Prolonged Amenorrhea and Oral Contraceptives*†

FERTILITY AND STERILITY Copyright

©

Vol. 32, No.3, September 1979 Printed in U.SA.

1979 The American Fertility Society

PROLONGED AMENORRHEA AND ORAL CONTRACEPTIVES*t

GEORGE TOLlS, M.D., M.Sc., C.S.P.Q., F.C.P.* DOREE RUGGERE, M.D. DAVID R. POPKIN, M.D., F.RC.S. JAMES CHOW, M.D., F.RC.S. MARK E. BOYD, M.D., F.RC.S. ALBERTO DE LEON, M.D., F.RC.S. ANDRE B. LALONDE, M.D., F.RC.S. ANTOINE ASSWAD, M.D., F.RC.S.

MEYER HENDELMAN, M.D., F.RC.S. VINCENT SCALI, M.D., F.RC.S.(C) ROBERT KOBY, M.D., F.RC.S. GEORGE ARRONET, M.D., F.R.C.S. BORIS YUFE, M.D., F.RC.S. FREDERICK J. TWEEDIE, M.D., F.RC.S. PAUL R FOURNIER, M.D., F.RC.S. FREDERICK NAFTOLIN, M.D., D.PmL.§

Departments of Obstetrics and Gynecology and Medicine, McGill Faculty of Medicine and Royal Victoria Hospital, Montreal, Quebec, Canada

Of 106 consecutive women referred for secondary amenorrhea of more than 1 year's duration, 65 were diagnosed as having functional amenorrhea. Of these 65, 29 had amenorrhea directly following discontinuation of oral contraceptives (OC group) and 36 had never used oral contraceptives (NOC group). There was no difference in the incidence of prior menstrual irregularlity in either group. Similarly, there was no difference in the resting serum estrone, estradiol, luteinizing hormone, follicle-stimulating hormone, and prolactin levels between the OC and NOC groups. Nor was there a difference between the OC and NOC groups in response to medroxyprogesterone acetate, clomiphene citrate, or luteinizing hormone-releasing factor. Of 106 patients, 17 were proven to have prolactinomas. Eight patients had a prior history of OC use, whereas nine did not. With the exception of elevated serum prolactin levels, there were no significant differences in biochemical tests or history of oral contraceptive use between the prolactinoma group and patients with prolonged "functional" amenorrhea (OC plus NOC groups). The lack of historical or biochemical difference between the OC and NOC subjects indicates homogeneity between groups, and does not support the existence of a "postpill" syndrome. Fertil Steril32:265, 1979

Secondary amenorrhea may follow the discontinuation of oral contraceptive steroids (OC's). Although the occurrence of 'this phenomenon has

been reported to be more frequent among women with pre-existing menstrual irregularities, 1 it does occur as well in females with regular cycles. Mter discontinuation of OC's, menstruation returns, usually within 6 months. 2 In cases of delayed return of menses, it has been suggested that this represents a slow recovery process of the hypothalamic-pituitary-ovarian axis which has been under the long-term negative feedback of the "pill," hence the term "oversuppression syndrome."3 It is becoming clear that such delay of menstrual function involves a very heterogeneous group and is the result of many etiologies. 4 Our study deals with historical, clinical, and biochemical analyses of 82 of 106 consecutive patients re-

Received February 6, 1979; revised May 9, 1979; accepted May 10, 1979. *Supported by the Fraser Memorial Fund, Royal Victoria Hospital, and the Medical Research Council of Canada (Grant MRC-297-47, to G. T.). t Presented at the Twenty-Fifth Annual Meeting of the Society for Gynecologic Investigation, March 1978, Atlanta, Ga. *Reprint requests: George Tolis, M.D., Department of Obstetrics and Gynecology, 9 Medical, Room M.9.28, 687 Pine Avenue West, Montreal, Que., Canada H3A 1Al. §Present address: Department of Obstetrics and Gynecology, Yale University School of Medicine, New Haven, Conn.

265

266

TOLlS ET AL.

ferred for secondary amenorrhea, correlating their history of OC use with multiple endocrine tests and final diagnosis.

September 1979 TABLE 1. Oral Contraceptive History in Patients with Secondary Amenorrhea OC users NOC users

PATIENTS

Of 106 consecutive women referred for amenorrhea of more than 1 year's duration, 65 were ultimately diagnosed as having "functional" amenorrhea, 17 as having prolactinomas (surgical diagnosis), and the remaining 24 patients as having various specific etiologic factors (see Fig. 1). This report deals with the 82 patients comprising the "functional" amenorrhea and prolactinoma groups. Among the "functional" amenorrhea group, 29 of 65 developed amenorrhea following the discontinuation ofOCs without any intervening bleeding (OC group). In 35 of the 65, there was no history of OC intake (NOC group) (Table 1). The mean ages of the OC and NOC groups were 24.8 and 25.8, respectively. The groups were similar in terms of gravidity and parity. In the NOC group, 9 of36 (25%) admitted to having had irregular menses prior to amenorrhea, whereas 13 of 29 (45%) of the OC group were irregular prior to OC intake. The menstrual history of the prolactinoma group is shown in Table 2. Galactorrhea was present in 3 of29 (10%) ofthe OC group and in 5 of36 (14%) of the NOC group, without elevations in serum prolactin5 (see Table 3). The diagnosis of prolactinoma was based upon the presence of primary hyperprolactinemia (not secondary to drugs or hypothyroidism)5 of autonomous character (that is, inability of thyrotropinreleasing factor or sleep to affect prolactin levels)6 and with tomographic confirmation. Amenorrhea and galactorrhea were present. By applying these

N.O.C. Functional (361 34.0'.

Functional amenorrhea

Prolactinoma

29/65 (45%) 36/65 (55%)

8/17 (47%) 9/17 (53%)

criteria, all 17 patients reported here were found to have a prolactin-secreting tumor upon transsphenoidal exploration. Eight of the seventeen patients (47%) had a history of OC use, whereas nine (53%) had never taken OCs (Table 1). METHODS

Hormone Measurements (Table 3) Initial studies in all patients included measurements of estrone (E I ), estradiol (E 2), folliclestimulating hormone (FSH), luteinizing hormone (LH), and prolactin (PRL), using radioimmunoassay techniques previously described. 7

Provocative Testing Medroxyprogesterone Acetate Withdrawal Test. Twenty-two of thirty-six patients in the NOC group, twenty-four of twenty-nine patients in the OC group, and seven of seventeen patients with prolactin-secreting tumors were given medroxyprogesterone acetate (Provera; Upjohn Co., Kalamazoo, Mich.) (10 mg daily for 5 days). The test was positive if vaginal bleeding ensued. Clomiphene Citrate Stimulation. Five of thirtysix patients in the NOC group, five oftwenty-nine in the OC group, and six of seventeen prolactinoma patients received clomiphene citrate (Clomid; Merrell-National Laboratories, Cincinnati, Ohio). (l00 mg orally daily for 7 days). The test was positive if ovulatory menses ensued. LH-Releasing Factor Test. Pituitary responsiveness was tested by the administration of 100 J45 of LH-releasing factor (LRF; Ayerst LaboraTABLE 2. Menstrual History in Seventeen Patients with Prolactinomas

A~~:::::===::::t~varian Tumor 11J 0.9%

No prior OC (N = 9)

Phenothiazine. (211.9% Asherman·. (31 2.S%

emature Menopause\4) 3.S°;'

FIG. 1. Description of diagnoses in 106 consecutive patients seen on referral for prolonged (more than 1 year) secondary amenorrhea. P.C.o., Polycystic ovarian disease.

Prior OC eN

=

2 Irregular since menarche 7 Regular until present illness 2 Spontaneous amenorrhea 4 Persistent postpartum amenorrhea 1 Persistent postabortal amenorrhea 8) 2 Irregular since menarche 3 Regular periods followed by menstrual irregularity 1 Persistent postpartum amenorrhea following ovulation induction 3 Regular until present illness 2 Persistent postpartum amenorrhea 1 Persistent postaborlal amenorrhea

ORAL CONTRACEPTIVES AND AMENORRHEA

Vol. 32, No.3

TABLE 3. Hormone Levels in Patients with Secondary

Amenorrhea NOCgroup

OC group Range

El (ng/dl) E2 (ng/dl) FSH (~/dl) Basal Peak (LRF)a LH (~/dl) Basal Peak (LRF)a PRL (ng/ml)

Mean

Range

Mean

1.8-6.0 2.2-7.3

4.7 4.5

1.9-11.2 1.3-6.2

4.5 3.2

10.0-44.0 ·19.0-100

25.6 52.2

22.0-54.2 45.0-120

36.0 76.6

1.0-12.5 6.8-74.0 1.0-12.0

5.3 35.3 5.8

1.0-17.2 16.0-160 1.8-15.8

7.2 54.2 6.6

aThe increment in both FSH and LH after LRF administration was comparable for both OC and NOC groups (j' > 0.1).

tories, New York, N. Y.) as previously described. 7 The LRF test was performed in 27 of 36 of the NOC group and in 23 of 29 of the OC group. Statistical~e~ods

For comparison of basal hormone values and increments in LH and FSH after LRF administration between OGand NOC groups, Student's t-test for unpaired data was used. For analysis of prevalence of galactorrhea among OC and NOC groups (patients with prolactinomas are included), history of spanomenorrhea, and responsiveness to Provera, a ->t test was used. 8 RESULTS

No statistical difference was found between the OC and NOC groups with respect to previous history of spanomenorrhea (t = 1.7430) or galactorrhea (x 2 = 0.0112).

Hormone

~easurements

Resting levels of Eh E 2 , FSH, and PRL were similar in both subgroups of patients with functional amenorrhea (Table 3). Patients with prolactinomas did not differ (P :> 0.1) from those with functional amenorrhea except for serum PRL levels, which ranged from 35 to 400 ng/ml (mean 135 ng/ml) (P < 0.001).

Provocative Testing Medroxyprogesterone Acetate Withdrawal Test. Fourteen of twenty-two patients in the NOC group, ten of twenty-four patients in the OC group, and three of seven patients with prolactinomas showed a positive response to Provera. A test revealed a value of 0.9483, indicating that the groups were not statistically different.

t

267

Clomiphene Citrate Stimulation. Two offive patients in the NOC group, four of five in the OC group, and three of six patients with prolactinomas responded to Clomid. LRF Test. As seen in Table 3, there was no difference (P > 0.1) in response (peak versus basal) to LRF between the OC and NOC groups. DISCUSSION

Despite many attempts to define the effect of oral contraceptives, there has been no study which confirmed or refuted a specific and persistent effect of OC's upon reproductive performance. In the absence of prospective studies comparing the development of amenorrhea in matched popUlations of OC and NOC subjects, several retrospective approaches to the problem have been employed. The incidence of amenorrhea following OC (the socalled "postpill amenorrhea" syndrome) has been observed. Figures from 0.2% to 2.64% have appeared in the literature. 9 This large range of incidence stems from the heterogeneity of the populations studied and the length of amenorrhea specified. For example, some authors consider even one missed period as amenorrhea. Furthermore, the incidence and pattern of secondary amenorrhea in the normal population remains unclear. 10 According to some, it has been considered to be 0.7%,11 a figure remarkably similar to the estimated incidence of amenorrhea of more than 6 months' duration following OC use. 12 Other studies have shown no difference in the incidence of amenorrhea following use of OCs as related to type of regimen employed or length of use. 1, 9, 10 However; the question of whether there is a syndrome of prolonged amenorrhea related to the use of OC's recurs and warrants reappraisal. Among 65 of 106 patients with "functional" amenorrhea, I.e., otherwise undiagnosable amenorrhea lasting 1 year or longer, approximately one-half had a history ofOC use (Table 1). Comparable data have been adduced by Jacobs et aJ.1° (who, however, defined amenorrhea as the absence of menses for 4 or more months' duration). Unfortunately, as in the other studies, there are no prevalence figures for the incidence of functional amenorrhea in our NOC population or for the use of OC's in the Quebec population. Similar to our OC and NOC groups, a history of oral contraception was present in about half of all patients with prolactinomas. These 8 of 17 patients with a history of OC use had taken them for contraception or regulation of cycles; none was taking OCs at the onset of the prolonged amenorrhea. The incidence of prolac-

TOLlS ET AL.

268

tinomas in our population of amenorrheic patients was 16% and was evenly distributed among those reporting OC and NOC use. A recent report has conformed the latter finding. 13 Finally, we employed extensive endocrine testing in an attempt to characterize theNOC and OC groups. On the basis of these tests we could not separate the groups or individual patients in each group. Taken together, the.historical and biochemicaldata do not support the existence of a "postpill" syndrome which· is separable from prolonged amenorrhea occurring in the absence of oral contraceptive use. Acknowledgments. We would like to acknowledge the nursing assistance of the metabolic day center and the outpatient department clinic in the Women's Pavillion, Royal Victoria Hospital. REFERENCES 1. Mishell DR: Current status of oral contraceptive steroids. Clin Obstet Gynecol19:743, 1976 2. Rice-Wray E, Correau S, Gorodorsky J, Esquivel J, Goldzieher JW: Return of ovulation after discontinuation of oral contraceptive treatment. Fertil Steril18:212, 1967 3. Good AE, Kempers RD: Prolonged over-suppression syndrome. Med Clin North Am 58:861, 1974

September 1979 4. Jacobs HS, Knuth UA, Hull MGR, Franks S: "Postpill" . amenorrhoea-----cause or coincidence? Br MedJ 2:940, 1977 5. Tolis G: Galactorrhoea-amenorrhoea and hyperprolactinaemia: pathophysiological aspects and diagnostic tests. Clin Endocrinol (OX!) 6:915, 1977 6. Tolis G, Woods I, Matthews E, Jones M: Prolactinomas: diagostic value of biochemical markers. Clin Invest Med 1:45,1978 7. Tolis G, Solomon S, Friesen H, LewisW, VerdyM, Pagalis G, Pavlatos F, Fessas P, Rochefort GJ: Anterior pituitary .function in the Prader_Labhart-Willi-syndrome. J Clin EndocrinolMetab 39:106,1974 8. Calton T:·Statistics in medicine. Boston, Little Brown and co,1974 9. Royal College of General Practitioners: Oral Contraceptives and Health. 1974, P 74 10. Speroff L: Which birth control pill should be prescribed? Fertil Steri127:997, 1976 11. Petterson F, Friesm H; Millins SJ: .Epidemiology of secondary amenorrhea: incidence and prevalence rates. Am J Ohstet Gynecol 117:80, 1973 .12. Shearman RP, Smith ID: Statistical analysis of relationship between oral contraceptives, secondary amenorrhoea and galactorrhoea. J Obstet Gynaecol Br Commonw 79:654, 1975 13. Van Campenhout J, Blanchet P, Beauregard H, Papas S: Amenorrhea following the use of oral contraceptives. Fertil Steril 28:728, 1977