Luteal phase dysfunction in endometriosis: elevated progesterone levels in peripheral and ovarian veins during the follicular phase*

Transcription

1 FERTILITY AND STERILITY Copyrightc 1987 The American Fertility Society Printed in U.BA. Luteal phase dysfunction in endometriosis: elevated progesterone levels in peripheral and ovarian veins during the follicular phase* Jonathan W. T. Ayers, M.D. t Debra L. Birenbaum, M.D. K. M. Jiaram Menon, Ph.D. Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, University of Michigan Medical Center, Ann Arbor, Michigan Endometriosis has been associated with corpus luteum inadequacy and abnormalities of luteal phase progesterone (P) secretion. In this study, abnormal luteolysis, as a second factor of luteal dysfunction, was assessed in 13 women with endometriosis and 25 control patients by measurement of ovarian vein estradiol (E 2) and P during the follicular phase. The results reveal that women with endometriosis have (1) significantly lower ovarian vein E 2, (2) significantly higher both peripheral and ovarian vein P, and (3) threefold higher PIE 2 ratios than controls during the follicular phase. These data support the concept of continued P production from an active corpus luteum well into the follicular phase of the following cycle in women with endometriosis. Failure of adequate luteolysis is a second aspect of luteal dysfunction in endometriosis and strongly supports the growing body of data confirming ovulatory asynchrony in the minimal endometriosis infertility syndrome. Fertil Steril47:925, 1987 The exact cause of infertility in women with minimal endometriosis and normal tubal-gonadal relationships is unclear. A growing body of data, however, strongly suggests that causal relationships may be in large part due to the associated hypothalamic-pituitary-ovarian axis dyssynchrony seen in the majority of endometriosis patients. The full spectrum of ovulatory dysfunction, from anovulation 1 to normoprolactinemic galactorrhea 2 and luteinized unruptured follicle syndrome,3 has been described with markedly increased frequency in women with endometriosis and infertility. Received June 6, 1986; revised and accepted February 27, *Supported by a grant from the Max Burnell Trust Fund, McLaren General Hospital, Flint, Michigan. treprint requests: Jonathan W. T. Ayers, M.D., Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, The University of Michigan, L-221 Women's Hospital, Ann Arbor, Michigan Although not all investigators concur on a primary endocrine "lesion" characteristic of endometriosis,4-6 most studies support a strong correlation between endometriosis and abnormalities of the luteal phase. 7-9 Because the normal menstrual cycle progression offolliculogenesis, ovulation, and subsequent fertilization and implantation depend on the normalcy of the previous luteal phase,1 it is clear that the luteal phase abnormalities associated with endometriosis may play a significant role in ovulatory dysfunction and infertility. Although a great deal of information has focused on the abnormalities of corpus luteum formation in endometriosis, there is little data concerning the equally important aspects of corpus luteum demise. The current study was designed to investigate possible abnormalities ofluteolysis as a second component ofluteal phase dysfunction in women with endometriosis by assessment of ovarian secretion of estradiol (E 2 ) and progester- Vo!' 47, No.6, June 1987 Ayers et al. Abnormal luteolysis in endometriosis 925

2 one (P) through the measurement of ovarian vein levels during the follicular phase. MATERIALS AND METHODS and ovarian venous blood samples were obtained from 12 women with previously diagnosed endometriosis at the time of laparotomy performed for surgical treatment of their infertility. Endometriosis had been previously assessed by laparoscopic evaluation as stage I or II by The American Fertility Society criteria. ll and ovarian venous samples were also obtained from a control population consisting of 25 women free of endometriosis undergoing laparotomy for reversal of tubal sterilization or salpingoplasty for tubal adhesions. All women had regular biphasic basal body temperature (BBT) graphs and luteal phase lengths of 11 to 13 days. Ages of both study and control patients ranged from 19 to 35 years, and none had taken steroid active medications for at least four cycles before the laparotomy. All 12 patients with endometriosis were sampled during the follicular phase between 6 and 9 days after the onset of a regular menstrual period. Control patients were sampled between days 6 and 1 of the follicular phase (13 patients) or on days 18 to 24 of the luteal phase (12 patients) of a normal cycle. The blood samples were obtained uder direct visualization by cannulation of the right and left ovarian veins with a 25-gauge butterfly catheter before performance of the intended surgical procedures. Superficial hematoma formation in the infundibulopelvic ligament occurred in 8 of 37 women and responded to simple manual pressure alone. blood samples were obtained concomitantly from an antecubital vein. P and E2 values were determined by standard radioimmunoassay. Statistical significance was -calculated with the unpaired Student'st"test. RESULTS Tabulation of the E2, P, and P/E2 ratios obtained during the study is presented in Table 1. Ovarian vein and peripheral E2 levels from the endometriosis and both control groups, follicular phase control (FC) and luteal phase control (LC), are graphically displayed in Figure 1. E2 values were similar in both endometriosis and FC groups of women. There were major differ- 926 Ayers et al. Abnormalluteolysis in endometriosis Table 1. Tabulated E2 (pglml) and P (nglml) Values ± Standard Error of the Mean Follicular phase endo- FC LC metriosis E2 84 ± ± 5 14 ± 23a P 1.9 ±.3a.65 ±.1 12 ± 1.9a P/E 2 ratio 35 ± loa 12 ± ± 22a Left ovarian vein E2 727 ± ± ± 145 P 3.7 ±.8 3. ± ± 14 P/E 2 ratio 48 ± ± ± 1 Right ovarian vein E2 55 ± ± ± 134 P 3. ±.6.95 ± ± 59 P/E 2 ratio 21 ± ± ± 36 Left and right ovarian venous mean E2 71a P 3.2a a P/E 2 ratio 32a a asignificantly different from Fe, P <.5. ences in the range of values of all three groups between right and left ovarian vein E2 levels. This significant disparity probably reflects the random predominance of the dominant follicle or corpus luteum on one side in a relatively small population. Indeed, at laparotomy during the early follicular phase in the majority of all patients, no dominant follicle could be identified. Moreover in the endometriosis patients, multiple corpora lutea were seen on both ovaries at this early stage in the cycle. Clearly absence of a morphologic "dominant ovary" in these early follicular phase evaluations suggests that both ovaries are significantly contributory to total steroid secretory activity. Thus, meaningful comparisons between right and left ovarian vein measurements reflect physiologic rather than pathologic differences and are not significant. Comparison of the mean E2 value of the right and left ovarian veins between the groups is, however, a meaningful reflection of total ovarian E2 production. The right and left mean E2 from endometriosis patients was significantly lower than the mean E2 in the FC group (71 pg/ml versus 154 pg/ml). In contrast to E2, P levels at both the peripheral and ovarian vein sites were significantly elevated in the endometriosis patients compared with the FC group (Fig. 2). Serum P mean values were 1.9 ±.3 ng/ml during the follicular phase in women with endometriosis compared with normal P values of.65 ±.1 ng/ml in the FC group. Mean left and right ovarian vein P values were also significantly higher in endometriosis patients (3.2 ng/ Fertility and Sterility

3 ::E I/) +1 E 2... Ot Q. <5 15 :: l I/) Left Ovarian [] Right Ovarian Mean of left and right values II :;::: { 1::: Figure 1 Mean E2 levels from the three groups show significant differences of the peripheral vein systemic site. However, the mean right and left E2 value in the endometriosis patients is significantly (P <.5) less than the mean FC group value. I:!!! I.'. ti ' ' ' ' ' ' ENDOMETRIOSIS NORMAL NORMAL ml) than in the FC women (1. 7 ng/ml). Although these follicular P values in endometriosis patients were far less than either the peripheral or ovarian vein P levels of control patients in the luteal phase (LC), these elevated P values are much higher than those observed during a normal follicular phase. Evaluation of P/E 2 ratios (Fig. 3) reveals even more significant differences in the follicular phase between endometriosis and FC normal patients. Although E 2/P ratios in endometriosis patients are less than those of the normal luteal phase (LC), they are significantly higher at both the peripheral (35 versus 12) and the mean left and right ovarian vein (32. versus 1.8) sites than normal FC values. DISCUSSION Minimal endometriosis and infertility probably represent a multifactorial hypothalamic-pituitary-ovarian dysfunction syndrome resulting in ovulatory abnormalities. Cheesman et al.12 found biphasic luteinizing hormone (LH) curves in patients with endometriosis and subsequent delayed ovarian luteinization and attenuated rises in pregnanediol-3-glucuronide and estriol-16- glucuronide until after the second LH peak. Further investigations by the same group suggested that P secretion and metabolism were abnormal in endometriosis patients, resulting in a significantly shortened "functional luteal phase," which contributed to ovulatory dysfunction.13 Similar luteal phase abnormalities in women with endometriosis have been demonstrated by Brosens et ai.,14 who documented a significant delay in the post-lh surge P peak, suggesting both a disturbance in ovum release and inappropriate luteinization. Hargrove and Abraham 15 confirmed both a delay in P peak and also a significantly decreased midluteal serum Pin 14 women with endometriosis. The conclusions of these investigators argue convincingly that abnormalities in luteinization, in both magnitude and timing of P secretion, are a commonly associated finding in women with endometriosis and infertility. The data from the current study suggest that the luteal phase abnormalities documented by previous investigators extend throughout the life span of the corpus luteum to include abnormalities of luteolysis as well. With P as a marker of luteal function, this study demonstrates significantly elevated peripheral and ovarian vein P 2 I/) +I E... c: Z :: III I- en III Cl) :: Lefl Ovarian 16 [] RighI Ovarian Mean of lef I and 15 righl values _ B o ' ' ENDOMETRIOSIS ' ' NORMAL. NORMAL Figure 2 Both peripheral and mean ovarian vein P values in endometriosis are significantly increased over FC group values. Ayers et al. Abnormal luteolysis in endometriosis 927

4 Left Ovarian m Right Ovarian Mean of left and right values I S :.:.:... 1i1. O '--vJ--\ J ENDOMETRIOSIS NORMAL NORMAL Figure 3 P/E 2 ratios in endometriosis women are three times higher at the peripheral vein and two times higher at the mean ovarian vein sites than those of normal FC women. values as well as significantly decreased ovarian vein E2 levels during the follicular phase of women with endometriosis as compared with normal control women. Together these steroid secretory abnormalities seen with endometriosis synergistically increase the follicular phase P/E2 ratios toward those values seen during the luteal phase, potentially exerting major alterations in ovulatory synchrony, tubal physiology, 16 and embryo implantation. 17 This finding of elevated P is consistent with failure of adequate luteolysis and prolongation of corpus luteum function into the next cycle of follicular growth. The subsequent follicular phase asynchrony induced by elevated P in women with endometriosis may explain Ronnberg's finding of decreased LH receptors, decreased follicle-stimulating hormone-priming granulosa cells, and subsequent suppressed LH action at midcycle in women with endometriosis. 18 Further support for the delayed luteolysis hypothesis in women with endometriosis is found in the Cohen and Katz 19 study showing similar elevated follicular phase levels of peripheral P in a small group of women with endometriosis. Recently, Swolin and Skogsberg2 have described an almost pathognomonic postmenstrual follicular phase basal temperature elevation in the majority of women with endometriosis. This follicular phase BBT elevation supports our find- ings of elevated P during the follicular phase of women with endometriosis. The pathophysiology of endometriosis-associated abnormalluteolysis is certainly speculative but may involve intraovarian alterations in estrogen-induced prostaglandins-the presumed luteolytic modulator.21 Indeed, the mean ovarian vein E2 levels in the endometriosis women were 5% lower than the mean E2 values in normal women at a comparable stage of follicular development (Fig. 1). The highly significant results obtained in this investigation confirm and broaden the well-established luteal dysfunction associated with endometriosis and infertility from abnormal corpus luteum formation to aberrant corpus luteum demise. Abnormal luteolysis with maintenance of an active corpus luteum secreting significant amounts of P well into the proceeding follicular phase may play a major role in subsequent abnormalities of folliculogenesis, ovulation, and ovum transplant. Conversely, the efficacy of antigonadotropin therapy such as Danocrine (Winthrop-Breon, New York, NY) or gonadotropin-releasing hormone antagonists may lie in the ability of these agents to suppress asynchronous luteolysis and follicular phase P as well as to reduce the amount of pelvic endometriosis. Clearly, at many levels of the hypothalamicpituitary-ovarian axis, the syndrome of minimal endometriosis and infertility encompasses a wide spectrum of ovulatory dysfunction and may well be a primary neuroendocrine rather than anatomic abnormality. REFERENCES 1. Soules MR, Malinak LR, Poindexter A: Endometriosis and anovulation: a coexisting problem in infertile women. Am J Obstet Gynecol 125:412, Hirschowitz JS, Soler NG, Wortsman J: The endometriosis-galactorrhea syndrome. Lancet 1:896, Dmowski WP, Rao R, Scommegna A: The luteinized unruptured follicle syndrome and endometriosis. Fertil Steril 33:3, Muse KN, Wilson EA: How does mild endometriosis cause infertility? Fertil Steril 38:145, Holtz G, Williamson HO, Mathur RS, Landgrebe SC, Moore EE: Luteinized unruptured follicle syndrome in mild endometriosis. J Reprod Med 3:643, Haney AF, Muscato JJ, Weinberg JB: Peritoneal fluid cell populations in infertility patients. Fertil Steril 35:696, Pittaway DE, Maxson W, Daniell J, Herbert C, Wentz AC: Luteal phase defects in infertility patients with endometriosis. Fertil Steril 39:712, Ayers et al. Abnormal luteolysis in endometriosis Fertility and Sterility

5 8. Pittaway DE, Wentz AC: Endometriosis and corpus luteum function. J Reprod Med 29:712, Jones GS: The luteal phase defect. Fertil Steril 27:351, Yen SSC: The human menstrual cycle. In Reproductive Endocrinology, Edited by SSC Yen, RB Jaffe. Philadelphia, WB Saunders, 1978, p 11. The American Fertility Society: Classification of endometriosis. Fertil Steril 43:351, Cheesman KL, Ben-Nun I, Chatterton RT Jr, Cohen MR: Relationship of luteinizing hormone, pregnanediol- 3-glucuronide, and estriol-16-glucuronide in urine of infertile women with endometriosis. Fertil Steril 38:542, Cheesman KL, Cheesman SD, Chatterton RT Jr, Cohen MR: Alterations in progesterone metabolism and luteal function in infertile women with endometriosis. Fertil Steril 4:59, Brosens la, Koninckx PR, Corvelyn PA: A study of pi asrna progesterone, estradiol-17j3, prolactin and luteinizing hormone levels and the luteal phase appearance of ovaries in women with endometriosis and infertility. Br J Obstet Gynaecol 85:246, Hargrove JT, Abraham GE: Abnormal luteal function in endometriosis. Fertil Steril 34:32, Brundin J: Hormonal regulation of egg transport through the mammalian oviduct. In Progress in Infertility, Edited by SJ Behrman, RW Kistner. Boston, Little Brown & Co, 1975, p Gidley-Baird AA, O'Neill C, Sinosich MJ, Porter RN, Pike IL, Saunders DM: Failure of implantation in human in vitro fertilization and embryo transfer patients: the effects of altered progesterone/estrogen ratios in humans and mice. Fertil Steril 45:69, Ronnberg L, Kauppila A, Rajaniemi H: Luteinizing hormone receptor disorders in endometriosis. Fertil Steril 42:64, Cohen BM, Katz M: The significance of the convoluted oviduct in infertile women. J Reprod Med 21:31, Swolin K, Skogsberg K: Endometriosis and basal body temperature. Acta Obstet Gynecol Scand 64:617, Sotrel G, Helvacioglu A, Dowers S, Scommegna A, Auletta FJ: Mechanism of luteolysis: effect of estradiol and prostaglandin F2n on corpus luteum luteinizing hormone/ human chorionic gonadotropin receptors and cyclic nucleotides in the rhesus monkey. Am J Obstet Gynecol 139:134, 1981 Ayers et al. Abnormalluteolysis in endometriosis 929