THERIOGENOLOGY
EVALUATION
OF ULTRASONOGRAPHY FOR MONITORING RHESUS MONKEYS
P.M. Morgan,
R.J.
Hutz,
Wisconsin
Received
FOLLICULAR
E.M. Kraus, J.A. Cormie, and B.D. Bavister
Regional Primate Research University of Wisconsin Madison, WI 53715-1299
for publication: Accepted:
September
Mm&
D.J.
GROWTH
IN
Dierschke,
Center
11, j9Jc
16, 1987
ABSTRACT if the The purpose of this study was to determine ovaries and of rhesus monkeys could be visualized by ultrasonography and to uterus detect changes associated with follicular and growth ovulation. examined during Animals were 15 menstrual cycles, for an average of nine consecutive days. Ultrasonic recordings were correlated with hormonal parameters (estradiol 17l3, Ep; luteinizing hormone, LH; and progesterone, P) and laparoscopic findings. The uterus and both were observed in more than 90% of the examinations. A dominant ovaries follicle (DF) was identified during all ovulatory cycles, on average 1 d preceding the Es peak. The maximal diameter of the DF ranged from Laparoscopic examinations to determine the site of 3 to 7 mm. the DF confirmed ultrasonic findings in 10 of 14 cycles (P < 0.1). There was no significant difference in the size of the dominant and contralateral ovaries; however, more follicles with a diameter of 2 to 7 mm were found on the dominant ovary (P < 0.05). Two animals stimulated with showed a linear exogenous gonadotropins increase in ovarian size for 6 d prior to oocyte recovery (P < 0.05), reflecting an increase in the number of developing follicles. Ultrasonography can be used to identify the DF during spontaneous cycles in rhesus monkeys and to monitor the response of monkeys to exogenous gonadotropins.
Key words:
rhesus
monkey,
ultrasonography,
dominant
follicle
Acknowledgments Publication No. 26-008 from the Wisconsin Regional Primate Research Center. Supported by NIH grants HD14765 and RR00167. We wish to thank Roger Pierson for his advice concerning the selection of the ultrasound equipment, Dr. Samuel Shall for use of computer programs, Pat Meller performing for progesterone assays, Bob Dodsworth for his advice and help with photography, Michael Erwin for assistance with animal handling, Linda Endlich for preparing the figures, and Elaine Anderson for typing the manuscript.
MAY
1987 VOL.
27 NO. 5
769
THERIOGENOLOGY
INTRODUCTION Ultrasonographic techniques allow repeated noninvasive examinahave tions and been used to study follicular development and luteal status in, among others, women (l), mares (Z), and heifers (3). The diameter of the unstimulated preovulatory follicle, as measured by ultrasonography, varies from > 13 mm in the heifer (3) to approximately 22 mm in women (1) and 45 mm in the mare (2). In the rhesus monkey, the diameter of the preovulatory follicle, measured at laparoscopy, has been reported to be approximately 9 mm on the day of the LH surge (4) or 6.6 mm on Day 10 of the cycle (5). This study initiated to assess was the use of a real-time ultrasound instrument in monitoring follicular growth and ovulation in normally cyclic and gonadotropin-stimulated rhesus monkeys.
MATERIALS
AND METHODS
Eleven mature rhesus monkeys (Macaca mulatta) were used in this _____ Animals were housed singly, in a controlled study. environment, at the Wisconsin Regional Primate Research Center. Animal care and handling were as previously described (6). Animals that demonstrated at least one normal cycle prior to the study were examined during a total of 15 menstrual cycles. Ultrasonography was performed for an average of nine Initially, animals were scanned beginning on consecutive days. Day 4 or 5 (n = 6 cycles), Day 1 being the first day of menses. On the basis of preliminary results from these animals indicating that few follicles > 2 mm were observed before Day 8, daily scans were initiated on Day 8 of the cycle for the remainder of the study. Blood was sampled on Days 8 through 14; beginning on Day 16, blood samples were 2, 4, 6, and taken on alternate days until the next menses. To determine whether ultrasonography affected the menstrual cycles of the animals during the from sex skin patterns (7) were comstudy, cycle lengths as determined pared to those during the previous cycle. The ultrasound machine used was an ATL MK 100 real-time instrument scanhead of high frequency and short focus (Advanced with a 7.5 MHZ Technology Laboratories, Inc, Bellevue, WA). Structures as small as 1 mm are within the resolving power of this ultrasound unit. This unit is equipped with an "image reverse" control that reverses the leftorientation of the ultrasound image on the screen, Using the right uterus as a reference point and considering the position of the animal, the left and right ovaries can be distinguished on the screen. Reversing the orientation on the screen with the image reverse could lead to confusion concerning which is the left or right ovary. position, Animals were restrained in supine without anesthesia. transducer was placed on the lower midline of the animal to locate The the uterus in a transverse plane. The widest and longest diameters of uterus in this plane were measured, using omnidirectional calipers the superimposed on the static (or frozen) image on the screen. Using the a landmark, the transducer was moved slowly in a transverse uterus as plane to locate the ovaries: these appeared as ovoid structures with hyperechoic borders. Scanning across an urine-filled bladder, when present, provides an anechoic window that enhances the view of the
MAY
1987 VOL.
27 NO. 5
THERIOGENOLOGY
this condition was not essential for success. The however, ovaries: muscles of the side wall of the pelvis may be mistaken for the ovary the monitor while scanning Examining the image on from a (8). that the remains transverse to a longitudinal plane will show ovary The position of the unlike the muscle which appears elongated. ovoid, ovarian vessels within the ligamentum suspensorium ovarii is stable in relation to the ovary: thus, these vessels may also be used as a reference to identify the ovaries (9). by passing the beam of The ovaries and follicles were scanned ultrasound from cranial to caudal through the ovary, thus ensuring that Follicles appear as dark (or hypothe entire ovary was observed. regions within the outline of the ovary. Follicles of < 2 mm echoic) diameter (estimated by comparison with the l-cm scale along the edge of were counted while scanning through the ovary: this techthe image) nique minimized errors due to loss of identity of individual follicles. > 2 mm were measured using omnidirectional calipers, and the Follicles largest diameter of each follicle was recorded. The largest follicle by ultrasonography was designated the DF. observed during the cycle One laparoscopy was performed per cycle to identify the ovary having a DF or corpus luteum (lo), and this ovary was designated as the dominant from ovary. The size of the ovaries and the uterus was determined twoassuming an elliptical shape (n x length x dimensional measurements, width/4). In the human (11) and the heifer (3), the corpus luteum has echogenic pattern than that of surrounding tissue and can a different In this study of distinguished by ultrasonography. the rhesus be we were unable to distinguish the corpus luteum by ultrasonomonkey, To minimize observer error (12), all examinations were pergraphy. formed by one person (JAC) who had no knowledge of the results of previous examinations. Sera were analyzed for Ez using a rapid assay (Diagnostic Products Corporation, Los Angeles, CA). The within- and between- assay coefficients of variation were 11.3 and 12.7%, respectively, at a control of LH and P were determined according value of 50 pg/ml. Concentrations to methods previously described (6). In a preliminary study using ultrasonography to monitor follicular growth in response to exogenous gonadotropins, two animals were one with pregnant mare serum gonadotropin (13) and treated the other with human menopausal gonadotropin (hMG; Perqonal, 0.5 ampules on Days 3-6, 1.0 ampules on Days 7-11). Data Analysis All ultrasonic recordings were normalized to the of the Es day peak (Day 0). The mean concentration of EZ on Day -1 was 127 + 15 pg/ml (SEM, n = 14), on Day 0 was 243 -t 30 pg/ml, and on Day 1 was 96 ? The LH peak was either the same day as (n = 6) or the day 16 pg/ml. after (n = 7) the E, peak. Results for the dominant ovary (DF determined by laparoscopy) and the contralateral ovary were pooled, respectively. Ovarian sizes at laparoscopy and ultrasonography were compared by Student's paired t-test. Characteristics of follicle development were analyzed by Chi-square goodness of fit. Incidence of ovulation was binomial analyzed by the test. Changes in uterine and ovarian size
MAY
1987 VOL.
27 NO. 5
771
THERIOGENOLOGY
evaluated by Student-Newman-Keuls
were
analysis of followed by one-way variance the P < 0.05 "as considered significant. test (14).
RESULTS Of the 11 animals studied, 10 had ovulatory cycles (n = 14) while were being studied, as indicated by appropriate hormonal changes they and observations at laparoscopy. The one did exhibit exception not peaks in E2 and LH, no follicles > 2 mm were observed by midcycle ultrasonography, P levels were basal during the luteal phase, and a observed at laparoscopy. This animal "as corpus luteum was not excluded when further data analysis "as performed. Ultrasonography had no. effect on cycle length, as determined from skin patterns and occurrence of menstruation. sex The lengths of the follicular phase in the previous cycle and the cycle under study were simi13.0 I! 0.6 d (X + SEM, n = 14) and 13.1 ? 0.6 d, respectively: in the length of the luteal phases larly, there "as no difference 0.5 d and 15.2 + 0.5 d, respectively). P levels were within (15.1 f the normal range (at least 1 ng/ml for at least 10 d). The uterus was visualized during each ultrasonic examination (Figand data were analyzed for Day -6 to Day 2. The mean size of ure 1A) (420 f 27 mm*, SEM, n = 16) the uterus was larger on Days -5 and;4 P < 0.05). than on Day -2 to Day 2 (308 f 11 mm , SEM, n = 60; Both ovaries were identified in 91.4% of the examinations (n = 128; Figure 1B). Neither ovary "as identified in24% of trials. Ovarian as measured at laparoscopy (47.9 f 5.8 mm [? * SEM, size n = 7I)or ultrasound (47.2 f 6.8 mm*) did not differ significantly. For all by animals showing hormonal patterns indicative of an ovulatory cycle, we were able to identify a dominant follicle during retrospective examinaThe DF "as identified, on tion of the ultrasonographic data. average, 1 d before the peak of E, (range, Day -4 to Day 1). Laparoscopic observations to determine the site of the DF were in agreement with ultrafindings in 10 of 14 cycles (P < 0.1). A reasonable explanation sonic to account for incorrectly assigning the DF to the contralateral ovary machine "as moved. is that the image reverse control on the ultrasound This possibility is supported by the fact that three of animals four of the study had the DF incorrectly during final phase scanned the while prior to that 9 of 10 assignments were correct. Howassigned, only for those animals for which ultrasonoever, results are presented graphic and laparoscopic findings agree. The maximal diameters of the 3 mm (n = l), 4 mm (n = 2), 5 mm (n = 2), 6 mm (n = 3) and 7 were DFs When the diammm (n = 2), with a mean of 5.3 -i 0.4 (SEM) (Figure 1C). the DF was 3 or 4 mm, no other follicles > 1 mm were observed eter of the DF "as no longer visible by ultrasoon the ovaries. On average, No follinography on the day of the E, peak (range, Day -2 to Day 2). cles > 3 mm were observed on the dominant ovary by 2 d after the E2 The corpus luteum "as not distinguishable by ultrasonography in peak. any of the cycles. Up to 10 follicles < 2 mm were observed per ovary throughout the cycle (data not presented). During 56% of the scans (n = menstrual The maximum number (Figure 2). 170), no follicles L 2 mm were observed
772
MAY
1987 VOL.
27 NO. 5
THERIOGENOLOGY
Dominant
Figure
1.
MAY
Folllclo
outline drawings of Ultrasonograms and the uterus and ovaries in rhesus monkeys. A) Transverse view of the The area between the cross-shaped uterus. cursors is the width of the uterus. B) Transverse view showing the uterus and both ovaries. Cl Longitudinal view of an ovary (11 x 5 mm) with a dominant follicle (6 x 4 mm). The uterus is evident in the foreground.
1987 VOL.
27 NO. 5
773
THERIOGENOLOGY
of follicles observed per ovary in a spontaneously cyclic animal was four (Figure 2). Seventy-five percent of all follicles were 2 to 3 mm in diameter were equally distributed on both ovaries (Table 1). and Follicles were observed more often on the dominant than the contralateral ovary (P < 0.05; Table 1). On Day -3, one follicle t 3 mm was observed on the dominant ovary in 80% of examinations, while a follicle seen on only 20% of the contralateral ovaries examined (P < 0.05); was no differences were observed on other days. The mean sizezof the dominant ovary, from Day -9 to Day 3, varied from 45 to 82 mm , and for the contralateral ovary from 46 to 71 mm=; there was no significant difference among days or between ovaries. were Both monkeys that stimulated showed a linear increase in ovarian size from Day -6 to Day 0 (Day 0 is the day of oocyte recovery; P < 0.05; Figure 3). The mean ovarian size was 364 f 30 mm2 (SEM, n = 4) on Day 0 compared to a mean of 53.3 ? 1.5 mm2 (SEM, n = 226) for spontaneously cyclic animals. In the animal stimulated with PMSG, there were 13 follicles of diameter > 3 mm on Day -3, of which 30% were L 6 mm. On Day -1, 55% of follicles observed were > 6 mm. On Day 0, 23 follicles were observed and 20 follicles of diameter > 3 mm were aspirated at laparoscopy (Figure 4A). In the hMGstimulated animal (Figure 4B) no follicles > 3 mm were observed on Day -4; however, on the day of oocyte recovery, 7 follicles > 3 mm diameter were observed by ultrasonography and this was confirmed at ovariectomy.
DISCUSSION To our knowledge, this is the first study to show that the ovaries and follicles in rhesus monkeys can be observed by ultrasonography. Our group has previously reported changes in endometrial indices during the menstrual cycle of rhesus monkeys (15). Pelvic ultrasonography in women during spontaneous and induced cycles has become routine (9). these studies, a full bladder (8) or two-thirds-full However, for A full bladder probladder (12) has been stated to be a prerequisite. vides an window which enhances the view of the ovaries, and anechoic equally importantly, displaces the bowel out of the pelvis (8). Reverare observed when sound waves encounter the highly beration artifacts reflective gas-filled bowel and bounce between the bowel and transducer results in bright ethos being recorded on the movement This (16). screen and thus interferes with reflections from the pelvic structures could not be controlled by noninvasive interest. Bladder volume of means in these monkeys, yet it was possible to visualize the uterus and both ovaries in more than 90% of the examinations. Since cycle lengths and progesterone production were within the normal ranges (17), animals were apparently not stressed by the procedures. Analysis of uterine measurements for Day -6 to Day 2 showed that mean larger in the early follicular phase. A similar the size was study in women showed that uterine size was smallest during the midcycle and increased significantly toward the end of the menstrual cycle (18). technique The accuracy of the ultrasonographic was comparing the ultrasonic with the laparoscopic findings.
774
assessed There was
MAY 1987 VOL. 27 NO. 5
by no
THERIOGENOLOGY
1
2
4
3
Number of follicles > 2 mm observed/ovary
Figure
2.
The distribution mm per ovary Numbers scans. scans.
of the number of as observed during above bars indicate
follicles of diameter L 2 a total of 170 ultrasound total the percentage of
500 r
2=0.44
.
p = 0.002
400
.
.
300
200
100 0 0
-8
-4
-6
-2
0
2
Days from oocyte recovery
Figure
3.
The ovarian area of hMG, respectively. spontaneously cyclic
MAY 1987 VOL. 27 NO. 5
stimulated with PMSG and two animals The 0 indicates the mean ovarian size in animals (n = 226).
775
2
3
4
5
6
7
No. observations
1 .:: =1 ;
3 z
t; u
1
Table 1.
3
11
-8
4
1
-7
4
-6
6
2
12
-5
13
1
-3
6
13
2
9
4
13121
-4
10
4
6
-2
2
2
1
0
10 10
13
6
1
-1
Dominant ovary
8
3
2
3
1
7
2
2
2
5
3
3
3
3
-8
2
-7
5
2
-6
6
13
2
-5
5
2
-4
9
13
4
-3
10
3
1
1
-2
4
0
910
3
11
-1
Contralateral ovary
9
4
1
1
6
1
14
2
5
1
3
The total number of follicles of each size observed (from 8 d before to 3 d after the E2 peak) on the dominant ovary (ovary with dominant follicle) and contralateral ovary
THERIOGENOLOGY
-
Figure
4.
MAY
Ovarv
-
Ultrasonograms and outline drawings of ovaries of stimulated Transverse view of both ovaries in a monkeys. A) rhesus oocyte recovery. A PMSG-stimulated animal on the day of of 11 follicles was observed on the left ovary (24 x total 20 mm) and 12 follicles on the right ovary (25 x 15 mm); all Foil icles visible in the plane shown. not follicles are Ovary (11 x 11 mm)ofan ranged in size from 5 to 11 mm. B) with are hMG on stimulated animal Day -3; two follicles visible.
1987 VOL.
27 NO. 5
777
I-HERIOGENOLOGY
significant difference in ovarian size as during measured ultrasonography or laparoscopy. Fourteen animals were judged by ultrasonography to have ovulatory cycles (the presence of a follicle t 3 mm), and one was judged to an anovulatory cycle. have These judgments were confirmed by laparoscopy and hormonal patterns. A DF was identified during the Ep peak. all ovulatory cycles, on average 1 d preceding Clark et al. (4), during sequential laparoscopies, positively identified the DF in the rhesus monkey 5 d before the LH surge: the size of the preovulatory follicle was reported to be 9 mm on the day of the LH surge. Using diameter of the DF ranged from 3 to 7 mm. ultrasonography, the However, the DF was not visible by ultrasonography in immediately the preovulatory period, this lack of visiblity may account for the and smaller diameters observed. In women, follicles disappeared during 30% the cycles on the day of the E2 peak (19), or in some cycles prior of In the latter case, there was a subto the serum E, elevation (20). normal secretion of E2; early disappearance of the DF in our study was not correlated with abnormal E, secretion. Follicles with a diameter of 7 mm were observed only on the dominant ovary. folliHowever, a &mm cle was observed on the contralateral ovary (Table 1); therefore, folup to this size cannot be positively identified as the DF. licles The site of the DF as predicted from ultrasonography was confirmed in 10 of errors in assignment 14 cycles by of the DF may have laparoscopy; resulted from confusion about the left-right orientation of the image the screen. Failure to localize the DF has also been reported in on studies of women, and errors have been made when both ovaries contained a preovulatory follicle of about the same size (21). Follicles with diameters S 2 mm were detected by ultrasonography throughout the study, while follicles of this size were not observed by These smaller follicles may be located deep within laparoscopy (4). visible at laparoscopy. the ovary and thus would not be Although the sizes of the dominant and contralateral ovaries did differ, not more were observed on the dominant ovary. follicles The lack of difference in size may reflect the presence of a corpus luteum on the contralawhich not be observed as a separate structure by teral ovary could ultrasonography in this study. Two monkeys treated with exogenous gonadotropins showed a dramatic increase in the number of developing follicles and a linear increase in A marked increase in ovarian size has also been reported ovarian size. Although we were not able to moniin women stimulated with hMi (22). follitor individual follicles during successive days, we did observe cles increasing in size during the treatment. Ultrasonography is noninvasive and suitable for repeated applicaimportant when working with nonhution, features that are particularly In addition, the cost is low after initial outlay man primates. the machine and the information gained can aid in immediate decithe for sions regarding treatment (e.g., hCG administration). This study has ovarian size and the number of developing follicles can be shown that determined during the normal cycle in the rhesus monkey by ultrasonothis knowledge facilitated monitoring the follicular that graphy and growth in response to exogenous gonadotropins.
778
MAY
1987 VOL.
27 NO. 5
THERIOGENOLOGY
REFERENCES cox, L.W., Warnes, G.M., Seamark, Kerin, J.F., Edmonds, D.K., Matthews, C.D., Young, G.B. and Baird, D.T. Morphological R.F., and functional relations of graafian follicle growth to ovulation laparoscopic and biochemical measureusing ultrasonic, in women aa:al-90 (1981). ments. Br. J. Obstet. Gynecol.
1.
Pierson, R.A. preovulatory (1985).
and Ginther, follicle in
Ultrasonic evaluation of O.J. the the mare. Theriogenology 24:359-368
Pierson, R.A. and Ginther, O.J. Theriogenology 21:495-504 ovary.
Ultrasonography (1984).
of
the
bovine
Clark, J.R., Dierschke, D.J., Meller, P.A. and Wolf, R.C. Horovarian folliculogenesis in rhesus monkeys. regulation of monal II. Serum concentrations of estradiol-17B and follicle stimulating hormone associated with growth and identification of the preovula21:497-503 (1979). Biol. Reprod. tory follicle. 5.
Markers of atresia in Hutz, R-J., Dierschke, D.J. and Wolf, R.C. from rhesus monkeys treated with ovarian follicular components Biol. Reprod. 34:65-70 (1986). estradiol-17B.
6.
Hormonal Clark, J.R., Dierschke, D.J. and Wolf, R.C. regulation of ovarian folliculogenesis in rhesus monkeys. I. Concentrations during of serum luteinizing hormone and progesterone laparoscopy and patterns of follicular development during successive menstrual Biol. Reprod. 18:779-783 (1978). cycles.
7.
Czaja, J.A., Robinson, J.A., Eisele, S.G., Scheffler, G. and Goy, Relationship between sexual skin colour of female rhesus R.W. monkeys and midcycle plasma levels of oestradiol and progesterone. 49:147-150 (1977). J. Reprod. Fertil.
a.
Sample, W.F., Lippe, graphy of the normal
9.
evaluation of normal Ritchie, W.G.M. Ultrasound in the Fertil. Steril. 43:167-181 (1985). induced ovulation.
B.M. and Gyepes, M.T. Gray-scale ultrasonofemale pelvis. Radiology 125:477-483 (1977). and
10.
D.J. and Clark, J.R. Laparoscopy in Macaca Dierschke, equipment employed and initial observations. specialized ?:lOO-110 (1976). Primatol.
11.
Nitschke-Dabelstein, S., Hackeloer, B.J. and Strum. G. Ovulation formation observed by corpus luteum ultrasonography. and 1:33-39 (1981). Ultrasound Med. Biol.
12.
Inherent Prins, G.S. and Vogelzang, R.L. variability in relation follicular to (1984). Fert. Embryo Transfer 1:221-225
MAY
1987 VOL.
27 NO. 5
sources of measurements.
mulatta: J. Med.
ultrasound J. Vitro.
779
THERIOGENOLOGY
13.
Boatman, D.E., Morgan, P.M. and Bavister, B.D. Variables affecting yield and developmental potential of embryos following the fertilization in rhesus superstimulation and in vitro monkeys. Gamete Res. 2:327-338 (1986).
14.
Zar, J.H. Biostatistical Cliffs, New Jersey.1984,
15.
Morgan, P.M., Huts, R.J., Kraus, E.M., Cormie, J.A. and Bavister, assessment of the endometrium in rhesus B.D. Ultrasonographic Biol. monkeys during the normal menstrual cycle. Reprod. (in press, 1987).
16.
R.A. Ultrasonic evaluation of Ginther, O.J. and Pierson, the of the mare: principles, equipment, and techreproductive tract J. Equine Vet. Sci. 2:195-201 (1983). niques.
17.
DiZerega, G.S. and Hodgen, G.D. to aberrant tility: a sequel %:489-499 (1981).
18.
Kaihola, H.-L. Uterine size measured by Piiroinen, 0. and ultrasound during the menstrual cycle. Acta Obstet. Gynec. Stand. =:247-250 (1975).
19.
Ehret, M-C., Renaud, R.L., Macler, J., Dervain, I., Aron, C., A. and Pollack, H. Echographic study of Plas-Roser, Spira, S-r follicular maturation and ovulation during the normal menstrual 2:272-276 (1980). cycle. Fertil. Steril.
20.
HaselPolan, M.L., Totora, M., Caldwell, B.V., DeCherney, A.H., Abnormal ovarian cycles as diagnosed by tine, F.P. and Kase, N. Fertil. Steril. ultrasound and serum estradiol levels. 37:342347 (1982).
21.
Th. and Wladimiroff, Leerentveld, R.A., Van Gent, I., Alberda, of follicular development in clomiphene induced Assessment J.W. and laparoscopy: a comparative ultrasound cycles by means of Ultrasound Med. Biol. 2:595-598 (1983). study.
22.
Ultrasonic Ylostalo, P., Lindgren, P.G. and Nillius, S.J. measand uterine size during ovarian follicles, ovarian urement of Endocriinduction of ovulation with human gonadotrophins. Acta (1981). nol. %:592-598
780
Analysis. Prentice-Hall pp 61-369.
Inc.,
Luteal phase dysfunction folliculogenesis. Fertil.
MAY
1987 VOL.
Englewood
inferSteril.
27 NO. 5