Plasma Hormone Profile in Anovulation*

Vol. 31, No.3, March 1979 Printed in U.SA.

FERTILITY AND STERILITY Copyright © 1979 The American Fertility Society

PLASMA HORMONE PROFILE IN ANOVULATION*

CHUNG H. WU, M.D.t GEORGE MIKHAIL, M.D. Endocrine Section, Department of Obstetrics and Gynecology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104

Daily plasma hormones, including luteinizing hormone (LH), follicle-stimulating hormone (FSH), estrone (E I ), estradiol (E 2 ), progesterone, androstenedione, and testosterone (T), were measured in 16 anovulatory patients for a span of 3 to 4 weeks. The clinical diagnoses in this group of patients included the following: anovulation-eumenorrhea (n = 5), anovulation-polymenorrhea (n = 1), anovulation-oligomenorrhea (n = 3), congenital adrenal hyperplasia (n = 1), polycystic ovarian disease (n = 4), severe hypothalamic amenorrhea (n = 1), and postpartum amenorrhea-galactorrhea (n = 1). Follicular activity was evident in polymenorrheic and oligomenorrheic patients, and menstruation occurred in these patients following estrogen withdrawal. No follicular maturation was noted in the group of patients with anovulation-eumenorrhea, and menstruation in these patients was considered breakthrough bleeding. Low FSH levels were observed in anovulatory patients with eumenorrhea, polymenorrhea, and oligomenorrhea. Significantly high LH values were noted in both classic and non-classic polycystic ovarian disease. Extremely low E1 and E2 levels were found in patients with severe hypothalamic amenorrhea and postpartum amenorrhea-galactorrhea. Slightly elevated progesterone levels were observed in polymenorrheic and oligomenorrheic patients prior to menstruation; this was frequently associated with an LH surge or elevation. Elevated T levels were consistently associated with hirsutism but not with obesity. Fertil Steril31:258, 1979

tradiollevels during the luteal phase. Androgens such as androstenedione (A) and testosterone (T) fluctuate throughout the cycle without distinct patterns except for a slight elevation of T in midcycle. 2 • 3 In the anovulatory cycle, it has been suggested that the hormone levels are tonic throughout the cycle. 4 Urinary LH, total estrogen, and pregnanediollevels in anovulatory patients have been reported. 5 Two types of anovulatory cycles were observed: one with tonic, low estrogen and LH excretion and the other with cyclic fluctuation of total estrogen and tonic LH excretion. 5 Plasma gonadotropin,6.7 estrogen, progesterone,1 and androgen8 levels in anovulatory cycles have also been previously reported. However, a composite report

The cyclic pattern of plasma hormones during the menstrual cycle has been well defined. 1 The ovulatory cycle has been characterized by elevated follicle-stimulating hormone (FSH) levels during the early follicular phase; rising estradiol levels during the late follicular phase, followed by midcycle luteinizing hormone (LH) and FSH surges; and increased postovulatory progesterone and es-

Received September 5, 1978; revised October 30, 1978; accepted November 1, 1978. *Supported in part by National Institutes of Health Grant HD-06274 and Clinical Research Center Grant 5-MOI-RR0040. tReprint requests: Chung H. Wu, M.D., Department of Obstetrics and Gynecology, Hospital ofthe University of Pennsylvania, 3400 Spruce Street, Philadelphia, Pa. 19104.

258

Vol. 31, No.3

HORMONES IN ANOVULATION

on all of these hormones in the same anovulatory cycle has not been presented. Anovulation is still one of the major causes of infertility, yet its hormonal state has not been well defined. In the present study, daily plasma hormones, including LH, FSH, estrone (E I ), estradiol (E 2 ), progesterone (P), A, and T were measured for 3 to 4 weeks in 16 anovulatory patients. The hormone patterns so obtained were compared with normal ovulatory cycles. MATERIALS AND METHODS

Patients. Sixteen anovulatory patients were selected for study from the fertility-gynecologic endocrine practice at the Hospital of the University of Pennsylvania. For 3 to 4 weeks daily blood samples were obtained from each patient at 8 to 10 A.M. through puncture of an antecubital vein. The plasma was separated and stored at - 20° C until assay. The clinical summary of the 16 patients is presented in Table 1. Five patients had anovulationeumenorrhea, and one of these five (patient L. L.) had non-classic polycystic ovarian disease (PCO). Anovulation-polymenorrhea and congenital adrenal hyperplasia (CAH) (confirmed by elevated plasma 17 a-hydroxyprogesterone) with anovulation-oligomenorrhea were noted in one patient in each category. Three additional patients had anovulation-oligomenorrhea, and one of these three (patient D. S.) had non-classic PCO. Patients with "classic PCO" were so diagnosed because they had the classic signs and symptoms described by Stein and Leventhal,9 i.e., oligo-secondary amenorrhea, hirsutism, obesity, and infertility, plus bilaterally enlarged polycystic ovaries. Patients with "non-classic PCO" included those who did not have one or two ofthe classic clinical signs yet did have polycystic ovaries. Two patients with secondary amenorrhea were studied in each category of PCO (classic and non-classic). Two additional patients with secondary amenorrhea-anovulation had severe hypothalamic amenorrhea (patient A. B.) and postpartum amenorrhea-galactorrhea (patient J. J.). Skull x-rays of all 16 patients ruled out pituitary lesions. T 4 - and T3-uptake tests suggested normal thyroid function. PCO was diagnosed on the basis of clinical signs and laparoscopic findings ofbilaterally enlarged polycystic ovaries. Plasma hormone levels were determined by radioimmunoassay (RIA). The double-antibody RIA technique was used to measure FSH and

259

LH.lO· 11 Steroid hormones, including E I , E 2 , P, A, and T, respectively, were measured by RIA as previously reported. 12- 14 On the basis of menstrual history, the patients were divided into three groups, namely those with eumenorrhea, polymenorrhea-oligomenorrhea, and secondary amenorrhea. Controls. Hormone profiles during the normal menstrual cycle were summarized from the daily plasma hormone levels of eight patients with normal ovulation. 3These data were used as controls in the present study. Since daily plasma hormone levels fluctuate significantly during both ovulatory and anovulatory cyles, comparison of individual or random values of plasma hormone levels tends to mislead. We therefore obtained mean values of daily plasma hormone levels for a span of 3 to 4 weeks and considered the mean values as representative ofthe hormone status for that individual in the comparison of ovulatory and anovulatory cycles. The values over two normal cycles were averaged to obtain the control values for the follicular and luteal phases. Student's t-test was used to compare mean hormone levels of patients and controls. RESULTS

Eumenorrhea. Three of five patients with anovulation-eumenorrhea had similar plasma hormone profiles. As represented by patient A. J. in Figure 1, all had gonadotropin and steroid hormone levels fluctuating within the normal range without any significant trend. Estradiol remained tonic throughout the cycle, suggesting its absence or slow follicular maturation in the ovary. Progesterone levels occasionally reached 2 ng/ml following some LH spikes, suggesting some luteinization of the follicular component without actual ovulation. Two other patients (patients D. H. and L. L.) with anovulation-eumenorrhea had elevated LH levels with occasional LH spikes (Figs. 2 and 3). Patient L. L. also had elevated A and T levels and was hirsute. The remaining hormones in these two patients showed tonic fluctuation within the normal range. Prior to menstrual bleeding, all five patients failed to show any increase or withdrawal in plasma estrogen. Thus, the menstrual bleeding in these patients was considered breakthrough bleeding rather than withdrawal bleeding. Polymenorrhea-Oligomenorrhea. In the polymenorrheic patient (patient B. S., Fig. 4), low FSH and normal LH levels were noted during the early

TABLE 1. Clinical Summary of the Sixteen Anovulatory Patients Patient

Age

Menarche and parity

Menstrual pattern

P-WD bleeding"

Hirsutismb

Clinical diagnosis'·

t-:l 0)

0

21

13 yr, P-O

q 25-29 days

No

172/58

Anovulation-eumenorrhea

J. R.

23

11 yr, P-O

q 26--29 days

No

169/50

Anovulation-eumenorrhea

E. R.

17

12 yr, P-O

q 23-30 days

No

161/53

Anovulation-eumenorrhea

D.H.

23

13 yr, P-O

q 26--28 days

No

164/59

L.L.

26

16 yr, P-O

q 23-28 days

Yes (++)

184/58

Anovulation-eumenorrhea; ? early PCO Anovulation-eumenorrhea; non-classic PCO

20

14 yr, P-O

q 12-16 days for 6 mo

No

174/54

Polymenorrheaoligomenorrhea B.S.

22

11 yr, P-O

q 2-5 mo

Yes

No

161/48

C. R.

27

12 yr, P-O

q 2-8 mo

Yes

No

169/57

D.S.

19

15 yr, P-O

q 4-9 mo

Yes

Yes (+)

164/50

S. P.

24

14 yr, P-O

q 2-3 mo

Yes

Yes (++)

163/59

31

9 yr, P-1, Ab-1

Amenorrhea for 2 yr

Yes

No

161/51

Non-classic PCO

R. B.

17

14 yr, P-O

Yes

Yes (+++)

160/57

Non-classic PCO

C. C.

22

13 yr, P-O

Yes

Yes (+)

170/113

Classic PCO

M.F.

28

13 yr, P-O

Yes

Yes (+++)

165/95

Classic PCO

A.B.

31

13 yr, P-O

No

No

166/48

J. J.

31

14 yr, P-1

Amenorrhea for 1 yr Amenorrhea for 2 yr Amenorrhea for 2 yr Amenorrhea for 2 yr Postpartum amenorrhea for 1 yr

No

No

165/54

Severe hypothalamic amenorrhea Amenorrhea-galactorrhea

ap_ WD bleeding, progesterone withdrawal bleeding. bHirsutism: +, mild; ++, moderate; +++, severe. epCO, Polycystic ovarian disease; CAH, congenital adrenal hyperplasia.

Remained eumenorrheic but occasionally ovulated Originally polymenorrheic for 3 mo prior to study; following study, took oral contraceptives for birth control Originally polymenorrheic fl:!r 2 mo prior to study; following study, remained eumenorrheic Remained eumenorrheic; took oral contraceptives for birth control Clomid treatment; ovulated and conceived; delivered at term; postpartum, continued eumenorrhea and anovulation

Anovulation-polymenorrhea Oral contraceptives to control polymenorrhea; right ovarian fibrothecoma removed 6 mo later; returned to ovulatory cycles after surgery Anovulation-oligomenorrhea Recovered to eumenorrhea WIth occasional ovulation Anovulation-oligomenorrhea Remained oligomenorrheic but occasionally recovered to eumenorrhea Anovulation-oligomenorrhea; Gradually recovered to eumenorrhea and spontaneous ovulation non-classic PCO Clomid treatment; ovulated and conceived twice; healthy term deCAH liveries

C.A.

Secondary amenorrhea F.P.

Follow-up

em/kg

yr

Eumenorrhea A. J.

Ht.lwt.

Secondary amenorrhea persists and withdrawal bleeding occurs after progesterone injections Ovarian suppression treatment; hirsutism markedly improved Ovarian suppression treatment; hirsutism improved Clomid treatment; ovulated on higher dose but no conception Improved by estrogen supplement Clomid-hCG treatment; ovulated on higher dose but no conception

~

~

!I

~

~

~ ;:,N (0

~

r HORMONES IN ANOVULATION

Vol. 31, No.3

E

"

CI

a.

f

200~ 150 100 50

E

"CI c:

E

"CI a. o

5 10 15 20 DAY OF OBSERVATION

25

FIG. 1. Daily plasma hormones in patient A. J. with anovulation-eumenorrhea.

phase of the cycle. Estradiol levels rose progressively during the late follicular phase yet failed to initiate an adequate LH surge, and ovulation did not occur. A slightly elevated progesterone level at this time suggested follicular luteinization without ovulation. Precipitously declining E2 and P levels induced menstrual bleeding; this cyclic phenomenon was repeated every 2 weeks. Oligomenorrheic patients, regardless of the diagnosis ofPCO, CAH, or simple oligomenorrhea-anovulation, showed similar plasma hormone profiles. As represented by patient D. S. (Fig. 5), there was an estradiol rise followed by an LH surge and a subsequent E2 decline prior to each menstrual flow. FSH levels tended to be low in these patients; LH levels were occasionally elevated in non-classic PCO (patient D. S.), but were within the normal range in CAH (patient S. P.) and in simple anovulation-oligomenorrhea (patients C. R. and C. A.). A slight elevation ofprogesterone following the LH surge also suggested follicular luteinization. Following menstrual bleeding, FSH levels gradually rose whereas estradiol

261

stayed at very low levels (less than 20 pg/ml) for a long period of time. Increased T levels were found in PCO and CAH, and elevated A and P levels were also noted in CAH. In both polymenorrheic and oligomenorrheic patients, menstrual bleeding occurred following estradiol withdrawal, and there was evidence offollicular maturation in every case prior to menstruation. Secondary Amenorrhea. In this group of patients with anovulation-secondary amenorrhea, there were two patients (patients F. P. and R. B.) with non-classic PCO, two patients (patients C. C. and M. F.) with classic PCO, one patient (patient A. B.) with severe hypothalamic amenorrhea, and one patient (patient J. J.) with postpartum amenorrhea-galactorrhea. Patient J. J. had persistently elevated plasma prolactin (PRL) levels (50 to 305 ng/ml) without evidence of pituitary tumor. All PCO patients (with both non-classic and classic disease) frequently showed elevated LH levels and normal FSH levels. In non-classic PCO without hirsutism or obesity (patientF. P., Fig. 6), normal androgen levels were noted, whereas in a severely hirsute but not obese patient (patient R. B.), highly elevated T (1539 ± D.H.

~rt~

E -....

0> C.

300

100

o

M

M

5

10 15 20 DAY OF OBSERVATION

25

30

FIG. 2. Daily plasma hormones in patient D. H. with anovulation-eumenorrhea.

March 1979

WU AND MIKHAIL

262

L.L.

~ r~: FSH

~[ o~.'·~ - rOO Eo

ir50-0~ -~ flOO E2 50-_ ____ _ .I\~

go~

~EI.OE.~

~2.0~

c:

~

'\.

1.5 1.0

c:

A.h4

~-

1\ ~ ~V.f ~

'z! 'rI

0.5 I

I

I

I

I

~t:'~ 400

t

iMmzn", 0

0.

5

0

0

I M!Z2Z7'>..

10. 15 20. DAY OF OBSERVATION

25

30

FIG. 3. Daily plasma hormones in patient L. L. with anovulation-eumenorrhea.

6

5

10. 15 20. DAY OF o.BSERVATlo.N

25

30.

FIG. 5. Daily plasma hormones in patient D. S. with anovulation-oligomenorrhea. F. P.

~~t'H~as. ~lo.~SH ~ 5 -E

E 20.0.f

~ _

ICC

.€ '" Q.

E

--1.

L-~~~~~~~-

30.0.~ ~ 20.0. ICCL -_ _ _ _ _ _ _ _ _ _ _ _ _ __

_ 6~ p

~;~

-

~ 1.5~

A

'" 1.0. <: 0.5 L -_ _ _ _ _ _ _ _ _ _ _ _ _ __

¥=fA

25

30

DAY o.F o.BSERVATlo.N

FIG. 4. Daily plasma hormones in patient B. S. with anovulation-polymenorrhea.

b

~

Ib

15

DAY OF OBSERVATION

20.

30.

FIG. 6. Daily plasma hormones in patient F. P. with nonclassic peo.

,

HORMONES IN ANOVULATION

Vol. 31, No.3 M.F

o DAY OF OBSERVATION

FIG. 7.

pea.

263

However, patient J. J. had elevated prolactin and galactorrhea and patient A. B. did not. Mean Values of Daily Plasma Hormones. The mean values of all hormones from these 16 patients are summarized and analyzed in Tables 2 and 3. Significantly high mean LH values were found in patients with classic PCO (patients C. C. and M. F.) and non-classic PCO (patients L. L., D. S., F. P., and R. B.). Patient D. H. also had elevated LH values which led us to suspect possible early PCO. Low LH values were noted in severe hypothalamic amenorrhea (patient A. B.), and slightly lower values in one patient (patient A. J.) with anovulation-eumenorrhea. Very low mean FSH values were observed in polymenorrheic and oligomenorrheic patients as well as in most of the anovulatory-eumenorrheic patients. However, no significant differences in FSH values were noted in severe hypothalamic amenorrhea, postpartum amenorrhea-galactorrhea, and in patients with PCO-oligomenorrhea, or PCO-secondary amenorrhea. Progesterone values were elevated in patients having some follicular activity followed by possible luteinization without ovulation. In contrast, patients without ovarian follicular activity

Daily plasma hormones in patient M. F. with classic

92 pg/ml) and A (2059 ± 76 pg/ml) levels were found. Estrogen and progesterone fluctuated within the normal range without suggestion of follicular activity or luteinization of the follicular component. In both cases of classic PCO, as represented by patient M. F. in Figure 7, elevated T and A levels were noted throughout the observation period, whereas E 1 , E 2, and P levels were similar to those in patients with non-classic PCO. In patients with severe hypothalamic amenorrhea (patient A. B., Fig. 8), low LH and midnormal levels ofFSH were noted. Extremely low or nondetectable levels ofE 1 and E2 were found in this case. This is compatible with the patient's complaint of dyspareunia and the failure to induce withdrawal bleeding following progesterone injection. Androgens fluctuated within the normal range, while progesterone stayed persistently less than 0.5 ng/ml. In patient J. J., with postpartum amenorrhea-galactorrhea, a hormonal profile similar to that of the patient with severe hypothalamic amenorrhea was found. Clinically, she also had hypoestrogenic signs such as dyspareunia and failure of progesterone withdrawal bleeding.

A.B.

E ...... 2 E

E ......

g

40l [300 200 100

b

DAY OF OBSERVATION

FIG. 8. Daily plasma hormones in patient A. B. with severe hypothalamic amenorrhea.

264

March 1979

WU AND MIKHAIL

TABLE 2. Daily Plasma LH, FSH, and Progesterone in Anovulatory Patients· Patient

No. of specimens

ill

FSH

Progesterone

mIUlml

mIUlml

ngiml

Eumenorrhea A.J. J.R. E. R. D.H. L.L.

24 30 23 28 21

5.3 ± 7.1 ± 4.2 ± 26 ± 76 ±

Polymenorrheaoligomenorrhea B. S. C.A. C.R. D. S. S. P.

26 28 27 28 27

6.7 ± 5.4 ± 7.6 ± 22 ± 6.5 ±

0.7 1.1 1.1

27 27 31 24 21 26

48 ± 19 ± 17 ± 30 ± 3.7 ± 6.6±

4.2C 1.2c 1.7c 1.6c O.4 c

28 28

7.3 ± 0.7 4.9± 0.6b

Secondary amenorrhea F.P.

R.B. C.C. M.F. A.B. J. J. Normal cycles Follicular phase Luteal phase

0.5 b

1.2 0.9 d 2.3 C 6.9C

5d

1.0

0.6

3.1 ± 2.3 ± 4.7 ± 2.6 ± 4.8±

0.6 c O.IC 0.8 b 0.5 c 0.6 d

0.87 ± 0.62 ± 0.82 ± 0.57 ± 0.35 ±

2.0 ± 1.4 ± 5.9± 9.1 ± 1.5 ±

0.4c 0.2c

0.20c 0.10c 0.12b

0.2c

2.11 ± 1.29± 0.98 ± 0.64± 1.98 ±

7.8 ± 8.4 ± 6.0 ± 9.2 ± 7.8± 8.9 ±

0.7 0.7 0.6 0.5 0.8 0.6

0.23 ± 0.49± 0.13 ± 0.55 ± 0.25± 0.52±

0.03 c 0.05b 0.02 c 0.03 b 0.02 d

0.9 1.2

7.5 ± 0.8 4.3 ± 0.9b

0.15 0.07 0.05b

0.05 0.04 c

0.10 0.22c

0.06

0.66 ± 0.04 7.1 ± 0.8 c

·Values are means ± standard error.

bp < 0.05 as compared with mean follicular phase value of normal cycles. cp < 0.001 as compared with mean follicular phase value of normal cycles. dp < 0.01 as compared with mean follicular phase value of normal cycles.

showed lower P values. Patients with CAH, as expected, had elevated mean progesterone values. Significantly high E2 mean values were observed in anovulation-polymenorrhea (patient B. S.), and extremely low E2 values were found in severe hypothalamic amenorrhea (patient A. B.) and in postpartum amenorrhea-galactorrhea (patient J. J .). Some differences in E2 values were seen in the remaining patients, as is shown in Table 3. High E1 values were also found in anovulationpolymenorrhea and in one patient (patient A. J.) with anovulation-eumenorrhea. Extremely low E1 levels were also noted in patients A. B. and J. J. In most patients with non-classic PCO (patients L. L., F. P., and R. B.) and in polymenorrheic patients, significant differences (either high or low) were found between mean values of A. However, in patients with classic PCO (patients C. C. and M. F.), A levels were not significantly high. Testosterone values were increased in all patients with classic and non-classic PCO except patient F. P., who also did not have hirsutism. The patient with CAH also had increased mean T values. In contrast, lower mean T values were observed in anovulation-eumenorrhea and in one of the anovula-

tory-oligomenorrheic patients (patient C. A.). In severe hypothalamic amenorrhea, the T level was decreased, but it was normal in postpartum amenorrhea-galactorrhea.

DISCUSSION

The incidence of anovulatory cycles in eumenorrheic females has been estimated to be 11% .15 Our follow-up observation of patient A. J. revealed occasionally ovulatory cycles, suggesting that anovulation-eumenorrhea-especially if plasma hormones are within the normal range-may be a transient phenomenon possibly resulting from mild hypothalamic dysfunction, and easily reversible to ovulatory cycles. However, the elevated LH and/or T seen in patients D. H. and L. L. may suggest more advanced dysfunction, difficult to reverse to normal ovulatory cycles. Uterine breakthrough bleeding has usually been related to a high level of unopposed estrogen; however, this was not the case in our study. On the contrary, breakthrough bleeding was observed even with normal or low E2 levels.

HORMONES IN ANOVULATION

Vol. 31, No.3

265

TABLE 3. Daily Plasma Estrogens and Androgens in Anovulatory Patients a Patient

No. of specimens

Estrone

Estradiol

Androstenedione

Testosterone

pg/ml

pg/ml

pg/ml

pg/ml

Eumenorrhea A.J. J. R. E.R. D.H. L. L.

24 30 23 28 21

135 ± 52 ± 35 ± 39 ± 40 ±

8.0 b 3.6 2.5 d 3.8 2.3

50 ± 50 ± 35 ± 38 ± 37 ±

5.7 6.5 5.0 c 3.4c 2.6 c

1200 1521 1055 1200 1621

Polymenorrheaoligomenorrhea B. S. C.A. C. R. D. S. S. P.

26 28 27 28 27

80 ± 54 ± 71 ± 55 ± 24 ±

9.0c 4.3 14 7 4.5 b

134 ± 56 ± 28 ± 58 ± 36 ±

14b 7.7 5b 12 7.0d

Secondary amenorrhea F.P. R. B. C. C. M.F. A.B. J. J.

27 27 31 24 21 26

37 ± 41 ± 28 ± 117± 7.1 ± 14 ±

5.4 4.4 2.8 b

37 ± 93 ± 42 ± 58 ± 8.8 ± 16 ±

2.1c 7.2c 2.7d 7 1.6b 3b

Normal cycles Follicular phase Luteal phase

28 28

11 b 1.9b 3b

49 ± 5.3 63 ± 6.8

62 ± 7.0 81 ± 5.3 d

72 92 30 c 71 56 c

276 ± 318 ± 176 ± 344 ± 833 ±

22b 13 c 19b 16 d 39b

765 ± 1660 ± 803 ± 1297 ± 2711 ±

32 b 64 c 38 b 55 13P

358 ± 150 ± 434 ± 559 ± 853 ±

24 5.5 b 36 36 C 32 b

958 ± 2059 ± 1421 ± 1405 ± 1200 ± 1263 ±

39 b 76 b 66 58 35 49

415 ± 1539 ± 525 ± 1187 ± 314 ± 439 ±

18 92 b 34 d 75 b 19c 22

± ± ± ± ±

1344 ± 86 1326 ± 102

430 ± 32 387 ± 18

aValues are means ± standard error.

bp < 0.001 as compared with mean follicular phase value of normal cycles. cp < 0.01 as compared with mean follicular phase value of normal cycles. dp < 0.05 as compared with mean follicular phase value of normal cycles.

PCO in eumenorrheic patients is rare but has been reported,t6 although a patient such as D. H., with possible early PCO whose only abnormality was elevated LH, has not been described. The etiology of PCO is still unclear, but it has been attributed to either hypothalamic-pituitary dysfunction 17 or an ovarian enzyme defect. 18 N onetheless, the elevated LH and relatively low E2 levels observed in patient D. H. also suggest early PCO. Anovulatory-eumenorrheic patients with evidence offollicular maturation 7,8 or without follicular maturation yet with midcycle LH surges 7 have been previously reported. It thus appears that a variety of hormonal profiles may be expected in such patients. In the anovulatory-polymenorrheic patient (patient B. S.), the plasma hormone profile paralleled the urinary hormone data, as previously observed. 6Low FSH levels during the early follicular phase of polymenorrheic patients may lead to inadequate follicular maturation, and the subsequent inadequate midcycle LH surge fails to initiate ovulation. These hormonal findings are similar to those found in luteal phase defect,t9 except that the latter induces abnormal ovulation.

Oligo-ovulation has been observed in oligomenorrheic patients, yet these patients tend to develop luteal phase defect. 19 Perhaps the low FSH levels during the follicular phase, as observed in this study, induce abnormal follicular maturation or require longer periods of stimulation (2 to 5 months) to initiate follicular maturation. Subsequently, with an inadequate (or even with an adequate) LH surge, ovulation fails to occur. If ovulation occurs, it tends to be abnormal, thus it leads to a luteal phase defect. Follicular luteinization seems to be a common phenomenon in these oligomenorrheic-anovulatory cases, as evidenced by the slightly elevated P levels consistently observed in this study. It has been a common belief that menstrual bleeding in oligomenorrheic patients occurs because of estrogen breakthrough bleeding. However, the data from this study fail to support this concept, suggesting rather that menstrual flow in these patients usually follows estrogen withdrawal. Random plasma LH and T measurements failed to show consistent elevation of these hormones in PCO.20 Our finding of great fluctuation of these

WU AND MIKHAIL

266

two hormones, and their frequent crossing of the upper limit of normal levels, may explain the previous observation. However, high E1 levels 20 and elevated A levels20 , 21 in peo were not consistently observed in this study. In severe hypothalamic amenorrhea and amenorrhea-galactorrhea, severe hypothalamic dysfunction or failure may lead to insufficient LH release. Subsequently, ovarian steroidogenesis decreases and the plasma estrogen level remains very low. Low estrogen levels, especially E 2, cause negative feedback at the hypothalamic-pituitary axis, releasing FSH so that plasma FSH levels tend to remain at the midnormal range. This finding concurs with a previous report22 and resembles findings in the prepubertal female. 23 The extremely low estrogen levels in these patients explain their hypo estrogenic signs and hypoplastic or inactive ovaries. The elevated PRL levels in a patient with amenorrhea-galactorrhea (patient J. J.) without pituitary tumor may also be explained by suppression of the PRL-inhibiting factor owing to severe hypothalamic dysfunction. Direct inhibition by PRL of follicular steroidogenesis in the ovary has been reported. 24 Thus, in amenorrhea-galactorrhea, both low LH and high PRL levels may synergistically accentuate suppression of follicular maturation so that the ovary becomes hypoactive and inefficient in steroidogenesis. Suppression of elevated PRL levels frequently leads to recovery of ovulation,25 suggesting that PRL alone may also suppress follicular maturation or interfere with gonadotropins at the ovarian level.

Acknowledgments. The materials for the radioimmunoassay ofLH and FSH were generously supplied by the National Pituitary Agency and the Endocrinology Study Section, National Institutes of Health. The technical assistance of Mr. Robert Krill and Mrs. Judith Wang is gratefully acknowledged.

REFERENCES 1. Vande Wiele RL, Bogumil J, Dyrenfurth I, Ferin M,

Jewelewicz R, Warren M, Rizkallah T, Mikhail G: Mechanisms regulating the menstrual cycle in women. Recent Prog Horm Res 26:63, 1970 2. Judd LL, Yen SSC: Serum androstenedione and testosterone levels during the menstrual cycle. J Clin Endocrinol Metab 36:475, 1973 3. Wu CH: Plasma hormones in clomiphene citrate therapy. Obstet GynecoI49:443, 1977 4. SperoffL, Glass RH, Kase NG: Clinical Gynecologic Endocrinology and Infertility, Second Edition. Baltimore, Williams & Wilkins Co, 1978, p 129

March 1979

5. Newton J: Hormone excretion patterns in anovulatory infertility. Postgrad Med J 48:5, 1972 6. Fraser IS, Michie EA, Wide L, Baird DT: Pituitary gonadotropins and ovarian function in adolescent dysfunctional uterine bleeding. J Clin Endocrinol Metab 37:407, 1973 7. Aksel S, Wiebe RH, Tyson JE, Jones GS: Hormonal findings associated with aluteal cycles. Obstet GynecoI48:598, 1976 8. Dawood MY, Saxena BB: Plasma testosterone and dihydrotestosterone in ovulatory and anovulatory cycles. Am J Obstet Gynecol 126:430, 1976 9. Stein IF Sr, Leventhal ML: Amenorrhea associated with bilateral polycystic ovaries. Am J Obstet Gynecol 29:181, 1935 10. Midgley AR: Radioimmunoassay for human folliclestimulating hormone. J Clin Endocrinol Metab 27:295, 1967 11. Midgley AR: Radioimmunoassay: a method for human chorionic gonadotropin and human luteinizing hormone. Endocrinology 79:10, 1966 12. Wu CH, Lundy LE: Radioimmunoassay of plasma estrogens. Steroids 18:91, 1971 13. DeVilla GO Jr, Roberts K, Wiest WG, Mikhail G, Flickinger G: A specific radioimmunoassay of plasma progesterone. J Clin Endocrinol Metab 35:458, 1972 14. Wu CH, Blasco L, Flickinger GL, Mikhail G: Ovarian function in the preovulatory rabbit. BioI Reprod 17:304, 1977 15. Cunningham GR, Caperton EM Jr, Goldzieher JW: Antiovulatory activity of synthetic corticoids. J Clin Endocrinol Metab 40:265, 1975 16. Raj SG, Thompson IE, Berger MJ, Taymor ML: Clinical aspects of the polycystic ovary syndrome. Obstet Gynecol 49:552, 1977 17. Berger MJ: Conceptual advances in the polycystic ovary syndrome. In Progress in Gynecology, Vol VI, Edited by ML Taymor, TH Green. New York, Grune and Stratton, 1975, p 237 18. Goldzieher JW: Polycystic ovarian disease. In Progress in Infertility, Second Edition, Edited by SJ Behrman, RW Kistner. Boston, Little, Brown and Co, 1975, p 325 19. Sherman BM, Korenman SG: Measurement of plasma LH, FSH, estradiol and progesterone in disorders of the human menstrual cycle: the inadequate luteal phase. J Clin Endocrinol Metab 39:145,1974 20. DeVane GW, Czekala NM, Judd HL, Yen SSC: Circulating gonadotropins, estrogens and androgens in polycystic ovarian disease. Am J Obstet Gynecol 121:496, 1975 21. Givens JR, Andersen RN, Umstot ES, Wiser WL: Clinical findings and hormonal responses in patients with polycystic ovarian disease with normal versus elevated LH levels. Obstet Gynecol 47:388, 1976 22. Kletzky OA, Davajan V, Nakamura RM, Mishell DR Jr: Classification of secondary amenorrhea based on distinct hormonal patterns. J Clin Endocrinol Metab 41:660,1975 23. Winter JSD, Faiman C: The development of cyclic pituitary gonadal function in adolescent female. J Clin Endocrinol Metab 37:714, 1973 24. McNatty KP, Sawers RS, McNeilly AS: A possible role for prolactin in control of steroid secretion by the human Graafian follicle. Nature 250:653, 1974 25. Yuen BH, Keye WR Jr, Jaffe RB: Human prolactin: secretion, regulation and pathophysiology. Obstet Gynecol Survey 8:527, 1973