, 210 simulated, 236 singleton, 195-250 twins, 287-330 Embryonic vesicle, ultrasonic anatomy of Days 16-50 (review), 220, 221 Days 17-19, 208-210 Days 20-22, 211, 212 Days 24-25, 212, 213 Days 28-33, 214, 215 Day 36, 216, 217 Days 40-50, 217-219 Days 9-16, 204-207 Enhancement artifacts description of, 55 examples of, 57 Estrous cycle effect of uterine fluid collections on, 263 estimating day of, 169 folliculogenesis during, 120 hormones during, 120 Examining area, centralized, 104 Exocoelom, 205 Expansion and contraction of embryonic vesicle, 235 Eye orbit, 227 Far field, 31, 36, 38, 75 Fetal debris in uterus, 192, 193 Fetal reduction, 312 Fetus, 224-227 ultrasonograms of, 227 374 Subject Index Film types multiformat, 97 negative-based, 92 Polaroid, 89 projection slides, 93 Firing of crystals, 27 even-odd, 28 Fixation of embryonic vesicle and reproductive status, 242 and uterine tone, 243-246 day of occurrence, 240 definition of, 229 site of, 241 Focal depth, 32 Focal point, 32 Focal zone, 32 and artifacts, 54, 55, 63 and transducer selection, 71 Focusing and transducer frequency, 32 dynamic, 32 elecric, 32 external, 31 internal, 31 Follicles, 133-154 double, 134 preovulatory changes, 143, 144, 147 selection of ovulatory, 138 ultrasonic anatomy of, 134, 147 ultrasound applications, 133 Folliculogenesis during estrous cycle, 120, 136 during pregnancy, 120, 140 Format, 41 Frame and raster scan, 42 and real-time, 29 definition of, 29 rate, 29 Freeze frame memory, 80 and photography, 80 examples of controls, 75 Frequency, see also transducer bandwidth, 25 definition of, 16 of transducers, 22 Gain, 37, 38 adjustment of, 79 and artifacts, 59, 60 Gray scale adjustment of, 76 and B-mode, 40 and nonspecular echoes, 53 and scan converters, 37, 39 and shade bars, 77 Hard copy, 37, 87-97, see also photography, video taping Heartbeat, 213 and embryonic loss, 271 Hemorrhagic follicles, 130 History of ultrasound, 7-10 Hydrosalpinx, 131 Impedance, acoustic, 19, 20 and shadowing, 54 definition of, 19 Intensity, 16 Interfaces, tissue, 19, 20 Internal reverberation artifacts, 59 Interpretation, 49-65 Intrarectal examinations, see transrectal Intrauterine fluid collections, 191, 192 and embryonic loss, 261-265 and length of estrous cycle, 263 and pregnancy rate, 262 and progesterone concentrations, 263 ultrasonic anatomy of, 191 Inversion of image, 77 Linear-array transducers arrangement of crystals, 26 definition of, 25 Luteal glands, 155-170 accessory and primary, 123, 166 detectability of, 156, 157 echogenicity of, 163 gray scale of, 165 gross morphogy, 119 transducer frequency to detect, 157 types of morphologies of, 161-165 ultrasonic anatomy of, in vivo, 160- 163 ultrasonic anatomy of, in water bath, 159 ultrasound applications for, 167 Magnification, 45, 75, 82 Matching layer, 21 Mesoderm, 205, 208, 211 Metritis, see uterine inflammation Microcotyledon, 225 M-mode, 10 Mobility of embryonic vesicle, 230-239 and expansion and contraction, 235 and to-and-fro movement, 235 characteristics of, 230 control of, 236 definition of, 229 examples of, 231, 233 progressive nature of, 234 rate of, 232 role of, 238-239 simulated vesicle to study, 237 Mucometra, 192 Multibit word, 39 - Multiformat cameras, 97 Near field, 31, 36, 38, 75 Negative-based photography, 92 Nonechogenic, 41 Subject Index 375 Nonspecular reflections and echoes description of, 52 examples of, 53 Orientation of embryonic vesicle definition of, 229 postulate of mechanism of, 247 Oscilloscope, 42, 43 Ovaries, 115-132, see also components of abnormalities of, 130-132 cortical and medullary areas of, 117 in cattle, 109 orientation of, 101, 102, 106, 116 scanning technique for, 129 shape of, 116 viewed in water bath, 107 Oviducts, 131, 132 Ovulation fossa, 116-119,129 Ovulation, 148, 149 and hormones, 120 breeding before, 150 changes preceding, 143, 144, 147, 150 during pregnancy, 123 predicting, 142 selection mechanism for, 138 superovulation, 152 ultrasonic anatomy, 148, 164 PGF2a, see uterine luteolytic mechanism Phantom, ultrasonic, 70, 78, 84 Photography, see also hard copy multiformat, 97 negative-based, 91-92 Polaroid, 88 projection slides, 93 thermal printers, 97 Picture element, 40, see also pixel Piezoelectric crystals, 16, 21, 26 and echoes, 18 firing of, 27, 28 Pixel, 40, 43, 45 376 Subject Index Polaroid photography, 88 Post processing, 37 Postpartum uterus, 181 Power packs, 81, 105 Pregnancy diagnosis, see specific species Pregnancy, see also embryonic vesicle predicting day of, 222, 223 Preprocessing, 37 Projection slides, 92 Pseudopregnancy and embryonic loss, 258-259 definition of, 168 detection of, 168 incidence of, 259 Pulser, 36 Pulses, production of, 21, 22 Pulsing rate, 36 Pyometra, 191-192 Quality of display screen, 45 of magnification, 45 of transducers, 28 Raster scan, pattern of, 42 Real-time, 29 and frame rate, 29 and raster scan, 42 Receiver, 36, see also gain definition of, 36 Refraction definition of, 20 Resolution, 30 and beam-width artifacts, 62 and transducer frequency, 70 axial, 30 lateral, 30 Reverberation artifacts description of, 58 examples of, 61 - Ribs, fetal, 227 Scan converter, 37 analog, 37 digital, 36, 37, 79 transfer of information of, to screen, 42, 43 Scanner components of, 35-37 selection of, 81 Scanning lines and firing of crystals, 27 and magnification, 45 and placement of echoes, 43 example of, 42 Scatter and nonspecular reflection, 52 and reverberation artifacts, 60 definiton of, 20 Screen, display, 37 and B-mode, 40 and raster scan, 42 scanning lines of, 43 transfer of information to, 42 Sector transducers, definition of, 25 Selecting a scanner, 81 Shade bars, 76, 77 Shadow artifacts description of, 54 examples of, 57 Sheep, 109, 111 Side effects, see bioeffects Simulated structures, 108 Sinus terminalis, 208, 211 Specular reflections and echoes description of, 50 examples of, 51 of embryonic vesicle, 207 of uterine cysts, 189 of uterine lumen, 182 Speed of ultrasound, 16 Superovulation, 152 Swine, 109, 111 Thermal printers, 97 Through-transmission artifact, 54 Time code system, 96 Time gain compensation, 38 Transabdominal imaging mares, 108 other species, 108 Transducer, 23, 24, 68-71 and beam-width artifacts, 62 and coaxial cable, 68 and orientation of image, 46 frequency of, 28 frequency of, to detect corpus luteum, Ss linear-array vs sector, 24, 25, 82 penetration and resolution of, 70 protection and care of, 69 quality of, 28 requirements for, 68 selection of frequency of, 70, 71 Sterilization of, 69 Transmitted beam, 50 Transrectal examination in nonequine species, 109 preparation of mare for, 105 technique of, 106, 129 versus transrectal palpation, 104 Twin conceptuses and association with ovulation pattern, 296 and fixation, 302 and mobility, 298-301 correction of, after fixation, 328 correction of, during mobility phase, 325-327 detection of, 319 diameter of, 320 Subject Index 377 interplay with uterus, 298, 322 location of, 320 management and correction of, 315-330 origin and development of, 287-312 reduction of, 303-312 searching for, 321 summary of management program, 329, 330 ultrasonic anatomy of, after fixation, 306-308, 323 during mobility phase, 298, 322 Ultrasonogram, 37 Umbilical cord, 218, 219 Uterine inflammation and intraluminal fluid collections, 261-265 and uterine luteolytic mechanism, 123, 265 Uterine luteolytic mechanism and PGF2a, 123 and pseudopregnancy, 259 and uterine inflammation, 123 comparison of, with sheep and cattle 238 importance of, to ultrasonographer, 123 nature of, 123 precocious activation of, 123 Uterine tone, 243-246 and fixation, 244 control of, 245,246 Uterus, 173-193 after breeding, 183 after parturition, 181 changes during estrous cycle, 183-187 division into segments, 176 during anestrus, 180 during diestrus, 179 378 Subject Index during estrus, 178 during pregnancy, 179 fetal debris in, 192, 193 fluid collections in lumen of, 191, 192, 261-265, 268 importance to ultrasonographer of, 173 layers of, 181 orientation of, 101, 102, 106, 174 origin of specular echoes of, 182 pathology of, 188-193 recording system for, 176 role of ovarian steroids in changes of, 187 scanning technique for, 176 shape of, 101, 102 tone of, and fixation, 243-246 ultrasonic anatomy of, 178-180, 183-187 Video inverse switch, 41 Videotape, 94, 95 time-code system, 96 Water bath built into transducer, 57 technique, 107 Wavelength, 16 White-on-black format, 41 Yolk sac, 205-221 origin of, 205 Zoom controls, 45, 75 University of Pennsylvania Libraries www.library.upenn.edu Circulation Department vpcircdk@pobox.upenn.ed» /** —eFE-1S56 AMAA 14 7/4LL1X