Real-time ultrasonography for determining pregnancy status and viable fetal numbers in ewes

THERIOGENOLOGY

REAL-TIME

ULTRASONOGRAPHY STATUS AND VIABLE

FOR FETAL

DETERMINING PREGNANCY NUMBERS IN EWES

M. A. Gearhart,’ W. E. lwingfield,’ p. P. Knight,’ J. 9. Smith,’ D. A. Dargatz, J. A. Boon and C. A. Stokes

College

1 Department of Clinical of Veterinary Medicine and

Sciences Biomedical

2

Department of Animal College of Agricultural

Colorado Received

State

University,

for publication: Accepted:

Fort

Sciences

Sciences Sciences Collins,

July 8, May 20,

CO 80523

1987

1988

ABSTRACT The reproductive tracts of 26 estrus synchronized, bred ewes were scanned with a portable 5.0 MHz real-time ultrasound unit within 1 to 6 d postbreeding. Intrarectal scanning was performed on alternate days until Days 28 to 30 and twice weekly until Day 50 of gestation. Transabdominal uterine scans were conducted twice weekly from Dayi 25 to 65 and continued weekly until parturition. A total of 24 ewes (92%) became pregnant. A nonpregnant ewe was recognized 100% of the time by both methods of ultrasonic screening. Correct identification of a gravid ewe as pregnant was 100% from Days 51 to 150 of gestation using transabdominal real-time ultrasonography. There was a significant association (P < 0.005) between the number of lambs born and the number of fetuses observed using transabdominal real-time ultrasonography after Day 25 of gestation. Accurate differentiation of fetal numbers by transabdominal scanning was 100.13% for ewes carrying one lamb and 97.3% for ewes carrying two lambs at Days 51 ts3 75 of gestation. Fetal attrition was documented in one ewe at Day 49 of gestation. Hydrops allantois was diagnosed in another ewe at 110 d of gestation. A total of 37 lambs were born to 23 ewes in the project flock. No congenital abnormalities were noted in any of the lambs. Transabdominal real-time ultrasonography is a safe, rapid, accurate and practical method for assessing pregnancy status, fetal number and fetal viability in sleep. Key

words:

ultrasound,

ovine,

pregnancy

diagnosis,

fetal

number

and

viability

Acknowledgments Funding for this project was provided Experiment Station, # l-54501.

through

a grant

The authors thank Dr. M. D. Salman for assistance with Drs. Z. Weixiong and R. A. Henik and Mr. G. L. Gearhart

AUGUST

1988 VOL. 30 NO. 2

by the Colorado

State

University

the statistical analysis; and for their technical support.

323

l-HERIOGENOLOGY

INTRODUCTION Many techniques for pregnancy diagnosis have been employed in sheep. Harnessed fertile or vasectomized rams, abdominal ballottement, udder scoring, rectal-abdominal palpation with a rod, doppler and A-mode ultrasonography, radiography, celiotomy, laparoscopy, plasma estrogen and progesterone levels, and vaginal biopsy are some of the techniques that have been employed (l-8). Most of the techniques developed for pregnancy diagnosis in ewes have been unsatisfactory due to factors such as expense, low accuracy rates, impracticality, ewe and human safety considerations and long delays in the availability of results. Early diagnosis of pregnancy is of considerable economic value to the sheep industry. Nonpregnant ewes could be sold, reducing feed expenses, while nonpregnant lambs could be marketed at higher prices than they would bring as mature ewes. In addition, differentiation of monotocous and polytocous ewes would allow for their appropriate nutritional management in late gestation. Ewes could be fed according to the number of fetuses they are carrying, thus minimizing prelambing feed costs and utilizing supplemental grain m5re efficiently. Improved control over the nutrition of ewes in late gestation would optimize birth weights, weaning weights and survivability of lambs (9-12). The incidence of dystocia would likewise be reduced. Pregnancy toxemia may be prevented if ewes carrying multiple fetuses were fed according to their actual energy needs (13). Prepartal identification and grouping of ewes according to stage of gestation and number of fetuses being carried would improve lambing management. Errors in selection of range flock replacement ewes that are not truly twins would be reduced. Feedlot operators could minimize reproductive losses and maximize feed efficiency and economic returns if pregnancy status and gestational age were accurately determined in ewe lambs upon arrival at the feedlot. Pregnancy can be detected in the cow as early as 10 d postbreeding using intrarectal real-time ultrasonography (14-16). In the sow, pregnancy status can be accurately determined from Day 22 of gestation by transabdominal real-time scanning from the flank (17). The equine embryonic vesicle can be identified via intrarectra1 real-time ultrasonography as early as 9 d post conception, and the dynamics of vesicle mobility, fixation and orientation within the uterine lumen have been documented (18,19). The objectives period of gestation ewes, 2) differentiate fetal viability and

of this investigation were to determine the optimum method and for using real-time ultrasonography to 1) detect pregnancy in single versus multiple fetuses in pregnant ewes, and 3) assess development in sheep. MATERIALS

Experimental

AND METHODS

Design

Twenty-four Suffolk and tw5 Dorset ewes in good health and condition, and exhibiting normal cyclic estrous activity were used in the study. The sheep represented a cross section of virgin, primiparous and multipargus animals. The ewes were vaccinated against Camovlobacter fetus subsp. w prior to breeding and subsequently against Clostridium oerfrinnens types C and D, 6 wk prior to lambing. Estrus synchronization was accomplished through the use of sustained release aVibrio Fetus Bacterin - Ovine. b Clostridium Perfringens Types City, MO.

Colorado Serum Company, Denver, CO. C & D Toxoid. Coopers Animal Health Inc., Kansas

AUGUST

1988 VOL.

30 NO. 2

THERIOGENOLOGY

progssterone ear implantsC for 7 d, followed by removal of the implants and intramuscular injection of 5 mg of prostaglandin F -alphad (PGF alpha). Thereafter, the ewes were placed in a pen for 7 d with three, ? ertility-tested rams s uffolk fitted with marking harnesses and observed daily for evidence of breeding activity. The date of marking was recorded for each ewe and was considered Day 0 of gestation. From Day 7 until Day 60, the flock was exposed to a different, fertilitytested yearling Suffolk ram whose marking crayon color was changed every 14 d. After Day 60, the ewes were not exposed to rams. The flock was fed 5 lbs of alfalfa hay per ewe per day and free choice trace mineralized salt. Six weeks prior to lambing the ewes were shorn. One pound of grain per head per day was supplementally fed to the pregnant ewes during the last 6 wk of gestation. Within 1 to 6 d postbreeding, the reproductive tracts of all ewes were scanned intrarectally withes portable real-time ultrasound machine using a 5.0 MHz linear Intrarectal scanning was performed on alternate days in each array transducer. ewe until Days 28 to 30, twice weekly until Day 50, then it was discontinued. Intrarectal ultrasound examinations were performed with the ewes restrained in right lateral recumbency in a portable squeeze chute. Cutaneous transabdominal real-time scans using the same 5.0 MHz transducer were initiated between Days 25 to 30 and performed twice weekly until Day 65, then continued weekly until the end of gestation. Transabdominal examinations were conducted with the ewes in a standing position. Both sides of the woolless, caudal ventral abdomen were scanned on each examination, starting at the right flank and proceeding systematically to the left flank area. Carboxymethylcellulosef was used liberally as an ultrasonic coupling gel for intrarectal and cutaneous scanning. From 60 to 120 ml of the water-soluble lubricant was instilled into the rectum of each ewe to stimulate fecal evacuation, displace rectal gas and facilitate introduction of the transducer into the rectum. The results of each ultrasound scan were labeled with identification number, breeding date, scan date and method recorded on videotape, using a portable VHS videocassette anal+sis and reevaluation. From each scan, data regarding number and fetal viability were collected. At parturition, born were compared to prepartal predictions of pregnancy as determined via real-time ultrasonography. Analytical

the ewe’s individual of examination, and recorderg for subsequent pregnancy status, fetal actual numbers of lambs status and fetal number

Method

The Chi-square association between time ultrasonography

test for independency was used to evaluate the statistical the recorded prepartal observations of fetal number using realand the actual outcome at parturition with respect to whether

Fchro-Mate-B. CEVA Laboratories, Inc., Overland Park, KS. Lutalyse. The Upjohn Company, Kalamazoo, MI eTechnicare 210DX. Technicare - Division of Johnson & Johnson FO. Hercules Cellulose Gum. Hercules Inc., Wilmington, gPanasonic NV-8420. Panasonic Company, Division of America, Secaucus, NJ.

AUGUST 1988 VOL. 30 NO. 2

DE. of Matsushita

Company,

Electric

Englewood,

Corporation

325

‘THERIOGENOLOGY

single or twin lambs were born (20). The conditional probabilities of the recorded observations were calculated as percentages to evaluate the likelihood of correctly predicting pregnancy status by using two different applications of real-time ultrasono, graphy (21). These conditional probabilities are analogous to the sensitivity and specificity of a diagnostic test (22). The conditional probabilities used to predict pregnancy status defined and calculated according to the following method: Truly Pregnant

Ultrasound Observation Pregnant Nonpregnant

were

Truly Nonpregnant

C

b d

a+c

b+d

a

+ -

in this study

a+b c+d -

The percentage [a/(a+c)] x 100 was defined as the likelihood of classifying a pregnant ewe as pregnant with ultrasonography (true positive). It was calculated as the [(number of ultrasonographic pregnant observations of pregnant ewes/total number of ultrasonographic pregnant and nonpregnant observations of pregnant ewes) x lOO]. The percentage [c/(a+c)] x 100 was defined as the likelihood of classifying a pregnant ewe as nonpregnant with ultrasonography (false-negative), It was calculated as the [(number of ultrasonographic nonpregnant observations of pregnant ewes/total number of ultrasonographic pregnant and nonpregnant observations of pregnant ewes) x IOO}.

other, equal

It should be noted i.e. the percentage 100.

The conditional defined and calculated

that the above two percentages are complimentary to each of true positives and the percentage of false negatives

probabilities according

Ultrasound observed no. of fetuses

Actual

0

1 2

The percentage

326

used to predict to the following

number

fetal number method:

in this study

were

of lambs

0

1

2

a d g

b

C

;

f i

-

~

-

a+d+g

b+e+h

[e/(b+e+h)]

x 100 was defined

a+b+c d+e+f g+h+i -

c+f+i

as the likelihood

AUGUST

of classifying

1988 VOL.

a

30 NO. 2

THERIOGENOLOGY

pregnant ewe carrying one lamb as pregnant with one fetus using ultrasonography. It was calculated as the [(number of ultrasonographic single fetus observations of ewes carrying one lamb/total number of ultrasonographic observations of zero, single and twin fetuses of ewes carrying one lamb) x 1001. The percentage [i/(c+f+i)] x 100 was defined as the likelihood of classifying a pregnant ewe carrying two lambs as pregnant with two fetuses using ultrasonography. It was calculated as the [(number of ultrasonographic observations of twin fetuses of ewes carrying two lambs/total number of ultrasonographic observations of zero, single and twin fetuses of ewes carrying two lambs) x lOO]. RESULTS Estrus

Synchronization

and

Breeding

All of the ewes were bred by Day 6 following removal of the progesterone implants and 24 out of 26 sheep (92%) became pregnant. Six of the ewes were rebred on subsequent estrus cycles. By comparing parturition dates to marking dams it was determined that two of the remarked ewes settled on their initial mating. Of the four remarked sheep that failed to conceive on their first breeding, two became pregnant on their second mating, while two ewes failed to become pregnant despite free access to a fertile ram for 2 mo. Pregnancy

Detection

Real-time ultrasonographic screening for detection of pregnancy in this study did not mistakenly indicate a nonpregnant ewe as pregnant. Thus the likelihood of recognizing a nonpregnant ewe as nonpregnant by either of the two different applications of ultrasonography was 100%. Alternatively, the likelihood of classifying a nonpregnant ewe as pregnant with real-time ultrasonography was 0%. The earliest that pregnancy was detected postbreeding was at Day 20 using intrarectal real-time ultrasonography (Figure 1). By using intrarectal ultrasonography, the conditional probabilities of correctly identifying a pregnant ewe as pregnant were 12.28% at Days 0 to 25 of gestation and 64.76% at Days 26 to 50 of gestation (Table 1). Table

1.

Percentage applications

likelihood of identifying pregnant ewes using two different of real-time ultrasonography at various lengths of gestation. lntrarectal

ultrasonography % (-Iv reanant) 87.72 35.24

% (+lpreanantla 12.28 64.16

!&s of gestation 0 to 25 26 to 50

Transabdominal Da\s 25 51 76 101 126

% (i/D reanantl 89.84

of gestation to 50 to 75 to 00 to 125 to 150

;The percentage The percentage negative).

AUGUST

likelihood likelihood

1988 VOL.

a

ultrasonography % (-/oreanant) 10.16

100.00

0.00

100.00

0.00

100.00

0.00

100.00

0.00

of identifying of identifying

30 NO. 2

a pregnant a pregnant

b

b

ewe as pregnant (true positive). ewe as nonpregnant (false

327

THERIOGENOLOGY

Figure

328

1.

Dorsal-ventral, intrarectal real-time scan using a 5.0 MHz transducer at Day 22 of gestation. The urinary bladder (B) is seen as a large, fluidfilled (dark area) structure in the upper left, ventral to the rectum. The convoluted, pregnant, fluid-distended uterine horn (U) is visible as smaller, paired, echolucent (dark areas) structures in the upper center, anterior (right of) to the urinary bladder.

AUGUST

1988 VOL.

30 NO. 2

THERIOGENOLOGY

From the transabdominal approach, pregnancy was first verified at Day 25 after breeding. The likelihood of identifying a gravid ewe as pregnant was 89.94% at Days 25 to 50 of gestation and 100% from Days 51 to 150 of gestation using transabdominal real-time ultrasonography (Table 1). Differentiation

of Fetal

Number

The earliest observation of multiple fetuses was made at Day 31 on an intrarectal scan. The Chi-square test of independence showed no significant association (P 5 0.005) at Days O-25 of gestation between the number of fetuses observed using intrarectal real-time ultrasonography and the actual number of lambs born. A significant association (P 5 0.005) was noted with the intrarectal approach at Days 26 to 50 of gestation. The likelihood of making a correct determination of fetal number at Days 26 to 50 of gestation scanning intrarectally was 80.0% for single and 4.62% for twin born lambs (Table 2). Table

2.

Percentage likelihood of correctly determining fetal number in pregnant ewes using two different applications of real-time ultrasonography at various lengths of gestation Intrarectal

ms

of gestation

0 to 25 26 to 50

ultrasonography

% (1 /sinalela

% (2/twins)

13.48 80.00

0.0 4.62

Transabdominal Qgys 25 51 76 101 126

of nestation to to to to to

50 75 100 125 150

“Thl: percentage likelihood of identifying pregnant with one lamb (true positive). Thl: percentage likelihood of identifying pregnant with twin lambs (true positive).

b

ultrasonography

% (l/sir~nle)~

% (2/twins)

78.85 100.00 78.78 66.67 100.00

56.58 97.30 81.25 47.05 33.33

b

a pregnant

ewe carrying

one lamb

a pregnant

ewe carrying

twin

lambs

as as

With the transabdominal technique, twin fetuses were first recognized at Day 35 of gestation (Figure 2). There was a significant association (P 5 0.005) between the actual number of lambs born and the number of fetuses observed using transabdominal real-time ultrasonography after Day 25 of gestation. The highest accuracy rate for differentiating fetal numbers using transabdominal scanning was 100.0% for ewe!: carrying one lamb and 97.3% for ewes carrying two lambs at Days 51 to 75 of gestation (Table 2).

AUGUST

1988 VOL.

30 NO. 2

329

rHERIOGENOLOGY

Figure

Fetal

2.

Cutaneous, transabdominal real-time scan using a 5.0 MHz transducer at Day 40 of gestation. Twin fetuses (F) are apparent (lower left and upper right) as distinct, circumscribed, echodense (white areas) structures within each fluid-filled (dark areas) uterine horn.

Viability

and

Development

Embryonic mortality was documented in one ewe that was first detected pregnant on an intrarectal scan at Day 24. Serial transabdominal scans identified a single, viable fetus from Days 33 to 45 as determined by fetal movement or heart beat. On Day 49 of gestation, the embryo was noted to be motionless and no heart beat was observed. Further serial scans repeatedly revealed a motionless fetus that was not increasing in size when compared with fetuses of similar gestational The amount of intrauterine fluid decreased age or previous scans of the same fetus, during this period until at Day 60 the fetus and uterus were no longer identifiable. An abortion or resorption was surmised and the ewe was penned separately with a fertile ram. Repeated weekly transabdominal scans revealed no uterine activity until Day 105 after the first mating, at which time some uterine fluid was again noted. Seven days later, a new developing fetus with a beating heart was seen. This ewe eventually lambed normally, but data on her were not included in the statistical analysis. Another ewe, identified as carrying twins, developed acute, marked bilateral abdominal distension and anorexia at 110 d of gestation. Physical examination, complete blood count, serum chemistries and urinalysis ruled out infectious disease and pregnancy ketosis. Hydrops allantois was diagnosed from serial, real-time ultrasound scans which demonstrated an increased amount of intrauterine fluid when compared with ewes carrying twin fetuses of similar gestational age (Figure 3).

AUGUST

1988 VOL. 30 NO. 2

THERIOGENOLOGY

Figure

3.

Hydrops allantois diagnosed at Day 110 of gestation. A transabdominal image demonstrates the increased echolucent fluid distension (dark area) within the uterine lumen (U). A placentome (P) (white area) is seen in the lower center area.

The ewe’s appetite returned to normal following percutaneous, ultrasoundguided allantocentesis. On Day 117, the fetal heart rates had decreased to 30 to 50 beats/minute. The uterine placentomes progressively lost definition on subsequent scans. Neither movement nor a heart beat were observed on Day 119 in the lifeless fetuses. The ewe remained bright and alert and aborted dead twin lambs without complications at Day 122 of gestation. Lambing

Performance

A total of 37 lambs, (9 singles and 14 pairs of twins) were born to 23 ewes in the project flock. In addition to the two lambs delivered prematurely by the ewe with hydrops allantois, four other lambs were either born dead or died during the perinatal period. All six lambs were autopsied to determine the cause of death. No anatomical abnormalities were noted on either gross necropsy or histological examination, and no pathogenic organisms were isolated from bacterial culture of fetal or placental tissues. DISCUSSION Real-time ultrasonography can be used successfully in sheep to determine pregnancy status, fetal number, viability and gestational age. Clear fluid distension of one or both uterine horns was the earliest detectable sign of pregnancy by either intrarectal or transabdominal scanning. The urinary bladder was always recognizable just cranial to the pubis on all intrarectal scans as an echolucent, fluid-filled viscus, regardless of how recently the ewe had urinated, and thus had to be differentiated from an early pregnancy. From the abdominal approach, the bladder was

AUGUST

1988 VOL.

30 NO. 2

331

I-HERIOGENOLOGY not always visible. Pyometra, hydrometra and metritis are potential differential diagnoses in very early pregnancy (23), but serial ultrasound examinations ruled uterine pathology, as fetuses were eventually demonstrated in all cases.

out

As gestation progressed, the developing conceptus and identification of placentomes became the major criteria for determining pregnancy status. A fetal heart beat was first detected via intrarectal scanning on Day 25 and via transabdominal examination on Day 29. The characteristic cotyledonary placentation of ruminants was easily recognizable when examining ewes from the ventral abdomen and facilitated a rapid, accurate determination of pregnancy status (Figure 4).

Figure

4.

TransabdominaI, real-time scan at Day 75 of gestation. A typical placentome (P) is readily imaged (white area, upper center) in this view. Placental structures appear as cup- or doughnut-shaped objects depending upon the directional axis or plane of the ultrasound beam. A fetal torso (F) (white area, center) is also visible. Fetal movement and cardiac motion were apparent in real-time. Echolucent (dark) areas are uterine fluid.

Recognition of placental or fetal structures as a criteria for pregnancy precludes the possibility of making a false positive diagnosis. However, prior to 45 d of gestation, it was possible to misdiagnose a ewe as nonpregnant by either method of scanning, when she was in fact pregnant, because fetal or placental structures were sometimes missed at these early stages. During the first 25 d of gestation, the probabilty of incorrectly diagnosing a pregnant ewe as nonpregnant is high (87.72%) because the ovaries, nonpregnant uterus, developing blastocyst and gravid uterus less than 20 to 25 d of gestation are not identifiable with the 5.0 MHz transducer by either the intrarectal or transabdominal techniques. Although abdominat scanning of the synchronized sheep did not begin until Day 25 of gestation, two ewes conceived on their second breeding and

AUGUST

1988 VOL. 30 NO. 2

THERIOGENOLOGY

therefore were scanned nonpregnant ewes were

transabdominally from Day 16 of gestation. examined repeatedly by both methods.

The two

From Days 25 to 50 of gestation, the likelihood of correctly determining pregnancy status was more favorable using the transabdominal approach (89.84%) than the intrarectal method (64.76%). With the intrarectal approach, inability to manually manipulate the transducer into a position of direct contact with the reproductive tract of the ewe in contrast to the cow or mare, yielded inconsistent assessments of each ewe’s pregnancy status from one scan date to the next. The fact that pregnancy was first determined by abdominal scanning within 5 d of the intrarectal technique indicates that the ovine gravid uterus descends over the pelvic brim early in gestation. Two pregnant ewes were consistently misdiagnosed as open on every intrarectal scan performed through Day 50, although each ewe was positively identified as pregnant from the transabdominal approach by 25 and 29 d, respectively. The probability of making a false negative diagnosis of pregnancy status by scanning intrarectally after Day 25 was high (35.24%), despite obtaining serial scans of the same sheep. Pregnancy determination using the transabdominal approach was more consistent on sequential examinations. From Day 46 of gestation until parturition, the ability to differentiate pregnant from open ewes was 100% accurate when using the transabdominal approach, regardless of the technician performing the scans and their previous experience with the technique. Intrarectal scanning was discontinued after Day 50, since examination of the gravid uterus was more readily and accurately conducted from the abdominal position. A. diagnosis of pregnancy infers that at least one fetus is present. Differentiation of single from multiple fetuses was made on the basis of observation of recognizably separate fetal heads, hearts, thoraces or torsos. Movement of the individual fetal lambs during scanning facilitated the determination of the number of fetuses present. Although the earliest observation of multiple fetuses was made on an intrarectal scan, the probability of correctly identifying twin fetuses from the intrarectal approach was low (4.62%), and the differentiation of all single from multiple bearing ewes was never acheived via this technique. The optimum period for accurately determining fetal number using the transabdominal approach was during the second trimester of gestation. The likelihood of correctly identifying ewes carrying single or twin fetuses was very high (100.0% and 97.3%, respectively) with transabdominal ultrasonography at Days 51 to 75 of gestation. Fetal number was correctly determined in all the pregnant ewes by Day 46 using the abdominal technique. In the last trimester, the marked increase in fetal size and normal reduction in the amount of amnioallantoic fluid often prevented reliable differentiation of fetal structures and an accurate assessment of fetal numbers. These findings are similar to those reported by others based on single determinations for pregnancy diagnosis and determination of fetal numbers in sheep employing a real-time ultrasound scanning device (24-29). A common error in determining fetal number late in the first trimester was to miss the second fetus. From late in the second trimester until parturition, it was common to underestimate or overestimate fetal numbers due to poor imaging of distinct fetal structures. The apparent increased ability to accurately identify ewes carrying a single lamb at Days 126 to IS0 of gestation, despite a continued decline in the likelihood of accurately identifying ewes bearing twin lambs during the same period, was actually due to superimposition of the ability to correctly detect pregnancy (i.e. the presence of at least one fetus) upon a rising inability to determine

AUGUST

1988 VOL.

30 NO. 2

333

rHERIOGENOLOGY

fetal number in late gestation (Table 2). Errors in predicting fetal numbers were less likely to occur if both fetuses were identified by scanning from only one flank, since it was easy to mistakenly identify different images of the same fetus as two different fetuses when scanning from both flanks. The probability of correctly determining pregnancy status and fetal number were highest using the transabdominal technique. In the commerical setting, where estrus synchronization is uncommon, natural service dates are unobserved or unrecorded, duration of gestation is unknown and ewes are examined only one time, abdominal scanning would be the preferred method to detect pregnancy or determine the number of fetuses present. Preliminary studies on ewes at various stages of gestation suggest that the lateral recumbent position for intrarectal scanning and the standing position for abdominal scanning are advantageous for acheiving the desired goals of determining pregnancy status and number of fetuses in ewes. Placing the ewes in right lateral recumbency often facilitates intrarectal scanning by retaining the uterus within the ewe’s pelvic canal, thus enhancing the likelihood of contact between the transducer and reproductive tract. Transabdominal scanning with the ewe standing upright produces better results in advanced pregnancy, as the gravid uterus descends along the ventral abdominal floor. Transabdominal scanning allows orientation of the transducer in multiple scan planes, while intrarectal scanning allows only a sagittal (longitudinal) scan plane. Multiple planes facilitate detection and evaluation of various intra-abdominal structures such as the uterus in early pregnancy. Although abdominal scanning in our study was done from both right and left flanks, pregnancy detection and determination of the number of fetuses was usually done from the right side, as the rumen occupies most of the area of the left flank. Either method of examination was well tolerated by the ewes. No problems with hemorrhaging, tenesmus or rectal prolapse occurred with the intrarectal technique. Both methods of scanning the ewes were rapid, nonstressful and were not labor intense. Withholding feed for 12 to 18 h prior to scanning resulted in an appreciable decrease in gastrointestinal tract gas accumulation, which is a major hindrance to pregnancy diagnosis via the intrarectal approach. This is less of a problem when scanning transabdominally, as the gravid uterus is usually found positioned directly against the ventral abdominal wall. The coincidental occurrence of a fetal death with subsequent resorption or abortion in one ewe and the development of a hydrops allantois in another ewe in the research flock gave a unique opportunity to monitor serially two pathological conditions of gestation in the ewe. Had the ewe that lost her fetus been scanned only once while pregnant, she would have been diagnosed pregnant but would have failed to produce any lambs. A positive pregnancy diagnosis is no guarantee of a lamb at parturition time! Lambing performance may be slightly lower than that predicted by ultrasonography due to prenatal mortality of one or more embryos or fetuses. Some fetal resorption or abortion unrelated to ultrasonic scanning is normal in most species (30). Prior to the development of diagnostic ultrasonography for the early detection of pregnancy, fetal attrition had not been easily documented in the ewe and in most cases has gone unrecorded (30). However, real-time ultrasonography has been successfully used to assess fetal viability and monitor fetal growth and development. Such monitoring capabilities could prove invaluable in managing infertility and gestational problems such as pregnancy toxemia.

334

AUGUST

1988 VOL.

30 NO. 2

THERIOGENOLOGY

The question of safety in using ultrasound for pregnancy detection is an ongoing concern in all species. The ewes in this study and their fetuses were repeatedly exposed to ultrasound shortly after conception and throughout gestation. Each ewe was scanned 34 different times, so that the cumulative total time her reproductive tract and fetus(es) were exposed to diagnostic ultrasound in this study was approximately 2.8 h. It is highly unlikely that sheep in a field situation would ever be exposed to such prolonged durations of diagnostic ultrasonic energy. No adverse effects were noted among the ewes and no abnormalities were found in any of the lambs born alive. No congenital abnormalities were found in the lambs which were aborted, born weak or died. The incidence of abortions, weak or dead lambs in the project flock was lower than the rest of the flock housed at the same facility. Perinatal lamb mortality was 16% among the ewes that were not subjected to real-time ultrasonography. The technology appears safe and provides more information on the reproductive status of the ewe than has ever been possible in the past. With experience, real-time ultrasonography will allow greater control of ovine reproduction, and as a result, more efficient and profitable sheep production REFERENCES 1.

Watt B.R., Anderson G.A., and Campbell used for detecting pregnancy in sheep.

2.

Fukui Y., Kimura T., and Ono H. Multiple pregnancy diagnosis an ultrasonic doppler method. Vet. Rec. 114:145, 1984.

3.

Made1 A.J. Detection Rec. 112:11-12, 1983.

4.

Rawlings NC., and assessment Theriogenology

5.

Turner C.B., and Hindson J.C. An assessment of a method nancy diagnosis in the ewe. Vet. Rec. 96:56-68, 1975.

6.

Hulet C.V. A rectal-abdominal palpation technique the ewe. J. Anim. Sci. s(4):814-819, 1972.

7.

Richardson 90:264-275,

8.

Meredith M.J., and Madoni M.O.K. The detection of pregnancy A-mode ultrasound. Br. Vet. J. -(4):325-330, 1980.

9.

Mellor D.J., and Murray L. Effects on the rate of increase refeeding ewes after short periods of severe undernutrition nancy. Res. Vet. Sci. 2377-382, 1982.

10.

Robinson J.J. Nutrition of the Pregnant Ewe. m: Treacher Requirements for Lactation in the Ewe. Sheep Production. Butterworths: London, Boston, Toronto 1983; pp. 11 l-153.

AUGUST

of pregnancy

I.P. A comparison of six methods Aust. Vet. J. 61:377-382, 1984.

in ewe lambs

by A-mode

in sheep

using

ultrasound.

Vet.

Jeffcoate LA., and Savage N.C., et al. Pregnancy diagnosis of fetal numbers in the ewe in a commercial setting. fi(5):655-663, 1983.

C. Pregnancy 1972.

1988 VOL.

diagnosis

30 NO. 2

in the ewe:

of manual

for diagnosing

a review.

preg-

pregnancy

in

Vet. Rec.

in sheep

in fetal during

by

girth of late preg-

T.T.: Nutrient Haresign W (ed.).

335

THERIOGENOLOGY

11.

Russel A.J.F., Doney J.M., and Reid R.L.: The use of biochemical parameters in controlling nutritional state in pregnant ewes, and the effect of undernourish ment during pregnancy on lamb birth-weight. J. Agric. Sci. @:351-358, 1967.

12.

Scott G.E. The sheepman’s production handbook. ment, Denver, Colorado 80206, 1983; p. 4-26.

13.

Ford E.J.H. Pregnancy Toxaemia. Blackwell Scientific Publications,

14.

Curran S., Pierson R.A., and Ginther O.J. Ultrasonographic appearance bovine conceptus from days 10 through 20. J. Am. Vet. Med. Assoc. m(lO):l289-1294, 1986.

of the

15.

Curran S., Pierson R.A., and Ginther O.J. Ultrasonographic appearance bovine conceptus from days 20 through 60. J. Am. Vet. Med. Assoc. m(10):1295-1302, 1986.

of the

16.

Pierson R.A., and Ginther O.J. study of embryonic development

17.

Inaba T., Nakazima sows by ultrasonic

18.

Ginther O.J. Ultrasonic evaluation of the reproductive single embryo. J. Eq. Vet. Sci. 3(2):75-81, 1984.

19.

Allen mares

20.

Ott L., and Mendenhall W. The chi-square test of independence. Understanding statistics, Fourth Edition, Duxbury Press, Boston 1985; pp. 433-438.

21.

Remington and Health

22.

Feinstein A.R. Clinical 1977; ~~214-226.

23.

Pieterse M.C., and Taverne M.A.M. Hydrometra in goats: diagnosis with time ultrasound and treatment with prostaglandins or oxytocin. Therio. X(6):813-821, 1986.

24.

Lindahl I.L. Pregnancy Q(6):1135-1140, 1976.

25.

Fowler D.G., and Wilkins J.F. Diagnosis of pregnancy and number of foetuses in sheep by real-time ultrasonic imaging. I. Effects of number of foetuses, stage of gestation, operator and breed of ewe on accuracy of diagnosis. Livestock Production Science 11:437-450, 1984.

26.

White I.R., and Russel A.J.F., in the diagnosis of pregnancy Vet. Rec. 115:140-143, 1984.

336

Sheep

Industry

In: Martin W.B. (ed.), Diseases London 1983; pp. 147-151.

Develop-

of Sheep.

Ultrasonography for detection of prenancy in heifers. Therio. 2(2):225-233, 1984.

Y., Matsui N., and Imori linear electronic scanning.

T.

Early Therio.

pregnancy x(1):97-101, tract

and

diagnosis 1983.

in

of the mare:

the

W.E., and Goddard P.J. Serial investigations of early pregnancy in pony using real-time ultrasound scanning. Equine Vet. J. x(6):509-514, 1984.

R.D., and Schork M.A. Statistics with applications: Sciences, Prentice Hall, New Jersey, 1970; pp 78-80. biostatistics.

diagnosis

The CV Mosby

in ewes by ultrasonic

Co.

The Biological

St. Louis,

scanning.

Missouri,

real-

J. An. Sci.

and Fowler D.G. Real-time ultrasonic scanning and the determination of fetal numbers in sheep.

AUGUST

1988 VOL.

30 NO. 2

THERIOGENOLOGY

27.

White I.R., and Russel A.J.F. rlsal-time ultrasonic scanning.

28.

Buckrell wetally

29.

Davey C.G. An evaluation ultrasound scanner. Aust.

30.

Chapman H.M. Theriogenology. 900.

AUlGUST

Determination In Practice

of fetal numbers 6(6):200-202, 1984.

in sheep

by

B.C., Bonnett B.N., and Johnson W.H. The use of real-time ultrasound for early pregnancy diagnosis in sheep. Therio. x(5):665-673, 1986. of pregnancy testing Vet. J. a(10):347-348,

Prenatal loss. WB Saunders,

1988 VOL.

in sheep 1986.

using

a real-time

In, Morrow DA (ed.). Current Therapy in Co., Philadelphia, London, Toronto 1980; pp. 896-

30 NO. 2

337