- Email: [email protected]
Neuroendocrine mechanism of anovulation in users of contraceptive subdermal implant of nomegestrol acetate (Uniplant)
FERTILITY AND STERIJ.ITY@ Copyright ” 1997 American Society far Reproductive Medicine
Vol. 67, No. 2, February Printed
Neuroendocrine mechanism of anovulation subdermal implant of nomegestrol acetate Kurt Barnhart, M.D.$§ Luigi Devoto, M.D.#j Ricardo Pommer, M.D.$** University Salvador,
of Chile, School Bahia, Brazil
on acid-free
paper
1997
m LT. S. A
in users of contraceptive (Uniplant)*t
Teresa Sir-Pettermann, M.D.$ Jorge Robinovic, M.D.$ Elsimar Coutinho, M.D.?? of Medicine,
San Borja-Arriardn
Clinical
Hospital,
Santiago,
Chile,
and
Federal
Univer.sity
of Bahia,
Objective: To evaluate a nomegestrol acetate subdermal contraceptive implant’s (Uniplant; Thermex, Monaco) effect on the hypothalamus-pituitary-ovarian axis. Design: A prospective clinical trial. Setting: San Borja-Arriarkn Clinical Hospital, University of Chile, School of Medicine. Patient(s): Normally cycling healthy women. Intervention(s): Insertion of Uniplant. Main Outcome Measure(s): Luteinizing hormone pulse and endocrine profiles were assessed before, 48 hours after insertion, and after prolonged use of the implant. Result(s): Anovulation was noted in 100% of users in the first month. Seventy percent of subjects demonstrated follicular development with the absence of ovulation and an endocrine profile similar to the follicular phase: (LH pulse/8 hours 6.85 + 0.67, LH amplitude 3.54 + 0.65 mIU/mL (conversion factor to SI unit, 1.001, and Ez 193 -C 29.4 pg/mL (conversion factor tO SI unit, 3.67), whereas 30% demonstrated no follicular activity with an endocrine profile similar to the luteal phase: (LH pulse/8 hours; 3.66 k 0.66, LH amplitude 5.76 2 1.73 mIU/mL, and Ez 67.5 t 4 pg/mL. Clinical characteristics, serum gonadotropin concentration, and LH pulse characteristics failed to predict which subjects would initiate or remain devoid of follicular activity. Conclusion(s): Uniplant results in anovulation via two mechanisms: hypothalamic suppression in subjects surge in subjects
Key Words:
who lack follicular development,, who initiate follicular activity.
Nomegestrol
acetate subdermal
and
likely
Fertil implants,
suppression
of the
pituitary
LH
Steril@ 1997;67:250-5 LH pulse hypothalamic pituitary fimc-
tion
A single subdermal implant of 55 mg of nomegestrol acetate (Uniplant; Thermex, Monaco), has been shown to be safe and effective as a l-year contracepReceived April 1, 1996; revised and accepted October 15, 1996. * Uniplant, Thermex, Monaco. 7 Supported in part by a grant to South-to-South Cooperation in Reproductive Health by the Rockefeller Foundation, New York, New York. f Research Institute for the Mother and Child, School of Medicine, University of Chile-San Borja ArriarBn-Clinical Hospital. Department of Obstetrics and Gynecology, School of Medicine, University of Chile. 5 Rockefeller Foundation Fellow at the University of Chile. 11Present address: Division of Human Reproduction, Department of Obstetrics and Gynecology, University of Pennsylvania, Philadelphia, Pennsylvania. 1 Reprint requests: Luigi Devoto, M.D., University of Chile, P.O. Box 226-3, Santiago, Chile (FAX: 56-2-554-6890). 250
Barnhart
et al.
Neuroendocrine
mechanism
of anovulation
tive device (1). The mechanism of action of progestagen implant(s) have included suppression of ovulation (2, 31, inadequate luteal phase (3), decreased sperm penetration of cervical mucus 113,41, and suboptimal endometrial development (5). Overall rates of anovulation range from 56% to 90% in users of subdermal progestin implant(s) (6), however, anovulation is the main contraceptive mechanism of action in the first few months of nomegestrol acetate use (7). Up to 25% of users of Uniplant d.emonstrate no follicular activity within the first 100 days of use, with an additional 60% demonstrated abnormal fol-
** Serono Chile. t t Maternidade Bahia.
Fellow in Reproductive Climerio de Oliveira,
Endocrinology. Federal University
Fertility
and
Sterilityycu
of
liculogenesis resulting in the development of an unluteinized enlarged follicle (7). The secretory patterns of gonadotropins change throughout the menstrual cycle and play a fundamental role in follicular development, ovulation, and corpus luteum function (8,9). The study of LH pulse patterns is an important tool to assess the hypothalamic-pituitary-ovarian axis and evaluate normal and abnormal folliculogenesis (10, 11). Progesterone alters normal LH pulsatility patterns and the LH surge (12,13). Incomplete suppression of gonadotropins or an altered ratio of gonadotropin secretion may result in abnormal folliculogenesis (14, 15) and the development of an enlarged follicle (15, 16). To enhance our understanding of anovulation caused by progestagen only subdermal implants, this study evaluates the hypothalamic-pituitaryovarian axis by measuring LH pulse patterns, serum gonadotropin concentrations, and follicular development in users of a single subdermal implant of nomegestrol acetate. Analysis of these patterns was conducted immediately after insertion and during prolonged use of the implant. Differences in the parameters also are evaluated in subjects who developed follicular growth and those devoid of follicular activity during use of the implant. MATERIALS
AND
METHODS
Subjects
Ten healthy women, 22 to 36 years of age, requesting a contraceptive method at the family planning clinic at San Borja-Arriaran Clinical HospitalNational Health Service-School of Medicine, University of Chile, participated in this study. Inclusion criteria included non-breast-feeding women with a normal cycle (28 to 32 days), a normal body mass index (BMI = weight [kg]/height [m21), a hemoglobin concentration > 12 mg/dL, and absence of use of any hormonal contraception in the previous 6 months. Each subject was able to understand and was willing to comply with the study protocol and signed an informed consent approved by the Institutional Review Board. Study
67, No. 2, February
Venipuncture
1997
and
Continuous
Blood
Sampling
An 18-gauge angiocatheter was placed in the forearm of the subject and blood collected continuously as described previously (18). Approximately 3 mL of blood was collected into a separate test tube every 10 minutes, using an automatic carousel (model 2110, Bio-Rad, Hercules, CA). The serum was separated by centrifugation and stored at -20°C until assayed. Assay
Both LH and FSH were measured in duplicates by RIA method using National Institute of Diabetes and Digestive and Kidney Disease reagents as reported previously (19). Luteinizing hormone and FSH are expressed in mIU/mL (conversion factor to SI unit, 1.00) using LER 907 as the standard. The intra-assay and interassay coefficients of variation (CV) were 7% and 8%, respectively. All s.amples from each patient were analyzed with the same RIA protocol. Estradiol and P were measured as reported previously (20). Statistics
Design
Each woman underwent three studies of LH pulsatile secretions. All study periods started between 8:00 A.M. and 9:00 A.M. and continued for 8 hours in an awake patient. Blood samples were collected every 10 minutes. Study I was conducted in the early follicular phase (days 3 to 4) of the menstrual cycle, just before insertion of the nomegestrol acetate implant. Study II was conducted 48 hours after insertion (days 5 to 6). Subjects then were evaluated proVol.
spectively with twice weekly transvaginal ultrasound (630 5 MHz vaginal probe; Aloka, Tokyo, Japan) and twice weekly serum Ez conce:ntrations to determine the presence or absence of follicular activity. Study III was conducted either during follicular growth (follicle size 14 to 20 mm) or, if no follicle activity was detected, after 30 to 45 days of nomegestrol acetate use. If follicular development was not initiated within 30 to 45 days, it was presumed folliculogenesis had been inhibited and the final LH pulse study was performed to compare with the LH pulse profile of subjects who had initiated folliculogenesis. In subjects who initiated folliculogenesis, the presence of subsequent ovulation was assessed with continued ultrasound to detect follicular rupture and serum P concentrations. At the end of this investigation, the subjects continued the use of the subdermal implant and were followed as part of the clinical trial evaluating the efficacy and. acceptability of this contraceptive device (17).
and Pulse
Analysis
The mean of duplicate values for LH of each sample was taken as the hormone value. Significant serum LH pulses were identified as described by Veldhuis et al. (21). To be detected as a significant peak, the LH concentration of two consecutive samples had to differ by at least twice the interassay CV at the level of the nadir, thus reducing the likelihood of false positive determinations to a ratio of ~5%. The analysis of pulsatile LH secretion was carried Barnhart
et al.
Neuroendocrine
mechanism
of anoculation
251
out using Cluster Pulse Analysis as reported previously (19). The transverse means for LH was calculated from 48 consecutive samples (one individual pulse study). The transverse means for FSH and Ea were determined from five samples of the same study. Luteinizing hormone pulse characteristics determined in the cluster analysis were compared by analysis of variance and individual differences determined by the Tacky Cramer test. Intragroup and intergroup differences in serum hormone concentrations and pulse characteristics were evaluated using a paired t-test and unpaired t-test, respectively. Data are presented as means ? SEM. Significance was established at the level of P < 0.05. RESULTS Acute Effect of Nomegestrol Study I (Before Insertion) Hours After Insertion)
of the Presence
of Follicular
The 10 patients were followed prospectively with ultrasound and hormonal evaluation: seven women were noted to develop follicles with maximum diameter ranging from 35 to 48 mm (group A), whereas the other three had no follicular activity detected (follicle < 10 mm), after 45 days of Uniplant insertion (group B). None of the patients in groups A or B subsequently ovulated, as determined by lack of rupture of the follicle as visualized by serial ultrasound in those with the initiation of follicular growth (group A) or by elevated serum P > 4 ng/mL (conversion factor to SI unit, 3.18) 1 week after study III in all subjects (groups A and B). Clinical characteristics between the two groups were not different. There was no difference in the mean age, (group A, 30.5 252
Barnhart
et al.
1 Endocrine Profile and LH Pulse Characteristics Subjects Using a Nomegestrol Acetate Implant, Uniplant, With Follicular Activity (Group A)*
Study
No. of pulses Amplitude ~mIU/mL) LH (treatment) bIU/mL) FSH (treatment) bnIU/mL~
E, (pg/mL)
I
Study II 48 h
preinsertion
postinsertion
(n = 7)
(II = 7)
of
Study III during folliculogenesis (n = 71
6.57
? 0.57
6.26
t
0.28
6.85
t
0.671
3.95
+ 0.74
6.13
+
0.98-t
3.54
t
0.65
6.41
i- 1.06
8.6
+
2.53
4.72
t
0.36+
i- 0.42 2 4.8
8.8 101.4
? 0.68 + 13
3.36 193.2
11.8 64.6
* Values are means t SEM. Conversion and FSH, 1.00; Ez, 3.67. f No significant difference. Z Significantly different from studies test. § Significantly different from studies ante.
factors
to Sl units
-t 0.238 * 29.43
are as follows:
LH
I and
III, P < 0.05,
I and
II, P < 0.05,
Tukey-Cramer analysis
of vati-
Acetate: Comparison Versus Study II (48
At study I, before the insertion of the implant, the 10 subjects have a mean of 7.1 + 0.52 pulses per 8 hours with an amplitude of LH 4.7 + 1.0 mIU/mL. Serum LH and FSH concentration were 10.3 ? 1.59 and 10.7 2 0.35 mIU/mL, respectively. The LH:FSH ratio was 0.78 5 0.05. Estradiol concentration was 62.57 2 3.61 pg/mL. At study II, 48 hours after implant insertion, there was no significant change in the number of pulses in 8 hours: (6.6 +- 0.4), FSH concentration (9.4 i 0.42 mIU/mL), LH:FSH ratio (0.85 2 O.l), or Ez concentration (89.9 -+ 9.42 pg/mL. There was an increase in LH pulse amplitude (7.1 i 1.84 mIU/mL; P = 0.08) and concentration (13.3 i 3.04 mIU/mL; P = 0.17) after the insertion of the implant but neither difference reached statistical significance. Determination Development
Table
Neuroendocrine
mechanism
of anouulation
-+-1.73 years; group B 30.6 -+ 3.84 years), BMI, parity, length of menstrual cycle, or day of implant insertion. Evaluation Initiated Ovulate
of Endocrine Profile of Subjects who Follicular Development but Failed to (Group A)
Table 1 illustrates the endocrine profiles of subjects who developed follicular growth (group A, n = 7). There was no change in pulse frequency per 8 hours, preinsertion (study I), 48 hours, postnomegestrol acetate use (study II), or during the development of follicular growth (study III). A significant increase in pulse amplitude was noted at stud,y II (P < 0.05) compared with study I and study III. Luteinizing hormone concentrations also were elevated at study II, but this was not a significant cha.nge compared with study I or study III. Serum FSH was not significantly different at study I and study II but was significantly lower at study III (P < 0.05). Serum E2 concentration was not significantly different at study I and study II but was significantly elevated at study III (P < 0.05). Evaluation of Endocrine Profile of Subjects Were Devoid of Follicular Activity (Group
who B)
Table 2 illustrates the endocrine profiles of subjects devoid of follicular maturation (group B, n = 3). There was no significant difference in pulse frequency per 8 hours, at preinsertion (study I), and 48 hours postinsertion of nomegestrol
and Sterility8
Table 2 Endocrine Profile and LH Pulse Characteristics Subjects Using a Nomegestrol Acetate Implant, Uniplant, Without Follicular Activity (Group B):’
Study I preinsertion (n = 3) No. of pulses Amplitude (mIU/mL) LH (treatment) (mIU/mL) FSH (treatment1 (mIU/mL) E2 WmL)
Study II 48 h postinsertion (n = 3)
of
Study III 30 to 45 d postinsertion (n = 3)
8.33
t
0.88
7.33
t 1.2
3.66
r 0.66?
3.00
+
0.21
3.92
t 0.73
5.76
t 1.73i
6.41
i-
3.15
8.4
r 6.00
7.05
f 1.98$
E 0.49 t 12
10.2 70.4
+ 0.39 t- 9.5
11.71 67.5
9.48 59.16
Table 3 Comparison of Endocrine Profile and LH Pulse Characteristics of Subjects After Prolonged use of’ a Nomegestrol Acetate Implant, Uniplant, Between Those With and Without Follicular Activity (Groun A Versus Groun Bl* Group A (n = 7) No. of pulses Amplitude (mIU/mLl LH (treatment) (mIU/mL 1 FSH (treatment) ImIUimL) LH:FSH ratio EZ (pg/mL)
lr 1.31 i 4j:
* Values are means r~ SEM. Conversion factors to SI units are as follows: LH and FSH, 1.00; Ez, 3.67. t Significantly different from studies I and II, P < 0.05, analysis of variance. $ No significant difference.
Group B (n = 3)
6.85
k
0.67
3.66
r 0.66.:
3.54
+
0.65
5.76
t 1.73t
4.72
L
0.3
7.05
t 1.981
11.71 0.6 67.5
2 1.3’1 -c 0.13-i+ 4f
3.36 t 0.23 1.36 2 0.33 193.2 i 29.4
* Values are means + SEM. Conversion factors to SI units as follows: LH and FSH, 1.00; Ey, 3.67. t Significantly different from group A, P < 0.05, unpaired test.
are t-
that subjects from group A developed a LH pulse profile similar to that found during the follicular phase of the normal menstrual cycle, whereas group B developed a LH pulse profile consistent with that of midluteal phase of the normal menskual cycle.
II, however, pulse amplitude was increased significantly at study III (P < 0.05). Serum LH concentration was not significantly different in the three study periods. Follicle-stimulating hormone concentration was not statistically different in study I and study II, whereas it was significantly higher at study III (P < 0.05). Serum Ez concentrations were not different across the three study periods.
DISCUSSION
This study indicates that the subdermal implant of nomegestrol acetate induces anovulation particu-
Comparison of Subjects With Follicular Development and With Subjects Devoid of Follieular Activity (Group A Versus Group B)
There was no significant difference in the mean number of pulses, amplitude of pulses, serum LH concentration, LH:FSH ratio, or mean Ez concentration between the two groups at study period I and study period II. Serum FSH concentrations were significantly higher in group A than B at study I (group A, 11.8 -+ 0.42; group B, 9.48 I 0.49; P < 0.05) but not at study II (group A, 8.8 ? 0.68; group B, 10.2 t- 0.39). Table 3 represents the comparison of the endocrine findings of groups A and B at study III. At study III, group A demonstrated a significantly higher number of pulses in 8 hours compared with group B (P < 0.05) with a significantly lower pulse amplitude (P < 0.05). Serum LH and FSH concentrations were significantly higher in group B fP < 0.05), whereas the LH:FSH ratio and serum E2 concentration was significantly greater in group A (P < 0.05). Figure 1 illustrates a representative LH pulse profile from a subject in group A and another subject in group B at study periods I, II, and III. Subjects from both groups showed a similar LH pulse profile preinsertion (study I) and acutely postinsertion of nomegestrol acetate (study II). Note, at study period III, Vol.
67, No. 2, February
1997
h.
II.
I
Figure 1 The LH pulse profile in two representative sub.jects illustrating the LH pulse patterns of subjects who developed a persistent nonluteinized follicle (group A, l ), and subjects who remained devoid of follicular activity (group B, 0). Study I was conducted in the early follicular phase (days 3 to 4) of the menstrual cycle, before insertion of the nomegestrol acetate implant. Study II was conducted 48 hours after insertion (days 5 to 61. Study III was conducted either during follicular growth (group A) or, if no follicle activity was detected, after 30 to 45 days of nomegestrol acetate use (group B).
Barnhart
et al.
Neuroendocrine
mechanism
of anovulation
253
larly in the first few months of use. Thirty percent (3110) of subjects were devoid of any follicular activity for the duration of the study, whereas the remaining 70% of subjects (7110) developed follicular growth that failed to luteinize or ovulate. Age, parity, BMI, or menstrual cycle characteristics were not different in those subjects who would initiate or became devoid of follicular activity. An elevated serum FSH in those who eventually developed follicular activity (group A) was the only significant difference between the two groups before insertion of the nomegestrol acetate implant. It is not clear if this difference is a result of multiple statistical comparisons or if it is of true clinical significance. A difference in FSH concentration was not noted at study II nor was there a significant difference in Ez concentration or follicular diameter as measured by ultrasound (data not shown). No changes in endocrine parameters or pulse characteristics noted immediately after insertion of the implant were predictive of subsequent presence or absence of follicular activity. Therefore, either the acute endocrine effect of nomegestrol acetate is not predictive of the suppression of folliculogenesis or the study of LH pulse characteristics is not a sensitive enough tool to detect such a change. Nevertheless, an increase in the LH pulse amplitude, with no associated change in LH pulse frequency, was noted after the insertion of the implant. This finding was statistically significant when evaluating the subjects in group A but did not reach significance when comparing the evaluation of all subjects (P = 0.08). This acute effect of a progestin is similar to the acute stimulatory effect of P on the pituitary, with no effect on the hypothalamus, as noted by Couzinet et al. cm. Seventy percent of subjects (group A) initiated folliculogenesis as demonstrated by an increase in the diameter of a leading follicle measured by vaginal ultrasound and a progressive increase in Ez concentration. In this group, serum FSH concentration decreased and the LH:FSH ratio increased at study III as compared with study I and study II. This change was likely a result of the negative feedback of Ez on the pituitary. None of these subjects ovulated or luteinized, as demonstrated by a lack of collapse of the leading follicle or rise in serum P concentrations. The LH pulse characteristics in this group were indistinguishable during all three study periods and remained similar to pulse characteristics consistent with the early follicular phase (8). Nomegestrol acetate had no effect on hypothalamic pulse generation in this group but likely had a direct effect on the pituitary reducing the periovulatory hypersensitivity to GnRH and Ez. Thus, the sustained elevated release of LH needed for ovulation and luteinization 254
Barnhart
et al.
Neuroendocrine
mechanism
of anovulation
of the dominant follicle was inhibited. Other possibilities are that nomegestrol acetate acted on the hypothalamus preventing the increase in pulse frequency associated with the late follicular phase (B), reduced the pituitary’s hypersensitivity to GnRH pulses by reducing the positive feedback of GnRH on GnRH receptors, or even acted directly on the ovary. Thirty percent of subjects (group El) were devoid of any follicular development, as demonstrated by a lack of follicular growth by vaginal ultrasound and serum E, concentrations. After 30 to .45 days use of the nomegestrol acetate implant (study III), pulse characteristics changed from those noted preinsertion and immediately postinsertion. In the initial study period, a pattern of high frequency and relative low amplitude, consistent with th.at of the follicular phase of a normal menstrual (cycle (81, was noted. In study III, LH pulse characteristics were noted to be of low frequency with an increased amplitude, consistent with the midluteal phase of a normal menstrual cycle (8). This alteration in pulse characteristics is consistent with the effect of P on the hypothalamus when given in the normal follicular phase, as demonstrated by Soules et al. (13). In addition, this LH pulse profile has been shown to be inappropriate to induce follicular growth (8, 11). A possible explanation of the initial two distinct effects of nomegestrol acetate on the hlypothalamuspituitary-ovarian axis could be related serum concentrations of nomegestrol acetate after the insertion. Higher doses may suppress the lhypothalamus and eliminate the initiation of folliculogenesis whereas lower concentrations may effect feedback at the level of the pituitary or act directly on the ovary, preventing follicle luteinization and ovulation. A dose-related effect on the pituitary-ovarian axis has in fact been reported with oral use of nomegestrol acetate (23). In conclusion, these data indicate two types of anovulation: type I is characterized by follicular development with absence of luteinization or ovulation (70%) and type II is characterized by the absence of any follicular growth (30%). Clinical characteristics of the patients, serum gonadotropin concentrations, and LH pulse characteristics, before and acutely after insertion, failed to predict which subjects would initiate follicular activity. Anovulation in nomegestrol acetate users who lack any follicular development is associated with a diminished frequency of GnRH pulse generation demonstrating a suppressive effect at the level of the hypothalamus. Seventy percent of subjects using an implant of nomegestrol acetate initiated follicular development but failed to ovulate. In this group of subjects, nomegestrol acetate likely acts preferentially at the level of Fertility
and Sterility@
the pituitary, reducing sulting in an persistent
Acknowledgments. ily Planning Clinic in particular, Mrs.
or suppressing non luteinized
LH surge, refollicle.
We are grateful to the midwives of the Famof the San Borja-ArriarBn Clinic Hospital and, Ivon Gallegos, for her technical assistance.
REFERENCES 1. Coutinho EM. One year contraception with a single implant containing Nomegestrol acetate (Uniplant). Contraception 1993;47:94-105. 2. Brache V, Alvarez-Sanchez F, Faundes A, Tejada AS, Cochon L. Ovarian endocrine function through five years of continuous treatment with Norplant subdermal contraceptive implants. Contraception 1990;41:169-77. 3. Brache V, Faundes A. Anovulation, inadequate luteal phase and poor sperm penetration in cervical mucus during prolonged use of Norplant implants. Contraception 1985;31:261. 4. Croxatto HB, Diaz S. Treatment with Norplant subdermal implants inhibits sperm penetration through cervical mucus in vitro. Contraception 1987;36:193. 5. Shaaban MM, Segal S, Salem HT, Ghaneimah SA, Khalifa E-AM, Ahmed A-G. Sonographic assessment of ovarian and endometrial changes during long-term Norplant use and their correlation with hormonal levels. Fertil Steril 1993;59:998-1002. 6. Ladipo 0, Coutinho EM. Contraceptive implants. Curr Opin Obstet Gynecol 1994;6:564-9. 7. Devoto I, Pommer R, Barnhart K, Alba F, Sir T, Robinovich J, et al. Ovarian function in subjects using Uniplant. Contracept Fertil Sex (Paris) 1995;(Suppl Nl-N9)23;S-32. 8. Filicori M, Santoro N, Merriam G, Crowley W. Characterization of the physiological pattern of episodic gonadotropin secretion throughout the human menstrual cycle. J Clin Endocrinol Metab 1986;62:1136. 9. Veldhuis J, Beitins I, Johnson M, Serabian M, Dufau L. Biologically active luteinizing hormone is secreted in episodic pulsations that vary in relation to the stage of the menstrual cycle. J Clin Endocrinol Metab 1984;58:1050. 10. Bennet A, Lacaze C, Berrada C, Barbe V, Louvet J. Correlations between mean LH levels and LH pulse characteristics: differences between normal and anovulatory women. Clin Endocrinol (Oxf) 1991;35:431-7. 11. Schweiger U, Reinhard T, Broocks A, Karl-Martin P. Gonadotropin secretion in the second half of the menstrual cycle: a comparison of women with normal cycles, luteal phase defects
Vol.
67, No. 2, February
1997
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
Barnhart
and disrupted follicular development Clin Endocrinol (0x0 1990;32:25-32. Vandenberg G, DeVane G, Yen SSC. Effects of exogenous estrogen and progestin on pituitary responsiveness to synthetic luteinizing hormone-releasing factor. J Clin Invest 1974;53:1750. Soules M, Steiner R, Clifton D, Cohen N, Aksel S, Bremner W. Progesterone modulation of pulsatile luteinizing hormone secretion in normal women. J Clin Endocrinol Metab 1984;58:378. Poindexter A, Dilby G, Brody S, Snabes M. The effects of a long-acting progestin on the hypothalamic-pituitary-ovarian axis in women with normal menstrual cycles. Contraception 1993;48:37-45. Alvarez-Sanchez F, Brache V, Faundes A, Tejada AS. Abnormal endocrine profile among women with confirmed presumed ovulation during long-term Norplant use. Contraception 1986;33:111-9. Cook D, Parfet J, Smith C, Moss G, Youngquist R, Garverick H. Secretory patterns of LH and FSH during development and hypothalamic and hypophysial characteristics following development of steroid-induced ovarian follicular cysts in dairy cattle. J Reprod Fert 1991;91:19-28. Coutinho E, De Souza JA, Athayde C, Barbo,sa I, Alverez F, Brache V, et al. Multicenter clinical trail on the efficacy and acceptability of a single contraceptive implant of nomegestrol acetate, Uniplant. Contraception 1996;53:121-5. Sir-Petterman T, Recabarren S, Bitt A, Jager W, Zimmermann V, Wila L. A simple device for serial blood collection in human subjects and animals. Exp Clin Endocrinol Diabetes 1995;103:398-401. Sir-Petterman T, Devoto L. Effect of clomiphene citrate on pulsatile luteinizing hormone profile in normal women and polycystic ovary syndrome. Horm Metab Res 1989;21:37036. Devoto L, Vega M, Navarro V, Sir T, Alba F, Castro 0. Regulation of steroid hormone synthesis by human corpora lutea: failure of follicle stimulating hormone to support luteal steroidogenesis in vivo and in vitro. Fertil Steril 1989;50:62833. Veldhuis JD, Royel AD, Johnson ML. Minimizing false-positive errors in hormonal pulse detection. Am J Physiol 1985;248:E475-81. Couzinet B, Brailly S, Bouchard P, Schaison G. Progesterone stimulates luteinizing hormone secretion by acting directly on the pituitary. J Clin Endocrinol Metab 1992;74:374-8. Brazin B, Thevenot R, Bursaux C, Paris J. Effect of nomegestrol acetate, a new 19-nor-progesterone derivative, on pituitary-ovarian function in women. Br J Olbstet Gynaecol 1987;94:1199-204.
et al.
Neuroendocrine
mechanism
of anovulation
255
Vol. 67, No. 2, February Printed
Neuroendocrine mechanism of anovulation subdermal implant of nomegestrol acetate Kurt Barnhart, M.D.$§ Luigi Devoto, M.D.#j Ricardo Pommer, M.D.$** University Salvador,
of Chile, School Bahia, Brazil
on acid-free
paper
1997
m LT. S. A
in users of contraceptive (Uniplant)*t
Teresa Sir-Pettermann, M.D.$ Jorge Robinovic, M.D.$ Elsimar Coutinho, M.D.?? of Medicine,
San Borja-Arriardn
Clinical
Hospital,
Santiago,
Chile,
and
Federal
Univer.sity
of Bahia,
Objective: To evaluate a nomegestrol acetate subdermal contraceptive implant’s (Uniplant; Thermex, Monaco) effect on the hypothalamus-pituitary-ovarian axis. Design: A prospective clinical trial. Setting: San Borja-Arriarkn Clinical Hospital, University of Chile, School of Medicine. Patient(s): Normally cycling healthy women. Intervention(s): Insertion of Uniplant. Main Outcome Measure(s): Luteinizing hormone pulse and endocrine profiles were assessed before, 48 hours after insertion, and after prolonged use of the implant. Result(s): Anovulation was noted in 100% of users in the first month. Seventy percent of subjects demonstrated follicular development with the absence of ovulation and an endocrine profile similar to the follicular phase: (LH pulse/8 hours 6.85 + 0.67, LH amplitude 3.54 + 0.65 mIU/mL (conversion factor to SI unit, 1.001, and Ez 193 -C 29.4 pg/mL (conversion factor tO SI unit, 3.67), whereas 30% demonstrated no follicular activity with an endocrine profile similar to the luteal phase: (LH pulse/8 hours; 3.66 k 0.66, LH amplitude 5.76 2 1.73 mIU/mL, and Ez 67.5 t 4 pg/mL. Clinical characteristics, serum gonadotropin concentration, and LH pulse characteristics failed to predict which subjects would initiate or remain devoid of follicular activity. Conclusion(s): Uniplant results in anovulation via two mechanisms: hypothalamic suppression in subjects surge in subjects
Key Words:
who lack follicular development,, who initiate follicular activity.
Nomegestrol
acetate subdermal
and
likely
Fertil implants,
suppression
of the
pituitary
LH
Steril@ 1997;67:250-5 LH pulse hypothalamic pituitary fimc-
tion
A single subdermal implant of 55 mg of nomegestrol acetate (Uniplant; Thermex, Monaco), has been shown to be safe and effective as a l-year contracepReceived April 1, 1996; revised and accepted October 15, 1996. * Uniplant, Thermex, Monaco. 7 Supported in part by a grant to South-to-South Cooperation in Reproductive Health by the Rockefeller Foundation, New York, New York. f Research Institute for the Mother and Child, School of Medicine, University of Chile-San Borja ArriarBn-Clinical Hospital. Department of Obstetrics and Gynecology, School of Medicine, University of Chile. 5 Rockefeller Foundation Fellow at the University of Chile. 11Present address: Division of Human Reproduction, Department of Obstetrics and Gynecology, University of Pennsylvania, Philadelphia, Pennsylvania. 1 Reprint requests: Luigi Devoto, M.D., University of Chile, P.O. Box 226-3, Santiago, Chile (FAX: 56-2-554-6890). 250
Barnhart
et al.
Neuroendocrine
mechanism
of anovulation
tive device (1). The mechanism of action of progestagen implant(s) have included suppression of ovulation (2, 31, inadequate luteal phase (3), decreased sperm penetration of cervical mucus 113,41, and suboptimal endometrial development (5). Overall rates of anovulation range from 56% to 90% in users of subdermal progestin implant(s) (6), however, anovulation is the main contraceptive mechanism of action in the first few months of nomegestrol acetate use (7). Up to 25% of users of Uniplant d.emonstrate no follicular activity within the first 100 days of use, with an additional 60% demonstrated abnormal fol-
** Serono Chile. t t Maternidade Bahia.
Fellow in Reproductive Climerio de Oliveira,
Endocrinology. Federal University
Fertility
and
Sterilityycu
of
liculogenesis resulting in the development of an unluteinized enlarged follicle (7). The secretory patterns of gonadotropins change throughout the menstrual cycle and play a fundamental role in follicular development, ovulation, and corpus luteum function (8,9). The study of LH pulse patterns is an important tool to assess the hypothalamic-pituitary-ovarian axis and evaluate normal and abnormal folliculogenesis (10, 11). Progesterone alters normal LH pulsatility patterns and the LH surge (12,13). Incomplete suppression of gonadotropins or an altered ratio of gonadotropin secretion may result in abnormal folliculogenesis (14, 15) and the development of an enlarged follicle (15, 16). To enhance our understanding of anovulation caused by progestagen only subdermal implants, this study evaluates the hypothalamic-pituitaryovarian axis by measuring LH pulse patterns, serum gonadotropin concentrations, and follicular development in users of a single subdermal implant of nomegestrol acetate. Analysis of these patterns was conducted immediately after insertion and during prolonged use of the implant. Differences in the parameters also are evaluated in subjects who developed follicular growth and those devoid of follicular activity during use of the implant. MATERIALS
AND
METHODS
Subjects
Ten healthy women, 22 to 36 years of age, requesting a contraceptive method at the family planning clinic at San Borja-Arriaran Clinical HospitalNational Health Service-School of Medicine, University of Chile, participated in this study. Inclusion criteria included non-breast-feeding women with a normal cycle (28 to 32 days), a normal body mass index (BMI = weight [kg]/height [m21), a hemoglobin concentration > 12 mg/dL, and absence of use of any hormonal contraception in the previous 6 months. Each subject was able to understand and was willing to comply with the study protocol and signed an informed consent approved by the Institutional Review Board. Study
67, No. 2, February
Venipuncture
1997
and
Continuous
Blood
Sampling
An 18-gauge angiocatheter was placed in the forearm of the subject and blood collected continuously as described previously (18). Approximately 3 mL of blood was collected into a separate test tube every 10 minutes, using an automatic carousel (model 2110, Bio-Rad, Hercules, CA). The serum was separated by centrifugation and stored at -20°C until assayed. Assay
Both LH and FSH were measured in duplicates by RIA method using National Institute of Diabetes and Digestive and Kidney Disease reagents as reported previously (19). Luteinizing hormone and FSH are expressed in mIU/mL (conversion factor to SI unit, 1.00) using LER 907 as the standard. The intra-assay and interassay coefficients of variation (CV) were 7% and 8%, respectively. All s.amples from each patient were analyzed with the same RIA protocol. Estradiol and P were measured as reported previously (20). Statistics
Design
Each woman underwent three studies of LH pulsatile secretions. All study periods started between 8:00 A.M. and 9:00 A.M. and continued for 8 hours in an awake patient. Blood samples were collected every 10 minutes. Study I was conducted in the early follicular phase (days 3 to 4) of the menstrual cycle, just before insertion of the nomegestrol acetate implant. Study II was conducted 48 hours after insertion (days 5 to 6). Subjects then were evaluated proVol.
spectively with twice weekly transvaginal ultrasound (630 5 MHz vaginal probe; Aloka, Tokyo, Japan) and twice weekly serum Ez conce:ntrations to determine the presence or absence of follicular activity. Study III was conducted either during follicular growth (follicle size 14 to 20 mm) or, if no follicle activity was detected, after 30 to 45 days of nomegestrol acetate use. If follicular development was not initiated within 30 to 45 days, it was presumed folliculogenesis had been inhibited and the final LH pulse study was performed to compare with the LH pulse profile of subjects who had initiated folliculogenesis. In subjects who initiated folliculogenesis, the presence of subsequent ovulation was assessed with continued ultrasound to detect follicular rupture and serum P concentrations. At the end of this investigation, the subjects continued the use of the subdermal implant and were followed as part of the clinical trial evaluating the efficacy and. acceptability of this contraceptive device (17).
and Pulse
Analysis
The mean of duplicate values for LH of each sample was taken as the hormone value. Significant serum LH pulses were identified as described by Veldhuis et al. (21). To be detected as a significant peak, the LH concentration of two consecutive samples had to differ by at least twice the interassay CV at the level of the nadir, thus reducing the likelihood of false positive determinations to a ratio of ~5%. The analysis of pulsatile LH secretion was carried Barnhart
et al.
Neuroendocrine
mechanism
of anoculation
251
out using Cluster Pulse Analysis as reported previously (19). The transverse means for LH was calculated from 48 consecutive samples (one individual pulse study). The transverse means for FSH and Ea were determined from five samples of the same study. Luteinizing hormone pulse characteristics determined in the cluster analysis were compared by analysis of variance and individual differences determined by the Tacky Cramer test. Intragroup and intergroup differences in serum hormone concentrations and pulse characteristics were evaluated using a paired t-test and unpaired t-test, respectively. Data are presented as means ? SEM. Significance was established at the level of P < 0.05. RESULTS Acute Effect of Nomegestrol Study I (Before Insertion) Hours After Insertion)
of the Presence
of Follicular
The 10 patients were followed prospectively with ultrasound and hormonal evaluation: seven women were noted to develop follicles with maximum diameter ranging from 35 to 48 mm (group A), whereas the other three had no follicular activity detected (follicle < 10 mm), after 45 days of Uniplant insertion (group B). None of the patients in groups A or B subsequently ovulated, as determined by lack of rupture of the follicle as visualized by serial ultrasound in those with the initiation of follicular growth (group A) or by elevated serum P > 4 ng/mL (conversion factor to SI unit, 3.18) 1 week after study III in all subjects (groups A and B). Clinical characteristics between the two groups were not different. There was no difference in the mean age, (group A, 30.5 252
Barnhart
et al.
1 Endocrine Profile and LH Pulse Characteristics Subjects Using a Nomegestrol Acetate Implant, Uniplant, With Follicular Activity (Group A)*
Study
No. of pulses Amplitude ~mIU/mL) LH (treatment) bIU/mL) FSH (treatment) bnIU/mL~
E, (pg/mL)
I
Study II 48 h
preinsertion
postinsertion
(n = 7)
(II = 7)
of
Study III during folliculogenesis (n = 71
6.57
? 0.57
6.26
t
0.28
6.85
t
0.671
3.95
+ 0.74
6.13
+
0.98-t
3.54
t
0.65
6.41
i- 1.06
8.6
+
2.53
4.72
t
0.36+
i- 0.42 2 4.8
8.8 101.4
? 0.68 + 13
3.36 193.2
11.8 64.6
* Values are means t SEM. Conversion and FSH, 1.00; Ez, 3.67. f No significant difference. Z Significantly different from studies test. § Significantly different from studies ante.
factors
to Sl units
-t 0.238 * 29.43
are as follows:
LH
I and
III, P < 0.05,
I and
II, P < 0.05,
Tukey-Cramer analysis
of vati-
Acetate: Comparison Versus Study II (48
At study I, before the insertion of the implant, the 10 subjects have a mean of 7.1 + 0.52 pulses per 8 hours with an amplitude of LH 4.7 + 1.0 mIU/mL. Serum LH and FSH concentration were 10.3 ? 1.59 and 10.7 2 0.35 mIU/mL, respectively. The LH:FSH ratio was 0.78 5 0.05. Estradiol concentration was 62.57 2 3.61 pg/mL. At study II, 48 hours after implant insertion, there was no significant change in the number of pulses in 8 hours: (6.6 +- 0.4), FSH concentration (9.4 i 0.42 mIU/mL), LH:FSH ratio (0.85 2 O.l), or Ez concentration (89.9 -+ 9.42 pg/mL. There was an increase in LH pulse amplitude (7.1 i 1.84 mIU/mL; P = 0.08) and concentration (13.3 i 3.04 mIU/mL; P = 0.17) after the insertion of the implant but neither difference reached statistical significance. Determination Development
Table
Neuroendocrine
mechanism
of anouulation
-+-1.73 years; group B 30.6 -+ 3.84 years), BMI, parity, length of menstrual cycle, or day of implant insertion. Evaluation Initiated Ovulate
of Endocrine Profile of Subjects who Follicular Development but Failed to (Group A)
Table 1 illustrates the endocrine profiles of subjects who developed follicular growth (group A, n = 7). There was no change in pulse frequency per 8 hours, preinsertion (study I), 48 hours, postnomegestrol acetate use (study II), or during the development of follicular growth (study III). A significant increase in pulse amplitude was noted at stud,y II (P < 0.05) compared with study I and study III. Luteinizing hormone concentrations also were elevated at study II, but this was not a significant cha.nge compared with study I or study III. Serum FSH was not significantly different at study I and study II but was significantly lower at study III (P < 0.05). Serum E2 concentration was not significantly different at study I and study II but was significantly elevated at study III (P < 0.05). Evaluation of Endocrine Profile of Subjects Were Devoid of Follicular Activity (Group
who B)
Table 2 illustrates the endocrine profiles of subjects devoid of follicular maturation (group B, n = 3). There was no significant difference in pulse frequency per 8 hours, at preinsertion (study I), and 48 hours postinsertion of nomegestrol
and Sterility8
Table 2 Endocrine Profile and LH Pulse Characteristics Subjects Using a Nomegestrol Acetate Implant, Uniplant, Without Follicular Activity (Group B):’
Study I preinsertion (n = 3) No. of pulses Amplitude (mIU/mL) LH (treatment) (mIU/mL) FSH (treatment1 (mIU/mL) E2 WmL)
Study II 48 h postinsertion (n = 3)
of
Study III 30 to 45 d postinsertion (n = 3)
8.33
t
0.88
7.33
t 1.2
3.66
r 0.66?
3.00
+
0.21
3.92
t 0.73
5.76
t 1.73i
6.41
i-
3.15
8.4
r 6.00
7.05
f 1.98$
E 0.49 t 12
10.2 70.4
+ 0.39 t- 9.5
11.71 67.5
9.48 59.16
Table 3 Comparison of Endocrine Profile and LH Pulse Characteristics of Subjects After Prolonged use of’ a Nomegestrol Acetate Implant, Uniplant, Between Those With and Without Follicular Activity (Groun A Versus Groun Bl* Group A (n = 7) No. of pulses Amplitude (mIU/mLl LH (treatment) (mIU/mL 1 FSH (treatment) ImIUimL) LH:FSH ratio EZ (pg/mL)
lr 1.31 i 4j:
* Values are means r~ SEM. Conversion factors to SI units are as follows: LH and FSH, 1.00; Ez, 3.67. t Significantly different from studies I and II, P < 0.05, analysis of variance. $ No significant difference.
Group B (n = 3)
6.85
k
0.67
3.66
r 0.66.:
3.54
+
0.65
5.76
t 1.73t
4.72
L
0.3
7.05
t 1.981
11.71 0.6 67.5
2 1.3’1 -c 0.13-i+ 4f
3.36 t 0.23 1.36 2 0.33 193.2 i 29.4
* Values are means + SEM. Conversion factors to SI units as follows: LH and FSH, 1.00; Ey, 3.67. t Significantly different from group A, P < 0.05, unpaired test.
are t-
that subjects from group A developed a LH pulse profile similar to that found during the follicular phase of the normal menstrual cycle, whereas group B developed a LH pulse profile consistent with that of midluteal phase of the normal menskual cycle.
II, however, pulse amplitude was increased significantly at study III (P < 0.05). Serum LH concentration was not significantly different in the three study periods. Follicle-stimulating hormone concentration was not statistically different in study I and study II, whereas it was significantly higher at study III (P < 0.05). Serum Ez concentrations were not different across the three study periods.
DISCUSSION
This study indicates that the subdermal implant of nomegestrol acetate induces anovulation particu-
Comparison of Subjects With Follicular Development and With Subjects Devoid of Follieular Activity (Group A Versus Group B)
There was no significant difference in the mean number of pulses, amplitude of pulses, serum LH concentration, LH:FSH ratio, or mean Ez concentration between the two groups at study period I and study period II. Serum FSH concentrations were significantly higher in group A than B at study I (group A, 11.8 -+ 0.42; group B, 9.48 I 0.49; P < 0.05) but not at study II (group A, 8.8 ? 0.68; group B, 10.2 t- 0.39). Table 3 represents the comparison of the endocrine findings of groups A and B at study III. At study III, group A demonstrated a significantly higher number of pulses in 8 hours compared with group B (P < 0.05) with a significantly lower pulse amplitude (P < 0.05). Serum LH and FSH concentrations were significantly higher in group B fP < 0.05), whereas the LH:FSH ratio and serum E2 concentration was significantly greater in group A (P < 0.05). Figure 1 illustrates a representative LH pulse profile from a subject in group A and another subject in group B at study periods I, II, and III. Subjects from both groups showed a similar LH pulse profile preinsertion (study I) and acutely postinsertion of nomegestrol acetate (study II). Note, at study period III, Vol.
67, No. 2, February
1997
h.
II.
I
Figure 1 The LH pulse profile in two representative sub.jects illustrating the LH pulse patterns of subjects who developed a persistent nonluteinized follicle (group A, l ), and subjects who remained devoid of follicular activity (group B, 0). Study I was conducted in the early follicular phase (days 3 to 4) of the menstrual cycle, before insertion of the nomegestrol acetate implant. Study II was conducted 48 hours after insertion (days 5 to 61. Study III was conducted either during follicular growth (group A) or, if no follicle activity was detected, after 30 to 45 days of nomegestrol acetate use (group B).
Barnhart
et al.
Neuroendocrine
mechanism
of anovulation
253
larly in the first few months of use. Thirty percent (3110) of subjects were devoid of any follicular activity for the duration of the study, whereas the remaining 70% of subjects (7110) developed follicular growth that failed to luteinize or ovulate. Age, parity, BMI, or menstrual cycle characteristics were not different in those subjects who would initiate or became devoid of follicular activity. An elevated serum FSH in those who eventually developed follicular activity (group A) was the only significant difference between the two groups before insertion of the nomegestrol acetate implant. It is not clear if this difference is a result of multiple statistical comparisons or if it is of true clinical significance. A difference in FSH concentration was not noted at study II nor was there a significant difference in Ez concentration or follicular diameter as measured by ultrasound (data not shown). No changes in endocrine parameters or pulse characteristics noted immediately after insertion of the implant were predictive of subsequent presence or absence of follicular activity. Therefore, either the acute endocrine effect of nomegestrol acetate is not predictive of the suppression of folliculogenesis or the study of LH pulse characteristics is not a sensitive enough tool to detect such a change. Nevertheless, an increase in the LH pulse amplitude, with no associated change in LH pulse frequency, was noted after the insertion of the implant. This finding was statistically significant when evaluating the subjects in group A but did not reach significance when comparing the evaluation of all subjects (P = 0.08). This acute effect of a progestin is similar to the acute stimulatory effect of P on the pituitary, with no effect on the hypothalamus, as noted by Couzinet et al. cm. Seventy percent of subjects (group A) initiated folliculogenesis as demonstrated by an increase in the diameter of a leading follicle measured by vaginal ultrasound and a progressive increase in Ez concentration. In this group, serum FSH concentration decreased and the LH:FSH ratio increased at study III as compared with study I and study II. This change was likely a result of the negative feedback of Ez on the pituitary. None of these subjects ovulated or luteinized, as demonstrated by a lack of collapse of the leading follicle or rise in serum P concentrations. The LH pulse characteristics in this group were indistinguishable during all three study periods and remained similar to pulse characteristics consistent with the early follicular phase (8). Nomegestrol acetate had no effect on hypothalamic pulse generation in this group but likely had a direct effect on the pituitary reducing the periovulatory hypersensitivity to GnRH and Ez. Thus, the sustained elevated release of LH needed for ovulation and luteinization 254
Barnhart
et al.
Neuroendocrine
mechanism
of anovulation
of the dominant follicle was inhibited. Other possibilities are that nomegestrol acetate acted on the hypothalamus preventing the increase in pulse frequency associated with the late follicular phase (B), reduced the pituitary’s hypersensitivity to GnRH pulses by reducing the positive feedback of GnRH on GnRH receptors, or even acted directly on the ovary. Thirty percent of subjects (group El) were devoid of any follicular development, as demonstrated by a lack of follicular growth by vaginal ultrasound and serum E, concentrations. After 30 to .45 days use of the nomegestrol acetate implant (study III), pulse characteristics changed from those noted preinsertion and immediately postinsertion. In the initial study period, a pattern of high frequency and relative low amplitude, consistent with th.at of the follicular phase of a normal menstrual (cycle (81, was noted. In study III, LH pulse characteristics were noted to be of low frequency with an increased amplitude, consistent with the midluteal phase of a normal menstrual cycle (8). This alteration in pulse characteristics is consistent with the effect of P on the hypothalamus when given in the normal follicular phase, as demonstrated by Soules et al. (13). In addition, this LH pulse profile has been shown to be inappropriate to induce follicular growth (8, 11). A possible explanation of the initial two distinct effects of nomegestrol acetate on the hlypothalamuspituitary-ovarian axis could be related serum concentrations of nomegestrol acetate after the insertion. Higher doses may suppress the lhypothalamus and eliminate the initiation of folliculogenesis whereas lower concentrations may effect feedback at the level of the pituitary or act directly on the ovary, preventing follicle luteinization and ovulation. A dose-related effect on the pituitary-ovarian axis has in fact been reported with oral use of nomegestrol acetate (23). In conclusion, these data indicate two types of anovulation: type I is characterized by follicular development with absence of luteinization or ovulation (70%) and type II is characterized by the absence of any follicular growth (30%). Clinical characteristics of the patients, serum gonadotropin concentrations, and LH pulse characteristics, before and acutely after insertion, failed to predict which subjects would initiate follicular activity. Anovulation in nomegestrol acetate users who lack any follicular development is associated with a diminished frequency of GnRH pulse generation demonstrating a suppressive effect at the level of the hypothalamus. Seventy percent of subjects using an implant of nomegestrol acetate initiated follicular development but failed to ovulate. In this group of subjects, nomegestrol acetate likely acts preferentially at the level of Fertility
and Sterility@
the pituitary, reducing sulting in an persistent
Acknowledgments. ily Planning Clinic in particular, Mrs.
or suppressing non luteinized
LH surge, refollicle.
We are grateful to the midwives of the Famof the San Borja-ArriarBn Clinic Hospital and, Ivon Gallegos, for her technical assistance.
REFERENCES 1. Coutinho EM. One year contraception with a single implant containing Nomegestrol acetate (Uniplant). Contraception 1993;47:94-105. 2. Brache V, Alvarez-Sanchez F, Faundes A, Tejada AS, Cochon L. Ovarian endocrine function through five years of continuous treatment with Norplant subdermal contraceptive implants. Contraception 1990;41:169-77. 3. Brache V, Faundes A. Anovulation, inadequate luteal phase and poor sperm penetration in cervical mucus during prolonged use of Norplant implants. Contraception 1985;31:261. 4. Croxatto HB, Diaz S. Treatment with Norplant subdermal implants inhibits sperm penetration through cervical mucus in vitro. Contraception 1987;36:193. 5. Shaaban MM, Segal S, Salem HT, Ghaneimah SA, Khalifa E-AM, Ahmed A-G. Sonographic assessment of ovarian and endometrial changes during long-term Norplant use and their correlation with hormonal levels. Fertil Steril 1993;59:998-1002. 6. Ladipo 0, Coutinho EM. Contraceptive implants. Curr Opin Obstet Gynecol 1994;6:564-9. 7. Devoto I, Pommer R, Barnhart K, Alba F, Sir T, Robinovich J, et al. Ovarian function in subjects using Uniplant. Contracept Fertil Sex (Paris) 1995;(Suppl Nl-N9)23;S-32. 8. Filicori M, Santoro N, Merriam G, Crowley W. Characterization of the physiological pattern of episodic gonadotropin secretion throughout the human menstrual cycle. J Clin Endocrinol Metab 1986;62:1136. 9. Veldhuis J, Beitins I, Johnson M, Serabian M, Dufau L. Biologically active luteinizing hormone is secreted in episodic pulsations that vary in relation to the stage of the menstrual cycle. J Clin Endocrinol Metab 1984;58:1050. 10. Bennet A, Lacaze C, Berrada C, Barbe V, Louvet J. Correlations between mean LH levels and LH pulse characteristics: differences between normal and anovulatory women. Clin Endocrinol (Oxf) 1991;35:431-7. 11. Schweiger U, Reinhard T, Broocks A, Karl-Martin P. Gonadotropin secretion in the second half of the menstrual cycle: a comparison of women with normal cycles, luteal phase defects
Vol.
67, No. 2, February
1997
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
Barnhart
and disrupted follicular development Clin Endocrinol (0x0 1990;32:25-32. Vandenberg G, DeVane G, Yen SSC. Effects of exogenous estrogen and progestin on pituitary responsiveness to synthetic luteinizing hormone-releasing factor. J Clin Invest 1974;53:1750. Soules M, Steiner R, Clifton D, Cohen N, Aksel S, Bremner W. Progesterone modulation of pulsatile luteinizing hormone secretion in normal women. J Clin Endocrinol Metab 1984;58:378. Poindexter A, Dilby G, Brody S, Snabes M. The effects of a long-acting progestin on the hypothalamic-pituitary-ovarian axis in women with normal menstrual cycles. Contraception 1993;48:37-45. Alvarez-Sanchez F, Brache V, Faundes A, Tejada AS. Abnormal endocrine profile among women with confirmed presumed ovulation during long-term Norplant use. Contraception 1986;33:111-9. Cook D, Parfet J, Smith C, Moss G, Youngquist R, Garverick H. Secretory patterns of LH and FSH during development and hypothalamic and hypophysial characteristics following development of steroid-induced ovarian follicular cysts in dairy cattle. J Reprod Fert 1991;91:19-28. Coutinho E, De Souza JA, Athayde C, Barbo,sa I, Alverez F, Brache V, et al. Multicenter clinical trail on the efficacy and acceptability of a single contraceptive implant of nomegestrol acetate, Uniplant. Contraception 1996;53:121-5. Sir-Petterman T, Recabarren S, Bitt A, Jager W, Zimmermann V, Wila L. A simple device for serial blood collection in human subjects and animals. Exp Clin Endocrinol Diabetes 1995;103:398-401. Sir-Petterman T, Devoto L. Effect of clomiphene citrate on pulsatile luteinizing hormone profile in normal women and polycystic ovary syndrome. Horm Metab Res 1989;21:37036. Devoto L, Vega M, Navarro V, Sir T, Alba F, Castro 0. Regulation of steroid hormone synthesis by human corpora lutea: failure of follicle stimulating hormone to support luteal steroidogenesis in vivo and in vitro. Fertil Steril 1989;50:62833. Veldhuis JD, Royel AD, Johnson ML. Minimizing false-positive errors in hormonal pulse detection. Am J Physiol 1985;248:E475-81. Couzinet B, Brailly S, Bouchard P, Schaison G. Progesterone stimulates luteinizing hormone secretion by acting directly on the pituitary. J Clin Endocrinol Metab 1992;74:374-8. Brazin B, Thevenot R, Bursaux C, Paris J. Effect of nomegestrol acetate, a new 19-nor-progesterone derivative, on pituitary-ovarian function in women. Br J Olbstet Gynaecol 1987;94:1199-204.
et al.
Neuroendocrine
mechanism
of anovulation
255