Premature plasma progesterone and androgen elevation are not prevented by adrenal suppression in in vitro fertilization

Transcription

1 FERTILITY AND STERILITY '~' Vol. 67, No. 1. January 1997 Cnpyrighi ' 1997 An'~eric,'m Society filr Repr,~duclive Medicine Printed on acid-free paper in U. S. A. Premature plasma progesterone and androgen elevation are not prevented by adrenal suppression in in vitro fertilization Renato Fanchin, M.D.*? Claudia Righini, M.D.* Frangois Olivennes, M.D.* Jo~lle Taieb, Ph.D.S Dominique de Ziegler, M.D.* Ren~ Frydman, M.D.* Hdpital Antoine Bdclere, Clamart, France Objective: To investigate the effects of adrenal suppression with dexamethasone (DEX) on P and androgen profiles during controlled ovarian hyperstimulation (COH) for IVF-ET. Design: Prospective controlled trial. Setting: In vitro fertilization program, Clamart, France. Patient(s): One hundred twenty IVF-ET candidates aged 25 to 39 years undergoing 120 COH cycles. Intervention(s): Group A: 60 women received a time-release GnRH agonist (GnRH-a) on cycle day 2. After pituitary desensitization was confirmed, 1 mg/d DEX was administered daily until hcg. Ovarian stimulation with hmg was started on the 7th day of DEX. Group B: 60 other women received an identical treatment except DEX was omitted. Main Outcome Measure(s): Plasma E.,, P, androstenedione (A), and T were measured 18 days after GnRH-a, on the 7th day of DEX in group A, and on the day of hcg. Result(s): Peak plasma E._, levels were similar in groups A and B. After GnRH-a, P and androgen levels were low in both groups. In group A, an additional decrease in these hormones was observed after 7 days of DEX. During COH, we observed similar absolute increases of P, A, and T in groups A (+0.52, +1.56, and ng/ml, respectively) and B (+0.55, +1.66, and ng/ml, respectivelyl. Conclusion(s): As expected, DEX lowered additionally P, A, and T levels from values achieved after GnRH-a alone, reflecting the adrenal contribution to the circulating levels of these hormones at baseline. Yet, the net increases in P, A, and T seen in COH were unaltered by DEX treatment, indicating that this phenomenon results solely from an effect of exogenous gonadotropins on the ovary. Fel~il Steril ~ 1997;67:115-9 Key Words: Progesterone, androgens, testosterone, androstenedione, controlled ovarian hyperstimulation, in vitro fertilization The physiopathological mechanisms implicated in premature P elevation occurring sporadically during COH in pituitary-desensitized IVF-ET cycles are puzzling (1-3). Numerous reports have confirmed that, in a sizable fraction of IVF-ET candidates (ranging from 10% to 30%}, P levels exceed an arbitrary mark of 0.9 ng/ml (conversion factor to SI unit, Received April 15, 1996; revised and accepted August 9, * Department of Obstetrics and Gynecology and Reproductive Endocrinology. + Reprint requests: Renato Fanchin, M.D., Department of Obstetrics and Gynecology, H6pital Antoine B~cl~re, 157, rue de la Porte de Trivaux, 92141, Clamart, France (FAX: ). $ Department of Biochemistry. 3.18) on the day of hcg administration. Although the consequences of this phenomenon on IVF-ET outcome have been matter of controversy, the fecundity of these cycles has been described as lower by the majority of reports (1-3), probably because of an altered endometrial receptivity (3, 4). As immunoreactive and bioactive levels of endogenous gonadotropins are uniformly low during COH when a GnRH agonist (GnRH-a) is used (5), we hypothesized a possible role of exogenous gonadotropins in the genesis of premature plasma P elevation. Supporting this view, we have evidenced recently that, in women undergoing COH with pituitary desensitization by GnRH-a, there is a progressive and significant increase in plasma P and androgens, culminating approximately 12 to 15 hours after the administration Vol. 67, No. 1, January 1997 Fanchin et al. Adrenal suppression during COH for IVF-ET 115

2 of hmg (6). Yet, in our first study (6), the peak in plasma P and androgens, occurring approximately 12 to 15 hours after the evening injection of hmg, took place in the morning, coincidentally to the circadian increase in ACTH and adrenal activity. Hence, we needed to rule out the possibility that the elevations in plasma P and androgens during the follicular phase of COH reflected the circadian increase in adrenal activity. For this, we compared P and androgen profiles in women undergoing COH with pituitary desensitization with GnRH-a with' and without adrenal suppression. MATERIALS AND METHODS Patient Characteristics We studied prospectively 120 consecutive COH cycles undertaken in 120 IVF candidates 25 to 39 years of age. Clinical indicatidns for IVF-ET were tubal factor (43%), male factor (32%), idiopathic infertility (17%), and endometriosis (8%). Body mass index (weight/height 2 100) was <27 in all patients. A regular ovulatory status was observed in 78% of cases. Patients suffering from chronic anovulation caused by peripheral endocrine disorders such as polycystic ovary syndrome or adrenal dysfunction were excluded from the protocol. Protocol for COH and Dexamethasone Treatment All patients enrolled in the study underwent COH as follows: a single injection of a time-release GnRHa, (3.75 mg IM Enanthone; Takeda Pharmaceuticals, Paris, France) was administered on the 2nd day of the cycle. Eighteen days later, complete pituitary desensitization was confirmed by plasma E2 < 50 pg/ml (conversion factor to SI unit, 3.671) and plasma LH levels <2.0 miu/ml (conversion factor to SI unit, 1.00). Patients had an ultrasound (US) to exclude ovarian cysts and to verify that endometrial thickness was <5 mm. On the same day, patients were assigned alternately to receive dexamethasone (DEX, Decadron; Merck Pharmaceuticals, Paris, France), 1 mg orally at bedtime (7), until the day of ET (group A, n = 60), or served as controls (group B, n = 60), according to the arrival order at the Institution. Human menopausal gonadotropin therapy (Humegon; Organon Laboratories, Saint Denis, France) was initiated on the 24th day after GnRH-a (7th day after confirming pituitary desensitization in all patients) at a dose set at 300 IU/d for the first 5 days of COH. All hmg injections were performed at approximately 6:00 P.M. Further hmg doses and the timing of hcg (10,000 IU IM Gonadotrophine Chorionique "Endo;" Organon Laboratories, Saint-Denis, France) admin- istration were decided according to the usual criteria of follicular maturation determined by US and plasma E2 levels. Patients received hcg when at least three follicles exceeded 17 mm in diameter and E.~ levels per mature follicle (>-17 mm in diameter) were >300 pg/ml. Oocyte retrieval was performed 35 hours after hcg administration by transvaginal US-guided aspiration. Follicles measuring < 12 mm in diameter were not aspirated. Luteal phase was supported with 300 mg of micronized P (Utrogestan; Besins-Iscovesco Pharmaceuticals, Paris, France) administered daily (100 mg in the morning, 200 mg in the evening) by vaginal route starting on the day of ET. Blood Sampling and Hormone Measurement Blood samples obtained on the 18th (baseline) and 24th day after GnRH-a administration (7th day of DEX administration) for patients included in group A (after DEX), and on the day ofhcg administration (after hmg) at 8:00 A.M. were assayed for E2, P, androstenedione (A), and T. Plasma E2 was determined by a immunometric technique using an Estradiol-60 Amerlite kit (Kodak Clinical Diagnostics, Ltd, Strasbourg, France). Sensitivity, calculated as the 95c~ coefficient of variation (CV) of repeated blank measurements, was 14 pg/ml for E2. Intra-assay and interassay CVs were 8c~, and 9% for E2, respectively. Plasma P was measured by RIA using a 112'~ Progesterone Coatria kit (Bio-Mdrieux, Paris, France). Sensitivity was 0.05 ng/ml (conversion factor to SI unit, 3.180). Intra-assay and interassay CVs were, respectively, 8% and 11% for P. Androstenedione and T were measured by RIA (Immunotech International, Marseille, France). Sensitivity was 0.05 ng/ml (conversion factors to SI units, 3.49 for A and 3.47 for TI. Intra-assay and interassay CV were, respectively, 5% and 10~ for both hormones. Plasma LH was measured by an immunometric technique using an LH-Amerlite kit, Sensitivity was 0.05 miu/ml and intra-assay and interassay CVs were 5% and 7c~, respectively, for LH. Statistics Measures of central tendency used were means and measures of variability, standard errors. Statistical assessment of our results was performed with a factorial analysis of variance, linear regression analysis, and paired or unpaired Student's t-test when appropriate. A P < 0.05 was considered statistically significant. RESULTS Patient characteristics in groups A and B were comparable concerning ages ( and Fanchin et al. Adrenal suppression during COH for IVF-ET Fertility and SteriliO, ~'

3 Table 1 Controlled Ovarian Hyperstimulation and Embryology Characteristics in Groups A and B* No. of No. of Day of No. of mature No. of available No. of embryos Group cycles hmg ampules Plasma E~ hcg oocytes embryos transferred p l,,/nll A (DEX) _ ,568 _ ± _ _ 0.2 B (No DEX) _ ,633 ± _ ± ± ± 0.2 * Values are means _+ SE. Differences between groups not sta-? On the day of hcg administration. Conversion factor to SI tistically significant, unit, years, respectively), ovulatory status (78% and 77%, respectively), and clinical indications for IVF- ET (tubal factor, 42~ and 43%; male factor, 32% and 33%; idiopathic infertility, 18% and 17%; and endometriosis, 8% and 7%, respectively). Data on COH characteristics and embryology in groups A and B are detailed in Table 1. Controlled ovarian hyperstimulation characteristics were similar in patients included in both groups (total number ofhmg ampules required, peak E2 levels, and the length of COH). The number of mature oocytes, available embryos obtained, and the number of embryos transferred also were comparable in both groups. There were no statistically significant differences in cleavage rates (defined as number of available embryos/ number of mature oocytes 100) between groups A and B (50~ and 44~, respectively). Concerning clinical pregnancy rates (PRs)/ET, we observed a trend for better results in group A (42%.) in comparison with group B (24%). Ongoing PRs/ET were significantly higher in group A (35% versus 17%, P < 0.04). Hormone Profiles Plasma P levels Plasma P profile in patients receiving or not receiving DEX is presented in Table 2. After GnRH-a Table 2 Plasma P, A, and T Levels 18 Days After GnRH-a (Baseline), on the 7th day of Dexamethasone (After DEX), and on the Day of bcg Administration {After hmgi* Absolute Af~ter hmg~ increases Group Baseline After DEX (day of hcg) ICOH effect) A I DEXI P (ng/mli A Ing/mL~ 1.74 _" 0.10 T (ng/mll 0.22 z 0.02 B tno DEX) Pqng/mLJ 0.22 : 0.05 A (ng/ml~ 1.89 z 0.10 T Ing/mL) 0.25 z _* _~ z _* ~ : z ~ t : z ~ ~ 0.02 * Values are means z SE. Conversion factors to SI units are as follows: P, 2.18; A, 3.49; and T, Group B values significantly higher than group A values, P < 0.01 for P; P < for A and T. (baseline), P was similarly low in both groups at _ 0.01 and ng/ml, respectively. In group A, an additional decrease in plasma P levels to 0.05 _ ng/ml (P < 0.04) was observed after 7 days of DEX treatment (after DEX). On the day of hcg administration (after hmg), plasma P levels were lower in group A (0.57 _ ng/ml) than in group B (0.77 _ ng/ml) (P < 0.01). However, considering exclusively the absolute increases in plasma P levels observed in groups A (calculated as after hmg - after DEX values) and B (calculated as after hmg - baseline values), both groups were similar (0.52 _ and 0.55 _ ng/ml, respectively). Plasma A and T levels Plasma A and T profiles in patients receiving or not receiving DEX are shown in Table 2. After GnRH-a, baseline values for plasma A and T were similarly low in both groups at 1.74 _ and 1.89 _ ng/ml for A, and and 0.25 _ ng/ml for T, respectively. In group A, A and T further decreased to and ng/ ml, respectively, after 7 days of treatment with DEX (P < 0.001) (after DEX). As for P, on the day ofhcg administration (after hmg), final levels of A and T were significantly lower in group A at 1.75 _ 0.12 and ng/ml, respectively, than in group B (3.55 +_ 0.20 and ng/ml, respectively) (P < 0.001). Considering, however, the absolute increases in these hormones in comparison with after DEX and baseline values, both groups were similar: 1.56 _ and _ ng/ml, respectively, in group A, and 1.66 _ and 0.17 _ ng/ml, respectively, in group B. Requirement for hmg and Duration of COH Total hmg requirement and duration of COH in the two groups according to whether plasma P levels on the day of hcg exceeded 0.9 ng/ml are displayed in Figure 1. Plasma P levels on the day of hcg were >0.9 ng/ml in 27% of patients treated with DEX and in 33% of controls and are in accordance with our prior experience (3) and that of others (4). These differences did not reach statistical significance. In Vol. 67, No. 1, Janual'y 1997 Fanchin et al. Adrenal suppression during COH for IVF-ET 117

4 u) m (3L E 0~.- "6 0 Z C9 o c- "6 C Group A Group A Group B Group B Figure 1 Comparison between the respective bmg requirement (to/)) and duration of hmg treatment (bottom) when plasma P levels exceeded (solid bar.~l or not (open bars) 0.9 ng/ml (conversion factor to SI unit, 3.18) on the day of hcg administration in patients treated with DEX (gl'oup A) and in controls (gl'oup B). Patients presenting P > 0.9 ng/ml have received larger amounts of hmg and their stimulation lasted longer than their respective counterparts in both gq'oups (P < 0.01 L both groups, women whose plasma P remained <0.9 ng/ml on the day of hcg received less hmg (32.2 +_ 1.7 and 36.1 _+ 2.0 ampules, for groups A and B, respectively) than those whose plasma P was >0.9 ng/ml ( and 43.5 _+ 2.5 ampules, for groups A and B, respectively} (P < 0.001). Furthermore, in both groups, COH was shorter in women whose plasma P levels were <0.9 ng/ml on the day ofhcg ( and _ 0.2 days, for groups A and B, respectively} in comparison with those whose plasma P was >0.9 ng/ml ( 12.6 _+ 0.2 and 12.4 _+_ 0.2 days for group A and B, respectively} (P < 0.01). As displayed in Figure 2, linear regression analysis showed a positive and significant association (r = 0.33; P < } between the total number of hmg ampules and the plasma P levels of the day of hcg. DISCUSSION This study was conceived to rule out the possible role of adrenal steroidogenesis in plasma P, A, and T elevations observed during hmg treatment in pituitary-desensitized COH cycles (1-3, 6-9). Prior work from our group (6) on the hormonal profiles during the 24 hours after the last hmg administra- T / tion (day of hcg) in women undergoing COH with a time-release GnRH-a and hmg has evidenced a progressive and significant increase in plasma P, A, and T levels, peaking approximately 12 to 15 hours after the hmg injection. However, as hmg was administered in the evening, the increase in P, A, and T coincided with the circadian increase in adrenal activity, raising the possibility of an adrenal contribution to this event. In the present study, we suppressed the adrenal function with DEX in 60 patients undergoing COH with the same protocol. Following a modified regimen previously described by Cedars et al. (7), the daily DEX dose (1 rag) was standardized in all patients. Sixty other women receiving an identical treatment except for DEX served as controls. Mean plasma P concentration during the follicular phase of the menstrual cycle is recognized as being approximately 0.5 ng/ml, and ovary and adrenal cortex contribute roughly 50~. (10). Consistently with this, after curtailing substantially the ovarian P secretion with pituitary desensitization in the present 120 patients, residual circulating P levels at baseline were approximately 0.2 ng/ml, which corresponds to the adrenal contribution to P. As expected, DEX lowered significantly plasma P to undetectable levels. As for P, mean T levels 18 days after time-release GnRH-a administration were comparable to castrate values at 0.24 ng/ml (11) and further fell to undetectable levels after DEX treatment. Plasma A levels exceeded slightly the castrate range at baseline but were completely lowered by DEX. Furthermore, as expected, plasma levels of P, A, and T at the end of COH (day of hcg) were significantly higher in group B than in group A. As detailed in Table 2, these differences in plasma P, A, and T correspond to those observed between baseline in group B and after DEX in group A and reflect the steady adrenal contribution to these hormones dur- P 2 (ng/ml) O ~ 1 go lh.e J No of hmg ampules Figure 2 Lineal" regression analysis between tile total number ofhmg ampules and plasma P levels on the day ofhcg administration (r = 0.33; P < ). 118 Fanchin et al, Adrelzal suppression during COH for IVF-ET Fertility and Sterility '~.~

5 ing COH. The net increases in these three hormones (after hmg - baseline values) were unaltered by DEX administration. This indicates that the increases in P (1-3, 6, 7) and androgens (6, 8, 12) observed during COH solely result from an effect of exogenous gonadotropins on the ovary. Another issue deserving consideration is that, in both groups, COH was longer and required more hmg in patients whose P levels on the day of hcg was >0.9 ng/ml. This supports the hypothesis that this phenomenon may correspond to a dose durationdependent effect ofhmg on ovarian follicles in which the adrenals play no contribution. Furthermore, patients receiving DEX presented a trend for better IVF-ET results. This is consistent with previous reports that found beneficial effects of glucocorticoid treatment on IVF-ET outcome (13, 14). The present study, however, was not designed primarily to address this important issue and the restricted number of cases analyzed does not offer enough statistical power for drawing conclusions about glucocorticoid administration and pregnancy outcome. Although an immunosuppressive effect commonly has been advocated for explaining the favorable effect of DEX on PRs (13, 14), our present data offer an alternative mechanism for this issue. Suppression of adrenal function with DEX may improve implantation by lowering the absolute values of P and androgens on the day of hcg, possibly reducing a negative impact of these hormones on endometrial receptivity. In conclusion, our results ruled out any adrenal involvement in follicular phase increments in P and androgens occurring in women whose endogenous gonadotropins are suppressed by GnRH-a. Hence, the present data reinforced our original hypothesis that this phenomenon results from an action of the exogenous gonadotropin treatment (hmg) on the ovary during COH (6). This leads us to recommend that, when plasma P rises to levels potentially harmful to endometrial receptivity (>0.9 ng/ml), it may be helpful to lower the overall exposure to hmg (less ampules for shorter time) in subsequent COH cycles. Furthermore, considering the hypothesis that the possible adverse effects of P and androgens on endometrial receptivity could be minimized by reducing the absolute plasma levels of these hormones during COH, the suppression of adrenal hormone production by DEX aiming at counterbalancing the direct effects ofhmg on the ovaries during pituitary-desen- sitized COH cycles may be profitable. These two latter points, however, deserve further investigation. REFERENCES 1. Schoolcraft W, Sinton E, Schlenker T, Huynh D, Hamilton F, Meldrum DR. Lower pregnancy rate with premature luteinization during pituitary suppression with leuprolide acetate. Fertil Steril 1991;55: Silverberg KM, Burns WN, Olive DL, Rieh] RM, Schenken RS. Serum progesterone levels predict success of in vitro fertilization/embryo transfer in patients stimulated with leuprolide acetate and human menopausal gonadotropins. J C]in Endocrinol Metab 1991;73: Fanchin R, de Ziegler D, Taieb J, Hazout A, Frydman R. Premature elevation of plasma progesterone alters pregnancy rates of in vitro fertilization and embryo transfer. Fertil Steri[ 1993;59: Legro RS, Ary BA, Paulson RA, Stanczynk FZ, Sauer MV. Premature luteinization as detected by elevated serum progesterone is associated with a higher pregnancy rate in donor oocyte in vitro fertilization. Hum Reprod 1993;8: Meldrum DR, Tsao Z, Monroe SE, Braunstein GD, Sladek J, Lu JKH, et al. Stimulation of LH fragments with reduced bioactivity following GnRH agonist administration in women. J C]in Endocrino] Metab 1984;58: Fanchin R, de Ziegler D, Castracane VD, Taieb J, Olivennes F, F,'ydman R. Physiopathology of premature progesterone elevation. Ferti] Steril 1995;64: Cedars MI, Steingold KA, de Ziegler D, Lapolt PS, Chang R J, Judd HL. Long-term administration of gonadotropin-releasing hormone agonist and dexamethasone: assessment of the adrenal role in ovarian dysfunction. Ferti] Steril 1992;57: Fanchin R, Righini C, Olivennes F, de Ziegler D, Selva J, Frydman R. Premature progesterone elevation does not alter oocyte quality in in vitro fertilization. Fertil Steril 1996;65: Cedars MI, Surey E, Hamilton F, Lapolt P, Meldrum DR. Leuprolide acetate lowers circulating bioactive luteinizing hormone and testosterone concentrations during ovarian stimulation for oocyte retrieval. Fertil Steri] 1990;53: Lippset MB. Steroid hormones. In: Yen SSC, Jaffe RB, editors. Reproductive endocrinology. Philadelphia: Saunders, 1986: Steingold K, de Ziegler D, Cedars M, Meldrum DR, Lu JKH, Judd HL, et al. Clinical and hormonal effects of chronic gonadotropin-releasing hormone agonist t,'eatment in polycystic ovarian disease. J Clin Endocrinol Metab 1987;65: YdingAndersen C, Ziebe S. Serum levels of free androstenedione, testosterone and estradiol are lower in the follicular phase of conceptional than of non-conceptional cycles after ovarian stimulation with gonadotropin-releasing hormone agonist protocol. Hum Reprod 1992;7: Cohen J, Malter H, Elsner C, Kort H, Massey J, Mayer MP. Immunosuppression supports implantation of zona pellucida dissected human embryos. Fertil Steril 1990;53: Polak de Fried E, Blanco L, Lancuba S, Asch RH. Improvement of clinical pregnancy rate ~nd implantation rate of invitro fertilization-embls, o transfer p~tients by using methylprednisone. Hum Reprod 1993;8: Vol. 67, No. 1, January 1997 Fanchin et al Adrenal suppression during COH for IVF-ET 119