Assessment of Ovulation by Ultrasound and Estradiol Levels during Spontaneous and Induced Cycles*

FERTILITY AND STERILITY Copyright " 1980 The American Fertility Society

Vol. 33, No.4, April 1980 Printed in U.s A.

ASSESSMENT OF OVULATION BY ULTRASOUND AND ESTRADIOL LEVELS DURING SPONTANEOUS AND INDUCED CYCLES*

DAVID H. SMITH, M.R.C.O.G., M.AuST.C.O.G. RICHARD H. PICKER, M.R.C.O.G., F.AuST.C.O.G., F.R.C.S.E., D.D.U. MICHAEL SINOSICH, B.Sc. DOUGLAS M. SAUNDERS, M.D., M.R.C.O.G., F.AuST.C.O.G., F.R.A.C.S.t Department of Obstetrics and Gynaecology, The Royal North Shore Hospital, St. Leonards, Sydney 2065, Australia

Both Graa{w.n follicle growth and subsequent ovulation were studied in 45 menstrual cycles of 28 patients by the estimation of plasma estradiol levels and by the measurement of follicle size and number by ultrasound. Twenty spontaneous ovulatory cycles were studied as controls compared with twenty cycles in which ovulation was induced by clomiphene and five cycles in which ovulation was induced by human pituitary gonadotropin. The means of the peak estradiol levels during the cycles in which one follicle was present were 1553.1 ± 87.8 pmoles/liter in the spontaneous cycles and 2296.8 ± 163.4 pmoles/liter in the clomiphene-treated cycles. Ultrasound was shown to be complementary to endocrine profiles because the number and diameter of Graa{w.n follicles could be measured accurately by this technique. Fertil Steril 33:387, 1980

of estradiol (E 2 ) levels, 9 in the timing of ovulation for artificial insemination. This study was undertaken to demonstrate how the monitoring of ovulation induction is improved by the addition of ultrasound to the currently used methods of measuring estrogen levels in either plasma3 ,11 or urine. 12 However, multiple pregnancies may still occur even when estrogen levels are closely controlledY In this study, information gained from the measurement of E2 levels and Graafian follicle diameters by ultrasound in spontaneous cycles of patients undergoing artificial insemination has been applied to the monitoring of ovulation induction.

The advent of modern ovulation-induction agents has meant that many pregnancies can now be successfully achieved in females whose infertility was due to ovulatory failure. 1 Ovulation induction should be monitored carefully to reduce the complications2 of ovarian hyperstimulation syndrome, multiple ovulations, and multiple pregnancies. Monitoring will improve the efficacy of treatment at both the hormonal and follicular level. 3-5 This will enable a more accurate titration of drug dosage and timing of the administration of human chorionic gonadotropin (heG). There are few reports6 -9 in the literature on the use of ultrasound for the measurement of Graafian follicle growth. A recent review 10 on the subject of monitoring ovulation induction makes no mention of the use of ultrasound for this purpose. Ultrasound has previously been shown to be of value, either alone6 -s or combined with the measurement

MATERIALS AND METHODS

Forty-five menstrual cycles of twenty-eight patients attending the fertility clinic at the Royal North Shore Hospital of Sydney were studied. Group I. This group consisted of 20 spontaneous cycles of 13 apparently normal women undergoing artificial insemination with donor semen; it served as a control group.

Received October 10, 1979; accepted November 28,1979. *Presented at the Thirty-Sixth Annual Meeting of The American Fertility Society, March 18 to 22, 1980, Houston, Tex. tTo whom reprint requests should be addressed.

387

388

April 19BO

SMITH ET AL.

TABLE 1. Results in Twenty Spontaneous Cycles of Thirteen Patients in Group 1 No. of follicles

No. of cycles

Follicle size em

pmole'"liter

1 2

19 1

2.55 ± O.la 1.7,2.5

1523 ± 87.8a 2335

E2 level

aYalues are,means ± standard error of the mean.

Group II. This group comprised 20 cycles of 11 infertile women in whom ovulation was induced with clomiphene citrate, 50 mg daily for 5 days beginning on day 5 of the menstrual cycle. Group III. This group consisted of five cycles of four infertile patients undergoing ovulation induction with human pituitary gonadotropin (hPG) using the method previously described by Brown et al. 12 Follicular growth and the timing of ovulation was assessed daily by plasma E2 levels and ultrasound examination in the periovulatory period. Estradiol levels were measured in peripheral blood samples obtained daily between 8 and 9 A.M. Unconjugated plasma E2 was measured by radioimmunoassay, using rabbit antiserum to 1713E 2-6-(O-carboxymethyl)oxime-bovine serum albumin obtained from Calbiochem (Carlingford, Australia). The limit of sensitivity of the assay was 0.02 pmole/liter. Cross-reaction with estrone was 0.6%; with estriol, 0.18%. Graafian follicle size was studied by B-mode gray-scale ultrasound using a Picker 80 L machine (Medtel Pty. Ltd., Lane Cove, Australia) equipped

FIG. 2. Transverse echogram of the pelvis on day 14 (same patient as in Figure 1) showing the follicle (F) having collapsed. B, Bladder; U, uterus. The E2 level was 874 pmoleslliter.

with a 3.5-MHz frequency transducer focused between 4 and 8 cm. Transverse echograms were taken at 0.5-cm intervals cephalad from the pubic symphysis. The patients were examined daily from day 10 of each cycle. RESULTS

Group I. The results of the 20 spontaneous ovulatory cycles are shown in Table 1. In 19 of these cycles only 1 follicle cyst was present (Figs. 1 and 2). The mean diameter (± standard error of the mean) of these follicles was 2.55 ± 0.1 cm and the mean peak E2 level (± standard error of the mean) was 1523.1 ± 87.8 pmoles/liter. Four pregnancies occurred in this group. However, in 1 of the 20 cycles, 2 Graafian follicle cysts were present. This cycle did not result in a pregnancy. Group II. The results of the 20 cycles in which ovulation was induced by clomiphene are shown in Table 2. Six pregnancies, including one set of twins (Fig. 3), occurred in the 11 patients in this group. The mean E2 value in this group when only one follicle was present was significantly higher (0.01 TABLE 2. Results in Twenty Clomiphene-Treated Cycles of Eleven Patients in Group 2 No. of follicles

Follicle size

No. of cycles

FIG. 1. Transverse echogram of the pelvis on day 13 showing a 3-cm follicle (F) on the right ovary. Posterior enhancement (P) can be noted behind the follicle. The full bladder (B) lies anterior to the uterus seen in transverse section (U). The top of the figure is anterior; the right (R) and left (L) sides of the patient are indicated.

1 2 2 3 Total

>1.5 >1.5, <1.5 >1.5 >1.5

E2 level

Pregnancies

pmoleslliter

em

9 3 7 1

2296.8 ± 163.4a 1750, 1922,2000 2942 ± 457a 6790

20

aYalues are means ± standard error of the mean. bOne set of twins.

3 0 3b 0

6

Vol. 33, No.4

389

ASSESSMENT OF OVULATION BY ULTRASOUND AND ESTRADIOL LEVELS

FIG. 3. Transverse echogram of the pelvis on day 17 showing two follicles (F), both of which are 2 cm in diameter, on the right ovary. B, Bladder; U, uterus. The left ovary is not seen. The E2 level was 3926 pmoleslliter.

FIG. 5. Transverse echogram ofthe pelvis on day 14 showing a 3-cm follicle (F) on the right ovary and a 2.5-cm follicle on the left ovary. B, Bladder; U, uterus. The E 2 level was 7711 pmoles/ liter.

> P > 0.001) than that in the cycles in group I in which a single follicle was present. There was no difference (P > 0.5) in follicle size. Group III. The results of the five cycles stimulated by hPG are shown in Table 3. In patient M. C., five large follicles were seen on ultrasound (Fig. 4). She also had a very high E 2 level (7802 pmoles! liter) and hence a significant risk of having a multiple pregnancy; accordingly, the ovulatory dose of hCG was not given. In comparison, however, patient P. R. also had a very high E2 level (7711 pmoles/liter), but ultrasound showed only two large follicles (Fig. 5). After the risk of a twin pregnancy was explained to the patient, she elected to receive hCG so that ovulation would occur. Pregnancy did in fact occur in this patient, but a single pregnancy was diagnosed by ultrasound on day 44.

There was no difference between pregnant and nonpregnant cycles in all groups studied, asjudged 'by peak E 2 levels and Graafian follicle diameters (P> 0.5). No conceptions have occurred in patients in whom the peak E2 level was less than 950 pmoles/ liter and the Graafian follicle size was less than 1.5 cm in diameter. Accordingly, these values have been defined as the minima necessary for successful ovulation. DISCUSSION

The advent of modem ultrasound now allows a comparison of the ovarian anatomy and its endocrine function as indicated by E 2 levels. However, E2 levels are important, and our past experience9 has indicated that in spontaneous ovulatory cycles an E2 level of approximately 1500 pmoleslliter is an accurate indicator of imminent ovulation. Appropriate E2 levels are necessary for adequate cervical mucus production and endometrial proliferation. Corpus luteum function may also be abnormal if there has been abnormal development of the Graafian follicle as indicated by low follicular phase E2 levels. 13 , 14 It was also apparent that the administration of TABLE 3. Results of the Five Cycles in Which Ovulation Was Stimulated by hPG Patient

FIG. 4. Transverse echogram of the pelvis on day 20 showing four large follicles (F) on the right ovary and a number of smaller follicles (F) on the left ovary. B, Bladder. The E 2 level was 7805 pmoleslliter.

M.e. M.S. L.G.

P.R.

No. of cycles

1 1 2 1

No. of follicles

5 1 1 2

Follicle

size

Peak E2 level

em

pmoleslliter

All >2.0 2.5 2.0 3.0,2.5

7805 1462 1024 7711

Pregnancies

0 1 0 1

390

April 1980

SMITH ET AL.

clomiphene resulted in greater ovarian stimulation, as reflected by the increased number offollicles that developed. The E2 values in the clomiphene-treated group in which only one follicle was present were also higher than the E2 values in the cycles of the control group in which one follicle was present. However, E2 levels will not distinguish between the presence of two small follicles, both producing estrogen but not ovulating until later, and a single mature follicle about to ovulate. They do not therefore give an accurate prediction of the multiple ovulatory potential of a cycle. Ultrasound, however, suggests the possibility of multiple ovulation and therefore the chance of multiple pregnancy.14 Ovulatory doses ofhCG can be withheld if the number of follicles is too great, as in the case ofM. C. in group III. Ultrasound also describes the maturity of the follicle and therefore the timing of the natural luteinizing hormone surge or the administration of hCG to mimic this surge and promote ovulation. The appearance and disappearnace of small follicles over a period of a few days has also been observed with ultrasound during the menstrual cycle, especially during nonovulatory cycles (both spontaneous and treated) and ovulation-induced cycles. The results show that ultrasound and the measurement of E2 levels are complementary investigations in the monitoring of ovulation-ultrasound because it reveals the physical properties of the Graafian follicle and the E2 level because it describes the hormonal milieu.

REFERENCES 1. Kistner RW: Use of clomiphene citrate, human chorionic gonadotropin, and human menopausal gonadotropin for induction of ovulation in the human female. Fertil Steril 17:569, 1966 2. Adams M, Cooke ID: Management of anovulation. Clin Obstet Gynaecol1:285, 1974 3. Shaaban MM, Klopper A: A study on the monitoring of gonadotropin therapy by the assay of plasma oestradiol and progesterone. J Obstet Gynaecol Br Commonw 80:783, 1973 4. Black WP, Coutts JRT, Dodson KS, Rao LGS: An assessment of urinary and plasma steroid estimations for monitoring treatment of anovulation with gonadotropins. J Obstet Gynaecol Br Commonw 81:667, 1974 5. Tredway DR, Goebelsmann V, Thorneycroft IH, Mishell DR Jr: Monitoring induction of ovulation with human menopausal gonadotropins by a rapid estrogen radioimmunoassay. Am J Obstet GynecoI120:1035, 1974 6. HackelOer BJ: Ultrasonic demonstration of follicular development. Lancet 1:941, 1978 7. ROnneberg L, Ylostalo P, JoupiUa P: Ultrasound to time insemination. Lancet 1:669, 1978 8. Renaud R, Dervain I, Macler J, Ehret C, Spira A: illtrasound monitoring of ovulation. Lancet 1:665, 1979 9. Robertson RD, Picker RH, Wilson PC, Saunders DM: The assessment of ovulation by ultrasound and plasma estradiol determination. Obstet Gynecol 54:686, 1979 10. Wu HC: Monitoring of ovulation induction. Fertil Steril 30:617, 1978 11. Muechler EK, Kohler D, Huang K: Plasma estrogen monitoring of ovulation induction with human menopausal gonadotropins. Fertil Steril 30:745, 1978 12. Brown JB, Evans JH, Adey FD, Taft HP, Townsend L: Factors involved in the induction of fertile ovulations with human gonadotropins. J Obstet Gynaecol Br Commonw 76:289, 1969 13. Lenton E, Cooke ID: Other disorders of ovulation. Clin Obstet Gynaecol1:313, 1974 14. Wentz AC: Physiologic and clinical considerations in luteal phase defects. Clin Obstet Gynecol 22:169, 1979 15. Smith DM, Picker RH, Saunders DM: Twin pregnancy suspected before implantation. Obstet Gynecol. In press